Skip to content

Pull requests: openxla/xla

New pull request New
451 Open 33,531 Closed
451 Open 33,531 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

Allow C64 and F32 buffers to be aliased (since a C64 tensor is just two F32 tensors stapled together).
#34579 opened Nov 28, 2025 by copybara-service bot Loading…
Removal of tsl-specific string type.
#34578 opened Nov 28, 2025 by copybara-service bot Loading…
[XLA:HLO Diff] Allow including backend config in fingerprint
#34577 opened Nov 28, 2025 by copybara-service bot Loading…
@ImanHosseini
Update XNNPack and Slinky dependency versions
#34576 opened Nov 28, 2025 by copybara-service bot Loading…
Removal of tsl-specific integral types.
#34575 opened Nov 28, 2025 by copybara-service bot Loading…
[xla:ffi] Do not depend on GPU backend from default FFI implementation
#34573 opened Nov 28, 2025 by copybara-service bot Loading…
@ezhulenev
[ROCM] Added command buffers support for convolutions 2nd attempt
#34572 opened Nov 28, 2025 by pemeliya Loading…
Lower CollectiveOpsTestE2E.MemcpyP2pLargeMessage test size to speed it up
#34570 opened Nov 28, 2025 by copybara-service bot Loading…
[XLA:GPU]: Track VMM allocation in cuda_executor.
#34569 opened Nov 28, 2025 by copybara-service bot Loading…
@sohaibiftikhar
[XLA:GPU] Derive replica and partition counts from HloModule in ExecuteReplicated.
#34568 opened Nov 28, 2025 by copybara-service bot Loading…
@olegshyshkov
Extract register spill information from ptxas log
#34567 opened Nov 28, 2025 by copybara-service bot Loading…
@vwbaker
Remove default backend config fields from test HLO strings.
#34564 opened Nov 28, 2025 by copybara-service bot Loading…
@chsigg
[XLA][XTile] Update SafeIntegerArithmeticPass to also work with tensors.
#34562 opened Nov 28, 2025 by copybara-service bot Loading…
@WillFroom
[XLA:GPU] Refactor CollectiveThunk buffer creation in ThunkEmitter.
#34561 opened Nov 28, 2025 by copybara-service bot Loading…
@olegshyshkov
[XLA:GPU] Add CreateSafeIntegerArithmeticPass to lowering pass.
#34560 opened Nov 28, 2025 by copybara-service bot Loading…
@WillFroom
Add session_id support to ProfilerSessionWrapper and ExportToTensorBoard.
#34557 opened Nov 28, 2025 by copybara-service bot Loading…
@subhamsoni-google
Add programmatic profiling test with session_id support in JAX.
#34556 opened Nov 28, 2025 by copybara-service bot Loading…
@subhamsoni-google
[XLA:GPU] Emit Triton into a local module.
#34552 opened Nov 28, 2025 by copybara-service bot Loading…
@pifon2a
PR #34520: [ROCm] Use rocminfo instead of lspci as it will report all connected gpus ev…
#34550 opened Nov 28, 2025 by copybara-service bot Loading…
Use GetInPlaceInputOutputPairs from AliasInfo instead of HloDataflowAnalysis.
#34548 opened Nov 28, 2025 by copybara-service bot Loading…
@akuegel
[XLA] Add googletest wrapper that adds status asserts to gmock.h
#34547 opened Nov 28, 2025 by copybara-service bot Loading…
@dextero
Prepare XLA Bazel dependencies for enabling the layering check
#34545 opened Nov 28, 2025 by copybara-service bot Loading…
@beckerhe
[DRAFT] Enable layering_check in XLA builds.
#34543 opened Nov 28, 2025 by beckerhe • Draft
[XLA:GPU] use embed_ir_in_executable flag instead of
#34535 opened Nov 27, 2025 by copybara-service bot Loading…
@metaflow
[ROCm] Make TmaPTX tests cuda specific
#34533 opened Nov 27, 2025 by hsharsha Loading…
1
ProTip! Updated in the last three days: updated:>2025-11-25.