Thread (9 messages) flat view 9 messages, 3 authors, 2021-06-22

Re: [PATCH] bitmaps: don't recurse into trees already in the bitmap

From: Jeff King <hidden>
Date: 2021-06-22 19:42:05

On Tue, Jun 22, 2021 at 12:47:57PM +0200, Patrick Steinhardt wrote:
Today I've been experimenting with the connectivity check of
git-receive-pack(1) once again to see whether I'm able to get a
performance improvement if the git-rev-list(1) command spawned as part
of the connectivity check uses `--use-bitmap-index`.

Turns out the answer is "no": it has exactly the same performance
characteristics when pushing into a bitmapped repository (linux.git)
compared to the version not using a bitmap index, except for once case
where it performs 70x worse: force-pushing "master~10:master" into a
bitmapped repo takes 11s instead of 0.15s with bitmaps enabled.

Just leaving this here as a note for anybody (or myself) to pick up at a
later point.
Thanks. I'd wager that's probably the "we find the exact bitmap for the
'haves'" problem. There are probably a lot of old refs with so-so bitmap
coverage, and we traverse all of them down to the nearest bitmap. If we
filled in the bitmap by traversing commits in timestamp or generation
order and ending at the merge-base, we could probably avoid looking at
them at all.

-Peff
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help