While testing a MSWord document, AbiWord segfaulted (current CVS with debug
build).
I got this:
Program received signal SIGSEGV, Segmentation fault.
0x8183390 in wvApplySprmFromBucket (ver=16838689, sprm=16443, apap=0x100f021,
achp=0x403b0021, asep=0xbfffc060, stsh=0xbfffe420,
pointer=0x8352d46 "CJ\020", pos=0xbfffce76, data=0x0) at sprm.c:971
971 return(RetSprm);
Current language: auto; currently c
(gdb) where
#0 0x8183390 in wvApplySprmFromBucket (ver=16838689, sprm=16443,
apap=0x100f021, achp=0x403b0021, asep=0xbfffc060, stsh=0xbfffe420,
pointer=0x8352d46 "CJ\020", pos=0xbfffce76, data=0x0) at sprm.c:971
#1 0x21 in ?? ()
#2 0x20403baf in ?? ()
Cannot access memory at address 0x18403baf.
(gdb) l
966 wvEatSprm(sprm,pointer,pos);
967 break;
968 }
969
970 wvGetSprmFromU16(&RetSprm,sprm);
971 return(RetSprm);
972 }
973
974 void wvApplysprmPIstdPermute(PAP *apap,U8 *pointer,U16 *pos)
975 {
I do have the sample file (CSP79readme.doc). I couldn't really reproduce the
problem since next time I opened it, it worked.
It look worse sinc AbiWord seems really not to be deterministic: I have a
file (GUIDEABI.doc) that has been rendered 3 times differently, even if each
time assert failure where issued. This is really weird.
OK. I know have a way to reproduce the crash: start AbiWord, open the second
file, then the first file. Et voil� ! But if I open another in between, it
does NOT crash.
Hub
