Fix threaded FastBroadcast for VectorOfArray#565
Merged
ChrisRackauckas merged 1 commit intoSciML:masterfrom Apr 4, 2026
Merged
Fix threaded FastBroadcast for VectorOfArray#565ChrisRackauckas merged 1 commit intoSciML:masterfrom
ChrisRackauckas merged 1 commit intoSciML:masterfrom
Conversation
The generic threaded path in FastBroadcast splits work along the last axis via views, which does not correctly partition VectorOfArray — each thread ended up operating on the full array. Add Threaded dispatch methods that iterate over inner arrays directly using Threads.@threads for SArray VectorOfArrays, and fall back to Serial for other types. Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
RecursiveArrayTools.VectorOfArrayis broken #564:@.. thread=trueonVectorOfArray{SArray}was applying the operation once per thread (each thread iterated the full array) instead of partitioning workThreadeddispatch forAbstractVectorOfSArrayusingThreads.@threadsto correctly parallelize over inner arraysThreadedfallback for generalAbstractVectorOfArraythat delegates toSerial(the generic view-based splitting doesn't work for VectorOfArray)Root cause
FastBroadcast's generic
fast_materialize_threaded!(from the Polyester extension) splits work along the last axis usingview, then callsfast_materialize!(Serial(), view(dst, ...), view(bc, ...))per chunk. ForVectorOfArray, these views don't correctly partition the underlyingustorage, so each thread ends up broadcasting over the entire array.Test plan
[2.0, 2.0]instead of[3.0, 3.0]with 2 threadsPkg.test()passes (131→133 in interface tests)🤖 Generated with Claude Code