Google Git
Sign in
opensecura / 3p / openxla / iree / b07c0b971873da12999ed2ede1e3f87515a14e2e
  1. b07c0b9 Merge pull request #8873 from google/benvanik-join-workers by Ben Vanik · 3 years ago
  2. 9ca7563 Make benchmark artifact generation a bit more flexible. (#8846) by Scott Todd · 3 years ago
  3. 519815e Merge pull request #8853 from google/benvanik-benchmark-executables by Ben Vanik · 3 years ago
  4. 4afc558 Enable CUDA on remainder of eligible bots. (#8868) by Stella Laurenzo · 3 years ago
  5. 22112e0 Force push latest-snapshot (#8869) by powderluv · 3 years ago
  6. 5943902 Always build CUDA on CI. (#8867) by Stella Laurenzo · 3 years ago
  7. 87b8cbf Update docker images to pick up iree_cuda_deps in base. (#8865) by Stella Laurenzo · 3 years ago
  8. 9d54563 Add a sript to fetch and trim the CUDA toolkit deps needed to build. (#8862) by Stella Laurenzo · 3 years ago
  9. 4f03b4c Add BenchmarkDriver to run common benchmark flow. (#8654) by Jerry Wu · 3 years ago
  10. 6af56a2 [ci] Publish source MLIR modules after benchmark compilation (#8858) by Lei Zhang · 3 years ago
  11. 2453fe5 matmul test, remove native_vector[] from config (#8857) by Thomas · 3 years ago
  12. d3f2fc3 Fix the format of driver and target_arch. (#8829) by Jerry Wu · 3 years ago
  13. 7cf793f Fix data race: different mutexes were guarding the same data (#8856) by bjacob · 3 years ago
  14. d30bd2b Order worker initialization vs other threads stealing work (#8854) by bjacob · 3 years ago
  15. fc22cb3 Fix memory handling problem in matmul_test and renable cuda tests (#8797) by Thomas · 3 years ago
  16. f8e11fd Adding dispatch support to iree-benchmark-module. by Ben Vanik · 3 years ago
  17. 7a0a746 Adding -iree-hal-dump-executable-benchmarks-to= flag. by Ben Vanik · 3 years ago
  18. 37f0e7d Make MapElementTypeToDType consistent with the enum (#8810) by Sean Silva · 3 years ago
  19. b788b1b Fix tracing macro for IREE_HAL_CUDA_ALLOCATOR_ID (#8852) by nirvedhmeshram · 3 years ago
  20. 697fe38 Merge pull request #8837 from matthias-springer/fix_bufferization_inparallel2 by Matthias Springer · 3 years ago
  21. 795bb53 Fix bufferization of in_parallel by Matthias Springer · 3 years ago
  22. 74f7772 Modify elementwise fusion for changes in D123236. (#8841) by MaheshRavishankar · 3 years ago
  23. d07d59c Add tests and benchmarks for transpose op targeting AVX2 (#8750) by Diego Caballero · 3 years ago
  24. fed9cca Add tracing macro for IREE_HAL_CUDA_ALLOCATOR_ID (#8845) by nirvedhmeshram · 3 years ago
  25. 24267c6 Proper fixes for padding in IREE CodegenStrategy (#8843) by Han-Chung Wang · 3 years ago
  26. 1ddd917 Fixes for CompilationInfoAttr::get methods. (#8842) by Han-Chung Wang · 3 years ago
  27. 58ed257 Handle offsets/strides correctly while rewriting destructive updates. (#8793) by MaheshRavishankar · 3 years ago
  28. 6b93fc7 Merge pull request #8838 from google/benvanik-buffer-utils by Ben Vanik · 3 years ago
  29. 9e0536e Generalize bazel_to_cmake iree-dialects conversions. (#8800) by Scott Todd · 3 years ago
  30. 8df3a79 Remove retired RISC-V LLVM option (#8830) by CindyLiu · 3 years ago
  31. 5c16bbb Update SwiftShader to d15c4248 (2022-04-07) by Lei Zhang · 3 years ago
  32. 2d5bb6f Adding dedicated tracy allocation tracking for CUDA. by Ben Vanik · 3 years, 1 month ago
  33. f9f670c Updating iree.natvis to the latest enums. by Ben Vanik · 3 years, 1 month ago
  34. 210f4f9 Adding utility for reusing the subspan buffer implementation. by Ben Vanik · 3 years, 1 month ago
  35. 435e309 Adding IREE_ASSERT_REF_COUNT_ZERO helper. by Ben Vanik · 3 years, 1 month ago
  36. 2444059 Adding IREE_HOST_SIZE_MAX and IREE_DEVICE_SIZE_MAX. by Ben Vanik · 3 years, 1 month ago
  37. ae8ecce Specially handle width-sensitive arith cast ops. (#8809) by Sean Silva · 3 years ago
  38. 6eac24d Add peeling support to fusion. (#8835) by Nicolas Vasilache · 3 years ago
  39. 08dfc09 Add support for non-string padding values. (#8834) by Nicolas Vasilache · 3 years ago
  40. 766a515 hide output of adb push (#8806) by bjacob · 3 years ago
  41. 210c1f5 redirect capture output (#8828) by bjacob · 3 years ago
  42. 7a3bbd3 Add support for tile interchange. (#8664) by Han-Chung Wang · 3 years ago
  43. 793b9cf Add workgroup swizzling for better cache reuse (#8789) by nirvedhmeshram · 3 years ago
  44. 1b9e84c Adopt TransformDialectExtension and add iree_bufferize + iree_set_num_workgroups_to_one transform ops (#8821) by Nicolas Vasilache · 3 years ago
  45. eb0d678 Improve compilation time and execution time for quantized matmul on ARM (#8815) by Han-Chung Wang · 3 years ago
  46. 7094918 Bump LLVM to 50de659adcc1 (#8819) by Nicolas Vasilache · 3 years ago
  47. 9f81fa0 [spirv] Fix SPIRVTileAndDistribute flow (#8808) by Lei Zhang · 3 years ago
  48. 05000ad [spirv] Remove obsolete pass for distributing copy (#8813) by Lei Zhang · 3 years ago
  49. fd0f3d1 Add Mobilenet V3 UINT8 to presubmits (#8812) by mariecwhite · 3 years ago
  50. b2076c8 Bump tolerance to 1e-5 and enable passing tests. (#8792) by Han-Chung Wang · 3 years ago
  51. 030e197 [spirv] Fix fp16 vectorization flow (#8799) by Lei Zhang · 3 years ago
  52. fa9b0e1 Expose tm_tensor input type to Python (#8803) by Sean Silva · 3 years ago
  53. d2fa8a2 Add Mobilebert Quant to Presubmits (#8796) by mariecwhite · 3 years ago
  54. e4f2143 Cherry pick MLIR "[mlir][vector] Fold extract(broadcast) of same rank" (#8804) by Lei Zhang · 3 years ago
  55. f4b23d7 Remove redundant linalg.fill op in mhlo.concatenate -> Linalg lowering. (#8795) by Han-Chung Wang · 3 years ago
  56. 7814a0a Delete most get_started/ docs and clean up what's left. (#8801) by Scott Todd · 3 years ago
  57. af961d7 Allow using capstone-next for the tracy profiler. (#8760) by bjacob · 3 years ago
  58. 773e654 Update releases to use bazel 5.1.0. by Stella Laurenzo · 3 years ago
  59. d52f47e [NFC] Apply some cleanups to HAL::EntryPointOp (#8787) by Nicolas Vasilache · 3 years ago
  60. 68ee844 Integrate llvm-project and bump dependencies. (#8786) by Han-Chung Wang · 3 years ago
  61. 3148a51 Adding a comment and print to iree-benchmark-module. (#8788) by Ben Vanik · 3 years ago
  62. daec83b Disabling CUDA e2e matmul tests pending #8784. (#8785) by Ben Vanik · 3 years ago
  63. 6954424 Update bug_report.md (#8783) by Han-Chung Wang · 3 years ago
  64. 8934c94 Fixes configurations for matvec and dot. (#8775) by Han-Chung Wang · 3 years ago
  65. 89c1f8a [vulkan] Improve iree-run-module with GUI (#8781) by Lei Zhang · 3 years ago
  66. 107b2a8 Remove default assignee for issues (#8776) by Geoffrey Martin-Noble · 3 years ago
  67. 4a0b85b [spirv] Invoke TransposeOp canonicalization after vectorization (#8726) by Lei Zhang · 3 years ago
  68. b47b7c3 Add BenchmarkSuite to load benchmarks. (#8753) by Jerry Wu · 3 years ago
  69. 90c0649 Integrate llvm-project and bump dependencies. (#8777) by Han-Chung Wang · 3 years, 1 month ago
  70. a9ce257 Merge pull request #8768 from google/benvanik-limit-host-visibility by Ben Vanik · 3 years, 1 month ago
  71. 406d7b3 Merge pull request #8738 from google/benvanik-generic-type-demotion by Ben Vanik · 3 years, 1 month ago
  72. 0339fa0 Removing IREE_HAL_BUFFER_USAGE_ALL and tightening up host visibility. by Ben Vanik · 3 years, 1 month ago
  73. 91f017e Copy iree-dialects/ to integrations/tensorflow (#8774) by Han-Chung Wang · 3 years, 1 month ago
  74. dc6bbbd Allow multiple nested InParallelOp -> HAL rewrite. (#8771) by Nicolas Vasilache · 3 years, 1 month ago
  75. 80bd1c5 [flow] Verify the dynamic dims for all ShapeAwareOp's (#8773) by Sean Silva · 3 years, 1 month ago
  76. 380e154 Forward tensor.insert_slice coming from in_parallel lowering to flow.… (#8757) by Nicolas Vasilache · 3 years, 1 month ago
  77. fde35ef Fix for InitializeEmptyTensors (#8772) by Sean Silva · 3 years, 1 month ago
  78. 45fc889 Swap Stage 5 and Stage 6 in Codegen Driver (#8722) by Diego Caballero · 3 years, 1 month ago
  79. 683fc8d Use official github.ref_name to refer to the current branch in builds (#8770) by powderluv · 3 years, 1 month ago
  80. f6aa9d0 Reject compat of device-local mappable CUDA buffers if not avail. by Ben Vanik · 3 years, 1 month ago
  81. f5137a1 Making buffer generation only request host visibility if needed. by Ben Vanik · 3 years, 1 month ago
  82. 82bee4c Preserving MHLO demotion behavior until #8745 is fixed. by Ben Vanik · 3 years, 1 month ago
  83. 271c4fe Unifying type promotion/demotion into a generic pass. by Ben Vanik · 3 years, 1 month ago
  84. 501a964 Fixing uninitialized return warning. by Ben Vanik · 3 years, 1 month ago
  85. e0d3fdb Updating to use the mlir::parseSourceString template argument. by Ben Vanik · 3 years, 1 month ago
  86. 43a4b76 Add a pass to initialize all remaining empty tensors to zero-filled tensors (#8749) by MaheshRavishankar · 3 years, 1 month ago
  87. 5b3ccb0 [vulkan] Enable passing MHLO round test (#8728) by Lei Zhang · 3 years, 1 month ago
  88. 023ba0d Add a oneshot build and allow building on a branch (#8746) by powderluv · 3 years, 1 month ago
  89. 62fa5b8 Integrate llvm-project at c50eec400c0edc73eec3c9e97b5c030492cb787f (#8747) by Thomas · 3 years, 1 month ago
  90. 8512b7f [CUDA] Add pass to try to reduce shared memory bank conflicts (#8764) by Thomas · 3 years, 1 month ago
  91. 891084e Upgrade Bazel version to 5.1.0. (#8765) by Scott Todd · 3 years, 1 month ago
  92. dcfeecf Add tips for debugging correctness issues. (#8763) by Han-Chung Wang · 3 years, 1 month ago
  93. 300223c Update iree-flow-trace-dispatch-tensors flag name. (#8762) by Han-Chung Wang · 3 years, 1 month ago
  94. 7cf5b84 TSan instrumentation of module code (#8474) by bjacob · 3 years, 1 month ago
  95. 0936266 Add config class to represent the parameters of benchmark script (#8744) by Jerry Wu · 3 years, 1 month ago
  96. 879f076 Add optional target parameter to PrintOp. (#8758) by Nicolas Vasilache · 3 years, 1 month ago
  97. 7d486be Adds a script to update the readme on supported TFLite models (#8707) by not-jenni · 3 years, 1 month ago
  98. 0a2f25d Adding an error message when iree-util-apply-patterns fails. (#8754) by Ben Vanik · 3 years, 1 month ago
  99. 95eb1e5 Add "hostonly" tag to the host features test suite. (#8748) by Han-Chung Wang · 3 years, 1 month ago
  100. fc301df Properly enable +dotprod in int8 benchmarks. (#8743) by bjacob · 3 years, 1 month ago