summaryrefslogtreecommitdiff
path: root/src/sna/sna_accel.c
AgeCommit message (Expand)Author
2013-06-29sna: Fix get_image_inplace to use the pixmap offsetChris Wilson
2013-06-29sna: Move the clone discard into free-gpuChris Wilson
2013-06-29sna: Improve checks for coherent access through CPU mappingsChris Wilson
2013-06-29sna: Promote assert(!priv->mapped) along migration pathsChris Wilson
2013-06-29sna: Add asserts around applying clearsChris Wilson
2013-06-29sna: Use inplace CPU mapping readback for GetImage on linear buffersChris Wilson
2013-06-28sna: Add debug control for disabling accelerated GetImageChris Wilson
2013-06-28sna: Assert that the kernel tiling mode matches our boChris Wilson
2013-06-27sna: Prefer operating inplace with a very large GPU boChris Wilson
2013-06-27sna: Add a fast path for reading back from tiled X boChris Wilson
2013-06-27sna: Support operations inplace on CPU mappings of a regionChris Wilson
2013-06-27sna: We can read from cloned pixmaps inplace - so long as we don't writeChris Wilson
2013-06-27sna: Tweak ordering of userptr temporary mappings for uploadsChris Wilson
2013-06-27sna: Fix DBG printing of can_upload_tiled_x()Chris Wilson
2013-06-27sna: Refactor freeing gpu_bo in manual tiled uploadChris Wilson
2013-06-27sna: Detect and handle cloned pixels for manual tiled uploadsChris Wilson
2013-06-26sna: Clear mapped state after performing manual tilingChris Wilson
2013-06-26sna: Do not force creation of CPU maps on !llcChris Wilson
2013-06-26sna: Rearrange tiled x upload so that damage accumulation is lastChris Wilson
2013-06-26sna: Optimize clears to whiteChris Wilson
2013-06-25sna: Do not perform tiling inplace if the destination is busyChris Wilson
2013-06-25sna: Be explicit when creating tiled-x pixmaps for manual tiling uplaodsChris Wilson
2013-06-25sna: Don't flag IGNORE_CPU for partial overwritten blocksChris Wilson
2013-06-24sna: Initialise gc->ops->PolyPoint for miWideDashChris Wilson
2013-06-24sna: Clear 'clear' hint upon uploading into tiledChris Wilson
2013-06-23sna: Discard proxy upload buffers before overwritting for PutImageChris Wilson
2013-06-23sna: Discard overwritten incomplete operations before migratingChris Wilson
2013-06-23sna: Allow tiled uploads to accumulate damageChris Wilson
2013-06-22sna: Refine the self-cow copy hack to reduce collateral damageChris Wilson
2013-06-21sna: Determine swizzling once during initialisation and choose memcpy_to_tiled_xChris Wilson
2013-06-21sna: Allow PutImage to write inplace using manual tilingChris Wilson
2013-06-21sna: Check if we may want to simply upload for a CopyAreaChris Wilson
2013-06-21sna: Fix inspection of transfer extents for deciding transportChris Wilson
2013-06-20sna: Mark overwriting CopyArea as not needing the dst damageChris Wilson
2013-06-20sna: Promote the CopyArea to the GPU if it subsumes the CPU damageChris Wilson
2013-06-19sna: Tidy a redundant conditional for CopyAreaChris Wilson
2013-06-19Revert "sna: Apply scanout stride limits to tiling selection"Chris Wilson
2013-06-19sna: Apply scanout stride limits to tiling selectionChris Wilson
2013-06-18sna: Allow wedged pixmaps to be accessed via GTTChris Wilson
2013-06-18sna: Specialise mi calls to PolyFillRectChris Wilson
2013-06-18sna: Add extra assertions to sanity check CPU access is coherentChris Wilson
2013-06-18sna: Mark the base bo as unreusable for temporary mappingsChris Wilson
2013-06-17sna: Refactor scanout flushingChris Wilson
2013-06-15sna: Be explicit in not marking an upload buffer as being on the cpuChris Wilson
2013-06-14sna: Tighten a few asserts on pixmap migrationChris Wilson
2013-06-12sna: Do not enlarge the read back region if there is already CPU damageChris Wilson
2013-06-11sna: Make sure the source is coherent on the CPU before uploadingChris Wilson
2013-06-10sna: Move the GTT dirty tracking from the pixmap to the boChris Wilson
2013-06-10sna: Make sure all sna_write_boxes* flag the GTT is dirtyChris Wilson
2013-06-07sna: Set the needs_flush flag on the bo when the scanout is dirtyChris Wilson