Commit 26208a7c authored by Yu Kuai's avatar Yu Kuai Committed by Song Liu

md/raid10: fix leak of 'r10bio->remaining' for recovery

raid10_sync_request() will add 'r10bio->remaining' for both rdev and
replacement rdev. However, if the read io fails, recovery_request_write()
returns without issuing the write io, in this case, end_sync_request()
is only called once and 'remaining' is leaked, cause an io hang.

Fix the problem by decreasing 'remaining' according to if 'bio' and
'repl_bio' is valid.

Fixes: 24afd80d ("md/raid10: handle recovery of replacement devices.")
Signed-off-by: default avatarYu Kuai <yukuai3@huawei.com>
Signed-off-by: default avatarSong Liu <song@kernel.org>
Link: https://lore.kernel.org/r/20230310073855.1337560-5-yukuai1@huaweicloud.com
parent 9fdfe6d4
......@@ -2615,11 +2615,22 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio)
{
struct r10conf *conf = mddev->private;
int d;
struct bio *wbio, *wbio2;
struct bio *wbio = r10_bio->devs[1].bio;
struct bio *wbio2 = r10_bio->devs[1].repl_bio;
/* Need to test wbio2->bi_end_io before we call
* submit_bio_noacct as if the former is NULL,
* the latter is free to free wbio2.
*/
if (wbio2 && !wbio2->bi_end_io)
wbio2 = NULL;
if (!test_bit(R10BIO_Uptodate, &r10_bio->state)) {
fix_recovery_read_error(r10_bio);
end_sync_request(r10_bio);
if (wbio->bi_end_io)
end_sync_request(r10_bio);
if (wbio2)
end_sync_request(r10_bio);
return;
}
......@@ -2628,14 +2639,6 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio)
* and submit the write request
*/
d = r10_bio->devs[1].devnum;
wbio = r10_bio->devs[1].bio;
wbio2 = r10_bio->devs[1].repl_bio;
/* Need to test wbio2->bi_end_io before we call
* submit_bio_noacct as if the former is NULL,
* the latter is free to free wbio2.
*/
if (wbio2 && !wbio2->bi_end_io)
wbio2 = NULL;
if (wbio->bi_end_io) {
atomic_inc(&conf->mirrors[d].rdev->nr_pending);
md_sync_acct(conf->mirrors[d].rdev->bdev, bio_sectors(wbio));
......
Markdown is supported
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