summaryrefslogtreecommitdiff
path: root/src/sna/kgem.h
AgeCommit message (Expand)Author
2012-09-12sna: Keep a very small, short-lived cache of large buffersChris Wilson
2012-09-12sna: Flush after operating on large buffersChris Wilson
2012-09-06sna: Apply the minimum 256 pitch to CREATE_USAGE_SHARED pixmaps as wellChris Wilson
2012-09-04sna: Fixup CREATE_USAGE_SHAREDChris Wilson
2012-09-04sna: Port prime interfacingChris Wilson
2012-08-28sna: Propagate the request to flush rather than directly submit the batchChris Wilson
2012-08-27sna: Track outstanding requests per-ringChris Wilson
2012-08-27sna: Flush the batch if it references a ShmPixmap and the GPU is idleChris Wilson
2012-08-26sna: Add some DBG to kgem_is_idle()Chris Wilson
2012-08-24sna: Allow the batch to be flushed if the GPU is idle upon a context switchChris Wilson
2012-08-21sna: Mark all levels of a proxy as dirtyChris Wilson
2012-08-20sna: Trim a parameter from kgem_bo_mark_dirty() and add some assertionsChris Wilson
2012-07-25Assume all unknown chipsets are future genChris Wilson
2012-07-21sna: Update assertion for cached io buffersChris Wilson
2012-07-21sna: Correct assertion for __kgem_bo_size()Chris Wilson
2012-07-21sna: Change the vmap interface name to userptrChris Wilson
2012-07-21sna: s/vmap/snoop/ since we use the flag more genericallyChris Wilson
2012-07-21sna: Fix role reversal of __kgem_bo_size() and kgem_bo_size()!Chris Wilson
2012-07-21sna: Refresh experimental userptr vmap supportChris Wilson
2012-07-21sna: Only update a buffer when it becomes dirtyChris Wilson
2012-07-20sna: Rename kgem_partial_bo to kgem_bufferChris Wilson
2012-07-20sna: Replace 'sync' flag with equivalent 'flush'Chris Wilson
2012-07-20sna: Remove unused scanout-is-dirty? flagChris Wilson
2012-07-19sna: Handle mixed bo/buffers in assertionsChris Wilson
2012-07-18sna: Enable runtime detection of set-cacheing ioctlChris Wilson
2012-07-18sna: Maintain a short-lived cache of snoopable CPU bo for older genChris Wilson
2012-07-16sna: Disable snoopable uplaod buffers for gen4Chris Wilson
2012-07-16Wrap defines to avoid redefinition warningsChris Wilson
2012-07-14sna: Avoid BLT to snoopable bo on older genChris Wilson
2012-07-14sna: Allow BLT support on future unknown hardwareChris Wilson
2012-07-14sna: Do not perform batch compaction on old kernelsChris Wilson
2012-07-14sna: Aim for consistency and use stdbool except for core X APIsChris Wilson
2012-07-11sna: Remove unused cached partials listChris Wilson
2012-07-10sna: Avoid the expensive recomputation of the region when copyingChris Wilson
2012-07-09sna: Simplify the DBG incarnationChris Wilson
2012-07-03sna: Use set-cache-level ioctl to create CPU boChris Wilson
2012-06-29sna: Only consider the request list when deciding whether the GPU is busyChris Wilson
2012-06-23sna: extend RandR to support super sized monitor configurationsChris Wilson
2012-06-21sna/gen3+: Keep vbo cachedChris Wilson
2012-06-20sna: Don't clear the needs_flush flag after emitting a flush on the busy boChris Wilson
2012-06-19sna: Explain why we ignore the busy status result during kgem_bo_flush()Chris Wilson
2012-06-18sna: Check if the busy is truly busy before commiting to an indirect uploadChris Wilson
2012-06-11sna: Instrument memory/bo allocations for monitoring over timeChris Wilson
2012-06-04sna: Make the bo-is-busy DBG more useful by saying which bo it isChris Wilson
2012-05-29sna: Also check for a free exec slot for the upload buffersChris Wilson
2012-05-28sna/dri: Requeue vblank throttling until the vsync'ed copy completesChris Wilson
2012-05-25sna: Trust the crtc-is-bound determination after modeset and hotplugChris Wilson
2012-05-18sna: Give the proxy a unique nameChris Wilson
2012-05-13sna: Tweak usage of kgem_bo_can_map vs kgem_bo_mappedChris Wilson
2012-05-13sna: Simplify partial buffer trackingChris Wilson