| Message ID | 20221026060506.1007830-1-alan.previn.teres.alexis@intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Fix Guc-Err-Capture sizing warning | expand |
AS per the other rev, below regressions are, once again, unrelated to this series because this series patch modified code that only get executed if GuC ADS is being initialized whereas the regressions are on platforms that do not have GuC enabled (which i further verified in the full dmesg).
On Wed, 2022-10-26 at 18:52 +0000, Patchwork wrote:
Patch Details
Series: Fix Guc-Err-Capture sizing warning
URL: https://patchwork.freedesktop.org/series/110155/
State: failure
Details: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/index.html
CI Bug Log - changes from CI_DRM_12295_full -> Patchwork_110155v1_full
Summary
FAILURE
Serious unknown changes coming with Patchwork_110155v1_full absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_110155v1_full, please notify your bug team to allow them
to document this new failure mode, which will reduce false positives in CI.
Participating hosts (9 -> 11)
Additional (2): shard-rkl shard-dg1
Possible new issues
Here are the unknown changes that may have been introduced in Patchwork_110155v1_full:
IGT changes
Possible regressions
* igt@i915_selftest@live@execlists:
* shard-skl: NOTRUN -> INCOMPLETE<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl5/igt@i915_selftest@live@execlists.html>
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render:
* shard-iclb: PASS<https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12295/shard-iclb6/igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render.html> -> FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-iclb1/igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render.html>
* igt@kms_sequence@queue-idle@edp-1-pipe-a:
* shard-skl: NOTRUN -> FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl2/igt@kms_sequence@queue-idle@edp-1-pipe-a.html>
* igt@syncobj_timeline@wait-all-for-submit-snapshot:
* shard-skl: PASS<https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12295/shard-skl2/igt@syncobj_timeline@wait-all-for-submit-snapshot.html> -> FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl1/igt@syncobj_timeline@wait-all-for-submit-snapshot.html>
Warnings
* igt@runner@aborted:
* shard-skl: (FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12295/shard-skl4/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12295/shard-skl10/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12295/shard-skl7/igt@runner@aborted.html>) (i915#3002<https://gitlab.freedesktop.org/drm/intel/issues/3002> / i915#4312<https://gitlab.freedesktop.org/drm/intel/issues/4312> / i915#6949<https://gitlab.freedesktop.org/drm/intel/issues/6949>) -> (FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl9/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl2/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl2/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl3/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl5/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl3/igt@runner@aborted.html>, FAIL<https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110155v1/shard-skl6/igt@runner@aborted.html>) (i915#3002<https://gitlab.freedesktop.org/drm/intel/issues/3002> / i915#4312<https://gitlab.freedesktop.org/drm/intel/issues/4312>)
GuC Error capture initialization calculates an estimation buffer size for worst case scenario of all engines getting reset. Fix the calculation change from drm_warn to drm_dbg since its a corner case Changes from prior revs: v3: - Rebase on latest drm-tip v2: - Reduce the guc-log-buffer error-capture-region allocation from 4MB to 1MB since the corrected math uncovers the larger headroom available to us. - Modify the code comment to focus on highlighting the headroom we have from how min_est is caclulated. - Add example min-est data from real hw in the patch comment. v1: - Change drm_dbg to drm_warn for the case of the mis-estated size not being met (John Harrison). Alan Previn (1): drm/i915/guc: Fix GuC error capture sizing estimation and reporting .../gpu/drm/i915/gt/uc/intel_guc_capture.c | 29 ++++++++++++------- drivers/gpu/drm/i915/gt/uc/intel_guc_log.c | 6 ++-- 2 files changed, 21 insertions(+), 14 deletions(-) base-commit: 3844a56a09225527d7d9148d7e05ef5a99ac282f