https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4009/shard-snb1/igt@drv_selftest@live_hangcheck.html dmesg: <5>[ 1124.912672] owatch: Using watchdog device /dev/watchdog0 <5>[ 1124.912727] owatch: Watchdog /dev/watchdog0 is a software watchdog <5>[ 1124.913273] owatch: timeout for /dev/watchdog0 set to 370 (requested 370) ... <4>[ 1221.407301] i915: probe of 0000:00:02.0 failed with error -25 <7>[ 1221.436150] [IGT] drv_selftest: exiting, ret=0 <7>[ 1221.498380] [IGT] drv_selftest: executing run.log: running: igt/drv_selftest/live_hangcheck [24/43] pass: 23, dmesg-warn: 1 | FATAL: command execution failed java.io.EOFException ... Finished: FAILURE Completed CI_IGT_test CI_DRM_3376/shard-snb1/35 : FAILURE CI_IGT_test runtime 451 seconds Rebooting shard-snb1 Note, SNB doesn't support pstore so it could be NMI/kernel OOPS. However, if it was OWATCH it would be noticed in run.log.
live_hangcheck should be capable of causing a machine hang in snb (because snb).
I will use this bug to pile up SNB incompletes timeout/system hang as for other machines.
https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4030/shard-snb6/igt@kms_flip@vblank-vs-hang.html <5>[ 5961.400179] owatch: Using watchdog device /dev/watchdog0 <5>[ 5961.400224] owatch: Watchdog /dev/watchdog0 is a software watchdog <5>[ 5961.400855] owatch: timeout for /dev/watchdog0 set to 370 (requested 370) ... <7>[ 6167.786992] [drm:verify_connector_state.isra.76 [i915]] [CONNECTOR:48:VGA-1] <7>[ 6167.787021] [drm:intel_atomic_commit_tail [i915]] [CRTC:47:pipe B] <7>[ 6167.787060] [drm:verify_single_dpll_state.isra.77 [i915]] PCH DPLL A followed by "stray" run.log: running: igt/kms_flip/vblank-vs-hang [59/75] skip: 35, pass: 24 \ FATAL: command execution failed java.io.EOFException ... Finished: FAILURE Completed CI_IGT_test CI_DRM_3427/shard-snb6/20 : FAILURE CI_IGT_test runtime 677 seconds Rebooting shard-snb6
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3439/shard-snb4/igt@gem_eio@suspend.html dmesg: <7>[ 1079.345572] [IGT] gem_eio: executing <4>[ 1079.361745] Setting dangerous option reset - tainting kernel <7>[ 1079.362293] [drm:i915_reset_device [i915]] resetting chip <5>[ 1079.362470] i915 0000:00:02.0: Resetting chip after gpu hang <7>[ 1079.362647] [drm:init_workarounds_ring [i915]] rcs0: Number of context specific w/a: 0 <7>[ 1079.363232] [IGT] gem_eio: starting subtest suspend <4>[ 1079.363287] Setting dangerous option reset - tainting kernel <7>[ 1079.363823] [drm:i915_reset_device [i915]] resetting chip <5>[ 1079.363887] i915 0000:00:02.0: Resetting chip after gpu hang <7>[ 1079.364026] [drm:i915_reset [i915]] GPU reset disabled run.log: Completed CI_IGT_test CI_DRM_3439/shard-snb4/33 : FAILURE CI_IGT_test runtime 154 seconds Rebooting shard-snb4
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3547/fi-snb-2600/igt@gem_exec_nop@basic-series.html
This is a Meta bug to capture all unexplained incompletes on SNB
Last seem: CI_DRM_3700: 2018-01-30 / 259 runs ago
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.