aboutsummaryrefslogtreecommitdiffstats
path: root/src/amd
Commit message (Expand)AuthorAgeFilesLines
* configure: check once for DRI3 dependenciesEmil Velikov2017-05-191-2/+1
* ac: add missing extern "C" guardsNicolai Hähnle2017-05-182-0/+16
* ac: add radeon_info::num_{sdma,compute}_ringsNicolai Hähnle2017-05-182-3/+15
* ac: add radeon_surf::htile_slice_sizeNicolai Hähnle2017-05-182-0/+6
* ac_surface: use radeon_info from ac_gpu_infoNicolai Hähnle2017-05-182-29/+29
* ac/radeonsi: move radeon_info initialization to amd/commonNicolai Hähnle2017-05-185-3/+289
* ac/radeonsi: move struct radeon_info to ac_gpu_info.hNicolai Hähnle2017-05-181-0/+93
* ac/radeonsi: move some aspects of sanity checking to ac_surfaceNicolai Hähnle2017-05-181-0/+33
* ac/radeonsi: add ac_compute_surface to automatically switch gfx6 vs. gfx9Nicolai Hähnle2017-05-182-16/+23
* ac/radeonsi: move the bulk of gfx9_surface_init to ac_surfaceNicolai Hähnle2017-05-182-0/+384
* ac/radeonsi: move the bulk of gfx6_surface_init to ac_surfaceNicolai Hähnle2017-05-182-0/+454
* ac/radeonsi: move amdgpu_addr_create to ac_surfaceNicolai Hähnle2017-05-185-0/+219
* ac/radeonsi: move surface definitions to new header ac_surface.hNicolai Hähnle2017-05-181-0/+178
* gallium/radeon: use a top-of-pipe timestamp for the start of TIME_ELAPSEDMarek Olšák2017-05-171-0/+11
* ac/debug: handle index field in SET_*_REG correctlyNicolai Hähnle2017-05-161-1/+7
* radeonsi/gfx9: add support for RavenMarek Olšák2017-05-151-0/+1
* amd/addrlib: import Raven supportMarek Olšák2017-05-153-3/+72
* radv: Save descriptor set even if vertex buffers are not saved.Bas Nieuwenhuizen2017-05-131-2/+3
* radeon: automake: remove unneeded elf Cflags/LibsEmil Velikov2017-05-111-0/+2
* Android: rework LLVM build supportRob Herring2017-05-111-1/+2
* Android: rework libelf dependenciesRob Herring2017-05-111-2/+4
* Android: amd/common: fix dependency on libmesa_nirRob Herring2017-05-111-1/+3
* Android: amd: use exported include dirs instead of explicit includesRob Herring2017-05-112-0/+9
* Android: remove remaining explicit libcxx includesRob Herring2017-05-111-1/+0
* radv: handle fragment shader srgb resolve pass betterDave Airlie2017-05-112-11/+38
* radv: always free nir shaders from modules on stackGrazvydas Ignotas2017-05-102-8/+4
* nir: Embed the shader_info in the nir_shader againJason Ekstrand2017-05-0912-111/+111
* ac: fix broken elimination of duplicated VS exportsMarek Olšák2017-05-081-14/+14
* radv: fix regression in blit2d push constant change.Dave Airlie2017-05-081-1/+1
* radv/meta: cleanup some unused code pathDave Airlie2017-05-082-34/+1
* radv/meta: port blit to using push constantsDave Airlie2017-05-081-100/+65
* radv/meta: move blit2d to using push constantsDave Airlie2017-05-082-75/+59
* radv/meta: move clear color to using push constantsDave Airlie2017-05-082-130/+78
* radv/meta: use novertex save path for resolve pass.Dave Airlie2017-05-082-2/+2
* radv: set base/ranges for push constant loads.Dave Airlie2017-05-086-0/+26
* radv: drop resolve hack workaroundsDave Airlie2017-05-073-104/+1
* radv/meta: select resolve pathsDave Airlie2017-05-071-8/+50
* radv/meta: add resolve pass using fragment/vertex shadersDave Airlie2017-05-076-2/+699
* radv: add subpass resolve compute pathDave Airlie2017-05-072-0/+94
* radv/resolve: split resolve emission out for computeDave Airlie2017-05-071-63/+78
* radv/meta: split out core part of resolve shaderDave Airlie2017-05-073-113/+140
* radv/meta: add srgb conversion to end of resolve shader.Dave Airlie2017-05-073-5/+61
* radv: set PERF_MOD in sample state like radeonsi.Dave Airlie2017-05-071-1/+2
* radv: apply the tess+GS hang workaround to Polaris12 as wellDave Airlie2017-05-071-1/+2
* radv/meta: fix restoring a push descriptor setFredrik Höglund2017-05-062-2/+7
* radv: enable POLARIS12 support.Dave Airlie2017-05-054-0/+8
* radeonsi: drop support for LLVM 3.8Marek Olšák2017-05-052-133/+53
* radeonsi: stop using v16i8Marek Olšák2017-05-051-1/+1
* radeonsi/gfx9: make some PA & DB registers match the closed Vulkan driverMarek Olšák2017-05-051-0/+4
* radv: don't advertise transfer props unless we can do anything elseDave Airlie2017-05-051-2/+2