Date: Wed, 04 Jan 2012 20:45:44 +0200 From: Alexander Motin <mav@FreeBSD.org> To: "Julian H. Stacey" <jhs@berklix.com> Cc: mobile@freebsd.org Subject: Re: powerd to use sysctl to import temps to drop freq to avoid heat crash Message-ID: <4F049E58.5040404@FreeBSD.org> In-Reply-To: <201201041822.q04IMlQJ060929@fire.js.berklix.net> References: <201201041822.q04IMlQJ060929@fire.js.berklix.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 01/04/12 20:22, Julian H. Stacey wrote:
> Question:
> In your
> dev.cpu.0.freq_levels: 2400/35000 2000/28000 1600/22000 1200/16000 800/14000
> 1st numbers are frequency, what are 2nd numbers after / ?
Relative power consumption under full CPU load.
> Presumably not voltages, as mine have such a wide span :
> dev.cpu.0.freq_levels: 1592/100000 1393/87500 1194/75000 995/62500
> 796/35457 696/31024 597/26592 497/22160 398/17728 298/13296
> 199/8864 99/4432
> (We should send-pr longer text to be produced by
> sysctl -d dev.cpu.0.freq_levels
> dev.cpu.0.freq_levels: CPU frequency levels
> )
For EIST and probably PowerNow! frequencies it is relative power levels.
But for frequencies created by throttling they are incorrect. If you
disable throttling, you should see only "real" value.
> Question
> "It is not recommended to set the system timer tick rate below 250 HZ and"
> Do you mean as shown by
> kern.clockrate: { hz = 1000, tick = 1000,
> Not kern.hz ?
Generally kern.hz, but for freqs below 1000Hz they are equal.
--
Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4F049E58.5040404>