From owner-freebsd-alpha Sat Jan 27 12:18:49 2001 Delivered-To: freebsd-alpha@freebsd.org Received: from server.baldwin.cx (server.geekhouse.net [64.81.6.52]) by hub.freebsd.org (Postfix) with ESMTP id 8533C37B401 for ; Sat, 27 Jan 2001 12:18:20 -0800 (PST) Received: from john.baldwin.cx (root@john.baldwin.cx [192.168.1.18]) by server.baldwin.cx (8.9.3/8.9.3) with ESMTP id MAA92137; Sat, 27 Jan 2001 12:19:35 -0800 (PST) (envelope-from john@baldwin.cx) Received: (from john@localhost) by john.baldwin.cx (8.9.3/8.9.3) id MAA44645; Sat, 27 Jan 2001 12:19:20 -0800 (PST) (envelope-from john) Message-Id: <200101272019.MAA44645@john.baldwin.cx> X-Mailer: XFMail 1.4.0 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20010127181650.A6504@cicely5.cicely.de> Date: Sat, 27 Jan 2001 12:19:20 -0800 (PST) From: John Baldwin To: Bernd Walter Subject: Re: -current failing on PC164 Cc: freebsd-alpha@FreeBSD.ORG, Andrew Gallatin Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On 27-Jan-01 Bernd Walter wrote: > On Wed, Jan 24, 2001 at 06:45:29AM -0800, John Baldwin wrote: >> >> On 24-Jan-01 Andrew Gallatin wrote: >> > >> > Bernd's problem was an unaligned access error, probably in the >> > witness >> > or mutex debug code. This (probably) has nothing to do with pmap >> > problems. >> >> No, they were the fault of my own stupidity. :-P I bothced some >> KASSERT()'s. *sigh* > > A newer source boots fine. > The only problem is that with the debugging options the system is > unbeleavable slow. > As an example the clock swi takes about 20% CPU... > > I added the following: > options KTR > options KTR_EXTEND > options KTR_ENTRIES=1024 > options KTR_COMPILE=0x3fffff > options KTR_MASK=0x201208 > options KTR_CPUMASK=0x3 >#options KTR_VERBOSE You probably want KTR_MASK not set to anything less you are actually debugging something specific. > options MUTEX_DEBUG > options WITNESS >#options WITNESS_DDB > options WITNESS_SKIPSPIN > > options DIAGNOSTIC > options INVARIANTS > options INVARIANT_SUPPORT You probably want to axe MUTEX_DEBUG (this is the big performance killer) and DIAGNOSTIC. > What are the current recomended options? > Hopefully I can remove some of them without loosing too much information > for the failure case. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.baldwin.cx/~john/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message