All of lore.kernel.org
 help / color / mirror / Atom feed
From: Daniel Lezcano <daniel.lezcano@linaro.org>
To: Quan Xu <quan.xu0@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Peter Zijlstra <peterz@infradead.org>
Cc: Yang Zhang <yang.zhang.wz@gmail.com>,
	Len Brown <len.brown@intel.com>,
	kvm@vger.kernel.org, linux-doc@vger.kernel.org, x86@kernel.org,
	LKML <linux-kernel@vger.kernel.org>,
	virtualization@lists.linux-foundation.org,
	Kyle Huey <me@kylehuey.com>, Ingo Molnar <mingo@redhat.com>,
	Borislav Petkov <bp@alien8.de>, Quan Xu <quan.xu03@gmail.com>,
	Andy Lutomirski <luto@kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>,
	linux-fsdevel@vger.kernel.org, xen-devel@lists.xenproject.org,
	Tom Lendacky <thomas.lendacky@amd.com>,
	Tobias Klauser <tklauser@distanz.ch>
Subject: Re: [PATCH RFC v3 3/6] sched/idle: Add a generic poll before enter real idle path
Date: Thu, 16 Nov 2017 10:45:03 +0100	[thread overview]
Message-ID: <93a26005-aa37-82e1-5c04-a82c9027bac8__33216.2534039982$1511414930$gmane$org@linaro.org> (raw)
In-Reply-To: <46086489-5a01-16e1-9314-70ae53c01952@gmail.com>

On 16/11/2017 10:12, Quan Xu wrote:
> 
> 
> On 2017-11-16 06:03, Thomas Gleixner wrote:
>> On Wed, 15 Nov 2017, Peter Zijlstra wrote:
>>
>>> On Mon, Nov 13, 2017 at 06:06:02PM +0800, Quan Xu wrote:
>>>> From: Yang Zhang <yang.zhang.wz@gmail.com>
>>>>
>>>> Implement a generic idle poll which resembles the functionality
>>>> found in arch/. Provide weak arch_cpu_idle_poll function which
>>>> can be overridden by the architecture code if needed.
>>> No, we want less of those magic hooks, not more.
>>>
>>>> Interrupts arrive which may not cause a reschedule in idle loops.
>>>> In KVM guest, this costs several VM-exit/VM-entry cycles, VM-entry
>>>> for interrupts and VM-exit immediately. Also this becomes more
>>>> expensive than bare metal. Add a generic idle poll before enter
>>>> real idle path. When a reschedule event is pending, we can bypass
>>>> the real idle path.
>>> Why not do a HV specific idle driver?
>> If I understand the problem correctly then he wants to avoid the heavy
>> lifting in tick_nohz_idle_enter() in the first place, but there is
>> already
>> an interesting quirk there which makes it exit early.  See commit
>> 3c5d92a0cfb5 ("nohz: Introduce arch_needs_cpu"). The reason for this
>> commit
>> looks similar. But lets not proliferate that. I'd rather see that go
>> away.
> 
> agreed.
> 
> Even we can get more benifit than commit 3c5d92a0cfb5 ("nohz: Introduce
> arch_needs_cpu")
> in kvm guest. I won't proliferate that..
> 
>> But the irq_timings stuff is heading into the same direction, with a more
>> complex prediction logic which should tell you pretty good how long that
>> idle period is going to be and in case of an interrupt heavy workload
>> this
>> would skip the extra work of stopping and restarting the tick and
>> provide a
>> very good input into a polling decision.
> 
> 
> interesting. I have tested with IRQ_TIMINGS related code, which seems
> not working so far.

I don't know how you tested it, can you elaborate what you meant by
"seems not working so far" ?

There are still some work to do to be more efficient. The prediction
based on the irq timings is all right if the interrupts have a simple
periodicity. But as soon as there is a pattern, the current code can't
handle it properly and does bad predictions.

I'm working on a self-learning pattern detection which is too heavy for
the kernel, and with it we should be able to detect properly the
patterns and re-ajust the period if it changes. I'm in the process of
making it suitable for kernel code (both math and perf).

One improvement which can be done right now and which can help you is
the interrupts rate on the CPU. It is possible to compute it and that
will give an accurate information for the polling decision.



-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

  parent reply	other threads:[~2017-11-16  9:45 UTC|newest]

Thread overview: 99+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-13 10:05 [PATCH RFC v3 0/6] x86/idle: add halt poll support Quan Xu
2017-11-13 10:06 ` [PATCH RFC v3 1/6] x86/paravirt: Add pv_idle_ops to paravirt ops Quan Xu
2017-11-13 10:53   ` Juergen Gross
2017-11-13 11:09     ` Wanpeng Li
2017-11-13 11:09       ` Wanpeng Li
2017-11-13 11:09     ` Wanpeng Li
2017-11-13 11:09     ` Wanpeng Li
2017-11-14  7:02     ` Quan Xu
2017-11-14  7:02     ` Quan Xu
2017-11-14  7:02     ` Quan Xu
2017-11-14  7:12       ` Wanpeng Li
2017-11-14  7:12         ` Wanpeng Li
2017-11-14  8:15         ` Quan Xu
2017-11-14  8:15         ` Quan Xu
2017-11-14  8:15           ` Quan Xu
2017-11-14  8:22           ` Wanpeng Li
2017-11-14  8:22             ` Wanpeng Li
2017-11-14  8:22             ` Wanpeng Li
2017-11-14  8:22             ` Wanpeng Li
2017-11-14 10:23             ` Quan Xu
2017-11-14 10:23               ` Quan Xu
2017-11-14 10:23             ` Quan Xu
2017-11-14 10:23             ` Quan Xu
2017-11-14  8:22           ` Wanpeng Li
2017-11-14  8:15         ` Quan Xu
2017-11-14  7:12       ` Wanpeng Li
2017-11-14  7:12       ` Wanpeng Li
2017-11-14  7:30       ` Juergen Gross
2017-11-14  7:30       ` Juergen Gross
2017-11-14  9:38         ` Quan Xu
2017-11-14 10:27           ` Juergen Gross
2017-11-14 11:43             ` Quan Xu
2017-11-14 11:43             ` Quan Xu
2017-11-14 11:43             ` Quan Xu
2017-11-14 11:58               ` Juergen Gross
2017-11-14 11:58               ` Juergen Gross
2017-11-14 11:58                 ` Juergen Gross
2017-11-14 10:27           ` Juergen Gross
2017-11-14 10:27           ` Juergen Gross
2017-11-14  9:38         ` Quan Xu
2017-11-14  9:38         ` Quan Xu
2017-11-14  7:30       ` Juergen Gross
2017-11-13 10:53   ` Juergen Gross
2017-11-13 10:53   ` Juergen Gross
2017-11-13 10:06 ` Quan Xu
2017-11-13 10:06 ` [PATCH RFC v3 2/6] KVM guest: register kvm_idle_poll for pv_idle_ops Quan Xu
2017-11-13 10:06 ` Quan Xu
2017-11-13 10:06 ` [PATCH RFC v3 3/6] sched/idle: Add a generic poll before enter real idle path Quan Xu
2017-11-15 12:11   ` Peter Zijlstra
2017-11-15 12:11     ` Peter Zijlstra
2017-11-15 22:03     ` Thomas Gleixner
2017-11-15 22:03     ` Thomas Gleixner
2017-11-15 22:03     ` Thomas Gleixner
2017-11-16  8:45       ` Peter Zijlstra
2017-11-16  8:45         ` Peter Zijlstra
2017-11-16  8:58         ` Thomas Gleixner
2017-11-16  8:58         ` Thomas Gleixner
2017-11-16  8:58         ` Thomas Gleixner
2017-11-16  9:29         ` Quan Xu
2017-11-16  9:29         ` Quan Xu
2017-11-16  9:47           ` Thomas Gleixner
2017-11-16  9:47           ` Thomas Gleixner
2017-11-16  9:47           ` Thomas Gleixner
2017-11-16  9:29         ` Quan Xu
2017-11-16  8:45       ` Peter Zijlstra
2017-11-16  9:12       ` Quan Xu
2017-11-16  9:45         ` Daniel Lezcano
2017-11-20  7:05           ` Quan Xu
2017-11-20  7:05           ` Quan Xu
2017-11-20 18:01             ` Daniel Lezcano
2017-11-20 18:01             ` Daniel Lezcano
2017-11-20 18:01             ` Daniel Lezcano
2017-11-20  7:05           ` Quan Xu
2017-11-16  9:45         ` Daniel Lezcano
2017-11-16  9:45         ` Daniel Lezcano [this message]
2017-11-16  9:53         ` Thomas Gleixner
2017-11-16  9:53         ` Thomas Gleixner
2017-11-17 11:23           ` Quan Xu
2017-11-17 11:23           ` Quan Xu
2017-11-17 11:23           ` Quan Xu
2017-11-17 11:36             ` Thomas Gleixner
2017-11-17 11:36             ` Thomas Gleixner
2017-11-17 11:36               ` Thomas Gleixner
2017-11-17 11:36               ` Thomas Gleixner
2017-11-17 12:21               ` Quan Xu
2017-11-17 12:21                 ` Quan Xu
2017-11-17 12:21               ` Quan Xu
2017-11-17 12:21               ` Quan Xu
2017-11-16  9:53         ` Thomas Gleixner
2017-11-16  9:12       ` Quan Xu
2017-11-16  9:12       ` Quan Xu
2017-11-15 12:11   ` Peter Zijlstra
2017-11-13 10:06 ` Quan Xu
2017-11-15 21:31 ` [PATCH RFC v3 0/6] x86/idle: add halt poll support Konrad Rzeszutek Wilk
2017-11-15 21:31 ` [Xen-devel] " Konrad Rzeszutek Wilk
2017-11-15 21:31   ` Konrad Rzeszutek Wilk
2017-11-20  7:18   ` Quan Xu
2017-11-20  7:18   ` Quan Xu
2017-11-20  7:18   ` Quan Xu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='93a26005-aa37-82e1-5c04-a82c9027bac8__33216.2534039982$1511414930$gmane$org@linaro.org' \
    --to=daniel.lezcano@linaro.org \
    --cc=bp@alien8.de \
    --cc=hpa@zytor.com \
    --cc=kvm@vger.kernel.org \
    --cc=len.brown@intel.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=me@kylehuey.com \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=quan.xu03@gmail.com \
    --cc=quan.xu0@gmail.com \
    --cc=tglx@linutronix.de \
    --cc=thomas.lendacky@amd.com \
    --cc=tklauser@distanz.ch \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=x86@kernel.org \
    --cc=xen-devel@lists.xenproject.org \
    --cc=yang.zhang.wz@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.