From owner-freebsd-current Sat Nov 28 08:14:51 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id IAA22700 for freebsd-current-outgoing; Sat, 28 Nov 1998 08:14:51 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [195.188.177.4]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id IAA22692 for ; Sat, 28 Nov 1998 08:14:49 -0800 (PST) (envelope-from kpielorz@tdx.co.uk) Received: from tdx.co.uk (lorca-tx.tdx.co.uk [195.188.177.242]) by caladan.tdx.co.uk (8.9.1/8.9.1) with ESMTP id QAA89994; Sat, 28 Nov 1998 16:14:37 GMT Message-ID: <3660216C.89446EB2@tdx.co.uk> Date: Sat, 28 Nov 1998 16:14:36 +0000 From: Karl Pielorz Organization: TDX - The Digital eXchange X-Mailer: Mozilla 4.5 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: Bill Woods CC: current@FreeBSD.ORG Subject: Re: Weird CPU useage References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Bill Woods wrote: > > I am running 3.0 -current, 2 day old code, on a dual PP200 with 128meg > Ram. The system is NOT overclocked. (This is inmportant) > > The problem is this, when I run rc5 clients and do a top, I see them > both running at >120% cpu useage each...that is too weird. I run two > sessions of the client, on for each cpu. Now, just to test this, I wrote > a small infinite loop progra (a cpu sucker) just to see how much CPU > useage was done, again, a top produced the same thing >120% per cpu. Hmmm... I have a dual PP200 system here, I'll see if I get a similar result (not obviously noticed this to be honest... ;-) > Now this would be all fine and dandy, (faster is better right?), except > for ythe heat produced. This weird situation produces a nice ammount of > heat. I have a fan in front, one in the back of the case, and each cpu has > a fan and heat sink on it. > > Any ideas would be appreciated. I'm sorry? - Are you saying that the CPU running RC5 or a 'tight-loop' shouldn't produce that much heat? RC5 runs as 'nice' - i.e. it will get the CPU if nothing else wants/needs it - and as such, on an idle system it will use all the CPU available for it's nasty DES/RC5 code cracking routines - this should generate the same amount of heat as a system running a deliberate 'tight' / CPU sucker (in fact it might generate more heat for the RC5 client - as it pushes the CPU more than a simple tight loop)... FreeBSD only _ever_ runs 'cooler' when it's doing _nothing_ as the HALT instruction get's called a lot by the idle loop AFAIK, this is why an idle FreeBSD box will not heat your cpu as much as an Idle 'other os' box (unless the 'other' OS does a similar thing)... It's also the reason why going into MS DOS edit and dropping down the 'File' menu causes the CPU to get nice and warm... Apparenlty dropping down the File menu involves some nice stressfull MS code ;-) -Kp ps. Please don't cross post - it almost always annoys people ;-) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message