Commit b49773e7 authored by Damien Le Moal's avatar Damien Le Moal Committed by Jens Axboe
Browse files

block: Disable write plugging for zoned block devices



Simultaneously writing to a sequential zone of a zoned block device
from multiple contexts requires mutual exclusion for BIO issuing to
ensure that writes happen sequentially. However, even for a well
behaved user correctly implementing such synchronization, BIO plugging
may interfere and result in BIOs from the different contextx to be
reordered if plugging is done outside of the mutual exclusion section,
e.g. the plug was started by a function higher in the call chain than
the function issuing BIOs.

         Context A                     Context B

   | blk_start_plug()
   | ...
   | seq_write_zone()
     | mutex_lock(zone)
     | bio-0->bi_iter.bi_sector = zone->wp
     | zone->wp += bio_sectors(bio-0)
     | submit_bio(bio-0)
     | bio-1->bi_iter.bi_sector = zone->wp
     | zone->wp += bio_sectors(bio-1)
     | submit_bio(bio-1)
     | mutex_unlock(zone)
     | return
   | -----------------------> | seq_write_zone()
  				| mutex_lock(zone)
     				| bio-2->bi_iter.bi_sector = zone->wp
     				| zone->wp += bio_sectors(bio-2)
				| submit_bio(bio-2)
				| mutex_unlock(zone)
   | <------------------------- |
   | blk_finish_plug()

In the above example, despite the mutex synchronization ensuring the
correct BIO issuing order 0, 1, 2, context A BIOs 0 and 1 end up being
issued after BIO 2 of context B, when the plug is released with
blk_finish_plug().

While this problem can be addressed using the blk_flush_plug_list()
function (in the above example, the call must be inserted before the
zone mutex lock is released), a simple generic solution in the block
layer avoid this additional code in all zoned block device user code.
The simple generic solution implemented with this patch is to introduce
the internal helper function blk_mq_plug() to access the current
context plug on BIO submission. This helper returns the current plug
only if the target device is not a zoned block device or if the BIO to
be plugged is not a write operation. Otherwise, the caller context plug
is ignored and NULL returned, resulting is all writes to zoned block
device to never be plugged.

Signed-off-by: default avatarDamien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent 9305d5d7
......@@ -688,7 +688,7 @@ bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio,
struct request *rq;
struct list_head *plug_list;
plug = current->plug;
plug = blk_mq_plug(q, bio);
if (!plug)
return false;
......
......@@ -1973,7 +1973,7 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
blk_mq_bio_to_request(rq, bio, nr_segs);
plug = current->plug;
plug = blk_mq_plug(q, bio);
if (unlikely(is_flush_fua)) {
/* bypass scheduler for flush rq */
blk_insert_flush(rq);
......
......@@ -233,4 +233,36 @@ static inline void blk_mq_clear_mq_map(struct blk_mq_queue_map *qmap)
qmap->mq_map[cpu] = 0;
}
/*
* blk_mq_plug() - Get caller context plug
* @q: request queue
* @bio : the bio being submitted by the caller context
*
* Plugging, by design, may delay the insertion of BIOs into the elevator in
* order to increase BIO merging opportunities. This however can cause BIO
* insertion order to change from the order in which submit_bio() is being
* executed in the case of multiple contexts concurrently issuing BIOs to a
* device, even if these context are synchronized to tightly control BIO issuing
* order. While this is not a problem with regular block devices, this ordering
* change can cause write BIO failures with zoned block devices as these
* require sequential write patterns to zones. Prevent this from happening by
* ignoring the plug state of a BIO issuing context if the target request queue
* is for a zoned block device and the BIO to plug is a write operation.
*
* Return current->plug if the bio can be plugged and NULL otherwise
*/
static inline struct blk_plug *blk_mq_plug(struct request_queue *q,
struct bio *bio)
{
/*
* For regular block devices or read operations, use the context plug
* which may be NULL if blk_start_plug() was not executed.
*/
if (!blk_queue_is_zoned(q) || !op_is_write(bio_op(bio)))
return current->plug;
/* Zoned block device write operation case: do not plug the BIO */
return NULL;
}
#endif
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment