Google Git
Sign in
opensecura / 3p / openxla / iree / a5c081c2152ca0cfd27369c67fc157c87ca52458
  1. a5c081c Migrate TOSA input conversion to a compiler plugin. (#15495) by Scott Todd · 1 year, 5 months ago
  2. 970e0f7 [GlobalOpt] Centralize data-tiling options under an if-statement. (#15537) by Han-Chung Wang · 1 year, 5 months ago
  3. 335e8b9 Disable check for TSC when validating tracy Python packages. (#15535) by Stella Laurenzo · 1 year, 5 months ago
  4. 1e41cbf Fix data race (TSan report) on worker thread startup on macOS in `iree_thread_request_affinity` (#15534) by Ben Vanik · 1 year, 5 months ago
  5. 9fc8f82 Fix assertion in VNNI ukernel (#15529) by bjacob · 1 year, 5 months ago
  6. d9f4e2e Document Python console scripts and instrumented runtime. (#15518) by Scott Todd · 1 year, 5 months ago
  7. 671b092 Split transform.tile_and_decompose_attention (#15516) by Kunwar Grover · 1 year, 5 months ago
  8. f458a88 Optimized vecmat ukernel tile functions for `i16 x u4 -> i32` on AVX-512-VNNI (#15525) by bjacob · 1 year, 5 months ago
  9. f9d7599 [SPIRV] Handle extraction from create_mask during load/store vectorization (#15524) by Quinn Dawkins · 1 year, 5 months ago
  10. 8633629 Fix data race (TSan report) on worker thread startup on macOS in `iree_thread_request_affinity` (#15499) by bjacob · 1 year, 5 months ago
  11. db7311b Use rewriter based methods for replacing in TileAndDecomposeAttention (#15514) by Kunwar Grover · 1 year, 5 months ago
  12. f66f28f Roll-up of minimal fixes needed to get external-parameter llama models working (#15523) by Stella Laurenzo · 1 year, 5 months ago
  13. 3ef0ea1 [CPU] Improve distribution tile sizes selection for mmt4d ops. (#15448) by Han-Chung Wang · 1 year, 5 months ago
  14. 7be3992 [Bindings][Python] Allow copies into a larger target buffer than the source (#15511) by Quinn Dawkins · 1 year, 5 months ago
  15. e2f8dc7 Merge CONTRIBUTING.md into website contributing guide. (#15449) by Scott Todd · 1 year, 5 months ago
  16. aa55bb6 Plumb 'torch' input to auto detect via plugin interface. (#15438) by Scott Todd · 1 year, 5 months ago
  17. 2f47c08 Add pytorch_aot_advanced sample Colab notebook using SHARK-Turbine. (#15329) by Scott Todd · 1 year, 5 months ago
  18. 094af3d Add note about registering vm list types. (#15279) by Scott Todd · 1 year, 5 months ago
  19. b3480e3 Set copyright as "The IREE Authors" in recently added torch files. (#15464) by Scott Todd · 1 year, 5 months ago
  20. f6ab15f Skip routinely-updated submodules in CODEOWNERS. (#15042) by Scott Todd · 1 year, 5 months ago
  21. 2fe00a5 Explain `PYTHONPATH` setup in build instructions. (#14834) by Scott Todd · 1 year, 5 months ago
  22. b55da2b [LLVMCPU] Re-enable vector masking for sub-byte element types (#15335) by Quinn Dawkins · 1 year, 5 months ago
  23. 74506f5 Supporting quantized block types in GGUF (kinda). (#15510) by Ben Vanik · 1 year, 5 months ago
  24. 13de9f3 CPUMaterializeEncoding: some tile sizes for `[i16, i16, i32]` and `[i16, ui4, i32]` on x86 (#15509) by bjacob · 1 year, 5 months ago
  25. 1582054 Setting worker local memory size based on L2 cache size by default. (#15444) by Ben Vanik · 1 year, 5 months ago
  26. fff16c7 [MicroKernels] Allow matching on CastOpInterface ops and use explicit signedness (#15472) by Max191 · 1 year, 5 months ago
  27. 5137b95 [DataTiling] Add supports for materializing elementwise ops. (#15507) by Max191 · 1 year, 5 months ago
  28. 4b1b11c [GlobalOptimization] Support ExpandVectors matvec cases with producer CastOpInterface ops (#15372) by Max191 · 1 year, 5 months ago
  29. d991dc9 Drop unused functions, but do not drop `-Wno-unused-function` (#15502) by bjacob · 1 year, 5 months ago
  30. bc98b9a [Codegen][GPU] Sink out shared memory and barriers in vector distribution (#15496) by Quinn Dawkins · 1 year, 5 months ago
  31. 1eb5713 [HAL] Add configuration stage in executable lowerings (#15362) by Quinn Dawkins · 1 year, 5 months ago
  32. 9b4606b Fix iree:span construction error using C++20 (#15489) by Julian Walker · 1 year, 5 months ago
  33. 675aafb bump torch-mlir to latest version (#15497) by saienduri · 1 year, 5 months ago
  34. ccc4c37 Add codeowners for Codegen/Common/GPU (#15498) by Quinn Dawkins · 1 year, 5 months ago
  35. 3c43fec [CPU] Improve distribution tile sizes selection for unpack ops. (#15447) by Han-Chung Wang · 1 year, 5 months ago
  36. c91759d Retire unused RISCV compile flag (#15484) by CindyLiu · 1 year, 5 months ago
  37. ed32801 Resolve `host` CPU to `generic` outside of x86 (#15481) by bjacob · 1 year, 5 months ago
  38. 3e2ae27 [SPIRV] Fix vector masking for transfer scalarization patterns (#15480) by Quinn Dawkins · 1 year, 5 months ago
  39. a2733b0 Add iree-dump-parameters python console script entrypoint. (#15490) by Stella Laurenzo · 1 year, 5 months ago
  40. d4f5368 LLVM integrate integrate-llvm-20231107 (#15470) by Stella Laurenzo · 1 year, 5 months ago
  41. d5ab0f0 Revert "[DataTiling] Add supports for materializing elementwise ops. (#15446)" by Stella Laurenzo · 1 year, 5 months ago
  42. bd60372 [DataTiling] Add supports for materializing elementwise ops. (#15446) by Max191 · 1 year, 5 months ago
  43. d115d41 Revert "Drop `-Wno-unused-function` on Clang, find out about actually unused functions, silence false positives." (#15483) by bjacob · 1 year, 5 months ago
  44. a4b8ccb Update github branch names etc (NFC) (#15482) by Jacques Pienaar · 1 year, 5 months ago
  45. 3a3c1a4 Fix `fp16` feature on arm64: the proper feature name is `fullfp16`, not `fp16`. (#15479) by bjacob · 1 year, 5 months ago
  46. 87ed5fc [GlobalOptimization] Lift `linalg.generic` ops to `linalg.batch_matmul/linalg.batch_vecmat/linalg.batch_matvec` (#15339) by Max191 · 1 year, 5 months ago
  47. 6db32c6 [GlobalOptimization] Support SetEncoding on batch matmul cases with p… (#15371) by Max191 · 1 year, 5 months ago
  48. 0ca4f62 Drop `-Wno-unused-function` on Clang, find out about actually unused functions, silence false positives. (#15471) by bjacob · 1 year, 5 months ago
  49. e604f9d Data-tiling: never use a generic fallback tile size (#15469) by bjacob · 1 year, 5 months ago
  50. d557095 Benchmark on c2-standard-60 in background (#15467) by Jerry Wu · 1 year, 5 months ago
  51. 88f41e5 [Reducer] Add initial check for interestingness (#15427) by Kunwar Grover · 1 year, 5 months ago
  52. d97270d Disable Metal tests when IREE_HAL_DRIVER_METAL is OFF. (#15476) by MaheshRavishankar · 1 year, 5 months ago
  53. 7ab3509 Record narrow static M/N sizes in `EncodingAttr` and rationalize MaterializeEncoding for narrow shapes. (#15431) by bjacob · 1 year, 5 months ago
  54. d509096 Treat `iree-llvmcpu-target-cpu` as an error outside of x86, and continue past errors (#15477) by bjacob · 1 year, 5 months ago
  55. 65fe91f Python implementation of the low level io parameters API. (#15457) by Stella Laurenzo · 1 year, 5 months ago
  56. 5b0fca4 Handle gguf zero-width types. (#15473) by Stella Laurenzo · 1 year, 5 months ago
  57. 2fda954 Add `chlo` dialect to auto input conversion pipeline (#15474) by Rob Suderman · 1 year, 5 months ago
  58. 82dc975 Start LLVM integrate integrate-llvm-20231103 (#15403) by Quinn Dawkins · 1 year, 5 months ago
  59. a0f3628 [CPU][SVE] Fix peeling for scalable-tiled loops (#15460) by Diego Caballero · 1 year, 5 months ago
  60. f4f3451 [GlobalOpt] Do not set encoding if they have preset compilation info. (#15455) by Han-Chung Wang · 1 year, 5 months ago
  61. d8a1643 Re-enable Pixel 6 tests and benchmarks (#15459) by Jerry Wu · 1 year, 5 months ago
  62. b6d7b83 [CMake] Add dep for ukernel internal headers (#15462) by Thomas Preud'homme · 1 year, 5 months ago
  63. 0f42d3f [GlobalOpt] Unset encodings for non-CPU backends. (#15453) by Han-Chung Wang · 1 year, 5 months ago
  64. e3f2ab3 [CPU] Improve tile sizes selection for tensor.pack ops. (#15397) by Han-Chung Wang · 1 year, 5 months ago
  65. 41a23ad [CPU] Materialize encodings into NOP if the target is not supported yet. (#15450) by Han-Chung Wang · 1 year, 5 months ago
  66. 327af04 Support output verification in Android benchmark tool (#15344) by Jerry Wu · 1 year, 5 months ago
  67. 38bfdba [GlobalOptimization] Add pattern to reassociate dequantization + matmul `linalg.gen… (#15278) by Max191 · 1 year, 5 months ago
  68. 1f61c88 Support fetching and streaming artifacts in benchmark tools (#15432) by Jerry Wu · 1 year, 5 months ago
  69. e14dff4 Add generated iree.runtime pyi file to enable IDE auto-complete. (#15454) by Stella Laurenzo · 1 year, 5 months ago
  70. a4b1a78 Remove redundant reshape checks in dot general preprocessing (#15319) by Rob Suderman · 1 year, 5 months ago
  71. d99527b Make default distribution logic divide work evenly. (#15414) by Han-Chung Wang · 1 year, 5 months ago
  72. af5d47d Fixing io_parameters.load VM lowering bug. (#15445) by Ben Vanik · 1 year, 5 months ago
  73. 4a77618 Generalization for ElementsAttr coverage (#15433) by saienduri · 1 year, 5 months ago
  74. f7de6ed Fix dashboard link on iree.perf.dev (#15443) by Jerry Wu · 1 year, 5 months ago
  75. d04da61 Always pass all local workgroup memory to each dispatch. (#15439) by Ben Vanik · 1 year, 5 months ago
  76. 4c71903 Fix `complex.bitcast` inputs to not generate bit manipulations (#15345) by Rob Suderman · 1 year, 5 months ago
  77. e94f3cb Add icons and tags to most website/docs/developers/ pages. (#15413) by Scott Todd · 1 year, 5 months ago
  78. 3e00192 Mention CPU .o files on tips page. (#15436) by Scott Todd · 1 year, 5 months ago
  79. f2e3260 Create oneshot stream command buffer in pending_queue_actions by Lei Zhang · 1 year, 5 months ago
  80. 618c835 [cuda] Port over CUDA stream-based command buffer impl by Lei Zhang · 1 year, 5 months ago
  81. e3671a5 NFC: Rename cuda to cuda2 by Lei Zhang · 1 year, 5 months ago
  82. a02ff0e NFC: Copy over existing stream command buffer impl by Lei Zhang · 1 year, 5 months ago
  83. ddb0d7d Let `EncodingAttr` use `struct(params)` assembly format (#15434) by bjacob · 1 year, 5 months ago
  84. 59122fd [metal] Fix unused variable when building for iOS (#15430) by Lei Zhang · 1 year, 5 months ago
  85. 717c7a0 Temporarily disable Pixel 6 tests and benchmarks (#15426) by Jerry Wu · 1 year, 5 months ago
  86. a306a28 Adding support for splat entries in parameter indices. (#15420) by Ben Vanik · 1 year, 5 months ago
  87. d46a708 Dump CPU .o files with `--iree-hal-dump-executable-intermediates-to=`. (#15412) by Ben Vanik · 1 year, 5 months ago
  88. eb3ec9c [spirv] Fix bufferization allocation in subgroup reduction pipeline (#15425) by Lei Zhang · 1 year, 5 months ago
  89. b28d1f0 [gpu] NFC: Tidy up the VectorReduceToGPU pass (#15424) by Lei Zhang · 1 year, 5 months ago
  90. 8ca883f Ignore Clangd configuration files for now (#15419) by Lei Zhang · 1 year, 5 months ago
  91. 4e1210c Make the schedule release GH job take an explicit override commit. (#15418) by Stella Laurenzo · 1 year, 5 months ago
  92. 640df51 Fix some broken links to benchmark_suites.md. (#15410) by Scott Todd · 1 year, 5 months ago
  93. 089e63b Add IO/Parameters dialect doc to website index page. (#15415) by Scott Todd · 1 year, 5 months ago
  94. cbe5799 Try to make enforce_glob less dumb and fix breakage at head. (#15409) by Stella Laurenzo · 1 year, 5 months ago
  95. 8ef62de Try to make enforce_glob less dumb and fix breakage at head. (#15409) by Stella Laurenzo · 1 year, 5 months ago
  96. 11ced0c Adding parameters as a concept to stream/hal/tooling. (#15104) by Ben Vanik · 1 year, 5 months ago
  97. 988f7c5 optimized s16s16s32 mmt4d tile functions on x86 (#15365) by bjacob · 1 year, 5 months ago
  98. a2a2e8d Build arm64 docker with ccache 4.7.4 (#15307) by Fredrik Knutsson · 1 year, 5 months ago
  99. 48fd11b Disable Pixel 4 e2e regression benchmarks (#15400) by Jerry Wu · 1 year, 5 months ago
  100. 904dfab Add Arm64 variants of build_all and test_all CI jobs (#15237) by Fredrik Knutsson · 1 year, 5 months ago