Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 29 Aug 2005 11:29:16 -0700
From:      "Kevin Oberman" <oberman@es.net>
To:        Bruno Ducrot <bruno@poupinou.org>
Cc:        acpi@freebsd.org, Hajimu UMEMOTO <ume@FreeBSD.org>
Subject:   Re: cvs commit: src/usr.sbin/powerd powerd.c 
Message-ID:  <20050829182916.F26405D07@ptavv.es.net>
In-Reply-To: Your message of "Mon, 29 Aug 2005 09:38:21 %2B0200." <20050829073821.GI7749@poupinou.org> 

next in thread | previous in thread | raw e-mail | index | archive | help
> Date: Mon, 29 Aug 2005 09:38:21 +0200
> From: Bruno Ducrot <bruno@poupinou.org>
> Sender: owner-cvs-all@freebsd.org
> 
> On Sun, Aug 28, 2005 at 10:36:38PM -0700, Nate Lawson wrote:
> > Another mitigating factor is a patch I hope to commit soon that removes 
> > levels that aren't useful.  The general idea is the same as a recent 
> > email from Tijl Coosemans but my approach is different.
> 
> I'm pretty sure it's incorrect to add p4tcc and acpi_throttle for power
> saving purpose.  I plan to add some flags in order to use only relevant
> frequencies to this end, but IMHO that should be done at low-level
> drivers.  On the other hand, it is usefull to keep the existing sysctl
> freqs, but for cooling purpose only.

Bruno,

As far as I know, Nate eliminated throttling when TCC was present a
while ago. I certainly don't see any combination of them on my P4m
system. 

The non-useful speeds Tjil eliminated were those where ICHSS and TCC (or
any of several other combos) would select:
1. The higher speed CPU clock when the same performance level could be
   obtained with a lower CPU clock and a higher throttle or TCC setting
2. A lower "performance level" actually results in higher power usage and
   a lower performance. (Never use a higher CPU clock speed if a lower
   CPU speed is available. the higher CPU speed NEVER seems to win or
   even come close.

These are the cases Tjil's patches were aimed at (and hit the target for
my case).
-- 
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman@es.net			Phone: +1 510 486-8634



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050829182916.F26405D07>