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;
Date:   Thu, 06 May 2021 12:44:21 +0100
Message-ID: <87pmy4qe7e.wl-maz@kernel.org>
From:   Marc Zyngier <maz@kernel.org>
To:     He Ying <heying24@huawei.com>
Cc:     <vincent.guittot@linaro.org>, <Valentin.Schneider@arm.com>,
        <andrew@lunn.ch>, <catalin.marinas@arm.com>,
        <f.fainelli@gmail.com>, <gregory.clement@bootlin.com>,
        <jason@lakedaemon.net>, <kernel-team@android.com>,
        <linux-arm-kernel@lists.infradead.org>,
        <linux-kernel@vger.kernel.org>, <linux@arm.linux.org.uk>,
        <saravanak@google.com>, <sumit.garg@linaro.org>,
        <tglx@linutronix.de>, <will@kernel.org>
Subject: Re: [PATCH v3 03/16] arm64: Allow IPIs to be handled as normal interrupts
In-Reply-To: <c66367b0-e8a0-2b7b-13c3-c9413462357c@huawei.com>
References: <CAKfTPtD3GkTd+qQvyCmyU7Atu1ictDQ82YbPRdY9a+Kkr2DjvA@mail.gmail.com>
        <c66367b0-e8a0-2b7b-13c3-c9413462357c@huawei.com>
User-Agent: Wanderlust/2.15.9 (Almost Unreal) SEMI-EPG/1.14.7 (Harue)
 FLIM-LB/1.14.9 (=?UTF-8?B?R29qxY0=?=) APEL-LB/10.8 EasyPG/1.0.0 Emacs/27.1
 (x86_64-pc-linux-gnu) MULE/6.0 (HANACHIRUSATO)
MIME-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue")
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Thu, 06 May 2021 08:50:42 +0100,
He Ying <heying24@huawei.com> wrote:
> 
> Hello Marc,
> 
> We have faced a performance regression for handling ipis since this
> commit. I think it's the same issue reported by Vincent.

Can you share more details on what regression you have observed?
What's the workload, the system, the performance drop?

> I found you pointed out the possible two causes:
> 
> (1) irq_enter/exit on the rescheduling IPI means we reschedule much
> more often.

It turned out to be a red herring. We don't reschedule more often, but
we instead suffer from the overhead of irq_enter()/irq_exit().
However, this only matters for silly benchmarks, and no real-life
workload showed any significant regression. Have you identified such
realistic workload?

> (2) irq_domain lookups add some overhead.

While this is also a potential source of overhead, it turned out not
to be the case.

> But I don't see any following patches in mainline. So, are you still
> working on this issue?  Looking forward to your reply.

See [1]. However, there is probably better things to do than this
low-level specialisation of IPIs, and Thomas outlined what needs to be
done (see v1 of the patch series).

Thanks,

	M.

[1] https://lore.kernel.org/lkml/20201124141449.572446-1-maz@kernel.org/

-- 
Without deviation from the norm, progress is not possible.