grit://cooking/bench/revs/count-since@deep-history
← cookingall benchmarksraw data

grit revs · count-since @deep-history

Bounded rev-list count with --since on deep-history vs grit rev-walk driver

repo: 50,000-commit linear synthetic history; no commit graph in measured worktree · 20,000 files · 50,000 commits · 1 refs

$ grit-bench daily --only revs/count-since@deep-history2026-10-27 · grit 8dbc58e77 · git version 2.43.0
toolmedianminmaxrunspeak memory
git1.5 ms1.2 ms1.7 ms208.1 MB
grit2.3 ms2.0 ms2.5 ms2016 MB
grit is 1.6× slowermemory: 1.9× more
every timed run (log scale)
git
grit
1.0 ms3.2 ms
How it's measured
git runs
$ git rev-list --count '--since=1970-01-01 13:53:00 +0000' -n 20 HEAD
grit runs
$ grit revs --count '--since=1970-01-01 13:53:00 +0000' -n 20 HEAD
Where the time goesone extra traced run per tool, outside the timed runs
git
— not traced
grit · median run 2.3 ms
grit.repository.open
0.2 ms
grit.config.load ×2
0.1 ms
grit.pack.index.open
0.0 ms
Every day8 days
dategrit commitgitgritgrit vs gitgrit peak
2026-10-278dbc58e771.5 ms2.3 ms1.6× slower16 MB
2026-10-26c7cf2c4432.2 ms3.6 ms1.6× slower15 MB
2026-10-252824ad3911.5 ms2.7 ms1.7× slower16 MB
2026-10-24157a05dde2.2 ms3.3 ms1.5× slower16 MB
2026-10-23bddf92f862.2 ms3.9 ms1.8× slower17 MB
2026-10-224c131b9422.2 ms3.3 ms1.5× slower15 MB
2026-10-21d4844e92a2.2 ms3.3 ms1.5× slower16 MB
2026-10-204754ad0b9—not run——