Re: [PATCH] packfile: fix perf regression with many packs

On Wed, Aug 12, 2026 at 05:29:55PM -0400, Jeff King wrote:
> On Wed, Aug 12, 2026 at 12:51:30PM -0700, Junio C Hamano wrote:
> 
> > "Johannes Schindelin via GitGitGadget" <[email protected]>
> > writes:
> > 
> > > In one reported use case (https://github.com/microsoft/git/issues/970),
> > > N equals 37,815 and caused a slow-down of a simple `git rev-parse
> > > --short HEAD` (which is regularly executed as part of `GIT_PS1`) from
> > > 0.4s to 4.5s. In another, heavily exercised CI scenario, clone times
> > > increased from under 2 minutes to over half an hour.
> > 
> > Face with Rolling Eyes (1f644) 🙄
> > 
> > As we grow older, more and more extreme use cases that we initially
> > thought were simply crazy become reality.
> 
> Sort of. The quadratic adding became a problem long ago, hence
> ec48540fe8 (packfile.c: speed up loading lots of packfiles, 2019-11-27).
> 
> So this was something we already dealt with that regressed. We can even
> see the regression in our perf suite:
> 
>   $ GIT_SKIP_TESTS='p5303.[1-9] p5303.1[0-9]' ./run 589127caa730^ 589127caa730 p5303-many-packs.sh
>   Test                         589127caa730^     589127caa730
>   ----------------------------------------------------------------------
>   5303.21: load 10,000 packs   0.13(0.11+0.02)   0.45(0.42+0.02) +246.2%
> 
> Unfortunately I don't think anybody pays close attention to the perf
> suite (partially because it's clunky and expensive to run, but also
> because it often requires human judgement to decide when something is a
> real change and not just a blip).

Yeah, that's a problem indeed. At GitLab we do have Bencher set up for
continuous benchmarking [1], but due to recent changes to our CI setup
those are now very flaky because seemingly, we flip-flop between two
different runners that have different specs. But we're obviously missing
a test there with lots of packfiles, so we didn't catch this regression.

Thanks!

Patrick

[1]: https://bencher.dev/perf/git/plots