Date: Sat, 08 Nov 2008 07:19:20 +0200 From: Alexander Motin <mav@FreeBSD.org> To: Ian Smith <smithi@nimnet.asn.au> Cc: Sam Leffler <sam@freebsd.org>, "Alexandre \"Sunny\" Kovalenko" <gaijin.k@gmail.com>, freebsd-mobile@freebsd.org Subject: Re: RFC: powerd algorithms enhancements Message-ID: <49152158.9090207@FreeBSD.org> In-Reply-To: <20081108012859.Y70117@sola.nimnet.asn.au> References: <491208D3.2050901@FreeBSD.org> <20081107033524.A70117@sola.nimnet.asn.au> <1226065673.1210.9.camel@RabbitsDen> <20081108012859.Y70117@sola.nimnet.asn.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Ian Smith wrote: > We're now seeing cpus that can vary freq, with absolute and relative > cpufreq drivers enabled, in ratios up to 32:1 or so, so the advice, > apart from 'disable powerd' :), seems to be to at least try setting > cpufreq.lowest to some reasonable speed for workload, maybe 300MHz? I surely should not be the default, but it is reasonable if systems should have some guarantied minimal performance. PS: At any modern SMP/HTT system, even if scheduler is unable to manage this IRQ situation, powerd running on different CPU will rise clock to required level just in second. It is hard to lock-out all CPUs same time. It's surely not solution, but still... -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49152158.9090207>