CUDA COMPILER DRIVER NVCC

TRM-06721-001_v10.2 | October 2019

Reference Guide
 CHANGES FROM PREVIOUS VERSION

‣ Major update to the document to reflect recent nvcc changes.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | ii
 TABLE OF CONTENTS

Chapter  1.  Introduction.........................................................................................1

1.1.  Overview................................................................................................... 1
1.1.1.  CUDA Programming Model......................................................................... 1
1.1.2.  CUDA Sources........................................................................................ 1
1.1.3.  Purpose of NVCC.................................................................................... 2
1.2.  Supported Host Compilers...............................................................................2
Chapter  2.  Compilation Phases................................................................................3
2.1.  NVCC Identification Macro.............................................................................. 3
2.2.  NVCC Phases............................................................................................... 3
2.3.  Supported Input File Suffixes...........................................................................4
2.4.  Supported Phases......................................................................................... 4
Chapter 3. The CUDA Compilation Trajectory............................................................. 7
Chapter 4. NVCC Command Options......................................................................... 9
4.1. Command Option Types and Notation.................................................................9
4.2.  Command Option Description......................................................................... 10
4.2.1. File and Path Specifications......................................................................10
4.2.1.1.  --output-file file (-o).........................................................................10
4.2.1.2. --pre-include file,... (-include).............................................................10
4.2.1.3.  --library library,... (-l)....................................................................... 10
4.2.1.4. --define-macro def,... (-D)..................................................................10
4.2.1.5. --undefine-macro def,... (-U).............................................................. 10
4.2.1.6. --include-path path,... (-I).................................................................. 10
4.2.1.7. --system-include path,... (-isystem).......................................................11
4.2.1.8. --library-path path,... (-L).................................................................. 11
4.2.1.9. --output-directory directory (-odir)....................................................... 11
4.2.1.10. --dependency-output file (-MF)........................................................... 11
4.2.1.11. --compiler-bindir directory (-ccbin)...................................................... 11
4.2.1.12. --cudart {none|shared|static} (-cudart).................................................11
4.2.1.13. --cudadevrt {none|static} (-cudadevrt)..................................................12
4.2.1.14. --libdevice-directory directory (-ldir).................................................... 12
4.2.1.15. --target-directory string (-target-dir).................................................... 12
4.2.2. Options for Specifying the Compilation Phase................................................ 12
4.2.2.1.  --link (-link)....................................................................................12
4.2.2.2.  --lib (-lib)...................................................................................... 12
4.2.2.3.  --device-link (-dlink)......................................................................... 13
4.2.2.4.  --device-c (-dc)............................................................................... 13
4.2.2.5.  --device-w (-dw).............................................................................. 13
4.2.2.6.  --cuda (-cuda)................................................................................. 13
4.2.2.7.  --compile (-c)................................................................................. 14
4.2.2.8.  --fatbin (-fatbin).............................................................................. 14

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | iii
 4.2.2.9.  --cubin (-cubin)............................................................................... 14
4.2.2.10.  --ptx (-ptx)................................................................................... 14
4.2.2.11.  --preprocess (-E)............................................................................ 14
4.2.2.12. --generate-dependencies (-M).............................................................15
4.2.2.13. --generate-nonsystem-dependencies (-MM)............................................. 15
4.2.2.14. --generate-dependencies-with-compile (-MD).......................................... 15
4.2.2.15. --generate-nonsystem-dependencies-with-compile (-MMD)...........................15
4.2.2.16.  --run (-run)................................................................................... 15
4.2.3. Options for Specifying Behavior of Compiler/Linker......................................... 16
4.2.3.1.  --profile (-pg)................................................................................. 16
4.2.3.2.  --debug (-g)....................................................................................16
4.2.3.3.  --device-debug (-G).......................................................................... 16
4.2.3.4. --extensible-whole-program (-ewp)........................................................16
4.2.3.5. --generate-line-info (-lineinfo)............................................................. 16
4.2.3.6.  --optimize level (-O)......................................................................... 16
4.2.3.7. --ftemplate-backtrace-limit limit (-ftemplate-backtrace-limit).......................16
4.2.3.8. --ftemplate-depth limit (-ftemplate-depth)..............................................16
4.2.3.9.  --shared (-shared)............................................................................ 16
4.2.3.10.  --x {c|c++|cu} (-x).......................................................................... 17
4.2.3.11. --std {c++03|c++11|c++14} (-std).........................................................17
4.2.3.12. --no-host-device-initializer-list (-nohdinitlist).......................................... 17
4.2.3.13. --expt-relaxed-constexpr (-expt-relaxed-constexpr).................................. 17
4.2.3.14. --extended-lambda (-extended-lambda).................................................18
4.2.3.15. --expt-extended-lambda (-expt-extended-lambda)....................................18
4.2.3.16. --machine {32|64} (-m).................................................................... 18
4.2.4. Options for Passing Specific Phase Options.................................................... 18
4.2.4.1. --compiler-options options,... (-Xcompiler).............................................. 18
4.2.4.2. --linker-options options,... (-Xlinker)..................................................... 18
4.2.4.3. --archive-options options,... (-Xarchive)..................................................18
4.2.4.4. --ptxas-options options,... (-Xptxas)...................................................... 18
4.2.4.5. --nvlink-options options,... (-Xnvlink)..................................................... 18
4.2.5. Options for Guiding the Compiler Driver...................................................... 19
4.2.5.1. --dont-use-profile (-noprof)................................................................. 19
4.2.5.2.  --dryrun (-dryrun).............................................................................19
4.2.5.3.  --verbose (-v)..................................................................................19
4.2.5.4.  --keep (-keep).................................................................................19
4.2.5.5. --keep-dir directory (-keep-dir)............................................................ 19
4.2.5.6. --save-temps (-save-temps)................................................................. 19
4.2.5.7.  --clean-targets (-clean)...................................................................... 19
4.2.5.8. --run-args arguments,... (-run-args)....................................................... 19
4.2.5.9. --input-drive-prefix prefix (-idp)........................................................... 19
4.2.5.10. --dependency-drive-prefix prefix (-ddp)................................................ 20
4.2.5.11. --drive-prefix prefix (-dp)................................................................. 20

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | iv
 4.2.5.12. --dependency-target-name target (-MT)................................................ 20
4.2.5.14. --no-device-link (-nodlink)................................................................. 20
4.2.6. Options for Steering CUDA Compilation........................................................ 20
4.2.6.1. --default-stream {legacy|null|per-thread} (-default-stream)......................... 20
4.2.7. Options for Steering GPU Code Generation................................................... 21
4.2.7.1. --gpu-architecture arch (-arch)............................................................ 21
4.2.7.2. --gpu-code code,... (-code).................................................................21
4.2.7.3. --generate-code specification (-gencode)................................................ 22
4.2.7.4. --relocatable-device-code {true|false} (-rdc)............................................ 22
4.2.7.5.  --entries entry,... (-e)....................................................................... 23
4.2.7.6. --maxrregcount amount (-maxrregcount).................................................23
4.2.7.7. --use_fast_math (-use_fast_math)......................................................... 23
4.2.7.8.  --ftz {true|false} (-ftz)...................................................................... 23
4.2.7.9. --prec-div {true|false} (-prec-div)......................................................... 24
4.2.7.10. --prec-sqrt {true|false} (-prec-sqrt)......................................................24
4.2.7.11. --fmad {true|false} (-fmad)............................................................... 24
4.2.7.12. --compile-as-tools-patch (-astoolspatch)................................................ 25
4.2.7.13. --keep-device-functions (-keep-device-functions)..................................... 25
4.2.8.  Generic Tool Options.............................................................................. 25
4.2.8.1.  --disable-warnings (-w)...................................................................... 25
4.2.8.2. --source-in-ptx (-src-in-ptx).................................................................25
4.2.8.3.  --restrict (-restrict)...........................................................................25
4.2.8.4. --Wno-deprecated-gpu-targets (-Wno-deprecated-gpu-targets).......................25
4.2.8.5. --Wno-deprecated-declarations (-Wno-deprecated-declarations).....................26
4.2.8.6.  --Wreorder (-Wreorder)...................................................................... 26
4.2.8.7. --Werror kind,... (-Werror)..................................................................26
4.2.8.8. --resource-usage (-res-usage)...............................................................26
4.2.8.9.  --help (-h)......................................................................................26
4.2.8.10.  --version (-V).................................................................................26
4.2.8.11. --options-file file,... (-optf)............................................................... 26
4.2.8.12. --time filename (-time).................................................................... 27
4.2.8.13. --qpp-config config (-qpp-config).........................................................27
4.2.9.  Phase Options.......................................................................................27
4.2.9.1.  Ptxas Options..................................................................................27
4.2.9.2.  NVLINK Options............................................................................... 30
Chapter  5.  GPU Compilation................................................................................. 31
5.1.  GPU Generations........................................................................................ 31
5.2.  GPU Feature List........................................................................................ 31
5.3.  Application Compatibility.............................................................................. 32
5.4.  Virtual Architectures....................................................................................32
5.5. Virtual Architecture Feature List..................................................................... 33
5.6.  Further Mechanisms.....................................................................................34
5.6.1.  Just-in-Time Compilation......................................................................... 34

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | v
 5.6.2.  Fatbinaries.......................................................................................... 35
5.7.  NVCC Examples.......................................................................................... 35
5.7.1.  Base Notation.......................................................................................35
5.7.2.  Shorthand............................................................................................35
5.7.2.1.  Shorthand 1....................................................................................35
5.7.2.2.  Shorthand 2....................................................................................36
5.7.2.3.  Shorthand 3....................................................................................36
5.7.3.  Extended Notation................................................................................. 36
5.7.4. Virtual Architecture Identification Macro...................................................... 37
Chapter 6. Using Separate Compilation in CUDA........................................................ 38
6.1. Code Changes for Separate Compilation............................................................ 38
6.2. NVCC Options for Separate Compilation............................................................ 38
6.3.  Libraries...................................................................................................39
6.4.  Examples.................................................................................................. 40
6.5. Potential Separate Compilation Issues...............................................................41
6.5.1.  Object Compatibility.............................................................................. 41
6.5.2.  JIT Linking Support................................................................................ 42
6.5.3.  Implicit CUDA Host Code......................................................................... 42
6.5.4.  Using __CUDA_ARCH__............................................................................ 42
6.5.5.  Device Code in Libraries..........................................................................43
Chapter 7. Miscellaneous NVCC Usage..................................................................... 44
7.1.  Cross Compilation....................................................................................... 44
7.2. Keeping Intermediate Phase Files.................................................................... 44
7.3.  Cleaning Up Generated Files.......................................................................... 44
7.4. Printing Code Generation Statistics.................................................................. 45

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | vi
 LIST OF FIGURES

Figure 1 CUDA Compilation Trajectory ....................................................................... 8

Figure 2 Two-Staged Compilation with Virtual and Real Architectures ................................ 33

Figure 3 Just-in-Time Compilation of Device Code ....................................................... 34

Figure 4 CUDA Separate Compilation Trajectory .......................................................... 39

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | vii
www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | viii
Chapter 1.
INTRODUCTION

1.1. Overview
1.1.1. CUDA Programming Model
The CUDA Toolkit targets a class of applications whose control part runs as a process
on a general purpose computing device, and which use one or more NVIDIA GPUs as
coprocessors for accelerating single program, multiple data (SPMD) parallel jobs. Such jobs
are self-contained, in the sense that they can be executed and completed by a batch of
GPU threads entirely without intervention by the host process, thereby gaining optimal
benefit from the parallel graphics hardware.
The GPU code is implemented as a collection of functions in a language that is
essentially C++, but with some annotations for distinguishing them from the host code,
plus annotations for distinguishing different types of data memory that exists on the
GPU. Such functions may have parameters, and they can be called using a syntax that is
very similar to regular C function calling, but slightly extended for being able to specify
the matrix of GPU threads that must execute the called function. During its life time, the
host process may dispatch many parallel GPU tasks.
For more information on the CUDA programming model, consult the CUDA C
Programming Guide.

1.1.2. CUDA Sources
Source files for CUDA applications consist of a mixture of conventional C++ host code,
plus GPU device functions. The CUDA compilation trajectory separates the device
functions from the host code, compiles the device functions using the proprietary
NVIDIA compilers and assembler, compiles the host code using a C++ host compiler
that is available, and afterwards embeds the compiled GPU functions as fatbinary
images in the host object file. In the linking stage, specific CUDA runtime libraries are
added for supporting remote SPMD procedure calling and for providing explicit GPU
manipulation such as allocation of GPU memory buffers and host-GPU data transfer.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 1
 Introduction

1.1.3. Purpose of NVCC
The compilation trajectory involves several splitting, compilation, preprocessing, and
merging steps for each CUDA source file. It is the purpose of nvcc, the CUDA compiler
driver, to hide the intricate details of CUDA compilation from developers. It accepts a
range of conventional compiler options, such as for defining macros and include/library
paths, and for steering the compilation process. All non-CUDA compilation steps are
forwarded to a C++ host compiler that is supported by nvcc, and nvcc translates its
options to appropriate host compiler command line options.

1.2. Supported Host Compilers

A general purpose C++ host compiler is needed by nvcc in the following situations:
‣ During non-CUDA phases (except the run phase), because these phases will be
forwarded by nvcc to this compiler.
‣ During CUDA phases, for several preprocessing stages and host code compilation
(see also The CUDA Compilation Trajectory).
nvcc assumes that the host compiler is installed with the standard method designed by
the compiler provider. If the host compiler installation is non-standard, the user must
make sure that the environment is set appropriately and use relevant nvcc compile
options.
The following documents provide detailed information about supported host compilers:
‣ NVIDIA CUDA Installation Guide for Linux
‣ NVIDIA CUDA Installation Guide for Mac OS X
‣ NVIDIA CUDA Installation Guide for Microsoft Windows
On all platforms, the default host compiler executable (gcc and g++ on Linux, clang
and clang++ on Mac OS X, and cl.exe on Windows) found in the current execution
search path will be used, unless specified otherwise with appropriate options (see File
and Path Specifications).

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 2
Chapter 2.
COMPILATION PHASES

2.1. NVCC Identification Macro

nvcc predefines the following macros:
__NVCC__
Defined when compiling C/C++/CUDA source files.
__CUDACC__
Defined when compiling CUDA source files.
__CUDACC_RDC__
Defined when compiling CUDA sources files in relocatable device code mode (see
NVCC Options for Separate Compilation).
__CUDACC_DEBUG__
Defined when compiler CUDA source files in the device-debug mode (see Options
for Specifying Behavior of Compiler/Linker).
__CUDACC_RELAXED_CONSTEXPR__
Defined when the --expt-relaxed-constexpr flag is specified on the command
line. Refer to CUDA C Programming Guide for more details.
__CUDACC_EXTENDED_LAMBDA__
Defined when the --expt-extended-lambda or --extended-lambda flag is
specified on the command line. Refer to CUDA C Programming Guide for more
details.
__CUDACC_VER_MAJOR__
Defined with the major version number of nvcc.
__CUDACC_VER_MINOR__
Defined with the minor version number of nvcc.
__CUDACC_VER_BUILD__
Defined with the build version number of nvcc.

2.2. NVCC Phases
A compilation phase is the a logical translation step that can be selected by command
line options to nvcc. A single compilation phase can still be broken up by nvcc into

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 3
 Compilation Phases

smaller steps, but these smaller steps are just implementations of the phase: they depend
on seemingly arbitrary capabilities of the internal tools that nvcc uses, and all of these
internals may change with a new release of the CUDA Toolkit. Hence, only compilation
phases are stable across releases, and although nvcc provides options to display the
compilation steps that it executes, these are for debugging purposes only and must not
be copied and used into build scripts.
nvcc phases are selected by a combination of command line options and input file
name suffixes, and the execution of these phases may be modified by other command
line options. In phase selection, the input file suffix defines the phase input, while the
command line option defines the required output of the phase.
The following paragraphs will list the recognized file name suffixes and the supported
compilation phases. A full explanation of the nvcc command line options can be found
in NVCC Command Options.

2.3. Supported Input File Suffixes

The following table defines how nvcc interprets its input files:
Input File Prefix Description
.cu CUDA source file, containing host code and device functions
.c C source file
.cc, .cxx, .cpp C++ source file
.ptx PTX intermediate assembly file (see Figure 1)
.cubin CUDA device code binary file (CUBIN) for a single GPU architecture
(see Figure 1)
.fatbin CUDA fat binary file that may contain multiple PTX and CUBIN files
(see Figure 1)
.o, .obj Object file
.a, .lib Library file
.res Resource file
.so Shared object file

Note that nvcc does not make any distinction between object, library or resource files. It
just passes files of these types to the linker when the linking phase is executed.

2.4. Supported Phases
The following table specifies the supported compilation phases, plus the option to
nvcc that enables execution of this phase. It also lists the default name of the output
file generated by this phase, which will take effect when no explicit output file name is
specified using option --output-file:

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 4
 Compilation Phases

nvcc Option
Phase Default Output File Name
Long Name Short Name
CUDA --cuda -cuda .cpp.ii appended to source file name, as in
compilation to x.cu.cpp.ii. This output file can be compiled
C/C++ source by the host compiler that was used by nvcc to
file preprocess the .cu file.
C/C++ --preprocess -E <result on standard output>
preprocessing
C/C++ --compile -c Source file name with suffix replaced by o on Linux
compilation to and Mac OS X, or obj on Windows
object file
Cubin --cubin -cubin Source file name with suffix replaced by cubin
generation
from CUDA
source files
Cubin --cubin -cubin Source file name with suffix replaced by cubin
generation
from PTX
intermediate
files.
PTX generation --ptx -ptx Source file name with suffix replaced by ptx
from CUDA
source files
Fatbinary --fatbin -fatbin Source file name with suffix replaced by fatbin
generation
from source,
PTX or cubin
files
Linking --device- -dlink a_dlink.obj on Windows or a_dlink.o on other
relocatable link platforms
device code.
Cubin --device- -dlink - a_dlink.cubin
generation link -- cubin
from linked cubin
relocatable
device code.
Fatbinary --device- -dlink - a_dlink.fatbin
generation link -- fatbin
from linked fatbin
relocatable
device code
Linking an <no phase option> a.exe on Windows or a.out on other platforms
executable
Constructing --lib -lib a.lib on Windows or a.a on other platforms
an object file
archive, or
library
make --generate- -M <result on standard output>
dependency dependencies
generation

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 5
 Compilation Phases

nvcc Option
Phase Default Output File Name
Long Name Short Name
make --generate- -MM <result on standard output>
dependency nonsystem-
generation dependencies
without
headers in
system paths.
Running an --run -run
executable

Notes:
‣ The last phase in this list is more of a convenience phase. It allows running the
compiled and linked executable without having to explicitly set the library path to
the CUDA dynamic libraries.
‣ Unless a phase option is specified, nvcc will compile and link all its input files.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 6
Chapter 3.
THE CUDA COMPILATION TRAJECTORY

CUDA compilation works as follows: the input program is preprocessed for device
compilation compilation and is compiled to CUDA binary (cubin) and/or PTX
intermediate code, which are placed in a fatbinary. The input program is preprocessed
once again for host compilation and is synthesized to embed the fatbinary and transform
CUDA specific C++ extensions into standard C++ constructs. Then the C++ host compiler
compiles the synthesized host code with the embedded fatbinary into a host object. The
exact steps that are followed to achieve this are displayed in Figure 1.
The embedded fatbinary is inspected by the CUDA runtime system whenever the device
code is launched by the host program to obtain an appropriate fatbinary image for the
current GPU.
CUDA programs are compiled in the whole program compilation mode by default,
i.e., the device code cannot reference an entity from a separate file. In the whole
program compilation mode, device link steps have no effect. For more information
on the separate compilation and the whole program compilation, see Using Separate
Compilation in CUDA.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 7
 The CUDA Compilation Trajectory

.cu

A B
C+ + Preprocessor C+ + Preprocessor
A is passed t o B as an input file.

.cpp4.ii .cpp1.ii A B
A is # include'd in B.

cudafe+ + .cudafe1.st ub.c cicc

# Repeat for each .cu input file.

# Repeat for each virt ual archit ect ure.

.cudafe1.cpp .pt x
# Repeat pt xas and nvlink for each virt ual/real

pt xas archit ect ure

com binat ion.

# Device linker consist s of st eps in .

.cubin

C+ + Com piler fat binary

.o / .obj .fat bin.c

nvlink a_dlink.reg.c

a_dlink.cubin

fat binary

a_dlink.fat bin.c link.st ub

C+ + Com piler

a_dlink.o / a_dlink.obj

Host Linker

execut able

Figure 1 CUDA Compilation Trajectory

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 8
Chapter 4.
NVCC COMMAND OPTIONS

4.1. Command Option Types and Notation

Each nvcc option has a long name and a short name, which are interchangeable with
each other. These two variants are distinguished by the number of hyphens that must
precede the option name: long names must be preceded by two hyphens, while short
names must be preceded by a single hyphen. For example, -I is the short name of --
include-path. Long options are intended for use in build scripts, where size of the
option is less important than descriptive value. In contrast, short options are intended
for interactive use.
nvcc recognizes three types of command options: boolean options, single value options,
and list options.
Boolean options do not have an argument; they are either specified on a command line
or not. Single value options must be specified at most once, and list options may be
repeated. Examples of each of these option types are, respectively: --verbose (switch
to verbose mode), --output-file (specify output file), and --include-path (specify
include path).
Single value options and list options must have arguments, which must follow the name
of the option itself by either one of more spaces or an equals character. When a one-
character short name such as -I, -l, and -L is used, the value of the option may also
immediately follow the option itself without being seperated by spaces or an equal
character. The individual values of list options may be separated by commas in a single
instance of the option, or the option may be repeated, or any combination of these two
cases.
Hence, for the two sample options mentioned above that may take values, the following
notations are legal:
-o file
-o=file
-Idir1,dir2 -I=dir3 -I dir4,dir5

Long option names are used throughout the document, unless specified otherwise,
however, short names can be used instead of long names to have the same effect.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 9
 NVCC Command Options

4.2. Command Option Description

This section presents tables of nvcc options. The option type in the tables can be
recognized as follows: boolean options do not have arguments specified in the first
column, while the other two types do. List options can be recognized by the repeat
indicator ,... at the end of the argument.
Long options are described in the first columns of the options tables, and short options
occupy the second columns.

4.2.1. File and Path Specifications

4.2.1.1. --output-file file (-o)
Specify name and location of the output file.

4.2.1.2. --pre-include file,... (-include)

Specify header files that must be pre-included during preprocessing.

4.2.1.3. --library library,... (-l)

Specify libraries to be used in the linking stage without the library file extension.
The libraries are searched for on the library search paths that have been specified using
option --library-path (see Libraries).

4.2.1.4. --define-macro def,... (-D)

Define macros to be used during preprocessing.
def can be either name or name=definition.

‣ name

‣ Predefine name as a macro.

‣ name=definition

‣ The contents of definition are tokenized and preprocessed as if they appear

during translation phase three in a #define directive. The definition will be
truncated by embedded new line characters.

4.2.1.5. --undefine-macro def,... (-U)

Undefine an existing macro during preprocessing or compilation.

4.2.1.6. --include-path path,... (-I)

Specify include search paths.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 10
 NVCC Command Options

4.2.1.7. --system-include path,... (-isystem)

Specify system include search paths.

4.2.1.8. --library-path path,... (-L)

Specify library search paths (see Libraries).

4.2.1.9. --output-directory directory (-odir)

Specify the directory of the output file.
This option is intended for letting the dependency generation step (see --generate-
dependencies) generate a rule that defines the target object file in the proper directory.

4.2.1.10. --dependency-output file (-MF)

Specify the dependency output file.
This option specifies the output file for the dependency generation step (see --
generate-dependencies). The option --generate-dependencies or --generate-
nonystem-dependencies must be specified if a dependency output file is set.

4.2.1.11. --compiler-bindir directory (-ccbin)

Specify the directory in which the compiler executable resides.
The host compiler executable name can be also specified to ensure that the correct host
compiler is selected. In addition, driver prefix options (--input-drive-prefix, --
dependency-drive-prefix, or --drive-prefix) may need to be specified, if nvcc is
executed in a Cygwin shell or a MinGW shell on Windows.

4.2.1.12. --cudart {none|shared|static} (-cudart)

Specify the type of CUDA runtime library to be used: no CUDA runtime library, shared/dynamic
CUDA runtime library, or static CUDA runtime library.

Allowed Values
‣ none
‣ shared
‣ static

Default
The static CUDA runtime library is used by default.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 11
 NVCC Command Options

4.2.1.13. --cudadevrt {none|static} (-cudadevrt)

Specify the type of CUDA device runtime library to be used: no CUDA device runtime library, or
static CUDA device runtime library.

Allowed Values
‣ none
‣ static

Default
The static CUDA device runtime library is used by default.

4.2.1.14. --libdevice-directory directory (-ldir)

Specify the directory that contains the libdevice library files.
Libdevice library files are located in the nvvm/libdevice directory in the CUDA
Toolkit.

4.2.1.15. --target-directory string (-target-dir)

Specify the subfolder name in the targets directory where the default include and library paths are
located.

4.2.2. Options for Specifying the Compilation Phase

Options of this category specify up to which stage the input files must be compiled.

4.2.2.1. --link (-link)
Specify the default behavior: compile and link all input files.

Default Output File Name

a.exe on Windows or a.out on other platforms is used as the default output file name.

4.2.2.2. --lib (-lib)
Compile all input files into object files, if necessary, and add the results to the specified library
output file.

Default Output File Name

a.lib on Windows or a.a on other platforms is used as the default output file name.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 12
 NVCC Command Options

4.2.2.3. --device-link (-dlink)
Link object files with relocatable device code and .ptx, .cubin, and .fatbin files into an
object file with executable device code, which can be passed to the host linker.

Default Output File Name

a_dlink.obj on Windows or a_dlink.o on other platforms is used as the
default output file name. When this option is used in conjunction with --fatbin,
a_dlink.fatbin is used as the default output file name. When this option is used in
conjunction with --cubin, a_dlink.cubin is used as the default output file name.

4.2.2.4. --device-c (-dc)
Compile each .c, .cc, .cpp, .cxx, and .cu input file into an object file that contains
relocatable device code.
It is equivalent to --relocatable-device-code=true --compile.

Default Output File Name

The source file name extension is replaced by .obj on Windows and .o on other
platforms to create the default output file name. For example, the default output file
name for x.cu is x.obj on Windows and x.o on other platforms.

4.2.2.5. --device-w (-dw)
Compile each .c, .cc, .cpp, .cxx, and .cu input file into an object file that contains
executable device code.
It is equivalent to --relocatable-device-code=false --compile.

Default Output File Name

The source file name extension is replaced by .obj on Windows and .o on other
platforms to create the default output file name. For example, the default output file
name for x.cu is x.obj on Windows and x.o on other platforms.

4.2.2.6. --cuda (-cuda)
Compile each .cu input file to a .cu.cpp.ii file.

Default Output File Name

.cu.cpp.ii is appended to the basename of the source file name to create the default
output file name. For example, the default output file name for x.cu is x.cu.cpp.ii.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 13
 NVCC Command Options

4.2.2.7. --compile (-c)
Compile each .c, .cc, .cpp, .cxx, and .cu input file into an object file.

Default Output File Name

The source file name extension is replaced by .obj on Windows and .o on other
platforms to create the default output file name. For example, the default output file
name for x.cu is x.obj on Windows and x.o on other platforms.

4.2.2.8. --fatbin (-fatbin)
Compile all .cu, .ptx, and .cubin input files to device-only .fatbin files.
nvcc discards the host code for each .cu input file with this option.

Default Output File Name

The source file name extension is replaced by .fatbin to create the default output file
name. For example, the default output file name for x.cu is x.fatbin.

4.2.2.9. --cubin (-cubin)
Compile all .cu and .ptx input files to device-only .cubin files.
nvcc discards the host code for each .cu input file with this option.

Default Output File Name

The source file name extension is replaced by .cubin to create the default output file
name. For example, the default output file name for x.cu is x.cubin.

4.2.2.10. --ptx (-ptx)
Compile all .cu input files to device-only .ptx files.
nvcc discards the host code for each .cu input file with this option.

Default Output File Name

The source file name extension is replaced by .ptx to create the default output file
name. For example, the default output file name for x.cu is x.ptx.

4.2.2.11. --preprocess (-E)
Preprocess all .c, .cc, .cpp, .cxx, and .cu input files.

Default Output File Name

The output is generated in stdout by default.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 14
 NVCC Command Options

4.2.2.12. --generate-dependencies (-M)
Generate a dependency file that can be included in a Makefile for the .c, .cc, .cpp, .cxx,
and .cu input file.

Default Output File Name

The output is generated in stdout by default.

4.2.2.13. --generate-nonsystem-dependencies (-MM)
Same as --generate-dependencies but skip headers files found in system directories
(Linux only).

Default Output File Name

The output is generated in stdout by default.

4.2.2.14. --generate-dependencies-with-compile (-MD)
Generate a dependency file and compile the input file. The dependency file can be included in a
Makefile for the .c, .cc, .cpp, .cxx, and .cu input file.

This option cannot be specified together with -E. The dependency file name is computed
as follows:

‣ If -MF is specified, then the specified file is used as the dependency file name.
‣ If -o is specified, the dependency file name is computed from the specified file name
by replacing the suffix with '.d'.
‣ Otherwise, the dependency file name is computed by replacing the input file
names's suffix with '.d'.
If the dependency file name is computed based on either -MF or -o, then multiple input
files are not supported.

4.2.2.15. --generate-nonsystem-dependencies-with-compile
(-MMD)
Same as --generate-dependencies-with-compile but skip header files found in system
directories (Linux only).

4.2.2.16. --run (-run)
Compile and link all input files into an executable, and executes it.
When the input is a single executable, it is executed without any compilation or linking.
This step is intended for developers who do not want to be bothered with setting the
necessary environment variables; these are set temporarily by nvcc.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 15
 NVCC Command Options

4.2.3. Options for Specifying Behavior of Compiler/

Linker
4.2.3.1. --profile (-pg)
Instrument generated code/executable for use by gprof.

4.2.3.2. --debug (-g)
Generate debug information for host code.

4.2.3.3. --device-debug (-G)
Generate debug information for device code.
This option turns off all optimizations on device code. It is not intended for profiling;
use --generate-line-info instead for profiling.

4.2.3.4. --extensible-whole-program (-ewp)
Generate extensible whole program device code, which allows some calls to not be resolved until
linking with libcudadevrt.

4.2.3.5. --generate-line-info (-lineinfo)
Generate line-number information for device code.

4.2.3.6. --optimize level (-O)

Specify optimization level for host code.

4.2.3.7. --ftemplate-backtrace-limit limit (-ftemplate-

backtrace-limit)
Set the maximum number of template instantiation notes for a single warning or error to limit.
A value of 0 is allowed, and indicates that no limit should be enforced. This value is also
passed to the host compiler if it provides an equivalent flag.

4.2.3.8. --ftemplate-depth limit (-ftemplate-depth)

Set the maximum instantiation depth for template classes to limit.
This value is also passed to the host compiler if it provides an equivalent flag.

4.2.3.9. --shared (-shared)
Generate a shared library during linking.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 16
 NVCC Command Options

Use option --linker-options when other linker options are required for more
control.

4.2.3.10. --x {c|c++|cu} (-x)

Explicitly specify the language for the input files, rather than letting the compiler choose a default
based on the file name suffix.

Allowed Values
‣ c
‣ c++
‣ cu

Default
The language of the source code is determined based on the file name suffix.

4.2.3.11. --std {c++03|c++11|c++14} (-std)

Select a particular C++ dialect.

Allowed Values
‣ c++03
‣ c++11
‣ c++14

Default
The default C++ dialect depends on the host compiler. nvcc matches the default C++
dialect that the host compiler uses.

4.2.3.12. --no-host-device-initializer-list (-
nohdinitlist)
Do not consider member functions of std::initializer_list as __host__ __device__
functions implicitly.

4.2.3.13. --expt-relaxed-constexpr (-expt-relaxed-
constexpr)
Experimental flag: Allow host code to invoke __device__ constexpr functions, and device
code to invoke __host__ constexpr functions.
Note that the behavior of this flag may change in future compiler releases.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 17
 NVCC Command Options

4.2.3.14. --extended-lambda (-extended-lambda)
Allow __host__, __device__ annotations in lambda declarations.

4.2.3.15. --expt-extended-lambda (-expt-extended-lambda)
Alias for --extended-lambda.

4.2.3.16. --machine {32|64} (-m)

Specify 32-bit vs. 64-bit architecture.

Allowed Values
‣ 32
‣ 64

Default
This option is set based on the host platform on which nvcc is executed.

4.2.4. Options for Passing Specific Phase Options

These allow for passing specific options directly to the internal compilation tools that
nvcc encapsulates, without burdening nvcc with too-detailed knowledge on these tools.
A table of useful sub-tool options can be found at the end of this chapter.

4.2.4.1. --compiler-options options,... (-Xcompiler)

Specify options directly to the compiler/preprocessor.

4.2.4.2. --linker-options options,... (-Xlinker)

Specify options directly to the host linker.

4.2.4.3. --archive-options options,... (-Xarchive)

Specify options directly to the library manager.

4.2.4.4. --ptxas-options options,... (-Xptxas)

Specify options directly to ptxas, the PTX optimizing assembler.

4.2.4.5. --nvlink-options options,... (-Xnvlink)

Specify options directly to nvlink, the device linker.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 18
 NVCC Command Options

4.2.5. Options for Guiding the Compiler Driver

4.2.5.1. --dont-use-profile (-noprof)
Do not use configurations from the nvcc.profile file for compilation.

4.2.5.2. --dryrun (-dryrun)
List the compilation sub-commands without executing them.

4.2.5.3. --verbose (-v)
List the compilation sub-commands while executing them.

4.2.5.4. --keep (-keep)
Keep all intermediate files that are generated during internal compilation steps.

4.2.5.5. --keep-dir directory (-keep-dir)

Keep all intermediate files that are generated during internal compilation steps in this directory.

4.2.5.6. --save-temps (-save-temps)
This option is an alias of --keep.

4.2.5.7. --clean-targets (-clean)
Delete all the non-temporary files that the same nvcc command would generate without this
option.
This option reverses the behavior of nvcc. When specified, none of the compilation
phases will be executed. Instead, all of the non-temporary files that nvcc would
otherwise create will be deleted.

4.2.5.8. --run-args arguments,... (-run-args)

Specify command line arguments for the executable when used in conjunction with --run.

4.2.5.9. --input-drive-prefix prefix (-idp)

Specify the input drive prefix.
On Windows, all command line arguments that refer to file names must be converted to
the Windows native format before they are passed to pure Windows executables. This
option specifies how the current development environment represents absolute paths.
Use /cygwin/ as prefix for Cygwin build environments and / as prefix for MinGW.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 19
 NVCC Command Options

4.2.5.10. --dependency-drive-prefix prefix (-ddp)

Specify the dependency drive prefix.
On Windows, when generating dependency files (see --generate-dependencies),
all file names must be converted appropriately for the instance of make that is used.
Some instances of make have trouble with the colon in absolute paths in the native
Windows format, which depends on the environment in which the make instance has
been compiled. Use /cygwin/ as prefix for a Cygwin make, and / as prefix for
MinGW. Or leave these file names in the native Windows format by specifying nothing.

4.2.5.11. --drive-prefix prefix (-dp)

Specify the drive prefix.
This option specifies prefix as both --input-drive-prefix and --dependency-
drive-prefix.

4.2.5.12. --dependency-target-name target (-MT)

Specify the target name of the generated rule when generating a dependency file (see --
generate-dependencies).

4.2.5.13. --no-align-double
Specify that -malign-double should not be passed as a compiler argument on 32-bit
platforms.
WARNING: this makes the ABI incompatible with the CUDA's kernel ABI for certain
64-bit types.

4.2.5.14. --no-device-link (-nodlink)
Skip the device link step when linking object files.

4.2.6. Options for Steering CUDA Compilation

4.2.6.1. --default-stream {legacy|null|per-thread} (-
default-stream)
Specify the stream that CUDA commands from the compiled program will be sent to by default.

Allowed Values
legacy
The CUDA legacy stream (per context, implicitly synchronizes with other streams)

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 20
 NVCC Command Options

per-thread
Normal CUDA stream (per thread, does not implicitly synchronize with other
streams)
null
Deprecated alias for legacy

Default
legacy is used as the default stream.

4.2.7. Options for Steering GPU Code Generation

4.2.7.1. --gpu-architecture arch (-arch)
Specify the name of the class of NVIDIA virtual GPU architecture for which the CUDA input
files must be compiled.
With the exception as described for the shorthand below, the architecture specified
with this option must be a virtual architecture (such as compute_50). Normally, this
option alone does not trigger assembly of the generated PTX for a real architecture (that
is the role of nvcc option --gpu-code, see below); rather, its purpose is to control
preprocessing and compilation of the input to PTX.
For convenience, in case of simple nvcc compilations, the following shorthand is
supported. If no value for option --gpu-code is specified, then the value of this
option defaults to the value of --gpu-architecture. In this situation, as only
exception to the description above, the value specified for --gpu-architecture
may be a real architecture (such as a sm_50), in which case nvcc uses the specified
real architecture and its closest virtual architecture as effective architecture values.
For example, nvcc --gpu-architecture=sm_50 is equivalent to nvcc --gpu-
architecture=compute_50 --gpu-code=sm_50,compute_50.

See Virtual Architecture Feature List for the list of supported virtual architectures and
GPU Feature List for the list of supported real architectures.

Default
sm_30 is used as the default value; PTX is generated for compute_30 then assembled
and optimized for sm_30.

4.2.7.2. --gpu-code code,... (-code)

Specify the name of the NVIDIA GPU to assemble and optimize PTX for.
nvcc embeds a compiled code image in the resulting executable for each specified code
architecture, which is a true binary load image for each real architecture (such as sm_50),
and PTX code for the virtual architecture (such as compute_50).

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 21
 NVCC Command Options

During runtime, such embedded PTX code is dynamically compiled by the CUDA
runtime system if no binary load image is found for the current GPU.
Architectures specified for options --gpu-architecture and --gpu-code may
be virtual as well as real, but the code architectures must be compatible with the
arch architecture. When the --gpu-code option is used, the value for the --gpu-
architecture option must be a virtual PTX architecture.

For instance, --gpu-architecture=compute_50 is not compatible with --gpu-

code=sm_30, because the earlier compilation stages will assume the availability of
compute_50 features that are not present on sm_30.

See Virtual Architecture Feature List for the list of supported virtual architectures and
GPU Feature List for the list of supported real architectures.

4.2.7.3. --generate-code specification (-gencode)

This option provides a generalization of the --gpu-architecture=arch --gpu-
code=code,... option combination for specifying nvcc behavior with respect to code
generation.
Where use of the previous options generates code for different real architectures
with the PTX for the same virtual architecture, option --generate-code allows
multiple PTX generations for different virtual architectures. In fact, --gpu-
architecture=arch --gpu-code=code,... is equivalent to --generate-
code=arch=arch,code=code,... .

--generate-code options may be repeated for different virtual architectures.

See Virtual Architecture Feature List for the list of supported virtual architectures and
GPU Feature List for the list of supported real architectures.

4.2.7.4. --relocatable-device-code {true|false} (-rdc)

Enable or disable the generation of relocatable device code.
If disabled, executable device code is generated. Relocatable device code must be linked
before it can be executed.

Allowed Values
‣ true
‣ false

Default
The generation of relocatable device code is disabled.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 22
 NVCC Command Options

4.2.7.5. --entries entry,... (-e)

Specify the global entry functions for which code must be generated.
PTX generated for all entry functions, but only the selected entry functions are
assembled. Entry function names for this option must be specified in the mangled name.

Default
nvcc generates code for all entry functions.

4.2.7.6. --maxrregcount amount (-maxrregcount)

Specify the maximum amount of registers that GPU functions can use.
Until a function-specific limit, a higher value will generally increase the performance of
individual GPU threads that execute this function. However, because thread registers
are allocated from a global register pool on each GPU, a higher value of this option will
also reduce the maximum thread block size, thereby reducing the amount of thread
parallelism. Hence, a good maxrregcount value is the result of a trade-off.
Value less than the minimum registers required by ABI will be bumped up by the
compiler to ABI minimum limit.
User program may not be able to make use of all registers as some registers are reserved
by compiler.

Default
No maximum is assumed.

4.2.7.7. --use_fast_math (-use_fast_math)
Make use of fast math library.
--use_fast_math implies --ftz=true --prec-div=false --prec-sqrt=false
--fmad=true.

4.2.7.8. --ftz {true|false} (-ftz)

Control single-precision denormals support.
--ftz=true flushes denormal values to zero and --ftz=false preserves denormal
values.
--use_fast_math implies --ftz=true.

Allowed Values
‣ true

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 23
 NVCC Command Options

‣ false

Default
This option is set to false and nvcc preserves denormal values.

4.2.7.9. --prec-div {true|false} (-prec-div)

This option controls single-precision floating-point division and reciprocals.
--prec-div=true enables the IEEE round-to-nearest mode and --prec-div=false
enables the fast approximation mode.
--use_fast_math implies --prec-div=false.

Allowed Values
‣ true
‣ false

Default
This option is set to true and nvcc enables the IEEE round-to-nearest mode.

4.2.7.10. --prec-sqrt {true|false} (-prec-sqrt)

This option controls single-precision floating-point square root.
--prec-sqrt=true enables the IEEE round-to-nearest mode and --prec-sqrt=false
enables the fast approximation mode.
--use_fast_math implies --prec-sqrt=false.

Allowed Values
‣ true
‣ false

Default
This option is set to true and nvcc enables the IEEE round-to-nearest mode.

4.2.7.11. --fmad {true|false} (-fmad)

This option enables (disables) the contraction of floating-point multiplies and adds/subtracts into
floating-point multiply-add operations (FMAD, FFMA, or DFMA).
--use_fast_math implies --fmad=true.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 24
 NVCC Command Options

Allowed Values
‣ true
‣ false

Default
This option is set to true and nvcc enables the contraction of floating-point multiplies
and adds/subtracts into floating-point multiply-add operations (FMAD, FFMA, or
DFMA).

4.2.7.12. --compile-as-tools-patch (-astoolspatch)
Compile patch code for CUDA tools. Implies --keep-device-functions.
May only be used in conjunction with --ptx or --cubin or --fatbin.
Shall not be used in conjunction with -rdc=true or -ewp.

4.2.7.13. --keep-device-functions (-keep-device-
functions)
In whole program compilation mode, preserve user defined external linkage __device__
function definitions in generated PTX.

4.2.8. Generic Tool Options

4.2.8.1. --disable-warnings (-w)
Inhibit all warning messages.

4.2.8.2. --source-in-ptx (-src-in-ptx)
Interleave source in PTX.
May only be used in conjunction with --device-debug or --generate-line-
info.

4.2.8.3. --restrict (-restrict)
Assert that all kernel pointer parameters are restrict pointers.

4.2.8.4. --Wno-deprecated-gpu-targets (-Wno-deprecated-
gpu-targets)
Suppress warnings about deprecated GPU target architectures.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 25
 NVCC Command Options

4.2.8.5. --Wno-deprecated-declarations (-Wno-deprecated-
declarations)
Suppress warning on use of a deprecated entity.

4.2.8.6. --Wreorder (-Wreorder)
Generate warnings when member initializers are reordered.

4.2.8.7. --Werror kind,... (-Werror)

Make warnings of the specified kinds into errors.
The following is the list of warning kinds accepted by this option:
all-warnings
Treat all warnings as errors.
cross-execution-space-call
Be more strict about unsupported cross execution space calls. The compiler will
generate an error instead of a warning for a call from a __host__ __device__ to a
__host__ function.
reorder
Generate errors when member initializers are reordered.
deprecated-declarations
Generate error on use of a deprecated entity.

4.2.8.8. --resource-usage (-res-usage)
Show resource usage such as registers and memory of the GPU code.
This option implies --nvlink-options=--verbose when --relocatable-
device-code=true is set. Otherwise, it implies --ptxas-options=--verbose.

4.2.8.9. --help (-h)
Print help information on this tool.

4.2.8.10. --version (-V)
Print version information on this tool.

4.2.8.11. --options-file file,... (-optf)

Include command line options from specified file.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 26
 NVCC Command Options

4.2.8.12. --time filename (-time)

Generate a comma separated value table with the time taken by each compilation phase, and
append it at the end of the file given as the option argument. If the file is empty, the column
headings are generated in the first row of the table.
If the file name is '-', the timing data is generated in stdout.

4.2.8.13. --qpp-config config (-qpp-config)

Specify the configuration ([[compiler/]version,][target]) when using q++ host compiler. The
argument will be forwarded to the q++ compiler with its -V flag.

4.2.9. Phase Options
The following sections lists some useful options to lower level compilation tools.

4.2.9.1. Ptxas Options
The following table lists some useful ptxas options which can be specified with nvcc
option -Xptxas.

4.2.9.1.1. --allow-expensive-optimizations (-allow-expensive-
optimizations)
Enable (disable) to allow compiler to perform expensive optimizations using maximum available
resources (memory and compile-time).
If unspecified, default behavior is to enable this feature for optimization level >= O2.

4.2.9.1.2. --compile-only (-c)
Generate relocatable object.

4.2.9.1.3. --def-load-cache (-dlcm)
Default cache modifier on global/generic load.
Default value: ca.

4.2.9.1.4. --def-store-cache (-dscm)
Default cache modifier on global/generic store.

4.2.9.1.5. --device-debug (-g)
Semantics same as nvcc option --device-debug.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 27
 NVCC Command Options

4.2.9.1.6. --disable-optimizer-constants (-disable-optimizer-
consts)
Disable use of optimizer constant bank.

4.2.9.1.7. --entry entry,... (-e)

Semantics same as nvcc option --entries.

4.2.9.1.8. --fmad (-fmad)
Semantics same as nvcc option --fmad.

4.2.9.1.9. --force-load-cache (-flcm)
Force specified cache modifier on global/generic load.

4.2.9.1.10. --force-store-cache (-fscm)
Force specified cache modifier on global/generic store.

4.2.9.1.11. --generate-line-info (-lineinfo)
Semantics same as nvcc option --generate-line-info.

4.2.9.1.12. --gpu-name gpuname (-arch)

Specify name of NVIDIA GPU to generate code for.
This option also takes virtual compute architectures, in which case code generation is
suppressed. This can be used for parsing only.
Allowed values for this option: compute_30, compute_35, compute_50, compute_52;
and sm_30, sm_32, sm_35, sm_50 and sm_52.
Default value: sm_30.

4.2.9.1.13. --help (-h)
Semantics same as nvcc option --help.

4.2.9.1.14. --machine (-m)
Semantics same as nvcc option --machine.

4.2.9.1.15. --maxrregcount amount (-maxrregcount)

Semantics same as nvcc option --maxrregcount.

4.2.9.1.16. --opt-level N (-O)
Specify optimization level.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 28
 NVCC Command Options

Default value: 3.

4.2.9.1.17. --options-file file,... (-optf)

Semantics same as nvcc option --options-file.

4.2.9.1.18. --output-file file (-o)

Specify name of output file.
Default value: elf.o.

4.2.9.1.19. --preserve-relocs (-preserve-relocs)
This option will make ptxas to generate relocatable references for variables and preserve
relocations generated for them in linked executable.

4.2.9.1.20. --sp-bound-check (-sp-bound-check)
Generate stack-pointer bounds-checking code sequence.
This option is turned on automatically when --device-debug or --opt-level=0 is
specified.

4.2.9.1.21. --verbose (-v)
Enable verbose mode which prints code generation statistics.

4.2.9.1.22. --version (-V)
Semantics same as nvcc option --version.

4.2.9.1.23. --warning-as-error (-Werror)
Make all warnings into errors.

4.2.9.1.24. --warn-on-double-precision-use (-warn-double-
usage)
Warning if double(s) are used in an instruction.

4.2.9.1.25. --warn-on-local-memory-usage (-warn-lmem-usage)
Warning if local memory is used.

4.2.9.1.26. --warn-on-spills (-warn-spills)
Warning if registers are spilled to local memory.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 29
 NVCC Command Options

4.2.9.2. NVLINK Options
The following table lists some useful nvlink options which can be specified with nvcc
option --nvlink-options.

4.2.9.2.1. --disable-warnings (-w)
Inhibit all warning messages.

4.2.9.2.2. --preserve-relocs (-preserve-relocs)
Preserve resolved relocations in linked executable.

4.2.9.2.3. --verbose (-v)
Enable verbose mode which prints code generation statistics.

4.2.9.2.4. --warning-as-error (-Werror)
Make all warnings into errors.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 30
Chapter 5.
GPU COMPILATION

This chapter describes the GPU compilation model that is maintained by nvcc, in
cooperation with the CUDA driver. It goes through some technical sections, with
concrete examples at the end.

5.1. GPU Generations
In order to allow for architectural evolution, NVIDIA GPUs are released in different
generations. New generations introduce major improvements in functionality and/
or chip architecture, while GPU models within the same generation show minor
configuration differences that moderately affect functionality, performance, or both.
Binary compatibility of GPU applications is not guaranteed across different generations.
For example, a CUDA application that has been compiled for a Fermi GPU will
very likely not run on a Kepler GPU (and vice versa). This is the instruction set and
instruction encodings of a geneartion is different from those of of other generations.
Binary compatibility within one GPU generation can be guaranteed under certain
conditions because they share the basic instruction set. This is the case between two GPU
versions that do not show functional differences at all (for instance when one version is
a scaled down version of the other), or when one version is functionally included in the
other. An example of the latter is the base Kepler version sm_30 whose functionality is
a subset of all other Kepler versions: any code compiled for sm_30 will run on all other
Kepler GPUs.

5.2. GPU Feature List

The following table lists the names of the current GPU architectures, annotated with the
functional capabilities that they provide. There are other differences, such as amounts of
register and processor clusters, that only affect execution performance.
In the CUDA naming scheme, GPUs are named sm_xy, where x denotes the GPU
generation number, and y the version in that generation. Additionally, to facilitate
comparing GPU capabilities, CUDA attempts to choose its GPU names such that if
x1y1 <= x2y2 then all non-ISA related capabilities of sm_x1y1 are included in those of

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 31
 GPU Compilation

sm_x2y2. From this it indeed follows that sm_30 is the base Kepler model, and it also
explains why higher entries in the tables are always functional extensions to the lower
entries. This is denoted by the plus sign in the table. Moreover, if we abstract from the
instruction encoding, it implies that sm_30's functionality will continue to be included
in all later GPU generations. As we will see next, this property will be the foundation for
application compatibility support by nvcc.
sm_30 and sm_32 Basic features
+ Kepler support
+ Unified memory programming
sm_35 + Dynamic parallelism support
sm_50, sm_52, and + Maxwell support
sm_53

sm_60, sm_61, and + Pascal support

sm_62

sm_70 and sm_72 + Volta support

sm_75 + Turing support

5.3. Application Compatibility
Binary code compatibility over CPU generations, together with a published instruction
set architecture is the usual mechanism for ensuring that distributed applications out
there in the field will continue to run on newer versions of the CPU when these become
mainstream.
This situation is different for GPUs, because NVIDIA cannot guarantee binary
compatibility without sacrificing regular opportunities for GPU improvements. Rather,
as is already conventional in the graphics programming domain, nvcc relies on a
two stage compilation model for ensuring application compatibility with future GPU
generations.

5.4. Virtual Architectures
GPU compilation is performed via an intermediate representation, PTX, which can be
considered as assembly for a virtual GPU architecture. Contrary to an actual graphics
processor, such a virtual GPU is defined entirely by the set of capabilities, or features,
that it provides to the application. In particular, a virtual GPU architecture provides a
(largely) generic instruction set, and binary instruction encoding is a non-issue because
PTX programs are always represented in text format.
Hence, a nvcc compilation command always uses two architectures: a virtual
intermediate architecture, plus a real GPU architecture to specify the intended processor
to execute on. For such an nvcc command to be valid, the real architecture must be an
implementation of the virtual architecture. This is further explained below.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 32
 GPU Compilation

The chosen virtual architecture is more of a statement on the GPU capabilities that
the application requires: using a smallest virtual architecture still allows a widest range
of actual architectures for the second nvcc stage. Conversely, specifying a virtual
architecture that provides features unused by the application unnecessarily restricts the
set of possible GPUs that can be specified in the second nvcc stage.
From this it follows that the virtual architecture should always be chosen as low as
possible, thereby maximizing the actual GPUs to run on. The real architecture should be
chosen as high as possible (assuming that this always generates better code), but this is
only possible with knowledge of the actual GPUs on which the application is expected
to run. As we will see later, in the situation of just in time compilation, where the driver
has this exact knowledge: the runtime GPU is the one on which the program is about to
be launched/executed.
virt ual com put e archit ect ure

NVCC
x.cu (device code)

St age 1
(PTX Generat ion)

x.pt x

CUDA Runt im e
real sm archit ect ure

St age 2
(Cubin Generat ion)

x.cubin Execut e

Figure 2 Two-Staged Compilation with Virtual and Real Architectures

5.5. Virtual Architecture Feature List

compute_30 and Basic features
compute_32
+ Kepler support
+ Unified memory programming
compute_35 + Dynamic parallelism support
compute_50, + Maxwell support
compute_52, and
compute_53

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 33
 GPU Compilation

compute_60, + Pascal support

compute_61, and
compute_62

compute_70 and + Volta support

compute_72

compute_75 + Turing support

The above table lists the currently defined virtual architectures. The virtual architecture
naming scheme is the same as the real architecture naming scheme shown in Section
GPU Feature List.

5.6. Further Mechanisms
Clearly, compilation staging in itself does not help towards the goal of application
compatibility with future GPUs. For this we need the two other mechanisms by CUDA
Samples: just in time compilation (JIT) and fatbinaries.

5.6.1. Just-in-Time Compilation
The compilation step to an actual GPU binds the code to one generation of GPUs. Within
that generation, it involves a choice between GPU coverage and possible performance.
For example, compiling to sm_30 allows the code to run on all Kepler-generation GPUs,
but compiling to sm_35 would probably yield better code if Kepler GK110 and later are
the only targets.
virt ual com put e archit ect ure

NVCC
x.cu (device code)

St age 1
(PTX Generat ion)

x.pt x

CUDA Runt im e
real sm archit ect ure

St age 2
(Cubin Generat ion)

x.cubin

Execut e

Figure 3 Just-in-Time Compilation of Device Code

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 34
 GPU Compilation

By specifying a virtual code architecture instead of a real GPU, nvcc postpones the
assembly of PTX code until application runtime, at which the target GPU is exactly
known. For instance, the command below allows generation of exactly matching GPU
binary code, when the application is launched on an sm_50 or later architecture.
nvcc x.cu --gpu-architecture=compute_50 --gpu-code=compute_50

The disadvantage of just in time compilation is increased application startup delay,

but this can be alleviated by letting the CUDA driver use a compilation cache (refer to
"Section 3.1.1.2. Just-in-Time Compilation" of CUDA C Programming Guide) which is
persistent over multiple runs of the applications.

5.6.2. Fatbinaries
A different solution to overcome startup delay by JIT while still allowing execution on
newer GPUs is to specify multiple code instances, as in
nvcc x.cu --gpu-architecture=compute_50 --gpu-code=compute_50,sm_50,sm_52

This command generates exact code for two Kepler variants, plus PTX code for use by
JIT in case a next-generation GPU is encountered. nvcc organizes its device code in
fatbinaries, which are able to hold multiple translations of the same GPU source code. At
runtime, the CUDA driver will select the most appropriate translation when the device
function is launched.

5.7. NVCC Examples
5.7.1. Base Notation
nvcc provides the options --gpu-architecture and --gpu-code for specifying the
target architectures for both translation stages. Except for allowed short hands described
below, the --gpu-architecture option takes a single value, which must be the name
of a virtual compute architecture, while option --gpu-code takes a list of values which
must all be the names of actual GPUs. nvcc performs a stage 2 translation for each of
these GPUs, and will embed the result in the result of compilation (which usually is a
host object file or executable).
Example
nvcc x.cu --gpu-architecture=compute_50 --gpu-code=sm_50,sm_52

5.7.2. Shorthand
nvcc allows a number of shorthands for simple cases.

5.7.2.1. Shorthand 1
--gpu-code arguments can be virtual architectures. In this case the stage 2 translation
will be omitted for such virtual architecture, and the stage 1 PTX result will be

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 35
 GPU Compilation

embedded instead. At application launch, and in case the driver does not find a better
alternative, the stage 2 compilation will be invoked by the driver with the PTX as input.
Example
nvcc x.cu --gpu-architecture=compute_50 --gpu-code=compute_50,sm_50,sm_52

5.7.2.2. Shorthand 2
The --gpu-code option can be omitted. Only in this case, the --gpu-architecture
value can be a non-virtual architecture. The --gpu-code values default to the
closest virtual architecture that is implemented by the GPU specified with --gpu-
architecture, plus the --gpu-architecture, value itself. The closest virtual
architecture is used as the effective --gpu-architecture, value. If the --gpu-
architecture value is a virtual architecture, it is also used as the effective --gpu-code
value.
Example
nvcc x.cu --gpu-architecture=sm_52
nvcc x.cu --gpu-architecture=compute_50

are equivalent to

nvcc x.cu --gpu-architecture=compute_52 --gpu-code=sm_52,compute_52

nvcc x.cu --gpu-architecture=compute_50 --gpu-code=compute_50

5.7.2.3. Shorthand 3
Both --gpu-architecture and --gpu-code options can be omitted.
Example
nvcc x.cu

is equivalent to

nvcc x.cu --gpu-architecture=compute_30 --gpu-code=sm_30,compute_30

5.7.3. Extended Notation
The options --gpu-architecture and --gpu-code can be used in all cases where
code is to be generated for one or more GPUs using a common virtual architecture. This
will cause a single invocation of nvcc stage 1 (that is, preprocessing and generation of
virtual PTX assembly code), followed by a compilation stage 2 (binary code generation)
repeated for each specified GPU.
Using a common virtual architecture means that all assumed GPU features are fixed
for the entire nvcc compilation. For instance, the following nvcc command assumes no
half-precision floating-point operation support for both the sm_50 code and the sm_53
code:
nvcc x.cu --gpu-architecture=compute_50 --gpu-code=compute_50,sm_50,sm_53

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 36
 GPU Compilation

Sometimes it is necessary to perform different GPU code generation steps, partitioned

over different architectures. This is possible using nvcc option --generate-code,
which then must be used instead of a --gpu-architecture and --gpu-code
combination.
Unlike option --gpu-architecture option --generate-code, may be repeated
on the nvcc command line. It takes sub-options arch and code, which must not
be confused with their main option equivalents, but behave similarly. If repeated
architecture compilation is used, then the device code must use conditional compilation
based on the value of the architecture identification macro __CUDA_ARCH__, which is
described in the next section.
For example, the following assumes absence of half-precision floating-point operation
support for the sm_50 and sm_52 code, but full support on sm_53:
nvcc x.cu \
--generate-code arch=compute_50,code=sm_50 \
--generate-code arch=compute_50,code=sm_52 \
--generate-code arch=compute_53,code=sm_53

Or, leaving actual GPU code generation to the JIT compiler in the CUDA driver:
nvcc x.cu \
--generate-code arch=compute_50,code=compute_50 \
--generate-code arch=compute_53,code=compute_53

The code sub-options can be combined with a slightly more complex syntax:
nvcc x.cu \
--generate-code arch=compute_50,code=[sm_50,sm_52] \
--generate-code arch=compute_53,code=sm_53

5.7.4. Virtual Architecture Identification Macro

The architecture identification macro __CUDA_ARCH__ is assigned a three-digit value
string xy0 (ending in a literal 0) during each nvcc compilation stage 1 that compiles for
compute_xy.
This macro can be used in the implementation of GPU functions for determining the
virtual architecture for which it is currently being compiled. The host code (the non-GPU
code) must not depend on it.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 37
Chapter 6.
USING SEPARATE COMPILATION IN CUDA

Prior to the 5.0 release, CUDA did not support separate compilation, so CUDA code
could not call device functions or access variables across files. Such compilation is
referred to as whole program compilation. We have always supported the separate
compilation of host code, it was just the device CUDA code that needed to all be within
one file. Starting with CUDA 5.0, separate compilation of device code is supported,
but the old whole program mode is still the default, so there are new options to invoke
separate compilation.

6.1. Code Changes for Separate Compilation

The code changes required for separate compilation of device code are the same as what
you already do for host code, namely using extern and static to control the visibility
of symbols. Note that previously extern was ignored in CUDA code; now it will be
honored. With the use of static it is possible to have multiple device symbols with
the same name in different files. For this reason, the CUDA API calls that referred to
symbols by their string name are deprecated; instead the symbol should be referenced
by its address.

6.2. NVCC Options for Separate Compilation

CUDA works by embedding device code into host objects. In whole program
compilation, it embeds executable device code into the host object. In separate
compilation, we embed relocatable device code into the host object, and run nvlink,
the device linker, to link all the device code together. The output of nvlink is then linked
together with all the host objects by the host linker to form the final executable.
The generation of relocatable vs executable device code is controlled by the --
relocatable-device-code option.
The --compile option is already used to control stopping a compile at a host object,
so a new option --device-c is added that simply does --relocatable-device-
code=true --compile .

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 38
 Using Separate Compilation in CUDA

To invoke just the device linker, the --device-link option can be used, which emits
a host object containing the embedded executable device code. The output of that must
then be passed to the host linker. Or:
nvcc <objects>

can be used to implicitly call both the device and host linkers. This works because if the
device linker does not see any relocatable code it does not do anything.
Figure 4 shows the flow (nvcc --device-c has the same flow as #unique_46/
unique_46_Connect_42_cuda-compilation-from-cu-to-o)
x.cu y.cu z.cpp

nvcc --device-c nvcc --device-c C+ + Com piler

x.o / x.obj y.o / y.obj z.o / z.obj

Device Linker

a_dlink.o / a_dlink.obj

Host Linker

execut able / library

Figure 4 CUDA Separate Compilation Trajectory

6.3. Libraries
The device linker has the ability to read the static host library formats (.a on Linux and
Mac OS X, .lib on Windows). It ignores any dynamic (.so or .dll) libraries. The --
library and --library-path options can be used to pass libraries to both the device
and host linker. The library name is specified without the library file extension when the
--library option is used.

nvcc --gpu-architecture=sm_50 a.o b.o --library-path=<path> --library=foo

Alternatively, the library name, including the library file extension, can be used without
the --library option on Windows.
nvcc --gpu-architecture=sm_50 a.obj b.obj foo.lib --library-path=<path>

Note that the device linker ignores any objects that do not have relocatable device code.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 39
 Using Separate Compilation in CUDA

6.4. Examples
Suppose we have the following files:
//---------- b.h ----------
#define N 8

extern __device__ int g[N];

extern __device__ void bar(void);

//---------- b.cu ----------
#include "b.h"

__device__ int g[N];

__device__ void bar (void)

{
g[threadIdx.x]++;
}
//---------- a.cu ----------
#include <stdio.h>
#include "b.h"

__global__ void foo (void) {

__shared__ int a[N];

a[threadIdx.x] = threadIdx.x;

__syncthreads();

g[threadIdx.x] = a[blockDim.x - threadIdx.x - 1];

bar();
}

int main (void) {

unsigned int i;
int *dg, hg[N];
int sum = 0;

foo<<<1, N>>>();

if(cudaGetSymbolAddress((void**)&dg, g)){
printf("couldn't get the symbol addr\n");
return 1;
}
if(cudaMemcpy(hg, dg, N * sizeof(int), cudaMemcpyDeviceToHost)){
printf("couldn't memcpy\n");
return 1;
}

for (i = 0; i < N; i++) {

sum += hg[i];
}
if (sum == 36) {
printf("PASSED\n");
} else {
printf("FAILED (%d)\n", sum);
}

return 0;
}

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 40
 Using Separate Compilation in CUDA

These can be compiled with the following commands (these examples are for Linux):
nvcc --gpu-architecture=sm_50 --device-c a.cu b.cu
nvcc --gpu-architecture=sm_50 a.o b.o

If you want to invoke the device and host linker separately, you can do:
nvcc --gpu-architecture=sm_50 --device-c a.cu b.cu
nvcc --gpu-architecture=sm_50 --device-link a.o b.o --output-file link.o
g++ a.o b.o link.o --library-path=<path> --library=cudart

Note that all desired target architectures must be passed to the device linker, as that
specifies what will be in the final executable (some objects or libraries may contain
device code for multiple architectures, and the link step can then choose what to put in
the final executable).
If you want to use the driver API to load a linked cubin, you can request just the cubin:
nvcc --gpu-architecture=sm_50 --device-link a.o b.o \
--cubin --output-file link.cubin

The objects could be put into a library and used with:

nvcc --gpu-architecture=sm_50 --device-c a.cu b.cu
nvcc --lib a.o b.o --output-file test.a
nvcc --gpu-architecture=sm_50 test.a

Note that only static libraries are supported by the device linker.
A PTX file can be compiled to a host object file and then linked by using:
nvcc --gpu-architecture=sm_50 --device-c a.ptx

An example that uses libraries, host linker, and dynamic parallelism would be:
nvcc --gpu-architecture=sm_50 --device-c a.cu b.cu
nvcc --gpu-architecture=sm_50 --device-link a.o b.o --output-file link.o
nvcc --lib --output-file libgpu.a a.o b.o link.o
g++ host.o --library=gpu --library-path=<path> \
--library=cudadevrt --library=cudart

It is possible to do multiple device links within a single host executable, as long as each
device link is independent of the other. This requirement of independence means that
they cannot share code across device executables, nor can they share addresses (e.g.,
a device function address can be passed from host to device for a callback only if the
device link sees both the caller and potential callback callee; you cannot pass an address
from one device executable to another, as those are separate address spaces).

6.5. Potential Separate Compilation Issues

6.5.1. Object Compatibility
Only relocatable device code with the same ABI version, link-compatible SM target
architecture, and same pointer size (32 or 64) can be linked together. Incompatible
objects will produce a link error. Link-compatible SM architectures are ones that have
compatible SASS binaries that can combine without translating, e.g. sm_52 and sm_50.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 41
 Using Separate Compilation in CUDA

An object could have been compiled for a different architecture but also have PTX
available, in which case the device linker will JIT the PTX to cubin for the desired
architecture and then link. Relocatable device code requires CUDA 5.0 or later Toolkit.
If a kernel is limited to a certain number of registers with the launch_bounds attribute
or the --maxrregcount option, then all functions that the kernel calls must not use
more than that number of registers; if they exceed the limit, then a link error will be
given.

6.5.2. JIT Linking Support

CUDA 5.0 does not support JIT linking, while CUDA 5.5 does. This means that to use JIT
linking you must recompile your code with CUDA 5.5 or later. JIT linking means doing
a relink of the code at startup time. The device linker (nvlink) links at the cubin level.
If the cubin does not match the target architecture at load time, the driver re-invokes
the device linker to generate cubin for the target architecture, by first JIT'ing the PTX for
each object to the appropriate cubin, and then linking together the new cubin.

6.5.3. Implicit CUDA Host Code

A file like b.cu above only contains CUDA device code, so one might think that the
b.o object doesn't need to be passed to the host linker. But actually there is implicit
host code generated whenever a device symbol can be accessed from the host side,
either via a launch or an API call like cudaGetSymbolAddress(). This implicit host
code is put into b.o, and needs to be passed to the host linker. Plus, for JIT linking to
work all device code must be passed to the host linker, else the host executable will not
contain device code needed for the JIT link. So a general rule is that the device linker
and host linker must see the same host object files (if the object files have any device
references in them—if a file is pure host then the device linker doesn't need to see it). If
an object file containing device code is not passed to the host linker, then you will see
an error message about the function __cudaRegisterLinkedBinary_name calling an
undefined or unresolved symbol __fatbinwrap_name.

6.5.4. Using __CUDA_ARCH__
In separate compilation, __CUDA_ARCH__ must not be used in headers such that
different objects could contain different behavior. Or, it must be guaranteed that all
objects will compile for the same compute_arch. If a weak function or template function
is defined in a header and its behavior depends on __CUDA_ARCH__, then the instances
of that function in the objects could conflict if the objects are compiled for different
compute arch. For example, if an a.h contains:
template<typename T>
__device__ T* getptr(void)
{
#if __CUDA_ARCH__ == 500
return NULL; /* no address */
#else
__shared__ T arr[256];
return arr;
#endif
}

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 42
 Using Separate Compilation in CUDA

Then if a.cu and b.cu both include a.h and instantiate getptr for the same type, and
b.cu expects a non-NULL address, and compile with:
nvcc --gpu-architecture=compute_50 --device-c a.cu
nvcc --gpu-architecture=compute_52 --device-c b.cu
nvcc --gpu-architecture=sm_52 a.o b.o

At link time only one version of the getptr is used, so the behavior would depend
on which version is picked. To avoid this, either a.cu and b.cu must be compiled for
the same compute arch, or __CUDA_ARCH__ should not be used in the shared header
function.

6.5.5. Device Code in Libraries

If a device function with non-weak external linkage is defined in a library as well as a
non-library object (or another library), the device linker will complain about the multiple
definitions (this differs from traditional host linkers that may ignore the function
definition from the library object, if it was already found in an earlier object).

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 43
Chapter 7.
MISCELLANEOUS NVCC USAGE

7.1. Cross Compilation
Cross compilation is controlled by using the following nvcc command line options:
‣ --compiler-bindir is used for cross compilation, where the underlying host
compiler is capable of generating objects for the target platform.
‣ --machine=32. This option signals that the target platform is a 32-bit platform. Use
this when the host platform is a 64-bit platform.

7.2. Keeping Intermediate Phase Files

nvcc stores intermediate results by default into temporary files that are deleted
immediately before it completes. The location of the temporary file directories used are,
depending on the current platform, as follows:
Windows
Value of environment variable TEMP is used. If it is not set, C:\Windows\temp is used
instead.
Other Platforms
Value of environment variable TMPDIR is used. If it is not set, /tmp is used instead.
Option --keep makes nvcc store these intermediate files in the current directory or in
the directory specified by --keep-dir instead, with names as described in Supported
Phases.

7.3. Cleaning Up Generated Files

All files generated by a particular nvcc command can be cleaned up by repeating the
command, but with additional option --clean-targets. This option is particularly
useful after using --keep, because the --keep option usually leaves quite an amount of
intermediate files around.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 44
 Miscellaneous NVCC Usage

Because using --clean-targets will remove exactly what the original nvcc command
created, it is important to exactly repeat all of the options in the original command. For
instance, in the following example, omitting --keep, or adding --compile will have
different cleanup effects.
nvcc acos.cu --keep
nvcc acos.cu --keep --clean-targets

7.4. Printing Code Generation Statistics

A summary on the amount of used registers and the amount of memory needed per
compiled device function can be printed by passing option --resource-usage to nvcc:
$ nvcc --resource-usage acos.cu
ptxas info : 1536 bytes gmem, 8 bytes cmem[14]
ptxas info : Compiling entry function 'acos_main' for 'sm_30'
ptxas info : Function properties for acos_main
0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info : Used 6 registers, 1536 bytes smem, 32 bytes cmem[0]

As shown in the above example, the amounts of statically allocated global memory
(gmem) and constant memory in bank 14 (cmem) are listed.
Global memory and some of the constant banks are module scoped resources and
not per kernel resources. Allocation of constant variables to constant banks is profile
specific.
Followed by this, per kernel resource information is printed.
Stack frame is per thread stack usage used by this function. Spill stores and loads
represent stores and loads done on stack memory which are being used for storing
variables that couldn't be allocated to physical registers.
Similarly number of registers, amount of shared memory and total space in constant
bank allocated is shown.

www.nvidia.com
CUDA Compiler Driver NVCC TRM-06721-001_v10.2 | 45
Notice
ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS,
DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY,
"MATERIALS") ARE BEING PROVIDED "AS IS." NVIDIA MAKES NO WARRANTIES,
EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE
MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF
NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR
PURPOSE.
Information furnished is believed to be accurate and reliable. However, NVIDIA
Corporation assumes no responsibility for the consequences of use of such
information or for any infringement of patents or other rights of third parties
that may result from its use. No license is granted by implication of otherwise
under any patent rights of NVIDIA Corporation. Specifications mentioned in this
publication are subject to change without notice. This publication supersedes and
replaces all other information previously supplied. NVIDIA Corporation products
are not authorized as critical components in life support devices or systems
without express written approval of NVIDIA Corporation.

Trademarks
NVIDIA and the NVIDIA logo are trademarks or registered trademarks of NVIDIA
Corporation in the U.S. and other countries. Other company and product names
may be trademarks of the respective companies with which they are associated.

Copyright
© 2007-2019 NVIDIA Corporation. All rights reserved.

www.nvidia.com