| Message ID | 20250224193115.2058512-1-imre.deak@intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | drm/dp: Fix link training interrupted by HPD pulse | expand |
Hi CI team, On Tue, Feb 25, 2025 at 01:41:55PM +0000, Patchwork wrote: > == Series Details == > > Series: drm/dp: Fix link training interrupted by HPD pulse (rev3) > URL : https://patchwork.freedesktop.org/series/145352/ > State : failure the issue reported below is unrelated, see my comment there, could you re-report this result? > == Summary == > > CI Bug Log - changes from CI_DRM_16178 -> Patchwork_145352v3 > ==================================================== > > Summary > ------- > > **FAILURE** > > Serious unknown changes coming with Patchwork_145352v3 absolutely need to be > verified manually. > > If you think the reported changes have nothing to do with the changes > introduced in Patchwork_145352v3, 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. > > External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/index.html > > Participating hosts (44 -> 43) > ------------------------------ > > Missing (1): fi-snb-2520m > > Possible new issues > ------------------- > > Here are the unknown changes that may have been introduced in Patchwork_145352v3: > > ### IGT changes ### > > #### Possible regressions #### > > * igt@kms_flip@basic-flip-vs-wf_vblank@b-dp2: > - fi-cfl-8109u: [PASS][1] -> [DMESG-WARN][2] +2 other tests dmesg-warn > [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/fi-cfl-8109u/igt@kms_flip@basic-flip-vs-wf_vblank@b-dp2.html > [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/fi-cfl-8109u/igt@kms_flip@basic-flip-vs-wf_vblank@b-dp2.html There is no HPD IRQ raised or handled on the above host, so the changes are not related to it. The issue is: <3> [158.833677] i915 0000:00:02.0: [drm] *ERROR* CPU pipe A FIFO underrun It seems to happen frequently, on the same host, leading to incorrect BAT failures. There is https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12935 but on a different host (MTL). > Known issues > ------------ > > Here are the changes found in Patchwork_145352v3 that come from known issues: > > ### IGT changes ### > > #### Issues hit #### > > * igt@i915_module_load@load: > - bat-mtlp-9: [PASS][3] -> [DMESG-WARN][4] ([i915#13494]) > [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-mtlp-9/igt@i915_module_load@load.html > [4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-mtlp-9/igt@i915_module_load@load.html > > * igt@i915_selftest@live: > - bat-twl-2: [PASS][5] -> [ABORT][6] ([i915#12435] / [i915#12919] / [i915#13503]) > [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-twl-2/igt@i915_selftest@live.html > [6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-twl-2/igt@i915_selftest@live.html > > * igt@i915_selftest@live@memory_region: > - bat-twl-2: [PASS][7] -> [ABORT][8] ([i915#12919]) > [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-twl-2/igt@i915_selftest@live@memory_region.html > [8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-twl-2/igt@i915_selftest@live@memory_region.html > > * igt@i915_selftest@live@requests: > - bat-atsm-1: [PASS][9] -> [INCOMPLETE][10] ([i915#12445]) > [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-atsm-1/igt@i915_selftest@live@requests.html > [10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-atsm-1/igt@i915_selftest@live@requests.html > > * igt@i915_selftest@live@workarounds: > - bat-arls-5: [PASS][11] -> [DMESG-FAIL][12] ([i915#12061]) +1 other test dmesg-fail > [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-arls-5/igt@i915_selftest@live@workarounds.html > [12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-arls-5/igt@i915_selftest@live@workarounds.html > > * igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence: > - bat-dg2-11: [PASS][13] -> [SKIP][14] ([i915#9197]) +3 other tests skip > [13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-dg2-11/igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence.html > [14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-dg2-11/igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence.html > > > #### Possible fixes #### > > * igt@i915_selftest@live@workarounds: > - bat-arlh-3: [DMESG-FAIL][15] ([i915#12061]) -> [PASS][16] +1 other test pass > [15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-arlh-3/igt@i915_selftest@live@workarounds.html > [16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-arlh-3/igt@i915_selftest@live@workarounds.html > > > #### Warnings #### > > * igt@i915_selftest@live: > - bat-atsm-1: [ABORT][17] ([i915#13679]) -> [INCOMPLETE][18] ([i915#12445]) > [17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-atsm-1/igt@i915_selftest@live.html > [18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-atsm-1/igt@i915_selftest@live.html > > > [i915#12061]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12061 > [i915#12435]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12435 > [i915#12445]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12445 > [i915#12919]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12919 > [i915#13494]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13494 > [i915#13503]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13503 > [i915#13679]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13679 > [i915#9197]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/9197 > > > Build changes > ------------- > > * Linux: CI_DRM_16178 -> Patchwork_145352v3 > > CI-20190529: 20190529 > CI_DRM_16178: d5debc4841e604372c74ebda0bbf54bd527c3475 @ git://anongit.freedesktop.org/gfx-ci/linux > IGT_8247: 8247 > Patchwork_145352v3: d5debc4841e604372c74ebda0bbf54bd527c3475 @ git://anongit.freedesktop.org/gfx-ci/linux > > == Logs == > > For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/index.html
Hi, https://patchwork.freedesktop.org/series/145352/ - Re-reported. i915.CI.BAT - Re-reported. Thanks, Ravali. -----Original Message----- From: I915-ci-infra <i915-ci-infra-bounces@lists.freedesktop.org> On Behalf Of Imre Deak Sent: 25 February 2025 19:55 To: I915-ci-infra@lists.freedesktop.org Cc: intel-gfx@lists.freedesktop.org Subject: Re: ✗ i915.CI.BAT: failure for drm/dp: Fix link training interrupted by HPD pulse (rev3) Hi CI team, On Tue, Feb 25, 2025 at 01:41:55PM +0000, Patchwork wrote: > == Series Details == > > Series: drm/dp: Fix link training interrupted by HPD pulse (rev3) > URL : https://patchwork.freedesktop.org/series/145352/ > State : failure the issue reported below is unrelated, see my comment there, could you re-report this result? > == Summary == > > CI Bug Log - changes from CI_DRM_16178 -> Patchwork_145352v3 > ==================================================== > > Summary > ------- > > **FAILURE** > > Serious unknown changes coming with Patchwork_145352v3 absolutely need to be > verified manually. > > If you think the reported changes have nothing to do with the changes > introduced in Patchwork_145352v3, 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. > > External URL: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/index.html > > Participating hosts (44 -> 43) > ------------------------------ > > Missing (1): fi-snb-2520m > > Possible new issues > ------------------- > > Here are the unknown changes that may have been introduced in Patchwork_145352v3: > > ### IGT changes ### > > #### Possible regressions #### > > * igt@kms_flip@basic-flip-vs-wf_vblank@b-dp2: > - fi-cfl-8109u: [PASS][1] -> [DMESG-WARN][2] +2 other tests dmesg-warn > [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/fi-cfl-8109u/igt@kms_flip@basic-flip-vs-wf_vblank@b-dp2.html > [2]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/fi-cfl-810 > 9u/igt@kms_flip@basic-flip-vs-wf_vblank@b-dp2.html There is no HPD IRQ raised or handled on the above host, so the changes are not related to it. The issue is: <3> [158.833677] i915 0000:00:02.0: [drm] *ERROR* CPU pipe A FIFO underrun It seems to happen frequently, on the same host, leading to incorrect BAT failures. There is https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12935 but on a different host (MTL). > Known issues > ------------ > > Here are the changes found in Patchwork_145352v3 that come from known issues: > > ### IGT changes ### > > #### Issues hit #### > > * igt@i915_module_load@load: > - bat-mtlp-9: [PASS][3] -> [DMESG-WARN][4] ([i915#13494]) > [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-mtlp-9/igt@i915_module_load@load.html > [4]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-mtlp-9 > /igt@i915_module_load@load.html > > * igt@i915_selftest@live: > - bat-twl-2: [PASS][5] -> [ABORT][6] ([i915#12435] / [i915#12919] / [i915#13503]) > [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-twl-2/igt@i915_selftest@live.html > [6]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-twl-2/ > igt@i915_selftest@live.html > > * igt@i915_selftest@live@memory_region: > - bat-twl-2: [PASS][7] -> [ABORT][8] ([i915#12919]) > [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-twl-2/igt@i915_selftest@live@memory_region.html > [8]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-twl-2/ > igt@i915_selftest@live@memory_region.html > > * igt@i915_selftest@live@requests: > - bat-atsm-1: [PASS][9] -> [INCOMPLETE][10] ([i915#12445]) > [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-atsm-1/igt@i915_selftest@live@requests.html > [10]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-atsm-1 > /igt@i915_selftest@live@requests.html > > * igt@i915_selftest@live@workarounds: > - bat-arls-5: [PASS][11] -> [DMESG-FAIL][12] ([i915#12061]) +1 other test dmesg-fail > [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-arls-5/igt@i915_selftest@live@workarounds.html > [12]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-arls-5 > /igt@i915_selftest@live@workarounds.html > > * igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence: > - bat-dg2-11: [PASS][13] -> [SKIP][14] ([i915#9197]) +3 other tests skip > [13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-dg2-11/igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence.html > [14]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-dg2-11 > /igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence.html > > > #### Possible fixes #### > > * igt@i915_selftest@live@workarounds: > - bat-arlh-3: [DMESG-FAIL][15] ([i915#12061]) -> [PASS][16] +1 other test pass > [15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-arlh-3/igt@i915_selftest@live@workarounds.html > [16]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-arlh-3 > /igt@i915_selftest@live@workarounds.html > > > #### Warnings #### > > * igt@i915_selftest@live: > - bat-atsm-1: [ABORT][17] ([i915#13679]) -> [INCOMPLETE][18] ([i915#12445]) > [17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_16178/bat-atsm-1/igt@i915_selftest@live.html > [18]: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/bat-atsm-1 > /igt@i915_selftest@live.html > > > [i915#12061]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12061 > [i915#12435]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12435 > [i915#12445]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12445 > [i915#12919]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12919 > [i915#13494]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13494 > [i915#13503]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13503 > [i915#13679]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13679 > [i915#9197]: > https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/9197 > > > Build changes > ------------- > > * Linux: CI_DRM_16178 -> Patchwork_145352v3 > > CI-20190529: 20190529 > CI_DRM_16178: d5debc4841e604372c74ebda0bbf54bd527c3475 @ git://anongit.freedesktop.org/gfx-ci/linux > IGT_8247: 8247 > Patchwork_145352v3: d5debc4841e604372c74ebda0bbf54bd527c3475 @ > git://anongit.freedesktop.org/gfx-ci/linux > > == Logs == > > For more details see: > https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_145352v3/index.html
On Mon, 24 Feb 2025, Imre Deak <imre.deak@intel.com> wrote: > This patchset fixes a problem during Display Port link training, where > the link training is interrupted by a short HPD pulse generated by the > sink. Generally the sink signals a bad link state via such short pulses, > but the sink should prevent such signaling during link training. > > Some sinks do generate a short pulse as above during link training, or a > pulse generated before the link training started may be handled from a > scheduled work during link training. Such a handling may interfere with > the link training (see patch 3 for the reason), so this patchset > prevents the handling of pulses during link training. I'm wondering if this series is made unnecessarily complicated by blocking the hotplug handling at the hpd pin level, while what is actually needed is blocking at the encoder level. I believe that would be simpler too. Thoughts? BR, Jani. > > Imre Deak (5): > drm/i915/hpd: Let an HPD pin be in the disabled state when handling > missed IRQs > drm/i915/hpd: Add support for suspending the IRQ handling on an HPD > pin > drm/i915/dp: Fix link training interrupted by a short HPD pulse > drm/i915/dp: Queue a link check after link training is complete > drm/i915/crt: Use intel_hpd_suspend/resume() instead of > intel_hpd_disable/enable() > > drivers/gpu/drm/i915/display/intel_crt.c | 7 +- > .../gpu/drm/i915/display/intel_display_core.h | 2 + > .../drm/i915/display/intel_dp_link_training.c | 23 +- > drivers/gpu/drm/i915/display/intel_hotplug.c | 264 ++++++++++++++---- > drivers/gpu/drm/i915/display/intel_hotplug.h | 5 +- > 5 files changed, 237 insertions(+), 64 deletions(-)
On Wed, Feb 26, 2025 at 02:55:32PM +0200, Jani Nikula wrote: > On Mon, 24 Feb 2025, Imre Deak <imre.deak@intel.com> wrote: > > This patchset fixes a problem during Display Port link training, where > > the link training is interrupted by a short HPD pulse generated by the > > sink. Generally the sink signals a bad link state via such short pulses, > > but the sink should prevent such signaling during link training. > > > > Some sinks do generate a short pulse as above during link training, or a > > pulse generated before the link training started may be handled from a > > scheduled work during link training. Such a handling may interfere with > > the link training (see patch 3 for the reason), so this patchset > > prevents the handling of pulses during link training. > > I'm wondering if this series is made unnecessarily complicated by > blocking the hotplug handling at the hpd pin level, while what is > actually needed is blocking at the encoder level. I believe that would > be simpler too. > > Thoughts? I think handling it at the HPD pin level makes more sense, since handling of the HPD IRQs should still happen after they are unblocked. Also in case of encoders sharing an HPD pin the handling of the IRQs should be blocked for all the encoders using this pin. The same block/unblock functionality is also needed for the CRT encoder, so it makes sense to extend that (intel_hpd_disable/enable) to support digital ports as well. > BR, > Jani. > > > > > Imre Deak (5): > > drm/i915/hpd: Let an HPD pin be in the disabled state when handling > > missed IRQs > > drm/i915/hpd: Add support for suspending the IRQ handling on an HPD > > pin > > drm/i915/dp: Fix link training interrupted by a short HPD pulse > > drm/i915/dp: Queue a link check after link training is complete > > drm/i915/crt: Use intel_hpd_suspend/resume() instead of > > intel_hpd_disable/enable() > > > > drivers/gpu/drm/i915/display/intel_crt.c | 7 +- > > .../gpu/drm/i915/display/intel_display_core.h | 2 + > > .../drm/i915/display/intel_dp_link_training.c | 23 +- > > drivers/gpu/drm/i915/display/intel_hotplug.c | 264 ++++++++++++++---- > > drivers/gpu/drm/i915/display/intel_hotplug.h | 5 +- > > 5 files changed, 237 insertions(+), 64 deletions(-) > > -- > Jani Nikula, Intel
This patchset fixes a problem during Display Port link training, where the link training is interrupted by a short HPD pulse generated by the sink. Generally the sink signals a bad link state via such short pulses, but the sink should prevent such signaling during link training. Some sinks do generate a short pulse as above during link training, or a pulse generated before the link training started may be handled from a scheduled work during link training. Such a handling may interfere with the link training (see patch 3 for the reason), so this patchset prevents the handling of pulses during link training. Imre Deak (5): drm/i915/hpd: Let an HPD pin be in the disabled state when handling missed IRQs drm/i915/hpd: Add support for suspending the IRQ handling on an HPD pin drm/i915/dp: Fix link training interrupted by a short HPD pulse drm/i915/dp: Queue a link check after link training is complete drm/i915/crt: Use intel_hpd_suspend/resume() instead of intel_hpd_disable/enable() drivers/gpu/drm/i915/display/intel_crt.c | 7 +- .../gpu/drm/i915/display/intel_display_core.h | 2 + .../drm/i915/display/intel_dp_link_training.c | 23 +- drivers/gpu/drm/i915/display/intel_hotplug.c | 264 ++++++++++++++---- drivers/gpu/drm/i915/display/intel_hotplug.h | 5 +- 5 files changed, 237 insertions(+), 64 deletions(-)