From: Matthew Wilcox <willy@infradead.org>
To: Jan Kara <jack@suse.cz>
Cc: "Kirill A . Shutemov" <kirill@shutemov.name>,
linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org, Hugh Dickins <hughd@google.com>,
William Kucharski <william.kucharski@oracle.com>
Subject: Re: [PATCH v2] page cache: Store only head pages in i_pages
Date: Wed, 13 Feb 2019 12:17:15 -0800 [thread overview]
Message-ID: <20190213201715.GU12668@bombadil.infradead.org> (raw)
In-Reply-To: <20190213144102.GA18351@quack2.suse.cz>
On Wed, Feb 13, 2019 at 03:41:02PM +0100, Jan Kara wrote:
> On Tue 12-02-19 10:34:54, Matthew Wilcox wrote:
> > Transparent Huge Pages are currently stored in i_pages as pointers to
> > consecutive subpages. This patch changes that to storing consecutive
> > pointers to the head page in preparation for storing huge pages more
> > efficiently in i_pages.
> >
> > Large parts of this are "inspired" by Kirill's patch
> > https://lore.kernel.org/lkml/20170126115819.58875-2-kirill.shutemov@linux.intel.com/
> >
> > Signed-off-by: Matthew Wilcox <willy@infradead.org>
>
> I like the idea!
Thanks! It's a step towards reducing the overhead of huge pages in the
page cache from (on x86) 520 pointers to a mere 8. Still not as good as
hugetlbfs, but I'm working on that too.
> > - pages[ret] = page;
> > + pages[ret] = find_subpage(page, xas.xa_index);
> > if (++ret == nr_pages) {
> > *start = page->index + 1;
> > goto out;
> > }
>
> So this subtly changes the behavior because now we will be returning in
> '*start' a different index. So you should rather use 'pages[ret]->index'
> instead.
You're right, I made a mistake there. However, seeing this:
https://lore.kernel.org/lkml/20190110030838.84446-1-yuzhao@google.com/
makes me think that I should be using xa_index + 1 there.
> Otherwise the patch looks good to me so feel free to add:
>
> Acked-by: Jan Kara <jack@suse.cz>
>
> after fixing these two.
Thanks!
next prev parent reply other threads:[~2019-02-13 20:17 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-12 18:34 [PATCH v2] page cache: Store only head pages in i_pages Matthew Wilcox
2019-02-13 14:41 ` Jan Kara
2019-02-13 20:17 ` Matthew Wilcox [this message]
2019-02-14 16:27 ` Jan Kara
2019-02-14 13:30 ` Kirill A. Shutemov
2019-02-14 20:53 ` Matthew Wilcox
2019-02-14 22:03 ` Kirill A. Shutemov
2019-02-14 22:25 ` Matthew Wilcox
2019-02-14 21:17 ` Matthew Wilcox
2019-02-14 22:08 ` Kirill A. Shutemov
2019-02-14 22:11 ` Matthew Wilcox
2019-02-14 22:29 ` Kirill A. Shutemov
2019-02-14 22:29 ` Kirill A. Shutemov
2019-02-15 20:28 ` Matthew Wilcox
2019-02-14 22:41 ` Kirill A. Shutemov
2019-02-15 20:20 ` Matthew Wilcox
2019-02-15 21:17 ` Kirill A. Shutemov
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=20190213201715.GU12668@bombadil.infradead.org \
--to=willy@infradead.org \
--cc=hughd@google.com \
--cc=jack@suse.cz \
--cc=kirill@shutemov.name \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=william.kucharski@oracle.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;
as well as URLs for NNTP newsgroup(s).