perf(gfql): HOLDBACK — host-thread the dense-count kernel O(E) passes - #1852
Draft
lmeyerov wants to merge 1 commit into
Draft
perf(gfql): HOLDBACK — host-thread the dense-count kernel O(E) passes#1852lmeyerov wants to merge 1 commit into
lmeyerov wants to merge 1 commit into
Conversation
Chunked 4-worker pool above 1M edges for the remaining bincount and the gather-sum (both release the GIL); partial tables are O(domain). Small inputs keep the exact single-thread lane; cupy untouched. Pool holds no query state, outside gfql_clear_caches(). Held back pending owner call on library-owned thread pools and on whether the thread-free round-4 lever (filter projection) suffices. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MF7uRZLKZaD6Q9FGWSmyXi
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Stacked on #1851 (elision core). Holds the concurrency half of round 3 pending owner judgment: a small process-lifetime thread pool chunks the remaining bincount + gather-sum above 1M edges (both release the GIL). Open questions the owner flagged: library-owned pools vs host pools (oversubscription, fork safety), and whether the thread-free round-4 lever (filter projection pushdown) closes the gap without this. Do not merge unless both answers favor it.
Pins: threaded-vs-plain equality on uneven chunk boundaries; kernel-level threaded/unthreaded/oracle triple.
🤖 Generated with Claude Code
https://claude.ai/code/session_01MF7uRZLKZaD6Q9FGWSmyXi