Skip to content

Pull requests: openxla/xla

New pull request New
363 Open 31,285 Closed
363 Open 31,285 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

PR #31855: [DOC] New document - hlo_pass
#32220 opened Oct 3, 2025 by copybara-service bot Loading…
Change HloModuleGroup so it can only contain at most one module.
#32219 opened Oct 3, 2025 by copybara-service bot Loading…
@hawkinsp
[XLA:Python] Fix Python profiler hooks under Python free threading.
#32218 opened Oct 3, 2025 by copybara-service bot Loading…
@hawkinsp
[Refactor] Completely Remove AsyncStreamKind
#32217 opened Oct 3, 2025 by chaserileyroberts • Draft
PR #31525: [GPU] Optimize collectives on sub-byte types.
#32216 opened Oct 3, 2025 by copybara-service bot Loading…
[XLA:GPU] Add CompositeRewriter to GPU compiler.
#32212 opened Oct 3, 2025 by copybara-service bot Loading…
@loislo
Automated Code Change
#32208 opened Oct 3, 2025 by copybara-service bot Loading…
Automated Code Change
#32206 opened Oct 3, 2025 by copybara-service bot Loading…
[XLA:GPU] fix scaled-dot emitter for handling 3d tensors with batch dimension.
#32204 opened Oct 3, 2025 by copybara-service bot Loading…
@loislo
Automated Code Change
#32199 opened Oct 3, 2025 by copybara-service bot Loading…
Automated Code Change
#32198 opened Oct 3, 2025 by copybara-service bot Loading…
Add FlattenDebugPayloadIntoMessage to XLA error utilities.
#32197 opened Oct 3, 2025 by copybara-service bot Loading…
@dubstack
Add comments for GPUCommunicationType, extend unit test.
#32196 opened Oct 3, 2025 by copybara-service bot Loading…
@felixwqp
Change PjRtStreamExecutorBuffer to inherit from CommonPjRtBufferImpl instead of
#32195 opened Oct 3, 2025 by copybara-service bot Loading…
Remove more code paths that can create >1 module module groups.
#32190 opened Oct 3, 2025 by copybara-service bot Loading…
@hawkinsp
keep presubmits happy
#32189 opened Oct 3, 2025 by copybara-service bot Loading…
@subhankarshah
[JAX] Add require_user_context_scope config in PjRt-IFRT and enable it for JAX
#32188 opened Oct 3, 2025 by copybara-service bot Loading…
PTX version 9.0 is supported starting with CUDA 13.0.
#32187 opened Oct 2, 2025 by dimvar Loading…
[IFRT] Add layout API migration helper functions
#32186 opened Oct 2, 2025 by copybara-service bot Loading…
Integrate LLVM at llvm/llvm-project@fd9a1dcc0176
#32184 opened Oct 2, 2025 by copybara-service bot Loading…
Integrate LLVM at llvm/llvm-project@ac0e99e19124
#32183 opened Oct 2, 2025 by copybara-service bot Loading…
Update rules_ml_toolchain and set LLVM version to 21 for Linux x86_64.
#32181 opened Oct 2, 2025 by copybara-service bot Loading…
add a IsGpuDevice util function.
#32180 opened Oct 2, 2025 by copybara-service bot Loading…
SPMD: DUS as select
#32179 opened Oct 2, 2025 by copybara-service bot Loading…
@wsmoses
Set optimize_scratch_bytes to false in Fusion Autotuner Pass
#32178 opened Oct 2, 2025 by copybara-service bot Loading…
@vwbaker
ProTip! Mix and match filters to narrow down what you’re looking for.