Red Hat Bugzilla – Bug 176090
Segfault on startup of fish
Last modified: 2007-11-30 17:11:19 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-GB; rv:1.8) Gecko/20051111 Firefox/1.5
Description of problem:
Segfault on startup. After installing the debuginfo packages, gdb shows:
Starting program: /usr/bin/fish
Detaching after fork from child process 10556.
Program received signal SIGSEGV, Segmentation fault.
*__GI_wcscmp (s1=0x53f370, s2=0x7fff00000000) at wcscmp.c:36
36 c2 = (wint_t) *s2++;
#0 *__GI_wcscmp (s1=0x53f370, s2=0x7fff00000000) at wcscmp.c:36
#1 0x0000000000406bcd in contains_str (a=0x53f370) at common.c:722
#2 0x000000000040f421 in parser_is_block (word=Variable "word" is not available.
) at parser.c:289
#3 0x00000000004111d5 in parse_job (p=0x53f320, j=0x53f600, tok=0x53f110) at parser.c:347
#4 0x00000000004117dc in eval (cmd=Variable "cmd" is not available.
) at parser.c:1666
#5 0x0000000000416a95 in reader_read () at reader.c:2979
#6 0x00000000004045ee in builtin_source (argv=0x53e430) at builtin.c:1811
#7 0x000000000040bd9c in exec (j=0x53e340) at exec.c:891
#8 0x0000000000411a4d in eval (cmd=Variable "cmd" is not available.
) at parser.c:1585
#9 0x000000000041e294 in main (argc=1, argv=0x7fffffc16bc8) at main.c:74
#10 0x00000031b471c3cf in __libc_start_main (main=0x41e0a0 <main>, argc=1, ubp_av=0x7fffffc16bc8,
init=0x41e4c0 <__libc_csu_init>, fini=Variable "fini" is not available.
) at ../sysdeps/generic/libc-start.c:216
#11 0x0000000000402db9 in _start ()
#12 0x00007fffffc16bb8 in ?? ()
#13 0x0000000000000000 in ?? ()
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Install fish - yum install fish
2. Run fish within gdb - gdb `which fish`
3. Run - gdb> run
This bug is fixed in later fish releases, Fedora is lagging behind the main fish
You can temporarily fix this by downloading an RPM from the fish homepage at
Hopefully, the Fedora packaging of fish will be better in synch in the future.
(volounteer packagers are welcomed)
I took the SRPM from the website and built rpms that way. I can comfirm that
they work perfectly. Should I resolve the bug? If so, to what?
You wouldn't happen to feel like maintaining fish on Fedora? In that case, you
could simply update the fish on fedora and mark the bug as fixed. Otherwise, I
guess the bug isn't resolved and should be left as is.
What would be involved with maintaining fish in Extras? I only have an AMD64
box, so my ability to test would be a little limited.
> Fedora is lagging behind the main fish releases.
You are the package maintainer, aren't you?