On Wed, Jun 11, 2003 at 09:22:36PM +0200, Martin Husemann wrote:
> Oops, I'll check that errors (think I never tried nell on 1.6.1).

Yep, there has been a pullup-botch. The sbus attach args changed, but only 
drivers that were active in GENERIC have been adapted.

Could you try the attached patch? It makes the kernel compile for me, but
I can not test it right now.

I'll create a pullup request for this change.

Martin

Index: stp4020.c
===================================================================
RCS file: /cvsroot/src/sys/dev/sbus/stp4020.c,v
retrieving revision 1.22
diff -u -r1.22 stp4020.c
--- stp4020.c	2002/03/25 09:02:54	1.22
+++ stp4020.c	2003/06/11 19:36:42
@@ -323,9 +323,9 @@
 			continue;
 
 		if (sbus_bus_map(sa->sa_bustag,
-				 sa->sa_reg[i].sbr_slot,
-				 sa->sa_reg[i].sbr_offset,
-				 sa->sa_reg[i].sbr_size,
+				 sa->sa_reg[i].oa_space,
+				 sa->sa_reg[i].oa_base,
+				 sa->sa_reg[i].oa_size,
 				 0, &bh) != 0) {
 			printf("%s: attach: cannot map registers\n",
 				self->dv_xname);
@@ -355,10 +355,10 @@
 	 * the lower level for PC card I/O.
 	 */
 	if (sa->sa_nintr != 0) {
-		bus_intr_establish(sa->sa_bustag, sa->sa_intr[1].sbi_pri,
+		bus_intr_establish(sa->sa_bustag, sa->sa_intr[1].oi_pri,
 				   IPL_NONE, 0, stp4020_statintr, sc);
 
-		bus_intr_establish(sa->sa_bustag, sa->sa_intr[0].sbi_pri,
+		bus_intr_establish(sa->sa_bustag, sa->sa_intr[0].oi_pri,
 				   IPL_NONE, 0, stp4020_iointr, sc);
 	}