From: John Hubbard <jhubbard@nvidia.com>
To: Matthew Wilcox <willy@infradead.org>, <linux-fsdevel@vger.kernel.org>
Cc: <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: Tue, 18 Feb 2020 16:01:43 -0800 [thread overview]
Message-ID: <e3671faa-dfb3-ceba-3120-a445b2982a95@nvidia.com> (raw)
In-Reply-To: <20200217184613.19668-12-willy@infradead.org>
On 2/17/20 10:45 AM, 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.
>
"Also, remove the gfp argument from read_pages(), now that read_pages()
no longer does allocation."
Generally looks accurate, just a few notes below:
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
> include/linux/pagemap.h | 24 ++++++++++++++++++++++++
> mm/readahead.c | 34 +++++++++++++++++-----------------
> 2 files changed, 41 insertions(+), 17 deletions(-)
>
> diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
> index 982ecda2d4a2..3613154e79e4 100644
> --- a/include/linux/pagemap.h
> +++ b/include/linux/pagemap.h
> @@ -639,8 +639,32 @@ struct readahead_control {
> /* private: use the readahead_* accessors instead */
> pgoff_t _start;
> unsigned int _nr_pages;
> + unsigned int _batch_count;
> };
>
> +static inline struct page *readahead_page(struct readahead_control *rac)
> +{
> + struct page *page;
> +
> + if (!rac->_nr_pages)
> + return NULL;
> +
> + page = xa_load(&rac->mapping->i_pages, rac->_start);
> + VM_BUG_ON_PAGE(!PageLocked(page), page);
> + rac->_batch_count = hpage_nr_pages(page);
> +
> + return page;
> +}
> +
> +static inline void readahead_next(struct readahead_control *rac)
> +{
> + rac->_nr_pages -= rac->_batch_count;
> + rac->_start += rac->_batch_count;
> +}
> +
> +#define readahead_for_each(rac, page) \
> + for (; (page = readahead_page(rac)); readahead_next(rac))
> +
How about this instead? It uses the "for" loop fully and more naturally,
and is easier to read. And it does the same thing:
static inline struct page *readahead_page(struct readahead_control *rac)
{
struct page *page;
if (!rac->_nr_pages)
return NULL;
page = xa_load(&rac->mapping->i_pages, rac->_start);
VM_BUG_ON_PAGE(!PageLocked(page), page);
rac->_batch_count = hpage_nr_pages(page);
return page;
}
static inline struct page *readahead_next(struct readahead_control *rac)
{
rac->_nr_pages -= rac->_batch_count;
rac->_start += rac->_batch_count;
return readahead_page(rac);
}
#define readahead_for_each(rac, page) \
for (page = readahead_page(rac); page != NULL; \
page = readahead_page(rac))
> /* The number of pages in this readahead block */
> static inline unsigned int readahead_count(struct readahead_control *rac)
> {
> diff --git a/mm/readahead.c b/mm/readahead.c
> index bdc5759000d3..9e430daae42f 100644
> --- a/mm/readahead.c
> +++ b/mm/readahead.c
> @@ -113,12 +113,11 @@ int read_cache_pages(struct address_space *mapping, struct list_head *pages,
>
> EXPORT_SYMBOL(read_cache_pages);
>
> -static void read_pages(struct readahead_control *rac, struct list_head *pages,
> - gfp_t gfp)
> +static void read_pages(struct readahead_control *rac, struct list_head *pages)
> {
> const struct address_space_operations *aops = rac->mapping->a_ops;
> + struct page *page;
> struct blk_plug plug;
> - unsigned page_idx;
>
> blk_start_plug(&plug);
>
> @@ -127,19 +126,13 @@ static void read_pages(struct readahead_control *rac, struct list_head *pages,
> readahead_count(rac));
> /* Clean up the remaining pages */
> put_pages_list(pages);
> - goto out;
> - }
> -
> - for (page_idx = 0; page_idx < readahead_count(rac); page_idx++) {
> - struct page *page = lru_to_page(pages);
> - list_del(&page->lru);
> - if (!add_to_page_cache_lru(page, rac->mapping, page->index,
> - gfp))
> + } else {
> + readahead_for_each(rac, page) {
> aops->readpage(rac->file, page);
> - put_page(page);
> + put_page(page);
> + }
> }
>
> -out:
> blk_finish_plug(&plug);
> }
>
> @@ -159,6 +152,7 @@ void __do_page_cache_readahead(struct address_space *mapping,
> unsigned long i;
> loff_t isize = i_size_read(inode);
> gfp_t gfp_mask = readahead_gfp_mask(mapping);
> + bool use_list = mapping->a_ops->readpages;
fwiw, "bool have_readpages" seems like a better name (after all, that's how read_pages()
effectively is written: "if you have .readpages, then..."), but I can see both sides
of that bikeshed. :)
> struct readahead_control rac = {
> .mapping = mapping,
> .file = filp,
> @@ -196,8 +190,14 @@ void __do_page_cache_readahead(struct address_space *mapping,
> page = __page_cache_alloc(gfp_mask);
> if (!page)
> break;
> - 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) {
It would be a little safer from a maintenance point of view, to check for !=0, rather
than checking for < 0. Most (all?) existing callers check that way, and it's good
to stay with the pack there.
> + put_page(page);
> + goto read;
> + }
> if (i == nr_to_read - lookahead_size)
> SetPageReadahead(page);
> rac._nr_pages++;
> @@ -205,7 +205,7 @@ void __do_page_cache_readahead(struct address_space *mapping,
> continue;
> read:
> if (readahead_count(&rac))
> - read_pages(&rac, &page_pool, gfp_mask);
> + read_pages(&rac, &page_pool);
> rac._nr_pages = 0;
> rac._start = ++offset;
> }
> @@ -216,7 +216,7 @@ void __do_page_cache_readahead(struct address_space *mapping,
> * will then handle the error.
> */
> if (readahead_count(&rac))
> - read_pages(&rac, &page_pool, gfp_mask);
> + read_pages(&rac, &page_pool);
> BUG_ON(!list_empty(&page_pool));
> }
>
>
thanks,
--
John Hubbard
NVIDIA
next prev parent reply other threads:[~2020-02-19 0:01 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
2020-02-19 0:01 ` John Hubbard [this message]
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=e3671faa-dfb3-ceba-3120-a445b2982a95@nvidia.com \
--to=jhubbard@nvidia.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).