From: William Lee Irwin III <wli@holomorphy.com>
To: Andrea Arcangeli <andrea@suse.de>
Cc: Andrew Morton <akpm@zip.com.au>,
Martin Schwidefsky <schwidefsky@de.ibm.com>,
linux-kernel@vger.kernel.org, Alan Cox <alan@lxorguk.ukuu.org.uk>,
Rik van Riel <riel@conectiva.com.br>,
kuznet@ms2.inr.ac.ru, Andrey Savochkin <saw@saw.sw.com.sg>
Subject: Re: inode highmem imbalance fix [Re: Bug with shared memory.]
Date: Fri, 24 May 2002 00:51:59 -0700 [thread overview]
Message-ID: <20020524075159.GG14918@holomorphy.com> (raw)
In-Reply-To: <OF6D316E56.12B1A4B0-ONC1256BB9.004B5DB0@de.ibm.com> <3CE16683.29A888F8@zip.com.au> <20020520043040.GA21806@dualathlon.random> <20020524073341.GJ21164@dualathlon.random>
On Fri, May 24, 2002 at 09:33:41AM +0200, Andrea Arcangeli wrote:
> Here it is, you should apply it together with vm-35 that you need too
> for the bh/highmem balance (or on top of 2.4.19pre8aa3). I tested it
> slightly on uml and it didn't broke so far, so be careful because it's not
> very well tested yet. On the lines of what Alexey suggested originally,
> if goal isn't reached, in a second pass we shrink the cache too, but
> only if the cache is the only reason for the "pinning" beahiour of the
> inode. If for example there are dirty blocks of metadata or of data
> belonging to the inode we wakeup_bdflush instead and we never shrink the
> cache in such case. If the inode itself is dirty as well we let the two
> passes fail so we will schedule the work for keventd. This logic should
> ensure we never fall into shrinking the cache for no good reason and
> that we free the cache only for the inodes that we actually go ahead and
> free. (basically only dirty pages set with SetPageDirty aren't trapped
> by the logic before calling the invalidate, like ramfs, but that's
> expected of course, those pages cannot be damaged by the non destructive
> invalidate anyways)
> Comments?
I haven't had the chance to give this a test run yet, but it looks very
promising. I have a slight concern about the hold time of the inode_lock
because prune_icache() already generates some amount of contention,
but what you've presented appears to be necessary to prevent lethal
cache bloat, and so that concern is secondary at most. I'll give it a
test run tomorrow if no one else on-site gets to it first, though with
the proviso that I've not been involved in workloads triggering this
specific KVA exhaustion issue, so what testing I can do is limited.
Thanks,
Bill
next prev parent reply other threads:[~2002-05-24 7:52 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-05-14 15:13 Bug with shared memory Martin Schwidefsky
2002-05-14 19:33 ` Andrew Morton
2002-05-15 22:42 ` Mike Kravetz
2002-05-15 23:07 ` Andrew Morton
2002-05-17 17:53 ` Bill Davidsen
2002-05-17 20:07 ` Mike Kravetz
2002-05-17 20:29 ` Anton Blanchard
2002-05-20 4:30 ` Andrea Arcangeli
2002-05-20 5:21 ` Andrew Morton
2002-05-20 11:34 ` Andrey Savochkin
2002-05-20 14:15 ` Andrea Arcangeli
2002-05-20 19:24 ` Rik van Riel
2002-05-20 23:46 ` Andrea Arcangeli
2002-05-21 0:14 ` Martin J. Bligh
2002-05-21 1:40 ` Andrea Arcangeli
2002-05-20 16:22 ` Martin J. Bligh
2002-05-20 19:38 ` Rik van Riel
2002-05-20 20:06 ` William Lee Irwin III
2002-05-20 16:13 ` Martin J. Bligh
2002-05-20 16:37 ` Andrea Arcangeli
2002-05-20 17:23 ` Martin J. Bligh
2002-05-20 17:32 ` William Lee Irwin III
2002-05-24 7:33 ` inode highmem imbalance fix [Re: Bug with shared memory.] Andrea Arcangeli
2002-05-24 7:51 ` William Lee Irwin III [this message]
2002-05-24 8:04 ` Andrew Morton
2002-05-24 15:20 ` Andrea Arcangeli
2002-05-24 11:47 ` Ed Tomlinson
2002-05-30 11:25 ` Denis Lunev
2002-05-30 17:59 ` Andrea Arcangeli
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=20020524075159.GG14918@holomorphy.com \
--to=wli@holomorphy.com \
--cc=akpm@zip.com.au \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=andrea@suse.de \
--cc=kuznet@ms2.inr.ac.ru \
--cc=linux-kernel@vger.kernel.org \
--cc=riel@conectiva.com.br \
--cc=saw@saw.sw.com.sg \
--cc=schwidefsky@de.ibm.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).