On Fri, Jun 20, 2025 at 08:09:01PM +0530, Nilay Shroff wrote:
>
>
> On 6/17/25 8:37 PM, Ming Lei wrote:
> > On Mon, Jun 16, 2025 at 11:02:25PM +0530, Nilay Shroff wrote:
> >> In preparation for allocating sched_tags before freezing the request
> >> queue and acquiring ->elevator_lock, move the elevator queue allocation
> >> logic from the elevator ops ->init_sched callback into blk_mq_init_sched.
> >>
> >> This refactoring provides a centralized location for elevator queue
> >> initialization, which makes it easier to store pre-allocated sched_tags
> >> in the struct elevator_queue during later changes.
> >>
> >> Signed-off-by: Nilay Shroff <nilay@xxxxxxxxxxxxx>
> >> ---
>
> [...]
>
> >> diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
> >> index 55a0fd105147..d914eb9d61a6 100644
> >> --- a/block/blk-mq-sched.c
> >> +++ b/block/blk-mq-sched.c
> >> @@ -475,6 +475,10 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
> >> q->nr_requests = 2 * min_t(unsigned int, q->tag_set->queue_depth,
> >> BLKDEV_DEFAULT_RQ);
> >>
> >> + eq = elevator_alloc(q, e);
> >> + if (!eq)
> >> + return -ENOMEM;
> >> +
> >> if (blk_mq_is_shared_tags(flags)) {
> >> ret = blk_mq_init_sched_shared_tags(q);
> >> if (ret)
> >
> > The above failure needs to be handled by kobject_put(&eq->kobj).
>
> I think here the elevator_alloc() failure occurs before we initialize
> eq->kobj. So we don't need to handle it with kobject_put(&eq->kobj)
> and instead simply returning -ENOMEM should be sufficient. Agree?
I meant the failure from blk_mq_init_sched_shared_tags(), which has to
call kobject_put() for correct cleanup.
Thanks,
Ming
