Google Git
Sign in
opensecura / 3p / openxla / iree / 573ff1ff02347266ed747dd316cefaeb4c710396
  1. 573ff1f Integrate LLVM at llvm/llvm-project@362aa434cc31ccca96749a6db8cd97f5b7d71206 (#16960) by Benoit Jacob · 1 year, 1 month ago
  2. 180e458 Add PDL pre-processing pass to compiler (#16945) by Nirvedh Meshram · 1 year, 1 month ago
  3. 29db67c Revert "[LLVMCPU][ArmSME] Add `2d-scalable-to-1d-scalable` pass" (#16963) by Scott Todd · 1 year, 1 month ago
  4. 782ac9d [LLVMCPU][ArmSME] Add `2d-scalable-to-1d-scalable` pass (#16712) by Benjamin Maxwell · 1 year, 1 month ago
  5. e913ff1 [CPU] Set vectorization options for Mmt4dTilingExpert pipeline. (#16954) by Han-Chung Wang · 1 year, 1 month ago
  6. 3fa9fbd Integrate LLVM at llvm/llvm-project@a6d932bca8875198fbf34564cda8a8d1640cdcbc (#16944) by Benoit Jacob · 1 year, 1 month ago
  7. 2c88e49 [LLVMGPU] Wmma layout for LLVMGPU vector distribute pipeline (#16928) by Stanley Winata · 1 year, 1 month ago
  8. d1eef77 Skip custom hip kernel sample if it would fail to build. (#16949) by Scott Todd · 1 year, 1 month ago
  9. e34c979 [torch-mlir] Cherrypick fix to fx_importer causing issues with int types. (#16950) by Stella Laurenzo · 1 year, 1 month ago
  10. c60dcc1 Fix bug in Horner's rule (#16865) by Pawel Paruzel · 1 year, 1 month ago
  11. 05ff73f [Flow] Do not propagate reshape when it's blocking unpack+generic fusion (#16930) by Han-Chung Wang · 1 year, 1 month ago
  12. 6ef1cfe Disable LLVM optional deps. (#16942) by Stella Laurenzo · 1 year, 1 month ago
  13. 52c8d52 Integrate torch-mlir at head (5325d3e6e6e0722ba78e14725b93107e0915710a). (#16940) by Stella Laurenzo · 1 year, 1 month ago
  14. 1204192 [GPU] Add workgroup transpose strategy to workgroup reordering pass (#16938) by Jakub Kuderski · 1 year, 1 month ago
  15. be9f097 [hip] Introduce options to control load of libamdhip64.so. (#16766) by Stella Laurenzo · 1 year, 1 month ago
  16. 41e312a [doc] Add documentation for HIP HAL backend (#16931) by Nithin Meganathan · 1 year, 1 month ago
  17. cc2ef92 [LLVMGPU] Allow workgroup reordering on ROCm (#16934) by Jakub Kuderski · 1 year, 1 month ago
  18. 623c1c8 [NFC] Simplify repeated vector push backs (#16936) by Jakub Kuderski · 1 year, 1 month ago
  19. 12a0b56 [NFC] Simplify type checks with isa predicates (#16935) by Jakub Kuderski · 1 year, 1 month ago
  20. 5cd0a0c Update Github runner to 2.315 (#16929) by Jerry Wu · 1 year, 1 month ago
  21. d884c54 [LLVMGPU][ROCm] Tweak preferred tile sizes in the MatmulSimt pipeline (#16923) by Jakub Kuderski · 1 year, 1 month ago
  22. 719a8a9 [LLVMGPU] Allow sending expanded convolutions down mfma pipeline (#16917) by Quinn Dawkins · 1 year, 1 month ago
  23. 501cb20 [VectorDistribution] Add better verifiers for anchors in layout analysis (#16924) by Kunwar Grover · 1 year, 1 month ago
  24. abe9aed [NFC] Fixing typo double and (#16904) by Jose Manuel Monsalve Diaz · 1 year, 1 month ago
  25. e8f8888 [Flow][Transforms] Add dynamic dim capture support to `scf.for` (#16889) by Markus Böck · 1 year, 1 month ago
  26. e942406 [cmake] Require runtime tracing for compiler tracing (#16922) by Jakub Kuderski · 1 year, 1 month ago
  27. 5acacb7 [Codegen] Fix layout analysis for vector.transpose (#16820) (#16921) by Quinn Dawkins · 1 year, 1 month ago
  28. bfdbd16 Cherrypick llvm/llvm-project@c43932ebdc40. (#16920) by Scott Todd · 1 year, 1 month ago
  29. c46daa6 Fix for iree-run-module with --input=not-a-path.bin (#16919) by James Newling · 1 year, 1 month ago
  30. e44cf32 Move external test suite configs out of experimental. (#16907) by Scott Todd · 1 year, 1 month ago
  31. ff820d6 Re-land "start testing real weight models ..." (#16918) by Scott Todd · 1 year, 1 month ago
  32. aacdd33 Tighten up the `lower_to_ukernel_ops.mlir` test (#16883) by Benoit Jacob · 1 year, 1 month ago
  33. daee1dd Ukernels: replace boilerplate by tables. (#16879) by Benoit Jacob · 1 year, 1 month ago
  34. ab1a65b Integrate LLVM at llvm/llvm-project@a9d1fead9614 (#16891) by Scott Todd · 1 year, 1 month ago
  35. d6357b4 [TD][Preprocessing] Speed up `match.cast_compatible_dag_from_root` (#16914) by Jakub Kuderski · 1 year, 1 month ago
  36. 831fcfa [docs][website] Fix up indexing (#16912) by Jakub Kuderski · 1 year, 1 month ago
  37. 7bda2ec Bump torch-mlir to HEAD (e2343cf4ce9a13e8fa09d6c5ade6524fa7cf2b02). (#16911) by Stella Laurenzo · 1 year, 1 month ago
  38. c160cb4 [LLVMGPU] Send skinny matmuls to the gpu reduction pipeline (#16898) by Jakub Kuderski · 1 year, 1 month ago
  39. cd1068b Revert "Start testing real weight models from external test suite." (#16910) by Scott Todd · 1 year, 1 month ago
  40. de65adf [docs][website] Add subsection on profiling with perf and pprof (#16908) by Jakub Kuderski · 1 year, 1 month ago
  41. 03749e7 Fix conv preprocessing filtering logic (#16897) by Jakub Kuderski · 1 year, 1 month ago
  42. 8ab68b6 Start testing real weight models from external test suite. (#16801) by Scott Todd · 1 year, 1 month ago
  43. 61a1f2e Mark regression tests as passing that now pass. (#16900) by Stella Laurenzo · 1 year, 1 month ago
  44. 07a854c [CPU][ArmSME] Add `-arm-sme-vector-legalization` to ArmSME pipeline (#16881) by Benjamin Maxwell · 1 year, 1 month ago
  45. e3ced3a [CodeGen][NFC] Remove unused encoding utils. (#16892) by Han-Chung Wang · 1 year, 1 month ago
  46. b96adf6 Bump torch-mlir to HEAD (17eeac880af409c6c0473c5930a2c08e25209f4c). (#16896) by Stella Laurenzo · 1 year, 1 month ago
  47. 2cdf145 fix build errors when tracing mode = 1 (#16884) by Okwan Kwon · 1 year, 1 month ago
  48. aa72368 Bump TF to nightly dev20240207 (#16871) by Julian Walker · 1 year, 1 month ago
  49. f3b6bcd Address comments by mariecwhite · 1 year, 1 month ago
  50. 76515a7 Add i8*i4 matmul microbenchmark by mariecwhite · 1 year, 1 month ago
  51. 5f2743b [hip] Move into hal/drivers and build by default (#16706) by Lei Zhang · 1 year, 1 month ago
  52. f19780a Add AMDGPU dialect to registerMlirDialects. (#16859) by Han-Chung Wang · 1 year, 1 month ago
  53. 565225e [CPU] Add data-tiling for s8s4s32 Arm64 ukernels by mariecwhite · 1 year, 1 month ago
  54. 767a611 [tracing] fix build errors when tracing mode = 1 (#16853) by Okwan Kwon · 1 year, 1 month ago
  55. 9e95c38 [Flow] Fix exponential blowup when optimizing dynamic `tensor.dim`s (#16847) by Markus Böck · 1 year, 1 month ago
  56. e7eef08 Adds a nullptr check around optional implementation in dynamic modules. (#16845) by Stella Laurenzo · 1 year, 1 month ago
  57. 253881e Fix verifier on stream.async.call to allow call to unknown lifetime. (#16844) by Stella Laurenzo · 1 year, 1 month ago
  58. ed59bed Enabling external HIP HSACO support ala CUDA external PTX support. (#16830) by Ben Vanik · 1 year, 1 month ago
  59. 21e067b Integrate LLVM at llvm/llvm-project@1a6ec906fb37 (#16753) by Han-Chung Wang · 1 year, 1 month ago
  60. 77cc34b Bump numpy dep up to the 2.0 prerelease. (#16800) by Scott Todd · 1 year, 1 month ago
  61. 2eae9b3 [CPU] Remove 8x8x16 i8mm microkernel by mariecwhite · 1 year, 1 month ago
  62. a2ed5d1 Trace allocate/deallocate in rocm_allocator. (#16822) by Scott Todd · 1 year, 1 month ago
  63. 92fe572 [CPU] Add s8s4s32 i8mm ukernel (#16678) by mariecwhite · 1 year, 1 month ago
  64. b61a918 Enable LTO optimization by default for runtime releases. (#16811) by Stella Laurenzo · 1 year, 1 month ago
  65. ee32fc7 [rocm] Fix crash when executable source information is missing (#16805) by Lei Zhang · 1 year, 1 month ago
  66. 2395046 [rocm] Excises (almost) dependence on /opt/rocm from the compiler. (#16803) by Stella Laurenzo · 1 year, 1 month ago
  67. d05b4a1 Splitting stack trace support out of status.c. (#16791) by Ben Vanik · 1 year, 1 month ago
  68. 94a0108 Improving fixupGlobalMutability in IREE::VM::GlobalInitializationPass. (#16783) by Ben Vanik · 1 year, 1 month ago
  69. 15d9039 Embedding executable source contents in binaries for tracing. (#16757) by Ben Vanik · 1 year, 1 month ago
  70. 05ff3e2 Don't link `opencl.bc` when compiling for ROCm. (#16778) by Scott Todd · 1 year, 1 month ago
  71. 3b30ab4 Revert "Ukernels: enable limited debug information that is useful in profilers like Tracy." (#16779) by Benoit Jacob · 1 year, 1 month ago
  72. cdff01f Ukernels: enable limited debug information that is useful in profilers like Tracy. (#15756) by Benoit Jacob · 1 year, 1 month ago
  73. e074a44 [ROCm] Add MI300 and MI300A target chips to doc (#16767) by Boian Petkantchin · 1 year, 1 month ago
  74. 20913f8 Read LLVM_VERSION_MAJOR as a directory property (#16771) by Benoit Jacob · 1 year, 1 month ago
  75. d2542cd Update tracy docs regarding `--iree-hal-dump-executable-sources-to=` (#15814) by Benoit Jacob · 1 year, 1 month ago
  76. 50aa9a3 Update 'iree-samples' -> 'iree-experimental' after rename. (#16761) by Scott Todd · 1 year, 1 month ago
  77. e9ee873 Static link when building RISC-V Linux benchmark tools (#16752) by Jerry Wu · 1 year, 1 month ago
  78. 8711c81 [EmitC] Fix some of the TODOs introduced in #16357 (#16759) by Simon Camphausen · 1 year, 1 month ago
  79. 858cce6 [LLVMGPU] Fix fused elementwise broadcasts in mfma pipeline (#16756) by Quinn Dawkins · 1 year, 1 month ago
  80. 5497435 Fixing iree-run-mlir error messages. (#16749) by Ben Vanik · 1 year, 1 month ago
  81. 26924e4 Adding `iree.tensor.trace` support for printf debugging. (#16746) by Ben Vanik · 1 year, 1 month ago
  82. e12ab47 Replace `unaryOperator` by EmitC LogicalNotOp (#16730) by Marius Brehler · 1 year, 1 month ago
  83. 0b8a096 Use EmitCBuilder for VariableOp (#16740) by Marius Brehler · 1 year, 1 month ago
  84. e9f7ecd [cuda][hip] Drop name_literal reference like Vulkan side (#16742) by Lei Zhang · 1 year, 1 month ago
  85. 3c296a5 Disabling inlining on the torch async function. (#16739) by Ben Vanik · 1 year, 1 month ago
  86. de398c3 Let RISCV_LINKER_FLAGS_EXE can be assigned by user (#16738) by Yun Hsiang · 1 year, 1 month ago
  87. a3603c6 [ROCM] Fix build with runtime tracing enabled (#16737) by Quinn Dawkins · 1 year, 1 month ago
  88. 18d73c7 [rocm] Fix IREE_ROCM_TRACE_ZONE symbol (#16736) by Lei Zhang · 1 year, 1 month ago
  89. c8081fd Adding legacy ROCM tracing zones. (#16735) by Ben Vanik · 1 year, 1 month ago
  90. 0077030 [pkgci] Enable on sdxl feature branch. by Stella Laurenzo · 1 year, 1 month ago
  91. 12fae0e Update JAX and TFLite MLIR artifacts for benchmarking by mariecwhite · 1 year, 1 month ago
  92. 7f9d97b Add optional attribute to set MFMA read layout (#16733) by harsh-nod · 1 year, 1 month ago
  93. 331801c bump torch to 80c7bc3f7ae12413836a2f610a6491794b4dbb08 (#16717) by Daniel Garvey · 1 year, 1 month ago
  94. 3baa82b [CUDA] Fix CUDA transform tests for generalized mmt (#16729) by Jakub Kuderski · 1 year, 1 month ago
  95. 27b837c Replace `binaryOperator` by EmitC ops (#16728) by Marius Brehler · 1 year, 1 month ago
  96. 69432b6 Delete compiler/src/iree/compiler/Dialect/Flow/Transforms/SubsetInsertionOpInterfaceImpl.cpp by Ben Vanik · 1 year, 1 month ago
  97. fc684f4 [EmitC] Fix error message in compiler driver (#16727) by Simon Camphausen · 1 year, 1 month ago
  98. 7c2b48c [LLVMGPU][SPIR-V] Run named op generalization early in configuration pipeline (#16726) by Jakub Kuderski · 1 year, 1 month ago
  99. d153b1c [LLVMGPU] Add shared memory prefetching (#16723) by Kunwar Grover · 1 year, 1 month ago
  100. 50714ae Retry failed pytest cases to try limiting flakes. (#16718) by Scott Todd · 1 year, 1 month ago