Re: git pull takes ~8 seconds on up-to-date Linux git tree

Jeff King [off-list ref] writes:

quoted

quoted

@@ -617,6 +618,8 @@ static struct commit_list *paint_down_to_common(struct commit *one, int n, struc
 
 	one->object.flags |= PARENT1;
 	commit_list_insert_by_date(one, &list);
+	if (!n)
+		return list;
 	for (i = 0; i < n; i++) {
 		twos[i]->object.flags |= PARENT2;
 		commit_list_insert_by_date(twos[i], &list);

This seems like an obvious optimization, but does it really have
anything to do with the patch at hand?

The function picks one and paints it against all others, but the
logic assumes there must be at least one other to paint against;
otherwise the traversal will not ever find a node that is painted
with both PARENT1 and PARENT2 to stop, leading us to traverse all
the way down to root.

On Fri, Oct 05, 2012 at 10:20:37PM -0700, Junio C Hamano wrote:

Jeff King [off-list ref] writes:

quoted

quoted

quoted

@@ -617,6 +618,8 @@ static struct commit_list *paint_down_to_common(struct commit *one, int n, struc
 
 	one->object.flags |= PARENT1;
 	commit_list_insert_by_date(one, &list);
+	if (!n)
+		return list;
 	for (i = 0; i < n; i++) {
 		twos[i]->object.flags |= PARENT2;
 		commit_list_insert_by_date(twos[i], &list);

This seems like an obvious optimization, but does it really have
anything to do with the patch at hand?

The function picks one and paints it against all others, but the
logic assumes there must be at least one other to paint against;
otherwise the traversal will not ever find a node that is painted
with both PARENT1 and PARENT2 to stop, leading us to traverse all
the way down to root.

Ah, OK. I was thinking it was just a way to skip the further logic,
which would come to the same answer (it does, just not quickly). Makes
sense.

-Peff