From: Nick Piggin <npiggin@suse.de>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
clameter@sgi.com, linux-mm@kvack.org, y-goto@jp.fujitsu.com
Subject: Re: Warning on memory offline (and possible in usual migration?)
Date: Tue, 22 Apr 2008 06:41:11 +0200 [thread overview]
Message-ID: <20080422044111.GF21993@wotan.suse.de> (raw)
In-Reply-To: <20080416113642.8ffd5684.akpm@linux-foundation.org>
On Wed, Apr 16, 2008 at 11:36:42AM -0700, Andrew Morton wrote:
> On Wed, 16 Apr 2008 20:00:36 +0900
> > With tons of printk, I think I found when it happens.
> >
> > Assume I use ia64/PAGE_SIZE=16k and ext3's blocksize=4k.
> > A page has 4 buffer_heads.
> >
> > Assume that a page is not Uptodate before issuing write_begin()
> >
> > At the end of writing to ext3, the kernel reaches here.
> > ==
> > static int __block_commit_write(struct inode *inode, struct page *page,
> > unsigned from, unsigned to)
> > {
> > int patrial=0;
> >
> > if (!All_buffers_to_this_page_is_uptodate)
> > partial = 1
> > if (!partial)
> > SetPageUptodate(page)
> > }
> > ==
> > To set a page as Uptodate, all buffers must be uptodate.
> >
> > But *all* buffers to this page is not necessary to be uptodate, here.
> > Then, the page can be not-up-to-date after commit-write.
> >
> > At page offlining, all buffers on the page seems to be marked as Uptodate
> > (by printk) but the page itself isn't. This seems strange.
> >
> > But I don't found who set Uptodate to the buffers.
> > And why page isn't up-to-date while all buffers are marked as up-to-date.
>
> That would imply that someone brought a buffer uptodate and didn't mark the
> page uptodate. That can happen if a read reads the buffer from disk or
> memsets all of it. Or if a write memsets all of it, or does
> copy_from_user() into all of it.
>
> > still chasing.
>
> umm..
>
> If you had some code which does
>
> pread(fd, buf, 1, 0);
> pread(fd, buf, 1, 4096);
> pread(fd, buf, 1, 8192);
> pread(fd, buf, 1, 12288);
>
> then I'd expect that each read would read a single buffer so we end up with
> four uptodate buffers, but nobody brings the entire page uptodate.
The generic read path AFAIK doesn't do partial buffers.
--
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:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2008-04-22 4:41 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-14 5:58 KAMEZAWA Hiroyuki
2008-04-14 17:46 ` Christoph Lameter
2008-04-15 10:13 ` KAMEZAWA Hiroyuki
2008-04-15 19:31 ` Christoph Lameter
2008-04-16 0:23 ` KAMEZAWA Hiroyuki
2008-04-16 2:23 ` KAMEZAWA Hiroyuki
2008-04-16 3:10 ` KAMEZAWA Hiroyuki
2008-04-16 5:22 ` KAMEZAWA Hiroyuki
2008-04-16 18:04 ` Christoph Lameter
2008-04-16 11:00 ` KAMEZAWA Hiroyuki
2008-04-16 18:36 ` Andrew Morton
2008-04-17 0:19 ` KAMEZAWA Hiroyuki
2008-04-17 6:38 ` Warning on memory offline (possible in migration ?) KAMEZAWA Hiroyuki
2008-04-17 6:43 ` Andrew Morton
2008-04-17 6:55 ` KAMEZAWA Hiroyuki
2008-04-22 4:41 ` Nick Piggin [this message]
2008-04-22 4:52 ` Warning on memory offline (and possible in usual migration?) Nick Piggin
2008-04-22 7:56 ` KAMEZAWA Hiroyuki
2008-04-22 9:43 ` Nick Piggin
2008-04-22 9:57 ` KAMEZAWA Hiroyuki
2008-04-22 19:16 ` Christoph Lameter
2008-04-23 0:48 ` Nick Piggin
2008-04-23 1:37 ` KAMEZAWA Hiroyuki
2008-04-23 2:41 ` KAMEZAWA Hiroyuki
2008-04-23 2:53 ` Nick Piggin
2008-04-23 3:44 ` KAMEZAWA Hiroyuki
2008-04-23 15:28 ` Nick Piggin
2008-04-24 1:34 ` KAMEZAWA Hiroyuki
2008-04-23 17:50 ` Christoph Lameter
2008-04-24 1:36 ` KAMEZAWA Hiroyuki
2008-04-24 19:11 ` Christoph Lameter
2008-04-25 0:11 ` KAMEZAWA Hiroyuki
2008-04-23 17:47 ` Christoph Lameter
2008-04-24 2:13 ` Nick Piggin
2008-04-29 7:20 ` KAMEZAWA Hiroyuki
2008-04-30 6:56 ` Nick Piggin
2008-04-30 7:04 ` KAMEZAWA Hiroyuki
2008-04-30 7:12 ` Andrew Morton
2008-04-30 7:22 ` KAMEZAWA Hiroyuki
2008-04-30 7:26 ` Nick Piggin
2008-04-30 17:59 ` Christoph Lameter
2008-04-30 18:01 ` Christoph Lameter
2008-05-01 1:44 ` Nick Piggin
2008-05-01 19:25 ` Christoph Lameter
2008-05-02 0:44 ` Nick Piggin
2008-05-02 1:07 ` Christoph Lameter
2008-05-02 1:23 ` Nick Piggin
2008-05-02 1:37 ` Christoph Lameter
2008-05-02 21:16 ` Christoph Lameter
2008-05-05 4:27 ` Nick Piggin
2008-05-05 17:28 ` Christoph Lameter
2008-05-06 8:52 ` Nick Piggin
2008-05-06 17:49 ` Christoph Lameter
2008-04-30 23:29 ` kamezawa.hiroyu
2008-05-01 0:34 ` Badari Pulavarty
2008-05-01 8:36 ` kamezawa.hiroyu
2008-04-22 4:50 ` Nick Piggin
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=20080422044111.GF21993@wotan.suse.de \
--to=npiggin@suse.de \
--cc=akpm@linux-foundation.org \
--cc=clameter@sgi.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=y-goto@jp.fujitsu.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