From: Alexander Viro <viro@math.psu.edu>
To: Jeff Garzik <jgarzik@mandrakesoft.com>
Cc: Andreas Dilger <adilger@turbolinux.com>,
kowalski@datrix.co.za, linux-kernel@vger.kernel.org,
Linus Torvalds <torvalds@transmeta.com>
Subject: [CFT][PATCH] Re: Fwd: Re: memory usage - dentry_cache
Date: Thu, 12 Apr 2001 03:10:46 -0400 (EDT) [thread overview]
Message-ID: <Pine.GSO.4.21.0104120257070.18135-100000@weyl.math.psu.edu> (raw)
In-Reply-To: <3AD550F0.8058FAA@mandrakesoft.com>
On Thu, 12 Apr 2001, Jeff Garzik wrote:
> Alexander Viro wrote:
> > We _have_ VM pressure there. However, such loads had never been used, so
> > there's no wonder that system gets unbalanced under them.
> >
> > I suspect that simple replacement of goto next; with continue; in the
> > fs/dcache.c::prune_dcache() may make situation seriously better.
>
> Awesome. With the obvious patch attached, some local ramfs problems
> disappeared, and my browser and e-mail program are no longer swapped out
> when doing a kernel build.
>
> Thanks :)
OK, how about wider testing? Theory: prune_dcache() goes through the
list of immediately killable dentries and tries to free given amount.
It has a "one warning" policy - it kills dentry if it sees it twice without
lookup finding that dentry in the interval. Unfortunately, as implemented
it stops when it had freed _or_ warned given amount. As the result, memory
pressure on dcache is less than expected.
Patch being:
--- fs/dcache.c Sun Apr 1 23:57:19 2001
+++ /tmp/dcache.c Thu Apr 12 03:07:39 2001
@@ -340,7 +340,7 @@
if (dentry->d_flags & DCACHE_REFERENCED) {
dentry->d_flags &= ~DCACHE_REFERENCED;
list_add(&dentry->d_lru, &dentry_unused);
- goto next;
+ continue;
}
dentry_stat.nr_unused--;
@@ -349,7 +349,6 @@
BUG();
prune_one_dentry(dentry);
- next:
if (!--count)
break;
}
next prev parent reply other threads:[~2001-04-12 7:11 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-04-11 11:36 Fwd: Re: memory usage - dentry_cache Marcin Kowalski
2001-04-12 4:48 ` Fwd: Re: memory usage - dentry_cacheg Andreas Dilger
2001-04-12 5:45 ` Alexander Viro
2001-04-12 6:53 ` [PATCH] " Jeff Garzik
2001-04-12 7:10 ` Alexander Viro [this message]
2001-04-12 8:44 ` [CFT][PATCH] Re: Fwd: Re: memory usage - dentry_cache David S. Miller
2001-04-12 12:27 ` Marcin Kowalski
2001-04-12 12:43 ` Yoann Vandoorselaere
2001-04-12 13:54 ` Alexander Viro
2001-04-12 17:27 ` Andreas Dilger
2001-04-12 7:00 ` Fwd: Re: memory usage - dentry_cacheg Andreas Dilger
2001-04-12 7:27 ` [race][RFC] d_flags use Alexander Viro
2001-04-12 8:01 ` David S. Miller
2001-04-12 8:06 ` Alexander Viro
2001-04-12 11:46 ` [PATCH] Re: memory usage - dentry_cacheg Ed Tomlinson
2001-04-12 14:56 ` Rik van Riel
2001-04-12 15:12 ` Alexander Viro
2001-04-13 1:34 ` Ed Tomlinson
2001-04-13 2:03 ` Alexander Viro
2001-04-13 4:45 ` Ed Tomlinson
2001-04-13 13:36 ` Ed Tomlinson
2001-04-12 15:30 ` Marcin Kowalski
2001-04-14 3:28 ` Paul
2001-04-12 14:34 ` [PATCH] Re: Fwd: " Jan Harkes
2001-04-12 14:50 ` Alexander Viro
2001-04-12 15:07 ` Rik van Riel
2001-04-12 15:27 ` Alexander Viro
2001-04-12 15:42 ` Alexander Viro
2001-04-12 15:48 ` Rik van Riel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Pine.GSO.4.21.0104120257070.18135-100000@weyl.math.psu.edu \
--to=viro@math.psu.edu \
--cc=adilger@turbolinux.com \
--cc=jgarzik@mandrakesoft.com \
--cc=kowalski@datrix.co.za \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@transmeta.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).