Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:6 as permitted sender) client-ip=2620:137:e000::3:6;
Message-ID: <35e3e173-8018-42d8-a6e8-7ba994ff6b17@acm.org>
Date:   Fri, 13 Oct 2023 14:13:08 -0700
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH V2] Consider inflight IO in io accounting for high latency
 devices
Content-Language: en-US
To:     Gulam Mohamed <gulam.mohamed@oracle.com>, axboe@kernel.dk,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20231013195559.1306345-1-gulam.mohamed@oracle.com>
From:   Bart Van Assche <bvanassche@acm.org>
In-Reply-To: <20231013195559.1306345-1-gulam.mohamed@oracle.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 10/13/23 12:55, Gulam Mohamed wrote:
> @@ -1015,7 +1018,13 @@ static inline void blk_account_io_start(struct request *req)
>   			req->part = req->q->disk->part0;
>   
>   		part_stat_lock();
> -		update_io_ticks(req->part, jiffies, false);
> +
> +		if (req->q->nr_hw_queues == 1) {
> +			hctx = xa_load(&req->q->hctx_table, 0);
> +			inflight = blk_mq_hctx_has_tags(hctx);
> +		}
> +
> +		update_io_ticks(req->part, jiffies, inflight);
>   		part_stat_unlock();
>   	}
>   }

blk_account_io_start() is called by blk_mq_bio_to_request(). So if I/O
statistics are enabled and if there is only a single hardware queue,
blk_mq_hctx_has_tags() will be called every time a bio is submitted?
The blk_mq_hctx_has_tags() function iterates over all tags. I would be
surprised if anyone would consider the overhead of this approach
acceptable.

Thanks,

Bart.