• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 102 subscribers
  • Views 935 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

How to build DSPLIB for C64x+

Eddie3909
Guru 10385 points
Other Parts Discussed in Thread: SPRC265, MATHLIB, SPRC542

Windows 7, CGT 7.4.4 (also tried CGT 7.2.6), C64x+, CCSv5.5

I downloaded DSP lib 3.4 from the following directory for the C64x+

http://www.ti.com/tool/sprc265

and followed the instructions to do the GNU Makefile method.

file:///C:/ti/dsplib_c64Px_3_4_0_0/docs/doxygen/html/gnu_build.html

The instructions said I had to change the tool defined in the makefile as shown below.

# (Mandatory) Specify where various tools are installed.

# C6X_GEN_INSTALL_DIR - C6x Code gen tools install directory
export C6X_GEN_INSTALL_DIR ?= C:\ti\C6000 Code Generation Tools 7.2.6

# PATH TO MATHLIB INSTALL DIR
export MATHLIB_INSTALL_DIR ?= c:/ti/mathlib_c66x_3_1_0_0

# PATH TO DSPLIB INSTALL DIR
export DSPLIB_INSTALL_DIR ?= c:/nightlybuilds/dsplib

# ROOT Directory
export ROOTDIR := ../..

But - this is going sound really dumb - I can't find the C64x+ mathlib in the wiki. I don't see it in the mathlib download page linked from here

http://processors.wiki.ti.com/index.php?title=Software_libraries

Could you point me to it? I can only seem to find the C66x and C67x math libraries.

FYI - for those who are interested in trying the CCS GUI build

http://e2e.ti.com/support/development_tools/code_composer_studio/f/81/p/367126/1292041.aspx#1292041

Cheers2u

Eddie

  • 0
    TI__Mastermind 35645 points

    Hi,

    Thanks for your post.

    Basically, MATHLIB is not a fixed point computation library, so, obviously, you cannot see math libraries for fixed point c64x+ core devices. May be, you could use the ELF version of the C64X+ DSP libraries (0042.ELF_libraries.zip) as attached by Rahul prabhu referred in the below E2E post:

    http://e2e.ti.com/support/dsp/tms320c6000_high_performance_dsps/f/112/t/119869.aspx

    Also, you could try for C64x+ IQmath Library which is a collection of highly optimized and high precision mathematical Function Library for C/C++ programmers to seamlessly port the floating-point algorithm into fixed point code on C64x+ devices. Please refer the below wiki source to download the IQmathlib:

    http://processors.wiki.ti.com/index.php/C64x_IQmath

    http://processors.wiki.ti.com/index.php/IQmath_Library_for_C64x%2B

    http://www.ti.com/tool/sprc542

    Thanks & regards,

    Sivaraj K

    ------------------------------------------------------------------------------------------------------

    Please click the Verify Answer button on this post if it answers your question

    -------------------------------------------------------------------------------------------------------

     

  • 0 in reply to Sivaraj Kuppuraj
    Guru 10385 points

    Hi Sivaraj

    What I'd really like to do at the end of this exercise is have the stock dsp lib built for the C64x+. So it concerns me that you are saying to try this or try that. Wasn't there a stock mathlib used to build the dsplib files that were in the zip I downloaded for dsplib 3.4.0.0 in the first place?

    I think it would be safest to use the mathlib that built the stock dsplib 3.4.0.0 release ......if possible.

    Cheers

    Eddie