Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v3] io_uring: reduce latency by reissueing the operation
To:     Olivier Langlois <olivier@trillion01.com>,
        Pavel Begunkov <asml.silence@gmail.com>,
        io-uring@vger.kernel.org, linux-kernel@vger.kernel.org
References: <4deda7761d61c189f4e2581828f852c8a1acb723.1624303174.git.olivier@trillion01.com>
From:   Jens Axboe <axboe@kernel.dk>
Message-ID: <82b94c49-b18d-94b2-cb8d-43cf54e402f3@kernel.dk>
Date:   Mon, 21 Jun 2021 14:05:24 -0600
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.10.0
MIME-Version: 1.0
In-Reply-To: <4deda7761d61c189f4e2581828f852c8a1acb723.1624303174.git.olivier@trillion01.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 6/21/21 1:22 PM, Olivier Langlois wrote:
> It is quite frequent that when an operation fails and returns EAGAIN,
> the data becomes available between that failure and the call to
> vfs_poll() done by io_arm_poll_handler().
> 
> Detecting the situation and reissuing the operation is much faster
> than going ahead and push the operation to the io-wq.

This now looks pretty good to me. I know you had some data associated
with this, would be great to include it in the commit message.

-- 
Jens Axboe