Bug 699211 - [RV570] gnome-shell hangs in __glxWaitVideoSyncXGI, only a complete reboot fixes [@ xf86PostMotionEventM]
Summary: [RV570] gnome-shell hangs in __glxWaitVideoSyncXGI, only a complete reboot f...
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Fedora
Classification: Fedora
Component: xorg-x11-drv-ati
Version: 15
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: Jérôme Glisse
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard: [cat:modesetting]
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-04-24 09:52 UTC by Hans de Goede
Modified: 2018-04-11 08:41 UTC (History)
9 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2012-08-07 16:28:48 UTC
Type: ---


Attachments (Terms of Use)
dmesg (80.95 KB, text/plain)
2011-04-24 09:52 UTC, Hans de Goede
no flags Details
xorg.log without any xorg.conf (74.54 KB, text/plain)
2011-04-24 09:54 UTC, Hans de Goede
no flags Details
xorg.log with xorg.conf disabling page-flipping (71.79 KB, text/plain)
2011-04-24 09:55 UTC, Hans de Goede
no flags Details
xorg.conf disabling page-flipping (148 bytes, text/plain)
2011-04-24 09:56 UTC, Hans de Goede
no flags Details
backtrace of initial gnome-shell hang (6.90 KB, text/plain)
2011-04-24 09:57 UTC, Hans de Goede
no flags Details
Backtrace of gnome-shell hang after killing it (so when restarted) (7.86 KB, text/plain)
2011-04-24 09:58 UTC, Hans de Goede
no flags Details
Backtrace of gnome-shell hang after killing (and auto restart) of X-server (2.03 KB, text/plain)
2011-04-24 09:59 UTC, Hans de Goede
no flags Details
backtrace of gnome-shell hang after starting X-server with page-flipping disabled (1.32 KB, text/plain)
2011-04-24 09:59 UTC, Hans de Goede
no flags Details
xorg.log with backtrace (106.08 KB, text/plain)
2011-04-28 13:59 UTC, Hans de Goede
no flags Details
backtrace - current F15 (6.46 KB, text/plain)
2011-08-01 10:04 UTC, Maciek Borzecki
no flags Details
dmidecode file from the hanged computer (9.94 KB, application/octet-stream)
2011-10-22 10:36 UTC, yanir.dj
no flags Details
messages log file after the gnome-shell hanged (675.21 KB, application/octet-stream)
2011-10-22 10:51 UTC, yanir.dj
no flags Details
strace out put of hanged gnome-session (15.50 KB, application/octet-stream)
2011-10-22 13:37 UTC, yanir.dj
no flags Details

Description Hans de Goede 2011-04-24 09:52:49 UTC
Created attachment 494502 [details]
dmesg

Hi,

I've been having this hang, where my X-session would essentially freeze except for the mouse. When this happens I can still switch to a text vc (or ssh in, etc.), killing gnome-shell temporarily unfreezes X, and I can even interact with the most top level app (no window manager) until gnome-shell is done restarting. Some times gnome-shell does not restart completely. But usually it reloads completely (after being killed), draw all window decorations etc. and then freezes the session again as soon as it is done.

I've had this happen to me again today (while scrolling fastly through a pdf with evince), and I've collected some more info this time around, including gdb backtraces of gnome-shell each time it hung. I've done backtraces of the X-server too, but those just show it waiting in its main loop for something to happen. Here is what happened / I did:

1) gnome-shell hang, bt points to  __glxWaitVideoSyncXGI
2) kill gnome-shell
3) hangs again, bt points to writev from libxcb blocking
4) kill x-server
5) gdm is fine, gnome-shell hangs as soon as it has finished loading (it does load completely)
6) bt again points to  __glxWaitVideoSyncXGI
7) Disable pageflipping in xorg.conf
8) Kill x-server
9) gdm is fine, gnome-shell hangs again, but this time before completing loading
10) bt points to  writev from libxcb blocking (from a different libX11 call then before)

I'll attach Xorg.log, dmesg, the 4 backtraces, and the xorg.conf used during the 4th backtrace (no xorg.conf was used until step 7).

Comment 1 Hans de Goede 2011-04-24 09:54:56 UTC
Created attachment 494503 [details]
xorg.log without any xorg.conf

Comment 2 Hans de Goede 2011-04-24 09:55:24 UTC
Created attachment 494504 [details]
xorg.log with xorg.conf disabling page-flipping

Comment 3 Hans de Goede 2011-04-24 09:56:38 UTC
Created attachment 494505 [details]
xorg.conf disabling page-flipping

Comment 4 Hans de Goede 2011-04-24 09:57:47 UTC
Created attachment 494506 [details]
backtrace of initial gnome-shell hang

Comment 5 Hans de Goede 2011-04-24 09:58:28 UTC
Created attachment 494507 [details]
Backtrace of gnome-shell hang after killing it (so when restarted)

Comment 6 Hans de Goede 2011-04-24 09:59:10 UTC
Created attachment 494508 [details]
Backtrace of gnome-shell hang after killing (and auto restart) of X-server

Comment 7 Hans de Goede 2011-04-24 09:59:47 UTC
Created attachment 494509 [details]
backtrace of gnome-shell hang after starting X-server with page-flipping disabled

Comment 8 Hans de Goede 2011-04-27 13:27:38 UTC
Update: I just hit this again, even though I'm running with page-flipping disabled in xorg.conf (and xorg.log confirms this). gdb bt showed gnome-shell was again hanging in glxWaitVideoSyncXGI.

Comment 9 Hans de Goede 2011-04-28 13:59:19 UTC
Created attachment 495546 [details]
xorg.log with backtrace

Hi,

Not sure if this is the same issue. This morning I upgraded xorg-x11-server to
1.10.1-14.fc15, with the hope that it might fix the hang I was seeing. This time Xorg itself hang (ssh to the machine worked, ctrl+alt+F2 did not), rather the gnome-shell, killing -9 Xorg recovered the machine, I've never had this before.

Either this is a new bug introduced by the "backport upstream DRI2 fixes that are being screwed around with upstream" changes, or it is a new variant of the issue reported in this bug, with the behavior changed because of the xorg-server changes. For now I'm assuming it is the latter and thus logging this under this bug.

The attached xorg.log contains a backtrace of Xorg detecting it is stuck in a loop and printing a backtrace.

Regards,

Hans

Comment 10 Jonathan Paulett 2011-05-14 09:06:11 UTC
I've been seeing much the same behavior on a Radeon X1300 (RV515) for the last few weeks: gnome-shell hangs, the mouse is still responsive, I can switch to a VC and kill gnome-session which sometimes fixes it (at least for a while).

Comment 11 Maciek Borzecki 2011-06-11 10:34:48 UTC
Hate to comment with 'same for me', but that's the current state after fresh install of F15. Once the problem happens, subsequent restarts of gnome shell do not help. Disabling page flipping does not help either.

Comment 12 Carl-Johan Kjellander 2011-06-15 23:49:22 UTC
It has happened to me as well several times since the switch to F15 on x86_64 with 01:00.0 VGA compatible controller: ATI Technologies Inc R580 [Radeon X1900]

To get a semi usable system back so you can save your data you can try switching VT and running:

 pkill -9 gnome-shell; DISPLAY=:0 metacity & DISPLAY=:0 gnome-panel &

Nothing in dmesg or Xorg.0.log that is usable, but the behaviour is exactly the same as for Hans.

Comment 13 Matěj Cepl 2011-06-16 22:12:43 UTC
Backtrace:
[  6651.885] 0: /usr/bin/Xorg (xorg_backtrace+0x2f) [0x4a13df]
[  6651.885] 1: /usr/bin/Xorg (mieqEnqueue+0x1e9) [0x4a08f9]
[  6651.886] 2: /usr/bin/Xorg (xf86PostMotionEventM+0xa3) [0x47dae3]
[  6651.886] 3: /usr/bin/Xorg (xf86PostMotionEventP+0x52) [0x47dc02]
[  6651.886] 4: /usr/lib64/xorg/modules/input/evdev_drv.so (0x7f55a30ec000+0x498b) [0x7f55a30f098b]
[  6651.886] 5: /usr/bin/Xorg (0x400000+0x6b538) [0x46b538]
[  6651.886] 6: /usr/bin/Xorg (0x400000+0x119bf0) [0x519bf0]
[  6651.886] 7: /lib64/libpthread.so.0 (0x7f55a7e1c000+0xf3c0) [0x7f55a7e2b3c0]
[  6651.886] 8: /lib64/libc.so.6 (0x7f55a73db000+0xeda8c) [0x7f55a74c8a8c]
[  6651.886] 9: /lib64/libc.so.6 (0x7f55a73db000+0x7eb51) [0x7f55a7459b51]
[  6651.886] 10: /lib64/libc.so.6 (__libc_malloc+0x67) [0x7f55a7457917]
[  6651.886] 11: /lib64/libc.so.6 (0x7f55a73db000+0x7234d) [0x7f55a744d34d]
[  6651.887] 12: /lib64/libc.so.6 (0x7f55a73db000+0x788aa) [0x7f55a74538aa]
[  6651.887] 13: /lib64/libc.so.6 (0x7f55a73db000+0x7a6cb) [0x7f55a74556cb]
[  6651.887] 14: /lib64/libc.so.6 (0x7f55a73db000+0x7b5e9) [0x7f55a74565e9]
[  6651.887] 15: /lib64/libc.so.6 (realloc+0xee) [0x7f55a745808e]
[  6651.887] 16: /usr/bin/Xorg (miAppendSpans+0x2fb) [0x55442b]
[  6651.887] 17: /usr/bin/Xorg (0x400000+0x157bae) [0x557bae]
[  6651.887] 18: /usr/bin/Xorg (0x400000+0x157ecc) [0x557ecc]
[  6651.887] 19: /usr/bin/Xorg (miWideDash+0x7a3) [0x55b0b3]
[  6651.887] 20: /usr/lib64/xorg/modules/libexa.so (0x7f55a51b8000+0x1295f) [0x7f55a51ca95f]
[  6651.888] 21: /usr/bin/Xorg (miPolyRectangle+0xc4) [0x5527c4]
[  6651.888] 22: /usr/bin/Xorg (0x400000+0xdacdc) [0x4dacdc]
[  6651.888] 23: /usr/bin/Xorg (0x400000+0x2b34f) [0x42b34f]
[  6651.888] 24: /usr/bin/Xorg (0x400000+0x2ec11) [0x42ec11]
[  6651.888] 25: /usr/bin/Xorg (0x400000+0x22e1a) [0x422e1a]
[  6651.888] 26: /lib64/libc.so.6 (__libc_start_main+0xed) [0x7f55a73fc43d]
[  6651.888] 27: /usr/bin/Xorg (0x400000+0x23101) [0x423101]

and the annotated version

Frame 2: /usr/bin/Xorg (xf86PostMotionEventM+0xa3) [0x47dae3]
/usr/src/debug/xorg-server-1.10.1/hw/xfree86/common/xf86Xinput.c:1056
xf86PostMotionEventM
Frame 3: /usr/bin/Xorg (xf86PostMotionEventP+0x52) [0x47dc02]
/usr/src/debug/xorg-server-1.10.1/hw/xfree86/common/xf86Xinput.c:1008
xf86PostMotionEventP
Frame 5: /usr/bin/Xorg (0x400000+0x6b538) [0x46b538]
/usr/src/debug/xorg-server-1.10.1/hw/xfree86/common/xf86Events.c:300
xf86SigioReadInput
Frame 6: /usr/bin/Xorg (0x400000+0x119bf0) [0x519bf0]
/usr/src/debug/xorg-server-1.10.1/hw/xfree86/os-support/linux/../shared/sigio.c:106
xf86SIGIO
Frame 16: /usr/bin/Xorg (miAppendSpans+0x2fb) [0x55442b]
/usr/src/debug/xorg-server-1.10.1/mi/mispans.c:192
miAppendSpans
Frame 17: /usr/bin/Xorg (0x400000+0x157bae) [0x557bae]
/usr/src/debug/xorg-server-1.10.1/mi/miwideline.c:132
fillSpans
Frame 18: /usr/bin/Xorg (0x400000+0x157ecc) [0x557ecc]
/usr/src/debug/xorg-server-1.10.1/mi/miwideline.c:235
miFillPolyHelper
Frame 19: /usr/bin/Xorg (miWideDash+0x7a3) [0x55b0b3]
/usr/src/debug/xorg-server-1.10.1/mi/miwideline.c:1851
miWideDashSegment
Frame 20: /usr/lib64/xorg/modules/libexa.so (0x7f55a51b8000+0x1295f) [0x7f55a51ca95f]
/usr/src/debug/xorg-server-1.10.1/exa/exa_unaccel.c:267
ExaCheckPolylines
Frame 21: /usr/bin/Xorg (miPolyRectangle+0xc4) [0x5527c4]
/usr/src/debug/xorg-server-1.10.1/mi/mipolyrect.c:166
miPolyRectangle
Frame 22: /usr/bin/Xorg (0x400000+0xdacdc) [0x4dacdc]
/usr/src/debug/xorg-server-1.10.1/miext/damage/damage.c:1207
damagePolyRectangle
Frame 23: /usr/bin/Xorg (0x400000+0x2b34f) [0x42b34f]
/usr/src/debug/xorg-server-1.10.1/dix/dispatch.c:1787
ProcPolyRectangle
Frame 24: /usr/bin/Xorg (0x400000+0x2ec11) [0x42ec11]
/usr/src/debug/xorg-server-1.10.1/dix/dispatch.c:431
Dispatch
Frame 25: /usr/bin/Xorg (0x400000+0x22e1a) [0x422e1a]
/usr/src/debug/xorg-server-1.10.1/dix/main.c:289
main

Comment 14 Carl-Johan Kjellander 2011-06-17 22:19:43 UTC
Now, every time I try the trick above to start metacity I get the "Oh, Noes. Something bad happened" dialog. I can still access at least firefox (to save my progress in the javascript Gameboy Color emulator and save progess) by making it go full screen with F11.

It is just stupid of Gnome 3 to of having that dialog without any option to ignore and kill it, or trying at least metacity before forcing a logout.

I tend to lean that it is a race condition in the radeon driver since restarting Xorg doesn't help and you can't unload the radeon driver since is says it is busy.

Comment 15 Kevin R. Porter 2011-07-04 21:32:55 UTC
(In reply to comment #10)
> I've been seeing much the same behavior on a Radeon X1300 (RV515) for the last
> few weeks: gnome-shell hangs, the mouse is still responsive, I can switch to a
> VC and kill gnome-session which sometimes fixes it (at least for a while).

Another one with a Radeon X1300 (RV515) - going to a VC to kill gnome-shell has cleared the hang in most instances, but I've also seen a couple instances where the shell immedieately hung again and only a reboot would clear it.

Running Gnome under Fedora 8 up through Fedora 14 was rock stable on this iron, I have to admit this has me tempted to dump F15 and reinstall F14 until F16 comes out, Gnome 3 doesn't appear to be ready for real use yet.

Comment 16 Maciek Borzecki 2011-08-01 10:04:35 UTC
Created attachment 516107 [details]
backtrace - current F15

gnome-shell-3.0.2-4.fc15.i686
xorg-x11-drv-ati-6.14.1-2.20110525gitfe5c42f51.fc15.i686

Comment 17 Linus Walleij 2011-09-24 22:48:56 UTC
I'm experiencing this bug all the time when browsings PDFs with Evince, and I have this in my dmesg:

[ 1113.257157] radeon 0000:01:00.0: GPU lockup CP stall for more than 10020msec
[ 1113.257162] ------------[ cut here ]------------
[ 1113.257193] WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:267 radeon_fence_wait+0x296/0x33d [radeon]()
[ 1113.257196] Hardware name: System Product Name
[ 1113.257199] GPU lockup (waiting for 0x0000E8F9 last fence id 0x0000E8F7)
[ 1113.257201] Modules linked in: fuse coretemp cryptoloop svcrdma rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr nfsd nfs_acl auth_rpcgss lockd sunrpc cpufreq_ondemand acpi_cpufreq mperf bnep bluetooth nf_conntrack_ipv4 nf_defrag_ipv4 ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables belkin_sa eeepc_wmi asus_wmi sparse_keymap rfkill snd_hda_codec_hdmi microcode snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm iTCO_wdt iTCO_vendor_support i2c_i801 xhci_hcd snd_timer snd r8169 mii soundcore snd_page_alloc virtio_net kvm ipv6 ata_generic pata_acpi pata_via wmi radeon ttm drm_kms_helper drm i2c_algo_bit i2c_core [last unloaded: scsi_wait_scan]
[ 1113.257251] Pid: 1248, comm: Xorg Not tainted 2.6.40.4-5.fc15.x86_64 #1
[ 1113.257253] Call Trace:
[ 1113.257263]  [<ffffffff81054c8e>] warn_slowpath_common+0x83/0x9b
[ 1113.257269]  [<ffffffff81054d49>] warn_slowpath_fmt+0x46/0x48
[ 1113.257297]  [<ffffffffa00d2589>] ? evergreen_gpu_is_lockup+0xba/0xc2 [radeon]
[ 1113.257317]  [<ffffffffa0098dcd>] radeon_fence_wait+0x296/0x33d [radeon]
[ 1113.257322]  [<ffffffff81070566>] ? remove_wait_queue+0x3a/0x3a
[ 1113.257341]  [<ffffffffa00993c1>] radeon_sync_obj_wait+0x11/0x13 [radeon]
[ 1113.257350]  [<ffffffffa0060642>] ttm_bo_wait+0xbf/0x17a [ttm]
[ 1113.257357]  [<ffffffffa0061075>] ? ttm_bo_list_ref_sub+0x29/0x2b [ttm]
[ 1113.257381]  [<ffffffffa00a9dcb>] radeon_bo_wait+0x7b/0x9f [radeon]
[ 1113.257405]  [<ffffffffa00aa32d>] radeon_gem_wait_idle_ioctl+0x3d/0x70 [radeon]
[ 1113.257417]  [<ffffffffa00157f7>] drm_ioctl+0x29e/0x37b [drm]
[ 1113.257439]  [<ffffffffa00aa2f0>] ? radeon_gem_busy_ioctl+0x86/0x86 [radeon]
[ 1113.257444]  [<ffffffff811f649c>] ? inode_has_perm+0x32/0x34
[ 1113.257448]  [<ffffffff811f6545>] ? file_has_perm+0xa7/0xc9
[ 1113.257453]  [<ffffffff81134f0a>] do_vfs_ioctl+0x460/0x4a1
[ 1113.257458]  [<ffffffff810a10b5>] ? audit_syscall_exit+0x12d/0x148
[ 1113.257462]  [<ffffffff81134fa1>] sys_ioctl+0x56/0x79
[ 1113.257467]  [<ffffffff8148eaf5>] ? int_check_syscall_exit_work+0x34/0x3d
[ 1113.257472]  [<ffffffff8148e842>] system_call_fastpath+0x16/0x1b
[ 1113.257475] ---[ end trace 1e2d256367bfeb48 ]---
[ 1113.258584] radeon 0000:01:00.0: GPU softreset 
[ 1113.258587] radeon 0000:01:00.0:   GRBM_STATUS=0xA0003828
[ 1113.258590] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[ 1113.258593] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[ 1113.258595] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[ 1113.258608] radeon 0000:01:00.0:   GRBM_SOFT_RESET=0x00007F6B
[ 1113.258712] radeon 0000:01:00.0:   GRBM_STATUS=0x00003828
[ 1113.258715] radeon 0000:01:00.0:  [ 1113.257157] radeon 0000:01:00.0: GPU lockup CP stall for more than 10020msec
[ 1113.257162] ------------[ cut here ]------------
[ 1113.257193] WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:267 radeon_fence_wait+0x296/0x33d [radeon]()
[ 1113.257196] Hardware name: System Product Name
[ 1113.257199] GPU lockup (waiting for 0x0000E8F9 last fence id 0x0000E8F7)
[ 1113.257201] Modules linked in: fuse coretemp cryptoloop svcrdma rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr nfsd nfs_acl auth_rpcgss lockd sunrpc cpufreq_ondemand acpi_cpufreq mperf bnep bluetooth nf_conntrack_ipv4 nf_defrag_ipv4 ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables belkin_sa eeepc_wmi asus_wmi sparse_keymap rfkill snd_hda_codec_hdmi microcode snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm iTCO_wdt iTCO_vendor_support i2c_i801 xhci_hcd snd_timer snd r8169 mii soundcore snd_page_alloc virtio_net kvm ipv6 ata_generic pata_acpi pata_via wmi radeon ttm drm_kms_helper drm i2c_algo_bit i2c_core [last unloaded: scsi_wait_scan]
[ 1113.257251] Pid: 1248, comm: Xorg Not tainted 2.6.40.4-5.fc15.x86_64 #1
[ 1113.257253] Call Trace:
[ 1113.257263]  [<ffffffff81054c8e>] warn_slowpath_common+0x83/0x9b
[ 1113.257269]  [<ffffffff81054d49>] warn_slowpath_fmt+0x46/0x48
[ 1113.257297]  [<ffffffffa00d2589>] ? evergreen_gpu_is_lockup+0xba/0xc2 [radeon]
[ 1113.257317]  [<ffffffffa0098dcd>] radeon_fence_wait+0x296/0x33d [radeon]
[ 1113.257322]  [<ffffffff81070566>] ? remove_wait_queue+0x3a/0x3a
[ 1113.257341]  [<ffffffffa00993c1>] radeon_sync_obj_wait+0x11/0x13 [radeon]
[ 1113.257350]  [<ffffffffa0060642>] ttm_bo_wait+0xbf/0x17a [ttm]
[ 1113.257357]  [<ffffffffa0061075>] ? ttm_bo_list_ref_sub+0x29/0x2b [ttm]
[ 1113.257381]  [<ffffffffa00a9dcb>] radeon_bo_wait+0x7b/0x9f [radeon]
[ 1113.257405]  [<ffffffffa00aa32d>] radeon_gem_wait_idle_ioctl+0x3d/0x70 [radeon]
[ 1113.257417]  [<ffffffffa00157f7>] drm_ioctl+0x29e/0x37b [drm]
[ 1113.257439]  [<ffffffffa00aa2f0>] ? radeon_gem_busy_ioctl+0x86/0x86 [radeon]
[ 1113.257444]  [<ffffffff811f649c>] ? inode_has_perm+0x32/0x34
[ 1113.257448]  [<ffffffff811f6545>] ? file_has_perm+0xa7/0xc9
[ 1113.257453]  [<ffffffff81134f0a>] do_vfs_ioctl+0x460/0x4a1
[ 1113.257458]  [<ffffffff810a10b5>] ? audit_syscall_exit+0x12d/0x148
[ 1113.257462]  [<ffffffff81134fa1>] sys_ioctl+0x56/0x79
[ 1113.257467]  [<ffffffff8148eaf5>] ? int_check_syscall_exit_work+0x34/0x3d
[ 1113.257472]  [<ffffffff8148e842>] system_call_fastpath+0x16/0x1b
[ 1113.257475] ---[ end trace 1e2d256367bfeb48 ]---
[ 1113.258584] radeon 0000:01:00.0: GPU softreset 
[ 1113.258587] radeon 0000:01:00.0:   GRBM_STATUS=0xA0003828
[ 1113.258590] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[ 1113.258593] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[ 1113.258595] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[ 1113.258608] radeon 0000:01:00.0:   GRBM_SOFT_RESET=0x00007F6B
[ 1113.258712] radeon 0000:01:00.0:   GRBM_STATUS=0x00003828
[ 1113.258715] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[ 1113.258717] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[ 1113.258720] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[ 1113.259725] radeon 0000:01:00.0: GPU reset succeed
[ 1113.281733] radeon 0000:01:00.0: WB enabled
[ 1113.298256] [drm] ring test succeeded in 1 usecs
[ 1113.298266] [drm] ib test succeeded in 1 usecs
 GRBM_STATUS_SE0=0x00000007[ 1113.257157] radeon 0000:01:00.0: GPU lockup CP stall for more than 10020msec
[ 1113.257162] ------------[ cut here ]------------
[ 1113.257193] WARNING: at drivers/gpu/drm/radeon/radeon_fence.c:267 radeon_fence_wait+0x296/0x33d [radeon]()
[ 1113.257196] Hardware name: System Product Name
[ 1113.257199] GPU lockup (waiting for 0x0000E8F9 last fence id 0x0000E8F7)
[ 1113.257201] Modules linked in: fuse coretemp cryptoloop svcrdma rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr nfsd nfs_acl auth_rpcgss lockd sunrpc cpufreq_ondemand acpi_cpufreq mperf bnep bluetooth nf_conntrack_ipv4 nf_defrag_ipv4 ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables belkin_sa eeepc_wmi asus_wmi sparse_keymap rfkill snd_hda_codec_hdmi microcode snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm iTCO_wdt iTCO_vendor_support i2c_i801 xhci_hcd snd_timer snd r8169 mii soundcore snd_page_alloc virtio_net kvm ipv6 ata_generic pata_acpi pata_via wmi radeon ttm drm_kms_helper drm i2c_algo_bit i2c_core [last unloaded: scsi_wait_scan]
[ 1113.257251] Pid: 1248, comm: Xorg Not tainted 2.6.40.4-5.fc15.x86_64 #1
[ 1113.257253] Call Trace:
[ 1113.257263]  [<ffffffff81054c8e>] warn_slowpath_common+0x83/0x9b
[ 1113.257269]  [<ffffffff81054d49>] warn_slowpath_fmt+0x46/0x48
[ 1113.257297]  [<ffffffffa00d2589>] ? evergreen_gpu_is_lockup+0xba/0xc2 [radeon]
[ 1113.257317]  [<ffffffffa0098dcd>] radeon_fence_wait+0x296/0x33d [radeon]
[ 1113.257322]  [<ffffffff81070566>] ? remove_wait_queue+0x3a/0x3a
[ 1113.257341]  [<ffffffffa00993c1>] radeon_sync_obj_wait+0x11/0x13 [radeon]
[ 1113.257350]  [<ffffffffa0060642>] ttm_bo_wait+0xbf/0x17a [ttm]
[ 1113.257357]  [<ffffffffa0061075>] ? ttm_bo_list_ref_sub+0x29/0x2b [ttm]
[ 1113.257381]  [<ffffffffa00a9dcb>] radeon_bo_wait+0x7b/0x9f [radeon]
[ 1113.257405]  [<ffffffffa00aa32d>] radeon_gem_wait_idle_ioctl+0x3d/0x70 [radeon]
[ 1113.257417]  [<ffffffffa00157f7>] drm_ioctl+0x29e/0x37b [drm]
[ 1113.257439]  [<ffffffffa00aa2f0>] ? radeon_gem_busy_ioctl+0x86/0x86 [radeon]
[ 1113.257444]  [<ffffffff811f649c>] ? inode_has_perm+0x32/0x34
[ 1113.257448]  [<ffffffff811f6545>] ? file_has_perm+0xa7/0xc9
[ 1113.257453]  [<ffffffff81134f0a>] do_vfs_ioctl+0x460/0x4a1
[ 1113.257458]  [<ffffffff810a10b5>] ? audit_syscall_exit+0x12d/0x148
[ 1113.257462]  [<ffffffff81134fa1>] sys_ioctl+0x56/0x79
[ 1113.257467]  [<ffffffff8148eaf5>] ? int_check_syscall_exit_work+0x34/0x3d
[ 1113.257472]  [<ffffffff8148e842>] system_call_fastpath+0x16/0x1b
[ 1113.257475] ---[ end trace 1e2d256367bfeb48 ]---
[ 1113.258584] radeon 0000:01:00.0: GPU softreset 
[ 1113.258587] radeon 0000:01:00.0:   GRBM_STATUS=0xA0003828
[ 1113.258590] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[ 1113.258593] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[ 1113.258595] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[ 1113.258608] radeon 0000:01:00.0:   GRBM_SOFT_RESET=0x00007F6B
[ 1113.258712] radeon 0000:01:00.0:   GRBM_STATUS=0x00003828
[ 1113.258715] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[ 1113.258717] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[ 1113.258720] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[ 1113.259725] radeon 0000:01:00.0: GPU reset succeed
[ 1113.281733] radeon 0000:01:00.0: WB enabled
[ 1113.298256] [drm] ring test succeeded in 1 usecs
[ 1113.298266] [drm] ib test succeeded in 1 usecs
[ 1113.258717] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[ 1113.258720] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[ 1113.259725] radeon 0000:01:00.0: GPU reset succeed
[ 1113.281733] radeon 0000:01:00.0: WB enabled
[ 1113.298256] [drm] ring test succeeded in 1 usecs
[ 1113.298266] [drm] ib test succeeded in 1 usecs

Comment 18 Linus Walleij 2011-09-25 05:40:02 UTC
Is it possible to disable the GL rendering in Evince or Cairo or whatever part it is that is talking to the GPU? Just for testers? I can reproduce the bug very easily on this system. Hints?

Comment 19 Jonathan Paulett 2011-10-08 23:26:11 UTC
This bug is still present in a fully updated F16 system. In fact, it's even worse now; because of the compositing in GDM, I now have to reboot every time it happens instead of just killing gnome-session.

Comment 20 yanir.dj 2011-10-22 10:34:17 UTC
hi 
i got the same problem .
Using fedora15 
gnome-shell-3.0.2-4.fc15.x86_64:
2.6.40.6-0.fc15.x86_64 #1 SMP  x86_64 x86_64 x86_64 GNU/Linux.

a lot of times i get the x-windows hanged ,i can only kill from another tty the gnome-session ,or else try "alt-ctl-backspace" which not always works .

attached dmidecode +messages.log .

i will try to had the trace of the hanging gnome-session 

Thanks 

YAnir

Comment 21 yanir.dj 2011-10-22 10:36:58 UTC
Created attachment 529602 [details]
dmidecode file from the hanged computer

Comment 22 yanir.dj 2011-10-22 10:51:14 UTC
Created attachment 529603 [details]
messages log file after the gnome-shell hanged

Comment 23 yanir.dj 2011-10-22 13:37:40 UTC
Created attachment 529616 [details]
strace out put of hanged gnome-session

Comment 24 bradejr 2011-12-21 16:18:34 UTC
I can verify that this is still a problem on F16:

gnome-shell-3.2.1-2.fc16.i686
xorg-x11-drv-ati-6.14.3-3.20111125git534fb6e41.fc16.i686
kernel-3.1.5-6.fc16.i686

01:00.0 VGA compatible controller: ATI Technologies Inc RV505 [Radeon X1550 Series]

Comment 25 Linus Walleij 2012-04-03 07:42:20 UTC
Reproduced lockup today with kernel 3.3.0:

[57593.846219] radeon 0000:01:00.0: GPU lockup CP stall for more than 10020msec
[57593.846221] GPU lockup (waiting for 0x001A308A last fence id 0x001A3084)
[57593.847303] radeon 0000:01:00.0: GPU softreset 
[57593.847305] radeon 0000:01:00.0:   GRBM_STATUS=0xA0003828
[57593.847307] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[57593.847308] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[57593.847310] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[57593.847325] radeon 0000:01:00.0:   GRBM_SOFT_RESET=0x00007F6B
[57593.847435] radeon 0000:01:00.0:   GRBM_STATUS=0x00003828
[57593.847436] radeon 0000:01:00.0:   GRBM_STATUS_SE0=0x00000007
[57593.847437] radeon 0000:01:00.0:   GRBM_STATUS_SE1=0x00000007
[57593.847439] radeon 0000:01:00.0:   SRBM_STATUS=0x200000C0
[57593.848441] radeon 0000:01:00.0: GPU reset succeed
[57593.869221] [drm] PCIE GART of 512M enabled (table at 0x0000000000040000).
[57593.869299] radeon 0000:01:00.0: WB enabled
[57593.869300] [drm] fence driver on ring 0 use gpu addr 0x40000c00 and cpu addr 0xff494c00
[57593.885474] [drm] ring test on 0 succeeded in 0 usecs
[57593.885529] [drm] ib test on ring 0 succeeded in 1 usecs

Symptom is that userspace gnome-shell hangs as an effect of the crash in the Radeon driver.

Comment 26 Fedora End Of Life 2012-08-07 16:28:50 UTC
This message is a notice that Fedora 15 is now at end of life. Fedora
has stopped maintaining and issuing updates for Fedora 15. It is
Fedora's policy to close all bug reports from releases that are no
longer maintained. At this time, all open bugs with a Fedora 'version'
of '15' have been closed as WONTFIX.

(Please note: Our normal process is to give advanced warning of this
occurring, but we forgot to do that. A thousand apologies.)

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, feel free to reopen
this bug and simply change the 'version' to a later Fedora version.

Bug Reporter: Thank you for reporting this issue and we are sorry that
we were unable to fix it before Fedora 15 reached end of life. If you
would still like to see this bug fixed and are able to reproduce it
against a later version of Fedora, you are encouraged to click on
"Clone This Bug" (top right of this page) and open it against that
version of Fedora.

Although we aim to fix as many bugs as possible during every release's
lifetime, sometimes those efforts are overtaken by events. Often a
more recent Fedora release includes newer upstream software that fixes
bugs or makes them obsolete.

The process we are following is described here:
http://fedoraproject.org/wiki/BugZappers/HouseKeeping


Note You need to log in before you can comment on or make changes to this bug.