From owner-freebsd-hackers@FreeBSD.ORG Tue Oct 24 09:12:18 2006 Return-Path: X-Original-To: freebsd-hackers@freebsd.org Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0D28116A40F; Tue, 24 Oct 2006 09:12:18 +0000 (UTC) (envelope-from kpielorz@tdx.co.uk) Received: from caladan.tdx.co.uk (caladan.tdx.co.uk [62.13.130.230]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4970A43D49; Tue, 24 Oct 2006 09:12:16 +0000 (GMT) (envelope-from kpielorz@tdx.co.uk) Received: from quadro.tdx.co.uk (rainbow.tdx.co.uk [62.13.130.232] (may be forged)) by caladan.tdx.co.uk (8.13.6/8.13.6/Kp) with ESMTP id k9O9CFKa049662; Tue, 24 Oct 2006 10:12:15 +0100 (BST) Date: Tue, 24 Oct 2006 10:13:05 +0100 From: Karl Pielorz To: Robert Watson Message-ID: In-Reply-To: <20061020134552.H38852@fledge.watson.org> References: <20061020110151.GU55428@deviant.kiev.zoral.com.ua> <82C14B40C9061504218D7585@Unsupported> <20061020134552.H38852@fledge.watson.org> X-Mailer: Mulberry/4.0.4 (Win32) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Cc: Kostik Belousov , freebsd-hackers@freebsd.org Subject: Re: 6.1-STABLE hangs, ddb shows 'acpi_timer_read'? X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 24 Oct 2006 09:12:18 -0000 --On 20 October 2006 13:47 +0100 Robert Watson wrote: >> I'm just a little hesitant to put it all in, and end up with a machine >> that's 80% slower :( > > Depends a lot on your workload. WITNESS used to really, really slow > things down for kernel lock intensive workloads (VFS especially); now it > just really slows things down. INVARIANTS overhead is generally > measurable, but for most workloads it is likely <20%. The place > INVARIANTS hits you is when the kernel is allocating and freeing lots of > memory, during which INVARIANTS will be scrubbing and checking for use > after free, etc. It's worth running with WITNESS when debugging > deadlocks if possible, because it is, essentially, a deadlock debugging > tool. A moderate number of people run with INVARIANTS in production, but > you generally don't want to do that with WITNESS. Well this is happening on an almost daily basis for me, so I threw everything in. I'd rather have lousy performance for a couple of days - and get all the info needed to fix the problem, than fsck a terrabyte, and mop up after a hung machine daily ;) Thanks for the replies, crossing fingers now :) -Karl