Make MKL the default when it's available #387
Merged
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
ChrisRackauckas
force-pushed
the
mklfactorization_default
branch
from
5156b96 to
321b0e1
Compare
The benchmarks have pretty conclusively shown that MKL's LU factorization is just so much better than OpenBLAS that we should effectively always use it. What this does is make MKL_jll into a dependency of LinearSolve.jl and then uses the direct calls to the binary as part of the default algorithm when it's available (it won't be available on systems where MKL does not exist, like M2 macbooks). This uses the direct calls instead of LibBLASTrampoline and thus does not effect the user's global state, thus only being a local change that simply accelerates packages using LinearSolve (i.e. all of SciML).
ChrisRackauckas
force-pushed
the
mklfactorization_default
branch
from
321b0e1 to
5c41c43
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Outdated suggestions cannot be applied.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.
The benchmarks have pretty conclusively shown that MKL's LU factorization is just so much better than OpenBLAS that we should effectively always use it. What this does is make MKL_jll into a dependency of LinearSolve.jl and then uses the direct calls to the binary as part of the default algorithm when it's available (it won't be available on systems where MKL does not exist, like M2 macbooks). This uses the direct calls instead of LibBLASTrampoline and thus does not effect the user's global state, thus only being a local change that simply accelerates packages using LinearSolve (i.e. all of SciML).