| Message ID | cover.1657522157.git.mchehab@kernel.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | Fix TLB invalidate issues with Broadwell | expand |
On Mon, Jul 11, 2022 at 08:22:33AM -0000, Patchwork wrote: > Patch Details > > Series: Fix TLB invalidate issues with Broadwell (rev5) > URL: [1]https://patchwork.freedesktop.org/series/105167/ > State: failure > Details: > [2]https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/index.ht > ml > > CI Bug Log - changes from CI_DRM_11862 -> Patchwork_105167v5 > > Summary > > FAILURE > > Serious unknown changes coming with Patchwork_105167v5 absolutely need > to be > verified manually. > > If you think the reported changes have nothing to do with the changes > introduced in Patchwork_105167v5, please notify your bug team 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_105167v5/index.html > > Participating hosts (45 -> 42) > > Additional (2): bat-rpls-1 bat-dg1-5 > Missing (5): fi-cml-u2 fi-hsw-4200u fi-icl-u2 fi-ctg-p8600 bat-jsl-3 > > Possible new issues > > Here are the unknown changes that may have been introduced in > Patchwork_105167v5: > > IGT changes > > Possible regressions > > * igt@i915_selftest@live@gem: > + fi-elk-e7500: [3]PASS -> [4]DMESG-FAIL I also believe this is a false positive... triggered a retest... hoping to get this series merged today... > > Suppressed > > The following results come from untrusted machines, tests, or statuses. > They do not affect the overall result. > * igt@i915_selftest@live@guc: > + {bat-rpls-1}: NOTRUN -> [5]DMESG-WARN > > Known issues > > Here are the changes found in Patchwork_105167v5 that come from known > issues: > > IGT changes > > Issues hit > > * igt@fbdev@nullptr: > + bat-dg1-5: NOTRUN -> [6]SKIP ([7]i915#2582) +4 similar issues > * igt@gem_mmap@basic: > + bat-dg1-5: NOTRUN -> [8]SKIP ([9]i915#4083) > * igt@gem_tiled_blits@basic: > + bat-dg1-5: NOTRUN -> [10]SKIP ([11]i915#4077) +2 similar > issues > * igt@gem_tiled_pread_basic: > + bat-dg1-5: NOTRUN -> [12]SKIP ([13]i915#4079) +1 similar issue > * igt@i915_pm_backlight@basic-brightness: > + bat-dg1-5: NOTRUN -> [14]SKIP ([15]i915#1155) > * igt@i915_suspend@basic-s2idle-without-i915: > + bat-dg1-5: NOTRUN -> [16]INCOMPLETE ([17]i915#6011) > * igt@kms_addfb_basic@basic-x-tiled-legacy: > + bat-dg1-5: NOTRUN -> [18]SKIP ([19]i915#4212) +7 similar > issues > * igt@kms_addfb_basic@basic-y-tiled-legacy: > + bat-dg1-5: NOTRUN -> [20]SKIP ([21]i915#4215) > * igt@kms_busy@basic: > + bat-dg1-5: NOTRUN -> [22]SKIP ([23]i915#1845 / [24]i915#4303) > * igt@kms_chamelium@common-hpd-after-suspend: > + fi-snb-2600: NOTRUN -> [25]SKIP ([26]fdo#109271 / > [27]fdo#111827) > + fi-rkl-11600: NOTRUN -> [28]SKIP ([29]fdo#111827) > * igt@kms_chamelium@dp-crc-fast: > + bat-dg1-5: NOTRUN -> [30]SKIP ([31]fdo#111827) +7 similar > issues > * igt@kms_flip@basic-flip-vs-modeset@b-edp1: > + bat-adlp-4: [32]PASS -> [33]DMESG-WARN ([34]i915#3576) +1 > similar issue > * igt@kms_force_connector_basic@force-load-detect: > + bat-dg1-5: NOTRUN -> [35]SKIP ([36]fdo#109285) > * igt@kms_pipe_crc_basic@nonblocking-crc: > + bat-dg1-5: NOTRUN -> [37]SKIP ([38]i915#4078) +13 similar > issues > * igt@kms_psr@primary_page_flip: > + bat-dg1-5: NOTRUN -> [39]SKIP ([40]i915#1072 / [41]i915#4078) > +3 similar issues > * igt@kms_setmode@basic-clone-single-crtc: > + bat-dg1-5: NOTRUN -> [42]SKIP ([43]i915#3555) > * igt@prime_vgem@basic-fence-flip: > + bat-dg1-5: NOTRUN -> [44]SKIP ([45]i915#1845 / [46]i915#3708) > * igt@prime_vgem@basic-fence-read: > + bat-dg1-5: NOTRUN -> [47]SKIP ([48]i915#3708) +2 similar > issues > * igt@prime_vgem@basic-gtt: > + bat-dg1-5: NOTRUN -> [49]SKIP ([50]i915#3708 / [51]i915#4077) > +1 similar issue > * igt@prime_vgem@basic-userptr: > + bat-dg1-5: NOTRUN -> [52]SKIP ([53]i915#3708 / [54]i915#4873) > * igt@runner@aborted: > + fi-elk-e7500: NOTRUN -> [55]FAIL ([56]fdo#109271 / > [57]i915#4312) > + bat-dg1-5: NOTRUN -> [58]FAIL ([59]i915#4312 / [60]i915#5257) > > Possible fixes > > * igt@i915_selftest@live@hangcheck: > + fi-snb-2600: [61]INCOMPLETE ([62]i915#3921) -> [63]PASS > + bat-dg1-6: [64]DMESG-FAIL ([65]i915#4494 / [66]i915#4957) -> > [67]PASS > * igt@i915_suspend@basic-s3-without-i915: > + fi-rkl-11600: [68]INCOMPLETE ([69]i915#5982) -> [70]PASS > * igt@vgem_basic@setversion: > + fi-kbl-soraka: [71]INCOMPLETE -> [72]PASS > > {name}: This element is suppressed. This means it is ignored when > computing > the status of the difference (SUCCESS, WARNING, or FAILURE). > > Build changes > > * Linux: CI_DRM_11862 -> Patchwork_105167v5 > > CI-20190529: 20190529 > CI_DRM_11862: ffee806d103b9604db7eb9cd689c098aca1ffa96 @ > git://anongit.freedesktop.org/gfx-ci/linux > IGT_6563: 7d43b49bf10788d4870668f93a800888fc8ab339 @ > https://gitlab.freedesktop.org/drm/igt-gpu-tools.git > Patchwork_105167v5: ffee806d103b9604db7eb9cd689c098aca1ffa96 @ > git://anongit.freedesktop.org/gfx-ci/linux > > Linux commits > > d83b60fff94b drm/i915/gt: Serialize TLB invalidates with GT resets > 0d56aa1c1499 drm/i915/gt: Serialize GRDOM access between multiple > engine resets > > References > > 1. https://patchwork.freedesktop.org/series/105167/ > 2. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/index.html > 3. https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11862/fi-elk-e7500/igt@i915_selftest@live@gem.html > 4. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-elk-e7500/igt@i915_selftest@live@gem.html > 5. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-rpls-1/igt@i915_selftest@live@guc.html > 6. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@fbdev@nullptr.html > 7. https://gitlab.freedesktop.org/drm/intel/issues/2582 > 8. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@gem_mmap@basic.html > 9. https://gitlab.freedesktop.org/drm/intel/issues/4083 > 10. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@gem_tiled_blits@basic.html > 11. https://gitlab.freedesktop.org/drm/intel/issues/4077 > 12. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@gem_tiled_pread_basic.html > 13. https://gitlab.freedesktop.org/drm/intel/issues/4079 > 14. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@i915_pm_backlight@basic-brightness.html > 15. https://gitlab.freedesktop.org/drm/intel/issues/1155 > 16. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@i915_suspend@basic-s2idle-without-i915.html > 17. https://gitlab.freedesktop.org/drm/intel/issues/6011 > 18. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_addfb_basic@basic-x-tiled-legacy.html > 19. https://gitlab.freedesktop.org/drm/intel/issues/4212 > 20. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_addfb_basic@basic-y-tiled-legacy.html > 21. https://gitlab.freedesktop.org/drm/intel/issues/4215 > 22. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_busy@basic.html > 23. https://gitlab.freedesktop.org/drm/intel/issues/1845 > 24. https://gitlab.freedesktop.org/drm/intel/issues/4303 > 25. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-snb-2600/igt@kms_chamelium@common-hpd-after-suspend.html > 26. https://bugs.freedesktop.org/show_bug.cgi?id=109271 > 27. https://bugs.freedesktop.org/show_bug.cgi?id=111827 > 28. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-rkl-11600/igt@kms_chamelium@common-hpd-after-suspend.html > 29. https://bugs.freedesktop.org/show_bug.cgi?id=111827 > 30. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_chamelium@dp-crc-fast.html > 31. https://bugs.freedesktop.org/show_bug.cgi?id=111827 > 32. https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11862/bat-adlp-4/igt@kms_flip@basic-flip-vs-modeset@b-edp1.html > 33. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-adlp-4/igt@kms_flip@basic-flip-vs-modeset@b-edp1.html > 34. https://gitlab.freedesktop.org/drm/intel/issues/3576 > 35. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_force_connector_basic@force-load-detect.html > 36. https://bugs.freedesktop.org/show_bug.cgi?id=109285 > 37. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_pipe_crc_basic@nonblocking-crc.html > 38. https://gitlab.freedesktop.org/drm/intel/issues/4078 > 39. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_psr@primary_page_flip.html > 40. https://gitlab.freedesktop.org/drm/intel/issues/1072 > 41. https://gitlab.freedesktop.org/drm/intel/issues/4078 > 42. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@kms_setmode@basic-clone-single-crtc.html > 43. https://gitlab.freedesktop.org/drm/intel/issues/3555 > 44. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@prime_vgem@basic-fence-flip.html > 45. https://gitlab.freedesktop.org/drm/intel/issues/1845 > 46. https://gitlab.freedesktop.org/drm/intel/issues/3708 > 47. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@prime_vgem@basic-fence-read.html > 48. https://gitlab.freedesktop.org/drm/intel/issues/3708 > 49. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@prime_vgem@basic-gtt.html > 50. https://gitlab.freedesktop.org/drm/intel/issues/3708 > 51. https://gitlab.freedesktop.org/drm/intel/issues/4077 > 52. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@prime_vgem@basic-userptr.html > 53. https://gitlab.freedesktop.org/drm/intel/issues/3708 > 54. https://gitlab.freedesktop.org/drm/intel/issues/4873 > 55. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-elk-e7500/igt@runner@aborted.html > 56. https://bugs.freedesktop.org/show_bug.cgi?id=109271 > 57. https://gitlab.freedesktop.org/drm/intel/issues/4312 > 58. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-5/igt@runner@aborted.html > 59. https://gitlab.freedesktop.org/drm/intel/issues/4312 > 60. https://gitlab.freedesktop.org/drm/intel/issues/5257 > 61. https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11862/fi-snb-2600/igt@i915_selftest@live@hangcheck.html > 62. https://gitlab.freedesktop.org/drm/intel/issues/3921 > 63. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-snb-2600/igt@i915_selftest@live@hangcheck.html > 64. https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11862/bat-dg1-6/igt@i915_selftest@live@hangcheck.html > 65. https://gitlab.freedesktop.org/drm/intel/issues/4494 > 66. https://gitlab.freedesktop.org/drm/intel/issues/4957 > 67. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/bat-dg1-6/igt@i915_selftest@live@hangcheck.html > 68. https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11862/fi-rkl-11600/igt@i915_suspend@basic-s3-without-i915.html > 69. https://gitlab.freedesktop.org/drm/intel/issues/5982 > 70. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-rkl-11600/igt@i915_suspend@basic-s3-without-i915.html > 71. https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_11862/fi-kbl-soraka/igt@vgem_basic@setversion.html > 72. https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/fi-kbl-soraka/igt@vgem_basic@setversion.html
On Mon, 11 Jul 2022 10:06:38 -0400 Rodrigo Vivi <rodrigo.vivi@intel.com> wrote: > On Mon, Jul 11, 2022 at 08:22:33AM -0000, Patchwork wrote: > > Patch Details > > > > Series: Fix TLB invalidate issues with Broadwell (rev5) > > URL: [1]https://patchwork.freedesktop.org/series/105167/ > > State: failure > > Details: > > [2]https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/index.ht > > ml > > > > CI Bug Log - changes from CI_DRM_11862 -> Patchwork_105167v5 > > > > Summary > > > > FAILURE > > > > Serious unknown changes coming with Patchwork_105167v5 absolutely need > > to be > > verified manually. > > > > If you think the reported changes have nothing to do with the changes > > introduced in Patchwork_105167v5, please notify your bug team 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_105167v5/index.html > > > > Participating hosts (45 -> 42) > > > > Additional (2): bat-rpls-1 bat-dg1-5 > > Missing (5): fi-cml-u2 fi-hsw-4200u fi-icl-u2 fi-ctg-p8600 bat-jsl-3 > > > > Possible new issues > > > > Here are the unknown changes that may have been introduced in > > Patchwork_105167v5: > > > > IGT changes > > > > Possible regressions > > > > * igt@i915_selftest@live@gem: > > + fi-elk-e7500: [3]PASS -> [4]DMESG-FAIL > > I also believe this is a false positive... > triggered a retest... Yeah, the latest test gave success: Subject: [Intel-gfx] ✓ Fi.CI.BAT: success for Fix TLB invalidate issues with Broadwell (rev6) Date: Mon, 11 Jul 2022 14:53:40 -0000 Reply-To: intel-gfx@lists.freedesktop.org Sender: "Intel-gfx" <intel-gfx-bounces@lists.freedesktop.org> == Series Details == Series: Fix TLB invalidate issues with Broadwell (rev6) URL : https://patchwork.freedesktop.org/series/105167/ State : success > > hoping to get this series merged today... Yeah, it would be great to have this merged, as it is a regression fix. Regards, Mauro
On Mon, Jul 11, 2022 at 05:11:23PM +0200, Mauro Carvalho Chehab wrote: > On Mon, 11 Jul 2022 10:06:38 -0400 > Rodrigo Vivi <rodrigo.vivi@intel.com> wrote: > > > On Mon, Jul 11, 2022 at 08:22:33AM -0000, Patchwork wrote: > > > Patch Details > > > > > > Series: Fix TLB invalidate issues with Broadwell (rev5) > > > URL: [1]https://patchwork.freedesktop.org/series/105167/ > > > State: failure > > > Details: > > > [2]https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_105167v5/index.ht > > > ml > > > > > > CI Bug Log - changes from CI_DRM_11862 -> Patchwork_105167v5 > > > > > > Summary > > > > > > FAILURE > > > > > > Serious unknown changes coming with Patchwork_105167v5 absolutely need > > > to be > > > verified manually. > > > > > > If you think the reported changes have nothing to do with the changes > > > introduced in Patchwork_105167v5, please notify your bug team 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_105167v5/index.html > > > > > > Participating hosts (45 -> 42) > > > > > > Additional (2): bat-rpls-1 bat-dg1-5 > > > Missing (5): fi-cml-u2 fi-hsw-4200u fi-icl-u2 fi-ctg-p8600 bat-jsl-3 > > > > > > Possible new issues > > > > > > Here are the unknown changes that may have been introduced in > > > Patchwork_105167v5: > > > > > > IGT changes > > > > > > Possible regressions > > > > > > * igt@i915_selftest@live@gem: > > > + fi-elk-e7500: [3]PASS -> [4]DMESG-FAIL > > > > I also believe this is a false positive... > > triggered a retest... > > Yeah, the latest test gave success: > > Subject: [Intel-gfx] ✓ Fi.CI.BAT: success for Fix TLB invalidate issues with Broadwell (rev6) > Date: Mon, 11 Jul 2022 14:53:40 -0000 > Reply-To: intel-gfx@lists.freedesktop.org > Sender: "Intel-gfx" <intel-gfx-bounces@lists.freedesktop.org> > > == Series Details == > > Series: Fix TLB invalidate issues with Broadwell (rev6) > URL : https://patchwork.freedesktop.org/series/105167/ > State : success > > > > > hoping to get this series merged today... > > Yeah, it would be great to have this merged, as it is a regression > fix. we got yet another false positive... I was tired and decided to ignore and move ahead. but then I noticed that the author and sign-off-bys are not right... :( please fix that... the author needs to sign-off-by... > > Regards, > Mauro
i915 selftest hangcheck is causing the i915 driver timeouts, as reported by Intel CI bot: http://gfx-ci.fi.intel.com/cibuglog-ng/issuefilterassoc/24297?query_key=42a999f48fa6ecce068bc8126c069be7c31153b4 When such test runs, the only output is: [ 68.811639] i915: Performing live selftests with st_random_seed=0xe138eac7 st_timeout=500 [ 68.811792] i915: Running hangcheck [ 68.811859] i915: Running intel_hangcheck_live_selftests/igt_hang_sanitycheck [ 68.816910] i915 0000:00:02.0: [drm] Cannot find any crtc or sizes [ 68.841597] i915: Running intel_hangcheck_live_selftests/igt_reset_nop [ 69.346347] igt_reset_nop: 80 resets [ 69.362695] i915: Running intel_hangcheck_live_selftests/igt_reset_nop_engine [ 69.863559] igt_reset_nop_engine(rcs0): 709 resets [ 70.364924] igt_reset_nop_engine(bcs0): 903 resets [ 70.866005] igt_reset_nop_engine(vcs0): 659 resets [ 71.367934] igt_reset_nop_engine(vcs1): 549 resets [ 71.869259] igt_reset_nop_engine(vecs0): 553 resets [ 71.882592] i915: Running intel_hangcheck_live_selftests/igt_reset_idle_engine [ 72.383554] rcs0: Completed 16605 idle resets [ 72.884599] bcs0: Completed 18641 idle resets [ 73.385592] vcs0: Completed 17517 idle resets [ 73.886658] vcs1: Completed 15474 idle resets [ 74.387600] vecs0: Completed 17983 idle resets [ 74.387667] i915: Running intel_hangcheck_live_selftests/igt_reset_active_engine [ 74.889017] rcs0: Completed 747 active resets [ 75.174240] intel_engine_reset(bcs0) failed, err:-110 [ 75.174301] bcs0: Completed 525 active resets After that, the machine just silently hangs. Bisecting the issue, the patch that introduced the regression is: 7938d61591d3 ("drm/i915: Flush TLBs before releasing backing store") Reverting it fix the issues, but introduce other problems, as TLB won't be invalidated anymore. So, instead, let's fix the root cause. It turns that the TLB flush logic ends conflicting with i915 reset, which is called during selftest hangcheck. So, the TLB cache should be serialized together with i915 reset. Tested on an Intel NUC5i7RYB with an i7-5557U Broadwell CPU. v4: - No functional changes. All changes are at the patch descriptions: - collected acked-by/reviewed-by; - use the same e-mail on Author and SoB on patch 1. v3: - Removed the logic that would check if the engine is awake before doing TLB flush invalidation as backporting PM logic up to Kernel 4.x could be too painful. After getting this one merged, I'll submit a separate patch with the PM awake logic. v2: - Reduced to bare minimum fixes, as this shoud be backported deeply into stable. Chris Wilson (2): drm/i915/gt: Serialize GRDOM access between multiple engine resets drm/i915/gt: Serialize TLB invalidates with GT resets drivers/gpu/drm/i915/gt/intel_gt.c | 15 ++++++++++- drivers/gpu/drm/i915/gt/intel_reset.c | 37 ++++++++++++++++++++------- 2 files changed, 42 insertions(+), 10 deletions(-)