[Bug 257830] New: mc abort (assertion failed)
https://bugzilla.novell.com/show_bug.cgi?id=257830 Summary: mc abort (assertion failed) Product: openSUSE 10.3 Version: Alpha 2 Platform: Other OS/Version: Other Status: NEW Severity: Major Priority: P5 - None Component: Other AssignedTo: nadvornik@novell.com ReportedBy: mmrazik@novell.com QAContact: qa@suse.de OtherBugsDependingO 245732 nThis: midnight is aborted when I use cursor keys with the following error: mc: xcb_xlib.c:42: xcb_xlib_lock: Assertion `!c->xlib.lock' failed. How to reproduce: - run mc - hit any cursor key (up, down,...) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 mmrazik@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|nadvornik@novell.com |sndirsch@novell.com Summary|mc abort (assertion failed) |xcb_xlib_lock: Assertion `!c->xlib.lock' failed. ------- Comment #1 from mmrazik@novell.com 2007-03-27 03:23 MST ------- okay - this apparently isn't mc bug. I'm getting the same assertion error e.g. when closing yast2 control center. I can reproduce this only when I'm using 'ssh -X' -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #2 from mmrazik@novell.com 2007-03-27 03:48 MST ------- "ssh -X root@openSuSE_10_3_system" from openSUSE 10.2 system, to be precise -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |mmrazik@novell.com ------- Comment #3 from sndirsch@novell.com 2007-03-27 04:15 MST ------- Since I can't reproduce this I need a gdb backtrace for both. mc and yast2 control center. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #4 from mmrazik@novell.com 2007-03-27 04:58 MST ------- mc backtrace: Program received signal SIGABRT, Aborted. [Switching to Thread 47180326708832 (LWP 12933)] 0x00002ae90600b395 in raise () from /lib64/libc.so.6 (gdb) bt #0 0x00002ae90600b395 in raise () from /lib64/libc.so.6 #1 0x00002ae90600c7f0 in abort () from /lib64/libc.so.6 #2 0x00002ae906004a76 in __assert_fail () from /lib64/libc.so.6 #3 0x00002ae906d0b871 in xcb_xlib_lock () from /usr/lib64/libxcb-xlib.so.0 #4 0x00002ae906a21475 in g_str_equal () from /usr/lib64/libX11.so #5 0x00002ae9069fdb48 in XFreeGC () from /usr/lib64/libX11.so #6 0x00002ae9069f6a8c in XCloseDisplay () from /usr/lib64/libX11.so #7 0x0000000000458c5a in x_error_handler (dpy=0x3285, ee=0x3285) at x11conn.c:106 #8 0x00002ae906a1aded in _XError () from /usr/lib64/libX11.so #9 0x00002ae906a223f8 in _XReply () from /usr/lib64/libX11.so #10 0x00002ae906a10a56 in XQueryPointer () from /usr/lib64/libX11.so #11 0x0000000000458b5a in mc_XQueryPointer (display=0x6f3010, win=121, root_return=0x7fffa5edfce8, child_return=0x7fffa5edfce0, root_x_return=0x7fffa5edfd04, root_y_return=0x7fffa5edfd00, win_x_return=0x7fffa5edfcfc, win_y_return=0x7fffa5edfcf8, mask_return=0x7fffa5edfcf4) at x11conn.c:216 #12 0x0000000000433772 in correct_key_code (code=<value optimized out>) at key.c:1286 #13 0x0000000000433b55 in get_key_code (no_delay=0) at key.c:906 #14 0x00000000004344dd in get_event (event=0x7fffa5edfeb0, redo_event=0, block=1) at key.c:964 #15 0x0000000000423d35 in run_dlg (h=0x78ab00) at dialog.c:775 #16 0x0000000000438183 in do_nc () at main.c:1672 #17 0x0000000000438a3e in main (argc=<value optimized out>, argv=0x7fffa5ee1198) at main.c:2239 (gdb) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 mmrazik@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|mmrazik@novell.com | ------- Comment #5 from mmrazik@novell.com 2007-03-27 05:03 MST ------- yast2 control center (btw I'm getting this assertion fail with any X application running via "ssh -X" - while closing the application): (gdb) bt #0 0x00002b0e3443c395 in raise () from /lib64/libc.so.6 #1 0x00002b0e3443d7f0 in abort () from /lib64/libc.so.6 #2 0x00002b0e34435a76 in __assert_fail () from /lib64/libc.so.6 #3 0x00002b0e34d7b871 in xcb_xlib_lock () from /usr/lib64/libxcb-xlib.so.0 #4 0x00002b0e33172475 in non-virtual thunk to QWidget::metric(int) const () from /usr/lib64/libX11.so.6 #5 0x00002b0e3314ec1a in XFreePixmap () from /usr/lib64/libX11.so.6 #6 0x00002b0e31423216 in QPixmap::deref () from /usr/lib/qt3/lib64/libqt-mt.so.3 #7 0x00002b0e314ceba7 in QPixmap::~QPixmap () from /usr/lib/qt3/lib64/libqt-mt.so.3 #8 0x00002b0e31414187 in QBitmap::~QBitmap () from /usr/lib/qt3/lib64/libqt-mt.so.3 #9 0x00002b0e31423188 in QPixmap::deref () from /usr/lib/qt3/lib64/libqt-mt.so.3 #10 0x00002b0e314ceb57 in QPixmap::~QPixmap () from /usr/lib/qt3/lib64/libqt-mt.so.3 #11 0x00002b0e3164beb7 in non-virtual thunk to QWidget::metric(int) const () from /usr/lib/qt3/lib64/libqt-mt.so.3 #12 0x00002b0e3443ee55 in __cxa_finalize () from /lib64/libc.so.6 #13 0x00002b0e313febe3 in non-virtual thunk to QWidget::metric(int) const () from /usr/lib/qt3/lib64/libqt-mt.so.3 #14 0x00007fff79b756c0 in ?? () #15 0x00002b0e317d2dd1 in _fini () from /usr/lib/qt3/lib64/libqt-mt.so.3 #16 0x0000000000000000 in ?? () (gdb) quit -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mhopf@novell.com, eich@novell.com ------- Comment #6 from sndirsch@novell.com 2007-03-27 05:07 MST ------- Looks like a bug in mc and libqt. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #7 from mmrazik@novell.com 2007-03-27 05:15 MST ------- just one more backtrace from konqueror: konqueror: xcb_xlib.c:42: xcb_xlib_lock: Assertion `!c->xlib.lock' failed. Program received signal SIGABRT, Aborted. [Switching to Thread 47082923447776 (LWP 13583)] 0x00002ad2523e5395 in raise () from /lib64/libc.so.6 (gdb) bt #0 0x00002ad2523e5395 in raise () from /lib64/libc.so.6 #1 0x00002ad2523e67f0 in abort () from /lib64/libc.so.6 #2 0x00002ad2523dea76 in __assert_fail () from /lib64/libc.so.6 #3 0x00002ad258486871 in xcb_xlib_lock () from /usr/lib64/libxcb-xlib.so.0 #4 0x00002ad254bfb475 in _XCBLockDisplay (dpy=0x618890) at xcb_lock.c:20 #5 0x00002ad254bd7c1a in XFreePixmap (dpy=0x350f, pixmap=31457462) at FreePix.c:41 #6 0x00002ad254417216 in QPixmap::deref () from /usr/lib/qt3/lib64/libqt-mt.so.3 #7 0x00002ad2544c2ba7 in QPixmap::~QPixmap () from /usr/lib/qt3/lib64/libqt-mt.so.3 #8 0x00002ad252774b57 in ~QBitmap (this=0x350f) at /usr/lib/qt3/include/qbitmap.h:47 #9 0x00002ad254417188 in QPixmap::deref () from /usr/lib/qt3/lib64/libqt-mt.so.3 #10 0x00002ad2544c2b57 in QPixmap::~QPixmap () from /usr/lib/qt3/lib64/libqt-mt.so.3 #11 0x00002ad25463feb7 in ?? () from /usr/lib/qt3/lib64/libqt-mt.so.3 #12 0x00002ad2523e7e55 in __cxa_finalize () from /lib64/libc.so.6 #13 0x00002ad2543f2be3 in ?? () from /usr/lib/qt3/lib64/libqt-mt.so.3 #14 0x00007fff58c1c000 in ?? () #15 0x00002ad2547c6dd1 in _fini () from /usr/lib/qt3/lib64/libqt-mt.so.3 #16 0x0000000000000000 in ?? () -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #8 from mmrazik@novell.com 2007-03-27 05:21 MST ------- I can not reproduce this when using "ssh -Y". While using -X everything is also much slower then usually... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|sndirsch@novell.com |nadvornik@novell.com ------- Comment #9 from sndirsch@novell.com 2007-03-27 07:36 MST ------- I can't see how this is related to X11 forwarding at all. I can't reproduce this on a STABLE system. Anyway these are bugs in applications or libs on top of new libX11/libxcb combo, which now makes sure that nobody locks/unlocks a display twice. Usually this is caused by unbalanced use of LockDisplay/UnlockDisplay. I suggest to search for such an unbalanced use in sources of mc and libQt. Reassigning to maintainer of mc. BTW, I still can't see that mc is a X11 application ... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #10 from mhopf@novell.com 2007-03-27 08:18 MST ------- Just tested SL10.2 -> root@SL10.3: All tested programs work flawlessly. For me that is a WORKSFORME. You could try to check that the qt-packages are recent enough on the remote machine. Are you using Xgl or is something else special in your environment? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #11 from mmrazik@novell.com 2007-03-27 08:22 MST ------- (In reply to comment #10)
Just tested SL10.2 -> root@SL10.3:
All tested programs work flawlessly. For me that is a WORKSFORME. You could try to check that the qt-packages are recent enough on the remote machine.
Are you using Xgl or is something else special in your environment?
No. Could you please try ssh -X root@dhcp9.suse.cz (standard password - thats the system which doesn't work for me) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #15 from mmrazik@novell.com 2007-03-27 10:20 MST ------- I have no idea as well. It is pretty obscure. I get rid of fglrx (using framebuffer ATM) but I'm, still able to reproduce. I asked 2 other people with the same laptop (lenovo T60) and the same setup (10.2 + fglrx) to try this, but they didn't observe this behaviour. I asked a few other people to reproduce but it seems I'm the only one with this problem. Really no idea. My laptop is pretty standard 10.2 installation with all updates installed. If you don't have any other ideas how to debug this further it will be probably WORKSFORME.... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |WORKSFORME ------- Comment #16 from sndirsch@novell.com 2007-03-27 10:31 MST ------- as you wish. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #17 from mmrazik@novell.com 2007-03-27 11:48 MST ------- well, I wish somebody comes with an idea how to debug/fix this ;-) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #18 from eich@novell.com 2007-03-27 12:03 MST ------- This problem has nothing to do with the fireglx driver. It's purely client side. the assertions have been added to xcb to help debug broken applications (or toolkit libs) but also broken libraries. If the problem is gone on a post 10.2 system it's a good sign as it indicates that a bug in an application library or application has been fixed. I don't recommend to run xcb on a production syustem on 10.2. That the problem goes away with a different form of X forwarding is not hard to explain: with the -Y forwarding the client uses the credentials of a secure X application while with -X it doesn't have access to certain resources and even extensions. Thus the client library needs to take different code paths ie. perform different actions (unless it fails completely). Most modern toolkits however are able to run without certain extensions - these code paths are however less tested. Thus it is possible that such a code path triggers the problem. The difference in performance can also be explained: usually the fallback code paths are slower. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #19 from mhopf@novell.com 2007-03-28 12:16 MST ------- Martin, you could diff the output of xdpyinfo on a working and a nonworking SL10.2 machine. Maybe the set of exported extensions is different. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #20 from mmrazik@novell.com 2007-03-29 03:03 MST ------- Nothing intereseting here (first time I had -DAMAGE in the diff, but I disabled DAMAGE and still can reproduce) :( conn:~/Documents:$ diff -Naur xdpyinfo_bad2.txt xdpyinfo.txt --- xdpyinfo_bad2.txt 2007-03-29 11:00:40.000000000 +0200 +++ xdpyinfo.txt 2007-03-29 09:59:24.000000000 +0200 @@ -17,7 +17,7 @@ depth 24, bits_per_pixel 32, scanline_pad 32 depth 32, bits_per_pixel 32, scanline_pad 32 keycode range: minimum 8, maximum 255 -focus: window 0x2e00007, revert to PointerRoot +focus: window 0x3600007, revert to PointerRoot number of extensions: 32 ATIFGLEXTENSION ATIFGLRXDRI -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #21 from sndirsch@novell.com 2007-03-29 03:08 MST ------- You already get another window ID, when you call xdpyinfo from another terminal, so this difference is unrelated. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #22 from mhopf@novell.com 2007-04-03 03:39 MST ------- Sorry, but I'm out of ideas. Maybe reinstalling the nonworking machine helps. But maybe it doesn't. As both Egbert and me already pointed out, this is in fact a client side problem, and the server *should* be unrelated, as long as the exported extensions are the same - which are in this case. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
https://bugzilla.novell.com/show_bug.cgi?id=257830 ------- Comment #23 from eich@novell.com 2007-04-03 06:06 MST ------- I'm not sure if I'm interpreting the top dump correctly. But it seems as if the error handler (x_error_handler) calls XCloseDisplay(). XCloseDisplay() doen't lock itself but XFreeGC() does. However XReply() gets called in a locked context already. This doesn't look right. #4 0x00002ae906a21475 in g_str_equal () from /usr/lib64/libX11.so #5 0x00002ae9069fdb48 in XFreeGC () from /usr/lib64/libX11.so #6 0x00002ae9069f6a8c in XCloseDisplay () from /usr/lib64/libX11.so #7 0x0000000000458c5a in x_error_handler (dpy=0x3285, ee=0x3285) at x11conn.c:106 #8 0x00002ae906a1aded in _XError () from /usr/lib64/libX11.so #9 0x00002ae906a223f8 in _XReply () from /usr/lib64/libX11.so The second and third dump is a lot harder to interpret. FreePixmap() is called (a rather normal thing) which of course tries to lock the display. Above that the trace ends up in a maze of Qt functions. No idea who else locks there. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.
participants (1)
-
bugzilla_noreply@novell.com