bitsandbytes, test existence of triton.ops.matmul_perf_model #1493
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
This fixes the bitsandbytes to work with newer triton versions which does not anymore include the triton.ops.matmul_perf_model which has been moved to triton-langs kernels project. https://github.com/triton-lang/kernels Fix is to simply catch the import error if triton is available but the triton.ops.matmul_perf_model is not. In such case the logic is implemented in a same way than earlier when the triton was not available at all. To my understanding there is currently no a way to install opt_matmul_perf.py from new tritons kernel project directly with pip, so in future it could propably be good to include that python class directly in bitsandbytes. bitsandbytes-foundation#1492 Signed-off-by: Mika Laitio <[email protected]>
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Outdated suggestions cannot be applied.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.
This fixes the bitsandbytes to work with newer triton versions which does not anymore include the triton.ops.matmul_perf_model which has been moved to triton-langs kernels project. https://github.com/triton-lang/kernels
Fix is to simply catch the import error if triton is available but the triton.ops.matmul_perf_model is not. In such case the logic is implemented in a same way than earlier when the triton was not available at all.
To my understanding there is currently no a way to install opt_matmul_perf.py from new tritons kernel project directly with pip, so in future it could propably be good to include that python class directly in bitsandbytes.
#1492