Date: Wed, 29 Jan 2003 19:30:55 -0800 From: Peter Wemm <peter@wemm.org> To: "Matthew D. Fuller" <fullermd@over-yonder.net> Cc: cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/dev/acpica acpi_cpu.c Message-ID: <20030130033055.D17962A89E@canning.wemm.org> In-Reply-To: <20030130031234.GB1224@over-yonder.net>
next in thread | previous in thread | raw e-mail | index | archive | help
"Matthew D. Fuller" wrote: > On Wed, Jan 29, 2003 at 05:27:06PM -0800 I heard the voice of > Peter Wemm, and lo! it spake thus: > > > > This is a bit old, but it can cause a slight slowdown, because it may take > > a little longer for a cpu to notice a job sitting on the run queue that it > > could have been running. Once it does a halt, it will stop checking until > > the next interrupt (sometimes a clock, sometimes disk/net/etc). This > > Did I miss something somewhere? Does this now work correctly on SMP? Or > is that only with ACPI? With HTT enabled, it is absolutely essential. Having the idle loop on a logical cpu sitting in a hard spin and stealing execution resources from the other logical cpus is not an option. It just so happens that the cost of doing it by default (slight slowdown) has some benefits to counteract the slight cost.. ie: significantly lower temperatures and power consumption. I've been thinking about adding some pretty basic tweaks to avoid the slowdown. It's easier to do on 4.x than 5.x. Cheers, -Peter -- Peter Wemm - peter@wemm.org; peter@FreeBSD.org; peter@yahoo-inc.com "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030130033055.D17962A89E>