xla
xla copied to clipboard
openxla
A machine learning compiler for GPUs, CPUs, and ML accelerators
[PJRT C API] Add some tests for PJRT C API implementation. - pjrt_c_api_test contains a basic set of tests. - A plugin can run the tests in pjrt_c_api_test via RegisterTestCApiFactory....
![copybara-service[bot] avatar](https://avatars.githubusercontent.com/in/44061?v=4 "copybara-service[bot] avatar"){kind=avatar}
Add CollectiveUpdateSliceOp It turns out there's a second form of CollectivePermute that involves sending slices and receiving them into slices, rather than sending and receiving tensors. In this CL, we're...
Generalize host callback support in JAX and IFRT This change introduces a general host callback support in IFRT and changes JAX to use this interface. * General host callback in...
Replace `tensorflow::Status::SetStackTrace` with `SetStackTrace(status, trace)`, to be compatible with the `absl::Status` API.
[xla:gpu] Enable CUDA Graphs by default
[SE] Fix compiler and correctness error in typed DeviceMemoryAllocator::Allocate. Currently, stream_executor::DeviceMemoryAllocator::Allocate will not compile for types that are not , and even if it did compile it would return ScopedDeviceMemory...
[NFC] Add opcode_string() to HloInstruction
set the flag --experimental_link_static_libraries=true.