Date: Sun, 28 Jun 2009 13:58:00 -0700 From: "Kevin Oberman" <oberman@es.net> To: Alexander Motin <mav@FreeBSD.org> Cc: freebsd-acpi@freebsd.org Subject: Re: acpi/cpu scaling probs? Dell D820 FreeBSD current Message-ID: <20090628205800.DF09A1CC09@ptavv.es.net> In-Reply-To: Your message of "Sun, 28 Jun 2009 23:42:48 %2B0300." <4A47D5C8.4090701@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> Date: Sun, 28 Jun 2009 23:42:48 +0300 > From: Alexander Motin <mav@FreeBSD.org> > > Kevin Oberman wrote: > >> Date: Sun, 28 Jun 2009 21:55:48 +0300 > >> From: Alexander Motin <mav@FreeBSD.org> > >> Sender: owner-freebsd-acpi@freebsd.org > >> > >> Ron Freidel wrote: > >>> I hope I am posting to the correct list... > >>> > >>> I have updated to FreeBSD current to try out acpi sleep, which works great > >>> by the way, and the improvements to wifi and the addition of sleep are > >>> enough to keep current on the laptop. > >>> > >>> Here's the problem, the cpu is maxed out, no scaling at all. > >>> > >>> Here's the output of powerd -v > >>> <snip> > >>> load 108%, current freq 2000 MHz ( 0), wanted freq 4000 MHz > >>> load 109%, current freq 2000 MHz ( 0), wanted freq 4000 MHz > >>> load 102%, current freq 2000 MHz ( 0), wanted freq 4000 MHz > >>> load 124%, current freq 2000 MHz ( 0), wanted freq 4000 MHz > >>> load 108%, current freq 2000 MHz ( 0), wanted freq 4000 MHz > >>> <snip> > >> ^^^ > >> This is the reason. powerd sees that one of your CPUs is constantly > >> busy. You should investigate why and what that CPU does. > > > > Agreed, but that still does not explain why it keeps trying to set the > > clock to 4 GHz on a 2 GHz system. Not too surprising that it does not > > work. :-) > > It is not a bug, but feature. It is specific of the "hiadaptive" mode. > It means that CPU is too busy now and should not drop frequency > immediately after load drops to not loose system interactivity and > performance. Thanks! This makes excellent sense. I think I need to switch mine to hiadaptive, at least on AC. Now, it Ron can get his CPU under control, I think things will be normal on his system. -- 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
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090628205800.DF09A1CC09>