On Wed, 30 May 2007 02:00:25 +0200
Jose Alberto Reguero <[email protected]> wrote:
> I have problems with a M2V motherboard and atl1 driver with msi and
> kernel 2.6.22-rc3. With kernel 2.6.21 I had no problems.
> I must add in drivers/pci/quirks.c (line 1723):
>
> DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS480,
> quirk_disable_msi);
> +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, 0xe238,
> quirk_disable_msi);
>
> to disable msi in the atl1.
See http://bugzilla.kernel.org/show_bug.cgi?id=8472.
A more general solution is to disable MSI for the VIA VT3351 (used by
the Asus M2V). I attempted to do this in the -stable tree with
http://lkml.org/lkml/2007/5/26/167, but it was shot down because the
patch isn't upstream yet.
Unfortunately, I can't submit it upstream because the enabling
patch hasn't yet been applied upstream by the PCI maintainer. My
request for info on the status of the patch went unanswered.
http://lkml.org/lkml/2007/5/19/74
The enabling patch was submitted by Tejun Heo 9 May.
http://lkml.org/lkml/2007/5/9/213
It was applied to -stable (2.6.21.2) 23 May.
http://lkml.org/lkml/2007/5/23/385
It has not yet been applied upstream. When it is, I'll push the atl1
MSI patch.
Jay
On Wed, May 30, 2007 at 06:47:11AM -0500, Jay Cliburn wrote:
> On Wed, 30 May 2007 02:00:25 +0200
> Jose Alberto Reguero <[email protected]> wrote:
>
> > I have problems with a M2V motherboard and atl1 driver with msi and
> > kernel 2.6.22-rc3. With kernel 2.6.21 I had no problems.
> > I must add in drivers/pci/quirks.c (line 1723):
> >
> > DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS480,
> > quirk_disable_msi);
> > +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, 0xe238,
> > quirk_disable_msi);
> >
> > to disable msi in the atl1.
>
> See http://bugzilla.kernel.org/show_bug.cgi?id=8472.
>
> A more general solution is to disable MSI for the VIA VT3351 (used by
> the Asus M2V). I attempted to do this in the -stable tree with
> http://lkml.org/lkml/2007/5/26/167, but it was shot down because the
> patch isn't upstream yet.
>
> Unfortunately, I can't submit it upstream because the enabling
> patch hasn't yet been applied upstream by the PCI maintainer. My
> request for info on the status of the patch went unanswered.
> http://lkml.org/lkml/2007/5/19/74
Sorry, I'm catching up on PCI stuff right now (am traveling in Tokyo
right now, gotta love jet lag...)
thanks for your patience.
greg k-h
On Wed, 30 May 2007 15:07:39 -0700
Greg KH <[email protected]> wrote:
> Sorry, I'm catching up on PCI stuff right now (am traveling in Tokyo
> right now, gotta love jet lag...)
>
> thanks for your patience.
Thanks a lot for the note, Greg. I appreciate it. I'll be patient.
Jay