man pages section 7: Device and

Network Interfaces
Beta

Part No: 819–2254–33

February 2010
Copyright ©2010 Sun Microsystems, Inc. All Rights Reserved 4150 Network Circle, Santa Clara, CA 95054 U.S.A. All rights reserved.

Sun Microsystems, Inc. has intellectual property rights relating to technology embodied in the product that is described in this document. In particular, and without
limitation, these intellectual property rights may include one or more U.S. patents or pending patent applications in the U.S. and in other countries.
U.S. Government Rights – Commercial software. Government users are subject to the Sun Microsystems, Inc. standard license agreement and applicable provisions
of the FAR and its supplements.
This distribution may include materials developed by third parties.
Parts of the product may be derived from Berkeley BSD systems, licensed from the University of California. UNIX is a registered trademark in the U.S. and other
countries, exclusively licensed through X/Open Company, Ltd.
Sun, Sun Microsystems, the Sun logo, the Solaris logo, the Java Coffee Cup logo, docs.sun.com, Java, and Solaris are trademarks or registered trademarks of Sun
Microsystems, Inc. in the U.S. and other countries. All SPARC trademarks are used under license and are trademarks or registered trademarks of SPARC
International, Inc. in the U.S. and other countries. Products bearing SPARC trademarks are based upon an architecture developed by Sun Microsystems, Inc.
The OPEN LOOK and SunTM Graphical User Interface was developed by Sun Microsystems, Inc. for its users and licensees. Sun acknowledges the pioneering efforts
of Xerox in researching and developing the concept of visual or graphical user interfaces for the computer industry. Sun holds a non-exclusive license from Xerox to
the Xerox Graphical User Interface, which license also covers Sun's licensees who implement OPEN LOOK GUIs and otherwise comply with Sun's written license
agreements.
Products covered by and information contained in this publication are controlled by U.S. Export Control laws and may be subject to the export or import laws in
other countries. Nuclear, missile, chemical or biological weapons or nuclear maritime end uses or end users, whether direct or indirect, are strictly prohibited. Export
or reexport to countries subject to U.S. embargo or to entities identified on U.S. export exclusion lists, including, but not limited to, the denied persons and specially
designated nationals lists is strictly prohibited.
DOCUMENTATION IS PROVIDED “AS IS” AND ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND WARRANTIES, INCLUDING ANY
IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT, ARE DISCLAIMED, EXCEPT TO
THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID.

Copyright ©2010 Sun Microsystems, Inc. All Rights Reserved 4150 Network Circle, Santa Clara, CA 95054 U.S.A. Tous droits réservés.
Sun Microsystems, Inc. détient les droits de propriété intellectuelle relatifs à la technologie incorporée dans le produit qui est décrit dans ce document. En particulier,
et ce sans limitation, ces droits de propriété intellectuelle peuvent inclure un ou plusieurs brevets américains ou des applications de brevet en attente aux Etats-Unis
et dans d'autres pays.
Cette distribution peut comprendre des composants développés par des tierces personnes.
Certaines composants de ce produit peuvent être dérivées du logiciel Berkeley BSD, licenciés par l'Université de Californie. UNIX est une marque déposée aux
Etats-Unis et dans d'autres pays; elle est licenciée exclusivement par X/Open Company, Ltd.
Sun, Sun Microsystems, le logo Sun, le logo Solaris, le logo Java Coffee Cup, docs.sun.com, Java et Solaris sont des marques de fabrique ou des marques déposées de
Sun Microsystems, Inc. aux Etats-Unis et dans d'autres pays. Toutes les marques SPARC sont utilisées sous licence et sont des marques de fabrique ou des marques
déposées de SPARC International, Inc. aux Etats-Unis et dans d'autres pays. Les produits portant les marques SPARC sont basés sur une architecture développée par
Sun Microsystems, Inc.
L'interface d'utilisation graphique OPEN LOOK et Sun a été développée par Sun Microsystems, Inc. pour ses utilisateurs et licenciés. Sun reconnaît les efforts de
pionniers de Xerox pour la recherche et le développement du concept des interfaces d'utilisation visuelle ou graphique pour l'industrie de l'informatique. Sun détient
une licence non exclusive de Xerox sur l'interface d'utilisation graphique Xerox, cette licence couvrant également les licenciés de Sun qui mettent en place l'interface
d'utilisation graphique OPEN LOOK et qui, en outre, se conforment aux licences écrites de Sun.
Les produits qui font l'objet de cette publication et les informations qu'il contient sont régis par la legislation américaine en matière de contrôle des exportations et
peuvent être soumis au droit d'autres pays dans le domaine des exportations et importations. Les utilisations finales, ou utilisateurs finaux, pour des armes nucléaires,
des missiles, des armes chimiques ou biologiques ou pour le nucléaire maritime, directement ou indirectement, sont strictement interdites. Les exportations ou
réexportations vers des pays sous embargo des Etats-Unis, ou vers des entités figurant sur les listes d'exclusion d'exportation américaines, y compris, mais de manière
non exclusive, la liste de personnes qui font objet d'un ordre de ne pas participer, d'une façon directe ou indirecte, aux exportations des produits ou des services qui
sont régis par la legislation américaine en matière de contrôle des exportations et la liste de ressortissants spécifiquement designés, sont rigoureusement interdites.
LA DOCUMENTATION EST FOURNIE "EN L'ETAT" ET TOUTES AUTRES CONDITIONS, DECLARATIONS ET GARANTIES EXPRESSES OU TACITES
SONT FORMELLEMENT EXCLUES, DANS LA MESURE AUTORISEE PAR LA LOI APPLICABLE, Y COMPRIS NOTAMMENT TOUTE GARANTIE
IMPLICITE RELATIVE A LA QUALITE MARCHANDE, A L'APTITUDE A UNE UTILISATION PARTICULIERE OU A L'ABSENCE DE CONTREFACON.

100217@23474
Contents

Preface ...................................................................................................................................................15

Introduction .........................................................................................................................................19
Intro(7) ................................................................................................................................................ 20

Device and Network Interfaces ......................................................................................................... 23

aac(7D) ................................................................................................................................................. 24
adp(7D) ................................................................................................................................................. 25
adpu320(7D) ........................................................................................................................................ 27
afb(7d) .................................................................................................................................................. 29
afe(7D) ................................................................................................................................................. 30
agpgart_io(7I) .................................................................................................................................... 32
ahci(7D) ............................................................................................................................................... 43
amd8111s(7D) ...................................................................................................................................... 44
amr(7D) ................................................................................................................................................. 45
arcmsr(7D) .......................................................................................................................................... 46
arn(7D) ................................................................................................................................................. 48
arp(7P) .................................................................................................................................................. 49
asy(7D) ................................................................................................................................................. 54
ata(7D) ................................................................................................................................................. 57
atge(7D) ............................................................................................................................................... 62
ath(7D) ................................................................................................................................................. 63
atu(7D) ................................................................................................................................................. 64
audio1575(7D) .................................................................................................................................... 65
audio(7D) ............................................................................................................................................. 66
audio(7I) .............................................................................................................................................. 69
audio810(7D) ...................................................................................................................................... 80

3
Contents

audiocmi(7D) ...................................................................................................................................... 81
audiocs(7D) ........................................................................................................................................ 82
audioemu10k(7D) ................................................................................................................................ 83
audioens(7D) ...................................................................................................................................... 84
audiohd(7D) ........................................................................................................................................ 85
audioixp(7D) ...................................................................................................................................... 86
audiols(7D) ........................................................................................................................................ 87
audiop16x(7D) .................................................................................................................................... 88
audiopci(7D) ...................................................................................................................................... 89
audiosolo(7D) .................................................................................................................................... 90
audiots(7D) ........................................................................................................................................ 91
audiovia823x(7D) .............................................................................................................................. 92
audiovia97(7D) .................................................................................................................................. 93
av1394(7D) .......................................................................................................................................... 94
bbc_beep(7D) ...................................................................................................................................... 95
bcm_sata(7D) ...................................................................................................................................... 96
bfe(7D) ................................................................................................................................................. 97
bge(7D) ................................................................................................................................................. 98
blk2scsa(7D) .................................................................................................................................... 101
bnx(7D) ............................................................................................................................................... 102
bnxe(7D) ............................................................................................................................................. 106
bpf(7D) ............................................................................................................................................... 111
bpp(7D) ............................................................................................................................................... 121
bscv(7D) ............................................................................................................................................. 126
bufmod(7M) ........................................................................................................................................ 127
cadp160(7D) ...................................................................................................................................... 131
cadp(7D) ............................................................................................................................................. 132
cdio(7I) .............................................................................................................................................. 136
ce(7D) ................................................................................................................................................. 144
chxge(7D) ........................................................................................................................................... 148
cmdk(7D) ............................................................................................................................................. 149
connld(7M) ........................................................................................................................................ 151
console(7D) ...................................................................................................................................... 153
cpqary3(7D) ...................................................................................................................................... 154
cpr(7) .................................................................................................................................................. 156
cpuid(7D) ........................................................................................................................................... 158

4 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
 Contents

ctfs(7FS) ........................................................................................................................................... 161

ctsmc(7D) ........................................................................................................................................... 162
cvc(7D) ............................................................................................................................................... 163
cvcredir(7D) .................................................................................................................................... 164
dad(7D) ............................................................................................................................................... 165
daplt(7D) ........................................................................................................................................... 168
dca(7D) ............................................................................................................................................... 169
dcam1394(7D) .................................................................................................................................... 171
dcfs(7FS) ........................................................................................................................................... 180
dda(7D) ............................................................................................................................................... 181
dev(7FS) .............................................................................................................................................. 182
devfs(7FS) ......................................................................................................................................... 183
devinfo(7D) ...................................................................................................................................... 184
dkio(7I) .............................................................................................................................................. 185
dlcosmk(7ipp) .................................................................................................................................... 197
dlpi(7P) ............................................................................................................................................. 198
dm2s(7D) ............................................................................................................................................. 207
dmd(7D) ............................................................................................................................................... 208
dmfe(7D) ............................................................................................................................................. 212
dnet(7D) ............................................................................................................................................. 214
dpt(7D) ............................................................................................................................................... 216
dr(7d) .................................................................................................................................................. 218
dscpmk(7ipp) ...................................................................................................................................... 219
dsp(7I) ................................................................................................................................................. 220
dtrace(7D) ........................................................................................................................................ 228
e1000g(7D) ........................................................................................................................................ 229
ecpp(7D) ............................................................................................................................................. 234
ehci(7D) ............................................................................................................................................. 240
elxl(7D) ............................................................................................................................................. 244
emlxs(7D) ........................................................................................................................................... 247
eri(7D) ............................................................................................................................................... 248
esp(7D) ............................................................................................................................................... 252
fas(7D) ............................................................................................................................................... 259
fasttrap(7D) .................................................................................................................................... 267
fbio(7I) .............................................................................................................................................. 268
fbt(7D) ............................................................................................................................................... 270

5
Contents

fcip(7D) ............................................................................................................................................. 271

fcoe(7D) ............................................................................................................................................. 274
fcoei(7D) ........................................................................................................................................... 275
fcoet(7D) ........................................................................................................................................... 276
fcp(7D) ............................................................................................................................................... 277
fctl(7D) ............................................................................................................................................. 278
fd(7D) ................................................................................................................................................. 279
fdio(7I) .............................................................................................................................................. 284
ffb(7D) ............................................................................................................................................... 288
flowacct(7ipp) .................................................................................................................................. 289
fp(7d) .................................................................................................................................................. 290
FSS(7) .................................................................................................................................................. 292
ge(7D) ................................................................................................................................................. 295
gld(7D) ............................................................................................................................................... 299
glm(7D) ............................................................................................................................................... 308
gpio_87317(7D) ................................................................................................................................ 314
grbeep(7d) ......................................................................................................................................... 315
hci1394(7D) ...................................................................................................................................... 316
hdio(7I) .............................................................................................................................................. 317
heci(7D) ............................................................................................................................................. 319
hermon(7D) ........................................................................................................................................ 320
hid(7D) ............................................................................................................................................... 322
hme(7D) ............................................................................................................................................... 325
hpfc(7D) ............................................................................................................................................. 330
hsfs(7FS) ........................................................................................................................................... 332
hubd(7D) ............................................................................................................................................. 336
hwahc(7D) ........................................................................................................................................... 339
hwarc(7D) ........................................................................................................................................... 340
hxge(7D) ............................................................................................................................................. 341
i915(7d) .............................................................................................................................................. 344
ib(7D) ................................................................................................................................................. 345
ibcm(7D) ............................................................................................................................................. 348
ibd(7D) ............................................................................................................................................... 349
ibdm(7D) ............................................................................................................................................. 353
ibdma(7D) ........................................................................................................................................... 354
ibmf(7) ................................................................................................................................................ 355

6 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
 Contents

ibtl(7D) ............................................................................................................................................. 356

icmp6(7P) ........................................................................................................................................... 357
icmp(7P) ............................................................................................................................................. 359
idn(7d) ................................................................................................................................................ 361
iec61883(7I) ...................................................................................................................................... 364
ieee1394(7D) .................................................................................................................................... 373
ifb(7d) ................................................................................................................................................ 375
ifp(7D) ............................................................................................................................................... 376
if_tcp(7P) ......................................................................................................................................... 380
igb(7D) ............................................................................................................................................... 389
ii(7D) ................................................................................................................................................. 392
inet6(7P) ........................................................................................................................................... 394
inet(7P) ............................................................................................................................................. 399
ip6(7P) ................................................................................................................................................ 402
ip(7P) .................................................................................................................................................. 410
ipge(7D) ............................................................................................................................................. 419
ipgpc(7ipp) ........................................................................................................................................ 423
ipnat(7I) ............................................................................................................................................ 425
ipnet(7D) ........................................................................................................................................... 432
ipqos(7ipp) ........................................................................................................................................ 434
iprb(7D) ............................................................................................................................................. 436
ipsec(7P) ........................................................................................................................................... 439
ipsecah(7P) ....................................................................................................................................... 443
ipsecesp(7P) ..................................................................................................................................... 444
ipw(7D) ............................................................................................................................................... 445
iscsi(7D) ........................................................................................................................................... 446
isdnio(7I) .......................................................................................................................................... 447
iser(7D) ............................................................................................................................................. 462
isp(7D) ............................................................................................................................................... 463
iwh(7D) ............................................................................................................................................... 470
iwi(7D) ............................................................................................................................................... 471
iwk(7D) ............................................................................................................................................... 472
iwp(7D) ............................................................................................................................................... 473
ixgb(7d) .............................................................................................................................................. 474
ixgbe(7D) ........................................................................................................................................... 476
jfb(7D) ............................................................................................................................................... 478

7
Contents

kb(7M) ................................................................................................................................................ 479

kdmouse(7D) ...................................................................................................................................... 488
kfb(7D) ............................................................................................................................................... 489
kmdb(7d) .............................................................................................................................................. 490
kstat(7D) ........................................................................................................................................... 491
ksyms(7D) ........................................................................................................................................... 492
ldterm(7M) ........................................................................................................................................ 494
llc1(7D) ............................................................................................................................................. 497
llc2(7D) ............................................................................................................................................. 500
lockstat(7D) .................................................................................................................................... 507
lofi(7D) ............................................................................................................................................. 508
lofs(7FS) ........................................................................................................................................... 510
log(7D) ............................................................................................................................................... 512
lp(7D) ................................................................................................................................................. 516
lsimega(7D) ...................................................................................................................................... 518
lx_systrace(7D) .............................................................................................................................. 519
m64(7D) ............................................................................................................................................... 520
marvell88sx(7D) .............................................................................................................................. 521
mc-opl(7D) ........................................................................................................................................ 525
md(7D) ................................................................................................................................................. 526
mediator(7D) .................................................................................................................................... 530
mega_sas(7D) .................................................................................................................................... 533
mem(7D) ............................................................................................................................................... 535
mhd(7i) ................................................................................................................................................. 536
mixer(7I) ............................................................................................................................................ 541
mpt(7D) ............................................................................................................................................... 552
mpt_sas(7D) ...................................................................................................................................... 558
mr_sas(7D) ........................................................................................................................................ 560
msglog(7D) ........................................................................................................................................ 562
mt(7D) ................................................................................................................................................. 563
mtio(7I) .............................................................................................................................................. 564
mwl(7D) ............................................................................................................................................... 578
mxfe(7D) ............................................................................................................................................. 579
myri10ge(7D) .................................................................................................................................... 580
n2cp(7d) .............................................................................................................................................. 582
n2rng(7d) ........................................................................................................................................... 584

8 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
 Contents

ncp(7D) ............................................................................................................................................... 586

ncrs(7D) ............................................................................................................................................. 588
nfb(7D) ............................................................................................................................................... 595
nge(7D) ............................................................................................................................................... 596
npe(7D) ............................................................................................................................................... 600
ntwdt(7D) ........................................................................................................................................... 601
ntxn(7D) ............................................................................................................................................. 602
null(7D) ............................................................................................................................................. 604
nulldriver(7D) ................................................................................................................................ 605
nv_sata(7D) ...................................................................................................................................... 606
nxge(7D) ............................................................................................................................................. 607
objfs(7FS) ......................................................................................................................................... 610
oce(7D) ............................................................................................................................................... 611
ohci(7D) ............................................................................................................................................. 612
openprom(7D) .................................................................................................................................... 614
oplkmdrv(7D) .................................................................................................................................... 620
oplmsu(7D) ........................................................................................................................................ 621
oplpanel(7D) .................................................................................................................................... 622
packet(7P) ......................................................................................................................................... 623
pcan(7D) ............................................................................................................................................. 626
pcata(7D) ........................................................................................................................................... 627
pcfs(7FS) ........................................................................................................................................... 629
pcic(7D) ............................................................................................................................................. 635
pcicmu(7D) ........................................................................................................................................ 636
pcie_pci(7D) .................................................................................................................................... 637
pcipsy(7D) ........................................................................................................................................ 638
pcisch(7D) ........................................................................................................................................ 639
pckt(7M) ............................................................................................................................................ 640
pcmcia(7D) ........................................................................................................................................ 641
pcn(7D) ............................................................................................................................................... 642
pcser(7D) ........................................................................................................................................... 644
pcwl(7D) ............................................................................................................................................. 646
pfb(7D) ............................................................................................................................................... 647
pf_key(7P) ......................................................................................................................................... 648
pfmod(7M) .......................................................................................................................................... 662
physmem(7D) ...................................................................................................................................... 666

9
Contents

pipemod(7M) ...................................................................................................................................... 667

pm(7D) ................................................................................................................................................. 668
poll(7d) .............................................................................................................................................. 674
prnio(7I) ............................................................................................................................................ 680
profile(7D) ...................................................................................................................................... 685
ptem(7M) ............................................................................................................................................ 686
ptm(7D) ............................................................................................................................................... 687
pts(7D) ............................................................................................................................................... 689
pty(7D) ............................................................................................................................................... 691
qfe(7d) ................................................................................................................................................ 694
qlc(7D) ............................................................................................................................................... 698
qlge(7D) ............................................................................................................................................. 700
quotactl(7I) ...................................................................................................................................... 701
qus(7D) ............................................................................................................................................... 703
radeon(7d) ......................................................................................................................................... 710
ral(7D) ............................................................................................................................................... 711
ramdisk(7D) ...................................................................................................................................... 712
random(7D) ........................................................................................................................................ 714
rarp(7P) ............................................................................................................................................. 716
rge(7D) ............................................................................................................................................... 717
route(7P) ........................................................................................................................................... 719
routing(7P) ....................................................................................................................................... 724
rtls(7D) ............................................................................................................................................. 726
rtw(7D) ............................................................................................................................................... 727
rum(7D) ............................................................................................................................................... 728
rwd(7D) ............................................................................................................................................... 729
rwn(7D) ............................................................................................................................................... 730
sad(7D) ............................................................................................................................................... 731
sata(7D) ............................................................................................................................................. 734
scfd(7D) ............................................................................................................................................. 737
schpc(7D) ........................................................................................................................................... 738
scsa1394(7D) .................................................................................................................................... 739
scsa2usb(7D) .................................................................................................................................... 742
scsi_vhci(7D) .................................................................................................................................. 747
sctp(7P) ............................................................................................................................................. 750
sd(7D) ................................................................................................................................................. 756

10 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
 Contents

sda(7D) ............................................................................................................................................... 763

SDC(7) .................................................................................................................................................. 764
sdcard(7D) ........................................................................................................................................ 765
sdhost(7D) ........................................................................................................................................ 767
sdp(7D) ............................................................................................................................................... 768
sdt(7D) ............................................................................................................................................... 771
se(7D) ................................................................................................................................................. 772
se_hdlc(7D) ...................................................................................................................................... 776
ses(7D) ............................................................................................................................................... 779
sesio(7I) ............................................................................................................................................ 781
sf(7D) ................................................................................................................................................. 783
sfe(7D) ............................................................................................................................................... 786
sgen(7D) ............................................................................................................................................. 788
sharefs(7FS) ..................................................................................................................................... 793
si3124(7D) ........................................................................................................................................ 794
sip(7P) ................................................................................................................................................ 795
sk98sol(7D) ...................................................................................................................................... 796
skfp(7D) ............................................................................................................................................. 802
slp(7P) ................................................................................................................................................ 804
smbfs(7FS) ......................................................................................................................................... 806
smbios(7D) ........................................................................................................................................ 808
smbus(7D) ........................................................................................................................................... 809
socal(7D) ........................................................................................................................................... 810
sockio(7I) .......................................................................................................................................... 812
sppptun(7M) ...................................................................................................................................... 813
spwr(7D) ............................................................................................................................................. 814
srpt(7D) ............................................................................................................................................. 815
ssd(7D) ............................................................................................................................................... 816
st(7D) ................................................................................................................................................. 821
streamio(7I) ...................................................................................................................................... 839
su(7D) ................................................................................................................................................. 854
sv(7D) ................................................................................................................................................. 857
symhisl(7D) ...................................................................................................................................... 858
sysmsg(7D) ........................................................................................................................................ 861
systrace(7D) .................................................................................................................................... 862
tavor(7D) ........................................................................................................................................... 863

11
Contents

tcp(7P) ................................................................................................................................................ 865

termio(7I) .......................................................................................................................................... 871
termiox(7I) ........................................................................................................................................ 892
ticlts(7D) ........................................................................................................................................ 898
timod(7M) .......................................................................................................................................... 900
tirdwr(7M) ........................................................................................................................................ 902
tmpfs(7FS) ......................................................................................................................................... 904
todopol(7D) ...................................................................................................................................... 906
tokenmt(7ipp) .................................................................................................................................... 907
tpf(7D) ............................................................................................................................................... 910
tsalarm(7D) ...................................................................................................................................... 911
tswtclmt(7ipp) .................................................................................................................................. 915
ttcompat(7M) .................................................................................................................................... 916
tty(7D) ............................................................................................................................................... 924
ttymux(7D) ........................................................................................................................................ 925
tzmon(7d) ........................................................................................................................................... 926
uata(7D) ............................................................................................................................................. 927
uath(7D) ............................................................................................................................................. 930
udfs(7FS) ........................................................................................................................................... 931
udp(7P) ................................................................................................................................................ 932
ufs(7FS) .............................................................................................................................................. 935
ugen(7D) ............................................................................................................................................. 938
uhci(7D) ............................................................................................................................................. 961
ural(7D) ............................................................................................................................................. 962
urtw(7D) ............................................................................................................................................. 963
usba(7D) ............................................................................................................................................. 964
usb_ac(7D) ........................................................................................................................................ 967
usb_ah(7M) ........................................................................................................................................ 969
usb_as(7D) ........................................................................................................................................ 970
usbftdi(7D) ...................................................................................................................................... 972
usb_ia(7D) ........................................................................................................................................ 975
usbkbm(7M) ........................................................................................................................................ 976
usb_mid(7D) ...................................................................................................................................... 978
usbms(7M) .......................................................................................................................................... 980
usbprn(7D) ........................................................................................................................................ 984
usbsacm(7D) ...................................................................................................................................... 989

12 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
 Contents

usbser_edge(7D) .............................................................................................................................. 992

usbsksp(7D) ...................................................................................................................................... 995
usbsprl(7D) ...................................................................................................................................... 998
usbvc(7D) ......................................................................................................................................... 1001
usbwcm(7M) ...................................................................................................................................... 1005
uscsi(7I) .......................................................................................................................................... 1007
usoc(7D) ........................................................................................................................................... 1011
uwba(7D) ........................................................................................................................................... 1013
virtualkm(7D) ................................................................................................................................ 1014
visual_io(7I) .................................................................................................................................. 1017
vni(7d) .............................................................................................................................................. 1024
vr(7D) ............................................................................................................................................... 1025
vt(7I) ................................................................................................................................................. 1027
vuidmice(7M) .................................................................................................................................. 1031
wpi(7D) ............................................................................................................................................. 1034
wscons(7D) ...................................................................................................................................... 1035
wusb_ca(7D) .................................................................................................................................... 1045
wusb_df(7D) .................................................................................................................................... 1046
xge(7D) ............................................................................................................................................. 1047
yge(7D) ............................................................................................................................................. 1049
zcons(7D) ......................................................................................................................................... 1051
zero(7D) ........................................................................................................................................... 1052
zs(7D) ............................................................................................................................................... 1053
zsh(7D) ............................................................................................................................................. 1056
zulu(7d) ............................................................................................................................................ 1060
zyd(7D) ............................................................................................................................................. 1061

13
14
 Preface

Both novice users and those familar with the SunOS operating system can use online man pages
to obtain information about the system and its features. A man page is intended to answer
concisely the question “What does it do?” The man pages in general comprise a reference
manual. They are not intended to be a tutorial.

Overview
The following contains a brief description of each man page section and the information it
references:
■ Section 1 describes, in alphabetical order, commands available with the operating system.
■ Section 1M describes, in alphabetical order, commands that are used chiefly for system
maintenance and administration purposes.
■ Section 2 describes all of the system calls. Most of these calls have one or more error returns.
An error condition is indicated by an otherwise impossible returned value.
■ Section 3 describes functions found in various libraries, other than those functions that
directly invoke UNIX system primitives, which are described in Section 2.
■ Section 4 outlines the formats of various files. The C structure declarations for the file
formats are given where applicable.
■ Section 5 contains miscellaneous documentation such as character-set tables.
■ Section 7 describes various special files that refer to specific hardware peripherals and device
drivers. STREAMS software drivers, modules and the STREAMS-generic set of system calls
are also described.
■ Section 9E describes the DDI (Device Driver Interface)/DKI (Driver/Kernel Interface),
DDI-only, and DKI-only entry-point routines a developer can include in a device driver.
■ Section 9F describes the kernel functions available for use by device drivers.
■ Section 9S describes the data structures used by drivers to share information between the
driver and the kernel.

Below is a generic format for man pages. The man pages of each manual section generally
follow this order, but include only needed headings. For example, if there are no bugs to report,

15
Preface

there is no BUGS section. See the intro pages for more information and detail about each
section, and man(1) for more information about man pages in general.
NAME This section gives the names of the commands or functions
documented, followed by a brief description of what they
do.
SYNOPSIS This section shows the syntax of commands or functions.
When a command or file does not exist in the standard
path, its full path name is shown. Options and arguments
are alphabetized, with single letter arguments first, and
options with arguments next, unless a different argument
order is required.

The following special characters are used in this section:

[ ] Brackets. The option or argument enclosed in
these brackets is optional. If the brackets are
omitted, the argument must be specified.
. . . Ellipses. Several values can be provided for the
previous argument, or the previous argument
can be specified multiple times, for example,
"filename . . ." .
| Separator. Only one of the arguments
separated by this character can be specified at a
time.
{ } Braces. The options and/or arguments
enclosed within braces are interdependent,
such that everything enclosed must be treated
as a unit.
PROTOCOL This section occurs only in subsection 3R to indicate the
protocol description file.
DESCRIPTION This section defines the functionality and behavior of the
service. Thus it describes concisely what the command
does. It does not discuss OPTIONS or cite EXAMPLES.
Interactive commands, subcommands, requests, macros,
and functions are described under USAGE.
IOCTL This section appears on pages in Section 7 only. Only the
device class that supplies appropriate parameters to the
ioctl(2) system call is called ioctl and generates its own
heading. ioctl calls for a specific device are listed
alphabetically (on the man page for that specific device).

16 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
 Preface

ioctl calls are used for a particular class of devices all of

which have an io ending, such as mtio(7I).
OPTIONS This secton lists the command options with a concise
summary of what each option does. The options are listed
literally and in the order they appear in the SYNOPSIS
section. Possible arguments to options are discussed under
the option, and where appropriate, default values are
supplied.
OPERANDS This section lists the command operands and describes
how they affect the actions of the command.
OUTPUT This section describes the output – standard output,
standard error, or output files – generated by the
command.
RETURN VALUES If the man page documents functions that return values,
this section lists these values and describes the conditions
under which they are returned. If a function can return
only constant values, such as 0 or –1, these values are listed
in tagged paragraphs. Otherwise, a single paragraph
describes the return values of each function. Functions
declared void do not return values, so they are not
discussed in RETURN VALUES.
ERRORS On failure, most functions place an error code in the global
variable errno indicating why they failed. This section lists
alphabetically all error codes a function can generate and
describes the conditions that cause each error. When more
than one condition can cause the same error, each
condition is described in a separate paragraph under the
error code.
USAGE This section lists special rules, features, and commands
that require in-depth explanations. The subsections listed
here are used to explain built-in functionality:

Commands
Modifiers
Variables
Expressions
Input Grammar
EXAMPLES This section provides examples of usage or of how to use a
command or function. Wherever possible a complete

17
Preface

example including command-line entry and machine

response is shown. Whenever an example is given, the
prompt is shown as example%, or if the user must be
superuser, example#. Examples are followed by
explanations, variable substitution rules, or returned
values. Most examples illustrate concepts from the
SYNOPSIS, DESCRIPTION, OPTIONS, and USAGE
sections.
ENVIRONMENT VARIABLES This section lists any environment variables that the
command or function affects, followed by a brief
description of the effect.
EXIT STATUS This section lists the values the command returns to the
calling program or shell and the conditions that cause these
values to be returned. Usually, zero is returned for
successful completion, and values other than zero for
various error conditions.
FILES This section lists all file names referred to by the man page,
files of interest, and files created or required by commands.
Each is followed by a descriptive summary or explanation.
ATTRIBUTES This section lists characteristics of commands, utilities,
and device drivers by defining the attribute type and its
corresponding value. See attributes(5) for more
information.
SEE ALSO This section lists references to other man pages, in-house
documentation, and outside publications.
DIAGNOSTICS This section lists diagnostic messages with a brief
explanation of the condition causing the error.
WARNINGS This section lists warnings about special conditions which
could seriously affect your working conditions. This is not
a list of diagnostics.
NOTES This section lists additional information that does not
belong anywhere else on the page. It takes the form of an
aside to the user, covering points of special interest.
Critical information is never covered here.
BUGS This section describes known bugs and, wherever possible,
suggests workarounds.

18 man pages section 7: Device and Network Interfaces • February 2010 (Beta)
R E F E R E N C E

Introduction

19
Intro(7)

Name Intro – introduction to special files

Description This section describes various device and network interfaces available on the system. The
types of interfaces described include character and block devices, STREAMS modules,
network protocols, file systems, and ioctl requests for driver subsystems and classes.

This section contains the following major collections:

(7D) The system provides drivers for a variety of hardware devices, such as disk,
magnetic tapes, serial communication lines, mice, and frame buffers, as well as
virtual devices such as pseudo-terminals and windows.

This section describes special files that refer to specific hardware peripherals and
device drivers. STREAMS device drivers are also described. Characteristics of both
the hardware device and the corresponding device driver are discussed where
applicable.

An application accesses a device through that device's special file. This section
specifies the device special file to be used to access the device as well as application
programming interface (API) information relevant to the use of the device driver.

All device special files are located under the /devices directory. The /devices
directory hierarchy attempts to mirror the hierarchy of system busses, controllers,
and devices configured on the system. Logical device names for special files in
/devices are located under the /dev directory. Although not every special file
under /devices will have a corresponding logical entry under /dev, whenever
possible, an application should reference a device using the logical name for the
device. Logical device names are listed in the FILES section of the page for the device
in question.

This section also describes driver configuration where applicable. Many device
drivers have a driver configuration file of the form driver_name.conf associated
with them (see driver.conf(4)). The configuration information stored in the driver
configuration file is used to configure the driver and the device. Driver
configuration files are located in /kernel/drv and /usr/kernel/drv. Driver
configuration files for platform dependent drivers are located in /platform/‘uname
-i‘/kernel/drv where ‘uname -i‘ is the output of the uname(1) command with the
-i option.

Some driver configuration files may contain user configurable properties. Changes
in a driver's configuration file will not take effect until the system is rebooted or the
driver has been removed and re-added (see rem_drv(1M) and add_drv(1M)).
(7FS) This section describes the programmatic interface for several file systems supported
by SunOS.

20 man pages section 7: Device and Network Interfaces • Last Revised 29 Sep 1994
 Intro(7)

(7I) This section describes ioctl requests which apply to a class of drivers or subsystems.
For example, ioctl requests which apply to most tape devices are discussed in
mtio(7I). Ioctl requests relevant to only a specific device are described on the man
page for that device. The page for the device in question should still be examined for
exceptions to the ioctls listed in section 7I.
(7M) This section describes STREAMS modules. Note that STREAMS drivers are
discussed in section 7D. streamio(7I) contains a list of ioctl requests used to
manipulate STREAMS modules and interface with the STREAMS framework. Ioctl
requests specific to a STREAMS module will be discussed on the man page for that
module.
(7P) This section describes various network protocols available in SunOS.

SunOS supports both socket-based and STREAMS-based network

communications. The Internet protocol family, described in inet(7P), is the
primary protocol family supported by SunOS, although the system can support a
number of others. The raw interface provides low-level services, such as packet
fragmentation and reassembly, routing, addressing, and basic transport for
socket-based implementations. Facilities for communicating using an
Internet-family protocol are generally accessed by specifying the AF_INET address
family when binding a socket; see socket(3SOCKET) for details.

Major protocols in the Internet family include:

■ The Internet Protocol (IP) itself, which supports the universal datagram format,
as described in ip(7P). This is the default protocol for SOCK_RAW type sockets
within the AF_INET domain.
■ The Transmission Control Protocol (TCP); see tcp(7P). This is the default
protocol for SOCK_STREAM type sockets.
■ The User Datagram Protocol (UDP); see udp(7P). This is the default protocol for
SOCK_DGRAM type sockets.
■ The Address Resolution Protocol (ARP); see arp(7P).
■ The Internet Control Message Protocol (ICMP); see icmp(7P).

See Also add_drv(1M), rem_drv(1M), Intro(3), ioctl(2), socket(3SOCKET), driver.conf(4),

arp(7P), icmp(7P), inet(7P), ip(7P), mtio(7I), st(7D), streamio(7I), tcp(7P), udp(7P)

System Administration Guide: IP Services

STREAMS Programming Guide

Writing Device Drivers

Introduction 21
22
R E F E R E N C E

Device and Network Interfaces

23
aac(7D)

Name aac – SCSI HBA driver for Adaptec AdvancedRAID Controller

Description The aac plain SCSI host bus adapter driver is a SCSA-compliant nexus driver that supports the
Adaptec 2200S/2120S SCSI RAID card, Dell PERC 3Di SCSI RAID controller, Dell PERC 3Si
SCSI RAID controller, Adaptec 2820SA SATA RAID card, Adaptec 4800SAS, 4805SAS SAS
RAID cards and SUN's STK RAID REM, STK RAID INT, and STK RAID EXT RAID cards.

The aac driver is ported from FreeBSD and supports RAID disk I/O functions and the RAID
management interface.

Driver There are no user configurable parameters available. Please configure your hardware through
Configuration BIOS.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability x86, SPARC (Limited to systems with AAC hardware

RAID cards.)

Files /kernel/drv/aac 32-bit ELF kernel module.

/kernel/drv/amd64/aac 64-bit ELF kernel module. (x86)
/kernel/drv/sparcv9/aac 64-bit ELF kernel module. (SPARC)
/kernel/drv/aac.conf Configuration file. (Contains no user-configurable options).

See Also prtconf(1M), attributes(5), scsi_hba_attach(9F), scsi_sync_pkt(9F),

scsi_transport(9F), scsi_device(9S), scsi_inquiry(9S), scsi_pkt(9S)

Small Computer System Interface-2 (SCSI-2)

24 man pages section 7: Device and Network Interfaces • Last Revised 27 Dec 2007
 adp(7D)

Name adp – Low-level module for controllers based on Adaptec AIC-7870P and AIC-7880P SCSI
chips

Description The adp module provides low-level interface routines between the common disk/tape I/O
system and SCSI (Small Computer System Interface) controllers based on the Adaptec
AIC-7870P and AIC-7880P SCSI chips. These controllers include the Adaptec AHA–2940,
AHA–2940W, AHA–2940U, AHA–2940UW, AHA–3940, and AHA–3940W, as well as
motherboards with embedded AIC-7870P and AIC-7880P SCSI chips.

Supported devices are AIC-7850, AIC-7860, AIC-7870, AIC-7880 and AIC-7895.

The adp module can be configured for disk and streaming tape support for one or more host
adapter boards, each of which must be the sole initiator on a SCSI bus. Auto-configuration
code determines if the adapter is present at the configured address and what types of devices
are attached to the adapter.

Preconfigure The Plug N Play SCAM Support option is not supported.

Known Problems and ■ To use the AHA-3940 or AHA-3940W adapters, the motherboard must have a BIOS that
Limitations supports the DEC PCI-to-PCI Bridge chip on the host bus adapter.
■ User-level programs have exhibited problems on some PCI systems with an Adaptec
AHA-2940 or AHA-2940W card and certain motherboards. If problems with user-level
programs occur, use the BIOS setup to disable write-back CPU caching (or all caching if
there is no control over the caching algorithm). The affected motherboards are:

- PCI motherboards with a 60-MHz Pentium chip, with PCI chipset numbers S82433LX Z852
and S82434LX Z850. The part numbers of the Intel motherboards are AA616393-007 and
AA615988-009.
- PCI motherboards with a 90-MHz Pentium chip, with PCI chipset numbers S82433NX
Z895, S82434NX Z895, and S82434NX Z896. The part number of the Intel motherboard is
541286-005. (Some Gateway 2000 systems use this motherboard.)
- AA-619772-002 motherboard with 82433LX Z852 and 82434LX Z882 chips causes random
memory inconsistencies. Return the motherboard to the vendor for a replacement.
■ If the AHA-2940 SCSI adapter does not recognize the Quantum Empire 1080S, HP 3323
SE or other SCSI disk drive, reduce the Synchronous Transfer rate on the Adaptec
controller to 8 Mbps.
■ The AHA-3940 has been certified by Adaptec to work on specific systems; however, some
testing has shown that the Solaris operating environment works properly in some of those
systems and not in others.

Configuration Use the Adaptec configuration utility to perform the following steps:

Device and Network Interfaces 25

adp(7D)

■ Configure each SCSI device to have a unique SCSI ID, then using the adapter's Advanced
Configuration Options setup menu, set the Plug N Play SCAM Support option to
Disabled.
■ If there is more than one controller (or an embedded controller), try to use one IRQ per
controller.
■ Enable bus mastering for the slots with your host bus adapters, when the choice is given.
■ For older disk drives, tape drives, and most CD-ROM devices, make sure the maximum
SCSI data transfer speed is set to 5.0 Mbps.
■ Enable support for disks larger than 1 Gbyte if applicable.
Files /kernel/drv/adp.conf Configuration file for the adp driver; there are no
user-configurable options in this file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5)

Solaris (Intel Platform Edition) Hardware Compatibility List

Notes Throughout the release, support of additional devices may be added. See the Solaris (Intel
Platform Edition) Hardware Compatibility List for additional information.

The adp driver supports Logical Unit Number (“LUN”) values of 0 through 15. This range
exceeds the standard SCSI-2 requirements which call for support of LUNs 0 through 7.

26 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 adpu320(7D)

Name adpu320 – Adaptec Ultra320 SCSI host bus adapter driver

Synopsis scsi@unit-address

Description The adpu320 host bus adapter driver is a SCSA-compliant nexus driver that supports the
following Adaptec Ultra320 SCSI Devices:
Chips AIC-7902

The adpu320 driver supports standard functions provided by the SCSA interface, including
tagged and untagged queuing, Wide/Fast/Ultra SCSI, and auto request sense. The adpu320
driver does not support linked commands. The adpu320 driver supports hot swap SCSI and
hot plug PCI.

Additionally, the adpu320 driver supports the following features:

■ 64-bit addressing (Dual address Cycle)
■ PCI-X v1.1 operating up to 133MHz and 64bits
■ PCI bus spec v2.2 operating up to 66MHz and 64bits
■ Packetized SCSI at 320 and 160 MB/s
■ QAS
■ DT
■ 40MB/sec in single-ended mode and up to 320MB/sec transfer rate in LVD mode
■ Domain Validation
■ Retained Training Information (RTI)
■ PCI and PCI-X Error handling

Note – The adpu320 driver does not support the HostRAID feature found on some Adaptec
SCSI controllers. For adpu320 to support a Adaptec SCSI adapter with HostRAID, you must
not use any HostRAID features.

Driver Configuration The adpu320 host bus adapter driver is configured by defining the properties found in
adpu320.conf. Properties in the adpu320.conf file that can be modified by the user include:
ADPU320_SCSI_RD_STRM, ADPU320_SCSI_NLUN_SUPPORT.
-------------------------------------------------------------
Option: ADPU320_SCSI_RD_STRM=[value]
Definition: Enables/disables read streaming negotiation
for all drives.
Possible Values: 0 (off), 1 (on)
Default Value: 0 (off)

Option: ADPU320_SCSI_NLUN_SUPPORT=[value]
Definition: Enables the number of logical units to be
scanned per drive.
Possible Values: 1-64
Default Value: 64
-------------------------------------------------------------

Device and Network Interfaces 27

adpu320(7D)

If you alter or add driver parameters incorrectly, you can render your system inoperable. Use
driver parameters with caution.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability x86

Files /kernel/drv/adpu320 Driver module.

/kernel/drv/amd64/adpu320 64–bit driver module.
/kernel/drv/adpu320.conf Configuration file.

See Also cfgadm(1M), prtconf(1M), attributes(5), scsi_abort(9F), scsi_hba_attach_setup(9F),

scsi_ifgetcap(9F), scsi_reset(9F), scsi_sync_pkt(9F), scsi_transport(9F),
scsi_device(9S), scsi_extended_sense(9S), scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

Small Computer System Interface-3 (SCSI-3)

28 man pages section 7: Device and Network Interfaces • Last Revised 23 May 2006
 afb(7d)

Name afb – Elite3D graphics accelerator driver

Description The afb driver is the device driver for the Sun Elite3D graphics accelerators. The
afbdaemonprocess loads the afb microcode at system startup time and during the resume
sequence of a suspend-resume cycle.
Files /dev/fbs/afbn Device special file
/usr/lib/afb.ucode afb microcode
/usr/sbin/afbdaemon afb microcode loader

See Also afbconfig(1M)

Device and Network Interfaces 29

afe(7D)

Name afe – ADMtek Fast Ethernet device driver

Synopsis /dev/afe

Description The afe driver is a multi-threaded, loadable, clonable, GLD-based STREAMS driver
supporting the Data Link Provider Interface dlpi(7P) on ADMtek (now Infineon) Centaur
and Comet Fast Ethernet controllers.

Application The afe driver can be used as either a "style 1" or a "style 2" Data Link Service Provider.
Programming Physical points of attachment (PPAs) are interpreted as the instance number of the afe
Interface controller as assigned by the Solaris operating environment.

The relevant fields returned as part of a DL_INFO_ACK response are:

■ Maximum SDU is 1500.
■ Minimum SDU is 0.
■ The dlsap address length is 8.
■ MAC type is DL_ETHER.
■ SAP length is -2. The 6–byte physical address is followed immediately by a 2–byte SAP.
■ Service mode is DL_CLDLS.
■ Broadcast address is the 6-byte Ethernet broadcast address (ff:ff:ff:ff:ff:ff).

If the SAP provided is zero, then IEEE 802.3 mode is assumed and outbound frames will have
the frame payload length written into the type field. Likewise, inbound frames with a SAP
between zero and 1500 are interpreted as IEEE 802.3 frames and delivered to any streams that
are bound to SAP zero (the 802.3 SAP).

Properties The following properties may be configured using either ndd(1M) or the afe.conf
configuration file as described by driver.conf(4):
adv_autoneg_cap
Enables (default) or disables IEEE 802.3u auto-negotiation of link speed and duplex
settings. If enabled, the device negotiates among the supported (and configured, see below)
link options with the link partner. If disabled, at least one of the link options below must be
specified. The driver selects the first enabled link option according to the IEEE 802.3u
specified preferences.
adv_100T4_cap
Enables the 100 BaseT4 link option. (Note that most hardware does not support this
unusual link style. Also, this uses two pairs of wires for data, rather than one.)
adv_100fdx_cap
Enables the 100 Base TX full-duplex link option. (This is generally the fastest mode if both
link partners support it. Most modern equipment supports this mode.)

30 man pages section 7: Device and Network Interfaces • Last Revised 1 Apr 2007
 afe(7D)

adv_100hdx_cap
Enables the 100 Base TX half-duplex link option. (Typically used when the link partner is a
100 Mbps hub.)
adv_10fdx_cap
Enables the 10 Base-T full-duplex link option. (This less-frequently used mode is typically
used when the link partner is a 10 Mbps switch.)
adv_10hdx_cap
Enables the 10 Base-T half-duplex link option. (This is the fall-back when no other option
is available. It is typically used when the link partner is a 10 Mbps hub or is an older
network card.)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC, x86

Interface Stability Committed

Files /dev/afe
Special character device.
/kernel/drv/afe
32-bit driver binary (x86).
/kernel/drv/amd64/afe
64-bit driver binary (x86).
/kernel/drv/sparcv9/afe
64-bit driver binary (SPARC).
/kernel/drv/afe.conf
Configuration file.

See Also ndd(1M), driver.conf(4), attributes(5), streamio(7I), dlpi(7P)

IEEE 802.3 — Institute of Electrical and Electronics Engineers, 2002

Device and Network Interfaces 31

agpgart_io(7I)

Name agpgart_io – Solaris agpgart driver I/O control operations

Synopsis #include <sys/agpgart.h>

Description The Accelerated Graphics Port (AGP) is a PCI bus technology enhancement that improves 3D
graphics performance by using low-cost system memory. AGP chipsets use the Graphics
Address Remapping Table (GART) to map discontiguous system memory into a contiguous
PCI memory range (known as the AGP Aperture), enabling the graphics card to utilize the
mapped aperture range as video memory.

The agpgart driver creates a pseudo device node at /dev/agpgart and provides a set of ioctls
for managing allocation/deallocation of system memory, setting mappings between system
memory and aperture range, and setting up AGP devices. The agpgart driver manages both
pseudo and real device nodes, but to initiate AGP-related operations you operate only on the
/dev/agpgart pseudo device node. To do this, open /dev/agpgart. The macro defined for
the pseudo device node name is:
#define AGP_DEVICE "/dev/agpgart"

The agpgart_io driver implementation is AGP architecture-dependent and cannot be made

generic. Currently, the agpgart_io driver only supports specific AGP systems. To determine
if a system is supported, run an open(2) system call on the AGP_DEVICE node. (Note that
open(2) fails if a system is not supported). After the AGP_DEVICE is opened, you can use
kstat(1M) to read the system architecture type.

In addition to AGP system support, the agpgart ioctls can also be used on Intel integrated
graphics devices (IGD). IGD devices usually have no dedicated video memory and must use
system memory as video memory. IGD devices contain translation tables (referred to as GTT
tables) that are similar to the GART translation table for address mapping purposes.

Processes must open the agpgart_io driver utilizing a GRAPHICS_ACCESS privilege. Then
all the ioctls can be called by this processes with the saved file descriptor. With the exception of
AGPIOC_INFO, the AGPIOC_ACQUIRE ioctl must be called before any other ioctl. Once a
process has acquired GART, it cannot be acquired by another process until the former process
calls AGPIOC_RELEASE.

If the AGP_DEVICE fails to open, it may be due to one of the following reasons:
EAGAIN
GART table allocation failed.
EIO
Internal hardware initialization failed.
ENXIO
Getting device soft state error. (This is unlikely to happen.)
EPERM
Without enough privilege.

32 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2008
 agpgart_io(7I)

ioctls With the exception of GPIOC_INFO, all ioctls shown in this section are protected by
GRAPHICS_ACCESS privilege. (Only processes with GRAPHICS_ACCESS privilege in its
effective set can access the privileged ioctls).

Common ioctl error codes are shown below. (Additional error codes may be displayed by
individual ioctls.)
ENXIO
Ioctl command not supported or getting device soft state error.
EPERM
Process not privileged.
AGPIOC_INFO
Get system wide AGP or IGD hardware information. This command can be called by any
process from user or kernel context.
The argument is a pointer to agp_info_t structure.

typedef struct _agp_info {

agp_version_t agpi_version; /* OUT: AGP version supported */
uint32_t agpi_devid; /* OUT: bridge vendor + device */
uint32_t agpi_mode; /* OUT: mode of bridge */
ulong_t agpi_aperbase; /* OUT: base of aperture */
size_t agpi_apersize; /* OUT: aperture size in MB */
uint32_t agpi_pgtotal; /* OUT: max aperture pages avail. */
uint32_t agpi_pgsystem; /* OUT: same as pg_total */
uint32_t agpi_pgused; /* OUT: no. of currently used pages */
} agp_info_t;

agpi_version The version of AGP protocol the bridge device is

compatible with, for example, major 3 and minor 0
means AGP version 3.0.

typedef struct _agp_version {

uint16_t agpv_major;
uint16_t agpv_minor;
} agp_version_t;

agpi_devid AGP bridge vendor and device ID.

agpi_mode Current AGP mode, read from AGP status register of
target device. The main bits are defined as below.
/* AGP status register bits definition */

#define AGPSTAT_RQ_MASK 0xff000000

#define AGPSTAT_SBA (0x1 << 9)
#define AGPSTAT_OVER4G (0x1 << 5)
#define AGPSTAT_FW (0x1 << 4)
#define AGPSTAT_RATE_MASK 0x7

Device and Network Interfaces 33

agpgart_io(7I)

/* AGP 3.0 only bits */

#define AGPSTAT_ARQSZ_MASK (0x7 << 13)
#define AGPSTAT_CAL_MASK (0x7 << 10)
#define AGPSTAT_GART64B (0x1 << 7)
#define AGPSTAT_MODE3 (0x1 << 3)
/* rate for 2.0 mode */
#define AGP2_RATE_1X 0x1
#define AGP2_RATE_2X 0x2
#define AGP2_RATE_4X 0x4
/* rate for 3.0 mode */
#define AGP3_RATE_4X 0x1
#define AGP3_RATE_8X 0x2

agpi_aperbase The base address of aperture in PCI memory space.

agpi_apersize The size of the aperture in megabytes.
agpi_pgtotal Represents the maximum memory
pages the system can allocate
according to aperture size and
system memory size (which may differ
from the maximum locked memory a process
can have. The latter is subject
to the memory resource limit imposed
by the resource_controls(5) for each
project(4)):

project.max-device-locked-memory

This value can be modified through system

utilities like prctl(1).

agpi_pgsystem Same as pg_total.

agpi_pgused System pages already allocated by the driver.

Return Values:

EFAULT Argument copy out error

EINVAL Command invalid
0 Success
AGPIOC_ACQUIRE
Acquire control of GART. With the exception of AGPIOC_INFO, a process must acquire
GART before can it call other agpgart ioctl commands. Additionally, only processes with
GRAPHICS_ACCESS privilege may access this ioctl. In the current agpgart
implementation, GART access is exclusive, meaning that only one process can perform
GART operations at a time. To release control over GART, call AGPIOC_RELEASE. This
command can be called from user or kernel context.

The argument should be NULL.

34 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2008
 agpgart_io(7I)

Return values:
EBUSY GART has been acquired
0 Success.
AGPIOC_RELEASE
Release GART control. If a process releases GART control, it cannot perform additional
GART operations until GART is reacquired. Note that this command does not free
allocated memory or clear GART entries. (All clear jobs are done by direct calls or by
closing the device). When a process exits without making this ioctl, the final close(2)
performs this automatically. This command can be called from user or kernel context.
The argument should be NULL.
Return values:
EPERM Not owner of GART.
0 Success.
AGPIOC_SETUP
Setup AGPCMD register. An AGPCMD register resides in both the AGP master and target
devices. The AGPCMD register controls the working mode of the AGP master and target
devices. Each device must be configured using the same mode. This command can be
called from user or kernel context.
The argument is a pointer to agp_setup_t structure:

typedef struct _agp_setup {

uint32_t agps_mode; /* IN: value to be set for AGPCMD */
} agp_setup_t;

agps_mode Specifying the mode to be set. Each bit of the value may have
a specific meaning, please refer to AGP 2.0/3.0 specification
or hardware datasheets for details.

/* AGP command register bits definition */

#define AGPCMD_RQ_MASK 0xff000000
#define AGPCMD_SBAEN (0x1 << 9)
#define AGPCMD_AGPEN (0x1 << 8)
#define AGPCMD_OVER4GEN (0x1 << 5)
#define AGPCMD_FWEN (0x1 << 4)
#define AGPCMD_RATE_MASK 0x7
/* AGP 3.0 only bits */
#define AGP3_CMD_ARQSZ_MASK (0x7 << 13)
#define AGP3_CMD_CAL_MASK (0x7 << 10)
#define AGP3_CMD_GART64BEN (0x1 << 7)
The final values set to the AGPCMD register of the master/target devices are decided by the
agps_mode value and AGPSTAT of the master and target devices.

Device and Network Interfaces 35

agpgart_io(7I)

Return Values:
EPERM Not owner of GART.
EFAULT Argument copy in error.
EINVAL Command invalid for non-AGP system.
EIO Hardware setup error.
0 Success.
AGPIOC_ALLOCATE
Allocate system memory for graphics device. This command returns a unique ID which
can be used in subsequent operations to represent the allocated memory. The memory is
made up of discontiguous physical pages. In rare cases, special memory types may be
required. The allocated memory must be bound to the GART table before it can be used by
graphics device. Graphics applications can also mmap(2) the memory to userland for data
storing. Memory should be freed when it is no longer used by calling
AGPIOC_DEALLOCATE or simply by closing the device. This command can be called
from user or kernel context.
The argument is a pointer to agp_allocate_t structure.

typedef struct _agp_allocate {

int32_t agpa_key; /* OUT:ID of allocated memory */
uint32_t agpa_pgcount;/* IN: no. of pages to be allocated */
uint32_t agpa_type;/* IN: type of memory to be allocated */
uint32_t agpa_physical; /* OUT: reserved */
} agp_allocate_t;
agpa_key Unique ID of the allocated memory.
agpa_pgcount Number of pages to be allocated. The driver currently
supports only 4K pages. The value cannot exceed the
agpi_pgtotal value returned by AGPIOC_INFO ioct and is
subject to the limit of project.max-device-locked-memory.
If the memory needed is larger than the resource limit but
not larger than agpi_pgtotal, use prctl(1) or other system
utilities to change the default value of memory resource
limit beforehand.
agpa_type Type of memory to be allocated. The valid value of
agpa_type should be AGP_NORMAL. It is defined as:
#define AGP_NORMAL 0
Above, AGP_NORMAL represents the discontiguous
non-cachable physical memory which doesn't consume

36 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2008
 agpgart_io(7I)

kernel virtual space but can be mapped to user space by

mmap(2). This command may support more type values in
the future.
agpa_physical Reserved for special uses. In normal operations, the value is
undefined.

Return Values:
EPERM Not owner of GART.
EINVAL Argument not valid.
EFAULT Argument copy in/out error.
ENOMEM Memory allocation error.
0 Success.
AGPIOC_DEALLOCATE Deallocate the memory identified by a key assigned in a
previous allocation. If the memory isn't unbound from
GART, this command unbinds it automatically. The
memory should no longer be used and those still in
mapping to userland cannot be deallocated. Always call
AGPIOC_DEALLOCATE explicitly (instead of
deallocating implicitly by closing the device), as the system
won't carry out the job until the last reference to the device
file is dropped. This command from user or kernel context.

The input argument is a key of type int32_t, no output

argument.

Return Values:
EPERM Not owner of GART.
EINVAL Key not valid or memory in use.
0 Success.
AGPIOC_BIND Bind allocated memory. This command binds the allocated
memory identified by a key to a specific offset of the GART
table, which enables GART to translate the aperture range
at the offset to system memory. Each GART entry
represents one physical page. If the GART range is
previously bound to other system memory, it returns an
error. Once the memory is bound, it cannot be bound to
other offsets unless it is unbound. To unbind the memory,
call AGPIOC_UNBIND or deallocate the memory. This
command can be called from user or kernel context.

Device and Network Interfaces 37

agpgart_io(7I)

The argument is a pointer to agp_bind_t structure:

typedef struct _agp_bind {

int32_t agpb_key; /* IN: ID of memory to be bound */
uint32_t agpb_pgstart; /* IN: offset in aperture */
} agp_bind_t;
agpb_key The unique ID of the memory to be
bound, which is previously allocated
by calling AGPIOC_ALLOCATE.
agpb_pgstart The starting page offset to be bound in
aperture space.

Return Values:
EPERM Not owner of GART.
EFAULT Argument copy in error.
EINVAL Argument not valid.
EIO Binding to the GTT table of IGD
devices failed.
0 Success.
AGPIOC_UNBIND Unbind memory identified by a key from the GART. This
command clears the corresponding entries in the GART
table. Only the memory not in mapping to userland is
allowed to be unbound.

This ioctl command can be called from user or kernel

context.
The argument is a pointer to agp_unbind_t structure.

typedef struct _agp_unbind {

int32_t agpu_key; /* IN: key of memory to be unbound*/
uint32_t agpu_pri; /* Not used: for compat. with Xorg */
} agp_unbind_t;
agpu_key Unique ID of the memory to be
unbound which was previously bound
by calling AGPIOC_BIND.
agpu_pri Reserved for compatibility with
X.org/XFree86, not used.

Return Values:

38 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2008
 agpgart_io(7I)

EPERM Not owner of GART.

EFAULT Argument copy in error.
EINVAL Argument not valid or memory in use.
EIO Unbinding from the GTT table of IGD
devices failed.
0 Success

Example Below is an sample program showing how agpgart ioctls can be used:
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h
#include <sys/ioccom.h>
#include <sys/types.h>
#include <fcntl.h>
#include <errno.h>
#include <sys/mman.h>
#include <sys/agpgart.h>

#define AGP_PAGE_SIZE 4096

int main(int argc, char *argv[])

{
int fd, ret;
agp_allocate_t alloc;
agp_bind_t bindinfo;
agp_info_t agpinfo;
agp_setup_t modesetup;
int *p = NULL;
off_t mapoff;
size_t maplen;

if((fd = open(AGP_DEVICE, O_RDWR))== -1) {

printf("open AGP_DEVICE error with %d\n", errno);\
exit(-1);
}
printf("device opened\n");

ret = ioctl(fd, AGPIOC_INFO, &agpinfo);

if(ret == -1) {
printf("Get info error %d\n", errno);
exit(-1);
}
printf("AGPSTAT is %x\n", agpinfo.agpi_mode);
printf("APBASE is %x\n", agpinfo.agpi_aperbase);

Device and Network Interfaces 39

agpgart_io(7I)

printf("APSIZE is %dMB\n", agpinfo.agpi_apersize);

printf("pg_total is %d\n", agpinfo.agpi_pgtotal);

ret = ioctl(fd, AGPIOC_ACQUIRE);

if(ret == -1) {
printf(" Acquire GART error %d\n", errno);
exit(-1);
}

modesetup.agps_mode = agpinfo.agpi_mode;
ret = ioctl(fd, AGPIOC_SETUP, &modesetup);
if(ret == -1) {
printf("set up AGP mode error\n", errno);
exit(-1);
}

printf("Please input the number of pages you want to allocate\n");

scanf("%d", &alloc.agpa_pgcount);
alloc.agpa_type = AGP_NORMAL;
ret = ioctl(fd, AGPIOC_ALLOCATE, &alloc);
if(ret == -1) {
printf("Allocate memory error %d\n", errno);
exit(-1);
}

printf("Please input the aperture page offset to bind\n");

scanf("%d", &bindinfo.agpb_pgstart);
bindinfo.agpb_key = alloc.agpa_key;
ret = ioctl(fd, AGPIOC_BIND, &bindinfo);
if(ret == -1) {
printf("Bind error %d\n", errno);
exit(-1);
}
printf("Bind successful\n");

/*
* Now gart aperture space from (bindinfo.agpb_pgstart) to
* (bindinfo.agpb_pgstart + alloc.agpa_pgcount) can be used for
* AGP graphics transactions
*/
...

/*
* mmap can allow user processes to store graphics data
* to the aperture space
*/
maplen = alloc.agpa_pgcount * AGP_PAGE_SIZE;

40 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2008
 agpgart_io(7I)

mapoff = bindinfo.agpb_pgstart * AGP_PAGE_SIZE;

p = (int *)mmap((caddr_t)0, maplen, (PROT_READ | PROT_WRITE),
MAP_SHARED, fd, mapoff);
if (p == MAP_FAILED) {
printf("Mmap error %d\n", errno);
exit(-1);
}
printf("Mmap successful\n");
...

/*
* When user processes finish access to the aperture space,
* unmap the memory range
*/
munmap((void *)p, maplen);
...

/*
* After finishing AGP transactions, the resources can be freed
* step by step or simply by close device.
*/
ret = ioctl(fd, AGPIOC_DEALLOCATE, alloc.agpa_key);
if(ret == -1) {
printf(" Deallocate memory error %d\n", errno);
exit(-1);
}

ret = ioctl(fd, AGPIOC_RELEASE);

if(ret == -1) {
printf(" Release GART error %d\n", errno);
exit(-1);
}

close(fd);
}

Files /dev/agpgart
Symbolic link to the pseudo agpgart device.
/platform/i86pc/kernel/drv/agpgart
agpgart pseudo driver.
/platform/i86pc/kernel/drv/agpgart.conf
Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

Device and Network Interfaces 41

agpgart_io(7I)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture X86

Availability SUNWagp, SUNWagph

Stability level Unstable

See Also prctl(1), kstat(1M), close(2), ioctl(2), open(2), mmap(2), project(4), privileges(5),
attributes(5), resource_controls(5)

42 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2008
 ahci(7D)

Name ahci – Advanced Host Controller Interface SATA controller driver

Synopsis sata@unit-address

Description The ahci driver is a SATA framework-compliant HBA driver that supports SATA HBA
controllers that are compatible with the Advanced Host Controller Interface 1.0 specification.
AHCI is an Intel-developed protocol that describes the register-level interface for host
controllers for serial ATA 1.0a and Serial ATA II. The AHCI 1.0 specification describes the
interface between the system software and the host controller hardware.

The ahci driver currently supports the Intel ICH6/7/8/9/10, VIA vt8251 and JMicron AHCI
controllers which are compliant with the Advanced Host Controller Interface 1.0
specification. The Intel ICH6/7/8/9 and VIA vt8251 controllers support standard SATA
features. The ahci driver currently supports hard disk, ATAPI DVD, ATAPI tape, ATAPI disk
(i.e. Dell RD1000), hotplug, NCQ (Native command queuing) and Port multipliers (Silicon
Image 3726/4726). Power management is not yet supported.

Configuration The ahci module contains no user configurable parameters.

Files /kernel/drv/ahci
32–bit ELF kernel module (x86).
/kernel/drv/amd64/ahci
64–bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWahci

See Also cfgadm(1M), cfgadm_sata(1M), prtconf(1M), sata(7D)

Advanced Host Controller Interface 1.0

Writing Device Drivers

Notes To bind the ahci driver to your controller, choose the [AHCI] BIOS option.

Note that booting is not supported if toggle exists between the [IDE] and [AHCI] BIOS
options

Device and Network Interfaces 43

amd8111s(7D)

Name amd8111s – AMD-8111 Fast Ethernet Network Adapter driver

Synopsis /dev/amd8111s
Description The amd8111s Fast Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on the AMD-8111
Fast Ethernet Network Adapter.
The amd8111s driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting.
Application The cloning, character-special device /dev/amd8111 is used to access all AMD-8111 Fast
Programming Ethernet devices installed within the system.
Interface
The amd8111s driver is managed by the dladm(1M) command line utility, which allows
VLANs to be defined on top of amd8111s instances and for amd8111s instances to be
aggregated. See dladm(1M) for more details.
Configuration By default, the amd8111s driver performs auto-negotiation to select the link speed and mode.
Link speed and mode can be any of the following:
100 Mbps, full-duplex.
100 Mbps, half-duplex.
10 Mbps, full-duplex.
10 Mbps, half-duplex.
Files /dev/amd8111s* Special character device.
/kernel/drv/amd8111s* 32-bit ELF kernel module (x86).
/kernel/drv/amd64/amd8111s* 64-bit ELF Kernel module (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNW8111s

Architecture x86

Interface Stability Committed

See Also dladm(1M), attributes(5), streamio(7I), dlpi(7P)

Writing Device Drivers
STREAMS Programming Guide
Network Interfaces Programmer's Guide

44 man pages section 7: Device and Network Interfaces • Last Revised 30 Jul 2007
 amr(7D)

Name amr – SCSI HBA driver for Dell PERC 3/DC, 4/SC, 4/DC and 4/DI

Description The amr plain SCSI host bus adapter driver is a SCSA-compliant nexus driver that supports the
Dell PERC 3DC/4SC/4DC/4Di RAID devices.

The amr driver ports from FreeBSD and only supports basic RAID disk I/O functions.

Driver There are no user configurable parameters available. Please configure your hardware through
Configuration BIOS.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability x86

Files /kernel/drv/amr 32-bit ELF kernel module.

/kernel/drv/amd64/amr 64–bit kernel module (x86 only).
/kernel/drv/amr.conf Driver configuration file (contains no user-configurable
options).

See Also prtconf(1M), attributes(5), lsimega(7D), scsi_hba_attach_setup(9F),

scsi_sync_pkt(9F), scsi_transport(9F), scsi_inquiry(9S), scsi_device(9S),
scsi_pkt(9S)

Small Computer System Interface-2 (SCSI-2)

Device and Network Interfaces 45

arcmsr(7D)

Name arcmsr – SAS and SATA HBA driver for Areca Hardware Raid devices

Description The arcmsr host bus adapter driver is a SCSA-compliant nexus driver that supports Areca
Technology SAS and SATA RAID devices.

Supported SATA RAID cards:

ARC-1110 pci17d3,1110
ARC-1120 pci17d3,1120
ARC-1130 pci17d3,1130
ARC-1160 pci17d3,1160
ARC-1170 pci17d3,1170
ARC-1201 pci17d3,1201
ARC-1210 pci17d3,1210
ARC-1220 pci17d3,1220
ARC-1230 pci17d3,1230
ARC-1260 pci17d3,1260
ARC-1270 pci17d3,1270
ARC-1280 pci17d3,1280

Supported SAS RAID cards:

ARC-1380 pci17d3,1380
ARC-1381 pci17d3,1381
ARC-1680 pci17d3,1680
ARC-1681 pci17d3,1681

Configuration There are no user configurable parameters available. Please configure your hardware through
the host system BIOS.
Files /kernel/drv/arcmsr 32-bit ELF kernel module.
/kernel/drv/amd64/arcmsr 64-bit kernel module (x64 only).
/kernel/drv/arcmsr.conf Driver configuration file (contains no user-configurable
options).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86, x64 only

Availability SUNWarcmsr

See Also prtconf(1M), attributes(5), scsi_hba_attach_setup(9F), scsi_sync_pkt(9F),

scsi_transport(9F), scsi_inquiry(9S), scsi_device(9S), scsi_pkt(9S)

Small Computer System Interface-2 (SCSI-2)

46 man pages section 7: Device and Network Interfaces • Last Revised 28 Feb 2008
 arcmsr(7D)

http://www.areca.com.tw/products/main.htm

http://developer.intel.com/design/iio/index.htm —(Intel Corp IO processors provide the

underlying RAID engine for the supported devices).

Device and Network Interfaces 47

arn(7D)

Name arn – Atheros AR9280/9281/9285 IEEE802.11 a/b/g/n wireless network device

Description The arn IEEE802.11 a/b/g/n wireless driver is a loadable, clonable, GLDv3-based STREAMS
driver supporting Atheros AR9280/9281/9285 IEEE802.11 a/b/g/n wireless network device.

Configuration The arn driver performs auto-negotiation to determine the data rate and mode. The driver
supports only BSS networks (also known as ap or infrastructure networks) and open(or
open-system) or shared system authentication. For wireless security, WEP encryption,
WPA-PSK, and WPA2-PSK are currently supported. Configuration and administration tasks
can be performed with the dladm(1M) and wificonfig(1M) utilities.
Files /dev/arn Special character device
/kernel/drv/arn 32-bit ELF kernel module (x86)
/kernel/drv/amd64/arn 64-bit ELF kernel driver module (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWarn

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), dlpi(7P), gld(7D)

48 man pages section 7: Device and Network Interfaces • Last Revised 24 Aug 2009
 arp(7P)

Name arp, ARP – Address Resolution Protocol

Synopsis #include <sys/fcntl.h>
#include <sys/socket.h>
#include <net/if_arp.h>
#include <netinet/in.h>
s = socket(AF_INET, SOCK_DGRAM, 0);
d = open ("/dev/arp", oflag);

Description ARP is a protocol used to map dynamically between Internet Protocol (IP) and Ethernet
addresses. It is used by all Ethernet datalink providers (network drivers) and can be used by
other datalink providers that support broadcast, including FDDI and Token Ring. The only
network layer supported in this implementation is the Internet Protocol, although ARP is not
specific to that protocol.

ARP caches IP-to-link-layer address mappings. When an interface requests a mapping for an
address not in the cache, ARP queues the message that requires the mapping and broadcasts a
message on the associated network requesting the address mapping. If a response is provided,
ARP caches the new mapping and transmits any pending message. ARP will queue a
maximum of four packets while awaiting a response to a mapping request. ARP keeps only the
first four transmitted packets.

Application The STREAMS device /dev/arp is not a Transport Level Interface (TLI) transport provider
Programming and may not be used with the TLI interface.
Interface
To facilitate communications with systems that do not use ARP, ioctl() requests are provided
to enter and delete entries in the IP-to-link address tables. Ioctls that change the table contents
require sys_net_config privilege. See privileges(5).
#include <sys/sockio.h>
#include <sys/socket.h>
#include <net/if.h>
#include <net/if_arp.h>
struct arpreq arpreq;
ioctl(s, SIOCSARP, (caddr_t)&arpreq);
ioctl(s, SIOCGARP, (caddr_t)&arpreq);
ioctl(s, SIOCDARP, (caddr_t)&arpreq);

SIOCSARP, SIOCGARP and SIOCDARP are BSD compatible ioctls. These ioctls do not
communicate the mac address length between the user and the kernel (and thus only work for
6 byte wide Ethernet addresses). To manage the ARP cache for media that has different sized
mac addresses, use SIOCSXARP, SIOCGXARP and SIOCDXARP ioctls.
#include <sys/sockio.h>
#include <sys/socket.h>
#include <net/if.h>

Device and Network Interfaces 49

arp(7P)

#include <net/if_dl.h>
#include <net/if_arp.h>
struct xarpreq xarpreq;
ioctl(s, SIOCSXARP, (caddr_t)&xarpreq);
ioctl(s, SIOCGXARP, (caddr_t)&xarpreq);
ioctl(s, SIOCDXARP, (caddr_t)&xarpreq);

Each ioctl() request takes the same structure as an argument. SIOCS[X]ARP sets an ARP
entry, SIOCG[X]ARP gets an ARP entry, and SIOCD[X]ARP deletes an ARP entry. These ioctl()
requests may be applied to any Internet family socket descriptors, or to a descriptor for the
ARP device. Note that SIOCS[X]ARP and SIOCD[X]ARP require a privileged user, while
SIOCG[X]ARP
does not.
The arpreq structure contains
/*
* ARP ioctl request
*/
struct arpreq {
struct sockaddr arp_pa; /* protocol address */
struct sockaddr arp_ha; /* hardware address */
int arp_flags; /* flags */
};

The xarpreq structure contains:

/*
* Extended ARP ioctl request
*/
struct xarpreq {
struct sockaddr_storage xarp_pa; /* protocol address */
struct sockaddr_dl xarp_ha; /* hardware address */
int xarp_flags; /* arp_flags field values */
};
#define ATF_COM 0x2 /* completed entry (arp_ha valid) */
#define ATF_PERM 0x4 /* permanent (non-aging) entry */
#define ATF_PUBL 0x8 /* publish (respond for other host) */
#define ATF_USETRAILERS 0x10 /* send trailer pckts to host */
#define ATF_AUTHORITY 0x20 /* hardware address is authoritative */

The address family for the [x]arp_pa sockaddr must be AF_INET. The ATF_COM flag bits
([x]arp_flags) cannot be altered. ATF_USETRAILERS is not implemented on Solaris and is
retained for compatibility only. ATF_PERM makes the entry permanent (disables aging) if the
ioctl() request succeeds. ATF_PUBL specifies that the system should respond to ARP requests
for the indicated protocol address coming from other machines. This allows a host to act as an
ARP server, which may be useful in convincing an ARP-only machine to talk to a non-ARP
machine. ATF_AUTHORITY indicates that this machine owns the address. ARP does not update
the entry based on received packets.

50 man pages section 7: Device and Network Interfaces • Last Revised 5 Feb 2009
 arp(7P)

The address family for the arp_ha sockaddr must be AF_UNSPEC.

Before invoking any of the SIOC*XARP ioctls, user code must fill in the xarp_pa field with the
protocol (IP) address information, similar to the BSD variant. The SIOC*XARP ioctls come in
two (legal) varieties, depending on xarp_ha.sdl_nlen:
1. if sdl_nlen = 0, it behaves as an extended BSD ioctl. The kernel uses the IP address to
determine the network interface.
2. if (sdl_nlen > 0) and (sdl_nlen < LIFNAMSIZ), the kernel uses the interface name in
sdl_data[0] to determine the network interface; sdl_nlen represents the length of the
string (excluding terminating null character).
3. if (sdl_nlen >= LIFNAMSIZ), an error (EINVAL) is flagged from the ioctl.
Other than the above, the xarp_ha structure should be 0-filled except for SIOCSXARP, where
the sdl_alen field must be set to the size of hardware address length and the hardware address
itself must be placed in the LLADDR/sdl_data[] area. (EINVAL will be returned if user specified
sdl_alen does not match the address length of the identified interface).
On return from the kernel on a SIOCGXARP ioctl, the kernel fills in the name of the interface
(excluding terminating NULL) and its hardware address, one after another, in the
sdl_data/LLADDR area; if the two are larger than can be held in the 244 byte sdl_data[] area,
an ENOSPC error is returned. Assuming it fits, the kernel will also set sdl_alen with the length
of hardware address, sdl_nlen with the length of name of the interface (excluding
terminating NULL), sdl_type with an IFT_* value to indicate the type of the media, sdl_slen
with 0, sdl_family with AF_LINK and sdl_index (which if not 0) with system given index for
the interface. The information returned is very similar to that returned via routing sockets on
an RTM_IFINFO message.
The ARP ioctls have several additional restrictions and enhancements when used in
conjunction with IPMP:
■ ARP mappings for IPMP data and test addresses are managed by the kernel and cannot be
changed through ARP ioctls, though they may be retrieved using SIOCGARP or SIOCGXARP.
■ ARP mappings for a given IPMP group must be consistent across the group. As a result,
ARP mappings cannot be associated with individual underlying IP interfaces in an IPMP
group and must instead be associated with the corresponding IPMP IP interface.
■ roxy ARP mappings for an IPMP group are automatically managed by the kernel.
Specifically, if the hardware address in a SIOCSARP or SIOCSXARP request matches the
hardware address of an IP interface in an IPMP group and the IP address is not local to the
system, the kernel regards this as a IPMP Proxy ARP entry. This IPMP Proxy ARP entry
will have its hardware address automatically adjusted in order to keep the IP address
reachable (provided the IPMP group has not entirely failed).

—
—

Device and Network Interfaces 51

arp(7P)

—P

ARP performs duplicate address detection for local addresses. When a logical interface is
brought up (IFF_UP) or any time the hardware link goes up (IFF_RUNNING), ARP sends probes
(ar$spa == 0) for the assigned address. If a conflict is found, the interface is torn down. See
ifconfig(1M) for more details.

ARP watches for hosts impersonating the local host, that is, any host that responds to an ARP
request for the local host's address, and any address for which the local host is an authority.
ARP defends local addresses and logs those with ATF_AUTHORITY set, and can tear down local
addresses on an excess of conflicts.

ARP also handles UNARP messages received from other nodes. It does not generate these
messages.

Packet Events The arp driver registers itself with the netinfo interface. To gain access to these events, a
handle from net_protocol_lookup must be acquired by passing it the value NHF_ARP. Through
this interface, two packet events are supported:

Physical in - ARP packets received via a network inter face

Physical out - ARP packets to be sent out via a network interface

For ARP packets, the hook_pkt_event structure is filled out as follows:

hpe_ifp
Identifier indicating the inbound interface for packets received with the physical in
event.
hpe_ofp
Identifier indicating the outbound interface for packets received with the physical out
event.
hpe_hdr
Pointer to the start of the ARP header (not the ethernet header).
hpe_mp
Pointer to the start of the mblk_t chain containing the ARP packet.
hpe_mb
Pointer to the mblk_t with the ARP header in it.

Network Interface In addition to events describing packets as they move through the system, it is also possible to
Events receive notification of events relating to network interfaces. These events are all reported back
through the same callback. The list of events is as follows:
plumb
A new network interface has been instantiated.

52 man pages section 7: Device and Network Interfaces • Last Revised 5 Feb 2009
 arp(7P)

unplumb
A network interface is no longer associated with ARP.

See Also arp(1M), ifconfig(1M), privileges(5), if_tcp(7P), inet(7P), netinfo(9F)

Plummer, Dave, An Ethernet Address Resolution Protocol or Converting Network Protocol

Addresses to 48 .bit Ethernet Addresses for Transmission on Ethernet Hardware, RFC 826, STD
0037, November 1982.

Malkin, Gary, ARP Extension - UNARP, RFC 1868, November, 1995

Diagnostics Several messages can be written to the system logs (by the IP module) when errors occur. In
the following examples, the hardware address strings include colon (:) separated ASCII
representations of the link layer addresses, whose lengths depend on the underlying media
(for example, 6 bytes for Ethernet).
Node %x:%x ... %x:%x is using our IP address %d.%d.%d.%d on %s.
Duplicate IP address warning. ARP has discovered another host on a local network that
responds to mapping requests for the Internet address of this system, and has defended the
system against this node by re-announcing the ARP entry.
%s has duplicate address %d.%d.%d.%d (in use by %x:%x ... %x:%x); disabled.
Duplicate IP address detected while performing initial probing. The newly-configured
interface has been shut down.
%s has duplicate address %d.%d.%d.%d (claimed by %x:%x ... %x:%x); disabled.
Duplicate IP address detected on a running IP interface. The conflict cannot be resolved,
and the interface has been disabled to protect the network.
Recovered address %d.%d.%d.%d on %s.
An interface with a previously-conflicting IP address has been recovered automatically and
reenabled. The conflict has been resolved.
Proxy ARP problem? Node '%x:%x ... %x:%x' is using %d.%d.%d.%d on %s
This message appears if arp(1M) has been used to create a published permanent
(ATF_AUTHORITY) entry, and some other host on the local network responds to mapping
requests for the published ARP entry.

Device and Network Interfaces 53

asy(7D)

Name asy – asynchronous serial port driver

Synopsis #include <fcntl.h>
#include <sys/termios.h>
open("/dev/term/n", mode);
open("/dev/tty/n", mode);
open("/dev/cua/n", mode);

Description The asy module is a loadable STREAMS driver that provides basic support for Intel-8250,
National Semiconductor-16450, 16550, and some 16650 and 16750 and equivalent UARTs
connected via the ISA-bus, in addition to basic asynchronous communication support. The
asy module supports those termio(7I) device control functions specified by flags in the
c_cflag word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK, INPCK, IXON,
IXANY, or IXOFF flags in the c_iflag word of the termios structure. All other termio(7I)
functions must be performed by STREAMS modules pushed atop the driver. When a device is
opened, the ldterm(7M) and ttcompat(7M) STREAMS modules are automatically pushed on
top of the stream, providing the standard termio(7I) interface.

The character-special devices /dev/term/a, /dev/term/b, /dev/term/c and /dev/term/d are

used to access the four standard serial ports (COM1, COM2, COM3 and COM4 at I/O
addresses 3f8, 2f8, 3e8 and 2e8 respectively). Serial ports on non-standard ISA-bus I/O
addresses are accessed via the character-special devices /dev/term/0, /dev/term/1, etc.
Device names are typically used to provide a logical access point for a dial-in line that is used
with a modem.

To allow a single tty line to be connected to a modem and used for incoming and outgoing
calls, a special feature is available that is controlled by the minor device number. By accessing
character-special devices with names of the form /dev/cua/n, it is possible to open a port
without the Carrier Detect signal being asserted, either through hardware or an equivalent
software mechanism. These devices are commonly known as dial-out lines.

Note – This module is affected by the setting of certain eeprom variables, ttya-ignore-cd and
ttya-rts-dtr-off (and similarly for ttyb-, ttyc-, and ttyd- parameters). For information on these
parameters, see the eeprom(1M) man page.

Note – For serial ports on the standard COM1 to COM4 I/O addresses above, the default
setting for ttya-ignore-cd and ttya-rts-dtr-off is true. If any of these ports are connected to a
modem, these settings should be changed to false. For serial ports on non-standard I/O
addresses, the default setting for ttya-ignore-cd and ttya-rts-dtr-off is false.

Application Once a /dev/cua/n line is opened, the corresponding tty line cannot be opened until the
Programming /dev/cua/n line is closed. A blocking open will wait until the /dev/cua/n line is closed
Interface (which will drop Data Terminal Ready, after which Carrier Detect will usually drop as
well) and carrier is detected again. A non-blocking open will return an error. If the
/dev/ttydn line has been opened successfully (usually only when carrier is recognized on the

54 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 asy(7D)

modem), the corresponding /dev/cua/n line cannot be opened. This allows a modem to be
attached to /dev/term/[n] (renamed from /dev/tty[n]) and used for dial-in (by enabling the
line for login in /etc/inittab) or dial-out (by tip(1) or uucp(1C)) as /dev/cua/n when no
one is logged in on the line.
ioctls The standard set of termio ioctl() calls are supported by asy.
Breaks can be generated by the TCSBRK, TIOCSBRK, and TIOCCBRK ioctl() calls.
The input and output line speeds may be set to any speed that is supported by termio. The
speeds cannot be set independently; for example, when the output speed is set, the input speed
is automatically set to the same speed.
When the asy module is used to service the serial console port, it supports a BREAK condition
that allows the system to enter the debugger or the monitor. The BREAK condition is
generated by hardware and it is usually enabled by default.
A BREAK condition originating from erroneous electrical signals cannot be distinguished
from one deliberately sent by remote DCE. The Alternate Break sequence can be used as a
remedy against this. Due to a risk of incorrect sequence interpretation, SLIP and certain other
binary protocols should not be run over the serial console port when Alternate Break sequence
is in effect. Although PPP is a binary protocol, it is able to avoid these sequences using the
ACCM feature in RFC 1662. For Solaris PPP 4.0, you do this by adding the following line to
the /etc/ppp/options file (or other configuration files used for the connection; see pppd(1M)
for details):
asyncmap 0x00002000

By default, the Alternate Break sequence is a three character sequence: carriage return, tilde
and control-B (CR ~ CTRL-B), but may be changed by the driver. For more information on
breaking (entering the debugger or monitor), see kbd(1) and kb(7M).
Errors An open() will fail under the following conditions:
ENXIO The unit being opened does not exist.
EBUSY The dial-out device is being opened while the dial-in device is already open, or the
dial-in device is being opened with a no-delay open and the dial-out device is
already open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
EINTR The open was interrupted by the delivery of a signal.
Files /dev/term/[a-d]
/dev/term/[012...] dial-in tty lines
/dev/cua/[a-d]
/dev/cua/[012...] dial-out tty lines

Device and Network Interfaces 55

asy(7D)

/kernel/drv/amd64/asy 64-bit kernel module for 64–bit x86 platform

/kernel/drv/asy.conf asy configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also tip(1), kbd(1), uucp(1C), eeprom(1M), pppd(1M), ioctl(2), open(2), termios(3C),
attributes(5), ldterm(7M), ttcompat(7M), kb(7M), termio(7I)
Diagnostics asyn : silo overflow. The hardware overrun occurred before the input
character could be serviced.
asyn : ring buffer overflow. The driver's character input ring buffer overflowed
before it could be serviced.

56 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 ata(7D)

Name ata – AT attachment disk driver

Synopsis ide@unit-address

Description The ata driver supports disk and ATAPI CD/DVD devices conforming to the AT Attachment
specification including IDE interfaces. Support is provided for both parallel ATA (PATA) and
serial ATA (SATA) interfaces.

Refer to the Solaris x86 Hardware Compatibility List for a list of supported controllers.

Preconfigure A PCI IDE controller can operate in compatibility mode or in PCI-native mode. If more than
one controller is present in the system, only one can operate in compatibility mode.

If two PATA drives share the same controller, you must set one to master and the other to
slave. If both a PATA disk drive and a PATA CD-ROM drive utilize the same controller, you
can designate the disk drive as the master with the CD-ROM drive as the slave, although this is
not mandatory.

Supported Settings Supported settings for the primary controller when in compatibility mode are:
■ IRQ Level: 14
■ I/O Address: 0x1F0

Supported settings for the secondary controller when in compatibility mode are:
■ IRQ Level: 15
■ I/O Address: 0x170

Note – When in PCI-native mode, the IRQ and I/O address resources are configured by the
system BIOS.
Known Problems and ■ This driver does not support any RAID features present on a PATA/SATA controller. As a
Limitations result, you should configure BIOS to select IDE mode rather than RAID mode. Some
systems may require updating BIOS to allow switching modes.
■ On some systems, the SATA controller must have option ROM enabled or BIOS will not
consider SATA drives as bootable devices.
■ Panasonic LK-MC579B and the Mitsumi FX34005 IDE CD-ROM drives are not supported
and cannot be used to install the Solaris operating environment.
■ CMD-604 is unable to handle simultaneous I/O on both IDE interfaces. This defect causes
the Solaris software to hang if both interfaces are used. Use only the primary IDE interface
at address 0x1F0.
■ NEC CDR-260/CDR-260R/CDR-273 and Sony CDU-55E ATAPI CD-ROM drives might
fail during installation.
■ Sony CDU-701 CD-ROM drives must be upgraded to use firmware version 1.0r or later to
support booting from the CD.

Device and Network Interfaces 57

ata(7D)

A Compact Flash(CF) card can work as an ATA disk through a CF-to-ATA adapter. If both
card and adapter implement Compact Flash Version 2.0, DMA is supported. If either of
them does not, you should set ata-disk-dma-enabled to '0.'

Configuration The ata driver properties are usually set in ata.conf. However, it may be convenient, or in
some cases necessary, for you to set some of the DMA related properties as a system global
boot environment property. You set or modify properties in the boot environment
immediately prior to booting the Solaris kernel using the GRUB boot loader kernel boot
command line. You can also set boot environment properties using the eeprom(1M)
command or by editing the bootenv.rc configuration file. If a property is set in both the
driver's ata.conf file and the boot environment, the ata.conf property takes precedence.

Property modifications other than with the GRUB kernel boot command line are not effective
until you reboot the system. Property modifications via the GRUB kernel boot command line
do not persist across future boots.

Direct Memory Access is enabled for disks and atapi CD/DVD by default. If you want to
disable DMA when booting from a CD/DVD, you must first set atapi-cd-dma-enabled to 0
using the GRUB kernel boot command line.
ata-dma-enabled This property is examined before the DMA properties
discussed below. If it is set to '0,' DMA is disabled for all
ATA/ATAPI devices, and no further property checks are
made. If this property is absent or is set to '1,' DMA status is
determined by further examining one of the other properties
listed below.
ata-disk-dma-enabled This property is examined only for ATA disk devices, and
only if ata-dma-enabled is not set to '0.'

If ata-disk-dma-enabled set to '0,' DMA is disabled for all

ATA disks in the system. If this property is absent or set to '1,'
DMA is enabled for all ATA disks and no further property
checks are made. If needed, this property should be created
by the administrator using the GRUB kernel boot command
line or the eeprom(1M) command.
atapi-cd-dma-enabled This property is examined only for ATAPI CD/DVD devices,
and only if ata-dma-enabled is not set to '0.'

If atapi-cd-dma-enabled is absent or set to '0,' DMA is

disabled for all ATAPI CD/DVD's. If set to '1,' DMA is
enabled and no further property checks are made.

The Solaris installation program creates this property in the

boot environment with a value of '1.' It can be changed with

58 man pages section 7: Device and Network Interfaces • Last Revised 18 Apr 2007
 ata(7D)

the GRUB kernel boot command line or eeprom(1M) as

shown in the Example section of this manpage.
atapi-other-dma-enabled This property is examined only for non-CD/DVD ATAPI
devices such as ATAPI tape drives, and only if
ata-dma-enabled is not set to '0.'

If atapi-other-dma-enabled is set to '0,' DMA is disabled for

all non-CD/DVD ATAPI devices. If this property is absent or
set to '1,' DMA is enabled and no further property checks are
made.

If needed, this property should be created by the

administrator using the GRUB kernel boot command line or
the eeprom(1M) command.
drive0_block_factor
drive1_block_factor ATA controllers support some amount of buffering
(blocking). The purpose is to interrupt the host when an
entire buffer full of data has been read or written instead of
using an interrupt for each sector. This reduces interrupt
overhead and significantly increases throughput. The driver
interrogates the controller to find the buffer size. Some
controllers hang when buffering is used, so the values in the
configuration file are used by the driver to reduce the effect of
buffering (blocking). The values presented may be chosen
from 0x1, 0x2, 0x4, 0x8 and 0x10.

The values as shipped are set to 0x1, and they can be tuned to
increase performance.

If your controller hangs when attempting to use higher block

factors, you may be unable to reboot the system. For x86
based systems, it is recommended that tuning be performed
using a duplicate of the /platform/i86pc/kernel directory
subtree. This ensures that a bootable kernel subtree exists in
the event of a failed test.
ata-revert-to-defaults
revert—<diskmodel> When rebooting or shutting down, the driver can set a feature
which allows the drive to return to the power-on settings
when the drive receives a software reset (SRST) sequence. If
this property is present and set to 1, the driver will set the
feature to revert to defaults during reset. Setting this property
to 1 may prevent some systems from soft-rebooting and

Device and Network Interfaces 59

ata(7D)

would require cycling the power to boot the system. If this

property is not present the system will not set the feature to
revert to defaults during reset.

To determine the string to substitute for <diskmodel>, boot

your system (you may have to press the reset button or
power-cycle) and then view /var/adm/messages. Look for
the string "IDE device at targ" or "ATAPI device at targ." The
next line will contain the word "model" followed by the
model number and a comma. Ignore all characters except
letters, digits, ".", "_", and "-". Change uppercase letters to
lower case. If the string revert-<diskmodel> is longer than 31
characters, use only the first 31 characters.

Examples EXAMPLE 1 Sample ata Configuration File

# for higher performance - set block factor to 16
drive0_block_factor=0x1 drive1_block_factor=0x1
max_transfer=0x100
flow_control="dmult" queue="qsort" disk="dadk" ;

EXAMPLE 2 Revert to defaults property

revert-st320420a=1;

Output of /var/adm/messages:
Aug 17 06:49:43 caesar ata:[ID 640982 kern.info] IDE device at targ 0,
lun 0 lastlun 0x0
Aug 17 06:49:43 caesar ata:[ID 521533 kern.info] model ST320420A, stat

EXAMPLE 3 Change DMA property using GRUB

To change a DMA property using the GRUB kernel boot command line:
1. Reset the system.
2. Press “e” to interrupt the timeout.
3. Select the kernel line.
4. Press “e.”
5. If there is no existing -B option:
Add: -B atapi-cd-dma-enabled=1
else...
Add: atapi-cd-dma-enabled=1 to the end of the current -B option. For example:-B
foo=bar,atapi-cd-dma-enabled=1.

60 man pages section 7: Device and Network Interfaces • Last Revised 18 Apr 2007
 ata(7D)

EXAMPLE 3 Change DMA property using GRUB (Continued)

6. Press Enter to commit the edited line to memory. (Does not write to the disk and is
non-persistent).
7. Press 'b' to boot the modified entry.

EXAMPLE 4 Change DMA Property with eeprom(1M)

To enable DMA for optical devices while the Solaris kernel is running with the eeprom(1M)
system command:
eeprom ’atapi-cd-dma-enabled=1’
Files /platform/i86pc/kernel/drv/ata Device driver.
/platform/i86pc/kernel/drv/ata.conf Configuration file.
/boot/solaris/bootenv.rc Boot environment variables file for Solaris x86.
eeprom(1M) can be used to modify properties
in this file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also eeprom(1M), attributes(5), grub(5)

INCITS T13 ATA/ATAPI-7 specifications

Device and Network Interfaces 61

atge(7D)

Name atge – Device driver for Atheros/Attansic Ethernet chipsets

Description The atge ethernet driver is GLD based supporting the Atheros/Attansic L1E Gigabit Ethernet
10/100/1000 Base (AR8121/AR8113) chipsets:
pciex1969,1026 Atheros/Attansic GigabitE 10/100/1000 Base (AR8121/AR8113)

The atge driver supports IEEE 802.3 auto-negotiation, flow control and VLAN tagging.

Configuration The default configuration is auto-negotiation with bi-directional flow control. The advertised
capabilities for auto-negotiation are based on the capabilities of the PHY.

You can set the capabilities advertised by the atge controlled device using dladm(1M). The
driver supports only those parameters which begin with en (enabled) in the parameters listed
by the command dladm(1M). Each of these boolean parameters determines if the device
advertises that mode of operation when the hardware supports it.
Files /dev/atge Special character device
/kernel/drv/atge 32-bit device drive (x86)
/kernel/drv/amd64/atge 64-bit device driver (x86)

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), ndd(1M), netstat(1M), driver.conf(4), attributes(5), ieee802.3(5),

dlpi(7P), streamio(7I)

Writing Device Drivers

Network Interface Guide

STREAMS Programmer's Guide

IEEE 802.3ae Specification, 2002

62 man pages section 7: Device and Network Interfaces • Last Revised 11 Sep 2009
 ath(7D)

Name ath – Atheros AR52xx 802.11b/g wireless NIC driver

Description The ath 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver for the Atheros AR52xx (AR5210/5211/5212) chipset-based wireless NIC.

The ath driver functions include controller initialization, wireless 802.11b/g infrastructure
network connection, WEP, frame transmit and receive, and promiscuous and multi-cast
support.

Driver The ath driver performs auto-negotiation to determine the data rates and mode. Supported
Configuration 802.11b data rates (Mbits/sec.) are 1, 2, 5.5 and 11. Supported 802.11g data rates (Mbits/sec.)
are 1, 2, 5.5, 11, 6, 9, 12, 18, 24, 36, 48 and 54.

The ath driver supports only BSS networks (also known as "ap" or "infrastructure" networks)
and the "open" (or "opensystem") authentication type. Only WEP encryption is currently
supported. Configuration and administration can be performed through the dladm(1M) or
wificonfig(1M) utilities.
Files /dev/ath* Special character device.
/kernel/drv/ath 32-bit ELF kernel module (x86).
/kernel/drv/amd64/ath 64-bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also dladm(1M), wificonfig(1M), gld(7D)

ANSI/IEEE Std 802.11– Standard for Wireless LAN Technology, 1999.

IEEE Std 802.11a– Standard for Wireless LAN Technology-Rev. A, 2003

IEEE Std 802.11b - Standard for Wireless LAN Technology-Rev.B, 2003

IEEE Std 802.11g— Standard for Wireless LAN Technology - Rev. G, 2003

Device and Network Interfaces 63

atu(7D)

Name atu – Atmel AT76C50x USB IEEE 802.11b Wireless Device Driver

Description The atu 802.11b wireless driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Atmel AT76C50x chipset-based wireless devices.

Configuration The atu driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5, and 11 Mbits/sec.

The atu driver supports only BSS networks (also known as ap or infrastructure networks).

open (or open-system) and shared key authentication modes are supported. Encryption
types WEP40 and WEP104 are supported.
Files /dev/atu* Special character device
/kernel/drv/atu 32-bit ELF kernel module (x86)
/kernel/drv/amd64/atu 64-bit ELF kernel module (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWatu

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), dlpi(7P), gld(7D)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification - IEEE, 2001

64 man pages section 7: Device and Network Interfaces • Last Revised 26 Mar 2009
 audio1575(7D)

Name audio1575 – Uli M1575 Super South Bridge audio digital controller interface

Description The audio1575 device uses the Uli M1575 AC97–compatible audio digital controller and an
AC-97 Codec to implement the audio device interface. This interface allows analog only
inputs and outputs.
Files /kernel/drv/sparcv9/audio1575 64–bit driver module
/kernel/drv/audio1575.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWaudd

Stability level Uncommitted

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Uli M1575 Super South Bridge Data Sheet Data Sheet— Uli USA Inc.

AD1981B AC '97 SoundMAX(R) Codec Data Sheet— Analog Devices Inc.

Diagnostics In addition to being logged, the following messages might appear on the system console:
play-interrupts too low
record-interrupts too low The interrupt rate specified in audio1575.conf is set too
low. It has been reset to the rate specified in the message.
Update audio1575.conf to a higher interrupt rate.
play-interrupts too high
record-interrupts too high The interrupt rate specified in audio1575.conf is set too
high. It has been reset to the rate specified in the message.
Update audio1575.conf to a lower interrupt rate.

Device and Network Interfaces 65

audio(7D)

Name audio – common audio framework

Description The audio driver provides common support routines for audio devices in Solaris.

The audio framework supports multiple personalities, allowing for devices to be accessed with
different programming interfaces.

The audio framework also provides a number of facilities, such as mixing of audio streams,
and data format and sample rate conversion.

Overview The audio framework provides a software mixing engine (audio mixer) for all audio devices,
allowing more than one process to play or record audio at the same time.

Multi-Stream Codecs
The audio mixer supports multi-stream Codecs. These devices have DSP engines that provide
sample rate conversion, hardware mixing, and other features. The use of such hardware
features is opaque to applications.

Backward Compatibility
It is not possible to disable the mixing function. Applications must not assume that they have
exclusive access to the audio device.

Audio Formats Digital audio data represents a quantized approximation of an analog audio signal waveform.
In the simplest case, these quantized numbers represent the amplitude of the input waveform
at particular sampling intervals. To achieve the best approximation of an input signal, the
highest possible sampling frequency and precision should be used. However, increased
accuracy comes at a cost of increased data storage requirements. For instance, one minute of
monaural audio recorded in u-Law format (pronounced mew-law) at 8 KHz requires nearly
0.5 megabytes of storage, while the standard Compact Disc audio format (stereo 16-bit linear
PCM data sampled at 44.1 KHz) requires approximately 10 megabytes per minute.

An audio data format is characterized in the audio driver by four parameters: sample Rate,
encoding, precision, and channels. Refer to the device-specific manual pages for a list of the
audio formats that each device supports. In addition to the formats that the audio device
supports directly, other formats provide higher data compression. Applications can convert
audio data to and from these formats when playing or recording.

Sample Rate
Sample rate is a number that represents the sampling frequency (in samples per second) of the
audio data.

The audio mixer always configures the hardware for the highest possible sample rate for both
play and record. This ensures that none of the audio streams require compute-intensive low
pass filtering. The result is that high sample rate audio streams are not degraded by filtering.

66 man pages section 7: Device and Network Interfaces • Last Revised 3 Aug 2009
 audio(7D)

Sample rate conversion can be a compute-intensive operation, dependingon the number of

channels and a device's sample rate. For example, an 8KHz signal can be easily converted to
48KHz, requiring a low cost up sampling by 6. However, converting from 44.1KHz to 48KHz
is computer intensive because it must be up sampled by 160 and then down sampled by 147.
This is only done using integer multipliers.

Applications can greatly reduce the impact of sample rate conversion by carefully picking the
sample rate. Applications should always use the highest sample rate the device supports. An
application can also do its own sample rate conversion (to take advantage of floating point and
accelerated instructions) or use small integers for up and down sampling.

All modern audio devices run at 48 kHz or a multiple thereof, hence just using 48 kHz can be a
reasonable compromise if the application is not prepared to select higher sample rates.

Encodings
An encoding parameter specifies the audiodata representation. u-Law encoding corresponds
to CCITT G.711, and is the standard for voice data used by telephone companies in the United
States, Canada, and Japan. A-Law encoding is also part of CCITT G.711 and is the standard
encoding for telephony elsewhere in the world. A-Law and u-Law audio data are sampled at a
rate of 8000 samples per second with 12-bit precision, with the data compressed to 8-bit
samples. The resulting audio data quality is equivalent to that of stan dard analog telephone
service.

Linear Pulse Code Modulation (PCM) is an uncompressed, signed audio format in which
sample values are directly proportional to audio signal voltages. Each sample is a 2's
complement number that represents a positive or negative amplitude.

Precision
Precision indicates the number of bits used to store each audio sample. For instance, u-Law
and A-Law data are stored with 8-bit precision. PCM data can be stored at various precisions,
though 16-bit is the most common.

Channels
Multiple channels of audio can be interleaved at sample boundaries. A sample frame consists
of a single sample from each active channel. For example, a sample frame of stereo 16-bit PCM
data consists of 2 16-bit samples, corresponding to the left and right channel data. The audio
mixer sets the hardware to the maximum number of channels supported. If a mono signal is
played or recorded, it is mixed on the first two (usually the left and right) channel only. Silence
is mixed on the remaining channels.

Supported Formats
The audio mixer supports the following audio formats:

Device and Network Interfaces 67

audio(7D)

Encoding Precision Channels

Signed Linear PCM 32-bit Mono or Stereo
Signed Linear PCM 16-bit Mono or Stereo
Signed Linear PCM 8-bit Mono or Stereo
u-Law 8-bit Mono or Stereo
A-Law 8-bit Mono or Stereo

The audio mixer converts all audio streams to 24-bit Linear PCM before mixing. After mixing,
conversion is made to the best possible Codec format. The conversion process is not compute
intensive and audio applications can choose the encoding format that best meets their needs.

The mixer discards the low order 8 bits of 32-bit Signed Linear PCM in order to perform
mixing. (This is done to allow for possible overflows to fit into 32-bits when mixing multiple
streams together.) Hence, the maximum effective precision is 24-bits.
Files /kernel/drv/audio 32-bit kernel driver module
/kernel/drv/amd64/audio 64-bit x86 kernel driver module
/kernel/drv/sparcv9/audio 64-bit SPARC kernel driver module
/kernel/drv/audio.conf audio configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWaudd

Interface Stability Uncommitted

See Also ioctl(2), attributes(5), audio(7I), dsp(7I)

68 man pages section 7: Device and Network Interfaces • Last Revised 3 Aug 2009
 audio(7I)

Name audio – generic audio device interface

Synopsis #include <sys/audio.h>

Overview An audio device is used to play and/or record a stream of audio data. Since a specific audio
device may not support all functionality described below, refer to the device-specific manual
pages for a complete description of each hardware device. An application can use the
AUDIO_GETDEV ioctl(2) to determine the current audio hardware associated with
/dev/audio.

The audio framework provides a software mixing engine (audio mixer) for all audio devices,
allowing more than one process to play or record audio at the same time.

Backward It is no longer possible to disable the mixing function. Applications must not assume that they
Compatibility have exclusive access to the audio device.

Multi-Stream Codecs The audio mixer supports multi-stream Codecs. These devices have DSP engines that provide
sample rate conversion, hardware mixing, and other features. The use of such hardware
features is opaque to applications.

Audio Formats Digital audio data represents a quantized approximation of an analog audio signal waveform.
In the simplest case, these quantized numbers represent the amplitude of the input waveform
at particular sampling intervals. To achieve the best approximation of an input signal, the
highest possible sampling frequency and precision should be used. However, increased
accuracy comes at a cost of increased data storage requirements. For instance, one minute of
monaural audio recorded in μ-Law format (pronounced mew-law) at 8 KHz requires nearly
0.5 megabytes of storage, while the standard Compact Disc audio format (stereo 16-bit linear
PCM data sampled at 44.1 KHz) requires approximately 10 megabytes per minute.

Audio data may be represented in several different formats. An audio device's current audio
data format can be determined by using the AUDIO_GETINFO ioctl(2) described below.

An audio data format is characterized in the audio driver by four parameters: Sample Rate,
Encoding, Precision, and Channels. Refer to the device-specific manual pages for a list of the
audio formats that each device supports. In addition to the formats that the audio device
supports directly, other formats provide higher data compression. Applications may convert
audio data to and from these formats when playing or recording.

Sample Rate Sample rate is a number that represents the sampling frequency (in samples per second) of the
audio data.

The audio mixer always configures the hardware for the highest possible sample rate for both
play and record. This ensures that none of the audio streams require compute-intensive low
pass filtering. The result is that high sample rate audio streams are not degraded by filter ing.

Sample rate conversion can be a compute-intensive operation, depending on the number of

channels and a device's sample rate. For example, an 8KHz signal can be easily converted to

Device and Network Interfaces 69

audio(7I)

48KHz, requiring a low cost up sampling by 6. However, converting from 44.1KHz to 48KHz
is compute intensive because it must be up sampled by 160 and then down sampled by 147.
This is only done using integer multipliers.
Applications can greatly reduce the impact of sample rate conversion by carefully picking the
sample rate. Applications should always use the highest sample rate the device supports. An
application can also do its own sample rate conversion (to take advantage of floating point and
accelerated instruction or use small integers for up and down sampling.
All modern audio devices run at 48 kHz or a multiple thereof, hence just using 48 kHz may be
a reasonable compromise if the application is not prepared to select higher sample rates.
Encodings An encoding parameter specifies the audio data representation. μ-Law encoding corresponds
to CCITT G.711, and is the standard for voice data used by telephone companies in the United
States, Canada, and Japan. A-Law encoding is also part of CCITT G.711 and is the standard
encoding for telephony elsewhere in the world. A-Law and μ-Law audio data are sampled at a
rate of 8000 samples per second with 12-bit precision, with the data compressed to 8-bit
samples. The resulting audio data quality is equivalent to that of standard analog telephone
service.
Linear Pulse Code Modulation (PCM) is an uncompressed, signed audio format in which
sample values are directly proportional to audio signal voltages. Each sample is a 2's
complement number that represents a positive or negative amplitude.
Precision Precision indicates the number of bits used to store each audio sample. For instance, u-Law
and A-Law data are stored with 8-bit precision. PCM data may be stored at various precisions,
though 16-bit is the most common.
Channels Multiple channels of audio may be interleaved at sample boundaries. A sample frame consists
of a single sample from each active channel. For example, a sample frame of stereo 16-bit PCM
data consists of two 16-bit samples, corresponding to the left and right channel data.
The audio mixer sets the hardware to the maximum number of channels supported. If a mono
signal is played or recorded, it is mixed on the first two (usually the left and right) channels
only. Silence is mixed on the remaining channels
Supported Formats The audio mixer supports the following audio formats:

Encoding Precision Channels

Signed Linear PCM 32-bit Mono or Stereo
Signed Linear PCM 16-bit Mono or Stereo
Signed Linear PCM 8-bit Mono or Stereo
u-Law 8-bit Mono or Stereo
A-Law 8-bit Mono or Stereo

The audio mixer converts all audio streams to 24-bit Linear PCM before mixing. After mixing,
conversion is made to the best possible Codec format. The conversion process is not compute
intensive and audio applications can choose the encoding format that best meets their needs.

70 man pages section 7: Device and Network Interfaces • Last Revised 6 May 2009
 audio(7I)

Note that the mixer discards the low order 8 bits of 32-bit Signed Linear PCM in order to
perform mixing. (This is done to allow for possible overflows to fit into 32-bits when mixing
multiple streams together.) Hence, the maximum effective precision is 24-bits.
Description The device /dev/audio is a device driver that dispatches audio requests to the appropriate
underlying audio hardware. The audio driver is implemented as a STREAMS driver. In order
to record audio input, applications open(2) the /dev/audio device and read data from it using
the read(2) system call. Similarly, sound data is queued to the audio output port by using the
write(2) system call. Device configuration is performed using the ioctl(2) interface.
Because some systems may contain more than one audio device, application writers are
encouraged to query the AUDIODEV environment variable. If this variable is present in the
environment, its value should identify the path name of the default audio device.
Opening the Audio The audio device is not treated as an exclusive resource. Each process may open the audio
Device device once.

Each open() completes as long as there are channels available to be allocated. If no channels
are available to be allocated:
■ if either the O_NDELAY or O_NONBLOCK flags are set in the open() oflag argument, then –1 is
immediately returned, with errno set to EBUSY.
■ if neither the O_NDELAY nor the O_NONBLOCK flag are set, then open() hangs until the device
is available or a signal is delivered to the process, in which case a –1 is returned with errno
set to EINTR.
Upon the initial open() of the audio channel, the audio mixer sets the data format of the audio
channel to the default state of 8-bit, 8Khz, mono u-Law data. If the audio device does not
support this configuration, it informs the audio mixer of the initial configuration. Audio
applications should explicitly set the encoding characteristics to match the audio data
requirements, and not depend on the default configuration.
Recording Audio Data The read() system call copies data from the system's buffers to the application. Ordinarily,
read() blocks until the user buffer is filled. The I_NREAD ioctl (see streamio(7I)) may be
used to determine the amount of data that may be read without blocking. The device may
alternatively be set to a non-blocking mode, in which case read() completes immediately, but
may return fewer bytes than requested. Refer to the read(2) manual page for a complete
description of this behavior.
When the audio device is opened with read access, the device driver immediately starts
buffering audio input data. Since this consumes system resources, processes that do not record
audio data should open the device write-only (O_WRONLY).
The transfer of input data to STREAMS buffers may be paused (or resumed) by using the
AUDIO_SETINFO ioctl to set (or clear) the record.pause flag in the audio information structure
(see below). All unread input data in the STREAMS queue may be discarded by using the
I_FLUSH STREAMS ioctl. See streamio(7I). When changing record parameters, the input

Device and Network Interfaces 71

audio(7I)

stream should be paused and flushed before the change, and resumed afterward. Otherwise,
subsequent reads may return samples in the old format followed by samples in the new
format. This is particularly important when new parameters result in a changed sample size.

Input data can accumulate in STREAMS buffers very quickly. At a minimum, it will
accumulate at 8000 bytes per second for 8-bit, 8 KHz, mono, u-Law data. If the device is
configured for 16-bit linear or higher sample rates, it will accumulate even faster. If the
application that consumes the data cannot keep up with this data rate, the STREAMS queue
may become full. When this occurs, the record.error flag is set in the audio information
structure and input sampling ceases until there is room in the input queue for additional data.
In such cases, the input data stream contains a discontinuity. For this reason, audio recording
applications should open the audio device when they are prepared to begin reading data,
rather than at the start of extensive initialization.

Playing Audio Data The write() system call copies data from an application's buffer to the STREAMS output
queue. Ordinarily, write() blocks until the entire user buffer is transferred. The device may
alternatively be set to a non-blocking mode, in which case write() completes immediately,
but may have transferred fewer bytes than requested. See write(2).

Although write() returns when the data is successfully queued, the actual completion of
audio output may take considerably longer. The AUDIO_DRAIN ioctl may be issued to allow an
application to block until all of the queued output data has been played. Alternatively, a
process may request asynchronous notification of output completion by writing a zero-length
buffer (end-of-file record) to the output stream. When such a buffer has been processed, the
play.eof flag in the audio information structure is incremented.

The final close(2) of the file descriptor hangs until all of the audio output has drained. If a
signal interrupts the close(), or if the process exits without closing the device, any remaining
data queued for audio output is flushed and the device is closed immediately.

The consumption of output data may be paused (or resumed) by using the AUDIO_SETINFO
ioctl to set (or clear) the play.pause flag in the audio information structure. Queued output
data may be discarded by using the I_FLUSH STREAMS ioctl. (See streamio(7I)).

Output data is played from the STREAMS buffers at a default rate of at least 8000 bytes per
second for μ-Law, A-Law or 8–bit PCM data (faster for 16-bit linear data or higher sampling
rates). If the output queue becomes empty, the play.error flag is set in the audio information
structure and output is stopped until additional data is written. If an application attempts to
write a number of bytes that is not a multiple of the current sample frame size, an error is
generated and the bad data is thrown away. Additional writes are allowed.

Asynchronous I/O The I_SETSIG STREAMS ioctl enables asynchronous notification, through the SIGPOLL
signal, of input and output ready condition changes. The O_NONBLOCK flag may be set using the
F_SETFL fcntl(2) to enable non-blocking read() and write() requests. This is normally
sufficient for applications to maintain an audio stream in the background.

72 man pages section 7: Device and Network Interfaces • Last Revised 6 May 2009
 audio(7I)

Audio Control It is sometimes convenient to have an application, such as a volume control panel, modify
Pseudo-Device certain characteristics of the audio device while it is being used by an unrelated process.

The /dev/audioctl pseudo-device is provided for this purpose. Any number of processes
may open /dev/audioctl simultaneously. However, read() and write() system calls are
ignored by /dev/audioctl. The AUDIO_GETINFO and AUDIO_SETINFO ioctl commands may
be issued to /dev/audioctl to determine the status or alter the behavior of /dev/audio. Note:
In general, the audio control device name is constructed by appending the letters "ctl" to the
path name of the audio device.

Audio Status Change Applications that open the audio control pseudo-device may request asynchronous
Notification notification of changes in the state of the audio device by setting the S_MSG flag in an I_SETSIG
STREAMS ioctl. Such processes receive a SIGPOLL signal when any of the following events
occur:
■ An AUDIO_SETINFO ioctl has altered the device state.
■ An input overflow or output underflow has occurred.
■ An end-of-file record (zero-length buffer) has been processed on output.
■ An open() or close() of /dev/audio has altered the device state.
■ An external event (such as speakerbox's volume control) has altered the device state.

ioctls

Audio Information The state of the audio device may be polled or modified using the AUDIO_GETINFO and
Structure AUDIO_SETINFO ioctl commands. These commands operate on the audio_info structure as
defined, in <sys/audio.h>, as follows:
/*
* This structure contains state information for audio device
* IO streams
*/

struct audio_prinfo {
/*
* The following values describe the
* audio data encoding
*/
uint_t sample_rate; /* samples per second */
uint_t channels; /* number of interleaved channels */
uint_t precision; /* number of bits per sample */
uint_t encoding; /* data encoding method */

/*
* The following values control audio device
* configuration
*/

Device and Network Interfaces 73

audio(7I)

uint_t gain; /* volume level */

uint_t port; /* selected I/O port */
uint_t buffer_size; /* I/O buffer size */

/*
* The following values describe the current device
* state
*/
uint_t samples; /* number of samples converted */
uint_t eof; /* End Of File counter (play only) */
uchar_t pause; /* non-zero if paused, zero to resume */
uchar_t error; /* non-zero if overflow/underflow */
uchar_t waiting; /* non-zero if a process wants access */
uchar_t balance; /* stereo channel balance */
/*
* The following values are read-only device state
* information
*/
uchar_t open;/* non-zero if open access granted */
uchar_t active; /* non-zero if I/O active */
uint_t avail_ports; /* available I/O ports */
uint_t mod_ports; /* modifiable I/O ports */
};
typedef struct audio_prinfo audio_prinfo_t;

/*
* This structure is used in AUDIO_GETINFO and AUDIO_SETINFO ioctl
* commands
*/
struct audio_info {
audio_prinfo_t record;/* input status info */
audio_prinfo_t play;/* output status info */
uint_t monitor_gain; /* input to output mix */
uchar_toutput_muted; /* non-zero if output muted */
uint_t hw_features; /* supported H/W features */
uint_t sw_features;/* supported S/W features */
uint_t sw_features_enabled;
/* supported S/W features enabled */
};
typedef struct audio_info audio_info_t;

/* Audio encoding types */

#define AUDIO_ENCODING_ULAW (1) /* u-Law encoding */
#define AUDIO_ENCODING_ALAW (2) /* A-Law encoding */
#define AUDIO_ENCODING_LINEAR (3) /* Signed Linear PCM encoding */
/*
* These ranges apply to record, play, and
* monitor gain values

74 man pages section 7: Device and Network Interfaces • Last Revised 6 May 2009
 audio(7I)

*/
#define AUDIO_MIN_GAIN (0)/* minimum gain value */
#define AUDIO_MAX_GAIN (255) /* maximum gain value */

/*
* These values apply to the balance field to adjust channel
* gain values
*/
#define AUDIO_LEFT_BALANCE(0) /* left channel only */
#define AUDIO_MID_BALANCE (32) /* equal left/right balance */
#define AUDIO_RIGHT_BALANCE (64) /* right channel only */

/*
* Define some convenient audio port names
* (for port, avail_ports and mod_ports)
*/

/* output ports (several might be enabled at once) */

#define AUDIO_SPEAKER (0x01)/* built-in speaker */
#define AUDIO_HEADPHONE (0x02)/* headphone jack */
#define AUDIO_LINE_OUT (0x04)/* line out */
#define AUDIO_SPDIF_OUT (0x08)/* SPDIF port */
#define AUDIO_AUX1_OUT (0x10)/* aux1 out */
#define AUDIO_AUX2_OUT (0x20)/* aux2 out */

/* input ports (usually only one may be

* enabled at a time)
*/
#define AUDIO_MICROPHONE (0x01) /* microphone */
#define AUDIO_LINE_IN (0x02) /* line in */
#define AUDIO_CD(0x04) /* on-board CD inputs */
#define AUDIO_SPDIF_IN (0x08) /* SPDIF port */
#define AUDIO_AUX1_IN (0x10) /* aux1 in */
#define AUDIO_AUX2_IN (0x20) /* aux2 in */
#define AUDIO_CODEC_LOOPB_IN (0x40) /* Codec inter.loopback */

/* These defines are for hardware features */

#define AUDIO_HWFEATURE_DUPLEX (0x00000001u)
/*simult. play & cap. supported */

#define AUDIO_HWFEATURE_MSCODEC (0x00000002u)

/* multi-stream Codec */

/* These defines are for software features *

#define AUDIO_SWFEATURE_MIXER (0x00000001u)
/* audio mixer audio pers. mod. */

/*

Device and Network Interfaces 75

audio(7I)

* Parameter for the AUDIO_GETDEV ioctl

* to determine current audio devices
*/#define MAX_AUDIO_DEV_LEN(16)
struct audio_device {
char name[MAX_AUDIO_DEV_LEN];
char version[MAX_AUDIO_DEV_LEN];
char config[MAX_AUDIO_DEV_LEN];
};
typedef struct audio_device audio_device_t;

The play.gain and record.gain fields specify the output and input volume levels. A value of
AUDIO_MAX_GAIN indicates maximum volume. Audio output may also be temporarily muted
by setting a non-zero value in the output_muted field. Clearing this field restores audio output
to the normal state.

The monitor_gain field is present for compatibility, and is no longer supported. See dsp(7I) for
more detail.

Likewise, the play.port, play.ports, play.mod_ports, record.port, record.ports, and

record.mod_ports are no longer supported. See dsp(7I) for more detail.

The play.balance and record.balance fields are fixed to AUDIO_MID_BALANCE. Changes to

volume levels for different channels can be made using the interfaces in dsp(7I).

The play.pause and record.pause flags may be used to pause and resume the transfer of data
between the audio device and the STREAMS buffers. The play.error and record.error flags
indicate that data underflow or overflow has occurred. The play.active and record.active flags
indicate that data transfer is currently active in the corresponding direction.

The play.open and record.open flags indicate that the device is currently open with the
corresponding access permission. The play.waiting and record.waiting flags provide an
indication that a process may be waiting to access the device. These flags are set automatically
when a process blocks on open(), though they may also be set using the AUDIO_SETINFO ioctl
command. They are cleared only when a process relinquishes access by closing the device.

The play.samples and record.samples fields are zeroed at open() and are incremented each
time a data sample is copied to or from the associated STREAMS queue. Some audio drivers
may be limited to counting buffers of samples, instead of single samples for their samples
accounting. For this reason, applications should not assume that the samples fields contain a
perfectly accurate count. The play.eof field increments whenever a zero-length output buffer is
synchronously processed. Applications may use this field to detect the completion of
particular segments of audio output.

The record.buffer_size field controls the amount of input data that is buffered in the device
driver during record operations. Applications that have particular requirements for low
latency should set the value appropriately. Note however that smaller input buffer sizes may
result in higher system overhead. The value of this field is specified in bytes and drivers will

76 man pages section 7: Device and Network Interfaces • Last Revised 6 May 2009
 audio(7I)

constrain it to be a multiple of the current sample frame size. Some drivers may place other
requirements on the value of this field. Refer to the audio device-specific manual page for
more details. If an application changes the format of the audio device and does not modify the
record.buffer_size field, the device driver may use a default value to compensate for the new
data rate. Therefore, if an application is going to modify this field, it should modify it during or
after the format change itself, not before. When changing the record.buffer_size parameters,
the input stream should be paused and flushed before the change, and resumed afterward.
Otherwise, subsequent reads may return samples in the old format followed by samples in the
new format. This is particularly important when new parameters result in a changed sample
size. If you change the record.buffer_size for the first packet, this protocol must be followed or
the first buffer will be the default buffer size for the device, followed by packets of the requested
change size.

The record.buffer_size field may be modified only on the /dev/audio device by processes that
have it opened for reading.

The play.buffer_size field is currently not supported.

The audio data format is indicated by the sample_rate, channels, precision and encoding fields.
The values of these fields correspond to the descriptions in the AUDIO FORMATS section of this
man page. Refer to the audio device-specific manual pages for a list of supported data format
combinations.

The data format fields can be modified only on the /dev/audio device.

If the parameter changes requested by an AUDIO_SETINFO ioctl cannot all be accommodated,

ioctl() returns with errno set to EINVAL and no changes are made to the device state.

Streamio IOCTLS All of the streamio(7I) ioctl commands may be issued for the /dev/audio device. Because
the /dev/audioctl device has its own STREAMS queues, most of these commands neither
modify nor report the state of /dev/audio if issued for the /dev/audioctl device. The
I_SETSIG ioctl may be issued for /dev/audioctl to enable the notification of audio status
changes, as described above.

Audio IOCTLS The audio device additionally supports the following ioctl commands:

AUDIO_DRAIN The argument is ignored. This command suspends the calling process
until the output STREAMS queue is empty and all queued samples have
been played, or until a signal is delivered to the calling process. It may not
be issued for the /dev/audioctldevice. An implicit AUDIO_DRAIN is
performed on the final close() of /dev/audio.
AUDIO_GETDEV The argument is a pointer to an audio_device_t structure. This
command may be issued for either /dev/audio or /dev/audioctl. The
returned value in the name field will be a string that will identify the
current /dev/audio hardware device, the value in version will be a string
indicating the current version of the hardware, and config will be a

Device and Network Interfaces 77

audio(7I)

device-specific string identifying the properties of the audio stream

associated with that file descriptor. Refer to the audio device-specific
manual pages to determine the actual strings returned by the device
driver.
AUDIO_GETINFO The argument is a pointer to an audio_info_t structure. This command
may be issued for either /dev/audio or /dev/audioctl. The current state
of the /dev/audio device is returned in the structure.

Values return pertain to a logical view of the device as seen by and private
to the process, and do not necessarily reflect the actual hardware device
itself.
AUDIO_SETINFO The argument is a pointer to an audio_info_t structure. This command
may be issued for either the /dev/audio or the /dev/audioctl device
with some restrictions. This command configures the audio device
according to the supplied structure and overwrites the existing structure
with the new state of the device. Note: The play.samples, record.samples,
play.error, record.error, and play.eof fields are modified to reflect the state
of the device when the AUDIO_SETINFO is issued. This allows programs to
automatically modify these fields while retrieving the previous value.

As with AUDIO_SETINFO, the settings managed by this ioctl deal with a

logical view of the device which is private to the process, and don't
necessarily have any impact on the hardware device itself.

Certain fields in the audio information structure, such as the pause flags, are treated as
read-only when /dev/audio is not open with the corresponding access permission. Other
fields, such as the gain levels and encoding information, may have a restricted set of acceptable
values. Applications that attempt to modify such fields should check the returned values to be
sure that the corresponding change took effect. The sample_rate, channels, precision, and
encoding fields treated as read-only for /dev/audioctl, so that applications can be guaranteed
that the existing audio format will stay in place until they relinquish the audio device.
AUDIO_SETINFO will return EINVAL when the desired configuration is not possible, or EBUSY
when another process has control of the audio device.

All of the logical device state is reset when the corresponding I/O stream of /dev/audio is
closed.

The audio_info_t structure may be initialized through the use of the AUDIO_INITINFO
macro. This macro sets all fields in the structure to values that are ignored by the
AUDIO_SETINFO command. For instance, the following code switches the output port from the
built-in speaker to the headphone jack without modifying any other audio parameters:
audio_info_t info;
AUDIO_INITINFO();

78 man pages section 7: Device and Network Interfaces • Last Revised 6 May 2009
 audio(7I)

info.play.port = AUDIO_HEADPHONE;
err = ioctl(audio_fd, AUDIO_SETINFO, );

This technique eliminates problems associated with using a sequence of AUDIO_GETINFO

followed by AUDIO_SETINFO.

Errors An open() will fail if:

EBUSY The requested play or record access is busy and either the O_NDELAY or O_NONBLOCK
flag was set in the open() request.
EINTR The requested play or record access is busy and a signal interrupted the open()
request.

An ioctl() will fail if:

EINVAL The parameter changes requested in the AUDIO_SETINFO() ioctl are invalid or are
not supported by the device.

Files The physical audio device names are system dependent and are rarely used by programmers.
Programmers should use the following generic device names:
/dev/audio Symbolic link to the system's primary audio device
/dev/audioctl Symbolic link to the control device for /dev/audio
/dev/sound/0 First audio device in the system
/dev/sound/0ctl Audio control device for /dev/sound/0
/usr/share/audio/samples Audio files

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWcsu, SUNWaudd, SUNWaudh

Stability Level Obsolete Uncommitted

See Also close(2), fcntl(2), ioctl(2), open(2), poll(2), read(2), write(2), attributes(5), dsp(7I),
streamio(7I)

Bugs Due to a feature of the STREAMS implementation, programs that are terminated or exit
without closing the audio device may hang for a short period while audio output drains. In
general, programs that produce audio output should catch the SIGINT signal and flush the
output stream before exiting.

Device and Network Interfaces 79

audio810(7D)

Name audio810 – Intel ICH series, nVidia nForce series and AMD 8111 audio core support

Description The audio810 driver provides support for AC 97 audio controllers embedded in Intel ICH,
nVidia nForce, and AMD 8111 chips.
Files /kernel/drv/audio810 32-bit kernel driver module
/kernel/drv/amd64/audio810 64-bit x86 kernel driver module
/kernel/drv/audio810.conf audio810 driver configuration file

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PC-based systems

Availability SUNWad810

Interface Stability Uncommitted

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

AMD-8111 HyperTransport I/O Hub Data Sheet — Advanced Micro Devices Inc.

ALC655 Specification — Realtek Inc.

Notes Some laptops (including Sony VAIO, among others), have their on-board amplifier powered
down by default, meaning that audio is suppressed even if hardware and the audio810 driver
are working normally. To correct this, set the ac97-invert-amp=1 property in the
/kernel/drv/audio810.conf to power-up the amplifier.

Diagnostics In addition to being logged, the following messages may appear on the system console:
play-interrupts too low
record-interrupts too low The interrupt rate in audio810.conf is set too low. It has
been reset to the rate specified in the message. Update
audio810.conf to a higher interrupt rate.
play-interrupts too high
record-interrupts too high The interrupt rate set inaudio810.conf is set too high.It
has been reset to the rate specified in the message. Update
audio810.confto a lower interrupt rate.

80 man pages section 7: Device and Network Interfaces • Last Revised 13 May 2009
 audiocmi(7D)

Name audiocmi – C-Media 8738, 8768, and 8338 driver support

Description The audiocmi driver provides support for the C-Media 8738, 8768, and 8338 audio
controllers. These are found on some motherboards and some add-in PCI cards.
Files /kernel/drv/audiocmi 32-bit kernel driver module
/kernel/drv/amd64/audiocmi 64-bit x86 kernel driver module
/kernel/drv/sparcv9/audiocmi 64-bit SPARC kernel driver module
/kernel/drv/audiocmi.conf Driver configuration file

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PC-based system

Availability SUNWaudiocmi

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Device and Network Interfaces 81

audiocs(7D)

Name audiocs – Crystal Semiconductor 4231 Audio driver

Description The audiocs driver supports the Crystal Semiconductor 4231 Codec to implement the audio
device interface.

The audiocs device provides support for the internal speaker, headphone, line out, line in,
microphone, and, on some platforms, internal CD-ROM audio in.

Errors audiocs errors are described in the audio(7I) manual page.

Files /kernel/drv/sparcv9/audiocs 64-bit audiocs driver
/kernel/drv/audiocs.conf audiocs driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWaudd

Stability level Uncommitted

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Crystal Semiconductor, Inc. CS4231 Data Sheet

Diagnostics In addition to being logged, the following messages can appear on the system console:
play-interrupts too low
record-interrupts too low The interrupt rate specified in audiocs.conf is set too
low. It is being reset to the rate specified in the message.
Update audiocs.conf to a higher interrupt rate.
play-interrupts too high
record-interrupts too high The interrupt rate specified in audiocs.conf is set too
high. It is being reset to the rate specified in the message.
Update audiocs.conf to a lower interrupt rate.

82 man pages section 7: Device and Network Interfaces • Last Revised 12 May 2009
 audioemu10k(7D)

Name audioemu10k – Creative EMU10K audio device support

Description The audioemu10k driver provides support for the Creative EMU 10K1 and 10K2 family of
audio devices. These are typically marketed under the Audigy or Sound Blaster Live! brands.

This device driver is capable of 5.1 or 7.1 surround sound and SPDIF playback and record,
depending on the capabilities of the individual device.
Files /kernel/drv/audioemu10k 32-bit kernel driver module
/kernel/drv/amd64/audioemu10k 64-bit kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWaudioemu10k

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Device and Network Interfaces 83

audioens(7D)

Name audioens – Ensoniq ESS 1371 and ESS 1373 audio driver

Description The audioens driver provides support for the Ensoniq ESS1371, ESS1373, and Creative 5880
AC'97 devices. These devices are commonly known by several different names, including the
Sound Blaster PCI128 and AudioPCI '97.
Files /kernel/drv/audioens 32-bit kernel module
/kernel/drv/amd64/audioens 64-bit x86 kernel module
/kernel/drv/sparcv9/audioens 64-bit SPARC kernel module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWaudd

Interface Stability Uncommitted

See Also attributes(5), audio(7I), dsp(7I), mixer(7I)

84 man pages section 7: Device and Network Interfaces • Last Revised 15 May 2009
 audiohd(7D)

Name audiohd – Intel High Definition Audio Controller support

Description The audiohd driver provides support for the generic codec chips which are compatible with
the Intel High-Definition Audio Controller 1.0 specification.
Files /kernel/drv/audiohd.conf audiohd driver configuration file
/kernel/drv/audiohd 32-bit x86 kernel driver module
/kernel/drv/amd64/audiohd 64-bit x86 kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PC-based system

Availability SUNWaudiohd

Stability level Uncommitted

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Intel High-Definition Audio Specification 1.0. - Intel Corporation

ALC880 Specification — Realtek Inc.

Diagnostics In addition to being logged, the following messages can appear on the system console:
play-interrupts too low
record-interrupts too low The interrupt rate in audiohd.conf is set too low. It has
been reset to the rate specified in the message. Update
audiohd.conf to a higher interrupt rate.
redcord-interrupts too low
record-interrupts too high The interrupt rate in audiohd.conf is set too low. It has
been reset to the rate specified in the message. Update
audiohd.conf to a higher interrupt rate

Device and Network Interfaces 85

audioixp(7D)

Name audioixp – ATI IXP400 south bridge audio digital controller interface

Description The audioixp device uses the IXP400 south bridge audio digital controller and a AC–97
Codec to implement the audio device interface.

The audioixp device provides support for the internal speaker, headphone, line out, line in,
and microphone.
Files /kernel/drv/audioixp.conf Driver configuration file
/kernel/drv/audioixp 32-bit kernel driver module
/kernel/drv/amd64/audioixp 64-bit kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PC-based system

Availability SUNWadixp

Stability level Uncommitted

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

ATI IXP400 South Bridge Data Sheet

Diagnostics In addition to being logged, the following messages may appear on the system console:
play-interrupts too low
record-interrupts too low The interrupt rate in audioixp.conf is set too low. It has
been reset to the rate specified in the message. Update
audioixp.conf to a higher interrupt rate.
play-interrupts too high
record-interrupts too high The interrupt rate set in audioixp.conf is set too high. It
has been reset to the rate specified in the message. Update
audioixp.conf to a lower interrupt rate.

86 man pages section 7: Device and Network Interfaces • Last Revised 12 May 2009
 audiols(7D)

Name audiols – Creative Audigy LS audio device support

Description The audiols driver provides support for the Creative Audigy LS audio device.

There are numerous devices marketed under the Audigy brand by Creative, but only Audigy
LS devices are supported by this driver.

This device is capable of 5.1 surround sound.

Files /kernel/drv/audiols 32-bit kernel driver module
/kernel/drv/amd64/audiols 64-bit kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWaudiols

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Device and Network Interfaces 87

audiop16x(7D)

Name audiop16x – Creative Sound Blaster Live! OEM support

Description The audiop16x driver provides support for the Creative Sound Blaster Live! products based
on the P16X device. These chips are also known as the EMU10K1X device, not to be confused
with the EMU10K1.

Add-in boards known to work with this driver are Sound Blaster Live! cards with model
numbers SB0200 or SB0213.

This device is capable of 5.1 surround sound.

Files /kernel/drv/audiop16x 32-bit kernel driver module
/kernel/drv/amd64/audiop16x 64-bit x86 kernel driver module
/kernel/drv/sparcv9/audiop16x 64-bit SPARC kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86, SPARC

Availability SUNWaudiop16x

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

88 man pages section 7: Device and Network Interfaces • Last Revised 8 Sep 2009
 audiopci(7D)

Name audiopci – Ensoniq 1370 driver support

Description The audiopci driver provides support for the Ensoniq 1370 audio controller. Ensoniq 1370
chips are found on add-in PCI cards commonly identified as Audio PCI and SoundBlaster
PCI.
Files /kernel/drv/audiopci 32-bit kernel driver module
/kernel/drv/amd64/audiopci 64-bit x86 kernel driver module
/kernel/drv/amd64/audiopci 64-bit SPARC kernel driver module
/kernel/drv/audiopci.conf Driver configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWaudd

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Creative Technology Ltd ES1370 Specification

http://www.sun.com

Device and Network Interfaces 89

audiosolo(7D)

Name audiosolo – ESS Solo-1 audio device support

Description The audiosolo driver provides support for the ESS Solo-1 audio device. This device is found
on certain motherboards and discrete audio cards. It supports 16-bit 48 kHZ stereo playback
and capture.
Files /kernel/drv/audiosolo 32-bit kernel driver module
/kernel/drv/amd64/audiosolo 64-bit kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWaudiosolo

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

90 man pages section 7: Device and Network Interfaces • Last Revised 18 Sep 2009
 audiots(7D)

Name audiots – Acer Laboratories Inc. M5451 audio processor interface

Description The audiots device uses the ALI M5451 audio processor and an AC-97 Codec to implement
the audio device interface.

The audiots device provides support for the internal speaker, headphone, line out, line in,
and microphone.
Files /kernel/drv/sparcv9/audiots 64-bit audiots driver
/kernel/drv/audiots.conf audiots driver configuration file

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWaudd

Stability level Uncommitted

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Acer Laboratories Inc. M5451 PCI Audio Processor Technical Specification

Diagnostics In addition to being logged, the following messages can appear on the system console:
play-interrupts too low
record-interrupts too low The interrupt rate in audiots.conf is set too low. It has
been reset to the rate specified in the message. Update
audiots.conf to a higher interrupt rate.
play-interrupts too high
record-interrupts too high The interrupt rate set in audiots.conf is set too high. It
has been reset to the rate specified in the message. Update
audiots.conf to a lower interrupt rate.

Device and Network Interfaces 91

audiovia823x(7D)

Name audiovia823x – VIA VT8233, VT8235, and VT8237) support

Description The audiovia823x driver provides support for the VIA VT8233, VT8235, and VT8237 AC'97
devices found on motherboards with certain VIA chip sets.
Files /kernel/drv/audiovia823x 32-bit x86 kernel module
/kernel/drv/amd64/audiovia823x 64-bit x86 kernel module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWvia823x

Interface Stability Committed

See Also attributes(5), audio(7I), dsp(7I), mixer(7I)

92 man pages section 7: Device and Network Interfaces • Last Revised 12 May 2009
 audiovia97(7D)

Name audiovia97 – Via 82C686 audio device support

Description The audiovia97 driver provides support for the integrated audio device found in the Via
82C686 southbridge chipset, found on certain motherboards.
Files /kernel/drv/audiovia97 32-bit kernel driver module

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Via 82C686-based systems

Availability SUNWaudiovia97

See Also ioctl(2), attributes(5), audio(7I), mixer(7I), streamio(7I)

Device and Network Interfaces 93

av1394(7D)

Name av1394 – 1394 audio/video driver

Synopsis unit@GUID

Description The av1394 driver implements iec61883(7I) interfaces for IEEE 1394 compliant devices.

Asynchronous The driver allows applications to act as FCP controllers, but not FCP targets. Only
Transactions IEC61883_FCP_CMD requests can be sent with write(2). Only IEC61883_FCP_RESP requests
can be received with read(2).

Isochronous When the read/write method of is used for transmit, the driver is capable of auto-detecting
Transactions and transmitting SD-DVCR 525/60 and 625/50 streams. See iec61883(7I) for details.

Files /dev/av/N/async device node for asynchronous data

/dev/av/N/isoch device node for isochronous data
kernel/drv/sparcv9/av1394 64-bit ELF kernel module
kernel/drv/av1394 32-bit ELF kernel module
kernel/drv/amd64/av1394 64-bit ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture All

Interface Stability Committed

See Also read(2), write(2), attributes(5), hci1394(7D), iec61883(7I)

IEEE Std 1394-1995 Standard for a High Performance Serial Bus

IEC 61883 Consumer audio/video equipment - Digital interface

94 man pages section 7: Device and Network Interfaces • Last Revised 3 Apr 2009
 bbc_beep(7D)

Name bbc_beep – Platform-dependent Beep driver for BBC-based hardware.

Synopsis beep@unit-address

Description The bbc_beep driver generates beeps on platforms (including Sun Blade 1000) that use
BBC-based registers and USB keyboards. When the KIOCCMD ioctl is issued to the USB
keyboard module (see usbkbm(7M)) with command KBD_CMD_BELL/KBD_CMD_NOBELL,
usbkbm(7M) passes the request to the bbc_beep driver to turn the beep on and off,
respectively.
Files /platform/sun4u/kernel/drv/sparcv9/bbc_beep
64–bit ELF kernel driver

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture BBC-based SPARC

Availability SUNWcarx.u

See Also kbd(1), attributes(5), grbeep(7d), kb(7M), usbkbm(7M)

Writing Device Drivers

Diagnostics None

Device and Network Interfaces 95

bcm_sata(7D)

Name bcm_sata – Broadcom HT1000 SATA controller driver

Synopsis sata@unit-address

Description The bcm_sata driver is a SATA HBA driver that supports Broadcom HT1000 SATA HBA
controllers.

HT1000 SATA controllers are compliant with the Serial ATA 1.0 specification and SATA II
Phase 1.0 specification (the extension to the SATA 1.0 specification). These HT1000
controllers support standard SATA features including SATA-II disks, NCQ, hotplug, ATAPI
devices and port multiplier.

The driver does not currently support NCQ and port multiplier features.

Configuration The bcm_sata module contains no user configurable parameters.

Files /kernel/drv/bcm_sata 32-bit ELF kernel module (x86)

/kernel/drv/amd64/bcm_sata 64-bit ELF kernel module (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWbcmsata

See Also cfgadm(1M), cfgadm_sata(1M), prtconf(1M), attributes(5), sata(7D), sd(7D)

Writing Device Drivers

96 man pages section 7: Device and Network Interfaces • Last Revised 31 Aug 2009
 bfe(7D)

Name bfe – Device driver for Broadcom BCM4401 100Base-T NIC

Description The bfe Fast Ethernet driver is GLD-based and supports the Broadcom BCM4401 100Base-T
NIC adapters :pci14e4,170c Broadcom BCM4401 100Base-T..

The bfe driver supports IEEE 802.3 auto-negotiation, flow control and VLAN tagging.

Configuration The default configuration is auto-negotiation with bidirectional flow control. The advertised
capabilities for auto-negotiation are based on the capabilities of the PHY.

You can set the capabilities advertised by the bfe controlled device using dladm(1M). The
driver supports only those parameters which begin with en (enabled) in the parameters listed
by the command dladm(1M). Each of these boolean parameters determines if the device
advertises that mode of operation when the hardware supports it.
Files /dev/bfe Special character device
/kernel/drv/bfe 32-bit device driver (x86)
/kernel/drv/amd64/bfe 64-bit device driver (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), netstat(1M), driver.conf(4), attributes(5), ieee802.3(5), dlpi(7P),

streamio(7I)

Writing Device Drivers

STREAMS Programmer's Guide

Network Interface Guide

IEEE 802.3ae Specification - 2002

Device and Network Interfaces 97

bge(7D)

Name bge – SUNW,bge Gigabit Ethernet driver for Broadcom BCM57xx

Synopsis /dev/bge*
Description The bge Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on Broadcom
BCM57xx (BCM5700/5701/5703/5704/5705/5705M/5714/5721/5751/5751M/5782/5788 on
x86) Gigabit Ethernet controllers fitted to the system motherboard. With the exception of
BCM5700/BCM5701/BCM5704S, these devices incorporate both MAC and PHY functions
and provide three-speed (copper) Ethernet operation on the RJ-45 connectors.
(BCM5700/BCM5701/BCM5704S do not have a PHY integrated into the MAC chipset.)
The bge driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting.
The bge driver and hardware support auto-negotiation, a protocol specified by the 1000
Base-T standard. Auto-negotiation allows each device to advertise its capabilities and discover
those of its peer (link partner). The highest common denominator supported by both link
partners is automatically selected, yielding the greatest available throughput, while requiring
no manual configuration. The bge driver also allows you to configure the advertised
capabilities to less than the maximum (where the full speed of the interface is not required), or
to force a specific mode of operation, irrespective of the link partner's advertised capabilities.
Application The cloning character-special device, /dev/bge, is used to access all BCM57xx devices (
Programming (BCM5700/5701/5703/5704, 5705/5714/5721/5751/5751M/5782 on x86) fitted to the system
Interface motherboard.

The bge driver is managed by the dladm(1M) command line utility, which allows VLANs to be
defined on top of bge instances and for bge instances to be aggregated. See dladm(1M) for
more details.
You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number for the system. The device is initialized on first attach and de-initialized (stopped) at
last detach.
The values returned by the driver in the DL_INFO_ACK primitive in response to a
DL_INFO_REQ are:
■ Maximum SDU (default 1500).
■ Minimum SDU (default 0).
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP length value is -2, meaning the physical address component is followed immediately
by a 2-byte SAP component within the DLSAP address.

98 man pages section 7: Device and Network Interfaces • Last Revised 9 Apr 2008
 bge(7D)

■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.

Configuration By default, the bge driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can be any one of the following, (as described in the IEEE803.2 standard):
■ 1000 Mbps, full-duplex
■ 1000 Mbps, half-duplex
■ 100 Mbps, full-duplex
■ 100 Mbps, half-duplex
■ 10 Mbps, full-duplex
■ 10 Mbps, half-duplex

The auto-negotiation protocol automatically selects:

■ Speed (1000 Mbps, 100 Mbps, or 10 Mbps)
■ Operation mode (full-duplex or half-duplex)

as the highest common denominator supported by both link partners. Because the bge device
supports all modes, the effect is to select the highest throughput mode supported by the other
device.

Alternatively, you can set the capabilities advertised by the bge device using dladm(1M). The
driver supports a number of parameters whose names begin with en_ (see below). Each of
these parameters contains a boolean value that determines whether the device advertises that
mode of operation. If en_autoneg_cap is set to 0, the driver forces the mode of operation
selected by the first non-zero parameter in priority order as listed below:
(highest priority/greatest throughput)
en_1000fdx_cap 1000Mbps full duplex
en_1000hdx_cap 1000Mpbs half duplex
en_100fdx_cap 100Mpbs full duplex
en_100hdx_cap 100Mpbs half duplex
en_10fdx_cap 10Mpbs full duplex
en_10hdx_cap 10Mpbs half duplex
(lowest priority/least throughput)

For example, to prevent the device 'bge2' from advertising gigabit capabilities, enter (as
super-user):
# dladm set-linkprop -p enable_1000hdx_cap=0 bge2
# dladm set-linkprop -p enable_1000fdx_cap=0 bge2

All capabilities default to enabled. Note that changing any capability parameter causes the link
to go down while the link partners renegotiate the link speed/duplex using the newly changed
capabilities.

Device and Network Interfaces 99

bge(7D)

The current settings of the parameters may be found using dladm show-ether. In addition, the
driver exports the current state, speed, duplex setting, and working mode of the link via kstat
parameters (these are read only and may not be changed). For example, to check link state of
device bge0:
# dladm show-ether -x bge0
LINK PTYPE STATE AUTO SPEED-DUPLEX PAUSE
bge0 current up yes 1G-f bi
-- capable -- yes 1G-fh,100M-fh,10M-fh bi
-- adv -- yes 1G-fh bi
-- peeradv -- yes 1G-f bi

The output above indicates that the link is up and running at 1Gbps full-duplex with its rx/tx
direction pause capability.

To extract link state information for the same link using kstat:
# kstat bge:0:mac:link_state
module: bge instance: 0
name: mac class: net
link_state

The default MTU is 1500. To enable Jumbo Frames support, you can configure the bge driver
by defining the default_mtu property via dladm(1M) or in driver.conf(4) to greater than
1500 bytes (for example: default_mtu=9000). Note that the largest jumbo size supported by
bge is 9000 bytes. Additionally, not all bge-derived devices currently support Jumbo Frames.
The following devices support Jumbo Frames up to 9KB: BCM5700, 5701, 5702, 5703C,
5703S, 5704C, 5704S, 5714C, 5714S, 5715C and 5715S. Other devices currently do not support
Jumbo Frames.
Files /kernel/drv/bge* 32–bit ELF kernel module. (x86)
/kernel/drv/amd64/bge 64–bit ELF kernel module (x86).
/kernel/drv/sparcv9/bge 64–bit ELF kernel module (SPARC).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), driver.conf(4), attributes(5), streamio(7I), dlpi(7P)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

100 man pages section 7: Device and Network Interfaces • Last Revised 9 Apr 2008
 blk2scsa(7D)

Name blk2scsa – SCSA block device emulation

Description The blk2scsa module provides support services for generic block devices so that they appear
to the system as devices on a virtual SCSI bus, thus allowing them to be serviced by the sd(7D)
SCSI disk driver. The blk2scsa device supports the SCSI-2 command set for Direct Access
Devices. The blk2scsa device supports multiple LUNs per physical device and creates a
separate child device for each LUN. All child nodes attach to sd(7D).

Device Special Disk block special file names are located in /dev/dsk. Raw file names are located in
Files /dev/rdsk. See sd(7D).

ioctls See dkio(7I)

Errors See sd(7D).

Files Device special files for the storage device are created in the same way as those for a SCSI disk.
See sd(7D) for more information.
/dev/dsk/cntndnsn Block files for disks.
/dev/rdsk/ctndnsn Raw files for disks.
/kernel/misc/blk2scsa 32-bit ELF kernel module (x86).
/kernel/misc/amd64/blk2scsa 64-bit ELF kernel module (x86).
/kernel/misc/sparcv9/blk2scsa 64-bit ELF kernel module (SPARC).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWckr

See Also eject(1), rmformat(1), rmmount(1), cfgadm_scsi(1M), fdisk(1M), mount(1M), umount(1M),

scsi(4), vfstab(4), attributes(5), sd(7D), dkio(7I), pcfs(7FS)

Device and Network Interfaces 101

bnx(7D)

Name bnx – Broadcom NetXtreme II Gigabit Ethernet Device Driver

Synopsis /dev/bnx
Description The bnx Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD v3-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), over Broadcom
NetXtreme II Ethernet controllers, including the BCM5706, BCM5708 and BCM5709
controllers. Driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support and error recovery and reporting.
Application The cloning, character-special device /dev/bnx is used to access all Broadcom NetXtreme II
Programming Ethernet devices installed within the system.
Interface
The bnx driver is dependent on /kernel/misc/mac, a loadable kernel module that provides
the bnx driver with the DLPI and STREAMS functionality required of a LAN driver.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are:
■ Maximum SDU (with jumbo frame) is 9000.
■ Minimum SDU is 0. The driver pads to 60-byte minimum packet size.
■ DSLAP address length is 8 bytes.
■ MAC type is DL_ETHER.
■ SAP length value is -2, meaning the physical address component is followed immediately
by a 2-byte sap component within the DLSAP address.
■ Version is DL_VERSION_2.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Configuration By default, the bnx driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can be any of the following:
2500 Mbps, full-duplex (fiber physical interface controller only)
1000 Mbps, full-duplex
100 Mbps, full-duplex
100 Mbps, half-duplex
10 Mbps, full-duplex
10 Mbps, half-duplex
To customize the driver parameters, edit the /kernel/drv/bnx.conf file. The driver
properties are:
adv_*
The adv parameters are advertised to the link partner and include:

102 man pages section 7: Device and Network Interfaces • Last Revised 30 Nov 2009
 bnx(7D)

adv_autoneg_cap
adv_pause_cap
adv_2500fdx_cap
adv_1000fdx_cap
adv_1000hdx_cap
adv_100fdx_cap
adv_100hdx_cap
adv_10fdx_cap
adv_10hdx_cap

transfer_speed
The driver attempts to auto-negotiate but is restricted to the specified speed. Duplex mode
is determined through auto-negotiation.
speed
full-duplex
Forces speed and duplex mode to a fixed value. This value take precedence over others.
speed
Configures link (or instance) to a designated speed. By default, AutoNegotiate (0) is set.
The setup is based on the following values:
0 AutoNegotiate.
10 10 Mbps speed mode (Copper only).
100 100 Mbps speed mode (Copper only).
1000 1000 Mbps speed mode (Copper and fiber).
2500 2500 Mbps speed mode (Fiber only).
Flow
Configures flow control parameters of a link. The setup is based on the following values:
0 Tx and Rx flow control are disabled.
1 Tx flow control is enabled. Pause frames are sent if resource is low, but
device does not process Rx Pause Frame.
2 Only Rx flow control is enabled. If device receives Pause Frame, it
stops sending.
3 Rx and TX flow control are enabled. Pause frames are sent if resource is
low. If device receives Pause Frame, it stops sending.
4 Advertise Rx and TX flow control are enabled and negotiating with link
partner. If link AutoNegotiate is not enabled, Tx and Rx Flow Control
are disabled.

Device and Network Interfaces 103

bnx(7D)

Jumbo
Configures Jumbo Frame link feature. Valid range for this parameter is 0 to 3800. If value
configured is less then 1500, Jumbo Frame feature is disabled.
RxBufs
Configures number of Rx packet descriptor. The valid value is 32 to 1024. More system
memory resource is used for larger number of Rx Packet Descriptors. Default value is 500.
RxTicks
Configures number of Rx Host Coalescing Ticks in microseconds. This determines the
maximum time interval in which the device generates an interrupt if one or more frames
are received. The default value is 25.
Coalesce
Configures number of Tx/Rx Maximum Coalesced Frames parameters. This determines
the maximum number of buffer descriptors the device processes before it generates an
interrupt. The default value is 16.
TxTicks
Configures number of Tx Host Coalescing Ticks in microseconds. This determines the
maximum time interval in which the device generates an interrupt if one or more frames
are sent. The default value is 45.
TxMaxCoalescedFrames
Configures number of Tx Maximum Coalesced Frames parameters. This determines the
maximum number of Tx buffer descriptors the device processes before it generates an
interrupt. The default value is 80.
RxTicksInt
Configures number of Rx Host Coalescing Ticks in microseconds during interrupt. This
determines the maximum time interval in which the device generates interrupt if one or
more frames are received during interrupt handling. The default value is 15.
TxTicksInt
Configures number of Tx Host Coalescing Ticks in microseconds during interrupt. This
determines the maximum time interval in which the device generates an interrupt if one or
more frames are received during interrupt handling. The default value is 15.
StatsTicks
Configures how often adapter statistics are DMA'd to host memory in microsecond.
Default is 1000000.

Configuring with You can also perform configuration tasks using ndd(1M). For example, to prevent the device
ndd(1M) bnx1 from advertising gigabit capabilities, do the following as super-user:

# ndd -set /dev/bnx1 adv_1000fdx_cap 0

All capabilities default to enabled and that changing any parameter causes the link to go down
while the link partners renegotiate the link speed/duplex. To view current parameters, use

104 man pages section 7: Device and Network Interfaces • Last Revised 30 Nov 2009
 bnx(7D)

ndd-get. In addition, the driver exports the current state, speed, duplex setting and working
mode of the link by way of the ndd parameters, which are read only and cannot be changed.
For example, to check the state of device bnx0:
# ndd -get /dev/bnx0 link_status
1
# ndd -get /dev/bnx0 link_speed
100
# ndd -get /dev/bnx0 link_duplex
2

The output above indicates that the link is up and running at 100Mbps full-duplex.
Files /dev/bnx Special character device
/kernel/drv/bnx 32-bit ELF kernel module (x86)
/kernel/drv/amd64/bnx 64-bit ELF Kernel module (x86)
/kernel/drv/bnx.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability BRCMbnx

Architecture x86

Interface Stability See below.

The bnx driver is Committed. The /kernel/drv/bnx.conf configuration file is Uncommitted.

See Also dladm(1M), ndd(1M), attributes(5), streamio(7I), dlpi(7P)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

Device and Network Interfaces 105

bnxe(7D)

Name bnxe – Broadcom NetXtreme II 10 Gigabit Ethernet Device Driver

Synopsis /dev/bnxe*
Description The bnxe Ethernet driver is a multi-threaded, loadable, clonable, GLDv3-based driver
supporting the Data Link Provider Interface, dlpi(7P), over Broadcom NetXtreme II 10
Gigabit Ethernet controllers. Multiple NetXtreme II controllers installed within the system are
supported by the driver.
The bnxe driver provides basic support for the NetXtreme II 10 Gigabit line of devices.
Functions include chip initialization, frame transit and receive, multicast and promiscuous
support, and error recovery and reporting. NetXtreme II 10 Gigabit devices provide
10/100/1000/10000 Mbps networking interfaces for Copper and 1000/2500/10000 Mbps for
Fiber physical interfaces.
The IEEE 802.3 standard specifies an auto-negotiation protocol to automatically select the
mode and speed of operation. The PHY device is capable of doing auto-negotiation with the
remote-end of the link (Link Partner) and receives the capabilities from the remote end. It
selects the Highest Common Denominator mode of operation. It also supports forced-mode
of operation where the driver can select desired mode of operation.
Driver Configuration There are two facilities by which the administrator can configure each NetXtreme II 10
Gigabit device in the system, the hardware configuration file and the NDD subsystem. The
hardware configuration file is located at /kernel/drv/bnxe.conf and contains a list of
options that are read when the driver is loaded. The NDD subsystem options are used as a way
to modify the device's configuration at runtime. All changes made with the NDD subsystem
are lost after device reset or system reboot. The remainder of this section discusses
configuration options common to both facilities.
Link Speed and Duplex Parameters
The primary way link speed and duplex settings are configured is through the following list of
options. Non-zero values turn them on, a value of zero disables them.
■ adv_autoneg_cap
■ adv_10000fdx_cap
■ adv_2500fdx_cap
■ adv_1000fdx_cap
■ adv_100fdx_cap
■ adv_100hdx_cap
■ adv_10fdx_cap
■ adv_10hdx_cap
When the adv_autoneg_cap option is set to a non-zero value, the remaining options control
which capabilities we advertise to the link partner during auto-negotiation. When the
adv_autoneg_cap option is set to zero, the driver walks down the list, from the highest speed /
duplex option to the lowest, and use the first non-zero option as the speed / duplex setting to
force the link with.

106 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2009
 bnxe(7D)

Flow Control Parameters

Flow control parameters configure the flow control properties of the physical link. The
properties are configured through the following list of options. Like the link speed and duplex
settings, non-zero values turn them on, a value of zero disables them.
autoneg_flow Controls whether or not the flow control properties are auto-negotiated or
forced.
txpause_cap Controls whether or not Tx flow control can be configured.
rxpause_cap Controls whether or not Rx flow control can be configured.

If Tx flow control is enabled, pause frames are sent to the link partner when Rx resources are
low. If Rx flow control is enabled, the hardware automatically stops transmitting if pause
frames are received. How Tx and Rx flow control are enabled depends on how the driver is
configured. When the autoneg_flow option is non-zero, the flow control capabilities become
advertisement settings and theauto-negotiation process dictates what actual flow control
settings are used. If the autoneg_flow option is zero, the flow control capabilities specified are
still advertised if the link is auto-negotiated, but the actual flow control settings are forced to
the specified settings.
Hardware checksum This parameter configures checksum calculation tasks to be
Configuration File offloaded to the hardware. If 0 then no checksums are offloaded. If 1
Options
then IPv4 header checksums offloaded for Rx/Tx. If 2 then
TCP/UDP/IPv4 header checksums are offloaded for Rx/Tx. The
default is 2 (TCP/UDP/IPv4 checksums).
mtu This parameter controls the MTU (Message Tranfer Unit) size of
the hardware. Egress Ethernet frames larger than this size is
fragmented and ingress Ethernet frames larger than this size is
dropped. The valid range is from 60 to 9216 (decimal). The default
value is 1500.
rx_descs
tx_descs These parameters control how many packets can be in-transit
within the driver. The greater the number of packets that are
allowed to be in-transit, the more memory the driver requires to
operate. The valid range is 1 to 32767. The default value is 1280.
rx_free_reclaim
tx_free_reclaim These parameters control the threshold of freely available packet
descriptors that are allowed to be free before reposting back to the
hardware so they can be reused for future packets. The valid range
is 0 to the value of rx_descs for Rx and tx_descs for Tx. A freely
available packet descriptor is, for Rx a packet that has been received

Device and Network Interfaces 107

bnxe(7D)

and processing finished, and for Tx a packet that has already been
sent. The default value is 1/16 of rx_descs for Rx and tx_descs for
Tx.
interrupt_coalesce This parameter gives the administrator the ability to allow the
hardware to collect more network events before interrupting the
host processor. Interrupt coalescing within the NetXtreme II 10
Gigabit hardware is quite aggressive resulting in great sustained
throughput but low latency for interactive traffic. Setting to 1 turns
it on and 0 off. Default is off.
rx_copy_threshold
tx_copy_threshold These parameters control the packet size threshold, in number of
bytes, when packets are double copied before processing. A value of
0 turns off double copies. For Tx a value of 1 means all packets are
copied. For Rx a really large value, that is, greater than the mtu, that
means all packets are copied. The default is 0 for both Rx and Tx
which implies that no copying is ever performed.

NDD Subsystem
Configuration Options
Hardware Capability Parameters
The following is a read-only list of capabilities the device is capable of supporting. A value of 1
means the capability is supported. A value of 0 means the capability is not supported.
■ autoneg_cap
■ 10000fdx_cap
■ 2500fdx_cap
■ 1000fdx_cap
■ 1000fdx_cap
■ 100hdx_cap
■ 10fdx_cap
■ 10hdx_cap
■ 10hdx_cap
■ txpause_cap

Hardware Capability Advertisement Parameters

The auto-negotiation advertisement parameters work exactly as described in the Hardware
Capability Parameters section of this man page with the difference that they show what is
actually being advertised

Miscellaneous Link Parameters

These parameters show the other phy options.
■ autoneg_flow

108 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2009
 bnxe(7D)

Link Status Parameters

These parameters show the current status of the physical link.
link_status Shows if the link is up or down.
link_speed Shows the current speed of the link.
link_duplex Shows the current duplex setting of the link.
link_txpause Shows whether or not TX flow control is enabled.
link_rxpause Shows whether or not RX flow control is enabled.

Convenience Parameters
The following parameters display multiple settings at once. They are intended for
convenience.
hw_cap Shows all the device hardware capabilities
adv_cap Shows all the capabilities we are advertising or have forced.

Debug
The following parameters are used to aid in debugging driver issues. These should be used in
conjunction with kstat statistic diagnostics.
debug Shows the state of all the internal packet descriptor queues.
Files /dev/bnx[instance] bnxe character special device.
/kernel/drv/bnxe.conf Configuration file of the bnxe driver.
/kernel/drv/bnxe 32-bit i386 driver binary
/kernel/drv/amd64/bnxe 64-bit i386 driver binary
release.txt Revision history of the driver

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability BRCMbnxe

Architecture x86

See Also ndd(1M), attributes(5), dlpi(7P), gld(7D),

Broadcom NetXtreme II 10 Gigabit Adapter Driver Installation Notes

Writing Device Drivers

Device and Network Interfaces 109

bnxe(7D)

STREAMS Programming Guide

Network Interfaces Programmer's Guide

110 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2009
 bpf(7D)

Name bpf – Berkeley Packet Filter raw network interface

Description The Berkeley Packet Filter provides a raw interface to data link layers in a protocol
independent fashion. All packets on the network, even those destined for other hosts, are
accessible through this mechanism.
The packet filter appears as a character special device, /dev/bpf. After opening the device, the
file descriptor must be bound to a specific network interface with the BIOSETIF ioctl. A
specific interface can be shared by multiple listeners, and the filter underlying each descriptor
sees an identical packet stream.
Associated with each open instance of a bpf file is a user-settable packet filter. Whenever a
packet is received by an interface, all file descriptors listening on that interface apply their
filter. Each descriptor that accepts the packet receives its own copy.
Reads from these files return the next group of packets that have matched the filter. To
improve performance, the buffer passed to read must be the same size as the buffers used
internally by bpf. This size is returned by the BIOCGBLEN ioctl , and under BSD, can be set with
BIOCSBLEN. An individual packet larger than this size is necessarily truncated.
The packet filter supports any link level protocol that has fixed length headers. Currently, only
Ethernet, SLIP and PPP drivers have been modified to interact with bpf.
Since packet data is in network byte order, applications should use the byteorder(3SOCKET)
macros to extract multi-byte values.
A packet can be sent out on the network by writing to a bpf file descriptor. The writes are
unbuffered, meaning that only one packet can be processed per write. Currently, only writes to
Ethernets and SLIP links are supported.
ioctls The ioctl(2) command codes in this section are defined in <net/bfp.h>. All commands
require these includes:
#include <sys/types.h>
#include <sys/time.h>
#include <sys/time.h>
#include <net/bpf.h>

Additionally, BIOCGETIF and BIOCSETIF require <net/if.h>.

The third argument to the ioctl(2)should be a pointer to the type indicated.
BIOCGBLEN (u_int)
Returns the required buffer length for reads on bpf files.
BIOCSBLEN (u_int)
Sets the buffer length for reads on bpf files. The buffer must be set before the file is attached
to an interface with BIOCSETIF. If the requested buffer size cannot be accommodated, the
closest allowable size is set and returned in the argument. A read call results in EINVAL if it
is passed a buffer that is not this size.

Device and Network Interfaces 111

bpf(7D)

BIOCGDLT (u_int)
Returns the type of the data link layer underlying the attached interface. EINVAL is returned
if no interface has been specified. The device types, prefixed with DLT_, are defined in
<net/bpf.h>.
BIOCGDLTLIST (struct bpf_dltlist)
Returns an array of available type of the data link layer underlying the attached interface:
struct bpf_dltlist {
u_int bfl_len;
u_int *bfl_list;
};
The available type is returned to the array pointed to the bfl_list field while its length in
u_int is supplied to the bfl_len field. NOMEM is returned if there is not enough buffer. The
bfl_len field is modified on return to indicate the actual length in u_int of the array
returned. If bfl_list is NULL, the bfl_len field is returned to indicate the required length
of an array in u_int.
BIOCSDLT (u_int)
Change the type of the data link layer underlying the attached interface. EINVAL is returned
if no interface has been specified or the specified type is not available for the interface.
BIOCPROMISC
Forces the interface into promiscuous mode. All packets, not just those destined for the
local host, are processed. Since more than one file can be listening on a given interface, a
listener that opened its interface non-promiscuously can receive packets promiscuously.
This problem can be remedied with an appropriate filter.

The interface remains in promiscuous mode until all files listening promiscuously are
closed.
BIOCFLUSH
Flushes the buffer of incoming packets, and resets the statistics that are returned by
BIOCGSTATS.
BIOCGETIF (struct ifreq)
Returns the name of the hardware interface that the file is listening on. The name is
returned in the ifr_name field of ifr. All other fields are undefined.
BIOCSETIF (struct ifreq)
Sets the hardware interface associate with the file. This command must be performed
before any packets can be read. The device is indicated by name using the ifr_name field of
the ifreq. Additionally, performs the actions of BIOCFLUSH.
BIOCSRTIMEOUT, BIOCGRTIMEOUT (struct timeval)
Set or get the read timeout parameter. The timeval specifies the length of time to wait
before timing out on a read request. This parameter is initialized to zero by open(2),
indicating no timeout.

112 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2009
 bpf(7D)

BIOCGSTATS (struct bpf_stat)

Returns the following structure of packet statistics:
struct bpf_stat {
uint64_t bs_recv;
uint64_t bs_drop;
uint64_t bs_capt;
uint64_t bs_padding[13];
};

The fields are:

bs_recv Number of packets received by the descriptor since opened or reset (including
any buffered since the last read call.
bs_drop Number of packets which were accepted by the filter but dropped by the
kernel because of buffer overflows, that is, the application's reads aren't
keeping up with the packet traffic.
bs_capt Number of packets accepted by the filter.
BIOCIMMEDIATE (u_int)
Enable or disable immediate mode, based on the truth value of the argument. When
immediate mode is enabled, reads return immediately upon packet reception. Otherwise, a
read blocks until either the kernel buffer becomes full or a timeout occurs. This is useful for
programs like rarpd(1M), which must respond to messages in real time. The default for a
new file is off.
BIOCSETF (struct bpf_program)
Sets the filter program used by the kernel to discard uninteresting packets. An array of
instructions and its length is passed in using the following structure:
struct bpf_program {
u_int bf_len;
struct bpf_insn *bf_insns;
};

The filter program is pointed to by the bf_insns field while its length in units of struct
bpf_insn is given by the bf_len field. The actions of BIOCFLUSH are also performed.

See the FILTER MACHINE section of this manual page for an explanation of the filter
language.
BIOCVERSION (struct bpf_version)
Returns the major and minor version numbers of the filter language currently recognized
by the kernel. Before installing a filter, applications must check that the current version is
compatible with the running kernel. Version numbers are compatible if the major numbers
match and the application minor is less than or equal to the kernel minor. The kernel
version number is returned in the following structure:

Device and Network Interfaces 113

bpf(7D)

struct bpf_version {
u_short bv_major;
u_short bv_minor;
};
The current version numbers are given by BPF_MAJOR_VERSION and BPF_MINOR_VERSION
from <net/bpf.h>.

An incompatible filter can result in undefined behavior, most likely, an error returned by
ioctl(2) or haphazard packet matching.
BIOCGHDRCMPLT BIOCSHDRCMPLT (u_int)
Enable/disable or get the header complete flag status. If enabled, packets written to the
bpf file descriptor does not have network layer headers rewritten in the interface output
routine. By default, the flag is disabled (value is 0).
BIOCGSEESENT BIOCSSEESENT (u_int)
Enable/disable or get the see sent flag status. If enabled, packets sent is passed to the filter.
By default, the flag is enabled (value is 1).

Standard Ioctls bpf supports several standard ioctl(2)'s that allow the user to do async or non-blocking I/O
to an open file descriptor.
FIONREAD (int) Returns the number of bytes that are immediately
available for reading.
SIOCGIFADDR (struct ifreq) Returns the address associated with the interface.
FIONBIO (int) Set or clear non-blocking I/O. If arg is non-zero, then
doing a read(2) when no data is available returns -1 and
errno is set to EAGAIN. If arg is zero, non-blocking I/O is
disabled. Setting this overrides the timeout set by
BIOCSRTIMEOUT.
FIOASYNC (int) Enable or disable async I/O. When enabled (arg is
non-zero), the process or process group specified by
FIOSETOWN starts receiving SIGIOs when packets arrive.
You must do an FIOSETOWN for this to take effect, as the
system does not default this for you. The signal can be
changed using BIOCSRSIG.
FIOSETOWN FIOGETOWN (int) Set or get the process or process group (if negative) that
should receive SIGIO when packets are available. The
signal can be changed using BIOCSRSIG.

bpf Header The following structure is prepended to each packet returned by read(2):

struct bpf_hdr {
struct timeval bh_tstamp;
uint32_t bh_caplen;

114 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2009
 bpf(7D)

uint32_t bh_datalen;
uint16_t bh_hdrlen;
};

The fields, whose values are stored in host order, and are:
bh_tstamp The time at which the packet was processed by the packet filter.
bh_caplen The length of the captured portion of the packet. This is the minimum of the
truncation amount specified by the filter and the length of the packet.
bh_datalen The length of the packet off the wire. This value is independent of the
truncation amount specified by the filter.
bh_hdrlen The length of the BPF header, which cannot be equal to sizeof (struct
bpf_hdr).

The bh_hdrlen field exists to account for padding between the header and the link level
protocol. The purpose here is to guarantee proper alignment of the packet data structures,
which is required on alignment sensitive architectures and improves performance on many
other architectures. The packet filter ensures that the bpf_hdr and the network layer header is
word aligned. Suitable precautions must be taken when accessing the link layer protocol fields
on alignment restricted machines. This is not a problem on an Ethernet, since the type field is
a short falling on an even offset, and the addresses are probably accessed in a bytewise
fashion).

Additionally, individual packets are padded so that each starts on a word boundary. This
requires that an application has some knowledge of how to get from packet to packet. The
macro BPF_WORDALIGN is defined in <net/bpf.h> to facilitate this process. It rounds up its
argument to the nearest word aligned value, where a word is BPF_ALIGNMENT bytes wide.

For example, if p points to the start of a packet, this expression advances it to the next packet:
p = (char *)p + BPF_WORDALIGN(p->bh_hdrlen + p->bh_caplen)

For the alignment mechanisms to work properly, the buffer passed to read(2) must itself be
word aligned. malloc(3C) always returns an aligned buffer.

Filter Machine A filter program is an array of instructions, with all branches forwardly directed, terminated
by a return instruction. Each instruction performs some action on the pseudo-machine state,
which consists of an accumulator, index register, scratch memory store, and implicit program
counter.

The following structure defines the instruction format:

struct bpf_insn {
uint16_t code;
u_char jt;
u_char jf;

Device and Network Interfaces 115

bpf(7D)

int32_t k;
};

The k field is used in different ways by different instructions, and the jt and jf fields are used
as offsets by the branch instructions. The opcodes are encoded in a semi-hierarchical fashion.
There are eight classes of instructions: BPF_LD, BPF_LDX, BPF_ST, BPF_STX, BPF_ALU, BPF_JMP,
BPF_RET, and BPF_MISC. Various other mode and operator bits are or'd into the class to give
the actual instructions. The classes and modes are defined in <net/bpf.h>.

Below are the semantics for each defined BPF instruction. We use the convention that A is the
accumulator, X is the index register, P[] packet data, and M[] scratch memory store. P[i:n]
gives the data at byte offset i in the packet, interpreted as a word (n=4), unsigned halfword
(n=2), or unsigned byte (n=1). M[i] gives the i'th word in the scratch memory store, which is
only addressed in word units. The memory store is indexed from 0 to BPF_MEMWORDS-1.k, jt,
and jf are the corresponding fields in the instruction definition. len refers to the length of the
packet.
BPF_LD These instructions copy a value into the accumulator. The type of the source
operand is specified by an addressing mode and can be a constant (BBPF_IMM),
packet data at a fixed offset (BPF_ABS), packet data at a variable offset (BPF_IND),
the packet length (BPF_LEN), or a word in the scratch memory store (BPF_MEM).
For BPF_IND and BPF_ABS, the data size must be specified as a word (BPF_W),
halfword (BPF_H), or byte (BPF_B). The semantics of all the recognized BPF_LD
instructions follow.
BPF_LD+BPF_W+BPF_ABS A <- P[k:4]
BPF_LD+BPF_H+BPF_ABS A <- P[k:2]
BPF_LD+BPF_B+BPF_ABS A <- P[k:1]
BPF_LD+BPF_W+BPF_IND A <- P[X+k:4]
BPF_LD+BPF_H+BPF_IND A <- P[X+k:2]
BPF_LD+BPF_B+BPF_IND A <- P[X+k:1]
BPF_LD+BPF_W+BPF_LEN A <- len
BPF_LD+BPF_IMM A <- k
BPF_LD+BPF_MEM A <- M[k]
BPF_LDX These instructions load a value into the index register. The addressing modes
are more restricted than those of the accumulator loads, but they include
BPF_MSH, a hack for efficiently loading the IP header length.
BPF_LDX+BPF_W+BPF_IMM X <- k
BPF_LDX+BPF_W+BPF_MEM X <- M[k]
BPF_LDX+BPF_W+BPF_LEN X <- len
BPF_LDX+BPF_B+BPF_MSH X <- 4*(P[k:1]&0xf)
BPF_ST This instruction stores the accumulator into the scratch memory. We do not
need an addressing mode since there is only one possibility for the destination.
BPF_ST M[k] <- A

116 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2009
 bpf(7D)

BPF_ALU The alu instructions perform operations between the accumulator and index
register or constant, and store the result back in the accumulator. For binary
operations, a source mode is required (BPF_K or BPF_X).
BPF_ALU+BPF_ADD+BPF_K A <- A + k
BPF_ALU+BPF_SUB+BPF_K A <- A - k
BPF_ALU+BPF_MUL+BPF_K A <- A * k
BPF_ALU+BPF_DIV+BPF_K A <- A / k
BPF_ALU+BPF_AND+BPF_K A <- A & k
BPF_ALU+BPF_OR+BPF_K A <- A | k
BPF_ALU+BPF_LSH+BPF_K A <- A << k
BPF_ALU+BPF_RSH+BPF_K A <- A >> k
BPF_ALU+BPF_ADD+BPF_X A <- A + X
BPF_ALU+BPF_SUB+BPF_X A <- A - X
BPF_ALU+BPF_MUL+BPF_X A <- A * X
BPF_ALU+BPF_DIV+BPF_X A <- A / X
BPF_ALU+BPF_AND+BPF_X A <- A & X
BPF_ALU+BPF_OR+BPF_X A <- A | X
BPF_ALU+BPF_LSH+BPF_X A <- A << X
BPF_ALU+BPF_RSH+BPF_X A <- A >> X
BPF_ALU+BPF_NEG A <- -A
BPF_JMP The jump instructions alter flow of control. Conditional jumps compare the
accumulator against a constant (BPF_K) or the index register (BPF_X). If the
result is true (or non-zero), the true branch is taken, otherwise the false branch
is taken. Jump offsets are encoded in 8 bits so the longest jump is 256
instructions. However, the jump always (BPF_JA) opcode uses the 32 bit k field
as the offset, allowing arbitrarily distant destinations. All condition also use
unsigned comparison conventions.
BPF_JMP+BPF_JA pc += k
BPF_JMP+BPF_JGT+BPF_K pc += (A > k) ? jt : jf
BPF_JMP+BPF_JGE+BPF_K pc += (A >= k) ? jt : jf
BPF_JMP+BPF_JEQ+BPF_K pc += (A == k) ? jt : jf
BPF_JMP+BPF_JSET+BPF_K pc += (A & k) ? jt : jf
BPF_JMP+BPF_JGT+BPF_X pc += (A > X) ? jt : jf
BPF_JMP+BPF_JGE+BPF_X pc += (A >= X) ? jt : jf
BPF_JMP+BPF_JEQ+BPF_X pc += (A == X) ? jt : jf
BPF_JMP+BPF_JSET+BPF_X pc += (A & X) ? jt : jf
BPF_RET The return instructions terminate the filter program and specify the amount of
packet to accept, that is, they return the truncation amount. A return value of
zero indicates that the packet should be ignored. The return value is either a
constant (BPF_K) or the accumulator (BPF_A).
BPF_RET+BPF_A accept A bytes
BPF_RET+BPF_K accept k bytes

Device and Network Interfaces 117

bpf(7D)

BPF_MISC The miscellaneous category was created for anything that does not fit into the
other classes in this section, and for any new instructions that might need to be
added. Currently, these are the register transfer instructions that copy the index
register to the accumulator or vice versa.
BPF_MISC+BPF_TAX X <- A
BPF_MISC+BPF_TXA A <- X

The BPF interface provides the following macros to facilitate array initializers:
BPF_STMT (opcode, operand)
BPF_JUMP (opcode, operand, true_offset, false_offset)

Sysctls The following sysctls are available when bpf is enabled:

net.bpf.maxbufsize Sets the maximum buffer size available for bpf peers.
net.bpf.stats Shows bpf statistics. They can be retrieved with the netstat(1M)
utility.
net.bpf.peers Shows the current bpf peers. This is only available to the super user
and can also be retrieved with the netstat(1M) utility.
Files /dev/bpf

Examples EXAMPLE 1 Using bfp to Accept Only Reverse ARP Requests

The following example shows a filter taken from the Reverse ARP Daemon. It accepts only
Reverse ARP requests.
struct bpf_insn insns[] = {
BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_REVARP, 0, 3),
BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, REVARP_REQUEST, 0, 1),
BPF_STMT(BPF_RET+BPF_K, sizeof(struct ether_arp) +
sizeof(struct ether_header)),
BPF_STMT(BPF_RET+BPF_K, 0),
};

EXAMPLE 2 Using bfp to Accept IP Packets

The following example shows filter that accepts only IP packets between host 128.3.112.15
and 128.3.112.35.
struct bpf_insn insns[] = {
BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 8),
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 26),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 2),

118 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2009
 bpf(7D)

EXAMPLE 2 Using bfp to Accept IP Packets (Continued)

BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 3, 4),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 0, 3),
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 1),
BPF_STMT(BPF_RET+BPF_K, (u_int)-1),
BPF_STMT(BPF_RET+BPF_K, 0),
};

EXAMPLE 3 Using bfp to Return Only TCP Finger Packets

The following example shows a filter that returns only TCP finger packets. The IP header must
be parsed to reach the TCP header. The BPF_JSET instruction checks that the IP fragment
offset is 0 so we are sure that we have a TCP header.
struct bpf_insn insns[] = {
BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 10),
BPF_STMT(BPF_LD+BPF_B+BPF_ABS, 23),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, IPPROTO_TCP, 0, 8),
BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20),
BPF_JUMP(BPF_JMP+BPF_JSET+BPF_K, 0x1fff, 6, 0),
BPF_STMT(BPF_LDX+BPF_B+BPF_MSH, 14),
BPF_STMT(BPF_LD+BPF_H+BPF_IND, 14),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 2, 0),
BPF_STMT(BPF_LD+BPF_H+BPF_IND, 16),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 0, 1),
BPF_STMT(BPF_RET+BPF_K, (u_int)-1),
BPF_STMT(BPF_RET+BPF_K, 0),
};

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Sparc, x86

Interface Stability Committed

See Also netstat(1M), rarpd(1M), lseek(2), ioctl(2), open(2), read(2), malloc(3C), select(3C),
byteorder(3SOCKET), signal(3C), attributes(5)

S. McCanne and V. Jacobson, The BSD Packet Filter: A New Architecture for User-level Packet
Capture, Proceedings of the 1993 Winter USENIX.

Device and Network Interfaces 119

bpf(7D)

Bugs The read buffer must be of a fixed size returned by the BIOCGBLEN ioctl.

A file that does not request promiscuous mode can receive promiscuous received packets as a
side effect of another file requesting this mode on the same hardware interface. This could be
fixed in the kernel with additional processing overhead. However, we favor the model where
all files must assume that the interface is promiscuous, and if so desired, must use a filter to
reject foreign packets.

Data link protocols with variable length headers are not currently supported.

Under SunOS, if a BPF application reads more than 2^31 bytes of data, read fails in
EINVALsignal(3C). You can either fix the bug in SunOS, or lseek(2) to 0 when read fails for
this reason.

Immediate mode and the read timeout are misguided features. This functionality can be
emulated with non-blocking mode and select(3C).

120 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2009
 bpp(7D)

Name bpp – bi-directional parallel port driver

Synopsis SUNW,bpp@slot,offset:bppn

Description The bpp driver provides a general-purpose bi-directional interface to parallel devices. It
supports a variety of output (printer) and input (scanner) devices, using programmable
timing relationships between the various handshake signals.

Application The bpp driver is an exclusive-use device. If the device has already been opened, subsequent
Programming opens fail with EBUSY.
Interface
Default Operation Each time the bpp device is opened, the default configuration is BPP_ACK_BUSY_HS for read
handshake, BPP_ACK_HS for write handshake, 1 microsecond for all setup times and strobe
widths, and 60 seconds for both timeouts. This configuration (in the write mode) drives many
common personal computer parallel printers with Centronics-type interfaces. The
application should use the BPPIOC_SETPARMS ioctl request to configure the bpp for the
particular device which is attached, if necessary.

Write Operation If a failure or error condition occurs during a write(2), the number of bytes successfully
written is returned (short write). Note that errno will not be set. The contents of certain status
bits will be captured at the time of the error, and can be retrieved by the application program,
using the BPPIOC_GETERR ioctl request. Subsequent write(2) calls may fail with the system
error ENXIO if the error condition is not rectified. The captured status information will be
overwritten each time an attempted transfer or a BPPIOC_TESTIO ioctl request occurs.

Read Operations If a failure or error condition occurs during a read(2), the number of bytes successfully read is
returned (short read). Note that errno will not be set. The contents of certain status bits will be
captured at the time of the error, and can be retrieved by the application, using the
BPPIOC_GETERR ioctl request. Subsequent read(2) calls may fail with ENXIO if the error
condition is not rectified. The captured register information will be overwritten each time an
attempted transfer or a BPPIOC_TESTIO ioctl request.

If the read_handshake element of the bpp_transfer_parms structure (see below) is set to

BPP_CLEAR_MEM or BPP_SET_MEM, zeroes or ones, respectively, are written into the user buffer.

Read/Write Operation When the driver is opened for reading and writing, it is assumed that scanning will take place,
as scanners are the only devices supported by this mode. Most scanners require that the
SLCT_IN or AFX pin be set to tell the scanner the direction of the transfer. The AFX line is set
when the read_handshake element of the bpp_transfer_parms structure is set to
BPP_HSCAN_HS, otherwise the SLCT_IN pin is set. Normally, scanning starts by writing a
command to the scanner, at which time the pin is set. When the scan data is read back, the pin
is reset.

ioctls The following ioctl requests are supported:

BPPIOC_SETPARMS Set transfer parameters.

Device and Network Interfaces 121

bpp(7D)

The argument is a pointer to a bpp_transfer_parms structure. See

below for a description of the elements of this structure. If a
parameter is out of range, EINVAL is returned.
BPPIOC_GETPARMS Get current transfer parameters.

The argument is a pointer to a bpp_transfer_parms structure. See

below for a description of the elements of this structure. If no
parameters have been configured since the device was opened, the
contents of the structure will be the default conditions of the
parameters (see Default Operation above).
BPPIOC_SETOUTPINS Set output pin values.

The argument is a pointer to a bpp_pins structure. See below for a

description of the elements of this structure. If a parameter is out of
range, EINVAL is returned.
BPPIOC_GETOUTPINS Read output pin values. The argument is a pointer to a bpp_pins
structure. See below for a description of the elements of this
structure.
BPPIOC_GETERR Get last error status.

The argument is a pointer to a bpp_error_status structure. See

below for a description of the elements of this structure. This
structure indicates the status of all the appropriate status bits at the
time of the most recent error condition during a read(2) or write(2)
call, or the status of the bits at the most recent BPPIOC_TESTIO ioctl
request. Note: The bits in the pin_status element indicate whether
the associated pin is active, not the actual polarity. The application
can check transfer readiness without attempting another transfer
using the BPPIOC_TESTIO ioctl. Note: The timeout_occurred and
bus_error fields will never be set by the BPPIOC_TESTIO ioctl, only
by an actual failed transfer.
BPPIOC_TESTIO Test transfer readiness.

This command checks to see if a read or write transfer would succeed

based on pin status, opened mode, and handshake selected. If a
handshake would succeed, 0 is returned. If a transfer would fail, -1 is
returned, and errno is set to EIO, and the error status information is
captured. The captured status can be retrieved using the
BPPIOC_GETERR ioctl call. Note that the timeout_occurred and
bus_error fields will never be set by this ioctl.

122 man pages section 7: Device and Network Interfaces • Last Revised 22 Aug 1994
 bpp(7D)

Transfer Parameters This structure is defined in <sys/bpp_io.h>.

Structure
struct bpp_transfer_parms {
enum handshake_t
read_handshake; /* parallel port read handshake mode */
int read_setup_time; /* DSS register - in nanoseconds */
int read_strobe_width; /* DSW register - in nanoseconds */
int read_timeout; /*
* wait this many seconds
* before aborting a transfer
*/
enum handshake_t
write_handshake; /* parallel port write handshake mode */
int write_setup_time; /* DSS register - in nanoseconds */
int write_strobe_width; /* DSW register - in nanoseconds */
int write_timeout; /*
* wait this many seconds
* before aborting a transfer
*/
};
/* Values for read_handshake and write_handshake fields */
enum handshake_t {
BPP_NO_HS, /* no handshake pins */
BPP_ACK_HS, /* handshake controlled by ACK line */
BPP_BUSY_HS, /* handshake controlled by BSY line */
BPP_ACK_BUSY_HS, /*
* handshake controlled by ACK and BSY lines
* read_handshake only!
*/
BPP_XSCAN_HS, /* xerox scanner mode,
* read_handshake only!
*/
BPP_HSCAN_HS, /*
* HP scanjet scanner mode
* read_handshake only!
*/
BPP_CLEAR_MEM, /* write 0’s to memory,
* read_handshake only!
*/
BPP_SET_MEM, /* write 1’s to memory,
* read_handshake only!
*/
/* The following handshakes are RESERVED. Do not use. */
BPP_VPRINT_HS, /* valid only in read/write mode */
BPP_VPLOT_HS /* valid only in read/write mode */
};

Device and Network Interfaces 123

bpp(7D)

The read_setup_time field controls the time between dstrb falling edge to bsy rising edge if
the read_handshake field is set to BPP_NO_HS or BPP_ACK_HS. It controls the time between
dstrb falling edge to ack rising edge if the read_handshake field is set to BPP_ACK_HS or
BPP_ACK_BUSY_HS. It controls the time between ack falling edge to dstrb rising edge if the
read_handshake field is set to BPP_XSCAN_HS.

The read_strobe_width field controls the time between ack rising edge and ack falling edge if
the read_handshake field is set to BPP_NO_HS or BPP_ACK_BUSY_HS. It controls the time
between dstrb rising edge to dstrb falling edge if the read_handshake field is set to
BPP_XSCAN_HS.

The values allowed for the write_handshake field are duplicates of the definitions for the
read_handshake field. Note that some of these handshake definitions are only valid in one
mode or the other.

The write_setup_time field controls the time between data valid to dstrb rising edge for all
values of the write_handshake field.

The write_strobe_width field controls the time between dstrb rising edge and dstrb falling
edge if the write_handshake field is not set to BPP_VPRINT_HS or BPP_VPLOT_HS. It controls
the minimum time between dstrb rising edge to dstrb falling edge if the write_handshake
field is set to BPP_VPRINT_HS or BPP_VPLOT_HS.

Transfer Pins Structure This structure is defined in <sys/bpp_io.h>.

struct bpp_pins {
uchar_t output_reg_pins; /* pins in P_OR register */
uchar_t input_reg_pins; /* pins in P_IR register */
};

/* Values for output_reg_pins field */

#define BPP_SLCTIN_PIN 0x01 /* Select in pin */
#define BPP_AFX_PIN 0x02 /* Auto feed pin */
#define BPP_INIT_PIN 0x04 /* Initialize pin */
#define BPP_V1_PIN 0x08 /* reserved pin 1 */
#define BPP_V2_PI 0x10 /* reserved pin 2 */
#define BPP_V3_PIN 0x20 /* reserved pin 3 */
#define BPP_ERR_PIN 0x01 /* Error pin */
#define BPP_SLCT_PIN 0x02 /* Select pin */
#define BPP_PE_PIN 0x04 /* Paper empty pin */

Error Pins Structure This structure is defined in the include file <sys/bpp_io.h>.

struct bpp_error_status {
char timeout_occurred; /* 1 if a timeout occurred */
char bus_error; /* 1 if an SBus bus error */
uchar_t pin_status; /*
* status of pins which could

124 man pages section 7: Device and Network Interfaces • Last Revised 22 Aug 1994
 bpp(7D)

* cause an error
*/
};
/* Values for pin_status field */
#define BPP_ERR_ERR 0x01 /* Error pin active */
#define BPP_SLCT_ERR 0x02 /* Select pin active */
#define BPP_PE_ERR 0x04 /* Paper empty pin active */
#define BPP_SLCTIN_ERR 0x10 /* Select in pin active */
#define BPP_BUSY_ERR 0x40 /* Busy pin active */
Errors EBADF The device is opened for write-only access and a read is attempted, or the device is
opened for read-only access and a write is attempted.
EBUSY The device has been opened and another open is attempted. An attempt has been
made to unload the driver while one of the units is open.
EINVAL A BPPIOC_SETPARMS ioctl is attempted with an out of range value in the
bpp_transfer_parms structure. A BPPIOC_SETOUTPINS ioctl is attempted with
an invalid value in the pins structure. An ioctl is attempted with an invalid value
in the command argument. An invalid command argument is received during
modload(1M) or modunload(1M).
EIO The driver encountered an SBus bus error when attempting an access.

A read or write does not complete properly, due to a peripheral error or a transfer
timeout.

A BPPIOC_TESTIO ioctl call is attempted while a condition exists which would

prevent a transfer (such as a peripheral error).
ENXIO The driver has received an open request for a unit for which the attach failed. The
driver has received a read or write request for a unit number greater than the
number of units available. The driver has received a write request for a unit which
has an active peripheral error.
Files /dev/bppn bi-directional parallel port devices

See Also ioctl(2), read(2), write(2), sbus(4)

Device and Network Interfaces 125

bscv(7D)

Name bscv, bscbus, i2bsc – Blade support chip interface driver

Description The bscv, bscbus and i2bsc drivers interface with the Blade support chip used on Sun
Microsystem's Blade server products. These drivers provide a conduit for passing control,
environmental, cpu signature and event information between Solaris and the Blade support
chip.

These drivers do not export public interfaces. Instead they make information available via picl,
prtdiag, prtfru and related tools. In addition, these drivers log Blade support chip
environmental event information into system logs.
Files /platform/sun4u/kernel/drv/sparcv9/bscbus 64-bit ELF kernel driver
/platform/sun4u/kernel/drv/sparcv9/bscv 64-bit ELF kernel driver
/platform/sun4u/kernel/drv/sparcv9/i2bsc 64-bit ELF kernel driver
/platform/i86pc/kernel/drv/bscbus 32–bit ELF kernel file (x86 only)
/platform/i86pc/kernel/drv/bscv 32–bit ELF kernel file (x86 only)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Limited to systems with Blade Support Chip

Availability SUNWcar.u, SUNWcar.i

126 man pages section 7: Device and Network Interfaces • Last Revised 22 August 2003
 bufmod(7M)

Name bufmod – STREAMS Buffer Module

Synopsis #include <sys/bufmod.h>
ioctl(fd, I_PUSH, "bufmod");

Description bufmod is a STREAMS module that buffers incoming messages, reducing the number of
system calls and the associated overhead required to read and process them. Although bufmod
was originally designed to be used in conjunction with STREAMS-based networking device
drivers, the version described here is general purpose so that it can be used anywhere
STREAMS input buffering is required.
Read-side Behavior The behavior of bufmod depends on various parameters and flags that can be set and queried
as described below under IOCTLS. bufmod collects incoming M_DATA messages into chunks,
passing each chunk upstream when the chunk becomes full or the current read timeout
expires. It optionally converts M_PROTO messages to M_DATA and adds them to chunks as well. It
also optionally adds to each message a header containing a timestamp, and a cumulative count
of messages dropped on the stream read side due to resource exhaustion or flow control.
Thedefault settings of bufmod allow it to drop messages when flow control sets in or resources
are exhausted; disabling headers and explicitly requesting no drops makes bufmod pass all
messages through. Finally, bufmod is capable of truncating upstream messages to a fixed,
programmable length.
When a message arrives, bufmod processes it in several steps. The following paragraphs
discuss each step in turn.
Upon receiving a message from below, if the SB_NO_HEADER flag is not set, bufmod immediately
timestamps it and saves the current time value for later insertion in the header described
below.
Next, if SB_NO_PROTO_CVT is not set, bufmod converts all leading M_PROTO blocks in the
message to M_DATA blocks, altering only the message type field and leaving the contents alone.
It then truncates the message to the current snapshot length, which is set with the SBIOCSSNAP
ioctl described below.
Afterwards, if SB_NO_HEADER is not set, bufmod prepends a header to the converted message.
This header is defined as follows.
struct sb_hdr {
uint_t sbh_origlen;
uint_t sbh_msglen;
uint_t sbh_totlen;
uint_t sbh_drops;
#if defined(_LP64) || defined(_I32LPx)
struct timeval32 sbh_timestamp;
#else
struct timeval sbh_timestamp;
#endif /* !_LP64 */
};

Device and Network Interfaces 127

bufmod(7M)

The sbh_origlen field gives the message's original length before truncation in bytes. The
sbh_msglen field gives the length in bytes of the message after the truncation has been done.
sbh_totlen gives the distance in bytes from the start of the truncated message in the current
chunk (described below) to the start of the next message in the chunk; the value reflects any
padding necessary to insure correct data alignment for the host machine and includes the
length of the header itself. sbh_drops reports the cumulative number of input messages that
this instance of bufmod has dropped due to flow control or resource exhaustion. In the current
implementation message dropping due to flow control can occur only if the SB_NO_DROPS flag
is not set. (Note: this accounts only for events occurring within bufmod, and does not count
messages dropped by downstream or by upstream modules.) The sbh_timestamp field
contains the message arrival time expressed as a struct timeval.
After preparing a message, bufmod attempts to add it to the end of the current chunk, using the
chunk size and timeout values to govern the addition. The chunk size and timeout values are
set and inspected using the ioctl() calls described below. If adding the new message would
make the current chunk grow larger than the chunk size, bufmod closes off the current chunk,
passing it up to the next module in line, and starts a new chunk. If adding the message would
still make the new chunk overflow, the module passes it upward in an over-size chunk of its
own. Otherwise, the module concatenates the message to the end of the current chunk.
To ensure that messages do not languish forever in an accumulating chunk, bufmod maintains
a read timeout. Whenever this timeout expires, the module closes off the current chunk and
passes it upward. The module restarts the timeout period when it receives a read side data
message and a timeout is not currently active. These two rules insure that bufmod minimizes
the number of chunks it produces during periods of intense message activity and that it
periodically disposes of all messages during slack intervals, but avoids any timeout overhead
when there is no activity.
bufmod handles other message types as follows. Upon receiving an M_FLUSH message
specifying that the read queue be flushed, the module clears the currently accumulating chunk
and passes the message on to the module or driver above. (Note: bufmod uses zero length
M_CTL messages for internal synchronization and does not pass them through.) bufmod passes
all other messages through unaltered to its upper neighbor, maintaining message order for
non high priority messages by passing up any accumulated chunk first.
If the SB_DEFER_CHUNK flag is set, buffering does not begin until the second message is received
within the timeout window.
If the SB_SEND_ON_WRITE flag is set, bufmod passes up the read side any buffered data when a
message is received on the write side. SB_SEND_ON_WRITE and SB_DEFER_CHUNK are often used
together.
Write-side Behavior bufmod intercepts M_IOCTL messages for the ioctls described below. The module passes all
other messages through unaltered to its lower neighbor. If SB_SEND_ON_WRITE is set, message
arrival on the writer side suffices to close and transmit the current read side chunk.
ioctls bufmod responds to the following ioctls.

128 man pages section 7: Device and Network Interfaces • Last Revised 11 Nov 1997
 bufmod(7M)

SBIOCSTIME Set the read timeout value to the value referred to by the struct timeval
pointer given as argument. Setting the timeout value to zero has the
side-effect of forcing the chunk size to zero as well, so that the module will
pass all incoming messages upward immediately upon arrival. Negative
values are rejected with an EINVAL error.
SBIOCGTIME Return the read timeout in the struct timeval pointed to by the
argument. If the timeout has been cleared with the SBIOCCTIME ioctl,
return with an ERANGE error.
SBIOCCTIME Clear the read timeout, effectively setting its value to infinity. This results in
no timeouts being active and the chunk being delivered when it is full.
SBIOCSCHUNK Set the chunk size to the value referred to by the uint_t pointer given as
argument. See Notes for a description of effect on stream head high water
mark.
SBIOCGCHUNK Return the chunk size in the uint_t pointed to by the argument.
SBIOCSSNAP Set the current snapshot length to the value given in the uint_t pointed to
by the ioctl's final argument. bufmod interprets a snapshot length value of
zero as meaning infinity, so it will not alter the message. See Notes for a
description of effect on stream head high water mark.
SBIOCGSNAP Returns the current snapshot length in the uint_t pointed to by the ioctl's
final argument.
SBIOCSFLAGS Set the current flags to the value given in the uint_t pointed to by the
ioctl's final argument. Possible values are a combination of the following.
SB_SEND_ON_WRITE Transmit the read side chunk on arrival of a
message on the write side.
SB_NO_HEADER Do not add headers to read side messages.
SB_NO_DROPS Do not drop messages due to flow control
upstream.
SB_NO_PROTO_CVT Do not convert M_PROTO messages into M_DATA.
SB_DEFER_CHUNK Begin buffering on arrival of the second read side
message in a timeout interval.
SBIOCGFLAGS Returns the current flags in the uint_t pointed to by the ioctl's final
argument.

See Also dlpi(7P), pfmod(7M)

Device and Network Interfaces 129

bufmod(7M)

Notes Older versions of bufmod did not support the behavioral flexibility controlled by the
SBIOCSFLAGS ioctl. Applications that wish to take advantage of this flexibility can guard
themselves against old versions of the module by invoking the SBIOCGFLAGS ioctl and
checking for an EINVAL error return.

When buffering is enabled by issuing an SBIOCSCHUNK ioctl to set the chunk size to a non zero
value, bufmod sends a SETOPTS message to adjust the stream head high and low water marks to
accommodate the chunked messages.

When buffering is disabled by setting the chunk size to zero, message truncation can have a
significant influence on data traffic at the stream head and therefore the stream head high and
low water marks are adjusted to new values appropriate for the smaller truncated message
sizes.

Bugs bufmod does not defend itself against allocation failures, so that it is possible, although very
unlikely, for the stream head to use inappropriate high and low water marks after the chunk
size or snapshot length have changed.

130 man pages section 7: Device and Network Interfaces • Last Revised 11 Nov 1997
 cadp160(7D)

Name cadp160 – Adaptec Ultra160 SCSI host bus adapter driver

Synopsis scsi@unit-address

Description The cadp160 host bus adapter driver is a SCSA-compliant nexus driver that supports the
following Adaptec Ultra160 SCSI devices:
■ Adapters: 39160, 29160, 29160N, 29160LP
■ Chips: AIC-7892B1, AIC-7899A, AIC-7899B2

The cadp160 driver supports standard functions provided by the SCSA interface including
tagged and untagged queuing, wide, fast and ultra SCSI, and auto request sense. The cadp160
driver does not support linked commands. The cadp160 driver supports hot swap SCSI, hot
plug PCI, 64-bit addressing (dual address cycle), domain validation, PCI bus clock rates up to
66MHZ and narrow and wide devices at 20MB/sec, 40MB/sec, 80MB/sec, and 160MB/sec.
Files /platform/i86pc/kernel/drv/cadp160 32–bit ELF kernel module.
/platform/i86pc/kernel/drv/amd64/cadp160 64–bit ELF kernel module.
/platform/i86pc/kernel/drv/cadp160.conf Optional configuration file.

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also prtconf(1M), driver.conf(4), pci(4), attributes(5), dlpi(7P), scsi_abort(9F),

scsi_hba_attach_setup(9F), scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F),
scsi_sync_pkt(9F), scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S),
scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

Solaris (x86 Edition) Hardware Compatibility List

ANSI Small Computer System Interface-2 (SCSI-2)

Device and Network Interfaces 131

cadp(7D)

Name cadp – Adaptec Ultra-2 SCSI host bus adapter driver

Synopsis scsi@unit-address

Description The cadp host bus adapter driver is a SCSA–compliant nexus driver that supports the
following Adaptec Ultra-2 SCSI devices:
■ Adapters: Adaptec AHA-2940U2W, AHA-2940U2B, AHA-2940U2, AHA-2950U2B,
AHA-3950U2B
■ Chips: AIC-7896/AIC-7897, AIC-7890/AIC-7890A, AIC-7891, AIC-7890AB, AIC-7890A

The cadp driver supports standard functions provided by the SCSA interface, including
tagged and untagged queuing, Wide/Fast/Ultra SCSI, and auto request sense. The cadp driver
does not support linked commands.
Preconfigure ■ The Plug N Play SCAM Support option is not supported.
■ If the BIOS is enabled on the card, ensure that the Adaptec SCSISelect BIOS option Reset
SCSI Bus at IC Initialization (under the Advanced Configuration Options menu) is set to
Enabled. Run the SCSISelect utility by pressing Ctrl-A when you see the Adaptec banner
during system boot.
■ If the adapter is being used in a multi-initiator configuration, do the following: (1) Ensure
that the system boot disk is not on the shared (clustered) bus. (2) Set the Reset SCSI Bus at
IC Initialization option to Disabled. (3) Set the Host Adapter BIOS option (under the
Advanced Configuration Options menu) to Disabled:Not scan. (4) Add the
allow-bus-reset=0 property to the /kernel/drv/cadp.conf file.
■ Reboot the system after you install patches.
Known Problems and ■ Running the format(1M) command on a Seagate ST19171W 9 GB disk drive fails.
Limitations
■ Some motherboards may have problems supporting channel B with boards based on the
Adaptec AIC-7896 chip. The problem arises because the BIOS doesn't properly assign two
interrupts for PCI interrupts INTA and INTB on the slot containing the AIC-7896 chip.
As a result, timeouts and resets on those devices appear on the console. For some
motherboards, you can work around the problem by setting the Advanced/PCI IRQ
Mapping feature to ISA Legacy IRQs.
■ If you experience problems when using a narrow SCSI CD-ROM drive on the internal
wide interface, disable "negotiate wide," "negotiate sync," or both for that device in the
Adaptec configuration utility.
■ The Fujitsu narrow disk (M1603SAU) can reselect with an invalid queue tag ID. This
violates the SCSI protocol and it causes the cadp driver to behave erroneously. Because this
is difficult to guard against, you should disable tagged queuing for these targets. Use the
iostat --E command to determine if you have a Fujitsu M1603S-512 disk. If you do, edit
the /kernel/drv/cadp.conf file and add the property targetn-scsi-options=0x1f78,
where n is the target number.
■ The IBM external wide disk (DFHSS2W, Revision 1717) is not supported.

132 man pages section 7: Device and Network Interfaces • Last Revised 23 May 2006
 cadp(7D)

■ When setting up a SCSI bus configuration, avoid connecting wide devices to a narrow bus.
However, if you have such a configuration, add the following entry to the cadp.conf file:
targetn-scsi-options=0x1df8 where n is the target ID of the wide device on the narrow
bus. This entry disables wide negotiation for the specified target. Also ensure that the
upper 8 bits of the bus are properly terminated at both ends of the SCSI chain.
■ If you experience installation problems on systems with Intel 440BX/440GX
motherboards, upgrade the motherboard BIOS with the latest revision.

Configuration You configure the cadp host bus adapter driver by defining the properties found in
cadp.conf. The cadp.conf file contains properties that you can modify, including:
scsi-options, target<n>-scsi-options, scsi-reset-delay, and scsi-initiator-id.
Properties in the cadp.conf file override global SCSI settings.

The property target<n>-scsi-options overrides the scsi-options property value for

target<n>, where <n> can vary from decimal 0 to 15. The cadp driver supports the following
scsi-options: SCSI_OPTIONS_DR, SCSI_OPTIONS_SYNC, SCSI_OPTIONS_TAG,
SCSI_OPTIONS_FAST, SCSI_OPTIONS_WIDE, SCSI_OPTIONS_FAST20, and
SCSI_OPTIONS_FAST40.

You configure the SCSI devices using the Adaptec configuration utility. When configuring the
devices, you should observe the following guidelines:
■ Configure each device using a unique SCSI ID. On the Advanced Configuration Options
menu, set Plug N Play SCAM Support to Disabled. Ensure that devices on either end of
the SCSI chain are terminated. When mixing wide (16 bits) and narrow (8 bits) devices on
the same wide chain, ensure that a wide device is at the end of the chain. If you place a
narrow device at the end of the chain, wide devices on the same chain will terminate the
low byte, resulting in a illegal configuration.
■ If there is more than one controller, or an embedded controller, attempt to use one IRQ
per controller.
■ When prompted, enable bus mastering for the slot(s) with your host bus adapter(s.)
■ Enable support for disks larger than 1 Gbyte, if applicable.

Examples Create a file called /kernel/drv/cadp.conf, then add the following line:
scsi-options=0x78;

The above line disables tagged queuing, Fast/Ultra SCSI, and wide mode for all cadp instances.

To set scsi-options more specifically per target, add the following lines to
/kernel/drv/cadp.conf:

target1-scsi-options=0x78;
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;

Device and Network Interfaces 133

cadp(7D)

seagate-scsi-options = 0x58;
scsi-options=0x3f8;

With the exception of one disk type that has scsi-options set to 0x58, the above example sets
scsi-options for target 1 to 0x78, and all remaining targets to 0x3f8.

The scsi-options properties that are specified per target ID have the highest precedence,
followed by scsi-options per device type. Global scsi-options for all cadp instances per
bus have the lowest precedence. You must reboot the system for the specified scsi options to
take effect.

Driver Capabilities To enable certain features on the cadp driver, the target driver must set capabilities. The
following capabilities can be queried and modified by the target driver: synchronous,
tagged-qing, wide-xfer, auto-rqsense, qfull-retries, and qfull-retry-interval. All
other capabilities are query only.

By default, the tagged-qing, auto-rqsense, and wide-xfer capabilities are disabled. The
disconnect, synchronous, and untagged-qing capabilities are always enabled. The cadp
driver capabilities can only be assigned binary values (0 or 1). The default value for
qfull-retries is 10 and the default value for qfull-retry-interval is 100. The
qfull-retries capability is au_char (0 to 255) while qfull-retry-interval is a u_short (0
to 65535).

If a conflict occurs between the value of scsi-options and a capability, the value set in
scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F)call. See
scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.
Files /kernel/drv/cadp ELF kernel module
/kernel/drv/cadp.conf Optional configuration file

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also prtconf(1M), driver.conf(4), pci(4), attributes(5), scsi_abort(9F),

scsi_hba_attach(9F), scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F),
scsi_sync_pkt(9F), scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S),
scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

Solaris (Intel Platform Edition) Hardware Compatibility List

ANSI Small Computer System Interface-2 (SCSI-2)

134 man pages section 7: Device and Network Interfaces • Last Revised 23 May 2006
 cadp(7D)

Notes The cadp driver supports the adapters and chipsets listed in this man page. For information
on support of additional devices, see the Solaris (Intel Platform Edition) Hardware
Compatibility List a component of the Information Library for Solaris 8 (Intel Platform
Edition).

The cadp driver exports properties indicating (per target) the negotiated transfer speed
(target<n>-sync-speed), whether wide bus (target<n>-wide), is supported for that
particular target (target<n>-scsi-options), and whether tagged queuing
(target<n>-tag-queue) has been enabled. The sync-speed property value is the data transfer
rate in KB/sec. The target<n>-tag-queue and the target<n>-wide property have value 1 to
indicate that the corresponding capability is enabled, or 0 to indicate that the capability is
disabled. See prtconf(1M) (verbose option) for information on viewing the cadp properties.

Sample output is provided below:

pci9005,f500, instance #2
System software properties:
name <interrupt-priorities> length <4>
value <0x05000000>.
name <tape> length <5>
value <0x7363747000>.
name <disk> length <5>
value <0x7363646b00>.
name <queue> length <6>
value <0x71736f727400>.
name <flow_control> length <6>
value <0x646d756c7400>.
Driver properties:
name <target0-tag-queue> length <4>
value <0x01000000>.
name <target0-wide> length <4>
value <0x01000000>.
name <target0-sync-speed> length <4>
value <0x28000000>.
name <chosen-interrupt> length <8>
value <0x0100000000000000>.
name <scsi-selection-timeout> length <4>
value <0xfa000000>.
name <scsi-options> length <4>
value <0xf81f0000>.
name <scsi-watchdog-tick> length <4>
value <0x0a000000>.
name <scsi-tag-age-limit> length <4>
value <0x02000000>.
name <scsi-reset-delay> length <4>
value <0xb80b0000>.

Device and Network Interfaces 135

cdio(7I)

Name cdio – CD-ROM control operations

Synopsis #include <sys/cdio.h>
Description The set of ioctl(2) commands described below are used to perform audio and CD-ROM specific
operations. Basic to these cdio ioctl requests are the definitions in <sys/cdio.h>.

Several CD-ROM specific commands can report addresses either in LBA (Logical Block Address)
format or in MSF (Minute, Second, Frame) format. The READ HEADER, READ SUBCHANNEL, and
READ TABLE OF CONTENTS commands have this feature.
LBA format represents the logical block address for the CD-ROM absolute address field or for
the offset from the beginning of the current track expressed as a number of logical blocks in a
CD-ROM track relative address field. MSF format represents the physical address written on
CD-ROM discs, expressed as a sector count relative to either the beginning of the medium or the
beginning of the current track.
ioctls The following I/O controls do not have any additional data passed into or received from them.
CDROMSTART This ioctl() spins up the disc and seeks to the last address requested.
CDROMSTOP This ioctl() spins down the disc.
CDROMPAUSE This ioctl() pauses the current audio play operation.
CDROMRESUME This ioctl() resumes the paused audio play operation.
CDROMEJECT This ioctl() ejects the caddy with the disc.
CDROMCLOSETRAY This ioctl() closes the caddy with the disc.

The following I/O controls require a pointer to the structure for that ioctl(), with data being
passed into the ioctl().
CDROMPLAYMSF This ioctl() command requests the drive to output the audio signals
at the specified starting address and continue the audio play until the
specified ending address is detected. The address is in MSF format. The
third argument of this ioctl() call is a pointer to the type struct
cdrom_msf.
/*
* definition of play audio msf structure
*/
struct cdrom_msf {
unsigned char cdmsf_min0; /* starting minute*/
unsigned char cdmsf_sec0; /* starting second*/
unsigned char cdmsf_frame0; /*starting frame*/
unsigned char cdmsf_min1; /* ending minute */
unsigned char cdmsf_sec1; /* ending second */
unsigned char cdmsf_frame1; /* ending frame */
};

136 man pages section 7: Device and Network Interfaces • Last Revised 4 Oct 2001
 cdio(7I)

The CDROMREADTOCENTRY ioctl request may be used to obtain the start

time for a track. An approximation of the finish time can be obtained
by using the CDROMREADTOCENTRY ioctl request to retrieve the start time
of the track following the current track.

The leadout track is the next consecutive track after the last audio track.
Hence, the start time of the leadout track may be used as the effective
finish time of the last audio track.
CDROMPLAYTRKIND This ioctl() command is similar to CDROMPLAYMSF. The starting and
ending address is in track/index format. The third argument of the
ioctl() call is a pointer to the type struct cdrom_ti.
/*
* definition of play audio track/index structure
*/
struct cdrom_ti {
unsigned char cdti_trk0; /* starting track*/
unsigned char cdti_ind0; /* starting index*/
unsigned char cdti_trk1; /* ending track */
unsigned char cdti_ind1; /* ending index */
};
CDROMVOLCTRL This ioctl() command controls the audio output level. The SCSI
command allows the control of up to four channels. The current
implementation of the supported CD-ROM drive only uses channel 0 and
channel 1. The valid values of volume control are between 0x00 and
0xFF, with a value of 0xFF indicating maximum volume. The third
argument of the ioctl() call is a pointer to struct cdrom_volctrl
which contains the output volume values.
/*
* definition of audio volume control structure
*/
struct cdrom_volctrl {
unsigned char channel0;
unsigned char channel1;
unsigned char channel2;
unsigned char channel3;
};

The following I/O controls take a pointer that will have data returned to the user program
from the CD-ROM driver.
CDROMREADTOCHDR This ioctl() command returns the header of the table of contents
(TOC). The header consists of the starting tracking number and the
ending track number of the disc. These two numbers are returned

Device and Network Interfaces 137

cdio(7I)

through a pointer of struct cdrom_tochdr. While the disc can start

at any number, all tracks between the first and last tracks are in
contiguous ascending order.
/*
* definition of read toc header structure
*/
struct cdrom_tochdr {
unsigned char cdth_trk0; /* starting track*/
unsigned char cdth_trk1; /* ending track*/
};
CDROMREADTOCENTRY This ioctl() command returns the information of a specified track.
The third argument of the function call is a pointer to the type
struct cdrom_tocentry. The caller needs to supply the track
number and the address format. This command will return a 4-bit
adr field, a 4-bit ctrl field, the starting address in MSF format or
LBA format, and the data mode if the track is a data track. The ctrl
field specifies whether the track is data or audio.
/*
* definition of read toc entry structure
*/
struct cdrom_tocentry {
unsigned char cdte_track;
unsigned char cdte_adr :4;
unsigned char cdte_ctrl :4;
unsigned char cdte_format;
union {
struct {
unsigned char minute;
unsigned char second;
unsigned char frame;
} msf;
int lba;
} cdte_addr;
unsigned char cdte_datamode;
};
To get the information from the leadout track, the following value is
appropriate for the cdte_track field:
CDROM_LEADOUT Leadout track

To get the information from the data track, the following value is
appropriate for the cdte_ctrl field:
CDROM_DATA_TRACK Data track

138 man pages section 7: Device and Network Interfaces • Last Revised 4 Oct 2001
 cdio(7I)

The following values are appropriate for the cdte_format field:

CDROM_LBA LBA format
CDROM_MSF MSF format
CDROMSUBCHNL This ioctl() command reads the Q sub-channel data of the current
block. The subchannel data includes track number, index number,
absolute CD-ROM address, track relative CD-ROM address, control data
and audio status. All information is returned through a pointer to
struct cdrom_subchnl. The caller needs to supply the address
format for the returned address.
struct cdrom_subchnl {
unsigned char cdsc_format;
unsigned char cdsc_audiostatus;
unsigned char cdsc_adr: 4;
unsigned char cdsc_ctrl: 4;
unsigned char cdsc_trk;
unsigned char cdsc_ind;
union {
struct {
unsigned char minute;
unsigned char second;
unsigned char frame;
} msf;
int lba;
} cdsc_absaddr;
union {
struct {
unsigned char minute;
unsigned char second;
unsigned char frame;
} msf;
int lba;
} cdsc_reladdr;
};
The following values are valid for the audio status field returned
from READ SUBCHANNEL command:
CDROM_AUDIO_INVALID Audio status not supported.
CDROM_AUDIO_PLAY Audio play operation in progress.
CDROM_AUDIO_PAUSED Audio play operation paused.
CDROM_AUDIO_COMPLETED Audio play successfully completed.
CDROM_AUDIO_ERROR Audio play stopped due to error.

Device and Network Interfaces 139

cdio(7I)

CDROM_AUDIO_NO_STATUS No current audio status to return.

CDROMREADOFFSET This ioctl() command returns the absolute CD-ROM address of the
first track in the last session of a Multi-Session CD-ROM. The third
argument of the ioctl() call is a pointer to an int.
CDROMCDDA This ioctl() command returns the CD-DA data or the subcode data.
The third argument of the ioctl() call is a pointer to the type
struct cdrom_cdda. In addition to allocating memory and
supplying its address, the caller needs to supply the starting address
of the data, the transfer length in terms of the number of blocks to be
transferred, and the subcode options. The caller also needs to issue
the CDROMREADTOCENTRY ioctl() to find out which tracks contain
CD-DA data before issuing this ioctl().
/*
* Definition of CD-DA structure
*/
struct cdrom_cdda {
unsigned int cdda_addr;
unsigned int cdda_length;
caddr_t cdda_data;
unsigned char cdda_subcode;
};
cdda_addr signifies the starting logical block address.

cdda_length signifies the transfer length in blocks. The length of the

block depends on the cdda_subcode selection, which is explained
below.

To get the subcode information related to CD-DA data, the following

values are appropriate for the cdda_subcode field:
CDROM_DA_NO_SUBCODE CD-DA data with no subcode.
CDROM_DA_SUBQ CD-DA data with sub Q code.
CDROM_DA_ALL_SUBCODE CD-DA data with all subcode.
CDROM_DA_SUBCODE_ONLY All subcode only.

To allocate the memory related to CD-DA and/or subcode data, the

following values are appropriate for each data block transferred:
CD-DA data with no subcode 2352 bytes
CD-DA data with sub Q code 2368 bytes
CD-DA data with all subcode 2448 bytes

140 man pages section 7: Device and Network Interfaces • Last Revised 4 Oct 2001
 cdio(7I)

All subcode only 96 bytes

CDROMCDXA This ioctl() command returns the CD-ROM XA (CD-ROM Extended
Architecture) data according to CD-ROM XA format. The third
argument of the ioctl() call is a pointer to the type struct
cdrom_cdxa. In addition to allocating memory and supplying its
address, the caller needs to supply the starting address of the data,
the transfer length in terms of number of blocks, and the format. The
caller also needs to issue the CDROMREADTOCENTRY ioctl() to find out
which tracks contain CD-ROM XA data before issuing this ioctl().
/*
* Definition of CD-ROM XA structure
*/
struct cdrom_cdxa {
unsigned int cdxa_addr;
unsigned int cdxa_length;
caddr_t cdxa_data;
unsigned char cdxa_format;
};
To get the proper CD-ROM XA data, the following values are
appropriate for the cdxa_format field:
CDROM_XA_DATA CD-ROM XA data only
CDROM_XA_SECTOR_DATA CD-ROM XA all sector data
CDROM_XA_DATA_W_ERROR CD-ROM XA data with error flags data

To allocate the memory related to CD-ROM XA format, the following

values are appropriate for each data block transferred:
CD-ROM XA data only 2048 bytes
CD-ROM XA all sector data 2352 bytes
CD-ROM XA data with error flags data 2646 bytes
CDROMSUBCODE This ioctl() command returns raw subcode data (subcodes P ~ W
are described in the "Red Book," see SEE ALSO) to the initiator while
the target is playing audio. The third argument of the ioctl() call is
a pointer to the type struct cdrom_subcode. The caller needs to
supply the transfer length in terms of number of blocks and allocate
memory for subcode data. The memory allocated should be a
multiple of 96 bytes depending on the transfer length.
/*
* Definition of subcode structure
*/

Device and Network Interfaces 141

cdio(7I)

struct cdrom_subcode {
unsigned int cdsc_length;
caddr_t cdsc_addr;
};
The next group of I/O controls get and set various CD-ROM drive parameters.
CDROMGBLKMODE This ioctl() command returns the current block size used by the
CD-ROM drive. The third argument of the ioctl() call is a pointer to an
integer.
CDROMSBLKMODE This ioctl() command requests the CD-ROM drive to change from the
current block size to the requested block size. The third argument of the
ioctl() call is an integer which contains the requested block size.

This ioctl() command operates in exclusive-use mode only. The caller

must ensure that no other processes can operate on the same CD-ROM
device before issuing this ioctl(). read(2) behavior subsequent to this
ioctl() remains the same: the caller is still constrained to read the raw
device on block boundaries and in block multiples.

To set the proper block size, the following values are appropriate:
CDROM_BLK_512 512 bytes
CDROM_BLK_1024 1024 bytes
CDROM_BLK_2048 2048 bytes
CDROM_BLK_2056 2056 bytes
CDROM_BLK_2336 2336 bytes
CDROM_BLK_2340 2340 bytes
CDROM_BLK_2352 2352 bytes
CDROM_BLK_2368 2368 bytes
CDROM_BLK_2448 2448 bytes
CDROM_BLK_2646 2646 bytes
CDROM_BLK_2647 2647 bytes
CDROMGDRVSPEED This ioctl() command returns the current CD-ROM drive speed. The
third argument of the ioctl() call is a pointer to an integer.
CDROMSDRVSPEED This ioctl() command requests the CD-ROM drive to change the current
drive speed to the requested drive speed. This speed setting is only
applicable when reading data areas. The third argument of the ioctl()
is an integer which contains the requested drive speed.

142 man pages section 7: Device and Network Interfaces • Last Revised 4 Oct 2001
 cdio(7I)

To set the CD-ROM drive to the proper speed, the following values are
appropriate:
CDROM_NORMAL_SPEED 150k/second
CDROM_DOUBLE_SPEED 300k/second
CDROM_QUAD_SPEED 600k/second
CDROM_MAXIMUM_SPEED 300k/second (2x drive) 600k/second (4x
drive)

Note that these numbers are only accurate when reading 2048 byte
blocks. The CD-ROM drive will automatically switch to normal speed
when playing audio tracks and will switch back to the speed setting when
accessing data.

See Also ioctl(2), read(2)

N. V. Phillips and Sony Corporation, System Description Compact Disc Digital Audio, ("Red
Book").

N. V. Phillips and Sony Corporation, System Description of Compact Disc Read Only Memory,
("Yellow Book").

N. V. Phillips, Microsoft, and Sony Corporation, System Description CD-ROM XA, 1991.

Volume and File Structure of CD-ROM for Information Interchange, ISO 9660:1988(E).

SCSI-2 Standard, document X3T9.2/86-109

SCSI Multimedia Commands, Version 2 (MMC-2)

Notes The CDROMCDDA, CDROMCDXA, CDROMSUBCODE, CDROMGDRVSPEED, CDROMSDRVSPEED, and some of

the block sizes in CDROMSBLKMODE are designed for new Sun-supported CD-ROM drives and
might not work on some of the older CD-ROM drives.

CDROMCDDA, CDROMCDXA and CDROMSUBCODE will return error if the transfer

length exceeds valid limits as determined appropriate. Example: for MMC-2 drives, length can
not exceed 3 bytes (i.e. 0xffffff). The same restriction is enforced for older, pre-MMC-2 drives,
as no limit was published for these older drives (and 3 bytes is reasonable for all media). Note
that enforcing this limit does not imply that values passed in below this limit will actually be
applicable for each and every piece of media.

The interface to this device is preliminary and subject to change in future releases. Programs
should be written in a modular fashion so that future changes can be easily incorporated.

Device and Network Interfaces 143

ce(7D)

Name ce – Cassini Gigabit-Ethernet device driver

Synopsis /dev/ce

Description The ce Sun Gigabit-Ethernet driver is a multi-threaded, loadable, clonable, STREAMS

hardware driver supporting the connectionless Data Link Provider Interface, dlpi(7P), over
all implementations of PCI Cassini Gigabit-Ethernet add-in adapters. Multiple Cassini-based
adapters installed within the system are supported by the driver. The ce driver provides basic
support for the Cassini-based Ethernet hardware and handles the pci108e,abba (PCI Cassini)
devices. Functions include chip initialization, frame transmit and receive, multicast and
promiscuous support, and error recovery and reporting. The Cassini device provides
1000BASE-SX networking interfaces using the Cassini ASIC external SERDES and fiber
optical transceiver, or 10/100/1000BASE-T using a Cassini ASIC attached to a GMII twisted
pair copper transceiver, or 10/100BASE-T using a Cassini ASIC attached to a MII twisted pair
copper transceiver.

The 1000Base-SX standard specifies an auto-negotiation protocol to automatically select the

mode of operation. In addition to the duplex mode of operation, the Cassini ASIC can
auto-negotiate for IEEE 802.3x frame-based flow control capabilities. The Cassini PCS can
perform auto-negotiation with the link's remote-end (link partner) and receives the
capabilities of the remote end. It selects the highest common denominator mode of operation
based on the priorities. It also supports forced-mode of operation where the driver selects the
mode of operation.

Application The /dev/ce cloning character-special device is used to access all ce controllers installed on
Programming the system.
Interface
ce and DLPI The ce driver is a Style 2 data link service provider. All M_PROTO and M_PCPROTO type messages
are interpreted as DLPI primitives. Valid DLPI primitives are defined in <sys/dlpi.h>. Refer
to dlpi(7P) for more information. An explicit DL_ATTACH_REQ message by the user is required
to associate the opened stream with a particular device (ppa). The ppa ID is interpreted as an
unsigned long data type and indicates the corresponding device instance (unit) number. An
error (DL_ERROR_ACK) is returned by the driver if the ppa field value does not correspond to a
valid device instance number for this system. The device is initialized on first attach and
de-initialized (stopped) upon last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are as follows:
■ Maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h>).
■ Minimum SDU is 0.
■ The dlsap address length is 8.
■ MAC type is DL_ETHER.

144 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 ce(7D)

■ The sap length value is –2 meaning the physical address component is followed
immediately by a 2 byte sap component within the DLSAP address.
■ Service mode is DL_CLDLS.
■ Optional quality of service (QOS) is not supported; the QOS fields are 0.
■ Provider style is DL_STYLE2.
■ Version is DL_VERSION_2.
■ Broadcast address value is Ethernet/IEEE broadcast address (0xFFFFFF.)

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular SAP
(Service Access Pointer) with the stream. The ce driver interprets the sap field within the
DL_BIND_REQ as an Ethernet “type,” therefore valid values for the sap field are in the range
[0-0xFFFF]. Only one Ethernet type can be bound to the stream at any time.

If you select a sap with a value of 0, the receiver will be in 802.3 mode. All frames received
from the media having a “type” field in the range [0-1500] are assumed to be 802.3 frames and
are routed up all open streams which are bound to sap value 0 . If more than one stream is in
802.3 mode, the frame will be duplicated and routed up multiple streams as DL_UNITDATA_IND
messages.

In transmission, the driver checks the sap field of the DL_BIND_REQ to verify that the sap value
is 0, and that the destination type field is in the range [0-1500]. If either is true, the driver
computes the length of the message, not including initial M_PROTO mblk (message block), of all
subsequent DL_UNITDATA_REQ messages and transmits 802.3 frames that have this value in the
MAC frame header length field.

The ce driver DLSAP address format consists of the 6 byte physical (Ethernet) address
component followed immediately by the 2 byte sap (type) component producing an 8 byte
DLSAP address. Applications should not hard code to this particular implementation-specific
DLSAP address format, but use information returned in the DL_INFO_ACK primitive to
compose and decompose DLSAP addresses. The sap length, full DLSAP length, and
sap/physical ordering are included within the DL_INFO_ACK. The physical address length can
be computed by subtracting the sap length from the full DLSAP address length or by issuing
the DL_PHYS_ADDR_REQ to obtain the current physical address associated with the stream.

Once in the DL_BOUND state, you can transmit frames on the Ethernet by sending
DL_UNITDATA_REQ messages to the ce driver. The ce driver will route received Ethernet frames
up all open and bound streams having a sap which matches the Ethernet type as
DL_UNITDATA_IND messages. Received Ethernet frames are duplicated and routed up multiple
open streams, if necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.

Device and Network Interfaces 145

ce(7D)

In addition to the mandatory connectionless DLPI message set, the driver additionally
supports the following primitives.

ce Primitives The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable/disable reception of

individual multicast group addresses. A set of multicast addresses may be iteratively created
and modified on a per-stream basis using these primitives. These primitives are accepted by
the driver in any state following DL_ATTACHED.

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives with the DL_PROMISC_PHYS

option set in the dl_level field enables/disables reception of all “promiscuous mode” frames
on the media, including frames generated by the local host. When used with the
DL_PROMISC_SAP option set, this enables/disables reception of all sap (Ethernet type) values.
When used with the DL_PROMISC_MULTI option set this enables/disables reception of all
multicast group addresses. The effect of each is always on a per-stream basis and independent
of the other sap and physical level configurations on this stream or other streams.

The DL_PHYS_ADDR_REQ primitive returns the 6 octet Ethernet address currently associated
(attached) to the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the 6 octet Ethernet address currently

associated (attached) to this stream. The credentials of the process which originally opened
this stream must be superuser. Otherwise EPERM is returned in the DL_ERROR_ACK. This
primitive is destructive because it affects all other current and future streams attached to this
device. An M_ERROR is sent up all other streams attached to this device when this primitive is
successful on this stream. Once changed, all streams subsequently opened and attached to this
device will obtain this new physical address. Once changed, the physical address will remain
until this primitive is used to change the physical address again or the system is rebooted,
whichever comes first.

ce DRIVER By default, the ce driver performs auto-negotiation to select the mode and flow control
capabilities of the link.

The link can assume one of the following modes:

■ 1000 Mbps, full-duplex
■ 1000 Mbps, half-duplex
■ Symmetric pause
■ Asymmetric pause

Speeds and modes are described in the 1000Base-TX standard.

The auto–negotiation protocol automatically selects:

■ Operation mode (half-duplex or full-duplex)
■ Flow control capability (symmetric and/or asymmetric)

146 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 ce(7D)

The auto–negotiation protocol does the following:

■ Gets all modes of operation supported by the link partner.
■ Advertises its capabilities to the link partner.
■ Selects the highest common denominator mode of operation based on the priorities.

The Cassini hardware can operate in all modes listed above, providing auto-negotiation is
used by default to bring up the link and select the common mode of operation with the link
partner. The PCS also supports forced-mode of operation in which the driver can select the
mode of operation and the flow control capabilities, using the ndd utility.

The Cassini device also supports programmable IPG (Inter-Packet Gap) parameters ipg1 and
ipg2. By default, the driver sets ipg1 and ipg2 to 8 and 4 byte-times respectively (which are the
standard values). If desired, you can alter these values from the standard 1000 Mpbs IPG set to
0.096 microseconds.

ce Parameter List The ce driver enables the setting and getting of various parameters for the Cassini device. The
parameter list includes current transceiver status, current link status, inter-packet gap,
PCS capabilities and link partner capabilities.

The PCS features two set of capabilities. One set reflects the capabilities of the hardware and
are read-only. The second set, which reflects the values you choose, are used in speed selection
and possess read/write capabilities. At boot time, these two sets of capabilities are the same.
The link partner capabilities are also read-only because the current default value of these
parameters can be read but not modified.
Files /dev/ce ce special character device.
/platform/sun4u/kernel/drv/ce.conf SPARC system-wide default device driver
properties.
/platform/amd64/kernel/drv/ce.conf 64–bit x86 system-wide default device driver
properties.
/kernel/drv/ce.conf 32–bit x86 system-wide default device driver
properties.
/kernel/drv/amd64/ce.conf 64–bit x86 system-wide default device driver
properties.

See Also ndd(1M), netstat(1M), driver.conf(4), ge(7D), hme(7D), qfe(7d), dlpi(7P)

Device and Network Interfaces 147

chxge(7D)

Name chxge – Chelsio Ethernet network interface controllers

Synopsis /dev/chxge

Description The chxge Ethernet driver is a multi-threaded, loadable, clonable, STREAMS hardware driver
supporting the connectionless Data Link Provider Interface, dlpi(7P), over Chelsio NIC
controllers. Multiple (and mixed) NIC controllers installed within the system are supported
by the driver. The chxge driver provides basic support for the NIC hardware. Functions
include chip initialization, frame transmit and receive, and error recovery and reporting.

Application The cloning, character-special device /dev/chxge is used to access NIC devices installed
Programming within the system.
Interface
chxge and Dlpi The chxge driver is dependent on /kernel/misc/gld, a loadable kernel module that provides
the chxge driver with the DLPI and STREAMS functionality required of a LAN driver. See
gld(7D) for more details on the primitives supported by the driver.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are:
■ Default Maximum SDU is 1500 (ETHERMTU).
■ dlsap address length is 8.
■ MAC type is DL_ETHER.
■ The sap length value is -2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Files /dev/chxge Character special device.
/kernel/drv/sparcv9/chxge SPARC chxge driver binary.
/kernel/drv/chxge x86 platform kernel module. (32-bit).
/kernel/drv/amd64/chxge x86 platform kernel module. (64-bit).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also netstat(1M), attributes(5), gld(7D), dlpi(7P), gld(9F), gld_mac_info(9S)

148 man pages section 7: Device and Network Interfaces • Last Revised 17 Jan 2005
 cmdk(7D)

Name cmdk – common disk driver

Synopsis cmdk@target, lun : [ partition | slice ]

Description The cmdk device driver is a common interface to various disk devices. The driver supports
magnetic fixed disks and magnetic removable disks.

The cmdk device driver supports three different disk labels: fdisk partition table, Solaris x86
VTOC and EFI/GPT.

The block-files access the disk using the system's normal buffering mechanism and are read
and written without regard to physical disk records. There is also a "raw" interface that
provides for direct transmission between the disk and the user's read or write buffer. A single
read or write call usually results in one I/O operation; raw I/O is therefore considerably more
efficient when many bytes are transmitted. The names of the block files are found in /dev/dsk.
Raw file names are found in /dev/rdsk.

I/O requests to the magnetic disk must have an offset and transfer length that is a multiple of
512 bytes or the driver returns an EINVAL error.

Slice 0 is normally used for the root file system on a disk, slice 1 as a paging area (for example,
swap), and slice 2 for backing up the entire fdisk partition for Solaris software. Other slices
may be used for usr file systems or system reserved area.

The fdisk partition 0 is to access the entire disk and is generally used by the fdisk(1M)
program.
Files /dev/dsk/cndn[s|p]n block device (IDE)
/dev/rdsk/cndn[s|p]n raw device (IDE)

where:
cn controller n.
dn lun n (0-1).
sn UNIX system slice n (0-15).
pn fdisk partition (0–36).
/kernel/drv/cmdk 32-bit kernel module.
/kernel/drv/amd64/cmdk 64-bit kernel module.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Device and Network Interfaces 149

cmdk(7D)

See Also fdisk(1M), mount(1M), lseek(2), read(2), write(2), readdir(3C), scsi(4), vfstab(4),
attributes(5), dkio(7I)

150 man pages section 7: Device and Network Interfaces • Last Revised 4 Nov 2008
 connld(7M)

Name connld – line discipline for unique stream connections

Synopsis #include </sys/steam.h>
int ioctl(fd,I_PUSH,"connld");

Description connld is a STREAMS-based module that provides unique connections between server and
client processes. It can only be pushed (see streamio(7I)) onto one end of a STREAMS-based
pipe that may subsequently be attached to a name in the file system name space with
fattach(3C). After the pipe end is attached, a new pipe is created internally when an
originating process attempts to open(2) or creat(2) the file system name. A file descriptor for
one end of the new pipe is packaged into a message identical to that for the ioctl I_SENDFD (see
streamio(7I)) and is transmitted along the stream to the server process on the other end. The
originating process is blocked until the server responds.

The server responds to the I_SENDFD request by accepting the file descriptor through the
I_RECVFD ioctl message. When this happens, the file descriptor associated with the other end
of the new pipe is transmitted to the originating process as the file descriptor returned from
open(2) or creat(2).

If the server does not respond to the I_SENDFD request, the stream that the connld module is
pushed on becomes uni-directional because the server will not be able to retrieve any data off
the stream until the I_RECVFD request is issued. If the server process exits before issuing the
I_RECVFD request, the open(2) or the creat(2) invocation will fail and return -1 to the
originating process.

When the connld module is pushed onto a pipe, it ignores messages going back and forth
through the pipe.

Errors On success, an open of connld returns 0. On failure, errno is set to the following values:
EINVAL A stream onto which connld is being pushed is not a pipe or the pipe does not
have a write queue pointer pointing to a stream head read queue.
EINVAL The other end of the pipe onto which connld is being pushed is linked under a
multiplexor.
EPIPE connld is being pushed onto a pipe end whose other end is no longer there.
ENOMEM An internal pipe could not be created.
ENXIO An M_HANGUP message is at the stream head of the pipe onto which connld is being
pushed.
EAGAIN Internal data structures could not be allocated.
ENFILE A file table entry could not be allocated.

Device and Network Interfaces 151

connld(7M)

See Also creat(2), open(2), fattach(3C), streamio(7I)

STREAMS Programming Guide

152 man pages section 7: Device and Network Interfaces • Last Revised 3 May 2004
 console(7D)

Name console – STREAMS-based console interface

Synopsis /dev/console

Description The file /dev/console refers to the system console device. /dev/console should be used for
interactive purposes only. Use of /dev/console for logging purposes is discouraged;
syslog(3C) or msglog(7D) should be used instead.

The identity of this device depends on the EEPROM or NVRAM settings in effect at the most
recent system reboot; by default, it is the ‘‘workstation console'' device consisting of the
workstation keyboard and frame buffer acting in concert to emulate an ASCII terminal (see
wscons(7D)).

Regardless of the system configuration, the console device provides asynchronous serial driver
semantics so that, in conjunction with the STREAMS line discipline module ldterm(7M), it
supports the termio(7I) terminal interface.

See Also syslog(3C), termios(3C), ldterm(7M), termio(7I), msglog(7D), wscons(7D)

Notes In contrast to pre-SunOS 5.0 releases, it is no longer possible to redirect I/O intended for
/dev/console to some other device. Instead, redirection now applies to the workstation
console device using a revised programming interface (see wscons(7D)). Since the system
console is normally configured to be the work station console, the overall effect is largely
unchanged from previous releases.

See wscons(7D) for detailed descriptions of control sequence syntax, ANSI control functions,
control character functions and escape sequence functions.

Device and Network Interfaces 153

cpqary3(7D)

Name cpqary3 – provides disk and SCSI tape support for HP Smart Array controllers

Description The cpqary3 module provides low-level interface routines between the common disk I/O
subsystem and the HP SMART Array controllers. The cpqary3 driver provides disk and SCSI
tape support for the HP Smart Array controllers.

Please refer to the cpqary3 Release Notes, for the supported HP Smart Array Controllers and
Storage boxes.

Each of the controller should be the sole initiator on a SCSI bus. Auto configuration code
determines if the adapter is present at the Configured address and what types of devices are
attached to it.

Configuration Use the Array Configuration Utility to configure the controllers. Each controller can support
up to 32 logical volumes. In addition, each controller supports up to a maximum of 28
connected SCSI tape drives. With 1.90 and later versions of cpqary3 driver, HP Smart Array
SAS controllers, having Firmware Revision 5.10 or later, support 64 logical drives. This
firmware also supports Dual Domian Multipath configurations.

The driver attempts to initialize itself in accordance with the information found in the
configuration file, /kernel/drv/cpqary3.conf.

The target driver's configuration file need entries if support is needed for targets numbering
greater than the default number of targets supported by the corresponding target driver.

By default, entries for SCSI target numbers 0 to 15 are present in sd.conf. Entries for target
numbers 16 and above are added in SCSI class in the sd.conf file for supporting
corresponding logical volumes.

If SCSI tape drives are connected to the supported controllers, entries for target IDs from 33 to
33+n must be added in the /kernel/drv/st.conf file under scsi class, where n is the total
number of SCSI tape drives connected to the controller with largest number of tape drives
connected to it, in the existing configuration. For example, two supported controllers, c1 and
c2 are present in the system. If controller c1 has two tape drives and controller c2 has five tape
drives connected, entries for target IDs 33 through 38 are required under scsi class in
/kernel/drv/st.conf file. The maximum number of tape drives that can be connected to a
controller is 28. With 1.90 and later versions of the cpqary3 driver, if tape drives are connected
to the Smart Array SAS controllers, target ID entries for tape drives from 65 to 65+n must be
added in the /kernel/drv/st.conf file under the scsi class.
Files /kernel/drv/cpqary3.conf Configuration file for CPQary3
/kernel/drv/sd.conf Configuration file for sd
/kernel/drv/st.conf Configuration file for st
/dev/dsk Block special file names for disk device

154 man pages section 7: Device and Network Interfaces • Last Revised 4 Jan 2010
 cpqary3(7D)

/dev/rdsk Character special file names for disk device

/dev/rmt Special file names for SCSI tape devices

See Also driver.conf(4), sd(7D), st(7D)

cpqary3 Release Notes

Notes The Smart Array controllers supported by the current version of the cpqary3 driver do not
support format unit SCSI command. Therefore, selecting the format option under the
format utility main menu is not supported. In addition, the repair option under format
utility main menu is not supported as this operation is not applicable to Logical volumes
connected to the supported Smart Array controllers.

The names of the block files can be found in /dev/dsk. The names of the raw files can be found
in /dev/rdsk.

Device and Network Interfaces 155

cpr(7)

Name cpr – Suspend and resume module

Synopsis /platform/’uname -m’/kernel/misc/cpr

Description The cpr module is a loadable module used to suspend and resume the entire system. You may
wish to suspend a system to save power or to power off temporarily for transport. The cpr
module should not be used in place of a normal shutdown when performing any hardware
reconfiguration or replacement. In order for the resume operation to succeed, it is important
that the hardware configuration remain the same. When the system is suspended, the entire
system state is preserved in non-volatile storage until a resume operation is conducted.

dtpower(1M) or power.conf(4) are used to configure the suspend-resume feature.

The speed of suspend and resume operations can range from 15 seconds to several minutes,
depending on the system speed, memory size, and load.

During resume operation, the SIGTHAW signal is sent to all processes to allow them to do any
special processing in response to suspend-resume operation. Normally applications are not
required to do any special processing because of suspend-resume, but some specialized
processes can use SIGTHAW to restore the state prior to suspend. For example, X can refresh the
screen in response to SIGTHAW.

In some cases the cpr module may be unable to perform the suspend operation. If a system
contains additional devices outside the standard shipped configuration, it is possible that
device drivers for these additional devices might not support suspend-resume operations. In
this case, the suspend fails and an error message is displayed. These devices must be removed
or their device drivers unloaded for the suspend operation to succeed. Contact the device
manufacturer to obtain a new version of device driver that supports suspend-resume.

A suspend may also fail when devices or processes are performing critical or time-sensitive
operations (such as realtime operations). The system will remain in its current running state.
Messages reporting the failure will be displayed on the console and status returned to the
caller. Once the system is successfully suspended the resume operation will succeed, barring
external influences such as a hardware reconfiguration.

Some network-based applications may fail across a suspend and resume cycle. This largely
depends on the underlying network protocol and the applications involved. In general,
applications that retry and automatically reestablish connections will continue to operate
transparently on a resume operation; those applications that do not will likely fail.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcpr

156 man pages section 7: Device and Network Interfaces • Last Revised 7 May 2001
 cpr(7)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface stability Unstable

See Also dtpower(1M) (OpenWindows Reference Manual), pmconfig(1M), uadmin(1M), uadmin(2),

power.conf(4), attributes(5)

Using Power Management

Writing Device Drivers

Notes Certain device operations such as tape and floppy disk activities are not resumable due to the
nature of removable media. These activities are detected at suspend time, and must be stopped
before the suspend operation will complete successfully.

Suspend-resume is currently supported only on a limited set of hardware platforms. Please see
the book Using Power Management for a complete list of platforms that support system Power
Management. See uname(2) to programatically determine if the machine supports
suspend-resume.

Device and Network Interfaces 157

cpuid(7D)

Name cpuid – CPU identification driver

Synopsis /dev/cpu/self/cpuid

Description

SPARC and x86 system This device provides implementation-private information via ioctls about various aspects of
the implementation to Solaris libraries and utilities.

x86 systems only This device also provides a file-like view of the namespace and return values of the x86 cpuid
instruction. The cpuid instruction takes a single 32-bit integer function code, and returns four
32-bit integer values corresponding to the input value that describe various aspects of the
capabilities and configuration of the processor.

The API for the character device consists of using the seek offset to set the function code value,
and using a read(2) or pread(2) of 16 bytes to fetch the four 32-bit return values of the
instruction in the order %eax, %ebx, %ecx and %edx.

No data can be written to the device. Like the cpuid instruction, no special privileges are
required to use the device.

The device is useful to enable low-level configuration information to be extracted from the
CPU without having to write any assembler code to invoke the cpuid instruction directly. It
also allows the kernel to attempt to correct any erroneous data returned by the instruction
(prompted by occassional errors in the information exported by various processor
implementations over the years).

See the processor manufacturers documentation for further information about the syntax and
semantics of the wide variety of information available from this instruction.

Example This example allows you to determine if the current x86 processor supports "long mode,"
which is a necessary (but not sufficient) condition for running the 64-bit Solaris kernel on the
processor.
/*

#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
#include <string.h>
#include <errno.h>
#include <stdio.h>

static const char devname[] = "/dev/cpu/self/cpuid";

/*ARGSUSED*/
int
main(int argc, char *argv[])

158 man pages section 7: Device and Network Interfaces • Last Revised 16 Sep 2004
 cpuid(7D)

{
struct {
uint32_t r_eax, r_ebx, r_ecx, r_edx;
} _r, *rp = &_r;
int d;
char *s;

if ((d = open(devname, O_RDONLY)) == -1) {

perror(devname);
return (1);
}

if (pread(d, rp, sizeof (*rp), 0) != sizeof (*rp)) {

perror(devname);
goto fail;
}

s = (char *)&rp->r_ebx;
if (strncmp(s, "Auth" "cAMD" "enti", 12) != 0 &&
strncmp(s, "Genu" "ntel" "ineI", 12) != 0)
goto fail;

if (pread(d, rp, sizeof (*rp), 0x80000001) == sizeof (*rp)) {

/*
* Read extended feature word; check bit 29
*/
(void) close(d);
if ((rp->r_edx >> 29) & 1) {
(void) printf("processor supports long mode\n");
return (0);
}
}
fail:
(void) close(d);
return (1);
}
Errors ENXIO Results from attempting to read data from the device on a system that does not
support the CPU identification interfaces
EINVAL Results from reading from an offset larger than UINT_MAX, or attempting to
read with a size that is not multiple of 16 bytes.
Files /dev/cpu/self/cpuid Provides access to CPU identification data.

Attributes See attributes(5) for descriptions of the following attributes:

Device and Network Interfaces 159

cpuid(7D)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWckr

Interface Stability Evolving

See Also psrinfo(1M), prtconf(1M), pread(2), read(2), attributes(5)

160 man pages section 7: Device and Network Interfaces • Last Revised 16 Sep 2004
 ctfs(7FS)

Name ctfs – contract file system

Description The ctfs filesystem is the interface to the contract sub-system. ctfs is mounted during boot
at /system/contract. For information on contracts and the contents of this filesystem, see
contract(4).
Files /system/contract Mount point for the ctfs file system

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWckr

See Also contract(4), vfstab(4), attributes(5), smf(5)

Device and Network Interfaces 161

ctsmc(7D)

Name ctsmc – System Management Controller driver

Description The ctsmc system management controller driver is a multithreaded, loadable, clonable
STREAMS hardware driver that supports communication with the system management
controller device on SUNW,NetraCT-410, SUNW,NetraCT-810 and SUNW,Netra-CP2300
platforms.

The smc device provides a Keyboard Controller Style (KCS) interface as described in the
Intelligent Platform Management Interface (IPMI) Version 1.5 specification. The ctsmc driver
enables user-land and kernel-land clients to access services provided by smc hardware.
Files /dev/ctsmc ctsmc special character device
/platform/sun4u/kernel/drv/sparcv9/ctsmc 64 bit ELF kernel driver

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWcar.u

See Also attributes(5)

STREAMS Programmers Guide

Writing Device Drivers

Intelligent Platform Management Interface (IPMI). Version 1.5 – PICMIG, February, 2001

162 man pages section 7: Device and Network Interfaces • Last Revised 7 Dec 2003
 cvc(7D)

Name cvc – virtual console driver

Description The cvc virtual console driver is a STREAMS-based pseudo driver that supports the network
console. The cvc driver interfaces with console(7D).

Logically, the cvc driver sits below the console driver. It redirects console output to the
cvcredir(7D) driver if a network console connection is active. If a network console
connection is not active, it redirects console output to an internal hardware interface.

The cvc driver receives console input from cvcredir and internal hardware and passes it to
the process associated with /dev/console.

Notes The cvc facility supersedes the SunOS wscons(7D) facility, which should not be used in
conjunction with cvc. The wscons driver is useful for systems with directly attached consoles
(frame buffers and keyboards), but is not useful with platforms using cvc, which have no local
keyboard or frame buffer.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Sun Enterprise 10000 servers, Sun Fire 15000 servers

Availability SUNWcvc.u

See Also cvcd(1M), attributes(5), console(7D), cvcredir(7D), wscons(7D)

Sun Enterprise 10000 SSP Reference Manual

Sun System Management Services (SMS) Reference Manual

Device and Network Interfaces 163

cvcredir(7D)

Name cvcredir – virtual console redirection driver

Description The cvcredir virtual console redirection driver is a STREAMS-based pseudo driver that
supports the network console provided on some platforms. The cvcredir driver interfaces
with the virtual console driver cvc(7D), and the virtual console daemon, cvcd(1M).

The cvcredir driver receives console output from cvc and passes it to cvcd. It receives
console input from cvcd and passes it to cvc.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Sun Enterprise 10000 servers, Sun Fire 15K servers

Availability SUNWcvc.u

See Also cvcd(1M), attributes(5), console(7D), cvc(7D)

Sun Enterprise 10000 SSP Reference Manual

Sun System Management Services (SMS) Reference Manual

164 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2000
 dad(7D)

Name dad – driver for IDE disk devices

Synopsis dad@ target,lun:partition

Description This driver handles the ide disk drives on SPARC platforms. The type of disk drive is
determined using the ATA IDE identify device command and by reading the volume label
stored on the drive. The dad device driver supports the Solaris SPARC VTOC and the
EFI/GPT disk volume labels.

The block-files access the disk using the system's normal buffering mechanism and are read
and written without regard to physical disk records. There is also a "raw" interface that
provides for direct transmission between the disk and the user's read or write buffer. A single
read or write call usually results in one I/O operation; raw I/O is therefore considerably more
efficient when many bytes are transmitted. The names of the block files are found in /dev/dsk.
Raw file names are found in /dev/rdsk.

I/O requests to the raw device must be aligned on a 512-byte (DEV_BSIZE) boundary and must
have a length that is a multiple of 512 bytes. Requests that do not meet the restrictions cause
the driver to return an EINVAL error. I/O requests to the block device have no alignment or
length restrictions.

Device Statistics Each device maintains I/O statistics both for the device and for each partition allocated on that
Support device. For each device/partition, the driver accumulates reads, writes, bytes read, and bytes
written. The driver also takes hi-resolution time stamps at queue entry and exit points, which
facilitates monitoring the residence time and cumulative residence-length product for each
queue.

Each device also has error statistics associated with it. These must include counters for hard
errors, soft errors and transport errors. Other data may be implemented as required.
Files /dev/dsk/cntndnsn block files
/dev/rdsk/cntndnsn raw files

where:
cn controller n
tn IDE target id n (0-3)
dn Always 0.
sn partition n (0-7)

The target ide numbers are assigned as:

0 Master disk on Primary channel.
1 Slave disk on Primary channel.
2 Master disk on Secondary channel

Device and Network Interfaces 165

dad(7D)

3 Slave disk on Secondary channel.

ioctls Refer to dkio(7I).

Errors EACCES Permission denied.
EBUSY The partition was opened exclusively by another thread.
EFAULT Argument was a bad address.
EINVAL Invalid argument.
EIO I/O error occurred.
ENOTTY The device does not support the requested ioctl function.
ENXIO The device did not exist during opening.
EROFS The device is a read-only device.

See Also format(1M), mount(1M), lseek(2), read(2), write(2), driver.conf(4), vfstab(4), dkio(7I)

X3T10 ATA-4 specifications.

Diagnostics Command:<number>, Error:<number>, Status:<number>
Indicates that the command failed with an error and provides status register contents.
Where <number> is a hexadecimal value.
offline
The driver has decided that the target disk is no longer there.
disk ok
The target disk is now responding again.
disk not responding to selection
The target disk is not responding.
i/o to invalid geometry
The geometry of the drive could not be established.
incomplete read/write - retrying/giving up
There was a residue after the command completed normally.
no bp for disk label
A bp with consistent memory could not be allocated.
no memory for disk label
Free memory pool exhausted.
ATA transport failed: reason ’nnnn’: {retrying|giving}
The host adapter has failed to transport a command to the target for the reason stated. The
driver will either retry the command or, ultimately, give up.

166 man pages section 7: Device and Network Interfaces • Last Revised 24 Oct 2005
 dad(7D)

no mem for property

Free memory pool exhausted.
transport rejected (<n>)
Host adapter driver was unable to accept a command.
Device Fault
Device fault - reason for such error is vendor specific.

Device and Network Interfaces 167

daplt(7D)

Name daplt – Tavor uDAPL service driver

Synopsis daplt@0:daplt

Description The daplt module is the driver component of the uDAPL service provider for Tavor which
implements the provider functions of the uDAPL Specification 1.2 described under
libdat(3LIB).

The daplt module is a child of the IB nexus driver, ib(7D), and layers on top of the Solaris
kernel IB Transport Layer, ibtl(7D). The daplt driver uses the InfiniBand Transport
Framework (IBTF). (See ibtl(7D), ibcm(7D), and ib(7D) to access privileged IB VERBS.)

The daplt driver copies out various HCA H/W object reference handles, including working
and completion queues and User Access Region registers, to its own uDAPL service provider
library for Tavor. The library can refer back to these object handles and use them to mmap(2) in
the mapping of these H/W queues and registers from the Tavor HCA driver, tavor(7D). This
process enables time-critical non-privileged IB VERBS such as send/receive work elements,
RDMA read/write and memory window bind, to be invoked in the userland library and
performed directly by the firmware or hardware. As a result, OS and network stack are
bypassed, achieving true zero data copy with the lowest possible latency.
Files /kernel/drv/sparcv9/daplt 64-bit SPARC ELF kernel driver
/kernel/drv/daplt 32–bit x86 ELF kernel driver
/kernel/drv/amd64/daplt 64–bit x86 ELF kernel driver
/kernel/drv/daplt.conf driver configuration file
/dev/daplt special character device.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWdapltr

See Also mmap(2), libdat(3LIB), driver.conf(4), attributes(5), ib(7D), ibcm(7D), ibdm(7D),

ibtl(7D), tavor(7D)

uDAPL Specification 1.2

168 man pages section 7: Device and Network Interfaces • Last Revised 12 Oct 2004
 dca(7D)

Name dca – Crypto Accelerator device driver

Synopsis pci108e,5454@pci-slot
pci108e,5455@pci-slot
pci108e,5456@pci-slot
pci14e4,5820@pci-slot
pci14e4,5821@pci-slot
pci14e4,5822@pci-slot

Description The dca device driver is a multi-threaded, loadable hardware driver supporting Sun
PCI-based (pci108e,5454) cryptographic accelerators, such as the Sun Crypto Accelerator
1000.

The dca driver requires the presence of Solaris Cryptographic Framework for applications and
kernel clients to access the provided services.

Extended The dca driver maintains the following statistics:

Description
3desjobs Total number of jobs submitted to the device for 3DES encryption.
3desbytes Total number of bytes submitted to the device for 3DES encryption.
rsapublic Total number of jobs submitted to the device for RSA public key
operations.
rsaprivate Total number of jobs submitted to the device for RSA private key
operations.
dsasign Total number of jobs submitted to the device for DSA signing.
dsaverify Total number of jobs submitted to the device for DSA verification.
rngjobs Total number of jobs submitted for pure entropy generation.
rngbytes Total number of bytes of pure entropy requested from the device.
rngsha1jobs Total number of jobs submitted for entropy generation, with SHA-1
post-processing.
rngsha1bytes Total number of bytes of entropy requested from the device, with SHA-1
post-processing.

Additional statistics may be supplied for Sun support personnel, but are not useful to end
users and are not documented here.

The dca driver can be configured by defining properties in /kernel/drv/dca.conf which

override the default settings. The following properties are supported:

Device and Network Interfaces 169

dca(7D)

nostats Disables the generation of statistics. This property may be used to help prevent
traffic analysis, but this may inhibit support personnel.
rngdirect Disables the SHA-1 post-processing of generated entropy. This may give
"truer" random numbers, but it may also introduce the risk of external biases
influencing the distribution of generated random numbers.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWdcar

Interface Stability Unstable

Files /kernel/drv/dca.conf dca configuration file

/kernel/drv/sparcv9/dca 64–bit ELF kernel driver (SPARC)
/kernel/drv/dca 32–bit ELF kernel driver (x86)
/kernel/drv/amd64/dca 64–bit ELF kernel driver (AMD64)

See Also cryptoadm(1M), kstat(1M), prtconf(1M), driver.conf(4), attributes(5)

Solaris Cryptographic Framework.

170 man pages section 7: Device and Network Interfaces • Last Revised 14 Aug 2005
 dcam1394(7D)

Name dcam1394 – 1394–based digital camera (IIDC) driver

Synopsis #include <sys/dcam/dcam1394_io.h>

Description The dcam1394 driver supports devices implementing the 1394 Trade Association Digital
Camera Specification (also referred to as the IIDC specification). Only a subset of the
specification is supported.

Reading Data Isochronous data is read from the driver frame-by-frame and is maintained within the driver
in a ring buffer.

Video frames are read from the isochronous input device using read(2).

The dcam1394_frame_t structure describes the frame layout and is defined as follows:
struct {
unsigned int vid_mode;
unsigned int seq_num;
hrtime_t timestamp;
unsigned char *buff;
};

The size to allocate for the structure is determined by the video mode for which the camera is
configured. Possible values for the vid_mode field are listed under
DCAM1394_PARAM_VID_MODE below.

Ioctl Requests The following ioctl requests are supported:

DCAM1394_CMD_CAM_RESET
Reset the device.
DCAM1394_CMD_REG_READ
Read the indicated dcam/IIDC register. The argument is a pointer to a dcam1394_reg_io_t
structure, which is defined as follows:
struct {
unsigned int offs;
unsigned int val;
};
The offs field should be set to the offset of the register from which to read. Register offset
values are defined in the 1394 Trade Association Digital Camera Specification.
After the operation is completed, the camera register value is put in the val field.
DCAM1394_CMD_REG_WRITE
Write the indicated dcam/IIDC register. The argument is a pointer to a
dcam1394_reg_io_t structure (described above).
The offs field should be set to the offset of the register from which to read. The register
offset values are defined in the 1394 Trade Association Digital Camera Specification.

Device and Network Interfaces 171

dcam1394(7D)

The val field should be set to the value to be written to the camera register.
DCAM1394_CMD_PARAM_GET
Gets a list of parameters associated with a camera. The argument is a pointer to a
dcam1394_param_list_t structure (described below). The parameter list is accessed
through macros defined below.
The paramter list only supports Format 1 video formats.
DCAM1394_CMD_PARAM_SET
Sets a list of parameters associated with a camera. The argument is a pointer to a
dcam1394_param_list_t structure (described below). The parameter list is accessed
through macros defined below.
The paramter list only supports Format 1 video formats.
DCAM1394_CMD_FRAME_RCV_START
Start receiving video frames from the camera.
The contents of the ring buffer may be accessed by reading the isochronous stream device.
DCAM1394_CMD_FRAME_RCV_STOP
Stop receiving frames from the camera.
DCAM1394_CMD_RING_BUFF_FLUSH
Flush the frames in the ring buffer.
DCAM1394_CMD_FRAME_SEQ_NUM_COUNT_RESET
Reset frame sequence number.
Parameter List The parameter list is initialized and access through macros. The data type for the parameter
Access list is dcam1394_param_list_t.

The following macros are used to access the parameter list:

PARAM_LIST_INIT(param_list)
Initialize the parameter list.
PARAM_LIST_ADD(param_list, param, subparam)
Add a parameter to the list.
PARAM_LIST_REMOVE(param_list, param, subparam)
Remove a parameter from the list.
PARAM_LIST_IS_ENTRY(param_list, param, subparam)
Indicates if a specific parameter is in the list.
PARAM_VAL(param_list, param, subparam)
Value of a specified parameter.
PARAM_ERR(param_list, param, subparam)
Indicates if a specific parameter is successfully set.

172 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2004
 dcam1394(7D)

When no subparam value is required, the value DCAM1394_SUBPARAM_NONE may be

used.
Parameters The following parameters may appear in the list:
DCAM1394_PARAM_CAP_POWER_CTRL
Queries if the camera can be turned off and on through software. The subparam value is
ignored.
DCAM1394_PARAM_POWER
Controls or queries if the camera is powered up. Verify this feature using
DCAM1394_PARAM_CAP_POWER_CTRL before use. The subparam field is ignored.
DCAM1394_PARAM_CAP_VID_MOD
Queries if a specific video mode is supported by the camera.
subparam is one of the following and is used to determine if a specified video mode is
supported by the camera:
DCAM1394_SUBPARAM_VID_MODE_0
DCAM1394_SUBPARAM_VID_MODE_YUV_444_160_120
Video mode is 4:4:4, YUV color space, 160x120 resolution.

DCAM1394_SUBPARAM_VID_MODE_1
DCAM1394_SUBPARAM_VID_MODE_YUV_422_320_240
Video mode is 4:2:2, YUV color space, 320x240 resolution.

DCAM1394_SUBPARAM_VID_MODE_2
DCAM1394_SUBPARAM_VID_MODE_YUV_411_640_480
Video mode is 4:1:1, YUV color space, 640x480 resolution.

DCAM1394_SUBPARAM_VID_MODE_3
DCAM1394_SUBPARAM_VID_MODE_YUV_422_640_480
Video mode is 4:2:2, YUV color space, 640x480 resolution.

DCAM1394_SUBPARAM_VID_MODE_4
DCAM1394_SUBPARAM_VID_MODE_RGB_640_480
Video mode is RGB color space, 640x480 resolution.

DCAM1394_SUBPARAM_VID_MODE_5
DCAM1394_SUBPARAM_VID_MODE_Y_640_480
Video mode is Y color space, 640x480 resolution.
DCAM1394_PARAM_VID_MODE
Controls or queries the current video mode of the camera. The subparam field is ignored.
When selecting the video mode, it should be compatible with the capability of the camera,
which may be determined by checking the DCAM1394_PARAM_CAP_VID_MODE
parameter.
The value of this parameter may be one of the following:

Device and Network Interfaces 173

dcam1394(7D)

DCAM1394_VID_MODE_0
DCAM1394_VID_MODE_YUV_444_160_120
Video mode is 4:4:4, YUV color space, 160x120 resolution.

DCAM1394_VID_MODE_1
DCAM1394_VID_MODE_YUV_422_320_240
Video mode is 4:2:2, YUV color space, 320x240 resolution.

DCAM1394_VID_MODE_2
DCAM1394_VID_MODE_YUV_411_640_480
Video mode is 4:1:1, YUV color space, 640x480 resolution.

DCAM1394_VID_MODE_3
DCAM1394_VID_MODE_YUV_422_640_480
Video mode is 4:2:2, YUV color space, 640x480 resolution.

DCAM1394_VID_MODE_4
DCAM1394_VID_MODE_RGB_640_480
Video mode is RGB color space, 640x480 resolution.

DCAM1394_VID_MODE_5
DCAM1394_VID_MODE_Y_640_480
Video mode is Y color space, 640x480 resolution.
DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_0
Queries if a specific frame rate is supported by the camera in video mode 0 (4:4:4, YUV,
160x120).

subparam is one of the following and used to determine if a specified frame rate is
supported by the camera:
DCAM1394_SUBPARAM_FRAME_RATE_0
DCAM1394_SUBPARAM_FRAME_RATE_3_75_FPS
Frame rate is 3.75 frames/second.

DCAM1394_SUBPARAM_FRAME_RATE_1
DCAM1394_SUBPARAM_FRAME_RATE_7_5_FPS
Frame rate is 7.5 frames/second.

DCAM1394_SUBPARAM_FRAME_RATE_2
DCAM1394_SUBPARAM_FRAME_RATE_15_FPS
Frame rate is 15 frames/second.

DCAM1394_SUBPARAM_FRAME_RATE_3
DCAM1394_SUBPARAM_FRAME_RATE_30_FPS
Frame rate is 30 frames/second.

DCAM1394_SUBPARAM_FRAME_RATE_4
DCAM1394_SUBPARAM_FRAME_RATE_60_FPS
Frame rate is 60 frames/second.

174 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2004
 dcam1394(7D)

DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_1
Queries if a specific frame rate is supported by the camera in video mode 1 (4:2:2, YUV,
320x240). See DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_0 for a listing of
valid subparam values.
DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_2
Queries if a specific frame rate is supported by the camera in video mode 2 (4:1:1, YUV,
640x480). See DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_0 for a listing of
valid subparam values.
DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_3
Queries if a specific frame rate is supported by the camera in video mode 3 (4:2:2, YUV,
640x480). See DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_0 for a listing of
valid subparam values.
DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_4
Queries if a specific frame rate is supported by the camera in video mode 4. (RGB,
640x480). See DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_0 for a listing of
valid subparam values.
DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_5
Queries if a specific frame rate is supported by the camera in video mode 5. (Y, 640x480).
See DCAM1394_PARAM_CAP_FRAME_RATE_VID_MODE_0 for a listing of valid
subparam values.
DCAM1394_PARAM_FRAME_RATE
Controls or queries the current frame rate of the camera. The subparam field is ignored.
When selecting a frame rate, it should be compatible with the capability of the camera,
which can be determined by querying one of the frame rate capability parameters above.

The value of this parameter may be one of the following:

DCAM1394_FRAME_RATE_0
DCAM1394_3_75_FPS
The frame rate is 3.75 frames per second.

DCAM1394_FRAME_RATE_1
DCAM1394_7_5_FPS
The frame rate is 7.5 frames per second.

DCAM1394_FRAME_RATE_2
DCAM1394_15_FPS
The frame rate is 15 frames per second.

DCAM1394_FRAME_RATE_3
DCAM1394_30_FPS
The frame rate is 30 frames per second.

DCAM1394_FRAME_RATE_4

Device and Network Interfaces 175

dcam1394(7D)

DCAM1394_60_FPS
The frame rate is 60 frames per second.
DCAM1394_PARAM_RING_BUFF_CAPACITY
Controls or queries the number of frames that the ring buffer may hold. This value can
range between 2 and 30. The subparam field is ignored.
DCAM1394_PARAM_RING_BUFF_NUM_FRAMES_READY
Queries the number of frames in the ring buffer ready to be accessed. The subparam field is
ignored.
DCAM1394_PARAM_RING_BUFF_READ_PTR_INCR
Controls or queries the number of bytes to advance the read pointer as it consumes data
from the ring buffer. The subparam field is ignored.
DCAM1394_PARAM_FRAME_NUM_BYTES
Queries the number of bytes in a frame at the current video mode. The subparam field is
ignored.
DCAM1394_PARAM_STATUS
Queries the parameter status. The subparam field is ignored.

The values for the parameter status is a bit field with the following values possibly set:
DCAM1394_STATUS_FRAME_RCV_DONE
Frame successfully received.
DCAM1394_STATUS_RING_BUFF_LOST_FRAME
A frame has been lost while processing the ring buffer.
DCAM1394_STATUS_PARAM_CHANGE
A parameter has been changed.
DCAM1394_STATUS_FRAME_SEQ_NUM_COUNT_OVERFLOW
Frame sequence number has reached its maximum possible value and has overflowed.
DCAM1394_STATUS_CAM_UNPLUG
Camera has been unplugged.
DCAM1394_PARAM_BRIGHTNESS
Query or control a camera feature. This feature queries or controls the brightness of the
camera.
DCAM1394_SUBPARAM_PRESENCE
Indicates if the feature is available.
DCAM1394_SUBPARAM_CAP_ON_OFF
Indicates if the feature may be enabled and disabled. May only be queried.
DCAM1394_SUBPARAM_ON_OFF
Indicates if the feature is enabled.

176 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2004
 dcam1394(7D)

DCAM1394_SUBPARAM_CAP_CTRL_AUTO
Indicates if the automatic control of this feature is supported by the camera. May only be
queried.
DCAM1394_SUBPARAM_CAP_CTRL_MANUAL
Indicates if the manual control of this feature is supported by the camera. May only be
queried.
DCAM1394_SUBPARAM_CTRL_MODE
Indicates if the feature is in auto or manual mode.
DCAM1394_SUBPARAM_MIN_VAL
Minimum value of the feature. May only be queried.
DCAM1394_SUBPARAM_MAX_VAL
Maximum value of the feature. May only be queried.
DCAM1394_SUBPARAM_VALUE
Current value of the feature.
DCAM1394_SUBPARAM_CAP_READ
Indicates if the feature may be read. May only be queried.
DCAM1394_PARAM_EXPOSURE
Query or control a camera feature. This feature queries or controls the exposure of the
camera. The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_SHARPNESS
Query or control a camera feature. This feature queries or controls the sharpness of the
camera. The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_WHITE_BALANCE
Query or control a camera feature. This feature queries or controls the white balance of the
camera. The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS, except for DCAM1394_SUBPARAM_VALUE.
DCAM1394_SUBPARAM_VALUE is replaced by two distinct subparams.
DCAM1394_SUBPARAM_U_VALUE U or B component of the white balance.
DCAM1394_SUBPARAM_V_VALUE V or R component of the white balance.
DCAM1394_PARAM_HUE
Query or control a camera feature. This feature queries or controls the hue of the camera.
The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.

Device and Network Interfaces 177

dcam1394(7D)

DCAM1394_PARAM_SATURATION
Query or control a camera feature. This feature queries or controls the saturation of the
camera. The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_GAMMA
Query or control a camera feature. This feature queries or controls the gamma of the
camera. The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_SHUTTER
Query or control a camera feature. This feature queries or controls the sharpness of the
camera. The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_GAIN
Query or control a camera feature. This feature queries or controls the gain of the camera.
The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_IRIS
Query or control a camera feature. This feature queries or controls the iris of the camera.
The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_FOCUS
Query or control a camera feature. This feature queries or controls the focus of the camera.
The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_ZOOM
Query or control a camera feature. This feature queries or controls the zoom of the camera.
The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_PAN
Query or control a camera feature. This feature queries or controls the pan of the camera.
The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
DCAM1394_PARAM_TILT
Query or control a camera feature. This feature queries or controls the tilt of the
camera.The subparams supported by this feature are described under
DCAM1394_PARAM_BRIGHTNESS.
Device Special /dev/dcamN Device node for isochronous input from camera.
Files
/dev/dcamctlN Device node for camera control.

178 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2004
 dcam1394(7D)

Files kernel/drv/sparcv9/dcam1394 64-bit ELF kernel module.

kernel/drv/dcam1394 32-bit ELF kernel module.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Evolving

See Also attributes(5), hci1394(7D)

1394 Trade Association Digital Camera Specification, Version 1.04 – 1996

IEEE Std 1394-2000 Standard for a High Performance Serial Bus – 2000

Device and Network Interfaces 179

dcfs(7FS)

Name dcfs – Compression file system

Synopsis #include <sys/filio.h>
#include <sys/fs/decomp.h>

Description The dcfs filesystem is a layered filesystem that you use to compress data when writing to a file
and decompress upon read. The primary function of the dcfs filesystem is to compress
individual files when constructing a boot archive and when reading or booting from the
archive.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Uncommitted

See Also boot(1M), bootadm(1M), fiocompress(1M), attributes(5), ufs(7FS)

Notes The dcfs compression/decompression file system works only with UFS.

180 man pages section 7: Device and Network Interfaces • Last Revised 28 Nov 2007
 dda(7D)

Name dda – MMS disk archiving driver

Synopsis /devices/pseudo/dda@lun:bn

Description The Media Management System (MMS) is a distributed removable media management
system based on the IEEE 1244 Media Management System specification. MMS provides a
uniform and consistent tape interface to client applications.

The dda driver is a pseudo tape drive with filesystem file media. The dda driver is a minimal
emulation of the st(7D) tape driver mtio(7I) interface with BSD no-rewind behavior and is
used for MMS disk archiving. The dda driver should be used with a MMS MMP mount
command handle. You configure DDA tapes, drives, and libraries for disk archiving using the
mmsadm(1M) command. You also use mmsadm(1M) to mount and unmount dda media.

Configuration To set the number of disk archiving tape drives available for MMS disk archiving, edit the
/kernel/drv/dda.conf configuration file. The default number of disk archiving tape drives is
20.

The disk archiving tape drive devlinks are located in the /dev/dda directory.

The DDA maximum block size is 262144 bytes and the minimum is 1 byte.
Files /kernel/drv/dda 32–bit kernel module (x86)
/kernel/drv/sparcv9/dda 64–bit kernel module (SPARC)
/kernel/drv/amd64/dda 64–bit kernel module (x86)
/kernel/drv/dda.conf dda configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWmmsr

Interface Stability Uncommitted

See Also mt(1), mmsadm(1M), attributes(5), dmd(7D), st(7D), mtio(7I)

IEEE 1244 Removable Media Standards Specification — IEEE, 2000

Device and Network Interfaces 181

dev(7FS)

Name dev – Device name file system

Description The dev filesystem manages the name spaces of devices under the Solaris operating
environment. The global zone's instance of the dev filesystem is mounted during boot on
/dev.

A subdirectory under /dev may have unique operational semantics. Most of the common
device names under /dev are created automatically by devfsadm(1M). Others, such as
/dev/pts, are dynamic and reflect the operational state of the system. You can manually
generate device names for newly attached hardware by invoking devfsadm(1M) or implicitly,
by indirectly causing a lookup or readdir operation in the filesystem to occur. For example,
you can discover a disk that was attached when the system was powered down (and generate a
name for that device) by invoking format(1M)).
Files /dev Mount point for the /dev filesystem in the global zone.

See Also devfsadm(1M), format(1M), devfs(7FS)

Notes The global /dev instance cannot be unmounted.

182 man pages section 7: Device and Network Interfaces • Last Revised 9 June 2006
 devfs(7FS)

Name devfs – Devices file system

Description The devfs filesystem manages a name space of all devices under the Solaris operating
environment and is mounted during boot on the /devices name space.

The /devices name space is dynamic and reflects the current state of accessible devices under
the Solaris operating environment. The names of all attached device instances are present
under /devices.

The content under /devices is under the exclusive control of the devfs filesystem and cannot
be changed.

The system may be configured to include a device in one of two ways:

By means of dynamic reconfiguration (DR), using, for example, cfgadm(1M).

For devices driven by driver.conf(4) enumeration, edit the driver.conf file to add a new
entry, then use update_drv(1M) to cause the system to re-read the driver.conf file and
thereby enumerate the instance.

The device may be attached through a number of system calls and programs, including
open(2), stat(2) and ls(1). During device attach, the device driver typically creates minor
nodes corresponding to the device via ddi_create_minor_node(9F). If the attach is
successful, one or more minor nodes referring to the device are created under /devices.

Operations like mknod(2), mkdir(2) and creat(2) are not supported in /devices.
Files /devices Mount point for devfs file system

See Also devfsadm(1M), vfstab(4), attach(9E)

Notes The /devices name space cannot be unmounted.

All content at or below the /devices name space is an implementation artifact and subject to
incompatible change or removal without notification.

Device and Network Interfaces 183

devinfo(7D)

Name devinfo – device information driver

Description The devinfo driver is a private mechanism used by the libdevinfo(3LIB) interfaces to access
kernel device configuration data and to guarantee data consistency.

Files /devices/pseudo/devinfo@0:devinfo

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Stability Level Private

See Also libdevinfo(3LIB), attributes(5)

Writing Device Drivers

184 man pages section 7: Device and Network Interfaces • Last Revised 9 Jan 1998
 dkio(7I)

Name dkio – disk control operations

Synopsis #include <sys/dkio.h>
#include <sys/vtoc.h>

Description Disk drivers support a set of ioctl(2) requests for disk controller, geometry, and partition
information. Basic to these ioctl() requests are the definitions in <sys/dkio.h>.

ioctls The following ioctl() requests set and/or retrieve the current disk controller, partitions, or
geometry information on all architectures:
DKIOCINFO
The argument is a pointer to a dk_cinfo structure (described below). This structure tells
the controller–type and attributes regarding bad-block processing done on the controller.
/*
* Structures and definitions for disk I/O control commands
*/
#define DK_DEVLEN 16 /* device name max length, */
/* including unit # and NULL */
/* Used for controller info */
struct dk_cinfo {
char dki_cname[DK_DEVLEN]; /* controller name */
/* (no unit #) */
ushort_t dki_ctype; /* controller type */
ushort_t dki_flags; /* flags */
ushort_t dki_cnum; /* controller number */
uint_t dki_addr; /* controller address */
uint_t dki_space; /* controller bus type */
uint_t dki_prio; /* interrupt priority */
uint_t dki_vec; /* interrupt vector */
char dki_dname[DK_DEVLEN]; /* drive name (no unit #) */
uint_t dki_unit; /* unit number */
uint_t dki_slave; /* slave number */
ushort_t dki_partition; /* partition number */
ushort_t dki_maxtransfer; /* maximum transfer size */
/* in DEV_BSIZE */

};
/*
* Controller types
*/

#define DKC_UNKNOWN 0
#define DKC_CDROM 1 /* CD-ROM, SCSI or other */
#define DKC_WDC2880 2
#define DKC_XXX_0 3 /* unassigned */
#define DKC_XXX_1 4 /* unassigned */
#define DKC_DSD5215 5

Device and Network Interfaces 185

dkio(7I)

#define DKC_ACB4000 7
#define DKC_XXX_2 9 /* unassigned */
#define DKC_NCRFLOPPY 10
#define DKC_SMSFLOPPY 12
#define DKC_SCSI_CCS 13 /* SCSI CCS compatible */
#define DKC_INTEL82072 14 /* native floppy chip */
#define DKC_MD 16 /* meta-disk (virtual-disk) */
/* driver */
#define DKC_INTEL82077 19 /* 82077 floppy disk */
/* controller */
#define DKC_DIRECT 20 /* Intel direct attached */
/* device (IDE) */
#define DKC_PCMCIA_MEM 21 /* PCMCIA memory disk-like */
/* type */
#define DKC_PCMCIA_ATA 22 /* PCMCIA AT Attached type */

/*
* Sun reserves up through 1023
*/

#define DKC_CUSTOMER_BASE 1024

/*
* Flags
*/

#define DKI_BAD144 0x01 /* use DEC std 144 */

/* bad sector fwding */
#define DKI_MAPTRK 0x02 /* controller does */
/* track mapping */
#define DKI_FMTTRK 0x04 /* formats only full
/* track at a time*/
#define DKI_FMTVOL 0x08 /* formats only full */
/* volume at a time*/
#define DKI_FMTCYL 0x10 /* formats only full */
/* cylinders at a time*/
#define DKI_HEXUNIT 0x20 /* unit number printed as */
/* 3 hexdigits */
#define DKI_PCMCIA_PFD 0x40 /* PCMCIA pseudo-floppy */
/* memory card */
DKIOCGAPART
The argument is a pointer to a dk_allmap structure (described below). This ioctl() gets
the controller's notion of the current partition table for disk drive.
DKIOCSAPART
The argument is a pointer to a dk_allmap structure (described below). This ioctl() sets
the controller's notion of the partition table without changing the disk itself.

186 man pages section 7: Device and Network Interfaces • Last Revised 16 Dec 2009
 dkio(7I)

/*
* Partition map (part of dk_label)
*/ struct dk_map {
daddr_t dkl_cylno; /* starting cylinder */
daddr_t dkl_nblk; /* number of blocks */
};
/*
* Used for all partitions
*/
struct dk_allmap {
struct dk_map dka_map[NDKMAP];
};
DKIOCGGEOM The argument is a pointer to a dk_geom structure (described below). This
ioctl() gets the controller's notion of the current geometry of the disk drive.
DKIOCSGEOM The argument is a pointer to a dk_geom structure (described below). This
ioctl() sets the controller's notion of the geometry without changing the
disk itself.
DKIOCGVTOC The argument is a pointer to a vtoc structure (described below). This
ioctl() returns the device's current volume table of contents (VTOC.) For
disks larger than 1TB, DKIOCGEXTVTOC must be used instead.
DKIOCSVTOC The argument is a pointer to a vtoc structure (described below). This
ioctl() changes the VTOC associated with the device. For disks larger than
1TB, DKIOCSEXTVTOC must be used instead.
struct partition {
ushort_t p_tag; /* ID tag of partition */
ushort_t p_flag; /* permission flags */
daddr_t p_start; /* start sector of partition */
long p_size; /* # of blocks in partition */
};

If DKIOCSVTOC is used with a floppy diskette, the p_start field must be the first sector of a
cylinder. To compute the number of sectors per cylinder, multiply the number of heads by the
number of sectors per track.
struct vtoc {
unsigned long v_bootinfo[3]; /* info needed by mboot
/* (unsupported)*/
unsigned long v_sanity; /* to verify vtoc */
/* sanity */
unsigned long v_version; /* layout version */
char v_volume[LEN_DKL_VVOL]; /* volume name */
ushort_t v_sectorsz; \*
sector size in bytes*/
ushort_t v_nparts; \*

Device and Network Interfaces 187

dkio(7I)

number of partitions*/
unsigned long v_reserved[10]; /* free space */
struct partition v_part[V_NUMPAR]; /* partition headers */
time_t timestamp[V_NUMPAR]; /* partition timestamp */
/* (unsupported) */
char v_asciilabel[LEN_DKL_ASCII]; /* compatibility */
};

/*
* Partition permission flags
*/

#define V_UNMNT 0x01 /* Unmountable partition */

#define V_RONLY 0x10 /* Read only */

/*
* Partition identification tags
*/

#define V_UNASSIGNED 0x00 /* unassigned partition */

#define V_BOOT 0x01 /* Boot partition */
#define V_ROOT 0x02 /* Root filesystem */
#define V_SWAP 0x03 /* Swap filesystem */
#define V_USR 0x04 /* Usr filesystem */
#define V_BACKUP 0x05 /* full disk */
#define V_VAR 0x07 /* Var partition */
#define V_HOME 0x08 /* Home partition */
#define V_ALTSCTR 0x09 /* Alternate sector partition */
DKIOCGEXTVTOC
The argument is a pointer to an extvtoc structure (described below). This ioctl returns the
device's current volume table of contents (VTOC). VTOC is extended to support a disk up
to 2TB in size. For disks larger than 1TB this ioctl must be used instead of DKIOCGVTOC.
DKIOCSEXTVTOC
The argument is a pointer to an extvtoc structure (described below). This ioctl changes the
VTOC associated with the device. VTOC is extended to support a disk up to 2TB in size.
For disks larger than 1TB this ioctl must be used instead of DKIOCSVTOC.
struct extpartition {
ushort_t p_tag; /* ID tag of partition */
ushort_t p_flag; /* permission flags */
ushort_t p_pad[2]; /* reserved */
diskaddr_t p_start; /* start sector no of partition */
diskaddr_t p_size; /* # of blocks in partition */
};

struct extvtoc {

188 man pages section 7: Device and Network Interfaces • Last Revised 16 Dec 2009
 dkio(7I)

uint64_t v_bootinfo[3]; /* info needed by mboot (unsupported) */

uint64_t v_sanity; /* to verify vtoc sanity */
uint64_t v_version; /* layout version */
char v_volume[LEN_DKL_VVOL]; /* volume name */
ushort_t v_sectorsz; /* sector size in bytes */
ushort_t v_nparts; /* number of partitions */
ushort_t pad[2];
uint64_t v_reserved[10];
struct extpartition v_part[V_NUMPAR]; /* partition headers */
uint64_t timestamp[V_NUMPAR]; /* partition timestamp (unsupported)*/
char v_asciilabel[LEN_DKL_ASCII]; /* for compatibility */
};

Partition permissions flags and identification tags

are defined the same as vtoc structure.
DKIOCEJECT
If the drive supports removable media, this ioctl() requests the disk drive to eject its disk.
DKIOCREMOVABLE
The argument to this ioctl() is an integer. After successful completion, this ioctl() sets
that integer to a non-zero value if the drive in question has removable media. If the media is
not removable, the integer is set to 0.
DKIOCHOTPLUGGABLE
The argument to this ioctl() is an integer. After successful completion, this ioctl() sets
that integer to a non-zero value if the drive in question is hotpluggable. If the media is not
hotpluggable, the integer is set to 0.
DKIOCREADONLY
The argument to this ioctl() is an integer. After successful completion, this ioctl() sets
that integer to a non-zero value if the drive in question has read-only media. If the media is
writable, or not present, the integer is set to 0.
DKIOCSTATE
This ioctl() blocks until the state of the drive, inserted or ejected, is changed. The
argument is a pointer to a dkio_state, enum, whose possible enumerations are listed
below. The initial value should be either the last reported state of the drive, or DKIO_NONE.
Upon return, the enum pointed to by the argument is updated with the current state of the
drive.
enum dkio_state {
DKIO_NONE, /* Return disk’s current state */
DKIO_EJECTED, /* Disk state is ’ejected’ */
DKIO_INSERTED /* Disk state is ’inserted’ */
};
DKIOCLOCK
For devices with removable media, this ioctl() requests the disk drive to lock the door.

Device and Network Interfaces 189

dkio(7I)

DKIOCUNLOCK
For devices with removable media, this ioctl() requests the disk drive to unlock the door.
DKIOCGMEDIAINFO
The argument to this ioctl() is a pointer to a dk_minfo structure. The structure indicates
the type of media or the command set profile used by the drive to operate on the media.
The dk_minfo structure also indicates the logical media block size the drive uses as the
basic unit block size of operation and the raw formatted capacity of the media in number of
logical blocks.
DKIOCGMEDIAINFOEXT
The argument to this ioctl() is a pointer to a dk_minfo_ext structure. The structure
indicates the type of media or the command set profile used by the drive to operate on the
media. The dk_minfo_ext structure also indicates the logical media block size the drive
uses as the basic unit block size of operation, the raw formatted capacity of the media in
number of logical blocks and the physical block size of the media.
/*
* Used for media info or profile info
*/
struct dk_minfo {
uint_t dki_media_type; /* Media type or profile info */
uint_t dki_lbsize; /* Logical blocksize of media */
diskaddr_t dki_capacity; /* Capacity as # of dki_lbsize blks */
};

/*
* Used for media info or profile info and physical blocksize
*/
struct dk_minfo_ext {
uint_t dki_media_type; /* Media type or profile info */
uint_t dki_lbsize; /* Logical blocksize of media */
diskaddr_t dki_capacity; /* Capacity as # of dki_lbsize blks */
uint_t dki_pbsize; /* Physical blocksize of media */
};

/*
* Media types or profiles known
*/
#define DK_UNKNOWN 0x00 /* Media inserted - type unknown */

/*
* SFF 8090 Specification Version 3, media types 0x01 - 0xfffe are
* retained to maintain compatibility with SFF8090. The following
* define the optical media type.
*/
#define DK_MO_ERASABLE 0x03 /* MO Erasable */

190 man pages section 7: Device and Network Interfaces • Last Revised 16 Dec 2009
 dkio(7I)

#define DK_MO_WRITEONCE 0x04 /* MO Write once */

#define DK_AS_MO 0x05 /* AS MO */
#define DK_CDROM 0x08 /* CDROM */
#define DK_CDR 0x09 /* CD-R */
#define DK_CDRW 0x0A /* CD-RW */
#define DK_DVDROM 0x10 /* DVD-ROM */
#define DK_DVDR 0x11 /* DVD-R */
#define DK_DVDRAM 0x12 /* DVD_RAM or DVD-RW */

/*
* Media types for other rewritable magnetic media
*/
#define DK_FIXED_DISK 0x10001 /* Fixed disk SCSI or otherwise */
#define DK_FLOPPY 0x10002 /* Floppy media */
#define DK_ZIP 0x10003 /* IOMEGA ZIP media */
#define DK_JAZ 0x10004 /* IOMEGA JAZ media */

If the media exists and the host can obtain a current profile list, the command succeeds and
returns the dk_minfo structure with data representing that media.

If there is no media in the drive, the command fails and the host returns an ENXIO error,
indicating that it cannot gather the information requested.

If the profile list is not available, the host attempts to identify the media-type based on the
available information.

If identification is not possible, the host returns media type DK_UNKNOWN. See NOTES for
blocksize usage and capacity information.
DKIOCSMBOOT
The argument is a pointer to struct mboot.

Copies the mboot information supplied in the argument to the absolute sector 0 of the
device. Prior to copying the information, this ioctl() performs the following checks on
the mboot data:
■ Ensures that the signature field is set to 0xAA55.
■ Ensures that partitions do not overlap.
■ On SPARC platforms, determines if the device is a removable media.

If the above verification fails, errno is set to EINVAL and the ioctl() command fails.

x86 Platforms — Upon successful write of mboot, the partition map structure maintained
in the driver is updated. If the new Solaris partition is different from the previous one, the
internal VTOC table maintained in the driver is set as follows:

If _SUNOS_VTOC_8 is defined:

Partition: 0. Start: 0. Capacity = Capacity of device.

Device and Network Interfaces 191

dkio(7I)

Partition: 2. Start: 0. Capacity = Capacity of device.

If _SUNOS_VTOC_16 is defined:

Partition: 2. Start: 0. Size = Size specified in mboot - 2 cylinders.

Partition: 8. Start: 0. Size = Sectors/cylinder.

Partition: 9. Start: Sectors/cylinder. Size = 2 * sectors/cylinder

To determine if the Solaris partition has changed:

If either offset or the size of the Solaris partition is different from the previous one then it
shall be deemed to have changed. In all other cases, the internal VTOC info remains as
before.

SPARC Platforms — The VTOC label and mboot both occupy the same location, namely
sector 0. As a result, following the successful write of mboot info, the internal VTOC table
maintained in the driver is set as follows:

Partition: 0. Start: 0. Size = Capacity of device.

Partition: 2. Start: 0. Size = Capacity of device.

See the NOTES section for usage of DKIOCSMBOOT when modifying Solaris partitions.
DKIOCGETVOLCAP
This ioctl provides information and status of available capabilities.

vc_info is a bitmap and the valid flag values are:

DKV_ABR_CAP - Capable of application-based recovery
DKV_DMR_CAP - Ability to read specific copy of data when
multiple copies exist. For example, in a two
way mirror, this ioctl is used to read each
side of the mirror.
vc_set is a bitmap and the valid flag values are:
DKV_ABR_CAP - This flag is set if ABR has been set on a device
that supports ABR functionality.
DKV_DMR_CAP - Directed read has been enabled.
These capabilities are not required to be persistent across a system reboot and their
persistence depends upon the implementation. For example, if the ABR capability for a
DRL mirror simply clears the dirty-region list and subsequently stops updating this list,
there is no reason for persistence because the VM recovery is a no-op. Conversely, if the
ABR capability is applied to a non-DRL mirror to indicate that the VM should not perform
a full recovery of the mirror following a system crash, the capability must be persistent so
that the VM know whether or not to perform recovery.

192 man pages section 7: Device and Network Interfaces • Last Revised 16 Dec 2009
 dkio(7I)

Return Errors:
EINVAL Invalid device for this operation.
ENOTSUP Functionality that is attempted to be set is not supported.
DKIOCSETVOLCAP
This ioctl sets the available capabilities for the device. If a capability flag is not set in vc_set,
that capability is cleared.

vc_info flags are ignored

vc_set valid flags are:

DKV_ABR_CAP - Flag to set application-based recovery. A device can
successfully support ABR only if it is capable.
DKV_DMR_CAP - Flag to set directed read.

int
ioctl(int , DKIODMR, vol_directed_rd *);
DKIODMR
This ioctl allows highly available applications to perform round-robin reads from the
underlying devices of a replicated device.
vdr_offset - offset at which the read should occur.
vdr_nbytes - number of bytes to be read
vdr_bytesread - number of bytes successfully read by the kernel.
vdr_data - pointer to a user allocated buffer to return the
data read
vdr_side - side to be read. Initialized to DKV_SIDE_INIT
vdr_side_name - The volume name that has been read.

Valid vdr_flags are:

DKV_DMR_NEXT_SIDE (set by user)
DKV_DMR_DONE (return value)
DKV_DMR_ERROR (return value)
DKV_DMR_SUCCESS(return value)
DKV_DMR_SHORT(return value)
The calling sequence is as follows: The caller sets the vdr_flags to DK_DMR_NEXT_SIDE and
vdr_side to DKV_SIDE_INIT at the start. Subsequent calls should be made without any
changes to these values. If they are changed the results of the ioctl are indeterminate.

When DKV_SIDE_INIT is set, the call results in the kernel reading from the first side. The
kernel updates vdr_side to indicate the side that was read, and vdr_side_name to contain
the name of that side. vdr_data contains the data that was read. Therefore to perform a
round-robin read all of the valid sides, there is no need for the caller to change the contents
of vdr_side.

Device and Network Interfaces 193

dkio(7I)

Subsequent ioctl calls result in reads from the next valid side until all valid sides have been
read. On success, the kernel sets DKV_DMR_SUCCESS. The following table shows the values of
vdr_flags that are returned when an error occurs:
vdr_flags | vdr_side | Notes
-------------|-------------------|----------------------------
DKV_DMR_ERROR| DKV_SIDE_INIT | No valid side to read
DKV_DMR_DONE | Not Init side | All valid sides read
DKV_DMR_SHORT| Any value | Bytes requested cannot
be read. vdr_bytesread
set to bytes actually
read.
Typical code fragment:

enable->vc_set |= DKV_ABR_SET;
retval = ioctl(filedes, DKIOSETVOLCAP, enable);
if (retval != EINVAL || retval != ENOTSUP) {
if (info->vc_set & DKV_DMR_SET) {
dr->vdr_flags |= DKV_DMR_NEXT_SIDE;
dr->vdr_side = DKV_SIDE_INIT;
dr->vdr_nbytes = 1024;
dr->vdr_offset = 0xff00;
do {
rval =ioctl(fildes, DKIODMR, dr);
if (rval != EINVAL) {
/* Process data */
}
} while (rval != EINVAL || dr->vdr_flags &
(DKV_DMR_DONE | DKV_DMR_ERROR | DKV_DMR_SHORT)
}
}

RETURN VALUES Upon successful completion, the value returned is 0. Otherwise, -1 is returned and errno is set
to indicate the error.

x86 Only The following ioctl() requests set and/or retrieve the current disk controller, partitions, or
geometry information on the x86 architecture.
DKIOCG_PHYGEOM
The argument is a pointer to a dk_geom structure (described below). This ioctl() gets the
driver's notion of the physical geometry of the disk drive. It is functionally identical to the
DKIOCGGEOM ioctl().
DKIOCG_VIRTGEOM
The argument is a pointer to a dk_geom structure (described below). This ioctl() gets the
controller's (and hence the driver's) notion of the virtual geometry of the disk drive. Virtual
geometry is a view of the disk geometry maintained by the firmware in a host bus adapter
or disk controller. If the disk is larger than 8 Gbytes, this ioctl fails because a CHS-based

194 man pages section 7: Device and Network Interfaces • Last Revised 16 Dec 2009
 dkio(7I)

geometry is not relevant or useful for this drive.

/*
* Definition of a disk’s geometry
*/
*/struct dk_geom {
unsigned shor dkg_ncyl; /* # of data cylinders */
unsigned shor dkg_acyl; /* # of alternate cylinders */
unsigned short dkg_bcyl; /* cyl offset (for fixed head */
/* area) */
unsigned short dkg_nhead; /* # of heads */
unsigned short dkg_obs1; /* obsolete */
unsigned short dkg_nsect; /* # of sectors per track*/
unsigned short dkg_intrlv; /* interleave factor */
unsigned short dkg_obs2; /* obsolete */
unsigned short dkg_obs3; /* obsolete */
unsigned short dkg_apc; /* alternates per cylinder */
/* (SCSI only) */
unsigned short dkg_rpm; /* revolutions per min*/
unsigned short dkg_pcyl; /* # of physical cylinders */
unsigned short dkg_write_reinstruct; /* # sectors to skip, writes*/
unsigned short dkg_read_reinstruct; /* # sectors to skip, reads*/
unsigned short dkg_extra[7]; /* for compatible expansion*/
};
DKIOCADDBAD
This ioctl() forces the driver to re-examine the alternates slice and rebuild the internal
bad block map accordingly. It should be used whenever the alternates slice is changed by
any method other than the addbadsec(1M) or format(1M) utilities. DKIOCADDBAD can only
be used for software remapping on IDE drives; SCSI drives use hardware remapping of
alternate sectors.
DKIOCPARTINFO
The argument is a pointer to a part_info structure (described below). This ioctl() gets
the driver's notion of the size and extent of the partition or slice indicated by the file
descriptor argument.
/*
* Used by applications to get partition or slice information
*/
struct part_info {
daddr_t p_start;
int p_length;
};
DKIOCEXTPARTINFO
The argument is a pointer to an extpart_info structure (described below). This ioctl gets
the driver's notion of the size and extent of the partition or slice indicated by the file
descriptor argument. On disks larger than 1TB, this ioctl must be used instead of
DKIOCPARTINFO.

Device and Network Interfaces 195

dkio(7I)

/*
* Used by applications to get partition or slice information
*/
struct extpart_info {
diskkaddr_t p_start;
diskaddr_t p_length;
};
DKIOCSETEXTPART
This ioctl is used to update the in-memory copy of the logical drive information
maintained by the driver. The ioctl takes no arguments. It causes a re-read of the partition
information and recreation of minor nodes if required. Prior to updating the data
structures, the ioctl ensures that the partitions do not overlap. Device nodes are created
only for valid partition entries. If there is any change in the partition offset, size or ID from
the previous read, the partition is deemed to have been changed and hence the device
nodes are recreated. Any modification to any of the logical partitions results in the
recreation of all logical device nodes.

See Also addbadsec(1M), fdisk(1M), format(1M), ioctl(2), cdio(7I), cmdk(7D), fdio(7I), hdio(7I),
sd(7D)

Notes Blocksize information provided in DKIOCGMEDIAINFO is the size (in bytes) of the device's basic
unit of operation and can differ from the blocksize that the Solaris operating environment
exports to the user. Capacity information provided in the DKIOCGMEDIAINFO are for reference
only and you are advised to use the values returned by DKIOCGGEOM or other appropriate ioctl
for accessing data using the standard interfaces.

For x86 only: If the DKIOCSMBOOT command is used to modify the Solaris partitions, the VTOC
information should also be set appropriately to reflect the changes to partition. Failure to do so
leads to unexpected results when the device is closed and reopened fresh at a later time. This is
because a default VTOC is assumed by driver when a Solaris partition is changed. The default
VTOC persists until the ioctl DKIOCSVTOC is called to modify VTOC or the device is closed and
reopened. At that point, the old valid VTOC is read from the disk if it is still available.

196 man pages section 7: Device and Network Interfaces • Last Revised 16 Dec 2009
 dlcosmk(7ipp)

Name dlcosmk – Data Layer Class of Service Marker

Description The dlcosmk marker is an action module that is executed as a result of classifying or metering
packets. It marks the packet with a user priority defined by the IEEE 801.D standard. This
feature is only possible on a VLAN device.

The 3-bit user priority is part of the 802.1Q VLAN header tag that is part of the ethernet
header (carrying the IP packet).

Statistics The dlcosmk module exports the following statistics through kstat:

Global statistics:
module: dlcosmk instance: <action id>
name: dlcosmk statistics class <action name>
crtime
snaptime
b_band <b_band value>
dl_max <dl_max value>
usr_pri <configured CoS>
npackets <number of packets>
epackets <number of packets in error>
ipackets <number of packets not processed>
Files /kernel/ipp/sparcv9/dlcosmk 64-bit module (SPARC only.)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos

See Also ipqosconf(1M), dscpmk(7ipp), flowacct(7ipp), ipqos(7ipp), ipgpc(7ipp), tokenmt(7ipp),

tswtclmt(7ipp)

Device and Network Interfaces 197

dlpi(7P)

Name dlpi – Data Link Provider Interface

Synopsis #include <sys/dlpi.h>

Description SunOS STREAMS-based device drivers wishing to support the STREAMS TCP/IP and other
STREAMS-based networking protocol suite implementations support Version 2 of the Data
Link Provider Interface (DLPI). DLPI V2 enables a data link service user to access and use any
of a variety of conforming data link service providers without special knowledge of the
provider's protocol. Specifically, the interface is intended to support Ethernet, X.25 LAPB,
SDLC, ISDN LAPD, CSMA/CD, FDDI, token ring, token bus, Bisync, and other datalink-level
protocols.

The interface specifies access to the data link service provider in the form of M_PROTO and
M_PCPROTO type STREAMS messages and does not define a specific protocol implementation.
The interface defines the syntax and semantics of primitives exchanged between the data link
user and the data link provider to attach a physical device with physical-level address to a
stream, bind a datalink-level address to the stream, get implementation-specific information
from the data link provider, exchange data with a peer data link user in one of three
communication modes (connection, connectionless, acknowledged connectionless),
enable/disable multicast group and promiscuous mode reception of datalink frames, get and
set the physical address associated with a stream, and several other operations.

Solaris conforms to The Open Group Technical Standard for DLPI, Version 2. For free access
to this specification, point your browser to www.opengroup.org/pubs/catalog/c811.htm. Solaris
also provides extensions to the DLPI standard, as detailed in this man page.
Solaris-SPECIFIC Notification Support Enables DLPI consumers to register for
Dlpi Extensions notification when events of interest occur at the
DLPI provider. The negotiation can be
performed on any attached DLPI stream, and
begins with the DLPI consumer, sending a
DL_NOTIFY_REQ to the provider, which is an
M_PROTO message with the following payload:
typedef struct {
t_uscalar_t dl_primitive;
uint32_t dl_notifications;
uint32_t dl_timelimit;
} dl_notify_req_t;
The dl_primitive field must be set to
DL_NOTIFY_REQ; the dl_timelimit field is
reserved for future use and must be set to zero.
The dl_notifications field is a bitmask containing
the event types the consumer is interested in
receiving, and must be zero or more of:

198 man pages section 7: Device and Network Interfaces • Last Revised 7 Sep 2009
 dlpi(7P)

DL_NOTE_LINK_DOWN Notify when link has gone down

DL_NOTE_LINK_UP Notify when link has come up
DL_NOTE_PHYS_ADDR Notify when address changes
DL_NOTE_SDU_SIZE Notify when MTU changes
DL_NOTE_SPEED Notify when speed changes
DL_NOTE_PROMISC_ON_PHYS Notify when DL_PROMISC_PHYS is set
DL_NOTE_PROMISC_OFF_PHYS Notify when DL_PROMISC_PHYS is cleared
Consumers might find it useful to send a
DL_NOTIFY_REQ message with no requested
types to check if the DLPI provider supports the
extension.

Upon receiving the DL_NOTIFY_REQ, the DLPI

provider must generate a DL_NOTIFY_ACK,
which is an M_PROTO message with the
following payload:
typedef struct {
t_uscalar_t dl_primitive;
uint32_t dl_notifications;
} dl_notify_ack_t;
The dl_primitive field must be set to
DL_NOTIFY_ACK. The dl_notifications
field must include any notifications that the
provider supports, along with any other
unrequested notifications that the provider
supports. However, regardless of the notifications
the provider supports, it is restricted to sending
only DL_NOTIFY_IND messages (see below)
that were requested in the DL_NOTIFY_REQ.

Since there are additional notification types

which are not yet available for public use, DLPI
consumers and providers must take care when
inspecting and setting the dl_notifications
field. Specifically, consumers must be careful to
only request the above notification types, and
providers must be careful to not include any
unrecognized notification types in the
dl_notifications field when constructing the
DL_NOTIFY_ACK. In addition,
DL_NOTIFY_IND's that are received with
undocumented dl_notification or dl_data values
must be ignored.

Device and Network Interfaces 199

dlpi(7P)

DLPI consumers might receive a

DL_ERROR_ACK message (with
dl_error_primitive set to DL_NOTIFY_REQ) in
response to the initial DL_NOTIFY_REQ
message. This message indicates that the DLPI
provider does not support the DLPI notification
extension. Otherwise, the DLPI consumer
receives a DL_NOTIFY_ACK and should expect
to receive DL_NOTIFY_IND messages for any
types that it requested that were still set in it. The
DL_NOTIFY_IND is an M_PROTO message
with the following payload:
typedef struct {
t_uscalar_t dl_primitive;
uint32_t dl_notification;
uint32_t dl_data;
t_uscalar_t dl_addr_length;
t_uscalar_t dl_addr_offset;
} dl_notify_ind_t;
The dl_primitive field must be set to
DL_NOTIFY_IND, and the dl_notification
field must be set to the event type that has
occurred (for example,
DL_NOTE_LINK_DOWN). Only a single event
type can be set in each DL_NOTIFY_IND.

For the DL_NOTE_SPEED event type, dl_data

must be set to the current interface speed in
kilobits per second. For the
DL_NOTE_PHYS_ADDR event type, dl_data
must be set to DL_CURR_PHYS_ADDR. For the
DL_NOTE_SDU_SIZE event type, dl_data must
be set to the current MTU in bytes. Otherwise,
dl_data must be set to zero.

For the DL_NOTE_PHYS_ADDR event type, the

dl_addr_length field must be set to the length of
the address, and the dl_addr_offset field must
be set to offset of the first byte of the address,
relative to b_rptr (for example, if the address
immediately follows the dl_notify_ind
structure, dl_addr_offset is set to 'sizeof
(dl_notify_ind)'). For all other event types, the
dl_addr_length and dl_addr_offset fields

200 man pages section 7: Device and Network Interfaces • Last Revised 7 Sep 2009
 dlpi(7P)

must be set to zero by DLPI providers and

ignored by DLPI consumers.

In addition to generating DL_NOTIFY_IND

messages when a requested event has occurred,
the DLPI provider must initially generate one or
more DL_NOTIFY_IND messages to notify the
DLPI consumer of the the current state of the
interface. For instance, if the consumer has
requested DL_NOTE_LINK_UP |
DL_NOTE_LINK_DOWN, the provider must
send a DL_NOTIFY_IND containing the current
state of the link (either DL_NOTE_LINK_UP or
DL_NOTE_LINK_DOWN) after sending the
DL_NOTIFY_ACK.

For the initial DL_NOTIFY_IND message, the

DLPI provider is strongly recommended against
sending DL_NOTE_LINK_DOWN, even if the
interface is still initializing and is not yet ready to
send or receive packets. Instead, either delaying
the DL_NOTIFY_IND message until the
interface is ready or optimistically reporting
DL_NOTIFY_LINK_UP and subsequently
reporting DL_NOTE_LINK_DOWN if the
negotation fails is strongly preferred. This
prevents DL_NOTIFY_IND consumers from
needlessly triggering network failover operations
and logging error messages during network
interface initialization.

The DLPI provider must continue to generate

DL_NOTIFY_IND messages until it receives a
new DL_NOTIFY_REQ message or the DLPI
stream is detached (or closed). Further, a DLPI
style 2 provider must keep track of the requested
events after a DL_DETACH_REQ operation, and
if a subsequent DL_ATTACH_REQ is received, it
must send gratuitous DL_NOTIFY_IND
messages to notify the consumer of the current
state of the device, since the state might have
changed while detached (or the consumer might
have simply discarded its previous state).

Device and Network Interfaces 201

dlpi(7P)

Passive Consumers of Aggregated Links Solaris link aggregations as configured by

dladm(1M) export DLPI nodes for both the link
aggregation, and individual links that comprises
the aggregation, to allow observability of the
aggregated links. To allow applications such as
snoop(1M) to open those individual aggregated
links while disallowing other consumers such as
ip(7P), DL_PASSIVE_REQ (a DLPI primitive),
must be issued by snoop(1M) and similar
applications.

The DL_PASSIVE_REQ primitive is an

M_PROTO message containing the following
payload:
typedef struct {
t_uscalar_t dl_primitive;
} dl_passive_req_t;
Issuing this primitive allows the consumer of a
DLPI link to coexist with a link aggregation that
also uses the link. Such a consumer is considered
passive.

Consumers that don't use this primitive while an

aggregation is using the link receive
DL_SYSERR/EBUSY when issuing the following
DLPI primitives:
DL_BIND_REQ
DL_ENABMULTI_REQ
DL_PROMISCON_REQ
DL_AGGR_REQ
DL_UNAGGR_REQ
DL_CONTROL_REQ
DL_SET_PHYS_ADDR_REQ
A consumer that has not issued a
DL_PASSIVE_REQ and has successfully issued
one of the above primitives is considered active.

The creation of a link aggregation using

dladm(1M) fails if one of the links included in the
aggregation has an active consumer, but
succeeds if the links do not have any DLPI
consumers or only passive consumers.

202 man pages section 7: Device and Network Interfaces • Last Revised 7 Sep 2009
 dlpi(7P)

Raw Mode The DLIOCRAW ioctl function is used by some

DLPI applications, most notably the snoop(1M)
command. The DLIOCRAW command puts the
stream into a raw mode, which, upon receive,
causes the the full MAC-level packet to be sent
upstream in an M_DATA message instead of it
being transformed into the DL_UNITDATA_IND
form normally used for reporting incoming
packets. Packet SAP filtering is still performed on
streams that are in raw mode. If a stream user
wants to receive all incoming packets it must also
select the appropriate promiscuous modes. After
successfully selecting raw mode, the application
is also allowed to send fully formatted packets to
the provider as M_DATA messages for
transmission. DLIOCRAW takes no arguments.
Once enabled, the stream remains in this mode
until closed.
Native Mode Some DLPI providers are able to represent their
link layer using more than one link-layer format.
In this case, the default link-layer format can
minimize impact to applications, but might not
allow truly native link-layer headers to be sent
or received. DLPI consumers who wish to use the
native link-layer format can use DLIOCNATIVE
to transition the stream. DLIOCNATIVE takes
no arguments and returns the DLPI mac type
associated with the new link-layer format upon
success. Once enabled, the stream remains in this
mode until closed. Note that DLIOCNATIVE
does not enable transition between dissimilar
DLPI mac types and (aside from the link-layer
format), the new DLPI mac type is guaranteed to
be semantically identical. In particular, the SAP
space and addressing format are not affected and
the effect of DLIOCNATIVE is only visible when
in raw mode, though any subsequent
DL_INFO_REQ requests generate responses
with dl_mac_type set to the native DLPI type.
Margin While a DLPI provider provides its maximum
SDU via dl_max_sdu in DL_INFO_ACK
messages, this value typically represents a

Device and Network Interfaces 203

dlpi(7P)

standard maximum SDU for the provider's

media (1500 for Ethernet for example), and not
necessarily the absolute maximum amount of
data that the provider is able to transmit in a
given data unit. The margin “is the extra amount
of data in bytes that the provider can transmit
beyond its advertised maximum SDU. For
example, if a DL_ETHER provider can handle
packets whose payload section is no greater than
1522 bytes and its dl_max_sdu is set to 1500 (as is
typical for Ethernet), then the margin would be
22. If a provider supports a non-zero margin, it
implements the DLIOCMARGININFO ioctl,
whose data is a t_uscalar_t representing the
margin size.

Dl_ETHER-SPECIFIC
Dlpi Semantics
VLAN Support

Traditional VLAN Access

Some DL_ETHER DLPI providers support IEEE 802.1Q Virtual LANs (VLAN). For these
providers, traffic for a particular VLAN can be accessed by opening a VLAN data-link.

Unless raw mode is enabled, a DLPI stream bound to a VLAN data-link behaves no differently
than a traditional DLPI stream. As with non-VLAN data-link access, data must be sent to a
DLPI provider without link-layer headers (which are added by the provider) and received data
is passed to interested DLPI consumers without link-layer headers. As a result, DLPI
consumers not require special-case logic to implement VLAN access.

SAP-Based VLAN Access

As per IEEE 802.1Q, all VLAN traffic is sent using Ether- Type 0x8100, meaning that in
addition to directly opening a VLAN data-link, all VLAN traffic for a given underline
data-link can also be accessed by opening the underlying data-link and binding to SAP
0x8100. Accordingly, all VLAN traffic (regardless of VLAN ID) can be sent and received by
the DLPI consumer. However, even when raw mode is disabled, packets are received starting
with their VLAN headers and must be sent to the DLPI provider with their VLAN headers
already pre-pended (but without Ethernet headers). Because adhering to these semantics
requires each DLPI consumer to have specialized knowledge of VLANs, VLANs should only
be accessed in this way when the traditional VLAN access method is insufficient (for example,
because access to all VLAN traffic, regardless of VLAN ID, is needed).

Because all VLAN traffic is sent with SAP 0x8100, VLAN traffic not filtered at the physical
(DL_PROMISC_PHYS) level is also visible if a DLPI consumer enables promiscuous mode of a

204 man pages section 7: Device and Network Interfaces • Last Revised 7 Sep 2009
 dlpi(7P)

stream at the DL_PROMISC_SAP level. As mentioned earlier, these packets are received starting
with their VLAN headers if raw mode is not enabled.

QoS Support
The IEEE 802.1Q standard defines eight classes of priority values used by QoS traffic control of
Ethernet packets. Although the priority values are encoded in the 802.1Q tags, they can be
used independently from VLANs. In particular, a special priority tagged packet (with VLAN
ID zero but priority bits non-zero) does not belong to any VLAN.

The priority value can be set on either a per-stream or per-packet basis. DLPI consumers can
specify the per-stream priority using the DL_UDQOS_REQ request (the priority value remains
unchanged until the next DL_UDQOS_REQ) and also specify the per-packet priority value
using the b_band field of a M_DATA message or the dl_priority field of a DL_UNITDATA_REQ.

Raw Mode

SAP-Based VLAN Access

When raw mode is enabled, the complete, unmodified MAC- level packet (including Ethernet
and VLAN headers) is passed to interested DLPI consumers. Similarly, the entire MAC-level
packet (including Ethernet and VLAN headers) must be sent to the DLPI provider for
transmission. Note that the priority value specified in the b_band field can be overridden by
encoding the priority value (if any) into the VLAN header.

Traditional VLAN Access

When raw mode is enabled, only packets with the correct VLAN ID are passed up to
interested DLPI consumers. With the exception of priority-tagged packets, DLPI providers
must strip off the VLAN headers (while retaining the preceding Ethernet headers) before
sending up the packets. For priority-tagged packets, DLPI providers must use the reserved tag
0 to encode the VLAN TCI and send up the packets.

On the transmit-side, DLPI consumers must send the packets down to the DLPI providers
without the VLAN headers (but with the Ethernet headers) unless certain QoS support is
required. If QoS support is needed, the packet can have the VLAN header to indicate the
priority value, however its VLAN ID must be zero. The DLPI providers then insert the VLAN
tags or encode the VLAN tags using the priority value specified in the VLAN headers and send
the packets.

Files Files in or under /dev.

Attributes See attributes(5) for descriptions of the following attributes:

Device and Network Interfaces 205

dlpi(7P)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Committed

(Notification support/Passive mode behavior)

See Also dladm(1M), snoop(1M), libdlpi(3LIB), gld(7D), ip(7P)

Notes A Solaris DLPI link name consists of a DLPI provider name followed by a numeric PPA
(physical point of attachment).

The DLPI provider name must be between 1 and 16 characters in length, though names
between 3 and 8 characters are preferred. The DLPI provider name can consist of any
alphanumeric character (a-z, A-Z, 0-9), and the underscore (_). The first and last character of
the DLPI provider name cannot be a digit.

The PPA must be a number between 0 and 4294967294 inclusive. Leading zeroes are not
permitted.

206 man pages section 7: Device and Network Interfaces • Last Revised 7 Sep 2009
 dm2s(7D)

Name dm2s – loadable STREAMS driver

Synopsis dm2s@0

Description The dm2s module is a loadable STREAMS driver that provides synchronous serial
communication support for DSCP communication. dm2s is specific to the SPARC Enterprise
Server family.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdscpr

Interface Stability Private

See Also attributes(5)

Device and Network Interfaces 207

dmd(7D)

Name dmd – Driver for MMS drive managers

Synopsis /devices/pseudo/dmd@0:watcher
/devices/pseudo/dmd@0:ndrm
/devices/pseudo/dmd@0:stat

Description The Media Management System (MMS) is a distributed removable media management
system based on the IEEE 1244 Media Management System specification. MMS provides a
uniform, consistent tape interface to client applications.

Drives managed by MMS have a drive manager on each host that processes tapes and accesses
the drive. The drive manager provides device-specific operations and label processing. (Note:
only ANSI labels are currently supported.) The handle returned by the MMS MMP mount
command belongs to the drive manager driver (dmd). When dmd receives an open or close
system call, it passes it to the drive manager for processing. The drive manager then validates
the file according to the file disposition flags. Note that only O_RDONLY, O_WRONLY and
O_RDWR flags are used in validation. If O_APPEND is specified, the drive manager positions
the file as if append is specified in the mount command and the record format and blocksize
are set.

ioctls The Device Manager driver supports the MTIOCTOP, MTREW, MTRETEN, MTFSF,
MTFSR, MTBSF, MTBSR, MTWEOF, MTEOM, MTSRSZ, MTGRSZ, MTIOCGET and
MTIOCLRERR driver ioctls. In addition, MMS provides the ioctls shown below for the device
manager driver. (Note that MMS_LOCATE is supported in both mms and raw modes and
MMS_LOCATE is supported only in mms mode).
MMS_BLK_LIMIT - returns the block limit of a drive.
arg:

typedef struct mms_blk_limit {

uint64_t mms_max; /* Max blocksize */
uint32_t mms_min; /* Min blocksize */
uint32_t mms_gran; /* granularity */
} mms_blk_limit_t;

MMS_GET_POS - gets the current position.

arg:

typedef struct mms_pos {

uint32_t mms_type; /* LBN or non LBN) */
uint64_t mms_pos;
} mms_pos_t;

/* Value of mms_pos_type */
#define MMS_LBYTEN 1 /* logical byte */
#define MMS_LBLKN 2 /* logical block */

208 man pages section 7: Device and Network Interfaces • Last Revised 17 Nov 2008
 dmd(7D)

MMS_LOCATE - locate to a specific location.

arg: arg must be a mms_pos_t obtained from a
previous MMS_GET_POS.

MMS_GET_CAPACITY - returns the tape’s capacity.

arg:
typedef struct mms_capacity {
/*
* Capacity is in megabytes (1048576)
*/
uint64_t mms_max; /* cartridge capacity */
uint64_t mms_avail; /* amount available */
/* from EOD */
uint32_t mms_pc_avail; /* percent available */
} mms_capacity_t;

MMS_SET_DENSITY - sets the tape’s density.

arg:
typedef struct mms_density {
uint32_t mms_den;
} mms_density_t;

MMS_GET_DENSITY - gets the current tape density.

arg:
typedef struct mms_density {
uint32_t mms_den;
} mms_density_t;

MMS_INQUIRY - gets inquiry data of drive.

arg:
typedef struct mms_inquiry {
uchar_t mms_inq[64];
} mms_inquiry_t;

Status of dmd

MMS uses the MTIOCGET ioctl to report status.

The status code in mt_dsreg is defined as follows:

/usr/include/sys/scsi/generic/status.h

#define STATUS_MASK 0x3E

#define STATUS_GOOD 0x00
#define STATUS_CHECK 0x02
#define STATUS_MET 0x04
#define STATUS_BUSY 0x08
#define STATUS_INTERMEDIATE 0x10

Device and Network Interfaces 209

dmd(7D)

#define STATUS_SCSI2 0x20

#define STATUS_INTERMEDIATE_MET 0x14
#define STATUS_RESERVATION_CONFLICT 0x18
#define STATUS_TERMINATED 0x22
#define STATUS_QFULL 0x28
#define STATUS_ACA_ACTIVE 0x30

The error code in mt_erreg are defined

as follows:

/usr/include/sys/scsi/generic/sense.h

/*
* Sense Key values for Extended Sense.
*/

#define KEY_NO_SENSE 0x00

#define KEY_RECOVERABLE_ERROR 0x01
#define KEY_NOT_READY 0x02
#define KEY_MEDIUM_ERROR 0x03
#define KEY_HARDWARE_ERROR 0x04
#define KEY_ILLEGAL_REQUEST 0x05
#define KEY_UNIT_ATTENTION 0x06
#define KEY_WRITE_PROTECT 0x07
#define KEY_DATA_PROTECT KEY_WRITE_PROTECT
#define KEY_BLANK_CHECK 0x08
#define KEY_VENDOR_UNIQUE 0x09
#define KEY_COPY_ABORTED 0x0A
#define KEY_ABORTED_COMMAND 0x0B
#define KEY_EQUAL 0x0C
#define KEY_VOLUME_OVERFLOW 0x0D
#define KEY_MISCOMPARE 0x0E
#define KEY_RESERVED 0x0F

/usr/include/sys/scsi/impl/sense.h:
/* Status returned by driver */
#define SUN_KEY_FATAL 0x10 /* handshake failure */
#define SUN_KEY_TIMEOUT 0x11 /* command timeout */
#define SUN_KEY_EOF 0x12 /* eof hit */
#define SUN_KEY_EOT 0x13 /* eot hit */
#define SUN_KEY_LENGTH 0x14 /* length error */
#define SUN_KEY_BOT 0x15 /* bot hit */
#define SUN_KEY_WRONGMEDIA 0x16 /* wrong tape media */
#define NUM_IMPL_SENSE_KEYS 7 /* seven extra keys */

In MMS mode, the following values may be in mt_erreg when using

210 man pages section 7: Device and Network Interfaces • Last Revised 17 Nov 2008
 dmd(7D)

mms mode. They are defined in mms.h

#define MMS_KEY_BOF 0xe0 /* beginning of file. */

#define MMS_KEY_EOF 0xe1 /* end of file. */
Files /kernel/drv/dmd 32–bit kernel module (x86)
/kernel/drv/sparcv9/dmd 64–bit kernel module (SPARC)
/kernel/drv/amd64/dmd 64–bit kernel module (x86)
/kernel/drv/dmd.conf Configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWmmsr

Interface Stability Uncommitted

See Also mt(1), mmsadm(1M), attributes(5), dda(7D), st(7D), mtio(7I)

IEEE 1244 Removable Media Standards Specification — IEEE, 2000

Device and Network Interfaces 211

dmfe(7D)

Name dmfe – Davicom Fast Ethernet driver for Davicom DM9102A

Synopsis /kernel/drv/sparcv9/dmfe

Description The dmfe Ethernet device provides 100Base-TX networking interfaces using the Davicom
DM9102A chip, which incorporates its own internal transceiver.

The dmfe driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting. Multiple controllers
installed within the system are supported by the driver.

The 100Base-TX standard specifies an auto-negotiation protocol to automatically select the

mode and speed of operation. The internal transceiver is capable of performing
auto-negotiation with the remote-end of the link (link partner) and receives the capabilities of
the remote end. It selects the highest common denominator mode of operation based on the
priorities. The internal transceiver also supports a forced-mode of operation under which the
driver selects the operational mode.

Application The /dev/dmfe cloning character-special device is used to access all Davicom DM9102A
Programming devices installed in the system.
Interface
You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (ppa). The ppa ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. If the ppa field value does not
correspond to a valid device instance number for this system, an error (DL_ERROR_ACK) is
returned. The device is initialized on first attach and de-initialized (stopped) at last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to a

DL_INFO_REQ are as follows:
■ Maximum SDU is 1500 (ETHERMTU - defined in sys/ethernet.h).
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ The sap length value is -2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.
■ The broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.

Configuration By default, the dmfe driver performs auto-negotiation to select the speed and mode of the link.
Link speed and mode can be 100 Mbps (full or half-duplex) or 10 Mbps (full or half-duplex) as
described in the 100Base-TX standard.

212 man pages section 7: Device and Network Interfaces • Last Revised 17 Sep 2007
 dmfe(7D)

The auto-negotiation protocol automatically selects speed mode (either 100 Mbps or 10
Mbps) and operation mode (either full-duplex or half-duplex) as the highest common
denominator supported by both link partners. Because the dmfe device supports all modes,
this effectively selects the highest-throughput mode supported by the other device.

Alternatively, you can explicitly specify the link parameters by adding entries to the dmfe
driver configuration file (/kernel/drv/dmfe.conf). You can set the speed parameter to 10 or
100 to force dmfe devices to operate at the specified speed. Additionally, you can set the
full-duplex parameter to 0 or 1 to disable or force full-duplex operation, respectively.

Note that specifying either "speed" or "full-duplex" explicitly disables auto-negotiation. To

enable the driver to determine the appropriate setting for each parameter, you should always
set both parameters. If it is necessary to force either speed or duplex setting (for example,
because the dmfe device is connected to an ancient device or hub that does not support
auto-negotiation), both parameters should be explicitly specified to match the requirements
of the external device.
Files /dev/dmfe Character special device
/kernel/drv/dmfe 32–bit kernel module (x86)
/kernel/drv/sparcv9/dmfe 64–bit kernel module (SPARC)
/kernel/drv/amd64/dmfe 64–bit kernel module (x86)
/kernel/drv/dmfe.conf dmfe configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also attributes(5), gld(7D), dlpi(7P), streamio(7I)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

Device and Network Interfaces 213

dnet(7D)

Name dnet – Ethernet driver for DEC 21040, 21041, 21140 Ethernet cards
Synopsis /kernel/drv/dnet

Description The dnet Ethernet driver is a multithreaded, loadable, clonable, STREAMS GLD driver.
Multiple controllers installed within the system are supported by the driver. The dnet driver
functions include controller initialization, frame transmit and receive, functional addresses,
promiscuous and multicast support, and error recovery and reporting.

Application The cloning character-special device, /dev/dnet, is used to access all DEC 21040/21041/21140
Programming devices installed in the system.
Interface
The dnet driver is dependent on /kernel/misc/gld, a loadable kernel module that provides
the dnet driver with the DLPI and STREAMS functionality required of a LAN driver. See
gld(7D) for more details on the primitives supported by the driver.

The device is initialized on the first attach and de-initialized (stopped) on the last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to a DL_INFO_REQ
from the user are as follows:
■ The maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h>).
■ The minimum SDU is 0.
■ The DLSAP address length is 8.
■ The MAC type is DL_ETHER.
■ The sap length value is −2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.
■ The broadcast address value is the Ethernet/IEEE broadcast address
(FF:FF:FF:FF:FF:FF).

Once in the DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.

Preconfiguration The PCI configuration process varies from system to system. Follow the instructions provided
by the vendor.
Known Problems and ■ On multiport cards (exception: Osicom (Rockwell) RNS2340), the first port is the top port.
Limitations (On the Osicom RNS2340, the first port is the bottom port.)
■ If the dnet driver fails to determine the correct speed and duplex mode resulting in a
corresponding drop in performance, set the speed and duplex mode using the dnet.conf
file.
■ The dnet driver incorrectly counts carrier lost or no carrier errors while in full-duplex
mode. There is no carrier signal present when in full-duplex mode and it should not be
counted as an error.
■ Version 4 SROM formats are not supported.

214 man pages section 7: Device and Network Interfaces • Last Revised 20 OCT 2000
 dnet(7D)

Configuration The /kernel/drv/dnet.conf file supports the following options:

full-duplex For full duplex operation use full-duplex=1, for half duplex use
full-duplex=0. Half-duplex operation gives better results on older 10mbit
networks.
speed For 10mbit operation use speed=10, for 100mbit operation use speed=100.
Certain 21140 based cards will operate at either speed. Use the speed
property to override the 100mbit default in this case.
Files /dev/dnet character special device
/kernel/drv/dnet.conf dnet configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), dlpi(7P), gld(7D) streamio(7I)

Writing Device Drivers

STREAMS Programming Guide

Device and Network Interfaces 215

dpt(7D)

Name dpt – DPT ServeRAID IV SCSI host bus adapter and RAID adapter driver

Description The dpt driver is a driver for the DPT (Distributed Processing Technology) family of
SmartRAID IV SCSI HBA and RAID adapters. The following HBA adapters are supported:
PM2024, PM2044UW, PM2044W, PM2124, PM2124W, PM2144UW, and PM2144W.

The following RAID adapters are supported: PM3224, PM3224W, PM3334UW, and
PM3334W.
Preconfiguration ■ DPT PM3224 only: Only EPROM 7A and later versions are supported.
■ DPT PM2024 and PM2124 only: Only EPROM 6D4 and later versions are supported.
■ Use adapters with SmartROM version 3.B or later versions only.
■ Be sure that the controller board is installed in a PCI bus-mastering slot.
■ Disable PCI parity checking if your firmware version is earlier than version 7A, if your
system memory is ECC, or if your system does not check parity.

Known Problems and During system boot, a message may be displayed saying a DPT controller driver cannot be
Limitations installed. This message indicates that the motherboard installed in your system may contain
ECC memory or may not check parity. If you see this message is displayed, disable PCI parity
checking.
Supported Settings ■ I/O Address: Auto

Configuration Auto-configuration code determines whether the adapter is present at the configured address
and what types of devices are attached to it. The DPT ServeRAID is primarily used as a disk
array (system drive) controller.

To configure the attached disk arrays, you must configure the controller (using the
configuration utilities provided by the hardware manufacturer) before you boot the Solaris
operating environment. You use the configuration utilities to set RAID levels, stripe
parameters, cache mechanisms and perform other functions. For more information, see the
user manual supplied with your hardware.
Files /kernel/drv/dpt.conf dpt configuration file
/dev/dsk/cndn[s|p]n block device
/dev/rdsk/cndn[s|p]n raw device where:
cn controller n
dn LUN n (0–7)
sn UNIX system slice n (0–15)
pn fdisk(1M) partition (0)

216 man pages section 7: Device and Network Interfaces • Last Revised 27 November 2000
 dpt(7D)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also fdisk(1M), attributes(5), cmdk(7D)

Device and Network Interfaces 217

dr(7d)

Name dr, drmach, ngdr, ngdrmach – Sun Enterprise 10000 dynamic reconfiguration driver
Synopsis dr
drmach
ngdr
ngdrmach

Description The dynamic reconfiguration (DR) driver consists of a platform-independent driver and a
platform-specific module. The DR driver uses standard features of the Solaris operating
environment whenever possible to control DR operations and calls the platform specific
module as needed. The DR driver creates minor nodes in the file system that serve as
attachment points for DR operations.

The DR driver provides a pseudo-driver interface to sequence attach and detach operations on
system boards using file system entry points referred to as ”attachment points.” The
attachment point form depends on the platform.

Sun Enterprise 10000 On the Sun Enterprise 10000 server, the DR driver consists of a platform-independent driver
Server (ngdr) and a platform-specific module (ngdrmach).

The domain configuration server (DCS) accepts DR requests from the system services
processor (SSP) and uses the libcfgadm(3LIB) interface to initiate the DR operation. After the
operation is performed, the results are returned to the SSP. For more information about the
DCS on the Sun Enterprise 10000, refer to the dcs(1M) man page and the Sun Enterprise
10000 Dynamic Reconfiguration User Guide.

The DR driver creates physical attachment points for system board slots that takes the
following form:
/devices/pseudo/ngdr@0:SBx

Where x represents the slot number (0 to 15) for a particular board.

The cfgadm_sbd(1M) plugin creates dynamic attachment points that refer to components on
system boards, including CPUs, memory, or I/O devices. Refer to the cfgadm_sbd(1M) man
page for more details.

See Also cfgadm_sbd(1M), ioctl(2), libcfgadm(3LIB)

Sun Enterprise 10000 Dynamic Reconfiguration User Guide

218 man pages section 7: Device and Network Interfaces • Last Revised 29 Sep 2003
 dscpmk(7ipp)

Name dscpmk – Differentiated Services Code Point Marker

Description The dscpmk marker is an action module that is executed as a result of classifying or metering
packets. It sets a codepoint in the IP header as defined in RFC-2474: Definition of the
Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers.

Statistics The dscpmk module exports the following statistics available through kstat:

Global statistics:
module: dscpmk instance: <action id>
name: dscpmk stats class <action name>
crtime
snaptime
npackets <number of packets>
epackets <number of packets in error>
ipackets <number of packets not processed>
dscp_unchanged <number of packets with DSCP unchanged>
dscp_changed <number of packets with DSCP changed>

Also, for each DSCP the following is exported:

module: dscpmk instance: <action id>
name: dscpmk_dscp0x<DSCP> value class: <action name>
dscp <DSCP value>
npackets <number of packets for this DSCP>
Files /kernel/ipp/sparcv9/dscpmk 64-bit module (SPARC only.)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos

See Also ipqosconf(1M), dlcosmk(7ipp), flowacct(7ipp), ipqos(7ipp), ipgpc(7ipp), tokenmt(7ipp),

tswtclmt(7ipp)

RFC 2474, Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6
Headers K. Nichols, S. Blake, F. Baker, D. Black — The Internet Society, 1998

Device and Network Interfaces 219

dsp(7I)

Name dsp – generic audio device interface

Synopsis #include <sys/soundcard.h>

Description To record audio input, applications open() the appropriate device and read data from it using
the read() system call. Similarly, sound data is queued to the audio output port by using the
write(2) system call. Device configuration is performed using the ioctl(2) interface.

Because some systems can contain more than one audio device, application writers are
encouraged to open the /dev/mixer device and determine the physical devices present on the
system using the SNDCTL_SYSINFO and SNDCTL_AUDIOINFO ioctls. See mixer(7I). The user
should be provided a the ability to select a different audio device, or alternatively, an
environment variable such as AUDIODSP can be used. In the absence of any specific
configuration from the user, the generic device file, /dev/dsp, can be used. This normally
points to a reasonably appropriate default audio device for the system.

Opening the Audio The audio device is not treated as an exclusive resource.
Device
Each open() completes as long as there are channels available to be allocated. If no channels
are available to be allocated, the call returns -1 with the errno set to EBUSY.

Audio applications should explicitly set the encoding characteristics to match the audio data
requirements after opening the device, and not depend on any default configuration.

Recording Audio Data The read() system call copies data from the system's buffers to the application. Ordinarily,
read() blocks until the user buffer is filled. The poll(2) system call can be used to determine
the presence of data that can be read without blocking. The device can alternatively be set to a
non-blocking mode, in which case read() completes immediately, but can return fewer bytes
than requested. Refer to the read(2) manual page for a complete description of this behavior.

When the audio device is opened with read access, the device driver allocates resources for
recording. Since this consumes system resources, processes that do not record audio data
should open the device write-only (O_WRONLY).

The recording process can be stopped by using the SNDCTL_DSP_HALT_INPUT ioctl, which also
discards all pending record data in underlying device FIFOs.

Before changing record parameters, the input should be stopped using the
SNDCTL_DSP_HALT_INPUT ioctl, which also flushes the any underlying device input FIFOs.
(This is not necessary if the process never started recording by calling read(2). Otherwise,
subsequent reads can return samples in the old format followed by samples in the new format.
This is particularly important when new parameters result in a changed sample size.

Input data can accumulate in device buffers very quickly. At a minimum, it accumulates at
8000 bytes per second for 8-bit, 8 KHz, mono, u-Law data. If the device is configured for more
channels, higher sample resolution, or higher sample rates, it accumulates even faster. If the
application that consumes the data cannot keep up with this data rate, the underlying FIFOs
can become full. When this occurs, any new incoming data is lost until the application makes

220 man pages section 7: Device and Network Interfaces • Last Revised 11 May 2009
 dsp(7I)

room available by consuming data. Additionally, a record overrun is noted, which can be
retrieved using the SNDCTL_DSP_GETERROR ioctl.

Record volume for a stream can be adjusted by issuing the SNDCTL_DSP_SETRECVOL ioctl. The
volume can also be retrieved using the SNDCTL_DSP_GETRECVOL.

Playing Audio Data The write() system call copies data from an application's buffer to the device output FIFO.
Ordinarily, write() blocks until the entire user buffer is transferred. The device can
alternatively be set to a non-blocking mode, in which case write()completes immediately,
but might have transferred fewer bytes than requested. See write(2).

Although write() returns when the data is successfully queued, the actual completion of
audio output might take considerably longer. The SNDCTL_DSP_SYNC ioctl can be issued to
allow an application to block until all of the queued output data has been played.

The final close(2) of the file descriptor waits until all of the audio output has drained. If a
signal interrupts the close(), or if the process exits without closing the device, any remaining
data queued for audio output is flushed and the device is closed immediately.

The output of playback data can be halted entirely, by calling the SNDCTL_DSP_HALT_OUTPUT
ioctl. This also discards any data that is queued for playback in device FIFOs.

Before changing playback parameters, the output should be drained using the
SNDCTL_DSP_SYNC ioctl, and then stopped using the SNDCTL_DSP_HALT_OUTPUT ioctl, which
also flushes the any underlying device output FIFOs. This is not necessary if the process never
started playback, such as by calling write(2). This is particularly important when new
parameters result in a changed sample size.

Output data is played from the playback buffers at a default rate of at least 8000 bytes per
second for u-Law, A-Law or 8-bit PCM data (faster for 16-bit linear data or higher sampling
rates). If the output FIFO becomes empty, the framework plays silence, resulting in audible
stall or click in the output, until more data is supplied by the application. The condition is also
noted as a play underrun, which can be determined using the SNDCTL_DSP_GETERROR ioctl.

Playback volume for a stream can be adjusted by issuing the SNDCTL_DSP_SETPLAYVOL ioctl.
The volume can also be retrieved using the SNDCTL_DSP_GETPLAYVOL.

Asynchronous I/O The O_NONBLOCK flag can be set using the F_SETFL fcntl(2) to enable non-blocking read()
and write() requests. This is normally sufficient for applications to maintain an audio stream
in the background.

It is also possible to determine the amount of data that can be transferred for playback or
recording without blocking using the SNDCTL_DSP_GETOSPACE or SNDCTL_DSP_GETISPACE
ioctls, respectively.

Device and Network Interfaces 221

dsp(7I)

Mixer Pseudo-Device The /dev/mixer provides access to global hardware settings such as master volume settings,
etc. It is also the interface used for determining the hardware configuration on the system.

Applications should open(2) /dev/mixer, and use the SNDCTL_SYSINFO and

SNDCTL_AUDIOINFO ioctls to determine the device node names of audio devices on the system.
See mixer(7I) for additional details.

ioctls

Information IOCTLs The following ioctls are supported on the audio device, as well as the mixer device. See
mixer(7I) for details.
OSS_GETVERSION
SNDCTL_SYSINFO
SNDCTL_AUDIOINFO
SNDCTL_MIXERINFO
SNDCTL_CARDINFO

Audio IOCTLs The dsp device supports the following ioctl commands:

SNDCTL_DSP_SYNC The argument is ignored. This command suspends the

calling process until the output FIFOs are empty and all
queued samples have been played, or until a signal is
delivered to the calling process. An implicit
SNDCTL_DSP_SYNC is performed on the final close() of the
dsp device.

This ioctl should not be used unnecessarily, as if it is used in

the middle of playback it causes a small click or pause, as the
FIFOs are drained. The correct use of this ioctl is just before
changing sample formats.
SNDCTL_DSP_HALT
SNDCTL_DSP_HALT_INPUT
SNDCTL_DSP_HALT_OUTPUT The argument is ignored. All input or output (or both)
associated with the file is halted, and any pending data is
discarded.
SNDCTL_DSP_SPEED The argument is a pointer to an integer, indicating the sample
rate (in Hz) to be used. The rate applies to both input and
output for the file descriptor. On return the actual rate, which
can differ from that requested, is stored in the integer pointed
to by the argument. To query the configured speed without
changing it the value 0 can be used by the application
SNDCTL_DSP_GETFMTS The argument is a pointer to an integer, which receives a bit
mask of encodings supported by the device. Possible values
are

222 man pages section 7: Device and Network Interfaces • Last Revised 11 May 2009
 dsp(7I)

AFMT_MU_LAW 8-bit unsigned u-Law

AFMT_A_LAW 8-bit unsigned a-Law
AFMT_U8 8-bit unsigned linear PCM
AFMT_S16_LE 16-bit signed
little-endian linear PCM
AFMT_S16_BE 16-bit signed
big-endian linear PCM
AFMT_S16_NE 16-bit signed native-endian
linear PCM
AFMT_U16_LE 16-bit unsigned
little-endian linear PCM
AFMT_U16_BE 16-bit unsigned big-endian
linear PCM
AFMT_U16_NE 16-bit unsigned big-endian
linear PCM
AFMT_S24_LE 24-bit signed little-endian
linear PCM, 32-bit aligned
AFMT_S24_BE 24-bit signed big-endian
linear PCM, 32-bit aligned
AFMT_S24_NE 24-bit signed native-endian
linear PCM, 32-bit aligned
AFMT_S32_LE 32-bit signed little-endian
linear PCM
AFMT_S32_BE 32-bit signed big-endian
linear PCM
AFMT_S32_NE 32-bit signed native-endian
linear PCM
AFMT_S24_PACKED 24-bit signed little-endian
packed linear PCM
Not all devices support all of these encodings. This
implementation uses AFMT_S24_LE or AFMT_S24_BE,
whichever is native, internally.
SNDCTL_DSP_SETFMT The argument is a pointer to an integer, which indicates the
encoding to be used. The same values as for
SNDCTL_DSP_GETFMT can be used, but the caller can only
specify a single option. The encoding is used for both input
and output performed on the file descriptor.
SNDCTL_DSP_CHANNELS The argument is a pointer to an integer, indicating the
number of channels to be used (1 for mono, 2 for stereo, etc.)
The value applies to both input and output for the file
descriptor. On return the actual channel configuration
(which can differ from that requested) is stored in the integer

Device and Network Interfaces 223

dsp(7I)

pointed to by the argument. To query the configured

channels without changing it the value 0 can be used by the
application.
SNDDCTL_DSP_GETCAPS The argument is a pointer to an integer bit mask, which
indicates the capabilities of the device. The bits returned can
include
PCM_CAP_OUTPUT Device supports playback
PCM_CAP_INPUT Device supports recording
PCM_CAP_DUPLEX Device supports simultaneous
playback and recording
SNDCTL_DSP_GETPLAYVOL
SNDCTL_DSP_GETRECVOL The argument is a pointer to an integer to receive the volume
level for either playback or record. The value is encoded as a
stereo value with the values for two channels in the least
significant two bytes. The value for each channel thus has a
range of 0-100. In this implementation, only the low order
byte is used, as the value is treated as a monophonic value, but
a stereo value (with both channel levels being identical) is
returned for compatibility.
SNDCTL_DSP_SETPLAYVOL
SNDCTL_DSP_SETRECVOL The argument is a pointer to an integer indicating volume
level for either playback or record. The value is encoded as a
stereo value with the values for two channels in the least
significant two bytes. The value for each channel has a range
of 0-100. Note that in this implementation, only the low order
byte is used, as the value is treated as a monophonic value.
Portable applications should assign the same value to both
bytes
SNDCTL_DSP_GETISPACE
SNDCTL_DSP_GETOSPACE The argument is a pointer to a struct audio_buf_info,
which has the following structure:
typedef struct audio_buf_info {
int fragments;* /# of available fragments */
int fragstotal;
/* Total # of fragments allocated */
int fragsize;
/* Size of a fragment in bytes */
int bytes;
/* Available space in bytes */
/* Note! ’bytes’ could be more than
fragments*fragsize */
} audio_buf_info;

224 man pages section 7: Device and Network Interfaces • Last Revised 11 May 2009
 dsp(7I)

The fields fragments, fragstotal, and fragsize are

intended for use with compatible applications (and in the
future with mmap(2)) only, and need not be used by typical
applications. On successful return the bytes member contains
the number of bytes that can be transferred without blocking.
SNDCTL_DSP_CURRENT_IPTR
SNDCTL_DSP_CURRENT_OPTR The argument is a pointer to an oss_count_t, which has the
following definition:
typedef struct {
long long samples;
/* Total # of samples */
int fifo_samples;
/* Samples in device FIFO */
int filler[32];/* For future use */
} oss_count_t;
The samples field contains the total number of samples
transferred by the device so far. The fifo_samples is the
depth of any hardware FIFO. This structure can be useful for
accurate stream positioning and latency calculations.
SNDCTL_DSP_GETIPTR
SNDCTL_DSP_GETOPTR The argument is a pointer to a struct count_info, which
has the following definition:
typedef struct count_info {
unsigned int bytes;
/* Total # of bytes processed */
int blocks;
/* # of fragment transitions since
last time */
int ptr;/* Current DMA pointer value */
} count_info;
These ioctls are primarily supplied for compatibility, and
should not be used by most applications.
SNDCTL_DSP_GETODELAY The argument is a pointer to an integer. On return, the
integer contains the number of bytes still to be played before
the next byte written are played. This can be used for accurate
determination of device latency. The result can differ from
actual value by up the depth of the internal device FIFO,
which is typically 64 bytes.
SNDCTL_DSP_GETERROR The argument is a pointer to a struct audio_errinfo,
defined as follows:

Device and Network Interfaces 225

dsp(7I)

typedef struct audio_errinfo {

int play_underruns;
int rec_overruns;
unsigned int play_ptradjust;
unsigned int rec_ptradjust;
int play_errorcount;
int rec_errorcount;
int play_lasterror;
int rec_lasterror;
int play_errorparm;
int rec_errorparm;
int filler[16];
} audio_errinfo;
For this implementation, only the play_underruns and
rec_overruns values are significant. No other fields are used
in this implementation.

These fields are reset to zero each time their value is retrieved
using this ioctl.

Compatibility IOCTLS These ioctls are supplied exclusively for compatibility with existing applications. Their use is
not recommended, and they are not documented here. Many of these are implemented as
simple no-ops.
SNDCTL_DSP_POST
SNDCTL_DSP_STEREO
SNDCTL_DSP_SETDUPLEX
SNDCTL_DSP_LOW_WATER
SNDCTL_DSP_PROFILE
SNDCTL_DSP_GETBLKSIZE
SNDCTL_DSP_SUBDIVIDE
SNDCTL_DSP_SETFRAGMENT
SNDCTL_DSP_COOKEDMODE
SNDCTL_DSP_READCTL
SNDCTL_DSP_WRITECTL
SNDCTL_DSP_SILENCE
SNDCTL_DSP_SKIP
SNDCTL_DSP_POST
SNDCTL_DSP_GET_RECSRC
SNDCTL_DSP_SET_RECSRC
SNDCTL_DSP_SET_RECSRC_NAMES
SNDCTL_DSP_GET_PLAYTGT
SNDCTL_DSP_SET_PLAYTGT
SNDCTL_DSP_SET_PLAYTGT_NAMES
SNDCTL_DSP_GETTRIGGER
SNDCTL_DSP_SETTRIGGER

226 man pages section 7: Device and Network Interfaces • Last Revised 11 May 2009
 dsp(7I)

SNDCTL_AUDIOINFO_EX
SNDCTL_ENGINEINFO

Errors An open() fails if:

EBUSY The requested play or record access isbusy and either the O_NDELAY or O_NONBLOCK
flag was set in the open() request.
EINTR The requested play or record access is busy and a signal interrupted the open()
request.
EINVAL The device cannot support the requested play or record access.

An ioctl() fails if:

EINVAL The parameter changes requested in the ioctl are invalid or are not supported by
the device.

Files The physical audio device names are system dependent and are rarely used by programmers.
Programmers should use the generic device names listed below.
/dev/dsp Symbolic link to the system's primary audio device
/dev/mixer Symbolic link to the pseudo mixer device for the system
/dev/sndstat Symbolic link to the pseudo mixer device for the system
/usr/share/audio/samples Audio files

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWcsu, SUNWaudd, SUNWaudh

Stability Level Uncommitted

See Also close(2), fcntl(2), ioctl(2), mmap(2), open(2), poll(2), read(2), write(2), attributes(5),
audio(7D), mixer(7I)

Device and Network Interfaces 227

dtrace(7D)

Name dtrace – DTrace dynamic tracing facility

Description The dtrace driver provides the dynamic instrumentation and tracing facilities for the DTrace
software, as well as the built-in dtrace provider. The dtrace driver is not a public interface
and you access the instrumentation offered by this provider through DTrace tools such as
dtrace(1M). Refer to the Solaris Dynamic Tracing Guide for a description of the public
documented interfaces available for the DTrace facility and dtrace provider probes.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), libdtrace(3LIB), attributes(5)

Solaris Dynamic Tracing Guide

228 man pages section 7: Device and Network Interfaces • Last Revised 4 Sep 2003
 e1000g(7D)

Name e1000g, e1000 – Intel PRO/1000 Gigabit family of network interface controllers
Synopsis /dev/e1000g

Description The e1000g Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), over Intel PRO/1000
family of Gigabit controllers. This driver supports multiple Intel Gigabit controllers installed
within the system. The e1000g driver provides basic support including chip initialization,
frame transmit and receive, multicast support, and error recovery and reporting.

Application The cloning, character-special device /dev/e1000g is used to access all Intel Gigabit devices
Programming installed within the system.
Interface
The e1000g driver is managed by the dladm(1M) command line utility, which allows VLANs
to be defined on top of e1000g instances and for e1000g instances to be aggregated. See
dladm(1M) for more details.

You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are as follows:
■ Maximum SDU (with jumbo frame) is as high as 16298.
■ Minimum SDU is 0. The driver pads to the mandatory 60-octet minimum packet size.
■ The dlsap address length is 8.
■ MAC type is DL_ETHER.
■ The sap length value is −2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.
■ The broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Configuration The e1000g driver does not support the use of shared RAM on the board.

To configure the e1000g driver:

■ Use prtconf -v | grep pci8086,[12][01][01][0-F] to obtain the instance number of
the driver.
■ Use ifconfig e1000ginstance plumb to plumb the controller.
■ Use ifconfig e1000ginstance inet ip_address netmask + broadcast + -trailers up
to bring up the interface.
■ Use the ping(1M) command to contact interfaces on the network to verify that the
configuration is operational.

Device and Network Interfaces 229

e1000g(7D)

Configuration File The following e1000g.conf configuration options are supported:

Options
AutoNegAdvertised
This is a bitmap for the speeds advertised during auto-negotiation.
Bit | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0
Setting| N/A | N/A | 1000F | N/A | 100F | 100H | 10F | 10H
The adapter only auto-negotiates to a speed that is advertised. For example:
AutoNegAdvertised = 4 causes an adapter to only advertise auto-negotiation at 100 Mbps,
full duplex. No other link speeds are accepted or given during auto-negotiation.
AutoNegAdvertised=47 advertises all speeds available, This is the same as using the default
setting of 0.
0–255 Allowed values
0 Default
ForceSpeedDuplex
Specify the speed and duplex mode for each instance.

If you set ForceSpeedDuplex=7,4, the e1000g0 is set to auto-negotiate and e1000g1 is set
to 100 Mbps, full duplex. Note that fiber optic ethernet adapters ignore this setting.

Allowed values are:

1 10 Megabits per second, Half Duplex.
2 10 Megabits per second, Full Duplex.
3 100 Megabits per second, Half Duplex.
4 100 Megabits per second, Full Duplex.
7 Auto-negotiate speed and duplex. (Default).
MaxFrameSize
Upper limit on the maximum MTU size the driver allows. All Intel gigabit adapters (except
the 82542-based Intel PRO/1000 adapter) allow the configuration of jumbo frames.

For a Intel PRO/1000 adapter that is later than 82571 (including 82571) the maximum
MTU accepted by the MAC is 9216. For others, the maximum MTU accepted by the MAC
is 16298. Use ifconfig(1M) to configure jumbo frames. Using ifconfig with the adapter
instance and the mtu argument (ifconfig e1000g0 mtu 9216) configures adapter
e1000g0 for the maximum allowable jumbo frame size.

Allowed values are:

0 Standard ethernet frames with a MTU equal to 1500. (Default).
1 Jumbo frames with a maximum MTU of 4010.
2 Jumbo frames with a maximum MTU of 8106.

230 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 e1000g(7D)

3 Jumbo frames with a maximum MTU of 16298.

FlowControl
Flow control utilizes ethernet XON and unicast and multicast XOFF packets to allow
ethernet equipment to slow down the stream of data between two ethernet devices.
Allowed values are:
0 Disable. Packets can get dropped in high-throughput situations, leading to reduced
network performance.
1 Receive only.
2 Transmit only.
3 Receive and transmit. (Default).
4 Use adapter's EEPROM-programmed factory default setting.
TbiCompatibilityEnable
You must enable this feature on Intel 82543CG-based copper adapters to operate correctly
with TBI mode ethernet hardware.
Allowed values are:
0 Disable.
1 Enable. (Default).
SetMasterSlave
Controls the PHY master/slave setting. Manually forcing master or slave can reduce time
needed to link with Planex 08TX and IO data switches. This setting should remain as the
hardware default.
Allowed values are:
0 Hardware default. (Default).
1 Force master.
2 Force slave.
3 Force auto.

By default, the following configuration options are not displayed in the e1000g.conf file.
Although they are configurable, you should not change these options:
NumRxDescriptors Number of available receive descriptors. Multiple receive
descriptors increase receive performance, but decrease available
memory.
80–4096 Allowed values.
1024 Default.

Device and Network Interfaces 231

e1000g(7D)

NumTxDescriptors Number of transmit descriptors available to the driver. Mulitiple

transmit descriptors increase transmit performance, but
decrease available memory.
80–4096 Allowed values.
1024 Default.
NumRxFreeList Number of pre-allocated buffers that the driver can use for
received data. Pre-allocating buffers can improve receive
performance but decrease available memory.
60–4096 Allowed values.
1024 Default.
MaxNumReceivePackets Maximum number of receive packets that the driver can handle
for each interrupt.

CPU utilization can be lowered through more efficient interrupt

management. If this value is increased, the time needed by the
CPU to process the individual interrupts increases, thereby
nullifying any performance gains realized by handling less
interrupts.
0–1024 Allowed values.
32 Default.

Configuration Options In addition to the e1000g.conf file, you can also use the dladm(1M) command to configure
Using dladm(1M) the e1000g driver.

To view supported configuration parameters, do the following step:

# dladm show-linkprop e1000g0

In addition, the current settings of the parameters can be found using dladm show-ether.
Using dladm(1M), you can set the link speed/duplex using the enabled capability parameters
supported by the e1000g device. Each parameter contains a boolean value that determines if
the device enables that mode of operation. The adv_autoneg_cap parameter controls
auto-negotiation. When adv_autoneg_cap is set to 0, the driver forces the mode of operation
selected by the first non-zero parameter in priority order as shown below:
en_1000fdx_cap 1000Mbps full duplex
en_100fdx_cap 100Mpbs full duplex
en_100hdx_cap 100Mpbs half duplex
en_10fdx_cap 10Mpbs full duplex
en_10hdx_cap 10Mpbs half duplex

Note – The link mode of 1000Mbps half duplex is not supported.

232 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 e1000g(7D)

Forced link mode of 1000Mbps full duplex is not supported.

Setting all the enabled link capabilities to 0 results in the link being reset to auto-negotiation
with full link capabilities advertised.
1 10Mpbs half duplex
2 10Mpbs full duplex
3 100Mpbs half duplex
4 100Mpbs full duplex
Files dev/e1000g Character special device.
/kernel/drv/e1000g.conf Driver configuration file.
/kernel/drv/sparcv9/e1000g 64–bit driver binary (SPARC).
/kernel/drv/e1000g 32–bit driver binary (x86)
/kernel/drv/amd64/e1000g 64–bit driver binary. (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), ifconfig(1M), kstat(1M), ping(1M), attributes(5), dlpi(7P)

Intel PRO/1000 Gigabit Adapter Driver Installation Notes for Solaris

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Guide

Device and Network Interfaces 233

ecpp(7D)

Name ecpp – IEEE 1284 compliant parallel port driver

Synopsis #include <sys/types.h>
#include <sys/ecppio.h>
ecpp@unit-address

Description The ecpp driver provides a bi-directional interface to IEEE 1284 compliant devices as well as a
forward single-directional interface to Centronics devices. In addition to the Centronics
protocol, the ecpp driver supports the IEEE 1284Compatibility, Nibble, and ECP protocols.
ECPP_COMPAT_MODE and ECPP_CENTRONICS modes of operation have logically identical
handshaking protocols, however devices that support ECPP_COMPAT_MODE are IEEE 1284
compliant devices. IEEE 1284 compliant devices support at least ECPP_COMPAT_MODE and
ECPP_NIBBLE_MODE. Centronics devices support only ECPP_CENTRONICS mode.
By default, ECPP_COMPAT_MODE devices have a strobe handshaking pulse width of 500ns. For
this mode, forward data transfers are conducted by DMA. By default, the strobe pulse width
for ECPP_CENTRONICS devices is two microseconds. Forward transfers for these devices are
managed through PIO. The default characteristics for both ECPP_COMPAT_MODE and
ECPP_CENTRONICS devices may be changed through tunable variables defined in ecpp.conf.
The ecpp driver is an exclusive-use device, meaning that if the device is already open,
subsequent opens fail with EBUSY.
Default Operation Each time the ecpp device is opened, the device is marked as EBUSY and the configuration
variables are set to their default values. The write_timeout period is set to 90 seconds.
The driver sets the mode variable according to the following algorithm: The driver initially
attempts to negotiate the link into ECPP_ECP_MODE during open(2). If it fails, the driver tries to
negotiate into ECPP_NIBBLE_MODE mode. If that fails, the driver operates in ECPP_CENTRONICS
mode. Upon successfully opening the device, IEEE 1284 compliant devices will be left idle in
either reverse idle phase of ECPP_ECP_MODE or in ECPP_NIBBLE_MODE. Subsequent calls to
write(2) invokes the driver to move the link into either ECPP_COMPAT_MODE or the forward
phase of ECPP_ECP_MODE. After the transfer completes, the link returns to idle state.
The application may attempt to negotiate the device into a specific mode or set the
write_timeout values through the ECPPIOC_SETPARMS ioctl(2) call. For mode negotiation to
be successful, both the host workstation and the peripheral must support the requested mode.
Tunables Characteristics of the ecpp driver may be tuned by the variables described in
/kernel/drv/ecpp.conf. These variables are read by the kernel during system startup. To
tune the variables, edit the ecpp.conf file and invoke update_drv(1M) to have the kernel read
the file again.
Some Centronics peripherals and certain IEEE 1284 compatible peripherals will not operate
with the parallel port operating in a fast handshaking mode. If printing problems occur, set
"fast-centronics" and "fast-1284-compatible" to "false.” See /kernel/drv/ecpp.conf for more
information.

234 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 ecpp(7D)

Read/Write Operation The ecpp driver is a full duplex STREAMS device driver. While an application is writing to an
IEEE 1284 compliant device, another thread may read from it.

Write Operation A write(2) operation returns the number of bytes successfully written to the stream head. If a
failure occurs while a Centronics device is transferring data, the content of the status bits will
be captured at the time of the error and can be retrieved by the application program using the
BPPIOC_GETERR ioctl(2) call. The captured status information is overwritten each time an
attempted transfer or a BPPIOC_TESTIO ioctl(2) occurs.

Read Operation If a failure or error condition occurs during a read(2), the number of bytes successfully read is
returned (short read). When attempting to read a port that has no data currently available,
read(2) returns 0 if O_NDELAY is set. If O_NONBLOCK is set, read(2) returns -1 and sets errno to
EAGAIN. If O_NDELAY and O_NONBLOCK are clear, read(2) blocks until data become available.

ioctls The ioctl(2) calls described below are supported. Note that when ecpp is transferring data,
the driver waits until the data has been sent to the device before processing the ioctl(2) call.

The ecpp driver supports prnio(7I) interfaces.

Note – The PRNIOC_RESET command toggles the nInit signal for 2 ms, followed by default
negotiation.

The following ioctl(2) calls are supported for backward compatibility and are not
recommended for new applications:
ECPPIOC_GETPARMS Get current transfer parameters. The argument is a pointer to a struct
ecpp_transfer_parms. See below for a description of the elements of
this structure. If no parameters have been configured since the device
was opened, the structure will be set to its default configuration. See
Default Operation above for more information.
ECPPIOC_SETPARMS Set transfer parameters. The argument is a pointer to a struct
ecpp_transfer_parms. If a parameter is out of range, EINVAL is
returned. If the peripheral or host device cannot support the
requested mode, EPROTONOSUPPORT is returned. See below for a
description of ecpp_transfer_parms and its valid parameters.

The Transfer Parameters Structure is defined in <sys/ecppio.h>.

struct ecpp_transfer_parms {
int write_timeout;
int mode;
};
The write_timeout field is set to the value of
ecpp-transfer-timeout specified in the ecpp.conf. The
write_timeout field specifies how long the driver will wait for the

Device and Network Interfaces 235

ecpp(7D)

peripheral to respond to a transfer request. The value must be greater

than 0 and less than ECPP_MAX_TIMEOUT. All other values are out of
range.

The mode field reflects the IEEE 1284 mode to which the parallel port
is currently configured. The mode may be set to one of the following
values only: ECPP_CENTRONICS, ECPP_COMPAT_MODE,
ECPP_NIBBLE_MODE, ECPP_ECP_MODE. All other values are invalid. If
the requested mode is not supported, ECPPIOC_SETPARMS will return
EPROTONOSUPPORT and the mode will be set to ECPP_CENTRONICS
mode. Afterwards, the application may change the mode back to the
original mode with ECPPIOC_SETPARMS.
ECPPIOC_GETDEVID This ioctl gets the IEEE 1284 device ID from the peripheral in
specified mode. Currently, the device ID can be retrieved only in
Nibble mode. A pointer to the structure defined in <sys/ecppsys.h>
must be passed as an argument.

The 1284 device ID structure:

struct ecpp_device_id {
int mode; /* mode to use for reading device id */
int len; /* length of buffer */
int rlen; /* actual length of device id string */
char *addr; /* buffer address */
};
The mode is the IEEE 1284 mode into which the port will be
negotiated to retrieve device ID information. If the peripheral or host
do not support the mode, EPROTONOSUPPORT is returned. Applications
should set mode to ECPP_NIBBLE_MODE. len is the length of the buffer
pointed to by addr. rlen is the actual length of the device ID string
returned from the peripheral. If the returned rlen is greater than len,
the application can call ECPPIOC_GETDEVID again with a buffer length
equal or greater than rlen. Note that the two length bytes of the IEEE
1284 device ID are not taken into account and are not returned in the
user buffer.

After ECPPIOC_GETDEVID successfully completes, the driver returns

the link to ECPP_COMPAT_MODE. The application is responsible for
determining the previous mode the link was operating in and
returning the link to that mode.
BPPIOC_TESTIO Tests the forward transfer readiness of a peripheral operating in
Centronics or Compatibility mode.

236 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 ecpp(7D)

TESTIO determines if the peripheral is ready to receive data by

checking the open flags and the Centronics status signals. If the
current mode of the device is ECPP_NIBBLE_MODE, the driver
negotiates the link into ECPP_COMPAT_MODE, check the status signals
and then return the link to ECPP_NIBBLE_MODE mode. If the current
mode is ECPP_CENTRONICS or ECPP_COMPAT_MODE, TESTIO examines
the Centronics status signals in the current mode. To receive data, the
device must have the nErr and Select signals asserted and must not
have the PE and Busy signals asserted. If ecpp is transferring data,
TESTIO waits until the previous data sent to the driver is delivered
before executing TESTIO. However if an error condition occurs while a
TESTIO is waiting, TESTIO returns immediately. If TESTIO determines
that the conditions are ok, 0 is returned. Otherwise, -1 is returned,
errno is set to EIO and the state of the status pins is captured. The
captured status can be retrieved using the BPPIOC_GETERR ioctl(2)
call. The timeout_occurred and bus_error fields will never be set by
this ioctl(2). BPPIOC_TESTIO and BPPIOC_GETERR are compatible to
the ioctls specified in bpp(7D).
BPPIOC_GETERR Get last error status. The argument is a pointer to a struct
bpp_error_status defined in <sys/bpp_io.h> header file. The error
status structure is:
struct bpp_error_status {
char timeout_occurred; /* 1=timeout */
char bus_error; /* not used */
uchar_t pin_status; /* status of pins which
/* could cause error */
};
The pin_status field indicates possible error conditions. The valid bits
for pin_status are: BPP_ERR_ERR, BPP_SLCT_ERR, BPP_PE_ERR,
BPP_BUSY_ERR. A set bit indicates that the associated pin is asserted.

This structure indicates the status of all the appropriate status bits at
the time of the most recent error condition during a write(2) call, or
the status of the bits at the most recent BPPIOC_TESTIO ioctl(2)call.

pin_status indicates possible error conditions under

ECPP_CENTRONICS or ECPP_COMPAT_MODE. Under these modes, the
state of the status pins will indicate the state of the device. For
instance, many Centronics printers lower the nErr signal when a
paper jam occurs. The behavior of the status pins depends on the
device. Additional status information may be retrieved through the
backchannel.

Device and Network Interfaces 237

ecpp(7D)

The timeout_occurred value is set when a timeout occurs during

write(2). bus_error is not used in this interface.

The following ioctls are used to directly read and write the parallel port status and control
signals. If the current mode of the device is ECPP_ECP_MODE or ECPP_NIBBLE_MODE, the driver
negotiates the link into ECPP_COMPAT_MODE, gets or sets the registers and then returns the link
to ECPP_NIBBLE_MODE. If the current mode is ECPP_CENTRONICS or ECPP_COMPAT_MODE, these
ioctls will get/set the register values in the current mode.
ECPPIOC_GETREGS Read register values. The argument is a pointer to a struct ecpp_regs.
See below for a description of this structure.
ECPPIOC_SETREGS Set ecpp register values. The argument is a pointer to a struct
ecpp_regs. See below for a description of this structure. If a parameter
is out of range, EINVAL is returned.

The Port Register Structure is defined in <sys/ecppio.h>.

struct ecpp_regs {
uchar dsr; /* status reg */
u_char dcr; /* control reg */
};
The status register is read-only. The ECPPIOC_SETREGS ioctl has no
affect on this register. Valid bit values for dsr are: ECPP_nERR,
ECPP_SLCT, ECPP_PE, ECPP_nACK, ECPP_nBUSY. All other bits are
reserved and always return 1.

The control register is read/write. Valid bit values for dcr are:
ECPP_STB, ECPP_AFX, ECPP_nINIT, ECPP_SLCTIN. All other bits are
reserved. Reading reserved bits always return 1. An attempt to write 0s
into these bits results in EINVAL.
Device Special /dev/lpN Solaris x86 only. (Backwards compatibility with former lp(7D)
Files devices.)
/dev/printers/N 1284 compliant parallel port device special files appears in both
namespaces.
Files kernel/drv/ecpp 32–bit ELF kernel module
kernel/drv/sparcv9/ecpp 64–bit SPARC ELF kernel module
kernel/drv/amd64/ecpp 64–bit x86 ELF kernel module
kernel/drv/ecpp.conf driver configuration file
kernel/drv/sparcv9/ecpp.conf driver configuration file for 64–bit SPARC
kernel/drv/amd64/ecpp.conf driver configuration file for 64–bit x86

238 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 ecpp(7D)

Errors EBADF The device is opened for write-only access and a read is attempted, or the device is
opened for read-only access and a write is attempted.
EBUSY The device has been opened and another open is attempted. An attempt has been
made to unload the driver while one of the units is open.
EINVAL A ECPPIOC_SETPARMS ioctl() is attempted with an out-of-range value in the
ecpp_transfer_parms structure. A ECPPIOC_SETREGS ioctl() is attempted with
an invalid value in the ecpp_regs structure. An ioctl() is attempted with an
invalid value in the command argument.An invalid command argument is
received during modload(1M) or modunload(1M).
EIO The driver encountered a bus error when attempting an access. A read or write did
not complete properly, due to a peripheral error or a transfer timeout.
ENXIO The driver has received an open request for a unit for which the attach failed. The
driver has received a write request for a unit which has an active peripheral error.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

ISA-based systems (x86)

Availability SUNWpd (Sparc)

SUNWpsdcr (x86)

Interface stability Evolving

See Also modload(1M), modunload(1M), update_drv(1M)ioctl(2), open(2), read(2), write(2),

attributes(5), bpp(7D), usbprn(7D), prnio(7I), streamio(7I)

IEEE Std 1284–1994

http://www.sun.com/io
Diagnostics Parallel port controller not supported Driver does not support parallel port controller on
the given host. Attach failed.

Device and Network Interfaces 239

ehci(7D)

Name ehci – Enhanced host controller driver

Synopsis usb@unit-address

Description The ehci driver is a USBA (Solaris USB Architecture) compliant nexus driver that supports
the Enhanced Host Controller Interface Specification 2.0, an industry standard developed by
Intel.

A USB 2.0 host controller includes one high-speed host controller and zero or more USB 1.1
host controllers. The high-speed host controller implements an EHCI (Enhanced Host
Controller Interface) that is used for all high-speed communications to high-speed-mode
devices.

All USB 2.0 devices connected to the root ports of the USB 2.0 host controller and all devices
connected to a high- speed-mode hub should be routed to the EHCI host controller.

All full- and low-speed devices connected to the root ports of the USB 2.0 host controller
should be routed to the companion USB 1.1 host controllers. (OHCI or UHCI host
controller).

The ehci supports bulk, interrupt, control and iso chronous transfers (on USB1.x devices
behind a USB2.0 hub).
Files /kernel/drv/ehci 32–bit ELF 86 kernel module
/kernel/drv/sparcv9/ehci 64–bit SPARC ELF kernel module
/kernel/drv/amd64/ehci 64–bit x86 ELF kernel module
/kernel/drv/ehci.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also add_drv(1M), prtconf(1M), rem_drv(1M), update_drv(1M), attributes(5), hubd(7D),

uhci(7D), ohci(7D), usba(7D)

Writing Device Drivers

Universal Serial Bus Specification 2.0

Enhanced Host Controller Interface Specification 1.0

System Administration Guide: Basic Administration

240 man pages section 7: Device and Network Interfaces • Last Revised 13 April 2006
 ehci(7D)

http://www.usb.org

http://www.sun.com/io

http://www.sun.com/bigadmin/hcl

http://www.intel.com/technology/usb/ehcispec.htm

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
WARNING: <device path> (ehci<instance number>): Message...
Unrecoverable USB hardware error.
There was an unrecoverable USB hardware error reported by the ehci controller. Reboot
the system. If this problem persists, contact your system vendor.
No SOF interrupts.
No SOF interrupts have been received. This USB EHCI controller is unusable.
Error recovery failure: Please hotplug the 2.0 hub at <device path>.
The driver failed to clear 2.0 hub's TT buffer. Remove and reinsert the external USB2.0 hub.
Revision<xx> is not supported.
High speed USB devices prior to revision 0.95 are not supported.

The following messages may be entered into the system log. They are formatted in the
following manner:
<device path> (ehci<instance number>): Message...
Unable to take control from BIOS. Failure is ignored.
The driver was unable to take control of the EHCI hardware from the system's BIOS. This
failure is ignored. To abort the attach on this take-over failure, comment out a property in
ehci.conf. (x86 only).
Unable to take control from BIOS.
The driver is unable to take control of the EHCI hardware from the system's BIOS and
aborts the attach. High speed (USB 2.0) support is disabled. In this case, all USB devices run
at full/low speed. Contact your system vendor or your system administror for possible
changes in BIOS settings. You can disable a property in ehci.conf to ignore this failure.
(x86 only.)
Low speed device is not supported.
Full speed device is not supported.
The driver detected a low or full speed device on its root hub port. Per USB 2.0
specification, the device should be routed to a companion host controller (OHCI or
UHCI). However, no attached companion host controller appears to be available.
Therefore, low and full speed devices are not supported.

Device and Network Interfaces 241

ehci(7D)

Low speed endpoint's poll interval of <n> ms is below threshold. Rounding up to 8 ms.
Low speed endpoints are limited to polling intervals between 8 ms and 255 ms. If a device
reports a polling interval that is less than 8 ms, the driver uses 8 ms instead.
Low speed endpoint's poll interval is greater than 255 ms.
The low speed device's polling interval is out of range. The host controller does not allocate
bandwidth for this device. This device is not usable.
Full speed endpoint's poll interval must be between 1 and 255 ms.
The full speed device's polling interval is out of range. The host controller does not allocate
bandwidth for this device. This device is not usable.
High speed endpoint's poll interval must be between 1 and 16 units.
The high speed device's polling interval is out of range. The host controller will not allocate
bandwidth for this device. This device will not be usable. Refer to the USB specification,
revision 2.0 for the unit definition.
ehci_modify_qh_status_bit: Failed to halt qh=<address>.
Error recovery failed. Please disconnect and reinsert all devices or reboot.

Note – Due to recently discovered incompatibilities with this USB controller, USB2.x transfer
support has been disabled. However, this device continues to function as a USB1.x controller.
Information on enabling USB2.x support is provided in this man page. Please refer to
www.sun.com/io for Solaris Ready products and to www.sun.com/bigadmin/hcl for additional
compatible USB products.

VIA chips may not be compatible with this driver. To bind ehci specifically to the chip and
eliminate the warnings, and to enable USB2.x suppport, a new, more specific driver alias (refer
to add_drv(1M) and update_drv(1M)) must be specified for ehci. By default, the ehci alias is
'pciclass,0c0320.' The compatible names in the prtconf(1M) output provides additional
aliases. For example:
# prtconf -vp | grep pciclass,0c0320
compatible: ’pci1106,3104.1106.3104.2063’ +
’pci1106,3104.1106.3104’ + ’pci1106,3104’ +
pci1106,3104.2063’ + ’pci1106,3104’ + ’pciclass,0c0320’ +
’pciclass,0c03’
....

A more specific alias is ’pci1106,3104.’ Perform the follow-

ing step to add this alias, then reboot the system:

# update_drv -a -i ’"pci1106,3104"’ ehci

# reboot

After you apply the above workaround, the following message is displayed in your system log:

242 man pages section 7: Device and Network Interfaces • Last Revised 13 April 2006
 ehci(7D)

Applying VIA workarounds.

Device and Network Interfaces 243

elxl(7D)

Name elxl – 3Com Ethernet device driver

Synopsis /kernel/drv/elxl
Description The elxl driver currently supports the following network cards: EtherLink XL (3C900-TPO,
3C900-COMBO, 3C900B-TPO, 3C900B-COMBO, and 3C900B-TPC), EtherLink XL 10/100
(3C905-TX Fast, 3C905-T4 Fast, 3C905B-TX Fast, 3C905B-T4 Fast, and 3C905C-TX-M Fast),
and EtherLink Server 10/100 (3C980-TX Fast and 3C980C-TXM).
The elxl Ethernet driver is a multi-threaded, loadable, clonable, STREAMS hardware driver
supporting the connectionless Data Link Provider Interface, dlpi(7P). Multiple EtherLink XL
controllers installed within the system are supported by the driver. The elxl driver provides
basic support for the EtherLink hardware. Functions include chip initialization, frame
transmit and receive, multicast and promiscuous mode support, and error recovery and
reporting.
The cloning, character-special device /dev/elxl is used to access all EtherLink devices
installed within the system.
The elxl driver is dependent on /kernel/misc/gld, a loadable kernel module that provides
the elxl driver with the DLPI and STREAMS functionality required of a LAN driver. See
gld(7D) for more details on the primatives supported by the driver.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ Maximum SDU is 1500 (ETHERMTU).
■ Minimum SDU is 0. The driver will pad to the mandatory 60-octet minimum packet size.
■ The dlsap address length is 8.
■ MAC type is DL_ETHER.
■ The sap length value is –2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.
■ The broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Preconfiguration
Supported Settings ■ Media Type: Auto Select
Known Problems and ■ 3C905B cards in a Compaq ProLiant 6500 can fail to generate interrupts. There is no
Limitations known workaround for this problem. However, because some slots appear to be more
prone to the problem than others, try correcting the problem by moving the card to
another PCI slot. If that fails, try rebooting the machine a number of times to free the card
from the wedged state.
■ Early versions of the 3Com 3C905C-TX-M adapter firmware do not support PXE network
boot on Solaris systems. If you are using a version earlier than 4.11, upgrade the firmware.
The PXE version is indicated by the Managed Boot Agent version number. This number is
not normally displayed during boot, but is shown on the PXE configuration screen.

244 man pages section 7: Device and Network Interfaces • Last Revised 16 October 2003
 elxl(7D)

Configuration The /kernel/drv/elxl.conf file supports the following tunable properties that you can set in
the elxl.conf file:
tx_start_thresh Minimum number of bytes in transmit FIFO (if lower then the
packet size), before NIC starts the frame transfer. The higher
the value, the less chance for a transmission underrun error
that triggers a transmission retry. Default value is 250,
maximum value is 1514.
recv-descriptors The number of frames the device may receive from the network
without system attention before the device begins dropping the
frames. Valid range is 1 to 256. Default is 24.
xmit-descriptors The number of outgoing frames that the system can queue on
the driver queue before blocking a send request. Valid range is
1 to 512. Default is 128.
min-recv-data-buffers Minimal number of allocated receive data buffers. Must be
equal or greater than effective number of recv-descriptors.
Default (max value) is 128.
max-recv-data-buffers Maximum number of allocated receive data buffers. Must be
equal or greater then min-recv-data-buffers number.
inter-frame-space Amount of time in addition to the standard inter-frame time
used by IEEE 802.3 deference rule. Expressed in 32 multiples of
bit time. Used only in half-duplex context. This small IFS is to
make the defer-after-transmit time slightly longer than the
defer-after-receive time. This greatly reduces the high collision
rate seen with heavy TCP traffic (almost exactly one collision
per ACK), giving slightly improved overall throughput in
half-duplex mode. Default IFS is 32. This property is ignored in
full-duplex mode.
full-duplex When set, it forces full-duplex mode for NIC capable of
working in full-duplex mode. If this property is set, the
auto-negotiation feature (if supported) of the network
controller is disabled. For full duplex operation use
full-duplex=1. For half duplex use full-duplex=0.
Half-duplex operation provides better results on older 10-Mbit
networks.
speed Link speed in Mbps. Valid values are 10 and 100. If this
property is set, the auto-negotiation feature of the network
controller (if supported) is disabled.

Device and Network Interfaces 245

elxl(7D)

Files /dev/elxl Special character device

/kernel/drv/elxl.conf Configuration file for elxl driver

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), gld(7D), streamio(7I), dlpi(7P).

246 man pages section 7: Device and Network Interfaces • Last Revised 16 October 2003
 emlxs(7D)

Name emlxs – Emulex-Sun LightPulse Fibre Channel host bus adapter driver
Synopsis SUNW,emlxs

Description The emlxs host bus adapter driver is a Sun Fibre Channel transport layer-compliant nexus
driver for the Emulex Light-Pulse family of Fibre Channel adapters. These adapters support
Fibre Channel SCSI and IP Protocols, FC-AL public loop profile, point-to-point fabric
connection and Fibre Channel service classes two and three.

The emlxs driver interfaces with the Sun Fibre Channel transport layer to support the
standard functions provided by the SCSA interface. It supports auto request sense and tagged
queueing by default. The driver requires that all devices have unique hard addresses in private
loop configurations. Devices with conflicting hard addresses are not accessible.
Files /kernel/drv/emlxs 32–bit ELF kernel module.
/kernel/drv/amd/emlxs 64-bit ELF kernel module (x86).
/kernel/drv/sparcv9/emlxs 64-bit ELF kernel module (SPARC).
/kernel/drv/emlxs.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWemlxs

See Also prtconf(1M), driver.conf(4), fcp(7D), fp(7d)

Writing Device Drivers

ANSI X3.230:1994, Fibre Channel Physical Signaling (FC-PH)

Project 1134-D, Fibre Channel Generic Services (FC-GS-2)

ANSI X3.269-1996, Fibre Channel Arbitrated Loop (FC-AL)

ANSI X3.270-1996, Fibre Channel Protocol for SCSI (FCP-SCSI)

ANSI X3.270-1996, SCSI-3 Architecture Model (SAM)

Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)

Fabric Loop Attachment (FC-FLA)

Device and Network Interfaces 247

eri(7D)

Name eri – eri Fast-Ethernet device driver

Synopsis /dev/eri

Description The eri Fast Ethernet driver is a multi-threaded, loadable, clonable, STREAMS—based
hardware driver supporting the connectionless Data Link Provider Interface dlpi(7P) over an
eri Fast-Ethernet controller. Multiple eri devices installed within the system are supported
by the driver.

The eri driver provides basic support for the eri hardware and handles the eri device.
Functions include chip initialization, frame transit and receive, multicast and promiscuous
support, and error recovery and reporting.

The eri device provides 100Base-TX networking interfaces using the SUN RIO ASIC and an
internal transceiver. The RIO ASIC provides the PCI interface and MAC functions. The
physical layer functions are provided by the internal transceiver which connects to a RJ-45
connector.

The 100Base-TX standard specifies an auto-negotiation protocol to automatically select the

mode and speed of operation. The internal transceiver is capable of performing
auto-negotiation using the remote-end of the link (link partner) and receives the capabilities
of the remote end. It selects the highest common denominator mode of operation based on the
priorities. It also supports a forced-mode of operation under which the driver selects the mode
of operation.

Application The cloning character-special device /dev/eri is used to access all eri controllers installed
Programming within the system.
Interface
eri and DLPI The eri driver is a “style 2” Data Link Service provider. All M_PROTO and M_PCPROTO type
messages are interpreted as DLPI primitives. Valid DLPI primitives are defined in
<sys/dlpi.h>. Refer to dlpi(7P) for more information.

An explicit DL_ATTACH_REQ message by the user is required to associate the opened stream
with a particular device (ppa). The ppa ID is interpreted as an unsigned integer data type
and indicates the corresponding device instance (unit) number. An error (DL_ERROR_ACK) is
returned by the driver if the ppa field value does not correspond to a valid device instance
number for this system. The device is initialized on first attach and de-initialized (stopped) at
last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ The maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h> ).
■ The minimum SDU is 0.
■ The dlsap address length is 8.
■ The MAC type is DL_ETHER.

248 man pages section 7: Device and Network Interfaces • Last Revised 1 Mar 2000
 eri(7D)

■ The sap length values is –2, meaning the physical address component is followed
immediately by a 2 byte sap component within the DLSAP address.
■ The service mode is DL_CLDLS.
■ Optional quality of service (QOS) is not currently supported so QOS fields are 0.
■ The provider style is DL_STYLE.
■ The version is DL_VERSION_2.
■ The broadcast address value is Ethernet/IEEE broadcast address (0xFFFFFF).

Once in the DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular
SAP (Service Access Pointer) with the stream. The eri driver interprets the sap field within
the DL_BIND_REQ as an Ethernet “type,” therefore valid values for the sap field are in the
[0-0xFFFF] range. Only one Ethernet type can be bound to the stream at any time.

If the user selects a sap with a value of 0, the receiver will be in IEEE 802.3 mode. All frames
received from the media having a Ethernet type field in the range [0-1500] are assumed to be
802.3 frames and are routed up all open Streams which are bound to sap value 0. If more than
one Stream is in 802.3 mode, the frame will be duplicated and routed up multiple Streams as
DL_UNITDATA_IND messages.

In transmission, the driver checks the sap field of the DL_BIND_REQ to determine if the value is
0 or if the Ethernet type field is in the range [0-1500]. If either is true, the driver computes the
length of the message, not including initial M_PROTO mblk (message block), of all subsequent
DL_UNITDATA_REQ messages, and transmits 802.3 frames that have this value in the MAC
frame header length field.

The eri driver's DLSAP address format consists of the 6 byte physical (Ethernet) address
component followed immediately by the 2 byte sap (type) component, producing an 8 byte
DLSAP address. Applications should not hardcode to this particular implementation-specific
DLSAP address format but use information returned in the DL_INFO_ACK primitive to
compose and decompose DLSAP addresses. The sap length, full DLSAP length, and
sap/physical ordering are included within the DL_INFO_ACK. The physical address length can
be computed by subtracting the sap length from the full DLSAP address length or by issuing
the DL_PHYS_ADDR_REQ to obtain the current physical address associated with the stream.

Once in the DL_BOUND state, the user may transmit frames on the Ethernet by sending
DL_UNITDATA_REQ messages to the eri driver. The eri driver will route received Ethernet
frames up all open and bound streams having a sap which matches the Ethernet type as
DL_UNITDATA_IND messages. Received Ethernet frames are duplicated and routed up multiple
open streams if necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.

Device and Network Interfaces 249

eri(7D)

eri Primitives In addition to the mandatory connectionless DLPI message set, the driver also supports the
following primitives:

The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable/disable reception of

individual multicast group addresses. A set of multicast addresses may be iteratively created
and modified on a per-stream basis using these primitives. These primitives are accepted by
the driver in any state following DL_ATTACHED.

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives with the DL_PROMISC_PHYS flag

set in the dl_level field enables/disables reception of all promiscuous mode frames on the
media, including frames generated by the local host. When used with the DL_PROMISC_SAP flag
set, this enables/disables reception of all sap (Ethernet type) values. When used with the
DL_PROMISC_MULTI flag set, this enables/disables reception of all multicast group addresses.
The effect of each is always on a per-stream basis and independent of the other sap and
physical level configurations on this stream or other streams.

The DL_PHYS_ADDR_REQ primitive returns the 6 octet Ethernet address currently associated
(attached) to the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the 6 octet Ethernet address currently

associated (attached) to this stream. The credentials of the process which originally opened
this stream must be superuser, or EPERM is returned in the DL_ERROR_ACK. This primitive is
destructive because it affects all current and future streams attached to this device. An M_ERROR
is sent up all other streams attached to this device when this primitive is successful on this
stream. Once changed, all streams subsequently opened and attached to this device will obtain
this new physical address. Once changed, the physical address will remain until this primitive
is used to change the physical address again or the system is rebooted, whichever comes first.

eri DRIVER By default, the eri driver performs auto-negotiation to select the mode and speed of the link,
which can be in one of the following modes, as described in the 100Base-TX standard:
■ 100 Mbps, full-duplex
■ 100 Mbps, half-duplex
■ 10 Mbps, full-duplex
■ 10 Mbps, half-duplex

The auto-negotiation protocol automatically selects:

■ Operation mode (half-duplex or full-duplex)
■ Speed (100 Mbps or 10 Mbps)

The auto–negotiation protocol does the following:

■ Gets all modes of operation supported by the link partner
■ Advertises its capabilities to the Link Partner
■ Selects the highest common denominator mode of operation based on the priorities

250 man pages section 7: Device and Network Interfaces • Last Revised 1 Mar 2000
 eri(7D)

The internal transceiver is capable of all of the operating speeds and modes listed above. By
default, auto-negotiation is used to select the speed and the mode of the link and the common
mode of operation with the link partner.

For users who want to select the speed and mode of the link, the eri device supports
programmable IPG (Inter-Packet Gap) parameters ipg1 and ipg2. Sometimes, the user may
want to alter these values depending on whether the driver supports 10 Mbps or 100 Mpbs and
accordingly, IPG will be set to 9.6 or 0.96 microseconds.

eri Parameter List The eri driver provides for setting and getting various parameters for the eri device. The
parameter list includes current transceiver status, current link status, inter-packet gap, local
transceiver capabilities and link partner capabilities.

The local transceiver has two set of capabilities: one set reflects hardware capabilities, which
are read-only (RO) parameters. The second set reflects the values chosen by the user and is
used in speed selection and possess read/write (RW) capability. At boot time, these two sets of
capabilities will be the same. Because the current default value of these parameters can only be
read and not modified, the link partner capabilities are also read only.
Files /dev/eri eri special character device.
/kernel/drv/eri.conf System wide default device driver properties
/kernel/drv/sparcv9/eri 64 bit device driver

See Also ndd(1M), netstat(1M), driver.conf(4), hme(7D), qfe(7d), dlpi(7P)

Device and Network Interfaces 251

esp(7D)

Name esp – ESP SCSI Host Bus Adapter Driver

Synopsis esp@sbus-slot,80000

Description The esp Host Bus Adapter driver is a SCSA compliant nexus driver that supports the Emulex
family of esp SCSI chips (esp100, esp100A, esp236, fas101, fas236).

The esp driver supports the standard functions provided by the SCSA interface. The driver
supports tagged and untagged queuing, fast SCSI (on FAS esp's only), almost unlimited
transfer size (using a moving DVMA window approach), and auto request sense; but it does
not support linked commands.

Configuration The esp driver can be configured by defining properties in esp.conf which override the global
SCSI settings. Supported properties are: scsi-options, target<n>-scsi-options,
scsi-reset-delay, scsi-watchdog-tick, scsi-tag-age-limit, scsi-initiator-id.

target<n>-scsi-options overrides the scsi-options property value for target<n>. <n>

can vary from 0 to 7.

Refer to scsi_hba_attach(9F) for details.

Examples EXAMPLE 1 A sample of esp configuration file.

Create a file /kernel/drv/esp.conf and add this line:

scsi-options=0x78;

This will disable tagged queuing, fast SCSI, and Wide mode for all esp instances. To disable an
option for one specific esp (refer to driver.conf(4)):
name="esp"
parent="/iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000"
reg=0xf,0x800000,0x40
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

Note that the default initiator ID in OBP is 7 and that the change to ID 6 will occur at attach
time. It may be preferable to change the initiator ID in OBP.

The above would set scsi-options for target 1 to 0x58 and for all other targets on this SCSI
bus to 0x178. The physical pathname of the parent can be determined using the /devices tree
or following the link of the logical device name:
example# ls -l /dev/rdsk/c0t3d0s0
lrwxrwxrwx 1 root root 88 Aug 22 13:29 /dev/rdsk/c0t3d0s0 ->
../../devices/iommu@f,e0000000/sbus@f,e0001000/espdma@f,400000/
esp@f,800000/sd@3,0:a,raw

The register property values can be determined from prtconf(1M) output (-v option):

252 man pages section 7: Device and Network Interfaces • Last Revised 7 Feb 1997
 esp(7D)

EXAMPLE 1 A sample of esp configuration file. (Continued)

esp, instance #0
....
Register Specifications:
Bus Type=0xf, Address=0x800000, Size=40

To set scsi-options more specifically per target:

target1-scsi-options=0x78;
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;
scsi-options=0x3f8;

The above would set scsi-options for target 1 to 0x78 and for all other targets on this SCSI
bus to 0x378 except for one specific disk type which will have scsi-options set to 0x58.

scsi-options specified per target ID has the highest precedence, followed by scsi-options
per device type. To get the inquiry string run probe-scsi or probe-scsi-all command at the ok
prompt before booting the system.

Global, for example. for all esp instances, scsi-options per bus has the lowest precedence.

The system needs to be rebooted before the specified scsi-options take effect.
Files /kernel/drv/esp ELF Kernel Module
/kernel/drv/esp.conf Configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SBus-based systems with esp-based

SCSI port and SSHA, SBE/S, FSBE/S,

and DSBE/S SBus SCSI Host Adapter options

See Also prtconf(1M), driver.conf(4), attributes(5), fas(7D), scsi_abort(9F),

scsi_hba_attach(9F), scsi_ifgetcap(9F), scsi_reset(9F), scsi_sync_pkt(9F),
scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S), scsi_inquiry(9S),
scsi_pkt(9S)
Writing Device Drivers
OpenBoot Command Reference
ANSI Small Computer System Interface-2 (SCSI-2)

Device and Network Interfaces 253

esp(7D)

ESP Technical Manuals, QLogic Corp.

Diagnostics The messages described below are some that may appear on the system console, as well as
being logged.

The first four messages may be displayed while the esp driver is trying to attach; these
messages mean that the esp driver was unable to attach. All of these messages are preceded by
"esp%d", where "%d" is the instance number of the esp controller.
Device in slave-only slot
The SBus device has been placed in a slave-only slot and will not be accessible; move to
non-slave-only SBus slot.
Device is using a hilevel intr
The device was configured with an interrupt level that cannot be used with this esp driver.
Check the SBus device.
Unable to map registers
Driver was unable to map device registers; check for bad hardware. Driver did not attach to
device; SCSI devices will be inaccessible.
Cannot find dma controller
Driver was unable to locate a dma controller. This is an auto-configuration error.
Disabled TQ since disconnects are disabled
Tagged queuing was disabled because disconnects were disabled in scsi-options.
Bad clock frequency- setting 20mhz, asynchronous mode
Check for bad hardware.
Sync pkt failed
Syncing a SCSI packet failed. Refer to scsi_sync_pkt(9F).
Slot %x: All tags in use!!!
The driver could not allocate another tag number. The target devices do not properly
support tagged queuing.
Target %d.%d cannot alloc tag queue\n
The driver could not allocate space for tag queue.
Gross error in esp status (%x)
The driver experienced severe SCSI bus problems. Check cables and terminator.
Spurious interrupt
The driver received an interrupt while the hardware was not interrupting.
Lost state in phasemanage
The driver is confused about the state of the SCSI bus.
Unrecoverable DMA error during selection
The DMA controller experienced host SBus problems. Check for bad hardware.

254 man pages section 7: Device and Network Interfaces • Last Revised 7 Feb 1997
 esp(7D)

Bad sequence step (0x%x) in selection

The esp hardware reported a bad sequence step. Check for bad hardware.
Undetermined selection failure
The selection of a target failed unexpectedly. Check for bad hardware.
>2 reselection IDs on the bus
Two targets selected simultaneously, which is illegal. Check for bad hardware.
Reconnect: unexpected bus free
A reconnect by a target failed. Check for bad hardware.
Timeout on receiving tag msg
Suspect target f/w failure in tagged queue handling.
Parity error in tag msg
A parity error was detected in a tag message. Suspect SCSI bus problems.
Botched tag
The target supplied bad tag messages. Suspect target f/w failure in tagged queue handling.
Parity error in reconnect msg’s
The reconnect failed because of parity errors.
Target <n> didn’t disconnect after sending <message>
The target unexpectedly did not disconnect after sending <message>.
No support for multiple segs
The esp driver can only transfer contiguous data.
No dma window?
Moving the DVMA window failed unexpectedly.
No dma window on <type> operation
Moving the DVMA window failed unexpectedly.
Cannot set new dma window
Moving the DVMA window failed unexpectedly.
Unable to set new window at <address> for <type> operation
Moving the DVMA window failed unexpectedly.
Illegal dma boundary? %x
An attempt was made to cross a boundary that the driver could not handle.
Unwanted data out/in for Target <n>
The target went into an unexpected phase.
Spurious <name> phase from target <n>
The target went into an unexpected phase.
SCSI bus DATA IN phase parity error
The driver detected parity errors on the SCSI bus.

Device and Network Interfaces 255

esp(7D)

SCSI bus MESSAGE IN phase parity error

The driver detected parity errors on the SCSI bus.
SCSI bus STATUS phase parity error
The driver detected parity errors on the SCSI bus.
Premature end of extended message
An extended SCSI bus message did not complete. Suspect a target f/w problem.
Premature end of input message
A multibyte input message was truncated. Suspect a target f/w problem.
Input message botch
The driver is confused about messages coming from the target.
Extended message <n> is too long
The extended message sent by the target is longer than expected.
<name> message <n> from Target <m> garbled
Target <m> sent message <name> of value <n> which the driver did not understand.
Target <n> rejects our message <name>
Target <n> rejected a message sent by the driver.
Rejecting message <name> from Target <n>
The driver rejected a message received from target <n>
Cmd dma error
The driver was unable to send out command bytes.
Target <n> refused message resend
The target did not accept a message resend.
Two-byte message <name> <value> rejected
The driver does not accept this two-byte message.
Unexpected selection attempt
An attempt was made to select this host adapter by another initiator.
Polled cmd failed (target busy)
A polled command failed because the target did not complete outstanding commands
within a reasonable time.
Polled cmd failed
A polled command failed because of timeouts or bus errors.
Disconnected command timeout for Target <id>.<lun>
A timeout occurred while target/lun was disconnected. This is usually a target f/w problem.
For tagged queuing targets, <n> commands were outstanding when the timeout was
detected.

256 man pages section 7: Device and Network Interfaces • Last Revised 7 Feb 1997
 esp(7D)

Disconnected tagged cmds (<n>) timeout for Target <id>.<lun>

A timeout occurred while target/lun was disconnected. This is usually a target f/w problem.
For tagged queuing targets, <n> commands were outstanding when the timeout was
detected.
Connected command timeout for Target <id>.<lun>
This is usually a SCSI bus problem. Check cables and termination.
Target <id>.<lun> reverting to async. mode
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Target <id>.<lun> reducing sync. transfer rate
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Reverting to slow SCSI cable mode
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Reset SCSI bus failed
An attempt to reset the SCSI bus failed.
External SCSI bus reset
Another initiator reset the SCSI bus.

Warnings The esp hardware does not support Wide SCSI mode. Only FAS-type esp's support fast SCSI
(10 MB/sec).

Notes The esp driver exports properties indicating per target the negotiated transfer speed
(target<n>-sync-speed) and whether tagged queuing has been enabled (target<n>-TQ).
The sync-speed property value is the data transfer rate in KB/sec. The target-TQ property
has no value. The existence of the property indicates that tagged queuing has been enabled.
Refer to prtconf(1M) (verbose option) for viewing the esp properties.
dma, instance #3
Register Specifications:
Bus Type=0x2, Address=0x81000, Size=10
esp, instance #3
Driver software properties:
name <target3-TQ> length <0> − <no
value>.
name <target3-sync-speed> length <4>
value <0x00002710>.
name <scsi-options> length <4>
value <0x000003f8>.
name <scsi-watchdog-tick> length <4>
value <0x0000000a>.
name <scsi-tag-age-limit> length <4>
value <0x00000008>.

Device and Network Interfaces 257

esp(7D)

name <scsi-reset-delay> length <4>

value <0x00000bb8>.

258 man pages section 7: Device and Network Interfaces • Last Revised 7 Feb 1997
 fas(7D)

Name fas – FAS SCSI Host Bus Adapter Driver

Synopsis fas@sbus-slot,0x8800000

Description The fas Host Bus Adapter driver is a SCSA compliant nexus driver that supports the Qlogic
FAS366 SCSI chip.

The fas driver supports the standard functions provided by the SCSA interface. The driver
supports tagged and untagged queuing, wide and fast SCSI, almost unlimited transfer size
(using a moving DVMA window approach), and auto request sense; but it does not support
linked commands.

Driver Configuration The fas driver can be configured by defining properties in fas.conf which override the global
SCSI settings. Supported properties are: scsi-options, target<n>-scsi-options,
scsi-reset-delay, scsi-watchdog-tick, scsi-tag-age-limit, scsi-initiator-id.

target<n>-scsi-options overrides the scsi-options property value for target<n>. <n>

can vary from decimal 0 to 15. The supported scsi-options are: SCSI_OPTIONS_DR,
SCSI_OPTIONS_SYNC, SCSI_OPTIONS_TAG, SCSI_OPTIONS_FAST, and SCSI_OPTIONS_WIDE.

After periodic interval scsi-watchdog-tick, the fas driver searches all current and
disconnected commands for timeouts.

scsi-tag-age-limit is the number of times that the fas driver attempts to allocate a
particular tag ID that is currently in use after going through all tag IDs in a circular fashion.
After finding the same tag ID in use scsi-tag-age-limit times, no more commands will be
submitted to this target until all outstanding commands complete or timeout.

Refer to scsi_hba_attach(9F) for details.

Examples EXAMPLE 1 A sample of fas configuration file

Create a file called /kernel/drv/fas.conf and add this line:

scsi-options=0x78;

This disables tagged queuing, Fast SCSI, and Wide mode for all fas instances. The following
example disables an option for one specific fas (refer to driver.conf(4) for more details):
name="fas" parent="/iommu@f,e0000000/sbus@f,e0001000"
reg=3,0x8800000,0x10,3,0x8810000,0x40
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

Note that the default initiator ID in OBP is 7 and that the change to ID 6 will occur at attach
time. It may be preferable to change the initiator ID in OBP.

The example above sets scsi-options for target 1 to 0x58 and all other targets on this SCSI
bus to 0x178.

Device and Network Interfaces 259

fas(7D)

EXAMPLE 1 A sample of fas configuration file (Continued)

The physical pathname of the parent can be determined using the /devices tree or following
the link of the logical device name:
# ls -l /dev/rdsk/c1t3d0s0
lrwxrwxrwx 1 root other 78 Aug 28 16:05 /dev/rdsk/c1t3d0s0 ->

. . /. . /devices/iommu@f,e0000000\
sbus@f,e0001000/SUNW,fas@3,8800000/sd@3,0:a,raw

Determine the register property values using the output from prtconf(1M) (with the -v
option):
SUNW,fas, instance #0
. . . .
Register Specifications:
Bus Type=0x3, Address=0x8800000, Size=10
Bus Type=0x3, Address=0x8810000, Size=40

scsi-options can also be specified per device type using the device inquiry string. All the
devices with the same inquiry string will have the same scsi-options set. This can be used to
disable some scsi-options on all the devices of the same type.
device-type-scsi-options-list=
"TOSHIBA XM5701TASUN12XCD", "cd-scsi-options";
cd-scsi-options = 0x0;

The above entry in /kernel/drv/fas.conf sets the scsi-options for all devices with inquiry
string TOSHIBA XM5701TASUN12XCD to cd-scsi-options. To get the inquiry string, run the
probe-scsi or probe-scsi-all command at the ok prompt before booting the system.

To set scsi-options more specifically per target:

target1-scsi-options=0x78;
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;
scsi-options=0x3f8;

The above sets scsi-options for target 1 to 0x78 and for all other targets on this SCSI bus to
0x3f8 except for one specific disk type which will have scsi-options set to 0x58.

scsi-options specified per target ID have the highest precedence, followed by scsi-options
per device type. Global fas scsi-options (effecting all instances) per bus have the lowest
precedence.

The system needs to be rebooted before the specified scsi-options take effect.

260 man pages section 7: Device and Network Interfaces • Last Revised 20 Jun 1997
 fas(7D)

Driver Capabilities The target driver needs to set capabilities in the fas driver in order to enable some driver
features. The target driver can query and modify these capabilities: synchronous,
tagged-qing, wide-xfer, auto-rqsense, qfull-retries, qfull-retry-interval. All other
capabilities can only be queried.
By default, tagged-qing, auto-rqsense, and wide-xfer capabilities are disabled, while
disconnect, synchronous, and untagged-qing are enabled. These capabilities can only have
binary values (0 or 1). The default value for qfull-retries is 10 and the default value for
qfull-retry-interval is 100. The qfull-retries capability is a uchar_t (0 to 255) while
qfull-retry-interval is a ushort_t (0 to 65535).
The target driver needs to enable tagged-qing and wide-xfer explicitly. The untagged-qing
capability is always enabled and its value cannot be modified, because fas can queue
commands even when tagged-qing is disabled.
Whenever there is a conflict between the value of scsi-options and a capability, the value set
in scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call.
Refer to scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.
Files /kernel/drv/fas ELF Kernel Module
/kernel/drv/fas.conf Optional configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Limited to Sparc SBus-based systems with FAS366-based SCSI

port and SunSWIFT SBus SCSI Host Adapter/Fast Ethernet
option.

See Also prtconf(1M), driver.conf(4), attributes(5), scsi_abort(9F), scsi_hba_attach(9F),

scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F), scsi_sync_pkt(9F),
scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S), scsi_inquiry(9S),
scsi_pkt(9S)
Writing Device Drivers
ANSI Small Computer System Interface-2 (SCSI-2)
QLogic Corporation, FAS366 Technical Manuals.
Diagnostics The messages described below are some that may appear on the system console, as well as
being logged.
The first five messages may be displayed while the fas driver is trying to attach; these messages
mean that the fas driver was unable to attach. All of these messages are preceded by "fas%d",
where "%d" is the instance number of the fas controller.

Device and Network Interfaces 261

fas(7D)

Device in slave-only slot

The SBus device has been placed in a slave-only slot and will not be accessible; move to
non-slave-only SBus slot.
Device is using a hilevel intr
The device was configured with an interrupt level that cannot be used with this fas driver.
Check the SBus device.
Cannot alloc dma handle
Driver was unable to allocate memory for a DMA controller.
Cannot alloc cmd area
Driver was unable to allocate memory for a command address.
Cannot create kmem_cache
Driver was unable to allocate memory for internal data structures.
Unable to map FAS366 registers
Driver was unable to map device registers; check for bad hardware. Driver did not attach to
device; SCSI devices will be inaccessible.
Cannot add intr
Driver could not add its interrupt service routine to the kernel.
Cannot map dma
Driver was unable to locate a DMA controller. This is an auto-configuration error.
Cannot bind cmdarea
Driver was unable to bind the DMA handle to an address.
Cannot create devctl minor node
Driver is unable to create a minor node for the controller.
Cannot attach
The driver was unable to attach; usually follows another warning that indicates why attach
failed.
Disabled TQ since disconnects are disabled
Tagged queuing was disabled because disconnects were disabled in scsi-options.
Bad clock frequency
Check for bad hardware.
Sync of pkt (<address>) failed
Syncing a SCSI packet failed. Refer to scsi_sync_pkt(9F).
All tags in use!
The driver could not allocate another tag number. The target devices do not properly
support tagged queuing.
Gross error in FAS366 status
The driver experienced severe SCSI bus problems. Check cables and terminator.

262 man pages section 7: Device and Network Interfaces • Last Revised 20 Jun 1997
 fas(7D)

Spurious interrupt
The driver received an interrupt while the hardware was not interrupting.
Lost state in phasemanage
The driver is confused about the state of the SCSI bus.
Unrecoverable DMA error during selection
The DMA controller experienced host SBus problems. Check for bad hardware.
Bad sequence step (<step number>) in selection
The FAS366 hardware reported a bad sequence step. Check for bad hardware.
Undetermined selection failure
The selection of a target failed unexpectedly. Check for bad hardware.
Target <n>: failed reselection (bad reselect bytes)
A reconnect failed, target sent incorrect number of message bytes. Check for bad hardware.
Target <n>: failed reselection (bad identify message)
A reconnect failed, target didn't send identify message or it got corrupted. Check for bad
hardware.
Target <n>: failed reselection (not in msgin phase)
Incorrect SCSI bus phase after reconnection. Check for bad hardware.
Target <n>: failed reselection (unexpected bus free)
Incorrect SCSI bus phase after reconnection. Check for bad hardware.
Target <n>: failed reselection (timeout on receiving tag msg)
A reconnect failed; target failed to send tag bytes. Check for bad hardware.
Target <n>: failed reselection (botched tag)
A reconnect failed; target failed to send tag bytes. Check for bad hardware.
Target <n>: failed reselection (invalid tag)
A reconnect failed; target sent incorrect tag bytes. Check for bad hardware.
Target <n>: failed reselection (Parity error in reconnect msg's)
A reconnect failed; parity error detected. Check for bad hardware.
Target <n>: failed reselection (no command)
A reconnect failed; target accepted abort or reset, but still tries to reconnect. Check for
bad hardware.
Unexpected bus free
Target disconnected from the bus without notice. Check for bad hardware.
Target <n> didn't disconnect after sending <message>
The target unexpectedly did not disconnect after sending <message>.
Bad sequence step (0x?) in selection
The sequence step register shows an improper value. The target might be misbehaving.

Device and Network Interfaces 263

fas(7D)

Illegal dma boundary?

An attempt was made to cross a boundary that the driver could not handle.
Unwanted data xfer direction for Target <n>
The target went into an unexpected phase.
Unrecoverable DMA error on dma <send/receive>
There is a DMA error while sending/receiving data. The host DMA controller is
experiencing some problems.
SCSI bus DATA IN phase parity error
The driver detected parity errors on the SCSI bus.
SCSI bus MESSAGE IN phase parity error
The driver detected parity errors on the SCSI bus.
SCSI bus STATUS phase parity error
The driver detected parity errors on the SCSI bus.
Premature end of extended message
An extended SCSI bus message did not complete. Suspect a target firmware problem.
Premature end of input message
A multibyte input message was truncated. Suspect a target firmware problem.
Input message botch
The driver is confused about messages coming from the target.
Extended message <n> is too long
The extended message sent by the target is longer than expected.
<name> message <n> from Target <m> garbled
Target <m> sent message <name> of value <n> which the driver did not understand.
Target <n> rejects our message <name>
Target <n> rejected a message sent by the driver.
Rejecting message <name> from Target <n>
The driver rejected a message received from target <n>.
Cmd transmission error
The driver was unable to send out command bytes.
Target <n> refused message resend
The target did not accept a message resend.
MESSAGE OUT phase parity error
The driver detected parity errors on the SCSI bus.
Two byte message <name> <value> rejected
The driver does not accept this two byte message.

264 man pages section 7: Device and Network Interfaces • Last Revised 20 Jun 1997
 fas(7D)

Gross error in fas status <stat>

The fas chip has indicated a gross error like FIFO overflow.
Polled cmd failed (target busy)
A polled command failed because the target did not complete outstanding commands
within a reasonable time.
Polled cmd failed
A polled command failed because of timeouts or bus errors.
Auto request sense failed
Driver is unable to get request sense from the target.
Disconnected command timeout for Target <id>.<lun>
A timeout occurred while target id/lun was disconnected. This is usually a target firmware
problem. For tagged queuing targets, <n> commands were outstanding when the timeout
was detected.
Disconnected tagged cmds (<n>) timeout for Target <id>.<lun>
A timeout occurred while target id/lun was disconnected. This is usually a target firmware
problem. For tagged queuing targets, <n> commands were outstanding when the timeout
was detected.
Connected command timeout for Target <id>.<lun>
This is usually a SCSI bus problem. Check cables and termination.
Target <id>.<lun> reverting to async. mode
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Target <id>.<lun> reducing sync. transfer rate
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Reverting to slow SCSI cable mode
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Target <id> reducing sync. transfer rate
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Target <id> reverting to async. mode
A data transfer hang was detected. The driver attempts to eliminate this problem by
reducing the data transfer rate.
Target <id> disabled wide SCSI mode
Due to problems on the SCSI bus, the driver goes into more conservative mode of
operation to avoid further problems.

Device and Network Interfaces 265

fas(7D)

Reset SCSI bus failed

An attempt to reset the SCSI bus failed.
External SCSI bus reset
Another initiator reset the SCSI bus.

Warnings The fas hardware (FAS366) supports both Wide and Fast SCSI mode, but fast20 is not
supported. The maximum SCSI bandwidth is 20 MB/sec. Initiator mode block sequence (IBS)
is not supported.

Notes The fas driver exports properties indicating per target the negotiated transfer speed
(target<n>-sync-speed), whether wide bus is supported (target<n>-wide), scsi-options
for that particular target (target<n>-scsi-options), and whether tagged queuing has been
enabled (target<n>-TQ). The sync-speed property value is the data transfer rate in KB/sec.
The target<n>-TQ and the target<n>-wide property have value 1 to indicate that the
corresponding capability is enabled, or 0 to indicate that the capability is disabled for that
target. Refer to prtconf(1M) (verbose option) for viewing the fas properties.
SUNW,fas,instance #1
Driver software properties:
name <target3-TQ> length <4>
value <0x00000001>.
name <target3-wide> length <4>
value <0x00000000>.
name <target3-sync-speed> length <4>
value <0x00002710>.
name <target3-scsi-options> length <4>
value <0x000003f8>.
name <target0-TQ> length <4>
value <0x00000001>.
name <pm_norm_pwr> length <4>
value <0x00000001>.
name <pm_timestamp> length <4>
value <0x30040346>.
name <scsi-options> length <4>
value <0x000003f8>.
name <scsi-watchdog-tick> length <4>
value <0x0000000a>.
name <scsi-tag-age-limit> length <4>
value <0x00000002>.
name <scsi-reset-delay> length <4>
value <0x00000bb8>.
Register Specifications:
Bus Type=0x3, Address=0x8800000, Size=10
Bus Type=0x3, Address=0x8810000, Size=40
Interrupt Specifications:
Interrupt Priority=0x35 (ipl 5)

266 man pages section 7: Device and Network Interfaces • Last Revised 20 Jun 1997
 fasttrap(7D)

Name fasttrap – DTrace user instruction tracing provider

Description The fasttrap driver is a DTrace dynamic tracing provider that performs dynamic
instrumentation of arbitrary instructions in Solaris processes. The fasttrap driver
implements the DTrace fasttrap and pid providers.

The fasttrap driver is not a public interface and you access instrumentation offered by this
provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description of the
public documented interfaces available for the DTrace facility and the probes offered by the
fasttrap provider.

Sparc Only The fasttrap provider provides a DTrace probe that fires each time a user process executes
an instruction. The pid provider allows for the dynamic creation of DTrace probes
corresponding to instruction locations inside any user process specified using a process ID
and an instruction address or symbol name. Together these providers permit DTrace users to
perform instrumentation of Solaris user processes and to trace the interactions between
processes and the operating system. See the chapter entitled “User Process Tracing" in the
Solaris Dynamic Tracing Guide for information on how to use these providers to instrument
processes.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), attributes(5), dtrace(7D)

Solaris Dynamic Tracing Guide

Device and Network Interfaces 267

fbio(7I)

Name fbio – frame buffer control operations

Description The frame buffers provided with this release support the same general interface that is defined
by <sys/fbio.h>. Each responds to an FBIOGTYPE ioctl(2) request which returns
information in a fbtype structure.

Each device has an FBTYPE which is used by higher-level software to determine how to
perform graphics functions. Each device is used by opening it, doing an FBIOGTYPE ioctl() to
see which frame buffer type is present, and thereby selecting the appropriate
device-management routines.

FBIOGINFO returns information specific to the GS accelerator.

FBIOSVIDEO and FBIOGVIDEO are general-purpose ioctl() requests for controlling possible
video features of frame buffers. These ioctl() requests either set or return the value of a flags
integer. At this point, only the FBVIDEO_ON option is available, controlled by FBIOSVIDEO.
FBIOGVIDEO returns the current video state.

The FBIOSATTR and FBIOGATTR ioctl() requests allow access to special features of newer
frame buffers. They use the fbsattr and fbgattr structures.

Some color frame buffers support the FBIOPUTCMAP and FBIOGETCMAP ioctl() requests,
which provide access to the colormap. They use the fbcmap structure.

Also, some framebuffers with multiple colormaps will either encode the colormap identifier in
the high-order bits of the index field in the fbcmap structure, or use the FBIOPUTCMAPI and
FBIOGETCMAPI ioctl() requests.

FBIOVERTICAL is used to wait for the start of the next vertical retrace period.

FBIOVRTOFFSET Returns the offset to a read-only vertical retrace page for those framebuffers
that support it. This vertical retrace page may be mapped into user space with mmap(2). The
first word of the vertical retrace page (type unsigned int) is a counter that is incremented every
time there is a vertical retrace. The user process can use this counter in a variety of ways.

FBIOMONINFO returns a mon_info structure which contains information about the monitor
attached to the framebuffer, if available.

FBIOSCURSOR, FBIOGCURSOR, FBIOSCURPOS and FBIOGCURPOS are used to control the hardware
cursor for those framebuffers that have this feature. FBIOGCURMAX returns the maximum sized
cursor supported by the framebuffer. Attempts to create a cursor larger than this will fail.

Finally FBIOSDEVINFO and FBIOGDEVINFO are used to transfer variable-length, device-specific

information into and out of framebuffers.

268 man pages section 7: Device and Network Interfaces • Last Revised 28 Oct 2009
 fbio(7I)

See Also ioctl(2), mmap(2)

Bugs The FBIOSATTR and FBIOGATTR ioctl() requests are only supported by frame buffers which
emulate older frame buffer types. If a frame buffer emulates another frame buffer, FBIOGTYPE
returns the emulated type. To get the real type, use FBIOGATTR.

The FBIOGCURPOS ioctl was incorrectly defined in previous operating systems, and older code
running in binary compatibility mode may get incorrect results.

Device and Network Interfaces 269

fbt(7D)

Name fbt – DTrace function boundary tracing provider

Description The fbt driver is a DTrace dynamic tracing provider that performs dynamic instrumentation
at function boundaries in the Solaris kernel.

The function is the fundamental unit of program text. In a well-designed system, the function
performs a discrete and well-defined operation on a specified object or series of like objects.
Most functions are implemented by themselves calling functions on encapsulated objects, but
some functions —so-called "leaf functions" — are implemented without making further
function calls. The Function Boundary Tracing fbt provider contains a mechanism for
instrumenting the vast majority of functions in the kernel and offering the instrumentation as
a set of DTrace probes.

The fbt driver is not a public interface and you access the instrumentation offered by this
provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description of the
public documented interfaces available for the DTrace facility and the probes offered by the
fbt provider.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), attributes(5), dtrace(7D)

Solaris Dynamic Tracing Guide

270 man pages section 7: Device and Network Interfaces • Last Revised 4 Sep 2003
 fcip(7D)

Name fcip – IP/ARP over Fibre Channel datagram encapsulation driver

Synopsis /dev/fcip

Description The fcip driver is a Fibre Channel upper layer protocol module for encapsulating IP (IPv4)
and ARP datagrams over Fibre Channel. The fcip driver is a loadable, clonable, STREAMS
driver supporting the connectionless Data Link Provider Interface, dlpi(7P) over any Sun
Fibre Channel transport layer-compliant host adapter.

The fcip driver complies with the RFC 2625 specification for encapsulating IP/ARP
datagrams over Fibre Channel, and allows encapsulation of IPv4 only, as specified in RFC
2625. The fcip driver interfaces with the fp(7d) Sun Fibre Channel port driver.

Application The cloning character-special device /dev/fcip is used to access all Fibre Channel ports
Programming capable of supporting IP/ARP traffic on the system.
Interface
fcip and DLPI The fcip driver is a "style 2" Data Link Service Provider. All M_PROTO and M_PCPROTO type
messages are interpreted as DLPI primitives. Valid DLPI primitives are defined in
<sys/dlpi.h>. Refer to dlpi(7P) for more information on DLPI primitives.

An explicit DL_ATTACH_REQ message must be sent to associate the opened stream with a
particular Fibre Channel port (ppa). The ppa ID is interpreted as an unsigned long data type
and indicates the corresponding Fibre Channel port driver instance number. An error
(DL_ERROR_ACK) is returned by the driver if the ppa field value does not correspond to a valid
port driver instance number or if the Fibre Channel port is not ONLINE. Refer to fp(7d) for
more details on the Fibre Channel port driver.

The values returned by the driver in the DL_INFO_ACK primitive in response to a DL_INFO_REQ
from the user are as follows:
■ Maximum SDU is 65280 (defined in RFC 2625).
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP length is -2.
■ Service mode is DL_CLDLS.
■ Optional quality of service (QOS) fields are set to 0.
■ Provider style is DL_STYLE2.
■ Provider version is DL_VERSION_2.
■ Broadcast address value is 0xFFFFFFFF.

Once in DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular SAP
(Service Access Point) with the stream. The fcip driver DLSAP address format consists of the
6–byte physical address component followed immediately by the 2–byte SAP component
producing an 8–byte DLSAP address. Applications should not be programmed to use this
implementation-specific DLSAP address format, but use information returned in the
DL_INFO_ACK primitive to compose and decompose DLSAP addresses. The SAP length, full

Device and Network Interfaces 271

fcip(7D)

DLSAP length, and SAP/physical ordering are included within the DL_INFO_ACK. The physical
address length is the full DLSAP address length minus the SAP length. The physical address
length can also be computed by issuing the DL_PHYS_ADDR_REQ primitive to obtain the current
physical address associated with the stream.

Once in the DL_BOUND state, the user can transmit frames on the fibre by sending
DL_UNITDATA_REQ messages to the fcip driver. The fcip driver will route received frames up
any of the open and bound streams having a SAP which matches the received frame's SAP
type as DL_UNITDATA_IND messages. Received Fibre Channel frames are duplicated and routed
up multiple open streams if necessary. The DLSAP address contained within the
DL_UNITDATA_REQ and DL_UNITDATA_IND messages consists of both the SAP (type) and
physical address (WorldWideName) components.

Other Primitives In Fibre Channel, multicasting is defined as an optional service for Fibre Channel classes three
and six only. If required, the Fibre Channel broadcast service can be used for multicasting. The
RFC 2625 specification does not support IP multicasting or promiscuous mode.

fcip Fibre Channel ELS The fcip driver will use the FARP Fibre Channel Extended Link Service (ELS), where
supported, to resolve WorldWide Names (MAC address) to FC Port Identifiers(Port_ID). The
fcip driver also supports InARP to resolve WorldWide Name and Port_ID to an IP address.
Files /dev/fcip fcip character-special device
/kernel/drv/fcip 32–bit ELF kernel driver (x86)
/kernel/drv/amd64/fcip 64–bit ELF kernel driver (x86)
/kernel/drv/sparcv9/fcip 64–bit ELF kernel driver (SPARC)
/kernel/drv/fcip.conf fcip driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWfcip

See Also netstat(1M), prtconf(1M), driver.conf(4), fp(7d), dlpi(7P)

Writing Device Drivers

IP and ARP over Fibre Channel, RFC 2625 M. Rajagopal, R. Bhagwat, W. Rickard. Gadzoox
Networks, June 1999

ANSI X3.230-1994, Fibre Channel Physical and Signalling Interface (FC-PH)

ANSI X3.272-1996, Fibre Channel Arbitrated Loop (FC-AL)

272 man pages section 7: Device and Network Interfaces • Last Revised 18 Oct 2004
 fcip(7D)

Notes If you use a Fibre Channel adapter with two or more ports that each share a common Node
WorldWideName, the fcip driver will likely attach to the first port on the adapter.

RFC 2625 requires that both source and destination WorldWideNames have their 4 bit NAA
identifiers set to binary '0001,' indicating that an IEEE 48–bit MAC address is contained in the
lower 48 bits of the network address fields. For additional details, see the RFC 2625
specification.

Device and Network Interfaces 273

fcoe(7D)

Name fcoe – fibre channel over Ethernet transport driver

Description The fcoe driver is a pseudo nexus driver which supports the transportation of FCoE
encapsualted frames. FCoE Ethernet frame will encapsulate the raw Fibre Channel frame.

The fcoe driver interfaces with FCoE target mode device driver, fcoet(7D).
Files /kernel/drv/fcoe 32-bit ELF kernel module (x86)
/kernel/drv/amd64/fcoe 64-bit ELF kernel module (x86)
kernel/drv/sparcv 64-bit ELF kernel module (SPARC)

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWfcoe

See Also driver.conf(4), attributes(5), fcoet(7D)

Writing Device Drivers

ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP)

274 man pages section 7: Device and Network Interfaces • Last Revised 18 Mar 2009
 fcoei(7D)

Name fcoei – Fibre Channel Over Ethernet Initiator Mode Driver

Synopsis fcoei@port,0

Description The fcoei driver is a pseudo device driver which encapsulates the raw Fibre Channel frames
into FCoE ethernet frames, or decapsulates FC frames from FCoE ethernet frames. The
supported FC frames include extended/basic link services, common transport frames and
initiator mode FCP frames.

The fcoei driver interfaces with the Sun Fibre Channel port driver, fp(7d), and the FCoE
transport driver, fcoe(7D).
Files /kernel/drv/fcoei 32-bit ELF kernel module (x86)
/kernel/drv/amd64/fcoei 64-bit ELF kernel module (x86)
kernel/drv/sparcv/fcoei 64-bit ELF kernel module (SPARC)

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWfcoei

See Also driver.conf(4), attributes(5), fcoe(7D), fcoet(7D), fp(7d)

Writing Device Drivers

ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP)

Device and Network Interfaces 275

fcoet(7D)

Name fcoet – fibre channel over Ethernet target mode driver

Description The fcoet driver is a pseudo device driver which encapsulates the raw Fibre Channel frames
into FCoE Ethernet frames, or decapsulates FC frames from FCoE Ethernet frames. The
supported FC frames contain extended/basic link services, common transport frames and
target mode FCP frames.

The fcoet driver interfaces with COMSTAR FC transport driver,fct, and FCoE transport
driver, fcoe(7D).
Files /kernel/drv/fcoet 32-bit ELF kernel module (x86)
/kernel/drv/amd64/fcoet 64-bit ELF kernel module (x86)
/kernel/drv/sparcv9/fcoet 64-bit ELF kernel module (SPARC)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWfcoet

See Also driver.conf(4), attributes(5), fcoe(7D)

Writing Device Drivers

ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP)

276 man pages section 7: Device and Network Interfaces • Last Revised 18 Mar 2009
 fcp(7D)

Name fcp – Fibre Channel protocol driver

Description The fcp driver is the upper layer protocol that supports mechanisms for transporting SCSI-3
commands over Fibre Channel. The fcp driver, which interfaces with the Sun Fibre Channel
transport library fctl(7D), supports the standard functions provided by the SCSA interface.
Files /kernel/drv/fcp 32–bit ELF kernel driver (x86)
/kernel/drv/amd64/fcp 64–bit ELF kernel driver (x86)
/kernel/drv/sparcv9/fcp 64–bit ELF kernel driver (SPARC)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Interface stability Unknown

Availability SUNWfcp

See Also prtconf(1M), driver.conf(4), fctl(7D), fp(7d), usoc(7D)

Writing Device Drivers

Fibre Channel Physical and Signaling Interface (FC-PH) ANSI X3.230: 1994

Fibre Channel Generic Services (FC-GS-2) Project 1134-D

Fibre Channel Arbitrated Loop (FC-AL) ANSI X3.272-1996

Fibre Channel Protocol for SCSI (FCP) ANSI X3.269-1996

SCSI-3 Architecture Model (SAM) Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)
ANSI X3.270-1996

Fabric Loop Attachment (FC-FLA), NCITS TR-20:1998

Device and Network Interfaces 277

fctl(7D)

Name fctl – Sun Fibre Channel transport library

Description The fctl kernel module interfaces the Sun Fibre Channel upper layer protocol (ULP)
mapping modules with Sun Fibre Channel adapter (FCA) drivers. There are no
user-configurable options for this module.
Files /kernel/misc/fctl 32–bit ELF kernel module (x86)
/kernel/misc/amd64/fctl 64–bit ELF kernel module (x86)
/kernel/misc/sparcv9/fctl 64–bit ELF kernel module (SPARC)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Interface stability Unknown

Availability SUNWfctl

See Also fp(7d)

278 man pages section 7: Device and Network Interfaces • Last Revised 20 Oct 2004
 fd(7D)

Name fd, fdc – drivers for floppy disks and floppy disk controllers

Synopsis
SPARC /dev/diskette0

/dev/rdiskette0
x86 /dev/diskette[0-1]

/dev/rdiskette[0-1]

Description The fd and fdc drivers provide the interfaces to floppy disks using the Intel 8272, Intel 82077,
NEC 765, or compatible disk controllers on x86 based systems.

The default partitions for the floppy driver are:

a All cylinders except the last
b Only the last cylinder
c Entire diskette

The fd driver autosenses the density of the diskette.

When the floppy is first opened the driver looks for a SunOS label in logical block 0 of the
diskette. If attempts to read the SunOS label fail, the open will fail. If block 0 is read
successfully but a SunOS label is not found, auto-sensed geometry and default partitioning are
assumed.

The fd driver supports both block and raw interfaces.

The block files (/dev/diskette*) access the diskette using the system's normal buffering
mechanism and may be read and written without regard to physical diskette records.

There is also a raw (/dev/rdiskette*) interface that provides for direct transmission between
the diskette and the user's read or write buffer. A single read(2) or write(2) call usually results
in one I/O operation; therefore raw I/O is considerably more efficient when larger blocking
factors are used. A blocking factor of no less than 8 Kbytes is recommended. See the Notes
section, below, for information on the number of sectors per track.

3.5" Diskettes For 3.5" double-sided diskettes, the following densities are supported:

SPARC 1.7 Mbyte density 80 cylinders, 21 sectors per track, 1.7 Mbyte capacity
high density 80 cylinders, 18 sectors per track, 1.44 Mbyte capacity
double density 80 cylinders, 9 sectors per track, 720 Kbyte capacity
x86 extended density 80 cylinders, 36 sectors per track, 2.88 Mbyte capacity
1.7 Mbyte density 80 cylinders, 21 sectors per track, 1.7 Mbyte capacity

Device and Network Interfaces 279

fd(7D)

high density 80 cylinders, 18 sectors per track, 1.44 Mbyte capacity

double density 80 cylinders, 9 sectors per track, 760 Kbyte capacity

5.25" Diskettes For 5.25" double-sided diskettes on x86 platforms, the densities listed below are supported:

SPARC 5.25" diskettes are not supported on SPARC platforms.

x86 high density 80 cylinders, 15 sectors per track, 1.2 Mbyte capacity
double density 40 cylinders, 9 sectors per track, 360 Kbyte capacity
double density 40 cylinders, 8 sectors per track, 320 Kbyte capacity
quad density 80 cylinders, 9 sectors per track, 720 Kbyte capacity
double density 40 cylinders, 16 sectors per track (256 bytes per sector), 320 Kbyte capacity
double density 40 cylinders, 4 sectors per track (1024 bytes per sector), 320 Kbyte capacity
Errors EBUSY During opening, the partition has been opened for exclusive access and another
process wants to open the partition. Once open, this error is returned if the floppy
disk driver attempted to pass a command to the floppy disk controller when the
controller was busy handling another command. In this case, the application
should try the operation again.
EFAULT An invalid address was specified in an ioctl command (see fdio(7I)).
EINVAL The number of bytes read or written is not a multiple of the diskette's sector size.
This error is also returned when an unsupported command is specified using the
FDIOCMD ioctl command (see fdio(7I)).
EIO During opening, the diskette does not have a label or there is no diskette in the
drive. Once open, this error is returned if the requested I/O transfer could not be
completed.
ENOSPC An attempt was made to write past the end of the diskette.
ENOTTY The floppy disk driver does not support the requested ioctl functions (see
fdio(7I)).
ENXIO The floppy disk device does not exist or the device is not ready.
EROFS The floppy disk device is opened for write access and the diskette in the drive is
write protected.
x86 Only ENOSYS The floppy disk device does not support the requested ioctl function ( FDEJECT).

x86 Configuration The driver attempts to initialize itself using the information found in the configuration file,
/platform/i86pc/kernel/drv/fd.conf.
name="fd" parent="fdc" unit=0;
name="fd" parent="fdc" unit=1;

280 man pages section 7: Device and Network Interfaces • Last Revised 2 Mar 2007
 fd(7D)

Files
SPARC /platform/sun4u/kernel/drv/fd driver module
/usr/include/sys/fdreg.h structs and definitions for Intel 82072 and 82077
controllers
/usr/include/sys/fdvar.h structs and definitions for floppy drivers
/dev/diskette device file
/dev/diskette0 device file
/dev/rdiskette raw device file
/dev/rdiskette0 raw device file

For ucb Compatibility

/dev/fd0[a-c] block file
/dev/rfd0[a-c] raw file
/dev/diskette0 directory containing volume management character device file
/dev/rdiskette0 directory containing the volume management raw character
device file
/dev/aliases/floppy0 symbolic link to the entry in /dev/rdiskette0
x86 /platform/i86pc/kernel/drv/fd driver module
/platform/i86pc/kernel/drv/fd.conf configuration file for floppy driver
/platform/i86pc/kernel/drv/fdc floppy-controller driver module
/platform/i86pc/kernel/drv/fdc.conf configuration file for the floppy-controller
/usr/include/sys/fdc.h structs and definitions for x86 floppy devices
/usr/include/sys/fdmedia.h structs and definitions for x86 floppy media

x86 First Drive

/dev/diskette device file
/dev/diskette0 device file
/dev/rdiskette raw device file
/dev/rdiskette0 raw device file

For ucb Compatibility

/dev/fd0[a-c] block file

Device and Network Interfaces 281

fd(7D)

/dev/rfd0[a-c] raw file

/dev/diskette0 directory containing volume management character device file
/dev/rdiskette0 directory containing the volume management raw character
device file
/dev/aliases/floppy0 symbolic link to the entry in /dev/rdiskette0

x86 Second Drive

/dev/diskette1 device file
/dev/rdiskette1 raw device file

For ucb Compatibility

/dev/fd1[a-c] block file
/dev/rfd1[a-c] raw file
/dev/diskette1 directory containing volume management character device file
/dev/rdiskette1 directory containing the volume management raw character
device file
/dev/aliases/floppy1 symbolic link to the entry in /dev/rdiskette1

See Also fdformat(1), dd(1M), drvconfig(1M), read(2), write(2), driver.conf(4), dkio(7I) fdio(7I)

Diagnostics
All Platforms fd<n>: <command name> failed (<sr1> <sr2> <sr3>)
The <command name> failed after several retries on drive <n>. The three hex values in
parenthesis are the contents of status register 0, status register 1, and status register 2 of the
Intel 8272, the Intel 82072, and the Intel 82077 Floppy Disk Controller on completion of
the command, as documented in the data sheet for that part. This error message is usually
followed by one of the following, interpreting the bits of the status register:
fd<n>: not writable
fd<n>: crc error blk <block number>

There was a data error on <block number>.

fd<n>: bad format
fd<n>: timeout
fd<n>: drive not ready
fd<n>: unformatted diskette or no diskette in drive
fd<n>: block <block number> is past the end!

282 man pages section 7: Device and Network Interfaces • Last Revised 2 Mar 2007
 fd(7D)

(nblk=<total number of blocks>)

The operation tried to access a block number that is greater than the total
number of blocks.
fd<n>: b_bcount 0x<op_size> not % 0x<sect_size>

The size of an operation is not a multiple of the sector size.

fd<n>: overrun/underrun
fd<n>: host bus error. There was a hardware error on a system bus.

SPARC Only Overrun/underrun errors occur when accessing a diskette while the system is heavily loaded.
Decrease the load on the system and retry the diskette access.

Notes 3.5" high density diskettes have 18 sectors per track and 5.25" high density diskettes have 15
sectors per track. They can cross a track (though not a cylinder) boundary without losing data,
so when using dd(1M) or read(2)/write(2) calls to or from the raw diskette, you should
specify bs=18k or multiples thereof for 3.5" diskettes, and bs=15k or multiples thereof for
5.25" diskettes.

The SPARC fd driver is not an unloadable module.

Under Solaris (x86 Edition), the configuration of the floppy drives is specified in CMOS
configuration memory. Use the BIOS setup program for the system to define the diskette size
and density/capacity for each installed drive. Note that MS-DOS may operate the floppy
drives correctly, even though the CMOS configuration may be in error. Solaris (x86 Edition)
relies on the CMOS configuration to be accurate.

Device and Network Interfaces 283

fdio(7I)

Name fdio – floppy disk control operations

Synopsis #include <sys/fdio.h>

Description The Solaris floppy driver supports a set of ioctl(2) requests for getting and setting the floppy
drive characteristics. Basic to these ioctl( ) requests are the definitions in <sys/fdio.h>.

ioctls The following ioctl( ) requests are available on the Solaris floppy driver.
FDDEFGEOCHAR x86 based systems: This ioctl( ) forces the floppy driver to restore the
diskette and drive characteristics and geometry, and partition information
to default values based on the device configuration.
FDGETCHANGE The argument is a pointer to an int. This ioctl( ) returns the status of the
diskette-changed signal from the floppy interface. The following defines
are provided for cohesion.

Note: For x86 based systems, use FDGC_DETECTED (which is available only on x86 based
systems) instead of FDGC_HISTORY.
/*
* Used by FDGETCHANGE, returned state of the sense disk change bit.
*/
#define FDGC_HISTORY 0x01 /* disk has changed since insertion or
last FDGETCHANGE call */
#define FDGC_CURRENT 0x02 /* if set, indicates drive has floppy,
> otherwise, drive is empty */
#define FDGC_CURWPROT 0x10 /* current state of write protect */
#define FDGC_DETECTED 0x20 /* previous state of DISK CHANGE */
FDIOGCHAR The argument is a pointer to an fd_char structure (described below). This
ioctl( ) gets the characteristics of the floppy diskette from the floppy
controller.
FDIOSCHAR The argument is a pointer to an fd_char structure (described below). This
ioctl( ) sets the characteristics of the floppy diskette for the floppy controller.
Typical values in the fd_char structure for a high density diskette:
field value
fdc_medium 0
fdc_transfer_rate 500
fdc_ncyl 80
fdc_nhead 2
fdc_sec_size 512
fdc_secptrack 18
fdc_steps -1 { This field doesn’t apply. }
/*
* Floppy characteristics
*/

284 man pages section 7: Device and Network Interfaces • Last Revised 26 April 2001
 fdio(7I)

struct fd_char {
uchar_t fdc_medium; /* equals 1 if floppy is medium density format */
int fdc_transfer_rate; /* transfer rate */
int fdc_ncyl; /* number of cylinders */
int fdc_nhead; /* number of heads */
int fdc_sec_size; /* sector size */
int fdc_secptrack; /* sectors per track */
int fdc_steps; /* no. of steps per data track */
};
FDGETDRIVECHAR The argument to this ioctl( ) is a pointer to an fd_drive structure
(described below). This ioctl( ) gets the characteristics of the floppy
drive from the floppy controller.
FDSETDRIVECHAR x86 based systems: The argument to this ioctl( ) is a pointer to an
fd_drive structure (described below). This ioctl( ) sets the
characteristics of the floppy drive for the floppy controller. Only
fdd_steprate, fdd_headsettle, fdd_motoron, and fdd_motoroff are
actually used by the floppy disk driver.
/*
* Floppy Drive characteristics
*/
struct fd_drive {
int fdd_ejectable; /* does the drive support eject? */
int fdd_maxsearch; /* size of per-unit search table */
int fdd_writeprecomp; /* cyl to start write precompensation */
int fdd_writereduce; /* cyl to start recucing write current */
int fdd_stepwidth; /* width of step pulse in 1 us units */
int fdd_steprate; /* step rate in 100 us units */
int fdd_headsettle; /* delay, in 100 us units */
int fdd_headload; /* delay, in 100 us units */
int fdd_headunload; /* delay, in 100 us units */
int fdd_motoron; /* delay, in 100 ms units */
int fdd_motoroff; /* delay, in 100 ms units */
int fdd_precomplevel; /* bit shift, in nano-secs */
int fdd_pins; /* defines meaning of pin 1, 2, 4 and 34 */
int fdd_flags; /* TRUE READY, Starting Sector #, & Motor On */
};
FDGETSEARCH Not available.
FDSETSEARCH Not available.
FDEJECT SPARC: This ioctl( ) requests the floppy drive to eject the diskette.
FDIOCMD The argument is a pointer to an fd_cmd structure (described below). This
ioctl( ) allows access to the floppy diskette using the floppy device driver.
Only the FDCMD_WRITE, FDCMD_READ, and FDCMD_FORMAT_TRACK commands

Device and Network Interfaces 285

fdio(7I)

are currently available.

struct fd_cmd {
ushort_t fdc_cmd; /* command to be executed */
int fdc_flags; /* execution flags (x86 only) */
daddr_t fdc_blkno; /* disk address for command */
int fdc_secnt; /* sector count for command */
caddr_t fdc_bufaddr; /* user’s buffer address */
uint_t fdc_buflen; /* size of user’s buffer */
};

Please note that the fdc_buflen field is currently unused. The fdc_secnt field is used to
calculate the transfer size, and the buffer is assumed to be large enough to accommodate the
transfer.
{
/*
* Floppy commands
*/
#define FDCMD_WRITE 1
#define FDCMD_READ 2
#define FDCMD_SEEK 3
#define FDCMD_REZERO 4
#define FDCMD_FORMAT_UNIT 5
#define FDCMD_FORMAT_TRACK 6
};
FDRAW The argument is a pointer to an fd_raw structure (described below). This ioctl( )
allows direct control of the floppy drive using the floppy controller. Refer to the
appropriate floppy-controller data sheet for full details on required command bytes
and returned result bytes. The following commands are supported.
/*
* Floppy raw commands
*/
#define FDRAW_SPECIFY 0x03
#define FDRAW_READID 0x0a (x86 only)
#define FDRAW_SENSE_DRV 0x04
#define FDRAW_REZERO 0x07
#define FDRAW_SEEK 0x0f
#define FDRAW_SENSE_INT 0x08 (x86 only)
#define FDRAW_FORMAT 0x0d
#define FDRAW_READTRACK 0x02
#define FDRAW_WRCMD 0x05
#define FDRAW_RDCMD 0x06
#define FDRAW_WRITEDEL 0x09
#define FDRAW_READDEL 0x0c

286 man pages section 7: Device and Network Interfaces • Last Revised 26 April 2001
 fdio(7I)

Please note that when using FDRAW_SEEK or FDRAW_REZERO, the driver automatically issues a
FDRAW_SENSE_INT command to clear the interrupt from the FDRAW_SEEK or the
FDRAW_REZERO. The result bytes returned by these commands are the results from the
FDRAW_SENSE_INT command. Please see the floppy-controller data sheet for more details on
FDRAW_SENSE_INT.
/*
* Used by FDRAW
*/
struct fd_raw {
char fdr_cmd[10]; /* user-supplied command bytes */
short fdr_cnum; /* number of command bytes */
char fdr_result[10]; /* controller-supplied result bytes */
ushort_t fdr_nbytes; /* number to transfer if read/write command */
char *fdr_addr; /* where to transfer if read/write command */
};

See Also ioctl(2), dkio(7I), fd(7D), hdio(7I)

Device and Network Interfaces 287

ffb(7D)

Name ffb – 24-bit UPA color frame buffer and graphics accelerator

Description ffb is a 24-bit UPA-based color frame buffer and graphics accelerator which comes in the two
configurations: single buffered frame and double buffered frame.
Single buffered frame buffer Consists of 32 video memory planes of 1280 x 1024 pixels,
including 24-bit single-buffering and 8-bit X planes.
Double buffered frame buffer Consists of 96 video memory planes of 1280 x 1024 pixels,
including 24-bit double-buffering, 8-bit X planes, 28-bit
Z-buffer planes and 4-bit Y planes.

The driver supports the following frame buffer ioctls which are defined in fbio(7I):
FBIOPUTCMAP, FBIOGETCMAP, FBIOSVIDEO, FBIOGVIDEO, FBIOVERTICAL,
FBIOSCURSOR, FBIOGCURSOR, FBIOSCURPOS, FBIOGCURPOS, FBIOGCURMAX,
FBIO_WID_PUT, FBIO_WID_GET
Files /dev/fbs/ffb0 device special file

See Also ffbconfig(1M), mmap(2), fbio(7I)

288 man pages section 7: Device and Network Interfaces • Last Revised 10 Jun 1997
 flowacct(7ipp)

Name flowacct – Flow Accouting module

Description The flow accounting module flowacct enables you to record flow details. You use flow details
to gather statistics and/or for billing purposes. Accounting consists of recording flow details in
a location you designate and in a format that you can retrieve at a later stage. IPQoS
accounting relies on the exacct mechanism to store and retrieve flow information.

A flow is defined by the 5-tuple - saddr, sport, daddr, dport and protocol.

Typically, the accounting module is the last datapath element in a sequence of actions. Flow
attributes include ToS/DS, user id, project id, creation time (time the flow was created), last
seen (when pkts for the flow were last seen), action name (instance that recorded the flow
information), nbytes and npackets. Attributes are split into groups entitled basic and extended.
The basic group records only the nbytes, npackets and action name, while the extended group
is a superset of the basic group and records all attributes. The attributes to be recorded, in
addition to the accounting file that contains flow details, are selected using acctadm(1M). The
flowacct module does not provide a mechanism to retrieve flow information from the
accounting file nor to interpret the retrieved information.

Statistics The flowacct module exports the following statistics available through kstat:
module: flowacct instance: <action id>
name: Flowacct statistics class <action name>
bytes_in_tbl <bytes in the flow table>
epackets <packets in error>
flows_in_tbl <flow records in the flow table>
nbytes <number of bytes through this instance>
npackets <number of packets>
usedmem <memory, in bytes, used by the flow table>
Files /kernel/ipp/sparcv9/flowacct 64-bit module (SPARC only.)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos

See Also ipqosconf(1M), acctadm(1M), libexacct3LIB, dlcosmk(7ipp), dscpmk(7ipp), ipqos(7ipp),

ipgpc(7ipp), tokenmt(7ipp), tswtclmt(7ipp)

Device and Network Interfaces 289

fp(7d)

Name fp – Sun Fibre Channel port driver

Description The fp driver is a Sun fibre channel nexus driver that enables fibre channel topology
discovery, device discovery, fibre channel adapter port management and other capabilities
through well-defined fibre channel adapter driver interfaces.

The fp driver requires the presence of a fabric name server in fabric and public loop topologies
to discover fibre channel devices. In private loop topologies, the driver discovers devices by
performing PLOGI to all valid AL_PAs, provided that devices do not participate in LIRP and
LILP stages of loop initialization. The fp driver also discovers devices in N_Port
point-to-point topologies.

Configuration The fp driver is configured by defining properties in the fp.conf file. Note that you must
reboot the system to have any changes you make to fp.conf take effect. The fp driver
supports the following properties:
mpxio-disable
Solaris I/O multipathing is enabled or disabled on fibre channel devices with the
mpxio-disable property. Specifying mpxio-disable="no" activates I/O multipathing, while
mpxio-disable="yes" disables the feature. Solaris I/O multipathing may be enabled or
disabled on a per port basis. Per port settings override the global setting for the specified
ports. The following example shows how to disable multipathing on port 0 whose parent is
/pci@8,600000/SUNW,qlc@4:
name="fp" parent="/pci@8,600000/SUNW,qlc@4" port=0
mpxio-disable="yes";
manual_configuration_only
Automatic configuration of SCSI devices in the fabric is enabled by default and thus allows
all devices discovered in the SAN zone to be enumerated in the kernel's device tree
automatically. The manual_configuration_only property may be configured to disable
the default behavior and force the manual configuration of the devices in the SAN.
Specifying manual_configuration_only=1 disables the automatic configuration of devices.
pwwn-lun-blacklist
Allows you to specify target port WWNs and LUN numbers you do not want configured.
LUN numbers are interpreted as decimals. White spaces and commas (',') can be used in
the list of LUN numbers.
#
# pwwn-lun-blacklist=
# "target-port-wwn,lun-list"
#
# To prevent LUNs 1 and 2 from being configured for target
# port 510000f010fd92a1 and target port 510000e012079df1, set:
#
# pwwn-lun-blacklist=
# "510000f010fd92a1,1,2",
# "510000e012079df1,1,2";

290 man pages section 7: Device and Network Interfaces • Last Revised 10 Jan 2008
 fp(7d)

#
Files /kernel/drv/fp 32-bit ELF kernel driver (x86)
/kernel/drv/amd64/fp 64-bit ELF kernel driver (x86)
/kernel/drv/sparcv9/fp 64–bit ELF kernel driver (SPARC)
/kernel/drv/fp.conf fp driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

mpxio-disable Uncommitted

manual_configuration_only Obsolete

Availability SUNWfctl

See Also cfgadm_fp(1M), prtconf(1M), stmsboot(1M), driver.conf(4), attributes(5), fcp(7D),

fctl(7D), scsi_vhci(7D)

Writing Device Drivers

Fibre Channel Physical and Signaling Interface (FC-PH) ANSI X3.230: 1994

Fibre Channel Generic Services (FC-GS-2) Project 1134-D

Fibre Channel Arbitrated Loop (FC-AL) ANSI X3.272-1996

Fibre Channel Protocol for SCSI (FCP) ANSI X3.269-1996

SCSI-3 Architecture Model (SAM) Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)
ANSI X3.270-1996

SCSI Direct Attach (FC-PLDA) ANSI X3.270-1996

SCSI Direct Attach (FC-PLDA) NCITS TR-19:1998

Fabric Loop Attachment (FC-FLA), NCITS TR-20:1998

Notes In N_Port point-to-point topologies, FCP error recovery does not work across events such as
link bounce/cable pull. I/O to devices with FCP-2/FCP-3 support (for example, FC tape
drives) will be disrupted by such events.

Device and Network Interfaces 291

FSS(7)

Name FSS – Fair share scheduler

Description The fair share scheduler (FSS) guarantees application performance by explicitly allocating
shares of CPU resources to projects. A share indicates a project's entitlement to available CPU
resources. Because shares are meaningful only in comparison with other project's shares, the
absolute quantity of shares is not important. Any number that is in proportion with the
desired CPU entitlement can be used.

The goals of the FSS scheduler differ from the traditional time-sharing scheduling class (TS).
In addition to scheduling individual LWPs, the FSS scheduler schedules projects against each
other, making it impossible for any project to acquire more CPU cycles simply by running
more processes concurrently.

A project's entitlement is individually calculated by FSS independently for each processor set
if the project contains processes bound to them. If a project is running on more than one
processor set, it can have different entitlements on every set. A project's entitlement is defined
as a ratio between the number of shares given to a project and the sum of shares of all active
projects running on the same processor set. An active project is one that has at least one
running or runnable process. Entitlements are recomputed whenever any project becomes
active or inactive, or whenever the number of shares is changed.

Processor sets represent virtual machines in the FSS scheduling class and processes are
scheduled independently in each processor set. That is, processes compete with each other
only if they are running on the same processor set. When a processor set is destroyed, all
processes that were bound to it are moved to the default processor set, which always exists.
Empty processor sets (that is, sets without processors in them) have no impact on the FSS
scheduler behavior.

If a processor set contains a mix of TS/IA and FSS processes, the fairness of the FSS scheduling
class can be compromised because these classes use the same range of priorities. Fairness is
most significantly affected if processes running in the TS scheduling class are CPU-intensive
and are bound to processors within the processor set. As a result, you should avoid having
processes from TS/IA and FSS classes share the same processor set. RT and FSS processes use
disjoint priority ranges and therefore can share processor sets.

As projects execute, their CPU usage is accumulated over time. The FSS scheduler periodically
decays CPU usages of every project by multiplying it with a decay factor, ensuring that more
recent CPU usage has greater weight when taken into account for scheduling. The FSS
scheduler continually adjusts priorities of all processes to make each project's relative CPU
usage converge with its entitlement.

While FSS is designed to fairly allocate cycles over a long-term time period, it is possible that
projects will not receive their allocated shares worth of CPU cycles due to uneven demand.
This makes one-shot, instantaneous analysis of FSS performance data unreliable.

Note that share is not the same as utilization. A project may be allocated 50% of the system,
although on the average, it uses just 20%. Shares serve to cap a project's CPU usage only when

292 man pages section 7: Device and Network Interfaces • Last Revised 1 Oct 2004
 FSS(7)

there is competition from other projects running on the same processor set. When there is no
competition, utilization may be larger than entitlement based on shares. Allocating a small
share to a busy project slows it down but does not prevent it from completing its work if the
system is not saturated.
The configuration of CPU shares is managed by the name server as a property of the
project(4) database. In the following example, an entry in the /etc/project file sets the
number of shares for project x-files to 10:
x-files:100::::project.cpu-shares=(privileged,10,none)

Projects with undefined number of shares are given one share each. This means that such
projects are treated with equal importance. Projects with 0 shares only run when there are no
projects with non-zero shares competing for the same processor set. The maximum number
of shares that can be assigned to one project is 65535.
You can use the prctl(1) command to determine the current share assignment for a given
project:
$ prctl -n project.cpu-shares -i project x-files

or to change the amount of shares if you have root privileges:

# prctl -r -n project.cpu-shares -v 5 -i project x-files

See the prctl(1) man page for additional information on how to modify and examine
resource controls associated with active processes, tasks, or projects on the system. See
resource_controls(5) for a description of the resource controls supported in the current
release of the Solaris operating system.
By default, project system (project ID 0) includes all system daemons started by initialization
scripts and has an “unlimited” amount of shares. That is, it is always scheduled first no matter
how many shares are given to other projects.
The following command sets FSS as the default scheduler for the system:
# dispadmin -d FSS

This change will take effect on the next reboot. Alternatively, you can move processes from the
time-share scheduling class (as well as the special case of init) into the FSS class without
changing your default scheduling class and rebooting by becoming root, and then using the
priocntl(1) command, as shown in the following example:
# priocntl -s -c FSS -i class TS
# priocntl -s -c FSS -i pid 1

Configuring You can use the dispadmin(1M) command to examine and tune the FSS scheduler's time
Scheduler With quantum value. Time quantum is the amount of time that a thread is allowed to run before it
Dispadmin must relinquish the processor. The following example dumps the current time quantum for
the fair share scheduler:

Device and Network Interfaces 293

FSS(7)

$ dispadmin -g -c FSS
#
# Fair Share Scheduler Configuration
#
RES=1000
#
# Time Quantum
#
QUANTUM=110

The value of the QUANTUM represents some fraction of a second with the fractional value
determied by the reciprocal value of RES. With the default value of RES = 1000, the reciprocal
of 1000 is .001, or milliseconds. Thus, by default, the QUANTUM value represents the time
quantum in milliseconds.

If you change the RES value using dispadmin with the -r option, you also change the
QUANTUM value. For example, instead of quantum of 110 with RES of 1000, a quantum of
11 with a RES of 100 results. The fractional unit is different while the amount of time is the
same.

You can use the -s option to change the time quantum value. Note that such changes are not
preserved across reboot. Please refer to the dispadmin(1M) man page for additional
information.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SUNWcsu

See Also prctl(1), priocntl(1), dispadmin(1M), psrset(1M), priocntl(2), project(4),

attributes(5), resource_controls(5)

System Administration Guide: Virtualization Using the Solaris Operating System

294 man pages section 7: Device and Network Interfaces • Last Revised 1 Oct 2004
 ge(7D)

Name ge – GEM Gigabit-Ethernet device driver

Synopsis /dev/ge
Description The ge Gigabit-Ethernet driver is a multi-threaded, loadable, clonable, STREAMS hardware
driver supporting the connectionless Data Link Provider Interface, dlpi(7P) over GEM, SBus
and PCI Gigabit-Ethernet add-in adapters. Multiple GEM-based adapters installed within the
system are supported by the driver. The ge driver provides basic support for the GEM-based
Ethernet hardware and handles the SUNW,sbus-gem (SBus GEM) and pci108e,2bad (PCI
GEM) devices. Functions include chip initialization, frame transmit and receive, multicast
and promiscuous support, and error recovery and reporting.
The GEM device provides 1000BASE-SX networking interfaces using the GEM ASIC external
SERDES and fiber optical transceiver. The GEM ASIC provides the appropriate bus interface,
MAC functions and physical code sub-layer (PCS) functions. The external SERDES connects
to a fiber transceiver and provides the physical connection.
The 1000Base-SX standard specifies an auto-negotiation protocol to automatically select the
mode of operation. In addition to duplex operation, the GEM ASIC can auto-negotiate for
IEEE 802.3x frame based flow control capabilities. The GEM PCS is capable of performing
auto-negotiation using the remote (or link partner) link end and receives the capabilities of
the remote end. It selects the highest common demoninator mode of operation based on
priorities. The ge driver also supports forced-mode operation under which the driver selects
the mode of operation.
Application The cloning character-special device /dev/ge is used to access all ge controllers installed
Programming within the system.
Interface
ge and DLPI The ge driver is a Style 2 data link service provider. All M_PROTO and M_PCPROTO type messages
are interpreted as DLPI primitives. Valid DLPI primitives are defined in <sys/dlpi.h>. Refer
to dlpi(7P) for more information.
You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (ppa). The ppa ID is interpreted as an unsigned long data type and indicates
the corresponding device instance (unit) number. An error (DL_ERROR_ACK) is returned by the
driver if the ppa field value does not correspond to a valid device instance number for the
system. The device is initialized on first attach and de-initialized (stopped) upon last detach.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are:
■ Maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h> ).
■ Minimum SDU is 0.
■ dlsap address length is 8.
■ MAC type is DL_ETHER.
■ sap length value is –2, meaning the physical address component is followed immediately
by a 2 byte sap component within the DLSAP address.

Device and Network Interfaces 295

ge(7D)

■ Service mode is DL_CLDLS.

■ Quality of service (QOS) is not supported; accordingly, QOS fields are 0.
■ Provider style is DL_STYLE2.
■ Version is DL_VERSION_2.
■ Broadcast address value is Ethernet/IEEE broadcast address (0xFFFFFF).

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular Service
Access Pointer (SAP) with the stream. The ge driver interprets the sap field within the
DL_BIND_REQ as an Ethernet type; accordingly, valid values for the sap field are in the
[0-0xFFFF] range. Only one Ethernet type can be bound to the stream at any time.

If you select a sap with a value of 0, the receiver will be in 802.3 mode. All frames received
from the media with a type field in the range [0-1500] are assumed to be 802.3 frames and are
routed up all open streams bound to sap value 0. If more than one stream is in 802.3 mode, the
frame will be duplicated and routed up multiple streams as DL_UNITDATA_IND messages.

In transmission, the driver checks the sap field of the DL_BIND_REQ to determine if the sap
value is 0 and the destination type field is in the range [0-1500]. If either is true, the driver
computes the length of the message, not including initial M_PROTO mblk (message block), of all
subsequent DL_UNITDATA_REQ messages and transmits 802.3 frames of that value in the MAC
frame header length field.

The ge driver DLSAP address format consists of the 6 byte physical (Ethernet) address
component followed immediately by the 2 byte sap (type) component producing an 8 byte
DLSAP address. Applications should not hard code to this particular implementation-specific
DLSAP address format, but use information returned in the DL_INFO_ACK primitive to
compose and decompose DLSAP addresses. The sap length, full DLSAP length and sap
physical ordering are included within the DL_INFO_ACK. The physical address length can be
computed by subtracting the sap length from the full DLSAP address length or by issuing the
DL_PHYS_ADDR_REQ to obtain the current physical address associated with the stream.

Once in the DL_BOUND state, you may transmit frames on the Ethernet by sending
DL_UNITDATA_REQ messages to the ge driver. The ge driver will route received Ethernet frames
up all open and bound streams having a sap which matches the Ethernet type as
DL_UNITDATA_IND messages. Received Ethernet frames are duplicated and routed up multiple
open streams if necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.

In addition to the mandatory connectionless DLPI message set, the driver additionally
supports ge primitives.

296 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 ge(7D)

ge Primitives The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable/disable reception of

individual multicast group addresses. A set of multicast addresses may be iteratively created
and modified on a per-stream basis using these primitives. The DL_ENABMULTI_REQ and
DL_DISABMULTI_REQ primitives are accepted by the driver in any state following
DL_ATTACHED.

With the DL_PROMISC_PHYS flag set in the dl_level field, the DL_PROMISCON_REQ and
DL_PROMISCOFF_REQ primitives enable/disable reception of all promiscuous mode frames on
the media including frames generated by the local host. When used with the DL_PROMISC_SAP
flag set, this enables/disables reception of all sap (Ethernet type) values. When used with the
DL_PROMISC_MULTI flag set, this enables/disables reception of all multicast group addresses.
The effect of each is always on a per-stream basis and independent of the other sap and
physical level configurations on the stream or other streams.

The DL_PHYS_ADDR_REQ primitive returns the six octet Ethernet address currently associated
(attached) to the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the 6 octet Ethernet address currently

associated (attached) to the stream. The credentials of the process which originally opened the
stream must be superuser or EPERM is returned in the DL_ERROR_ACK. The
DL_SET_PHYS_ADDR_REQ primitive is destructive and affects all other current and future
streams attached to this device. A M_ERROR is sent up all other streams attached to the device
when DL_SET_PHYS_ADDR_REQ is successful on the stream. Once changed, all streams
subsequently opened and attached to the device will obtain the new physical address. Once
changed, the physical address will remain until DL_SET_PHYS_ADDR_REQ is used to change the
physical address again or the system is rebooted, whichever comes first.

ge DRIVER By default, the ge driver performs auto-negotiation to select the mode and flow control
capabilities of the link. The link can be in one of the following modes:
■ 1000 Mbps, full-duplex
■ 1000 Mbps, half-duplex
■ Symmetric pause
■ Asymmetric pause

Speeds and modes are described in the 1000Base-TX standard.

The auto-negotation protocol automatically selects:

■ Operation mode (half-duplex or full-duplex)
■ Flow control capability (Symmetric and/or Asymmetric)

The auto–negotiation protocol:

■ Gets all the modes of operation supported by the link partner.
■ Advertises its capabilities to the link partner.

Device and Network Interfaces 297

ge(7D)

■ Selects the highest common denominator mode of operation based on the priorities.

When by default, auto-negotiation is used to bring up the link and select the common mode of
operation with the link partner, the GEM hardware is capable of all of the operating modes
listed above. The PCS also supports forced-mode of operation under which the driver can
select the mode of operation and flow control capabilities using the ndd utility.

The GEM device also supports programmable Inter-Packet Gap (IPG) parameters ipg1 and
ipg2. By default, the driver sets ipg1 to 8 byte-times and ipg2 to 4 byte-times, (the standard
values.) You may want to alter these values from the standard 1000 Mpbs IPG set to 0.096
microseconds.

ge Parameter List You can use the ge driver to set and get parameters for the GEM device. The parameter list
includes current transceiver status, current link status, inter-packet gap, PCS capabilities and
link partner capabilities.

The PCS has two set of capabilities. One set reflects the capabilities of the hardware and are
read-only. The second set are read/write and are used in speed selection and reflect the values
you choose. At boot time, both sets will be the same. The link partner capabilities are read only
and cannot be modified.
Files /dev/ge ge special character device
/kernel/drv/ge.conf System wide default device driver properties

See Also ndd(1M), netstat(1M), driver.conf(4), dlpi(7P), hme(7D), qfe(7d)

298 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 gld(7D)

Name gld – Generic LAN Driver

Synopsis #include <sys/stropts.h>
#include <sys/stream.h>
#include <sys/dlpi.h>
#include <sys/gld.h>

Interface Level Solaris architecture specific (Solaris DDI).

Description GLD is a multi-threaded, clonable, loadable kernel module providing support for Solaris local
area network (LAN) device drivers. LAN drivers in Solaris are STREAMS-based drivers that
use the Data Link Provider Interface (DLPI) to communicate with network protocol stacks.
These protocol stacks use the network drivers to send and receive packets on a local area
network. A network device driver must implement and adhere to the requirements imposed
by the DDI/DKI specification, STREAMS specification, DLPI specification, and
programmatic interface of the device itself.

GLD implements most STREAMS and DLPI functionality required of a Solaris LAN driver.
Several Solaris network drivers are implemented using GLD.

A Solaris network driver implemented using GLD comprises two distinct parts: a generic
component that deals with STREAMS and DLPI interfaces, and a device-specific component
that deals with the particular hardware device. The device-specific module indicates its
dependency on the GLD module and registers itself with GLD from within the driver's
attach(9E) function. Once it is successfully loaded, the driver is DLPI-compliant. The
device-specific part of the driver calls gld(9F) functions when it receives data or needs some
service from GLD. GLD makes calls into the gld(9E) entry points of the device-specific driver
through pointers provided to GLD by the device-specific driver when it registered itself with
GLD. The gld_mac_info(9S) structure is the main data interface between GLD and the
device-specific driver.

The GLD facility currently supports devices of type DL_ETHER, DL_TPR, and DL_FDDI. GLD
drivers are expected to process fully-formed MAC-layer packets and should not perform
logical link control (LLC) handling.

Note – Support for the DL_TPR and DL_FDDI media types in GLD is obsolete and may be
removed in a future release of Solaris.

In some cases, it may be necessary or desirable to implement a full DLPI-compliant driver

without using the GLD facility. This is true for devices that are not IEEE 802-style LAN
devices, or where a device type or DLPI service not supported by GLD is required.

Device Naming The name of the device-specific driver module must adhere to the naming constraints
Constraints outlined in the NOTES section of dlpi(7P).

Device and Network Interfaces 299

gld(7D)

Type DL_ETHER: For devices designated type DL_ETHER, GLD provides support for both Ethernet V2 and ISO
Ethernet V2 and ISO 8802-3 (IEEE 802.3) packet processing. Ethernet V2 enables a data link service user to access
8802-3 (IEEE 802.3)
and use any of a variety of conforming data link service providers without special knowledge
of the provider's protocol. A service access point (SAP) is the point through which the user
communicates with the service provider.

SAP 0 denotes that the user wishes to use 802.3 mode. In transmission, GLD checks the
destination SAP value of the DL_UNITDATA_REQ and the SAP value to which the stream is
bound. If both are 0, the GLD computes the length of the packet payload and transmits 802.3
frames having that length in the MAC frame header type field. Such lengths will never exceed
1500.

All frames received from the media that have a type field in the range [0-1500] are assumed to
be 802.3 frames and are routed up all open streams that are in 802.3 mode, (those streams
bound to a SAP value in of 0. If more than one stream is in 802.3 mode, the incoming frame is
duplicated and routed up each such stream.

Streams bound to a SAP value of 1536 or greater receive incoming packets whose Ethernet
MAC header type value exactly matches the value of the SAP to which the stream is bound.
SAP values in the range [1-1535] are undefined and should not be used.

Types DL_TPR and Note – Support for the DL_TPR and DL_FDDI media types in GLD is obsolete and may be
DL_FDDI: SNAP removed in a future release of Solaris.
Processing
For media types DL_TPR and DL_FDDI, GLD implements minimal SNAP (Sub-Net Access
Protocol) processing for SAP values of 1536 or greater. A SAP value of 0 denotes that the user
wishes to use LLC mode. SAP values in the range [1-1535] have undefined semantics and
should not be used.

SNAP headers are carried under LLC headers with destination SAP 0xAA. For outgoing
packets with SAP values greater than 1535, GLD creates an LLC+SNAP header that always
looks like:

‘‘AA AA 03 00 00 00 XX XX''

where ‘‘XX XX'' represents the 16-bit SAP, corresponding to the Ethernet V2 style ‘‘type.'' This
is the only class of SNAP header that is processed - non-zero OUI fields, and LLC control fields
other than 03 are considered to be LLC packets with SAP 0xAA.

A DL_UNITDATA_REQ message specifying a destination SAP value of 0, passed down a

stream bound to SAP 0, is assumed to contain an LLC packet and will not undergo SNAP
processing.

Incoming packets are examined to ascertain whether they fall into the format specified above.
Packets that do will be passed to streams bound to the packet's 16-bit SNAP type, as well as
being passed to any stream in LLC mode (those bound to a SAP value of 0).

300 man pages section 7: Device and Network Interfaces • Last Revised 10 Nov 2005
 gld(7D)

Type DL_TPR: Source Note – Support for the DL_TPR media type in GLD is obsolete and may be removed in a future
Routing release of Solaris.

For type DL_TPR devices, GLD implements minimal support for source routing. Source
routing enables a station that is sending a packet across a bridged medium to specify (in the
packet MAC header) routing information that determines the route that the packet will take
through the network.

Functionally, the source routing support provided by GLD learns routes, solicits and responds
to requests for information about possible multiple routes and selects among the multiple
routes that are available. It adds Routing Information Fields to the MAC headers of outgoing
packets and recognizes such fields in incoming packets.

GLD's source routing support does not implement the full Route Determination Entity (RDE)
specified in ISO 8802-2 (IEEE 802.2) Section 9. However, it is designed to interoperate with
any such implementations that may exist in the same (or a bridged) network.

Style 1 and 2 Providers GLD implements both Style 1 and Style 2 providers. A physical point of attachment (PPA) is
the point at which a system attaches itself to a physical communication medium. All
communication on that physical medium funnels through the PPA. The Style 1 provider
attaches the stream to a particular PPA based on the major/minor device that has been
opened. The Style 2 provider requires the DLS user to explicitly identify the desired PPA using
DL_ATTACH_REQ. In this case, open(9E) creates a stream between the user and GLD and
DL_ATTACH_REQ subsequently associates a particular PPA with that stream. Style 2 is denoted
by a minor number of zero. If a device node whose minor number is not zero is opened, Style 1
is indicated and the associated PPA is the minor number minus 1. In both Style 1 and Style 2
opens, the device is cloned.

Implemented DLPI GLD implements the following DLPI primitives:

Primitives
The DL_INFO_REQ primitive requests information about the DLPI stream. The message
consists of one M_PROTO message block. GLD returns device-dependent values in the
DL_INFO_ACK response to this request, based on information the GLD-based driver specified
in the gld_mac_info(9S) structure passed to gld_register(). However GLD returns the
following values on behalf of all GLD-based drivers:
■ The version is DL_VERSION_2.
■ The service mode is DL_CLDLS — GLD implements connectionless-mode service.
■ The provider style is DL_STYLE1 or DL_STYLE2, depending on how the stream was opened.

The DL_ATTACH_REQ primitive is called to associate a PPA with a stream. This request is needed
for Style 2 DLS providers to identify the physical medium over which the communication will
transpire. Upon completion, the state changes from DL_UNATTACHED to DL_UNBOUND. The

Device and Network Interfaces 301

gld(7D)

message consists of one M_PROTO message block. This request may not be issued when using
the driver in Style 1 mode; streams opened using Style 1 are already attached to a PPA by the
time the open completes.

The DL_DETACH_REQ primitive requests to detach the PPA from the stream. This is only
allowed if the stream was opened using Style 2.

The DL_BIND_REQ and DL_UNBIND_REQ primitives bind and unbind a DLSAP to the stream.
The PPA associated with each stream will have been initialized upon completion of the
processing of the DL_BIND_REQ. Multiple streams may be bound to the same SAP; each such
stream receives a copy of any packets received for that SAP.

The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable and disable reception of

individual multicast group addresses. A set of multicast addresses may be iteratively created
and modified on a per-stream basis using these primitives. The stream must be attached to a
PPA for these primitives to be accepted.

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives enable and disable promiscuous

mode on a per-stream basis, either at a physical level or at the SAP level. The DL Provider will
route all received messages on the media to the DLS user until either a DL_DETACH_REQ or a
DL_PROMISCOFF_REQ is received or the stream is closed. Physical level promiscuous mode may
be specified for all packets on the medium or for multicast packets only. The stream must be
attached to a PPA for these primitives to be accepted.

The DL_UNITDATA_REQ primitive is used to send data in a connectionless transfer. Because this
is an unacknowledged service, there is no guarantee of delivery. The message consists of one
M_PROTO message block followed by one or more M_DATA blocks containing at least one byte of
data.

The DL_UNITDATA_IND type is used when a packet is received and is to be passed upstream.
The packet is put into an M_PROTO message with the primitive set to DL_UNITDATA_IND.

The DL_PHYS_ADDR_REQ primitive returns the MAC address currently associated with the PPA
attached to the stream, in the DL_PHYS_ADDR_ACK primitive. When using style 2, this primitive
is only valid following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the MAC address currently associated with
the PPA attached to the stream. This primitive affects all other current and future streams
attached to this device. Once changed, all streams currently or subsequently opened and
attached to this device will obtain this new physical address. The new physical address will
remain in effect until this primitive is used to change the physical address again or the driver is
reloaded.

302 man pages section 7: Device and Network Interfaces • Last Revised 10 Nov 2005
 gld(7D)

The DL_GET_STATISTICS_REQ primitive requests a DL_GET_STATISTICS_ACK response

containing statistics information associated with the PPA attached to the stream. Style 2
streams must be attached to a particular PPA using DL_ATTACH_REQ before this primitive will
be successful.

GLD supports the DL_NOTE_LINK_UP, DL_NOTE_LINK_DOWN and

DL_NOTE_SPEED notifications using the DL_NOTIFY_IND primitive. See dlpi(7P).

Implemented ioctl GLD implements the DLIOCRAW ioctl described in dlpi(7P). For any other ioctl command,
Functions GLD passes it to the device-specific driver's gldm_ioctl() function as described in gld(9E).

Requirements on GLD GLD-based drivers must include the header file <sys/gld.h>.
Drivers
GLD-based drivers must also specify a link dependency on "misc/gld". (See the -N option in
ld(1)).

GLD implements the open(9E) and close(9E) functions and the required STREAMS put(9E)
and srv(9E) functions on behalf of the device-specific driver. GLD also implements the
getinfo(9E) function for the driver.

The mi_idname element of the module_info(9S) structure is a string specifying the name of
the driver. This must exactly match the name of the driver module as it exists in the file system.

The read-side qinit(9S) structure should specify the following elements as shown below:
qi_putp NULL
qi_srvp gld_rsrv
qi_qopen gld_open
qi_qclose gld_close

The write-side qinit(9S) structure should specify the following elements as shown below:
qi_putp gld_wput
qi_srvp gld_wsrv
qi_qopen NULL
qi_qclose NULL

The devo_getinfo element of the dev_ops(9S) structure should specify gld_getinfo as the
getinfo(9E) routine.

The driver's attach(9E) function does all the work of associating the hardware-specific device
driver with the GLD facility and preparing the device and driver for use.

Device and Network Interfaces 303

gld(7D)

The attach(9E) function allocates a gld_mac_info(9S) (‘‘macinfo'') structure using

gld_mac_alloc(). The driver usually needs to save more information per device than is
defined in the macinfo structure; it should allocate the additional required data structure and
save a pointer to it in the gldm_private member of the gld_mac_info(9S) structure.

The attach(9E) routine must initialize the macinfo structure as described in

gld_mac_info(9S) and then call gld_register() to link the driver with the GLD module. The
driver should map registers if necessary and be fully initialized and prepared to accept
interrupts before calling gld_register(). The attach(9E) function should add interrupts but
not enable the device to generate them. The driver should reset the hardware before calling
gld_register() to ensure it is quiescent; the device must not be started or put into a state
where it may generate an interrupt before gld_register() is called. That will be done later
when GLD calls the driver's gldm_start() entry point described in gld(9E). Once
gld_register() succeeds, the gld(9E) entry points may be called by GLD at any time.

The attach(9E) routine should return DDI_SUCCESS if gld_register() succeeds. If

gld_register() fails, it returns DDI_FAILURE and the attach(9E) routine should deallocate
any resources it allocated before calling gld_register() and then also return DDI_FAILURE.
Under no circumstances should a failed macinfo structure be reused; it should be deallocated
using gld_mac_free().

The detach(9E) function should attempt to unregister the driver from GLD. This is done by
calling gld_unregister() described in gld(9F). The detach(9E) routine can get a pointer to
the needed gld_mac_info(9S) structure from the device's private data using
ddi_get_driver_private(9F). gld_unregister() checks certain conditions that could
require that the driver not be detached. If the checks fail, gld_unregister() returns
DDI_FAILURE, in which case the driver's detach(9E) routine must leave the device operational
and return DDI_FAILURE. If the checks succeed, gld_unregister() ensures that the device
interrupts are stopped, calling the driver's gldm_stop() routine if necessary, unlinks the
driver from the GLD framework, and returns DDI_SUCCESS. In this case, the detach(9E)
routine should remove interrupts, deallocate any data structures allocated in the attach(9E)
routine, using gld_mac_free() to deallocate the macinfo structure, and return DDI_SUCCESS.
It is important to remove the interrupt before calling gld_mac_free().

Network Statistics Solaris network drivers must implement statistics variables. GLD itself tallies some network
statistics, but other statistics must be counted by each GLD-based driver. GLD provides
support for GLD-based drivers to report a standard set of network driver statistics. Statistics
are reported by GLD using the kstat(7D) and kstat(9S) mechanism. The
DL_GET_STATISTICS_REQ DLPI command may also be used to retrieve the current statistics
counters. All statistics are maintained as unsigned, and all are 32 bits unless otherwise noted.

GLD maintains and reports the following statistics.

rbytes64 Total bytes successfully received on the interface (64 bits).
rbytes Total bytes successfully received on the interface.

304 man pages section 7: Device and Network Interfaces • Last Revised 10 Nov 2005
 gld(7D)

obytes64 Total bytes requested to be transmitted on the interface (64 bits).

obytes Total bytes requested to be transmitted on the interface.
ipackets64 Total packets successfully received on the interface (64 bits).
ipackets Total packets successfully received on the interface.
opackets64 Total packets requested to be transmitted on the interface (64 bits).
opackets Total packets requested to be transmitted on the interface.
multircv Multicast packets successfully received, including group and functional
addresses (long).
multixmt Multicast packets requested to be transmitted, including group and
functional addresses (long).
brdcstrcv Broadcast packets successfully received (long).
brdcstxmt Broadcast packets requested to be transmitted (long).
unknowns Valid received packets not accepted by any stream (long).
noxmtbuf Packets discarded on output because transmit buffer was busy, or no buffer
could be allocated for transmit (long).
blocked Times a received packet could not be put up a stream because the queue was
flow controlled (long).
xmtretry Times transmit was retried after having been delayed due to lack of resources
(long).
promisc Current ‘‘promiscuous'' state of the interface (string).

The device dependent driver counts the following statistics, keeping track of them in a private
per-instance structure. When GLD is asked to report statistics, it calls the driver's
gldm_get_stats() entry point, as described in gld(9E), to update the device-specific statistics
in the gld_stats(9S) structure. GLD then reports the updated statistics using the named
statistics variables below.
ifspeed Current estimated bandwidth of the interface in bits per second (64 bits).
media Current media type in use by the device (string).
intr Times interrupt handler was called and claimed the interrupt (long).
norcvbuf Times a valid incoming packet was known to have been discarded because no
buffer could be allocated for receive (long).
ierrors Total packets received that couldn't be processed because they contained errors
(long).
oerrors Total packets that weren't successfully transmitted because of errors (long).

Device and Network Interfaces 305

gld(7D)

missed Packets known to have been dropped by the hardware on receive (long).
uflo Times FIFO underflowed on transmit (long).
oflo Times receiver overflowed during receive (long).

The following group of statistics applies to networks of type DL_ETHER; these are maintained
by device-specific drivers of that type, as above.
align_errors Packets received with framing errors (not an integral number of
octets) (long).
fcs_errors Packets received with CRC errors (long).
duplex Current duplex mode of the interface (string).
carrier_errors Times carrier was lost or never detected on a transmission attempt
(long).
collisions Ethernet collisions during transmit (long).
ex_collisions Frames where excess collisions occurred on transmit, causing
transmit failure (long).
tx_late_collisions Times a transmit collision occurred late (after 512 bit times) (long).
defer_xmts Packets without collisions where first transmit attempt was delayed
because the medium was busy (long).
first_collisions Packets successfully transmitted with exactly one collision.
multi_collisions Packets successfully transmitted with multiple collisions.
sqe_errors Times SQE test error was reported.
macxmt_errors Packets encountering transmit MAC failures, except carrier and
collision failures.
macrcv_errors Packets received with MAC errors, except align, fcs, and toolong
errors.
toolong_errors Packets received larger than the maximum permitted length.
runt_errors Packets received smaller than the minimum permitted length
(long).

The following group of statistics applies to networks of type DL_TPR; these are maintained by
device-specific drivers of that type, as above.
line_errors Packets received with non-data bits or FCS errors.
burst_errors Times an absence of transitions for five half-bit timers was
detected.

306 man pages section 7: Device and Network Interfaces • Last Revised 10 Nov 2005
 gld(7D)

signal_losses Times loss of signal condition on the ring was detected.

ace_errors Times an AMP or SMP frame in which A is equal to C is equal to
0, was followed by another such SMP frame without an
intervening AMP frame.
internal_errors Times the station recognized an internal error.
lost_frame_errors Times the TRR timer expired during transmit.
frame_copied_errors Times a frame addressed to this station was received with the FS
field A bit set to 1.
token_errors Times the station acting as the active monitor recognized an error
condition that needed a token transmitted.
freq_errors Times the frequency of the incoming signal differed from the
expected frequency.

The following group of statistics applies to networks of type DL_FDDI; these are maintained by
device-specific drivers of that type, as above.
mac_errors Frames detected in error by this MAC that had not been detected in
error by another MAC.
mac_lost_errors Frames received with format errors such that the frame was stripped.
mac_tokens Number of tokens received (total of non-restricted and restricted).
mac_tvx_expired Number of times that TVX has expired.
mac_late Number of TRT expirations since this MAC was reset or a token was
received.
mac_ring_ops Number of times the ring has entered the ‘‘Ring_Operational'' state
from the ‘‘Ring Not Operational'' state.
Files /kernel/misc/gld loadable kernel module

See Also ld(1), kstat(7D), dlpi(7P), attach(9E), gld(9E), open(9E), gld(9F), gld_mac_info(9S),
gld_stats(9S), kstat(9S)

Writing Device Drivers

Warnings Contrary to the DLPI specification, GLD returns the device's correct address length and
broadcast address in DL_INFO_ACK even before the stream has been attached to a PPA.

Promiscuous mode may only be entered by streams that are attached to a PPA.

The physical address of a PPA may be changed by the superuser while other streams are bound
to the same PPA.

Device and Network Interfaces 307

glm(7D)

Name glm – GLM SCSI Host Bus Adapter Driver

Synopsis scsi@unit-address

Description The glm Host Bus Adapter driver is a SCSA compliant nexus driver that supports the LSI
53c810, LSI 53c875, LSI 53c876, LSI 53C896 and LSI 53C1010 SCSI chips

It supports the standard functions provided by the SCSA interface. That is, it supports tagged
and untagged queuing, Narrow/Wide/Fast/Ultra SCSI/Ultra SCSI 2/Ultra SCSI 3, and auto
request sense, but it does not support linked commands.

Driver Configuration Configure the glm driver by defining properties in glm.conf. These properties override the
global SCSI settings. glm supports these properties which can be modified by the user:
scsi-options, target<n>-scsi-options, scsi-reset-delay, scsi-tag-age-limit,
scsi-watchdog-tick, and scsi-initiator-id.

target<n>-scsi-options overrides the scsi-options property value for target<n>. <n>

can vary from decimal 0 to 15. glm supports these scsi-options: SCSI_OPTIONS_DR,
SCSI_OPTIONS_SYNC, SCSI_OPTIONS_TAG, SCSI_OPTIONS_FAST, SCSI_OPTIONS_WIDE,
SCSI_OPTIONS_FAST20, SCSI_OPTIONS_FAST40 and SCSI_OPTIONS_FAST80.

After periodic interval scsi-watchdog-tick, the glm driver searches through all current and
disconnected commands for timeouts.

scsi-tag-age-limit is the number of times that the glm driver attempts to allocate a
particular tag ID that is currently in use after going through all tag IDs in a circular fashion.
After finding the same tag ID in use scsi-tag-age-limit times, no more commands will be
submitted to this target until all outstanding commands complete or timeout.

Refer to scsi_hba_attach(9F).

Examples EXAMPLE 1 Using the glm Configuration File

Create a file called /kernel/drv/glm.conf and add the following line:

scsi-options=0x78;

This disables tagged queuing, Fast/Ultra SCSI and wide mode for all glm instances.

The following example disables an option for one specific glm (refer to driver.conf(4) and
pci(4) for more details):
name="glm" parent="/pci@1f,4000"
unit-address="3"
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

308 man pages section 7: Device and Network Interfaces • Last Revised 15 Nov 2005
 glm(7D)

EXAMPLE 1 Using the glm Configuration File (Continued)

Note that the default initiator ID in OBP is 7 and that the change to ID 6 will occur at attach
time. It may be preferable to change the initiator ID in OBP.

The example above sets scsi-options for target 1 to 0x58 and all other targets on this SCSI
bus to 0x178.

The physical pathname of the parent can be determined using the /devices tree or following
the link of the logical device name:
# ls -l /dev/rdsk/c0t0d0s0
lrwxrwxrwx 1 root root 45 May 16 10:08 /dev/rdsk/c0t0d0s0 ->
. . / . . /devices/pci@1f,4000/scsi@3/sd@0,0:a,raw

In this case, like the example above, the parent is /pci@1f,4000 and the unit-address is the
number bound to the scsi@3 node.

To set scsi-options more specifically per target:

target1-scsi-options=0x78;
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;
scsi-options=0x3f8;

The above sets scsi-options for target 1 to 0x78 and for all other targets on this SCSI bus to
0x3f8 except for one specific disk type which will have scsi-options set to 0x58.

scsi-options specified per target ID have the highest precedence, followed by scsi-options
per device type. Global scsi-options (for all glm instances) per bus have the lowest
precedence.

The system needs to be rebooted before the specified scsi-options take effect.

Driver Capabilities The target driver needs to set capabilities in the glm driver in order to enable some driver
features. The target driver can query and modify these capabilities: synchronous,
tagged-qing, wide-xfer, auto-rqsense, qfull-retries, qfull-retry-interval. All other
capabilities can only be queried.

By default, tagged-qing, auto-rqsense, and wide-xfer capabilities are disabled, while

disconnect, synchronous, and untagged-qing are enabled. These capabilities can only have
binary values (0 or 1). The default value for qfull-retries is 10 and the default value for
qfull-retry-interval is 100. The qfull-retries capability is a uchar_t (0 to 255) while
qfull-retry-interval is a ushort_t (0 to 65535).

The target driver needs to enable tagged-qing and wide-xfer explicitly. The untagged-qing
capability is always enabled and its value cannot be modified.

Device and Network Interfaces 309

glm(7D)

Whenever there is a conflict between the value of scsi-options and a capability, the value set
in scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call.

Refer to scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.

Files /kernel/drv/glm 32–bit ELF kernel module (x86).
/kernel/drv/amd64/glm 64–bit ELF kernel module (x86).
/kernel/drv/sparcv9/glm 64–bit ELF kernel module (SPARC).
/kernel/drv/glm.conf Optional configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Limited to PCI-based systems with LSI 53c810, LSI 53c875, LSI
53c876, LSI 53c896 and LSI 53c1010 SCSI I/O processors

See Also prtconf(1M), driver.conf(4), pci(4), attributes(5), scsi_abort(9F),

scsi_hba_attach(9F), scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F),
scsi_sync_pkt(9F), scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S),
scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

ANSI Small Computer System Interface-2 (SCSI-2),

LSI Logi Inc (formerly Symbios Logic Inc.):

■ SYM53c810 PCI-SCSI I/O processor with Narrow operation
■ SYM53c875 PCI-SCSI I/O Processor With Fast-20
■ SYM53c876 PCI-SCSI I/O processor Dual channel Fast-20
■ SYM53c896 PCI-SCSI I/O processor Dual channel Fast-40
■ SYM53c1010 PCI-SCSI I/O processor Dual Channel Fast-80

Diagnostics The messages described below are some that may appear on the system console, as well as
being logged.
Device is using a hilevel intr
The device was configured with an interrupt level that cannot be used with this glm driver.
Check the PCI device.
map setup failed
Driver was unable to map device registers; check for bad hardware. Driver did not attach to
device; SCSI devices will be inaccessible.

310 man pages section 7: Device and Network Interfaces • Last Revised 15 Nov 2005
 glm(7D)

glm_script_alloc failed
The driver was unable to load the SCRIPTS for the SCSI processor, check for bad hardware.
Driver did not attach to device; SCSI devices will be inaccessible.
cannot map configuration space.
The driver was unable to map in the configuration registers. Check for bad hardware. SCSI
devices will be inaccessible.
attach failed
The driver was unable to attach; usually preceded by another warning that indicates why
attach failed. These can be considered hardware failures.
SCSI bus DATA IN phase parity error
The driver detected parity errors on the SCSI bus.
SCSI bus MESSAGE IN phase parity error
The driver detected parity errors on the SCSI bus.
SCSI bus STATUS phase parity error
The driver detected parity errors on the SCSI bus.
Unexpected bus free
Target disconnected from the bus without notice. Check for bad hardware.
Disconnected command timeout for Target <id>.<lun>
A timeout occurred while target id/lun was disconnected. This is usually a target firmware
problem. For tagged queuing targets, <n> commands were outstanding when the timeout
was detected.
Disconnected tagged cmd(s) (<n>) timeout for Target <id>.<lun>
A timeout occurred while target id/lun was disconnected. This is usually a target firmware
problem. For tagged queuing targets, <n> commands were outstanding when the timeout
was detected.
Connected command timeout for Target <id>.<lun>
This is usually a SCSI bus problem. Check cables and termination.
Target <id> reducing sync. transfer rate
A data transfer hang or DATA-IN phase parity error was detected. The driver attempts to
eliminate this problem by reducing the data transfer rate.
Target <id> reverting to async. mode
A second data transfer hang was detected for this target. The driver attempts to eliminate
this problem by reducing the data transfer rate.
Target <id> disabled wide SCSI mode
A second data phase hang was detected for this target. The driver attempts to eliminate this
problem by disabling wide SCSI mode.

Device and Network Interfaces 311

glm(7D)

auto request sense failed

An attempt to start an auto request packet failed. Another auto request packet may already
be in transport.
invalid reselection (<id>.<lun>)
A reselection failed; target accepted abort or reset, but still tries to reconnect. Check for
bad hardware.
invalid intcode
The SCRIPTS processor generated an invalid SCRIPTS interrupt. Check for bad hardware.

Notes The x4422a card uses an OBP (forth) firmware and is incompatible with x86 BIOS. As a result,
the x4422a cannot be used as a boot device on x86.

The glm driver supports the following LSI chips:

■ LSI 53C810, which supports Narrow, Fast SCSI mode. The maximum SCSI bandwidth is
10 MB/sec.
■ LSI 53C875, which supports Wide, Fast, and Ultra SCSI mode. The maximum SCSI
bandwidth is 40 MB/sec.
■ LSI 53C896, which supports Wide, Fast and Ultra SCSI 2 mode. The maximum LVD SCSI
bandwidth is 80 MB/sec.
■ LSI 53c1010, which supports wide, Fast and Ultra SCSI 3 mode. The maximum LVD SCSI
bandwidth is 160 MB/sec.

The glm driver exports properties indicating per target the negotiated transfer speed
(target<n>-sync-speed), whether wide bus is supported (target<n>-wide), for that
particular target (target<n>-scsi-options), and whether tagged queuing has been enabled
(target<n>-TQ). The sync-speed property value is the data transfer rate in KB/sec. The
target<n>-TQ and the target<n>-wide property have value 1 to indicate that the
corresponding capability is enabled, or 0 to indicate that the capability is disabled for that
target. Refer to prtconf(1M) (verbose option) for viewing the glm properties.
scsi, instance #0
Driver properties:
name <target6-TQ> length <4>
value <0x00000000>.
name <target6-wide> length <4>
value <0x00000000>.
name <target6-sync-speed> length <4>
value <0x00002710>.
name <target1-TQ> length <4>
value <0x00000001>.
name <target1-wide> length <4>
value <0x00000000>.
name <target1-sync-speed> length <4>
value <0x00002710>.

312 man pages section 7: Device and Network Interfaces • Last Revised 15 Nov 2005
 glm(7D)

name <target0-TQ> length <4>

value <0x00000001>.
name <target0-wide> length <4>
value <0x00000001>.
name <target0-sync-speed> length <4>
value <0x00009c40>.
name <scsi-options> length <4>
value <0x000007f8>.
name <scsi-watchdog-tick> length <4>
value <0x0000000a>.
name <scsi-tag-age-limit> length <4>
value <0x00000002>.
name <scsi-reset-delay> length <4>
value <0x00000bb8>.
name <latency-timer> length <4>
value <0x00000088>.
name <cache-line-size> length <4>
value <0x00000010>.

Device and Network Interfaces 313

gpio_87317(7D)

Name gpio_87317 – General purpose I/O driver for SuperIO

Description The gpio_87317 driver is the general purpose I/O driver for the National Semiconductor
SuperIO (PC87317) chipset. It supports remote system controller (RSC) administration via an
interface to the SuperIO's general purpose I/O bits.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Limited to SPARC systems with SuperIO

Availability SUNWcar

Interface Stability Unstable

See Also PC87317VUL/PC97317VUL SuperI/O Data Sheet — National Semiconductor

314 man pages section 7: Device and Network Interfaces • Last Revised 12 April 2000
 grbeep(7d)

Name grbeep – Platform-dependent beep driver for SMBus-based hardware

Synopsis beep@unit-address

Description The grbeep driver generates beeps on platforms (including Sun Blade 100, 150, 1500, 2500)
that use SMBbus-based registers and USB keyboards. When the KIOCCMD ioctl is issued to the
USB keyboard module (see usbkbm(7M)) with command KBD_CMD_BELL/KBD_CMD_NOBELL,
usbkbm(7M) passes the request to the grbeep driver to turn the beep on and off, respectively.
Files /platform/sun4u/kernel/drv/sparcv9/grbeep 64–bit ELF kernel driver

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SMBus-based SPARC

Availability SUNWcarx.u

See Also kbd(1), attributes(5), bbc_beep(7D), kb(7M), usbkbm(7M)

Writing Device Drivers

Diagnostics None

Device and Network Interfaces 315

hci1394(7D)

Name hci1394 – 1394 OpenHCI host controller driver

Synopsis firewire@unit-address

Description The hci1394 host controller driver is an IEEE 1394 compliant nexus driver that supports the
1394 Open Host Controller Interface Specification 1.0, an industry standard developed by Sun,
Apple, Compaq, Intel, Microsoft, National Semconductor, and Texas Instruments. The
hci1394 driver supports asynchronous transfers, isochronous transfers, and bus reset
management. The hci1394 driver also supports the nexus device control interface.
Files /kernel/drv/sparcv9/hci1394 64–bit SPARC ELF kernel module
/kernel/drv/hci1394 32-bit x86 ELF kernel module
/kernel/drv/amd64/hci1394 64-bit x86 ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNW1394x

Interface Stability Unstable

See Also attributes(5), ieee1394(7D)

IEEE 1394 - IEEE Standard for a High Performance Serial Bus

1394 Open Host Controller Interface Specification 1.0

316 man pages section 7: Device and Network Interfaces • Last Revised 30 August 2005
 hdio(7I)

Name hdio – SMD and IPI disk control operations

Synopsis #include <sys/hdio.h>

Description Note – The SMC and IPI drivers have been discontinued. dkio(7I) is now the preferred method
for retrieving disk information.

The SMD and IPI disk drivers supplied with this release support a set of ioctl(2) requests for
diagnostics and bad sector information. Basic to these ioctl() requests are the definitions in
<sys/hdio.h>.
ioctls HDKIOCGTYPE The argument is a pointer to a hdk_type structure (described below). This
ioctl() gets specific information from the hard disk.
HDKIOCSTYPE The argument is a pointer to a hdk_type structure (described below). This
ioctl() sets specific information about the hard disk.
/*
* Used for drive info
*/
struct hdk_type {
ushort_t hdkt_hsect; /* hard sector count (read only) */
ushort_t hdkt_promrev; /* prom revision (read only) */
uchar_t hdkt_drtype; /* drive type (ctlr specific) */
uchar_t hdkt_drstat; /* drive status (ctlr specific, ro) */
};
HDKIOCGBAD The argument is a pointer to a hdk_badmap structure (described below). This
ioctl() is used to get the bad sector map from the disk.
HDKIOCSBAD The argument is a pointer to a hdk_badmap structure (described below). This
ioctl() is used to set the bad sector map on the disk.
/*
* Used for bad sector map
*/
struct hdk_badmap {
caddr_t hdkb_bufaddr; /* address of user’s map buffer */
};
HDKIOCGDIAG The argument is a pointer to a hdk_diag structure (described below). This
ioctl() gets the most recent command that failed along with the sector and
error number from the hard disk.
/*
* Used for disk diagnostics
*/
struct hdk_diag {
ushort_t hdkd_errcmd; /* most recent command in error */
daddr_t hdkd_errsect; /* most recent sector in error */
uchar_t hdkd_errno; /* most recent error number */

Device and Network Interfaces 317

hdio(7I)

uchar_t hdkd_severe; /* severity of most recent error */

};

See Also ioctl(2), dkio(7I)

318 man pages section 7: Device and Network Interfaces • Last Revised 13 Aug 2002
 heci(7D)

Name heci – Intel(R) AMT Manageability Interface Driver

Description The Intel AMT Manageability Interface driver allows applications to access the Intel Active
Management Technology (Intel AMT) FW by way of the host interface (as opposed to a
network interface).

The Intel AMT Manageability Interface driver is meant to be used by the Local Manageability
Service. When the Intel AMT machine is in Legacy Mode, the Intel AMT Manageability
Interface driver functions. Messages from the Intel AMT Manageability Interface driver are
sent to the system's log, that is, /var/log/messages.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWamt-lms

Interface Stability Volatile

See Also lms(1M), attributes(5), e1000g(7D), iwk(7D)

Device and Network Interfaces 319

hermon(7D)

Name hermon – ConnectX MT25408/MT25418/MT25428 InfiniBand (IB) Driver

Description The hermon driver is an IB Architecture-compliant implementation of an HCA, which

operates on the Mellanox MT25408, MT25418 and MT25428 InfiniBand ASSPs using host
memory for context storage rather than locally attached memory on the card. Cards based on
these ASSP's utilize the PCI-Express I/O bus. These ASSP's support the link and physical
layers of the InfiniBand specification while the ASSP and the driver support the transport
layer.

The hermon driver interfaces with the InfiniBand Transport Framework (IBTF) and provides
an implementation of the Channel Interfaces that are defined by that framework. It also
enables management applications and agents to access the IB fabric.
Files /kernel/drv/hermon
32–bit ELF kernel module. (x86)
/kernel/drv/amd64/hermon
64–bit ELF kernel module. (x86)
/kernel/drv/sparcv9/hermon
64-bit ELF Kernel Module. (SPARC)
/kernel/drv/hermon.conf
Driver configuration file.

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCIe-based systems

Availability SUNWhermon

See Also driver.conf(4), printers.conf(4), attributes(5)

Writing Device Drivers

Diagnostics In addition to being logged, the following messages may appear on the system console:
hermoni: driver attached for maintenance mode only.
There was a failure in the boot process of the hermon ASSP and the only function that can
be performed is to re-flash firmware on the ASSP. (Note that i represents the instance of the
hermon device number.)
hermoni: driver failed to attach
The ASSP could not boot into either operational (HCA) mode or into maintenance mode.
The device is inoperable. (Note that i represents the instance of the hermon device
number.)

320 man pages section 7: Device and Network Interfaces • Last Revised 31 Jul 2008
 hermon(7D)

Unexpected port number in port state change event.

A port state change event occurred, but the port number in the message does not exist on
this HCA. This message also indicates the port number that was in the port state changed.
Hermon driver successfully detached.
The driver has been removed from the system and the HCA is no longer available for
transfer operations.
hermoni: port m up.
A port up asynchronous event has occurred. (Note that i represents the instance of the
Hermon device number while "m" represents the port number on the Hermon device.
hermoni: port m down.
A port up asynchronous event has occurred. Similar to port up event.
hermon: <command name> command failed.
A internal firmware command failed to execute.

Device and Network Interfaces 321

hid(7D)

Name hid – Human interface device (HID) class driver

Synopsis keyboard@unit-address
mouse@unit-address
input@unit-address:consumer_control
#include <sys/hid.h>
int ioctl(int fildes, int command, ... /*arg*/);

Description The hid driver is a USBA (Solaris USB Architecture) compliant client driver that supports the
Human Interface Device Class (HID) 1.0 specification. The Human Interface Device (HID)
class encompasses devices controlled by humans to operate computer systems. Typical
examples of HID devices include keyboards, mice, trackballs, and joysticks. HID also covers
front-panel controls such as knobs, switches, and buttons. A USB device with multiple
interfaces may have one interface for audio and a HID interface to define the buttons that
control the audio.
The hid driver is general and primarily handles the USB functionality of the device and
generic HID functionality. For example, HID interfaces are required to have an interrupt pipe
for the device to send data packets, and the hid driver opens the pipe to the interrupt endpoint
and starts polling. The hid driver is also responsible for managing the device through the
default control pipe. In addition to being a USB client driver, the hid driver is also a
STREAMS driver so that modules may be pushed on top of it.
The HID specification is flexible, and HID devices dynamically describe their packets and
other parameters through a HID report descriptor. The HID parser is a misc module that
parses the HID report descriptor and creates a database of information about the device. The
hid driver queries the HID parser to find out the type and characteristics of the HID device.
The HID specification predefines packet formats for the boot protocol keyboard and mouse.
ioctls HIDIOCKMGDIRECT This ioctl should only be addressed to a USB keyboard or mouse
device. The hid driver maintains two streams for each USB
keyboard/mouse instance: an internal one for the use of the kernel and
an external one for the use of user applications. This ioctl returns the
information of which stream gets the input for the moment.

arg must point to a variable of int type. Upon return, 0 means the
internal stream gets the input, 1 means the external stream gets the
input.
HIDIOCKMSDIRECT This ioctl should only be addressed to a USB keyboard or mouse
device. The hid driver maintains two streams for each USB
keyboard/mouse instance: an internal one for the use of the kernel and
an external one for the use of user applications. This ioctl sets which
stream should get the input for the moment.

arg must point to a variable of int type. The argument 0 means the
internal stream gets the input, 1 means the external stream gets the
input.

322 man pages section 7: Device and Network Interfaces • Last Revised 21 Aug 2009
 hid(7D)

Files /kernel/drv/hid 32–bit x86 ELF kernel hid module

/kernel/drv/amd64/hid 64–bit x86 ELF kernel hid module
/kernel/drv/sparcv9/hid 64–bit SPARC ELF kernel hid module
/kernel/misc/hidparser 32–bit x86 ELF kernel hidparser module
/kernel/misc/amd64/hidparser 64–bit x86 ELF kernel hidparser module
/kernel/misc/sparcv9/hidparser 64–bit SPARC ELF kernel hidparser module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also cfgadm_usb(1M), attributes(5), usba(7D), virtualkm(7D)

Writing Device Drivers

STREAMS Programming Guide

Universal Serial Bus Specification 1.0 and 1.1

Device Class Definition for Human Interface Devices (HID) 1.1

System Administration Guide: Basic Administration

http://www.sun.com
Diagnostics hid_attach: Unsupported HID device. The device requires a protocol not supported by the
hid driver.
Parsing of hid descriptor failed. The HID report descriptor cannot be parsed
correctly. The device cannot be supported by the
hid driver.
Invalid report descriptor. The HID report descriptor is invalid. The device
cannot be supported by the hid driver.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><hid<instance number>): message...
hid_attach: Unsupported HID device. The device cannot be supported by this version of
the HID driver.

Device and Network Interfaces 323

hid(7D)

Parsing of HID descriptor failed. The device cannot be supported by this version of
the HID driver.
Invalid report descriptor. The device cannot be supported by this version of
the HID driver.

Notes The hid driver currently supports only keyboard, mouse and audio HID control devices.

Normally a mouse is not power managed and consquently, screen darkening can be undone
with a mouse movement. If power management of the mouse is required, add the following
line to hid.conf then reboot the system:
hid-mouse-pm-enable;

Modern mice that are power managed require a 'click' to wake up. Occasionally, this may
cause unexpected results.

324 man pages section 7: Device and Network Interfaces • Last Revised 21 Aug 2009
 hme(7D)

Name hme – SUNW,hme Fast-Ethernet device driver

Synopsis /dev/hme

Description The SUNW,hme Fast-Ethernet driver is a multi-threaded, loadable, clonable, STREAMS

hardware driver supporting the connectionless Data Link Provider Interface, dlpi(7P), over a
SUNW,hme Fast-Ethernet controller. The motherboard and add-in SBus SUNW,hme controllers
of several varieties are supported. Multiple SUNW,hme controllers installed within the system
are supported by the driver.

The hme driver provides basic support for the SUNW,hme hardware. It is used to handle the
SUNW,hme device. Functions include chip initialization, frame transit and receive, multicast
and promiscuous support, and error recovery and reporting. SUNW,hme The SUNW,hme device
provides 100Base-TX networking interfaces using SUN's FEPS ASIC and an Internal
Transceiver. The FEPS ASIC provides the Sbus interface and MAC functions and the Physical
layer functions are provided by the Internal Transceiver which connects to a RJ-45 connector.
In addition to the RJ-45 connector, an MII (Media Independent Interface) connector is also
provided on all SUNW,hme devices except the SunSwith SBus adapter board. The MII interface
is used to connect to an External Transceiver which may use any physical media (copper or
fiber) specified in the 100Base-TX standard. When an External Transceiver is connected to the
MII, the driver selects the External Transceiver and disables the Internal Transceiver.

The 100Base-TX standard specifies an “auto-negotiation” protocol to automatically select the

mode and speed of operation. The Internal transceiver is capable of doing “auto-negotiation”
with the remote-end of the link (Link Partner) and receives the capabilities of the remote end.
It selects the Highest Common Denominator mode of operation based on the priorities. It also
supports forced-mode of operation where the driver can select the mode of operation.

Application The cloning character-special device /dev/hme is used to access all SUNW,hme controllers
Programming installed within the system.
Interface
hme and DLPI The hme driver is a “style 2” Data Link Service provider. All M_PROTO and M_PCPROTO type
messages are interpreted as DLPI primitives. Valid DLPI primitives are defined in
<sys/dlpi.h>. Refer to dlpi(7P) for more information. An explicit DL_ATTACH_REQ message
by the user is required to associate the opened stream with a particular device (ppa). The ppa
ID is interpreted as an unsigned long data type and indicates the corresponding device
instance (unit) number. An error (DL_ERROR_ACK) is returned by the driver if the ppa field
value does not correspond to a valid device instance number for this system. The device is
initialized on first attach and de-initialized (stopped) at last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ The maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h> ).
■ The minimum SDU is 0.

Device and Network Interfaces 325

hme(7D)

■ The dlsap address length is 8.

■ The MAC type is DL_ETHER.
■ The sap length values is −2 meaning the physical address component is followed
immediately by a 2 byte sap component within the DLSAP address.
■ The service mode is DL_CLDLS.
■ No optional quality of service (QOS) support is included at present so the QOS fields are 0.
■ The provider style is DL_STYLE2.
■ The version is DL_VERSION_2.
■ The broadcast address value is Ethernet/IEEE broadcast address (0xFFFFFF).

Once in the DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular
SAP (Service Access Pointer) with the stream. The hme driver interprets the sap field within
the DL_BIND_REQ as an Ethernet “type” therefore valid values for the sap field are in the
[0-0xFFFF] range. Only one Ethernet type can be bound to the stream at any time.

If the user selects a sap with a value of 0, the receiver will be in “802.3 mode”. All frames
received from the media having a “type” field in the range [0-1500] are assumed to be 802.3
frames and are routed up all open Streams which are bound to sap value 0. If more than one
Stream is in “802.3 mode” then the frame will be duplicated and routed up multiple Streams as
DL_UNITDATA_IND messages.

In transmission, the driver checks the sap field of the DL_BIND_REQ if the sap value is 0, and if
the destination type field is in the range [0-1500]. If either is true, the driver computes the
length of the message, not including initial M_PROTO mblk (message block), of all subsequent
DL_UNITDATA_REQ messages and transmits 802.3 frames that have this value in the MAC frame
header length field.

The hme driver DLSAP address format consists of the 6 byte physical (Ethernet) address
component followed immediately by the 2 byte sap (type) component producing an 8 byte
DLSAP address. Applications should not hardcode to this particular implementation-specific
DLSAP address format but use information returned in the DL_INFO_ACK primitive to compose
and decompose DLSAP addresses. The sap length, full DLSAP length, and sap/physical ordering
are included within the DL_INFO_ACK. The physical address length can be computed by
subtracting the sap length from the full DLSAP address length or by issuing the
DL_PHYS_ADDR_REQ to obtain the current physical address associated with the stream.

Once in the DL_BOUND state, the user may transmit frames on the Ethernet by sending
DL_UNITDATA_REQ messages to the hme driver. The hme driver will route received Ethernet
frames up all those open and bound streams having a sap which matches the Ethernet type as
DL_UNITDATA_IND messages. Received Ethernet frames are duplicated and routed up multiple

326 man pages section 7: Device and Network Interfaces • Last Revised 5 Sep 1995
 hme(7D)

open streams if necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.

In addition to the mandatory connectionless DLPI message set the driver additionally supports
the following primitives.

hme Primitives The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable/disable reception of

individual multicast group addresses. A set of multicast addresses may be iteratively created
and modified on a per-stream basis using these primitives. These primitives are accepted by
the driver in any state following DL_ATTACHED.

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives with the DL_PROMISC_PHYS flag

set in the dl_level field enables/disables reception of all (“promiscuous mode”) frames on the
media including frames generated by the local host. When used with the DL_PROMISC_SAP flag
set this enables/disables reception of all sap (Ethernet type) values. When used with the
DL_PROMISC_MULTI flag set this enables/disables reception of all multicast group addresses.
The effect of each is always on a per-stream basis and independent of the other sap and
physical level configurations on this stream or other streams.

The DL_PHYS_ADDR_REQ primitive returns the 6 octet Ethernet address currently associated
(attached) to the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the 6 octet Ethernet address currently

associated (attached) to this stream. The credentials of the process which originally opened
this stream must be superuser. Otherwise EPERM is returned in the DL_ERROR_ACK. This
primitive is destructive in that it affects all other current and future streams attached to this
device. An M_ERROR is sent up all other streams attached to this device when this primitive is
successful on this stream. Once changed, all streams subsequently opened and attached to this
device will obtain this new physical address. Once changed, the physical address will remain
until this primitive is used to change the physical address again or the system is rebooted,
whichever comes first.

hme DRIVER By default, the hme driver performs “auto-negotiation” to select the mode and speed of the
link, when the Internal Transceiver is used.

When an External Transceiver is connected to the MII interface, the driver selects the External
Transceiver for networking operations. If the External Transceiver supports
“auto-negotiation”, the driver uses the auto-negotiation procedure to select the link speed and
mode. If the External Transceiver does not support auto-negotiation, it will select the highest
priority mode supported by the transceiver.
■ 100 Mbps, full-duplex
■ 100 Mbps, half-duplex
■ 10 Mbps, full-duplex

Device and Network Interfaces 327

hme(7D)

■ 10 Mbps, half-duplex

The link can be in one of the 4 following modes:

These speeds and modes are described in the 100Base-TX standard.

The auto−negotiation protocol automatically selects:

■ Operation mode (half-duplex or full-duplex)
■ Speed (100 Mbps or 10 Mbps)

The auto−negotiation protocol does the following:

■ Gets all the modes of operation supported by the Link Partner
■ Advertises its capabilities to the Link Partner
■ Selects the highest common denominator mode of operation based on the priorities

The internal transceiver is capable of all of the operating speeds and modes listed above. When
the internal transceiver is used, by default, auto-negotiation is used to select the speed and the
mode of the link and the common mode of operation with the Link Partner.

When an external transceiver is connected to the MII interface, the driver selects the external
transceiver for networking operations. If the external transceiver supports auto-negotiation:
■ The driver uses the auto-negotiation procedure to select the link speed and mode.

If the external transceiver does not support auto-negotiation

■ The driver selects the highest priority mode supported by the transceiver.

Sometimes, the user may want to select the speed and mode of the link. The SUNW,hme device
supports programmable “IPG” (Inter-Packet Gap) parameters ipg1 and ipg2. By default, the
driver sets ipg1 to 8 byte-times and ipg2 to 4 byte-times (which are the standard values).
Sometimes, the user may want to alter these values depending on whether the driver supports
10 Mbps or 100 Mpbs and accordingly, IPG will be set to 9.6 or 0.96 microseconds.

hme Parameter List The hme driver provides for setting and getting various parameters for the SUNW,hme device.
The parameter list includes:

current transceiver status

current link status
inter-packet gap
local transceiver capabilities
link partner capabilities

The local transceiver has two set of capabilities: one set reflects the capabilities of the
hardware, which are read-only (RO) parameters and the second set reflects the values chosen

328 man pages section 7: Device and Network Interfaces • Last Revised 5 Sep 1995
 hme(7D)

by the user and is used in speed selection. There are read/write (RW) capabilities. At boot
time, these two sets of capabilities will be the same. The Link Partner capabilities are also read
only parameters because the current default value of these parameters can only be read and
cannot be modified.
Files /dev/hme hme special character device
/kernel/drv/hme.conf System-wide default device driver properties

See Also ndd(1M), netstat(1M), driver.conf(4), dlpi(7P)

Device and Network Interfaces 329

hpfc(7D)

Name hpfc – Agilent fibre channel host bus adapter

Synopsis PCI pci103c

Description The hpfc fibre channel host bus adapter is a SCSA compliant nexus driver that supports all
Agilent fibre channel host bus adapters, including the HHBA5100x, HHBA5101x, and
HHBA5121x models. Agilent host bus adapters support the fibre channel protocol on private
fibre channel arbitrated loops and fabrics. The driver supports up to ten host bus adapters,
with a maximum of 125 fibre channel devices on each host bus adapter. The hpfc driver
supports a maximum of 256 LUNs per target.

The hpfc driver does not support the BIOS Int 13 feature, which enables the booting of an
operating system. As a result, you should not install an operating system on devices attached
to the hpfc driver.

Configuration The hpfc driver attempts to configure itself using the information in the
/kernel/drv/hpfc.conf configuration file.

By default, the driver supports only LUN 0 for each target device. To add multiple LUN
support, modify the /kernel/drv/sd.conf file.

Before upgrading the hpfc driver, backup the sd.conf file to save customized LUN settings
and then use pkgrm(1M) to remove the old version of the driver.

The host bus adapter port is initialized to FL_Port when connected to a fabric switch. To
change it to F_Port, add the init_as_nport=1 entry to the hpfc.conf file and reboot the
system.

To conserve system resources, at least one disk drive must be attached to the hpfc driver. If no
devices are attached, the driver will not load.
Files /kernel/drv/hpfc 32–bit ELF kernel module
/kernel/drv/sparcv9/hpfc 64–bit ELF kernel module
/kernel/drv/hpfc.conf Driver configuration file
/kernel/drv/sd.conf SCSI disk configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86, SPARC

See Also luxadm(1M), pkgrm(1M), prtconf(1M), driver.conf(4), attributes(5), ses(7D), ssd(7D)

ANSI X3.272–1996, Fibre Channel Arbitrated Loop (FC-AL),

330 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 hpfc(7D)

ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP),

ANSI X3.270-1996, SCSI-3 Architecture Model (SAM),

Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)

Device and Network Interfaces 331

hsfs(7FS)

Name hsfs – High Sierra & ISO 9660 CD-ROM file system

Description HSFS is a file system type that allows users to access files on High Sierra or ISO 9660 format
CD-ROM disks from within the SunOS operating system. Once mounted, a HSFS file system
provides standard SunOS read-only file system operations and semantics, meaning that you
can read and list files in a directory on a High Sierra or ISO 9660 CD-ROM and applications
can use standard UNIX system calls on these files and directories.

This file system contains support for Rock Ridge, ISO 9660 Version 2 and Joliet extensions.
These extensions provide support for file names with a length of at least 207 bytes, but only
Rock Ridge extensions (with the exception of writability and hard links) can provide file
system semantics and file types as they are found in UFS. The presence of Rock Ridge, ISO
9660 Version 2 and Joliet is autodetected and the best-suitable available extension is used by
the HSFS driver for file name and attribute lookup.

If your /etc/vfstab file contains a line similar to the following:

/dev/dsk/c0t6d0s0 −/hsfs hsfs -no ro

and /hsfs exists, you can mount an HSFS file system with either of the following commands:
mount -F hsfs -o ro device-special directory-name

or
mount /hsfs

By default, Rock Ridge extensions are used if available, otherwise ISO 9660 Version 2, then
Joliet are used. If neither extension is present HSFS defaults to the standard capabilities of ISO
9660. Since so-called hybrid CD-ROMs that contain multiple extensions are possible, you can
use the following mount options to deliberately disable the search for a specific extension or to
force the use of a specific extension even if a preferable type is present:
mount -F hsfs -o ro,nrr device-special directory-name

Mount options are:

rr—request HSFS to use Rock Ridge extensions, if present. This is the default behavior and
does not need to be explicitly specified.

nrr—disable detection and use of Rock Ridge extensions, even if present.

vers2—request HSFS to use ISO 9660 Version 2 extensions, even if Rock Ridge is available.

novers2—disable detection and use of ISO 9660 Version 2 extensions.

joliet—request HSFS to use Joliet extensions, even if Rock Ridge or ISO 9660 Version 2
extensions are available.

nojoliet—disable detection and use of Joliet extensions.

332 man pages section 7: Device and Network Interfaces • Last Revised 1 Nov 2006
 hsfs(7FS)

Files on a High Sierra or ISO 9660 CD-ROM disk have names of the form filename.ext;version,
where filename and the optional ext consist of a sequence of uppercase alphanumeric
characters (including ‘‘_''), while the version consists of a sequence of digits, representing the
version number of the file. HSFS converts all the uppercase characters in a file name to
lowercase, and truncates the ‘‘;'' and version information. If more than one version of a file is
present on the CD-ROM, only the file with the highest version number is accessible.

Conversion of uppercase to lowercase characters may be disabled by using the -o nomaplcase

option to mount(1M). (See mount_hsfs(1M)).

If the CD-ROM contains Rock Ridge, ISO 9660 version 2 or Joliet extensions, the file names
and directory names may contain any character supported under UFS. The names may also be
upper and/or lower case and are case sensitive. File name lengths can be as long as those of
UFS.

Files accessed through HSFS have mode 555 (owner, group and world readable and
executable), uid 0 and gid 3. If a directory on the CD-ROM has read permission, HSFS grants
execute permission to the directory, allowing it to be searched.

With Rock Ridge extensions, files and directories can have any permissions that are supported
on a UFS file system. However, under all write permissions, the file system is read-only, with
EROFS returned to any write operations.

Like High Sierra and ISO 9660 CD-ROMs, HSFS supports only regular files and directories. A
Rock Ridge CD-ROM can support regular files, directories, and symbolic links, as well as
device nodes, such as block, character, and FIFO.

Examples EXAMPLE 1 Sample Display of File System Files

If there is a file BIG.BAR on a High Sierra or ISO 9660 format CD-ROM it will show up as
big.bar when listed on a HSFS file system.

If there are three files

BAR.BAZ;1
BAR.BAZ;2

and
BAR.BAZ;3

on a High Sierra or ISO 9660 format CD-ROM, only the file BAR.BAZ;3 will be accessible. It
will be listed as bar.baz.

See Also mount(1M), mount_hsfs(1M), vfstab(4)

N. V. Phillips and Sony Corporation, System Description Compact Disc Digital Audio, ("Red
Book").

Device and Network Interfaces 333

hsfs(7FS)

N. V. Phillips and Sony Corporation, System Description of Compact Disc Read Only Memory,
("Yellow Book").

IR "Volume and File Structure of CD-ROM for Information Interchange", ISO 9660:1988(E).
Diagnostics hsfs: Warning: the file system...
does not conform to the ISO-9660 spec The specific reason appears on the following
line. You might be attempting to mount a
CD-ROM containing a different file system,
such as UFS.
hsfs: Warning: the file system...
contains a file [with an] unsupported type The hsfs file system does not support the
format of some file or directory on the
CD-ROM, for example a record structured file.
hsfs: hsnode table full, %d nodes allocated There are not enough HSFS internal data
structure elements to handle all the files
currently open. This problem may be overcome
by adding a line of the form set
hsfs:nhsnode=number to the /etc/system
system configuration file and rebooting. See
system(4).

Warnings Do not physically eject a CD-ROM while the device is still mounted as a HSFS file system.

Under MS-DOS (for which CD-ROMs are frequently targeted), files with no extension may be
represented either as:
filename.

or
filename

that is, with or without a trailing period. These names are not equivalent under UNIX systems.
For example, the names:
BAR.

and
BAR

are not names for the same file under the UNIX system. This may cause confusion if you are
consulting documentation for CD-ROMs originally intended for MS-DOS systems.

Use of the -o notraildot option to mount(1M) makes it optional to specify the trailing dot.
(See mount_hsfs(1M)).

334 man pages section 7: Device and Network Interfaces • Last Revised 1 Nov 2006
 hsfs(7FS)

Notes No translation of any sort is done on the contents of High Sierra or ISO 9660 format
CD-ROMs; only directory and file names are subject to interpretation by HSFS.

Device and Network Interfaces 335

hubd(7D)

Name hubd – USB hub driver

Synopsis hub@unit-address

Description The hubd is a USBA (Solaris USB Architecture) compliant client driver that supports USB
hubs conforming to the Universal Serial Bus Specification 2.0. The hubd driver supports
bus–powered and self–powered hubs. The driver supports hubs with individual port power,
ganged power and no power switching.

When a device is attached to a hub port, the hubd driver enumerates the device by determining
its type and assigning an address to it. For multi-configuration devices, hubd sets the preferred
configuration (refer to cfgadm_usb(1M) to select a configuration). The hubd driver attaches a
driver to the device if one is available for the default or selected configuration. When the
device is disconnected from the hub port, the hubd driver offlines any driver instance attached
to the device.
Files /kernel/drv/hubd 32– bit x86 ELF kernel module
/kernel/drv/amd64/hubd 64– bit x86 ELF kernel module
/kernel/drv/sparcv9/hubd 64–bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also cfgadm_usb(1M), attributes(5), usba(7D)

Writing Device Drivers

Universal Serial Bus Specification 2.0

System Administration Guide: Basic Administration

http://www.sun.com/io

Diagnostics In addition to being logged, the following messages may also appear on the system console.
Messages are formatted in the following manner:
WARNING: <device path> <hubd<instance number>): Message...

where <instance number> is the instance number of hubd and <device path> is the physical
path to the device in /devices directory. Messages from the root hub are displayed with a
usb<instance number> prefix instead of hub<instance number> as the root hub is an
integrated part of the host controller.

336 man pages section 7: Device and Network Interfaces • Last Revised 6 July 2006
 hubd(7D)

Connecting device on port <number> failed.

The driver failed to enumerate the device connected on port <number> of hub. If
enumeration fails, disconnect and re-connect.
Use of a USB 1.0 hub behind a high speed port may cause unexpected failures.
Devices connected to a USB 1.0 hub which are in turn connected to an external USB 2.0
hub, may misbehave unexpectedly or suddenly go offline. This is due to a documented
incompatibility between USB 1.0 hubs and USB 2.0 hub Transaction Translators. Please
use only USB 2.0 or USB 1.1 hubs behind high-speed ports.
Connecting a high speed device to a non-high speed hub (port x) will result in a loss of
performance. Please connect the device to a high speed port to get the maximum
performance.
USB 2.0 devices connected to USB 1.0 or 1.1 hubs cannot run at their highest speed, even
when the hub is in turn connected to a high-speed port. For best performance, reconnect
without going through a USB 1.0 or 1.1 hub.
Cannot access <device>. Please reconnect.
This hub has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.
Port <n> overcurrent.
An overcurrent condition was detected. Please remove the device on this port.
Devices not identical to the previous one on this port. Please disconnect and reconnect.
Same condition as described above; however in this case, the driver is unable to identify the
original device with a name string.
Hub driver supports max of <n> ports on hub. Hence, using the first <number of physical
ports> of <n> ports available.
The current hub driver supports hubs that have <n> ports or less. A hub with more than
<n> ports has been plugged in. Only the first <n> out of the total <number of physical
ports> ports are usable.
Hub global over current condition, please disconnect the devices connected to the hub to clear
the condition. You may need to re-connect the hub if the ports do not work.
An overcurrent condition was detected on the hub. This means that the aggregate current
being drawn by the devices on the downstream ports exceeds a preset value. Refer to
section 7.2.1.2 and 11.13 of the Universal Serial Bus Specification 2.0. If this message
continues to display, you may need to remove downstream devices to eliminate the
problem. If any port does not work after the overcurrent condition is cleared, re-connect
the hub to re-enable the ports.
Root hub over current condition, please check your system to clear the condition as soon as
possible. You may need to reboot the system if the root hub does not recover automatically.
An overcurrent condition was detected on the root hub, indicating that malfunctioning
devices on the downstream ports are drawing too much current. Please disconnect the
problematic downstream devices to eliminate the problem. If the root hub doesn't work

Device and Network Interfaces 337

hubd(7D)

after the overcurrent condition is cleared, you may need to reboot the system.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><hubd<instance number>): message...
Local power has been lost, please disconnect hub.
A USB self-powered hub has lost external power. All USB devices connected down-stream
from this hub will cease to function. Disconnect the hub, plug in the external power-supply
and then plug in the hub again.
Local power has been lost, the hub could draw <x> mA power from the USB bus.
A USB self/bus-powered hub has lost external power. Some USB devices connected
down-stream from this hub may cease to function. Disconnect the external power-supply
and then plug in the hub again.
Two bus-powered hubs cannot be concatenated.
A bus-powered hub was connected to a bus powered hub port. Please remove this
bus-powered hub and connect it to a self-powered hub or a root hub port.
Configuration <n> for device <device> at port <m> exceeds power available for this port.
Please re-insert your device into another hub port which has enough power.
The device requires more power than is available on this port.
Port <n> in over current condition, please check the attached device to clear the condition.
The system will try to recover the port, but if not successful, you need to re-connect the hub or
reboot the system to bring the port back to work.
An overcurrent condition was detected on port <n>. This means the device connected to
the port is drawing more current than the hub can supply. If this message continues to
display, please disconnect the device to eliminate the problem. If the port doesn't work after
the overcurrent condition is cleared, please re-connect the hub or reboot the system to
enable the port again.

338 man pages section 7: Device and Network Interfaces • Last Revised 6 July 2006
 hwahc(7D)

Name hwahc – Host Wire Adapter Host Controller Driver

Description The hwahc driver is a USBA (Solaris USB Architecture) compliant nexus driver that supports
the Wireless USB 1.0 Host Wire Adapter Host Controller, an industry standard developed by
USB-IF.

A Host Wire Adapter (HWA) is a USB device whose upstream connection is a USB 2.0 wired
interface. The HWA operates as a host to a cluster of downstream Wireless USB devices.

The hwahc driver supports bulk, interrupt and control transfers.

Files /kernel/drv/hwahc 32-bit ELF 86 kernel module
/kernel/drv/sparcv9/hwahc 64-bit SPARC ELF kernel module
/kernel/drv/amd64/hwahc 64-bit x86 ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWusb

See Also add_drv(1M), prtconf(1M), rem_drv(1M), update_drv(1M), attributes(5), ehci(7D),

hubd(7D), usba(7D)

Writing Device Drivers

System Administration Guide: Basic Administration

Universal Serial Bus Specification 2.0

Wireless Universal Serial Bus Specification 1.0

http://www.usb.org

http://www.sun.com

Diagnostics All host controller errors are passed to the client drivers. In addition to being logged, the
following messages can appear on the system console. All messages are formatted in the
following way:
WARNING: device_path hwahc instance_number: Message ...

Connection device on WUSB port port_number fails

The connecting device fails to connect to the HWA. Make sure the device has been associated
with the host.

Device and Network Interfaces 339

hwarc(7D)

Name hwarc – HWA Radio Controller Driver

Synopsis hwa-radio@unit-address

Description The hwarc driver is a USBA (Solaris USB Architecture) compliant client driver that supports
Host Wire Adapter Radio Controller, specified in Wireless Universal Serial Bus Specification,
Version 1.0.

The hwarc driver handles the Radio Controller Interface of an HWA device and properly
controls the UWB (Ultra Wideband) Radio in the device. The driver controls an HWA device
to Scan, Start/Stop Beacon, Get IE, and so forth.
Files /kernel/drv/hwarc 32-bit ELF 86 kernel module
/kernel/drv/sparcv9/hwarc 64-bit SPARC ELF kernel module
/kernel/drv/amd64/hwarc 64-bit x86 ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWusb

See Also add_drv(1M), prtconf(1M), rem_drv(1M), update_drv(1M), attributes(5), hwahc(7D),

usba(7D), uwba(7D),

Writing Device Drivers

Universal Serial Bus Specification 1.0, 1.1 and 2.0 - 1996, 1998, 2000

Wireless Universal Serial Bus Specification 1.0

http://www.usb.org

http://www.sun.com

340 man pages section 7: Device and Network Interfaces • Last Revised 17 Apr 2009
 hxge(7D)

Name hxge – Sun Blade 10 Gigabit Ethernet network driver

Synopsis /dev/hxge*

Description The hxge Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on the Sun Blade
Shared 10Gb Ethernet Interface.

The Shared PCI-Express 10 Gb networking interface provides network I/O consolidation for
up to six Constellation blades, with each blade seeing its own portion of the network interface.

The hxge driver functions include chip initialization, frame transmit and receive, flow
classification, multicast and promiscuous support and error recovery and reporting in the
blade domain.

Application The cloning character-special device, /dev/hxge, is used to access Sun Blade Shared 10Gb
Programming Ethernet Interface devices installed within the system.
Interface
The hxge driver is managed by the dladm(1M) command line utility, which allows VLANs to
be defined on top of hxge instances and for hxge instances to be aggregated. See dladm(1M)
for more details.

You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number for the system. The device is initialized on first attach and de-initialized (stopped) at
last detach

The values returned by the driver in the DL_INFO_ACK primitive in response to a

DL_INFO_REQ are:
■ Maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h>).
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP length value is -2, meaning the physical address component is followed immediately
by a 2-byte SAP component within the DLSAP address.
■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Due to the nature of the link address definition for IPoIB, the
DL_SET_PHYS_ADDR_REQ DLPI primitive is not supported.
In the transmit case for streams that have been put in raw mode via the DLIOCRAW ioctl,
the dlpi application must prepend the 20 byte IPoIB destination address to the data it
wants to transmit over-the-wire. In the receive case, applications receive the IP/ARP
datagram along with the IETF defined 4 byte header.

Device and Network Interfaces 341

hxge(7D)

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.
Configuration The link speed and mode are fixed at 10 Gbps full-duplex.
The default MTU is 1500. To enable jumbo frame support, you configure the hxge driver by
defining the accept-jumbo property to 1 in the hxge.conf file. Note that the largest jumbo size
is 9178 bytes.
The driver may be configured to discard certain classes of traffic. By default, no class of traffic
is allowed. You configure the hxge driver by defining the class option property to 0x20000 in
hxge.conf to discard the specified class of traffic. For example, the following line in
hxge.conf discards all IP Version 4 TCP traffic:
class-opt-ipv4-tcp = 0x20000;

You can also use the ndd(1M) command to configure the hxge driver at runtime to discard any
classes of traffic.
The hxgedriver supports the self-healing functionality of Solaris OS. By default it is configured
to DDI_FM_EREPORT_CAPABLE | DDI_FM_ERRCB_CAPABLE. You configure the hxge
driver by defining the fm-capable property in hxge.conf to other capabilities or to 0x0 to
disable it entirely.
The hxge driver may be configured using the standard ifconfig(1M) command.
The hxge driver also reports various hardware and software statistics data. You can view these
statistics using the kstat(1M) command.
Files /dev/hxge* Special character device.
/kernel/drv/hxge 32–bit device driver (x86).
/kernel/drv/sparcv9/hxge 64–bit device driver (SPARC).
/kernel/drv/amd64/hxge 64–bit device driver (x86).
/kernel/drv/hxge.conf Configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), ifconfig(1M), kstat(1M), ndd(1M), netstat(1M), driver.conf(4),

attributes(5), streamio(7I), dlpi(7P)
Writing Device Drivers
STREAMS Programming Guide

342 man pages section 7: Device and Network Interfaces • Last Revised 10 Feb 2007
 hxge(7D)

Network Interfaces Programmer's Guide

Device and Network Interfaces 343

i915(7d)

Name i915 – DRI-compliant kernel driver providing graphic hardware acceleration support

Description The i915 driver is a Direct Rendering Infrastructure (DRI)– compliant kernel driver that
provides graphics hardware acceleration support. DRI is a framework for coordinating OS
kernel, 3D graphics hardware, X window system and OpenGL applications.

The i915 driver currently supports the Intel i845, i865, i915, i945, i965 and G33 series
integrated graphics controllers.
Files /platform/i86pc/kernel/drv/i915 32–bit ELF kernel module (x86).
/platform/i86pc/kernel/drv/amd64/i915 64–bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdrmr

Architecture x86

See Also attributes(5)

/usr/X11/share/man/man1/Xserver.1

/usr/X11/share/man/man1/Xorg.1

/usr/X11/share/man/man5/X11.5

344 man pages section 7: Device and Network Interfaces • Last Revised 24 Jun 2009
 ib(7D)

Name ib – InfiniBand Bus Nexus Driver

Description The ib (IB nexus) driver is a pseudo nexus driver that supports enumeration of port devices,
VPPA (Virtual Physical Point Attachment), HCA_SVC (HCA Service) devices, and I/O
controllers (IOC) on the InfiniBand fabric that are visible to the host and provides interfaces
to cfgadm_ib(1M) to manage hot-plugging of IB devices. The ib nexus driver enumerates the
port device, VPPA devices and HCA_SVC devices based on entries specified in the ib.conf
file. IOC devices are enumerated on demand. The IB nexus driver uses InfiniBand Device
Manager services (ibdm(7D)) to enumerate port devices, VPPA devices, HCA_SVC devices,
and IOCs on the IB fabric.

Configuration You configure the ib driver by defining properties in the ib.conf file. The IB nexus driver
supports the following properties:

PROPERTY NAME DEFAULT POSSIBLE VALUES

port-svc-list "" List of service names, for example:

srv

vppa-svc-list "" List of service names, for example:

ipib

hca-svc-list "" List of service names, for example:

hca_nfs

The port-svc-list property defines the list of port communication service names per port.
The IB nexus driver creates a device instance for each entry in this property per Host Channel
Adapter (HCA) port. The ib.conf file contains a port-svc-list="" entry by default. You
update port-svc-list with service names you want to add to the system.

The vppa-svc-list property defines the list of VPPA communication service names per port
per partition key. The IB nexus driver creates a device instance for each entry in this property
per Host Channel Adapter (HCA) port. The ib.conf file contains a vppa-svc-list="" entry
by default. You update vppa-svc-list with service names you want to add to the system.

The hca-svc-list property defines the list of HCA_SVC communication service names per
HCA. The IB nexus driver creates a device instance for each entry in this property per Host
Channel Adapter (HCA). The ib.conf file contains a hca-svc-list="" entry by default. You
update hca-svc-list with service names you want to add to the system.

The service name specified in port-svc-list, vppa-svc-list and hca-svc-list must be

unique, be a maximum of four characters long, and is limited to digits 0-9 and letters a-z and
A-Z.

IOC drivers (which are parented by the IB nexus driver) may themselves have .conf files. To
distinguish those cases from pseudo drivers parented by IB nexus, such drivers should include

Device and Network Interfaces 345

ib(7D)

the "ib-node-type" property with value merge in the IOC driver.conf file. That property
ensures that properties from the .conf file are merged with other properties found through
hardware probing.
Examples Example 1: A sample ib.conf file with one service name entry for PORT communication
services.
#
# Copyright 2001-2003 Sun Microsystems, Inc. All rights reserved.
# Use is subject to license terms.
#
port-svc-list=""
vppa-svc-list="";
hca-svc-list="";

In Example 1, the IB nexus driver does not create any port/vppa/hca_svc device instances.
Example 2: A sample ib.conf file with one entry for "srv" service:

port-svc-list="srv"
vppa-svc-list="";
hca-svc-list="";

The IB nexus driver creates one srv service instance for every HCA port that exists on the
host. For example, if there are two HCAs, each with two ports on the host, the IB nexus driver
creates four instances of the srv service.
Example 3: A sample ib.conf file with one service name entry for each of Port and VPPA
communication services:

port-svc-list="srv"
vppa-svc-list="ipib";
hca-svc-list="";

If there are two HCAs in the system with two ports each and each port has two valid PKEY
values, the IB nexus driver creates four instances of srv service (one for each port). It also
creates eight instances of ipd service (one per each port/PKEY combination).
Example 4: A sample ib.conf file with one service name entry for each of Port, VPPA and
HCA_SVC communication services:

port-svc-list="srv";
vppa-svc-list="ipib";
hca-svc-list="hca_nfs";

The IB nexus driver creates one instance of hca_nfs service for each HCA in the system.
Example 5: IOC driver .conf

346 man pages section 7: Device and Network Interfaces • Last Revised 26 Jan 2009
 ib(7D)

ib-node-type="merge";
enable-special-mode="on";
Files /kernel/drv/ib 32-bit x86 ELF kernel module
/kernel/drv/amd64/ib 64-bit x86 ELF kernel module
/kernel/drv/sparcv9/ib 64-bit SPARC ELF kernel module
/kernel/drv/ib.conf driver configuration file

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWhea, SUNWib

Interface Stability Consolidation Private

See Also cfgadm_ib(1M), driver.conf(4), ib(4), attributes(5), ibcm(7D), ibdm(7D), ibtl(7D)

Writing Device Drivers

InfiniBand Architecture Specification, Volume 1: Release 1.1

System Administration Guide: Basic Administration

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
ib: WARNING: Error message...
unit-address property in %s.conf not well-formed. The driver.conf file does not have a
valid "unit-addr" property defined.
This property is an array of strings.
cannot find unit-address in %s.conf. The driver.conf file does not have a
valid "unit-addr" property defined.
This property is an array of strings.
Waiting for Port %d initialization. Waiting for port initialization from
subnet manager.

Device and Network Interfaces 347

ibcm(7D)

Name ibcm – Solaris InfiniBand Communication Manager

Description The Solaris InfiniBand Communication Manager (IBCM) is a Solaris kernel misc module that
adheres to the InfiniBand Architecture Specification, Volume 1: Release 1.1 for InfiniBand
Communication Management Class.

IBCM provides a transport layer abstraction to IB clients to set up reliable connected channels
along with service, multicast, and path lookup-related functionality. IBCM implements the
CM protocol as per the InfiniBand Architecture Specification, Volume 1: Release 1.1 and
utilizes the InfiniBand Management Framework module for all IB management-related
functionality and the InfiniBand Transport Layer (see ibtl(7D)) for all IB Verbs-related
functionality.
Files /kernel/misc/ibcm 32-bit x86 ELF kernel module
/kernel/misc/amd64/ibcm 64-bit x86 ELF kernel module
/kernel/misc/sparcv9/ibcm 64-bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Consolidation Private

Availability SUNWhea

See Also attributes(5), ibtl(7D)

InfiniBand Architecture Specification, Volume 1: Release 1.1

348 man pages section 7: Device and Network Interfaces • Last Revised 2 Apr 2004
 ibd(7D)

Name ibd – Infiniband IPoIB device driver

Synopsis /dev/ibd*
Description The ibd driver implements the IETF IP over Infiniband protocol and provides IPoIB service
for all IBA ports present in the system.
The ibd driver is a multi-threaded, loadable, clonable, STREAMS hardware driver supporting
the connectionless Data Link Provider Interface, dlpi(7P)).
By default, datagram mode is used by each ibd instance, unless the enable_rc is set to 1 for
that instance in the .conf file. This change can be made on a per instance basis by changing the
corresponding value of the variable. So the Nth value of enable_rc changes the setting for the
Nth instance of ibd. Any value other than 1, or no .conf file at all is equivalent to specifying
datagram mode.
Because ibd over connected mode attempts to use a large MTU (65520 bytes), applications
should adapt to the large MTU to get better performance, for example, adopting a large TCP
window size.
Use the cloning, character-special device /dev/ibd to access all ibd devices installed within
the system.
The ibd driver is dependent on GLD, a loadable kernel module that provides the ibd driver
with the DLPI and STREAMS functionality required of a LAN driver. Except as noted in the
Application Programming Interface section of this man page, see gld(7D) for more details
on the primitives supported by the driver. The GLD module is located at
/kernel/misc/sparcv9/gld on 64 bit systems and at /kernel/misc/gld on 32 bit systems.
The ibd driver expects certain configuration of the IBA fabric prior to operation (which also
implies the SM must be active and managing the fabric). Specifically, the IBA multicast group
representing the IPv4 limited broadcast address 255.255.255.255 (also defined as
broadcast-GID in IETF documents) should be created prior to initializing the device. IBA
properties (including mtu, qkey and sl) of this group is used by the driver to create any other
IBA multicast group as instructed by higher level (IP) software. The driver probes for the
existance of this broadcast-GID during attach(9E).
Application The values returned by the driver in the DL_INFO_ACK primitive in response to your
Programming DL_INFO_REQ are:
Interface (DLPI)
■ Maximum SDU is the MTU associated with the broadcast-GID group, less the 4 byte IPoIB
header.
■ Minimum SDU is 0.
■ dlsap address length is 22.
■ MAC type is DL_IB.
■ The sap length value is -2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.

Device and Network Interfaces 349

ibd(7D)

■ Broadcast address value is the MAC address consisting of the 4 bytes of QPN 00:FF:FF:FF
prepended to the IBA multicast address of the broadcast-GID.
Due to the nature of link address definition for IPoIB, the DL_SET_PHYS_ADDR_REQ
DLPI primitive is not supported.
In the transmit case for streams that have been put in raw mode via the DLIOCRAW ioctl,
the DLPI application must prepend the 20 byte IPoIB destination address to the data it
wants to transmit over-the-wire. In the receive case, applications receive the IP/ARP
datagram along with the IETF defined 4 byte header.

Warning This section describes warning messages that might be generated by the driver. Please note
that while the format of these messages can be modified in future versions, the same general
information is provided.

While joining IBA multicast groups corresponding to IP multicast groups as part of multicast
promiscuous operations as required by IP multicast routers, or as part of running snoop(1M),
it is possible that joins to some multicast groups can fail due to inherent resource constraints
in the IBA components. In such cases, warning message similar to the following appear in the
system log, indicating the interface on which the failure occurred:
NOTICE: ibd0: Could not get list of IBA multicast groups
NOTICE: ibd0: IBA promiscuous mode missed multicast group
NOTICE: ibd0: IBA promiscuous mode missed new multicast gid

Also, if the IBA SM indicates that multicast trap support is suspended or unavailable, the
system log contains a message similar to:
NOTICE: ibd0: IBA multicast support degraded due to
unavailability of multicast traps

And when the SM indicates trap support is restored:

NOTICE: ibd0: IBA multicast support restored due to
availability of multicast traps

Additionally, if the IBA link transitions to an unavailable state (that is, the IBA link state
becomes Down, Initialize or Armed) and then becomes active again, the driver tries to rejoin
previously joined groups if required. Failure to rejoin multicast groups triggers messages such
as:
NOTICE: ibd0: Failure on port up to rejoin multicast gid

If the corresponding HCA port is in the unavailable state defined above when initializing an
ibd interface using ifconfig(1M), a message is emitted by the driver:
NOTICE: ibd0: Port is not active

Further, as described above, if the broadcast-GID is not found, or the associated MTU is
higher than what the HCA port can support, the following messages are printed to the system
log:

350 man pages section 7: Device and Network Interfaces • Last Revised 19 Jan 2010
 ibd(7D)

NOTICE: ibd0: IPoIB broadcast group absent

NOTICE: ibd0: IPoIB broadcast group MTU 4096 greater than port’s
maximum MTU 2048

In all cases of these reported problems when running ifconfig(1M), it should be checked that
IBA cabling is intact, an SM is running on the fabric, and the broadcast-GID with appropriate
properties has been created in the IBA partition.
The MTU of Reliable Connected mode can be larger than the MTU of Unreliable Datagram
mode.
When Reliable Connected mode is enabled, ibd still uses Unreliable Datagram mode to
transmit and receive multicast packets. If the payload size (excluding 4 byte IPoIB header) of a
multicast packet is larger than the IP link MTU specified by the broadcast group, ibd drops it.
A message appears in the system log when drops occur:
NOTICE: ibd0: Reliable Connected mode is on. Multicast packet
length (<packet length> > <IP_LINK_MTU>) is too long to send

If only one side has enabled Reliable Connected mode, communication falls back to datagram
mode. The connected mode instance uses Path MTU discovery to automatically adjust the
MTU of a unicast packet if an MTU difference exists. Before Path MTU discovery reduces the
MTU for a specific destination, several packets whose size exceeds the MTU of Unreliable
Datagram mode is dropped.
Configuration The IPoIB service comes preconfigured on all HCA ports in the system. To turn the service off,
or back on after turning it off, refer to documentation in cfgadm_ib(1M).

Examples EXAMPLE 1 Enabling Connected Mode

The following example driver .conf file enables Connected Mode for ibd instances 0 and 1.
Instances 2 and 3 use datagram mode.
# 1: unicast packets is sent over Reliable Connected Mode
# 0: unicast packets will be sent over Unreliable Datagram Mode
#
# Each element in the list below maps to the corresponding ibd
# instance; the first element is for ibd instance 0, the second
# element is for instance 1 and so on.
#
enable_rc=1,1,0,0
Files /dev/ibd* Special character device
/kernel/drv/ib.conf Configuration file to start IPoIB service
/kernel/drv/ibd.conf Configuration file for IPoIB driver
/kernel/drv/sparcv9/ibd 64–bit SPARC device driver
/kernel/drv/amd64/ibd 64–bit x86 device driver

Device and Network Interfaces 351

ibd(7D)

/kernel/drv/ibd 32–bit x86 device driver

See Also cfgadm(1M), cfgadm_ib(1M), ifconfig(1M), syslogd(1M), gld(7D), ib(7D), kstat(7D),

streamio(7I), dlpi(7P), attributes(5), attach(9E)

Notes IBD is a GLD-based driver and provides the statistics described by gld(7D). Note that valid
received packets not accepted by any stream (long) increases when IBD transmits broadcast IP
packets. This happens because the infiniband hardware copies and loops back the transmitted
broadcast packets to the source. These packets are discarded by GLD and are recorded as
'unknowns'.

352 man pages section 7: Device and Network Interfaces • Last Revised 19 Jan 2010
 ibdm(7D)

Name ibdm – Solaris InfiniBand Device Manager

Description The Infiniband Device Manager (IBDM) is an IBTF-compliant kernel misc module. IBDM
adheres to the InfiniBand Device Management class as described in InfiniBand Architecture
Specification, Volume 1: Release 1.1 and enumerates all the devices which are visible from a
given host and maintains a data base of all IB devices visible to the host. IBDM provides
interfaces to the IB nexus driver that enables the driver to retrieve information about IB
devices on the fabric.
Files /kernel/misc/ibdm 32-bit x86 ELF kernel module
/kernel/misc/amd64/ibdm 64-bit x86 ELF kernel module
/kernel/misc/sparcv9/ibdm 64-bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Contract Consolidation Private

Availability SUNWhea

See Also attributes(5), ib(7D), ibtl(7D), ibcm(7D)

InfiniBand Architecture Specification, Volume 1: Release 1.1

Diagnostics None.

Device and Network Interfaces 353

ibdma(7D)

Name ibdma – Solaris InfiniBand Device Manager Agent

Description The Infiniband Device Manager Agent (ibdma) is an IBTF-compliant kernel misc module.

IBDMA implements limited portions of the target (agent) side of the InfiniBand Device
Management class as described in InfiniBand Architecture Specification, Volume 1: Release
1.2.1.

IBDMA responds to incoming Device Management Datagrams (MADS) by enumerating

available target-side Infiniband services. Initiator systems can use this service to discover
target-side resources such as the virtual I/O Controllers exported by srpt(7D).
Files /kernel/misc/ibdma 32-bit x86 ELF kernel module
/kernel/misc/amd64/ibdma 64-bit x86 ELF kernel module
/kernel/misc/sparcv9/ibdma 64-bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWibdmar

See Also attributes(5), ib(7D), ibdm(7D), ibtl(7D), srpt(7D)

InfiniBand Architecture Specification, Volume 1: Release 1.2.1

354 man pages section 7: Device and Network Interfaces • Last Revised 21 May 2009
 ibmf(7)

Name ibmf – InfiniBand Management Transport Framework

Description The InfiniBand (IB) Management Transport Framework provides the mechanisms for IB
management modules to communicate with other InfiniBand management modules such as
the Subnet Administration process. It also provides helper functions such as Subnet
Administration Access (SAA) for commonly performed operations.
Files /kernel/misc/ibmf 32-bit ELF kernel misc module (x86 platform only).
/kernel/misc/amd64/ibmf 64-bit ELF kernel misc module (x86 platform only).
/kernel/misc/sparcv9/ibmf 64-bit ELF kernel misc module (SPARC platform only).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Interface stability Consolidation Private

Availability SUNWib

See Also ibtl(7D)

InfiniBand Architecture Specification, Version 1.1

www.infinibandta.org

Device and Network Interfaces 355

ibtl(7D)

Name ibtl – Solaris InfiniBand Transport Layer

Description InfiniBand (IB) is an I/O technology based on switched fabrics. The Solaris InfiniBand
Transport Layer (IBTL) is a Solaris kernel misc module and adheres to the IB Architecture
Version 1.1 specification and provides a transport layer abstraction to IB client drivers.

IBTL implements the programming interfaces for the Solaris InfiniBand Transport
Framework (IBTF), consisting of the IB Channel Interface (CI) and the IB Transport Interface
(TI).

The CI consists of Host Channel Adapters (HCAs) and HCA drivers. A host is attached to the
IB fabric through the CI layer. The Solaris InfiniBand CI is Sun's API rendering of the
InfiniBand Architecture (IBTA) "verbs" specification.

The Solaris InfiniBand TI is the kernel service driver interface into the Solaris InfiniBand
Transport Framework. It provides transport and communications setup programming
interfaces for Unreliable Datagram (UD) and Reliable Connected (RC) transport types only.
Files /kernel/misc/ibtl 32-bit x86 ELF kernel misc module
/kernel/misc/amd64/ibtl 64-bit x86 ELF kernel misc module
/kernel/misc/sparcv9/ibtl 64-bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Consolidation Private

Availability SUNWhea, SUNWib

See Also attributes(5), ib(7D), ibcm(7D), ibdm(7D)

InfiniBand Architecture Specification, Volume 1: Release 1.1

356 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 2004
 icmp6(7P)

Name icmp6 – Internet Control Message Protocol for Internet Protocol Version 6
Synopsis #include <sys/socket.h>
#include <netinet/in.h>
#include <netinet/ip_icmp.h>
#include <netinet/icmp6.h>
s = socket(AF_INET6, SOCK_RAW, proto);
t = t_open("/dev/icmp6", O_RDWR);

Description The ICMP6 protocol is the error and control message protocol used with Version 6 of the
Internet Protocol. It is used by the kernel to handle and report errors in protocol processing. It
is also used for IPv6 neighbor and router discovery, and for multicast group membership
queries and reports. It may also be accessed by programs using the socket interface or the
Transport Level Interface (TLI) for network monitoring and diagnostic functions. When used
with the socket interface, a “raw socket” type is used. The protocol number for ICMP6, used in
the proto parameter to the socket call, can be obtained from getprotobyname(3SOCKET).
ICMP6 file descriptors and sockets are connectionless and are normally used with the
t_sndudata / t_rcvudata and the sendto() / recvfrom() calls. They may also be used with
the sendmsg()/recvgmsg() calls when sending or receiving ancillary data.

Outgoing packets automatically have an Internet Protocol Version 6 (IPv6) header and zero
or more IPv6 extension headers prepended. These headers are prepended by the kernel.
Unlike ICMP for IPv4, the IP_HDRINCL option is not supported for ICMP6, so ICMP6
applications neither build their own outbound IPv6 headers, nor do they receive the inbound
IPv6 headers with received data. IPv6 extension headers and relevant fields of the IPv6 header
may be set or received as ancillary data to a sendmsg(3SOCKET) or recvmsg(3SOCKET)
system call. Each of these fields and extension headers may also be set on a per socket basis
with the setsockopt(3SOCKET) system call. Such "sticky" options are used on all outgoing
packets unless overridden by ancillary data. When any ancillary data is present with a
sendmsg(3SOCKET) system call, all sticky options are ignored for that system call, but
subsequently remain configured.

ICMP6 is a datagram protocol layered above IPv6. Received ICMP6 messages may be reflected
back to users of higher-level protocols such as TCP or UDP as error returns from system calls.
A copy of each ICMP6error message received by the system is provided to every holder of an
open ICMP6 socket or TLI descriptor.

See Also getprotobyname(3SOCKET), recv(3SOCKET), recvmsg(3SOCKET), send(3SOCKET),

sendmsg(3SOCKET), setsockopt(3SOCKET), t_rcvudata(3NSL), t_sndudata(3NSL),
inet6(7P), ip6(7P), routing(7P)

Conta, A. and Deering, S., RFC 2463, Internet Control Message Protocol (ICMPv6) for the
Internet Protocol Version 6 (IPv6) Specification, The Internet Society, December 1998.

Device and Network Interfaces 357

icmp6(7P)

Diagnostics A socket operation may fail with one of the following errors returned:
EISCONN An attempt was made to establish a connection on a socket which already
has one, or when trying to send a datagram with the destination address
specified and the socket is already connected.
ENOTCONN An attempt was made to send a datagram, but no destination address is
specified, and the socket has not been connected.
ENOBUFS The system ran out of memory for an internal data structure.
EADDRNOTAVAIL An attempt was made to create a socket with a network address for which
no network interface exists.
ENOMEM The system was unable to allocate memory for an internal data structure.
ENOPROTOOPT An attempt was made to set an IPv4 socket option on an IPv6 socket.
EINVAL An attempt was made to set an invalid or malformed socket option.
EAFNOSUPPORT An attempt was made to bind or connect to an IPv4 or mapped address,
or to specify an IPv4 or mapped address as the next hop.

358 man pages section 7: Device and Network Interfaces • Last Revised 10 Nov 1999
 icmp(7P)

Name icmp, ICMP – Internet Control Message Protocol

Synopsis #include <sys/socket.h>
#include <netinet/in.h>
#include <netinet/ip_icmp.h>
s = socket(AF_INET, SOCK_RAW, proto);
t = t_open("/dev/icmp", O_RDWR);

Description ICMP is the error and control message protocol used by the Internet protocol family. It is used
by the kernel to handle and report errors in protocol processing. It may also be accessed by
programs using the socket interface or the Transport Level Interface (TLI) for network
monitoring and diagnostic functions. When used with the socket interface, a “raw socket”
type is used. The protocol number for ICMP, used in the proto parameter to the socket call, can
be obtained from getprotobyname(3SOCKET). ICMP file descriptors and sockets are
connectionless, and are normally used with the t_sndudata / t_rcvudata and the sendto() /
recvfrom() calls.

Outgoing packets automatically have an Internet Protocol (IP) header prepended to them.
Incoming packets are provided to the user with the IP header and options intact.

ICMP is an datagram protocol layered above IP. It is used internally by the protcol code for
various purposes including routing, fault isolation, and congestion control. Receipt of an
ICMP “redirect” message will add a new entry in the routing table, or modify an existing one.
ICMP messages are routinely sent by the protocol code. Received ICMP messages may be
reflected back to users of higher-level protocols such as TCP or UDP as error returns from
system calls. A copy of all ICMP message received by the system is provided to every holder of
an open ICMP socket or TLI descriptor.

See Also getprotobyname(3SOCKET), recv(3SOCKET), send(3SOCKET), t_rcvudata(3NSL),

t_sndudata(3NSL), inet(7P), ip(7P), routing(7P)

Postel, Jon, Internet Control Message Protocol — DARPA Internet Program Protocol
Specification, RFC 792, Network Information Center, SRI International, Menlo Park, Calif.,
September 1981.

Diagnostics A socket operation may fail with one of the following errors returned:
EISCONN An attempt was made to establish a connection on a socket which already
has one, or when trying to send a datagram with the destination address
specified and the socket is already connected.
ENOTCONN An attempt was made to send a datagram, but no destination address is
specified, and the socket has not been connected.
ENOBUFS The system ran out of memory for an internal data structure.
EADDRNOTAVAIL An attempt was made to create a socket with a network address for which
no network interface exists.

Device and Network Interfaces 359

icmp(7P)

Notes Replies to ICMP “echo” messages which are source routed are not sent back using inverted
source routes, but rather go back through the normal routing mechanisms.

360 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 idn(7d)

Name idn – inter-domain network device driver

Synopsis /dev/idn

Description The idn driver is a multi-thread, loadable, clonable, STREAMS-based pseudo driver that
supports the connectionless Data Link Provider Interface dlpi(7P) over the Sun Enterprise
10000 Gigplane-XB Interconnect. This connection is permitted only between domains within
the same Sun Enterprise 10000 server.

The idn driver supports 1 to 32 logical network interfaces that can be connected to domains
linked to the local domain through the domain_link(1M) command. (See domain_link(1M)
in the for more information.) The idn driver works in conjunction with the System Service
Processor (SSP) to perform domain linking/unlinking and automated linking upon host
bootup.

The /dev/idn device is used to access all IDN services provided by the system.

IDN and DLPI The idn driver is a style-2 Data Link Service provider. All M_PROTO and M_PCPROTO–type
messages are interpreted as DLPI primitives. For the idn driver to associate the opened stream
with a particular device (ppa), you must send an explicit DL_ATTACH_REQ message. The ppa ID
is interpreted as an unsigned long and indicates the corresponding device instance (unit)
number. The DL_ERROR_ACK error is returned by the driver if the ppa field value does not
correspond to a valid device-instance number for the system. The device is initialized on first
attach and de-initialized (stopped) on the last detach.
■ The maximum SDU is configurable by using the idn.conf file and has a range of 512 bytes
to 512 Kbytes. The default value is 16384 bytes.
■ The minimum SDU is 0.
■ The Service Access Pointer (SAP) address length is 8.
■ The MAC type is DL_ETHER.
■ The SAP length value is -2, meaning the physical address component is followed
immediately by a 2-byte SAP component within the DLSAP address.
■ The service mode is DL_CLDLS.
■ Optional quality of service (QOS) is not presently supported; accordingly, the QOS fields
are 0.
■ The provider style is DL_STYLE2.
■ The version is DL_VERSION_2.
■ The broadcast address value is Ethernet/IEEE broadcast address (0xFFFFFF). The idn
driver supports broadcast by issuing messages to each target individually. The idn driver is
inherently a point-to-point network between domains. When the idn driver is in the
DL_ATTACHED state, the user must send a DL_BIND_REQ request to associate a particular SAP
with the stream. The idn driver interprets the SAP field within the DL_BIND_REQ message

Device and Network Interfaces 361

idn(7d)

as an Ethernet type and valid values for the SAP field are in the range of 0 to 0xFFFF. Only
one Ethernet type can be bound to the stream at any time.

If a SAP with a value of 0 is selected, the receiver will be in 802.3 mode. All frames received
from the media having a type field in the range of 0 to 1500 are assumed to be 802.3 frames and
are routed up all open streams which are bound to SAP value 0. If more than one stream is in
802.3 mode, then the frame will be duplicated and routed up as multiple stream
DL_UNITDATA_IND messages.

In transmission, the driver checks the SAP field of the DL_BIND_REQ to determine if the SAP
value is 0, and if the destination type field is in the range of 0 to 1500. If either is true, the driver
computes the length of the message, (excluding the initial message block M_PROTO mblk) of all
subsequent DL_UNITDATA_REQ messages and transmits 802.3 frames that have this value in the
MAC frame header length field.

The driver also supports raw M_DATA mode. When the user sends a DLIOCRAW ioctl, the
particular stream is put in raw mode. A complete frame and a proper ether header is expected
as part of the data.

The DLSAP address format consists of the 6-byte, physical address component (Ethernet)
followed immediately by the 2-byte SAP component (type), producing an 8-byte DLSAP
address. Applications should not hardcode to this particular implementation-specific DLSAP
address format, but instead should use information returned in the DL_INFO_ACK primitive to
compose and decompose DLSAP addresses. The SAP length, full DLSAP length, and SAP
physical ordering are included within the DL_INFO_ACK primitive. The physical address length
can be computed by subtracting the SAP length from the full DLSAP address length or by
issuing the DL_PHYS_ADDR_REQ message to obtain the current physical address associated with
the stream.

When the idn driver is in the DL_BOUND state, you can transmit frames on the IDN by sending
DL_UNITDATA_REQ messages to the driver. The driver then routes received IDN frames up the
open and bound streams having a SAP which matches the Ethernet type as DL_UNITDATA_IND
messages. If necessary, received IDN frames are duplicated and routed up multiple open
streams. The DLSAP address contained within the DL_UNITDATA_REQ and DL_UNITDATA_IND
messages consists of both the SAP (type) and physical (Ethernet) components.

IDN Primitives In addition to the mandatory connectionless DLPI message set, the idn driver supports the
following primitives:

The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives which enable or disable,

respectively, the reception of individual multicast group addresses. A set of multicast
addresses may be iteratively created and modified on a per-stream basis using these
primitives. These primitives are accepted by the driver in any state following the DL_ATTACHED
state.

362 man pages section 7: Device and Network Interfaces • Last Revised 3 Jun 1999
 idn(7d)

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives, which with the

DL_PROMISC_PHYS flag set in the dl_level field, enable or disable, respectively, the reception
of all promiscuous frames on the media, including frames generated by the local domain.
When used with the DL_PROMISC_SAP flag set in the dl_level field, these primitives enable or
disable, respectively, the reception of all SAP (Ethernet type) values. When used with the
DL_PROMISC_MULTI flag set in the dl_level field, these primitives enable or disable,
respectively, the reception of all multicast group addresses. The effect of each is always on a
per-stream basis and independent of the other SAP and physical level configurations on this
stream or other streams.

The DL_PHYS_ADDR_REQ primitive which returns the 6-octet, Ethernet address associated with
(or attached to) the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ request.

Because the driver maintains domain address information in the address to direct packets to
the correct destination, the DL_SET_PHYS_ADDR_REQ primitive is not allowed.

Files The following files are supported:

/dev/idn
IDN special character device
/platform/SUNW,Ultra-Enterprise-10000/kernel/drv/idn.conf
System-wide and per-interface default device driver properties

See Also netstat(1M), ndd(1M), dlpi(7P)

domain_link(1M) in the .

Notes The idn driver supports a set of properties that can be set by using the driver.conf file for the
IDN. See the for more information about the properties in the driver.conf(4), (idn.conf, for
IDNs).

Device and Network Interfaces 363

iec61883(7I)

Name iec61883 – IEC 61883 interfaces

Synopsis #include <sys/av/iec61883.h>

Description The set of interfaces described in this man page can be used to control and exchange data with
consumer audio/video devices using protocols specified inIEC 61883 Consumer Electronic
Audio/Video Equipment - Digital Interface, including Common Isochronous Packet (CIP),
Connection Management Procedures (CMP) and Function Control Protocol (FCP).

An iec61883 compliant driver exports two device nodes for isochronous and for
asynchronous transactions. See the FILES section of this man page for the namespace
definition.

Isochronous Transfers Two methods are provided to receive/transmit isochronous data: using mmap(2) in
combination with ioctl(2), and read(2) or write(2).

Mmap/Ioctl This method provides better performance and finer-grained control than read/write, and is a
method of choice for most applications. The data buffer is mapped into a user process address
space, which means no data copying between the kernel and an application is necessary.
Synchronization between user processes and the driver is performed using ioctl(2)
commands.

An application allocates resources for isochronous transfer using IEC61883_ISOCH_INIT.

Then the data buffer can be mapped into the process space using mmap(2).

A circular data buffer consists of one or more equal size frame buffers (further referred to as
frames, unless to avoid ambiguity with AV frames). Frames are numbered starting with zero
and are always transferred sequentially. Frames consist equal sized packets. Each packet
contains a CIP header and one or more data blocks.

A driver and an application act as a producer and a consumer: producer supplies full frames
(filled with data) to the consumer, and the producer is not allowed to access those frames until
the consumer claims them empty.

A transfer can be initiated and suspended with IEC61883_START and IEC61883_STOP

commands respectively. IEC61883_RECV or IEC61883_XMIT is used for producer-consumer
synchronization.

Read/Write Using this method, an application calls read(2) or write(2) to receive or transmit a specified
amount of data. Bus-specific overhead, such as isochronous packet headers, is handled by the
driver and is not exposed to applications. Data returned by read(2) contains CIP headers and
data blocks. Empty packets are not returned by read(2). write(2) data should meet the same
requirements.

If one or more channels have been allocated since open(2) (see IEC61883_ISOCH_INIT), the
data is received/transmitted using channel that was created the last.

364 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 2009
 iec61883(7I)

If no channels were allocated, the driver uses the broadcast channel by default and allocates
the default-size data buffer. During transmit, the first packet's CIP header is used to
auto-detect the data format. If it is one of the formats supported by the driver, it is properly
transmitted (with inserted empty packets and timestamps).

For both methods, if during transmit the driver runs out of data, it transmits empty packets
containing only a CIP header of the next to be transmitted packet, as defined in IEC 61883-1.

Connection Applications wishing to follow Connection Management Procedures (CMP) in combination

Management with isochronous transfers should use the ioctl(2) IEC61883_PLUG_INIT,
Procedures
IEC61883_PLUG_FINI, IEC61883_PLUG_REG_READ and IEC61883_PLUG_REG_CAS commands.

Asynchronous read(2), write(2), ioctl(2), and poll(2) can be used with asynchronous nodes.
Transactions Asynchronous data exchange between a driver and an application utilizes a common data
structure called asynchronous request (ARQ):
typedef struct iec61883_arq {
int arq_type;
int arq_len;
union {
uint32_t quadlet;
uint64_t octlet;
uint8_t buf[8];
} arq_data;
} iec61883_arq_t;

arq_type contains ARQ type:

IEC61883_ARQ_FCP_CMD
IEC61883_ARQ_FCP_RESP
FCP command and response frame respectively. Outgoing frames are sent using write(2),
incoming frames are received with read(2).

See IEC 61883-1 for the FCP frame structure definition.

IEC61883_ARQ_BUS_RESET
Returned by the driver when a bus reset occurs. There is no data associated with this
request type, and arq_len is set to 0.

If arq_len is 4 or 8, then data should be supplied in arq_data.quadlet or arq_data.octlet

respectively, otherwise up to 8 bytes can be put in arq_data.buf, with the rest of the data
following immediately after.

write(2) For a request to be sent to a target, an iec61883_arq_t structure along with associated data is
passed to the driver using write(2). write() blocks until the request is completed.

Device and Network Interfaces 365

iec61883(7I)

read(2) A driver collects incoming ARQs in the internal buffer. Buffer size can be changed using the
ioctl(2) command IEC61883_FCP_SET_IBUF_SIZE.

Reading an ARQ takes one or two steps depending on data length. An application first reads
sizeof (iec61883_arq_t) bytes: if arq_len is less than or equal 4, which is usually the case,
no additional step is needed. Otherwise, the remaining arq_len - 4 bytes should be read and
concatenated.

read(2) blocks until the specified amount of data is available, unless O_NONBLOCK or O_NDELAY
flag was set during open(2), in which caseread(2) returns immediately.

poll(2) Applications can poll(2) asynchronous nodes on the POLLIN event.

Bus Reset In case of a bus reset, the driver notifies an application by generating an ARQ of type
IEC61883_ARQ_BUS_RESET.

If there were established isochronous connections before bus reset, the driver attempts to
restore all connections as described in IEC 61883 and resume any active transfers that were in
progress.

ioctls The following commands only apply to isochronous nodes:

IEC61883_ISOCH_INIT
This command allocates a data buffer and isochronous resources (if necessary) for the
isochronous transfer. The argument is a pointer to the structure:
typedef struct iec61883_isoch_init {
int ii_version; /* interface version */
int ii_pkt_size; /* packet size */
int ii_frame_size; /* packets/frame */
int ii_frame_cnt; /* # of frames */
int ii_direction; /* xfer direction */
int ii_bus_speed; /* bus speed */
uint64_t ii_channel; /* channel mask */
int ii_dbs; /* DBS */
int ii_fn; /* FN */
int ii_rate_n; /* rate numerator */
int ii_rate_d; /* rate denominator */
int ii_ts_mode; /* timestamp mode */
int ii_flags; /* flags */
int ii_handle; /* isoch handle */
int ii_frame_rcnt; /* # of frames */
off_t *ii_mmap_off /* mmap offset */
int ii_rchannel; /* channel */
int ii_error; /* error code */
} iec61883_isoch_init_t;

ii_version should be set to IEC61883_V1_0.

366 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 2009
 iec61883(7I)

The driver attempts to allocate a data buffer consisting of ii_frame_cnt frames, with
ii_frame_size packets in each frame. Packet size in bytes is specified by ii_pkt_size
specifies and should be a multiple of 512 and compatible with ii_bus_speed.

ii_direction can take one of the following values:

IEC61883_DIR_RECV
Receiving isochronous data.
IEC61883_DIR_XMIT
Transmitting isochronous data.

ii_bus_speed chooses bus speed to be used and can be either IEC61883_S100,

IEC61883_S200 or IEC61883_S400.

ii_channel is a mask that specifies an isochronous channel number to be used, with the
Nth bit representing channel N. When transmitting data, several bits can be set at a time, in
which case the driver chooses one, for example, 0x3FF means a range from 0 to 9. In case of
receive, only one bit can be set.

ii_dbs specifies data block size in quadlets, for example, DBS value for SD-DVCR is 0x78.
Refer to IEC 61883 for more details on DBS.

ii_fn specifies fraction number, which defines the number of blocks in which a source
packet is divided. Allowed values are from 0 to 3. Refer to IEC 61883 for more details on
FN.

Data rate expected by the AV device can be lower than the bus speed, in which case the
driver has to periodically insert empty packets into the data stream to avoid device buffer
overflows. This rate is specified with a fraction N/D, set by ii_rate_n and ii_rate_d
respectively. Any integer numbers can be used, or the following predefined constants:
IEC61883_RATE_N_DV_NTSC IEC61883_RATE_D_DV_NTSC
Data rate expected by DV-NTSC devices.
IEC61883_RATE_N_DV_PAL IEC61883_RATE_D_DV_PAL
Data rate expected by DV-PAL devices.

During data transmission, a timestamp based on the current value of the cycle timer is
usually required. ii_ts_mode defines timestamp mode to be used:
IEC61883_TS_SYT
Driver puts a timestamp in the SYT field of the first CIP header of each frame.
IEC61883_TS_NONE
No timestamps.

ii_dbs, ii_fn, ii_rate_n, ii_rate_d and ii_ts_mode are only required for
transmission. In other case these should be set to 0.

Device and Network Interfaces 367

iec61883(7I)

ii_flags should be set to 0.

If command succeeds, ii_handle contains a handle that should be used with other
isochronous commands. ii_frame_rcnt contains the number of allocated frames (can be
less than ii_frame_cnt). ii_mmap_off contains an offset to be used in mmap(2), for
example, to map an entire data receive buffer:
pa = mmap(NULL, init.ii_pkt_size *
init.ii_frame_size * init.ii_frame_rcnt,
PROT_READ, MAP_PRIVATE, fd, init.ii_mmap_off);

ii_rchannel contains channel number.

In case of command success, ii_error is set to 0; otherwise one of the following values can
be returned:
IEC61883_ERR_NOMEM
Not enough memory for the data buffer.
IEC61883_ERR_NOCHANNEL
Cannot allocate isochronous channel.
IEC61883_ERR_PKT_SIZE
Packet size is not allowed at this bus speed.
IEC61883_ERR_VERSION
Interface version is not supported.
IEC61883_ERR_INVAL
One or more the parameters are invalid
IEC61883_ERR_OTHER
Unspecified error type.
IEC61883_ISOCH_FINI
Argument is a handle returned by IEC61883_ISOCH_INIT. This command frees any
resources associated with this handle. There must be no active transfers and the data buffer
must be unmapped; otherwise the command fails.
IEC61883_START
This command starts an isochronous transfer. The argument is a handle returned by
IEC61883_ISOCH_INIT.
IEC61883_STOP
This command stops an isochronous transfer. The argument is a handle returned by
IEC61883_ISOCH_INIT.
IEC61883_RECV
This command is used to receive full frames and return empty frames to the driver. The
argument is a pointer to the structure:

368 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 2009
 iec61883(7I)

typedef struct iec61883_recv {

int rx_handle; /* isoch handle */
int rx_flags; /* flags */
iec61883_xfer_t rx_xfer; /* xfer params */
} iec61883_recv_t;

typedef struct iec61883_xfer {

int xf_empty_idx; /* first empty frame */
int xf_empty_cnt; /* empty frame count */
int xf_full_idx; /* first full frame */
int xf_full_cnt; /* full frame count */
int xf_error; /* error */
} iec61883_xfer_t;
rx_flags should be set to 0.
An application sets xf_empty_idx and xf_empty_cnt to indicate frames it no longer needs.
E.g. if a buffer consists of 6 frames, xf_empty_idx is 4, xf_empty_cnt is 3 - means that
frames 4, 5 and 0 can now be reused by the driver. If there are no empty frames, for
example, the first time this command is called, xf_empty_cnt should be set to 0.
When the command returns, xf_full_idx and xf_full_cnt specifies the frames that are
full. xf_error is always 0.
In general, AV frame boundaries are not aligned with the frame buffer boundaries, because
the first received packet might not be the first packet of an AV frame, and, in contrast with
the read/write method, the driver does not remove empty CIP packets.
Applications should detect empty packets by comparing adjacent packets' continuity
counters (DBC field of the CIP header).
IEC61883_XMIT
This command is used to transmit full frames and get more empty frames from the driver.
The argument is a pointer to the structure:
typedef struct iec61883_xmit {
int tx_handle; /* isoch handle */
int tx_flags; /* flags */
iec61883_xfer_t tx_xfer; /* xfer params */
int tx_miss_cnt; /* missed cycles */
} iec61883_xmit_t;
tx_flags should be set to zero.
The application sets xf_full_idx and xf_full_cnt to specify frames it wishes to transmit.
If there are no frames to transmit (e.g. the first time this command is called), xf_full_cnt
should be set to 0.
When the command returns, xf_empty_idx and xf_empty_cnt specifies empty frames
which can be to transmit more data. xf_error is always 0.

Device and Network Interfaces 369

iec61883(7I)

tx_miss_cnt contains the number of isochronous cycles missed since last transfer due to
data buffer under run. This can happen when an application does not supply data fast
enough.

For the purposes of time stamping, the driver considers the first packet in a frame buffer to
be the first packet of an AV frame.
IEC61883_PLUG_INIT
This command returns a handle for the specified plug. The argument is a pointer to the
structure:
typedef struct iec61883_plug_init {
int pi_ver; /* interface version */
int pi_loc; /* plug location */
int pi_type; /* plug type */
int pi_num; /* plug number */
int pi_flags; /* flags */
int pi_handle; /* plug handle */
int pi_rnum; /* plug number */
} iec61883_plug_init_t;

pi_ver should be set to IEC61883_V1_0.

pi_loc specifies plug location:

IEC61883_LOC_LOCAL
On the local unit (local plug). A plug control register (PCR) is allocated. Command fails
if the plug already exists
IEC61883_LOC_REMOTE
On the remote unit (remote plug). The plug should exist on the remote unit, otherwise
the command fails.

pi_type specifies isochronous plug type:

IEC61883_PLUG_IN IEC61883_PLUG_OUT
Input or output plugs.
IEC61883_PLUG_MASTER_IN IEC61883_PLUG_MASTER_OUT
Master input or master output plug. These plugs always exist on the local unit.

pi_num specifies plug number. This should be 0 for master plugs, and from 0 to 31 for
input/output plugs. Alternatively, a special value IEC61883_PLUG_ANY can be used to let the
driver choose a free plug number, create the plug and return the number in pi_rnum.

pi_flags should be set to 0.

If the command succeeds, pi_handle contains a handle that should be used with other plug
commands.

370 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 2009
 iec61883(7I)

IEC61883_PLUG_FINI
Argument is a handle returned by IEC61883_PLUG_INIT. This command frees any
resources associated with this handle, including the PCR.
IEC61883_PLUG_REG_READ
Read plug register value. The argument is a pointer to the structure:
typedef struct iec61883_plug_reg_val {
int pr_handle; /* plug handle */
uint32_t pr_val; /* register value */
} iec61883_plug_reg_val_t;
pr_handle is a handle returned by IEC61883_PLUG_INIT. Register value is returned in
pr_val.
IEC61883_PLUG_REG_CAS
Atomically compare and swap plug register value. The argument is a pointer to the
structure:
typedef struct iec61883_plug_reg_lock {
int pl_handle; /* plug handle */
uint32_t pl_arg; /* compare arg */
uint32_t pl_data; /* write value */
UINT32_t pl_old; /* original value */
} iec61883_plug_reg_lock_t;
pr_handle is a handle returned by IEC61883_PLUG_INIT.

Original register value is compared with pl_arg and if they are equal, register value is
replaced with pl_data. In any case, the original value is stored in pl_old.

The following commands only apply to asynchronous nodes:

IEC61883_ARQ_GET_IBUF_SIZE
This command returns current incoming ARQ buffer size. The argument is a pointer to
int.
IEC61883_ARQ_SET_IBUF_SIZE
This command changes incoming ARQ buffer size. The argument is the new buffer size in
bytes.
Files /dev/av/N/async Device node for asynchronous data
/dev/av/N/isoch Device has been disconnected
Errors EIO Bus operation failed.

DMA failure.
EFAULT ioctl(2) argument points to an illegal address.
EINVAL Invalid argument or argument combination.

Device and Network Interfaces 371

iec61883(7I)

ENODEV Device has been disconnected.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture All

Stability level Committed

See Also ioctl(2), mmap(2), open(2), poll(2), read(2), write(2), attributes(5), av1394(7D)

IEC 61883 Consumer audio/video equipment - Digital interface

IEEE Std 1394-1995 Standard for a High Performance Serial Bus

372 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 2009
 ieee1394(7D)

Name ieee1394, firewire, 1394 – Solaris IEEE-1394 Architecture

Description IEEE-1394 provides a means for interconnecting devices in computer and home
entertainment systems. (The IEEE-1394 architecture is also known as Firewire, an Apple
Computer trademark, and i.Link, a Sony trademark). The most common IEEE-1394 devices
are digital camcorders, mass-storage devices and cameras (including webcam-type devices).
For more information on USB, refer to the 1394 Trade Association website at
http://www.1394ta.org.

The Solaris IEEE-1394 architecture supports up to 63 hot-pluggable IEEE-1394 devices per

IEEE-1394 bus. The maximum data transfer rate is 400 Mbits, depending on the capabilities of
the attached device.

The Solaris IEEE-1394 architecture supports devices implementing a number of different

specifications. The basic behavior of the IEEE-1394 bus is described in the IEEE 1394-1995
and IEEE 1394a-2000 specifications.

IEEE-1394 host controllers implementing the 1394 Open Host Controller Interface
specification are supported. Camcorders implementing the IEC 61883 and 1394 Trade
Association AV/C specifications are supported. Mass-storage devices implementing the ANSI
SBP-2 specification are supported. Digital cameras implementing the 1394 Trade Association
1394-based Digital Camera (IIDC) specification are supported.

Files Listed below are drivers and modules which either utilize or are utilized by the Solaris
IEEE-1394 architecture. Drivers in /kernel/drv are 32 bit drivers (only). Drivers in
/kernel/drv/sparcv9 or /kernel/drv/amd64 are 64 bit drivers.

SUPPORT MODULE(S) FUNCTION

/kernel/misc/[sparcv9|amd64/]s1394 IEEE-1394 framework

/kernel/misc/[sparcv9|amd64/]sbp2 Serial Bus Protocol-2 (SBP-2)

TARGET DRIVER DEVICE CLASS

/kernel/drv/[sparcv9|amd64/]s1394 IEEE-1394 framework

/kernel/drv/[sparcv9|amd64/]scsa1394 mass storage class

/kernel/drv/[sparcv9|amd64/]av1394 camcorder (AV/C) class

/kernel/drv/[sparcv9|amd64/]dcam1394 digital camera (IIDC) class

HOST CONTROLLER INTERFACE DRIVER(S) DEVICE

/kernel/drv/[sparcv9|amd64/]hci1394 Open HCI

Device and Network Interfaces 373

ieee1394(7D)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNW1394, SUNW1394h, SUNWav1394,

SUNWscsa1394, SUNWsbp2, SUNWfwdc,
SUNWfwdcu

See Also attributes(5), av1394(7D), dcam1394(7D), hci1394(7D), scsa1394(7D), iec61883(7I)

http://www.sun.com

IEEE 1394a Specification – 1394 Trade Association, 2000

IEEE 1394 Specification – 1394 Trade Association, 1995

Notes Booting from IEEE-1394 mass-storage devices is not supported, but may be possible if
supported by the BIOS of the computer system.

374 man pages section 7: Device and Network Interfaces • Last Revised 3 Apr 2009
 ifb(7d)

Name ifb – 24-bit PCI color frame buffer and graphics accelerator driver

Description The ifb driver is the device driver for the Sun Expert3D graphics accelerators. The Expert3D
is a high resolution, high performance PCI graphics framebuffer providing hardware texture
mapping. The Expert3D also supports 1920x1200 double buffered, z-buffered display and
1280 x 1024 stereo.

The ifbdaemon process loads the ifb microcode at system startup time and during the resume
sequence of a suspend-resume cycle.
Files /dev/fbs/ifbn Device special file
/usr/lib/ifb.ucode ifb microcode
/usr/sbin/ifbdaemon ifb microcode loader

See Also SUNWifb_config(1M)

Device and Network Interfaces 375

ifp(7D)

Name ifp – ISP2100 Family Fibre Channel Host Bus Adapter Driver
Synopsis PCI SUNW,ifp@pci-slot

Description The ifp Host Bus Adapter is a SCSA compliant nexus driver for the Qlogic ISP2100/ISP2100A
chips. These chips support Fibre Channel Protocol for SCSI on Private Fibre Channel
Arbitrated loops.

The ifp driver interfaces with SCSI disk target driver, ssd(7D), and the SCSI-3 Enclosure
Services driver, ssd(7D). Only SCSI devices of type disk and ses are supported at present
time.

The ifp driver supports the standard functions provided by the SCSA interface. It supports
auto request sense (cannot be turned off) and tagged queueing by default. The driver requires
that all devices have unique hard addresses defined by switch settings in hardware. Devices
with conflicting hard addresses will not be accessible.
Files /kernel/drv/ifp ELF Kernel Module
/kernel/drv/sparcv9/ifp ELF Kernel Module (64–bit version)
/kernel/drv/ifp.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC

See Also luxadm(1M),prtconf(1M),driver.conf(4),attributes(5),ses(7D),ssd(7D)

Writing Device Drivers,

ANSI X3.272–1996, Fibre Channel Arbitrated Loop (FC-AL),

ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP),

ANSI X3.270-1996, SCSI-3 Architecture Model (SAM),

Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA),

ISP2100 Firmware Interface Specification, QLogic Corporation

Diagnostics The messages described below are some that may appear on the system console, as well as
being logged.

This first set of messages may be displayed while the ifp driver is initially trying to attach. All of
these messages mean that the ifp driver was unable to attach. These messages are preceded by
"ifp<number>", where "<number>" is the instance number of the ISP2100 Host Bus Adapter.

376 man pages section 7: Device and Network Interfaces • Last Revised 22 Jul 1998
 ifp(7D)

Device is using a hilevel intr, unused The device was configured with an interrupt level
that cannot be used with this ifp driver. Check the
device.
Failed to alloc soft state Driver was unable to allocate space for the internal
state structure. Driver did not attach to device; SCSI
devices will be inaccessible.
Bad soft state Driver requested an invalid internal state structure.
Driver did not attach to device; SCSI devices will be
inaccessible.
Unable to map pci config registers
Unable to map biu registers Driver was unable to map device registers; check for
bad hardware. Driver did not attach to device; SCSI
devices will be inaccessible.
Cannot alloc tran Driver was unable to obtain a transport handle to be
able to communicate with SCSA framework. Driver
did not attach to device; SCSI devices will be
inaccessible.
ddi_create_minor_node failed Driver was unable to create devctl minor node that
is used by luxadm(1M) for administering the loop.
Driver did not attach to device; SCSI devices will be
inaccessible.
Cannot alloc dma handle Driver was unable allocate a dma handle for
communicating with the Host Bus Adapter. Driver
did not attach to device; SCSI devices will be
inaccessible.
Cannot alloc cmd area Driver was unable allocate dma memory for request
and response queues. Driver did not attach to device;
SCSI devices will be inaccessible.
Cannot bind cmd area Driver was unable to bind dma handle to the cmd area.
Driver did not attach to device; SCSI devices will be
inaccessible.
Cannot alloc fcal handle Driver was unable allocate a dma handle for retrieving
loop map from the Host Bus Adapter. Driver did not
attach to device; SCSI devices will be inaccessible.
Cannot bind portdb Driver was unable to bind fcal port handle to the
memory used for obtaining port database. Driver did
not attach to device; SCSI devices will be inaccessible.

Device and Network Interfaces 377

ifp(7D)

scsi_hba_attach failed Driver was unable to attach to the SCSA framework.

Driver did not attach to device; SCSI devices will be
inaccessible.
Unable to create hotplug thread Driver was not able to create the kernel thread used
for hotplug support. Driver did not attach to device;
SCSI devices will be inaccessible.
Cannot add intr Driver was not able to add the interrupt routine to
the kernel. Driver did not attach to device; SCSI
devices will be inaccessible.
Unable to attach Driver was unable to attach to the hardware for some
reason that may be printed. Driver did not attach to
device; SCSI devices will be inaccessible.

The following set of messages may be display at any time. They will be printed with the full
device pathname followed by the shorter form described above.
Firmware checksum incorrect Firmware has an invalid checksum and will not be
downloaded.
Chip reset timeout ISP chip failed to reset in the time allocated; may be bad
hardware.
Stop firmware failed Stopping the firmware failed; may be bad hardware.
Load ram failed Unable to download new firmware into the ISP chip.
DMA setup failed The DMA setup failed in the host adapter driver on a
scsi_pkt. This will return TRAN_BADPKT to a SCSA target
driver.
Bad request pkt type
Bad request pkt
Bad request pkt hdr
Bad req pkt order The ISP Firmware rejected the packet as being set up
incorrectly. This will cause the ifp driver to call the target
completion routine with the reason of CMD_TRAN_ERR set
in the scsi_pkt. Check the target driver for correctly
setting up the packet.
Firmware error The ISP chip encountered a firmware error of some kind.
This error will cause the ifp driver to do error recovery
by resetting the chip.
DMA Failure (event) The ISP chip encountered a DMA error while reading
from the request queue (event is 8003) or writing to the

378 man pages section 7: Device and Network Interfaces • Last Revised 22 Jul 1998
 ifp(7D)

response queue (event is 8004). This error will cause the

ifp driver to do error recovery by resetting the chip.
Fatal error, resetting interface This is an indication that the ifp driver is doing error
recovery. This will cause all outstanding commands that
have been transported to the ifp driver to be completed
via the scsi_pkt completion routine in the target driver
with reason of CMD_RESET and status of STAT_BUS_RESET
set in the scsi_pkt.
target t, duplicate port wwns The driver detected target t to be having the same port
WWN as a different target; this is not supposed to
happen. Target t will become inaccessible.
target t, duplicate switch settings The driver detected devices with the same switch setting
t. All such devices will become inaccessible.
WWN changed on target t The World Wide Name (WWN) has changed on the
device with switch setting t.
target t, unknown device type dt The driver does not know the device type dt reported by
the device with switch setting t.

Device and Network Interfaces 379

if_tcp(7P)

Name if_tcp, if – general properties of Internet Protocol network interfaces

Description A network interface is a device for sending and receiving packets on a network. It is usually a
hardware device, although it can be implemented in software. Network interfaces used by the
Internet Protocol (IPv4 or IPv6) must be STREAMS devices conforming to the Data Link
Provider Interface (DLPI). See dlpi(7P).

Application An interface becomes available to IP when it is opened and the IP module is pushed onto the
Programming stream with the I_PUSH ioctl(2) command. (See streamio(7I)). The SIOCSLIFNAME ioctl(2)
Interface is issued to specify the name of the interface and to indicate whether it is IPv4 or IPv6. This
may be initiated by the kernel at boot time or by a user program after the system is running.
Each interface must be assigned an IP address with the SIOCSLIFADDR ioctl() before it can be
used. On interfaces where the network-to-link layer address mapping is static, only the
network number is taken from the ioctl() request; the remainder is found in a hardware
specific manner. On interfaces which provide dynamic network-to-link layer address
mapping facilities (for example, Ethernets using arp(7P)), the entire address specified in the
ioctl() is used. A routing table entry for destinations on the network of the interface is
installed automatically when an interface's address is set.

You cannot create IPMP IP interfaces using the procedure described above. Instead, use
ifconfig(1M).

ioctls The following ioctl() calls may be used to manipulate IP network interfaces. Unless
specified otherwise, the request takes an lifreq structure as its parameter. This structure has
the form:
struct lifreq {
#define LIFNAMSIZ 32
char lifr_name[LIFNAMSIZ]; /* if name, e.g. "le1" */
union {
int lifru_addrlen; /* for subnet/token etc */
uint_t lifru_ppa; /* SIOCSLIFNAME */
} lifr_lifru1;
union {
struct sockaddr_storage lifru_addr;
struct sockaddr_storage lifru_dstaddr;
struct sockaddr_storage lifru_broadaddr;
struct sockaddr_storage lifru_token; /* With lifr_addrlen */
struct sockaddr_storage lifru_subnet; /* With lifr_addrlen */
int lifru_index; /* interface index */
uint64_t lifru_flags; /* SIOC?LIFFLAGS */
int lifru_metric;
uint_t lifru_mtu;
int lif_muxid[2]; /* mux id’s for arp & ip */
struct lif_nd_req lifru_nd_req;
struct lif_ifinfo_req lifru_ifinfo_req;
zoneid_t lifru_zone; /* SIOC[GS]LIFZONE */

380 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 if_tcp(7P)

} lifr_lifru;

#define lifr_addrlen lifr_lifru1.lifru_addrlen

#define lifr_ppa lifr_lifru1.lifru_ppa /* Driver’s ppa */
#define lifr_addr lifr_lifru.lifru_addr /* address */
#define lifr_dstaddr lifr_lifru.lifru_dstaddr
#define lifr_broadaddr lifr_lifru.lifru_broadaddr /* broadcast addr. */
#define lifr_token lifr_lifru.lifru_token /* address token */
#define lifr_subnet lifr_lifru.lifru_subnet /* subnet prefix */
#define lifr_index lifr_lifru.lifru_index /* interface index */
#define lifr_flags lifr_lifru.lifru_flags /* flags */
#define lifr_metric lifr_lifru.lifru_metric /* metric */
#define lifr_mtu lifr_lifru.lifru_mtu /* mtu */
#define lifr_ip_muxid lifr_lifru.lif_muxid[0]
#define lifr_arp_muxid lifr_lifru.lif_muxid[1]
#define lifr_nd lifr_lifru.lifru_nd_req /* SIOCLIF*ND */
#define lifr_ifinfo lifr_lifru.lifru_ifinfo_req /* SIOC[GS]LIFLNKINFO */
#define lifr_zone lifr_lifru.lifru_zone /* SIOC[GS]LIFZONE */
};
SIOCSLIFADDR Set interface address.
SIOCGLIFADDR Get interface address.
SIOCSLIFDSTADDR Set point to point address for interface.
SIOCGLIFDSTADDR Get point to point address for interface.
SIOCSLIFFLAGS Set interface flags field. If the interface is marked down, any processes
currently routing packets through the interface are notified.
SIOCGLIFFLAGS Get interface flags.
SIOCGLIFCONF Get interface configuration list. This request takes a lifconf structure
(see below) as a value-result parameter. The lifc_family field can be
set to AF_UNSPEC to retrieve both AF_INET and AF_INET6 interfaces. The
lifc_len field should be set to the size of the buffer pointed to by
lifc_buf.

The lifc_flags field should usually be set to zero, but callers that need
low-level knowledge of the underlying IP interfaces that comprise an
IPMP group can set it to LIFC_UNDER_IPMP to request that those
interfaces be included in the result. Upon success, lifc_len contains
the length, in bytes, of the array of lifreq structures pointed to by
lifc_req. For each lifreq structure, the lifr_name and lifr_addr
fields are valid.

Device and Network Interfaces 381

if_tcp(7P)

SIOCGLIFNUM Get number of interfaces. This request returns an integer which is the
number of interface descriptions (struct lifreq) returned by the
SIOCGLIFCONF ioctl (in other words, indicates how large lifc_len
must be).

This request takes a struct lifnum (see below) as a value-result

parameter. The lifn_family field can be set to AF_UNSPEC to count
both AF_INET and AF_INET6 interfaces. The lifn_flags field should
usually be set to zero, but callers that need low-level knowledge of the
underlying IP interfaces that comprise an IPMP group can set it to
LIFC_UNDER_IPMP to request that those interfaces be included in the
count.
SIOCSLIFMTU Set the maximum transmission unit (MTU) size for interface. Place the
request in the lifru_mtu field. The MTU can not exceed the physical
MTU limitation (which is reported in the DLPI DL_INFO_ACK message).
SIOCGLIFMTU Get the maximum transmission unit size for interface.
SIOCSLIFMETRIC Set the metric associated with the interface. The metric is used by
routing daemons such as in.routed(1M).
SIOCGLIFMETRIC Get the metric associated with the interface.
SIOCGLIFMUXID Get the ip and arp muxid associated with the interface.
SIOCSLIFMUXID Set the ip and arp muxid associated with the interface.
SIOCGLIFINDEX Get the interface index associated with the interface.
SIOCSLIFINDEX Set the interface index associated with the interface.
SIOCGLIFZONE Get the zone associated with the interface.
SIOCSLIFZONE Set the zone associated with the interface. Only applies for zones that
use the shared-IP instance.
SIOCLIFADDIF Add a new logical interface on a physical interface using an unused
logical interface number.
SIOCLIFREMOVEIF Remove a logical interface by specifying its IP address or logical
interface name.
SIOCSLIFTOKEN Set the address token used to form IPv6 link-local addresses and for
stateless address autoconfiguration.
SIOCGLIFTOKEN Get the address token used to form IPv6 link-local addresses and for
stateless address autoconfiguration.
SIOCSLIFSUBNET Set the subnet prefix associated with the interface.
SIOCGLIFSUBNET Get the subnet prefix associated with the interface.

382 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 if_tcp(7P)

SIOCSLIFLNKINFO Set link specific parameters for the interface.

SIOCGLIFLNKINFO Get link specific parameters for the interface.
SIOCLIFDELND Delete a neighbor cache entry for IPv6.
SIOCLIFGETND Get a neighbor cache entry for IPv6.
SIOCLIFSETND Set a neighbor cache entry for IPv6.
SIOCSLIFUSESRC Set the interface from which to choose a source address. The
lifr_index field has the interface index corresponding to the interface
whose address is to be used as the source address for packets going out
on the interface whose name is provided by lifr_name. If the
lifr_index field is set to zero, the previous setting is cleared. See
ifconfig(1M) for examples of the usesrc option.
SIOCGLIFUSESRC Get the interface index of the interface whose address is used as the
source address for packets going out on the interface provided by
lifr_name field. The value is retrieved in the lifr_index field. See
ifconfig(1M) for examples of the usesrc option.
SIOCGLIFSRCOF Get the interface configuration list for interfaces that use an address
hosted on the interface provided by the lifs_ifindex field in the
lifsrcof struct (see below), as a source address. The application sets
lifs_maxlen to the size (in bytes) of the buffer it has allocated for the
data. On return, the kernel sets lifs_len to the actual size required.
Note, the application could set lifs_maxlen to zero to query the kernel
of the required buffer size instead of estimating a buffer size. The
application tests lifs_len <= lifs_maxlen -- if that's true, the buffer
was big enough and the application has an accurate list. If it is false, it
needs to allocate a bigger buffer and try again, and lifs_len provides a
hint of how big to make the next trial. See ifconfig(1M) for examples
of the usesrc option.
SIOCTONLINK Test if the address is directly reachable, for example, that it can be
reached without going through a router. This request takes an
sioc_addrreq structure (see below) as a value-result parameter. The
sa_addr field should be set to the address to test. The sa_res field will
contain a non-zero value if the address is onlink.
SIOCTMYADDR Test if the address is assigned to this node. This request takes an
sioc_addrreq structure (see below) as a value-result parameter. The
sa_addr field should be set to the address to test. The sa_res field will
contain a non-zero value if the address is assigned to this node.
SIOCTMYSITE Test if the address is part of the same site as this node. This request
takes an sioc_addrreq structure (see below) as a value-result

Device and Network Interfaces 383

if_tcp(7P)

parameter. The sa_addr field should be set to the address to test. The
sa_res field will contain a non-zero value if the address is in the same
site.

The structure used by SIOCGLIFCONF has the form:

struct lifconf {
sa_family_t lifc_family;
int lifc_flags; /* request specific
/* interfaces */
int lifc_len; /* size of assoc. buffer */
union {
caddr_t lifcu_buf;
struct lifreq *lifcu_req;
} lifc_lifcu;

#define lifc_buf lifc_lifcu.lifcu_buf /* buffer address */

#define lifc_req lifc_lifcu.lifcu_req /* array of structs returned */
};

The structure used by SIOCGLIFNUM has the form:

struct lifnum {
sa_family_t lifn_family;
int lifn_flags; /* req. specf. interfaces */
int lifn_count; /* Result */
};

The structure used by SIOCTONLINK, SIOCTMYADDR and SIOCTMYSITE has the form:
struct sioc_addrreq {
struct sockaddr_storage sa_addr; /* Address to test */
int sa_res; /* Result - 0/1 */
};

The structure used by SIOCGLIFSRCOF has the form:

struct lifsrcof {
uint_t lifs_ifindex; /* addr on this interface */
/* used as the src addr */
size_t lifs_maxlen; /* size of buffer: input */
size_t lifs_len; /* size of buffer: output */
union {
caddr_t lifsu_buf;
struct lifreq *lifsu_req;
} lifs_lifsu;
#define lifs_buf lifs_lifsu.lifsu_buf /* buffer addr. */
#define lifs_req lifs_lifsu.lifsu_req /* array returned */
};

384 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 if_tcp(7P)

The following ioctl() calls are maintained for compatibility but only apply to IPv4 network
interfaces, since the data structures are too small to hold an IPv6 address. Unless specified
otherwise, the request takes an ifreq structure as its parameter. This structure has the form:
struct ifreq {
#define IFNAMSIZ 16
char ifr_name[IFNAMSIZ]; /* interface name - e.g. "hme0" */
union {
struct sockaddr ifru_addr;
struct sockaddr ifru_dstaddr;
struct sockaddr ifru_broadaddr;
short ifru_flags;
int ifru_metric;
int if_muxid[2]; /* mux id’s for arp and ip */
int ifru_index; /* interface index */
} ifr_ifru;

#define ifr_addr ifr_ifru.ifru_addr /* address */

#define ifr_dstaddr ifr_ifru.ifru_dstaddr /* other end of p-to-p link */
#define ifr_broadaddr ifr_ifru.ifru_broadaddr /* broadcast address */
#define ifr_flags ifr_ifru.ifru_flags /* flags */
#define ifr_index ifr_ifru.ifru_index /* interface index */
#define ifr_metric ifr_ifru.ifru_metric /* metric */
};
SIOCSIFADDR Set interface address.
SIOCGIFADDR Get interface address.
SIOCSIFDSTADDR Set point to point address for interface.
SIOCGIFDSTADDR Get point to point address for interface.
SIOCSIFFLAGS Set interface flags field. If the interface is marked down, any processes
currently routing packets through the interface are notified.
SIOCGIFFLAGS Get interface flags.
SIOCGIFCONF Get interface configuration list. This request takes an ifconf structure
(see below) as a value-result parameter. The ifc_len field should be set
to the size of the buffer pointed to by ifc_buf. Upon success, ifc_len
will contain the length, in bytes, of the array of ifreq structures pointed
to by ifc_req. For each ifreq structure, the ifr_name and ifr_addr
fields are valid. Though IPMP IP interfaces are included in the array,
underlying IP interfaces that comprise those IPMP groups are not.
SIOCGIFNUM Get number of interfaces. This request returns an integer which is the
number of interface descriptions (struct ifreq) returned by the
SIOCGIFCONF ioctl (in other words, indicates how large ifc_len must

Device and Network Interfaces 385

if_tcp(7P)

be). Though IPMP IP interfaces are included in the array, underlying IP

interfaces that comprise those IPMP groups are not.
SIOCSIFMTU Set the maximum transmission unit (MTU) size for interface. Place the
request in the ifr_metric field. The MTU has to be smaller than
physical MTU limitation (which is reported in the DLPI DL_INFO_ACK
message).
SIOCGIFMTU Get the maximum transmission unit size for interface. Upon success, the
request is placed in the ifr_metric field.
SIOCSIFMETRIC Set the metric associated with the interface. The metric is used by
routine daemons such as in.routed(1M).
SIOCGIFMETRIC Get the metric associated with the interface.
SIOCGIFMUXID Get the ip and arp muxid associated with the interface.
SIOCSIFMUXID Set the ip and arp muxid associated with the interface.
SIOCGIFINDEX Get the interface index associated with the interface.
SIOCSIFINDEX Set the interface index associated with the interface.

The ifconf structure has the form:

struct ifconf {
int ifc_len; /* size of assoc. buffer */
union {
caddr_t ifcu_buf;
struct ifreq *ifcu_req;
} ifc_ifcu;

#define ifc_buf ifc_ifcu.ifcu_buf /* buffer address */

#define ifc_req ifc_ifcu.ifcu_req /* array of structs returned */
};

IFF_ Flags You can use the ifconfig(1M) command to display the IFF_ flags listed below (with the
leading IFF_ prefix removed). See the ifconfig(1M) manpage for a definition of each flag.
#define IFF_UP 0x0000000001 /* Address is up */
#define IFF_BROADCAST 0x0000000002 /* Broadcast address valid */
#define IFF_DEBUG 0x0000000004 /* Turn on debugging */
#define IFF_LOOPBACK 0x0000000008 /* Loopback net */

#define IFF_POINTOPOINT 0x0000000010 /* Interface is p-to-p */

#define IFF_NOTRAILERS 0x0000000020 /* Avoid use of trailers */
#define IFF_RUNNING 0x0000000040 /* Resources allocated */
#define IFF_NOARP 0x0000000080 /* No address res. protocol */

#define IFF_PROMISC 0x0000000100 /* Receive all packets */

386 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 if_tcp(7P)

#define IFF_ALLMULTI 0x0000000200 /* Receive all multicast pkts */

#define IFF_INTELLIGENT 0x0000000400 /* Protocol code on board */
#define IFF_MULTICAST 0x0000000800 /* Supports multicast */

#define IFF_MULTI_BCAST 0x0000001000 /* Multicast using broadcst. add. */

#define IFF_UNNUMBERED 0x0000002000 /* Non-unique address */
#define IFF_DHCPRUNNING 0x0000004000 /* DHCP controls interface */
#define IFF_PRIVATE 0x0000008000 /* Do not advertise */

#define IFF_NOXMIT 0x0000010000 /* Do not transmit pkts */

#define IFF_NOLOCAL 0x0000020000 /* No address - just on-link subnet */
#define IFF_DEPRECATED 0x0000040000 /* Address is deprecated */
#define IFF_ADDRCONF 0x0000080000 /* Addr. from stateless addrconf */

#define IFF_ROUTER 0x0000100000 /* Router on interface */

#define IFF_NONUD 0x0000200000 /* No NUD on interface */
#define IFF_ANYCAST 0x0000400000 /* Anycast address */
#define IFF_NORTEXCH 0x0000800000 /* Don’t xchange rout. info */

#define IFF_IPV4 0x0001000000 /* IPv4 interface */

#define IFF_IPV6 0x0002000000 /* IPv6 interface */
#define IFF_NOFAILOVER 0x0008000000 /* in.mpathd test address */
#define IFF_FAILED 0x0010000000 /* Interface has failed */

#define IFF_STANDBY 0x0020000000 /* Interface is a hot-spare */

#define IFF_INACTIVE 0x0040000000 /* Functioning but not used */
#define IFF_OFFLINE 0x0080000000 /* Interface is offline */
#define IFF_COS_ENABLED 0x0200000000 /* If CoS marking is supported

#define IFF_COS_ENABLED 0x0200000000 /* If CoS marking is supported */

#define IFF_PREFERRED 0x0400000000 /* Prefer as source address */
#define IFF_TEMPORARY 0x0800000000 /* RFC3041 */
#define IFF_FIXEDMTU 0x1000000000 /* MTU set with SIOCSLIFMTU */

#define IFF_VIRTUAL 0x2000000000 /* Cannot send/receive pkts */

#define IFF_DUPLICATE 0x4000000000 /* Local address in use */
#define IFF_IPMP 0x8000000000 /* IPMP IP interface */
Errors EPERM Calling process has insufficient privileges.
ENXIO The lifr_name member of the lifreq structure contains an invalid value.

For SIOCGLIFSRCOF, the lifs_ifindex member of the lifsrcof structure

contains an invalid value.

For SIOCSLIFUSESRC, this error is returned if the lifr_index is set to an invalid

value.
EBADADDR Wrong address family or malformed address.

Device and Network Interfaces 387

if_tcp(7P)

EINVAL For SIOCSLIFMTU, this error is returned when the requested MTU size is invalid.
This error indicates the MTU size is greater than the MTU size supported by the
DLPI provider or less than 68 (for IPv4) or less than 1280 (for IPv6).

For SIOCSLIFUSESRC, this error is returned if either the lifr_index or

lifr_name identify interfaces that are already part of an existing IPMP group.
EEXIST For SIOCLIFADDIF, this error is returned if the lifr_name member in the
lifreq structure corresponds to an interface that already has the PPA specified
by lifr_ppa plumbed.

See Also ifconfig(1M), in.routed(1M), ioctl(2), streamio(7I), arp(7P), dlpi(7P), ip(7P), ip6(7P)

388 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 igb(7D)

Name igb – Intel 82575 1Gb PCI Express NIC Driver

Synopsis /dev/igb*

Description The igb Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on Intel 82575
Gigabit Ethernet controllers.

The igb driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting.

The igb driver and hardware support auto-negotiation, a protocol specified by the 1000
Base-T standard. Auto-negotiation allows each device to advertise its capabilities and discover
those of its peer (link partner). The highest common denominator supported by both link
partners is automatically selected, yielding the greatest available throughput, while requiring
no manual configuration. The igb driver also allows you to configure the advertised
capabilities to less than the maximum (where the full speed of the interface is not required), or
to force a specific mode of operation, irrespective of the link partner's advertised capabilities.

Application The cloning character-special device, /dev/igb, is used to access all Intel 82575 Gigabit
Programming devices installed within the system.
Interface
The igb driver is managed by the dladm(1M) command line utility, which allows VLANs to be
defined on top of igb instances and for igb instances to be aggregated. See dladm(1M) for
more details.

You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number for the system. The device is initialized on first attach and de-initialized (stopped) at
last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to your
DL_INFO_REQ are:
■ Maximum SDU is 9000.
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP (Service Access Point) length value is -2, meaning the physical address component is
followed immediately by a 2-byte SAP component within the DLSAP address.
■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Device and Network Interfaces 389

igb(7D)

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a

particular SAP with the stream.

Configuration By default, the igb driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can be any one of the following, (as described in the IEEE803.2 standard):

1000 Mbps, full-duplex.

100 Mbps, full-duplex.

100 Mbps, half-duplex.

10 Mbps, full-duplex.

10 Mbps, half-duplex.

The auto-negotiation protocol automatically selects speed (1000 Mbps, 100 Mbps, or 10
Mbps) and operation mode (full-duplex or half-duplex) as the highest common denominator
supported by both link partners.

Alternatively, you can set the capabilities advertised by the igb device using ndd(1M). The
driver supports a number of parameters whose names begin with adv_ (see below). Each of
these parameters contains a boolean value that determines if the device advertises that mode
of operation. For example, the adv_1000fdx_cap parameter indicates if 1000M full duplex is
advertised to link partner. The adv_autoneg cap parameter controls whether auto-negotiation
is performed. If adv_autoneg_cap is set to 0, the driver forces the mode of operation selected
by the first non-zero parameter in priority order as shown below:
(highest priority/greatest throughput)
adv_1000fdx_cap 1000Mbps full duplex
adv_100fdx_cap 100Mpbs full duplex
adv_100hdx_cap 100Mbps half duplex
adv_10fdx_cap 10Mpbs full duplex
adv_10hdx_cap 10Mpbs half duplex
(lowest priority/least throughput)

All capabilities default to enabled. Note that changing any capability parameter causes the link
to go down while the link partners renegotiate the link speed/duplex using the newly changed
capabilities.
Files /dev/igb* Special character device.
/kernel/drv/igb 32–bit device driver (x86).
/kernel/drv/amd64/igb 64–bit device driver (x86).
/kernel/drv/sparcv9/igb 64–bit device driver (SPARC).
/kernel/drv/igb.conf Configuration file.

390 man pages section 7: Device and Network Interfaces • Last Revised 20 Jul 2007
 igb(7D)

See Also dladm(1M), ndd(1M), netstat(1M), driver.conf(4), attributes(5), streamio(7I),

dlpi(7P),

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

Device and Network Interfaces 391

ii(7D)

Name ii – Instant Image control device

Description The ii device is a control interface for Instant Image devices and controls the Instant Image
module through the ioctl(2) interface.

Instant Image is a point-in-time volume copy facility for the Solaris operating environment
that is administered through the iiadm(1M) command. With Instant Image, you can create an
independent point-in-time copy of a volume or a master volume-dependent point-in-time
view. You can also independently access the master and shadow volume for read and write
operations. Instant Image also lets you update the shadow volume from the master volume or
restore the master volume from the shadow. (Restore operations to volumes can be full or
incremental). Instant Image supports fast volume re-synchronization, letting you create a new
point-in-time volume copy by updating the specified volume with only changed data.

To create a shadow volume you need a:

1. Master volume to be shadowed.
2. Shadow volume where the copy will reside. This volume must be equal to or larger than the
master volume.
3. Administrative bitmap volume or file for tracking differences between the shadow and
master volumes. The administrative bitmap volume or file must be at least 24Kbytes in size
and requires 8KBytes for each GByte (or part thereof) of master volume size, plus an
additional 8KBytes overhead. For example, to shadow a 3GByte master volume, the
administration volume must be 8KBytes + (3 * 8KBytes) =32KBytes in size.

The Instant Image module uses services provided by the SDBC and SD_GEN modules. The
SV module is required to present a conventional block device interface to the storage product
interface of the Instant Image, SDBC and SD_GEN modules.

When a shadow operation is suspended or resumed, the administration volumes may be

stored in permanent SDBC storage or loaded and saved to and from kernel memory. The
ii_bitmap variable in the /kernel/drv/ii.conf configuration file determines the
administration volume storage type. A value of 0 indicates kernel memory, while a value of 1
indicates permanent SDBC storage. If the system is part of a storage products cluster, use the 1
value (permanent storage), otherwise use kernel memory (0 value).
Files kernel/drv/ii 32– bit ELF kernel module (x86).
/kernel/drv/ii.conf Configuration file.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

392 man pages section 7: Device and Network Interfaces • Last Revised 8 Jun 2007
 ii(7D)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWiu

Interface Stability Committed

See Also iiadm(1M), ioctl(2), attributes(5), sv(7D)

Device and Network Interfaces 393

inet6(7P)

Name inet6 – Internet protocol family for Internet Protocol version 6

Synopsis #include <sys/types.h>
#include <netinet/in.h>

Description The inet6 protocol family implements a collection of protocols that are centered around the
Internet Protocol version 6 (IPv6) and share a common address format. The inet6 protocol
family can be accessed using the socket interface, where it supports the SOCK_STREAM,
SOCK_DGRAM, and SOCK_RAW socket types, or the Transport Level Interface (TLI), where it
supports the connectionless (T_CLTS) and connection oriented (T_COTS_ORD) service types.

Protocols The Internet protocol family for IPv6 included the Internet Protocol Version 6 (IPv6), the
Neighbor Discovery Protocol (NDP), the Internet Control Message Protocol (ICMPv6), the
Transmission Control Protocol (TCP), and the User Datagram Protocol (UDP).

TCP supports the socket interface's SOCK_STREAM abstraction and TLI's T_COTS_ORD service
type. UDP supports the SOCK_DGRAM socket abstraction and the TLI T_CLTS service type. See
tcp(7P) and udp(7P). A direct interface to IPv6 is available using the socket interface. See
ip6(7P). ICMPv6 is used by the kernel to handle and report errors in protocol processing. It is
also accessible to user programs. See icmp6(7P). NDP is used to translate 128-bit IPv6
addresses into 48–bit Ethernet addresses.

IPv6 addresses come in three types: unicast, anycast, and multicast. A unicast address is an
identifier for a single network interface. An anycast address is an identifier for a set of
interfaces; a packet sent to an anycast address is delivered to the "nearest" interface identified
by that address, pursuant to the routing protocol's measure of distance. A multicast address is
an identifier for a set of interfaces; a packet sent to a multicast address is delivered to all
interfaces identified by that address. There are no broadcast addresses as such in IPv6; their
functionality is superseded by multicast addresses.

For IPv6 addresses, there are three scopes within which unicast addresses are guaranteed to be
unique. The scope is indicated by the address prefix. The three varieties are link-local (the
address is unique on that physical link), site-local (the address is unique within that site), and
global (the address is globally unique).

The three highest order bits for global unicast addresses are set to 001. The ten highest order
bits for site-local addresses are set to 1111 1110 11. The ten highest order bits for link-local
addresses are set to 1111 1110 11. For multicast addresses, the eight highest order bits are set
to 1111 1111. Anycast addresses have the same format as unicast addresses.

IPv6 addresses do not follow the concept of "address class" seen in IP.

A global unicast address is divided into the following segments:

■ The first three bits are the Format Prefix identifying a unicast address.
■ The next 13 bits are the Top-Level Aggregation (TLA) identifier. For example, the
identifier could specify the ISP.

394 man pages section 7: Device and Network Interfaces • Last Revised 3 Oct 2002
 inet6(7P)

■ The next eight bits are reserved for future use.

■ The next 24 bits are the Next-Level Aggregation (NLA) identifier.
■ The next 16 bits are the Site-Level Aggregation (SLA) identifier.
■ The last 64 bits are the interface ID. This will most often be the hardware address of the
link in IEEE EUI-64 format.

Link-local unicast addresses are divided in this manner:

■ The first ten bits are the Format Prefix identifying a link-local address.
■ The next 54 bits are zero.
■ The last 64 bits are the interface ID. This will most often be the hardware address of the
link in IEEE EUI-64 format.

Site-local unicast addresses are divided in this manner:

■ The first ten bits are the Format Prefix identifying a site-local address.
■ The next 38 bits are zero.
■ The next 16 bits are the subnet ID.
■ The last 64 bits are the interface ID. This will most often be the hardware address of the
link in IEEE EUI-64 format.

Addressing IPv6 addresses are sixteen byte quantities, stored in network byte order. The socket API uses
the sockaddr_in6 structure when passing IPv6 addresses between an application and the
kernel. The sockaddr_in6 structure has the following members:
sa_familty_t sin6_family;
in_port_t sin6_port;
uint32_t sin6_flowinfo;
struct in6_addr sin6_addr;
uint32_t sin6_scope_id;
uint32_t __sin6_src_id;

Library routines are provided to manipulate structures of this form. See inet(3SOCKET).

The sin6_addr field of the sockaddr_in6 structure specifies a local or remote IPv6 address.
Each network interface has one or more IPv6 addresses configured, that is, a link-local
address, a site-local address, and one or more global unicast IPv6 addresses. The special value
of all zeros may be used on this field to test for "wildcard" matching. Given in a
bind(3SOCKET) call, this value leaves the local IPv6 address of the socket unspecified, so that
the socket will receive connections or messages directed at any of the valid IPv6 addresses of
the system. This can prove useful when a process neither knows nor cares what the local IPv6
address is, or when a process wishes to receive requests using all of its network interfaces.

The sockaddr_in6 structure given in the bind() call must specify an in6_addr value of either
all zeros or one of the system's valid IPv6 addresses. Requests to bind any other address will

Device and Network Interfaces 395

inet6(7P)

elicit the error EADDRNOTAVAI. When a connect(3SOCKET) call is made for a socket that has a
wildcard local address, the system sets the sin6_addr field of the socket to the IPv6 address of
the network interface through which the packets for that connection are routed.
The sin6_port field of the sockaddr_in6 structure specifies a port number used by TCP or
UDP. The local port address specified in a bind() call is restricted to be greater than
IPPORT_RESERVED (defined in <netinet/in.h>) unless the creating process is running as the
super-user, providing a space of protected port numbers. In addition, the local port address
cannot be in use by any socket of the same address family and type. Requests to bind sockets to
port numbers being used by other sockets return the error EADDRINUSE. If the local port
address is specified as 0, the system picks a unique port address greater than
IPPORT_RESERVED. A unique local port address is also selected when a socket which is not
bound is used in a connect(3SOCKET) or sendto() call. See send(3SOCKET). This allows
programs that do not care which local port number is used to set up TCP connections by
simply calling socket(3SOCKET) and then connect(3SOCKET), and then sending UDP
datagrams with a socket() call followed by a sendto() call.
Although this implementation restricts sockets to unique local port numbers, TCP allows
multiple simultaneous connections involving the same local port number so long as the
remote IPv6 addresses or port numbers are different for each connection. Programs may
explicitly override the socket restriction by setting the SO_REUSEADDR socket option with
setsockopt(). See getsockopt(3SOCKET).
In addition, the same port may be bound by two separate sockets if one is an IP socket and the
other an IPv6 socket.
TLI applies somewhat different semantics to the binding of local port numbers. These
semantics apply when Internet family protocols are used using the TLI.
Source Address IPv6 source address selection is done on a per destination basis, and utilizes a list of rules from
Selection which the best source address is selected from candidate addresses. The candidate set
comprises a set of local addresses assigned on the system which are up and not anycast. If just
one candidate exists in the candidate set, it is selected.
Conceptually, each selection rule prefers one address over another, or determines their
equivalence. If a rule produces a tie, a subsequent rule is used to break the tie.
The sense of some rules may be reversed on a per-socket basis using the
IPV6_SRC_PREFERENCES socket option (see ip6(7P)). The flag values for this option are
defined in <netinet/in.h> and are referenced in the description of the appropriate rules
below.
As the selection rules indicate, the candidate addresses are SA and SB and the destination is D.
Prefer the same address If SA == D, prefer SA. If SB == D, prefer SB.
Prefer appropriate scope Here, Scope(X) is the scope of X according to the IPv6
Addressing Architecture.

396 man pages section 7: Device and Network Interfaces • Last Revised 3 Oct 2002
 inet6(7P)

If Scope(SA) < Scope(SB): If Scope(SA) < Scope(D), then

prefer SB and otherwise prefer SA.

If Scope(SB) < Scope(SA): If Scope(SB) < Scope(D), then

prefer SA and otherwise prefer SB.
Avoid deprecated addresses If one of the addresses is deprecated (IFF_DEPRECATED)
and the other is not, prefer the one that isn't deprecated.
Prefer preferred addresses If one of the addresses is preferred (IFF_PREFERRED) and
the other is not, prefer the one that is preferred.
Prefer outgoing interface If one of the addresses is assigned to the interface that will be
used to send packets to D and the other is not, then prefer the
former.
Prefer matching label This rule uses labels which are obtained through the IPv6
default address selection policy table. See ipaddrsel(1M) for
a description of the default contents of the table and how the
table is configured.

If Label(SA) == Label(D) and Label(SB) != Label(D), then

prefer SA.

If Label(SB) == Label(D) and Label(SA) != Label(D), then

prefer SB.
Prefer public addresses This rule prefers public addresses over temporary addresses,
as defined in RFC 3041. Temporary addresses are disabled by
default and may be enabled by appropriate settings in
ndpd.conf(4).

The sense of this rule may be set on a per-socket basis using

the IPV6_SRC_PREFERENCES socket option. Passing the
flag IPV6_PREFER_SRC_TMP or
IPV6_PREFER_SRC_PUBLIC will cause temporary or public
addresses to be preferred, respectively, for that particular
socket. See ip6(7P) for more information about IPv6 socket
options.
Use longest matching prefix. This rule prefers the source address that has the longer
matching prefix with the destination. Because this is the last
rule and because both source addresses could have equal
matching prefixes, this rule does an xor of each source
address with the destination, then selects the source address
with the smaller xor value in order to break any potential tie.

If SA ^ D < SB ^ D, then prefer SA.

Device and Network Interfaces 397

inet6(7P)

If SB ^ D < SA ^ D, then prefer SB.

Applications can override this algorithm by calling bind(3SOCKET) and specifying an

address.

See Also ioctl(2), bind(3SOCKET), connect(3SOCKET), getipnodebyaddr(3SOCKET),

getipnodebyname(3SOCKET),getprotobyname(3SOCKET), getservbyname(3SOCKET),
getsockopt(3SOCKET), inet(3SOCKET), send(3SOCKET), icmp6(7P), ip6(7P), tcp(7P),
udp(7P)

Conta, A. and Deering, S., Internet Control Message Protocol (ICMPv6) for the Internet
Protocol Version 6 (IPv6) Specification, RFC 1885, December 1995.

Deering, S. and Hinden, B., Internet Protocol, Version 6 (IPv6) Specification, RFC 1883,
December 1995.

Hinden, B. and Deering, S., IP Version 6 Addressing Architecture, RFC 1884, December 1995.

Draves, R., RFC 3484, Default Address Selection for IPv6. The Internet Society. February 2003.

Narten, T., and Draves, R. RFC 3041, Privacy Extensions for Stateless Address
Autoconfiguration in IPv6. The Internet Society. January 2001.

Notes The IPv6 support is subject to change as the Internet protocols develop. Users should not
depend on details of the current implementation, but rather the services exported.

398 man pages section 7: Device and Network Interfaces • Last Revised 3 Oct 2002
 inet(7P)

Name inet – Internet protocol family

Synopsis #include <sys/types.h>
#include <netinet/in.h>

Description The Internet protocol family implements a collection of protocols which are centered around
the Internet Protocol (“IP”) and which share a common address format. The Internet family
protocols can be accessed using the socket interface, where they support the SOCK_STREAM,
SOCK_DGRAM, and SOCK_RAW socket types, or the Transport Level Interface (TLI), where they
support the connectionless (T_CLTS) and connection oriented (T_COTS_ORD) service types.

Protocols The Internet protocol family is comprised of the Internet Protocol (“IP”), the Address
Resolution Protocol (“ARP”), the Internet Control Message Protocol (“ICMP”), the
Transmission Control Protocol (“TCP”), and the User Datagram Protocol (“UDP”).

TCP supports the socket interface's SOCK_STREAM abstraction and TLI's T_COTS_ORD service
type. UDP supports the SOCK_DGRAM socket abstraction and the TLI T_CLTS service type. See
tcp(7P) and udp(7P). A direct interface to IP is available using both TLI and the socket
interface (see ip(7P)). ICMP is used by the kernel to handle and report errors in protocol
processing. It is also accessible to user programs (see icmp(7P)). ARP is used to translate 32-bit
IP addresses into 48-bit Ethernet addresses. See arp(7P).

The 32-bit IP address is divided into network number and host number parts. It is
frequency-encoded. The most-significant bit is zero in Class A addresses, in which the
high-order 8 bits represent the network number. Class B addresses have their high order two
bits set to 10 and use the high-order 16 bits as the network number field. Class C addresses
have a 24-bit network number part of which the high order three bits are 110. Sites with a
cluster of IP networks may chose to use a single network number for the cluster; this is done
by using subnet addressing. The host number portion of the address is further subdivided into
subnet number and host number parts. Within a subnet, each subnet appears to be an
individual network. Externally, the entire cluster appears to be a single, uniform network
requiring only a single routing entry. Subnet addressing is enabled and examined by the
following ioctl(2) commands. They have the same form as the SIOCSIFADDR command.
SIOCSIFNETMASK Set interface network mask. The network mask defines the network part
of the address; if it contains more of the address than the address type
would indicate, then subnets are in use.
SIOCGIFNETMASK Get interface network mask.

Addressing IP addresses are four byte quantities, stored in network byte order. IP addresses should be
manipulated using the byte order conversion routines. See byteorder(3SOCKET).

Addresses in the Internet protocol family use the sockaddr_in structure, which has that
following members:

Device and Network Interfaces 399

inet(7P)

short sin_family;
ushort_t sin_port;
struct in_addr sin_addr;
char sin_zero[8];

Library routines are provided to manipulate structures of this form; See inet(3SOCKET).

The sin_addr field of the sockaddr_in structure specifies a local or remote IP address. Each
network interface has its own unique IP address. The special value INADDR_ANY may be used in
this field to effect “wildcard” matching. Given in a bind(3SOCKET) call, this value leaves the
local IP address of the socket unspecified, so that the socket will receive connections or
messages directed at any of the valid IP addresses of the system. This can prove useful when a
process neither knows nor cares what the local IP address is or when a process wishes to
receive requests using all of its network interfaces. The sockaddr_in structure given in the
bind(3SOCKET) call must specify an in_addr value of either INADDR_ANY or one of the
system's valid IP addresses. Requests to bind any other address will elicit the error
EADDRNOTAVAIL. When a connect(3SOCKET) call is made for a socket that has a wildcard
local address, the system sets the sin_addr field of the socket to the IP address of the network
interface that the packets for that connection are routed through.

The sin_port field of the sockaddr_in structure specifies a port number used by TCP or
UDP. The local port address specified in a bind(3SOCKET) call is restricted to be greater than
IPPORT_RESERVED (defined in <<netinet/in.h>>) unless the creating process is running as
the superuser, providing a space of protected port numbers. In addition, the local port address
must not be in use by any socket of same address family and type. Requests to bind sockets to
port numbers being used by other sockets return the error EADDRINUSE. If the local port
address is specified as 0, then the system picks a unique port address greater than
IPPORT_RESERVED. A unique local port address is also picked when a socket which is not
bound is used in a connect(3SOCKET) or sendto (see send(3SOCKET)) call. This allows
programs which do not care which local port number is used to set up TCP connections by
simply calling socket(3SOCKET) and then connect(3SOCKET), and to send UDP
datagrams with a socket(3SOCKET) call followed by a sendto() call.

Although this implementation restricts sockets to unique local port numbers, TCP allows
multiple simultaneous connections involving the same local port number so long as the
remote IP addresses or port numbers are different for each connection. Programs may
explicitly override the socket restriction by setting the SO_REUSEADDR socket option with
setsockopt (see getsockopt(3SOCKET)).

TLI applies somewhat different semantics to the binding of local port numbers. These
semantics apply when Internet family protocols are used using the TLI.

400 man pages section 7: Device and Network Interfaces • Last Revised 3 Aug 2000
 inet(7P)

See Also ioctl(2), bind(3SOCKET), byteorder(3SOCKET), connect(3SOCKET),

gethostbyname(3NSL), getnetbyname(3SOCKET), getprotobyname(3SOCKET),
getservbyname(3SOCKET), getsockopt(3SOCKET), send(3SOCKET), socket(3SOCKET),
arp(7P), icmp(7P), ip(7P), tcp(7P), udp(7P)

Network Information Center, DDN Protocol Handbook (3 vols.), Network Information

Center, SRI International, Menlo Park, Calif., 1985.

Notes The Internet protocol support is subject to change as the Internet protocols develop. Users
should not depend on details of the current implementation, but rather the services exported.

Device and Network Interfaces 401

ip6(7P)

Name ip6 – Internet Protocol Version 6

Synopsis #include <sys/socket.h>
#include <netinet/in.h>
#include <netinet/ip6.h>
s = socket(AF_INET6, SOCK_RAW, proto);
t = t_open ("/dev/rawip6", O_RDWR);

Description The IPv6 protocol is the next generation of the internetwork datagram delivery protocol of the
Internet protocol family. Programs may use IPv6 through higher-level protocols such as the
Transmission Control Protocol (TCP) or the User Datagram Protocol (UDP), or may
interface directly to IPv6. See tcp(7P) and udp(7P). Direct access may be by means of the
socket interface, using a “raw socket,” or by means of the Transport Level Interface (TLI). The
protocol options and IPv6 extension headers defined in the IPv6 specification may be set in
outgoing datagrams.

Application The STREAMS driver /dev/rawip6 is the TLI transport provider that provides raw access to
Programming IPv6.
Interface
Raw IPv6 sockets are connectionless and are normally used with the sendto() and
recvfrom() calls (see send(3SOCKET) and recv(3SOCKET)), although the
connect(3SOCKET) call may also be used to fix the destination for future datagrams. In this
case, the read(2) or recv(3SOCKET) and write(2) or send(3SOCKET) calls may be used.
Ancillary data may also be sent or received over raw IPv6 sockets using the
sendmsg(3SOCKET) and recvmsg(3SOCKET) system calls.

Unlike raw IP, IPv6 applications do not include a complete IPv6 header when sending; there is
no IPv6 analog to the IP IP_HDRINCL socket option. IPv6 header values may be specified or
received as ancillary data to a sendmsg(3SOCKET) or recvmsg(3SOCKET) system call, or may
be specified as "sticky" options on a per-socket basis by using the setsockopt(3SOCKET)
system call. Such sticky options are applied to all outbound packets unless overridden by
ancillary data. If any ancillary data is specified in a sendmsg(3SOCKET) call, all sticky options
not explicitly overridden revert to default values for that datagram only; the sticky options
persist as set for subsequent datagrams.

Since sendmsg(3SOCKET) is not supported for SOCK_STREAM upper level protocols such as
TCP, ancillary data is unsupported for TCP. Sticky options, however, are supported.

Since sendmsg(3SOCKET) is supported for SOCK_DGRAM upper level protocols, both ancillary
data and sticky options are supported for UDP, ICMP6, and raw IPv6 sockets.

The socket options supported at the IPv6 level are:

IPV6_BOUND_IF Limit reception and transmission of packets to this interface.
Takes an integer as an argument; the integer is the selected
interace index.

402 man pages section 7: Device and Network Interfaces • Last Revised 2 Dec 2008
 ip6(7P)

IPV6_UNSPEC_SRC Boolean. Allow/disallow sending with a zero source address.

IPV6_UNICAST_HOPS Default hop limit for unicast datagrams. This option takes an
integer as an argument. Its value becomes the new default value
for ip6_hops that IPv6 will use on outgoing unicast datagrams
sent from that socket. The initial default is 60.
IPV6_CHECKSUM Specify the integer offset in bytes into the user data of the
checksum location. Does not apply to the ICMP6 protocol. Note:
checksums are required for all IPv6 datagrams; this is different
from IP, in which datagram checksums were optional. IPv6 will
compute the ULP checksum if the value in the checksum field is
zero.
IPV6_SEC_OPT Enable or obtain IPsec security settings for this socket. For more
details on the protection services of IPsec, see ipsec(7P).
IPV6_DONTFRAG Boolean. Control fragmentation.
IPV6_USE_MIN_MTU Controls whether path MTU discovery is used. If set to 1, path
MTU discovery is never used and IPv6 packets are sent with the
IPv6 minimum MTU. If set to -1, path MTU discovery is not
used for multicast and multicast packets are sent with the IPv6
minimum MTU. If set to 0, path MTU is always performed.
IPV6_V6ONLY Boolean. If set, only V6 packets can be sent or received
IPV6_SRC_PREFERENCES Enable or obtain Source Address Selection rule settings for this
socket. For more details on the Source Address Selection rules,
see inet6(7P).

The following options are boolean switches controlling the reception of ancillary data:
IPV6_RECVPKTINFO Enable/disable receipt of the index of the interface the packet
arrived on, and of the inbound packet's destination address.
IPV6_RECVHOPLIMIT Enable/disable receipt of the inbound packet's current hoplimit.
IPV6_RECVHOPOPTS Enable/disable receipt of the inbound packet's IPv6 hop-by-hop
extension header.
IPV6_RECVDSTOPTS Enable/disable receipt of the inbound packet's IPv6 destination
options extension header.
IPV6_RECVRTHDR Enable/disable receipt of the inbound packet's IPv6 routing
header.

Device and Network Interfaces 403

ip6(7P)

IPV6_RECVRTHDRDSTOPTS Enable/disable receipt of the inbound packet's

intermediate-hops options extension header. This option is
obsolete. IPV6_RECVDSTOPTS turns on receipt of both
destination option headers.
IPV6_RECVTCLASS Enable/disable receipt of the traffic class of the inbound packet.
IPV6_RECVPATHMTU Enable/disable receipt of the path mtu of the inbound packet.

The following options may be set as sticky options with setsockopt(3SOCKET) or as

ancillary data to a sendmsg(3SOCKET) system call:
IPV6_PKTINFO Set the source address and/or interface out which the packet(s) will
be sent. Takes a struct in6_pktinfo as the parameter.
IPV6_HOPLIMIT Set the initial hoplimit for outbound datagrams. Takes an integer as
the parameter. Note: This option sets the hoplimit only for ancillary
data or sticky options and does not change the default hoplimit for
the socket; see IPV6_UNICAST_HOPS and IPV6_MULTICAST_HOPS to
change the socket's default hoplimit.
IPV6_NEXTHOP Specify the IPv6 address of the first hop, which must be a neighbor of
the sending host. Takes a struct sockaddr_in6 as the parameter.
When this option specifies the same address as the destination IPv6
address of the datagram, this is equivalent to the existing
SO_DONTROUTE option.
IPV6_HOPOPTS Specify one or more hop-by-hop options. Variable length. Takes a
complete IPv6 hop-by-hop options extension header as the
parameter.
IPV6_DSTOPTS Specify one or more destination options. Variable length. Takes a
complete IPv6 destination options extension header as the
parameter.
IPV6_RTHDR Specify the IPv6 routing header. Variable length. Takes a complete
IPv6 routing header as the parameter. Currently, only type 0 routing
headers are supported.
IPV6_RTHDRDSTOPTS Specify one or more destination options for all intermediate hops.
May be configured, but will not be applied unless an IPv6 routing
header is also configured. Variable length. Takes a complete IPv6
destination options extension header as the parameter.
IPV6_PATHMTU Get the path MTU associated with a connected socket. Takes a
ip6_mtuinfo as the parameter.
IPV6_TCLASS Set the traffic class associated with outgoing packets. The parameter
is an integer. If the parameter is less then -1 or greater then 256,

404 man pages section 7: Device and Network Interfaces • Last Revised 2 Dec 2008
 ip6(7P)

EINVAL is returned. If the parameter is equal to -1, use the default. If

the parameter is between 0 and 255 inclusive, use that value.

The following options affect the socket's multicast behavior:

IPV6_JOIN_GROUP Join a multicast group. Takes a struct ipv6_mreq as the
parameter; the structure contains a multicast address and an
interface index.
IPV6_LEAVE_GROUP Leave a multicast group. Takes a struct ipv6_mreq as the
parameter; the structure contains a multicast address and an
interface index.
MCAST_JOIN_GROUP Functionally equivalent to IPV6_JOIN_GROUP. Takes a
struct group_req as the parameter. The structure contains
a multicast address and an interface index.
MCAST_BLOCK_SOURCE Block multicast packets on a particular multicast group
whose source address matches the given source address. The
specified group must be joined previously using
IPV6_JOIN_GROUP or MCAST_JOIN_GROUP. Takes a
struct group_source_req as the parameter. The structure
contains an interface index, a multicast address, and a
source address.
MCAST_UNBLOCK_SOURCE Unblock multicast packets which were previously blocked
using MCAST_BLOCK_SOURCE. Takes a struct
group_source_req as the parameter. The structure contains
an interface index, a multicast address, and a source address.
MCAST_LEAVE_GROUP Functionally equivalent to IPV6_LEAVE_GROUP. Takes a
struct group_req as the parameter. The structure contains
a multicast address and an interface index.
MCAST_JOIN_SOURCE_GROUP Begin receiving packets for the given multicast group whose
source address matches the specified address. Takes a
struct group_source_req as the parameter. The structure
contains an interface index, a multicast address, and a
source address.
MCAST_LEAVE_SOURCE_GROUP Stop receiving packets for the given multicast group whose
source address matches the specified address. Takes a
struct group_source_req as the parameter. The structure
contains an interface index, a multicast address, and a
source address.

Device and Network Interfaces 405

ip6(7P)

IPV6_MULTICAST_IF The outgoing interface for multicast packets. This option

takes an integer as an argument; the integer is the interface
index of the selected interface.
IPV6_MULTICAST_HOPS Default hop limit for multicast datagrams. This option takes
an integer as an argument. Its value becomes the new default
value for ip6_hops that IPv6 will use on outgoing multicast
datagrams sent from that socket. The initial default is 1.
IPV6_MULTICAST_LOOP Loopback for multicast datagrams. Normally multicast
datagrams are delivered to members on the sending host.
Setting the unsigned character argument to 0 will cause the
opposite behavior.

The multicast socket options can be used with any datagram socket type in the IPv6 family.

At the socket level, the socket option SO_DONTROUTE may be applied. This option forces
datagrams being sent to bypass routing and forwarding by forcing the IPv6 hoplimit field to 1,
meaning that the packet will not be forwarded by routers.

Raw IPv6 datagrams can also be sent and received using the TLI connectionless primitives.

Datagrams flow through the IPv6 layer in two directions: from the network up to user
processes and from user processes down to the network. Using this orientation, IPv6 is layered
above the network interface drivers and below the transport protocols such as UDP and TCP.
The Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6)
is logically a part of IPv6. See icmp6(7P).

Unlike IP, IPv6 provides no checksum of the IPv6 header. Also unlike IP, upper level protocol
checksums are required. IPv6 will compute the ULP/data portion checksum if the checksum
field contains a zero (see IPV6_CHECKSUM option above).

IPv6 extension headers in received datagrams are processed in the IPv6 layer according to the
protocol specification. Currently recognized IPv6 extension headers include hop-by-hop
options header, destination options header, routing header (currently, only type 0 routing
headers are supported), and fragment header.

By default, the IPv6 layer will not forward IPv6 packets that are not addressed to it. This
behavior can be overridden by using routeadm(1M) to enable the ipv6-forwarding option.
IPv6 forwarding is configured at boot time based on the setting of routeadm(1M)'s
ipv6-forwarding option.

For backwards compatibility, IPv6 forwarding can be enabled or disabled using ndd(1M)'s
ip_forwarding variable. It is set to 1 if IPv6 forwarding is enabled, or 0 if it is disabled.

Additionally, finer-grained forwarding can be configured in IPv6. Each interface can be

configured to forward IPv6 packets by setting the IFF_ROUTER interface flag. This flag can be

406 man pages section 7: Device and Network Interfaces • Last Revised 2 Dec 2008
 ip6(7P)

set and cleared using ifconfig(1M)'s router and –router options. If an interface's
IFF_ROUTER flag is set, packets can be forwarded to or from the interface. If it is clear,
packets will neither be forwarded from this interface to others, nor forwarded to this interface.
Setting the ip6_forwarding variable sets all of the IPv6 interfaces' IFF_ROUTER flags.

For backwards compatibility, each interface creates an <ifname>ip6_forwarding /dev/ip6

variable that can be modified using ndd(1M). An interface's :ip6_forwarding ndd variable is
a boolean variable that mirrors the status of its IFF_ROUTER interface flag. It is set to 1 if the
flag is set, or 0 if it is clear. This interface specific <ifname>:ip6_forwarding ndd variable is
obsolete and may be removed in a future release of Solaris. The ifconfig(1M) router and
–router interfaces are preferred.

The IPv6 layer will send an ICMP6 message back to the source host in many cases when it
receives a datagram that can not be handled. A "time exceeded" ICMP6 message will be sent if
the ip6_hops field in the IPv6 header drops to zero in the process of forwarding a datagram. A
"destination unreachable" message will be sent by a router or by the originating host if a
datagram can not be sent on because there is no route to the final destination; it will be sent by
a router when it encounters a firewall prohibition; it will be sent by a destination node when
the transport protocol (that is, TCP) has no listener. A "packet too big" message will be sent
by a router if the packet is larger than the MTU of the outgoing link (this is used for Path MTU
Discovery). A "parameter problem" message will be sent if there is a problem with a field in
the IPv6 header or any of the IPv6 extension headers such that the packet cannot be fully
processed.

The IPv6 layer supports fragmentation and reassembly. Datagrams are fragmented on output
if the datagram is larger than the maximum transmission unit (MTU) of the network
interface. Fragments of received datagrams are dropped from the reassembly queues if the
complete datagram is not reconstructed within a short time period.

Errors in sending discovered at the network interface driver layer are passed by IPv6 back up
to the user process.

See Also svcs(1), ndd(1M), routeadm(1M), svcadm(1M), read(2), write(2), bind(3SOCKET),

connect(3SOCKET), getsockopt(3SOCKET), recv(3SOCKET), recvmsg(3SOCKET),
send(3SOCKET), sendmsg(3SOCKET), setsockopt(3SOCKET), defaultrouter(4), smf(5),
icmp6(7P), if_tcp(7P), ipsec(7P), inet6(7P), routing(7P), tcp(7P), udp(7P)

Deering, S. and Hinden, B. RFC 2460, Internet Protocol, Version 6 (IPv6) Specification. The
Internet Society. December, 1998.

Stevens, W., and Thomas, M. RFC 2292, Advanced Sockets API for IPv6. Network Working
Group. February 1998.

Diagnostics A socket operation may fail with one of the following errors returned:
EPROTONOSUPPORT Unsupported protocol (for example, IPPROTO_RAW.)

Device and Network Interfaces 407

ip6(7P)

EACCES A bind() operation was attempted with a “reserved” port number and
the effective user ID of the process was not the privileged user.
EADDRINUSE A bind() operation was attempted on a socket with a network
address/port pair that has already been bound to another socket.
EADDRNOTAVAIL A bind() operation was attempted for an address that is not configured
on this machine.
EINVAL A sendmsg() operation with a non-NULL msg_accrights was
attempted.
EINVAL A getsockopt() or setsockopt() operation with an unknown socket
option name was given.
EINVAL A getsockopt() or setsockopt() operation was attempted with the
IPv6 option field improperly formed; an option field was shorter than
the minimum value or longer than the option buffer provided; the
value in the option field was invalid.
EISCONN A connect() operation was attempted on a socket on which a
connect() operation had already been performed, and the socket
could not be successfully disconnected before making the new
connection.
EISCONN A sendto() or sendmsg() operation specifying an address to which the
message should be sent was attempted on a socket on which a
connect() operation had already been performed.
EMSGSIZE A send(), sendto(), or sendmsg() operation was attempted to send a
datagram that was too large for an interface, but was not allowed to be
fragmented (such as broadcasts).
ENETUNREACH An attempt was made to establish a connection via connect(), or to
send a datagram by means of sendto() or sendmsg(), where there was
no matching entry in the routing table; or if an ICMP “destination
unreachable” message was received.
ENOTCONN A send() or write() operation, or a sendto() or sendmsg() operation
not specifying an address to which the message should be sent, was
attempted on a socket on which a connect() operation had not already
been performed.
ENOBUFS The system ran out of memory for fragmentation buffers or other
internal data structures.
ENOMEM The system was unable to allocate memory for an IPv6 socket option or
other internal data structures.

408 man pages section 7: Device and Network Interfaces • Last Revised 2 Dec 2008
 ip6(7P)

ENOPROTOOPT An IP socket option was attempted on an IPv6 socket, or an IPv6 socket

option was attempted on an IP socket.
ENOPROTOOPT Invalid socket type for the option.

Notes Applications using the sockets API must use the Advanced Sockets API for IPv6 (RFC 2292) to
see elements of the inbound packet's IPv6 header or extension headers.

The ip6 service is managed by the service management facility, smf(5), under the service
identifier:
svc:/network/initial:default

Administrative actions on this service, such as enabling, disabling, or requesting restart, can
be performed using svcadm(1M). The service's status can be queried using the svcs(1)
command.

Device and Network Interfaces 409

ip(7P)

Name ip, IP – Internet Protocol

Synopsis #include <sys/socket.h>
#include <netinet/in.h>
s = socket(AF_INET, SOCK_RAW, proto);
t = t_open ("/dev/rawip", O_RDWR);

Description IP is the internetwork datagram delivery protocol that is central to the Internet protocol
family. Programs may use IP through higher-level protocols such as the Transmission Control
Protocol (TCP) or the User Datagram Protocol (UDP), or may interface directly to IP. See
tcp(7P) and udp(7P). Direct access may be by means of the socket interface, using a “raw
socket,” or by means of the Transport Level Interface (TLI). The protocol options defined in
the IP specification may be set in outgoing datagrams.

Packets sent to or from this system may be subject to IPsec policy. See ipsec(7P) for more
information.

Application The STREAMS driver /dev/rawip is the TLI transport provider that provides raw access to IP.
Programming
Interface Raw IP sockets are connectionless and are normally used with the sendto() and recvfrom()
calls (see send(3SOCKET) and recv(3SOCKET)), although the connect(3SOCKET) call may
also be used to fix the destination for future datagram. In this case, the read(2) or
recv(3SOCKET) and write(2) or send(3SOCKET) calls may be used. If proto is IPPROTO_RAW
or IPPROTO_IGMP, the application is expected to include a complete IP header when sending.
Otherwise, that protocol number will be set in outgoing datagrams and used to filter incoming
datagrams and an IP header will be generated and prepended to each outgoing datagram. In
either case, received datagrams are returned with the IP header and options intact.

If an application uses IP_HDRINCL and provides the IP header contents, the IP stack does not
modify the following supplied fields under any conditions: Type of Service, DF Flag, Protocol,
and Destination Address. The IP Options and IHL fields are set by use of IP_OPTIONS, and
Total Length is updated to include any options. Version is set to the default. Identification is
chosen by the normal IP ID selection logic. The source address is updated if none was
specified and the TTL is changed if the packet has a broadcast destination address. Since an
applicaton cannot send down fragments (as IP assigns the IP ID), Fragment Offset is always 0.
The IP Checksum field is computed by IP. None of the data beyond the IP header are changed,
including the application-provided transport header.

The socket options supported at the IP level are:

IP_OPTIONS IP options for outgoing datagrams. This socket option may be used
to set IP options to be included in each outgoing datagram. IP
options to be sent are set with setsockopt() (see
getsockopt(3SOCKET)). The getsockopt(3SOCKET) call returns
the IP options set in the last setsockopt() call. IP options on

410 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 ip(7P)

received datagrams are visible to user programs only using raw IP

sockets. The format of IP options given in setsockopt() matches
those defined in the IP specification with one exception: the list of
addresses for the source routing options must include the first-hop
gateway at the beginning of the list of gateways. The first-hop
gateway address will be extracted from the option list and the size
adjusted accordingly before use. IP options may be used with any
socket type in the Internet family.
IP_SEC_OPT Enable or obtain IPsec security settings for this socket. For more
details on the protection services of IPsec, see ipsec(7P).
IP_ADD_MEMBERSHIP Join a multicast group.
IP_DROP_MEMBERSHIP Leave a multicast group.
IP_BOUND_IF Limit reception and transmission of packets to this interface. Takes
an integer as an argument. The integer is the selected interface
index.

The following options take in_pktinfo_t as the parameter:

IP_PKTINFO
Set the source address and/or transmit interface of the packet(s). Note that the
IP_BOUND_IF socket option takes precedence over the interface index passed in
IP_PKTINFO.
struct in_pktinfo {
unsigned int ipi_ifindex;/* send/recv interface index */
struct in_addr ipi_spec_dst;/* matched source addr. */
struct in_addr ipi_addr;/* src/dst addr. in IP hdr */
} in_pktinfo_t;
When passed in (on transmit) via ancillary data with IP_PKTINFO, ipi_spec_dst is used as
the source address and ipi_ifindex is used as the interface index to send the packet out.
IP_RECVPKTINFO
Enable/disable receipt of the index of the interface the packet arrived on, the local address
that was matched for reception, and the inbound packet's actual destination address. Takes
boolean as the parameter. Returns struct in_pktinfo_t as ancillary data.
The following options take a struct ip_mreq as the parameter. The structure contains a
multicast address which must be set to the CLASS-D IP multicast address and an interface
address. Normally the interface address is set to INADDR_ANY which causes the kernel to choose
the interface on which to join.
IP_BLOCK_SOURCE Block multicast packets whose source address matches the
given source address. The specified group must be joined
previously using IP_ADD_MEMBERSHIP or
MCAST_JOIN_GROUP.

Device and Network Interfaces 411

ip(7P)

IP_UNBLOCK_SOURCE Unblock (begin receiving) multicast packets which were

previously blocked using IP_BLOCK_SOURCE.
IP_ADD_SOURCE_MEMBERSHIP Begin receiving packets for the given multicast group
whose source address matches the specified address.
IP_DROP_SOURCE_MEMBERSHIP Stop receiving packets for the given multicast group whose
source address matches the specified address.

The following options take a struct ip_mreq_source as the parameter. The structure
contains a multicast address (which must be set to the CLASS-D IP multicast address), an
interface address, and a source address.
MCAST_JOIN_GROUP Join a multicast group. Functionally equivalent to
IP_ADD_MEMBERSHIP.
MCAST_BLOCK_SOURCE Block multicast packets whose source address matches the
given source address. The specified group must be joined
previously using IP_ADD_MEMBERSHIP or
MCAST_JOIN_GROUP.
MCAST_UNBLOCK_SOURCE Unblock (begin receiving) multicast packets which were
previously blocked using MCAST_BLOCK_SOURCE.
MCAST_LEAVE_GROUP Leave a multicast group. Functionally equivalent to
IP_DROP_MEMBERSHIP.
MCAST_JOIN_SOURCE_GROUP Begin receiving packets for the given multicast group whose
source address matches the specified address.
MCAST_LEAVE_SOURCE_GROUP Stop receiving packets for the given multicast group whose
source address matches the specified address.

The following options take a struct group_req or struct group_source_req as the parameter.
The ‘group_req structure contains an interface index and a multicast address which must be
set to the CLASS-D multicast address. The group_source_req structure is used for those
options which include a source address. It contains an interface index, multicast address, and
source address.
IP_MULTICAST_IF The outgoing interface for multicast packets. This option takes a
struct in_addr as an argument, and it selects that interface for
outgoing IP multicast packets. If the address specified is INADDR_ANY,
it uses the unicast routing table to select the outgoing interface
(which is the default behavior).
IP_MULTICAST_TTL Time to live for multicast datagrams. This option takes an unsigned
character as an argument. Its value is the TTL that IP uses on
outgoing multicast datagrams. The default is 1.

412 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 ip(7P)

IP_MULTICAST_LOOP Loopback for multicast datagrams. Normally multicast datagrams

are delivered to members on the sending host (or sending zone).
Setting the unsigned character argument to 0 causes the opposite
behavior, meaning that when multiple zones are present, the
datagrams are delivered to all zones except the sending zone.
IP_RECVIF Receive the inbound interface index.
IP_TOS This option takes an integer argument as its input value. The least
significant 8 bits of the value are used to set the Type Of Service field
in the IP header of the outgoing packets.
IP_DONTFRAG This option controls whether IP will allow fragmentation both
locally (fragmenting the packets before sending them out on the
wire), and in the network (whether or not the Don’t Fragment flag
will be set in the IPv4 header). Setting the option to any non-zero
value disables fragmentation. Setting the option to zero enables
fragmentation. When fragmentation is disabled then IP will not
create any Path MTU state on behalf of this socket.
IP_NEXTHOP This option specifies the address of the onlink nexthop for traffic
originating from that socket. It causes the routing table to be
bypassed and outgoing traffic is sent directly to the specified
nexthop. This option takes an ipaddr_t argument representing the
IPv4 address of the nexthop as the input value. The IP_NEXTHOP
option takes precedence over IPOPT_LSRR. IP_BOUND_IF and
SO_DONTROUTE take precedence over IP_NEXTHOP. This
option has no meaning for broadcast and multicast packets. The
application must ensure that the specified nexthop is alive. An
application may want to specify the IP_NEXTHOP option on a TCP
listener socket only for incoming requests to a particular IP address.
In this case, it must avoid binding the socket to INADDR_ANY and
instead must bind the listener socket to the specific IP address. In
addition, typically the application may want the incoming and
outgoing interface to be the same. In this case, the application must
select a suitable nexthop that is onlink and reachable via the desired
interface and do a setsockopt (IP_NEXTHOP) on it. Then it must
bind to the IP address of the desired interface. Setting the
IP_NEXTHOP option requires the PRIV_SYS_NET_CONFIG
privilege.

The multicast socket options (IP_MULTICAST_IF, IP_MULTICAST_TTL,

IP_MULTICAST_LOOP and IP_RECVIF) can be used with any datagram socket type in the
Internet family.

Device and Network Interfaces 413

ip(7P)

At the socket level, the socket option SO_DONTROUTE may be applied. This option forces
datagrams being sent to bypass routing and forwarding by forcing the IP Time To Live field to
1, meaning that the packet will not be forwarded by routers.

Raw IP datagrams can also be sent and received using the TLI connectionless primitives.

Datagrams flow through the IP layer in two directions: from the network up to user processes
and from user processes down to the network. Using this orientation, IP is layered above the
network interface drivers and below the transport protocols such as UDP and TCP. The
Internet Control Message Protocol (ICMP) is logically a part of IP. See icmp(7P).

IP provides for a checksum of the header part, but not the data part, of the datagram. The
checksum value is computed and set in the process of sending datagrams and checked when
receiving datagrams.

IP options in received datagrams are processed in the IP layer according to the protocol
specification. Currently recognized IP options include: security, loose source and record route
(LSRR), strict source and record route (SSRR), record route, and internet timestamp.

By default, the IP layer will not forward IPv4 packets that are not addressed to it. This behavior
can be overridden by using routeadm(1M) to enable the ipv4-forwarding option. IPv4
forwarding is configured at boot time based on the setting of routeadm(1M)'s ipv4-forwarding
option.

For backwards compatibility, IPv4 forwarding can be enabled or disabled using ndd(1M)'s
ip_forwarding variable. It is set to 1 if IPv4 forwarding is enabled, or 0 if it is disabled.

Additionally, finer-grained forwarding can be configured in IP. Each interface can be

configured to forward IP packets by setting the IFF_ROUTER interface flag. This flag can be
set and cleared using ifconfig(1M)'s router and router options. If an interface's
IFF_ROUTER flag is set, packets can be forwarded to or from the interface. If it is clear,
packets will neither be forwarded from this interface to others, nor forwarded to this interface.
Setting the ip_forwarding variable sets all of the IPv4 interfaces' IFF_ROUTER flags.

For backwards compatibility, each interface creates an <ifname>:ip_forwarding /dev/ip

variable that can be modified using ndd(1M). An interface's :ip_forwarding ndd variable is a
boolean variable that mirrors the status of its IFF_ROUTER interface flag. It is set to 1 if the
flag is set, or 0 if it is clear. This interface specific <ifname> :ip_forwarding ndd variable is
obsolete and may be removed in a future release of Solaris. The ifconfig(1M) router and
–router interfaces are preferred.

The IP layer sends an ICMP message back to the source host in many cases when it receives a
datagram that can not be handled. A “time exceeded” ICMP message is sent if the “time to
live” field in the IP header drops to zero in the process of forwarding a datagram. A
“destination unreachable” message is sent if a datagram can not be forwarded because there is
no route to the final destination, or if it can not be fragmented. If the datagram is addressed to

414 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 ip(7P)

the local host but is destined for a protocol that is not supported or a port that is not in use, a
destination unreachable message is also sent. The IP layer may send an ICMP “source quench”
message if it is receiving datagrams too quickly. ICMP messages are only sent for the first
fragment of a fragmented datagram and are never returned in response to errors in other
ICMP messages.

The IP layer supports fragmentation and reassembly. Datagrams are fragmented on output if
the datagram is larger than the maximum transmission unit (MTU) of the network interface.
Fragments of received datagrams are dropped from the reassembly queues if the complete
datagram is not reconstructed within a short time period.

Errors in sending discovered at the network interface driver layer are passed by IP back up to
the user process.

Multi-Data Transmit allows more than one packet to be sent from the IP module to another in
a given call, thereby reducing the per-packet processing costs. The behavior of Multi-Data
Transmit can be overrideen by using ndd(1M) to set the /dev/ip variable,
ip_multidata_outbound to 0. Note, the IP module will only initiate Multi-Data Transmit if the
network interface driver supports it.

Packet Events Through the netinfo framework, this driver provides the following packet events:
Physical in Packets received on a network interface from an external source.
Physical out Packets to be sent out a network interface.
Forwarding Packets being forwarded through this host to another network.
loopback in Packets that have been sent by a local application to another.
loopback out Packets about to be received by a local application from another.

Currently, only a single function may be registered for each event. As a result, if the slot for an
event is already occupied by someone else, a second attempt to register a callback fails.

To receive packet events in a kernel module, it is first necessary to obtain a handle for either
IPv4 or IPv6 traffic. This is achieved by passing NHF_INET or NHF_INET6 through to a
net_protocol_lookup() call. The value returned from this call must then be passed into a call
to net_register_hook(), along with a description of the hook to add. For a description of the
structure passed through to the callback, please see hook_pkt_event(9S). For IP packets, this
structure is filled out as follows:
hpe_ifp Identifier indicating the inbound interface for packets received with the physical
in event.
hpe_ofp Identifier indicating the outbound interface for packets received with the
physical out event.
hpe_hdr Pointer to the start of the IP header (not the ethernet header).

Device and Network Interfaces 415

ip(7P)

hpe_mp Pointer to the start of the mblk_t chain containing the IP packet.
hpe_mb Pointer to the mblk_t with the IP header in it.

Network Interface In addition to events describing packets as they move through the system, it is also possible to
Events receive notification of events relating to network interfaces. These events are all reported back
through the same callback. The list of events is as follows:
plumb A new network interface has been instantiated.
unplumb A network interface is no longer associated with this protocol.
up At least one logical interface is now ready to receive packets.
down There are no logical interfaces expecting to receive packets.
address change An address has changed on a logical interface.

See Also ifconfig(1M), routeadm(1M), ndd(1M), read(2), write(2), bind(3SOCKET),

connect(3SOCKET), getsockopt(3SOCKET), recv(3SOCKET), send(3SOCKET),
defaultrouter(4), icmp(7P), if_tcp(7P), inet(7P), ip6(7P), ipsec(7P), routing(7P),
tcp(7P), udp(7P), net_hook_register(9F), hook_pkt_event(9S)

Braden, R., RFC 1122, Requirements for Internet Hosts − Communication Layers, Information
Sciences Institute, University of Southern California, October 1989.

Postel, J., RFC 791, Internet Protocol − DARPA Internet Program Protocol Specification,
Information Sciences Institute, University of Southern California, September 1981.

Diagnostics A socket operation may fail with one of the following errors returned:
EACCES A bind() operation was attempted with a “reserved” port number and the
effective user ID of the process was not the privileged user.

Setting the IP_NEXTHOP was attempted by a process lacking the

PRIV_SYS_NET_CONFIG privilege.
EADDRINUSE A bind() operation was attempted on a socket with a network
address/port pair that has already been bound to another socket.
EADDRNOTAVAIL A bind() operation was attempted for an address that is not configured
on this machine.
EINVAL A sendmsg() operation with a non-NULL msg_accrights was
attempted.
EINVAL A getsockopt() or setsockopt() operation with an unknown socket
option name was given.

416 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 ip(7P)

EINVAL A getsockopt() or setsockopt() operation was attempted with the IP

option field improperly formed; an option field was shorter than the
minimum value or longer than the option buffer provided.
EISCONN A connect() operation was attempted on a socket on which a connect()
operation had already been performed, and the socket could not be
successfully disconnected before making the new connection.
EISCONN A sendto() or sendmsg() operation specifying an address to which the
message should be sent was attempted on a socket on which a connect()
operation had already been performed.
EMSGSIZE A send(), sendto(), or sendmsg() operation was attempted to send a
datagram that was too large for an interface, but was not allowed to be
fragmented (such as broadcasts).
ENETUNREACH An attempt was made to establish a connection by means of connect(),
or to send a datagram by means of sendto() or sendmsg(), where there
was no matching entry in the routing table; or if an ICMP “destination
unreachable” message was received.
ENOTCONN A send() or write() operation, or a sendto() or sendmsg() operation
not specifying an address to which the message should be sent, was
attempted on a socket on which a connect() operation had not already
been performed.
ENOBUFS The system ran out of memory for fragmentation buffers or other internal
data structures.
ENOBUFS SO_SNDBUF or SO_RCVBUF exceeds a system limit.
EINVAL Invalid length for IP_OPTIONS.
EHOSTUNREACH Invalid address for IP_MULTICAST_IF.

Invalid (offlink) nexthop address for IP_NEXTHOP.

EINVAL Not a multicast address for IP_ADD_MEMBERSHIP and
IP_DROP_MEMBERSHIP.
EADDRNOTAVAIL Bad interface address for IP_ADD_MEMBERSHIP and IP_DROP_MEMBERSHIP.
EADDRINUSE Address already joined for IP_ADD_MEMBERSHIP.
ENOENT Address not joined for IP_DROP_MEMBERSHIP.
ENOPROTOOPT Invalid socket type.
EPERM No permissions.

Device and Network Interfaces 417

ip(7P)

Notes Raw sockets should receive ICMP error packets relating to the protocol; currently such
packets are simply discarded.

Users of higher-level protocols such as TCP and UDP should be able to see received IP
options.

418 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 ipge(7D)

Name ipge – PCI-E Gigabit-Ethernet device driver for Intel 82571–based ethernet controller.
Synopsis /dev/ipge

Description The ipge Sun Gigabit-Ethernet driver is a multi-threaded, loadable, clonable, STREAMS
hardware driver supporting the connectionless Data Link Provider Interface, dlpi(7P).
Multiple PCI-E based adapters installed within the system are supported by the driver. The
ipge driver provides basic support for the PCI-E–based Ethernet hardware and is used to
handle pci8086,105e (PCI-E) devices. Functions include chip initialization, frame transmit
and receive, multicast and promiscuous support, and error recovery and reporting. The PCI-E
device provides 1000BASE-SX networking interfaces using PCI-E ASIC, external SERDES
and fiber optical transceiver, or 10/100/1000BASE-T using a PCI-E ASIC attached to a GMII
twisted pair copper transceiver, or 10/100BASE-T using a PCI-E ASIC attached to a MII
twisted pair copper transceiver.

The 1000Base-SX standard specifies an "auto-negotiation" protocol to automatically select the

mode of operation. In addition to duplex mode of operation, the MAC controller can
auto-negotiate for IEEE 802.3x frame based flow control capabilities. The PCI-E PCS is
capable of doing auto-negotiation with the remote-end of the link (link partner) and receives
the capabilities of the remote end. It selects the highest common denominator mode of
operation based on the priorities and also supports forced-mode of operation, in which the
driver selects the mode of operation.

Application The /dev/ipge cloning character-special device is used to access all ipge controllers installed
Programming within the system.
Interface
ipge and Dlpi The ipge driver is a "style 2" Data Link Service provider. All M_PROTO and M_PCPROTO
type messages are interpreted as DLPI primitives. Valid DLPI primitives are defined in
<sys/dlpi.h>. Refer to dlpi(7P) for more information. An explicit DL_ATTACH_REQ
message by a DLS user is required to associate an opened stream to a particular device (ppa).
The ppa ID is interpreted as an unsigned long data type and indicates the corresponding
device instance (unit) number. An error (DL_ERROR_ACK) is returned by the driver if the
ppa field value does not correspond to a valid device instance number in the system. The
device is initialized on first attach and un-initialized (stopped) during last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are as follows:
■ Maximum SDU is 1500. (ETHERMTU - defined in <sys/ethernet>.
■ Minimum SDU is 0.
■ DSLAP address length is 8 bytes.
■ MAC type is DL_ETHER.
■ SAP length value is -2 meaning the physical address component is followed immediately
by a 2-byte sap component within the DLSAP address.

Device and Network Interfaces 419

ipge(7D)

■ Service mode is DL_CLDLS.

■ No optional quality of service (QOS) support is currently included and the QOS fields are
0.
■ Provider style is DL_STYLE2.
■ Version is DL_VERSION_2.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
SAP (Service Access Point) with the stream. The ipge driver interprets the sap field within the
DL_BIND_REQ as an Ethernet "type," meaning valid values for the sap field are in the
[0-0xFFFF] range. Only one Ethernet type can be bound to the stream at any time.

When you select a sap with a value of 0, the receiver is in "802.3 mode." All frames received
from the media having a "type" field in the range [0-1500] are assumed to be 802.3 frames and
are routed up all open streams which are bound to sap value 0. If more than one stream is in
"802.3 mode" then the frame is duplicated and routed up multiple streams as
DL_UNITDATA_IND messages.

During transmission, the driver checks if either the sap value is 0 or destination type field is in
the range [0-1500]. If true, the driver sets MAC frame header length field with the length of
DL_UNITDATA_REQ message blocks, excluding initial M_PROTO message block, and
transmits as 802.3 frames.

The ipge driver DLSAP address format consists of the 6 byte physical (Ethernet) address
component followed by 2 byte sap (type) component producing an 8 byte DLSAP address.
Applications should not hard code to this particular implementation-specific DLSAP address
format but use information returned in the DL_INFO_ACK primitive to compose and
decompose DLSAP addresses. The sap length, full DLSAP length, and sap/physical ordering
are included within the DL_INFO_ACK. The physical address length can be computed by
subtracting the sap length from the full DLSAP address length or by issuing the
DL_PHYS_ADDR_REQ to obtain the current physical address associated with the stream.

Once the stream is in the DL_BOUND state, you may begin transmitting by sending
DL_UNITDATA_REQ messages to the driver.

During receive, the driver routes all received Ethernet frames as DL_UNITDATA_IND
messages to all open and bound streams whose sap matches the Ethernet type of the received
frame. Received Ethernet frames are duplicated and routed up multiple open streams if
necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.

420 man pages section 7: Device and Network Interfaces • Last Revised 23 June 2005
 ipge(7D)

dlpi Primitives In addition to the mandatory connectionless DLPI messages, the driver supports the
primitives described below.

The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable/disable

reception of individual multicast group addresses. A set of multicast addresses may be
iteratively created and modified on a per-stream basis using DL_ENABMULTI_REQ and
DL_DISABMULTI_REQ. DL_ENABMULTI_REQ and DL_DISABMULTI_REQ are
accepted by the driver in any state following DL_ATTACHED state.

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives with the

DL_PROMISC_PHYS flag set in dl_level field enables/disables reception of all promiscuous
mode frames on the media including frames generated by the local host. When used with the
DL_PROMISC_SAP flag set, this enables/disables reception of all sap (Ethernet type) values.
When used with the DL_PROMISC_MULTI flag set, this enables/disables reception of all
multicast group addresses. The effect of each is always on a per-stream basis and independent
of the other sap and physical level configurations on this stream or other streams.

The DL_PHYS_ADDR_REQ primitive returns the 6 octet Ethernet address currently

associated (attached) to the stream in the DL_PHYS_ADDR_ACK primitive.
DL_PHYS_ADDR_REQ is valid only in states following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the 6 octet Ethernet address currently

associated (attached) to the stream. The credentials of the process which originally opened the
stream must be superuser or EPERM is returned in the DL_ERROR_ACK. Because it affects
all current and future streams attached to the device, the DL_SET_PHYS_ADDR_REQ is
destructive. An M_ERROR is sent up all other streams attached to the device when
DL_SET_PHYS_ADDR_REQ is successful on the stream. Once changed, all streams
subsequently opened and attached to the device obtain the new physical address. Once
changed, the physical address remains until DL_SET_PHYS_ADDR_REQ is used to change
the physical address again or the system is rebooted, whichever occurs first.

Configuration By default, the ipge driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can be any one of the following, (as described in the IEEE803.2 standard):
■ 1000 Mbps, full-duplex
■ 1000 Mbps, half-duplex
■ 100 Mbps, full-duplex
■ 100 Mbps, half-duplex
■ 10 Mbps, full-duplex
■ 10 Mbps, half-duplex

The auto-negotiation protocol automatically selects:

■ Speed (1000 Mbps, 100 Mbps, or 10 Mbps)
■ Operation mode (full-duplex or half-duplex)

Device and Network Interfaces 421

ipge(7D)

The auto-negotiation protocol:

Gets all the modes of operation supported by the link partner.

Advertises its capabilities to the link partner.
Selects the highest common denominator mode of operation based on the priorities

The PCI-E hardware is capable of all of the operating modes listed above, when by default,
auto-negotiation is used to bring up the link and select the common mode of operation with
the link partner. Forced-mode of operation is supported (in which the driver selects the mode
of operation and the flow control capabilities) using the ndd(1M) utility.

Parameters The ipge driver enables setting/getting of various parameters for the PCI-E device. The
parameter list includes current transceiver status, current link status, interpacket gap, PCS
capabilities and link partner capabilities. PCS capabilities consist of two sets: one reflects the
capabilities of the hardware (which are read-only (RO) parameters), while the second reflects
the values you choose and is used in speed selection. At boot time, thse two sets of capabilities
are identical. By default, the link partner capabilities are read only and cannot be modified.
Files /dev/ipge Character special device.
/kernel/drv/ipge.conf System wide default device driver properties.

See Also ndd(1M), netstat(1M), driver.conf(4), dlpi(7P)

422 man pages section 7: Device and Network Interfaces • Last Revised 23 June 2005
 ipgpc(7ipp)

Name ipgpc – IP Generic Packet Classifier

Description The IP Generic Packet Classifier (ipgpc) module provides packet classification at the Solaris
IP layer. ipgpc is an implementation of the Multi-Field (MF) classifier as described in
RFC2475: An Architecture for Differentiated Services.

The classifier is configured, at startup or dynamically, using a set of “filters.” Filters describe
selectors that are matched against input packets that are processed by the classifier. Some
selectors feature exact matching data points, while others utilize non-exact or wildcard data
points.

Each filter is associated with a class describing the next actions to process a packet. There is a
many-to-one (M-to-1) mapping relationship between filters and a class. Additionally, each
class is aware of which filters are associated with it. A class is configured with a class name and
a next action.

Unlike traditional classifiers used in edge routers, ipgpc is designed for a host or server device.
A host-based classifier provides access to more resources and data than edge routers. User,
project, and interface information are available at the host.

Statistics The ipgpc module exports global and per-class statistics (available through kstat:)

Global statistics:
module: ipgpc instance:<action id>
name: ipgpc global stats class: <action name>
crtime
snaptime
nbytes <number of classified bytes>
nclasses <number of classes>
nfilters <number of filters>
npackets <number of classified packets>
epackets <number of packets in error>

Per-class statistics:
module: ipgpc instance:<action id>
name: <class name> class: <action name>
crtime
snaptime
last match <time of last match>
nbytes <number of classified bytes>
npackets <number of classified packets>
Files /kernel/ipp/sparcv9/ipgpc 64-bit module (SPARC only.)

Device and Network Interfaces 423

ipgpc(7ipp)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos

See Also ipqosconf(1M), dlcosmk(7ipp), dscpmk(7ipp), flowacct(7ipp), ipqos(7ipp), tokenmt(7ipp),

tswtclmt(7ipp)

RFC 2475, An Architecture for Differentiated Services S. Blake, D. Black, M. Carlson, E. Davies,
Z. Wang, W. Weiss — The Internet Society, 1998

424 man pages section 7: Device and Network Interfaces • Last Revised 29 Sep 2004
 ipnat(7I)

Name ipnat – IP Filter/NAT module interface

Description The ipnat device provides interfaction with the NAT features of the Solaris IPFilter.

Application The NAT features programming model is a component of the Solaris IP Filter and is accessed
Programming via the NAT device file /dev/ipnat. Opening the device for reading or writing determines
Interface which ioctl calls can be successfully made.

ioctls The caller must construct a ipfobj structure when issuing a SIOCGNATL or SIOCSTPUT. The
ipfobj structure is then passed to the ioctl call and is filled out with ipfo_type set to
IPFOBJ_value. IPFOBJ_ value provides a matching name for the structure, while ipfo_size is
set to the total size of the structure being passed and ipfo_ptr is set to the structure address.
The ipfo_rev structure should be set to the current value of IPFILTER_VERSION, while
ipfo_offset and ipfo_xxxpad should be set to 0.
/*
* Structure used with SIOCGNATL/SIOCSTPUT.
*/
/*
* Object structure description. For passing through in ioctls.
*/
typedef struct ipfobj {
u_32_t ipfo_rev; /* IPFilter version (IPFILTER_VERSION) */
u_32_t ipfo_size; /* size of object at ipfo_ptr */
void *ipfo_ptr; /* pointer to object */
int ipfo_type; /* type of object being pointed to */
int ipfo_offset; /* bytes from ipfo_ptr where to start */
u_char ipfo_xxxpad[32]; /* reserved for future use */
} ipfobj_t;

#define IPFILTER_VERSION 4010901 /* IPFilter version */

#define IPFOBJ_NATSAVE 8 /* struct nat_save */
#define IPFOBJ_NATLOOKUP 9 /* struct natlookup */

The following ioctl() calls may be used to manipulate the ipnat sub-system inside of ipf. Note
that the ipnat driver only accept calls from applications using the same data model as the
kernel. In other words, 64–bit kernels can only accept calls from 64–bit applications. Calls
from 32–bit applications fail with EINVAL.
SIOCSTLCK Set or clear the NAT lock to prevent table updates attributable to packet
flow-through.
SIOCGNATL Search the NAT table for the rdr entry that matches the fields in the natlookup
structure. The caller must populate the structure with the address/port
information of the accepted TCP connection (nl_inip, nl_inport) and the
address/port information of the peer (nl_outip, nl_outport). The nl_flags field
must have the IPN_TCP option set. All other fields must be set to 0. If the call

Device and Network Interfaces 425

ipnat(7I)

succeeds, nl_realip and nl_realport are set to the real destination address and
port, respectively. The nl_inport and nl_outport fields must be in host byte
order.

If IPN_FINDFORWARD is set in nl_flags, a check is made to see if it is possible to

create an outgoing NAT session by checking if a packet coming from
(nl_realip,nl_realport) and destined for (nl_outip,nl_outport) can be
translated. If translation is possible, the flag remains set, otherwise it is cleared
in the structure returned to the caller.
/*
* Structure used with SIOCGNATL.
*/
typedef struct natlookup {
i6addr_t nl_inipaddr;
i6addr_t nl_outipaddr;
i6addr_t nl_realipaddr;
int nl_v;
int nl_flags;
u_short nl_inport;
u_short nl_outport;
u_short nl_realport;
} natlookup_t

#define nl_inip nl_inipaddr.in4

#define nl_outip nl_outipaddr.in4
#define nl_realip nl_realipaddr.in4
#define nl_inip6 nl_inipaddr.in6
#define nl_outip6 nl_outipaddr.in6
#define nl_realip6 nl_realipaddr.in6

/*
* Accepted values for nl_flags
*/
#define IPN_TCP 0x00001
#define IPN_FINDFORWARD 0x400000
SIOCSTPUT Move a NAT mapping structure from user space into the kernel. This ioctl is
used by ipfs(1M) to restore NAT sessions saved in /var/db/ipf/ipnat.ipf.
The nat_save structure must have its ipn_nat and ipn_ipnat structures filled
out correctly. Fields not assigned a value must be initialised to 0. All pointer
fields are adjusted, as appropriate, once the structure is passed into the kernel
and none are preserved.

To create a translation, the following fields must be set:

426 man pages section 7: Device and Network Interfaces • Last Revised 22 May 2008
 ipnat(7I)

Interface name - The interface name on which the host is to be exited must be
set in nat_ifnames[0].

Local IP address and port number - The connection's local IP address and port
number are stored in network byte order using nat_inip/nat_inport.

Destination address/port - The destination address/port are stored in

nat_oip/nat_oport.

Target address/port - The translation's target address/port is stored in

nat_outip/nat_outport.

The caller must also precalculate the checksum adjustments necessary to

complete the translation and store those values in nat_sumd (delta required
for TCP header) and nat_ipsumd (delta required for IP header).
/*
* Structures used with SIOCSTPUT.
*/
typedef struct nat_save {
void *ipn_next;
struct nat ipn_nat;
struct ipnat ipn_ipnat;
struct frentry ipn_fr;
int ipn_dsize;
char ipn_data[4];
} nat_save_t;

typedef struct nat {

ipfmutex_t nat_lock;
struct nat *nat_next;
struct nat **nat_pnext;
struct nat *nat_hnext[2];
struct nat **nat_phnext[2];
struct hostmap *nat_hm;
void *nat_data;
struct nat **nat_me;
struct ipstate *nat_state;
struct ap_session *nat_aps;
frentry_t *nat_fr;
struct ipnat *nat_ptr;
void *nat_ifps[2];
void *nat_sync;
ipftqent_t nat_tqe;
u_32_t nat_flags;

Device and Network Interfaces 427

ipnat(7I)

u_32_t nat_sumd[2];
u_32_t nat_ipsumd;
u_32_t nat_mssclamp;
i6addr_t nat_inip6;
i6addr_t nat_outip6;
i6addr_t nat_oip6;
U_QUAD_T nat_pkts[2];
U_QUAD_T nat_bytes[2];
union {
udpinfo_t nat_unu;
tcpinfo_t nat_unt;
icmpinfo_t nat_uni;
greinfo_t nat_ugre;
} nat_un;
u_short nat_oport;
u_short nat_use;
u_char nat_p;
int nat_dir;
int nat_ref;
int nat_hv[2];
char nat_ifnames[2][LIFNAMSIZ];
int nat_rev;
int nat_v;
} nat_t;

#define nat_inip nat_inip6.in4

#define nat_outip nat_outip6.in4
#define nat_oip nat_oip6.in4
#define nat_inport nat_un.nat_unt.ts_sport
#define nat_outport nat_un.nat_unt.ts_dport
/*
* Values for nat_dir
*/
#define NAT_INBOUND 0
#define NAT_OUTBOUND 1
/*
* Definitions for nat_flags
*/
#define NAT_TCP 0x0001 /* IPN_TCP */

Examples The following example shows how to prepare and use SIOCSTPUT to insert a NAT session
directly into the table. Note that the usual TCP/IP code is omitted is this example.

In the code segment below, incoming_fd is the TCP connection file descriptor that is accepted
as part of the redirect process, while remote_fd is the outgoing TCP connection to the remote
server being translated back to the original IP address/port pair.

428 man pages section 7: Device and Network Interfaces • Last Revised 22 May 2008
 ipnat(7I)

Note – The following ipnat headers must be included before you can use the code shown in this
example:
#include <netinet/in.h>
#include <arpa/inet.h>
#include <net/if.h>
#include <netinet/ipl.h>
#include <netinet/ip_compat.h>
#include <netinet/ip_fil.h>
#include <netinet/ip_nat.h>
#include <string.h>
#include <fcntl.h>

Note – In the example below, various code fragments have been excluded to enhance clarity.
int
translate_connection(int incoming_fd)
{
struct sockaddr_in usin;
struct natlookup nlp;
struct nat_save ns;
struct ipfobj obj;
struct nat *nat;
int remote_fd;
int nat_fd;
int onoff;

memset(&ns, 0, sizeof(ns));
nat = &ns.ipn_nat

namelen = sizeof(usin);
getsockname(remote_fd, (struct sockaddr *)&usin, &namelen);

namelen = sizeof(sin);
getpeername(incoming_fd, (struct sockaddr *) &sin, &namelen);

namelen = sizeof(sloc);
getsockname(incoming_fd, (struct sockaddr *) &sloc, &namelen);

bzero((char *) &obi, sizeof(obj));

obj.ipfo_rev = IPFILTER_VERSION;
obj.ipfo_size = sizeof(nlp);
obj.ipfo_ptr = &nip;
obj.ipfo_type = IPFOBJ_NATLOOKUP;

/*
* Build up the NAT natlookup structure.
*/

Device and Network Interfaces 429

ipnat(7I)

bzero((char *) &nlp, sizeof(nlp));

nlp.nl_outip = sin.sin_addr;
nlp.nl_inip = sloc.sin_addr;
nlp.nl_flags = IPN_TCP;
nlp.nl_outport = ntohs(sin.sin_port);
nlp.nl_inport = ntohs(sloc.sin_port);

/*
* Open the NAT device and lookup the mapping pair.
*/
nat_fd = open(IPNAT_NAME, O_RDWR);
if (ioctl(nat_fd, SIOCGNATL, &obj) != 0)
return -1;

nat->nat_inip = usin.sin_addr;
nat->nat_outip = nlp.nl_outip;
nat->nat_oip = nlp.nl_realip;

sum1 = LONG_SUM(ntohl(usin.sin_addr.s_addr)) +
ntohs(usin.sin_port);
sum2 = LONG_SUM(ntohl(nat->nat_outip.s_addr)) +
ntohs(nlp.nl_outport);
CALC_SUMD(sum1, sum2, sumd);
nat->nat_sumd[0] = (sumd & 0xffff) + (sumd >> 16);
nat->nat_sumd[1] = nat->nat_sumd[0];

sum1 = LONG_SUM(ntohl(usin.sin_addr.s_addr));
sum2 = LONG_SUM(ntohl(nat->nat_outip.s_addr));
CALC_SUMD(sum1, sum2, sumd);
nat->nat_ipsumd = (sumd & 0xffff) + (sumd >> 16);

nat->nat_inport = usin.sin_port;
nat->nat_outport = nlp.nl_outport;
nat->nat_oport = nlp.nl_realport;

nat->nat_flags = IPN_TCPUDP;

/*
* Prepare the ipfobj structure, accordingly.
*/
bzero((char *)&obi, sizeof(obj));
obj.ipfo_rev = IPFILTER_VERSION;
obj.ipfo_size = sizeof(*nsp);
obj.ipfo_ptr = nsp;
obj.ipfo_type = IPFOBJ_NATSAVE;

onoff = 1;

430 man pages section 7: Device and Network Interfaces • Last Revised 22 May 2008
 ipnat(7I)

if (ioctl(nat_fd, SIOCSTPUT, &obj) != 0)

fprintf(stderr, "Error occurred\n");

return connect(rem_fd, (struct sockaddr ) &usin, sizeof(usin));

}
Errors EPERM The device has been opened for reading only. To succeed, the ioctl call must be
opened for both reading and writing. The call may be returned if it is
privileged and the calling process did not assert {PRIV_SYS_NET_CONFIG} in
the effective set.
ENOMEM More memory was allocated than the kernel can provide. The call may also be
returned if the application inserts a NAT entry that exceeds the hash bucket
chain's maximum length.
EFAULT The calling process specified an invalid pointer in the ipfobj structure.
EINVAL The calling process detected a parameter or field set to an unacceptable value.
EEXIST The calling process, via SIOCSTPUT, attempted to add a NAT entry that already
exists in the NAT table.
ESRCH The calling process called SIOCSTPUT before setting the SI_NEWFR flag and
providing a pointer in the nat_fr field that cannot be found in the current rule
set.
EACESS The calling process issued a SIOCSTPUT before issuing a SIOCSTLCK.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Committed

See Also ipfs(1M), ipnat(1M), ioctl(2), attributes(5)

Device and Network Interfaces 431

ipnet(7D)

Name ipnet, lo0 – ipnet device driver

Synopsis /dev/ipnet/*, /dev/lo0

Description The ipnet device driver creates, removes and manages nodes in the /dev/ipnet/ namespace.

A node is created in /dev/ipnet for every IP interface on the system, including interfaces that
exist only in software and for which there is no hardware device. The ipnet device also
provides access to all IP traffic to and from the system. To provide access to packets that are
internally looped-back in IP, the ipnet driver creates a /dev/lo0 DLPI device.

Application
Programming
Interface
ipnet and DLPI Device nodes created in /dev/ipnet are DLPI style-1 devices. All M_PROTO and
M_PCPROTO–type messages are interpreted as DLPI primitives. Because the device is read-only
and packets can only be observed by opening them, the following subset of DLPI primitives is
supported:
DL_INFO_REQ
DL_BIND_REQ
DL_UNBIND_REQ
DL_PROMISCON_REQ
DL_PROMISCOFF_REQ
DLIOCRAW

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are:
■ Maximum SDU is INT_MAX
■ Minimum SDU is 0.
■ DLSAP address length is 2.
■ MAC type is DL_IPNET.
■ SAP length value is 2.
■ Service mode is DL_CLDLS.
■ No optional quality of service (QOS) support is provided. Accordingly, the QOS fields are
0.
■ Provider style is DL_STYLE1.
■ Version is DL_VERSION_2.

The /dev/ipnet/* and /dev/lo0 devices only accept DL_BIND_REQ requests for SAPs 4 (IPv4
packets), 6 (IPv6 packets), or 0 (all IP packets). DL_BIND_REQ requests for other SAP values
result in a DL_ERROR_ACK of DL_BADSAP.

432 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 ipnet(7D)

ipnet primitives For /dev/ipnet/* devices, the DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives with
the DL_PROMISC_PHYS flag set in the dl_level field enables/disables the reception of all
packets. When disabled, only packets with addresses matching any of the configured
addresses on the IP interface are received. When used with the DL_PROMISC_MULTI flag set,
reception of all multicast group addresses can be enabled/disabled. DL_PROMISC_PHYS and
DL_PROMISC_MULTI have no effect for /dev/lo0. When the DL_PROMISC_SAP flag is set,
reception of all IPv4/IPv6 can be enabled/disabled.

The DLIOCRAW ioctl is supported but has no effect on the data returned from the device.

The DL_IOC_IPNET_INFO ioctl enables/disables the inclusion of a dl_ipnetinfo_t structure

that is prepended to the IP header when receiving packet data. When enabled, a non-zero
integer is returned reflecting the current DL_IOC_IPNET_INFO version. The dl_ipnetinfo_t
data structure is defined in <sys/dlpi.h> and includes the following fields:
uint8_t dli_version; /* DL_IPNETINFO_* version */
uint8_t dli_ipver; /* packet IP header version */
uint16_t dli_len; /* length of dl_ipnetinfo_t */
uint64_t dli_srczone; /* packet source zone ID (if any) */
uint64_t dli_dstzone; /* packet dest zone ID (if any) */

The current dli_version is 1. To robustly support future dl_ipnetinfo_t versions,

consumers should check that dli_version is a value they recognize, and must use the
dli_len field to advance past the dl_ipnetinfo_t header.
Files /dev/ipnet/*, /dev/lo0 Special character devices.
/kernel/drv/ipnet.conf Configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWckr

Architecture SPARC, x86

Interface Stability Committed

See Also attributes(5), dlpi(7P)

Device and Network Interfaces 433

ipqos(7ipp)

Name ipqos – IP Quality of Service

Description ipqos is an implementation of the Differentiated Services model defined in RFC2475: An

Architecture for Differentiated Services, which defines the following entities: multi-field
classifier, meter, marker, and dropper. The Solaris implementation of ipqos adds a flow
accounting entity.

These entities can be combined into processing paths that constitute a series of actions that are
performed on groups of flows. The classifier groups together flows and directs them in a given
processing path. Classifier configuration and path construction are achieved using the
ipqosconf(1M) command.

A summary of the ipqos entities follows. For more information, refer to the corresponding
man page for each entity.
ipgpc An implementation of the classifier defined in the model. ipgpc has
been extended and is able to select traffic based on IP header
parameters, user id, project id, interface name, interface group and
direction.
tokenmt, tswtclmt These modules implement different metering algorithms. tokenmt
implements both RFC2697: A Single Rate Three Color Marker and RFC
2698: A Two Rate Three Color Marker. tswtclmt implements RFC2859:
A Time Sliding Window Three Color Marker. These modules only
implement the metering functions defined in the RFCs.
dlcosmk A marker entity that allows the setting of the user priority field of
Ethernet frames as defined in the IEEE 802.1D specification. dlcosmk is
only available with VLAN capable network interfaces.
dscpmk A marker entity that enables the setting of the Differentiated Services
Code Point Value in the IP header as defined in RFC 2474: Definition of
the Differentiated Services Field (DS Field) in the IPv4 and IPv6
headers.
flowacct An accounting module that utilizes the Solaris extended accounting
facility. flowacct logs all flows with parameters used to build a charge
back mechanism.

Statistics ipqos modules export statistics through the kstat facility. Exported statistics contain the
following common parameters:
module module name
instance dynamic parameter identifying a specific instance
name a string for global statistics (for example, ipgpc global stat) or a class name for
per-class statistics for a classifier action

434 man pages section 7: Device and Network Interfaces • Last Revised 25 Feb 2002
 ipqos(7ipp)

To verify classifier configuration, generate traffic for each of the configured classes and check
that the statistic counters for the expected class are increased. If you're unsure about the
parameters for your traffic, you can use snoop(1M) to determine them.

Some actions have the instance id of the next configured action in their statistics. This instance
id can be used to follow the action processing path. Instance id's -1 and -2 are the built-in
actions continue and drop, respectively.

Examples:

To retrieve all statistics for ipgpc:

kstat -m ipgpc

To retrieve statistics for the class http:

kstat -m ipgpc -c http

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos (32–bit) SUNWqosx (64–bit)

See Also ipqosconf(1M), dlcosmk(7ipp), dscpmk(7ipp), flowacct(7ipp), ipgpc(7ipp), tokenmt(7ipp),

tswtclmt(7ipp)

RFC 2475, An Architecture for Differentiated Services S. Blake, D. Black, M. Carlson, E. Davies,
Z. Wang, W. Weiss — The Internet Society, 1998

RFC 2474, Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6
Headers K. Nichols, S. Blake, F. Baker, D. Black — The Internet Society, 1998

RFC 2697, A Single Rate Three Color Marker J. Heinanen, R. Guerin — The Internet Society,
1999

RFC 2698, A Two Rate Three Color Marker J. Heinanen, R. Guerin — The Internet Society,
1999

RFC 2859, A Time Sliding Window Three Colour Marker (TSWTCM) W. Fang, N. Seddigh, B.
Nandy — The Internet Society, 2000

Device and Network Interfaces 435

iprb(7D)

Name iprb – Intel 82557, 82558, 82559–controlled network interface controllers

Synopsis /dev/iprb
Description The iprb Ethernet driver is a multi-threaded, loadable, clonable, STREAMS hardware driver
supporting the connectionless Data Link Provider Interface, dlpi(7P), over Intel D100 82557,
82558, and 82559 controllers. Multiple 82557, 82558, and 82559 controllers installed within
the system are supported by the driver. The iprb driver provides basic support for the 82557,
82558, and 82559 hardware. Functions include chip initialization, frame transmit and receive,
multicast support, and error recovery and reporting.
Application The cloning, character-special device /dev/iprb is used to access all 82557, 82558, and 82559
Programming devices installed within the system.
Interface
iprb and DLPI The iprb driver is dependent on /kernel/misc/gld, a loadable kernel module that provides
the iprb driver with the DLPI and STREAMS functionality required of a LAN driver. See
gld(7D) for more details on the primitives supported by the driver.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ Maximum SDU is 1500 (ETHERMTU).
■ Minimum SDU is 0. The driver will pad to the mandatory 60-octet minimum packet size.
■ The dlsap address length is 8.
■ MAC type is DL_ETHER.
■ The sap length value is −2, meaning the physical address component is followed
immediately by a 2-byte sap component within the DLSAP address.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Known Problems x86 based systems with the Intel EtherExpress PRO/100B or the Intel EtherExpress PRO/100+
And Limitations might hang when the interface is brought down at the very instant that a packet is being
received. To avoid this, wait until the system is experiencing light or no network traffic before
bringing the interface down.
Early versions of the firmware on Intel EtherExpress PRO/100+ and Intel PRO/100+
Management adapters do not support PXE network boot on Solaris systems. Upgrade the
firmware if the version is lower than 078. PXE firmware versions are expressed as three-digit
build numbers. The build number is typically displayed by the firmware during boot. If the
PXE build number is not displayed during boot, change the system BIOS or adapter BIOS
configuration to display PXE messages during boot.
Files iprb Device special file
/kernel/drv/iprb.conf iprb configuration file
<sys/stropts.h> stropts network header file
<sys/ethernet.h> Ethernet network header file

436 man pages section 7: Device and Network Interfaces • Last Revised 17 November 2000
 iprb(7D)

<sys/dlpi.h> dlpi network header file

<sys/gld.h> gld network header file

The iprb.conf configuration file options include:

-TxURRetry Default: 3

Allowed Values: 0, 1, 2, 3

Sets the number of retransmissions. Modified when tuning performance.

-MWIEnable Default: 0 (Disable)

Allowed Values: 0 (Disable), 1 (Enable)

Should only be set for 82558 adapters and systems in which the PCI bus
supports Memory Write & Invalidate operations. Can improve the
performance for some configurations.
-FlowControl Default: 0 (Disable)

Allowed Values: 0 (Disable), 1 (Enable)

Setting this value can improve the performance for some configurations
-CollisionBackOffModification Default: 0 (Disable)

Allowed Values: 0 (Disable), 1 (Enable)

Setting this value can improve the performance for

some configurations
-PhyErrataFrequency Default: 0 (Disable)

Allowed Values: 0 (Disable), 10 (Enable)

If you have problems establishing links with cables length = 70 Ft,

set this field to 10
-CpuCycleSaver Default: 0

Allowed Values: 1 through FFFFh

Reasonable Values: 200h through 800h

The CPUSaver algorithm improves the system's P/E ratio by reducing

the number of interrupts generated by the card. The algorithm bundles
multiple receive frames together, then generates a single interrupt for
the bundle. Because the microcode does not support run-time

Device and Network Interfaces 437

iprb(7D)

configuration, configuration must be done prior to the micro code being

loaded into the chip. Changing this value from its default means that the
driver will have to be unloaded and loaded for the change to take affect.
Setting the CpuCycleSaver option to 0 prevents the algorithm from
being used. Because it varies for different network environments, the
optimal value for this parameter is impossible to predict. Accordingly,
developers should run tests to determine the effect that changing this
value has on bandwidth and CPU utilization.
-ForceSpeedDuplex Default: 5 (Auto-negotiate)

Allowed Values: 4 (100 FDX)

3 (100 HDX)

2 (10 FDX)

1 (10 HDX)

Specify the speed and duplex mode for each instance.

Example: ForceSpeedDuplex=5,4;

Sets iprb0 to autonegotiate and iprb1 to 100 FDX.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), dlpi(7P), gld(7D)

438 man pages section 7: Device and Network Interfaces • Last Revised 17 November 2000
 ipsec(7P)

Name ipsec – Internet Protocol Security Architecture

Description The IP Security Architecture (IPsec) provides protection for IP datagrams. The protection can
include confidentiality, strong integrity of the data, partial sequence integrity (replay
protection), and data authentication. IPsec is performed inside the IP processing, and it can be
applied with or without the knowledge of an Internet application.

IPsec applies to both IPv4 and IPv6. See ip(7P) and ip6(7P).

Protection Mechanisms IPsec provides two mechanisms for protecting data. The Authentication Header (AH)
provides strong integrity, replay protection, and data authentication. AH protects as much of
the IP datagram as it can. AH cannot protect fields that change nondeterministically between
sender and receiver.

The Encapsulating Security Payload (ESP) provides confidentiality over what it encapsulates,
as well as the services that AH provides, but only over that which it encapsulates. ESP's
authentication services are optional, which allow ESP and AH to be used together on the same
datagram without redundancy.

Authentication and encryption algorithms are used for IPsec. Authentication algorithms
produce an integrity checksum value or “digest”based on the data and a key. Encryption
algorithms operate on data in units of a “block size”.

NAT Traversal IPsec's ESP can also encapsulate itself in UDP if IKE (see in.iked(1M)) discovers a Network
Address Translator (NAT) between two communicating endpoints.

A UDP socket can be specified as a NAT-Traversal endpoint. See udp(7P) for details.

Security Associations AH and ESP use Security Associations (SA). SA's are entities that specify security properties
from one host to another. Two communicating machines require two SAs (at a minimum) to
communicate securely. However, communicating machines that use multicast can share the
same multicast SA. SAs are managed through the pf_key(7P) interface. For IPv4, automatic
SA management is available through the Internet Key Exchange (IKE), as implemented by
in.iked(1M). A command-line front-end is available by means of ipseckey(1M). An IPsec
SA is identified by a tuple of <AH or ESP, destination IP address, and SPI>. The Security
Parameters Index (SPI) is an arbitrary 32-bit value that is transmitted on the wire with an AH
or ESP packet. See ipsecah(7P) or ipsecesp(7P) for an explanation about where the SPI falls
in a protected packet.

Protection Policy and Mechanism and policy are separate. The policy for applying IPsec is enforced on a
Enforcement system-wide or per-socket level. Configuring system-wide policy and per-tunnel policy (see
Mechanisms
Transport Mode and Tunnel Mode sections) is done via the ipsecconf(1M) command.
Configuring per-socket policy is discussed later in this section.

System-wide IPsec policy is applied to incoming and outgoing datagrams. Some additional
rules can be applied to outgoing datagrams because of the additional data known by the
system. Inbound datagrams can be accepted or dropped. The decision to drop or accept an

Device and Network Interfaces 439

ipsec(7P)

inbound datagram is based on several criteria which sometimes overlap or conflict. Conflict
resolution is resolved by which rule is parsed first, with one exception: if a policy entry states
that traffic should bypass all other policy, it is automatically be accepted. Outbound datagrams
are sent with or without protection. Protection may (or may not) indicate specific algorithms.
If policy normally would protect a datagram, it can be bypassed either by an exception in
system-wide policy or by requesting a bypass in per-socket policy.

Intra-machine traffic policies are enforced, but actual security mechanisms are not applied.
Instead, the outbound policy on an intra-machine packet translates into an inbound packet
with those mechanisms applied.

IPsec policy is enforced in the ip(7P) driver. Several ndd tunables for /dev/ip affect policy
enforcement, including:
icmp_accept_clear_messages If equal to 1 (the default), allow certain cleartext icmp
messages to bypass policy. For ICMP echo requests
(“ping”messages), protect the response like the request. If
zero, treat icmp messages like other IP traffic.
igmp_accept_clear_messages If 1, allow inbound cleartext IGMP messages to bypass IPsec
policy.
pim_accept_clear_messages If 1, allow inbound cleartext PIM messages to bypass IPsec
policy.
ipsec_policy_log_interval IPsec logs policy failures and errors to /var/adm/messages.
To prevent syslog from being overloaded, the IPsec kernel
modules limit the rate at which errors can be logged. You
can query/set ipsec_policy_log_interval using ndd(1M).
The value is in milliseconds. Only one message can be
logged per interval.

Transport Mode and If IPsec is used on a tunnel, Tunnel Mode IPsec can be used to protect distinct flows within a
Tunnel Mode tunnel or to cause packets that do not match per-tunnel policy to drop. System-wide policy is
always Transport Mode. A tunnel can use Transport Mode IPsec or Tunnel Mode IPsec.

Per-Socket Policy The IP_SEC_OPT or IPV6_SEC_OPT socket option is used to set per-socket IPsec policy. The
structure used for an IP_SEC_OPT request is:
typedef struct ipsec_req {
uint_t ipsr_ah_req; /* AH request */
uint_t ipsr_esp_req; /* ESP request */
uint_t ipsr_self_encap_req; /* Self-Encap request */
uint8_t ipsr_auth_alg; /* Auth algs for AH */
uint8_t ipsr_esp_alg; /* Encr algs for ESP */
uint8_t ipsr_esp_auth_alg; /* Auth algs for ESP */
} ipsec_req_t;

440 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2009
 ipsec(7P)

The IPsec request has fields for both AH and ESP. Algorithms may or may not be specified.
The actual request for AH or ESP services can take one of the following values:
IPSEC_PREF_NEVER Bypass all policy. Only the superuser may request this service.
IPSEC_PREF_REQUIRED Regardless of other policy, require the use of the IPsec service.

The following value can be logically ORed to an IPSEC_PREF_REQUIRED value:

IPSEC_PREF_UNIQUE Regardless of other policy, enforce a unique SA for traffic originating
from this socket.

In the event IP options not normally encapsulated by ESP need to be, the
ipsec_self_encap_req is used to add an additional IP header outside the original one.
Algorithm values from <net/pfkeyv2.h> are as follows:
SADB_AALG_MD5HMAC Uses the MD5-HMAC (RFC 2403) algorithm for authentication.
SADB_AALG_SHA1HMAC Uses the SHA1-HMAC (RFC 2404) algorithm for authentication.
SADB_EALG_DESCBC Uses the DES (RFC 2405) algorithm for encryption.
SADB_EALG_3DESCBC Uses the Triple DES (RFC 2451) algorithm for encryption.
SADB_EALG_BLOWFISH Uses the Blowfish (RFC 2451) algorithm for encryption.
SADB_EALG_AES Uses the Advanced Encryption Standard algorithm for
encryption.
SADB_AALG_SHA256HMAC
SADB_AALG_SHA384HMAC
SADB_AALG_SHA512HMAC Uses the SHA2 hash algorithms with HMAC (RFC 4868)for
authentication.

An application should use either the getsockopt(3SOCKET) or the setsockopt(3SOCKET)

call to manipulate IPsec requests. For example:
#include <sys/socket.h>
#include <netinet/in.h>
#include <net/pfkeyv2.h> /* For SADB_*ALG_* */
/* .... socket setup skipped */
rc = setsockopt(s, IPPROTO_IP, IP_SEC_OPT,
(const char *)&ipsec_req, sizeof (ipsec_req_t));

Security While IPsec is an effective tool in securing network traffic, it will not make security problems
disappear. Security issues beyond the mechanisms that IPsec offers may be discussed in
similar ”Security” or “Security Consideration” sections within individual reference manual
pages.

While a non-root user cannot bypass IPsec, a non-root user can set policy to be different from
the system-wide policy. For ways to prevent this, consult the ndd(1M) variables in /dev/ip.

Device and Network Interfaces 441

ipsec(7P)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Committed

See Also in.iked(1M), ipsecconf(1M), ipseckey(1M), ndd(1M), getsockopt(3SOCKET),

setsockopt(3SOCKET), attributes(5), inet(7P), ip(7P), ip6(7P), ipsecah(7P),
ipsecesp(7P), pf_key(7P), udp(7P)

Kent, S., and Atkinson, R., RFC 2401, Security Architecture for the Internet Protocol, The
Internet Society, 1998.

Kent, S. and Atkinson, R., RFC 2406, IP Encapsulating Security Payload (ESP), The Internet
Society, 1998.

Madson, C., and Doraswamy, N., RFC 2405, The ESP DES-CBC Cipher Algorithm with Explicit
IV, The Internet Society, 1998.

Madsen, C. and Glenn, R., RFC 2403, The Use of HMAC-MD5-96 within ESP and AH, The
Internet Society, 1998.

Madsen, C. and Glenn, R., RFC 2404, The Use of HMAC-SHA-1-96 within ESP and AH, The
Internet Society, 1998.

Pereira, R. and Adams, R., RFC 2451, The ESP CBC-Mode Cipher Algorithms, The Internet
Society, 1998.

Kelly, S. and Frankel, S., RFC 4868, Using HMAC-SHA-256, HMAC-SHA-384, and
HMAC-SHA-512 with IPsec, 2007.

Huttunen, A., Swander, B., Volpe, V., DiBurro, L., Stenberg, M., RFC 3948, UDP
Encapsulation of IPsec ESP Packets, The Internet Society, 2005.

442 man pages section 7: Device and Network Interfaces • Last Revised 25 Sep 2009
 ipsecah(7P)

Name ipsecah, AH – IPsec Authentication Header

Synopsis drv/ipsecah

Description The ipsecah module (AH) provides strong integrity, authentication, and partial sequence
integrity (replay protection) to IP datagrams. AH protects the parts of the IP datagram that
can be predicted by the sender as it will be received by the receiver. For example, the IP TTL
field is not a predictable field, and is not protected by AH.

AH is inserted between the IP header and the transport header. The transport header can be
TCP, UDP, ICMP, or another IP header, if tunnels are being used.

AH Device AH is implemented as a module that is auto-pushed on top of IP. The entry /dev/ipsecah is
used for tuning AH with ndd(1M).

Authentication Current authentication algorithms supported include HMAC-MD5 and HMAC-SHA-1. Each
Algorithms authentication algorithm has its own key size and key format properties. You can obtain a list
of authentication algorithms and their properties by using the ipsecalgs(1M) command.
You can also use the functions described in the getipsecalgbyname(3NSL) man page to
retrieve the properties of algorithms.

Security Without replay protection enabled, AH is vulnerable to replay attacks. AH does not protect
Considerations against eavesdropping. Data protected with AH can still be seen by an adversary.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsr

Interface Stability Committed

See Also ipsecalgs(1M), ipsecconf(1M), ndd(1M), attributes(5), getipsecalgbyname(3NSL),

ip(7P), ipsec(7P), ipsecesp(7P)

Kent, S. and Atkinson, R.RFC 2402, IP Authentication Header, The Internet Society, 1998.

Device and Network Interfaces 443

ipsecesp(7P)

Name ipsecesp, ESP – IPsec Encapsulating Security Payload

Synopsis drv/ipsecesp

Description The ipsecesp module provides confidentiality, integrity, authentication, and partial sequence
integrity (replay protection) to IP datagrams. The encapsulating security payload (ESP)
encapsulates its data, enabling it to protect data that follows in the datagram. For TCP packets,
ESP encapsulates the TCP header and its data only. If the packet is an IP in IP datagram, ESP
protects the inner IP datagram. Per-socket policy allows "self-encapsulation" so ESP can
encapsulate IP options when necessary. See ipsec(7P).

Unlike the authentication header (AH), ESP allows multiple varieties of datagram protection.
(Using a single datagram protection form can expose vulnerabilities.) For example, only ESP
can be used to provide confidentiality. But protecting confidentiality alone exposes
vulnerabilities in both replay attacks and cut-and-paste attacks. Similarly, if ESP protects only
integrity and does not fully protect against eavesdropping, it may provide weaker protection
than AH. See ipsecah(7P).

ESP Device ESP is implemented as a module that is auto-pushed on top of IP. Use the /dev/ipsecesp
entry to tune ESP with ndd(1M).

Algorithms ESPuses encryption and authentication algorithms. Authentication algorithms include

HMAC-MD5 and HMAC-SHA-1. Encryption algorithms include DES, Triple-DES, Blowfish
and AES. Each authentication and encryption algorithm contain key size and key format
properties. You can obtain a list of authentication and encryption algorithms and their
properties by using the ipsecalgs(1M) command. You can also use the functions described
in the getipsecalgbyname(3NSL) man page to retrieve the properties of algorithms. Because
of export laws in the United States, not all encryption algorithms are available outside of the
United States.

Security ESP without authentication exposes vulnerabilities to cut-and-paste cryptographic attacks as

Considerations well as eavesdropping attacks. Like AH, ESP is vulnerable to eavesdropping when used
without confidentiality.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsr (32-bit)

Interface Stability Evolving

See Also ipsecalgs(1M), ipsecconf(1M), ndd(1M), attributes(5), getipsecalgbyname(3NSL),

ip(7P), ipsec(7P), ipsecah(7P)

Kent, S. and Atkinson, R.RFC 2406, IP Encapsulating Security Payload (ESP), The Internet
Society, 1998.

444 man pages section 7: Device and Network Interfaces • Last Revised 18 May 2003
 ipw(7D)

Name ipw – Intel Pro. Wireless 802.11b IPW2100B driver

Description The ipw 802.11b wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on Intel Pro
Wireless 2100B chipset-based wireless NIC's. Driver functions include controller
initialization, wireless 802.11b infrastructure network connection, WEP, frame transmit and
receive and promiscuous support.

Driver The ipw driver performs auto-negotiation to determine the data rate and mode. Supported
Configuration 802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec.

The ipw driver supports only BSS networks (also known as "ap" or "infrastructure" networks)
and the "open" ("open-system") or "shared system" authentication. Only WEP encryption is
currently supported. You perform configuration and administration tasks using the
dladm(1M) and wificonfig(1M) utilities.
Files /dev/ipw* Special character device.
/kernel/drv/ipw 32-bit ELF kernel module (x86).
/kernel/drv/amd64/ipw 64-bit ELF kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability x86

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

ANSI/IEEE Std 802.11- Standard for Wireless LAN Technology — 1999

IEEE Std 802.11b - Standard for Wireless LAN Technology-Rev. B - 2003

Device and Network Interfaces 445

iscsi(7D)

Name iscsi – iSCSI software initiator driver and service

Description The iscsi driver is a software initiator that transports SCSI commands over TCP/IP as
described in RFC 3720.

The initiator driver is administered through iscsiadm(1M). The iscsi initiator service is
managed by the service management facility, smf(5), under the following FMRI:
svc:/network/iscsi/initiator:default

The iscsi initiator acts as a host adapter driver that attaches the appropriate target driver, for
example, sd(7D) for disks, or st(7D) for tapes) for devices it discovers. See the System
Administration Guide: Devices and File Systems for more information.

Once enabled, the iscsi initiator service ensures the right timing to start the discovery and
enumeration of iSCSI devices during boot, but it doesn't guarantee the success of discovery for
certain iSCSI devices. If the service is disabled, iscsi driver stops the discovery and
enumeration of iSCSI devices and also tries to offline all existing iSCSI devices. iscsiadm(1M)
works only when the service is enabled.

iSCSI boot(1M) is not affected by the status of the iscsi initiator service.
Files /kernel/drv/iscsi 32-bit ELF kernel driver
/kernel/drv/sparcv9/iscsi 64-bit SPARC ELF kernel driver
/kernel/drv/amd64/iscsi 64-bit AMD64 ELF kernel driver
/kernel/drv/iscsi.conf Driver configuration file
/etc/iscsi/* iscsi persistent store

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWiscsir, SUNWiscsiu

See Also iscsiadm(1M), attributes(5), smf(5), sd(7D), st(7D)

RFC 3720 Internet Small Computer Systems Interface (iSCSI)

System Administration Guide: Devices and File Systems

446 man pages section 7: Device and Network Interfaces • Last Revised 30 Mar 2009
 isdnio(7I)

Name isdnio – ISDN interfaces

Synopsis #include <sun/audioio.h>
#include <sun/isdnio.h>

int ioctl(int fd, int command, /* arg */ ...);

Description ISDN ioctl commands are a subset of ioctl(2) commands that perform a variety of control
functions on Integrated Services Digital Network (ISDN) STREAMS devices. The arguments
command and arg are passed to the file designated by fd and are interpreted by the ISDN device
driver.

fd is an open file descriptor that refers to a stream. command determines the control function to
be performed as described in the IOCTLS section of this document. arg represents additional
information that is needed by command. The type of arg depends upon the command, but
generally it is an integer or a pointer to a command-specific data structure.

Since these ISDN commands are a subset of ioctl and streamio(7I), they are subject to errors
as described in those interface descriptions.

This set of generic ISDN ioctl commands is meant to control various types of ISDN
STREAMS device drivers. The following paragraphs give some background on various types
of ISDN hardware interfaces and data formats, and other device characteristics.

Controllers, Interfaces, This manual page discusses operations on, and facilities provided by ISDN controllers,
and Channels interfaces and channels. A controller is usually a hardware peripheral device that provides one
or more ISDN interfaces and zero or more auxiliary interfaces. In this context, the term
interface is synonymous with the term “port”. Each interface can provide one or more
channels.

Time Division ISDN BRI-TE, BRI-NT, and PRI interfaces are all examples of Time Division Multiplexed
Multiplexed Serial Serial Interfaces. As an example, a Basic Rate ISDN (BRI) Terminal Equipment (TE) interface
Interfaces
provides one D-channel and two B-channels on the same set of signal wires. The BRI
interface, at the S reference point, operates at a bit rate of 192,000 bits per second. The bits are
encoded using a pseudoternary coding system that encodes a logic one as zero volts, and a
logic zero as a positive or negative voltage. Encoding rules state that adjacent logic zeros must
be encoded with opposite voltages. Violations of this rule are used to indicate framing
information such that there are 4000 frames per second, each containing 48 bits. These 48 bits
are divided into channels. Not including framing and synchronization bits, the frame is
divided into 8 bits for the B1-channel, 1 bit for the D-channel, 8 bits for B2, 1 bit for D, 8 bits
for B1, 1 bit for D, and 8 bits for B2. This results in a 64,000 bps B1-channel, a 64,000 bps
B2-channel, and a 16,000 bps D-channel, all on the same serial interface.

Basic Rate ISDN A Basic Rate ISDN (BRI) interface consists of a 16000 bit per second Delta Channel
(D-channel) for signaling and X.25 packet transmission, and two 64000 bit per second Bearer
Channels (B-channels) for transmission of voice or data.

Device and Network Interfaces 447

isdnio(7I)

The CCITT recommendations on ISDN Basic Rate interfaces, I.430, identify several
“reference points” for standardization. From (Stallings89): Reference point T (terminal)
corresponds to a minimal ISDN network termination at the customer's premises. It separates
the network provider's equipment from the user's equipment. Reference point S (system)
corresponds to the interface of individual ISDN terminals. It separates user terminal
equipment from network-related communications functions. Reference point R (rate)
provides a non-ISDN interface between user equipment that is not ISDN-compatible and
adaptor equipment. . . . The final reference point . . . is reference point U (user). This
interface describes the full-duplex data signal on the subscriber line.

Some older technology components of some ISDN networks occasionally steal the low order
bit of an ISDN B-channel octet in order to transmit in-band signaling information between
switches or other components of the network. Even when out-of-band signaling has been
implemented in these networks, and the in-band signaling is no longer needed, the
bit-robbing mechanism may still be present. This bit robbing behavior does not appreciably
affect a voice call, but it will limit the usable bandwidth of a data call to 56000 bits per second
instead of 64000 bits per second. These older network components only seem to exist in the
United States of America, Canada and Japan. ISDN B-channel data calls that have one end
point in the United States, Canada or Japan may be limited to 56000 bps usable bandwidth
instead of the normal 64000 bps. Sometimes the ISDN service provider may be able to supply
56kbps for some calls and 64kbps for other calls. On an international call, the local ISDN
service provider may advertise the call as 64kbps even though only 56kbps are reliably
delivered because of bit-robbing in the foreign ISDN that is not reported to the local switch.

A Basic Rate Interface implements either a Terminal Equipment (TE) interface or a Network
Termination (NT) interface. TE's can be ISDN telephones, a Group 4 fax, or other ISDN
terminal equipment. A TE connects to an NT in order to gain access to a public or private
ISDN network. A private ISDN network, such as provided by a Private Branch Exchange
(PBX), usually provides access to the public network.

If multi-point configurations are allowed by an NT, it may be possible to connect up to eight

TE's to a single NT interface. All of the TE's in a multipoint configuration share the same D
and B-channels. Contention for B-Channels by multiple TEs is resolved by the ISDN switch
(NT) through signaling protocols on the D-channel.

Contention for access to the D-channel is managed by a collision detection and priority
mechanism. D-channel call control messages have higher priority than other packets. This
media access function is managed at the physical layer.

A BRI-TE interface may implement a “Q-channel”, the Q-channel is a slow speed, 800 bps,
data path from a TE to an NT. Although the structure of the Q-channel is defined in the I.430
specification, the use of the Q-channel is for further study.

A BRI-NT interface may implement an “S-channel”, the S-channel is a slow speed, 4000 bps,
data path from a NT to an TE. The use of the S-channel is for further study.

448 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

Primary Rate ISDN Primary Rate ISDN (PRI) interfaces are either 1.544Mbps (T1 rate) or 2.048Mbps (E1 rate)
and are typically organized as 23 B-channels and one D-Channel (23B+D) for T1 rates, and 30
B-Channels and one D-Channel (30B+D) for E1 rates. The D-channels on a PRI interface
operate at 64000 bits per second. T1 rate PRI interface is the standard in the United States,
Canada and Japan while E1 rate PRI interface is the standard in European countries. Some E1
rate PRI interface implementations allow access to channel zero which is used for framing.

Channel Types ISDN channels fall into several categories; D-channels, bearer channels, and management
pseudo channels. Each channel has a corresponding device name somewhere under the
directory /dev/isdn/ as documented in the appropriate hardware specific manual page.
D-channels There is at most one D-channel per ISDN interface. The
D-channel carries signaling information for the
management of ISDN calls and can also carry X.25 packet
data. In the case of a PRI interface, there may actually be
no D-channel if Non-Facility Associated Signaling is used.
D-channels carry data packets that are framed and
checked for transmission errors according to the LAP-D
protocol. LAP-D uses framing and error checking
identical to the High Speed Data Link (HDLC) protocol.
B-channels BRI interfaces have two B-channels, B1 and B2. On a BRI
interface, the only other type of channel is an H-channel
which is a concatenation of the B1 and B2 channels. An
H-channel is accessed by opening the “base” channel, B1
in this case, and using the ISDN_SET_FORMAT ioctl to
change the configuration of the B-channel from 8-bit, 8
kHz to 16-bit, 8kHz.

On a primary rate interface, B channels are numbered

from 0 to 31 in Europe and 1 to 23 in the United States,
Canada and Japan.
H-Channels A BRI or PRI interface can offer multiple B-channels
concatenated into a single, higher bandwidth channel.
These concatenated B-channels are referred to as an
“H-channels” on a BRI interface. The PRI interface
version of an H-channel is referred to as an Hn-channels
where n is a number indicating how the B-channels have
been aggregated into a single channel.
■ A PRI interface H0 channel is 384 kbps allowing
3H0+D on a T1 rate PRI interface and 4H0+D
channels on an E1 rate PRI interface.

Device and Network Interfaces 449

isdnio(7I)

■ A T1 PRI interface H11 channel is 1536 kbps

(24×64000bps). This will consume the channel
normally reserved for the D-channel, so signaling
must be done with Non-Facility Associated Signaling
(NFAS) from another PRI interface.
■ An E1 PRI interface H12 channel is 1920 kbps
(30×64000bps). An H12-channel leaves room for the
framing-channel as well as the D-channel.
Auxiliary channels Auxiliary channels are non-ISDN hardware interfaces
that are closely tied to the ISDN interfaces. An example
would be a video or audio coder/decoder (codec). The
existence of an auxiliary channel usually implies that one
or more B-channels can be “connected” to an auxiliary
interface in hardware.
Management pseudo-channels A management pseudo-channel is used for the
management of a controller, interface, or hardware
channel. Management channels allow for out-of-band
control of hardware interfaces and for out-of-band
notification of status changes. There is at least one
management device per hardware interface.

There are three different types of management channels

implemented by ISDN hardware drivers:
■ A controller management device handles all ioctls that
simultaneously affect hardware channels on different
interfaces. Examples include resetting a controller,
mu-code (as in the Greek letter mu) downloading of a
controller, or the connection of an ISDN B-channel to
an auxiliary channel that represents an audio
coder/decoder (codec). The latter case would be
accomplished using the ISDN_SET_CHANNEL ioctl.
■ An interface management device handles all ioctls that
affect multiple channels on the same interface.
Messages associated with the activation and
deactivation of an interface arrive on the management
device associated with the D channel of an ISDN
interface.
■ Auxiliary interfaces may also have management
devices. See the hardware specific man pages for
operations on auxiliary devices.

450 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

Trace pseudo-channels A device driver may choose to implement a trace device

for a data or management channel. Trace channels receive
a special M_PROTO header with the original channel's
original M_PROTO or M_DATA message appended to the
special header. The header is described by:
typedef struct {
uint_t seq; /* Sequence number */
int type; /* device dependent */
struct timeval timestamp;
char _f[8]; /* filler */
} audtrace_hdr_t;

ISDN Channel types The isdn_chan_t type enumerates the channels available on ISDN interfaces. If a particular
controller implements any auxiliary channels then those auxiliary channels will be described
in a controller specific manual page. The defined channels are described by the isdn_chan_t
type as shown below:
/* ISDN channels */
typedef enum {
ISDN_CHAN_NONE = 0x0, /* No channel given */
ISDN_CHAN_SELF, /* The channel performing the ioctl */
ISDN_CHAN_HOST, /* Unix STREAMS*/
ISDN_CHAN_CTRL_MGT, /* Controller management */

/* TE channel defines */

ISDN_CHAN_TE_MGT, /* Receives activation/deactivation */

ISDN_CHAN_TE_D_TRACE, /* Trace device for protocol analysis apps */
ISDN_CHAN_TE_D,
ISDN_CHAN_TE_B1,
ISDN_CHAN_TE_B2,

/* NT channel defines */

ISDN_CHAN_NT_MGT, /* Receives activation/deactivation */

ISDN_CHAN_NT_D_TRACE, /* Trace device for protocol analysis apps */
ISDN_CHAN_NT_D,
ISDN_CHAN_NT_B1,
ISDN_CHAN_NT_B2,

/* Primary rate ISDN */

ISDN_CHAN_PRI_MGT,
ISDN_CHAN_PRI_D,
ISDN_CHAN_PRI_B0, ISDN_CHAN_PRI_B1,
ISDN_CHAN_PRI_B2, ISDN_CHAN_PRI_B3,
ISDN_CHAN_PRI_B4, ISDN_CHAN_PRI_B5,

Device and Network Interfaces 451

isdnio(7I)

ISDN_CHAN_PRI_B6, ISDN_CHAN_PRI_B7,
ISDN_CHAN_PRI_B8, ISDN_CHAN_PRI_B9,
ISDN_CHAN_PRI_B10, ISDN_CHAN_PRI_B11,
ISDN_CHAN_PRI_B12, ISDN_CHAN_PRI_B13,
ISDN_CHAN_PRI_B14, ISDN_CHAN_PRI_B15,
ISDN_CHAN_PRI_B16, ISDN_CHAN_PRI_B17,
ISDN_CHAN_PRI_B18, ISDN_CHAN_PRI_B19,
ISDN_CHAN_PRI_B20, ISDN_CHAN_PRI_B21,
ISDN_CHAN_PRI_B22, ISDN_CHAN_PRI_B23,
ISDN_CHAN_PRI_B24, ISDN_CHAN_PRI_B25,
ISDN_CHAN_PRI_B26, ISDN_CHAN_PRI_B27,
ISDN_CHAN_PRI_B28, ISDN_CHAN_PRI_B29,
ISDN_CHAN_PRI_B30, ISDN_CHAN_PRI_B31,

/* Auxiliary channel defines */

ISDN_CHAN_AUX0, ISDN_CHAN_AUX1, ISDN_CHAN_AUX2, ISDN_CHAN_AUX3,

ISDN_CHAN_AUX4, ISDN_CHAN_AUX5, ISDN_CHAN_AUX6, ISDN_CHAN_AUX7
} isdn_chan_t;

ISDN Interface types The isdn_interface_t type enumerates the interfaces available on ISDN controllers. The
defined interfaces are described by the isdn_interface_t type as shown below:
/* ISDN interfaces */
typedef enum {
ISDN_TYPE_UNKNOWN = -1, /* Not known or applicable */
ISDN_TYPE_SELF = 0, /*
* For queries, application may
* put this value into "type" to
* query the state of the file
* descriptor used in an ioctl.
*/
ISDN_TYPE_OTHER, /* Not an ISDN interface */
ISDN_TYPE_TE,
ISDN_TYPE_NT,
ISDN_TYPE_PRI,
} isdn_interface_t;

Activation and The management device associated with an ISDN D-channel is used to request activation,
Deactivation of ISDN deactivation and receive information about the activation state of the interface. See the
Interfaces
descriptions of the ISDN_PH_ACTIVATE_REQ and ISDN_MPH_DEACTIVATE_REQ ioctls. Changes
in the activation state of an interface are communicated to the D-channel application through
M_PROTO messages sent up-stream on the management device associated with the D-channel.
If the D-channel protocol stack is implemented as a user process, the user process can retrieve
the M_PROTO messages using the getmsg(2) system call.

These M_PROTO messages have the following format:

452 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

typedef struct isdn_message {

unsigned int magic; /* set to ISDN_PROTO_MAGIC */
isdn_interface_t type; /* Interface type */
isdn_message_type_t message; /* CCITT or vendor Primitive */
unsigned int vendor[5]; /* Vendor specific content */
} isdn_message_t;
typedef enum isdn_message_type {
ISDN_VPH_VENDOR = 0, /* Vendor specific messages */
ISDN_PH_AI, /* Physical: Activation Ind */
ISDN_PH_DI, /* Physical: Deactivation Ind */
ISDN_MPH_AI, /* Management: Activation Ind */
ISDN_MPH_DI, /* Management: Deactivation Ind */
ISDN_MPH_EI1, /* Management: Error 1 Indication */
ISDN_MPH_EI2, /* Management: Error 2 Indication */
ISDN_MPH_II_C, /* Management: Info Ind, connection */
ISDN_MPH_II_D /* Management: Info Ind, disconn. */
} isdn_message_type_t;

ioctls

STREAMS IOCTLS All of the streamio(7I) ioctl commands may be issued for a device conforming to the the
isdnio interface.

ISDN interfaces that allow access to audio data should implement a reasonable subset of the
audio(7I) interface.
ISDN ioctls ISDN_PH_ACTIVATE_REQ Request ISDN physical layer activation. This command is valid
for both TE and NT interfaces. fd must be a D-channel file
descriptor. arg is ignored.

TE activation will occur without use of the

ISDN_PH_ACTIVATE_REQ ioctl if the device corresponding to
the TE D-channel is open, “on”, and the ISDN switch is
requesting activation.
ISDN_MPH_DEACTIVATE_REQ fd must be an NT D-channel file descriptor. arg is ignored.

This command requests ISDN physical layer de-activation.

This is not valid for TE interfaces. A TE interace may be
turned off by use of the ISDN_PARAM_POWER command or by
close(2) on the associated fd.
ISDN_ACTIVATION_STATUS fd is the file descriptor for a D-channel, the management
device associated with an ISDN interface, or the management
device associated with the controller. arg is a pointer to an
isdn_activation_status_t structure. Although it is
possible for applications to determine the current activation
state with this ioctl, a D-channel protocol stack should

Device and Network Interfaces 453

isdnio(7I)

instead process messages from the management pseudo

channel associated with the D-channel.
typedef struct isdn_activation_status {
isdn_interface_t type;
enum isdn_activation_state activation;
} isdn_activation_status_t;
typedef enum isdn_activation_state {
ISDN_OFF = 0, /* Interface is powered down */
ISDN_UNPLUGGED, /* Power but no-physical connection */
ISDN_DEACTIVATED_REQ, /* Pending Deactivation, NT Only */
ISDN_DEACTIVATED, /* Activation is permitted */
ISDN_ACTIVATE_REQ, /* Attempting to activate */
ISDN_ACTIVATED, /* Interface is activated */
} isdn_activation_state_t;
The type field should be set to ISDN_TYPE_SELF. The device
specific interface type will be returned in the type field.

The isdn_activation_status_t structure contains the

interface type and the current activation state. type is the
interface type and should be set by the caller to
ISDN_TYPE_SELF.
ISDN_INTERFACE_STATUS The ISDN_INTERFACE_STATUS ioctl retrieves the status and
statistics of an ISDN interface. The requesting channel must
own the interface whose status is being requested or the ioctl
will fail. fd is the file descriptor for an ISDN interface
management device. arg is a pointer to a struct
isdn_interface_info. If the interface field is set to
ISDN_TYPE_SELF, it will be changed in the returned structure
to reflect the proper device-specific interface of the
requesting fd.
typedef struct isdn_interface_info {
isdn_interface_t interface;
enum isdn_activation_state activation;
unsigned int ph_ai; /* Physical: Activation Ind */
unsigned int ph_di; /* Physical: Deactivation Ind */
unsigned int mph_ai; /* Management: Activation Ind */
unsigned int mph_di; /* Management: Deactivation Ind */
unsigned int mph_ei1; /* Management: Error 1 Indication */
unsigned int mph_ei2; /* Management: Error 2 Indication */
unsigned int mph_ii_c; /* Management: Info Ind, connection */
unsigned int mph_ii_d; /* Management: Info Ind, disconn. */
} isdn_interface_info_t;

454 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

ISDN_CHANNEL_STATUS The ISDN_CHANNEL_STATUS ioctl retrieves the status and

statistics of an ISDN channel. The requesting channel must
own the channel whose status is being requested or the ioctl
will fail. fd is any file descriptor. arg is a pointer to a struct
isdn_channel_info. If the interface field is set to
ISDN_CHAN_SELF, it will be changed in the returned structure
to reflect the proper device-specific channel of the requesting
fd.
typedef struct isdn_channel_info {
isdn_chan_t channel;
enum isdn_iostate iostate;
struct isdn_io_stats {
ulong_t packets; /* packets transmitted or received */
ulong_t octets; /* octets transmitted or received */
ulong_t errors; /* errors packets transmitted or received */
} transmit, receive;
} isdn_channel_info_t;
ISDN_PARAM_SET fd is the file descriptor for a management device. arg is a
pointer to a struct isdn_param. This command allows the
setting of various ISDN physical layer parameters such as
timers. This command uses the same arguments as the
ISDN_PARAM_GET command.
ISDN_PARAM_GET fd is the file descriptor for a management device. arg is a
pointer to a struct isdn_param This command provides for
querying the value of a particular ISDN physical layer
parameter.
typedef enum {
ISDN_PARAM_NONE = 0,
ISDN_PARAM_NT_T101, /* NT Timer, 5-30 s, in milliseconds */
ISDN_PARAM_NT_T102, /* NT Timer, 25-100 ms, in milliseconds */
ISDN_PARAM_TE_T103, /* TE Timer, 5-30 s, in milliseconds */
ISDN_PARAM_TE_T104, /* TE Timer, 500-1000 ms, in milliseconds */
ISDN_PARAM_MAINT, /* Manage the TE Maintenance Channel */
ISDN_PARAM_ASMB, /* Modify Activation State Machine Behavior */
ISDN_PARAM_POWER, /* Take the interface online or offline */
ISDN_PARAM_PAUSE, /* Paused if == 1, else not paused == 0 */
} isdn_param_tag_t;
enum isdn_param_asmb {
ISDN_PARAM_TE_ASMB_CCITT88, /* 1988 bluebook */
ISDN_PARAM_TE_ASMB_CTS2, /* Conformance Test Suite 2 */
};
typedef struct isdn_param {
isdn_param_tag_t tag;
union {

Device and Network Interfaces 455

isdnio(7I)

unsigned int us; /* micro seconds */

unsigned int ms; /* Timer value in ms */
unsigned int flag; /* Boolean */
enum isdn_param_asmb asmb;
enum isdn_param_maint maint;
struct {
isdn_chan_t channel; /* Channel to Pause */
int paused; /* TRUE or FALSE */
} pause;
unsigned int reserved[2]; /* reserved, set to zero */
} value;
} isdn_param_t;
ISDN_PARAM_POWER If an implementation provides power on and off functions,
then power should be on by default. If flag is
ISDN_PARAM_POWER_OFF then a TE interface is forced into
state F0, NT interfaces are forced into state G0. If flag is
ISDN_PARAM_POWER_ON then a TE interface will immediately
transition to state F3 when the TE D-channel is opened. If
flag is one, an NT interface will transition to state G1 when
the NT D-channel is opened.

Implementations that do not provide ISDN_POWER return

failure with errno set to ENXIO.ISDN_POWER is different from
ISDN_PH_ACTIVATE_REQ since CCITT specification requires
that if a BRI-TE interface device has power, then it permits
activation.
ISDN_PARAM_NT_T101 This parameter accesses the NT timer value T1. The CCITT
recommendations specify that timer T1 has a value from 5 to
30 seconds. Other standards may differ.
ISDN_PARAM_NT_T102 This parameter accesses the NT timer value T2. The CCITT
recommendations specify that timer T2 has a value from 25
to 100 milliseconds. Other standards may differ.
ISDN_PARAM_TE_T103 This parameter accesses the TE timer value T3. The CCITT
recommendations specify that timer T3 has a value from 5 to
30 seconds. Other standards may differ.
ISDN_PARAM_TE_T104 This parameter accesses the TE timer value T4. The CTS2
specifies that timer T4 is either not used or has a value from
500 to 1000 milliseconds. Other standards may differ. CTS2
requires that timer T309 be implemented if T4 is not
available.
ISDN_PARAM_MAINT This parameter sets the multi-framing mode of a BRI-TE
interface. For normal operation this parameter should be set

456 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

to ISDN_PARAM_MAINT_ECHO. Other uses of this parameter

are dependent on the definition and use of the BRI interface S
and Q channels.
ISDN_PARAM_ASMB There are a few differences in the BRI-TE interface activation
state machine standards. This parameter allows the selection
of the appropriate standard. At this time, only
ISDN_PARAM_TE_ASMB_CCITT88 and
ISDN_PARAM_TE_ASMB_CTS2 are available.
ISDN_PARAM_PAUSE This parameter allows a management device to pause the IO
on a B-channel. pause.channel is set to indicate which
channel is to be paused or un-paused. pause.paused is set to
zero to un-pause and one to pause. fd is associated with an
ISDN interface management device. arg is a pointer to a
struct isdn_param.
ISDN_SET_LOOPBACK fd is the file descriptor for an ISDN interface's management
device. arg is a pointer to an isdn_loopback_request_t
structure.
typedef enum {
ISDN_LOOPBACK_LOCAL,
ISDN_LOOPBACK_REMOTE,
} isdn_loopback_type_t;
typedef enum {
ISDN_LOOPBACK_B1 = 0x1,
ISDN_LOOPBACK_B2 = 0x2,
ISDN_LOOPBACK_D = 0x4,
ISDN_LOOPBACK_E_ZERO = 0x8,
ISDN_LOOPBACK_S = 0x10,
ISDN_LOOPBACK_Q = 0x20,
} isdn_loopback_chan_t;
typedef struct isdn_loopback_request {
isdn_loopback_type_t type;
int channels;
} isdn_loopback_request_t;
An application can receive D-channel data during
D-Channel loopback but cannot transmit data. The field type
is the bitwise OR of at least one of the following values:
ISDN_LOOPBACK_B1 (0x1) /* loopback on B1-channel */
ISDN_LOOPBACK_B2 (0x2) /* loopback on B2-channel */
ISDN_LOOPBACK_D (0x4) /* loopback on D-channel */
ISDN_LOOPBACK_E_ZERO (0x8) /* force E-channel to Zero if */
/* fd is for NT interface */
ISDN_LOOPBACK_S (0x10) /* loopback on S-channel */

Device and Network Interfaces 457

isdnio(7I)

ISDN_LOOPBACK_Q (0x20) /* loopback on Q-channel */

ISDN_RESET_LOOPBACK arg is a pointer to an isdn_loopback_request_t structure.
ISDN_RESET_LOOPBACK turns off the selected loopback
modes.

ISDN Data Format The isdn_format_t type is meant to be a complete description of the various data modes and
rates available on an ISDN interface. Several macros are available for setting the format fields.
The isdn_format_t structure is shown below:
/* ISDN channel data format */
typedef enum {
ISDN_MODE_NOTSPEC, /* Not specified */
ISDN_MODE_HDLC, /* HDLC framing and error checking */
ISDN_MODE_TRANSPARENT /* Transparent mode */
} isdn_mode_t;

/* Audio encoding types (from audioio.h) */

#define AUDIO_ENCODING_NONE (0) /* no encoding*/

#define AUDIO_ENCODING_ULAW (1) /* mu-law */
#define AUDIO_ENCODING_ALAW (2) /* A-law */
#define AUDIO_ENCODING_LINEAR (3) /* Linear PCM */
typedef struct isdn_format {
isdn_mode_t mode;
unsigned int sample_rate; /* sample frames/sec*/
unsigned int channels; /* # interleaved chans */
unsigned int precision; /* bits per sample */
unsigned int encoding; /* data encoding */
} isdn_format_t;
/*
* These macros set the fields pointed
* to by the macro argument (isdn_format_t*)fp in preparation
* for the ISDN_SET_FORMAT ioctl.
*/
ISDN_SET_FORMAT_BRI_D(fp) /* BRI D-channel */
ISDN_SET_FORMAT_PRI_D(fp) /* PRI D-channel */
ISDN_SET_FORMAT_HDLC_B64(fp) /* BRI B-ch @ 56kbps */
ISDN_SET_FORMAT_HDLC_B56(fp) /* BRI B-ch @ 64kbps */
ISDN_SET_FORMAT_VOICE_ULAW(fp) /* BRI B-ch voice */
ISDN_SET_FORMAT_VOICE_ALAW(fp) /* BRI B-ch voice */
ISDN_SET_FORMAT_BRI_H(fp) /* BRI H-channel */

ISDN Datapath Types Every STREAMS stream that carries data to or from the ISDN serial interfaces is classified as a
channel-stream datapath. A possible ISDN channel-stream datapath device name for a TE
could be /dev/isdn/0/te/b1.

458 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

On some hardware implementations, it is possible to route the data from hardware channel to
hardware channel completely within the chip or controller. This is classified as a
channel-channel datapath. There does not need to be any open file descriptor for either
channel in this configuration. Only when data enters the host and utilizes a STREAMS stream
is this classified as an ISDN channel-stream datapath.

ISDN Management A management stream is a STREAMS stream that exists solely for control purposes and is not
Stream intended to carry data to or from the ISDN serial interfaces. A possible management device
name for a TE could be /dev/isdn/0/te/mgt.

Channel The following ioctls describe operations on individual channels and the connection of
Management multiple channels.
Ioctls
ISDN_SET_FORMAT fd is a data channel, the management pseudo-channel associated with
the data channel, or the management channel associated with the data
channel's interface or controller. arg is a pointer to a struct
isdn_format_req. The ISDN_SET_FORMAT ioctl sets the format of an
ISDN channel-stream datapath. It may be issued on both an open
ISDN channel-stream datapath Stream or an ISDN Management
Stream. Note that an open(2) call for a channel-stream datapath will
fail if an ISDN_SET_FORMAT has never been issued after a reset, as the
mode for all channel-stream datapaths is initially biased to
ISDN_MODE_NOTSPEC. arg is a pointer to an ISDN format type
(isdn_format_req_t*).
typedef struct isdn_format_req {
isdn_chan_t channel;
isdn_format_t format; /* data format */
int reserved[4]; /* future use - must be 0 */
} isdn_format_req_t;
If there is not an open channel-stream datapath for a requested
channel, the default format of that channel will be set for a subsequent
open(2).

To modify the format of an open stream, the driver will disconnect the
hardware channel, flush the internal hardware queues, set the new
default configuration, and finally reconnect the data path using the
newly specified format. Upon taking effect, all state information will
be reset to initial conditions, as if a channel was just opened. It is
suggested that the user flush the interface as well as consult the
hardware specific documentation to insure data integrity.

If a user desires to connect more than one B channel, such as an

H-channel, the B-channel with the smallest offset should be specified,
then the precision should be specified multiples of 8. For an

Device and Network Interfaces 459

isdnio(7I)

H-channel the precision value would be 16. The user should

subsequently open the base B-channel. If any of the sequential
B-channels are busy the open will fail, otherwise all of the B-channels
that are to be used in conjunction will be marked as busy.

The returned failure codes and their descriptions are listed below:
EPERM /* No permission for intented operation */
EINVAL /* Invalid format request */
EIO /* Set format attempt failed. */
ISDN_SET_CHANNEL The ISDN_SET_CHANNEL ioctl sets up a data connection within an
ISDN controller. The ISDN_SET_CHANNEL ioctl can only be issued
from an ISDN management stream to establish or modify
channel-channel datapaths. The ioctl parameter arg is a pointer to an
ISDN connection request (isdn_conn_req_t*). Once a data path is
established, data flow is started as soon as the path endpoints become
active. Upon taking effect, all state information is reset to initial
conditions, as if a channel was just opened.

The isdn_conn_req_t structure is shown below. The five fields

include the receive and transmit ISDN channels, the number of
directions of the data path, as well as the data format. The reserved
field must always be set to zero.
/* Number of directions for data flow */
typedef enum {
ISDN_PATH_NOCHANGE = 0, /* Invalid value */
ISDN_PATH_DISCONNECT, /* Disconnect data path */
ISDN_PATH_ONEWAY, /* One way data path */
ISDN_PATH_TWOWAY, /* Bi-directional data path */
} isdn_path_t;
typedef struct isdn_conn_req {
isdn_chan_t from;
isdn_chan_t to;
isdn_path_t dir; /* uni/bi-directional or disconnect */
isdn_format_t format; /* data format */
int reserved[4]; /* future use - must be 0 */
} isdn_conn_req_t;
To specify a read-only, write-only, or read-write path, or to
disconnect a path, the dir field should be set to ISDN_PATH_ONEWAY,
ISDN_PATH_TWOWAY, and ISDN_PATH_DISCONNECT respectively. To
modify the format of a channel-channel datapath, a user must
disconnect the channel and then reconnect with the desired format.

The returned failure codes and their descriptions are listed below:

460 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 isdnio(7I)

EPERM /* No permission for intented operation */

EBUSY /* Connection in use */
EINVAL /* Invalid connection request */
EIO /* Connection attempt failed */
ISDN_GET_FORMAT The ISDN_GET_FORMAT ioctl gets the ISDN data format of the
channel-stream datapath described by fd. arg is a pointer to an ISDN
data format request type (isdn_format_req_t*). ISDN_GET_FORMAT
can be issued on any channel to retrieve the format of any channel it
owns. For example, if issued on the TE management channel, the
format of any other te channel can be retrieved.
ISDN_GETCONFIG The ISDN_GETCONFIG ioctl is used to get the current connection status
of all ISDN channels associated with a particular management stream.
ISDN_GETCONFIG also retrieves a hardware identifier and the generic
interface type. arg is an ISDN connection table pointer
(isdn_conn_tab_t*). The isdn_conn_tab_t structure is shown
below:
typedef struct isdn_conn_tab {
char name[ISDN_ID_SIZE]; /* identification string */
isdn_interface_t type;
int maxpaths; /* size in entries of app’s array int npaths; */
/* number of valid entries returned by driver */
isdn_conn_req_t *paths; /* connection table in app’s memory */
} isdn_conn_tab_t;
The table contains a string which is the interface's unique
identification string. The second element of this table contains the
ISDN transmit and receive connections and configuration for all
possible data paths for each type of ISDN controller hardware. Entries
that are not connected will have a value of ISDN_NO_CHAN in the from
and to fields. The number of entries will always be ISDN_MAX_CHANS,
and can be referenced in the hardware specific implementation
documentation. An isdn_conn_tab_t structure is allocated on a per
controller basis.

See Also getmsg(2), ioctl(2), open(2), poll(2), read(2), write(2), audio(7I), streamio(7I)

ISDN, An Introduction – William Stallings, Macmillan Publishing Company. ISBN

0-02-415471-7

Device and Network Interfaces 461

iser(7D)

Name iser – iSCSI Extensions for Remote DMA driver

Description The iSER driver accelerates the iSCSI protocol by mapping the data transfer phases to Remote
DMA (RDMA) operations. No iSER configuration is required for its use, but an
RDMA-capable protocol (RCaP) must be configured and enabled on both target and initiator
endpoints.

Currently, InfiniBand RC is the supported RCaP, and for discovery IP over IB must be
configured on both the initiator and target. If Infiniband (IB) hardware is present and an
Infiniband reliable-connected (RC) connection can be established then an iSER-enabled
initiator uses iSER connections to iSER-enabled targets. Otherwise the connection is
established using IP-based connectivity.
Files /kernel/drv/iser 32–bit ELF kernel driver
/kernel/drv/sparcv9/iser 64-bit SPARC ELF kernel drive
/kernel/drv/amd64/iser 64-bit AMD64 ELF kernel driver
/kernel/drv/iser.conf Driver configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWiscsidmr

See Also iscsiadm(1M), itadm(1M), attributes(5), ibd(7D)

System Administration Guide: Devices and File Systems

RFC 3720 Internet Small Computer Systems Interface (iSCSI)

RFC 5046 iSCSI Extensions for RDM

462 man pages section 7: Device and Network Interfaces • Last Revised 30 Mar 2009
 isp(7D)

Name isp – ISP SCSI Host Bus Adapter Driver

Synopsis
Sbus QLGC,isp@sbus-slot,10000
PCI SUNW,isptwo@pci-slot

Description The ISP Host Bus Adapter is a SCSA compliant nexus driver that supports the Qlogic ISP1000
SCSI and the ISP1040B SCSI chips. The ISP1000 chip works on SBus and the ISP1040B chip
works on PCI bus. The ISP is an intelligent SCSI Host Bus Adapter chip that reduces the
amount of CPU overhead used in a SCSI transfer.
The isp driver supports the standard functions provided by the SCSA interface. The driver
supports tagged and untagged queuing, fast and wide SCSI, and auto request sense, but does
not support linked commands. The PCI version ISP Host bus adapter based on ISP1040B also
supports Fast-20 scsi devices.
Configuration The isp driver can be configured by defining properties in isp.conf which override the global
SCSI settings. Supported properties are scsi-options, target<n>-scsi-options,
scsi-reset-delay, scsi-watchdog-tick, scsi-tag-age-limit, scsi-initiator-id, and
scsi-selection-timeout.
target<n>-scsi-options overrides the scsi-options property value for target<n>. <n> is
a hex value that can vary from 0 to f. Refer to scsi_hba_attach(9F) for details.
Both the ISP1000 and ISP1040B support only certain SCSI selection timeout values. The valid
values are 25, 50, 75, 100, 250, 500, 750 and 1000. These properties are in units of milliseconds.

Examples EXAMPLE 1 SCSI Options

Create a file called /kernel/drv/isp.conf and add this line:
scsi-options=0x78;

This will disable tagged queuing, fast SCSI, and Wide mode for all isp instances. The
following will disable an option for one specific ISP (refer to driver.conf(4)):
name="isp" parent="/iommu@f,e0000000/sbus@f,e0001000"
reg=1,0x10000,0x450
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

Note that the default initiator ID in OBP is 7 and that the change to ID 6 will occur at attach
time. It may be preferable to change the initiator ID in OBP.
The above would set scsi-options for target 1 to 0x58 and for all other targets on this SCSI
bus to 0x178.
The physical pathname of the parent can be determined using the /devices tree or following
the link of the logical device name:

Device and Network Interfaces 463

isp(7D)

EXAMPLE 1 SCSI Options (Continued)

example# ls -l /dev/rdsk/c2t0d0s0
lrwxrwxrwx 1 root root 76 Aug 22 13:29 /dev/rdsk/c2t0d0s0 ->
../../devices/iommu@f,e0000000/sbus@f,e0001000/QLGC,isp@1,10000/sd@0,0:a,raw

Determine the register property values using the output of prtconf(1M) with the -v option:
QLGC,isp, instance #0
...
Register Specifications:
Bus Type=0x1, Address=0x10000, Size=450

EXAMPLE 2 ISP Properties

The isp driver exports properties indicating per target the negotiated transfer speed
(target<n>-sync-speed), whether tagged queuing has been enabled (target<n>-TQ), and
whether the wide data transfer has been negotiated (target<n>-wide). The sync-speed
property value is the data transfer rate in KB/sec. The target-TQ and target-wide properties
have no value. The existence of these properties indicate that tagged queuing or wide transfer
has been enabled. Refer to prtconf(1M) (verbose option) for viewing the isp properties.
QLGC,isp, instance #2
Driver software properties:
name <target0-TQ> length <0> -- <no value>.
name <target0-wide> length <0> -- <no value>.
name <target0-sync-speed> length <4>
value <0x000028f5>.
name <scsi-options> length <4>
value <0x000003f8>.
name <scsi-watchdog-tick> length <4>
value <0x0000000a>.
name <scsi-tag-age-limit> length <4>
value <0x00000008>.
name <scsi-reset-delay> length <4>
value <0x00000bb8>.

EXAMPLE 3 PCI Bus

To achieve the same setting of SCSI-options as in instance #0 above on a PCI machine,

create a file called /kernel/drv/isp.conf and add the following entries.
name="isp" parent="/pci@1f,2000/pci@1"
unit-address="4"
scsi-options=0x178
target3-scsi-options=0x58 scsi-initiator-id=6;

464 man pages section 7: Device and Network Interfaces • Last Revised 12 Jan 1998
 isp(7D)

EXAMPLE 3 PCI Bus (Continued)

The physical pathname of the parent can be determined using the /devices tree or following
the link of the logical device name:

To set scsi-options more specifically per device type, add the following line in the
/kernel/drv/isp.conf file:
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;

All device which are of this specific disk type will have scsi-options set to 0x58.

scsi-options specified per target ID has the highest precedence, followed by scsi-options
per device type. Global (for all isp instances) scsi-options per bus has the lowest
precedence.

The system needs to be rebooted before the specified scsi-options take effect.

EXAMPLE 4 Driver Capabilities

The target driver needs to set capabilities in the isp driver in order to enable some driver
features. The target driver can query and modify these capabilities: synchronous,
tagged-qing, wide-xfer, auto-rqsense, qfull-retries, qfull-retry-interval. All other
capabilities can only be queried.

By default, tagged-qing, auto-rqsense, and wide-xfer capabilities are disabled, while

disconnect, synchronous, and untagged-qing are enabled. These capabilities can only have
binary values (0 or 1). The default values for qfull-retries and qfull-retry-interval are
both 10. The qfull-retries capability is a uchar_t (0 to 255) while qfull-retry-interval
is a ushort_t (0 to 65535).

The target driver needs to enable tagged-qing and wide-xfer explicitly. The untagged-qing
capability is always enabled and its value cannot be modified, because isp can queue
commands even when tagged-qing is disabled.

Whenever there is a conflict between the value of scsi-options and a capability, the value set
in scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call.

Refer to scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.

Files /kernel/drv/isp ELF Kernel Module
/kernel/drv/isp.conf Configuration file

Device and Network Interfaces 465

isp(7D)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also prtconf(1M), driver.conf(4), attributes(5), scsi_abort(9F), scsi_hba_attach(9F),

scsi_ifgetcap(9F), scsi_reset(9F), scsi_transport(9F), scsi_device(9S),
scsi_extended_sense(9S), scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

ANSI Small Computer System Interface-2 (SCSI-2)

QLogic Corporation, ISP1000 Firmware Interface Specification

QLogic Corporation, ISP1020 Firmware Interface Specification

QLogic Corporation, ISP1000 Technical Manual

QLogic Corporation, ISP1020a/1040a Technical Manual

QLogic Corporation, Differences between the ISP1020a/1040a and the ISP1020B/1040B -

Application Note

Diagnostics The messages described below may appear on the system console as well as being logged.

The first set of messages may be displayed while the isp driver is first trying to attach. All of
these messages mean that the isp driver was unable to attach. These messages are preceded by
"isp<number>", where "<number>" is the instance number of the ISP Host Bus Adapter.
Device in slave-only slot, unused The SBus device has been placed in a
slave-only slot and will not be accessible;
move to non-slave-only SBus slot.
Device is using a hilevel intr, unused The device was configured with an interrupt
level that cannot be used with this isp
driver. Check the device.
Failed to alloc soft state Driver was unable to allocate space for the
internal state structure. Driver did not attach
to device; SCSI devices will be inaccessible.
Bad soft state Driver requested an invalid internal state
structure. Driver did not attach to device;
SCSI devices will be inaccessible.

466 man pages section 7: Device and Network Interfaces • Last Revised 12 Jan 1998
 isp(7D)

Unable to map registers Driver was unable to map device registers;

check for bad hardware. Driver did not
attach to device; SCSI devices will be
inaccessible.
Cannot add intr Driver was not able to add the interrupt
routine to the kernel. Driver did not attach
to device; SCSI devices will be inaccessible.
Unable to attach Driver was unable to attach to the hardware
for some reason that may be printed. Driver
did not attach to device; SCSI devices will be
inaccessible.

The next set of messages can be displayed at any time. They will be printed with the full device
pathname followed by the shorter form described above.
Firmware should be < 0x<number> bytes Firmware size exceeded allocated space
and will not download firmware. This
could mean that the firmware was
corrupted somehow. Check the isp
driver.
Firmware checksum incorrect Firmware has an invalid checksum and
will not be downloaded.
Chip reset timeout ISP chip failed to reset in the time
allocated; may be bad hardware.
Stop firmware failed Stopping the firmware failed; may be
bad hardware.
Load ram failed Unable to download new firmware into
the ISP chip.
DMA setup failed The DMA setup failed in the host
adapter driver on a scsi_pkt. This will
return TRAN_BADPKT to a SCSA target
driver.
Bad request pkt The ISP Firmware rejected the packet as
being set up incorrectly. This will cause
the isp driver to call the target
completion routine with the reason of
CMD_TRAN_ERR set in the scsi_pkt.
Check the target driver for correctly
setting up the packet.

Device and Network Interfaces 467

isp(7D)

Bad request pkt header The ISP Firmware rejected the packet as
being set up incorrectly. This will cause
the isp driver to call the target
completion routine with the reason of
CMD_TRAN_ERR set in the scsi_pkt.
Check the target driver for correctly
setting up the packet.
Polled command timeout on <number>.<number> A polled command experienced a
timeout. The target device, as noted by
the target lun (<number>.<number>)
information, may not be responding
correctly to the command, or the ISP
chip may be hung. This will cause an
error recovery to be initiated in the isp
driver. This could mean a bad device or
cabling.
SCSI Cable/Connection problem
Hardware/Firmware error The ISP chip encountered a firmware
error of some kind. The problem is
probably due to a faulty scsi cable or
improper cable connection. This error
will cause the isp driver to do error
recovery by resetting the chip.
Received unexpected SCSI Reset The ISP chip received an unexpected
SCSI Reset and has initiated its own
internal error recovery, which will
return all the scsi_pkt with reason set
to CMD_RESET.
Fatal timeout on target <number>.<number> The isp driver found a command that
had not completed in the correct
amount of time; this will cause error
recovery by the isp driver. The device
that experienced the timeout was at
target lun (<number>.<number>).
Fatal error, resetting interface This is an indication that the isp driver
is doing error recovery. This will cause
all outstanding commands that have
been transported to the isp driver to be
completed via the scsi_pkt completion
routine in the target driver with reason

468 man pages section 7: Device and Network Interfaces • Last Revised 12 Jan 1998
 isp(7D)

of CMD_RESET and status of

STAT_BUS_RESET set in the scsi_pkt.

Device and Network Interfaces 469

iwh(7D)

Name iwh – Intel(R) WiFi Link 5100/5300 Driver

Description The iwh 802.11a/g/n wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Intel Shirley Peak WiFi chipset-based NIC's. Driver
functions include controller initialization, wireless 802.11 infrastructure network connection,
WEP and frame transmit and receive.

Configuration The iwh driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec.
Files /dev/iwh Special character device.
/kernel/drv/iwh 32-bit ELF kernel module (x86).
/kernel/drv/amd64/iwh 64-bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWiwh

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification– IEEE, 2001.

470 man pages section 7: Device and Network Interfaces • Last Revised 26 Aug 2008
 iwi(7D)

Name iwi – Intel Pro. Wireless 802.11a/b/g IPW2200B/G IPW2915A/B/G Driver

Description The iwi 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on Intel Pro
Wireless 2200BG 2915ABG chipset-based wireless NIC's. Driver functions include controller
initialization, wireless 802.11b infrastructure network connection, WEP and frame transmit
and receive.

Driver The iwi driver performs auto-negotiation to determine the data rate and mode. Supported
Configuration 802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec.

The iwi driver supports only BSS networks (also known as "ap" or "infrastructure" networks)
and "open"(or "open-system") or "shared system" authentication. Only WEP encryption is
currently supported. You perform configuration and administration tasks using the
dladm(1M) and wificonfig(1M) utilities.
Files /dev/iwi* Special character device.
/kernel/drv/iwi 32-bit ELF kernel module (x86).
/kernel/drv/amd64/iwi 64-bit ELF kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability x86

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

ANSI/IEEE Std 802.11- Standard for Wireless LAN Technology — 1999

IEEE Std 802.11a- Standard for Wireless LAN Technology-Rev. A— 2003

IEEE Std 802.11b - Standard for Wireless LAN Technology-Rev. B — 2003

IEEE Std 802.11g- Standard for Wireless LAN Technology -Rev. G— 2003

Device and Network Interfaces 471

iwk(7D)

Name iwk – Intel Pro. Wireless 802.11a/g/n 4965 driver

Description The iwk 802.11a/g/n wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Intel Pro Wireless 4965AGN chipset-based wireless NIC's.

Driver The iwk driver supports only 802.11g BSS networks (also known as "ap" or "infrastructure"
Configuration networks) and "open" (or "open-system") or "shared system" authentication. For wireless
security, WEP encryption and WPA-PSK are currently supported. You perform configuration
and administration tasks using the dladm(1M) and wificonfig(1M) utilities.
Files /dev/iwk* Special character device.
/kernel/drv/iwk 32-bit ELF kernel module (x86).
/kernel/drv/amd64/iwk 64-bit ELF kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWiwk

Interface stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

IEEE 802.11g- Wireless LAN Standard— IEEE, 2003

472 man pages section 7: Device and Network Interfaces • Last Revised 14 Nov 2007
 iwp(7D)

Name iwp – Intel WiFi Link 6000 Series Device Driver

Description The iwp 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver which supports the Intel WiFi Link 6000 series chipset-based NICs.

Configuration The iwp driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec.

The iwp driver only supports BSS networks, (also known as ap or infrastructure networks),
open(or open-system) , or shared system authentication.
Files /kernel/drv/iwp 32-bit ELF kernel module (x86)
/kernel/drv/amd64/iwp 64-bit ELF kernel module (x86)
/dev/iwp* Special character device

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWiwp

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5)gld(7D), dlpi(7P)

IEEE 802.11 - Wireless LAN Media Access Control and Physical Layer Specification, 2001

Device and Network Interfaces 473

ixgb(7d)

Name ixgb – SUNWixgb, 10 Gigabit Ethernet driver for Intel 82597ex controllers and Sun Ethernet
PCI-X Adapter (X5544A-4) adapters.
Synopsis /dev/ixgb
Description The ixgb 10 Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, DLPI on Intel 82597ex 10
Gigabit Ethernet controllers and Sun 10 Gigabit Ethernet PCI-X Adapter (X5544A-4) on x86
Platforms. The Intel 10G controller incorporates both MAC and PHY functions and provides
10G (fiber) Ethernet operation on the SR and LR connectors. The Sun 10 Gigabit Ethernet
PCI-X Adapter (X5544A-4) is a 133 MHz PCI-X 10 Gigabit Ethernet card utilizing the Intel
82597EX PCI-X MAC controller with XFP-based 10GigE optics.
The ixgb driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support and error recovery and reporting.
The ixgb driver and hardware support auto-negotiation, a protocol specified by the IEEE
802.3ae specification.
Configuration The following ixgb.conf configuration option is supported:
default_mtu
Upper limit on the maximum MTU size the driver allows. Intel 82597EX controller allows
the configuration of jumbo frames. To configure jumbo frame, use ifconfig(1M). Use
ifconfig with the adapter instance and the mtu argument (for example: ifconfig ixgb0 mtu
9000) to configure the adapter for the maximum allowable jumbo frame size. Allowed
range is 1500 - 9000.
Application The cloning character-special device /dev/ixgb is used to access all Intel 10G controllers and
Programming Sun 10 Gigabit Ethernet PCI-X adapters (X5544A-4) installed within the system
Interface
The ixgb driver is managed by the dladm(1M) command line utility, which allows VLANs to
be defined on top of ixgb instances and for ixgb instances to be aggregated. See dladm(1M)
for more details.
You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance number
for the system. The device is initialized on first attach and de-initialized (stopped) at last
detach.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are:
■ Maximum SDU is 9000 (ETHERMTU, as defined in <sys/ethernet.h>).
■ Minimum SDU is 0.
■ DLSAP address length is 8.

474 man pages section 7: Device and Network Interfaces • Last Revised 16 Jul 2008
 ixgb(7d)

■ MAC type is DL_ETHER.

■ SAP length value is −2 meaning the physical address component is followed immediately
by a 2 byte SAP component within the DLSAP address.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular Service
Access Point (SAP) with the stream.

By default, the ixgb driver performs auto-negotiation to select the 10G link speed.
Files /dev/ixgb Special character device.
/kernel/drv/sparcv9/ixgb Driver binary.
/kernel/drv/ixgb 32-bit kernel module. (x86 only).
/kernel/drv/amd64/ixgb 64-bit kernel module (x86 only).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also dladm(1M), ifconfig(1M), attributes(5), gld(7D), streamio(7I), dlpi(7P)

IEEE 802.3ae 10 Gigabit Ethernet Specification — June, 2002

Sun 10 Gigabit Ethernet PCI-X Adapter (X5544A-4) Driver Installation Notes for Solaris

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

Device and Network Interfaces 475

ixgbe(7D)

Name ixgbe – Intel 10Gb PCI Express NIC Driver

Synopsis /dev/ixgbe*

Description The ixgbe 10 Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on Intel 10–Gigabit
PCI Express Ethernet controllers.

The ixgbe driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting.

The ixgbe driver and hardware support auto-negotiation, a protocol specified by the IEEE
802.3ae specification.

Application The cloning character-special device, /dev/ixgbe, is used to access all Intel 10 –Gigabit PCI
Programming Express Ethernet devices installed within the system.
Interface
The ixgbe driver is managed by the dladm(1M) command line utility, which allows VLANs to
be defined on top of ixgbe instances and for ixgbe instances to be aggregated. See dladm(1M)
for more details.

You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number for the system. The device is initialized on first attach and de-initialized (stopped) at
last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to your
DL_INFO_REQ are:
■ Maximum SDU is 16366.
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP (Service Access Point) length value is -2, meaning the physical address component is
followed immediately by a 2-byte SAP component within the DLSAP address.
■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a
particular SAP with the stream.

Configuration By default, the ixgbe driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can only be 10000 Mbps full-duplex. See the IEEE802.3 standard for more
information.

476 man pages section 7: Device and Network Interfaces • Last Revised 3 Jun 2009
 ixgbe(7D)

Files /dev/ixgbe* Special character device.

/kernel/drv/ixgbe 32–bit device driver (x86).
/kernel/drv/amd64/ixgbe 64–bit device driver (x86).
/kernel/drv/sparcv9/ixgbe 64–bit device driver (SPARC).
/kernel/drv/ixgbe.conf Configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWixgbe

Architecture SPARC, x86

Interface Stability Committed

See Also dladm(1M), netstat(1M), driver.conf(4), attributes(5), streamio(7I), dlpi(7P)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

IEEE 802.3ae Specificiation, IEEE – 2002

Device and Network Interfaces 477

jfb(7D)

Name jfb – XVR-1200 Graphics Accelerator device driver

Description The jfb module is the device driver for the Sun XVR-600 and Sun XVR-1200 Graphics
Accelerators. The XVR-1200 Graphics Accelerator is a high resolution, high performance PCI
graphics framebuffer providing hardware 2D and 3D acceleration. Sun XVR-600 is the single
pipeline version of the Sun XVR-1200.

The jfbdaemon process loads the jfb microcode at system startup time and during the resume
sequence of a suspend-resume cycle.
Files /dev/fbs/jfb0 Device special file for XVR-600 and XVR-1200 high performance
single screen.
/dev/fbs/jfb0a Device special file for the XVR-1200 first video out.
/dev/fbs/jfb0b Device special file for the XVR-1200 second video out.
/usr/lib/jfb.ucode jfb microcode.
/usr/sbin/jfbdaemon jfb microcode loader.

See Also SUNWjfb_config(1M)

478 man pages section 7: Device and Network Interfaces • Last Revised 21 June 2004
 kb(7M)

Name kb – keyboard STREAMS module

Synopsis #include <sys/types.h>
#include <sys/stream.h>
#include <sys/stropts.h>
#include <sys/vuid_event.h>
#include <sys/kbio.h>
#include <sys/kbd.h>
ioctl(fd, I_PUSH, "kb");

Description The kb STREAMS module processes byte streams generated by a keyboard attached to a CPU
serial port. Definitions for altering keyboard translation and reading events from the keyboard
are contained in <sys/kbio.h> and <sys/kbd.h>.

The kb STREAMS module utilizes a set of keyboard tables to recognize which keys have been
typed. Each translation table is an array of 128 16-bit words (unsigned shorts). If a table
entry is less than 0x100, the entry is treated as an ISO 8859/1 character. Higher values indicate
special characters that invoke more complicated actions.

Keyboard Translation The keyboard can be in one of the following translation modes:
Mode
TR_NONE Keyboard translation is turned off and up/down key codes are
reported.
TR_ASCII ISO 8859/1 codes are reported.
TR_EVENT firm_events are reported.
TR_UNTRANS_EVENT firm_events containing unencoded keystation codes are reported for
all input events within the window system.

Keyboard All instances of the kb module share seven translation tables that convert raw keystation codes
Translation-Table to event values. The tables are:
Entries
Unshifted Used when a key is depressed and no shifts are in effect.
Shifted Used when a key is depressed and a Shift key is held down.
Caps Lock Used when a key is depressed and Caps Lock is in effect.
Alt Graph Used when a key is depressed and the Alt Graph key is held down.
Num Lock Used when a key is depressed and Num Lock is in effect.
Controlled Used when a key is depressed and the Control key is held down. (Regardless of
whether a Shift key or the Alt Graph is being held down, or whether Caps Lock
or Num Lock is in effect).

Device and Network Interfaces 479

kb(7M)

Key Up Used when a key is released.

Each key on the keyboard has a key station code that represents a number from 0 to 127.
The number is used as an index into the translation table that is currently in effect. If the
corresponding entry in the translation table is a value from 0 to 255, the value is treated as an
ISO 8859/1 character, and the character is the result of the translation.

If the entry in the translation table is higher than 255, it is a special entry. Special entry values
are classified according to the value of the high-order bits. The high-order value for each class
is defined as a constant, as shown below. When added to the constant, the value of the
low-order bits distinguish between keys within each class:
SHIFTKEYS 0x100 A shift key. The value of the particular shift key is added to determine
which shift mask to apply:
CAPSLOCK 0 Caps Lock key.
SHIFTLOCK 1 “Shift Lock” key.
LEFTSHIFT 2 Left-hand Shift key.
RIGHTSHIFT 3 Right-hand Shift key.
LEFTCTRL 4 Left-hand (or only) Control key.
RIGHTCTRL 5 Right-hand Control key.
ALTGRAPH 9 Alt Graph key.
ALT 10 Alternate or Alt key.
NUMLOCK 11 Num Lock key.
BUCKYBITS 0x200 Used to toggle mode-key-up/down status without altering the value of
an accompanying ISO 8859/1 character. The actual bit-position value,
minus 7, is added.
METABIT 0 The Meta key was pressed along with the key. This
is the only user-accessible bucky bit. It is ORed in
as the 0x80 bit; since this bit is a legitimate bit in a
character, the only way to distinguish between, for
example, 0xA0 as META+0x20 and 0xA0 as an
8-bit character is to watch for META key up and
META key down events and keep track of whether
the META key was down.
SYSTEMBIT 1 The System key was pressed. This is a place holder
to indicate which key is the system-abort key.
FUNNY 0x300 Performs various functions depending on the value of the low 4 bits:

480 man pages section 7: Device and Network Interfaces • Last Revised 26 Feb 2004
 kb(7M)

NOP 0x300 Does nothing.

OOPS 0x301 Exists, but is undefined.
HOLE 0x302 There is no key in this position on the
keyboard, and the position-code should not be
used.
RESET 0x306 Keyboard reset.
ERROR 0x307 The keyboard driver detected an internal error.
IDLE 0x308 The keyboard is idle (no keys down).
COMPOSE 0x309 The COMPOSE key; the next two keys should
comprise a two-character COMPOSE key
sequence.
NONL 0x30A Used only in the Num Lock table; indicates that
this key is not affected by the Num Lock state,
so that the translation table to use to translate
this key should be the one that would have been
used had Num Lock not been in effect.
0x30B — 0x30F Reserved for non-parameterized functions.
FA_CLASS 0x400 A floating accent or “dead key.” When this key is pressed, the next key
generates an event for an accented character; for example, “floating
accent grave” followed by the “a” key generates an event with the ISO
8859/1 code for the “a with grave accent” character. The low-order
bits indicate which accent; the codes for the individual “floating
accents” are as follows:
FA_UMLAUT 0x400 umlaut
FA_CFLEX 0x401 circumflex
FA_TILDE 0x402 tilde
FA_CEDILLA 0x403 cedilla
FA_ACUTE 0x404 acute accent
FA_GRAVE 0x405 grave accent
STRING 0x500 The low-order bits index a table of strings. When a key with a STRING
entry is depressed, the characters in the null-terminated string for that
key are sent, character-by-character. The maximum length is defined
as:
KTAB_STRLEN 10

Device and Network Interfaces 481

kb(7M)

Individual string numbers are defined as:

HOMEARROW 0x00
UPARROW 0x01
DOWNARROW 0x02
LEFTARROW 0x03
RIGHTARROW 0x04

String numbers 0x05 — 0x0F are available for custom entries.

FUNCKEYS 0x600 There are 64 keys reserved for function keys. The actual positions are
usually on the left/right/top/bottom of the keyboard.

The next-to-lowest 4 bits indicate the group of function keys:

LEFTFUNC 0x600
RIGHTFUNC 0x610
TOPFUNC 0x610 0x610
BOTTOMFUNC 0x630

The low 4 bits indicate the function key number within the group:
LF(n) (LEFTFUNC+(n)-1)
RF(n) (RIGHTFUNC+(n)-1)
TF(n) (TOPFUNC+(n)-1)
BF(n) (BOTTOMFUNC+(n)-1)
PADKEYS 0x700 A “numeric keypad key.” These entries should appear only in the Num
Lock translation table; when Num Lock is in effect, these events will be
generated by pressing keys on the right-hand keypad. The low-order
bits indicate which key. The codes for the individual keys are:
PADEQUAL 0x700 “=” key
PADSLASH 0x701 “/” key
PADSTAR 0x702 “*” key
PADMINUS 0x703 “-” key
PADSEP 0x704 “,” key
PAD7 0x705 “7” key
PAD8 0x706 “8” key

482 man pages section 7: Device and Network Interfaces • Last Revised 26 Feb 2004
 kb(7M)

PAD9 0x707 “9” key

PADPLUS 0x708 “+” key
PAD4 0x709 “4” key
PAD5 0x70A “5” key
PAD6 0x70B “6” key
PAD1 0x70C “1” key
PAD2 0x70D “2” key
PAD3 0x70E “3” key
PAD0 0x70F “0” key
PADDOT 0x710 “.” key
PADENTER 0x711 “Enter” key

When a function key is pressed in TR_ASCII mode, the following escape sequence is sent:

ESC[0 . . .. 9z

where ESC is a single escape character and “0 . .. 9” indicates the decimal representation of the
function-key value. For example, function key R1 sends the sequence:

ESC[208z

because the decimal value of RF(1) is 208. In TR_EVENT mode, if there is a VUID event code for
the function key in question, an event with that event code is generated; otherwise, individual
events for the characters of the escape sequence are generated.

Keyboard When started, the kb STREAMS module is in the compatibility mode. When the keyboard is
Compatibility Mode in the TR_EVENT translation mode, ISO 8859/1 characters from the upper half of the character
set (that is, characters with the eighth bit set) , are presented as events with codes in the
ISO_FIRST range (as defined in <<sys/vuid_event.h>>). For backwards compatibility with
older versions of the keyboard driver, the event code is ISO_FIRST plus the character value.
When compatibility mode is turned off, ISO 8859/1 characters are presented as events with
codes equal to the character code.

Description The following ioctl() requests set and retrieve the current translation mode of a keyboard:
KIOCTRANS Pointer to an int. The translation mode is set to the value in the int pointed
to by the argument.
KIOCGTRANS Pointer to an int. The current translation mode is stored in the int pointed
to by the argument.

Device and Network Interfaces 483

kb(7M)

ioctl() requests for changing and retrieving entries from the keyboard translation table use
the kiockeymap structure:
struct kiockeymap {
int kio_tablemask; /* Translation table (one of: 0, CAPSMASK,
* SHIFTMASK, CTRLMASK, UPMASK,
* ALTGRAPHMASK, NUMLOCKMASK)
*/
#define KIOCABORT1 –1 /* Special “mask”: abort1 keystation */
#define KIOCABORT2 –2 /* Special “mask”: abort2 keystation */
uchar_t kio_station; /* Physical keyboard key station (0-127) */
ushort_t kio_entry; /* Translation table station’s entry */
char kio_string[10]; /* Value for STRING entries-null terminated */
};
KIOCSKEY Pointer to a kiockeymap structure. The translation table entry referred to by the
values in that structure is changed. The kio_tablemask request specifies which
of the following translation tables contains the entry to be modified:
UPMASK 0x0080 “Key Up” translation table.
NUMLOCKMASK 0x0800 “Num Lock” translation table.
CTRLMASK 0x0030 “Controlled” translation table.
ALTGRAPHMASK 0x0200 “Alt Graph” translation table.
SHIFTMASK 0x000E “Shifted” translation table.
CAPSMASK 0x0001 “Caps Lock” translation table.
(No shift keys pressed or locked) “Unshifted” translation table.

The kio_station request specifies the keystation code for the entry to be modified. The value
of kio_entry is stored in the entry in question. If kio_entry is between STRING and
STRING+15, the string contained in kio_string is copied to the appropriate string table
entry. This call may return EINVAL if there are invalid arguments.

Special values of kio_tablemask can affect the two step “break to the PROM monitor”
sequence. The usual sequence is L1-a or Stop-. If kio_tablemask is KIOCABORT1, then the
value of kio_station is set to be the first keystation in the sequence. If kio_tablemask, is
KIOCABORT2 then the value of kio_station is set to be the second keystation in the sequence.
An attempt to change the “break to the PROM monitor” sequence without having superuser
permission results in an EPERM error.
KIOCGKEY The argument is a pointer to a kiockeymap structure. The current value of the
keyboard translation table entry specified by kio_tablemask and kio_station
is stored in the structure pointed to by the argument. This call may return
EINVAL if there are invalid arguments.

484 man pages section 7: Device and Network Interfaces • Last Revised 26 Feb 2004
 kb(7M)

KIOCTYPE The argument is a pointer to an int. A code indicating the type of the keyboard
is stored in the int pointed to by the argument:
KB_SUN3 Sun Type 3 keyboard
KB_SUN4 Sun Type 4 or 5 keyboard, or non-USB Sun Type 6 keyboard
KB_USB USB standard HID keyboard, including Sun Type 6 USB
keyboards
KB_ASCII ASCII terminal masquerading as keyboard
KB_PC Type 101 PC keyboard
KB_DEFAULT Stored in the int pointed to by the argument if the keyboard
type is unknown. In case of error, -1 is stored in the int
pointed to by the argument.
KIOCLAYOUT The argument is a pointer to an int. On a Sun Type 4 keyboard, the layout
code specified by the keyboard's DIP switches is stored in the int pointed to
by the argument.
KIOCCMD The argument is a pointer to an int. The command specified by the value of
the int pointed to by the argument is sent to the keyboard. The commands
that can be sent are:

Commands to the Sun Type 3 and Sun Type 4 keyboards:

KBD_CMD_RESET Reset keyboard as if power-up.
KBD_CMD_BELL Turn on the bell.
KBD_CMD_NOBELL Turn off the bell.
KBD_CMD_CLICK Turn on the click annunciator.
KBD_CMD_NOCLICK Turn off the click annunciator.

Commands to the Sun Type 4 keyboard:

KBD_CMD_SETLED Set keyboard LEDs.
KBD_CMD_GETLAYOUT Request that keyboard indicate layout.

Inappropriate commands for particular keyboard types are ignored. Since there is no reliable
way to get the state of the bell or click (because the keyboard cannot be queried and a process
could do writes to the appropriate serial driver — circumventing this ioctl() request) an
equivalent ioctl() to query its state is not provided.
KIOCSLED The argument is a pointer to an char. On the Sun Type 4 keyboard, the LEDs
are set to the value specified in that char. The values for the four LEDs are:
LED_CAPS_LOCK “Caps Lock” light.

Device and Network Interfaces 485

kb(7M)

LED_COMPOSE “Compose” light.

LED_SCROLL_LOCK “Scroll Lock” light.
LED_NUM_LOCK “Num Lock” light.

On some Japanese layouts, the value for the fifth LED is:
LED_KANA “Kana” light.
KIOCGLED Pointer to a char. The current state of the LEDs is stored in the char pointed
to by the argument.
KIOCSCOMPAT Pointer to an int. “Compatibility mode” is turned on if the int has a value
of 1, and is turned off if the int has a value of 0.
KIOCGCOMPAT Pointer to an int. The current state of “compatibility mode” is stored in the
int pointed to by the argument.

The following ioctl() request allows the default effect of the keyboard abort sequence to be
changed.
KIOCSKABORTEN Pointer to an int. The keyboard abort sequence effect (typically L1-A or
Stop-A on the keyboard on SPARC systems, F1–A on x86 systems, and
BREAK on the serial console device) is enabled if the int has a value of
KIOCABORTENABLE(1). If the value is KIOCABORTDISABLE(0) , the
keyboard abort sequence effect is disabled. If the value is
KIOCABORTALTERNATE(2), the Alternate Break sequence is in effect
and is defined by the serial console drivers zs(7D)se(7D) and asy(7D).
Any other value of the parameter for this ioctl() is treated as enable.
The Alternate Break sequence is applicable to the serial console devices
only.

Due to a risk of incorrect sequence interpretation, SLIP and certain other

binary protocols should not be run over the serial console port when
Alternate Break sequence is in effect. Although PPP is a binary protocol, it
is able to avoid these sequences using the ACCM feature in RFC 1662. For
Solaris PPP 4.0, you do this by adding the following line to the
/etc/ppp/options file (or other configuration files used for the
connection; see pppd(1M) for details):
asyncmap 0x00002000
SLIP has no comparable capability, and must not be used if the Alternate
Break sequence is in use.

This ioctl() will be active and retain state even if there is no physical
keyboard in the system. The default effect (enable) causes the operating
system to suspend and enter the kernel debugger (if present) or the

486 man pages section 7: Device and Network Interfaces • Last Revised 26 Feb 2004
 kb(7M)

system prom (on most systems with OpenBoot proms). The default effect
is enabled on most systems, but may be different on server systems with
key switches in the 'secure' position. On these systems, the effect is always
disabled when the key switch is in the 'secure' position. This
ioctl()returns EPERM if the caller is not the superuser.

These ioctl() requests are supported for compatibility with the system keyboard device
/dev/kbd.
KIOCSDIRECT Has no effect.
KIOCGDIRECT Always returns 1.

The following ioctl() requests are used to set and get the keyboard autorepeat delay and rate.
KIOCSRPTDELAY This argument is a pointer to an int, which is the kb autorepeat delay, unit
in millisecond.
KIOCGRPTDELAY This argument is a pointer to an int. The current auto repeat delay setting
is stored in the integer pointed by the argument, unit in millisecond.
KIOCSRPTRATE This argument is a pointer to an int, which is the kb autorepeat rate, unit
in millisecond.
KIOCGRPTRATE This argument is a pointer to an int. The current auto repeat rate setting is
stored in the integer pointed by the argument, unit in millisecond.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Stable

See Also kbd(1), kmdb(1), loadkeys(1), pppd(1M), keytables(4), attributes(5), zs(7D), se(7D),
asy(7D), virtualkm(7D), termio(7I), usbkbm(7M)

Notes Many keyboards released after Sun Type 4 keyboard also report themselves as Sun Type 4
keyboards.

Device and Network Interfaces 487

kdmouse(7D)

Name kdmouse – built-in mouse device interface

Description The kdmouse driver supports machines with built-in PS/2 mouse interfaces. It allows
applications to obtain information about the mouse's movements and the status of its buttons.

Programs are able to read directly from the device. The data returned corresponds to the byte
sequences as defined in the IBM PS/2 Technical Reference Manual.
Files /dev/kdmouse device file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), vuidmice(7M)

IBM PS/2 Technical Reference Manual.

488 man pages section 7: Device and Network Interfaces • Last Revised 24 Feb 1998
 kfb(7D)

Name kfb – Sun XVR-2500 Graphics Accelerator device driver

Description The kfb driver is the Sun XVR-2500 graphics accelerator device driver. The Sun XVR-2500
graphics accelerator is a high resolution, high performance PCI graphics framebuffer
providing hardware 2D and 3D acceleration.

The kfbdaemon process provides memory management for the XVR-2500 product family.
Files dev/fbs/kfbn Device special file for XVR-2500 high performance single screen.
/usr/sbin/kfbdaemon Memory manager.

See Also SUNWkfb_config(1M)

Device and Network Interfaces 489

kmdb(7d)

Name kmdb – Kernel debugger

Description The kmdb driver is the mechanism used by mdb to invoke and control kmdb. This is not a public
interface.
Files /dev/kmdb Kernel debugger driver.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Private

See Also attributes(5)

490 man pages section 7: Device and Network Interfaces • Last Revised 8 May 2004
 kstat(7D)

Name kstat – kernel statistics driver

Description The kstat driver is the mechanism used by the kstat(3KSTAT) library to extract kernel
statistics. This is NOT a public interface.
Files /dev/kstat kernel statistics driver

See Also kstat(3KSTAT), kstat(9S)

Device and Network Interfaces 491

ksyms(7D)

Name ksyms – kernel symbols

Synopsis /dev/ksyms

Description The file /dev/ksyms is a character special file that allows read-only access to an ELF format
image containing two sections: a symbol table and a corresponding string table. The contents
of the symbol table reflect the symbol state of the currently running kernel. You can determine
the size of the image with the fstat( ) system call. The recommended method for accessing
the /dev/ksyms file is by using the ELF access library. See elf(3ELF) for details. If you are not
familiar with ELF format, see a.out(4).

/dev/ksyms is an executable for the processor on which you are accessing it. It contains ELF
program headers which describe the text and data segment(s) in kernel memory. Since
/dev/ksyms has no text or data, the fields specific to file attributes are initialized to NULL. The
remaining fields describe the text or data segment(s) in kernel memory.
Symbol table The SYMTAB section contains the symbol table entries present in the
currently running kernel. This section is ordered as defined by the ELF
definition with locally-defined symbols first, followed by globally-defined
symbols. Within symbol type, the symbols are ordered by kernel module
load time. For example, the kernel file symbols are first, followed by the first
module's symbols, and so on, ending with the symbols from the last
module loaded.

The section header index (st_shndx) field of each symbol entry in the
symbol table is set to SHN_ABS, because any necessary symbol relocations
are performed by the kernel link editor at module load time.
String table The STRTAB section contains the symbol name strings that the symbol
table entries reference.

See Also kernel(1M), stat(2), elf(3ELF), kvm_open(3KVM), a.out(4), mem(7D)

Warnings The kernel is dynamically configured. It loads kernel modules when necessary. Because of this
aspect of the system, the symbol information present in the running system can vary from
time to time, as kernel modules are loaded and unloaded.

When you open the /dev/ksyms file, you have access to an ELF image which represents a
snapshot of the state of the kernel symbol information at that instant in time. While the
/dev/ksyms file remains open, kernel module autounloading is disabled, so that you are
protected from the possibility of acquiring stale symbol data. Note that new modules can still
be loaded, however. If kernel modules are loaded while you have the /dev/ksyms file open, the
snapshot held by you will not be updated. In order to have access to the symbol information of
the newly loaded modules, you must first close and then reopen the /dev/ksyms file. Be aware
that the size of the /dev/ksyms file will have changed. You will need to use the fstat()
function (see stat(2)) to determine the new size of the file.

492 man pages section 7: Device and Network Interfaces • Last Revised 11 Dec 2000
 ksyms(7D)

Avoid keeping the /dev/ksyms file open for extended periods of time, either by using
kvm_open(3KVM) of the default namelist file or with a direct open. There are two reasons why
you should not hold /dev/ksyms open. First, the system's ability to dynamically configure
itself is partially disabled by the locking down of loaded modules. Second, the snapshot of
symbol information held by you will not reflect the symbol information of modules loaded
after your initial open of /dev/ksyms.

Note that the ksyms driver is a loadable module, and that the kernel driver modules are only
loaded during an open system call. Thus it is possible to run stat(2) on the /dev/ksyms file
without causing the ksyms driver to be loaded. In this case, the file size returned is
UNKNOWN_SIZE. A solution for this behavior is to first open the /dev/ksyms file, causing the
ksyms driver to be loaded (if necessary). You can then use the file descriptor from this open in
a fstat( ) system call to get the file's size.

Notes The kernel virtual memory access library (libkvm) routines use /dev/ksyms as the default
namelist file. See kvm_open(3KVM) for details.

Device and Network Interfaces 493

ldterm(7M)

Name ldterm – standard STREAMS terminal line discipline module

Synopsis #include <sys/stream.h>
#include <sys/termios.h>
int ioctl(fd,I_PUSH,"ldterm");

Description The ldterm STREAMS module provides most of the termio(7I) terminal interface. The vis
module does not perform the low-level device control functions specified by flags in the
c_cflag word of the termio/termios structure, or by the IGNBRK, IGNPAR, PARMRK, or INPCK
flags in the c_iflag word of the termio/termios structure. Those functions must be
performed by the driver or by modules pushed below the ldterm module. The ldterm module
performs all other termio/termios functions, though some may require the cooperation of
the driver or modules pushed below ldterm and may not be performed in some cases. These
include the IXOFF flag in the c_iflag word and the delays specified in the c_oflag word.

The ldterm module also handles single and multi-byte characters from various codesets
including both Extended Unix Code (EUC) and non-EUC codesets.

The remainder of this section describes the processing of various STREAMS messages on the
read- and write-side.

Read-side Behavior Various types of STREAMS messages are processed as follows:

M_BREAK Depending on the state of the BRKINT flag, either an interrupt signal is generated
or the message is treated as if it were an M_DATA message containing a single
ASCII NUL character when this message is received.
M_DATA This message is normally processed using the standard termio input processing.
If the ICANON flag is set, a single input record (‘‘line'') is accumulated in an
internal buffer and sent upstream when a line-terminating character is received.
If the ICANON flag is not set, other input processing is performed and the
processed data are passed upstream.

If output is to be stopped or started as a result of the arrival of characters (usually

CNTRL-Q and CNTRL-S), M_STOP and M_START messages are sent downstream.
If the IXOFF flag is set and input is to be stopped or started as a result of
flow-control considerations, M_STOPI and M_STARTI messages are sent
downstream.

M_DATA messages are sent downstream, as necessary, to perform echoing.

If a signal is to be generated, an M_FLUSH message with a flag byte of FLUSHR is

placed on the read queue. If the signal is also to flush output, an M_FLUSH message
with a flag byte of FLUSHW is sent downstream.

494 man pages section 7: Device and Network Interfaces • Last Revised 7 Jun1999
 ldterm(7M)

All other messages are passed upstream unchanged.

Write-side Behavior Various types of STREAMS messages are processed as follows:

M_FLUSH The write queue of the module is flushed of all its data messages and the
message is passed downstream.
M_IOCTL The function of this ioctl is performed and the message is passed downstream
in most cases. The TCFLSH and TCXONC ioctls can be performed entirely in the
ldterm module, so the reply is sent upstream and the message is not passed
downstream.
M_DATA If the OPOST flag is set, or both the XCASE and ICANON flags are set, output
processing is performed and the processed message is passed downstream
along with any M_DELAY messages generated. Otherwise, the message is passed
downstream without change.
M_CTL If the size of the data buffer associated with the message is the size of struct
iocblk, ldterm will perform functional negotiation to determine where the
termio(7I) processing is to be done. If the command field of the iocblk
structure (ioc_cmd) is set to MC_NO_CANON, the input canonical processing
normally performed on M_DATA messages is disabled and those messages are
passed upstream unmodified. (This is for the use of modules or drivers that
perform their own input processing, such as a pseudo-terminal in TIOCREMOTE
mode connected to a program that performs this processing). If the command
is MC_DO_CANON, all input processing is enabled. If the command is
MC_PART_CANON, then an M_DATA message containing a termios structure is
expected to be attached to the original M_CTL message. The ldterm module will
examine the iflag, oflag, and lflag fields of the termios structure and from
that point on, will process only those flags that have not been turned ON. If
none of the above commands are found, the message is ignored. In any case, the
message is passed upstream.
M_FLUSH The read queue of the module is flushed of all its data messages and all data in
the record being accumulated are also flushed. The message is passed upstream.
M_IOCACK The data contained within the message, which is to be returned to the process,
are augmented if necessary, and the message is passed upstream.

All other messages are passed downstream unchanged.

ioctls The ldterm module processes the following TRANSPARENT ioctls. All others are passed
downstream.
TCGETS/TCGETA
The message is passed downstream. If an acknowledgment is seen, the data provided by the
driver and modules downstream are augmented and the acknowledgement is passed
upstream.

Device and Network Interfaces 495

ldterm(7M)

TCSETS/TCSETSW/TCSETSF/TCSETA/TCSETAW/TCSETAF
The parameters that control the behavior of the ldterm module are changed. If a mode
change requires options at the stream head to be changed, an M_SETOPTS message is sent
upstream. If the ICANON flag is turned on or off, the read mode at the stream head is
changed to message-nondiscard or byte-stream mode, respectively. If the TOSTOP flag is
turned on or off, the tostop mode at the stream head is turned on or off, respectively. In any
case, ldterm passes the ioctl on downstream for possible additional processing.
TCFLSH
If the argument is 0, an M_FLUSH message with a flag byte of FLUSHR is sent downstream and
placed on the read queue. If the argument is 1, the write queue is flushed of all its data
messages and an M_FLUSH message with a flag byte of FLUSHW is sent upstream and
downstream. If the argument is 2, the write queue is flushed of all its data messages and an
M_FLUSH message with a flag byte of FLUSHRW is sent downstream and placed on the read
queue.
TCXONC
If the argument is 0 and output is not already stopped, an M_STOP message is sent
downstream. If the argument is 1 and output is stopped, an M_START message is sent
downstream. If the argument is 2 and input is not already stopped, an M_STOPI message is
sent downstream. If the argument is 3 and input is stopped, an M_STARTI message is sent
downstream.
TCSBRK
The message is passed downstream, so the driver has a chance to drain the data and then
send an M_IOCACK message upstream.
EUC_WSET
This call takes a pointer to an eucioc structure, and uses it to set the EUC line discipline's
local definition for the code set widths to be used for subsequent operations. Within the
stream, the line discipline may optionally notify other modules of this setting using M_CTL
messages. When this call is received and the euciocstructure contains valid data, the line
discipline changes into EUC handling mode once the euciocdata is completely transferred
to an internal data structure.
EUC_WGET
This call takes a pointer to an eucioc structure, and returns in it the EUC code set widths
currently in use by the EUC line discipline. If the current codeset of the line discipline is not
an EUC one, the result is meaningless.

See Also termios(3C), console(7D), termio(7I)

STREAMS Programming Guide

496 man pages section 7: Device and Network Interfaces • Last Revised 7 Jun1999
 llc1(7D)

Name llc1 – Logical Link Control Protocol Class 1 Driver

Synopsis #include <sys/stropts.h>
#include <sys/ethernet.h>
#include <sys/dlpi.h>
#include <sys/llc1.h>

Description The llc1 driver is a multi-threaded, loadable, clonable, STREAMS multiplexing driver
supporting the connectionless Data Link Provider Interface, dlpi(7P), implementing IEEE
802.2 Logical Link Control Protocol Class 1 over a STREAM to a MAC level driver. Multiple
MAC level interfaces installed within the system can be supported by the driver. The llc1
driver provides basic support for the LLC1 protocol. Functions provided include frame
transmit and receive, XID, and TEST, multicast support, and error recovery and reporting.
The cloning, character-special device, /dev/llc1, is used to access all LLC1 controllers
configured under llc1.
The llc1 driver is a “Style 2” Data Link Service provider. All messages of types M_PROTO and
M_PCPROTO are interpreted as DLPI primitives. An explicit DL_ATTACH_REQ message by the user
is required to associate the opened stream with a particular device (ppa). The ppa ID is
interpreted as an unsigned long and indicates the corresponding device instance (unit)
number. An error (DL_ERROR_ACK) is returned by the driver if the ppa field value does not
correspond to a valid device instance number for this system.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ The maximum Service Data UNIT (SDU) is derived from the MAC layer linked below the
driver. In the case of an Ethernet driver, the SDU will be 1497.
■ The minimum SDU is 0.
■ The MAC type is DL_CSMACD or DL_TPR as determined by the driver linked under llc1. If
the driver reports that it is DL_ETHER, it will be changed to DL_CSMACD; otherwise the type
is the same as the MAC type.
■ The sap length value is −1, meaning the physical address component is followed
immediately by a 1-octet sap component within the DLSAP address.
■ The service mode is DL_CLDLS.
■ The MAC type is DL_CSMACD or DL_TPR as determined by the driver linked under llc1. If
the driver reports that it is DL_ETHER, it will be changed to DL_CSMACD; otherwise the type
is the same as the MAC type.
■ The dlsap address length is 7.
■ No optional quality of service (QOS) support is included at present, so the QOS fields
should be initialized to 0.
■ The DLPI version is DL_VERSION_2.

Device and Network Interfaces 497

llc1(7D)

■ The provider style is DL_STYLE2.

■ The broadcast address value is the broadcast address returned from the lower level driver.
Once in the DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream. The llc1 driver interprets the sap field within the
DL_BIND_REQ as an IEEE 802.2 “SAP,” therefore valid values for the sap field are in the
[0-0xFF] range with only even values being legal.
The llc1 driver DLSAP address format consists of the 6-octet physical (e.g., Ethernet) address
component followed immediately by the 1-octet sap (type) component producing a 7-octet
DLSAP address. Applications should not hard-code to this particular implementation-specific
DLSAP address format, but use information returned in the DL_INFO_ACK primitive to
compose and decompose DLSAP addresses. The sap length, full DLSAP length, and
sap/physical ordering are included within the DL_INFO_ACK. The physical address length can
be computed by subtracting the absolute value of the sap length from the full DLSAP address
length or by issuing the DL_PHYS_ADDR_REQ to obtain the current physical address associated
with the stream.
Once in the DL_BOUND state, the user may transmit frames on the LAN by sending
DL_UNITDATA_REQ messages to the llc1 driver. The llc1 driver will route received frames up
all open and bound streams having a sap which matches the IEEE 802.2 DSAP as
DL_UNITDATA_IND messages. Received frames are duplicated and routed up multiple open
streams if necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.
In addition to the mandatory, connectionless DLPI message set, the driver additionally
supports the following primitives:
The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable/disable reception of
specific multicast group addresses. A set of multicast addresses may be iteratively created and
modified on a per-stream basis using these primitives. These primitives are accepted by the
driver in any driver state that is valid while still being attached to the ppa.
The DL_PHYS_ADDR_REQ primitive returns the 6-octet physical address currently associated
(attached) to the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ.
The DL_SET_PHYS_ADDR_REQ primitive changes the 6-octet physical address currently
associated (attached) to this stream. Once changed, all streams subsequently opened and
attached to this device will obtain this new physical address. Once changed, the physical
address will remain set until this primitive is used to change the physical address again or the
system is rebooted, whichever occurs first.
The DL_XID_REQ/DL_TEST_REQ primitives provide the means for a user to issue an LLC XID or
TEST request message. A response to one of these messages will be in the form of a
DL_XID_CON/DL_TEST_CON message.

498 man pages section 7: Device and Network Interfaces • Last Revised 13 Feb 1997
 llc1(7D)

The DL_XID_RES/DL_TEST_RES primitives provide a way for the user to respond to the receipt
of an XID or TEST message that was received as a DL_XID_IND/DL_TEST_IND message.

XID and TEST will be automatically processed by llc1 if the DL_AUTO_XID/DL_AUTO_TEST bits
are set in the DL_BIND_REQ.
Files /dev/llc1 cloning, character-special device

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), dlpi(7P)

Device and Network Interfaces 499

llc2(7D)

Name llc2 – Class II logical link control driver

Description The llc2 logical link control driver interfaces network software (NetBIOS, SNA, OSI, and so
on) running under the Solaris operating environment to a physical LAN network controlled
by one of the supported communications adapters. The llc2 driver, which appears as a
STREAMS driver to the network software, resides in the kernel and is accessed by standard
UNIX STREAMS functions.
This version of the llc2 driver includes support for both connectionless and
connection-oriented logical link control class II (llc2) operations for Ethernet, Token Ring,
and FDDI adapters when accessed through the appropriate Solaris MAC layer driver. The
Data Link Provider Interface (DLPI) to the llc2 driver enables multiple and different protocol
stacks, (including NetBIOS and SNA), to operate simultaneously over one or more local area
networks.
To start the llc2 driver by default, rename file /etc/llc2/llc2_start.default to
/etc/llc2/llc2_start. This allows the /etc/rc2.d/S40llc2 script to build up the
configuration file for each ppa interface in /etc/llc2/default/llc2.* and start llc2 on
each interface. To verify the configuration files, manually run
/usr/lib/llc2/llc2_autoconfig.
For more information on the llc2 driver, see the IEEE standard 802.2 Logical Link Control.
Obtaining Llc2 You can obtain LLC2 statistics or reset the statistics counter to zero using the ILD_LLC2 ioctl.
Statistics The ILD_LLC2 ioctl has a number of subcommands. The following retrieve LLC2 statistics:

Name Function

LLC2_GET_STA_STATS Get station statistics

LLC2_GET_SAP_STATS Get SAP statistics

LLC2_GET_CON_STATS Get connection statistics

The structure used depends on the subcommand sent.

Llc2_GET_STA_ The LLC2_GET_STA_STATS command retrieves statistics on a particular Physical Point of
Stats Attachment (PPA).

When sending the LLC2_GET_STA_STATS command, the llc2GetStaStats structure is used:

typedef struct llc2GetStaStats {
uint_t ppa;
uint_t cmd;
uchar_t clearFlag;
uchar_t state;
ushort_t numSaps;
uchar_t saps[LLC2_MAX_SAPS];

500 man pages section 7: Device and Network Interfaces • Last Revised 12 Aug 1999
 llc2(7D)

uint_t nullSapXidCmdRcvd;
uint_t nullSapXidRspSent;
uint_t nullSapTestCmdRcvd;
uint_t nullSapTestRspSent;
uint_t outOfState;
uint_t allocFail;
uint_t protocolError;
} llc2GetStaStats_t;

The members of the structure are:

Member Description

cmd LLC2_GET_STA_STATS

clearFlag Clear counters flag. Set this to 0 to retreive statistics and to 1 to

reset all counters to 0.

state Station component state. Possible values are ?????

numSaps Number of active SAPs in the saps array

saps An array of active SAP values

nullSapXidCmdRcvd Number of NULL SAP XID commands received

nullSapXidRspSent Number of NULL SAP XID responses sent

nullSapTestCmdRcvd Number of NULL SAP TEST commands received

nullSapTestRspSent Number of NULL SAP TEST responses sent

outOfState Number of invalid events received

allocFail Number of buffer allocation failures

protocolError Number of protocol errors

Llc2_GET_SAP_ The LLC2_GET_SAP_STATS command retrieves statistics related to a particular SAP. When
Stats sending the LLC2_GET_SAP_STATS command, the llc2GetSapStats structure is used:
typedef struct llc2GetSapStats {
uint_t ppa;
uint_t cmd;
uchar_t sap;
uchar_t clearFlag;
uchar_t state;
uint_t numCons;
ushort_t cons[LLC2_MAX_CONS];
uint_t xidCmdSent;
uint_t xidCmdRcvd;
uint_t xidRspSent;
uint_t xidRspRcvd;

Device and Network Interfaces 501

llc2(7D)

uint_t testCmdSent;
uint_t testCmdRcvd;
uint_t testRspSent;
uint_t testRspRcvd;
uint_t uiSent;
uint_t uiRcvd;
uint_t outOfState;
uint_t allocFail;
uint_t protocolError;
} llc2GetSapStats_t;

The members are:

Member Description

ppa Physical Point of Attachment number

cmd LLC2_GET_SAP_STATS

sap SAP value

clearFlag Clear counters flag. Set this to 0 to retreive statistics and to 1 to reset all
counters to 0.

state SAP component state

numCons Number of active connections in the cons array

cons Array of active connection indexes

xidCmdSent Number of XID commands sent

xidCmdRcvd Number of XID responses received

xidRspSent Number of XID responses sent

xidRspRcvd Number of XID responses received

testCmdSent Number of TEST commands sent

testCmdRcvd Number of TEST commands received

testRspSent Number of TEST responses sent

testRspRcvd Number of TEST responses received

uiSent Number of UI frames sent

uiRcvd Number of UI frames received

outOfState Number of invalid events received

allocFail Number of buffer allocation failures

502 man pages section 7: Device and Network Interfaces • Last Revised 12 Aug 1999
 llc2(7D)

Member Description

protocolError Number of protocol errors

Llc2_GET_CON_STATSThe LLC2_GET_CON_STATS command retrieves statistics related to a particular connection

component. When sending the LLC2_GET_CON_STATS command, the llc2GetConStats
structure is used:
typedef struct llc2GetConStats {
uint_t ppa;
uint_t cmd;
uchar_t sap;
ushort_t con;
uchar_t clearFlag;
uchar_t stateOldest;
uchar_t stateOlder;
uchar_t stateOld;
uchar_t state;
ushort_t sid;
dlsap_t rem;
ushort_t flag;
uchar_t dataFlag;
uchar_t k;
uchar_t vs;
uchar_t vr;
uchar_t nrRcvd;
ushort_t retryCount;
uint_t numToBeAcked;
uint_t numToResend;
uint_t macOutSave;
uint_t macOutDump;
uchar_t timerOn;
uint_t iSent;
uint_t iRcvd;
uint_t frmrSent;
uint_t frmrRcvd;
uint_t rrSent;
uint_t rrRcvd;
uint_t rnrSent;
uint_t rnrRcvd;
uint_t rejSent;
uint_t rejRcvd;
uint_t sabmeSent;
uint_t sabmeRcvd;
uint_t uaSent;
uint_t uaRcvd;
uint_t discSent;
uint_t outOfState;
uint_t allocFail;

Device and Network Interfaces 503

llc2(7D)

uint_t protocolError;
uint_t localBusy;
uint_t remoteBusy;
uint_t maxRetryFail;
uint_t ackTimerExp;
uint_t pollTimerExp;
uint_t rejTimerExp;
uint_t remBusyTimerExp;
uint_t inactTimerExp;
uint_t sendAckTimerExp;
} llc2GetConStats_t;

The members of the structure are:

Member Description

ppa Physical Point of Attachment number

cmd LLC2_GET_CON_STATS

sap SAP value

con Connection index

clearFlag Clear counters flag. Set this to 0 to retreive statistics and to

1 to reset all counters to 0.

stateOldest, stateOlder, stateOld, state The four previous dlpi states of the connection

sid SAP value and connection index

dlsap_t rem Structure containing the remote MAC address and SAP

flag Connection component processing flag

dataFlag DATA_FLAG

k transmit window size

vs Sequence number of the next I-frame to send

vr Sequence number of the next I-frame expected

nrRcvd Sequence number of the last I-frame acknowledged by the

remote node

retryCount Number of timer expirations

numToBeAcked Number of outbound I-frames to be acknowledged

numToResend Number of outbound I-frames to be re-sent

macOutSave Number of outbound I-frames held by the MAC driver to

be saved on return to LLC2

504 man pages section 7: Device and Network Interfaces • Last Revised 12 Aug 1999
 llc2(7D)

Member Description

macOutDump Number of outbound I-frames held by the MAC driver to

be dumped on return to LLC2

timerOn Timer activity flag

iSent Number of I-frames sent

iRcvd Number of I-frames received

frmrSent Number of frame rejects sent

frmrRcvd Number of frame rejects received

rrSent Number of RRs sent

rrRcvd Number of RRs received

rnrRcvd Number of RNRs received

rejSent Number of rejects sent

rejRcvd Number of rejects received

sabmeSent Number of SABMEs sent

sabmeRcvd Number of SABMEs received

uaSent Number of UAs sent

uaRcvd Number of UAs received

discSent Number of DISCs sent

outOfState Number of invalid events received

allocFail Number of buffer allocation failures

protocolError Number of protocol errors

localBusy Number of times in a local busy state

remoteBusy Number of times in a remote busy state

maxRetryFail Number of failures due to reaching maxRetry

ackTimerExp Number of ack timer expirations

pollTimerExp Number of P-timer expirations

rejTimerExp Number of reject timer expirations

remBusyTimerExp Number of remote busy timer expirations

inactTimerExp Number of inactivity timer expirations

sendAckTimerExp Number of send ack timer expirations

Device and Network Interfaces 505

llc2(7D)

Member Description

Files /dev/llc2 Clone device used to access the driver /etc/llc2/default/llc2.?

configuration files (One file per ppa interface.)

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWllc

See Also llc2_autoconfig(1), llc2_config(1), llc2(4)

506 man pages section 7: Device and Network Interfaces • Last Revised 12 Aug 1999
 lockstat(7D)

Name lockstat – DTrace kernel lock instrumentation provider

Description The lockstat driver is a DTrace dynamic tracing provider that performs dynamic
instrumentation for locking primitives in the Solaris kernel.

The lockstat provider makes probes available that you can use to discern lock contention
statistics, or to understand virtually any aspect of locking behavior inside the operating system
kernel. The lockstat(1M) command is implemented as a DTrace consumer that uses the
lockstat provider to gather raw data.

The lockstat driver is not a public interface and you access the instrumentation offered by
this provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description of
the public documented interfaces available for the DTrace facility and the probes offered by
the lockstat provider.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), lockstat(1M), attributes(5), dtrace(7D)

Solaris Dynamic Tracing Guide

Device and Network Interfaces 507

lofi(7D)

Name lofi – Loopback file driver

Description The lofi file driver exports a file as a block device, enabling system utilities such as
fstyp(1M), fsck(1M) and mount(1M) to operate on underlying file system images (including
CD-ROM and FAT floppy images) contained on the block device. Reads and writes to the
block device are translated to reads and writes on the exported file. See lofiadm(1M) for
examples.

File block device entries are contained in /dev/lofi. The /dev/rlofi file contains the
character (or raw) device entries. Entries are in the form of decimal numbers and are assigned
through lofiadm(1M). When created, these device entries are owned by root, in group sys and
have permissions of 0600. Ownership, group, and permission settings can be altered, however
there may be ramifications. See lofiadm(1M) for more information.

lofi devices can be compressed. See lofiadm(1M) for more information.

Files exported through lofi can also be encrypted. See lofiadm(1M) for details on how to
specify encryption keys.
Files /dev/lofictl
Master control device
/dev/lofi/n
Block device for file n
/dev/rlofi/n
Character device for file n
/kernel/drv/lofi
32-bit driver
/kernel/drv/sparcv9/lofi
64-bit driver (SPARC)
/kernel/drv/amd64/lofi
64-bit driver (x86)
/kernel/drv/lofi.conf
Driver configuration file. (Do not alter).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWcsr, SUNWcarx.u

Interface Stability Committed

508 man pages section 7: Device and Network Interfaces • Last Revised 25 Nov 2008
 lofi(7D)

See Also fstyp(1M), fsck(1M), mount(1M), lofiadm(1M), newfs(1M), attributes(5), lofs(7FS)

Notes Just as you would not directly access a disk device that has mounted file systems, you should
not access a file associated with a block device except through the lofi file driver.

For compatibility purposes, a raw device is also exported with the block device. (For example,
newfs(1M)).

The lofi driver isn't available in a zone and will not work inside a zone.

Device and Network Interfaces 509

lofs(7FS)

Name lofs – loopback virtual file system

Synopsis #include <sys/param.h>
#include <sys/mount.h>

int mount (const char* dir, const char* virtual, int mflag, lofs, NULL, 0);

Description The loopback file system device allows new, virtual file systems to be created, which provide
access to existing files using alternate pathnames. Once the virtual file system is created, other
file systems can be mounted within it, without affecting the original file system. However, file
systems which are subsequently mounted onto the original file system are visible to the virtual
file system, unless or until the corresponding mount point in the virtual file system is covered
by a file system mounted there.

virtual is the mount point for the virtual file system. dir is the pathname of the existing file
system. mflag specifies the mount options; the MS_DATA bit in mflag must be set. If the
MS_RDONLY bit in mflag is not set, accesses to the loop back file system are the same as for the
underlying file system. Otherwise, all accesses in the loopback file system will be read-only. All
other mount(2) options are inherited from the underlying file systems.

A loopback mount of '/' onto /tmp/newroot allows the entire file system hierarchy to appear
as if it were duplicated under /tmp/newroot, including any file systems mounted from remote
NFS servers. All files would then be accessible either from a pathname relative to '/' or from a
pathname relative to /tmp/newroot until such time as a file system is mounted in
/tmp/newroot, or any of its subdirectories.

Loopback mounts of '/' can be performed in conjunction with the chroot(2) system call, to
provide a complete virtual file system to a process or family of processes.

Recursive traversal of loopback mount points is not allowed. After the loopback mount of
/tmp/newroot, the file /tmp/newroot/tmp/newroot does not contain yet another file system
hierarchy; rather, it appears just as /tmp/newroot did before the loopback mount was
performed (for example, as an empty directory).

Examples lofs file systems are mounted using:

mount -F lofs /tmp /mnt

See Also lofiadm(1M), mount(1M), chroot(2), mount(2), sysfs(2), vfstab(4), lofi(7D)

Notes All access to entries in lofs mounted file systems map to their underlying file system. If a
mount point is made available in multiple locations via lofs and is busy in any of those
locations, an attempt to mount a file system at that mount point fails unless the overlay flag is
specified. See mount(1M). Examples of a mount point being busy within a lofs mount include
having a file system mounted on it or it being a processes' current working directory.

510 man pages section 7: Device and Network Interfaces • Last Revised 31 Aug 2009
 lofs(7FS)

Warnings Because of the potential for confusing users and applications, you should use loopback
mounts with care. A loopback mount entry in /etc/vfstab must be placed after the mount
points of both directories it depends on. This is most easily accomplished by making the
loopback mount entry the last in /etc/vfstab.

Device and Network Interfaces 511

log(7D)

Name log – interface to STREAMS error logging and event tracing

Synopsis #include <sys/strlog.h>
#include <sys/log.h>

Description log is a STREAMS software device driver that provides an interface for console logging and
for the STREAMS error logging and event tracing processes (see strerr(1M), and
strace(1M)). log presents two separate interfaces: a function call interface in the kernel
through which STREAMS drivers and modules submit log messages; and a set of ioctl(2)
requests and STREAMS messages for interaction with a user level console logger, an error
logger, a trace logger, or processes that need to submit their own log messages.

Kernel Interface log messages are generated within the kernel by calls to the function strlog():

strlog(short mid,
short sid,
char level,
ushort_t flags,
char *fmt,
unsigned arg1 . . .
);

Required definitions are contained in <sys/strlog.h>, <sys/log.h>, and <sys/syslog.h>.

mid is the STREAMS module id number for the module or driver submitting the log message.
sid is an internal sub-id number usually used to identify a particular minor device of a driver.
level is a tracing level that allows for selective screening out of low priority messages from the
tracer. flags are any combination of SL_ERROR (the message is for the error logger), SL_TRACE
(the message is for the tracer), SL_CONSOLE (the message is for the console logger), SL_FATAL
(advisory notification of a fatal error), and SL_NOTIFY (request that a copy of the message be
mailed to the system administrator). fmt is a printf(3C) style format string, except that %s, %e,
%E, %g, and %G conversion specifications are not handled. Up to NLOGARGS (in this release,
three) numeric or character arguments can be provided.

User Interface log is implemented as a cloneable device, it clones itself without intervention from the system
clone device. Each open of /dev/log obtains a separate stream to log. In order to receive log
messages, a process must first notify log whether it is an error logger, trace logger, or console
logger using a STREAMS I_STR ioctl call (see below). For the console logger, the I_STR
ioctl has an ic_cmd field of I_CONSLOG, with no accompanying data. For the error logger, the
I_STR ioctl has an ic_cmd field of I_ERRLOG, with no accompanying data. For the trace
logger, the ioctl has an ic_cmd field of I_TRCLOG, and must be accompanied by a data buffer
containing an array of one or more struct trace_ids elements.
struct trace_ids {
short ti_mid;
short ti_sid;
char ti_level;
};

512 man pages section 7: Device and Network Interfaces • Last Revised 11 Mar 1998
 log(7D)

Each trace_ids structure specifies a mid, sid, and level from which messages will be accepted.
strlog(9F) will accept messages whose mid and sid exactly match those in the trace_ids
structure, and whose level is less than or equal to the level given in the trace_ids structure. A
value of −1 in any of the fields of the trace_ids structure indicates that any value is accepted
for that field.

Once the logger process has identified itself using the ioctl call, log will begin sending up
messages subject to the restrictions noted above. These messages are obtained using the
getmsg(2) function. The control part of this message contains a log_ctl structure, which
specifies the mid, sid, level, flags, time in ticks since boot that the message was submitted, the
corresponding time in seconds since Jan. 1, 1970, a sequence number, and a priority. The time
in seconds since 1970 is provided so that the date and time of the message can be easily
computed, and the time in ticks since boot is provided so that the relative timing of log
messages can be determined.
struct log_ctl {
short mid;
short sid;
char level; /* level of message for tracing */
short flags; /* message disposition */
#if defined(_LP64) || defined(_I32LPx)
clock32_t ltime; /* time in machine ticks since boot */
time32_t ttime; /* time in seconds since 1970 */
#else
clock_t ltime;
time_t ttime;
#endif
int seq_no; /* sequence number */
int pri; /* priority = (facility|level) */
};

The priority consists of a priority code and a facility code, found in <sys/syslog.h>. If
SL_CONSOLE is set in flags, the priority code is set as follows: If SL_WARN is set, the priority code
is set to LOG_WARNING; If SL_FATAL is set, the priority code is set to LOG_CRIT; If SL_ERROR is set,
the priority code is set to LOG_ERR; If SL_NOTE is set, the priority code is set to LOG_NOTICE; If
SL_TRACE is set, the priority code is set to LOG_DEBUG; If only SL_CONSOLE is set, the priority
code is set to LOG_INFO. Messages originating from the kernel have the facility code set to
LOG_KERN. Most messages originating from user processes will have the facility code set to
LOG_USER.

Different sequence numbers are maintained for the error and trace logging streams, and are
provided so that gaps in the sequence of messages can be determined (during times of high
message traffic some messages may not be delivered by the logger to avoid hogging system
resources). The data part of the message contains the unexpanded text of the format string
(null terminated), followed by NLOGARGS words for the arguments to the format string, aligned
on the first word boundary following the format string.

Device and Network Interfaces 513

log(7D)

A process may also send a message of the same structure to log, even if it is not an error or
trace logger. The only fields of the log_ctl structure in the control part of the message that are
accepted are the level, flags, and pri fields; all other fields are filled in by log before being
forwarded to the appropriate logger. The data portion must contain a null terminated format
string, and any arguments (up to NLOGARGS) must be packed, 32-bits each, on the next 32-bit
boundary following the end of the format string.

ENXIO is returned for I_TRCLOG ioctls without any trace_ids structures, or for any
unrecognized ioctl calls. The driver silently ignores incorrectly formatted log messages sent
to the driver by a user process (no error results).

Processes that wish to write a message to the console logger may direct their output to
/dev/conslog, using either write(2) or putmsg(2).

Driver Configuration The following driver configuration properties may be defined in the log.conf file.

msgid=1 If msgid=1, each message will be preceded by a message ID as described in

syslogd(1M).
msgid=0 If msgid=0, message IDs will not be generated. This property is unstable and may
be removed in a future release.

Examples EXAMPLE 1 I_ERRLOG registration.

struct strioctl ioc;
ioc.ic_cmd = I_ERRLOG;
ioc.ic_timout = 0; /* default timeout (15 secs.) */
ioc.ic_len = 0;
ioc.ic_dp = NULL;
ioctl(log, I_STR, &ioc);

EXAMPLE 2 I_TRCLOG registration.

struct trace_ids tid[2];
tid[0].ti_mid = 2;
tid[0].ti_sid = 0;
tid[0].ti_level = 1;
tid[1].ti_mid = 1002;
tid[1].ti_sid = −1; /* any sub-id will be allowed */
tid[1].ti_level = −1; /* any level will be allowed */
ioc.ic_cmd = I_TRCLOG;
ioc.ic_timout = 0;
ioc.ic_len = 2 * sizeof(struct trace_ids);
ioc.ic_dp = (char *)tid;
ioctl(log, I_STR, &ioc);

Example of submitting a log message (no arguments):

514 man pages section 7: Device and Network Interfaces • Last Revised 11 Mar 1998
 log(7D)

EXAMPLE 2 I_TRCLOG registration. (Continued)

struct strbuf ctl, dat;

struct log_ctl lc;
char *message = "Don’t forget to pick up some milk
on the way home";
ctl.len = ctl.maxlen = sizeof(lc);
ctl.buf = (char *)&lc;
dat.len = dat.maxlen = strlen(message);
dat.buf = message;
lc.level = 0;
lc.flags = SL_ERROR|SL_NOTIFY;
putmsg(log, &ctl, &dat, 0);
Files /dev/log Log driver.
/dev/conslog Write only instance of the log driver, for console logging.
/kernel/drv/log.conf Log configuration file.

See Also strace(1M), strerr(1M), Intro(3), getmsg(2), ioctl(2), putmsg(2), write(2), printf(3C),
strlog(9F)

STREAMS Programming Guide

Device and Network Interfaces 515

lp(7D)

Name lp – driver for parallel port

Synopsis include <sys/bpp_io.h>
fd = open("/dev/lpn", flags);

Description The lp driver provides the interface to the parallel ports used by printers for x86 based
systems. The lp driver is implemented as a STREAMS device.
ioctls BPPIOC_TESTIO Test transfer readiness. This command checks to see if a read or write
transfer would succeed based on pin status. If a transfer would succeed, 0
is returned. If a transfer would fail, −1 is returned, and errno is set to EIO.
The error status can be retrieved using the BPPIOC_GETERR ioctl() call.
BPPIOC_GETERR Get last error status. The argument is a pointer to a struct
bpp_error_status. See below for a description of the elements of this
structure. This structure indicates the status of all the appropriate status
bits at the time of the most recent error condition during a read(2) or
write(2) call, or the status of the bits at the most recent BPPIOC_TESTIO
ioctl(2) call. The application can check transfer readiness without
attempting another transfer using the BPPIOC_TESTIO ioctl().

Error Pins Structure This structure and symbols are defined in the include file <sys/bpp_io.h>:

struct bpp_error_status {
char timeout_occurred; /* Not use */
char bus_error; /* Not use */
uchar_t pin_status; /* Status of pins which could cause an error */
};

/* Values for pin_status field */

#define BPP_ERR_ERR 0x01 /* Error pin active */
#define BPP_SLCT_ERR 0x02 /* Select pin active */
#define BPP_PE_ERR 0x04 /* Paper empty pin active */

Note: Other pin statuses are defined in <sys/bpp_io.h>, but BPP_ERR_ERR, BPP_SLCT_ERR
and BPP_PE_ERR are the only ones valid for the x86 lp driver.
Errors EIO A BPPIOC_TESTIO ioctl() call is attempted while a condition exists that would
prevent a transfer (such as a peripheral error).
EINVAL An ioctl() is attempted with an invalid value in the command argument.
Files /platform/i86pc/kernel/drv/lp.conf configuration file for lp driver

Attributes See attributes(5) for descriptions of the following attributes:

516 man pages section 7: Device and Network Interfaces • Last Revised 21 May 1997
 lp(7D)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also sysbus(4), attributes(5), streamio(7I)

Notes A read operation on a bi-directional parallel port is not supported.

Device and Network Interfaces 517

lsimega(7D)

Name lsimega – SCSI HBA driver for LSI MegaRAID 320-2x SCSI RAID Controller

Description The lsimega SCSI RAID host bus adapter driver is a SCSA-compliant nexus driver that
supports LSI MegaRAID 320-2x SCSI RAID devices. The driver also supports LSI MegaRAID
320-1, 320-2e, Dell PERC 4e/Si, 4e/Di, 4e/DC RAID controllers.

The lsimega driver supports standard functions provided by the SCSA interface, including
RAID disk I/O and passthrough I/O to support SCSI tape and CDROM.

Driver The lsimega.conf file contains no user configurable parameters. Please configure your
Configuration hardware through BIOS.

The LSI MegaRAID 320-2x device can support up to 40 logic drivers. Note that BIOS numbers
the logic drives as 1 through 40, however in Solaris these drives are numbered from 0 to 39.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability x86

Files /kernel/drv/lsimega 32-bit ELF kernel module.

/kernel/drv/amd64/lsimega 64–bit kernel module (x86 only).
/kernel/drv/lsimega.conf Driver configuration file (contains no user-configurable
options).

See Also prtconf(1M), attributes(5), amr(7D), scsi_hba_attach_setup(9F), scsi_sync_pkt(9F),

scsi_transport(9F), scsi_inquiry(9S), scsi_device(9S), scsi_pkt(9S)

Small Computer System Interface-2 (SCSI-2)

518 man pages section 7: Device and Network Interfaces • Last Revised 6 April 2006
 lx_systrace(7D)

Name lx_systrace – DTrace Linux system call tracing provider

Description The lx_systrace driver implements the DTrace lxsyscall dynamic tracing provider. The
lxsyscall provider performs dynamic instrumentation to offer probes that fire whenever a
thread enters or returns from a Linux system call entry point.

The lx_systrace driver is not a public interface and you access the instrumentation offered
by this provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description
of the public documented interfaces available for the DTrace facility and the probes offered by
the lxsyscall provider.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWlxr

Interface Stability Private

See Also dtrace(1M), attributes(5), lx(5), dtrace(7D)

Solaris Dynamic Tracing Guide

Device and Network Interfaces 519

m64(7D)

Name m64 – PGX, PGX24, and PGX64 frame buffers device driver

Description The m64 driver is the Sun PGX graphics accelerator device driver.
Files /dev/fbs/m64\fIn Device special file

See Also m64config(1M)

520 man pages section 7: Device and Network Interfaces • Last Revised 05 Feb 2001
 marvell88sx(7D)

Name marvell88sx – Marvell 88SX SATA controller driver

Synopsis sata@unit-address

Description The marvell88sx driver is a SATA framework-compliant HBA driver that supports the
Marvell 88SX5081, 88SX5080, 88SX5040, 88SX5041, 88SX6081, and 88SX6041 controllers.

The 88SX5081, 88SX5080, 88SX5040 and 88SX5041 Marvell controllers are fully compliant
with the Serial ATA 1.0 specification and support the SATA device hot-swap compliant 1.5
Gbps speed feature.

The 88SX6081 and 88SX6041 Marvell controllers are fully-compliant with the SATA II Phase
1.0 specification (the extension to the SATA 1.0 specification) and support SATA II native
command queuing and backwards compatibility with SATA I 1.5 Gpbs speed and devices. In
addition, the 88SX6081 device supports the SATA II Phase 1.0 specification features,
including SATA II 3.0 Gbps speed, SATA II Port Multiplier functionality and SATA II Port
Selector. Currently the driver does not support native command queuing, port multiplier or
port selector functionality.

Configuration There are no tunable parameters in the marvell88sx.conf file.

Files /kernel/drv/marvell88sx 32–bit ELF 86 kernel module.
/kernel/drv/amd64/marvell88sx 64–bit ELF kernel module.
/kernel/drv/marvell88sx.conf Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWmv88sx

See Also cfgadm(1M), prtconf(1M), cfgadm_sata(1M), attributes(5), nv_sata(7D), sata(7D)

Diagnostics In addition to being logged, the following messages may appear on the system console:
marvell88sx<n>:PCI error address 0x<high addr>:<low addr>
PCI command 0x<command>DAC [true|false] attribute
0x<attribute><pci error type>.

The nth instance of a marvell88sx reports a PCI bus status message. (A hardware issue needs
attention). The PCI bus address, PCI command (whether or not it was a dual address
command), the PCI-X attribute bit, and the error type are displayed.
marvell88sx<n>: port <port #>: error in PIO command 0<cmd>x:
status 0x<status>.

Device and Network Interfaces 521

marvell88sx(7D)

The port number on the nth marvell88sx controller received an error while performing a
programmed I/O command <cmd> with status <status>.
marvell88sx<n>: error on port<port#>: One or more of the following:
ATA UDMA data parity error
ATA UDMA PRD parity error
device error
device disconnected
device connected
SError interrupt
reserved bit 6
EDMA self disabled
BIST FIS or asynchronous notification
command request queue parity error
command response queue parity error
internal memory parity error
I/O ready time-out
link control receive error - crc
link control receive error - fifo
link control receive error - reset
link control receive error - state
link data receive error - crc
link data receive error - fifo
link data receive error - reset
link data receive error - state
link control transmit error - crc
link control transmit error - fifo
link control transmit error - reset
link control transmit error - DMAT
link control transmit error - collision
link data transmit error -crc
link data transmit error - fifo
link data transmit error - reset
link data transmit error - DMAT
link data transmit error - collision
transport protocol error

The port number on the nth marvell88sx controller received one or more error conditions as
listed.
marvell88sx<n>: device on port <port #> still busy.

The port number on the nth marvell88sx remains busy. (Indicates a hardware problem).
Check the disk and the controller.
marvell88sx<n>: pci_config_setup failed.

Could not access PCI configuration space for the nth marvell88sx controller.
marvell88sx<n>:failed to get device id.

522 man pages section 7: Device and Network Interfaces • Last Revised 17 Aug 2007
 marvell88sx(7D)

The device-id property for the nth marvell88sx controller cannot be read.
marvell88sx<n>: Unrecognized device - device id 0x<device id>
assuming <n> ports.

The device id associated with the nth marvell88sx controller is not supported and the number
of ports could not be determined. n ports are being assumed.
marvell88sx<n>:Unrecognized device - device idOx<device id>.

The device id associated with the nth marvell88sx controller is not supported.
marvell88sx<n>: Could not attach. Could not allocate softstate.

A call to ddi_soft_state_zalloc() failed for the nth marvell88sx controller. The system may be
low on resources. The driver failed to attach.
marvell88sx<n>: Could not attach, unknown device model.

The nth marvell88sx controller is unsupported hardware. The driver failed to attach.
marvell88sx<n>: Could not attach, unsupported chip step-
ping or unable to get the chip stepping.

The nth marvell88sx controller is not supported due to a known bad chip stepping or a
stepping of an unknown model.
marvell88sx<n>: ddi_intr_get_supported_types failed.

The driver failed to attach.

marvell88sx<n>: power management component not created.

Power management is not supported.

mavell88sx<n>: unable to attach to sata framework.

The driver failed to attach.

marvell88sx<n>: unable to detach from sata framework.

The driver failed to detach.

marvell88sx<n>: Could not attach, failed interrupt registration.

The driver failed to attach.

marvell88sx<n>: Cannot get number interrupts, rc

The number of interrupts for the nth marvell88sx device could not be determined.
marvell88sx<n>: 0 is not a valid number of interrupts.

The number of interrupts for the nth marvell88sx device was returned as 0.
marvell88sx<n>: Failed to get the number of available interrupts.

Device and Network Interfaces 523

marvell88sx(7D)

The number of available interrupts for the nth marvell88sx controller could not be
determined.
marvell88sx<n>: Number of available interrupts is 0.

No interrupts were available for the nth marvell88sx device.

marvell88sx<n>: could not allocate interrupts.

The interrupts for the nth marvell88sx device could not be allocated.
marvell88sx<n>: could not get interrupt priority.

The interrupt priority for the nth marvell88sx device could not be determined.
marvell88sx<n>: Could not add interrupt handler.

An interrupt service routine could not be added to the system for the nth marvell88sx device.
marvell88sx<n>:polled read/write request never completed- port <num>.

A polled read or write did not complete in a reasonable amount of time. If this problem
persists, there may be a hardware problem with (a) the controller, (b) the controller port, (c)
the disk attached to controller port or (d) the cabling.
marvell88sx<n>: EDMA never disabled.

Could not disable EDMA. (Indicates a hardware problem).

marvell88sx<n>: Could not attach.

The nth marvell8ssx device could not attach. This message is usually preceded by another
warning indicating why the attach failed.

524 man pages section 7: Device and Network Interfaces • Last Revised 17 Aug 2007
 mc-opl(7D)

Name mc-opl – memory controller driver for the SPARC Enterprise Server family

Description The mc-opl driver is the memory controller driver for the SPARC Enterprise Server family.
This driver manages the hardware memory-scrubbing operations.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr

Interface Stability Private

See Also attributes(5)

Device and Network Interfaces 525

md(7D)

Name md – user configurable pseudo device driver

Description md is a user configurable pseudo device driver that provides disk concatenation, striping,
mirroring, RAID5 metadevices, trans metadevices, and hot spare utilities. Trans devices are
no longer supported and have been replaced by UFS logging. See mount_ufs(1M).

The block devices access the disk using the system's normal buffering mechanism and are read
and written without regard to physical disk records. There is also a ‘‘raw'' device which
provides for direct transmission between the disk and the user's read or write buffer. A single
read or write call usually results in one I/O operation; raw I/O is therefore considerably more
efficient when many bytes are transmitted. The names of the block devices are found in
/dev/md/dsk; the names of the raw devices are found in /dev/md/rdsk. Metadevices have the
appearance of whole disks; there are no slices (partitions).

I/O requests (such as lseek(2)) to the metadevices must have an offset that is a multiple of 512
bytes (DEV_BSIZE), or the driver returns an EINVAL error. If the transfer length is not a
multiple of 512 bytes, the tranfer count is rounded up by the driver.

The md pseudo device drivers support all disk devices on all Solaris 2.4 or later Solaris systems.

ioctls This section provides a list of the ioctls supported by the metadisk driver.

The following ioctls are valid when issued to the raw metadevice, such as /dev/md/rdsk/d0.
See dkio(7I) for additional information.
DKIOCGGEOM This ioctl is used to get the disk geometry. The metadisk driver fills in the
dkg_nhead, dkg_nsect, dkg_rpm, dkg_write_reinstruct and
dkg_read_reinstruct from the first component of the metadevice (at
metainit time). dkg_ncyl is calculated using the size of the metadevice
(reported by metastat) divided by (dkg_nhead * dkg_nsect). The total size is
always a multiple of (dkg_nhead * dkg_nsect). If the first component of a
metadevice does not start on cylinder number 0, then the dkg_ncyl is
increased by one cylinder; because DKIOCGVTOC reports the metadevice as
starting on cylinder 1. The side effect here is that it looks like cylinder 0 is not
being used, but all the arithmetic works out correctly. If the metadevice is not
set up, then ENXIO is returned.
DKIOCINFO When issued to the administrative device or metadevice, this ioctl sets
dki_unit to the unit number of the metadevice, dki_ctype to a value of
DKC_MD, and dki_partition to 0, because there are no slices.
DKIOCGVTOC This ioctl returns the current vtoc. If one has not been written, then a default
vtoc is returned. v_nparts is always 1. v_part[0].p_start is 0 if the first
component of the metadevice starts on cylinder 0. Otherwise, the p_start
field is the starting sector of cylinder 1. v_part[0].p_size is the same as the
total size reported by metastat.

526 man pages section 7: Device and Network Interfaces • Last Revised 29 Aug 2003
 md(7D)

DKIOCSVTOC This ioctl stores the vtoc in the metadevice state database so it is persistent
across reboots.

Diagnostics

Notice Log Messages The informative log messages include:

md: dnum: Hotspared device dev with dev

The first device name listed has been hot spare replaced with the second device name listed.
md: dnum: Hotspared device dev(num,num) with dev(num,num)

The first device number listed has been hot spare replaced with the second device number
listed.
md: Could not load misc /dev

The named misc module is not loadable. It is possibly missing, or something else has been
copied over it.
md: dnum: no mem for property dev

Memory could not be allocated in the prop_op entry point.

md: db: Parsing error on ’dev’

Set command in /kernel/drv/md.conf for the mddb.bootlist <number> is not in the

correct format. metadb -p can be run to put the correct set commands into the
/kernel/drv/md.conf file.
md: dnum: dev(num,num) needs maintenance
md: dnum: dev needs maintenance

An I/O or open error has occurred on a device within a mirror causing a component in the
mirror to change to the Maintenance state.

md: dnum: dev(num,num) last erred md: dnum: dev last erred

An I/O or open error has occurred on a device within a mirror and the data is not replicated
elsewhere in the mirror. This is causing the component in the mirror to change to the Last
Erred state.

Warning Log Messages The warning log messages include:

md: State database is stale

This error message comes when there are not enough usable replicas for the state database to
be able to update records in the database. All accesses to the metadevice driver will fail. To fix
this problem, more replicas need to be added or inaccessible replicas need to be deleted.

Device and Network Interfaces 527

md(7D)

md: dnum: read error on devmd: dnum: write error on dev

A read or write error has occurred on the specified submirror, at the specified device name.
This happens if any read or write errors occur on a submirror.
md: dnum: read error on dev(num,num)md: dnum: write error on dev(
num,num)

A read or write error has occurred on the specified submirror, at the specified device number.
This happens if any read or write errors occur on a submirror.
md: State database commit failed
md: State database delete failed

These messages occur when there have been device errors on components where the state
database replicas reside. These errors only occur when more than half of the replicas have had
device errors returned to them. For instance, if you have three components with state database
replicas and two of the components report errors, then these errors may occur. The state
database commit or delete is retried periodically. If a replica is added, then the commit or
delete will finish and the system will be operational. Otherwise the system will timeout and
panic.
md: dnum: Cannot load dev driver

Underlying named driver module is not loadable (for example, sd, id, xy, or a third-party
driver). This could indicate that the driver module has been removed.
md: Open error of hotspare devmd: Open error of hotspare dev(num,num)

Named hotspare is not openable, or underlying driver is not loadable.

Panic Log Messages The panic log messages include:

md: dnum: Unknown close typemd: dnum: Unknown open type

Metadevice is being opened/closed with an unknown open type (OTYP).

md: State database problem

Failed metadevice state database commit or delete has been retried the default 100 times.
Files /dev/md/dsk/dn block device (where n is the device number)
/dev/md/rdsk/dn raw device (where n is the device number)
/dev/md/setname/dsk/dn block device (where setname is the name of the diskset and
n is the device number)
/dev/md/setname/rdsk/dn raw device (where setname is the name of the diskset and n
is the device number)
/dev/md/admin administrative device

528 man pages section 7: Device and Network Interfaces • Last Revised 29 Aug 2003
 md(7D)

/kernel/drv/md driver module

/kernel/drv/md.conf driver configuration file
/kernel/misc/md_stripe stripe driver misc module
/kernel/misc/md_mirror mirror driver misc module
/kernel/misc/md_hotspares hotspares driver misc module
/kernel/misc/md_trans metatrans driver for UFS logging
/kernel/misc/md_raid RAID5 driver misc module

See Also mdmonitord(1M), metaclear(1M), metadb(1M), metadetach(1M), metahs(1M),

metainit(1M), metaoffline(1M), metaonline(1M), metaparam(1M), metarecover(1M),
metarename(1M), metareplace(1M), metaroot(1M), metassist(1M), metaset(1M),
metastat(1M), metasync(1M), metattach(1M), md.cf(4), md.tab(4), attributes(5),

Solaris Volume Manager Administration Guide

Notes Trans metadevices have been replaced by UFS logging. Existing trans devices are not
logging--they pass data directly through to the underlying device. See mount_ufs(1M) for
more information about UFS logging.

Device and Network Interfaces 529

mediator(7D)

Name mediator – support for HA configurations consisting of two strings of drives

Description Beginning with a prior version, Solaris Volume Manager provided support for
high-availability (HA) configurations consisting of two hosts that share at least three strings of
drives and that run software enabling exclusive access to the data on those drives from one
host. (Note: Volume Manager, by itself, does not actually provide a high-availability
environment. The diskset feature is an enabler for HA configurations.)

Volume Manager provides support for a low-end HA solution consisting of two hosts that
share only two strings of drives. The hosts in this type of configuration, referred to as
mediators, run a special daemon, rpc.metamedd(1M). The mediator hosts take on additional
responsibilities to ensure that data is available in the case of host or drive failures.

In a mediator configuration, two hosts are physically connected to two strings of drives. This
configuration can survive the failure of a single host or a single string of drives, without
administrative intervention. If both a host and a string of drives fail (multiple failures), the
integrity of the data cannot be guaranteed. At this point, administrative intervention is
required to make the data accessible.

The following definitions pertain to a mediator configuration:

diskset A set of drives containing metadevices and hot spares
that can be shared exclusively (but not concurrently) by
two hosts.
Volume Manager state database A replicated database that stores metadevice
configuration and state information.
mediator host A host that runs the rpc.metamedd(1M) daemon and
that has been added to a diskset. The mediator host
participates in checking the state database and the
mediator quorum.
mediator quorum The condition achieved when the number of accessible
mediator hosts is equal to half+1 the total number of
configured mediator hosts. Because it is expected that
there will be two mediator hosts, this number will
normally be 2 ([(2/2) + 1] = 2.)
replica A single copy of the Volume Manager metadevice state
database.
replica quorum The condition achieved when the number of accessible
replicas is equal to half+1 the total number of
configured replicas. For example, if a system is
configured with ten replicas, the quorum is met when
six are accessible ([(10/2) + 1 = 6]).

530 man pages section 7: Device and Network Interfaces • Last Revised 20 Jun 2008
 mediator(7D)

A mediator host running the rpc.metamedd(1M) daemon keeps track of replica updates. As
long as the following conditions are met, access to data occurs without any administrative
intervention:
■ The replica quorum is not met.
■ Half of the replicas are still accessible.
■ The mediator quorum is met.

The following conditions describe the operation of mediator hosts:

1. If the is met, access to the diskset is granted. At this point no mediator host is involved.
2. If the replica quorum is not met, half of the replicas are accessible, the mediator quorum is
met, and the replica and mediator data match, access to the diskset is granted. The
mediator host contributes the deciding vote.
3. If the replica quorum is not met, half of the replicas are accessible, the mediator quorum is
not met, half of the mediator hosts is accessible, and the replica and mediator data match,
the system prompts you to grant or deny access to the diskset.
4. If the replica quorum is not met, half of the replicas are accessible, the mediator quorum is
met, and the replica and mediator data do not match, access to the diskset is read-only.
You can delete replicas, release the diskset, and retake the diskset to gain read-write access
to the data in the diskset.
5. In all other cases, the diskset access is read-only. You can delete replicas, release the
diskset, and retake the diskset to gain read-write access to the data in the diskset.

The metaset(1M) command administers disksets and mediator hosts. The following options
to the metaset command pertain only to administering mediator hosts.
-a -m mediator_host_list Adds mediator hosts to the named set. A mediator_host_list is
the nodename of the mediator host to be added and up to 2 other
aliases for the mediator host. The nodename and aliases for each
mediator host are separated by commas. Up to 3 mediator hosts
can be specified for the named diskset.
-d -m mediator_host_list Deletes mediator hosts from the named diskset. Mediator hosts
are deleted from the diskset by specifying the nodename of
mediator host to delete.
-q Displays an enumerated list of tags pertaining to ‘‘tagged data''
that may be encountered during a take of the ownership of a
diskset.
-t [-f] -y Takes ownership of a diskset safely, unless -f is used, in which
case the take is unconditional. If metaset finds that another host
owns the set, this host will not be allowed to take ownership of
the set. If the set is not owned by any other host, all the disks
within the set will be owned by the host on which metaset was

Device and Network Interfaces 531

mediator(7D)

executed. The metadevice state database is read in and the shared

metadevices contained in the set become accessible. The -t
option will take a diskset that has stale databases. When the
databases are stale, metaset will exit with code 66, and a message
will be printed. At that point, the only operations permitted are
the addition and deletion of replicas. Once the addition or
deletion of the replicas has been completed, the diskset should be
released and retaken to gain full access to the data. If mediator
hosts have been configured, some additional exit codes are
possible. If half of the replicas and half of the mediator hosts are
operating properly, the take will exit with code 3. At this point,
you can add or delete replicas, or use the -y option on a
subsequent take. If the take operation encounters ‘‘tagged data,''
the take operation will exit with code 2. You can then run the
metaset command with the -q option to see an enumerated list
of tags.
-t [-f] -u tagnumber Once a tag has been selected, a subsequent take with -u
tagnumber can be executed to select the data associated with the
given tagnumber.

See Also metaset(1M), md(7D), rpc.metamedd(1M), rpc.metad(1M)

Sun Cluster documentation, Solaris Volume Manager Administration Guide

Notes Diskset administration, including the addition and deletion of hosts and drives, requires all
hosts in the set to be accessible from the network.

532 man pages section 7: Device and Network Interfaces • Last Revised 20 Jun 2008
 mega_sas(7D)

Name mega_sas – SCSI HBA driver for LSI MegaRAID SAS controller

Description The mega_sas MegaRAID controller host bus adapter driver is a SCSA-compliant nexus
driver that supports the Dell PERC 5/E, 5/i, 6/E and 6/i RAID controllers, the IBM
ServeRAID-MR10k SAS/SATA controller and the LSI MegaRAID SAS/SATA 8308ELP,
8344ELP, 84016E, 8408ELP, 8480ELP, 8704ELP, 8704EM2, 8708ELP, 8708EM2, 8880EM2 and
8888ELP series of controllers.

Supported RAID features include RAID levels 0, 1, 5, and 6, RAID spans 10, 50 and 60, online
capacity expansion (OCE), online RAID level migration (RLM), auto resume after loss of
system power during arrays, array rebuild or reconstruction (RLM) and configurable stripe
size up to 1MB. Additional supported RAID features include check consistency for
background data integrity, patrol read for media scanning and repairing, 64 logical drive
support, up to 64TB LUN support, automatic rebuild and global and dedicated hot spare
support.

Configuration The mega_sas.conf file contains no user configurable parameters. Please configure your
hardware through the related BIOS utility or the MegaCli configuration utility. If you want to
install to a drive attached to a mega_sas HBA, you should create the virtual drive first from the
BIOS before running the Solaris install. You can obtain the MegaCli utility from the LSI
website.

The mega_sas device can support up to 64 virtual disks. Note that BIOS numbers the virtual
disks as 1 through 64, however in the Solaris operating environment virtual disks are
numbered from 0 to 63. Also note that SAS and SATA drives cannot be configured into the
same virtual disk.

Known Problems The mega_sas driver does not support the LSI MegaRAID SAS 8204ELP, 8204XLP, 8208ELP,
and Limitations and 8208XLP controllers.

Files /kernel/drv/mega_sas 32-bit ELF kernel module. (x86)

/kernel/drv/amd64/mega_sas 64-bit kernel module. (x86)
/kernel/drv/mega_sas.conf Driver configuration file (contains no user-configurable
options).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86-based systems

Availability SUNWmegasas

Interface stability Uncommitted

Device and Network Interfaces 533

mega_sas(7D)

See Also prtconf(1M), attributes(5), sata(7D), scsi_hba_attach_setup(9F), scsi_sync_pkt(9F),

scsi_transport(9F), scsi_inquiry(9S), scsi_device(9S), scsi_pkt(9S)

Small Computer System Interface-2 (SCSI-2)

534 man pages section 7: Device and Network Interfaces • Last Revised 14 Aug 2008
 mem(7D)

Name mem, kmem, allkmem – physical or virtual memory access

Synopsis /dev/mem
/dev/kmem
/dev/allkmem

Description The file /dev/mem is a special file that provides access to the physical memory of the computer.

The file /dev/kmem is a special file that provides access to the virtual address space of the
operating system kernel, excluding memory that is associated with an I/O device.

The file /dev/allkmem is a special file that provides access to the virtual address space of the
operating system kernel, including memory that is associated with an I/O device. You can use
any of these devices to examine and modify the system.

Byte addresses in /dev/mem are interpreted as physical memory addresses. Byte addresses in
/dev/kmem and /dev/allkmem are interpreted as kernel virtual memory addresses. A reference
to a non-existent location returns an error. See ERRORS for more information.

The file /dev/mem accesses physical memory; the size of the file is equal to the amount of
physical memory in the computer. This size may be larger than 4GB on a system running the
32-bit operating environment. In this case, you can access memory beyond 4GB using a series
of read(2) and write(2) calls, a pread64() or pwrite64() call, or a combination of llseek(2)
and read(2) or write(2).
Errors EFAULT Occurs when trying to write(2) a read-only location (allkmem), read(2) a
write-only location (allkmem), or read(2) or write(2) a non-existent or
unimplemented location (mem, kmem, allkmem).
EIO Occurs when trying to read(2) or write(2) a memory location that is associated
with an I/O device using the /dev/kmem special file.
ENXIO Results from attempting to mmap(2) a non-existent physical (mem) or virtual (kmem,
allkmem) memory address.
Files /dev/mem Provides access to the computer's physical memory.
/dev/kmem Provides access to the virtual address space of the operating system kernel,
excluding memory that is associated with an I/O device.
/dev/allkmem Provides access to the virtual address space of the operating system kernel,
including memory that is associated with an I/O device.

See Also llseek(2), mmap(2), read(2), write(2)

Warnings Using these devices to modify (that is, write to) the address space of a live running operating
system or to modify the state of a hardware device is extremely dangerous and may result in a
system panic if kernel data structures are damaged or if device state is changed.

Device and Network Interfaces 535

mhd(7i)

Name mhd – multihost disk control operations

Synopsis #include <sys/mhd.h>

Description The mhd ioctl(2) control access rights of a multihost disk, using disk reservations on the disk
device.

The stability level of this interface (see attributes(5)) is evolving. As a result, the interface is
subject to change and you should limit your use of it.

The mhd ioctls fall into two major categories: (1) ioctls for non-shared multihost disks and (2)
ioctls for shared multihost disks.

One ioctl, MHIOCENFAILFAST, is applicable to both non-shared and shared multihost disks. It is
described after the first two categories.

All the ioctls require root privilege.

For all of the ioctls, the caller should obtain the file descriptor for the device by calling open(2)
with the O_NDELAY flag; without the O_NDELAY flag, the open may fail due to another host
already having a conflicting reservation on the device. Some of the ioctls below permit the
caller to forcibly clear a conflicting reservation held by another host, however, in order to call
the ioctl, the caller must first obtain the open file descriptor.

Non-shared multihost Non-shared multihost disks ioctls consist of MHIOCTKOWN, MHIOCRELEASE, HIOCSTATUS, and
disks MHIOCQRESERVE. These ioctl requests control the access rights of non-shared multihost disks.
A non-shared multihost disk is one that supports serialized, mutually exclusive I/O mastery by
the connected hosts. This is in contrast to the shared-disk model, in which concurrent access
is allowed from more than one host (see below).

A non-shared multihost disk can be in one of two states:

■ Exclusive access state, where only one connected host has I/O access
■ Non-exclusive access state, where all connected hosts have I/O access. An external
hardware reset can cause the disk to enter the non-exclusive access state.

Each multihost disk driver views the machine on which it's running as the "local host"; each
views all other machines as "remote hosts". For each I/O or ioctl request, the requesting host is
the local host.

Note that the non-shared ioctls are designed to work with SCSI-2 disks. The SCSI-2
RESERVE/RELEASE command set is the underlying hardware facility in the device that
supports the non-shared ioctls.

The function prototypes for the non-shared ioctls are:

ioctl(fd, MHIOCTKOWN);
ioctl(fd, MHIOCRELEASE);

536 man pages section 7: Device and Network Interfaces • Last Revised 1 Sep 2005
 mhd(7i)

ioctl(fd, MHIOCSTATUS);
ioctl(fd, MHIOCQRESERVE);
MHIOCTKOWN Forcefully acquires exclusive access rights to the multihost disk for the
local host. Revokes all access rights to the multihost disk from remote
hosts. Causes the disk to enter the exclusive access state.

Implementation Note: Reservations (exclusive access rights) broken via

random resets should be reinstated by the driver upon their detection, for
example, in the automatic probe function described below.
MHIOCRELEASE Relinquishes exclusive access rights to the multihost disk for the local
host. On success, causes the disk to enter the non- exclusive access state.
MHIOCSTATUS Probes a multihost disk to determine whether the local host has access
rights to the disk. Returns 0 if the local host has access to the disk, 1 if it
doesn't, and -1 with errno set to EIO if the probe failed for some other
reason.
MHIOCQRESERVE Issues, simply and only, a SCSI-2 Reserve command. If the attempt to
reserve fails due to the SCSI error Reservation Conflict (which implies
that some other host has the device reserved), then the ioctl will return –1
with errno set to EACCES. The MHIOCQRESERVE ioctl does NOT issue a bus
device reset or bus reset prior to attempting the SCSI-2 reserve
command. It also does not take care of re-instating reservations that
disappear due to bus resets or bus device resets; if that behavior is desired,
then the caller can call MHIOCTKOWN after the MHIOCQRESERVE has returned
success. If the device does not support the SCSI-2 Reserve command,
then the ioctl returns –1 with errno set to ENOTSUP. The MHIOCQRESERVE
ioctl is intended to be used by high-availability or clustering software for
a "quorum" disk, hence, the "Q" in the name of the ioctl.

Shared Multihost Disks Shared multihost disks ioctls control access to shared multihost disks. The ioctls are merely a
veneer on the SCSI-3 Persistent Reservation facility. Therefore, the underlying semantic
model is not described in detail here, see instead the SCSI-3 standard. The SCSI-3 Persistent
Reservations support the concept of a group of hosts all sharing access to a disk.

The function prototypes and descriptions for the shared multihost ioctls are as follows:
ioctl(fd, MHIOCGRP_INKEYS, (mhioc_inkeys_t) *k);
Issues the SCSI-3 command Persistent Reserve In Read Keys to the device. On input, the
field k->li should be initialized by the caller with k->li.listsize reflecting how big of an
array the caller has allocated for the k->li.list field and with k->li.listlen == 0. On
return, the field k->li.listlen is updated to indicate the number of reservation keys the
device currently has: if this value is larger than k->li.listsize then that indicates that the
caller should have passed a bigger k->li.list array with a bigger k->li.listsize. The
number of array elements actually written by the callee into k->li.list is the minimum of

Device and Network Interfaces 537

mhd(7i)

k->li.listlen and k->li.listsize. The field k->generation is updated with the

generation information returned by the SCSI-3 Read Keys query. If the device does not
support SCSI-3 Persistent Reservations, then this ioctl returns –1 with errno set to
ENOTSUP.
ioctl(fd, MHIOCGRP_INRESV, (mhioc_inresvs_t) *r);
Issues the SCSI-3 command Persistent Reserve In Read Reservations to the device.
Remarks similar to MHIOCGRP_INKEYS apply to the array manipulation. If the device does
not support SCSI-3 Persistent Reservations, then this ioctl returns –1 with errno set to
ENOTSUP.
ioctl(fd, MHIOCGRP_REGISTER, (mhioc_register_t) *r);
Issues the SCSI-3 command Persistent Reserve Out Register. The fields of structure r are all
inputs; none of the fields are modified by the ioctl. The field r->aptpl should be set to true
to specify that registrations and reservations should persist across device power failures, or
to false to specify that registrations and reservations should be cleared upon device power
failure; true is the recommended setting. The field r->oldkey is the key that the caller
believes the device may already have for this host initiator; if the caller believes that that this
host initiator is not already registered with this device, it should pass the special key of all
zeros. To achieve the effect of unregistering with the device, the caller should pass its
current key for the r->oldkey field and an r->newkey field containing the special key of all
zeros. If the device returns the SCSI error code Reservation Conflict, this ioctl returns –1
with errno set to EACCES.
ioctl(fd, MHIOCGRP_RESERVE, (mhioc_resv_desc_t) *r);
Issues the SCSI-3 command Persistent Reserve Out Reserve. The fields of structure r are all
inputs; none of the fields are modified by the ioctl. If the device returns the SCSI error code
Reservation Conflict, this ioctl returns –1 with errno set to EACCES.
ioctl(fd, MHIOCGRP_PREEMPTANDABORT, (mhioc_preemptandabort_t) *r);
Issues the SCSI-3 command Persistent Reserve Out Preempt-And-Abort. The fields of
structure r are all inputs; none of the fields are modified by the ioctl. The key of the victim
host is specified by the field r->victim_key. The field r->resvdesc supplies the
preempter's key and the reservation that it is requesting as part of the SCSI-3
Preempt-And-Abort command. If the device returns the SCSI error code Reservation
Conflict, this ioctl returns –1 with errno set to EACCES.
ioctl(fd, MHIOCGRP_PREEMPT, (mhioc_preemptandabort_t) *r);
Similar to MHIOCGRP_PREEMPTANDABORT, but instead issues the SCSI-3 command Persistent
Reserve Out Preempt. (Note: This command is not implemented).
ioctl(fd, MHIOCGRP_CLEAR, (mhioc_resv_key_t) *r);
Issues the SCSI-3 command Persistent Reserve Out Clear. The input parameter r is the
reservation key of the caller, which should have been already registered with the device, by
an earlier call to MHIOCGRP_REGISTER. (Note: This command is not implemented).

For each device, the non-shared ioctls should not be mixed with the Persistent Reserve Out
shared ioctls, and vice-versa, otherwise, the underlying device is likely to return errors,

538 man pages section 7: Device and Network Interfaces • Last Revised 1 Sep 2005
 mhd(7i)

because SCSI does not permit SCSI-2 reservations to be mixed with SCSI-3 reservations on a
single device. It is, however, legitimate to call the Persistent Reserve In ioctls, because these are
query only. Issuing the MHIOCGRP_INKEYS ioctl is the recommended way for a caller to
determine if the device supports SCSI-3 Persistent Reservations (the ioctl will return –1 with
errno set to ENOTSUP if the device does not).

MHIOCENFAILFAST The MHIOCENFAILFAST ioctl is applicable for both non-shared and shared disks, and may be
Ioctl used with either the non-shared or shared ioctls.

ioctl(fd, MHIOENFAILFAST, (unsigned int *) millisecs);

Enables or disables the failfast option in the multihost disk driver and enables or disables
automatic probing of a multihost disk, described below. The argument is an unsigned
integer specifying the number of milliseconds to wait between executions of the automatic
probe function. An argument of zero disables the failfast option and disables automatic
probing. If the MHIOCENFAILFAST ioctl is never called, the effect is defined to be that both
the failfast option and automatic probing are disabled.

Automatic Probing The MHIOCENFAILFAST ioctl sets up a timeout in the driver to periodically schedule automatic
probes of the disk. The automatic probe function works in this manner: The driver is
scheduled to probe the multihost disk every n milliseconds, rounded up to the next integral
multiple of the system clock's resolution. If
1. the local host no longer has access rights to the multihost disk, and
2. access rights were expected to be held by the local host,

the driver immediately panics the machine to comply with the failfast model.

If the driver makes this discovery outside the timeout function, especially during a read or
write operation, it is imperative that it panic the system then as well.

Return Values Each request returns –1 on failure and sets errno to indicate the error.
EPERM Caller is not root.
EACCES Access rights were denied.
EIO The multihost disk or controller was unable to successfully complete the
requested operation.
EOPNOTSUP The multihost disk does not support the operation. For example, it does not
support the SCSI-2 Reserve/Release command set, or the SCSI-3 Persistent
Reservation command set.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWhea

Device and Network Interfaces 539

mhd(7i)

Stability Evolving

See Also ioctl(2), open(2), attributes(5), open(2)

540 man pages section 7: Device and Network Interfaces • Last Revised 1 Sep 2005
 mixer(7I)

Name mixer – generic mixer device interface

Synopsis #include <sys/soundcard.h>
Description .

Mixer Pseudo-Device The /dev/mixer pseudo-device is provided for two purposes:

■ The first purpose is for applications that wish to learn about the list of audio devices on the
system, so that they can select (or provide for users to select) an appropriate audio device.
The /dev/mixer pseudo-device provides interfaces to enumerate all of the audio devices
on the system.
■ The second purpose is for mixer panel type applications which need to control master
settings for the audio hardware in the system, such as gain levels, balance, port
functionality, and other device features.

Ordinary audio applications should not attempt to adjust their playback or record volumes or
other device settings using this device. Instead, they should use the SNDCTL_DSP_SETPLAYVOL
and SNDCTL_DSP_SETRECVOL ioctls that are documented in dsp(7I).

Sndstat Device The /dev/sndstat device supports read(2), and can be read to retrieve human-readable
information about the audio devices on the system. Software should not attempt to interpret
the contents of this device.

ioctls

Information IOCTLs The following ioctls are intended to aid applications in identifying the audio devices available
on the system. These ioctls can be issued against either the pseudo-device /dev/mixer, or a
against a file descriptor open to any other audio device in the system.

Applications should issue SNDCTL_SYSINFO first to learn what audio devices and mixers are
available on the system, and then use SNDCTL_AUDIOINFO or SNDCTL_MIXERINFO to obtain
more information about the audio devices or mixers, respectively.
OSS_GETVERSION The argument is a pointer to an integer, which retrieves the version of
the OSS API used. The value is encoded with the major version
(currently 4) encoded in the most significant 16 bits, and a minor
version encoded in the lower 16 bits.
SNDCTL_SYSINFO The argument is a pointer to an oss_sysinfo structure, which has the
following definition:
typedef struct oss_sysinfo {
char product[32]; /* E.g. SunOS Audio */
char version[32]; /* E.g. 4.0a */
int versionnum; /* See OSS_GETVERSION */
char options[128]; /* NOT SUPPORTED */

int numaudios; /* # of audio/dsp devices */

Device and Network Interfaces 541

mixer(7I)

int openedaudio[8]; /* Reserved, always 0 */

int numsynths; /* NOT SUPPORTED, always 0 */

int nummidis; /* NOT SUPPORTED, always 0 */
int numtimers; /* NOT SUPPORTED, always 0 */
int nummixers; /* # of mixer devices */

int openedmidi[8]; /* Mask of midi devices are

busy */
int numcards; /* Number of sound cards in
the system */
int numaudioengines; /* Number of audio engines in
the system */
char license[16]; /* E.g. "GPL" or "CDDL" */
char revision_info[256]; /* Reserved */
int filler[172]; /* Reserved */
} oss_sysinfo;
The important fields here are numaudios, which is used to determine
the number of audio devices that can be queried with
SNDCTL_AUDIOINFO, nummixers which provides a count of mixers on
the system, and numcards which counts to total number of aggregate
devices. A card can consist of one or more audio devices and one or
more mixers, although more typically there is exactly one audio
device and one mixer for each card.
SNDCTL_AUDIOINFO The argument is a pointer to an oss_audioinfo structure, which has
the following structure:
typedef struct oss_audioinfo {
int dev; /* Device to query */
char name[64]; /* Human readable name */
int busy; /* reserved */
int pid; /* reserved */
int caps; /* PCM_CAP_INPUT, PCM_CAP_OUTPUT */
int iformats; /* Supported input formats */
int oformats; /* Supported output formats */
int magic; /* reserved */
char cmd[64]; /* reserved */
int card_number;
int port_number; /* reserved */
int mixer_dev;
int legacy_device; /* Obsolete field.
Replaced by devnode */
int enabled; /* reserved */
int flags; /* reserved */
int min_rate; /* Minimum sample rate */
int max_rate; /* Maximum sample rate */

542 man pages section 7: Device and Network Interfaces • Last Revised 21 May 2009
 mixer(7I)

int min_channels; /* Minimum number

of channels */
int max_channels; /* Maximum number
of channels */
int binding; /* reserved */
int rate_source; /* reserved */
char handle[32]; /* reserved */
unsigned int nrates; /* reserved */
unsigned int rates[20]; /* reserved */
char song_name[64]; /* reserved */
char label[16]; /* reserved */
int latency; /* reserved */
char devnode[32]; /* Device special file
name (absolute path) */
int next_play_engine; /* reserved */
int next_rec_engine; /* reserved */
int filler[184]; /* reserved */
} oss_audioinfo;
In the above structure, all of the fields are reserved except the
following: dev, name, card_number, mixer_dev, caps,
min_rate, max_rate, min_channels, max_channels, and
devnode. The reserved fields are provided for compatibility with other
OSS implementations, and available for legacy applications. New
applications should not attempt to use these fields.

The dev field should be initialized by the application to the number of

the device to query. This is a number between zero (inclusive) and
value of numaudios (exclusive) returned by SNDCTL_SYSINFO.
Alternatively, when issuing the ioctl against a real mixer or dsp device,
the special value -1 can be used to indicate that the query is being
made against the device opened. If -1 is used, the field is overwritten
with the device number for the current device on successful return.

No other fields are significant upon entry, but a successful return

contains details of the device.

The name field is a human readable name representing the device.

Applications should not try to interpret it.

The card_number field indicates the number assigned to the aggregate

device. This can be used with the SNDCTL_CARDINFO ioctl.

The mixer_dev is the mixer device number for the mixing device
associated with the audio device. This can be used with the
SNDCTL_MIXERINFO ioctl.

Device and Network Interfaces 543

mixer(7I)

The caps field contains any of the bits PCM_CAP_INPUT,

PCM_CAP_OUTPUT, and PCM_CAP_DUPLEX. Indicating whether the device
support input, output, and whether input and output can be used
simultaneously. All other bits are reserved.

The min_rate and max_rate fields indicate the minimum and

maximum sample rates supported by the device. Most applications
should try to use the maximum supported rate for the best audio
quality and lowest system resource consumption.

The min_channels and max_channels provide an indication of the

number of channels (1 for mono, 2 for stereo, 6 for 5.1, etc.) supported
by the device.

The devnode field contains the actual full path to the device node for
this device, such as /dev/sound/audio810:0dsp. Applications should
open this file to access the device.
SNDCTL_CARDINFO The argument is a pointer to a struct oss_card_info, which has the
following definition:
typedef struct oss_card_info {
int card;
char shortname[16];
char longname[128];
int flags;/* reserved */
char hw_info[400];
int intr_count;/* reserved */
int ack_count;/* reserved */
int filler[154];
} oss_card_info;
This ioctl is used to query for information about the aggregate audio
device.

The card field should be initialized by the application to the number

of the card to query. This is a number between zero inclusive and
value of numcards (exclusive) returned by SNDCTL_SYSINFO.)
Alternatively, when issuing the ioctl against a real mixer or dsp device,
the special value -1 can be used to indicate that the query is being
made against the device opened. If -1 is used, the field is overwritten
with the number for the current hardware device on successful return.

The shortname, longname, and hw_info contain ASCIIZ strings

describing the device in more detail. The hw_info member can
contain multiple lines of detail, each line ending in a NEWLINE.

544 man pages section 7: Device and Network Interfaces • Last Revised 21 May 2009
 mixer(7I)

The flag, intr_count, and ack_count fields are not used by this
implementation.
SNDCTL_MIXERINFO The argument is a pointer to a struct oss_mixer_info, which has
the following definition:
typedef struct oss_mixerinfo {
int dev;
char id[16];/* Reserved */
char name[32];
int modify_counter;
int card_number;
int port_number;/* Reserved */
char handle[32];/* Reserved */
int magic;/* Reserved */
int enabled;/* Reserved */
int caps;/* Reserved */
int flags;/* Reserved */
int nrext;
int priority;
char devnode[32];/* Device special file name
(absolute path) */
int legacy_device;/* Reserved */
int filler[245];/* Reserved */
} oss_mixerinfo;
In the above structure, all of the fields are reserved except the
following: dev, name, modify_counter, card_number, nrext,
priority, and devnode. The reserved fields are provided for
compatibility with other OSS implementations, and available for
legacy applications. New applications should not attempt to use these
fields.

The dev field should be initialized by the application to the number of

the device to query. This is a number between zero inclusive and value
of nummixers (exclusive) returned by SNDCTL_SYSINFO, or by
SNDCTL_MIX_NRMIX. Alternatively, when issuing the ioctl against a real
mixer or dsp device, the special value -1 can be used to indicate that
the query is being made against the device opened. If -1 is used, the
field is overwritten with the mixer number for the current open file on
successful return.

No other fields are significant upon entry, but on successful return

contains details of the device.

The name field is a human readable name representing the device.

Applications should not try to interpret it.

Device and Network Interfaces 545

mixer(7I)

The modify_counter is changed by the mixer framework each time

the settings for the various controls or extensions of the device are
changed. Applications can poll this value to learn if any other changes
need to be searched for.

The card_number field is the number of the aggregate audio device

this mixer is located on. It can be used with the SNDCTL_CARDINFO
ioctl.

The nrext field is the number of mixer extensions available on this

mixer. See the SNDCTL_MIX_NREXT description.

The priority is used by the framework to assign a preference that

applications can use in choosing a device. Higher values are
preferable. Mixers with priorities less than -1 should never be selected
by default.

The devnode field contains the actual full path to the device node for
the physical mixer, such as /dev/sound/audio810:0mixer.
Applications should open this file to access the mixer settings.

Mixer Extension IOCTLs The pseudo /dev/mixer device supports ioctls that can change the various settings for the
audio hardware in the system.

Those ioctls should only be used by dedicated mixer applications or desktop volume controls,
and not by typical ordinary audio applications such as media players. Ordinary applications
that wish to adjust their own volume settings should use the SNDCTL_DSP_SETPLAYVOL or
SNDCTL_DSP_SETRECVOL ioctls for that purpose. See dsp(7I) for more information. Ordinary
applications should never attempt to change master port selection or hardware settings such
as monitor gain settings.

The ioctls in this section can only be used to access the mixer device that is associated with the
current file descriptor.

Applications should not assume that a single /dev/mixer node is able to access any physical
settings. Instead, they should use the ioctl SNDCTL_MIXERINFO to determine the device path for
the real mixer device, and issue ioctls on a file descriptor opened against the corresponding
devnode field.

When a dev member is specified in each of the following ioctls, the application should specify
-1, although for compatibility the mixer allows the application to specify the mixer device
number.
SNDCTL_MIX_NRMIX The argument is a pointer to an integer, which receives the
number of mixer devices in the system. Each can be queried by

546 man pages section 7: Device and Network Interfaces • Last Revised 21 May 2009
 mixer(7I)

using its number with the SNDCTL_MIXERINFO ioctl. The same

information is available using the SNDCTL_SYSINFO ioctl.
SNDCTL_MIX_NREXT The argument is a pointer to an integer. On entry, the integer
should contain the special value -1. On return the argument
receives the number of mixer extensions (or mixer controls)
supported by the mixer device. More details about each extension
can be obtained by SNDCTL_MIX_EXTINFO ioctl.
SNDCTL_MIX_EXTINFO The argument is a pointer to an oss_mixext structure which is
defined as follows:
typedef struct oss_mixext {
int dev; /* Mixer device number */
int ctrl; /* Extension number */
int type; /* Entry type */
int maxvalue;
int minvalue;
int flags;
char id[16]; /* Mnemonic ID (internal use) */
int parent; /* Entry# of parent
(-1 if root) */
int dummy; /* NOT SUPPORTED */
int timestamp;
char data[64]; /* Reserved */
unsigned char enum_present[32]; /* Mask
of allowed
enum
values */
int control_no; /* Reserved */
unsigned int desc; /* NOT SUPPORTED */
char extname[32];
int update_counter;
int filler[7]; /* Reserved */
} oss_mixext;
On entry, the dev field should be initialized to the value -1, and the
ctrl field should be initialized with the number of the extension
being accessed. Between 0, inclusive, and the value returned by
SNDCTL_MIX_NREXT, exclusive.

Mixer extensions are organized as a logical tree, starting with a

root node. The root node always has a ctrl value of zero. The
structure of the tree can be determined by looking at the parent
field, which contains the extension number of the parent
extension, or -1 if the extension is the root extension.

Device and Network Interfaces 547

mixer(7I)

The type indicates the type of extension used. This

implementation supports the following values:
MIXT_DEVROOT Root node for extension tree
MIXT_GROUP Logical grouping of controls
MXIT_ONOFF Boolean value, 0 = off, 1 = on.
MIXT_ENUM Enumerated value, 0 to maxvalue.
MIXT_MONOSLIDER Monophonic slider, 0 to 255.
MIXT_STEREOSLIDER Stereophonic slider, 0 to 255
(encoded as
lower two bytes in value.)
MIXT_MARKER Place holder, can ignore.
The flags field is a bit array. This implementation makes use of the
following possible bits:
MIXF_READABLE Extension’s value is readable.
MIXF_WRITEABLE Extension’s value is modifiable.
MIXF_POLL Extension can self-update.
MIXF_PCMVOL Extension is for master
PCM playback volume.
MIXF_MAINVOL Extension is for a typical
analog volume
MIXF_RECVOL Extension is for master
record gain.
MIXF_MONVOL Extension is for a monitor
source’s gain.
The id field contains an ASCIIZ identifier for the extension.

The timestamp field is set when the extension tree is first

initialized. Applications must use the same timestamp value when
attempting to change the values. A change in the timestamp
indicates a change a in the structure of the extension tree.

The enum_present field is a bit mask of possible enumeration

values. If a bit is present in the enum_present mask, then the
corresponding enumeration value is legal. The mask is in little
endian order.

The desc field provides information about scoping, which can be

useful as layout hints to applications. The following hints are
available:
MIXEXT_SCOPE_MASK Mask of possible scope
values.
MIXEXT_SCOPE_INPUT Extension is an input
control.
MIXEXT_SCOPE_OUTPUT Extension is an

548 man pages section 7: Device and Network Interfaces • Last Revised 21 May 2009
 mixer(7I)

output control.
MIXEXT_SCOPE_MONITOR Extension relates to
input monitoring.
MIXEXT_SCOPE_OTHER No scoping hint provided.
The extname is the full name of the extension.

The update_counter is incremented each time the control's value

is changed.
SNDCTL_MIX_ENUMINFO The argument is a pointer to an oss_mixer_enuminfo structure,
which is defined as follows:
typedef struct oss_mixer_enuminfo {
int dev;
int ctrl;
int nvalues;
int version;
short strindex[255];
char strings[3000];
} oss_mixer_enuminfo;
On entry, the dev field should be initialized to the value -1, and the
ctrl field should be initialized with the number of the extension
being accessed. Between 0, inclusive, and the value returned by
SNDCTL_MIX_NREXT, exclusive.

On return the nvalues field contains the number of values, and

strindex contains an array of indices into the strings member,
each index pointing to an ASCIIZ describing the enumeration
value.
SNDCTL_MIX_READ
SNDCTL_MIX_WRITE The argument is a pointer to an oss_mixer_value structure,
defined as follows:
typedef struct oss_mixer_value {
int dev;
int ctrl;
int value;
int flags; /* Reserved for future use.
Initialize to 0 */
int timestamp; /* Must be set to
oss_mixext.timestamp */
int filler[8]; /* Reserved for future use.
Initialize to 0 */
} oss_mixer_value;

Device and Network Interfaces 549

mixer(7I)

On entry, the dev field should be initialized to the value -1, and the
ctrl field should be initialized with the number of the extension
being accessed. Between 0, inclusive, and the value returned by
SNDCTL_MIX_NREXT, exclusive. Additionally, the timestamp
member must be initialized to the same value as was supplied in
the oss_mixext structure used with SNDCTL_MIX_EXTINFO.

For SNDCTL_MIX_WRITE, the application should supply the new

value for the extension. For SNDCTL_MIX_READ, the mixer returns
the extensions current value in value.

Compatibility IOCTLs The following ioctls are for compatibility use only:

SOUND_MIXER_READ_VOLUME
SOUND_MIXER_READ_PCM
SOUND_MIXER_READ_OGAIN
SOUND_MIXER_READ_RECGAIN
SOUND_MIXER_READ_RECLEV
SOUND_MIXER_READ_IGAIN
SOUND_MIXER_READ_RECSRC
SOUND_MIXER_READ_RECMASK
SOUND_MIXER_READ_DEVMASK
SOUND_MIXER_READ_STEREODEVS
SOUND_MIXER_WRITE_VOLUME
SOUND_MIXER_WRITE_PCM
SOUND_MIXER_WRITE_OGAIN
SOUND_MIXER_WRITE_RECGAIN
SOUND_MIXER_WRITE_RECLEV
SOUND_MIXER_WRITE_IGAIN
SOUND_MIXER_WRITE_RECSRC
SOUND_MIXER_WRITE_RECMASK
SOUND_MIXER_INFO
SNDCTL_AUDIOINFO_EX
SNDCTL_ENGINEINFO

These ioctls can affect the software volume levels associated with the calling process. They
have no effect on the physical hardware levels or settings. They should not be used in new
applications.

Errors An ioctl() fails if:

EINVAL The parameter changes requested in the ioctl are invalid or are not supported by
the device.
ENXIO The device or extension referenced does not exist.

550 man pages section 7: Device and Network Interfaces • Last Revised 21 May 2009
 mixer(7I)

Files /dev/mixer Symbolic link to the pseudo mixer device for the system
/dev/sndstat Sound status device

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWcsu, SUNWaudd, SUNWaudh

Stability Level See below.

The information and mixer extension IOCTLs are Uncommitted. The Compatibility IOCTLs
are Obsolete Uncommitted. The extension names are Uncommitted.

See Also close(2), ioctl(2), open(2), , read(2), attributes(5), dsp(7I)

Bugs The names of mixer extensions are not guaranteed to be predictable.

Device and Network Interfaces 551

mpt(7D)

Name mpt – SCSI host bus adapter driver

Synopsis scsi@unit-address
Description The mpt host bus adapter driver is a SCSA compliant nexus driver that supports the LSI
53C1030 SCSI, SAS1064, SAS1068 and Dell SAS 6i/R controllers.
The mpt driver supports the standard functions provided by the SCSA interface, including
tagged and untagged queuing, Narrow/Wide/Fast/Ultra SCSI/Ultra SCSI 2/Ultra SCSI 3/Ultra
SCSI 4, and auto request sense. The mpt driver does not support linked commands. The mpt
driver also supports SATA and Serial-Attached SCSI devices when connected to LSI SAS1064
(PCI-X), SAS1068 and Dell SAS 6i/R (PCI-Express) controllers.
Driver The mpt driver obtains configuration parameters from the /kernel/drv/mpt.conf file. These
Configuration parameters can override global SCSI settings.

The following configurable properties are applicable for parallel SCSI controllers and devices:
scsi-options, target<n>-scsi-options, scsi-reset-delay, scsi-tag-age-limit,
scsi-watchdog-tick, and scsi-initiator-id.
The property target<n>-scsi-options overrides the scsi-options property value for
target<n>, where <n> can vary from decimal 0 to 15 for parallel SCSI operations. The mpt
driver supports the following parallel SCSI options: SCSI_OPTIONS_DR, SCSI_OPTIONS_SYNC,
SCSI_OPTIONS_TAG, SCSI_OPTIONS_FAST, SCSI_OPTIONS_WIDE, SCSI_OPTIONS_FAST20,
SCSI_OPTIONS_FAST40, SCSI_OPTIONS_FAST80, SCSI_OPTIONS_FAST160, and
SCSI_OPTIONS_QAS. To view the numeric values of these options, see
/usr/include/sys/scsi/conf/autoconf.h.
The scsi-reset-delay and scsi-watchdog-tick properties are applicable for
Serial-Attached SCSI (SAS) controllers and SAS or SATA devices.
After periodic interval scsi-watchdog-tick, the mpt driver searches through all current and
disconnected commands for timeouts.
The scsi-tag-age-limit property is ignored by mpt, regardless of controller or devices type.
Refer to scsi_hba_attach_setup(9F) for more details of parallel SCSI properties and flags.
When supported, multipath-capable storage is attached with Serial-Attached SCSI or SATA.
Solaris I/O Multipathing may be enabled for mpt instances. This feature is configured with the
mpxio-disable property in the mpt.conf file. To perform mutipathing tasks, we recommend
that you use stmsboot(1M). Specifying mpxio-disable="no" enables the feature, while
specifying mpxio-disable="yes" disables the feature. Solaris I/O Multipathing may be
enabled or disabled on a per-controller basis. The following example shows how to disable
multipathing on a controller whose parent is /pci@7c0/pci@0/pci@9 and unit-address is 0:
name="mpt" parent="/pci@7c0/pci@0/pci@9" unit-address="0" mpxio-disable="yes";

Currently, mpt supports the mpt_offline_delay property. This property delays the offlining
of a device until the timer has expired. The default value is 20 seconds.

552 man pages section 7: Device and Network Interfaces • Last Revised 24 Nov 2008
 mpt(7D)

mpt supports the mpt-on-bus-time property, which controls a timer that resets a bus when a
bus connection exceeds the timer value. The default value of mpt-on-bus-time is 15 seconds.
A value of 0 disables this feature. The property can be configured in /kernel/drv/mpt.conf
as mpt-on-bus-time. In the following example, the timeout is disabled for unit 4 and set to
two minutes for unit 4,1:
name="mpt" parent="/pci@1d,700000"
unit-address="4"
mpt-on-bus-time=0;
name="mpt" parent="/pci@1d,700000"
unit-address="4,1"
mpt-on-bus-time=120;

Values have the following effect:

No property configured: Default, 15 second timeout

n = 0: Disables bus timeout feature
0 < n <= 15: Minimum (and default), 15 seconds
15 < n <= 3435: The actual value in seconds
3435 < n: Maximum, 3435 seconds

Examples EXAMPLE 1 Using the mpt Configuration File

Create a file called /kernel/drv/mpt.conf, then add the following line:
scsi-options=0x78;

The above example disables tagged queuing, Fast/Ultra SCSI, and wide mode for all mpt
instances. The property value is calculated by or-ing the individual SCSI_OPTIONS_xxx
values defined in /usr/include/sys/scsi/conf/autoconf.h.
The following example disables an option for one specific parallel SCSI mpt device. See
driver.conf(4) and pci(4) for more details.
name="mpt" parent="/pci@1f,4000"
unit-address="3"
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

Note that the default initiator ID is 7 and that the change to ID 6 occurs at attach time. It may
be preferable to change the initiator ID with eeprom(1M).
The example above sets scsi-options for target 1 to 0x58 and all other targets on this SCSI
bus to 0x178.
You can determine the physical path name of the parent by using the /devices tree or by
following the link of the logical device name:
# ls -l /dev/rdsk/c0t0d0s0
lrwxrwxrwx 1 root root 45 May 16 10:08 /dev/rdsk/c0t0d0s0 ->
. . / . . /devices/pci@1f,4000/scsi@3/sd@0,0:a,raw

Device and Network Interfaces 553

mpt(7D)

EXAMPLE 1 Using the mpt Configuration File (Continued)

As in the previous example, the parent is /pci@1f,4000 and the unit-address is 3.

To set scsi-options more specifically per target, do the following:

target1-scsi-options=0x78;
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;
scsi-options=0x3f8;

The above sets scsi-options for target 1 to 0x78. All other targets on the SCSI bus are set to
0x3f8 (with the exception of one specific disk type for which scsi-options is set to 0x58).

scsi-options specified per target ID have the highest precedence, followed by scsi-options
per device type. Global scsi-options (for all mpt instances) per bus have the lowest
precedence.

You must reboot the system for the specified scsi-options to take effect.

SCSI Transport SCSI transport capabilities as set by the target driver. The following capabilities can be queried
Capabilities and modified by the target driver: synchronous, tagged-qing, wide-xfer, auto-rqsense,
qfull-retries, and qfull-retry-interval. All other capabilities are query only.

By default, tagged-qing, auto-rqsense, and wide-xfer capabilities are disabled, while

disconnect, synchronous, and untagged-qing are enabled. These capabilities can only have
binary values (0 or 1).The default value for qfull-retries is 10, while the default value for
qfull-retry-interval is 100. The qfull-retries capability is a uchar_t (0 to 255), while
qfull-retry-interval is a ushort_t (0 to 65535).

The target driver must enable tagged-qing and wide-xfer explicitly. The untagged-qing
capability is always enabled and its value cannot be modified.

If a conflict exists between the value of scsi-options and a capability, the value set in
scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call.

Refer to scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.

Files /kernel/drv/mpt 32–bit ELF kernel module
/kernel/drv/sparcv9/mpt 64–bit SPARC ELF kernel module
/kernel/drv/amd64/mpt 64–bit x86 ELF kernel module
/kernel/drv/mpt.conf Optional configuration file

554 man pages section 7: Device and Network Interfaces • Last Revised 24 Nov 2008
 mpt(7D)

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86, SPARC (Limited to systems with LSI Fusion

family SCSI I/O processors).

See Also eeprom(1M), prtconf(1M), stmsboot(1M), driver.conf(4), pci(4), attributes(5),

scsi_vhci(7D), scsi_abort(9F), scsi_hba_attach_setup(9F), scsi_ifgetcap(9F),
scsi_ifsetcap(9F), scsi_reset(9F), scsi_sync_pkt(9F), scsi_transport(9F),
scsi_device(9S), scsi_extended_sense(9S), scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

ANSI Small Computer System Interface-2 (SCSI-2) (and later revisions).

ANSI Serial-Attached SCSI-2 (SAS2)

SYM53c1030 PCI-SCSI I/O processor Dual Channel Fast-160 — LSI Logic Inc.

LSISASI1064 PCI-X to 4-port 3 Gb/s SAS Controller - LSI Logic Inc.

LSISASI1068/E 4-Port PCI Express to 3 Gb/s SAS Controller - LSI Logic Inc.

Sun StorEdge Traffic Manager Installation and Configuration Guide

Diagnostics The messages described below are logged and may also appear on the system console.
Device is using a hilevel intr The device was configured with an
interrupt level that cannot be used with
this mtp driver. Check the PCI device.
Map setup failed The driver was unable to map device
registers; check for bad hardware. Driver
did not attach to device; SCSI devices are
inaccessible.
Cannot map configuration space The driver was unable to map in the
configuration registers. Check for bad
hardware. SCSI devices will be
inaccessible.
Attach failed The driver was unable to attach; usually
preceded by another warning that
indicates why attach failed. These can be
considered hardware failures.
Connected command timeout for Target <id>. This is usually a SCSI bus problem. Check
cables and termination.

Device and Network Interfaces 555

mpt(7D)

Target <id> reducing sync. transfer rate A data transfer hang or DATA-IN phase
parity error was detected. The driver
attempts to eliminate this problem by
reducing the data transfer rate.
Target <id> reverting to async. mode A second data transfer hang was detected
for this target. The driver attempts to
eliminate this problem by reducing the
data transfer rate.
Target <id> disabled wide SCSI mode A second data phase hang was detected for
this target. The driver attempts to
eliminate this problem by disabling wide
SCSI mode.

Notes The mpt driver supports the parallel SCSI LSI 53c1030 controller. The LSI 53c1030 controller
series supports Wide, Fast and Ultra SCSI 4 mode. The maximum LVD SCSI bandwidth is 320
MB/sec.

The mpt driver exports properties indicating the negotiated transfer speed per target
(target<n>-sync-speed), whether wide bus is supported (target<n>-wide) for that
particular target (target<n>-scsi-options), and whether tagged queuing has been enabled
(target<n>-TQ). The sync-speed property value indicates the data transfer rate in KB/sec.
The target<n>-TQ and the target<n>-wide property have value 1 (to indicate that the
corresponding capability is enabled for that target), or 0 (to indicate that the capability is
disabled for that target). See prtconf(1M) (verbose option) for details on viewing the mpt
properties.
scsi, instance #4
Driver properties:
name=’target8-TQ’ type=int items=1 dev=none
value=00000001
name=’target8-wide’ type=int items=1 dev=none
value=00000001
name=’target8-sync-speed’ type=int items=1 dev=none
value=00013880
name=’target5-TQ’ type=int items=1 dev=none
value=00000001
name=’target5-wide’ type=int items=1 dev=none
value=00000001
name=’target5-sync-speed’ type=int items=1 dev=none
value=00013880
name=’target4-TQ’ type=int items=1 dev=none
value=00000001
name=’target4-wide’ type=int items=1 dev=none
value=00000001
name=’target4-sync-speed’ type=int items=1 dev=none

556 man pages section 7: Device and Network Interfaces • Last Revised 24 Nov 2008
 mpt(7D)

value=00013880
name=’pm-components’ type=string items=3 dev=none
value=’NAME=mpt4’ + ’0=Off (PCI D3 State)’ + ’3=On (PCI \
D0 State)’
name=’scsi-selection-timeout’ type=int items=1 dev=(238,0)
value=000000fa
name=’scsi-options’ type=int items=1 dev=(238,0)
value=00103ff8
name=’scsi-watchdog-tick’ type=int items=1 dev=(238,0)
value=0000000a
name=’scsi-tag-age-limit’ type=int items=1 dev=(238,0)
value=00000002
name=’scsi-reset-delay’ type=int items=1 dev=(238,0)
value=00000bb8

Device and Network Interfaces 557

mpt_sas(7D)

Name mpt_sas – SAS-2 host bus adapter driver

Synopsis scsi@unit-address

Description The mpt_sas host bus adapter driver is a nexus driver that supports the LSI SAS200x/2108
series of chips. These chips support SAS/SATA interfaces, including tagged and untagged
queuing, SATA 3G/SAS 3G/SAS 6G. The mpt_sas driver supports the following Dell cards:
PERC H200 Integrated, PERC H200 Adapter, PERC H200 Modular, and 6Gbps SAS HBA.

Configuration The mpt_sas driver is configured by defining properties in mpt_sas.conf. These properties
override the global SCSI settings. The mpt_sas driver supports one modifiable property:
mpxio-disable
Solaris I/O multipathing is enabled or disabled on SAS devices with the mpxio-disable
property. Specifying mpxio-disable="no" activates I/O multipathing, while
mpxio-disable="yes" disables I/O multipathing.

Solaris I/O multipathing can be enabled or disabled on a per port basis. Per port settings
override the global setting for the specified ports.

The following example shows how to disable multipathing on port 0 whose parent is
/pci@0,0/pci8086,2940@1c/pci1000,72@0:
name="mpt_sas" parent="/pci@0,0/pci8086,2940@1c/pci1000,72@0"
mpxio-disable="yes";

Examples EXAMPLE 1 Using the mpt_sas Configuration File to Disable MPXIO

Create a file called /kernel/drv/mpt_sas.conf and add the following line:

name="mpt_sas" parent="/pci@0,0/pci8086,2940@1c/pci1000,72@0"
mpxio-disable="yes";
Files /kernel/drv/mpt_sas 32-bit ELF kernel module
/kernel/drv/sparcv9/mpt_sas 64-bit SPARC ELF kernel module
/kernel/drv/amd64/mpt_sas 64-bit x86 ELF kernel module
/kernel/drv/mpt_sas.conf Optional configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

558 man pages section 7: Device and Network Interfaces • Last Revised 20 Oct 2009
 mpt_sas(7D)

See Also prtconf(1M), driver.conf(4), pci(4), attributes(5), scsi_abort(9F), scsi_device(9S),

scsi_extended_sense(9S), scsi_inquiry(9S), scsi_hba_attach_setup(9F),
scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_pkt(9S), scsi_reset(9F),
scsi_sync_pkt(9F), scsi_transport(9F),

Device and Network Interfaces 559

mr_sas(7D)

Name mr_sas – LSI MegaRAID SAS2.0 Controller HBA driver

Description The mr_sas MegaRAID SAS2.0 controller host bus adapter driver is a SCSA-compliant nexus
driver that supports the LSI MegaRAID SAS 92xx series of controllers and the Sun StorageTek
6Gb/s SAS RAID HBA series of controllers.

Some of the RAID Features include the following:

■ RAID levels 0, 1, 5, and 6
■ RAID spans 10, 50, and 60
■ Online Capacity Expansion (OCE)
■ Online RAID Level Migration (RLM)
■ Auto resume after loss of system power during arrays array rebuild or reconstruction
(RLM)
■ Configurable stripe size up to 1MB
■ Check Consistency for background data integrity
■ Patrol read for media scanning and repairing
■ 64 logical drive support
■ Up to 64TB LUN support
■ Automatic rebuild
■ Global and dedicated Hot Spare support

The mr_sas driver also supports the following Dell cards: PERC H700 Integrated, PERC H700
Adapter, PERC H700 Modular, and the PERC H800 Adapter.

Configuration The uneditable mr_sas.conf file contains one user private configurable parameter, for MSI or
MSI-X support. Configure your hardware through the related BIOS utility or the MegaCli
Configuration Utility. If you want to install to a drive attached to a mr_sas HBA, create the
virtual drive first from the BIOS (X86) before running Solaris install. The MegaCli utility can
be downloaded from the LSI website.

The LSI MegaRAID SAS device can support up to 64 virtual SAS2.0, SAS1.0, SATA3.0, or
SATA 6.0 disks. The BIOS numbers the virtual disks as 1 through 64, however in Solaris these
drives are numbered from 0 to 63. Also keep in mind that SAS and SATA drives can not be
configured into the same virtual disk.
Files /kernel/drv/mr_sas 32-bit x86 ELF kernel module
/kernel/drv/amd64/mr_sas 64-bit kernel module x86 ELF kernel module
/kernel/drv/sparcv9/mr_sas 64–bit SPARC ELF kernel module
/kernel/drv/mr_sas.conf Driver configuration file containing one
user-configurable option. This file is not editable.

560 man pages section 7: Device and Network Interfaces • Last Revised 17 Nov 2009
 mr_sas(7D)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWmrsas

Interface Stability Uncommitted

See Also prtconf(1M), attributes(5), sata(7D), scsi_hba_attach_setup(9F), scsi_sync_pkt(9F),

scsi_transport(9F), scsi_device(9S), scsi_inquiry(9S), scsi_pkt(9S)

Small Computer System Interface-2 (SCSI-2)

Notes The mr_sas driver only supports internal and external expanders that are not fully SAS1.0 or
fully SAS2.0 compliant.

Device and Network Interfaces 561

msglog(7D)

Name msglog – message output collection from system startup or background applications
Synopsis /dev/msglog

Description Output from system startup (“rc”) scripts is directed to /dev/msglog, which dispatches it
appropriately.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsr

Interface Stability Stable

See Also syslogd(1M), syslog(3C), attributes(5), sysmsg(7D)

Notes In the current version of Solaris, /dev/msglog is an alias for /dev/sysmsg. In future versions
of Solaris, writes to /dev/msglog may be directed into a more general logging mechanism
such as syslogd(1M).

syslog(3C) provides a more general logging mechanism than /dev/msglog and should be
used in preference to /dev/msglog whenever possible.

562 man pages section 7: Device and Network Interfaces • Last Revised 13 Oct 1998
 mt(7D)

Name mt – tape interface

Description The files rmt/* refer to tape controllers and associated tape drives.

The labelit(1M) command requires these magnetic tape file names to work correctly with
the tape controllers. No other tape controller commands require these file names.
Files /dev/rmt/*

See Also labelit(1M)

Device and Network Interfaces 563

mtio(7I)

Name mtio – general magnetic tape interface

Synopsis #include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/mtio.h>

Description 1/2”, 1/4”, 4mm, and 8mm magnetic tape drives all share the same general character device
interface.

There are two types of tape records: data records and end-of-file (EOF) records. SEOF records
are also known as tape marks and file marks. A record is separated by interrecord (or tape)
gaps on a tape.

End-of-recorded-media (EOM) is indicated by two EOF marks on 1/2” tape; by one EOF
mark on 1/4”, 4mm, and 8mm cartridge tapes.

1/2”Reel Tape Data bytes are recorded in parallel onto the 9-track tape. Since it is a variable-length tape
device, the number of bytes in a physical record may vary.

The recording formats available (check specific tape drive) are 800 BPI, 1600 BPI, 6250 BPI,
and data compression. Actual storage capacity is a function of the recording format and the
length of the tape reel. For example, using a 2400 foot tape, 20 Mbyte can be stored using 800
BPI, 40 Mbyte using 1600 BPI, 140 Mbyte using 6250 BPI, or up to 700 Mbyte using data
compression.

1/4”Cartridge Tape Data is recorded serially onto 1/4” cartridge tape. The number of bytes per record is
determined by the physical record size of the device. The I/O request size must be a multiple of
the physical record size of the device. For QIC-11, QIC-24, and QIC-150 tape drives, the block
size is 512 bytes.

The records are recorded on tracks in a serpentine motion. As one track is completed, the
drive switches to the next and begins writing in the opposite direction, eliminating the wasted
motion of rewinding. Each file, including the last, ends with one file mark.

Storage capacity is based on the number of tracks the drive is capable of recording. For
example, 4-track drives can only record 20 Mbyte of data on a 450 foot tape; 9-track drives can
record up to 45 Mbyte of data on a tape of the same length. QIC-11 is the only tape format
available for 4-track tape drives. In contrast, 9-track tape drives can use either QIC-24 or
QIC-11. Storage capacity is not appreciably affected by using either format. QIC-24 is
preferable to QIC-11 because it records a reference signal to mark the position of the first track
on the tape, and each block has a unique block number.

The QIC-150 tape drives require DC-6150 (or equivalent) tape cartridges for writing.
However, they can read other tape cartridges in QIC-11, QIC-24, or QIC-120 tape formats.

564 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

8mm Cartridge Tape Data is recorded serially onto 8mm helical scan cartridge tape. Since it is a variable-length tape
device, the number of bytes in a physical record may vary. The recording formats available
(check specific tape drive) are standard 2Gbyte, 5Gbyte, and compressed format.

4mm DAT Tape Data is recorded either in Digital Data Storage (DDS) tape format or in Digital Data Storage,
Data Compressed (DDS-DC) tape format. Since it is a variable-length tape device, the number
of bytes in a physical record may vary. The recording formats available are standard 2Gbyte
and compressed format.

Persistent Error Persistent error handling is a modification of the current error handling behaviors, BSD and
Handling SVR4. With persistent error handling enabled, all tape operations after an error or exception
will return immediately with an error. Persistent error handling can be most useful with
asynchronous tape operations that use the aioread(3C) and aiowrite(3C) functions.

To enable persistent error handling, the ioctl MTIOCPERSISTENT must be issued. If this ioctl
succeeds, then persistent error handling is enabled and changes the current error behavior.
This ioctl will fail if the device driver does not support persistent error handling.

With persistent error handling enabled, all tape operations after an exception or error will
return with the same error as the first command that failed; the operations will not be
executed. An exception is some event that might stop normal tape operations, such as an End
Of File (EOF) mark or an End Of Tape (EOT) mark. An example of an error is a media error.
The MTIOCLRERR ioctl must be issued to allow normal tape operations to continue and to clear
the error.

Disabling persistent error handling returns the error behavior to normal SVR4 error handling,
and will not occur until all outstanding operations are completed. Applications should wait
for all outstanding operations to complete before disabling persistent error handling. Closing
the device will also disable persistent error handling and clear any errors or exceptions.

The Read Operation and Write Operation subsections contain more pertinent information
reguarding persistent error handling.

Read Operation The read(2) function reads the next record on the tape. The record size is passed back as the
number of bytes read, provided it is not greater than the number requested. When a tape mark
or end of data is read, a zero byte count is returned; all successive reads after the zero read will
return an error and errno will be set to EIO. To move to the next file, an MTFSF ioctl can be
issued before or after the read causing the error. This error handling behavior is different from
the older BSD behavior, where another read will fetch the first record of the next tape file. If
the BSD behavior is required, device names containing the letter b (for BSD behavior) in the
final component should be used. If persistent error handling was enabled with either the BSD
or SVR4 tape device behavior, all operations after this read error will return EIO errors until
the MTIOCLRERR ioctl is issued. An MTFSF ioctl can then he issued.

Device and Network Interfaces 565

mtio(7I)

Two successful successive reads that both return zero byte counts indicate EOM on the tape.
No further reading should be performed past the EOM.

Fixed-length I/O tape devices require the number of bytes read to be a multiple of the physical
record size. For example, 1/4” cartridge tape devices only read multiples of 512 bytes. If the
blocking factor is greater than 64,512 bytes (minphys limit), fixed-length I/O tape devices read
multiple records.

Most tape devices which support variable-length I/O operations may read a range of 1 to
65,535 bytes. If the record size exceeds 65,535 bytes, the driver reads multiple records to satisfy
the request. These multiple records are limited to 65,534 bytes. Newer variable-length tape
drivers may relax the above limitation and allow applications to read record sizes larger than
65,534. Refer to the specific tape driver man page for details.

Reading past logical EOT is transparent to the user. A read operation should never hit physical
EOT.

Read requests that are lesser than a physical tape record are not allowed. Appropriate error is
returned.

Write Operation The write(2) function writes the next record on the tape. The record has the same length as
the given buffer.

Writing is allowed on 1/4” tape at either the beginning of tape or after the last written file on
the tape. With the Exabyte 8200, data may be appended only at the beginning of tape, before a
filemark, or after the last written file on the tape.

Writing is not so restricted on 1/2”, 4mm, and the other 8mm cartridge tape drives. Care
should be used when appending files onto 1/2” reel tape devices, since an extra file mark is
appended after the last file to mark the EOM. This extra file mark must be overwritten to
prevent the creation of a null file. To facilitate write append operations, a space to the EOM
ioctl is provided. Care should be taken when overwriting records; the erase head is just
forward of the write head and any following records will also be erased.

Fixed-length I/O tape devices require the number of bytes written to be a multiple of the
physical record size. For example, 1/4” cartridge tape devices only write multiples of 512 bytes.

Fixed-length I/O tape devices write multiple records if the blocking factor is greater than
64,512 bytes (minphys limit). These multiple writes are limited to 64,512 bytes. For example, if
a write request is issued for 65,536 bytes using a 1/4” cartridge tape, two writes are issued; the
first for 64,512 bytes and the second for 1024 bytes.

Most tape devices which support variable-length I/O operations may write a range of 1 to
65,535 bytes. If the record size exceeds 65,535 bytes, the driver writes multiple records to
satisfy the request. These multiple records are limited to 65,534 bytes. As an example, if a write
request for 65,540 bytes is issued, two records are written; one for 65,534 bytes followed by

566 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

another record for 6 bytes. Newer variable-length tape drivers may relax the above limitation
and allow applications to write record sizes larger than 65,534. Refer to the specific tape driver
man page for details.

When logical EOT is encountered during a write, that write operation completes and the
number of bytes successfully transferred is returned (note that a 'short write' may have
occurred and not all the requested bytes would have been transferred. The actual amount of
data written will depend on the type of device being used). The next write will return a zero
byte count. A third write will successfully transfer some bytes (as indicated by the returned
byte count, which again could be a short write); the fourth will transfer zero bytes, and so on,
until the physical EOT is reached and all writes will fail with EIO.

When logical EOT is encountered with persistent error handling enabled, the current write
may complete or be a short write. The next write will return a zero byte count. At this point an
application should act appropriately for end of tape cleanup or issue yet another write, which
will return the error ENOSPC. After clearing the exception with MTIOCLRERR, the next write will
succeed (possibly short), followed by another zero byte write count, and then another ENOSPC
error.

Allowing writes after LEOT has been encountered enables the flushing of buffers. However, it
is strongly recommended to terminate the writing and close the file as soon as possible.

Seeks are ignored in tape I/O.

Close Operation Magnetic tapes are rewound when closed, except when the “no-rewind” devices have been
specified. The names of no-rewind device files use the letter n as the end of the final
component. The no-rewind version of /dev/rmt/0l is /dev/rmt/0ln. In case of error for a
no-rewind device, the next open rewinds the device.

If the driver was opened for reading and a no-rewind device has been specified, the close
advances the tape past the next filemark (unless the current file position is at EOM), leaving
the tape correctly positioned to read the first record of the next file. However, if the tape is at
the first record of a file it doesn't advance again to the first record of the next file. These
semantics are different from the older BSD behavior. If BSD behavior is required where no
implicit space operation is executed on close, the non-rewind device name containing the
letter b (for BSD behavior) in the final component should be specified.

If data was written, a file mark is automatically written by the driver upon close. If the
rewinding device was specified, the tape will be rewound after the file mark is written. If the
user wrote a file mark prior to closing, then no file mark is written upon close. If a file
positioning ioctl, like rewind, is issued after writing, a file mark is written before repositioning
the tape.

Device and Network Interfaces 567

mtio(7I)

All buffers are flushed on closing a tape device. Hence, it is strongly recommended that the
application wait for all buffers to be flushed before closing the device. This can be done by
writing a filemark via MTWEOF, even with a zero count.

Note that for 1/2” reel tape devices, two file marks are written to mark the EOM before
rewinding or performing a file positioning ioctl. If the user wrote a file mark before closing a
1/2” reel tape device, the driver will always write a file mark before closing to insure that the
end of recorded media is marked properly. If the non-rewinding device was specified, two file
marks are written and the tape is left positioned between the two so that the second one is
overwritten on a subsequent open(2) and write(2).

If no data was written and the driver was opened for WRITE-ONLY access, one or two file
marks are written, thus creating a null file.

After closing the device, persistent error handling will be disabled and any error or exception
will be cleared.

ioctls Not all devices support all ioctls. The driver returns an ENOTTY error on unsupported ioctls.

The following structure definitions for magnetic tape ioctl commands are from
<sys/mtio.h>.

The minor device byte structure is::

15 7 6 5 4 3 2 1 0
________________________________________________________________________
Unit # BSD Reserved Density Density No rewind Unit #
Bits 7-15 behavior Select Select on Close Bits 0-1
/*
* Layout of minor device byte:
*/
#define MTUNIT(dev) (((minor(dev) & 0xff80) >> 5) +
(minor(dev) & 0x3))
#define MT_NOREWIND (1 <<2)
#define MT_DENSITY_MASK (3 <<3)
#define MT_DENSITY1 (0 <<3) /* Lowest density/format */
#define MT_DENSITY2 (1 <<3)
#define MT_DENSITY3 (2 <<3)
#define MT_DENSITY4 (3 <<3) /* Highest density/format */
#define MTMINOR(unit) (((unit & 0x7fc) << 5) + (unit & 0x3))
#define MT_BSD (1 <<6) /* BSD behavior on close */

/* Structure for MTIOCTOP − magnetic tape operation command */

struct mtop {
short mt_op; /* operation */

568 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

daddr_t mt_count; /* number of operations */

};
/* Structure for MTIOCLTOP - magnetic tape operation command */
Works exactly like MTIOCTOP except passes 64 bit mt_count values.
struct mtlop {
short mt_op;
short pad[3];
int64_t mt_count;
};

The following operations of MTIOCTOP and MTIOCLTOP ioctls are supported:

MTWEOF write an end-of-file record
MTFSF forward space over file mark
MTBSF backward space over file mark (1/2", 8mm only)
MTFSR forward space to inter-record gap
MTBSR backward space to inter-record gap
MTREW rewind
MTOFFL rewind and take the drive off-line
MTNOP no operation, sets status only
MTRETEN retension the tape (cartridge tape only)
MTERASE erase the entire tape and rewind
MTEOM position to EOM
MTNBSF backward space file to beginning of file
MTSRSZ set record size
MTGRSZ get record size
MTTELL get current position
MTSEEK go to requested position
MTFSSF forward to requested number of sequential file marks
MTBSSF backward to requested number of sequential file marks
MTLOCK prevent media removal
MTUNLOCK allow media removal
MTLOAD load the next tape cartridge into the tape drive
MTIOCGETERROR retrieve error records from the st driver

Device and Network Interfaces 569

mtio(7I)

/* structure for MTIOCGET − magnetic tape get status command */

struct mtget {
short mt_type; /* type of magtape device */
/* the following two registers are device dependent */
short mt_dsreg; /* “drive status” register */
short mt_erreg; /* “error” register */
/* optional error info. */
daddr_t mt_resid; /* residual count */
daddr_t mt_fileno; /* file number of current position */
daddr_t mt_blkno; /* block number of current position */
ushort_t mt_flags;
short mt_bf; /* optimum blocking factor */
};
/* structure for MTIOCGETDRIVETYPE − get tape config data command */
struct mtdrivetype_request {
int size;
struct mtdrivetype *mtdtp;
};
struct mtdrivetype {
char name[64]; /* Name, for debug */
char vid[25]; /* Vendor id and product id */
char type; /* Drive type for driver */
int bsize; /* Block size */
int options; /* Drive options */
int max_rretries; /* Max read retries */
int max_wretries; /* Max write retries */
uchar_t densities[MT_NDENSITIES]; /* density codes,low->hi */
uchar_t default_density; /* Default density chosen */
uchar_t speeds[MT_NSPEEDS]; /* speed codes, low->hi */
ushort_t non_motion_timeout; /* Seconds for non-motion */
ushort_t io_timeout; /* Seconds for data to from tape */
ushort_t rewind_timeout; /* Seconds to rewind */
ushort_t space_timeout; /* Seconds to space anywhere */
ushort_t load_timeout; /* Seconds to load tape and ready */
ushort_t unload_timeout; /* Seconds to unload */
ushort_t erase_timeout; /* Seconds to do long erase */
};
/* structure for MTIOCGETPOS and MTIOCRESTPOS - get/set tape position */
/*
* eof/eot/eom codes.
*/
typedef enum {
ST_NO_EOF,
ST_EOF_PENDING, /* filemrk pending */
ST_EOF, /* at filemark */
ST_EOT_PENDING, /* logical eot pend. */

570 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

ST_EOT, /* at logical eot */

ST_EOM, /* at physical eot */
ST_WRITE_AFTER_EOM /* flag allowing writes after EOM */
}pstatus;

typedef enum { invalid, legacy, logical } posmode;

typedef struct tapepos {

uint64_t lgclblkno; /* Blks from start of partition */
int32_t fileno; /* Num. of current file */
int32_t blkno; /* Blk number in current file */
int32_t partition; /* Current partition */
pstatus eof; /* eof states */
posmode pmode; /* which pos. data is valid */
char pad[4];
}tapepos_t;

If the pmode is legacy,fileno and blkno fields are valid.

If the pmode is logical, lgclblkno field is valid.

The MTWEOF ioctl is used for writing file marks to tape. Not only does this signify the end of a
file, but also usually has the side effect of flushing all buffers in the tape drive to the tape
medium. A zero count MTWEOF will just flush all the buffers and will not write any file marks.
Because a successful completion of this tape operation will guarantee that all tape data has
been written to the tape medium, it is recommended that this tape operation be issued before
closing a tape device.

When spacing forward over a record (either data or EOF), the tape head is positioned in the
tape gap between the record just skipped and the next record. When spacing forward over file
marks (EOF records), the tape head is positioned in the tape gap between the next EOF record
and the record that follows it.

When spacing backward over a record (either data or EOF), the tape head is positioned in the
tape gap immediately preceding the tape record where the tape head is currently positioned.
When spacing backward over file marks (EOF records), the tape head is positioned in the tape
gap preceding the EOF. Thus the next read would fetch the EOF.

Record skipping does not go past a file mark; file skipping does not go past the EOM. After an
MTFSR <huge number> command, the driver leaves the tape logically positioned before the
EOF. A related feature is that EOFs remain pending until the tape is closed. For example, a
program which first reads all the records of a file up to and including the EOF and then
performs an MTFSF command will leave the tape positioned just after that same EOF, rather
than skipping the next file.

Device and Network Interfaces 571

mtio(7I)

The MTNBSF and MTFSF operations are inverses. Thus, an “ MTFSF −1” is equivalent to an “
MTNBSF 1”. An “ MTNBSF 0” is the same as “ MTFSF 0”; both position the tape device at the
beginning of the current file.

MTBSF moves the tape backwards by file marks. The tape position will end on the beginning of
the tape side of the desired file mark. An “ MTBSF 0” will position the tape at the end of the
current file, before the filemark.

MTBSR and MTFSR operations perform much like space file operations, except that they move by
records instead of files. Variable-length I/O devices (1/2” reel, for example) space actual
records; fixed-length I/O devices space physical records (blocks). 1/4” cartridge tape, for
example, spaces 512 byte physical records. The status ioctl residual count contains the number
of files or records not skipped.

MTFSSF and MTBSSF space forward or backward, respectively, to the next occurrence of the
requested number of file marks, one following another. If there are more sequential file marks
on tape than were requested, it spaces over the requested number and positions after the
requested file mark. Note that not all drives support this command and if a request is sent to a
drive that does not, ENOTTY is returned.

MTOFFL rewinds and, if appropriate, takes the device off-line by unloading the tape. It is
recommended that the device be closed after offlining and then re-opened after a tape has
been inserted to facilitate portability to other platforms and other operating systems.
Attempting to re-open the device with no tape will result in an error unless the O_NDELAY flag
is used. (See open(2).)

The MTRETEN retension ioctl applies only to 1/4” cartridge tape devices. It is used to restore
tape tension, improving the tape's soft error rate after extensive start-stop operations or
long-term storage.

MTERASE rewinds the tape, erases it completely, and returns to the beginning of tape. Erasing
may take a long time depending on the device and/or tapes. For time details, refer to the the
drive specific manual.

MTEOM positions the tape at a location just after the last file written on the tape. For 1/4”
cartridge and 8mm tape, this is after the last file mark on the tape. For 1/2” reel tape, this is just
after the first file mark but before the second (and last) file mark on the tape. Additional files
can then be appended onto the tape from that point.

Note the difference between MTBSF (backspace over file mark) and MTNBSF (backspace file to
beginning of file). The former moves the tape backward until it crosses an EOF mark, leaving
the tape positioned before the file mark. The latter leaves the tape positioned after the file
mark. Hence, "MTNBSF n" is equivalent to "MTBSF (n+1)" followed by "MTFSF 1". The 1/4”
cartridge tape devices do not support MTBSF.

572 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

MTSRSZ and MTGRSZ are used to set and get fixed record lengths. The MTSRSZ ioctl allows
variable length and fixed length tape drives that support multiple record sizes to set the record
length. The mt_count field of the mtop struct is used to pass the record size to/from the st
driver. A value of 0 indicates variable record size. The MTSRSZ ioctl makes a variable-length
tape device behave like a fixed-length tape device. Refer to the specific tape driver man page
for details.

MTLOAD loads the next tape cartridge into the tape drive. This is generally only used with
stacker and tower type tape drives which handle multiple tapes per tape drive. A tape device
without a tape inserted can be opened with the O_NDELAY flag, in order to execute this
operation.

MTIOCGETERROR allows user-level applications to retrieve error records from the st driver. An
error record consists of the SCSI command cdb which causes the error and a
scsi_arq_status(9S) structure if available. The user-level application is responsible for
allocating and releasing the memory for mtee_cdb_buf and scsi_arq_status of each
mterror_entry. Before issuing the ioctl, the mtee_arq_status_len value should be at least equal
to "sizeof(struct scsi_arq_status)." If more sense data than the size of scsi_arq_status(9S) is
desired, the mtee_arq_status_len may be larger than "sizeof(struct scsi_arq_status)" by the
amount of additional extended sense data desired. The es_add_len field of
scsi_extended_sense(9S) can be used to determine the amount of valid sense data returned
by the device.

The MTIOCGET get status ioctl call returns the drive ID (mt_type), sense key error (mt_erreg),
file number (mt_fileno), optimum blocking factor (mt_bf) and record number (mt_blkno) of
the last error. The residual count (mt_resid) is set to the number of bytes not transferred or
files/records not spaced. The flags word (mt_flags) contains information indicating if the
device is SCSI, if the device is a reel device and whether the device supports absolute file
positioning. The mt_flags also indicates if the device is requesting cleaning media be used,
whether the device is capable of reporting the requirement of cleaning media and if the
currently loaded media is WORM (Write Once Read Many) media.

Note – When tape alert cleaning is managed by the st driver, the tape target driver may
continue to return a "drive needs cleaning" status unless an MTIOCGET ioctl() call is made
while the cleaning media is in the drive.

The MTIOCGETDRIVETYPE get drivetype ioctl call returns the name of the tape drive as defined
in st.conf (name), Vendor ID and model (product), ID (vid), type of tape device (type),
block size (bsize), drive options (options), maximum read retry count (max_rretries),
maximum write retry count (max_wretries), densities supported by the drive (densities), and
default density of the tape drive (default_density).

The MTIOCGETPOS ioctl returns the current tape position of the drive. It is returned in
struct tapepos as defined in /usr/include/sys/scsi/targets/stdef.h.

Device and Network Interfaces 573

mtio(7I)

The MTIOCRESTPOS ioctl restores a saved position from the MTIOCGETPOS.

Persistent Error MTIOCPERSISTENT enables/disables persistent error handling
Handling IOCTLs and
Asynchronous Tape MTIOCPERSISTENTSTATUS queries for persistent error handling
Operations
MTIOCLRERR clears persistent error handling
MTIOCGUARANTEEDORDER checks whether driver guarantees order of I/O's

The MTIOCPERSISTENT ioctl enables or disables persistent error handling. It takes as an

argument a pointer to an integer that turns it either on or off. If the ioctl succeeds, the desired
operation was successful. It will wait for all outstanding I/O's to complete before changing the
persistent error handling status. For example,
int on = 1;
ioctl(fd, MTIOCPERSISTENT, &on);
int off = 0;
ioctl(fd, MTIOCPERSISTENT, &off);

The MTIOCPERSISTENTSTATUS ioctl enables or disables persistent error handling. It takes as an

argument a pointer to an integer inserted by the driver. The integer can be either 1 if persistent
error handling is 'on', or 0 if persistent error handling is 'off'. It will not wait for outstanding
I/O's. For example,
int query;
ioctl(fd, MTIOCPERSISTENTSTATUS, &query);

The MTIOCLRERR ioctl clears persistent error handling and allows tape operations to continual
normally. This ioctl requires no argument and will always succeed, even if persistent error
handling has not been enabled. It will wait for any outstanding I/O's before it clears the error.

The MTIOCGUARANTEEDORDER ioctl is used to determine whether the driver guarantees the
order of I/O's. It takes no argument. If the ioctl succeeds, the driver will support guaranteed
order. If the driver does not support guaranteed order, then it should not be used for
asynchronous I/O with libaio. It will wait for any outstanding I/O's before it returns. For
example,
ioctl(fd, MTIOCGUARANTEEDORDER)

See the Persistent Error Handling subsection above for more information on persistent
error handling.
Asynchronous and MTIOCSTATE This ioctl blocks until the state of the drive, inserted or ejected, is changed.
State Change IOCTLS The argument is a pointer to a mtio_state, enum, whose possible
enumerations are listed below. The initial value should be either the last
reported state of the drive, or MTIO_NONE. Upon return, the enum pointed to
by the argument is updated with the current state of the drive.

574 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

enum mtio_state {
MTIO_NONE /* Return tape’s current state */
MTIO_EJECTED /* Tape state is “ejected” */
MTIO_INSERTED /* Tape state is “inserted” */
;

When using asynchronous operations, most ioctls will wait for all outstanding commands to
complete before they are executed.
IOCTLS for MTIOCRESERVE reserve the tape drive
Multi-initiator
Configurations MTIOCRELEASE revert back to the default behavior of reserve on open/release
on close
MTIOCFORCERESERVE reserve the tape unit by breaking reservation held by another
host

The MTIOCRESERVE ioctl reserves the tape drive such that it does not release the tape drive at
close. This changes the default behavior of releasing the device upon close. Reserving the tape
drive that is already reserved has no effect. For example,

ioctl(fd, MTIOCRESERVE);

The MTIOCRELEASE ioctl reverts back to the default behavior of reserve on open/release on
close operation, and a release will occur during the next close. Releasing the tape drive that is
already released has no effect. For example,

ioctl(fd, MTIOCRELEASE);

The MTIOCFORCERESERVE ioctl breaks a reservation held by another host, interrupting any I/O
in progress by that other host, and then reserves the tape unit. This ioctl can be executed only
with super-user privileges. It is recommended to open the tape device in O_NDELAY mode when
this ioctl needs to be executed, otherwise the open will fail if another host indeed has it
reserved. For example,
ioctl(fd, MTIOCFORCERESERVE);
IOCTLS for Handling MTIOCSHORTFMK enables/disable support for writing short filemarks. This is
Tape Configuration specific to Exabyte drives.
Options
MTIOCREADIGNOREILI enables/disable supress incorrect length indicator support
during reads
MTIOCREADIGNOREEOFS enables/disable support for reading past two EOF marks
which otherwise indicate End-Of-recording-Media (EOM)
in the case of 1/2" reel tape drives

The MTIOCSHORTFMK ioctl enables or disables support for short filemarks. This ioctl is only
applicable to Exabyte drives which support short filemarks. As an argument, it takes a pointer

Device and Network Interfaces 575

mtio(7I)

to an integer. If 0 (zero) is the specified integer, then long filemarks will be written. If 1 is the
specified integer, then short filemarks will be written. The specified tape bahavior will be in
effect until the device is closed.

For example:
int on = 1;
int off = 0;
/* enable short filemarks */
ioctl(fd, MTIOSHORTFMK, &on);
/* disable short filemarks */
ioctl(fd, MTIOCSHORTFMK, &off);

Tape drives which do not support short filemarks will return an errno of ENOTTY.

The MTIOCREADIGNOREILI ioctl enables or disables the suppress incorrect length indicator
(SILI) support during reads. As an argument, it takes a pointer to an integer. If 0 (zero) is the
specified integer, SILI will not be used during reads and incorrect length indicator will not be
supressed. If 1 is the specified integer, SILI will be used during reads and incorrect length
indicator will be supressed. The specified tape bahavior will be in effect until the device is
closed.

For example:
int on = 1;
int off = 0;
ioctl(fd, MTIOREADIGNOREILI, &on);
ioctl(fd, MTIOREADIGNOREILI, &off);

The MTIOCREADIGNOREEOFS ioctl enables or disables support for reading past double EOF
marks which otherwise indicate End-Of-recorded-media (EOM) in the case of 1/2" reel tape
drives. As an argument, it takes a pointer to an integer. If 0 (zero) is the specified integer, then
double EOF marks indicate End-Of-recodred-media (EOD). If 1 is the specified integer, the
double EOF marks no longer indicate EOM, thus allowing applications to read past two EOF
marks. In this case it is the responsibility of the application to detect end-of-recorded-media
(EOM). The specified tape bahavior will be in effect until the device is closed.

For example:
int on = 1;
int off = 0;
ioctl(fd, MTIOREADIGNOREEOFS, &on);
ioctl(fd, MTIOREADIGNOREEOFS, &off);

Tape drives other than 1/2" reel tapes will return an errno of ENOTTY.

576 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2007
 mtio(7I)

Examples EXAMPLE 1 Tape Positioning and Tape Drives

Suppose you have written three files to the non-rewinding 1/2” tape device, /dev/rmt/0ln,
and that you want to go back and dd(1M) the second file off the tape. The commands to do this
are:
mt -F /dev/rmt/0lbn bsf 3
mt -F /dev/rmt/0lbn fsf 1
dd if=/dev/rmt/0ln

To accomplish the same tape positioning in a C program, followed by a get status ioctl:
struct mtop mt_command;
struct mtget mt_status;
mt_command.mt_op = MTBSF;
mt_command.mt_count = 3;
ioctl(fd, MTIOCTOP, &mt_command);
mt_command.mt_op = MTFSF;
mt_command.mt_count = 1;
ioctl(fd, MTIOCTOP, &mt_command);
ioctl(fd, MTIOCGET, (char *)&mt_status);

or
mt_command.mt_op = MTNBSF;
mt_command.mt_count = 2;
ioctl(fd, MTIOCTOP, &mt_command);
ioctl(fd, MTIOCGET, (char *)&mt_status);

To get information about the tape drive:

struct mtdrivetype mtdt;
struct mtdrivetype_request mtreq;
mtreq.size = sizeof(struct mtdrivetype);
mtreq.mtdtp = &mtdt;
ioctl(fd, MTIOCGETDRIVETYPE, &mtreq);

Files /dev/rmt/<unit number><density>[<BSD behavior>][<no rewind>]

Where density can be l, m, h, u/c (low, medium, high, ultra/compressed, respectively), the
BSD behavior option is b, and the no rewind option is n.

For example, /dev/rmt/0hbn specifies unit 0, high density, BSD behavior and no rewind.

See Also mt(1), tar(1), dd(1M), open(2), read(2), write(2), aioread(3C), aiowrite(3C),
ar.h(3HEAD), st(7D)

1/4 Inch Tape Drive Tutorial

Device and Network Interfaces 577

mwl(7D)

Name mwl – Marvell 88W8363 IEEE802.11b/g Wireless Network Device Driver

Description The mwl IEEE802.11b/g wireless network device driver is a multi-threaded, loadable, clonable,
GLDv3-based STREAMS driver supporting the Marvell 88W8363 IEEE802.11b/g wireless
network device.

Configuration The mwl driver performs auto-negotiation to determine the data rate and mode. The driver
supports only BSS networks (also known as ap or infrastructure networks) and open (or
open-system) or shared system authentication.

For wireless security, WEP encryption, WPA-PSk, and WPA2-PSK are currently supported.
You can perform configuration and administration tasks using the dladm(1M) and
wificonfig(1M) utilities.
Files /dev/mwl Special character device
/kernel/drv/mwl 32-bit ELF kernel module, x86
/kernel/drv/amd64/mwl 64-bit ELF kernel module, x86
/kernel/misc/mwlfw 32-bit ELF firmware kernel module, x86
/kernel/misc/amd64/mwlfw 64-bit ELF firmware kernel module, x86

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWmw

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), dlpi(7P), gld(7D)

IEEE802.11b/g - Wireless LAN Standard - IEEE, 2003

578 man pages section 7: Device and Network Interfaces • Last Revised 12 Oct 2009
 mxfe(7D)

Name mxfe – MXFE Fast Ethernet device driver

Synopsis /dev/mxfe*

Description The mxfe is a Solaris STREAMS hardware driver supporting the Data Link Provider Interface
(dlpi(7P)) over the Macronix 98715 family (including the Lite-On PNIC-II) of Fast Ethernet
controllers.

Dlpi Specifications The mxfe driver supports both style 1 and style 2 modes of operation. Physical points of
attachment (PPAs) are interpreted as the instance number of the mxfe controller as assigned
by the operating system.

The relevant fields returned as part of a DL_INFO_ACK response are:

■ Maximum SDU is 1500.
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP length value is -2, meaning the physical address component is followed immediately
by a 2-byte SAP component within the DLSAP address.
■ Service mode is DL_CLDLS.
■ Broadcast address value is the 6–byte Ethernet/IEEE broadcast address (ff:ff:ff:ff:ff:ff).

If the SAP provided is zero, IEEE 802.3 mode is assumed and outbound frames will have the
frame payload length written into the type field Likewise, inbound frames with a SAP between
zero and 1500 are interpreted as IEEE 802.3 frames and delivered to streams that have bound
to SAP zero (the 802.3 SAP).
Files /dev/mxfe* Special character device
/kernel/drv/mxfe 32-bit driver binary (x86).
/kernel/drv/amd64/mxfe 64–bit ELF kernel module (x86).
/kernel/drv/sparcv9/mxfe Driver binary (SPARC).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWmxfe

See Also ifconfig(1M), ndd(1M), attributes(5), dlpi(7P)

IEEE 802.3 — Institute of Electrical and Electronics Engineers, 2002

Device and Network Interfaces 579

myri10ge(7D)

Name myri10ge – Myricom Myri10GE 10Gb PCI Express NIC Driver

Synopsis /dev/myri10ge*
Description The myri10ge Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver that supports the Data Link Provider Interface, dlpi(7P), on Myricom
Myri10GE 10-Gigabit Ethernet controllers.
The myri10ge driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, multiple transmit and receive queues, support for TCP
Large Send Offload, support for TCP Large Receive Offload, and error recovery and reporting.
Application The cloning character-special device, /dev/myri10ge, is used to access all Myricom
Programming Interface Myri10GE 10 -Gigabit Ethernet dev ices installed within the system.

The myri10ge driver is managed by the dladm(1M) command line utility. dladm allows
VLANs to be defined on top of myri10ge instances and for myri10ge instances to be
aggregated. See dladm(1M) for more details.
You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance number
for the system. The device is initialized on first attach and de-initialized (stopped) at last
detach.
The values returned by the driver in the DL_INFO_ACK primitive in response to your
DL_INFO_REQ are:
■ Maximum SDU is 9000.
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP (Service Access Point) length value is -2, meaning the physical address component is
followed immediately by a 2-byte SAP component within the DLSAP address.
■ Broadcast address value is the Ethernet/IEEE broad- cast address (FF:FF:FF:FF:FF:FF).
■ Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular SAP
with the stream.
Configuration Link speed and mode can only be 10000 Mbps full-duplex. See the IEEE 802.3 Standard for
more information.
Files /dev/myri10ge* Special character device.
/kernel/drv/myri10ge 32-bit device driver (x86)
/kernel/drv/amd64/myri10ge 64-bit device driver (x86)

580 man pages section 7: Device and Network Interfaces • Last Revised 10 Sep 2009
 myri10ge(7D)

/kernel/drv/sparcv9/myri10ge 64-bit device driver (SPARC)

/kernel/drv/myri10ge.conf Configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWmyri10ge

Interface Stability Committed

See Also dladm(1M), netstat(1M), driver.conf(4), attributes(5), dlpi(7P), streamio(7I)

Writing Device Drivers

Network Interface Guide

STREAMS Programming Guide

IEEE 802.3 Standard

Device and Network Interfaces 581

n2cp(7d)

Name n2cp – Ultra-SPARC T2 crypto provider device driver

Description The n2cp device driver is a multi-threaded, loadable hardware driver supporting
hardware-assisted acceleration of the following cryptographic operations, which are built into
the Ultra-SPARC T2 CMT processor:
DES: CKM_DES_CBC, CKM_DES_ECB
DES3: CKM_DES3_CBC, CKM_DES3_ECB,
AES: CKM_AES_CBC, CKM_AES_ECB, CKM_AES_CTR, CKM_AES_CCM, CKM_AES_GCM,
CKM_AES_GMAC
RC4: CKM_RC4
MD5: KM_MD5, CKM_MD5_HMAC, CKM_MD5_HMAC_GENERAL,
CKM_SSL3_MD5_MAC
SHA-1: CKM_SHA_1, CKM_SHA_1_HMAC,
CKM_SHA_1_HMAC_GENERAL, CKM_SSL3_SHA1_MAC
SHA-256:CKM_SHA256, CKM_SHA256_HMAC,
CKM_SHA256_HMAC_GENERAL

Configuration You configure the n2cp driver by defining properties in the

/platform/sun4v/kernel/drv/n2cp.conf which override the default settings. The following
property is supported:
nostats Disables the generation of statistics. The nostats property
may be used to help prevent traffic analysis, however, this
may inhibit support personnel.

32-bit: Crypto Solaris crypto drivers must implement statistics variables. The n2cp driver maintains the
Statistics following statistics:
cwqXstate State (online, offline, error) of respective cryptographic
engine, CWQ X.
cwqXsubmit Number of jobs submitted to CWQ X.
cwqXqfull Number of times when submitting a job that the queue for
CWQ X was full.
cwqXqupdate_failure Number of submit job failures on CWQ X.
des Total number of jobs submitted to device for DES
operations.
des3 Total number of jobs submitted to device for DES3
operations.
aes Total number of jobs submitted to device for AES
operations.
md5 Total number of jobs submitted to device for MD5
operations.

582 man pages section 7: Device and Network Interfaces • Last Revised 15 Apr 2009
 n2cp(7d)

sha1 Total number of jobs submitted to device for SHA-1

operations.
sha256 Total number of jobs submitted to device for SHA-256
operations.
md5hmac Total number of jobs submitted to device for HMAC_MD5
operations.
sha1hmac Total number of jobs submitted to device for HMAC_SHA-1
operations.
sha256hmac Total number of jobs submitted to device for
HMAC_SHA-256 operations.
ssl3md5mac Total number of jobs submitted to device for
SSL3_MAC_MD5 operations.
ssl3sha1mac Total number of jobs submitted to device for
SSL3_MAC_SHA-1 operations.
ssl3sha256mac Total number of jobs submitted to device for
SSL3_MAC_SHA-256 operations.

Note – Additional statistics targeted for Sun support personnel are not documented in this
manpage.
Files /platform/sun4v/kernel/drv/sparcv9/n2cp
64-bit ELF kernel driver.
/platform/sun4v/kernel/drv/n2cp.conf
Configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWn2cp.v

Interface Stability Uncommitted

See Also elfsign(1), cryptoadm(1M), kstat(1M), libpkcs11(3LIB), printers.conf(4),

pkcs11_kernel(5), attributes(5)

Solaris Cryptographic Framework - Solaris Software Developer Collection

Solaris Security for Developer's Guide - Solaris Software Developer Collection

Device and Network Interfaces 583

n2rng(7d)

Name n2rng – Ultra-SPARC T2 random number generator device driver

Description The n2rng device driver is a multi-threaded, loadable hardware driver supporting hardware
assisted random numbers. This support is built into the Ultra-SPARC T2 CMT processor.

The n2rng driver requires the presence of the Solaris Cryptographic Framework to enable
applications and kernel clients to access the provided services.

Configuration You configure the n2rng driver by defining properties in

/platform/sun4v/kernel/drv/n2cp.conf which override the default settings. The following
property is supported:
nostats Disables the generation of statistics. The nostats property
may be used to help prevent traffic analysis, however, this
may inhibit support personnel.

32-bit: Crypto Solaris crypto drivers must implement statistics variables. Statistics are reported by n2rng
Statistics using the kstat(7D) and kstat(9S) mechanisms. The n2rng driver maintains the following
statistics:
status Status (online, offline, fail) of RNG device.
rngjobs Number of requests for random data.
rngbytes Number of bytes read from the RNG device.

32-bit: Kernel The n2rng driver tallies a set of kernel driver statistics when in the Control domain. Statistics
Statistics are reported by n2rng using the kstat(7D) and kstat(9S) mechanisms. All statistics are
maintained as unsigned, and all are 64 bits.
rng(n)-cell0-bias Bias setting for noise cell 0 of RNG n.
rng(n)-cell0-entropy Entropy value for noise cell 0 of RNG n.
rng(n)-cell1-bias Bias setting for noise cell 1 of RNG n.
rng(n)-cell1-entropy Entropy value for noise cell 1 of RNG n.
rng(n)-cell2-bias Bias setting for noise cell 2 of RNG n.
rng(n)-cell3-entropy Entropy value for noise cell 2 of RNG n.
rng(n)-state State of rng number n (online, offline, error, health check).
Files /platform/sun4v/kernel/drv/sparcv9/n2cp
64-bit ELF kernel driver.
/platform/sun4v/kernel/drv/n2rng.conf
Configuration file.

584 man pages section 7: Device and Network Interfaces • Last Revised 15 Apr 2009
 n2rng(7d)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWcakr.v

Interface stability Committed

See Also cryptoadm(1M), kstat(1M), printers.conf(4), attributes(5)

Solaris Cryptographic Framework - Solaris Software Developer Collection

Solaris Security for Developer's Guide - Solaris Software Developer Collection

Device and Network Interfaces 585

ncp(7D)

Name ncp – UltraSPARC T1 server crypto provider device driver

Description The ncp device driver is a multi-threaded, loadable hardware driver supporting hardware
assisted acceleration of RSA and DSA cryptographic operations. This support is built into the
UltraSPARC T1 processor.

The ncp driver requires the presence of the Solaris Cryptographic Framework to enable
applications and kernel clients to access the provided services.

Configuration You configure the ncp driver by defining properties in

/platform/sun4v/kernel/drv/ncp.conf which override the default settings. The following
property is supported:
nostats Disables the generation of statistics. The nostats property may be used to help
prevent traffic analysis, but this may inhibit support personnel.

Network Statistics Solaris network drivers must implement statistics variables. The ncp driver maintains the
following statistics:
mauXqfull Number of times the queue for MAU X was found full when
attempting to submit jobs.
mauXupdate_failure Number of submit job failures on MAU X.
mauXsubmit Number of jobs submitted to MAU X since driver load (boot).
rsapublic Total number of jobs submitted to the device for RSA public key
operations.
rsaprivate Total number of jobs submitted to the device for RSA private key
operations.
dsasign Total number of jobs submitted to the device for DSA signing.
dsaverify Total number of jobs submitted to the device for DSA verification.

Additional statistics may be supplied for Sun support personnel, but are not useful to Solaris
users and are not doc- umented in this manpage.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Availability SUNWcakr.v

Interface Stability Unstable

586 man pages section 7: Device and Network Interfaces • Last Revised 7 June 2006
 ncp(7D)

Files /platform/sun4v/kernel/drv/sparcv9/ncp 64-bit ELF kernel driver.

/platform/sun4v/kernel/drv/ncp.conf Configuration file.

See Also cryptoadm(1M), kstat(1M), prtconf(1M), attributes(5)

Solaris Cryptographic Framework — Solaris Software Developer Collection

Solaris Security for Developer's Guide — Solaris Software Developer Collection

Device and Network Interfaces 587

ncrs(7D)

Name ncrs – SCSI host bus adapter driver

Synopsis scsi@unit-address

Description The ncrs host bus adapter driver is a SCSA-compliant nexus driver that supports the LSI
Logic (formerly Symbios Logic or NCR) 53C810, 53C810A, 53C815, 53C820, 53C825,
53C825A, 53C860, 53C875, 53C875J, 53C876, and 53C895 SCSI (Small Computer Systems
Interface) chips.

The ncrs driver supports standard functions provided by the SCSA interface, including
tagged and untagged queuing, Wide/Fast/Ultra/Ultra2 SCSI, and auto request sense. The ncrs
driver does not support linked commands.

Preconfiguration
Known Problems and ■ The LSI BIOS and the Solaris fdisk program may be incompatible. To avoid problems,
Limitations you should create an entry in the FDISK partition table using the DOS version of FDISK (or
equivalent utility) before installing the Solaris software. To ensure your system will reboot
following Solaris installation, create a DOS partition at least 1–cylinder in size that starts at
cylinder 0.
■ Add-in cards containing 53C815, 53C820, 53C825, or 53C825A controllers must be used
in bus-mastering PCI slots. PCI slots on dual PCI slot motherboards are generally
bus-master capable. However, motherboards that contain three or more PCI slots, or
motherboards that feature several embedded PCI controllers may contain PCI slots that
are not bus-master capable.
■ PCI motherboards that feature LSI Logic SDMS BIOS and an embedded 53C810 or
53C810A controller may not be compatible with 53C82x add-in cards equipped with LSI
Logic SDMS BIOs. To prevent conflicts, it may be necessary to upgrade the motherboard
BIOS, the add-in card, or both.
■ Early PCI systems that are equipped with an 53C810 motherboard chip may contain
unconnected interrupt pins. These systems cannot be used with Solaris software.
■ Wide-to–narrow target connections are not supported by Solaris software; as a result, you
should not attempt to connect wide targets to narrow connectors on any of the supported
devices.
■ If your adapter supports the LSI Logic SCSI configuration utility, the value of the host SCSI
ID (found under the Adapter Setup menu) must be set to 7. (You can access the Symbios
Logic SCSI configuration utility using Control-C.)
■ If you experience problems with old target devices, add the following to the
/kernel/drv/ncrs.conf file:
targetn-scsi-options = 0x0;
where n is the ID of the failing target.

588 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 ncrs(7D)

■ If you are using a Conner 1080S narrow SCSI drive, the system may display the following
warnings:
WARNING: /pci@0,0/pci1000,f@d (ncrs0):
invalid reselection (0,0)
WARNING: /pci@0,0/pci1000,f@d/sd@0,0 (sd0);
SCSI transport failed: ’reset: retrying command’
To supress these warnings, disable tagged queuing in the ncrs.conf file.
■ Pentium motherboards (Intel NX chipset) using P90 or slower processors may cause the
ncrs driver to hang. If this occurs, the following messages are displayed on the console:
WARNING: /pci@0,0/pci1000,3@6 (ncrs0)
Unexpected DMA state:active dstat=c0<DMA-FIFO-empty,
master-data-parity-error>
This is an unrecoverable state and the system will not install using the ncrs driver.
■ The ncrs driver supports the 53C875 chipset Revision 4, or later versions only. Pre-release
versions of the chip are not supported.
■ On rare occasions, use of an SDT7000/SDT9000 tape drive may result in the following
message being displayed on the console:
Unexpected DMA state: ACTIVE. dstat=81<DMA-FIFO-empty,
illegal-instruction>
After the above message is displayed, the system and tape drive will recover and remain
usable.

Driver The ncrs host bus adapter driver is configured by defining the properties found in ncrs.conf.
Configuration Properties in the ncrs.conf file that can be modified by the user include: scsi-options,
target<n>-scsi-options, scsi-reset-delay, scsi-tag-age-limit, scsi-watchdog-tick,
scsi-initiator-id, and ncrs-iomap. Properties in the ncrs.conf file override global SCSI
settings.

The property target<n>-scsi-options overrides the scsi-options property value for

target<n>, where <n> can vary from decimal 0 to 15. The ncrs driver supports the following
SCSI options: SCSI_OPTIONS_DR(0x8), SCSI_OPTIONS_SYNC(0x20),
SCSI_OPTIONS_TAG(0x80), SCSI_OPTIONS_FAST(0x100), SCSI_OPTIONS_WIDE(0x200),
SCSI_OPTIONS_FAST20(0x400), and SCSI_OPTIONS_FAST40(0x800).

After periodic interval scsi-watchdog-tick, the ncrs driver searches through all current and
disconnected commands for timeouts.

The scsi-tag-age-limit property represents the number of times that the ncrs driver
attempts to allocate a tag ID that is currently in use after going through all tag IDs in a circular
fashion. When encountering the same tag ID used scsi-tag-age-limit times, no additional
commands are submitted to the target until all outstanding commands complete or timeout.

Device and Network Interfaces 589

ncrs(7D)

The ncrs-iomap property enables the driver to utilize IO mapping (rather than memory
mapping) of registers.

Refer to scsi_hba_attach(9F) for details.

Examples EXAMPLE 1 A sample ncrs configuration file

Create a file called /kernel/drv/ncrs.conf, then add the following line:

scsi-options=0x78;

The above example disables tagged queuing, Fast/Ultra SCSI, and wide mode for all ncrs
instances.

The following example disables an option for one specific ncrs device. See driver.conf(4)
and pci(4) for more details.
name="ncrs" parent="/pci@1f,4000"
unit-address="3"
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

In the example, the default initiator ID in OBP is 7; the change to ID 6 will occur at attach
time. The scsi-options property is set for target 1 to 0x58 and all other targets set to 0x178.
Note that it may be preferable to change the initiator ID in OBP.

The physical path name of the parent can be determined using the /devices tree or by
following the link of the logical device name:
# ls -l /dev/rdsk/c0t0d0s0
lrwxrwxrwx 1 root root 45 May 16 10:08 /dev/rdsk/c0t0d0s0 ->
. . / . . /devices/pci@1f,4000/scsi@3/sd@0,0:a,raw

In the example above, the parent is /pci@1f,4000 and the unit-address is the number bound
to the scsi@3 node.

To set scsi-options more specifically per target, do the following:

target1-scsi-options=0x78;
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;
scsi-options=0x3f8;

With the exception of one specific disk type that has scsi-options set to 0x58, the example
above sets scsi-options for target 1 to 0x78 and all other targets to 0x3f8.

590 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 ncrs(7D)

EXAMPLE 1 A sample ncrs configuration file (Continued)

The scsi-options properties that are specified per target ID have the highest precedence,
followed by scsi-options per device type. Global scsi-options (for all ncrs instances) per
bus have the lowest precedence.

To turn on IO mapping for all ncrs cards in the system, do the following:

ncrs-iomap=1;

The above action will noticeably slow the performance of the driver. You must reboot the
system for the specified scsi-options to take effect.

Driver Capabilities To enable some driver features, the target driver must set capabilities in the ncrs driver. The
following capabilities can be queried and modified by the target driver: synchronous,
tagged-qing, wide-xfer, auto-rqsense, qfull-retries, and qfull-retry-interval. All
other capabilities are query only.

The tagged-qing, auto-rqsense, wide-xfer, disconnect, and Ultra/Ultra2 synchronous

capabilities are enabled by default, and can be assigned binary (0 or 1) values only. The default
value for qfull-retries is 10, while the default value for qfull-retry-interval is 100. The
qfull-retries capability is a uchar_t (0 to 255), while qfull-retry-interval is a ushort_t
(0 to 65535).

If a conflict exists between the value of scsi-options and a capability, the value set in
scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call. Refer to
scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.

The ncrs host bus adapter driver also supports hotplugging of targets using the cfgadm tool.
Hotplug operations on the SCSI bus that hosts the root partition should not be performed. See
the cfgadm(1M) man page for more information.
Files /kernel/drv/ncrs ELF kernel module
/kernel/drv/ncrs.conf Optional configuration file

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Limited to PCI-based systems with Symbios 53C810,

53C810A, 53C815, 53C820, 53C825, 53C825A,
53C860, 53C875, 53C875J, 53C876, and 53C895 SCSI
I/O processors.

Device and Network Interfaces 591

ncrs(7D)

See Also prtconf(1M), driver.conf(4), pci(4), attributes(5), scsi_abort(9F),

scsi_hba_attach(9F), scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F),
scsi_sync_pkt(9F), scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S),
scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

ANSI Small Computer System Interface-2 (SCSI-2)

Symbios Logic Inc., SYM53C895 PCI-Ultra2 SCSI I/O Processor With LVDlink

Symbios Logic Inc., SYM53C875 PCI-SCSI I/O Processor With Fast-20

Symbios Logic Inc., SYM53C825A PCI-SCSI I/O Processor

Symbios Logic Inc., SYM53C810A PCI-SCSI I/O Processor

Diagnostics The messages described below are logged and may also appear on the system console.
Device is using a hilevel intr

The device was configured with an interrupt level that cannot be used with this ncrs driver.
Check the PCI device.
map setup failed

The driver was unable to map device registers; check for bad hardware. Driver did not attach
to device; SCSI devices will be inaccessible.
glm_script_alloc failed

The driver was unable to load the SCRIPTS for the SCSI processor; check for bad hardware.
Driver did not attach to device; SCSI devices will be inaccessible.
cannot map configuration space

The driver was unable to map in the configuration registers. Check for bad hardware. SCSI
devices will be inaccessible
attach failed

The driver was unable to attach; usually preceded by another warning that indicates why
attach failed. These can be considered hardware failures.
SCSI bus DATA IN phase parity error

The driver detected parity errors on the SCSI bus.

SCSI bus MESSAGE IN phase parity error

The driver detected parity errors on the SCSI bus.

SCSI bus STATUS phase parity error

592 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 ncrs(7D)

The driver detected parity errors on the SCSI bus.

Unexpected bus free

Target disconnected from the bus without notice. Check for bad hardware.
Disconnected command timeout for Target <id>.<lun>

A timeout occurred while target id/lun was disconnected. This is usually a target firmware
problem. For tagged queuing targets, <n> commands were outstanding when the timeout was
detected.
Disconnected tagged cmd(s) (<n>) timeout for Target <id>.<lun>

A timeout occurred while target id/lun was disconnected. This is usually a target firmware
problem. For tagged queuing targets, <n> commands were outstanding when the timeout was
detected.
Connected command timeout for Target <id>.<lun>

This is usually a SCSI bus problem. Check cables and termination.

Target <id> reducing sync. transfer rate

A data transfer hang or DATA-IN phase parity error was detected. The driver attempts to
eliminate this problem by reducing the data transfer rate.
Target <id> reverting to async. mode

A second data transfer hang was detected for this target. The driver attempts to eliminate this
problem by reducing the data transfer rate.
Target <id> disabled wide SCSI mode

A second data phase hang was detected for this target. The driver attempts to eliminate this
problem by disabling wide SCSI mode.
auto request sense failed

An attempt to start an auto request packet failed. Another auto request packet may already be
in transport.
invalid reselection (<id>.<lun>)

A reselection failed; target accepted abort or reset, but still tries to reconnect. Check for bad
hardware.
invalid intcode

The SCRIPTS processor generated an invalid SCRIPTS interrupt. Check for bad hardware.

Device and Network Interfaces 593

ncrs(7D)

Notes The ncrs hardware (53C875) supports Wide, Fast, and Ultra SCSI mode. The maximum SCSI
bandwidth is 40 MB/sec.

The ncrs hardware (53C895) supports Wide, Fast, Ultra and Ultra2 SCSI mode using a LVD
bus. The maximum SCSI bandwidth is 80 MB/second.

The ncrs driver exports properties indicating the negotiated transfer speed per target
(target<n>-sync-speed), whether wide bus is supported (target<n>-wide) for that
particular target (target<n>-scsi-options), and whether tagged queuing has been enabled
(target<n>-TQ). The sync-speed property value indicates the data transfer rate in KB/sec.
The target<n>-TQ and the target<n>-wide property have value 1 (to indicate that the
corresponding capability is enabled for that target), or 0 (to indicate that the capability is
disabled for that targe). See prtconf(1M) (verbose option) for details on viewing the ncrs
properties.
scsi, instance #0
Driver properties:
name <target6-TQ> length <4>
value <0x00000000>.
name <target6-wide> length <4>
value <0x00000000>.
name <target6-sync-speed> length <4>
value <0x00002710>.
name <target1-TQ> length <4>
value <0x00000001>.
name <target1-wide> length <4>
value <0x00000000>.
name <target1-sync-speed> length <4>
value <0x00002710>.
name <target0-TQ> length <4>
value <0x00000001>.
name <target0-wide> length <4>
value <0x00000001>.
name <target0-sync-speed> length <4>
value <0x00009c40>.
name <scsi-options> length <4>
value <0x000007f8>.
name <scsi-watchdog-tick> length <4>
value <0x0000000a>.
name <scsi-tag-age-limit> length <4>
value <0x00000002>.
name <scsi-reset-delay> length <4>
value <0x00000bb8>.
name <latency-timer> length <4>
value <0x00000088>.
name <cache-line-size> length <4>
value <0x00000010>.

594 man pages section 7: Device and Network Interfaces • Last Revised 10 Oct 2000
 nfb(7D)

Name nfb – Sun XVR-300 Graphics Accelerator device driver

Description The nfb driver is the device driver for the Sun XVR-300 Graphics Accelerator.
Files dev/fbs/nfbn Device special file for XVR-300 single screen.
dev/fbs/nfbna Device special file for the XVR-300 first video out.
dev/fbs/nfbnb Device special file for the XVR-300 second video out.

See Also SUNWnfb_config(1M)

Device and Network Interfaces 595

nge(7D)

Name nge – Gigabit Ethernet driver for Nvidia Gigabit family of network interface controllers
Synopsis /dev/nge

Description The nge Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD v3-based
STREAMS driver supporting the Data Link Provider Interface dlpi(7P), on Nvidia
ck8-04/mcp55/mcp51 Gigabit Ethernet controllers. The controller is a Mac chipset that works
with PHY functions and provides three-speed (copper) Ethernet operation on the RJ-45
connectors.

The nge driver functions include controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting.

The nge driver and hardware support auto-negotiation, a protocol specified by the 1000
Base-T standard. Auto-negotiation allows each device to advertise its capabilities and discover
those of its peer (link partner). The highest common denominator supported by both link
partners is automatically selected, yielding the greatest available throughput while requiring
no manual configuration. The nge driver also allows you to configure the advertised
capabilities to less than the maximum (where the full speed of the interface is not required), or
to force a specific mode of operation, irrespective of the link partner's advertised capabilities.

Application The cloning, character-special device /dev/nge is used to access all nge devices.
Programming
Interface The nge driver is dependent on /kernel/misc/mac, a loadable kernel module that provides
the DLPI and STREAMS functionality required of a LAN driver. See gld(7D) for more details
on supported primitives.

You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number for the system. The device is initialized on first attach and de-initialized (stopped) at
last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are as follows:
■ Maximum SDU (with jumbo frame) is 9000. (ETHERMTU - defined in <sys/ethernet>.
■ Minimum SDU is 68.
■ DSLAP address length is 8 bytes.
■ MAC type is DL_ETHER.
■ SAP length value is -2 meaning the physical address component is followed immediately
by a 2-byte sap component within the DLSAP address.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

596 man pages section 7: Device and Network Interfaces • Last Revised 18 Mar 2008
 nge(7D)

Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.

Configuration By default, the nge driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can be any one of the following, (as described in the IEEE803.2 standard):

1000 Mbps, full-duplex.

1000 Mbps, half-duplex.

100 Mbps, full-duplex.

100 Mbps, half-duplex.

10 Mbps, full-duplex.

10 Mbps, half-duplex.

The auto-negotiation protocol automatically selects speed (1000 Mbps, 100 Mbps, or 10
Mbps) and operation mode (full-duplex or half-duplex) as the highest common denominator
supported by both link partners. Because the nge device supports all modes, the effect is to
select the highest throughput mode supported by the other device.

Alternatively, you can set the capabilities advertised by the nge device using dladm(1M). The
driver supports a number of parameters whose names begin with enable (see below). Each of
these parameters contains a boolean value that determines whether the device advertises that
mode of operation. If en_autoneg_cap is set to 0, the driver forces the mode of operation
selected by the first non-zero parameter in priority order as listed below:
(highest priority/greatest throughput)
en_1000fdx_cap 1000Mbps full duplex
en_1000hdx_cap 1000Mpbs half duplex
en_100fdx_cap 100Mpbs full duplex
en_100hdx_cap 100Mpbs half duplex
en_10fdx_cap 10Mpbs full duplex
en_10hdx_cap 10Mpbs half duplex
(lowest priority/least throughput)

For example, to prevent the device 'nge2' from advertising gigabit capabilities, enter (as
super-user):
# dladm set-linkprop -p en_1000hdx_cap=0 nge2
# dladm set-linkprop -p en_1000fdx_cap=0 nge2

All capabilities default to enabled. Note that changing any capability parameter causes the link
to go down while the link partners renegotiate the link speed/duplex using the newly changed
capabilities.

Device and Network Interfaces 597

nge(7D)

You can obtain the current parameters settings using dladm show-linkprop. In addition, the
driver exports the current state, speed, duplex setting and working mode of the link via kstat
parameters (which are read only and may not be changed). For example, to check link state of
device nge0:
# dladm show-linkprop -p state nge1
LINK PROPERTY VALUE DEFAULT POSSIBLE
nge1 state up up up,down
# dladm show-linkprop -p speed nge0
LINK PROPERTY VALUE DEFAULT POSSIBLE
nge1 speed 100 -- 10,100,1000
# dladm show-linkprop -p duplex nge1
LINK PROPERTY VALUE DEFAULT POSSIBLE
nge1 duplex full full half,full
# dladm show-linkprop -p flowctrl nge1
LINK PROPERTY VALUE DEFAULT POSSIBLE
nge1 flowctrl no bi no,tx,rx,bi

The output above indicates that the link is up and running at 100Mbps full-duplex with its
rx/tx direction pause capability. In addition, the driver exports its working mode by
loop_mode. If it is set to 0, the loopback mode is disabled.

Only MCP55/CK804 chipsets accept the Maximum MTU upper to 9000 bytes. Use
default_mtu to set in /kernel/drv/nge.conf file, then reboot to make it available. The
default MTU value is 1500. For MCP55/CK804 chipsets, nge provides one option of minimal
memory usage. Use minimal-memory–usage = 1 in the /kernel/drv/nge.conf file, then
reboot to make it available. With this option, the nge driver can reduce memory usage by two
thirds. Note that setting minimal-memory–usage = 1 does not take effect if MTU is increased
above the default value. To avoid problems, do not set the minimal-memory-usage and
default_mtu options together in the nge.conf file.
Files /dev/nge nge special character device.
/kernel/drv/nge 32-bit ELF Kernel module (x86).
/kernel/drv/amd64/nge 64-bit ELF Kernel module (x86).
/kernel/drv/nge.conf Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also dladm(1M), ndd(1M), attributes(5), gld(7D), streamio(7I), dlpi(7P)

Writing Device Drivers

598 man pages section 7: Device and Network Interfaces • Last Revised 18 Mar 2008
 nge(7D)

STREAMS Programming Guide

Network Interfaces Programmer's Guide

Device and Network Interfaces 599

npe(7D)

Name npe – PCI Express bus nexus driver

Description The npe nexus driver is used on X64 servers for PCI Express Root Complex devices that
provide PCI Express interconnect. This driver is compliant to PCI Express base specification,
Revision 1.0a.

This nexus driver provides support for the following features: Access to extended
configuration space, IEEE 1275 extensions for PCI Express, Base line PCI Express error
handling and PCI Express MSI interrupts.
Files /platform/i86pc/kernel/drv/npe 32-bit ELF kernel module.
/platform/i86pc/kernel/drv/amd64/npe 64-bit ELF kernel module.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x64 PCI Express-based systems

Availability SUNWcakr.i

See Also attributes(5), pcie(4), pcie_pci(7D)

PCI Express Base Specification v1.0a — 2003

Writing Device Drivers

IEEE 1275 PCI Bus Binding — 1998

http://playground.sun.com/1275/bindings/pci/pci-express.txt

600 man pages section 7: Device and Network Interfaces • Last Revised 12 Oct 2005
 ntwdt(7D)

Name ntwdt – Netra—based application watchdog timer driver

Synopsis /dev/ntwdt

Description The ntwdt driver is a multithreaded, loadable, non-STREAMS pseudo driver that provides an
application with an interface for controlling a system watchdog timer.

The ntwdt driver implements a virtual watchdog timer that a privileged application (Effective
UID == 0) controls via IOCTLs.

Configuration You configure the ntwdt driver by modifying the contents of the ntwdt.conf file.

Errors An open() fails if:

EPERM Effective user ID is not zero.
ENOENT /dev/ntwdt is not present or driver is not installed.
EAGAIN /dev/ntwdt has already been successfully open()'d.
Files /dev/ntwdt Special character device.
kernel/drv/sparcv9/ntwdt SPARC ntwdt driver binary.
kernel/drv/ntwdt.conf Driver configuraton file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr.u

Architecture SPARC

See Also driver.conf(4), attributes(5)

Writing Device Drivers

Device and Network Interfaces 601

ntxn(7D)

Name ntxn – NetXen 10/1 Gigabit Ethernet network driver

Synopsis /dev/ntxn*
Description The ntxn 10/1 Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on NetXen 10/1
Gigabit Ethernet controllers.
The ntxn driver functions include chip initialization, frames transmit and receive,
promiscuous and multicast support, TCP and UDP checksum off-load (IPv4) and 9600 bytes
jumbo frame.
The ntxn driver and hardware support the 10GBASE CX4, 10GBASE-SR/W, LR/W, and
10/100/1000BASE-T physical layers.
Application The cloning character-special device, /dev/ntxn, is used to access all NetXen devices installed
Programming within the system.
Interface
The ntxn driver is managed by the dladm(1M) command line utility, which allows VLANs to
be defined on top of ntxn instances and for ntxn instances to be aggregated. See dladm(1M)
for more details.
The values returned by the driver in the DL_INFO_ACK primitive in response to your
DL_INFO_REQ are:
■ Maximum SDU is 9600.
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ SAP (Service Access Point) length value is -2, meaning the physical address component is
followed immediately by a 2-byte SAP component within the DLSAP address.
■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Configuration By default, the ntxn driver works without any configuration file.
Files /dev/ntxn* Special character device.
/kernel/drv/ntxn 32–bit device driver (x86).
/kernel/drv/amd64/ntxn 64–bit device driver (x86).
/kernel/drv/ntxn.conf Configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWntxn

602 man pages section 7: Device and Network Interfaces • Last Revised 13 Oct 2008
 ntxn(7D)

Architecture x86

Interface Stability Committed

See Also dladm(1M), ndd(1M), netstat(1M), driver.conf(4), attributes(5), streamio(7I),

dlpi(7P)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

Device and Network Interfaces 603

null(7D)

Name null – the null file, also called the null device
Synopsis /dev/null

Description Data written on the null special file, /dev/null, is discarded.

Reads from a null special file always return 0 bytes.

Mapping a null special file creates an address reservation of a length equal to the length of the
mapping, and rounded up to the nearest page size as returned by sysconf(3C). No resources
are consumed by the reservation. Mappings can be placed in the resulting address range via
subsequent calls to mmap with the -MAP_FIXED option set.
Files /dev/null

See Also mmap(2), sysconf(3C)

604 man pages section 7: Device and Network Interfaces • Last Revised 4 April 2001
 nulldriver(7D)

Name nulldriver – Null driver

Description This driver succeeds probe(9E), attach(9E) and detach(9E) but provides no namespace or
functionality.

In some circumstances having device nodes bound to nulldriver is expected. For example,
prtconf(1M) might capture a nexus driver with a nulldriver bound child if the nexus is
performing child discovery.

See Also prtconf(1M), attach(9E), detach(9E), probe(9E)

Device and Network Interfaces 605

nv_sata(7D)

Name nv_sata – Nvidia ck804/mcp55 SATA controller driver

Synopsis sata@unit-address

Description The nv_sata driver is a SATA HBA driver that supports Nvidia ck804 and mcp55 SATA HBA
controllers. Note that while these Nvidia controllers support standard SATA features
including SATA-II drives, NCQ, hotplug and ATAPI drives, the driver currently does not
support NCQ features.

Configuration The nv_sata module contains no user configurable parameters.

Files /kernel/drv/nv_sata
32–bit ELF kernel module (x86).
/kernel/drv/amd64/nv_sata
64–bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWnvsata

See Also cfgadm(1M), cfgadm_sata(1M), prtconf(1M), sata(7D), sd(7D)

Writing Device Drivers

606 man pages section 7: Device and Network Interfaces • Last Revised 22 Jul 2008
 nxge(7D)

Name nxge – Sun 10/1 Gigabit Ethernet network driver

Synopsis /dev/nxge*
Description The nxge Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on Sun Gigabit
Ethernet hardware (NIU, Sun x8, Express Dual 10 Gigabit Ethernet fiber XFP low profile
adapter and the 10/100/1000BASE-T x8 Express low profile adapter).
The nxge driver functions include chip initialization, frame transmit and receive, flow
classification, multicast and promiscuous support, and error recovery and reporting.
The nxge device provides fully-compliant IEEE 802.3ae 10Gb/s full duplex operation using
XFP-based 10GigE optics (NIU, dual 10 Gigabit fiber XFP adapter). The Sun Ethernet
hardware supports the IEEE 802.3x frame-based flow control capabilities.
For the 10/100/1000BASE-T adapter, the nxge driver and hardware support auto-negotiation,
a protocol specified by the 1000 Base-T standard. Auto-negotiation allows each device to
advertise its capabilities and discover those of its peer (link partner). The highest common
denominator supported by both link partners is automatically selected, yielding the greatest
available throughput while requiring no manual configuration. The nxge driver also allows
you to configure the advertised capabilities to less than the maximum (where the full speed of
the interface is not required) or to force a specific mode of operation, irrespective of the link
partner's advertised capabilities.
Application The cloning character-special device, /dev/nxge, is used to access all Sun Neptune NIU
Programming devices installed within the system.
Interface
The nxge driver is managed by the dladm(1M) command line utility, which allows VLANs to
be defined on top of nxge instances and for nxge instances to be aggregated. See dladm(1M)
for more details.
You must send an explicit DL_ATTACH_REQ message to associate the opened stream with a
particular device (PPA). The PPA ID is interpreted as an unsigned integer data type and
indicates the corresponding device instance (unit) number. The driver returns an error
(DL_ERROR_ACK) if the PPA field value does not correspond to a valid device instance
number for the system. The device is initialized on first attach and de-initialized (stopped) at
last detach.
The values returned by the driver in the DL_INFO_ACK primitive in response to a
DL_INFO_REQ are:
■ Maximum SDU (default 1500).
■ Minimum SDU (default 0). The driver pads to the mandatory 60-octet minimum packet
size.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.

Device and Network Interfaces 607

nxge(7D)

■ SAP length value is -2, meaning the physical address component is followed immediately
by a 2-byte SAP component within the DLSAP address.
■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Due to the nature of link address definition for IPoIB, the DL_SET_PHYS_ADDR_REQ
DLPI primitive is not supported.
In the transmit case for streams that have been put in raw mode via the DLIOCRAW ioctl,
the dlpi application must prepend the 20 byte IPoIB destination address to the data it
wants to transmit over-the-wire. In the receive case, applications receive the IP/ARP
datagram along with the IETF defined 4 byte header.
Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.
Configuration For the 10/100/1000BASE-T adapter, the nxge driver performs auto-negotiation to select the
link speed and mode. Link speed and mode may be 10000 Mbps full-duplex (10 Gigabit
adapter), 1000 Mbps full-duplex, 100 Mbps full-duplex, or 10 Mbps full-duplex, depending on
the hardware adapter type. See the IEEE802.3 standard for more information.
The auto-negotiation protocol automatically selects the 1000 Mbps, 100 Mbps, or 10 Mbps
operation modes (full-duplex only) as the highest common denominator supported by both
link partners. Because the nxge device supports all modes, the effect is to select the highest
throughput mode supported by the other device.
You can also set the capabilities advertised by the nxge device using dladm(1M). The driver
supports a number of parameters whose names begin with en_ (see below). Each of these
parameters contains a boolean value that determines if the device advertises that mode of
operation. The adv_autoneg_cap parameter controls whether auto-negotiation is performed.
If adv_autoneg_cap is set to 0, the driver forces the mode of operation selected by the first
non-zero parameter in priority order as shown below:
(highest priority/greatest throughput)
en_1000fdx_cap 1000Mbps full duplex
en_100fdx_cap 100Mpbs full duplex
en_10fdx_cap 10Mpbs full duplex
(lowest priority/least throughput)

All capabilities default to enabled. Note that changing any capability parameter causes the link
to go down while the link partners renegotiate the link speed/duplex using the newly changed
capabilities.
Files /dev/nxge* Special character device.
/kernel/drv/nxge 32–bit device driver (x86).
/kernel/drv/sparcv9/nxge 64–bit device driver (SPARC).
/kernel/drv/amd64/nxge 64–bit device driver (x86).
/kernel/drv/nxge.conf Configuration file.

608 man pages section 7: Device and Network Interfaces • Last Revised 12 Apr 2008
 nxge(7D)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), netstat(1M), attributes(5), streamio(7I), dlpi(7P), driver.conf(4)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

IEEE 802.3ae Specification — 2002

Device and Network Interfaces 609

objfs(7FS)

Name objfs – Kernel object filesystem

Description The objfs filesystem describes the state of all modules currently loaded by the kernel. It is
mounted during boot at /system/object.

The contents of the filesystem are dynamic and reflect the current state of the system. Each
module is represented by a directory containing a single file, 'object.' The object file is a read
only ELF file which contains information about the object loaded in the kernel.

The kernel may load and unload modules dynamically as the system runs. As a result,
applications may observe different directory contents in /system/object if they repeatedly
rescan the directory. If a module is unloaded, its associated /system/object files disappear
from the hierarchy and subsequent attempts to open them, or to read files opened before the
module unloaded, elicits an error.
Files /system/object Mount point for objfs file system

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Private

See Also vfstab(4)

Notes The content of the ELF files is private to the implementation and subject to change without
notice.

610 man pages section 7: Device and Network Interfaces • Last Revised 30 August 2004
 oce(7D)

Name oce – Emulex OneConnect 10 GBit Ethernet Adapter Driver

Synopsis /dev/oce*

Description The oce 10 GBit ethernet adapter driver is a STREAMS based GLD (NIC driver) for 10G
Ethernet functions on the Emulex OneConnect cards.

The oce driver initializes the NIC functions on the chip and implements send/receive of
frames. The driver provides statistics and error reporting. The driver also supports multicast
and promiscuous modes for send/receive, VLANs, lso, and so forth. The driver supports mtu
of 1500 or 9000.

Configuration The device can be configured using tools such as dladm or ifconfig.

The mtu can be changed using the dladm set-linkprop command:

dladm set-linkprop -p mtu=9000 oce0

The only valid value for speed/mode is 10 Gbps/full-duplex.

The interfaces created by the oce driver can be configured through ifconfig:
ifconfig oce0 plumb xxx.xxx.xxx.xxx up ifconfig oce0 down unplumb
Files /kernel/drv/oce 32-bit ELF kernel module
/kernel/drv/amd64/oce 64-bit ELF kernel module, x86
/kernel/drv/sparcv9/oce 64-bit ELF kernel module, SPARC

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC, x86

Availability SUNWemlxs

See Also dladm(1M), ifconfig(1M), netstat(1M), prtconf(1M), attributes(5), dlpi(7P)

Writing Device Drivers

Network Interface Guide

STREAMS Programming Guide

IEEE 802.3ae Specificiation, IEEE - 2002

Device and Network Interfaces 611

ohci(7D)

Name ohci – OpenHCI host controller driver

Synopsis usb@unit-address
Description The ohci driver is a USBA (Solaris USB Architecture) compliant nexus driver that supports
the Open Host Controller Interface Specification 1.1, an industry standard developed by
Compaq, Microsoft, and National Semiconductor.
The ohci driver supports bulk, interrupt, control and isochronous transfers.
Files /kernel/drv/ohci 32–bit x86 ELF kernel module
/kernel/drv/amd64/ohci 64–bit x86 ELF kernel module
/kernel/drv/sparcv9/ohci 64–bit SPARC ELF kernel module
/kernel/drv/ohci.conf driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also attributes(5), ehci(7D), hubd(7D), uhci(7D), usba(7D)

Writing Device Drivers
Universal Serial Bus Specification 2.0
Open Host Controller Interface Specification for USB 1.0a
System Administration Guide: Basic Administration
http://www.sun.com/io
Diagnostics All host controller errors are passed to the client drivers. Root hub errors are documented in
hubd(7D).
In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
WARNING: <device path> <ohci><instance number>>: Error message...
Unrecoverable USB Hardware Error.
There was an unrecoverable USB hardware error reported by the OHCI Controller. Please
reboot the system. If this problem persists, contact your system vendor.
No SOF interrupts have been received. This OHCI USB controller is unusable.
The USB hardware is not generating Start Of Frame interrupts. Please reboot the system. If
this problem persists, contact your system vendor.

612 man pages section 7: Device and Network Interfaces • Last Revised 23 Nov 2006
 ohci(7D)

The following messages may be entered into the system log. They are formatted in the
following manner:
<device path> <ohci<instance number>): Message...
Failed to program frame interval register. For an unspecified reason, the frame interval
register has been nulled out by the Uli M1575
chip. Please reboot the system. If this problem
persists, contact your system vendor.

Device and Network Interfaces 613

openprom(7D)

Name openprom – PROM monitor configuration interface

Synopsis #include <sys/fcntl.h>
#include <sys/types.h>
#include <sys/openpromio.h>
open("/dev/openprom", mode);

Description The internal encoding of the configuration information stored in EEPROM or NVRAM varies
from model to model, and on some systems the encoding is “hidden” by the firmware. The
openprom driver provides a consistent interface that allows a user or program to inspect and
modify that configuration, using ioctl(2) requests. These requests are defined in
<sys/openpromio.h>:
struct openpromio {
uint_t oprom_size; /* real size of following data */
union {
char b[1]; /* NB: Adjacent, Null terminated */
int i;
} opio_u;
};
#define oprom_array opio_u.b /* property name/value array */
#define oprom_node opio_u.i /* nodeid from navigation config-ops */
#define oprom_len opio_u.i /* property len from OPROMGETPROPLEN */
#define OPROMMAXPARAM 32768 /* max size of array (advisory) */

For all ioctl(2) requests, the third parameter is a pointer to a struct openpromio. All
property names and values are null-terminated strings; the value of a numeric option is its
ASCII representation.

For the raw ioctl(2) operations shown below that explicitly or implicitly specify a nodeid, an
error may be returned. This is due to the removal of the node from the firmware device tree by
a Dynamic Reconfiguration operation. Programs should decide if the appropriate response is
to restart the scanning operation from the beginning or terminate, informing the user that the
tree has changed.
ioctls OPROMGETOPT This ioctl takes the null-terminated name of a property in the
oprom_array and returns its null-terminated value (overlaying its
name). oprom_size should be set to the size of oprom_array; on return
it will contain the size of the returned value. If the named property does
not exist, or if there is not enough space to hold its value, then
oprom_size will be set to zero. See BUGS below.
OPROMSETOPT This ioctl takes two adjacent strings in oprom_array; the
null-terminated property name followed by the null-terminated value.

614 man pages section 7: Device and Network Interfaces • Last Revised 13 Jan 1997
 openprom(7D)

OPROMSETOPT2 This ioctl is similar to OPROMSETOPT, except that it uses the difference
between the actual user array size and the length of the property name
plus its null terminator.
OPROMNXTOPT This ioctl is used to retrieve properties sequentially. The
null-terminated name of a property is placed into oprom_array and on
return it is replaced with the null-terminated name of the next property
in the sequence, with oprom_size set to its length. A null string on input
means return the name of the first property; an oprom_size of zero on
output means there are no more properties.
OPROMNXT
OPROMCHILD
OPROMGETPROP
OPROMNXTPROP These ioctls provide an interface to the raw config_ops operations in the
PROM monitor. One can use them to traverse the system device tree;
see prtconf(1M).
OPROMGETPROPLEN This ioctl provides an interface to the property length raw config op. It
takes the name of a property in the buffer, and returns an integer in the
buffer. It returns the integer -1 if the property does not exist; 0 if the
property exists, but has no value (a boolean property); or a positive
integer which is the length of the property as reported by the PROM
monitor. See BUGS below.
OPROMGETVERSION This ioctl returns an arbitrary and platform-dependent
NULL-terminated string in oprom_array, representing the underlying
version of the firmware.
Errors EAGAIN There are too many opens of the /dev/openprom device.
EFAULT A bad address has been passed to an ioctl(2) routine.
EINVAL The size value was invalid, or (for OPROMSETOPT) the property does not exist, or an
invalid ioctl is being issued, or the ioctl is not supported by the firmware, or the
nodeid specified does not exist in the firmware device tree.
ENOMEM The kernel could not allocate space to copy the user's structure.
EPERM Attempts have been made to write to a read-only entity, or read from a write only
entity.
ENXIO Attempting to open a non-existent device.

Examples EXAMPLE 1 oprom_array Data Allocation and Reuse

The following example shows how the oprom_array is allocated and reused for data returned
by the driver.

Device and Network Interfaces 615

openprom(7D)

EXAMPLE 1 oprom_array Data Allocation and Reuse (Continued)

/*
* This program opens the openprom device and prints the platform
* name (root node name property) and the prom version.
*
* NOTE: /dev/openprom is readable only by user ’root’ or group ’sys’.
*/
#include <stdio.h>
#include <string.h>
#include <fcntl.h>
#include <errno.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/openpromio.h>
#define min(a, b) (a < b ? a : b)
#define max(a, b) (a > b ? a : b)
#define MAXNAMESZ 32 /* Maximum property *name* size */
#define BUFSZ 1024 /* A Handly default buffer size */
#define MAXVALSZ (BUFSZ - sizeof (int))
static char *promdev = "/dev/openprom";
/*
* Allocate an openpromio structure big enough to contain
* a bufsize’d oprom_array. Zero out the structure and
* set the oprom_size field to bufsize.
*/
static struct openpromio *
opp_zalloc(size_t bufsize)
{
struct openpromio *opp;
opp = malloc(sizeof (struct openpromio) + bufsize);
(void) memset(opp, 0, sizeof (struct openpromio) + bufsize);
opp->oprom_size = bufsize;
return (opp);
}
/*
* Free a ’struct openpromio’ allocated by opp_zalloc
*/
static void
opp_free(struct openpromio *opp)
{
free(opp);
}
/*
* Get the peer node of the given node. The root node is the peer of zero.
* After changing nodes, property lookups apply to that node. The driver
* ’remembers’ what node you are in.

616 man pages section 7: Device and Network Interfaces • Last Revised 13 Jan 1997
 openprom(7D)

EXAMPLE 1 oprom_array Data Allocation and Reuse (Continued)

*/
static int
peer(int nodeid, int fd)
{
struct openpromio *opp;
int i;
opp = opp_zalloc(sizeof (int));
opp->oprom_node = nodeid;
if (ioctl(fd, OPROMNEXT, opp) < 0) {
perror("OPROMNEXT");
exit(1);
}
i = opp->oprom_node;
opp_free(opp);
return(i);
}
int
main(void)
{
struct openpromio *opp;
int fd, proplen;
size_t buflen;
if ((fd = open(promdev, O_RDONLY)) < 0) {
fprintf(stderr, "Cannot open openprom device\n");
exit(1);
}
/*
* Get and print the length and value of the
* root node ’name’ property
*/
(void) peer(0, fd); /* Navigate to the root node */
/*
* Allocate an openpromio structure sized big enough to
* take the string "name" as input and return the int-sized
* length of the ’name’ property.
* Then, get the length of the ’name’ property.
*/
buflen = max(sizeof (int), strlen("name") + 1);
opp = opp_zalloc(buflen);
(void) strcpy(opp->oprom_array, "name");
if (ioctl(fd, OPROMGETPROPLEN, opp) < 0) {
perror("OPROMGETPROPLEN");
/* exit(1); */
proplen = 0; /* down-rev driver? */
} else

Device and Network Interfaces 617

openprom(7D)

EXAMPLE 1 oprom_array Data Allocation and Reuse (Continued)

proplen = opp->oprom_len;
opp_free(opp);
if (proplen == -1) {
printf("’name’ property does not exist!\n");
exit (1);
}
/*
* Allocate an openpromio structure sized big enough
* to take the string ’name’ as input and to return
* ’proplen + 1’ bytes. Then, get the value of the
* ’name’ property. Note how we make sure to size the
* array at least one byte more than the returned length
* to guarantee NULL termination.
*/
buflen = (proplen ? proplen + 1 : MAXVALSZ);
buflen = max(buflen, strlen("name") + 1);
opp = opp_zalloc(buflen);
(void) strcpy(opp->oprom_array, "name");
if (ioctl(fd, OPROMGETPROP, opp) < 0) {
perror("OPROMGETPROP");
exit(1);
}
if (opp->oprom_size != 0)
printf("Platform name <%s> property len <%d>\n",
opp->oprom_array, proplen);
opp_free(opp);
/*
* Allocate an openpromio structure assumed to be
* big enough to get the ’prom version string’.
* Get and print the prom version.
*/
opp_zalloc(MAXVALSZ);
opp->oprom_size = MAXVALSZ;
if (ioctl(fd, OPROMGETVERSION, opp) < 0) {
perror("OPROMGETVERSION");
exit(1);
}
printf("Prom version <%s>\n", opp->oprom_array);
opp_free(opp);
(void) close(fd);
return (0);
}
Files /dev/openprom PROM monitor configuration interface

618 man pages section 7: Device and Network Interfaces • Last Revised 13 Jan 1997
 openprom(7D)

See Also eeprom(1M), monitor(1M), prtconf(1M), ioctl(2), mem(7D)

Bugs There should be separate return values for non-existent properties as opposed to not enough
space for the value.

An attempt to set a property to an illegal value results in the PROM setting it to some legal
value, with no error being returned. An OPROMGETOPT should be performed after an
OPROMSETOPT to verify that the set worked.

Some PROMS lie about the property length of some string properties, omitting the NULL
terminator from the property length. The openprom driver attempts to transparently
compensate for these bugs when returning property values by NULL terminating an extra
character in the user buffer if space is available in the user buffer. This extra character is
excluded from the oprom_size field returned from OPROMGETPROP and OPROMGETOPT and
excluded in the oprom_len field returned from OPROMGETPROPLEN but is returned in the user
buffer from the calls that return data, if the user buffer is allocated at least one byte larger than
the property length.

Device and Network Interfaces 619

oplkmdrv(7D)

Name oplkmdrv – key management driver for the SPARC Enterprise Server family
Synopsis kmdrv

Description oplkmdrv is a character driver that implements a framework for exchanging the security keys
with the Service Processor on a SPARC Enterprise Server. The oplkmdrv driver is specific to
the SPARC Enterprise Server family.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdscpr

Interface Stability Private

attributes(5)

See Also attributes(5)

620 man pages section 7: Device and Network Interfaces • Last Revised 18 Apr 2006
 oplmsu(7D)

Name oplmsu – Serial I/O multiplexing STREAMS device driver

Synopsis /pseudo-console

Description The oplmsu driver is a STREAMS multiplexer driver that connects multiple serial devices to
the system console.

Currently, this support is provided only on a SPARC Enterprise Server.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr

Interface Stability Private

See Also attributes(5)

Device and Network Interfaces 621

oplpanel(7D)

Name oplpanel – device driver for the SPARC Enterprise Server family

Description The oplpanel device driver monitors the panel reset button. If the button is pressed, a
high-level interrupt is generated, and the oplpanel driver causes a system panic.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr

Interface Stability Private

See Also attributes(5)

622 man pages section 7: Device and Network Interfaces • Last Revised 18 Apr 2006
 packet(7P)

Name packet, PF_PACKET – packet interface on device level

Synopsis #include <sys/socket.h>
#include <netpacket/packet.h>
#include <sys/ethernet.h>
packet_socket = socket(2,7,n)(PF_PACKET, int socket_type, int protocol);

Description Packet sockets are used to receive or send (2,n) raw ( 3x,7, 8,3x cbreak) packets at the device
driver (OSI Layer 2) level. These allow users to implement protocol modules in (1,8) user
space on top of the physical layer.
The socket_type is either SOCK_RAW for raw (3x,7,8,3x cbreak) packets including the link (1,2)
level header or SOCK_DGRAM for cooked packets with the link (1,2) level header removed. The
link (1,2) level header information is available in (1,8) a common format in (1,8) a
sockaddr_ll. protocol is the IEEE 802.3 protocol number in (1,8) network order. See the
<sys/ethernet.h> include file (1,n) for a list of allowed protocols. When protocol is set (7,n,1
builtins) to htons (ETH_P_ALL) then all protocols are received. All incoming packets of that
protocol type is passed to the packet socket (2,7,n) before they are passed to the protocols
implemented in (1,8) the kernel.
Only process with the net_rawaccesss privilege may create PF_PACKET sockets. Processes in
the global zone may bind to any network interface that is displayed using the command: dladm
show-link.
SOCK_RAW packets are passed to and from the device driver without any changes in (1,8) the
packet data. When receiving a packet, the address is still parsed and passed in (1,8) a standard
sockaddr_ll address structure. When transmitting a packet, the user supplied buffer should
contain the physical layer header. That packet is then queued unmodified to the network
driver of the interface defined by the destination address.
SOCK_DGRAM operates on a slightly higher level. The physical header is removed before the
packet is passed to the user. Packets sent through a SOCK_DGRAM packet socket (2,7,n) get a
suitable physical layer header based on the information in (1,8) the sockaddr_ll destination
address before they are queued.
By default, all packets of the specified protocol type are passed to a packet socket. To only get
packets from a specific interface use bind (2,n,1 builtins)(2) specifying an address in (1,8) a
struct sockaddr_ll to bind (2,n,1 builtins) the packet socket (2,7,n) to an interface. Only the
sll_protocol and the sll_ifindex address fields are used for purposes of binding.
The connect(3SOCKET) operation is not supported on packet sockets.
Address Types The sockaddr_ll is a device independent physical layer address.

struct sockaddr_ll {
unsigned short sll_family; /* Always AF_PACKET */
unsigned short sll_protocol; /* Physical layer protocol */

Device and Network Interfaces 623

packet(7P)

int sll_ifindex; /* Interface number */

unsigned short sll_hatype; /* Header type */
unsigned char sll_pkttype; /* Packet type */
unsigned char sll_halen; /* Length of address */
unsigned char sll_addr[8]; /* Physical layer address */
};

sll_protocol is the standard ethernet protocol type in (1,8) network order as defined in (1,8)
the sys/ethernet.h include file. It defaults to the socket (2,7,n)'s protocol. sll_ifindex is the
interface index of the interface. See netdevice(7). 0 matches any interface (only legal for
binding). sll_hatype is a ARP type as defined in (1,8) the sys/ethernet.h include file.
sll_pkttype contains the packet type. Valid types are PACKET_HOST for a packet addressed to
the local host(1,5), PACKET_BROADCAST for a physical layer broadcast packet,
PACKET_MULTICAST for a packet sent to a physical layer multicast address, PACKET_OTHERHOST
for a packet to some other host (1,5) that has been caught by a device driver in(1,8)
promiscuous mode, and PACKET_OUTGOING for a packet originated from the local host(1,5)
that is looped back to a packet socket. These types make only sense for receiving. sll_addr
and sll_halen contain the physical layer, for example, IEEE 802.3, address and its length. The
exact interpretation depends on the device.

When you send (2,n) packets it is enough to specify sll_family, sll_addr, sll_halen,
sll_ifindex. The other fields should be 0. sll_hatype and sll_pkttype are set (7,n,1
builtins) on received packets for your information. For bind (2,n,1 builtins) only
sll_protocol and sll_ifindex are used.

Socket Options Packet sockets can be used to configure physical layer multicasting and promiscuous mode. It
works by calling setsockopt(3SOCKET) on a packet socket (2,7,n) for SOL_PACKET and one
of the options PACKET_ADD_MEMBERSHIP to add a binding or PACKET_DROP_MEMBERSHIP to
drop it. They both expect a packet_mreq structure as argument:
struct packet_mreq
{
int mr_ifindex; /* interface index */
unsigned short mr_type; /* action */
unsigned short mr_alen; /* address length */
unsigned char mr_address[8]; /* physical layer address */
};

mr_ifindex contains the interface index for the interface whose status should be changed. The
mr_type parameter specifies which action to perform. PACKET_MR_PROMISC enables receiving
all packets on a shared medium often known as promiscuous mode, PACKET_MR_MULTICAST
binds the socket (2,7,n) to the physical layer multicast group specified in (1,8) mr_address and
mr_alen. PACKET_MR_ALLMULTI sets the socket (2,7,n) up to receive all multicast packets
arriving at the interface.

In addition the traditional ioctls, SIOCSIFFLAGS, SIOCADDMULTI, and SIOCDELMULTI can be

used for the same purpose.

624 man pages section 7: Device and Network Interfaces • Last Revised 28 Oct 2009
 packet(7P)

See Also connect(3SOCKET), setsockopt(3SOCKET)

Notes For portable programs it is suggested to usepcap(3C) instead of PF_PACKET; although this only
covers a subset of the PF_PACKET features.

The SOCK_DGRAM packet sockets make no attempt to create or parse the IEEE 802.2 LLC header
for a IEEE 802.3 frame. When ETH_P_802_3 is specified as protocol for sending the kernel
creates the 802.3 frame and fills out the length field; the user has to supply the LLC header to
get a fully conforming packet. Incoming 802.3 packets are not multiplexed on the DSAP/SSAP
protocol fields; instead they are supplied to the user as protocol ETH_P_802_2 with the LLC
header prepended. It is therefore not possible to bind (2,n,1 builtins) to ETH_P_802_3; bind
(2,n,1 builtins) to ETH_P_802_2 instead and do the protocol multiplex yourself. The default for
sending is the standard Ethernet DIX encapsulation with the protocol filled in.

Packet sockets are not subject to the input or output firewall chains.

Device and Network Interfaces 625

pcan(7D)

Name pcan – Cisco Aironet 802.11b wireless NIC driver

Description The pcan wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver. It supports the pccard and PCI/MiniPCI cards with the Cisco Aironet
802.11b chipset. For pccard, the driver works in both SPARC and x86 (32–bit/64–bit) modes.
For PCI/MiniPCI card, the driver works in 32–bit x86 mode only.

Driver The pcan driver supports 802.11b data rates of 1, 2, 5.5 and 11 (Mbits/sec). The default is 11.
Configuration
The pcan driver supports BSS networks (also known as "ap" or "infrastructure" networks) and
IBSS networks (also known as "ad-hoc" networks). For authentication type, the pcan driver
supports the "open" (or "open-system") mode. For encryption type, only WEP is currently
supported. You perform configuration and administration tasks using the dladm(1M) and
wificonfig(1M) utilities.
Files /dev/pcan* Special character device.
/kernel/drv/pcan 32-bit ELF kernel module (x86).
/kernel/drv/amd64/pcan 64-bit ELF kernel module (x86).
/kernel/drv/sparcv9/pcan 64-bit ELF kernel module (SPARC).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC, x86

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D)

802.11b Standard for Wireless Local Area Networks (WLANs) - IEEE

626 man pages section 7: Device and Network Interfaces • Last Revised 24 Jan 2007
 pcata(7D)

Name pcata – PCMCIA ATA card device driver

Synopsis pcata@socket#:a -u
pcata@socket#:a -u,raw

Description The PCMCIA ATA card device driver supports PCMCIA ATA disk and flash cards that follow
the following standards:
■ PC card 2.01 compliance (MBR+fdisk table required for all platforms).
■ PC card ATA 2.01 compliance.
■ PC card services 2.1 compliance.

The driver supports standard PCMCIA ATA cards that contain a Card Information Structure
(CIS). For PCMCIA, nodes are created in /devices that include the socket number as one
component of the device name referred to by the node. However, the names in /dev,
/dev/dsk, and /dev/rdsk follow the current conventions for ATA devices, which do not
encode the socket number in any part of the name. For example, you may have the following:

Platform /devices name /dev/dsk name

x86 /devices/isa/pcic@1,3e0 /dev/dsk/c1d0s0

/disk@0:a

SPARC /devices/iommu@f,e0000000 /dev/dsk/c1d0s0

/sbus@f,e0001000 /SUNW,
pcmcia@3,0 /disk@0:a

Preconfiguration If a PC Card ATA device is recognized, the pcata driver is automatically loaded, IRQs
allocated, devices nodes created, and special files created (if they do not already exist).
Known Problems and ■ You need to umount the file system before removing the disk.
Limitations
■ The ufs file systems on removable media (PC Card ATA) should have one of the
onerror={panic, lock, umount} mount options set.

Configuration Configuration topics include initial installation and configuration, identifying an

unrecognized device, special files and hot-plugging.
Initial Installation and 1. Install the Solaris software.
Configuration 2. Boot the system.
3. Insert the PC card ATA device.

Identifying an If you insert a PC card ATA device and it is not recognized (no special files created), use the
Unrecognized Device prtconf command to identify the problem.

1. Run the prtconf -D command to see if your pcata card is recognized. (A recognized
device will appear at the end of the prtconf output. For example:
# prtconf -D
. . .

Device and Network Interfaces 627

pcata(7D)

pcic, instance #0 (driver name: pcic)

. . .
disk, instance #0
2. If pcata does not appear in the prtconf output, there is a problem with the PC card
adapter configuration or with the hardware. Check to see whether the problem is with the
card or the adapter by trying to use the card on another machine and by seeing if it works
on the same machine using DOS.

Special Files For PC card devices, nodes are created in /devices that include the socket number as one
component of a device name that the node refers to. However, the /prtc/dev names and the
names in /dev/dsk and /dev/rdsk do follow the current convention for ATA devices, which
do not encode the socket number in any part of the name.
Hot-Plugging ■ If you want to remove the disk, you must unmount the file system.
■ Use the mkfs_pcfs(1M) command to create a pcfs file system:
# mkfs -F pcfs /dev/rdsk/c#d#p0:d
■ To mount a pcfs file system, type:
# mount -F pcfs /dev/dsk/c#d#p0:c /mnt
■ If you want to create a ufs file system, use the newfs command and type:
# newfs /dev/rdsk/c#d#s#
■ To mount a ufs file system, type:
# mount -F ufs /dev/dsk/c#d#s# /mnt
■ To create a Solaris partition, run the format command and go to the Partition menu. For
more information, see the format(1M) man page.
Files /kernel/drv/pcata pcata driver

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWpsdpr

See Also format(1M), mount(1M), newfs(1M), pcmcia(7D), attributes(5), pcfs(7FS)

628 man pages section 7: Device and Network Interfaces • Last Revised 2 Mar 2007
 pcfs(7FS)

Name pcfs – FAT formatted file system

Synopsis #include <sys/param.h>
#include <sys/mount.h>
#include <sys/fs/pc_fs.h>

int mount(const char *spec,

const char * dir, int mflag,
"pcfs", NULL, 0, char *optptr,
int optlen);

Description pcfs is a file system type that enables direct access to files on FAT formatted disks from within
the SunOS operating system.
Once mounted, pcfs provides standard SunOS file operations and semantics. Using pcfs, you
can create, delete, read, and write files on a FAT formatted disk. You can also create and delete
directories and list files in a directory.
pcfs supports FAT12 (floppies) and FAT16 and FAT32 file systems.
pcfs file systems can be force umounted using the -f argument to umount(1M).
The pcfs file system contained on the block special file identified by spec is mounted on the
directory identified by dir. spec and dir are pointers to pathnames. mflag specifies the mount
options. The MS_DATA bit in mflag must be set. Mount options can be passed to pcfs using
the optptr and optlen arguments. See mount_pcfs(1M) for a list of mount options supported
by pcfs.
Because FAT formatted media can record file timestamps between January 1st 1980 and
December 31st 2127, it's not possible to fully represent UNIX time_t in pcfs for 32 bit or 64 bit
programs. In particular, if post-2038 timestamps are present on a FAT formatted medium and
pcfs returns these, 32bit applications may unexpectedly fail with EOVERFLOW errors. To
prevent this, the default behaviour of pcfs has been modified to clamp post-2038 timestamps
to the latest possible value for a 32bit time_t, which is January 19th 2038, 03:14:06 UTC when
setting and retrieving file timestamps. You can override this behavior using the noclamptime
mount option, as described in mount_pcfs(1M).
Timestamps on FAT formatted media are recorded in local time. If the recording and
receiving systems use different timezones, the representation of timestamps shown on the two
systems for the same medium might vary. To correct this, pcfs provides a timezone mount
option to force interpretation of timestamps as read from a FAT formatted medium in a given
timezone (that of the recorder). By default, the local timezone of the receiver is used. See
mount_pcfs(1M) for details.
The root directory of a FAT formatted medium has no timestamps and pcfs returns the time
when the mount was done as timestamp for the root of the filesystem.
The FAT filesystem doesn't support multiple links. As a result, the link count for all files and
directories in pcfs is hard-coded as 1.

Device and Network Interfaces 629

pcfs(7FS)

Mounting File Systems Use the following command to mount pcfs from diskette:

mount -F pcfs device-special directory-name

You can use:

mount directory-name

if the following line is in your /etc/vfstab file:

device-special - directory-namepcfs − no rw

Use the following command to mount pcfs from non-diskette media:

mount -F pcfs device-special:logical-drive directory-name

You can use:

mount directory-name

if the following line is in your /etc/vfstab file:

device-special:logical_drive − directory-name pcfs − no rw

device-special specifies the special block device file for the diskette (/dev/disketteN) or the
entire hard disk (/dev/dsk/cNtNdNp0 for a SCSI disk, and /dev/dsk/cNdNp0 for IDE disks)
or the PCMCIA pseudo-floppy memory card (/dev/dsk/cNtNdNsN).
logical-drive specifies either the DOS logical drive letter (c through z) or a drive number (1
through 24). Drive letter c is equivalent to drive number 1 and represents the Primary DOS
partition on the disk; drive letters d through z are equivalent to drive numbers 2 through 24,
and represent DOS drives within the Extended FAT partition. Note that device-special and
logical-drive must be separated by a colon.
directory-name specifies the location where the file system is mounted.
For example, to mount the Primary DOS partition from a SCSI hard disk, use:
mount -F pcfs /dev/dsk/cNtNdNp0:c /pcfs/c

To mount the first logical drive in the Extended DOS partition from an IDE hard disk, use:
mount -F pcfs /dev/dsk/cNdNp0:d /pcfs/d

To mount a DOS diskette in the first floppy drive when volume management is not running
use:
mount -F pcfs /dev/diskette /pcfs/a

If Volume Management is running, run volcheck(1) to automatically mount the floppy and
some removable disks.
To mount a PCMCIA pseudo-floppy memory card, with Volume Management not running
(or not managing the PCMCIA media), use:

630 man pages section 7: Device and Network Interfaces • Last Revised 1 Dec 2009
 pcfs(7FS)

mount -F pcfs /dev/dsk/cNtNdNsN /pcfs

Conventions Files and directories created through pcfs must comply with either the FAT short file name
convention or the long file name convention introduced with Windows 95. The FAT short file
name convention is of the form filename[.ext], where filename generally consists of from one
to eight upper-case characters, while the optional ext consists of from one to three upper-case
characters.

The long file name convention is much closer to Solaris file names. A long file name can
consist of any characters valid in a short file name, lowercase letters, non-leading spaces, the
characters +,;=[], any number of periods, and can be up to 255 characters long. Long file
names have an associated short file name for systems that do not support long file names
(including earlier releases of Solaris). The short file name is not visible if the system recognizes
long file names. pcfs generates a unique short name automatically when creating a long file
name.

Given a long file name such as This is a really long filename.TXT, the short file name
will generally be of the form THISIS~N.TXT, where N is a number. The long file name will
probably get the short name THISIS~1.TXT, or THISIS~2.TXT if THISIS~1.TXT already exits
(or THISIS~3.TXT if both exist, and so forth). If you use pcfs file systems on systems that do
not support long file names, you may want to continue following the short file name
conventions. See EXAMPLES.

When creating a file name, pcfs creates a short file name if it fits the FAT short file name
format, otherwise it creates a long file name. This is because long file names take more
directory space. Because the root directory of a pcfs file system is fixed size, long file names in
the root directory should be avoided if possible.

When displaying file names, pcfs shows them exactly as they are on the media. This means
that short names are displayed as uppercase and long file names retain their case. Earlier
versions of pcfs folded all names to lowercase, which can be forced with the
PCFS_MNT_FOLDCASE mount option. All file name searches within pcfs, however, are treated as
if they were uppercase, so readme.txt and ReAdMe.TxT refer to the same file.

To format a diskette or a PCMCIA pseudo-floppy memory card in FAT format in the SunOS
system, use either the fdformat -d or the DOS FORMAT command.

Boot Partitions On x86 systems, hard drives may contain an fdisk partition reserved for the Solaris boot
utilities. These partitions are special instances of pcfs. You can mount an x86 boot partition
with the command:
mount -F pcfs device-special:boot directory-name

or you can use:

mount directory-name

if the following line is in your /etc/vfstab file:

Device and Network Interfaces 631

pcfs(7FS)

device-special:boot − directory-name pcfs − no rw

device-special specifies the special block device file for the entire hard disk
(/dev/dsk/cNtNdNp0)

directory-name specifies the location where the file system is mounted.

All files on a boot partition are owned by super-user. Only the super-user may create, delete,
or modify files on a boot partition.

Examples EXAMPLE 1 Sample Displays of File Names

If you copy a file financial.data from a UNIX file system to pcfs, it displays as
financial.data in pcfs, but may show up as FINANC~1.DAT in systems that do not support
long file names.

The following are legal long file names. They are also illegal short file names:

test.sh.orig
data+
.login

Other systems that do not support long file names may see:

TESTSH~1.ORI
DATA~1
LOGIN~1

The short file name is generated from the initial characters of the long file name, so
differentiate names in the first few characters. For example, these names:

WorkReport.January.Data
WorkReport.February.Data
WorkReport.March.Data

result in these short names, which are not distinguishable:

WORKRE~1.DAT
WORKRE~2.DAT
WORKRE~13.DAT

These names, however:

January.WorkReport.Data

632 man pages section 7: Device and Network Interfaces • Last Revised 1 Dec 2009
 pcfs(7FS)

EXAMPLE 1 Sample Displays of File Names (Continued)

February.WorkReport.Data
March.WorkReport.Data

result in the more descriptive short names:

JANUAR~1.DAT
FEBRUA~1.DAT
MARCHW~1.DAT
Files /usr/lib/fs/pcfs/mount pcfs mount command
/usr/kernel/fs/pcfs 32-bit kernel module

Environment See environ(5) for descriptions of the following environment variables for the current locale
Variables setting: LANG, LC_ALL, LC_CTYPE, and LC_COLLATE.

See Also chgrp(1), chown(1), dos2unix(1), eject(1), fdformat(1), unix2dos(1), volcheck(1),

mount(1M), mount_pcfs(1M), umount(1M), ctime(3C), vfstab(4), environ(5),

Warnings Do not physically eject a FAT floppy while the device is mounted as pcfs. If Volume
Management is managing a device, use the eject(1) command before physically removing
media.

When mounting pcfs on a hard disk, make sure the first block on that device contains a valid
fdisk partition table.

Because pcfs has no provision for handling owner-IDs or group-IDs on files, chown(1) or
chgrp(1) may generate various errors. This is a limitation of pcfs, but it should not cause
problems other than error messages.

Notes Only the following characters are allowed in pcfs short file names and extensions:

0-9
A-Z
$#&@!%()-{}<>‘_^~|’

SunOS and FAT use different character sets and have different requirements for the text file
format. Use the dos2unix(1) and unix2dos(1) commands to convert files between them.

pcfs offers a convenient transportation vehicle for files between Sun workstations and PCs.
Because the FAT disk format was designed for use under DOS, it does not operate efficiently
under the SunOS system and should not be used as the format for a regular local storage.
Instead, use ufs for local storage within the SunOS system.

Device and Network Interfaces 633

pcfs(7FS)

Although long file names can contain spaces (just as in UNIX file names), some utilities may
be confused by them.

This implementation of pcfs conforms to the behavior exhibited by Windows 95 version

4.00.950.

When pcfs encounters long file names with non-ASCII characters, it converts such long file
names in Unicode scalar values into UTF-8 encoded filenames so that they are legible and
usable with any of Solaris UTF-8 locales. In the same context, when new file names with
non-ASCII characters are created, pcfs expects that such file names are in UTF-8. This feature
increases the interoperability of pcfs on Solaris with other operating systems.

Bugs pcfs should handle the disk change condition in the same way that DOS does, so you do not
need to unmount the file system to change floppies.

634 man pages section 7: Device and Network Interfaces • Last Revised 1 Dec 2009
 pcic(7D)

Name pcic – Intel i82365SL PC Card Interface Controller

Description The Intel i82365SL PC Card interface controller provides one or more PCMCIA PC card
sockets. The pcic driver implements a PCMCIA bus nexus driver.

The driver provides basic support for the Intel 82365SL and compatible chips. Tested chips
are:
■ Intel — 82365SL.
■ Cirrus Logic — PD6710/PD6720/PD6722.
■ Vadem — VG365/VG465/VG468/VG469.
■ Toshiba — PCIC and ToPIC
■ Ricoh — RF5C366/RL5C466/RL5C475/RL5C476/RL5C477/RL5C478.
■ 02Micro — OZ6912/6972.
■ Texas Instruments — PCI1130/PCI1131/PCI1031/PCI1221/PCI1225/PCI1520/PCI
1410/PCI1420/PCI4520/PCI7510/PCI7621.

While most systems using one of the above chips will work, some systems are not supported
due to hardware designs options that may not be software detectable.

Direct access to the PCMCIA hardware is not supported. All device access must be through
the DDI.

Configuration Configuration of PC Card interface controllers are automatically done in the system by
leveraging ACPI on x86 (or OBP on SPARC). Configuration includes allocation of device
memory, I/O ports, CardBus subordinary bus number and interrupts. There is no
user-interference required. Note that the controller may not work when ACPI is disabled.

There is one driver configuration property defined in the pcic.conf file:

interrupt-priorities=6; This property must be defined and must be below 10.
Files /kernel/drv/pcic pcic driver
/kernel/drv/pcic.conf pcic configuration file

See Also cardbus(4), pcmcia(7D)

Device and Network Interfaces 635

pcicmu(7D)

Name pcicmu – PCI bus nexus driver for the SPARC Enterprise Server family

Description The pcicmu nexus driver is used for onboard devices for the SPARC Enterprise Server family.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr

Interface Stability Private

See Also attributes(5)

636 man pages section 7: Device and Network Interfaces • Last Revised 18 Apr 2006
 pcie_pci(7D)

Name pcie_pci – PCI Express bridge nexus driver

Description The pcie_pci nexus driver is used on X64 servers for PCI Express bridge class devices
including PCI Express root ports which are implemented as virtual bridges and PCI Express to
PCI/PCI-X bridges.

The pcie_pci driver is compliant with the PCI Express Base, Revision 1.0a specification and
supports Base line PCI Express error handling and PCI Express Hot Plug.
Files /platform/i86pc/kernel/drv/pcie_pci 32-bit ELF kernel module.
/platform/i86pc/kernel/drv/amd64/pcie_pci 64-bit ELF kernel module.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x64 PCI Express-based systems

Availability SUNWcakr.i

See Also attributes(5), pcie(4), npe(7D)

PCI Express Base Specification v1.0a —2003

Writing Device Drivers

IEEE 1275 PCI Bus Binding — 1998

http://playground.sun.com/1275/bindings/pci/pci-express.txt

Device and Network Interfaces 637

pcipsy(7D)

Name pcipsy – PCI bus to Safari bus nexus driver

Description The pcipsy nexus driver is used for certain IO assemblies for sun4u and high-end Enterprise
E10000 servers..
Files /platform/SUNW,Ultra-Enterprise-10000/kernel/drv/sparcv9/pcipsy
32–bit ELF kernel module.
/platform/sun4u/kernel/drv/sparcv9/pcipsy
64-bit ELF kernel module.

See Also pcisch(7D)

638 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2005
 pcisch(7D)

Name pcisch – PCI Bus to Safari bus nexus driver

Description The pcisch nexus driver is used for Schizo and XMITS-based I/O assemblies for the following
high-end and midrange Sun enterprise servers: Sun Fire E15K, Sun Fire E25K, Sun Fire E2900,
Sun Fire E4900 and Sun Fire E6900.
Files /platform/sun4u/kernel/drv/sparcv9/pcisch 64-bit ELF kernel module.

See Also pcipsy(7D)

Device and Network Interfaces 639

pckt(7M)

Name pckt – STREAMS Packet Mode module

Synopsis int ioctl( fd, I_PUSH, "pckt");

Description pckt is a STREAMS module that may be used with a pseudo terminal to packetize certain
messages. The pckt module should be pushed (see I_PUSH on streamio(7I)) onto the master
side of a pseudo terminal.

Packetizing is performed by prefixing a message with an M_PROTO message. The original

message type is stored in the 1 byte data portion of the M_PROTO message.

On the read-side, only the M_PROTO, M_PCPROTO, M_STOP, M_START, M_STOPI, M_STARTI,
M_IOCTL, M_DATA, M_FLUSH, and M_READ messages are packetized. All other message types are
passed upstream unmodified.

Since all unread state information is held in the master's stream head read queue, flushing of
this queue is disabled.

On the write-side, all messages are sent down unmodified.

With this module in place, all reads from the master side of the pseudo terminal should be
performed with the getmsg(2) or getpmsg() function. The control part of the message
contains the message type. The data part contains the actual data associated with that message
type. The onus is on the application to separate the data into its component parts.

See Also getmsg(2), ioctl(2), ldterm(7M), ptem(7M), streamio(7I), termio(7I)

STREAMS Programming Guide

640 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 pcmcia(7D)

Name pcmcia – PCMCIA nexus driver

Description The PCMCIA nexus driver supports PCMCIA card client device drivers. There are no
user-configurable options for this driver.
Files /kernel/misc/pcmcia pcmcia driver

Device and Network Interfaces 641

pcn(7D)

Name pcn – AMD PCnet Ethernet controller device driver

Synopsis /dev/pcn

Description The pcn Ethernet driver is a multi-threaded, loadable, clonable driver for the AMD PCnet
family of Ethernet controllers that use the Generic LAN Driver (GLD) facility to implement the
required STREAMS and Data Link Provider (see dlpi(7P)) interfaces.

This driver supports a number of integrated motherboards and add-in adapters based on the
AMD PCnet-ISA, PCnet-PCI, and PCnet-32 controller chips. The pcn driver functions
include controller initialization, frame transmit and receive, functional addresses,
promiscuous and multicast support, and error recovery and reporting.

Application The cloning character-special device, /dev/pcn, is used to access all PCnet devices installed in
Programming the system.
Interface
pcn and DLPI The pcn driver uses the Solaris GLD module which handles all the STREAMS and DLPI specific
functions of the driver. It is a style 2 DLPI driver and therefore supports only the
connectionless mode of data transfer. Thus, a DLPI user should issue a DL_ATTACH_REQ
primitive to select the device to be used. Valid DLPI primitives are defined in <sys/dlpi.h>.
Refer to dlpi(7P) for more information.

The device is initialized on the first attach and de-initialized (stopped) on the last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to a DL_INFO_REQ
from the user are:
■ Maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h>).
■ Minimum SDU is 0.
■ DLSAP address length is 8.
■ MAC type is DL_ETHER.
■ sap length value is −2, meaning the physical address component is followed immediately
by a 2-byte sap component within the DLSAP address.
■ Service mode is DL_CLDLS.
■ No optional quality of service (QOS) support is included at present, accordingly, the QOS
fields are 0.
■ Provider style is DL_STYLE2.
■ Version is DL_VERSION_2.
■ Broadcast address value is the Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Once in the DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.

642 man pages section 7: Device and Network Interfaces • Last Revised 20 Oct 2000
 pcn(7D)

Known Problems and ■ Occasional data corruption has occurred when pcn and pcscsi drivers in HP Vectra XU
Limitations 5/90 and Compaq Deskpro XL systems are used under high network and SCSI loads.
These drivers do not perform well in a production server. A possible workaround is to
disable the pcn device with the system BIOS and use a separate add-in network interface.
■ The Solaris pcn driver does not support IRQ 4.
Files /dev/pcn Character special device
/kernel/drv/pcn.conf Configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), standards(5), dlpi(7P), streamio(7I)

Writing Device Drivers

STREAMS Programming Guide

Device and Network Interfaces 643

pcser(7D)

Name pcser – PCMCIA serial card device driver

Synopsis serial@<socket>:pcser
serial@<socket>:pcser,cu

Description The pcser PCMCIA serial card device driver supports asynchronous serial I/O access to any
PCMCIA card that that complies with Revision 2.1 of the PCMCIA standard and which
represents an 8250-type UART interface.
Preconfigure If a PC card modem or serial device is recognized, the pcser device driver is automatically
loaded, ports and IRQs allocated, and special files created (if they don't already exist).
Configuration Configuration steps include initial installation and configuration, identifying an unrecognized
device and misidentifying a recognized device.
Initial Installation and 1. Install the Solaris software.
Configuration 2. Boot the system.
3. Insert the modem or serial device.
Identifying an If you insert a PC card modem or serial device and it is not recognized (that is, no special files
Unrecognized Device are created under /dev/cua or /dev/term), use the prtconf command to find the problem:

1. Become root.
2. Run the prtconf -D command to see if your modem or serial device is recognized. An
unrecognized device will appear at the end of the prtconf output. For example:
# prtconf -D
. . .
pcic, instance #0 (driver name: pcic)
. . .
pccard111.222 (driver not attached)
3. If your device is not recognized, use the add_drv command to add the name of your device
as another known alias for pcser devices. For example, type the following at the command
line:
# add_drv -i’"pccard111.222"’ pcser
Note – Include the double quotes in single quotes to keep the shell from stripping out the
double quotes. Use the identification string listed in the prtconf output. Use the entire
string in the add_drv command. See add_drv(1M).
Misidentifying a 1. Run the prtconf -D command to see if your modem or serial device is erroneously
Recognized Device recognized as a memory card. If the device is incorrectly recognized as a memory card, the
output of the prtconf command could show:
# prtconf -D
. . .
pcic, instance #0 (driver name: pcic)
. . .

644 man pages section 7: Device and Network Interfaces • Last Revised 10 Sep 2002
 pcser(7D)

memory, instance #0 (driver name: pcmem)

pcram, instance #0 (driver name: pcram)
2. Use the Configuration Assistant to identify the memory resource conflict, and add correct
information for the device on the View/Edit Devices menu. Typically, the problem may be
a resource conflict between device memory settings. A PC Card adapter chip that is not
fully supported may also be the cause of the problem.
3. To work properly with the Solaris operating environment, all devices must be accounted
for, even those the Solaris environment does not support. The Configuration Assistant
software accounts for all devices in your system.

Additional When adding a new serial port or modem to the system, you often need to edit configuration
Configuration files so that applications can use the new communications port. For example, the
/etc/uucp/Devices file needs to be updated to use UUCP. See Overview of UUCP in the
System Administration Guide. For PPP on the serial port, see pppd(1M) and Solaris PPP
Overview in the System Administration Guide.

Special Files The serial devices in /dev/term and /dev/cua are named by socket number. A card inserted
in socket 0 is pc0, and socket 1 is pc1.

Hot Plugging If a PC Card modem or serial device is unplugged while in use, the device driver returns errors
until the card is replaced in the socket.

The device must be closed and reopened with the card reinserted before the device begins
working again. The restart process depends on the application. For example, a tip session
automatically exits when a card in use is unplugged. To restart the system, you must restart the
tip session.
Files /kernel/drv/pcser pcser driver
/dev/term/pcn dial-in devices
/dev/cua/pcn dial-out devices where: n is the PCMCIA physical socket number.

See Also cu(1C), tip(1), uucp(1C), autopush(1M), ports(1M), ioctl(2), open(2), pcmcia(7D),
termio(7I), ldterm(7M), ttcompat(7M)
Diagnostics pcser: socketn soft silo overflow
The driver's character input ring buffer overflowed before it could be serviced.
pcser: socketn unable to get CIS information
The CIS on the card has incorrect information or is in an incorrect format. This message
usually indicates a non-compliant card.

Device and Network Interfaces 645

pcwl(7D)

Name pcwl – Lucent/PrismII 802.11b wireless NIC driver

Description The pcwl 802.11b wireless NIC driver is a multi- threaded, loadable, clonable, GLDv3-based
STREAMS driver. It supports the pccard and PCI/MiniPCI cards with the Lucent and PrismII
802.11b chipsets on x86 and SPARC.

Driver The pcwl driver supports 802.11b data rates of 1, 2, 5.5 and 11 (Mbits/sec). The default is 11.
Configuration
The pcwl driver supports BSS networks (also known as "ap" or "infrastructure" networks) and
IBSS (or "ad-hoc") networks. For authentication type, the pcwl driver supports the "open" (or
"open-system") mode and the "shared-key" mode. For encryption type, only WEP is currently
supported. You perform configuration and administration tasks using the dladm(1M) and
wificonfig(1M) utilities.
Files /dev/pcwl* Special character device.
/kernel/drv/pcwl 32-bit ELF kernel module (x86).
/kernel/drv/amd64/pcwl 64-bit ELF kernel module (x86).
/kernel/drv/sparcv9/pcwl 64-bit ELF kernel module (SPARC).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC, x86

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D)

802.11b Standard for Wireless Local Area Networks (WLANs) - IEEE

646 man pages section 7: Device and Network Interfaces • Last Revised 24 Jan 2007
 pfb(7D)

Name pfb – Sun XVR-50 and XVR-100 Graphics Accelerator device driver

Description The pfb driver is the device driver for the Sun XVR-50 and XVR-100 Graphics Accelerator.
Files /dev/fbs/pfb\fIn Device special file for XVR-50 or XVR-100 single screen.
/dev/fbs/pfb\fIn\fPa Device special file for the XVR-100 first video out.
/dev/fbs/pfb\fIn\fPb Device special file for the XVR-100 second video out.

See Also SUNWpfb_config(1M)

Device and Network Interfaces 647

pf_key(7P)

Name pf_key – Security association database interface

Synopsis #include <sys/types.h>
#include <sys/socket.h>
#include <net/pfkeyv2.h>

int socket(PF_KEY,SOCK_RAW,PF_KEY_V2);

Description Keying information for IPsec security services is maintained in security association databases
(SADBs). The security associations (SAs) are used to protect both inbound and outbound
packets.

A user process (or possibly multiple co-operating processes) maintains SADBs by sending
messages over a special kind of socket. This is analogous to the method described in
route(7P). Only a superuser may access an SADB.

SunOS applications that use PF_KEY include ipseckey(1M) and in.iked(1M).

The operating system can spontaneously send pf_key messages to listening processes, such as
a request for a new SA for an outbound datagram or to report the expiration of an existing SA.

One opens the channel for passing SADB control messages by using the socket call shown in
the Synopsis section above. More than one key socket can be open per system.

Messages are formed by a small base header, followed by zero or more extension messages,
some of which require additional data following them. The base message and all extensions
must be eight-byte aligned. An example message is the GET message, which requires the base
header, the SA extension, and the ADDRESS_DST extension.

Messages Messages include:

#define SADB_GETSPI /* Get a new SPI value from the system. */

#define SADB_UPDATE /* Update an SA. */
#define SADB_ADD /* Add a fully-formed SA. */
#define SADB_DELETE /* Delete an SA. */
#define SADB_GET /* Get an SA */
#define SADB_ACQUIRE /* Kernel needs a new SA. */
#define SADB_REGISTER /* Regis. to receive ACQUIRE msgs. */
#define SADB_EXPIRE /* SA has expired. */
#define SADB_FLUSH /* Flush all SAs. */
#define SADB_DUMP /* Get all SAs. (Unreliable) */
#define SADB_X_PROMISC /* Listen promiscuously */
#define SADB_X_INVERSE_ACQUIRE /* Query kernel policy,
get an ACQUIRE in return. */
#define SADB_X_UPDATEPAIR /* Update an SA and its pair SA */
#define SADB_X_DELPAIR /* Delete an SA pair. */

648 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

The base message header consists of:

struct sadb_msg {
uint8_t sadb_msg_version; /* Set to PF_KEY_V2, for compat. */
uint8_t sadb_msg_type; /* Msg. type */
uint8_t sadb_msg_errno; /* Why message failed */
uint8_t sadb_msg_satype; /* Which security service */
uint16_t sadb_msg_len; /* Length in 8-byte units */
uint16_t sadb_msg_reserved; /* Zero out */
#define sadb_x_msg_diagnostic sadb_msg_reserved
/* Extended diagnostics for errors */
uint32_t sadb_msg_seq; /* For msg. originator */
uint32_t sadb_msg_pid; /* ID originator */
};

Extension types include:

#define SADB_EXT_SA /* SA info */
#define SADB_EXT_LIFETIME_HARD /* Hard lifetime */
#define SADB_EXT_LIFETIME_SOFT /* Soft lifetime */
#define SADB_EXT_ADDRESS_SRC /* Source address */
#define SADB_EXT_ADDRESS_DST /* Destination address */
#define SADB_EXT_ADDRESS_PROXY /* Proxy address - DEPRECATED */
#define SADB_EXT_KEY_AUTH /* Authen. key */
#define SADB_EXT_KEY_ENCRYPT /* Encryption key */
#define SADB_EXT_IDENTITY_SRC /* Source certif. ID */
#define SADB_EXT_IDENTITY_DST /* Destination certif. ID */
#define SADB_EXT_SENSITIVITY /* Sensitivity info */
#define SADB_EXT_PROPOSAL /* Security proposal */
#define SADB_EXT_SUPPORTED_AUTH /* Supported authen. algo’s */
#define SADB_EXT_SUPPORTED_ENCRYPT /* Supported encryption algo’s */
#define SADB_EXT_SPIRANGE /* Range of possible SPIs *
#define SADB_X_EXT_EREG /* Reg. for extended ACQUIRE */
#define SADB_X_EXT_EPROP /* Extended ACQUIRE proposals */
#define SADB_X_EXT_KM_COOKIE /* Indicates which KM derived SA. */
#define SADB_X_EXT_ADDRESS_NATT_LOC /* NAT-Traversal local (my public) */
#define SADB_X_EXT_ADDRESS_NATT_REM /* NAT-T remote (peer’s private) */
#define SADB_X_EXT_ADDRESS_INNER_SRC /* Tunnel-mode inner source */
#define SADB_X_EXT_ADDRESS_INNER_DST /* Tunnel-mode inner dest */
#define SADB_X_EXT_REPLAY_VALUE /* Replay Value */
#define SADB_X_EXT_LIFETIME_IDLE /* Idle lifetime */
#define SADB_X_EXT_PAIR /* SA pair extension*/
#define SADB_X_EXT_OUTER_SENS /*outer sensitivity */

Security Association Information Extension flags:

#define SADB_SAFLAGS_PFS 0x1 /* Perfect forward secrecy? */
#define SADB_SAFLAGS_NOREPLAY 0x2 /* Replay field NOT PRESENT. */
#define SADB_X_SAFLAGS_USED 0x80000000 /* SA used/not used */

Device and Network Interfaces 649

pf_key(7P)

#define SADB_X_SAFLAGS_UNIQUE 0x40000000 /* SA unique/reusable */

#define SADB_X_SAFLAGS_AALG1 0x20000000 /* Auth-alg specif. flag 1 */
#define SADB_X_SAFLAGS_AALG2 0x10000000 /* Auth-alg specif. flag 2 */
#define SADB_X_SAFLAGS_EALG1 0x8000000 /* Encr-alg specif. flag 1 */
#define SADB_X_SAFLAGS_EALG2 0x4000000 /* Encr-alg specif. flag 2 */
#define SADB_X_SAFLAGS_KM1 0x2000000 /* Key mgmt. specif. flag 1 */
#define SADB_X_SAFLAGS_KM2 0x1000000 /* Key mgmt. specif. flag 2 */
#define SADB_X_SAFLAGS_KM3 0x800000 /* Key mgmt. specif. flag 3 */
#define SADB_X_SAFLAGS_KM4 0x400000 /* Key mgmt. specif. flag 4 */
#define SADB_X_SAFLAGS_KRES1 0x200000 /* Reserved by the kernel */
#define SADB_X_SAFLAGS_NATT_LOC 0x100000 /* this has a natted srcSA */
#define SADB_X_SAFLAGS_NATT_REM 0x80000 /* this has a natted dstSA */
#define SADB_X_SAFLAGS_KRES2 0x40000 /* Reserved by the kernel */
#define SADB_X_SAFLAGS_TUNNEL 0x20000 /* tunnel mode */
#define SADB_X_SAFLAGS_PAIRED 0x10000 /* inbound/outbound pair*/
#define SADB_X_SAFLAGS_OUTBOUND 0x8000 /* SA direction bit */
#define SADB_X_SAFLAGS_INBOUND 0x4000 /* SA direction bit */

Sensitivity Extension flags:

#define SADB_X_SENS_IMPLICIT 0x1 /* implicit labelling */
#define SADB_X_SENS_UNLABELED 0x2 /* peer is unlabeled */

Extension headers include:

Generic Extension Header

struct sadb_ext {
uint16_t sadb_ext_len; /* In 64-bit words, inclusive */
uint16_t sadb_ext_type; /* 0 is reserved */
};
Security Association Information Extension
struct sadb_sa {
uint16_t sadb_sa_len;
uint16_t sadb_sa_exttype; /* ASSOCIATION */
uint32_t sadb_sa_spi;
uint8_t sadb_sa_replay;
uint8_t sadb_sa_state;
uint8_t sadb_sa_auth;
uint8_t sadb_sa_encrypt;
uint32_t sadb_sa_flags;
};
Lifetime Extension
struct sadb_lifetime {
uint16_t sadb_lifetime_len;
uint16_t sadb_lifetime_exttype; /* SOFT, HARD, CURRENT */

650 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

uint32_t sadb_lifetime_allocations;
uint64_t sadb_lifetime_bytes;
uint64_t sadb_lifetime_addtime;
uint64_t sadb_lifetime_usetime;
};
Address Extension
struct sadb_address {
uint16_t sadb_address_len;
uint16_t sadb_address_exttype; /* SRC, DST, NATT_*, INNER_* */
uint8_t sadb_address_proto; /* Proto for ports... */
uint8_t sadb_address_prefixlen; /* Prefix length for INNER_*. */
uint16_t sadb_address_reserved; /* Padding */
/* Followed by a sockaddr
structure.*/
};
Keying Material Extension
struct sadb_key {
uint16_t sadb_key_len;
uint16_t sadb_key_exttype; /* AUTH, ENCRYPT */
uint16_t sadb_key_bits;
uint16_t sadb_key_reserved;
/* Followed by actual key(s) in
canonical (outbound proc.) order. */
};
Indentity Extension
struct sadb_ident {
uint16_t sadb_ident_len;
uint16_t sadb_ident_exttype; /* SRC, DST, PROXY */
uint16_t sadb_ident_type; /* FQDN, USER_FQDN, etc. */
uint16_t sadb_ident_reserved; /* Padding */
uint64_t sadb_ident_id; /* For userid, etc. */
/* Followed by an identity null-terminate C string if present. */
};
Sensitivity/Integrity Extension
struct sadb_sens {
uint16_t sadb_sens_len;
uint16_t sadb_sens_exttype; /* SENSITIVITY, OUTER_SENS */
uint32_t sadb_sens_dpd;
uint8_t sadb_sens_sens_level;
uint8_t sadb_sens_sens_len; /* 64-bit words */
uint8_t sadb_sens_integ_level;
uint8_t sadb_sens_integ_len; /* 64-bit words */
uint32_t sadb_x_sens_flags;

Device and Network Interfaces 651

pf_key(7P)

/*
* followed by two uint64_t arrays
* uint64_t sadb_sens_bitmap[sens_bitmap_len];
* uint64_t integ_bitmap[integ_bitmap_len];
*/
};
Proposal Extension
struct sadb_prop {
uint16_t sadb_prop_len;
uint16_t sadb_prop_exttype; /* PROPOSAL, X_EPROP */
uint8_t sadb_prop_replay;
uint8_t sadb_X_prop_ereserved;
uint16_t sadb_x_prop_numecombs;
/* Followed by sadb_comb[] array or sadb_ecomb[] array. */
};
Combination Instance for a Proposal
struct sadb_comb {
uint8_t sadb_comb_auth;
uint8_t sadb_comb_encrypt;
uint16_t sadb_comb_flags;
uint16_t sadb_comb_auth_minbits;
uint16_t sadb_comb_auth_maxbits;
uint16_t sadb_comb_encrypt_minbits;
uint16_t sadb_comb_encrypt_maxbits;
uint32_t sadb_comb_reserved;
uint32_t sadb_comb_soft_allocations;
uint32_t sadb_comb_hard_allocations;
uint64_t sadb_comb_soft_bytes;
uint64_t sadb_comb_hard_bytes;
uint64_t sadb_comb_soft_addtime;
uint64_t sadb_comb_hard_addtime;
uint64_t sadb_comb_soft_usetime;
uint64_t sadb_comb_hard_usetime;
};
Extended Combination
struct sadb_x_ecomb {
uint8_t sadb_x_ecomb_numalgs;
uint8_t sadb_x_ecomb_reserved;
uint16_t sadb_x_ecomb_flags; /* E.g. PFS? */
uint32_t sadb_x_ecomb_reserved2;
uint32_t sadb_x_ecomb_soft_allocations;
uint32_t sadb_x_ecomb_hard_allocations;
uint64_t sadb_x_ecomb_soft_bytes;
uint64_t sadb_x_ecomb_hard_bytes;

652 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

uint64_t sadb_x_ecomb_soft_addtime;
uint64_t sadb_x_ecomb_hard_addtime;
uint64_t sadb_x_ecomb_soft_usetime;
uint64_t sadb_x_ecomb_hard_usetime;
};
Extended Combination Algorithm Descriptors
struct sadb_x_algdesc {
uint8_t sadb_x_algdesc_satype; /* ESP, AH, etc. */
uint8_t sadb_x_algdesc_algtype; /* AUTH, CRYPT, COMPRESS */
uint8_t sadb_x_algdesc_alg; /* DES, 3DES, MD5, etc. */
uint8_t sadb_x_algdesc_reserved;
uint16_t sadb_x_algdesc_minbits; /* Bit strengths. */
uint16_t sadb_x_algdesc_maxbits;
};
Extended Register
struct sadb_x_ereg {
uint16_t sadb_x_ereg_len;
uint16_t sadb_x_ereg_exttype; /* X_EREG */
uint8_t sadb_x_ereg_satypes[4]; /* Array of SA types, 0-terminated.
|};
Key Management Cookie
struct sadb_x_kmc {
uint16_t sadb_x_kmc_len;
uint16_t sadb_x_kmc_exttype; /* X_KM_COOKIE */
uint32_t sadb_x_kmc_proto; /* KM protocol */
uint32_t sadb_x_kmc_cookie; /* KMP-specific */
uint32_t sadb_x_kmc_reserved; /* Reserved; must be zero */
};
Supported Algorithms Extension
struct sadb_supported {
uint16_t sadb_supported_len;
uint16_t sadb_supported_exttype;
uint32_t sadb_supported_reserved;
};
Algorithm Instance
struct sadb_alg {
uint8_t sadb_alg_id; /* Algorithm type. */
uint8_t sadb_alg_ivlen; /* IV len, in bits */
uint16_t sadb_alg_minbits; /* Min. key len (in bits) */
uint16_t sadb_alg_maxbits; /* Max. key length */
uint16_t sadb_alg_reserved;
};

Device and Network Interfaces 653

pf_key(7P)

SPI Extension Range

struct sadb_spirange {
uint16_t sadb_spirange_len;
uint16_t sadb_spirange_exttype; /* SPI_RANGE */
uint32_t sadb_spirange_min
uint32_t sadb_spirange_max;
uint32_t sadb_spirange_reserved;
};
Security Association Pair Extension
struct sadb_x_pair {
uint16_t sadb_x_pair_len;
uint16_t sadb_x_pair_exttype; /* SADB_X_EXT_PAIR */
uint32_t sadb_x_pair_spi; /* SPI of paired SA */
};
Replay Value
struct sadb_x_replay_ctr {
uint16_t sadb_x_rc_len;
uint16_t sadb_x_rc_exttype;
uint32_t sadb_x_rc_replay32; /* For 240x SAs. */
uint64_t sadb_x_rc_replay64; /* For 430x SAs. */
};

Message Use and Each message has a behavior. A behavior is defined as where the initial message travels, for
Behavior example, user to kernel, and what subsequent actions are expected to take place. Contents of
messages are illustrated as:
<base, REQUIRED EXTENSION, REQ., (OPTIONAL EXTENSION,) (OPT)>

The SA extension is sometimes used only for its SPI field. If all other fields must be ignored,
this is represented by SA(*).

The lifetime extensions are represented with one to three letters after the word lifetime,
representing (H)ARD, (S)OFT, and (C)URRENT.

The address extensions are represented with one to three letters after the word address,
representing (S)RC, (D)ST, (Nl)NAT-T local, (Nr)NAT-T remote, (Is)Inner source, and
(Id)Inner destination.

Source and destination address extensions reflect outer-header selectors for an IPsec SA. An
SA is inbound or outbound depending on which of the source or destination address is local to
the node. Inner-source and inner-destination selectors represent inner-header selectors for
Tunnel Mode SAs. A Tunnel Mode SA must have either IPPROTO_ENCAP or
IPPROTO_IPV6 in its outer-headers as protocol selector, in addition to filled-in
Inner-address extensions.

654 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

NAT-T local and NAT-T remote addresses store local and remote ports used for ESP-in-UDP
encapsulation. A non-zero local NAT-T address extension represents the local node's external
IP address if it is not equivalent to the SA's local address. A non-zero remote NAT-T address
represents a peer's behind-a-NAT address if it is not equivalent to the SA's remote address. An
SA with NAT-T extensions protects-and-transmits outbound traffic. Processing of inbound
NAT-T traffic requires a UDP socket bound to the appropriate local port and it must have the
UDP_NAT_T_ENDPOINT (see udp(7P)) socket option enabled.

Note that when an error occurs, only the base header is sent. In the event of an error, an
extended diagnostic can be set (see DIAGNOSTICS). Typical errors include:
EINVAL Various message improprieties, including SPI ranges that are malformed, weak
keys, and others. If EINVAL is returned, an application should look at the
sadb_x_msg_diagnostic field of the sadb_msg structure. It contains one of
many possible causes for EINVAL. See net/pfkeyv2.h for values, all of the form
SADB_X_DIAGNOSTIC_.
ENOMEM Needed memory was not available.
ENSGSIZ Message exceeds the maximum length allowed.
EEXIST SA (that is being added or created with GETSPI) already exists.
ESRCH SA could not be found.

The following are examples of message use and behavior:

SADB_GETSPI
Send a SADB_GETSPI message from a user process to the kernel.
<base, address, SPI range>

The kernel returns the SADB_GETSPI message to all listening processes.

<base, SA(*), address (SD)>

SADB_UPDATE
Send a SADB_UPDATE message from a user process to the kernel.
<base, SA, (lifetime(HS),) address(SD), (address(Is,Id),
address(Nl,Nr),key (AE), (identity(SD),) (sensitivity, outer sensitivity)>

The kernel returns the SADB_UPDATE message to all listening processes.

<base, SA(*), address (SD), (pair)>

Adding a sadb_x_pair extension to an SADB_UPDATE or SADB_ADD message updates the

security association pair linkage with the SPI of the security association contained in that

Device and Network Interfaces 655

pf_key(7P)

extension. The resulting security association pair can be updated or as a single entity using the
SADB_X_UPDATEPAIR or SADB_X_DELPAIR message types.

SADB_ADD
Send a SADB_ADD message from a user process to the kernel.
<base, SA, (lifetime(HS),) address(SD), (address(Is,Id),)
(address(Nl,Nr),) key(AE), (identity(SD),) (sensitivity, outer sensitivity) (pair)>

The kernel returns the SADB_ADD message to all listening processes.

<base, SA, (lifetime(HS),) address (SD), (address(Is,Id),)
(address(Nl,Nr),) (identity (SD),) (sensitivity, outer sensitivity)>

SADB_X_UPDATEPAIR
Send a SADB_X_UPDATEPAIR message from a user process to the kernel. This message type is
used to update the lifetime values of a security association and the lifetime values of the
security association it is paired with.
<base, SA, lifetime(HS), address(SD)>

SADB_DELETE | SADB_X_DELPAIR
Send a SADB_DELETE message from a user process to the kernel. The SADB_X_DELPAIR message
type requests deletion of the security association and the security association it is paired with.
<base, SA (*), address (SD)>

The kernel returns the SADB_DELETE message to all listening processes.

<base, SA (*), address (SD)>

SADB_GET
Send a SADB_GET message from a user process to the kernel.
<base, SA (*), address (SD)>

The kernel returns the SADB_GET message to the socket that sent the SADB_GET message.
<base, SA , (lifetime (HSC),) address SD), (address (P),) key (AE),
(identity (SD),) (sensitivity, outer sensitivity)>

SADB_ACQUIRE
The kernel sends a SADB_ACQUIRE message to registered sockets. Note that any GETSPI, ADD, or
UPDATE calls in reaction to an ACQUIRE must fill in the sadb_msg_seq of those messages with
the one in the ACQUIRE message. The address (SD) extensions must have the port fields filled in
with the port numbers of the session requiring keys if appropriate.

656 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

<base, address (SD), (address(Is,Id)), (identity(SD),)

(sensitivity) proposal>

Extended ACQUIRE has a slightly different format. The sadb_msg_satype field is 0, and the
extension contains the desired combination(s) of security protocols.
<base, address (SD), (address(Is,Id)), (identity(SD),)
(sensitivity,) eprop>

If key management fails, send an SADB_ACQUIRE to indicate failure.

<base>

SADB_X_INVERSE_ACQUIRE
For inbound Key Management processing, a Key Management application can wish to consult
the kernel for its policy. The application should send to the kernel:
<base, address (SD), (address(Is,Id))>

The kernel returns a message similar to a kernel-generated extended ACQUIRE:

<base, address (SD), (address(Is,Id)), (identity(SD),)
(sensitivity,) eprop>

SADB_REGISTER
Send a SADB_REGISTER message from a user process to the kernel.
<base>

The kernel returns the SADB_REGISTER message to registered sockets, with algorithm types
supported by the kernel being indicated in the supported algorithms field. Note that this
message can arrive asynchronously due to an algorithm being loaded or unloaded into a
dynamically linked kernel.
<base, supported>

There is also the extended REGISTER, which allows this process to receive extended
ACQUIREs.
<base, ereg>

Which returns a series of SADB_REGISTER replies (one for each security protocol registered)
from the kernel.

SADB_EXPIRE
The kernel sends a SADB_EXPIRE message to all listeners when the soft limit of a security
association has been expired.
<base, SA, lifetime (C and one of HS), address (SD)>

Device and Network Interfaces 657

pf_key(7P)

SADB_FLUSH
Send a SADB_FLUSH message from a user process to the kernel.
<base>

The kernel returns the SADB_FLUSH message to all listening sockets.

<base>

SADB_DUMP
Send a SADB_DUMP message from a user process to the kernel.
<base>

Several SADB_DUMP messages returns from the kernel to the sending socket.
<base, SA, (lifetime (HSC),) address (SD), (address (Is,Id),)
(address (Nl,Nr),) key (AE), (identity (SD),) sensitivity, outer sensitivity)>

To mark the end of a dump a single base header arrives with its sadb_mdg_seq set to 0.
<base>

SADB_X_PROMISC
Send a SADB_X_PROMISC message from a user process to the kernel.
<base>

The kernel returns the SADB_X_PROMISC message to all listening processes.

<base>

Diagnostics The message returning from the kernel contains a diagnostic value in the base message header,
the diagnostic value indicates if action requested by the original message was a success.

Diagnostic Values:
#define SADB_X_DIAGNOSTIC_NONE 0
#define SADB_X_DIAGNOSTIC_UNKNOWN_MSG 1
#define SADB_X_DIAGNOSTIC_UNKNOWN_EXT 2
#define SADB_X_DIAGNOSTIC_BAD_EXTLEN 3
#define SADB_X_DIAGNOSTIC_UNKNOWN_SATYPE 4
#define SADB_X_DIAGNOSTIC_SATYPE_NEEDED 5
#define SADB_X_DIAGNOSTIC_NO_SADBS 6
#define SADB_X_DIAGNOSTIC_NO_EXT 7
/* Bad address family value */
#define SADB_X_DIAGNOSTIC_BAD_SRC_AF 8
/* in sockaddr->sa_family. */
#define SADB_X_DIAGNOSTIC_BAD_DST_AF 9
/* These two are synonyms. */

658 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

#define SADB_X_DIAGNOSTIC_BAD_PROXY_AF 10
#define SADB_X_DIAGNOSTIC_BAD_INNER_SRC_AF 10

#define SADB_X_DIAGNOSTIC_AF_MISMATCH 11

#define SADB_X_DIAGNOSTIC_BAD_SRC 12
#define SADB_X_DIAGNOSTIC_BAD_DST 13

#define SADB_X_DIAGNOSTIC_ALLOC_HSERR 14
#define SADB_X_DIAGNOSTIC_BYTES_HSERR 15
#define SADB_X_DIAGNOSTIC_ADDTIME_HSERR 16
#define SADB_X_DIAGNOSTIC_USETIME_HSERR 17

#define SADB_X_DIAGNOSTIC_MISSING_SRC 18
#define SADB_X_DIAGNOSTIC_MISSING_DST 19
#define SADB_X_DIAGNOSTIC_MISSING_SA 20
#define SADB_X_DIAGNOSTIC_MISSING_EKEY 21
#define SADB_X_DIAGNOSTIC_MISSING_AKEY 22
#define SADB_X_DIAGNOSTIC_MISSING_RANGE 23

#define SADB_X_DIAGNOSTIC_DUPLICATE_SRC 24
#define SADB_X_DIAGNOSTIC_DUPLICATE_DST 25
#define SADB_X_DIAGNOSTIC_DUPLICATE_SA 26
#define SADB_X_DIAGNOSTIC_DUPLICATE_EKEY 27
#define SADB_X_DIAGNOSTIC_DUPLICATE_AKEY 28
#define SADB_X_DIAGNOSTIC_DUPLICATE_RANGE 29

#define SADB_X_DIAGNOSTIC_MALFORMED_SRC 30
#define SADB_X_DIAGNOSTIC_MALFORMED_DST 31
#define SADB_X_DIAGNOSTIC_MALFORMED_SA 32
#define SADB_X_DIAGNOSTIC_MALFORMED_EKEY 33
#define SADB_X_DIAGNOSTIC_MALFORMED_AKEY 34
#define SADB_X_DIAGNOSTIC_MALFORMED_RANGE 35

#define SADB_X_DIAGNOSTIC_AKEY_PRESENT 36
#define SADB_X_DIAGNOSTIC_EKEY_PRESENT 37
#define SADB_X_DIAGNOSTIC_PROP_PRESENT 38
#define SADB_X_DIAGNOSTIC_SUPP_PRESENT 39
#define SADB_X_DIAGNOSTIC_BAD_AALG 40
#define SADB_X_DIAGNOSTIC_BAD_EALG 41
#define SADB_X_DIAGNOSTIC_BAD_SAFLAGS 42
#define SADB_X_DIAGNOSTIC_BAD_SASTATE 43

#define SADB_X_DIAGNOSTIC_BAD_AKEYBITS 44
#define SADB_X_DIAGNOSTIC_BAD_EKEYBITS 45

#define SADB_X_DIAGNOSTIC_ENCR_NOTSUPP 46

Device and Network Interfaces 659

pf_key(7P)

#define SADB_X_DIAGNOSTIC_WEAK_EKEY 47
#define SADB_X_DIAGNOSTIC_WEAK_AKEY 48

#define SADB_X_DIAGNOSTIC_DUPLICATE_KMP 49
#define SADB_X_DIAGNOSTIC_DUPLICATE_KMC 50

#define SADB_X_DIAGNOSTIC_MISSING_NATT_LOC 51
#define SADB_X_DIAGNOSTIC_MISSING_NATT_REM 52
#define SADB_X_DIAGNOSTIC_DUPLICATE_NATT_LOC 53
#define SADB_X_DIAGNOSTIC_DUPLICATE_NATT_REM 54
#define SADB_X_DIAGNOSTIC_MALFORMED_NATT_LOC 55
#define SADB_X_DIAGNOSTIC_MALFORMED_NATT_REM 56
#define SADB_X_DIAGNOSTIC_DUPLICATE_NATT_PORTS 57

#define SADB_X_DIAGNOSTIC_MISSING_INNER_SRC 58
#define SADB_X_DIAGNOSTIC_MISSING_INNER_DST 59
#define SADB_X_DIAGNOSTIC_DUPLICATE_INNER_SRC 60
#define SADB_X_DIAGNOSTIC_DUPLICATE_INNER_DST 61
#define SADB_X_DIAGNOSTIC_MALFORMED_INNER_SRC 62
#define SADB_X_DIAGNOSTIC_MALFORMED_INNER_DST 63

#define SADB_X_DIAGNOSTIC_PREFIX_INNER_SRC 64
#define SADB_X_DIAGNOSTIC_PREFIX_INNER_DST 65
#define SADB_X_DIAGNOSTIC_BAD_INNER_DST_AF 66
#define SADB_X_DIAGNOSTIC_INNER_AF_MISMATCH 67

#define SADB_X_DIAGNOSTIC_BAD_NATT_REM_AF 68
#define SADB_X_DIAGNOSTIC_BAD_NATT_LOC_AF 69

#define SADB_X_DIAGNOSTIC_PROTO_MISMATCH 70
#define SADB_X_DIAGNOSTIC_INNER_PROTO_MISMATCH 71

#define SADB_X_DIAGNOSTIC_DUAL_PORT_SETS 72

#define SADB_X_DIAGNOSTIC_PAIR_INAPPROPRIATE 73
#define SADB_X_DIAGNOSTIC_PAIR_ADD_MISMATCH 74
#define SADB_X_DIAGNOSTIC_PAIR_ALREADY 75
#define SADB_X_DIAGNOSTIC_PAIR_SA_NOTFOUND 76
#define SADB_X_DIAGNOSTIC_BAD_SA_DIRECTION 77

#define SADB_X_DIAGNOSTIC_SA_NOTFOUND 78
#define SADB_X_DIAGNOSTIC_SA_EXPIRED 79
#define SADB_X_DIAGNOSTIC_BAD_CTX 80
#define SADB_X_DIAGNOSTIC_INVALID_REPLAY 81
#define SADB_X_DIAGNOSTIC_MISSING_LIFETIME 82
#define SADB_X_DIAGNOSTIC_BAD_LABEL 83

660 man pages section 7: Device and Network Interfaces • Last Revised 12 Nov 2009
 pf_key(7P)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsr

Interface Stability Committed

See Also in.iked(1M), ipseckey(1M), ipsec(7P), ipsecah(7P), ipsecesp(7P), route(7P), udp(7P)

McDonald, D.L., Metz, C.W., and Phan, B.G., RFC 2367, PF_KEY Key Management API,
Version 2, The Internet Society, July 1998.

Notes Time-based lifetimes might not expire with exact precision in seconds because kernel load can
affect the aging of SA's.

Device and Network Interfaces 661

pfmod(7M)

Name pfmod – STREAMS Packet Filter Module

Synopsis #include <sys/pfmod.h>
ioctl(fd, IPUSH, "pfmod");

Description pfmod is a STREAMS module that subjects messages arriving on its read queue to a packet
filter and passes only those messages that the filter accepts on to its upstream neighbor. Such
filtering can be very useful for user-level protocol implementations and for networking
monitoring programs that wish to view only specific types of events.
Read-side Behavior pfmod applies the current packet filter to all M_DATA and M_PROTO messages arriving on its read
queue. The module prepares these messages for examination by first skipping over all leading
M_PROTO message blocks to arrive at the beginning of the message's data portion. If there is no
data portion, pfmod accepts the message and passes it along to its upstream neighbor.
Otherwise, the module ensures that the part of the message's data that the packet filter might
examine lies in contiguous memory, calling the pullupmsg(9F) utility routine if necessary to
force contiguity. (Note: this action destroys any sharing relationships that the subject message
might have had with other messages.) Finally, it applies the packet filter to the message's data,
passing the entire message upstream to the next module if the filter accepts, and discarding the
message otherwise. See PACKET FILTERS below for details on how the filter works.
If there is no packet filter yet in effect, the module acts as if the filter exists but does nothing,
implying that all incoming messages are accepted. The ioctls section below describes how to
associate a packet filter with an instance of pfmod.
pfmod passes all other messages through unaltered to its upper neighbor.
Write-side Behavior pfmod intercepts M_IOCTL messages for the ioctl described below. The module passes all other
messages through unaltered to its lower neighbor.
ioctls pfmod responds to the following ioctl.
PFIOCSETF This ioctl directs the module to replace its current packet filter, if any, with the
filter specified by the struct packetfilt pointer named by its final
argument. This structure is defined in <sys/pfmod.h> as:
struct packetfilt {
uchar_t Pf_Priority; /* priority of filter */
uchar_t Pf_FilterLen; /* length of filter cmd list */
ushort_t Pf_Filter[ENMAXFILTERS]; /* filter command list */
};

The Pf_Priority field is included only for compatibility with other packet filter
implementations and is otherwise ignored. The packet filter itself is specified in the Pf_Filter
array as a sequence of two-byte commands, with the Pf_FilterLen field giving the number of
commands in the sequence. This implementation restricts the maximum number of
commands in a filter (ENMAXFILTERS) to 255. The next section describes the available
commands and their semantics.

662 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2006
 pfmod(7M)

Packet Filters A packet filter consists of the filter command list length (in units of ushort_ts), and the filter
command list itself. (The priority field mentioned above is ignored in this implementation.)
Each filter command list specifies a sequence of actions that operate on an internal stack of
ushort_ts ("shortwords") or an offset register. The offset register is initially zero. Each
shortword of the command list specifies an action and a binary operator. Using _n_ as
shorthand for the next shortword of the instruction stream and _%oreg_ for the offset register,
the list of actions is:
COMMAND SHORTWORDS ACTION
ENF_PUSHLIT 2 Push _n_ on the stack.
ENF_PUSHZERO 1 Push zero on the stack.
ENF_PUSHONE 1 Push one on the stack.
ENF_PUSHFFFF 1 Push 0xFFFF on the stack.
ENF_PUSHFF00 1 Push 0xFF00 on the stack.
ENF_PUSH00FF 1 Push 0x00FF on the stack.
ENF_LOAD_OFFSET 2 Load _n_ into _%oreg_.
ENF_BRTR 2 Branch forward _n_ shortwords if
the top element of the stack is
non-zero.
ENF_BRFL 2 Branch forward _n_ shortwords if
the top element of the stack is zero.
ENF_POP 1 Pop the top element from the stack.
ENF_PUSHWORD+m 1 Push the value of shortword (_m_ +
_%oreg_) of the packet onto the stack.

The binary operators can be from the set {ENF_EQ, ENF_NEQ, ENF_LT, ENF_LE, ENF_GT,ENF_GE,
ENF_AND, ENF_OR, ENF_XOR} which operate on the top two elements of the stack and replace
them with its result.

When both an action and operator are specified in the same shortword, the action is
performed followed by the operation.

The binary operator can also be from the set {ENF_COR, ENF_CAND, ENF_CNOR, ENF_CNAND}.
These are “short-circuit” operators, in that they terminate the execution of the filter
immediately if the condition they are checking for is found, and continue otherwise. All pop
two elements from the stack and compare them for equality; ENF_CAND returns false if the
result is false; ENF_COR returns true if the result is true; ENF_CNAND returns true if the result is
false; ENF_CNOR returns false if the result is true. Unlike the other binary operators, these four
do not leave a result on the stack, even if they continue.

The short-circuit operators should be used when possible, to reduce the amount of time spent
evaluating filters. When they are used, you should also arrange the order of the tests so that the
filter will succeed or fail as soon as possible; for example, checking the IP destination field of a
UDP packet is more likely to indicate failure than the packet type field.

The special action ENF_NOPUSH and the special operator ENF_NOP can be used to only perform
the binary operation or to only push a value on the stack. Since both are (conveniently)

Device and Network Interfaces 663

pfmod(7M)

defined to be zero, indicating only an action actually specifies the action followed by ENF_NOP,
and indicating only an operation actually specifies ENF_NOPUSH followed by the operation.

After executing the filter command list, a non-zero value (true) left on top of the stack (or an
empty stack) causes the incoming packet to be accepted and a zero value (false) causes the
packet to be rejected. (If the filter exits as the result of a short-circuit operator, the top-of-stack
value is ignored.) Specifying an undefined operation or action in the command list or
performing an illegal operation or action (such as pushing a shortword offset past the end of
the packet or executing a binary operator with fewer than two shortwords on the stack) causes
a filter to reject the packet.

Examples The packet filter module is not dependent on any particular device driver or module but is
commonly used with datalink drivers such as the Ethernet driver. If the underlying datalink
driver supports the Data Link Provider Interface (DLPI) message set, the appropriate
STREAMS DLPI messages must be issued to attach the stream to a particular hardware device
and bind a datalink address to the stream before the underlying driver will route received
packets upstream. Refer to the DLPI Version 2 specification for details on this interface.

The reverse ARP daemon program may use code similar to the following fragment to
construct a filter that rejects all but RARP packets. That is, it accepts only packets whose
Ethernet type field has the value ETHERTYPE_REVARP. The filter works whether a VLAN is
configured or not.
struct ether_header eh; /* used only for offset values */
struct packetfilt pf;
register ushort_t *fwp = pf.Pf_Filter;
ushort_t offset;
int fd;
/*
* Push packet filter streams module.
*/
if (ioctl(fd, I_PUSH, "pfmod") < 0)
syserr("pfmod");

/*
* Set up filter. Offset is the displacement of the Ethernet
* type field from the beginning of the packet in units of
* ushort_ts.
*/
offset = ((uint_t) &eh.ether_type - (uint_t) &eh.ether_dhost) /
sizeof (us_short);
*fwp++ = ENF_PUSHWORD + offset;
*fwp++ = ENF_PUSHLIT;
*fwp++ = htons(ETHERTYPE_VLAN);
*fwp++ = ENF_EQ;
*fwp++ = ENF_BRFL;
*fwp++ = 3; /* If this isn’t ethertype VLAN, don’t change oreg */

664 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2006
 pfmod(7M)

*fwp++ = ENF_LOAD_OFFSET;
*fwp++ = 2; /* size of the VLAN tag in words */
*fwp++ = ENF_POP;
*fwp++ = ENF_PUSHWORD + offset;
*fwp++ = ENF_PUSHLIT;
*fwp++ = htons(ETHERTYPE_REVARP);
*fwp++ = ENF_EQ;
pf.Pf_FilterLen = fwp - &pf.PF_Filter[0];

This filter can be abbreviated by taking advantage of the ability to combine actions and
operations:
*fwp++ = ENF_PUSHWORD + offset;
*fwp++ = ENF_PUSHLIT | ENF_EQ;
*fwp++ = htons(ETHERTYPE_REVARP);
*fwp++ = htons(ETHERTYPE_VLAN);
*fwp++ = ENF_BRFL | ENF_NOP;
*fwp++ = 3;
*fwp++ = ENF_LOAD_OFFSET | ENF_NOP;
*fwp++ = 2;
*fwp++ = ENF_POP | ENF_NOP;
*fwp++ = ENF_PUSHWORD + offset;
*fwp++ = ENF_PUSHLIT | ENF_EQ;
*fwp++ = htons(ETHERTYPE_REVARP);

See Also bufmod(7M), dlpi(7P), pullupmsg(9F)

Device and Network Interfaces 665

physmem(7D)

Name physmem – physmem driver

Description The physmem driver is a private mechanism used by diagnostic test suites to test the physical
memory of the system.
Files /dev/physmem
Kernel module.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface stability Private

Caution This driver is used by Sun internal diagnostic programs only. Any other use may have a
harmful impact on the system.

666 man pages section 7: Device and Network Interfaces • Last Revised 15 Nov 2006
 pipemod(7M)

Name pipemod – STREAMS pipe flushing module

Description The typical stream is composed of a stream head connected to modules and terminated by a
driver. Some stream configurations such as pipes and FIFOs do not have a driver and hence
certain features commonly supported by the driver need to be provided by other means.
Flushing is one such feature, and it is provided by the pipemod module.

Pipes and FIFOs in their simplest configurations only have stream heads. A write side is
connected to a read side. This remains true when modules are pushed. The twist occurs at a
point known as the mid-point. When an M_FLUSH message is passed from a write queue to a
read queue the FLUSHR and/or FLUSHW bits have to be switched. The mid-point of a pipe is not
always easily detectable, especially if there are numerous modules pushed on either end of the
pipe. In that case there needs to be a mechanism to intercept all message passing through the
stream. If the message is an M_FLUSH message and it is at the mid-point, the flush bits need to
be switched. This bit switching is handled by the pipemod module.

pipemod should be pushed onto a pipe or FIFO where flushing of any kind will take place. The
pipemod module can be pushed on either end of the pipe. The only requirement is that it is
pushed onto an end that previously did not have modules on it. That is, pipemod must be the
first module pushed onto a pipe so that it is at the mid-point of the pipe itself.

The pipemod module handles only M_FLUSH messages. All other messages are passed on to the
next module using the putnext() utility routine. If an M_FLUSH message is passed to pipemod
and the FLUSHR and FLUSHW bits are set, the message is not processed but is passed to the next
module using the putnext() routine. If only the FLUSHR bit is set, the FLUSHR bit is turned off
and the FLUSHW bit is set. The message is then passed on to the next module using putnext().
Similarly, if the FLUSHW bit is the only bit set in the M_FLUSH message, the FLUSHW bit is turned
off and the FLUSHR bit is turned on. The message is then passed to the next module on the
stream.

The pipemod module can be pushed on any stream that desires the bit switching. It must be
pushed onto a pipe or FIFO if any form of flushing must take place.

See Also STREAMS Programming Guide

Device and Network Interfaces 667

pm(7D)

Name pm – Power Management driver

Synopsis /dev/pm

Description The Power Management ( pm) driver provides an interface for applications to configure
devices within the system for Power Management. The interface is provided through ioctl(2)
commands. The pm driver may be accessed using /dev/pm.

Power Management The Power Management framework model allows the system to be viewed as a collection of
Framework devices. Each device is a collection of components that comprise the smallest power
manageable units. The device driver controls the definition of a device's power manageable
components.

A component can either be busy or idle at the current power level. Normally, the Power
Management framework takes an idle component to the next lower power level. The Power
Management framework uses two factors to determine this transition: the component must
have been idle for at least the threshold time, and the device to which the component belongs
must satisfy any dependency requirements. A dependency occurs when a device requires
another device to be power managed before it can be power managed. Dependencies occur on
a per device basis: when a dependency exists, no components of a device may be managed
unless all the devices it depends upon are first power managed.

Using the commands below, an application may take control of the Power Management of a
device from the Power Management framework driver and manage the transition of device
power levels directly.

For this set of ioctl commands, arg (see ioctl(2)) points to a structure of type pm_req defined
in <sys/pm.h>:
typedef struct pm_req {
char *physpath; /* physical path of device */
/* to configure. See libdevinfo(3LIB) */
int component; /* device component */
int value; /* power level, threshold value, or count */
void *data; /* command-dependent variable-sized data */
size_t datasize; /* size of data buffer */
} pm_req_t;

The fields should contain the following data:

physpath Pointer to the physical path of a device. See libdevinfo(3LIB). For example,
for the device /devices/pseudo/pm@0:pm the physpath value would be
/pseudo/pm@0.
component Non-negative integer specifying which component is being configured. The
numbering starts at zero.
value Non-negative integer specifying the threshold value in seconds or the desired
power level, or the number of levels being specified.

668 man pages section 7: Device and Network Interfaces • Last Revised 20 Sep 1999
 pm(7D)

data Pointer to a buffer which contains or receives variable-sized data, such as the
name of a device upon which this device has a dependency.
size Size of the data buffer.

Not all fields are used in each command.

PM_DIRECT_PM The device named by physpath is disabled
from being power managed by the
framework. The caller will power manage
the device directly using the
PM_DIRECT_NOTIFY, PM_GET_TIME_IDLE
and PM_GET_CURRENT_POWER,
PM_GET_FULL_POWER and
PM_SET_CURRENT_POWER commands. If the
device needs to have its power level
changed either because its driver calls
pm_raise_power(9F),
pm_lower_power(9F), or
pm_power_has_changed(9F) or because the
device is the parent of another device that is
changing power level or a device that this
device depends on is changing power level,
then the power level change of the device
will be blocked and the caller will be
notified as described below for the
PM_DIRECT_NOTIFY command.

Error codes:
EBUSY Device already disabled for
Power Management by
framework.
EPERM Caller is neither superuser nor
effective group ID of 0.
PM_RELEASE_DIRECT_PM The device named by physpath (which
must have been the target of a
PM_DIRECT_PM command) is re-enabled for
Power Management by the framework.

Error codes:
EINVAL Device component out of
range.

Device and Network Interfaces 669

pm(7D)

PM_DIRECT_NOTIFY PM_DIRECT_NOTIFY_WAIT These commands allow the process that is

directly power managing a device to be
notified of events that could change the
power level of the device. When such an
event occurs, this command returns
information about the event.

arg (see ioctl(2)) points to a structure of

type pm_state_change defined in
<sys/pm.h>:
typedef struct pm_state_change {
char *physpath; /* device which has changed state */
int component; /* which component changed state */
#if defined(_BIG_ENDIAN)
ushort_t flags; /* PSC_EVENT_LOST, PSC_ALL_LOWEST */
ushort_t event; /* type of event */
#else
ushort_t event; /* type of event *
ushort_t flags; /* PSC_EVENT_LOST, PSC_ALL_LOWEST */
#endif
time_t timestamp; /* time of state change */+
int old_level; /* power level changing from */
int new_level; /* power level changing to */
size_t size; /* size of buffer physpath points to */
} pm_state_change_t;
When an event occurs, the struct pointed
to by arg is filled in. If the event type is
PSC_PENDING_CHANGE, then the
information in the rest of the struct
describes an action that the framework
would have taken if the device were not
directly power managed by the caller. The
caller is responsible for completing the
indicated level changes using
PM_SET_CURRENT_POWER below.

An event type of PSC_HAS_CHANGED

indicates that the driver for the directly
power managed device has called
pm_power_has_changed(9F) due to the
device changing power on its own. It is
provided to allow the caller to track the
power state of the device.

670 man pages section 7: Device and Network Interfaces • Last Revised 20 Sep 1999
 pm(7D)

The system keeps events in a circular

buffer. If the buffer overflow, the oldest
events are lost and when the event that next
follows a lost event is retrieved it will have
PSC_EVENT_LOST set in flags.

PM_DIRECT_NOTIFY returns EWOULDBLOCK if

no event is pending, and
PM_DIRECT_NOTIFY_WAIT blocks until an
event is available.

pm also supports the poll(2) interface.

When an event is pending a poll(2) call
that includes a file descriptor for /dev/pm
and that has POLLIN or POLLRDNORM set in
its event mask will return.
PM_SET_CURRENT_POWER Component component of the device
named by physpath (which must contain
the physical path of a device against which
the process has issued a PM_DIRECT_PM
command) is set to power level value. If all
components of the device named by
physpath were at level 0, value is non-zero
and some device has a dependency on this
device, then all components of that device
will be brought to full power before this
command returns. Similarly, if the parent
of the target device is powered off, then it
will be brought up as needed before this
command returns. When
PM_SET_CURRENT_POWER is issued
against a device, the resulting power
change is included in the event list for
PM_DIRECT_NOTIFY.

Error codes:
EINVAL Device component out of
range, or power level < 0.
EIO Failed to power device or its
ancestors or the devices on
which this device has
dependency or their ancestors.
Note that this may not indicate

Device and Network Interfaces 671

pm(7D)

a failure, the device driver may

have rejected the command as
inappropriate because the
component has become busy.
EPERM Caller has not previously issued
a successful PM_DIRECT_PM
command against this device.
PM_GET_FULL_POWER The highest supported power level of
component component of the device
named by physpath is returned.
PM_GET_CURRENT_POWER The current power level of component
component of the device named by
physpath is returned.

Error codes:
EAGAIN Device component power level
is not currently known.
PM_GET_TIME_IDLE PM_GET_TIME_IDLE returns the number of
seconds that component component of the
device named by physpath has been idle. If
the device is not idle, then 0 is returned.

Note that because the state of the device

may change between the time the process
issues the PM_GET_TIME_IDLE command
and the time the process issues a
PM_SET_CURRENT_POWER command to
reduce the power level of an idle
component, the process must be prepared
to deal with a PM_SET_CURRENT_POWER
command returning failure because the
driver has rejected the command as
inappropriate because the device
component has become busy. This can be
differentiated from other types of failures
by issuing the PM_GET_TIME_IDLE
command again to see if the component
has become busy.

Errors Upon error, the commands will return −1, and set errno. In addition to the error codes listed
above by command, the following error codes are common to all commands:

672 man pages section 7: Device and Network Interfaces • Last Revised 20 Sep 1999
 pm(7D)

EFAULT Bad address passed in as argument.

ENODEV Device is not power manageable, or device is not configured.
ENXIO Too many opens attempted.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface stability Unstable

See Also pmconfig(1M), Intro(2), ioctl(2), libdevinfo(3LIB), power.conf(4), attributes(5),

attach(9E), detach(9E), power(9E), pm_busy_component(9F), pm_idle_component(9F),
pm_lower_power(9F), pm_power_has_changed(9F), pm_raise_power(9F)

Writing Device Drivers

Device and Network Interfaces 673

poll(7d)

Name poll – driver for fast poll on many file descriptors

Synopsis #include <sys/devpoll.h>
int fd = open("/dev/poll", O_RDWR);
ssize_t n = write(int fd, struct pollfd buf[], int bufsize);
int n = ioctl(int fd, DP_POLL, struct dvpoll* arg);
int n = ioctl(int fd, DP_ISPOLLED, struct pollfd* pfd);
Parameters fd Open file descriptor that refers to the /dev/poll driver.
path /dev/poll
buf Array of pollfd structures.
bufsize Size of buf in bytes.
arg Pointer to pollcall structure.
pfd Pointer to pollfd structure.

Description Note – The /dev/poll device, associated driver and corresponding manpages may be removed
in a future Solaris release. For similar functionality in the event ports framework, see
port_create(3C).

The /dev/poll driver is a special driver that enables you to monitor multiple sets of polled file
descriptors. By using the /dev/poll driver, you can efficiently poll large numbers of file
descriptors. Access to the /dev/poll driver is provided through open(2), write(2), and
ioctl(2) system calls.

Writing an array of pollfd struct to the /dev/poll driver has the effect of adding these file
descriptors to the monitored poll file descriptor set represented by the fd. To monitor
multiple file descriptor sets, open the /dev/poll driver multiple times. Each fd corresponds
to one set. For each pollfd struct entry (defined in sys/poll.h):
struct pollfd {
int fd;
short events;
short revents;
}

The fd field specifies the file descriptor being polled. The events field indicates the interested
poll events on the file descriptor. If a pollfd array contains multiple pollfd entries with the
same fd field, the "events" field in each pollfd entry is OR'ed. A special POLLREMOVE event in
the events field of the pollfd structure removes the fd from the monitored set. The revents
field is not used. Write returns the number of bytes written successfully or -1 when write fails.

The DP_POLL ioctl is used to retrieve returned poll events occured on the polled file
descriptors in the monitored set represented by fd. arg is a pointer to the devpoll structures
which are defined as follows:

674 man pages section 7: Device and Network Interfaces • Last Revised 28 Mar 2007
 poll(7d)

struct dvpoll {
struct pollfd* dp_fds;
int dp_nfds;
int dp_timeout;
}

The dp_fds points to a buffer that holds an array of returned pollfd structures. The dp_nfds
field specifies the size of the buffer in terms of the number of pollfd entries it contains. The
dp_nfds field also indicates the maximum number of file descriptors from which poll
information can be obtained. If there is no interested events on any of the polled file
descriptors, the DP_POLL ioctl call will wait dp_timeout milliseconds before returning. If
dp_timeout is 0, the ioctl call returns immediately. If dp_timeout is -1, the call blocks until an
interested poll events is available or the call is interrupted. Upon return, if the ioctl call has
failed, -1 is returned. The memory content pointed by dp_fds is not modified. A return value
0 means the ioctl is timed out. In this case, the memory content pointed by dp_fds is not
modified. If the call is successful, it returns the number of valid pollfd entries in the array
pointed by dp_fds; the contents of the rest of the buffer is undefined. For each valid pollfd
entry, the fd field indicates the file desciptor on which the polled events happened. The
events field is the user specified poll events. The revents field contains the events
occurred. –1 is returned if the call fails.

DP_ISPOLLED ioctl allows you to query if a file descriptor is already in the monitored set
represented by fd. The fd field of the pollfd structure indicates the file descriptor of interest.
The DP_ISPOLLED ioctl returns 1 if the file descriptor is in the set. The events field contains 0.
The revents field contains the currently polled events. The ioctl returns 0 if the file
descriptor is not in the set. The pollfd structure pointed by pfd is not modified. The ioctl
returns a -1 if the call fails.

Examples The following example shows how /dev/poll may be used.

{
...
/*
* open the driver
*/
if ((wfd = open("/dev/poll", O_RDWR)) < 0) {
exit(-1);
}
pollfd = (struct pollfd* )malloc(sizeof(struct pollfd) * MAXBUF);
if (pollfd == NULL) {
close(wfd);
exit(-1);
}
/*
* initialize buffer
*/
for (i = 0; i < MAXBUF; i++) {

Device and Network Interfaces 675

poll(7d)

pollfd[i].fd = fds[i];
pollfd[i].events = POLLIN;
pollfd[i].revents = 0;
}
if (write(wfd, &pollfd[0], sizeof(struct pollfd) * MAXBUF) !=
sizeof(struct pollfd) * MAXBUF) {
perror("failed to write all pollfds");
close (wfd);
free(pollfd);
exit(-1);
}
/*
* read from the devpoll driver
*/
dopoll.dp_timeout = -1;
dopoll.dp_nfds = MAXBUF;
dopoll.dp_fds = pollfd;
result = ioctl(wfd, DP_POLL, &dopoll);
if (result < 0) {
perror("/dev/poll ioctl DP_POLL failed");
close (wfd);
free(pollfd);
exit(-1);
}
for (i = 0; i < result; i++) {
read(dopoll.dp_fds[i].fd, rbuf, STRLEN);
}
...
}

The following example is part of a test program which shows how DP_ISPOLLED() ioctl may be
used.
{
...

loopcnt = 0;
while (loopcnt < ITERATION) {
rn = random();
rn %= RANGE;
if (write(fds[rn], TESTSTRING, strlen(TESTSTRING)) !=
strlen(TESTSTRING)) {
perror("write to fifo failed.");
close (wfd);
free(pollfd);
error = 1;
goto out1;
}

676 man pages section 7: Device and Network Interfaces • Last Revised 28 Mar 2007
 poll(7d)

dpfd.fd = fds[rn];
dpfd.events = 0;
dpfd.revents = 0;
result = ioctl(wfd, DP_ISPOLLED, &dpfd);
if (result < 0) {
perror("/dev/poll ioctl DP_ISPOLLED failed");
printf("errno = %d\n", errno);
close (wfd);
free(pollfd);
error = 1;
goto out1;
}
if (result != 1) {
printf("DP_ISPOLLED returned incorrect result: %d.\n",
result);
close (wfd);
free(pollfd);
error = 1;
goto out1;
}
if (dpfd.fd != fds[rn]) {
printf("DP_ISPOLLED returned wrong fd %d, expect %d\n",
dpfd.fd, fds[rn]);
close (wfd);
free(pollfd);
error = 1;
goto out1;
}
if (dpfd.revents != POLLIN) {
printf("DP_ISPOLLED returned unexpected revents %d\n",
dpfd.revents);
close (wfd);
free(pollfd);
error = 1;
goto out1;
}
if (read(dpfd.fd, rbuf, strlen(TESTSTRING)) !=
strlen(TESTSTRING)) {
perror("read from fifo failed");
close (wfd);
free(pollfd);
error = 1;
goto out1;
}
loopcnt++;
}

Device and Network Interfaces 677

poll(7d)

Errors EACCES A process does not have permission to access the content cached in /dev/poll.
EINTR A signal was caught during the execution of the ioctl(2) function.
EFAULT The request argument requires a data transfer to or from a buffer pointed to by
arg, but arg points to an illegal address.
EINVAL The request or arg parameter is not valid for this device, or field of the dvpoll
struct pointed by arg is not valid (for example, when using write/pwrite dp_nfds is
greater than {OPEN_MAX}, or when using the DPPOLL ioctl dp_nfds is greater
than or equal to {OPEN_MAX}}.
ENXIO The O_NONBLOCK flag is set, the named file is a FIFO, the O_WRONLY flag is set, and
no process has the file open for reading; or the named file is a character special or
block special file and the device associated with this special file does not exist.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWcar ( Solaris)

SUNWcsr, SUNWcsu (Solaris on x86)

SUNWhea (header files)

Interface Stability Obsolete

MT-Level Safe

See Also open(2), poll(2), write(2), attributes(5)

Notes The /dev/poll API is particularly beneficial to applications that poll a large number of file
descriptors repeatedly. Applications will exhibit the best performance gain if the polled file
descriptor list rarely change.

When using the /dev/poll driver, you should remove a closed file descriptor from a
monitored poll set. Failure to do so may result in a POLLNVAL revents being returned for the
closed file descriptor. When a file descriptor is closed but not removed from the monitored
set, and is reused in subsequent open of a different device, you will be polling the device
associated with the reused file descriptor. In a multithreaded application, careful coordination
among threads doing close and DP_POLL ioctl is recommended for consistent results.

The /dev/poll driver caches a list of polled file descriptors, which are specific to a process.
Therefore, the /dev/poll file descriptor of a process will be inherited by its child process, just
like any other file descriptors. But the child process will have very limited access through this

678 man pages section 7: Device and Network Interfaces • Last Revised 28 Mar 2007
 poll(7d)

inherited /dev/poll file descriptor. Any attempt to write or do ioctl by the child process will
result in an EACCES error. The child process should close the inherited /dev/poll file
descriptor and open its own if desired.

The /dev/poll driver does not yet support polling. Polling on a /dev/poll file descriptor will
result in POLLERR being returned in the revents field of pollfd structure.

Device and Network Interfaces 679

prnio(7I)

Name prnio – generic printer interface

Synopsis #include <sys/prnio.h>

Description The prnio generic printer interface defines ioctl commands and data structures for printer
device drivers.

prnio defines and provides facilities for five basic phases of the printing process:
■ Identification — Retrieve device information/attributes
■ Setup — Set device attributes
■ Transfer — Transfer data to or from the device
■ Cleanup — Transfer phase conclusion
■ Abort — Transfer phase interruption

During the Identification phase, the application retrieves a set of device capabilities and
additional information using the PRNIOC_GET_IFCAP, PRNIOC_GET_STATUS,
PRNIOC_GET_TIMEOUTS, PRNIOC_GET_IFINFO and PRNIOC_GET_1284_DEVID commands.

During the Setup phase the application sets some interface attributes and probably resets the
printer as described in the PRNIOC_SET_IFCAP, PRNIOC_SET_TIMEOUTS and PRNIOC_RESET
sections.

During the Transfer phase, data is transferred in a forward (host to peripheral) or reverse
direction (peripheral to host). Transfer is accomplished using write(2) and read(2) system
calls. For prnio compliant printer drivers, forward transfer support is mandatory, while
reverse transfer support is optional. Applications can also use PRNIOC_GET_STATUS and
PRNIOC_GET_1284_STATUS commands during the transfer to monitor the device state.

The Cleanup phase is accomplished by closing the device using close(2). Device drivers
supporting prnio may set non-zero error code as appropriate. Applications should explicitly
close(2) a device before exiting and check errno value.

The Abort phase is accomplished by interrupting the write(2) and read(2) system calls. The
application can perform some additional cleanup during the Abort phase as described in
PRNIOC_GET_IFCAP section.
ioctls PRNIOC_GET_IFCAP Application can retrieve printer interface capabilities using this
command. The ioctl(2) argument is a pointer to uint_t, a bit
field representing a set of properties and services provided by a
printer driver. Set bit means supported capability. The following
values are defined:

PRN_BIDI - When this bit is set, the interface operates in a

bidirectional mode, instead of forward-only mode.
PRN_HOTPLUG - If this bit is set, the interface allows device
hot-plugging.

680 man pages section 7: Device and Network Interfaces • Last Revised 2 Jan 2002
 prnio(7I)

PRN_1284_DEVID - If this bit is set, the device is capable of

returning 1284 device ID (see PRNIOC_GET_1284_DEVID.)
PRN_1284_STATUS - If this bit is set, the device driver can return
device status lines (see PRNIOC_GET_1284_STATUS). Some
devices support this ioctl in unidirectional mode only.
PRN_TIMEOUTS - If this bit is set the peripheral may stall during
the transfer phase and the driver can timeout and return from
the write(2) and read(2) returning the number of bytes that
have been transferred. If PRN_TIMEOUTS is set, the driver
supports this functionality and the timeout values can be
retrieved and modified via the PRNIOC_GET_TIMEOUTS and
PRNIOC_SET_TIMEOUTS ioctls. Otherwise, applications can
implement their own timeouts and abort phase.
PRN_STREAMS - This bit impacts the application abort phase
behaviour. If the device claimed PRN_STREAMS capability, the
application must issue an I_FLUSH ioctl(2) before close(2) to
dismiss the untransferred data. Only STREAMS drivers can
support this capability.
PRNIOC_SET_IFCAP This ioctl can be used to change interface capabilities. The
argument is a pointer to uint_t bit field that is described in
detail in the PRNIOC_GET_IFCAP section. Capabilities should be
set one at a time; otherwise the command will return EINVAL.
The following capabilities can be changed by this ioctl:

PRN_BIDI - When this capability is set, the interface operates in a

bidirectional mode, instead of forward-only mode. Devices that
support only one mode will not return error; applications should
use PRNIOC_GET_IFCAP to check if the mode was successfully
changed. Because some capabilities may be altered as a side
effect of changing other capabilities, this command should be
followed by PRNIOC_GET_IFCAP.
PRNIOC_GET_IFINFO This command can be used to retrieve printer interface info
string, which is an arbitrary format string usually describing the
bus type. The argument is a pointer to struct
prn_interface_info as described below.
struct prn_interface_info {
uint_t if_len; /* length of buffer */
uint_t if_rlen; /* actual info length */
char *if_data; /* buffer address */
};

Device and Network Interfaces 681

prnio(7I)

The application allocates a buffer and sets if_data and if_len values to its address and
length, respectively. The driver returns the string to this buffer and sets if_len to its length. If
if_len is less that if_rlen, the driver must return the first if_len bytes of the string. The
application may then repeat the command with a bigger buffer.

Although prnio does not limit the contents of the interface info string, some values are
recommended and defined in <sys/prnio.h> by the following macros:

PRN_PARALLEL - Centronics or IEEE 1284 compatible devices

PRN_SERIAL - EIA-232/EIA-485 serial ports
PRN_USB - Universal Serial Bus printers
PRN_1394 - IEEE 1394 peripherals
Printer interface info string is for information only: no implications should be made from its
value.
PRNIOC_RESET Some applications may want to reset the printer state
during Setup and/or Cleanup phase using PRNIOC_RESET
command. Reset semantics are device-specific, and in
general, applications using this command should be aware
of the printer type.

Each prnio compliant driver is required to accept this

request, although performed actions are completely
driver-dependent. More information on the PRNIOC_RESET
implementation for the particular driver is available in the
corresponding man page and printer manual.
PRNIOC_GET_1284_DEVID This command can be used to retrieve printer device ID as
defined by IEEE 1284-1994.The ioctl(2) argument is a
pointer to struct prn_1284_device_id as described
below.
struct prn_1284_device_id {
uint_t id_len; /* length of buffer */
uint_t id_rlen; /* actual ID length */
char *id_data; /* buffer address */
};

For convenience, the two-byte length field is not considered part of device ID string and is not
returned in the user buffer. Instead, id_rlen value shall be set to (length - 2) by the driver,
where length is the ID length field value. If buffer length is less than id_rlen, the driver
returns the first id_len bytes of the ID.

The printer driver must return the most up-to-date value of the device ID.

682 man pages section 7: Device and Network Interfaces • Last Revised 2 Jan 2002
 prnio(7I)

PRNIOC_GET_STATUS This command can be used by applications to retrieve current

device status. The argument is a pointer to uint_t, where the
status word is returned. Status is a combination of the following
bits:

PRN_ONLINE - For devices that support PRN_HOTPLUG capability, this bit is set when the device is
online, otherwise the device is offline. Devices without PRN_HOTPLUG support should always
have this bit set.
PRN_READY - This bit indicates if the device is ready to receive/send data. Applications may use
this bit for an outbound flow control
PRNIOC_GET_1284_STATUS Devices that support PRN_1284_STATUS capability accept this
ioctl to retrieve the device status lines defined in IEEE 1284 for
use in Compatibility mode. The following bits may be set by
the driver:

PRN_1284_NOFAULT - Device is not in error state

PRN_1284_SELECT - Device is selected
PRN_1284_PE - Paper error
PRN_1284_BUSY - Device is busy
PRNIOC_GET_TIMEOUTS This command retrieves current transfer timeout values for
the driver. The argument is a pointer to struct prn_timeouts
as described below.
struct prn_timeouts {
uint_t tmo_forward; /* forward transfer timeout */
uint_t tmo_reverse; /* reverse transfer timeout */
};

tmo_forward and tmo_reverse define forward and reverse transfer timeouts in seconds. This
command is only valid for drivers that support PRN_TIMEOUTS capability.
PRNIOC_SET_TIMEOUTS This command sets current transfer timeout values for the driver.
The argument is a pointer to struct prn_timeouts. See
PRNIOC_GET_TIMEOUTS for description of this structure. This
command is only valid for drivers that support PRN_TIMEOUTS
capability.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, IA

Device and Network Interfaces 683

prnio(7I)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Evolving

See Also close(2), ioctl(2), read(2), write(2), attributes(5), ecpp(7D), usbprn(7D), lp(7D)

IEEE Std 1284-1994

684 man pages section 7: Device and Network Interfaces • Last Revised 2 Jan 2002
 profile(7D)

Name profile – DTrace profile interrupt provider

Description The profile driver is a DTrace dynamic tracing provider that adds time-based interrupt
event sources that can be used as DTrace probes.

Each profile event source is a time-based interrupt firing every fixed, specified time interval.
You can use these probes to sample some aspect of system state every unit time and the
samples can then be used to infer system behavior. If the sampling rate is high, or the sampling
time is long, an accurate inference is possible. By using the DTrace facility to bind arbitrary
actions to probes, you can use the profile provider to sample practically anything in the
system. For example, you could sample the state of the current thread, the CPU state, or the
current machine instruction each time a probe fires.

The profile driver is not a public interface and you access the instrumentation offered by this
provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description of the
public documented interfaces available for the DTrace facility and the probes offered by the
profile provider.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), attributes(5), dtrace(7D)

Solaris Dynamic Tracing Guide

Device and Network Interfaces 685

ptem(7M)

Name ptem – STREAMS Pseudo Terminal Emulation module

Synopsis int ioctl(fd, I_PUSH, "ptem");

Description ptem is a STREAMS module that, when used in conjunction with a line discipline and pseudo
terminal driver, emulates a terminal.

The ptem module must be pushed (see I_PUSH, streamio(7I)) onto the slave side of a pseudo
terminal STREAM, before the ldterm(7M) module is pushed.

On the write-side, the TCSETA, TCSETAF, TCSETAW, TCGETA, TCSETS, TCSETSW, TCSETSF, TCGETS,
TCSBRK, JWINSIZE, TIOCGWINSZ, and TIOCSWINSZ termio ioctl(2) messages are processed and
acknowledged. If remote mode is not in effect, ptem handles the TIOCSTI ioctl by copying the
argument bytes into an M_DATA message and passing it back up the read side. Regardless of the
remote mode setting, ptem acknowledges the ioctl and passes a copy of it downstream for
possible further processing. A hang up (that is, stty 0) is converted to a zero length M_DATA
message and passed downstream. Termio cflags and window row and column information
are stored locally one per stream. M_DELAY messages are discarded. All other messages are
passed downstream unmodified.

On the read-side all messages are passed upstream unmodified with the following exceptions.
All M_READ and M_DELAY messages are freed in both directions. A TCSBRK ioctl is converted to
an M_BREAK message and passed upstream and an acknowledgement is returned downstream.
A TIOCSIGNAL ioctl is converted into an M_PCSIG message, and passed upstream and an
acknowledgement is returned downstream. Finally a TIOCREMOTE ioctl is converted into an
M_CTL message, acknowledged, and passed upstream; the resulting mode is retained for use in
subsequent TIOCSTI parsing.
Files <sys/ptem.h>

See Also stty(1), ioctl(2), ldterm(7M), pckt(7M), streamio(7I), termio(7I)

STREAMS Programming Guide

686 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 ptm(7D)

Name ptm – STREAMS pseudo-tty master driver

Description The pseudo-tty subsystem simulates a terminal connection, where the master side represents
the terminal and the slave represents the user process's special device end point. In order to
use the pseudo-tty subsystem, a node for the master side driver /dev/ptmx and N number of
nodes for the slave driver must be installed. See pts(7D). The master device is set up as a
cloned device where its major device number is the major for the clone device and its minor
device number is the major for the ptm driver. There are no nodes in the file system for master
devices. The master pseudo driver is opened using the open(2) system call with /dev/ptmx as
the device parameter. The clone open finds the next available minor device for the ptm major
device.

A master device is available only if it and its corresponding slave device are not already open.
When the master device is opened, the corresponding slave device is automatically locked out.
Only one open is allowed on a master device. Multiple opens are allowed on the slave device.
After both the master and slave have been opened, the user has two file descriptors which are
the end points of a full duplex connection composed of two streams which are automatically
connected at the master and slave drivers. The user may then push modules onto either side of
the stream pair.

The master and slave drivers pass all messages to their adjacent queues. Only the M_FLUSH
needs some processing. Because the read queue of one side is connected to the write queue of
the other, the FLUSHR flag is changed to the FLUSHW flag and vice versa. When the master device
is closed an M_HANGUP message is sent to the slave device which will render the device
unusable. The process on the slave side gets the errno EIO when attempting to write on that
stream but it will be able to read any data remaining on the stream head read queue. When all
the data has been read, read() returns 0 indicating that the stream can no longer be used. On
the last close of the slave device, a 0-length message is sent to the master device. When the
application on the master side issues a read() or getmsg() and 0 is returned, the user of the
master device decides whether to issue a close() that dismantles the pseudo-terminal
subsystem. If the master device is not closed, the pseudo-tty subsystem will be available to
another user to open the slave device.

If O_NONBLOCK or O_NDELAY is set, read on the master side returns −1 with errno set to EAGAIN if
no data is available, and write returns −1 with errno set to EAGAIN if there is internal flow
control.

ioctls The master driver supports the ISPTM and UNLKPT ioctls that are used by the functions
grantpt(3C), unlockpt(3C) and ptsname(3C). The ioctl ISPTM determines whether the file
descriptor is that of an open master device. On success, it returns the 0. The ioctl UNLKPT
unlocks the master and slave devices. It returns 0 on success. On failure, the errno is set to
EINVAL indicating that the master device is not open.

Device and Network Interfaces 687

ptm(7D)

Files /dev/ptmx master clone device

/dev/pts/M slave devices (M = 0 -> N-1)

See Also grantpt(3C), ptsname(3C), unlockpt(3C), pckt(7M), pts(7D)

STREAMS Programming Guide

688 man pages section 7: Device and Network Interfaces • Last Revised 5 Feb 1997
 pts(7D)

Name pts – STREAMS pseudo-tty slave driver

Description The pseudo-tty subsystem simulates a terminal connection, where the master side represents
the terminal and the slave represents the user process's special device end point. In order to
use the pseudo-tty subsystem, a node for the master side driver /dev/ptmx and N nodes for
the slave driver (N is determined at installation time) must be installed. The names of the slave
devices are /dev/pts/M where M has the values 0 through N-1. When the master device is
opened, the corresponding slave device is automatically locked out. No user may open that
slave device until its permissions are adjusted and the device unlocked by calling functions
grantpt(3C) and unlockpt(3C). The user can then invoke the open system call with the name
that is returned by the ptsname(3C) function. See the example below.

Only one open is allowed on a master device. Multiple opens are allowed on the slave device.
After both the master and slave have been opened, the user has two file descriptors which are
end points of a full duplex connection composed of two streams automatically connected at
the master and slave drivers. The user may then push modules onto either side of the stream
pair. The user needs to push the ptem(7M) and ldterm(7M) modules onto the slave side of the
pseudo-terminal subsystem to get terminal semantics.

The master and slave drivers pass all messages to their adjacent queues. Only the M_FLUSH
needs some processing. Because the read queue of one side is connected to the write queue of
the other, the FLUSHR flag is changed to the FLUSHW flag and vice versa. When the master device
is closed an M_HANGUP message is sent to the slave device which will render the device
unusable. The process on the slave side gets the errno EIO when attempting to write on that
stream but it will be able to read any data remaining on the stream head read queue. When all
the data has been read, read returns 0 indicating that the stream can no longer be used. On the
last close of the slave device, a 0-length message is sent to the master device. When the
application on the master side issues a read() or getmsg() and 0 is returned, the user of the
master device decides whether to issue a close() that dismantles the pseudo-terminal
subsystem. If the master device is not closed, the pseudo-tty subsystem will be available to
another user to open the slave device. Since 0-length messages are used to indicate that the
process on the slave side has closed and should be interpreted that way by the process on the
master side, applications on the slave side should not write 0-length messages. If that occurs,
the write returns 0, and the 0-length message is discarded by the ptem module.

The standard STREAMS system calls can access the pseudo-tty devices. The slave devices
support the O_NDELAY and O_NONBLOCK flags.
Examples int fdm fds;
char *slavename;
extern char *ptsname();

fdm = open("/dev/ptmx", O_RDWR); /* open master */

grantpt(fdm); /* change permission of slave */
unlockpt(fdm); /* unlock slave */
slavename = ptsname(fdm); /* get name of slave */

Device and Network Interfaces 689

pts(7D)

fds = open(slavename, O_RDWR); /* open slave */

ioctl(fds, I_PUSH, "ptem"); /* push ptem */
ioctl(fds, I_PUSH, "ldterm"); /* push ldterm*/
Files /dev/ptmx master clone device
/dev/pts/M slave devices (M = 0 -> N-1)

See Also grantpt(3C), ptsname(3C), unlockpt(3C), ldterm(7M), ptm(7D), ptem(7M)

STREAMS Programming Guide

690 man pages section 7: Device and Network Interfaces • Last Revised 21 Aug 1992
 pty(7D)

Name pty – pseudo-terminal driver

Description The pty driver provides support for a pair of devices collectively known as a pseudo-terminal.
The two devices comprising a pseudo-terminal are known as a controller and a slave. The slave
device distinguishes between the B0 baud rate and other baud rates specified in the c_cflag
word of the termios structure, and the CLOCAL flag in that word. It does not support any of the
other termio(7I) device control functions specified by flags in the c_cflag word of the
termios structure and by the IGNBRK, IGNPAR, PARMRK, or INPCK flags in the c_iflag word of
the termios structure, as these functions apply only to asynchronous serial ports. All other
termio(7I) functions must be performed by STREAMS modules pushed atop the driver; when
a slave device is opened, the ldterm(7M) and ttcompat(7M) STREAMS modules are
automatically pushed on top of the stream, providing the standard termio(7I) interface.

Instead of having a hardware interface and associated hardware that supports the terminal
functions, the functions are implemented by another process manipulating the controller
device of the pseudo-terminal.

The controller and the slave devices of the pseudo-terminal are tightly connected. Any data
written on the controller device is given to the slave device as input, as though it had been
received from a hardware interface. Any data written on the slave terminal can be read from
the controller device (rather than being transmitted from a UAR).

By default, 48 pseudo-terminal pairs are configured as follows:

/dev/pty[p-r][0-9a-f] controller devices
/dev/tty[p-r][0-9a-f] slave devices

ioctls The standard set of termio ioctls are supported by the slave device. None of the bits in the
c_cflag word have any effect on the pseudo-terminal, except that if the baud rate is set to B0,
it will appear to the process on the controller device as if the last process on the slave device
had closed the line; thus, setting the baud rate to B0 has the effect of ‘‘hanging up'' the
pseudo-terminal, just as it has the effect of ‘‘hanging up'' a real terminal.

There is no notion of ‘‘parity'' on a pseudo-terminal, so none of the flags in the c_iflag word
that control the processing of parity errors have any effect. Similarly, there is no notion of a
‘‘break'', so none of the flags that control the processing of breaks, and none of the ioctls that
generate breaks, have any effect.

Input flow control is automatically performed; a process that attempts to write to the
controller device will be blocked if too much unconsumed data is buffered on the slave device.
The input flow control provided by the IXOFF flag in the c_iflag word is not supported.

The delays specified in the c_oflag word are not supported.

As there are no modems involved in a pseudo-terminal, the ioctls that return or alter the
state of modem control lines are silently ignored.

Device and Network Interfaces 691

pty(7D)

A few special ioctls are provided on the controller devices of pseudo-terminals to provide the
functionality needed by applications programs to emulate real hardware interfaces:
TIOCSTOP The argument is ignored. Output to the pseudo-terminal is suspended, as if a
STOP character had been typed.
TIOCSTART The argument is ignored. Output to the pseudo-terminal is restarted, as if a
START character had been typed.
TIOCPKT The argument is a pointer to an int. If the value of the int is non-zero, packet
mode is enabled; if the value of the int is zero, packet mode is disabled. When
a pseudo-terminal is in packet mode, each subsequent read(2) from the
controller device will return data written on the slave device preceded by a
zero byte (symbolically defined as TIOCPKT_DATA), or a single byte reflecting
control status information. In the latter case, the byte is an inclusive-or of
zero or more of the bits:
TIOCPKT_FLUSHREAD whenever the read queue for the terminal is
flushed.
TIOCPKT_FLUSHWRITE whenever the write queue for the terminal is
flushed.
TIOCPKT_STOP whenever output to the terminal is stopped using
^S.
TIOCPKT_START whenever output to the terminal is restarted.
TIOCPKT_DOSTOP whenever XON/XOFF flow control is enabled after
being disabled; it is considered ‘‘enabled'' when the
IXON flag in the c_iflag word is set, the VSTOP
member of the c_cc array is ^S and the VSTART
member of the c_cc array is ^Q.
TIOCPKT_NOSTOP whenever XON/XOFF flow control is disabled
after being enabled.
TIOCREMOTE The argument is a pointer to an int. If the value of the int is non-zero,
remote mode is enabled; if the value of the int is zero, remote mode is
disabled. This mode can be enabled or disabled independently of packet
mode. When a pseudo-terminal is in remote mode, input to the slave device
of the pseudo-terminal is flow controlled and not input edited (regardless of
the mode the slave side of the pseudo-terminal). Each write to the controller
device produces a record boundary for the process reading the slave device.
In normal usage, a write of data is like the data typed as a line on the terminal;
a write of 0 bytes is like typing an EOF character. Note: this means that a
process writing to a pseudo-terminal controller in remote mode must keep
track of line boundaries, and write only one line at a time to the controller. If,

692 man pages section 7: Device and Network Interfaces • Last Revised 8 Aug 1994
 pty(7D)

for example, it were to buffer up several NEWLINE characters and write

them to the controller with one write(), it would appear to a process reading
from the slave as if a single line containing several NEWLINE characters had
been typed (as if, for example, a user had typed the LNEXT character before
typing all but the last of those NEWLINE characters). Remote mode can be
used when doing remote line editing in a window manager, or whenever flow
controlled input is required.
Examples #include <fcntl.h>
#include <sys/termios.h>

int fdm fds;

fdm = open("/dev/ptyp0, O_RDWR); /* open master */
fds = open("/dev/ttyp0, O_RDWR); /* open slave */
Files /dev/pty[p-z][0-9a-f] pseudo-terminal controller devices
/dev/tty[p-z][0-9a-f] pseudo-terminal slave devices

See Also rlogin(1), rlogind(1M), ldterm(7M), termio(7I), ttcompat(7M),

Notes It is apparently not possible to send an EOT by writing zero bytes in TIOCREMOTE mode.

Device and Network Interfaces 693

qfe(7d)

Name qfe – SUNW,qfe Quad Fast-Ethernet device driver

Synopsis /dev/qfe

Description The SUNW,qfe Quad Fast-Ethernet driver is a multi-threaded, loadable, clonable, STREAMS
hardware driver supporting the connectionless Data Link Provider Interface, dlpi(7P), over a
SUNW,qfe Quad Fast-Ethernet controller. Multiple SUNW,qfe controllers installed within
the system are supported by the driver. The qfe driver provides basic support for the
SUNW,qfe hardware. It is used to handle the SUNW,qfe device. Functions include chip
initialization, frame transit and receive, multicast and promiscuous support, and error
recovery and reporting.

SUNW,qfe The SUNW,qfe device provides a 100Base-TX networking interface. There are two types of
SUNW,qfe device: one supporting Sbus and the other supporting the PCI bus interface. The
Sbus SUNW,qfe device uses Sun's FEPS ASIC, which provides the Sbus interface and MAC
functions. The PCI SUNW,qfe device uses Sun's PFEX ASIC to provide the PCI interface and
MAC functions. Both connect with the 100Base-TX on-board transceiver, which connects to a
RJ45 connector to provide the Physical layer functions and external connection.

The 100Base-TX standard specifies an “auto-negotiation” protocol to automatically select the

mode and speed of operation. The internal transceiver is capable of doing auto-negotiation
with the remote-end of the link (link partner) and receives the capabilities of the remote end.
It selects the Highest Common Denominator mode of operation based on the priorities. It also
supports forced-mode of operation where the driver can select the mode of operation.

Application The cloning character-special device /dev/qfe is used to access all SUNW,qfe controllers
Programming installed within the system.
Interface
qfe and DLPI The qfe driver is a “style 2” data link service provider. All M_PROTO and M_PCPROTO type
messages are interpreted as DLPI primitives. Valid DLPI primitives are defined in
<sys/dlpi.h>. Refer to dlpi(7P) for more information. An explicit DL_ATTACH_REQ message
by the user is required to associate the opened stream with a particular device (ppa). The ppa
ID is interpreted as an unsigned long data type and indicates the corresponding device
instance (unit) number. The driver returns an error (DL_ERROR_ACK) if the ppa field value does
not correspond to a valid device instance number for this system. The device is initialized on
first attach and de-initialized (stopped) at last detach.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ The maximum SDU is 1500 (ETHERMTU - defined in <sys/ethernet.h>).
■ The minimum SDU is 0.
■ The dlsap address length is 8.
■ The MAC type is DL_ETHER.

694 man pages section 7: Device and Network Interfaces • Last Revised 6 May 1998
 qfe(7d)

■ The sap length values is −2 meaning the physical address component is followed
immediately by a 2 byte sap component within the DLSAP address.
■ The service mode is DL_CLDLS.
■ No optional quality of service (QOS) support is included at present so the QOS fields are 0.
■ The provider style is DL_STYLE2.
■ The version is DL_VERSION_2.
■ The broadcast address value is Ethernet/IEEE broadcast address (0xFFFFFF).

Once in the DL_ATTACHED state, the user must send a DL_BIND_REQ to associate a particular
service access pointer SAP with the stream. The qfe driver interprets the sap field within the
DL_BIND_REQ as an Ethernet “type” therefore valid values for the sap field are in the
[0-0xFFFF] range. Only one Ethernet type can be bound to the stream at any time.

If the user selects a sap with a value of 0, the receiver will be in “802.3 mode”. All frames
received from the media having a “type” field in the range [0-1500] are assumed to be 802.3
frames and are routed up all open streams which are bound to sap value 0. If more than one
stream is in “802.3 mode” then the frame will be duplicated and routed up multiple streams as
DL_UNITDATA_IND messages.

In transmission, the driver checks the sap field of the DL_BIND_REQ if the sap value is 0, and if
the destination type field is in the range [0-1500]. If either is true, the driver computes the
length of the message, not including initial M_PROTO mblk (message block), of all subsequent
DL_UNITDATA_REQ messages and transmits 802.3 frames that have this value in the MAC frame
header length field.

The qfe driver DLSAP address format consists of the 6 byte physical (Ethernet) address
component followed immediately by the 2 byte sap (type) component producing an 8 byte
DLSAP address. Applications should not hardcode to this particular implementation-specific
DLSAP address format but use information returned in the DL_INFO_ACK primitive to compose
and decompose DLSAP addresses. The sap length, full DLSAP length, and sap/physical ordering
are included within the DL_INFO_ACK. The physical address length can be computed by
subtracting the sap length from the full DLSAP address length or by issuing the
DL_PHYS_ADDR_REQ to obtain the current physical address associated with the stream.

Once in the DL_BOUND state, the user may transmit frames on the Ethernet by sending
DL_UNITDATA_REQ messages to the qfe driver. The qfe driver will route received Ethernet
frames up all those open and bound streams having a sap which matches the Ethernet type as
DL_UNITDATA_IND messages. Received Ethernet frames are duplicated and routed up multiple
open streams if necessary. The DLSAP address contained within the DL_UNITDATA_REQ and
DL_UNITDATA_IND messages consists of both the sap (type) and physical (Ethernet)
components.

Device and Network Interfaces 695

qfe(7d)

In addition to the mandatory connectionless DLPI message set the driver also supports the
following primitives.

qfe Primitives The DL_ENABMULTI_REQ and DL_DISABMULTI_REQ primitives enable or disable reception of
individual multicast group addresses. A set of multicast addresses may be iteratively created
and modified on a per-stream basis using these primitives. The driver accepts these primitives
in any state following DL_ATTACHED.

The DL_PROMISCON_REQ and DL_PROMISCOFF_REQ primitives with the DL_PROMISC_PHYS flag

set in the dl_level field enables or disables reception of all frames on the media
(“promiscuous mode”), including frames generated by the local host.

When used with the DL_PROMISC_SAP flag set this enables or disables reception of all sap
(Ethernet type) values. When used with the DL_PROMISC_MULTI flag set this enables or disables
reception of all multicast group addresses. The effect of each is always on a per-stream basis
and independent of the other sap and physical level configurations on this stream or other
streams.

The DL_PHYS_ADDR_REQ primitive returns the 6 octet Ethernet address currently associated
(attached) to the stream in the DL_PHYS_ADDR_ACK primitive. This primitive is valid only in
states following a successful DL_ATTACH_REQ.

The DL_SET_PHYS_ADDR_REQ primitive changes the 6 octet Ethernet address currently

associated (attached) to this stream. The credentials of the process which originally opened
this stream must be root. Otherwise EPERM is returned in the DL_ERROR_ACK. This primitive is
destructive in that it affects all other current and future streams attached to this device. An
M_ERROR is sent up all other streams attached to this device when this primitive is successful on
this stream. Once changed, all streams subsequently opened and attached to this device will
obtain this new physical address. Once changed, the physical address will remain until this
primitive is used to change the physical address again or the system is rebooted, whichever
comes first.

qfe Driver By default, the qfe driver performs “auto-negotiation” to select the mode and speed of the link.

The link can be in one of the four following modes:

■ 100 Mbps, full-duplex
■ 100 Mbps, half-duplex
■ 10 Mbps, full-duplex
■ 10 Mbps, half-duplex

These speeds and modes are described in the 100Base-TX standard.

The auto−negotiation protocol automatically selects:

■ Operation mode (half-duplex or full-duplex)
■ Speed (100 Mbps or 10 Mbps)

696 man pages section 7: Device and Network Interfaces • Last Revised 6 May 1998
 qfe(7d)

The auto−negotiation protocol does the following:

■ Gets all the modes of operation supported by the Link Partner
■ Advertises its capabilities to the Link Partner
■ Selects the highest common denominator mode of operation based on the priorities.
■ The highest priority is given to the 100 Mbps, full-duplex; lowest priority is given to 10
Mbps, half-duplex.

The 100Base-TX transceiver is capable of all of the operating speeds and modes listed above.
By default, auto-negotiation is used to select the speed and the mode of the link and the
common mode of operation with the link partner.

Sometimes, the user may want to select the speed and mode of the link. The SUNW,qfe device
supports programmable "IPG" (Inter-Packet Gap) parameters ipg1 and ipg2. By default, the
driver sets ipg1 to 8 byte-times and ipg2 to 4 byte-times (which are the standard values).
Sometimes, the user may want to alter these values depending on whether the driver supports
10 Mbps or 100 Mpbs and accordingly, IPG will be set to 9.6 or 0.96 microseconds.

qfe Parameter List The qfe driver provides for setting and getting various parameters for the SUNW,qfe device.
The parameter list includes:
■ current transceiver status
■ current link status
■ inter-packet gap
■ local transceiver capabilities
■ link partner capabilities

The local transceiver has two sets of capabilities: one set reflects the capabilities of the
hardware, which are read-only (RO) parameters, and the second set, which reflects the values
chosen by the user, is used in speed selection. There are read/write (RW) capabilities. At boot
time, these two sets of capabilities will be the same. The Link Partner capabilities are also
read-only parameters because the current default value of these parameters can only be read
and cannot be modified.
Files /dev/qfe qfe special character device
/kernel/drv/qfe.conf system wide default device driver properties

See Also ndd(1M), netstat(1M), driver.conf(4), dlpi(7P)

Device and Network Interfaces 697

qlc(7D)

Name qlc – ISP2200, ISP2300, and SP212 Family Fibre Channel host bus adapter driver.
Synopsis SUNW,qlc

Description The qlc host bus adapter driver is a Sun Fibre Channel transport layer-compliant nexus driver
for the Qlogic ISP2200, ISP2200A, ISP2310, ISP2312, and SP212 adapters. These adapters
support Fibre Channel SCSI and IP Protocols, FC-AL public loop profile, point-to-point
fabric connection and Fibre Channel service classes two and three (see NOTES section below).

The qlc driver interfaces with the Sun Fibre Channel transport layer to support the standard
functions provided by the SCSA interface. It supports auto request sense and tagged queueing
by default. The driver requires that all devices have unique hard addresses in private loop
configurations. Devices with conflicting hard addresses are not accessible.
Files /kernel/drv/qlc 32–bit ELF kernel module (x86)
/kernel/drv/amd64/qlc 64-bit ELF kernel module (x86)
/kernel/drv/sparcv9/qlc 64-bit ELF kernel module (SPARC)
/kernel/drv/qlc.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWqlc

See Also prtconf(1M), driver.conf(4), fcp(7D), fp(7d)

Writing Device Drivers

ANSI X3.230:1994, Fibre Channel Physical Signaling (FC-PH)

Project 1134-D, Fibre Channel Generic Services (FC-GS-2)

ANSI X3.269-1996, Fibre Channel Arbitrated Loop (FC-AL)

ANSI X3.270-1996, Fibre Channel Protocol for SCSI (FCP-SCSI)

ANSI X3.270-1996, SCSI-3 Architecture Model (SAM)

Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)

Fabric Loop Attachment (FC-FLA)

ISP2200 Firmware Interface Specification, QLogic Corporation

ISP2300 Series Firmware Specification, QLogic Corporation

698 man pages section 7: Device and Network Interfaces • Last Revised 21 Feb 2005
 qlc(7D)

Notes SP-212–based host bus adapters (including QLA-210) are supported on x86 platforms only
and are limited to a maximum of 8 targets in fabric and sixteen targets in local loop topology.
FL topology is not supported with the SP-212–based host bus adapter.

Device and Network Interfaces 699

qlge(7D)

Name qlge – 10 Gigabit Ethernet driver for QLogic QLE81XX Converged Network Adapter Family
Synopsis SUNW,qlge

Description The qlge 10 Gigabit Ethernet driver is a multi-threaded, Loadable, clonable, GLDv3-based
driver. The qlge driver provides basic support including chip initialization, auto-negotiation,
packet transmit and receive, Jumbo Frame, promiscuous and multicast support, 802.3x
Standard Ethernet Flow Control and Class Based Flow Control (CBFC), Checksum Offload,
Large Send Offload (LSO).

Configuration The qlge driver is managed by the dladm(1M)command line utility, which allows VLANs to
be defined on top of qlge instances and for qlge instances to be aggregated. See dladm(1M) for
details.

Users can also modify qlge.conf to change default settings, like mtu, flow control mode, and
so forth.
Files /kernel/drv/qlge 32-bit ELF kernel module, x86
/kernel/drv/amd64/qlge 64-bit ELF kernel module, x86
/kernel/drv/sparcv9/qlge 64-bit ELF kernel module, SPARC
/kernel/drv/qlge.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC, x86

Availability SUNWqlc

See Also dladm(1M), prtconf(1M), attributes(5)

Writing Device Drivers

700 man pages section 7: Device and Network Interfaces • Last Revised 14 Dec 2009
 quotactl(7I)

Name quotactl – manipulate disk quotas

Synopsis #include <sys/fs/ufs_quota.h>
int ioctl(int fd, Q_QUOTACTL, struct quotcl *qp)

Description This ioctl() call manipulates disk quotas. fd is the file descriptor returned by the open()
system call after opening the quotas file (located in the root directory of the filesystem
running quotas.) Q_QUOTACTL is defined in /usr/include/sys/fs/ufs_quota.h. qp is the
address of the quotctl structure which is defined as
struct quotctl {
int op;
uid_t uid;
caddr_t addr;
};

op indicates an operation to be applied to the user ID uid. (See below.) addr is the address of an
optional, command specific, data structure which is copied in or out of the system. The
interpretation of addr is given with each value of op below.
Q_QUOTAON Turn on quotas for a file system. addr points to the full pathname of the
quotas file. uid is ignored. It is recommended that uid have the value of 0.
This call is restricted to the super-user.
Q_QUOTAOFF Turn off quotas for a file system. addr and uid are ignored. It is recommended
that addr have the value of NULL and uid have the value of 0. This call is
restricted to the super-user.
Q_GETQUOTA Get disk quota limits and current usage for user uid. addr is a pointer to a
dqblk structure (defined in <sys/fs/ufs_quota.h>). Only the super-user
may get the quotas of a user other than himself.
Q_SETQUOTA Set disk quota limits and current usage for user uid. addr is a pointer to a
dqblk structure (defined in sys/fs/ufs_quota.h). This call is restricted to
the super-user.
Q_SETQLIM Set disk quota limits for user uid. addr is a pointer to a dqblk structure
(defined in sys/fs/ufs_quota.h). This call is restricted to the super-user.
Q_SYNC Update the on-disk copy of quota usages for this file system. addr and uid are
ignored.
Q_ALLSYNC Update the on-disk copy of quota usages for all file systems with active
quotas. addr and uid are ignored.

Return Values This ioctl() returns:

0 on success.
−1 on failure and sets errno to indicate the error.

Device and Network Interfaces 701

quotactl(7I)

Errors EFAULT addr is invalid.

EINVAL The kernel has not been compiled with the QUOTA option. op is invalid.
ENOENT The quotas file specified by addr does not exist.
EPERM The call is privileged and the calling process did not assert {PRIV_SYS_MOUNT}
in the effective set.
ESRCH No disk quota is found for the indicated user. Quotas have not been turned on for
this file system.
EUSERS The quota table is full.

If op is Q_QUOTAON, ioctl() may set errno to:

EACCES The quota file pointed to by addr exists but is not a regular file. The quota file
pointed to by addr exists but is not on the file system pointed to by special.
EIO Internal I/O error while attempting to read the quotas file pointed to by addr.
Files /usr/include/sys/fs/ufs_quota.h quota-related structure/function definitions and
defines

See Also quota(1M), quotacheck(1M), quotaon(1M), getrlimit(2), mount(2)

Bugs There should be some way to integrate this call with the resource limit interface provided by
setrlimit() and getrlimit(2).

This call is incompatible with Melbourne quotas.

702 man pages section 7: Device and Network Interfaces • Last Revised 14 June 2004
 qus(7D)

Name qus – Qlogic Ultra3 SCSI ISP10160 Host Bus Adapter Driver
Synopsis pci@pci-slot/scsi@4 - Scsi bus-1
pci@pci-slot/scsi@5 - Scsi bus-2

Description The ISP10160 host bus adapter is a SCSA-compliant nexus driver that supports Qlogic
ISP10160 SCSI chips on the PCI bus. The ISP10160 is an intelligent SCSI host bus adapter chip
that reduces the amount of CPU overhead used in a SCSI transfer.

The qus driver supports the standard functions provided by the SCSA interface. The driver
supports tagged and untagged queuing, fast, fast-20, fast-40, fast-80, and auto request sense,
but does not support linked commands.

Configuration You configure the qus driver by defining properties in qus.conf which override the global
SCSI settings. Supported properties are scsi-options, target<n>-scsi-options,
scsi-reset-delay, scsi-watchdog-tick, scsi-tag-age-limit, scsi- initiator-id, and
scsi-selection-timeout.

target<n>-scsi-options overrides the scsi-options property value for target<n>.<n> is a

hex value that can vary from 0 to f. Refer to scsi_hba_attach(9F) for details.

Examples Example 1: SCSI Options:

Create a file called /kernel/drv/qus.conf, then add the following line to disable tagged
queuing, fast SCSI, and Wide mode for all qus instances:
scsi-options=0x78;

To disable an option for a specific ISP10160 (see driver.conf(4)), do the following:

name="qus" parent="/pci@1f,2000/pci@1"
unit-address="4"
scsi-options=0x178
target3-scsi-options=0x58 scsi-initiator-id=6;

The default initiator ID in OBP is 7 and that the change to ID 6 occurs at attach time. You
might prefer to change the initiator ID in OBP.

Example 1 sets scsi-options for target 3 to 0x58 and all other targets on this SCSI bus to
0x178.

To determine the physical pathname of the parent, use the /devices tree or follow the link of
the logical device name:
example# ls -l /dev/rdsk/c2t0d0s0
lrwxrwxrwx 1 root root 76 Aug 22 13:29 /dev/rdsk/c2t0d0
s0 -> ../../devices/pci@1f,2000/pci@1/scsi@5/sd@0,0:a,raw

Device and Network Interfaces 703

qus(7D)

To determine the hardware property values, use the output of prtconf(1M) with the -v
option:
pci, instance #0
Driver properties:
name=’device_type’ type=string items=1 dev=none
value=’pci’
Hardware properties:
name=’ranges’ type=int items=8
value=82000000.00000000.00100000.82000000.00000000.00100000./
00000000.00100000
name=’latency-timer’ type=int items=1
value=00000040
name=’cache-line-size’ type=int items=1
value=00000010
scsi, instance #0
Driver properties:
name=’scsi-selection-timeout’ type=int items=1 dev=(249,0)
value=000000fa
name=’scsi-options’ type=int items=1 dev=(249,0)
value=00107ff8
name=’scsi-watchdog-tick’ type=int items=1 dev=(249,0)
value=0000000a
name=’scsi-tag-age-limit’ type=int items=1 dev=(249,0)
value=00000002
name=’scsi-reset-delay’ type=int items=1 dev=(249,0)
value=00000bb8
Hardware properties:
name=’cache-line-size’ type=int items=1
value=00000010
sd (driver not attached)
st (driver not attached)
scsi, instance #1
Driver properties:
name=’scsi-selection-timeout’ type=int items=1 dev=(249,0)
value=000000fa
name=’scsi-options’ type=int items=1 dev=(249,0)
value=00107ff8
name=’scsi-watchdog-tick’ type=int items=1 dev=(249,0)
value=0000000a
name=’scsi-tag-age-limit’ type=int items=1 dev=(249,0)
value=00000002
name=’scsi-reset-delay’ type=int items=1 dev=(249,0)
value=00000bb8
Hardware properties:
name=’cache-line-size’ type=int items=1
value=00000010

704 man pages section 7: Device and Network Interfaces • Last Revised 29 May 2003
 qus(7D)

sd (driver not attached)

st (driver not attached)

Example 2: ISP10160 Properties

The qus driver exports properties indicating (per target) the negotiated transfer speed
(target<n>-sync-speed), whether tagged queuing has been enabled (target<n>-TQ), and
whether the wide data transfer has been negotiated (target<n>-wide). The sync-speed
property value is the data transfer rate in KB/sec. The target-TQ and target-wide properties
have no value. The existence of these properties indicate that tagged queuing or wide transfer
is enabled. Refer to prtconf(1M) (verbose option) for information on qus properties.
scsi, instance #1
Driver properties:
name=’target2-wide’ type=boolean dev=none
name=’target2-TQ’ type=boolean dev=none
name=’target2-sync-speed’ type=int items=1 dev=none
value=00027100
name=’target0-wide’ type=boolean dev=none
name=’target0-TQ’ type=boolean dev=none
name=’target0-sync-speed’ type=int items=1 dev=none
value=00027100

To determine the physical pathname of the parent, use the /devices tree or follow the link of
the logical deiice name.

To set scsi-options more specifically per device type, add the following line to the
/kernel/drv/qus.conf file:
device-type-scsi-options-list =
"SEAGATE ST32550W", "seagate-scsi-options" ;
seagate-scsi-options = 0x58;

All devices of this specific disk type have scsi-options set to 0x58.

scsi-options specified per target ID has the highest precedence, followed by scsi-options
per device type. Global (for all qus instances) scsi-options per bus has the lowest
precedence.

You must reboot your system for the specified scsi-options to take effect.

Example 3: Driver Capabilities

To enable some driver features, the target driver must set capabilities in the qus driver. The
target driver can query and modify the following capabilities: synchronous, tagged-qing,
wide-xfer, auto-rqsense, qfull-retries, qfull-retry-interval. All other capabilities are query only.

Device and Network Interfaces 705

qus(7D)

By default, tagged-qing, auto-rqsense, and wide-xfer capabilities are disabled, while

disconnect, synchronous, and untagged-qing are enabled. These capabilities can have binary
values (0 or 1) only. The default value for both qfull-retries and qfull-retry-interval is 10. The
qfull-retries capability is a uchar_t (0 to 255), while qfull-retry-interval is a ushort_t (0 to
65535).

The target driver must enable tagged-qing and wide-xfer explicitly. The untagged-qing
capability is always enabled and its value cannot be modified due to the qus driver's ability to
queue commands even when tagged-qing is disabled.

When a conflict occurs between the value of scsi-options and a capability, the value set in
scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call.

Refer to scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.

Files /kernel/drv/sparcv9/qus 64–bitELF kernel module
/kernel/drv/qus.conf Configuration file

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also prtconf(1M), driver.conf(4) , attributes(5), scsi_abort(9F), scsi_hba_attach(9F),

scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F), scsi_transport(9F),
scsi_device(9S), scsi_extended_sense(9S), scsi_inquiry(9S), scsi_pkt(9S)
Writing Device Drivers
ANSI Small Computer System Interface-2 (SCSI-2)
SCSI Parallel Interface-3 (SPI-3)
QLogic Corporation, ISP1080/1280/10160A/12160A Initiator Firmware Interface Specification
QLogic Corporation, ISP10160A/12160A Technical Manual
QLogic Corporation, ISP1080 Technical Manual
QLogic Corporation, ISP1280 Technical Manual
Diagnostics In addition to being logged, the following messages can appear on the system console.
The first set of messages shown below can be displayed while the qus driver is first trying to
attach. All messages in the first set indicate that the qus driver was unable to attach. Each
message is preceded by "qus<number>)", where "<number>" is the instance number of the
JASPER Host Bus Adapter.

706 man pages section 7: Device and Network Interfaces • Last Revised 29 May 2003
 qus(7D)

Device in slave-only slot, unused The SBus device has been placed in a
slave-only slot and are not accessible; move
to non- slave-only SBus slot.
Device is using a hilevel intr, unused The device was configured with an interrupt
level that cannot be used with the qus driver.
Check the device.
Failed to alloc soft state Driver is unable to allocate space for the
internal state structure. Driver did not attach
to device; SCSI devices are inaccessible.
Bad soft state Driver requested an invalid internal state
structure. Driver did not attach to device;
SCSI devices are inaccessible.
Can't reload firmware: failing attach/resume Driver is unable to reload firmware; check
for bad hardware. Driver did not attach to
device; SCSI devices are inaccessible.
Can't reset interface during attach/resume Driver is unable to reset the hardware. Driver
did not attach to device; SCSI devices are
inaccessible.
Cannot find PCI device-id Driver is unable to find PCI device-id. Driver
did not attach to device; SCSI devices are
inaccessible.
Unable to support ISP chip Driver is unable to support the ISP chip,
which is in the interface. Driver did not
attach to device; SCSI devices are
inaccessible.
Unable to map pci config registers Driver is unable to map device registers.
Check for bad hardware. Driver did not
attach to device. SCSI devices are
inaccessible.
Unable to attach: check for hardware problem Driver is unable to attach to the hardware.
Driver did not attach to device; SCSI devices
are inaccessible.

The following messages can be displayed at any time and are printed with the full device
pathname followed by the shorter form described above.
Firmware (<actual-size>) should be <allowed_size> bytes
Firmware id(<id>) verification failed
Firmware length too short
Firmware checksum incorrect

Device and Network Interfaces 707

qus(7D)

Can't find firmware rev. string

reset/init ISP chip failed
reset/init ISP chip failed
Load ram failed
Start firmware mailbox command failed
Can't set clock rate
Can't get RAM info
These messages indicate a firmware download failure and possible corruption of the
firmware. Check the ISP driver.
Chip reset timeout
ISP chip failed to reset in the time allocated. Potential hardware problem.
Bad request pkt payload
The ISP Firmware rejected the packet, indicating that the packet was set up incorrectly. As
a result, the qus driver calls the target completion routine with the reason of
CMD_TRAN_ERR set in the scsi_pkt. To correctly set up the packet, check the target
driver.
Bad request pkt header
The ISP Firmware rejected the packet because the packet was set up incorrectly. As a result,
the qus driver calls the target completion routine with the reason of CMD_TRAN_ERR set
in the scsi_pkt. To correctly set up the packet, check the target driver.
Target synch. rate reduced. tgt <target-id>
Target<target-id> reducing transfer rate
These messages indicate that the target is reducing its transfer rate. Reboot the system to
obtain the maximum transfer rate.
Failed to Get Features
Chip reset detected
These messages indicate a possible ISP chip failure. Driver attempts to recover from this
condition by reloading and restarting the firmware.
Interface going offline
Although all driver recovery procedure are completed, the interface did not come online
and might need replacement.
SCSI Cable/Connection problem
The SCSI cable is faulty or is connected improperly.
Hardware/Firmware error
The ISP chip encountered a firmware error that is probably due to a faulty SCSI cable or
improper cable connection. As a result, the qus driver attempts to do error recovery by
resetting the chip.
Received unexpected SCSI Reset
The ISP chip received an unexpected SCSI Reset and has initiated its own internal error
recovery, which returns scsi_pkt with reason set to CMD_RESET.

708 man pages section 7: Device and Network Interfaces • Last Revised 29 May 2003
 qus(7D)

Fatal error, resetting interface

The qus driver is performing error recovery. As a result, all outstanding commands that
have been transported to the qus driver is completed by way of the scsi_pkt completion
routine in the target driver with reason of CMD_RESET and status of STAT_BUS_RESET
set in scsi_pkt.
Phase skipped: Command completed with good status — Potential data path failure:
Possible SCSI cable failure. Driver might still be able to communicate with the target and
continue sending commands. Leaving the system in the same state triggers this message
once per hour. System I/O throughput might be reduced. The SCSI might need to be
replaced.
LVD Error detected
SCSI Cable/Connection problem
Bus Not Terminated
f/w initiated BDR fails
These messages indicate a possible cable failure. Outstanding commands are returned with
CMD_RESET or CMD_TRAN_ERR set in scsi_pkt.

Device and Network Interfaces 709

radeon(7d)

Name radeon – DRI (Direct Rendering Infrastructure)-compliant kernel driver providing

3-dimensional graphic hardware acceleration support.

Description The radeon driver is a DRI-compliant kernel driver that provides graphics hardware
acceleration support. DRI is a framework for coordinating OS kernel, 3D graphics hardware,
X window system and OpenGL applications.

The radeon driver currently supports certain low-end ATI radeon graphics cards, including
Radeon X700.
Files /kernel/drv/radeon 32-bit ELF kernel module (x86).
/kernel/drv/amd64/radeon 64-bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdrmr

Architecture x86

Interface Stability Committed

See Also /usr/X11/share/man/man1/Xserver.1

/usr/X11/share/man/man1/Xorg.1

/usr/X11/share/man/man5/X11.5

attributes(5)

710 man pages section 7: Device and Network Interfaces • Last Revised 24 Jun 2009
 ral(7D)

Name ral – Ralink RT2500 802.11b/g Wireless driver

Description The ral 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting Ralink RT2500 chipset-based NIC's.

Configuration The ral driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 6, 9,
11, 12, 18, 24, 36, 48 and 54 Mbits/sec. The ral driver supports only BSS networks (also
known as "ap" or "infrastructure" networks) and "open"(or "open-system") or "shared
system" authentication.
Files /dev/ral*
Special character device.
/kernel/drv/ral
32-bit ELF kernel module (x86).
/kernel/drv/amd64/ral
64-bit ELF kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWralink

Interface stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification — IEEE, 2001

Device and Network Interfaces 711

ramdisk(7D)

Name ramdisk – RAM disk device driver

Synopsis ramdisk@0:diskname

Description The ramdisk driver supports numerous ramdisk devices that are created by the system during
the boot process (see boot(1M)) or during normal system operation (see ramdiskadm(1M) for
more information).

Device Special Each ramdisk can be accessed either as a block device or as a raw device. When accessed as a
Files block device, the normal buffering mechanism is used when reading from and writing to the
device, without regard to physical disk records. Accessing the ramdisk as a raw device enables
direct transmission between the disk and the read or write buffer. A single read or write call
usually results in a single I/O operation, meaning that raw I/O is more efficient when many
bytes are transmitted. You can find block files names in /dev/ramdisk. Raw file names are
found in /dev/rramdisk.

There are no alignment or length restrictions on I/O requests to either block or character
devices.
Errors EFAULT The argument features a bad address.
EINVAL Invalid argument. EIO. An I/O error occurred.
EPERM Cannot create or delete a ramdisk without write permission on /dev/ramdiskctl.
ENOTTY The device does not support the requested ioctl function.
ENXIO The device did not exist during opening.
EBUSY Cannot exclusively open /dev/ramdiskctl. One or more ramdisks are still open.
EEXIST A ramdisk with the indicated name already exists.
EAGAIN Cannot allocate resource for ramdisk. Try again later.
Files /dev/ramdisk/diskname Block device for ramdisk named diskname.
/dev/rramdisk/diskname Raw device for ramdisk name diskname
/kernel/drv/ramdisk 32-bit driver
/kernel/drv/ramdisk.conf Driver configuration file. (Do not alter).
/kernel/drv/sparcv9/ramdisk 64-bit driver

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Evolving

712 man pages section 7: Device and Network Interfaces • Last Revised 04 Mar 2003
 ramdisk(7D)

See Also ramdiskadm(1M), fsck(1M), fstyp(1M), mount(1M), newfs(1M), driver.conf(4),

filesystem(5), dkio(7I)

Notes The percentage of available physical memory that can be allocated to ramdisks is constrained
by the variable rd_percent_physmem. You can tune the rd_percent_physmem variable in
/etc/system. By default, the percentage of available physical memory that can be allocated to
ramdisks is fixed at 25%.

A ramdisk may not be the best possible use of system memory. Accordingly, use ramdisks only
when absolutely necessary.

Device and Network Interfaces 713

random(7D)

Name random, urandom – Strong random number generator device

Synopsis /dev/random
/dev/urandom

Description The /dev/random and /dev/urandom files are special files that are a source for random bytes
generated by the kernel random number generator device. The /dev/random and
/dev/urandom files are suitable for applications requiring high quality random numbers for
cryptographic purposes.

The generator device produces random numbers from data and devices available to the kernel
and estimates the amount of randomness (or "entropy”) collected from these sources. The
entropy level determines the amount of high quality random numbers that are produced at a
given time.

Applications retrieve random bytes by reading /dev/random or /dev/urandom. The

/dev/random interface returns random bytes only when sufficient amount of entropy has been
collected. If there is no entropy to produce the requested number of bytes, /dev/random
blocks until more entropy can be obtained. Non-blocking I/O mode can be used to disable the
blocking behavior. The /dev/random interface also supports poll(2). Note that using poll(2)
will not increase the speed at which random numbers can be read.

Bytes retrieved from /dev/random provide the highest quality random numbers produced by
the generator, and can be used to generate long term keys and other high value keying
material.

The /dev/urandom interface returns bytes regardless of the amount of entropy available. It
does not block on a read request due to lack of entropy. While bytes produced by the
/dev/urandom interface are of lower quality than bytes produced by /dev/random, they are
nonetheless suitable for less demanding and shorter term cryptographic uses such as short
term session keys, paddings, and challenge strings.

Data can be written to /dev/random and /dev/urandom. Data written to either special file is
added to the generator's internal state. Data that is difficult to predict by other users may
contribute randomness to the generator state and help improve the quality of future generated
random numbers.

/dev/random collects entropy from providers that are registered with the kernel-level
cryptographic framework and implement random number generation routines. The
cryptoadm(1M) utility allows an administrator to configure which providers will be used with
/dev/random.
Errors EAGAIN O_NDELAY or O_NONBLOCK was set and no random bytes are available for reading
from /dev/random.
EINTR A signal was caught while reading and no data was transferred.
ENOXIO open(2) request failed on /dev/random because no entropy provider is available.

714 man pages section 7: Device and Network Interfaces • Last Revised 1 Sep 2008
 random(7D)

Files /dev/random
/dev/urandom

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsr

Interface Stability Evolving

See Also cryptoadm(1M), open(2), poll(2), attributes(5)

Notes /dev/random can be configured to use only the hardware-based providers registered with the
kernel-level cryptographic framework by disabling the software-based provider using
cryptoadm(1M). You can also use cryptoadm(1M) to obtain the name of the software-based
provider.

Because no entropy is available, disabling all randomness providers causes read(2) and
poll(2) on /dev/random to block indefinitely and results in a warning message being logged
and displayed on the system console. However, read(2) and poll(2) on /dev/urandom
continue to work in this case.

An implementation of the /dev/random and /dev/urandom kernel-based random number

generator first appeared in Linux 1.3.30.

A /dev/random interface for Solaris first appeared as part of the CryptoRand implementation.

Device and Network Interfaces 715

rarp(7P)

Name rarp, RARP – Reverse address resolution protocol

Description You use the RARP protocol to map dynamically between the Internet Protocol (IP) and
network interface MAC addresses. RARP is often used to boot a Solaris client. RARP clients
include the SPARC boot PROM, x86 boot floppy, SunOS kernel, and ifconfig(1M).
in.rarpd(1M) provides the server-side implementation.

RARP request timeout behavior in application-layer clients is governed by the

/etc/inet/rarp default file. To tune the number of retries an application attempts before
giving up, set the RARP_RETRIES variable in /etc/inet/rarp. If the file is not present or
RARP_RETRIES is not initialized within it, applications retry a maximum of five times with a
eight second wait between retries.

Files /etc/inet/rarp

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability (protocol) Standard

Interface Stability (defaults file) Unstable

Interface Stability (RARP_RETRIES) Unstable

See Also ifconfig(1M), in.rarpd(1M), arp(7P)

Reverse Address Resolution Protocol RFC 903. June, 1984 R. Finlayson, T. Mann, J.C. Mogul, M.
Theimer

716 man pages section 7: Device and Network Interfaces • Last Revised 28 May 2005
 rge(7D)

Name rge – Realtek Gigabit/Fast Ethernet Network Adapter driver

Synopsis /dev/rge
Description The rge Gigabit/Fast Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on the Realtek
Gigabit/Fast Ethernet Network Adapter.
The rge driver functions includes controller initialization, frame transmit and receive,
promiscuous and multicast support, and error recovery and reporting.
Application The cloning, character-special device /dev/rge is used to access all Realtek Gigabit/Fast
Programming Ethernet devices installed within the system.
Interface
The rge driver is managed by the dladm(1M) command line utility, which allows VLANs to be
defined on top of rge instances and for rge instances to be aggregated. See dladm(1M) for
more details.
The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are as follows:
■ Maximum SDU (with jumbo frame) is 7000.
■ Minimum SDU is 0.
■ DSLAP address length is 8 bytes.
■ MAC type is DL_ETHER.
■ SAP length value is -2, meaning the physical address component is followed immediately
by a 2-byte sap component within the DLSAP address.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Once in the DL_ATTACHED state, you must send a DL_BIND_REQ to associate a particular
Service Access Point (SAP) with the stream.
Configuration By default, the rge driver performs auto-negotiation to select the link speed and mode. Link
speed and mode can be any one of the following:
■ 1000 Mbps, full-duplex
■ 100 Mbps, full-duplex
■ 100 Mbps, half-duplex
■ 10 Mbps, full-duplex
■ 10 Mbps, half-duplex
Alternatively, you can set the capabilities advertised by the rge device using ndd(1M). The
driver supports a number of parameters whose names begin with adv_. Each of these
parameters contains a boolean value that determines if the device advertises that mode of
operation. The adv_pause_cap indicates if half/full duplex pause is advertised to link partner.
You can set adv_asym_pause_cap to advertise to the link partner that asymmetric pause is
desired.

Device and Network Interfaces 717

rge(7D)

For example, to prevent the device 'rge2' from advertising gigabit capabilities, enter (as
super-user):
# ndd -set /dev/rge2 adv_1000fdx_cap 0

All capabilities default to enabled. Note that changing any capability parameter causes the link
to go down while the link partners renegotiate the link speed/duplex using the newly changed
capabilities.

You can find the current parameter settings by using ndd -get. In addition, the driver exports
the current state, speed, duplex setting, and working mode of the link via ndd parameters
(these are read only and may not be changed). For example, to check link state of device rge0:
# ndd -get /dev/rge0 link_status
1
# ndd -get /dev/rge0 link_speed
100
# ndd -get /dev/rge0 link_duplex
2

The output above indicates that the link is up and running at 100Mbps full-duplex. In
addition, the driver exports its working mode by loop_mode. If it is set to 0, the loopback mode
is disabled.
Files /dev/rge* Character special device.
/kernel/drv/rge 32–bit x86 rge driver binary.
/kernel/drv/amd64/rge 64–bit x86 rge driver binary.
/kernel/drv/sparcv9/rge SPARC rge driver binary.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), attributes(5), streamio(7I), dlpi(7P)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

718 man pages section 7: Device and Network Interfaces • Last Revised 21 Dec 2007
 route(7P)

Name route – kernel packet forwarding database

Synopsis #include <sys/types.h>
#include <sys/socket.h>
#include <net/if.h>
#include <net/route.h>

int socket(PF_ROUTE, SOCK_RAW, int protocol);

Description UNIX provides some packet routing facilities. The kernel maintains a routing information
database, which is used in selecting the appropriate network interface when transmitting
packets.
A user process (or possibly multiple co-operating processes) maintains this database by
sending messages over a special kind of socket. This supplants fixed size ioctl(2)'s specified in
routing(7P). Routing table changes can only be carried out by the superuser.
The operating system might spontaneously emit routing messages in response to external
events, such as receipt of a re-direct, or failure to locate a suitable route for a request. The
message types are described in greater detail below.
Routing database entries come in two flavors: entries for a specific host, or entries for all hosts
on a generic subnetwork (as specified by a bit mask and value under the mask). The effect of
wildcard or default route can be achieved by using a mask of all zeros, and there can be
hierarchical routes.
When the system is booted and addresses are assigned to the network interfaces, the internet
protocol family installs a routing table entry for each interface when it is ready for traffic.
Normally the protocol specifies the route through each interface as a direct connection to the
destination host or network. If the route is direct, the transport layer of a protocol family
usually requests the packet be sent to the same host specified in the packet. Otherwise, the
interface is requested to address the packet to the gateway listed in the routing entry, that is,
the packet is forwarded.
When routing a packet, the kernel attempts to find the most specific route matching the
destination. If no entry is found, the destination is declared to be unreachable, and a
routing-miss message is generated if there are any listeners on the routing control socket
(described below). If there are two different mask and value-under-the-mask pairs that match,
the more specific is the one with more bits in the mask. A route to a host is regarded as being
supplied with a mask of as many ones as there are bits in the destination.
A wildcard routing entry is specified with a zero destination address value, and a mask of all
zeroes. Wildcard routes are used when the system fails to find other routes matching the
destination. The combination of wildcard routes and routing redirects can provide an
economical mechanism for routing traffic.
One opens the channel for passing routing control messages by using the socket call. There
can be more than one routing socket open per system.

Device and Network Interfaces 719

route(7P)

Messages are formed by a header followed by a small number of sockaddrs, whose length
depend on the address family. sockaddrs are interpreted by position. An example of a type of
message with three addresses might be a CIDR prefix route: Destination, Netmask, and
Gateway. The interpretation of which addresses are present is given by a bit mask within the
header, and the sequence is least significant to most significant bit within the vector.

Any messages sent to the kernel are returned, and copies are sent to all interested listeners.
The kernel provides the process ID of the sender, and the sender can use an additional
sequence field to distinguish between outstanding messages. However, message replies can be
lost when kernel buffers are exhausted.

The protocol parameter specifies which messages an application listening on the routing
socket is interested in seeing, based on the the address family of the sockaddrs present.
Currently, you can specify AF_INET and AF_INET6 to filter the messages seen by the listener, or
alternatively, you can specify AF_UNSPEC to indicate that the listener is interested in all routing
messages.

The kernel might reject certain messages, and indicates this by filling in the rtm_errno field of
the rt_msghdr struct (see below). The following codes are returned:
EEXIST If requested to duplicate an existing entry
ESRCH If requested to delete a non-existent entry
ENOBUFS If insufficient resources were available to install a new route.
EPERM If the calling process does not have appropriate privileges to alter the routing
table.

In the current implementation, all routing processes run locally, and the values for rtm_errno
are available through the normal errno mechanism, even if the routing reply message is lost.

A process can avoid the expense of reading replies to its own messages by issuing a
setsockopt(3SOCKET) call indicating that the SO_USELOOPBACK option at the SOL_SOCKET
level is to be turned off. A process can ignore all messages from the routing socket by doing a
shutdown(3SOCKET) system call for further input.

By default, underlying IP interfaces in an IPMP group are not visible to routing sockets. As
such, routing sockets do not receive events related to underlying IP interface in an IPMP
group. For consistency, when an IP interface is placed into an IPMP group, RTM_DELADDR
messages are generated for each IFF_UP address that is not migrated to the corresponding
IPMP IP interface and an RTM_IFINFO message is sent indicating the interface is down.
Similarly, when an underlying interface is removed from an IPMP group, an RTM_IFINFO
message is sent indicating the interface is again up and RTM_NEWADDR messages are generated
for each IFF_UP address found on the interface.

720 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 route(7P)

The RT_AWARE socket option at the SOL_ROUTE level allows an application to indicate its
awareness of certain features, which control routing socket behavior. The supported values
are:
RTAW_DEFAULT Default awareness.
RTAW_UNDER_IPMP IPMP underlying interface awareness. When enabled, underlying IP
interfaces in an IPMP group remain visible to the routing socket and
events related to them continue to be generated.

An RTM_ADD request tied to an underlying IP interface in an IPMP group is translated to an

RTM_ADD request for its corresponding IPMP IP interface. All routing socket requests other
than RTM_ADD and RTM_GET fail when issued on an underlying IP interface in an IPMP group.

If a route is in use when it is deleted, the routing entry is marked down and removed from the
routing table, but the resources associated with it are not reclaimed until all references to it are
released.

The RTM_IFINFO, RTM_NEWADDR, and RTM_ADD messages associated with interface configuration
(setting the IFF_UP bit) are normally delayed until after Duplicate Address Detection
completes. Thus, applications that configure interfaces and wish to wait until the interface is
ready can wait until RTM_IFINFO is returned and SIOCGLIFFLAGS shows that IFF_DUPLICATE is
not set.

Messages User processes can obtain information about the routing entry to a specific destination by
using a RTM_GET message.

Messages include:
#define RTM_ADD 0x1 /* Add Route */
#define RTM_DELETE 0x2 /* Delete Route */
#define RTM_CHANGE 0x3 /* Change Metrics, Flags, or Gateway */
#define RTM_GET 0x4 /* Report Information */
#define RTM_LOSING 0x5 /* Kernel Suspects Partitioning */
#define RTM_REDIRECT 0x6 /* Told to use different route */
#define RTM_MISS 0x7 /* Lookup failed on this address */
#define RTM_LOCK 0x8 /* fix specified metrics */
#define RTM_OLDADD 0x9 /* caused by SIOCADDRT */
#define RTM_OLDDEL 0xa /* caused by SIOCDELRT */
#define RTM_RESOLVE 0xb /* request to resolve dst to LL addr */
#define RTM_NEWADDR 0xc /* address being added to iface */
#define RTM_DELADDR 0xd /* address being removed from iface */
#define RTM_IFINFO 0xe /* iface going up/down etc. */

A message header consists of:

struct rt_msghdr {
ushort_t rtm_msglen; /* to skip over non-understood messages */

Device and Network Interfaces 721

route(7P)

uchar_t rtm_version; /* future binary compatibility */

uchar_t rtm_type; /* message type */
ushort_t rtm_index; /* index for associated ifp */
pid_t rtm_pid; /* identify sender */
int rtm_addrs; /* bitmask identifying sockaddrs in msg */
int rtm_seq; /* for sender to identify action */
int rtm_errno; /* why failed */
int rtm_flags; /* flags, incl kern & message, e.g., DONE */
int rtm_use; /* from rtentry */
uint_t rtm_inits; /* which values we are initializing */

struct rt_metrics rtm_rmx; /* metrics themselves */

};

where
struct rt_metrics {
uint32_t rmx_locks; /* Kernel must leave these values alone */
uint32_t rmx_mtu; /* MTU for this path */
uint32_t rmx_hopcount; /* max hops expected */
uint32_t rmx_expire; /* lifetime for route, e.g., redirect */
uint32_t rmx_recvpipe; /* inbound delay-bandwidth product */
uint32_t rmx_sendpipe; /* outbound delay-bandwidth product */
uint32_t rmx_ssthresh; /* outbound gateway buffer limit */
uint32_t rmx_rtt; /* estimated round trip time */
uint32_t rmx_rttvar; /* estimated rtt variance */
uint32_t rmx_pksent; /* packets sent using this route */
};

/* Flags include the values */

#define RTF_UP 0x1 /* route usable */

#define RTF_GATEWAY 0x2 /* destination is a gateway */
#define RTF_HOST 0x4 /* host entry (net otherwise) */
#define RTF_REJECT 0x8 /* host or net unreachable */
#define RTF_DYNAMIC 0x10 /* created dynamically(by redirect) */
#define RTF_MODIFIED 0x20 /* modified dynamically(by redirect) */
#define RTF_DONE 0x40 /* message confirmed */
#define RTF_MASK 0x80 /* subnet mask present */
#define RTF_CLONING 0x100 /* generate new routes on use */
#define RTF_XRESOLVE 0x200 /* external daemon resolves name */
#define RTF_LLINFO 0x400 /* generated by ARP */
#define RTF_STATIC 0x800 /* manually added */
#define RTF_BLACKHOLE 0x1000 /* just discard pkts (during updates) */
#define RTF_PRIVATE 0x2000 /* do not advertise this route */
#define RTF_PROTO2 0x4000 /* protocol specific routing flag #2 */
#define RTF_PROTO1 0x8000 /* protocol specific routing flag #1 */

722 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2009
 route(7P)

#define RTF_MULTIRT 0x10000 /* multiroute */

#define RTF_SETSRC 0x20000 /* set default outgoing src address */
#define RTF_INDIRECT 0x40000 /* gateway not directly reachable */
#define RTF_KERNEL 0x80000 /* created by kernel; can’t delete */

/* Specifiers for metric values in rmx_locks and rtm_inits are */

#define RTV_MTU 0x1 /* init or lock _mtu */

#define RTV_HOPCOUNT 0x2 /* init or lock _hopcount */
#define RTV_EXPIRE 0x4 /* init or lock _expire */
#define RTV_RPIPE 0x8 /* init or lock _recvpipe */
#define RTV_SPIPE 0x10 /* init or lock _sendpipe */
#define RTV_SSTHRESH 0x20 /* init or lock _ssthresh */
#define RTV_RTT 0x40 /* init or lock _rtt */
#define RTV_RTTVAR 0x80 /* init or lock _rttvar */

/* Specifiers for which addresses are present in the messages are */

#define RTA_DST 0x1 /* destination sockaddr present */

#define RTA_GATEWAY 0x2 /* gateway sockaddr present */
#define RTA_NETMASK 0x4 /* netmask sockaddr present */
#define RTA_GENMASK 0x8 /* cloning mask sockaddr present */
#define RTA_IFP 0x10 /* interface name sockaddr present */
#define RTA_IFA 0x20 /* interface addr sockaddr present */
#define RTA_AUTHOR 0x40 /* sockaddr for author of redirect */
#define RTA_BRD 0x80 /* for NEWADDR, broadcast or p-p dest addr */

See Also ioctl(2), setsockopt(3SOCKET), shutdown(3SOCKET), routing(7P)

Notes Some of the metrics might not be implemented and return zero. The implemented metrics are
set in rtm_inits.

The RTF_INDIRECT flag allows adding routes where the gateway is not directly reachable.
When an indirect route is the best match for a packet to be sent or forwarded, then IP proceeds
to lookup that gateway to find a route that is directly reachable. The RTF_INDIRECT flag can
be used even if the gateway is directly reachable.

When the routing table contains several equal routes, that is, routes for the same destination
and mask, then IP attempts to spread the traffic over those routes. The spreading is such that
an individual transport connection uses the same route to avoid packet reordering as seen by
e.g., TCP. The details of the spreading algoritm is not documented and is likely to evolve over
time.

Device and Network Interfaces 723

routing(7P)

Name routing – system support for packet network routing

Description The network facilities provide general packet routing. The routing interface described here
can be used to maintain the system's IPv4 routing table. It has been maintained for
compatibility with older applications. The recommended interface for maintaining the
system's routing tables is the routing socket, described at route(7P). The routing socket can be
used to manipulate both the IPv4 and IPv6 routing tables of the system. Routing table
maintenance may be implemented in applications processes.

A simple set of data structures compose a “routing table” used in selecting the appropriate
network interface when transmitting packets. This table contains a single entry for each route
to a specific network or host. The routing table was designed to support routing for the
Internet Protocol (IP), but its implementation is protocol independent and thus it may serve
other protocols as well. User programs may manipulate this data base with the aid of two
ioctl(2) commands, SIOCADDRT and SIOCDELRT. These commands allow the addition and
deletion of a single routing table entry, respectively. Routing table manipulations may only be
carried out by privileged user.

A routing table entry has the following form, as defined in /usr/include/net/route.h:

struct rtentry {
unit_t rt_hash; /* to speed lookups */
struct sockaddr rt_dst; /* key */
struct sockaddr rt_gateway; /* value */
short rt_flags; /* up/down?, host/net */
short rt_refcnt; /* # held references */
unit_t rt_use; /* raw # packets forwarded */
/*
* The kernel does not use this field, and without it the structure is
* datamodel independent.
*/
#if !defined(_KERNEL)
struct ifnet *rt_ifp; /* the answer: interface to use */
#endif /* !defined(_KERNEL) */
};

with rt_flags defined from:

#define RTF_UP 0x1 /* route usable */
#define RTF_GATEWAY 0x2 /* destination is a gateway */
#define RTF_HOST 0x4 /* host entry (net otherwise) */

There are three types of routing table entries: those for a specific host, those for all hosts on a
specific network, and those for any destination not matched by entries of the first two types,
called a wildcard route. Each network interface installs a routing table entry when it is
initialized. Normally the interface specifies if the route through it is a “direct” connection to
the destination host or network. If the route is direct, the transport layer of a protocol family

724 man pages section 7: Device and Network Interfaces • Last Revised 9 Nov 1999
 routing(7P)

usually requests the packet be sent to the same host specified in the packet. Otherwise, the
interface may be requested to address the packet to an entity different from the eventual
recipient; essentially, the packet is forwarded.

Routing table entries installed by a user process may not specify the hash, reference count, use,
or interface fields; these are filled in by the routing routines. If a route is in use when it is
deleted, meaning its rt_refcnt is non-zero, the resources associated with it will not be
reclaimed until all references to it are removed.

User processes read the routing tables through the /dev/ip device.

The rt_use field contains the number of packets sent along the route. This value is used to
select among multiple routes to the same destination. When multiple routes to the same
destination exist, the least used route is selected.

A wildcard routing entry is specified with a zero destination address value. Wildcard routes
are used only when the system fails to find a route to the destination host and network. The
combination of wildcard routes and routing redirects can provide an economical mechanism
for routing traffic.
Errors EEXIST A request was made to duplicate an existing entry.
ESRCH A request was made to delete a non-existent entry.
ENOBUFS Insufficient resources were available to install a new route.
ENOMEM Insufficient resources were available to install a new route.
ENETUNREACH The gateway is not directly reachable. For example, it does not match the
destination/subnet on any of the network interfaces.
Files /dev/ip IP device driver

See Also route(1M), ioctl(2), route(7P)

Device and Network Interfaces 725

rtls(7D)

Name rtls – driver for Realtek 8139 fast Ethernet controllers

Description The rtls driver supports network interfaces based on the Realtek 8139 family of fast Ethernet
controllers. These devices have an integrated 10BASE-T and 100BASE-TX PHY, and support
IEEE 802.3 auto-negotiation of link speed and duplex mode.

The link settings can be viewed or modified using dladm(1M).

Files /kernel/drv/rtls 32–bit driver binary (x86)
/kernel/drv/amd64/rtls 64–bit driver binary (x86)
/kernel/drv/sparcv9/rtls 64–bit driver binary (SPARC)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

See Also dladm(1M), netstat(1M), attributes(5), ieee802.3(5), dlpi(7P)

726 man pages section 7: Device and Network Interfaces • Last Revised 16 Jan 2010
 rtw(7D)

Name rtw – RealTek 8180L 802.11b Wireless NIC driver

Description The rtw 802.11b wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting RealTek 8180L chipset-based NIC's.

Configuration The rtw driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. The default is 11.

The rtw driver supports only BSS networks (also known as "ap" or "infrastructure" networks)
and "open"(or "open-system") or "shared system" authentication.
Files /dev/rtw*
Special character device.
/kernel/drv/rtw
32-bit ELF kernel module (x86).
/kernel/drv/amd64/rtw
64-bit ELF kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWrtw

Interface stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification — IEEE, 2001

Device and Network Interfaces 727

rum(7D)

Name rum – Ralink RT2501/RT2601/RT73USB 802.11b/g Wireless Driver

Description The rum 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Ralink RT2501/RT2601/RT73USB chipset-based NIC's.

Configuration The rum driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec. The rum driver supports only BSS networks (also
known as "ap" or "infrastructure" networks) and "open" (or "open-system") or "shared
system" authentication.
Files /dev/rum*
Special character device.
/kernel/drv/rum
32-bit ELF kernel module. (x86)
/kernel/drv/amd64/rum
64-bit ELF kernel module. (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWrum

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification - IEEE, 2001

728 man pages section 7: Device and Network Interfaces • Last Revised 29 Oct 2008
 rwd(7D)

Name rwd – Ralink RT2561/RT2561S/RT2661 IEEE802.11b/g wireless network driver

Description The rwd IEEE802.11b/g wireless network driver is a multithreaded, loadable, clonable,
GLDv3-based STREAMS driver supporting Ralink RT2561/RT2561S/RT2661 IEEE802.11b/g
wireless network driver.

Configuration The rwd driver performs auto-negotiation to determine the data rate and mode. The driver
supports only BSS networks (also known as “ap” or “infrastructure” networks) and “open”(or
“open-system”) or “shared system” authentication. For wireless security, WEP encryption,
WPA-PSk, and WPA2-PSK are currently supported. You can perform configuration and
administration tasks using the dladm(1M)and wificonfig(1M) utilities.
Files /dev/rwd Special character device
/kernel/drv/rwd 32-bit ELF kernel module (x86)
/kernel/drv/amd64/rwd 64-bit ELF kernel module (x86)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWrwd

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

IEEE802.11b/g - Wireless LAN Standard - IEEE, 2003

Device and Network Interfaces 729

rwn(7D)

Name rwn – Ralink RT2700/2800 IEEE802.11 a/b/g/n wireless network device

Description The rwn IEEE802.11 a/b/g/n wireless driver is a multi-threaded, loadable, clonable,
GLDv3-based STREAMS driver supporting Ralink RT2700/RT2800 IEEE802.11 a/b/g/n
wireless network device.

Configuration The rwn driver performs auto-negotiation to determine the data rate and mode. The driver
supports only BSS networks (also known as ap or infrastructure networks) and open
(open-system) or shared system authentication.

For wireless security, WEP encryption, WPA-PSK, and WPA2-PSK are currently supported.
You can perform configuration and administration tasks using the dladm(1M) and
wificonfig(1M) utilities.
Files /dev/rwn Special character device
/kernel/drv/rwn 32-bit ELF kernel module, x86
/kernel/drv/amd64/rwn 64-bit ELF kernel module, x86

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWrwn

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5)

IEEE802.11b/g - Wireless LAN Standard - IEEE, 2003

730 man pages section 7: Device and Network Interfaces • Last Revised 12 Oct 2009
 sad(7D)

Name sad – STREAMS Administrative Driver

Synopsis #include <sys/types.h>
#include <sys/conf.h>
#include <sys/sad.h>
#include <sys/stropts.h>
int ioctl(int fildes, int command, int arg);

Description The STREAMS Administrative Driver provides an interface for applications to perform
administrative operations on STREAMS modules and drivers. The interface is provided
through ioctl(2) commands. Privileged operations may access the sad driver using
/dev/sad/admin. Unprivileged operations may access the sad driver using /dev/sad/user.

The fildes argument is an open file descriptor that refers to the sad driver. The command
argument determines the control function to be performed as described below. The arg
argument represents additional information that is needed by this command. The type of arg
depends upon the command, but it is generally an integer or a pointer to a command-specific
data structure.

Command The autopush facility (see autopush(1M)) allows one to configure a list of modules to be
Functions automatically pushed on a stream when a driver is first opened. Autopush is controlled by the
following commands:
SAD_SAP Allows the administrator to configure the given device's autopush information.
arg points to a strapush structure, which contains the following members:
unit_t ap_cmd;
major_t sap_major;
minor_t sap_minor;
minor_t sap_lastminor;
unit_t sap_npush;
unit_t sap_list [MAXAPUSH] [FMNAMESZ + 1];
The sap_cmd field indicates the type of configuration being done. It may take on
one of the following values:
SAP_ONE Configure one minor device of a driver.
SAP_RANGE Configure a range of minor devices of a driver.
SAP_ALL Configure all minor devices of a driver.
SAP_CLEAR Undo configuration information for a driver.

The sap_major field is the major device number of the device to be configured.
The sap_minor field is the minor device number of the device to be configured.
The sap_lastminor field is used only with the SAP_RANGE command, which
configures a range of minor devices between sap_minor and sap_lastminor,

Device and Network Interfaces 731

sad(7D)

inclusive. The minor fields have no meaning for the SAP_ALL command. The
sap_npush field indicates the number of modules to be automatically pushed
when the device is opened. It must be less than or equal to MAXAPUSH , defined in
sad.h. It must also be less than or equal to NSTRPUSH, the maximum number of
modules that can be pushed on a stream, defined in the kernel master file. The
field sap_list is an array of NULL-terminated module names to be pushed in
the order in which they appear in the list.

When using the SAP_CLEAR command, the user sets only sap_major and
sap_minor. This will undo the configuration information for any of the other
commands. If a previous entry was configured as SAP_ALL, sap_minor should be
set to zero. If a previous entry was configured as SAP_RANGE , sap_minor should
be set to the lowest minor device number in the range configured.

On failure, errno is set to the following value:

EFAULT arg points outside the allocated address space.
EINVAL The major device number is invalid, the number of modules is
invalid, or the list of module names is invalid.
ENOSTR The major device number does not represent a STREAMS driver.
EEXIST The major-minor device pair is already configured.
ERANGE The command is SAP_RANGE and sap_lastminor is not greater than
sap_minor, or the command is SAP_CLEAR and sap_minor is not
equal to the first minor in the range.
ENODEV The command is SAP_CLEAR and the device is not configured for
autopush.
ENOSR An internal autopush data structure cannot be allocated.
SAD_GAP Allows any user to query the sad driver to get the autopush configuration
information for a given device. arg points to a strapush structure as described in
the previous command.

The user should set the sap_major and sap_minor fields of the strapush
structure to the major and minor device numbers, respectively, of the device in
question. On return, the strapush structure will be filled in with the entire
information used to configure the device. Unused entries in the module list will
be zero-filled.

On failure, errno is set to one of the following values:

EFAULT arg points outside the allocated address space.
EINVAL The major device number is invalid.

732 man pages section 7: Device and Network Interfaces • Last Revised 16 Apr 1997
 sad(7D)

ENOSTR The major device number does not represent a STREAMS driver.
ENODEV The device is not configured for autopush.
SAD_VML Allows any user to validate a list of modules (that is, to see if they are installed on
the system). arg is a pointer to a str_list structure with the following members:
int sl_nmods;
struct str_mlist *sl_modlist;
The str_mlist structure has the following member:
char l_name[FMNAMESZ+1];
sl_nmods indicates the number of entries the user has allocated in the array and
sl_modlist points to the array of module names. The return value is 0 if the list
is valid, 1 if the list contains an invalid module name, or −1 on failure. On failure,
errno is set to one of the following values:
EFAULT arg points outside the allocated address space.
EINVAL The sl_nmods field of the str_list structure is less than or equal to
zero.

See Also Intro(2), ioctl(2), open(2)

STREAMS Programming Guide

Diagnostics Unless otherwise specified, the return value from ioctl() is 0 upon success and −1 upon
failure with errno set as indicated.

Device and Network Interfaces 733

sata(7D)

Name sata – Solaris SATA framework

Description Serial ATA is an interconnect technology designed to replace parallel ATA technology. It is
used to connect hard drives, optical drives, removable magnetic media devices and other
peripherals to the host system. For complete information on Serial ATA technology, visit the
Serial ATA web site at http://www.serialata.org.

Up to 32 SATA devices may be plugged directly to each SATA HBA and up to 15 SATA devices
may be plugged directly to each SATA port multiplier supported by the Solaris SATA
framework. The actual number of pluggable devices my be lower, and is limited by the
number of device ports on the SATA HBA or the SATA port multiplier. The maximum data
rate is either 1.5Gb/sec. or 3.0Gb/sec., depending on the capability of a SATA device, port
multiplier and SATA HBA controller.

The Solaris SATA framework adheres to the Serial ATA 1.0a specification and supports
SATA-2 signaling speed 3.0Gb/sec. SATA devices that are connected to SATA HBAs
controlled by a SATA framework-compliant HBA driver are treated by the system as SCSI
devices. The Solaris SCSI disk driver (sd(7D)) is attached as a target driver for each device
node created by the SATA framework. You can use the cfgadm(1M) utility to manage hot
plugged and unplugged SATA devices.
Files /kernel/misc/sata 32–bit ELF kernel module (x86).
/kernel/misc/amd64/sata 64–bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWckr

See Also cfgadm(1M), prtconf(1M), cfgadm_sata(1M), attributes(5), ahci(7D), marvell88sx(7D),

nv_sata(7D), sd(7D), si3124(7D)
Serial ATA 1.0a Specification — Serial ATA International Organization.
Serial ATA II (Extension to Serial ATA 1.0.a.) — Serial ATA International Organization.
http://www.sun.com/
Diagnostics The messages described below may appear on the system console as well as being logged. All
messages are presented in one of the following formats and are followed by the diagnostic
message:
sata: WARNING: <controller/devices/.. path>:

or:

734 man pages section 7: Device and Network Interfaces • Last Revised 10 Sep 2009
 sata(7D)

sata: NOTICE: <controller/devices/.. path>:

...where <controller/devices/.. path> identifies a specific SATA HBA issuing a diagnostic

message shown below.
SATA port X: link lost.
Communication (via serial link) between the HBA and the device plugged to the specified
SATA device port has been lost.
SATA port X: link established.
Communication (via serial link) between the HBA and the device plugged to the specified
SATA device port has been established.
SATA port X: device reset.
The device plugged to the specified SATA device port has been reset. The reset may be due
to a communication or command error, command timeout, or an explicit request from the
host.
SATA port X failed.
The specified SATA device port failed and is in an unusable state. You can change the port
state by deactivating the port and activating it again using cfgadm SATA hardware-specific
commands (see cfgadm_sata(1M)).
SATA port X error.
An error was detected in specified SATA device port operations.
SATA device detached at port X.
Communication (via serial link) between the HBA and the device plugged to the specified
SATA device port has been lost and could not be re-established. The SATA framework
assumes that the device is unplugged from the specified SATA device port.
SATA device detected at port X.
Communication( via serial link) between the HBA and the device plugged to the specified
empty SATA device port has been established. The SATA framework assumes that the new
device is plugged to the specified SATA device port.
SATA disk device at port X.
This message is followed by a disk description specifying the disk vendor, serial number,
firmware revision number and the disk capabilities.
SATA CD/DVD (ATAPI) device at port X.
This message is followed by a SATA CD/DVD description specifying the DVD vendor,
serial number, firmware revision number and the DVD capabilities.
SATA device at port X cannot be configured. Application(s) accessing previously attached
device have to release it before newly inserted device can be made accessible.
The port cannot be configured because there is application using the previous attached
device, so the application must release it, then the newly inserted device can be configured.

Device and Network Interfaces 735

sata(7D)

Application(s) accessing previously attached SATA device have to release it before newly
inserted device can be made accessible.
The target node remained and it belongs to a previously attached device. This happens
when the file was open or the node was waiting for resources at the time the associated
device was removed. Instruct event daemon to retry the cleanup later.
sata: error recovery request for non-attached device at cport X.
When error recovery is requested, the device is not yet attached.
SATA device at port X is not power-managed.
When property pm-capable on the target device node setting fails, the SATA device won't
be power-managed.
SATA disk device at port X does not support LBA.
The disk device plugged into specified SATA device port does not support LBA addressing
and cannot be used.
Cannot identify SATA device at port X - device is attached.
IDENTIFY (PACKET) DEVICE data cannot be retrieved successfully after the device is
attached to the SATA port.
sata: <HBA driver name><instance number>:hba attached failed.
The SATA HBA instance attach operation failed. This HBA instance cannot be configured
and is not available.
sata: invalid ATAPI cdb length<command cdb length>.
The length of the command cdb is greater than that the device can support.
sata: invalid sata_hba_tran version X for driver <HBA driver name>.
The specified SATA HBA driver and the SATA framework are incompatible. The driver
cannot attach and SATA HBAs controlled by this driver (and devices plugged to this SATA
HBA ports) are not available.
sata_hba_attach: cannot create SATA attachment point for port X.
The specified SATA device port cannot be configured in the system and a device plugged to
this port could not be not be configured and used.
sata_create_target_node: cannot create target node for device at port X.
The device target node for the device plugged to the specified SATA device port could not
be created. As a result, the device cannot be configured and used.

736 man pages section 7: Device and Network Interfaces • Last Revised 10 Sep 2009
 scfd(7D)

Name scfd – System Control Facility (SCF) driver

Synopsis scfd@unit-address

Description The System Control Facility (SCF) driver is a device driver that communicates with the
eXtended System Control Facility (XSCF) firmware on a SPARC Enterprise Server.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr

Interface Stability Private

See Also attributes(5)

Device and Network Interfaces 737

schpc(7D)

Name schpc – StarCat Hot Plug Controller Driver

Description The schpc driver controls all hot-plug operations on high-end Sun Fire E15K and E25K
enterprise servers.
Files /platform/SUNW,Sun-Fire-15000/kernel/drv/sparcv9/schpc
64-bit ELF kernel module.
/platform/SUNW,Sun-Fire-15000/kernel/drv/schpc.conf
Driver configuration file.

See Also cfgadm(1M)

738 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2005
 scsa1394(7D)

Name scsa1394 – SCSI to 1394 bridge driver

Synopsis unit@GUID

Description The scsa1394 driver is a 1394 target and an SCSA HBA driver that supports 1394 mass storage
devices compliant with the Serial Bus Protocol 2 (SBP-2) specification. It supports both
bus-powered and self-powered 1394 mass storage devices.

The scsa1394 nexus driver maps SCSA target driver requests to SBP-2 Operation Request
Blocks (ORB's).

The scsa1394 driver creates a child device info node for each logical unit (LUN) on the mass
storage device. The standard Solaris SCSI disk driver is attached to those nodes. Refer to
sd(7D).

This driver supports multiple LUN devices and creates a separate child device info node for
each LUN. All child LUN nodes attach to sd(7D).

In previous releases, all 1394 mass storage devices were treated as removable media devices
and managed by rmformat(1) and volume management software. In the current release,
however, only mass storage devices with a removable bit (RMB) value of 1 are removable.
(The RMB is part of the device's SCSI INQUIRY data.) See SCSI specifications T10/995D
Revision 11a, T10/1236-D Revision 20 or T10/1416-D Revision 23 for more information.
However, for backward compatibility, all 1394 mass storage devices can still be managed by
rmformat(1). With or without a volume manager, you can mount, eject, hot remove and hot
insert a 1394 mass storage device as the following sections explain.

Using Volume Mass storage devices are managed by a volume manager. Software that manages removable
Management media creates a device nickname that can be listed with eject(1) or rmmount(1). A device that
is not mounted automatically can be mounted using rmmount(1) under /rmdisk/label. Note
that the mount(1M) and mount(1M) commands do not accept nicknames; you must use
explicit device names with these commands.

See rmmount(1) to unmount the device and eject(1) to eject the media. If the device is ejected
while it is mounted, volume management software unmounts the device before ejecting it. It
also might kill any active applications that are accessing the device.

Volume management software is hotplug-aware and normally mounts file systems on USB
mass storage devices if the file system is recognized. Before hot removing the USB device, use
eject(1) to unmount the file system.

You can disable the automatic mounting and unmounting of removable devices by inserting a
entry for a removable device in /etc/vfstab. In this entry, you must set the mount at boot
field to no. See vfstab(4). See the System Administration Guide, Volume I and Solaris
Common Desktop Environment: User's Guide for details on how to manage a removable device
with CDE and Removable Media Manager. See dtfile.1X under CDE for information on
how to use Removable Media Manager.

Device and Network Interfaces 739

scsa1394(7D)

Using mount And Use mount(1M) to explicitly mount the device and umount(1M) to unmount the device. Use
umount eject(1) to eject the media. After you have explicitly mounted a removable device, you cannot
use a nickname as an argument to eject.

Removing the storage device while it is being accessed or mounted fails with a console
warning. To hot remove the storage device from the system, unmount the file system, then kill
all applications accessing the device. Next, hot remove the device. A storage device can be hot
inserted at any time.

For a comprehensive listing of (non-bootable) 1394 mass-storage devices that are compatible
with this driver, see www.sun.com/io.

Device Special Block special file names are located in /dev/dsk. Raw file names are located in /dev/rdsk.
Files Input/output requests to the devices must follow the same restrictions as those for SCSI disks.
Refer to sd(7D).

ioctls Refer to cdio(7I) and dkio(7I).

Errors Refer to sd(7D).

Files The device special files for the 1394 mass storage device are created like those for a SCSI disk.
Refer to sd(7D).
/dev/dsk/cntndnsn Block files
/dev/rdsk/cntndnsn Raw files
/vol/dev/aliases/rmdisk0 Symbolic link to the character device for the media in
removable drive 0. This is a generic removable media
device.
/kernel/drv/scsa1394 32–bit x86 ELF kernel module
/kernel/drv/amd64/scsa1394 64–bit x86 ELF kernel module
/kernel/drv/sparcv9/scsa1394 64–bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWscsa1394

See Also cdrw(1), eject(1), rmformat(1), rmmount(1), cfgadm_scsi(1M), fdisk(1M), mount(1M),

umount(1M), dtfile.1X (in CDE man pages), scsi(4), vfstab(4), attributes(5),
hci1394(7D), sd(7D), pcfs(7FS), cdio(7I), dkio(7I)

IEEE Std 1394-1995 Standard for a High Performance Serial Bus

740 man pages section 7: Device and Network Interfaces • Last Revised 2 Mar 2007
 scsa1394(7D)

ANSI NCITS 325-1998 - Serial Bus Protocol 2 (SBP-2)

System Administration Guide: Devices and File Systems

SCSI Specification T10/995D Revision 11a — March 1997

SCSI SpecificationT10/1236-D Revision 20 — July 2001

SCSI SpecificationT10/1416-D Revision 23— May 2005

Solaris Common Desktop Environment: User's Guide

http://www.sun.com/io

Device and Network Interfaces 741

scsa2usb(7D)

Name scsa2usb – SCSI to USB bridge driver

Synopsis storage@unit-address

Description The scsa2usb driver is a USBA (Solaris USB architecture) compliant nexus driver that
supports the USB Mass Storage Bulk Only Transport Specification 1.0 and USB
Control/Bulk/Interrupt (CBI) Transport Specification 1.0. The scsa2usb driver also supports
USB storage devices that implement CBI Transport without the interrupt completion for
status (that is, Control/Bulk (CB) devices.) It supports bus-powered and self-powered USB
mass storage devices. This nexus driver is both a USB client driver and a SCSA HBA driver. As
such, the scsa2usb driver only supports storage devices that utilize the above two transports.

The scsa2usb driver also supports a ugen(7D) interface allowing raw access to the device, for
example by libusb(3LIB) applications, bypassing the child sd(7D) or st(7D) driver. Because
a libusb application might change the state of the device, you should not access the disk or tape
concurrently.

The scsa2usb nexus driver maps SCSA target driver requests to USBA client driver requests.

The scsa2usb driver creates a child device info node for each logical unit (LUN) on the mass
storage device. The standard Solaris SCSI disk driver or tape driver is attached to those nodes.
Refer to sd(7D) or st(7D).

This driver supports multiple LUN devices and creates a separate child device info node for
each LUN. All child LUN nodes attach to sd(7D) for disks or st(7D) for tapes.

In previous releases, all USB disk storage devices were treated as removable media devices and
managed by rmformat(1) and volume management software. In the current release, however,
only disk storage devices with a removable bit (RMB) value of 1 are removable. (The RMB is
part of the device's SCSI INQUIRY data.) See SCSI specifications T10/995D Revision 11a,
T10/1236-D Revision 20 or T10/1416-D Revision 23 for more information. However, for
backward compatibility, all USB disk storage devices can still be managed by rmformat(1).
With or without a volume manager, you can mount, eject, hot remove and hot insert a 1394
mass storage device as the following sections explain.

Some devices may be supported by the USB mass storage driver even though they do not
identify themselves as compliant with the USB mass storage class.

The scsa2usb.conf file contains an attribute-override-list that lists the vendor ID,
product ID, and revision for matching mass storage devices, as well as fields for overriding the
default device attributes. The entries in this list are commented out by default and may be
uncommented to enable support of particular devices.

Follow the information given in the scsa2usb.conf file to see if a particular device can be
supported using the override information. Also see http://www.sun.com/io. For example, by
adding the following to the scsa2usb.conf file, many USB memory sticks and card readers
might operate more reliably:

742 man pages section 7: Device and Network Interfaces • Last Revised 28 Jul 2008
 scsa2usb(7D)

attribute-override-list = "vid=* reduced-cmd-support=true";

Note that this override applies to all USB mass storage devices and might be inappropriate for
a USB CD writer. If so, you can add an entry for each device to the attribute override list.

If USB mass storage support is considered a security risk, this driver can be disabled in
/etc/system as follows:
exclude: scsa2usb

Alternatively, you can disable automatic handling of a device as described in the following
subsection.

Using Volume Disk storage devices are managed by Volume Manager. Software that manages removable
Management media creates a device nickname that can be listed with eject(1) or rmmount(1). A device that
is not mounted automatically can be mounted using rmmount(1) under /rmdisk/label. Note
that the mount(1M) and mount(1M) commands do not accept nicknames; you must use
explicit device names with these commands.

See rmmount(1) to unmount the device and eject(1) to eject the media. If the device is ejected
while it is mounted, volume management software unmounts the device before ejecting it. It
also might kill any active applications that are accessing the device.

Volume management software is hotplug-aware and normally mounts file systems on USB
mass storage devices if the file system is recognized. Before hot removing the USB device, use
eject(1) to unmount the file system. After the device is removed, a console warning, such as
“The disconnected device was busy, please reconnect,” might display. The warning is harmless
and you can ignore it.

You can disable the automatic mounting and unmounting of removable devices by inserting a
entry for a removable device in /etc/vfstab. In this entry, you must set the mount at boot
field to no. See vfstab(4). See the System Administration Guide, Volume I and Solaris
Common Desktop Environment: User's Guide for details on how to manage a removable device
with CDE and Removable Media Manager. See dtfile.1X under CDE for information on
how to use Removable Media Manager.

Using mount and Use mount(1M) to explicitly mount the device and umount(1M) to unmount the device. Use
umount eject(1) to eject the media. After you have explicitly mounted a removable device, you cannot
use a nickname as an argument to eject.

Removing the disk device while it is being accessed or mounted fails with a console warning.
To hot remove the disk device from the system, unmount the file system, then kill all
applications accessing the device. Next, hot remove the device. A storage device can be hot
inserted at any time.

For a comprehensive listing of (non-bootable) USB mass-storage devices that are compatible
with this driver, see www.sun.com/io.

Device and Network Interfaces 743

scsa2usb(7D)

Device Special Disk block special file names are located in /dev/dsk, while raw file names are located in
Files /dev/rdsk. Tape raw file names are located in /dev/rmt. Input/output requests to the devices
must follow the same restrictions as those for SCSI disks or tapes. Refer to sd(7D) or st(7D).

ioctls Refer to dkio(7I) and cdio(7I).

Errors Refer to sd(7D) for disks or st(7D) for tapes.

Files The device special files for the USB mass storage device are created like those for a SCSI disk or
SCSI tape. Refer to sd(7D) or st(7D).
/dev/dsk/cntndnsn Block files for disks.
/dev/rdsk/cntndnsn Raw files for disks.
/dev/usb/*/*/* ugen(7D) nodes
/dev/rmt/[0- 127][l,m,h,u,c][b][n] Raw files for tapes.
/vol/dev/aliases/zip0 Symbolic link to the character device for the
media in Zip drive 0
/vol/dev/aliases/jaz0 Symbolic link to the character device for the
media in Jaz drive 0.
/vol/dev/aliases/rmdisk0 Symbolic link to the character device for the
media in removable drive 0. This is a generic
removable media device.
/kernel/drv/scsa2usb 32–bit x86 ELF kernel module
/kernel/drv/amd64/scsa2usb 64–bit x86 ELF kernel module
/kernel/drv/sparcv9/scsa2usb 64–bit SPARC ELF kernel module
/kernel/drv/scsa2usb.conf Can be used to override specific characteristics.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also cdrw(1), eject(1), rmformat(1), rmmount(1), cfgadm_scsi(1M), cfgadm_usb(1M),

fdisk(1M), mount(1M), umount(1M), dtfile.1X (in CDE man pages), libusb(3LIB),
scsi(4), vfstab(4), attributes(5), ieee1394(7D)sd(7D), st(7D), ugen(7D), usba(7D),
pcfs(7FS), cdio(7I), dkio(7I)

Writing Device Drivers

744 man pages section 7: Device and Network Interfaces • Last Revised 28 Jul 2008
 scsa2usb(7D)

System Administration Guide, Volume I

Solaris Common Desktop Environment: User's Guide
Universal Serial Bus Specification 2.0
Universal Serial Bus Mass Storage Class Specification Overview 1.0
Universal Serial Bus Mass Storage Class Bulk-Only Transport Specification 1.0
Universal Serial Bus Mass Storage Class Control/Bulk/Interrupt (CBI) Transport Specification
1.0
System Administration Guide: Basic Administration
SCSI Specification T10/995D Revision 11a — March 1997
SCSI SpecificationT10/1236-D Revision 20 — July 2001
SCSI SpecificationT10/1416-D Revision 23— May 2005
http://www.sun.com/io
Diagnostics Refer to sd(7D) and st(7D).
In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (scsa2usb<instance number>): Error Message...
Cannot access <device>. Please reconnect.
There was an error in accessing the mass-storage device during reconnect. Please reconnect
the device.
Device is not identical to the previous one on this port. Please disconnect and reconnect.
Another USB device has been inserted on a port that was connected to a mass-storage
device. Please disconnect the USB device and reconnect the mass-storage device back into
that port.
Reinserted device is accessible again.
The mass-storage device that was hot-removed from its USB slot has been re-inserted to
the same slot and is available for access.
Please disconnect and reconnect this device.
A hotplug of the device is needed before it can be restored.
The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><scsa2usb<instance number>): message...
Invalid <record> in scsa2usb.conf file entry.
An unrecognized record was specified in the scsa2usb.conf file.

Device and Network Interfaces 745

scsa2usb(7D)

Pkt submitted with 0 timeout which may cause indefinite hangs.

An application submitted a request but did not specify a timeout.
Syncing not supported.
Syncing after a panic is not supported. The filesystem may be corrupted.
scsa2usb.conf override: <record>.
An override record specified in scsa2usb.conf was applied. Examples of an override
record applied to a device with vendor ID 123 and product ID 456 are:
vid=0x123 pid=0x456 reduced-cmd-support=true

or

vid=* reduced-cmd-support=true
...meaning that the override record is applied to this device and all other USB mass storage
devices.

Notes The Zip 100 drive does not comply with Universal Serial Bus Specification 1.0 and cannot be
power managed. Power Management support for Zip 100 has been disabled.

If the system panics while a UFS file system is mounted on the mass storage media, no syncing
will take place for the disk mass-storage device. (Syncing is not supported by the scsa2usb
driver.) As a result, the file system on the media will not be consistent on reboot.

If a PCFS file system is mounted, no syncing is needed and the filesystem will be consistent on
reboot.

If a mass-storage device is busy, system suspend cannot proceed and the system will
immediately resume again.

Attempts to remove a mass-storage device from the system will fail. The failure will be logged
to the console. An attempt to replace the removed device with some other USB device will also
fail. To successfully remove a USB mass-storage device you must “close” all references to it.

An Iomega Zip 100Mb disk cannot be formatted on an Iomega Zip250 drive. See the Iomega
web site at http://www.iomega.com for details.

Concurrent I/O to devices with multiple LUNs on the same device is not supported.

Some USB CD-RW devices may perform inadequately at their advertised speeds. To
compensate, use USB CD-RW devices at lower speeds (2X versus 4X). See cdrw(1) for details.

This driver also supports CBI devices that do not use USB interrupt pipe for status
completion.

746 man pages section 7: Device and Network Interfaces • Last Revised 28 Jul 2008
 scsi_vhci(7D)

Name scsi_vhci – SCSI virtual host controller interconnect driver

Description The scsi_vhci driver is a SCSA compliant pseudo nexus driver that supports Solaris
operating system I/O multipathing services for SCSI-3 devices. This driver introduces a
fundamental restructuring of the Solaris device tree to enable a multipath device to be
represented as single device instance rather than as an instance per physical path as in earlier
Solaris versions.

The logical units (LUNs) associated multipath SCSI target devices managed by this driver are
identified and represented by using the SCSI-3 VPD page (0x83) LUN global unique identifier
(GUID) represented as hexadecimal number (64/128 bits)

Symbolic links in /dev/[r]dsk and /dev/scsi/changer continue to adhere to the cNtNdNsN

format. cN is the logical controller number assigned to this driver instance. tN is the GUID.

Symbolic links in /dev/rmt/#[l|m|h|c|u],[b],[n] also adhere to the same format as

non-multipath devices. Because of persistent binding of tape devices, you may want to remove
old non-multipath links when enabling them for multipath.

The following is an example of a system with an A5000 storage array:

...
/dev/rdsk/c4t200000203709C3F5d0s0 -> ../../devices/
scsi_vhci/ssd@g200000203709c3f5:a,raw
...
/dev/rdsk/c4t200000203709C3F5d0s7 -> ../../devices/
scsi_vhci/ssd@g200000203709c3f5:h,ra
...

The following is an example of a system with a T300 storage array:

...
/dev/rdsk/c1t60020F200000033939C2C2B60008D4AEd0s0 ->
../../devices/scsi_vhci/
ssd@g60020f200000033939a2c2b60008d4ae:a,raw
...
/dev/rdsk/c1t60020F200000033939A2C2B60008D4AEd0s7 ->
../../devices/scsi_vhci/
ssd@g60020f200000033939a2c2b60008d4ae:h,raw

The scsi_vhci driver receives naming and transport services from one or more physical HBA
(host bus adapter) devices. To support multi-pathing, a physical HBA driver must have its
multipathing enabled and comply with the multipathing services provided by this driver.

The scsi_vhci driver supports the standard functions provided by the SCSA interface.

Device and Network Interfaces 747

scsi_vhci(7D)

Configuration For each candidate SCSI target device, the scsi_vhci code must identify a failover module to
support the device. If a failover module can't be identified, the device will not function under
scsi_vhci multipathing control. For SCSI target devices that support the standard Target
Port Group Select, no special vendor/product knowledge is needed. For other SCSI target
devices, each failover module understands which devices it supports.

When autoconfiguration does not result in the desired configuration, a vendor/product

specific override mechanism is available. This scsi_vhci.conf base mechanism can be used
to direct a device to a specific failover module (or to indicate that a device should not be under
scsi_vhci multipathing control by way of NONE). In scsi_vhci.conf, the property
’scsi-vhci-failover-override’ defines overrides in scsi_get_device_type_string(9F)
form. To add a third-party (non-Sun) symmetric storage device to run under scsi_vhci (and
thereby take advantage of scsi_vhci multipathing), you add the vendor ID and product ID for
the device, as those strings are returned by the SCSI Inquiry command. For example, to add a
device from a vendor with the ID of "Acme" and a product ID of "MSU", you would add:
device-type-scsi-options-list =
"Acme MSU", "f_sym",

In addition to "Acme", you also might want to add another entry, for example, a device from
"XYZ" vendor with a product ID of "ABC":
scsi-vhci-failover-override =
"Acme MSU", "f_sym",
"XYZ ABC", "f_sym";

As a last override, you might add an entry so that no devices from "ABC" vendor use scsi_vhci
multipathing:
scsi-vhci-failover-override =
"Acme MSU", "f_sym",
"XYZ ABC", "f_sym",
"ABC ", "NONE";
Files /kernel/drv/sparcv9/scsi_vhci 64–bit kernel module (SPARC).
/kernel/drv/scsi_vhci 32-bit kernel module (x86).
/kernel/drv/amd64/scsi_vhci 64-bit kernel module (amd64).
/kernel/drv/scsi_vhci.conf Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWckr

748 man pages section 7: Device and Network Interfaces • Last Revised 23 Jun 2008
 scsi_vhci(7D)

See Also eeprom(1M), prtconf(1M), stmsboot(1M), mpathadm(1M), attributes(5), fcp(7D),

fctl(7D), fp(7d), mpt(7D), ssd(7D), sd(7D), st(7D), sgen(7D), scsi_abort(9F),
scsi_get_device_type_scsi_options(9F), scsi_get_device_type_string(9F),
scsi_ifgetcap(9F), scsi_reset(9F), scsi_transport(9F), scsi_inquiry(9S),
scsi_extended_sense(9S), scsi_pkt(9S)

Writing Device Drivers

Small Computer System Interface-3 (SCSI-3)

Notes In previous releases, the scsi_vhci.conf file supported the mpxio-disable property, which
allowed you to disable Solaris I/O multipathing on a system-wide basis. This property is not
present in the current release of the Solaris operating system. Multipathing is always enabled
in scsi_vhci. If you want to disable multipathing, use the mechanisms provided by the HBA
drivers. See fp(7d) and mpt(7D).

In previous releases, the override mechanism was based on the

scsi_get_device_type_scsi_options(9F) defined "device-type-scsi-options-list"
property. During upgrade, scsi_vhci.conf is converted to the new form. After upgrade, a
scsi_vhci.conf modification based on the old mechanism is silently ignored.

In previous releases, Solaris I/O multipathing was also known as MPxIO and Sun StorEdge
Traffic Manager (STMS).

Device and Network Interfaces 749

sctp(7P)

Name sctp, SCTP – Stream Control Transmission Protocol

Synopsis #include <sys/socket.h>
#include <netinet/in.h>

s = socket(AF_INET, SOCK_STREAM, IPPROTO_SCTP);

s = socket(AF_INET, SOCK_SEQPACKET, IPPROTO_SCTP);
s = socket(AF_INET6, SOCK_STREAM, IPPROTO_SCTP);
s = socket(AF_INET6, SOCK_SEQPACKET, IPPROTO_SCTP);

Description SCTP is a transport protocol layered above the Internet Protocol (IP), or the Internet Protocol
Version 6 (IPv6). SCTP provides a reliable, session oriented, flow-controlled, two-way
transmission of data. It is a message– oriented protocol and supports framing of individual
messages boundaries. An SCTP association is created between two endpoints for data transfer
which is maintained during the lifetime of the transfer. An SCTP association is setup between
two endpoints using a four-way handshake mechanism with the use of a cookie to guard
against some types of denial of service (DoS) attacks. These endpoints may be represented by
multiple IP addresses.

An SCTP message includes a common SCTP header followed by one or more chunks.
Included in the common header is a 32-bit field which contains the checksum (computed
using CRC-32c polynomial) of the entire SCTP packet.

SCTP transfers data payloads in the form of DATA chunks. Each DATA chunk contains a
Transmission Sequence Number (TSN), which governs the transmission of messages and
detection of loss. DATA chunk exchanges follow the Transmission Control Protocol's (TCP)
Selective ACK (SACK) mechanism. The receiver acknowledges data by sending SACK
chunks, which not only indicate the cumulative TSN range received, but also non-cumulative
TSNs received, implying gaps in the received TSN sequence. SACKs are sent using the delayed
acknowledgment method similar to TCP, that is, one SCTP per every other received packet
with an upper bound on the delay (when there are gaps detected the frequency is increased to
one every received packet). Flow and congestion control follow TCP algorithms: Slow Start,
Congestion Avoidance, Fast Recovery and Fast retransmit. But unlike TCP, SCTP does not
support half-close connection and “urgent” data.

SCTP is designed to support a number of functions that are critical for telephony signalling
transport, including multi-streaming. SCTP allows data to be partitioned into multiple
streams that have the property of independent sequenced delivery so that message loss in any
one stream only affects delivery within that stream. In many applications (particularly
telephony signalling), it is only necessary to maintain sequencing of messages that affect some
resource. Other messages may be delivered without having to maintain overall sequence
integrity. A DATA chunk on an SCTP association contains the Stream Id/Stream Sequence
Number pair, in addition to the TSN, which is used for sequenced delivery within a stream.

750 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2009
 sctp(7P)

SCTP uses IP's host level addressing and adds its own per-host collection of port addresses.
The endpoints of an SCTP association are identified by the combination of IP address(es) and
an SCTP port number. By providing the ability for an endpoint to have multiple IP addresses,
SCTP supports multi-homing, which makes an SCTP association more resilient in the
presence of network failures (assuming the network is constructed to provided redundancy).
For a multi-homed SCTP association, a single address is used as the primary address, which is
used as the destination address for normal DATA chunk transfers. Retransmitted DATA
chunks are sent over alternate address(es) to increase the probability of reaching the remote
endpoint. Continued failure to send DATA chunks over the primary address results in
selecting an alternate address as the primary address. Additionally, SCTP monitors the
accessibility of all alternate addresses by sending periodic “heartbeats” chunks. An SCTP
association supports multi-homing by exchanging the available list of addresses during
association setup (as part of its four-way handshake mechanism). An SCTP endpoint is
associated with a local address using the bind(3SOCKET) call. Subsequently, the endpoint can
be associated with additional addresses using sctp_bindx(3SOCKET). By using a special
value of INADDR_ANY with IP or the unspecified address (all zeros) with IPv6 in the bind() or
sctp_bindx() calls, an endpoint can be bound to all available IP or IPv6 addresses on the
system.

SCTP uses a three-way mechanism to allow graceful shutdown, where each endpoint has
confirmation of the DATA chunks received by the remote endpoint prior to completion of the
shutdown. An Abort is provided for error cases when an immediate shutdown is needed.

Applications can access SCTP using the socket interface as a SOCK_STREAM (one-to-one style)
or SOCK_SEQPACKET (one-to-many style) socket type.

One-to-one style socket interface supports similar semantics as sockets for connection
oriented protocols, such as TCP. Thus, a passive socket is created by calling the
listen(3SOCKET) function after binding the socket using bind(). Associations to this
passive socket can be received using accept(3SOCKET) function. Active sockets use the
connect(3SOCKET) function after binding to initiate an association. If an active socket is not
explicitly bound, an implicit binding is performed. If an application wants to exchange data
during the association setup phase, it should not call connect(), but use
sendto(3SOCKET)/sendmsg(3SOCKET) to implicitly initiate an association. Once an
association has been established, read(2) and write(2) can used to exchange data.
Additionally, send(3SOCKET), recv(3SOCKET), sendto(), recvfrom(3SOCKET),
sendmsg(), and recvmsg(3SOCKET) can be used.

One-to-many socket interface supports similar semantics as sockets for connection less
protocols, such as UDP (however, unlike UDP, it does not support broadcast or multicast
communications). A passive socket is created using the listen() function after binding the
socket using bind(). An accept() call is not needed to receive associations to this passive
socket (in fact, an accept() on a one-to-many socket will fail). Associations are accepted
automatically and notifications of new associations are delivered in recvmsg() provided
notifications are enabled. Active sockets after binding (implicitly or explicitly) need not call

Device and Network Interfaces 751

sctp(7P)

connect() to establish an association, implicit associations can be created using

sendmsg()/recvmsg() or sendto()/recvfrom() calls. Such implicit associations cannot be
created using send() and recv() calls. On an SCTP socket (one-to-one or one-to-many), an
association may be established using sendmsg(). However, if an association already exists for
the destination address specified in the msg_name member of the msg parameter, sendmsg()
must include the association id in msg_iov member of the msg parameter (using
sctp_sndrcvinfo structure) for a one-to-many SCTP socket. If the association id is not
provided, sendmsg() fails with EADDRINUSE. On a one-to-one socket the destination
information in the msg parameter is ignored for an established association.

A one-to-one style association can be created from a one-to-many association by branching it

off using the sctp_peeloff(3SOCKET) call; send() and recv() can be used on such peeled
off associations. Calling close(2) on a one-to-many socket will gracefully shutdown all the
associations represented by that one-to-many socket.

The sctp_sendmsg(3SOCKET) and sctp_recvmsg(3SOCKET) functions can be used to

access advanced features provided by SCTP.

SCTP provides the following socket options which are set using setsockopt(3SOCKET) and
read using getsockopt(3SOCKET). The option level is the protocol number for SCTP,
available from getprotobyname(3SOCKET).
SCTP_NODELAY
Turn on/off any Nagle-like algorithm (similar to TCP_NODELAY).
SO_RCVBUF
Set the receive buffer.
SO_SNDBUF
Set the send buffer.
SCTP_AUTOCLOSE
For one-to-many style socket, automatically close any association that has been idle for
more than the specified number of seconds. A value of '0' indicates that no associations
should be closed automatically.
SCTP_EVENTS
Specify various notifications and ancillary data the user wants to receive.
SCTP_STATUS
Retrieve current status information about an SCTP association.

In addition SCTP provides the following option to handle gathering of a limited set of per
endpoint association statistics from a one-to-one socket.
SCTP_GET_ASSOC_STATS Gather and reset per endpoint association statistics.

Example Usage:

752 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2009
 sctp(7P)

#include <netinet/sctp.h>

struct sctp_assoc_stats stat;

int rc;

int32_t len = sizeof (stat);

/*
* Per endpoint stats use the socket descriptor for sctp association.
*/

/* Gather per endpoint association statistics */

rc = getsockopt(sd, IPPROTO_SCTP, SCTP_GET_ASSOC_STATS, &stat, &len);

-----
sctp.h

/*
* SCTP socket option used to read per endpoint association statistics.
*/
#define SCTP_GET_ASSOC_STATS 24

/*
* A socket user request reads local per endpoint association stats.
* All stats are counts except sas_maxrto, which is the max value
* since the last user request for stats on this endpoint.
*/
typedef struct sctp_assoc_stats {
uint64_t sas_rtxchunks; /* Retransmitted Chunks */
uint64_t sas_gapcnt; /* Gap Acknowledgements Received */
uint64_t sas_maxrto; /* Maximum Observed RTO this period */
uint64_t sas_outseqtsns; /* TSN received > next expected */
uint64_t sas_osacks; /* SACKs sent */
uint64_t sas_isacks; /* SACKs received */
uint64_t sas_octrlchunks; /* Control chunks sent - no dups */
uint64_t sas_ictrlchunks; /* Control chunks received - no dups */
uint64_t sas_oodchunks; /* Ordered data chunks sent */
uint64_t sas_iodchunks; /* Ordered data chunks received */
uint64_t sas_ouodchunks; /* Unordered data chunks sent */
uint64_t sas_iuodchunks; /* Unordered data chunks received */
uint64_t sas_idupchunks; /* Dups received (ordered+unordered) */
} sctp_assoc_stats_t;

Multihoming The ability of SCTP to use multiple addresses in an association can create issues with some
network utilities. This requires a system administrator to be careful in setting up the system.

Device and Network Interfaces 753

sctp(7P)

For example, the tcpd allows an administrator to use a simple form of address/hostname
access control. While tcpd can work with SCTP, the access control part can have some
problems. The tcpd access control is only based on one of the addresses at association setup
time. Once as association is allowed, no more checking is performed. This means that during
the life time of the association, SCTP packets from different addresses of the peer host can be
received in the system. This may not be what the system administrator wants as some of the
peer's addresses are supposed to be blocked.

Another example is the use of IP Filter, which provides several functions such as IP packet
filtering (ipf(1M)) and NAT ipnat(1M)). For packet filtering, one issue is that a filter policy
can block packets from some of the addresses of an association while allowing packets from
other addresses to go through. This can degrade SCTP's performance when failure occurs.
There is a more serious issue with IP address rewrite by NAT. At association setup time, SCTP
endpoints exchange IP addresses. But IP Filter is not aware of this. So when NAT is done on a
packet, it may change the address to an unacceptable one. Thus the SCTP association setup
may succeed but packets cannot go through afterwards when a different IP address is used for
the association.

See Also ipf(1M), ipnat(1M), ndd(1M), ioctl(2), close(2), read(2), write(2), accept(3SOCKET),
bind(3SOCKET), connect(3SOCKET), getprotobyname(3SOCKET),
getsockopt(3SOCKET), libsctp(3LIB), listen(3SOCKET), recv(3SOCKET),
recvfrom(3SOCKET), recvmsg(3SOCKET), sctp_bindx(3SOCKET),
sctp_getladdrs(3SOCKET), sctp_getpaddrs(3SOCKET), sctp_freepaddrs(3SOCKET),
sctp_opt_info(3SOCKET), sctp_peeloff(3SOCKET), sctp_recvmsg(3SOCKET),
sctp_sendmsg(3SOCKET), send(3SOCKET), sendmsg(3SOCKET), sendto(3SOCKET),
socket(3SOCKET), ipfilter(5), tcp(7P), udp(7P), inet(7P), inet6(7P), ip(7P), ip6(7P)

R. Stewart, Q. Xie, K. Morneault, C. Sharp, H. Schwarzbauer, T. Taylor, I. Rytina, M. Kalla, L.

Zang, V. Paxson, RFC 2960, Stream Control Transmission Protocol, October 2000

L. Ong, J. Yoakum, RFC 3286, An Introduction to Stream Control Transmission Protocol

(SCTP), May 2002

J. Stone, R. Stewart, D. Otis, RFC 3309, Stream Control Transmission Protocol (SCTP)
Checksum Change, September 2002.

Diagnostics A socket operation may fail if:

EPROTONOSUPPORT The socket type is other than SOCK_STREAM and SOCK_SEQPACKET.
ETIMEDOUT An association was dropped due to excessive retransmissions.
ECONNREFUSED The remote peer refused establishing an association.
EADDRINUSE A bind() operation was attempted on a socket with a network
address/port pair that has already been bound to another socket.

754 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2009
 sctp(7P)

EINVAL A bind() operation was attempted on a socket with an invalid network

address.
EPERM A bind() operation was attempted on a socket with a “reserved“ port
number and the effective user ID of the process was not the privileged
user.

Device and Network Interfaces 755

sd(7D)

Name sd – SCSI disk and ATAPI/SCSI CD-ROM device driver

Synopsis sd@target,lun:partition

Description To open a device without checking if the vtoc is valid, use the O_NDELAY flag. When the
device is opened using O_NDELAY, the first read or write to the device that happens after the
open results in the label being read if the label is not currently valid. Once read, the label
remains valid until the last close of the device. Except for reading the label, O_NDELAY has
no impact on the driver.

SPARC The sd SCSI and SCSI/ATAPI driver supports embedded SCSI-2 and CCS-compatible SCSI
disk and CD-ROM drives, ATAPI 2.6 (SFF-8020i)-compliant CD-ROM drives,
SFF-8090–compliant SCSI/ATAPI DVD-ROM drives, IOMEGA SCSI/ATAPI ZIP drives,
SCSI JAZ drives, and USB mass storage devices (refer to scsa2usb(7D)).

To determine the disk drive type, use the SCSI/ATAPI inquiry command and read the volume
label stored on block 0 of the drive. (The volume label describes the disk geometry and
partitioning and must be present for the disk to be mounted by the system.) A volume label is
not required for removable, re-writable or read-only media.

x86 Only The sddriver supports embedded SCSI-2 and CCS-compatible SCSI disk and CD-ROM
drives, ATAPI 2.6 (SFF-8020i)-compliant CD-ROM drives, SFF-8090-compliant SCSI/ATAPI
DVD-ROM drives, IOMEGA SCSI/ATAPI ZIP drives, and SCSI JAZ drives.

The x86 BIOS legacy requires a master boot record (MBR) and fdisk table in the first physical
sector of the bootable media. If the x86 hard disk contains a Solaris disk label, it is located in
the second 512-byte sector of the FDISK partition.

Device Special Block-files access the disk using normal buffering mechanism and are read-from and
Files written-to without regard to physical disk records. A raw interface enables direct transmission
between the disk and the user's read or write buffer. A single read or write call usually results
in a single I/O operation, therefore raw I/O is more efficient when many bytes are transmitted.
Block files names are found in /dev/dsk; raw file names are found in /dev/rdsk.

I/O requests to the raw device must be aligned on a 512-byte (DEV_BSIZE) boundary and all
I/O request lengths must be in multiples of 512 bytes. Requests that do not meet these
requirements will trigger an EINVAL error. There are no alignment or length restrictions on
I/O requests to the block device.

Cd-ROM Drive A CD-ROM disk is single-sided and contains approximately 640 megabytes of data or 74
Support minutes of audio. When the CD-ROM is opened, the eject button is disabled to prevent
manual removal of the disk until the last close() is called. No volume label is required for a
CD-ROM. The disk geometry and partitioning information are constant and never change. If
the CD-ROM contains data recorded in a Solaris-aware file system format, it can be mounted
using the appropriate Solaris file system support.

756 man pages section 7: Device and Network Interfaces • Last Revised 8 Sep 2009
 sd(7D)

Dvd-ROM Drive DVD-ROM media can be single or double-sided and can be recorded upon using a single or
Support double layer structure. Double-layer media provides parallel or opposite track paths. A
DVD-ROM can hold from between 4.5 Gbytes and 17 Gbytes of data, depending on the layer
structure used for recording and if the DVD-ROM is single or double-sided.

When the DVD-ROM is opened, the eject button is disabled to prevent the manual removal of
a disk until the last close() is called. No volume label is required for a DVD-ROM. If the
DVD-ROM contains data recorded in a Solaris-aware file system format, it can be mounted
using the appropriate Solaris file system support.

Zip/JAZ Drive ZIP/JAZ media provide varied data capacity points; a single JAZ drive can store up to 2 GBytes
Support of data, while a ZIP-250 can store up to 250MBytes of data. ZIP/JAZ drives can be read-from
or written-to using the appropriate drive.

When a ZIP/JAZ drive is opened, the eject button is disabled to prevent the manual removal of
a disk until the last close() is called. No volume label is required for a ZIP/JAZ drive. If the
ZIP/JAZ drive contains data recorded in a Solaris-aware file system format, it can be mounted
using the appropriate Solaris file system support.

Device Statistics Each device maintains I/O statistics for the device and for partitions allocated for that device.
Support For each device/partition, the driver accumulates reads, writes, bytes read, and bytes written.
The driver also initiates hi-resolution time stamps at queue entry and exit points to enable
monitoring of residence time and cumulative residence-length product for each queue.

Not all device drivers make per-partition IO statistics available for reporting. sd and ssd(7D)
per-partition statistics are enabled by default but may disabled in their configuration files.

ioctls Refer to dkio(7I), and cdio(7I)

ERRORS EACCES Permission denied
EBUSY The partition was opened exclusively by another thread
EFAULT The argument features a bad address
EINVAL Invalid argument
ENOTTY The device does not support the requested ioctl function
ENXIO During opening, the device did not exist. During close, the drive unlock failed
EROFS The device is read-only
EAGAIN Resource temporarily unavailable
EINTR A signal was caught during the execution of the ioctl() function
ENOMEM Insufficient memory
EPERM Insufficent access permission

Device and Network Interfaces 757

sd(7D)

EIO An I/O error occurred. Refer to notes for details on copy-protected DVD-ROM
media.

Configuration The sd driver can be configured by defining properties in the sd.conf file. The sd driver
supports the following properties:
enable-partition-kstats The default value is 1, which causes partition IO statistics to
be maintained. Set this value to zero to prevent the driver
from recording partition statistics. This slightly reduces the
CPU overhead for IO, mimimizes the amount of sar(1) data
collected and makes these statistics unavailable for reporting
by iostat(1M) even though the -p/-P option is specified.
Regardless of this setting, disk IO statistics are always
maintained.
qfull-retries The supplied value is passed as the qfull-retries capability
value of the HBA driver. See scsi_ifsetcap(9F) for details.
qfull-retry-interval The supplied value is passed as the qfull-retry interval
capability value of the HBA driver. See scsi_ifsetcap(9F)
for details.
allow-bus-device-reset The default value is 1, which allows resetting to occur. Set this
value to 0 (zero) to prevent the sd driver from calling
scsi_reset(9F) with a second argument of RESET_TARGET
when in error-recovery mode. This scsi_reset(9F) call may
prompt the HBA driver to send a SCSI Bus Device Reset
message. The scsi_reset(9F) call with a second argument of
RESET_TARGET may result from an explicit request via the
USCSICMD ioctl. Some high-availability multi-initiator
systems may wish to prohibit the Bus Device Reset message;
to do this, set the allow-bus-device-reset property to 0.
optical-device-bind Controls the binding of the driver to non self-identifying
SCSI target optical devices. (See scsi(4)). The default value is
1, which causes sd to bind to DTYPE_OPTICAL devices (as
noted in scsi(4)). Setting this value to 0 prevents automatic
binding. The default behavior for the SPARC-based sd driver
prior to Solaris 9 was not to bind to optical devices.
power-condition Boolean type, when set to False, it indicates that the disk
does not support power condition field in the START STOP
UNIT command.

In addition to the above properties, some device-specific tunables can be configured in

sd.conf using the sd-config-list global property. The value of this property is a list of
duplets. The formal syntax is:

758 man pages section 7: Device and Network Interfaces • Last Revised 8 Sep 2009
 sd(7D)

sd-config-list = <duplet> [, <duplet> ]* ;

where

<duplet>:= "<vid+pid>" , "<tunable-list>"

and

<tunable-list>:= <tunable> [, <tunable> ]*;

<tunable> = <name> : <value>

The <vid+pid> is the string that is returned by the target device

on a SCSI inquiry command.

The <tunable-list> contains one or more tunables to apply to

all target devices with the specified <vid+pid>.

Each <tunable> is a <name> : <value> pair. Supported

tunable names are:

delay-busy: when busy, nsecs of delay before retry.

retries-timeout: retries to perform on an IO timeout.

mmc-gesn-polling For optical drives compliant with MMC-3 and supporting the GET
EVENT STATUS NOTIFICATION command, this command is used for
periodic media state polling, usually initiated by the DKIOCSTATE
dkio(7I) ioctl. To disable the use of this command, set this boolean
property to false. In that case, either the TEST UNIT READY or
zero-length WRITE(10) command is used instead.
Examples The following is an example of a global sd-config-list property:

sd-config-list =
"SUN T4", "delay-busy:600, retries-timeout:6",
"SUN StorEdge_3510", "retries-timeout:3";
Files /kernel/drv/sd.conf Driver configuration file
/dev/dsk/cntndnsn Block files
/dev/rdsk/cntndnsn Raw files

Where:
cn controller n
tn SCSI target id n (0-6)

Device and Network Interfaces 759

sd(7D)

dn SCSI LUN n (0-7 normally; some HBAs support LUNs to 15 or 32. See the specific
manpage for details)
sn partition n (0-7)
x86 Only /dev/rdsk/cntndnpn raw files

Where:
pn Where n=0 the node corresponds to the entire disk.

See Also sar(1), cfgadm_scsi(1M), fdisk(1M), format(1M), iostat(1M), close(2), ioctl(2),

lseek(2), read(2), write(2), driver.conf(4), scsi(4), filesystem(5), scsa2usb(7D),
ssd(7D), hsfs(7FS), pcfs(7FS), udfs(7FS), cdio(7I), dkio(7I), scsi_ifsetcap(9F),
scsi_reset(9F)

ANSI Small Computer System Interface-2 (SCSI-2)

ATA Packet Interface for CD-ROMs, SFF-8020i

Mt.Fuji Commands for CD and DVD, SFF8090v3

http://www.sun.com/io
Diagnostics Error for Command:<command name>
Error Level: Fatal
Requested Block: <n>
Error Block: <m>
Vendor:’<vendorname>’
Serial Number:’<serial number>’
Sense Key:<sense key name>

ASC: 0x<a> (<ASC name>), ASCQ: 0x<b>, FRU: 0x<c>

The command indicated by <command name> failed. The Requested Block is the block
where the transfer started and the Error Block is the block that caused the error. Sense Key,
ASC, and ASCQ information is returned by the target in response to a request sense
command.
Caddy not inserted in drive
The drive is not ready because no caddy has been inserted.
Check Condition on REQUEST SENSE
A REQUEST SENSE command completed with a check condition. The original command
will be retried a number of times.
Label says <m> blocks Drive says <n> blocks
There is a discrepancy between the label and what the drive returned on the READ CAPACITY
command.

760 man pages section 7: Device and Network Interfaces • Last Revised 8 Sep 2009
 sd(7D)

Not enough sense information

The request sense data was less than expected.
Request Sense couldn't get sense data
The REQUEST SENSE command did not transfer any data.
Reservation Conflict
The drive was reserved by another initiator.
SCSI transport failed: reason ’xxxx’: {retrying|giving up}
The host adapter has failed to transport a command to the target for the reason stated. The
driver will either retry the command or, ultimately, give up.
Unhandled Sense Key<n>
The REQUEST SENSE data included an invalid sense.
Unit not ready. Additional sense code 0x
<n> The drive is not ready.
Can't do switch back to mode 1
A failure to switch back to read mode 1.
Corrupt label - bad geometry
The disk label is corrupted.
Corrupt label - label checksum failed
The disk label is corrupted.
Corrupt label - wrong magic number
The disk label is corrupted.
Device busy too long
The drive returned busy during a number of retries.
Disk not responding to selection
The drive is powered down or died
Failed to handle UA
A retry on a Unit Attention condition failed.
I/O to invalid geometry
The geometry of the drive could not be established.
Incomplete read/write - retrying/giving up
There was a residue after the command completed normally.
No bp for direct access device format geometry
A bp with consistent memory could not be allocated.
No bp for disk label
A bp with consistent memory could not be allocated.

Device and Network Interfaces 761

sd(7D)

No bp for fdisk
A bp with consistent memory could not be allocated.
No bp for rigid disk geometry
A bp with consistent memory could not be allocated.
No mem for property
Free memory pool exhausted.
No memory for direct access device format geometry
Free memory pool exhausted.
No memory for disk label
Free memory pool exhausted.
No memory for rigid disk geometry
The disk label is corrupted.
No resources for dumping
A packet could not be allocated during dumping.
Offline
Drive went offline; probably powered down.
Requeue of command fails
Driver attempted to retry a command and experienced a transport error.
sdrestart transport failed()
Driver attempted to retry a command and experienced a transport error.
Transfer length not modulo
Illegal request size.
Transport of request sense fails()
Driver attempted to submit a request sense command and failed.
Transport rejected()
Host adapter driver was unable to accept a command.
Unable to read label
Failure to read disk label.
Unit does not respond to selection
Drive went offline; probably powered down.

Notes DVD-ROM media containing DVD-Video data may follow/adhere to the requirements of
content scrambling system or copy protection scheme. Reading of copy-protected sector will
cause I/O error. Users are advised to use the appropriate playback software to view video
contents on DVD-ROM media containing DVD-Video data.

762 man pages section 7: Device and Network Interfaces • Last Revised 8 Sep 2009
 sda(7D)

Name sda – SD/MMC architecture

Description The sda module provides support services for Secure Digital (SD) and MultiMediaCard
(MMC) slot and card device drivers.
Files /kernel/misc/sda 32-bit ELF kernel module (x86)
/kernel/misc/amd64/sda 64-bit ELF kernel module (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWsdcard

See Also cfgadm_sdcard(1M), attributes(5), blk2scsa(7D), sd(7D), sdcard(7D), sdhost(7D),

scsa2usb(7D)

System Administration Guide, Volume I

System Administration Guide: Basic Administration

Notes The sda module provides support only for SD/MMC devices that are connected via a
supported slot driver. Notably, slots that are on USB busses are normally treated as USB mass
storage devices and are serviced by the scsa2usb(7D) driver.

Device and Network Interfaces 763

SDC(7)

Name SDC – System Duty Cycle scheduling class

Description The System Duty Cycle (SDC) scheduling class is used for some CPU-intensive kernel thread
workloads. Like the SYS class, it cannot be used for user processes.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Private

See Also attributes(5)

764 man pages section 7: Device and Network Interfaces • Last Revised 25 Nov 2009
 sdcard(7D)

Name sdcard – SD/MMC memory card driver

Description The sdcard memory card driver supports MultiMediaCard (MMC), Secure Digital (SD), and
Secure Digital High Capacity (SDHC) memory cards. It uses the blk2scsa(7D) service to
present these cards to the system as SCSI disks on a virtual SCSI bus, creating a child device to
be serviced with sd(7D). Each card appears as its own SCSI LUN. Cards are hot-pluggable and
removable.

Device Special Disk block special file names are located in /dev/dsk. Raw file names are located in
Files /dev/rdsk. See sd(7D).

ioctls See dkio(7I)

Errors See sd(7D) and blk2scsa(7D). Additionally, sdcard may issue the following warnings, which
indicate a failure to identify the card as a supported type:
"Unknown SD CSD version (%d)"
"Unknown MMC CSD version (%d)"
"Unknown MMCA version (%d)"
"Card type unknown"

Files Device special files for the storage device are created in the same way as those for a SCSI disk.
See sd(7D) for more information.
/dev/dsk/cntndnsn Block files for disks
/dev/rdsk/cntndnsn Raw files for disks
/kernel/drv/sdcard 32-bit ELF kernel module (x86)
/kernel/misc/amd64/sdcard 64-bit ELF kernel module (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWsdcard

See Also rmformat(1), rmmount(1), cfgadm_scsi(1M), cfgadm_sdcard(1M), fdisk(1M), mount(1M),

umount(1M), scsi(4), vfstab(4), attributes(5), blk2scsa(7D), sd(7D), sda(7D), dkio(7I),
pcfs(7FS)

System Administration Guide, Volume I

System Administration Guide: Basic Administration

Device and Network Interfaces 765

sdcard(7D)

Diagnostics See sd(7D).

766 man pages section 7: Device and Network Interfaces • Last Revised 28 Oct 2009
 sdhost(7D)

Name sdhost – Standard-compliant Secure Digital slot driver

Synopsis pciclass,080500@unit-address
pciclass,080501@unit-address

Description The sdhost driver supports Secure Digital (SD) standard media slots commonly found on
mobile computers.

Files Memory card device files are created by the sdcard(7D) driver. An attachment point device
file is created for each physical slot on the system:
/dev/sdcardx/y Attachment point for slot y on controller x. Typically this is
named /dev/sdcard0/0.
/kernel/drv/sdhost 32-bit ELF kernel module (x86).
/kernel/drv/amd64/sdhost 64-bit ELF kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWsdcard

See Also cfgadm_sdcard(1M), attributes(5), sda(7D), sdcard(7D)

System Administration Guide, Volume I

System Administration Guide: Basic Administration

Device and Network Interfaces 767

sdp(7D)

Name sdp – Sockets Direct Protocol driver

Synopsis #include <socket.h>
#include <netinet/in.h>
s = socket(AF_INET, SOCK_STREAM, PROTO_SDP);
s = socket(AF_INET6, SOCK_STREAM, PROTO_SDP);

Description The Sockets Direct Protocol (SDP) is a transport protocol layered over the Infiniband
Transport Framework (IBTF). SDP is a standard implementation based on Annex 4 of the
Infiniband Architecture Specification Vol 1 and provides reliable byte-stream, flow controlled
two-way data transmission that closely mimics the Transmission Control Protocol (TCP).

SDP supports a sockets-based SOCK_STREAM interface to application programs. It also

supports graceful close (including half-closed sockets), IP addressing (IPv4 or IPv6), the
connecting/accepting connect model, out-of-band (OOB) data and common socket options.
The SDP protocol also supports kernel bypass data transfers and data transfers from
send-upper-layer-protocol (ULP) buffers to receive ULP buffers. A SDP message includes a
BSDH header followed by data. (A BSDH header advertises the amount of available buffers on
the local side).

SDP networking functionality is broken into the sdp driver and a function call-based sockfs
implementation. A new protocol family of PROTO_SDP is introduced to use the SDP
transport provided by the driver.

Sockets utilizing SDP are either active or passive. Active sockets initiate connections to passive
sockets. Both active and passive sockets must have their local IP or IPv6 address and SDP port
number bound with the bind(3SOCKET) system call after the socket is created. By default,
SDP sockets are active. A passive socket is created by calling the listen(3SOCKET) system
call after binding the socket with bind(). This process establishes a queueing parameter for the
passive socket. Connections to the passive socket can be received with the accept(3SOCKET)
system call. Active sockets use the connect(3SOCKET) call after binding to initiate
connections.

In most cases, SDP sends data when it is presented. When outstanding data is not yet
acknowledged, SDP gathers small amounts of output to be sent in a single packet once an
acknowledgement is received. For a small number of clients this packetization may cause
significant delays. To circumvent this problem, SDP provided by the driver supplies
SDP_NODELAY, a socket-level boolean option. Note that this behavior is similar to the
TCP_NODELAY option.

SDP provides an urgent data mechanism that can be invoked using the out-of-band
provisions of send(3SOCKET). The out-of-band delivery behavior is identical to TCP. The
caller may mark one byte as "urgent" with the MSG_OOB flag to send(3SOCKET). This sets
an "urgent pointer" pointing to the byte in the SDP stream. The receiver of the stream is
notified of the urgent data by a SIGURG signal. The SIOCATMARK ioctl(2) request returns a

768 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2006
 sdp(7D)

value indicating whether the stream is at the urgent mark. Because the system never returns
data across the urgent mark in a single read(2) call, it is possible to advance to the urgent data
in a simple loop which reads data, testing the socket with the SIOCATMARK ioctl() request until
it reaches the mark.

Address Formats SDP uses IP/IPv6 addresses to refer to local and remote devices and opens a reliable connected
IB connection between two end points. The sdp driver supports a point-to-point connection,
however broadcasting and multicasting are not supported.

Socket Options SDP supports setsockopt and getsockopt to set and read socket options. Very few socket
options affect SDP protocol operations. Other common socket options are processed but do
not affect SDP protocol operation. All socket options are checked for validity. A getsockopt
returns the values set or toggled by setsockopt. Socket options that affect protocol operations
are SO_LINGER, SO_DEBUG, SO_REUSEADDR and SO_OOBINLINE.
Errors EISCONN A connect() operation was attempted on a socket on which a
connect() operation had already been performed.
ECONNRESET The remote peer forced the connection to be closed. This
usually occurs when the remote machine loses state
information about the connection due to a crash.
ECONNREFUSED The remote peer actively refused connection establishment.
This usually occurs because no process is listening to the
port.
EADDRINUSE A bind() operation was attempted on a socket with a
network address/port pair that has already been bound to
another socket.
EADDRNOTAVAIL A bind() operation was attempted on a socket with a
network address for which no network interface exists.
EACCES A bind() operation was attempted with a reserved port
number and the effective user ID of the process was not the
privileged user.
ENOBUFS The system ran out of memory for internal data structures.
Files /kernel/drv/sdp
32–bit ELF kernel module (x86).
/kernel/drv/amd64/sdp
64–bit ELF kernel module (x86).
/kernel/drv/sparcv9/sdp
64–bit ELF kernel module (SPARC).
/kernel/drv/sdpib
32–bit ELF kernel module (x86).

Device and Network Interfaces 769

sdp(7D)

/kernel/drv/amd64/sdpib
64–bit ELF kernel module (x86).
/kernel/drv/sparcv9/sdpib
64–bit ELF kernel module (SPARC).

Attributes See attributes(5) for descriptions of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86, SPARC

Availability SUNWibsdp, SUNWibsdp

See Also read(2), getsockopt(3XNET), socket.h(3HEAD), accept(3SOCKET), bind(3SOCKET),

connect(3SOCKET), send(3SOCKET), attributes(5), standards(5)

Infiniband Architecture Specification Vol 1– Annex 4 — November, 2002

770 man pages section 7: Device and Network Interfaces • Last Revised 13 Nov 2006
 sdt(7D)

Name sdt – DTrace statically defined tracing provider

Description The sdt driver is a DTrace dynamic tracing provider that performs dynamic instrumentation
at statically-defined locations in the Solaris kernel.

The sdt provider allows kernel developers to explicitly create probes at formally designated
locations in the operating system kernel and loadable modules, allowing the implementor to
consciously choose the points in their code that are desired probe points, and to convey some
semantic knowledge about that point with the choice of probe name and a relevant set of
arguments.

The sdt driver is not a public interface and you access instrumentation offered by this
provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description of the
public documented interfaces available for the DTrace facility and the probes offered by the
sdt provider.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), attributes(5), dtrace(7D)

Solaris Dynamic Tracing Guide

Device and Network Interfaces 771

se(7D)

Name se – Siemens 82532 ESCC serial communications driver

Synopsis se@bus_address:port_name[,cu]

Description The se module is a loadable STREAMS driver that provides basic support for the 82532 ESCC
hardware and basic asynchronous and synchronous communication support. This manual
page describes the asynchronous protocol interface; for information on the synchronous
interface, please see the se_hdlc(7D) manual page.

This module is affected by the setting of specific eeprom variables. For information on
parameters that are persistent across reboots, see the eeprom(1M) man page.

The platform specific device bus address for the se module is bus_address. The se module's
port_name is a single letter (a-z).

Note – During boot up, ttya/b characteristics are read from the /kernel/drv/options.conf
file and changed from the PROM defaults to reflect Solaris defaults. Messages displayed on the
console after this point are based on settings in that file. If you switch a characteristic, (for
example, the baud rate of the console terminal), you must revise the
/kernel/drv/options.conf or the console will be configured to an unusable configuration
and console messages will be garbled by the mismatched serial port settings.

Application The Siemens 82532 provides two serial input/output channels capable of supporting a variety
Programming of communication protocols. A typical system will use one of these devices to implement two
Interface serial ports (port_name), usually configured for RS-423 (which also supports most RS-232
equipment). The Siemens 82532 uses 64 character input and output FIFOs to reduce system
overhead. When receiving characters, the CPU is notified when 32 characters have arrived
(one-half of receive buffer is full) or no character has arrived in the time it would take to
receive four characters at the current baud rate.

When sending characters, the Siemens 82532 places the first 64 characters to be sent into its
output FIFO and then notifies the CPU when it is half empty (32 characters left). Because the
se module waits for the Siemens 82532 to transmit the remaining characters within its output
FIFO before making requested changes, delays may occur when the port's attributes are being
modified.

The se module implements CTS/RTS flow control in hardware. To prevent data overruns,
remove CTS/RTS flow control responsibility from the CPU during periods of high system
load.

In async mode (obtained by opening /dev/cua/[a-z], /dev/term/[a-z] or /dev/tty[a-z]),

the driver supports the termio(7I) device control functions specified by flags in the c_cflag
word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK, or INPCK flags in the
c_iflag word. All other termio(7I) functions must be performed by STREAMS modules
pushed atop the driver. When a device is opened, the ldterm(7M) and ttcompat(7M)
STREAMS modules are automatically pushed on top of the stream, providing the standard
termio interface.

772 man pages section 7: Device and Network Interfaces • Last Revised 16 Jul 2009
 se(7D)

You can connect a single tty line to a modem for incoming and outgoing calls using a special
feature controlled by the minor device number. By accessing character-special devices with
names of the form /dev/cua/[a-z], it is possible to open a port without the Carrier Detect
signal being asserted, either through hardware or an equivalent software mechanism. These
devices are commonly known as dial-out lines.

After a /dev/cua/[a-z] line is opened, the corresponding tty line cannot be opened until the
/dev/cua/[a-z] line is closed. A blocking open will wait until the /dev/cua/[a-z] line is closed
(which will drop Data Terminal Ready and Carrier Detect) and carrier is detected again. A
non-blocking open will return an error. If the tty line has been opened successfully (usually
only when carrier is recognized on the modem), the corresponding /dev/cua/[a-z] line
cannot be opened. This allows a modem to be attached to a device, (for example, /dev/term/
[a-z] renamed from /dev/tty[a-z]) and used for dial-in (by enabling the line for login in
/etc/inittab) and dial-out (by tip(1) or uucp(1C)) as /dev/cua/[a-z] when no one is logged
in on the line.

The se driver can be configured to support mice as well as applications requiring no buffering
on the receive fifo. You can do this in one of two ways:
1. Using the se configuration file — To configure the se device to support mice using this
approach, create an se.conf under /kernel/drv, then add keywords of the form:
disable-rfifo-port<a/b><instance number>. For example, if your system has two se
devices and you want port b on the device (associated with instance 0) and port a
(associated with instance 1) to have their receive fifo disabled, the se.conf file must
contain the following entries:
disable-rfifo-portb0;
disable-rfifo-porta1;
2. Programmatically — You can also configure the se device to support mice programatically
by using the SERVICEIMM stream call to turn buffering off on the receive fifo, and/or
SERVICEDEF to turn it back on again.

ioctls The se module supports the standard set of termio ioctl( ) calls.

Breaks can be generated by the TCSBRK, TIOCSBRK, and TIOCCBRK ioctl() calls.

The state of the DCD, CTS, RTS, and DTR interface signals can be queried through the use of the
TIOCM_CAR, TIOCM_CTS, TIOCM_RTS, and TIOCM_DTR arguments to the TIOCMGET ioctl
command, respectively. Due to hardware limitations, only the RTS and DTR signals may be set
through their respective arguments to the TIOCMSET, TIOCMBIS, and TIOCMBIC ioctl
commands.

The input and output line speeds may be set to all baud rates supported by termio. Input and
output line speeds cannot be set independently; when you set the output speed, the input
speed is automatically set to the same speed.

Device and Network Interfaces 773

se(7D)

When using baud rates over 100,000 baud, the software changes the line driver configuration
to handle the higher data rates. This action decreases the theoretical maximum cable length
from 70 meters to 30 meters.

When the se module is used to service the serial console port, it supports a BREAK condition
that allows the system to enter the debugger or the monitor. The BREAK condition is
generated by hardware and it is usually enabled by default. A BREAK condition originating
from erroneous electrical signals cannot be distinguished from one deliberately sent by
remote DCE. Due to a risk of incorrect sequence interpretation, SLIP and certain other binary
protocols should not be run over the serial console port when Alternate Break sequence is in
effect. Although PPP is a binary protocol, it is able to avoid these sequences using the ACCM
feature in RFC 1662. For Solaris PPP 4.0, you do this by adding the following line to the
/etc/ppp/options file (or other configuration files used for the connection; see pppd(1M) for
details):
asyncmap 0x00002000

By default, the Alternate Break sequence is a three character sequence: carriage return, tilde
and control-B (CR ~ CTRL-B), but may be changed by the driver. For information on
breaking (entering the debugger or monitor), see kmdb(1) and kb(7M).

Errors An open() will fail under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The dial-out device is being opened and the dial-in device is already open, or the
dial-in device is being opened with a no-delay open and the dial-out device is
already open.
EBUSY The port is in use by another serial protocol.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl( ) call.
EINTR The open was interrupted by the delivery of a signal.
Files /dev/cua/[a-z] dial-out tty lines
/dev/term/[a-z] dial-in tty lines
/dev/tty[a-z] binary compatibility package device names
/dev/se_hdlc[0-9] synchronous devices - see se_hdlc(7D).
/dev/se_hdlc synchronous control clone device
/kernel/drv/options.conf System wide default device driver properties

774 man pages section 7: Device and Network Interfaces • Last Revised 16 Jul 2009
 se(7D)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also kmdb(1), tip(1), cu(1C), uucp(1C), eeprom(1M), ports(1M), pppd(1M), ioctl(2), open(2),
attributes(5),zs(7D), zsh(7D), se_hdlc(7D), termio(7I), ldterm(7M), ttcompat(7M),
kb(7M)
Diagnostics sen : fifo overrun The Siemens 82532 internal FIFO received more data than it
could handle. This indicates that Solaris was not servicing data
interrupts fast enough and suggests a system with too many
interrupts or a data line with a data rate that is too high.
sen : buffer overrun The se module was unable to store data it removed from the
Siemens 82532 FIFO. The user process is not reading data fast
enough, and suggests an overloaded system. If possible, the
application should enable flow control (either CTSRTS or
XONXOFF) to allow the driver to backpressure the remote
system when the local buffers fill up.

Device and Network Interfaces 775

se_hdlc(7D)

Name se_hdlc – on-board high-performance serial HDLC interface

Synopsis se@bus_address:port_number[, hdlc]

Description The se_hdlc devices are a synchronous hdlc-framing interface for the se serial devices. Both
built-in serial ports (port_number) on platforms which have the se serial devices, support
synchronous data transfer at a maximum rate of 384 kbps. bus_address is the platform specific
se device bus address. port_number is a single digit number (0-9).

Application The se_hdlcn devices provide a data path which supports the transfer of data via read(2) and
Programming write(2) system calls, as well as ioctl(2) calls. Data path opens are exclusive in order to
Interface protect against injection or diversion of data by another process.

The se_hdlc device provides a separate control path for use by programs that need to
configure or monitor a connection independent of any exclusive access restrictions imposed
by data path opens. Up to three control paths may be active on a particular serial channel at
any one time. Control path accesses are restricted to ioctl(2) calls only; no data transfer is
possible.

When used in synchronous modes, the SAB 82532 ESCC supports several options for clock
sourcing and data encolding. Both the transmit and receive clock sources can be set to be the
external Transmit clock (TRxC), external Receive Clock (RTxC), the internal Baud Rate
Generator (BRG), or the output of the ESCC 's Digital Phase-Lock Loop (DPLL).

The BRG is a programmable divisor that derives a clock frequency from the PCLK input signal
to the ESCC. The programmed baud rate is translated into a floating point (6-bit mantissa,
4–bit exponent) number time constant that is stored in the ESCC.

A local loopback mode is available, primarily for use by syncloop(1M) for testing purposes,
and should not be confused with SDLC loop mode, which is not supported on this interface.
Also, an auto-echo feature may be selected that causes all incoming data to be routed to the
transmit data line, allowing the port to act as the remote end of a digital loop. Neither of these
options should be selected casually, or left in use when not needed.

The se driver keeps running totals of various hardware generated events for each channel.
These include numbers of packets and characters sent and received, abort conditions detected
by the receiver, receive CRC errors, transmit underruns, receive overruns, input errors and
output errors, and message block allocation failures. Input errors are logged whenever an
incoming message must be discarded, such as when an abort or CRC error is detected, a
receive overrun occurs, or when no message block is available to store incoming data. Output
errors are logged when the data must be discarded due to underruns, CTS drops during
transmission, CTS timeouts, or excessive watchdog timeouts caused by a cable break.

ioctls The se driver supports the following ioctl() commands.

776 man pages section 7: Device and Network Interfaces • Last Revised 1 Jan 1997
 se_hdlc(7D)

S_IOCGETMODE Return a struct scc_mode containing parameters currently in use. These

include the transmit and receive clock sources, boolean loopback and
NRZI mode flags and the integer baud rate.
S_IOCSETMODE The argument is a struct scc_mode from which the ESCC channel will
be programmed.
S_IOCGETSTATS Return a struct sl_stats containing the current totals of
hardware-generated events. These include numbers of packets and
characters sent and received by the driver, aborts and CRC errors
detected, transmit underruns, and receive overruns.
S_IOCCLRSTATS Clear the hardware statistics for this channel.
S_IOCGETSPEED Returns the currently set baud rate as an integer. This may not reflect the
actual data transfer rate if external clocks are used.
S_IOCGETMCTL Returns the current state of the CTS and DCD incoming modem
interface signals as an integer.

The following structures are used with se hdlc ioctl() commands:

struct scc_mode {
char sm_txclock; /* transmit clock sources */
char sm_rxclock; /* receive clock sources */
char sm_iflags; /* data and clock inversion flags (non-zsh) */
uchar_t sm_config; /* boolean configuration options */
int sm_baudrate; /* real baud rate */
int sm_retval; /* reason codes for ioctl failures */
};
struct sl_stats {
long ipack; /* input packets */
long opack; /* output packets */
long ichar; /* input bytes */
long ochar; /* output bytes */
long abort; /* abort received */
long crc; /* CRC error */
long cts; /* CTS timeouts */
long dcd; /* Carrier drops */
long overrun; /* receive overrun */
long underrun; /* transmit underrun */
long ierror; /* input error */
long oerror; /* output error */
long nobuffers; /* receive side memory allocation failure */
};

Errors An open() will fail if a STREAMS message block cannot be allocated or under the following
conditions:

Device and Network Interfaces 777

se_hdlc(7D)

ENXIO The unit being opened does not exist.

EBUSY The device is in use by another serial protocol.

An ioctl() will fail under the following conditions:

EINVAL An attempt was made to select an invalid clocking source.
EINVAL The baud rate specified for use with the baud rate generator would translate to a
null time constant in the ESCC's registers.
Files /dev/se_hdlc[0-1], /dev/se_hdlc character-special devices
/usr/include/sys/ser_sync.h header file specifying synchronous serial
communication definitions

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also syncinit(1M), syncloop(1M), syncstat(1M), ioctl(2), open(2), read(2), write(2),

attributes(5), se(7D), zsh(7D)

Siemens ESCC2 SAB 82532 Enhanced Serial Communication Controller User's Manual
Diagnostics se_hdlc clone open failed, no memory, rq=nnn
A kernel memory allocation failed for one of the private data structures. The value of nnn is
the address of the read queue passed to open(2).
se_hdlc: clone device must be attached before use!
An operation was attempted through a control path before that path had been attached to a
particular serial channel.
se_hdlcn: not initialized, can't send message
An M_DATA message was passed to the driver for a channel that had not been programmed
at least once since the driver was loaded. The ESCC's registers were in an unknown state.
The S_IOCSETMODE ioctl command performs the programming operation.
sen hdlc_start: Invalid message type d on write queue
driver received an invalid message type from streams.
se_hdlcn: transmit hung
The transmitter was not successfully restarted after the watchdog timer expired. This is
usually caused by a bad or disconnected cable.

778 man pages section 7: Device and Network Interfaces • Last Revised 1 Jan 1997
 ses(7D)

Name ses – SCSI enclosure services device driver

Synopsis ses@target,lun

Description The ses device driver is an interface to SCSI enclosure services devices. These devices sense
and monitor the physical conditions in an enclosure as well as allow access to the status
reporting and configuration features of the enclosure (such as indicator LEDs on the
enclosure.)

ioctl(9E) calls may be issued to ses to determine the state of the enclosure and to set
parameters on the enclosure services device.

No ses driver properties are defined. Use the ses.conf file to configure the ses driver.

Examples EXAMPLE 1 ses.conf File Format

The following is an example of the ses.conf file format:

#
# Copyright (c) 1996, by Sun Microsystems, Inc.
# All rights reserved.
#
#
#ident "@(#)ses.conf 1.1 97/02/10 SMI"
#

name="ses" parent="sf"
target=15;

name="ses" parent="SUNW,pln" port=0 target=15;

name="ses" parent="SUNW,pln" port=1 target=15;
name="ses" parent="SUNW,pln" port=2 target=15;
name="ses" parent="SUNW,pln" port=3 target=15;
name="ses" parent="SUNW,pln" port=4 target=15;
name="ses" parent="SUNW,pln" port=5 target=15;

name="ses" class="scsi"
target=15 lun=0;

ioctls The SES driver currently supports the SES, SAFTE and SEN enclosure service chipsets. SEN
and SAFTE protocols are translated internally in the driver into SES compliant data
structures. This enables the SES driver to work seamlessly with different protocols and
eliminates the need to enhance user applications.
SESIOC_GETNOBJ Returns an unsigned integer that represents the number of SES data
structures in the enclosure services chip.
SESIOC_GETOBJMAP Returns a size array containing ses_object elements communicated
through SESIOC_GETNOBJ(). ses_object is defined in sesio.h.

Device and Network Interfaces 779

ses(7D)

SESIOC_INIT Instructs the device to perform a self-diagnostic test. Currently SES

& SEN devices always return success.
SESIOC_GETENCSTAT Returns an unsigned character that represents status enclosure as
defined by Table 25 in Section 7.1.2 of the SES specification NCITS
305-199x.
SESIOC_GETOBJSTAT This ioctl is passed an ses_objarg containing the obj_id you want
to set, then fills in the remaining fields according to element status
page of the SES specification.
SESIOC_SETOBJSTAT Sets options in the control field. You set control field options by
filling out all fields in ses_objarg. Field definitions are presented in
Section 7.2.2 of the SES specification.
Files /kernel/drv/ses.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also driver.conf(4), scsi(4), attributes(5), esp(7D), isp(7D), ioctl(9E)

780 man pages section 7: Device and Network Interfaces • Last Revised 27 May 2008
 sesio(7I)

Name sesio – enclosure services device driver interface

Synopsis #include <sys/sesio.h>

Description The ses device driver provides the following ioctls as a means to access SCSI enclosure
services devices.

ioctls The ses driver supports the following ioctls:

SES_IOCTL_GETSTATE This ioctl obtains enclosure state in the ses_ioctl structure.
SES_IOCTL_SETSTATE This ioctl is used to set parameters on the enclosure services device.
The ses_ioctl structure is used to pass information into the
driver.
Errors EIO The ses driver was unable to obtain data from the enclosure services device or the
data transfer could not be completed.
ENOTTY The ses driver does not support the requested ioctl function.
ENXIO The enclosure services device does not exist.
EFAULT The user specified a bad data length.

Structures The ses_ioctl structure has the following fields:

uint32_t; /* Size of buffer that follows */
uint8_t page_code: /* Page to be read/written */
uint8_t reserved[3]; /* Reserved; Set to 0 */
unit8t buffer[1]; /* Size arbitrary, user specifies */

Examples EXAMPLE 1 Using the SES_IOCTL_GETSTATE ioctl

The following example uses the SES_IOCTL_GETSTATE ioctl to recover 20 bytes of page 4 from
a previously opened device.
char abuf[30];
struct ses_ioctl *sesp;
int status;
sesp = (ses_ioctl *)abuf;
sesp->size = 20;
sesp->page_code = 4;
status = ioctl(fd, SES_IOCTL_GETSTATE, abuf);

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Device and Network Interfaces 781

sesio(7I)

See Also ses(7D), ioctl(9E)

782 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 1997
 sf(7D)

Name sf – SOC+ FC-AL FCP Driver

Synopsis sf@port,0

Description The sf driver is a SCSA compliant nexus driver which supports the Fibre Channel Protocol
for SCSI on Private Fibre Channel Arbitrated loops. An SBus card called the SOC+ card (see
socal(7D)) connects the Fibre Channel loop to the host system.

The sf driver interfaces with the SOC+ device driver, socal(7D), the SCSI disk target driver,
ssd(7D), and the SCSI-3 Enclosure Services driver, ses(7D). It only supports SCSI devices of
type disk and ses.

The sf driver supports the standard functions provided by the SCSA interface. The driver
supports auto request sense and tagged queueing by default.

The driver requires that all devices have unique hard addresses defined by switch settings in
hardware. Devices with conflicting hard addresses will not be accessible.
Files /platform/architecture/kernel/drv/sf ELF kernel module
/platform/architecture/kernel/drv/sf.conf sf driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also luxadm(1M), prtconf(1M), driver.conf(4), socal(7D), ssd(7D)

Writing Device Drivers

ANSI X3.272-1996, Fibre Channel Arbitrated Loop (FC-AL)

ANSI X3.269-1996, Fibre Channel Protocol for SCSI (FCP)

ANSI X3.270-1996, SCSI-3 Architecture Model (SAM)

Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)

Diagnostics In addition to being logged, the messages below may display on the system console.

The first set of messages indicate that the attachment was unsuccessful, and will only display
while the sf driver is initially attempting to attach. Each message is preceded by sf%d , where
%d is the instance number of the sf device.
Failed to alloc soft state Driver was unable to allocate space for the
internal state structure. Driver did not attach
to device, SCSI devices will be inaccessible.

Device and Network Interfaces 783

sf(7D)

Bad soft state Driver requested an invalid internal state

structure. Driver did not attach to device, SCSI
devices will be inaccessible.
Failed to obtain transport handle Driver was unable to obtain a transport handle
to communicate with the socal driver. Driver
did not attach to device, SCSI devices will be
inaccessible
Failed to allocate command/response pool Driver was unable to allocate space for
commands and responses. Driver did not
attach to device, SCSI devices will be
inaccessible.
Failed to allocate kmem cache Driver was unable to allocate space for the
packet cache. Driver did not attach to device,
SCSI devices will be inaccessible.
Failed to allocate dma handle for Driver was unable to allocate a dma handle for
the loop map. Driver did not attach to device,
SCSI devices will be inaccessible.
Failed to allocate lilp map Driver was unable to allocate space for the loop
map. Driver did not attach to device, SCSI
devices will be inaccessible.
Failed to bind dma handle for Driver was unable to bind a dma handle for the
loop map. Driver did not attach to device, SCSI
devices will be inaccessible.
Failed to attach Driver was unable to attach for some reason
that may be printed. Driver did not attach to
device, SCSI devices will be inaccessible.

The next set of messages may display at any time. The full device pathname, followed by the
shorter form described above, will precede the message.
Invalid lilp map The driver did not obtain a valid lilp map from the socal
driver. SCSI device will be inaccessible.
Target t, AL-PA x and hard The device with a switch setting t has an AL-PA x which
does not match its hard address y. The device will not be
accessible.
Duplicate switch settings The driver detected devices with the same switch setting.
All such devices will be inaccessible.
WWN changed on target t The World Wide Name (WWN) has changed on the device
with switch setting t.

784 man pages section 7: Device and Network Interfaces • Last Revised 27 Mar 1997
 sf(7D)

Target t, unknown device type The driver does not know the device type reported by the
device with switch setting t.

Device and Network Interfaces 785

sfe(7D)

Name sfe – SiS900 series Fast Ethernet device driver

Synopsis /dev/sfe

Description The sfe driver is a loadable, clonable, GLD-based STREAMS driver supporting the Data Link
Provider Interface dlpi(7P) on Silicon Integrated Systems 900 series Fast Ethernet controllers.

Application The relevant fields returned as part of a DL_INFO_ACK response are:

Programming
Interface ■ Maximum SDU is 1500.
■ Minimum SDU is 0.
■ The dlsap address length is 8.
■ MAC type is DL_ETHER.
■ SAP length is -2. The 6–byte physical address is followed immediately by a 2–byte SAP.
■ Service mode is DL_CLDLS.
■ Broadcast address is the 6-byte Ethernet broadcast address (ff:ff:ff:ff:ff:ff).

If the SAP provided is zero, then IEEE 802.3 mode is assumed and outbound frames will have
the frame payload length written into the type field. Likewise, inbound frames with a SAP
between zero and 1500 are interpreted as IEEE 802.3 frames and delivered to any streams that
are bound to SAP zero (the 802.3 SAP).

Properties The following properties may be configured using either ndd(1M) or the sfe.conf
configuration file as described by driver.conf(4):
adv_100fdx_cap
Enables the 100 Base TX full-duplex link option. (This is generally the fastest mode if both
link partners support it. Most modern equipment supports this mode.)
adv_100hdx_cap
Enables the 100 Base TX half-duplex link option. (Typically used when the link partner is a
100 Mbps hub.)
adv_10fdx_cap
Enables the 10 Base-T full-duplex link option. (This less-frequently used mode is typically
used when the link partner is a 10 Mbps switch.)
adv_10hdx_cap
Enables the 10 Base-T half-duplex link option. (This is the fall-back when no other option
is available. It is typically used when the link partner is a 10 Mbps hub or is an older
network card.)

Attributes See attributes(5) for a description of the following attributes:

786 man pages section 7: Device and Network Interfaces • Last Revised 12 Dec 2007
 sfe(7D)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SPARC, x86

Interface Stability Committed

Files /dev/sfe
Special character device.
/kernel/drv/sfe
32-bit driver binary (x86).
/kernel/drv/amd64/sfe
64-bit driver binary (x86).
/kernel/drv/sparcv9/sfe
64-bit driver binary (SPARC).
/kernel/drv/sfe.conf
Configuration file.

See Also ndd(1M), driver.conf(4), attributes(5), streamio(7I), dlpi(7P)

IEEE 802.3 — Institute of Electrical and Electronics Engineers, 2002

Device and Network Interfaces 787

sgen(7D)

Name sgen – Generic SCSI device driver

Synopsis #include <sys/scsi/targets/sgendef.h>
sgen@target,lun:<devtype>

Description The sgen driver exports the uscsi(7I) interfaces to user processes. The sgen driver can be
configured to bind to SCSI devices for which no system driver is available. Examples of such
devices include SCSI scanners and SCSI processor devices.

Security Typically, drivers which export the uscsi(7I) interface unconditionally require that the user
present superuser credentials. The sgen driver does not, and relies on the filesystem
permissions on its device special file to govern who may access that device. By default, access is
restricted and device nodes created by the sgen driver are readable and writable by the
superuser exclusively.

It is important to understand that SCSI devices coexisting on the same SCSI bus may
potentially interact with each other. This may result from firmware bugs in SCSI devices, or
may be made to happen programmatically by sending appropriate SCSI commands to a
device. Potentially, any application controlling a device via the sgen driver can introduce data
integrity or security problems in that device or any other device sharing the same SCSI bus.

Granting unprivileged users access to an sgen-controlled SCSI device may create other
problems. It may be possible for a user to instruct a target device to gather data from another
target device on the same bus. It may also be possible for malicious users to install new
firmware onto a device to which they are granted access. In environments where security is a
concern but user access to devices controlled by the sgen driver is nontheless desired, it is
recommended that the devices be separated onto a dedicated SCSI bus to mitigate the risk of
data corruption and security violations.

Configuration The sgen driver is configurable via the sgen.conf file. In addition to standard SCSI device
configuration directives (see scsi(4)), administrators can set several additional properties for
the sgen driver.

By default, the sgen driver will not claim or bind to any devices on the system. To do so, it
must be configured by the administrator using the inquiry-config-list and/or the
device-type-config-list properties.

As with other SCSI drivers, the sgen.conf configuration file enumerates the targets sgen
should use. See scsi(4) for more details. For each target enumerated in the sgen.conf file, the
sgen driver sends a SCSI INQUIRY command to gather information about the device present at
that target. The inquiry-config-list property specifies that the sgen driver should bind to a
particular device returning a particular set of inquiry data. The device-type-config-list
specifies that the sgen driver should bind to every device that is of a particular SCSI device
type. When examining the device, the sgen driver tests to see if it matches an entry in the
device-type-config-list or the inquiry-config-list. For more detail on these two
properties, see the PROPERTIES section.

788 man pages section 7: Device and Network Interfaces • Last Revised 29 Mar 2008
 sgen(7D)

When a match against the INQUIRY data presented by a device is made, the sgen driver
attaches to that device and creates a device node and link in the /devices and /dev
hierarchies. See the FILES section for more information about how these files are named.

It is important for the administrator to ensure that devices claimed by the sgen driver do not
conflict with existing target drivers on the system. For example, if the sgen driver is configured
to bind to a direct access device, the standard sd.conf file will usually cause sd to claim the
device as well. This can cause unpredictable results. In general, the uscsi(7I) interface
exported by sd(7D) or st(7D) should be used to gain access to direct access and sequential
devices.

The sgen driver is disabled by default. The sgen.conf file is shipped with all of the
'name="sgen" class="scsi" target=...' entries commented out to shorten boot time and to
prevent the driver from consuming kernel resources. To use the sgen driver effectively on
desktop systems, simply uncomment all of the name="sgen" lines in sgen.conf file. On larger
systems with many SCSI controllers, carefully edit the sgen.conf file so that sgen binds only
where needed. Refer to driver.conf(4) for further details.
Properties inquiry-config-list The inquiry-config-list property is a list of pairs of strings that
enumerates a list of specific devices to which the sgen driver will
bind. Each pair of strings is referred to as <vendorid, productid>
in the discussion below.
vendorid is used to match the Vendor ID reported by the device. The SCSI specification
limits Vendor IDs to eight characters. Correspondingly, the length of this string
should not exceed eight characters. As a special case, "*" may be used as a
wildcard which matches any Vendor ID. This is useful in situations where more
than one vendor produces a particular model of a product. vendorid is matched
against the Vendor ID reported by the device in a case-insensitive manner.
productid is used to match the product ID reported by the device. The SCSI specification
limits product IDs to sixteen characters (unused characters are filled with the
whitespace characters). Correspondingly, the length of productid should not
exceed sixteen characters. When examining the product ID of the device, sgen
examines the length l of productid and performs a match against only the first
l characters in the device's product ID. productid is matched against the
product ID reported by the device in a case-insensitive manner.

For example, to match some fictitious devices from ACME corp, the inquiry-config-list
can be configured as follows:

inquiry-config-list = "ACME", "UltraToast 3000",

"ACME", "UltraToast 4000",

"ACME", "UltraToast 5000";

Device and Network Interfaces 789

sgen(7D)

To match "UltraToast 4000" devices, regardless of vendor, inquiry-config-list is modified

as follows:

inquiry-config-list = "*", "UltraToast 4000";

To match every device from ACME in the "UltraToast" series (i.e UltraToast 3000, 4000, 5000,
...), inquiry-config-list is modified as follows:

inquiry-config-list = "ACME" "UltraToast";

Whitespace characters are significant when specifying productid. For example, a productid
of "UltraToast 1000" is fifteen characters in length. If a device reported its ID as "UltraToast
10000", the sgen driver would bind to it because only the first fifteen characters are considered
significant when matching. To remedy this situation, specify productid as "UltraToast 1000 ",
(note trailing space). This forces the sgen driver to consider all sixteen characters in the
product ID to be significant.
device-type-config-list The device-type-config-list property is a list of strings
that enumerate a list of device types to which the sgen driver
will bind. The valid device types correspond to those defined
by the SCSI-3 SPC Draft Standard, Rev. 11a. These types are:

Type Name Inquiry Type ID

direct 0x00

sequential 0x01

printer 0x02

processor 0x03

worm 0x04

rodirect 0x05

scanner 0x06

optical 0x07

changer 0x08

comm 0x09

prepress1 0x0a

prepress2 0x0b

790 man pages section 7: Device and Network Interfaces • Last Revised 29 Mar 2008
 sgen(7D)

Type Name Inquiry Type ID

array_ctrl 0x0c

ses 0x0d

rbc 0x0e

ocrw 0x0f

bridge 0x10

type_unknown 0x1f

Alternately, you can specify device types by INQUIRY type ID. To do this, specify
type_0x<typenum> in the sgen-config-list. Case is not significant when specifying device
type names.
sgen-diag The sgen-diag property sets the diagnostic output level. This property can be
set globally and/or per target/lun pair. sgen-diag is an integer property, and
can be set to 0, 1, 2 or 3. Illegal values will silently default to 0. The meaning of
each diagnostic level is as follows:
0 No error reporting [default]
1 Report driver configuration information, unusual conditions, and indicate when sense
data has been returned from the device.
2 Trace the entry into and exit from routines inside the driver, and provide extended
diagnostic data. No error reporting [default].
3 Provide detailed output about command characteristics, driver state, and the contents of
each CDB passed to the driver.

In ascending order, each level includes the diagnostics that the previous level reports. See the
IOCTLS section for more infomation on the SGEN_IOC_DIAG ioctl.
Files sgen.conf Driver configuration file. See CONFIGURATION for
more details.
/dev/scsi/<devtype>/cntndn The sgen driver categorizes each device in a separate
directory by its SCSI device type. The files inside the
directory are named according to their controller
number, target ID and LUN as follows:

cn is the controller number, tn is the SCSI target id and

dn is the SCSI LUN

This is analogous to the {controller;target;device}

naming scheme, and the controller numbers correspond
to the same controller numbers which are used for

Device and Network Interfaces 791

sgen(7D)

naming disks. For example, /dev/dsk/c0t0d0s0 and

/dev/scsi/scanner/c0t5d0 are both connected to
controller c0.

ioctls The sgen driver exports the uscsi(7I) interface for each device it manages. This allows a user
process to talk directly to a SCSI device for which there is no other driver installed in the
system. Additionally, the sgen driver supports the following ioctls:
SGEN_IOC_READY Send a TEST UNIT READY command to the device and return 0 upon
success, non-zero upon failure. This ioctl accepts no arguments.
SGEN_IOC_DIAG Change the level of diagnostic reporting provided by the driver. This
ioctl accepts a single integer argument between 0 and 3. The levels have
the same meaning as in the sgen-diag property discussed in
PROPERTIES above.
Errors EBUSY The device was opened by another thread or process using the O_EXCL flag, or
the device is currently open and O_EXCL is being requested.
ENXIO During opening, the device did not respond to a TEST UNIT READY SCSI
command.
ENOTTY Indicates that the device does not support the requested ioctl function.

Examples Here is an example of how sgen can be configured to bind to scanner devices on the system:

device-type-config-list = "scanner";

The administrator should subsequently uncomment the appropriate name="sgen"... lines for
the SCSI target ID to which the scanner corresponds. In this example, the scanner is at target
4.

name= "sgen" class= "scsi" target=4 lun=0;

If it is expected that the scanner will be moved from target to target over time, or that more
scanners might be added in the future, it is recommended that all of the name="sgen"... lines
be uncommented, so that sgen checks all of the targets on the bus.

For large systems where boot times are a concern, it is recommended that the parent=""
property be used to specify which SCSI bus sgen should examine.

See Also driver.conf(4), scsi(4), sd(7D), st(7D), uscsi(7I)

Writing Device Drivers

ANSI Small Computer System Interface-2 (SCSI-2)

SCSI-3 SPC Draft Standard, Rev. 11a

792 man pages section 7: Device and Network Interfaces • Last Revised 29 Mar 2008
 sharefs(7FS)

Name sharefs – Kernel sharetab filesystem

Description The sharefs filesystem describes the state of all shares currently loaded by the kernel. It is
mounted during boot time as a read-only file at /etc/dfs/sharetab.

Filesystem contents are dynamic and reflect the current set of shares in the system. File
contents are described in sharetab(4).

File contents can be modified as a result of share(1M), sharectl(1M), sharemgr(1M) and

changing properties of a zfs(1M) data set.

The module may not be unloaded dynamically by the kernel.

Files /etc/dfs/sharetab System record of shared file systems.

See Also share(1M), sharectl(1M), sharemgr(1M), zfs(1M), sharetab(4)

Device and Network Interfaces 793

si3124(7D)

Name si3124 – SiliconImage 3124/3132 SATA controller driver

Synopsis sata@unit-address

Description The si3124 driver is a SATA framework-compliant HBA driver that supports Silicon Image
3124 and 3132 SATA controllers. Note that while the Silicon Image controllers supports
standard SATA features including SATA-II disks, NCQ, hotplug, port multiplier and ATAPI
disks, the si3124 driver currently does not support NCQ, port multiplier or ATAPI features.

Configuration There are no tunable parameters in the si3124.conf file.

Files /kernel/drv/si3124 32–bit ELF kernel module (x86).
/kernel/drv/amd64/si3124 64–bit ELF kernel module. (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWsi3124

See Also cfgadm(1M), prtconf(1M), cfgadm_sata(1M), attributes(5), nv_sata(7D), sata(7D)

Writing Device Drivers

794 man pages section 7: Device and Network Interfaces • Last Revised 17 August 2007
 sip(7P)

Name sip – SIP Proxy/registrar/redirect server

Description Solaris supports deployment of VoIP/SIP services by providing an RFC 3261–compliant SIP
proxy/registrar/redirect server called SER from iptel.org.

See the ser(8) man page under /usr/sfw/man.

Files /etc/sfw/ser/ser.cfg

/etc/sfw/ser/README.solaris.ser

/usr/sfw/share/doc/ser/README

Device and Network Interfaces 795

sk98sol(7D)

Name sk98sol – SysKonnect Gigabit Ethernet SK-98xx device driver

Synopsis /dev/skge
/kernel/drv/sk98sol

Description The sk98sol driver is a multi-threaded, loadable, clonable, STREAMS hardware driver
supporting the connectionless Data Link Provider Interface (DLPI), over a SysKonnect
Gigabit Ethernet adapter (SK-98xx series). The driver supports multiple installed SysKonnect
SK-98xx adapters. Functions include chip initialization, frame transmit and receive, multicast
and promiscuous support, and error recovery and reporting.

Application The driver provides the /dev/skge cloning character-special device as well as per-adapter
Programming character-special devices /dev/skgex, where x represents the device instance number.
Interface
sk98sol and DLPI The sk98sol driver is a Style 1 and Style 2 Data Link Service (DLS) provider. All M_PROTO and
M_PCPROTO type messages are interpreted as DLPI primitives. Valid DLPI primitives are
defined in <sys/dlpi.h>. See dlpi(7P).

An explicit DL_ATTACH_REQ message by the user is required to associate the opened stream
with a particular device (ppa). This is unnecessary and invalid for DLPI Style 1. The ppa ID is
interpreted as an unsigned long data type and indicates the corresponding device instance
(unit) number. An error (DL_ERROR_ACK) is returned by the driver if the ppa field value does
not correspond to a valid device instance number for the system.

The device is initialized on first attach and de-initialized (stopped) upon last detach. Valid
device numbers for all detected adapters are displayed on the console at driver startup time
and are written to the /var/adm/messages log file.

The values returned in the DL_INFO_ACK primitive in response to the DL_INFO_REQ request are:
■ Maximum SDU is 1500 (9000 if JumboFrames are enabled).
■ Minimum SDU is 0.
■ DLSAP address length is 8 bytes.
■ MAC type is DL_CSMACD.
■ SAP length value is –2, meaning the physical address component is followed immediately
by a 2-byte SAP component within the DLSAP address.
■ Service mode is DL_CLDLS.
■ Optional quality of service (QOS) support is not included; as a result, the QOS field values
are 0.
■ Provider style is DL_STYLE2.
■ Version is DL_VERSION_2.

796 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2003
 sk98sol(7D)

Options Parameters are set in the /kernel/drv/sk98sol.conf configuration file, which is created
during installation. See driver.conf(4). You can edit the /kernel/drv/sk98sol.conf file to
reflect your settings and reboot the system to use the new parameter values. If the file exists
prior to driver installation, the new parameter values will be used as soon as the driver is
installed.

String parameter values must be surrounded with double quotes ("), while integer parameter
values are not. Parameter names and values are case sensitive and you should use them exactly
as shown.

Note – You can increase sk98sol performance by tuning certain TCP and UDP parameters.
However, you should be aware that this act may adversely impact the performance of other
network cards.

To tune specific parameters to increase sk98sol performance, do the following:

ndd -set /dev/tcp tcp_xmit_hiwat 65536
ndd -set /dev/tcp tcp_xmit_lowat 32768
ndd -set /dev/tcp tcp_recv_hiwat 65536

ndd -set /dev/udp udp_xmit_hiwat 65536

ndd -set /dev/udp udp_xmit_lowat 327
ndd -set /dev/udp udp_recv_hiwat 65536

Per-Port Parameters The parameters discussed in this section can be set for each port on the adapter.

In each of the following descriptions, ? represents port A or B.

AutoNegotiation_?

Type: String

Values: On, Off, Sense

Default: Sense (for SK-984x fiber adapters).

Default: On (for SK-982x copper adapters.)

The Sense value automatically detects whether the link partner supports autonegotiation. If
your link partner is configured to half duplex with autonegotiation turned off, set the
AutoNegotiation_? and DuplexCapabilities_? parameters manually. Do not set the
AutoNegotiation_? parameter value to Sense, as it will fail.

Do not use Sense for 1000Base-T (copper) adapters. If Sense is selected, it will be mapped to
On automatically.

DuplexCapabilities_?

Device and Network Interfaces 797

sk98sol(7D)

Type: String

Values: Half, Full, Both

Default: Both

Set the DuplexCapabilities_? parameter only if the AutoNegotiation_? parameter is set to the
On or Off values. If AutoNegotiation_? is set to On, all three AutoNegotiation_? values are
possible; however, if set to Off, only the Full and Half values are allowed.

Use the DuplexCapabilities_? parameter if your link partner does not support all possible
combinations.

FlowControl_?

Type: String

Values: Sym, SymOrRem, LocSend, None

Default: SymOrRem

Use the FlowControl_? parameter to set the flow control capabilities reported by the port
during autonegotiation:
Sym Symetric flow control, where both link partners are allowed to send PAUSE
frames.
SymOrRem SymetricOrRemote flow control, where both link partners or only the remote
partner are allowed to send PAUSE frames.
LocSend LocalSend flow control, where only the local link partner is allowed to send
PAUSE frames.
None No flow control, where no link partner is allowed to send PAUSE frames.

The FlowControl_? parameter is ignored if AutoNegotiation_? is set to "Off."

Role_?

Type: String

Values: Auto, Master, Slave

Default: Auto

Use the Role_? parameter only for the SK-9821 and SK-9822 adapters.

798 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2003
 sk98sol(7D)

1000Base-T communication between two ports requires one port to act as the master (and
provide timing information) and the other as slave. Normally, this is negotiated between the
two ports during link establishment. If this fails, use the Role_? parameter to force the master
and slave roles on the ports. If AutoNegotiation_? is set to "Off," then the Role_? parameter
must be set manually.

Per-Adapter PreferredPort
Parameters

Type: String

Values: A, B

Default: A

Use the PreferredPort parameter to force the preferred port to A or B (on two-port NICs). The
preferred port is the port selected if both ports are detected as fully functional.

RlmtMode

Type: integer

Values: 1, 2, 3

Default: 1

RLMT (Redundant Link Management Technology) provides three modes to determine if a

port is available for use.
1. Check link state only: use the link state reported by the adapter hardware for each
individual port.
2. Check other port: RLMT sends test frames from one port to another and checks if they are
received. The ports must be connected to the network that allow LLC test frames to be
exchanged (that is, networks without routers between the ports).
3. Check other port and segmentation: RLMT checks the other port and also requests
information from the Gigabit Ethernet switch next to each port to determine if the
network is segmented between the ports. Only use this mode if you have Gigabit Ethernet
switches installed and configured to use the Spanning Tree protocol.

Note that modes 2 and 3 are meant to operate in configurations where a network path exists
between the ports on a single adapter. They are not designed to work in networks where
adapters are connected back-to-back.

JumboFrames

Device and Network Interfaces 799

sk98sol(7D)

Type: String

Values: Off, On

Default: Off

To enable support for JumboFrames (frames with a length of up to 9014 bytes), set
JumboFrames to "On." Because longer frames reduce operating system overhead, JumboFrames
increases network throughput.

For full JumboFrames support, the maximum transfer unit (MTU) size used by TCP/IP must
also be changed by using the ifconfig(1M) command. To do this, remove the comment sign
(#) before the ifconfig line in the /etc/rcS.d/S50sk98sol file. You should also ensure that
the adapter device number (skge0) matches the attach number displayed during system
startup. The MTU must be set to 9000, not including the 14 bytes of MAC address header.

JumboFrames can only be used if all equipment in your subnetwork supports them; currently
many switches do not support JumboFrames). Devices without Jumbo Support drop the
longer frames (and might report them as error frames). If you experience problems with this,
connect two SK-98xx adapters (with JumboFrames enabled) back-to-back.)

CopyThreshold

Type: Integer

Values: 0–1500

Default: 1500

During transmit, the driver relies on the frame's physical memory address to tell the hardware
where to find the frame data. Setting up the DMA address can take time on Solaris; it may be
more convenient to copy the frame data to a buffer that you have set up in advance. All frames
with a length less than or equal to the CopyThreshold parameter value are copied into buffers;
for longer frames, the real DMA setup is done. By default (without JumboFrames support), all
frames are copied. You can experiment with this parameter to find out if your system
performs better with only smaller frames copied.

To use more complex syntax for setting different parameters on multiple adapters, see
driver.conf(4). For example:

name="sk98sol" parent="/pci@1f,4000" unit-address="2" AutoNegotiation_A="Off";

name="sk98sol" parent="/pci@1f,2000" unit-address="2" AutoNegotiation_B="Sense";

800 man pages section 7: Device and Network Interfaces • Last Revised 19 May 2003
 sk98sol(7D)

Diagnostics If multiple NICs are installed in the system, the following message may appear on the console
and in the /var/adm/messages log file:

Allocation of descriptor memory failed

You can avoid this message by tuning the lomempages kernel parameter. By default, the value
of this parameter is 36 pages. Each SK-98xx adapter requires a determined number of pages,
so increase the value of the lomempages parameter in increments of ten pages until all NICs in
the system run correctly.

To modify the value of this parameter to 46 pages, append the set lomempages=46 line to the
/etc/system file and reboot the system.
Files /dev/skge Character special device
/dev/skgex Per-adapter character special device, where x is the
adapter ppa
/kernel/drv/sk98sol ELF kernel module
/kernel/drv/sparcv9/sk98sol 64–bit ELF kernel module
/kernel/drv/sk98sol.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture IA, SPARC

See Also ifconfig(1M), netstat(1M), driver.conf(4), attributes(5), dlpi(7P).

sk98sol.txt driver README file — Included in the driver package; also available from
www.syskonnect.com.

Device and Network Interfaces 801

skfp(7D)

Name skfp – SysKonnect FDDI PCI device driver

Synopsis /dev/skfp
Description The skfp FDDI driver is a multi-threaded, loadable, clonable, STREAMS hardware driver
supporting the connectionless Data Link Provider Interface (DLPI) over a SysKonnect FDDI
PCI adapter. The driver supports multiple installed SysKonnect FDDI PCI adapters.
Functions include chip initialization, frame transit and receive, multicast and promiscuous
support, and error recovery and reporting.
The skfp driver supports all SysKonnect SK-NET FDDI PCI adapters (SK-55xx (32-bit) and
SK-58xx (64-bit) series) on 32-bit systems, and the SK-58xx series on 64-bit systems.
Application The skfp driver provides the /dev/skfp cloning character-special device that accesses all
Programming SK-NET FDDI PCI adapters using Data Link Service (DLS) Style 2. It also provides
Interface per-adapter character-special devices /dev/skfpx, (where x represents the device instance
number) that access a special NIC using DLS Style 1.
skfp and DLPI The skfp driver is a Style 1 and Style 2 DLS provider. All M_PROTO and M_PCPROTO type
messages are interpreted as DLPI primitives. Valid DLPI primitives are defined in
<sys/dlpi.h>. See dlpi(7P).
An explicit DL_ATTACH_REQ message by the user is required to associate the opened Stream
with a particular device (ppa). This is unnecessary and invalid for DLPI Style 1. The ppa ID is
interpreted as an unsigned long data type and indicates the corresponding device instance
(unit) number. An error (DL_ERROR_ACK) is returned by the driver if the ppa field value does
not correspond to a valid device instance number for the system.
The device is initialized on first attach and de-initialized (stopped) upon last detach. Valid
device numbers for all detected adapters are displayed on the console at driver startup time
and written to the /var/adm/messages log file.
The values returned in the DL_INFO_ACK primitive in response to the DL_INFO_REQ request are:
■ Maximum SDU is 4470.
■ Minimum SDU is 0.
■ DSLAP address length is 8 bytes.
■ MAC type is DL_FDDI.
■ SAP length value is –2, meaning the physical address component is followed immediately
by a 2-byte SAP component within the DLSAP address.
■ Service mode is DL_CLDLS.
■ Optional quality of service (QOS) support is not included; as a result, the QOS field values
are 0.
■ Provider style is DL_STYLE2.
■ Version is DL_VERSION_2.

802 man pages section 7: Device and Network Interfaces • Last Revised 17 May 2003
 skfp(7D)

Options Options are not required for normal operation. In special cases, FDDI Station Management
(SMT) parameters can be modified by using the /usr/bin/smtpara utility (see the driver
README files). The smtpara utility should be used only by those very familiar with FDDI.

DIAGNOSTICS If multiple NICs are installed in the system, the following message may appear on the console
and in the /var/adm/messages log file:

skfp: DMA memory allocation failed !

You can avoid this message by tuning the lomempages kernel parameter. By default, the value
of this parameter is 36 pages. Each SK-FDDI PCI adapter requires nine pages, so increase the
value of the lomempages parameter in increments of nine pages until all NICs in the system
run correctly.

To modify the value of this parameter to 45 pages, you can, for example, append the set
lomempages=45 line to the /etc/system file and reboot the system.
Files /dev/skfp Character special device
/dev/skfpx Per-adapter character special device, where x is the adapter ppa
/kernel/drv/skfp ELF kernel module
/kernel/drv/skfp.conf Driver configuration file
/usr/bin/smtpara SMT parameter utility
/etc/fddi.cfg smtpara configuration file
<sys/dlpi.h> DLPI definitions

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture IA, SPARC

See Also ifconfig(1M), netstat(1M), attributes(5), dlpi(7P)

skfp.txt README file — Included in the driver package or available from

www.syskonnect.com.

Device and Network Interfaces 803

slp(7P)

Name slp – Service Location Protocol

Description The Service Location Protocol (SLP) is a dynamic service discovery protocol that runs on top
of the Internet Protocol (IP). The protocol is specified by the IETF standard-track documents
RFC 2165, RFC 2608, RFC 2609; the API is documented in RFC 2614. .

There are two components to the SLP technology. The first is a daemon, slpd(1M), which
coordinates SLP operations. The second is a software library, slp_api(3SLP), through which
processes access a public API. Both components are configured by means of the SLP
configuration file, slp.conf(4).

The SLP API is useful for two types of processes:

Client Applications Services and service information can be requested from the API.
Clients do not need to know the location of a required service, only
the type of service, and optionally, the service characteristics. SLP will
supply the location and other information to the client through the
API.
Server Processes Programs that offer network services use the SLP API to advertise
their location as well as other service information. The advertisement
can optionally include attributes describing the service.
Advertisements are accompanied by a lifetime; when the lifetime
expires, the advertisement is flushed, unless it is refreshed prior to
expiration.

API libraries are available for both the C and Java languages.

SLP provides the following additional features:

■ slpd(1M) can be configured to function as a transparent directory agent. This feature
makes SLP scalable to the enterprise. System administrators can configure directory agents
to achieve a number of different strategies for scalability.
■ SLP service advertising and discovery is performed in scopes. Unless otherwise configured,
all discovery and all advertisements are in the scope default. In the case of a larger network,
scopes can be used to group services and client systems so that users will only find those
services which are physically near them, belong to their department, or satisfy the specified
criteria. Administrators can configure these scopes to achieve different service provider
strategies.
■ Services may be registered by proxy through a serialized registration file. This is an
alternative to registering services through the API. See slpd.reg(4) for more information.

Attributes See attributes(5) for descriptions of the following attributes:

804 man pages section 7: Device and Network Interfaces • Last Revised 17 Nov 1999
 slp(7P)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWslpu

CSI CSI-enabled

Interface Stability Standard

MT-Level MT-Safe

See Also slpd(1M), slp_api(3SLP), slp.conf(4), slpd.reg(4), attributes(5)

Guttman, E., Perkins, C., Veizades, J., and Day, M., RFC 2608, Service Location Protocol,
Version 2, The Internet Society, June 1999.

Guttman, E., Perkins, C., and Kempf, J., RFC 2609, Service Templates and Service: Schemes,
The Internet Society, June 1999.

Kempf, J. and Guttman, E., RFC 2614, An API for Service Location, The Internet Society, June
1999.

Veizades, J., Guttman, E., Perkins, C., and Kaplan, S., RFC 2165, Service Location Protocol,
Network Working Group, 1997.

Device and Network Interfaces 805

smbfs(7FS)

Name smbfs – CIFS/SMB file system

Description The smbfs file system allows you to mount CIFS shares that are exported from Windows or
compatible systems. SMB is the historical name for the CIFS protocol, which stands for Server
Message Block and is more commonly used in technical contexts.

The smbfs file system permits ordinary UNIX applications to change directory into an smbfs
mount and perform simple file and directory operations. Supported operations include open,
close, read, write, rename, delete, mkdir, rmdir and ls.

Limitations Some local UNIX file systems (for example UFS) have features that are not supported by
smbfs. These include:
■ No mapped-file access because mmap(2) returns ENOSYS.
■ Locking is local only and is not sent to the server.

The following are limitations in the CIFS protocol:

■ unlink() or rename() of open files returns EBUSY.
■ rename() of extended attribute files returns EINVAL.
■ Creation of files with any of the following illegal characters returns EINVAL: colon (:),
backslash (\\), slash (/), asterisk (*), question mark (?), double quote ("), less than (<),
greater than (>), and vertical bar (|).
■ chmod and chown settings are silently discarded.
■ Links are not supported.
■ Symbolic links are not supported.
■ mknod is not supported. (Only file and directory objects are supported.)

The current smbfs implementation does not support multi-user mounts. Instead, each Unix
user needs to make their own private mount points.

Currently, all access through an smbfs mount point uses the Windows credentials established
by the user that ran the mount command. Normally, permissions on smbfs mount points
should be 0700 to prevent Unix users from using each others' Windows credentials. See the
dirperms option to mount_smbfs(1M) for details regarding how to control smbfs mount
point permissions.

An important implication of this limitation is that system-wide mounts, such as those made
using /etc/vfstab or automount maps are only useful in cases where access control is not a
concern, such as for public read-only resources.

Attributes See attributes(5) for descriptions of the following attributes:

806 man pages section 7: Device and Network Interfaces • Last Revised 9 Sep 2009
 smbfs(7FS)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWsmbfsu

Interface Stability Uncommitted

See Also smbutil(1), mount_smbfs(1M), nsmbrc(4), attributes(5)

Device and Network Interfaces 807

smbios(7D)

Name smbios – System Management BIOS image

Synopsis /dev/smbios

Description The smbios device is a character special file that provides access to a snapshot of the System
Management BIOS (SMBIOS) image exported by the current system. SMBIOS is an
industry-standard mechanism that enables low-level system software to export hardware
configuration information to higher-level system management software. The SMBIOS data
format is defined by the Distributed Management Task Force (DMTF). For more information
on SMBIOS and to obtain a copy of the SMBIOS specification and implementation guidelines,
refer to http://www.dmtf.org.

The SMBIOS image consists of a table of structures, each describing some aspect of the system
software or hardware configuration. The content of the image varies widely by platform and
BIOS vendor and may not exist on some systems. You can use the smbios(1M) utility to
inspect the contents of the SMBIOS image and copy it to a file.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWkvm.i

Interface Stability Stable

See Also prtdiag(1M), smbios(1M), attributes(5)

System Management BIOS Reference Specification, Version 2, Release 4 — 2005

Notes The implementation of a System Management BIOS image is entirely at the discretion of the
system and BIOS vendors. Not all systems export a SMBIOS. The SMBIOS structure content
varies widely between systems and BIOS vendors and frequently does not comply with the
guidelines included in the specification. For example, some structure fields may not be filled in
by the BIOS, while others may be filled in with non-conforming values.

808 man pages section 7: Device and Network Interfaces • Last Revised 26 August 2005
 smbus(7D)

Name smbus – System Management Bus controller driver

Description The smbus driver is a I2C (Inter IC) nexus driver that allows the system to communicate with
various system component chips. SMBus is a two-wire control bus based on the I2C protocol
through which systems can communicate with various I2C devices connected to the bus.

The smbus driver supports byte and block level transfer based on interrupt and polled mode.
Files /platform/sun4u/kernel/drv/sparcv9/smbus 64 bit ELF kernel module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcarx

Architecture SPARC

See Also attributes(5)

Writing Device Drivers

System Management Bus (SMBus) Specification 2.0 — SBS Implementation Forum

The I2C Bus and How To Use It —Philips Semiconductor Document # 98-8080-575-01

Device and Network Interfaces 809

socal(7D)

Name socal – Serial Optical Controller for Fibre Channel Arbitrated Loop (SOC+) device driver
Synopsis socal@sbus-slot,0

Description The Fibre Channel Host Bus Adapter is an SBus card which implements two full duplex Fibre
Channel interfaces. Each Fibre Channel interface can connect to a Fibre Channel Arbitrated
Loop (FC-AL).

The socal device driver is a nexus driver and implements portions of the FC-2 and FC-4
layers of FC-AL.
Files /kernel/drv/socal ELF Kernel Module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also sbus(4), sf(7D), ssd(7D)

Writing Device Drivers

ANSI X3.230-1994, Fibre Channel Physical and Signalling Interface (FC-PH)

ANSI X3.272-1996, Fibre Channel Arbitrated Loop (FC-AL)

Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA)

Diagnostics The messages described below may appear on system console in addition to being logged.

On the console, these messages are preceded by:

socal%d: port %a

where %d is the instance number of the socal controller and %a is the port on the host
adapter.
Fibre Channel Loop is ONLINE The Fibre Channel loop is now online.
Fibre Channel Loop is OFFLINE The Fibre Channel loop is now offline.
attach failed: device in slave-only slot. Move soc+ card to another slot.
attach failed: bad soft state. Driver did not attach, devices will be
inaccessible.
attach failed: unable to alloc xport struct. Driver did not attach, devices will be
inaccessible.

810 man pages section 7: Device and Network Interfaces • Last Revised 9 May 1997
 socal(7D)

attach failed: unable to map eeprom Driver was unable to map device memory;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.
attach failed: unable to map XRAM Driver was unable to map device memory;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.
attach failed: unable to map registers Driver was unable to map device registers;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.
attach failed: unable to access status register Driver was unable to map device registers;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.
attach failed: unable to install interrupt handler Driver was not able to add the interrupt
routine to the kernel. Driver did not
attach to device, devices will be
inaccessible.
attach failed: unable to access host adapter XRAM Driver was unable to access device RAM;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.
attach failed: unable to write host adapter XRAM Driver was unable to write device RAM;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.
attach failed: read/write mismatch in XRAM Driver was unable to verify device RAM;
check for bad hardware. Driver did not
attach to device, devices will be
inaccessible.

Device and Network Interfaces 811

sockio(7I)

Name sockio – ioctls that operate directly on sockets

Synopsis #include <sys/sockio.h>

Description The ioctls listed in this manual page apply directly to sockets, independent of any underlying
protocol. The setsockopt() call (see getsockopt(3SOCKET)) is the primary method for
operating on sockets, rather than on the underlying protocol or network interface. ioctls for
a specific network interface or protocol are documented in the manual page for that interface
or protocol.
SIOCSPGRP The argument is a pointer to an int. Set the process-group ID that will
subsequently receive SIGIO or SIGURG signals for the socket referred to by
the descriptor passed to ioctl to the value of that int. The argument must
be either positive (in which case it must be a process ID) or negative (in
which case it must be a process group).
SIOCGPGRP The argument is a pointer to an int. Set the value of that int to the
process-group ID that is receiving SIGIO or SIGURG signals for the socket
referred to by the descriptor passed to ioctl.
SIOCCATMARK The argument is a pointer to an int. Set the value of that int to 1 if the read
pointer for the socket referred to by the descriptor passed to ioctl points to
a mark in the data stream for an out-of-band message. Set the value of that
int to 0 if the read pointer for the socket referred to by the descriptor passed
to ioctl does not point to a mark in the data stream for an out-of-band
message.

See Also ioctl(2), getsockopt(3SOCKET)

812 man pages section 7: Device and Network Interfaces • Last Revised 8 Nov 1996
 sppptun(7M)

Name sppptun – PPP tunneling pseudo-driver

Synopsis /dev/sppptun

Description The /dev/sppptun pseudo-driver provides an interface for tunneling PPP sessions. This
interface provides PPP over Ethernet (PPPoE) service with Solaris PPP.
Files /dev/sppptun Solaris PPP tunneling device driver.

See Also pppoec(1M), pppoed(1M), sppptun(1M)

RFC 2516 — A Method for Transmitting PPP Over Ethernet (PPPoE). Mamakos, et. al.
February 1999.

Device and Network Interfaces 813

spwr(7D)

Name spwr – SMC EtherPower II 10/100 (9432) Ethernet device driver

Synopsis /dev/spwr

Description The spwr Ethernet driver is a multi-threaded, loadable, clonable, STREAMS hardware driver
supporting the connectionless Data Link Provider Interface, dlpi(7P), over SMC EtherPower
II 10/100 controllers. Multiple EtherPower II controllers installed within the system are
supported by the driver. The spwr driver provides basic support for the SMC EtherPower II
hardware. Functions include chip initialization, frame transmit and receive, multicast
support, and error recovery and reporting.

Application The cloning character-special device /dev/spwr is used to access all SMC EtherPower II
Programming devices installed within the system.
Interface
The spwr driver is dependent on /kernel/misc/gld, a loadable kernel module that provides
the spwr driver with the DLPI and STREAMS functionality required of a LAN driver. See
gld(7D) for more details on the primitives supported by the driver.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ from the user are as follows:
■ The maximum SDU is 1500 ( ETHERMTU).
■ The minimum SDU is 0. The spwr driver will pad to the mandatory 60-octet minimum
packet size.
■ The DLSAP address length is 8.
■ The MAC type is DL_ETHER.
■ The SAP length value is –2, meaning the physical address component is followed
immediately by a 2-byte SAP component within the DLSAP address.
■ The broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).
Files /dev/spwr Character special device.
/kernel/drv/spwr.conf Driver configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), dlpi(7P), gld(7D)

814 man pages section 7: Device and Network Interfaces • Last Revised 2 Dec 1998
 srpt(7D)

Name srpt – SCSI RDMA Protocol Target Driver for Infiniband (IB)

Description The srpt kernel pseudo device driver is an IB Architecture-compliant implementation of the
target side of the SCSI RDMA Protocol (SRP). SRP accelerates the SCSI protocol by mapping
SCSI data transfer phases to RDMA operations using InfiniBand as the underlying transport.

SRP target services are enabled and disabled through smf(5), using the FMRI
svc:/system/ibsrp/target.

When enabled, srpt enumerates each IB Host Channel Adapter (HCA) on the system and
registers each one as a SCSI target using the SCSI Target Mode Framework (STMF).
Files /kernel/drv/srpt 32-bit x86 ELF kernel module
/kernel/drv/amd64/srpt 64-bit x86 ELF kernel module
/kernel/drv/sparcv9/srpt 64-bit SPARC ELF kernel module
/kernel/drv/srpt.conf Driver configuration file

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWsrptr

See Also stmfadm(1M), ib(7D), ibdma(7D), ibtl(7D), attributes(5)

COMSTAR Administration Guide

SCSI RDMA Protocol (SRP) T10 Project 1415-D, Revision

Device and Network Interfaces 815

ssd(7D)

Name ssd – Fibre Channel Arbitrated Loop disk device driver

Synopsis ssd@port,target:partition

Description The ssd driver supports Fibre Channel disk devices.

The specific type of each disk is determined by the SCSI inquiry command and reading the
volume label stored on block 0 of the drive. The volume label describes the disk geometry and
partitioning; it must be present or the disk cannot be mounted by the system.

The block-files access the disk using the system's normal buffering mechanism and are read
and written without regard to physical disk records. A “raw” interface provides for direct
transmission between the disk and the read or write buffer. A single read or write call usually
results in one I/O operation; raw I/O is therefore more efficient when many bytes are
transmitted. Block file names are found in /dev/dsk; the names of the raw files are found in
/dev/rdsk.

I/O requests (such as lseek(2)) to the SCSI disk must have an offset that is a multiple of 512
bytes (DEV_BSIZE), or the driver returns an EINVAL error. If the transfer length is not a multiple
of 512 bytes, the transfer count is rounded up by the driver.

Partition 0 is normally used for the root file system on a disk, with partition 1 as a paging area
(for example, swap). Partition 2 is used to back up the entire disk. Partition 2 normally maps
the entire disk and may also be used as the mount point for secondary disks in the system. The
rest of the disk is normally partition 6. For the primary disk, the user file system is located here.

The device has associated error statistics. These must include counters for hard errors, soft
errors and transport errors. Other data may be implemented as required.

Device Statistics The device maintains I/O statistics for the device and for partitions allocated for that device.
Support For each device/partition, the driver accumulates reads, writes, bytes read, and bytes written.
The driver also initiates hi-resolution time stamps at queue entry and exit points to enable
monitoring of residence time and cumulative residence-length product for each queue.

Not all device drivers make per-partition IO statistics available for reporting. ssd and sd(7D)
per-partition statistics are enabled by default but may be disabled in their configuration files.

ioctls Refer to dkio(7I).

Errors EACCES Permission denied.
EBUSY The partition was opened exclusively by another thread.
EFAULT The argument was a bad address.
EINVAL Invalid argument.
EIO An I/O error occurred.
ENOTTY The device does not support the requested ioctl function.

816 man pages section 7: Device and Network Interfaces • Last Revised 9 Aug 2008
 ssd(7D)

ENXIO When returned during open(2), this error indicates the device does not exist.
EROFS The device is a read-only device.

Configuration You configure the ssd driver by defining properties in the ssd.conf file. The ssd driver
supports the following properties:
enable-partition-kstats The default value is 1, which causes partition IO statistics to
be maintained. Set this value to zero to prevent the driver
from recording partition statistics. This slightly reduces the
CPU overhead for IO, mimimizes the amount of sar(1) data
collected and makes these statistics unavailable for reporting
by iostat(1M) even though the -p/-P option is specified.
Regardless of this setting, disk IO statistics are always
maintained.

In addition to the above properties, some device-specific tunables can be configured in

ssd.conf using the 'ssd-config-list' global property. The value of this property is a list of
duplets. The formal syntax is:
ssd-config-list = <duplet> [, <duplet> ]* ;

where

<duplet>:= "<vid+pid>" , "<tunable-list>"

and

<tunable-list>:= <tunable> [, <tunable> ]*;

<tunable> = <name> : <value>

The <vid+pid> is the string that is returned by the target device

on a SCSI inquiry command.

The <tunable-list> contains one or more tunables to apply to

all target devices with the specified <vid+pid>.

Each <tunable> is a <name> : <value> pair. Supported

tunable names are:

delay-busy: when busy, nsecs of delay before retry.

retries-timeout: retries to perform on an IO timeout.

Examples The following is an example of a global ssd-config-list property:

ssd-config-list =
"SUN T4", "delay-busy:600, retries-timeout:6",

Device and Network Interfaces 817

ssd(7D)

"SUN StorEdge_3510", "retries-timeout:3";

Files ssd.conf Driver configuration file
/dev/dsk/cntndnsn block files
/dev/rdsk/cntndnsn raw files
cn is the controller number on the system.
tn 7-bit disk loop identifier, such as switch setting
dn SCSI lun n
sn partition n (0-7)

See Also sar(1), format(1M), iostat(1M), ioctl(2), lseek(2), open(2), read(2), write(2),
scsi(4)driver.conf(4), cdio(7I), dkio(7I)

ANSI Small Computer System Interface-2 (SCSI-2)

ANSI X3.272-1996, Fibre Channel Arbitrated Loop (FC-AL)

Fibre Channel - Private Loop SCSI Direct Attach (FC-PLDA)

Diagnostics Error for command ’<command name>’ Error Level: Fatal Requested Block <n>,
Error Block: <m>, Sense Key: <sense key name>, Vendor ’<vendor name>’:
ASC = 0x<a> (<ASC name>), ASCQ = 0x<b>, FRU = 0x<c>

The command indicated by <command name> failed. The Requested Block is the block where
the transfer started and the Error Block is the block that caused the error. Sense Key, ASC, and
ASCQ information is returned by the target in response to a request sense command.
Check Condition on REQUEST SENSE

A REQUEST SENSE command completed with a check condition. The original command
will be retried a number of times.
Label says <m> blocks Drive says <n> blocks

There is a discrepancy between the label and what the drive returned on the READ
CAPACITY command.
Not enough sense information

The request sense data was less than expected.

Request Sense couldn’t get sense data

The REQUEST SENSE command did not transfer any data.

Reservation Conflict

818 man pages section 7: Device and Network Interfaces • Last Revised 9 Aug 2008
 ssd(7D)

The drive was reserved by another initiator.

SCSI transport failed: reason ’xxxx’ : {retrying|giving up}

The host adapter has failed to transport a command to the target for the reason stated. The
driver will either retry the command or, ultimately, give up.
Unhandled Sense Key <n>

The REQUEST SENSE data included an invalid sense key.

Unit not Ready. Additional sense code 0x<n>

The drive is not ready.

corrupt label - bad geometry

The disk label is corrupted.

corrupt label - label checksum failed

The disk label is corrupted.

corrupt label - wrong magic number

The disk label is corrupted.

device busy too long

The drive returned busy during a number of retries.

disk not responding to selection

The drive was probably powered down or died.

i/o to invalid geometry

The geometry of the drive could not be established.

incomplete read/write - retrying/giving up

There was a residue after the command completed normally.

logical unit not ready

The drive is not ready.

no bp for disk label

A bp with consistent memory could not be allocated.

no mem for property

Free memory pool exhausted.

Device and Network Interfaces 819

ssd(7D)

no memory for disk label

Free memory pool exhausted.

no resources for dumping

A packet could not be allocated during dumping.

offline

Drive went offline; probably powered down.

requeue of command fails<n>

Driver attempted to retry a command and experienced a transport error.

ssdrestart transport failed <n>

Driver attempted to retry a command and experienced a transport error.

transfer length not modulo <n>

Illegal request size.

transport rejected <n>

Host adapter driver was unable to accept a command.

unable to read label

Failure to read disk label.

unit does not respond to selection

Drive went offline; probably powered down.

820 man pages section 7: Device and Network Interfaces • Last Revised 9 Aug 2008
 st(7D)

Name st – driver for SCSI tape devices

Synopsis st@target,lun:l,m,h,c,ubn

Description The st device driver provides a standard interface to various SCSI tape devices. See mtio(7I)
for details.

To determine if the st device driver supports your tape device, SPARC users should enter the
following on a command line:
% strings /kernel/drv/sparcv9/st | grep -i <tape device name>

x86 users can do the following to determine if the st device driver supports a particular tape
device:
% strings /kernel/drv/st | grep -i <tape device name>

The driver can be opened with either rewind on close or no rewind on close options. It can also
be opened with the O_NDELAY (see open(2)) option when there is no tape inserted in the drive.
A maximum of four tape formats per device are supported (see FILES below). The tape format
is specified using the device name. (Tape format is also referred to as tape density).

Following are a list of SCSI commands that can be executed while another host reserves the
tape drive. The commands are:
SCMD_TEST_UNIT_READY
SCMD_REQUEST_SENSE
SCMD_READ_BLKLIM
SCMD_INQUIRY
SCMD_RESERVE
SCMD_RELEASE
SCMD_DOORLOCK
SCMD_REPORT_DENSITIES
SCMD_LOG_SENSE_G1
SCMD_PERSISTENT_RESERVE_IN
SCMD_PERSISTENT_RESERVE_OUT
SCMD_REPORT_LUNS

In multi-initiator environments, the driver will not reserve the tape drive if above commands
are issued. For other SCSI commands, the driver reserves the tape drive and releases the drive
at close if it has been reserved. Refer to the MTIOCRESERVE and MTIOCRELEASE ioctls in
mtio(7I) for information about how to allow a tape drive to remain reserved upon close. See
the flag options below for information about disabling this feature.

If a SCSI-3 persistent reservation is done through the driver, the driver disables all existing
SCSI-2 reservations.

Device and Network Interfaces 821

st(7D)

If the tape drive is opened in O_NDELAY mode, no reservation occurs during the open, as per
the POSIX standard (see standards(5)). However, if a command not found in the above list is
used, a reservation will occur to provide reserve/release functionality before the command is
issued.

Persistent Errors and The st driver now supports persistent errors (see mtio(7I) and asynchronous tape operations
Asynchronous Tape (see mtio(7I), aioread(3C), and aiowrite(3C)).
Operation
Read Operation If the driver is opened for reading in a different format than the tape is written in, the driver
overrides the user-selected format. For example, if a 1/4” cartridge tape is written in QIC-24
format and opened for reading in QIC-150, the driver detects a read failure on the first read
and automatically switches to QIC-24 to read the data.

Note that if the low density format is used, no indication is given that the driver has
overridden the format you selected. Other formats issue a warning message to inform you of
an overridden format selection. Some devices automatically perform this function and do not
require driver support (1/2” reel tape drive, for example).

Write Operation Writing from the beginning of tape is performed in the user-specified format. The original
tape format is used for appending onto previously written tapes.

Tape Configuration The st driver has a built-in configuration table for most Sun-supported tape drives. For those
tape drives that are not in the table, the st driver tries to read the configuration from the tape
drive through optional SCSI-3 commands. To support the addition of third party tape devices
which are not in the built-in configuration table or not able to report their configuration,
device information can be supplied in st.conf as global properties that apply to each node, or
as properties that are applicable to one node only. By supplying the information in st.conf,
the built-in configuration is overridden and the st driver will not query the configuration
from tape drives. The st driver looks for the property called tape-config-list. The value of
this property is a list of triplets, where each triplet consists of three strings.

The formal syntax is:

tape-config-list = <triplet> [, <triplet> *];

where
<triplet> := <vid+pid>, <pretty print>, <data-property-name>

and
<data-property-name> = <version>, <type>, <bsize>,
<options>, <number of densities>,
<density> [, <density>*], <default-density>;

or

822 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

<data-property-name> = <version 2>, <type>, <bsize>,

<options>, <number of densities>,
<density> [, <density>*], <default-density>,
<non-motion time-out>, <I/O time-out>,
<rewind time-out>, <space time-out>,
<load time-out>, <unload time-out>,
<erase time-out>;

A semicolon (;) is used to terminate a prototype devinfo node specification. Individual

elements listed within the specification should not be separated by a semicolon. (Refer to
driver.conf(4) for more information.)

<vid+pid> is the string that is returned by the tape device on a SCSI inquiry command. This
string may contain any character in the range 0x20-0x7e. Characters such as “ " ” (double
quote) or “ ' ” (single quote), which are not permitted in property value strings, are represented
by their octal equivalent (for example, \042 and \047). Trailing spaces may be truncated.

<pretty print> is used to report the device on the console. This string may have zero length,
in which case the <vid+pid> will be used to report the device.

<data-property-name> is the name of the property which contains all the tape configuration
values (such as <type>, <bsize>, etc.) corresponding for the tape drive for the specified
<vid+pid>.

<version> is a version number and should be 1 or 2. In the future, higher version numbers
may be used to allow for changes in the syntax of the <data-property-name> value list.

<type> is a type field. Valid types are defined in /usr/include/sys/mtio.h. For third party
tape configuration, the following generic types are recommended:

MT_ISQIC 0x32

MT_ISREEL 0x33

MT_ISDAT 0x34

MT_IS8MM 0x35

MT_ISOTHER 0x36

MT_ISTAND25G 0x37

MT_ISDLT 0x38

MT_ISSTK9840 0x39

MT_ISBMDLT1 0x3A

Device and Network Interfaces 823

st(7D)

MT_LTO 0x3B

<bsize> is the preferred block size of the tape device. The value should be 0 for variable block
size devices.

<options> is a bit pattern representing the devices, as defined in

/usr/include/sys/scsi/targets/stdef.h. Valid flags for tape configuration are shown in
the following table. Note that this table does not list flags that are non-configurable in st.conf
(including ST_KNOWS_MEDIA which uses the media type reported from the mode select data to
select the correct density code).

ST_VARIABLE 0x0001

ST_QIC 0x0002

ST_REEL 0x0004

ST_BSF 0x0008

ST_BSR 0x0010

ST_LONG_ERASE 0x0020

ST_AUTODEN_OVERRIDE 0x0040

ST_NOBUF 0x0080

ST_KNOWS_EOD 0x0200

ST_UNLOADABLE 0x0400

ST_SOFT_ERROR_REPORTING 0x0800

ST_LONG_TIMEOUTS 0x1000

ST_NO_RECSIZE_LIMIT 0x8000

ST_MODE_SEL_COMP 0x10000

ST_NO_RESERVE_RELEASE 0x20000

ST_READ_IGNORE_ILI 0x40000

ST_READ_IGNORE_EOFS 0x80000

ST_SHORT_FILEMARKS 0x100000

ST_EJECT_TAPE_ON_CHANGER_FAILURE 0x200000

ST_RETRY_ON_RECOVERED_DEFERRED_ERROR 0x400000

ST_WORMABLE 0x1000000

824 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

ST_VARIABLE The flag indicates the tape device supports

variable length record sizes.
ST_QIC The flag indicates a Quarter Inch Cartridge
(QIC) tape device.
ST_REEL The flag indicates a 1/2−inch reel tape device.
ST_BSF If flag is set, the device supports backspace
over EOF marks (bsf - see mt(1)).
ST_BSR If flag is set, the tape device supports the
backspace record operation (bsr - see mt(1)).
If the device does not support bsr, the st
driver emulates the action by rewinding the
tape and using the forward space record (fsf)
operation to forward the tape to the correct
file. The driver then uses forward space
record (fsr - see mt(1)) to forward the tape to
the correct record.
ST_LONG_ERASE The flag indicates the tape device needs a
longer time than normal to erase.
ST_AUTODEN_OVERRIDE The auto-density override flag. The device is
capable of determining the tape density
automatically without issuing a
“mode-select”/“mode-sense command.”
ST_NOBUF The flag disables the device's ability to
perform buffered writes. A buffered write
occurs when the device acknowledges the
completion of a write request after the data
has been written to the device's buffer, but
before all of the data has been written to the
tape.
ST_KNOWS_EOD If flag is set, the device can determine when
EOD (End of Data) has been reached. When
this flag is set, the st driver uses fast file
skipping. Otherwise, file skipping happens
one file at a time.
ST_UNLOADABLE The flag indicates the device will not
complain if the st driver is unloaded and
loaded again (see modload(1M) and
modunload(1M)). That is, the driver will
return the correct inquiry string.

Device and Network Interfaces 825

st(7D)

ST_SOFT_ERROR_REPORTING The flag indicates the tape device will

perform a “request sense” or “log sense”
command when the device is closed.
Currently, only Exabyte and DAT drives
support this feature.
ST_LONG_TIMEOUTS The flag indicates the tape device requires
timeouts that are five times longer than usual
for normal operation.
ST_NO_RECSIZE_LIMIT The flag applies to variable-length tape
devices. If this flag is set, the record size is not
limited to a 64 Kbyte record size. The record
size is only limited by the smaller of either the
record size supported by the device or the
maximum DMA transfer size of the system.
(Refer to Large Record Sizes and
WARNINGS.) The maximum block size that will
not be broken into smaller blocks can be
determined from the mt_bf returned from
the MTIOCGET ioctl(). This number is the
lesser of the upper block limit returned by the
drive from READ BLOCK LIMITS command
and the dma-max property set by the Host
Bus Adapter (HBA) to which the drive is
attached.
ST_MODE_SEL_COMP If the ST_MODE_SEL_COMP flag is set, the driver
determines which of the two mode pages the
device supports for selecting or deselecting
compression. It first tries the Data
Compression mode page (0x0F); if this fails, it
tries the Device Configuration mode page
(0x10). Some devices, however, may need a
specific density code for selecting or
deselecting compression. Please refer to the
device specific SCSI manual. When the flag is
set, compression is enabled only if the "c" or
"u" device is used. Note that when the lower 2
densities of a drive are identically configured
and the upper 2 densities are identically
configured, but the lower and upper differ
from each other and
ST_MODE_SEL_COMP is set, the “m” node
sets compression on for the lower density

826 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

code (for example, 0x42) and the "c" and "u"

nodes set compression on for the higher
density (for example, 0x43). For any other
device densities, compression is disabled.
ST_NO_RESERVE_RELEASE The ST_NO_RESERVE_RELEASE flag disables
the use of reserve on open and release on
close. If an attempt to use a ioctl of MTRESERVE
or MTRELEASE on a drive with this flag set, it
will return an error of ENOTTY (inappropriate
ioctl for device).
ST_READ_IGNORE_ILI The ST_READ_IGNORE_ILI flag is applicable
only to variable block devices which support
the SILI bit option. The ST_READ_IGNORE_ILI
flag indicates that SILI (supress incorrect
length indicator) bit will be set during reads.
When this flag is set, short reads (requested
read size is less than the record size on the
tape) will be successful and the number of
bytes transferred will be equal to the record
size on the tape. The tape will be positioned at
the start of the next record skipping over the
extra data (the remaining data has been has
been lost). Long reads (requested read size is
more than the record size on the tape) will see
a large performance gain when this flag is set,
due to overhead reduction. When this flag is
not set, short reads will return an error of
ENOMEM.
ST_READ_IGNORE_EOFS The ST_READ_IGNORE_EOFS flag is applicable
only to 1/2" Reel Tape drives and when
performing consecutive reads only. It should
not be used for any other tape command.
Usually End-of-recorded-media (EOM) is
indicated by two EOF marks on 1/2" tape and
application cannot read past EOM. When
this flag is set, two EOF marks no longer
indicate EOM allowing applications to read
past two EOF marks. In this case it is the
responsibility of the application to detect
end-of-recorded-media (EOM). When this
flag is set, tape operations (like MTEOM)
which positions the tape at

Device and Network Interfaces 827

st(7D)

end-of-recorded-media will fail since

detection of end-of-recorded-media (EOM)
is to be handled by the application. This flag
should be used when backup applications
have embedded double filemarks between
files.
ST_SHORT_FILEMARKS The ST_SHORT_FILEMARKS flag is applicable
only to EXABYTE 8mm tape drives which
supports short filemarks. When this flag is
set, short filemarks is used for writing
filemarks. Short filemarks could lead to tape
incompatible with some otherwise
compatible device. By default long filemarks
will be used for writing filemarks.
ST_EJECT_TAPE_ON_CHANGER_FAILURE If ST_EJECT_TAPE_ON_CHANGER_FAILURE flag
is set, the tape is ejected automatically if the
tape cartridge is trapped in the medium due
to positioning problems of the medium
changer.

The following ASC/ASCQ keys are defined to

the reasons for causing tape ejection if
ST_EJECT_TAPE_ON_CHANGER_FAILURE
option is set to 0x200000:

Sense ASC/ASCQ Description

Key

4 15/01 Mechanical Failure

4 44/00 Internal Target Failure

2 53/00 Media Load or Eject Failed

4 53/00 Media Load or Eject Failed

4 53/01 Unload Tape Failure

ST_RETRY_ON_RECOVERED_DEFERRED_ERROR If
ST_RETRY_ON_RECOVERED_DEFERRED_ERROR
flag is set, the st driver will retry the last write
if this cmd caused a check condition with

828 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

error code 0x71 and sense code 0x01. Some

tape drives, notably the IBM 3090, require
this option.
ST_WORMABLE When ST_WORMABLE is set, st attempts to
detect the presence of WORM media in the
device.

<number of densities> is the number of densities specified. Each tape drive can support up
to four densities. The value entered should therefore be between 1 and 4; if less than 4, the
remaining densities will be assigned a value of 0x0.

<density> is a single-byte hexadecimal number. It can either be found in the device

specification manual or be obtained from the device vendor.

<default-density> has a value between 0 and (<number of densities> - 1).

<non-motion time-out> Time in seconds that the drive should be able to perform any SCSI
command that doesn't require tape to be moved. This includes mode sense, mode select,
reserve, release, read block limits, and test unit ready.

<I/O time-out> Time in seconds to perform data transfer I/O to or from tape including worst
case error recovery.

<rewind time-out> Time in seconds to rewind from anywhere on tape to BOT including
worst case recovery forcing buffered write data to tape.

<space time-out> Time in seconds to space to any file, block or end of data on tape.
Including worst case when any form of cataloging is invalid.

<load time-out> Time in seconds to load tape and be ready to transfer first block. This
should include worst case recovery reading tape catalog or drive specific operations done at
load.

<unload time-out> Time in seconds to unload tape. Should include worst case time to write
to catalog, unthread, and tape cartridge unloading. Also should include worst case time for
any drive specific operations that are preformed at unload. Should not include rewind time as
the driver rewinds tape before issuing the unload.

<erase time-out> Time in seconds to preform a full (BOT to EOT) erase of longest medium
with worst case error recovery.

Device Statistics Each device maintains I/O statistics both for the device and for each partition allocated on that
Support device. For each device/partition, the driver accumulates reads, writes, bytes read, and bytes
written. The driver also takes hi-resolution time stamps at queue entry and exit points, which
facilitates monitoring the residence time and cumulative residence-length product for each
queue.

Device and Network Interfaces 829

st(7D)

Each device also has error statistics associated with it. These must include counters for hard
errors, soft errors and transport errors. Other data may be implemented as required.

ioctls The behavior of SCSI tape positioning ioctls is the same across all devices which support them.
(Refer to mtio(7I).) However, not all devices support all ioctls. The driver returns an ENOTTY
error on unsupported ioctls.

The retension ioctl only applies to 1/4” cartridge tape devices. It is used to restore tape tension,
thus improving the tape's soft error rate after extensive start-stop operations or long-term
storage.

In order to increase performance of variable-length tape devices (particularly when they are
used to read/write small record sizes), two operations in the MTIOCTOP ioctl, MTSRSZ and
MTGRSZ, can be used to set and get fixed record lengths. The ioctl also works with fixed-length
tape drives which allow multiple record sizes. The min/max limits of record size allowed on a
driver are found by using a SCSI-2 READ BLOCK LIMITS command to the device. If this
command fails, the default min/max record sizes allowed are 1 byte and 63k bytes. An
application that needs to use a different record size opens the device, sets the size with the
MTSRSZ ioctl, and then continues with I/O. The scope of the change in record size remains
until the device is closed. The next open to the device resets the record size to the default
record size (retrieved from st.conf).

Note that the error status is reset by the MTIOCGET get status ioctl call or by the next read, write,
or other ioctl operation. If no error has occurred (sense key is 0), the current file and record
position is returned.
Errors EACCES The driver is opened for write access and the tape is write-protected or the tape
unit is reserved by another host.
EBUSY The tape drive is in use by another process. Only one process can use the tape
drive at a time. The driver will allow a grace period for the other process to finish
before reporting this error.
EINVAL The number of bytes read or written is not a multiple of the physical record size
(fixed-length tape devices only).
EIO During opening, the tape device is not ready because either no tape is in the drive,
or the drive is not on-line. Once open, this error is returned if the requested I/O
transfer could not be completed.
ENOTTY This indicates that the tape device does not support the requested ioctl function.
ENXIO During opening, the tape device does not exist.
ENOMEM This indicates that the record size on the tape drive is more than the requested size
during read operation.

830 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

Examples EXAMPLE 1 Global tape-config list property

The following is an example of a global tape-config-list property:

tape-config-list =
"Magic DAT", "Magic 4mm Helical Scan", "magic-data",
"Major Appliance", "Major Appliance Tape", "major-tape";

magic-data = 1,0x34,1024,0x1639,4,0,0x8c,0x8c,0x8c,3;
major-tape = 2,0x3c,0,0x18619,4,0x0,0x0,0x0,0x0,
3,0,0,30,120,0,0,36000;

name="st" class="scsi"
target=0 lun=0;
name="st" class="scsi"
target=1 lun=0;
name="st" class="scsi"
target=2 lun=0;
.
.
.
name="st" class="scsi"
target=6 lun=0;

EXAMPLE 2 Tape-config-list property applicable to target 2 only

The following is an example of a tape-config-list property applicable to target 2 only:

name="st" class="scsi"
target=0 lun=0;
name="st" class="scsi"
target=1 lun=0;
name="st" class="scsi"
target=2 lun=0
tape-config-list =
"Magic DAT", "Magic 4mm Helical Scan", "magic-data"
magic-data = 1,0x34,1024,0x1639,4,0,0x8c,0x8c,0x8c,3;
name="st" class="scsi"
target=3 lun=0;
.
.
.
name="st" class="scsi"
target=6 lun=0;

Device and Network Interfaces 831

st(7D)

Large Record Sizes To support applications such as seismic programs that require large record sizes, the flag
ST_NO_RECSIZE_LIMIT must be set in drive option in the configuration entry. A SCSI tape
drive that needs to transfer large records should OR this flag with other flags in the 'options'
field in st.conf. (Refer to Tape Configuration.) By default, this flag is set for the built-in
config entries of Archive DAT and Exabyte drives.

If this flag is set, the st driver issues a SCSI-2 READ BLOCK LIMITS command to the device to
determine the maximum record size allowed by it. If the command fails, st continues to use
the maximum record sizes mentioned in the mtio(7I) man page.

If the command succeeds, st restricts the maximum transfer size of a variable-length device to
the minimum of that record size and the maximum DMA size that the host adapter can
handle. Fixed-length devices are bound by the maximum DMA size allocated by the machine.
Note that tapes created with a large record size may not be readable by earlier releases or on
other platforms.

(Refer to the WARNINGS section for more information.)

EOT Handling The Emulex drives have only a physical end of tape (PEOT); thus it is not possible to write past
EOT. All other drives have a logical end of tape (LEOT) before PEOT to guarantee flushing the
data onto the tape. The amount of storage between LEOT and PEOT varies from less than 1
Mbyte to about 20 Mbyte, depending on the tape drive.

If EOT is encountered while writing an Emulex, no error is reported but the number of bytes
transferred is 0 and no further writing is allowed. On all other drives, the first write that
encounters EOT will return a short count or 0. If a short count is returned, then the next write
will return 0. After a zero count is returned, the next write returns a full count or short count.
A following write returns 0 again. It is important that the number and size of trailer records be
kept as small as possible to prevent data loss. Therefore, writing after EOT is not
recommended.

Reading past EOT is transparent to the user. Reading is stopped only by reading EOF's. For
1/2” reel devices, it is possible to read off the end of the reel if one reads past the two file marks
which mark the end of recorded media.
Files /kernel/drv/st.conf driver configuration file
/usr/include/sys/mtio.h structures and definitions for mag tape io
control commands
/usr/include/sys/scsi/targets/stdef.h definitions for SCSI tape drives
/dev/rmt/[0− 127][l,m,h,u,c][b][n] where l,m,h,u,c specifies the density (low,
medium, high, ultra/compressed), b the
optional BSD behavior (see mtio(7I)), and n
the optional no rewind behavior. For

832 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

example, /dev/rmt/0lbn specifies unit 0,

low density, BSD behavior, and no rewind.

For 1/2” reel tape devices (HP-88780), the

densities are:

l 800 BPI density

m 1600 BPI density

h 6250 BPI density

c data compression

(not supported on all

modules)

For 8mm tape devices (Exabyte

8200/8500/8505):

l Standard 2 Gbyte
format

m 5 Gbyte format (8500,

8505 only)

h,c 5 Gbyte compressed

format (8505 only)

For 4mm DAT tape devices (Archive

Python):

l Standard format

m,h,c data compression

For all QIC (other than QIC-24) tape

devices:

l,m,h,c density of the tape

cartridge type

(not all devices can read

and

Device and Network Interfaces 833

st(7D)

write all formats)

For QIC-24 tape devices (Emulex MT−02):

l QIC-11 Format

m,h,c QIC-24 Format

See Also mt(1), modload(1M), modunload(1M), open(2), read(2), write(2), aioread(3C),

aiowrite(3C), kstat(3KSTAT), driver.conf(4), scsi(4), standards(5), esp(7D), isp(7D),
mtio(7I), ioctl(9E)

Diagnostics The st driver diagnostics may be printed to the console or messages file.

Each diagnostic is dependent on the value of the system variable st_error_level.

st_error_level may be set in the /etc/system file. The default setting for st_error_level
is 4 (SCSI_ERR_RETRYABLE) which is suitable for most configurations since only actual
fault diagnostics are printed. Settings range from values 0 (SCSI_ERR_ALL) which is most
verbose, to 6 (SCSI_ERR_NONE) which is least verbose. See stdef.h for the full list of
error-levels. SCSI_ERR_ALL level the amount of diagnostic information is likely to be
excessive and unnecessary.

The st driver diagnostics are described below:

Error for Command: <scsi_cmd_name()> Error Level:<error_class>
Requested Block: <blkno> Error Block: <err_blkno>
Vendor: <name>: Serial Number: <inq_serial>
Sense Key: <es_key> ASC: 0x<es_add_code> (scsi_asc_ascq_name()>), ASCQ:
0x<es_qual_code>, FRU: 0x<ex_fru_code>

where <error_class> may be any one of the following: "All," "Unknown," "Informational,"
"Recovered," "Retryable," "Fatal"

The command indicated by <scsi_cmd_name> failed. Requested Block represents the block
where the transfer started. Error Block represents the block that caused the error. Sense Key,
ASC, ASCQ and FRU information is returned by the target in response to a request sense
command. See SCSI protocol documentation for description of Sense Key, ASC, ASCQ, FRU.

The st driver attempts to validate entries in the st.conf file. Each field in the entry is checked
for upper and lower limits and invalid bits set. The fields are named as follows in config string
order:
conf version
drive type
block size
options
number of densities

834 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

density code
default density
non motion timeout
I/O timeout
space timeout
load timeout
unload timeout
erase timeout

The st.conf diagnostics are described below:

<con-name> <field-in-err> <problem-with-field>

where <con-name> is the name of the config string. Where <field-in-err> is the field
containing invalid entries and where <problem-with-field> describes the nature of the invalid
entry.
Write/read: not modulo <n> block size

The request size for fixed record size devices must be a multiple of the specified block size.
Recovery by resets failed

After a transport error, the driver attempted to recover by issuing a device reset and then a bus
reset if device reset failed. These recoveries failed.
Periodic head cleaning required

The driver reported that periodic head cleaning is now required. This diagnostic is generated
either due to a threshold number of retries, or due to the device communicating to the driver
that head cleaning is required.
Soft error rate (<n>%) during writing/reading was too high

The soft error rate has exceeded the threshold specified by the vendor.
SCSI transport failed: reason ’xxxx’: {retrying|giving up}

The Host Bus Adapter (HBA) has failed to transport a command to the target for the reason
stated. The driver will either retry the command or, ultimately, give up.
Tape not inserted in drive

A media access command was attempted while there was no tape inserted into the specified
drive. In this case, the drive returns sense key of DRIVE NOT READY.
Transport rejected

The Host Bus Adapter (HBA) driver is not accepting commands after failing to successfully
transport a scsi packet to the target. The actual status received by the st driver from the
underlying HBA driver was either TRAN_FATAL_ERROR or TRAN_BADPKT.
Retrying command

Device and Network Interfaces 835

st(7D)

The st driver failed to complete a command. However the command is retryable and will be
retried.
Giving up

The st driver has exhausted retries or otherwise is unable to retry the command and so is
giving up.
No target struct for st%d

The st driver failed to obtain state information because the requested state structure was not
allocated. The specified device was probably not attached.
File mark detected

The operation detected an end of file mark. (File marks signify the end of a file on the tape
media).
End-of-media detected

The operation reached the end of the tape media.

Exabyte soft error reporting failed. DAT soft error reporting failed

The st driver was unable to determine if the soft error threshold had been exceeded because it
did not successfully read the data it requires or did not obtain enough data. This data is
retrieved using the log sense command.
Log sense parameter code does not make sense

The log sense command retrieves hardware statistics that are stored on the drive (for example,
soft error counts and retries.) If the data retrieved from the drive is invalid, this message is
printed and the data is not used.
Restoring tape position at fileno=%x, blkno=%lx....

The st driver is positioning to the specified file and block. This occurs on an open.
Failed to restore the last <file/block> position:
In this state, tape will be loaded at BOT during next open

The st driver could not position to the specified location and will revert to the beginning of
the tape when the next open is attempted.
Device does not support compression

The compression facility of the device was requested. However the device does not have a
hardware compression capability.
DAT soft error reset failed

After DAT soft error reporting, the counters within the device that accumulate this sense data
need to be re-set. This operation failed.

836 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 st(7D)

Errors after pkt alloc (b_flags=0x%x, b_error=0x%x)

Memory allocation for a scsi packet failed.

Incorrect length indicator set

The drive reported the length of data requested in a READ operation, was incorrect. Incorrect
Length Indicator (ILI) is a very commonly used facility in SCSI tape protocol and should not
be seen as an error per-se. Applications typically probe a new tape with a read of any length,
using the returned length to the read system call for future reads. Along with this operation, an
underlying ILI error is received. ILI errors are therefore informational only and are masked at
the default st_error_level.
Data property (%s) has no value
Data property (%s) incomplete
Version # for data property (%s) greater than 1

These diagnostics indicate problems in retrieving the values of the various property settings.
The st driver is in the process of setting the property/parameter values for the tape drive using
information from either the built-in table within the driver or from uncommented entries in
the st.conf file. The effect on the system may be that the tape drive may be set with default or
generic driver settings which may not be appropriate for the actual type of tape drive being
used.
st_attach-RESUME: tape failure tape position will be lost

On a resume after a power management suspend, the previously known tape position is no
longer valid. This can occur if the tape was changed while the system was in power
management suspend. The operation will not be retried.
Write Data Buffering has been deprecated. Your applications should
continue to work normally. However, they should be ported to use
Asynchronous I/O.

Indicates that buffering has been removed from Solaris.

Cannot detach: fileno=%x, blkno=%lx

The st driver cannot unload because the tape is not positioned at BOT (beginning of tape).
May indicate hardware problems with the tape drive.
Variable record length I/O
Fixed record length (%d byte blocks) I/O

Tape-drives can use either Fixed or Variable record length. If the drive uses Fixed length
records, then the built in property table or the st.conf file will contain a non-zero
record-length property. Most DAT, Exabyte and DLT drives support Variable record lengths.
Many QIC format tape drives have historically been of Fixed record length.
Command will be retried
un_ncmds: %d can’t retry cmd

Device and Network Interfaces 837

st(7D)

These diagnostics are only seen with tape drives with the
ST_RETRY_ON_RECOVERED_DEFERRED_ERROR bit set. See stdef.h for explanation of
the specific usage of this setting.

Warnings Effective with Solaris 2.4, the ST_NO_RECSIZE_LIMIT flag is set for the built-in config entries of
the Archive DAT and Exabyte drivers by default. (Refer to Large Record Sizes.) Tapes
written with large block sizes prior to Solaris 2.4 may cause some applications to fail if the
number of bytes returned by a read request is less than the requested block size (for example,
asking for 128 Kbytes and receiving less than 64 Kbytes).

The ST_NO_RECSIZE_LIMIT flag can be disabled in the config entry for the device as a
work-around. (Refer to Tape Configuration.) This action disables the ability to read and
write with large block sizes and allows the reading of tapes written prior to Solaris 2.4 with
large block sizes.

(Refer to mtio(7I) for a description of maximum record sizes.)

Bugs Tape devices that do not return a BUSY status during tape loading prevent user commands
from being held until the device is ready. The user must delay issuing any tape operations until
the tape device is ready. This is not a problem for tape devices supplied by Sun Microsystems.

Tape devices that do not report a blank check error at the end of recorded media may cause file
positioning operations to fail. Some tape drives, for example, mistakenly report media error
instead of blank check error.

838 man pages section 7: Device and Network Interfaces • Last Revised 22 Oct 2008
 streamio(7I)

Name streamio – STREAMS ioctl commands

Synopsis #include <sys/types.h>
#include <stropts.h>
#include <sys/conf.h>

int ioctl(int fildes, int command, ... /*arg*/);

Description STREAMS (see Intro(3)) ioctl commands are a subset of the ioctl(2) commands and
perform a variety of control functions on streams.

The fildes argument is an open file descriptor that refers to a stream. The command argument
determines the control function to be performed as described below. The arg argument
represents additional information that is needed by this command. The type of arg depends
upon the command, but it is generally an integer or a pointer to a command-specific data
structure. The command and arg arguments are interpreted by the STREAM head. Certain
combinations of these arguments may be passed to a module or driver in the stream.

Since these STREAMS commands are ioctls, they are subject to the errors described in
ioctl(2). In addition to those errors, the call will fail with errno set to EINVAL, without
processing a control function, if the STREAM referenced by fildes is linked below a
multiplexor, or if command is not a valid value for a stream.

Also, as described in ioctl(2), STREAMS modules and drivers can detect errors. In this case,
the module or driver sends an error message to the STREAM head containing an error value.
This causes subsequent calls to fail with errno set to this value.

ioctls The following ioctl commands, with error values indicated, are applicable to all STREAMS
files:
I_PUSH Pushes the module whose name is pointed to by arg onto the top of the
current stream, just below the STREAM head. If the STREAM is a pipe, the
module will be inserted between the stream heads of both ends of the pipe. It
then calls the open routine of the newly-pushed module. On failure, errno is
set to one of the following values:
EINVAL Invalid module name.
EFAULT arg points outside the allocated address space.
ENXIO Open routine of new module failed.
ENXIO Hangup received on fildes.
ENOTSUP Pushing a module is not supported on this stream.
I_POP Removes the module just below the STREAM head of the STREAM pointed
to by fildes. To remove a module from a pipe requires that the module was
pushed on the side it is being removed from. arg should be 0 in an I_POP
request. On failure, errno is set to one of the following values:

Device and Network Interfaces 839

streamio(7I)

EINVAL No module present in the stream.

ENXIO Hangup received on fildes.
EPERM Attempt to pop through an anchor by an unprivileged process.
ENOTSUP Removal is not supported.
I_ANCHOR Positions the stream anchor to be at the stream's module directly below the
stream head. Once this has been done, only a privileged process may pop
modules below the anchor on the stream. arg must be 0 in an I_ANCHOR
request. On failure, errno is set to the following value:
EINVAL Request to put an anchor on a pipe.
I_LOOK Retrieves the name of the module just below the stream head of the stream
pointed to by fildes, and places it in a null terminated character string
pointed at by arg. The buffer pointed to by arg should be at least FMNAMESZ+1
bytes long. This requires the declaration #include <sys/conf.h>. On
failure, errno is set to one of the following values:
EFAULT arg points outside the allocated address space.
EINVAL No module present in stream.
I_FLUSH This request flushes all input and/or output queues, depending on the value
of arg. Legal arg values are:
FLUSHR Flush read queues.
FLUSHW Flush write queues.
FLUSHRW Flush read and write queues.

If a pipe or FIFO does not have any modules pushed, the read queue of the
stream head on either end is flushed depending on the value of arg.

If FLUSHR is set and fildes is a pipe, the read queue for that end of the pipe is
flushed and the write queue for the other end is flushed. If fildes is a FIFO,
both queues are flushed.

If FLUSHW is set and fildes is a pipe and the other end of the pipe exists, the
read queue for the other end of the pipe is flushed and the write queue for
this end is flushed. If fildes is a FIFO, both queues of the FIFO are flushed.

If FLUSHRW is set, all read queues are flushed, that is, the read queue for the
FIFO and the read queue on both ends of the pipe are flushed.

Correct flush handling of a pipe or FIFO with modules pushed is achieved

via the pipemod module. This module should be the first module pushed
onto a pipe so that it is at the midpoint of the pipe itself.

840 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

On failure, errno is set to one of the following values:

ENOSR Unable to allocate buffers for flush message due to insufficient
stream memory resources.
EINVAL Invalid arg value.
ENXIO Hangup received on fildes.
I_FLUSHBAND Flushes a particular band of messages. arg points to a bandinfo structure
that has the following members:
unsigned char bi_pri;
int bi_flag;
The bi_flag field may be one of FLUSHR, FLUSHW, or FLUSHRW as described
earlier.
I_SETSIG Informs the stream head that the user wishes the kernel to issue the SIGPOLL
signal (see signal(3C)) when a particular event has occurred on the stream
associated with fildes. I_SETSIG supports an asynchronous processing
capability in streams. The value of arg is a bitmask that specifies the events
for which the user should be signaled. It is the bitwise OR of any
combination of the following constants:
S_INPUT Any message other than an M_PCPROTO has arrived on a
stream head read queue. This event is maintained for
compatibility with previous releases. This event is triggered
even if the message is of zero length.
S_RDNORM An ordinary (non-priority) message has arrived on a stream
head read queue. This event is triggered even if the message
is of zero length.
S_RDBAND A priority band message (band > 0) has arrived on a stream
head read queue. This event is triggered even if the message
is of zero length.
S_HIPRI A high priority message is present on the stream head read
queue. This event is triggered even if the message is of zero
length.
S_OUTPUT The write queue just below the stream head is no longer full.
This notifies the user that there is room on the queue for
sending (or writing) data downstream.
S_WRNORM This event is the same as S_OUTPUT.
S_WRBAND A priority band greater than 0 of a queue downstream exists
and is writable. This notifies the user that there is room on
the queue for sending (or writing) priority data downstream.

Device and Network Interfaces 841

streamio(7I)

S_MSG A STREAMS signal message that contains the SIGPOLL

signal has reached the front of the stream head read queue.
S_ERROR An M_ERROR message has reached the stream head.
S_HANGUP An M_HANGUP message has reached the stream head.
S_BANDURG When used in conjunction with S_RDBAND, SIGURG is
generated instead of SIGPOLL when a priority message
reaches the front of the stream head read queue.

A user process may choose to be signaled only of high priority messages by

setting the arg bitmask to the value S_HIPRI.

Processes that wish to receive SIGPOLL signals must explicitly register to

receive them using I_SETSIG. If several processes register to receive this
signal for the same event on the same stream, each process will be signaled
when the event occurs.

If the value of arg is zero, the calling process will be unregistered and will not
receive further SIGPOLL signals. On failure, errno is set to one of the
following values:
EINVAL arg value is invalid or arg is zero and process is not registered to
receive the SIGPOLL signal.
EAGAIN Allocation of a data structure to store the signal request failed.
I_GETSIG Returns the events for which the calling process is currently registered to be
sent a SIGPOLL signal. The events are returned as a bitmask pointed to by
arg, where the events are those specified in the description of I_SETSIG
above. On failure, errno is set to one of the following values:
EINVAL Process not registered to receive the SIGPOLL signal.
EFAULT arg points outside the allocated address space.
I_FIND Compares the names of all modules currently present in the stream to the
name pointed to by arg, and returns 1 if the named module is present in the
stream. It returns 0 if the named module is not present. On failure, errno is
set to one of the following values:
EFAULT arg points outside the allocated address space.
EINVAL arg does not contain a valid module name.
I_PEEK Allows a user to retrieve the information in the first message on the stream
head read queue without taking the message off the queue. I_PEEK is

842 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

analogous to getmsg(2) except that it does not remove the message from the
queue. arg points to a strpeek structure, which contains the following
members:
struct strbuf ctlbuf;
struct strbuf databuf;
long flags;
The maxlen field in the ctlbuf and databuf strbuf structures (see
getmsg(2)) must be set to the number of bytes of control information and/or
data information, respectively, to retrieve. flags may be set to RS_HIPRI or
0. If RS_HIPRI is set, I_PEEK will look for a high priority message on the
stream head read queue. Otherwise, I_PEEK will look for the first message on
the stream head read queue.

I_PEEK returns 1 if a message was retrieved, and returns 0 if no message was

found on the stream head read queue. It does not wait for a message to
arrive. On return, ctlbuf specifies information in the control buffer,
databuf specifies information in the data buffer, and flags contains the
value RS_HIPRI or 0. On failure, errno is set to the following value:
EFAULT arg points, or the buffer area specified in ctlbuf or databuf is,
outside the allocated address space.
EBADMSG Queued message to be read is not valid for I_PEEK.
EINVAL Illegal value for flags.
ENOSR Unable to allocate buffers to perform the I_PEEK due to
insufficient STREAMS memory resources.
I_SRDOPT Sets the read mode (see read(2)) using the value of the argument arg. Legal
arg values are:
RNORM Byte-stream mode, the default.
RMSGD Message-discard mode.
RMSGN Message-nondiscard mode.

In addition, the stream head's treatment of control messages may be

changed by setting the following flags in arg:
RPROTNORM Reject read() with EBADMSG if a control message is at
the front of the stream head read queue.
RPROTDAT Deliver the control portion of a message as data when a
user issues read(). This is the default behavior.
RPROTDIS Discard the control portion of a message, delivering any
data portion, when a user issues a read( ).

Device and Network Interfaces 843

streamio(7I)

On failure, errno is set to the following value:

EINVAL arg is not one of the above legal values, or arg is the bitwise
inclusive OR of RMSGD and RMSGN.
I_GRDOPT Returns the current read mode setting in an int pointed to by the argument
arg. Read modes are described in read( ). On failure, errno is set to the
following value:
EFAULT arg points outside the allocated address space.
I_NREAD Counts the number of data bytes in data blocks in the first message on the
stream head read queue, and places this value in the location pointed to by
arg. The return value for the command is the number of messages on the
stream head read queue. For example, if zero is returned in arg, but the
ioctl return value is greater than zero, this indicates that a zero-length
message is next on the queue. On failure, errno is set to the following value:
EFAULT arg points outside the allocated address space.
I_FDINSERT Creates a message from specified buffer(s), adds information about another
stream and sends the message downstream. The message contains a control
part and an optional data part. The data and control parts to be sent are
distinguished by placement in separate buffers, as described below.

The arg argument points to a strfdinsert structure, which contains the

following members:
struct strbuf ctlbuf;
struct strbuf databuf;
t_uscalar_t flags;
int fildes;
int offset;
The len member in the ctlbuf strbuf structure (see putmsg(2)) must be
set to the size of a t_uscalar_t plus the number of bytes of control
information to be sent with the message. The fildes member specifies the
file descriptor of the other stream, and the offset member, which must be
suitably aligned for use as a t_uscalar_t, specifies the offset from the start
of the control buffer where I_FDINSERT will store a t_uscalar_t whose
interpretation is specific to the stream end. The len member in the databuf
strbuf structure must be set to the number of bytes of data information to
be sent with the message, or to 0 if no data part is to be sent.

The flags member specifies the type of message to be created. A normal

message is created if flags is set to 0, and a high-priority message is created
if flags is set to RS_HIPRI. For non-priority messages, I_FDINSERT will
block if the stream write queue is full due to internal flow control

844 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

conditions. For priority messages, I_FDINSERT does not block on this

condition. For non-priority messages, I_FDINSERT does not block when the
write queue is full and O_NDELAY or O_NONBLOCK is set. Instead, it fails and
sets errno to EAGAIN.

I_FDINSERT also blocks, unless prevented by lack of internal resources,

waiting for the availability of message blocks in the stream, regardless of
priority or whether O_NDELAY or O_NONBLOCK has been specified. No partial
message is sent.

The ioctl() function with the I_FDINSERT command will fail if:
EAGAIN A non-priority message is specified, the O_NDELAY or
O_NONBLOCK flag is set, and the stream write queue is full due to
internal flow control conditions.
ENOSR Buffers can not be allocated for the message that is to be created.
EFAULT The arg argument points, or the buffer area specified in ctlbuf
or databuf is, outside the allocated address space.
EINVAL One of the following: The fildes member of the strfdinsert
structure is not a valid, open stream file descriptor; the size of a
t_uscalar_t plus offset is greater than the len member for
the buffer specified through ctlptr; the offset member does
not specify a properly-aligned location in the data buffer; or an
undefined value is stored in flags.
ENXIO Hangup received on the fildes argument of the ioctl call or
the fildes member of the strfdinsert structure.
ERANGE The len field for the buffer specified through databuf does not
fall within the range specified by the maximum and minimum
packet sizes of the topmost stream module; or the len member
for the buffer specified through databuf is larger than the
maximum configured size of the data part of a message; or the
len member for the buffer specified through ctlbuf is larger
than the maximum configured size of the control part of a
message.

I_FDINSERT can also fail if an error message was received by the stream head
of the stream corresponding to the fildes member of the strfdinsert
structure. In this case, errno will be set to the value in the message.
I_STR Constructs an internal STREAMS ioctl message from the data pointed to by
arg, and sends that message downstream.

Device and Network Interfaces 845

streamio(7I)

This mechanism is provided to send user ioctl requests to downstream

modules and drivers. It allows information to be sent with the ioctl, and
will return to the user any information sent upstream by the downstream
recipient. I_STR blocks until the system responds with either a positive or
negative acknowledgement message, or until the request times out after
some period of time. If the request times out, it fails with errno set to ETIME.

To send requests downstream, arg must point to a strioctl structure

which contains the following members:
int ic_cmd;
int ic_timout;
int ic_len;
char *ic_dp;
ic_cmd is the internal ioctl command intended for a downstream module
or driver and ic_timout is the number of seconds (-1 = infinite, 0 = use
default, >0 = as specified) an I_STR request will wait for acknowledgement
before timing out. ic_len is the number of bytes in the data argument and
ic_dp is a pointer to the data argument. The ic_len field has two uses: on
input, it contains the length of the data argument passed in, and on return
from the command, it contains the number of bytes being returned to the
user (the buffer pointed to by ic_dp should be large enough to contain the
maximum amount of data that any module or the driver in the stream can
return).

At most one I_STR can be active on a stream. Further I_STR calls will block
until the active I_STR completes via a positive or negative acknowlegment, a
timeout, or an error condition at the stream head. By setting the ic_timout
field to 0, the user is requesting STREAMS to provide the DEFAULT timeout.
The default timeout is specific to the STREAMS implementation and may
vary depending on which release of Solaris you are using. For Solaris 8 (and
earlier versions), the default timeout is fifteen seconds. The O_NDELAY and
O_NONBLOCK (see open(2)) flags have no effect on this call.

The stream head will convert the information pointed to by the strioctl
structure to an internal ioctl command message and send it downstream.
On failure, errno is set to one of the following values:
ENOSR Unable to allocate buffers for the ioctl message due to
insufficient STREAMS memory resources.
EFAULT Either arg points outside the allocated address space, or the
buffer area specified by ic_dp and ic_len (separately for data
sent and data returned) is outside the allocated address space.

846 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

EINVAL ic_len is less than 0 or ic_len is larger than the maximum

configured size of the data part of a message or ic_timout is less
than -1.
ENXIO Hangup received on fildes.
ETIME A downstream ioctl timed out before acknowledgement was
received.

An I_STR can also fail while waiting for an acknowledgement if a message

indicating an error or a hangup is received at the stream head. In addition,
an error code can be returned in the positive or negative acknowledgement
message, in the event the ioctl command sent downstream fails. For these
cases, I_STR will fail with errno set to the value in the message.
I_SWROPT Sets the write mode using the value of the argument arg. Legal bit settings
for arg are:
SNDZERO Send a zero-length message downstream when a write of 0
bytes occurs.

To not send a zero-length message when a write of 0 bytes occurs, this bit
must not be set in arg.

On failure, errno may be set to the following value:

EINVAL arg is not the above legal value.
I_GWROPT Returns the current write mode setting, as described above, in the int that is
pointed to by the argument arg.
I_SENDFD Requests the stream associated with fildes to send a message, containing a
file pointer, to the stream head at the other end of a stream pipe. The file
pointer corresponds to arg, which must be an open file descriptor.

I_SENDFD converts arg into the corresponding system file pointer. It

allocates a message block and inserts the file pointer in the block. The user id
and group id associated with the sending process are also inserted. This
message is placed directly on the read queue (see Intro(3)) of the stream
head at the other end of the stream pipe to which it is connected. On failure,
errno is set to one of the following values:
EAGAIN The sending stream is unable to allocate a message block to
contain the file pointer.
EAGAIN The read queue of the receiving stream head is full and cannot
accept the message sent by I_SENDFD.
EBADF arg is not a valid, open file descriptor.

Device and Network Interfaces 847

streamio(7I)

EINVAL fildes is not connected to a stream pipe.

ENXIO Hangup received on fildes.
I_RECVFD Retrieves the file descriptor associated with the message sent by an I_SENDFD
ioctl over a stream pipe. arg is a pointer to a data buffer large enough to
hold an strrecvfd data structure containing the following members:
int fd;
uid_t uid;
gid_t gid;
fd is an integer file descriptor. uid and gid are the user id and group id,
respectively, of the sending stream.

If O_NDELAY and O_NONBLOCK are clear (see open(2)), I_RECVFD will block
until a message is present at the stream head. If O_NDELAY or O_NONBLOCK is
set, I_RECVFD will fail with errno set to EAGAIN if no message is present at
the stream head.

If the message at the stream head is a message sent by an I_SENDFD, a new

user file descriptor is allocated for the file pointer contained in the message.
The new file descriptor is placed in the fd field of the strrecvfd structure.
The structure is copied into the user data buffer pointed to by arg. On
failure, errno is set to one of the following values:
EAGAIN A message is not present at the stream head read queue, and
the O_NDELAY or O_NONBLOCK flag is set.
EBADMSG The message at the stream head read queue is not a message
containing a passed file descriptor.
EFAULT arg points outside the allocated address space.
EMFILE NOFILES file descriptors are currently open.
ENXIO Hangup received on fildes.
EOVERFLOW uid or gid is too large to be stored in the structure pointed to
by arg.
I_LIST Allows the user to list all the module names on the stream, up to and
including the topmost driver name. If arg is NULL, the return value is the
number of modules, including the driver, that are on the stream pointed to
by fildes. This allows the user to allocate enough space for the module
names. If arg is non-null, it should point to an str_list structure that has
the following members:
int sl_nmods;
struct str_mlist *sl_modlist;

848 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

The str_mlist structure has the following member:

char l_name[FMNAMESZ+1];
The sl_nmods member indicates the number of entries the process has
allocated in the array. Upon return, the sl_modlist member of the
str_list structure contains the list of module names, and the number of
entries that have been filled into the sl_modlist array is found in the
sl_nmods member (the number includes the number of modules including
the driver). The return value from ioctl() is 0. The entries are filled in
starting at the top of the stream and continuing downstream until either the
end of the stream is reached, or the number of requested modules
(sl_nmods) is satisfied. On failure, errno may be set to one of the following
values:
EINVAL The sl_nmods member is less than 1.
EAGAIN Unable to allocate buffers
I_ATMARK Allows the user to see if the current message on the stream head read queue
is ‘‘marked'' by some module downstream. arg determines how the checking
is done when there may be multiple marked messages on the stream head
read queue. It may take the following values:
ANYMARK Check if the message is marked.
LASTMARK Check if the message is the last one marked on the queue.

The return value is 1 if the mark condition is satisfied and 0 otherwise. On

failure, errno is set to the following value:
EINVAL Invalid arg value.
I_CKBAND Check if the message of a given priority band exists on the stream head read
queue. This returns 1 if a message of a given priority exists, 0 if not, or −1 on
error. arg should be an integer containing the value of the priority band in
question. On failure, errno is set to the following value:
EINVAL Invalid arg value.
I_GETBAND Returns the priority band of the first message on the stream head read queue
in the integer referenced by arg. On failure, errno is set to the following
value:
ENODATA No message on the stream head read queue.
I_CANPUT Check if a certain band is writable. arg is set to the priority band in question.
The return value is 0 if the priority band arg is flow controlled, 1 if the band
is writable, or −1 on error. On failure, errno is set to the following value:
EINVAL Invalid arg value.

Device and Network Interfaces 849

streamio(7I)

I_SETCLTIME Allows the user to set the time the stream head will delay when a stream is
closing and there are data on the write queues. Before closing each module
and driver, the stream head will delay for the specified amount of time to
allow the data to drain. Note, however, that the module or driver may itself
delay in its close routine; this delay is independent of the stream head's delay
and is not settable. If, after the delay, data are still present, data will be
flushed. arg is a pointer to an integer containing the number of milliseconds
to delay, rounded up to the nearest legal value on the system. The default is
fifteen seconds. On failure, errno is set to the following value:
EINVAL Invalid arg value.
I_GETCLTIME Returns the close time delay in the integer pointed by arg.
I_SERROPT Sets the error mode using the value of the argument arg.

Normally stream head errors are persistent; once they are set due to an
M_ERROR or M_HANGUP, the error condition will remain until the stream is
closed. This option can be used to set the stream head into non-persistent
error mode i.e. once the error has been returned in response to a read(2),
getmsg(2), ioctl(2), write(2), or putmsg(2) call the error condition will be
cleared. The error mode can be controlled independently for read and write
side errors. Legal arg values are either none or one of:
RERRNORM Persistent read errors, the default.
RERRNONPERSIST Non-persistent read errors.

OR'ed with either none or one of:

WERRNORM Persistent write errors, the default.
WERRNONPERSIST Non-persistent write errors.

When no value is specified e.g. for the read side error

behavior then the behavior for that side will be left
unchanged.

On failure, errno is set to the following value:

EINVAL arg is not one of the above legal values.
I_GERROPT Returns the current error mode setting in an int pointed to by the argument
arg. Error modes are described above for I_SERROPT. On failure,errno is set
to the following value:
EFAULT arg points outside the allocated address space.

The following four commands are used for connecting and disconnecting multiplexed
STREAMS configurations.

850 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

I_LINK Connects two streams, where fildes is the file descriptor of the stream
connected to the multiplexing driver, and arg is the file descriptor of the
stream connected to another driver. The stream designated by arg gets
connected below the multiplexing driver. I_LINK requires the multiplexing
driver to send an acknowledgement message to the stream head regarding the
linking operation. This call returns a multiplexor ID number (an identifier
used to disconnect the multiplexor, see I_UNLINK) on success, and -1 on
failure. On failure, errno is set to one of the following values:
ENXIO Hangup received on fildes.
ETIME Time out before acknowledgement message was received at stream
head.
EAGAIN Temporarily unable to allocate storage to perform the I_LINK.
ENOSR Unable to allocate storage to perform the I_LINK due to
insufficient STREAMS memory resources.
EBADF arg is not a valid, open file descriptor.
EINVAL fildes stream does not support multiplexing.
EINVAL arg is not a stream, or is already linked under a multiplexor.
EINVAL The specified link operation would cause a ‘‘cycle'' in the resulting
configuration; that is, a driver would be linked into the
multiplexing configuration in more than one place.
EINVAL fildes is the file descriptor of a pipe or FIFO.
EINVAL Either the upper or lower stream has a major number >= the
maximum major number on the system.

An I_LINK can also fail while waiting for the multiplexing driver to
acknowledge the link request, if a message indicating an error or a hangup is
received at the stream head of fildes. In addition, an error code can be returned
in the positive or negative acknowledgement message. For these cases, I_LINK
will fail with errno set to the value in the message.
I_UNLINK Disconnects the two streams specified by fildes and arg. fildes is the file
descriptor of the stream connected to the multiplexing driver. arg is the
multiplexor ID number that was returned by the I_LINK. If arg is -1, then all
streams that were linked to fildes are disconnected. As in I_LINK, this
command requires the multiplexing driver to acknowledge the unlink. On
failure, errno is set to one of the following values:
ENXIO Hangup received on fildes.

Device and Network Interfaces 851

streamio(7I)

ETIME Time out before acknowledgement message was received at stream

head.
ENOSR Unable to allocate storage to perform the I_UNLINK due to
insufficient STREAMS memory resources.
EINVAL arg is an invalid multiplexor ID number or fildes is not the stream
on which the I_LINK that returned arg was performed.
EINVAL fildes is the file descriptor of a pipe or FIFO.

An I_UNLINK can also fail while waiting for the multiplexing driver to
acknowledge the link request, if a message indicating an error or a hangup is
received at the stream head of fildes. In addition, an error code can be returned
in the positive or negative acknowledgement message. For these cases,
I_UNLINK will fail with errno set to the value in the message.
I_PLINK Connects two streams, where fildes is the file descriptor of the stream
connected to the multiplexing driver, and arg is the file descriptor of the
stream connected to another driver. The stream designated by arg gets
connected via a persistent link below the multiplexing driver. I_PLINK
requires the multiplexing driver to send an acknowledgement message to the
stream head regarding the linking operation. This call creates a persistent link
that continues to exist even if the file descriptor fildes associated with the upper
stream to the multiplexing driver is closed. This call returns a multiplexor ID
number (an identifier that may be used to disconnect the multiplexor, see
I_PUNLINK) on success, and -1 on failure. On failure, errno is set to one of the
following values:
ENXIO Hangup received on fildes.
ETIME Time out before acknowledgement message was received at the
stream head.
EAGAIN Unable to allocate STREAMS storage to perform the I_PLINK.
EBADF arg is not a valid, open file descriptor.
EINVAL fildes does not support multiplexing.
EINVAL arg is not a stream or is already linked under a multiplexor.
EINVAL The specified link operation would cause a ‘‘cycle'' in the resulting
configuration; that is, if a driver would be linked into the
multiplexing configuration in more than one place.
EINVAL fildes is the file descriptor of a pipe or FIFO.

An I_PLINK can also fail while waiting for the multiplexing driver to
acknowledge the link request, if a message indicating an error on a hangup is

852 man pages section 7: Device and Network Interfaces • Last Revised 8 Apr 2009
 streamio(7I)

received at the stream head of fildes. In addition, an error code can be returned
in the positive or negative acknowledgement message. For these cases,
I_PLINK will fail with errno set to the value in the message.
I_PUNLINK Disconnects the two streams specified by fildes and arg that are connected with
a persistent link. fildes is the file descriptor of the stream connected to the
multiplexing driver. arg is the multiplexor ID number that was returned by
I_PLINK when a stream was linked below the multiplexing driver. If arg is
MUXID_ALL then all streams that are persistent links to fildes are disconnected.
As in I_PLINK, this command requires the multiplexing driver to
acknowledge the unlink. On failure, errno is set to one of the following values:
ENXIO Hangup received on fildes.
ETIME Time out before acknowledgement message was received at the
stream head.
EAGAIN Unable to allocate buffers for the acknowledgement message.
EINVAL Invalid multiplexor ID number.
EINVAL fildes is the file descriptor of a pipe or FIFO.

An I_PUNLINK can also fail while waiting for the multiplexing driver to
acknowledge the link request if a message indicating an error or a hangup is
received at the stream head of fildes. In addition, an error code can be returned
in the positive or negative acknowledgement message. For these cases,
I_PUNLINK will fail with errno set to the value in the message.

Return Values Unless specified otherwise above, the return value from ioctl() is 0 upon success and −1
upon failure, with errno set as indicated.

See Also Intro(3), close(2), fcntl(2), getmsg(2), ioctl(2), open(2), poll(2), putmsg(2), read(2),
write(2), signal(3C), signal.h(3HEAD),

STREAMS Programming Guide

Device and Network Interfaces 853

su(7D)

Name su – asynchronous serial port driver

Synopsis #include <fcntl.h>
#include <sys/termios.h>
open("/dev/tty[a-z]", _mode);
open("/dev/term[a-z]", _mode);
open("/dev/cua[a-z]", _mode);

Description The su module is a loadable STREAMS driver that provides basic support for standard
UARTS that use Intel-8250, National Semiconductor-16450/16550 hardware and Southbridge
1535D (16550 compatable) Super I/O hardware. The module also provides keyboard and
mouse I/O support for Sun machines using those same Intel, National Semiconductor and
Southbridge chipsets. The su driver provides basic asynchronous communication support for
serial ports. Both the serial devices and keyboard/mouse devices will have streams built with
appropriate modules pushed atop the su driver by means of either the autopush(1M) or
dacf.conf(4) facilities, depending on the OS revision and architecture in use.

The su module supports those termio(7I) device control functions specified by flags in the
c_cflag word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK, or INPCK flags
in the c_iflag word of the termios structure. All other termio(7I) functions must be
performed by STREAMS modules pushed atop the driver. When a device is opened, the
ldterm(7M) and ttcompat(7M) STREAMS modules are automatically pushed on top of the
stream, providing the standard termio(7I) interface.

The character-special devices /dev/ttya and /dev/ttyb are used to access the two standard
serial ports. The su module supports up to ten serial ports, including the standard ports. The
tty[a-z] devices have minor device numbers in the range 00-03, and may be assigned names
of the form /dev/ttyd_n_, where _n_ denotes the line to be accessed. These device names are
typically used to provide a logical access point for a _dial-in_ line that is used with a modem.

To allow a single tty line to be connected to a modem and used for incoming and outgoing
calls, a special feature is available that is controlled by the minor device number. By accessing
character-special devices with names of the form /dev/cua_n, it is possible to open a port
without the Carrier Detect signal being asserted, either through hardware or an equivalent
software mechanism. These devices are commonly known as _dial-out_ lines.

Application Once a /dev/cua_n_ line is opened, the corresponding tty, or ttyd line cannot be opened until
Programming the /dev/cua_n_ line is closed. A blocking open will wait until the /dev/cua_n_ line is closed
Interface (which will drop Data Terminal Ready, after which Carrier Detect will usually drop as
well) and carrier is detected again. A non-blocking open will return an error. If the
/dev/ttyd_n_ line has been opened successfully (usually only when carrier is recognized on
the modem), the corresponding /dev/cua_n_ line cannot be opened. This allows a modem to
be attached to a device, (for example, /dev/ttyd0, which is renamed from /dev/tty00) and
used for dial-in (by enabling the line for login in /etc/inittab) or dial-out (by tip(1) or
uucp(1C)) as /dev/cua0 when no one is logged in on the line.

854 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2004
 su(7D)

ioctls The standard set of termio ioctl() calls are supported by su.

Breaks can be generated by the TCSBRK, TIOCSBRK, and TIOCCBRK ioctl() calls.

The input and output line speeds may be set to any of the following baud rates: 0, 50, 75, 110,
134, 150, 200, 300, 600, 1200, 1800, 2400, 4800, 9600, 19200, 38400, 57600 or 115200. The
speeds cannot be set independently; for example, when the output speed is set, the input speed
is automatically set to the same speed.

When the su module is used to service the serial console port, it supports a BREAK condition
that allows the system to enter the debugger or the monitor. The BREAK condition is
generated by hardware and it is usually enabled by default.

A BREAK condition originating from erroneous electrical signals cannot be distinguished

from one deliberately sent by remote DCE. The Alternate Break sequence can be used as a
remedy against this. Due to a risk of incorrect sequence interpretation, SLIP and certain other
binary protocols should not be run over the serial console port when Alternate Break sequence
is in effect. Although PPP is a binary protocol, it is able to avoid these sequences using the
ACCM feature in RFC 1662. For Solaris PPP 4.0, you do this by adding the following line to
the /etc/ppp/options file (or other configuration files used for the connection; see pppd(1M)
for details):
asyncmap 0x00002000

By default, the Alternate Break sequence is a three character sequence: carriage return, tilde
and control-B (CR ~ CTRL-B), but may be changed by the driver. For more information on
breaking (entering the debugger or monitor), see kbd(1) and kb(7M).

Errors An open() will fail under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The dial-out device is being opened while the dial-in device is already open, or the
dial-in device is being opened with a no-delay open and the dial-out device is
already open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
Files /dev/cua/[a-z] dial-out tty lines
/dev/term/[a-z] dial-in tty lines
/dev/tty[a-z] binary compatibility package device names

Attributes See attributes(5) for descriptions of the following attributes:

Device and Network Interfaces 855

su(7D)

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also strconf(1), kbd(1), tip(1),uucp(1C), autopush(1M), kstat(1M), pppd(1M), ioctl(2),

open(2), termios(3C), dacf.conf(4), attributes(5), kb(7M), ldterm(7M), ttcompat(7M),
termio(7I)

Diagnostics The su driver keeps track of various warning and error conditions using kstat counters. The
output of the kstat su command provides kstat counters. The counters and their meaning
follow:
silo overflow The internal chip FIFO received more data than it could handle. This
indicates that the Solaris operating environment was not servicing
data interrupts fast enough possibly due to a system with too many
interrupts or a data line with a data rate that is too high.
ring buffer overflow The su module was unable to store data it removed from the chips
internal FIFO into a software buffer. The user process is not reading
data fast enough, possibly due to an overloaded system. If possible,
the application should enable flow control (either CTSRTS or
XONXOFF) to allow the driver to backpressure the remote system
when the local buffers fill up.

856 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2004
 sv(7D)

Name sv – Storage Volume system call device

Description The sv driver allows standard system call access (see Intro(2)) to a disk device to be redirected
into the StorageTek architecture software. This enables standard applications to use Sun
StorageTek Availability Suite components such as Point-in-Time Copy and Remote Mirror
software.
Files kernel/drv/sv SV control and administration driver.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWspsvu

Interface Stability Committed

See Also iiadm(1M)

Device and Network Interfaces 857

symhisl(7D)

Name symhisl – symhisl SCSI Host Bus Adapter Driver

Synopsis scsi@unit-address
Description The symhisl host bus adapter driver is a SCSA-compliant nexus driver that supports the LSI
Logic SYM53C895A, SYM53C1010-33, and SYM53C1010-66 SCSI controller chips.
The symhisl driver supports the standard functions provided by the SCSA interface,
including tagged and untagged queuing, Wide, Fast, Ultra, Ultra2, and Ultra3 SCSI, and auto
request sense. The symhisl driver does not support linked commands.
Configuration You configure the symhisl driver by defining properties in the symhisl.conf file. Properties
in the symhisl.conf file override the global SCSI settings. The driver supports the following
user-modifiable properties:

scsi-options
target<n>-scsi-options
scsi-reset-delay
scsi-watchdog-tick
scsi-initiator-id
symFlags
target<n>-scsi-options overrides the scsi-options property value for target<n>. <n>
can vary from hex 0 to f. symhisl supports the following scsi-options:
SCSI_OPTIONS_DR, SCSI_OPTIONS_SYNC, SCSI_OPTIONS_FAST,
SCSI_OPTIONS_ULTRA, SCSI_OPTIONS_ULTRA2, SCSI_OPTIONS_TAG, and
SCSI_OPTIONS_WIDE.
SCSI_OPTIONS_PARITY is supported for the scsi-options setting only and disables host
adapter parity checking.
After periodic interval scsi-watchdog-tick (seconds), the symhisl driver searches through
all current and disconnected commands for timeouts.
symFlags is a driver-specific bit-mask you can use to enable or disable driver properties.
bit 0 When set, the driver will not reset the SCSI bus at initialization. Certain CD-ROM,
tape, and other devices will not work properly when this bit is set. The default state
for this bit is cleared.
bit 1 When set, the driver will not export the DMI ioctl interface. Set this bit only if you
want to disable the ioctl interface for security reasons. The default state for this bit is
cleared.
bit 2 When set, the driver disables 64-bit addressing capability. When clear, the driver
enables 64-bit addressing capability. The default state for this bit is cleared.
bit 3 When set, the driver disables SCSI domain validation for all devices on any adapters
controlled by the driver.

858 man pages section 7: Device and Network Interfaces • Last Revised 20 July 2006
 symhisl(7D)

Refer to scsi_hba_attach(9F) for more information on driver configuration.

Examples Edit the file /kernel/drv/symhisl.conf and add the following line:
scsi-options=0x78;

This disables tagged queuing, Fast, Ultra, and Ultra2 SCSI and wide mode for all symhisl
instances.

The following example disables an option for one specific symhisl instance (refer to
driver.conf(4) and pci(4) for more details):
name="symhisl" parent="/pci@1f,4000"
unit-address="3"
target1-scsi-options=0x58
scsi-options=0x178 scsi-initiator-id=6;

Note that the initiator ID can only be changed for symhisl adapters that do not use the LSI
Logic Boot ROM Configuration Utility. For adapters that can use the LSI Logic Boot ROM
Configuration Utility, scsi-initiator-id has no effect.

The example above sets scsi-options for target 1 to 0x58 and all other targets on this SCSI
bus to 0x178.

The physical path name of the parent can be determined using the /devices tree or following
the link of the logical device name:
# ls –l /dev/rdsk/c0t0d0s0
lrwxrwxrwx 1 root root 45 May 16 10:08 /dev/rdsk/c0t0d0s0 ->
. . / . . /devices/pci@1f,4000/scsi@3/sd@0,0:a,raw

In this case, the parent is /pci@1f,4000 and the unit-address is the number bound to the scsi@3
node.

scsi-options specified per target ID have the highest precedence, followed by scsi-options
per device type. Global scsi-options (for all symhisl instances) per bus have the lowest
precedence.

The system must be rebooted for the specified scsi-options to take effect.

Driver Capabilities The target driver sets capabilities in the symhisl driver to enable some driver features. The
target driver can query and modify the following capabilities: disconnect, synchronous,
wide-xfer, tagged-qing, and auto-rqsense. All other capabilities are query only.

By default, tagged-qing capabilities are disabled, while disconnect, synchronous,

wide-xfer, auto-rqsense, and untagged-qing are enabled. These capabilities can only have
binary values (0 or 1).

Device and Network Interfaces 859

symhisl(7D)

The target driver must enable tagged-qing explicitly. The untagged-qing capability is always
enabled and its value cannot be modified.

If a conflict exists between the value of scsi-options and a capability, the value set in
scsi-options prevails. Only whom != 0 is supported in the scsi_ifsetcap(9F) call. Refer to
scsi_ifsetcap(9F) and scsi_ifgetcap(9F) for details.
Files /kernel/drv/symhisl ELF kernel module
/kernel/drv/symhisl.conf Configuration file

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture Limited to PCI-based systems with LSI Logic

SYM53C895A, SYM53C1010-33, and
SYM53C1010-66 SCSI I/O processors.

See Also prtconf(1M), driver.conf(4), pci(4), attributes(5), scsi_abort(9F),

scsi_hba_attach(9F), scsi_ifgetcap(9F), scsi_ifsetcap(9F), scsi_reset(9F),
scsi_sync_pkt(9F), scsi_transport(9F), scsi_device(9S), scsi_extended_sense(9S),
scsi_inquiry(9S), scsi_pkt(9S)

Writing Device Drivers

ANSI Small Computer System Interface-2 (SCSI-2),

LSI Logic Corporation, SYM53C896 PCI-SCSI I/O Processor

LSI Logic Corporation, SYM53C895A PCI-SCSI I/O Processor

LSI Logic Corporation, SYM53C1010 PCI-SCSI I/O Processor

Notes The symhisl SYM53C895A and SYM53C896 (SYM21002 and SYM22910) hardware and
software support Wide, Fast, SCSI Ultra, and Ultra2 synchronous speeds. SYM53C1010-33
and SYM53C1010-66 also support Ultra3 synchronous speeds. The maximum SCSI
bandwidth for Ultra2 transfers is 80 Mbytes/sec and 160 Mbytes/sec for Ultra3.

860 man pages section 7: Device and Network Interfaces • Last Revised 20 July 2006
 sysmsg(7D)

Name sysmsg – system message routing to console devices

Synopsis /dev/sysmsg

Description The file /dev/sysmsg routes output to a variable set of console devices. Writes to /dev/sysmsg
are always directed to the system console /dev/console, and are in addition directed to a set
of auxiliary console devices managed by consadm(1m).

Only root has permission to write to this device.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsr

Interface Stability Stable

See Also consadm(1m), syslogd(1M) , attributes(5), console(7D)

Device and Network Interfaces 861

systrace(7D)

Name systrace – DTrace system call tracing provider

Description The systrace driver implements the DTrace syscall dynamic tracing provider. The syscall
provider performs dynamic instrumentation to offer probes that fire whenever a thread enters
or returns from a kernel system call entry point.

The systrace driver is not a public interface and you access the instrumentation offered by
this provider through DTrace. Refer to the Solaris Dynamic Tracing Guide for a description of
the public documented interfaces available for the DTrace facility and the probes offered by
the syscall provider.

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWdtrp

Interface Stability Private

See Also dtrace(1M), attributes(5), dtrace(7D)

Solaris Dynamic Tracing Guide

862 man pages section 7: Device and Network Interfaces • Last Revised 4 Sep 2003
 tavor(7D)

Name tavor – InfiniHost MT23108/MT25208 InfiniBand (IB) Driver

Synopsis PCI pci15b3,5a44@pci-slot, pci15b3,5a45@pci-slot
PCI-E pci15b3,6278@pci-e-slot, pci15b3,6279@pci-e-slot

Description The tavor driver is an IB Architecture-compliant implementation of an HCA, which operates

on the Mellanox MT23108 InfiniBand ASSP and the Mellanox MT25208 InfiniBand ASSP.
These ASSP's support the link and physical layers of the InfiniBand specification, while the
ASSP and the driver support the transport layer.

The tavor driver interfaces with the InfiniBand Transport Framework (IBTF) and provides an
implementation of the Channel Interfaces that are defined by that framework. It also enables
management applications and agents to access the IB fabric.
Files /kernel/drv/tavor 32-bit ELF kernel module (x86 platform only).
/kernel/drv/amd64/tavor 64-bit ELF kernel module (x86 platform only).
/kernel/drv/sparcv9/tavor 64–bit ELF Kernel Module (SPARC platform only).
/kernel/drv/tavor.conf Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWtavor

See Also driver.conf(4), printers.conf(4), attributes(5)

Writing Device Drivers

Diagnostics In addition to being logged, the following messages may appear on the system console.
tavor : driver attached for maintenance mode only. There was a failure in the boot process
of the tavor ASSP and the only
function that can be performed is to
re-flash firmware on the ASSP.
driver failed to attach. The ASSP could not boot into either
operational (HCA) mode or into
maintenance mode. The device in
inoperable.
Unexpected port number in port state change event. A port state change event occurred,
but the port number in the message
does not exist on this HCA. This

Device and Network Interfaces 863

tavor(7D)

message also indicates the port

number that was in the port state
change.
Tavor driver successfully detached. The driver has been removed from the
system, and the HCA is no longer
available for transfer operations.
tavor"n": port "m" up. A port up asynchronous event has
occured. “n” represents the instance of
the Tavor device number, and “m”
represents the port number on the
Tavor device.
tavor"n": port "m" down. A port down aynchronous event has
occurred.
Tavor: <command name> command failed. A internal firmware command failed
to execute.

864 man pages section 7: Device and Network Interfaces • Last Revised 11 Feb 2005
 tcp(7P)

Name tcp, TCP – Internet Transmission Control Protocol

Synopsis #include <sys/socket.h>
#include <netinet/in.h>
s = socket(AF_INET, SOCK_STREAM, 0);
s = socket(AF_INET6, SOCK_STREAM, 0);
t = t_open("/dev/tcp", O_RDWR);
t = t_open("/dev/tcp6", O_RDWR);

Description TCP is the virtual circuit protocol of the Internet protocol family. It provides reliable,
flow-controlled, in order, two-way transmission of data. It is a byte-stream protocol layered
above the Internet Protocol (IP), or the Internet Protocol Version 6 (IPv6), the Internet
protocol family's internetwork datagram delivery protocol.

Programs can access TCP using the socket interface as a SOCK_STREAM socket type, or using the
Transport Level Interface (TLI) where it supports the connection-oriented (T_COTS_ORD)
service type.

TCP uses IP's host-level addressing and adds its own per-host collection of “port addresses.”
The endpoints of a TCP connection are identified by the combination of an IP or IPv6 address
and a TCP port number. Although other protocols, such as the User Datagram Protocol
(UDP), may use the same host and port address format, the port space of these protocols is
distinct. See inet(7P) and inet6(7P) for details on the common aspects of addressing in the
Internet protocol family.

Sockets utilizing TCP are either “active” or “passive.” Active sockets initiate connections to
passive sockets. Both types of sockets must have their local IP or IPv6 address and TCP port
number bound with the bind(3SOCKET) system call after the socket is created. By default,
TCP sockets are active. A passive socket is created by calling the listen(3SOCKET) system
call after binding the socket with bind(). This establishes a queueing parameter for the passive
socket. After this, connections to the passive socket can be received with the
accept(3SOCKET) system call. Active sockets use the connect(3SOCKET) call after binding
to initiate connections.

By using the special value INADDR_ANY with IP, or the unspecified address (all zeroes) with
IPv6, the local IP address can be left unspecified in the bind() call by either active or passive
TCP sockets. This feature is usually used if the local address is either unknown or irrelevant. If
left unspecified, the local IP or IPv6 address will be bound at connection time to the address of
the network interface used to service the connection.

Note that no two TCP sockets can be bound to the same port unless the bound IP addresses
are different. IPv4 INADDR_ANY and IPv6 unspecified addresses compare as equal to any IPv4
or IPv6 address. For example, if a socket is bound to INADDR_ANY or unspecified address and

Device and Network Interfaces 865

tcp(7P)

port X, no other socket can bind to port X, regardless of the binding address. This special
consideration of INADDR_ANY and unspecified address can be changed using the socket option
SO_REUSEADDR. If SO_REUSEADDR is set on a socket doing a bind, IPv4 INADDR_ANY and IPv6
unspecified address do not compare as equal to any IP address. This means that as long as the
two sockets are not both bound to INADDR_ANY/unspecified address or the same IP address, the
two sockets can be bound to the same port.

If an application does not want to allow another socket using the SO_REUSEADDR option to bind
to a port its socket is bound to, the application can set the socket level option SO_EXCLBIND on
a socket. The option values of 0 and 1 mean enabling and disabling the option respectively.
Once this option is enabled on a socket, no other socket can be bound to the same port.

Once a connection has been established, data can be exchanged using the read(2) and
write(2) system calls.

Under most circumstances, TCP sends data when it is presented. When outstanding data has
not yet been acknowledged, TCP gathers small amounts of output to be sent in a single packet
once an acknowledgement has been received. For a small number of clients, such as window
systems that send a stream of mouse events which receive no replies, this packetization may
cause significant delays. To circumvent this problem, TCP provides a socket-level boolean
option, TCP_NODELAY. TCP_NODELAY is defined in <netinet/tcp.h>, and is set with
setsockopt(3SOCKET) and tested with getsockopt(3SOCKET). The option level for the
setsockopt() call is the protocol number for TCP, available from
getprotobyname(3SOCKET).

For some applications, it may be desirable for TCP not to send out data unless a full TCP
segment can be sent. To enable this behavior, an application can use the TCP_CORK socket
option. When TCP_CORK is set with a non-zero value, TCP sends out a full TCP segment only.
When TCP_CORK is set to zero after it has been enabled, all buffered data is sent out (as
permitted by the peer's receive window and the current congestion window). TCP_CORK is
defined in <netinet/tcp.h>, and is set with setsockopt(3SOCKET) and tested with
getsockopt(3SOCKET). The option level for the setsockopt() call is the protocol number
for TCP, available from getprotobyname(3SOCKET).

The SO_RCVBUF socket level option can be used to control the window that TCP advertises
to the peer. IP level options may also be used with TCP. See ip(7P) and ip6(7P).

Another socket level option, SO_RCVBUF, can be used to control the window that TCP
advertises to the peer. IP level options may also be used with TCP. See ip(7P) and ip6(7P).

TCP provides an urgent data mechanism, which may be invoked using the out-of-band
provisions of send(3SOCKET). The caller may mark one byte as “urgent” with the MSG_OOB
flag to send(3SOCKET). This sets an “urgent pointer” pointing to this byte in the TCP stream.
The receiver on the other side of the stream is notified of the urgent data by a SIGURG signal.
The SIOCATMARK ioctl(2) request returns a value indicating whether the stream is at the

866 man pages section 7: Device and Network Interfaces • Last Revised 30 June 2006
 tcp(7P)

urgent mark. Because the system never returns data across the urgent mark in a single read(2)
call, it is possible to advance to the urgent data in a simple loop which reads data, testing the
socket with the SIOCATMARK ioctl() request, until it reaches the mark.

Incoming connection requests that include an IP source route option are noted, and the
reverse source route is used in responding.

A checksum over all data helps TCP implement reliability. Using a window-based flow control
mechanism that makes use of positive acknowledgements, sequence numbers, and a
retransmission strategy, TCP can usually recover when datagrams are damaged, delayed,
duplicated or delivered out of order by the underlying communication medium.

If the local TCP receives no acknowledgements from its peer for a period of time, (for
example, if the remote machine crashes), the connection is closed and an error is returned.

TCP follows the congestion control algorithm described in RFC 2581, and also supports the
initial congestion window (cwnd) changes in RFC 3390. The initial cwnd calculation can be
overridden by the socket option TCP_INIT_CWND. An application can use this option to set
the initial cwnd to a specified number of TCP segments. This applies to the cases when the
connection first starts and restarts after an idle period. The process must have the
PRIV_SYS_NET_CONFIG privilege if it wants to specify a number greater than that
calculated by RFC 3390.

SunOS supports TCP Extensions for High Performance (RFC 1323) which includes the
window scale and time stamp options, and Protection Against Wrap Around Sequence
Numbers (PAWS). SunOS also supports Selective Acknowledgment (SACK) capabilities (RFC
2018) and Explicit Congestion Notification (ECN) mechanism (RFC 3168).

Turn on the window scale option in one of the following ways:

■ An application can set SO_SNDBUF or SO_RCVBUF size in the setsockopt() option to be
larger than 64K. This must be done before the program calls listen() or connect(),
because the window scale option is negotiated when the connection is established. Once
the connection has been made, it is too late to increase the send or receive window beyond
the default TCP limit of 64K.
■ For all applications, use ndd(1M) to modify the configuration parameter
tcp_wscale_always. If tcp_wscale_always is set to 1, the window scale option will always
be set when connecting to a remote system. If tcp_wscale_always is 0, the window scale
option will be set only if the user has requested a send or receive window larger than 64K.
The default value of tcp_wscale_always is 1.
■ Regardless of the value of tcp_wscale_always, the window scale option will always be
included in a connect acknowledgement if the connecting system has used the option.

Turn on SACK capabilities in the following way:

Device and Network Interfaces 867

tcp(7P)

■ Use ndd to modify the configuration parameter tcp_sack_permitted. If

tcp_sack_permitted is set to 0, TCP will not accept SACK or send out SACK information. If
tcp_sack_permitted is set to 1, TCP will not initiate a connection with SACK permitted
option in the SYN segment, but will respond with SACK permitted option in the SYN|ACK
segment if an incoming connection request has the SACK permitted option. This means
that TCP will only accept SACK information if the other side of the connection also accepts
SACK information. If tcp_sack_permitted is set to 2, it will both initiate and accept
connections with SACK information. The default for tcp_sack_permitted is 2 (active
enabled).

Turn on TCP ECN mechanism in the following way:

■ Use ndd to modify the configuration parameter tcp_ecn_permitted. If
tcp_ecn_permitted is set to 0, TCP will not negotiate with a peer that supports ECN
mechanism. If tcp_ecn_permitted is set to 1 when initiating a connection, TCP will not
tell a peer that it supports ECN mechanism. However, it will tell a peer that it supports
ECN mechanism when accepting a new incoming connection request if the peer indicates
that it supports ECN mechanism in the SYN segment. If tcp_ecn_permitted is set to 2, in
addition to negotiating with a peer on ECN mechanism when accepting connections, TCP
will indicate in the outgoing SYN segment that it supports ECN mechanism when TCP
makes active outgoing connections. The default for tcp_ecn_permitted is 1.

Turn on the time stamp option in the following way:

■ Use ndd to modify the configuration parameter tcp_tstamp_always. If
tcp_tstamp_always is 1, the time stamp option will always be set when connecting to a
remote machine. If tcp_tstamp_always is 0, the timestamp option will not be set when
connecting to a remote system. The default for tcp_tstamp_always is 0.
■ Regardless of the value of tcp_tstamp_always, the time stamp option will always be
included in a connect acknowledgement (and all succeeding packets) if the connecting
system has used the time stamp option.

Use the following procedure to turn on the time stamp option only when the window scale
option is in effect:
■ Use ndd to modify the configuration parameter tcp_tstamp_if_wscale. Setting
tcp_tstamp_if_wscale to 1 will cause the time stamp option to be set when connecting to
a remote system, if the window scale option has been set. If tcp_tstamp_if_wscale is 0,
the time stamp option will not be set when connecting to a remote system. The default for
tcp_tstamp_if_wscale is 1.

Protection Against Wrap Around Sequence Numbers (PAWS) is always used when the time
stamp option is set.

SunOS also supports multiple methods of generating initial sequence numbers. One of these
methods is the improved technique suggested in RFC 1948. We HIGHLY recommend that you

868 man pages section 7: Device and Network Interfaces • Last Revised 30 June 2006
 tcp(7P)

set sequence number generation parameters as close to boot time as possible. This prevents
sequence number problems on connections that use the same connection-ID as ones that used
a different sequence number generation. The svc:/network/initial:default service
configures the initial sequence number generation. The service reads the value contained in
the configuration file /etc/default/inetinit to determine which method to use.

The /etc/default/inetinit file is an unstable interface, and may change in future releases.

TCP may be configured to report some information on connections that terminate by means
of an RST packet. By default, no logging is done. If the ndd(1M) parameter tcp_trace is set to 1,
then trace data is collected for all new connections established after that time.

The trace data consists of the TCP headers and IP source and destination addresses of the last
few packets sent in each direction before RST occurred. Those packets are logged in a series of
strlog(9F) calls. This trace facility has a very low overhead, and so is superior to such utilities
as snoop(1M) for non-intrusive debugging for connections terminating by means of an RST.

SunOS supports the keep-alive mechanism described in RFC 1122. It is enabled using the
socket option SO_KEEPALIVE. When enabled, the first keep-alive probe is sent out after a
TCP is idle for two hours If the peer does not respond to the probe within eight minutes, the
TCP connection is aborted. You can alter the interval for sending out the first probe using the
socket option TCP_KEEPALIVE_THRESHOLD. The option value is an unsigned integer in
milliseconds. The system default is controlled by the TCP ndd parameter
tcp_keepalive_interval. The minimum value is ten seconds. The maximum is ten days, while
the default is two hours. If you receive no response to the probe, you can use the
TCP_KEEPALIVE_ABORT_THRESHOLD socket option to change the time threshold for
aborting a TCP connection. The option value is an unsigned integer in milliseconds. The value
zero indicates that TCP should never time out and abort the connection when probing. The
system default is controlled by the TCP ndd parameter tcp_keepalive_abort_interval. The
default is eight minutes.

See Also svcs(1), ndd(1M), ioctl(2), read(2), svcadm(1M), write(2), accept(3SOCKET),

bind(3SOCKET), connect(3SOCKET), getprotobyname(3SOCKET),
getsockopt(3SOCKET), listen(3SOCKET), send(3SOCKET), smf(5), inet(7P), inet6(7P),
ip(7P), ip6(7P)

Ramakrishnan, K., Floyd, S., Black, D., RFC 3168, The Addition of Explicit Congestion
Notification (ECN) to IP, September 2001.

Mathias, M. and Hahdavi, J. Pittsburgh Supercomputing Center; Ford, S. Lawrence Berkeley

National Laboratory; Romanow, A. Sun Microsystems, Inc. RFC 2018, TCP Selective
Acknowledgement Options, October 1996.

Bellovin, S., RFC 1948, Defending Against Sequence Number Attacks, May 1996.

Device and Network Interfaces 869

tcp(7P)

Jacobson, V., Braden, R., and Borman, D., RFC 1323, TCP Extensions for High Performance,
May 1992.

Postel, Jon, RFC 793, Transmission Control Protocol - DARPA Internet Program Protocol
Specification, Network Information Center, SRI International, Menlo Park, CA., September
1981.

Diagnostics A socket operation may fail if:

EISCONN A connect() operation was attempted on a socket on which a connect()
operation had already been performed.
ETIMEDOUT A connection was dropped due to excessive retransmissions.
ECONNRESET The remote peer forced the connection to be closed (usually because the
remote machine has lost state information about the connection due to a
crash).
ECONNREFUSED The remote peer actively refused connection establishment (usually
because no process is listening to the port).
EADDRINUSE A bind() operation was attempted on a socket with a network
address/port pair that has already been bound to another socket.
EADDRNOTAVAIL A bind() operation was attempted on a socket with a network address for
which no network interface exists.
EACCES A bind() operation was attempted with a “reserved” port number and the
effective user ID of the process was not the privileged user.
ENOBUFS The system ran out of memory for internal data structures.

Notes The tcp service is managed by the service management facility, smf(5), under the service
identifier:
svc:/network/initial:default

Administrative actions on this service, such as enabling, disabling, or requesting restart, can
be performed using svcadm(1M). The service's status can be queried using the svcs(1)
command.

870 man pages section 7: Device and Network Interfaces • Last Revised 30 June 2006
 termio(7I)

Name termio – general terminal interface

Synopsis #include <termio.h>

ioctl(int fildes, int request, struct termio *arg);

ioctl(int fildes, int request, int arg);
#include <termios.h>

ioctl(int fildes, int request, struct termios *arg);

Description This release supports a general interface for asynchronous communications ports that is
hardware-independent. The user interface to this functionality is using function calls (the
preferred interface) described in termios(3C) or ioctl commands described in this section.
This section also discusses the common features of the terminal subsystem which are relevant
with both user interfaces.

When a terminal file is opened, it normally causes the process to wait until a connection is
established. In practice, user programs seldom open terminal files; they are opened by the
system and become a user's standard input, output, and error files. The first terminal file
opened by the session leader that is not already associated with a session becomes the
controlling terminal for that session. The controlling terminal plays a special role in handling
quit and interrupt signals, as discussed below. The controlling terminal is inherited by a child
process during a fork(2). A process can break this association by changing its session using
setsid() (see setsid(2)).

A terminal associated with one of these files ordinarily operates in full-duplex mode.
Characters may be typed at any time, even while output is occurring, and are only lost when
the character input buffers of the system become completely full, which is rare. For example,
the number of characters in the line discipline buffer may exceed {MAX_CANON} and IMAXBEL
(see below) is not set, or the user may accumulate { MAX_INPUT} number of input characters
that have not yet been read by some program. When the input limit is reached, all the
characters saved in the buffer up to that point are thrown away without notice.

Session Management A control terminal will distinguish one of the process groups in the session associated with it
(Job Control) to be the foreground process group. All other process groups in the session are designated as
background process groups. This foreground process group plays a special role in handling
signal-generating input characters, as discussed below. By default, when a controlling
terminal is allocated, the controlling process's process group is assigned as foreground process
group.

Background process groups in the controlling process's session are subject to a job control line
discipline when they attempt to access their controlling terminal. Process groups can be sent
signals that will cause them to stop, unless they have made other arrangements. An exception
is made for members of orphaned process groups.

Device and Network Interfaces 871

termio(7I)

An orphaned process group is one where the process group (see getpgid(2)) has no members
with a parent in a different process group but sharing the same controlling terminal. When a
member of an orphaned process group attempts to access its controlling terminal, EIO is
returned because there would be no way to restart the process if it were stopped on one of
these signals.

If a member of a background process group attempts to read its controlling terminal, its
process group will be sent a SIGTTIN signal, which will normally cause the members of that
process group to stop. If, however, the process is ignoring or holding SIGTTIN, or is a member
of an orphaned process group, the read will fail with errno set to EIO, and no signal is sent.

If a member of a background process group attempts to write its controlling terminal and the
TOSTOP bit is set in the c_lflag field, its process group is sent a SIGTTOU signal, which will
normally cause the members of that process group to stop. If, however, the process is ignoring
or holding SIGTTOU, the write will succeed. If the process is not ignoring or holding SIGTTOU
and is a member of an orphaned process group, the write will fail with errno set to EIO, and no
signal will be sent.

If TOSTOP is set and a member of a background process group attempts to ioctl its controlling
terminal, and that ioctl will modify terminal parameters (for example, TCSETA, TCSETAW,
TCSETAF, or TIOCSPGRP), its process group will be sent a SIGTTOU signal, which will normally
cause the members of that process group to stop. If, however, the process is ignoring or
holding SIGTTOU, the ioctl will succeed. If the process is not ignoring or holding SIGTTOU and
is a member of an orphaned process group, the write will fail with errno set to EIO, and no
signal will be sent.

Canonical Mode Input Normally, terminal input is processed in units of lines. A line is delimited by a newline (ASCII
Processing LF) character, an end-of-file (ASCII EOT) character, or an end-of-line character. This means
that a program attempting to read will block until an entire line has been typed. Also, no
matter how many characters are requested in the read call, at most one line will be returned. It
is not necessary, however, to read a whole line at once; any number of characters may be
requested in a read, even one, without losing information.

During input, erase and kill processing is normally done. The ERASE character (by default,
the character DEL) erases the last character typed. The WERASE character (the character
Control-w) erases the last “word” typed in the current input line (but not any preceding
spaces or tabs). A “word” is defined as a sequence of non-blank characters, with tabs counted
as blanks. Neither ERASE nor WERASE will erase beyond the beginning of the line. The KILL
character (by default, the character NAK) kills (deletes) the entire input line, and optionally
outputs a newline character. All these characters operate on a key stroke basis, independent of
any backspacing or tabbing that may have been done. The REPRINT character (the character
Control-r) prints a newline followed by all characters that have not been read. Reprinting also
occurs automatically if characters that would normally be erased from the screen are fouled by
program output. The characters are reprinted as if they were being echoed; consequencely, if
ECHO is not set, they are not printed.

872 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

The ERASE and KILL characters may be entered literally by preceding them with the escape
character. In this case, the escape character is not read. The erase and kill characters may be
changed.

Non-canonical Mode In non-canonical mode input processing, input characters are not assembled into lines, and
Input Processing erase and kill processing does not occur. The MIN and TIME values are used to determine
how to process the characters received.

MIN represents the minimum number of characters that should be received when the read is
satisfied (that is, when the characters are returned to the user). TIME is a timer of 0.10-second
granularity that is used to timeout bursty and short-term data transmissions. The four
possible values for MIN and TIME and their interactions are described below.
Case A: MIN > 0, TIME > 0 In this case, TIME serves as an intercharacter timer and is
activated after the first character is received. Since it is an
intercharacter timer, it is reset after a character is received.
The interaction between MIN and TIME is as follows: as soon
as one character is received, the intercharacter timer is started.
If MIN characters are received before the intercharacter timer
expires (note that the timer is reset upon receipt of each
character), the read is satisfied. If the timer expires before
MIN characters are received, the characters received to that
point are returned to the user. Note that if TIME expires, at
least one character will be returned because the timer would
not have been enabled unless a character was received. In this
case (MIN > 0, TIME > 0), the read sleeps until the MIN and
TIME mechanisms are activated by the receipt of the first
character. If the number of characters read is less than the
number of characters available, the timer is not reactivated
and the subsequent read is satisfied immediately.
Case B: MIN > 0, TIME = 0 In this case, since the value of TIME is zero, the timer plays no
role and only MIN is significant. A pending read is not
satisfied until MIN characters are received (the pending read
sleeps until MIN characters are received). A program that uses
this case to read record based terminal I/O may block
indefinitely in the read operation.
Case C: MIN = 0, TIME > 0 In this case, since MIN = 0, TIME no longer represents an
intercharacter timer: it now serves as a read timer that is
activated as soon as a read is done. A read is satisfied as soon
as a single character is received or the read timer expires. Note
that, in this case, if the timer expires, no character is returned.
If the timer does not expire, the only way the read can be
satisfied is if a character is received. In this case, the read will

Device and Network Interfaces 873

termio(7I)

not block indefinitely waiting for a character; if no character is

received within TIME *.10 seconds after the read is initiated,
the read returns with zero characters.
Case D: MIN = 0, TIME = 0 In this case, return is immediate. The minimum of either the
number of characters requested or the number of characters
currently available is returned without waiting for more
characters to be input.

Comparing Different Some points to note about MIN and TIME :

Cases of MIN, TIME
Interaction ■ In the following explanations, note that the interactions of MIN and TIME are not
symmetric. For example, when MIN > 0 and TIME = 0, TIME has no effect. However, in
the opposite case, where MIN = 0 and TIME > 0, both MIN and TIME play a role in that
MIN is satisfied with the receipt of a single character.
■ Also note that in case A (MIN > 0, TIME > 0), TIME represents an intercharacter timer,
whereas in case C ( MIN = 0, TIME > 0), TIME represents a read timer.

These two points highlight the dual purpose of the MIN/TIME feature. Cases A and B, where
MIN > 0, exist to handle burst mode activity (for example, file transfer programs), where a
program would like to process at least MIN characters at a time. In case A, the intercharacter
timer is activated by a user as a safety measure; in case B, the timer is turned off.

Cases C and D exist to handle single character, timed transfers. These cases are readily
adaptable to screen-based applications that need to know if a character is present in the input
queue before refreshing the screen. In case C, the read is timed, whereas in case D, it is not.

Another important note is that MIN is always just a minimum. It does not denote a record
length. For example, if a program does a read of 20 bytes, MIN is 10, and 25 characters are
present, then 20 characters will be returned to the user.

Writing Characters When one or more characters are written, they are transmitted to the terminal as soon as
previously written characters have finished typing. Input characters are echoed as they are
typed if echoing has been enabled. If a process produces characters more rapidly than they can
be typed, it will be suspended when its output queue exceeds some limit. When the queue is
drained down to some threshold, the program is resumed.

Special Characters Certain characters have special functions on input. These functions and their default
character values are summarized as follows:
INTR (Control-c or ASCII ETX) generates a SIGINT signal. SIGINT is sent to all
foreground processes associated with the controlling terminal. Normally, each
such process is forced to terminate, but arrangements may be made either to
ignore the signal or to receive a trap to an agreed upon location. (See
signal.h(3HEAD)).

874 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

QUIT (Control-| or ASCII FS) generates a SIGQUIT signal. Its treatment is identical
to the interrupt signal except that, unless a receiving process has made other
arrangements, it will not only be terminated but a core image file (called core)
will be created in the current working directory.
ERASE (DEL) erases the preceding character. It does not erase beyond the start of a
line, as delimited by a NL, EOF, EOL, or EOL2 character.
WERASE (Control-w or ASCII ETX) erases the preceding “word”. It does not erase
beyond the start of a line, as delimited by a NL, EOF, EOL, or EOL2 character.
KILL (Control-u or ASCII NAK) deletes the entire line, as delimited by a NL, EOF,
EOL, or EOL2 character.
REPRINT (Control-r or ASCII DC2) reprints all characters, preceded by a newline, that
have not been read.
EOF (Control-d or ASCII EOT) may be used to generate an end-of-file from a
terminal. When received, all the characters waiting to be read are immediately
passed to the program, without waiting for a newline, and the EOF is discarded.
Thus, if no characters are waiting (that is, the EOF occurred at the beginning of
a line) zero characters are passed back, which is the standard end-of-file
indication. Unless escaped, the EOF character is not echoed. Because EOT is
the default EOF character, this prevents terminals that respond to EOT from
hanging up.
NL (ASCII LF) is the normal line delimiter. It cannot be changed or escaped.
EOL (ASCII NULL) is an additional line delimiter, like NL . It is not normally used.
EOL2 is another additional line delimiter.
SWTCH (Control-z or ASCII EM) Header file symbols related to this special character
are present for compatibility purposes only and the kernel takes no special
action on matching SWTCH (except to discard the character).
SUSP (Control-z or ASCII SUB) generates a SIGTSTP signal. SIGTSTP stops all
processes in the foreground process group for that terminal.
DSUSP (Control-y or ASCII EM). It generates a SIGTSTP signal as SUSP does, but the
signal is sent when a process in the foreground process group attempts to read
the DSUSP character, rather than when it is typed.
STOP (Control-s or ASCII DC3) can be used to suspend output temporarily. It is
useful with CRT terminals to prevent output from disappearing before it can be
read. While output is suspended, STOP characters are ignored and not read.
START (Control-q or ASCII DC1) is used to resume output. Output has been
suspended by a STOP character. While output is not suspended, START
characters are ignored and not read.

Device and Network Interfaces 875

termio(7I)

DISCARD (Control-o or ASCII SI) causes subsequent output to be discarded. Output is

discarded until another DISCARD character is typed, more input arrives, or
the condition is cleared by a program.
LNEXT (Control-v or ASCII SYN) causes the special meaning of the next character to
be ignored. This works for all the special characters mentioned above. It allows
characters to be input that would otherwise be interpreted by the system (for
example KILL, QUIT). The character values for INTR, QUIT, ERASE,
WERASE, KILL, REPRINT, EOF, EOL, EOL2, SWTCH, SUSP, DSUSP, STOP,
START, DISCARD, and LNEXT may be changed to suit individual tastes. If the
value of a special control character is _POSIX_VDISABLE (0), the function of
that special control character is disabled. The ERASE, KILL, and EOF
characters may be escaped by a preceding backslash (\) character, in which case
no special function is done. Any of the special characters may be preceded by
the LNEXT character, in which case no special function is done.

Modem Disconnect When a modem disconnect is detected, a SIGHUP signal is sent to the terminal's controlling
process. Unless other arrangements have been made, these signals cause the process to
terminate. If SIGHUP is ignored or caught, any subsequent read returns with an end-of-file
indication until the terminal is closed.

If the controlling process is not in the foreground process group of the terminal, a SIGTSTP is
sent to the terminal's foreground process group. Unless other arrangements have been made,
these signals cause the processes to stop.

Processes in background process groups that attempt to access the controlling terminal after
modem disconnect while the terminal is still allocated to the session will receive appropriate
SIGTTOU and SIGTTIN signals. Unless other arrangements have been made, this signal causes
the processes to stop.

The controlling terminal will remain in this state until it is reinitialized with a successful open
by the controlling process, or deallocated by the controlling process.

Terminal Parameters The parameters that control the behavior of devices and modules providing the termios
interface are specified by the termios structure defined by termios.h. Several ioctl(2)
system calls that fetch or change these parameters use this structure that contains the
following members:
tcflag_t c_iflag; /* input modes */
tcflag_t c_oflag; /* output modes */
tcflag_t c_cflag; /* control modes */
tcflag_t c_lflag; /* local modes */
cc_t c_cc[NCCS]; /* control chars */

876 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

The special control characters are defined by the array c_cc. The symbolic name NCCS is the
size of the Control-character array and is also defined by <termios.h>. The relative positions,
subscript names, and typical default values for each function are as follows:

Relative Position Subscript Name Typical Default Value

0 VINTR ETX

1 VQUIT FS

2 VERASE DEL

3 VKILL NAK

4 VEOF EOT

5 VEOL NUL

6 VEOL2 NUL

7 VWSTCH NUL

8 VSTART NUL

9 VSTOP DC3

10 VSUSP SUB

11 VDSUSP EM

12 VREPRINT DC2

13 VDISCARD SI

14 VWERASE ETB

15 VLNEXT SYN

16-19 Reserved

Input Modes The c_iflag field describes the basic terminal input control:

IGNBRK Ignore break condition.

BRKINT Signal interrupt on break.
IGNPAR Ignore characters with parity errors.
PARMRK Mark parity errors.
INPCK Enable input parity check.
ISTRIP Strip character.
INLCR Map NL to CR on input.

Device and Network Interfaces 877

termio(7I)

IGNCR Ignore CR.

ICRNL Map CR to NL on input.
IUCLC Map upper-case to lower-case on input.
IXON Enable start/stop output control.
IXANY Enable any character to restart output.
IXOFF Enable start/stop input control.
IMAXBEL Echo BEL on input line too long.

If IGNBRK is set, a break condition (a character framing error with data all zeros) detected on
input is ignored, that is, not put on the input queue and therefore not read by any process. If
IGNBRK is not set and BRKINT is set, the break condition shall flush the input and output queues
and if the terminal is the controlling terminal of a foreground process group, the break
condition generates a single SIGINT signal to that foreground process group. If neither IGNBRK
nor BRKINT is set, a break condition is read as a single '\0' (ASCII NULL) character, or if
PARMRK is set, as '\377', '\0', c, where '\377' is a single character with value 377 octal (0xff hex,
255 decimal), '\0' is a single character with value 0, and c is the errored character received.

If IGNPAR is set, a byte with framing or parity errors (other than break) is ignored.

If PARMRK is set, and IGNPAR is not set, a byte with a framing or parity error (other than break)
is given to the application as the three-character sequence: '\377', '\0', c, where '\377' is a single
character with value 377 octal (0xff hex, 255 decimal), '\0' is a single character with value 0, and
c is the errored character received. To avoid ambiguity in this case, if ISTRIP is not set, a valid
character of '\377' is given to the application as ‘\377.' If neither IGNPAR nor PARMRK is set, a
framing or parity error (other than break) is given to the application as a single '\0' (ASCII
NULL) character.

If INPCK is set, input parity checking is enabled. If INPCK is not set, input parity checking is
disabled. This allows output parity generation without input parity errors. Note that whether
input parity checking is enabled or disabled is independent of whether parity detection is
enabled or disabled. If parity detection is enabled but input parity checking is disabled, the
hardware to which the terminal is connected will recognize the parity bit, but the terminal
special file will not check whether this is set correctly or not.

If ISTRIP is set, valid input characters are first stripped to seven bits, otherwise all eight bits are
processed.

If INLCR is set, a received NL character is translated into a CR character. If IGNCR is set, a

received CR character is ignored (not read). Otherwise, if ICRNL is set, a received CR character
is translated into a NL character.

878 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

If IUCLC is set, a received upper case, alphabetic character is translated into the corresponding
lower case character.

If IXON is set, start/stop output control is enabled. A received STOP character suspends output
and a received START character restarts output. The STOP and START characters will not be
read, but will merely perform flow control functions. If IXANY is set, any input character
restarts output that has been suspended.

If IXOFF is set, the system transmits a STOP character when the input queue is nearly full, and
a START character when enough input has been read so that the input queue is nearly empty
again.

If IMAXBEL is set, the ASCII BEL character is echoed if the input stream overflows. Further
input is not stored, but any input already present in the input stream is not disturbed. If
IMAXBEL is not set, no BEL character is echoed, and all input present in the input queue is
discarded if the input stream overflows.

Output Modes The c_oflag field specifies the system treatment of output:

OPOST Post-process output.

OLCUC Map lower case to upper on output.
ONLCR Map NL to CR-NL on output.
OCRNL Map CR to NL on output.
ONOCR No CR output at column 0.
ONLRET NL performs CR function.
OFILL Use fill characters for delay.
OFDEL Fill is DEL, else NULL.
NLDLY Select newline delays:

NL0
NL1
CRDLY Select carriage-return delays:

CR0
CR1
CR2
CR3
TABDLY Select horizontal tab delays or tab expansion:

Device and Network Interfaces 879

termio(7I)

TAB0
TAB1
TAB2
TAB3 Expand tabs to spaces
XTABS Expand tabs to spaces
BSDLY Select backspace delays:

BS0
BS1
VTDLY Select vertical tab delays:

VT0
VT1
FFDLY Select form feed delays:

FF0
FF1

If OPOST is set, output characters are post-processed as indicated by the remaining flags;
otherwise, characters are transmitted without change.

If OLCUC is set, a lower case alphabetic character is transmitted as the corresponding upper case
character. This function is often used in conjunction with IUCLC.

If ONLCR is set, the NL character is transmitted as the CR-NL character pair. If OCRNL is set, the
CR character is transmitted as the NL character. If ONOCR is set, no CR character is transmitted
when at column 0 (first position). If ONRET is set, the NL character is assumed to do the
carriage-return function; the column pointer is set to 0 and the delays specified for CR are
used. Otherwise, the NL character is assumed to do just the line-feed function; the column
pointer remains unchanged. The column pointer is also set to 0 if the CR character is actually
transmitted.

The delay bits specify how long transmission stops to allow for mechanical or other
movement when certain characters are sent to the terminal. In all cases, a value of 0 indicates
no delay. If OFILL is set, fill characters are transmitted for delay instead of a timed delay. This is
useful for high baud rate terminals that need only a minimal delay. If OFDEL is set, the fill
character is DEL ; otherwise it is NULL.

If a form-feed or vertical-tab delay is specified, it lasts for about 2 seconds.

880 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

Newline delay lasts about 0.10 seconds. If ONLRET is set, the carriage-return delays are used
instead of the newline delays. If OFILL is set, two fill characters are transmitted.

Carriage-return delay type 1 is dependent on the current column position, type 2 is about 0.10
seconds, and type 3 is about 0.15 seconds. If OFILL is set, delay type 1 transmits two fill
characters, and type 2 transmits four fill characters.

Horizontal-tab delay type 1 is dependent on the current column position. Type 2 is about 0.10
seconds. Type 3 specifies that tabs are to be expanded into spaces. If OFILL is set, two fill
characters are transmitted for any delay.

Backspace delay lasts about 0.05 seconds. If OFILL is set, one fill character is transmitted.

The actual delays depend on line speed and system load.

Control Modes The c_cflag field describes the hardware control of the terminal:

CBAUD Baud rate:

B0 Hang up
B50 50 baud
B75 75 baud
B110 110 baud
B134 134 baud
B150 150 baud
B200 200 baud
B300 300 baud
B600 600 baud
B1200 1200 baud
B1800 1800 baud
B2400 2400 baud
B4800 4800 baud
B9600 9600 baud
B19200 19200 baud
EXTA External A
B38400 38400 baud
EXTB External B

Device and Network Interfaces 881

termio(7I)

B57600 57600 baud

B76800 76800 baud
B115200 115200 baud
B153600 153600 baud
B230400 230400 baud
B307200 307200 baud
B460800 460800 baud
CSIZE Character size:
CS5 5 bits
CS6 6 bits
CS7 7 bits
CS8 8 bits
CSTOPB Send two stop bits, else one
CREAD Enable receiver
PARENB Parity enable
PARODD Odd parity, else even
HUPCL Hang up on last close
CLOCAL Local line, else dial-up
CIBAUD Input baud rate, if different from output rate
PAREXT Extended parity for mark and space parity
CRTSXOFF Enable inbound hardware flow control
CRTSCTS Enable outbound hardware flow control
CBAUDEXT Bit to indicate output speed > B38400
CIBAUDEXT Bit to indicate input speed > B38400

The CBAUD bits together with the CBAUDEXT bit specify the output baud rate. To retrieve the
output speed from the termios structure pointed to by termios_p see the following code
segment.
speed_t ospeed;
if (termios_p->c_cflag & CBAUDEXT)
ospeed = (termios_p->c_cflag & CBAUD) + CBAUD + 1;

882 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

else
ospeed = termios_p->c_cflag & CBAUD;

To store the output speed in the termios structure pointed to by termios_p see the following
code segment.
speed_t ospeed;
if (ospeed > CBAUD) {
termios_p->c_cflag |= CBAUDEXT;
ospeed -= (CBAUD + 1);
} else
termios_p->c_cflag &= ~CBAUDEXT;
termios_p->c_cflag =
(termios_p->c_cflag & ~CBAUD) | (ospeed & CBAUD);

The zero baud rate, B0, is used to hang up the connection. If B0 is specified, the
data-terminal-ready signal is not asserted. Normally, this disconnects the line.

If the CIBAUDEXT or CIBAUD bits are not zero, they specify the input baud rate, with the
CBAUDEXT and CBAUD bits specifying the output baud rate; otherwise, the output and input
baud rates are both specified by the CBAUDEXT and CBAUD bits. The values for the CIBAUD bits
are the same as the values for the CBAUD bits, shifted left IBSHIFT bits. For any particular
hardware, impossible speed changes are ignored. To retrieve the input speed in the termios
structure pointed to by termios_p see the following code segment.
speed_t ispeed;
if (termios_p->c_cflag & CIBAUDEXT)
ispeed = ((termios_p->c_cflag & CIBAUD) >> IBSHIFT)
+ (CIBAUD >> IBSHIFT) + 1;
else
ispeed = (termios_p->c_cflag & CIBAUD) >> IBSHIFT;

To store the input speed in the termios structure pointed to by termios_p see the following
code segment.
speed_t ispeed;
if (ispeed == 0) {
ispeed = termios_p->c_cflag & CBAUD;
if (termios_p->c_cflag & CBAUDEXT)
ispeed += (CBAUD + 1);
}
if ((ispeed << IBSHIFT) > CIBAUD) {
termios_p->c_cflag |= CIBAUDEXT;
ispeed -= ((CIBAUD >> IBSHIFT) + 1);
} else
termios_p->c_cflag &= ~CIBAUDEXT;
termios_p->c_cflag =

Device and Network Interfaces 883

termio(7I)

(termios_p->c_cflag & ~CIBAUD) |

((ispeed << IBSHIFT) & CIBAUD);

The CSIZE bits specify the character size in bits for both transmission and reception. This size
does not include the parity bit, if any. If CSTOPB is set, two stop bits are used; otherwise, one
stop bit is used. For example, at 110 baud, two stops bits are required.

If PARENB is set, parity generation and detection is enabled, and a parity bit is added to each
character. If parity is enabled, the PARODD flag specifies odd parity if set; otherwise, even parity
is used.

If CREAD is set, the receiver is enabled. Otherwise, no characters are received.

If HUPCL is set, the line is disconnected when the last process with the line open closes it or
terminates. That is, the data-terminal-ready signal is not asserted.

If CLOCAL is set, the line is assumed to be a local, direct connection with no modem control;
otherwise, modem control is assumed.

If CRTSXOFF is set, inbound hardware flow control is enabled.

If CRTSCTS is set, outbound hardware flow control is enabled.

The four possible combinations for the state of CRTSCTS and CRTSXOFF bits and their
interactions are described below.
Case A: CRTSCTS off, CRTSXOFF off. In this case the hardware flow control is disabled.
Case B: CRTSCTS on, CRTSXOFF off. In this case only outbound hardware flow control is
enabled. The state of CTS signal is used to do outbound flow control. It is expected
that output will be suspended if CTS is low and resumed when CTS is high.
Case C: CRTSCTS off, CRTSXOFF on. In this case only inbound hardware flow control is
enabled. The state of RTS signal is used to do inbound flow control. It is expected
that input will be suspended if RTS is low and resumed when RTS is high.
Case D: CRTSCTS on, CRTSXOFF on. In this case both inbound and outbound hardware flow
control are enabled. Uses the state of CTS signal to do outbound flow control and
RTS signal to do inbound flow control.

Local Modes The c_lflag field of the argument structure is used by the line discipline to control terminal
functions. The basic line discipline provides the following:
ISIG Enable signals.
ICANON Canonical input (erase and kill processing).
XCASE Canonical upper/lower presentation.
ECHO Enable echo.

884 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

ECHOE Echo erase character as BS-SP-BS &.

ECHOK Echo NL after kill character.
ECHONL Echo NL .
NOFLSH Disable flush after interrupt or quit.
TOSTOP Send SIGTTOU for background output.
ECHOCTL Echo control characters as char, delete as ^?.
ECHOPRT Echo erase character as character erased.
ECHOKE BS-SP-BS erase entire line on line kill.
FLUSHO Output is being flushed.
PENDIN Retype pending input at next read or input character.
IEXTEN Enable extended (implementation-defined) functions.

If ISIG is set, each input character is checked against the special control characters INTR,
QUIT, SWTCH, SUSP, STATUS, and DSUSP. If an input character matches one of these
control characters, the function associated with that character is performed. (Note: If SWTCH
is set and the character matches, the character is simply discarded. No other action is taken.) If
ISIG is not set, no checking is done. Thus, these special input functions are possible only if
ISIG is set.

If ICANON is set, canonical processing is enabled. This enables the erase and kill edit functions,
and the assembly of input characters into lines delimited by NL-c, EOF, EOL, and EOL . If
ICANON is not set, read requests are satisfied directly from the input queue. A read is not
satisfied until at least MIN characters have been received or the timeout value TIME has
expired between characters. This allows fast bursts of input to be read efficiently while still
allowing single character input. The time value represents tenths of seconds.

If XCASE is set and ICANON is set, an upper case letter is accepted on input if preceded by a
backslash (\) character, and is output preceded by a backslash (\) character. In this mode, the
following escape sequences are generated on output and accepted on input:

FOR: USE:

‘ \'

| \!

≈ \^

{ \(

} \)

Device and Network Interfaces 885

termio(7I)

FOR: USE:

\ \\

For example, input A as \a, \n as \\n, and \N as \\\n.

If ECHO is set, characters are echoed as received.

When ICANON is set, the following echo functions are possible.

■ If ECHO and ECHOE are set, and ECHOPRT is not set, the ERASE and WERASE characters are
echoed as one or more ASCII BS SP BS, which clears the last character(s) from a CRT
screen.
■ If ECHO, ECHOPRT, and IEXTEN are set, the first ERASE and WERASE character in a
sequence echoes as a backslash (\), followed by the characters being erased. Subsequent
ERASE and WERASE characters echo the characters being erased, in reverse order. The
next non-erase character causes a ‘/' (slash) to be typed before it is echoed. ECHOPRT should
be used for hard copy terminals.
■ If ECHOKE and IEXTEN are set, the kill character is echoed by erasing each character on the
line from the screen (using the mechanism selected by ECHOE and ECHOPRa).
■ If ECHOK is set, and ECHOKE is not set, the NL character is echoed after the kill character to
emphasize that the line is deleted. Note that a ‘\' (escape) character or an LNEXT character
preceding the erase or kill character removes any special function.
■ If ECHONL is set, the NL character is echoed even if ECHO is not set. This is useful for
terminals set to local echo (so called half-duplex).

If ECHOCTL and IEXTEN are set, all control characters (characters with codes between 0 and 37
octal) other than ASCII TAB, ASCII NL, the START character, and the STOP character,
ASCII CR, and ASCII BS are echoed as ^ X, where X is the character given by adding 100 octal
to the code of the control character (so that the character with octal code 1 is echoed as ^ A),
and the ASCII DEL character, with code 177 octal, is echoed as ^ ?.

If NOFLSH is set, the normal flush of the input and output queues associated with the INTR,
QUIT, and SUSP characters is not done. This bit should be set when restarting system calls
that read from or write to a terminal (see sigaction(2) ).

If TOSTOP and IEXTEN are set, the signal SIGTTOU is sent to a process that tries to write to its
controlling terminal if it is not in the foreground process group for that terminal. This signal
normally stops the process. Otherwise, the output generated by that process is output to the
current output stream. Processes that are blocking or ignoring SIGTTOU signals are excepted
and allowed to produce output, if any.

If FLUSHO and IEXTEN are set, data written to the terminal is discarded. This bit is set when the
FLUSH character is typed. A program can cancel the effect of typing the FLUSH character by
clearing FLUSHO.

886 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

If PENDIN and IEXTEN are set, any input that has not yet been read is reprinted when the next
character arrives as input. PENDIN is then automatically cleared.

If IEXTEN is set, the following implementation-defined functions are enabled: special

characters ( WERASE, REPRINT, DISCARD, and LNEXT) and local flags ( TOSTOP, ECHOCTL,
ECHOPRT, ECHOKE, FLUSHO, and PENDIN).

Minimum and Timeout The MIN and TIME values were described previously, in the subsection, Non-canonical
Mode Input Processing. The initial value of MIN is 1, and the initial value of TIME is 0.

Terminal Size The number of lines and columns on the terminal's display is specified in the winsize
structure defined by sys/termios.h and includes the following members:
unsigned short ws_row; /* rows, in characters */
unsigned short ws_col; /* columns, in characters */
unsigned short ws_xpixel; /* horizontal size, in pixels */
unsigned short ws_ypixel; /* vertical size, in pixels */

Termio Structure The SunOS/SVR4 termio structure is used by some ioctls; it is defined by sys/termio.h and
includes the following members:
unsigned short c_iflag; /* input modes */
unsigned short c_oflag; /* output modes */
unsigned short c_cflag; /* control modes */
unsigned short c_lflag; /* local modes */
char c_line; /* line discipline */
unsigned char c_cc[NCC]; /* control chars */

The special control characters are defined by the array c_cc. The symbolic name NCC is the size
of the Control-character array and is also defined by termio.h. The relative positions,
subscript names, and typical default values for each function are as follows:

Relative Positions Subscript Names Typical Default Values

0 VINTR EXT

1 VQUIT FS

2 VERASE DEL

3 VKILL NAK

4 VEOF EOT

5 VEOL NUL

6 VEOL2 NUL

7 Reserved

Device and Network Interfaces 887

termio(7I)

The MIN values is stored in the VMIN element of the c_cc array; the TIME value is stored in the
VTIME element of the c_cc array. The VMIN element is the same element as the VEOF element;
the VTIME element is the same element as the VEOL element.

The calls that use the termio structure only affect the flags and control characters that can be
stored in the termio structure; all other flags and control characters are unaffected.

Modem Lines On special files representing serial ports, modem control lines can be read. Control lines (if the
underlying hardware supports it) may also be changed. Status lines are read-only. The
following modem control and status lines may be supported by a device; they are defined by
sys/termios.h:
TIOCM_LE line enable
TIOCM_DTR data terminal ready
TIOCM_RTS request to send
TIOCM_ST secondary transmit
TIOCM_SR secondary receive
TIOCM_CTS clear to send
TIOCM_CAR carrier detect
TIOCM_RNG ring
TIOCM_DSR data set ready

TIOCM_CD is a synonym for TIOCM_CAR, and TIOCM_RI is a synonym for TIOCM_RNG. Not all of
these are necessarily supported by any particular device; check the manual page for the device
in question.

The software carrier mode can be enabled or disabled using the TIOCSSOFTCAR ioctl. If the
software carrier flag for a line is off, the line pays attention to the hardware carrier detect
(DCD) signal. The tty device associated with the line cannot be opened until DCD is asserted.
If the software carrier flag is on, the line behaves as if DCD is always asserted.

The software carrier flag is usually turned on for locally connected terminals or other devices,
and is off for lines with modems.

To be able to issue the TIOCGSOFTCAR and TIOCSSOFTCAR ioctl calls, the tty line should be
opened with O_NDELAY so that the open(2) will not wait for the carrier.

Default Values The initial termios values upon driver open is configurable. This is accomplished by setting
the “ttymodes” property in the file /kernel/drv/options.conf. Since this property is
assigned during system initialization, any change to the “ttymodes” property will not take
effect until the next reboot. The string value assigned to this property should be in the same
format as the output of the stty(1) command with the -g option.

888 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

If this property is undefined, the following termios modes are in effect. The initial input
control value is BRKINT, ICRNL, IXON, IMAXBEL. The initial output control value is OPOST,
ONLCR, TAB3. The initial hardware control value is B9600, CS8, CREAD. The initial line-discipline
control value is ISIG, ICANON, IEXTEN, ECHO, ECHOK, ECHOE, ECHOKE, ECHOCTL.

ioctls The ioctls supported by devices and STREAMS modules providing the termios(3C)
interface are listed below. Some calls may not be supported by all devices or modules. The
functionality provided by these calls is also available through the preferred function call
interface specified on termios.
TCGETS The argument is a pointer to a termios structure. The current terminal
parameters are fetched and stored into that structure.
TCSETS The argument is a pointer to a termios structure. The current terminal
parameters are set from the values stored in that structure. The change is
immediate.
TCSETSW The argument is a pointer to a termios structure. The current terminal
parameters are set from the values stored in that structure. The change
occurs after all characters queued for output have been transmitted. This
form should be used when changing parameters that affect output.
TCSETSF The argument is a pointer to a termios structure. The current terminal
parameters are set from the values stored in that structure. The change
occurs after all characters queued for output have been transmitted; all
characters queued for input are discarded and then the change occurs.
TCGETA The argument is a pointer to a termio structure. The current terminal
parameters are fetched, and those parameters that can be stored in a
termio structure are stored into that structure.
TCSETA The argument is a pointer to a termio structure. Those terminal
parameters that can be stored in a termio structure are set from the values
stored in that structure. The change is immediate.
TCSETAW The argument is a pointer to a termio structure. Those terminal
parameters that can be stored in a termio structure are set from the values
stored in that structure. The change occurs after all characters queued for
output have been transmitted. This form should be used when changing
parameters that affect output.
TCSETAF The argument is a pointer to a termio structure. Those terminal
parameters that can be stored in a termio structure are set from the values
stored in that structure. The change occurs after all characters queued for
output have been transmitted; all characters queued for input are
discarded and then the change occurs.

Device and Network Interfaces 889

termio(7I)

TCSBRK The argument is an int value. Wait for the output to drain. If the argument
is 0, then send a break (zero valued bits for 0.25 seconds).
TCXONC Start/stop control. The argument is an int value. If the argument is 0,
suspend output; if 1, restart suspended output; if 2, suspend input; if 3,
restart suspended input.
TCFLSH The argument is an int value. If the argument is 0, flush the input queue; if
1, flush the output queue; if 2, flush both the input and output queues.
TIOCGPGRP The argument is a pointer to a pid_t. Set the value of that pid_t to the
process group ID of the foreground process group associated with the
terminal. See termios(3C) for a description of TCGETPGRP.
TIOCSPGRP The argument is a pointer to a pid_t. Associate the process group whose
process group ID is specified by the value of that pid_t with the terminal.
The new process group value must be in the range of valid process group
ID values. Otherwise, the error EPERM is returned.
TIOCGSID The argument is a pointer to a pid_t. The session ID of the terminal is
fetched and stored in the pid_t.
TIOCGWINSZ The argument is a pointer to a winsize structure. The terminal driver's
notion of the terminal size is stored into that structure.
TIOCSWINSZ The argument is a pointer to a winsize structure. The terminal driver's
notion of the terminal size is set from the values specified in that structure.
If the new sizes are different from the old sizes, a SIGWINCH signal is set to
the process group of the terminal.
TIOCMBIS The argument is a pointer to an int whose value is a mask containing
modem control lines to be turned on. The control lines whose bits are set in
the argument are turned on; no other control lines are affected.
TIOCMBIC The argument is a pointer to an int whose value is a mask containing
modem control lines to be turned off. The control lines whose bits are set
in the argument are turned off; no other control lines are affected.
TIOCMGET The argument is a pointer to an int. The current state of the modem status
lines is fetched and stored in the int pointed to by the argument.
TIOCMSET The argument is a pointer to an int containing a new set of modem control
lines. The modem control lines are turned on or off, depending on whether
the bit for that mode is set or clear.
TIOCSPPS The argument is a pointer to an int that determines whether
pulse-per-second event handling is to be enabled (non-zero) or disabled
(zero). If a one-pulse-per-second reference clock is attached to the serial

890 man pages section 7: Device and Network Interfaces • Last Revised 14 Sep 2005
 termio(7I)

line's data carrier detect input, the local system clock will be calibrated to it.
A clock with a high error, that is, a deviation of more than 25 microseconds
per tick, is ignored.
TIOCGPPS The argument is a pointer to an int, in which the state of the even handling
is returned. The int is set to a non-zero value if pulse-per-second (PPS)
handling has been enabled. Otherwise, it is set to zero.
TIOCGSOFTCAR The argument is a pointer to an int whose value is 1 or 0, depending on
whether the software carrier detect is turned on or off.
TIOCSSOFTCAR The argument is a pointer to an int whose value is 1 or 0. The value of the
integer should be 0 to turn off software carrier, or 1 to turn it on.
TIOCGPPSEV The argument is a pointer to a struct ppsclockev. This structure contains
the following members:
struct timeval tv;
uint32_t serial;
“tv” is the system clock timestamp when the event (pulse on the DCD pin)
occurred. “serial” is the ordinal of the event, which each consecutive event
being assigned the next ordinal. The first event registered gets a “serial”
value of 1. The TIOCGPPSEV returns the last event registered; multiple calls
will persistently return the same event until a new one is registered. In
addition to time stamping and saving the event, if it is of one-second
period and of consistently high accuracy, the local system clock will
automatically calibrate to it.

Files Files in or under /dev

See Also stty(1), fork(2), getpgid(2), getsid(2), ioctl(2), setsid(2), sigaction(2), signal(3C),
tcsetpgrp(3C), termios(3C), signal.h(3HEAD), streamio(7I)

Device and Network Interfaces 891

termiox(7I)

Name termiox – extended general terminal interface

Description The extended general terminal interface supplements the termio(7I) general terminal
interface by adding support for asynchronous hardware flow control, isochronous flow
control and clock modes, and local implementations of additional asynchronous features.
Some systems may not support all of these capabilities because of either hardware or software
limitations. Other systems may not permit certain functions to be disabled. In these cases the
appropriate bits will be ignored. See <sys/termiox.h> for your system to find out which
capabilities are supported.

Hardware Flow Control Hardware flow control supplements the termio(7I) IXON, IXOFF, and IXANY character flow
Modes control. Character flow control occurs when one device controls the data transfer of another
device by the insertion of control characters in the data stream between devices. Hardware
flow control occurs when one device controls the data transfer of another device using
electrical control signals on wires (circuits) of the asynchronous interface. Isochronous
hardware flow control occurs when one device controls the data transfer of another device by
asserting or removing the transmit clock signals of that device. Character flow control and
hardware flow control may be simultaneously set.

In asynchronous, full duplex applications, the use of the Electronic Industries Association's
EIA-232-D Request To Send (RTS) and Clear To Send (CTS) circuits is the preferred method
of hardware flow control. An interface to other hardware flow control methods is included to
provide a standard interface to these existing methods.

The EIA-232-D standard specified only unidirectional hardware flow control - the Data
Circuit-terminating Equipment or Data Communications Equipment (DCE) indicates to the
Data Terminal Equipment (DTE) to stop transmitting data. The termiox interface allows
both unidirectional and bidirectional hardware flow control; when bidirectional flow control
is enabled, either the DCE or DTE can indicate to each other to stop transmitting data across
the interface. Note: It is assumed that the asynchronous port is configured as a DTE. If the
connected device is also a DTE and not a DCE, then DTE to DTE (for example, terminal or
printer connected to computer) hardware flow control is possible by using a null modem to
interconnect the appropriate data and control circuits.

Clock Modes Isochronous communication is a variation of asynchronous communication whereby two

communicating devices may provide transmit and/or receive clock signals to one another.
Incoming clock signals can be taken from the baud rate generator on the local isochronous
port controller, from CCITT V.24 circuit 114, Transmitter Signal Element Timing - DCE
source (EIA-232-D pin 15), or from CCITT V.24 circuit 115, Receiver Signal Element Timing
- DCE source (EIA-232-D pin 17). Outgoing clock signals can be sent on CCITT V.24 circuit
113, Transmitter Signal Element Timing - DTE source (EIA-232-D pin 24), on CCITT V.24
circuit 128, Receiver Signal Element Timing - DTE source (no EIA-232-D pin), or not sent at
all.

892 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 termiox(7I)

In terms of clock modes, traditional asynchronous communication is implemented simply by

using the local baud rate generator as the incoming transmit and receive clock source and not
outputting any clock signals.

Terminal Parameters The parameters that control the behavior of devices providing the termiox interface are
specified by the termiox structure defined in the <sys/termiox.h> header. Several ioctl(2)
system calls that fetch or change these parameters use this structure:
#define NFF 5
struct termiox {
unsigned short x_hflag; /* hardware flow control modes */
unsigned short x_cflag; /* clock modes */
unsigned short x_rflag[NFF]; /* reserved modes */
unsigned short x_sflag; /* spare local modes */
};

The x_hflag field describes hardware flow control modes:

RTSXOFF 0000001 Enable RTS hardware flow control on input.

CTSXON 0000002 Enable CTS hardware flow control on output.

DTRXOFF 0000004 Enable DTR hardware flow control on input.

CDXON 0000010 Enable CD hardware flow control on output.

ISXOFF 0000020 Enable isochronous hardware flow control on input

The EIA-232-D DTR and CD circuits are used to establish a connection between two systems.
The RTS circuit is also used to establish a connection with a modem. Thus, both DTR and RTS
are activated when an asynchronous port is opened. If DTR is used for hardware flow control,
then RTS must be used for connectivity. If CD is used for hardware flow control, then CTS
must be used for connectivity. Thus, RTS and DTR (or CTS and CD) cannot both be used for
hardware flow control at the same time. Other mutual exclusions may apply, such as the
simultaneous setting of the termio(7I) HUPCL and the termiox DTRXOFF bits, which use the
DTE ready line for different functions.

Variations of different hardware flow control methods may be selected by setting the the
appropriate bits. For example, bidirectional RTS/CTS flow control is selected by setting both
the RTSXOFF and CTSXON bits and bidirectional DTR/CTS flow control is selected by setting
both the DTRXOFF and CTSXON. Modem control or unidirectional CTS hardware flow control is
selected by setting only the CTSXON bit.

As previously mentioned, it is assumed that the local asynchronous port (for example,
computer) is configured as a DTE. If the connected device (for example, printer) is also a DTE,
it is assumed that the device is connected to the computer's asynchronous port using a null
modem that swaps control circuits (typically RTS and CTS). The connected DTE drives RTS

Device and Network Interfaces 893

termiox(7I)

and the null modem swaps RTS and CTS so that the remote RTS is received as CTS by the local
DTE. In the case that CTSXON is set for hardware flow control, printer's lowering of its RTS
would cause CTS seen by the computer to be lowered. Output to the printer is suspended until
the printer's raising of its RTS, which would cause CTS seen by the computer to be raised.

If RTSXOFF is set, the Request To Send (RTS) circuit (line) will be raised, and if the
asynchronous port needs to have its input stopped, it will lower the Request To Send (RTS)
line. If the RTS line is lowered, it is assumed that the connected device will stop its output until
RTS is raised.

If CTSXON is set, output will occur only if the Clear To Send (CTS) circuit (line) is raised by the
connected device. If the CTS line is lowered by the connected device, output is suspended until
CTS is raised.

If DTRXOFF is set, the DTE Ready (DTR) circuit (line) will be raised, and if the asynchronous
port needs to have its input stopped, it will lower the DTE Ready (DTR) line. If the DTR line is
lowered, it is assumed that the connected device will stop its output until DTR is raised.

If CDXON is set, output will occur only if the Received Line Signal Detector (CD) circuit (line) is
raised by the connected device. If the CD line is lowered by the connected device, output is
suspended until CD is raised.

If ISXOFF is set, and if the isochronous port needs to have its input stopped, it will stop the
outgoing clock signal. It is assumed that the connected device is using this clock signal to
create its output. Transit and receive clock sources are programmed using the x_cflag fields.
If the port is not programmed for external clock generation, ISXOFF is ignored. Output
isochronous flow control is supported by appropriate clock source programming using the
x_cflag field and enabled at the remote connected device.

The x_cflag field specifies the system treatment of clock modes.

XMTCLK 0000007 Transmit clock source:

XCIBRG 0000000 Get transmit clock from internal baud rate generator.

XCTSET 0000001 Get transmit clock from transmitter signal element timing
(DCE source) lead, CCITT V.24 circuit 114, EIA-232-D
pin 15.

XCRSET 0000002 Get transmit clock from receiver signal element timing
(DCE source) lead, CCITT V.24 circuit 115, EIA-232-D
pin 17.

RCVCLK 0000070 Receive clock source:

RCIBRG 0000000 Get receive clock from internal baud rate generator.

894 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 termiox(7I)

RCTSET 0000010 Get receive clock from transmitter signal element timing
(DCE source) lead, CCITT V.24 circuit 114, EIA-232-D
pin 15.

RCRSET 0000020 Get receive clock from receiver signal element timing
(DCE source) lead, CCITT V.24 circuit 115, EIA-232-D
pin 17.

TSETCLK 0000700 Transmitter signal element timing (DTE source) lead,

CCITT V.24 circuit 113, EIA-232-D pin 24, clock source:

TSETCOFF 0000000 TSET clock not provided.

TSETCRBRG 0000100 Output receive baud rate generator on circuit 113.

TSETCTBRG 0000200 Output transmit baud rate generator on circuit 113

TSETCTSET 0000300 Output transmitter signal element timing (DCE source)

on circuit 113.

TSETCRSET 0000400 Output receiver signal element timing (DCE source) on

circuit 113.

RSETCLK 0007000 Receiver signal element timing (DTE source) lead, CCITT
V.24 circuit 128, no EIA-232-D pin, clock source:

RSETCOFF 0000000 RSET clock not provided.

RSETCRBRG 0001000 Output receive baud rate generator on circuit 128.

RSETCTBRG 0002000 Output transmit baud rate generator on circuit 128.

RSETCTSET 0003000 Output transmitter signal element timing (DCE source)

on circuit 128.

RSETCRSET 0004000 Output receiver signal element timing (DCE) on circuit

128.

If the XMTCLK field has a value of XCIBRG the transmit clock is taken from the hardware internal
baud rate generator, as in normal asynchronous transmission. If XMTCLK = XCTSET the
transmit clock is taken from the Transmitter Signal Element Timing (DCE source) circuit. If
XMTCLK = XCRSET the transmit clock is taken from the Receiver Signal Element Timing (DCE
source) circuit.

If the RCVCLK field has a value of RCIBRG the receive clock is taken from the hardware Internal
Baud Rate Generator, as in normal asynchronous transmission. If RCVCLK = RCTSET the receive
clock is taken from the Transmitter Signal Element Timing (DCE source) circuit. If RCVCLK =
RCRSET the receive clock is taken from the Receiver Signal Element Timing (DCE source)
circuit.

If the TSETCLK field has a value of TSETCOFF the Transmitter Signal Element Timing (DTE
source) circuit is not driven. If TSETCLK = TSETCRBRG the Transmitter Signal Element Timing

Device and Network Interfaces 895

termiox(7I)

(DTE source) circuit is driven by the Receive Baud Rate Generator. If TSETCLK = TSETCTBRG
the Transmitter Signal Element Timing (DTE source) circuit is driven by the Transmit Baud
Rate Generator. If TSETCLK = TSETCTSET the Transmitter Signal Element Timing (DTE
source) circuit is driven by the Transmitter Signal Element Timing (DCE source). If TSETCLK
= TSETCRBRG the Transmitter Signal Element Timing (DTE source) circuit is driven by the
Receiver Signal Element Timing (DCE source).

If the RSETCLK field has a value of RSETCOFF the Receiver Signal Element Timing (DTE source)
circuit is not driven. If RSETCLK = RSETCRBRG the Receiver Signal Element Timing (DTE
source) circuit is driven by the Receive Baud Rate Generator. If RSETCLK = RSETCTBRG the
Receiver Signal Element Timing (DTE source) circuit is driven by the Transmit Baud Rate
Generator. If RSETCLK = RSETCTSET the Receiver Signal Element Timing (DTE source) circuit
is driven by the Transmitter Signal Element Timing (DCE source). If RSETCLK = RSETCRBRG
the Receiver Signal Element Timing (DTE source) circuit is driven by the Receiver Signal
Element Timing (DCE source).

The x_rflag is reserved for future interface definitions and should not be used by any
implementations. The x_sflag may be used by local implementations wishing to customize
their terminal interface using the termiox ioctl system calls.

ioctls The ioctl(2) system calls have the form:

ioctl (fildes, command, arg) struct termiox * arg;

The commands using this form are:

TCGETX The argument is a pointer to a termiox structure. The current terminal
parameters are fetched and stored into that structure.
TCSETX The argument is a pointer to a termiox structure. The current terminal
parameters are set from the values stored in that structure. The change is
immediate.
TCSETXW The argument is a pointer to a termiox structure. The current terminal
parameters are set from the values stored in that structure. The change occurs
after all characters queued for output have been transmitted. This form should be
used when changing parameters that will affect output.
TCSETXF The argument is a pointer to a termiox structure. The current terminal
parameters are set from the values stored in that structure. The change occurs
after all characters queued for output have been transmitted; all characters
queued for input are discarded and then the change occurs.

Files /dev/*

See Also stty(1), ioctl(2), termio(7I)

896 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 termiox(7I)

Notes The termiox(7I) system call is provided for compatibility with previous releases and its use is
discouraged. Instead, the termio(7I) system call is recommended. See termio(7I) for usage
information.

Device and Network Interfaces 897

ticlts(7D)

Name ticlts, ticots, ticotsord – loopback transport providers

Synopsis #include <sys/ticlts.h>
#include <sys/ticots.h>
#include <sys/ticotsord.h>

Description The devices known as ticlts, ticots, and ticotsord are ‘‘loopback transport providers,''
that is, stand-alone networks at the transport level. Loopback transport providers are
transport providers in every sense except one: only one host (the local machine) is ‘‘connected
to'' a loopback network. Loopback transports present a TPI (STREAMS-level) interface to
application processes and are intended to be accessed via the TLI (application-level) interface.
They are implemented as clone devices and support address spaces consisting of
‘‘flex-addresses,'' that is, arbitrary sequences of octets of length > 0, represented by a netbuf
structure.

ticlts is a datagram-mode transport provider. It offers (connectionless) service of type

T_CLTS. Its default address size is TCL_DEFAULTADDRSZ. ticlts prints the following error
messages (see t_rcvuderr(3NSL)):
TCL_BADADDR bad address specification
TCL_BADOPT bad option specification
TCL_NOPEER bound
TCL_PEERBADSTATE peer in wrong state

ticots is a virtual circuit-mode transport provider. It offers (connection-oriented) service of

type T_COTS. Its default address size is TCO_DEFAULTADDRSZ. ticots prints the following
disconnect messages (see t_rcvdis(3NSL)):
TCO_NOPEER no listener on destination address
TCO_PEERNOROOMONQ peer has no room on connect queue
TCO_PEERBADSTATE peer in wrong state
TCO_PEERINITIATED peer-initiated disconnect
TCO_PROVIDERINITIATED provider-initiated disconnect

ticotsord is a virtual circuit-mode transport provider, offering service of type T_COTS_ORD

(connection-oriented service with orderly release). Its default address size is
TCOO_DEFAULTADDRSZ. ticotsord prints the following disconnect messages (see
t_rcvdis(3NSL)):
TCOO_NOPEER no listener on destination address
TCOO_PEERNOROOMONQ peer has no room on connect queue

898 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 ticlts(7D)

TCOO_PEERBADSTATE peer in wrong state

TCOO_PEERINITIATED provider-initiated disconnect
TCOO_PROVIDERINITIATED peer-initiated disconnect

Usage Loopback transports support a local IPC mechanism through the TLI interface. Applications
implemented in a transport provider-independent manner on a client-server model using this
IPC are transparently transportable to networked environments.

Transport provider-independent applications must not include the headers listed in the
synopsis section above. In particular, the options are (like all transport provider options)
provider dependent.

ticlts and ticots support the same service types (T_CLTS and T_COTS) supported by the OSI
transport-level model.

ticotsord supports the same service type (T_COTSORD) supported by the TCP/IP model.

Files /dev/ticlts

/dev/ticots

/dev/ticotsord

See Also t_rcvdis(3NSL), t_rcvuderr(3NSL)

Device and Network Interfaces 899

timod(7M)

Name timod – Transport Interface cooperating STREAMS module

Synopsis #include <sys/stropts.h>
ioctl(fildes, I_STR, &my_strioctl);

Description timod is a STREAMS module for use with the Transport Interface (“TI”) functions of the
Network Services library. The timod module converts a set of ioctl(2) calls into STREAMS
messages that may be consumed by a transport protocol provider that supports the Transport
Interface. This allows a user to initiate certain TI functions as atomic operations.

The timod module must be pushed onto only a stream terminated by a transport protocol
provider that supports the TI.

All STREAMS messages, with the exception of the message types generated from the ioctl
commands described below, will be transparently passed to the neighboring module or driver.
The messages generated from the following ioctl commands are recognized and processed
by the timod module. The format of the ioctl call is:
#include <sys/stropts.h>
-
-
struct strioctl my_strioctl;
-
-
strioctl.ic_cmd = cmd;
strioctl.ic_timout = INFTIM;
strioctl.ic_len = size;
strioctl.ic_dp = (char *)buf
ioctl(fildes, I_STR, &my_strioctl);

On issuance, size is the size of the appropriate TI message to be sent to the transport provider
and on return size is the size of the appropriate TI message from the transport provider in
response to the issued TI message. buf is a pointer to a buffer large enough to hold the contents
of the appropriate TI messages. The TI message types are defined in <sys/tihdr.h>. The
possible values for the cmd field are:
TI_BIND Bind an address to the underlying transport protocol provider. The message
issued to the TI_BIND ioctl is equivalent to the TI message type T_BIND_REQ
and the message returned by the successful completion of the ioctl is
equivalent to the TI message type T_BIND_ACK.
TI_UNBIND Unbind an address from the underlying transport protocol provider. The
message issued to the TI_UNBIND ioctl is equivalent to the TI message type
T_UNBIND_REQ and the message returned by the successful completion of the
ioctl is equivalent to the TI message type T_OK_ACK.
TI_GETINFO Get the TI protocol specific information from the transport protocol
provider. The message issued to the TI_GETINFO ioctl is equivalent to the TI

900 man pages section 7: Device and Network Interfaces • Last Revised 26 Mar 1993
 timod(7M)

message type T_INFO_REQ and the message returned by the successful

completion of the ioctl is equivalent to the TI message type T_INFO_ACK.
TI_OPTMGMT Get, set, or negotiate protocol specific options with the transport protocol
provider. The message issued to the TI_OPTMGMT ioctl is equivalent to the TI
message type T_OPTMGMT_REQ and the message returned by the successful
completion of the ioctl is equivalent to the TI message type
T_OPTMGMT_ACK.
Files <sys/timod.h> ioctl definitions
<sys/tiuser.h> TLI interface declaration and structure file
<sys/tihdr.h> TPI declarations and user-level code
<sys/errno.h> system error messages file. Please see errno(3C).

See Also Intro(3), ioctl(2), errno(3C), tirdwr(7M)

STREAMS Programming Guide

Diagnostics If the ioctl returns with a value greater than 0, the lower 8 bits of the return value will be one
of the TI error codes as defined in <sys/tiuser.h>. If the TI error is of type TSYSERR, then the
next 8 bits of the return value will contain an error as defined in <sys/errno.h> (see
Intro(3)).

Device and Network Interfaces 901

tirdwr(7M)

Name tirdwr – Transport Interface read/write interface STREAMS module

Synopsis int ioctl( fd, I_PUSH, "tirdwr");

Description tirdwr is a STREAMS module that provides an alternate interface to a transport provider
which supports the Transport Interface (“TI”) functions of the Network Services library (see
Section 3N). This alternate interface allows a user to communicate with the transport protocol
provider using the read(2) and write(2) system calls. The putmsg(2) and getmsg(2) system
calls may also be used. However, putmsg and getmsg can only transfer data messages between
user and stream; control portions are disallowed.

The tirdwr module must only be pushed (see I_PUSH in streamio(7I)) onto a stream
terminated by a transport protocol provider which supports the TI. After the tirdwr module
has been pushed onto a stream, none of the TI functions can be used. Subsequent calls to TI
functions cause an error on the stream. Once the error is detected, subsequent system calls on
the stream return an error with errno set to EPROTO.

The following are the actions taken by the tirdwr module when pushed on the stream,
popped (see I_POP in streamio(7I)) off the stream, or when data passes through it.
push When the module is pushed onto a stream, it checks any existing data destined for
the user to ensure that only regular data messages are present. It ignores any
messages on the stream that relate to process management, such as messages that
generate signals to the user processes associated with the stream. If any other
messages are present, the I_PUSH will return an error with errno set to EPROTO.
write The module takes the following actions on data that originated from a write
system call:
■ All messages with the exception of messages that contain control portions (see
the putmsg and getmsg system calls) are transparently passed onto the module's
downstream neighbor.
■ Any zero length data messages are freed by the module and they will not be
passed onto the module's downstream neighbor.
■ Any messages with control portions generate an error, and any further system
calls associated with the stream fails with errno set to EPROTO.
read The module takes the following actions on data that originated from the transport
protocol provider.

All messages with the exception of those that contain control portions (see the
putmsg and getmsg system calls) are transparently passed onto the module's
upstream neighbor. The action taken on messages with control portions will be as
follows:
■ Any data messages with control portions have the control portions removed
from the message before to passing the message on to the upstream neighbor.

902 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 tirdwr(7M)

■ Messages that represent an orderly release indication from the transport

provider generate a zero length data message, indicating the end of file, which
will be sent to the reader of the stream. The orderly release message itself is freed
by the module.
■ Messages that represent an abortive disconnect indication from the transport
provider cause all further write and putmsg system calls to fail with errno set to
ENXIO. All further read and getmsg system calls return zero length data
(indicating end of file) once all previous data has been read.
■ With the exception of the above rules, all other messages with control portions
generate an error and all further system calls associated with the stream will fail
with errno set to EPROTO.

Any zero length data messages are freed by the module and they are not passed onto
the module's upstream neighbor.
pop When the module is popped off the stream or the stream is closed, the module takes
the following action:
■ If an orderly release indication has been previously received, then an orderly
release request will be sent to the remote side of the transport connection.

See Also Intro(3), getmsg(2), putmsg(2), read(2), write(2), Intro(3), streamio(7I), timod(7M)

STREAMS Programming Guide

Device and Network Interfaces 903

tmpfs(7FS)

Name tmpfs – memory based file system

Synopsis #include <sys/mount.h>

mount (special, directory, MS_DATA, "tmpfs", NULL, 0);

Description tmpfs is a memory based file system which uses kernel resources relating to the VM system
and page cache as a file system. Once mounted, a tmpfs file system provides standard file
operations and semantics. tmpfs is so named because files and directories are not preserved
across reboot or unmounts, all files residing on a tmpfs file system that is unmounted will be
lost.

tmpfs file systems can be mounted with the command:

mount -F tmpfs swap directory

Alternatively, to mount a tmpfs file system on /tmp at multi-user startup time (maximizing
possible performance improvements), add the following line to /etc/vfstab:
swap −/tmp tmpfs − yes −

tmpfs is designed as a performance enhancement which is achieved by caching the writes to

files residing on a tmpfs file system. Performance improvements are most noticeable when a
large number of short lived files are written and accessed on a tmpfs file system. Large
compilations with tmpfs mounted on /tmp are a good example of this.

Users of tmpfs should be aware of some constraints involved in mounting a tmpfs file system.
The resources used by tmpfs are the same as those used when commands are executed (for
example, swap space allocation). This means that large sized tmpfs files can affect the amount
of space left over for programs to execute. Likewise, programs requiring large amounts of
memory use up the space available to tmpfs. Users running into this constraint (for example,
running out of space on tmpfs) can allocate more swap space by using the swap(1M)
command.

Another constraint is that the number of files available in a tmpfs file system is calculated
based on the physical memory of the machine and not the size of the swap device/partition. If
you have too many files, tmpfs will print a warning message and you will be unable to create
new files. You cannot increase this limit by adding swap space.

Normal file system writes are scheduled to be written to a permanent storage medium along
with all control information associated with the file (for example, modification time, file
permissions). tmpfs control information resides only in memory and never needs to be
written to permanent storage. File data remains in core until memory demands are sufficient
to cause pages associated with tmpfs to be reused at which time they are copied out to swap.

An additional mount option can be specified to control the size of an individual tmpfs file
system.

904 man pages section 7: Device and Network Interfaces • Last Revised 9 Oct 1990
 tmpfs(7FS)

See Also df(1M), mount(1M), mount_tmpfs(1M), swap(1M), mmap(2), mount(2), umount(2), vfstab(4)

System Administration Guide: Basic Administration

Diagnostics If tmpfs runs out of space, one of the following messages will display in the console.
directory: File system full, swap space limit exceeded
This message appears because a page could not be allocated while writing to a file. This can
occur if tmpfs is attempting to write more than it is allowed, or if currently executing
programs are using a lot of memory. To make more space available, remove unnecessary
files, exit from some programs, or allocate more swap space using swap(1M).
directory: File system full, memory allocation failed
tmpfs ran out of physical memory while attempting to create a new file or directory.
Remove unnecessary files or directories or install more physical memory.

Warnings Files and directories on a tmpfs file system are not preserved across reboots or unmounts.
Command scripts or programs which count on this will not work as expected.

Notes Compilers do not necessarily use /tmp to write intermediate files therefore missing some
significant performance benefits. This can be remedied by setting the environment variable
TMPDIR to /tmp. Compilers use the value in this environment variable as the name of the
directory to store intermediate files.

swap to a tmpfs file is not supported.

df(1M) output is of limited accuracy since a tmpfs file system size is not static and the space
available to tmpfs is dependent on the swap space demands of the entire system.

Device and Network Interfaces 905

todopol(7D)

Name todopol – Time-Of-Day driver for SPARC Enterprise Server family

Description The todopl driver is the Time-Of-Day (TOD) driver for the SPARC Enterprise Server family.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcakr

Interface Stability Private

See Also attributes(5)

906 man pages section 7: Device and Network Interfaces • Last Revised 18 Apr 2006
 tokenmt(7ipp)

Name tokenmt – Single and Two Rate Three Conformance Level Meter
Description The tokenmt module can be configured as a Single or a Two Rate meter. Packets are deemed to
belong to one of the three levels - Red, Yellow or Green - depending on the configured rate(s)
and the burst sizes. When configured as a Single Rate meter, tokenmt can operate with just the
Green and Red levels.
Configuration parameters for tokenmt correspond to definitions in RFC– 2697 and RFC–
2698 as follows:
Configuring tokenmt as a Single Rate meter (from RFC– 2697):

committed_rate - CIR
committed_burst - CBS
peak_burst - EBS
(thus 'peak_burst' for a single rate meter is actually the 'excess burst' in the RFC. However,
throughout the text the parameter name “peak burst” is used.)
Configuring tokenmt as a Two Rate meter (from RFC– 2698):

committed_rate - CIR
peak_rate - PIR
committed_burst - CBS
peak_burst - PBS
The meter is implemented using token buckets C and P, which initially hold tokens equivalent
to committed and peak burst sizes (bits) respectively. When a packet of size B bits arrive at
time t, the following occurs:
When operating as a Single Rate meter, the outcome (level)
is decided as follows:
- Update tokens in C and P
o Compute no. of tokens accumulated since the
last time packet was seen at the committed rate as
T(t) = committed rate * (t - t’)
(where t’ is the time the last packet was seen)
o Add T tokens to C up to a maximum of committed burst
size. Add remaining tokens ((C+T) - Commited Burst),
if any, to P, to a maximum of peak burst size.
- Decide outcome
o If not color aware
o If B <= C, outcome is GREEN and C -= B.
o Else, if B <= P, outcome is YELLOW and P -= B.
o Else, outcome is Red.
o Else,
o obtain DSCP from packet

Device and Network Interfaces 907

tokenmt(7ipp)

o obtain color from color_map, color_map[DSCP]

o if (color is GREEN) and (B <= C), outcome is
GREEN and C -= B.
o Else, if (color is GREEN or YELLOW) and
(B <= P), outcome is YELLOW and P -= B.
o Else, outcome is RED.
Note that if peak_burst and yellow_next_actions are
not specified (that is, a single rate meter with two
outcomes), the outcome is never YELLOW.

When operating as a Two Rate meter, the outcome (level) is decided as follows:
- Update tokens in C and P
o Compute no. of tokens accumulated since the last time a
packet was seen at the committed and peak rates as
Tc(t) = committed rate * (t - t’)
Tp(t) = peak rate * (t - t’)
(where t’ is the time the last packet was seen)
o Add Tc to C up to a maximum of committed burst size
o Add Tp to P up to a maximum of peak burst size
- Decide outcome
o If not color aware
o If B > P, outcome is RED.
o Else, if B > C, outcome is YELLOW and P -= B
o Else, outcome is GREEN and C -= B & P -= B
o Else,
o obtain DSCP from packet
o obtain color from color_map, color_map[DSCP]
o if (color is RED) or (B > P), outcome is RED
o Else, if (color is YELLOW) or (B > C),
outcome is YELLOW and P -= B
o Else, outcome is GREEN and C -= B & P -= B

Statistics The tokenmt module exports the following statistics through kstat:

Global statistics:
module: tokenmt instance: <action id>
name: tokenmt statistics class <action name>
epackets <number of packets in error>
green_bits <number of bits in green>
green_packets <number of packets in green>
red_bits <number of bits in red>
red_packets <number of packets in red>
yellow_bits <number of bits in yellow>
yellow packets <number of packets in yellow>

908 man pages section 7: Device and Network Interfaces • Last Revised 29 Sep 2004
 tokenmt(7ipp)

Files /kernel/ipp/sparcv9/tokenmt 64-bit module (SPARC only.)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos

See Also ipqosconf(1M), dlcosmk(7ipp), dscpmk(7ipp), flowacct(7ipp), ipqos(7ipp), ipgpc(7ipp),

tswtclmt(7ipp)

RFC 2697, A Single Rate Three Color Marker J. Heinanen, R. Guerin — The Internet Society,
1999

RFC 2698, A Two Rate Three Color Marker J. Heinanen, R. Guerin — The Internet Society,
1999

Device and Network Interfaces 909

tpf(7D)

Name tpf – Platform Specific Module (PSM) for Tricord Systems Enterprise Server Models ES3000,
ES4000 and ES5000.

Description tpf provides the platform dependent functions for Solaris x86 MP support. These functions
adhere to the PSMI Specifications. (Platform Specific Module Interface Specifications.)
Tricord Systems Enterprise Servers are Intel APIC based MP platforms which run from 1 to 12
Intel processors. The tpf psm supports dynamic interrupt distribution across all processors in
an MP configuration.

The psm is automatically invoked on an ESxxxx platform at system boot time.

Files /kernel/mach/tpf MP module.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5)

910 man pages section 7: Device and Network Interfaces • Last Revised 1 Jan 1997
 tsalarm(7D)

Name tsalarm – Alarm device driver

Synopsis tsalarm@0:ctl
Description The tsalarm driver is a Multi-threaded, loadable non-STREAMS pseudo driver that manages
ALOM alarms. The tsalarm driver provides an interface through which alarm relays can be
controlled on SUNW,Netra-240 and SUNW,Netra-440 platforms.
Hardware The alarm hardware differs depending on platform. The Netra 240 and 440 platforms features
Interface four dry contact alarm relays which are controlled by ALOM. You can set each alarm to “on”
or “off” by using ioctl interfaces provided from the host. The four alarms are labeled as
“critical,” “major,” “minor,” and “user.” The user alarm is set by a user application depending
on system condition. LED's in front of the box provide a visual indication of the four alarms.
The number of alarms and their meanings/labels may vary across platforms.
ioctls The interface provided by the tsalarm driver comprises ioctls that enable applications to
manipulate the alarm module. The alarm module is accessed via two device nodes: i)
/dev/lom and /dev/tsalarm:ctl.
The following ioctls are supported by the /dev/lom and /dev/tsalarm:ctl devices:
TSIOCALCTL - Turn an alarm on or off. The argument is a pointer to the
ts_aldata_t/lom_aldata_t
structure. This structure is described
below. alarm_no member is an integer
which specifies the alarm to which the
command is to be applied. The
alarm_state/state structure
member indicates the state to which
the alarm should be set (where 0 ==
off). An error (EINVAL) is returned if
either an invalid alarm_no or invalid
alarm_state is provided.
TSIOCALSTATE - Get the state of the alarms. The argument is a pointer to the
ts_aldata_t/lom_aldata_t
structure. This structure is described
below. alarm_no member is an integer
which indicates the alarm to which the
command will be applied. The
alarm_state member holds the
alarm's current state and is filled in by
the driver. A zero indicates that the
alarm is off. An error (EINVAL) is
returned if an invalid alarm_no is
provided. The structures and
definitions for the values are defined

Device and Network Interfaces 911

tsalarm(7D)

below.

Alarm values:
The following old style values are defined in <lom.io.h>

#define ALARM_NUM_0 0 /* number of zero’th alarm */

#define ALARM_NUM_1 1 /* number of first alarm */

#define ALARM_NUM_2 2 /* number of second alarm */

#define ALARM_NUM_3 3 /* number of third alarm */

Alarm values defined in <lom.io.h>

#define ALARM_OFF 0 /* Turn off alarm */

#define ALARM_ON 1 /* Turn on alarm */

Alarm Data Structure:

This structure is defined in <lom.io.h>

typedef struct {

int alarm_no; /* alarm to apply command to */

int alarm_state; /* state of alarm (0 == off) */

} ts_aldata_t;

Use the following LOM interfaces to get and set the alarms. These definitions are included in
<lom_io.h>
#define ALARM_CRITICAL 0 /* number of critical alarm */

#define ALARM_MAJOR 1 /* number of major alarm */

#define ALARM_MINOR 2 /* number of minor alarm */

#define ALARM_USER 3 /* number of user alarm */

The following alarm data structure is provided in <lom_io.h>:

typedef struct {

int alarm_no;

912 man pages section 7: Device and Network Interfaces • Last Revised 16 Mar 2004
 tsalarm(7D)

int state;

} lom_aldata_t;

Errors An open() will fail if:

ENXIO The driver is not installed in the system.

An ioctl() will fail if:

EFAULT There was a hardware failure during the specified operation.
EINVAL The alarm number specified is not valid or an invalid value was supplied.
ENXIO The driver is not installed in the system or the monitor callback routine could
not be scheduled.
Examples How to set an alarm:

#include <sys/unistd.h>
#include <fcntl.h>
#include <stdio.h>
#include <lom_io.h>

#define LOM_DEVICE "/dev/lom"

int
main()
{
lom_aldata_t lld;
int fd = open(LOM_DEVICE, O_RDWR);

if (fd == -1) {
printf("Error opening device: %s\n", LOM_DEVICE);
exit (1);
}

lld.alarm_no = ALARM_CRITICAL; /* Set the critical alarm */

lld.state = ALARM_ON; /* Set the alarm */

if (ioctl(fd, LOMIOCALCTL, (char *)&lld) != 0)

printf("Setting alarm failed");
else
printf("Alarm set successfully");

close(fd);

Device and Network Interfaces 913

tsalarm(7D)

Files /dev/lom
LOM device.
/dev/tsalarm:ctl
Alarm control device.
/platform/platform/kernel/drv/sparcv9/tsalarm
Device driver module.
/platform/SUNW,Netra-240/kernel/drv/tsalarm.conf
Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcarx.u

Architecture SPARC

See Also attributes(5)

Writing Device Drivers

914 man pages section 7: Device and Network Interfaces • Last Revised 16 Mar 2004
 tswtclmt(7ipp)

Name tswtclmt – Time Sliding Window Three Conformance Level Meter

Description The Time Sliding Window Three Conformance level meter (tswtcl) meters a traffic stream
and determines the conformance level of its packets.

Packets are deemed to belong to one of the three levels, Red, Yellow or Green, depending on
the committed and peak rate.

The meter provides an estimate of the running average bandwidth. It takes into account
burstiness and smoothes out its estimate to approximate the longer-term measured sending
rate of the traffic stream.

The estimated bandwidth approximates the running average bandwidth of the traffic stream
over a specific window (time interval). tswtcl estimates the average bandwidth using a
time-based estimator. When a packet arrives for a class, tswtcl re-computes the average rate
by using the rate in the last window (time interval) and the size of the arriving packet. The
window is then slid to start at the current time (the packet arrival time). If the computed rate is
less than the committed configuration parameter, the packet is deemed Green; else if the rate
is less than the peak rate, it is Yellow; else Red. To avoid dropping multiple packets within a
TCP window, tswtcl probabilistically assigns one of the three conformance level to the
packet.

Statistics The tswtcl module exports global and per-class statistics through kstat:

Global statistics:
module: tswtclmt instance: <action id>
name: tswtclmt statistics class <action name>
green_bits <number of bit in green>
green_packets <number of packets in green>
red_bits <number of bits in red>
red_packets <number of packets in red>
yellow_bits <number of bits in yellow>
yellow_packets <number of packets in yellow>
Files /kernel/ipp/sparcv9/tswtclmt 64-bit module (SPARC only.)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWqos

See Also ipqosconf(1M), dlcosmk(7ipp), dscpmk(7ipp), flowacct(7ipp), ipqos(7ipp), ipgpc(7ipp),

tokenmt(7ipp)

RFC 2859, A Time Sliding Window Three Colour Marker (TSWTCM) W. Fang, N. Seddigh, B.
Nandy — The Internet Society, 2000

Device and Network Interfaces 915

ttcompat(7M)

Name ttcompat – V7, 4BSD and XENIX STREAMS compatibility module

Synopsis #define BSD_COMP
#include <sys/stropts.h>
#include <sys/ioctl.h>
ioctl(fd, I_PUSH, "ttcompat");

Description ttcompat is a STREAMS module that translates the ioctl calls supported by the older
Version 7, 4BSD, and XENIX terminal drivers into the ioctl calls supported by the termio
interface (see termio(7I)). All other messages pass through this module unchanged; the
behavior of read and write calls is unchanged, as is the behavior of ioctl calls other than the
ones supported by ttcompat.
This module can be automatically pushed onto a stream using the autopush mechanism when
a terminal device is opened; it does not have to be explicitly pushed onto a stream. This
module requires that the termios interface be supported by the modules and the application
can push the driver downstream. The TCGETS, TCSETS, and TCSETSF ioctl calls must be
supported. If any information set or fetched by those ioctl calls is not supported by the
modules and driver downstream, some of the V7/4BSD/XENIX functions may not be
supported. For example, if the CBAUD bits in the c_cflag field are not supported, the functions
provided by the sg_ispeed and sg_ospeed fields of the sgttyb structure (see below) will not
be supported. If the TCFLSH ioctl is not supported, the function provided by the TIOCFLUSH
ioctl will not be supported. If the TCXONC ioctl is not supported, the functions provided by
the TIOCSTOP and TIOCSTART ioctl calls will not be supported. If the TIOCMBIS and TIOCMBIC
ioctl calls are not supported, the functions provided by the TIOCSDTR and TIOCCDTR ioctl
calls will not be supported.
The basic ioctl calls use the sgttyb structure defined by <sys/ttold.h> (included by
<sys/ioctl.h>):
struct sgttyb {
char sg_ispeed;
char sg_ospeed;
char sg_erase;
char sg_kill;
int sg_flags;
};

The sg_ispeed and sg_ospeed fields describe the input and output speeds of the device. If the
speed set on the device is over B38400, then it is reported as B38400 for compatibility reasons.
If it is set to B38400 and the current speed is over B38400, the change is ignored. See
TIOCGETP and TIOCSETP below. The sg_erase and sg_kill fields of the argument
structure specify the erase and kill characters respectively, and reflect the values in the
VERASE and VKILL members of the c_cc field of the termios structure.
The sg_flags field of the argument structure contains several flags that determine the
system's treatment of the terminal. They are mapped into flags in fields of the terminal state,
represented by the termios structure.

916 man pages section 7: Device and Network Interfaces • Last Revised 2 Oct 2001
 ttcompat(7M)

Delay type 0 (NL0, TAB0, CR0, FF0, BS0) is always mapped into the equivalent delay type 0 in
the c_oflag field of the termios structure. Other delay mappings are performed as follows:

sg_flags c_oflag

BS1 BS1

FF1 VT1

CR1 CR2

CR2 CR3

CR3 CR0 (not supported)

TAB1 TAB1

TAB2 TAB2

XTABS TAB3

NL1 ONLRET|CR1

NL2 NL1

NL3 NL0 (not supported)

If previous TIOCLSET or TIOCLBIS ioctl calls have not selected LITOUT or PASS8 mode, and if
RAW mode is not selected, the ISTRIP flag is set in the c_iflag field of the termios structure,
and the EVENP and ODDP flags control the parity of characters sent to the terminal and accepted
from the terminal, as follows:
0 (neither EVENP nor ODDP) Parity is not to be generated on output or checked on
input. The character size is set to CS8 and the PARENB flag is
cleared in the c_cflag field of the termios structure.
EVENP Even parity characters are to be generated on output and
accepted on input. The INPCK flag is set in the c_iflag
field of the termios structure, the character size is set to
CS7 and the PARENB flag is set in the c_iflag field of the
termios structure.
ODDP Odd parity characters are to be generated on output and
accepted on input. The INPCK flag is set in the c_iflag, the
character size is set to CS7 and the PARENB and PARODD flags
are set in the c_iflag field of the termios structure.
EVENP|ODDP or ANYP Even parity characters are to be generated on output and
characters of either parity are to be accepted on input. The
INPCK flag is cleared in the c_iflag field, the character size

Device and Network Interfaces 917

ttcompat(7M)

is set to CS7 and the PARENB flag is set in the c_iflag field
of the termios structure.

The RAW flag disables all output processing (the OPOST flag in the c_oflag field, and the XCASE
and IEXTEN flags in the c_iflag field are cleared in the termios structure) and input
processing (all flags in the c_iflag field other than the IXOFF and IXANY flags are cleared in the
termios structure). Eight bits of data, with no parity bit are accepted on input and generated
on output; the character size is set to CS8 and the PARENB and PARODD flags are cleared in the
c_cflag field of the termios structure. The signal-generating and line-editing control
characters are disabled by clearing the ISIG and ICANON flags in the c_iflag field of the
termios structure.

The CRMOD flag turns input carriage return characters into linefeed characters, and output
linefeed characters to be sent as a carriage return followed by a linefeed. The ICRNL flag in the
c_iflag field, and the OPOST and ONLCR flags in the c_oflag field, are set in the termios
structure.

The LCASE flag maps upper-case letters in the ASCII character set to their lower-case
equivalents on input (the IUCLC flag is set in the c_iflag field), and maps lower-case letters in
the ASCII character set to their upper-case equivalents on output (the OLCUC flag is set in the
c_oflag field). Escape sequences are accepted on input, and generated on output, to handle
certain ASCII characters not supported by older terminals (the XCASE flag is set in the c_lflag
field).

Other flags are directly mapped to flags in the termios structure:

sg_flags Flags in termios structure

CBREAK Complement of ICANON in c_lflag field

ECHO ECHO in c_lflag field

TANDEM IXOFF in c_iflag field

Another structure associated with each terminal specifies characters that are special in both
the old Version 7 and the newer 4BSD terminal interfaces. The following structure is defined
by <sys/ttold.h>:
struct tchars {
char t_intrc; /* interrupt */
char t_quitc; /* quit */
char t_startc; /* start output */
char t_stopc; /* stop output */
char t_eofc; /* end-of-file */
char t_brkc; /* input delimiter (like nl) */
};

918 man pages section 7: Device and Network Interfaces • Last Revised 2 Oct 2001
 ttcompat(7M)

XENIX defines the tchar structure as tc. The characters are mapped to members of the c_cc
field of the termios structure as follows:
tchars c_cc index
t_intrc VINTR
t_quitc VQUIT
t_startc VSTART
t_stopc VSTOP
t_eofc VEOF
t_brkc VEOL

Also associated with each terminal is a local flag word (TIOCLSET and TIOCLGET), specifying
flags supported by the new 4BSD terminal interface. Most of these flags are directly mapped to
flags in the termios structure:

Local flags Flags in termios structure

LCRTBS Not supported

LPRTERA ECHOPRT in the c_lflag field

LCRTERA ECHOE in the c_lflag field

LTILDE Not supported

LMDMBUF Not supported

LTOSTOP TOSTOP in the c_lflag field

LFLUSHO FLUSHO in the c_lflag field

LNOHANG CLOCAL in the c_cflag field

LCRTKIL ECHOKE in the c_lflag field

LPASS8 CS8 in the c_cflag field

LCTLECH CTLECH in the c_lflag field

LPENDIN PENDIN in the c_lflag field

LDECCTQ Complement of IXANY in the c_iflag field

LNOFLSH NOFLSH in the c_lflag field

Each flag has a corresponding equivalent sg_flags value. The sg_flags definitions omit the
leading "L”; for example, TIOCSETP with sg_flags set to TOSTOP is equivalent to
TIOCLSET with LTOSTOP.

Another structure associated with each terminal is the ltchars structure which defines
control characters for the new 4BSD terminal interface. Its structure is:

Device and Network Interfaces 919

ttcompat(7M)

struct ltchars {
char t_suspc; /* stop process signal */
char t_dsuspc; /* delayed stop process signal */
char t_rprntc; /* reprint line */
char t_flushc; /*flush output (toggles) */
char t_werasc; /* word erase */
char t_lnextc; /* literal next character */
};

The characters are mapped to members of the c_cc field of the termios structure as follows:

ltchars c_cc index

t_suspc VSUS

t_dsuspc VDSUSP

t_rprntc VREPRINT

t_flushc VDISCARD

t_werasc VWERASE

t_lnextc VLNEXT

ioctls ttcompat responds to the following ioctl calls. All others are passed to the module below.
TIOCGETP The argument is a pointer to an sgttyb structure. The current terminal state is
fetched; the appropriate characters in the terminal state are stored in that
structure, as are the input and output speeds. If the speed is over B38400, then
B38400 is returned. The values of the flags in the sg_flags field are derived
from the flags in the terminal state and stored in the structure.
TIOCEXCL Set ‘‘exclusive-use'' mode; no further opens are permitted until the file has
been closed.
TIOCNXCL Turn off ‘‘exclusive-use'' mode.
TIOCSETP The argument is a pointer to an sgttyb structure. The appropriate characters
and input and output speeds in the terminal state are set from the values in
that structure, and the flags in the terminal state are set to match the values of
the flags in the sg_flags field of that structure. The state is changed with a
TCSETSF ioctl so that the interface delays until output is quiescent, then
throws away any unread characters, before changing the modes. If the current
device speed is over B38400 for either input or output speed, and B38400 is
specified through this interface for that speed, the actual device speed is not
changed. If the device speed is B38400 or lower or if some speed other than
B38400 is specified, then the actual speed specified is set.

920 man pages section 7: Device and Network Interfaces • Last Revised 2 Oct 2001
 ttcompat(7M)

TIOCSETN The argument is a pointer to an sgttyb structure. The terminal state is

changed as TIOCSETP would change it, but a TCSETS ioctl is used, so that the
interface neither delays nor discards input.
TIOCHPCL The argument is ignored. The HUPCL flag is set in the c_cflag word of the
terminal state.
TIOCFLUSH The argument is a pointer to an int variable. If its value is zero, all characters
waiting in input or output queues are flushed. Otherwise, the value of the int
is treated as the logical OR of the FREAD and FWRITE flags defined by
<sys/file.h>. If the FREAD bit is set, all characters waiting in input queues are
flushed, and if the FWRITE bit is set, all characters waiting in output queues are
flushed.
TIOCSBRK The argument is ignored. The break bit is set for the device. (This is not
supported by ttcompat. The underlying driver must support TIOCSBRK.)
TIOCCBRK The argument is ignored. The break bit is cleared for the device. (This is not
supported by ttcompat. The underlying driver must support TIOCCBRK.)
TIOCSDTR The argument is ignored. The Data Terminal Ready bit is set for the device.
TIOCCDTR The argument is ignored. The Data Terminal Ready bit is cleared for the
device.
TIOCSTOP The argument is ignored. Output is stopped as if the STOP character had been
typed.
TIOCSTART The argument is ignored. Output is restarted as if the START character had
been typed.
TIOCGETC The argument is a pointer to a tchars structure. The current terminal state is
fetched, and the appropriate characters in the terminal state are stored in that
structure.
TIOCSETC The argument is a pointer to a tchars structure. The values of the appropriate
characters in the terminal state are set from the characters in that structure.
TIOCLGET The argument is a pointer to an int. The current terminal state is fetched, and
the values of the local flags are derived from the flags in the terminal state and
stored in the int pointed to by the argument.
TIOCLBIS The argument is a pointer to an int whose value is a mask containing flags to
be set in the local flags word. The current terminal state is fetched, and the
values of the local flags are derived from the flags in the terminal state; the
specified flags are set, and the flags in the terminal state are set to match the
new value of the local flags word.
TIOCLBIC The argument is a pointer to an int whose value is a mask containing flags to
be cleared in the local flags word. The current terminal state is fetched, and the

Device and Network Interfaces 921

ttcompat(7M)

values of the local flags are derived from the flags in the terminal state; the
specified flags are cleared, and the flags in the terminal state are set to match
the new value of the local flags word.
TIOCLSET The argument is a pointer to an int containing a new set of local flags. The
flags in the terminal state are set to match the new value of the local flags word.
(This ioctl was added because sg_flags was once a 16 bit value. The local
modes controlled by TIOCLSET are equivalent to the modes controlled by
TIOCSETP and sg_flags.)
TIOCGLTC The argument is a pointer to an ltchars structure. The values of the
appropriate characters in the terminal state are stored in that structure.
TIOCSLTC The argument is a pointer to an ltchars structure. The values of the
appropriate characters in the terminal state are set from the characters in that
structure.
FIORDCHK Returns the number of immediately readable characters. The argument is
ignored. (This ioctl is handled in the stream head, not in the ttcompat
module.)
FIONREAD Returns the number of immediately readable characters in the int pointed to
by the argument. (This ioctl is handled in the stream head, not in the
ttcompat module.)

The following ioctls are returned as successful for the sake of compatibility. However, nothing
significant is done (that is, the state of the terminal is not changed in any way, and no message
is passed through to the underlying tty driver).

DIOCSETP
DIOCSETP
DIOCGETP
LDCLOSE
LDCHG
LDOPEN
LDGETT
LDSETT
TIOCGETD
TIOCSETD

The following old ioctls are not supported by ttcompat, but are supported by Solaris tty
drivers. As with all ioctl not otherwise listed in this documentation, these are passed through
to the underlying driver and are handled there.
TIOCREMOTE
TIOCGWINSZ
TIOCSWINSZ

922 man pages section 7: Device and Network Interfaces • Last Revised 2 Oct 2001
 ttcompat(7M)

The following ioctls are not supported by ttcompat, and are generally not supported by
Solaris tty drivers. They are passed through, and the tty drivers return EINVAL.
LDSMAP
LDGMAP
LDNMAP
TIOCNOTTY
TIOCOUTQ

(Note: LDSMAP, LDGMAP, and LDNMAP are defined in <sys/termios.h>.)

See Also ioctl(2), termios(3C), ldterm(7M), termio(7I)

Device and Network Interfaces 923

tty(7D)

Name tty – controlling terminal interface

Description The file /dev/tty is, in each process, a synonym for the control terminal associated with the
process group of that process, if any. It is useful for programs or shell sequences that wish to be
sure of writing messages on the terminal no matter how output has been redirected. It can also
be used for programs that demand the name of a file for output, when typed output is desired
and it is tiresome to find out what terminal is currently in use.

Files /dev/tty

/dev/tty*

See Also ports(1M), console(7D)

924 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 ttymux(7D)

Name ttymux – Serial I/O multiplexing STREAMS device driver

Synopsis multiplexer@0,0:input
multiplexer@0,0:output

Description ttymux is a STREAMS multiplexer driver that connects multiple serial devices to the system
console. Using this driver, input from multiple physical devices can be multiplexed onto a
single input stream for the system console. Output written to the console can be distributed to
multiple physical devices to provide redundant console interfaces to a system. Input and
output can be multiplexed to or from a separate list of devices.

ttymux is a STREAMS multiplexer for serial drivers (such as se(7D)) that comply with the
Solaris terminal subsystem interface.

Currently, multiplexer interfaces are provided for system console I/O only and not for general
serial I/O multiplexing. Multiplexer interfaces are currently not available for all platforms.
Please see NOTES.
Files /kernel/drv/sparcv9/ttymux 64– bit ELF kernel module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC (NetraCT series only)

See Also se(7D), termio(7I)

Writing Device Drivers

Notes Successful loading of this driver and its services depends on the EEPROM or NVRAM settings
in effect at the most recent system reboot. Without the platform firmware support, this feature
cannot be enabled. Currently, this support is provided only on a NetraCT product family.

Use caution when enabling this feature to perform console input multiplexing, particularly
during super-user login. Because no security measures are enabled when the driver is in
operation, you must clearly understand the security implications involved in using this feature
and take appropriate measures to provide maximum protection to the host. This can include
such steps as enabling input to physically secured console devices only.

The ttymux driver does not handle the behavioral differences in control characteristics of
different terminal types (for example, an ESCAPE sequence.) As a result, multiple terminal
types are not supported simultaneously. Please refer to the platform user guide for more
information.

Device and Network Interfaces 925

tzmon(7d)

Name tzmon – ACPI Thermal Zone Monitor

Description The tzmon is a pseudo driver that serves as an ACPI thermal zone monitor. Thermal zones are
logical regions within a computer system for which ACPI performs temperature monitoring
and control functions. The number of thermal zones on a system with ACPI support varies.
For example, some systems may have one or more thermal zones, while others may have none.
See the Advanced Configuration and Power Interface Specification, (ACPI) Version 3.0A. for
more details.

The tzmon handles thermal Zone events from ACPI and polls the temperature for each zone
exposed by the ACPI implementation. If threshold temperatures are reached, tzmon takes
appropriate action. For example, if the temperature is sufficiently high and the ACPI
implementation supports it, tzmon initiates system shutdown.

Note that by default, system temperature control functions are usually performed by the BIOS
and may supersede tzmon functions, depending on the BIOS implementation. Also, many
ACPI implementations expose no thermal zones and in these cases, tzmon performs no
functions.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWckr

Architecture x86/x64 only

Interface stability Private

See Also attributes(5)

Advanced Configuration and Power Interface Specification, (ACPI), Version 3.0A.

926 man pages section 7: Device and Network Interfaces • Last Revised 31 Oct 2006
 uata(7D)

Name uata – IDE Host Bus Adapter Driver

Synopsis ide@unit-address

Description The uata host bus adapter driver is a nexus driver that supports the ide interface on SPARC
platforms. The driver attempts to set the disk and ATAPI CD-ROM drive to maximum
supported speed. The uata driver supports ultra DMA mode-4 (ATA66).

Currently, the uata driver supports the CMD646U, Sil680a and Acer Southbridge M5229 IDE
controllers. The uata driver supports two channels concurrently with two devices connected
per channel. The devices are logically numbered from 0 to 3:
0 Master disk on primary channel.
1 Slave disk on primary channel.
2 Master disk on secondary channel.
3 Slave disk on secondary channel.

For ATAPI devices, an ATAPI DRIVE RESET command is issued to facilitate recovery from
timeouts and errors. The BSY bit of the drive's status register is polled to check for the drive
reset completion. If the drive reset fails, a warning message is displayed and the recovery
process continues. This logic is subject to change.

To control the maximum time spent waiting for the ATAPI drive reset to complete, the
atapi-device-reset-waittime tunable property is available through the
/kernel/drv/uata.conf file. The default and maximum/minimum values are shown below.
Please see /kernel/drv/uata.conf for more info.
Default value: 3000000
Minimum value: 20
Maximum value: 3000000

The atapi-device-reset-waittime property units are in microseconds.

Files /kernel/drv/uata 32–bit ELF kernel module.
/kernel/drv/uata.conf Driver configuration file.

See Also prtconf(1M), driver.conf(4), attributes(5)

Writing Device Drivers

X3T10 ATA-4 specifications

Diagnostics In addition to being logged, the following messages may appear on the system console:
ddi_get_iblock_cookie failed.
The driver could not obtain the interrupt cookie. The attach may fail.

Device and Network Interfaces 927

uata(7D)

Drive not ready before set_features.

Indicates a fatal problem. The drives are not ready to be programmed and features cannot
be set. (During the driver initialization process, the driver must set the features for the
drive, including dma and pio).
Error set after issuing Set Feature command.
Indicates a fatal problem. The driver's error bit was set after the set feature command was
issued. (During the driver initialization process, the driver must set the features for the
drive, including dma and pio).
Interrupt not seen after set_features.
Indicates a fatal problem with the drive. Features cannot be set.
ata_controller - set features failed.
Indicates a fatal problem with the drive. Features cannot be set.
? target <number> lun 0.
Displayed at boot up time to indicate that the target <number> was identified, where
<number> is a decimal value.
resid
Residual number of bytes in data transfer and the I/O operation could not be finished
completely.
ghd_timer_newstate: HBA reset failed.
Generally indicates a fatal condition. I/O operation cannot be completed following reset of
the channel.
timeout: <message> chno =<number> target=<number>.
A timeout occured because of <message> on device (target=<number>) on channel (chno
=<number). Where <message> could be either early abort, early timeout, abort request,
abort device, reset target or reset bus.
ata_controller - Drive not ready before command <number>.
The drive did not respond before issuing the command <number> to the controller;
command <number> will not be issued to the drive. (<number> is the hexadecimal opcode
for the sleep or standby commands, which are issued when the drive transitions between
power management states).
ATAPI drive reset failed for target: <number>;Continuing the recovery process.
If this message is displayed after you modify /kernel/drv/uata.conf, try to increase the
atapi-device-reset-waittime property value within the maximum value allowed,
otherwise contact Sun support.
ata_controller - Command <number> failed.
Command <number> failed on the drive. (<number> is the hexadecimal opcode for the
sleep or standby commands, which are issued when the drive transitions between power
management states).

928 man pages section 7: Device and Network Interfaces • Last Revised 20 April 2005
 uata(7D)

ata_controller - Command <number> returned error.

The command returned an error. (<number> is the hexadecimal opcode for the sleep or
standby commands, which are issued when the drive transitions between power
management states).
ata_controller - Cannot take drive <number> to sleep.
The disk will not transition to sleep state. (Indicates that the driver could not set the device
to sleep mode while performing power management functions).
ata_controller - Cannot reset secondary/primary channel.
The disk will not transition from sleep to active state.
ata_controller - Unsupported Controller Vendor 0x13d0, Device 0x43f1, Revision 0x034.
An unsupported ata controller was found on the system and prints <ID>, device id and
revision of the controller, where <ID> represents the hexidecimal vendor ID.
Changing the mode of targ: <number> to Ultra DMA mode: <number>.
For the timedout command, the driver attempts to recover by changing speed to lower
values and retrying the command. This message indicates to which mode the driver is
attempting to re-program the drive, where <number> is a decimal value.
Changing the mode of targ: <number> to Multi DMA mode: <number>.
For the timedout command, the driver attempts to recover by changing speed to lower
values and retrying the command. This message indicates to which mode the driver is
attempting to re-program the drive, where <number> is a decimal value.

These messages are informational and indicate that a timeout occured for a I/O request. The
uata driver recovers from these states automatically unless there is a fatal error.

Device and Network Interfaces 929

uath(7D)

Name uath – Atheros AR5523 USB IEEE802.11a/b/g Wireless Network Driver

Description The uath IEEE802.11a/b/g wireless network driver is a multi-threaded, loadable, clonable,
GLDv3-based STREAMS driver supporting Atheros AR5523 USB IEEE802.11a/b/g wireless
network driver.

Configuration The uath driver performs auto-negotiation to determine the data rate and mode. The driver
supports only BSS networks (also known as ap or infrastructure networks) and open (or
open-system) or shared system authentication. For wireless security, WEP encryption,
WPA-PSk, and WPA2-PSK are currently supported. You can perform configuration and
administration tasks using the dladm(1M) and wificonfig(1M) utilities.
Files /dev/uath Special character device
/kernel/drv/uath 32-bit ELF 86 kernel module (x86)
/kernel/drv/amd64/uat 64-bit ELF kernel module (x86)
/kernel/misc/uathfw 32-bit ELF firmware kernel module (x86)
/kernel/misc/amd64/uathfw 64-bit ELF firmware kernel module (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWuath

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), dlpi(7P), gld(7D)

IEEE802.11b/g - Wireless LAN Standard - IEEE, 2003

930 man pages section 7: Device and Network Interfaces • Last Revised 10 Sep 2009
 udfs(7FS)

Name udfs – universal disk format file system

Description The udfs file system is a file system type that allows user access to files on Universal Disk
Format (UDF) disks from within the Solaris operating environment. Once mounted, a udfs
file system provides standard Solaris file system operations and semantics. That is, users can
read files, write files, and list files in a directory on a UDF device and applications can use
standard UNIX system calls on these files and directories.

Because udfs is a platform-independent file system, the same media can be written to and read
from by any operating system or vendor.

Mounting File Systems udfs file systems are mounted using:

mount-F udfs -o rw/ro device-special

Use:
mount /udfs

if the /udfs and device special file /dev/dsk/c0t6d0s0 are valid and the following line (or
similar line) appears in your /etc/vfstab file:
/dev/dsk/c0t6d0s0 - /udfs udfs - no ro

The udfs file system provides read-only support for ROM, RAM, and sequentially–recordable
media and read-write support on RAM media.

The udfs file system also supports regular files, directories, and symbolic links, as well as
device nodes such as block, character, FIFO, and Socket.

See Also mount(1M), mount_udfs(1M), vfstab(4)

Notes Invalid characters such as “NULL” and "/” and invalid file names such as "." and ".." will be
translated according to the following rule:

Replace the invalid character with an “_," then append the file name with # followed by a 4
digit hex representation of the 16-bit CRC of the original FileIdentifier. For example, the
file name ".." will become "__#4C05"

Device and Network Interfaces 931

udp(7P)

Name udp, UDP – Internet User Datagram Protocol

Synopsis #include <sys/socket.h>
#include <netinet/in.h>
s = socket(AF_INET, SOCK_DGRAM, 0);
s = socket(AF_INET6, SOCK_DGRAM, 0);
t = t_open("/dev/udp", O_RDWR);
t = t_open("/dev/udp6", O_RDWR);

Description UDP is a simple datagram protocol which is layered directly above the Internet Protocol (“IP”)
or the Internet Protocol Version 6 (“IPv6”). Programs may access UDP using the socket
interface, where it supports the SOCK_DGRAM socket type, or using the Transport Level Interface
(“TLI”), where it supports the connectionless (T_CLTS) service type.

Within the socket interface, UDP is normally used with the sendto(), sendmsg(),
recvfrom(), and recvmsg() calls (see send(3SOCKET) and recv(3SOCKET)). If the
connect(3SOCKET) call is used to fix the destination for future packets, then the
recv(3SOCKET) or read(2) and send(3SOCKET) or write(2) calls may be used.

UDP address formats are identical to those used by the Transmission Control Protocol
(“TCP”). Like TCP, UDP uses a port number along with an IPor IPv6 address to identify the
endpoint of communication. The UDP port number space is separate from the TCP port
number space, that is, a UDP port may not be “connected” to a TCP port. The
bind(3SOCKET) call can be used to set the local address and port number of a UDP socket.
The local IP or IPv6 address may be left unspecified in the bind() call by using the special
value INADDR_ANY for IP, or the unspecified address (all zeroes) for IPv6. If the bind() call is
not done, a local IP or IPv6 address and port number will be assigned to the endpoint when
the first packet is sent. Broadcast packets may be sent, assuming the underlying network
supports this, by using a reserved “broadcast address" This address is network interface
dependent. Broadcasts may only be sent by the privileged user.

Note that no two UDP sockets can be bound to the same port unless the bound IP addresses
are different. IPv4 INADDR_ANY and IPv6 unspecified addresses compare as equal to any IPv4
or IPv6 address. For example, if a socket is bound to INADDR_ANY or unspecified address and
port X, no other socket can bind to port X, regardless of the binding address. This special
consideration of INADDR_ANY and unspecified address can be changed using the SO_REUSEADDR
socket option. If SO_REUSEADDR is set on a socket doing a bind, IPv4 INADDR_ANY and IPv6
unspecified address do not compare as equal to any IP address. This means that as long as the
two sockets are not both bound to INADDR_ANY/unspecified address or the same IP address, the
two sockets can be bound to the same port.

If an application does not want to allow another socket using the SO_REUSEADDR option to bind
to a port its socket is bound to, the application can set the socket level option SO_EXCLBIND on

932 man pages section 7: Device and Network Interfaces • Last Revised 4 July 2006
 udp(7P)

a socket. The option values of 0 and 1 represent enabling and disabling the option,
respectively. Once this option is enabled on a socket, no other socket can be bound to the same
port.

IPv6 does not support broadcast addresses; their function is supported by IPv6 multicast
addresses.

Options at the IP level may be used with UDP. See ip(7P) or ip6(7P). Additionally, there is
one UDP-level option of interest to IPsec Key Management applications (see ipsec(7P)and
pf_key(7P)):
UDP_NAT_T_ENDPOINT
If this boolean option is set, datagrams sent via this socket will have a non-ESP marker
inserted between the UDP header and the data. Likewise, inbound packets that match the
endpoint's local-port will be demultiplexed between ESP or the endpoint itself if a non-ESP
marker is present. This option is only available on IPv4 sockets (AF_INET), and the
application must have sufficient privilege to use PF_KEY sockets to also enable this option.

There are a variety of ways that a UDP packet can be lost or corrupted, including a failure of
the underlying communication mechanism. UDP implements a checksum over the data
portion of the packet. If the checksum of a received packet is in error, the packet will be
dropped with no indication given to the user. A queue of received packets is provided for each
UDP socket. This queue has a limited capacity. Arriving datagrams which will not fit within its
high-water capacity are silently discarded.

UDP processes Internet Control Message Protocol (“ICMP”) and Internet Control Message
Protocol Version 6 (“ICMP6”) error messages received in response to UDP packets it has sent.
See icmp(7P) and icmp6(7P).

ICMP “source quench” messages are ignored. ICMP “destination unreachable,” “time
exceeded” and “parameter problem” messages disconnect the socket from its peer so that
subsequent attempts to send packets using that socket will return an error. UDP will not
guarantee that packets are delivered in the order they were sent. As well, duplicate packets
may be generated in the communication process.

ICMP6 “destination unreachable” packets are ignored unless the enclosed code indicates that
the port is not in use on the target host, in which case, the application is notified. ICMP6
“parameter problem” notifications are similarly passed upstream. All other ICMP6 messages
are ignored.

See Also read(2), write(2), bind(3SOCKET), connect(3SOCKET), recv(3SOCKET),

send(3SOCKET), icmp(7P), icmp6(7P), inet(7P), inet6(7P), ip(7P), ipsec(7P), ip6(7P),
pf_key(7P), tcp(7P)

Postel, Jon, RFC 768, User Datagram Protocol, Network Information Center, SRI
International, Menlo Park, Calif., August 1980

Device and Network Interfaces 933

udp(7P)

Huttunen, A., Swander, B., Volpe, V., DiBurro, L., Stenberg, M., RFC 3948, UDP
Encapsulation of IPsec ESP Packets, The Internet Society, 2005.

Diagnostics A socket operation may fail if:

EISCONN A connect() operation was attempted on a socket on which a connect()
operation had already been performed, and the socket could not be
successfully disconnected before making the new connection.
EISCONN A sendto() or sendmsg() operation specifying an address to which the
message should be sent was attempted on a socket on which a connect()
operation had already been performed.
ENOTCONN A send() or write() operation, or a sendto() or sendmsg() operation
not specifying an address to which the message should be sent, was
attempted on a socket on which a connect() operation had not already
been performed.
EADDRINUSE A bind() operation was attempted on a socket with a network
address/port pair that has already been bound to another socket.
EADDRNOTAVAIL A bind() operation was attempted on a socket with a network address for
which no network interface exists.
EINVAL A sendmsg() operation with a non-NULL msg_accrights was
attempted.
EACCES A bind() operation was attempted with a “reserved” port number and the
effective user ID of the process was not the privileged user.
ENOBUFS The system ran out of memory for internal data structures.

934 man pages section 7: Device and Network Interfaces • Last Revised 4 July 2006
 ufs(7FS)

Name ufs – UFS file system

Synopsis #include <sys/param.h>
#include <sys/types.h>
#include <sys/fs/ufs_fs.h>
#include <sys/fs/ufs_inode.h>

Description UFS is the default disk-based file system for the Solaris environment. The UFS file system is
hierarchical, starting with its root directory (/) and continuing downward through a number
of directories. The root of a UFS file system is inode 2. A UFS file system's root contents
replace the contents of the directory upon which it is mounted.

Subsequent sections of this manpage provide details of the UFS file systems.

State Flags (fs_state UFS uses state flags to identify the state of the file system. fs_state is FSOKAY - fs_time.
and fs_clean) fs_time is the timestamp that indicates when the last system write occurred. fs_state is
updated whenever fs_clean changes. Some fs_clean values are:
FSCLEAN Indicates an undamaged, cleanly unmounted file system.
FSACTIVE Indicates a mounted file system that has modified data in memory. A mounted
file system with this state flag indicates that user data or metadata would be lost
if power to the system is interrupted.
FSSTABLE Indicates an idle mounted file system. A mounted file system with this state flag
indicates that neither user data nor metadata would be lost if power to the
system is interrupted.
FSBAD Indicates that this file system contains inconsistent file system data.
FSLOG Indicates that the file system has logging enabled. A file system with this flag set
is either mounted or unmounted. If a file system has logging enabled, the only
flags that it can have are FSLOG or FSBAD. A non-logging file system can have
FSACTIVE, FSSTABLE, or FSCLEAN.

It is not necessary to run the fsck command on unmounted file systems with a state of
FSCLEAN, FSSTABLE, or FSLOG. mount(2) returns ENOSPC if an attempt is made to mount a UFS
file system with a state of FSACTIVE for read/write access.

As an additional safeguard, fs_clean should be trusted only if fs_state contains a value

equal to FSOKAY - fs_time, where FSOKAY is a constant integer defined in the
/usr/include/sys/fs/ufs_fs.h file. Otherwise, fs_clean is treated as though it contains
the value of FSACTIVE.

Extended Fundamental Extended Fundamental Types (EFT) provide 32-bit user ID (UID), group ID (GID), and
Types (EFT) device numbers.

Device and Network Interfaces 935

ufs(7FS)

If a UID or GID contains an extended value, the short variable (ic_suid, ic_sgid) contains
the value 65535 and the corresponding UID or GID is in ic_uid or ic_gid. Because numbers
for block and character devices are stored in the first direct block pointer of the inode
(ic_db[0]) and the disk block addresses are already 32 bit values, no special encoding exists
for device numbers (unlike UID or GID fields).

Multiterabyte File A multiterabyte file system enables creation of a UFS file system up to approximately 16
System terabytes of usable space, minus approximately one percent overhead. A sparse file can have a
logical size of one terabyte. However, the actual amount of data that can be stored in a file is
approximately one percent less than one terabyte because of file system overhead.

On-disk format changes for a multiterabyte UFS file system include:

■ The magic number in the superblock changes from FS_MAGIC to MTB_UFS_MAGIC. For more
information, see the /usr/include/sys/fs/ufs_fs file.
■ The fs_logbno unit is a sector for UFS that is less than 1 terabyte in size and fragments for
a multiterabyte UFS file system.

UFS Logging UFS logging bundles the multiple metadata changes that comprise a complete UFS operation
into a transaction. Sets of transactions are recorded in an on-disk log and are applied to the
actual UFS file system's metadata.

UFS logging provides two advantages:

1. A file system that is consistent with the transaction log eliminates the need to run fsck
after a system crash or an unclean shutdown.
2. UFS logging often provides a significant performance improvement. This is because a file
system with logging enabled converts multiple updates to the same data into single
updates, thereby reducing the number of overhead disk operations.

The UFS log is allocated from free blocks on the file system and is sized at approximately 1
Mbyte per 1 Gbyte of file system, up to 256 Mbytes. The log size may be larger (up to a
maximum of 512 Mbytes), depending upon the number of cylinder groups present in the file
system. The log is continually flushed as it fills up. The log is also flushed when the file system
is unmounted or as a result of a lockfs(1M) command.

Mounting UFS File You can mount a UFS file system in various ways using syntax similar to the following:
Systems
1. Use mount from the command line:
# mount -F ufs /dev/dsk/c0t0d0s7 /export/home
2. Include an entry in the /etc/vfstab file to mount the file system at boot time:
/dev/dsk/c0t0d0s7 /dev/rdsk/c0t0d0s7 /export/home ufs 2 yes -
For more information on mounting UFS file systems, see mount_ufs(1M).

936 man pages section 7: Device and Network Interfaces • Last Revised 18 Jun 2009
 ufs(7FS)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Uncommitted

See Also df(1M), fsck(1M), fsck_ufs(1M), fstyp(1M), lockfs(1M), mkfs_ufs(1M), newfs(1M),

ufsdump(1M), ufsrestore(1M), tunefs(1M), mount(2), attributes(5)

Writing Device Drivers

Notes For information about internal UFS structures, see newfs(1M) and mkfs_ufs(1M). For
information about the ufsdump and ufsrestore commands, see ufsdump(1M),
ufsrestore(1M), and /usr/include/protocols/dumprestore.h.

If you experience difficulty in allocating space on the ufs filesystem, it may be due to
framentation. Fragmentation can occur when you do not have sufficient free blocks to satisfy
an allocation request even though df(1M) indicates that enough free space is available. (This
may occur because df only uses the available fragment count to calculate available space, but
the file system requires contiguous sets of fragments for most allocations). If you suspect that
you have exhausted contiguous fragments on your file system, you can use the fstyp(1M)
utility with the -v option. In the fstyp output, look at the nbfree (number of blocks free) and
nffree (number of fragments free) fields. On unmounted filesystems, you can use fsck(1M)
and observe the last line of output, which reports, among other items, the number of
fragments and the degree of fragmentation. To correct a fragmentation problem, run
ufsdump(1M) and ufsrestore(1M) on the ufs filesystem.

Device and Network Interfaces 937

ugen(7D)

Name ugen – USB generic driver

Synopsis Node Name@unit-address
#include <sys/usb/clients/ugen/usb_ugen.h>

Description ugen is a generic USBA (Solaris USB Architecture) compliant client character driver that
presents USB devices to applications through a standard open(2), close(2), read(2), write(2),
aioread(3C), aiowrite(3C) Unix interface. Uninterpreted raw data are transfered to and
from the device via file descriptors created for each USB endpoint. Status is obtained by
reading file descriptors created for endpoint and full device status.
ugen supports control, bulk, isochronous and interrupt (in and out) transfers. libusb(3LIB)
uses ugen to access devices that do not contain drivers (such as digital cameras and PDAs).
Refer to /usr/sfw/share/doc/libusb/libusb.txt for details.
Binding In general, no explicit binding of the ugen driver is necessary because usb_mid(7D) is the
default driver for devices without a class or vendor unique driver. usb_mid(7D) creates the
same logical device names as ugen, but only if no child interfaces are explicitly bound to ugen.
If it is necessary to bind ugen explicitly to a device or interface, the following section explains
the necessary steps.
ugen can bind to a device with one or more interfaces in its entirety, or to a single interface of
that device. The binding type depends on information that is passed to add_drv(1M) or
update_drv(1M).
An add_drv(1M) command binds ugen to a list of device types it is to control.
update_drv(1M) adds an additional device type to the list of device types being managed by
the driver.
Names used to bind drivers can be found in /var/adm/messages. When a device is on-lined
after hot insertion, and no driver is found, there will be an entry containing:
USB 2.0 device (usb<vid>,<pid>)...

where vid is the USB vendor identifier in hex and pid is the product identifier in hex supplied
by the device descriptor usb_dev_descr(9S).
When using ugen for the first time, you must add the driver utilizing add_drv(1M), using a
command of the following form:
Assuming that the vid is 472 and pid is b0b0:

add_drv -n -m ’* <device perms> <owner> <group>’

-i ’"usb472,b0b0"’ ugen

If the command fails with:

(ugen) already in use as a driver or alias.

...add the device using update_drv(1M):

938 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

update_drv -a -m ’* <device perms> <owner> <group>’

-i ’"usb472,b0b0"’ ugen

This binds ugen to the entire device.

If ugen only binds to one interface of the device, use the following driver_alias instead of
usb<vid>,<pid>:
usbif<vid>,<pid>.config<cfg value>.<interface number>

where cfg value is the value of bConfigurationValue in the configuration descriptor

(usb_cfg_descr(9S)). For example "usbif1234,4567.config1.0."

Note that you can use update_drv to also remove bindings. Please see update_drv(1M) for
more information.

After a successful add_drv or update_drv, remove the device and reinsert. Check with the
prtconf(1M) -D option to determine if ugen is successfully bound to the device and the nodes
created in /dev/usb/<vid>.<pid> (see below).

An example showing how to bind a child device representing interface 0 of configuration 1 of

a composite device follows:
update_drv -a -m ’* 0666 root sys’
-i ’"usbif472,b0b0.config1.0"’ ugen

Note that you can completely uninstall the ugen driver and delete it from the system by doing:
pkgrm SUNWugen

Any pkgadd of SUNWugen after the pkgrm reactivates any pre-existing ugen driver
device-bindings.

Any pre-existing ugen driver device-bindings are preserved across operating system upgrades.

Logical Device For each device or child device it manages, ugen creates one logical device name for
Name Format device-wide status and one logical device name for endpoint 0. ugen also creates logical device
names for all other endpoints within the device node's binding scope (interface or device),
plus logical device names for their status.

If separate ugen instances control different interfaces of the same device, the device-wide
status and endpoint logical device names created for each instance will share access to the
same source or endpoint pipes. For example, a device with two interfaces, each operated by
their own ugen instance, will show endpoint0 as if0cntrl0 to the first interface, and will
show it as if1cntrl0 to the second interface. Both of these logical device names share
endpoint0. Likewise for the same device, ugen makes the device-wide status available as
if0devstat to the first interface and as if1devstat to the second interface. if0devstat and
if1devstat both return the same data.

Device and Network Interfaces 939

ugen(7D)

Any ugen logical device name can be held open by only one user at a time, regardless of
whether the O_EXCL flag passed to open(2). When a single pipe or data source is shared by
multiple logical device names, such as if[0,1]cntrl0 or if[0,1]devstat above, more than one
logical device name sharing the pipe or data source can be open at a time. However, only one
user may access the shared pipe or data source at a time, regardless of the logical device name
used for access.

When ugen is bound to an entire device, the following logical device names are created (each
on a single line). N represents the instance number of the device type.
Endpoint 0 (default endpoint):

/dev/usb/<vid>.<pid>/<N>/cntrl0
/dev/usb/<vid>.<pid>/<N>/cntrl0stat

For example:

/dev/usb/472.b0b0/0/cntrl0
/dev/usb/472.b0b0/0/cntrl0stat

Configuration index 1, Endpoints > 0, alternate 0:

/dev/usb/<vid>.<pid>/<N>/if<interface#>
<in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/if<interface#>
<in|out|cntrl><endpoint#>stat

For example:

/dev/usb/472.b0b0/0/if0in1
/dev/usb/472.b0b0/0/if0in1stat

Configuration index 1, Endpoints > 0, alternate > 0:

/dev/usb/<vid>.<pid>/<N>/if<interface#>.
<alternate><in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/if<interface#>.
<alternate<in|out|cntrl><endpoint#>stat

For example:

/dev/usb/472.b0b0/0/if0.1in3
/dev/usb/472.b0b0/0/if0.1in3stat

Configuration index> 1, Endpoints > 0, alternate 0:

/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>
<in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>

940 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

<in|out|cntrl><endpoint#>stat

For example:

/dev/usb/472.b0b0/0/cfg2if0in1
/dev/usb/472.b0b0/0/cfg2if0in1stat

Note that the configuration value from the configuration

descriptor indexed by the configuration index is used in
the node name and not the configuration index itself.

Configuration index> 1, Endpoints > 0, alternate > 0:

/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>.
<alternate<in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>.
<alternate<in|out|cntrl><endpoint#>stat

For example:

/dev/usb/472.b0b0/0/cfg2if0.1in1
/dev/usb/472.b0b0/0/cfg2if0.1in1stat

Device status:

/dev/usb/<vid>.<pid>/<N>/devstat

For example:

/dev/usb/472.b0b0/0/devstat

When ugen is bound to a single device interface, the following logical device nodes are created:
Endpoint 0 (default endpoint):

/dev/usb/<vid>.<pid>/<N>/if<interface#>cntrl0
/dev/usb/<vid>.<pid>/<N>/if<interface#>cntrl0stat

For example:

/dev/usb/472.b0b0/0/if0cntrl0
/dev/usb/472.b0b0/0/if0cntrl0stat

Device status:
/dev/usb/<vid>.<pid>/<N>/if<interface#>devstat

For example:
/dev/usb/472.b0b0/0/if0devstat

Device and Network Interfaces 941

ugen(7D)

The format for all other logical device names is identical to the format used when ugen is
bound to the entire device.

Opening the endpoint of a different configuration or different alternate interface will cause an
implicit change of configuration or a switch to an alternate interface. A configuration change
is prohibited when any non-zero endpoint device nodes are open. An alternate interface
switch is prohibited if any endpoint in the same interface is open.

Hot-PLUGGING A device may be hot-removed at any time. Following hot-removal, the device status changes
to USB_DEV_STAT_DISCONNECTED, the status of open endpoints change to
USB_LC_STAT_DISCONNECTED upon their access, and all subsequent transfer requests
fail. Endpoints are reactivated by first reinserting the device and then closing and reopening
all endpoints that were open when the device was disconnected.

Cpr CPR (Checkpoint/Resume) may be initiated at any time and is treated similarly to a
(CHECKPOINT/RESUME)
hot-removal. Upon successful suspend and resume, all subsequent transfer requests fail as an
indication to the application to reinitialize. Applications should close and reopen all
endpoints to reinstate them. All endpoint and device status on Resume (before close and
reopen) is USB_LC_STAT_SUSPENDED. A system suspend will fail while ugen is performing
a transfer.

Device Power Devices which support remote wakeup can be power managed when they have no open logical
Management device nodes. When an application opens the first logical device node of a device, that
application should assume that a reinitialization of device state is required.

Device Status Applications can monitor device status changes by reading the device status from the device
Management status logical name. When opened without O_NONBLOCK and O_NDELAY, all reads from
that file descriptor (with the exception of the the intial read that follows the open) block until a
device status change occurs. Calls to read will always return immediately if opened with
O_NONBLOCK or O_NDELAY. Nonblocking calls to read which have no data to return, return no
error and zero bytes read.

Device statuses are:

USB_DEV_STAT_ONLINE Device is available.
USB_DEV_STAT_DISCONNECTED Device has been disconnected.
USB_DEV_STAT_RESUMED Device has been resumed, however, endpoints
which were open on suspend have not yet been
closed and reopened.
USB_DEV_STAT_UNAVAILABLE Device has been reconnected, however, endpoints
which were open on disconnect have not yet been
closed and reopened.

The following code reads the device status device logical name:

942 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

int fd;
int status;

if ((fd = open("/dev/usb/472.b0b0/0/devstat",
O_RDONLY)) < 0) {
/* handle error */
}

if (read(fd, &status, sizeof(status)) != sizeof(status)) {

/* handle error */
}

switch (status) {
case USB_DEV_STAT_DISCONNECTED:
printf ("Terminating as device has been disconnected.\n");
exit (0);

case USB_DEV_STAT_RESUMED:
case USB_DEV_STAT_UNAVAILABLE:
/*
* Close and reopen endpoints to reestablish device access,
* then reset device.
*/
break;

case USB_DEV_STAT_ONLINE:
default:
break;
}

Use poll(2) to block on several logical names simultaneously, including device status logical
names. Poll indicates when reading a logical name would return data. See poll(2) for details.
Calls to read may be done whether or not they follow calls to poll.

Endpoint Status Each data endpoint has a corresponding status logical name. Use the status logical name to
Management retrieve the state of the data endpoint, including detail on how its most recent transfer failed.
Reads of the status file descriptors always return immediately. See the ERRORS section for
more information on endpoint status values. All logical device name files created for returning
status must be opened with O_RDONLY.

The following code illustrates reading the status file descriptor of an endpoint which just failed
a data transfer in order to get more information on the failure.
int data_xfered, status;
int ep1_data_fd, ep1_stat_fd;
uchar_t request[8];

ep1_data_fd = open ("/dev/usb/472.b0b0/0/if0out1", O_WRONLY);

Device and Network Interfaces 943

ugen(7D)

if (ep1_data_fd < 0) {
/* Handle open error. */
}

ep1_stat_fd = open ("/dev/usb/472.b0b0/0/if0out1stat",

O_RDONLY);
if (ep1_stat_fd < 0) {
/* Handle open error. */
}

data_xfered = write(ep1_data_fd, request, sizeof (request));

/* An error occured during the data transfer. */

if (data_xfered != sizeof (request)) {

/* Read status file descriptor for details on failure. */

if (read(ep1_stat_fd, (int *)&status, sizeof (status)) !=
sizeof (status)) {
status = USB_LC_STAT_UNSPECIFIED_ERR;
}

/* Take appropriate action. */

switch (status) {
case USB_LC_STAT_STALL:
printf ("Endpoint stalled.\n");
break;
case ...
...
}

Control Transfers The control endpoint is typically used to set up the device and to query device status or
configuration.

Applications requiring I/O on a control endpoint should open the corresponding logical
device name and use regular UNIX I/O system calls. For example: read(2), write(2),
aioread(3C) and aiowrite(3C). poll(2) is not supported on control endpoints.

A control endpoint must be opened with O_RDWR since it is bidirectional. It cannot be opened
with O_NONBLOCK or O_NDELAY.

For example:
fd = open("/dev/usb/472.b0b0/0/cntrl0", O_RDWR);

944 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

fdstat = open("/dev/usb/472.b0b0/0/cntrl0stat", O_RDONLY);

Control endpoints can be read and written. A read operation receives data from the device
and a write operation sends data to the device.

To perform a control-IN transfer, perform a write(2) of USB setup data (see section 9.3 of the
USB 1.1 or 2.0 specifications) followed by a read(2) on the same control endpoint to fetch the
desired data. For example:
void init_cntrl_req(
uchar_t *req, uchar_t bmRequestType, uchar_t bRequest,
ushort_t wValue, ushort_t wIndex, ushort_t wLength) {
req[0] = bmRequestType;
req[1] = bRequest;
req[2] = 0xFF & wValue;
req[3] = 0xFF & (wValue >> 8);
req[4] = 0xFF & wIndex;
req[5] = 0xFF & (wIndex >> 8);
req[6] = 0xFF & wLength;
req[7] = 0xFF & (wLength >> 8);
}

....

uchar_t dev_descr_req[8];
usb_dev_descr_t descr;

init_cntrl_req(dev_descr_req,
USB_DEV_REQ_DEV_TO_HOST, USB_REQ_GET_DESCR,
USB_DESCR_TYPE_SETUP_DEV, 0, sizeof (descr));

count = write(fd, dev_descr_req, sizeof (dev_descr_req));

if (count != sizeof (dev_descr_req)) {
/* do some error recovery */
...
}

count = read(fd, &descr, sizeof (descr));

if (count != sizeof (descr)) {
/* do some error recovery */
}

The application can issue any number of reads to read data received on a control endpoint.
ugen successfully completes all reads, returning the number of bytes transferred. Zero is
returned when there is no data to transfer.

Device and Network Interfaces 945

ugen(7D)

If the read/write fails and returns –1, you can access the endpoint's status device logical name
for precise error information:
int status;

count = read(fdstat, &status, sizeof (status));

if (count == sizeof (status)) {
switch (status) {
case USB_LC_STAT_SUSPENDED:
case USB_LC_STAT_DISCONNECTED:
/* close all endpoints */
...
break;
default:
...
break;
}
}

Refer to the ERRORS section for all possible error values.

To perform a control-OUT transfer, send in a single transfer, the USB setup data followed by
any accompanying data bytes.
/* 1st 8 bytes of wbuf are setup. */
init_cntrl_req(wbuf, .......);

/* Data bytes begin at byte 8 of wbuf. */

bcopy(data, &wuf[8], sizeof (data));

/* Send it all in a single transfer. */

count = write(fd, wbuf, sizeof (wbuf));

A write(2) returns the number of bytes (both setup and data) actually transferred, (whether
or not the write is completely successful), provided that some data is actually transferred.
When no data is transferred, write(2) returns -1. Applications can read the corresponding
endpoint status to retrieve detailed error information. Note that it is an error to specify a size
different than:

(number of data bytes + number of setup bytes).

Here is a more extensive example which gets all descriptors of a device configuration. For sake
of brevity, uninteresting parts are omitted.
#include <sys/usb/usba.h>
#include <sys/usb/clients/ugen/usb_ugen.h>

uchar_t *config_cloud;
uchar_t *curr_descr;

946 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

uchar_t *bytes;

int curr_descr_len;
int curr_descr_type;

usb_cfg_descr_t cfg_descr;
usb_if_descr_t if_descr;
usb_ep_descr_t ep_descr;

/* See 9.13 of USB 2.0 spec for ordering. */

static char *pipetypes[] = {
"Control", "Isochronous", "Bulk", "Interrupt"
};

/*
* Setup to send a request to read just the config descriptor. The
* size of the whole cloud, containing all cfg, interface, endpoint,
* class and vendor-specific descriptors, will be returned as part of
* the config descriptor.
*/
init_cntrl_req(&setup_data, USB_DEV_REQ_DEV_TO_HOST, USB_REQ_GET_DESCR,
USB_DESCR_TYPE_SETUP_CFG, 0, USB_CFG_DESCR_SIZE);

/*
* Write setup data. USB device will prepare to return the whole
* config cloud as a response to this. We will read this separately.
*/
count = write(ctrl_fd, &setup_data, sizeof (setup_data));
if (count != sizeof (setup_data)) {
/* Error recovery. */
} else {
count = read(ctrl_fd, &cfg_descr, USB_CFG_DESCR_SIZE);
if (count != USB_CFG_DESCR_SIZE) {
/* Error recovery. */
}
}

/* USB data is little endian. */

bytes = (uchar_t *)(&cfg_descr.wTotalLength);
totalLength = bytes[0] + (bytes[1] << 8);

/*
* The size of the whole cloud is in the bLength field. Set up
* to read this amount of data, to get the whole cloud.
*/
config_cloud = malloc(totalLength);

Device and Network Interfaces 947

ugen(7D)

init_cntrl_req(&setup_data, USB_DEV_REQ_DEV_TO_HOST, USB_REQ_GET_DESCR,

USB_DESCR_TYPE_SETUP_CFG, 0, totalLength);

count = write(ctrl_fd, &setup_data, sizeof (setup_data));

if (count != sizeof (setup_data)) {
/* Error recovery. */
} else {
count = read(ctrl_fd, config_cloud, totalLength);
if (count != totalLength) {
/* Error recovery. */
}
}

/* Got the data. Now loop, dumping out the descriptors found. */

curr_descr = config_cloud;
offset = 0;
while (offset < totalLength) {

/* All descr have length and type at offset 0 and 1 */

curr_descr_len = curr_descr[0];
curr_descr_type = curr_descr[1];

switch (curr_descr_type) {
case USB_DESCR_TYPE_CFG:

/*
* Copy data into separate structure, needed for
* proper alignment of all non char fields. Note:
* non-char fields of all descriptors begin on aligned
* boundaries. The issue is that some structures may
* be adjacent to others which have an odd-numbered
* byte size, and may thus start on an odd-numbered
* boundary. */
bcopy(curr_descr, &cfg_descr, curr_descr_len);

/* Remember to read any words in endian-neutral way. */

(void) printf("\nConfig %d found.\n",

cfg_descr.bConfigurationValue);
break;

case USB_DESCR_TYPE_IF:
bcopy(curr_descr, &if_descr, curr_descr_len);
(void) printf("\n\tInterface %d, Alt %d found.\n",
if_descr.bInterfaceNumber,

948 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

if_descr.bAlternateSetting);
break;

case USB_DESCR_TYPE_EP:
bcopy(curr_descr, &ep_descr, curr_descr_len);
(void) printf("\n\t\tEndpoint %d (%s-%s) found.\n",
(ep_descr.bEndpointAddress & USB_EP_NUM_MASK),
(pipetypes[
ep_descr.bmAttributes & USB_EP_ATTR_MASK]),
((ep_descr.bEndpointAddress &
USB_EP_DIR_IN) ? "IN" : "OUT"));
break;

default:
(void) printf(
"\n\t\t\tOther descriptor found. Type:%d\n",
curr_descr_type);
break;
}

offset += curr_descr_len;
curr_descr = &config_cloud[offset];
}

Interrupt-IN Applications requiring data from an interrupt-IN endpoint should open the corresponding
Transfers logical device name and use read(2), aioread(3C) and poll(2) system calls.

An interrupt-IN endpoint must be opened with O_RDONLY. It can also be opened using
O_NONBLOCK or O_NDELAY if desired.
fd = open("/dev/usb/472.b0b0/0/if0in1", O_RDONLY);
fdstat = open("/dev/usb/472.b0b0/0/if0in1stat", O_RDONLY);

ugen starts polling interrupt—IN endpoints immediately upon opening them and stops
polling them upon closure. (Polling refers to interrogation of the device by the driver and
should not be confused with poll(2), which is an interrogation of the driver by the
application.)

A read(2) of an endpoint opened with the O_NONBLOCK or O_NDELAY flags set will not block
when there is insufficient data available to satisfy the request. The read simply returns what it
can without signifying any error.

Applications should continuously check for and consume interrupt data. ugen enables
buffering of up to one second of incoming data. In case of buffer overflow, ugen stops polling
the interrupt-IN endpoint until the application consumes all the data. In this case, a read(2) of
an empty buffer returns -1, sets the endpoint status to USB_LC_STAT_INTR_BUF_FULL (to

Device and Network Interfaces 949

ugen(7D)

indicate that the buffer had been full and polling had been stopped) and causes ugen to start
polling the endpoint again. To retrieve the status, the application can open and read the
corresponding endpoint's status device logical name.
for (;;) {
count = read(fd, buf, sizeof(buf));
if (count == -1) {
int cnt, status;

cnt = read(fdstat, &status, sizeof (status));

if (cnt == -1) {
/* more error recovery here */
} else {
switch (status) {
case USB_LC_STAT_INTR_BUF_FULL:
...
break;
default:
...
break;
}
}
}
/* process the data */
....
}

ugen will never drop data. However, the device may drop data if the application cannot read it
at the rate that it is produced.

Applications requiring unbuffered data from an interrupt-IN endpoint should open the
associated status endpoint with O_RDWR before opening the associated interrupt-IN
endpoint and write a control byte with USB_EP_INTR_ONE_XFER set. All other bits are
reserved and should be 0.

"One transfer" mode will persist until disabled explicitly after the associated interrupt-IN
endpoint has been closed by writing a control byte with USB_EP_INTR_ONE_XFER cleared.

"One transfer" mode is implicitly disabled when the status/control endpoint is closed.

Attempts to change the "one transfer" mode while the endpoint is open will result in EINVAL.

An application can open multiple interrupt-IN endpoints and can call poll(2) to monitor the
availability of new data. (Note: poll works with interrupt-IN data endpoints, not their status
endpoints.)
struct pollfd pfd[2];

950 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

bzero(pfd, sizeof (pfd));

pfd[0].fd = fd1; /* fd1 is one interrupt-IN endpoint. */
pfd[0].events = POLLIN;
pfd[1].fd = fd2; /* fd2 is another interrupt-IN endpoint. */
pfd[1].events = POLLIN;

for (;;) {
poll(pfd, 2, -1);

if (pfd[0].revents & POLLIN) {

count = read(fd1, buf, sizeof (buf));
....
}
if (pfd[1].revents & POLLIN) {
count = read(fd2, buf, sizeof (buf));
....
}
}

You can monitor the device status endpoint via poll(2) concurrently with the multiple
interrupt-IN endpoints. Simply add another pollfd element to the pfd array in the previous
code example, and initialize the new element's fd field with the file descriptor of the device
status endpoint (opened without O_NONBLOCK or O_NDELAY). Set the new element's
event field to POLLIN like the other elements. Note that only interrupt–IN endpoints and the
device status endpoint can be monitored using poll(2).

Interrupt-OUT Applications requiring output on an interrupt-OUT endpoint can open the corresponding
Transfers logical device name and perform regular UNIX I/O system calls such as write(2) and
aiowrite(3C).

An interrupt-OUT endpoint must be opened with O_WRONLY.

fd = open("/dev/usb/472.b0b0/0/if0out3", O_WRONLY);

fdstat = open("/dev/usb/472.b0b0/0/if0out3stat", O_RDONLY);

Data can be written to an interrupt-OUT endpoint as follows:

count = write(fd, buf, sizeof (buf)):
if (count == -1) {
/* error recovery */
}

Bulk Transfers Applications requiring I/O on a bulk endpoint can open the corresponding logical device
name and perform regular UNIX I/O system calls. For example: read(2), write(2),
aioread(3C) and aiowrite(3C). poll(2) is not supported on bulk endpoints.

Device and Network Interfaces 951

ugen(7D)

A bulk endpoint must be opened with O_RDONLY or O_WRONLY and cannot be opened with
O_NONBLOCK or O_NDELAY:
fd = open("/dev/usb/472.b0b0/0/if0in2", O_RDONLY);
fdstat = open("/dev/usb/472.b0b0/0/if0in2stat", O_RDONLY);

Data can be read from a bulk-IN endpoint as follows:

count = read(fd, buf, sizeof (buf)):
if (count == -1) {
/* error recovery */
}

Data can be written to a bulk-OUT endpoint as follows:

count = write(fd, buf, sizeof (buf)):

if (count == -1) {
/* error recovery */
}

Isochronous Applications requiring I/O on an isochronous endpoint can open the corresponding logical
Transfers device name and perform regular UNIX I/O system calls such as read(2), write(2), poll(2),
aioread(3C) and aiowrite(3C). An isochronous endpoint must be opened with O_RDWR.
fd = open("/dev/usb/472.b0b0/0/if0.3in2", O_RDWR);

fdstat = open("/dev/usb/472.b0b0/0/if0.3in2stat", O_RDONLY);

Applications can use the status logical name to retrieve the state of the isochronous data
endpoint, including details on why the most recent transfer failed.

Applications have the flexibility to specify the number of isochronous packets and the size of
individual packets they want to transfer. Applications should use the following data structures
to exchange isochronous packet information with the ugen driver:
typedef struct ugen_isoc_pkt_descr {
/*
* Set by the application, for all isochro.
* requests, to the num. of bytes to xfer
* in a packet.
*/
ushort_t dsc_isoc_pkt_len;

/*
* Set by ugen to actual num. of bytes sent/received
* in a packet.
*/
ushort_t dsc_isoc_pkt_actual_len;

952 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

/*
* Per pkt. status set by ugen driver both for the
* isochronous IN and OUT requests. Application can
* use USB_LC_STAT_* to parse the status.
*/
int dsc_isoc_pkt_status;
} ugen_isoc_pkt_descr_t;

typedef struct ugen_isoc_req_head {

/* pkt count of the isoc request */
int req_isoc_pkts_count;

/* pkt descriptors */
ugen_isoc_pkt_descr_t req_isoc_pkt_descrs[1];
} ugen_isoc_req_head_t;

req_isoc_pkts_count is limited by the capability of the USB host controller driver. The
current upper bound for the uhci and ohci drivers is 512. The upper bound for the ehci
driver is 1024.

For an isochronous-IN endpoint, applications must first use the ugen_isoc_req_head_t

structure followed by ugen_isoc_pkt_descr_t to write packet request information to the
ugen node. The ugen driver then checks the validity of the request. If it is valid, the driver
immediately begins isochronous polling on the IN endpoint and applications can proceed
with read(2) of the data on the isochronous-IN endpoint. Upon successful return of read(2),
isochronous packet descriptors (whose dsc_isoc_pkt_actual_len and
dsc_isoc_pkt_status fields were filled by the driver) are returned, followed by the request's
device payload data.

Applications should continuously check for and consume isochronous data. The ugen driver
enables buffering of up to eight seconds of incoming data for full-speed isochronous endpoint,
one second of data for high-speed isochronous endpoints who request one transaction per
microframe and 1/3 of a second of incoming data for high-speed high-bandwidth isochronous
endpoints who request three transactions per microframe. In case of buffer overflow, ugen
discards the oldest data.

The isochronous-IN polling can only be stopped by a close(2) associated file descriptor. If
applications want to change packet information, they must first close(2) the endpoint to stop
the isochronous-IN polling, then open(2) the endpoint and write(2) new packets request.

The following example shows how to read an isochronous-IN endpoint:

#include <sys/usb/clients/ugen/usb_ugen.h>

char *buf, *p;

ushort_t pktlen;

Device and Network Interfaces 953

ugen(7D)

int pktcnt, i;
int len;
ugen_isoc_req_head_t *req;
ugen_isoc_pkt_descr_t *pktdesc;
char rdbuf[5000];

pktcnt = 4; /* 4 packets in this request */

len = sizeof(int) +
sizeof(ugen_isoc_pkt_descr_t) * pktcount;

buf = malloc(len);
if (!buf) {
/* Error recovery. */
}

req = (ugen_isoc_req_head_t *)buf;

req->req_isoc_pkts_count = pktcnt;

pktdesc = (ugen_isoc_pkt_descr_t *)
(req->req_isoc_pkt_descrs);

for (i = 0; i < pktcnt; i++) {

/*
* pktlen should not exceed xfer
* capability of an endpoint
*/
pktdesc[i].dsc_isoc_pkt_len = pktlen;

pktdesc[i].dsc_isoc_pkt_actual_len = 0;
pktdesc[i].dsc_isoc_pkt_status = 0;
}

/*
* write request info to driver and len must
* be exactly the sum of
* sizeof(int) + sizeof(ugen_isoc_pkt_descr_t) * pktcnt.
* Otherwise, an error is returned.
*/
if (write(fd, buf, len) < 0) {
/* Error recovery. */
}

/*
* Read length should be sum of all pkt descriptors
* length + payload data length of all pkts
* (sizeof(ugen_isoc_pkt_descr_t) + pktlen) * pktcnt

954 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

*/
if (read(fd, rdbuf, (sizeof(ugen_isoc_pkt_descr_t) +
pktlen) * pktcnt) < 0) {
/* Error recovery. */
}

pktdesc = (ugen_isoc_pkt_descr_t *) rdbuf;

/* points to payload beginning */

p = rdbuf + pktcnt * sizeof(ugen_isoc_pkt_descr_t);

for (i = 0; i < pktcnt; i++) {

printf("packet %d len = %d,"
" actual_len = %d, status = 0x%x\n",
i, pktdesc->dsc_isoc_pkt_len,
pktdesc->dsc_isoc_pkt_actual_len,
pktdesc->dsc_isoc_pkt_status);

/* Processing data */

/*
* next packet data payload, do NOT use
* dsc_isoc_pkt_actual_len
*/
p += pktdesc->dsc_isoc_pkt_len;

pktdesc++;
}

For an isochronous-OUT endpoint, applications use the same packet descriptor and request
structures to write request information to the ugen node. Following the packet request head
information is the packet payload data. Upon successful return of write(2), applications can
read(2) the same ugen file immediately to retrieve the individual packet transfer status of the
last request. If the application isn't concerned about the status, it can omit it.

In the following example, an application transfers data on an isochronous-OUT endpoint:

#include <sys/usb/clients/ugen/usb_ugen.h>
char *buf, *p;
ushort_t i, pktlen;
int len, pktcnt;
ugen_isoc_req_head_t *req;
ugen_isoc_pkt_descr_t *pktdesc;
char rdbuf[4096];

pktcnt = 4;

/*

Device and Network Interfaces 955

ugen(7D)

* set packet length to a proper value, don’t

* exceed endpoint’s capability
*/
pktlen = 1024;

len = sizeof(int) +
sizeof(ugen_isoc_pkt_descr_t) * pktcount;

len += pktlen * pktcnt;

buf = malloc(len);
if (!buf) {
/* Error recovery. */
}

req = (ugen_isoc_req_head_t *)buf;

req->req_isoc_pkts_count = pktcnt;

pktdesc =
(ugen_isoc_pkt_descr_t *)(req->req_isoc_pkt_descrs);

for (i = 0; i < pktcnt; i++) {

pktdesc[i].dsc_isoc_pkt_len = pktlen;
pktdesc[i].dsc_isoc_pkt_actual_len = 0;
pktdesc[i].dsc_isoc_pkt_status = 0;
}

/* moving to beginning of payload data */

p = buf + sizeof(int) + sizeof(*pktdesc) * pktcnt;
for (i = 0; i < pktcnt; i++) {

/* fill in the data buffer */

p += pktlen;
}

/*
* write packet request information and data to ugen driver
*
* len should be the exact value of sizeof(int) +
* sizeof(ugen_isoc_pkt_descr_t) * pktcnt + payload length
*/
if (write(fd, buf, len) < 0) {
/* Error recovery. */
}

/* read packet status */

956 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

if (read(fd, rdbuf, sizeof(*pktdesc) * pktcnt) < 0) {

/* Error recovery. */

} else {

/* Parse every packet’s transfer status */

Errors The following statuses are returned by endpoint status device logical names:
USB_LC_STAT_NOERROR No error.
USB_LC_STAT_CRC CRC error detected.
USB_LC_STAT_BITSTUFFING Bit stuffing error.
USB_LC_STAT_DATA_TOGGLE_MM Data toggle did not match.
USB_LC_STAT_STALL Endpoint returned stall.
USB_LC_STAT_DEV_NOT_RESP Device not responding.
USB_LC_STAT_UNEXP_PID Unexpected Packet Identifier (PID).
USB_LC_STAT_PID_CHECKFAILURE Check bits on PID failed.
USB_LC_STAT_DATA_OVERRUN Data overrun.
USB_LC_STAT_DATA_UNDERRUN Data underrun.
USB_LC_STAT_BUFFER_OVERRUN Buffer overrun.
USB_LC_STAT_BUFFER_UNDERRUN Buffer underrun.
USB_LC_STAT_TIMEOUT Command timed out.
USB_LC_STAT_NOT_ACCESSED Not accessed by the hardware.
USB_LC_STAT_UNSPECIFIED_ERR Unspecified USBA or HCD error.
USB_LC_STAT_NO_BANDWIDTH No bandwidth available.
USB_LC_STAT_HW_ERR Host Controller h/w error.
USB_LC_STAT_SUSPENDED Device was suspended.
USB_LC_STAT_DISCONNECTED Device was disconnected.
USB_LC_STAT_INTR_BUF_FULL Polling was stopped as the interrupt-IN data
buffer was full. Buffer is now empty and
polling has been resumed.
USB_LC_STAT_INTERRUPTED Request was interrupted.

Device and Network Interfaces 957

ugen(7D)

USB_LC_STAT_NO_RESOURCES No resources available for request.

USB_LC_STAT_INTR_POLLING_FAILED Failed to restart polling.
USB_LC_STAT_ISOC_POLLING_FAILED Failed to start isochronous polling.
USB_LC_STAT_ISOC_UNINITIALIZED Isochronous packet information not
initialized.
USB_LC_STAT_ISOC_PKT_ERROR All packets in this isochronous request have
errors. The polling on this isochronous-IN
endpoint is suspended and can be resumed
on next read(2).

The following system call errno values are returned:

EINVAL An attempt was made to enable or disable "one transfer" mode while the
associated endpoint was open.
EBUSY The endpoint has been opened and another open is attempted.
EACCES An endpoint open was attempted with incorrect flags.
ENOTSUP Operation not supported.
ENXIO Device associated with the file descriptor does not exist.
ENODEV Device has been hot-removed or a suspend/resume happened before this
command.
EIO An I/O error occurred. Send a read on the endpoint status minor node to get the
exact error information.
EINTR Interrupted system call.
ENOMEM No memory for the allocation of internal structures.
Files /kernel/drv/ugen 32 bit ELF kernel module (x86 platform only)
/kernel/drv/sparcv9/ugen 64 bit ELF kernel module

/dev/usb/<vid>.<pid>/<N>/cntrl0
/dev/usb/<vid>.<pid>/<N>/cntrl0stat

/dev/usb/<vid>.<pid>/<N>/if<interface#>
<in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/if<interface#>
<in|out|cntrl><endpoint#>stat

/dev/usb/<vid>.<pid>/<N>/if<interface#>.
<alternate><in|out|cntrl<endpoint#>
/dev/usb/<vid>.<pid>/<N>/if<interface#>.
<alternate><in|out|cntrl><endpoint#>stat

958 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 ugen(7D)

/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>
<in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>
<in|out|cntrl<endpoint#stat>

/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>.
<alternate><in|out|cntrl><endpoint#>
/dev/usb/<vid>.<pid>/<N>/cfg<value>if<interface#>.
<alternate><in|out|cntrl><endpoint#>stat

/dev/usb/<vid>.<pid>/<N>/devstat

/dev/usb/<vid>.<pid>/<N>/if<interface#>cntrl0
/dev/usb/<vid>.<pid>/<N>/if<interface#>cntrl0stat

where N is an integer representing the instance number of this type of device. (All logical
device names for a single device share the same N.)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based SPARC

Availability SUNWugen, SUNWugenu

See Also libusb(3LIB), close(2), poll(2), read(2), write(2), aioread(3C), aiowrite(3C), usba(7D),
usb_dev_descr(9S).

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (ugen<instance num>): Error Message...
Too many minor nodes.
Device has too many minor nodes. Not all are available.
Instance number too high (<number>).
Too many devices are using this driver.
Cannot access <device>. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.
Device is not identical to the previous one on this port. Please disconnect and reconnect.
Same condition as described above; however in this case, the driver is unable to identify the
original device with a name string.

Device and Network Interfaces 959

ugen(7D)

Notes ugen returns -1 for all commands and sets errno to ENODEV when device has been
hot-removed or resumed from a suspend. The application must close and reopen all open
minor nodes to reinstate successful communication.

960 man pages section 7: Device and Network Interfaces • Last Revised 3 Dec 2008
 uhci(7D)

Name uhci – host controller driver

Synopsis pcivid,pid@unit-address

Description The uhci host controller driver is a USBA (Solaris USB Architecture) compliant nexus driver
that supports the Universal Host Controller Interface Specification 1.1, an industry standard
developed by Intel. The uhci driver supports all USB transfers, including interrupt, control,
isochronous and bulk.
Files /kernel/drv/uhci 32–bit ELF kernel module. (SPARC or x86).
/kernel/drv/amd64/uhci 64–bit ELF kernel module. (x86).
/kernel/drv/sparcv9/uhci 64-bit SPARC ELF kernel module.
/kernel/drv/uhci.conf Driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC or x86 PCI-based systems

Availability SUNWusb

See Also attributes(5), ehci(7D), hubd(7D), ohci(7D), usba(7D)

Writing Device Drivers

Universal Host Controller Interface Specification for USB 1.1

Universal Serial Bus Specification 2.0

System Administration Guide: Basic Administration

http://www.sun.com/io

Diagnostics All host controller errors are passed to the client drivers. Root errors are documented in
hubd(7D).

In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
WARNING: <device path> <uhci><instance number>>: Error message...
No SOF interrupts have been received. This USB UHCI host controller is unusable.
The USB hardware is not generating Start Of Frame interrupts. Please reboot the system. If
this problem persists, contact your system vendor.

Device and Network Interfaces 961

ural(7D)

Name ural – Ralink RT2500USB 802.11b/g Wireless Driver

Description The ural 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Ralink RT2500USB chipset-based NIC's.

Configuration The ural driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec. The ural driver supports only BSS networks (also
known as "ap" or "infrastructure" networks) and "open" (or "open-system") or "shared
system" authentication.
Files /dev/ural*
Special character device.
/kernel/drv/ural
32-bit ELF kernel module. (x86)
/kernel/drv/amd64/ural
64-bit ELF kernel module. (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWural

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification - IEEE, 2001

962 man pages section 7: Device and Network Interfaces • Last Revised 6 May 2008
 urtw(7D)

Name urtw – RealTek RTL8187L/RTL8187B USB 802.11b/g Wireless Driver

Description The urtw 802.11b/g wireless driver is a multi-threaded, loadable, clonable,

GLDv3-basedSTREAMS driver supporting the RealTek RTL8187L chipset-based wireless
devices.

Configuration The urtw driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 8, 24, 36, 48 and 54 Mbits/sec. The atu driver supports only BSS networks (also known
asap or infrastructure networks) and open (or open-system) or shared system
authentication.
Files /dev/urtw* Special character device.
/kernel/drv/urtw 32-bit ELF kernel module. (x86)
/kernel/drv/amd64/urtw 64-bit ELF kernel module. (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWurtw

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 — Wireless LAN Media Access Control and Physical Layer Specification, IEEE, 2001

Device and Network Interfaces 963

usba(7D)

Name usba, usb – Solaris USB Architecture (USBA)

Description USB provides a low-cost means for attaching peripheral devices, including mass-storage
devices, keyboards, mice, and printers, to a system. For complete information on the USB
architecture, visit the USB website at http://www.usb.org.

USBA supports 126 hot-pluggable USB devices per USB bus. The maximum data transfer rate
is 1.5 Mbits (low speed USB 1.x) or 12 Mbits (full speed USB 1.x) or 480 MBits (high speed
USB 2.0) Mbits per second (Mbps).

USBA adheres to the Universal Serial Bus 2.0 specification and provides a transport layer
abstraction to USB client drivers.

For information on how to write USB client drivers, see Writing Device Drivers. For the latest
information on writing USB drivers, visit
http://developers.sun.com/solaris/developer/support/driver/usb.html. For a complete list of
USBA interfaces, see Intro(9F) and Intro(9S).

Devices without a driver may have a libusb(3LIB) application. For more information, see
/usr/sfw/share/doc/libusb/libusb.txt.

Files Listed below are drivers and modules which either utilize or are utilized by USBA. Drivers in
/kernel/drv are 32 bit drivers (x86 only). Drivers in /kernel/drv/sparcv9 or
/kernel/drv/amd64 are 64 bit drivers.
Client Driver Function/Device

kernel/drv/[sparcv9|amd64/]hid HID class

kernel/drv/[sparcv9|amd64/]hubd hub class
kernel/drv/[sparcv9|amd64/]hwahc HWA Host Controller class
kernel/drv/[sparcv9|amd64/]hwarc HWA Radio Controller class
kernel/drv/[sparcv9|amd64/]scsa2usb mass storage class
kernel/drv/[sparcv9|amd64/]usbprn printer class
kernel/drv/[sparcv9|amd64/]usb_as audio streaming class
kernel/drv/[sparcv9|amd64/]usb_ac audio control class
kernel/drv/[sparcv9|amd64/]usbvc video class
kernel/drv/[sparcv9|amd64/]usb_mid multi-interface device
kernel/drv/[sparcv9|amd64/]usb_ia interface-association driver
kernel/drv/[sparcv9|amd64/]usbser_edge Edgeport USB to serial port
kernel/drv/[sparcv9|amd64/]usbsksp Keyspan USB to serial port
kernel/drv/[sparcv9|amd64/]usbsprl pl2303 USB to serial port
kernel/drv/[sparcv9|amd64/]usbsacm CDC ACM class to serial port
kernel/drv/[sparcv9|amd64/]ugen generic USB driver
kernel/drv/[sparcv9|amd64/]wusb_ca WUSB Cable Association class
kernel/drv/[sparcv9|amd64/]ohci open host controller driver
kernel/drv/[sparcv9|amd64/]uhci universal host controller driver
kernel/drv/[sparcv9|amd64/]ehci enhanced host controller driver

964 man pages section 7: Device and Network Interfaces • Last Revised 22 Apr 2009
 usba(7D)

Client Streams Modules Function/Device

/kernel/strmod/[sparcv9|amd64]usbkbm Keyboad
/kernel/strmod/[sparcv9|amd64]usbms Mouse
/kernel/strmod/[sparcv9|amd64]usb_ah Audio HID
Host Controller Interface Drivers Device

/kernel/drv/[sparcv9|amd64]ehci Enhanced HCI

/kernel/drv/[sparcv9|amd64]ohci Open HCI
/kernel/drv/[sparcv|amd64/]uhci Univeral HCI

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWusb, SUNWusbu

See Also cfgadm_usb(1M), libusb(3LIB), attributes(5), ehci(7D), hid(7D), hubd(7D), ohci(7D),

scsa2usb(7D), uhci(7D), usb_ac(7D), usb_as(7D), usb_ia(7D), usb_mid(7D), usbprn(7D),
usbsacm(7D), usbser_edge(7D), usbsksp(7D), usbsprl(7D), usbvc(7D), ugen(7D),
virtualkm(7D). Intro(9F), Intro(9S)

Writing Device Drivers

Universal Serial Bus Specification 2.0.

Interface Association Descriptor Engineering Change Notice (ECN)

System Administration Guide: Basic Administration

http://www.sun.com

Notes Booting from USB mass-storage devices is not supported on SPARC, but is supported on X86.

Diagnostics The messages described below may appear on the system console as well as being logged. All
messages are formatted in the following manner:
WARNING: Error message...
No driver found for device <device_name> (interface <number> node name=<node_name>)
The installed Solaris software does not contain a supported driver for this hardware.
<number> is the interface number. <name> is either the device path name or the device
name.
Draining callbacks timed out!
An internal error occured. Please reboot your system. If this problem persists, contact your
system vendor.

Device and Network Interfaces 965

usba(7D)

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usba<instance number>): message...
Incorrect USB driver version for <n.m>. Driver is incompatible with USBA framework.

966 man pages section 7: Device and Network Interfaces • Last Revised 22 Apr 2009
 usb_ac(7D)

Name usb_ac – USB audio control driver

Synopsis sound-control@unit-address

Description The usb_ac driver is a USBA (Solaris USB Architecture) compliant client driver that supports
the USB Audio Class 1.0 specification.

The audio control driver is a USB class driver and offers functionality similar to the audiocs
(sun4u) and audiots (Sun Blade 100) drivers which use the Solaris audio mixer framework
(mixer(7I)). Unlike the audiocs and audiots drivers, the USB audio device can have
play-only or record-only capability.

Drivers corresponding to other USB audio interfaces on the device, including the usb_as(7D)
audio streaming driver or the hid(7D) driver, are plumbed under the USB audio control
driver and do not directly interface with user applications.

The usb_ac driver supports USB audio class compliant devices with a feature unit.

Errors If a device is hot-removed while it is active, all subsequent opens returns EIO. All other errors
are defined in the audio(7I) man page.
Files /kernel/drv/usb_ac 32-bit x86 ELF kernel module
/kernel/drv/amd64/usb_ac 64-bit x86 ELF kernel module
/kernel/drv/sparcv9/usb_ac 64-bit SPARC ELF kernel module.
/kernel/drv/usb_ac.conf USB audio driver configuration file.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

Interface Stability Uncommitted

See Also cfgadm_usb(1M), ioctl(2), attributes(5), hid(7D), usba(7D), usb_as(7D), audio(7I),

mixer(7I), streamio(7I), usb_ah(7M)

Writing Device Drivers

Universal Serial Bus Specification 1.0 and 1.1

Universal Serial Bus Device Class Definition for Audio Devices, Release 1.0

System Administration: Basic Administration

Device and Network Interfaces 967

usb_ac(7D)

Diagnostics In addition to being logged, the following messages can appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usb_ac<instance num>): Error Message...
Failure to plumb audio streams drivers. The usb audio streaming driver or the hid driver
could not be plumbed under the audio control
driver and the device is not usable.

968 man pages section 7: Device and Network Interfaces • Last Revised 13 May 2009
 usb_ah(7M)

Name usb_ah – USB audio HID STREAMS module

Description The usb_ah STREAMS module enables the USB input control device which is a member of the
Human Interface Device (HID) class and provides support for volume change and mute
button. The usb_ah module is pushed on top of a HID class driver instance (see hid(7D)) and
below an Audio Control class driver instance (see usb_ac(7D)). It translates the HID specific
events to the events that are supported by the Solaris audio mixer framework.
Files /kernel/strmod/usb_ah 32-bit ELF kernel STREAMS module. (x86 platform
only.)
/kernel/strmod/sparcv9/usb_ah SPARC 64-bit ELF kernel STREAMS module
/kernel/strmod/amd64/usb_ah x8664-bit ELF kernel STREAMS module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWusb

Interface Stability Committed

See Also hid(7D), usba(7D), usb_ac(7D), usb_as(7D), usb_mid(7D), audio(7I),

STREAMS Programming Guide

System Administration Guide: Basic Administration

Universal Serial Bus Specification 1.0 and 1.1

Device Class Definition for Human Interface Devices (HID) 1.1

Diagnostics None

Notes If USB audio drivers are not loaded, buttons are not active.

Device and Network Interfaces 969

usb_as(7D)

Name usb_as – USB audio streaming driver

Synopsis sound@unit-address

Description The usb_as driver is a USBA (Solaris USB Architecture) compliant client driver that supports
the USB Audio Class 1.0 specification.

The usb_as driver processes audio data messages during play and record and sets sample
frequency, precision, encoding and other functions on request from the USB audio control
driver. See usb_ac(7D).

This driver is plumbed under the USB audio control driver and does not directly interface
with the user application.
Files /kernel/drv/usb_as 32–bit x86 ELF kernel module
/kernel/drv/amd64/usb_as 64–bit x86 ELF kernel module
/kernel/drv/sparcv9/usb_as 64–bit SPARC ELF kernel module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

Stability level Committed

See Also attributes(5), usba(7D), usb_ac(7D), audio(7I)

Writing Device Drivers

Universal Serial Bus Specification 1.0 and 1.1

System Administration Guide: Basic Administration

Diagnostics In addition to being logged, the following messages can appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usb_as<instance num>): Error Message...

where <device path> is the physical path to the device in /devices directory.
No bandwidth available.
There is no bandwidth available for the isochronous pipe. As a result, no data is transferred
during play and record.

970 man pages section 7: Device and Network Interfaces • Last Revised 15 May 2009
 usb_as(7D)

Operating a full/high speed audio device on a high speed port is not supported.
The USB software does not currently support full or high speed audio devices connected to
an external USB 2.0 hub that is linked to a port of a USB 2.0 host controller. Audio devices
must be connected directly to a port of a USB 2.0 controller or to any USB 1.1 port.
Cannot access device. Please reconnect <name>.
There was an error in accessing the device during reconnect. Please reconnect the device.
Device is not identical to the previous one on this port. Please disconnect and reconnect.
A USB audio streaming interface was hot-removed while open. A new device was
hot-inserted which is not identical to the original USB audio device. Please disconnect the
USB device and reconnect the device to the same port.

Notes The USB audio streaming interface is power managed if the device is idle.

Device and Network Interfaces 971

usbftdi(7D)

Name usbftdi – FTDI USB to serial converter driver

Synopsis #include <fcntl.h>
#include <sys/termio.h>
usbftdi@unit

Description The usbftdi driver is a loadable STREAMS and USBA (Solaris USB Architecture) compliant
client driver that provides basic asynchronous communication support for FTDI
USB-to-serial converters. Serial device streams are built with appropriate modules that are
pushed atop the usbftdi driver by the autopush(1M) facility.

Application The usbftdi module supports the termio(7I) device control functions specified by flags in the
Programming Interface c_cflag word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK, and INPCK flags
in the c_iflag word of the termios structure. All other termio(7I) functions must be
performed by STREAMS modules pushed atop the driver. When a device is opened, the ,
ldterm(7M) and ttcompat(7M) STREAMS modules are automatically pushed on top of the
stream, providing the standard termio(7I) interface.

Use device logical names /dev/term/[0-9]* to access the serial ports for a dial-in line that is
used with a modem.

Use device logical names /dev/cua/[0-9]* to access the serial ports for other applications.
These names are also used to provide a logical access point for a dial-out line.

Device hot-removal is functionally equivalent to a modem disconnect event, as defined in

termio(7I).

Input and output line speeds can be set to the following baud rates: 300, 600, 1200, 2400, 4800,
9600, 19200, 38400, 57600, 115200, 230400, 460800 or 921600. Input and output line speeds
can not be set independently. For example, when the output speed is set, the input speed is
automatically set to the same speed.

Soft Carrier Capabilities Many devices that use this USB serial interface component are not, in fact dial-in lines
connected to carefully configured RS-232 modems. They are often intelligent peripherals
whose manufacturers want to present a serial port interface to application software. Some
applications use only three wire connections, or are otherwise somewhat casual about the state
of the Carrier Detect (electrical) signal, and the other modem control lines.

The configuration file delivered with this driver, usbftdi.conf, acknowledges this by setting
the driver property ignore-cd to 1. This enables soft carrier mode where the kernel does
not block opens waiting for DCD to be asserted.

This behavior also matches the default ignore carrier detect behavior of the onboard
serial ports of machines that have them. See eeprom(1M) for further details.

The hardware carrier behavior (the driver's internal default) can be selected by either
unsetting (commenting out) the ignore-cd property, or by setting the value of the property to
zero.

972 man pages section 7: Device and Network Interfaces • Last Revised 13 Apr 2009
 usbftdi(7D)

More sophisticated selection of which devicesl ignore or obey the DCD signal can be effected
using port-%d-ignore-cd properties.

Dial-In and Dial-Out A related feature is available for traditional usage that enables a single tty line to be connected
Support to a modem and used for incoming and outgoing calls. By accessing through device logical
name /dev/cua/[0-9]*, you can open a port without the carrier detect signal being asserted,
either through hardware or an equivalent software mechanism. These devices are commonly
known as dial-out lines.

A dial-in line can be opened only if the corresponding dial-out line is closed. A blocking
/dev/term open waits until the /dev/cua line is closed, which drops Data Terminal Ready,
after which Carrier Detect usually drops as well. When the carrier is detected again with the
/dev/cua device remaining closed, this indicates an incoming call and the blocking open
seizes exclusive use of the line.

A non-blocking /dev/term open returns an error if the /dev/cua device is open.

If the /dev/term line is opened successfully (usually only when carrier is recognized on the
modem, though see Soft Carrier Capabilities section of this manual page), the
corresponding /dev/cua line can not be opened. This allows a modem and port to be used for
dial-in (enabling the line for login in /etc/inittab) or dial-out (using tip(1) or uucp(1C))
when no-one is logged in on the line.

Errors An open() fails under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The /dev/cua (dial-out) device is being opened while the /dev/term (dial-in
device) is open, or the dial-in device is being opened with a no-delay open while the
dial-out device is open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
EIO USB device I/O error.
Files /usr/kernel/drv/usbftdi 32-bit x86 ELF kernel module
/usr/kernel/drv/usbftdi.conf Kernel module configuration file
/usr/kernel/drv/amd64/usbftdi 64-bit x86 ELF kernel module
/usr/kernel/drv/sparcv9/usbftdi 64-bit SPARC ELF kernel module
/dev/cua/[0-9]* Dial-out tty lines
/dev/term/[0-9]* Dial-in tty lines

Device and Network Interfaces 973

usbftdi(7D)

Attributes See attributes(5) for a description of the following attribute:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWuftdi

See Also strconf(1), tip(1), uucp(1C), autopush(1M), eeprom(1M), ioctl(2), open(2), termios(3C),
usba(7D), termio(7I), ldterm(7M), ttcompat(7M), eeprom(1M), attributes(5),

Diagnostics In addition to being logged, the following messages might appear on the system console. All
messages are formatted in the following manner:
Warning: device_path usbftdiinstance num): Error Message ...
Device was disconnected while open. Data may have been lost.
The device has been hot-removed or powered off while it was open and a possible data
transfer was in progress. The job might be aborted.
Device is not identical to the previous one on this port. Please disconnect and
reconnect.
The device was hot-removed while open. A new device was hot-inserted which is not
identical to the original device. Please disconnect the device and reconnect the original
device to the same port.
Device has been reconnected, but data may have been lost.
The device that was hot-removed from its USB port has been re-inserted again to the same
port. It is available for access but data from a previous transfer might be lost.
Cannot access device. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.

The following messages might be logged into the system log. They are formatted in the
following manner:
device_path usbftdiiinstance number): message ...
Input overrun. Data was lost.

974 man pages section 7: Device and Network Interfaces • Last Revised 13 Apr 2009
 usb_ia(7D)

Name usb_ia – USB interface association driver

Synopsis interface association@unit-address

Description The usb_ia driver is a USBA (Solaris Universal Serial Bus Architecture)-compliant nexus
driver that binds to a device's interface association nodes when no vendor or class specific
driver is available. To do this, usb_ia creates nodes for the internal interfaces and then
attempts to bind drivers to each child interface.

Each interface association node has a parent device node that is created by usb_mid(7D) and
all ugen(7D) interfaces are exported by usb_mid(7D). (Note: attempting to export ugen(7D)
interfaces using usb_ia is prohibited.)
Files /kernel/drv/usb_ia
32-bit ELF kernel module. (x86).
/kernel/drv/amd64/usb_ia
64-bit ELF kernel module. (x86).
/kernel/drv/sparcv9/usb_ia
64-bit ELF kernel module. (SPARC).
Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC & x86 PCI-based systems

Availability SUNWusb

See Also attributes(5),ugen(7D), usb_mid(7D), usba(7D)

Writing Device Drivers
System Administration Guide: Basic Administration
Universal Serial Bus Specification 2.0 — 2000
Interface Association Descriptor Engineering Change Notice (ECN)—2003
http://www.sun.com/io
http://docs.sun.com
Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
<device path> (usb_ia<instance num>):message...
No driver found for interface <n> (nodename: <string>) of <device>.
No driver is available for this interface.

Device and Network Interfaces 975

usbkbm(7M)

Name usbkbm – keyboard STREAMS module for Sun USB Keyboard

Synopsis open("/dev/kbd", O_RDWR)

Description The usbkbm STREAMS module processes byte streams generated by a keyboard attached to a
USB port. USB keyboard is a member of Human Interface Device (HID) Class, and usbkbm
only supports the keyboard protocol defined in the specification. Definitions for altering
keyboard translation and reading events from the keyboard are in <sys/kbio.h> and
<sys/kbd.h>.

The usbkbm STREAMS module adheres to the interfaces exported by kb(7M). Refer to the
DESCRIPTION section of kb(7M) for a discussion of the keyboard translation modes and the
IOCTL section for the supported ioctl(2) requests.

IOCTLS USB Keyboard usbkbm returns different values for the following ioctls than kb(7M):

KIOCTYPE This ioctl() returns a new keyboard type defined for the USB keyboard. All
types are listed below:
KB_SUN3 Sun Type 3 keyboard
KB_SUN4 Sun Type 4 keyboard
KB_ASCII ASCII terminal masquerading as keyboard
KB_PC Type 101 PC keyboard
KB_USB USB keyboard

The USB keyboard type is KB_USB; usbkbm will return KB_USB in response to the KIOCTYPE
ioctl.
KIOCLAYOUT The argument is a pointer to an int. The layout code specified by the
bCountryCode value returned in the HID descriptor is returned in the int
pointed to by the argument. The countrycodes are defined in 6.2.1 of the
HID 1.0 specifications.
KIOCCMD
KBD_CMD_CLICK/KBD_CMD_NOCLICK The kb(7M) indicates that
inappropriate commands for
particular keyboards are ignored.
Because clicking is not supported on
the USB keyboard, usbkbm ignores
this command
KBD_CMD_SETLED Set keyboard LEDs. Same as kb(7M).
KBD_CMD_GETLAYOUT The country codes defined in 6.2.1
of the HID 1.0 specification are
returned.

976 man pages section 7: Device and Network Interfaces • Last Revised 27 June 2005
 usbkbm(7M)

KBD_CMD_BELL/KBD_CMD_NOBELL This command is supported

although the USB keyboard does not
have a buzzer. The request for the
bell is rerouted.
KBD_CMD_RESET There is no notion of resetting the
keyboard as there is for the type4
keyboard. usbkbm ignores this
command and does not return an
error.
Files /kernel/strmod/usbkbm 32-bit ELF kernel STREAMS module (x86 platform
only)
/kernel/strmod/sparcv9/usbkbm SPARC 64-bit ELF kernel STREAMS module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWusb

See Also dumpkeys(1), kbd(1), loadkeys(1), ioctl(2), keytables(4), attributes(5), hid(7D),

usba(7D), virtualkm(7D), kb(7M)

STREAMS Programming Guide

System Administration Guide: Basic Administration

http://www.sun.com/io

Diagn0STICS None

Device and Network Interfaces 977

usb_mid(7D)

Name usb_mid – USB Multi Interface Driver

Synopsis device@unit-address

Description The usb_mid driver is a USBA (Solaris Universal Serial Bus Architecture)-compliant nexus
driver that binds to device level nodes of a composite (multi interface) device if no vendor or
class-specific driver is available. The usb_mid driver creates interface nodes or interface
association nodes and attempts to bind drivers to them. If no driver is found for interface
association nodes, usb_ia(7D) is bound by default.

UGEN (Generic USB) The usb_mid driver also supports a ugen(7D) interface allowing raw access to the device, for
example by libusb(3LIB) applications, by passing the drivers bound to each interface. Since a
libusb application might change the state of the device, you should not access the device
through the child interface drivers. Note that the usb_mid driver creates a ugen interface only
if none of its children are explicitly bound to the ugen(7D) driver. Additionally, usb_mid does
not create children.
Files /kernel/drv/usb_mid 32-bit x86 ELF kernel module
/kernel/drv/amd64/usb_mid 64-bit x86 ELF kernel module
/kernel/drv/sparcv9/usb_mid 64-bit SPARC ELF kernel module
/dev/usb/*/*/* ugen(7D) nodes.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC & x86, PCI-based systems

Availability SUNWusb

See Also cfgadm_usb(1M), libusb(3LIB), attributes(5), usba(7D), usb_ia(7D)

Writing Device Drivers

Universal Serial Bus Specification 2.0—2000

Interface Association Descriptor Engineering Change Notice (ECN)—2003

System Administration Guide: Basic Administration

http://www.sun.com/io

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usb_mid<instance number>): Error Message...

978 man pages section 7: Device and Network Interfaces • Last Revised 22 Dec 2006
 usb_mid(7D)

Cannot access <device>. Please reconnect.

This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.
Device not identical to the previous one on this port. Please disconnect and reconnect.
Same condition as described above; however in this case, the driver is unable to identify the
original device with a name string.
Please disconnect and reconnect this device.
A hotplug of the device is needed before it can be restored.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usb_mid<instance number>): message...
No driver found for interface <n> (nodename: <string>) of <device>.
No driver is available for this interface.
No driver found for device <device>.
No driver is available for this interface.
Can't support ugen for multiple configuration devices that have attached child interface
drivers.
No ugen interface is available and libusb(3LIB) cannot work with this device.

Device and Network Interfaces 979

usbms(7M)

Name usbms – USB mouse STREAMS module

Synopsis #include <sys/vuid_event.h>
#include <sys/vuid_wheel.h>
#include <sys/msio.h>
#include <sys/msreg.h>

Description The usbms STREAMS module processes byte streams generated by a USB mouse. A USB
mouse is a member of the Human Interface Device (HID) class and the usbms module
supports only the mouse boot protocol defined in the HID specification.

The usbms module must be pushed on top of the HID class driver (see hid(7D)). In the
VUID_FIRM_EVENT mode, the usbms module translates packets from the USB mouse into Firm
events. The Firm event structure is defined in <sys/vuid_event.h>. The STREAMS module
state is initially set to raw or VUID_NATIVE mode which performs no message processing. See
the HID 1.0 specification for the raw format of the mouse packets. To initiate mouse protocol
conversion to Firm events, change the state to VUID_FIRM_EVENT.

When the usb mouse is opened or hot plugged in, the MOUSE_TYPE_ABSOLUTE event
(Firm event) is sent to the upper level to notify the VUID application that it is the absolute
mouse.
ioctls VUIDGFORMAT This option returns the current state of the STREAMS module. The state of
the usbms STREAMS module may be either VUID_NATIVE (no message
processing) or VUID_FIRM_EVENT (convert to Firm events).
VUIDSFORMAT The argument is a pointer to an int. Set the state of the STREAMS module
to the int pointed to by the argument.
typedef struct vuid_addr_probe {
short base; /* default vuid device addr directed too */
union {
short next; /* next addr for default when VUIDSADDR */
short current; /* current addr of default when VUIDGADDR */
} data;
} Vuid_addr_probe;
VUIDSADDR The argument is a pointer to a Vuid_addr_probe structure. VUIDSADDR sets the
virtual input device segment address indicated by base to next.

If base does not equal VKEY_FIRST, ENODEV is returned.

VUIDGADDR The argument is a pointer to a Vuid_addr_probe structure. Return the
address of the virtual input device segment indicated by base to current.

If base does not equal VKEY_FIRST, ENODEV is returned.

980 man pages section 7: Device and Network Interfaces • Last Revised 1 Dec 2005
 usbms(7M)

VUIDGWHEELCOUNT This ioctl takes a pointer to an integer as

argument and sets the value of the
integer to the number of wheels available
on this device. This ioctl returns 1 if
wheel(s) are present and zero if no
wheels are present.
VUIDGWHEELINFO This command returns static
information about the wheel that does
not change while a device is in use.
Currently the only information defined
is the wheel orientation which is either
VUID_WHEEL_FORMAT_VERTICAL
or
VUID_WHEEL_FORMAT_HORIZONTAL.
If the module cannot distinguish the
orientation of the wheel or the wheel is
of some other format, the format is set to
VUID_WHEEL_FORMAT_UNKNOWN.
typedef struct {
int vers;
int id;
int format;
} wheel_info;
The ioctl takes a pointer to "wheel_info"
structure with the "vers" set to the
current version of the "wheel_info"
structure and "id" set to the id of the
wheel for which the information is
desired.
VUIDSWHEELSTATE/VUIDGWHEELSTATE VUIDSWHEELSTATE sets the state of
the wheel to that specified in the
stateflags. VUIDGWHEELSTATE
returns the current state settings in the
stateflags field.

stateflags is an OR'ed set of flag bits. The

only flag currently defined is
VUID_WHEEL_STATE_ENABLED.

When stateflags is set to

VUID_WHEEL_STATE_ENABLED the

Device and Network Interfaces 981

usbms(7M)

module converts motion of the specified

wheel into VUID events and sends those
up stream.

Wheel events are enabled by default.

Applications that want to change the

stateflags should first get the current
stateflags and then change only the bit
they want.
typedef struct {
int vers;
int id;
uint32_t stateflags;
} wheel_state;
These ioctls take a pointer to
"wheel_state" as an argument with the
"vers" and "id" members filled in. These
members have the same meaning as that
for 'VUIDGWHEEL INFO' ioctl.

ioctl() requests for changing and retrieving mouse parameters use the Ms_parms structure:
typedef struct {
int jitter_thresh;
int speed_law;
int speed_limit;
} Ms_parms;

jitter_thresh is the "jitter threshold" of the mouse. Motions fewer than jitter_thresh
units along both axes are accumulated and then sent up the stream after 1/12 second.

speed_law indicates whether extremely large motions are to be ignored. If it is 1, a "speed

limit" is applied to mouse motions. Motions along either axis of more than speed_limit units
are discarded.
MSIOGETPARMS The argument is a pointer to a Ms_params structure. The usbms module
parameters are returned in the structure.
MSIOSETPARMS The argument is a pointer to a Ms_params structure. The usbms module
parameters are set according to the values in the structure.
MSIOSRESOLUTION Used by the absolute mouse to get the current screen resolution. The
parameter is a pointer to the Ms_screen_resolution structure:

982 man pages section 7: Device and Network Interfaces • Last Revised 1 Dec 2005
 usbms(7M)

int height; /* height of the screen */

int width; /* width of the screen */
}Ms_screen_resolution;
The usbms module parameters are set according to the values in the
structure and used to calculate the correct coordinates.
Files /kernel/strmod/usbms 32-bit ELF kernel STREAMS module (x86 platform only.)
/kernel/strmod/sparcv9/usbms SPARC 64-bit ELF kernel STREAMS module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWusb

See Also ioctl(2), attributes(5), hid(7D), virtualkm(7D), usba(7D)

System Administration Guide: Basic Administration

http://www/sun.com/io

Diagnostics The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usbms<instance number>): message...
Invalid Hid descriptor tree. Set to default value (3 buttons).
The mouse supplied incorrect information in its HID report.
Mouse buffer flushed when overrun.
Mouse data was lost.

Device and Network Interfaces 983

usbprn(7D)

Name usbprn – USB printer class driver

Synopsis #include <sys/usb/clients/printer/usb_printer.h>
#include <sys/ecppio.h>
usbprn@unit-address

Description The usbprn driver is a USBA (Solaris USB Architecture) compliant client driver that supports
the USB Printer Class 1.0 specification. The usbprn driver supports a subset of the ecpp(7D)
parallel port driver functionality. However, unlike the STREAMS-based ecpp driver, usbprn is
a character driver.

The usbprn driver supports all USB printer-class compliant printers. For a list of
recommended printers and USB parallel printer adapters, visit http://www.sun.com/io.

The usbrpn driver includes support for communicating with many different printers. To use
these printers, it may be nesessary to install and configure additional format conversion
packages available in the Solaris distribution. Configuration of these conversion packages
under the Solaris printing system can be simplified through the use of the printmgr(1M). This
tool allows selection of printer manufacturer/model information while creating a print queue.
For USB connected printers, it attempts to pre-select the the manufacturer and model
information based on the 1284 device id supplied by the printer.

UGEN (Generic USB) The usbprn driver also supports a ugen(7D) interface allowing raw access to the device, for
example by libusb(3LIB) applications, by passing the drivers bound to each interface.
Because a libusb application might change the state of the device, you should not access the
device through the child interface drivers.

Default Operation With certain minor exceptions (outlined in the Notes sections below), the usbprn driver
supports a subset of the ecpp(7D) ioctl interfaces:

Configuration variables are set to their default values each time the USB printer device is
attached. The write_timeout period (defined in the ECPPIOC_SETPARMS ioctl description
below) is set to 90 seconds. The mode is set to centronics mode (ECPP_CENTRONICS).
Parameters can be changed through the ECPPIOC_SETPARMS ioctl and read through the
ECPPIOC_GETPARMS ioctl. Each time the USB printer device is opened, the device is
marked as busy and all further opens will return EBUSY. Once the device is open, applications
can write to the device and the driver can send data and obtain device id and status.

Note – Unlike the ecpp(7D) driver, usbprn resets configuration variables to their default values
with each attach(9E). (The ecpp(7D) driver resets configuration variables with each open(2).)

Write Operation A write(2) operation returns the number of bytes successfully written to the device. If a
failure occurs while a driver is transferring data to printer, the contents of the status bits are
captured at the time of the error and can be retrieved by the application program using the
ECPPIOC_GETERR ioctl(2) call. The captured status information is overwritten each time
an ECPPIOC_TESTIO ioctl(2) occurs.

984 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2005
 usbprn(7D)

ioctls The usbprn driver supports prnio(7I) interfaces. Note that the PRNIOC_RESET command has
no effect on USB printers.

The following ioctl(2) calls are supported for backward compatibility and are not
recommended for new applications.
ECPPIOC_GETPARMS Gets current transfer parameters. The argument is a pointer to struct
ecpp_transfer_parms. If parameters are not configured after the
device is opened, the structure will be set to its default configuration.

Note – Unlike the ecpp(7D) driver, only the ECPP_CENTRONICS

mode is currently supported in usbprn.
ECPPIOC_SETPARMS Sets transfer parameters. The argument is a pointer to a struct
ecpp_transfer_parms. If a parameter is out of range, EINVAL is
returned. If the peripheral or host device cannot support the
requested mode, EPROTONOSUPPORT is returned.

The transfer parameters structure is defined in <sys/ecppio.h>:

struct ecpp_transfer_parms {
int write_timeout;
int mode;
};
The write_timeout field, which specifies how long the driver will take
to transfer 8192 bytes of data to the device, is set to a default value of
90 seconds. The write_timeout field must be greater than one second
and less than 300 seconds (five minutes.)

Note – Unlike the ecpp(7D) driver, only the ECPP_CENTRONICS

mode is currently supported in usbprn. Also, the semantics of
write_timeout in usbprn differ from ecpp(7D). Refer to ecpp(7D)
for information.
BPPIOC_TESTIO Tests the transfer readiness of a print device and checks status bits to
determine if a write(2) will succeed. If status bits are set, a transfer
will fail. If a transfer will succeed, zero is returned. If a transfer fails,
the driver returns EIO and the state of the status bits are captured. The
captured status can be retrieved using the BPPIOC_GETERR
ioctl(2) call. BPPIOC_TESTIO and BPPIOC_GETERR are
compatible to the ioctls specified in bpp(7D).

Note – Unlike the ecpp(7D) driver, only the ECPP_CENTRONICS

mode is currently supported in usbprn. Additionally, bus_error and
timeout_occurred fields are not used in the usbprn interface. (In
ecpp(7D), timeout_occurred is used.)

Device and Network Interfaces 985

usbprn(7D)

BPPIOC_GETERR Get last error status. The argument is a pointer to a struct

bpp_error_status. This structure indicates the status of all the
appropriate status bits at the time of the most recent error condition
during a write(2) call, or the status of the bits at the most recent
BPPIOC_TESTIO ioctl(2) call.
struct bpp_error_status {
char timeout_occurred; /* not used */
char bus_error; /* not used */
uchar_t pin_status; /* status of pins which
/* could cause error */
};
The pin_status field indicates possible error conditions. The error
status structure bpp_error_status is defined in the include file
<sys/bpp_io.h>. The valid bits for pin_status can be BPP_ERR_ERR,
BPP_SLCT_ERR, and BPP_PE_ERR. A set bit indicates that the associated
pin is asserted.

Note – Unlike the ecpp(7D) driver, only the ECPP_CENTRONICS

mode is currently supported in usbprn. Additionally, the bus_error
and timeout_occurred fields are not used in the usbprn interface. (In
ecpp(7D), timeout_occurred is used.) Unlike ecpp(7D), the
BPP_BUSY_ERR status bit is not supported by USB printers.
ECPPIOC_GETDEVID Gets the IEEE 1284 device ID from the peripheral. The argument is a
pointer to a struct ecpp_device_id. Applications should set mode
to ECPP_CENTRONICS. If another mode is used, the driver will
return EPROTONOSUPPORT. len is the length of the buffer pointed to by
addr. rlen is the actual length of the device ID string returned from
the peripheral. If the returned rlen is greater than len, the application
should call ECPPIOC_GETDEVID a second time with a buffer length
equal to rlen.

The 1284 device ID structure:

struct ecpp_device_id {
int mode; /* mode to use for reading device id */
int len; /* length of buffer */
int rlen; /* actual length of device id string */
char *addr; /* buffer address */
Note – Unlike ecpp(7D), only the ECPP_CENTRONICS mode is
currently supported in usbprn.

Read Operation The read operation is not supported and returns EIO.

986 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2005
 usbprn(7D)

Errors EBUSY The device has been opened and another open is attempted. An
attempt has been made to unload the driver while one of the units is
open.
EINVAL An unsupported IOCTL has been received. A ECPPIOC_SETPARMS
ioctl(2) is attempted with an out of range value in the
ecpp_transfer_parms structure.
EIO The driver has received an unrecoverable device error, or the device is
not responding, or the device has stalled when attempting an access. A
write(2) or ioctl(2) did not complete due to a peripheral access. A
read(2) system call has been issued.
ENXIO The driver has received an open(2) request for a unit for which the
attach failed.
ENODEV The driver has received an open(2) request for a device that has been
disconnected.
EPROTONOSUPPORT The driver has received a ECPPIOC_SETPARMS ioctl(2) for a mode
argument other than ECPP_CENTRONICS in the
ecpp_transfer_parms structure.
Files /kernel/drv/usbprn 32–bit x86 ELF kernel module
/kernel/drv/amd64/usbprn 64–bit x86 ELF kernel module
/kernel/drv/sparcv9/usbprn 64–bit SPARC ELF kernel module
/dev/usb/*/*/* ugen(7D) nodes.
/dev/printers/n Character special files

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also cfgadm_usb(1M), printmgr(1M), ioctl(2), open(2), read(2), write(2), libusb(3LIB),

attributes(5), bpp(7D), ecpp(7D), ugen(7D), usba(7D)prnio(7I), attach(9E)

Writing Device Drivers

Universal Serial Bus Specification 1.0 and 1.1

USB Device Class Definition for Printing Devices 1.0

System Administration Guide: Basic Administration

Device and Network Interfaces 987

usbprn(7D)

http://www.sun.com/io

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usbprn<instance num>): Error Message...
Device was disconnected while open. Data may have been lost.
The device has been hot-removed or powered off while it was open and a possible data
transfer was in progress. The job may be aborted.
Cannot access <device>. Please reconnect.
There was an error in accessing the printer during reconnect. Please reconnect the device.
Device is not identical to the previous one on this port. Please disconnect and reconnect.
A USB printer was hot-removed while open. A new device was hot-inserted which is not
identical to the original USB printer. Please disconnect the USB device and reconnect the
printer to the same port.
Printer has been reconnected but data may have been lost.
The printer that was hot-removed from its USB port has been re-inserted again to the same
port. It is available for access but the job that was running prior to the hot-removal may be
lost.

Notes The USB printer will be power managed if the device is closed.

If a printer is hot-removed before a job completes, the job is terminated and the driver will
return EIO. All subsequent opens will return ENODEV. If a printer is hot-removed, an LP
reconfiguration may not be needed if a printer is re-inserted on the same port. If re-inserted
on a different port, an LP reconfiguration may be required.

The USB Parallel Printer Adapter is not hotpluggable. The printer should be connected to
USB Parallel Printer Adapter before plugging the USB cable into host or hub port and should
be removed only after disconnecting the USB cable of USB Parallel Printer Adapter from the
host or hub port.

988 man pages section 7: Device and Network Interfaces • Last Revised 18 June 2005
 usbsacm(7D)

Name usbsacm – USB communication device class ACM driver

Synopsis #include <sys/termio.h>

usbsacm@unit

Description The usbsacm driver is a loadable STREAMS and USBA (Solaris USB architecture)-compliant
client driver that provides basic asynchronous communication support for USB modems and
ISDN terminal adapters that conform to the Universal Serial Bus Communication Device Class
Abstract Control Model (USB CDC ACM) specification. You can download the USB CDC
specification from the USB website at
http://www.usb.org/developers/devclass_docs/usbcdc11.pdf. Supported devices include mobile
phones and PCMCIA cards which provide modem function by the usb cable. Serial device
streams are built with appropriate modules that are pushed atop the usbsacm driver by the
autopush(1M) facility.

The usbsacm module supports the termio(7I) device control functions specified by flags in the
c_cflag word of termios structure, and by the IGNBRK, IGNPAR, PARMRK and INPCK flags
in the c_iflag word of the termios structure. All other termio(7I) functions must be performed
by STREAMS modules pushed atop the driver. When a device is opened, the ldterm(7M) and
ttcompat(7M) STREAMS modules are automatically pushed on top of the stream, providing
the standard termio(7I) interface.

You use device logical names /dev/term/[0-9]* to access the serial ports. These names are
typically used to provide a logical access point for a dial-in line that is used with a modem. You
can use pppd(1M) to transmit datagrams over these serial ports.

A special feature (controlled by the minor device number) is available that enables a single tty
line to be connected to a modem and used for incoming and outgoing calls. By accessing
through device logical name /dev/cua/[0-9]*, you can open a port without the carrier detect
signal being asserted, either through hardware or an equivalent software mechanism. These
devices are commonly known as 'dial-out' lines.

Unlike onboard serial ports, the usbsacm ports cannot serve as a local serial console.

Application A dial-in line can be opened only if the corresponding dial-out line is closed. A blocking
Programming /dev/term open waits until the /dev/cua line is closed (which drops Data Terminal Ready, after
Interface which Carrier Detect usually drops as well) and carrier is detected again. A non-blocking
/dev/term open returns an error if the /dev/cua is open.

If the /dev/term line is opened successfully (usually only when carrier is recognized on the
modem), the corresponding /dev/cua line cannot be opened. This allows a modem and port to
be used for dial-in (by enabling the line for login in /etc/inittab) or dial-out (by tip(1) or
uucp(1C)) as /dev/cua0 when no one is logged in on the line.

Device hot-removal is functionally equivalent to a modem disconnect event, as defined in

termio(7I).

Device and Network Interfaces 989

usbsacm(7D)

ioctls The usbsacm driver supports the standard set of termio(7I) ioctl calls.

The input and output line speeds may be set to any of the following baud rates: 75, 150, 300,
600, 1200, 1800, 2400, 4800, 9600, 19200, 38400, 57600, 115200, 230400 or 460800. The speeds
cannot be set independently. For example, when the output speed is set, the input speed is
automatically set to the same speed.

Errors An open() fails under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The /dev/cua (dial-out) device is being opened while the /dev/term (dial-in device)
is open, or the dial-in device is being opened with a no-delay open while the
dial-out device is open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
EIO USB device I/O error.
Files /kernel/drv/usbsacm 32-bit ELF kernel module. (x86)
/kernel/drv/amd64/usbsacm 64-bit ELF kernel module. (x86)
/kernel/drv/sparcv9/usbsacm 64-bit ELF kernel module. (SPARC)
/dev/cua/[0-9] dial-out tty lines
/dev/term/[0-9] dial-in tty lines

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86 PCI-based systems

Availability SUNWuacm

See Also strconf(1), tip(1),uucp(1C), autopush(1M), pppd(1M), ioctl(2), open(2), termios(3C),

attributes(5), usba(7D), termio(7I), ldterm(7M), ttcompat(7M)

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usbsacm<instance num>):Error Message...

Device was disconnected while open. Data may have been lost.
The device has been hot-removed or powered off while it was open and a possible data
transfer was in progress. The job may be aborted.

990 man pages section 7: Device and Network Interfaces • Last Revised 1 Nov 2006
 usbsacm(7D)

Device is not identical to the previous one on this port. Please disconnect and reconnect.
The device was hot-removed while open. A new device was hot-inserted which is not
identical to the original device. Please disconnect the device and reconnect the original
device to the same port.
Device has been reconnected, but data may have been lost.
The device that was hot-removed from its USB port has been re-inserted again to the same
port. It is available for access but data from a previous transfer may be lost.
Cannot access <device>. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usbsacm<instance number>): message...
Input overrun. Data was lost.

Device and Network Interfaces 991

usbser_edge(7D)

Name usbser_edge – Digi Edgeport USB to serial converter driver

Synopsis #include <fcntl.h>
#include <sys/termios.h>
usbser_edge@unit

Description The usbser_edge driver is a loadable STREAMS and USBA (Solaris USB architecture)
compliant client driver that provides basic asynchronous communication support for Digi
Edgeport USB-to-serial converters. Supported devices include Edgeport/1, Edgeport/2,
Edgeport/21, Edgeport/4, Edgeport/421, Edgeport/8, and Edgeport/416. Serial device streams
are built with appropriate modules that are pushed atop the usbser_edge driver by the
autopush(1M) facility.

The usbser_edge module supports the termio(7I) device control functions specified by flags
in the c_cflag word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK and INPCK
flags in the c_iflag word of the termios structure. All other termio(7I) functions must be
performed by STREAMS modules pushed atop the driver. When a device is opened, the
ldterm(7M) and ttcompat(7M) STREAMS modules are automatically pushed on top of the
stream, providing the standard termio(7I) interface.

Use device logical names /dev/term/[0-9]* to access the serial ports. These names are
typically used to provide a logical access point for a dial-in line that is used with a modem.

To allow a single tty line to be connected to a modem and used for incoming and outgoing
calls, a special feature is available that is controlled by the minor device number. By accessing
through device logical name/dev/cua/[0-9]*, you can open a port without the Carrier
Detect signal being asserted, either through hardware or an equivalent software mechanism.
These devices are commonly known as dial-out lines.

Unlike onboard serial ports, the usbser_edge ports cannot serve as a local serial console.

Application A dial-in line can be opened only if the corresponding dial-out line is closed. A blocking
Programming /dev/term open waits until the /dev/cua line is closed (which drops Data Terminal Ready,
Interface after which Carrier Detect usually drops as well) and carrier is detected again. A
non-blocking/dev/term open returns an error if the /dev/cua is open.

If the /dev/term line is opened successfully (usually only when carrier is recognized on the
modem), the corresponding /dev/cua line cannot be opened. This allows a modem and port
to be used for dial-in (by enabling the line for login in /etc/inittab) or dial-out (by tip(1),
or uucp(1C)) when no one is logged in on the line.

Device hot-removal is functionally equivalent to modem disconnect event, as defined in

termio(7I).

992 man pages section 7: Device and Network Interfaces • Last Revised 18 Sep 2009
 usbser_edge(7D)

ioctls The usbser_edge driver supports the standard set of termio(7I) ioctl calls.

Input and output line speeds can be set to the following baud rates: 0, 50, 75, 110, 134, 150,
200, 300, 600, 1200, 1800, 2400, 4800, 9600, 19200, 38400, 57600, 115200, or 230400. Input and
output line speeds cannot be set independently; for example, when the output speed is set, the
input speed is automatically set to the same speed.

Errors An open() fails under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The /dev/cua (dial-out) device is being opened while the/dev/term (dial-in device)
is open, or the dial-in device is being opened with a no-delay open while the
dial-out device is open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
EIO USB device I/O error.
Files /kernel/drv/usbser_edge 32–bit x86 ELF kernel module
/kernel/drv/amd64/usbser_edge 64–bit x86 ELF kernel module
/kernel/drv/sparcv9/usbser_edge 64–bit SPARC ELF kernel module
/kernel/drv/usbser_edge.conf configures communication mode
/dev/cua/[0-9]* dial-out tty lines
/dev/term/[0-9]* dial-in tty lines

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWuedg

See Also strconf(1), tip(1), uucp(1C), autopush(1M), ioctl(2), open(2), termios(3C),

attributes(5), usba(7D), termio(7I), ldterm(7M), ttcompat(7M)

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usbser_edge<instance num>): Error Message...

Device was disconnected while open. Data may have been lost.
The device was hot-removed or powered off while it was open and a possible data transfer
was in progress.

Device and Network Interfaces 993

usbser_edge(7D)

Device is not identical to the previous one on this port. Please disconnect and reconnect.
The device was hot-removed while open. A new device was hot-inserted which is not
identical to the original device. Please disconnect the device and reconnect the original
device to the same port.
Device has been reconnected, but data may have been lost.
The device that was hot-removed from its USB port has been re-inserted again to the same
port. It is available for access but data from a previous transfer may be lost.
Cannot access <device>. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usbser_edge<instance number>): message...
Input overrun Data was lost.

994 man pages section 7: Device and Network Interfaces • Last Revised 18 Sep 2009
 usbsksp(7D)

Name usbsksp – Keyspan USB to serial converter driver

Synopsis #include <fcntl.h>
#include <sys/termio.h>
usbskpsp@unit

Description The usbsksp driver is a loadable STREAMS and USBA (Solaris USB architecture) compliant
client driver that provides basic asynchronous communication support for Keyspan
USB-to-serial converters. The usbsksp driver supports the Keyspan USA19HS, USA49WG
and USA49WLC models. By default, the USA19HS and USA49WG models are compatible
with the usbsksp driver and no configuration or installation steps are required. (The
USA49WG model is a USB 2.0 device comforming to Universal Serial Bus Specification 2.0
and the USB 2.0 host controller is required to support the USA49WG model. Note that the
USA49WG is not compatible with USB 1.1 host controllers). If you use the Keyspan
USA49WLC model, you must download and install a firmware package to enable the device to
work with the usbsksp driver. See the Keyspan website
(http://www.keyspan.com/downloads/sun/) for more information. Serial device streams are
built with appropriate modules that are pushed atop the usbsksp driver by the autopush(1M)
facility.

The usbsksp module supports the termio(7I) device control functions specified by flags in the
c_cflag word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK and INPCK
flags in the c_iflag word of the termios structure. All other termio(7I) functions must be
performed by STREAMS modules pushed atop the driver. When a device is opened, the
ldterm(7M) and ttcompat(7M) STREAMS modules are automatically pushed on top of the
stream, providing the standard termio(7I) interface.

Use device logical names /dev/term/[0-9]* to access the serial ports. These names are
typically used to provide a logical access point for a dial-in line that is used with a modem.

A special feature (controlled by the minor device number) is available that enables a single tty
line to be connected to a modem and used for incoming and outgoing calls. By accessing
through device logical name /dev/cua/[0-9]*, you can open a port without the carrier detect
signal being asserted, either through hardware or an equivalent software mechanism. These
devices are commonly known as 'dial-out' lines.

Unlike onboard serial ports, the usbsksp ports cannot serve as a local serial console.

Application A dial-in line can be opened only if the corresponding dial-out line is closed. A blocking
Programming /dev/term open waits until the /dev/cua line is closed (which drops Data Terminal Ready,
Interface after which Carrier Detect usually drops as well) and carrier is detected again. A non-blocking
/dev/term open returns an error if the /dev/cua is open.

If the /dev/term line is opened successfully (usually only when carrier is recognized on the
modem), the corresponding /dev/cua line cannot be opened. This allows a modem and port

Device and Network Interfaces 995

usbsksp(7D)

to be used for dial-in (by enabling the line for login in /etc/inittab) or dial-out (by tip(1),
or uucp(1C)) when no one is logged in on the line.

Device hot-removal is functionally equivalent to a modem disconnect event, as defined in

termio(7I).

ioctls The usbsksp driver supports the standard set of termio(7I) ioctl calls.

Input and output line speeds can be set to the following baud rates: 0, 50, 75, 110, 134, 150,
200, 300, 600, 1200, 1800, 2400, 4800, 9600, 19200, 38400, 57600, 115200, or 230400. Input and
output line speeds cannot be set independently. For example, when the output speed is set, the
input speed is automatically set to the same speed.

Errors An open() fails under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The /dev/cua (dial-out) device is being opened while the /dev/term (dial-in
device) is open, or the dial-in device is being opened with a no-delay open while the
dial-out device is open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
EIO USB device I/O error.
Files /kernel/drv/usbsksp 32-bit x86 ELF kernel module.
/kernel/drv/amd64/usbsksp 64-bit x86 ELF kernel module.
/kernel/drv/sparcv9/usbsksp 64-bit SPARC ELF kernel module.
/dev/cua/[0-9]* dial-out tty lines.
/dev/term/[0-9]* dial-in tty lines.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWuksp

See Also strconf(1), tip(1), uucp(1C), autopush(1M), ioctl(2), open(2), termios(3C),

attributes(5), usba(7D), termio(7I), ldterm(7M), ttcompat(7M)

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usbsksp<instance num>): Error Message...

996 man pages section 7: Device and Network Interfaces • Last Revised 23 August 2006
 usbsksp(7D)

Device was disconnected while open. Data may have been lost.
The device has been hot-removed or powered off while it was open and a possible data
transfer was in progress. The job may be aborted.
Device is not identical to the previous one on this port. Please disconnect and reconnect.
The device was hot-removed while open. A new device was hot-inserted which is not
identical to the original device. Please disconnect the device and reconnect the original
device to the same port.
Device has been reconnected, but data may have been lost.
The device that was hot-removed from its USB port has been re-inserted again to the same
port. It is available for access but data from a previous transfer may be lost.
Cannot access <device>. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.
No valid firmware available for Keyspan usa49wlc usb-to-serial adapter. Please download it
from Keyspan website and install it.
By default, only an empty firmware package is installed for the usa49wlc model. Please
download the SUNWukspfw package from Keyspan's web site and install it.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usbsksp<instance number>): message...
Input overrun Data was lost.

Device and Network Interfaces 997

usbsprl(7D)

Name usbsprl – Prolific PL2303 USB to serial converter driver

Synopsis #include <fcntl.h>
#include <sys/termio.h>
usbsprl@unit

Description The usbsprl driver is a loadable STREAMS and USBA (Solaris USB architecture) compliant
client driver that provides basic asynchronous communication support for Prolific PL2303
USB-to-serial converters. Supported devices include PL2303H, PL2303HX and PL2303X.
Serial device streams are built with appropriate modules that are pushed atop the usbsprl
driver by the autopush(1M) facility.

The usbsprl module supports the termio(7I) device control functions specified by flags in the
c_cflag word of the termios structure, and by the IGNBRK, IGNPAR, PARMRK and INPCK
flags in the c_iflag word of the termios structure. All other termio(7I) functions must be
performed by STREAMS modules pushed atop the driver. When a device is opened, the
ldterm(7M) and ttcompat(7M) STREAMS modules are automatically pushed on top of the
stream, providing the standard termio(7I) interface.

Use device logical names /dev/term/[0-9]* to access the serial ports. These names are
typically used to provide a logical access point for a dial-in line that is used with a modem.

A special feature (controlled by the minor device number) is available that enables a single tty
line to be connected to a modem and used for incoming and outgoing calls. By accessing
through device logical name /dev/cua/[0-9]*, you can open a port without the carrier detect
signal being asserted, either through hardware or an equivalent software mechanism. These
devices are commonly known as 'dial-out' lines.

Application A dial-in line can be opened only if the corresponding dial-out line is closed. A blocking
Programming /dev/term open waits until the /dev/cua line is closed (which drops Data Terminal Ready,
Interface after which Carrier Detect usually drops as well) and carrier is detected again. A non-blocking
/dev/term open returns an error if the /dev/cua is open.

If the /dev/term line is opened successfully (usually only when carrier is recognized on the
modem), the corresponding /dev/cua line cannot be opened. This allows a modem and port
to be used for dial-in (by enabling the line for login in /etc/inittab) or dial-out (by tip(1),
or uucp(1C)) when no one is logged in on the line.

Device hot-removal is functionally equivalent to a modem disconnect event, as defined in

termio(7I).

ioctls The usbsprl driver supports the standard set of termio(7I) ioctl calls.

998 man pages section 7: Device and Network Interfaces • Last Revised 23 Nov 2006
 usbsprl(7D)

Input and output line speeds can be set to the following baud rates: 75, 150, 300, 600, 1200,
1800, 2400, 4800, 9600, 19200, 38400, 57600, 115200, 230400 or 460800. Input and output line
speeds cannot be set independently. For example, when the output speed is set, the input
speed is automatically set to the same speed.

Errors An open() fails under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The /dev/cua (dial-out) device is being opened while the /dev/term (dial-in
device) is open, or the dial-in device is being opened with a no-delay open while the
dial-out device is open.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl() call.
EIO USB device I/O error.
Files /kernel/drv/usbsprl 32-bit x86 ELF kernel module.
/kernel/drv/amd64/usbsprl 64-bit x86 ELF kernel module.
/kernel/drv/sparcv9/usbsprl 64-bit SPARC ELF kernel module.
/dev/cua/[0-9]* dial-out tty lines.
/dev/term/[0-9]* dial-in tty lines.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWuprl

See Also strconf(1), tip(1), uucp(1C), autopush(1M), ioctl(2), open(2), termios(3C),

attributes(5), usba(7D), termio(7I), ldterm(7M), ttcompat(7M)

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usbsprl<instance num>): Error Message...

Device was disconnected while open. Data may have been lost.
The device has been hot-removed or powered off while it was open and a possible data
transfer was in progress. The job may be aborted.

Device and Network Interfaces 999

usbsprl(7D)

Device is not identical to the previous one on this port. Please disconnect and reconnect.
The device was hot-removed while open. A new device was hot-inserted which is not
identical to the original device. Please disconnect the device and reconnect the original
device to the same port.
Device has been reconnected, but data may have been lost.
The device that was hot-removed from its USB port has been re-inserted again to the same
port. It is available for access but data from a previou transfer may be lost.
Cannot access <device>. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.

The following messages may be logged into the system log. They are formatted in the
following manner:
<device path><usbsprl<instance number>): message...
Input overrun.
Data was lost.

1000 man pages section 7: Device and Network Interfaces • Last Revised 23 Nov 2006
 usbvc(7D)

Name usbvc – USB video class driver

Synopsis #include <sys/usb/clients/video/usbvc/usbvc.h>

#include <sys/videodev2.h>

usbvc@unit-address

Description The usbvc driver is a USBA (Solaris USB Architecture)-compliant client driver that supports
the USB Device Class Definition for Video Devices specification, Versions 1.0 and 1.1. The
usbvc driver supports a subset of the video controls and formats described in the USB
specification.

The usbvc driver also implements the Video4Linux2 API (V4L2), Version 0.20 for
applications. For more information on the V4L2 API, visit http://www.thedirks.org/v4l2.

Note that the usbvc driver supports the video capture function only and that video output is
not supported. For more information on supported USB video-class devices and functions,
visit http://www.sun.com/io.

Reading Data The usbvc driver reads video data from the isochronous endpoint of the device. Bulk data
endpoints are not supported.

MJPEG and UNCOMPRESSED video formats are supported. Isochronous data are read from
the isochronous input device frame-by-frame and are maintained in a buffer array within the
driver. Video frames are read from the driver using the read(2) or mmap(2) I/O method. For
read(2), each read returns a buffer of a video frame. For mmap(2), each VIDIOC_DQBUF ioctl
returns the buffer structure v4l2_buffer. (A video frame buffer pointer is included in the
structure). See the V4L2 API for buffer structure and other related data structure information.

ioctls A brief overview of supported ioctl requests appears below. For more detailed information,
refer to the V4L2 API document. Note: ioctl information presented in the V4L2 API document
may differ slightly from the content of this manpage. In such cases, you should rely on the
information in this manpage.
VIDIOC_QUERYCAP
Query the device capabilities. Besides device capabilities, the usbvc driver returns structure
v4l2_capability which includes information on the driver, data bus and OS kernel. Please
note that the "Version" structure member has no meaning in Solaris and is always set to 1.
VIDIOC_ENUM_FMT
Enumerate the video formats supported by the device.
VIDIOC_S_FMT
Set a video format.
VIDIOC_G_FMT
Get a video format.

Device and Network Interfaces 1001

usbvc(7D)

VIDIOC_REQBUFS
Request the usbvc driver to allocate video data buffers. If a buffer is set to zero, the driver
stops reading video data from the device and releases all allocated buffers. (For mmap(2)
only).
VIDIOC_QUERYBUF
Query a given buffer's status. (For mmap(2) only).
VIDIOC_QBUF
Enqueue an empty buffer to the video data buffer array. (For mmap(2) only).
VIDIOC_DQBUF
Dequeue a done buffer from the video data buffer array. (For mmap(2) only).
VIDIOC_STREAMON
Start reading video data.
VIDIOC_STREAMOFF
Stop reading video data.
VIDIOC_ENUMINPUT
Enumerate all device inputs. Currently, the usbvc driver supports one input only.
VIDIOC_G_INPUT
Get the device's current input. At this time, the usbvc driver supports one input only.
VIDIOC_S_INPUT
Set the device's current input. At this time, the usbvc driver supports one input only.
VIDIOC_QUERYCTRL
Query the device and driver for supported video controls. Currently, the usbvc driver
supports the brightness, contrast, saturation, hue, and gamma video controls.
VIDIOC_G_CTRL
Get the device's current video control.
VIDIOC_S_CTRL
Set the device's current video control.
VIDIOC_G_PARM
Get streaming parameters, the number of frames per second and number of buffers used
internally by driver in read/write mode.
VIDIOC_S_PARM
Set streaming parameters, the number of frames per second and number of buffers used
internally by driver in read/write mode.
Errors EBUSY An open was attempted after the device has already been opened.
EINVAL An unsupported ioctl is received or an ioctl is attempted with an out-of-range
value.

1002 man pages section 7: Device and Network Interfaces • Last Revised 18 July 2008
 usbvc(7D)

EIO The driver received an unrecoverable device error or the device did not respond or
the device stalled when attempting an access. A read(2) or ioctl(2) did not
complete due to a peripheral access.
ENXIO The driver received an open(2) request for a device for which the attach failed.
ENODEV The driver received an open(2) request for a disconnected device.
Files /kernel/drv/usbvc
32-bit ELF kernel module. (x86)
/kernel/drv/amd64/usbvc
64-bit ELF kernel module. (x86)
/kernel/drv/sparcv9/usbvc
64-bit ELF kernel module. (SPARC)
/dev/usb/*/*/*
ugen(7D) nodes.
/dev/videoN
Device node for isochronous input from USB video device and device control.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusbvc

See Also cfgadm_usb(1M), ioctl(2), open(2), mmap(2), read(2), libusb(3LIB),

attributes(5),ugen(7D), usba(7D), attach(9E)

Writing Device Drivers

System Administration Guide: Basic Administration

Universal Serial Bus Specification 1.0, 1.1 and 2.0— 1996, 1998, 2000

USB Device Class Definition for Video Devices 1.0 and 1.1— 2003, 2005

Video4Linux2 API (V4L2), Version 0.20

I/O Technologies and Solutions —http://www.sun.com/io

http://docs.sun.com

http://www.usb.org

http://www.thedirks.org/v4l2

Device and Network Interfaces 1003

usbvc(7D)

Diagnostics In addition to being logged, the following messages may appear on the system console. All
messages are formatted in the following manner:
Warning: <device path> (usbvc<instance num>):Error Message...
Device was disconnected while open. Data may have been lost.
The device has been hot-removed or powered off while it was open and a possible data
transfer was in progress. The job may be aborted.
Cannot access <device>. Please reconnect.
This device has been disconnected because a device other than the original one has been
inserted. The driver informs you of this fact by displaying the name of the original device.
Device is not identical to the previous one on this port. Please disconnect and reconnect.
The device was hot-removed while open. A new device was hot-inserted which is not
identical to the original device. Please disconnect the device and reconnect the original
device to the same port.

Notes The USB video device will be power-managed when the device is idle.

If a USB video device is hot-removed while active, a console warning is displayed requesting
you to put the device back in the same port and telling you of potential data loss. Hot-removal
of an active video device is strongly discouraged.

Always close all applications before hot-removing or hot-inserting a device. If an application

is open when a device is hot-removed, inserting the device in a different port will create new
/dev/videoN links. Moving an active device to another port is not recommended.

1004 man pages section 7: Device and Network Interfaces • Last Revised 18 July 2008
 usbwcm(7M)

Name usbwcm – STREAMS module for Wacom USB Tablets

Synopsis # include <sys/usb/clients/usbinput/usbwcm/usbwcm.h>

Description The usbwcm STREAMS module processes byte streams generated by a Wacom USB tablet.

The usbwcm module must be pushed on top of the HID class driver. See hid(7D). The usbwcm
module translates data from Wacom USB tablet into formatted events expected by Wacom
X.org XInputdriver. The event structure is the same as that of FreeBSD uwacom driver.

Event Structure The event_input structure is defined in:

usr/include/sys/usb/clients/usbinput/usbwcm/usbwcm.h
struct event_input {
#if defined(_LP64) || defined(_I32LPx)
struct timeval32 time;
#else
struct timeval time;
#endif
uint16_t type;
uint16_t index;
int32_t value;
};
time The event's timestamp. When the event occurred. The timestamp is not defined to
be meaningful except by being compared with other input event timestamps.
type The event's unique type: button, relative/absolute valuator, sync, and so forth. type is
among the following: EVT_SYN, EVT_BTN, EVT_REL, EVT_ABS and EVT_MSC.
index The event's sub-type. The index in a button event identifies which button status was
changed. Typical button index includes: BTN_LEFT, BTN_RIGHT, BTN_MIDDLE,
BTN_SIDE, BTN_EXTRA, BTN_TOOL_PEN, BTN_TOOL_ERASER, BTN_TOOL_PAD,
BTN_TOOL_MOUSE, BTN_TIP, BTN_STYLUS_1, BTN_STYLUS_2

For absolute valuators, index is among the following: ABS_X, ABS_Y, ABS_Z,
ABS_RX, ABS_RY, ABS_WHEEL, ABS_PRESSURE, ABS_DISTANCE, ABS_TILT_X,
ABS_TILT_Y, and ABS_MISC.
Ioctls EVTIOCGVERSION The argument is a pointer to an int. This option returns the current
version of the event interface implemented by the STREAMS module.
EVTIOCGDEVID The argument is a pointer to event_dev_id structure. This ioctl returns
the identifiers of the device.
struct event_dev_id {
uint16_t bus;
#defineID_BUS_USB 3
uint16_t vendor;
uint16_t product;

Device and Network Interfaces 1005

usbwcm(7M)

uint16_t version;
};
EVTIOCGBM The argument is a pointer to an variable-length char array. This ioctl
returns the event types reported by device:

EVT_SYN, EVT_BTN, EVT_REL, EVT_ABS, EVT_MSC

EVTIOCGABS The argument is a pointer to an event_abs_axis structure. This ioctl
returns the ranges, and other parameters for the specified axis.
struct event_abs_axis {
int32_t value;
int32_t min;
int32_t max;
int32_t fuzz;
int32_t flat;
};

Files /kernel/strmod/usbwcm 32-bit ELF kernel STREAMS module

/kernel/strmod/amd64/usbwcm 64-bit ELF kernel STREAMS module
/kernel/strmod/sparcv9/usbwcm SPARC 64-bit ELF kernel STREAMS module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86-based systems

Availability SUNWusb

See Also ioctl(2), attributes(5), hid(7D)

System Administration Guide: Basic Administration

http://linuxwacom.sourceforge.net

http://www.sun.com

1006 man pages section 7: Device and Network Interfaces • Last Revised 15 Jan 2010
 uscsi(7I)

Name uscsi – user SCSI command interface

Synopsis #include <sys/scsi/impl/uscsi.h>

ioctl(int fildes, int request, struct uscsi_cmd *cmd);

Description The uscsi command is very powerful and somewhat dangerous; therefore it has some
permission restrictions. See WARNINGS for more details.
Drivers supporting this ioctl(2) provide a general interface allowing user-level applications
to cause individual SCSI commands to be directed to a particular SCSI or ATAPI device under
control of that driver. The uscsi command is supported by the sd driver for SCSI disks and
ATAPI CD-ROM drives, and by the st driver for SCSI tape drives. uscsi may also be
supported by other device drivers; see the specific device driver manual page for complete
information.
Applications must not assume that all Solaris disk device drivers support the uscsi ioctl
command. The SCSI command may include a data transfer to or from that device, if
appropriate for that command. Upon completion of the command, the user application can
determine how many bytes were transferred and the status returned by the device. Also,
optionally, if the command returns a Check Condition status, the driver will automatically
issue a Request Sense command and return the sense data along with the original status. See
the USCSI_RQENABLE flag below for this Request Sense processing. The uscsi_cmd structure is
defined in <sys/scsi/impl/uscsi.h> and includes the following members:
int uscsi_flags; /* read, write, etc. see below */
short uscsi_status; /* resulting status */
short uscsi_timeout; /* Command Timeout */
caddr_t uscsi_cdb /* CDB to send to target */
caddr_t uscsi_bufaddr; /* i/o source/destination */
size_t uscsi_buflen; /* size of i/o to take place*/
size_t uscsi_resid; /* resid from i/o operation */
uchar_t uscsi_cdblen; /* # of valid CDB bytes */
uchar_t uscsi_rqlen; /* size of uscsi_rqbuf */
uchar_t uscsi_rqstatus; /* status of request sense cmd */
uchar_t uscsi_rqresid; /* resid of request sense cmd */
caddr_t uscsi_rqbuf; /* request sense buffer */
void *uscsi_reserved_5; /* Reserved for future use */

The fields of the uscsi_cmd structure have the following meanings:

uscsi_flags The I/O direction and other details of how to carry out the SCSI
command. Possible values are described below.
uscsi_status The SCSI status byte returned by the device is returned in this field.
uscsi_timeout Time in seconds to allow for completion of the command.
uscsi_cdb A pointer to the SCSI CDB (command descriptor block) to be
transferred to the device in command phase.

Device and Network Interfaces 1007

uscsi(7I)

uscsi_bufaddr The user buffer containing the data to be read from or written to the
device.
uscsi_buflen The length of uscsi_bufaddr.
uscsi_resid If a data transfer terminates without transferring the entire requested
amount, the remainder, or residue, is returned in this field.
uscsi_cdblen The length of the SCSI CDB to be transferred to the device in
command phase.
uscsi_rqlen The length of uscsi_rqbuf, the application's Request Sense buffer.
uscsi_rqstatus The SCSI status byte returned for the Request Sense command
executed automatically by the driver in response to a Check
Condition status return.
uscsi_rqresid The residue, or untransferred data length, of the Request Sense data
transfer (the number of bytes, less than or equal to uscsi_rqlen,
which were not filled with sense data).
uscsi_rqbuf Points to a buffer in application address space to which the results of
an automatic Request Sense command are written.
uscsi_reserved_5 Reserved for future use.

The uscsi_flags field defines the following:

USCSI_WRITE /* send data to device */
USCSI_SILENT /* no error messages */
USCSI_DIAGNOSE /* fail if any error occurs */
USCSI_ISOLATE /* isolate from normal commands */
USCSI_READ /* get data from device */
USCSI_ASYNC /* set bus to asynchronous mode */
USCSI_SYNC /* return bus to sync mode if possible */
USCSI_RESET /* reset target */
USCSI_RESET_TARGET /* reset target */
USCSI_RESET_LUN /* reset logical unit */
USCSI_RESET_ALL /* reset all targets */
USCSI_RQENABLE /* enable request sense extensions */
USCSI_RENEGOT /* renegotiate wide/sync on next I/O */

The uscsi_flags bits have the following interpretation:

USCSI_WRITE Data will be written from the initiator to the target.
USCSI_SILENT The driver should not print any console error messages or warnings
regarding failures associated with this SCSI command.

1008 man pages section 7: Device and Network Interfaces • Last Revised 29 May 2007
 uscsi(7I)

USCSI_DIAGNOSE The driver should not attempt any retries or other recovery
mechanisms if this SCSI command terminates abnormally in any
way.
USCSI_ISOLATE This SCSI command should not be executed with other commands.
USCSI_READ Data will be read from the target to the initiator.
USCSI_ASYNC Set the SCSI bus to asynchronous mode before running this
command.
USCSI_SYNC Set the SCSI bus to synchronous mode before running this
command.
USCSI_RESET Send a SCSI bus device reset message to this target.
USCSI_RESET_TARGET Same as USCSI_RESET. Use this flag to request TARGET RESET.
(USCSI_RESET is maintained only for compatibility with old
applications).
USCSI_RESET_LUN Send a SCSI logical unit reset message to this target.
USCSI_RESET_ALL USCSI_RESET_ALL, USCSI_RESET/USCSI_RESET_TARGET
and USCSI_RESET_LUN are mutually exclusive options and
issuing them in any simultaneous combination will result in
implementation-dependent behavior

When a USCSI reset request is combined with other SCSI

commands, the following semantics take effect:

If the USCSI RESET flag is specified, the other fields (other than
uscsi_flags) in the uscsi_cmd are ignored. The uscsi_cdblen must be
set to zero.
USCSI_RQENABLE Enable Request Sense extensions. If the user application is prepared
to receive sense data, this bit must be set, the fields uscsi_rqbuf
and uscsi_rqbuflen must be non-zero, and the uscsi_rqbuf must
point to memory writable by the application.
USCSI_RENEGOT Tells USCSI to renegotiate wide mode and synchronous transfer
speed before the transmitted SCSI command is executed. This flag
in effects tells the target driver to pass the
FLAG_RENEGOTIATE_WIDE_SYNC flag in the SCSI packet before
passing the command to an adapter driver for transport.

See the scsi_pkt(9S) flag FLAG_RENEGOTIATE_WIDE_SYNC for more

information.

Device and Network Interfaces 1009

uscsi(7I)

ioctls The ioctl supported by drivers providing the uscsi interface is:
USCSICMD The argument is a pointer to a uscsi_cmd structure. The SCSI device addressed
by that driver is selected, and given the SCSI command addressed by
uscsi_cdb. If this command requires a data phase, the uscsi_buflen and
uscsi_bufaddr fields must be set appropriately; if data phase occurs, the
uscsi_resid is returned as the number of bytes not transferred. The status of
the command, as returned by the device, is returned in the uscsi_status field.
If the command terminates with Check Condition status, and Request Sense is
enabled, the sense data itself is returned in uscsi_rqbuf. The uscsi_rqresid
provides the residue of the Request Sense data transfer.
Errors EINVAL A parameter has an incorrect, or unsupported, value.
EIO An error occurred during the execution of the command.
EPERM A process without root credentials tried to execute the USCSICMD ioctl.
EFAULT The uscsi_cmd itself, the uscsi_cdb, the uscsi_buf, or the uscsi_rqbuf point to
an invalid address.

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWhea

Interface Stability Committed

See Also ioctl(2), attributes(5), sd(7D), st(7D)

ANSI Small Computer System Interface-2 (SCSI-2)
Warnings The uscsi command is very powerful, but somewhat dangerous, and so its use is restricted to
processes running as root, regardless of the file permissions on the device node. The device
driver code expects to own the device state, and uscsi commands can change the state of the
device and confuse the device driver. It is best to use uscsi commands only with no side
effects, and avoid commands such as Mode Select, as they may cause damage to data stored on
the drive or system panics. Also, as the commands are not checked in any way by the device
driver, any block may be overwritten, and the block numbers are absolute block numbers on
the drive regardless of which slice number is used to send the command.
The uscsi interface is not recommended for very large data transfers (typically more than
16MB). If the requested transfer size exceeds the maximum transfer size of the DMA engine, it
will not be broken up into multiple transfers and DMA errors may result.
The USCSICMD ioctl associates a struct uscsi_cmd with a device by using an open file
descriptor to the device. Other APIs might provide the same struct uscsi_cmd
programming interface, but perform device association in some other manner.

1010 man pages section 7: Device and Network Interfaces • Last Revised 29 May 2007
 usoc(7D)

Name usoc – universal serial optical controller for Fibre Channel arbitrated loop (SOC+) device
driver
Description The Fibre Channel adapter is an SBus card that implements two full duplex Fibre Channel
interfaces. Each interface can connect to a Fibre Channel arbitrated loop (FC-AL). The usoc
device driver is a nexus driver and implements portions of the FC-2 and FC-4 layers of FC-AL.
Files /kernel/drv/usoc 32–bit ELF kernel module
/kernel/drv/sparcv9/usoc 64–bit ELF kernel module

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

Interface stability Unknown

Availability SUNWusoc

See Also fctl(7D), sbus(4), fcp(7D), fp(7d), ssd(7D)

Writing Device Drivers
Fibre Channel Physical and Signaling Interface (FC-PH) ANSI X3.230: 1994
Fibre Channel Arbitrated Loop (FC-AL) ANSI X3.272-1996
Fibre Channel Private Loop SCSI Direct Attach (FC-PLDA) NCITS TR-19:1998
Fabric Channel Loop Attachment (FC-FLA), NCITS TR-20:1998
Diagnostics The following messages are logged and may also appear on the system console. On the console
these messages are preceded by:
usoc%d:

where
usoc%d:

is the per-port instance number of the usoc controller.

Fibre Channel is ONLINE

The Fibre Channel loop is now online.

Fibre Channel Loop is ONLINE

The Fibre Channel loop is now online.

Fibre Channel Loop is OFFLINE

Device and Network Interfaces 1011

usoc(7D)

The Fibre Channel loop is now offline.

attach failed: device in slave-only slot.

Move soc+ card to another slot.

attach failed: alloc soft state.

Driver did not attach, devices will be inaccessible.

attach failed: bad soft state.

Driver did not attach, devices will be inaccessible.

attach failed: unable to map eeprom

Driver was unable to map device memory; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.
attach failed: unable to map XRAM

Driver was unable to map device memory; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.
attach failed: unable to map registers

Driver was unable to map device registers; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.
attach failed: unable to access status register

Driver was unable to map device registers; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.
attach failed: unable to install interrupt handler

Driver was not able to add the interrupt routine to the kernel. Driver did not attach to device,
devices will be inaccessible.
attach failed: unable to access host adapter XRAM

Driver was unable to access device RAM; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.
attach failed: unable to write host adapter XRAM

Driver was unable to write device RAM; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.
attach failed: read/write mismatch in XRAM

Driver was unable to verify device RAM; check for bad hardware. Driver did not attach to
device, devices will be inaccessible.

1012 man pages section 7: Device and Network Interfaces • Last Revised 20 Jul 1999
 uwba(7D)

Name uwba, uwb – Solaris UWB Architecture (UWBA)

Description Ultra-WideBand (UWB) radio technology supports high bandwidth for wireless devices.
UWBA is a miscellaneous module and it supports radio controller drivers for UWB based
devices like HWA (Host Wire Adapter), WHCI (Wireless Host Controller Interface) and so
forth. For example, both HWA radio controller driver (hwarc) and whci driver register to
uwba during attach.

UWBA provides a series of common interfaces for drivers that support UWB radio
technology. Each radio controller driver register itself as a UWB dev to the uwba model in the
attach entry, then other driver or module can control this device to perform the UWB
functions through a list of common interface. For example, a hwahc driver can control the
hwarc driver to scan in a specific channel, start/stop beacon, manage device/MAC address,
and so forth.
Files /kernel/misc/uwba 32-bit ELF 86 kernel module
/kernel/misc/amd64/uwba 64-bit x86 ELF kernel module
/kernel/misc/sparcv9/uwba 64-bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture PCI-based systems

Availability SUNWuwb

See Also attributes(5), hwahc(7D), hwarc(7D), usba(7D)

Writing Device Drivers

ECMA-368 High Rate Ultra Wideband PHY and MAC Standard, 1st Edition

Wireless Host Controller Interface Specification for Certified Wireless Universal Serial Bus,
Version 0.95

Wireless Universal Serial Bus Specification 1.0

Device and Network Interfaces 1013

virtualkm(7D)

Name virtualkm – Virtual keyboard and mouse

Synopsis /dev/kbd
/dev/mouse
#include <sys/types.h>
#include <sys/kbio.h>
int ioctl(int fildes, int command, ... /*arg*/);

Description A virtual keyboard or mouse is an abstraction of one or more physical keyboards or mice
(USB or PS2) connected to a system. Input streams for these physical devices are coalesced
into a single input stream and appear as a single device to the upper layers.
/dev/kbd is the virtual keyboard device file. Inputs from multiple keyboards are coalesced into
a single input stream, meaning that all keyboards appear as a single keyboard to a console or
window system and accordingly, are treated as a single device. The virtual keyboard layout is
consistent with the layout of the first keyboard plugged into the system. Note that on x86
platforms, the virtual keyboard layout can be overloaded by eeprom(1M).
/dev/mouse is the virtual mouse device file. Inputs from multiple mice are coalesced into a
single input stream, meaning that all mice appear as single mouse to the window system.
Commands from applications are dispatched by the virtual keyboard/mouse facility to the
underlying physical devices and will succeed provided that one of the underlying devices
responds with success. For example, a single command issued to turn on LED's will turn on
corresponding LED's for all underlying physical keyboards.
Although physical keyboards/mice are linked to the virtual keyboard/mouse facility, each may
be opened separately by accessing its associated device file. (For example, /dev/usb/hid0 for a
usb mouse). Directly accessing a device file can be useful for multi-seat, gok(1) or similar
purposes.
When a single physical device is opened via its associated device file, it is automatically
removed from the single virtual input stream. When closed, it is automatically re– coalesced
into the single virtual input stream.
Under the virtualkm facility, the PS/2 mouse is coalesced into a virtual mouse single input
stream and can be accessed using the /dev/mouse file. (Note that in previous releases, the PS/2
mouse was accessed via the /dev/kdmouse physical device file). In the current release, you use
the /dev/kdmouse file to directly access the physical PS/2 mouse.
INTERFACES The virtual mouse provides the following event ID's for mouse capability changes:

MOUSE_CAP_CHANGE_NUM_BUT This event is reported when the total number

of mouse buttons changes. The
Firm_event.value is set to the new button
total, which is the maximum number of all
mice buttons. Other fields are ignored.

1014 man pages section 7: Device and Network Interfaces • Last Revised 16 Jan 2007
 virtualkm(7D)

MOUSE_CAP_CHANGE_NUM_WHEEL This event is reported when the total number

of mouse wheels changes. The
Firm_event.value is set to the new wheel
total. Other fields are ignored. The event
value (Firm_event.value) can be 0 (no
wheel), 1 (vertical wheel), or 2 (vertical and
horizontal wheel).

The Firm_event structure is described in <sys/vuid_event.h>. As with other events, firm

events are received using read(2).

Event ID's are used by applications (including certain mouse demo applications) that are
programmed to graphically represent the actual number of buttons and wheels on a mouse.
When an application of this type receives a Firm_event with a ID
MOUSE_CAP_CHANGE_NUM_BUT or MOUSE_CAP_CHANGE_NUM_WHEEL event,
it is instructed to update its state information using the new value. Consider, for example, a
mouse demo application whose sole function is to display a mouse with buttons that
graphically correspond to the actual number of buttons on the mouse. If, for example, the
system has a single two-button USB mouse attached, the application, by default, will
graphically display the mouse with a left and a right button. However, if a another
three-button USB mouse is hot-plugged into the system, a
MOUSE_CAP_CHANGE_NUM_BUT Firm event with Firm_event.value of three instructs
the demo application to update the mouse display to indicate three buttons.
ioctls KIOCSETFREQ Sets the frequency for either keyboard beeper or console beeper. To set
the corresponding beeper frequency, arg must point to a freq_request
structure:
struct freq_request {
enum fr_beep_type type; /* beep type */
int16_t freq; /* frequency */
};
Where type is the corresponding beeper type defined as:
enum fr_beep_type { CONSOLE_BEEP =1, KBD_BEEP =2 };
and freq is the frequency value to be set as the beeper frequency indicated
by type. This value should be between 0 and 32767 with border inclusive.
Files /dev/kbd Virtual Keyboard device file.
/dev/mouse Virtual Mouse device file.
/dev/kdmouse Physical PS/2 mouse device file.
/dev/usb/hid* Physical USB keyboard/mouse device file.
/etc/dacf.conf Device auto-configuration file.

Device and Network Interfaces 1015

virtualkm(7D)

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Availability SUNWckr, SUNWcsd, SUNWusb, SUNWpsdcr,

SUNWcakr.i

Interface Stability Evolving

See Also kbd(1), eeprom(1M), read(2), attributes(5), hid(7D), usba(7D), kb(7M), usbkbm(7M),
usbms(7M), vuidmice(7M)

See gok(1) in the GNOME man pages, available in the SUNWgnome package.

Diagnostics The messages described below may appear on the system console as well as being logged. All
messages are formatted in the following manner:
WARNING: Error message...
conskbd: keyboard is not available for system debugging: device_path.
Errors were encountered while entering kmdb during initialization for debugger mode. As
a result, the keyboard is not available.
conskbd: keyboard is not available: <device_path>
Errors were encountered while exiting kmdb during un-initialization for debugger mode.
As a result, the keyboard is not available.
Failed to relink the mouse <device_path> underneath virtual mouse
An error was encountered and the mouse is unavailable. (When a mouse is physically
opened via a physical device file such as /dev/usb/hid0, it is removed from the single
virtual input stream (/dev/mouse). When closed, it is re-coalesced into a single virtual
input stream beneath /dev/mouse. If an error is encountered, (for example, the mouse has
been physically removed), it is unavailable beneath /dev/mouse.

Notes Currently, the virtualkm device supports only USB and PS2 keyboards and mice.

The virtualkm device maintains complete compatibility on select legacy systems, (including
Ultra 10's), that are equipped with serial keyboard/mouse.

1016 man pages section 7: Device and Network Interfaces • Last Revised 16 Jan 2007
 visual_io(7I)

Name visual_io – Solaris VISUAL I/O control operations

Synopsis #include <sys/visual_io.h>

Description The Solaris VISUAL environment defines a small set of ioctls for controlling graphics and
imaging devices.

The VIS_GETIDENTIFIER ioctl is mandatory and must be implemented in device drivers for
graphics devices using the Solaris VISUAL environment. The VIS_GETIDENTIFIER ioctl is
defined to return a device identifier from the device driver. This identifier must be a
uniquely-defined string.

There are two additional sets of ioctls. One supports mouse tracking via hardware cursor
operations. Use of this set is optional, however, if a graphics device has hardware cursor
support and implements these ioctls, the mouse tracking performance is improved. The
remaining set supports the device acting as the system console device. Use of this set is
optional, but if a graphics device is to be used as the system console device, it must implement
these ioctls.

The VISUAL environment also defines interfaces for non-ioctl entry points into the driver
that the Solaris operating environment calls when it is running in standalone mode (for
example, when using a stand-alone debugger, entering the PROM monitor, or when the
system panicking). These are also known as Polled I/O entry points, which operate under an
explicit set of restrictions, described below.
ioctls VIS_GETIDENTIFIER This ioctl() returns an identifier string to uniquely identify a
device used in the Solaris VISUAL environment. This is a mandatory
ioctl and must return a unique string. We suggest that the name be
formed as <companysymbol><devicetype>.

VIS_GETIDENTIFIER takes a vis_identifier structure as its

parameter. This structure has the form:
#define VIS_MAXNAMELEN 128
struct vis_identifier {
char name[VIS_MAXNAMELEN];
};
VIS_GETCURSOR
VIS_SETCURSOR These ioctls fetch and set various cursor attributes, using the
vis_cursor structure.
struct vis_cursorpos {
short x; /* cursor x coordinate */
short y; /* cursor y coordinate */
};

struct vis_cursorcmap {
int version; /* version */

Device and Network Interfaces 1017

visual_io(7I)

int reserved;
unsigned char *red; /* red color map elements */
unsigned char *green;/* green color map elements */
unsigned char *blue; /* blue color map elements */
};

#define VIS_CURSOR_SETCURSOR 0x01 /* set cursor */

#define VIS_CURSOR_SETPOSITION 0x02 /* set cursor position */
#define VIS_CURSOR_SETHOTSPOT 0x04 /* set cursor hot spot */
#define VIS_CURSOR_SETCOLORMAP 0x08 /* set cursor colormap */
#define VIS_CURSOR_SETSHAPE 0x10 /* set cursor shape */
#define VIS_CURSOR_SETALL \
(VIS_CURSOR_SETCURSOR | VIS_CURSOR_SETPOSITION | \
VIS_CURSOR_SETHOTSPOT | VIS_CURSOR_SETCOLORMAP | \
VIS_CURSOR_SETSHAPE)

struct vis_cursor {
short set; /* what to set */
short enable; /* cursor on/off */
struct vis_cursorpos pos; /* cursor position */
struct vis_cursorpos hot; /* cursor hot spot */
struct vis_cursorcmap cmap; /* color map info */
struct vis_cursorpos size; /* cursor bitmap size */
char *image; /* cursor image bits */
char *mask; /* cursor mask bits */
};

The vis_cursorcmap structure should contain pointers to two elements, specifying the red,
green, and blue values for foreground and background.
VIS_SETCURSORPOS
VIS_MOVECURSOR These ioctls fetch and move the current cursor position, using the
vis_cursorpos structure.

Console Optional Ioctls The following ioctl sets are used by graphics drivers that are part of the system console device.
All of the ioctls must be implemented to be a console device. In addition, if the system does not
have a prom or the prom goes away during boot, the special standalone ioctls (listed below)
must also be implemented.

The coordinate system for the console device places 0,0 at the upper left corner of the device,
with rows increasing toward the bottom of the device and columns increasing from left to
right.
VIS_PUTCMAP
VIS_GETCMAP Set or get color map entries.

The argument is a pointer to a vis_cmap structure, which contains the following fields:

1018 man pages section 7: Device and Network Interfaces • Last Revised 28 Oct 2009
 visual_io(7I)

struct vis_cmap {
int index;
int count;
uchar_t *red;
uchar_t *green;
uchar_t *blue;
}

index is the starting index in the color map where you want to start setting or getting color
map entries.

count is the number of color map entries to set or get. It also is the size of the red, green, and
blue color arrays.

*red, *green, and *blue are pointers to unsigned character arrays which contain the color
map info to set or where the color map info is placed on a get.
VIS_DEVINIT Initializes the graphics driver as a console device.

The argument is a pointer to a vis_devinit structure. The graphics driver is expected to

allocate any local state information needed to be a console device and fill in this structure.
struct vis_devinit {
int version;
screen_size_t width;
screen_size_t height;
screen_size_t linebytes;
unit_t size;
int depth;
short mode;
struct vis_polledio *polledio;
vis_modechg_cb_t modechg_cb;
struct vis_modechg_arg *modechg_arg;
};

version is the version of this structure and should be set to VIS_CONS_REV.

width and height are the width and height of the device. If mode (see below) is VIS_TEXT then
width and height are the number of characters wide and high of the device. If mode is
VIS_PIXEL then width and height are the number of pixels wide and high of the device.

linebytes is the number of bytes per line of the device.

size is the total size of the device in pixels.

depth is the pixel depth in device bits. Currently supported depths are: 1, 4, 8 and 24.

mode is the mode of the device. Either VIS_PIXEL (data to be displayed is in bitmap format) or
VIS_TEXT (data to be displayed is in ascii format).

Device and Network Interfaces 1019

visual_io(7I)

polledio is used to pass the address of the structure containing the standalone mode polled
I/O entry points to the device driver back to the terminal emulator. The vis_polledio
interfaces are described in the Console Standalone Entry Points section of this manpage.
These entry points are where the operating system enters the driver when the system is
running in standalone mode. These functions perform identically to the
VIS_CONSDISPLAY, VIS_CONSCURSOR and VIS_CONSCOPY ioctls, but are called
directly by the Solaris operating environment and must operate under a very strict set of
assumptions.

modechg_cb is a callback function passed from the terminal emulator to the framebuffer driver
which the frame-buffer driver must call whenever a video mode change event occurs that
changes the screen height, width or depth. The callback takes two arguments, an opaque
handle, modechg_arg, and the address of a vis_devinit struct containing the new video mode
information.

modechg_arg is an opaque handle passed from the terminal emulator to the driver, which the
driver must pass back to the terminal emulator as an argument to the modechg_cb function
when the driver notifies the terminal emulator of a video mode change.
VIS_DEVFINI Tells the graphics driver that it is no longer the system console device.
There is no argument to this ioctl. The driver is expected to free any
locally kept state information related to the console.
VIS_CONSCURSOR Describes the size and placement of the cursor on the screen. The
graphics driver is expected to display or hide the cursor at the indicated
position.

The argument is a pointer to a vis_conscursor structure which contains the following fields:
struct vis_conscursor {
screen_pos_t row;
screen_pos_t col;
screen_size_t width;
screen_size_t height
color_t fg_color;
color_t bg_color;
short action;
};

row and col are the first row and column (upper left corner of the cursor).

width and height are the width and height of the cursor.

If mode in the VIS_DEVINIT ioctl is set to VIS_PIXEL, then col, row, width and height are in
pixels. If mode in the VIS_DEVINIT ioctl was set to VIS_TEXT, then col, row, width and height
are in characters.

1020 man pages section 7: Device and Network Interfaces • Last Revised 28 Oct 2009
 visual_io(7I)

fg_color and bg_color are the foreground and background color map indexes to use when
the action (see below) is set to VIS_DISPLAY_CURSOR.

action indicates whether to display or hide the cursor. It is set to either VIS_HIDE_CURSOR or
VIS_DISPLAY_CURSOR.
VIS_CONSDISPLAY Display data on the graphics device. The graphics driver is expected to
display the data contained in the vis_display structure at the specified
position on the console.

The vis_display structure contains the following fields:

struct vis_display {
screen_pos_t row;
screen_pos_t col;
screen_size_t width;
screen_size_t height;
uchar_t *data;
color_t fg_color;
color_t bg_color;
};

row and col specify at which starting row and column the date is to be displayed. If mode in the
VIS_DEVINIT ioctl was set to VIS_TEXT, row and col are defined to be a character offset from
the starting position of the console device. If mode in the VIS_DEVINIT ioctl was set to
VIS_PIXEL, row and col are defined to be a pixel offset from the starting position of the
console device.

width and height specify the size of the data to be displayed. If mode in the VIS_DEVINIT ioctl
was set to VIS_TEXT, width and height define the size of data as a rectangle that is width
characters wide and height characters high. If mode in the VIS_DEVINIT ioctl was set to
VIS_PIXEL, width and height define the size of data as a rectangle that is width pixels wide
and height pixels high.

*data is a pointer to the data to be displayed on the console device. If mode in the VIS_DEVINIT
ioctl was set to VIS_TEXT, data is an array of ASCII characters to be displayed on the console
device. The driver must break these characters up appropriately and display it in the retangle
defined by row, col, width, and height. If mode in the VIS_DEVINIT ioctl was set to VIS_PIXEL,
data is an array of bitmap data to be displayed on the console device. The driver must break
this data up appropriately and display it in the retangle defined by row, col, width, and
height.

The fg_color and bg_color fields define the foreground and background color map indexes
to use when displaying the data. fb_color is used for on pixels and bg_color is used for off
pixels.

Device and Network Interfaces 1021

visual_io(7I)

VIS_CONSCOPY Copy data from one location on the device to another. The driver is
expected to copy the specified data. The source data should not be
modified. Any modifications to the source data should be as a side effect of
the copy destination overlapping the copy source.

The argument is a pointer to a vis_copy structure which contains the following fields:
struct vis_copy {
screen_pos_t s_row;
screen_pos_t s_col;
screen_pos_t e_row;
screen_pos_t e_col;
screen_pos_t t_row;
screen_pos_t t_col;
short direction;
};

s_row, s_col, e_row, and e_col define the source rectangle of the copy. s_row and s_col are
the upper left corner of the source rectangle. e_row and e_col are the lower right corner of the
source rectangle. If mode in the VIS_DEVINIT ioctl() was set to VIS_TEXT, s_row, s_col,
e_row, and e_col are defined to be character offsets from the starting position of the console
device. If mode in the VIS_DEVINIT ioctl was set to VIS_PIXEL, s_row, s_col, e_row, and
e_col are defined to be pixel offsets from the starting position of the console device.

t_row and t_col define the upper left corner of the destination rectangle of the copy. The
entire rectangle is copied to this location. If mode in the VIS_DEVINIT ioctl was set to VIS_TEXT,
t_row, and t_col are defined to be character offsets from the starting position of the console
device. If mode in the VIS_DEVINIT ioctl was set to VIS_PIXEL, t_row, and t_col are defined to
be pixel offsets from the starting position of the console device.

direction specifies which way to do the copy. If direction is VIS_COPY_FORWARD the graphics
driver should copy data from position (s_row, s_col) in the source rectangle to position
(t_row, t_col) in the destination rectangle. If direction is VIS_COPY_BACKWARDS the graphics
driver should copy data from position (e_row, e_col) in the source rectangle to position
(t_row+(e_row-s_row), t_col+(e_col-s_col)) in the destination rectangle.

Console Standalone Console standalone entry points are necessary only if the driver is implementing
Entry Points (Polled I/O console-compatible extensions. All console vectored standalone entry points must be
Interfaces)
implemented along with all console-related ioctls if the console extension is implemented.
struct vis_polledio {
struct vis_polledio_arg *arg;
void (*display)(vis_polledio_arg *, struct vis_consdisplay *);
void (*copy)(vis_polledio_arg *, struct vis_conscopy *);
void (*cursor)(vis_polledio_arg *, struct vis_conscursor *);
};

1022 man pages section 7: Device and Network Interfaces • Last Revised 28 Oct 2009
 visual_io(7I)

The vis_polledio structure is passed from the driver to the Solaris operating environment,
conveying the entry point addresses of three functions which perform the same operations of
their similarly named ioctl counterparts. The rendering parameters for each entry point are
derived from the same structure passed as the respective ioctl. See the Console Optional Ioctls
section of this manpage for an explanation of the specific function each of the entry points,
display(), copy() and cursor() are required to implement. In addition to performing the
prescribed function of their ioctl counterparts, the standalone vectors operate in a special
context and must adhere to a strict set of rules. The polled I/O vectors are called directly
whenever the system is quisced (running in a limited context) and must send output to the
display. Standalone mode describes the state in which the system is running in single-threaded
mode and only one processor is active. Solaris operating environment services are stopped,
along with all other threads on the system, prior to entering any of the polled I/O interfaces.
The polled I/O vectors are called when the system is running in a standalone debugger, when
executing the PROM monitor (OBP) or when panicking.

The following restrictions must be observed in the polled I/O functions:

1. The driver must not allocate memory.
2. The driver must not wait on mutexes.
3. The driver must not wait for interrupts.
4. The driver must not call any DDI or LDI services.
5. The driver must not call any system services.

The system is single-threaded when calling these functions, meaning that all other threads are
effectively halted. Single-threading makes mutexes (which cannot be held) easier to deal with,
so long as the driver does not disturb any shared state. See Writing Device Drivers for more
information about implementing polled I/O entry points.

See Also ioctl(2)

Writing Device Drivers

Notes On SPARC systems, compatible drivers supporting the kernel terminal emulator should
export the tem-support DDI property.tem-support indicates that the driver supports the kernel
terminal emulator. By exporting tem-support it's possible to avoid premature handling of an
incompatible driver.
tem-support This DDI property, set to 1, means driver is compatible with the console
kernel framebuffer interface.

Device and Network Interfaces 1023

vni(7d)

Name vni – STREAMS virtual network interface driver

Description The vni pseudo device is a multi-threaded, loadable, clonable, STREAMS pseudo-device
supporting the connectionless Data Link Provider Interface dlpi(7P) Style 2. Note that DLPI
is intended to interact with IP, meaning that DLPI access to applications is not supported. (For
example, snoop fails on the vni interface.)

The vni device is a software-only interface and does not send or receive data. The device
provides a DLPI upper interface that identifies itself to IP with a private media type. It can be
configured via ifconfig(1M) and can have IP addresses assigned to it, making aliases
possible.

The vni pseudo device is particularly useful in hosting an IP address when used in
conjunction with the 'usesrc' ifconfig option (see ifconfig(1M) for examples). The logical
instances of the device can also be used to host addresses as an alternative to hosting them over
the loopback interface.

Multicast is not supported on this device. More specifically, the following options return an
error when used with an address specified on vni: IP_MULTICAST_IF,
IP_ADD_MEMBERSHIP, IP_DROP_MEMBERSHIP, IPV6_MULTICAST_IF,
IPV6_JOIN_GROUP, IPV6_LEAVE_GROUP. In addition, broadcast is not supported.

Because there is no physical hardware configured below it, no traffic can be received through
nor transmitted on a virtual interface. All packet transmission and reception is accomplished
with existing physical interfaces and tunnels. Because applications that deal with packet
transmission and reception (such as packet filters) cannot filter traffic on virtual interfaces,
you cannot set up a packet filter on a virtual interface. Instead, you should configure the policy
rules to apply to the physical interfaces and tunnels, and if necessary, use the virtual IP
addresses themselves as part of the rule configuration. Also, note that the virtual interface
cannot be part of an IP multipathing (IPMP) group.
Files /dev/vni 64–bit ELF kernel driver

See Also ifconfig(1M), in.mpathd(1M), ip(7P), ip6(7P)

1024 man pages section 7: Device and Network Interfaces • Last Revised 18 July 2004
 vr(7D)

Name vr – driver for VIA Rhine fast Ethernet controllers

Description The vr Fast Ethernet driver is GLD based and supporting the VIA Rhine family of Fast
Ethernet adapters:
pci1106,3043 VIA VT86C100A Fast Ethernet
pci1106,3065 VT6102 VIA Rhine II
pci1106,3106 VT6105 VIA Rhine III
pci1106,3053 VT6105 VIA Rhine III Management Adapter

The vr driver supports IEEE 802.3 auto-negotiation, flow control and VLAN tagging.

Configuration The default configuration is autonegotiation with bidirectional flow control. The advertised
capabilities for autonegotiation are based on the capabilities of the PHY.

You can set the capabilities advertised by the vr controlled device using dladm(1M). The
driver supports a number of parameters, the names of which begin with en_ (enabled). Each of
these boolean parameters determines if the device advertises that mode of operation when the
hardware supports it.

The adv_autoneg_cap parameter controls whether auto-negotiation is performed. If

adv_autoneg_cap is 0, the driver selects the speed/duplex combination from the first
non-zero parameter from this list:
en_100fdx_cap 100Mpbs full duplex
en_100hdx_cap 100Mpbs half duplex
en_10fdx_cap 10Mpbs full duplex
en_10hdx_cap 10Mpbs half duplex

All capabilities default to enabled. Changing any capability parameter causes the link to go
down while the link partners renegotiate the link using the newly changed capabilities.

Limitations The vr driver does not support asymmetric flowcontrol. VT86C100A and Rhine II adapters
are not capable of transmitting flowcontrol messages
Files /dev/vr Special character device
/kernel/drv/vr 32-bit device driver (x86)
/kernel/drv/sparcv9/vr 64-bit device driver (SPARC)
/kernel/drv/amd64/vr 64-bit device driver (x86)

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86

Device and Network Interfaces 1025

vr(7D)

See Also dladm(1M), netstat(1M), driver.conf(4), attributes(5), ieee802.3(5), dlpi(7P),

streamio(7I)

Writing Device Drivers

STREAMS Programmer's Guide

Network Interfaces Programmer's Guide

IEEE 802.3ae Specification - 2002

1026 man pages section 7: Device and Network Interfaces • Last Revised 5 May 2009
 vt(7I)

Name vt – Solaris virtual console interface

Synopsis #include <sys/kd.h>
#include <sys/vt.h>

Description The virtual console device driver — also known as virtual terminal (VT) — is a layer of
management functions that provides facilities to support and switch between multiple screen
faces on a single physical device.

VT's are accessed in the same way as other devices. The open(2) system call is used to open the
virtual console and read(2), write(2) and ioctl(2) are used in the normal way and support
the functionality of the underlying device. In addition, some virtual console-specific ioctls are
provided and described below.

The VT provides a link between different screen faces and the device. The active virtual
console corresponds to the currently visible screen face. Device input is directed to the active
console and any device-specific modes that change on a per virtual terminal basis are set to the
characteristics associated with the active console.

You manage VT's by intercepting keyboard sequences (“hot key”). To maintain consistency
with Xserver, the virtual console device driver supports the Ctrl, Alt, F# and arrow keys.

The sequence AltL + F# (where AltL represents the Alt key and F# represents function keys 1
through 12) is used to select virtual console 1-12. The sequence AltGraph + F# (where
AltGraph represents the right Alt key and F# represent function keys 1 through 12) is for
virtual console 13-24. Alt + F1 chooses the system console (also known as virtual console 1).
The sequence Alt + -> (where “>” represents the right directional arrow) selects the next VT
in a circular ring fashion and Alt + <- ( where “<” represents the left directional arrow)
changes to the previous console in a circular fashion. The sequence Alt + ^ (where “^”
represents the up directional arrow) is for the last used console.

Virtual console switching can be done automatically (VT_AUTO) on receipt of a ‘‘hot-key'' or by

the process owning the VT (VT_PROCESS). When performed automatically, the process
associated with the virtual console is unaware of the switch. Saving and restoring the device
are handled by the underlying device driver and the virtual console manager. Note that
automatic switching is the default mode.

When a ‘‘hot-key'' is sent when in process-controlled switch mode, the process owning the VT
is sent a signal (relsig) it has specified to the virtual console manager (see signal(3C))
requesting the process to release the physical device. At this point, the virtual console manager
awaits the VT_RELDISP ioctl from the process. If the process refuses to release the device
(meaning the switch does not occur), it performs a VT_RELDISP ioctl with an argument of 0
(zero). If the process desires to release the device, it saves the device state (keyboard, display,
and I/O registers) and then performs a VT_RELDISP with an argument of 1 to complete the
switch.

Device and Network Interfaces 1027

vt(7I)

A ring of VT's can contain intermixed auto mode and process control mode consoles. When
an auto mode process becomes active, the underlying device driver and the virtual console
manager handle the restoring of the device. Process control mode processes are sent a
specified signal (acqsig) when they become the active console. The process then restores the
device state (keyboard, display, and I/O registers) and performs VT_RELDISP ioctl with an
argument of VT_ACKACQ to complete the switching protocol.

The modify-operations ioctls (VT_SETMODE, VT_RELDISP, VT_WAITACTIVE, KDSETMODE) check if

the VT is the controlling tty of the calling process. If not, the sys_devices privilege is enforced.
VT_ACTIVATE requires the sys_devices privilege. Note that there is no controlling tty and
privilege check for query/view operations.

ioctls The following ioctls apply to devices that support virtual consoles:
VT_ENABLED
Queries to determine if VT functionality is available on the system. The argument is a
pointer to an integer. If VT functionality is available, the integer is 1, otherwise it is 0.
VT_OPENQRY
Finds an available VT. The argument is a pointer to an integer. The integer is filled in with
the number of the first available console that no other process has open (and hence, is
available to be opened). If there are no available VT's, -1 is filled in.
VT_GETMODE
Determines the VT's current mode, either VT_AUTO or VT_PROCESS. The argument is the
address of the following structure, as defined in <sys/vt.h>
struct vt_mode {
char mode; /* VT mode */
char waitv; /* not used */
short relsig;/* signal to use for release request */
short acqsig;/* signal to use for display acquired */
short frsig;/* not used */
}

/* Virtual console Modes */

#define VT_AUTO 0 /* automatic VT switching */
#define VT_PROCESS 1 /* process controls switching */

The structure will be filled in with the current value

for each field.
VT_SETMODE
Sets the VT mode. The argument is a pointer to a vt_mode structure as defined above. The
structure should be filled in with the desired mode. If process-control mode is specified, the
signals used to communicate with the process should be specified. If any signals are not
specified (value is zero), the signal default is SIGUSR1 (for relsig and acqsig).

1028 man pages section 7: Device and Network Interfaces • Last Revised 22 Sep 2008
 vt(7I)

VT_RELDISP
Tells the VT manager if the process releases (or refuses to release) the display. An argument
of 1 indicates the VT is released. An argument of 0 indicates refusal to release. The
VT_ACKACQ argument indicates if acquisition of the VT has been completed.
VT_ACTIVATE
Makes the VT specified in the argument the active VT (in the same manner as if a hotkey
initiated the switch). If the specified VT is not open or does not exist, the call fails and errno
is set to ENXIO.
VT_WAITACTIVE
If the specified VT is currently active, this call returns immediately. Otherwise, it sleeps
until the specified VT becomes active, at which point it returns.
VT_GETSTATE
Obtains the active VT number and a list of open VTs. The argument is an address to the
following structure:
struct vt_stat {
unsigned short v_active, /* number of the active VT */
v_signal, /* not used */
v_state; /* count of open VTs. For every 1 in this
field, there is an open VT */
}

With VT_GETSTATE, the VT manager first gets the number of the active VT, then
determines the number of open VTs in the system and sets a 1 for each open VT in v_state.
Next, the VT manager transfers the information in structure vt_stat passed by the user
process.
KDGETMODE
Obtains the text/graphics mode associated with the VT.
#define KD_TEXT 0
#define KD_GRAPHICS 1

KDSETMODE
Sets the text/graphics mode to the VT.

KD_TEXT indicates that console text is displayed on the screen. Normally KD_TEXT is
combined with VT_AUTO mode for text console terminals, so that the console text display
automatically is saved and restored on the hot key screen switches.

KD_GRAPHICS indicates that the user/application (usually Xserver) has direct control of the
display for this VT in graphics mode. Normally KD_GRAPHICS is combined with
VT_PROCESS mode for this VT indicating direct control of the display in graphics mode. In
this mode, all writes to the VT using the write system call are ignored, and you must save
and restore the display on the hot key screen switches.

Device and Network Interfaces 1029

vt(7I)

When the mode of the active VT is changed from KD_TEXT to KD_GRAPHICS or a VT of

KD_GRAPHICS mode is made active from a previous active VT of KD_TEXT mode, the virtual
console manager initiates a KDSETMODE ioctl with KD_GRAPHICS as the argument to the
underlying console frame buffer device indicating that current display is running into
graphics mode.

When the mode of the active VT is changed from KD_GRAPHICS to KD_TEXT or a VT of

KD_TEXT mode is actived from a previous active VT of KD_GRAPHICS mode, the virtual
console manager initiates a KDSETMODE ioctl with KD_TEXT as the argument to the
underlying console frame buffer device indicating that current display is running into
console text mode.
Files /dev/vt/# VT devices.

See Also ioctl(2), signal(3C), wscons(7D)

Notes By default, there are only five virtual console instance login prompts running on /dev/vt/#
(where “#” represents 2 to 6) in addition to the system console running on /dev/console.
Normally Xorg uses the seventh virtual console (/dev/vt/7.) To switch from consoles to
Xserver (which normally picks up the first available virtual console), use [ Ctrl + ] Alt + F7 .
# svcs | grep login
online 17:49:11 svc:/system/console-login:default
online 17:49:11 svc:/system/console-login:vt2
online 17:49:11 svc:/system/console-login:vt3
online 17:49:11 svc:/system/console-login:vt4
online 17:49:11 svc:/system/console-login:vt5
online 17:49:11 svc:/system/console-login:vt6

console-login:default is for the system console, others for

virtual consoles.

You can modify properties/disable/enable and remove/add

virtual consoles using smf(5):

# svccfg -s console-login add vt8

# svccfg -s console-login:vt8 setprop ttymon/device=astring: "/dev/vt/8"
# svcadm enable console-login:vt8

1030 man pages section 7: Device and Network Interfaces • Last Revised 22 Sep 2008
 vuidmice(7M)

Name vuidmice, vuidm3p, vuidm4p, vuidm5p, vuid2ps2, vuid3ps2 – converts mouse protocol to
Firm Events
Synopsis #include <sys/stream.h>
#include <sys/vuid_event.h>
#include <sys/vuid_wheel.h>
int ioctl(fd, I_PUSH, vuidm3p);
int ioctl(fd, I_PUSH, vuidm4p);
int ioctl(fd, I_PUSH, vuidm5p);
int ioctl(fd, I_PUSH, vuid2ps2);
int ioctl(fd, I_PUSH, vuid3ps2);

Description The STREAMS modules vuidm3p, vuidm4p, vuidm5p, vuid2ps2, and vuid3ps2 convert mouse
protocols to Firm events. The Firm event structure is described in <sys/vuid_event.h>.
Pushing a STREAMS module does not automatically enable mouse protocol conversion to
Firm events. The STREAMS module state is initially set to raw or VUID_NATIVE mode which
performs no message processing. You must change the state to VUID_FIRM_EVENT mode to
initiate mouse protocol conversion to Firm events. This can be accomplished by the following
code:
int format;
format = VUID_FIRM_EVENT;
ioctl(fd, VUIDSFORMAT, &format);

You can also query the state of the STREAMS module by using the VUIDGFORMAT option.
int format;
int fd; /* file descriptor */
ioctl(fd, VUIDGFORMAT, &format);
if ( format == VUID_NATIVE );
/* The state of the module is in raw mode.
* Message processing is not enabled.
*/
if ( format == VUID_FIRM_EVENT );
/* Message processing is enabled.
* Mouse protocol conversion to Firm events
* are performed.

The remainder of this section describes the processing of STREAMS messages on the read-
and write-side.
Read Side Behavior M_DATA Incoming messages are queued and converted to Firm events.
M_FLUSH The read queue of the module is flushed of all its data messages and all data in
the record being accumulated are also flushed. The message is passed upstream.

Device and Network Interfaces 1031

vuidmice(7M)

Write Side Behavior M_IOCTL Messages sent downstream as a result of an ioctl(2) system call. The two
valid ioctl options processed by the vuidmice modules are VUIDGFORMAT
and VUIDSFORMAT.
M_FLUSH The write queue of the module is flushed of all its data messages and the
message is passed downstream.
VUIDGFORMAT This option returns the current state of the STREAMS module. The state of
the vuidmice STREAMS module may either be VUID_NATIVE (no message
processing) or VUID_FIRM_EVENT (convert to Firm events).
VUIDSFORMAT This option sets the state of the STREAMS module to VUID_FIRM_EVENT. If
the state of the STREAMS module is already in VUID_FIRM_EVENT, this
option is non-operational. It is not possible to set the state back to
VUID_NATIVE once the state becomes VUID_FIRM_EVENT. To disable
message processing, pop the STREAMS module out by calling ioctl(fd,
1I_POP, vuid*).

The following wheel support ioctls are defined for PS/2 mouse only:
VUIDGWHEELCOUNT This ioctl takes a pointer to an integer as argument and sets the
value of the integer to the number of wheels available on this
device.
VUIDGWHEELINFO This command returns static information about the wheel that
does not change while a device is in use. Currently the only
information defined is the wheel orientation which is either
VUID_WHEEL_FORMAT_VERTICAL or
VUID_WHEEL_FORMAT_HORIZONTAL.
typedef struct {
int vers;
int id;
int format;
} wheel_info;
The ioctl takes a pointer to "wheel_info" structure with the
"vers" set to the current version of the "wheel_info" structure
and "id" set to the id of the wheel for which the information is
desired.
VUIDSWHEELSTATE
VUIDGWHEELSTATE VUIDSWHEELSTATE sets the state of the wheel to that
specified in the stateflags. VUIDGWHEELSTATE returns the
current state settings in the stateflags field.

stateflags is an OR'ed set of flag bits. The only flag currently

defined is VUID_WHEEL_STATE_ENABLED.

1032 man pages section 7: Device and Network Interfaces • Last Revised 21 June 2005
 vuidmice(7M)

When stateflags is set to VUID_WHEEL_STATE_ENABLED

the module converts motion of the specified wheel into VUID
events and sends those up stream.

Wheel events are disabled by default.

Applications that want to change a flag should first get the

current flags and then change only the bit they want.
typedef struct {
int vers;
int id;
uint32_t stateflags;
} wheel_state;
These ioctls take pointer to 'wheel_state' as an argument with
the 'vers' and 'id' members filled up. These members have the
same meaning as that for 'VUIDGWHEELINFO' ioctl.

Mouse Configurations Module Protocol Type Device

vuidm3p 3-Byte Protocol Microsoft 2 Button Serial /dev/tty*

Mouse

vuidm4p 4-Byte Protocol Logitech 3 Button Mouseman /dev/tty*

vuidm5p Logitech 3 Button Bus Mouse Microsoft Bus /dev/logi/ dev/msm

Mouse

vuid2ps2 PS/2 Protocol 2 Button PS/2 Compatible /dev/kdmouse

Mouse

vuid3ps2 PS/2 Protocol 3 Button PS/2 Compatible /dev/kdmouse

Mouse

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also attributes(5), virtualkm(7D)

STREAMS Programming Guide

Device and Network Interfaces 1033

wpi(7D)

Name wpi – Intel Pro Wireless 802.11a/b/g 3945 driver

Description The wpi 802.11b/g wireless NIC driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the Intel Pro Wireless 3945ABG chipset-based NIC's.

Configuration The wpi driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec.
Files /dev/wpi
Special character device.
/kernel/drv/wpi
32–bit ELF kernel module (x86).
/kernel/drv/amd64/wpi
64–bit ELF kernel module (x86).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWwpi

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 — Wireless LAN Media Access Control and Physical Layer Specification — IEEE, 2001

1034 man pages section 7: Device and Network Interfaces • Last Revised 23 Apr 2007
 wscons(7D)

Name wscons – workstation console

Synopsis #include <sys/strredir.h>
ioctl(fd, SRIOCSREDIR, target);
ioctl(fd, SRIOCISREDIR, target);

Description The wscons workstation console consists of a workstation keyboard and frame buffer that act
together to emulate an ASCII terminal. It includes a redirection facility that allows I/O issued
to the workstation console to be diverted to a STREAMS device, enabling window systems to
redirect output that would otherwise appear directly on the frame buffer in corrupted form.

Redirection The wscons redirection facility maintains a list of devices that are designated as redirection
targets through the SRIOCSREDIR ioctl described below. Only the current entry is active; when
the active entry is closed, the most recent remaining entry becomes active. The active entry
acts as a proxy for the device being redirected and handles all read(2), write(2), ioctl(2), and
poll(2) calls issued against the redirectee.

The ioctls described below control the redirection facility. In both cases, fd is a descriptor for
the device being redirected (or workstation console) and target is a descriptor for a STREAMS
device.
SRIOCSREDIR Designates target as the source and destination of I/O ostensibly directed
to the device denoted by fd.
SRIOCISREDIR Returns 1 if target names the device currently acting as proxy for the device
denoted by fd, and 0 if it is not.

ANSI Standard Terminal The Solaris kernel terminal emulator provides ANSI X3.64 emulation both on SPARC and x86
Emulation systems.

On SPARC systems, the PROM monitor is used to emulate an ANSI X3.64 terminal if the
kernel terminal emulator is not available for emulation. See visual_io(7I) for more details.

Note: The VT100 adheres the ANSI X3.64 standard. However, because the VT100 features
nonstandard extensions to ANSI X3.64, it is incompatible with Sun terminal emulators.

The SPARC console displays 34 lines of 80 ASCII characters per line. The x86 console displays
25 lines of 80 ASCII characters per line. Devices with smaller text capacities may display less.
On SPARC systems, the screen-#rows screen-#columns should be set to 34 or 80
respectively or text capacities will vary from those described above. On SPARC systems, the
screen-#rows and screen-#columns fields are stored in NVRAM/EEPROM. See
eeprom(1M) for more information. Both SPARC and x86 consoles offer scrolling, (x, y) cursor
addressing ability and a number of other control functions.

The console cursor marks the current line and character position on the screen. ASCII
characters between 0x20 (space) and 0x7E (tilde) inclusive are printing characters. When a

Device and Network Interfaces 1035

wscons(7D)

print character is written to the console (and is not part of an escape sequence), it is displayed
at the current cursor position and the cursor moves one position to the right on the current
line.

On SPARC based systems, later PROM revisions have the full 8-bit ISO Latin-1 (ISO 8859-1)
character set. Earlier PROM revisions display characters in the range 0xA0 through 0xFE as
spaces.

When the cursor is at the right edge of the screen, it moves to the first character position on
the next line. When the cursor is at the screen's right-bottom edge, the line-feed function is
performed (see CTRL-J below). The line-feed function scrolls the screen up by one or more
lines before moving the cursor to the first character position on the next line.

Control Sequence The wscons console defines a number of control sequences that may occur during input.
Syntax When a control sequence is written to the console, it affects one of the control functions
described below. Control sequences are not displayed on screen.

A number of control sequences (or control character functions) are of the form:

CTRL-x

where x represents a singe character., such as CNTRL-J for a line feed.

Other ANSI control sequences are of the form:

ESC [ params char

Note – Spaces are included only for readability; these characters must occur in the given
sequence without the intervening spaces.
ESC ASCII escape character (ESC, CTRL-[, 0x1B).
[ Left square bracket ‘[' (0x5B).
params Sequence of zero or more decimal numbers made up of digits between 0 and 9,
separated by semicolons. Parameters are represented by n in the syntax
descriptions for escape sequence functions.
char Function character, which is different for each control sequence and it
represented by x in the syntax descriptions for control character functions.

In the following examples of syntactically valid escape sequences, ESC represent the single
ASCII character, Escape:
ESC[m Select graphic rendition with default parameter
ESC[7m Select graphic rendition with reverse image
ESC[33;54H Set cursor position

1036 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2006
 wscons(7D)

ESC[123;456;0;;3;B Move cursor down

Syntactically valid control characters and ANSI escape sequences that are not currently
interpreted by the console are ignored.

Each control function requires a specified number of parameters. If fewer parameters are
supplied, the remaining parameters (with certain exceptions noted below) default to 1. If more
parameters are supplied, the first n parameters are used by kernel terminal emulator. In
contrast, only the last n parameters are used by PROM based emulator, where n is the number
required by that particular command character.

Parameters which are omitted or set to 0 are reset to the default value of 1 (with certain
exceptions). For example, the command character M requires one parameter. ESC[;M, ESC[0M,
ESC[M and ESC[23;15;32;1M are all equivalent to ESC[1M and provide a parameter value of 1.
Note that ESC[;5M (interpreted as ‘ESC[5M') is not equivalent to ESC[5;M (interpreted as
‘ESC[5;1M') which is ultimately interpreted as ‘ESC[1M').

ANSI Control Functions The following paragraphs specify the ANSI control functions implemented by the console.
Each description provides:
■ Control sequence syntax
■ Hexadecimal equivalent of control characters where applicable
■ Control function name and ANSI or Sun abbreviation (if any).
■ Description of parameters required, if any
■ Description of the control function
■ Initial setting of the mode for functions that set a mode. To restore the initial settings, use
the SUNRESET escape sequence.

Control Character The wscons control character functions are:

Functions
Bell (BEL),
CTRL-G
0x7 Used for consoles that are not equipped with an audible bell. Current
Sun workstation models also flash the screen if the keyboard is not the
console input device.
Backspace (BS),
CTRL-H,
0x8 The cursor moves one position to the left on the current line. If it is
already at the left edge of the screen, no change takes place.
Tab (TAB),
CTRL-I,
0x9 The cursor moves right on the current line to the next tab stop. The tab
stops are fixed at every multiple of eight columns. If the cursor is

Device and Network Interfaces 1037

wscons(7D)

already at the right edge of the screen, nothing change takes place.
Otherwise, the cursor moves right a minimum of one and a maximum
of eight character positions.
Line-feed (LF),
CTRL-J,
0xA The cursor, while remaining at the same character position on the line,
moves down one line. If the cursor is at the bottom line, the screen
either scrolls up or wraps around depending on the setting of an
internal variable n (initially 1) . The internal variable can be changed
using the ESC[r control sequence. If n is greater than zero, the entire
screen (including the cursor) is scrolled up by n lines before executing
the line-feed. The top n lines scroll off the screen and are lost. New
blank lines n scroll onto the bottom of the screen. After scrolling, move
the cursor down one line to execute the line feed.

If n is zero, wrap-around mode is entered. The ESC [ 1 r exits back to

scroll mode. If a line-feed occurs on the bottom line in wrap mode, the
cursor goes to the same character position in the top line of the screen.
During line-feeds, the line that the cursor moves to is cleared and no
scrolling occurs. Wrap-around mode is not implemented in the
window system.

On SPARC based systems, the speed at which the screen scrolls is

dependent on the amount of data waiting to be printed. Whenever a
scroll occurs and the console is in normal scroll mode (ESC [ 1 r), it
scans the rest of the data awaiting printing to see how many line-feeds
occur in it. This scan stops when the console finds a control character
from the set {VT, FF, SO, SI, DLE, DC1, DC2, DC3, DC4, NAK, SYN, ETB, CAN,
EM, SUB, ESC, FS, GS, RS, US} . At that point, the screen is scrolled by n
lines (n ≥ 1) and processing continues. The scanned text is processed
normally and fills in the newly created lines. As long as escape codes or
other control characters are not intermixed with the text, this results in
faster scrolling
Reverse Line-feed,
CTRL-K,
0xB With kernel terminal emulator (while remaining at the same character
position on the line), the cursor moves down one line. However, with
PROM based emulator (while remaining at the same character
position on the line), the cursor moves up one line. If the cursor is
already at the top line, no change takes place.
Form-feed (FF)
CTRL-L,

1038 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2006
 wscons(7D)

0xC The cursor is positioned to the home position (upper-left corner) and
the entire screen is cleared.
Return (CR),
CTRL-M,
0xD The cursor moves to the leftmost character position on the current
line.

Escape Sequence The wscons escape sequence functions are:

Functions
Escape (ESC),
CTRL-[,
0x1B The escape character. Escape initiates a
multi-character control sequence.
Insert Character (ICH)
ESC[#@ Takes one parameter, n (default 1). Inserts n
spaces at the current cursor position. The
current line, starting at the current cursor
position inclusive, is shifted to the right by n
character positions to make room for the spaces.
The rightmost n character positions shift off the
line and are lost. The position of the cursor is
unchanged.
Cursor Up (CUU),
ESC[#A Takes one parameter, n (default 1). Moves the
cursor up n lines. If the cursor is fewer than n
lines from the top of the screen, moves the
cursor to the topmost line on the screen. The
character position of the cursor on the line is
unchanged.
Cursor Down (CUD),
ESC[#B Takes one parameter, (default 1). Moves the
cursor down n lines. If the cursor is fewer than n
lines from the bottom of the screen, move the
cursor to the last line on the screen. The
character position of the cursor on the line is
unchanged.
Cursor Forward (CUF),
ESC[#C Takes one parameter, n (default 1). Moves the
cursor to the right by n character positions on
the current line. If the cursor is fewer than n

Device and Network Interfaces 1039

wscons(7D)

positions from the right edge of the screen,

moves the cursor to the rightmost position on
the current line.
Cursor Backward (CUB),
ESC[#D Takes one parameter, n (default 1). Moves the
cursor to the left by n character positions on the
current line. If the cursor is fewer than n
positions from the left edge of the screen, moves
the cursor to the leftmost position on the current
line.
Cursor Next Line (CNL),
ESC[#E Takes one parameter, n (default 1). Positions the
cursor at the leftmost character position on the
n-th line below the current line. If the current
line is less than n lines from the bottom of the
screen, positions the cursor at the leftmost
character position on the bottom line.
Horizontal and Vertical Position (HVP),
ESC[#1;#2f or
Cursor Position (CUP),
ESC[#1;#2H Takes two parameters, n1 and n2 (default 1, 1).
Moves the cursor to the n2-th character position
on the n1-th line. Character positions are
numbered from 1 at the left edge of the screen;
line positions are numbered from 1 at the top of
the screen. Hence, if both parameters are
omitted, the default action moves the cursor to
the home position (upper left corner). If only
one parameter is supplied, the cursor moves to
column 1 of the specified line.
Erase in Display (ED),
ESC[J Takes no parameters. Erases from the current
cursor position inclusive to the end of the screen,
that is, to the end of the current line and all lines
below the current line. The cursor position is
unchanged.
Erase in Line (EL),
ESC[K Takes no parameters. Erases from the current
cursor position inclusive to the end of the
current line. The cursor position is unchanged.

1040 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2006
 wscons(7D)

Insert Line (IL),

ESC[#L Takes one parameter, n (default 1). Makes room
for n new lines starting at the current line by
scrolling down by n lines the portion of the
screen from the current line inclusive to the
bottom. The n new lines at the cursor are filled
with spaces; the bottom n lines shift off the
bottom of the screen and are lost. The position of
the cursor on the screen is unchanged.
Delete Line (DL),
ESC[#M Takes one parameter, n (default 1). Deletes n
lines beginning with the current line. The
portion of the screen from the current line
inclusive to the bottom is scrolled upward by n
lines. The n new lines scrolling onto the bottom
of the screen are filled with spaces; the n old lines
beginning at the cursor line are deleted. The
position of the cursor on the screen is
unchanged.
Delete Character (DCH),
ESC[#P Takes one parameter, n (default 1). Deletes n
characters starting with the current cursor
position. Shifts the tail of the current line to the
left by n character positions from the current
cursor position, inclusive, to the end of the line.
Blanks are shifted into the rightmost n character
positions. The position of the cursor on the
screen is unchanged.
Select Graphic Rendition (SGR),
ESC[#m Takes one parameter, n (default 0). Note that
unlike most escape sequences, the parameter
defaults to zero if omitted. Invokes the graphic
rendition specified by the parameter. All
following printing characters in the data stream
are rendered according to the parameter until
the next occurrence of this escape sequence in
the data stream. With PROM-based emulator,
only two graphic renditions are defined:
0 Normal rendition
7 Negative (reverse) image

Device and Network Interfaces 1041

wscons(7D)

Negative image displays characters as

white-on-black if the screen mode is currently
black-on white, and vice-versa. Any non-zero
value of n is currently equivalent to 7 and selects
the negative image rendition.

In addition to the two renditions mentioned

above, the following ISO 6429-1983 graphic
rendition values support color text with kernel
terminal emulator:
30 black foreground
31 red foreground
32 green foreground
33 brown foreground
34 blue foreground
35 magenta foreground
36 cyan foreground
37 white foreground
40 black background
41 red background
42 green background
43 brown background
44 blue background
45 magenta background
46 cyan background
47 white background
Black On White (SUNBOW),
ESC[p Takes no parameters. On SPARC, sets the screen
mode to black-on-white. If the screen mode is
already black-on-white, has no effect. In this
mode, spaces display as solid white, other
characters as black-on-white. The cursor is a
solid black block. Characters displayed in
negative image rendition (see ‘Select Graphic
Rendition' above) are white-on-black. This

1042 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2006
 wscons(7D)

comprises the initial setting of the screen mode

on reset. On x86 systems, use ESC[q to set
black-on-white.
White On Black (SUNWOB),
ESC[q Takes no parameters. On SPARC, sets the screen
mode to white-on-black. If the screen mode is
already white-on-black, has no effect. In this
mode spaces display as solid black, other
characters as white-on-black. The cursor is a
solid white block. Characters displayed in
negative image rendition (see ‘Select Graphic
Rendition' above) are black-on-white. Initial
setting of the screen mode on reset is black on
white. On x86 systems, use ESC[p to set
white-on-black.
ESC[#r
Set Scrolling (SUNSCRL) Takes one parameter, n (default 0). Sets to n an
internal register which determines how many
lines the screen scrolls up when a line-feed
function is performed with the cursor on the
bottom line. A parameter of 2 or 3 introduces a
small amount of jump when a scroll occurs. A
parameter of 34 clears the screen rather than
scrolling. The initial setting is 1 on reset.

A parameter of zero initiates wrap mode instead

of scrolling. If a linefeed occurs on the bottom
line during wrap mode, the cursor goes to the
same character position in the top line of the
screen. When a line feed occurs, the line that the
cursor moves to is cleared and no scrolling
occurs. ESC [ 1 r exits back to scroll mode.

For more information, see the description of the

Line-feed (CTRL-J) control function above.
ESC[s
Reset terminal emulator (SUNRESET) Takes no parameters. Resets all modes to default,
restores current font from PROM. Screen and
cursor position are unchanged.

Return Values When there are no errors, the redirection ioctls have return values as described above.
Otherwise, they return −1 and set errno to indicate the error. If the target stream is in an error
state, errno is set accordingly.

Device and Network Interfaces 1043

wscons(7D)

If the target stream is in an error state, errno is set accordingly.

Errors EBADF target does not denote an open file.
ENOSTR target does not denote a STREAMS device.
Files /dev/wscons Workstation console, accessed via the redirection facility
/dev/systty Devices that must be opened for the SRIOCSREDIR and SRIOCISREDIR
ioctls.
/dev/syscon Access system console
/dev/console Access system console

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Interface Stability Stable

See Also cvcd(1M), eeprom(1M), ioctl(2), poll(2), read(2), write(2), cvc(7D), console(7D),
visual_io(7I)

Warnings The redirection ioctls block while there is I/O outstanding on the device instance being
redirected. If you try to redirect the workstation console while there is a outstanding read, the
workstation console will hang until the read completes.

Notes The cvc facility supersedes the SunOS wscons facility and should not be used with wscons.

1044 man pages section 7: Device and Network Interfaces • Last Revised 26 May 2006
 wusb_ca(7D)

Name wusb_ca – WUSB Cable Association Driver

Description The wusb_ca driver is a USBA (Solaris USB Architecture) compliant client driver that
supports the cable association model which is defined in Association Models Supplement to
the Certified WUSB specification.

The wireless USB cable association driver is a USB class driver that provides interfaces for
establishing a first-time connection between Wireless USB hosts and devices. This process of
establishing a first-time connection is called association in WUSB standard. It is a prerequisite
process that must be completed by hosts and devices prior to implementing the security
requirements outlined in Wireless Universal Serial Bus Specification 1.0.

Users should use wusbadm(1M) to do cable association for WUSB devices.

Files /kernel/drv/wusb_ca 32-bit ELF 86 kernel module
/kernel/drv/amd64/wusb_ca 64-bit x86 ELF kernel module
/kernel/drv/sparcv9/wusb_ca 64-bit SPARC ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also wusbadm(1M), attributes(5), hwahc(7D), hwarc(7D), usba(7D)

Writing Device Drivers

System Administration Guide: Basic Administration

Wireless Universal Serial Bus Specification 1.0

http://www.usb.org

http://www.sun.com

Device and Network Interfaces 1045

wusb_df(7D)

Name wusb_df, hwa1480_fw – WUSB firmware download driver and firmware module for Intel
i1480 chipset

Description The wusb_df driver is a Solaris USB Architecture (USBA) compliant client driver that is used
to download firmware for Host Wire Adapter (HWA) dongles that use Intel i1480 chipsets.

Currently, the wusb_df driver can only download driver for Intel i1480 based HWA dongles.
The hwa1480_fw is a miscellaneous module which is transformed from Intel's firmware binary
version RC1.3PA2-20070828. wusb_df reads firmware data from hwa1480_fw module and
downloads it to HWA hardware.

Users can use elfwrap(1) to transform new firmware binary. Users must use the same name as
hwa1480_fw, since wusb_df only recognizes this symbol.
Files /kernel/drv/wusb_df 32-bit ELF 86 kernel module
/kernel/drv/sparcv9/wusb_df 64-bit SPARC ELF kernel module
/kernel/drv/amd64/wusb_df 64-bit x86 ELF kernel module
/kernel/misc/hwa_1480 32-bit ELF 86 kernel module
/kernel/misc/sparcv9/hwa_1480 64-bit SPARC ELF kernel module
/kernel/drv/amd64/hwa_1480 64-bit x86 ELF kernel module

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC, x86, PCI-based systems

Availability SUNWusb

See Also elfwrap(1)add_drv(1M), rem_drv(1M), update_drv(1M), attributes(5)

Writing Device Drivers

System Administration Guide: Basic Administration

Wireless Universal Serial Bus Specification 1.0

http://www.usb.org

http://www.sun.com

1046 man pages section 7: Device and Network Interfaces • Last Revised 18 Sep 2009
 xge(7D)

Name xge – Neterion Xframe 10Gigabit Ethernet Network Adapter driver

Synopsis /dev/xge

Description The xge 10 Gigabit Ethernet driver is a multi-threaded, loadable, clonable, GLD-based
STREAMS driver supporting the Data Link Provider Interface, dlpi(7P), on S2IO Xframe
10-Gigabit Ethernet Network Adapter.

The xge driver functions includes controller initialization, frame transmit and receive,
promiscuous and multicast support, TCP and UDP checksum offload (IPv4 and IPv6),
9622-byte jumbo frame, and error recovery and reporting.

The xge driver and hardware support the 10GBase-SR/W, LR/W, and ER/W 802.3 physical
layer.

Application The cloning, character-special device /dev/xge is used to access all Xframe devices installed
Programming within the system.
Interface
The xge driver is managed by the dladm(1M) command line utility, which allows VLANs to be
defined on top of xge instances and for xge instances to be aggregated. See dladm(1M) for
more details.

The values returned by the driver in the DL_INFO_ACK primitive in response to the
DL_INFO_REQ are as follows:
■ Maximum SDU is 9600.
■ Minimum SDU is 0.
■ DSLAP address length is 8 bytes.
■ MAC type is DL_ETHER.
■ SAP length value is -2 meaning the physical address component is followed immediately
by a 2-byte sap component within the DLSAP address.
■ Broadcast address value is Ethernet/IEEE broadcast address (FF:FF:FF:FF:FF:FF).

Configuration By default, the xge driver works without any configuration file.

You can check the running-time status of a device instance using ndd(1M). Currently, the
driver provides an interface to print all hardware statistics.

For example, to print statistics of device xge0:

#ndd /dev/xge0 stats
tmac_data_octets 772
tmac_frms 15
tmac_drop_frms 0
tmac_bcst_frms 6
tmac_mcst_frms 6

Device and Network Interfaces 1047

xge(7D)

...

rmac_vld_frms 13
rmac_fcs_err_frms 0
rmac_drop_frms 0
rmac_vld_bcst_frms 7
rmac_vld_mcst_frms 11
rmac_out_rng_len_err_frms 0
rmac_in_rng_len_err_frms 0
rmac_long_frms 0

...

not_traffic_intr_cnt 242673
traffic_intr_cnt 28

...
Files /dev/xge xge special character device

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also dladm(1M), ndd(1M), attributes(5), streamio(7I), dlpi(7P)

Writing Device Drivers

STREAMS Programming Guide

Network Interfaces Programmer's Guide

1048 man pages section 7: Device and Network Interfaces • Last Revised 3 Oct 2005
 yge(7D)

Name yge – Marvell Yukon 2 Ethernet device driver

Synopsis /dev/net/yge

Description The yge driver supports Marvell Yukon 2 Fast Ethernet and Gigabit Ethernet controllers.

Properties The following properties can be configured using dladm(1M):

adv_autoneg_cap Enables (default) or disables IEEE 802.3 auto-negotiation of link speed

and duplex settings. If enabled, the device negotiates among the
supported (and configured, see below) link options with the link
partner. If disabled, at least one of the link options below must be
specified. The driver selects the first enabled link option according to
the IEEE 802.3 specified preferences.
adv_1000fdx_cap Enables the 1000 Mbps full-duplex link option.
adv_1000hdx_cap Enables the 1000 Mbps half-duplex link option.
adv_100T4_cap Enables the 100 BaseT4 link option. (Note that most hardware does not
support this unusual link style. Also, this uses two pairs of wires for
data, rather than one.)
adv_100fdx_cap Enables the 1000 Mbps full-duplex link option.
adv_100hdx_cap Enables the 1000 Mbps half-duplex link option.
adv_10fdx_cap Enables the 10 Base-T full-duplex link option.
adv_10hdx_cap Enables the 10 Base-T half-duplex link option.
mtu On most devices, can be set between 1500 (default) and 9000. This
property can only be changed when the device is not in use.
Files /dev/net/yge Special network device
/kernel/drv/yge 32-bit driver binary (x86)
/kernel/drv/amd64/yge 64-bit driver binary (x86)
/kernel/drv/sparcv9/yge 64-bit driver binary (SPARC)

Attributes See attributes(5) for a descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86, SPARC

Interface Stability Committed

Device and Network Interfaces 1049

yge(7D)

See Also dladm(1M), attributes(5), dlpi(7P)

1050 man pages section 7: Device and Network Interfaces • Last Revised 15 Sep 2009
 zcons(7D)

Name zcons – Zone console device driver

Description The zcons character driver exports the console for system zones. The driver is comprised of
two "sides:" a master side with which applications in the global zone communicate, and a slave
side, which receives I/O from the master side. The slave side is available in the global zones.

Applications must not depend on the location of /dev or /devices entries exported by zcons.
Inside a zone, the zcons slave side is fronted by /dev/console and other console-related
symbolic links, which are used by applications that expect to write to the system console.

The zcons driver is Sun Private, and may change in future releases.
Files /dev/zcons/<zonename>/masterconsole Global zone master side console for zone
<zonename>.
/dev/zcons/<zonename>/slaveconsole Global zone slave side console for zone
<zonename>.
/dev/zconsole Non-global zone console (slave side).

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWcsu

Interface Stability Sun Private

See Also zoneadm(1M), zonecfg(1M), attributes(5), zones(5)

Device and Network Interfaces 1051

zero(7D)

Name zero – source of zeroes

Description A zero special file is a source of zeroed unnamed memory.

Reads from a zero special file always return a buffer full of zeroes. The file is of infinite length.

Writes to a zero special file are always successful, but the data written is ignored.

Mapping a zero special file creates a zero-initialized unnamed memory object of a length equal
to the length of the mapping and rounded up to the nearest page size as returned by sysconf.
Multiple processes can share such a zero special file object provided a common ancestor
mapped the object MAP_SHARED.
Files /dev/zero

See Also fork(2), mmap(2), sysconf(3C)

1052 man pages section 7: Device and Network Interfaces • Last Revised 3 Jul 1990
 zs(7D)

Name zs – Zilog 8530 SCC serial communications driver

Synopsis #include <fcntl.h>
#include <sys/termios.h>
open("/dev/term/n", mode);
open("/dev/ttyn", mode);
open("/dev/cua/n", mode);

Description The Zilog 8530 provides two serial input/output channels capable of supporting a variety of
communication protocols. A typical system uses two or more of these devices to implement
essential functions, including RS-423 ports (which also support most RS-232 equipment), and
the console keyboard and mouse devices.

The zs module is a loadable STREAMS driver that provides basic support for the Zilog 8530
hardware and basic asynchronous communication support. The driver supports the
termio(7I) device control functions specified by flags in the c_cflag word of the termios
structure and by the IGNBRK, IGNPAR, PARMRK, or INPCK flags in the c_iflag word. All other
termio(7I) functions must be performed by STREAMS modules pushed atop the driver.
When a device is opened, the ldterm(7M) and ttcompat(7M) STREAMS modules are
automatically pushed on top of the stream, providing the standard termio(7I) interface.

The character-special devices /dev/term/a and /dev/term/b are used to access the two serial
ports on the CPU board.

Valid name space entries are /dev/cua/[a-z], /dev/term/[a-z] and /dev/tty[a-z]. The
number of entries used in a name space are machine dependent.

To allow a single tty line to be connected to a modem and used for both incoming and
outgoing calls, a special feature is available that is controlled by the minor device number. By
accessing character-special devices with names of the form /dev/cua/[n], it is possible to
open a port without the Carrier Detect signal being asserted, either through hardware or an
equivalent software mechanism. These devices are commonly known as dial-out lines.

Once a /dev/cua/[n] line is opened, the corresponding tty line cannot be opened until the
/dev/cua/n line is closed. A blocking open will wait until the /dev/cua/[n] line is closed
(which will drop Data Terminal Ready, and Carrier Detect) and carrier is detected again. A
non-blocking open will return an error. If the tty line has been opened successfully (usually
only when carrier is recognized on the modem) , the corresponding /dev/cua/[n] line cannot
be opened. This allows a modem to be attached to /dev/term/[n] (renamed from
/dev/tty[n]) and used for dial-in (by enabling the line for login in /etc/inittab) and also
used for dial-out (by tip(1) or uucp(1C)) as /dev/cua/[n] when no one is logged in on the
line.

Device and Network Interfaces 1053

zs(7D)

Note – This module is affected by the setting of specific eeprom variables. For information on
parameters that are persistent across reboots, see the eeprom(1M) man page.

ioctls The zs module supports the standard set of termio ioctl( ) calls.

If the CRTSCTS flag in the c_cflag field is set, output will be generated only if CTS is high; if CTS
is low, output will be frozen. If the CRTSCTS flag is clear, the state of CTS has no effect.

If the CRTSXOFF flag in the c_cflag field is set, input will be received only if RTS is high; if RTS
is low, input will be frozen. If the CRTSXOFF flag is clear, the state of RTS has no effect.

The termios CRTSCTS (respectively CRTSXOFF) flag and termiox CTSXON (respectively
RTSXOFF) can be used interchangeably.

Breaks can be generated by the TCSBRK, TIOCSBRK, and TIOCCBRK ioctl( ) calls.

The state of the DCD, CTS, RTS, and DTR interface signals may be queried through the use of
the TIOCM_CAR, TIOCM_CTS, TIOCM_RTS, and TIOCM_DTR arguments to the TIOCMGET ioctl
command, respectively. Due to hardware limitations, only the RTS and DTR signals may be set
through their respective arguments to the TIOCMSET, TIOCMBIS, and TIOCMBIC ioctl
commands.

The input and output line speeds may be set to any of the speeds supported by termio. The
input and output line speeds cannot be set independently; for example, when you set the the
output speed, the input speed is automatically set to the same speed.

When the driver is used to service the serial console port, it supports a BREAK condition that
allows the system to enter the debugger or the monitor. The BREAK condition is generated by
hardware and it is usually enabled by default. A BREAK condition originating from erroneous
electrical signals cannot be distinguished from one deliberately sent by remote DCE. The
Alternate Break sequence can be used to remedy this.

Due to a risk of incorrect sequence interpretation, SLIP and certain other binary protocols
should not be run over the serial console port when Alternate Break sequence is in effect.
Although PPP is a binary protocol, it is able to avoid these sequences using the ACCM feature
in RFC 1662. For Solaris PPP 4.0, you do this by adding the following line to the
/etc/ppp/options file (or other configuration files used for the connection; see pppd(1M) for
details):
asyncmap 0x00002000

By default, the Alternate Break sequence is three characters: carriage return, tilde and
control-B (CR ~ CTRL-B), but may be changed by the driver. For more information on
breaking (entering the debugger or monitor), see kbd(1) and kb(7M).

1054 man pages section 7: Device and Network Interfaces • Last Revised 16 Jul 2009
 zs(7D)

Errors An open will fail under the following conditions:

ENXIO The unit being opened does not exist.
EBUSY The dial-out device is being opened and the dial-in device is already open, or the
dial-in device is being opened with a no-delay open and the dial-out device is
already open.
EBUSY The port is in use by another serial protocol.
EBUSY The unit has been marked as exclusive-use by another process with a TIOCEXCL
ioctl( ) call.
EINTR The open was interrupted by the delivery of a signal.
Files /dev/cua/[a-z] dial-out tty lines
/dev/term/[a-z] dial-in tty lines
/dev/tty[a-z] binary compatibility package device names

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture SPARC

See Also eeprom(1M), kmdb(1), tip(1), cu(1C), uucp(1C), ports(1M), pppd(1M), ioctl(2), open(2),
attributes(5), zsh(7D), termio(7I), kb(7M), ldterm(7M), ttcompat(7M)
Diagnostics zsn : silo overflow. The Zilog 8530 character input silo overflowed before it
could be serviced.
zsn : ring buffer overflow. The driver's character input ring buffer overflowed
before it could be serviced.

Device and Network Interfaces 1055

zsh(7D)

Name zsh – On-board serial HDLC/SDLC interface

Synopsis #include <fcntl.h>
open(/dev/zshn, mode );
open(/dev/zsh, mode );

Description The zsh module is a loadable STREAMS driver that implements the sending and receiving of
data packets as HDLC frames over synchronous serial lines. The module is not a standalone
driver, but instead depends upon the zs module for the hardware support required by all
on-board serial devices. When loaded this module acts as an extension to the zs driver,
providing access to an HDLC interface through character-special devices.

The zshn devices provide what is known as a data path which supports the transfer of data
via read(2) and write(2) system calls, as well as ioctl(2) calls. Data path opens are exclusive
in order to protect against injection or diversion of data by another process.

The zsh device provides a separate control path for use by programs that need to configure
or monitor a connection independent of any exclusive access restrictions imposed by data
path opens. Up to three control paths may be active on a particular serial channel at any one
time. Control path accesses are restricted to ioctl(2) calls only; no data transfer is possible.

When used in synchronous modes, the Z8530 SCC supports several options for clock
sourcing and data encoding. Both the transmit and receive clock sources can be set to be the
external Transmit Clock (TRxC), external Receive Clock (RTxC), the internal Baud Rate Generator
(BRG), or the output of the SCC's Digital Phase-Lock Loop (DPLL).

The Baud Rate Generator is a programmable divisor that derives a clock frequency from the
PCLK input signal to the SCC. A programmed baud rate is translated into a 16-bit time
constant that is stored in the SCC. When using the BRG as a clock source the driver may answer
a query of its current speed with a value different from the one specified. This is because baud
rates translate into time constants in discrete steps, and reverse translation shows the change.
If an exact baud rate is required that cannot be obtained with the BRG, an external clock source
must be selected.

Use of the DPLL option requires the selection of NRZI data encoding and the setting of a
non-zero value for the baud rate, because the DPLL uses the BRG as its reference clock source.

A local loopback mode is available, primarily for use by the syncloop(1M) utility for testing
purposes, and should not be confused with SDLC loop mode, which is not supported on this
interface. Also, an auto-echo feature may be selected that causes all incoming data to be
routed to the transmit data line, allowing the port to act as the remote end of a digital loop.
Neither of these options should be selected casually, or left in use when not needed.

The zsh driver keeps running totals of various hardware generated events for each channel.
These include numbers of packets and characters sent and received, abort conditions detected

1056 man pages section 7: Device and Network Interfaces • Last Revised 1 Jan 1997
 zsh(7D)

by the receiver, receive CRC errors, transmit underruns, receive overruns, input errors and
output errors, and message block allocation failures. Input errors are logged whenever an
incoming message must be discarded, such as when an abort or CRC error is detected, a receive
overrun occurs, or when no message block is available to store incoming data. Output errors
are logged when the data must be discarded due to underruns, CTS drops during transmission,
CTS timeouts, or excessive watchdog timeouts caused by a cable break.

ioctls The zsh driver supports several ioctl() commands, including:

S_IOCGETMODE Return a struct scc_mode containing parameters currently in use. These
include the transmit and receive clock sources, boolean loopback and
NRZI mode flags and the integer baud rate.
S_IOCSETMODE The argument is a struct scc_mode from which the SCC channel will be
programmed.
S_IOCGETSTATS Return a struct sl_stats containing the current totals of
hardware-generated events. These include numbers of packets and
characters sent and received by the driver, aborts and CRC errors detected,
transmit underruns, and receive overruns.
S_IOCCLRSTATS Clear the hardware statistics for this channel.
S_IOCGETSPEED Returns the currently set baud rate as an integer. This may not reflect the
actual data transfer rate if external clocks are used.
S_IOCGETMCTL Returns the current state of the CTS and DCD incoming modem interface
signals as an integer.

The following structures are used with zsh ioctl() commands:

struct scc_mode {
char sm_txclock; /* transmit clock sources */
char sm_rxclock; /* receive clock sources */
char sm_iflags; /* data and clock inversion flags (non-zsh) */
uchar_t sm_config; /* boolean configuration options */
int sm_baudrate; /* real baud rate */
int sm_retval; /* reason codes for ioctl failures */
};
struct sl_stats {
long ipack; /* input packets */
long opack; /* output packets */
long ichar; /* input bytes */
long ochar; /* output bytes */
long abort; /* abort received */
long crc; /* CRC error */
long cts; /* CTS timeouts */
long dcd; /* Carrier drops */
long overrun; /* receive overrun */

Device and Network Interfaces 1057

zsh(7D)

long underrun; /* transmit underrun */

long ierror; /* input error */
long oerror; /* output error */
long nobuffers; /* receive side memory allocation failure */
};

Errors An open() will fail if a STREAMS message block cannot be allocated, or:
ENXIO The unit being opened does not exist.
EBUSY The device is in use by another serial protocol.

An ioctl() will fail if:

EINVAL An attempt was made to select an invalid clocking source.
EINVAL The baud rate specified for use with the baud rate generator would translate to a
null time constant in the SCC's registers.
Files /dev/zsh[0-1],/dev/zsh character-special devices
/usr/include/sys/ser_sync.h header file specifying synchronous serial
communication definitions

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

See Also syncinit(1M), syncloop(1M), syncstat(1M), ioctl(2), open(2), read(2), write(2),

attributes(5), zs(7D)

Refer to the Zilog Z8530 SCC Serial Communications Controller Technical Manual for details
of the SCC's operation and capabilities.
Diagnostics zsh data open failed, no memory, rq=nnn
zsh clone open failed, no memory, rq=nnn
A kernel memory allocation failed for one of the private data structures. The value of nnn is
the address of the read queue passed to open(2).
zsh_open: can’t alloc message block
The open could not proceed because an initial STREAMS message block could not be made
available for incoming data.
zsh: clone device d must be attached before use!
An operation was attempted through a control path before that path had been attached to a
particular serial channel.

1058 man pages section 7: Device and Network Interfaces • Last Revised 1 Jan 1997
 zsh(7D)

zshn: invalid operation for clone dev.

An inappropriate STREAMS message type was passed through a control path. Only
M_IOCTL and M_PROTO message types are permitted.
zshn: not initialized, can’t send message
An M_DATA message was passed to the driver for a channel that had not been programmed
at least once since the driver was loaded. The SCC's registers were in an unknown state. The
S_IOCSETMODE ioctl command performs the programming operation.
zshn: transmit hung
The transmitter was not successfully restarted after the watchdog timer expired.

Device and Network Interfaces 1059

zulu(7d)

Name zulu – Sun XVR-4000 Graphics Accelerator driver

Description The zulu driver is the device driver for the Sun XVR-4000 Graphics Accelerator.
Files /dev/fbs/zulun device special file
/usr/lib/zulu.ucode zulu microcode

Attributes See attributes(5) for descriptions of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Availability SUNWzulux

See Also SUNWzulu_config(1M), zuludaemon(1M), attributes(5)

1060 man pages section 7: Device and Network Interfaces • Last Revised 8 Oct 2002
 zyd(7D)

Name zyd – ZyDAS ZD1211/ZD1211B USB 802.11b/g Wireless Driver

Description The zyd 802.11b/g wireless driver is a multi-threaded, loadable, clonable, GLDv3-based
STREAMS driver supporting the ZyDAS ZD1211/ZD1211B USB chipset-based wireless
devices.

Configuration The zyd driver performs auto-negotiation to determine the data rate and mode. Supported
802.11b data rates are 1, 2, 5.5 and 11 Mbits/sec. Supported 802.11g data rates are 1, 2, 5.5, 11,
6, 9, 12, 18, 24, 36, 48 and 54 Mbits/sec. The zyd driver supports only BSS networks (also
known as "ap" or "infrastructure" networks) and open (or "open-system"), shared key and
WPA/WPA2 authentication. Supported encryption types are WEP40, WEP104, TKIP and
AES-CCMP.
Files /dev/zyd* Special character device.
/kernel/drv/zyd 32–bit kernel module (x86).
/kernel/drv/amd64/zyd 64–bit kernel module (x86).

Attributes See attributes(5) for a description of the following attributes:

ATTRIBUTE TYPE ATTRIBUTE VALUE

Architecture x86

Availability SUNWzyd

Interface Stability Committed

See Also dladm(1M), wificonfig(1M), attributes(5), gld(7D), dlpi(7P)

802.11 - Wireless LAN Media Access Control and Physical Layer Specification - IEEE, 2001

Device and Network Interfaces 1061

1062