From: William Lee Irwin III <wli@holomorphy.com>
To: Andrew Morton <akpm@osdl.org>
Cc: Jeff Garzik <jgarzik@pobox.com>,
mbligh@aracnet.com, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, bzolnier@gmail.com, rddunlap@osdl.org,
axboe@suse.de
Subject: Re: [PATCH] Re: news about IDE PIO HIGHMEM bug (was: Re: 2.6.9-mm1)
Date: Wed, 27 Oct 2004 14:34:41 -0700 [thread overview]
Message-ID: <20041027213441.GC12934@holomorphy.com> (raw)
In-Reply-To: <20041027142914.197c72ed.akpm@osdl.org>
Jeff Garzik <jgarzik@pobox.com> wrote:
>> However, pfn_to_page(page_to_pfn(page) + 1) might be safer. If
>> rather slower. Is this patch acceptable to everyone? Andrew?
On Wed, Oct 27, 2004 at 02:29:14PM -0700, Andrew Morton wrote:
> spose so. The scatterlist API is being a bit silly there.
> It might be worthwhile doing:
> #ifdef CONFIG_DISCONTIGMEM
> #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + n)
> #else
> #define nth_page(page,n) ((page)+(n))
> #endif
This is actually not quite good enough. Zones are not guaranteed
to have adjacent mem_map[]'s even with CONFIG_DISCONTIGMEM=n. It may
make sense to prevent merging from spanning zones, but frankly the
overhead of the pfn_to_page()/page_to_pfn() is negligible in comparison
to the data movement and (when applicable) virtual windowing, where in
the merging code cpu overhead is a greater concern, particularly for
devices that don't require manual data movement.
-- wli
WARNING: multiple messages have this Message-ID (diff)
From: William Lee Irwin III <wli@holomorphy.com>
To: Andrew Morton <akpm@osdl.org>
Cc: Jeff Garzik <jgarzik@pobox.com>,
mbligh@aracnet.com, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, bzolnier@gmail.com, rddunlap@osdl.org,
axboe@suse.de
Subject: Re: [PATCH] Re: news about IDE PIO HIGHMEM bug (was: Re: 2.6.9-mm1)
Date: Wed, 27 Oct 2004 14:34:41 -0700 [thread overview]
Message-ID: <20041027213441.GC12934@holomorphy.com> (raw)
In-Reply-To: <20041027142914.197c72ed.akpm@osdl.org>
Jeff Garzik <jgarzik@pobox.com> wrote:
>> However, pfn_to_page(page_to_pfn(page) + 1) might be safer. If
>> rather slower. Is this patch acceptable to everyone? Andrew?
On Wed, Oct 27, 2004 at 02:29:14PM -0700, Andrew Morton wrote:
> spose so. The scatterlist API is being a bit silly there.
> It might be worthwhile doing:
> #ifdef CONFIG_DISCONTIGMEM
> #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + n)
> #else
> #define nth_page(page,n) ((page)+(n))
> #endif
This is actually not quite good enough. Zones are not guaranteed
to have adjacent mem_map[]'s even with CONFIG_DISCONTIGMEM=n. It may
make sense to prevent merging from spanning zones, but frankly the
overhead of the pfn_to_page()/page_to_pfn() is negligible in comparison
to the data movement and (when applicable) virtual windowing, where in
the merging code cpu overhead is a greater concern, particularly for
devices that don't require manual data movement.
-- wli
--
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-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
next prev parent reply other threads:[~2004-10-27 21:42 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <58cb370e041027074676750027@mail.gmail.com>
2004-10-27 15:14 ` news about IDE PIO HIGHMEM bug (was: Re: 2.6.9-mm1) Jeff Garzik
2004-10-27 15:14 ` Jeff Garzik
2004-10-27 15:52 ` Martin J. Bligh
2004-10-27 15:52 ` Martin J. Bligh
2004-10-27 15:59 ` Jeff Garzik
2004-10-27 15:59 ` Jeff Garzik
2004-10-27 17:36 ` Martin J. Bligh
2004-10-27 17:36 ` Martin J. Bligh
2004-10-27 16:01 ` Martin J. Bligh
2004-10-27 16:01 ` Martin J. Bligh
2004-10-27 16:35 ` [PATCH] " Jeff Garzik
2004-10-27 21:29 ` Andrew Morton
2004-10-27 21:29 ` Andrew Morton
2004-10-27 21:31 ` Jeff Garzik
2004-10-27 21:31 ` Jeff Garzik
2004-10-27 21:34 ` William Lee Irwin III [this message]
2004-10-27 21:34 ` William Lee Irwin III
2004-10-27 18:08 ` Christoph Hellwig
2004-10-27 18:08 ` Christoph Hellwig
2004-10-27 18:33 ` news about IDE PIO HIGHMEM bug Jeff Garzik
2004-10-27 18:33 ` Jeff Garzik
2004-10-27 18:48 ` William Lee Irwin III
2004-10-27 18:48 ` William Lee Irwin III
2004-10-28 0:18 ` William Lee Irwin III
2004-10-28 0:18 ` William Lee Irwin III
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=20041027213441.GC12934@holomorphy.com \
--to=wli@holomorphy.com \
--cc=akpm@osdl.org \
--cc=axboe@suse.de \
--cc=bzolnier@gmail.com \
--cc=jgarzik@pobox.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mbligh@aracnet.com \
--cc=rddunlap@osdl.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.