Numba: combine n-ary Add/Mul terms pairwise - #2363
Draft
velochy wants to merge 1 commit into
Draft
Conversation
Log-depth dependency chains and pairwise-summation accuracy instead of one flat left-to-right chain. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Member
|
Does this still shows compile cost benefits beyond #2354 ? |
Contributor
Author
|
Measured: no compile benefit beyond #2354. Interleaved A/B/A/B at n=80 on top of #2354+#2361, cold caches: flat 354/361 s vs pairwise 361/322 s, RSS identical — pure noise. The flat chain was already linear to compile; this PR's value is only pairwise-summation float accuracy and log-depth dependency chains, per your suggestion on #2354. If that alone doesn't justify it, happy to close. |
Member
|
Let's leave it open, just no rush to merge |
ricardoV94
marked this pull request as draft
August 20, 2026 13:38
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 numba dispatch for n-ary scalar
Add/Mulemitted one flat left-to-right chain (x0 + x1 + ... + xn). Combining terms pairwise instead gives log-depth dependency chains and pairwise-summation float accuracy, at identical cost — as suggested in review of #2354.