MATHLIB

DSP Math Library for Floating Point Devices

MATHLIB

Overview

The Texas Instruments math library is an optimized floating-point math function library for C programmers using TI floating point devices. These routines are typically used in computationally intensive real-time applications where optimal execution speed is critical. By using these routines instead of the routines found in the existing run-time-support libraries, considerably faster execution speeds can be achieved without rewriting existing code. The MATHLIB library includes all the floating-point math routines that are currently provided in the existing run-time-support libraries. These new functions can be called with the current run-time-support library names or the new names included in the math library.

The MATHLIB library includes all the floating-point math routines that are currently provided in the existing run-time-support libraries. These new functions can be called with the current run-time-support library names or the new names included in the math library.

See the benchmarks: DSP core benchmarks

Features

Types of functions included:
- Trigonometric and hyperbolic: Sin, Cos, Tan, Arctan, etc.
- Power, exponential, and logarithmic
- Reciprocal
- Square root
- Division
Natural C Source Code
Optimized C code with Intrinsics
Hand-coded assembly-optimized routines
C-callable routines, which can be inlined and are fully compatible with the TMS320C6000 compiler
Routines which accept single sample or vector inputs
Provided functions are tested against C model and existing run-time-support functions
Benchmarks (cycle and code size)
Compiled using Code Generation Tools v7.2.0

Download View video with transcript Video

Downloads

Driver or library

C674X-C66X-MATHLIB — C674x and C66x math library

Download options

C674X-C66X-MATHLIB — C674x and C66x math library

close

Latest version

Version: 3.1.2.1

Release date: 17 May 2016

Release notes

View all versions

MATHLIB_c66x_MisraC_Policy.xlsx — 40 K

C66x MATHLIB MISRA Compliance

MD5 checksum

mathlib_c66x_3_1_2_1_Win32.exe — 26394 K

MATHLIB for C66x Windows Installer

MD5 checksum

mathlib_c66x_3_1_2_1_Linux.bin — 26228 K

MATHLIB for C66x Linux Installer

MD5 checksum

mathlib_c674x_3_1_2_1_Win32.exe — 14794 K

MATHLIB for C674x Windows Installer

MD5 checksum

mathlib_c674x_3_1_2_1_Linux.bin — 14629 K

MATHLIB for C674x Linux Installer

MD5 checksum

mathlib_rts_c66x_3_1_2_1_Win32.exe — 23406 K

MATHLIB RTS override library for C66x Windows Installer

MD5 checksum

mathlib_rts_c66x_3_1_2_1_Linux.bin — 23241 K

MATHLIB RTS override library for C66x Linux Installer

MD5 checksum

mathlib_rts_c674x_3_1_2_1_Win32.exe — 13312 K

MATHLIB RTS override library for C674x Windows Installer

MD5 checksum

mathlib_rts_c674x_3_1_2_1_Linux.bin — 13146 K

MATHLIB RTS override library for C674x Linux Installer

MD5 checksum

Documentation

MATHLIB_Function_Reference.chm

MATHLIB Function Reference

MATHLib_c674x_TestReport.html

C674x MATHLIB Benchmarks

MATHLib_c66x_TestReport.html

C66x MATHLIB Benchmarks

MATHLIB 3.1.0

MATHLIB Post Production Release

MATHLIB 3.0.2

MATHLIB Post Production Release

MATHLIB 3.0.1

MATHLIB Post Production Release

MATHLIB 3.0.0

MATHLIB Post Production Release

Software_Manifest.html

MATHLIB Software Manifest

Release Information

This is a GA release of the Floating-Point Math Function Library (MATHLIB) for C674x/C66x processors.

MATHLIB contains optimized versions of most floating-point math routines currently provided in existing run-time-support libraries.

MATHLIB RTS override library contains floating-point math routines with same symbol names as in run-time-support library. These routines can be used to replace existing run-time-support library function calls.

The deliverables in this release include a Windows or Linux installation executable for either the C674x or C66x processors. Each executable installs a component package repository, a documentation directory, an Eclipse plugin directory and an expanded component directory structure with component libraries, header files and test examples.

Software Defects

A list of tracked software defects can be found here.

(Note: If prompted for login/password, close the window and try the link again)

Driver or library

C67X-MATHLIB — DSP Math Library for C67x Floating Point Devices

Supported products & hardware

Supported products & hardware

Products

Arm-based processors

OMAPL137-HT — High temperature low power C674x floating-point DSP + Arm processor - up to 456 MHz

OMAPL138B-EP — Enhanced product low power C674x floating-point DSP + Arm9 processor - 345 MHz

Digital signal processors (DSPs)

DM505 — SoC for vision analytics 15mm package

SM320C6678-HIREL — High reliability product high performance 8-core C6678 fixed and floating-point DSP

SM320C6727B — Military grade C6727B floating-point DSP

SM320C6727B-EP — Enhanced product C6727 floating-point DSP

SMV320C6727B-SP — Space grade C6727B floating-point DSP - rad-tolerant class V with ceramic package

TMS320C6701 — C67x floating-point DSP- up to 167MHz, McBSP

TMS320C6711D — C67x floating-point DSP- up to 250MHz, McBSP, 32-Bit EMIFA

TMS320C6712D — C67x floating-point DSP- up to 150MHz, McBSP, 16-Bit EMIFA

TMS320C6720 — C67x floating-point DSP - 200MHz, McASP, 16-Bit EMIFA

TMS320C6722B — C67x floating-point DSP- up to 250MHz, McASP, 16-Bit EMIFA

TMS320C6726B — C67x floating-point DSP- up to 266MHz, McASP, 16-Bit EMIFA

TMS320C6727 — C67x floating-point DSP- up to 250MHz, McASP, 32-Bit EMIFA

TMS320C6727B — C67x floating-point DSP- up to 350MHz, McASP, 32-Bit EMIFA

TMS320C6743 — Low power C674x floating-point DSP- 375MHz

TMS320C6745 — Low power C674x floating-point DSP- 456MHz, QFP

TMS320C6747 — Low power C674x floating-point DSP- 456MHz, PBGA

Download options

C67X-MATHLIB — DSP Math Library for C67x Floating Point Devices

close

Latest version

Version: 2.01.00.00

Release date: 20 Oct 2010

Release notes

View all versions

c67xmathlib_2_01_00_00_Windows_Setup.exe — 3286 K

Current Windows installer

MD5 checksum

c67xmathlib_2_01_00_00_Linux-x86_Setup.bin — 3045 K

Current Linux installer

MD5 checksum

Products

Arm-based processors

OMAPL137-HT — High temperature low power C674x floating-point DSP + Arm processor - up to 456 MHz

OMAPL138B-EP — Enhanced product low power C674x floating-point DSP + Arm9 processor - 345 MHz

Digital signal processors (DSPs)

DM505 — SoC for vision analytics 15mm package

SM320C6678-HIREL — High reliability product high performance 8-core C6678 fixed and floating-point DSP

SM320C6727B — Military grade C6727B floating-point DSP

SM320C6727B-EP — Enhanced product C6727 floating-point DSP

SMV320C6727B-SP — Space grade C6727B floating-point DSP - rad-tolerant class V with ceramic package

TMS320C6701 — C67x floating-point DSP- up to 167MHz, McBSP

TMS320C6711D — C67x floating-point DSP- up to 250MHz, McBSP, 32-Bit EMIFA

TMS320C6712D — C67x floating-point DSP- up to 150MHz, McBSP, 16-Bit EMIFA

TMS320C6720 — C67x floating-point DSP - 200MHz, McASP, 16-Bit EMIFA

TMS320C6722B — C67x floating-point DSP- up to 250MHz, McASP, 16-Bit EMIFA

TMS320C6726B — C67x floating-point DSP- up to 266MHz, McASP, 16-Bit EMIFA

TMS320C6727 — C67x floating-point DSP- up to 250MHz, McASP, 32-Bit EMIFA

TMS320C6727B — C67x floating-point DSP- up to 350MHz, McASP, 32-Bit EMIFA

TMS320C6743 — Low power C674x floating-point DSP- 375MHz

TMS320C6745 — Low power C674x floating-point DSP- 456MHz, QFP

TMS320C6747 — Low power C674x floating-point DSP- 456MHz, PBGA

Release Information

C67X-MATHLIB

Additional resources you might need

Driver or library

C674X-C66X-MATHLIB — C674x and C66x math library

Download options

C674X-C66X-MATHLIB — C674x and C66x math library

close

Latest version

Version: 3.1.2.1

Release date: 17 May 2016

Release notes

View all versions

MATHLIB_c66x_MisraC_Policy.xlsx — 40 K

C66x MATHLIB MISRA Compliance

MD5 checksum

mathlib_c66x_3_1_2_1_Win32.exe — 26394 K

MATHLIB for C66x Windows Installer

MD5 checksum

mathlib_c66x_3_1_2_1_Linux.bin — 26228 K

MATHLIB for C66x Linux Installer

MD5 checksum

mathlib_c674x_3_1_2_1_Win32.exe — 14794 K

MATHLIB for C674x Windows Installer

MD5 checksum

mathlib_c674x_3_1_2_1_Linux.bin — 14629 K

MATHLIB for C674x Linux Installer

MD5 checksum

mathlib_rts_c66x_3_1_2_1_Win32.exe — 23406 K

MATHLIB RTS override library for C66x Windows Installer

MD5 checksum

mathlib_rts_c66x_3_1_2_1_Linux.bin — 23241 K

MATHLIB RTS override library for C66x Linux Installer

MD5 checksum

mathlib_rts_c674x_3_1_2_1_Win32.exe — 13312 K

MATHLIB RTS override library for C674x Windows Installer

MD5 checksum

mathlib_rts_c674x_3_1_2_1_Linux.bin — 13146 K

MATHLIB RTS override library for C674x Linux Installer

MD5 checksum

Documentation

MATHLIB_Function_Reference.chm

MATHLIB Function Reference

MATHLib_c674x_TestReport.html

C674x MATHLIB Benchmarks

MATHLib_c66x_TestReport.html

C66x MATHLIB Benchmarks

MATHLIB 3.1.0

MATHLIB Post Production Release

MATHLIB 3.0.2

MATHLIB Post Production Release

MATHLIB 3.0.1

MATHLIB Post Production Release

MATHLIB 3.0.0

MATHLIB Post Production Release

Software_Manifest.html

MATHLIB Software Manifest

Release Information

This is a GA release of the Floating-Point Math Function Library (MATHLIB) for C674x/C66x processors.

MATHLIB contains optimized versions of most floating-point math routines currently provided in existing run-time-support libraries.

MATHLIB RTS override library contains floating-point math routines with same symbol names as in run-time-support library. These routines can be used to replace existing run-time-support library function calls.

The deliverables in this release include a Windows or Linux installation executable for either the C674x or C66x processors. Each executable installs a component package repository, a documentation directory, an Eclipse plugin directory and an expanded component directory structure with component libraries, header files and test examples.

Software Defects

A list of tracked software defects can be found here.

(Note: If prompted for login/password, close the window and try the link again)

Supported products & hardware

Arm-based processors

66AK2E05 — High performance multicore DSP+Arm - 4x Arm A15 cores, 1x C66x DSP core, NetCP, 10GbE

66AK2H06 — High performance multicore DSP+Arm - 2x Arm A15 cores, 4x C66x DSP cores

66AK2H12 — High performance multicore DSP+Arm - 4x Arm A15 cores, 8x C66x DSP cores

66AK2H14 — High performance multicore DSP+Arm - 4x Arm A15 cores, 8x C66x DSP cores, 10GbE

AM5706 — Sitara processor: cost optimized Arm Cortex-A15 & DSP and secure boot

AM5708 — Sitara processor: cost optimized Arm Cortex-A15 & DSP, multimedia and secure boot

AM5716 — Sitara processor: Arm Cortex-A15 & DSP

AM5718 — Sitara processor: Arm Cortex-A15 & DSP, multimedia

AM5726 — Sitara processor: dual Arm Cortex-A15 & dual DSP

AM5728 — Sitara processor: dual Arm Cortex-A15 & dual DSP, multimedia

AM5746 — Sitara processor: dual arm Cortex-A15 & dual DSP, ECC on DDR and secure boot

AM5748 — Sitara processor: dual arm Cortex-A15 & dual DSP, multimedia, ECC on DDR and secure boot

AM5749 — Sitara processor: dual Arm Cortex-A15 & dual DSP, multimedia, ECC on DDR, secure boot, deep learning

OMAP-L132 — Low power C674x floating-point DSP + Arm9 processor - 200MHz

OMAP-L137 — Low power C674x floating-point DSP + Arm9 processor - up to 456 MHz

OMAP-L138 — Low power C674x floating-point DSP + Arm9 processor - up to 456MHz

OMAPL137-HT — High temperature low power C674x floating-point DSP + Arm processor - up to 456 MHz

OMAPL138B-EP — Enhanced product low power C674x floating-point DSP + Arm9 processor - 345 MHz

SMOMAPL138B-HIREL — High reliability product low power C674x floating-point DSP + Arm9 processor - 375 MHz

Digital signal processors (DSPs)

66AK2L06 — Multicore DSP+ARM KeyStone II System-on-Chip (SoC)

DM505 — SoC for vision analytics 15mm package

SM320C6678-HIREL — High reliability product high performance 8-core C6678 fixed and floating-point DSP

SM320C6727B — Military grade C6727B floating-point DSP

SM320C6727B-EP — Enhanced product C6727 floating-point DSP

SMV320C6727B-SP — Space grade C6727B floating-point DSP - rad-tolerant class V with ceramic package

TMS320C6654 — High performance single-core C66x fixed and floating-point DSP- up to 850MHz

TMS320C6655 — High performance single-core C66x fixed and floating-point DSP- up to 1.25GHz

TMS320C6657 — High performance dual-core C66x fixed and floating-point DSP- up to 1.25GHz, 2 UART

TMS320C6670 — 4 core fixed and floating point DSP for Communications and Telecom

TMS320C6671 — High performance single-core C66x fixed and floating-point DSP - 1GHz

TMS320C6672 — High performance dual-core C66x fixed and floating-point DSP- up to 1.25GHz

TMS320C6674 — High performance quad-core C66x fixed and floating-point DSP- up to 1.25GHz

TMS320C6678 — High performance octo-core C66x fixed and floating-point DSP- up to 1.25GHz

TMS320C6701 — C67x floating-point DSP- up to 167MHz, McBSP

TMS320C6711D — C67x floating-point DSP- up to 250MHz, McBSP, 32-Bit EMIFA

TMS320C6712D — C67x floating-point DSP- up to 150MHz, McBSP, 16-Bit EMIFA

TMS320C6713B — C67x floating-point DSP- up to 300MHz, McBSP, 16-Bit EMIFA

TMS320C6720 — C67x floating-point DSP - 200MHz, McASP, 16-Bit EMIFA

TMS320C6722B — C67x floating-point DSP- up to 250MHz, McASP, 16-Bit EMIFA

TMS320C6726B — C67x floating-point DSP- up to 266MHz, McASP, 16-Bit EMIFA

TMS320C6727 — C67x floating-point DSP- up to 250MHz, McASP, 32-Bit EMIFA

TMS320C6727B — C67x floating-point DSP- up to 350MHz, McASP, 32-Bit EMIFA

TMS320C6742 — Low power C674x floating-point DSP- 200MHz

TMS320C6743 — Low power C674x floating-point DSP- 375MHz

TMS320C6745 — Low power C674x floating-point DSP- 456MHz, QFP

TMS320C6746 — Low power C674x floating-point DSP- 456MHz

TMS320C6747 — Low power C674x floating-point DSP- 456MHz, PBGA

TMS320C6748 — Low power C674x floating-point DSP- 456MHz, SATA

Technical documentation

No results found. Please clear your search and try again.

View all 1

	Type	Title		Date
	Technical article	Leveraging TI’s free libraries in your development	PDF \| HTML	15 Mar 2017

Support & training

TI E2E™ forums with technical support from TI engineers

View all forum topics

Content is provided "as is" by TI and community contributors and does not constitute TI specifications. See terms of use.

If you have questions about quality, packaging or ordering TI products, see TI support.

Video series

View all videos

Videos

千亿体育app官网登录(中国)官方网站IOS/安卓通用版/手机APP | 米乐app下载官网（中国）|ios|Android/通用版APP最新版 | 米乐|米乐·M6（中国大陆）官方网站 | 千亿体育登陆地址 | 华体会体育（中国）HTH·官方网站 | 千赢qy国际_全站最新版千赢qy国际V6.2.14安卓/IOS下载 | 18新利网v1.2.5|中国官方网站 | bob电竞真人（中国官网）安卓/ios苹果/电脑版【1.97.95版下载】 | 千亿体育app官方下载(中国)官方网站IOS/安卓/手机APP下载安装 |