git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH 3/3] Add parse_commit_for_rev() to take advantage of sha1-cache

From
Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Date
Apr 3, 2012, 06:55 UTC
Message-ID
<1333436109-16526-4-git-send-email-pclouds@gmail.com>
In-Reply-To
<53707c0a-3782-47a4-8a35-da7136ff4822@email.android.com>

This function tries to lookup sha1-cache. If it's found, struct commit is filled, no actual commit parsing is done. Otherwise parse_commit() is called.

Because sha1-cache only has information enough for rev machinery (tree, parent and date), this function is hardly suitable for general use.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
 builtin/reflog.c |    2 +-
 commit.c         |   36 +++++++++++++++++++++++++++++++-----
 commit.h         |    1 +
 log-tree.c       |    2 +-
 revision.c       |   10 +++++-----
 upload-pack.c    |    2 +-
 walker.c         |    2 +-
 7 files changed, 41 insertions(+), 14 deletions(-)
diff --git a/builtin/reflog.c b/builtin/reflog.c
index 062d7da..b15ff98 100644
--- a/builtin/reflog.c
+++ b/builtin/reflog.c
@@ -240,7 +240,7 @@ static void mark_reachable(struct expire_reflog_cb *cb)
 		free(entry);
 		if (commit->object.flags & REACHABLE)
 			continue;
-		if (parse_commit(commit))
+		if (parse_commit_limited(commit))
 			continue;
 		commit->object.flags |= REACHABLE;
 		if (commit->date < expire_limit) {
diff --git a/commit.c b/commit.c
index 946ea70..aa32658 100644
--- a/commit.c
+++ b/commit.c
@@ -7,6 +7,7 @@
 #include "revision.h"
 #include "notes.h"
 #include "gpg-interface.h"
+#include "sha1_cache.h"
 
 int save_commit_buffer = 1;
 
@@ -28,7 +29,11 @@ static struct commit *check_commit(struct object *obj,
 struct commit *lookup_commit_reference_gently(const unsigned char *sha1,
 					      int quiet)
 {
-	struct object *obj = deref_tag(parse_object(sha1), NULL, 0);
+	struct object *obj;
+	struct commit *c = lookup_commit(sha1);
+	if (c)
+		return c;
+	obj = deref_tag(parse_object(sha1), NULL, 0);
 
 	if (!obj)
 		return NULL;
@@ -258,6 +263,12 @@ int parse_commit_buffer(struct commit *item, const void *buffer, unsigned long s
 
 	if (item->object.parsed)
 		return 0;
+
+	if (item->parents) {
+		free_commit_list(item->parents);
+		item->parents = NULL;
+	}
+
 	item->object.parsed = 1;
 	tail += size;
 	if (tail <= bufptr + 46 || memcmp(bufptr, "tree ", 5) || bufptr[45] != '\n')
@@ -332,6 +343,21 @@ int parse_commit(struct commit *item)
 	return ret;
 }
 
+int parse_commit_limited(struct commit *c)
+{
+	unsigned char tree[20];
+	unsigned char parent[20];
+
+	if (c->object.parsed || c->parents)
+		return 0;
+	if (has_commit_cache(c->object.sha1, tree, parent, &c->date)) {
+		commit_list_insert(lookup_commit(parent), &c->parents);
+		c->tree = lookup_tree(tree);
+		return 0;
+	}
+	return parse_commit(c);
+}
+
 int find_commit_subject(const char *commit_buffer, const char **subject)
 {
 	const char *eol;
@@ -413,7 +439,7 @@ struct commit *pop_most_recent_commit(struct commit_list **list,
 
 	while (parents) {
 		struct commit *commit = parents->item;
-		if (!parse_commit(commit) && !(commit->object.flags & mark)) {
+		if (!parse_commit_limited(commit) && !(commit->object.flags & mark)) {
 			commit->object.flags |= mark;
 			commit_list_insert_by_date(commit, list);
 		}
@@ -605,10 +631,10 @@ static struct commit_list *merge_bases_many(struct commit *one, int n, struct co
 			return commit_list_insert(one, &result);
 	}
 
-	if (parse_commit(one))
+	if (parse_commit_limited(one))
 		return NULL;
 	for (i = 0; i < n; i++) {
-		if (parse_commit(twos[i]))
+		if (parse_commit_limited(twos[i]))
 			return NULL;
 	}
 
@@ -645,7 +671,7 @@ static struct commit_list *merge_bases_many(struct commit *one, int n, struct co
 			parents = parents->next;
 			if ((p->object.flags & flags) == flags)
 				continue;
-			if (parse_commit(p))
+			if (parse_commit_limited(p))
 				return NULL;
 			p->object.flags |= flags;
 			commit_list_insert_by_date(p, &list);
diff --git a/commit.h b/commit.h
index 154c0e3..113303b 100644
--- a/commit.h
+++ b/commit.h
@@ -47,6 +47,7 @@ struct commit *lookup_commit_or_die(const unsigned char *sha1, const char *ref_n
 
 int parse_commit_buffer(struct commit *item, const void *buffer, unsigned long size);
 int parse_commit(struct commit *item);
+int parse_commit_limited(struct commit *item);
 
 /* Find beginning and length of commit subject. */
 int find_commit_subject(const char *commit_buffer, const char **subject);
diff --git a/log-tree.c b/log-tree.c
index cea8756..047ddec 100644
--- a/log-tree.c
+++ b/log-tree.c
@@ -643,7 +643,7 @@ void show_log(struct rev_info *opt)
 		show_mergetag(opt, commit);
 	}
 
-	if (!commit->buffer)
+	if (!commit->buffer && parse_commit(commit) < 0)
 		return;
 
 	/*
diff --git a/revision.c b/revision.c
index c97d834..4c229fd 100644
--- a/revision.c
+++ b/revision.c
@@ -273,7 +273,7 @@ static struct commit *handle_commit(struct rev_info *revs, struct object *object
 	 */
 	if (object->type == OBJ_COMMIT) {
 		struct commit *commit = (struct commit *)object;
-		if (parse_commit(commit) < 0)
+		if (parse_commit_limited(commit) < 0)
 			die("unable to parse commit %s", name);
 		if (flags & UNINTERESTING) {
 			commit->object.flags |= UNINTERESTING;
@@ -465,7 +465,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit)
 		 */
 		if (revs->first_parent_only && nth_parent++)
 			break;
-		if (parse_commit(p) < 0)
+		if (parse_commit_limited(p) < 0)
 			die("cannot simplify commit %s (because of %s)",
 			    sha1_to_hex(commit->object.sha1),
 			    sha1_to_hex(p->object.sha1));
@@ -498,7 +498,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit)
 				 * IOW, we pretend this parent is a
 				 * "root" commit.
 				 */
-				if (parse_commit(p) < 0)
+				if (parse_commit_limited(p) < 0)
 					die("cannot simplify commit %s (invalid %s)",
 					    sha1_to_hex(commit->object.sha1),
 					    sha1_to_hex(p->object.sha1));
@@ -561,7 +561,7 @@ static int add_parents_to_list(struct rev_info *revs, struct commit *commit,
 			parent = parent->next;
 			if (p)
 				p->object.flags |= UNINTERESTING;
-			if (parse_commit(p) < 0)
+			if (parse_commit_limited(p) < 0)
 				continue;
 			if (p->parents)
 				mark_parents_uninteresting(p);
@@ -588,7 +588,7 @@ static int add_parents_to_list(struct rev_info *revs, struct commit *commit,
 	for (parent = commit->parents; parent; parent = parent->next) {
 		struct commit *p = parent->item;
 
-		if (parse_commit(p) < 0)
+		if (parse_commit_limited(p) < 0)
 			return -1;
 		if (revs->show_source && !p->util)
 			p->util = commit->util;
diff --git a/upload-pack.c b/upload-pack.c
index bb08e2e..d30e604 100644
--- a/upload-pack.c
+++ b/upload-pack.c
@@ -694,7 +694,7 @@ static void receive_needs(void)
 				/* make sure the real parents are parsed */
 				unregister_shallow(object->sha1);
 				object->parsed = 0;
-				if (parse_commit((struct commit *)object))
+				if (parse_commit_limited((struct commit *)object))
 					die("invalid commit");
 				parents = ((struct commit *)object)->parents;
 				while (parents) {
diff --git a/walker.c b/walker.c
index be389dc..7b818f5 100644
--- a/walker.c
+++ b/walker.c
@@ -71,7 +71,7 @@ static struct commit_list *complete = NULL;
 
 static int process_commit(struct walker *walker, struct commit *commit)
 {
-	if (parse_commit(commit))
+	if (parse_commit_limited(commit))
 		return -1;
 
 	while (complete && complete->item->date >= commit->date) {

-- 
1.7.3.1.256.g2539c.dirty
Previous: Nguyễn Thái Ngọc DuyNext: Nguyen Thai Ngoc Duy
Message 29 of 37 in “Git push performance problems with ~100K refs”
  1. Martin FickMar 30, 2012
  2. Junio C HamanoMar 30, 2012
  3. Martin FickMar 30, 2012
  4. Jeff KingMar 30, 2012
  5. Jeff KingMar 30, 2012
  6. Martin FickMar 30, 2012
  7. 1/3 add mergesort() for linked listsRené Scharfe, Mar 31, 2012
  8. Junio C HamanoApr 5, 2012
  9. René ScharfeApr 8, 2012
  10. Junio C HamanoApr 9, 2012
  11. Stephen BoydApr 11, 2012
  12. Junio C HamanoApr 11, 2012
  13. 2/3 commit: use mergesort() in commit_list_sort_by_date()René Scharfe, Mar 31, 2012
  14. 3/3 revision: insert unsorted, then sort in prepare_revision_walk()René Scharfe, Mar 31, 2012
  15. Martin FickMar 31, 2012
  16. Junio C HamanoMar 31, 2012
  17. Martin FickApr 2, 2012
  18. Shawn PearceApr 2, 2012
  19. Martin FickApr 2, 2012
  20. Shawn PearceApr 2, 2012
  21. Jeff KingApr 2, 2012
  22. Jeff KingApr 2, 2012
  23. Martin FickApr 2, 2012
  24. Nguyen Thai Ngoc DuyApr 3, 2012
  25. Martin FickApr 3, 2012
  26. 0/3 Commit cacheNguyễn Thái Ngọc Duy, Apr 3, 2012
  27. 1/3 parse_commit_buffer: rename a confusing variable nameNguyễn Thái Ngọc Duy, Apr 3, 2012
  28. 2/3 Add commit cache to help speed up commit traversalNguyễn Thái Ngọc Duy, Apr 3, 2012
  29. 3/3 Add parse_commit_for_rev() to take advantage of sha1-cacheNguyễn Thái Ngọc Duy, Apr 3, 2012
  30. Nguyen Thai Ngoc DuyApr 5, 2012
  31. Shawn PearceApr 6, 2012
  32. Nguyen Thai Ngoc DuyApr 7, 2012
  33. Nguyen Thai Ngoc DuyApr 3, 2012
  34. Jeff KingApr 2, 2012
  35. René ScharfeApr 2, 2012
  36. Jeff KingApr 3, 2012
  37. Jeff KingApr 3, 2012

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.