diff-index: don't queue unchanged filepairs with diff_change()

diff_cache() queues unchanged filepairs if the flag find_copies_harder is set, and uses diff_change() for that. This function allocates a filespec for each side, does a few other things that are unnecessary for unchanged filepairs and always sets the diff_flag has_changes, which is simply misleading in this case. Add a new streamlined function for queuing unchanged filepairs and use it in show_modified(), which is called by diff_cache() via oneway_diff() and do_oneway_diff(). It allocates only a single filespec for each filepair and uses it twice with reference counting. This has a measurable effect if there are a lot of them, like in the Linux repo: Benchmark 1: ./git_v2.52.0 -C ../linux diff --cached --find-copies-harder Time (mean ± σ): 31.8 ms ± 0.2 ms [User: 24.2 ms, System: 6.3 ms] Range (min … max): 31.5 ms … 32.3 ms 85 runs Benchmark 2: ./git -C ../linux diff --cached --find-copies-harder Time (mean ± σ): 23.9 ms ± 0.2 ms [User: 18.1 ms, System: 4.6 ms] Range (min … max): 23.5 ms … 24.4 ms 111 runs Summary ./git -C ../linux diff --cached --find-copies-harder ran 1.33 ± 0.01 times faster than ./git_v2.52.0 -C ../linux diff --cached --find-copies-harder Signed-off-by: René Scharfe <l.s.r@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>

René Scharfe committed Nov 30, 2025 at 12:47 UTC 38f88051dae6ddb2f1cdb9c7415d4ba6caef04af
3 files changed +31 -7
diff-lib.c
+6 -7
@@ -418,13 +418,12 @@ static int show_modified(struct rev_info *revs,
418 }
419
420 oldmode = old_entry->ce_mode;
421 - if (mode == oldmode && oideq(oid, &old_entry->oid) && !dirty_submodule &&
422 - !revs->diffopt.flags.find_copies_harder)
423 - return 0;
424 -
425 - diff_change(&revs->diffopt, oldmode, mode,
426 - &old_entry->oid, oid, 1, !is_null_oid(oid),
427 - old_entry->name, 0, dirty_submodule);
421 + if (mode != oldmode || !oideq(oid, &old_entry->oid) || dirty_submodule)
422 + diff_change(&revs->diffopt, oldmode, mode,
423 + &old_entry->oid, oid, 1, !is_null_oid(oid),
424 + old_entry->name, 0, dirty_submodule);
425 + else if (revs->diffopt.flags.find_copies_harder)
426 + diff_same(&revs->diffopt, mode, oid, old_entry->name);
427 return 0;
428 }
429
diff.c
+20
@@ -7347,6 +7347,26 @@ void diff_change(struct diff_options *options,
7347 concatpath, old_dirty_submodule, new_dirty_submodule);
7348 }
7349
7350 +void diff_same(struct diff_options *options,
7351 + unsigned mode,
7352 + const struct object_id *oid,
7353 + const char *concatpath)
7354 +{
7355 + struct diff_filespec *one;
7356 +
7357 + if (S_ISGITLINK(mode) && is_submodule_ignored(concatpath, options))
7358 + return;
7359 +
7360 + if (options->prefix &&
7361 + strncmp(concatpath, options->prefix, options->prefix_length))
7362 + return;
7363 +
7364 + one = alloc_filespec(concatpath);
7365 + fill_filespec(one, oid, 1, mode);
7366 + one->count++;
7367 + diff_queue(&diff_queued_diff, one, one);
7368 +}
7369 +
7370 struct diff_filepair *diff_unmerge(struct diff_options *options, const char *path)
7371 {
7372 struct diff_filepair *pair;
diff.h
+5
@@ -572,6 +572,11 @@ void diff_change(struct diff_options *,
572 const char *fullpath,
573 unsigned dirty_submodule1, unsigned dirty_submodule2);
574
575 +void diff_same(struct diff_options *,
576 + unsigned mode,
577 + const struct object_id *oid,
578 + const char *fullpath);
579 +
580 struct diff_filepair *diff_unmerge(struct diff_options *, const char *path);
581
582 void compute_diffstat(struct diff_options *options, struct diffstat_t *diffstat,