RE: ✗ Fi.CI.BAT: failure for drm/i915: Add Display Port tunnel BW allocation support (rev6)

Tue Feb 27 11:34:01 UTC 2024

Hi,

https://patchwork.freedesktop.org/series/129082/ - Re-reported.

Thanks,
Tejasree

-----Original Message-----
From: I915-ci-infra <i915-ci-infra-bounces at lists.freedesktop.org> On Behalf Of Imre Deak
Sent: Tuesday, February 27, 2024 4:29 PM
To: intel-gfx at lists.freedesktop.org; I915-ci-infra at lists.freedesktop.org
Cc: Saarinen, Jani <jani.saarinen at intel.com>
Subject: Re: ✗ Fi.CI.BAT: failure for drm/i915: Add Display Port tunnel BW allocation support (rev6)

On Tue, Feb 27, 2024 at 01:18:36AM +0000, Patchwork wrote:
> == Series Details ==
> 
> Series: drm/i915: Add Display Port tunnel BW allocation support (rev6)
> URL   : https://patchwork.freedesktop.org/series/129082/
> State : failure
> 
> == Summary ==
> 
> CI Bug Log - changes from CI_DRM_14345 -> Patchwork_129082v6 
> ====================================================
> 
> Summary
> -------
> 
>   **FAILURE**
> 
>   Serious unknown changes coming with Patchwork_129082v6 absolutely need to be
>   verified manually.
>   
>   If you think the reported changes have nothing to do with the changes
>   introduced in Patchwork_129082v6, please notify your bug team (I915-ci-infra at lists.freedesktop.org) to allow them
>   to document this new failure mode, which will reduce false positives in CI.
> 
>   External URL: 
> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_129082v6/index.html
> 
> Participating hosts (39 -> 38)
> ------------------------------
> 
>   Missing    (1): fi-snb-2520m 
> 
> Possible new issues
> -------------------
> 
>   Here are the unknown changes that may have been introduced in Patchwork_129082v6:
> 
> ### IGT changes ###
> 
> #### Possible regressions ####
> 
>   * igt at kms_pipe_crc_basic@hang-read-crc at pipe-d-edp-1:
>     - bat-arls-2:         [PASS][1] -> [ABORT][2]
>    [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14345/bat-arls-2/igt@kms_pipe_crc_basic@hang-read-crc@pipe-d-edp-1.html
>    [2]: 
> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_129082v6/bat-arls-2
> /igt at kms_pipe_crc_basic@hang-read-crc at pipe-d-edp-1.html

It's an unrelated issue:

There is only an eDP panel connected to the above ARL, so no TBT/DP tunnel where the patchset would have an effect.

Looks like some core memory management issue:

<0> [223.728871] BUG: spinlock bad magic on CPU#1, kworker/1:1/224 <0> [223.728880]  lock: obj_hash+0xec060/0x120000, .magic: d6ad4ead, .owner: <none>/-1, .owner_cpu: -1 <4> [223.728890] CPU: 1 PID: 224 Comm: kworker/1:1 Not tainted 6.8.0-rc6-Patchwork_129082v6-gb3552bf1e79d+ #1 <4> [223.728894] Hardware name: Intel Corporation Arrow Lake Client Platform/MTL-S UDIMM 2DPC EVCRB, BIOS MTLSFWI1.R00.3473.D80.2311222130 11/22/2023 <4> [223.728898] Workqueue: events delayed_fput <7> [223.728778] i915 0000:00:02.0: [drm:pipe_config_mismatch [i915]] [CRTC:233:pipe D] fastset requirement not met in hw.adjusted_mode.crtc_htotal (expected 3040, found 0) <4> [223.728906] Call Trace:
<4> [223.728909]  <TASK>
<4> [223.728911]  dump_stack_lvl+0x64/0xb0 <4> [223.728919]  do_raw_spin_lock+0x74/0xc0 <4> [223.728925]  _raw_spin_lock_irqsave+0x45/0x60 <4> [223.728931]  ? debug_check_no_obj_freed+0xcc/0x240
<4> [223.728936]  debug_check_no_obj_freed+0xcc/0x240
<4> [223.728940]  ? lock_acquire+0xd8/0x2d0 <4> [223.728943]  ? find_held_lock+0x2b/0x80 <4> [223.728949]  free_unref_page_prepare+0xec/0x350
<4> [223.728955]  free_unref_page+0x33/0x160 <4> [223.728959]  release_pages+0x6c/0x580 <4> [223.728966]  __folio_batch_release+0x1a/0x30 <4> [223.728970]  shmem_undo_range+0x248/0x6c0 <4> [223.728977]  ? cpu_attach_domain+0x74f/0x9a0 <4> [223.728982]  shmem_truncate_range+0x14/0x40 <4> [223.728985]  shmem_evict_inode+0xd7/0x270 <4> [223.728989]  evict+0xc8/0x170

probably CI noticing this and aborting with:

[224.019721] Subtest hang-read-crc: SUCCESS (11.568s [224.948083] Aborting: Lockdep not active

> 
>   
> Known issues
> ------------
> 
>   Here are the changes found in Patchwork_129082v6 that come from known issues:
> 
> ### CI changes ###
> 
> #### Issues hit ####
> 
>   * boot:
>     - fi-cfl-8109u:       [PASS][3] -> [FAIL][4] ([i915#8293])
>    [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14345/fi-cfl-8109u/boot.html
>    [4]: 
> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_129082v6/fi-cfl-810
> 9u/boot.html
> 
>   
> 
> ### IGT changes ###
> 
> #### Possible fixes ####
> 
>   * igt at i915_selftest@live at gem_contexts:
>     - bat-atsm-1:         [INCOMPLETE][5] ([i915#10094] / [i915#10137]) -> [PASS][6]
>    [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14345/bat-atsm-1/igt@i915_selftest@live@gem_contexts.html
>    [6]: 
> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_129082v6/bat-atsm-1
> /igt at i915_selftest@live at gem_contexts.html
> 
>   * igt at i915_selftest@live at hangcheck:
>     - {bat-rpls-3}:       [DMESG-WARN][7] ([i915#5591]) -> [PASS][8]
>    [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14345/bat-rpls-3/igt@i915_selftest@live@hangcheck.html
>    [8]: 
> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_129082v6/bat-rpls-3
> /igt at i915_selftest@live at hangcheck.html
> 
>   
>   {name}: This element is suppressed. This means it is ignored when computing
>           the status of the difference (SUCCESS, WARNING, or FAILURE).
> 
>   [i915#10094]: https://gitlab.freedesktop.org/drm/intel/issues/10094
>   [i915#10137]: https://gitlab.freedesktop.org/drm/intel/issues/10137
>   [i915#5591]: https://gitlab.freedesktop.org/drm/intel/issues/5591
>   [i915#8293]: https://gitlab.freedesktop.org/drm/intel/issues/8293
> 
> 
> Build changes
> -------------
> 
>   * Linux: CI_DRM_14345 -> Patchwork_129082v6
> 
>   CI-20190529: 20190529
>   CI_DRM_14345: b3552bf1e79de200a6ca49c518ead46a1256618c @ git://anongit.freedesktop.org/gfx-ci/linux
>   IGT_7731: 17f897a81868fb35c6a7033a8b07256659742248 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
>   Patchwork_129082v6: b3552bf1e79de200a6ca49c518ead46a1256618c @ 
> git://anongit.freedesktop.org/gfx-ci/linux
> 
> 
> ### Linux commits
> 
> 6f0303321631 drm/i915/dp: Enable DP tunnel BW allocation mode
> 8b160bd78946 drm/i915/dp: Read DPRX for all long HPD pulses
> 84f928739de0 drm/i915/dp: Suspend/resume DP tunnels 232a81c3545b 
> drm/i915/dp: Call intel_dp_sync_state() always for DDI DP encoders 
> d6e68e72eeea drm/i915/dp: Handle DP tunnel IRQs 77a316a5547c 
> drm/i915/dp: Allocate/free DP tunnel BW in the encoder enable/disable 
> hooks 904f8ff4847c drm/i915/dp: Compute DP tunnel BW during encoder 
> state computation
> e30c2b294042 drm/i915/dp: Account for tunnel BW limit in 
> intel_dp_max_link_data_rate()
> cfe003896226 drm/i915/dp: Add DP tunnel atomic state and check BW 
> limit
> e845d3e65f57 drm/i915/dp: Add support for DP tunnel BW allocation
> 1e491673dd40 drm/i915/dp: Sync instead of try-sync commits when 
> getting active pipes f9a23c72455c drm/i915/dp: Add 
> intel_dp_max_link_data_rate() d02f06d5e00c drm/i915/dp: Factor out 
> intel_dp_read_dprx_caps()
> 95a664a77474 drm/i915/dp: Factor out intel_dp_update_sink_caps()
> e587647c4a35 drm/i915/dp: Export intel_dp_max_common_rate/lane_count()
> cc6689daacbe drm/i915/dp: Factor out intel_dp_config_required_rate()
> e76c8cac43f0 drm/i915/dp: Use drm_dp_max_dprx_data_rate()
> 8fea4c43fb32 drm/i915/dp: Add support to notify MST connectors to 
> retry modesets
> a50d19596539 drm/i915: Fix display bpp limit computation during system 
> resume
> cfae6a8751d6 drm/dp: Add support for DP tunneling 5bed9d5b20dc drm/dp: 
> Add drm_dp_max_dprx_data_rate()
> 
> == Logs ==
> 
> For more details see: 
> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_129082v6/index.html