summaryrefslogtreecommitdiff
path: root/src/sna/kgem.c
AgeCommit message (Expand)Author
2016-04-06sna: Manually expand sse2 memcpy to compensate for a bad compilerChris Wilson
2016-03-16sna: Relax tiling requirements to cope with kernel errorsChris Wilson
2016-02-18sna: Use dirtyfb for fallback direct renderingChris Wilson
2016-02-18sna: Restrict set-domain short-circuit to readonlyChris Wilson
2016-02-13sna: Also mark imported Prime objects as uncleanChris Wilson
2016-02-13sna/dri2: Initialize priv->cache_sizeChris Wilson
2016-02-13sna: Reorder busyness checks after finding the proxy's parentsChris Wilson
2015-12-30sna: Disable WC mmaps for 945 and olderChris Wilson
2015-12-30sna: Skip sync if we have an idle coherent boChris Wilson
2015-11-22sna: Remove obsolete assertionChris Wilson
2015-11-18sna: Only reap the VMA if tiling actually changesChris Wilson
2015-11-18sna: Add some DBG around tiling changesChris Wilson
2015-11-18sna: Remove the VMA from the list when unmappingChris Wilson
2015-11-13sna: Limit the allocation retriesChris Wilson
2015-11-13sna: Fix typo when comparing '=' ! '=='Chris Wilson
2015-11-12sna: Wait upon the same ring when out-of-memoryChris Wilson
2015-11-11sna: Clear the whole tilingv2 structChris Wilson
2015-11-08sna: Simply retire-requestsChris Wilson
2015-11-06sna: Remove bogus assertion from last patchChris Wilson
2015-11-06sna: Avoid handing back a cached pinned batchChris Wilson
2015-10-31sna: Add a couple more asserts to catch static requests being queuedChris Wilson
2015-10-19sna: Use the new context param for report GTT sizeChris Wilson
2015-10-19sna: Manually clear the GPU bo needs_flush flag after doing an explicit flushChris Wilson
2015-10-13sna: Handle proxy objects in undo assertionsChris Wilson
2015-09-23sna: Explicitly retire any buffers underneath the static_requestChris Wilson
2015-09-18sna: Check alternate userptr mmappingsChris Wilson
2015-08-22sna: Fix build on squeezeChris Wilson
2015-08-07sna: Remove incorrect assertionChris Wilson
2015-08-07sna: Add a handle=%d to a DBG for consistencyChris Wilson
2015-08-07sna: More simple DBG tracing around cleanup/retire handlingChris Wilson
2015-08-07sna: Release the reference on the rq->bo during forced cleanupChris Wilson
2015-08-07sna: Always prefer a fresh allocation for the batch if !llcChris Wilson
2015-08-07sna: Add a little more DBG information to kgem_retire()Chris Wilson
2015-08-07sna: Fix sync'ing to the most recent request on forced cache cleanupChris Wilson
2015-08-07sna: Add a DBG for when we discard an invalid old mmappingChris Wilson
2015-08-07sna: Treat being wedged as unlikely during rendering with HWChris Wilson
2015-08-07sna: Stop allocating requests once wedgedChris Wilson
2015-08-07sna: Tweak DBG traces for cache cleanupChris Wilson
2015-08-07sna: Add DBG trace when doing a forced cache cleanupChris Wilson
2015-08-07sna: Harden batch allocation against resource starvationChris Wilson
2015-08-07sna: Fallback after a bo allocation failure for the batchChris Wilson
2015-08-07sna: Handle batch allocation failureChris Wilson
2015-07-25sna: Put the assert(!kgem->can_fence) back to where they wereChris Wilson
2015-07-24sna: Only reset the bo->needs_flush flag if the buffer is busy-writeChris Wilson
2015-07-24sna: Add logging for set-tiling failureChris Wilson
2015-07-24sna: Use DBG_NO_TILING to disable all fencingChris Wilson
2015-07-24sna: Ensure that DRI2 buffers are created with a fenceChris Wilson
2015-07-16sna: Stop trying to CPU mmap a prime object after failureChris Wilson
2015-06-24sna: Fix misspelling of "caching"Chris Wilson
2015-06-09sna: Pass scanout flag when creating PRIME boChris Wilson
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-22 23:14:21 +0000