Fix a bug that affects machines which has non-zero base GPE blocks.

This bug is already reported to Intel ACPICA people.
This commit is contained in:
kochi 2003-03-05 15:13:01 +00:00
parent be084ba94f
commit 7629f23bb8

View File

@ -115,14 +115,14 @@
*****************************************************************************/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: evgpeblk.c,v 1.2 2003/03/04 17:25:15 kochi Exp $");
__KERNEL_RCSID(0, "$NetBSD: evgpeblk.c,v 1.3 2003/03/05 15:13:01 kochi Exp $");
#include "acpi.h"
#include "acevents.h"
#include "acnamesp.h"
#define _COMPONENT ACPI_EVENTS
ACPI_MODULE_NAME ("evgpe")
ACPI_MODULE_NAME ("evgpeblk")
/*******************************************************************************
@ -209,7 +209,7 @@ AcpiEvSaveMethodInfo (
/* Ensure that we have a valid GPE number for this GPE block */
if ((GpeNumber < GpeBlock->BlockBaseNumber) ||
(GpeNumber >= (GpeBlock->RegisterCount * 8)))
(GpeNumber - GpeBlock->BlockBaseNumber >= (GpeBlock->RegisterCount * 8)))
{
/* Not valid, all we can do here is ignore it */