valgrind on FC2?

From: Jack Howarth (howarth_at_bromo.msbb.uc.edu)
Date: 06/22/05

  • Next message: Dr. Don Harden: "Re: ping with crossover cable" 
    To: fedora-list@redhat.com
Date: Wed, 22 Jun 2005 16:38:42 -0400 (EDT)

         Does anyone know if there are specific limitations to the
    types of binaries that can be used with valgrind? I am attempting
    to find a memory leak in a commercial scientific package we use,
    InsightII by Accelrys. After installing the valgrind 2.1.1-1.1.fc2.dag
    rpms, I tried running...

    valgrind --tool=memcheck --leak-check=yes /usr/accelrys/I2000.3L/Linux_2_Intel_32/biosym_exe/insightII

    The executable at /usr/accelrys/I2000.3L/Linux_2_Intel_32/biosym_exe/insightII
    can be used to run the program normally outside of valgrind. However when I
    attempt to run it within valgrind I get the error...

    ==8055== Memcheck, a memory error detector for x86-linux.
    ==8055== Copyright (C) 2002-2004, and GNU GPL'd, by Julian Seward.
    ==8055== Using valgrind-2.1.1, a program supervision framework for x86-linux.
    ==8055== Copyright (C) 2000-2004, and GNU GPL'd, by Julian Seward.
    ==8055== Valgrind library directory: /usr/lib/valgrind
    ==8055== Command line
    ==8055== /usr/accelrys/I2000.3L/Linux_2_Intel_32/biosym_exe/insightII
    ==8055== Startup, with flags:
    ==8055== --tool=memcheck
    ==8055== --leak-check=yes
    ==8055== -v
    ==8055== Reading syms from /usr/accelrys/I2000.3L/Linux_2_Intel_32/exe/insight/insightII (0x8048000)
    ==8055== Reading syms from /lib/ld-2.3.3.so (0x3C000000)
    ==8055== object doesn't have any debug info
    ==8055== Reading syms from /lib/ld-2.3.3.so (0xB0000000)
    ==8055== object doesn't have any debug info
    ==8055== Reading syms from /usr/lib/valgrind/vgskin_memcheck.so (0xB7CA6000)
    ==8055== Reading syms from /lib/tls/libc-2.3.3.so (0xB7ECD000)
    ==8055== object doesn't have any debug info
    ==8055== Reading syms from /lib/libdl-2.3.3.so (0xB7FEA000)
    ==8055== object doesn't have any debug info
    ==8055== Reading syms from /usr/lib/valgrind/stage2 (0xB8000000)
    ==8055== object doesn't have any debug info
    ==8055== Reading suppressions file: /usr/lib/valgrind/default.supp
    ==8055== REDIRECT soname:libc.so.6(__GI___errno_location) to soname:libpthread.so.0(__errno_location)
    ==8055== REDIRECT soname:libc.so.6(__errno_location) to soname:libpthread.so.0(__errno_location)
    ==8055== REDIRECT soname:libc.so.6(__GI___h_errno_location) to soname:libpthread.so.0(__h_errno_location)
    ==8055== REDIRECT soname:libc.so.6(__h_errno_location) to soname:libpthread.so.0(__h_errno_location)
    ==8055== REDIRECT soname:libc.so.6(__GI___res_state) to soname:libpthread.so.0(__res_state)
    ==8055== REDIRECT soname:libc.so.6(__res_state) to soname:libpthread.so.0(__res_state)
    ==8055== REDIRECT soname:libc.so.6(stpcpy) to *vgpreload_memcheck.so*(stpcpy)
    ==8055== REDIRECT soname:libc.so.6(strnlen) to *vgpreload_memcheck.so*(strnlen)
    ==8055== REDIRECT soname:ld-linux.so.2(stpcpy) to *vgpreload_memcheck.so*(stpcpy)
    ==8055== REDIRECT soname:ld-linux.so.2(strchr) to *vgpreload_memcheck.so*(strchr)
    ==8055==
    Segmentation fault (core dumped)

    I was under the impression that I didn't need to have debug versions of
    the binaries and libraries for valgrind to work. Or isn't that really
    the case?
                   Jack 

    -- 
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-list
  • Next message: Dr. Don Harden: "Re: ping with crossover cable"

    Relevant Pages

    • Re: Suspicious _init() in core dump
      ... AFAICT glibc1 predates ELF and linuxthreads, ... The reason I believe that is that there is no debug info for either ... When did you last run your exe with valgrind? ... In order to understand recursion you must first understand recursion. ...
      (comp.os.linux.development.apps)
    • Re: kadmin.local segfault gdb output (valgrind output)
      ... Very weird, when running kadmin.local under valgrind, ... it does NOT segfault. ... Julian Seward et al. ... binary translation. ...
      (comp.protocols.kerberos)
    • Re: valgrind on 5.3BETA1
      ... $ valgrind --skin=memcheck date ... There doesn't seem to be any docs on these umtx syscalls so I don't ...
      (freebsd-hackers)
    • kadmin.local segfault
      ... Very weird, when running kadmin.local under valgrind, ... it does NOT segfault. ... Julian Seward et al. ... kadmin.local: cpw test ...
      (comp.protocols.kerberos)