From: Andrea Arcangeli <andrea@suse.de>
To: Andrew Morton <akpm@osdl.org>
Cc: nickpiggin@yahoo.com.au, linux-kernel@vger.kernel.org, hugh@veritas.com
Subject: Re: smp race fix between invalidate_inode_pages* and do_no_page
Date: Wed, 11 Jan 2006 10:02:25 +0100 [thread overview]
Message-ID: <20060111090225.GY15897@opteron.random> (raw)
In-Reply-To: <20060111005134.3306b69a.akpm@osdl.org>
On Wed, Jan 11, 2006 at 12:51:34AM -0800, Andrew Morton wrote:
> Andrea Arcangeli <andrea@suse.de> wrote:
> >
> > On Wed, Jan 11, 2006 at 03:08:31PM +1100, Nick Piggin wrote:
> > > I'd be inclined to think a lock_page is not a big SMP scalability
> > > problem because the struct page's cacheline(s) will be written to
> > > several times in the process of refcounting anyway. Such a workload
> > > would also be running into tree_lock as well.
> >
> > I seem to recall you wanted to make the tree_lock a readonly lock for
> > readers for the exact same scalability reason? do_no_page is quite a
> > fast path for the tree lock too. But I totally agree the unavoidable is
> > the atomic_inc though, good point, so it worth more to remove the
> > tree_lock than to remove the page lock, the tree_lock can be avoided the
> > atomic_inc on page->_count not.
> >
> > The other bonus that makes this attractive is that then we can drop the
> > *whole* vm_truncate_count mess... vm_truncate_count and
> > inode->trunate_count exists for the only single reason that do_no_page
> > must not map into the pte a page that is under truncation.
>
> I think you'll find this hard - filemap_nopage() is the first to find the
> page but we need lock coverage up in do_no_page(). So the ->nopage
> protocol will need to be changed to "must return with the page locked". Or
> we add a new ->nopage_locked and call that if the vm_ops implements it.
Can't we avoid to change the protocol and use lock_page in do_no_page
instead? All we need to check before mailing out trying again is that
page->mapping is still there and then we have to set "page_mapping() ==
True" before unlocking (then the other side will have to block the pte
pte_lock running unmap_mapping_pages a second time with the page lock
held).
The main scary thing as far as I can tell, is the blocking lock_page. We
can't just do TryLockPage...
> But I agree it's a good change if we can pull it off.
Ok good thanks!
next prev parent reply other threads:[~2006-01-11 9:02 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-13 19:37 smp race fix between invalidate_inode_pages* and do_no_page Andrea Arcangeli
2005-12-13 21:02 ` Andrew Morton
2005-12-13 21:14 ` Andrea Arcangeli
2005-12-16 13:51 ` Andrea Arcangeli
2006-01-10 6:24 ` Andrea Arcangeli
2006-01-10 6:48 ` Andrea Arcangeli
2006-01-11 4:08 ` Nick Piggin
2006-01-11 8:23 ` Andrea Arcangeli
2006-01-11 8:51 ` Andrew Morton
2006-01-11 9:02 ` Andrea Arcangeli [this message]
2006-01-11 9:06 ` Andrew Morton
2006-01-11 9:13 ` Andrea Arcangeli
2006-01-11 20:49 ` Hugh Dickins
2006-01-11 21:05 ` Andrew Morton
2006-01-13 7:35 ` Nick Piggin
2006-01-13 7:47 ` Andrew Morton
2006-01-13 10:37 ` Nick Piggin
2006-03-31 12:36 ` Andrea Arcangeli
2006-04-02 5:17 ` Nick Piggin
2006-04-02 5:21 ` Andrew Morton
2006-04-07 19:18 ` Hugh Dickins
2006-01-11 9:39 ` Nick Piggin
2006-01-11 9:34 ` 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=20060111090225.GY15897@opteron.random \
--to=andrea@suse.de \
--cc=akpm@osdl.org \
--cc=hugh@veritas.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nickpiggin@yahoo.com.au \
/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).