issues
search
openxla
/
xla
A machine learning compiler for GPUs, CPUs, and ML accelerators
Apache License 2.0
2.71k
stars
436
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Integrate LLVM at llvm/llvm-project@556ea5265a25
#19701
copybara-service[bot]
opened
26 minutes ago
0
[ROCm] failed to legalize operation 'math.exp' for exponential op with bf16 dtype
#19700
hugomano
opened
47 minutes ago
0
Explicit stream annotation: Set ExecutionStreamId based on frontend attribute
#19699
chaserileyroberts
opened
48 minutes ago
0
[XLA:CPU] Update RunHloBenchmark to enable running with HLO with inferred arguments
#19698
copybara-service[bot]
opened
1 hour ago
0
[XLA] Go back to using a glob for including dialects in the `mlir_interpreter`.
#19697
copybara-service[bot]
closed
1 hour ago
0
Upgrade Abseil to latest LTS branch (lts_2024_07_22).
#19696
copybara-service[bot]
opened
2 hours ago
0
[XLA:GPU] Move dot_algorithm_rewriter from xla/servive/gpu/transforms to xla/hlo/transforms
#19695
copybara-service[bot]
opened
2 hours ago
0
Remove custom logging implementation from TSL
#19694
copybara-service[bot]
opened
3 hours ago
0
[XLA:GPU] Fix a bug in dot_algorithm_rewriter.
#19693
copybara-service[bot]
closed
2 hours ago
0
Integrate LLVM at llvm/llvm-project@a12e79a85fc1
#19692
copybara-service[bot]
closed
1 hour ago
0
[ROCm] Use -fno-canonical-system-headers for gcc
#19691
alekstheod
opened
5 hours ago
0
Re-enable deterministic scatter expander pass by default.
#19690
copybara-service[bot]
closed
5 hours ago
0
Add more flexible custom hermetic Python setup
#19689
copybara-service[bot]
opened
7 hours ago
0
Fix two issues in `PartitionScatterIndexPassthroughDimensions`.
#19688
copybara-service[bot]
opened
7 hours ago
0
Integrate Triton up to [9732c047](https://github.com/openai/triton/commits/9732c04701bd856daca89bde38bafa4636ca56a8)
#19687
copybara-service[bot]
opened
8 hours ago
0
PR #19679: [XLA:CPU][oneDNN] Relocate Addend Shape Validation to the Contraction Rewriter
#19686
copybara-service[bot]
closed
7 hours ago
1
Experiment with removing hermetic_cuda_data_dir argument.
#19685
copybara-service[bot]
closed
7 hours ago
1
Add cuda::CompilationProvider interface and first implementation for subprocess compilation
#19684
copybara-service[bot]
closed
4 hours ago
0
PR #19656: Fix implicit index handling in ScatterDeterminismExpander
#19683
copybara-service[bot]
closed
7 hours ago
0
Change parameter type in LinkUsingNvlink
#19682
copybara-service[bot]
closed
7 hours ago
0
[xla] Add S4/U4 support to reshape
#19681
copybara-service[bot]
opened
11 hours ago
0
[xla:collectives] Initial xla/core/collectives component commit
#19680
copybara-service[bot]
opened
11 hours ago
0
[XLA:CPU][oneDNN] Relocate Addend Shape Validation to the Contraction Rewriter
#19679
akhilgoe
closed
7 hours ago
0
Remove obsolete PjRtClient::AsyncSendPlaceholder API.
#19678
copybara-service[bot]
closed
10 hours ago
0
Set implicitTrunc on APInt creation
#19677
copybara-service[bot]
closed
11 hours ago
0
[XLA:GPU] Support cross-replica cps in collective-permute decomposer
#19676
copybara-service[bot]
opened
16 hours ago
0
Integrate StableHLO at openxla/stablehlo@f21104d0
#19675
copybara-service[bot]
opened
16 hours ago
0
Move `tsl/platform/profile_utils` to `xla/tsl/platform/profile_utils`
#19674
copybara-service[bot]
closed
11 hours ago
0
Integrate LLVM at llvm/llvm-project@a12e79a85fc1
#19673
copybara-service[bot]
opened
17 hours ago
0
[IFRT] Implement BytecodeDialectInterface for VIFRT.
#19672
copybara-service[bot]
closed
14 hours ago
0
[XLA:GPU] Remove BuildInitializerThunk and thunk_util.
#19671
copybara-service[bot]
opened
18 hours ago
0
[xla:cpu] Add a benchmark for creating zero-copy PjRt buffer
#19670
copybara-service[bot]
closed
3 hours ago
0
Replace custom free-threading flag by rules_python is_py_freethreaded in Nanobind
#19669
vfdev-5
opened
20 hours ago
0
[xla:codegen] Add a testonly KernelEmitter for testing XLA:CPU kernels
#19668
copybara-service[bot]
closed
19 hours ago
0
Stop using AsGpuStreamValue in gpu_cudamallocasync_allocator_test.
#19667
copybara-service[bot]
closed
16 hours ago
0
Eliminate static_casts in GpuCommandBuffer.
#19666
copybara-service[bot]
closed
17 hours ago
0
Add a simple test for the symbol_finder
#19665
copybara-service[bot]
closed
8 hours ago
0
[XLA:GPU] Dump the failing HLO fusion to a file when Triton numerics verification fails.
#19664
copybara-service[bot]
closed
20 hours ago
0
Legalize more dialects in shardy
#19663
copybara-service[bot]
opened
23 hours ago
0
Revert: [XLA:GPU] Enable Triton normalization fusions by default.
#19662
copybara-service[bot]
closed
19 hours ago
0
[tsl:concurrency] Fix asan error in CountDownAsyncValueRef
#19661
copybara-service[bot]
closed
23 hours ago
0
[ROCm] switch rocm build to clang
#19660
alekstheod
opened
1 day ago
0
[xla:cpu] NFC: Remove ExecuteState alias from Thunk
#19659
copybara-service[bot]
closed
4 hours ago
0
#sdy Refactor `xla-sdy-mhlo-round-trip-shard-map-export` from a `ConversionPattern` to a walk.
#19658
copybara-service[bot]
closed
3 hours ago
0
[XLA:GPU] Consolidate sort optimizations in a dedicated compiler pass.
#19657
copybara-service[bot]
opened
1 day ago
0
Fix implicit index handling in ScatterDeterminismExpander
#19656
sergey-kozub
closed
7 hours ago
0
[ROCm] Make MLIR Math dialect lowering more deterministic
#19655
draganmladjenovic
opened
1 day ago
0
PR #19463: [XLA:GPU] Add an option to disable GPU multi thread sharing
#19654
copybara-service[bot]
opened
1 day ago
0
Reverts 93f9dda11dff8eb32aa0e287ed1350ba334ddd6d
#19653
copybara-service[bot]
opened
1 day ago
0
[XLA:GPU] Copy final bufferize patterns that were removed in upstream MLIR.
#19652
copybara-service[bot]
closed
1 day ago
0
Next