[Bug 1160352] Re: Still reachable leak is reported with dlopen and pthread

[Colin Watson]

Fixed independently in Debian, and making its way into saucy now via a
merge.  Thanks.

valgrind (1:3.8.1-3ubuntu1) saucy; urgency=low

  * Resynchronise with Debian.  Remaining changes:
    - Allow specification of an extra directory in which to look for
      debuginfo objects.
    - Remove valgrind-mpi package from Ubuntu and any references to it, as
      mpi-default-dev is in universe for the time being.
    - Lower over-inflated valgrind-dbg Recommends to Suggests instead.
    - Don't strip vgpreload* on ARM; this results in unusable stack traces
      without valgrind-dbg.
    - Apply patch from https://bugs.kde.org/show_bug.cgi?id=308717,
      implementing fixed-point VCVT.F64.[SU]32 on ARM.  This instruction
      shows up quite frequently in desktop components.
  * Drop remaining ppc64 patches; we no longer care about this in Ubuntu,
    and valgrind seems to build fine on Debian ppc64.
  * Renumber Ubuntu patches to make future merges easier.

 -- Colin Watson <cjwatson at ubuntu.com>  Thu, 30 May 2013 11:09:43 +0100

valgrind (1:3.8.1-3) unstable; urgency=low

  * Add 11_glibc-2.17.patch to fix FTBFS with glibc 2.17 (Closes:
#707438)

 -- Alessandro Ghedini <ghedo at debian.org>  Fri, 10 May 2013 18:10:39
+0200

valgrind (1:3.8.1-2) unstable; urgency=low

  [ Alessandro Ghedini ]
  * Bump Standards-Version to 3.9.4 (no changes needed)
  * Quote LD_LIBRARY_PATH in the wrapper script to avoid errors with spaced
    paths (LP: #880685)
  * Enable support for mipsel too (Closes: #696850)
  * Add suppression for dlopen() leak when linking with -lpthread
    (Closes: #700899)

  [ Samuel Bronson ]
  * Allow building more than once from the same tree
    - Build and ship FAQ.txt from sources, since the bundled copy is deleted
      on "make clean"
  * Drop docs/valgrind.1 changes from 08_fix-spelling-in-manpage.patch
  * Add 10_unbreak-xinclude-in-manpage.patch to fix a broken XInclude
    (Closes: #696694)

 -- Alessandro Ghedini <ghedo at debian.org>  Wed, 20 Feb 2013 11:21:07
+0100

** Bug watch added: Debian Bug tracker #700899
   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=700899

** Also affects: valgrind (Debian) via
   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=700899
   Importance: Unknown
       Status: Unknown

** Bug watch added: KDE Bug Tracking System #308717
   https://bugs.kde.org/show_bug.cgi?id=308717

** Changed in: valgrind (Ubuntu)
       Status: New => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to valgrind in Ubuntu.
https://bugs.launchpad.net/bugs/1160352

Title:
  Still reachable leak is reported with dlopen and pthread

Status in “valgrind” package in Ubuntu:
  Fix Released
Status in “valgrind” package in Debian:
  Unknown

Bug description:
  Valgrind reports still reachable leaks in dlopen()/dlclose() only when
  linked with -lpthread with the following code:

  $ cat /tmp/valgrind-dlopen-pthread-leak.c
  #include <stdio.h>
  #include <dlfcn.h>

  int main() {
    char *error = 0;

    void *handle = dlopen(NULL, RTLD_NOW | RTLD_GLOBAL);

    if((error = dlerror())) {
      printf("dlopen failed: %s\n", error);
      return 1;
    }
    printf("%p\n", handle);

    dlclose(handle);

    if((error = dlerror())) {
      printf("dlclose failed: %s\n", error);
      return 1;
    }

    return 0;
  }

  This code correctly dlopen()s and dlclose()s, so there should be no
  leak whatsoever. And this can be confirmed from the result of valgrind
  without -lpthread:

  $ gcc /tmp/valgrind-dlopen-pthread-leak.c -ldl
  $ valgrind ./a.out
  ...
  ==31371== All heap blocks were freed -- no leaks are possible
  ...

  But this doesn't hold once I compile the sample code with -lpthread:

  $ gcc /tmp/valgrind-dlopen-pthread-leak.c -ldl -lpthread
  $ valgrind ./a.out
  ...
  ==31363== LEAK SUMMARY:
  ==31363==    definitely lost: 0 bytes in 0 blocks
  ==31363==    indirectly lost: 0 bytes in 0 blocks
  ==31363==      possibly lost: 0 bytes in 0 blocks
  ==31363==    still reachable: 32 bytes in 1 blocks
  ==31363==         suppressed: 0 bytes in 0 blocks
  ...

  I've created a suppresion for this:

  $ valgrind --gen-suppressions=all --leak-check=full --show-reachable=yes ./a.out
  ...
  ==31414== 32 bytes in 1 blocks are still reachable in loss record 1 of 1
  ==31414==    at 0x4C29DB4: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
  ==31414==    by 0x4E3359F: _dlerror_run (dlerror.c:142)
  ==31414==    by 0x4E32FC0: dlopen@@GLIBC_2.2.5 (dlopen.c:88)
  ==31414==    by 0x400662: main (in /home/ryoqun/dev/rubinius/rubinius-master/a.out)
  ...
  {
     Still-reachable leak in dlopen/dlclose when linked to libpthread
     Memcheck:Leak
     fun:calloc
     fun:_dlerror_run
     fun:dlopen@@GLIBC_2.2.5
     fun:main
  }
  ...

  I don't know this should rather be fixed at libc...

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/valgrind/+bug/1160352/+subscriptions