xla
xla copied to clipboard
Published
20 hours ago •
openxla
openxla
[PJRT IFRT] Pass distributed client into the PJRT IFRT layer for TPU (already done for CPU, GPU will be a separate CL).
[PJRT IFRT] Pass distributed client into the PJRT IFRT layer for TPU (already done for CPU, GPU will be a separate CL).
Objective: Let IFRT handle topology exchange and other global-cluster info, while PJRT strictly handles host-local info. This is a step towards that direction.
![copybara-service[bot] avatar](https://avatars.githubusercontent.com/in/44061?v=4 "Published by a pub.dev verified publisher"){kind=small}