Bug 91447

Summary: [BSW-BDW-U] igt/kms_frontbuffer_tracking GPU HANG
Product: DRI Reporter: cprigent <christophe.prigent>
Component: DRM/IntelAssignee: dog <paul.a.parenteau>
Status: CLOSED DUPLICATE QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: critical    
Priority: high CC: intel-gfx-bugs
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: BDW, BSW/CHT i915 features:
Attachments:
Description Flags
kern.log.tar.gz
none
Kernel configuration none

Description cprigent 2015-07-24 07:38:55 UTC
Created attachment 117334 [details]
kern.log.tar.gz

Braswell-M
Hardware
Platform: Braswell M
CPU : Intel(R) CPU  @ 1.52 GHz (family: 6, model: 76 stepping: 3)
SoC : BSW C0 CRB : BRASWELL RVP Fab2
Mandatory Reworks : All Feature Reworks: F28, F32,F33 & F37
Optional reworks : O-01a Software
BIOS : BRAS.X64.B074.R00.1505311952 TXE FW : 2.0.0.2064
Ksc : 1.08
Linux distribution: Ubuntu 14.04 LTS 64 bits
Kernel 4.2-rc2 from git://anongit.freedesktop.org/drm-intel branch drm-intel-nightly
Mesa: mesa-10.6.2 from http://cgit.freedesktop.org/mesa/mesa/
Xf86_video_intel: 2.99.917 from
http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/
Libdrm: libdrm-2.4.62 from http://cgit.freedesktop.org/mesa/drm/
Cairo: 1.14.2 from http://cgit.freedesktop.org/cairo
libva: libva-1.6.0 from http://cgit.freedesktop.org/libva/
intel-driver: 1.6.0 from http://cgit.freedesktop.org/vaapi/intel-driver
xorg: 1.17.99 installed with script git_xorg.sh
Xserver: xorg-server-1.17.2 from http://cgit.freedesktop.org/xorg/xserver
Intel-gpu-tools: 8ad1e4077879a111f341dbfd2e0fee84efc9f57e from http://cgit.freedesktop.org/xorg/app/intel-gpu-tools/

Steps:
------
1. Execute command:
cd <...>/intel-gpu-tools/scripts/
./run-tests.sh -r <path>/results  -x .*suspend.* -x gem_userptr_blits@swapping.* -x gem_dummy_reloc_loop -x gem_concurrent_all.* -x gem_evict_everything@swapping-hang -x gem_close_race -x gem_cpu_reloc -x gem_ctx_thrash.* -x drv_hangman.* -x drv_missed_irq_hang -x kms_setmode.* -x gem_fence_thrash@bo-write-verify-threaded-y -x gem_multi_bsd_sync_loop -x gem_reloc_vs_gpu@forked-interruptible-faulting-reloc-thrashing -x gem_seqno_wrap -x gem_userptr_blits@forked-sync-swapping-multifd-mempressure-interruptible -x kms_render@gpu-blit -x kms_cursor_crc@cursor-size-change -x gem_render_linear_blits@swap-thrash -x gem_render_tiled_blits@swap-thrash -x gem_evict_everything@mlocked-hang -x gem_userptr_blits@mlocked-normal-sync -x gem_userptr_blits@mlocked-sync-interruptible -x gem_userptr_blits@mlocked-unsync-normal -x gem_userptr_blits@mlocked-unsync-interruptible -x gem_exec_big -x gem_mmap_gtt@medium-copy -x gem_mmap_gtt@medium-copy-xy -x gem_pwrite@huge-gtt -x gem_pwrite@big-cpu -x gem_pwrite@big-gtt -x gem_pwrite@huge-cpu
2. Wait a moment

Actual result:
--------------
2. A lot of tests are failing. Log shows a GPU hang and Failed to reset chip

Expected result:
----------------
2. No GPU hang
Comment 1 cprigent 2015-07-24 07:40:04 UTC
Log:
Jul 23 11:08:01 BSWM4 kernel: [ 2479.420274] [drm:intel_dp_start_link_train [i915]] *ERROR* too many full retries, give up
Jul 23 11:08:01 BSWM4 kernel: [ 2479.420706] [drm:intel_dp_complete_link_train [i915]] *ERROR* failed to train DP, aborting
Jul 23 11:08:01 BSWM4 kernel: [ 2479.420991] [drm:intel_enable_dp] Enabling DP audio on pipe C
Jul 23 11:08:01 BSWM4 kernel: [ 2479.420999] [drm:intel_audio_codec_enable] ELD on [CONNECTOR:50:DP-2], [ENCODER:49:TMDS-49]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.421006] [drm:ilk_audio_codec_enable] Enable audio codec on port D, pipe C, 32 bytes ELD
Jul 23 11:08:01 BSWM4 kernel: [ 2479.421303] [drm:intel_enable_pipe] enabling pipe C
Jul 23 11:08:01 BSWM4 kernel: [ 2479.421333] [drm:intel_psr_enable] PSR not supported by this panel
Jul 23 11:08:01 BSWM4 kernel: [ 2479.423478] [drm:vlv_pipe_set_fifo_size] Pipe C FIFO split 511 / 511 / 511
Jul 23 11:08:01 BSWM4 kernel: [ 2479.423506] [drm:vlv_update_wm] Setting FIFO watermarks - C: plane=391, cursor=63, sprite0=0, sprite1=0, SR: plane=0, cursor=0 level=0 cxsr=0
Jul 23 11:08:01 BSWM4 kernel: [ 2479.438191] [drm:vlv_pipe_set_fifo_size] Pipe C FIFO split 511 / 511 / 511
Jul 23 11:08:01 BSWM4 kernel: [ 2479.454935] [drm:intel_connector_check_state] [CONNECTOR:42:eDP-1]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.454967] [drm:intel_connector_check_state] [CONNECTOR:50:DP-2]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.454975] [drm:check_encoder_state] [ENCODER:34:TMDS-34]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.454983] [drm:check_encoder_state] [ENCODER:39:TMDS-39]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.454990] [drm:check_encoder_state] [ENCODER:41:TMDS-41]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.454997] [drm:check_encoder_state] [ENCODER:47:TMDS-47]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.455003] [drm:check_encoder_state] [ENCODER:49:TMDS-49]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.455011] [drm:check_crtc_state] [CRTC:21]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.465516] [drm:check_crtc_state] [CRTC:26]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.465524] [drm:check_crtc_state] [CRTC:31]
Jul 23 11:08:01 BSWM4 kernel: [ 2479.523922] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.523948] [drm:i915_pages_create_for_stolen] offset=0x0, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.525127] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.525140] [drm:i915_pages_create_for_stolen] offset=0x20000, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.525257] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.525269] [drm:i915_pages_create_for_stolen] offset=0x40000, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.525437] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.525456] [drm:i915_pages_create_for_stolen] offset=0x60000, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.642842] [drm:i915_gem_open] 
Jul 23 11:08:01 BSWM4 kernel: [ 2479.643284] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.643299] [drm:i915_pages_create_for_stolen] offset=0x0, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645154] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645171] [drm:i915_pages_create_for_stolen] offset=0x20000, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645297] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645309] [drm:i915_pages_create_for_stolen] offset=0x40000, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645487] [drm:i915_gem_object_create_stolen] creating stolen object: size=20000
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645499] [drm:i915_pages_create_for_stolen] offset=0x60000, size=131072
Jul 23 11:08:01 BSWM4 kernel: [ 2479.645953] [drm:i915_gem_open] 
Jul 23 11:08:01 BSWM4 kernel: [ 2479.647421] Setting dangerous option reset - tainting kernel
Jul 23 11:08:07 BSWM4 kernel: [ 2485.798500] [drm] stuck on render ring
Jul 23 11:08:07 BSWM4 kernel: [ 2485.828278] [drm] GPU HANG: ecode 8:0:0xe757fffe, in gem_eio [16000], reason: Ring hung, action: reset
Jul 23 11:08:07 BSWM4 kernel: [ 2485.828630] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Jul 23 11:08:07 BSWM4 kernel: [ 2485.828636] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Jul 23 11:08:07 BSWM4 kernel: [ 2485.828640] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Jul 23 11:08:07 BSWM4 kernel: [ 2485.828644] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Jul 23 11:08:07 BSWM4 kernel: [ 2485.828661] [drm] GPU crash dump saved to /sys/class/drm/card0/error
Jul 23 11:08:07 BSWM4 kernel: [ 2485.830201] [drm:i915_reset_and_wakeup] resetting chip
Jul 23 11:08:07 BSWM4 kernel: [ 2485.830954] drm/i915: Resetting chip after gpu hang
Jul 23 11:08:07 BSWM4 kernel: [ 2485.831005] [drm:i915_reset [i915]] *ERROR* Failed to reset chip: -19
Jul 23 11:08:07 BSWM4 kernel: [ 2485.834819] Setting dangerous option reset - tainting kernel

cat /sys/class/drm/card0/error
no error state collected
Comment 2 cprigent 2015-07-24 07:41:23 UTC
Created attachment 117335 [details]
Kernel configuration
Comment 3 cprigent 2015-07-24 07:49:32 UTC
Reproduced with kernel 4.2-rc2 from git://anongit.freedesktop.org/drm-intel tag drm-intel-testing-2015-07-17
Comment 4 cprigent 2015-07-24 08:38:09 UTC
Reproduced on BDW-U

Broadwell-U
Hardware
Platform: Wilson Beach DVT2 Ultrabook
CPU: Intel(R) CPU @ 1.60GHz (family: 6, model: 61 stepping: 3)
SoC: Broadwell 2+2 D0 (QDF : QGHA)
Chipset PCH: Wildcat Point B0
Software
Linux distribution: Ubuntu 14.10 LTS 64 bits
kernel 4.2-rc2 from git://anongit.freedesktop.org/drm-intel tag drm-intel-testing-2015-07-17
Mesa: mesa-10.6.2 from http://cgit.freedesktop.org/mesa/mesa/
Xf86_video_intel: 2.99.917 from
http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/
Libdrm: libdrm-2.4.62 from http://cgit.freedesktop.org/mesa/drm/
Cairo: 1.14.2 from http://cgit.freedesktop.org/cairo
libva: libva-1.6.0 from http://cgit.freedesktop.org/libva/
intel-driver: 1.6.0 from http://cgit.freedesktop.org/vaapi/intel-driver
xorg: 1.17.99 installed with script git_xorg.sh
Xserver: xorg-server-1.17.2 from http://cgit.freedesktop.org/xorg/xserver
Intel-gpu-tools: 8ad1e4077879a111f341dbfd2e0fee84efc9f57e from http://cgit.freedesktop.org/xorg/app/intel-gpu-tools/
Comment 5 cprigent 2015-07-28 16:47:52 UTC
Bug scrub:
Assign to Paul. Please assign to someone available.
Comment 6 cprigent 2015-07-28 16:49:20 UTC
To cprigent:
Try with a commercial device
Comment 7 Chris Wilson 2015-07-28 16:58:04 UTC
The culprit was misidentified.

*** This bug has been marked as a duplicate of bug 91349 ***
Comment 8 Jari Tahvanainen 2017-07-03 12:51:58 UTC
Closing resolved+duplicate as duplicate of closed+fixed.

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.