Received-SPF: pass (google.com: domain of linux-kernel+bounces-149425-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Sender: Tejun Heo <htejun@gmail.com>
Date: Wed, 17 Apr 2024 16:05:14 -1000
From: Tejun Heo <tj@kernel.org>
To: Yu Kuai <yukuai1@huaweicloud.com>
Cc: axboe@kernel.dk, chenhuacai@kernel.org, josef@toxicpanda.com,
	jhs@mojatatu.com, svenjoac@gmx.de, raven@themaw.net,
	pctammela@mojatatu.com, qde@naccy.de, zhaotianrui@loongson.cn,
	linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
	loongarch@lists.linux.dev, cgroups@vger.kernel.org,
	yi.zhang@huawei.com, yangerkun@huawei.com,
	"yukuai (C)" <yukuai3@huawei.com>
Subject: Re: [PATCH RFC v2 5/6] blk-throttle: support to destroy throtl_data
 when blk-throttle is disabled
Message-ID: <ZiB_2ryBOxq2_IyG@slm.duckdns.org>
References: <20240406080059.2248314-1-yukuai1@huaweicloud.com>
 <20240406080059.2248314-6-yukuai1@huaweicloud.com>
 <Zhl37slglnnTSMO7@slm.duckdns.org>
 <1bb85208-1224-77dc-f0b2-7b7a228ef70b@huaweicloud.com>
 <Zh6wx4mXZy_EOViH@slm.duckdns.org>
 <19086c09-3060-a4ce-4ac6-811a29653979@huaweicloud.com>
 <Zh8kPGAu2TG4Su2M@slm.duckdns.org>
 <1ed63126-d2e6-f0b6-42ef-127ecb464955@huaweicloud.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1ed63126-d2e6-f0b6-42ef-127ecb464955@huaweicloud.com>

Hello,

On Wed, Apr 17, 2024 at 09:39:43AM +0800, Yu Kuai wrote:
..
> I guess I'll do lazy init first, and then modularization for rq_qos,
> and leave blk-throtl there for now. Perhaps add a new throtl model in
> iocost can replace blk-throtl in the future.

That sounds like a plan.

> BTW, currently during test of iocost, I found that iocost can already
> achieve that, for example, by following configure:
> 
> echo "$dev enable=1 min=100 max=100" > qos
> echo "$dev wbps=4096 wseqiops=1 wrandiops=1" > model
> 
> In the test, I found that wbps and iops is actually limited to the
> set value.

Yeah, it shouldn't be too difficult to add .max support to iocost so that
you can say something like "this cgroup subtree can't use more than 60% of
available capacity". That'd be really cool to have.

Thanks.

-- 
tejun