c62b4549 · [XLA:Runtime] Moved the nccl_clique_key target to runtime folder. · 4月 05, 2024

f7740404 · Make RemoveSendDone and RemoveRecvDone virtual and return the new custom calls · 4月 05, 2024

9467a53a · Reverts 276dc805 · 4月 05, 2024

373a11dd · Prepare move of CUDA compilation functionality for jaxlib · 4月 05, 2024

9db78cb6 · To ensure that C++20 is fully supported in TensorFlow, · 4月 05, 2024

865fb957 · [XLA:GPU] Compute a graph of TiledHloInstruction for given tile parameters. · 4月 05, 2024

893ca433 · Adding support for transpose to LayoutCanonicalizer pass. · 4月 05, 2024

04e404a8 · #shlo_ref Fix I4 tests. · 4月 05, 2024

82370641 · Add F16 wrapper type · 4月 05, 2024

e98a555d · Move CUDA specific functions from asm_compiler to cuda_asm_compiler target · 4月 05, 2024

36e7e729 · Testing the new tensorflow workflow · 4月 05, 2024

ab4f3131 · Move LoadKernelOrGetPtr into :redzone_allocator · 4月 05, 2024

eb5dfeb7 · Add an XNNPACK delegate for quantized `BatchMatMul`. · 4月 05, 2024

cf50bea0 · patch https://github.com/openxla/xla/pull/11225 to fix indexing maps · 4月 05, 2024

3383efcc · Integrate LLVM at llvm/llvm-project@8487e05967aa · 4月 05, 2024

d26646ef · [XLA] Add argument and return shardings to while body and cond based on result... · 4月 05, 2024

842f819c · Integrate LLVM at llvm/llvm-project@5ed60ffd790a · 4月 05, 2024

fbd0a74d · Set while computation back pointer more consistently. · 4月 05, 2024

34d7cde6 · Unpack per-channel hybrid quantized MHLO ops to float ops · 4月 05, 2024