[xla:gpu] Add runtime optimization using frontend attribute #63430
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
![copybara-service[bot]](https://avatars.githubusercontent.com/in/44061?s=60&v=4){kind=small}
copybara-service
bot
force-pushed
the
test_611192382
branch
from
648dc02
to
717fcbc
Compare
_xla_send_recv_validation. A collective-permute instruction inside a loop may not always send or receive data that affect the output of the whole module in all iterations. Assume this information is encoded in frontend attribute _xla_send_recv_validation attached to the Send and Recv instructions decomposed from such a collective-permute instruction, the runtime can use this information to skip the invocation of the NCCL API that performs the Send and Recv operations. Add tests. PiperOrigin-RevId: 614751869
copybara-service
bot
force-pushed
the
test_611192382
branch
from
717fcbc
to
6924b9d
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
[xla:gpu] Add runtime optimization using frontend attribute
_xla_send_recv_validation.
A collective-permute instruction inside a loop may not always send or receive
data that affect the output of the whole module in all iterations. Assume this
information is encoded in frontend attribute _xla_send_recv_validation attached
to the Send and Recv instructions decomposed from such a collective-permute
instruction, the runtime can use this information to skip the invocation of the
NCCL API that performs the Send and Recv operations.
Add tests.