Analysis of fails:

* igt@kms_ccs@crc-sprite-planes-basic-4-tiled-bmg-ccs@pipe-a-dp-2:

This display test required a compressed framebuffer, but got uncompressed one. It is not possible to lint that to the series under test.

It is bad that we have a test which may fail due to KMDs decision. Someone from display team should make sure that test design is adjusted to KMDs logic.

* igt@kms_flip@2x-wf_vblank-ts-check-interruptible@ac-dp2-hdmi-a3:

I guess the test checks if vblank wait is interruptible? And it is, though it looks like the error path is not clean of errors. But this is display issue, unrelated to the series under test.

* igt@kms_flip@2x-wf_vblank-ts-check-interruptible@ad-dp2-hdmi-a3:

Another vblank interruption test, a different problem in error path. Display team should fix their error paths. Not related to series under test.

* igt@kms_plane_alpha_blend@constant-alpha-max:

That's new - not a display problem, but GuC problem. I don't think it's a good idea to try dumping GuC log to dmesg though - we should have separate mechanisms for capturing GuC log. Dumping 10MB of encoded binary data into dmesg does not look right, and makes the log harder to read. I don't think GuC should ever crash, so GuC team should take a look. Unrelated to series under test.

* igt@xe_fault_injection@vm-create-fail-xe_exec_queue_create_bind:

Despite the test name, issue did not happened during exec queue operations but during unbind. It looks like - not a surprise - display code caused a general protection fault. The display code seem to be unsafe during unbind. "sleeping function called from invalid context" - this should never happen. The display code is clearly broken. If an issue with such message became a sporadic, this also means the code lacks asserts.

Almost all fails being related to display support, could mean that display is way better tested than other blocks of the KMD. Unfortunately looking at the fails suggests that the reason is simpler - we have some serious problems within our display code. The code has invalid flows in error paths, and races in kmd unbind path. The sporadics will not go away until fixed.

I did not found any indication of any of the issues being related to the series under test. Also none of these cases reproduced in previous revision of the series.

(though some did reproduced two versions earlier, which means the fails aren't that rare - there is no excuse for not fixing a sporadic with 33% repro rate; in the series under test we worked hard to fix a 0.3% (3/1000) problem, and that was the repro rate after creating a custom crafted test case which maximized reproduction)

-Tomasz

Patch Details

Series:	drm/xe/vf: Fix exec queue creation during post-migration recovery (rev4)
URL:	https://patchwork.freedesktop.org/series/161287/
State:	failure
Details:	https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-161287v4/index.html

CI Bug Log - changes from xe-4626-340c78371713a8fdd88fbfe77fd0f7165294c31f_FULL -> xe-pw-161287v4_FULL

Summary

FAILURE

Serious unknown changes coming with xe-pw-161287v4_FULL absolutely need to be
verified manually.

If you think the reported changes have nothing to do with the changes
introduced in xe-pw-161287v4_FULL, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
to document this new failure mode, which will reduce false positives in CI.

Participating hosts (2 -> 2)

No changes in participating hosts

Possible new issues

Here are the unknown changes that may have been introduced in xe-pw-161287v4_FULL:

IGT changes

Possible regressions

• igt@kms_ccs@crc-sprite-planes-basic-4-tiled-bmg-ccs@pipe-a-dp-2:

  • shard-bmg: PASS -> FAIL +2 other tests fail

• igt@kms_flip@2x-wf_vblank-ts-check-interruptible@ac-dp2-hdmi-a3:

  • shard-bmg: PASS -> DMESG-FAIL

• igt@kms_flip@2x-wf_vblank-ts-check-interruptible@ad-dp2-hdmi-a3:

  • shard-bmg: PASS -> ABORT +2 other tests abort

• igt@kms_plane_alpha_blend@constant-alpha-max:

  • shard-bmg: PASS -> DMESG-WARN +2 other tests dmesg-warn

• igt@xe_fault_injection@vm-create-fail-xe_exec_queue_create_bind:

  • shard-bmg: NOTRUN -> ABORT +1 other test abort

Known issues

Here are the changes found in xe-pw-161287v4_FULL that come from known issues:

IGT changes

Issues hit

• igt@kms_big_fb@linear-32bpp-rotate-90:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2327)

• igt@kms_big_fb@yf-tiled-64bpp-rotate-90:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#1124) +1 other test skip

• igt@kms_bw@linear-tiling-2-displays-2560x1440p:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#367) +1 other test skip

• igt@kms_ccs@crc-sprite-planes-basic-4-tiled-dg2-rc-ccs:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2887) +2 other tests skip

• igt@kms_chamelium_color@ctm-limited-range:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2325)

• igt@kms_chamelium_color@degamma:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#306)

• igt@kms_chamelium_hpd@dp-hpd-with-enabled-mode:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2252) +4 other tests skip

• igt@kms_content_protection@legacy@pipe-a-dp-1:

  • shard-bmg: NOTRUN -> FAIL (Intel XE#3304) +1 other test fail

• igt@kms_content_protection@lic-type-1:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2341)

• igt@kms_content_protection@uevent-hdcp14:

  • shard-bmg: NOTRUN -> FAIL (Intel XE#6707) +1 other test fail

• igt@kms_cursor_crc@cursor-random-32x10:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#1424)

• igt@kms_cursor_legacy@cursora-vs-flipa-varying-size:

  • shard-bmg: PASS -> DMESG-WARN (Intel XE#5354)

• igt@kms_dirtyfb@drrs-dirtyfb-ioctl:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#1508)

• igt@kms_dp_link_training@uhbr-mst:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#4354)

• igt@kms_fbc_dirty_rect@fbc-dirty-rectangle-out-visible-area:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#4422)

• igt@kms_flip@2x-nonexisting-fb:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#1421)

• igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1:

  • shard-lnl: PASS -> FAIL (Intel XE#301) +1 other test fail

• igt@kms_flip@flip-vs-suspend-interruptible@d-dp2:

  • shard-bmg: PASS -> INCOMPLETE (Intel XE#2049 / Intel XE#2597)

• igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#7178)

• igt@kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-render:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2311) +9 other tests skip

• igt@kms_frontbuffer_tracking@drrs-abgr161616f-draw-mmap-wc:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#7061) +1 other test skip

• igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-pgflip-blt:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#4141) +3 other tests skip

• igt@kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-mmap-wc:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#6312 / Intel XE#651)

• igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2313) +10 other tests skip

• igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-move:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#656) +1 other test skip

• igt@kms_hdmi_inject@inject-audio:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#7308)

• igt@kms_joiner@basic-force-ultra-joiner:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#6911)

• igt@kms_plane@pixel-format-4-tiled-mtl-rc-ccs-cc-modifier:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#7283) +3 other tests skip

• igt@kms_plane_lowres@tiling-4@pipe-b-dp-2:

  • shard-bmg: NOTRUN -> ABORT (Intel XE#5545 / Intel XE#6652)

• igt@kms_plane_multiple@2x-tiling-y:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#5021)

• igt@kms_pm_dc@deep-pkgc:

  • shard-lnl: PASS -> FAIL (Intel XE#2029 / Intel XE#7314)

• igt@kms_psr2_sf@fbc-psr2-primary-plane-update-sf-dmg-area:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#1489)

• igt@kms_psr@fbc-psr-primary-render:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2234 / Intel XE#2850) +2 other tests skip

• igt@kms_rotation_crc@primary-y-tiled-reflect-x-270:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#3414 / Intel XE#3904)

• igt@kms_vrr@flip-suspend:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#1499) +1 other test skip

• igt@xe_eudebug@basic-vm-access-parameters-userptr:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#4837) +1 other test skip

• igt@xe_eudebug@discovery-empty:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#4837) +1 other test skip

• igt@xe_eudebug_online@pagefault-write-stress:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#6665 / Intel XE#6681)

• igt@xe_evict@evict-beng-cm-threads-small-multi-vm:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#6540 / Intel XE#688) +3 other tests skip

• igt@xe_evict@evict-beng-mixed-many-threads-small:

  • shard-bmg: PASS -> INCOMPLETE (Intel XE#6321)

• igt@xe_exec_balancer@many-execqueues-parallel-userptr-invalidate:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#7482) +1 other test skip

• igt@xe_exec_basic@multigpu-many-execqueues-many-vm-rebind:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2322) +1 other test skip

• igt@xe_exec_basic@multigpu-no-exec-bindexecqueue-rebind:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#1392)

• igt@xe_exec_fault_mode@many-execqueues-multi-queue-userptr:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#7136) +2 other tests skip

• igt@xe_exec_fault_mode@twice-multi-queue-rebind-prefetch:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#7136) +2 other tests skip

• igt@xe_exec_multi_queue@many-queues-preempt-mode-priority-smem:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#6874) +13 other tests skip

• igt@xe_exec_multi_queue@one-queue-preempt-mode-fault-dyn-priority:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#6874)

• igt@xe_exec_threads@threads-multi-queue-cm-shared-vm-userptr-invalidate:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#7138) +3 other tests skip

• igt@xe_multigpu_svm@mgpu-atomic-op-basic:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#6964)

• igt@xe_multigpu_svm@mgpu-coherency-fail-prefetch:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#6964) +1 other test skip

• igt@xe_oa@oa-tlb-invalidate:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2248 / Intel XE#7325)

• igt@xe_pat@pat-index-xehpc:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#1420)

• igt@xe_pm@d3cold-basic-exec:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#2284)

• igt@xe_pxp@pxp-termination-key-update-post-termination-irq:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#4733)

• igt@xe_query@multigpu-query-pxp-status:

  • shard-bmg: NOTRUN -> SKIP (Intel XE#944) +1 other test skip

• igt@xe_sriov_auto_provisioning@selfconfig-reprovision-increase-numvfs:

  • shard-bmg: PASS -> FAIL (Intel XE#5937) +1 other test fail

• igt@xe_sriov_vram@vf-access-after-resize-up:

  • shard-lnl: NOTRUN -> SKIP (Intel XE#6376 / Intel XE#7330)

Possible fixes

• igt@kms_flip@flip-vs-suspend@c-hdmi-a3:

  • shard-bmg: INCOMPLETE (Intel XE#2049 / Intel XE#2597) -> PASS +1 other test pass

• igt@kms_vrr@seamless-rr-switch-virtual@pipe-a-edp-1:

  • shard-lnl: FAIL (Intel XE#2142) -> PASS +1 other test pass

• igt@xe_evict@evict-mixed-many-threads-small:

  • shard-bmg: INCOMPLETE (Intel XE#6321) -> PASS

• igt@xe_exec_system_allocator@fault-process-same-page-benchmark:

  • shard-bmg: FAIL -> PASS

• igt@xe_exec_system_allocator@pat-index-madvise-pat-idx-uc-multi-vma:

  • shard-lnl: FAIL (Intel XE#5625) -> PASS

• igt@xe_wedged@basic-wedged:

  • shard-bmg: ABORT -> PASS

Warnings

• igt@kms_tiled_display@basic-test-pattern:

  • shard-bmg: FAIL (Intel XE#1729) -> SKIP (Intel XE#2426)

• igt@xe_fault_injection@probe-fail-guc-xe_guc_ct_send_recv:

  • shard-bmg: ABORT (Intel XE#5466) -> ABORT (Intel XE#5466 / Intel XE#6652)

Build changes

• Linux: xe-4626-340c78371713a8fdd88fbfe77fd0f7165294c31f -> xe-pw-161287v4

IGT_8774: 8892452285d691f612ecb3b2a6fefc50983af15f @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-4626-340c78371713a8fdd88fbfe77fd0f7165294c31f: 340c78371713a8fdd88fbfe77fd0f7165294c31f
xe-pw-161287v4: 161287v4