Specification of Fixed Point Math Routines

AUTOSAR CP R23-11

Specification of Fixed Point Math

Document Title Routines
Document Owner AUTOSAR
Document Responsibility AUTOSAR
Document Identification No 394

Document Status published

Part of AUTOSAR Standard Classic Platform
Part of Standard Release R23-11

Document Change History

Date Release Changed by Description
AUTOSAR
2023-11-23 R23-11 Release • No content changes.
Management
• Update [SWS_Mfx_00064] for the
Function IDs 0x152 to 0x156 with the
new ’Associated maximum shift’ value

• Corrected the datatype of

AUTOSAR Mfx_Add_u32s32_u8 function
2022-11-24 R22-11 Release (SWS_Mfx_00008)
Management
• Correct parameter description for
(SWS_Mfx_00053)

• New Signum function Mfx_Sgn_s32_s8

is added
• Missing input parameter and return
AUTOSAR value description of Mfx_DivShLeft
2021-11-25 R21-11 Release function (SWS_Mfx_00058) added
Management
• Editorial change (converted to LaTex)
AUTOSAR
2020-11-30 R20-11 Release • Chapter 7.1 Error sections updated
Management
AUTOSAR • Editorial changes
2019-11-28 R19-11 Release • Changed Document Status from Final to
Management published
5

1 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
AUTOSAR
2018-10-31 4.4.0 Release • Editorial changes
Management
• Added requirement tracing information
for SWS_Mfx_00024

AUTOSAR • Removal of (**) from Mul variants in

2017-12-08 4.3.1 Release SWS_Mfx_00024
Management
• Addition of (*) for 0x078

• Renamed "Development Error Tracer" to

"Default Error Tracer" in abbreviations
• The renaming of "Development Error
Tracer" to "Default Error Tracer" is done
in abbreviations

• Removal of the requirement SWS_Mfx_

AUTOSAR 00204
2016-11-30 4.3.0 Release
• Maximum shift value updated for SWS_
Management
Mfx_00064

• Updated SWS_Mfx_00073 for clarity in

minmax handling

• Clarifications
• Updated SWS_Mfx_00017 for shift value
AUTOSAR of Function ID 0x200 to 0x205 from 64 to
2015-07-31 4.2.2 Release 63
Management
• Updated SWS_Mfx_00001 under
Section 5.1 File Structure
AUTOSAR
2014-10-31 4.2.1 Release • Minor corrections and clarifications
Management
AUTOSAR
2014-03-31 4.1.3 Release • Editorial Changes
Management
AUTOSAR • Corrections and removals of duplicate
Release functions
2013-10-31 4.1.2
Management • Editorial changes
AUTOSAR
2013-03-15 4.1.1 • Editorial Changes
Administration
5

2 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
• Addition to the list of function for
AUTOSAR consistency and completeness
2011-12-22 4.0.3
Administration
• Fix typing errors in document
• New API created to achieve completion
of the need
AUTOSAR
2010-09-30 3.1.5
Administration • File structure has been detailed for what
concerns naming conventions
AUTOSAR
2010-02-02 3.1.4 • Initial Release
Administration

3 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

Disclaimer

This work (specification and/or software implementation) and the material contained in
it, as released by AUTOSAR, is for the purpose of information only. AUTOSAR and the
companies that have contributed to it shall not be liable for any use of the work.
The material contained in this work is protected by copyright and other types of intel-
lectual property rights. The commercial exploitation of the material contained in this
work requires a license to such intellectual property rights.
This work may be utilized or reproduced without any modification, in any form or by
any means, for informational purposes only. For any other purpose, no part of the work
may be utilized or reproduced, in any form or by any means, without permission in
writing from the publisher.
The work has been developed for automotive applications only. It has neither been
developed, nor tested for non-automotive applications.
The word AUTOSAR and the AUTOSAR logo are registered trademarks.

4 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

Contents
1 Introduction and functional overview 8

2 Acronyms and Abbreviations 9

3 Related documentation 10
3.1 Input documents & related standards and norms . . . . . . . . . . . . . 10
3.2 Related specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4 Constraints and assumptions 11
4.1 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2 Applicability to car domains . . . . . . . . . . . . . . . . . . . . . . . . . 11
5 Dependencies to other modules 12
5.1 File structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6 Requirements Tracing 13

7 Functional specification 15
7.1 Error classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7.1.1 Development Errors . . . . . . . . . . . . . . . . . . . . . . . . 15
7.1.2 Runtime Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7.1.3 Transient Faults . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7.1.4 Production Error . . . . . . . . . . . . . . . . . . . . . . . . . . 15
7.1.5 Extended Production Errors . . . . . . . . . . . . . . . . . . . . 15
7.2 Initialization and shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7.3 Using Library API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
7.4 Library implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
8 API specification 18
8.1 Imported types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
8.2 Type definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
8.3 Comment about rounding . . . . . . . . . . . . . . . . . . . . . . . . . . 19
8.4 Comment about routines optimization . . . . . . . . . . . . . . . . . . . . 19
8.4.1 Optimized with constants . . . . . . . . . . . . . . . . . . . . . 19
8.5 Mathematical routines definitions . . . . . . . . . . . . . . . . . . . . . . 20
8.5.1 Additions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
8.5.2 Subtractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
8.5.3 Absolute value . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
8.5.4 Absolute value of a difference . . . . . . . . . . . . . . . . . . 24
8.5.5 Multiplications . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
8.5.6 Divisions rounded towards 0 . . . . . . . . . . . . . . . . . . . 28
8.5.7 Divisions rounded off . . . . . . . . . . . . . . . . . . . . . . . 30
8.5.8 Combinations of multiplication and division rounded towards 0 32
8.5.9 Combinations of multiplication and division rounded off . . . . 34
8.5.10 Combinations of multiplication and shift right . . . . . . . . . . 36

5 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.11 Combinations of division and shift left . . . . . . . . . . . . . . 38

8.5.12 Modulo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
8.5.13 Limiting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
8.5.14 Limitations with only one value for minimum and maximum . . 42
8.5.15 Minimum and maximum . . . . . . . . . . . . . . . . . . . . . . 43
8.5.16 Signum Function . . . . . . . . . . . . . . . . . . . . . . . . . . 44
8.6 2n Scaled Integer Math Functions . . . . . . . . . . . . . . . . . . . . . . 44
8.6.1 Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.6.1.1 16-Bit to 8-Bit 2n Scaled Integer Conversion . . . . . . 46
8.6.1.2 8-Bit to 16-Bit 2n Scaled Integer Conversion . . . . . . 47
8.6.1.3 32-Bit to 16-Bit 2n Scaled Integer Conversion . . . . . 48
8.6.1.4 16-Bit to 32-Bit 2n Scaled Integer Conversion . . . . . 49
8.6.2 Multiplication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
8.6.2.1 16-Bit Multiplication of 2n Scaled Integer . . . . . . . . 50
8.6.2.2 32-Bit Multiplication of 2n Scaled Integer . . . . . . . . 51
8.6.3 Division . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
8.6.3.1 16-Bit Division of 2n Scaled Integer . . . . . . . . . . . 52
8.6.3.2 32-Bit Division of 2n Scaled Integer . . . . . . . . . . . 54
8.6.4 Addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
8.6.4.1 16-Bit Addition of 2n Scaled Integer . . . . . . . . . . 55
8.6.4.2 32-Bit Addition of 2n Scaled Integer . . . . . . . . . . 57
8.6.5 Subtraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
8.6.5.1 16-Bit Subtraction of 2n Scaled Integer . . . . . . . . . 58
8.6.5.2 32-Bit Subtraction of 2n Scaled Integer . . . . . . . . . 60
8.6.6 Absolute Difference of 2n Scaled Integer . . . . . . . . . . . . 61
8.6.7 Absolute Value . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
8.6.7.1 16-Bit Absolute Value of 2n Scaled Integer . . . . . . . 63
8.6.7.2 32-Bit Absolute Value of 2n Scaled Integer . . . . . . . 64
8.7 Examples of use of functions . . . . . . . . . . . . . . . . . . . . . . . . 65
8.7.1 Combinations of multiplication and shift right . . . . . . . . . . 65
8.7.2 Combinations of division and shift left . . . . . . . . . . . . . . 65
8.8 Version API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8.8.1 Mfx_GetVersionInfo . . . . . . . . . . . . . . . . . . . . . . . . 66
8.9 Callback notifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8.10 Scheduled functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8.11 Expected interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8.11.1 Mandatory interfaces . . . . . . . . . . . . . . . . . . . . . . . 67
8.11.2 Optional interfaces . . . . . . . . . . . . . . . . . . . . . . . . . 67
8.11.3 Configurable interfaces . . . . . . . . . . . . . . . . . . . . . . 67
8.12 Service Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
9 Sequence diagrams 68

10 Configuration specification 69
10.1 How to read this chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
10.2 Containers and configuration parameters . . . . . . . . . . . . . . . . . . 69
10.3 Published Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

6 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

A Not applicable requirements 70

B History of Specification Items 71

B.1 Specification Item History of this document compared to AUTOSAR
R22-11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
B.1.1 Added Specification Items in R23-11 . . . . . . . . . . . . . . . 71
B.1.2 Changed Specification Items in R23-11 . . . . . . . . . . . . . 71
B.1.3 Deleted Specification Items in R23-11 . . . . . . . . . . . . . . 71

7 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

1 Introduction and functional overview

This specification specifies the functionality, API and the configuration of the AUTOSAR
library dedicated to arithmetic routines for fixed point values.
This mathematical library (MFX) contains the following routines :
• addition
• subtraction
• absolute value
• absolute value of differences
• multiplication
• division
• combination of multiplication and division
• combination of multiplication and shift right
• combination of division and shift left
• modulo
• limitation
Some of these functions are proposed too for 2n Scaled Integers :
• addition
• subtraction
• absolute value
• absolute value of differences
• multiplication
• division
• conversion (specific to 2n Scaled Integers)
All routines are re-entrant and can be used by multiple runnables at the same time.

8 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

2 Acronyms and Abbreviations

The glossary below includes acronyms and abbreviations relevant to the MFXLibrary
module that are not included in the [1, AUTOSAR glossary].
Abbreviation / Acronym: Description:
Abs Absolute value
AbsDiff Absolute value of a difference
Add Addition
AR Autosar
BSW Basic Software
DET Default Error Tracer
Div Division
DivShLeft Combination of division and shift left
ECU Electronic Control Unit
Limit Limitation routine
Max Maximum
MFX/Mfx Math - Fixed Point library
Min Minimum
Minmax Limitation with only one value for min and max
Mod Modulo routine
Mul Multiplication
MulDiv Combination of multiplication and division
MulShRight Combination of multiplication and shift right
s16 Mnemonic for the sint16, specified in AUTOSAR_SWS_
PlatformTypes
s32 Mnemonic for the sint32, specified in AUTOSAR_SWS_
PlatformTypes
s8 Mnemonic for the sint8, specified in AUTOSAR_SWS_
PlatformTypes
Sub Subtraction
SWS Software Specification
u16 Mnemonic for the uint16, specified in AUTOSAR_SWS_
PlatformTypes
u32 Mnemonic for the uint32, specified in AUTOSAR_SWS_
PlatformTypes
u8 Mnemonic for the uint8, specified in AUTOSAR_SWS_
PlatformTypes

9 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

3 Related documentation

3.1 Input documents & related standards and norms

[1] Glossary
AUTOSAR_FO_TR_Glossary
[2] ISO/IEC 9899:1990 Programming Language - C
https://www.iso.org
[3] General Specification of Basic Software Modules
AUTOSAR_CP_SWS_BSWGeneral
[4] General Requirements on Basic Software Modules
AUTOSAR_CP_SRS_BSWGeneral
[5] Requirements on Libraries
AUTOSAR_CP_SRS_Libraries

3.2 Related specification

AUTOSAR provides a General Specification on Basic Software modules [3, SWS BSW
General], which is also valid for MFXLibrary.
Thus, the specification SWS BSW General shall be considered as additional and re-
quired specification for MFXLibrary.

10 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4 Constraints and assumptions

4.1 Limitations
• No requirements on Service library can be implemented in multiple ways. Many
small routines can be combined into one implementation file. For bigger routines,
one file shall contain one routine implementation. Generally one routine per ob-
ject file is recommended from linker optimization point of view. For Bit handling
routines more routines can contribute to form one object file. This kind of group-
ing is not achieved in Release 4.0, Rev001 and will be addressed in Release 4.0,
rev002

4.2 Applicability to car domains

No restrictions.

11 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

5 Dependencies to other modules

5.1 File structure

[SWS_Mfx_00001] dThe MFX module shall provide the following files:
• C files, Mfx_<name>.c used to implement the library. All C files shall be prefixed
with ’Mfx’.
c(SRS_LIBS_00005)
Implementation & grouping of routines with respect to C files is recommended as per
below options and there is no restriction to follow the same.
Option 1 : <Name> can be function name providing one C file per function,
eg.: Mfx_Add_u8u8_u8.c etc.
Option 2 : <Name> can have common name of group of functions:
• 2.1 Group by object family:
eg.: Mfx_NomMath.c, Mfx_ScaledMath.c
• 2.2 Group by routine family:
eg.: Mfx_Add.c
• 2.3 Group by method family: if it makes sense
• 2.4 Group by architecture:
eg.: Mfx_Add8.c
• 2.5 Group by other methods: (individual grouping allowed)
Option 3 : <Name> can be removed so that single C file shall contain all MFX functions,
eg.: Mfx.c.
Using above options gives certain flexibility of choosing suitable granularity with re-
duced number of C files. Linking only on-demand is also possible in case of some
options.

12 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

6 Requirements Tracing
The following tables reference the requirements specified in [4], [5] and links to the
fulfillment of these. Please note that if column “Satisfied by” is empty for a specific
requirement this means that this requirement is not fulfilled by this document.
Requirement Description Satisfied by
[SRS_BSW_00003] All software modules shall provide [SWS_Mfx_00215]
version and identification information
[SRS_BSW_00007] All Basic SW Modules written in C [SWS_Mfx_00209]
language shall conform to the MISRA
C 2012 Standard.
[SRS_BSW_00304] All AUTOSAR Basic Software [SWS_Mfx_00212]
Modules shall use only AUTOSAR
data types instead of native C data
types
[SRS_BSW_00306] AUTOSAR Basic Software Modules [SWS_Mfx_00213]
shall be compiler and platform
independent
[SRS_BSW_00318] Each AUTOSAR Basic Software [SWS_Mfx_00215]
Module file shall provide version
numbers in the header file
[SRS_BSW_00321] The version numbers of AUTOSAR [SWS_Mfx_00215]
Basic Software Modules shall be
enumerated according specific rules
[SRS_BSW_00348] All AUTOSAR standard types and [SWS_Mfx_00211]
constants shall be placed and
organized in a standard type header
file
[SRS_BSW_00374] All Basic Software Modules shall [SWS_Mfx_00214]
provide a readable module vendor
identification
[SRS_BSW_00378] AUTOSAR shall provide a boolean [SWS_Mfx_00212]
type
[SRS_BSW_00379] All software modules shall provide a [SWS_Mfx_00214]
module identifier in the header file
and in the module XML description
file.
[SRS_BSW_00402] Each module shall provide version [SWS_Mfx_00214]
information
[SRS_BSW_00407] Each BSW module shall provide a [SWS_Mfx_00215] [SWS_Mfx_00216]
function to read out the version
information of a dedicated module
implementation
[SRS_BSW_00411] All AUTOSAR Basic Software [SWS_Mfx_00216]
Modules shall apply a naming rule for
enabling/disabling the existence of
the API
[SRS_BSW_00437] Memory mapping shall provide the [SWS_Mfx_00210]
possibility to define RAM segments
which are not to be initialized during
startup
[SRS_LIBS_00001] The functional behavior of each [SWS_Mfx_00218]
library functions shall not be
configurable
5

13 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Requirement Description Satisfied by
[SRS_LIBS_00002] A library shall be operational before [SWS_Mfx_00200]
all BSW modules and application
SW-Cs
[SRS_LIBS_00003] A library shall be operational until the [SWS_Mfx_00201]
shutdown
[SRS_LIBS_00004] Using libraries shall not pass through [SWS_Mfx_00203]
a port interface
[SRS_LIBS_00005] Each library shall provide one header [SWS_Mfx_00001] [SWS_Mfx_91001]
file with its public interface
[SRS_LIBS_00007] Using a library should be documented [SWS_Mfx_00205]
[SRS_LIBS_00009] All library functions shall be re-entrant [SWS_Mfx_91001]
[SRS_LIBS_00011] All function names and type names [SWS_Mfx_91001]
shall start with "Library short name_"
[SRS_LIBS_00015] It shall be possible to configure the [SWS_Mfx_00206]
microcontroller so that the library
code is shared between all callers
[SRS_LIBS_00017] Usage of macros should be avoided [SWS_Mfx_00207]
[SRS_LIBS_00018] A library function may only call library [SWS_Mfx_00208]
functions
Table 6.1: RequirementsTracing

14 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

7 Functional specification

7.1 Error classification

[SWS_Mfx_00227] dSection 7.1 "Error Handling" of the document "General Specifica-
tion of Basic Software Modules" describes the error handling of the Basic Software in
detail. Above all, it constitutes a classification scheme consisting of five error types
which may occur in BSW modules.c()
Based on this foundation, the following section specifies particular errors arranged in
the respective subsections below.

7.1.1 Development Errors

There are no development errors.

7.1.2 Runtime Errors

There are no runtime errors.

7.1.3 Transient Faults

There are no transient faults.

7.1.4 Production Error

There are no production errors.

7.1.5 Extended Production Errors

There are no extended production errors.

15 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

7.2 Initialization and shutdown

[SWS_Mfx_00200] dMFX library shall not require initialization phase. A Library func-
tion may be called at the very first step of ECU initialization, e.g. even by the OS or
EcuM, thus the library shall be ready.c(SRS_LIBS_00002)
[SWS_Mfx_00201] dMFX library shall not require a shutdown operation phase.c
(SRS_LIBS_00003)

7.3 Using Library API

[SWS_Mfx_00203] dMFX API can be directly called from BSW modules or SWC. No
port definition is required. It is a pure function call.c(SRS_LIBS_00004)
[SWS_Mfx_00205] dUsing a library should be documented. if a BSW module or a
SWC uses a Library, the developer should add an Implementation-DependencyOnAr-
tifact in the BSW/SWC template.
minVersion and maxVersion parameters correspond to the supplier version. In case
of AUTOSAR library, these parameters may be left empty because a SWC or BSW
module may rely on a library behaviour, not on a supplier implementation. However,
the SWC or BSW modules shall be compatible with the AUTOSAR platform where they
are integrated.c(SRS_LIBS_00007)

7.4 Library implementation

[SWS_Mfx_00206] dThe MFX library shall be implemented in a way that the code can
be shared among callers in different memory partitions.c(SRS_LIBS_00015)
[SWS_Mfx_00207] dUsage of macros should be avoided. The function should be de-
clared as function or inline function. Macro #define should not be used.c(SRS_LIBS_-
00017)
[SWS_Mfx_00208] dA library function shall not call any BSW modules functions, e.g.
the DET. A library function can call other library functions. Because a library function
shall be re-entrant. But other BSW modules functions may not be re-entrant.c(SRS_-
LIBS_00018)
[SWS_Mfx_00209] dThe library, written in C programming language, should conform
to the MISRA C Standard.
Please refer to SWS_BSW_00115 for more details.c(SRS_BSW_00007)
[SWS_Mfx_00210] dEach AUTOSAR library Module implementation <library>*.c and
<library>*.h shall map their code to memory sections using the AUTOSAR memory
mapping mechanism.c(SRS_BSW_00437)

16 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00211] dEach AUTOSAR library Module implementation <library>*.c, that

uses AUTOSAR integer data types and/or the standard return, shall include the header
file Std_Types.h.c(SRS_BSW_00348)
[SWS_Mfx_00212] dAll AUTOSAR library Modules should use the AUTOSAR data
types (integers, boolean) instead of native C data types, unless this library is clearly
identified to be compliant only with a platform.c(SRS_BSW_00378, SRS_BSW_00304)
[SWS_Mfx_00213] dAll AUTOSAR library Modules should avoid direct use of compiler
and platform specific keyword, unless this library is clearly identified to be compliant
only with a platform.c(SRS_BSW_00306)
[SWS_Mfx_00225] dIntegral promotion has to be adhered to when implementing Mfx
services. Thus, to obtain maximal precision, intermediate results shall not be limited.c
()

17 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8 API specification

8.1 Imported types

In this chapter, all types included from the following modules are listed:
Module Imported Type
Std_Types.h sint8, uint8, sint16, uint16, sint32, uint32

It is observed that since the sizes of the integer types provided by the C language are
implementation-defined, the range of values that may be represented within each of
the integer types will vary between implementations.
Thus, in order to improve the portability of the software, these types are defined in
Platform_Types.h [6].The following mnemonics are used in the library routine names.
Size Platform Type Mnemonic
signed 8-Bit sint8 s8
signed 16-Bit sint16 s16
signed 32-Bit sint32 s32
unsigned 8-Bit uint8 u8
unsigned 16-Bit uint16 u16
unsigned 32-Bit uint32 u32

Table 8.1: Base Types

As described in [6], the ranges for each of the base types are shown in Table 2.
Base Type Range
uint8 [ 0, 255 ]
sint8 [ -128, 127 ]
uint16 [ 0, 65535 ]
sint16 [ -32768, 32767 ]
uint32 [ 0, 4294967295 ]
sint32 [ -2147483648, 2147483647 ]

Table 8.2: Ranges for Base Types

As a convention in the rest of the document:

• mnemonics will be used in the name of the routines (using <InTypeMn1> that
means Type Mnemonic for Input 1)
• the real type will be used in the description of the prototypes of the routines (using
<InType1> or <OutType>).

18 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.2 Type definitions

None.

8.3 Comment about rounding

Two types of rounding can be applied:
Results are "rounded off", it means:
• 0 <= X < 0.5 rounded to 0
• 0.5 <= X < 1 rounded to 1
• -0.5 < X <= 0 rounded to 0
• -1 < X <= -0.5 rounded to -1
Results are rounded towards zero:
• 0 <= X < 1 rounded to 0
• -1 < X <= 0 rounded to 0

8.4 Comment about routines optimization

8.4.1 Optimized with constants

For optimization purpose, in some routines, it is mandatory that an argument of the

function "must be constant".
The requirement is that the expression must be fully evaluated at compile time. It may
be a constant literal, macro, or arithmetic expression that can be computed at compile
time. It may not contain a variable or function call.
For example, the parameters for the radix points are constant expressions so that they
may be eliminated after the pre-process phase of compilation. When implemented
properly as an inline function or macro, the calculations for the number of shifts nec-
essary are done at compile time, not at run time. There is a ROM/throughput penalty
when constant expressions are not used.

19 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5 Mathematical routines definitions

8.5.1 Additions

[SWS_Mfx_00002] Definition of API function Mfx_Add_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_Add_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_Add_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value
)
Service ID [hex] 0x001 to 0x024
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description This routine makes an addition between the two arguments.
Available via Mfx.h

c()
[SWS_Mfx_00006] dThis routine makes an addition between the two arguments:
Return-value = x_value + y_valuec()
[SWS_Mfx_00007] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00008] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x001 uint8 Mfx_Add_u8u8_u8( uint8 , uint8);
0x002 uint8 Mfx_Add_u8s8_u8( uint8 , sint8);
0x003 sint8 Mfx_Add_u8s8_s8( uint8 , sint8);
0x004 uint8 Mfx_Add_s8s8_u8( sint8 , sint8);
0x005 sint8 Mfx_Add_s8s8_s8( sint8 , sint8);
0x006 uint16 Mfx_Add_u16u16_u16( uint16 , uint16);
0x007 uint16 Mfx_Add_u16s16_u16( uint16 , sint16);
0x008 sint16 Mfx_Add_u16s16_s16( uint16 , sint16);
0x009 uint8 Mfx_Add_s16s16_u8( sint16 , sint16);
0x00A sint8 Mfx_Add_s16s16_s8( sint16 , sint16);
0x00B uint16 Mfx_Add_s16s16_u16( sint16 , sint16);
0x00C sint16 Mfx_Add_s16s16_s16( sint16 , sint16);
0x00D sint8 Mfx_Add_u32u32_s8( uint32 , uint32);
0x00E sint16 Mfx_Add_u32u32_s16( uint32 , uint32);
5

20 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype
0x00F uint32 Mfx_Add_u32u32_u32( uint32 , uint32);
0x010 sint32 Mfx_Add_u32u32_s32( uint32 , uint32);
0x011 uint32 Mfx_Add_u32s32_u32( uint32 , sint32);
0x012 sint32 Mfx_Add_u32s32_s32( uint32 , sint32);
0x013 uint32 Mfx_Add_s32s32_u32( sint32 , sint32);
0x014 sint32 Mfx_Add_s32s32_s32( sint32 , sint32);
0x015 uint8 Mfx_Add_s32s32_u8( sint32 , sint32);
0x016 sint8 Mfx_Add_s32s32_s8( sint32 , sint32);
0x017 uint16 Mfx_Add_s32s32_u16( sint32 , sint32);
0x018 sint16 Mfx_Add_s32s32_s16( sint32 , sint32);
0x019 sint16 Mfx_Add_u32s32_s16( uint32 , sint32);
0x01A sint8 Mfx_Add_u32s32_s8( uint32 , sint32);
0x01B uint16 Mfx_Add_u32s32_u16( uint32 , sint32);
0x01C uint8 Mfx_Add_u32s32_u8( uint32 , sint32);
0x01D uint16 Mfx_Add_u32u32_u16( uint32 , uint32);
0x01E uint8 Mfx_Add_u32u32_u8( uint32 , uint32);
0x01F sint16 Mfx_Add_u16u16_s16( uint16 , uint16);
0x020 uint8 Mfx_Add_u16u16_u8( uint16 , uint16);
0x021 uint8 Mfx_Add_u16s16_u8( uint16 , sint16);
0x022 sint8 Mfx_Add_u16u16_s8( uint16 , uint16);
0x023 sint8 Mfx_Add_u16s16_s8( uint16 , sint16);
0x024 sint8 Mfx_Add_u8u8_s8( uint8 , uint8);

8.5.2 Subtractions

[SWS_Mfx_00009] Definition of API function Mfx_Sub_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_Sub_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_Sub_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value
)
Service ID [hex] 0x025 to 0x054
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description This routine makes a subtraction between the two arguments.
Available via Mfx.h

c()

21 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00010] dThis routine makes a subtraction between the two arguments:

Return-value = x_value - y_valuec()
[SWS_Mfx_00011] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00012] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x025 uint8 Mfx_Sub_u8u8_u8( uint8 , uint8);
0x026 sint8 Mfx_Sub_u8u8_s8( uint8 , uint8);
0x027 uint8 Mfx_Sub_u8s8_u8( uint8 , sint8);
0x028 sint8 Mfx_Sub_s8u8_s8( sint8 , uint8);
0x029 sint8 Mfx_Sub_s8s8_s8( sint8 , sint8);
0x02A uint8 Mfx_Sub_u16u16_u8( uint16 , uint16);
0x02B sint8 Mfx_Sub_u16u16_s8( uint16 , uint16);
0x02C uint8 Mfx_Sub_s16s16_u8( sint16 , sint16);
0x02D sint8 Mfx_Sub_s16s16_s8( sint16 , sint16);
0x02E uint8 Mfx_Sub_s32s32_u8( sint32 , sint32);
0x02F sint8 Mfx_Sub_s32s32_s8( sint32 , sint32);
0x030 uint16 Mfx_Sub_u16u16_u16( uint16 , uint16);
0x031 uint16 Mfx_Sub_u16s16_u16( uint16 , sint16);
0x032 sint16 Mfx_Sub_s16u16_s16( sint16 , uint16);
0x033 sint16 Mfx_Sub_u16s16_s16( uint16 , sint16);
0x034 uint16 Mfx_Sub_s16s16_u16( sint16 , sint16);
0x035 sint16 Mfx_Sub_u16u16_s16( uint16 , uint16);
0x036 sint16 Mfx_Sub_s16s16_s16( sint16 , sint16);
0x037 uint8 Mfx_Sub_s32u32_u8( sint32 , uint32);
0x038 sint8 Mfx_Sub_u32s32_s8( uint32 , sint32);
0x039 uint16 Mfx_Sub_s32u32_u16( sint32 , uint32);
0x03A uint16 Mfx_Sub_u32u32_u16( uint32 , uint32);
0x03B sint16 Mfx_Sub_u32u32_s16( uint32 , uint32);
0x03C uint16 Mfx_Sub_s32s32_u16( sint32 , sint32);
0x03D sint16 Mfx_Sub_s32s32_s16( sint32 , sint32);
0x03E uint32 Mfx_Sub_u32u32_u32( uint32 , uint32);
0x03F uint32 Mfx_Sub_u32s32_u32( uint32 , sint32);
0x040 uint32 Mfx_Sub_s32u32_u32( sint32 , uint32);
0x041 sint32 Mfx_Sub_u32u32_s32( uint32 , uint32);
0x042 sint32 Mfx_Sub_s32u32_s32( sint32 , uint32);
0x043 sint32 Mfx_Sub_u32s32_s32( uint32 , sint32);
0x044 uint32 Mfx_Sub_s32s32_u32( sint32 , sint32);
0x045 sint32 Mfx_Sub_s32s32_s32( sint32 , sint32);
0x046 sint16 Mfx_Sub_s32u32_s16( sint32 , uint32);
0x047 sint8 Mfx_Sub_s32u32_s8( sint32 , uint32);
0x048 sint16 Mfx_Sub_u32s32_s16( uint32 , sint32);
0x049 uint16 Mfx_Sub_u32s32_u16( uint32 , sint32);
0x04A uint8 Mfx_Sub_u32s32_u8( uint32 , sint32);
5

22 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype
0x04B sint8 Mfx_Sub_u32u32_s8( uint32 , uint32);
0x04C uint8 Mfx_Sub_u32u32_u8( uint32 , uint32);
0x04D uint16 Mfx_Sub_s16u16_u16( sint16 , uint16);
0x04E uint8 Mfx_Sub_u16s16_u8( uint16 , sint16);
0x04F uint8 Mfx_Sub_s16u16_u8( sint16 , uint16);
0x050 sint8 Mfx_Sub_u16s16_s8( uint16 , sint16);
0x051 sint8 Mfx_Sub_s16u16_s8( sint16 , uint16);
0x052 uint8 Mfx_Sub_s8u8_u8( sint8 , uint8);
0x053 uint8 Mfx_Sub_s8s8_u8( sint8 , sint8);
0x054 sint8 Mfx_Sub_u8s8_s8( uint8 , sint8);

8.5.3 Absolute value

[SWS_Mfx_00013] Definition of API function Mfx_Abs_<InTypeMn1>_<OutType

Mn> d
Service Name Mfx_Abs_<InTypeMn1>_<OutTypeMn>
Syntax <OutType> Mfx_Abs_<InTypeMn1>_<OutTypeMn> (
<InType1> x_value
)
Service ID [hex] 0x055 to 0x05E
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description This routine computes the absolute value of a signed value.
Available via Mfx.h

c()
[SWS_Mfx_00014] dThis routine computes the absolute value of a signed value:
Return-value = | x_value |c()
[SWS_Mfx_00015] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00016] dHere is the list of implemented functions.c()

23 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

Function ID[hex] Function prototype

0x055 uint8 Mfx_Abs_s8_u8( sint8 );
0x056 sint8 Mfx_Abs_s8_s8( sint8 );
0x057 uint8 Mfx_Abs_s32_u8( sint32 );
0x058 uint16 Mfx_Abs_s16_u16( sint16 );
0x059 sint16 Mfx_Abs_s16_s16( sint16 );
0x05A sint16 Mfx_Abs_s32_s16( sint32 );
0x05B uint32 Mfx_Abs_s32_u32( sint32 );
0x05C sint32 Mfx_Abs_s32_s32( sint32 );
0x05D sint8 Mfx_Abs_s32_s8( sint32 );
0x05E uint16 Mfx_Abs_s32_u16( sint32 );

8.5.4 Absolute value of a difference

[SWS_Mfx_00017] Definition of API function Mfx_AbsDiff_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_AbsDiff_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_AbsDiff_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value
)
Service ID [hex] 0x05F to 0x082
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description This routine computes the absolute value of a difference between 2 values.
Available via Mfx.h

c()
[SWS_Mfx_00018] dThis routine computes the absolute value of a difference between
2 values:
Return-value = | x_value - y_value |c()
[SWS_Mfx_00019] dReturn-value shall be saturated to boundary values in the event
of overflow.c()
[SWS_Mfx_00020] dHere is the list of implemented functions.c()

24 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

Function ID[hex] Function prototype

0x05F uint8 Mfx_AbsDiff_u8u8_u8( uint8 , uint8);
0x060 uint16 Mfx_AbsDiff_u16u16_u16( uint16 , uint16);
0x061 uint8 Mfx_AbsDiff_s16s16_u8( sint16 , sint16);
0x062 uint16 Mfx_AbsDiff_s16s16_u16( sint16 , sint16);
0x063 uint8 Mfx_AbsDiff_u32s32_u8( uint32 , sint32);
0x064 uint16 Mfx_AbsDiff_u32s32_u16( uint32 , sint32);
0x065 uint32 Mfx_AbsDiff_u32s32_u32( uint32 , sint32);
0x066 uint32 Mfx_AbsDiff_u32u32_u32( uint32 , uint32);
0x067 uint8 Mfx_AbsDiff_s32s32_u8( sint32 , sint32);
0x068 sint16 Mfx_AbsDiff_s32s32_s16( sint32 , sint32);
0x069 sint32 Mfx_AbsDiff_s32s32_s32( sint32 , sint32);
0x06A sint8 Mfx_AbsDiff_s32s32_s8( sint32 , sint32);
0x06B uint16 Mfx_AbsDiff_s32s32_u16( sint32 , sint32);
0x06C uint32 Mfx_AbsDiff_s32s32_u32( sint32 , sint32);
0x06D uint16 Mfx_AbsDiff_u32u32_u16( uint32 , uint32);
0x06E uint8 Mfx_AbsDiff_u32u32_u8( uint32 , uint32);
0x06F sint8 Mfx_Absdiff_u32u32_s8( uint32 , uint32);
0x070 sint16 Mfx_Absdiff_u32u32_s16( uint32 , uint32);
0x071 sint32 Mfx_Absdiff_u32u32_s32( uint32 , uint32);
0x072 sint8 Mfx_Absdiff_u32s32_s8( uint32 , sint32);
0x073 sint16 Mfx_Absdiff_u32s32_s16( uint32 , sint32);
0x074 sint32 Mfx_Absdiff_u32s32_s32( uint32 , sint32);
0x075 uint16 Mfx_AbsDiff_u16s16_u16( uint16 , sint16);
0x076 sint16 Mfx_AbsDiff_u16u16_s16( uint16 , uint16);
0x077 sint16 Mfx_AbsDiff_u16s16_s16( uint16 , sint16);
0x078 sint16 Mfx_AbsDiff_s16s16_s16( sint16 , sint16);
0x079 uint8 Mfx_AbsDiff_u16u16_u8( uint16 , uint16);
0x07A uint8 Mfx_AbsDiff_u16s16_u8( uint16 , sint16);
0x07B sint8 Mfx_AbsDiff_u16u16_s8( uint16 , uint16);
0x07C sint8 Mfx_AbsDiff_u16s16_s8( uint16 , sint16);
0x07D sint8 Mfx_AbsDiff_s16s16_s8( sint16 , sint16);
0x07E uint8 Mfx_AbsDiff_u8s8_u8( uint8 , sint8);
0x07F uint8 Mfx_AbsDiff_s8s8_u8( sint8 , sint8);
0x080 sint8 Mfx_AbsDiff_u8u8_s8( uint8 , uint8);
0x081 sint8 Mfx_AbsDiff_u8s8_s8( uint8 , sint8);
0x082 sint8 Mfx_AbsDiff_s8s8_s8( sint8 , sint8);

25 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.5 Multiplications

[SWS_Mfx_00021] Definition of API function Mfx_Mul_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_Mul_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_Mul_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value
)
Service ID [hex] 0x083 to 0x0A7
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description This routine makes a multiplication between the two arguments.
Available via Mfx.h

c()
[SWS_Mfx_00022] dThis routine makes a multiplication between the two arguments:
Return-value = x_value * y_valuec()
[SWS_Mfx_00023] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()

26 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00024] dHere is the list of implemented functions.c()

Function ID[hex] Function prototype
0x083 uint8 Mfx_Mul_u8u8_u8( uint8 , uint8);
0x084 uint8 Mfx_Mul_s8s8_u8( sint8 , sint8);
0x085 sint8 Mfx_Mul_s8s8_s8( sint8 , sint8);
0x086 uint16 Mfx_Mul_u16u16_u16( uint16 , uint16);
0x087 uint16 Mfx_Mul_s16s16_u16( sint16 , sint16);
0x088 uint8 Mfx_Mul_s16s16_u8( sint16 , sint16);
0x089 sint8 Mfx_Mul_s16s16_s8( sint16 , sint16);
0x08A sint16 Mfx_Mul_s16s16_s16( sint16 , sint16);
0x08B uint32 Mfx_Mul_u32u32_u32( uint32 , uint32);
0x08C sint32 Mfx_Mul_u32u32_s32( uint32 , uint32);
0x08D uint32 Mfx_Mul_s32s32_u32( sint32 , sint32);
0x08E uint8 Mfx_Mul_s32s32_u8( sint32 , sint32);
0x08F sint8 Mfx_Mul_u32u32_s8( uint32 , uint32);
0x090 sint8 Mfx_Mul_s32s32_s8( sint32 , sint32);
0x091 sint16 Mfx_Mul_u32u32_s16( uint32 , uint32);
0x092 sint16 Mfx_Mul_s32s32_s16( sint32 , sint32);
0x093 uint16 Mfx_Mul_s32s32_u16( sint32 , sint32);
0x094 sint32 Mfx_Mul_s32s32_s32( sint32 , sint32);
0x095 sint16 Mfx_Mul_u32s32_s16( uint32 , sint32);
0x096 sint8 Mfx_Mul_u32s32_s8( uint32 , sint32);
0x097 uint8 Mfx_Mul_u32s32_u8( uint32 , sint32);
0x098 uint16 Mfx_Mul_u32u32_u16( uint32 , uint32);
0x099 uint8 Mfx_Mul_u32u32_u8( uint32 , uint32);
0x09A uint8 Mfx_Mul_u8s8_u8( uint8 , sint8);
0x09B sint8 Mfx_Mul_u8s8_s8( uint8 , sint8);
0x09C uint16 Mfx_Mul_u16s16_u16( uint16 , sint16);
0x09D sint16 Mfx_Mul_u16s16_s16( uint16 , sint16);
0x09E sint32 Mfx_Mul_u16s16_s32( uint16 , sint16);
0x09F uint16 Mfx_Mul_u32s32_u16( uint32 , sint32);
0x0A0 uint32 Mfx_Mul_u32s32_u32( uint32 , sint32);
0x0A1 sint32 Mfx_Mul_u32s32_s32( uint32 , sint32);
0x0A2 sint16 Mfx_Mul_u16u16_s16( uint16 , uint16);
0x0A3 uint8 Mfx_Mul_u16u16_u8( uint16 , uint16);
0x0A4 uint8 Mfx_Mul_u16s16_u8( uint16 , sint16);
0x0A5 sint8 Mfx_Mul_u16u16_s8( uint16 , uint16);
0x0A6 sint8 Mfx_Mul_u16s16_s8( uint16 , sint16);
0x0A7 sint8 Mfx_Mul_u8u8_s8( uint8 , uint8);

27 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.6 Divisions rounded towards 0

[SWS_Mfx_00025] Definition of API function Mfx_Div_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_Div_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_Div_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value
)
Service ID [hex] 0x0A8 to 0x0D7
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description These routines make a division between the two arguments.
Available via Mfx.h

c()
[SWS_Mfx_00026] dThese routines make a division between the two arguments:
Return-value = x_value / y_valuec()
[SWS_Mfx_00027] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00028] dThe result after division by zero is defined by:
* If x_value ≥ 0 then the function returns the maximum value of the output type
* If x_value < 0 then the function returns the minimum value of the output typec()
[SWS_Mfx_00030] dThe result is rounded towards 0.c()
[SWS_Mfx_00031] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x0A8 uint8 Mfx_Div_u8u8_u8( uint8 , uint8);
0x0A9 uint8 Mfx_Div_s8u8_u8( sint8 , uint8);
0x0AA uint8 Mfx_Div_u8s8_u8( uint8 , sint8);
0x0AB uint8 Mfx_Div_s8s8_u8( sint8 , sint8);
0x0AC sint8 Mfx_Div_u8s8_s8( uint8 , sint8);
0x0AD sint8 Mfx_Div_s8u8_s8( sint8 , uint8);
0x0AE sint8 Mfx_Div_s8s8_s8( sint8 , sint8);
0x0AF uint16 Mfx_Div_u16u16_u16( uint16 , uint16);
0x0B0 uint16 Mfx_Div_s16u16_u16( sint16 , uint16);
5

28 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype
0x0B1 uint16 Mfx_Div_u16s16_u16( uint16 , sint16);
0x0B2 sint16 Mfx_Div_u16s16_s16( uint16 , sint16);
0x0B3 sint16 Mfx_Div_s16u16_s16( sint16 , uint16);
0x0B4 uint16 Mfx_Div_s16s16_u16( sint16 , sint16);
0x0B5 uint8 Mfx_Div_s16s16_u8( sint16 , sint16);
0x0B6 sint8 Mfx_Div_s16s16_s8( sint16 , sint16);
0x0B7 sint16 Mfx_Div_s16s16_s16( sint16 , sint16);
0x0B8 sint16 Mfx_Div_s32u32_s16( sint32 , uint32);
0x0B9 uint32 Mfx_Div_u32u32_u32( uint32 , uint32);
0x0BA uint32 Mfx_Div_s32u32_u32( sint32 , uint32);
0x0BB uint32 Mfx_Div_u32s32_u32( uint32 , sint32);
0x0BC sint32 Mfx_Div_u32s32_s32( uint32 , sint32);
0x0BD sint32 Mfx_Div_s32u32_s32( sint32 , uint32);
0x0BE uint32 Mfx_Div_s32s32_u32( sint32 , sint32);
0x0BF uint8 Mfx_Div_s32s32_u8( sint32 , sint32);
0x0C0 sint8 Mfx_Div_s32s32_s8( sint32 , sint32);
0x0C1 uint16 Mfx_Div_s32s32_u16( sint32 , sint32);
0x0C2 sint16 Mfx_Div_s32s32_s16( sint32 , sint32);
0x0C3 sint32 Mfx_Div_s32s32_s32( sint32 , sint32);
0x0C4 sint8 Mfx_Div_u32u32_s8( uint32 , uint32);
0x0C5 sint16 Mfx_Div_u32u32_s16( uint32 , uint32);
0x0C6 sint32 Mfx_Div_u32u32_s32( uint32 , uint32);
0x0C7 sint8 Mfx_Div_s32u32_s8( sint32 , uint32);
0x0C8 uint16 Mfx_Div_s32u32_u16( sint32 , uint32);
0x0C9 uint8 Mfx_Div_s32u32_u8( sint32 , uint32);
0x0CA sint16 Mfx_Div_u32s32_s16( uint32 , sint32);
0x0CB sint8 Mfx_Div_u32s32_s8( uint32 , sint32);
0x0CC uint16 Mfx_Div_u32s32_u16( uint32 , sint32);
0x0CD uint8 Mfx_Div_u32s32_u8( uint32 , sint32);
0x0CE uint16 Mfx_Div_u32u32_u16( uint32 , uint32);
0x0CF uint8 Mfx_Div_u32u32_u8( uint32 , uint32);
0x0D0 sint16 Mfx_Div_u16u16_s16( uint16 , uint16);
0x0D1 uint8 Mfx_Div_u16u16_u8( uint16 , uint16);
0x0D2 uint8 Mfx_Div_u16s16_u8( uint16 , sint16);
0x0D3 uint8 Mfx_Div_s16u16_u8( sint16 , uint16);
0x0D4 sint8 Mfx_Div_u16u16_s8( uint16 , uint16);
0x0D5 sint8 Mfx_Div_u16s16_s8( uint16 , sint16);
0x0D6 sint8 Mfx_Div_s16u16_s8( sint16 , uint16);
0x0D7 sint8 Mfx_Div_u8u8_s8( uint8 , uint8);

29 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.7 Divisions rounded off

[SWS_Mfx_00032] Definition of API function Mfx_RDiv_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_RDiv_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_RDiv_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value
)
Service ID [hex] 0x0D8 to 0x107
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description These routines make a division between the two arguments.
Available via Mfx.h

c()
[SWS_Mfx_00033] dThese routines make a division between the two arguments:
Return-value = x_value / y_valuec()
[SWS_Mfx_00034] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00035] dThe result after division by zero is defined by:
* If x_value ≥ 0 then the function returns the maximum value of the output type
* If x_value < 0 then the function returns the minimum value of the output typec()
[SWS_Mfx_00037] dThe result is rounded off.c()
[SWS_Mfx_00038] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype : RDiv
0x0D8 uint8 Mfx_RDiv_u8u8_u8( uint8 , uint8);
0x0D9 uint8 Mfx_RDiv_s8u8_u8( sint8 , uint8);
0x0DA uint8 Mfx_RDiv_u8s8_u8( uint8 , sint8);
0x0DB uint8 Mfx_RDiv_s8s8_u8( sint8 , sint8);
0x0DC sint8 Mfx_RDiv_u8s8_s8( uint8 , sint8);
0x0DD sint8 Mfx_RDiv_s8u8_s8( sint8 , uint8);
0x0DE sint8 Mfx_RDiv_s8s8_s8( sint8 , sint8);
0x0DF uint16 Mfx_RDiv_u16u16_u16( uint16 , uint16);
0x0E0 uint16 Mfx_RDiv_s16u16_u16( sint16 , uint16);
0x0E1 uint16 Mfx_RDiv_u16s16_u16( uint16 , sint16);
5

30 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype : RDiv
0x0E2 sint16 Mfx_RDiv_u16s16_s16( uint16 , sint16);
0x0E3 sint16 Mfx_RDiv_s16u16_s16( sint16 , uint16);
0x0E4 uint16 Mfx_RDiv_s16s16_u16( sint16 , sint16);
0x0E5 uint8 Mfx_RDiv_s16s16_u8( sint16 , sint16);
0x0E6 sint8 Mfx_RDiv_s16s16_s8( sint16 , sint16);
0x0E7 sint16 Mfx_RDiv_s16s16_s16( sint16 , sint16);
0x0E8 sint16 Mfx_RDiv_s32u32_s16( sint32 , uint32);
0x0E9 uint32 Mfx_RDiv_u32u32_u32( uint32 , uint32);
0x0EA uint32 Mfx_RDiv_s32u32_u32( sint32 , uint32);
0x0EB uint32 Mfx_RDiv_u32s32_u32( uint32 , sint32);
0x0EC sint32 Mfx_RDiv_u32s32_s32( uint32 , sint32);
0x0ED sint32 Mfx_RDiv_s32u32_s32( sint32 , uint32);
0x0EE uint32 Mfx_RDiv_s32s32_u32( sint32 , sint32);
0x0EF uint8 Mfx_RDiv_s32s32_u8( sint32 , sint32);
0x0F0 sint8 Mfx_RDiv_s32s32_s8( sint32 , sint32);
0x0F1 uint16 Mfx_RDiv_s32s32_u16( sint32 , sint32);
0x0F2 sint16 Mfx_RDiv_s32s32_s16( sint32 , sint32);
0x0F3 sint32 Mfx_RDiv_s32s32_s32( sint32 , sint32);
0x0F4 sint8 Mfx_RDiv_u32u32_s8( uint32 , uint32);
0x0F5 sint16 Mfx_RDiv_u32u32_s16( uint32 , uint32);
0x0F6 sint32 Mfx_RDiv_u32u32_s32( uint32 , uint32);
0x0F7 sint8 Mfx_RDiv_s32u32_s8( sint32 , uint32);
0x0F8 uint16 Mfx_RDiv_s32u32_u16( sint32 , uint32);
0x0F9 uint8 Mfx_RDiv_s32u32_u8( sint32 , uint32);
0x0FA sint16 Mfx_RDiv_u32s32_s16( uint32 , sint32);
0xoFB sint8 Mfx_RDiv_u32s32_s8( uint32 , sint32);
0x0FC uint16 Mfx_RDiv_u32s32_u16( uint32 , sint32);
0x0FD uint8 Mfx_RDiv_u32s32_u8( uint32 , sint32);
0x0FE uint16 Mfx_RDiv_u32u32_u16( uint32 , uint32);
0x0FF uint8 Mfx_RDiv_u32u32_u8( uint32 , uint32);
0x100 sint16 Mfx_RDiv_u16u16_s16( uint16 , uint16);
0x101 uint8 Mfx_RDiv_u16u16_u8( uint16 , uint16);
0x102 uint8 Mfx_RDiv_u16s16_u8( uint16 , sint16);
0x103 uint8 Mfx_RDiv_s16u16_u8( sint16 , uint16);
0x104 sint8 Mfx_RDiv_u16u16_s8( uint16 , uint16);
0x105 sint8 Mfx_RDiv_u16s16_s8( uint16 , sint16);
0x106 sint8 Mfx_RDiv_s16u16_s8( sint16 , uint16);
0x107 sint8 Mfx_RDiv_u8u8_s8( uint8 , uint8);

31 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.8 Combinations of multiplication and division rounded towards 0

[SWS_Mfx_00039] Definition of API function Mfx_MulDiv_<InTypeMn1><InType

Mn2><InTypeMn3>_<OutTypeMn> d
Service Name Mfx_MulDiv_<InTypeMn1><InTypeMn2><InTypeMn3>_<OutTypeMn>
Syntax <OutType> Mfx_MulDiv_<InTypeMn1><InTypeMn2><InTypeMn3>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value,
<InType3> z_value
)
Service ID [hex] 0x108 to 0x121
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
z_value Third argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description These routines make a multiplication between the two arguments and a division by the third
argument.
Available via Mfx.h

c()
[SWS_Mfx_00040] dThese routines make a multiplication between the two arguments
and a division by the third argument:
Return-value = x_value * y_value / z_valuec()
[SWS_Mfx_00041] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00042] dThe result after division by zero is defined by:
* If x_value*y_value ≥ 0 then the function returns the maximum value of the output
type
* If x_value*y_value < 0 then the function returns the minimum value of the output typec
()
[SWS_Mfx_00044] dThe result is rounded towards 0.c()
[SWS_Mfx_00045] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype : Div
0x108 uint16 Mfx_MulDiv_s32s32s32_u16( sint32 , sint32 , sint32);
0x109 sint16 Mfx_MulDiv_s32s32s32_s16( sint32 , sint32 , sint32);
0x10A uint16 Mfx_MulDiv_u32u32u16_u16( uint32 , uint32 ,
uint16);
5

32 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype : Div
0x10B sint16 Mfx_MulDiv_s32s32s16_s16( sint32 , sint32 , sint16);
0x10C uint16 Mfx_MulDiv_s16u16s16_u16( sint16 , uint16 , sint16);
0x10D uint16 Mfx_MulDiv_s16u16u16_u16( sint16 , uint16 , uint16);
0x10E uint16 Mfx_MulDiv_u16u16u16_u16( uint16 , uint16 ,
uint16);
0x10F sint16 Mfx_MulDiv_s16u16s16_s16( sint16 , uint16 , sint16);
0x110 sint16 Mfx_MulDiv_s16s16u16_s16( sint16 , sint16 , uint16);
0x111 sint16 Mfx_MulDiv_s16u16u16_s16( sint16 , uint16 , uint16);
0x112 sint16 Mfx_MulDiv_s16s16s16_s16( sint16 , sint16 , sint16);
0x113 uint32 Mfx_MulDiv_u32u32u32_u32( uint32 , uint32 ,
uint32);
0x114 uint32 Mfx_MulDiv_u32u32s32_u32( uint32 , uint32 , sint32);
0x115 uint32 Mfx_MulDiv_u32s32u32_u32( uint32 , sint32 , uint32);
0x116 uint32 Mfx_MulDiv_u32s32s32_u32( uint32 , sint32 , sint32);
0x117 sint32 Mfx_MulDiv_s32s32u32_s32( sint32 , sint32 , uint32);
0x118 sint32 Mfx_MulDiv_s32u32s32_s32( sint32 , uint32 , sint32);
0x119 sint32 Mfx_MulDiv_s32u32u32_s32( sint32 , uint32 , uint32);
0x11A sint32 Mfx_MulDiv_s32s32s32_s32( sint32 , sint32 , sint32);
0x11B uint16 Mfx_MulDiv_u32u32u32_u16( uint32 , uint32 ,
uint32);
0x11C uint16 Mfx_MulDiv_u16s16s16_u16( uint16 , sint16 , sint16);
0x11D uint16 Mfx_MulDiv_u16s16u16_u16( uint16 , sint16 , uint16);
0x11E sint16 Mfx_MulDiv_u16s16s16_s16( uint16 , sint16 , sint16);
0x11F sint16 Mfx_MulDiv_u16s16u16_s16( uint16 , sint16 , uint16);
0x120 sint32 Mfx_MulDiv_u32s32s32_s32( uint32 , sint32 , sint32);
0x121 sint32 Mfx_MulDiv_u32s32u32_s32( uint32 , sint32 , uint32);

Note : The redundancy due to commutativity will be reduced in the next version

33 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.9 Combinations of multiplication and division rounded off

[SWS_Mfx_00046] Definition of API function Mfx_RMulDiv_<InTypeMn1><InType

Mn2><InTypeMn3>_<OutTypeMn> d
Service Name Mfx_RMulDiv_<InTypeMn1><InTypeMn2><InTypeMn3>_<OutTypeMn>
Syntax <OutType> Mfx_RMulDiv_<InTypeMn1><InTypeMn2><InTypeMn3>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value,
<InType3> z_value
)
Service ID [hex] 0x122 to 0x13B
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
z_value Third argument
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description These routines make a multiplication between the two arguments and a division by the third
argument.
Available via Mfx.h

c()
[SWS_Mfx_00047] dThese routines make a multiplication between the two arguments
and a division by the third argument:
Return-value = x_value * y_value / z_valuec()
[SWS_Mfx_00048] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00049] dThe result after division by zero is defined by:
* If x_value*y_value ≥ 0 then the function returns the maximum value of the output
type
* If x_value*y_value < 0 then the function returns the minimum value of the output typec
()
[SWS_Mfx_00051] dThe result is rounded off.c()
[SWS_Mfx_00052] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype : RDiv
0x122 uint16 Mfx_RMulDiv_s32s32s32_u16( sint32 , sint32 ,
sint32);
0x123 sint16 Mfx_RMulDiv_s32s32s32_s16( sint32 , sint32 ,
sint32);
5

34 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype : RDiv
0x124 uint16 Mfx_RMulDiv_u32u32u16_u16( uint32 , uint32 ,
uint16);
0x125 sint16 Mfx_RMulDiv_s32s32s16_s16( sint32 , sint32 ,
sint16);
0x126 uint16 Mfx_RMulDiv_s16u16s16_u16( sint16 , uint16 ,
sint16);
0x127 uint16 Mfx_RMulDiv_s16u16u16_u16( sint16 , uint16 ,
uint16);
0x128 uint16 Mfx_RMulDiv_u16u16u16_u16( uint16 , uint16 ,
uint16);
0x129 sint16 Mfx_RMulDiv_s16u16s16_s16( sint16 , uint16 ,
sint16);
0x12A sint16 Mfx_RMulDiv_s16s16u16_s16( sint16 , sint16 ,
uint16);
0x12B sint16 Mfx_RMulDiv_s16u16u16_s16( sint16 , uint16 ,
uint16);
0x12C sint16 Mfx_RMulDiv_s16s16s16_s16( sint16 , sint16 ,
sint16);
0x12D uint32 Mfx_RMulDiv_u32u32u32_u32( uint32 , uint32 ,
uint32);
0x12E uint32 Mfx_RMulDiv_u32u32s32_u32( uint32 , uint32 ,
sint32);
0x12F uint32 Mfx_RMulDiv_u32s32u32_u32( uint32 , sint32 ,
uint32);
0x130 uint32 Mfx_RMulDiv_u32s32s32_u32( uint32 , sint32 ,
sint32);
0x131 sint32 Mfx_RMulDiv_s32s32u32_s32( sint32 , sint32 ,
uint32);
0x132 sint32 Mfx_RMulDiv_s32u32s32_s32( sint32 , uint32 ,
sint32);
0x133 sint32 Mfx_RMulDiv_s32u32u32_s32( sint32 , uint32 ,
uint32);
0x134 sint32 Mfx_RMulDiv_s32s32s32_s32( sint32 , sint32 ,
sint32);
0x135 uint16 Mfx_RMulDiv_u32u32u32_u16( uint32 , uint32 ,
uint32);
0x136 uint16 Mfx_RMulDiv_u16s16s16_u16( uint16 , sint16 ,
sint16);
0x137 uint16 Mfx_RMulDiv_u16s16u16_u16( uint16 , sint16 ,
uint16);
0x138 sint16 Mfx_RMulDiv_u16s16s16_s16( uint16 , sint16 ,
sint16);
0x139 sint16 Mfx_RMulDiv_u16s16u16_s16( uint16 , sint16 ,
uint16);
0x13A sint32 Mfx_RMulDiv_u32s32s32_s32( uint32 , sint32 ,
sint32);
0x13B sint32 Mfx_RMulDiv_u32s32u32_s32( uint32 , sint32 ,
uint32);

Note : The redundancy due to commutativity will be reduced in the next version

35 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.10 Combinations of multiplication and shift right

[SWS_Mfx_00053] Definition of API function Mfx_MulShRight_<InTypeMn1><In

TypeMn2>_<OutTypeMn> d
Service Name Mfx_MulShRight_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_MulShRight_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value,
uint8 shift
)
Service ID [hex] 0x13C to 0x151
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First factor
y_value Second factor
shift Shift left of the fixed point result. Must be a constant expression.
Maximum shift according to SWS_Mfx_00057
Parameters (inout) None
Parameters (out) None
Return value <OutType> Result of the calculation
Description This routine makes a multiplication between the two arguments and applies a shift right defined
by the third argument.
Available via Mfx.h

c()
[SWS_Mfx_00054] dThis routine makes a multiplication between the two arguments
and applies a shift right defined by the third argument:
Return-value = (x_value * y_value) >> shiftc()
[SWS_Mfx_00055] dWe precise that for the shift right of a negative number, we always
keep the bit of sign.c()
[SWS_Mfx_00056] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00057] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype Associated
maximum shift
0x13C uint8 Mfx_MulShRight_s16s16u8_u8( 30
sint16 , sint16 , uint8 );
0x13D sint8 Mfx_MulShRight_s16s16u8_s8( 30
sint16 , sint16 , uint8 );
0x13E sint16 Mfx_MulShRight_s16s16u8_ 30
s16( sint16 , sint16 , uint8 );
0x13F uint16 Mfx_MulShRight_s16s16u8_ 30
u16( sint16 , sint16 , uint8 );
5

36 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype Associated
maximum shift
0x140 uint8 Mfx_MulShRight_u32s32u8_u8( 63
uint32 , sint32 , uint8 );
0x141 sint8 Mfx_MulShRight_u32s32u8_s8( 63
uint32 , sint32 , uint8 );
0x142 uint16 Mfx_MulShRight_u32s32u8_ 63
u16( uint32 , sint32 , uint8 );
0x143 sint16 Mfx_MulShRight_u32s32u8_ 63
s16( uint32 , sint32 , uint8 );
0x144 uint32 Mfx_MulShRight_u32s32u8_ 63
u32( uint32 , sint32 , uint8 );
0x145 sint32 Mfx_MulShRight_u32s32u8_ 63
s32( uint32 , sint32 , uint8 );
0x146 sint8 Mfx_MulShRight_s32s32u8_s8( 62
sint32 , sint32 , uint8 );
0x147 uint8 Mfx_MulShRight_s32s32u8_u8( 62
sint32 , sint32 , uint8 );
0x148 sint16 Mfx_MulShRight_s32s32u8_ 62
s16( sint32 , sint32 , uint8 );
0x149 uint16 Mfx_MulShRight_s32s32u8_ 62
u16( sint32 , sint32 , uint8 );
0x14A uint32 Mfx_MulShRight_s32s32u8_ 62
u32( sint32 , sint32 , uint8 );
0x14B sint32 Mfx_MulShRight_s32s32u8_ 62
s32( sint32 , sint32 , uint8 );
0x14C uint8 Mfx_MulShRight_u32u32u8_u8( 63
uint32 , uint32 , uint8 );
0x14D sint8 Mfx_MulShRight_u32u32u8_s8( 63
uint32 , uint32 , uint8 );
0x14E uint16 Mfx_MulShRight_u32u32u8_ 63
u16( uint32 , uint32 , uint8 );
0x14F sint16 Mfx_MulShRight_u32u32u8_ 63
s16( uint32 , uint32 , uint8 );
0x150 uint32 Mfx_MulShRight_u32u32u8_ 63
u32( uint32 , uint32 , uint8 );
0x151 sint32 Mfx_MulShRight_u32u32u8_ 63
s32( uint32 , uint32 , uint8 );

If you want to see an example of the use of these functions, see subsection 8.7.1 .

37 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.11 Combinations of division and shift left

[SWS_Mfx_00058] Definition of API function Mfx_DivShLeft_<InTypeMn1><In

TypeMn2>u8_<OutTypeMn> d
Service Name Mfx_DivShLeft_<InTypeMn1><InTypeMn2>u8_<OutTypeMn>
Syntax <OutType> Mfx_DivShLeft_<InTypeMn1><InTypeMn2>u8_<OutTypeMn> (
<InType1> x_value,
<InType2> y_value,
uint8 shift
)
Service ID [hex] 0x152 to 0x16E
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value Numerator
y_value Denominator
shift Shift left of the fixed point result. Must be a constant expression.
Maximum shift according to SWS_Mfx_00064
Parameters (inout) None
Parameters (out) None
Return value <OutType> Quotient result
Description This routine applies a shift left defined by the third argument to the first argument, and then
makes a division by the second argument.
Available via Mfx.h

c()
[SWS_Mfx_00059] dThis routine applies a shift left defined by the third argument to
the first argument, and then makes a division by the second argument:
Return-value = (x_value << shift) / y_valuec()
[SWS_Mfx_00060] dWe precise that for the shift left of a negative number, we always
keep the bit of sign.c()
[SWS_Mfx_00061] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00062] dThe result after division by zero is defined by:
* If x_value ≥ 0 then the function returns the maximum value of the output type
* If x_value < 0 then the function returns the minimum value of the output typec()
[SWS_Mfx_00064] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype Associated maximum shift
0x152 uint8 Mfx_DivShLeft_u8u8u8_u8( uint8 8
, uint8 , uint8 );
0x153 uint8 Mfx_DivShLeft_u16u16u8_u8( 16
uint16 , uint16 , uint8 );
5

38 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype Associated maximum shift
0x154 uint16 Mfx_DivShLeft_u16u16u8_u16( 16
uint16 , uint16 , uint8 );
0x155 sint16 Mfx_DivShLeft_s16s16u8_s16( 16
sint16 , sint16 , uint8 );
0x156 sint16 Mfx_DivShLeft_s16u16u8_s16( 16
sint16 , uint16 , uint8 );
0x157 uint16 Mfx_DivShLeft_u32u32u8_u16( 31
uint32 , uint32 , uint8 );
0x158 uint32 Mfx_DivShLeft_u32u32u8_u32( 31
uint32 , uint32 , uint8 );
0x159 sint32 Mfx_DivShLeft_s32s32u8_s32( 31
sint32 , sint32 , uint8 );
0x15A sint32 Mfx_DivShLeft_s32u32u8_s32( 31
sint32 , uint32 , uint8 );
0x15B uint8 Mfx_DivShLeft_u32s32u8_u8( 31
uint32 , sint32 , uint8 );
0x15C sint8 Mfx_DivShLeft_u32s32u8_s8( 31
uint32 , sint32 , uint8 );
0x15D uint16 Mfx_DivShLeft_u32s32u8_u16( 31
uint32 , sint32 , uint8 );
0x15E sint16 Mfx_DivShLeft_u32s32u8_s16( 31
uint32 , sint32 , uint8 );
0x15F uint32 Mfx_DivShLeft_u32s32u8_u32( 31
uint32 , sint32 , uint8 );
0x160 sint32 Mfx_DivShLeft_u32s32u8_s32( 31
uint32 , sint32 , uint8 );
0x161 sint8 Mfx_DivShLeft_s32s32u8_s8( 31
sint32 , sint32 , uint8 );
0x162 uint8 Mfx_DivShLeft_s32s32u8_u8( 31
sint32 , sint32 , uint8 );
0x163 sint16 Mfx_DivShLeft_s32s32u8_s16( 31
sint32 , sint32 , uint8 );
0x164 uint16 Mfx_DivShLeft_s32s32u8_u16( 31
sint32 , sint32 , uint8 );
0x165 uint32 Mfx_DivShLeft_s32s32u8_u32( 31
sint32 , sint32 , uint8 );
0x166 uint8 Mfx_DivShLeft_u32u32u8_u8( 31
uint32 , uint32 , uint8 );
0x167 sint8 Mfx_DivShLeft_u32u32u8_s8( 31
uint32 , uint32 , uint8 );
0x168 sint16 Mfx_DivShLeft_u32u32u8_s16( 31
uint32 , uint32 , uint8 );
0x169 sint32 Mfx_DivShLeft_u32u32u8_s32( 31
uint32 , uint32 , uint8 );
0x16A uint8 Mfx_DivShLeft_s32u32u8_u8( 31
sint32 , uint32 , uint8 );
0x16B sint8 Mfx_DivShLeft_s32u32u8_s8( 31
sint32 , uint32 , uint8 );
0x16C uint16 Mfx_DivShLeft_s32u32u8_u16( 31
sint32 , uint32 , uint8 );
0x16D sint16 Mfx_DivShLeft_s32u32u8_s16( 31
sint32 , uint32 , uint8 );
0x16E uint32 Mfx_DivShLeft_s32u32u8_u32( 31
sint32 , uint32 , uint8 );

39 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

If you want to see an example of the use of these functions, see subsection 8.7.2 .

8.5.12 Modulo

[SWS_Mfx_00065] Definition of API function Mfx_Mod_<TypeMn> d

Service Name Mfx_Mod_<TypeMn>
Syntax <Type> Mfx_Mod_<TypeMn> (
<Type> x_value,
<Type> y_value
)
Service ID [hex] 0x16F to 0x178
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <Type> Result of the calculation
Description This routine returns the remainder of the division x_value / y_value if y_value is not zero.
Available via Mfx.h

c()
[SWS_Mfx_00066] dIf y_value is zero, the result is zero.c()
[SWS_Mfx_00068] dIn other cases, Return-value = x_value mod y_valuec()
[SWS_Mfx_00069] dThe sign of the remainder is the same than the sign of x_value.c
()
[SWS_Mfx_00070] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x16F uint8 Mfx_Mod_u8( uint8 , uint8);
0x170 sint8 Mfx_Mod_s8( sint8 , sint8);
0x171 uint16 Mfx_Mod_u16( uint16 , uint16);
0x172 sint16 Mfx_Mod_s16( sint16 , sint16);
0x173 uint32 Mfx_Mod_u32( uint32 , uint32);
0x174 sint32 Mfx_Mod_s32( sint32 , sint32);
0x175 uint8 Mfx_Mod_u32u32_u8( uint32 , uint32)
0x176 sint8 Mfx_Mod_s32s32_s8( sint32 , sint32)
0x177 uint16 Mfx_Mod_u32u32_u16( uint32 , uint32)
0x178 sint16 Mfx_Mod_s32s32_s16( sint32 , sint32)

40 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.13 Limiting

[SWS_Mfx_00073] Definition of API function Mfx_Limit_<TypeMn> d

Service Name Mfx_Limit_<TypeMn>
Syntax <Type> Mfx_Limit_<TypeMn> (
<Type> value,
<Type> min_value,
<Type> max_value
)
Service ID [hex] 0x179 to 0x17E
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) value input value.
min_value Lower Bound. min_value shall not be strictly greater than max_
value.
max_value Upper Bound. max_value shall not be strictly lower than min_
value.
Parameters (inout) None
Parameters (out) None
Return value <Type> Result of the calculation
Description This routine limits the input value between Lower Bound and Upper Bound.
Available via Mfx.h

c()
[SWS_Mfx_00074] dReturn-value = min_value if value < min_valuec()
[SWS_Mfx_00075] dReturn-value = max_value if value > max_valuec()
[SWS_Mfx_00076] dReturn-value = value in the other casesc()
[SWS_Mfx_00079] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x179 uint8 Mfx_Limit_u8( uint8 , uint8, uint8);
0x17A sint8 Mfx_Limit_s8( sint8 , sint8, sint8);
0x17B uint16 Mfx_Limit_u16( uint16 , uint16, uint16);
0x17C sint16 Mfx_Limit_s16( sint16 , sint16, sint16);
0x17D uint32 Mfx_Limit_u32( uint32 , uint32, uint32);
0x17E sint32 Mfx_Limit_s32( sint32 , sint32, sint32);

41 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.14 Limitations with only one value for minimum and maximum

[SWS_Mfx_00082] Definition of API function Mfx_Minmax_<TypeMn> d

Service Name Mfx_Minmax_<TypeMn>
Syntax <Type> Mfx_Minmax_<TypeMn> (
<Type> value,
<Type> minmax_value
)
Service ID [hex] 0x17F to 0x184
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) value First argument
minmax_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <Type> Result of the calculation
Description The routine limits a value to a minimum or a maximum that depends on the sign of the minmax_
value.
Available via Mfx.h

c() The result value is :

[SWS_Mfx_00083] dminmax_value if minmax_value ≥ 0 and value > minmax_valuec
()
[SWS_Mfx_00084] dminmax_value if minmax_value < 0 and value < minmax_valuec()
[SWS_Mfx_00085] dvalue in the other casesc()
[SWS_Mfx_00086] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x17F uint8 Mfx_Minmax_u8( uint8 , uint8);
0x180 sint8 Mfx_Minmax_s8( sint8 , sint8);
0x181 uint16 Mfx_Minmax_u16( uint16 , uint16);
0x182 sint16 Mfx_Minmax_s16( sint16 , sint16);
0x183 uint32 Mfx_Minmax_u32( uint32 , uint32);
0x184 sint32 Mfx_Minmax_s32( sint32 , sint32);

42 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.5.15 Minimum and maximum

[SWS_Mfx_00090] Definition of API function Mfx_Min_<TypeMn> d

Service Name Mfx_Min_<TypeMn>
Syntax <Type> Mfx_Min_<TypeMn> (
<Type> x_value,
<Type> y_value
)
Service ID [hex] 0x185 to 0x18A
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <Type> Result of the calculation
Description This routine returns the minimum between two values.
Available via Mfx.h

c()
[SWS_Mfx_00091] dReturn-value = x_value if x_value < y_valuec()
[SWS_Mfx_00092] dReturn-value = y_value in the other casec()
[SWS_Mfx_00093] Definition of API function Mfx_Max_<TypeMn> d
Service Name Mfx_Max_<TypeMn>
Syntax <Type> Mfx_Max_<TypeMn> (
<Type> x_value,
<Type> y_value
)
Service ID [hex] 0x18B to 0x190
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
y_value Second argument
Parameters (inout) None
Parameters (out) None
Return value <Type> Result of the calculation
Description This routine returns the maximum between two values.
Available via Mfx.h

c()
[SWS_Mfx_00094] dReturn-value = x_value if x_value > y_valuec()
[SWS_Mfx_00095] dReturn-value = y_value in the other casec()
[SWS_Mfx_00096] dHere is the list of implemented functions.c()

43 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

Function ID[hex] Function prototype

0x185 uint8 Mfx_Min_u8( uint8 , uint8);
0x186 sint8 Mfx_Min_s8( sint8 , sint8);
0x187 uint16 Mfx_Min_u16( uint16 , uint16);
0x188 sint16 Mfx_Min_s16( sint16 , sint16);
0x189 uint32 Mfx_Min_u32( uint32 , uint32);
0x18A sint32 Mfx_Min_s32( sint32 , sint32);
0x18B uint8 Mfx_Max_u8( uint8 , uint8);
0x18C sint8 Mfx_Max_s8( sint8 , sint8);
0x18D uint16 Mfx_Max_u16( uint16 , uint16);
0x18E sint16 Mfx_Max_s16( sint16 , sint16);
0x18F uint32 Mfx_Max_u32( uint32 , uint32);
0x190 sint32 Mfx_Max_s32( sint32 , sint32);

8.5.16 Signum Function

[SWS_Mfx_91001] Definition of API function Mfx_Sgn_s32_s8 d

Service Name Mfx_Sgn_s32_s8
Syntax sint8 Mfx_Sgn_s32_s8 (
sint32 x_value
)
Service ID [hex] 0x1dc
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x_value First argument
Parameters (inout) None
Parameters (out) None
Return value sint8 Sign of the first argument
Description Signum function. Extract the sign of an integer value.
Available via Mfx.h

c(SRS_LIBS_00005, SRS_LIBS_00009, SRS_LIBS_00011)

[SWS_Mfx_00223] dExtract the sign of an integer value. It is defined as follows:
Return-value = -1, if x_value < 0; 0, if x_value == 0; 1, if x_value > 0c()

8.6 2n Scaled Integer Math Functions

For all the following functions, upper case letters will be used for operands, and lower
case letters will be used for radix.
For example :
• "x" is the operand, "a" is the parameter that represents its radix,
• "C" is the result, "c" is the parameter for its radix.

44 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

A Radix will always be a signed integer on 16 bits (sint16). For that reason, the
mnemonic will not appear in the name of the functions in order to have shorter names.
For all operations, the valid range is given for information. Indeed, operations with
parameters outside of the valid range will be saturated within the range of the output
type. It can help for optimization purpose.

45 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.1 Conversion

8.6.1.1 16-Bit to 8-Bit 2n Scaled Integer Conversion

[SWS_Mfx_00100] Definition of API function Mfx_ConvertP2_<InTypeMn>_<Out

TypeMn> d
Service Name Mfx_ConvertP2_<InTypeMn>_<OutTypeMn>
Syntax <OutType> Mfx_ConvertP2_<InTypeMn>_<OutTypeMn> (
<InType> x,
sint16 a,
sint16 c
)
Service ID [hex] 0x191 to 0x192
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed-point operand.
a Radix point position of the fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -15 <= (c - a) <= 7
Parameters (inout) None
Parameters (out) None
Return value <OutType> =2ˆ(c-a) * x
Description The routine converts a scaled 16-bit integer to a scaled 8-bit integer.
Available via Mfx.h

c()
[SWS_Mfx_00101] dThe function returns the integer value of the fixed point conversion
(C), determined by C = 2ˆ(c-a) * x.c()
[SWS_Mfx_00102] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00103] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00104] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x191 uint8 Mfx_ConvertP2_u16_u8(uint16 x, sint16 a, sint16 c)
0x192 sint8 Mfx_ConvertP2_s16_s8(sint16 x, sint16 a, sint16 c)

46 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.1.2 8-Bit to 16-Bit 2n Scaled Integer Conversion

[SWS_Mfx_00106] Definition of API function Mfx_ConvertP2_<InTypeMn>_<Out

TypeMn> d
Service Name Mfx_ConvertP2_<InTypeMn>_<OutTypeMn>
Syntax <OutType> Mfx_ConvertP2_<InTypeMn>_<OutTypeMn> (
<InType> x,
sint16 a,
sint16 c
)
Service ID [hex] 0x193 to 0x194
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed-point operand.
a Radix point position of the fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -7 <= (c - a) <= 15
Parameters (inout) None
Parameters (out) None
Return value <OutType> =2ˆ(c-a) * x
Description The routine converts a scaled 8-bit integer to a scaled 16-bit integer.
Available via Mfx.h

c()
[SWS_Mfx_00107] dThe function returns the integer value of the fixed point conversion
(C), determined by C = 2ˆ(c-a) * x.c()
[SWS_Mfx_00108] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00109] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00110] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x193 uint16 Mfx_ConvertP2_u8_u16(uint8 x, sint16 a, sint16 c)
0x194 sint16 Mfx_ConvertP2_s8_s16 (sint8 x, sint16 a, sint16 c)

47 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.1.3 32-Bit to 16-Bit 2n Scaled Integer Conversion

[SWS_Mfx_00112] Definition of API function Mfx_ConvertP2_<InTypeMn>_<Out

TypeMn> d
Service Name Mfx_ConvertP2_<InTypeMn>_<OutTypeMn>
Syntax <OutType> Mfx_ConvertP2_<InTypeMn>_<OutTypeMn> (
<InType> x,
sint16 a,
sint16 c
)
Service ID [hex] 0x195 to 0x196
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed-point operand.
a Radix point position of the fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -31 <= (c - a) <= 15
Parameters (inout) None
Parameters (out) None
Return value <OutType> =2ˆ(c-a) * x
Description The routine converts a scaled 32-bit integer to a scaled 16-bit integer.
Available via Mfx.h

c()
[SWS_Mfx_00113] dThe function returns the integer value of the fixed point conversion
(C), determined by C = 2ˆ(c-a) * x.c()
[SWS_Mfx_00114] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00115] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00116] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x195 uint16 Mfx_ConvertP2_u32_u16 (uint32 x, sint16 a, sint16 c)
0x196 sint16 Mfx_ConvertP2_s32_s16 (sint32 x, sint16 a, sint16 c)

48 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.1.4 16-Bit to 32-Bit 2n Scaled Integer Conversion

[SWS_Mfx_00118] Definition of API function Mfx_ConvertP2_<InTypeMn>_<Out

TypeMn> d
Service Name Mfx_ConvertP2_<InTypeMn>_<OutTypeMn>
Syntax <OutType> Mfx_ConvertP2_<InTypeMn>_<OutTypeMn> (
<InType> x,
sint16 a,
sint16 c
)
Service ID [hex] 0x197 to 0x198
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed-point operand.
a Radix point position of the fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -15 <= (c - a) <= 31
Parameters (inout) None
Parameters (out) None
Return value <OutType> =2ˆ(c-a) * x
Description The routine converts a scaled 16-bit integer to a scaled 32-bit integer.
Available via Mfx.h

c()
[SWS_Mfx_00119] dThe function returns the integer value of the fixed point conversion
(C), determined by C = 2ˆ(c-a) * x.c()
[SWS_Mfx_00120] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00121] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00122] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x197 uint32 Mfx_ConvertP2_u16_u32(uint16 x, sint16 a, sint16 c)
0x198 sint32 Mfx_ConvertP2_s16_s32(sint16 x, sint16 a, sint16 c)

49 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.2 Multiplication

8.6.2.1 16-Bit Multiplication of 2n Scaled Integer

[SWS_Mfx_00124] Definition of API function Mfx_MulP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_MulP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_MulP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x199 to 0x19E
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -31 <= (c - b - a) <= 15
Parameters (inout) None
Parameters (out) None
Return value <OutType> = 2ˆ(c-b-a) * [x * y]
Description The routine multiplies two 16-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00125] dThe function returns the integer value of the fixed point multiplica-
tion (C), determined by C = 2ˆ(c-b-a) * [x * y].c()
[SWS_Mfx_00126] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00127] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00128] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x199 uint16 Mfx_MulP2_u16u16_u16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x19A uint16 Mfx_MulP2_u16s16_u16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
5

50 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Function ID[hex] Function prototype
0x19B uint16 Mfx_MulP2_s16s16_u16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x19C sint16 Mfx_MulP2_u16u16_s16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x19D sint16 Mfx_MulP2_u16s16_s16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x19E sint16 Mfx_MulP2_s16s16_s16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)

8.6.2.2 32-Bit Multiplication of 2n Scaled Integer

[SWS_Mfx_00130] Definition of API function Mfx_MulP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_MulP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_MulP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x19F to 0x1A4
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -63 <= (c - b - a) <= 31
Parameters (inout) None
Parameters (out) None
Return value <OutType> = 2ˆ(c-b-a) * [x * y]
Description The routine multiplies two 32-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00131] dThe function returns the integer value of the fixed point multiplica-
tion (C), determined by C = 2ˆ(c-b-a) * [x * y].c()
[SWS_Mfx_00132] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00133] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()

51 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00134] dHere is the list of implemented functions.c()

Function ID[hex] Function prototype
0x19F uint32 Mfx_MulP2_u32u32_u32(uint32 x, uint32 y, sint16 a,
sint16 b, sint16 c)
0x1A0 uint32 Mfx_MulP2_u32s32_u32(uint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)
0x1A1 uint32 Mfx_MulP2_s32s32_u32(sint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)
0x1A2 sint32 Mfx_MulP2_u32u32_s32(uint32 x, uint32 y, sint16 a,
sint16 b, sint16 c)
0x1A3 sint32 Mfx_MulP2_u32s32_s32(uint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)
0x1A4 sint32 Mfx_MulP2_s32s32_s32(sint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)

8.6.3 Division

8.6.3.1 16-Bit Division of 2n Scaled Integer

[SWS_Mfx_00136] Definition of API function Mfx_DivP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_DivP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_DivP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1A5 to 0x1AC
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -15 <= (c + b - a) <= 31
Parameters (inout) None
Parameters (out) None
Return value <OutType> = [2ˆ(c+b-a) * x] / y
Description The routine divides two 16-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()

52 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00137] dThe function returns the integer value of the fixed point quotient
(C), determined by C = [2ˆ(c+b-a) * x] / y.c()
[SWS_Mfx_00138] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00139] dIf the divisor, y, is zero, the result is defined by:
* If x ≥ 0 then the function returns the maximum value of the output type
* If x < 0 then the function returns the minimum value of the output typec()
[SWS_Mfx_00141] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00142] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x1A5 uint16 Mfx_DivP2_u16u16_u16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1A6 uint16 Mfx_DivP2_u16s16_u16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1A7 uint16 Mfx_DivP2_s16u16_u16(sint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1A8 uint16 Mfx_DivP2_s16s16_u16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1A9 sint16 Mfx_DivP2_u16u16_s16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1AA sint16 Mfx_DivP2_u16s16_s16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1AB sint16 Mfx_DivP2_s16u16_s16(sint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1AC sint16 Mfx_DivP2_s16s16_s16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)

53 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.3.2 32-Bit Division of 2n Scaled Integer

[SWS_Mfx_00144] Definition of API function Mfx_DivP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_DivP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_DivP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1AD to 0x1B4
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -31 <= (c + b - a) <= 63
Parameters (inout) None
Parameters (out) None
Return value <OutType> = [2ˆ(c+b-a) * x] / y
Description The routine divides two 32-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00145] dThe function returns the integer value of the fixed point quotient
(C), determined by C = [2ˆ(c+b-a) * x] / y.c()
[SWS_Mfx_00146] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00147] dIf the divisor, y, is zero, the result is defined by:
* If x ≥ 0 then the function returns the maximum value of the output type
* If x < 0 then the function returns the minimum value of the output typec()
[SWS_Mfx_00149] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00150] dHere is the list of implemented functions.c()

54 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

Function ID[hex] Function prototype

0x1AD uint32 Mfx_DivP2_u32u32_u32(uint32 x, uint32 y, sint16 a,
sint16 b, sint16 c)
0x1AE uint32 Mfx_DivP2_u32s32_u32(uint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)
0x1AF uint32 Mfx_DivP2_s32u32_u32(sint32 x, uint32 y, sint16 a,
sint16 b, sint16 c)
0x1B0 uint32 Mfx_DivP2_s32s32_u32(sint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)
0x1B1 sint32 Mfx_DivP2_u32u32_s32(uint32 x, uint32 y, sint16 a,
sint16 b, sint16 c)
0x1B2 sint32 Mfx_DivP2_u32s32_s32(uint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)
0x1B3 sint32 Mfx_DivP2_s32u32_s32(sint32 x, uint32 y, sint16 a,
sint16 b, sint16 c)
0x1B4 sint32 Mfx_DivP2_s32s32_s32(sint32 x, sint32 y, sint16 a,
sint16 b, sint16 c)

8.6.4 Addition

8.6.4.1 16-Bit Addition of 2n Scaled Integer

[SWS_Mfx_00152] Definition of API function Mfx_AddP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_AddP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_AddP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1B5 to 0x1BA
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: 0 <= |a - b| <= 15 (c - b) <= 15, (a - c) <=
15, a >= b (c - a) <= 15, (b - c) <= 15, a < b
Parameters (inout) None
Parameters (out) None
Return value <OutType> a >= b: 2ˆ(c-a) * [x + (y * 2ˆ(a-b))],
a < b: 2ˆ(c-b) * [(x * 2ˆ(b-a)) + y]
5

55 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Description The routine adds two 16-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00153] dThe function returns the integer value of the fixed point sum (C),
determined by
a ≥ b: C = 2ˆ(c-a) * [x + (y * 2ˆ(a-b))],
a < b: C = 2ˆ(c-b) * [(x * 2ˆ(b-a)) + y].c()
[SWS_Mfx_00154] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00155] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00156] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x1B5 uint16 Mfx_AddP2_u16u16_u16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1B6 uint16 Mfx_AddP2_u16s16_u16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1B7 uint16 Mfx_AddP2_s16s16_u16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1B8 sint16 Mfx_AddP2_u16u16_s16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1B9 sint16 Mfx_AddP2_u16s16_s16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1BA sint16 Mfx_AddP2_s16s16_s16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)

56 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.4.2 32-Bit Addition of 2n Scaled Integer

[SWS_Mfx_00158] Definition of API function Mfx_AddP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_AddP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_AddP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1BB to 0x1C0
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: 0 <= |a - b| <= 31 (c - b) <= 31, (a - c) <=
31, a >= b (c - a) <= 31, (b - c) <= 31, a < b
Parameters (inout) None
Parameters (out) None
Return value <OutType> a >= b: 2ˆ(c-a) * [x + (y * 2ˆ(a-b))],
a < b: 2ˆ(c-b) * [(x * 2ˆ(b-a)) + y]
Description The routine adds two 32-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00159] dThe function returns the integer value of the fixed point sum (C),
determined by
a ≥ b: C = 2ˆ(c-a) * [x + (y * 2ˆ(a-b))],
a < b: C = 2ˆ(c-b) * [(x * 2ˆ(b-a)) + y]c()
[SWS_Mfx_00160] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00161] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()

57 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00162] dHere is the list of implemented functions.c()

Function ID[hex] Function prototype
0x1BB uint32 Mfx_AddP2_u32u32_u32(uint32 x, uint32 y, sint32 a,
sint32 b, sint32 c)
0x1BC uint32 Mfx_AddP2_u32s32_u32(uint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)
0x1BD uint32 Mfx_AddP2_s32s32_u32(sint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)
0x1BE sint32 Mfx_AddP2_u32u32_s32(uint32 x, uint32 y, sint32 a,
sint32 b, sint32 c)
0x1BF sint32 Mfx_AddP2_u32s32_s32(uint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)
0x1C0 sint32 Mfx_AddP2_s32s32_s32(sint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)

8.6.5 Subtraction

8.6.5.1 16-Bit Subtraction of 2n Scaled Integer

[SWS_Mfx_00164] Definition of API function Mfx_SubP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_SubP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_SubP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1C1 to 0x1C8
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: 0 <= |a - b| <= 15 (c - b) <= 15, (a - c) <=
15, a >= b (c - a) <= 15, (b - c) <= 15, a < b
Parameters (inout) None
Parameters (out) None
Return value <OutType> a >= b: 2ˆ(c-a) * [x - (y * 2ˆ(a-b))]
a < b: 2ˆ(c-b) * [(x * 2ˆ(b-a)) - y]
Description The routine subtracts two 16-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()

58 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00165] dThe function returns the integer value of the fixed point difference
(C), determined by
a ≥ b: C = 2ˆ(c-a) * [x - (y * 2ˆ(a-b))],
a < b: C = 2ˆ(c-b) * [(x * 2ˆ(b-a)) - y]c()
[SWS_Mfx_00166] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00167] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00168] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x1C1 uint16 Mfx_SubP2_u16u16_u16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1C2 uint16 Mfx_SubP2_u16s16_u16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1C3 uint16 Mfx_SubP2_s16u16_u16(sint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1C4 uint16 Mfx_SubP2_s16s16_u16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1C5 sint16 Mfx_SubP2_u16u16_s16(uint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1C6 sint16 Mfx_SubP2_u16s16_s16(uint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)
0x1C7 sint16 Mfx_SubP2_s16u16_s16(sint16 x, uint16 y, sint16 a,
sint16 b, sint16 c)
0x1C8 sint16 Mfx_SubP2_s16s16_s16(sint16 x, sint16 y, sint16 a,
sint16 b, sint16 c)

59 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.5.2 32-Bit Subtraction of 2n Scaled Integer

[SWS_Mfx_00170] Definition of API function Mfx_SubP2_<InTypeMn1><InType

Mn2>_<OutTypeMn> d
Service Name Mfx_SubP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_SubP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1C9 to 0x1D0
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: 0 <= |a - b| <= 31 (c - b) <= 31, (a - c) <=
31, a >= b (c - a) <= 31, (b - c) <= 31, a < b
Parameters (inout) None
Parameters (out) None
Return value <OutType> a >= b: 2ˆ(c-a) * [x - (y * 2ˆ(a-b))]
a < b: 2ˆ(c-b) * [(x * 2ˆ(b-a)) - y]
Description The routine subtracts two 32-bit integers with scaling factors set by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00171] dThe function returns the integer value of the fixed point difference
(C), determined by
a ≥ b: C = 2ˆ(c-a) * [x - (y * 2ˆ(a-b))],
a < b: C = 2ˆ(c-b) * [(x * 2ˆ(b-a)) - y].c()
[SWS_Mfx_00172] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00173] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()

60 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

[SWS_Mfx_00174] dHere is the list of implemented functions.c()

Function ID[hex] Function prototype
0x1C9 uint32 Mfx_SubP2_u32u32_u32(uint32 x, uint32 y, sint32 a,
sint32 b, sint32 c)
0x1CA uint32 Mfx_SubP2_u32s32_u32(uint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)
0x1CB uint32 Mfx_SubP2_s32u32_u32(sint32 x, uint32 y, sint32 a,
sint32 b, sint32 c)
0x1CC uint32 Mfx_SubP2_s32s32_u32(sint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)
0x1CD sint32 Mfx_SubP2_u32u32_s32(uint32 x, uint32 y, sint32 a,
sint32 b, sint32 c)
0x1CE sint32 Mfx_SubP2_u32s32_s32(uint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)
0x1CF sint32 Mfx_SubP2_s32u32_s32(sint32 x, uint32 y, sint32 a,
sint32 b, sint32 c)
0x1D0 sint32 Mfx_SubP2_s32s32_s32(sint32 x, sint32 y, sint32 a,
sint32 b, sint32 c)

8.6.6 Absolute Difference of 2n Scaled Integer

[SWS_Mfx_00176] Definition of API function Mfx_AbsDiffP2_<InTypeMn1><In

TypeMn2>_<OutTypeMn> d
Service Name Mfx_AbsDiffP2_<InTypeMn1><InTypeMn2>_<OutTypeMn>
Syntax <OutType> Mfx_AbsDiffP2_<InTypeMn1><InTypeMn2>_<OutTypeMn> (
<InType1> x,
<InType2> y,
sint16 a,
sint16 b,
sint16 c
)
Service ID [hex] 0x1D1 to 0x1D6
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
y Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
b Radix point position of the second fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: 0 <= |a - b| <= 15 (c - b) <= 15, (a - c) <=
15, a >= b (c - a) <= 15, (b - c) <= 15, a < b
Parameters (inout) None
Parameters (out) None
Return value <OutType> a >= b: 2ˆ(c-a) * |x - (y * 2ˆ(a-b))|
a < b: 2ˆ(c-b) * |(x * 2ˆ(b-a)) - y|
5

61 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

4
Description The routine subtracts and takes the absolute value of two 16-bit integers with scaling factors set
by input parameters.
Available via Mfx.h

c()
[SWS_Mfx_00177] dThe function returns the integer value of the fixed point absolute
difference (C), determined by
a ≥ b: C = 2ˆ(c-a) * |x - (y * 2ˆ(a-b))|,
a < b: C = 2ˆ(c-b) * |(x * 2ˆ(b-a)) - y|.c()
[SWS_Mfx_00178] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00179] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00180] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x1D1 uint16 Mfx_AbsDiffP2_u16u16_u16(uint16 x, uint16 y, sint16
a, sint16 b, sint16 c)
0x1D2 uint16 Mfx_AbsDiffP2_u16s16_u16(uint16 x, sint16 y, sint16
a, sint16 b, sint16 c)
0x1D3 uint16 Mfx_AbsDiffP2_s16s16_u16(sint16 x, sint16 y, sint16
a, sint16 b, sint16 c)
0x1D4 sint16 Mfx_AbsDiffP2_u16u16_s16(uint16 x, uint16 y, sint16
a, sint16 b, sint16 c)
0x1D5 sint16 Mfx_AbsDiffP2_u16s16_s16(uint16 x, sint16 y, sint16
a, sint16 b, sint16 c)
0x1D6 sint16 Mfx_AbsDiffP2_s16s16_s16(sint16 x, sint16 y, sint16
a, sint16 b, sint16 c)

62 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.7 Absolute Value

8.6.7.1 16-Bit Absolute Value of 2n Scaled Integer

[SWS_Mfx_00182] Definition of API function Mfx_AbsP2_s16_<OutTypeMn> d

Service Name Mfx_AbsP2_s16_<OutTypeMn>
Syntax <OutType> Mfx_AbsP2_s16_<OutTypeMn> (
<InType1> x,
sint16 a,
sint16 c
)
Service ID [hex] 0x1D7 to 0x1D8
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -15 <= (c - a) <= 15
Parameters (inout) None
Parameters (out) None
Return value <OutType> 2ˆ(c-a) * |x|
Description The routine takes the absolute value of a 16-bit integer with scaling factors set by input
parameters.
Available via Mfx.h

c()
[SWS_Mfx_00183] dThe function returns the integer value of the fixed point absolute
value (C), deter-mined by C = 2ˆ(c-a) * |x|.c()
[SWS_Mfx_00184] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00185] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00186] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
()
()
0x1D7 uint16 Mfx_AbsP2_s16_u16(sint16 x, sint16 a, sint16 c)
0x1D8 sint16 Mfx_AbsP2_s16_s16(sint16 x, sint16 a, sint16 c)

63 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.6.7.2 32-Bit Absolute Value of 2n Scaled Integer

[SWS_Mfx_00188] Definition of API function Mfx_AbsP2_s32_<OutTypeMn> d

Service Name Mfx_AbsP2_s32_<OutTypeMn>
Syntax <OutType> Mfx_AbsP2_s32_<OutTypeMn> (
<InType1> x,
sint16 a,
sint16 c
)
Service ID [hex] 0x1D9 to 0x1DA
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) x Integer value of the fixed point operand.
a Radix point position of the first fixed point operand. Must be a
constant expression.
c Radix point position of the fixed point result. Must be a constant
expression. Valid range: -31 <= (c - a) <= 31
Parameters (inout) None
Parameters (out) None
Return value <OutType> 2ˆ(c-a) * |x|
Description The routine takes the absolute value of a 32-bit integer with scaling factors set by input
parameters.
Available via Mfx.h

c()
[SWS_Mfx_00189] dThe function returns the integer value of the fixed point absolute
value (C), deter-mined by C = 2ˆ(c-a) * |x|.c()
[SWS_Mfx_00190] dReturn-value shall be saturated to boundary values in the event
of negative or positive overflow.c()
[SWS_Mfx_00191] dIf it is necessary to round the result of this equation, it is rounded
toward zero.c()
[SWS_Mfx_00192] dHere is the list of implemented functions.c()
Function ID[hex] Function prototype
0x1D9 uint32 Mfx_AbsP2_s32_u32(sint32 x, sint16 a, sint16 c)
0x1DA sint32 Mfx_AbsP2_s32_s32(sint32 x, sint16 a, sint16 c)

64 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.7 Examples of use of functions

8.7.1 Combinations of multiplication and shift right

The function that multiplies an argument by a factor of a given range can be interpreted
as the combination of multiplication and shift right.
If we consider the factor that is a power of two : 2n1
If we consider the maximum of the type used to code the factor : 2n2-1
Then, the shift right we shall apply to the result of the multiplication is given by :
(n2-n1)
For example, we multiply a s8 value (argument1) by a factor of 1 (20) coded with an u8
(Max(u8)=28-1).
The physical range of the factor is [0 , 0.996]
The result is :
Mfx_MulShRight_s16s16u8_s8(argument1, factor, 8)

8.7.2 Combinations of division and shift left

In the domain of power train, the function that divides two arguments to compute a
factor of a given range can be interpreted as the combination of division and shift left.
If we consider the factor that is a power of two : 2n1
If we consider the maximum of the type used to code the result (factor) : 2n2-1
Then, the shift left we shall apply to the result of the division is given by : (n2-n1)
For example, we divide two u16 values (argument1 and argument2) to obtain a factor
of 1 (20) coded with an u16 (Max(u16)=216-1).
The physical range of the result is [0 , 0.999985]
The result is :
Mfx_DivShLeft_u16u16u8_u16(argument1, argument2, 16)

65 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.8 Version API

8.8.1 Mfx_GetVersionInfo

[SWS_Mfx_00215] Definition of API function Mfx_GetVersionInfo d

Service Name Mfx_GetVersionInfo
Syntax void Mfx_GetVersionInfo (
Std_VersionInfoType* versioninfo
)
Service ID [hex] 0x1DB
Sync/Async Synchronous
Reentrancy Reentrant
Parameters (in) None
Parameters (inout) None
Parameters (out) versioninfo Pointer to where to store the version information of this module.
Format according [BSW00321]
Return value None
Description Returns the version information of this library.
Available via Mfx.h

c(SRS_BSW_00407, SRS_BSW_00003, SRS_BSW_00318, SRS_BSW_00321) The

version information of a BSW module generally contains:
Module Id
Vendor Id
Vendor specific version numbers (SRS_BSW_00407).
[SWS_Mfx_00216] dIf source code for caller and callee of Mfx_ GetVersionInfo is avail-
able, the Mfx library should realize Mfx_ GetVersionInfo as a macro defined in the
module’s header file.c(SRS_BSW_00407, SRS_BSW_00411)

8.9 Callback notifications

None.

8.10 Scheduled functions

The MFX library does not have scheduled functions.

8.11 Expected interfaces

None.

66 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

8.11.1 Mandatory interfaces

None.

8.11.2 Optional interfaces

None.

8.11.3 Configurable interfaces

None.

8.12 Service Interfaces

None.

67 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

9 Sequence diagrams
Not applicable.

68 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

10 Configuration specification
In general, this chapter defines configuration parameters and their clustering into con-
tainers. In order to support the specification Chapter 10.1 describes fundamentals.
It also specifies a template (table) you shall use for the parameter specification. We
intend to leave Chapter 10.1 in the specification to guarantee comprehension.
Chapter 10.2 specifies the structure (containers) and the parameters of the module
MFXLibrary.
Chapter 10.3 specifies published information of the module MFXLibrary.

10.1 How to read this chapter

For details refer to the chapter 10.1 “Introduction to configuration specification” in
SWS_BSWGeneral.

10.2 Containers and configuration parameters

[SWS_Mfx_00218] dThe MFX library shall not have any configuration options that may
affect the functional behavior of the routines. I.e. for a given set of input parameters,
the outputs shall be always the same. For example, the returned value in case of error
shall not be configurable.c(SRS_LIBS_00001)
However, a library vendor is allowed to add specific configuration options concerning
library implementation, e.g. for resources consumption optimization.

10.3 Published Information

For details refer to the chapter 10.3 “Published Information” in SWS_BSWGeneral.
[SWS_Mfx_00214] dThe standardized common published parameters as required by
SRS_BSW_00402 in the SRS General on Basic Software Modules [chapter 3.1] shall
be published
within the header file of this module and need to be provided in the BSW Module De-
scription. The according module abbreviation can be found in the List of Basic Software
Modules [chapter 3.1] .c(SRS_BSW_00402, SRS_BSW_00374, SRS_BSW_00379)
Additional module-specific published parameters are listed below if applicable.

69 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

A Not applicable requirements

[SWS_Mfx_00222] dThese requirements are not applicable to this specification.c()

70 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary

 Specification of Fixed Point Math Routines
AUTOSAR CP R23-11

B History of Specification Items

Please note that the lists in this chapter also include specification items that have been
removed from the specification in a later version. These specification items do not
appear as hyperlinks in the document.

B.1 Specification Item History of this document compared to

AUTOSAR R22-11.

B.1.1 Added Specification Items in R23-11

none

B.1.2 Changed Specification Items in R23-11

none

B.1.3 Deleted Specification Items in R23-11

none

71 of 71 Document ID 394: AUTOSAR_CP_SWS_MFXLibrary