190 Pull requests merged by 5 people

• Bump the github-actions group with 2 updates

  #96243 merged Jul 4, 2025

• Add some very slow algorithms to the cuDNN algorithm deny list

  #96392 merged Jul 4, 2025

• Remove GOOGLE_CUDA macro usage from conv_algorithm_picker

  #96377 merged Jul 4, 2025

• Make HloAlgorithmDenylist more robust and more testable

  #96375 merged Jul 4, 2025

• [xla:cpu] Delete unused files

  #96341 merged Jul 4, 2025

• [XLA] Fix typo in LHS causing bug

  #96288 merged Jul 4, 2025

• Modify MLIR location of TFLite op to contain TFLite tensor names

  #95992 merged Jul 4, 2025

• Remove tsl_workspace*.bzl files in XLA.

  #96324 merged Jul 3, 2025

• Optimize mlir::ConvertMlirHloToHlo by moving the proto from the XlaComputation local variable in mlir::ConvertToHloModule::ConsumeMainProto.

  #96343 merged Jul 3, 2025

• [IFRT Proxy] Change loaded host callback polling to use continuation to avoid

  #96266 merged Jul 3, 2025

• PR #28471: [XLA:GPU] allow lowering DynamicMemCopy thunk when it depends on loop iteration

  #96338 merged Jul 3, 2025

• PR #28497: Create cudnn handle for all threads to reuse during cudnn compilation

  #96383 merged Jul 3, 2025

• [XLA:GPU] NFC: slightly improve TritonXLAExtractInsertToTritonPass.

  #96380 merged Jul 3, 2025

• PR #28487: chore: add Arm Limited to AUTHORS

  #96379 merged Jul 3, 2025

• Optimize DiceSimLimitedSubgraph by pre allocating vectors.

  #96316 merged Jul 3, 2025

• Mark PjRtFuture::IsReady and PjRtFuture::IsKnownReady as const

  #96352 merged Jul 3, 2025

• kernels: Remove use of icu::UnicodeStringAppendable

  #96194 merged Jul 3, 2025

• Integrate LLVM at llvm/llvm-project@696c0f92e0fe

  #96330 merged Jul 3, 2025

• Port ThunkPassPipeline with CommandBufferConversionPass into gpu_compiler. Introduce a flag, that switches creation of command buffers from HLO level to Thunk level. This is a temporary flag to provide a smooth transition.

  #95801 merged Jul 3, 2025

• [XLA:GPU] do not fail in nest gemm fusion if we cannot convert the computation

  #96314 merged Jul 3, 2025

• [XLA:GPU] Add DotOperandDims::InsertDimension()

  #96310 merged Jul 3, 2025

• #sdy #mixed_serialization don't make JAX export use SdyRoundTripExportPipeline to stringify attributes and convert ops to StableHLO CustomCallOps and back.

  #94871 merged Jul 3, 2025

• [XLA:GPU] Remove ScheduleGpuModuleWithMemoryScheduler from public API.

  #96358 merged Jul 3, 2025

• [XLA:GPU] Extract common pattern in combiner tests.

  #96306 merged Jul 3, 2025

• [XLA:GPU] Implement upwards tile propagation for ConcatenateOp.

  #96294 merged Jul 3, 2025

• [XLA] Add ShapeUtil::InsertDimensionAtIndex.

  #96309 merged Jul 3, 2025

• Add tfl-optimize pass

  #96269 merged Jul 3, 2025

• [XLA:GPU] Annotate module with suggested combiner threshold and use it in combiners.

  #96305 merged Jul 3, 2025

• Move expensive variables on their last use to avoid copies.

  #96360 merged Jul 3, 2025

• [XLA:GPU] Add triton support test for sort

  #96250 merged Jul 3, 2025

• [HLO Diff] Replace usages of boost::bimap with a custom bidirectional map.

  #96276 merged Jul 3, 2025

• [XLA:CPU][nanort] Store program shape in NanoRtExecutable.

  #96336 merged Jul 3, 2025

• [XLA:GPU] nest gemm fusion test: separate tests that are parametrized on reshape op

  #96313 merged Jul 3, 2025

• [HLO Diff] Optimize Ancestor LCS implementation by reducing memory allocations.

  #96315 merged Jul 3, 2025

• Automated Code Change

  #96296 merged Jul 3, 2025

• Reverts 60077e5a6c50a9ad655fab6879e2bb432475b57b

  #96112 merged Jul 3, 2025

• Fix XLA crash when the product of tensor dimensions overflows.

  #96337 merged Jul 3, 2025

• [xla:cpu] Parallelize compiled dot along the batch dimension

  #96322 merged Jul 3, 2025

• [xla:cpu] Add batched gemv benchmark

  #96334 merged Jul 3, 2025

• Add CommonPjRtBufferImpl::logical_on_device_shape()

  #95708 merged Jul 3, 2025

• [xla:cpu] Add gemv benchmark

  #96277 merged Jul 3, 2025

• Remove extra space in stablehlo.while pretty print

  #96127 merged Jul 2, 2025

• [xla:cpu] Remove accidental VLOG from raw_buffer

  #96331 merged Jul 2, 2025

• Make IsDeleted() a const method

  #96159 merged Jul 2, 2025

• Return IllegalArgumentError instead of InternalError when deserializing DeviceAssignmentProto. Some of the users have special treatment for non InternalError cases.

  #96270 merged Jul 2, 2025

• Add a tuple sharding when creating get-tuple-element(tuple(single_result)).

  #96080 merged Jul 2, 2025

• PR #28360: [GPU] Collectives prefer stream id over stream kind

  #96327 merged Jul 2, 2025

• Separate stacks for python frames and c frames.

  #96255 merged Jul 2, 2025

• Further lower the max bytes threshold used by the Proto Splitter

  #96278 merged Jul 2, 2025

• [XLA:GPU] Fix host-offloading case for unified estimator.

  #96292 merged Jul 2, 2025

• PR #24579: [XLA:CPU][oneDNN] Add SiLU activation for oneDNN contractions

  #96245 merged Jul 2, 2025

• [XLA] Clarify memory is HBM related, Fix description of xla_tpu_rwb_fusion, remove xla_tpu_overlap_compute_collective_tc from the public doc.

  #96317 merged Jul 2, 2025

• PR #28402: Bump github/codeql-action from 3.29.0 to 3.29.1

  #96301 merged Jul 2, 2025

• Add estimation for fragmentation size when computing peak memory usage.

  #95635 merged Jul 2, 2025

• Add the testcase to validate that collective is used when spmd partitioner handles cross-partition large pad.

  #96258 merged Jul 2, 2025

• [xla:cpu] NFC: Move XfeedManager to new location

  #96267 merged Jul 2, 2025

• Automated Code Change

  #96300 merged Jul 2, 2025

• Change HloModule::add_frontend_attributes to allow adding a single kv pair

  #96251 merged Jul 2, 2025

• Remove --xnnpack_slinky option

  #96287 merged Jul 2, 2025

• Automated Code Change

  #96158 merged Jul 2, 2025

• Set half_pixel_centers to negation of align_corners in image resize composite legalization

  #96257 merged Jul 2, 2025

• [XLA:GPU] Disable horizontal input fusion.

  #96248 merged Jul 2, 2025

• [XLA:GPU] Add PipeliningCollectiveAnalyzer.

  #95480 merged Jul 2, 2025

• Implement RaggedDot in HloEvaluator

  #95946 merged Jul 2, 2025

• [XLA][Numerics][HLO Value Tracking] Add the original value recovery table

  #96188 merged Jul 2, 2025

• Automated Code Change

  #96237 merged Jul 2, 2025

• [xla] Add an AUTHORS file.

  #96198 merged Jul 2, 2025

• Automated Code Change

  #96240 merged Jul 2, 2025

• Fix a bug where BufferFromHostLiteral events are not fulfilled upon allocation failures

  #96281 merged Jul 2, 2025

• [XLA] Refactoring Reduce Window Rewriter to reduce complexity

  #96043 merged Jul 2, 2025

• Remove redundant logs and cast.

  #96212 merged Jul 2, 2025

• [XLA][SC] Add additional flags to the backend config for the sparsedense grad ops to identify how many values are in the various tuples.

  #96256 merged Jul 2, 2025

• [xla] Temporarily skip tests in //xla/tools/hlo_diff on CI due to service outage

  #96268 merged Jul 2, 2025

• Refactor the code for getting a sharding attribute from mlir::TF::XlaShardingOp.

  #96081 merged Jul 2, 2025

• [XLA:CPU] Refactor simple_jit_runner for better benchmarking.

  #96259 merged Jul 1, 2025

• is_ci_build: Move from XLA to TSL.

  #96092 merged Jul 1, 2025

• Ensure there are no isolated TPUDummyInput nodes left behind after simplify_ici_dummy_variables_pass.

  #96264 merged Jul 1, 2025

• Enable tests and fix for duplicate registration tests error

  #96116 merged Jul 1, 2025

• Move tensorflow/third_party/triton to xla/third_party/triton

  #95981 merged Jul 1, 2025

• [SparseReshapeOp] Microoptimization: Don't call OpKernelContext::input() multiple times.

  #96249 merged Jul 1, 2025

• Do not build xnn fusions inside reducers.

  #96219 merged Jul 1, 2025

• Add a test case for int4 scatter with unique_indices=true

  #96247 merged Jul 1, 2025

• Pass AliasInfo to HloAliasAnalysis.

  #95797 merged Jul 1, 2025

• [XLA:GPU] Deprecate CUBLAS_GEMM_DEFAULT_TENSOR_OP.

  #96186 merged Jul 1, 2025

• [oneDNN][CPU] fuse a matmul pattern

  #86172 merged Jul 1, 2025

• [XLA:GPU] check if producer / consumer can be fused in CanFuseTriton

  #96184 merged Jul 1, 2025

• [XLA:CPU] Set loaded workgroup as invariant & fix incorrect alignment metadata.

  #96067 merged Jul 1, 2025

• [XLA:GPU] Fix the implementation of the TMA viability filter to conform to the Nvidia documentation

  #96185 merged Jul 1, 2025

• Add use_shardy_partitioner in TPUCompileMetadataProto and TpuAotCompilationOptions. The default value is false.

  #95808 merged Jul 1, 2025

• [XLA] Add after target rule kDelay rule and use it in aggressive flexible scheduling

  #96173 merged Jul 1, 2025

• unicode_ops: Encode noncharacters

  #96193 merged Jul 1, 2025

• Fix compilation error in tensorflow/python/tfcompile_wrapper.cc on s390x

  #95322 merged Jul 1, 2025

• Automated Code Change

  #96156 merged Jul 1, 2025

• Automated Code Change

  #96154 merged Jul 1, 2025

• PR #28394: [XLA:GPU] Add DynamicSliceCopyFusion command to command buffer.

  #96196 merged Jul 1, 2025

• Add CreateRawAliasOfBuffer() to CommonPjRtBufferImpl.

  #95618 merged Jul 1, 2025

• Simplify ComputationPlacer interface.

  #96102 merged Jul 1, 2025

• Add initial support for reductions in XnnGraphFusion.

  #96140 merged Jul 1, 2025

• Fix invalid process count bug?

  #96091 merged Jul 1, 2025

• Fix: Shardy should propagate use_auto_spmd_partitioning field from config

  #96206 merged Jul 1, 2025

• Update version of rules_ml_toolchain repo.

  #96202 merged Jul 1, 2025

• 1. Limit the max-width and max-height of a cell.

  #96189 merged Jun 30, 2025

• Integrate StableHLO at openxla/stablehlo@9df3b556

  #96200 merged Jun 30, 2025

• Refactor py_import macros to avoid unpacking pypi wheels twice.

  #95889 merged Jun 30, 2025

• Announcement about deprecating tf lite within TF.

  #96204 merged Jun 30, 2025

• add no_oss_py313 tag to following max python test

  #96201 merged Jun 30, 2025

• Check operands of elementwise ops offloaded to xnnpack.

  #96161 merged Jun 30, 2025

• Added a XLA Flags Guidance doc

  #95699 merged Jun 30, 2025

• Remove device shape & device size bytes functions from HLO runners.

  #96045 merged Jun 30, 2025

• [xla:pjrt] Add PjRtFuture::TryMap to map futures when map functor itself can fail

  #96197 merged Jun 30, 2025

• [xla:cpu] Decouple xfeed manager from legacy cpu_runtime

  #96096 merged Jun 30, 2025

• Rollback due to suspected correctness issues.

  #96192 merged Jun 30, 2025

• [XLA] Add a length limit on op_name when concatenating in call_inliner.

  #96191 merged Jun 30, 2025

• Remove the auto assignment list

  #96174 merged Jun 30, 2025

• Don't handle empty reductions in XNNPACK delegate

  #96105 merged Jun 30, 2025

• [XLA:GPU] Record reification cost only for supported instructions.

  #95960 merged Jun 30, 2025

• [XLA:CPU] Add dereferenceable metadata to loaded argument pointer.

  #96070 merged Jun 30, 2025

• [XLA:GPU] Code cleanups:

  #96182 merged Jun 30, 2025

• Fix a problem in Shape::Equal in comparing buffer types.

  #95910 merged Jun 30, 2025

• [XLA:GPU] Make gpu_hlo_schedule testing cover unified latency estimator cases.

  #95588 merged Jun 30, 2025

• [XLA:CPU][XLA:GPU] Use llvm array types for GEP when lowering tensors.

  #96015 merged Jun 30, 2025

• Added support for the following to allow PtrVec to replace

  #96109 merged Jun 30, 2025

• [XLA:GPU] add lock guards for containers in priority_fusion

  #96181 merged Jun 30, 2025

• Make XNNPack weight cache work with files bigger than 2 GiB on Windows.

  #96090 merged Jun 30, 2025

• PR #26217: [XLA:GPU] Add Intel GPU specific tags for xla and sysl_status component

  #96179 merged Jun 30, 2025

• Automated Code Change

  #96168 merged Jun 30, 2025

• [XLA:GPU] Enforce compile-time exhaustiveness of GemmFusionAutotuner::BackendConfig visitors.

  #96178 merged Jun 30, 2025

• Automated Code Change

  #96155 merged Jun 30, 2025

• Automated Code Change

  #96167 merged Jun 30, 2025

• Automated Code Change

  #96151 merged Jun 30, 2025

• [HLO Graph Dumper] Add the computation name to the op name when annotating parameters.

  #95978 merged Jun 30, 2025

• PR #25298: [XLA:GPU] cudnn sdpa flex attention

  #96176 merged Jun 30, 2025

• [XLA:GPU] Lifting int4_pass above TritonXlaExtractInsertPass. This is necessary to allow both the legacy Triton emitter and the Generic emitter to correctly handle int4. This also includes accommodating new instructions in the pass.

  #95944 merged Jun 29, 2025

• Automated Code Change

  #96145 merged Jun 29, 2025

• Do not propagate name_stacks into lower_jaxpr_to_fun.

  #96089 merged Jun 28, 2025

• Populate the HLO op_type field from JAX StableHLO.

  #96085 merged Jun 28, 2025

• Automated Code Change

  #96136 merged Jun 28, 2025

• Refactor checks in xnn_fusion.cc/xnn_emitter.cc.

  #96126 merged Jun 28, 2025

• Automated Code Change

  #96132 merged Jun 28, 2025

• Automated Code Change

  #96012 merged Jun 28, 2025

• Porting the grappler batch prioritization rewriter to TFRT.

  #96022 merged Jun 28, 2025

• Automated Code Change

  #96121 merged Jun 28, 2025

• Add test and refactor Device Assignment.

  #94904 merged Jun 28, 2025

• Nit, ifrt proxy: additional logging to debug connection failures.

  #96120 merged Jun 28, 2025

• [xla:pjrt] Add PjRtFuture::Map API to map PjRt futures

  #96115 merged Jun 28, 2025

• Moving two functions from dot_handler.cc to spmd_partitioner_util.cc to be used by other files.

  #95725 merged Jun 28, 2025

• [XLA:benchmarks] Add README guide for onboarding new benchmarks to OpenXLA.

  #95275 merged Jun 28, 2025

• Remove JAX_IFRT_VERSION_NUMBER check for types in XLA/JAX

  #96035 merged Jun 28, 2025

• Fix typo in custom_call.md

  #95976 merged Jun 28, 2025

• Reduce latency by copy constructing device id vector directly rather than using for loop to create the vector followed by move.

  #96117 merged Jun 28, 2025

• #HLODiff Don't print parameter numbers in HLO canonical fingerprint.

  #96113 merged Jun 28, 2025

• [IFRT] Rename Client::GetDefaultLayout() to Client::GetDefaultPjRtLayout()

  #96108 merged Jun 28, 2025

• [XLA:MSA] Decrease copy resource scaling from 2^50 to 2^40. The 2^50 scaling factor sometimes causes total scores to exceed the max value of int64_t causing ubsan/asan errors. Enable the test failing ubsan errors.

  #96027 merged Jun 27, 2025

• [XLA] Add a Hlo print option for printing ENTRY keyword

  #95970 merged Jun 27, 2025

• Cleanup: rename variable_index to variable_arg_index in IfrtServingExecutable to improve code readability

  #96107 merged Jun 27, 2025

• Fix handling of None values with as_numpy_iterator.

  #96009 merged Jun 27, 2025

• [IFRT] Rename Array::layout() to Array::pjrt_layout()

  #96099 merged Jun 27, 2025

• Upgrade Bazel version in tf_keras build script.

  #96101 merged Jun 27, 2025

• Schedule computation in post order.

  #96095 merged Jun 27, 2025

• tsl: Add config_setting for running in CI.

  #96097 merged Jun 27, 2025

• Integrate LLVM at llvm/llvm-project@7a3356951053

  #96023 merged Jun 27, 2025

• Disable internal precompilation for local runs.

  #96029 merged Jun 27, 2025

• Add initial support for broadcasts in XnnGraphFusion.

  #95912 merged Jun 27, 2025

• add tpu_dynamic_registration

  #96040 merged Jun 27, 2025

• [xla:cpu:onednn] Enable matching dot + eltwise to oneDNN fusions in DotLibraryRewriter

  #96094 merged Jun 27, 2025

• [XLA] Handle sub-computations more correctly in constant folding.

  #96093 merged Jun 27, 2025

• Integrate StableHLO at openxla/stablehlo@955fa7e6

  #96031 merged Jun 27, 2025

• [xla:cpu] Delete ffi support from legacy runtime

  #96049 merged Jun 27, 2025

• [XLA] Slightly simplify explanation message code in HLO constant floding.

  #96082 merged Jun 27, 2025

• [xla:cpu:onednn] Refactor onednn_fusion. Put things that depend on oneDNN Graph in a separate file.

  #96086 merged Jun 27, 2025

• Split utilities for the weight cache code/tests to their own files.

  #96013 merged Jun 27, 2025

• Update xla_test to set the device as a specific GPU (e.g. h100, a100, etc) and update uses of xla_test_backend_predicates.h utilities to reflect this change

  #95983 merged Jun 27, 2025

• [xla] Remove unused thunk type

  #96068 merged Jun 27, 2025

• [xla:cpu] Delete unused cpu runtime symbols

  #96048 merged Jun 27, 2025

• [XLA] Factor out foldability check in HLO constant folding.

  #96084 merged Jun 27, 2025

• [xla:gpu] Add support for rendering command buffer execution graphs

  #96042 merged Jun 27, 2025

• [XLA] Add call folding test to constant propagation.

  #96083 merged Jun 27, 2025

• [XLA:GPU] Refactor code figuring out a support for unified latency estimator.

  #95887 merged Jun 27, 2025

• [XLA:GPU] Document interpolation API.

  #96072 merged Jun 27, 2025

• Fix an error in GetDotGroupPartitionContractingOutputShardings in SPMD dot handler.

  #96076 merged Jun 27, 2025

• [xla:copy_insertion] Fixed a problem in finding a rotated non-copyable chain.

  #95905 merged Jun 27, 2025

• PR #28225: [ROCm] Fix invalid rocblas version for rocm 7

  #95996 merged Jun 27, 2025

• Remove the old Triton search space code now that the new search space landed.

  #96004 merged Jun 27, 2025

• [xla][gpu] Nest gemm fusion: Update result shape and operand shape consistently.

  #96058 merged Jun 27, 2025

• PR #27836: Add support for NVSHMEM put/get

  #96074 merged Jun 27, 2025

• [XLA:GPU] Don't use optional for PropagateTileToInput[Broadcast|Transpose|Slice].

  #96078 merged Jun 27, 2025

• [XLA:GPU] don't fail in autotuner if a specific config does not fly

  #96077 merged Jun 27, 2025

• Iterate through all fusion heroes when looking for a transpose emitter.

  #96014 merged Jun 27, 2025

• [XLA:GPU] Create input buffers from executable instead of the root instruction.

  #96007 merged Jun 27, 2025

• Reverts fcdc6de87afe454721a374b503899ea7a9f05360

  #96037 merged Jun 27, 2025

98 Pull requests opened by 6 people

• Migrate from CanShareBuffer hook to AliasInfo.

  #96075 opened Jun 27, 2025

• [XLA:GPU] Flip unified cost model on.

  #96087 opened Jun 27, 2025

• strict_cc_test: Only set --gtest_fail_if_no_test_selected on CI.

  #96088 opened Jun 27, 2025

• Eliminate circular dependency between device assignment and Computation Placement. Use Device Assignment for looking up logical and global ids.

  #96103 opened Jun 27, 2025

• [xla:cpu] Remove infeed/outfeed support from XLA:CPU

  #96104 opened Jun 27, 2025

• Remove all uses of `xla_test_library`

  #96106 opened Jun 27, 2025

• [XLA:SCHEDULING] Defer scheduling of allocate buffer custom calls

  #96110 opened Jun 27, 2025

• Add an async version of FunctionalHloRunner::Run that runs a pjrt executable.

  #96111 opened Jun 27, 2025

• Update VLOG message to make it more accurate.

  #96114 opened Jun 27, 2025

• [Phase Compilation] Part-3: Augment extension with buffer deletion callback.

  #96118 opened Jun 28, 2025

• Add traceme for IFRT executable launch

  #96119 opened Jun 28, 2025

• Integrate LLVM at llvm/llvm-project@67a5fc8e12dc

  #96123 opened Jun 28, 2025

• Automated Code Change

  #96143 opened Jun 29, 2025

• Automated Code Change

  #96149 opened Jun 29, 2025

• PR #19067: [XLA:CPU][oneDNN] Move simplification pass before oneDNN pass

  #96177 opened Jun 30, 2025

• Update protobuf to v6.30.1

  #96183 opened Jun 30, 2025

• Upgrade Abseil to [LTS 20250127.1](https://github.com/abseil/abseil-cpp/releases/tag/20250127.1).

  #96187 opened Jun 30, 2025

• build: Set is_ci flag in OSS CI.

  #96190 opened Jun 30, 2025

• #sdy Move `AddAxisOrMergeInserter` to Shardy OSS utils.

  #96195 opened Jun 30, 2025

• Example Extension with cpp dependency injection

  #96199 opened Jun 30, 2025

• Experiment: Try to remove lite rt dependencies from TF wheel.

  #96205 opened Jun 30, 2025

• Fix TPU detection in Google Colab V2 environments

  #96208 opened Jun 30, 2025

• Construct XLA GPU client with coordination service client.

  #96210 opened Jul 1, 2025

• Use schedule order to calculate resource usage.

  #96211 opened Jul 1, 2025

• Integrate LLVM at llvm/llvm-project@67a5fc8e12dc

  #96213 opened Jul 1, 2025

• Add 'mode' attribute to AllReduce and ReduceScatter.

  #96214 opened Jul 1, 2025

• [XLA:GPU] Remove horizontal fusion.

  #96241 opened Jul 1, 2025

• PR #19067: [XLA:CPU][oneDNN] Move simplification pass before oneDNN pass

  #96244 opened Jul 1, 2025

• Bump urllib3 from 2.4.0 to 2.5.0 in /ci/official/requirements_updater/numpy1_requirements

  #96246 opened Jul 1, 2025

• Add AutotunerPass.

  #96252 opened Jul 1, 2025

• [HLO Diff] Precompute HLO values dependencies for each instruction in graph.

  #96253 opened Jul 1, 2025

• #sdy Clean up `AddAxisOrMergeInserter` in dedup_meshes

  #96254 opened Jul 1, 2025

• Integrate with `rules_ml_toolchain`

  #96260 opened Jul 1, 2025

• PR #28314: [ROCm] added allreduce kernel registration

  #96261 opened Jul 1, 2025

• kernels: Remove use of icu::UnicodeStringAppendable

  #96262 opened Jul 1, 2025

• Pass to change arg attrs

  #96263 opened Jul 1, 2025

• Simplify HloComputationGraphMatcher by re-using leaf matcher for constant and params.

  #96265 opened Jul 1, 2025

• [XLA] Refactoring Reduce Window Rewriter to reduce complexity

  #96271 opened Jul 1, 2025

• PR #28314: [ROCm] added allreduce kernel registration

  #96272 opened Jul 1, 2025

• Reverts 628eaad6c61ef9f8586658a01147df8028d1132c

  #96273 opened Jul 1, 2025

• #HLODiff Add lcs parent similarity to MatchInstructionsWithMultipleCandidates

  #96274 opened Jul 1, 2025

• If a host xpose copy goes into a host-call, the copy cannot be moved.

  #96275 opened Jul 1, 2025

• [xla] test CI

  #96280 opened Jul 2, 2025

• Add support for deserializing xplanes to Jaxlib

  #96282 opened Jul 2, 2025

• Automated Code Change

  #96289 opened Jul 2, 2025

• Exclude engine_id 0 from cuDNN autotuning if alternatives are available

  #96303 opened Jul 2, 2025

• [XLA:GPU] Support command buffer conversion for async operations on thunks level.

  #96307 opened Jul 2, 2025

• Return kTfLiteDelegateError if delegate_kernel is null

  #96308 opened Jul 2, 2025

• Automated Code Change

  #96311 opened Jul 2, 2025

• [Hlo Diff] Simplify ExactSubgraphMatcher impl.

  #96312 opened Jul 2, 2025

• code cleanup: shorten variable names

  #96318 opened Jul 2, 2025

• [xla:cpu] Make DotLibraryRewriter support greedy fusion mode.

  #96319 opened Jul 2, 2025

• [TFLite] Fix incorrect `FuseBinaryOpWithTransposeConvNoneBias` optimization

  #96320 opened Jul 2, 2025

• Use `-Wl,-install_name` for CPU plugin on MacOS

  #96321 opened Jul 2, 2025

• Implement slinky thread pool using `Eigen::ThreadPoolInterface`. This should work, and be *reasonably* efficient, but it is not as efficient as slinky::thread_pool_impl would be directly. This is effectively an extra layer of task overhead to run each worker task (it should still better than enqueuing a new task for each iteration of a parallel loop).

  #96323 opened Jul 2, 2025

• [XLA] Add should_inline callback to CallInliner

  #96325 opened Jul 2, 2025

• Reproduce an OSS test failure

  #96326 opened Jul 2, 2025

• [XLA] Refactor HloDCE to improve readability.

  #96328 opened Jul 2, 2025

• Erros with missing .so files when running wheel test

  #96332 opened Jul 2, 2025

• Add TraceMes to PyGrain for measuring input wait time on host and attributing device idleness to input pipeline slowness.

  #96335 opened Jul 2, 2025

• fixed missing .so files in tf wheel audit

  #96340 opened Jul 3, 2025

• [Autotuner] Add block level emitter backend for Triton fusion.

  #96342 opened Jul 3, 2025

• Automated Code Change

  #96350 opened Jul 3, 2025

• [XLA:CPU][nanort] Test that compilation pipeline propagates int4 packing information to executable.

  #96359 opened Jul 3, 2025

• [XLA:CPU] Add tile size to WorkDimensions.

  #96361 opened Jul 3, 2025

• [XLA:CPU][XLA:GPU] Use nsw rather than nuw

  #96362 opened Jul 3, 2025

• Normalize node names to be consistent with tfdbg dump filepath behavior. Ultimately tfdbg calls [TranslateName](https://source.corp.google.com/piper///depot/google3/third_party/tensorflow/compiler/xla/tsl/platform/file_system.cc;rcl=779041561;l=64) which then calls [CleanPath](https://source.corp.google.com/piper///depot/google3/third_party/tensorflow/compiler/xla/tsl/platform/file_system.cc;rcl=779041561;l=376) when dumping tensors which normalizes the filepath

  #96363 opened Jul 3, 2025

• [XLA:CPU] Don't enable verifier by default.

  #96365 opened Jul 3, 2025

• #sdy Fix forward of making XLA C++ changes so we can fall back to GSPMD in JAX export if the loaded module was lowered for GSPMD.

  #96368 opened Jul 3, 2025

• [XLA:CPU][XLA:GPU] Add fast min/max rewrite pattern.

  #96369 opened Jul 3, 2025

• Automated Code Change

  #96370 opened Jul 3, 2025

• Call get_logical_on_device_shape within thread pool.

  #96372 opened Jul 3, 2025

• [XLA:CPU] Add pass to add loop unroll flags to nested scf.ForOps

  #96373 opened Jul 3, 2025

• [XLA:CPU] Add pass to peel loop for workgroups that cross the constraints.

  #96374 opened Jul 3, 2025

• [XLA:GPU] Don't convert thunks into command buffer, if amount of thunks is less then `xla_gpu_graph_min_graph_size`.

  #96376 opened Jul 3, 2025

• Use `absl::NoDestructor` in more places.

  #96378 opened Jul 3, 2025

• [XLA:CPU] Add flag to flatten the call graph after fusion.

  #96381 opened Jul 3, 2025

• Solve BUILD deps conflict for linear_solver

  #96382 opened Jul 3, 2025

• Sort and remove duplicate symbols.

  #96384 opened Jul 3, 2025

• Add missing includes.

  #96385 opened Jul 3, 2025

• Move rng_mutex to the only using method.

  #96386 opened Jul 3, 2025

• Add validation for label probability distribution in softmax_cross_entropy_with_logits

  #96387 opened Jul 3, 2025

• Automated Code Change

  #96388 opened Jul 4, 2025

• Automated Code Change

  #96389 opened Jul 4, 2025

• Automated Code Change

  #96390 opened Jul 4, 2025

• Automated Code Change

  #96391 opened Jul 4, 2025

• Automated Code Change

  #96393 opened Jul 4, 2025

• Automated Code Change

  #96394 opened Jul 4, 2025

• Automated Code Change

  #96395 opened Jul 4, 2025

• Automated Code Change

  #96397 opened Jul 4, 2025

• Automated Code Change

  #96398 opened Jul 4, 2025

• Integrate Triton up to [d25df428](https://github.com/openai/triton/commits/d25df428072a7374307c44376267a68e44b636d1)

  #96399 opened Jul 4, 2025

• Automated Code Change

  #96400 opened Jul 4, 2025

• Do not populate floating point data with nans

  #96402 opened Jul 4, 2025

• Introduce fatbin extractor tool that extracts fatbins from object files.

  #96403 opened Jul 4, 2025

• Add e2e ragged dot test for gpu

  #96404 opened Jul 4, 2025

• Automated Code Change

  #96405 opened Jul 4, 2025

• Remove unused parameter (NFC).

  #96406 opened Jul 4, 2025

11 Issues closed by 4 people

• Criminal Justice

  #96329 closed Jul 3, 2025

• Insufficient documentation on GPU use, especially in MPS

  #60647 closed Jul 3, 2025

• //tensorflow/python/ops/ragged:ragged_cross_op_test is flaky

  #60670 closed Jul 3, 2025

• api_compatibility_test fails on Python 3.11

  #60679 closed Jul 3, 2025

• typing_extensions >= 4.6.0 causes pip unit test failure

  #60687 closed Jul 3, 2025

• Mismatch Between Quantized TFLite Layer Outputs and Expected Mathematical Values When Using get_tensor()

  #93917 closed Jul 3, 2025

• tf.nn.conv2d with invalid input dims crashes in TF ≤2.19 — now raises InvalidArgumentError in nightly

  #95415 closed Jul 1, 2025

• My trained model detects almost every classes include target class.

  #96060 closed Jun 30, 2025

• erorr

  #96160 closed Jun 30, 2025

• Protobuf 4.24.0 break tensorflow and causes segfault with TF 2.12

  #61551 closed Jun 29, 2025

• Numpy and tf experimental Numpy differ in vander matrix creation case for N=0

  #60628 closed Jun 28, 2025

8 Issues opened by 7 people

• The native library arm64-v8a/libtask_audio_jni.so (from org.tensorflow:tensorflow-lite-task-audio:0.4.4) is not 16 KB aligned

  #96401 opened Jul 4, 2025

• Inefficient 2D convolution compared to PyTorch

  #96396 opened Jul 4, 2025

• crash when inference use libtensorflowlite_c.so and config threadnum >1 backend cpu

  #96347 opened Jul 3, 2025

• OSX : 'Chrome 137.0.7151.120' + 'TensorFlow.js v4.22.0' / webgpu - Err: "...Failed to execute 'mapAsync' on 'GPUBuffer'..."

  #96339 opened Jul 3, 2025

• how to get profile of per operation that delegate gpu opencl like cpu enable_op_profiling result, rather than only ModifyGraphWithDelegate?

  #96239 opened Jul 1, 2025

• Multiple segmentation faults and aborted in some modules

  #96209 opened Jun 30, 2025

• Unable to use Tensorflow and Tensorflow-TPU on Google Colab

  #96203 opened Jun 30, 2025

• Inconsistent `tf.math.reciprocal` Behavior for complex128 `inf` between CPU and GPU

  #96180 opened Jun 30, 2025

49 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [oneDNN] Add Infer after last allow to be added to Allow list

  #84874 commented on Jul 1, 2025 • 1 new comment

• build(aarch64): Update to oneDNN-3.7 + ACL-24.12 (fix)

  #93951 commented on Jul 3, 2025 • 0 new comments

• Fix comparison functions and add unit tests

  #94484 commented on Jun 30, 2025 • 0 new comments

• Move Mosaic CC sources into XLA

  #94877 commented on Jul 1, 2025 • 0 new comments

• [XLA] Add stack trace breakdown to `HloLiveRange::ToString` for peak memory usage

  #94954 commented on Jun 27, 2025 • 0 new comments

• Fix: Safely Capture and Store TF Operation Stack Trace at Creation to Prevent Dangling Reference Errors

  #95034 commented on Jul 1, 2025 • 0 new comments

• [Phase Compilation] Part-3: Add C++ layers to test and interact with C PJRT API.

  #95089 commented on Jun 27, 2025 • 0 new comments

• [PROTOTYPE] Cleanup TFL dependencies in tosa

  #95262 commented on Jul 1, 2025 • 0 new comments

• Remove old heartbeat options.

  #95294 commented on Jul 2, 2025 • 0 new comments

• Fix subprocess.check_output decoding issue in pip_smoke_test.py to handle byte output safely

  #95335 commented on Jul 1, 2025 • 0 new comments

• io_utils: prevent `input()` crash in non-interactive mode

  #95525 commented on Jul 1, 2025 • 0 new comments

• Expose Auto Sharding to StableHLO.

  #95535 commented on Jul 3, 2025 • 0 new comments

• [mlir][tosa] Support negative axis in gather lowering

  #95608 commented on Jul 1, 2025 • 0 new comments

• Added `PjrtClient::UpdateGlobalProcessInfo` method.

  #95611 commented on Jul 1, 2025 • 0 new comments

• Add `use_shardy_partitioner` in `tf.tpu.XlaOptions`, `TPUReplicateMetadata`. The default value is false.

  #95799 commented on Jul 1, 2025 • 0 new comments

• Update Protobuf to 6.30.1

  #95873 commented on Jul 3, 2025 • 0 new comments

• [xla:gpu] Handle buffer related custom calls.

  #95957 commented on Jun 30, 2025 • 0 new comments

• [ #HLODiff ] Filter ignored opcodes before rendering results.

  #95971 commented on Jul 2, 2025 • 0 new comments

• lite: Propagate an error during OpInit()

  #95989 commented on Jul 2, 2025 • 0 new comments

• Move `tensorflow/third_party/shardy` to `xla/third_party/shardy`

  #96028 commented on Jul 1, 2025 • 0 new comments

• Create new CI configurations

  #96034 commented on Jul 1, 2025 • 0 new comments

• Add attribute to configure different lane sizes in sparsecore preprocessing ops.

  #96036 commented on Jul 1, 2025 • 0 new comments

• [ #HLODiff ] Sort diff patterns by diff size instead of group size.

  #96044 commented on Jul 2, 2025 • 0 new comments

• [ #HLODiff ] Refine computation diff.

  #96046 commented on Jul 2, 2025 • 0 new comments

• Add random perturbations to the xla_tpu_msa_sort_order_overrides flag

  #96050 commented on Jun 28, 2025 • 0 new comments

• YoloX different Model Output for Python and Android

  #95489 commented on Jun 27, 2025 • 0 new comments

• /libtensorflow_cc.so.2: undefined symbol: _ZNK10tensorflow8OpKernel11TraceStringB5cxx11ERKNS_15OpKernelContextEb

  #95758 commented on Jun 30, 2025 • 0 new comments

• Aborted (core dumped) in `tf.image.non_max_suppression`\`tf.raw_ops.NonMaxSuppressionV2`\`tf.raw_ops.NonMaxSuppressionV3`\`tf.raw_ops.NonMaxSuppressionV4`

  #95760 commented on Jun 30, 2025 • 0 new comments

• Aborted (core dumped) in `tf.distribute.Server`

  #95762 commented on Jun 30, 2025 • 0 new comments

• how to use libtensorflowlite_c.so C API and delegate gpu opencl correctly?

  #95795 commented on Jul 1, 2025 • 0 new comments

• Aborted (core dumped) in `tf.raw_ops.QuantizedMaxPool`\`tf.raw_ops.QuantizedAvgPool`

  #95761 commented on Jul 2, 2025 • 0 new comments

• how to use libtensorflowlite_c.so C API and delegate gpu opencl correctly?

  #95634 commented on Jul 2, 2025 • 0 new comments

• `tf.linalg.solve` behaves inconsistently on GPU for singular matrices depending on shape

  #94657 commented on Jul 2, 2025 • 0 new comments

• Missing GPU op for zeros_like for RaggedTensorVariant, error occurs when Ragged Tensor fed thru tf.map_fn

  #46635 commented on Jul 2, 2025 • 0 new comments

• Loss suddenly becoming NaN on TPUs

  #93548 commented on Jul 2, 2025 • 0 new comments

• TPU nan issue

  #86953 commented on Jul 2, 2025 • 0 new comments

• It doesn't support on python3.13

  #78774 commented on Jul 2, 2025 • 0 new comments

• Runtime error when using empty axes for reduce operator

  #95663 commented on Jul 3, 2025 • 0 new comments

• graph execution error bug with tfm.nlp.layers.MultiHeadRelativeAttention

  #94599 commented on Jul 3, 2025 • 0 new comments

• TensorFlow Docker `tensorflow/tensorflow:latest-gpu` fails to detect GPU due to CUDA/cuDNN mismatch

  #94593 commented on Jul 4, 2025 • 0 new comments

• Incorrect gradient after divide operation when result contains inf

  #60695 commented on Jul 4, 2025 • 0 new comments

• can we customize memory allocation functions(like malloc/free) for inference with C api?

  #60662 commented on Jul 4, 2025 • 0 new comments

• standalone pip package for tf.io.gfile.GFile

  #45942 commented on Jul 4, 2025 • 0 new comments

• Fix compile error in tensorflow/python/tfcompile_wrapper.cc on s390x

  #87676 commented on Jun 30, 2025 • 0 new comments

• Fix: Ensure boolean_mask_v2() only accepts boolean dtype for mask

  #89370 commented on Jun 30, 2025 • 0 new comments

• Move duplicate CUDA/XLA registration logs from INFO to VLOG

  #89808 commented on Jun 30, 2025 • 0 new comments

• TfLite elementwise_ops add type support (#104)

  #92706 commented on Jun 30, 2025 • 0 new comments

• [XLA:GPU] NFC, refactor DotDecomposer.

  #93521 commented on Jul 2, 2025 • 0 new comments

• Stable delegate python api

  #93850 commented on Jul 1, 2025 • 0 new comments