First Last Prev Next    No search results available      Search page      Enter new bug
Bug#: 8100
Alias:
Product:
Component:
Status: RESOLVED
Resolution: FIXED
Assigned To: Gentoo Linux Gnome Desktop Team <gnome@gentoo.org>
Hardware:
OS:
Version:
Priority:
Severity:
Reporter: Jan Gutter <jangutter@tuks.co.za>
Add CC:
CC:
Remove selected CCs
URL:
Summary:
Status Whiteboard:
Keywords:

Filename Description Type Creator Created Size Actions
abiword.bug4 debug info grabed from bug-buddy (abiword 1.0.4_pre1-r1) text/plain Sylvain Meilard 2002-12-18 11:13 0000 10.51 KB Details
Create a New Attachment (proposed patch, testcase, etc.) View All

Bug 8100 depends on: Show dependency tree
Show dependency graph
Bug 8100 blocks:
Votes: 1    Show votes for this bug    Vote for this bug

Additional Comments: (this is where you put emerge --info)







View Bug Activity   |   Format For Printing   |   XML   |   Clone This Bug


Description:   Opened: 2002-09-19 05:34 0000
Abiword 1.0.2 crashes on print preview, print and exit... Is this Gentoo 1.4rc1
specific? I've also had this problem with a system compiled with GCC 3.1.1...
Any info is appreciated

------- Comment #1 From Modig 2002-10-02 19:28:56 0000 -------
<error text>
Applicatoin "/usr/bin/AbiWord_d" (process 28781) has crashed due to a fatal error.
(Segmentation fault)
</error text>

------- Comment #2 From Shad Nelson 2002-10-07 16:35:10 0000 -------
I get the same error with GCC 3.2

------- Comment #3 From foser (RETIRED) 2002-10-10 11:06:03 0000 -------
*** Bug 8850 has been marked as a duplicate of this bug. ***

------- Comment #4 From foser (RETIRED) 2002-10-16 13:59:14 0000 -------
Seems to be an abiword bug http://bugzilla.abisource.com/show_bug.cgi?id=3846

------- Comment #5 From foser (RETIRED) 2002-10-17 10:32:00 0000 -------
And another one to keep an eye on
http://bugzilla.abisource.com/show_bug.cgi?id=3457 .

------- Comment #6 From Sylvain Meilard 2002-12-16 18:40:54 0000 -------
I'm also a Gentoo user, and i have the same bug.
I'm (only) a C programmer, and not a a C++ programmer,
but i 've had a look at the code of the
release 1.0.2 :
it seem's that the crasch occurs line 963 in
abi/src/af/xap/unix/xap_UnixFrame.cpp, which is
gtk_widget_destroy(getTopLevelWindow());

That's in the method bool XAP_UnixFrame::close(),
called from the file
abi/src/wp/ap/xp/ap_EditMethods.cpp, line 2804, at
the end of the function
static bool s_closeWindow (AV_View * pAV_View,
EV_EditMethodCallData *pCallDate, bool bCanExit )

Could it come from the callback function
for the "destroy" signal ?
(I'm not a gtk guru...)

Hope this could help

------- Comment #7 From foser (RETIRED) 2002-12-16 18:51:04 0000 -------
interesting, for me the problem still exists in 1.4_pre1. Can you confirm that
one too ?

The problem is that on some systems which seem quite similar it does happen and
on other it's doesn't.

To give any sensible futher comment i need to look at it myself, will do so later.

------- Comment #8 From Sylvain Meilard 2002-12-18 07:47:23 0000 -------
Yes, i'm also using gentoo 1.4_pre1.
I build it from stage1 (stage1-x86-1.4_rc1-20020908-1208.tar.bz2)
CHOST="i686-pc-linux-gnu"
CFLAGS="-mcpu=athlon-xp -O3 -pipe"
with an athlon XP 1900+

------- Comment #9 From Sylvain Meilard 2002-12-18 10:08:22 0000 -------
I removed my -O3 option, and add a -g option to CFLAGS in /etc/make.conf, and
abiword don't crash anymore !
now, my -g option is useless :-)

------- Comment #10 From Sylvain Meilard 2002-12-18 11:05:43 0000 -------
In fact, it still crash when i ask for a print preview. But it don't crash when
quitting.

------- Comment #11 From Sylvain Meilard 2002-12-18 11:13:34 0000 -------
Created an attachment (id=6589) [edit]
debug info grabed from bug-buddy (abiword 1.0.4_pre1-r1)

here is the debug infos i get when i choose "print preview" in the "file" menu.

My CFLAGS has a -g option, and no -Ox option
I tried to rebuild also libgnome, libgnomeui, libgnomeprint, libgnomeprintui
and gtk+-1.2.10-r9 with the same options, but i didn't get more debug
informations.

------- Comment #12 From Matthew Smith 2003-01-03 00:26:06 0000 -------
I just did a 'USE="-gnome" emerge abiword' as per
http://bugzilla.abisource.com/show_bug.cgi?id=3457#c7 and neither print nor exit
causes a crash anymore.

------- Comment #13 From foser (RETIRED) 2003-01-13 10:25:34 0000 -------
Well, i think the print and exit crashes are unrelated. And the print crash i
havent seen in a while, the exit crash though is still there when i USE=-gnome.
Only you probably don't notice because there's no gnome crash dialog. Try
running abi from the commandline and check.

------- Comment #14 From Sander Graswinckel 2003-01-19 09:09:56 0000 -------
I can confirm that. Using -gnome I still get the "crash on exit":

/usr/bin/abiword: line 67:  2740 Aborted                
$ABISUITE_LIBEXEC/AbiWord_d "$@"

gdb reports:
Program received signal SIGSEGV, Segmentation fault.
0x401146b1 in gtk_handlers_run () from /usr/lib/libgtk-1.2.so.0

(gdb) bt
#0  0x401146b1 in gtk_handlers_run () from /usr/lib/libgtk-1.2.so.0
Cannot access memory at address 0x0

------- Comment #15 From Brad Laue (RETIRED) 2003-02-23 00:54:26 0000 -------
I've just built 1.4rc2 and AbiWord 1.0.4_pre1-r1 and AbiWord-1.0.4 both crash
on print, print preview, and exit as per the above.

My compiler optimizations are the standard -march=pentium3 -O3 -pipe and
nothing else special. GCC 3.2.1 does like to complain about the abiword source
though, while it's building it.

------- Comment #16 From Sander Graswinckel 2003-03-23 17:03:00 0000 -------
This problem (crash on exit) goes away if -O1 is used instead of -O2 in the
CXXFLAGS. I added CXXFLAGS="${CXXFLAGS//-O?} -O1" to the ebuild just above the
econf line. 
I hope this helps to narrow down this problem a little.

BTW I tried this on abiword-1.0.5 (which still has this problem)

------- Comment #17 From foser (RETIRED) 2003-03-24 07:47:52 0000 -------
this is amazing ... such a longstanding problem with such an easy solution.
Well, at least it works for me.. i'm adding it to the 1.0.5 ebuild in the tree
now.

Though i remember trying different optimizations to get it right.. duh mustve
missed this.

------- Comment #18 From foser (RETIRED) 2003-04-03 17:09:55 0000 -------
since 1.0.5 is stable now and has this optimization cap we should be fine from
now on.

thnx for your patience and help.

First Last Prev Next    No search results available      Search page      Enter new bug