November 2020 Archives by date
Starting: Mon Nov 2 08:12:48 UTC 2020
Ending: Mon Nov 30 10:58:08 UTC 2020
Messages: 2658
- [PATCH 01/61] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/61] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/61] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/61] drm/i915: Add missing -EDEADLK handling to execbuf pinning
Maarten Lankhorst
- [PATCH 05/61] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/61] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/61] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/61] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/61] drm/i915: Convert i915_gem_object_attach_phys() to ww locking
Maarten Lankhorst
- [PATCH 10/61] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/61] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/61] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/61] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/61] drm/i915: Reject UNSYNCHRONIZED for userptr
Maarten Lankhorst
- [PATCH 15/61] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v4.
Maarten Lankhorst
- [PATCH 16/61] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 17/61] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 18/61] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 19/61] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 20/61] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 21/61] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 22/61] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 23/61] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 24/61] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 25/61] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 26/61] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/61] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 28/61] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 29/61] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 30/61] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 31/61] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 32/61] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 33/61] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 34/61] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 35/61] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 36/61] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 37/61] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 38/61] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 39/61] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 40/61] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 41/61] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/61] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/61] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/61] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/61] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/61] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/61] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/61] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/61] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 50/61] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/61] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/61] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/61] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/61] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/61] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/61] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/61] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/61] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/61] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/61] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 61/61] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- ✗ Fi.CI.BAT: failure for series starting with [01/61] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH 01/61] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/61] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/61] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/61] drm/i915: Add missing -EDEADLK handling to execbuf pinning
Maarten Lankhorst
- [PATCH 05/61] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/61] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/61] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/61] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/61] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/61] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/61] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/61] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/61] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/61] drm/i915: Reject UNSYNCHRONIZED for userptr
Maarten Lankhorst
- [PATCH 15/61] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v4.
Maarten Lankhorst
- [PATCH 16/61] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 17/61] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 18/61] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 19/61] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 20/61] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 21/61] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 22/61] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 23/61] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 24/61] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 25/61] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 26/61] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/61] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 28/61] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 29/61] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 30/61] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 31/61] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 32/61] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 33/61] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 34/61] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 35/61] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 36/61] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 37/61] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 38/61] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 39/61] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 40/61] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 41/61] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/61] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/61] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/61] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/61] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/61] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/61] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/61] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/61] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 50/61] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/61] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/61] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/61] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/61] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/61] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/61] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/61] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/61] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/61] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/61] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 61/61] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- ✓ Fi.CI.BAT: success for series starting with [01/61] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH 01/76] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 02/76] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 03/76] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 04/76] drm/i915/gt: Expose more parameters for emitting writes into the ring
Chris Wilson
- [PATCH 05/76] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 06/76] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 07/76] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 08/76] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 09/76] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 10/76] drm/i915/gem: Perform all asynchronous waits prior to marking payload start
Chris Wilson
- [PATCH 11/76] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 12/76] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 13/76] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 14/76] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 15/76] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 16/76] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 17/76] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 18/76] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 19/76] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 20/76] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 21/76] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 22/76] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 23/76] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 24/76] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 25/76] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 26/76] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 27/76] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 28/76] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 29/76] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 30/76] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 31/76] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 32/76] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 33/76] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 34/76] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 35/76] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 36/76] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 37/76] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 38/76] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 39/76] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 40/76] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 41/76] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 42/76] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 43/76] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 44/76] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 45/76] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 46/76] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 47/76] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 48/76] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 49/76] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 50/76] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 51/76] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 52/76] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 53/76] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 54/76] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 55/76] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 56/76] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 57/76] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 58/76] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 59/76] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 60/76] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 61/76] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 62/76] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 63/76] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 64/76] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 65/76] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 66/76] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 67/76] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 68/76] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 69/76] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 70/76] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 71/76] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 72/76] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 73/76] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 74/76] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 75/76] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 76/76] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/76] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/61] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH 01/78] drm/i915: Hold onto an explicit ref to i915_vma_work.pinned
Chris Wilson
- [PATCH 02/78] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 03/78] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 04/78] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 05/78] drm/i915/gt: Expose more parameters for emitting writes into the ring
Chris Wilson
- [PATCH 06/78] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 07/78] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 08/78] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 09/78] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 10/78] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 11/78] drm/i915/gem: Perform all asynchronous waits prior to marking payload start
Chris Wilson
- [PATCH 12/78] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 13/78] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 14/78] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 15/78] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 16/78] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 17/78] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 18/78] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 19/78] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 20/78] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 21/78] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 22/78] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 23/78] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 24/78] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 25/78] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 26/78] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 27/78] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 28/78] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 29/78] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 30/78] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 31/78] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 32/78] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 33/78] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 34/78] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 35/78] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 36/78] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 37/78] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 38/78] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 39/78] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 40/78] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 41/78] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 42/78] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 43/78] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 44/78] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 45/78] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 46/78] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 47/78] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 48/78] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 49/78] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 50/78] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 51/78] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 52/78] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 53/78] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 54/78] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 55/78] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 56/78] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 57/78] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 58/78] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 59/78] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 60/78] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 61/78] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 62/78] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 63/78] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 64/78] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 65/78] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 66/78] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 67/78] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 68/78] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 69/78] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 70/78] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 71/78] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 72/78] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 73/78] cl-g6
Chris Wilson
- [PATCH 74/78] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 75/78] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 76/78] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 77/78] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 78/78] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/78] drm/i915: Hold onto an explicit ref to i915_vma_work.pinned
Patchwork
- [PATCH 01/79] drm/i915/gt: Expose more parameters for emitting writes into the ring
Chris Wilson
- [PATCH 02/79] drm/i915/gt: Flush xcs before tgl breadcrumbs
Chris Wilson
- [PATCH 03/79] drm/i915: Hold onto an explicit ref to i915_vma_work.pinned
Chris Wilson
- [PATCH 04/79] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 05/79] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 06/79] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 07/79] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 08/79] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 09/79] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 10/79] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 11/79] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 12/79] drm/i915/gem: Perform all asynchronous waits prior to marking payload start
Chris Wilson
- [PATCH 13/79] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 14/79] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 15/79] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 16/79] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 17/79] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 18/79] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 19/79] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 20/79] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 21/79] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 22/79] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 23/79] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 24/79] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 25/79] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 26/79] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 27/79] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 28/79] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 29/79] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 30/79] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 31/79] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 32/79] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 33/79] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 34/79] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 35/79] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 36/79] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 37/79] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 38/79] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 39/79] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 40/79] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 41/79] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 42/79] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 43/79] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 44/79] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 45/79] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 46/79] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 47/79] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 48/79] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 49/79] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 50/79] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 51/79] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 52/79] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 53/79] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 54/79] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 55/79] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 56/79] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 57/79] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 58/79] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 59/79] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 60/79] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 61/79] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 62/79] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 63/79] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 64/79] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 65/79] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 66/79] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 67/79] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 68/79] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 69/79] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 70/79] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 71/79] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 72/79] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 73/79] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 74/79] cl-g6
Chris Wilson
- [PATCH 75/79] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 76/79] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 77/79] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 78/79] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 79/79] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- [PATCH 02/79] drm/i915/gt: Flush xcs before tgl breadcrumbs
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/79] drm/i915/gt: Expose more parameters for emitting writes into the ring
Patchwork
- [PATCH 01/27] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 02/27] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 03/27] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 04/27] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 05/27] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 06/27] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 07/27] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 08/27] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 09/27] drm/i915/gem: Perform all asynchronous waits prior to marking payload start
Chris Wilson
- [PATCH 10/27] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 11/27] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 12/27] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 13/27] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 14/27] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 15/27] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 16/27] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 17/27] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 18/27] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 19/27] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 20/27] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 21/27] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 22/27] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 23/27] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 24/27] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 25/27] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 26/27] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 27/27] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/27] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Patchwork
- [PATCH 01/21] drm/i915/gvt: Remove incorrect kerneldoc marking
Chris Wilson
- [PATCH 02/21] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 03/21] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 04/21] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 05/21] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 06/21] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 07/21] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 08/21] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 09/21] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 10/21] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 11/21] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 12/21] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 13/21] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 14/21] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 15/21] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 16/21] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 17/21] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 18/21] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 19/21] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 20/21] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 21/21] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/21] drm/i915/gvt: Remove incorrect kerneldoc marking
Patchwork
- [PATCH 01/62] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/62] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/62] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/62] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/62] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/62] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/62] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/62] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/62] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/62] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/62] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/62] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/62] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/62] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/62] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/62] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/62] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 18/62] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 19/62] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 20/62] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 21/62] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 22/62] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 23/62] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 24/62] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 25/62] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 26/62] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/62] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 28/62] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 29/62] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 30/62] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 31/62] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 32/62] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 33/62] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 34/62] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 35/62] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 36/62] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 37/62] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 38/62] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 39/62] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 40/62] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 41/62] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/62] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/62] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/62] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/62] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/62] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/62] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/62] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/62] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/62] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/62] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/62] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/62] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/62] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/62] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/62] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/62] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/62] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/62] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/62] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/62] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/62] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- ✓ Fi.CI.BAT: success for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH 01/22] drm/i915/gem: Allow backends to override pread implementation
Chris Wilson
- [PATCH 02/22] drm/i915/gem: Pull phys pread/pwrite implementations to the backend
Chris Wilson
- [PATCH 03/22] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 04/22] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 05/22] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 06/22] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 07/22] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 08/22] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 09/22] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 10/22] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 11/22] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 12/22] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 13/22] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 14/22] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 15/22] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 16/22] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 17/22] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 18/22] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 19/22] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 20/22] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 21/22] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 22/22] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- ✓ Fi.CI.BAT: success for series starting with [01/22] drm/i915/gem: Allow backends to override pread implementation
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/22] drm/i915/gem: Allow backends to override pread implementation
Patchwork
- [PATCH] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- ✓ Fi.CI.BAT: success for drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- ✗ Fi.CI.IGT: failure for drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- [RFC] consolidated wa table
Tvrtko Ursulin
- [RFC] consolidated wa table
Chris Wilson
- [RFC v2] consolidated wa table
Tvrtko Ursulin
- ✓ Fi.CI.BAT: success for consolidated wa table
Patchwork
- ✓ Fi.CI.BAT: success for consolidated wa table (rev2)
Patchwork
- [RFC] consolidated wa table
Tvrtko Ursulin
- [RFC] consolidated wa table
Chris Wilson
- [RFC] consolidated wa table
Tvrtko Ursulin
- ✗ Fi.CI.IGT: failure for consolidated wa table (rev2)
Patchwork
- [RFC] consolidated wa table
Tvrtko Ursulin
- ✗ Fi.CI.BAT: failure for consolidated wa table (rev3)
Patchwork
- [RFC] consolidated wa table
Chris Wilson
- [PATCH] drm/i915/gt: Start timeline with a wrap
Chris Wilson
- ✗ Fi.CI.BAT: failure for drm/i915/gt: Start timeline with a wrap
Patchwork
- [PATCH] cheapo
Matthew Auld
- [PATCH 1/2] nb-1
Matthew Auld
- [PATCH 2/2] hmmmm
Matthew Auld
- [PATCH] invrange
Chris Wilson
- [PATCH 01/62] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/62] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/62] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/62] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/62] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/62] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/62] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/62] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/62] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/62] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/62] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/62] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/62] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/62] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/62] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/62] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/62] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 18/62] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 19/62] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 20/62] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 21/62] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 22/62] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 23/62] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 24/62] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 25/62] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 26/62] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/62] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 28/62] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 29/62] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 30/62] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 31/62] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 32/62] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 33/62] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 34/62] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 35/62] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 36/62] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 37/62] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 38/62] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 39/62] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 40/62] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 41/62] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/62] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/62] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/62] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/62] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/62] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/62] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/62] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/62] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/62] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/62] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/62] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/62] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/62] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/62] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/62] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/62] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/62] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/62] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/62] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/62] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/62] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- ✓ Fi.CI.BAT: success for cheapo
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [1/2] nb-1
Patchwork
- ✗ Fi.CI.BUILD: warning for series starting with [1/2] nb-1
Patchwork
- ✓ Fi.CI.BAT: success for invrange
Patchwork
- [PATCH 1/2] drm/i915: stop registering if drm_dev_register() fails
Lucas De Marchi
- [PATCH 2/2] reorder register
Lucas De Marchi
- ✓ Fi.CI.BAT: success for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.IGT: failure for cheapo
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: stop registering if drm_dev_register() fails
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/2] nb-1
Patchwork
- ✗ Fi.CI.IGT: failure for invrange
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: stop registering if drm_dev_register() fails
Patchwork
- [PATCH 01/35] drm/i915/gt: Include semaphore status in print_request()
Chris Wilson
- [PATCH 02/35] drm/i915: Lift i915_request_show()
Chris Wilson
- [PATCH 03/35] drm/i915/gt: Show all active timelines for debugging
Chris Wilson
- [PATCH 04/35] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 05/35] drm/i915: Show timeline dependencies for debug
Chris Wilson
- [PATCH 06/35] drm/i915: Disable atomics in L3 for gen9
Chris Wilson
- [PATCH 07/35] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 08/35] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 09/35] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 10/35] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 11/35] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 12/35] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 13/35] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 14/35] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 15/35] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 16/35] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 17/35] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 18/35] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 19/35] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 20/35] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 21/35] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 22/35] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 23/35] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 24/35] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 25/35] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 26/35] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 27/35] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 28/35] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 29/35] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 30/35] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 31/35] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 32/35] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 33/35] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 34/35] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 35/35] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/35] drm/i915/gt: Include semaphore status in print_request()
Patchwork
- [PATCH 01/66] drm/i915/gt: Include semaphore status in print_request()
Chris Wilson
- [PATCH 02/66] drm/i915: Lift i915_request_show()
Chris Wilson
- [PATCH 03/66] drm/i915/gt: Show all active timelines for debugging
Chris Wilson
- [PATCH 04/66] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 05/66] drm/i915: Show timeline dependencies for debug
Chris Wilson
- [PATCH 06/66] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 07/66] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 08/66] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 09/66] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 10/66] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 11/66] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 12/66] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 13/66] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 14/66] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 15/66] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 16/66] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 17/66] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 18/66] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 19/66] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 20/66] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 21/66] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 22/66] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 23/66] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 24/66] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 25/66] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 26/66] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 27/66] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 28/66] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 29/66] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 30/66] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 31/66] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 32/66] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 33/66] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 34/66] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 35/66] drm/i915: Disable atomics in L3 for gen9
Chris Wilson
- [PATCH 36/66] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 37/66] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 38/66] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 39/66] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 40/66] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 41/66] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 42/66] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 43/66] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 44/66] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 45/66] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 46/66] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 47/66] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 48/66] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 49/66] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 50/66] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 51/66] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 52/66] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 53/66] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 54/66] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 55/66] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 56/66] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 57/66] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 58/66] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 59/66] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 60/66] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 61/66] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 62/66] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 63/66] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 64/66] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 65/66] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 66/66] drm/i915: Fair low-latency scheduling
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/66] drm/i915/gt: Include semaphore status in print_request()
Patchwork
- [PATCH] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- ✗ Fi.CI.BAT: failure for drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH 01/65] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/65] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/65] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/65] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/65] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/65] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/65] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/65] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/65] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/65] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/65] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/65] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/65] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/65] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/65] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/65] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/65] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 18/65] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 19/65] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 20/65] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 21/65] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 22/65] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 23/65] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 24/65] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 25/65] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 26/65] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/65] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 28/65] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 29/65] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 30/65] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 31/65] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 32/65] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 33/65] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 34/65] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 35/65] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 36/65] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 37/65] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 38/65] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 39/65] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 40/65] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 41/65] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/65] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/65] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/65] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/65] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/65] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/65] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/65] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/65] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/65] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/65] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/65] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/65] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/65] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/65] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/65] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/65] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/65] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/65] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/65] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/65] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/65] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- [PATCH 63/65] drm/i915: Idea to implement eviction locking
Maarten Lankhorst
- [PATCH 64/65] drm/i915: Keep lazy reference to userptr pages
Maarten Lankhorst
- [PATCH 65/65] fence annotations on gpu reset
Maarten Lankhorst
- [PATCH 0/7] Allow privileged user to map the OA buffer
Umesh Nerlige Ramappa
- [PATCH 1/7] drm/i915/perf: Ensure observation logic is not clock gated
Umesh Nerlige Ramappa
- [PATCH 2/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Umesh Nerlige Ramappa
- [PATCH 3/7] drm/i915/gt: Add a reference to the engine in i915_wa_list
Umesh Nerlige Ramappa
- [PATCH 4/7] drm/i915/perf: Whitelist OA report trigger registers
Umesh Nerlige Ramappa
- [PATCH 5/7] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Umesh Nerlige Ramappa
- [PATCH 6/7] drm/i915/perf: Whitelist OA counter and buffer registers
Umesh Nerlige Ramappa
- [PATCH 7/7] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Umesh Nerlige Ramappa
- [PATCH 0/7] Allow privileged user to map the OA buffer
Umesh Nerlige Ramappa
- [PATCH 1/7] drm/i915/perf: Ensure observation logic is not clock gated
Umesh Nerlige Ramappa
- [PATCH 2/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Umesh Nerlige Ramappa
- [PATCH 3/7] drm/i915/gt: Add a reference to the engine in i915_wa_list
Umesh Nerlige Ramappa
- [PATCH 4/7] drm/i915/perf: Whitelist OA report trigger registers
Umesh Nerlige Ramappa
- [PATCH 5/7] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Umesh Nerlige Ramappa
- [PATCH 6/7] drm/i915/perf: Whitelist OA counter and buffer registers
Umesh Nerlige Ramappa
- [PATCH 7/7] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Umesh Nerlige Ramappa
- [PATCH 01/34] drm/i915/selftests: Small tweak to put the termination conditions together
Chris Wilson
- [PATCH 02/34] drm/i915/gt: Include semaphore status in print_request()
Chris Wilson
- [PATCH 03/34] drm/i915: Lift i915_request_show()
Chris Wilson
- [PATCH 04/34] drm/i915/gt: Show all active timelines for debugging
Chris Wilson
- [PATCH 05/34] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 06/34] drm/i915: Show timeline dependencies for debug
Chris Wilson
- [PATCH 07/34] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 08/34] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 09/34] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 10/34] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 11/34] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 12/34] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 13/34] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 14/34] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 15/34] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 16/34] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 17/34] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 18/34] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 19/34] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 20/34] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 21/34] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 22/34] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 23/34] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 24/34] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 25/34] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 26/34] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 27/34] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 28/34] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 29/34] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 30/34] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 31/34] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 32/34] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 33/34] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 34/34] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 0/7] Allow privileged user to map the OA buffer
Umesh Nerlige Ramappa
- [PATCH 1/7] drm/i915/perf: Ensure observation logic is not clock gated
Umesh Nerlige Ramappa
- [PATCH 2/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Umesh Nerlige Ramappa
- [PATCH 3/7] drm/i915/gt: Add a reference to the engine in i915_wa_list
Umesh Nerlige Ramappa
- [PATCH 4/7] drm/i915/perf: Whitelist OA report trigger registers
Umesh Nerlige Ramappa
- [PATCH 5/7] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Umesh Nerlige Ramappa
- [PATCH 6/7] drm/i915/perf: Whitelist OA counter and buffer registers
Umesh Nerlige Ramappa
- [PATCH 7/7] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Umesh Nerlige Ramappa
- ✗ Fi.CI.BAT: failure for series starting with [01/65] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH 1/3] drm/i915: stop registering if drm_dev_register() fails
Lucas De Marchi
- [PATCH 2/3] drm/i915: group display-related register calls
Lucas De Marchi
- [PATCH 3/3] drm/i915/display: move register functions to display/
Lucas De Marchi
- ✗ Fi.CI.BAT: failure for series starting with [01/34] drm/i915/selftests: Small tweak to put the termination conditions together
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915: stop registering if drm_dev_register() fails
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/3] drm/i915: stop registering if drm_dev_register() fails
Patchwork
- [PATCH] Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Chris Wilson
- ✓ Fi.CI.BAT: success for Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Patchwork
- ✗ Fi.CI.IGT: failure for Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Patchwork
- [PATCH 01/14] Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Chris Wilson
- [PATCH 02/14] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 03/14] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 04/14] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 05/14] drm/i915: Track runtime spent in unreachable intel_contexts
Chris Wilson
- [PATCH 06/14] drm/i915: Track runtime spent in closed GEM contexts
Chris Wilson
- [PATCH 07/14] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/14] drm/i915: Expose per-engine client busyness
Chris Wilson
- [PATCH 09/14] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/14] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/14] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/14] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 13/14] proc: Show GPU runtimes
Chris Wilson
- [PATCH 14/14] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✓ Fi.CI.BAT: success for series starting with [01/14] Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Patchwork
- ✓ Fi.CI.IGT: success for series starting with [01/14] Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Patchwork
- [PATCH 01/14] Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Chris Wilson
- [PATCH 02/14] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 03/14] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 04/14] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 05/14] drm/i915: Track runtime spent in unreachable intel_contexts
Chris Wilson
- [PATCH 06/14] drm/i915: Track runtime spent in closed GEM contexts
Chris Wilson
- [PATCH 07/14] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/14] drm/i915: Expose per-engine client busyness
Chris Wilson
- [PATCH 09/14] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/14] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/14] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/14] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 13/14] proc: Show GPU runtimes
Chris Wilson
- [PATCH 14/14] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/14] Revert "drm: convert drm_atomic_uapi.c to new debug helpers"
Patchwork
- [PATCH] need-resched
Chris Wilson
- ✗ Fi.CI.BAT: failure for need-resched
Patchwork
- [PATCH 01/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 02/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 03/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 04/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 05/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 06/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 07/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/13] drm/i915: Expose per-engine client busyness
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/13] drm/i915/gt: Track the overall busy time
Patchwork
- [PATCH 01/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 02/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 03/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 04/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 05/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 06/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 07/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/13] drm/i915: Expose per-engine client busyness
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/13] drm/i915/gt: Track the overall busy time
Patchwork
- [PATCH 01/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 02/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 03/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 04/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 05/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 06/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 07/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/13] drm/i915: Expose per-engine client busyness
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/13] drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- [PATCH 0/7] Try OA workaround changes
Umesh Nerlige Ramappa
- [PATCH 1/7] drm/i915/perf: Ensure observation logic is not clock gated
Umesh Nerlige Ramappa
- [PATCH 2/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Umesh Nerlige Ramappa
- [PATCH 3/7] drm/i915/gt: Add a reference to the engine in i915_wa_list
Umesh Nerlige Ramappa
- [PATCH 4/7] drm/i915/perf: Whitelist OA report trigger registers
Umesh Nerlige Ramappa
- [PATCH 5/7] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Umesh Nerlige Ramappa
- [PATCH 6/7] drm/i915/perf: Whitelist OA counter and buffer registers
Umesh Nerlige Ramappa
- [PATCH 7/7] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Umesh Nerlige Ramappa
- [PATCH 1/6] drm/i915/perf: Ensure observation logic is not clock gated
Chris Wilson
- [PATCH 2/6] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Chris Wilson
- [PATCH 3/6] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Chris Wilson
- [PATCH 4/6] drm/i915/perf: Whitelist OA report trigger registers
Chris Wilson
- [PATCH 5/6] drm/i915/perf: Whitelist OA counter and buffer registers
Chris Wilson
- [PATCH 6/6] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Chris Wilson
- [PATCH 01/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 02/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 03/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 04/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 05/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 06/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 07/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/13] drm/i915: Expose per-engine client busyness
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✗ Fi.CI.BUILD: failure for Try OA workaround changes
Patchwork
- ✗ Fi.CI.BUILD: failure for series starting with [1/6] drm/i915/perf: Ensure observation logic is not clock gated
Patchwork
- [PATCH 0/7] Try OA workaround changes (rebase)
Umesh Nerlige Ramappa
- [PATCH 1/7] drm/i915/perf: Ensure observation logic is not clock gated
Umesh Nerlige Ramappa
- [PATCH 2/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Umesh Nerlige Ramappa
- [PATCH 3/7] drm/i915/gt: Add a reference to the engine in i915_wa_list
Umesh Nerlige Ramappa
- [PATCH 4/7] drm/i915/perf: Whitelist OA report trigger registers
Umesh Nerlige Ramappa
- [PATCH 5/7] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Umesh Nerlige Ramappa
- [PATCH 6/7] drm/i915/perf: Whitelist OA counter and buffer registers
Umesh Nerlige Ramappa
- [PATCH 7/7] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Umesh Nerlige Ramappa
- [PATCH 1/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Chris Wilson
- [PATCH 2/7] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Chris Wilson
- [PATCH 3/7] allow/deny-white
Chris Wilson
- [PATCH 4/7] drm/i915/perf: Ensure observation logic is not clock gated
Chris Wilson
- [PATCH 5/7] drm/i915/perf: Whitelist OA report trigger registers
Chris Wilson
- [PATCH 6/7] drm/i915/perf: Whitelist OA counter and buffer registers
Chris Wilson
- [PATCH 7/7] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/13] drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- [PATCH 01/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 02/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 03/13] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 04/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 05/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 06/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 07/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 08/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- ✓ Fi.CI.BAT: success for series starting with [1/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Patchwork
- ✗ Fi.CI.BAT: failure for series starting with [01/13] drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/7] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Patchwork
- [PATCH 01/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 02/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 03/13] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 04/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 05/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 06/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 07/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 08/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- [PATCH 1/8] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Chris Wilson
- [PATCH 2/8] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Chris Wilson
- [PATCH 3/8] drm/i915/gt: Check for conflicting RING_NONPRIV
Chris Wilson
- [PATCH 4/8] drm/i915/gt: Enable dynamic adjustment of RING_NONPRIV
Chris Wilson
- [PATCH 5/8] drm/i915/perf: Ensure observation logic is not clock gated
Chris Wilson
- [PATCH 6/8] drm/i915/perf: Whitelist OA report trigger registers
Chris Wilson
- [PATCH 7/8] drm/i915/perf: Whitelist OA counter and buffer registers
Chris Wilson
- [PATCH 8/8] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Chris Wilson
- [PATCH 1/8] drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
Chris Wilson
- [PATCH 2/8] drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
Chris Wilson
- [PATCH 3/8] drm/i915/gt: Check for conflicting RING_NONPRIV
Chris Wilson
- [PATCH 4/8] drm/i915/gt: Enable dynamic adjustment of RING_NONPRIV
Chris Wilson
- [PATCH 5/8] drm/i915/perf: Ensure observation logic is not clock gated
Chris Wilson
- [PATCH 6/8] drm/i915/perf: Whitelist OA report trigger registers
Chris Wilson
- [PATCH 7/8] drm/i915/perf: Whitelist OA counter and buffer registers
Chris Wilson
- [PATCH 8/8] drm/i915/perf: Map OA buffer to user space for gen12 performance query
Chris Wilson
- [PATCH 01/13] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 02/13] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 03/13] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 04/13] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 05/13] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 06/13] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 07/13] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 08/13] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 09/13] drm/i915: Track context current active time
Chris Wilson
- [PATCH 10/13] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 11/13] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 12/13] proc: Show GPU runtimes
Chris Wilson
- [PATCH 13/13] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- [PATCH] drm/i915/display: Print backlight controller and type
José Roberto de Souza
- [PATCH 01/72] drm/i915/phy: Quieten state loss across suspend
Chris Wilson
- [PATCH 02/72] drm/i915/gem: Remove incorrect early dbg print
Chris Wilson
- [PATCH 03/72] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 04/72] drm/i915/selftests: Small tweak to put the termination conditions together
Chris Wilson
- [PATCH 05/72] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 06/72] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 07/72] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 08/72] drm/i915/gt: Include semaphore status in print_request()
Chris Wilson
- [PATCH 09/72] drm/i915: Lift i915_request_show()
Chris Wilson
- [PATCH 10/72] drm/i915/gt: Show all active timelines for debugging
Chris Wilson
- [PATCH 11/72] drm/i915: Lift waiter/signaler iterators
Chris Wilson
- [PATCH 12/72] drm/i915: Show timeline dependencies for debug
Chris Wilson
- [PATCH 13/72] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 14/72] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 15/72] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 16/72] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 17/72] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 18/72] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 19/72] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 20/72] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 21/72] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 22/72] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 23/72] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 24/72] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 25/72] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 26/72] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 27/72] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 28/72] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 29/72] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 30/72] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 31/72] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 32/72] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 33/72] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 34/72] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 35/72] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 36/72] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 37/72] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 38/72] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 39/72] drm/i915: Disable atomics in L3 for gen9
Chris Wilson
- [PATCH 40/72] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 41/72] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 42/72] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 43/72] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 44/72] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 45/72] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 46/72] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 47/72] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 48/72] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 49/72] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 50/72] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 51/72] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 52/72] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 53/72] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 54/72] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 55/72] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 56/72] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 57/72] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 58/72] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 59/72] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 60/72] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 61/72] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 62/72] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 63/72] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 64/72] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 65/72] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 66/72] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 67/72] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 68/72] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 69/72] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 70/72] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 71/72] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 72/72] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 1/7] drm/i915/phy: Quieten state loss across suspend
Chris Wilson
- [PATCH 2/7] drm/i915/gem: Remove incorrect early dbg print
Chris Wilson
- [PATCH 3/7] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 4/7] drm/i915/selftests: Small tweak to put the termination conditions together
Chris Wilson
- [PATCH 5/7] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 6/7] drm/i915/gt: Ignore dt==0 for reporting underflows
Chris Wilson
- [PATCH 7/7] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH] HAX: Experimenting a different init sequence...
Rodrigo Vivi
- ✗ Fi.CI.BAT: failure for series starting with [01/13] drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [01/72] drm/i915/phy: Quieten state loss across suspend
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [1/7] drm/i915/phy: Quieten state loss across suspend
Patchwork
- ✓ Fi.CI.BAT: success for HAX: Experimenting a different init sequence...
Patchwork
- ✗ Fi.CI.IGT: failure for drm/i915/display: Print backlight controller and type
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/72] drm/i915/phy: Quieten state loss across suspend
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/7] drm/i915/phy: Quieten state loss across suspend
Patchwork
- [PATCH] drm/i915/display: Warn about types of backlight not handled
José Roberto de Souza
- ✗ Fi.CI.IGT: failure for HAX: Experimenting a different init sequence...
Patchwork
- ✗ Fi.CI.BAT: failure for drm/i915/display: Warn about types of backlight not handled
Patchwork
- [PATCH 01/12] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 02/12] drm/i915/gt: Track the overall busy time
Chris Wilson
- [PATCH 03/12] drm/i915: Expose list of clients in sysfs
Chris Wilson
- [PATCH 04/12] drm/i915: Update client name on context create
Chris Wilson
- [PATCH 05/12] drm/i915: Make GEM contexts track DRM clients
Chris Wilson
- [PATCH 06/12] drm/i915: Track runtime spent in closed and unreachable GEM contexts
Chris Wilson
- [PATCH 07/12] drm/i915: Track all user contexts per client
Chris Wilson
- [PATCH 08/12] drm/i915: Track context current active time
Chris Wilson
- [PATCH 09/12] drm/i915: Prefer software tracked context busyness
Chris Wilson
- [PATCH 10/12] drm/i915: Look up clients by pid
Chris Wilson
- [PATCH 11/12] proc: Show GPU runtimes
Chris Wilson
- [PATCH 12/12] drm/i915/gt: Export device and per-process runtimes via procfs
Chris Wilson
- [PATCH 01/62] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/62] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/62] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/62] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/62] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/62] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/62] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/62] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/62] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/62] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/62] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/62] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/62] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/62] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/62] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/62] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/62] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 18/62] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 19/62] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 20/62] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 21/62] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 22/62] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 23/62] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 24/62] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 25/62] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 26/62] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/62] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 28/62] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 29/62] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 30/62] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 31/62] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 32/62] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 33/62] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 34/62] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 35/62] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 36/62] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 37/62] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 38/62] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 39/62] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 40/62] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 41/62] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/62] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/62] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/62] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/62] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/62] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/62] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/62] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/62] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/62] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/62] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/62] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/62] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/62] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/62] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/62] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/62] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/62] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/62] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/62] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/62] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/62] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- [PATCH 01/62] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/62] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/62] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/62] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/62] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/62] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/62] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/62] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/62] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/62] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/62] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/62] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/62] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/62] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/62] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/62] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/62] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 18/62] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 19/62] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 20/62] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 21/62] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 22/62] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 23/62] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 24/62] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 25/62] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 26/62] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/62] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 28/62] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 29/62] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 30/62] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 31/62] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 32/62] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 33/62] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 34/62] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 35/62] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 36/62] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 37/62] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 38/62] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 39/62] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 40/62] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 41/62] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/62] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/62] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/62] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/62] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/62] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/62] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/62] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/62] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/62] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/62] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/62] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/62] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/62] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/62] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/62] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/62] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/62] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/62] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/62] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/62] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/62] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- [PATCH 01/63] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/63] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/63] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/63] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/63] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/63] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/63] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/63] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/63] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/63] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/63] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/63] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/63] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/63] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/63] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/63] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/63] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 18/63] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 19/63] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 20/63] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 21/63] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 22/63] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 23/63] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 24/63] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 25/63] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 26/63] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 27/63] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 28/63] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 29/63] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 30/63] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 31/63] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 32/63] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 33/63] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 34/63] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 35/63] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 36/63] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 37/63] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 38/63] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 39/63] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 40/63] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 41/63] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 42/63] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 43/63] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 44/63] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 45/63] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 46/63] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/63] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/63] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/63] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/63] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/63] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/63] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/63] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/63] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/63] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/63] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/63] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/63] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/63] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/63] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/63] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/63] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- [PATCH 63/63] drm/i915: Move gt_revoke() slightly
Maarten Lankhorst
- [PATCH 01/23] drm/i915/gem: Remove incorrect early dbg print
Chris Wilson
- [PATCH 02/23] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 03/23] drm/i915/selftests: Small tweak to put the termination conditions together
Chris Wilson
- [PATCH 04/23] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 05/23] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 06/23] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 07/23] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 08/23] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 09/23] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 10/23] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 11/23] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 12/23] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 13/23] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 14/23] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 15/23] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 16/23] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 17/23] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 18/23] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 19/23] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 20/23] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 21/23] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 22/23] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 23/23] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- ✗ Fi.CI.BUILD: failure for series starting with [01/12] drm/i915/gem: Drop free_work for GEM contexts
Patchwork
- ✗ Fi.CI.BUILD: failure for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.BUILD: failure for series starting with [01/63] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.BAT: failure for series starting with [01/23] drm/i915/gem: Remove incorrect early dbg print
Patchwork
- [PATCH i-g-t 1/2] i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH i-g-t 2/2] HAX: Run OA buffer wraparound
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✗ Fi.CI.BAT: failure for series starting with [1/2] i915/perf: Test OA buffer wrap around
Patchwork
- ✗ Fi.CI.BAT: failure for Drop reason check from report
Patchwork
- [PATCH i-g-t] i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✓ Fi.CI.BAT: success for i915/perf: Test OA buffer wrap around
Patchwork
- ✓ Fi.CI.BAT: success for Drop reason check from report (rev2)
Patchwork
- [PATCH i-g-t 1/2] igt fork problem with sanitize
Zbigniew Kempczyński
- [PATCH i-g-t 2/2] HAX: try asan and signals
Zbigniew Kempczyński
- ✓ Fi.CI.IGT: success for i915/perf: Test OA buffer wrap around
Patchwork
- ✗ Fi.CI.IGT: failure for Drop reason check from report (rev2)
Patchwork
- [PATCH] drm/i915/display: Warn about types of backlight not handled
José Roberto de Souza
- [PATCH 02/23] drm/i915/selftests: Improve granularity for mocs reset checks
Mika Kuoppala
- [PATCH 03/23] drm/i915/selftests: Small tweak to put the termination conditions together
Mika Kuoppala
- [PATCH i-g-t] i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- [PATCH 001/162] drm/i915/selftest: also consider non-contiguous objects
Matthew Auld
- [PATCH 002/162] drm/i915/selftest: assert we get 2M GTT pages
Matthew Auld
- [PATCH 003/162] drm/i915/selftest: handle local-memory in perf_memcpy
Matthew Auld
- [PATCH 004/162] drm/i915/gt: Move move context layout registers and offsets to lrc_reg.h
Matthew Auld
- [PATCH 005/162] drm/i915/gt: Rename lrc.c to execlists_submission.c
Matthew Auld
- [PATCH 006/162] drm/i915: split gen8+ flush and bb_start emission functions to their own file
Matthew Auld
- [PATCH 007/162] drm/i915: split wa_bb code to its own file
Matthew Auld
- [PATCH 008/162] HAX drm/i915: Work around the selftest timeline lock splat workaround
Matthew Auld
- [PATCH 009/162] drm/i915: Introduce drm_i915_lock_isolated
Matthew Auld
- [PATCH 010/162] drm/i915: Lock hwsp objects isolated for pinning at create time
Matthew Auld
- [PATCH 011/162] drm/i915: Pin timeline map after first timeline pin, v5.
Matthew Auld
- [PATCH 012/162] drm/i915: Move cmd parser pinning to execbuffer
Matthew Auld
- [PATCH 013/162] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Matthew Auld
- [PATCH 014/162] drm/i915: Ensure we hold the object mutex in pin correctly v2
Matthew Auld
- [PATCH 015/162] drm/i915: Add gem object locking to madvise.
Matthew Auld
- [PATCH 016/162] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Matthew Auld
- [PATCH 017/162] drm/i915: Rework struct phys attachment handling
Matthew Auld
- [PATCH 018/162] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Matthew Auld
- [PATCH 019/162] drm/i915: make lockdep slightly happier about execbuf.
Matthew Auld
- [PATCH 020/162] drm/i915: Disable userptr pread/pwrite support.
Matthew Auld
- [PATCH 021/162] drm/i915: No longer allow exporting userptr through dma-buf
Matthew Auld
- [PATCH 022/162] drm/i915: Reject more ioctls for userptr
Matthew Auld
- [PATCH 023/162] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Matthew Auld
- [PATCH 024/162] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Matthew Auld
- [PATCH 025/162] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Matthew Auld
- [PATCH 026/162] drm/i915: Flatten obj->mm.lock
Matthew Auld
- [PATCH 027/162] drm/i915: Populate logical context during first pin.
Matthew Auld
- [PATCH 028/162] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Matthew Auld
- [PATCH 029/162] drm/i915: Handle ww locking in init_status_page
Matthew Auld
- [PATCH 030/162] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Matthew Auld
- [PATCH 031/162] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Matthew Auld
- [PATCH 032/162] drm/i915: Add object locking to vm_fault_cpu
Matthew Auld
- [PATCH 033/162] drm/i915: Move pinning to inside engine_wa_list_verify()
Matthew Auld
- [PATCH 034/162] drm/i915: Take reservation lock around i915_vma_pin.
Matthew Auld
- [PATCH 035/162] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Matthew Auld
- [PATCH 036/162] drm/i915: Make __engine_unpark() compatible with ww locking v2
Matthew Auld
- [PATCH 037/162] drm/i915: Take obj lock around set_domain ioctl
Matthew Auld
- [PATCH 038/162] drm/i915: Defer pin calls in buffer pool until first use by caller.
Matthew Auld
- [PATCH 039/162] drm/i915: Fix pread/pwrite to work with new locking rules.
Matthew Auld
- [PATCH 040/162] drm/i915: Fix workarounds selftest, part 1
Matthew Auld
- [PATCH 041/162] drm/i915: Prepare for obj->mm.lock removal
Matthew Auld
- [PATCH 042/162] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Matthew Auld
- [PATCH 043/162] drm/i915: Add ww locking around vm_access()
Matthew Auld
- [PATCH 044/162] drm/i915: Increase ww locking for perf.
Matthew Auld
- [PATCH 045/162] drm/i915: Lock ww in ucode objects correctly
Matthew Auld
- [PATCH 046/162] drm/i915: Add ww locking to dma-buf ops.
Matthew Auld
- [PATCH 047/162] drm/i915: Add missing ww lock in intel_dsb_prepare.
Matthew Auld
- [PATCH 048/162] drm/i915: Fix ww locking in shmem_create_from_object
Matthew Auld
- [PATCH 049/162] drm/i915: Use a single page table lock for each gtt.
Matthew Auld
- [PATCH 050/162] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 051/162] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Matthew Auld
- [PATCH 052/162] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 053/162] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 054/162] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 055/162] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 056/162] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 057/162] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 058/162] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 059/162] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Matthew Auld
- [PATCH 060/162] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Matthew Auld
- [PATCH 061/162] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Matthew Auld
- [PATCH 062/162] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Matthew Auld
- [PATCH 063/162] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Matthew Auld
- [PATCH 064/162] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Matthew Auld
- [PATCH 065/162] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Matthew Auld
- [PATCH 066/162] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Matthew Auld
- [PATCH 067/162] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Matthew Auld
- [PATCH 068/162] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Matthew Auld
- [PATCH 069/162] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Matthew Auld
- [PATCH 070/162] drm/i915: Finally remove obj->mm.lock.
Matthew Auld
- [PATCH 071/162] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Matthew Auld
- [PATCH 072/162] drm/i915: Avoid some false positives in assert_object_held()
Matthew Auld
- [PATCH 073/162] drm/i915: Reference contending lock objects
Matthew Auld
- [PATCH 074/162] drm/i915: Break out dma_resv ww locking utilities to separate files
Matthew Auld
- [PATCH 075/162] drm/i915: Introduce a for_i915_gem_ww(){}
Matthew Auld
- [PATCH 076/162] drm/i915: Untangle the vma pages_mutex
Matthew Auld
- [PATCH 077/162] drm/i915/fbdev: Use lmem physical addresses for fb_mmap() on discrete
Matthew Auld
- [PATCH 078/162] drm/i915: Return error value when bo not in LMEM for discrete
Matthew Auld
- [PATCH 079/162] drm/i915/dmabuf: Disallow LMEM objects from dma-buf
Matthew Auld
- [PATCH 080/162] drm/i915/lmem: Fail driver init if LMEM training failed
Matthew Auld
- [PATCH 081/162] HAX drm/i915/lmem: support CPU relocations
Matthew Auld
- [PATCH 082/162] HAX drm/i915/lmem: support pread and pwrite
Matthew Auld
- [PATCH 083/162] drm/i915: Update the helper to set correct mapping
Matthew Auld
- [PATCH 084/162] drm/i915: introduce kernel blitter_context
Matthew Auld
- [PATCH 085/162] drm/i915/region: support basic eviction
Matthew Auld
- [PATCH 086/162] drm/i915: Add blit functions that can be called from within a WW transaction
Matthew Auld
- [PATCH 087/162] drm/i915: Delay publishing objects on the eviction lists
Matthew Auld
- [PATCH 088/162] drm/i915: support basic object migration
Matthew Auld
- [PATCH 089/162] drm/i915/dg1: Fix occasional migration error
Matthew Auld
- [PATCH 090/162] drm/i915/query: Expose memory regions through the query uAPI
Matthew Auld
- [PATCH 091/162] drm/i915: Store gt in memory region
Matthew Auld
- [PATCH 092/162] drm/i915/uapi: introduce drm_i915_gem_create_ext
Matthew Auld
- [PATCH 093/162] drm/i915/lmem: allocate cmd ring in lmem
Matthew Auld
- [PATCH 094/162] drm/i915/dg1: Do not check r->sgt.pfn for NULL
Matthew Auld
- [PATCH 095/162] drm/i915/dg1: Introduce dmabuf mmap to LMEM
Matthew Auld
- [PATCH 096/162] drm/i915: setup the LMEM region
Matthew Auld
- [PATCH 097/162] drm/i915: Distinction of memory regions
Matthew Auld
- [PATCH 098/162] drm/i915/gtt: map the PD up front
Matthew Auld
- [PATCH 099/162] drm/i915/gtt/dgfx: place the PD in LMEM
Matthew Auld
- [PATCH 100/162] drm/i915/gtt: make flushing conditional
Matthew Auld
- [PATCH 101/162] drm/i915/gtt/dg1: add PTE_LM plumbing for PPGTT
Matthew Auld
- [PATCH 102/162] drm/i915/gtt/dg1: add PTE_LM plumbing for GGTT
Matthew Auld
- [PATCH 103/162] drm/i915: allocate context from LMEM
Matthew Auld
- [PATCH 104/162] drm/i915: move engine scratch to LMEM
Matthew Auld
- [PATCH 105/162] drm/i915: Provide a way to disable PCIe relaxed write ordering
Matthew Auld
- [PATCH 106/162] drm/i915: i915 returns -EBUSY on thread contention
Matthew Auld
- [PATCH 107/162] drm/i915: setup GPU device lmem region
Matthew Auld
- [PATCH 108/162] drm/i915: Fix object page offset within a region
Matthew Auld
- [PATCH 109/162] drm/i915: add i915_gem_object_is_devmem() function
Matthew Auld
- [PATCH 110/162] drm/i915: finish memory region support for stolen objects.
Matthew Auld
- [PATCH 111/162] drm/i915/lmem: support optional CPU clearing for special internal use
Matthew Auld
- [PATCH 112/162] drm/i915/guc: put all guc objects in lmem when available
Matthew Auld
- [PATCH 113/162] drm/i915: Create stolen memory region from local memory
Matthew Auld
- [PATCH 114/162] drm/i915/lmem: Bypass aperture when lmem is available
Matthew Auld
- [PATCH 115/162] drm/i915/lmem: reset the lmem buffer created by fbdev
Matthew Auld
- [PATCH 116/162] drm/i915/dsb: Enable lmem for dsb
Matthew Auld
- [PATCH 117/162] drm/i915: Reintroduce mem->reserved
Matthew Auld
- [PATCH 118/162] drm/i915/dg1: Reserve first 1MB of local memory
Matthew Auld
- [PATCH 119/162] drm/i915/dg1: Read OPROM via SPI controller
Matthew Auld
- [PATCH 120/162] drm/i915/oprom: Basic sanitization
Matthew Auld
- [PATCH 121/162] drm/i915: WA for zero memory channel
Matthew Auld
- [PATCH 122/162] drm/i915/dg1: Compute MEM Bandwidth using MCHBAR
Matthew Auld
- [PATCH 123/162] drm/i915/dg1: Double memory bandwidth available
Matthew Auld
- [PATCH 124/162] drm/i915/lmem: allocate HWSP in lmem
Matthew Auld
- [PATCH 125/162] drm/i915/lmem: Limit block size to 4G
Matthew Auld
- [PATCH 126/162] drm/i915/gem: Update shmem available memory
Matthew Auld
- [PATCH 127/162] drm/i915: Allow non-uniform subslices in gen12+
Matthew Auld
- [PATCH 128/162] drm/i915/dg1: intel_memory_region_evict() changes for eviction
Matthew Auld
- [PATCH 129/162] drm/i915/dg1: i915_gem_object_memcpy(..) infrastructure
Matthew Auld
- [PATCH 130/162] drm/i915/dg1: Eviction logic
Matthew Auld
- [PATCH 131/162] drm/i915/dg1: Add enable_eviction modparam
Matthew Auld
- [PATCH 132/162] drm/i915/dg1: Add lmem_size modparam
Matthew Auld
- [PATCH 133/162] drm/i915/dg1: Track swap in/out stats via debugfs
Matthew Auld
- [PATCH 134/162] drm/i915/dg1: Measure swap in/out timing stats
Matthew Auld
- [PATCH 135/162] drm/i915: define intel_partial_pages_for_sg_table
Matthew Auld
- [PATCH 136/162] drm/i915: create and destroy dummy vma
Matthew Auld
- [PATCH 137/162] drm/i915: blt copy between objs using pre-created vma windows
Matthew Auld
- [PATCH 138/162] drm/i915/dg1: Eliminate eviction mutex
Matthew Auld
- [PATCH 139/162] drm/i915/dg1: Keep engine awake across whole blit
Matthew Auld
- [PATCH 140/162] drm/i915: window_blt_copy is used for swapin and swapout
Matthew Auld
- [PATCH 141/162] drm/i915: Lmem eviction statistics by category
Matthew Auld
- [PATCH 142/162] drm/i915/gem/selftest: test and measure window based blt cpy
Matthew Auld
- [PATCH 143/162] drm/i915: suspend/resume eviction
Matthew Auld
- [PATCH 144/162] drm/i915: Reset blitter context when unpark engine
Matthew Auld
- [PATCH 145/162] drm/i915/dg1: Add dedicated context for blitter eviction
Matthew Auld
- [PATCH 146/162] drm/i915/pm: suspend and restore ppgtt mapping
Matthew Auld
- [PATCH 147/162] drm/i915/gt: Allocate default ctx objects in SMEM
Matthew Auld
- [PATCH 148/162] drm/i915: suspend/resume enable blitter eviction
Matthew Auld
- [PATCH 149/162] drm/i915: suspend/resume handling of perma-pinned objects
Matthew Auld
- [PATCH 150/162] drm/i915: need consider system BO snoop for dgfx
Matthew Auld
- [PATCH 151/162] drm/i915: move eviction to prepare hook
Matthew Auld
- [PATCH 152/162] drm/i915: Perform execbuffer object locking as a separate step
Matthew Auld
- [PATCH 153/162] drm/i915: Implement eviction locking v2
Matthew Auld
- [PATCH 154/162] drm/i915: Support ww eviction
Matthew Auld
- [PATCH 155/162] drm/i915: Use a ww transaction in the fault handler
Matthew Auld
- [PATCH 156/162] drm/i915: Use a ww transaction in i915_gem_object_pin_map_unlocked()
Matthew Auld
- [PATCH 157/162] drm/i915: Improve accuracy of eviction stats
Matthew Auld
- [PATCH 158/162] drm/i915: Support ww locks in suspend/resume
Matthew Auld
- [PATCH 159/162] drm/i915/dg1: Fix mapping type for default state object
Matthew Auld
- [PATCH 160/162] drm/i915/dg1: Fix GPU hang due to shmemfs page drop
Matthew Auld
- [PATCH 161/162] drm/i915/dg1: allow pci to auto probe
Matthew Auld
- [PATCH 162/162] drm/i915: drop fake lmem
Matthew Auld
- [PATCH i-g-t] i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✗ Fi.CI.BAT: failure for series starting with [1/2] igt fork problem with sanitize
Patchwork
- ✓ Fi.CI.BAT: success for i915/perf: Test OA buffer wrap around (rev3)
Patchwork
- ✗ Fi.CI.BAT: failure for Drop reason check from report
Patchwork
- ✓ Fi.CI.BAT: success for drm/i915/display: Warn about types of backlight not handled (rev2)
Patchwork
- ✗ Fi.CI.BAT: failure for series starting with [001/162] drm/i915/selftest: also consider non-contiguous objects
Patchwork
- ✓ Fi.CI.BAT: success for Drop reason check from report
Patchwork
- ✓ Fi.CI.IGT: success for i915/perf: Test OA buffer wrap around (rev3)
Patchwork
- [PATCH] drm/i915/dp: Attempt to fix the FEC enable timeout error for Bigjoiner master
Manasi Navare
- ✗ Fi.CI.IGT: failure for drm/i915/display: Warn about types of backlight not handled (rev2)
Patchwork
- ✓ Fi.CI.BAT: success for drm/i915/dp: Attempt to fix the FEC enable timeout error for Bigjoiner master
Patchwork
- ✓ Fi.CI.IGT: success for Drop reason check from report
Patchwork
- ✓ Fi.CI.IGT: success for drm/i915/dp: Attempt to fix the FEC enable timeout error for Bigjoiner master
Patchwork
- [PATCH 01/66] drm/i915/gem: Remove incorrect early dbg print
Chris Wilson
- [PATCH 02/66] drm/i915/selftests: Improve granularity for mocs reset checks
Chris Wilson
- [PATCH 03/66] drm/i915/selftests: Small tweak to put the termination conditions together
Chris Wilson
- [PATCH 04/66] drm/i915/gt: Plug IPS into intel_rps_set
Chris Wilson
- [PATCH 05/66] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 06/66] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 07/66] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 08/66] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 09/66] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 10/66] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 11/66] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 12/66] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 13/66] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 14/66] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 15/66] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 16/66] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 17/66] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 18/66] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 19/66] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 20/66] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 21/66] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 22/66] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 23/66] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 24/66] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 25/66] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 26/66] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 27/66] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 28/66] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 29/66] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 30/66] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 31/66] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 32/66] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 33/66] drm/i915: Disable atomics in L3 for gen9
Chris Wilson
- [PATCH 34/66] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 35/66] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 36/66] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 37/66] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 38/66] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 39/66] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 40/66] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 41/66] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 42/66] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 43/66] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 44/66] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 45/66] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 46/66] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 47/66] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 48/66] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 49/66] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 50/66] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 51/66] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 52/66] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 53/66] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 54/66] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 55/66] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 56/66] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 57/66] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 58/66] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 59/66] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 60/66] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 61/66] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 62/66] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 63/66] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 64/66] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 65/66] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 66/66] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- ✗ Fi.CI.BUILD: failure for series starting with [01/66] drm/i915/gem: Remove incorrect early dbg print
Patchwork
- [PATCH 01/63] drm/i915/gt: Plug IPS into intel_rps_set
Chris Wilson
- [PATCH 02/63] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 03/63] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 04/63] drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission
Chris Wilson
- [PATCH 05/63] drm/i915/gt: Track signaled breadcrumbs outside of the breadcrumb spinlock
Chris Wilson
- [PATCH 06/63] drm/i915/gt: Don't cancel the interrupt shadow too early
Chris Wilson
- [PATCH 07/63] drm/i915/gt: Free stale request on destroying the virtual engine
Chris Wilson
- [PATCH 08/63] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 09/63] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 10/63] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 11/63] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 12/63] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 13/63] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 14/63] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 15/63] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 16/63] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 17/63] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 18/63] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 19/63] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 20/63] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 21/63] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 22/63] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 23/63] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 24/63] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 25/63] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 26/63] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 27/63] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 28/63] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 29/63] drm/i915/gt: Limit VFE threads based on GT
Chris Wilson
- [PATCH 30/63] drm/i915: Disable atomics in L3 for gen9
Chris Wilson
- [PATCH 31/63] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 32/63] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 33/63] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 34/63] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 35/63] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 36/63] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 37/63] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 38/63] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 39/63] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 40/63] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 41/63] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 42/63] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 43/63] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 44/63] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 45/63] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 46/63] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 47/63] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 48/63] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 49/63] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 50/63] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 51/63] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 52/63] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 53/63] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 54/63] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 55/63] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 56/63] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 57/63] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 58/63] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 59/63] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 60/63] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 61/63] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 62/63] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 63/63] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/63] drm/i915/gt: Plug IPS into intel_rps_set
Patchwork
- [PATCH i-g-t v2 1/2] igt fork problem with sanitize
Zbigniew Kempczyński
- [PATCH i-g-t v2 2/2] HAX: try asan and signals
Zbigniew Kempczyński
- ✗ Fi.CI.BUILD: failure for series starting with [v2,1/2] igt fork problem with sanitize
Patchwork
- [PATCH 01/62] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 02/62] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 03/62] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 04/62] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 05/62] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 06/62] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 07/62] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 08/62] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 09/62] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 10/62] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 11/62] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 12/62] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 13/62] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 14/62] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 15/62] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 16/62] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 17/62] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 01/63] drm/i915: Do not share hwsp across contexts any more, v6
Maarten Lankhorst
- [PATCH 18/62] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 02/63] drm/i915: Pin timeline map after first timeline pin, v3.
Maarten Lankhorst
- [PATCH 19/62] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 03/63] drm/i915: Move cmd parser pinning to execbuffer
Maarten Lankhorst
- [PATCH 20/62] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 04/63] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Maarten Lankhorst
- [PATCH 21/62] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 05/63] drm/i915: Ensure we hold the object mutex in pin correctly.
Maarten Lankhorst
- [PATCH 22/62] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 06/63] drm/i915: Add gem object locking to madvise.
Maarten Lankhorst
- [PATCH 23/62] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 07/63] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Maarten Lankhorst
- [PATCH 24/62] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 08/63] drm/i915: Rework struct phys attachment handling
Maarten Lankhorst
- [PATCH 25/62] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 09/63] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Maarten Lankhorst
- [PATCH 26/62] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 10/63] drm/i915: make lockdep slightly happier about execbuf.
Maarten Lankhorst
- [PATCH 27/62] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 11/63] drm/i915: Disable userptr pread/pwrite support.
Maarten Lankhorst
- [PATCH 28/62] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 12/63] drm/i915: No longer allow exporting userptr through dma-buf
Maarten Lankhorst
- [PATCH 29/62] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 13/63] drm/i915: Reject more ioctls for userptr
Maarten Lankhorst
- [PATCH 30/62] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 14/63] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Maarten Lankhorst
- [PATCH 31/62] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 15/63] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Maarten Lankhorst
- [PATCH 32/62] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 16/63] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Maarten Lankhorst
- [PATCH 33/62] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 17/63] drm/i915: Flatten obj->mm.lock
Maarten Lankhorst
- [PATCH 34/62] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 18/63] drm/i915: Populate logical context during first pin.
Maarten Lankhorst
- [PATCH 35/62] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 19/63] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Maarten Lankhorst
- [PATCH 36/62] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 20/63] drm/i915: Handle ww locking in init_status_page
Maarten Lankhorst
- [PATCH 37/62] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 21/63] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Maarten Lankhorst
- [PATCH 38/62] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 22/63] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Maarten Lankhorst
- [PATCH 39/62] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 23/63] drm/i915: Add object locking to vm_fault_cpu
Maarten Lankhorst
- [PATCH 40/62] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 24/63] drm/i915: Move pinning to inside engine_wa_list_verify()
Maarten Lankhorst
- [PATCH 41/62] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 25/63] drm/i915: Take reservation lock around i915_vma_pin.
Maarten Lankhorst
- [PATCH 42/62] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 26/63] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Maarten Lankhorst
- [PATCH 43/62] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 27/63] drm/i915: Make __engine_unpark() compatible with ww locking.
Maarten Lankhorst
- [PATCH 44/62] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 28/63] drm/i915: Take obj lock around set_domain ioctl
Maarten Lankhorst
- [PATCH 45/62] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 29/63] drm/i915: Defer pin calls in buffer pool until first use by caller.
Maarten Lankhorst
- [PATCH 46/62] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 30/63] drm/i915: Fix pread/pwrite to work with new locking rules.
Maarten Lankhorst
- [PATCH 47/62] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 31/63] drm/i915: Fix workarounds selftest, part 1
Maarten Lankhorst
- [PATCH 48/62] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 32/63] drm/i915: Prepare for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 49/62] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 33/63] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Maarten Lankhorst
- [PATCH 50/62] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 34/63] drm/i915: Add ww locking around vm_access()
Maarten Lankhorst
- [PATCH 51/62] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 35/63] drm/i915: Increase ww locking for perf.
Maarten Lankhorst
- [PATCH 52/62] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 36/63] drm/i915: Lock ww in ucode objects correctly
Maarten Lankhorst
- [PATCH 53/62] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 37/63] drm/i915: Add ww locking to dma-buf ops.
Maarten Lankhorst
- [PATCH 54/62] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 38/63] drm/i915: Add missing ww lock in intel_dsb_prepare.
Maarten Lankhorst
- [PATCH 55/62] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 39/63] drm/i915: Fix ww locking in shmem_create_from_object
Maarten Lankhorst
- [PATCH 56/62] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 40/63] drm/i915: Use a single page table lock for each gtt.
Maarten Lankhorst
- [PATCH 57/62] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 41/63] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 58/62] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 42/63] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 59/62] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 43/63] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 60/62] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 44/63] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 61/62] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 45/63] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 62/62] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- [PATCH 46/63] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 47/63] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 48/63] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 49/63] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Maarten Lankhorst
- [PATCH 50/63] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 51/63] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 52/63] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 53/63] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 54/63] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 55/63] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 56/63] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 57/63] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 58/63] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 59/63] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 60/63] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Maarten Lankhorst
- [PATCH 61/63] drm/i915: Finally remove obj->mm.lock.
Maarten Lankhorst
- [PATCH 62/63] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Maarten Lankhorst
- [PATCH 63/63] drm/i915: Move gt_revoke() slightly
Maarten Lankhorst
- [PATCH 001/163] drm/i915/selftest: also consider non-contiguous objects
Matthew Auld
- [PATCH 002/163] drm/i915/selftest: assert we get 2M GTT pages
Matthew Auld
- [PATCH 003/163] drm/i915/selftest: handle local-memory in perf_memcpy
Matthew Auld
- [PATCH 004/163] drm/i915/gt: Move move context layout registers and offsets to lrc_reg.h
Matthew Auld
- [PATCH 005/163] drm/i915/gt: Rename lrc.c to execlists_submission.c
Matthew Auld
- [PATCH 006/163] drm/i915: split gen8+ flush and bb_start emission functions to their own file
Matthew Auld
- [PATCH 007/163] drm/i915: split wa_bb code to its own file
Matthew Auld
- [PATCH 008/163] HAX drm/i915: Work around the selftest timeline lock splat workaround
Matthew Auld
- [PATCH 009/163] drm/i915: Introduce drm_i915_lock_isolated
Matthew Auld
- [PATCH 010/163] drm/i915: Lock hwsp objects isolated for pinning at create time
Matthew Auld
- [PATCH 011/163] drm/i915: Pin timeline map after first timeline pin, v5.
Matthew Auld
- [PATCH 012/163] drm/i915: Move cmd parser pinning to execbuffer
Matthew Auld
- [PATCH 013/163] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Matthew Auld
- [PATCH 014/163] drm/i915: Ensure we hold the object mutex in pin correctly v2
Matthew Auld
- [PATCH 015/163] drm/i915: Add gem object locking to madvise.
Matthew Auld
- [PATCH 016/163] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Matthew Auld
- [PATCH 017/163] drm/i915: Rework struct phys attachment handling
Matthew Auld
- [PATCH 018/163] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Matthew Auld
- [PATCH 019/163] drm/i915: make lockdep slightly happier about execbuf.
Matthew Auld
- [PATCH 020/163] drm/i915: Disable userptr pread/pwrite support.
Matthew Auld
- [PATCH 021/163] drm/i915: No longer allow exporting userptr through dma-buf
Matthew Auld
- [PATCH 022/163] drm/i915: Reject more ioctls for userptr
Matthew Auld
- [PATCH 023/163] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Matthew Auld
- [PATCH 024/163] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Matthew Auld
- [PATCH 025/163] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Matthew Auld
- [PATCH 026/163] drm/i915: Flatten obj->mm.lock
Matthew Auld
- [PATCH 027/163] drm/i915: Populate logical context during first pin.
Matthew Auld
- [PATCH 028/163] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Matthew Auld
- [PATCH 029/163] drm/i915: Handle ww locking in init_status_page
Matthew Auld
- [PATCH 030/163] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Matthew Auld
- [PATCH 031/163] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Matthew Auld
- [PATCH 032/163] drm/i915: Add object locking to vm_fault_cpu
Matthew Auld
- [PATCH 033/163] drm/i915: Move pinning to inside engine_wa_list_verify()
Matthew Auld
- [PATCH 034/163] drm/i915: Take reservation lock around i915_vma_pin.
Matthew Auld
- [PATCH 035/163] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Matthew Auld
- [PATCH 036/163] drm/i915: Make __engine_unpark() compatible with ww locking v2
Matthew Auld
- [PATCH 037/163] drm/i915: Take obj lock around set_domain ioctl
Matthew Auld
- [PATCH 038/163] drm/i915: Defer pin calls in buffer pool until first use by caller.
Matthew Auld
- [PATCH 039/163] drm/i915: Fix pread/pwrite to work with new locking rules.
Matthew Auld
- [PATCH 040/163] drm/i915: Fix workarounds selftest, part 1
Matthew Auld
- [PATCH 041/163] drm/i915: Prepare for obj->mm.lock removal
Matthew Auld
- [PATCH 042/163] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Matthew Auld
- [PATCH 043/163] drm/i915: Add ww locking around vm_access()
Matthew Auld
- [PATCH 044/163] drm/i915: Increase ww locking for perf.
Matthew Auld
- [PATCH 045/163] drm/i915: Lock ww in ucode objects correctly
Matthew Auld
- [PATCH 046/163] drm/i915: Add ww locking to dma-buf ops.
Matthew Auld
- [PATCH 047/163] drm/i915: Add missing ww lock in intel_dsb_prepare.
Matthew Auld
- [PATCH 048/163] drm/i915: Fix ww locking in shmem_create_from_object
Matthew Auld
- [PATCH 049/163] drm/i915: Use a single page table lock for each gtt.
Matthew Auld
- [PATCH 050/163] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 051/163] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Matthew Auld
- [PATCH 052/163] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 053/163] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 054/163] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 055/163] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 056/163] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 057/163] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 058/163] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 059/163] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Matthew Auld
- [PATCH 060/163] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Matthew Auld
- [PATCH 061/163] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Matthew Auld
- [PATCH 062/163] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Matthew Auld
- [PATCH 063/163] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Matthew Auld
- [PATCH 064/163] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Matthew Auld
- [PATCH 065/163] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Matthew Auld
- [PATCH 066/163] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Matthew Auld
- [PATCH 067/163] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Matthew Auld
- [PATCH 068/163] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Matthew Auld
- [PATCH 069/163] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Matthew Auld
- [PATCH 070/163] drm/i915: Finally remove obj->mm.lock.
Matthew Auld
- [PATCH 071/163] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Matthew Auld
- [PATCH 072/163] drm/i915: Avoid some false positives in assert_object_held()
Matthew Auld
- [PATCH 073/163] drm/i915: Reference contending lock objects
Matthew Auld
- [PATCH 074/163] drm/i915: Break out dma_resv ww locking utilities to separate files
Matthew Auld
- [PATCH 075/163] drm/i915: Introduce a for_i915_gem_ww(){}
Matthew Auld
- [PATCH 076/163] drm/i915: Untangle the vma pages_mutex
Matthew Auld
- [PATCH 077/163] drm/i915/fbdev: Use lmem physical addresses for fb_mmap() on discrete
Matthew Auld
- [PATCH 078/163] drm/i915: Return error value when bo not in LMEM for discrete
Matthew Auld
- [PATCH 079/163] drm/i915/dmabuf: Disallow LMEM objects from dma-buf
Matthew Auld
- [PATCH 080/163] drm/i915/lmem: Fail driver init if LMEM training failed
Matthew Auld
- [PATCH 081/163] HAX drm/i915/lmem: support CPU relocations
Matthew Auld
- [PATCH 082/163] HAX drm/i915/lmem: support pread and pwrite
Matthew Auld
- [PATCH 083/163] drm/i915: Update the helper to set correct mapping
Matthew Auld
- [PATCH 084/163] drm/i915: introduce kernel blitter_context
Matthew Auld
- [PATCH 085/163] drm/i915/region: support basic eviction
Matthew Auld
- [PATCH 086/163] drm/i915: Add blit functions that can be called from within a WW transaction
Matthew Auld
- [PATCH 087/163] drm/i915: Delay publishing objects on the eviction lists
Matthew Auld
- [PATCH 088/163] drm/i915: support basic object migration
Matthew Auld
- [PATCH 089/163] drm/i915/dg1: Fix occasional migration error
Matthew Auld
- [PATCH 090/163] drm/i915/query: Expose memory regions through the query uAPI
Matthew Auld
- [PATCH 091/163] drm/i915: Store gt in memory region
Matthew Auld
- [PATCH 092/163] drm/i915/uapi: introduce drm_i915_gem_create_ext
Matthew Auld
- [PATCH 093/163] drm/i915/lmem: allocate cmd ring in lmem
Matthew Auld
- [PATCH 094/163] drm/i915/dg1: Do not check r->sgt.pfn for NULL
Matthew Auld
- [PATCH 095/163] drm/i915/dg1: Introduce dmabuf mmap to LMEM
Matthew Auld
- [PATCH 096/163] drm/i915: setup the LMEM region
Matthew Auld
- [PATCH 097/163] drm/i915: Distinction of memory regions
Matthew Auld
- [PATCH 098/163] drm/i915/gtt: map the PD up front
Matthew Auld
- [PATCH 099/163] drm/i915/gtt/dgfx: place the PD in LMEM
Matthew Auld
- [PATCH 100/163] drm/i915/gtt: make flushing conditional
Matthew Auld
- [PATCH 101/163] drm/i915/gtt/dg1: add PTE_LM plumbing for PPGTT
Matthew Auld
- [PATCH 102/163] drm/i915/gtt/dg1: add PTE_LM plumbing for GGTT
Matthew Auld
- [PATCH 103/163] drm/i915: allocate context from LMEM
Matthew Auld
- [PATCH 104/163] drm/i915: move engine scratch to LMEM
Matthew Auld
- [PATCH 105/163] drm/i915: Provide a way to disable PCIe relaxed write ordering
Matthew Auld
- [PATCH 106/163] drm/i915: i915 returns -EBUSY on thread contention
Matthew Auld
- [PATCH 107/163] drm/i915: setup GPU device lmem region
Matthew Auld
- [PATCH 108/163] drm/i915: Fix object page offset within a region
Matthew Auld
- [PATCH 109/163] drm/i915: add i915_gem_object_is_devmem() function
Matthew Auld
- [PATCH 110/163] drm/i915: finish memory region support for stolen objects.
Matthew Auld
- [PATCH 111/163] drm/i915/lmem: support optional CPU clearing for special internal use
Matthew Auld
- [PATCH 112/163] drm/i915/guc: put all guc objects in lmem when available
Matthew Auld
- [PATCH 113/163] drm/i915: Create stolen memory region from local memory
Matthew Auld
- [PATCH 114/163] drm/i915/lmem: Bypass aperture when lmem is available
Matthew Auld
- [PATCH 115/163] drm/i915/lmem: reset the lmem buffer created by fbdev
Matthew Auld
- [PATCH 116/163] drm/i915/dsb: Enable lmem for dsb
Matthew Auld
- [PATCH 117/163] drm/i915: Reintroduce mem->reserved
Matthew Auld
- [PATCH 118/163] drm/i915/dg1: Reserve first 1MB of local memory
Matthew Auld
- [PATCH 119/163] drm/i915/dg1: Read OPROM via SPI controller
Matthew Auld
- [PATCH 120/163] drm/i915/oprom: Basic sanitization
Matthew Auld
- [PATCH 121/163] drm/i915: WA for zero memory channel
Matthew Auld
- [PATCH 122/163] drm/i915/dg1: Compute MEM Bandwidth using MCHBAR
Matthew Auld
- [PATCH 123/163] drm/i915/dg1: Double memory bandwidth available
Matthew Auld
- [PATCH 124/163] drm/i915/lmem: allocate HWSP in lmem
Matthew Auld
- [PATCH 125/163] drm/i915/lmem: Limit block size to 4G
Matthew Auld
- [PATCH 126/163] drm/i915/gem: Update shmem available memory
Matthew Auld
- [PATCH 127/163] drm/i915: Allow non-uniform subslices in gen12+
Matthew Auld
- [PATCH 128/163] drm/i915/dg1: intel_memory_region_evict() changes for eviction
Matthew Auld
- [PATCH 129/163] drm/i915/dg1: i915_gem_object_memcpy(..) infrastructure
Matthew Auld
- [PATCH 130/163] drm/i915/dg1: Eviction logic
Matthew Auld
- [PATCH 131/163] drm/i915/dg1: Add enable_eviction modparam
Matthew Auld
- [PATCH 132/163] drm/i915/dg1: Add lmem_size modparam
Matthew Auld
- [PATCH 133/163] drm/i915/dg1: Track swap in/out stats via debugfs
Matthew Auld
- [PATCH 134/163] drm/i915/dg1: Measure swap in/out timing stats
Matthew Auld
- [PATCH 135/163] drm/i915: define intel_partial_pages_for_sg_table
Matthew Auld
- [PATCH 136/163] drm/i915: create and destroy dummy vma
Matthew Auld
- [PATCH 137/163] drm/i915: blt copy between objs using pre-created vma windows
Matthew Auld
- [PATCH 138/163] drm/i915/dg1: Eliminate eviction mutex
Matthew Auld
- [PATCH 139/163] drm/i915/dg1: Keep engine awake across whole blit
Matthew Auld
- [PATCH 140/163] drm/i915: window_blt_copy is used for swapin and swapout
Matthew Auld
- [PATCH 141/163] drm/i915: Lmem eviction statistics by category
Matthew Auld
- [PATCH 142/163] drm/i915/gem/selftest: test and measure window based blt cpy
Matthew Auld
- [PATCH 143/163] drm/i915: suspend/resume eviction
Matthew Auld
- [PATCH 144/163] drm/i915: Reset blitter context when unpark engine
Matthew Auld
- [PATCH 145/163] drm/i915/dg1: Add dedicated context for blitter eviction
Matthew Auld
- [PATCH 146/163] drm/i915/pm: suspend and restore ppgtt mapping
Matthew Auld
- [PATCH 147/163] drm/i915/gt: Allocate default ctx objects in SMEM
Matthew Auld
- [PATCH 148/163] drm/i915: suspend/resume enable blitter eviction
Matthew Auld
- [PATCH 149/163] drm/i915: suspend/resume handling of perma-pinned objects
Matthew Auld
- [PATCH 150/163] drm/i915: need consider system BO snoop for dgfx
Matthew Auld
- [PATCH 151/163] drm/i915: move eviction to prepare hook
Matthew Auld
- [PATCH 152/163] drm/i915: Perform execbuffer object locking as a separate step
Matthew Auld
- [PATCH 153/163] drm/i915: Implement eviction locking v2
Matthew Auld
- [PATCH 154/163] drm/i915: Support ww eviction
Matthew Auld
- [PATCH 155/163] drm/i915: Use a ww transaction in the fault handler
Matthew Auld
- [PATCH 156/163] drm/i915: Use a ww transaction in i915_gem_object_pin_map_unlocked()
Matthew Auld
- [PATCH 157/163] drm/i915: Improve accuracy of eviction stats
Matthew Auld
- [PATCH 158/163] drm/i915: Support ww locks in suspend/resume
Matthew Auld
- [PATCH 159/163] drm/i915/dg1: Fix mapping type for default state object
Matthew Auld
- [PATCH 160/163] drm/i915/dg1: Fix GPU hang due to shmemfs page drop
Matthew Auld
- [PATCH 161/163] drm/i915/dg1: allow pci to auto probe
Matthew Auld
- [PATCH 162/163] drm/i915: drop fake lmem
Matthew Auld
- [PATCH 163/163] HAX
Matthew Auld
- ✓ Fi.CI.BAT: success for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [01/63] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- [PATCH i-g-t 1/9] tests/kms_fence_pin_leak: Remove vestigal 'ctx' local
Chris Wilson
- [PATCH i-g-t 2/9] tests/fbdev: Move existing tests into separate subgroups
Chris Wilson
- [PATCH i-g-t 3/9] tests/fbdev: Map framebuffer in igt_fixture
Chris Wilson
- [PATCH i-g-t 4/9] tests/fbdev: Add tests for read operations on framebuffer
Chris Wilson
- [PATCH i-g-t 5/9] tests/fbdev: Add tests for unaligned reads on framebuffer memory
Chris Wilson
- [PATCH i-g-t 6/9] tests/fbdev: Add tests for write operations on framebuffer
Chris Wilson
- [PATCH i-g-t 7/9] tests/fbdev: Add tests for unaligned writes on framebuffer memory
Chris Wilson
- [PATCH i-g-t 8/9] tests/fbdev: Add tests for accessing framebuffer near EOF
Chris Wilson
- [PATCH i-g-t 9/9] tests/fbdev: Add tests for read/writing with NULL-pointer buffers
Chris Wilson
- ✓ Fi.CI.BAT: success for series starting with [1/9] tests/kms_fence_pin_leak: Remove vestigal 'ctx' local
Patchwork
- ✗ Fi.CI.BAT: failure for series starting with [001/163] drm/i915/selftest: also consider non-contiguous objects
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/62] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/63] drm/i915: Do not share hwsp across contexts any more, v6
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/9] tests/kms_fence_pin_leak: Remove vestigal 'ctx' local
Patchwork
- [PATCH i-g-t] i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✓ Fi.CI.BAT: success for i915/perf: Test OA buffer wrap around (rev4)
Patchwork
- ✓ Fi.CI.BAT: success for Drop reason check from report
Patchwork
- ✗ Fi.CI.IGT: failure for i915/perf: Test OA buffer wrap around (rev4)
Patchwork
- ✗ Fi.CI.IGT: failure for Drop reason check from report
Patchwork
- [PATCH 001/163] drm/i915/selftest: also consider non-contiguous objects
Matthew Auld
- [PATCH 002/163] drm/i915/selftest: assert we get 2M GTT pages
Matthew Auld
- [PATCH 003/163] drm/i915/selftest: handle local-memory in perf_memcpy
Matthew Auld
- [PATCH 004/163] drm/i915/gt: Move move context layout registers and offsets to lrc_reg.h
Matthew Auld
- [PATCH 005/163] drm/i915/gt: Rename lrc.c to execlists_submission.c
Matthew Auld
- [PATCH 006/163] drm/i915: split gen8+ flush and bb_start emission functions to their own file
Matthew Auld
- [PATCH 007/163] drm/i915: split wa_bb code to its own file
Matthew Auld
- [PATCH 008/163] HAX drm/i915: Work around the selftest timeline lock splat workaround
Matthew Auld
- [PATCH 009/163] drm/i915: Introduce drm_i915_lock_isolated
Matthew Auld
- [PATCH 010/163] drm/i915: Lock hwsp objects isolated for pinning at create time
Matthew Auld
- [PATCH 011/163] drm/i915: Pin timeline map after first timeline pin, v5.
Matthew Auld
- [PATCH 012/163] drm/i915: Move cmd parser pinning to execbuffer
Matthew Auld
- [PATCH 013/163] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Matthew Auld
- [PATCH 014/163] drm/i915: Ensure we hold the object mutex in pin correctly v2
Matthew Auld
- [PATCH 015/163] drm/i915: Add gem object locking to madvise.
Matthew Auld
- [PATCH 016/163] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Matthew Auld
- [PATCH 017/163] drm/i915: Rework struct phys attachment handling
Matthew Auld
- [PATCH 018/163] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Matthew Auld
- [PATCH 019/163] drm/i915: make lockdep slightly happier about execbuf.
Matthew Auld
- [PATCH 020/163] drm/i915: Disable userptr pread/pwrite support.
Matthew Auld
- [PATCH 021/163] drm/i915: No longer allow exporting userptr through dma-buf
Matthew Auld
- [PATCH 022/163] drm/i915: Reject more ioctls for userptr
Matthew Auld
- [PATCH 023/163] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Matthew Auld
- [PATCH 024/163] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Matthew Auld
- [PATCH 025/163] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Matthew Auld
- [PATCH 026/163] drm/i915: Flatten obj->mm.lock
Matthew Auld
- [PATCH 027/163] drm/i915: Populate logical context during first pin.
Matthew Auld
- [PATCH 028/163] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Matthew Auld
- [PATCH 029/163] drm/i915: Handle ww locking in init_status_page
Matthew Auld
- [PATCH 030/163] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Matthew Auld
- [PATCH 031/163] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Matthew Auld
- [PATCH 032/163] drm/i915: Add object locking to vm_fault_cpu
Matthew Auld
- [PATCH 033/163] drm/i915: Move pinning to inside engine_wa_list_verify()
Matthew Auld
- [PATCH 034/163] drm/i915: Take reservation lock around i915_vma_pin.
Matthew Auld
- [PATCH 035/163] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Matthew Auld
- [PATCH 036/163] drm/i915: Make __engine_unpark() compatible with ww locking v2
Matthew Auld
- [PATCH 037/163] drm/i915: Take obj lock around set_domain ioctl
Matthew Auld
- [PATCH 038/163] drm/i915: Defer pin calls in buffer pool until first use by caller.
Matthew Auld
- [PATCH 039/163] drm/i915: Fix pread/pwrite to work with new locking rules.
Matthew Auld
- [PATCH 040/163] drm/i915: Fix workarounds selftest, part 1
Matthew Auld
- [PATCH 041/163] drm/i915: Prepare for obj->mm.lock removal
Matthew Auld
- [PATCH 042/163] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Matthew Auld
- [PATCH 043/163] drm/i915: Add ww locking around vm_access()
Matthew Auld
- [PATCH 044/163] drm/i915: Increase ww locking for perf.
Matthew Auld
- [PATCH 045/163] drm/i915: Lock ww in ucode objects correctly
Matthew Auld
- [PATCH 046/163] drm/i915: Add ww locking to dma-buf ops.
Matthew Auld
- [PATCH 047/163] drm/i915: Add missing ww lock in intel_dsb_prepare.
Matthew Auld
- [PATCH 048/163] drm/i915: Fix ww locking in shmem_create_from_object
Matthew Auld
- [PATCH 049/163] drm/i915: Use a single page table lock for each gtt.
Matthew Auld
- [PATCH 050/163] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 051/163] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Matthew Auld
- [PATCH 052/163] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 053/163] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 054/163] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 055/163] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 056/163] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 057/163] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 058/163] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 059/163] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Matthew Auld
- [PATCH 060/163] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Matthew Auld
- [PATCH 061/163] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Matthew Auld
- [PATCH 062/163] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Matthew Auld
- [PATCH 063/163] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Matthew Auld
- [PATCH 064/163] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Matthew Auld
- [PATCH 065/163] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Matthew Auld
- [PATCH 066/163] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Matthew Auld
- [PATCH 067/163] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Matthew Auld
- [PATCH 068/163] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Matthew Auld
- [PATCH 069/163] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Matthew Auld
- [PATCH 070/163] drm/i915: Finally remove obj->mm.lock.
Matthew Auld
- [PATCH 071/163] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Matthew Auld
- [PATCH 072/163] drm/i915: Avoid some false positives in assert_object_held()
Matthew Auld
- [PATCH 073/163] drm/i915: Reference contending lock objects
Matthew Auld
- [PATCH 074/163] drm/i915: Break out dma_resv ww locking utilities to separate files
Matthew Auld
- [PATCH 075/163] drm/i915: Introduce a for_i915_gem_ww(){}
Matthew Auld
- [PATCH 076/163] drm/i915: Untangle the vma pages_mutex
Matthew Auld
- [PATCH 077/163] drm/i915/fbdev: Use lmem physical addresses for fb_mmap() on discrete
Matthew Auld
- [PATCH 078/163] drm/i915: Return error value when bo not in LMEM for discrete
Matthew Auld
- [PATCH 079/163] drm/i915/dmabuf: Disallow LMEM objects from dma-buf
Matthew Auld
- [PATCH 080/163] drm/i915/lmem: Fail driver init if LMEM training failed
Matthew Auld
- [PATCH 081/163] HAX drm/i915/lmem: support CPU relocations
Matthew Auld
- [PATCH 082/163] HAX drm/i915/lmem: support pread and pwrite
Matthew Auld
- [PATCH 083/163] drm/i915: Update the helper to set correct mapping
Matthew Auld
- [PATCH 084/163] drm/i915: introduce kernel blitter_context
Matthew Auld
- [PATCH 085/163] drm/i915/region: support basic eviction
Matthew Auld
- [PATCH 086/163] drm/i915: Add blit functions that can be called from within a WW transaction
Matthew Auld
- [PATCH 087/163] drm/i915: Delay publishing objects on the eviction lists
Matthew Auld
- [PATCH 088/163] drm/i915: support basic object migration
Matthew Auld
- [PATCH 089/163] drm/i915/dg1: Fix occasional migration error
Matthew Auld
- [PATCH 090/163] drm/i915/query: Expose memory regions through the query uAPI
Matthew Auld
- [PATCH 091/163] drm/i915: Store gt in memory region
Matthew Auld
- [PATCH 092/163] drm/i915/uapi: introduce drm_i915_gem_create_ext
Matthew Auld
- [PATCH 093/163] drm/i915/lmem: allocate cmd ring in lmem
Matthew Auld
- [PATCH 094/163] drm/i915/dg1: Do not check r->sgt.pfn for NULL
Matthew Auld
- [PATCH 095/163] drm/i915/dg1: Introduce dmabuf mmap to LMEM
Matthew Auld
- [PATCH 096/163] drm/i915: setup the LMEM region
Matthew Auld
- [PATCH 097/163] drm/i915: Distinction of memory regions
Matthew Auld
- [PATCH 098/163] drm/i915/gtt: map the PD up front
Matthew Auld
- [PATCH 099/163] drm/i915/gtt/dgfx: place the PD in LMEM
Matthew Auld
- [PATCH 100/163] drm/i915/gtt: make flushing conditional
Matthew Auld
- [PATCH 101/163] drm/i915/gtt/dg1: add PTE_LM plumbing for PPGTT
Matthew Auld
- [PATCH 102/163] drm/i915/gtt/dg1: add PTE_LM plumbing for GGTT
Matthew Auld
- [PATCH 103/163] drm/i915: allocate context from LMEM
Matthew Auld
- [PATCH 104/163] drm/i915: move engine scratch to LMEM
Matthew Auld
- [PATCH 105/163] drm/i915: Provide a way to disable PCIe relaxed write ordering
Matthew Auld
- [PATCH 106/163] drm/i915: i915 returns -EBUSY on thread contention
Matthew Auld
- [PATCH 107/163] drm/i915: setup GPU device lmem region
Matthew Auld
- [PATCH 108/163] drm/i915: Fix object page offset within a region
Matthew Auld
- [PATCH 109/163] drm/i915: add i915_gem_object_is_devmem() function
Matthew Auld
- [PATCH 110/163] drm/i915: finish memory region support for stolen objects.
Matthew Auld
- [PATCH 111/163] drm/i915/lmem: support optional CPU clearing for special internal use
Matthew Auld
- [PATCH 112/163] drm/i915/guc: put all guc objects in lmem when available
Matthew Auld
- [PATCH 113/163] drm/i915: Create stolen memory region from local memory
Matthew Auld
- [PATCH 114/163] drm/i915/lmem: Bypass aperture when lmem is available
Matthew Auld
- [PATCH 115/163] drm/i915/lmem: reset the lmem buffer created by fbdev
Matthew Auld
- [PATCH 116/163] drm/i915/dsb: Enable lmem for dsb
Matthew Auld
- [PATCH 117/163] drm/i915: Reintroduce mem->reserved
Matthew Auld
- [PATCH 118/163] drm/i915/dg1: Reserve first 1MB of local memory
Matthew Auld
- [PATCH 119/163] drm/i915/dg1: Read OPROM via SPI controller
Matthew Auld
- [PATCH 120/163] drm/i915/oprom: Basic sanitization
Matthew Auld
- [PATCH 121/163] drm/i915: WA for zero memory channel
Matthew Auld
- [PATCH 122/163] drm/i915/dg1: Compute MEM Bandwidth using MCHBAR
Matthew Auld
- [PATCH 123/163] drm/i915/dg1: Double memory bandwidth available
Matthew Auld
- [PATCH 124/163] drm/i915/lmem: allocate HWSP in lmem
Matthew Auld
- [PATCH 125/163] drm/i915/lmem: Limit block size to 4G
Matthew Auld
- [PATCH 126/163] drm/i915/gem: Update shmem available memory
Matthew Auld
- [PATCH 127/163] drm/i915: Allow non-uniform subslices in gen12+
Matthew Auld
- [PATCH 128/163] drm/i915/dg1: intel_memory_region_evict() changes for eviction
Matthew Auld
- [PATCH 129/163] drm/i915/dg1: i915_gem_object_memcpy(..) infrastructure
Matthew Auld
- [PATCH 130/163] drm/i915/dg1: Eviction logic
Matthew Auld
- [PATCH 131/163] drm/i915/dg1: Add enable_eviction modparam
Matthew Auld
- [PATCH 132/163] drm/i915/dg1: Add lmem_size modparam
Matthew Auld
- [PATCH 133/163] drm/i915/dg1: Track swap in/out stats via debugfs
Matthew Auld
- [PATCH 134/163] drm/i915/dg1: Measure swap in/out timing stats
Matthew Auld
- [PATCH 135/163] drm/i915: define intel_partial_pages_for_sg_table
Matthew Auld
- [PATCH 136/163] drm/i915: create and destroy dummy vma
Matthew Auld
- [PATCH 137/163] drm/i915: blt copy between objs using pre-created vma windows
Matthew Auld
- [PATCH 138/163] drm/i915/dg1: Eliminate eviction mutex
Matthew Auld
- [PATCH 139/163] drm/i915/dg1: Keep engine awake across whole blit
Matthew Auld
- [PATCH 140/163] drm/i915: window_blt_copy is used for swapin and swapout
Matthew Auld
- [PATCH 141/163] drm/i915: Lmem eviction statistics by category
Matthew Auld
- [PATCH 142/163] drm/i915/gem/selftest: test and measure window based blt cpy
Matthew Auld
- [PATCH 143/163] drm/i915: suspend/resume eviction
Matthew Auld
- [PATCH 144/163] drm/i915: Reset blitter context when unpark engine
Matthew Auld
- [PATCH 145/163] drm/i915/dg1: Add dedicated context for blitter eviction
Matthew Auld
- [PATCH 146/163] drm/i915/pm: suspend and restore ppgtt mapping
Matthew Auld
- [PATCH 147/163] drm/i915/gt: Allocate default ctx objects in SMEM
Matthew Auld
- [PATCH 148/163] drm/i915: suspend/resume enable blitter eviction
Matthew Auld
- [PATCH 149/163] drm/i915: suspend/resume handling of perma-pinned objects
Matthew Auld
- [PATCH 150/163] drm/i915: need consider system BO snoop for dgfx
Matthew Auld
- [PATCH 151/163] drm/i915: move eviction to prepare hook
Matthew Auld
- [PATCH 152/163] drm/i915: Perform execbuffer object locking as a separate step
Matthew Auld
- [PATCH 153/163] drm/i915: Implement eviction locking v2
Matthew Auld
- [PATCH 154/163] drm/i915: Support ww eviction
Matthew Auld
- [PATCH 155/163] drm/i915: Use a ww transaction in the fault handler
Matthew Auld
- [PATCH 156/163] drm/i915: Use a ww transaction in i915_gem_object_pin_map_unlocked()
Matthew Auld
- [PATCH 157/163] drm/i915: Improve accuracy of eviction stats
Matthew Auld
- [PATCH 158/163] drm/i915: Support ww locks in suspend/resume
Matthew Auld
- [PATCH 159/163] drm/i915/dg1: Fix mapping type for default state object
Matthew Auld
- [PATCH 160/163] drm/i915/dg1: Fix GPU hang due to shmemfs page drop
Matthew Auld
- [PATCH 161/163] drm/i915/dg1: allow pci to auto probe
Matthew Auld
- [PATCH 162/163] drm/i915: drop fake lmem
Matthew Auld
- [PATCH 163/163] HAX
Matthew Auld
- [PATCH 1/2] drm/i915: call i915_init_ggtt() earlier for DSB
Matthew Auld
- [PATCH 2/2] drm/i915/dsb: don't call I915_VMA_RELEASE_MAP
Matthew Auld
- [PATCH 1/2] drm/i915: call i915_init_ggtt() earlier for DSB
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [001/163] drm/i915/selftest: also consider non-contiguous objects
Patchwork
- [PATCH i-g-t] HAX: i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✗ Fi.CI.BAT: failure for HAX: i915/perf: Test OA buffer wrap around
Patchwork
- ✗ Fi.CI.BAT: failure for series starting with [1/2] drm/i915: call i915_init_ggtt() earlier for DSB
Patchwork
- ✗ Fi.CI.BAT: failure for Drop reason check from report
Patchwork
- [PATCH 01/72] drm/i915/gt: Limit frequency drop to RPe on parking
Chris Wilson
- [PATCH 02/72] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 03/72] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 04/72] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 05/72] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 06/72] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 07/72] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 08/72] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 09/72] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 10/72] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 11/72] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 12/72] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 13/72] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 14/72] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 15/72] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 16/72] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 17/72] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 18/72] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 19/72] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 20/72] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 21/72] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 22/72] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 23/72] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 24/72] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 25/72] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 26/72] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 27/72] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 28/72] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 29/72] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 30/72] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 31/72] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 32/72] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 33/72] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 34/72] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 35/72] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 36/72] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 37/72] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 38/72] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 39/72] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 40/72] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 41/72] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 42/72] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 43/72] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 44/72] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 45/72] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 46/72] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 47/72] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 48/72] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 49/72] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 50/72] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 51/72] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 52/72] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 53/72] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 54/72] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 55/72] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 56/72] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 57/72] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 58/72] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 59/72] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 60/72] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 61/72] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 62/72] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 63/72] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 64/72] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 65/72] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 66/72] cl-g6
Chris Wilson
- [PATCH 67/72] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 68/72] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 69/72] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 70/72] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 71/72] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- [PATCH 72/72] active
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/72] drm/i915/gt: Limit frequency drop to RPe on parking
Patchwork
- [PATCH i-g-t] HAX: i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✗ Fi.CI.BAT: failure for HAX: i915/perf: Test OA buffer wrap around (rev2)
Patchwork
- ✗ Fi.CI.BAT: failure for Drop reason check from report
Patchwork
- [PATCH 001/165] placeholder
Matthew Auld
- [PATCH 002/165] drm/i915/dsb: don't call I915_VMA_RELEASE_MAP
Matthew Auld
- [PATCH 003/165] drm/i915/selftest: also consider non-contiguous objects
Matthew Auld
- [PATCH 004/165] drm/i915/selftest: assert we get 2M GTT pages
Matthew Auld
- [PATCH 005/165] drm/i915/selftest: handle local-memory in perf_memcpy
Matthew Auld
- [PATCH 006/165] drm/i915/gt: Move move context layout registers and offsets to lrc_reg.h
Matthew Auld
- [PATCH 007/165] drm/i915/gt: Rename lrc.c to execlists_submission.c
Matthew Auld
- [PATCH 008/165] drm/i915: split gen8+ flush and bb_start emission functions to their own file
Matthew Auld
- [PATCH 009/165] drm/i915: split wa_bb code to its own file
Matthew Auld
- [PATCH 010/165] HAX drm/i915: Work around the selftest timeline lock splat workaround
Matthew Auld
- [PATCH 011/165] drm/i915: Introduce drm_i915_lock_isolated
Matthew Auld
- [PATCH 012/165] drm/i915: Lock hwsp objects isolated for pinning at create time
Matthew Auld
- [PATCH 013/165] drm/i915: Pin timeline map after first timeline pin, v5.
Matthew Auld
- [PATCH 014/165] drm/i915: Move cmd parser pinning to execbuffer
Matthew Auld
- [PATCH 015/165] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Matthew Auld
- [PATCH 016/165] drm/i915: Ensure we hold the object mutex in pin correctly v2
Matthew Auld
- [PATCH 017/165] drm/i915: Add gem object locking to madvise.
Matthew Auld
- [PATCH 018/165] drm/i915: Move HAS_STRUCT_PAGE to obj->flags
Matthew Auld
- [PATCH 019/165] drm/i915: Rework struct phys attachment handling
Matthew Auld
- [PATCH 020/165] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2.
Matthew Auld
- [PATCH 021/165] drm/i915: make lockdep slightly happier about execbuf.
Matthew Auld
- [PATCH 022/165] drm/i915: Disable userptr pread/pwrite support.
Matthew Auld
- [PATCH 023/165] drm/i915: No longer allow exporting userptr through dma-buf
Matthew Auld
- [PATCH 024/165] drm/i915: Reject more ioctls for userptr
Matthew Auld
- [PATCH 025/165] drm/i915: Reject UNSYNCHRONIZED for userptr, v2.
Matthew Auld
- [PATCH 026/165] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER.
Matthew Auld
- [PATCH 027/165] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v5.
Matthew Auld
- [PATCH 028/165] drm/i915: Flatten obj->mm.lock
Matthew Auld
- [PATCH 029/165] drm/i915: Populate logical context during first pin.
Matthew Auld
- [PATCH 030/165] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2.
Matthew Auld
- [PATCH 031/165] drm/i915: Handle ww locking in init_status_page
Matthew Auld
- [PATCH 032/165] drm/i915: Rework clflush to work correctly without obj->mm.lock.
Matthew Auld
- [PATCH 033/165] drm/i915: Pass ww ctx to intel_pin_to_display_plane
Matthew Auld
- [PATCH 034/165] drm/i915: Add object locking to vm_fault_cpu
Matthew Auld
- [PATCH 035/165] drm/i915: Move pinning to inside engine_wa_list_verify()
Matthew Auld
- [PATCH 036/165] drm/i915: Take reservation lock around i915_vma_pin.
Matthew Auld
- [PATCH 037/165] drm/i915: Make intel_init_workaround_bb more compatible with ww locking.
Matthew Auld
- [PATCH 038/165] drm/i915: Make __engine_unpark() compatible with ww locking v2
Matthew Auld
- [PATCH 039/165] drm/i915: Take obj lock around set_domain ioctl
Matthew Auld
- [PATCH 040/165] drm/i915: Defer pin calls in buffer pool until first use by caller.
Matthew Auld
- [PATCH 041/165] drm/i915: Fix pread/pwrite to work with new locking rules.
Matthew Auld
- [PATCH 042/165] drm/i915: Fix workarounds selftest, part 1
Matthew Auld
- [PATCH 043/165] drm/i915: Prepare for obj->mm.lock removal
Matthew Auld
- [PATCH 044/165] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner.
Matthew Auld
- [PATCH 045/165] drm/i915: Add ww locking around vm_access()
Matthew Auld
- [PATCH 046/165] drm/i915: Increase ww locking for perf.
Matthew Auld
- [PATCH 047/165] drm/i915: Lock ww in ucode objects correctly
Matthew Auld
- [PATCH 048/165] drm/i915: Add ww locking to dma-buf ops.
Matthew Auld
- [PATCH 049/165] drm/i915: Add missing ww lock in intel_dsb_prepare.
Matthew Auld
- [PATCH 050/165] drm/i915: Fix ww locking in shmem_create_from_object
Matthew Auld
- [PATCH 051/165] drm/i915: Use a single page table lock for each gtt.
Matthew Auld
- [PATCH 052/165] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 053/165] drm/i915/selftests: Prepare client blit for obj->mm.lock removal.
Matthew Auld
- [PATCH 054/165] drm/i915/selftests: Prepare coherency tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 055/165] drm/i915/selftests: Prepare context tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 056/165] drm/i915/selftests: Prepare dma-buf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 057/165] drm/i915/selftests: Prepare execbuf tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 058/165] drm/i915/selftests: Prepare mman testcases for obj->mm.lock removal.
Matthew Auld
- [PATCH 059/165] drm/i915/selftests: Prepare object tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 060/165] drm/i915/selftests: Prepare object blit tests for obj->mm.lock removal.
Matthew Auld
- [PATCH 061/165] drm/i915/selftests: Prepare igt_gem_utils for obj->mm.lock removal
Matthew Auld
- [PATCH 062/165] drm/i915/selftests: Prepare context selftest for obj->mm.lock removal
Matthew Auld
- [PATCH 063/165] drm/i915/selftests: Prepare hangcheck for obj->mm.lock removal
Matthew Auld
- [PATCH 064/165] drm/i915/selftests: Prepare execlists for obj->mm.lock removal
Matthew Auld
- [PATCH 065/165] drm/i915/selftests: Prepare mocs tests for obj->mm.lock removal
Matthew Auld
- [PATCH 066/165] drm/i915/selftests: Prepare ring submission for obj->mm.lock removal
Matthew Auld
- [PATCH 067/165] drm/i915/selftests: Prepare timeline tests for obj->mm.lock removal
Matthew Auld
- [PATCH 068/165] drm/i915/selftests: Prepare i915_request tests for obj->mm.lock removal
Matthew Auld
- [PATCH 069/165] drm/i915/selftests: Prepare memory region tests for obj->mm.lock removal
Matthew Auld
- [PATCH 070/165] drm/i915/selftests: Prepare cs engine tests for obj->mm.lock removal
Matthew Auld
- [PATCH 071/165] drm/i915/selftests: Prepare gtt tests for obj->mm.lock removal
Matthew Auld
- [PATCH 072/165] drm/i915: Finally remove obj->mm.lock.
Matthew Auld
- [PATCH 073/165] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2.
Matthew Auld
- [PATCH 074/165] drm/i915: Avoid some false positives in assert_object_held()
Matthew Auld
- [PATCH 075/165] drm/i915: Reference contending lock objects
Matthew Auld
- [PATCH 076/165] drm/i915: Break out dma_resv ww locking utilities to separate files
Matthew Auld
- [PATCH 077/165] drm/i915: Introduce a for_i915_gem_ww(){}
Matthew Auld
- [PATCH 078/165] drm/i915: Untangle the vma pages_mutex
Matthew Auld
- [PATCH 079/165] drm/i915/fbdev: Use lmem physical addresses for fb_mmap() on discrete
Matthew Auld
- [PATCH 080/165] drm/i915: Return error value when bo not in LMEM for discrete
Matthew Auld
- [PATCH 081/165] drm/i915/dmabuf: Disallow LMEM objects from dma-buf
Matthew Auld
- [PATCH 082/165] drm/i915/lmem: Fail driver init if LMEM training failed
Matthew Auld
- [PATCH 083/165] HAX drm/i915/lmem: support CPU relocations
Matthew Auld
- [PATCH 084/165] HAX drm/i915/lmem: support pread and pwrite
Matthew Auld
- [PATCH 085/165] drm/i915: Update the helper to set correct mapping
Matthew Auld
- [PATCH 086/165] drm/i915: introduce kernel blitter_context
Matthew Auld
- [PATCH 087/165] drm/i915/region: support basic eviction
Matthew Auld
- [PATCH 088/165] drm/i915: Add blit functions that can be called from within a WW transaction
Matthew Auld
- [PATCH 089/165] drm/i915: Delay publishing objects on the eviction lists
Matthew Auld
- [PATCH 090/165] drm/i915: support basic object migration
Matthew Auld
- [PATCH 091/165] drm/i915/dg1: Fix occasional migration error
Matthew Auld
- [PATCH 092/165] drm/i915/query: Expose memory regions through the query uAPI
Matthew Auld
- [PATCH 093/165] drm/i915: Store gt in memory region
Matthew Auld
- [PATCH 094/165] drm/i915/uapi: introduce drm_i915_gem_create_ext
Matthew Auld
- [PATCH 095/165] drm/i915/lmem: allocate cmd ring in lmem
Matthew Auld
- [PATCH 096/165] drm/i915/dg1: Do not check r->sgt.pfn for NULL
Matthew Auld
- [PATCH 097/165] drm/i915/dg1: Introduce dmabuf mmap to LMEM
Matthew Auld
- [PATCH 098/165] drm/i915: setup the LMEM region
Matthew Auld
- [PATCH 099/165] drm/i915: Distinction of memory regions
Matthew Auld
- [PATCH 100/165] drm/i915/gtt: map the PD up front
Matthew Auld
- [PATCH 101/165] drm/i915/gtt/dgfx: place the PD in LMEM
Matthew Auld
- [PATCH 102/165] drm/i915/gtt: make flushing conditional
Matthew Auld
- [PATCH 103/165] drm/i915/gtt/dg1: add PTE_LM plumbing for PPGTT
Matthew Auld
- [PATCH 104/165] drm/i915/gtt/dg1: add PTE_LM plumbing for GGTT
Matthew Auld
- [PATCH 105/165] drm/i915: allocate context from LMEM
Matthew Auld
- [PATCH 106/165] drm/i915: move engine scratch to LMEM
Matthew Auld
- [PATCH 107/165] drm/i915: Provide a way to disable PCIe relaxed write ordering
Matthew Auld
- [PATCH 108/165] drm/i915: i915 returns -EBUSY on thread contention
Matthew Auld
- [PATCH 109/165] drm/i915: setup GPU device lmem region
Matthew Auld
- [PATCH 110/165] drm/i915: Fix object page offset within a region
Matthew Auld
- [PATCH 111/165] drm/i915: add i915_gem_object_is_devmem() function
Matthew Auld
- [PATCH 112/165] drm/i915: finish memory region support for stolen objects.
Matthew Auld
- [PATCH 113/165] drm/i915/lmem: support optional CPU clearing for special internal use
Matthew Auld
- [PATCH 114/165] drm/i915/guc: put all guc objects in lmem when available
Matthew Auld
- [PATCH 115/165] drm/i915: Create stolen memory region from local memory
Matthew Auld
- [PATCH 116/165] drm/i915/lmem: Bypass aperture when lmem is available
Matthew Auld
- [PATCH 117/165] drm/i915/lmem: reset the lmem buffer created by fbdev
Matthew Auld
- [PATCH 118/165] drm/i915/dsb: Enable lmem for dsb
Matthew Auld
- [PATCH 119/165] drm/i915: Reintroduce mem->reserved
Matthew Auld
- [PATCH 120/165] drm/i915/dg1: Reserve first 1MB of local memory
Matthew Auld
- [PATCH 121/165] drm/i915/dg1: Read OPROM via SPI controller
Matthew Auld
- [PATCH 122/165] drm/i915/oprom: Basic sanitization
Matthew Auld
- [PATCH 123/165] drm/i915: WA for zero memory channel
Matthew Auld
- [PATCH 124/165] drm/i915/dg1: Compute MEM Bandwidth using MCHBAR
Matthew Auld
- [PATCH 125/165] drm/i915/dg1: Double memory bandwidth available
Matthew Auld
- [PATCH 126/165] drm/i915/lmem: allocate HWSP in lmem
Matthew Auld
- [PATCH 127/165] drm/i915/lmem: Limit block size to 4G
Matthew Auld
- [PATCH 128/165] drm/i915/gem: Update shmem available memory
Matthew Auld
- [PATCH 129/165] drm/i915: Allow non-uniform subslices in gen12+
Matthew Auld
- [PATCH 130/165] drm/i915/dg1: intel_memory_region_evict() changes for eviction
Matthew Auld
- [PATCH 131/165] drm/i915/dg1: i915_gem_object_memcpy(..) infrastructure
Matthew Auld
- [PATCH 132/165] drm/i915/dg1: Eviction logic
Matthew Auld
- [PATCH 133/165] drm/i915/dg1: Add enable_eviction modparam
Matthew Auld
- [PATCH 134/165] drm/i915/dg1: Add lmem_size modparam
Matthew Auld
- [PATCH 135/165] drm/i915/dg1: Track swap in/out stats via debugfs
Matthew Auld
- [PATCH 136/165] drm/i915/dg1: Measure swap in/out timing stats
Matthew Auld
- [PATCH 137/165] drm/i915: define intel_partial_pages_for_sg_table
Matthew Auld
- [PATCH 138/165] drm/i915: create and destroy dummy vma
Matthew Auld
- [PATCH 139/165] drm/i915: blt copy between objs using pre-created vma windows
Matthew Auld
- [PATCH 140/165] drm/i915/dg1: Eliminate eviction mutex
Matthew Auld
- [PATCH 141/165] drm/i915/dg1: Keep engine awake across whole blit
Matthew Auld
- [PATCH 142/165] drm/i915: window_blt_copy is used for swapin and swapout
Matthew Auld
- [PATCH 143/165] drm/i915: Lmem eviction statistics by category
Matthew Auld
- [PATCH 144/165] drm/i915/gem/selftest: test and measure window based blt cpy
Matthew Auld
- [PATCH 145/165] drm/i915: suspend/resume eviction
Matthew Auld
- [PATCH 146/165] drm/i915: Reset blitter context when unpark engine
Matthew Auld
- [PATCH 147/165] drm/i915/dg1: Add dedicated context for blitter eviction
Matthew Auld
- [PATCH 148/165] drm/i915/pm: suspend and restore ppgtt mapping
Matthew Auld
- [PATCH 149/165] drm/i915/gt: Allocate default ctx objects in SMEM
Matthew Auld
- [PATCH 150/165] drm/i915: suspend/resume enable blitter eviction
Matthew Auld
- [PATCH 151/165] drm/i915: suspend/resume handling of perma-pinned objects
Matthew Auld
- [PATCH 152/165] drm/i915: need consider system BO snoop for dgfx
Matthew Auld
- [PATCH 153/165] drm/i915: move eviction to prepare hook
Matthew Auld
- [PATCH 154/165] drm/i915: Perform execbuffer object locking as a separate step
Matthew Auld
- [PATCH 155/165] drm/i915: Implement eviction locking v2
Matthew Auld
- [PATCH 156/165] drm/i915: Support ww eviction
Matthew Auld
- [PATCH 157/165] drm/i915: Use a ww transaction in the fault handler
Matthew Auld
- [PATCH 158/165] drm/i915: Use a ww transaction in i915_gem_object_pin_map_unlocked()
Matthew Auld
- [PATCH 159/165] drm/i915: Improve accuracy of eviction stats
Matthew Auld
- [PATCH 160/165] drm/i915: Support ww locks in suspend/resume
Matthew Auld
- [PATCH 161/165] drm/i915/dg1: Fix mapping type for default state object
Matthew Auld
- [PATCH 162/165] drm/i915/dg1: Fix GPU hang due to shmemfs page drop
Matthew Auld
- [PATCH 163/165] drm/i915/dg1: allow pci to auto probe
Matthew Auld
- [PATCH 164/165] drm/i915: drop fake lmem
Matthew Auld
- [PATCH 165/165] HAX
Matthew Auld
- ✗ Fi.CI.BAT: failure for series starting with [001/165] placeholder
Patchwork
- [PATCH 1/2] drm/i915: prevent GGTT pinning during nogem display init
Matthew Auld
- [PATCH 2/2] drm/i915/dsb: don't call I915_VMA_RELEASE_MAP
Matthew Auld
- ✗ Fi.CI.BAT: failure for series starting with [1/2] drm/i915: prevent GGTT pinning during nogem display init
Patchwork
- [PATCH 1/2] drm/i915: prevent GGTT pinning during nogem display init
Chris Wilson
- [PATCH 1/2] drm/i915/display: Record the plane update times for debugging
Chris Wilson
- [PATCH 2/2] drm/i915/display: Defer initial modeset until after GGTT is initialised
Chris Wilson
- [PATCH 1/3] drm/i915: forbid dsb from GGTT pinning during nogem init
Matthew Auld
- [PATCH 2/3] drm/i915: sanity check uninitialised GGTT pin
Matthew Auld
- [PATCH 3/3] drm/i915: don't call I915_VMA_RELEASE_MAP
Matthew Auld
- ✗ Fi.CI.BUILD: failure for series starting with [1/2] drm/i915/display: Record the plane update times for debugging
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915: forbid dsb from GGTT pinning during nogem init
Patchwork
- [PATCH 1/2] drm/i915/display: Record the plane update times for debugging
Chris Wilson
- [PATCH 2/2] drm/i915/display: Defer initial modeset until after GGTT is initialised
Chris Wilson
- [PATCH 1/6] drm/i915/gt: Limit frequency drop to RPe on parking
Chris Wilson
- [PATCH 2/6] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 3/6] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 4/6] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 5/6] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 6/6] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/display: Record the plane update times for debugging
Patchwork
- ✓ Fi.CI.BAT: success for series starting with [1/6] drm/i915/gt: Limit frequency drop to RPe on parking
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/3] drm/i915: forbid dsb from GGTT pinning during nogem init
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915/display: Record the plane update times for debugging
Patchwork
- [PATCH 01/72] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 02/72] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 03/72] drm/i915/gt: Limit frequency drop to RPe on parking
Chris Wilson
- [PATCH 04/72] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 05/72] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 06/72] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 07/72] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 08/72] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 09/72] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 10/72] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 11/72] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 12/72] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 13/72] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 14/72] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 15/72] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 16/72] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 17/72] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 18/72] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 19/72] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 20/72] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 21/72] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 22/72] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 23/72] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 24/72] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 25/72] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 26/72] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 27/72] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 28/72] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 29/72] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 30/72] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 31/72] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 32/72] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 33/72] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 34/72] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 35/72] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 36/72] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 37/72] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 38/72] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 39/72] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 40/72] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 41/72] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 42/72] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 43/72] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 44/72] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 45/72] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 46/72] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 47/72] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 48/72] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 49/72] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 50/72] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 51/72] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 52/72] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 53/72] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 54/72] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 55/72] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 56/72] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 57/72] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 58/72] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 59/72] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 60/72] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 61/72] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 62/72] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 63/72] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 64/72] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 65/72] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 66/72] cl-g6
Chris Wilson
- [PATCH 67/72] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 68/72] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 69/72] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 70/72] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 71/72] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- [PATCH 72/72] active
Chris Wilson
- ✗ Fi.CI.IGT: failure for series starting with [1/6] drm/i915/gt: Limit frequency drop to RPe on parking
Patchwork
- ✗ Fi.CI.BAT: failure for series starting with [01/72] drm/i915/gt: Decouple completed requests on unwind
Patchwork
- [PATCH i-g-t] HAX: i915/perf: Test OA buffer wrap around
Umesh Nerlige Ramappa
- [PATCH 0/1] Drop reason check from report
Umesh Nerlige Ramappa
- [PATCH 1/1] i915/perf: Drop report reason check
Umesh Nerlige Ramappa
- ✗ Fi.CI.BAT: failure for HAX: i915/perf: Test OA buffer wrap around (rev3)
Patchwork
- ✗ Fi.CI.BAT: failure for Drop reason check from report
Patchwork
- [PATCH 01/72] drm/i915/gt: Decouple completed requests on unwind
Chris Wilson
- [PATCH 02/72] drm/i915/gt: Check for a completed last request once
Chris Wilson
- [PATCH 03/72] drm/i915/gt: Protect context lifetime with RCU
Chris Wilson
- [PATCH 04/72] drm/i915/gt: Split the breadcrumb spinlock between global and contexts
Chris Wilson
- [PATCH 05/72] drm/i915/gt: Move the breadcrumb to the signaler if completed upon cancel
Chris Wilson
- [PATCH 06/72] drm/i915/gt: Limit frequency drop to RPe on parking
Chris Wilson
- [PATCH 07/72] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 08/72] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 09/72] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 10/72] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 11/72] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 12/72] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 13/72] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 14/72] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 15/72] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 16/72] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 17/72] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 18/72] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 19/72] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 20/72] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 21/72] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 22/72] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 23/72] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 24/72] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 25/72] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 26/72] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 27/72] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 28/72] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 29/72] drm/i915/gem: Delay attach mmu-notifier until we acquire the pinned userptr
Chris Wilson
- [PATCH 30/72] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 31/72] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 32/72] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 33/72] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 34/72] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 35/72] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 36/72] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 37/72] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 38/72] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 39/72] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 40/72] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 41/72] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 42/72] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 43/72] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 44/72] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 45/72] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 46/72] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 47/72] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 48/72] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 49/72] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 50/72] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 51/72] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 52/72] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 53/72] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 54/72] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 55/72] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 56/72] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 57/72] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 58/72] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 59/72] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 60/72] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 61/72] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 62/72] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 63/72] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 64/72] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 65/72] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 66/72] cl-g6
Chris Wilson
- [PATCH 67/72] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 68/72] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 69/72] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 70/72] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 71/72] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- [PATCH 72/72] active
Chris Wilson
- ✗ Fi.CI.BAT: failure for series starting with [01/72] drm/i915/gt: Decouple completed requests on unwind
Patchwork
- ✗ Fi.CI.BUILD: warning for series starting with [01/72] drm/i915/gt: Decouple completed requests on unwind
Patchwork
- [PATCH] Revert "drm/i915: re-order if/else ladder for hpd_irq_setup"
Chris Wilson
- [PATCH 01/12] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 02/12] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 03/12] drm/i915/gt: Limit frequency drop to RPe on parking
Chris Wilson
- [PATCH 04/12] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 05/12] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 06/12] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 07/12] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 08/12] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 09/12] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 10/12] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 11/12] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 12/12] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH] HAX sound: Disable probing sna_hda with DG1
Chris Wilson
- ✓ Fi.CI.BAT: success for series starting with [01/12] drm/i915/gt: Replace direct submit with direct call to tasklet
Patchwork
- ✓ Fi.CI.BAT: success for drm/i915/gt: Replace direct submit with direct call to tasklet
Patchwork
- ✓ Fi.CI.BAT: success for HAX sound: Disable probing sna_hda with DG1
Patchwork
- ✗ Fi.CI.IGT: failure for series starting with [01/12] drm/i915/gt: Replace direct submit with direct call to tasklet
Patchwork
- ✗ Fi.CI.IGT: failure for drm/i915/gt: Replace direct submit with direct call to tasklet
Patchwork
- ✓ Fi.CI.IGT: success for HAX sound: Disable probing sna_hda with DG1
Patchwork
- [PATCH] HAX sound: Disable probing snd_hda with DG1
Chris Wilson
- [PATCH 1/3] drm/i915: Disable outputs during unregister
Chris Wilson
- [PATCH 2/3] drm/i915/gem: Differentiate oom failures from invalid map types
Chris Wilson
- [PATCH 3/3] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 01/68] drm/i915: Disable outputs during unregister
Chris Wilson
- [PATCH 02/68] drm/i915/gem: Differentiate oom failures from invalid map types
Chris Wilson
- [PATCH 03/68] drm/i915/gt: Replace direct submit with direct call to tasklet
Chris Wilson
- [PATCH 04/68] drm/i915/gt: ce->inflight updates are now serialised
Chris Wilson
- [PATCH 05/68] drm/i915/gt: Limit frequency drop to RPe on parking
Chris Wilson
- [PATCH 06/68] drm/i915/gem: Drop free_work for GEM contexts
Chris Wilson
- [PATCH 07/68] drm/i915/gt: Track the overall awake/busy time
Chris Wilson
- [PATCH 08/68] drm/i915/gt: Use virtual_engine during execlists_dequeue
Chris Wilson
- [PATCH 09/68] drm/i915/gt: Decouple inflight virtual engines
Chris Wilson
- [PATCH 10/68] drm/i915/gt: Defer schedule_out until after the next dequeue
Chris Wilson
- [PATCH 11/68] drm/i915/gt: Remove virtual breadcrumb before transfer
Chris Wilson
- [PATCH 12/68] drm/i915/gt: Shrink the critical section for irq signaling
Chris Wilson
- [PATCH 13/68] drm/i915/gt: Resubmit the virtual engine on schedule-out
Chris Wilson
- [PATCH 14/68] drm/i915/gt: Simplify virtual engine handling for execlists_hold()
Chris Wilson
- [PATCH 15/68] drm/i915: Encode fence specific waitqueue behaviour into the wait.flags
Chris Wilson
- [PATCH 16/68] drm/i915/gt: Wrap intel_timeline.has_initial_breadcrumb
Chris Wilson
- [PATCH 17/68] drm/i915/gt: Track timeline GGTT offset separately from subpage offset
Chris Wilson
- [PATCH 18/68] drm/i915/gt: Add timeline "mode"
Chris Wilson
- [PATCH 19/68] drm/i915/gt: Use indices for writing into relative timelines
Chris Wilson
- [PATCH 20/68] drm/i915/selftests: Exercise relative timeline modes
Chris Wilson
- [PATCH 21/68] drm/i915/gt: Use ppHWSP for unshared non-semaphore related timelines
Chris Wilson
- [PATCH 22/68] drm/i915/selftests: Confirm RING_TIMESTAMP / CTX_TIMESTAMP share a clock
Chris Wilson
- [PATCH 23/68] drm/i915/gt: Consolidate the CS timestamp clocks
Chris Wilson
- [PATCH 24/68] drm/i915/gt: Prefer recycling an idle fence
Chris Wilson
- [PATCH 25/68] drm/i915/gem: Optimistically prune dma-resv from the shrinker.
Chris Wilson
- [PATCH 26/68] drm/i915: Drop i915_request.lock serialisation around await_start
Chris Wilson
- [PATCH 27/68] drm/i915: Drop i915_request.lock requirement for intel_rps_boost()
Chris Wilson
- [PATCH 28/68] drm/i915/gem: Reduce ctx->engine_mutex for reading the clone source
Chris Wilson
- [PATCH 29/68] drm/i915/gem: Reduce ctx->engines_mutex for get_engines()
Chris Wilson
- [PATCH 30/68] drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()
Chris Wilson
- [PATCH 31/68] drm/i915/gt: Drop atomic for engine->fw_active tracking
Chris Wilson
- [PATCH 32/68] drm/i915/gt: Extract busy-stats for ring-scheduler
Chris Wilson
- [PATCH 33/68] drm/i915/gt: Convert stats.active to plain unsigned int
Chris Wilson
- [PATCH 34/68] drm/i915/gt: Refactor heartbeat request construction and submission
Chris Wilson
- [PATCH 35/68] drm/i915: Strip out internal priorities
Chris Wilson
- [PATCH 36/68] drm/i915: Remove I915_USER_PRIORITY_SHIFT
Chris Wilson
- [PATCH 37/68] drm/i915/gt: Defer the kmem_cache_free() until after the HW submit
Chris Wilson
- [PATCH 38/68] drm/i915: Prune empty priolists
Chris Wilson
- [PATCH 39/68] drm/i915: Replace engine->schedule() with a known request operation
Chris Wilson
- [PATCH 40/68] drm/i915/gt: Do not suspend bonded requests if one hangs
Chris Wilson
- [PATCH 41/68] drm/i915: Teach the i915_dependency to use a double-lock
Chris Wilson
- [PATCH 42/68] drm/i915: Restructure priority inheritance
Chris Wilson
- [PATCH 43/68] drm/i915/selftests: Measure set-priority duration
Chris Wilson
- [PATCH 44/68] drm/i915/selftests: Exercise priority inheritance around an engine loop
Chris Wilson
- [PATCH 45/68] drm/i915: Improve DFS for priority inheritance
Chris Wilson
- [PATCH 46/68] drm/i915/gt: Remove timeslice suppression
Chris Wilson
- [PATCH 47/68] drm/i915: Extract request submission from execlists
Chris Wilson
- [PATCH 48/68] drm/i915: Extract request suspension from the execlists backend
Chris Wilson
- [PATCH 49/68] drm/i915: Extract the ability to defer and rerun a request later
Chris Wilson
- [PATCH 50/68] drm/i915: Fix the iterative dfs for defering requests
Chris Wilson
- [PATCH 51/68] drm/i915: Fair low-latency scheduling
Chris Wilson
- [PATCH 52/68] drm/i915/gt: Specify a deadline for the heartbeat
Chris Wilson
- [PATCH 53/68] drm/i915: Extend the priority boosting for the display with a deadline
Chris Wilson
- [PATCH 54/68] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
- [PATCH 55/68] drm/i915: Move scheduler queue
Chris Wilson
- [PATCH 56/68] drm/i915: Move tasklet from execlists to sched
Chris Wilson
- [PATCH 57/68] drm/i915/gt: Another tweak for flushing the tasklets
Chris Wilson
- [PATCH 58/68] Restore "drm/i915: drop engine_pin/unpin_breadcrumbs_irq"
Chris Wilson
- [PATCH 59/68] drm/i915/gt: Couple tasklet scheduling for all CS interrupts
Chris Wilson
- [PATCH 60/68] drm/i915/gt: Support creation of 'internal' rings
Chris Wilson
- [PATCH 61/68] drm/i915/gt: Use client timeline address for seqno writes
Chris Wilson
- [PATCH 62/68] cl-g6
Chris Wilson
- [PATCH 63/68] drm/i915/gt: Infrastructure for ring scheduling
Chris Wilson
- [PATCH 64/68] drm/i915/gt: Enable busy-stats for ring-scheduler
Chris Wilson
- [PATCH 65/68] drm/i915/gt: Implement ring scheduler for gen6/7
Chris Wilson
- [PATCH 66/68] drm/i915/gt: Enable ring scheduling for gen6/7
Chris Wilson
- [PATCH 67/68] drm/i915: Move saturated workload detection back to the context
Chris Wilson
- [PATCH 68/68] active
Chris Wilson
Last message date:
Mon Nov 30 10:58:08 UTC 2020
Archived on: Mon Nov 30 11:00:25 UTC 2020
This archive was generated by
Pipermail 0.09 (Mailman edition).