-
Notifications
You must be signed in to change notification settings - Fork 75.3k
Pull requests: tensorflow/tensorflow
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[XLA:CPU] Pass LLVM data layout information to the XLA:CPU CompilationResultProto
#113087
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:GPU] Introduce CombinedGpuPerformanceModel
#113086
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:GPU] Default dot precision to BF16_BF16_F32 for F32 arguments.
#113085
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:GPU] remove nested gemm fusion code
#113084
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:CPU] Get rid of the compilation hooks.
#113083
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:GPU] Make GpuPerformanceModelCache thread-safe
#113070
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:GPU] Add Tie method to GpuClique and CollectiveCliques.
#113066
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
Validate num_reduction_axes against XNN_MAX_TENSOR_DIMS in VisitReduceNode
#113064
opened Mar 23, 2026 by
matejsmycka
Loading…
Push shardy outliner up past StablehloRoundTripShardMapExport pass.
#113059
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA] Add utilities to read backend config (for 1P/3P users)
#113052
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
Store xla::cpu::TargetMachineOptions in GpuExecutable and pass it to thunk deserializer.
#113050
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
Disable tf-reduce-identity.mlir test for open source.
#113049
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
[XLA:GPU]: Unify the host callback registry across all streams
#113044
opened Mar 23, 2026 by
copybara-service
bot
•
Draft
Previous Next
ProTip!
Adding no:label will show everything without a label.