Re: [PATCH] cvsimport: introduce -L<imit> option to workaround memory leaks

From: Linus Torvalds <torvalds@osdl.org>
Date: May 23, 2006, 16:25 UTC
Message-ID: <Pine.LNX.4.64.0605230919190.5623@g5.osdl.org>
In-Reply-To: <Pine.LNX.4.64.0605230848060.5623@g5.osdl.org>

On Tue, 23 May 2006, Linus Torvalds wrote:

> 
> So an unpacked git archive on ext3 (but not ext2, I believe: ext2 should 
> use the page cache for directories) ends up being very buffer-cache 
> intensive. And the buffer cache is basically deprecated..

A few notes: I'm not 100% sure things really fit in the 2GB (*), so it really may be IO limited, and the low CPU use is just because that machine also happens to have a dang fast next-gen Intel CPU that hasn't even been released yet.

Also, I do realize that hashed directories should actually decrease the buffer cache pressure too, just because we wouldn't need to read all of the directory for a lookup.

		Linus

(*) cvsps itself grows to 1.6GB of the 2GB and while that memory should be largely idle, the problem may simply be that we don't swap it out eagerly enough. Allowing filesystem metadata to swap out processes is something we've tuned against, because it tends to result in horrible interactive behaviour after a nightly "updatedb" run. So it's entirely possible that this is all normal..