Intel i915 hangs

ilikepie · 2012-02-24 10:58:06

Hi,

Recently my system has been exhibiting seemingly random freezes once every couple days. All of a sudden, nothing responds except I can still move the mouse cursor around; typing has no effect, trying to switch to a tty does nothing. The system still responds to ping though. Sshd isn't running by default so I couldn't try logging in from another machine.

For what it's worth it happened once when I used Firefox, and another time when I was doing Java development.

When it's stuck like that I just force reboot the thing. I don't really know what's going on -- although if I had to guess, the most likely culprit would be something related to X, bumblebee or bbswitch. How could I be sure though? In what log files (or other things) could I look to find out what has crashed?

Thanks for any input!

Last edited by ilikepie (2012-02-29 19:25:37)

ManuP · 2012-02-24 13:43:17

/var/log/errors.log would be my guess.
I got an error with the same description once and it was a bug in intel driver. But since that was almost 2 years ago i highly doubt that would be your issue.

ilikepie · 2012-02-24 22:27:07

Alas /var/log/errors.log contains no relevant info.

The system crashed again twice today and this time sshd was on, and I could log into the computer successfully. However, killing X and lxdm was kind of weird (they didn't die at first), and then when they finally died, trying a Ctrl-Alt-F1 to get to a tty resulted in random characters on the screen in text mode. So I just rebooted via ssh.

I did a full system upgrade today too, but it didn't seem to fix anything.

EDIT: I have an Intel graphics chip, too...

Last edited by ilikepie (2012-02-24 22:37:15)

Avant-texte · 2012-02-25 00:51:33

ilikepie wrote:

Alas /var/log/errors.log contains no relevant info.
The system crashed again twice today and this time sshd was on, and I could log into the computer successfully. However, killing X and lxdm was kind of weird (they didn't die at first), and then when they finally died, trying a Ctrl-Alt-F1 to get to a tty resulted in random characters on the screen in text mode. So I just rebooted via ssh.
I did a full system upgrade today too, but it didn't seem to fix anything.
EDIT: I have an Intel graphics chip, too...

The whole system may not be to blame. A single service choking for some reason could be causing your problems. Saying your system crashed implies a kernel panic, but that's not the case if you're still getting responses from the system and have the ability to execute kill operations.

You might want to try turning off as many daemons as you can to see if your freezing problems stops. If so, incrementally turn your services back on until you find where the promlem actually is.

Posting your logs with "no relevant info" might show something useful to someone else here. Also, posting the configs of any seemingly problematic (and relevant) services is probably a good idea too. If you want help troubleshooting, you should give as much info as you can. Afterall you're asking for help with a machine we can't see.

NicolaOrritos · 2012-02-25 09:01:50

Also you can post the file .xsession-error.log, found into your home folder.
Plus, posting the file /var/log/Xorg.0.log may be helpful.
Feed us with data please :-)

Nicola

ilikepie · 2012-02-29 19:16:24

Thank you for the replies.

It happened again today -- but instead of turning it off right away, I decided to let it sit there for fifteen minutes or so. When I came back, it still didn't let me do anything, so I closed the laptop lid and open it, and lo! xscreensaver popped up and allowed me to log back in.

dmesg and Xorg.0.log now show useful information which they didn't when I just sshed into the freshly crashed machine earlier. The timestamp of the last log messages in xorg.0.log seems to coincide with when I got back to the machine and closed/reopened the lid (and X seemingly restarted, with the reloading of input devices, screen resolutions and such?).

Here goes:

dmesg:

[ 6303.231361] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 6303.231421] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state
[ 6303.241630] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 586211 at 586210, next 586216)

Xorg.0.log:

[  6312.672] [mi] EQ overflowing. The server is probably stuck in an infinite loop.
[  6312.672] 
Backtrace:
[  6312.740] 0: /usr/bin/X (xorg_backtrace+0x26) [0x566c26]
[  6312.741] 1: /usr/bin/X (mieqEnqueue+0x191) [0x547531]
[  6312.741] 2: /usr/bin/X (0x400000+0x46fa4) [0x446fa4]
[  6312.741] 3: /usr/bin/X (xf86PostKeyboardEvent+0x4b) [0x48222b]
[  6312.741] 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x7f109f611000+0x481e) [0x7f109f61581e]
[  6312.741] 5: /usr/bin/X (0x400000+0x6ca27) [0x46ca27]
[  6312.741] 6: /usr/bin/X (0x400000+0x9267a) [0x49267a]
[  6312.741] 7: /lib/libpthread.so.0 (0x7f10a4115000+0xf270) [0x7f10a4124270]
[  6312.741] 8: /lib/libc.so.6 (ioctl+0x7) [0x7f10a3087c17]
[  6312.741] 9: /usr/lib/libdrm.so.2 (drmIoctl+0x28) [0x7f10a1aa5418]
[  6312.741] 10: /usr/lib/libdrm_intel.so.1 (0x7f10a0fb3000+0x6303) [0x7f10a0fb9303]
[  6312.741] 11: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7f10a11ce000+0xb1b2) [0x7f10a11d91b2]
[  6312.741] 12: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7f10a11ce000+0x11d3a) [0x7f10a11dfd3a]
[  6312.741] 13: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7f10a11ce000+0x32603) [0x7f10a1200603]
[  6312.741] 14: /usr/lib/xorg/modules/drivers/intel_drv.so (0x7f10a11ce000+0x38d3c) [0x7f10a1206d3c]
[  6312.741] 15: /usr/bin/X (0x400000+0xf7684) [0x4f7684]
[  6312.741] 16: /usr/bin/X (0x400000+0xefc14) [0x4efc14]
[  6312.741] 17: /usr/bin/X (0x400000+0x33cd1) [0x433cd1]
[  6312.741] 18: /usr/bin/X (0x400000+0x22f0a) [0x422f0a]
[  6312.741] 19: /lib/libc.so.6 (__libc_start_main+0xed) [0x7f10a2fcd38d]
[  6312.741] 20: /usr/bin/X (0x400000+0x231fd) [0x4231fd]
[  6312.741] (EE) intel(0): Detected a hung GPU, disabling acceleration.
[  6312.741] (EE) intel(0): When reporting this, please include i915_error_state from debugfs and the full dmesg.
[  6670.819] (II) AIGLX: Suspending AIGLX clients for VT switch
[  6672.632] (II) Open ACPI successful (/var/run/acpid.socket)
[  6672.632] (II) AIGLX: Resuming AIGLX clients after VT switch
[  6672.682] (II) intel(0): EDID vendor "SEC", prod id 12874
[  6672.682] (II) intel(0): Printing DDC gathered Modelines:
[  6672.682] (II) intel(0): Modeline "1366x768"x0.0   70.70  1366 1414 1446 1486  768 770 775 792 -hsync -vsync (47.6 kHz)
[  6672.938] (--) synaptics: ETPS/2 Elantech ETF1059 Click-Pad: touchpad found

Also, /sys/kernel/debug/dri/0/i915_error_state contains some sort of raw, huge (2 megs) dump which may be better suited to a bug report.

EDIT: I haven't rebooted yet; glxspheres and glxgears just segfault, nothing is being displayed. Programs that require OpenGL refuse to open up, but for some reason, xfwm's built-in compositor still works.

Last edited by ilikepie (2012-02-29 19:27:35)

Gusar · 2012-02-29 19:32:29

Which graphic card?

ilikepie · 2012-02-29 19:48:30

The main gpu is an i915, but there's also a geforce gt520 mx that sits there deactivated at boot with bbswitch.

Gusar · 2012-02-29 20:28:43

i915 doesn't tell me anything, *every* intel card is handled by the i915 driver. But ok, you still did give me some info - I was checking if you have a 8xx card, which you don't.

GPU hangs are nasty. But I haven't seen them on non-8xx cards since ages. You should probably report this upstream.

ilikepie · 2012-02-29 21:37:57

Gusar wrote:

i915 doesn't tell me anything, *every* intel card is handled by the i915 driver. But ok, you still did give me some info - I was checking if you have a 8xx card, which you don't.
GPU hangs are nasty. But I haven't seen them on non-8xx cards since ages. You should probably report this upstream.

I think I'll report it, yup.

I'm not sure what chip it is exactly, lspci gives a generic name and the exact specs of the gpu in this laptop are nowhere to be found. They just detail what the Nvidia GPU is, not the Intel one. Maybe the Windows driver could report the exact name, I'll give that a try.

RomaHagen · 2012-02-29 22:33:19

I also experience same freezes. Are you using NetworkManager? Older versions of it showed this nasty behaviour.

ilikepie · 2012-02-29 23:08:13

Yeah. I used to use awesome and wicd and I didn't experience crashes with them. I've recently migrated to xfce and networkmanager, and crashes started to appear then. (I left wicd because I couldn't manage to get it to auto-connect to more than one wireless network, which is inconvenient since I'm often on the go, but that's another story)

However, I have an up-to-date system. I use networkmanager 0.9.2.0-2, and network-manager-applet 0.9.2.0-1. I might try going back to wicd for a while to see if that's the fix. I find it weird that a network daemon would mess with the graphics though.

Last edited by ilikepie (2012-02-29 23:11:45)

don_crissti · 2012-03-01 11:05:08

ilikepie wrote:

I'm not sure what chip it is exactly, lspci gives a generic name and the exact specs of the gpu in this laptop are nowhere to be found.

Geforce GT520MX was released last year so most likely your laptop has a Sandy Bridge processor. Your iGPU should be either HD Graphics 3000 or HD Graphics 2000, depending on your processor model.

kaname · 2012-03-14 06:33:16

I'm having the same problem, its on an Asus N53SV, Main graphic card is . Also have an optimus card, in my experience I get random freezes but no information about the reason of the crash, I checked every log in /var/log but but no trace of any problem.

Last edited by kaname (2012-03-14 06:41:22)

alexey · 2012-03-29 13:41:02

Hi,

I faced same problem more then a month ago. The last kernel verion was worknig 3.2.6. New kenels does not allow to switch resolutions via xrandr and as a result could play doom3 on lower res. I saved 3.2.6 kernel and just booted into it when i was goint to shot some...
About a week or two new error in dmesg

[ 288.141730] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 288.141745] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state
[ 288.147389] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 21113 at 21095, next 21114)
[ 296.475104] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 296.475148] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 21140 at 21124, next 21141)
[ 304.855064] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 304.855103] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 21182 at 21170, next 21183)
[ 313.148485] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 313.148564] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 21265 at 21261, next 21266)
[ 321.441896] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 321.441970] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 21276 at 21272, next 21277)

Sine both kernels were working the lastest one just could now switch to lower res. Now both are not working . Game loads , A few movement you do it hungs, and stays in this state for some seconds... so i have to kill it.

Did anyone faced just bug? Or any idea what from userspace can be broken. you know there x86_64 drivers and libs . for doom3 i need libe32-*

My current kernel is Linux localhost 3.2.13-1-ARCH #1 SMP PREEMPT Sat Mar 24 09:10:39 CET 2012 x86_64 Intel(R) Core(TM) i5-2557M CPU @ 1.70GHz GenuineIntel GNU/Linux

Kernel line ( I usee it on ASUS Zenbook 31UX with SSD) Command line: root=/dev/mapper/root cryptdevice=/dev/sda2:root ro elevator=noop nohz=on

alexey · 2012-03-29 13:43:24

It's me again. i forgot to say i tried to downgrade different X releted libs, drivers.... no affect still

cfr · 2012-03-29 22:15:59

I'm seeing similar freezes. I have intel graphics (the integrated 3000 series stuff) but I don't have any other graphics card. Intel is all there is. I don't see the problem with the lts kernel and the problem only appeared for me in the mainline kernel recently.

alexey · 2012-03-30 08:47:40

linux-lits solved problem with changing screen resolution ( as all =< 3.2.8 ). but u can still see
Mar 30 12:31:34 localhost kernel: [ 123.553756] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 30 12:31:34 localhost kernel: [ 123.553777] [drm] capturing error event; look for more information in /debug/dri/0/i9
15_error_state
Mar 30 12:31:34 localhost kernel: [ 123.560998] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 1
1322 at 11315, next 11323)
Mar 30 12:31:45 localhost kernel: [ 135.199808] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 30 12:31:45 localhost kernel: [ 135.199875] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 1
3380 at 13350, next 13381)
Mar 30 12:32:25 localhost kernel: [ 174.652570] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 30 12:32:25 localhost kernel: [ 174.652631] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 1
4023 at 14006, next 14025)

Can this be a GPU overheating bug ? This Zenbook 31UX is pretty hot

litemotiv · 2012-03-30 08:56:50

alexey wrote:

linux-lits solved problem with changing screen resolution ( as all =< 3.2.8 ). but u can still see
Mar 30 12:31:34 localhost kernel: [ 123.553756] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 30 12:31:34 localhost kernel: [ 123.553777] [drm] capturing error event; look for more information in /debug/dri/0/i9
15_error_state
Mar 30 12:31:34 localhost kernel: [ 123.560998] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 1
1322 at 11315, next 11323)
Mar 30 12:31:45 localhost kernel: [ 135.199808] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 30 12:31:45 localhost kernel: [ 135.199875] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 1
3380 at 13350, next 13381)
Mar 30 12:32:25 localhost kernel: [ 174.652570] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 30 12:32:25 localhost kernel: [ 174.652631] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 1
4023 at 14006, next 14025)

Can this be a GPU overheating bug ? This Zenbook 31UX is pretty hot

Do you have i915 RC6 enabled?

alexey · 2012-03-30 11:46:31

No i do not. i reacall it worked with i915.modeset=1, but after i completely removed all i915 options.

alexey · 2012-03-30 11:51:33

More over
on my Linux localhost 3.2.13-1-ARCH #1 SMP PREEMPT Sat Mar 24 09:10:39 CET 2012 x86_64 Intel(R) Core(TM) i5-2557M CPU @ 1.70GHz GenuineIntel GNU/Linux

I have this set up:
$ xrandr
Screen 0: minimum 320 x 200, current 1600 x 900, maximum 8192 x 8192
eDP1 connected 1600x900+0+0 (normal left inverted right x axis y axis) 293mm x 164mm
1600x900 60.0*+
1024x768 60.0 60.0
800x600 60.3 60.0 56.2
640x480 59.9
VGA1 disconnected (normal left inverted right x axis y axis)
HDMI1 disconnected (normal left inverted right x axis y axis)
DP1 disconnected (normal left inverted right x axis y axis)

And the command xrandr --output eDP1 --mode 1024x768 give me blank screen. so i have blindly type it back.
The lts kernel does not has this bug, as well all above 3.2.6

Last edited by alexey (2012-03-30 11:55:35)

cfr · 2012-03-30 16:26:18

I do have RC6 enabled but I only see the hangcheck errors with the mainline kernel, not the lts version. Although my laptop may have overheated on Monday evening (when it woke up from suspend in transit and I didn't realise), I didn't get any of these errors at that time even though the mainline kernel was in use. Rather I got such errors earlier in the day. Generally speaking, my laptop does not run especially hot (at least in comparison with my old one!) and so I certainly don't think it is an overheating issue in my case. grepping i915 in kernel.log:

Mar 25 16:27:30 localhost kernel: [125351.620968] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 25 16:27:30 localhost kernel: [125351.620982] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 25 16:27:30 localhost kernel: [125351.622711] i915 0000:00:02.0: setting latency timer to 64
Mar 25 21:05:36 localhost kernel: [    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-linux root=UUID=<UUID> ro quiet add_efi_memmap resume=/dev/disk/by-uuid/<UUID> pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 i915.semaphores=1
Mar 25 21:05:36 localhost kernel: [    0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-linux root=UUID=<UUID> ro quiet add_efi_memmap resume=/dev/disk/by-uuid/<UUID> pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 i915.semaphores=1
Mar 25 21:05:36 localhost kernel: [    2.786357] i915 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Mar 25 21:05:36 localhost kernel: [    2.786364] i915 0000:00:02.0: setting latency timer to 64
Mar 25 21:05:36 localhost kernel: [    2.879557] i915 0000:00:02.0: irq 40 for MSI/MSI-X
Mar 25 21:05:36 localhost kernel: [    3.693054] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Mar 26 12:27:16 localhost kernel: [18335.425865] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 26 12:27:16 localhost kernel: [18335.425879] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 26 12:27:16 localhost kernel: [18335.427586] i915 0000:00:02.0: setting latency timer to 64
Mar 26 12:35:32 localhost kernel: [18834.108099] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 12:35:38 localhost kernel: [18840.120747] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 12:35:45 localhost kernel: [18847.936499] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 12:35:45 localhost kernel: [18847.936513] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state
Mar 26 12:35:45 localhost kernel: [18847.954065] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 291299 at 291290, next 291304)
Mar 26 12:55:28 localhost kernel: [20030.749987] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 12:55:34 localhost kernel: [20036.762624] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 12:55:40 localhost kernel: [20042.778602] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 12:55:40 localhost kernel: [20042.778667] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 991598 at 991589, next 991605)
Mar 26 13:25:31 localhost kernel: [21832.875276] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 13:25:37 localhost kernel: [21838.887919] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 13:25:43 localhost kernel: [21844.900559] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 13:25:43 localhost kernel: [21844.900594] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 2053554 at 2053551, next 2053574)
Mar 26 13:25:49 localhost kernel: [21851.246493] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 13:25:55 localhost kernel: [21857.259141] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 13:26:01 localhost kernel: [21863.271772] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
Mar 26 13:26:01 localhost kernel: [21863.271839] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 2053595 at 2053575, next 2053597)
Mar 26 13:59:50 localhost kernel: [23892.078870] Modules linked in: fuse aes_x86_64 cryptd aes_generic cpufreq_ondemand coretemp rfcomm bnep ip6t_REJECT ip6t_LOG nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables ipt_REJECT ipt_LOG xt_limit xt_recent xt_tcpudp nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack iptable_filter ip_tables x_tables nls_cp437 vfat fat snd_hda_codec_hdmi snd_hda_codec_conexant joydev snd_hda_intel uvcvideo snd_hda_codec arc4 snd_hwdep snd_pcm snd_page_alloc snd_timer iwlwifi thinkpad_acpi videodev nvram snd v4l2_compat_ioctl32 mac80211 iTCO_wdt tpm_tis btusb bluetooth media atl1c iTCO_vendor_support cfg80211 serio_raw soundcore rts_pstor(C) mei(C) rfkill tpm psmouse wmi i2c_i801 thermal battery evdev pcspkr tpm_bios ac acpi_cpufreq mperf processor freq_table ext4 crc16 jbd2 mbcache usbhid hid sd_mod ahci libahci libata scsi_mod ehci_hcd usbcore usb_common i915 video button i2c_algo_bit intel_agp intel_gtt drm_kms_helper drm i2c_core
Mar 26 22:17:49 localhost kernel: [    2.562799] i915 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Mar 26 22:17:49 localhost kernel: [    2.562806] i915 0000:00:02.0: setting latency timer to 64
Mar 26 22:17:49 localhost kernel: [    2.664602] i915 0000:00:02.0: irq 40 for MSI/MSI-X
Mar 26 22:17:49 localhost kernel: [    3.488876] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Mar 26 22:24:41 localhost kernel: [  463.839027] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 26 22:24:41 localhost kernel: [  463.839041] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 26 22:24:41 localhost kernel: [  463.840793] i915 0000:00:02.0: setting latency timer to 64
Mar 26 22:48:29 localhost kernel: [    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-linux-lts root=UUID=<UUID> ro quiet add_efi_memmap resume=/dev/disk/by-uuid/<UUID> pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 i915.semaphores=1
Mar 26 22:48:29 localhost kernel: [    0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-linux-lts root=UUID=<UUID> ro quiet add_efi_memmap resume=/dev/disk/by-uuid/<UUID> pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 i915.semaphores=1
Mar 26 22:48:29 localhost kernel: [    3.836110] i915 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Mar 26 22:48:29 localhost kernel: [    3.836117] i915 0000:00:02.0: setting latency timer to 64
Mar 26 22:48:29 localhost kernel: [    3.927448] i915 0000:00:02.0: irq 40 for MSI/MSI-X
Mar 26 22:48:29 localhost kernel: [    4.683794] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Mar 27 11:54:37 localhost kernel: [12884.470794] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 27 11:54:37 localhost kernel: [12884.470808] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 27 11:54:37 localhost kernel: [12884.472925] i915 0000:00:02.0: setting latency timer to 64
Mar 27 13:07:42 localhost kernel: [17265.136961] Modules linked in: fuse aes_x86_64 cryptd aes_generic coretemp rfcomm bnep ip6t_REJECT ip6t_LOG nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables ipt_REJECT ipt_LOG xt_limit xt_recent xt_tcpudp nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack iptable_filter ip_tables x_tables nls_cp437 vfat fat joydev snd_hda_codec_hdmi snd_hda_codec_conexant arc4 uvcvideo videodev v4l2_compat_ioctl32 media iwlagn thinkpad_acpi snd_hda_intel snd_hda_codec mac80211 snd_hwdep snd_pcm nvram snd_page_alloc snd_timer snd iTCO_wdt btusb soundcore rts_pstor(C) bluetooth mei(C) wmi cfg80211 atl1c i2c_i801 serio_raw psmouse rfkill iTCO_vendor_support pcspkr tpm_tis thermal battery tpm tpm_bios evdev ac acpi_cpufreq mperf processor ext4 crc16 jbd2 mbcache usbhid hid sd_mod ahci libahci libata scsi_mod ehci_hcd usbcore i915 video button i2c_algo_bit intel_agp intel_gtt drm_kms_helper drm i2c_core
Mar 27 20:39:37 localhost kernel: [41052.392955] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 27 20:39:37 localhost kernel: [41052.392969] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 27 20:39:37 localhost kernel: [41052.395084] i915 0000:00:02.0: setting latency timer to 64
Mar 28 09:46:36 localhost kernel: [60281.034170] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 28 09:46:36 localhost kernel: [60281.034184] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 28 09:46:36 localhost kernel: [60281.036271] i915 0000:00:02.0: setting latency timer to 64
Mar 28 13:57:51 localhost kernel: [75063.283483] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 28 13:57:51 localhost kernel: [75063.283507] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 28 13:57:51 localhost kernel: [75063.285919] i915 0000:00:02.0: setting latency timer to 64
Mar 28 22:56:56 localhost kernel: [    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-linux-lts root=UUID=<UUID> ro quiet add_efi_memmap resume=/dev/disk/by-uuid/<UUID> pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 i915.semaphores=1
Mar 28 22:56:56 localhost kernel: [    0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-linux-lts root=UUID=<UUID> ro quiet add_efi_memmap resume=/dev/disk/by-uuid/<UUID> pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 i915.semaphores=1
Mar 28 22:56:56 localhost kernel: [    3.833207] i915 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Mar 28 22:56:56 localhost kernel: [    3.833214] i915 0000:00:02.0: setting latency timer to 64
Mar 28 22:56:56 localhost kernel: [    3.924413] i915 0000:00:02.0: irq 40 for MSI/MSI-X
Mar 28 22:56:56 localhost kernel: [    4.710371] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
Mar 29 15:51:17 localhost kernel: [17898.288307] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 29 15:51:17 localhost kernel: [17898.288322] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 29 15:51:17 localhost kernel: [17898.290403] i915 0000:00:02.0: setting latency timer to 64
Mar 29 21:58:24 localhost kernel: [37602.220789] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 29 21:58:24 localhost kernel: [37602.220813] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 29 21:58:24 localhost kernel: [37602.223185] i915 0000:00:02.0: setting latency timer to 64
Mar 30 15:13:57 localhost kernel: [51582.920102] i915 0000:00:02.0: restoring config space at offset 0xf (was 0x100, writing 0x10b)
Mar 30 15:13:57 localhost kernel: [51582.920116] i915 0000:00:02.0: restoring config space at offset 0x1 (was 0x900007, writing 0x900407)
Mar 30 15:13:57 localhost kernel: [51582.922059] i915 0000:00:02.0: setting latency timer to 64

The only plausible overheating occurred between around 19:30 and 22:30/23:00 on 25 March. And, as I say, I get these problems repeatedly with the mainline kernel when the machine is nowhere near overheating. (The laptop doesn't feel especially hot, the fan isn't especially fast, reported temperatures are typically below 50, with a critical level being 100.)

Last edited by cfr (2012-03-30 16:30:30)

alexey · 2012-03-30 19:36:14

I am still geting:

[ 257.358202] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 257.358217] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state
[ 257.363941] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 25927 at 25924, next 25967)
[ 310.108397] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 310.108472] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 39622 at 39589, next 39623)

And doom3 claims
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow
Posix_QueEvent: overflow

It was working before at least on linux 3.2.6 , bu i belive dome lib32 drm stuff changed. crap!

alexey · 2012-03-30 20:21:01

With lts kernel it's getting evern worse - hangs forever ( as supposed by new i915 features)

i would say i915 drm fscked up.... why did not did i rollback points when the doom3 woas working.... shame on me.

alexey · 2012-03-30 20:22:10

guys, reallly on 3.2.6 it wroked untill i updated some userspace... and i can not find it

Arch Linux

#1 2012-02-24 10:58:06

Intel i915 hangs

#2 2012-02-24 13:43:17

Re: Intel i915 hangs

#3 2012-02-24 22:27:07

Re: Intel i915 hangs

#4 2012-02-25 00:51:33

Re: Intel i915 hangs

#5 2012-02-25 09:01:50

Re: Intel i915 hangs

#6 2012-02-29 19:16:24

Re: Intel i915 hangs

#7 2012-02-29 19:32:29

Re: Intel i915 hangs

#8 2012-02-29 19:48:30

Re: Intel i915 hangs

#9 2012-02-29 20:28:43

Re: Intel i915 hangs

#10 2012-02-29 21:37:57

Re: Intel i915 hangs

#11 2012-02-29 22:33:19

Re: Intel i915 hangs

#12 2012-02-29 23:08:13

Re: Intel i915 hangs

#13 2012-03-01 11:05:08

Re: Intel i915 hangs

#14 2012-03-14 06:33:16

Re: Intel i915 hangs

#15 2012-03-29 13:41:02

Re: Intel i915 hangs

#16 2012-03-29 13:43:24

Re: Intel i915 hangs

#17 2012-03-29 22:15:59

Re: Intel i915 hangs

#18 2012-03-30 08:47:40

Re: Intel i915 hangs

#19 2012-03-30 08:56:50

Re: Intel i915 hangs

#20 2012-03-30 11:46:31

Re: Intel i915 hangs

#21 2012-03-30 11:51:33

Re: Intel i915 hangs

#22 2012-03-30 16:26:18

Re: Intel i915 hangs

#23 2012-03-30 19:36:14

Re: Intel i915 hangs

#24 2012-03-30 20:21:01

Re: Intel i915 hangs

#25 2012-03-30 20:22:10

Re: Intel i915 hangs

Board footer