summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/r600/r600_hw_context.c
Commit message (Expand)AuthorAgeFilesLines
* r600g: use last_gfx_fence like radeonsiMarek Olšák2016-08-031-3/+12
* gallium/radeon: remove RADEON_FLUSH_KEEP_TILING_FLAGS flagMarek Olšák2016-07-191-3/+0
* r600g: don't flush caches when binding shader resourcesMarek Olšák2016-06-041-0/+4
* r600g: only do necessary cache flushes in cp_dma_copy_bufferMarek Olšák2016-06-041-14/+1
* r600g: fix CP DMA hazard with index buffer fetches (v3)Marek Olšák2016-06-041-1/+68
* r600g: properly sync CP with CP DMA on R6xxMarek Olšák2016-06-041-1/+8
* r600g: write WAIT_UNTIL in the correct placeMarek Olšák2016-06-041-8/+11
* radeon/winsys: introduce radeon_winsys_cs_chunkNicolai Hähnle2016-06-011-1/+2
* gallium/radeon: use cs_check_space throughoutNicolai Hähnle2016-06-011-4/+2
* gallium/radeon: add radeon_emitted to check for non-trivial IBsNicolai Hähnle2016-05-171-4/+2
* gallium/radeon: use radeon_emitNicolai Hähnle2016-05-171-18/+18
* gallium/radeon: flush if DMA IB memory usage is too highMarek Olšák2016-05-101-1/+1
* gallium/radeon: fix (S)DMA read-after-write hazardsMarek Olšák2016-05-101-0/+1
* r600g: use common scissor and viewport codeMarek Olšák2016-04-121-6/+4
* gallium/radeon: remove R600_QUERY_HW_FLAG_TIMERMarek Olšák2016-04-121-1/+1
* gallium/radeon: merge timer and non-timer query listsMarek Olšák2016-04-121-2/+1
* r600g: add pausing pipeline & streamout queries into set_active_query_stateMarek Olšák2016-04-121-0/+10
* gallium/radeon: remove old CS tracingMarek Olšák2016-03-201-9/+2
* r600g: remove support for DRM < 2.12.0Marek Olšák2016-03-011-3/+1
* r600/eg: workaround bug with tess shader and dynamic GPRs.Dave Airlie2015-12-071-1/+1
* r600: add shader key entries for tcs and tes.Dave Airlie2015-12-071-0/+4
* r600: move to using hw stages array for hw stage atomsDave Airlie2015-12-071-4/+4
* gallium/radeon: atomize render condition (SET_PREDICATION)Marek Olšák2015-11-131-0/+1
* gallium/radeon: simplify restoring render condition after flushMarek Olšák2015-11-131-5/+0
* gallium/radeon: inline the r600_rings structureMarek Olšák2015-11-131-17/+17
* gallium/radeon: remove the IB flushing flagMarek Olšák2015-11-131-3/+0
* gallium/radeon: move GFX/DMA flushing from add_to_buffer_list to need_cs_spaceMarek Olšák2015-11-131-0/+5
* gallium/radeon: tell the winsys the exact resource binding typesMarek Olšák2015-10-031-4/+4
* r600g: simplify dirty atom trackingGrazvydas Ignotas2015-09-031-5/+4
* r600g: make all viewport states use single atomGrazvydas Ignotas2015-09-031-4/+4
* r600g: make all scissor states use single atomGrazvydas Ignotas2015-09-031-1/+3
* gallium/radeon: rename r600_context_bo_reloc -> radeon_add_to_buffer_listMarek Olšák2015-09-011-4/+4
* gallium/radeon: rename write_*_reg functionsMarek Olšák2015-09-011-2/+2
* r600g: use a bitfield to track dirty atomsGrazvydas Ignotas2015-08-111-6/+6
* r600g: don't mark unused atom dirtyGrazvydas Ignotas2015-08-111-1/+3
* gallium/radeon: use helper functions to mark atoms dirtyGrazvydas Ignotas2015-08-111-27/+27
* gallium/radeon: allow the winsys to choose the IB sizeMarek Olšák2015-08-061-1/+1
* gallium/radeon: suspend timer queries between IBsMarek Olšák2015-08-061-1/+2
* Fix a few typosZoë Blade2015-04-271-1/+1
* r600g: Use R600_MAX_VIEWPORTS instead of 16Alexandre Demers2015-03-091-1/+1
* r600g: add support for primitive id without geom shader (v2)Dave Airlie2015-01-281-1/+1
* r600g: move r6xx-specific streamout flush flagging into r600gMarek Olšák2015-01-071-3/+6
* r600g: use gpu_address from r600_resourceMarek Olšák2014-08-091-2/+2
* r600g: fix hang on RV740 by using DX_RASTERIZATION_KILL instead of SX_MISCMarek Olšák2014-04-251-2/+2
* r600g: fix flushing on RV670, RS780, RS880 againMarek Olšák2014-04-251-0/+9
* r600g,radeonsi: don't skip the context flush if a fence should be returnedMarek Olšák2014-04-181-1/+1
* r600g,radeonsi: share some of gfx flush codeMarek Olšák2014-04-161-41/+3
* r600g: merge r600_flush with r600_context_flushMarek Olšák2014-04-161-2/+29
* gallium/radeon: create and return a fence in the flush functionMarek Olšák2014-04-161-4/+5
* r600g: use more appropriate names for async DMA functionsMarek Olšák2014-03-201-6/+6