Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Thu, 15 Dec 2022 06:42:34 -1000
From:   Tejun Heo <tj@kernel.org>
To:     hanjinke <hanjinke.666@bytedance.com>
Cc:     josef@toxicpanda.com, axboe@kernel.dk, cgroups@vger.kernel.org,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [External] Re: [RFC PATCH] blk-throtl: Introduce sync queue for
 write ios
Message-ID: <Y5tOejLZrpMzH8/w@slm.duckdns.org>
References: <20221206163826.10700-1-hanjinke.666@bytedance.com>
 <Y5et48VryiKgL/eD@slm.duckdns.org>
 <f42d81c1-47c8-e283-1259-8c554bf46c46@bytedance.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <f42d81c1-47c8-e283-1259-8c554bf46c46@bytedance.com>
Precedence: bulk

Hello,

Sorry about the delay.

On Tue, Dec 13, 2022 at 11:55:16AM +0800, hanjinke wrote:
> If sq->queue[] were only classfied SYNC and ASYNC, some things may become a
> little difficult to handle。As we put sync write and read together into SYNC
> queue, the two may influence each other.
> Whit wbps=1M and rbps=100M configured, sync io likely be throtled while read
> ios after it may can be dispatched within the limit. In that case,
> maybe we should scan the whole SYNC queue to check read io.

Ah, yeah, you're right. The configuration per-rw so we have to split within.
Will review your latest patch.

Thanks.

-- 
tejun