On Thu, Jun 26, 2014 at 4:01 PM, Andy Goth <[email protected]> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Was trying to run find, eventually got this:
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x00007ffff6574230 in __strncmp_ssse3 () from /lib64/libc.so.6
> (gdb) ba
> #0 0x00007ffff6574230 in __strncmp_ssse3 () from /lib64/libc.so.6
> #1 0x000000000045384c in manifest_file_seek_base (p=p@entry=0xc2a920,
> zName=zName@entry=0x8612b0 "~/", bBest=1) at ./src/manifest.c:1267
> #2 0x0000000000455b3d in manifest_file_seek (p=0xc2a920,
> zName=zName@entry=0x8612b0 "~/", bBest=bBest@entry=1)
> at ./src/manifest.c:1285
> #3 0x000000000044239a in fusefs_getattr (zPath=<optimized out>,
> stbuf=0x7fffffffdf20) at ./src/fusefs.c:167
> #4 0x00007ffff6806be3 in ?? () from /lib64/libfuse.so.2
> #5 0x00007ffff6807b15 in ?? () from /lib64/libfuse.so.2
> #6 0x00007ffff6809ada in fuse_session_loop () from /lib64/libfuse.so.2
> #7 0x00007ffff680f394 in ?? () from /lib64/libfuse.so.2
> #8 0x00000000004424e1 in fusefs_cmd () at ./src/fusefs.c:341
> #9 0x0000000000417e39 in main (argc=<optimized out>, argv=<optimized
> out>)
> at ./src/main.c:725
>
> A little sniffing:
>
> (gdb) p zName
> $1 = 0x8612b0 "~/"
> (gdb) p p
> $2 = (Manifest *) 0xc2a920
> (gdb) p lwr
> $3 = <optimized out>
> (gdb) p p->aFile
> $4 = (ManifestFile *) 0xc843d0
> (gdb) p p->aFile[lwr]
> value has been optimized out
> (gdb) p i
> $5 = <optimized out>
> (gdb) ptype p->aFile
> type = struct ManifestFile {
> char *zName;
> char *zUuid;
> char *zPerm;
> char *zPrior;
> } *
> (gdb) l
> 1262 return &p->aFile[i];
> 1263 }
> 1264 }
> 1265 if( bBest ){
> 1266 i = (int)strlen(zName);
> 1267 if( strncmp(zName, p->aFile[lwr].zName, i)==0 ) return
> &p->aFile[lwr];
> 1268 }
> 1269 return 0;
> 1270 }
> 1271
>
> While I have the debugger open (better ask soon), is there anything
> more you'd like me to check?
>
Since you are able to repro this, can you rebuild Fossil w/o the -O2 option
(so that variables are not optimized out) and rerun in the debugger yet
again, then do the same analysis?
>
> This happened while exploring the fossil repository (current through
> cfb8d6604f8a159e8a5c1fa8843be12060f0fff4 2014-06-26 07:40:11 UTC),
> looking at checkins/tip.
>
> I actually had another SIGSEGV earlier, not sure if this is the same
> crash or not.
>
> - --
> Andy Goth | <andrew.m.goth/at/gmail/dot/com>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v2.0.22 (MingW32)
> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
>
> iQEcBAEBAgAGBQJTrHwlAAoJELtYwrrr47Y4phoH/iVz7/gnObnYrbBKml+ydSWJ
> zxpILKE2Lib4ob+I0UuPtwTLvMfD9okLbjk3/ZR1cVGpHexEI+BXDdxMqwdllV1J
> hG4fExp2vKpPwz9Q7n6Pf4Ty4KqANAPdVKcnhliI+RBkFRzuCN7MXc45cfehf0y/
> /ffQyIAYKfs0g+zz0Cv3Ym2Nc9nGoYwDo3ZEMagO+KOZLSeI4PI0iDo6XKMxmwaq
> lS3Iq436rojCIdtkoY6xc8l0htMaZiGrmDdsIenJpLcMnI3i1Iamy48q25P+dDa3
> SDGaCmlvF1nZUnV6lgAGy57RDlxuxHmPAhpkixaaZ6X3nb0XmW+GZrCfm+1mucY=
> =aGaY
> -----END PGP SIGNATURE-----
> _______________________________________________
> fossil-users mailing list
> [email protected]
> http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
>
--
D. Richard Hipp
[email protected]
_______________________________________________
fossil-users mailing list
[email protected]
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
