From: andrea@e-mind.com
To: Mark Hemment <markhe@veritas.com>
Cc: Jens Axboe <axboe@kernel.org>,
Linux Kernel <linux-kernel@vger.kernel.org>,
Rik van Riel <riel@conectiva.com.br>
Subject: Re: [patch] 4GB I/O, cut three
Date: Wed, 30 May 2001 16:26:07 +0200 [thread overview]
Message-ID: <20010530162607.D1408@athlon.random> (raw)
In-Reply-To: <20010530115538.B15089@suse.de> <Pine.LNX.4.21.0105301113550.7153-100000@alloc>
In-Reply-To: <Pine.LNX.4.21.0105301113550.7153-100000@alloc>; from markhe@veritas.com on Wed, May 30, 2001 at 11:59:50AM +0100
On Wed, May 30, 2001 at 11:59:50AM +0100, Mark Hemment wrote:
> Now, when HIGHMEM allocations come in (for page cache pages), they
> skip the HIGH zone and use the NORMAL zone (as it now has plenty
> of free pages) - the code at the top of __alloc_pages(), which
> checks against ->pages_low.
btw, I think such heuristic is horribly broken ;), the highmem zone
simply needs to be balanced if it is under the pages_low mark, just
skipping it and falling back into the normal zone that happens to be
above the low mark is the wrong thing to do.
> Also, the problem isn't as bad as it first looks - HIGHMEM page-cache
> pages do get "recycled" (reclaimed), but there is a slight imbalance.
there will always be some imbalance unless all allocations would be
capable of highmem (which will never happen). The only thing we can do
is to optimize the zone usage so we won't run out of normal pages unless
there was a good reason. Once we run out of normal pages we'll simply
return NULL and the reserved pool of highmem bounces will be used
instead (other callers will behave differently).
Andrea
next prev parent reply other threads:[~2001-05-30 14:26 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-05-29 14:07 [patch] 4GB I/O, cut three Jens Axboe
2001-05-29 14:11 ` Jens Axboe
2001-05-30 9:43 ` Mark Hemment
2001-05-30 9:55 ` Jens Axboe
2001-05-30 10:59 ` Mark Hemment
2001-05-30 14:26 ` andrea [this message]
2001-05-30 18:42 ` Rik van Riel
2001-05-30 18:57 ` Andrea Arcangeli
2001-05-30 18:57 ` Yoann Vandoorselaere
2001-05-30 19:18 ` Andrea Arcangeli
2001-05-30 19:23 ` Rik van Riel
2001-05-30 14:00 ` Andrea Arcangeli
2001-05-30 14:06 ` Jens Axboe
2001-05-30 18:36 ` Rik van Riel
2001-05-30 13:03 ` Mark Hemment
2001-05-30 13:24 ` Jens Axboe
2001-05-30 13:37 ` Mark Hemment
2001-05-30 13:40 ` Jens Axboe
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=20010530162607.D1408@athlon.random \
--to=andrea@e-mind.com \
--cc=axboe@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=markhe@veritas.com \
--cc=riel@conectiva.com.br \
/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