svn commit: r215327 - head/sys/dev/nfe

[Pyun YongHyeon]

On Mon, Nov 15, 2010 at 02:11:15AM +0200, Alexander Motin wrote:
> Pyun YongHyeon wrote:
> > Author: yongari
> > Date: Sun Nov 14 23:37:43 2010
> > New Revision: 215327
> > URL: http://svn.freebsd.org/changeset/base/215327
> > 
> > Log:
> >   P5N32-SLI PREMIUM from ASUSTeK is known to have MSI/MSI-X issue
> >   such that nfe(4) does not work with MSI-X. When MSI-X support was
> >   introduced, I remember MCP55 controller worked without problems so
> >   the issue could be either PCI bridge or BIOS issue. But I also
> >   noticed snd_hda(4) disabled MSI on all MCP55 chipset so I'm still
> >   not sure this is generic issue of MCP55 chipset. If this was PCI
> >   bridge issue we would have added it to a system wide black-list
> >   table but it's not clear to me at this moment whether it was caused
> >   by either broken BIOS or silicon bug of MCP55 chipset.
> 
> MCP5x seem to be infinite source of surprises. Some reports I remember:

Yeah, they even do not release any errata.

>  - snd_hda not working with MSI enabled - AFAIR not just loosing
> interrupts but completely stops responding;
>  - using regular HPET interrupts breaks HDA sound after some time
> (interrupts are not shared), while legacy_route mode operates properly;
>  - at least on one system I've seen non-functioning SATA interrupts.
> It would be nice to find what's going on there. I've got tired to add
> workarounds for it. :(
> 

Yeah, I agree but I'm sure not all MCP55 was broken because one
user who tested MCP55 with MSI-X reported success. This is reason
why I didn't add it to system-wide blacklist table.
If you remember exact PCI bridge chipset I think you can compare it
with the pciconf output in PR/152150 and if it's the same we can
move it to PCI quirk table.