Skip to content

Pull requests: openxla/xla

New pull request New
341 Open 38,453 Closed
341 Open 38,453 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

Automated Code Change
#39442 opened Mar 18, 2026 by copybara-service bot Loading…
Increase unroll factor of kConcatenate to align with other Blackwell ops' vectorization.
#39441 opened Mar 18, 2026 by copybara-service bot Loading…
Automated Code Change
#39440 opened Mar 18, 2026 by copybara-service bot Loading…
Automated Code Change
#39439 opened Mar 18, 2026 by copybara-service bot Loading…
[XLA:GPU] Refactor NcclCommunicator::RegisterBufferOnce to reduce nesting.
#39438 opened Mar 18, 2026 by copybara-service bot Loading…
@olegshyshkov
[XLA:GPU] Add support for packing RegisteredBufferHandle as a kernel argument.
#39437 opened Mar 18, 2026 by copybara-service bot Loading…
@olegshyshkov
memfile_builtin: Disallow empty FileToc names.
#39436 opened Mar 18, 2026 by copybara-service bot Loading…
@pizzud
[xla:gpu] Add async thunk passes for optimizing async execution
#39435 opened Mar 18, 2026 by ezhulenev Loading…
Reverts 3bb04640fd36ccd086503235438445987998668d
#39434 opened Mar 18, 2026 by copybara-service bot Loading…
PR #39428: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks
#39432 opened Mar 18, 2026 by copybara-service bot Loading…
Disable HLO Sharding V3 in V1/V2 SPMD partitioner tests.
#39430 opened Mar 18, 2026 by copybara-service bot Loading…
Integrate StableHLO at openxla/stablehlo@78429ebf
#39429 opened Mar 18, 2026 by copybara-service bot Loading…
@abhigunj
[xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks
#39428 opened Mar 18, 2026 by ezhulenev Loading…
1
Add RTX 6000 Pro support
#39427 opened Mar 18, 2026 by copybara-service bot Loading…
Add instrumentation IDs to GPU HLOs and collect aggregated stats.
#39425 opened Mar 18, 2026 by copybara-service bot Loading…
@loislo
[XLA:GPU] Remove unnecessary if in tiled_hlo.cc.
#39424 opened Mar 18, 2026 by copybara-service bot Loading…
@pifon2a
Enable experimental Triton scaled dot product by default.
#39423 opened Mar 18, 2026 by copybara-service bot Loading…
@loislo
Enable Modulo 1 simplification in SymbolicExpr.
#39421 opened Mar 18, 2026 by copybara-service bot Loading…
[XLA:CPU] Optimize Gather with non-contiguous slice sizes
#39420 opened Mar 18, 2026 by copybara-service bot Loading…
@seantalts
[XLA] Backend config on disk inlining: schema & read support
#39419 opened Mar 18, 2026 by copybara-service bot Loading…
@thcmbs
[ROCm] Implement empty graph node support and safeguards for HIP command buffer AMD-GPU XLA on AMD GPU
#39417 opened Mar 18, 2026 by phambinhfin Loading…
Integrate LLVM at llvm/llvm-project@b03b58be38f9
#39416 opened Mar 18, 2026 by copybara-service bot Loading…
PR #39097: [GPU] Enable PDL by default.
#39415 opened Mar 18, 2026 by copybara-service bot Loading…
Add fusion debug logging
#39413 opened Mar 18, 2026 by complexbulb Loading…
Add simplification for (x % a) % b when a % b == 0.
#39412 opened Mar 18, 2026 by copybara-service bot Loading…
ProTip! What’s not been updated in a month: updated:<2026-02-18.