| Message ID | ZxqJS8bCWc9ZgIav@fedora (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [PULL] drm-xe-next | expand |
On Thu, Oct 24, 2024 at 07:52:11PM +0200, Thomas Hellstrom wrote: > Hi, Dave & Simona, > > This week's drm-xe-next PR > > Thanks, > Thomas > > > drm-xe-next-2024-10-24: > UAPI Changes: > - Define and parse OA sync properties (Ashutosh) > > Driver Changes: > - Add caller info to xe_gt_reset_async (Nirmoy) > - A large forcewake rework / cleanup (Himal) > - A g2h response timeout fix (Badal) > - A PTL workaround (Vinay) > - Handle unreliable MMIO reads during forcewake (Shuicheng) > - Ufence user-space access fixes (Nirmoy) > - Annotate flexible arrays (Matthew Brost) > - Enable GuC lite restore (Fei) > - Prevent GuC register capture on VF (Zhanjun) > - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) > - Parallel queues fix on GT reset (Nirmoy) > - Move reference grabbing to a job's dma-fence (Matt Brost) > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) This breaks CI [1] - my mistake. Maybe omit these in this weeks PR. We need [2] merged to fix this. Waiting on an RB but I'd like to get all of this in 6.12. Matt [1] https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-140135v2/bat-lnl-1/igt@xe_exec_fault_mode@twice-invalid-fault.html [2] https://patchwork.freedesktop.org/series/140406/ > - OA synchronization support (Ashutosh) > > The following changes since commit 2eb460ab9f4bc5b575f52568d17936da0af681d8: > > drm/xe: Enlarge the invalidation timeout from 150 to 500 (2024-10-16 16:11:10 +0100) > > are available in the Git repository at: > > https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-2024-10-24 > > for you to fetch changes up to 85d3f9e84e0628c412b69aa99b63654dfa08ad68: > > drm/xe/oa: Allow only certain property changes from config (2024-10-23 12:42:20 -0700) > > ---------------------------------------------------------------- > UAPI Changes: > - Define and parse OA sync properties (Ashutosh) > > Driver Changes: > - Add caller info to xe_gt_reset_async (Nirmoy) > - A large forcewake rework / cleanup (Himal) > - A g2h response timeout fix (Badal) > - A PTL workaround (Vinay) > - Handle unreliable MMIO reads during forcewake (Shuicheng) > - Ufence user-space access fixes (Nirmoy) > - Annotate flexible arrays (Matthew Brost) > - Enable GuC lite restore (Fei) > - Prevent GuC register capture on VF (Zhanjun) > - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) > - Parallel queues fix on GT reset (Nirmoy) > - Move reference grabbing to a job's dma-fence (Matt Brost) > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > - OA synchronization support (Ashutosh) > > ---------------------------------------------------------------- > Ashutosh Dixit (7): > drm/xe/oa: Separate batch submission from waiting for completion > drm/xe/oa/uapi: Define and parse OA sync properties > drm/xe/oa: Add input fence dependencies > drm/xe/oa: Signal output fences > drm/xe/oa: Move functions up so they can be reused for config ioctl > drm/xe/oa: Add syncs support to OA config ioctl > drm/xe/oa: Allow only certain property changes from config > > Badal Nilawar (1): > drm/xe/guc/ct: Flush g2h worker in case of g2h response timeout > > Fei Yang (1): > drm/xe: enable lite restore > > Himal Prasad Ghimiray (26): > drm/xe: Add member initialized_domains to xe_force_wake() > drm/xe/forcewake: Change awake_domain datatype > drm/xe/forcewake: Add a helper xe_force_wake_ref_has_domain() > drm/xe: Error handling in xe_force_wake_get() > drm/xe: Modify xe_force_wake_put to handle _get returned mask > drm/xe/device: Update handling of xe_force_wake_get return > drm/xe/hdcp: Update handling of xe_force_wake_get return > drm/xe/gsc: Update handling of xe_force_wake_get return > drm/xe/gt: Update handling of xe_force_wake_get return > drm/xe/xe_gt_idle: Update handling of xe_force_wake_get return > drm/xe/devcoredump: Update handling of xe_force_wake_get return > drm/xe/tests/mocs: Update xe_force_wake_get() return handling > drm/xe/mocs: Update handling of xe_force_wake_get return > drm/xe/xe_drm_client: Update handling of xe_force_wake_get return > drm/xe/xe_gt_debugfs: Update handling of xe_force_wake_get return > drm/xe/guc: Update handling of xe_force_wake_get return > drm/xe/huc: Update handling of xe_force_wake_get return > drm/xe/oa: Handle force_wake_get failure in xe_oa_stream_init() > drm/xe/pat: Update handling of xe_force_wake_get return > drm/xe/gt_tlb_invalidation_ggtt: Update handling of xe_force_wake_get return > drm/xe/xe_reg_sr: Update handling of xe_force_wake_get return > drm/xe/query: Update handling of xe_force_wake_get return > drm/xe/vram: Update handling of xe_force_wake_get return > drm/xe: forcewake debugfs open fails on xe_forcewake_get failure > drm/xe: Ensure __must_check for xe_force_wake_get() return > drm/xe: Change return type to void for xe_force_wake_put > > Matthew Brost (5): > drm/xe: Use __counted_by for flexible arrays > drm/xe: Take ref to job's fence in arm > drm/xe: Mark GGTT work queue with WQ_MEM_RECLAIM > drm/xe: Mark G2H work queue with WQ_MEM_RECLAIM > drm/xe: Mark GT work queue with WQ_MEM_RECLAIM > > Michal Wajdeczko (1): > drm/xe/pf: Show VFs LMEM provisioning summary over debugfs > > Nirmoy Das (4): > drm/xe: Add caller info to xe_gt_reset_async > drm/xe/ufence: Prefetch ufence addr to catch bogus address > drm/xe/ufence: Warn if mmget_not_zero() fails > drm/xe: Don't restart parallel queues multiple times on GT reset > > Shuicheng Lin (1): > drm/xe: Handle unreliable MMIO reads during forcewake > > Vinay Belgaumkar (1): > drm/xe/ptl: Apply Wa_14022866841 > > Zhanjun Dong (1): > drm/xe/guc: Prevent GuC register capture running on VF > > drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + > drivers/gpu/drm/xe/display/xe_hdcp_gsc.c | 6 +- > drivers/gpu/drm/xe/tests/xe_mocs.c | 18 +- > drivers/gpu/drm/xe/xe_debugfs.c | 27 +- > drivers/gpu/drm/xe/xe_devcoredump.c | 14 +- > drivers/gpu/drm/xe/xe_device.c | 25 +- > drivers/gpu/drm/xe/xe_drm_client.c | 8 +- > drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 +- > drivers/gpu/drm/xe/xe_execlist.c | 2 +- > drivers/gpu/drm/xe/xe_force_wake.c | 134 ++++-- > drivers/gpu/drm/xe/xe_force_wake.h | 23 +- > drivers/gpu/drm/xe/xe_force_wake_types.h | 6 +- > drivers/gpu/drm/xe/xe_ggtt.c | 2 +- > drivers/gpu/drm/xe/xe_gsc.c | 23 +- > drivers/gpu/drm/xe/xe_gsc_proxy.c | 9 +- > drivers/gpu/drm/xe/xe_gt.c | 110 +++-- > drivers/gpu/drm/xe/xe_gt_debugfs.c | 13 +- > drivers/gpu/drm/xe/xe_gt_idle.c | 26 +- > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 35 ++ > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + > drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c | 5 + > drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 5 +- > drivers/gpu/drm/xe/xe_guc.c | 15 +- > drivers/gpu/drm/xe/xe_guc_ads.c | 5 + > drivers/gpu/drm/xe/xe_guc_capture.c | 8 +- > drivers/gpu/drm/xe/xe_guc_ct.c | 20 +- > drivers/gpu/drm/xe/xe_guc_fwif.h | 1 + > drivers/gpu/drm/xe/xe_guc_log.c | 9 +- > drivers/gpu/drm/xe/xe_guc_pc.c | 50 +- > drivers/gpu/drm/xe/xe_guc_submit.c | 29 +- > drivers/gpu/drm/xe/xe_huc.c | 8 +- > drivers/gpu/drm/xe/xe_mocs.c | 14 +- > drivers/gpu/drm/xe/xe_oa.c | 678 +++++++++++++++++++--------- > drivers/gpu/drm/xe/xe_oa_types.h | 12 + > drivers/gpu/drm/xe/xe_pat.c | 65 ++- > drivers/gpu/drm/xe/xe_query.c | 10 +- > drivers/gpu/drm/xe/xe_reg_sr.c | 24 +- > drivers/gpu/drm/xe/xe_sched_job.c | 2 +- > drivers/gpu/drm/xe/xe_sched_job_types.h | 3 +- > drivers/gpu/drm/xe/xe_sync.c | 5 +- > drivers/gpu/drm/xe/xe_vram.c | 12 +- > drivers/gpu/drm/xe/xe_wa_oob.rules | 2 + > include/uapi/drm/xe_drm.h | 17 + > 43 files changed, 997 insertions(+), 487 deletions(-)
On Thu, 2024-10-24 at 19:22 +0000, Matthew Brost wrote: > On Thu, Oct 24, 2024 at 07:52:11PM +0200, Thomas Hellstrom wrote: > > Hi, Dave & Simona, > > > > This week's drm-xe-next PR > > > > Thanks, > > Thomas > > > > > > drm-xe-next-2024-10-24: > > UAPI Changes: > > - Define and parse OA sync properties (Ashutosh) > > > > Driver Changes: > > - Add caller info to xe_gt_reset_async (Nirmoy) > > - A large forcewake rework / cleanup (Himal) > > - A g2h response timeout fix (Badal) > > - A PTL workaround (Vinay) > > - Handle unreliable MMIO reads during forcewake (Shuicheng) > > - Ufence user-space access fixes (Nirmoy) > > - Annotate flexible arrays (Matthew Brost) > > - Enable GuC lite restore (Fei) > > - Prevent GuC register capture on VF (Zhanjun) > > - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) > > - Parallel queues fix on GT reset (Nirmoy) > > - Move reference grabbing to a job's dma-fence (Matt Brost) > > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > > This breaks CI [1] - my mistake. Maybe omit these in this weeks PR. > > We need [2] merged to fix this. Waiting on an RB but I'd like to get > all of this in 6.12. > > Matt > > [1] > https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-140135v2/bat-lnl-1/igt@xe_exec_fault_mode@twice-invalid-fault.html > [2] https://patchwork.freedesktop.org/series/140406/ So this CI failure is a warning only and IMHO for drm-xe-next (6.13) it's not catastrophic. There might be a window in the bisect history where this warning appears. It's perhaps more important for -fixes, though. If we need to wait for the scheduler patch going into drm-misc-next / drm-next/ backmerge we'd hold off this branch for too long I fear. @Dave, @Sima If you feel differently please skip this PR for this week and we'll work to get the scheduler patch merged asap. Thanks, Thomas > > > - OA synchronization support (Ashutosh) > > > > The following changes since commit > > 2eb460ab9f4bc5b575f52568d17936da0af681d8: > > > > drm/xe: Enlarge the invalidation timeout from 150 to 500 (2024- > > 10-16 16:11:10 +0100) > > > > are available in the Git repository at: > > > > https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe- > > next-2024-10-24 > > > > for you to fetch changes up to > > 85d3f9e84e0628c412b69aa99b63654dfa08ad68: > > > > drm/xe/oa: Allow only certain property changes from config (2024- > > 10-23 12:42:20 -0700) > > > > ---------------------------------------------------------------- > > UAPI Changes: > > - Define and parse OA sync properties (Ashutosh) > > > > Driver Changes: > > - Add caller info to xe_gt_reset_async (Nirmoy) > > - A large forcewake rework / cleanup (Himal) > > - A g2h response timeout fix (Badal) > > - A PTL workaround (Vinay) > > - Handle unreliable MMIO reads during forcewake (Shuicheng) > > - Ufence user-space access fixes (Nirmoy) > > - Annotate flexible arrays (Matthew Brost) > > - Enable GuC lite restore (Fei) > > - Prevent GuC register capture on VF (Zhanjun) > > - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) > > - Parallel queues fix on GT reset (Nirmoy) > > - Move reference grabbing to a job's dma-fence (Matt Brost) > > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > > - OA synchronization support (Ashutosh) > > > > ---------------------------------------------------------------- > > Ashutosh Dixit (7): > > drm/xe/oa: Separate batch submission from waiting for > > completion > > drm/xe/oa/uapi: Define and parse OA sync properties > > drm/xe/oa: Add input fence dependencies > > drm/xe/oa: Signal output fences > > drm/xe/oa: Move functions up so they can be reused for config > > ioctl > > drm/xe/oa: Add syncs support to OA config ioctl > > drm/xe/oa: Allow only certain property changes from config > > > > Badal Nilawar (1): > > drm/xe/guc/ct: Flush g2h worker in case of g2h response > > timeout > > > > Fei Yang (1): > > drm/xe: enable lite restore > > > > Himal Prasad Ghimiray (26): > > drm/xe: Add member initialized_domains to xe_force_wake() > > drm/xe/forcewake: Change awake_domain datatype > > drm/xe/forcewake: Add a helper xe_force_wake_ref_has_domain() > > drm/xe: Error handling in xe_force_wake_get() > > drm/xe: Modify xe_force_wake_put to handle _get returned mask > > drm/xe/device: Update handling of xe_force_wake_get return > > drm/xe/hdcp: Update handling of xe_force_wake_get return > > drm/xe/gsc: Update handling of xe_force_wake_get return > > drm/xe/gt: Update handling of xe_force_wake_get return > > drm/xe/xe_gt_idle: Update handling of xe_force_wake_get > > return > > drm/xe/devcoredump: Update handling of xe_force_wake_get > > return > > drm/xe/tests/mocs: Update xe_force_wake_get() return handling > > drm/xe/mocs: Update handling of xe_force_wake_get return > > drm/xe/xe_drm_client: Update handling of xe_force_wake_get > > return > > drm/xe/xe_gt_debugfs: Update handling of xe_force_wake_get > > return > > drm/xe/guc: Update handling of xe_force_wake_get return > > drm/xe/huc: Update handling of xe_force_wake_get return > > drm/xe/oa: Handle force_wake_get failure in > > xe_oa_stream_init() > > drm/xe/pat: Update handling of xe_force_wake_get return > > drm/xe/gt_tlb_invalidation_ggtt: Update handling of > > xe_force_wake_get return > > drm/xe/xe_reg_sr: Update handling of xe_force_wake_get return > > drm/xe/query: Update handling of xe_force_wake_get return > > drm/xe/vram: Update handling of xe_force_wake_get return > > drm/xe: forcewake debugfs open fails on xe_forcewake_get > > failure > > drm/xe: Ensure __must_check for xe_force_wake_get() return > > drm/xe: Change return type to void for xe_force_wake_put > > > > Matthew Brost (5): > > drm/xe: Use __counted_by for flexible arrays > > drm/xe: Take ref to job's fence in arm > > drm/xe: Mark GGTT work queue with WQ_MEM_RECLAIM > > drm/xe: Mark G2H work queue with WQ_MEM_RECLAIM > > drm/xe: Mark GT work queue with WQ_MEM_RECLAIM > > > > Michal Wajdeczko (1): > > drm/xe/pf: Show VFs LMEM provisioning summary over debugfs > > > > Nirmoy Das (4): > > drm/xe: Add caller info to xe_gt_reset_async > > drm/xe/ufence: Prefetch ufence addr to catch bogus address > > drm/xe/ufence: Warn if mmget_not_zero() fails > > drm/xe: Don't restart parallel queues multiple times on GT > > reset > > > > Shuicheng Lin (1): > > drm/xe: Handle unreliable MMIO reads during forcewake > > > > Vinay Belgaumkar (1): > > drm/xe/ptl: Apply Wa_14022866841 > > > > Zhanjun Dong (1): > > drm/xe/guc: Prevent GuC register capture running on VF > > > > drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + > > drivers/gpu/drm/xe/display/xe_hdcp_gsc.c | 6 +- > > drivers/gpu/drm/xe/tests/xe_mocs.c | 18 +- > > drivers/gpu/drm/xe/xe_debugfs.c | 27 +- > > drivers/gpu/drm/xe/xe_devcoredump.c | 14 +- > > drivers/gpu/drm/xe/xe_device.c | 25 +- > > drivers/gpu/drm/xe/xe_drm_client.c | 8 +- > > drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 +- > > drivers/gpu/drm/xe/xe_execlist.c | 2 +- > > drivers/gpu/drm/xe/xe_force_wake.c | 134 ++++-- > > drivers/gpu/drm/xe/xe_force_wake.h | 23 +- > > drivers/gpu/drm/xe/xe_force_wake_types.h | 6 +- > > drivers/gpu/drm/xe/xe_ggtt.c | 2 +- > > drivers/gpu/drm/xe/xe_gsc.c | 23 +- > > drivers/gpu/drm/xe/xe_gsc_proxy.c | 9 +- > > drivers/gpu/drm/xe/xe_gt.c | 110 +++-- > > drivers/gpu/drm/xe/xe_gt_debugfs.c | 13 +- > > drivers/gpu/drm/xe/xe_gt_idle.c | 26 +- > > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 35 ++ > > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + > > drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c | 5 + > > drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 5 +- > > drivers/gpu/drm/xe/xe_guc.c | 15 +- > > drivers/gpu/drm/xe/xe_guc_ads.c | 5 + > > drivers/gpu/drm/xe/xe_guc_capture.c | 8 +- > > drivers/gpu/drm/xe/xe_guc_ct.c | 20 +- > > drivers/gpu/drm/xe/xe_guc_fwif.h | 1 + > > drivers/gpu/drm/xe/xe_guc_log.c | 9 +- > > drivers/gpu/drm/xe/xe_guc_pc.c | 50 +- > > drivers/gpu/drm/xe/xe_guc_submit.c | 29 +- > > drivers/gpu/drm/xe/xe_huc.c | 8 +- > > drivers/gpu/drm/xe/xe_mocs.c | 14 +- > > drivers/gpu/drm/xe/xe_oa.c | 678 > > +++++++++++++++++++--------- > > drivers/gpu/drm/xe/xe_oa_types.h | 12 + > > drivers/gpu/drm/xe/xe_pat.c | 65 ++- > > drivers/gpu/drm/xe/xe_query.c | 10 +- > > drivers/gpu/drm/xe/xe_reg_sr.c | 24 +- > > drivers/gpu/drm/xe/xe_sched_job.c | 2 +- > > drivers/gpu/drm/xe/xe_sched_job_types.h | 3 +- > > drivers/gpu/drm/xe/xe_sync.c | 5 +- > > drivers/gpu/drm/xe/xe_vram.c | 12 +- > > drivers/gpu/drm/xe/xe_wa_oob.rules | 2 + > > include/uapi/drm/xe_drm.h | 17 + > > 43 files changed, 997 insertions(+), 487 deletions(-)
On Fri, 25 Oct 2024, Thomas Hellström <thomas.hellstrom@linux.intel.com> wrote: > On Thu, 2024-10-24 at 19:22 +0000, Matthew Brost wrote: >> On Thu, Oct 24, 2024 at 07:52:11PM +0200, Thomas Hellstrom wrote: >> > Hi, Dave & Simona, >> > >> > This week's drm-xe-next PR >> > >> > Thanks, >> > Thomas >> > >> > >> > drm-xe-next-2024-10-24: >> > UAPI Changes: >> > - Define and parse OA sync properties (Ashutosh) >> > >> > Driver Changes: >> > - Add caller info to xe_gt_reset_async (Nirmoy) >> > - A large forcewake rework / cleanup (Himal) >> > - A g2h response timeout fix (Badal) >> > - A PTL workaround (Vinay) >> > - Handle unreliable MMIO reads during forcewake (Shuicheng) >> > - Ufence user-space access fixes (Nirmoy) >> > - Annotate flexible arrays (Matthew Brost) >> > - Enable GuC lite restore (Fei) >> > - Prevent GuC register capture on VF (Zhanjun) >> > - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) >> > - Parallel queues fix on GT reset (Nirmoy) >> > - Move reference grabbing to a job's dma-fence (Matt Brost) >> > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) >> >> This breaks CI [1] - my mistake. Maybe omit these in this weeks PR. How did this pass CI and get merged in the first place?!? It's now botching unrelated pre-merge testing all over the place, e.g. [3] and [4]. BR, Jani. [3] https://lore.kernel.org/r/172981565466.1330037.6238046952250769671@2413ebb6fbb6 [4] https://lore.kernel.org/r/172981849964.1330038.16133455483045565936@2413ebb6fbb6 >> >> We need [2] merged to fix this. Waiting on an RB but I'd like to get >> all of this in 6.12. >> >> Matt >> >> [1] >> https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-140135v2/bat-lnl-1/igt@xe_exec_fault_mode@twice-invalid-fault.html >> [2] https://patchwork.freedesktop.org/series/140406/ > > So this CI failure is a warning only and IMHO for drm-xe-next (6.13) > it's not catastrophic. There might be a window in the bisect history > where this warning appears. It's perhaps more important for -fixes, > though. > > If we need to wait for the scheduler patch going into drm-misc-next / > drm-next/ backmerge we'd hold off this branch for too long I fear. > > @Dave, @Sima > If you feel differently please skip this PR for this week and we'll > work to get the scheduler patch merged asap. > > Thanks, > Thomas > > >> >> > - OA synchronization support (Ashutosh) >> > >> > The following changes since commit >> > 2eb460ab9f4bc5b575f52568d17936da0af681d8: >> > >> > drm/xe: Enlarge the invalidation timeout from 150 to 500 (2024- >> > 10-16 16:11:10 +0100) >> > >> > are available in the Git repository at: >> > >> > https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe- >> > next-2024-10-24 >> > >> > for you to fetch changes up to >> > 85d3f9e84e0628c412b69aa99b63654dfa08ad68: >> > >> > drm/xe/oa: Allow only certain property changes from config (2024- >> > 10-23 12:42:20 -0700) >> > >> > ---------------------------------------------------------------- >> > UAPI Changes: >> > - Define and parse OA sync properties (Ashutosh) >> > >> > Driver Changes: >> > - Add caller info to xe_gt_reset_async (Nirmoy) >> > - A large forcewake rework / cleanup (Himal) >> > - A g2h response timeout fix (Badal) >> > - A PTL workaround (Vinay) >> > - Handle unreliable MMIO reads during forcewake (Shuicheng) >> > - Ufence user-space access fixes (Nirmoy) >> > - Annotate flexible arrays (Matthew Brost) >> > - Enable GuC lite restore (Fei) >> > - Prevent GuC register capture on VF (Zhanjun) >> > - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) >> > - Parallel queues fix on GT reset (Nirmoy) >> > - Move reference grabbing to a job's dma-fence (Matt Brost) >> > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) >> > - OA synchronization support (Ashutosh) >> > >> > ---------------------------------------------------------------- >> > Ashutosh Dixit (7): >> > drm/xe/oa: Separate batch submission from waiting for >> > completion >> > drm/xe/oa/uapi: Define and parse OA sync properties >> > drm/xe/oa: Add input fence dependencies >> > drm/xe/oa: Signal output fences >> > drm/xe/oa: Move functions up so they can be reused for config >> > ioctl >> > drm/xe/oa: Add syncs support to OA config ioctl >> > drm/xe/oa: Allow only certain property changes from config >> > >> > Badal Nilawar (1): >> > drm/xe/guc/ct: Flush g2h worker in case of g2h response >> > timeout >> > >> > Fei Yang (1): >> > drm/xe: enable lite restore >> > >> > Himal Prasad Ghimiray (26): >> > drm/xe: Add member initialized_domains to xe_force_wake() >> > drm/xe/forcewake: Change awake_domain datatype >> > drm/xe/forcewake: Add a helper xe_force_wake_ref_has_domain() >> > drm/xe: Error handling in xe_force_wake_get() >> > drm/xe: Modify xe_force_wake_put to handle _get returned mask >> > drm/xe/device: Update handling of xe_force_wake_get return >> > drm/xe/hdcp: Update handling of xe_force_wake_get return >> > drm/xe/gsc: Update handling of xe_force_wake_get return >> > drm/xe/gt: Update handling of xe_force_wake_get return >> > drm/xe/xe_gt_idle: Update handling of xe_force_wake_get >> > return >> > drm/xe/devcoredump: Update handling of xe_force_wake_get >> > return >> > drm/xe/tests/mocs: Update xe_force_wake_get() return handling >> > drm/xe/mocs: Update handling of xe_force_wake_get return >> > drm/xe/xe_drm_client: Update handling of xe_force_wake_get >> > return >> > drm/xe/xe_gt_debugfs: Update handling of xe_force_wake_get >> > return >> > drm/xe/guc: Update handling of xe_force_wake_get return >> > drm/xe/huc: Update handling of xe_force_wake_get return >> > drm/xe/oa: Handle force_wake_get failure in >> > xe_oa_stream_init() >> > drm/xe/pat: Update handling of xe_force_wake_get return >> > drm/xe/gt_tlb_invalidation_ggtt: Update handling of >> > xe_force_wake_get return >> > drm/xe/xe_reg_sr: Update handling of xe_force_wake_get return >> > drm/xe/query: Update handling of xe_force_wake_get return >> > drm/xe/vram: Update handling of xe_force_wake_get return >> > drm/xe: forcewake debugfs open fails on xe_forcewake_get >> > failure >> > drm/xe: Ensure __must_check for xe_force_wake_get() return >> > drm/xe: Change return type to void for xe_force_wake_put >> > >> > Matthew Brost (5): >> > drm/xe: Use __counted_by for flexible arrays >> > drm/xe: Take ref to job's fence in arm >> > drm/xe: Mark GGTT work queue with WQ_MEM_RECLAIM >> > drm/xe: Mark G2H work queue with WQ_MEM_RECLAIM >> > drm/xe: Mark GT work queue with WQ_MEM_RECLAIM >> > >> > Michal Wajdeczko (1): >> > drm/xe/pf: Show VFs LMEM provisioning summary over debugfs >> > >> > Nirmoy Das (4): >> > drm/xe: Add caller info to xe_gt_reset_async >> > drm/xe/ufence: Prefetch ufence addr to catch bogus address >> > drm/xe/ufence: Warn if mmget_not_zero() fails >> > drm/xe: Don't restart parallel queues multiple times on GT >> > reset >> > >> > Shuicheng Lin (1): >> > drm/xe: Handle unreliable MMIO reads during forcewake >> > >> > Vinay Belgaumkar (1): >> > drm/xe/ptl: Apply Wa_14022866841 >> > >> > Zhanjun Dong (1): >> > drm/xe/guc: Prevent GuC register capture running on VF >> > >> > drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + >> > drivers/gpu/drm/xe/display/xe_hdcp_gsc.c | 6 +- >> > drivers/gpu/drm/xe/tests/xe_mocs.c | 18 +- >> > drivers/gpu/drm/xe/xe_debugfs.c | 27 +- >> > drivers/gpu/drm/xe/xe_devcoredump.c | 14 +- >> > drivers/gpu/drm/xe/xe_device.c | 25 +- >> > drivers/gpu/drm/xe/xe_drm_client.c | 8 +- >> > drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 +- >> > drivers/gpu/drm/xe/xe_execlist.c | 2 +- >> > drivers/gpu/drm/xe/xe_force_wake.c | 134 ++++-- >> > drivers/gpu/drm/xe/xe_force_wake.h | 23 +- >> > drivers/gpu/drm/xe/xe_force_wake_types.h | 6 +- >> > drivers/gpu/drm/xe/xe_ggtt.c | 2 +- >> > drivers/gpu/drm/xe/xe_gsc.c | 23 +- >> > drivers/gpu/drm/xe/xe_gsc_proxy.c | 9 +- >> > drivers/gpu/drm/xe/xe_gt.c | 110 +++-- >> > drivers/gpu/drm/xe/xe_gt_debugfs.c | 13 +- >> > drivers/gpu/drm/xe/xe_gt_idle.c | 26 +- >> > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 35 ++ >> > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + >> > drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c | 5 + >> > drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 5 +- >> > drivers/gpu/drm/xe/xe_guc.c | 15 +- >> > drivers/gpu/drm/xe/xe_guc_ads.c | 5 + >> > drivers/gpu/drm/xe/xe_guc_capture.c | 8 +- >> > drivers/gpu/drm/xe/xe_guc_ct.c | 20 +- >> > drivers/gpu/drm/xe/xe_guc_fwif.h | 1 + >> > drivers/gpu/drm/xe/xe_guc_log.c | 9 +- >> > drivers/gpu/drm/xe/xe_guc_pc.c | 50 +- >> > drivers/gpu/drm/xe/xe_guc_submit.c | 29 +- >> > drivers/gpu/drm/xe/xe_huc.c | 8 +- >> > drivers/gpu/drm/xe/xe_mocs.c | 14 +- >> > drivers/gpu/drm/xe/xe_oa.c | 678 >> > +++++++++++++++++++--------- >> > drivers/gpu/drm/xe/xe_oa_types.h | 12 + >> > drivers/gpu/drm/xe/xe_pat.c | 65 ++- >> > drivers/gpu/drm/xe/xe_query.c | 10 +- >> > drivers/gpu/drm/xe/xe_reg_sr.c | 24 +- >> > drivers/gpu/drm/xe/xe_sched_job.c | 2 +- >> > drivers/gpu/drm/xe/xe_sched_job_types.h | 3 +- >> > drivers/gpu/drm/xe/xe_sync.c | 5 +- >> > drivers/gpu/drm/xe/xe_vram.c | 12 +- >> > drivers/gpu/drm/xe/xe_wa_oob.rules | 2 + >> > include/uapi/drm/xe_drm.h | 17 + >> > 43 files changed, 997 insertions(+), 487 deletions(-) >
On Fri, 2024-10-25 at 12:34 +0300, Jani Nikula wrote: > On Fri, 25 Oct 2024, Thomas Hellström > <thomas.hellstrom@linux.intel.com> wrote: > > On Thu, 2024-10-24 at 19:22 +0000, Matthew Brost wrote: > > > On Thu, Oct 24, 2024 at 07:52:11PM +0200, Thomas Hellstrom wrote: > > > > Hi, Dave & Simona, > > > > > > > > This week's drm-xe-next PR > > > > > > > > Thanks, > > > > Thomas > > > > > > > > > > > > drm-xe-next-2024-10-24: > > > > UAPI Changes: > > > > - Define and parse OA sync properties (Ashutosh) > > > > > > > > Driver Changes: > > > > - Add caller info to xe_gt_reset_async (Nirmoy) > > > > - A large forcewake rework / cleanup (Himal) > > > > - A g2h response timeout fix (Badal) > > > > - A PTL workaround (Vinay) > > > > - Handle unreliable MMIO reads during forcewake (Shuicheng) > > > > - Ufence user-space access fixes (Nirmoy) > > > > - Annotate flexible arrays (Matthew Brost) > > > > - Enable GuC lite restore (Fei) > > > > - Prevent GuC register capture on VF (Zhanjun) > > > > - Show VFs VRAM / LMEM provisioning summary over debugfs > > > > (Michal) > > > > - Parallel queues fix on GT reset (Nirmoy) > > > > - Move reference grabbing to a job's dma-fence (Matt Brost) > > > > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > > > > > > This breaks CI [1] - my mistake. Maybe omit these in this weeks > > > PR. > > How did this pass CI and get merged in the first place?!? > > It's now botching unrelated pre-merge testing all over the place, > e.g. [3] and [4]. > > BR, > Jani. This appears to have been a partial merge of a passing series.... /Thomas > > > [3] > https://lore.kernel.org/r/172981565466.1330037.6238046952250769671@2413ebb6fbb6 > [4] > https://lore.kernel.org/r/172981849964.1330038.16133455483045565936@2413ebb6fbb6 > > > > > > > > We need [2] merged to fix this. Waiting on an RB but I'd like to > > > get > > > all of this in 6.12. > > > > > > Matt > > > > > > [1] > > > https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-140135v2/bat-lnl-1/igt@xe_exec_fault_mode@twice-invalid-fault.html > > > [2] https://patchwork.freedesktop.org/series/140406/ > > > > So this CI failure is a warning only and IMHO for drm-xe-next > > (6.13) > > it's not catastrophic. There might be a window in the bisect > > history > > where this warning appears. It's perhaps more important for -fixes, > > though. > > > > If we need to wait for the scheduler patch going into drm-misc-next > > / > > drm-next/ backmerge we'd hold off this branch for too long I fear. > > > > @Dave, @Sima > > If you feel differently please skip this PR for this week and we'll > > work to get the scheduler patch merged asap. > > > > Thanks, > > Thomas > > > > > > > > > > > - OA synchronization support (Ashutosh) > > > > > > > > The following changes since commit > > > > 2eb460ab9f4bc5b575f52568d17936da0af681d8: > > > > > > > > drm/xe: Enlarge the invalidation timeout from 150 to 500 > > > > (2024- > > > > 10-16 16:11:10 +0100) > > > > > > > > are available in the Git repository at: > > > > > > > > https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe- > > > > next-2024-10-24 > > > > > > > > for you to fetch changes up to > > > > 85d3f9e84e0628c412b69aa99b63654dfa08ad68: > > > > > > > > drm/xe/oa: Allow only certain property changes from config > > > > (2024- > > > > 10-23 12:42:20 -0700) > > > > > > > > --------------------------------------------------------------- > > > > - > > > > UAPI Changes: > > > > - Define and parse OA sync properties (Ashutosh) > > > > > > > > Driver Changes: > > > > - Add caller info to xe_gt_reset_async (Nirmoy) > > > > - A large forcewake rework / cleanup (Himal) > > > > - A g2h response timeout fix (Badal) > > > > - A PTL workaround (Vinay) > > > > - Handle unreliable MMIO reads during forcewake (Shuicheng) > > > > - Ufence user-space access fixes (Nirmoy) > > > > - Annotate flexible arrays (Matthew Brost) > > > > - Enable GuC lite restore (Fei) > > > > - Prevent GuC register capture on VF (Zhanjun) > > > > - Show VFs VRAM / LMEM provisioning summary over debugfs > > > > (Michal) > > > > - Parallel queues fix on GT reset (Nirmoy) > > > > - Move reference grabbing to a job's dma-fence (Matt Brost) > > > > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > > > > - OA synchronization support (Ashutosh) > > > > > > > > --------------------------------------------------------------- > > > > - > > > > Ashutosh Dixit (7): > > > > drm/xe/oa: Separate batch submission from waiting for > > > > completion > > > > drm/xe/oa/uapi: Define and parse OA sync properties > > > > drm/xe/oa: Add input fence dependencies > > > > drm/xe/oa: Signal output fences > > > > drm/xe/oa: Move functions up so they can be reused for > > > > config > > > > ioctl > > > > drm/xe/oa: Add syncs support to OA config ioctl > > > > drm/xe/oa: Allow only certain property changes from > > > > config > > > > > > > > Badal Nilawar (1): > > > > drm/xe/guc/ct: Flush g2h worker in case of g2h response > > > > timeout > > > > > > > > Fei Yang (1): > > > > drm/xe: enable lite restore > > > > > > > > Himal Prasad Ghimiray (26): > > > > drm/xe: Add member initialized_domains to xe_force_wake() > > > > drm/xe/forcewake: Change awake_domain datatype > > > > drm/xe/forcewake: Add a helper > > > > xe_force_wake_ref_has_domain() > > > > drm/xe: Error handling in xe_force_wake_get() > > > > drm/xe: Modify xe_force_wake_put to handle _get returned > > > > mask > > > > drm/xe/device: Update handling of xe_force_wake_get > > > > return > > > > drm/xe/hdcp: Update handling of xe_force_wake_get return > > > > drm/xe/gsc: Update handling of xe_force_wake_get return > > > > drm/xe/gt: Update handling of xe_force_wake_get return > > > > drm/xe/xe_gt_idle: Update handling of xe_force_wake_get > > > > return > > > > drm/xe/devcoredump: Update handling of xe_force_wake_get > > > > return > > > > drm/xe/tests/mocs: Update xe_force_wake_get() return > > > > handling > > > > drm/xe/mocs: Update handling of xe_force_wake_get return > > > > drm/xe/xe_drm_client: Update handling of > > > > xe_force_wake_get > > > > return > > > > drm/xe/xe_gt_debugfs: Update handling of > > > > xe_force_wake_get > > > > return > > > > drm/xe/guc: Update handling of xe_force_wake_get return > > > > drm/xe/huc: Update handling of xe_force_wake_get return > > > > drm/xe/oa: Handle force_wake_get failure in > > > > xe_oa_stream_init() > > > > drm/xe/pat: Update handling of xe_force_wake_get return > > > > drm/xe/gt_tlb_invalidation_ggtt: Update handling of > > > > xe_force_wake_get return > > > > drm/xe/xe_reg_sr: Update handling of xe_force_wake_get > > > > return > > > > drm/xe/query: Update handling of xe_force_wake_get return > > > > drm/xe/vram: Update handling of xe_force_wake_get return > > > > drm/xe: forcewake debugfs open fails on xe_forcewake_get > > > > failure > > > > drm/xe: Ensure __must_check for xe_force_wake_get() > > > > return > > > > drm/xe: Change return type to void for xe_force_wake_put > > > > > > > > Matthew Brost (5): > > > > drm/xe: Use __counted_by for flexible arrays > > > > drm/xe: Take ref to job's fence in arm > > > > drm/xe: Mark GGTT work queue with WQ_MEM_RECLAIM > > > > drm/xe: Mark G2H work queue with WQ_MEM_RECLAIM > > > > drm/xe: Mark GT work queue with WQ_MEM_RECLAIM > > > > > > > > Michal Wajdeczko (1): > > > > drm/xe/pf: Show VFs LMEM provisioning summary over > > > > debugfs > > > > > > > > Nirmoy Das (4): > > > > drm/xe: Add caller info to xe_gt_reset_async > > > > drm/xe/ufence: Prefetch ufence addr to catch bogus > > > > address > > > > drm/xe/ufence: Warn if mmget_not_zero() fails > > > > drm/xe: Don't restart parallel queues multiple times on > > > > GT > > > > reset > > > > > > > > Shuicheng Lin (1): > > > > drm/xe: Handle unreliable MMIO reads during forcewake > > > > > > > > Vinay Belgaumkar (1): > > > > drm/xe/ptl: Apply Wa_14022866841 > > > > > > > > Zhanjun Dong (1): > > > > drm/xe/guc: Prevent GuC register capture running on VF > > > > > > > > drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + > > > > drivers/gpu/drm/xe/display/xe_hdcp_gsc.c | 6 +- > > > > drivers/gpu/drm/xe/tests/xe_mocs.c | 18 +- > > > > drivers/gpu/drm/xe/xe_debugfs.c | 27 +- > > > > drivers/gpu/drm/xe/xe_devcoredump.c | 14 +- > > > > drivers/gpu/drm/xe/xe_device.c | 25 +- > > > > drivers/gpu/drm/xe/xe_drm_client.c | 8 +- > > > > drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 +- > > > > drivers/gpu/drm/xe/xe_execlist.c | 2 +- > > > > drivers/gpu/drm/xe/xe_force_wake.c | 134 ++++-- > > > > drivers/gpu/drm/xe/xe_force_wake.h | 23 +- > > > > drivers/gpu/drm/xe/xe_force_wake_types.h | 6 +- > > > > drivers/gpu/drm/xe/xe_ggtt.c | 2 +- > > > > drivers/gpu/drm/xe/xe_gsc.c | 23 +- > > > > drivers/gpu/drm/xe/xe_gsc_proxy.c | 9 +- > > > > drivers/gpu/drm/xe/xe_gt.c | 110 +++-- > > > > drivers/gpu/drm/xe/xe_gt_debugfs.c | 13 +- > > > > drivers/gpu/drm/xe/xe_gt_idle.c | 26 +- > > > > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 35 ++ > > > > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + > > > > drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c | 5 + > > > > drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 5 +- > > > > drivers/gpu/drm/xe/xe_guc.c | 15 +- > > > > drivers/gpu/drm/xe/xe_guc_ads.c | 5 + > > > > drivers/gpu/drm/xe/xe_guc_capture.c | 8 +- > > > > drivers/gpu/drm/xe/xe_guc_ct.c | 20 +- > > > > drivers/gpu/drm/xe/xe_guc_fwif.h | 1 + > > > > drivers/gpu/drm/xe/xe_guc_log.c | 9 +- > > > > drivers/gpu/drm/xe/xe_guc_pc.c | 50 +- > > > > drivers/gpu/drm/xe/xe_guc_submit.c | 29 +- > > > > drivers/gpu/drm/xe/xe_huc.c | 8 +- > > > > drivers/gpu/drm/xe/xe_mocs.c | 14 +- > > > > drivers/gpu/drm/xe/xe_oa.c | 678 > > > > +++++++++++++++++++--------- > > > > drivers/gpu/drm/xe/xe_oa_types.h | 12 + > > > > drivers/gpu/drm/xe/xe_pat.c | 65 ++- > > > > drivers/gpu/drm/xe/xe_query.c | 10 +- > > > > drivers/gpu/drm/xe/xe_reg_sr.c | 24 +- > > > > drivers/gpu/drm/xe/xe_sched_job.c | 2 +- > > > > drivers/gpu/drm/xe/xe_sched_job_types.h | 3 +- > > > > drivers/gpu/drm/xe/xe_sync.c | 5 +- > > > > drivers/gpu/drm/xe/xe_vram.c | 12 +- > > > > drivers/gpu/drm/xe/xe_wa_oob.rules | 2 + > > > > include/uapi/drm/xe_drm.h | 17 + > > > > 43 files changed, 997 insertions(+), 487 deletions(-) > > >
On Fri, Oct 25, 2024 at 12:45:26PM +0200, Thomas Hellström wrote: > On Fri, 2024-10-25 at 12:34 +0300, Jani Nikula wrote: > > On Fri, 25 Oct 2024, Thomas Hellström > > <thomas.hellstrom@linux.intel.com> wrote: > > > On Thu, 2024-10-24 at 19:22 +0000, Matthew Brost wrote: > > > > On Thu, Oct 24, 2024 at 07:52:11PM +0200, Thomas Hellstrom wrote: > > > > > Hi, Dave & Simona, > > > > > > > > > > This week's drm-xe-next PR > > > > > > > > > > Thanks, > > > > > Thomas > > > > > > > > > > > > > > > drm-xe-next-2024-10-24: > > > > > UAPI Changes: > > > > > - Define and parse OA sync properties (Ashutosh) > > > > > > > > > > Driver Changes: > > > > > - Add caller info to xe_gt_reset_async (Nirmoy) > > > > > - A large forcewake rework / cleanup (Himal) > > > > > - A g2h response timeout fix (Badal) > > > > > - A PTL workaround (Vinay) > > > > > - Handle unreliable MMIO reads during forcewake (Shuicheng) > > > > > - Ufence user-space access fixes (Nirmoy) > > > > > - Annotate flexible arrays (Matthew Brost) > > > > > - Enable GuC lite restore (Fei) > > > > > - Prevent GuC register capture on VF (Zhanjun) > > > > > - Show VFs VRAM / LMEM provisioning summary over debugfs > > > > > (Michal) > > > > > - Parallel queues fix on GT reset (Nirmoy) > > > > > - Move reference grabbing to a job's dma-fence (Matt Brost) > > > > > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > > > > > > > > This breaks CI [1] - my mistake. Maybe omit these in this weeks > > > > PR. > > > > How did this pass CI and get merged in the first place?!? > > > > It's now botching unrelated pre-merge testing all over the place, > > e.g. [3] and [4]. > > > > BR, > > Jani. > > This appears to have been a partial merge of a passing series.... > /Thomas > Yea again my mistake on the partial merge - will be more careful going forward. Have RBs on the scheduler patch which will fix our CI but getting conflicts on drm-misc-next so need some maintainer help there. Friday so won't get this fixed up until Monday. Matt > > > > > > > [3] > > https://lore.kernel.org/r/172981565466.1330037.6238046952250769671@2413ebb6fbb6 > > [4] > > https://lore.kernel.org/r/172981849964.1330038.16133455483045565936@2413ebb6fbb6 > > > > > > > > > > > > We need [2] merged to fix this. Waiting on an RB but I'd like to > > > > get > > > > all of this in 6.12. > > > > > > > > Matt > > > > > > > > [1] > > > > https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-140135v2/bat-lnl-1/igt@xe_exec_fault_mode@twice-invalid-fault.html > > > > [2] https://patchwork.freedesktop.org/series/140406/ > > > > > > So this CI failure is a warning only and IMHO for drm-xe-next > > > (6.13) > > > it's not catastrophic. There might be a window in the bisect > > > history > > > where this warning appears. It's perhaps more important for -fixes, > > > though. > > > > > > If we need to wait for the scheduler patch going into drm-misc-next > > > / > > > drm-next/ backmerge we'd hold off this branch for too long I fear. > > > > > > @Dave, @Sima > > > If you feel differently please skip this PR for this week and we'll > > > work to get the scheduler patch merged asap. > > > > > > Thanks, > > > Thomas > > > > > > > > > > > > > > > - OA synchronization support (Ashutosh) > > > > > > > > > > The following changes since commit > > > > > 2eb460ab9f4bc5b575f52568d17936da0af681d8: > > > > > > > > > > drm/xe: Enlarge the invalidation timeout from 150 to 500 > > > > > (2024- > > > > > 10-16 16:11:10 +0100) > > > > > > > > > > are available in the Git repository at: > > > > > > > > > > https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe- > > > > > next-2024-10-24 > > > > > > > > > > for you to fetch changes up to > > > > > 85d3f9e84e0628c412b69aa99b63654dfa08ad68: > > > > > > > > > > drm/xe/oa: Allow only certain property changes from config > > > > > (2024- > > > > > 10-23 12:42:20 -0700) > > > > > > > > > > --------------------------------------------------------------- > > > > > - > > > > > UAPI Changes: > > > > > - Define and parse OA sync properties (Ashutosh) > > > > > > > > > > Driver Changes: > > > > > - Add caller info to xe_gt_reset_async (Nirmoy) > > > > > - A large forcewake rework / cleanup (Himal) > > > > > - A g2h response timeout fix (Badal) > > > > > - A PTL workaround (Vinay) > > > > > - Handle unreliable MMIO reads during forcewake (Shuicheng) > > > > > - Ufence user-space access fixes (Nirmoy) > > > > > - Annotate flexible arrays (Matthew Brost) > > > > > - Enable GuC lite restore (Fei) > > > > > - Prevent GuC register capture on VF (Zhanjun) > > > > > - Show VFs VRAM / LMEM provisioning summary over debugfs > > > > > (Michal) > > > > > - Parallel queues fix on GT reset (Nirmoy) > > > > > - Move reference grabbing to a job's dma-fence (Matt Brost) > > > > > - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) > > > > > - OA synchronization support (Ashutosh) > > > > > > > > > > --------------------------------------------------------------- > > > > > - > > > > > Ashutosh Dixit (7): > > > > > drm/xe/oa: Separate batch submission from waiting for > > > > > completion > > > > > drm/xe/oa/uapi: Define and parse OA sync properties > > > > > drm/xe/oa: Add input fence dependencies > > > > > drm/xe/oa: Signal output fences > > > > > drm/xe/oa: Move functions up so they can be reused for > > > > > config > > > > > ioctl > > > > > drm/xe/oa: Add syncs support to OA config ioctl > > > > > drm/xe/oa: Allow only certain property changes from > > > > > config > > > > > > > > > > Badal Nilawar (1): > > > > > drm/xe/guc/ct: Flush g2h worker in case of g2h response > > > > > timeout > > > > > > > > > > Fei Yang (1): > > > > > drm/xe: enable lite restore > > > > > > > > > > Himal Prasad Ghimiray (26): > > > > > drm/xe: Add member initialized_domains to xe_force_wake() > > > > > drm/xe/forcewake: Change awake_domain datatype > > > > > drm/xe/forcewake: Add a helper > > > > > xe_force_wake_ref_has_domain() > > > > > drm/xe: Error handling in xe_force_wake_get() > > > > > drm/xe: Modify xe_force_wake_put to handle _get returned > > > > > mask > > > > > drm/xe/device: Update handling of xe_force_wake_get > > > > > return > > > > > drm/xe/hdcp: Update handling of xe_force_wake_get return > > > > > drm/xe/gsc: Update handling of xe_force_wake_get return > > > > > drm/xe/gt: Update handling of xe_force_wake_get return > > > > > drm/xe/xe_gt_idle: Update handling of xe_force_wake_get > > > > > return > > > > > drm/xe/devcoredump: Update handling of xe_force_wake_get > > > > > return > > > > > drm/xe/tests/mocs: Update xe_force_wake_get() return > > > > > handling > > > > > drm/xe/mocs: Update handling of xe_force_wake_get return > > > > > drm/xe/xe_drm_client: Update handling of > > > > > xe_force_wake_get > > > > > return > > > > > drm/xe/xe_gt_debugfs: Update handling of > > > > > xe_force_wake_get > > > > > return > > > > > drm/xe/guc: Update handling of xe_force_wake_get return > > > > > drm/xe/huc: Update handling of xe_force_wake_get return > > > > > drm/xe/oa: Handle force_wake_get failure in > > > > > xe_oa_stream_init() > > > > > drm/xe/pat: Update handling of xe_force_wake_get return > > > > > drm/xe/gt_tlb_invalidation_ggtt: Update handling of > > > > > xe_force_wake_get return > > > > > drm/xe/xe_reg_sr: Update handling of xe_force_wake_get > > > > > return > > > > > drm/xe/query: Update handling of xe_force_wake_get return > > > > > drm/xe/vram: Update handling of xe_force_wake_get return > > > > > drm/xe: forcewake debugfs open fails on xe_forcewake_get > > > > > failure > > > > > drm/xe: Ensure __must_check for xe_force_wake_get() > > > > > return > > > > > drm/xe: Change return type to void for xe_force_wake_put > > > > > > > > > > Matthew Brost (5): > > > > > drm/xe: Use __counted_by for flexible arrays > > > > > drm/xe: Take ref to job's fence in arm > > > > > drm/xe: Mark GGTT work queue with WQ_MEM_RECLAIM > > > > > drm/xe: Mark G2H work queue with WQ_MEM_RECLAIM > > > > > drm/xe: Mark GT work queue with WQ_MEM_RECLAIM > > > > > > > > > > Michal Wajdeczko (1): > > > > > drm/xe/pf: Show VFs LMEM provisioning summary over > > > > > debugfs > > > > > > > > > > Nirmoy Das (4): > > > > > drm/xe: Add caller info to xe_gt_reset_async > > > > > drm/xe/ufence: Prefetch ufence addr to catch bogus > > > > > address > > > > > drm/xe/ufence: Warn if mmget_not_zero() fails > > > > > drm/xe: Don't restart parallel queues multiple times on > > > > > GT > > > > > reset > > > > > > > > > > Shuicheng Lin (1): > > > > > drm/xe: Handle unreliable MMIO reads during forcewake > > > > > > > > > > Vinay Belgaumkar (1): > > > > > drm/xe/ptl: Apply Wa_14022866841 > > > > > > > > > > Zhanjun Dong (1): > > > > > drm/xe/guc: Prevent GuC register capture running on VF > > > > > > > > > > drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + > > > > > drivers/gpu/drm/xe/display/xe_hdcp_gsc.c | 6 +- > > > > > drivers/gpu/drm/xe/tests/xe_mocs.c | 18 +- > > > > > drivers/gpu/drm/xe/xe_debugfs.c | 27 +- > > > > > drivers/gpu/drm/xe/xe_devcoredump.c | 14 +- > > > > > drivers/gpu/drm/xe/xe_device.c | 25 +- > > > > > drivers/gpu/drm/xe/xe_drm_client.c | 8 +- > > > > > drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 +- > > > > > drivers/gpu/drm/xe/xe_execlist.c | 2 +- > > > > > drivers/gpu/drm/xe/xe_force_wake.c | 134 ++++-- > > > > > drivers/gpu/drm/xe/xe_force_wake.h | 23 +- > > > > > drivers/gpu/drm/xe/xe_force_wake_types.h | 6 +- > > > > > drivers/gpu/drm/xe/xe_ggtt.c | 2 +- > > > > > drivers/gpu/drm/xe/xe_gsc.c | 23 +- > > > > > drivers/gpu/drm/xe/xe_gsc_proxy.c | 9 +- > > > > > drivers/gpu/drm/xe/xe_gt.c | 110 +++-- > > > > > drivers/gpu/drm/xe/xe_gt_debugfs.c | 13 +- > > > > > drivers/gpu/drm/xe/xe_gt_idle.c | 26 +- > > > > > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 35 ++ > > > > > drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + > > > > > drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c | 5 + > > > > > drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 5 +- > > > > > drivers/gpu/drm/xe/xe_guc.c | 15 +- > > > > > drivers/gpu/drm/xe/xe_guc_ads.c | 5 + > > > > > drivers/gpu/drm/xe/xe_guc_capture.c | 8 +- > > > > > drivers/gpu/drm/xe/xe_guc_ct.c | 20 +- > > > > > drivers/gpu/drm/xe/xe_guc_fwif.h | 1 + > > > > > drivers/gpu/drm/xe/xe_guc_log.c | 9 +- > > > > > drivers/gpu/drm/xe/xe_guc_pc.c | 50 +- > > > > > drivers/gpu/drm/xe/xe_guc_submit.c | 29 +- > > > > > drivers/gpu/drm/xe/xe_huc.c | 8 +- > > > > > drivers/gpu/drm/xe/xe_mocs.c | 14 +- > > > > > drivers/gpu/drm/xe/xe_oa.c | 678 > > > > > +++++++++++++++++++--------- > > > > > drivers/gpu/drm/xe/xe_oa_types.h | 12 + > > > > > drivers/gpu/drm/xe/xe_pat.c | 65 ++- > > > > > drivers/gpu/drm/xe/xe_query.c | 10 +- > > > > > drivers/gpu/drm/xe/xe_reg_sr.c | 24 +- > > > > > drivers/gpu/drm/xe/xe_sched_job.c | 2 +- > > > > > drivers/gpu/drm/xe/xe_sched_job_types.h | 3 +- > > > > > drivers/gpu/drm/xe/xe_sync.c | 5 +- > > > > > drivers/gpu/drm/xe/xe_vram.c | 12 +- > > > > > drivers/gpu/drm/xe/xe_wa_oob.rules | 2 + > > > > > include/uapi/drm/xe_drm.h | 17 + > > > > > 43 files changed, 997 insertions(+), 487 deletions(-) > > > > > >
Hi, Dave & Simona, This week's drm-xe-next PR Thanks, Thomas drm-xe-next-2024-10-24: UAPI Changes: - Define and parse OA sync properties (Ashutosh) Driver Changes: - Add caller info to xe_gt_reset_async (Nirmoy) - A large forcewake rework / cleanup (Himal) - A g2h response timeout fix (Badal) - A PTL workaround (Vinay) - Handle unreliable MMIO reads during forcewake (Shuicheng) - Ufence user-space access fixes (Nirmoy) - Annotate flexible arrays (Matthew Brost) - Enable GuC lite restore (Fei) - Prevent GuC register capture on VF (Zhanjun) - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) - Parallel queues fix on GT reset (Nirmoy) - Move reference grabbing to a job's dma-fence (Matt Brost) - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) - OA synchronization support (Ashutosh) The following changes since commit 2eb460ab9f4bc5b575f52568d17936da0af681d8: drm/xe: Enlarge the invalidation timeout from 150 to 500 (2024-10-16 16:11:10 +0100) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-2024-10-24 for you to fetch changes up to 85d3f9e84e0628c412b69aa99b63654dfa08ad68: drm/xe/oa: Allow only certain property changes from config (2024-10-23 12:42:20 -0700) ---------------------------------------------------------------- UAPI Changes: - Define and parse OA sync properties (Ashutosh) Driver Changes: - Add caller info to xe_gt_reset_async (Nirmoy) - A large forcewake rework / cleanup (Himal) - A g2h response timeout fix (Badal) - A PTL workaround (Vinay) - Handle unreliable MMIO reads during forcewake (Shuicheng) - Ufence user-space access fixes (Nirmoy) - Annotate flexible arrays (Matthew Brost) - Enable GuC lite restore (Fei) - Prevent GuC register capture on VF (Zhanjun) - Show VFs VRAM / LMEM provisioning summary over debugfs (Michal) - Parallel queues fix on GT reset (Nirmoy) - Move reference grabbing to a job's dma-fence (Matt Brost) - Mark a number of local workqueues WQ_MEM_RECLAIM (Matt Brost) - OA synchronization support (Ashutosh) ---------------------------------------------------------------- Ashutosh Dixit (7): drm/xe/oa: Separate batch submission from waiting for completion drm/xe/oa/uapi: Define and parse OA sync properties drm/xe/oa: Add input fence dependencies drm/xe/oa: Signal output fences drm/xe/oa: Move functions up so they can be reused for config ioctl drm/xe/oa: Add syncs support to OA config ioctl drm/xe/oa: Allow only certain property changes from config Badal Nilawar (1): drm/xe/guc/ct: Flush g2h worker in case of g2h response timeout Fei Yang (1): drm/xe: enable lite restore Himal Prasad Ghimiray (26): drm/xe: Add member initialized_domains to xe_force_wake() drm/xe/forcewake: Change awake_domain datatype drm/xe/forcewake: Add a helper xe_force_wake_ref_has_domain() drm/xe: Error handling in xe_force_wake_get() drm/xe: Modify xe_force_wake_put to handle _get returned mask drm/xe/device: Update handling of xe_force_wake_get return drm/xe/hdcp: Update handling of xe_force_wake_get return drm/xe/gsc: Update handling of xe_force_wake_get return drm/xe/gt: Update handling of xe_force_wake_get return drm/xe/xe_gt_idle: Update handling of xe_force_wake_get return drm/xe/devcoredump: Update handling of xe_force_wake_get return drm/xe/tests/mocs: Update xe_force_wake_get() return handling drm/xe/mocs: Update handling of xe_force_wake_get return drm/xe/xe_drm_client: Update handling of xe_force_wake_get return drm/xe/xe_gt_debugfs: Update handling of xe_force_wake_get return drm/xe/guc: Update handling of xe_force_wake_get return drm/xe/huc: Update handling of xe_force_wake_get return drm/xe/oa: Handle force_wake_get failure in xe_oa_stream_init() drm/xe/pat: Update handling of xe_force_wake_get return drm/xe/gt_tlb_invalidation_ggtt: Update handling of xe_force_wake_get return drm/xe/xe_reg_sr: Update handling of xe_force_wake_get return drm/xe/query: Update handling of xe_force_wake_get return drm/xe/vram: Update handling of xe_force_wake_get return drm/xe: forcewake debugfs open fails on xe_forcewake_get failure drm/xe: Ensure __must_check for xe_force_wake_get() return drm/xe: Change return type to void for xe_force_wake_put Matthew Brost (5): drm/xe: Use __counted_by for flexible arrays drm/xe: Take ref to job's fence in arm drm/xe: Mark GGTT work queue with WQ_MEM_RECLAIM drm/xe: Mark G2H work queue with WQ_MEM_RECLAIM drm/xe: Mark GT work queue with WQ_MEM_RECLAIM Michal Wajdeczko (1): drm/xe/pf: Show VFs LMEM provisioning summary over debugfs Nirmoy Das (4): drm/xe: Add caller info to xe_gt_reset_async drm/xe/ufence: Prefetch ufence addr to catch bogus address drm/xe/ufence: Warn if mmget_not_zero() fails drm/xe: Don't restart parallel queues multiple times on GT reset Shuicheng Lin (1): drm/xe: Handle unreliable MMIO reads during forcewake Vinay Belgaumkar (1): drm/xe/ptl: Apply Wa_14022866841 Zhanjun Dong (1): drm/xe/guc: Prevent GuC register capture running on VF drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + drivers/gpu/drm/xe/display/xe_hdcp_gsc.c | 6 +- drivers/gpu/drm/xe/tests/xe_mocs.c | 18 +- drivers/gpu/drm/xe/xe_debugfs.c | 27 +- drivers/gpu/drm/xe/xe_devcoredump.c | 14 +- drivers/gpu/drm/xe/xe_device.c | 25 +- drivers/gpu/drm/xe/xe_drm_client.c | 8 +- drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 +- drivers/gpu/drm/xe/xe_execlist.c | 2 +- drivers/gpu/drm/xe/xe_force_wake.c | 134 ++++-- drivers/gpu/drm/xe/xe_force_wake.h | 23 +- drivers/gpu/drm/xe/xe_force_wake_types.h | 6 +- drivers/gpu/drm/xe/xe_ggtt.c | 2 +- drivers/gpu/drm/xe/xe_gsc.c | 23 +- drivers/gpu/drm/xe/xe_gsc_proxy.c | 9 +- drivers/gpu/drm/xe/xe_gt.c | 110 +++-- drivers/gpu/drm/xe/xe_gt_debugfs.c | 13 +- drivers/gpu/drm/xe/xe_gt_idle.c | 26 +- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 35 ++ drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c | 5 + drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 5 +- drivers/gpu/drm/xe/xe_guc.c | 15 +- drivers/gpu/drm/xe/xe_guc_ads.c | 5 + drivers/gpu/drm/xe/xe_guc_capture.c | 8 +- drivers/gpu/drm/xe/xe_guc_ct.c | 20 +- drivers/gpu/drm/xe/xe_guc_fwif.h | 1 + drivers/gpu/drm/xe/xe_guc_log.c | 9 +- drivers/gpu/drm/xe/xe_guc_pc.c | 50 +- drivers/gpu/drm/xe/xe_guc_submit.c | 29 +- drivers/gpu/drm/xe/xe_huc.c | 8 +- drivers/gpu/drm/xe/xe_mocs.c | 14 +- drivers/gpu/drm/xe/xe_oa.c | 678 +++++++++++++++++++--------- drivers/gpu/drm/xe/xe_oa_types.h | 12 + drivers/gpu/drm/xe/xe_pat.c | 65 ++- drivers/gpu/drm/xe/xe_query.c | 10 +- drivers/gpu/drm/xe/xe_reg_sr.c | 24 +- drivers/gpu/drm/xe/xe_sched_job.c | 2 +- drivers/gpu/drm/xe/xe_sched_job_types.h | 3 +- drivers/gpu/drm/xe/xe_sync.c | 5 +- drivers/gpu/drm/xe/xe_vram.c | 12 +- drivers/gpu/drm/xe/xe_wa_oob.rules | 2 + include/uapi/drm/xe_drm.h | 17 + 43 files changed, 997 insertions(+), 487 deletions(-)