mm: don't cap request size based on read-ahead setting

We ran into a funky issue, where someone doing 256K buffered reads saw
128K requests at the device level.  Turns out it is read-ahead capping
the request size, since we use 128K as the default setting.  This
doesn't make a lot of sense - if someone is issuing 256K reads, they
should see 256K reads, regardless of the read-ahead setting, if the
underlying device can support a 256K read in a single command.

This patch introduces a bdi hint, io_pages.  This is the soft max IO
size for the lower level, I've hooked it up to the bdev settings here.
Read-ahead is modified to issue the maximum of the user request size,
and the read-ahead max size, but capped to the max request size on the
device side.  The latter is done to avoid reading ahead too much, if the
application asks for a huge read.  With this patch, the kernel behaves
like the application expects.

Link: http://lkml.kernel.org/r/1479498073-8657-1-git-send-email-axboe@fb.com
Signed-off-by: Jens Axboe <axboe@fb.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 28 insertions, 11 deletions

diff --git a/mm/readahead.c b/mm/readahead.c
index c8a955b1297e..c4ca70239233 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -207,12 +207,21 @@ out:
  * memory at once.
  */
 int force_page_cache_readahead(struct address_space *mapping, struct file *filp,
-		pgoff_t offset, unsigned long nr_to_read)
+			       pgoff_t offset, unsigned long nr_to_read)
 {
+	struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+	struct file_ra_state *ra = &filp->f_ra;
+	unsigned long max_pages;
+
 	if (unlikely(!mapping->a_ops->readpage && !mapping->a_ops->readpages))
 		return -EINVAL;
 
-	nr_to_read = min(nr_to_read, inode_to_bdi(mapping->host)->ra_pages);
+	/*
+	 * If the request exceeds the readahead window, allow the read to
+	 * be up to the optimal hardware IO size
+	 */
+	max_pages = max_t(unsigned long, bdi->io_pages, ra->ra_pages);
+	nr_to_read = min(nr_to_read, max_pages);
 	while (nr_to_read) {
 		int err;
 
@@ -369,10 +378,18 @@ ondemand_readahead(struct address_space *mapping,
 		   bool hit_readahead_marker, pgoff_t offset,
 		   unsigned long req_size)
 {
-	unsigned long max = ra->ra_pages;
+	struct backing_dev_info *bdi = inode_to_bdi(mapping->host);
+	unsigned long max_pages = ra->ra_pages;
 	pgoff_t prev_offset;
 
 	/*
+	 * If the request exceeds the readahead window, allow the read to
+	 * be up to the optimal hardware IO size
+	 */
+	if (req_size > max_pages && bdi->io_pages > max_pages)
+		max_pages = min(req_size, bdi->io_pages);
+
+	/*
 	 * start of file
 	 */
 	if (!offset)
@@ -385,7 +402,7 @@ ondemand_readahead(struct address_space *mapping,
 	if ((offset == (ra->start + ra->size - ra->async_size) ||
 	     offset == (ra->start + ra->size))) {
 		ra->start += ra->size;
-		ra->size = get_next_ra_size(ra, max);
+		ra->size = get_next_ra_size(ra, max_pages);
 		ra->async_size = ra->size;
 		goto readit;
 	}
@@ -400,16 +417,16 @@ ondemand_readahead(struct address_space *mapping,
 		pgoff_t start;
 
 		rcu_read_lock();
-		start = page_cache_next_hole(mapping, offset + 1, max);
+		start = page_cache_next_hole(mapping, offset + 1, max_pages);
 		rcu_read_unlock();
 
-		if (!start || start - offset > max)
+		if (!start || start - offset > max_pages)
 			return 0;
 
 		ra->start = start;
 		ra->size = start - offset;	/* old async_size */
 		ra->size += req_size;
-		ra->size = get_next_ra_size(ra, max);
+		ra->size = get_next_ra_size(ra, max_pages);
 		ra->async_size = ra->size;
 		goto readit;
 	}
@@ -417,7 +434,7 @@ ondemand_readahead(struct address_space *mapping,
 	/*
 	 * oversize read
 	 */
-	if (req_size > max)
+	if (req_size > max_pages)
 		goto initial_readahead;
 
 	/*
@@ -433,7 +450,7 @@ ondemand_readahead(struct address_space *mapping,
 	 * Query the page cache and look for the traces(cached history pages)
 	 * that a sequential stream would leave behind.
 	 */
-	if (try_context_readahead(mapping, ra, offset, req_size, max))
+	if (try_context_readahead(mapping, ra, offset, req_size, max_pages))
 		goto readit;
 
 	/*
@@ -444,7 +461,7 @@ ondemand_readahead(struct address_space *mapping,
 
 initial_readahead:
 	ra->start = offset;
-	ra->size = get_init_ra_size(req_size, max);
+	ra->size = get_init_ra_size(req_size, max_pages);
 	ra->async_size = ra->size > req_size ? ra->size - req_size : ra->size;
 
 readit:
@@ -454,7 +471,7 @@ readit:
 	 * the resulted next readahead window into the current one.
 	 */
 	if (offset == ra->start && ra->size == ra->async_size) {
-		ra->async_size = get_next_ra_size(ra, max);
+		ra->async_size = get_next_ra_size(ra, max_pages);
 		ra->size += ra->async_size;
 	}