[PATCH v2 0/2] Optimization batch 6: make full use of exact renames

["Elijah Newren via GitGitGadget" <gitgitgadget@gmail.com>]

This series depends on en/merge-ort-perf and makes full use of exact
renames; see commit messages for details.

Thanks to Stolee and Junio for reviewing v1.

Changes since v1:

 * Update rename_src_nr when updating rename_src
 * Introduce want_copies in the first patch and use it in a few more places
 * Move a comment below a few exit-early if-checks.

Elijah Newren (2):
  diffcore-rename: no point trying to find a match better than exact
  diffcore-rename: filter rename_src list when possible

 diffcore-rename.c | 69 +++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 61 insertions(+), 8 deletions(-)


base-commit: 557ac0350d9efa1f59c708779ca3fb3aee121131
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-842%2Fnewren%2Fort-perf-batch-6-v2
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-842/newren/ort-perf-batch-6-v2
Pull-Request: https://github.com/gitgitgadget/git/pull/842

Range-diff vs v1:

 1:  3e69857f37e ! 1:  770e894b4ab diffcore-rename: no point trying to find a match better than exact
     @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
       	struct diff_score *mx;
       	int i, j, rename_count, skip_unmodified = 0;
       	int num_destinations, dst_cnt;
     -+	int num_sources;
     ++	int num_sources, want_copies;
       	struct progress *progress = NULL;
       
       	trace2_region_enter("diff", "setup", options->repo);
     ++	want_copies = (detect_rename == DIFF_DETECT_COPY);
     + 	if (!minimum_score)
     + 		minimum_score = DEFAULT_RENAME_SCORE;
     + 
     +@@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
     + 				p->one->rename_used++;
     + 			register_rename_src(p);
     + 		}
     +-		else if (detect_rename == DIFF_DETECT_COPY) {
     ++		else if (want_copies) {
     + 			/*
     + 			 * Increment the "rename_used" score by
     + 			 * one, to indicate ourselves as a user.
      @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
       	 * files still remain as options for rename/copies!)
       	 */
       	num_destinations = (rename_dst_nr - rename_count);
      +	num_sources = rename_src_nr;
     -+	if (detect_rename != DIFF_DETECT_COPY)
     ++	if (!want_copies)
      +		num_sources -= rename_count;
       
       	/* All done? */
     @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
       			struct diff_filespec *one = rename_src[j].p->one;
       			struct diff_score this_src;
       
     -+			if (one->rename_used &&
     -+			    detect_rename != DIFF_DETECT_COPY)
     ++			if (one->rename_used && !want_copies)
      +				continue;
      +
       			if (skip_unmodified &&
     @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
       	}
       	stop_progress(&progress);
       
     +@@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
     + 	STABLE_QSORT(mx, dst_cnt * NUM_CANDIDATE_PER_DST, score_compare);
     + 
     + 	rename_count += find_renames(mx, dst_cnt, minimum_score, 0);
     +-	if (detect_rename == DIFF_DETECT_COPY)
     ++	if (want_copies)
     + 		rename_count += find_renames(mx, dst_cnt, minimum_score, 1);
     + 	free(mx);
     + 	trace2_region_leave("diff", "inexact renames", options->repo);
 2:  580ba9a10f5 ! 2:  7ae9460d3db diffcore-rename: filter rename_src list when possible
     @@ diffcore-rename.c: static int find_renames(struct diff_score *mx, int dst_cnt, i
       	return count;
       }
       
     -+static int remove_unneeded_paths_from_src(int num_src,
     -+					  int detecting_copies)
     ++static void remove_unneeded_paths_from_src(int detecting_copies)
      +{
      +	int i, new_num_src;
      +
     ++	if (detecting_copies)
     ++		return; /* nothing to remove */
     ++	if (break_idx)
     ++		return; /* culling incompatbile with break detection */
     ++
      +	/*
      +	 * Note on reasons why we cull unneeded sources but not destinations:
      +	 *   1) Pairings are stored in rename_dst (not rename_src), which we
     @@ diffcore-rename.c: static int find_renames(struct diff_score *mx, int dst_cnt, i
      +	 *      sources N times each, so avoid that by removing the sources
      +	 *      from rename_src here.
      +	 */
     -+	if (detecting_copies)
     -+		return num_src; /* nothing to remove */
     -+	if (break_idx)
     -+		return num_src; /* culling incompatbile with break detection */
     -+
     -+	for (i = 0, new_num_src = 0; i < num_src; i++) {
     ++	for (i = 0, new_num_src = 0; i < rename_src_nr; i++) {
      +		/*
      +		 * renames are stored in rename_dst, so if a rename has
      +		 * already been detected using this source, we can just
     @@ diffcore-rename.c: static int find_renames(struct diff_score *mx, int dst_cnt, i
      +		new_num_src++;
      +	}
      +
     -+	return new_num_src;
     ++	rename_src_nr = new_num_src;
      +}
      +
       void diffcore_rename(struct diff_options *options)
       {
       	int detect_rename = options->detect_rename;
     -@@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
     - 	struct diff_score *mx;
     - 	int i, j, rename_count, skip_unmodified = 0;
     - 	int num_destinations, dst_cnt;
     --	int num_sources;
     -+	int num_sources, want_copies;
     - 	struct progress *progress = NULL;
     - 
     - 	trace2_region_enter("diff", "setup", options->repo);
     -+	want_copies = (detect_rename == DIFF_DETECT_COPY);
     - 	if (!minimum_score)
     - 		minimum_score = DEFAULT_RENAME_SCORE;
     - 
      @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
       		goto cleanup;
       
     @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
      +	 * Calculate how many renames are left
       	 */
       	num_destinations = (rename_dst_nr - rename_count);
     --	num_sources = rename_src_nr;
     --	if (detect_rename != DIFF_DETECT_COPY)
     ++	remove_unneeded_paths_from_src(want_copies);
     + 	num_sources = rename_src_nr;
     +-	if (!want_copies)
      -		num_sources -= rename_count;
     -+	num_sources = remove_unneeded_paths_from_src(rename_src_nr, want_copies);
       
       	/* All done? */
       	if (!num_destinations || !num_sources)
      @@ diffcore-rename.c: void diffcore_rename(struct diff_options *options)
     - 		for (j = 0; j < NUM_CANDIDATE_PER_DST; j++)
     - 			m[j].dst = -1;
     - 
     --		for (j = 0; j < rename_src_nr; j++) {
     -+		for (j = 0; j < num_sources; j++) {
       			struct diff_filespec *one = rename_src[j].p->one;
       			struct diff_score this_src;
       
     --			if (one->rename_used &&
     --			    detect_rename != DIFF_DETECT_COPY)
     +-			if (one->rename_used && !want_copies)
      -				continue;
     -+			assert(!one->rename_used ||
     -+			       detect_rename == DIFF_DETECT_COPY ||
     -+			       break_idx);
     ++			assert(!one->rename_used || want_copies || break_idx);
       
       			if (skip_unmodified &&
       			    diff_unmodified_pair(rename_src[j].p))

-- 
gitgitgadget