Google Git
Sign in
opensecura / 3p / openxla / iree / 024c48b23c0e8721d94e544fa78bcfd291afa439 / compiler
  1. 7b58c71 Integrates/llvm 20240621 (#17723) by Nirvedh Meshram · 9 months ago
  2. ac418d1 Integrate llvm/llvm-project@27ac46e6bea2 (#17662) by Lei Zhang · 9 months ago
  3. f427965 Add extra info to error message in transfer_read operation with element and thread count info (#17695) by RattataKing · 9 months ago
  4. 9fd55d2 [Codegen][GPU] Update greedy tile + fuse pipeline to generate mfma (#17617) by Quinn Dawkins · 9 months ago
  5. d01fb23 add indexing maps for `iree_linalg_ext.scatter`'s out operand (#17704) by Ian Wood · 9 months ago
  6. 643a7cd [Flow] move tensor lowerings out of FormDispatchWorkgroupsPass (#17282) by Ian Wood · 9 months ago
  7. 12d43e8 [Codegen][GPU] Allow serial tiling of online_attention op (#17702) by Kunwar Grover · 9 months ago
  8. 90f29a6 Reland "[spirv] Switch to use common target description" (#17699) by Lei Zhang · 9 months ago
  9. 7c41049 Fixing broken fill builtins that were double offsetting. (#17696) by Ben Vanik · 9 months ago
  10. d792d24 Revert "[spirv] Switch to use common target description" (#17698) by Scott Todd · 9 months ago
  11. 7b9fb12 [spirv] Switch to use common target description (#17623) by Lei Zhang · 9 months ago
  12. 6f17869 Only set one narrow M/N at a time (#17647) by lialan · 9 months ago
  13. 3461314 Drop tile sizes specific to the ukernels-disabled case. (#17631) by lialan · 9 months ago
  14. 2b3c46c [GPUDistributionPatterns] Propagate predicate attribute for cmpf op (#17664) by Avinash Sharma · 9 months ago
  15. 1f954b2 [LLVMGPU] Generalize AMDGPUChainedMatmul pass to multiple dimensions (#17684) by Kunwar Grover · 9 months ago
  16. 7b782a8 [LinalgExt] Reland: Add online_attention op (#17681) by Kunwar Grover · 10 months ago
  17. 3428231 [LLVMCPU] Populate ArmSVE to LLVM conversion patterns (#17665) by Benjamin Maxwell · 10 months ago
  18. 045bf32 Change calculation of reassociation indicies in ConvertConvToChannelsLast.cpp (#17668) by Ian Wood · 10 months ago
  19. dc10693 Enable Workgroup Reordering Based on Translation Info Config Entries (#17645) by Bangtian Liu · 10 months ago
  20. 2ff4102 Revert "[LinalgExt] Add online_attention op" (#17658) by Scott Todd · 10 months ago
  21. 71c07fa [CPU] Signal errors if there are large vectors. (#17620) by Han-Chung Wang · 10 months ago
  22. 0a561c4 [Codegen][GPU] Make operand promotion pattern work with generics (#17650) by Quinn Dawkins · 10 months ago
  23. abf0087 [LinalgExt] Add online_attention op (#17536) by Kunwar Grover · 10 months ago
  24. 52b21f8 [GPUHeuristic] Modify schedule generator to consider distribution of tranfer_read layout anchor (#17636) by Stanley Winata · 10 months ago
  25. 6e1d80a [Flow] Make the output indexing_map of elementwise ops identity. (#17583) by Ian Wood · 10 months ago
  26. db7974c [util] Add serialization support for `f64` resources (#17640) by Markus Böck · 10 months ago
  27. cda3ccb [GPU] Enable tensor.pack e2e tests for rocm backend. (#17587) by Han-Chung Wang · 10 months ago
  28. d7744b7 [Codegen][GPU] Loosen dim mapping restrictions on forall fusion (#17612) by Max191 · 10 months ago
  29. 8ab07d2 [Codegen][LLVMGPU][NFC] Cleanup contract distribution pattern for LayoutAttr (#17581) by Kunwar Grover · 10 months ago
  30. 363e088 [Vecdist][GPU] Distribute LayoutConflict to roundtrip to shared memory. (#17618) by Stanley Winata · 10 months ago
  31. 088aef8 [LLVMGPU] Generalize VectorContractOpInfo based on indexing maps (#17625) by Kunwar Grover · 10 months ago
  32. 1943bc6 Remove AVX-512 tile sizes for non-ukernel case. (#17628) by lialan · 10 months ago
  33. f4cfb55 Enable end-of-file-fixer and trailing-whitespace hooks. (#17630) by Scott Todd · 10 months ago
  34. f062b19 [LLVMGPU] Fix linear dim selection in GPUApplyTilingLevel (#17611) by Max191 · 10 months ago
  35. ae04c67 [Codegen][LLVMGPU] Add pass pipeline for greedy tile + fuse (#17559) by Quinn Dawkins · 10 months ago
  36. 3b5d269 Enable the `mmt4d` ukernel by default on `x86_64` and on `arm_64` (outside of SVE/SME). (#17502) by Benoit Jacob · 10 months ago
  37. 58feff3 [CPU] Add support for unpack ukernel preparation (#17498) by Prashant Kumar · 10 months ago
  38. 5404ad7 Add versioned, automatically installed 'buildifier' pre-commit hook. (#17589) by Scott Todd · 10 months ago
  39. 6a43e05 add fused op to linalgext annotation (#17474) by Ian Wood · 10 months ago
  40. 29472a1 [CPU] Reland "Data tiling: transpose narrow-N into narrow-M" (#17545) by lialan · 10 months ago
  41. 5a44639 [CodeGen] Fix `in_bounds` attribute bug in tensor.extract_slice folding patterns. (#17563) by lialan · 10 months ago
  42. b44581a [LLVMGPU][ROCM][Layoutv1] Landing Implementation of WMMA on layoutV1 (#17580) by Stanley Winata · 10 months ago
  43. aef06ed [iree][global] Control the demotion of ops (#17515) by Prashant Kumar · 10 months ago
  44. 6291224 Bump llvm-project@534590144f7c7ec34b8e5e95aba3e4f214b074eb (#17572) by Rob Suderman · 10 months ago
  45. a5bd834 Fix conversion of pathlib.Path to str (#17573) by patosgui · 10 months ago
  46. 7388d75 [CPU][ArmSME] Enable transposes for f32 and f64 (#17440) by Cullen Rhodes · 10 months ago
  47. 14fd6ac [Codegen] Add pass to normalize loop bounds (#17542) by Quinn Dawkins · 10 months ago
  48. 63a2d14 [Codegen][GPU] Allow fusion of thread foralls into lane foralls (#17524) by Quinn Dawkins · 10 months ago
  49. 9edbb34 [Codegen][GPU] Make GPUApplyTilingLevel run cleanup patterns (#17533) by Quinn Dawkins · 10 months ago
  50. 9161303 [Codegen][GPU] Fix result replacement given multiple tiled ops (#17537) by Quinn Dawkins · 10 months ago
  51. b246fa6 [Codegen][GPU] Add pattern to distribute multi_mma ops to lanes (#17521) by Quinn Dawkins · 10 months ago
  52. e876020 Strided pack/unpack ukernels (#17516) by Benoit Jacob · 10 months ago
  53. 62efaee Format files across the project using pre-commit. (#17534) by Scott Todd · 10 months ago
  54. 2f36f8b [CPU] Do not unroll outer dimensions for elementwise ops (#17513) by Han-Chung Wang · 10 months ago
  55. 7df9985 [CPU] Do not decompose the ops when they are not in ukernels list. (#17529) by Han-Chung Wang · 10 months ago
  56. ff80728 [CPU] Remove outdated llvm::cl::opt cmd flag. (#17528) by Han-Chung Wang · 10 months ago
  57. 2587078 Integrate LLVM at 191e64ffa9061e9c0d5387b38bd2d5a9ffebb0fb (#17510) by Kunwar Grover · 10 months ago
  58. 6035f81 [Codegen][GPU] Update forall fusion to look through reshapes (#17523) by Quinn Dawkins · 10 months ago
  59. 2c59505 [Codegen] Add destination fusion to fuse_and_hoist pass (#17517) by Quinn Dawkins · 10 months ago
  60. 26e4c6b [Codegen][GPU] Enable vectorization of iree_gpu.shuffle_tensor + lowering to iree_gpu barriers (#17506) by Quinn Dawkins · 10 months ago
  61. ce1be9c [Codegen][GPU] Add semi-generic tile + fuse pass (#17477) by Quinn Dawkins · 10 months ago
  62. 5b243a8 [Backend][ROCM] Add gfx1150 support. (#17508) by Stanley Winata · 10 months ago
  63. aa0bc40 [Codegen][GPU] Add pass to fuse and hoist scf.forall ops (#17505) by Quinn Dawkins · 10 months ago
  64. 29e70ab Update onnx package version minimum to 1.16.0. (#17504) by Scott Todd · 10 months ago
  65. a6a56a9 Add `LinalgFusionInterface` to support fusion for linalg_ext ops (added `scatter` and `reverse`) (#17428) by Ian Wood · 10 months ago
  66. 3d1364e [Codegen][GPU] Add pattern to lower iree_gpu.multi_mma to intrinsics (#17457) by Quinn Dawkins · 10 months ago
  67. ab8f668 Revert "Data tiling: transpose narrow-N into narrow-M" (#17503) by Benoit Jacob · 10 months ago
  68. e33ca89 [LinalgExt] Split TileAndDecomposeAttention (#17468) by Kunwar Grover · 10 months ago
  69. 322d688 [Codegen][GPU] Add pattern to drop lead unit dims of multi_mma ops (#17456) by Quinn Dawkins · 10 months ago
  70. 117cb43 Test 'console' provider in 'tracing' job. (#16454) by Scott Todd · 10 months ago
  71. 16bdaa9 Data tiling: transpose narrow-N into narrow-M (#17446) by lialan · 10 months ago
  72. 6c75aa1 [Codegen][GPU] Allow iree_gpu.tensor_barrier to take vectors (#17479) by Quinn Dawkins · 10 months ago
  73. 1750e2b Integrate LLVM at 855eef2abd81cb8c7543d4748353d5e378fdd4c2 (#17501) by Benoit Jacob · 10 months ago
  74. 051c361 NFC: Make a few loop transformations more accessible (#17489) by Quinn Dawkins · 10 months ago
  75. 9e3d27a Upgrade to nanobind 2.0. (#17497) by Stella Laurenzo · 10 months ago
  76. cad02f9 [Codegen][GPU] Add unrolling pattern for iree_gpu.multi_mma (#17454) by Quinn Dawkins · 10 months ago
  77. 46c6bf5 [CPU] Add support for pack ukernel preparation. (#17472) by Han-Chung Wang · 10 months ago
  78. abdf550 Update IREE onnx import to be in sync with Torch-MLIR (#17476) by saienduri · 10 months ago
  79. a842527 [Codegen][GPU] Drop dead PassDetail.h file (#17490) by Quinn Dawkins · 10 months ago
  80. 63dff03 [Codegen][GPU] Add iree_gpu.tensor_barrier op (#17478) by Quinn Dawkins · 10 months ago
  81. 31e1a30 [Codegen][GPU] Add dictionary based lowering config attribute (#17463) by Quinn Dawkins · 10 months ago
  82. ea7d01e Pin nanobind to 1.9.2 to defer 2.0.0 API changes. (#17481) by Scott Todd · 10 months ago
  83. 008add9 [CodeGen][NFC] Rename DecomposeBatchMmt4DOps to CPUPrepareUkernels. (#17471) by Han-Chung Wang · 10 months ago
  84. 30e0238 Bump LLVM to llvm/llvm-project@bd3f5a4bd3d9d7ee8ae801c24c5081073b20abd4 (#17470) by MaheshRavishankar · 10 months ago
  85. 9fe159d [LinalgExt] Generalize attention tiling interface implementation (#17408) by Kunwar Grover · 10 months ago
  86. 1316c92 [Codegen] NFC: Move the lowering config to an attribute interface (#17439) by Quinn Dawkins · 10 months ago
  87. 7813fd3 [CPU] Fix a distribution bug and limiting distribution tile sizes. (#17436) by Han-Chung Wang · 10 months ago
  88. d4aa849 [CPU] Add support for pack/unpack ukernels enablement on llvm-cpu path. (#17427) by Han-Chung Wang · 10 months ago
  89. 6c5198d Folding no-op stream.async.update ops away. (#17458) by Ben Vanik · 10 months ago
  90. 006af5d [GPU] Support specifying LLVMGPU backend target features (#17451) by Lei Zhang · 10 months ago
  91. a36773a [Codegen][GPU] Add vectorization pattern for iree_gpu.multi_mma (#17453) by Quinn Dawkins · 10 months ago
  92. f6a38ac [GPU] Thread through a common target description (#17217) by Lei Zhang · 10 months ago
  93. 62a996b [Codegen] Add lane distribution for scf.forall (#17373) by Quinn Dawkins · 10 months ago
  94. 080b1fa [Codegen][GPU] Add a contraction like operation for mma intrinsics (#17374) by Quinn Dawkins · 10 months ago
  95. e0f3c05 [Codegen][GPU] Change iree_gpu.shuffle_tensor to take a region for the read (#17425) by Quinn Dawkins · 10 months ago
  96. dc61fcc Register ShapeDialect in StableHLO plugin. (#17444) by Scott Todd · 10 months ago
  97. a3b74bc [CPU][ArmSME] Update tiling to use all SME accumulators (#16389) by Benjamin Maxwell · 11 months ago
  98. 6d95f8c Integrate LLVM at `74a87548` (clean) (#17423) by Ingo Müller · 11 months ago
  99. 4f8ee51 Moving demotion/promotion passes to input conversion. (#17422) by Ben Vanik · 11 months ago
  100. dece30e [CPU] Do not decompose pack/unpack ops on x86 backends. (#17366) by Prashant Kumar · 11 months ago