aboutsummaryrefslogtreecommitdiffstats
path: root/src/amd
Commit message (Expand)AuthorAgeFilesLines
* nir: Make "divergent" a property of an SSA valueJason Ekstrand2020-05-132-31/+28
* aco: improve phi affinities with p_split_vectorRhys Perry2020-05-131-0/+19
* aco: consider affinities when creating v_mac_f32Rhys Perry2020-05-131-2/+8
* aco: mark phi definitions as last-seen phi operandsRhys Perry2020-05-131-14/+14
* aco: fix consecutively written vgprs from vmem instructionsRhys Perry2020-05-131-10/+26
* aco: simplify consecutive ordered vmem/lds writes optimizationRhys Perry2020-05-131-10/+2
* radv: add a LLVM version string workaround for SotTR and ACOSamuel Pitoiset2020-05-131-3/+22
* radv: use the common code for generating extensions and dispatch tablesSamuel Pitoiset2020-05-132-348/+13
* radv: implement VK_EXT_private_dataSamuel Pitoiset2020-05-132-0/+52
* radv: use the base object struct typesSamuel Pitoiset2020-05-1312-7/+138
* radv: use the common base object type for VkDeviceSamuel Pitoiset2020-05-1312-90/+87
* aco: optimize add/sub(a, cndmask(b, 0, 1, cond)) -> addc/subbrev_co(0, a, b)Samuel Pitoiset2020-05-121-13/+79
* aco: coalesce parallelcopies during register allocationDaniel Schürmann2020-05-121-6/+3
* radv: update the list of allowed Android extensionsSamuel Pitoiset2020-05-121-0/+13
* radv: handle different Vulkan API versions correctlySamuel Pitoiset2020-05-121-12/+43
* radv: limit the Vulkan version to 1.1 for AndroidSamuel Pitoiset2020-05-121-2/+7
* amd/common: Fix incorrect use of asprintf instead of vasprintfSerge Martin2020-05-111-2/+2
* ac/surface: remove shadowing declarationPierre-Eric Pelloux-Prayer2020-05-111-2/+2
* aco: prevent invalid loads/stores vectorization if robustness is enabledSamuel Pitoiset2020-05-111-2/+10
* nir: do not vectorize load/store if offset can overflow and robustness enabledSamuel Pitoiset2020-05-111-1/+2
* aco: fix 64-bit trunc with negative exponents on GFX6Samuel Pitoiset2020-05-111-2/+3
* ac/surface: fix broken pitch override on gfx8Marek Olšák2020-05-081-3/+5
* radv: report correct backend IR in hang reports when ACO is usedSamuel Pitoiset2020-05-081-1/+3
* radv: do not print the LLVM version string twice in hang reportsSamuel Pitoiset2020-05-081-2/+1
* radv: remove the LLVM version string when ACO is usedSamuel Pitoiset2020-05-081-4/+5
* aco: either copy-propagate or inline create_vector operandsDaniel Schürmann2020-05-071-4/+6
* ac/surface: override all offsets including metadata offsetsMarek Olšák2020-05-071-0/+15
* amd: unify code for overriding offset and stride for imported buffersMarek Olšák2020-05-073-21/+34
* ac/surface,radeonsi: move the set/get_umd_metadata code into ac_surface.cMarek Olšák2020-05-072-0/+176
* ac/surface,radeonsi: move the set/get_bo_metadata code to ac_surface.cMarek Olšák2020-05-072-0/+120
* amd: assume HTILE is always rb/pipe_aligned, remove ac_surface.u.gfx9.htileMarek Olšák2020-05-074-14/+16
* amd: assume CMASK is always rb/pipe_aligned, remove ac_surface.u.gfx9.cmaskMarek Olšák2020-05-074-22/+18
* amd: remove duplicated definitions from amdgpu_drm.hMarek Olšák2020-05-071-3/+0
* radv/winsys: do not count visible VRAM buffers twice in the budgetSamuel Pitoiset2020-05-062-17/+33
* radv: display an error message if the winsys init failedSamuel Pitoiset2020-05-061-1/+2
* radv: use a linked list for physical devicesSamuel Pitoiset2020-05-062-52/+69
* radv: don't report error with other vendor DRM devicesSamuel Pitoiset2020-05-061-9/+15
* radv: report INITIALIZATION_FAILED when the amdgpu winsys init failedSamuel Pitoiset2020-05-061-1/+1
* radv: fix a memleak if the physical device initialization failedSamuel Pitoiset2020-05-061-6/+8
* radv: rename radv_devices() to radv_enumerate_physical_devices()Samuel Pitoiset2020-05-061-3/+3
* radv: cleanup radv_CreateInstance()Samuel Pitoiset2020-05-061-35/+22
* ac/surface: unset RADEON_SURF_TC_COMPATIBLE_HTILE if HTILE hasn't been computedMarek Olšák2020-05-051-0/+8
* ac/nir: export some undef as zeroPierre-Eric Pelloux-Prayer2020-05-052-6/+36
* radv: enable shaderInt16 unconditionally with LLVM and only GFX8+ with ACOSamuel Pitoiset2020-05-051-1/+1
* radeonsi: add workaround for issue 2647Pierre-Eric Pelloux-Prayer2020-05-052-3/+26
* aco: remove unecessary p_split_vector with v2b reg classSamuel Pitoiset2020-05-051-80/+33
* radv: Remove RANGE_SIZE usageJoshua Ashton2020-05-055-12/+26
* ac/surface: fix MSAA crash with FORCE_SWIZZLE_MODE on gfx9Marek Olšák2020-05-041-1/+3
* Revert "ac: reassociate FP expressions for inexact instructions for radeonsi"Marek Olšák2020-05-041-9/+0
* radeonsi: fix export countPierre-Eric Pelloux-Prayer2020-05-041-9/+7