Repository navigation
The 1930th performance column, and every release beside it on one machine - #1204
Merged
Merged
Conversation
…hine The release checklist owes a measured performance column with the previous one re-measured on the same machine. This is that column for 308384b: every release from v2.1.0 to v2.4.0 and master, each measured on 2026-09-07 with nothing else on the machine, one entry per run because a single command here is capped at ten minutes. Since the 1915th, SimplifyHard is +0.18%, SolveHard +0.18% and SolveMediumHard +0.34% -- to the digit the cost recorded for #1192 when it repointed the CanonicalOrder sets at their data form -- so the thirteen pull requests after it read as allocation-neutral on these benchmarks, and the section says that is a reading of two agreeing measurements rather than a bisection. SimplifyEasy's +360 bytes is the one figure that reading does not cover, and is left unattributed. Since 2.4.0, the pair a release publishes, the Simplify and Solve family is up between 0.28% and 0.84%, the compile rows sit inside their two-percent floor, and everything else is flat. v2.4.0 measured twice a fortnight apart agrees to the byte on SimplifyHard and SolveMediumHard and to 0.0005% on SolveHard, so the file's determinism claim holds again with the same rider on the compile rows. Part of #746, and item 3 of the release checklist. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012sonx8iAspMiwRwokT1Ura
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Release checklist item 3: a measured performance column for the current master (
308384b2), withevery release from v2.1.0 to v2.4.0 re-measured on the same machine on the same day, nothing else
running. Documentation only.
What it says
SimplifyHard+0.18%,SolveHard+0.18%,SolveMediumHard+0.34% — to the digit the cost recorded for #1192;SimplifyEasy+360 bytes, unattributed; everything else flatSimplifyHard+0.84%,SolveMediumHard+0.65%,SolveHard+0.41%,SimplifyEasy+0.28%; compile rows inside their 2% floor; all inside the 3% gateSo the thirteen PRs after #1192 (#1191's codomain identity, the 29 growth declarations, the constant
fold, the graph-first collapse, bottom-up extraction) read as allocation-neutral on these benchmarks.
The section is explicit that this is two agreeing measurements and not a bisection: the commits
either side of #1192 were not re-measured, because an older commit builds from scratch and one such
run overran the ten minutes a single command gets in this session.
v2.4.0measured a fortnight apart agrees to the byte onSimplifyHardandSolveMediumHardandto 0.0005% on
SolveHard, so the file's determinism claim holds again, with its rider on thecompile rows.
Timings are in the generated table and deliberately not in the document, per its 1844th section.
Part of #746.
🤖 Generated with Claude Code
https://claude.ai/code/session_012sonx8iAspMiwRwokT1Ura