From owner-freebsd-current@FreeBSD.ORG Sun Dec 20 06:33:32 2009 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A186C1065672; Sun, 20 Dec 2009 06:33:32 +0000 (UTC) (envelope-from oberman@es.net) Received: from mailgw.es.net (mail2.es.net [IPv6:2001:400:107:1::2]) by mx1.freebsd.org (Postfix) with ESMTP id 844558FC26; Sun, 20 Dec 2009 06:33:32 +0000 (UTC) Received: from ptavv.es.net (ptavv.es.net [IPv6:2001:400:910::29]) by mailgw.es.net (8.14.3/8.14.3) with ESMTP id nBK6XIjX002003 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Sat, 19 Dec 2009 22:33:20 -0800 Received: from ptavv.es.net (localhost [127.0.0.1]) by ptavv.es.net (Tachyon Server) with ESMTP id 49A8C1CC0B; Sat, 19 Dec 2009 22:33:18 -0800 (PST) To: Doug Barton In-reply-to: Your message of "Sat, 19 Dec 2009 19:20:19 PST." <4B2D97F3.703@FreeBSD.org> Date: Sat, 19 Dec 2009 22:33:18 -0800 From: "Kevin Oberman" Message-Id: <20091220063318.49A8C1CC0B@ptavv.es.net> X-Proofpoint-Virus-Version: vendor=fsecure engine=1.12.8161:2.4.5, 1.2.40, 4.0.166 definitions=2009-12-20_02:2009-12-12, 2009-12-20, 2009-12-19 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=5.0.0-0908210000 definitions=main-0912190246 Cc: Paul G Webster , freebsd-current@freebsd.org, Glen Barber Subject: Re: Systems running hot? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 20 Dec 2009 06:33:32 -0000 > Date: Sat, 19 Dec 2009 19:20:19 -0800 > From: Doug Barton > Sender: owner-freebsd-current@freebsd.org > > Glen Barber wrote: > > On Sat, Dec 19, 2009 at 9:27 PM, Doug Barton wrote: > >> Paul G Webster wrote: > >>> tried running powerd on it? > >> I was not running powerd, but I just enabled it with: > >> performance_cx_lowest="HIGH" > >> performance_cpu_freq="HIGH" > >> economy_cx_lowest="HIGH" > >> economy_cpu_freq="HIGH" > >> > >> If I'm reading the rc.conf man page correctly, that will give me the > >> benefit of good performance while at the same time dropping the cpu > >> frequency and power usage when idle? > >> > >> Any other suggestions are welcome. > >> > > > > You can also set settings dependent on the power source: > > > > powerd_flags="-a max -b adaptive" > > I set it to "-a adaptive -b adaptive" (I'm on AC power most of the > time) and the good news is that the temp dropped by about 10 degrees C > and the fan went down to the low speed for the first time today. > > My concern about using this feature is that the last time I used it I > ended up with problems with my (build into the motherboard) video > card. I was never sure at the time whether or not the problem with the > video card was a result of powerd, or of the heat problems I was > having at the time (back in 8-current when everyone agreed that there > was a heat problem). I'd like to think that powerd would not fry a > graphics card, and I don't think that it's even likely, but the system > is out of warranty now and I can't afford to foul it up. > > The other negative to running powerd is that it causes my display to > flash occasionally (which of course feeds my "powerd is destroying my > graphics card" paranoia). It's not life threatening, but it is annoying. > > Thanks to everyone for their suggestions, we'll see how this goes ... > > > Doug powerd has undergone major revision lately (I think in 7.2 and 8.0) to greatly enhance responsiveness. I have moved to running adaptive at all times and no longer notice slowness when I kick off some CPU intensive command. Beyond that, read Alexander Motin's power management article at http://wiki.freebsd.org/TuningPowerConsumption when it's available. (It seems that the wiki is down at the moment,) Note the comments on throttling/TCC and deep sleep states. He and I both found TCC and throttling not worthwhile and possibly causing very unpleasant side-effects. C (sleep) states are a really big win, but cause problems on a few systems. mav did a great job on this, duplicating my work back in V5 days and taking it a lot farther than I did. Anyone who wants to save power and heat really should read it. I also found the original article at http://lists.freebsd.org/pipermail/freebsd-current/2009-May/006436.html. I'm not sure it's identical to the wiki entry, but it is very close and is currently available. -- 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 Key fingerprint:059B 2DDF 031C 9BA3 14A4 EADA 927D EBB3 987B 3751