Re: Save Copy to HTML segfault - latest CVS


Subject: Re: Save Copy to HTML segfault - latest CVS
From: Adrian Phillips (adrianp@powertech.no)
Date: Fri Jan 18 2002 - 02:49:34 CST


>>>>> "Adrian" == Adrian Phillips <adrianp@powertech.no> writes:

Adrian> I'm having a hard time trying to debug a crash with the
Adrian> latest CVS (as of 07 GMT).

Adrian> - Save Copy to .html gives :-

Adrian> 09:31:32:virusscan 511 $ ~/abiword/bin/abiword_debug
Adrian> mail_hub.abw GNU gdb 5.0 Copyright 2000 Free Software
Adrian> Foundation, Inc. GDB is free software, covered by the GNU
Adrian> General Public License, and you are welcome to change it
Adrian> and/or distribute copies of it under certain conditions.
Adrian> Type "show copying" to see the conditions. There is
Adrian> absolutely no warranty for GDB. Type "show warranty" for
Adrian> details. This GDB was configured as
Adrian> "i686-pc-linux-gnu"... [New Thread 1024 (LWP 16011)]

Adrian> Program received signal SIGSEGV, Segmentation fault.
Adrian> [Switching to Thread 1024 (LWP 16011)] 0x815e100 in
Adrian> pt_PieceTable::_tellAndMaybeAddListener (this=Cannot
Adrian> access memory at address 0x676e7036 ) at
Adrian> pt_PT_Listener.cpp:77 77 bool bStatus2 =
Adrian> pListener->populate(sfh,pcr); (gdb) bt #0 0x815e100 in
Adrian> pt_PieceTable::_tellAndMaybeAddListener (this=Cannot
Adrian> access memory at address 0x676e7036 ) at
Adrian> pt_PT_Listener.cpp:77 Cannot access memory at address
Adrian> 0x676e702e (gdb)

Adrian> - this is when compiling using -O2 -g - when compiling
Adrian> with -g (no -O2) it saves without problem and I can
Adrian> continue editing

Adrian> I don't know whether its related to these changes to
Adrian> pt_PT_Listener.cpp or something else :-

Adrian> ---------------------------- revision 1.14 date:
Adrian> 2001/02/06 22:54:38; author: cuenca; state: Exp; lines:
Adrian> +36 -36 Now you can forget that UT_Bool, UT_TRUE &
Adrian> UT_FALSE were never existed :)

Adrian> This is with :- ii gcc 2.95.4-5 The GNU C compiler.

Just thought I'd mention (for the archives at least) that compiling
with gcc-3.0 with -O2 doesn't have this problem so this seems to be a
2.95 bug.

Sincerely,

Adrian

-- 
Your mouse has moved.
Windows NT must be restarted for the change to take effect.
Reboot now?  [OK]



This archive was generated by hypermail 2b25 : Fri Jan 18 2002 - 02:49:44 CST