Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH V2] sched: fair: Use the earliest break even
To:     Vincent Guittot <vincent.guittot@linaro.org>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        Qais Yousef <qais.yousef@arm.com>,
        Valentin Schneider <valentin.schneider@arm.com>
References: <20200311202625.13629-1-daniel.lezcano@linaro.org>
 <CAKfTPtAqeHhVCeSgE1DsaGGkM6nY-9oAvGw_6zWvv1bKyE85JQ@mail.gmail.com>
 <e6e8ff94-64f2-6404-e332-2e030fc7e332@linaro.org>
 <CAKfTPtBU1fyxWhR04QTCbvn07KgTqAHRVOt18D3TxmZSeiHQQQ@mail.gmail.com>
 <ee3bbfeb-ddd5-e4dc-3999-39370e7a6c73@linaro.org>
 <CAKfTPtDUmqYB1i7UcYXxcNjnQOoGufsB9do-9NxTMeWdJAfP2w@mail.gmail.com>
From:   Daniel Lezcano <daniel.lezcano@linaro.org>
Autocrypt: addr=daniel.lezcano@linaro.org; prefer-encrypt=mutual; keydata=
 xsFNBFv/yykBEADDdW8RZu7iZILSf3zxq5y8YdaeyZjI/MaqgnvG/c3WjFaunoTMspeusiFE
 sXvtg3ehTOoyD0oFjKkHaia1Zpa1m/gnNdT/WvTveLfGA1gH+yGes2Sr53Ht8hWYZFYMZc8V
 2pbSKh8wepq4g8r5YI1XUy9YbcTdj5mVrTklyGWA49NOeJz2QbfytMT3DJmk40LqwK6CCSU0
 9Ed8n0a+vevmQoRZJEd3Y1qXn2XHys0F6OHCC+VLENqNNZXdZE9E+b3FFW0lk49oLTzLRNIq
 0wHeR1H54RffhLQAor2+4kSSu8mW5qB0n5Eb/zXJZZ/bRiXmT8kNg85UdYhvf03ZAsp3qxcr
 xMfMsC7m3+ADOtW90rNNLZnRvjhsYNrGIKH8Ub0UKXFXibHbafSuq7RqyRQzt01Ud8CAtq+w
 P9EftUysLtovGpLSpGDO5zQ++4ZGVygdYFr318aGDqCljKAKZ9hYgRimPBToDedho1S1uE6F
 6YiBFnI3ry9+/KUnEP6L8Sfezwy7fp2JUNkUr41QF76nz43tl7oersrLxHzj2dYfWUAZWXva
 wW4IKF5sOPFMMgxoOJovSWqwh1b7hqI+nDlD3mmVMd20VyE9W7AgTIsvDxWUnMPvww5iExlY
 eIC0Wj9K4UqSYBOHcUPrVOKTcsBVPQA6SAMJlt82/v5l4J0pSQARAQABzSpEYW5pZWwgTGV6
 Y2FubyA8ZGFuaWVsLmxlemNhbm9AbGluYXJvLm9yZz7Cwa4EEwEIAEECGwEFCwkIBwIGFQoJ
 CAsCBBYCAwECHgECF4ACGQEWIQQk1ibyU76eh+bOW/SP9LjScWdVJwUCXAkeagUJDRnjhwAh
 CRCP9LjScWdVJxYhBCTWJvJTvp6H5s5b9I/0uNJxZ1Un69gQAJK0ODuKzYl0TvHPU8W7uOeu
 U7OghN/DTkG6uAkyqW+iIVi320R5QyXN1Tb6vRx6+yZ6mpJRW5S9fO03wcD8Sna9xyZacJfO
 UTnpfUArs9FF1pB3VIr95WwlVoptBOuKLTCNuzoBTW6jQt0sg0uPDAi2dDzf+21t/UuF7I3z
 KSeVyHuOfofonYD85FkQJN8lsbh5xWvsASbgD8bmfI87gEbt0wq2ND5yuX+lJK7FX4lMO6gR
 ZQ75g4KWDprOO/w6ebRxDjrH0lG1qHBiZd0hcPo2wkeYwb1sqZUjQjujlDhcvnZfpDGR4yLz
 5WG+pdciQhl6LNl7lctNhS8Uct17HNdfN7QvAumYw5sUuJ+POIlCws/aVbA5+DpmIfzPx5Ak
 UHxthNIyqZ9O6UHrVg7SaF3rvqrXtjtnu7eZ3cIsfuuHrXBTWDsVwub2nm1ddZZoC530BraS
 d7Y7eyKs7T4mGwpsi3Pd33Je5aC/rDeF44gXRv3UnKtjq2PPjaG/KPG0fLBGvhx0ARBrZLsd
 5CTDjwFA4bo+pD13cVhTfim3dYUnX1UDmqoCISOpzg3S4+QLv1bfbIsZ3KDQQR7y/RSGzcLE
 z164aDfuSvl+6Myb5qQy1HUQ0hOj5Qh+CzF3CMEPmU1v9Qah1ThC8+KkH/HHjPPulLn7aMaK
 Z8t6h7uaAYnGzjMEXZLIEhYJKwYBBAHaRw8BAQdAGdRDglTydmxI03SYiVg95SoLOKT5zZW1
 7Kpt/5zcvt3CwhsEGAEIACAWIQQk1ibyU76eh+bOW/SP9LjScWdVJwUCXZLIEgIbAgCvCRCP
 9LjScWdVJ40gBBkWCAAdFiEEbinX+DPdhovb6oob3uarTi9/eqYFAl2SyBIAIQkQ3uarTi9/
 eqYWIQRuKdf4M92Gi9vqihve5qtOL396pnZGAP0c3VRaj3RBEOUGKxHzcu17ZUnIoJLjpHdk
 NfBnWU9+UgD/bwTxE56Wd8kQZ2e2UTy4BM8907FsJgAQLL4tD2YZggwWIQQk1ibyU76eh+bO
 W/SP9LjScWdVJ5CaD/0YQyfUzjpR1GnCSkbaLYTEUsyaHuWPI/uSpKTtcbttpYv+QmYsIwD9
 8CeH3zwY0Xl/1fE9Hy59z6Vxv9YVapLx0nPDOA1zDVNq2MnutxHb8t+Imjz4ERCxysqtfYrv
 gao3E/h0c8SEeh+bh5MkjwmU8CwZ3doWyiVdULKESe7/Gs5OuhFzaDVPCpWdsKdCAGyUuP/+
 qRWwKGVpWP0Rrt6MTK24Ibeu3xEZO8c3XOEXH5d9nf6YRqBEIizAecoCr00E9c+6BlRS0AqR
 OQC3/Mm7rWtco3+WOridqVXkko9AcZ8AiM5nu0F8AqYGKg0y7vkL2LOP8us85L0p57MqIR1u
 gDnITlTY0x4RYRWJ9+k7led5WsnWlyv84KNzbDqQExTm8itzeZYW9RvbTS63r/+FlcTa9Cz1
 5fW3Qm0BsyECvpAD3IPLvX9jDIR0IkF/BQI4T98LQAkYX1M/UWkMpMYsL8tLObiNOWUl4ahb
 PYi5Yd8zVNYuidXHcwPAUXqGt3Cs+FIhihH30/Oe4jL0/2ZoEnWGOexIFVFpue0jdqJNiIvA
 F5Wpx+UiT5G8CWYYge5DtHI3m5qAP9UgPuck3N8xCihbsXKX4l8bdHfziaJuowief7igeQs/
 WyY9FnZb0tl29dSa7PdDKFWu+B+ZnuIzsO5vWMoN6hMThTl1DxS+jc7ATQRb/8z6AQgAvSkg
 5w7dVCSbpP6nXc+i8OBz59aq8kuL3YpxT9RXE/y45IFUVuSc2kuUj683rEEgyD7XCf4QKzOw
 +XgnJcKFQiACpYAowhF/XNkMPQFspPNM1ChnIL5KWJdTp0DhW+WBeCnyCQ2pzeCzQlS/qfs3
 dMLzzm9qCDrrDh/aEegMMZFO+reIgPZnInAcbHj3xUhz8p2dkExRMTnLry8XXkiMu9WpchHy
 XXWYxXbMnHkSRuT00lUfZAkYpMP7La2UudC/Uw9WqGuAQzTqhvE1kSQe0e11Uc+PqceLRHA2
 bq/wz0cGriUrcCrnkzRmzYLoGXQHqRuZazMZn2/pSIMZdDxLbwARAQABwsGNBBgBCAAgFiEE
 JNYm8lO+nofmzlv0j/S40nFnVScFAlv/zPoCGwwAIQkQj/S40nFnVScWIQQk1ibyU76eh+bO
 W/SP9LjScWdVJ/g6EACFYk+OBS7pV9KZXncBQYjKqk7Kc+9JoygYnOE2wN41QN9Xl0Rk3wri
 qO7PYJM28YjK3gMT8glu1qy+Ll1bjBYWXzlsXrF4szSqkJpm1cCxTmDOne5Pu6376dM9hb4K
 l9giUinI4jNUCbDutlt+Cwh3YuPuDXBAKO8YfDX2arzn/CISJlk0d4lDca4Cv+4yiJpEGd/r
 BVx2lRMUxeWQTz+1gc9ZtbRgpwoXAne4iw3FlR7pyg3NicvR30YrZ+QOiop8psWM2Fb1PKB9
 4vZCGT3j2MwZC50VLfOXC833DBVoLSIoL8PfTcOJOcHRYU9PwKW0wBlJtDVYRZ/CrGFjbp2L
 eT2mP5fcF86YMv0YGWdFNKDCOqOrOkZVmxai65N9d31k8/O9h1QGuVMqCiOTULy/h+FKpv5q
 t35tlzA2nxPOX8Qj3KDDqVgQBMYJRghZyj5+N6EKAbUVa9Zq8xT6Ms2zz/y7CPW74G1GlYWP
 i6D9VoMMi6ICko/CXUZ77OgLtMsy3JtzTRbn/wRySOY2AsMgg0Sw6yJ0wfrVk6XAMoLGjaVt
 X4iPTvwocEhjvrO4eXCicRBocsIB2qZaIj3mlhk2u4AkSpkKm9cN0KWYFUxlENF4/NKWMK+g
 fGfsCsS3cXXiZpufZFGr+GoHwiELqfLEAQ9AhlrHGCKcgVgTOI6NHg==
Message-ID: <c652094b-83bf-1f4c-4b9a-ff911864125f@linaro.org>
Date:   Fri, 13 Mar 2020 14:17:55 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.4.1
MIME-Version: 1.0
In-Reply-To: <CAKfTPtDUmqYB1i7UcYXxcNjnQOoGufsB9do-9NxTMeWdJAfP2w@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 13/03/2020 14:15, Vincent Guittot wrote:
> On Fri, 13 Mar 2020 at 13:15, Daniel Lezcano <daniel.lezcano@linaro.org> wrote:

[ ... ]

>>>>>> +
>>>>>> +       if (idle_state)
>>>>>> +               idle_set_break_even(rq, ktime_get_ns() +
>>>>>
>>>>> What worries me a bit is that it adds one ktime_get call each time a
>>>>> cpu enters idle
>>>>
>>>> Right, we can improve this in the future by folding the local_clock() in
>>>> cpuidle when entering idle with this ktime_get.
>>>
>>> Using local_clock() would be more latency friendly
>>
>> Unfortunately we are comparing the deadline across CPUs, so the
>> local_clock() can not be used here.
>>
>> But if we have one ktime_get() instead of a local_clock() + ktime_get(),
>> that should be fine, no?
> 
> Can't this computation of break_even be done in cpuidle framework
> while computing other statistics for selecting the idle state instead
> ? cpuidle already uses ktime_get for next hrtimer as an example.
> So cpuidle compute break_even and make it available to scheduler like
> exit_latency. And I can imagine that system wide time value will also
> be needed when looking at next wakeup event of cluster/group of CPUs

Ok, so you suggest to revisit and consolidate the whole time capture in
cpuidle? I think that makes sense.


-- 
 <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