Bug 92117 - [gen4] GPU crash playing HTML5 video on Intel 82Q963/Q965, Ubuntu 14.04
Summary: [gen4] GPU crash playing HTML5 video on Intel 82Q963/Q965, Ubuntu 14.04
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: unspecified
Hardware: x86 (IA32) Linux (All)
: medium normal
Assignee: Ian Romanick
QA Contact: Intel 3D Bugs Mailing List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-09-25 16:17 UTC by Cedders
Modified: 2015-09-27 21:42 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
/sys/class/drm/card0/error following crash on Ubuntu 14.04 (33.21 KB, text/plain)
2015-09-25 16:17 UTC, Cedders
Details
Compressed /sys/class/drm/card0/error following crash with 2.99.917 (32.10 KB, application/octet-stream)
2015-09-25 16:37 UTC, Cedders
Details

Description Cedders 2015-09-25 16:17:21 UTC
Created attachment 118445 [details]
/sys/class/drm/card0/error following crash on Ubuntu 14.04

Reproducible: yes

Summary: playing https://www.youtube.com/watch?v=mY9x0AHUa-E in Firefox 41.0 or Chromium 45.0.2454.85, a second or two of video plays, then GPU crashes (and browser crash if Firefox); mouse pointer is fixed in position but can change shape until reboot.  Ctrl+Alt+Fn won't display other virtual console.  (Also occurs on other HTML5 YouTube video, but less reliably.)  

Sep 25 10:45:26 PC-2 kernel: [  397.000093] [drm] stuck on render ring
Sep 25 10:45:26 PC-2 kernel: [  397.000101] [drm] GPU crash dump saved to /sys/class/drm/card0/error
Sep 25 10:45:26 PC-2 kernel: [  397.000104] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Sep 25 10:45:26 PC-2 kernel: [  397.000106] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Sep 25 10:45:26 PC-2 kernel: [  397.000108] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Sep 25 10:45:26 PC-2 kernel: [  397.000111] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Sep 25 10:45:26 PC-2 kernel: [  397.000984] [drm:i915_set_reset_status] *ERROR* render ring hung inside bo (0x48ab000 ctx 0) at 0x48ab708
Sep 25 10:45:26 PC-2 kernel: [  397.072060] [drm] GMBUS [i915 gmbus vga] timed out, falling back to bit banging on pin 2
Sep 25 10:45:27 PC-2 kernel: [  397.508017] [drm:i915_reset] *ERROR* Failed to reset chip.
Sep 25 10:45:27 PC-2 kernel: [  397.870517] Chrome_ChildThr[2998]: segfault at 0 ip b76da11e sp b1bfea00 error 6 in plugin-container[b76d3000+3f000]

Sep 25 10:47:19 PC-2 kernel: [  509.880363] WARNING: CPU: 1 PID: 1960 at /build/linux-zqVCY7/linux-3.13.0/drivers/gpu/drm/i915/intel_display.c:922 assert_pll+0x73/0x80 [i915]()
Sep 25 10:47:19 PC-2 kernel: [  509.880364] PLL state assertion failure (expected on, current off)
Sep 25 10:47:19 PC-2 kernel: [  509.880387] Modules linked in: snd_hrtimer pci_stub vboxpci(OX) vboxnetadp(OX) vboxnetflt(OX) vboxdrv(OX) cuse dm_crypt bnep rfcomm bluetooth gpio_ich hp_wmi sparse_keymap snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_page_alloc snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device snd_timer coretemp kvm serio_raw lpc_ich shpchp snd soundcore parport_pc mac_hid binfmt_misc ppdev lp parport raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq raid1 raid0 multipath linear hid_generic usbhid hid i915 i2c_algo_bit floppy drm_kms_helper drm video pata_acpi e1000e wmi psmouse ptp pps_core

Youtube video characteristics:
Video ID: mY9x0AHUa-E
Dimensions: 854 x 480
Resolution: 640 x 360
Volume: 100%
Stream Type: https
CPN:jZE_mQvy-9ca8qqq
Mime Type: video/mp4; codecs="avc1.42001E, mp4a.40.2"
DASH: no (18)

Linux 3.13.0-63-generic #103-Ubuntu SMP Fri Aug 14 21:43:30 UTC 2015 i686
Ubuntu 14.04.3 LTS
xserver-xorg-video-intel:i386/trusty-updates 2:2.99.910-0ubuntu1.6 uptodate
intel-gpu-tools:i386/trusty-updates 1.3-0ubuntu2.1 uptodate
libdrm-intel1:i386/trusty-updates 2.4.60-2~ubuntu14.04.1 uptodate
libegl1-mesa:i386/trusty-updates 10.1.3-0ubuntu0.4 uptodate
libegl1-mesa-drivers:i386/trusty-updates 10.1.3-0ubuntu0.4 uptodate
libgl1-mesa-dev:i386/trusty-updates 10.1.3-0ubuntu0.4 uptodate
libgl1-mesa-dri:i386/trusty-updates 10.1.3-0ubuntu0.4 uptodate
libgl1-mesa-glx:i386/trusty-updates 10.1.3-0ubuntu0.4 uptodate

00:00.0 Host bridge: Intel Corporation 82Q963/Q965 Memory Controller Hub (rev 02)
00:02.0 VGA compatible controller: Intel Corporation 82Q963/Q965 Integrated Graphics Controller (rev 02) (prog-if 00 [VGA controller])
        Subsystem: Hewlett-Packard Company Device 2801
        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-

       description: VGA compatible controller
       product: 82Q963/Q965 Integrated Graphics Controller
       vendor: Intel Corporation
       physical id: 2
       bus info: pci@0000:00:02.0
       version: 02
       width: 64 bits
       clock: 33MHz
       capabilities: msi pm vga_controller bus_master cap_list rom
       configuration: driver=i915 latency=0
       resources: irq:41 memory:f0400000-f04fffff memory:e0000000-efffffff ioport:1100(size=8)

Possibly related: bug 89263, bug 89990, bug 77918, bug 91254
Comment 1 Cedders 2015-09-25 16:37:44 UTC
Created attachment 118447 [details]
Compressed /sys/class/drm/card0/error following crash with 2.99.917

This is also fairly reproducible with
libdrm-intel1:i386/vivid 2.4.60-2 uptodate
xserver-xorg-video-intel:i386/vivid-updates 2:2.99.917-1~exp1ubuntu2.2 uptodate
but seems to require video to play twice.
Comment 2 Chris Wilson 2015-09-25 16:43:57 UTC
It is worth checking with mesa-10.5.3+ as that fixed many gen4 issues.
Comment 3 Cedders 2015-09-25 19:03:39 UTC
Thanks, Chris.  Upgrading libgl1-mesa-dri libraries from 10.1.3 to 10.5.9-2ubuntu1~vivid2 seems to fix the crash, as does "always_flush_cache=true" or "always_flush_batch=true" workarounds mentioned in bug 80568.

Is there nothing to fix in the i915 driver then?
Comment 4 Kenneth Graunke 2015-09-27 21:42:49 UTC
Nope, this was a Mesa bug.  Glad it's working for you now.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.