Skip to content

chore(release): prepare for 0.4.11 - #424

Open
github-actions[bot] wants to merge 1 commit into
masterfrom
release/v0.4.11
Open

github-actions[bot] wants to merge 1 commit into
masterfrom
release/v0.4.11

Conversation

@github-actions

Copy link
Copy Markdown
Contributor

Prepares v0.4.11, bumped by cargo release patch.

The version and the changelog are the whole diff. Merging this tags
v0.4.11 and starts the release, which builds the binaries and
plays the strength match against the previous release.

Close it without merging to call the release off, nothing has been
tagged or published yet.

@github-actions

Copy link
Copy Markdown
Contributor Author

Speed against ec68cfab

Measured on AMD EPYC 7763 64-Core Processor.
Both sides built with rustc 1.99.1 (58cf8f9c3 2026-10-13).

Both sides built and run on this runner in this job, the way
scripts/speed.sh measures a perf commit. Each round runs
both sides on a layout of its own, the same compiled code with
its code and data shuffled and moved, so the 95% interval carries
where the code landed as well as the run. The verdict holds the
interval against a 1% threshold. The default layout, the one a
release ships, is measured after as a diagnostic. The node
counts are the search's: they move when the search does, and a
speed change leaves them alone.

round     base nps  candidate nps  change
    1      4190486        4330158   +3.3%
    2      4118450        4212504   +2.3%
    3      4323913        4554237   +5.3%
    4      4583020        4568220   -0.3%
    5      4513272        4546011   +0.7%
    6      4401134        4506083   +2.4%
    7      4600966        4636804   +0.8%
    8      4358433        4338738   -0.5%
    9      4277489        4365156   +2.0%
   10      4381669        4449722   +1.6%
   11      4315322        4644037   +7.6%
   12      4485167        4583563   +2.2%
   13      4316752        4281019   -0.8%
   14      4242491        4380968   +3.3%
   15      4501688        4475084   -0.6%
   16      4104368        4402807   +7.3%
   17      4228425        4434876   +4.9%
   18      4374428        4461359   +2.0%
   19      4448496        4512962   +1.4%
   20      4379122        4513856   +3.1%
   21      4366380        4262892   -2.4%
   22      4236244        4269271   +0.8%
   23      4292903        4418751   +2.9%
   24      4466623        4644667   +4.0%
   25      4383445        4424667   +0.9%
   26      4657610        4395774   -5.6%
   27      4557625        4530991   -0.6%
   28      4612289        4635804   +0.5%
   29      4460436        4681653   +5.0%
   30      4376266        4637598   +6.0%
   31      4333084        4446784   +2.6%
   32      4514119        4493284   -0.5%
   33      4582571        4546315   -0.8%
   34      4342398        4497542   +3.6%
   35      4381716        4224390   -3.6%
   36      4340991        4255879   -2.0%
   37      4398158        4463198   +1.5%
   38      4455749        4380119   -1.7%
   39      4455018        4343565   -2.5%
   40      4414806        4281767   -3.0%

             nodes    time  median nps  faster half
base       9312660  2.13 s     4381692      4493695
candidate  9312660  2.09 s     4455540      4556663
change                           +1.7%        +1.4%

paired change +1.3%, 95% interval +0.3% to +2.2%
diagnostic, on the default layout alone -0.4%, 95% interval -1.7% to +1.0%, 13 rounds

not resolved: the interval reaches past ±1.0% without clearing it, so
more rounds are needed to say either way

Speed: +1.3% (bench nps, 95% interval +0.3% to +2.2%, 40 interleaved rounds over shuffled layouts vs ec68cfab)

The instructions each side's bench executed, counted under
cachegrind. The count repeats to within a few hundred
instructions, so a small change here is a real one, but it
prices instructions only: cache misses, mispredicted branches
and where the code lands are the speed's to show.

instructions against ec68cfab, one cachegrind run a side, bench

             instructions    nodes  per node
base       13,159,580,875  9312660    1413.1
candidate  13,159,580,863  9312660    1413.1
change             -0.00%             -0.00%

within ±0.7%, as far as edits not made for speed have moved the count

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant