Re: git-diff in a worktree is an order of magnitude slower?
- From
Junio C Hamano <gitster@pobox.com>
- Date
- Jun 21, 2026, 00:53 UTC
- Message-ID
- <xmqqa4sog1e9.fsf@gitster.g>
- In-Reply-To
- <CALnO6CAx91kbJ84d6Ef655UNG0y0rhyknBRh6Y+0o7Xn-uVytQ@mail.gmail.com>
"D. Ben Knoble" <ben.knoble@gmail.com> writes:
> But the refresh_index_quietly call is guarded by (effectively; the > actual code uses rev.diffopt.skip_stat_unmatch) > > 1 < !!diff_auto_refresh_index
It is not quite that, is it? In aecbf914 (git-diff: resurrect the traditional empty "diff --git" behaviour, 2007-08-31), it read more like
if (1 < rev.diffopt.skip_stat_unmatch) refresh_index_quietly();
where rev.diffopt.skip_stat_unmatch was initialized to 1 if diff_auto_refresh_index (boolean) is set to true.
Now, cmd_diff() dispatches to various diff backends to compare two sets (like "a tree object vs the index", "the index vs the working tree files"), each of which ends with a call to diffcore_std() and diffcore_flush() to conclude. In diffcore_std() there is a call to diffcore_skip_stat_unmatch() ONLY when skip_stat_unmatch member is set (we initialize it to 1 when auto-refrresh-index is enabled, as you saw above). The function is used to squelch the paths that remain in diff_queued_diff only because they were stat-dirty without having an actual content change, and _counts_ how many such ghost changes existed by incrementing the .skip_stat_unmatch counter.
> which dates to aecbf914c4 (git-diff: resurrect the traditional empty > "diff --git" behaviour, 2007-08-31). On my system that comparison is > false because the double-negation produces 1 > (diff_auto_refresh_index=1 or the result of git_config_bool).
Not quite. It was false because double-negation initializes the member to 1, which causes a call to diffcore_skip_stat_unmatch() be made, *and* the diffcore_skip_stat_unmatch() function did not find any ghost changes, i.e., paths that were only stat-dirty hence needed a call to refresh_index_quietly().
> So… has that conditional been quietly dead all this time? I can't > imagine that's right, but…
I initially thought it was an embarrassing thinko, but after seeing how .skip_stat_unmatch is used as a 1-based counter (i.e., if the member says 42, it means it saw 41 paths that were stat-dirty but without actual content change), I do not think so.
Now, it is a different matter if such a "dual" purpose "more than a simple boolean" counter is a good idea. Apparently it confused both of us in this case ;-).
Thanks.