Date: Tue, 7 Sep 2004 19:41:19 -0700 From: Marcel Moolenaar <marcel@xcllnt.net> To: Andreas Kohn <andreas.kohn@gmx.net> Cc: freebsd-alpha@freebsd.org Subject: Re: 5.3-BETA3 and tga... Message-ID: <20040908024119.GA7290@alpha.pn.xcllnt.net> In-Reply-To: <1094592557.845.7.camel@klamath.ankon.de.eu.org> References: <1094587559.845.4.camel@klamath.ankon.de.eu.org> <Pine.LNX.4.33.0409072211070.24958-100000@servww6.ww.uni-erlangen.de> <20040907211357.GA96722@ns1.xcllnt.net> <1094592557.845.7.camel@klamath.ankon.de.eu.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Sep 07, 2004 at 11:29:17PM +0200, Andreas Kohn wrote: > > > > Some background: The machine check is caused by the load prior to > > the and instruction and it's this load that used to cause a MMU > > fault. The load is the very first register access for the TGA card. > > I'll see if newbusification makes a difference... > I did a quick experiment and it looks newbusification will solve the problem. Of course this opens pandora's box. The whole frame buffer code is unsuitable of being newbusified. By design, I might add. It'll be a lot of work. I think a kluge is called for but I'm not entirely sure it's something I want or should spend my time on. I need to give it some thought... -- Marcel Moolenaar USPA: A-39004 marcel@xcllnt.net
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040908024119.GA7290>