Re: acpi-20040715: functional regression on ASUS M2N

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

> I added your suggested patch to
> 
>   arch/i386/kernel/i8259.c
> 
> and recompiled. 
> 
> Problem persists, no difference I could tell.

I don't know why the patch masks out the reserved interrupts. None of 
the other code that sets ELCR is masking out those bits, and it all 
works fine. Possibly some chipsets use different edge/level behavior for 
these, and the save code is forcing them back to edge. Does the problem 
persist if you go into arch/i386/kernel/i8259.c and change:

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;
}

to:

static void save_ELCR(char *trigger)
{
         /* IRQ 0,1,2,8,13 are marked as reserved */
         trigger[0] = inb(0x4d0);
         trigger[1] = inb(0x4d1);
}



-------------------------------------------------------
This SF.Net email is sponsored by OSTG. Have you noticed the changes on
Linux.com, ITManagersJournal and NewsForge in the past few weeks? Now,
one more big change to announce. We are now OSTG- Open Source Technology
Group. Come see the changes on the new OSTG site. www.ostg.com