[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e137c315-22e9-3ca9-b4ab-1ce58a53b14d@linux.alibaba.com>
Date: Wed, 25 Nov 2020 21:36:11 +0800
From: Baolin Wang <baolin.wang@...ux.alibaba.com>
To: Tejun Heo <tj@...nel.org>
Cc: axboe@...nel.dk, baolin.wang7@...il.com,
linux-block@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/7] blk-iocost: Move the usage ratio calculation to the
correct place
在 2020/11/25 20:19, Tejun Heo 写道:
> Hello,
>> @@ -2225,6 +2207,25 @@ static void ioc_timer_fn(struct timer_list *timer)
>> time_before64(vtime, now.vnow - ioc->margins.low))) {
>> u32 hwa, old_hwi, hwm, new_hwi;
>>
>> + if (vdone != vtime) {
>> + u64 inflight_us = DIV64_U64_ROUND_UP(
>> + cost_to_abs_cost(vtime - vdone, hw_inuse),
>> + ioc->vtime_base_rate);
>> +
>> + usage_us = max(usage_us, inflight_us);
>> + }
>> +
>> + /* convert to hweight based usage ratio */
>> + if (time_after64(iocg->activated_at, ioc->period_at))
>> + usage_dur = max_t(u64, now.now - iocg->activated_at, 1);
>> + else
>> + usage_dur = max_t(u64, now.now - ioc->period_at, 1);
>> +
>> + usage = clamp_t(u32,
>> + DIV64_U64_ROUND_UP(usage_us * WEIGHT_ONE,
>> + usage_dur),
>> + 1, WEIGHT_ONE);
>
> Can you please move the variable declarations inside the block together with
> the code?
Yes, sure. Will do in next version. Thanks.
Powered by blists - more mailing lists