Re: How git affects kernel.org performance
[Posted January 10, 2007 by corbet]
| From: |
| Linus Torvalds <torvalds-AT-osdl.org> |
| To: |
| "H. Peter Anvin" <hpa-AT-zytor.com> |
| Subject: |
| Re: How git affects kernel.org performance |
| Date: |
| Sat, 6 Jan 2007 21:39:42 -0800 (PST) |
| Cc: |
| git-AT-vger.kernel.org, nigel-AT-nigel.suspend2.net,
"J.H." <warthog9-AT-kernel.org>,
Randy Dunlap <randy.dunlap-AT-oracle.com>,
Andrew Morton <akpm-AT-osdl.org>, Pavel Machek <pavel-AT-ucw.cz>,
kernel list <linux-kernel-AT-vger.kernel.org>,
webmaster-AT-kernel.org |
On Sat, 6 Jan 2007, H. Peter Anvin wrote:
>
> During extremely high load, it appears that what slows kernel.org down more
> than anything else is the time that each individual getdents() call takes.
> When I've looked this I've observed times from 200 ms to almost 2 seconds!
> Since an unpacked *OR* unpruned git tree adds 256 directories to a cleanly
> packed tree, you can do the math yourself.
"getdents()" is totally serialized by the inode semaphore. It's one of the
most expensive system calls in Linux, partly because of that, and partly
because it has to call all the way down into the filesystem in a way that
almost no other common system call has to (99% of all filesystem calls can
be handled basically at the VFS layer with generic caches - but not
getdents()).
So if there are concurrent readdirs on the same directory, they get
serialized. If there is any file creation/deletion activity in the
directory, it serializes getdents().
To make matters worse, I don't think it has any read-ahead at all when you
use hashed directory entries. So if you have cold-cache case, you'll read
every single block totally individually, and serialized. One block at a
time (I think the non-hashed case is likely also suspect, but that's a
separate issue)
In other words, I'm not at all surprised it hits on filldir time.
Especially on ext3.
Linus
