|
From: Duane G. <d.g...@ps...> - 2005-03-11 17:00:21
|
Hi folks,=20 2.4.0.rc2 crashes trying to read the debugging symbols from one of our libraries. On the other hand, so does 2.2, so I doubt it is a regression :) To be honest, I'm not surprised it is falling over. We are running on a very old RedHat 7.2 system, and the library that crashes on includes C and C++ code compiled with gcc 2.96, plus FORTRAN code compiled by an old version (3.2-4) of the PGI compiler. On the other hand the PGI compiler is supposedly generating GNU debugging info, and gdb doesn't seem to mind it. The error is the same with or without the --workaround-gcc296-bugs flag. If I run with the release version of the library, i.e. without debugging information, it all seems to work fine. > uname -a Linux cucumbertree.psenterprise.com 2.4.9-13SGI_XFS_1.0.2smp #1 SMP Thu Nov 15 14:05:48 CST 2001 i686 unknown > valgrind -v --workaround-gcc296-bugs=3Dyes gPROMSconsoled.exe =3D=3D5130=3D=3D Memcheck, a memory error detector for x86-linux. =3D=3D5130=3D=3D Copyright (C) 2002-2005, and GNU GPL'd, by Julian = Seward et al. =3D=3D5130=3D=3D Using valgrind-2.4.0.rc2, a program supervision = framework for x86-linux. =3D=3D5130=3D=3D Copyright (C) 2000-2005, and GNU GPL'd, by Julian = Seward et al. =3D=3D5130=3D=3D Valgrind library directory: /usr/local/lib/valgrind =3D=3D5130=3D=3D Command line =3D=3D5130=3D=3D gPROMSconsoled.exe =3D=3D5130=3D=3D Startup, with flags: =3D=3D5130=3D=3D -v =3D=3D5130=3D=3D --workaround-gcc296-bugs=3Dyes =3D=3D5130=3D=3D Contents of /proc/version: =3D=3D5130=3D=3D Linux version 2.4.9-13SGI_XFS_1.0.2smp (root@gibble) = (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #1 SMP Thu Nov 15 14:05:48 CST 2001 =3D=3D5130=3D=3D Reading syms from /home/duaneg/projects/trunk/src/build/LINUX/debug/bin/gPROMSconsoled.exe (0x8048000) =3D=3D5130=3D=3D Reading syms from /lib/ld-2.2.4.so (0x1B8E4000) =3D=3D5130=3D=3D Reading syms from /usr/local/lib/valgrind/stage2 = (0xB0000000) =3D=3D5130=3D=3D Reading syms from /lib/ld-2.2.4.so (0xB1000000) =3D=3D5130=3D=3D Reading syms from /lib/libdl-2.2.4.so (0xB1032000) =3D=3D5130=3D=3D Reading syms from /lib/i686/libc-2.2.4.so (0xB1036000) =3D=3D5130=3D=3D Reading syms from = /usr/local/lib/valgrind/vgskin_memcheck.so (0xB1300000) =3D=3D5130=3D=3D Reading suppressions file: = /usr/local/lib/valgrind/default.supp =3D=3D5130=3D=3D =3D=3D5130=3D=3D Reading syms from /usr/local/lib/valgrind/vg_inject.so (0x1B8FE000) =3D=3D5130=3D=3D Reading syms from = /usr/local/lib/valgrind/vgpreload_memcheck.so (0x1B901000) =3D=3D5130=3D=3D Reading syms from /home/duaneg/projects/trunk/src/build/LINUX/debug/lib/libgLinkedMsgFiled .so (0x1B909000) =3D=3D5130=3D=3D Reading syms from /home/duaneg/projects/trunk/src/build/LINUX/debug/lib/libInterruptCheckd .so (0x1B922000) =3D=3D5130=3D=3D Reading syms from /home/duaneg/projects/trunk/src/build/LINUX/debug/lib/libgSERVER3d.so (0x1B928000) =3D=3D5130=3D=3D warning: mmap failed on /home/duaneg/projects/trunk/src/build/LINUX/debug/lib/libgSERVER3d.so =3D=3D5130=3D=3D no symbols or debug info loaded =3D=3D5130=3D=3D Reading syms from /home/duaneg/projects/trunk/src/build/LINUX/debug/lib/libgSERVERd.so (0x1C5EF000) =20 @@ expected ';' at FP-TYPE extra (remains=3D"double = precision:t7=3DR2;8;") @@ parsing 6=3DR1;4; gave NULL type (R1;4; remains) --5130-- INTERNAL ERROR: Valgrind received a signal 11 (SIGSEGV) - exiting --5130-- si_code=3D1 Fault EIP: 0xB003F013; Faulting address: 0x0 --5130-- esp=3D0xB03FEBE8 =20 =20 valgrind: the `impossible' happened: Killed by fatal signal Basic block ctr is approximately 9330 =3D=3D5130=3D=3D at 0xB003F013: vgPlain_st_basetype = (vg_symtypes.c:423) =3D=3D5130=3D=3D by 0xB003D357: initSym (vg_stabs.c:1079) =3D=3D5130=3D=3D by 0xB003DE8F: vgPlain_read_debuginfo_stabs (vg_stabs.c:1655) =3D=3D5130=3D=3D by 0xB00383F0: vg_read_lib_symbols = (vg_symtab2.c:1485) =3D=3D5130=3D=3D by 0xB00385F6: vgPlain_read_seg_symbols = (vg_symtab2.c:1564) =3D=3D5130=3D=3D by 0xB002E2F5: vgPlain_map_file_segment = (vg_memory.c:408) =3D=3D5130=3D=3D by 0xB002E43D: vgPlain_map_fd_segment = (vg_memory.c:465) =3D=3D5130=3D=3D by 0xB003FFCD: mmap_segment (vg_syscalls.c:192) =3D=3D5130=3D=3D by 0xB004B431: vgArch_gen_old_mmap_before (vg_syscalls.c:4339) =3D=3D5130=3D=3D by 0xB0050D65: vgPlain_client_syscall = (vg_syscalls.c:6138) =3D=3D5130=3D=3D by 0xB0017139: handle_syscall (vg_scheduler.c:632) =3D=3D5130=3D=3D by 0xB0017451: vgPlain_scheduler = (vg_scheduler.c:733) =20 sched status: running_tid=3D1 =20 Thread 1: status =3D VgTs_Runnable =3D=3D5130=3D=3D at 0x1B8F66ED: mmap (in /lib/ld-2.2.4.so) =3D=3D5130=3D=3D by 0x1B8EBC28: _dl_map_object (dl-load.c:1740) =3D=3D5130=3D=3D by 0x1B8F11AB: openaux (dl-deps.c:68) =3D=3D5130=3D=3D by 0x1B8F1712: _dl_catch_error (dl-error.c:152) =3D=3D5130=3D=3D by 0x1B8F0677: _dl_map_object_deps (dl-deps.c:248) =3D=3D5130=3D=3D by 0x1B8E6732: dl_main (rtld.c:881) =3D=3D5130=3D=3D by 0x1B8F4006: _dl_sysdep_start (../sysdeps/generic/dl-sysdep.c:207) =3D=3D5130=3D=3D by 0x1B8E640E: _dl_start_final (rtld.c:262) =3D=3D5130=3D=3D by 0x1B8E6202: _dl_start (rtld.c:214) =3D=3D5130=3D=3D by 0x1B8E5E65: (within /lib/ld-2.2.4.so) Cheers,=20 Duane. _____________________________________________________________ Duane Griffin Senior Software Developer Process Systems Enterprise Limited - "The model company" Bridge Studios, 107a Hammersmith Bridge Road, London W6 9DA Call +44 (0)20 8563 0888 or visit us at www.psenterprise.com |
|
From: Tom H. <to...@co...> - 2005-03-11 18:38:21
|
In message <E0D...@pe...>
"Duane Griffin" <d.g...@ps...> wrote:
> 2.4.0.rc2 crashes trying to read the debugging symbols from one of our
> libraries. On the other hand, so does 2.2, so I doubt it is a regression
> :)
Can you enter a bug for this please.
Tom
--
Tom Hughes (to...@co...)
http://www.compton.nu/
|
|
From: Jeremy F. <je...@go...> - 2005-03-12 00:44:13
|
Tom Hughes wrote:
>In message <E0D...@pe...>
> "Duane Griffin" <d.g...@ps...> wrote:
>
>
>
>>2.4.0.rc2 crashes trying to read the debugging symbols from one of our
>>libraries. On the other hand, so does 2.2, so I doubt it is a regression
>>:)
>>
>>
>
>Can you enter a bug for this please.
>
Are you planning on looking at this? Given that this bug seems old and
hasn't caused too many problems, I'm not inclined to hold up the 2.4
release pending its resolution, but if there's a simple obviously
correct fix (as much as anything in the stabs parser can be) it would be
nice to get in.
J
|
|
From: Tom H. <to...@co...> - 2005-03-12 07:36:09
|
In message <423...@go...>
Jeremy Fitzhardinge <je...@go...> wrote:
> Tom Hughes wrote:
>
> >In message <E0D...@pe...>
> > "Duane Griffin" <d.g...@ps...> wrote:
> >
> >>2.4.0.rc2 crashes trying to read the debugging symbols from one of our
> >>libraries. On the other hand, so does 2.2, so I doubt it is a regression
> >>:)
> >
> >Can you enter a bug for this please.
>
> Are you planning on looking at this? Given that this bug seems old and
> hasn't caused too many problems, I'm not inclined to hold up the 2.4
> release pending its resolution, but if there's a simple obviously
> correct fix (as much as anything in the stabs parser can be) it would be
> nice to get in.
I put a patch u plast night, although I'm worried it might break
with stabs from other compilers. I agree this probably shouldn't
hold up the release.
--
Tom Hughes (to...@co...)
http://www.compton.nu/
|
|
From: Jeremy F. <je...@go...> - 2005-03-12 08:28:57
|
Tom Hughes wrote:
>I put a patch u plast night, although I'm worried it might break
>with stabs from other compilers. I agree this probably shouldn't
>hold up the release.
>
>
Looks fine to me; I just checked it in. We'll see if it breaks someone
else. I'm not sure where the 3rd parameter to R came from.
J
|