linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Michael Ellerman <patch-notifications@ellerman.id.au>
To: "Gautham R. Shenoy" <ego@linux.vnet.ibm.com>,
	"Rafael J. Wysocki" <rjw@rjwysocki.net>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Stewart Smith <stewart@linux.vnet.ibm.com>,
	Michael Neuling <mikey@neuling.org>,
	Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
	Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>,
	Akshay Adiga <akshay.adiga@linux.vnet.ibm.com>,
	Nicholas Piggin <npiggin@gmail.com>
Cc: "Gautham R. Shenoy" <ego@linux.vnet.ibm.com>,
	linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org,
	linux-pm@vger.kernel.org
Subject: Re: cpuidle:powernv: Make the snooze timeout dynamic.
Date: Wed,  6 Jun 2018 01:19:32 +1000 (AEST)	[thread overview]
Message-ID: <410b7m522wz9s47@ozlabs.org> (raw)
In-Reply-To: <1527768909-32637-1-git-send-email-ego@linux.vnet.ibm.com>

On Thu, 2018-05-31 at 12:15:09 UTC, "Gautham R. Shenoy" wrote:
> From: "Gautham R. Shenoy" <ego@linux.vnet.ibm.com>
> 
> The commit 78eaa10f027c ("cpuidle: powernv/pseries: Auto-promotion of
> snooze to deeper idle state") introduced a timeout for the snooze idle
> state so that it could be eventually be promoted to a deeper idle
> state. The snooze timeout value is static and set to the target
> residency of the next idle state, which would train the cpuidle
> governor to pick the next idle state eventually.
> 
> The unfortunate side-effect of this is that if the next idle state(s)
> is disabled, the CPU will forever remain in snooze, despite the fact
> that the system is completely idle, and other deeper idle states are
> available.
> 
> This patch fixes the issue by dynamically setting the snooze timeout
> to the target residency of the next enabled state on the device.
> 
> Before Patch
> Reviewed-by: Balbir Singh <bsingharora@gmail.com>
> 
> ==================
> POWER8 : Only nap disabled.
>  $cpupower monitor sleep 30
> sleep took 30.01297 seconds and exited with status 0
>               |Idle_Stats
> PKG |CORE|CPU | snoo | Nap  | Fast
>    0|   8|   0| 96.41|  0.00|  0.00
>    0|   8|   1| 96.43|  0.00|  0.00
>    0|   8|   2| 96.47|  0.00|  0.00
>    0|   8|   3| 96.35|  0.00|  0.00
>    0|   8|   4| 96.37|  0.00|  0.00
>    0|   8|   5| 96.37|  0.00|  0.00
>    0|   8|   6| 96.47|  0.00|  0.00
>    0|   8|   7| 96.47|  0.00|  0.00
> 
> POWER9: Shallow states (stop0lite, stop1lite, stop2lite, stop0, stop1,
> stop2) disabled:
> $cpupower monitor sleep 30
> sleep took 30.05033 seconds and exited with status 0
>               |Idle_Stats
> PKG |CORE|CPU | snoo | stop | stop | stop | stop | stop | stop | stop | stop
>    0|  16|   0| 89.79|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00
>    0|  16|   1| 90.12|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00
>    0|  16|   2| 90.21|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00
>    0|  16|   3| 90.29|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00
> 
> After Patch
> ======================
> POWER8 : Only nap disabled.
> $ cpupower monitor sleep 30
> sleep took 30.01200 seconds and exited with status 0
>               |Idle_Stats
> PKG |CORE|CPU | snoo | Nap  | Fast
>    0|   8|   0| 16.58|  0.00| 77.21
>    0|   8|   1| 18.42|  0.00| 75.38
>    0|   8|   2|  4.70|  0.00| 94.09
>    0|   8|   3| 17.06|  0.00| 81.73
>    0|   8|   4|  3.06|  0.00| 95.73
>    0|   8|   5|  7.00|  0.00| 96.80
>    0|   8|   6|  1.00|  0.00| 98.79
>    0|   8|   7|  5.62|  0.00| 94.17
> 
> POWER9: Shallow states (stop0lite, stop1lite, stop2lite, stop0, stop1,
> stop2) disabled:
> 
> $cpupower monitor sleep 30
> sleep took 30.02110 seconds and exited with status 0
>               |Idle_Stats
> PKG |CORE|CPU | snoo | stop | stop | stop | stop | stop | stop | stop | stop
>    0|   0|   0|  0.69|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  9.39| 89.70
>    0|   0|   1|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.05| 93.21
>    0|   0|   2|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00| 89.93
>    0|   0|   3|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00|  0.00| 93.26
> 
> Signed-off-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>

Applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/0a4ec6aa035a52c422eceb2ed51ed8

cheers

      parent reply	other threads:[~2018-06-05 15:19 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-31 12:15 [PATCH] cpuidle:powernv: Make the snooze timeout dynamic Gautham R. Shenoy
2018-05-31 14:51 ` Balbir Singh
2018-06-01  4:54   ` Gautham R Shenoy
2018-06-02  0:16     ` Balbir Singh
2018-06-04 11:27 ` Michael Ellerman
2018-06-05  3:47   ` Stewart Smith
2018-06-05  8:45   ` Gautham R Shenoy
2018-06-05 12:18     ` Michael Ellerman
2018-06-05 15:19 ` Michael Ellerman [this message]

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=410b7m522wz9s47@ozlabs.org \
    --to=patch-notifications@ellerman.id.au \
    --cc=akshay.adiga@linux.vnet.ibm.com \
    --cc=daniel.lezcano@linaro.org \
    --cc=ego@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mikey@neuling.org \
    --cc=npiggin@gmail.com \
    --cc=rjw@rjwysocki.net \
    --cc=shilpa.bhat@linux.vnet.ibm.com \
    --cc=stewart@linux.vnet.ibm.com \
    --cc=svaidy@linux.vnet.ibm.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).