Re: acpi-20040715: functional regression on ASUS M2N

[Nathan Bryant <nbryant-p32f3XyCuykqcZcGjlUOXw@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 513 bytes --]

Georg C. F. Greve wrote:

> nb> Georg, can you add some printk's to save_ELCR and log the values
> nb> that we are actually saving?
>
>I'll be glad to do that, but will have to see how to do it in a way
>that I actually see what is printed. Haven't done much kernel hacking
>really, is there "the right way" to do it?
>  
>
The attached patch is good enough. Since the call to restore_ELCR is 
crashing you, we comment it out, and log the raw values from save_ELCR:

By the way, can you send the output of lspci?

[-- Attachment #2: elcrlog.patch --]
[-- Type: text/x-patch, Size: 870 bytes --]

===== arch/i386/kernel/i8259.c 1.35 vs edited =====
--- 1.35/arch/i386/kernel/i8259.c	2004-07-14 16:00:16 -04:00
+++ edited/arch/i386/kernel/i8259.c	2004-08-06 09:33:02 -04:00
@@ -244,21 +244,23 @@
  */
 static void restore_ELCR(char *trigger)
 {
-	outb(trigger[0], 0x4d0);
-	outb(trigger[1], 0x4d1);
+	/* IRQ 0,1,2,8,13 are marked as reserved */
+	outb(trigger[0] & 0xF8, 0x4d0);
+	outb(trigger[1] & 0xDE, 0x4d1);
 }
 
 static void save_ELCR(char *trigger)
 {
-	/* IRQ 0,1,2,8,13 are marked as reserved */
-	trigger[0] = inb(0x4d0) & 0xF8;
-	trigger[1] = inb(0x4d1) & 0xDE;
+	trigger[0] = inb(0x4d0);
+	trigger[1] = inb(0x4d1);
+	printk(KERN_DEBUG "i8259: saving ELCR: 0x%02x 0x%02x\n",
+		trigger[0], trigger[1]);
 }
 
 static int i8259A_resume(struct sys_device *dev)
 {
 	init_8259A(0);
-	restore_ELCR(irq_trigger);
+	/*restore_ELCR(irq_trigger);*/
 	return 0;
 }