• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 54 subscribers
  • Views 2113 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.

Breakpoint for NaN, +Inf, -Inf

David Luca
Expert 2795 points


Hello, 

I was wondering how can you setup in CCS a breakpoint condition to stop the program when a variable of type float becomes NaN or +Inf, -Inf?

Thank you in advance,

David.

  • 0
    TI__Guru**** 472291 points

    Hi David,

    Are you using a version of the compiler that supports NaN and Inf values correctly? What device and version of the compiler are you using?

    Thanks

    ki

  • 0 in reply to Ki
    Expert 2795 points

    Hi Ki,

    I'm using CCSv5.4 with 7.4.4 compiler for C6000, just that I would like to setup breakpoint condition MyValue==NaN in breakpoint properties and don't know if is possible. I tried that condition already and it reported some error.

    Best regards,

    David.

  • 0 in reply to David Luca
    TI__Guru**** 472291 points

    David Luca said:
    I'm using CCSv5.4 with 7.4.4 compiler for C6000

    I think you need at least 7.5 for proper support. I'll have to check on this.

    David Luca said:
    just that I would like to setup breakpoint condition MyValue==NaN in breakpoint properties and don't know if is possible. I tried that condition already and it reported some error.

    I don't think that would work. the debugger won't be able to resolve NaN... it has to be some actual value stored in memory

  • 0 in reply to Ki
    Expert 2795 points

    Hi Ki,

    Is it possible to set a float/double MyValue = NaN or MyValue = Inf? I will try tomorrow the same and then if that would work I can breakpoint on memory pattern change. I tried checking for updates but 7.5 compiler is not yet available.

    Best regards,

    David.

  • 0 in reply to David Luca
    TI__Guru**** 472291 points

    David Luca said:
    Is it possible to set a float/double MyValue = NaN or MyValue = Inf?

    I am not aware of any way to do this.

    David Luca said:
    I tried checking for updates but 7.5 compiler is not yet available.

    It has not been released yet. Looking at what was fixed in 7.5, it really pertains more to the issue where the optimizer does not  take into account NaN when comparing floating point values as described here: http://e2e.ti.com/support/development_tools/compiler/f/343/p/169199/617850.aspx