On 3/31/19 11:09 AM, Ming Lei wrote:
> Now resources for dispatch is freed in queue's release handler,
> we don't need to worry about the possible race between blk_cleanup_queue
> and run queue.
>
> So don't grab the queue usage counter before run queue in
> scsi_end_request().
Similar to prior one.
If something similar to do in the future, can we have "this is revert of
8dc765d438f1 ("SCSI: fix queue cleanup race before queue initialization is
done")......"
Thank you very much!
Dongli Zhang
>
> Cc: James Smart <james.smart@xxxxxxxxxxxx>
> Cc: Bart Van Assche <bart.vanassche@xxxxxxx>
> Cc: linux-scsi@xxxxxxxxxxxxxxx,
> Cc: Martin K . Petersen <martin.petersen@xxxxxxxxxx>,
> Cc: Christoph Hellwig <hch@xxxxxx>,
> Cc: James E . J . Bottomley <jejb@xxxxxxxxxxxxxxxxxx>,
> Cc: jianchao wang <jianchao.w.wang@xxxxxxxxxx>
> Signed-off-by: Ming Lei <ming.lei@xxxxxxxxxx>
> ---
> drivers/scsi/scsi_lib.c | 7 -------
> 1 file changed, 7 deletions(-)
>
> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
> index 601b9f1de267..18bf341d1236 100644
> --- a/drivers/scsi/scsi_lib.c
> +++ b/drivers/scsi/scsi_lib.c
> @@ -604,12 +604,6 @@ static bool scsi_end_request(struct request *req, blk_status_t error,
> */
> scsi_mq_uninit_cmd(cmd);
>
> - /*
> - * queue is still alive, so grab the ref for preventing it
> - * from being cleaned up during running queue.
> - */
> - percpu_ref_get(&q->q_usage_counter);
> -
> __blk_mq_end_request(req, error);
>
> if (scsi_target(sdev)->single_lun ||
> @@ -618,7 +612,6 @@ static bool scsi_end_request(struct request *req, blk_status_t error,
> else
> blk_mq_run_hw_queues(q, true);
>
> - percpu_ref_put(&q->q_usage_counter);
> put_device(&sdev->sdev_gendev);
> return false;
> }
>
