Add optional all-dims-small (AND) rule for SUP threshold#944
Draft
jeffhammond wants to merge 1 commit into
Draft
Add optional all-dims-small (AND) rule for SUP threshold#944jeffhammond wants to merge 1 commit into
jeffhammond wants to merge 1 commit into
Conversation
BLIS's default SUP threshold rule routes to the skinny/unpacked path when ANY dimension is below its threshold (m<MT || n<NT || k<KT). That sends rank-k updates and tall/wide panels -- skinny in one dimension but large in another -- to SUP, where they run far slower than the packed native path (e.g. 4000x4000x64: SUP 474 vs native 683 GFLOP/s on GB10). Add a config-gated ALL-dims variant (BLIS_SUP_THRESH_ALL) of bli_cntx_l3_sup_thresh_is_met: take SUP only when m, n AND k are all below threshold, i.e. genuinely small matrices. A lone small dimension no longer forces SUP; small squares (which do want SUP) are unaffected. Opt-in; default OR behavior preserved when the macro is undefined. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
jeffhammond
marked this pull request as draft
July 16, 2026 15:24
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
BLIS's default SUP threshold rule routes to the skinny/unpacked path when ANY dimension is below its threshold (m<MT || n<NT || k<KT). That sends rank-k updates and tall/wide panels -- skinny in one dimension but large in another -- to SUP, where they run far slower than the packed native path (e.g. 4000x4000x64: SUP 474 vs native 683 GFLOP/s on GB10).
Add a config-gated ALL-dims variant (BLIS_SUP_THRESH_ALL) of bli_cntx_l3_sup_thresh_is_met: take SUP only when m, n AND k are all below threshold, i.e. genuinely small matrices. A lone small dimension no longer forces SUP; small squares (which do want SUP) are unaffected. Opt-in; default OR behavior preserved when the macro is undefined.