Summary: | [CI][BAT]igt@i915_selftest@live_execlists - incomplete - i915_request_retire: bcs0 fence 18b59:382, current 382 | ||
---|---|---|---|
Product: | DRI | Reporter: | Lakshmi <lakshminarayana.vudum> |
Component: | DRM/Intel | Assignee: | Intel GFX Bugs mailing list <intel-gfx-bugs> |
Status: | RESOLVED FIXED | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> |
Severity: | not set | ||
Priority: | not set | CC: | intel-gfx-bugs, nagendra.kanagala |
Version: | DRI git | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | SKL | i915 features: | GEM/Other |
Description
Lakshmi
2019-11-21 08:04:13 UTC
The CI Bug Log issue associated to this bug has been updated. ### New filters associated * SKL: igt@i915_selftest@live_execlists - incomplete - i915_request_retire: bcs0 fence 18b59:382, current 382 - https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7387/fi-skl-6770hq/igt@i915_selftest@live_execlists.html We have to look at the pstore to get the final death throes... <0>[ 534.577515] i915_sel-5332 7d..1 521975229us : trace_ports: vcs1: submit { 19005:2, 19004:2 } <0>[ 534.577567] <idle>-0 1..s1 521975268us : process_csb: vcs1 cs-irq head=5, tail=0 <0>[ 534.577619] <idle>-0 1..s1 521975269us : process_csb: vcs1 csb[0]: status=0x00000802:0x00000060 <0>[ 534.577670] <idle>-0 1..s1 521975270us : trace_ports: vcs1: preempted { 19004:2*, 0:0 } <0>[ 534.577719] <idle>-0 1..s1 521975271us : trace_ports: vcs1: promote { 19005:2!, 19004:2 } <0>[ 534.577784] i915_sel-5332 7.... 521975305us : i915_request_retire: vcs1 fence 19004:2, current 2 <0>[ 534.577836] i915_sel-5332 7.... 521975308us : intel_context_unpin: vcs1 context:19004 retire <0>[ 534.577886] i915_sel-5332 7.... 521975311us : __intel_context_retire: vcs1 context:19004 retire <0>[ 534.577944] <idle>-0 1..s1 521975324us : assert_pending_valid: Unpinned context:19004 in pending[1] <0>[ 534.577997] <idle>-0 1..s1 521975329us : process_csb: process_csb:2166 GEM_BUG_ON(!assert_pending_valid(execlists, "promote")) <0>[ 534.578002] --------------------------------- Which looks like a valid race in the debug; that is the HW managed to complete and we processed that completion on another CPU, before we were able to perform our assertions! *** Bug 112362 has been marked as a duplicate of this bug. *** commit c95d31c3df1b9ff29bcd85c47324e16d430bbd5a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 21 10:35:46 2019 +0000 drm/i915/execlists: Lock the request while validating it during promotion Since the request is already on the HW as we perform its validation, it and even its subsequent barrier may be concurrently retired before we process the assertions. If it is retired already and so off the HW, our assertions become void and we need to ignore them. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=112363 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191121103546.146487-1-chris@chris-wilson.co.uk |
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.