2022-11-15 13:54:01

by Yu Kuai

Subject: [PATCH v3 01/10] block: clear ->slave_dir when dropping the main slave_dir reference

From: Christoph Hellwig <[email protected]>

Zero out the pointer to ->slave_dir so that the holder code doesn't
incorrectly treat the object as alive when add_disk failed or after
del_gendisk was called.

Fixes: 89f871af1b26 ("dm: delay registering the gendisk")
Reported-by: Yu Kuai <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Yu Kuai <[email protected]>
---
block/genhd.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/block/genhd.c b/block/genhd.c
index 74026ce31405..e9501c66ba4d 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -530,6 +530,7 @@ int __must_check device_add_disk(struct device *parent, struct gendisk *disk,
rq_qos_exit(disk->queue);
out_put_slave_dir:
kobject_put(disk->slave_dir);
+ disk->slave_dir = NULL;
out_put_holder_dir:
kobject_put(disk->part0->bd_holder_dir);
out_del_integrity:
@@ -634,6 +635,7 @@ void del_gendisk(struct gendisk *disk)

kobject_put(disk->part0->bd_holder_dir);
kobject_put(disk->slave_dir);
+ disk->slave_dir = NULL;

part_stat_set_all(disk->part0, 0);
disk->part0->bd_stamp = 0;
--
2.31.1

2022-11-16 22:40:43

by Mike Snitzer

Subject: Re: [PATCH v3 01/10] block: clear ->slave_dir when dropping the main slave_dir reference

On Tue, Nov 15 2022 at 9:10P -0500,
Yu Kuai <[email protected]> wrote:

> From: Christoph Hellwig <[email protected]>
>
> Zero out the pointer to ->slave_dir so that the holder code doesn't
> incorrectly treat the object as alive when add_disk failed or after
> del_gendisk was called.
>
> Fixes: 89f871af1b26 ("dm: delay registering the gendisk")
> Reported-by: Yu Kuai <[email protected]>
> Signed-off-by: Christoph Hellwig <[email protected]>
> Signed-off-by: Yu Kuai <[email protected]>

Reviewed-by: Mike Snitzer <[email protected]>