issues
search
openxla
/
xla
A machine learning compiler for GPUs, CPUs, and ML accelerators
Apache License 2.0
2.64k
stars
418
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Move GpuDriver::GetModuleXxx functions into the appropriate Executor class.
#17994
copybara-service[bot]
closed
6 days ago
0
Move GpuDriver Load functions into appropriate Executor classes.
#17993
copybara-service[bot]
closed
6 days ago
0
[HLO Componentization] Remove ununsed hlo/parser dependencies
#17992
copybara-service[bot]
closed
6 days ago
0
Skip `test_ragged_copy_on_host` if `xla_extension_version` < 290
#17991
copybara-service[bot]
closed
6 days ago
0
Remove unused stream_executor/platform/port.h.
#17990
copybara-service[bot]
closed
6 days ago
0
[easy] [XLA] s/DoProducerConsumerMultiOutputFusion/DoBackendSpecificMultiOutputFusion
#17989
copybara-service[bot]
opened
1 week ago
0
Properly plumb rocm_compiler through configure.py
#17988
copybara-service[bot]
closed
1 week ago
0
Reverts f6370d12d4b97e29b409a4f6c77200f471983f2c
#17987
copybara-service[bot]
opened
1 week ago
0
Remove some unnecessary stream_executor/platform dependencies.
#17986
copybara-service[bot]
closed
1 week ago
0
[TEST] Intentionally break dependency violation test
#17985
copybara-service[bot]
opened
1 week ago
0
Better error diagnostics for the dependency violation check
#17984
copybara-service[bot]
closed
1 week ago
0
Eliminate some bad strategy combinations for gather operands/outputs from the search space.
#17983
copybara-service[bot]
closed
6 days ago
0
[DRAFT] Stream Annotation Prototype
#17982
chaserileyroberts
opened
1 week ago
0
[xla:ffi] Add support for DenseElementAttr attributes.
#17981
copybara-service[bot]
closed
6 days ago
0
[ROCm] Fix pjrt_c_api_gpu_test for ROCm
#17980
hsharsha
opened
1 week ago
0
[XLA:CPU] Add efficient 1D sort thunk implementation
#17979
copybara-service[bot]
closed
6 hours ago
1
Split GpuTimer into CUDA and ROCm specific implementations
#17978
copybara-service[bot]
closed
5 days ago
0
[XLA:GPU] Compute computation layout for module
#17977
copybara-service[bot]
closed
6 days ago
0
Add PTX CustomKernel.
#17976
copybara-service[bot]
closed
6 days ago
0
[HLO Componentization] Create hlo/parser sub-component (Phase II).
#17975
copybara-service[bot]
closed
1 week ago
0
[HLO Componentization] Create hlo/parser sub-component (Phase II).
#17974
copybara-service[bot]
closed
1 week ago
0
[HLO Componentization] Create hlo/parser sub-component (Phase II).
#17973
copybara-service[bot]
closed
6 days ago
0
[HLO Componentization] Create hlo/parser sub-component (Phase II).
#17972
copybara-service[bot]
closed
1 week ago
0
Avoid compile error on MacOS.
#17971
copybara-service[bot]
closed
1 week ago
0
#sdy Add support for inlined meshes in sdy round trip.
#17970
copybara-service[bot]
closed
5 days ago
0
PR #17814: [ROCM] buffer_comparator init bugfix
#17969
copybara-service[bot]
closed
6 days ago
0
PR #17789: [nfc] Remove loop iter from dynamic slice thunk
#17968
copybara-service[bot]
closed
1 week ago
0
[XLA:GPU] Fix triton build stubs and add a test
#17967
copybara-service[bot]
closed
1 week ago
0
PR #15577: [PJRT:GPU] Add setting for mocked number of hosts per slice
#17966
copybara-service[bot]
closed
1 week ago
0
[XLA:GPU] Check that a small constant is supported by Triton emitter before fusion.
#17965
copybara-service[bot]
closed
1 week ago
0
Integrate LLVM at llvm/llvm-project@82f5acfbec65
#17963
copybara-service[bot]
closed
1 week ago
0
PR #17900: [ROCm] Fix build break in executor and kernel test introduced in f896afd
#17962
copybara-service[bot]
closed
1 week ago
0
Optimized lowering for `8xi4 -> 8xbf16` conversion in TritonGPUToLLVM.
#17961
copybara-service[bot]
closed
1 week ago
0
PR #17814: [ROCM] buffer_comparator init bugfix
#17960
copybara-service[bot]
opened
1 week ago
0
[XLA:GPU] Remove the use of GpuTimer::ReturnRandomDurationsForTesting() in determinism_test.cc.
#17959
copybara-service[bot]
closed
1 week ago
0
transpose injection by copy
#17958
wenscarl
closed
1 week ago
0
Automated Code Change
#17957
copybara-service[bot]
opened
1 week ago
0
[host_callback] Remove the outfeed received machinery
#17956
copybara-service[bot]
closed
1 week ago
0
Handle gather/scatter batching dims in Get(Gather/Scatter)SizeInChunkRatio for cost analysis.
#17955
copybara-service[bot]
closed
1 week ago
0
[XLA] Refactor LayoutMode / MemorySpaceColor handling to avoid code duplication.
#17954
copybara-service[bot]
opened
1 week ago
0
[XLA:GPU] Support dot_bf16_bf16_f32 algorithm with cuBLAS by adding convert before the dot call.
#17953
copybara-service[bot]
closed
6 days ago
0
Rollback of PR #76831 because it was related to a build error due to openmp symbol missing.
#17952
copybara-service[bot]
closed
1 week ago
0
[XLA:GPU] Support 0D Tensors in the Generic Triton Emitter.
#17951
copybara-service[bot]
closed
1 week ago
0
Remove Post Layout Assignment Collective Pipeliner
#17950
philipphack
closed
3 days ago
0
Remove stream_executor/platform/platform.h in favor of simply using tsl/platform/platform.h.
#17949
copybara-service[bot]
closed
1 week ago
0
Skip nextafter test with IFRT for unsupported dtype.
#17948
copybara-service[bot]
opened
1 week ago
0
Reverts a2f9ed71b6a5f4bc1109682c12a9286c83d1b5f9
#17947
copybara-service[bot]
closed
6 days ago
0
[XLA:GPU] Split `triton_fusion_emitter` into three files.
#17946
copybara-service[bot]
closed
1 week ago
0
PR #17295: [ROCm] clang support
#17945
copybara-service[bot]
closed
1 week ago
0
Reverts f8d9def577b86a8f7201b6eeaa2850847ba96b1e
#17944
copybara-service[bot]
closed
1 week ago
0
Previous
Next