|
From: Scott M. <sm...@no...> - 2005-03-30 18:28:49
|
Hi all, I'm running Valgrind 2.4.0 - Memcheck (liking the new version so far!) but getting an assertion error on a certain process.. a snippet from the log is below. Any clue what the cause could be? First time I've seen an assertion error! Thx, Scott. ==6917== Reading syms from <myprocessname>.so.1.0.0 (0x1BEAA000) ==6917== badly formed extended line op encountered! ==6917== badly formed extended line op encountered! ==6917== badly formed extended line op encountered! ==6917== badly formed extended line op encountered! valgrind: vg_symtab2.c:248 (vgPlain_addLineInfo): Assertion `lineno >= 0' failed. ==6917== at 0xB0033A87: vgPlain_skin_assert_fail (vg_mylibc.c:1170) ==6917== by 0xB0033A86: assert_fail (vg_mylibc.c:1166) ==6917== by 0xB0033AF4: vgPlain_core_assert_fail (vg_mylibc.c:1177) ==6917== by 0xB003941F: vgPlain_addLineInfo (vg_symtab2.c:189) ==6917== by 0xB003FB19: vgPlain_read_debuginfo_dwarf2 (vg_dwarf.c:422) ==6917== by 0xB003BB85: vg_read_lib_symbols (vg_symtab2.c:1490) ==6917== by 0xB003BDAC: vgPlain_read_seg_symbols (vg_symtab2.c:1564) ==6917== by 0xB0030483: vgPlain_map_file_segment (vg_memory.c:408) ==6917== by 0xB00305F8: vgPlain_map_fd_segment (vg_memory.c:465) ==6917== by 0xB0044932: mmap_segment (vg_syscalls.c:192) ==6917== by 0xB0055B71: vgArch_gen_old_mmap_before (vg_syscalls.c:4417) ==6917== by 0xB005DFB7: vgPlain_client_syscall (vg_syscalls.c:6234) ==6917== by 0xB0015E68: handle_syscall (vg_scheduler.c:649) ==6917== by 0xB001618B: vgPlain_scheduler (vg_scheduler.c:750) ==6917== by 0xB008F0E5: vgArch_thread_wrapper (core_os.c:69) |
|
From: Jeremy F. <je...@go...> - 2005-03-31 21:25:42
|
Scott Mills wrote:
> Hi all,
>
> I'm running Valgrind 2.4.0 - Memcheck (liking the new version so far!)
> but getting an assertion error on a certain process.. a snippet from
> the log is below.
>
> Any clue what the cause could be? First time I've seen an assertion error!
>
What compiler generated this debug output? The base problem is either
that the compiler generated bad debug info, or Valgrind is parsing it
wrongly. This ends up passing a negative line-number, which doesn't
make much sense. I think the dwarf parser is just getting derailed;
it's a bit hard to tell what's going on.
Can you post the output of "readelf -wl <yourfile>.so.1.0.0"
(compressed, if its huge)?
J
|