Hi Patrick, On Wed, Jun 03, 2020 at 02:27:50PM +0200, Patrick Steinhardt wrote: > In order to test the impact on the case where we don't have any > "reference-transaction" hook installed in the repository, this commit > introduces a new performance test for git-update-refs(1). Run against an > empty repository, it produces the following results: > > Test HEAD~ HEAD > ------------------------------------------------------------------------------ > 1400.2: update existing reference 2.05(1.58+0.54) 2.08(1.58+0.57) +1.5% > 1400.3: create and destroy reference 1.79(1.38+0.49) 1.82(1.39+0.51) +1.7% > > So the overhead is around ~1.5%. Given that git-update-refs(1) is a > near-direct wrapper around reference transactions, there likely is no > other command that is impacted much worse than this. This is a serious performance regression that is worth considering. For example, a 1.5% slow-down on average in reference transactions would cause be enough for me to bisect the regression down to see what changed. What are ways that this could be avoided? I bet that this would work quite well with Emily's work on hooks, where we could check in the configuration first whether a hook is even configured. Could this be integrated with that? If not, could you cache the result of whether or not the hook exists, and/or implement some mechanism to say something like, for eg., "only run reference transaction hooks core.blah = 1" is true? I haven't myself looked seriously at your patch (although I do plan on doing so, but haven't yet had time), but I think that this should be considered thoroughly before moving forward. Thanks, Taylor
