aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/i915_request.c (follow)
AgeCommit message (Expand)AuthorFilesLines
2020-07-13drm/i915: Skip signaling a signaled requestChris Wilson1-10/+13
2020-06-03drm/i915: Drop i915_request.i915 backpointerChris Wilson1-6/+6
2020-05-29drm/i915: Check for awaits on still currently executing requestsChris Wilson1-1/+48
2020-05-29drm/i915: Add a few asserts around handling of i915_request_is_active()Chris Wilson1-2/+3
2020-05-27drm/i915/gt: Do not schedule normal requests immediately along virtualChris Wilson1-4/+21
2020-05-27drm/i915: Reorder await_execution before await_requestChris Wilson1-132/+132
2020-05-26drm/i915: Improve execute_cb struct packingChris Wilson1-9/+14
2020-05-21drm/i915: Avoid using rq->engine after free during i915_fence_releaseChris Wilson1-2/+33
2020-05-14drm/i915: Drop no-semaphore boostingChris Wilson1-36/+4
2020-05-13drm/i915: Mark the addition of the initial-breadcrumb in the requestChris Wilson1-1/+6
2020-05-09drm/i915: Replace the hardcoded I915_FENCE_TIMEOUTChris Wilson1-1/+2
2020-05-08drm/i915: Prevent using semaphores to chain up to external fencesChris Wilson1-0/+26
2020-05-08drm/i915: Peel dma-fence-chains for awaitLionel Landwerlin1-1/+28
2020-05-08drm/i915: Pull waiting on an external dma-fence into its routineChris Wilson1-6/+10
2020-05-08drm/i915: Ignore submit-fences on the same timelineChris Wilson1-0/+3
2020-05-07drm/i915: Remove wait priority boostingChris Wilson1-3/+0
2020-05-07drm/i915: Mark concurrent submissions with a weak-dependencyChris Wilson1-2/+6
2020-05-06drm/i915: Propagate error from completed fencesChris Wilson1-1/+3
2020-04-03drm/i915: Keep a per-engine request poolChris Wilson1-5/+22
2020-03-23drm/i915: Rely on direct submission to the queueChris Wilson1-2/+0
2020-03-10drm/i915: Mark up racy read of active rq->engineChris Wilson1-3/+5
2020-03-10drm/i915: Defer semaphore priority bumping to a workqueueChris Wilson1-5/+17
2020-03-10drm/i915: Improve the start alignment of bonded pairsChris Wilson1-5/+36
2020-03-09drm/i915: Mark racy read of intel_engine_cs.saturatedChris Wilson1-1/+1
2020-03-07drm/i915: Do not poison i915_request.link on removalChris Wilson1-1/+1
2020-03-06drm/i915: Assert requests within a context are submitted in orderChris Wilson1-0/+11
2020-03-05drm/i915: Return early for await_start on same timelineChris Wilson1-2/+2
2020-03-05drm/i915: Actually emit the await_startChris Wilson1-1/+1
2020-03-04drm/i915: Apply i915_request_skip() on submissionChris Wilson1-20/+49
2020-03-03drm/i915/gem: Check that the context wasn't closed during setupChris Wilson1-46/+8
2020-02-28drm/i915/gt: Expose busywait duration to sysfsChris Wilson1-9/+10
2020-02-28drm/i915: Protect i915_request_await_start from early waitsChris Wilson1-13/+28
2020-02-20drm/i915: remove the other slab_dependenciesMatthew Auld1-11/+0
2020-02-12drm/i915: Poison rings after useChris Wilson1-9/+17
2020-02-11drm/i915: Disable use of hwsp_cacheline for kernel_contextChris Wilson1-5/+9
2020-02-05drm/i915: Flush execution tasklets before checking request statusChris Wilson1-1/+2
2020-02-03drm/i915: Initialise basic fence before acquiring seqnoChris Wilson1-7/+14
2020-01-22drm/i915: Mark the removal of the i915_request from the sched.linkChris Wilson1-0/+2
2020-01-16drm/i915: Keep track of request among the scheduling listsChris Wilson1-1/+3
2020-01-06drm/i915: Merge i915_request.flags with i915_request.fence.flagsChris Wilson1-1/+0
2019-12-23drm/i915: Mark the GEM context link as RCU protectedChris Wilson1-3/+3
2019-12-21drm/i915: Remove i915->kernel_contextChris Wilson1-1/+4
2019-12-20drm/i915: Push the use-semaphore marker onto the intel_contextChris Wilson1-5/+3
2019-12-20drm/i915: Drop GEM context as a direct link from i915_requestChris Wilson1-8/+13
2019-12-18drm/i915/gt: Remove direct invocation of breadcrumb signalingChris Wilson1-1/+1
2019-12-17drm/i915/gt: Eliminate the trylock for reading a timeline's hwspChris Wilson1-2/+2
2019-12-16drm/i915: Eliminate the trylock for awaiting an earlier requestChris Wilson1-18/+21
2019-12-13drm/i915: Use EAGAIN for trylock failuresChris Wilson1-1/+1
2019-12-13drm/i915: Introduce new macros for tracingVenkata Sandeep Dhanalakota1-18/+5
2019-12-11drm/i915: Use the i915_device name for identifying our request fencesChris Wilson1-2/+2