use threadpoolctl for detecting BLAS threadpools #692
Merged
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Outdated suggestions cannot be applied.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.
We were using numpy.config.get_info to figure out what BLAS library was being used, in an effort to warn people if they hadn't disabled the internal threadpool for the BLAS they were using. This method is removed in numpy 1.26, causing errors.
Fix by using the threadpoolctl library to detect both the BLAS library and the number of threads its configured for.
While we could automatically configure the BLAS library to reduce the threadpool size to 1, this is process wide - and would have side effects for our users. Instead just warn here, and give instructions for people on how to configure themselves.