From: Bart Van Assche <bvanassche@acm.org>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org, Christoph Hellwig <hch@lst.de>,
Damien Le Moal <dlemoal@kernel.org>,
Bart Van Assche <bvanassche@acm.org>
Subject: [PATCH v17 06/14] blk-zoned: Track the write pointer per zone
Date: Wed, 15 Jan 2025 14:46:40 -0800 [thread overview]
Message-ID: <20250115224649.3973718-7-bvanassche@acm.org> (raw)
In-Reply-To: <20250115224649.3973718-1-bvanassche@acm.org>
Derive the write pointer from successfully completed zoned writes. This
patch prepares for restoring the write pointer after a write has failed
either by the device (e.g. a unit attention or an unaligned write) or by
the driver (e.g. BLK_STS_RESOURCE).
Cc: Christoph Hellwig <hch@lst.de>
Cc: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
block/blk-zoned.c | 37 +++++++++++++++++++++++++++++--------
block/blk.h | 4 +++-
2 files changed, 32 insertions(+), 9 deletions(-)
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 9d08a54c201e..089c6740df4a 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -51,6 +51,8 @@ static const char *const zone_cond_name[] = {
* @zone_no: The number of the zone the plug is managing.
* @wp_offset: The zone write pointer location relative to the start of the zone
* as a number of 512B sectors.
+ * @wp_offset_compl: End offset for completed zoned writes as a number of 512
+ * byte sectors.
* @bio_list: The list of BIOs that are currently plugged.
* @bio_work: Work struct to handle issuing of plugged BIOs
* @rcu_head: RCU head to free zone write plugs with an RCU grace period.
@@ -63,6 +65,7 @@ struct blk_zone_wplug {
unsigned int flags;
unsigned int zone_no;
unsigned int wp_offset;
+ unsigned int wp_offset_compl;
struct bio_list bio_list;
struct work_struct bio_work;
struct rcu_head rcu_head;
@@ -554,6 +557,7 @@ static struct blk_zone_wplug *disk_get_and_lock_zone_wplug(struct gendisk *disk,
zwplug->flags = 0;
zwplug->zone_no = zno;
zwplug->wp_offset = bdev_offset_from_zone_start(disk->part0, sector);
+ zwplug->wp_offset_compl = zwplug->wp_offset;
bio_list_init(&zwplug->bio_list);
INIT_WORK(&zwplug->bio_work, blk_zone_wplug_bio_work);
zwplug->disk = disk;
@@ -612,6 +616,7 @@ static void disk_zone_wplug_set_wp_offset(struct gendisk *disk,
/* Update the zone write pointer and abort all plugged BIOs. */
zwplug->flags &= ~BLK_ZONE_WPLUG_NEED_WP_UPDATE;
zwplug->wp_offset = wp_offset;
+ zwplug->wp_offset_compl = zwplug->wp_offset;
disk_zone_wplug_abort(zwplug);
/*
@@ -1148,6 +1153,7 @@ void blk_zone_write_plug_bio_endio(struct bio *bio)
struct gendisk *disk = bio->bi_bdev->bd_disk;
struct blk_zone_wplug *zwplug =
disk_get_zone_wplug(disk, bio->bi_iter.bi_sector);
+ unsigned int end_sector;
unsigned long flags;
if (WARN_ON_ONCE(!zwplug))
@@ -1165,11 +1171,24 @@ void blk_zone_write_plug_bio_endio(struct bio *bio)
bio->bi_opf |= REQ_OP_ZONE_APPEND;
}
- /*
- * If the BIO failed, abort all plugged BIOs and mark the plug as
- * needing a write pointer update.
- */
- if (bio->bi_status != BLK_STS_OK) {
+ if (bio->bi_status == BLK_STS_OK) {
+ switch (bio_op(bio)) {
+ case REQ_OP_WRITE:
+ case REQ_OP_ZONE_APPEND:
+ case REQ_OP_WRITE_ZEROES:
+ end_sector = bdev_offset_from_zone_start(disk->part0,
+ bio->bi_iter.bi_sector + bio_sectors(bio));
+ if (end_sector > zwplug->wp_offset_compl)
+ zwplug->wp_offset_compl = end_sector;
+ break;
+ default:
+ break;
+ }
+ } else {
+ /*
+ * If the BIO failed, mark the plug as having an error to
+ * trigger recovery.
+ */
spin_lock_irqsave(&zwplug->lock, flags);
disk_zone_wplug_abort(zwplug);
zwplug->flags |= BLK_ZONE_WPLUG_NEED_WP_UPDATE;
@@ -1772,7 +1791,7 @@ EXPORT_SYMBOL_GPL(blk_zone_issue_zeroout);
static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
struct seq_file *m)
{
- unsigned int zwp_wp_offset, zwp_flags;
+ unsigned int zwp_wp_offset, zwp_wp_offset_compl, zwp_flags;
unsigned int zwp_zone_no, zwp_ref;
unsigned int zwp_bio_list_size;
unsigned long flags;
@@ -1782,11 +1801,13 @@ static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
zwp_flags = zwplug->flags;
zwp_ref = refcount_read(&zwplug->ref);
zwp_wp_offset = zwplug->wp_offset;
+ zwp_wp_offset_compl = zwplug->wp_offset_compl;
zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
spin_unlock_irqrestore(&zwplug->lock, flags);
- seq_printf(m, "%u 0x%x %u %u %u\n", zwp_zone_no, zwp_flags, zwp_ref,
- zwp_wp_offset, zwp_bio_list_size);
+ seq_printf(m, "zone_no %u flags 0x%x ref %u wp_offset %u wp_offset_compl %u bio_list_size %u\n",
+ zwp_zone_no, zwp_flags, zwp_ref, zwp_wp_offset,
+ zwp_wp_offset_compl, zwp_bio_list_size);
}
int queue_zone_wplugs_show(void *data, struct seq_file *m)
diff --git a/block/blk.h b/block/blk.h
index 4904b86d5fec..2274253cfa58 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -470,8 +470,10 @@ static inline void blk_zone_update_request_bio(struct request *rq,
* the original BIO sector so that blk_zone_write_plug_bio_endio() can
* lookup the zone write plug.
*/
- if (req_op(rq) == REQ_OP_ZONE_APPEND || bio_zone_write_plugging(bio))
+ if (req_op(rq) == REQ_OP_ZONE_APPEND || bio_zone_write_plugging(bio)) {
bio->bi_iter.bi_sector = rq->__sector;
+ bio->bi_iter.bi_size = rq->__data_len;
+ }
}
void blk_zone_write_plug_bio_endio(struct bio *bio);
static inline void blk_zone_bio_endio(struct bio *bio)
next prev parent reply other threads:[~2025-01-15 22:47 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-15 22:46 [PATCH v17 00/14] Improve write performance for zoned UFS devices Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 01/14] block: Support block drivers that preserve the order of write requests Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 02/14] dm-linear: Report to the block layer that the write order is preserved Bart Van Assche
2025-01-17 18:04 ` Mikulas Patocka
2025-01-21 21:38 ` Bart Van Assche
2025-01-27 17:55 ` Mikulas Patocka
2025-01-15 22:46 ` [PATCH v17 03/14] block: Rework request allocation in blk_mq_submit_bio() Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 04/14] block: Support allocating from a specific software queue Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 05/14] blk-mq: Restore the zoned write order when requeuing Bart Van Assche
2025-01-15 22:46 ` Bart Van Assche [this message]
2025-01-15 22:46 ` [PATCH v17 07/14] blk-zoned: Defer error handling Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 08/14] blk-zoned: Add an argument to blk_zone_plug_bio() Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 09/14] blk-zoned: Support pipelining of zoned writes Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 10/14] scsi: core: Retry unaligned " Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 11/14] scsi: sd: Increase retry count for " Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 12/14] scsi: scsi_debug: Add the preserves_write_order module parameter Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 13/14] scsi: scsi_debug: Support injecting unaligned write errors Bart Van Assche
2025-01-15 22:46 ` [PATCH v17 14/14] scsi: ufs: Inform the block layer about write ordering Bart Van Assche
2025-01-16 7:43 ` Can Guo
2025-01-16 15:58 ` Bao D. Nguyen
2025-01-23 0:52 ` Bart Van Assche
2025-01-17 22:47 ` [PATCH v17 00/14] Improve write performance for zoned UFS devices Damien Le Moal
2025-01-21 21:57 ` Bart Van Assche
2025-01-23 4:16 ` Damien Le Moal
2025-01-27 23:01 ` Bart Van Assche
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=20250115224649.3973718-7-bvanassche@acm.org \
--to=bvanassche@acm.org \
--cc=axboe@kernel.dk \
--cc=dlemoal@kernel.org \
--cc=hch@lst.de \
--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