Only pass one wrapper object in SYCL reductions #6047
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
masterleinad
force-pushed
the
simplify_sycl_reductions
branch
from
f819811
to
2e51c67
Compare
Rombur
approved these changes
May 15, 2023
dalg24
approved these changes
May 19, 2023
Closed
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Outdated suggestions cannot be applied.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.
Passing a combined functor reducer object to
ParallelReduceallows us to also pass the combined object (wrapped) into the kernel. ForSYCL, we were still passing functor and reducer separately which could cause worse performance in case there was no explicit reducer (and therefore we effectively only use the functor). Also, passing two objects viaIndirectKernelMemcould cause problems when deciding if we can use the device-copyable code path since the limits there are only really designed for one object.I didn't notice any significant differences in performance with respect to #6035.