Google Git
Sign in
opensecura / 3p / openxla / iree / 31de207c4ffaf6f4dfb5d4f9e219201aeda5b26e
  1. 31de207 Disable failing ukernel tests on Windows (#13276) by bjacob · 2 years ago
  2. 8cf1db2 Using FindThreads instead of hardcode `-lpthread`. (#13118) by Rechie Kho · 2 years ago
  3. 49d0123 ukernels: pack: move the generation of i64 padding_value to codegen (#13264) by bjacob · 2 years ago
  4. 1fd449b ukernels: fold type enums into flags (#13260) by bjacob · 2 years ago
  5. 9a8e63e Ukernel interface: take offsets, reorder fields (#13235) by bjacob · 2 years ago
  6. 47da9cf Disable TF Bert-Large due to #13211 (#13212) by mariecwhite · 2 years ago
  7. 5f16489 ukernels: let `pack` take `padding_value` by value (#13233) by bjacob · 2 years ago
  8. 6bd3211 ukernels: drop the unused `i8` case in `unpack`. (#13231) by bjacob · 2 years ago
  9. 4bca308 Ukernels: separate public vs internal headers (#13230) by bjacob · 2 years ago
  10. 772a335 [runtime] Fix std::array parameter unpacking (#13222) by Eugene Zhulenev · 2 years ago
  11. 52861ca Integrate llvm-project at 9e9b1effac34 and bump dependencies (#13232) by Kojo Acquah · 2 years ago
  12. c6c713c [StableHLO] Add initial input conversion pipeline and e2e tests (#13223) by Jakub Kuderski · 2 years ago
  13. 786e1a0 Add missing canonicalization to LLVMCPUVectorLowering (#13169) by Diego Caballero · 2 years ago
  14. 84d938e Make python dependency in lit test configurable (#13224) by Tori Baker · 2 years ago
  15. ddbb3dd Relax the application conditions for the matmul tensorcore transform … (#13228) by Nicolas Vasilache · 2 years ago
  16. c59cf83 NFC - Improve debug messages around transform dialect matching (#13226) by Nicolas Vasilache · 2 years ago
  17. 585d387 Plumb compiler version info through to built Python wheels. (#13214) by Stella Laurenzo · 2 years ago
  18. e199a08 Split mhlo.optimization_barrier lowering into several util.optimization_barrier ops if there are multiple inputs (#13210) by NatashaKnk · 2 years ago
  19. 69236fa Remove outdated python version requirement for Pytype (#13209) by Phoenix Meadowlark · 2 years ago
  20. 43ceb15 Update matmul tensorcore strategy to properly trigger pipelining with… (#13194) by Nicolas Vasilache · 2 years ago
  21. fe604d3 [StableHLO][NFC] Trim down stablehlo to linalg conversion deps (#13180) by Jakub Kuderski · 2 years ago
  22. 978754a tidy up elementwise ukernels (#13204) by bjacob · 2 years ago
  23. 4dd4eac [llvmgpu] check if operands can be distributable for shmem promotion (#13201) by Okwan Kwon · 2 years ago
  24. ab37989 Integrate llvm-project at 571eaead173b and bump dependencies (#13200) by Kojo Acquah · 2 years ago
  25. e1a4a2b Remove the `matmul` ukernel (#13175) by bjacob · 2 years ago
  26. e6b7fae port `pack` and `unpack` to `ukernel.generic` (#13173) by bjacob · 2 years ago
  27. 0d82968 Use git directly instead of `actions/checkout` on Windows (#13186) by Geoffrey Martin-Noble · 2 years ago
  28. 0ab01b6 Use GCS for Windows ccache (#13183) by Geoffrey Martin-Noble · 2 years ago
  29. de2ecca Remove Pixel 4 GPU benchmarks (#13188) by Jerry Wu · 2 years ago
  30. eafc042 Fix ASAN issue casting to uint32 (#13193) by Tori Baker · 2 years ago
  31. 0e9c462 Add lowering for BF16 arith/math operations to F32 with truncf/extf wrappers (#12841) by Rob Suderman · 2 years ago
  32. 302a289 Generalize matmul tensorcore strategy to work with arbitrary unaligned f32 tens… (#13192) by Nicolas Vasilache · 2 years ago
  33. 3730042 Refactor strategy towards generalization (#13190) by Nicolas Vasilache · 2 years ago
  34. a806149 Disable models due to recent regression: https://github.com/openxla/iree/issues/13189 by mariecwhite · 2 years ago
  35. 7758993 Remove pull_request by mariecwhite · 2 years ago
  36. af46cd8 Add Bert-Large TF models in various batch sizes by mariecwhite · 2 years, 1 month ago
  37. ed174dc Fix batch_matmul + transpose using nvgpu.mma.sync.1688.f32 (#13075) (#13170) by Manish Gupta · 2 years ago
  38. ab5bcbd Add compiler C API `ireeCompilerGetRevision`. (#13184) by Stella Laurenzo · 2 years ago
  39. 8a1182c Put ccache hint in docs on one line. (#13185) by Stella Laurenzo · 2 years ago
  40. 20e014b Enable Shared Memory Swizzle Optimization for Batched Matmul (#13060) by Manish Gupta · 2 years ago
  41. 4880d00 [Dev Container] - Adding VSCode Dev Container Capability (#12921) by Jonathan DEKHTIAR · 2 years ago
  42. 8548216 Fix exportTargetAttr construction in InlineExecutablesPass. (#13144) by Scott Todd · 2 years ago
  43. 6a7f69f Fixing vm::ref operator& after type consistency changes. (#13178) by Ben Vanik · 2 years ago
  44. b8527c6 NFC - Parameterize the matmul tensor core strategy towards generaliza… (#13177) by Nicolas Vasilache · 2 years ago
  45. 04d2ba5 Updating architecture diagram with "plugins" in a few places. (#12615) by Ben Vanik · 2 years ago
  46. d2f5a7c Update `experimental/web/sample_dynamic/` after runtime changes. (#13174) by Scott Todd · 2 years ago
  47. 56c4eff Update TensorFlow package in `samples` Dockerfile to 2.12.0. (#13171) by Scott Todd · 2 years ago
  48. e4e2398 Add CI workflow to run android tests on mobile phones (#13024) by Jerry Wu · 2 years ago
  49. 81cf28c Simplifying iree-run-mlir by making it run only a single function. (#13149) by Ben Vanik · 2 years ago
  50. 0d5f837 [xla] Drop arg & res attributes on XLA import (#13154) by Jacques Pienaar · 2 years ago
  51. 9040c43 Opt in to `-Wc++20-extensions` warning in clang. (#13158) by Scott Todd · 2 years ago
  52. 48d7a42 Replace `ukernel.mmt4d` by `ukernel.generic`. (#13143) by bjacob · 2 years ago
  53. 479900a [mhlo] Remove redundant preprocessing pattern. (#13153) by Jacques Pienaar · 2 years ago
  54. abbe3da [StableHLO][NFC] Do not alias the stablehlo namespace (#13150) by Jakub Kuderski · 2 years ago
  55. cc783ff [StableHLO] Port convolution to linalg lowering (#13151) by Jakub Kuderski · 2 years ago
  56. 73f29cf Appease Windows compiler (#13155) by Nicolas Vasilache · 2 years ago
  57. cf3d786 Add a transform dialect CUDA strategy for unaligned f32 matmul that are still divisible by 4 (#13098) by Thomas · 2 years ago
  58. 8d1a790 Integrate llvm-project at c2c49f464435 and bump dependencies (#13135) by Kojo Acquah · 2 years ago
  59. 0b92d79 Enable passing tests on CPU. (#13147) by Han-Chung Wang · 2 years ago
  60. 1859381 [StableHLO] Port rng to linalg lowering (#13137) by Jakub Kuderski · 2 years ago
  61. 9461d3b Adding support for loading VM modules from dynamic libraries. (#13112) by Ben Vanik · 2 years ago
  62. 62153df Generalize `ukernel.generic` with `strided_outer_dims` attribute (#13138) by bjacob · 2 years ago
  63. 169e6a4 Update mmperf docker image by mariecwhite · 2 years ago
  64. 26a145a Update benchmark documentation to store artifacts in build directory (#13141) by Phoenix Meadowlark · 2 years ago
  65. 5f81fd3 [StableHLO] Port reduce_window to linalg lowering (#13128) by Jakub Kuderski · 2 years ago
  66. fe53bc6 [NVIDIA Container] Fix `PATH` to include NVIDIA binaries (#13126) by Jonathan DEKHTIAR · 2 years ago
  67. 27179e2 Use MPI for NCCL unique ID exchange by default (#12902) by Okwan Kwon · 2 years ago
  68. c205355 fatal non-numerical errors in e2e matmul test (#13139) by bjacob · 2 years ago
  69. 0c3a30e Revert "Reorder ukernel operands to match what `ukernel.generic_raw` can generate" (#13136) by bjacob · 2 years ago
  70. 6f3c0a4 [codeGen] Added math.exp2 expansion. (#13124) by Balaji V. Iyer · 2 years ago
  71. 4b09dc5 Add warpDims argument to buildMapToBlockAndThreads (#13134) by Nicolas Vasilache · 2 years ago
  72. 5abc05f Integrate llvm-project at de088dd3a0aa and bump dependencies (#13130) by Kojo Acquah · 2 years ago
  73. 63fee48 Adding support for passing processor_id/data to dynamic imports. (#13109) by Ben Vanik · 2 years ago
  74. 1b1571d [StableHLO][NFC] Port select_and_scatter tests (#13122) by Jakub Kuderski · 2 years ago
  75. e4c0a11 [StableHLO][NFC] Port reduction tests (#13121) by Jakub Kuderski · 2 years ago
  76. 7171c45 [TransformExtensions] Add a `pad(extract_slice(fill))` -> `fill` pattern to `tiling_canonicalization` (#13093) by qcolombet · 2 years ago
  77. e19fc8e Adding a local executable plugin mechanism. (#12625) by Ben Vanik · 2 years ago
  78. 51fbedf Update KernelConfig.cpp to properly tile batch dim for convolutions. (#13108) by Kojo Acquah · 2 years ago
  79. 70985b5 Expand the CUDA header file pattern for building with CUDA 12.x. (#13106) by pjannaty · 2 years ago
  80. e68e0ba [ci] Make vendor/arch definitions in benchmarks better (#13099) by Lei Zhang · 2 years ago
  81. 3f1c154 Reorder ukernel operands to match what `ukernel.generic_raw` can generate (#13103) by bjacob · 2 years ago
  82. f98b5d8 Compute reduction across batch dims (#13091) by harsh-nod · 2 years ago
  83. b798319 Fix MSVC warning: wrong pointer type in `_mm_prefetch` (#13102) by bjacob · 2 years ago
  84. e2ce623 `ukernel.generic_raw` op passing only buffer and offset for memref operands (#13053) by bjacob · 2 years ago
  85. ff19816 Integrate LLVM at llvm/llvm-project@75f1f158 (#13094) by Thomas · 2 years ago
  86. c582ec7 Make boot disks bigger instead of futzing with ramdisk and local ssd (#13079) by Geoffrey Martin-Noble · 2 years ago
  87. 544f785 [TransformExtensions] Fix the vector_to_mma_conversion op (#13092) by qcolombet · 2 years ago
  88. 245a4f2 Some of the recently added tests require sm80 (#13077) by Tori Baker · 2 years ago
  89. 1b06f74 Allow more general accumulators when distributing reduction + broadcast + transpose (#13039) by harsh-nod · 2 years ago
  90. 186cdd4 [ROCm] Fix package name prefix and package root dir (#13088) by Boian Petkantchin · 2 years ago
  91. 644820c Add support to emit masked cp.async with 2-D masks (#13084) by Nicolas Vasilache · 2 years ago
  92. 09630d6 Finally moving VM type registration to iree_vm_instance_t. (#12650) by Ben Vanik · 2 years ago
  93. 8be370c [StableHLO] Port patterns to handle scalar op to arith lowering (#13083) by Jakub Kuderski · 2 years ago
  94. df166ed Adding flow.tensor.alloc op for unique allocations. (#13081) by Ben Vanik · 2 years ago
  95. 5c005b7 [StableHLO] Re-enable linalg.map lowering tests for non-pointwise ops (#13082) by Jakub Kuderski · 2 years ago
  96. 87ef655 Add a structured op matcher for 2d convolutions (#12805) by Quinn Dawkins · 2 years ago
  97. 40ecb03 [StableHLO] Port more non-pointwise linalg lowering tests (#13080) by Jakub Kuderski · 2 years ago
  98. 586f19e Add codegen side plumbing for calling micro kernels from codegen backends. (#12831) by MaheshRavishankar · 2 years ago
  99. e635b2f Disable gather vectorization (#13067) by Jacques Pienaar · 2 years ago
  100. 60b8ef4 [StableHLO][NFC] Enable FileCheck variable scope in linalg lowering tests (#13063) by Jakub Kuderski · 2 years ago