summaryrefslogtreecommitdiff
Commit message (Expand)AuthorAgeFilesLines
* winsys/amdgpu: handle fence and dependencies mergeamdgpuChristian König2015-07-091-3/+3
* winsys/radeon: fix build breakage since 248f552Marek Olšák2015-07-091-1/+1
* winsys/amdgpu: switch to new amdgpu_cs_query_fence_status interfaceJammy Zhou2015-07-091-3/+4
* winsys/amdgpu: query the GART page size from the kernelMarek Olšák2015-07-083-8/+18
* winsys/amdgpu: set the ring type at CS initilizationMarek Olšák2015-07-081-26/+27
* gallium/radeon: add RING_COMPUTE instead of RADEON_FLUSH_COMPUTEMarek Olšák2015-07-083-8/+9
* winsys/amdgpu: add ip_instance into amdgpu_fenceMarek Olšák2015-07-082-7/+12
* radeonsi: don't count the exact needed CS space if the CS is large enoughMarek Olšák2015-07-081-2/+11
* winsys/amdgpu: allow maximum IB size of 4 MBMarek Olšák2015-07-081-2/+4
* gallium/radeon: allow the winsys to choose the IB sizeMarek Olšák2015-07-0812-23/+26
* winsys/amdgpu: use safer access to amdgpu_fence_wait::signalledMarek Olšák2015-07-082-6/+10
* winsys/amdgpu: simplify waiting on a variable in amdgpu_fence_waitMarek Olšák2015-07-081-25/+12
* winsys/amdgpu: track fences per ring and be thread-safeMarek Olšák2015-07-086-42/+122
* gallium/radeon: unify buffer_wait and buffer_is_busy in the winsys interfaceMarek Olšák2015-07-0810-80/+58
* winsys/amdgpu: use amdgpu_device_handle as a unique device ID instead of fdMarek Olšák2015-07-082-43/+42
* winsys/amdgpu: remove useless flushing and waiting in amdgpu_bo_set_tilingMarek Olšák2015-07-081-11/+0
* winsys/amdgpu: reduce BO cache timeoutMarek Olšák2015-07-081-1/+1
* winsys/amdgpu: align BO size to page sizeMarek Olšák2015-07-081-0/+6
* winsys/amdgpu: remove amdgpu_winsys::num_cpusMarek Olšák2015-07-082-5/+1
* radeonsi: don't flush an empty IB if the only thing we need is a fenceMarek Olšák2015-07-083-3/+15
* gallium/radeon: enable the GPU load query for amdgpuMarek Olšák2015-07-081-1/+3
* gallium/os: add conversion and wait functions for absolute timeoutsMarek Olšák2015-07-082-0/+67
* gallium/os: add os_wait_until_zero (v2)Marek Olšák2015-07-082-1/+48
* radeon/llvm: Handle LLVM backend rename from R600 to AMDGPUTom Stellard2015-07-072-5/+16
* winsys/amdgpu: use dependencies instead of waiting for last fence v2Christian König2015-06-292-1/+25
* radeonsi: fix a hang with DrawTransformFeedback on 4 SE chipsMarek Olšák2015-06-261-0/+4
* radeon/vce: implement VCE dual instance supportLeo Liu2015-06-263-5/+30
* st/omx/enc: stack frame tasks for the gatheringLeo Liu2015-06-262-2/+22
* radeon/video: config encode stacked frame number based on HWLeo Liu2015-06-261-0/+2
* vl: add cap for stacking framesLeo Liu2015-06-261-1/+2
* radeon/vce: make reloc offset signedChristian König2015-06-264-6/+6
* radeon/vce: add config task and put task info into encoder v2Leo Liu2015-06-264-33/+47
* st/omx/enc: flush after eos handling v2Leo Liu2015-06-261-0/+1
* winsys/amdgpu: sync buffer accesses to different ringsChristian König2015-06-191-0/+13
* winsys/amdgpu: require LLVM 3.6.1 for VI because of bug fixes thereMarek Olšák2015-06-161-3/+5
* configure.ac: rename LLVM_VERSION_PATCH to avoid conflict with llvm-config.hMarek Olšák2015-06-161-1/+1
* Revert "radeon/llvm: require LLVM 3.6.1 because it has stable VI support"Marek Olšák2015-06-162-1/+8
* Revert "radeon/llvm: drop support for LLVM < 3.6"Marek Olšák2015-06-1610-19/+308
* radeon/vce: fix VCE fail after rebaseLeo Liu2015-06-161-8/+6
* radeon/vce: add dual pipe support for VILeo Liu2015-06-154-22/+21
* radeonsi: properly handler raster_config setup on CZAlex Deucher2015-06-111-1/+1
* radeonsi: properly set the raster_config for KVAlex Deucher2015-06-111-5/+9
* radeon/llvm: drop support for LLVM < 3.6Marek Olšák2015-06-1110-308/+19
* radeon/llvm: require LLVM 3.6.1 because it has stable VI supportMarek Olšák2015-06-112-8/+1
* radeon/vce: add new firmware support for VI and CILeo Liu2015-06-111-1/+7
* gallium/radeon: fix VCE build after rebaseMarek Olšák2015-06-101-4/+4
* winsys/amdgpu: don't use VRAM with APUs that don't have much of itMarek Olšák2015-06-102-1/+15
* winsys/amdgpu: don't use KMS handles as reloc hash keysMarek Olšák2015-06-104-21/+7
* winsys/amdgpu: add IBs to the buffer list, adapt to interface changesMarek Olšák2015-06-101-5/+4
* winsys/amdgpu: allocate IBs like normal buffersMarek Olšák2015-06-102-11/+52
View Lorry Controller Status