Skip to content

Pull requests: openxla/xla

New pull request New
717 Open 44,749 Closed
717 Open 44,749 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

[XLA:GPU]: Add support for Record API to FFI
#46218 opened Jul 25, 2026 by copybara-service Bot Loading…
@sohaibiftikhar
Add support for uint64_t in RAFT select-k execution.
#46217 opened Jul 25, 2026 by copybara-service Bot Loading…
@apivovarov
Introduce PjRtStreamExecutorRawClient to allow decoupling
#46216 opened Jul 25, 2026 by copybara-service Bot Loading…
@pschuh
[IFRT IR] Add IfrtString element type, which represents an IFRT string dtype
#46215 opened Jul 25, 2026 by copybara-service Bot Loading…
Automated Code Change
#46210 opened Jul 25, 2026 by copybara-service Bot Loading…
Automated Code Change
#46208 opened Jul 25, 2026 by copybara-service Bot Loading…
Automated Code Change
#46207 opened Jul 25, 2026 by copybara-service Bot Loading…
[XLA:MSA] Re-write block prefetching to follow a buffer based approach. Also, supports async fusions in block prefetching.
#46204 opened Jul 24, 2026 by copybara-service Bot Loading…
@sparc1998
[XLA:CPU] Use xla.apply_indexing for tile ID calculation in CPU tiled lowering.
#46203 opened Jul 24, 2026 by copybara-service Bot Loading…
@pifon2a
Fix label paths in tensorflow/compiler/xla/xla.default.bzl
#46202 opened Jul 24, 2026 by copybara-service Bot Loading…
diffbase change is not going to be pushed
#46200 opened Jul 24, 2026 by copybara-service Bot Loading…
@pizzud
functional_hlo_runner: Support stablehlo input.
#46199 opened Jul 24, 2026 by copybara-service Bot Loading…
@pizzud
[XLA:GPU][oneAPI] Skip PTX kernel tests on oneAPI backend
#46198 opened Jul 24, 2026 by bhavani-subramanian Contributor Loading…
Annotate the op tracker graph metadata to include the name of the MLIR op that triggered the creation of each op tracker.
#46196 opened Jul 24, 2026 by copybara-service Bot Loading…
Handle V3 shardings in infer_dispatch_info.
#46195 opened Jul 24, 2026 by copybara-service Bot Loading…
Adding support for multi-dim reduce. This is implemented via nested scf.for loops followed by a chain of reduce operations for each reduction axis.
#46194 opened Jul 24, 2026 by copybara-service Bot Loading…
@Moerafaat
Implement IsTpuCancellationPlaceholderError through PJRT C API extension.
#46193 opened Jul 24, 2026 by copybara-service Bot Loading…
Support dump/load of autotune results in new cache format
#46189 opened Jul 24, 2026 by copybara-service Bot Loading…
@shyamliagrawal
[xla:gpu] Refactor ThunkEmitter and collective (group) lowering
#46188 opened Jul 24, 2026 by ezhulenev Member Loading…
5
[XLA:GPU] Reuse existing channel IDs in AllReduceSplitter.
#46186 opened Jul 24, 2026 by copybara-service Bot Loading…
@mkuperst
Don't dump HLO module when loading the CPU executable.
#46185 opened Jul 24, 2026 by copybara-service Bot Loading…
Convert to V2 before deciding dims for group sharding in ReplicatePartial.
#46184 opened Jul 24, 2026 by copybara-service Bot Loading…
[XLA][TPU] Save original HLO sharding in recovery computation
#46183 opened Jul 24, 2026 by copybara-service Bot Loading…
@jcai19
[ROCm: XLA profiler] Use real HIP stream_id instead of HSA queue handle
#46182 opened Jul 24, 2026 by cj401-amd Contributor Loading…
4 tasks done
[XLA:GPU][oneAPI] Integrate oneDNN matmul for SYCL backend
#46181 opened Jul 24, 2026 by kanvi-nervana Contributor Loading…
1
ProTip! Follow long discussions with comments:>50.