summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/i915/i915_active.c
AgeCommit message (Expand)Author
2020-10-15Merge tag 'drm-next-2020-10-15' of git://anongit.freedesktop.org/drm/drmLinus Torvalds
2020-09-24treewide: Make all debug_obj_descriptors constStephen Boyd
2020-09-07drm/i915: Reduce locking around i915_active_acquire_preallocate_barrier()Chris Wilson
2020-09-07drm/i915: Make the stale cached active node available for any timelineChris Wilson
2020-09-07drm/i915: Keep the most recently used active-fence upon discardChris Wilson
2020-09-07drm/i915: Export a preallocate variant of i915_active_acquire()Chris Wilson
2020-09-07drm/i915: Skip taking acquire mutex for no ref->active callbackChris Wilson
2020-08-17drm/i915: Fix wrong return valueTianjia Zhang
2020-04-06drm/i915: Allow asynchronous waits on the i915_active barriersChris Wilson
2020-04-06drm/i915: Make exclusive awaits on i915_active optionalChris Wilson
2020-03-30drm/i915: Wrap i915_active in a simple kreffed structChris Wilson
2020-03-30drm/i915: Allow for different modes of interruptible i915_active_waitChris Wilson
2020-03-23drm/i915/gt: Delay release of engine-pm after last retirementChris Wilson
2020-03-11drm/i915: Extend i915_request_await_active to use all timelinesChris Wilson
2020-02-28drm/i915: Skip barriers inside waitsChris Wilson
2020-02-25drm/i915: Flush idle barriers when waitingChris Wilson
2020-02-25drm/i915: Drop assertion that active->fence is unchangedChris Wilson
2020-02-03drm/i915: Hold reference to previous active fence as we queueChris Wilson
2020-01-30drm/i915: Use the async worker to avoid reclaim tainting the ggtt->mutexChris Wilson
2020-01-30drm/i915: Fix preallocated barrier list appendJosé Roberto de Souza
2020-01-27drm/i915: Tighten atomicity of i915_active_acquire vs i915_active_releaseChris Wilson
2020-01-17drm/i915: Satisfy smatch that a loop has at least one iterationChris Wilson
2020-01-06drm/i915/gt: Drop mutex serialisation between context pin/unpinChris Wilson
2019-12-21drm/i915: Remove i915->kernel_contextChris Wilson
2019-12-05drm/i915: Serialise i915_active_acquire() with __active_retire()Chris Wilson
2019-12-02drm/i915: Serialise i915_active_wait() with its retirementChris Wilson
2019-12-02drm/i915: Specialise i915_active.work lock classesChris Wilson
2019-11-27drm/i915: Serialise i915_active_fence_set() with itselfChris Wilson
2019-11-20drm/i915: Mark up the calling context for intel_wakeref_put()Chris Wilson
2019-11-14drm/i915: Split i915_active.mutex into an irq-safe spinlock for the rbtreeChris Wilson
2019-10-24drm/i915/gt: Split intel_ring_submissionChris Wilson
2019-10-21drm/i915/gt: Introduce barrier pulses along enginesChris Wilson
2019-10-18drm/i915: Make for_each_engine_masked work on intel_gtTvrtko Ursulin
2019-10-04drm/i915: Move idle barrier cleanup into engine-pmChris Wilson
2019-10-04drm/i915: Coordinate i915_active with its own mutexChris Wilson
2019-10-04drm/i915: Push the i915_active.retire into a workerChris Wilson
2019-10-04drm/i915: Pull i915_vma_pin under the vm->mutexChris Wilson
2019-09-20drm/i915: Mark i915_request.timeline as a volatile, rcu pointerChris Wilson
2019-08-27drm/i915: Only activate i915_active debugobject onceChris Wilson
2019-08-19drm/i915: i915_active.retire() is optionalChris Wilson
2019-08-16drm/i915: Markup expected timeline locks for i915_activeChris Wilson
2019-08-14drm/i915: Serialise read/write of the barrier's engineChris Wilson
2019-08-09drm/i915: Lift timeline into intel_contextChris Wilson
2019-08-02drm/i915: Allow sharing the idle-barrier from other kernel requestsChris Wilson
2019-07-26drm/i915: Do not rely on for loop caching the maskTvrtko Ursulin
2019-07-26drm/i915: Capture vma contents outside of spinlockChris Wilson
2019-07-03drm/i915: Markup potential lock for i915_activeChris Wilson
2019-07-02drm/i915: Report if i915_active is still busy upon waitingChris Wilson
2019-06-21drm/i915: Provide an i915_active.acquire callbackChris Wilson
2019-06-21drm/i915: Track i915_active using debugobjectsChris Wilson