From: Benjamin Marzinski <bmarzins@redhat.com>
To: Mikulas Patocka <mpatocka@redhat.com>,
Mike Snitzer <snitzer@redhat.com>, Jens Axboe <axboe@kernel.dk>
Cc: dm-devel@lists.linux.dev, linux-block@vger.kernel.org,
Damien Le Moal <dlemoal@kernel.org>,
Christoph Hellwig <hch@lst.de>
Subject: [PATCH 4/7] dm: fix dm_blk_report_zones
Date: Sun, 9 Mar 2025 18:29:00 -0400 [thread overview]
Message-ID: <20250309222904.449803-5-bmarzins@redhat.com> (raw)
In-Reply-To: <20250309222904.449803-1-bmarzins@redhat.com>
If dm_get_live_table() returned NULL, dm_put_live_table() was never
called. Also, if md->zone_revalidate_map is set, check that
dm_blk_report_zones() is being called from the process that set it in
__bind(). Otherwise the zone resources could change while accessing
them. Finally, it is possible that md->zone_revalidate_map will change
while calling this function. Only read it once, so that we are always
using the same value. Otherwise we might miss a call to
dm_put_live_table().
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
---
drivers/md/dm-core.h | 1 +
drivers/md/dm-zone.c | 23 +++++++++++++++--------
2 files changed, 16 insertions(+), 8 deletions(-)
diff --git a/drivers/md/dm-core.h b/drivers/md/dm-core.h
index 3637761f3585..f3a3f2ef6322 100644
--- a/drivers/md/dm-core.h
+++ b/drivers/md/dm-core.h
@@ -141,6 +141,7 @@ struct mapped_device {
#ifdef CONFIG_BLK_DEV_ZONED
unsigned int nr_zones;
void *zone_revalidate_map;
+ struct task_struct *revalidate_map_task;
#endif
#ifdef CONFIG_IMA
diff --git a/drivers/md/dm-zone.c b/drivers/md/dm-zone.c
index 681058feb63b..11e19281bb64 100644
--- a/drivers/md/dm-zone.c
+++ b/drivers/md/dm-zone.c
@@ -56,24 +56,29 @@ int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
{
struct mapped_device *md = disk->private_data;
struct dm_table *map;
- int srcu_idx, ret;
+ struct dm_table *zone_revalidate_map = md->zone_revalidate_map;
+ int srcu_idx, ret = -EIO;
- if (!md->zone_revalidate_map) {
- /* Regular user context */
+ if (!zone_revalidate_map || md->revalidate_map_task != current) {
+ /*
+ * Regular user context or
+ * Zone revalidation during __bind() is in progress, but this
+ * call is from a different process
+ */
if (dm_suspended_md(md))
return -EAGAIN;
map = dm_get_live_table(md, &srcu_idx);
- if (!map)
- return -EIO;
} else {
/* Zone revalidation during __bind() */
- map = md->zone_revalidate_map;
+ map = zone_revalidate_map;
}
- ret = dm_blk_do_report_zones(md, map, sector, nr_zones, cb, data);
+ if (map)
+ ret = dm_blk_do_report_zones(md, map, sector, nr_zones, cb,
+ data);
- if (!md->zone_revalidate_map)
+ if (!zone_revalidate_map)
dm_put_live_table(md, srcu_idx);
return ret;
@@ -175,7 +180,9 @@ int dm_revalidate_zones(struct dm_table *t, struct request_queue *q)
* our table for dm_blk_report_zones() to use directly.
*/
md->zone_revalidate_map = t;
+ md->revalidate_map_task = current;
ret = blk_revalidate_disk_zones(disk);
+ md->revalidate_map_task = NULL;
md->zone_revalidate_map = NULL;
if (ret) {
--
2.48.1
next prev parent reply other threads:[~2025-03-09 22:29 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-09 22:28 [RFC PATCH 0/7] dm: fix issues with swapping dm tables Benjamin Marzinski
2025-03-09 22:28 ` [PATCH 1/7] dm: don't change md if dm_table_set_restrictions() fails Benjamin Marzinski
2025-03-09 23:18 ` Damien Le Moal
2025-03-10 16:39 ` Benjamin Marzinski
2025-03-09 22:28 ` [PATCH 2/7] dm: free table mempools if not used in __bind Benjamin Marzinski
2025-03-09 23:19 ` Damien Le Moal
2025-03-09 22:28 ` [PATCH 3/7] dm: handle failures in dm_table_set_restrictions Benjamin Marzinski
2025-03-09 23:25 ` Damien Le Moal
2025-03-10 17:37 ` Benjamin Marzinski
2025-03-10 18:15 ` Benjamin Marzinski
2025-03-10 23:27 ` Damien Le Moal
2025-03-14 13:38 ` Mikulas Patocka
2025-03-14 13:46 ` Mikulas Patocka
2025-03-10 23:16 ` Damien Le Moal
2025-03-09 22:29 ` Benjamin Marzinski [this message]
2025-03-09 23:27 ` [PATCH 4/7] dm: fix dm_blk_report_zones Damien Le Moal
2025-03-09 22:29 ` [PATCH 5/7] blk-zoned: clean up zone settings for devices without zwplugs Benjamin Marzinski
2025-03-09 23:31 ` Damien Le Moal
2025-03-09 22:29 ` [RFC PATCH 6/7] blk-zoned: modify blk_revalidate_disk_zones for bio-based drivers Benjamin Marzinski
2025-03-09 22:29 ` [RFC PATCH 7/7] dm: allow devices to revalidate existing zones Benjamin Marzinski
2025-03-09 23:59 ` Damien Le Moal
2025-03-10 17:43 ` Benjamin Marzinski
2025-03-10 23:19 ` Damien Le Moal
2025-03-10 23:42 ` Benjamin Marzinski
2025-03-11 0:00 ` Damien Le Moal
2025-03-09 23:16 ` [RFC PATCH 0/7] dm: fix issues with swapping dm tables Damien Le Moal
2025-03-10 16:38 ` Benjamin Marzinski
2025-03-10 23:13 ` Damien Le Moal
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=20250309222904.449803-5-bmarzins@redhat.com \
--to=bmarzins@redhat.com \
--cc=axboe@kernel.dk \
--cc=dlemoal@kernel.org \
--cc=dm-devel@lists.linux.dev \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=mpatocka@redhat.com \
--cc=snitzer@redhat.com \
/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