Google Git
Sign in
chromium / external / github.com / tensorflow / tensorflow.git / HEAD
  1. 89045aa Go: Update generated wrapper functions for TensorFlow ops. by A. Unique TensorFlower · 18 minutes ago master
  2. a339764 Merge pull request #101517 from ILCSFNO:patch-1 by TensorFlower Gardener · 30 minutes ago
  3. cbdc771 Merge pull request #101522 from ILCSFNO:patch-4 by TensorFlower Gardener · 38 minutes ago
  4. f27d9d6 Automated Code Change by A. Unique TensorFlower · 49 minutes ago
  5. 2096501 Remove register everything. by Jacques Pienaar · 67 minutes ago
  6. 1ddcd85 Move absl_thread_pool to XLA as YnnThreadpool by A. Unique TensorFlower · 76 minutes ago
  7. c9d8d37 [xla:gpu] Relax nested gemm fusion constraints. by Christian Sigg · 87 minutes ago
  8. d46c1b9 Automated Code Change by A. Unique TensorFlower · 2 hours ago
  9. 46f983d Enable lowering from FQ Composite for 2-bit by Majid Dadashi · 2 hours ago
  10. c0d9a60 Internal changes to project structure by Gregory Pataky · 3 hours ago
  11. b2f2568 [xla:cpu:xnn] Temporarily disable XNNPACK by default. by Penporn Koanantakool · 3 hours ago
  12. a0e060a Automated Code Change by A. Unique TensorFlower · 3 hours ago
  13. f67cb87 Add support for int2/int4 in tfl.cast by Majid Dadashi · 4 hours ago nightly
  14. 5592d36 Automated Code Change by A. Unique TensorFlower · 4 hours ago
  15. a8a7474 Update XNNPACK in XLA by A. Unique TensorFlower · 4 hours ago
  16. ef3a678 [xla:cpu] Fix BufferAllocationInfo::InOutParameter constructor by Eugene Zhulenev · 7 hours ago
  17. e0f3263 [StableHLO Builder] Add API to set frontend attributes by Kevin Gleason · 7 hours ago
  18. c19b0d8 [XLA] Relax restrictions on the `iteration_space` parameter in `Schedule`. by Benjamin Chetioui · 7 hours ago
  19. c3ce8a9 Add `PjRtDeviceDimensions` struct and proto. by Haibo Huang · 8 hours ago
  20. c986bf1 Add a function to support internal logging. by A. Unique TensorFlower · 8 hours ago
  21. 180445d Add more textual hdrs to `mkl_dnn_acl` target to enable usage of `--config=mkl_aarch64_threadpool` with `--config=rbe_cross_compile_linux_aarch64` in JAX wheel builds. by Yulia Baturina · 9 hours ago
  22. a04d8ee Update XNNPACK in XLA by A. Unique TensorFlower · 9 hours ago
  23. a2ab65d [xla:ffi] Fix XLA_FFI_REGISTER_TYPE_X macro to support optional args by Eugene Zhulenev · 9 hours ago
  24. fd09236 Rollforward of: Implement PjRtStreamExecutorRawBuffer::CopyToLiteralAsync and allow by Parker Schuh · 9 hours ago
  25. 878a485 Add peak private footprint memory measurement tracking by Marissa Ikonomidis · 10 hours ago
  26. 1f3321f * Adds flag `tf_serialize_mlir_to_compressed_bytecode` to serialize to compressed bytecode. by A. Unique TensorFlower · 10 hours ago
  27. 5863476 Integrate LLVM at llvm/llvm-project@bfee9db78577 by A. Unique TensorFlower · 10 hours ago
  28. a6a2128 Add IFTTT for named sharding types by Kanish Anand · 10 hours ago
  29. 5c18a50 Use nvml impl lib based wrapper by Maxim Ermilov · 11 hours ago
  30. 83a98ac [XLA] Remove dead module-group related code from HloPassPipelineTest by Michael Kuperstein · 11 hours ago
  31. 168eb2c [XLA:CPU][XTile] Add lowering for tensor extract and from_elements & fix dot with scalar output. by Will Froom · 11 hours ago
  32. 4e29e9d [XLA] Remove unused RecordPassStartMetadata overload. by Michael Kuperstein · 11 hours ago
  33. 9607c16 Support Windows memory info by Marissa Ikonomidis · 11 hours ago
  34. c8b47aa In profile_data_lib.cc, throw std::runtime_error instead of check fail. by A. Unique TensorFlower · 12 hours ago
  35. 83c4070 [XLA:GPU] Don't fail Autotuner::GetSupportedConfigs if one of the backend fails by A. Unique TensorFlower · 14 hours ago
  36. 3c991bd Remove linking libnvidia-ml.so from hermetic CUDA forward compatibility mode. by Yulia Baturina · 14 hours ago
  37. 7a51446 Propagate NCCL aborts into user exceptions. by Michael Whittaker · 15 hours ago
  38. c12ac78 [XLA] Add `asinh` as a native HLO opcode. by A. Unique TensorFlower · 16 hours ago
  39. e2315cd Add a metric on enqueue request size by A. Unique TensorFlower · 16 hours ago
  40. b81b331 PR #31030: [XLA:GPU] Move ReduceScatterCreator after AlgebraicSimplifier by Sevin Fide Varoglu · 17 hours ago
  41. edf3f8b [XLA] Implement a `TiledHloSchedule` that transposes the iteration order over the non-contracting dimensions of a `dot`. by Benjamin Chetioui · 17 hours ago
  42. 78e0281 [XLA:CPU][XTile] Add lowering for StableHLO DotGeneral. by Will Froom · 17 hours ago
  43. 2ec0b2b Cache uncompress functions by A. Unique TensorFlower · 17 hours ago
  44. d52ccd4 [xla:ffi] Fix clang macro expansion warnings by Eugene Zhulenev · 17 hours ago
  45. f7524f0 Add proto [de]serialization for `SelectKThunk`. by Aliia Khasanova · 18 hours ago
  46. 45c685e Automated Code Change by A. Unique TensorFlower · 18 hours ago
  47. 26d0e97 Automated Code Change by A. Unique TensorFlower · 19 hours ago
  48. 373e68f [XLA:GPU] Change SDC names to more descriptive ones by Marcin Radomski · 19 hours ago
  49. deac368 [XLA:GPU] Unconditionally emit func.func from triton emitter. by Will Froom · 20 hours ago
  50. ba9c639 [XLA:CPU][XTile] Create simple lowering for tiled ops. by Will Froom · 20 hours ago
  51. 9d2df1c [Triton] Fixing getLastInductionValue utility to also accept Index type. This would otherwise crash when warp specialization is enabled. by Mohammed Anany · 20 hours ago
  52. b31aff7 Update mesh definition to better match it's use cases of querying tile index from device id's or vice-versa. Refactor into separate classes. by Kanish Anand · 21 hours ago
  53. 5dd154d Automated Code Change by A. Unique TensorFlower · 21 hours ago
  54. e17b782 NFC: Simplify padding logic in Split-K GEMM rewriter. by Christian Sigg · 21 hours ago
  55. 72dc641 [XLA:GPU] Add experimental buffer checksum tracing by Marcin Radomski · 21 hours ago
  56. 4154902 Remove obsolete Triton LLVM integration patches. by A. Unique TensorFlower · 21 hours ago
  57. 379d3eb Set call result shardings to the out shardings of func that is created or found from cache. by A. Unique TensorFlower · 22 hours ago
  58. 6ac8b2e [xla:gpu] Add padding to split-k to allow pipelining. by Christian Sigg · 22 hours ago
  59. ba08bd8 Update GraphDef version to 2382. by A. Unique TensorFlower · 22 hours ago
  60. acd4869 compat: Update forward compatibility horizon to 2025-10-16 by A. Unique TensorFlower · 22 hours ago
  61. c1e8fc6 [XLA:GPU] Add abstract class for multicast memory to GpuExecutor. by A. Unique TensorFlower · 23 hours ago
  62. 4dab5ef Add proto [de]serialization for `Memset32BitValueThunk`. by Aliia Khasanova · 23 hours ago
  63. 83f3904 [XLA:GPU] Consider multi-output fusions supported by Triton codegen. by Adrian Kuegel · 24 hours ago
  64. 6c44013 [XLA:GPU] Move AsyncStreamKind and CollectiveOpGroupMode to xla_data.proto. by Alexander Belyaev · 24 hours ago
  65. 06ea670 [XLA:GPU] Run hlo lit tests on several GPU platforms. by Adrian Kuegel · 25 hours ago
  66. b63d273 Automated Code Change by A. Unique TensorFlower · 26 hours ago
  67. 2edf355 Sort op's first operand is now generated without duplicates if the by A. Unique TensorFlower · 26 hours ago
  68. 7b8e21a Automated Code Change by A. Unique TensorFlower · 26 hours ago
  69. 12afb3d Automated Code Change by A. Unique TensorFlower · 26 hours ago
  70. 6872f47 Automated Code Change by A. Unique TensorFlower · 26 hours ago
  71. 88d1adf Automated Code Change by A. Unique TensorFlower · 27 hours ago
  72. 490206b Automated Code Change by A. Unique TensorFlower · 27 hours ago
  73. d33383d Introduce `tsl::WithCurrentContext` for capturing the current context. by A. Unique TensorFlower · 27 hours ago
  74. 4df1a3c [XLA:MSA] When block prefetching, finalize the original value if a sliced value is prefetched successfully and the original value is not. by Subhankar Shah · 29 hours ago
  75. 55371df [PjRt-IFRT] `ifrt::PjRtArray::pjrt_layout()` uses `nullptr` to indicate a default layout by Hyeontaek Lim · 30 hours ago
  76. 0c8f3ea Change EnterHostCallback() and by Parker Schuh · 30 hours ago
  77. 61785a4 [xla:ffi] Add a test for automatic FFI handler signature inference from C++ function by Eugene Zhulenev · 30 hours ago
  78. 8f60f24 Go: Update generated wrapper functions for TensorFlow ops. by A. Unique TensorFlower · 31 hours ago
  79. 57a6012 [tf2xla] Move allocator testing to allocator_test.cc by Eugene Zhulenev · 31 hours ago
  80. 0ce64af Introduce `HERMETIC_PYTHON_VERSION_KIND` for the Bzlmod build. by Yun Peng · 32 hours ago
  81. 16f7038 Support the Shardy dialect in ConvertSerializedStableHloModuleToBfloat16. by A. Unique TensorFlower · 32 hours ago
  82. a6a11a603 Implement StreamExecutorGpuClient::ScheduleRemoteSend. This allows migrating by Parker Schuh · 32 hours ago
  83. dd90f5f [XLA:GPU][codegen] Emit stablehlo for iota and implement lowering of stablehlo.iota to tt.make_range by Karlo Basioli · 33 hours ago
  84. 644b4a8 Replace `stream->BlockHostUntilDone()` with `BlockHostUntilDoneWithHostCallback()`. by A. Unique TensorFlower · 33 hours ago
  85. e2bc3e5 Update tfl.transpose version inconsistency in register_ref.cc by Majid Dadashi · 34 hours ago
  86. 2582934 [XLA:GPU] Add verbose tracing for BlockHostUntilDone and stream synchronization by Frederik Gossen · 34 hours ago
  87. f147bdd Extract launch information from the Triton compilation pipeline and use it instead of XLA's calculation. This is necessary in cases where the pipeline overrides the expected launch configuration. by Mohammed Anany · 35 hours ago
  88. c265f58 Integrate LLVM at llvm/llvm-project@267fa8dd1efc by A. Unique TensorFlower · 35 hours ago
  89. ccd8759 [XLA:CPU] Use asm to set name of intrinsic generated IR functions. by Sean Talts · 35 hours ago
  90. 2fe50c5 Fix MacOS nightly wheel builds by adding h5py version limit. by Yulia Baturina · 35 hours ago
  91. 3c7395e [XLA:CPU] Fix intrinsic library failing when passed an already vectorized call. From Will Froom. by Sean Talts · 35 hours ago
  92. 1108cc9 Add proto [de]serialization for CholeskyThunk. by Aliia Khasanova · 35 hours ago
  93. 9e80aec Make file handling utilities compatible with files larger than 4GiB on 32 bit Windows. by Quentin Khan · 36 hours ago
  94. 23736ec [XLA:CPU] Add test showing exp intrinsic vectorizations. by Sean Talts · 2 days ago
  95. 948d0df [XLA:GPU][codegen] Emit tensor dialect for bitcast and implement lowering of bitcast from tensor dialect to triton. by Karlo Basioli · 2 days ago
  96. 503198f [xla:cpu] Construct BufferAllocationInfo from BufferAssignment by Eugene Zhulenev · 2 days ago
  97. 7e4627b Add support for kTfLiteInt2 type export/import. by Majid Dadashi · 2 days ago
  98. 6c32106 Integrate Triton up to [de2ba394](https://github.com/openai/triton/commits/de2ba3946bc2a7a55ad77331ed60fd7c685156bf) by A. Unique TensorFlower · 2 days ago
  99. b545b61 [XLA:GPU] Provide functions to setup multicast from a single process. by A. Unique TensorFlower · 2 days ago
  100. 0ab9f48 Refactor `SelectKThunk` to accept `ThunkInfo` instead of `HloInstruction` pointer. by Aliia Khasanova · 2 days ago