|
From: Rob <ca...@cs...> - 2005-06-28 19:57:51
|
Hi, I've used valgrind for a long time w/o significant problem. However, since I've upgraded to RedHat FC3 (glibc-2.3.3), when I drop to a debugger, none of the symbols resolve correctly. I upgraded to valgrind 2.4.0 (from 2.2.0) and I have the same problem. When I invoke gdb directly on the code, it finds all the symbols without problem. Also, before valgrind attaches gdb, it can obviously read the symbols, b/c it says "you have a problem at foo.c:152". Appologies if this is a known problem... I checked the FAQ and the last couple of months of the archives. Anyone have any ideas? Running with 2.4.0, RedHat FC3, glibc-2.3.3 and `gdb --version` reports "GNU gdb Red Hat Linux (6.1post-1.20040607.41rh)". Thanks in advance, - Rob . ----- capveg@grr:~/nicework/fileswarm> val treesim -config dunnoconfig ==21195== Memcheck, a memory error detector for x86-linux. ==21195== Copyright (C) 2002-2005, and GNU GPL'd, by Julian Seward et al. ==21195== Using valgrind-2.4.0, a program supervision framework for x86-linux. ==21195== Copyright (C) 2000-2005, and GNU GPL'd, by Julian Seward et al. ==21195== For more details, rerun with: -v ==21195== [program output] ==21195== Conditional jump or move depends on uninitialised value(s) ==21195== at 0x25A9AE3F: setup_community (DunnoPeer.c:696) ==21195== by 0x25A9ACF5: recv_leaf_fyi (DunnoPeer.c:637) ==21195== by 0x25A9A234: recv (DunnoPeer.c:292) ==21195== by 0x2598B677: recvqueueDequeue (xmitqueue.c:145) ==21195== by 0x8048BDE: main (main.c:48) ==21195== ==21195== ---- Attach to debugger ? --- [Return/N/n/Y/y/C/c] ---- y starting debugger ==21195== starting debugger with cmd: /usr/bin/gdb -nw /proc/21196/fd/1015 21196 GNU gdb Red Hat Linux (6.1post-1.20040607.41rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-redhat-linux-gnu"...Using host libthread_db library "/lib/tls/libthread_db.so.1". Attaching to program: /proc/21196/fd/1015, process 21196 0x25a9ae3f in ?? () (gdb) where #0 0x25a9ae3f in ?? () #1 0x259920a8 in ?? () #2 0x259920a8 in ?? () #3 0x70dfe448 in ?? () #4 0x2598ad35 in ?? () #5 0x259920a8 in ?? () #6 0x25999bb9 in ?? () #7 0x2599aea1 in ?? () #8 0x00000000 in ?? () - Rob . |
|
From: Rob <ca...@cs...> - 2005-07-01 05:05:56
|
On Tue, Jun 28, 2005 at 03:57:44PM -0400, Rob wrote: > Hi, > > I've used valgrind for a long time w/o significant problem. However, since > I've upgraded to RedHat FC3 (glibc-2.3.3), when I drop to a debugger, > none of the symbols resolve correctly. I upgraded to valgrind 2.4.0 > (from 2.2.0) and I have the same problem. When I invoke gdb directly > on the code, it finds all the symbols without problem. Also, before > valgrind attaches gdb, it can obviously read the symbols, b/c it says > "you have a problem at foo.c:152". > > Appologies if this is a known problem... I checked the FAQ and the last > couple of months of the archives. > > Anyone have any ideas? Running with 2.4.0, RedHat FC3, glibc-2.3.3 > and `gdb --version` reports "GNU gdb Red Hat Linux (6.1post-1.20040607.41rh)". Fwiw, I was able to resolve this by building my own gdb from source, and then bulding my own valgrind to use the new gdb. Is it known that RedHat's gdb has a bad interaction with valgrind or is it something specific to my system? Cheers, - Rob . |
|
From: Nicholas N. <nj...@cs...> - 2005-07-01 12:59:20
|
On Fri, 1 Jul 2005, Rob wrote: > Fwiw, I was able to resolve this by building my own gdb from source, > and then bulding my own valgrind to use the new gdb. Is it known that > RedHat's gdb has a bad interaction with valgrind or is it something > specific to my system? No idea, sorry. N |
|
From: Kjartan M. <km...@br...> - 2005-07-01 16:04:05
|
fre, 01,.07.2005 kl. 01.05 -0400, skrev Rob: > On Tue, Jun 28, 2005 at 03:57:44PM -0400, Rob wrote: > > Hi, > > > > I've used valgrind for a long time w/o significant problem. However, since > > I've upgraded to RedHat FC3 (glibc-2.3.3), when I drop to a debugger, > > none of the symbols resolve correctly. I upgraded to valgrind 2.4.0 > > (from 2.2.0) and I have the same problem. When I invoke gdb directly > > on the code, it finds all the symbols without problem. Also, before > > valgrind attaches gdb, it can obviously read the symbols, b/c it says > > "you have a problem at foo.c:152". > > > > Appologies if this is a known problem... I checked the FAQ and the last > > couple of months of the archives. > > > > Anyone have any ideas? Running with 2.4.0, RedHat FC3, glibc-2.3.3 > > and `gdb --version` reports "GNU gdb Red Hat Linux (6.1post-1.20040607.41rh)". > > Fwiw, I was able to resolve this by building my own gdb from source, > and then bulding my own valgrind to use the new gdb. Is it known that > RedHat's gdb has a bad interaction with valgrind or is it something > specific to my system? > I've seen this myself the last couple of Fedora releases at least. Maybe filing a bug in Red Hat's bug tracker is a good thing to do... Cheers Kjartan |