Description of problem: I'm seeing this in my dmesg: [95566.456090] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [95566.460453] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 276360 at 276359, next 276363) [95568.716017] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [95569.225052] [drm:i915_reset] *ERROR* Failed to reset chip. The X session just dies, though I can switch to VTs. I've put in some information here: https://bugzilla.kernel.org/show_bug.cgi?id=28882#c4 Chris Wilson mentioned it's a userspace bug in a follow-up comment in that bug report. I saw this with the kernel-2.6.38-0.rc6.git6.1.fc15.x86_64 kernel on an F14 userspace. I've not seen this yet with kernel-2.6.38-0.rc8.git0.1.fc15.x86_64. This bug has triggered thrice: once when mplayer was starting, once when totem was starting, once when I closed the laptop lid (which is set to just blank the screen). H/w details: 00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller]) Subsystem: Lenovo Device 20e4 Flags: bus master, fast devsel, latency 0, IRQ 47 Memory at f2000000 (64-bit, non-prefetchable) [size=4M] Memory at d0000000 (64-bit, prefetchable) [size=256M] I/O ports at 1800 [size=8] Expansion ROM at <unassigned> [disabled] Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit- Capabilities: [d0] Power Management version 3 Kernel driver in use: i915 Kernel modules: i915
Thanks for the bug report. We have reviewed the information you have provided above, and there is some additional information we require that will be helpful in our diagnosis of this issue. Please add drm.debug=0x04 to the kernel command line, restart computer, and attach * your X server config file (/etc/X11/xorg.conf, if available), * X server log file (/var/log/Xorg.*.log) * output of the dmesg command, and * system log (/var/log/messages) to the bug report as individual uncompressed file attachments using the bugzilla file attachment link above. We will review this issue again once you've had a chance to attach this information. Thanks in advance.
I've not seen this since the move to 2.6.38-rc8. So looks like this was a kernel issue that got fixed before rc8. I'll re-open if I do see it again.
I was just hit by this bug so it seems that it's back with the kernel-2.6.38.2-8.fc15.i686 package. Downgrading to previous (kernel-2.6.38.1-6.fc15.i686) solves it. Mar 29 16:58:16 work kernel: [ 2.245334] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0 Mar 30 12:18:51 work kernel: [75280.628024] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung Mar 30 12:18:51 work kernel: [75280.630762] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 7550520 at 7550508, next 7550521) Mar 30 12:18:54 work kernel: [75282.768014] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung Mar 30 12:18:54 work kernel: [75282.769145] [drm:i915_reset] *ERROR* Failed to reset chip.
Re-opening. When does it happen for you? Any sure way of reproducing?
FYI I was able to reproduce this on vanilla 2.6.38 but not 2.6.38.2 (so far at least). This was not on Fedora, but I believe that's irrelevant here.
Unfortunately downgrade of kernel doesn't help - I was just hit by it again. (In reply to comment #4) > Re-opening. When does it happen for you? Any sure way of reproducing? So far I don't know how to reproduce it. I'll try to downgrade mesa and see what happens.
Same thing happening to me. Especially bad with GL-on-GL-compositor. (Like running a Clutter app. inside GNOME Shell.)
This is 100% reproducible on my Lenovo T520 Sandy Bridge. Each time I move the mouse to the upper left in Gnome Shell it happens. F15 all updates as of 4-1-2011. kernel: [ 329.203437] [drm:i915_hangcheck_ring_idle] *ERROR* Hangcheck timer elapsed... blt ring idle [waiting on 72182, at 72182], missed IRQ? I will upload the log files.
Created attachment 489571 [details] dmesg dump
Created attachment 489572 [details] xorg.0.log
Created attachment 489573 [details] messages dump
Sorry, ^ Linux ace 2.6.38.2-10.fc15.x86_64 #1 SMP Thu Mar 31 03:11:25 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
*** Bug 696798 has been marked as a duplicate of this bug. ***
No change with 2.6.38.5.
2.6.38.2-14.fc15.x86_64 was fine for me for a long time. I then updated to 2.6.38.5-22.fc15.x86_64 and got this within 5 mins of booting. I've now dropped to 2.6.38.4-20.fc15.x86_64, this one's working fine for a few hours now.
(In reply to comment #15) > 2.6.38.2-14.fc15.x86_64 was fine for me for a long time. I then updated to > 2.6.38.5-22.fc15.x86_64 and got this within 5 mins of booting. I've now > dropped to 2.6.38.4-20.fc15.x86_64, this one's working fine for a few hours > now. Unfortunately I cant confirm that any of the kernels listed above worked for me - I'm still experiencing infrequent GPU hangs irrespective of kernel version. FYI I found that GPU dies quite often when some window pop ups (for example, popup windows in Transmission or Gajim). I even stopped to look at contact's details in Gajim :). Hope it helps fixing this annoying issue.
Yes, it is a userspace bug (also is against the xorg-drv-intel component). However, for me, looks like some kernels trigger it, some don't. 2.6.38.4 triggered this one as well after a while. I'm back to 2.6.38.2.
This issue is fixed in 2.6.39-rc6 and -rc7 from rawhide. (Although that kernel adds a new problem.)
2.6.39rc7 Vanilla git pull from yesterday, this is still happening for me.
I'm seeing this message with F15 beta, kernel kernel-2.6.38.6-26.rc1.fc15.x86_64 Mostly I see it when running Minecraft, but I just got it in normal gnome-shell use.
Me too: 2.6.38.6-27.fc15.x86_64 kernel Perhaps https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/761065 is related. They talk about a semaphore issue.
Doing the below seems to fix/prevent/work-a-round the issue for me. As root: echo 1 > /sys/module/i915/parameters/semaphores
This workaround doesn't fix entirely the problem. Globally I have much better performances. However, I can now encounter bigger hangs in OpenGL apps. I tried Bullet demos. When using the program AppAllBulletDemos, switching to another demo is veeerrry long (a few seconds or minutes). The first time I saw that I thought about a hard lock and I rebooted the computer! When performances go very bad, I see those messages from the kernel: [ 179.786033] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [ 179.786059] [drm:kick_ring] *ERROR* Kicking stuck semaphore on render ring [ 186.139002] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [ 186.139027] [drm:kick_ring] *ERROR* Kicking stuck semaphore on render ring [ 192.228321] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung [ 192.228345] [drm:kick_ring] *ERROR* Kicking stuck semaphore on render ring and so on... So the semaphore trick is related with the problem but there is something else to take into account.
I tried "echo 1 > /sys/module/i915/parameters/semaphores" and it seemed to help for a while. But then I got a total system lockup, which required a power-cycle reboot to recover from. I had never had one of those before.
Jeremy, maybe it was not a system lockup but a very long hang like I had. On my machine, I had to wait from a few seconds to several minutes (5 minutes ?). As I said, the program AppAllBulletDemos of Bullet demos shows this problem very well! http://bulletphysics.org/
Perhaps, I only waited about 10-15 secs. But it was a complete lockup - no pointer movement, no ping response, no caplock toggle, no magic-sysrq, sound playback buffer looping.
I am seeing exactly this on my fully patched F15-beta2 system (ie it's more F15 than F15-beta2) It's a Dell Latitude E6320 laptop with Sandybridge and every few days it totally locks up. I have walked up to it first thing in the morning to discover it's hung - so this is no "freezes for 10 minutes" thing - it's dead. This is with 2.6.38.6-27.fc15.x86_64
Have it on an Dell D630 (i965 chipset, no sandybridge) running the current Fedora 15 with kernel 2.6.38.7-30.fc15.x86_64. After some time the GUI gets completly unresponsive. It can go away again, but often I just restart. Very annoying. In the Ubuntu-Bugreport there is a hint to activate semaphores in the intel driver to solve this. I try it.
(In reply to comment #28) > After some time the GUI gets completly unresponsive. It can go away again, but > often I just restart. Very annoying. That is the not the same bug as this bug.
I'm seeing the same issue with fully updated F15 on a Dell E6320 (Sandybridge), i.e. kernel 2.6.38.7-30.fc15.x86_64, libdrm-2.4.26-1.fc15.x86_64, etc. I can repeatably and consistently produce serious hanging by attempting to run some OpenGL apps, for example Oolite (http://www.oolite.org) renders both itself and the system unusable. The suggested workaround simply turns a hard lockup into repeated lockups of several seconds at a time. The Fedora version for this bug should be changes to 15 as it is definitely still present.
This kernel patch is supposed to fix the issue: https://patchwork.kernel.org/patch/879532/ Can we get it backported into F15 asap?
Whoops, that appears to be a slightly outdated version of the patch. The patch we want is 498e720b96379d8ee9c294950a01534a73defcf3 in Linus' git tree.
I've compiled and am currently running a custom kernel based on 2.6.38.8-32.fc15 with the following two patches applied, and I get no more random GPU hangs. Please consider adding both patches to the next kernel update for F15.
Created attachment 509674 [details] SNB GPU hang fix, part 1 This patch is the abovementioned 498e720b96379d8ee9c294950a01534a73defcf3
Created attachment 509675 [details] SNB GPU hang fix, part 2 This patch is from the upstream bug report here: https://bugs.freedesktop.org/show_bug.cgi?id=38529
I am getting this same errors in dell latitude D630 Jun 27 18:15:03 kernel: [12443.368076] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung Jun 27 18:15:03 kernel: [12443.368089] [drm:kick_ring] *ERROR* Kicking stuck wait on render ring If i lock the screen and then try to login in around 10 mins it is hung and have to do a hard reboot. This happens with 2.6.38.8-32.fc15.x86_64 and 2.6.38.7-30.fc15.x86_64 kernels right now downgraded to kernel 2.6.38.6-27.fc15.x86_64 and this does not occurs. Can the above patches be added if its is the fix. Regards Ranjith
+1 for adding these patches in please. I had intermittent hanging on a Dell Optiplex 790, reliably triggered by opening the activities overview or bringing up the notification area in GNOME Shell. I'm now running a custom kernel (2.6.38.8-32.fc15 plus these two patches) and the problem is gone.
Hmm, bug filed against F14, reporting errors in various F15 and rawhide kernels.
Linux version 2.6.40.4-5.fc15.x86_64 Sep 17 18:52:55 rruban kernel: [14580.168147] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung Sep 17 18:52:55 rruban kernel: [14580.168160] [drm:kick_ring] *ERROR* Kicking stuck wait on render ring still hangs. Is there a fix yet ?. Can you tell me which kernel have this fix for f15 ?. Thanks Ranjith
I've got the same problem. Everytime I play Minecraft and pause the game to browse the internet or something else, X randomly just freezes. GDM restart doesn't help. Oct 12 09:52:50 arno-ThinkPad-T400 kernel: [ 2689.904090] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung Oct 12 09:52:50 arno-ThinkPad-T400 kernel: [ 2689.908614] [drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting 4288918 at 4288899, next 4288929) Oct 12 09:52:52 arno-ThinkPad-T400 kernel: [ 2691.956013] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung Oct 12 09:52:52 arno-ThinkPad-T400 kernel: [ 2692.464076] [drm:i915_reset] *ERROR* Failed to reset chip. Oct 12 09:52:59 arno-ThinkPad-T400 kernel: Kernel logging (proc) stopped. _____________ arno-ThinkPad-T400 ~ # lsb_release -a No LSB modules are available. Distributor ID: LinuxMint Description: Linux Mint 11 Katya Release: 11 Codename: katya arno-ThinkPad-T400 ~ # uname -a Linux arno-ThinkPad-T400 2.6.38-11-generic-pae #50-Ubuntu SMP Mon Sep 12 22:21:04 UTC 2011 i686 i686 i386 GNU/Linux Minecraft downloaded from Minecraft.net (paid version) Running with: java -Xmx1024M -Xms512M -cp minecraft.jar net.minecraft.LauncherFrame >/dev/null 2>&1 & _____________ arno-ThinkPad-T400 ~ # lspci |grep -i graph 00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) 00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) arno-ThinkPad-T400 ~ # lspci -vvv -s 00:02.0 00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller]) Subsystem: Lenovo Device 20e4 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0 Interrupt: pin A routed to IRQ 46 Region 0: Memory at f4400000 (64-bit, non-prefetchable) [size=4M] Region 2: Memory at d0000000 (64-bit, prefetchable) [size=256M] Region 4: I/O ports at 1800 [size=8] Expansion ROM at <unassigned> [disabled] Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit- Address: fee0300c Data: 41a9 Capabilities: [d0] Power Management version 3 Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-) Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME- Kernel driver in use: i915 Kernel modules: i915 arno-ThinkPad-T400 ~ # lspci -vvv -s 00:02.1 00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) Subsystem: Lenovo Device 20e4 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0 Region 0: Memory at f4200000 (64-bit, non-prefetchable) [size=1M] Capabilities: [d0] Power Management version 3 Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-) Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME- arno-ThinkPad-T400 ~ # lsmod |grep 915 i915 451068 2 drm_kms_helper 40745 1 i915 drm 184164 3 i915,drm_kms_helper i2c_algo_bit 13184 1 i915 video 18951 1 i915 arno-ThinkPad-T400 ~ # dmesg |grep -Ei $(lsmod |grep 915 |awk '{print $1}' |tr '\n' '|' |sed 's/.$//g') |grep -i init [ 1.637683] [drm] Initialized drm 1.1.0 20060810 [ 2.454311] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0 arno-ThinkPad-T400 ~ # dmesg |grep -Ei $(lsmod |grep 915 |awk '{print $1}' |tr '\n' '|' |sed 's/.$//g') [ 0.526510] pci 0000:00:02.0: Boot video device [ 1.637683] [drm] Initialized drm 1.1.0 20060810 [ 1.912331] i915 0000:00:02.0: power state changed by ACPI to D0 [ 1.912335] i915 0000:00:02.0: power state changed by ACPI to D0 [ 1.912340] i915 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 [ 1.912345] i915 0000:00:02.0: setting latency timer to 64 [ 1.944182] i915 0000:00:02.0: irq 46 for MSI/MSI-X [ 1.944188] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010). [ 1.944190] [drm] Driver supports precise vblank timestamp query. [ 2.437758] fbcon: inteldrmfb (fb0) is primary device [ 2.437843] fb0: inteldrmfb frame buffer device [ 2.437844] drm: registered panic notifier [ 2.454144] input: Video Bus as /devices/LNXSYSTM:00/device:00/PNP0A08:00/LNXVIDEO:00/input/input4 [ 2.454183] ACPI: Video Device [VID] (multi-head: yes rom: no post: no) [ 2.454311] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Seems that this is a post-2.6.37 regression. (Andrew Morton) https://lkml.org/lkml/2011/3/2/518
(In reply to comment #40) > I've got the same problem. > Everytime I play Minecraft and pause the game to browse the internet or > something else, X randomly just freezes. GDM restart doesn't help. I also get graphical glitches even when it's working OK. I suspect this is actually a regression in Minecraft since 1.8, but I think the driver stack should be a little more resilient. I often get complete system hangs after a couple of hangcheck warnings.