From owner-cvs-all Fri Jan 31 10:55:42 2003 Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E43CD37B401; Fri, 31 Jan 2003 10:55:40 -0800 (PST) Received: from canning.wemm.org (canning.wemm.org [192.203.228.65]) by mx1.FreeBSD.org (Postfix) with ESMTP id 95A1343F43; Fri, 31 Jan 2003 10:55:40 -0800 (PST) (envelope-from peter@wemm.org) Received: from wemm.org (localhost [127.0.0.1]) by canning.wemm.org (Postfix) with ESMTP id 84A122A8A1; Fri, 31 Jan 2003 10:55:40 -0800 (PST) (envelope-from peter@wemm.org) X-Mailer: exmh version 2.5 07/13/2001 with nmh-1.0.4 To: Nate Lawson Cc: John Baldwin , cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/i386/i386 mp_machdep.c In-Reply-To: Date: Fri, 31 Jan 2003 10:55:40 -0800 From: Peter Wemm Message-Id: <20030131185540.84A122A8A1@canning.wemm.org> Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Nate Lawson wrote: > On Fri, 31 Jan 2003, John Baldwin wrote: > > On 30-Jan-2003 Matthew Dillon wrote: > > > Ouch. That's rather counter-intuitive since idle_hlt usually > > > degrades performance, but I see how it would improve performance on > > > an HTT box. > > > > Well, idle_hlt can only really hurt if you are doing compute bound stuff. > > You get at least 128 clock interrupts per second from the RTC that get > > IPI'd to all the other CPUs, so no CPU would errantly stay halted for more > > than 1/128th of a second with idle_hlt on anyways. As Peter has mentioned, > > the idle_hlt thing shows some rather impressive thermal benefits, and I > > imagine it can provide power savings of some sort. I would argue that > > being fully compute bound is not the common case and that we should probabl y > > default to having it on in general. > > Why not check CPU utilization and dynamically en/disable HLT? I was thinking of doing an idle cpu mask and adding an IPI to do a wakeup. That way we can just kill the problem and be done with it. Have you considered promisciously turning on the idle_hlt flag by the HTT detection code? ie: if booting with >1 logical cores, then turn it on at boot. Also, some folks measured a slight increase in performance by turning on idle_hlt, which presumably was due to less cache line thrashing. This will obviously have different effects on different cpu architectures (p3, p4, athlon etc). Anyway, I really like it, because I've been able to stop my 6-month cpu fan replacement cycle that has been as regular as clockwork for the last 5 years. PPro fans are getting harder to find these days. :-] The last set have lasted just over a year and have no signs of wearing out. (I bought a small box of them, figuring I was using them so fast and they were hard to find....) 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