aboutsummaryrefslogtreecommitdiff
path: root/mlir/lib/Dialect/GPU
AgeCommit message (Expand)AuthorFilesLines
6 hours[mlir][gpu] Validate argument count in gpu.launch parser (#180388)Longsheng Mou1-12/+17
5 days[MLIR][XeVM] Remove xevm to llvm from convert to llvm (#175672)Sang Ik Lee1-0/+3
5 days[mlir][gpu][pipeline] Add missing passes in XeGPU to XeVM pipeline (#179320)Charitha Saumya1-7/+17
6 days[milr][gpu] Make barrier elimination address-space aware (#178101)Krzysztof Drewniak1-69/+173
8 days[MLIR] Apply clang-tidy fixes for bugprone-argument-comment in PromoteShuffle...Mehdi Amini1-1/+2
11 days[MLIR][GPU] Update serializeToObject to use SerializedObject wrapper and incl...Zichen Lu2-1/+3
13 days[mlir] Fix new clang-tidy warning llvm-type-switch-case-types. NFC. (#178487)Jakub Kuderski3-4/+4
13 days[mlir] Update all-reduce (& vector tests) to use workgroup barriers (#178285)Krzysztof Drewniak2-3/+24
14 days[MLIR] Apply clang-tidy fixes for llvm-else-after-return in EliminateBarriers...Mehdi Amini1-1/+2
2026-01-26[mlir][gpu] Add address space modifier to gpu.barrier (#177425)Krzysztof Drewniak1-4/+31
2026-01-25[MLIR] Fix GCC's `-Wreturn-type` warnings (#177654)foxtran1-0/+2
2026-01-23[mlir][gpu] Enforce async keyword when parsing gpu.launch with results (#176570)Akimasa Watanuki1-1/+6
2026-01-22[MLIR][GPU][NVVM] Add verify-target-arch option to nvvm-attach-target pass (#...Srinivasa Ravi1-1/+2
2026-01-21[MLIR][GPU] Make sure to propagate known cluster sizes in kernel outlining (#...Adam Paszke1-0/+4
2026-01-16[mlir] Fix build after #175815 (#176332)Matthias Springer1-1/+1
2026-01-16[mlir][Interfaces] Split successor inputs from region successor (#175815)Matthias Springer1-1/+4
2026-01-14[MLIR][XeGPU] Add 2D `vector.multi_reduction` optimization (#171154)Artem Kroviakov1-0/+3
2026-01-14[mlir][Interfaces][NFC] Simplify and align `RegionSuccessor` design / API (#1...Matthias Springer1-1/+1
2026-01-12 [mlir][Utils] Add VerificationUtils (NFC) (#174336)Nick Kreeger1-3/+4
2026-01-12[mlir][amdgpu] Remove redundant barriers (#175436)Ivan Butygin1-6/+2
2026-01-06[MLIR] Propagate known cluster sizes from gpu.launch to gpu.func (#174404)Adam Paszke2-1/+21
2026-01-02[mlir][gpu] Fold `subgroup_broadcast(subgroup_broadcast(%val))` chains (#174159)Ivan Butygin1-0/+8
2025-12-14[mlir][nvgpu] Move memref memspace attributes conversion to single place (#17...Ivan Butygin1-20/+3
2025-12-13[mlir][gpu] Use `arith` dialect to lower gpu.global_id (#171614)Longsheng Mou1-6/+8
2025-12-11[mlir][amdgpu] Move GPU memory spaces conversion to single place (#171876)Ivan Butygin2-15/+6
2025-12-01[MLIR][GPU] subgroup_mma fp64 extension - take 2 (#169061)Giacomo Castiglioni1-2/+2
2025-11-26Reland: [GPUToXeVMPipeline][Pipeline] Modify pipeline to add `convert-vector-...Md Abdullah Shahneous Bari1-0/+3
2025-11-25Revert "[GPUToXeVMPipeline][Pipeline] Modify pipeline to add `convert-vector-...Md Abdullah Shahneous Bari1-3/+0
2025-11-25[mlir] Fix build failure with BUILD_SHARED_LIBS=ONKrzysztof Parzyszek1-0/+1
2025-11-25[GPUToXeVMPipeline][Pipeline] Modify pipeline to add `convert-vector-to-llvm`...Md Abdullah Shahneous Bari1-0/+3
2025-11-21Revert "[MLIR][GPU] subgroup_mma fp64 extension" (#169049)Fabian Mora1-2/+2
2025-11-21[MLIR][GPU] subgroup_mma fp64 extension (#165873)Giacomo Castiglioni1-2/+2
2025-11-18[mlir][NVVM] Add no-rollback option to NVVM lowering passes (#168477)Matthias Springer1-0/+1
2025-11-13[MLIR] Apply clang-tidy fixes for misc-use-internal-linkage in AsyncRegionRew...Mehdi Amini1-2/+2
2025-11-10[MLIR][XeGPU] Decouple `inst_data` and `lane_layout` in propagation (#166941)Artem Kroviakov1-1/+8
2025-11-02[ADT] Prepare to deprecate variadic `StringSwitch::Cases`. NFC. (#166020)Jakub Kuderski1-4/+4
2025-10-31[MLIR][GPU] Ensure all lanes in cluster have final reduction value (#165764)James Newling1-1/+27
2025-10-30[mlir] Simplify Default cases in type switches. NFC. (#165767)Jakub Kuderski1-3/+3
2025-10-28 [MLIR] Revamp RegionBranchOpInterface (#165429)Mehdi Amini1-1/+1
2025-10-28Revert " [MLIR] Revamp RegionBranchOpInterface " (#165356)Mehdi Amini1-1/+1
2025-10-28 [MLIR] Revamp RegionBranchOpInterface (#161575)Mehdi Amini1-1/+1
2025-10-17[mlir][gpu]Add GPUToXeVM lowering pipeline pass. (#161216)Md Abdullah Shahneous Bari2-0/+145
2025-10-11[mlir] Use llvm accumulate wrappers. NFCI. (#162957)Jakub Kuderski2-4/+3
2025-10-06[mlir] Simplify unreachable type switch cases. NFC. (#162032)Jakub Kuderski2-4/+2
2025-09-30[MLIR] Fix gpu.launch attribution argument printing (#161408)Mehdi Amini1-29/+16
2025-09-29[MLIR] Apply clang-tidy fixes for bugprone-argument-comment in SubgroupReduce...Mehdi Amini1-1/+1
2025-09-25[mlir][gpu] Add innermost-first policy when mapping loops to GPU IDs (#160634)Georgios Pinitas1-10/+55
2025-09-14[mlir][LLVM] Add LLVMAddrSpaceAttrInterface and NVVMMemorySpaceAttr (#157339)Fabian Mora1-5/+3
2025-09-08[mlir][gpu] Revert gpu.subgroup_broadcast with any_lane (#157373)Jakub Kuderski1-4/+0
2025-08-30[mlir][gpu] Add `subgroup_broadcast` op (#152808)Ivan Butygin1-0/+40