PERFORCE change 36551 for review

Marcel Moolenaar marcel at FreeBSD.org
Thu Aug 21 00:31:02 PDT 2003


http://perforce.freebsd.org/chv.cgi?CH=36551

Change 36551 by marcel at marcel_nfs on 2003/08/21 00:30:38

	Aarrgghh. The interrupt issue on pluto1 and pluto2 are
	very likely not driver problems. I forgot we have a hack
	on the ia64 branch to deal with the fact that the non-
	legacy interrupts used by the UARTs are edge sensitive,
	active high. We really need to get that fixed in CVS...

Affected files ...

.. //depot/projects/uart/ia64/ia64/interrupt.c#4 edit

Differences ...

==== //depot/projects/uart/ia64/ia64/interrupt.c#4 (text+ko) ====

@@ -266,9 +266,18 @@
 		struct sapic *sa = ia64_sapics[i];
 		if (irq < sa->sa_base || irq > sa->sa_limit)
 			continue;
+		/*
+		 * KLUDGE: Not all interrupts higher or equal to 16 are
+		 * active low and level sensitive. We don't know yet how
+		 * to check for this, so we hardcode the 2 cases we have
+		 * wrong explicitly. This kludge is specific to the HP
+		 * rx2600...
+		 */
 		sapic_enable(sa, irq - sa->sa_base, vector,
-		    (irq < 16) ? SAPIC_TRIGGER_EDGE : SAPIC_TRIGGER_LEVEL,
-		    (irq < 16) ? SAPIC_POLARITY_HIGH : SAPIC_POLARITY_LOW);
+		    ((irq < 16 || vector == 66 || vector == 67)
+			? SAPIC_TRIGGER_EDGE : SAPIC_TRIGGER_LEVEL),
+		    ((irq < 16 || vector == 66 || vector == 67)
+			? SAPIC_POLARITY_HIGH : SAPIC_POLARITY_LOW));
 	}
 }
 


More information about the p4-projects mailing list