I'm trying to run valgrind on inkscape in Ubunto 12.04LTS but
 valgrind inkscape
blows up with a segfault in libgc. Â (Inkscape without valgrind does not
segfault.)
After finding this debian bug post (and one offline communication from
you)
I tried putting suppression lines in
/usr/lib/valgrind/default.supp
Then ran it with --gen-suppressions=all to find more,
ending up finally with:
{
<gc1>
Memcheck:Cond
fun:GC_*
}
{
<gc2>
Memcheck:Value8
fun:GC_*
}
{
<gc3>
Memcheck:Value4
fun:GC_*
}
{ <gc4>
Memcheck:Addr4
fun:memset
fun:GC_stopped_mark
fun:GC_try_to_collect_inner
fun:GC_init
}
{
<gc5>
Memcheck:Addr4
fun:GC_*
}
Unfortunately at that point valgrind explodes when run, with or without
--gen-suppressions=all, like this:
e# valgrind src/inkscape
==15375== Memcheck, a memory error detector
==15375== Copyright (C) 2002-2011, and GNU GPL'd, by Julian Seward et
al.
==15375== Using Valgrind-3.7.0 and LibVEX; rerun with -h for copyright
info
==15375== Command: src/inkscape
==15375==
==15375==
==15375== Process terminating with default action of signal 11
(SIGSEGV)
==15375== Access not within mapped region at address 0xBECF2000
==15375== at 0x5439B7F: GC_push_all_eager (in
/usr/lib/libgc.so.1.0.3)
==15375== by 0x5439BD2: GC_push_all_stack (in
/usr/lib/libgc.so.1.0.3)
==15375== by 0x54422DA: GC_push_all_stacks (in
/usr/lib/libgc.so.1.0.3)
==15375== by 0x543CD73: GC_default_push_other_roots (in
/usr/lib/libgc.so.1.0.3)
==15375== by 0x543B254: GC_push_roots (in /usr/lib/libgc.so.1.0.3)
==15375== by 0x543A73D: GC_mark_some (in /usr/lib/libgc.so.1.0.3)
==15375== by 0x54314D4: GC_stopped_mark (in /usr/lib/libgc.so.1.0.3)
==15375== by 0x5438BBF: ??? (in /usr/lib/libgc.so.1.0.3)
==15375== If you believe this happened as a result of a stack
==15375== overflow in your program's main thread (unlikely but
==15375== possible), you can try to increase the size of the
==15375== main thread stack using the --main-stacksize= flag.
==15375== The main thread stack size used in this run was 8388608.
==15375==
==15375== HEAP SUMMARY:
==15375== in use at exit: 237,220 bytes in 4,876 blocks
==15375== total heap usage: 5,872 allocs, 996 frees, 304,317 bytes
allocated
==15375==
==15375== LEAK SUMMARY:
==15375== definitely lost: 0 bytes in 0 blocks
==15375== indirectly lost: 0 bytes in 0 blocks
==15375== possibly lost: 77,250 bytes in 875 blocks
==15375== still reachable: 159,970 bytes in 4,001 blocks
==15375== suppressed: 0 bytes in 0 blocks
==15375== Rerun with --leak-check=full to see details of leaked memory
==15375==
==15375== For counts of detected and suppressed errors, rerun with: -v
==15375== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 396 from
25)
Segmentation fault (core dumped)
Any thoughts on getting past this?
Thank you,
David Mathog
mat...@caltech.edu
Manager, Sequence Analysis Facility, Biology Division, Caltech
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
