From: Linus Torvalds <torvalds@transmeta.com>
To: "Quintela Carreira Juan J." <quintela@vexeta.dc.fi.udc.es>
Cc: Rajagopal Ananthanarayanan <ananth@sgi.com>,
Andrea Arcangeli <andrea@suse.de>,
Benjamin Redelings I <bredelin@ucla.edu>,
linux-mm@kvack.org
Subject: Re: [DATAPOINT] pre7-6 will not swap
Date: Mon, 8 May 2000 19:33:26 -0700 (PDT) [thread overview]
Message-ID: <Pine.LNX.4.10.10005081927200.839-100000@penguin.transmeta.com> (raw)
In-Reply-To: <ytt66sov6a9.fsf@vexeta.dc.fi.udc.es>
On 9 May 2000, Quintela Carreira Juan J. wrote:
> Hi Linus,
> I have tested two versions of the patch (against vanilla
> pre7-6), the first was to remove the test altogether (I think this is
> from Rajagopal):
I'll make my current pre7-7 available right away, to head off the
discussion.
I found out the real reason for the problem, and it was quite a lot more
subtle than I originally thought.
The "don't page out pages from zones that don't need it" test is a good
test, but it turns out that it triggers a rather serious problem: the way
the buffer cache dirty page handling is done is by having shrink_mmap() do
a "try_to_free_buffers()" on the pages it encounters that have
"page->buffer" set.
And doing that is quite important, because without that logic the buffers
don't get written to disk in a timely manner, nor do already-written
buffers get refiled to their proper lists. So you end up being "out of
memory" - not because the machine is really out of memory, but because
those buffers have a tendency to stick around if they aren't constantly
looked after by "try_to_free_buffers()".
So the real fix ended up being to re-order the tests in shrink_mmap() a
bit, so that try_to_free_buffers() is called even for pages that are on
a good zone that doesn't need any real balancing..
[ time passes ]
pre7-7 is there now.
Linus
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux.eu.org/Linux-MM/
next prev parent reply other threads:[~2000-05-09 2:33 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <8evk0f$7jote$1@fido.engr.sgi.com>
2000-05-06 17:12 ` Rajagopal Ananthanarayanan
2000-05-06 4:25 ` Benjamin Redelings I
2000-05-06 19:35 ` Linus Torvalds
2000-05-06 5:35 ` Benjamin Redelings I
2000-05-06 21:46 ` Rik van Riel
2000-05-06 22:24 ` Rajagopal Ananthanarayanan
2000-05-06 14:03 ` Benjamin Redelings I
2000-05-07 0:22 ` Rik van Riel
2000-05-07 2:23 ` Linus Torvalds
2000-05-07 17:40 ` Rik van Riel
2000-05-07 17:53 ` Linus Torvalds
2000-05-07 19:13 ` Rajagopal Ananthanarayanan
2000-05-07 19:30 ` Linus Torvalds
2000-05-08 20:40 ` gprof data for pre7-6 Rajagopal Ananthanarayanan
2000-05-09 1:52 ` [DATAPOINT] pre7-6 will not swap Quintela Carreira Juan J.
2000-05-09 2:28 ` Rajagopal Ananthanarayanan
2000-05-09 2:33 ` Linus Torvalds [this message]
2000-05-09 3:31 ` Rajagopal Ananthanarayanan
2000-05-09 15:56 ` [DATAPOINT] pre7-8 swaps with FREE mem? Benjamin Redelings I
2000-05-06 20:12 ` PG_referenced and lru_cache (cpu%) Roger Larsson
2000-05-06 18:31 ` Rik van Riel
2000-05-06 22:16 ` Roger Larsson
2000-05-05 8:07 [DATAPOINT] pre7-6 will not swap Benjamin Redelings I
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.LNX.4.10.10005081927200.839-100000@penguin.transmeta.com \
--to=torvalds@transmeta.com \
--cc=ananth@sgi.com \
--cc=andrea@suse.de \
--cc=bredelin@ucla.edu \
--cc=linux-mm@kvack.org \
--cc=quintela@vexeta.dc.fi.udc.es \
/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