Re: [PATCH] Fix incorrect Asus k7m irq router detection

Re: [PATCH] Fix incorrect Asus k7m irq router detection

[Dave Jones]

On Wed, Jul 27, 2005 at 10:10:35AM -0700, Aleksey Gorelov wrote:
 
 > > >  static __init int via_router_probe(struct irq_router *r, 
 > >struct pci_dev *router, u16 device)
 > > >  {
 > > >  	/* FIXME: We should move some of the quirk fixup stuff here */
 > > > +
 > > > +	if (router->device == PCI_DEVICE_ID_VIA_82C686 &&
 > > > +			device == PCI_DEVICE_ID_VIA_82C586_0) {
 > > > +		/* Asus k7m bios wrongly reports 82C686A as 
 > >586-compatible */
 > > > +		device = PCI_DEVICE_ID_VIA_82C686;
 > > > +	}
 > > > +
 > > >  	switch(device)
 > > >  	{
 > > >  		case PCI_DEVICE_ID_VIA_82C586_0:
 > >
 > >If this really is a problem with that board, it should have a DMI entry
 > >for that board alone, not for every VIA chipset that uses the 
 > >586/686 combo,
 > >as I'm fairly certain there are some that legitimately use 
 > >this combination,
 > >and the patch above will force them all to be reported as 82C686's.
 > 
 >   How can they use it legitimately if 586 & 686 routers are not
 > programming register compatible ? Any board which reports  that 686 is
 > compatible with 586 will have that issue.

Actually now that I've woken up properly, your patch looks ok to me.
For some reason I was thinking we were comparing a north and south
bridge here.

		Dave

[PATCH] Fix incorrect Asus k7m irq router detection

[Giancarlo Formicuccia]

Hi,

this patch:
http://marc.theaimsgroup.com/?l=bk-commits-head&m=111955644929114&w=2
uncovered a k7m bios bug, where the VT82C686A router is reported as
being "586-compatible". The two chips have different pirq mapping, so
this leads to "irq routing conflict" on many pci devices.

The suggested fix was discussed with Aleksey Gorelov, who helped me
to identify the problem as a probable bios bug.

Patch for 2.6.13-git4.

Signed-off-by: Giancarlo Formicuccia <giancarlo.formicuccia@gmail.com>

--- linux-2.6.13-rc3-git4/arch/i386/pci/irq.c.org	2005-07-27 08:58:05.000000000 +0200
+++ linux-2.6.13-rc3-git4/arch/i386/pci/irq.c	2005-07-27 08:59:29.000000000 +0200
@@ -550,6 +550,13 @@
 static __init int via_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
 	/* FIXME: We should move some of the quirk fixup stuff here */
+
+	if (router->device == PCI_DEVICE_ID_VIA_82C686 &&
+			device == PCI_DEVICE_ID_VIA_82C586_0) {
+		/* Asus k7m bios wrongly reports 82C686A as 586-compatible */
+		device = PCI_DEVICE_ID_VIA_82C686;
+	}
+
 	switch(device)
 	{
 		case PCI_DEVICE_ID_VIA_82C586_0:

Re: [PATCH] Fix incorrect Asus k7m irq router detection

[Dave Jones]

On Wed, Jul 27, 2005 at 10:16:30AM +0200, Giancarlo Formicuccia wrote:

 > this patch:
 > http://marc.theaimsgroup.com/?l=bk-commits-head&m=111955644929114&w=2
 > uncovered a k7m bios bug, where the VT82C686A router is reported as
 > being "586-compatible". The two chips have different pirq mapping, so
 > this leads to "irq routing conflict" on many pci devices.
 > 
 > The suggested fix was discussed with Aleksey Gorelov, who helped me
 > to identify the problem as a probable bios bug.
 > 
 > Patch for 2.6.13-git4.
 > 
 > Signed-off-by: Giancarlo Formicuccia <giancarlo.formicuccia@gmail.com>
 > 
 > --- linux-2.6.13-rc3-git4/arch/i386/pci/irq.c.org	2005-07-27 08:58:05.000000000 +0200
 > +++ linux-2.6.13-rc3-git4/arch/i386/pci/irq.c	2005-07-27 08:59:29.000000000 +0200
 > @@ -550,6 +550,13 @@
 >  static __init int via_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 >  {
 >  	/* FIXME: We should move some of the quirk fixup stuff here */
 > +
 > +	if (router->device == PCI_DEVICE_ID_VIA_82C686 &&
 > +			device == PCI_DEVICE_ID_VIA_82C586_0) {
 > +		/* Asus k7m bios wrongly reports 82C686A as 586-compatible */
 > +		device = PCI_DEVICE_ID_VIA_82C686;
 > +	}
 > +
 >  	switch(device)
 >  	{
 >  		case PCI_DEVICE_ID_VIA_82C586_0:

If this really is a problem with that board, it should have a DMI entry
for that board alone, not for every VIA chipset that uses the 586/686 combo,
as I'm fairly certain there are some that legitimately use this combination,
and the patch above will force them all to be reported as 82C686's.

		Dave