Linux block layer
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Jens Axboe <axboe@kernel.dk>
Cc: Damien Le Moal <Damien.LeMoal@wdc.com>,
	Hans Holmberg <hans@owltronix.com>,
	linux-block@vger.kernel.org
Subject: [PATCH 7/8] block: don't handle bio based drivers in blk_revalidate_disk_zones
Date: Tue,  3 Dec 2019 10:39:07 +0100	[thread overview]
Message-ID: <20191203093908.24612-8-hch@lst.de> (raw)
In-Reply-To: <20191203093908.24612-1-hch@lst.de>

bio based drivers only need to update q->nr_zones.  Do that manually
instead of overloading blk_revalidate_disk_zones to keep that function
simpler for the next round of changes that will rely even more on the
request based functionality.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/blk-zoned.c             | 16 +++++-----------
 drivers/block/null_blk_main.c | 12 +++++++++---
 drivers/md/dm-table.c         | 12 +++++++-----
 include/linux/blkdev.h        |  5 -----
 4 files changed, 21 insertions(+), 24 deletions(-)

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 0131f9e14bd1..51d427659ce7 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -419,8 +419,9 @@ static int blk_revalidate_zone_cb(struct blk_zone *zone, unsigned int idx,
  *
  * Helper function for low-level device drivers to (re) allocate and initialize
  * a disk request queue zone bitmaps. This functions should normally be called
- * within the disk ->revalidate method. For BIO based queues, no zone bitmap
- * is allocated.
+ * within the disk ->revalidate method for blk-mq based drivers.  For BIO based
+ * drivers only q->nr_zones needs to be updated so that the sysfs exposed value
+ * is correct.
  */
 int blk_revalidate_disk_zones(struct gendisk *disk)
 {
@@ -433,15 +434,8 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 
 	if (WARN_ON_ONCE(!blk_queue_is_zoned(q)))
 		return -EIO;
-
-	/*
-	 * BIO based queues do not use a scheduler so only q->nr_zones
-	 * needs to be updated so that the sysfs exposed value is correct.
-	 */
-	if (!queue_is_mq(q)) {
-		q->nr_zones = args.nr_zones;
-		return 0;
-	}
+	if (WARN_ON_ONCE(!queue_is_mq(q)))
+		return -EIO;
 
 	/*
 	 * Ensure that all memory allocations in this context are done as
diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c
index dd6026289fbf..068cd0ae6e2c 100644
--- a/drivers/block/null_blk_main.c
+++ b/drivers/block/null_blk_main.c
@@ -1576,11 +1576,17 @@ static int null_gendisk_register(struct nullb *nullb)
 	disk->queue		= nullb->q;
 	strncpy(disk->disk_name, nullb->disk_name, DISK_NAME_LEN);
 
+#ifdef CONFIG_BLK_DEV_ZONED
 	if (nullb->dev->zoned) {
-		ret = blk_revalidate_disk_zones(disk);
-		if (ret)
-			return ret;
+		if (queue_is_mq(nullb->q)) {
+			ret = blk_revalidate_disk_zones(disk);
+			if (ret)
+				return ret;
+		} else {
+			nullb->q->nr_zones = blkdev_nr_zones(disk);
+		}
 	}
+#endif
 
 	add_disk(disk);
 	return 0;
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 2ae0c1913766..0a2cc197f62b 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1954,12 +1954,14 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 	/*
 	 * For a zoned target, the number of zones should be updated for the
 	 * correct value to be exposed in sysfs queue/nr_zones. For a BIO based
-	 * target, this is all that is needed. For a request based target, the
-	 * queue zone bitmaps must also be updated.
-	 * Use blk_revalidate_disk_zones() to handle this.
+	 * target, this is all that is needed.
 	 */
-	if (blk_queue_is_zoned(q))
-		blk_revalidate_disk_zones(t->md->disk);
+#ifdef CONFIG_BLK_DEV_ZONED
+	if (blk_queue_is_zoned(q)) {
+		WARN_ON_ONCE(queue_is_mq(q));
+		q->nr_zones = blkdev_nr_zones(t->md->disk);
+	}
+#endif
 
 	/* Allow reads to exceed readahead limits */
 	q->backing_dev_info->io_pages = limits->max_sectors >> (PAGE_SHIFT - 9);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 503c4d4c5884..47eb22a3b7f9 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -375,11 +375,6 @@ static inline unsigned int blkdev_nr_zones(struct gendisk *disk)
 	return 0;
 }
 
-static inline int blk_revalidate_disk_zones(struct gendisk *disk)
-{
-	return 0;
-}
-
 static inline int blkdev_report_zones_ioctl(struct block_device *bdev,
 					    fmode_t mode, unsigned int cmd,
 					    unsigned long arg)
-- 
2.20.1


  parent reply	other threads:[~2019-12-03  9:39 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-03  9:39 avoid out of bounds zone bitmap access Christoph Hellwig
2019-12-03  9:39 ` [PATCH 1/8] null_blk: fix zone size paramter check Christoph Hellwig
2019-12-03  9:39 ` [PATCH 2/8] null_blk: cleanup null_gendisk_register Christoph Hellwig
2019-12-03  9:39 ` [PATCH 3/8] block: remove the empty line at the end of blk-zoned.c Christoph Hellwig
2019-12-03  9:39 ` [PATCH 4/8] block: simplify blkdev_nr_zones Christoph Hellwig
2019-12-03  9:39 ` [PATCH 5/8] block: replace seq_zones_bitmap with conv_zones_bitmap Christoph Hellwig
2019-12-03 14:02   ` Javier González
2019-12-03  9:39 ` [PATCH 6/8] block: allocate the zone bitmaps lazily Christoph Hellwig
2019-12-03 14:03   ` Javier González
2019-12-03  9:39 ` Christoph Hellwig [this message]
2019-12-03 14:04   ` [PATCH 7/8] block: don't handle bio based drivers in blk_revalidate_disk_zones Javier González
2019-12-03  9:39 ` [PATCH 8/8] block: set the zone size in blk_revalidate_disk_zones atomically Christoph Hellwig
2019-12-03 14:00   ` Javier González
2019-12-03 15:18     ` Christoph Hellwig
2019-12-03 15:34       ` Javier González
2019-12-03 15:42         ` Christoph Hellwig
2019-12-03 17:17           ` Javier González
2019-12-03 16:00 ` avoid out of bounds zone bitmap access Jens Axboe

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=20191203093908.24612-8-hch@lst.de \
    --to=hch@lst.de \
    --cc=Damien.LeMoal@wdc.com \
    --cc=axboe@kernel.dk \
    --cc=hans@owltronix.com \
    --cc=linux-block@vger.kernel.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