From: Dave Chinner <david@fromorbit.com>
To: Matthew Wilcox <willy@infradead.org>
Cc: linux-fsdevel@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, linux-btrfs@vger.kernel.org,
linux-erofs@lists.ozlabs.org, linux-ext4@vger.kernel.org,
linux-f2fs-devel@lists.sourceforge.net, cluster-devel@redhat.com,
ocfs2-devel@oss.oracle.com, linux-xfs@vger.kernel.org
Subject: Re: [PATCH v6 07/19] mm: Put readahead pages in cache earlier
Date: Wed, 19 Feb 2020 11:59:15 +1100 [thread overview]
Message-ID: <20200219005915.GV10776@dread.disaster.area> (raw)
In-Reply-To: <20200218154222.GQ7778@bombadil.infradead.org>
On Tue, Feb 18, 2020 at 07:42:22AM -0800, Matthew Wilcox wrote:
> On Tue, Feb 18, 2020 at 05:14:59PM +1100, Dave Chinner wrote:
> > On Mon, Feb 17, 2020 at 10:45:52AM -0800, Matthew Wilcox wrote:
> > > From: "Matthew Wilcox (Oracle)" <willy@infradead.org>
> > >
> > > At allocation time, put the pages in the cache unless we're using
> > > ->readpages. Add the readahead_for_each() iterator for the benefit of
> > > the ->readpage fallback. This iterator supports huge pages, even though
> > > none of the filesystems to be converted do yet.
> >
> > This could be better written - took me some time to get my head
> > around it and the code.
> >
> > "When populating the page cache for readahead, mappings that don't
> > use ->readpages need to have their pages added to the page cache
> > before ->readpage is called. Do this insertion earlier so that the
> > pages can be looked up immediately prior to ->readpage calls rather
> > than passing them on a linked list. This early insert functionality
> > is also required by the upcoming ->readahead method that will
> > replace ->readpages.
> >
> > Optimise and simplify the readpage loop by adding a
> > readahead_for_each() iterator to provide the pages we need to read.
> > This iterator also supports huge pages, even though none of the
> > filesystems have been converted to use them yet."
>
> Thanks, I'll use that.
>
> > > +static inline struct page *readahead_page(struct readahead_control *rac)
> > > +{
> > > + struct page *page;
> > > +
> > > + if (!rac->_nr_pages)
> > > + return NULL;
> >
> > Hmmmm.
> >
> > > +
> > > + page = xa_load(&rac->mapping->i_pages, rac->_start);
> > > + VM_BUG_ON_PAGE(!PageLocked(page), page);
> > > + rac->_batch_count = hpage_nr_pages(page);
> >
> > So we could have rac->_nr_pages = 2, and then we get an order 2
> > large page returned, and so rac->_batch_count = 4.
>
> Well, no, we couldn't. rac->_nr_pages is incremented by 4 when we add
> an order-2 page to the readahead.
I don't see any code that does that. :)
i.e. we aren't actually putting high order pages into the page
cache here - page_alloc() allocates order-0 pages) - so there's
nothing in the patch that tells me how rac->_nr_pages behaves
when allocating large pages...
IOWs, we have an undocumented assumption in the implementation...
> I can put a
> BUG_ON(rac->_batch_count > rac->_nr_pages)
> in here to be sure to catch any logic error like that.
Definitely necessary given that we don't insert large pages for
readahead yet. A comment explaining the assumptions that the
code makes for large pages is probably in order, too.
> > > - page->index = offset;
> > > - list_add(&page->lru, &page_pool);
> > > + if (use_list) {
> > > + page->index = offset;
> > > + list_add(&page->lru, &page_pool);
> > > + } else if (add_to_page_cache_lru(page, mapping, offset,
> > > + gfp_mask) < 0) {
> > > + put_page(page);
> > > + goto read;
> > > + }
> >
> > Ok, so that's why you put read code at the end of the loop. To turn
> > the code into spaghetti :/
> >
> > How much does this simplify down when we get rid of ->readpages and
> > can restructure the loop? This really seems like you're trying to
> > flatten two nested loops into one by the use of goto....
>
> I see it as having two failure cases in this loop. One for "page is
> already present" (which already existed) and one for "allocated a page,
> but failed to add it to the page cache" (which used to be done later).
> I didn't want to duplicate the "call read_pages()" code. So I reshuffled
> the code rather than add a nested loop. I don't think the nested loop
> is easier to read (we'll be at 5 levels of indentation for some statements).
> Could do it this way ...
Can we move the update of @rac inside read_pages()? The next
start offset^Windex we start at is rac._start + rac._nr_pages, right?
so read_pages() could do:
{
if (readahead_count(rac)) {
/* do readahead */
}
/* advance the readahead cursor */
rac->_start += rac->_nr_pages;
rac._nr_pages = 0;
}
and then we only need to call read_pages() in these cases and so
the requirement for avoiding duplicating code is avoided...
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
next prev parent reply other threads:[~2020-02-19 0:59 UTC|newest]
Thread overview: 111+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-17 18:45 [PATCH v6 00/19] Change readahead API Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 01/19] mm: Return void from various readahead functions Matthew Wilcox
2020-02-18 4:47 ` Dave Chinner
2020-02-18 21:05 ` John Hubbard
2020-02-18 21:21 ` Matthew Wilcox
2020-02-18 21:52 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 02/19] mm: Ignore return value of ->readpages Matthew Wilcox
2020-02-18 4:48 ` Dave Chinner
2020-02-18 21:33 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 03/19] mm: Use readahead_control to pass arguments Matthew Wilcox
2020-02-18 5:03 ` Dave Chinner
2020-02-18 13:56 ` Matthew Wilcox
2020-02-18 22:46 ` Dave Chinner
2020-02-18 22:52 ` Matthew Wilcox
2020-02-18 22:22 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 04/19] mm: Rearrange readahead loop Matthew Wilcox
2020-02-18 5:08 ` Dave Chinner
2020-02-18 13:57 ` Matthew Wilcox
2020-02-18 22:48 ` Dave Chinner
2020-02-18 22:33 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 04/16] mm: Tweak readahead loop slightly Matthew Wilcox
2020-02-18 22:57 ` John Hubbard
2020-02-18 23:00 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 05/16] mm: Put readahead pages in cache earlier Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 05/19] mm: Remove 'page_offset' from readahead loop Matthew Wilcox
2020-02-18 5:14 ` Dave Chinner
2020-02-18 23:08 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 06/16] mm: Add readahead address space operation Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 06/19] mm: rename readahead loop variable to 'i' Matthew Wilcox
2020-02-18 5:33 ` Dave Chinner
2020-02-18 23:11 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 07/16] mm: Add page_cache_readahead_limit Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 07/19] mm: Put readahead pages in cache earlier Matthew Wilcox
2020-02-18 6:14 ` Dave Chinner
2020-02-18 15:42 ` Matthew Wilcox
2020-02-19 0:59 ` Dave Chinner [this message]
2020-02-19 0:01 ` John Hubbard
2020-02-19 1:02 ` Matthew Wilcox
2020-02-19 1:13 ` John Hubbard
2020-02-19 3:24 ` John Hubbard
2020-02-19 14:41 ` Matthew Wilcox
2020-02-19 14:52 ` Christoph Hellwig
2020-02-19 15:01 ` Matthew Wilcox
2020-02-19 20:24 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 08/16] fs: Convert mpage_readpages to mpage_readahead Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 08/19] mm: Add readahead address space operation Matthew Wilcox
2020-02-18 6:21 ` Dave Chinner
2020-02-18 16:10 ` Matthew Wilcox
2020-02-19 1:04 ` Dave Chinner
2020-02-19 0:12 ` John Hubbard
2020-02-19 3:10 ` Eric Biggers
2020-02-19 3:35 ` Eric Biggers
2020-02-19 16:52 ` Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 09/16] btrfs: Convert from readpages to readahead Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 09/19] mm: Add page_cache_readahead_limit Matthew Wilcox
2020-02-18 6:31 ` Dave Chinner
2020-02-18 19:54 ` Matthew Wilcox
2020-02-19 1:08 ` Dave Chinner
2020-02-19 1:32 ` John Hubbard
2020-02-19 2:23 ` Matthew Wilcox
2020-02-19 2:46 ` John Hubbard
2020-02-17 18:45 ` [PATCH v6 10/16] erofs: Convert uncompressed files from readpages to readahead Matthew Wilcox
2020-02-17 18:45 ` [PATCH v6 10/19] fs: Convert mpage_readpages to mpage_readahead Matthew Wilcox
2020-02-18 1:51 ` [Ocfs2-devel] " Joseph Qi
2020-02-18 6:37 ` Dave Chinner
2020-02-19 2:48 ` John Hubbard
2020-02-19 3:28 ` Eric Biggers
2020-02-19 3:47 ` Matthew Wilcox
2020-02-19 3:55 ` Eric Biggers
2020-02-17 18:45 ` [PATCH v6 11/19] btrfs: Convert from readpages to readahead Matthew Wilcox
2020-02-18 6:57 ` Dave Chinner
2020-02-18 21:12 ` Matthew Wilcox
2020-02-19 1:23 ` Dave Chinner
2020-02-17 18:46 ` [PATCH v6 11/16] erofs: Convert compressed files " Matthew Wilcox
2020-02-19 2:34 ` Gao Xiang
2020-02-17 18:46 ` [PATCH v6 12/19] erofs: Convert uncompressed " Matthew Wilcox
2020-02-19 2:39 ` Gao Xiang
2020-02-19 3:04 ` Dave Chinner
2020-02-17 18:46 ` [PATCH v6 12/16] ext4: Convert " Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 13/19] erofs: Convert compressed files " Matthew Wilcox
2020-02-19 3:08 ` Dave Chinner
2020-02-17 18:46 ` [PATCH v6 13/16] f2fs: Convert " Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 14/19] ext4: " Matthew Wilcox
2020-02-19 3:16 ` Dave Chinner
2020-02-19 3:29 ` Eric Biggers
2020-02-17 18:46 ` [PATCH v6 14/16] fuse: " Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 15/19] f2fs: " Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 15/16] iomap: " Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 16/19] fuse: " Matthew Wilcox
2020-02-19 3:22 ` Dave Chinner
2020-02-17 18:46 ` [PATCH v6 16/16] mm: Use memalloc_nofs_save in readahead path Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 17/19] iomap: Restructure iomap_readpages_actor Matthew Wilcox
2020-02-19 3:17 ` John Hubbard
2020-02-19 5:35 ` Matthew Wilcox
2020-02-19 3:29 ` Dave Chinner
2020-02-19 6:04 ` Matthew Wilcox
2020-02-19 6:40 ` Dave Chinner
2020-02-19 17:06 ` Matthew Wilcox
2020-02-17 18:46 ` [PATCH v6 18/19] iomap: Convert from readpages to readahead Matthew Wilcox
2020-02-19 3:40 ` Dave Chinner
2020-02-17 18:46 ` [PATCH v6 19/19] mm: Use memalloc_nofs_save in readahead path Matthew Wilcox
2020-02-19 3:43 ` Dave Chinner
2020-02-19 5:22 ` Matthew Wilcox
2020-02-17 18:48 ` [PATCH v6 00/19] Change readahead API Matthew Wilcox
2020-02-18 4:56 ` Dave Chinner
2020-02-18 13:42 ` Matthew Wilcox
2020-02-18 21:26 ` Dave Chinner
2020-02-19 3:45 ` Dave Chinner
2020-02-19 3:48 ` Matthew Wilcox
2020-02-19 3:57 ` Dave Chinner
2020-02-18 20:49 ` John Hubbard
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=20200219005915.GV10776@dread.disaster.area \
--to=david@fromorbit.com \
--cc=cluster-devel@redhat.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-erofs@lists.ozlabs.org \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-xfs@vger.kernel.org \
--cc=ocfs2-devel@oss.oracle.com \
--cc=willy@infradead.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 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).