• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 54 subscribers
  • Views 254 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.

Linux CCSv5.1 M7 Variables windows has wrong data...

Joe Cossette
Intellectual 780 points

Again, like prior posts...this post continues this thread from prior posts with the same subject (except different revisions for M5 and M6).  Basically, the same problem appears to exist and no satisfactory solutions or answers for this issue.

On my last post on this subject (for M6 revision) I didn't get any reply from TI folks in regards to my assertions that the existing documentation was "outdated".

Upon further pondering I'm not sure that this is a "Linux OS debug" issue at all.  It appears more like a toolchain issue because the TI debugger doesn't appear to be getting any symbolic info correct inasfar as data is concerned.  As the previous post shows, the debugger should have all the symbolic info it needs to resolve the variable "i" to it's proper value (look at the disassembly).  But it does not get it correct.

Hopefully, we'll get a little more action on this issue...

  • 0
    TI__Intellectual 2625 points

    Hi Joe,

    In your last posts, you were debugging a Linux kernel module.

    You were loading the kernel module symbol file by using add-symbol.

    Are you still using this procedure to load the module symbol file ?

    I do not think that this will load the symbol properly.

    I can have a webex with you to show you the proper way to load the module symbol.

    Regards,

    Raymond

  • 0 in reply to Raymond Pang
    Intellectual 780 points

    I am still using this procedure to load the symbols for Linux Kernel Modules...I don't know of any other procedure...under GDB this procedure works fine, admittedly, GDB takes parameters for all sections, not just  the code and data sections as the CCS dialog only accepts.  If you've got a better way of doing this that might fix this issue then I'm all ears...;)...

    I would like to get this problem cleared up once and for all...but if you remember from our past webex sessions...the system I'm running on doesn't have a proper JVM for hosting webex sessions...I've tried a couple of different JVM's, they all seem to work Ok when you test them against some web sites that purport to be able to tell if you've got java installed correctly, but when I run them in a webex session it doesn't work properly...I'm running 32-bit Ubuntu 10.04 LTS.