dm: do not allow readahead to limit IO size

mainline inclusion
from mainline-5.0-rc1
commit c6d6e9b0f6b4201c77f2cea3964dd122697e3543
category: bugfix
bugzilla: 5879
CVE: NA
---------------------------

Update DM to set the bdi's io_pages.  This fixes reads to be capped at
the device's max request size (even if user's read IO exceeds the
established readahead setting).

Conflicts:
	drivers/md/dm-table.c

Fixes: 9491ae4a ("mm: don't cap request size based on read-ahead setting")
Cc: stable@vger.kernel.org
Reviewed-by: NJens Axboe <axboe@kernel.dk>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Signed-off-by: NYufen Yu <yuyufen@huawei.com>
Reviewed-by: NJason Yan <yanaijie@huawei.com>
Signed-off-by: NYang Yingliang <yangyingliang@huawei.com>

drivers/md/dm-table.c

@@ -1937,6 +1937,9 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	 */
 	if (blk_queue_add_random(q) && dm_table_all_devices_attribute(t, device_is_not_random))
 		blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, q);
+
+	/* Allow reads to exceed readahead limits */
+	q->backing_dev_info->io_pages = limits->max_sectors >> (PAGE_SHIFT - 9);
 }
 
 unsigned int dm_table_get_num_targets(struct dm_table *t)