summaryrefslogtreecommitdiff
path: root/src/amd/llvm
AgeCommit message (Expand)AuthorFilesLines
2021-06-22ac/llvm: don't set skip-uniform-regions to fix atomic.cmpswapMarek Olšák1-0/+3
2021-06-21nir,amd: Suffix nir_op_cube_face_coord/index with _amdJason Ekstrand1-4/+4
2021-06-20ac/llvm: add a callback to ac_cull_triangle to generate code in inner-most blockMarek Olšák2-4/+16
2021-06-09amd: Add extra source to the mbcnt_amd NIR intrinsic.Timur Kristóf3-3/+9
2021-06-08ac/llvm: implement byte/word extract/insert instructionsRhys Perry1-0/+25
2021-06-07nir: Rename WORK_GROUP (and similar) to WORKGROUPCaio Marcelo de Oliveira Filho1-2/+2
2021-06-07nir: Rename nir_intrinsic_load_local_group_size to nir_intrinsic_load_workgro...Caio Marcelo de Oliveira Filho1-1/+1
2021-05-25radeonsi: rewrite the prefix sum computation for shader cullingMarek Olšák2-64/+7
2021-05-25ac/llvm: don't draw the primitive for the dummy export workaround for Navi1xMarek Olšák1-1/+2
2021-05-25ac/llvm: set range metadata on mbcnt and deduplicate get_thread_idMarek Olšák1-17/+5
2021-05-25ac/llvm: allow ac_build_optimization_barrier with SGPRs, pointers, and metadataMarek Olšák3-14/+35
2021-05-25ac/llvm: expose set_range_metadata to more usersMarek Olšák2-3/+5
2021-05-25ac/llvm: set target features per function instead of per target machineMarek Olšák3-31/+21
2021-05-24amd: add Beige Goby supportMarek Olšák1-0/+1
2021-05-24amd: add Yellow Carp supportAaron Liu1-0/+1
2021-05-03radv,ac/llvm: use a dword alignment for descriptor loadsRhys Perry1-1/+2
2021-04-27radv,aco: use nir_address_format_vec2_index_32bit_offsetRhys Perry1-4/+9
2021-04-27radv: implement vulkan_resource_reindexRhys Perry1-14/+0
2021-04-23amd: Fix warnings around variable sizesJames Park1-3/+3
2021-04-23ac: move ac_lower_indirect_derefs() outside of the LLVM dirSamuel Pitoiset2-43/+0
2021-04-20ac/nir: set TRUNC_COORD=0 for nir_texop_tg4Rhys Perry1-0/+7
2021-04-16amd: remove some references to older LLVM versions in commentsMarek Olšák2-10/+1
2021-04-16amd: drop support for LLVM 10Marek Olšák2-11/+6
2021-04-16amd: drop support for LLVM 9Marek Olšák3-26/+1
2021-04-16amd: drop support for LLVM 8Samuel Pitoiset3-159/+43
2021-04-13ac/llvm: implement 16-bit packed VS outputs and FS inputsMarek Olšák3-8/+29
2021-04-08nir: Extract shader_info->cs.shared_size out of union.Bas Nieuwenhuizen1-1/+1
2021-04-02ac/llvm: don't set unsupported xnack options to fix LLVM crashes on gfx6-8Marek Olšák2-5/+1
2021-03-18ac/llvm: Fix alignment of shared load intrinsics.Timur Kristóf1-2/+2
2021-03-17ac/llvm: Emit more efficient code for load_shared.Timur Kristóf1-9/+7
2021-03-17ac/llvm: Add constant offset to load/store_shared.Timur Kristóf1-4/+7
2021-03-17ac/llvm: Make sure to always emit integer comparison for nir_op_ieq.Timur Kristóf1-11/+2
2021-03-17ac/llvm: Make shared loads/stores work correctly for non-CS stages.Timur Kristóf1-1/+5
2021-03-17ac/llvm: Implement new Geometry Shader intrinsics.Timur Kristóf2-0/+17
2021-03-17ac/llvm: Implement the new tessellation intrinsics.Timur Kristóf2-0/+28
2021-03-17ac/llvm: Implement AMD-specific buffer load/store intrinsics.Timur Kristóf1-0/+59
2021-03-17ac/llvm: handle demote in LLVM 13 that just added support for itMarek Olšák2-3/+20
2021-03-10ac/llvm: unpack thread IDs on AldebaranMarek Olšák1-1/+13
2021-03-10ac,radeonsi: add sampler changes for AldebaranMarek Olšák3-1/+9
2021-03-10amd: add Aldebaran chip enumJames Zhu1-0/+2
2021-03-03ac/llvm: open code fpow on LLVM 12 using fmul.legacyMarek Olšák1-0/+11
2021-03-03ac/llvm: add type parameter into ac_build_buffer_load to fix 16-bit TES inputsMarek Olšák3-9/+10
2021-03-03ac/llvm: fix visit_load_ubo_buffer to use SMEM for 16 bits instead of VMEMMarek Olšák1-20/+18
2021-03-03ac/llvm: implement 16-bit and 64-bit fpow correctlyMarek Olšák1-0/+9
2021-03-03ac/llvm: add support for 16-bit source operands for samplersMarek Olšák3-13/+57
2021-03-03ac/llvm: fix ac_build_atomic_rmw with LLVM 13Marek Olšák1-0/+6
2021-01-27radv,ac/nir: implement non-uniform get_ssbo_sizeRhys Perry2-7/+8
2021-01-08ac/nir: implement sparse image/texture loadsRhys Perry3-18/+126
2021-01-08ac/nir: implement nir_op_vec5Rhys Perry1-0/+2
2021-01-05ac/nir: use llvm.readcyclecounter for LLVM9+Rhys Perry1-2/+3
generated by cgit v1.2.3 (git 2.39.1) at 2024-04-28 05:50:03 +0000