hi,

i'm currently testing current gcc-4.4.4(svn) on the large c++ application.
the application uses few threads, contains over 100+ shared components
and the tested debug build (-O1 -g2 -fno-inline) has over 2GB of disk space,
so creating a reduced testcase at this point isn't trivial thing :(

running application compiled with gcc-4.4 trigs abort() in glibc malloc
checker. in the other side gcc-4.3.5 and msvc8 builds work fine.

i've tried to valgrind the application ;)

# valgrind --leak-check=no --read-var-info=yes --track-origins=yes ...

==23309== Memcheck, a memory error detector
==23309== Copyright (C) 2002-2009, and GNU GPL'd, by Julian Seward et al.
==23309== Using Valgrind-3.5.0 and LibVEX; rerun with -h for copyright info
==23309== Command: /home/users/(...)
==23309== Parent PID: 23305
==23309== 
--23309-- warning: addVar: in range 0x1fa2 .. 0x28ff outside segment 0x43a6d0 
.. 0x48beb5 (l_func_4623)
--23309-- warning: addVar: in range 0x2911 .. 0x292e outside segment 0x43a6d0 
.. 0x48beb5 (l_func_4623)
--23309-- warning: addVar: in range 0x1fa2 .. 0x28ff outside segment 0x43a6d0 
.. 0x48beb5 (l_4628counters)
--23309-- warning: addVar: in range 0x2911 .. 0x292e outside segment 0x43a6d0 
.. 0x48beb5 (l_4628counters)
--23309-- warning: addVar: in range 0x29dd .. 0x344e outside segment 0x43a6d0 
.. 0x48beb5 (l_4630bufg)
--23309-- warning: addVar: in range 0x345d .. 0x3477 outside segment 0x43a6d0 
.. 0x48beb5 (l_4630bufg)
--23309-- warning: addVar: in range 0x29dd .. 0x344e outside segment 0x43a6d0 
.. 0x48beb5 (l_4638counter)
--23309-- warning: addVar: in range 0x345d .. 0x3477 outside segment 0x43a6d0 
.. 0x48beb5 (l_4638counter)
==23309== Conditional jump or move depends on uninitialised value(s)
==23309==    at 0x400ABEE: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309==    by 0x40039F2: dl_main (in /lib64/ld-2.11.1.so)
==23309==    by 0x4014780: _dl_sysdep_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4001452: _dl_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4000B27: ??? (in /lib64/ld-2.11.1.so)
==23309==  Uninitialised value was created by a stack allocation
==23309==    at 0x400AA90: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309== 
==23309== Conditional jump or move depends on uninitialised value(s)
==23309==    at 0x400ABF7: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309==    by 0x40039F2: dl_main (in /lib64/ld-2.11.1.so)
==23309==    by 0x4014780: _dl_sysdep_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4001452: _dl_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4000B27: ??? (in /lib64/ld-2.11.1.so)
==23309==  Uninitialised value was created by a stack allocation
==23309==    at 0x400AA90: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309== 
==23309== Conditional jump or move depends on uninitialised value(s)
==23309==    at 0x400ABEE: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309==    by 0x40038B2: dl_main (in /lib64/ld-2.11.1.so)
==23309==    by 0x4014780: _dl_sysdep_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4001452: _dl_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4000B27: ??? (in /lib64/ld-2.11.1.so)
==23309==  Uninitialised value was created by a stack allocation
==23309==    at 0x400AA90: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309== 
==23309== Conditional jump or move depends on uninitialised value(s)
==23309==    at 0x400ABF7: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309==    by 0x40038B2: dl_main (in /lib64/ld-2.11.1.so)
==23309==    by 0x4014780: _dl_sysdep_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4001452: _dl_start (in /lib64/ld-2.11.1.so)
==23309==    by 0x4000B27: ??? (in /lib64/ld-2.11.1.so)
==23309==  Uninitialised value was created by a stack allocation
==23309==    at 0x400AA90: _dl_relocate_object (in /lib64/ld-2.11.1.so)
==23309== 

valgrind: m_debuginfo/d3basics.c:696 (vgModuleLocal_evaluate_GX): Assertion 
'aMax == ~(Addr)0' failed.
==23309==    at 0x3802AC14: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x3802AE87: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x38095BA4: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x3804F94F: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x380508C7: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x38024EDB: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x380251B3: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x380284D0: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x3802556C: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x38006865: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x38089C13: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x380650C1: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x38061CCD: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x38062B29: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)
==23309==    by 0x3808AD14: ??? (in /usr/lib64/valgrind/memcheck-amd64-linux)

sched status:
  running_tid=1

Thread 1: status = VgTs_Runnable
==23309==    at 0x5ABB34E: __libc_sigaction (in /lib64/libpthread-2.11.1.so)
==23309==    by 0x5824093: (anonymous namespace)::registerSignalFilter() 
(eh.cpp:428)
==23309==    by 0x586FFF5: ??? (in /home/users/(...)/build-release-x86_64-gnu-
linux/libeh.so)
==23309==    by 0x5820C9A: ??? (in /home/users/(...)/build-release-x86_64-gnu-
linux/libeh.so)


it looks like a heisenbug which also crashes valgrind.
could you please help me with tracking this problem?

thanks in advance,
Pawel.

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Valgrind-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/valgrind-users

Reply via email to