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.

CCS: ccs10.1 compilation speed is very slow

Tool/software: Code Composer Studio

Hi,Experts,

we using ccs10.1 to compiler our code,the compile speed is very slow!! Has anyone encountered this problem?

System environment: windows10.

We did not use the ccs IDE to compile, we made our own makefile and directly called ti-cgt-arm_20.2.1.LTS to compile our code. It takes an average of one minute to compile a .c file.

How to solve this problem?

Regards,

Xie

  • For one file that takes a long time to compile, please follow the directions in the article How to Submit a Compiler Test Case.

    Thanks and regards,

    -George

  • Hi,George

    We used to compile very quickly with the same set of code that we used for Windows 7 and CCS10.0, but now compile very slowly with Windows 10 and CCS10.1.

    sorry I can't support own source files.

    Regards,

    Xie

  • Hi,George

    Below are our compiler options:

    [Project_info]
    compiler_opt=--common=off --asm_listing --embed_inline_assembly -DMAKEFILE_BUILD -Dxdc__nolocalnames -Dj7200_evm=j7200_evm -DIPC_SUPPORT_SCICLIENT -DAUTOSAR_421 -D__CC_ARM -DBUILD_MCU -DBUILD_MCU1_0 -DSBL_LOG_LEVEL=3 -DSOC_J7200 -DUART_ENABLED -DOSAL_DEBUG --code_state=32 --diag_warning=225 --diag_wrap=off --opt_for_speed=2 --program_level_compile --display_error_number -ms --c99 -O0 -g --preproc_with_compile --float_support=vfpv3d16 --endian=little -mv7R5 --abi=eabi
    link_opt=-qq -mv7R5 --diag_warning=225 --diag_wrap=off --display_error_number --warn_sections --diag_suppress=10063 --diag_suppress=10068 --diag_suppress=23000 -x --zero_init=on --define=BUILD_MCU -i"@ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\lib" -i"@ccs/tools/compiler/ti-cgt-arm_20.2.1.LTS/include" --rom_model
    link_opt_m=
    buildpath=\Customer
    compiler_path=D:\ti\ccs1011
    chip=TI
    mcu=j7200_evm
    build_toolchain=ARM
    build_thread=10
    compiler_tool=.\ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\bin\armcl.exe
    third_tool=.\ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\bin\armcl.exe
    third_opt=.\Customer\Build\Link\linker_r5.lds -llibc.a
    burn_tool=.\ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\bin\armhex.exe
    burn_opt= -memwidth=8 --intel -o
    debug_tool=.\ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\bin\armcl.exe
    debug_opt=
    link_tool=.\ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\bin\armcl.exe
    assembler=.\ccs\tools\compiler\ti-cgt-arm_20.2.1.LTS\bin\armar.exe
    assembler_opt=rq
    target=.\bin\

    Regards,

    Xie

  • Remove the option --program_level_compile. When used in a make file that builds one file at a time, it has no impact on optimization.  It also slows down the compiler, though not by much.  I would be surprised if removing it solves the problem.

    A better way to achieve the whole program optimization is with the option --opt_level=4.  To learn more about it, please search the TI ARM compiler manual for the sub-chapter titled Link-Time Optimization.  Because this option does slow down the compiler, you probably want to delay experimenting with it.

    Linda Xie said:
    We used to compile very quickly with the same set of code that we used for Windows 7 and CCS10.0

    What version of the compiler builds so quickly?

    Linda Xie said:
    sorry I can't support own source files.

    Unfortunately, that is the only way to resolve your problem.  Are you willing to send the files to me privately?

    Thanks and regards,

    -George