From owner-freebsd-questions@freebsd.org Sat Oct 17 12:40:05 2015 Return-Path: Delivered-To: freebsd-questions@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 4EA1CA1796B for ; Sat, 17 Oct 2015 12:40:05 +0000 (UTC) (envelope-from smithi@nimnet.asn.au) Received: from sola.nimnet.asn.au (paqi.nimnet.asn.au [115.70.110.159]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id C0D341571 for ; Sat, 17 Oct 2015 12:40:03 +0000 (UTC) (envelope-from smithi@nimnet.asn.au) Received: from localhost (localhost [127.0.0.1]) by sola.nimnet.asn.au (8.14.2/8.14.2) with ESMTP id t9HCdupU026881; Sat, 17 Oct 2015 23:39:56 +1100 (EST) (envelope-from smithi@nimnet.asn.au) Date: Sat, 17 Oct 2015 23:39:56 +1100 (EST) From: Ian Smith To: john.haraden@yahoo.com cc: freebsd-questions@freebsd.org Subject: Re: est In-Reply-To: <849B8A68-B2CA-4BDD-80D9-B367943F58F5@yahoo.com> Message-ID: <20151017232655.D15983@sola.nimnet.asn.au> References: <20151016013912.B15983@sola.nimnet.asn.au> <849B8A68-B2CA-4BDD-80D9-B367943F58F5@yahoo.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 17 Oct 2015 12:40:05 -0000 On Thu, 15 Oct 2015 15:02:45 -0700, john.haraden@yahoo.com wrote: > I must do some more research on power management. This whole issue > is more complicated than I previously believed. However, I can > disable est by adding hint.est.0.disable="1" in /boot/loader.conf. Ah, right. Well, that's a good deal easier than building a new kernel. Can you see or measure ANY difference in performance with est disabled? Could you share the result of `sysctl dev.cpu.0` with est disabled? I'm just (perennially) curious as to what comes up without it. No need to see it for your other 23 CPUs, they should all be the same .. > There is also some speculation about increasing performance by > appropriately combining est and powerd. Can you point to any links for such speculation? Running powerd implies having est (or powernow on AMD) working anyway. > My goal is to maximize the performance of a server isolated from the > net for scientific research. Fair enough. May you get all 24 cores up to 100% then! :) cheers, Ian