Bug 90511

Summary: [SNB] gpu freeze / hang on ubuntu 15.04 (kernel 3.19), sandy bridge i7
Product: DRI Reporter: ghostrider
Component: DRM/IntelAssignee: Intel GFX Bugs mailing list <intel-gfx-bugs>
Status: CLOSED INVALID QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: normal    
Priority: medium CC: intel-gfx-bugs
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
Output from "dmesg" (drm.debug=0xe) none

Description ghostrider 2015-05-18 15:37:06 UTC 
The system would frequently slow down (1-2 a day) slow down, hang, or throw me out back to the gdm login screen, respectively.

dmesg gives:

WARNING: CPU: 3 PID: 0 at /build/buildd/linux-3.19.0/drivers/gpu/drm/i915/intel_uncore.c:125 gen6_gt_check_fifodbg.isra.8+0x40/0x50 [i915]()
[  395.267274] GT wake FIFO error 0x3
[  395.267275] Modules linked in: pci_stub vboxpci(OE) vboxnetadp(OE) vboxnetflt(OE) vboxdrv(OE) arc4 md4 nls_utf8 cifs fscache cfg80211 rfcomm bnep joydev intel_rapl hid_logitech iosf_mbi x86_pkg_temp_thermal ff_memless hid_generic intel_powerclamp coretemp kvm_intel usbhid btusb gpio_ich hid bluetooth kvm crct10dif_pclmul snd_hda_codec_hdmi crc32_pclmul snd_hda_codec_realtek ghash_clmulni_intel snd_hda_codec_generic aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd serio_raw nuvoton_cir snd_hda_intel rc_core mac_hid 8250_fintek snd_hda_controller snd_hda_codec snd_hwdep snd_pcm i915 snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq video drm_kms_helper snd_seq_device snd_timer mei_me drm snd lpc_ich mei shpchp i2c_algo_bit soundcore parport_pc ppdev lp parport autofs4 btrfs xor
[  395.267320]  raid6_pq uas usb_storage e1000e psmouse ahci libahci ptp pps_core
[  395.267327] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G        W  OE  3.19.0-16-generic #16-Ubuntu
[  395.267329] Hardware name:                  /DH67CL, BIOS BLH6710H.86A.0076.2010.1115.1959 11/15/2010
[  395.267331]  ffffffffc077cdb8 ffff88041fac3c98 ffffffff817c2235 0000000000000007
[  395.267334]  ffff88041fac3ce8 ffff88041fac3cd8 ffffffff8107595a 0000000000000f72
[  395.267337]  ffff88040c610060 0000000000000003 000000000001f001 000000000002203c
[  395.267340] Call Trace:
[  395.267341]  <IRQ>  [<ffffffff817c2235>] dump_stack+0x45/0x57
[  395.267348]  [<ffffffff8107595a>] warn_slowpath_common+0x8a/0xc0
[  395.267351]  [<ffffffff810759d6>] warn_slowpath_fmt+0x46/0x50
[  395.267368]  [<ffffffffc0707b80>] gen6_gt_check_fifodbg.isra.8+0x40/0x50 [i915]
[  395.267389]  [<ffffffffc0707bb8>] __gen6_gt_force_wake_put+0x28/0x30 [i915]
[  395.267397]  [<ffffffffc070af52>] gen6_read32+0x132/0x140 [i915]
[  395.267405]  [<ffffffffc06fa039>] i915_hangcheck_elapsed+0x319/0x440 [i915]
[  395.267406]  [<ffffffff810dc290>] ? internal_add_timer+0x80/0x80
[  395.267413]  [<ffffffffc06f9d20>] ? i915_queue_hangcheck+0x60/0x60 [i915]
[  395.267415]  [<ffffffff810dc2d9>] call_timer_fn+0x39/0x110
[  395.267422]  [<ffffffffc06f9d20>] ? i915_queue_hangcheck+0x60/0x60 [i915]
[  395.267423]  [<ffffffff810de140>] run_timer_softirq+0x250/0x320
[  395.267425]  [<ffffffff81079d89>] __do_softirq+0x109/0x270
[  395.267427]  [<ffffffff8107a055>] irq_exit+0x95/0xa0
[  395.267428]  [<ffffffff817cc366>] smp_apic_timer_interrupt+0x46/0x60
[  395.267429]  [<ffffffff817ca3fd>] apic_timer_interrupt+0x6d/0x80
[  395.267430]  <EOI>  [<ffffffff81664402>] ? cpuidle_enter_state+0x62/0x160
[  395.267432]  [<ffffffff816643f1>] ? cpuidle_enter_state+0x51/0x160
[  395.267434]  [<ffffffff816645e7>] cpuidle_enter+0x17/0x20
[  395.267436]  [<ffffffff810b6a41>] cpu_startup_entry+0x311/0x3b0
[  395.267437]  [<ffffffff81048f97>] start_secondary+0x197/0x1c0
[  395.267438] ---[ end trace 68e780780d50c524 ]---
[  395.267444] [drm] stuck on render ring
[  395.267449] [drm] stuck on blitter ring
[  395.268190] [drm] GPU HANG: ecode 6:0:0xf4e9fffe, in Xorg [1327], reason: Ring hung, action: reset
[  395.268191] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[  395.268192] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[  395.268192] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[  395.268193] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[  395.268194] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[  395.268311] [drm:i915_set_reset_status [i915]] *ERROR* gpu hanging too fast, banning!
[  395.276285] drm/i915: Resetting chip after gpu hang

Please find attached /sys/class/drm/card0/error.
Comment 1 Ander Conselvan de Oliveira 2015-05-19 07:01:10 UTC 
Attachment is missing.

While at it, add drm.debug=0xe to your kernel command line, reproduce the bug again and attach full dmesg.
Comment 2 ghostrider 2015-05-22 15:06:00 UTC 
Created attachment 115975 [details]
Output from "dmesg" (drm.debug=0xe)

The error had been reproduced using "drm.debug=0x3" added to the kernel command line.
Comment 3 Ander Conselvan de Oliveira 2015-05-25 05:12:31 UTC 
You are using the kernel shipped with Ubuntu, which may contain a number of patches the upstream developers aren't aware of. Could you verify the issue occurs with an upstream kernel? Ideally, use the brach drm-intel-nightly from http://cgit.freedesktop.org/drm-intel for the latest i915 code.
Comment 4 Jani Nikula 2015-08-18 15:18:29 UTC 
Timeout, closing. Please reopen if you can reproduce the issue with a recent upstream kernel.

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.