Date: Mon, 15 Mar 2004 23:09:37 -0800 From: Marcel Moolenaar <marcel@xcllnt.net> To: Ken Smith <kensmith@cse.Buffalo.EDU> Cc: freebsd-sparc64@freebsd.org Subject: Re: Console patch part II... Message-ID: <20040316070937.GA13045@dhcp01.pn.xcllnt.net> In-Reply-To: <20040316070004.GA16684@electra.cse.Buffalo.EDU> References: <20040316055650.GA15182@electra.cse.Buffalo.EDU> <20040316064541.GA12961@dhcp01.pn.xcllnt.net> <20040316070004.GA16684@electra.cse.Buffalo.EDU>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Mar 16, 2004 at 02:00:04AM -0500, Ken Smith wrote: > On Mon, Mar 15, 2004 at 10:45:41PM -0800, Marcel Moolenaar wrote: > > > Your patch looks like a big kluge. Mostly because you're delaying > > cninit() and hardcode a dependency on ofw_console. That sure breaks > > my box and I'm not going to add ofw_console simply because I > > already have a console (using uart(4)). > > Do you know what happens if you call panic() before cninit()? You need to fix the code to call cninit() soon enough that you never need to panic before it? > The > reason I tried the big kluge was that I haven't been able to > successfully move enough of what seem to be the pre-requisites to > cninit() to before the first spot panic() gets called. This mostly tells that the prerequisites are bogus. > If panic() > can be called before cninit() gets run then moving stuff around might > be easier. You can never call panic() before cninit(), because panic() is meaningless without a console. Where is the problem exactly? Is it syscons again (like on alpha)? -- 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?20040316070937.GA13045>