From owner-freebsd-hackers Wed Mar 17 11:23:16 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from ceia.nordier.com (m1-38-dbn.dial-up.net [196.34.155.38]) by hub.freebsd.org (Postfix) with ESMTP id 9E92A14BCD for ; Wed, 17 Mar 1999 11:22:59 -0800 (PST) (envelope-from rnordier@nordier.com) Received: (from rnordier@localhost) by ceia.nordier.com (8.8.7/8.6.12) id VAA00398; Wed, 17 Mar 1999 21:19:35 +0200 (SAT) From: Robert Nordier Message-Id: <199903171919.VAA00398@ceia.nordier.com> Subject: Re: KVA size changes in 3.1-stable In-Reply-To: from Chuck Robey at "Mar 16, 99 11:01:39 pm" To: chuckr@mat.net (Chuck Robey) Date: Wed, 17 Mar 1999 21:19:33 +0200 (SAT) Cc: crossd@cs.rpi.edu, rnordier@nordier.com, dillon@apollo.backplane.com, freebsd-hackers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Chuck Robey wrote: > On Tue, 16 Mar 1999, David E. Cross wrote: > > > AUGH! > > > > I compiled the /boot/loader from the 4.0 tree CVS-ed a couple of hours ago > > (we have our own copy which periodically syncs with cvsup.freebsd.org). I > > tell it to boot *any* kernel, and it just *hangs* (caps lock/scroll lock > > still works, but nothing else.)... Normally a /boot/loader startup looks > > something like: > > /-\|kernel /-\|/... etc you get the idea. > > > > this one looks like: > > > > /-\|/ (and hangs here) > > How long since you last used disklabel to update your bootblocks? Old > ones will do that (they did for me, using 2.2.6 bootblocks a while > back). I'm really skeptical that the loader problems you experienced had anything directly to do with the 2.2.6 bootblocks. I have a machine here, still running 2.2.5-RELEASE, and regularly test loader on it. To double-check, I just upgraded the bootblocks to 2.2.6, and loader works fine. Because loader runs under BTX, the protected mode environment passed by the bootblocks is completely torn down, and the machine is brought back to real address mode before BTX starts. So I just can't relate to this kind of cause producing this kind of effect. To answer David's original question, I don't know of anything likely to cause loader to behave this way. I think Daniel's suggestion, made elsewhere, of trying /boot/loader.old would help in establishing whether the problem is in the new code, or whether something in the environment has changed and is causing problems. However, I'll try the latest loader, once today's make world finishes here. -- Robert Nordier To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message