[pallas:triton] The lowering now uses PTX instead of Triton IR #25903
+176
−28
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_715379979
branch
from
7962127 to
50ef04f
Compare
This change improves the stability and backward compatibility of Pallas Triton calls, because unlike PTX, the Triton dialect has no stability guarantees and does change in practice. A few notes * The implementation only supports CUDA at the moment. More work is needed to support ROCm. * Pallas Triton no longer delegates compilation to PTX to XLA:GPU. Instead, compilation is done via a new PjRt extension, which uses its own compilation pipeline mirrored after the one in the Triton Python bindings. * The implementation of the old custom call used by Pallas Triton is deprecated and will be removed after 6 months as per [compatibility guarantees] [*] [*]: https://jax.readthedocs.io/en/latest/export/export.html#compatibility-guarantees PiperOrigin-RevId: 715379979
copybara-service
bot
force-pushed
the
test_715379979
branch
from
50ef04f to
794963d
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Outdated suggestions cannot be applied.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.
[pallas:triton] The lowering now uses PTX instead of Triton IR
This change improves the stability and backward compatibility of Pallas Triton
calls, because unlike PTX, the Triton dialect has no stability guarantees
and does change in practice.
A few notes
to support ROCm.
compilation is done via a new PjRt extension, which uses its own compilation
pipeline mirrored after the one in the Triton Python bindings.
deprecated and will be removed after 6 months as per
[compatibility guarantees] *