Re: Tin 1.9.0 core dump

In <[email protected]>, Urs Janßen wrote:
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x0 in ?? ()
> > (gdb) bt
> > #0  0x0 in ?? ()
> > #1  0x8099bc1 in pcre_compile2 ()
> > #2  0x8098bed in pcre_compile ()
> > #3  0x8086e31 in compile_regex (
> >     regex=0x80e2aaf "(R[eE](\\^\\d+|\\[\\d\\])?|A[wW]|Odp):\\s",
> >     cache=0x8122c04, options=16) at ./regex.c:124
> 
> looks ok to me (but could be optimized as it's done in the newwer version)
> 
> > #4  0x806403e in postinit_regexp () at ./init.c:922
> > #5  0x806c1a8 in main (argc=3, argv=0xbffff564) at ./main.c:217
> > 
> > 
> > So, a problem in the new PCRE version... and a strip_re_regex value in the
> > tinrc file that can be REALLY old...
> > 
> > But after removing the tinrc file and creating a new one with tin 1.8.1
> > I still get the core dump and the trace
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x0 in ?? ()
> > (gdb) bt
> > #0  0x0 in ?? ()
> > #1  0x8099bc1 in pcre_compile2 ()
> > #2  0x8098bed in pcre_compile ()
> > #3  0x8086e31 in compile_regex (
> >     regex=0x80e2aaf "(?:R[eE](?:\\^\\d+|\\[\\d\\])?|A[wW]|Odp|Sv):\\s",
> >     cache=0x8122c04, options=16) at ./regex.c:124
> > #4  0x806403e in postinit_regexp () at ./init.c:922
> > #5  0x806c1a8 in main (argc=3, argv=0xbffff564) at ./main.c:217
> 
> again, looks ok. strange. I'll ask the other tin-dev folks if they have an
> idea.

the was another report - same problem, different OS/architectucre
(mips-sei-seiux):

| Program received signal SIGSEGV, Segmentation fault.
| 0x0 in ?? ()
| (gdb) bt
| #0  0x0 in ?? ()
| #1  0x46a670 in pcre_compile2 (
|     pattern=0x10027f80 "(?:R[eE](?:\\^\\d+|\\[\\d\\])?|A[wW]|Odp|Sv):\\s",
|     options=16, errorcodeptr=0x0, errorptr=0x7fffdbb0, erroroffset=0x7fffdbb4,
|     tables=0x1001b220 "") at ./pcre_compile.c:4939
| #2  0x4687e8 in pcre_compile (
|     pattern=0x10027f80 "(?:R[eE](?:\\^\\d+|\\[\\d\\])?|A[wW]|Odp|Sv):\\s",
|     options=16, errorptr=0x7fffdbb0, erroroffset=0x7fffdbb4, tables=0x0)
|     at ./pcre_compile.c:3905
| #3  0x447d30 in compile_regex (
|     regex=0x10027f80 "(?:R[eE](?:\\^\\d+|\\[\\d\\])?|A[wW]|Odp|Sv):\\s",
|     cache=0x10033970, options=115) at ./regex.c:124
| #4  0x41f954 in postinit_regexp () at ./init.c:923
| #5  0x4286a4 in main (argc=1, argv=0x0) at ./main.c:217
| 
| 4939 in pcre_compile.c: re = (real_pcre *)(pcre_malloc)(size);

I'm going to revert the pcre-6.4->pcre-6.6 patch.

urs
-- 
Jef Poskanzer:
"When people aren't stupid Usenet is even more useful. Too bad this happens
 so rarely."