[PATCH] Re: Catching NForce2 lockup with NMI watchdog - found

[Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>]

It is possible :-).  Here is a completly untested patch.

[PATCH] fix lockups with APIC support on nForce2

Add PCI quirk to disable Halt Disconnect and Stop Grant Disconnect
(based on athcool program by Osamu Kayasono).

 arch/i386/pci/fixup.c |   13 +++++++++++++
 1 files changed, 13 insertions(+)

diff -puN arch/i386/pci/fixup.c~nforce2_disconnect_quirk arch/i386/pci/fixup.c
--- linux-2.6.0-test11/arch/i386/pci/fixup.c~nforce2_disconnect_quirk	2003-12-06 13:36:56.147911576 +0100
+++ linux-2.6.0-test11-root/arch/i386/pci/fixup.c	2003-12-06 14:03:41.655837272 +0100
@@ -187,6 +187,18 @@ static void __devinit pci_fixup_transpar
 		dev->transparent = 1;
 }
 
+/*
+ * Halt Disconnect and Stop Grant Disconnect (bit 4 at offset 0x6F)
+ * must be disabled when APIC is used (or lockups will happen).
+ */
+static void __devinit pci_fixup_nforce2_disconnect(struct pci_dev *d)
+{
+	u8 t;
+
+	pci_read_config_byte(d, 0x6F, &t);
+	pci_write_config_byte(d, 0x6F, (t & 0xef));
+}
+
 struct pci_fixup pcibios_fixups[] = {
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82451NX,	pci_fixup_i450nx },
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82454GX,	pci_fixup_i450gx },
@@ -205,5 +217,6 @@ struct pci_fixup pcibios_fixups[] = {
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8367_0,	pci_fixup_via_northbridge_bug },
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_NCR,	PCI_DEVICE_ID_NCR_53C810,	pci_fixup_ncr53c810 },
 	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_INTEL,	PCI_ANY_ID,			pci_fixup_transparent_bridge },
+	{ PCI_FIXUP_HEADER,	PCI_VENDOR_ID_NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2,	pci_fixup_nforce2_disconnect },
 	{ 0 }
 };

_

On Saturday 06 of December 2003 13:24, Prakash K. Cheemplavam wrote:
> cheuche+lkml@free.fr wrote:
> > On Sat, Dec 06, 2003 at 12:49:50AM +0100, Prakash K. Cheemplavam wrote:
> >>So gals and guys, try disabling cpu disconnect in bios and see whether
> >>aopic now runs stable.
> >
> > Yes that fix it. Well time will tell but I cannot make it crash with
> > hdparm -tT or cat /dev/hda so far. I'm dumping hda to /dev/null right
> > now.
> >
> > After testing to make it crash, I used athcool to reenable CPU
> > disconnect, and guess what, test after that just crashed the box.
> > You found the problem, congratulations.
> >
> :-)
>
> Isn't it possible to ad athcool's code into the kernel, maybe into the
> pm section or even make it an kernel option. It seems to be a nice
> workaround for the time-being.
>
> Prakash