Add fp8 min/max reduction kernels#10604
Merged
Merged
Conversation
copybara-service
Bot
force-pushed
the
test_938792741
branch
2 times, most recently
from
d76a252 to
7d8f9c9
Compare
This change refactors the `float16_wrapper` helper, it is now `sign_magnitude<T>`, which is just a wrapper to convert from/to sign-magnitude types, which floating point values are. The floating point aspect of it was just noise that can be removed. Then, update kernels: - Instead of instantiating fp16 and bf16 kernels, just do one `xf16` kernel, which works for both. - Add `xf8` kernels, which support fp8 types. PiperOrigin-RevId: 938852162
copybara-service
Bot
force-pushed
the
test_938792741
branch
from
7d8f9c9 to
6ca2888
Compare
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add fp8 min/max reduction kernels
This change refactors the
float16_wrapperhelper, it is nowsign_magnitude<T>, which is just a wrapper to convert from/to sign-magnitude types, which floating point values are. The floating point aspect of it was just noise that can be removed.Then, update kernels:
xf16kernel, which works for both.xf8kernels, which support fp8 types.