issues
search
openxla
/
xla
A machine learning compiler for GPUs, CPUs, and ML accelerators
Apache License 2.0
2.74k
stars
440
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Reduce the number of runs for AllReduce test
#19847
shraiysh
closed
2 days ago
0
[xla:cpu] NFC: Extract RuntimeSymbolGenerator into a separate library
#19846
copybara-service[bot]
closed
1 day ago
0
Explicilty interface for AcquireExternalReference and WaitExternalReference
#19845
yliu120
opened
3 days ago
1
Integrate LLVM at llvm/llvm-project@b214ca82daee
#19844
copybara-service[bot]
opened
3 days ago
0
[XLA:CPU] Fix the bug in transposed convolution async execution.
#19843
copybara-service[bot]
closed
3 days ago
1
Reverts eab45d5da2fab59de8c02678c2b7b9ae69d9fef8
#19842
copybara-service[bot]
opened
3 days ago
0
#sdy bump version due to JAX MacOS breakage
#19841
copybara-service[bot]
closed
3 days ago
0
[XLA:GPU] Add intra-warp reduce of reduce test.
#19840
copybara-service[bot]
closed
3 days ago
0
Integrate LLVM at llvm/llvm-project@c0192a008c4a
#19839
copybara-service[bot]
closed
3 days ago
0
PR #18840: [NVIDIA] Support larger head dim for cudnn fmha
#19838
copybara-service[bot]
closed
2 days ago
0
Reverts 27352402f3c65c41e7c897138d6ad3e015a04014
#19837
copybara-service[bot]
closed
3 days ago
0
Replace gpu_asm_extra_flags string option by individual flags
#19836
copybara-service[bot]
opened
3 days ago
0
PR #19026: [NVIDIA GPU] LHS enhancement for multiple collective resources
#19835
copybara-service[bot]
closed
2 days ago
0
Add support for async dynamic slice fusion
#19834
shraiysh
opened
3 days ago
2
PR #19026: [NVIDIA GPU] LHS enhancement for multiple collective resources
#19833
copybara-service[bot]
opened
3 days ago
0
Make CudaExecutor::CreateDeviceDescrition not fail even if CUDA is not available
#19832
copybara-service[bot]
closed
3 days ago
0
Add DeferRelocatableCompilationCompilationProvider
#19831
copybara-service[bot]
closed
3 days ago
0
//xla/service/cpu:vectorized_reduce_with_no_vector_registers_test fails on Apple Silicon
#19830
majnemer
closed
6 hours ago
3
//xla/stream_executor/cuda:compilation_provider_test should be skipped by on CPU builds
#19829
majnemer
closed
3 days ago
2
PR #18840: [NVIDIA] Support larger head dim for cudnn fmha
#19828
copybara-service[bot]
opened
3 days ago
0
[XLA] Guarantee ordering of infeeds/outfeeds across called computations
#19827
copybara-service[bot]
opened
3 days ago
0
Clarify index parallel dims in gather/scatter instructions.
#19826
copybara-service[bot]
closed
21 hours ago
0
Fix `PropagateShardingAlongDimsAndReplicateOthers` and expose it as a public util function.
#19825
copybara-service[bot]
closed
3 days ago
0
//xla/tests:complex_unary_op_test_cpu fails on macOS Apple Silicon
#19824
majnemer
opened
3 days ago
3
Limit the number of stragglers we log to avoid `RESOURCE_EXHAUSTED` errors in the RPC layer from sending overly verbose errors.
#19823
copybara-service[bot]
closed
3 days ago
0
Automated Code Change
#19822
copybara-service[bot]
opened
3 days ago
0
[xla:cpu] Move TargetMachineFeatures to xla/backends/codegen
#19821
copybara-service[bot]
closed
3 days ago
0
Modify XlaOp Exp to accept result accuracy as an argument. We want to be able to select implementation of exp depending on this config.
#19820
copybara-service[bot]
opened
3 days ago
0
[XLA:Collective] Support normalizing all-reduce
#19819
copybara-service[bot]
opened
3 days ago
0
Update target_config to be a text proto and populate it on the
#19818
copybara-service[bot]
closed
3 days ago
0
Create op_metircs_to_record to deal with Roofline Analysis
#19817
copybara-service[bot]
closed
3 days ago
0
[XLA:GPU] Fix an ASAN error
#19816
copybara-service[bot]
closed
2 days ago
0
[IFRT] Add IFRT IR program SerDeRoundTrip helper method for tests.
#19815
copybara-service[bot]
closed
3 days ago
0
Add Megascale topology stat type.
#19814
copybara-service[bot]
opened
3 days ago
0
Add lax.composite primitive
#19813
copybara-service[bot]
opened
3 days ago
0
[XLA] Alias ragged all-to-all output with operand 1.
#19812
copybara-service[bot]
closed
3 days ago
0
[JAX] Add Python binding for building a colocated Python program
#19811
copybara-service[bot]
closed
3 days ago
0
Enable explicit batch dims of gather/scatter operations in GSPMD. There are two components.
#19810
copybara-service[bot]
opened
3 days ago
0
Change the default partitioning method for gather and scatter to kExplicitBatch.
#19809
copybara-service[bot]
opened
3 days ago
0
Relocates all ShardingConfig <--> ShardingConfigProto conversion from platforms/ to third_party/.
#19808
copybara-service[bot]
closed
3 days ago
0
When computing the set of instructions to shard in the presence of SPMDShardToFullShape and SPMDFullToShardShape custom calls, handle the case where parameters of a called computation may not flow to the roots of the computation.
#19807
copybara-service[bot]
closed
3 days ago
0
[XLA:SPMD] Fix a bug in `PartitionGatherTrivialSlicedOperandDimensions`.
#19806
copybara-service[bot]
closed
3 days ago
0
[hlo-opt] Add a placeholder method to register passes from the CPU/GPU providers.
#19805
copybara-service[bot]
closed
3 days ago
0
Reverts 783d6c98e36b7d7cdabeb11b34b6c3d88e716e74
#19804
copybara-service[bot]
closed
3 days ago
0
Reverts 046f3dc59a0c67a0ce144cc01a5af5aeac58977c
#19803
copybara-service[bot]
closed
3 days ago
0
Factor out test config for better readability
#19802
copybara-service[bot]
opened
4 days ago
0
[Cleanup] Use push_back instead of emplace_back where appropriate
#19801
copybara-service[bot]
closed
3 days ago
0
[Cleanup] Use push_back instead of emplace_back where appropriate
#19800
copybara-service[bot]
closed
2 days ago
0
[Cleanup] Use push_back instead of emplace_back where appropriate
#19799
copybara-service[bot]
closed
3 days ago
0
[xla:cpu] NFC: Extract XLA:CPU alignment requirements into a separate library
#19798
copybara-service[bot]
closed
4 days ago
0
Previous
Next