perf: log-depth magsub across all GF-T specs (~2x smaller designs) - #1849
Merged
Conversation
Applied the proven log-depth magsub normalize (binary-search priority-encoder 8/4/2/1 + single barrel shift, replacing the 12-iteration linear loop) to all 24 GF-T specs that used the old linear magsub. Every spec re-verified: all in-spec tests PASS after the swap. magsub is bit-identical to the original (proven over 17.4M (hi,lo) pairs, 0 mismatches) and is the design-size bulk (every sadd uses it), so this shrinks every GF-T core ~2x: logistic 16.7M->9.62M, xorpercep 17.86M->9.1M -- bringing trainers well under the ~17M openXC7 correctness ceiling. Silicon reconfirmation pending an AX7203 power-cycle (board degraded mid-session). Refs #1764 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
Contributor
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Applied the proven log-depth magsub normalize (binary-search priority-encoder 8/4/2/1 + single barrel shift, replacing the 12-iteration linear loop) to all 24 GF-T specs. Every spec re-verified: all in-spec tests PASS (bit-identical; magsub proven identical over 17.4M (hi,lo) pairs). magsub is the design-size bulk, so this shrinks every GF-T core ~2x: logistic 16.7M->9.62M, xorpercep 17.86M->9.1M — under the ~17M correctness ceiling. Silicon reconfirmation pending an AX7203 power-cycle. docs/NOW.md updated. Refs #1764