Re: pthread_key_create memory leakage in pthreads?

From: Simon Barner (barner_at_FreeBSD.org)
Date: 09/29/05

  • Next message: Billy Newsom: "Re: [PATCH] option to re-enable aggressive ATA probing"
    Date: Thu, 29 Sep 2005 20:28:36 +0200
    To: Lefteris Tsintjelis <lefty@ene.asda.gr>
    
    
    

    Lefteris Tsintjelis wrote:
    > >Can you provide a (minimal) example where this happens? Also, what is
    > >the exact command line?
    >
    > valgrind --trace-children=yes --leak-check=yes --show-reachable=yes ./sql-m

    Two other things you can try to determine whether it's you code or a
    internal error in valgrind:

    - make sure you compile and link your code with debugging information
      (-g), and at an reasonable optimization level (say -O).

    - Use valgrinds -v option to get more information.

    If you are sure, that's it's valgrind's fault, you can generate a
    suppression rule:

    valgrind -v --gen-suppressions=yes <other valgrind options> <command>

    You can put them into a file, say ~/.valgrind-suppressions, and tell
    valgrind to them with the --suppressions=<filename> switch. If you want
    to use them by default, create a ~/.valgrindrc file and put the
    appropriate options there.

    If all that doesn't help you, you're welcome to send me a copy of your
    code (in a private mail), and I'll have a look.

    -- 
    Best regards / Viele Grüße,                             barner@FreeBSD.org
     Simon Barner                                                barner@gmx.de
    
    



  • Next message: Billy Newsom: "Re: [PATCH] option to re-enable aggressive ATA probing"

    Relevant Pages

    • Re: Valgrind false alarm?
      ... Valgrind doesn't deal with C ... >> I was thinking that the definition was at file scope. ... > If the variable has file scope the loader will initialize it in either ... machine code(and the debugging information in the executables)'. ...
      (comp.unix.programmer)
    • Re: pthread_key_create memory leakage in pthreads?
      ... > Simon Barner wrote: ... >>the exact command line? ... It's probably leakage internal to valgrind, since the stack trace stops ...
      (freebsd-stable)
    • Re: pthread_key_create memory leakage in pthreads?
      ... Lefteris Tsintjelis wrote: ... > I am using valgrind to check my program but it always seems to come with ... I use a proper pthread_exit in my ... the exact command line? ...
      (freebsd-stable)