summaryrefslogtreecommitdiff
path: root/src/amd
AgeCommit message (Expand)AuthorFilesLines
2021-07-22aco: don't create v_madmk_f32/v_madak_f32 from v_fma_legacy_f16Rhys Perry1-0/+5
2021-07-22aco/optimizer: ensure to not erase high bits when propagating packed constantsDaniel Schürmann1-1/+2
2021-07-22radv: only init the TC-compat ZRANGE metadata for the depth aspectSamuel Pitoiset1-1/+1
2021-07-13radeonsi,radv: fix a late alloc deadlock with <= 6 CUs per SAMarek Olšák1-2/+0
2021-07-13ac/surface/tests: fix the ARM buildMarek Olšák1-4/+1
2021-07-13aco: fix shared_atomic_comp_swap if the second source isn't a VGPRSamuel Pitoiset1-2/+4
2021-07-13ac/surface: Handle non-retiled displayable DCC correctly for modifiers.Bas Nieuwenhuizen1-6/+13
2021-07-13Convert most remaining free-form fall-through comments to FALLTHROUGHMichel Dänzer4-5/+5
2021-07-13radv: disable DCC for DOOM 2016 and Wolfenstein IISamuel Pitoiset1-1/+5
2021-06-30ac/surface: don't print stencil info if tex has no stencilPierre-Eric Pelloux-Prayer1-1/+1
2021-06-26amd/addrlib: remove Meson debug message()Simon Ser1-2/+0
2021-06-26aco/gfx10: Emit barrier at the start of NGG VS and TES.Timur Kristóf1-0/+7
2021-06-26aco/gfx10: NGG zero output workaround for conservative rasterization.Timur Kristóf1-2/+8
2021-06-26radv/llvm: Emit s_barrier at the beginning of NGG non-GS shaders.Timur Kristóf1-0/+4
2021-06-26ac/nir: Update TCS output barriers with nir_var_mem_shared.Timur Kristóf1-4/+23
2021-06-18radv: always decompress both aspects of a depth/stencil imageSamuel Pitoiset2-17/+1
2021-06-18radv: reject binding buffer/image when the device memory is too smallSamuel Pitoiset1-2/+28
2021-06-16radv: Allow DCC images to be compressed with foreign queues.Bas Nieuwenhuizen1-0/+3
2021-06-16radv: Actually return correct value for read-only DCC compressedness.Bas Nieuwenhuizen1-1/+1
2021-06-16radv: Don't skip barriers that only change queues.Bas Nieuwenhuizen1-3/+3
2021-06-16radv: fix aligning the image offset by using align64()Samuel Pitoiset1-1/+1
2021-06-10radv: do not launch an IB2 for secondary cmdbuf with INDIRECT_MULTI on GFX7Samuel Pitoiset4-49/+94
2021-06-10aco: use v1b/v2b for ds_read_u8/ds_read_u16Rhys Perry2-4/+6
2021-06-10aco: fix emitting literal offsets with SMEM on GFX7Samuel Pitoiset1-2/+4
2021-06-10aco: fix range checking for SSBO loads/stores with SGPR offset on GFX6-7Samuel Pitoiset2-0/+19
2021-06-10aco/ra: Fix off-by-one-error in print_regsTony Wasserka1-3/+2
2021-06-10radv: emit PA_SC_CONSERVATIVE_RASTERIZATION_CNTL only on GFX9+Samuel Pitoiset1-19/+21
2021-06-05aco: don't create 4 and 5 dword NSA instructions on GFX10Rhys Perry2-1/+10
2021-06-05aco: do not clause NSA instructionsRhys Perry4-5/+12
2021-05-31radv: add radv_absolute_depth_biasRhys Perry2-21/+24
2021-05-31aco: fix emitting discard when the program just endsSamuel Pitoiset1-5/+2
2021-05-31ac/gpu_info: set has_zero_index_buffer_bug for Navi12 tooMarek Olšák1-2/+2
2021-05-31radv: fix fast clearing DCC if one level can't be compressed on GFX10+Samuel Pitoiset1-11/+18
2021-05-31radv: Fix compatible image handle type for dmabufs.Georg Lehmann1-1/+1
2021-05-31aco: fix derivatives/intrinsics with SGPR sourcesSamuel Pitoiset1-1/+16
2021-05-19aco: disallow SGPRs on DPP instructionsRhys Perry2-1/+3
2021-05-18radv: Use correct border swizzle on GFX9+.Bas Nieuwenhuizen1-9/+9
2021-05-18aco/ra: initialize temp_in_scc earlierRhys Perry1-1/+2
2021-05-18aco/ra: also prevent overflow register for p_create_vector operandsDaniel Schürmann1-7/+7
2021-05-18radv: Handle unnormalized samplers in YCbCr loweringJoshua Ashton1-9/+32
2021-05-18radv: call nir_copy_prop() after load-store vectorizationDaniel Schürmann1-0/+1
2021-05-18aco/ra: prevent underflow register for p_create_vector operandsDaniel Schürmann1-6/+11
2021-05-18aco/scheduler: Fix register demand computation for upwards movesTony Wasserka1-6/+6
2021-05-18aco/scheduler: Fix register demand computation for downwards movesTony Wasserka1-3/+15
2021-05-18aco: fix additional register requirements for spillingDaniel Schürmann1-13/+24
2021-05-18radv: fix extending the dirty bits to 64-bitSamuel Pitoiset2-4/+4
2021-05-12radv: fix missing ITERATE_256 for D/S MSAA images that are TC-compat HTILESamuel Pitoiset6-33/+33
2021-05-12nir: Move shared_memory_explicit_layout bit into common shader_infoCaio Marcelo de Oliveira Filho1-1/+1
2021-05-03radv,ac/llvm: use a dword alignment for descriptor loadsRhys Perry2-1/+5
2021-05-03radv: Only require DRM 3.23.Bas Nieuwenhuizen2-14/+33
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-25 23:28:56 +0000