[PATCH] revision: trace topo-walk statistics

["Derrick Stolee via GitGitGadget" <gitgitgadget@xxxxxxxxx>]

From: Derrick Stolee <dstolee@xxxxxxxxxxxxx>

We trace statistics about the effectiveness of changed-path Bloom
filters since 42e50e78 (revision.c: add trace2 stats around Bloom
filter usage, 2020-04-06). Add similar tracing for the topo-walk
algorithm that uses generation numbers to limit the walk size.

This information can help investigate and describe benefits to
heuristics and other changes.

The information that is printed is in JSON format and can be formatted
nicely to present as follows:

    {
	"count_explort_walked":2603,
	"count_indegree_walked":2603,
	"count_topo_walked":473
    }

Each of these values count the number of commits are visited by each of
the three "stages" of the topo-walk as detailed in b4542418 (revision.c:
generation-based topo-order algorithm, 2018-11-01).

Signed-off-by: Derrick Stolee <dstolee@xxxxxxxxxxxxx>
---
    revision: trace topo-walk statistics
    
    Here is a patch I thought useful when investigating the performance
    implications of Abhishek's series [1]
    
    [1]
    https://lore.kernel.org/git/pull.676.v5.git.1609154168.gitgitgadget@xxxxxxxxx/

Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-828%2Fderrickstolee%2Ftrace2-topo-walk-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-828/derrickstolee/trace2-topo-walk-v1
Pull-Request: https://github.com/gitgitgadget/git/pull/828

 revision.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/revision.c b/revision.c
index 9dff845bed6..1bb590ece78 100644
--- a/revision.c
+++ b/revision.c
@@ -3308,6 +3308,26 @@ struct topo_walk_info {
 	struct author_date_slab author_date;
 };
 
+static int topo_walk_atexit_registered;
+static unsigned int count_explore_walked;
+static unsigned int count_indegree_walked;
+static unsigned int count_topo_walked;
+
+static void trace2_topo_walk_statistics_atexit(void)
+{
+	struct json_writer jw = JSON_WRITER_INIT;
+
+	jw_object_begin(&jw, 0);
+	jw_object_intmax(&jw, "count_explore_walked", count_explore_walked);
+	jw_object_intmax(&jw, "count_indegree_walked", count_indegree_walked);
+	jw_object_intmax(&jw, "count_topo_walked", count_topo_walked);
+	jw_end(&jw);
+
+	trace2_data_json("topo_walk", the_repository, "statistics", &jw);
+
+	jw_release(&jw);
+}
+
 static inline void test_flag_and_insert(struct prio_queue *q, struct commit *c, int flag)
 {
 	if (c->object.flags & flag)
@@ -3329,6 +3349,8 @@ static void explore_walk_step(struct rev_info *revs)
 	if (repo_parse_commit_gently(revs->repo, c, 1) < 0)
 		return;
 
+	count_explore_walked++;
+
 	if (revs->sort_order == REV_SORT_BY_AUTHOR_DATE)
 		record_author_date(&info->author_date, c);
 
@@ -3367,6 +3389,8 @@ static void indegree_walk_step(struct rev_info *revs)
 	if (repo_parse_commit_gently(revs->repo, c, 1) < 0)
 		return;
 
+	count_indegree_walked++;
+
 	explore_to_depth(revs, commit_graph_generation(c));
 
 	for (p = c->parents; p; p = p->next) {
@@ -3476,6 +3500,11 @@ static void init_topo_walk(struct rev_info *revs)
 	 */
 	if (revs->sort_order == REV_SORT_IN_GRAPH_ORDER)
 		prio_queue_reverse(&info->topo_queue);
+
+	if (trace2_is_enabled() && !topo_walk_atexit_registered) {
+		atexit(trace2_topo_walk_statistics_atexit);
+		topo_walk_atexit_registered = 1;
+	}
 }
 
 static struct commit *next_topo_commit(struct rev_info *revs)
@@ -3502,6 +3531,8 @@ static void expand_topo_walk(struct rev_info *revs, struct commit *commit)
 			    oid_to_hex(&commit->object.oid));
 	}
 
+	count_topo_walked++;
+
 	for (p = commit->parents; p; p = p->next) {
 		struct commit *parent = p->item;
 		int *pi;

base-commit: 71ca53e8125e36efbda17293c50027d31681a41f
-- 
gitgitgadget