Re: [PATCH V4 3/5] blk-mq: stop to handle IO and drain IO before hctx becomes dead

[John Garry <john.garry@xxxxxxxxxx>]

On 14/10/2019 02:50, Ming Lei wrote:
> 		break;
> +		msleep(5);
> +	}
> +}
> +
>   static int blk_mq_hctx_notify_online(unsigned int cpu, struct hlist_node *node)
>   {
> +	struct blk_mq_hw_ctx *hctx = hlist_entry_safe(node,
> +			struct blk_mq_hw_ctx, cpuhp_online);
> +
> +	if ((cpumask_next_and(-1, hctx->cpumask, cpu_online_mask) == cpu) &&
> +	    (cpumask_next_and(cpu, hctx->cpumask, cpu_online_mask) >=
> +	     nr_cpu_ids)) {
> +		set_bit(BLK_MQ_S_INTERNAL_STOPPED, &hctx->state);
> +		blk_mq_hctx_drain_inflight_rqs(hctx);

Hi Ming,

> +        }

This goes a bit beyond a nit: The indentation needs to be fixed here.

Thanks,
John

>   	return 0;
>   }
>   
> @@ -2246,6 +2284,8 @@ static int blk_mq_hctx_notify_dead(unsigned int cpu, struct hlist_node *node)
>   	ctx