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: [RFC 4/4] io_uring: implement futex wait
To:     Pavel Begunkov <asml.silence@gmail.com>, io-uring@vger.kernel.org
Cc:     Andres Freund <andres@anarazel.de>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Darren Hart <dvhart@infradead.org>,
        Davidlohr Bueso <dave@stgolabs.net>,
        linux-kernel@vger.kernel.org
References: <cover.1622558659.git.asml.silence@gmail.com>
 <e91af9d8f8d6e376635005fd111e9fe7a1c50fea.1622558659.git.asml.silence@gmail.com>
 <bd824ec8-48af-b554-67a1-7ce20fcf608c@kernel.dk>
 <409a624c-de75-0ee5-b65f-ee09fff34809@gmail.com>
From:   Jens Axboe <axboe@kernel.dk>
Message-ID: <bdc55fcd-b172-def4-4788-8bf808ccf6d6@kernel.dk>
Date:   Tue, 1 Jun 2021 10:01:11 -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: <409a624c-de75-0ee5-b65f-ee09fff34809@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 6/1/21 9:58 AM, Pavel Begunkov wrote:
> On 6/1/21 4:45 PM, Jens Axboe wrote:
>> On 6/1/21 8:58 AM, Pavel Begunkov wrote:
>>> Add futex wait requests, those always go through io-wq for simplicity.
>>
>> Not a huge fan of that, I think this should tap into the waitqueue
>> instead and just rely on the wakeup callback to trigger the event. That
>> would be a lot more efficient than punting to io-wq, both in terms of
>> latency on trigger, but also for efficiency if the app is waiting on a
>> lot of futexes.
> 
> Yes, that would be preferable, but looks futexes don't use
> waitqueues but some manual enqueuing into a plist_node, see
> futex_wait_queue_me() or mark_wake_futex().
> Did I miss it somewhere?

Yes, we'd need to augment that with a callback. I do think that's going
to be necessary, I don't see the io-wq solution working well outside of
the most basic of use cases. And even for that, it won't be particularly
efficient for single waits.

-- 
Jens Axboe