summaryrefslogtreecommitdiff
path: root/src/sna/sna_accel.c
AgeCommit message (Expand)Author
2014-03-19sna: Assert after applying clipping that the draw rectangle is wholly containedChris Wilson
2014-03-14sna: Ensure we flush SHM pixmaps if rendering to a GPU shadowChris Wilson
2014-03-11sna: Avoid using the wrong pitch for comparing replacement sizesChris Wilson
2014-03-11sna: Assert that the pixmap pitch is initialised before useChris Wilson
2014-03-10sna: Unroll finding bbox of pointsChris Wilson
2014-03-10sna: Emit points using the BLT primitive when appropriateChris Wilson
2014-02-27sna: Limit the WHOLE_HINT to large PutImagesChris Wilson
2014-02-27sna: Avoid promoting region-to-whole migration and discarding damageChris Wilson
2014-02-23sna: Avoid signed overflow when printing allocated bytes in DBGChris Wilson
2014-02-22sna: Allow SIGBUS/SEGV to kill the threadChris Wilson
2014-02-22sna: Assert that sigtrap is activeChris Wilson
2014-02-21sna: Use a hint to do whole image uploads inplaceChris Wilson
2014-02-21sna: Allow allocation to fail even when debuggingChris Wilson
2014-02-12sna: Delete redundant branchChris Wilson
2014-02-11sna: Defer move-to-gpu until we need to transfer GPU damage to CPUChris Wilson
2014-02-10sna: Remove damaged region after converting CPU bo to GPU boChris Wilson
2014-02-10sna: Check for overwriting damage before applying clearChris Wilson
2014-02-10sna: Clear GPU damage flag for 1x1 writes onto a clear bufferChris Wilson
2014-02-06sna: And clear drawable->id on reused pixmapsChris Wilson
2014-02-06sna: Add some DBG to clarify the source pixmap (for tiles)Chris Wilson
2014-02-06sna: Reset composite offsets when reusing freed pixmapsChris Wilson
2014-02-05sna: Add some DBG around tiled bltsChris Wilson
2014-02-05sna: Pass read hints from move-to-cpu to wait_for_shadowChris Wilson
2014-02-02sna: Reconstruct damage for the partially replaced discarded CPU boChris Wilson
2014-02-01sna: If IGNORE_CPU is not set we must mark the move as MOVE_READChris Wilson
2014-02-01sna: Skip discarding CPU bo when using as a render targetChris Wilson
2014-02-01sna: Allow more inplace promotions of CPU to GPU boChris Wilson
2014-02-01sna: Undo region translation before returningChris Wilson
2014-02-01sna: Apply drawable offsets to partial CPU damage being replacedChris Wilson
2014-02-01sna: Don't discard the GPU prior to using it for stashing CPU damageChris Wilson
2014-02-01sna: Remark the region as damaged after uploading the rest to the GPUChris Wilson
2014-02-01sna: Only discard CPU damage for an replacing regionChris Wilson
2014-01-31sna: remove short-circuit for move-to-CPU when damage covers regionChris Wilson
2014-01-31sna: Balance memory accounting for buffer objectsChris Wilson
2014-01-31sna: Reorder assignments vs DBGChris Wilson
2014-01-31sna: Actually reuse pixmap headers for ordinary pixmapsChris Wilson
2014-01-31sna: Tidy up pixmap header allocationsChris Wilson
2014-01-31sna: Show the memory debug info againChris Wilson
2014-01-30sna: Constify GCFunc to quieten the compiler.Chris Wilson
2014-01-29sna: Do not use color corrected pixel values for comparing against gcChris Wilson
2014-01-21sna: Restore gc->serialNumber correctly after falling backChris Wilson
2014-01-21sna: Drop bogus assert added in recent patchesChris Wilson
2014-01-21sna: Reduce zeal of new gc->pCompositeClipChris Wilson
2014-01-21sna: Fix typo in assertion from previous commitChris Wilson
2014-01-21sna: More assertions for tracking gc->pCompositeClipChris Wilson
2014-01-20sna: Include serial numbers in ValidateGC DBGChris Wilson
2014-01-20sna: Short-cut the fallback for XCopyArea with depth < 8Chris Wilson
2014-01-20sna: Add some more DBG output around the clipping in sna_do_copy()Chris Wilson
2014-01-17sna: Nullify pixmap->devPrivate.ptr after promoting CPU bo to GPUChris Wilson
2014-01-17sna: Intensify the move-to-cpu pointer paranoiaChris Wilson