Re: [PATCH] pci: don't enable too many HT MSI mapping

[Prakash Punnoor]

On Donnerstag 05 März 2009 18:15:14 Matthew Wilcox wrote:
> On Wed, Mar 04, 2009 at 08:15:37AM +0100, Prakash Punnoor wrote:
> > 2.6.29-rc7 is released, but the last two patches here in the thread have
> > not been picked up. Why?
>
> Oh, I meant to say ... we're at -rc7.  What is the danger here of this
> fixing your machine but breaking somebody else's?

There is a danger that less host bridges get HT MSI enabled by the quirk -
which might need it after all.

> Is there something we
> can minimally do that fixes your machine today and then put in a patch
> for 30-rc1 that is likely to fix other machines?

Something like this would be minimal and works for me:

--- drivers/pci/quirks.c.old    2009-03-06 00:34:40.996532222 +0100
+++ drivers/pci/quirks.c        2009-03-06 00:37:06.915532269 +0100
@@ -2141,6 +2141,10 @@
        int pos;
        int found;

+       /* Enabling HT MSI mapping on this device breaks MCP51 */
+       if (dev->device == 0x270)
+               return;
+
        /* check if there is HT MSI cap or enabled on this device */
        found = ht_check_msi_mapping(dev);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/