| Message ID | 20211102020705.2321858-1-yukuai3@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | blk-cgroup: fix missing put device in error path from blkg_conf_pref() | expand |
On 2021/11/02 10:07, Yu Kuai wrote: friendly ping... > If blk_queue_enter() failed due to queue is dying, the > blkdev_put_no_open() is needed because blkcg_conf_open_bdev() succeeded. > > Fixes: 0c9d338c8443 ("blk-cgroup: synchronize blkg creation against policy deactivation") > Signed-off-by: Yu Kuai <yukuai3@huawei.com> > --- > block/blk-cgroup.c | 9 +++++---- > 1 file changed, 5 insertions(+), 4 deletions(-) > > diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c > index 88b1fce90520..663aabfeba18 100644 > --- a/block/blk-cgroup.c > +++ b/block/blk-cgroup.c > @@ -640,7 +640,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, > */ > ret = blk_queue_enter(q, 0); > if (ret) > - return ret; > + goto fail; > > rcu_read_lock(); > spin_lock_irq(&q->queue_lock); > @@ -676,13 +676,13 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, > new_blkg = blkg_alloc(pos, q, GFP_KERNEL); > if (unlikely(!new_blkg)) { > ret = -ENOMEM; > - goto fail; > + goto fail_exit_queue; > } > > if (radix_tree_preload(GFP_KERNEL)) { > blkg_free(new_blkg); > ret = -ENOMEM; > - goto fail; > + goto fail_exit_queue; > } > > rcu_read_lock(); > @@ -722,9 +722,10 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, > fail_unlock: > spin_unlock_irq(&q->queue_lock); > rcu_read_unlock(); > +fail_exit_queue: > + blk_queue_exit(q); > fail: > blkdev_put_no_open(bdev); > - blk_queue_exit(q); > /* > * If queue was bypassing, we should retry. Do so after a > * short msleep(). It isn't strictly necessary but queue >
On Tue, Nov 02, 2021 at 10:07:05AM +0800, Yu Kuai wrote: > If blk_queue_enter() failed due to queue is dying, the > blkdev_put_no_open() is needed because blkcg_conf_open_bdev() succeeded. > > Fixes: 0c9d338c8443 ("blk-cgroup: synchronize blkg creation against policy deactivation") > Signed-off-by: Yu Kuai <yukuai3@huawei.com> Acked-by: Tejun Heo <tj@kernel.org> Thanks.
On 2021/11/02 10:07, Yu Kuai wrote: Hi, Jens friendly ping ... Thanks, Kuai > If blk_queue_enter() failed due to queue is dying, the > blkdev_put_no_open() is needed because blkcg_conf_open_bdev() succeeded. > > Fixes: 0c9d338c8443 ("blk-cgroup: synchronize blkg creation against policy deactivation") > Signed-off-by: Yu Kuai <yukuai3@huawei.com> > --- > block/blk-cgroup.c | 9 +++++---- > 1 file changed, 5 insertions(+), 4 deletions(-) > > diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c > index 88b1fce90520..663aabfeba18 100644 > --- a/block/blk-cgroup.c > +++ b/block/blk-cgroup.c > @@ -640,7 +640,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, > */ > ret = blk_queue_enter(q, 0); > if (ret) > - return ret; > + goto fail; > > rcu_read_lock(); > spin_lock_irq(&q->queue_lock); > @@ -676,13 +676,13 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, > new_blkg = blkg_alloc(pos, q, GFP_KERNEL); > if (unlikely(!new_blkg)) { > ret = -ENOMEM; > - goto fail; > + goto fail_exit_queue; > } > > if (radix_tree_preload(GFP_KERNEL)) { > blkg_free(new_blkg); > ret = -ENOMEM; > - goto fail; > + goto fail_exit_queue; > } > > rcu_read_lock(); > @@ -722,9 +722,10 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, > fail_unlock: > spin_unlock_irq(&q->queue_lock); > rcu_read_unlock(); > +fail_exit_queue: > + blk_queue_exit(q); > fail: > blkdev_put_no_open(bdev); > - blk_queue_exit(q); > /* > * If queue was bypassing, we should retry. Do so after a > * short msleep(). It isn't strictly necessary but queue >
On Tue, 2 Nov 2021 10:07:05 +0800, Yu Kuai wrote: > If blk_queue_enter() failed due to queue is dying, the > blkdev_put_no_open() is needed because blkcg_conf_open_bdev() succeeded. > > Applied, thanks! [1/1] blk-cgroup: fix missing put device in error path from blkg_conf_pref() commit: 15c30104965101b8e76b24d27035569d6613a7d6 Best regards,
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 88b1fce90520..663aabfeba18 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -640,7 +640,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, */ ret = blk_queue_enter(q, 0); if (ret) - return ret; + goto fail; rcu_read_lock(); spin_lock_irq(&q->queue_lock); @@ -676,13 +676,13 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, new_blkg = blkg_alloc(pos, q, GFP_KERNEL); if (unlikely(!new_blkg)) { ret = -ENOMEM; - goto fail; + goto fail_exit_queue; } if (radix_tree_preload(GFP_KERNEL)) { blkg_free(new_blkg); ret = -ENOMEM; - goto fail; + goto fail_exit_queue; } rcu_read_lock(); @@ -722,9 +722,10 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, fail_unlock: spin_unlock_irq(&q->queue_lock); rcu_read_unlock(); +fail_exit_queue: + blk_queue_exit(q); fail: blkdev_put_no_open(bdev); - blk_queue_exit(q); /* * If queue was bypassing, we should retry. Do so after a * short msleep(). It isn't strictly necessary but queue
If blk_queue_enter() failed due to queue is dying, the blkdev_put_no_open() is needed because blkcg_conf_open_bdev() succeeded. Fixes: 0c9d338c8443 ("blk-cgroup: synchronize blkg creation against policy deactivation") Signed-off-by: Yu Kuai <yukuai3@huawei.com> --- block/blk-cgroup.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)