Subject: Re: uvm_fault with 1-9-2002 sources
To: Christopher SEKIYA <wileyc@rezrov.net>
From: Jaromir Dolecek <jdolecek@netbsd.org>
List: current-users
Date: 01/10/2002 15:56:19
Christopher SEKIYA wrote:
> On Thu, Jan 10, 2002 at 12:04:14AM -0800, Brian Stark wrote:
> 
> > stopped in pid0 (swapper) at sbdsp_probe+0x26 movi 0x10(xedx),%eax
> 
> Seeing this on my alpha as well.  The problem is in sbdsp.c:
> 
>         /* if flags set, go and probe the jazz16 stuff */
>         if (sc->sc_dev.dv_cfdata->cf_flags & 1)
>                 sbdsp_jazz16_probe(sc); 
> 	else
> 		...
> 
> dv_cfdata seems to be null.  A quick look at the relevent CVS logs does not
> reveal anything obvious, so my quick fix was to comment out the comparison and
> the jazz probe -- resulted in a working kernel.

Well, following should be "right" fix (not tried if it compiles,
it should). Of course, what the code is trying
to do is not particularily clean.

Jaromir 


--- sb_isa.c.orig	Thu Jan 10 15:53:37 2002
+++ sb_isa.c	Thu Jan 10 15:53:52 2002
@@ -98,6 +98,7 @@ sb_isa_match(parent, match, aux)
 
 	memset(sc, 0, sizeof *sc);
 	strcpy(sc->sc_dev.dv_xname, "sb");
+	sc->sc_dev.dv_cfdata = match;
 	return sbfind(parent, sc, 1, aux);
 }
 
-- 
Jaromir Dolecek <jdolecek@NetBSD.org> http://www.NetBSD.org/Ports/i386/ps2.html
-=  Those who would give up liberty for a little temporary safety deserve  =-
-=  neither liberty nor safety, and will lose both.  -- Benjamin Franklin  =-