• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 53 subscribers
  • Views 619 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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/CCSTUDIO-C2000: Headles Commands not correctly working in CCS 6.2(linux)

Ozino Odharo
Mastermind 20662 points

Part Number: CCSTUDIO-C2000

Tool/software: Code Composer Studio

Hi,

I am using headless commands to import, clean, and build projects in CCS 6.2 on a linux machine. There sets of projects that have inconsistent build times. For instance, a set that would normally take about 2.5 hours(119 projects) to build recently took 30 mins. These are the commands that I am currently using to perform the actions:

-os linux -ws gtk -arch x86_64 -application com.ti.ccstudio.apps.importProject -data <path_to_workspace> -ccs.location <path_to_project> -ccs.overwrite
-os linux -ws gtk -arch x86_64 -application com.ti.ccstudio.apps.buildProject -data <path_to_workspace > -ccs.projects <project_name>  -ccs.configuration <name> -ccs.clean
-os linux -ws gtk -arch x86_64 -application com.ti.ccstudio.apps.buildProject -data  <path_to_workspace> -ccs.projects <project_name> -ccs.configuration <name> -ccs.buildType full

I am wondering what is causing the inconsistent build times and whether the projects are actually getting built. I have previously used these commands in CCSv6.1. Not sure if the headless commands are up to date for CCS 6.2.

Additionally, I also keep having this error appear anytime I run theses commands.

xvfb-run: error: Xvfb failed to start

Regards,

Ozino

  • 0
    TI__Guru**** 464391 points
    Hi Ozino,
    Can your provide the full build output (preferably streamed to a file) for the 6.2 build?

    Thanks
    ki
  • 0 in reply to Ki
    TI__Mastermind 20662 points

    Hi Ki,

    I was able to figure out what was wrong with the test harness. It turns out that the problem was with headless command. It seems that CCS had some sort of a virtual screen that got locked up at certain points during the testing process. The solution I found  is to use the -a option when calling the xvfb-run command. This allows CCS to autoselect the virtual screen whenever running certain commands. After adding this option to the calls to the build, import, and clean headless commands, I am no longer seeing the error (XVFB Fatal server error: Server is already active for display 99 issue).

    Thanks,

    Ozino