[opensuse-buildservice] Dinotrace crashes in X on X86_64/2 CPU Opensuse 11
Hello, installing dinotrace-9.3f-1.4.x86_64.rpm from science/openSUSE_11.0 on OpenSuse 11.0/X86_64/2 CPU the program crashes:
gdb /usr/bin/dinotrace GNU gdb 6.8 Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-suse-linux"... (gdb) r Starting program: /usr/bin/dinotrace
Program received signal SIGSEGV, Segmentation fault. 0x00007f49aded107f in XNextEvent () from /usr/lib64/libX11.so.6 (gdb) bt #0 0x00007f49aded107f in XNextEvent () from /usr/lib64/libX11.so.6 #1 0x00007f49ae2088b9 in XtAppNextEvent () from /usr/lib64/libXt.so.6 #2 0x000000000043e177 in main () (gdb) The same crash happens when compiling from the sources at http://www.veripool.org/wiki/dinotrace with standard ./configure&& make. Removing all optimizations however, the 64-bit build runs with no crash so long, and so does dinotrace-9.3f-1.4.i586.rpm on the X86_64 machine or on a I386 1CPU machines. Should we expect the error in dinotrace or in some library? I also just mailed the author, Wilson Snyder. Cheers -- Uwe Bonnes bon@elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
* Uwe Bonnes (bon@elektron.ikp.physik.tu-darmstadt.de) [20080715 15:00]:
Program received signal SIGSEGV, Segmentation fault. 0x00007f49aded107f in XNextEvent () from /usr/lib64/libX11.so.6 (gdb) bt #0 0x00007f49aded107f in XNextEvent () from /usr/lib64/libX11.so.6 #1 0x00007f49ae2088b9 in XtAppNextEvent () from /usr/lib64/libXt.so.6 #2 0x000000000043e177 in main () (gdb)
After a bit of debugging it looks like it could be a bug in libXt. I've opened a new bug in http://rudin.suse.de:8888/show_bug.cgi?id=409324 . You might want to add yourself in CC of that bug. Philipp --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
On Dienstag, 15. Juli 2008, Philipp Thomas wrote:
* Uwe Bonnes (bon@elektron.ikp.physik.tu-darmstadt.de) [20080715 15:00]:
Program received signal SIGSEGV, Segmentation fault. 0x00007f49aded107f in XNextEvent () from /usr/lib64/libX11.so.6 (gdb) bt #0 0x00007f49aded107f in XNextEvent () from /usr/lib64/libX11.so.6 #1 0x00007f49ae2088b9 in XtAppNextEvent () from /usr/lib64/libXt.so.6 #2 0x000000000043e177 in main () (gdb)
After a bit of debugging it looks like it could be a bug in libXt. I've opened a new bug in http://rudin.suse.de:8888/show_bug.cgi?id=409324 . You might want to add yourself in CC of that bug.
I've played with it, too. Seems there's a problem with one of this two -O2 optimisation flags: -fschedule-insns -fschedule-insns2 Added the longer description to the bugtracker: https://bugzilla.novell.com/show_bug.cgi?id=409324 I'm changing the optimisation option to -O1 for dinotrace. Regards Werner --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
On Tue, 15 Jul 2008 21:44:20 +0200, Werner Hoch wrote:
I've played with it, too. Seems there's a problem with one of this two -O2 optimisation flags: -fschedule-insns -fschedule-insns2
Added the longer description to the bugtracker: https://bugzilla.novell.com/show_bug.cgi?id=409324
I'm changing the optimisation option to -O1 for dinotrace.
Didn't work in my tests :( dinotrace compiled with -O1 still segfaulted. Only -O0 i.e. no optimization at all got me an executable that ran without problems. Philipp --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
"Philipp" == Philipp Thomas <philipp.thomas2@gmx.net> writes:
Philipp> On Tue, 15 Jul 2008 21:44:20 +0200, Werner Hoch wrote: >> I've played with it, too. Seems there's a problem with one of this >> two -O2 optimisation flags: -fschedule-insns -fschedule-insns2 >> Added the longer description to the bugtracker: >> https://bugzilla.novell.com/show_bug.cgi?id=409324 >> >> I'm changing the optimisation option to -O1 for dinotrace. Philipp> Didn't work in my tests :( dinotrace compiled with -O1 still Philipp> segfaulted. Only -O0 i.e. no optimization at all got me an Philipp> executable that ran without problems. For me -O1 works too. -- Uwe Bonnes bon@elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
On Wed, Jul 16, 2008 at 10:23:27AM +0200, Uwe Bonnes wrote:
"Philipp" == Philipp Thomas <philipp.thomas2@gmx.net> writes:
Philipp> On Tue, 15 Jul 2008 21:44:20 +0200, Werner Hoch wrote: >> I've played with it, too. Seems there's a problem with one of this >> two -O2 optimisation flags: -fschedule-insns -fschedule-insns2
>> Added the longer description to the bugtracker: >> https://bugzilla.novell.com/show_bug.cgi?id=409324 >> >> I'm changing the optimisation option to -O1 for dinotrace.
Philipp> Didn't work in my tests :( dinotrace compiled with -O1 still Philipp> segfaulted. Only -O0 i.e. no optimization at all got me an Philipp> executable that ran without problems.
For me -O1 works too.
Is this for building dinotrace or for building xorg-x11-libXt? Ciao, Marcus --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
* Marcus Meissner (meissner@suse.de) [20080716 10:47]:
Is this for building dinotrace or for building xorg-x11-libXt?
For building dinotrace. In this loop /* loop forever */ /*This code ~= XtAppMainLoop (global->appcontext);*/ while (1) { XEvent event; if (global->redraw_needed && !XtAppPending (global->appcontext) && (!global->redraw_manually || (global->redraw_needed & GRD_MANUAL))) { draw_perform(); } XtAppNextEvent (global->appcontext, &event); /*if (DTPRINT) { printf ("[Event %s] ", /events[event.type]);fflush(stdout); }*/ XtDispatchEvent (&event); } At some point in the loop the lower four bytes of the address of event passed to XtAppNextEvent are cleared and a later memcpy leads to a segfault. Philipp --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org
participants (5)
-
Marcus Meissner -
Philipp Thomas -
Philipp Thomas -
Uwe Bonnes -
Werner Hoch