From: Hannes Reinecke <hare@suse.de>
To: Matthew Wilcox <willy@infradead.org>
Cc: Luis Chamberlain <mcgrof@kernel.org>,
Christoph Hellwig <hch@lst.de>, Jens Axboe <axboe@kernel.dk>,
Pankaj Raghav <p.raghav@samsung.com>,
linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org,
Hannes Reinecke <hare@suse.de>
Subject: [PATCH 01/18] mm/readahead: rework loop in page_cache_ra_unbounded()
Date: Mon, 18 Sep 2023 13:04:53 +0200 [thread overview]
Message-ID: <20230918110510.66470-2-hare@suse.de> (raw)
In-Reply-To: <20230918110510.66470-1-hare@suse.de>
Rework the loop in page_cache_ra_unbounded() to advance with
the number of pages in a folio instead of just one page at a time.
Signed-off-by: Hannes Reinecke <hare@suse.de>
---
mm/readahead.c | 17 +++++++++--------
1 file changed, 9 insertions(+), 8 deletions(-)
diff --git a/mm/readahead.c b/mm/readahead.c
index e815c114de21..40a5f1f65281 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -208,7 +208,7 @@ void page_cache_ra_unbounded(struct readahead_control *ractl,
struct address_space *mapping = ractl->mapping;
unsigned long index = readahead_index(ractl);
gfp_t gfp_mask = readahead_gfp_mask(mapping);
- unsigned long i;
+ unsigned long i = 0;
/*
* Partway through the readahead operation, we will have added
@@ -226,7 +226,7 @@ void page_cache_ra_unbounded(struct readahead_control *ractl,
/*
* Preallocate as many pages as we will need.
*/
- for (i = 0; i < nr_to_read; i++) {
+ do {
struct folio *folio = xa_load(&mapping->i_pages, index + i);
if (folio && !xa_is_value(folio)) {
@@ -239,8 +239,8 @@ void page_cache_ra_unbounded(struct readahead_control *ractl,
* not worth getting one just for that.
*/
read_pages(ractl);
- ractl->_index++;
- i = ractl->_index + ractl->_nr_pages - index - 1;
+ ractl->_index += folio_nr_pages(folio);
+ i = ractl->_index + ractl->_nr_pages - index;
continue;
}
@@ -251,15 +251,16 @@ void page_cache_ra_unbounded(struct readahead_control *ractl,
gfp_mask) < 0) {
folio_put(folio);
read_pages(ractl);
- ractl->_index++;
- i = ractl->_index + ractl->_nr_pages - index - 1;
+ ractl->_index += folio_nr_pages(folio);
+ i = ractl->_index + ractl->_nr_pages - index;
continue;
}
if (i == nr_to_read - lookahead_size)
folio_set_readahead(folio);
ractl->_workingset |= folio_test_workingset(folio);
- ractl->_nr_pages++;
- }
+ ractl->_nr_pages += folio_nr_pages(folio);
+ i += folio_nr_pages(folio);
+ } while (i < nr_to_read);
/*
* Now start the IO. We ignore I/O errors - if the folio is not
--
2.35.3
next prev parent reply other threads:[~2023-09-18 11:06 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-18 11:04 [RFC PATCH 00/18] block: update buffer_head for Large-block I/O Hannes Reinecke
2023-09-18 11:04 ` Hannes Reinecke [this message]
[not found] ` <CGME20230920115645eucas1p1c8ed9bf515c4532b3e6995f8078a863b@eucas1p1.samsung.com>
2023-09-20 11:56 ` [PATCH 01/18] mm/readahead: rework loop in page_cache_ra_unbounded() Pankaj Raghav
2023-09-20 14:13 ` Hannes Reinecke
2023-09-21 9:06 ` Pankaj Raghav
2023-09-20 14:18 ` Matthew Wilcox
2023-09-18 11:04 ` [PATCH 02/18] fs/mpage: use blocks_per_folio instead of blocks_per_page Hannes Reinecke
2023-09-18 13:15 ` Matthew Wilcox
2023-09-18 17:45 ` Hannes Reinecke
2023-09-18 11:04 ` [PATCH 03/18] block/buffer_head: introduce block_{index_to_sector,sector_to_index} Hannes Reinecke
2023-09-18 16:36 ` Matthew Wilcox
2023-09-18 17:42 ` Hannes Reinecke
2023-09-18 21:01 ` Matthew Wilcox
2023-09-18 11:04 ` [PATCH 04/18] fs/buffer.c: use accessor function to translate page index to sectors Hannes Reinecke
2023-10-20 19:37 ` Matthew Wilcox
2023-10-21 5:08 ` Matthew Wilcox
2023-10-23 5:03 ` Hannes Reinecke
2023-09-18 11:04 ` [PATCH 05/18] fs/mpage: " Hannes Reinecke
2023-09-18 11:04 ` [PATCH 06/18] fs: Allow fine-grained control of folio sizes Hannes Reinecke
2023-09-18 11:04 ` [PATCH 07/18] mm/filemap: allocate folios with mapping order preference Hannes Reinecke
2023-09-18 13:41 ` Matthew Wilcox
2023-09-18 17:34 ` Hannes Reinecke
2023-09-18 11:05 ` [PATCH 08/18] mm/readahead: " Hannes Reinecke
2023-09-18 13:11 ` kernel test robot
2023-09-18 20:46 ` kernel test robot
2023-09-18 11:05 ` [PATCH 09/18] fs/buffer: use mapping order in grow_dev_page() Hannes Reinecke
2023-09-18 14:00 ` Matthew Wilcox
2023-09-18 17:38 ` Hannes Reinecke
2023-09-18 11:05 ` [PATCH 10/18] block/bdev: lift restrictions on supported blocksize Hannes Reinecke
2023-09-18 11:05 ` [PATCH 11/18] block/bdev: enable large folio support for large logical block sizes Hannes Reinecke
2023-09-18 11:05 ` [PATCH 12/18] brd: convert to folios Hannes Reinecke
2023-09-18 11:05 ` [PATCH 13/18] brd: abstract page_size conventions Hannes Reinecke
2023-09-18 11:05 ` [PATCH 14/18] brd: use memcpy_{to,from}_folio() Hannes Reinecke
2023-09-18 11:05 ` [PATCH 15/18] brd: make sector size configurable Hannes Reinecke
2023-09-18 11:05 ` [PATCH 16/18] brd: make logical " Hannes Reinecke
2023-09-18 11:05 ` [PATCH 17/18] xfs: remove check for block sizes smaller than PAGE_SIZE Hannes Reinecke
2023-09-20 2:13 ` Dave Chinner
2023-09-18 11:05 ` [PATCH 18/18] nvme: enable logical block size > PAGE_SIZE Hannes Reinecke
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=20230918110510.66470-2-hare@suse.de \
--to=hare@suse.de \
--cc=axboe@kernel.dk \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=mcgrof@kernel.org \
--cc=p.raghav@samsung.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).