Google Git
Sign in
opensecura / 3p / openxla / iree / 02c52150c14ad18363c272bfcb9697602c395869 / compiler
  1. 02c5215 Revert "[Codegen] Re-Enable transform dialect configuration strategy sample (#15787)" (#16097) by Quinn Dawkins · 1 year, 3 months ago
  2. 562098f [VectorDistribution] Add infrastructure to support vector distribution based on layout (#16009) by Kunwar Grover · 1 year, 3 months ago
  3. 3b534c4 [Codegen] Re-Enable transform dialect configuration strategy sample (#15787) by Quinn Dawkins · 1 year, 3 months ago
  4. 8fb2680 Disable loop unrolling in LLVM IR optimization passes (#16092) by Benoit Jacob · 1 year, 3 months ago
  5. dc81beb [CPU] Do not fuse ukernel ops into tiling loops. (#16054) by Han-Chung Wang · 1 year, 3 months ago
  6. a7a7ad6 Add vm.buffer.hash and util.buffer.hash ops (#16003) by Quinn Dawkins · 1 year, 3 months ago
  7. 198f271 [CPU] Fix multiconfig bug with tensor.pack op (#16082) by Max191 · 1 year, 3 months ago
  8. c8ecc1c Reland "[spirv][vulkan] Enable device query generation and execution" (#16075) by Lei Zhang · 1 year, 3 months ago
  9. 282ab77 Revert "[spirv][vulkan] Enable device query generation and execution" (#16077) by Han-Chung Wang · 1 year, 3 months ago
  10. 852684a [spirv][vulkan] Enable device query generation and execution (#15977) by Lei Zhang · 1 year, 3 months ago
  11. b55ba25 Fixing/silencing some warnings that have crept in over time. (#16072) by Ben Vanik · 1 year, 3 months ago
  12. 776789e [GlobalOpt] Add a pass to simplify tensor pack/unpack ops. (#15993) by Han-Chung Wang · 1 year, 3 months ago
  13. c1edb82 [CodeGen] Implement MemoryEffectsOpInterface for ukernel ops. (#16053) by Han-Chung Wang · 1 year, 3 months ago
  14. 6aa310c [CPU] Move checking stack allocation cmd flag to Passes.cpp (#16062) by Han-Chung Wang · 1 year, 3 months ago
  15. bd2c92d [Stream] Update more op folders to verify matching types (#16070) by Quinn Dawkins · 1 year, 3 months ago
  16. 46b06d0 [minor code simplification] Implement algorithm without stack (#15999) by James Newling · 1 year, 3 months ago
  17. b16cee3 Bump LLVM to llvm/llvm-project@054b5fc0fd41 (#16055) by Han-Chung Wang · 1 year, 3 months ago
  18. ccd576c [LLVMGPU] Add AMDGPUToArith conversion patterns to ROCDL lowering (#16067) by Quinn Dawkins · 1 year, 3 months ago
  19. 6647a5b [CPU] Skip tiling if the compute op is not a TilingInterface op. (#16052) by Han-Chung Wang · 1 year, 3 months ago
  20. 124d562 Bump StableHLO to f8dcebfa1ec166806974f6ae0dfb902d36b47238 (#16049) by Jacques Pienaar · 1 year, 3 months ago
  21. ef344ac Bump LLVM to llvm/llvm-project@6b65d79 and deps (2023-12-29) (#16012) by Kunwar Grover · 1 year, 3 months ago
  22. b3200c8 [CPU] Enable mmt4d distribution for large reduction size cases. (#16037) by Han-Chung Wang · 1 year, 3 months ago
  23. 0f0e0e7 [CodeGen] Carry over lowering_config when decomposing batch_mmt4d ops. (#16043) by Han-Chung Wang · 1 year, 3 months ago
  24. 6711155 Add folding arithmetic extensions (#15953) by erman-gurses · 1 year, 3 months ago
  25. c4739bc [LinalgExt] Delete LinalgExt tiling patterns and passes. (#15921) by Han-Chung Wang · 1 year, 3 months ago
  26. e4aa589 [CodeGen] Add aflag to allow potentially to remove unnecessary code to improve performance. (#15862) by Lubomir Litchev · 1 year, 3 months ago
  27. ce282c8 [stablehlo] Add missing nullptr check for unregistered dialects (#16032) by Jakub Kuderski · 1 year, 3 months ago
  28. 41da229 [CPU][NFC] Retire LLVMCPUTensorPad pass. (#16027) by Han-Chung Wang · 1 year, 3 months ago
  29. 80efa38 [GlobalOpt] Add f32->bf16 demotion cases for transposed matmuls (#16022) by Max191 · 1 year, 3 months ago
  30. 895645b [VectorLayoutAnalysis] Add transfer functions for vector.contract (#15996) by Kunwar Grover · 1 year, 3 months ago
  31. f9cdcfd [python] Expose python bindings for scf in iree.compiler.dialects (#16013) by Kunwar Grover · 1 year, 3 months ago
  32. e7384a1 [VectorLayoutAnalysis] Add debug printing (#16007) by Kunwar Grover · 1 year, 3 months ago
  33. c35d8e9 Standardizes CMake setup of C directory trees behind a macro. (#16011) by Stella Laurenzo · 1 year, 3 months ago
  34. 1ae94a5 [ROCM] Expose amdgpu-waves-per-eu opt hint (#16010) by harsh-nod · 1 year, 3 months ago
  35. 15c306f Build functioning dev packages for IREECompiler and IREERuntime. (#16008) by Stella Laurenzo · 1 year, 3 months ago
  36. b0e8f3c [VectorLayoutAnalysis] Fix bug in scf.for transfer functions (#15989) by Kunwar Grover · 1 year, 3 months ago
  37. 5bac47b [NFC][Codegen] Move Codegen/Dialect to Codegen/Dialect/Codegen/IR (#15944) by Kunwar Grover · 1 year, 4 months ago
  38. a3af0b0 Fix producer-consumer graph in `ConstExprAnalysis` (#15913) by Jerry Wu · 1 year, 4 months ago
  39. 0af34bd [GlobalOpt] Add quantized matmul reassociation support for f16 types (#15964) by Max191 · 1 year, 4 months ago
  40. 8d9638e [Flow] Loosen restrictions for dequantization fusion (#15663) by Max191 · 1 year, 4 months ago
  41. 5e33995 [onnx] Enable --iree-input-type=onnx (#15995) by Stella Laurenzo · 1 year, 4 months ago
  42. 7b7ffeb [onnx] Add ONNX importer and iree-import-onnx tool to compiler package. (#15920) by Stella Laurenzo · 1 year, 4 months ago
  43. 09c9f5b [spirv] Provide same entry point set in variants when linking (#15935) by Lei Zhang · 1 year, 4 months ago
  44. 5b8e870 [Flow] Relax dynamic shape restrictions on slice -> Flow patterns (#15861) by Quinn Dawkins · 1 year, 4 months ago
  45. 9ba681c [GlobalOpt][NFC] Follow LLVM comment style and apply few cleanups. (#15990) by Han-Chung Wang · 1 year, 4 months ago
  46. ded5262 [GlobalOpt] Remove specialized check-prefix from lit tests. (#15991) by Han-Chung Wang · 1 year, 4 months ago
  47. 109d174 Add a f32-to-bf16 demotion pass for matmuls, and new compiler flag `--iree-global-opt-enable-demote-contraction-inputs-to-bf16` (#15946) by bjacob · 1 year, 4 months ago
  48. 86511e9 drop redundant `populateExpandBFloat16Patterns` call in `ConvertToLLVM` (#15986) by bjacob · 1 year, 4 months ago
  49. 73f5947 Align the Flow::DispatchTensorStoreOp verifier with tensor.insert_slice (#15959) by Quinn Dawkins · 1 year, 4 months ago
  50. 829b398 [VM] Add a vm.rodata.table.inline op (#15971) by Quinn Dawkins · 1 year, 4 months ago
  51. 4e9199b Integrate llvm/llvm-project@99045b60b575 and bump deps (2023-12-19) (#15972) by Han-Chung Wang · 1 year, 4 months ago
  52. 92df2b4 Make iree.compiler.api.Output.map_memory() retain its backing reference. (#15975) by Stella Laurenzo · 1 year, 4 months ago
  53. da8570c [GlobalOpt] Turn transpose propagation off by default. (#15979) by Han-Chung Wang · 1 year, 4 months ago
  54. 166dcc5 [GlobalOptimization] Add pass for transpose propagation (#15858) by Quinn Dawkins · 1 year, 4 months ago
  55. 39576d0 [spirv] Trim provided executable target before linking (#15888) by Lei Zhang · 1 year, 4 months ago
  56. a0919c5 [Codegen] Bail TileAndDistributeToWorkgroups if the workgroup count is lowered (#15885) by Quinn Dawkins · 1 year, 4 months ago
  57. e2ba0a2 Updating HAL passes to use tablegen. (#15952) by Ben Vanik · 1 year, 4 months ago
  58. 0bd654f [compiler] NFC: Move op equivalence into a file in Utils/ (#15966) by Lei Zhang · 1 year, 4 months ago
  59. a5f0b29 [CPU] Unify distribution tile size selection heuristic. (#15928) by Han-Chung Wang · 1 year, 4 months ago
  60. 873ba77 [CPU][NFC] Update variable names to be camelCase. (#15967) by Han-Chung Wang · 1 year, 4 months ago
  61. 0ecbe75 [SPIRV] Switch tiling on reductions to use common helpers. (#15940) by Han-Chung Wang · 1 year, 4 months ago
  62. a0bf55c [Codegen] Switch to Util helper for getting bit widths (#15957) by Quinn Dawkins · 1 year, 4 months ago
  63. 6c1b1ab [GPU] Simplify gpu warp reduction lowering (#15963) by Jakub Kuderski · 1 year, 4 months ago
  64. e2dcbe1 Add support for serializing complex resource attributes (#15960) by Quinn Dawkins · 1 year, 4 months ago
  65. 18ed8d5 Bump torch-mlir to 9c655d0bfb166785dc17e51b0afa1f937c227cef (#15956) by Quinn Dawkins · 1 year, 4 months ago
  66. 0c0eb27 [spirv] Cleanup and reorder functions in SPIR-V passes (#15958) by Lei Zhang · 1 year, 4 months ago
  67. fac5c59 [Flow] Enable fusion of `tensor.unpack` with non-intersecting reductions (#15945) by Max191 · 1 year, 4 months ago
  68. 7606729 [spirv] NFC: sort pass pipelines update comment headers (#15950) by Lei Zhang · 1 year, 4 months ago
  69. 6f2c98f [LLVMGPU] Add multi-row vector reduction configuration for ROCm (#15941) by Jakub Kuderski · 1 year, 4 months ago
  70. 416e4b4 [Transform] Add transform.iree.gpu_distribute_shared_memory_copy (#15942) by Kunwar Grover · 1 year, 4 months ago
  71. 88c9992 [CodeGen] Switch tiling only cases to use scf::tileUsingSCFForOp (#15939) by Han-Chung Wang · 1 year, 4 months ago
  72. a910e74 [Codegen] Add VectorLayoutAnalysis (#15726) by Kunwar Grover · 1 year, 4 months ago
  73. 6dbc227 [CPU] Retire CPUDoubleTilingPadExpert pipeline. (#15931) by Han-Chung Wang · 1 year, 4 months ago
  74. 5dd6f34 [CodeGen][NFC] Rename tileLinalgOpsWithFilter to distributeLinalgOpsWithFilter. (#15938) by Han-Chung Wang · 1 year, 4 months ago
  75. 0842feb [bf16] Rework vector+bf16 support to avoid invalid conversion (#15911) by Rob Suderman · 1 year, 4 months ago
  76. 5519e24 Removing old HAL device matching attributes. (#15933) by Ben Vanik · 1 year, 4 months ago
  77. 8c99394 [GlobalOpt][DT] Retire ExpandVectors pass. (#15932) by Han-Chung Wang · 1 year, 4 months ago
  78. bc0b7d4 Fixing flow->stream affinity mapping. (#15929) by Ben Vanik · 1 year, 4 months ago
  79. a4a6b4b [CPU] Add support for converting math.powf from fp16 to fp32. (#15927) by Han-Chung Wang · 1 year, 4 months ago
  80. 98fa0b3 [LLVMGPU] Re-enable multidim distribution on gridDimension. (#15924) by Stanley Winata · 1 year, 4 months ago
  81. 9a59315 [CPU] Add missing passes for ukernel fallback (#15917) by Jerry Wu · 1 year, 4 months ago
  82. 9726ead [spirv] Respect entry point ordinal when serializing executables (#15905) by Lei Zhang · 1 year, 4 months ago
  83. 80e70ca Replacing hal.ex.shared_device with hal.devices.* ops. (#15916) by Ben Vanik · 1 year, 4 months ago
  84. 605aca9 [CPU][ArmSME] Add (initial) tiling and lowering pipeline for ArmSME (#15794) by Benjamin Maxwell · 1 year, 4 months ago
  85. 6ebc2b8 Annotate dispatches with pure data-movement linalg operations better (#15745) by Kunwar Grover · 1 year, 4 months ago
  86. 33a331b [CPU] Enable codegen fallback when ukernels are not available. (#15883) by Han-Chung Wang · 1 year, 4 months ago
  87. 82ae1ee [Codegen] Add VectorExt to CommonDialectRegistration (#15729) by Kunwar Grover · 1 year, 4 months ago
  88. 3402f2c [NFC] Internalize getMaterializeEncodingValueFn method to where it is used. (#15871) by Han-Chung Wang · 1 year, 4 months ago
  89. e2c7d81 [Codegen] Cherry-pick vector transfer write distribution changes. NFC. (#15901) by Jakub Kuderski · 1 year, 4 months ago
  90. 138af17 Cleanup compiler target plugin files. (#15898) by Scott Todd · 1 year, 4 months ago
  91. dee9169 Convert VMVX compiler target into a plugin. (#15896) by Scott Todd · 1 year, 4 months ago
  92. 3aa8421 [NFC] Replace LinalgExt::TilingPattern with walk-based transforms. (#15894) by Han-Chung Wang · 1 year, 4 months ago
  93. 49456a0 [Stream] Improve the function signature verifier for CmdDispatchOp (#15886) by Quinn Dawkins · 1 year, 4 months ago
  94. 7e06b6e [NFC] Remove VMVX deps from Codegen/Transforms. (#15895) by Han-Chung Wang · 1 year, 4 months ago
  95. 2529fb3 [CPU] Remove unnecessary factors from getMaxVectorTileSize. (#15843) by Han-Chung Wang · 1 year, 4 months ago
  96. 5a4e764 Adding stream.dispatch.workgroup.* info ops. (#15889) by Ben Vanik · 1 year, 4 months ago
  97. 8d01698 [vulkan] Enable initial executable linking (#15802) by Lei Zhang · 1 year, 4 months ago
  98. 15d6417 [SPIRV] Implements SPIRVTileAndDistribute as walk-based manner. (#15870) by Han-Chung Wang · 1 year, 4 months ago
  99. c216468 Integrate llvm 20231207 (#15855) by Stella Laurenzo · 1 year, 4 months ago
  100. 6829973 [NFC] Move RaiseSpecialOps to GlobalOptimization (#15881) by Quinn Dawkins · 1 year, 4 months ago