AMD 780G chipset major issues 2/3 (snd_hda)

[Robert Noland]

On Wed, 2009-04-08 at 11:38 +0300, Alexander Motin wrote:
> Robert Noland wrote:
> > On Tue, 2009-04-07 at 23:58 +0300, Alexander Motin wrote:
> >> Diego Depaoli wrote:
> >>> On Mon, Apr 6, 2009 at 11:06 PM, Paul B. Mahol <onemda at gmail.com> wrote:
> >>>> Because pcm0 switched position with pcm1 :-)
> >>> Already noticed.
> >>> Do you know why?
> >> It is really interesting question, but probably to the PCI guys.
> > 
> > I think it is due to bus enumeration... the first one found (i.e. lower
> > bus id) becomes pcm0.  If I plug in the radeon HD 3850 which has hdmi
> > audio, it ends up being pcm0, instead of my rear ports.  I also have a
> > seperate codec for front ports, which is a bit of a pain, but...
> 
> I understand this, I am surprised that attach orders on boot and later
> are different:
> 
> on boot:
> hdac0: <ATI RS780 High Definition Audio Controller> mem
> 0xfeae8000-0xfeaebfff irq 19 at device 5.1 on pci1
> hdac1: <ATI SB600 High Definition Audio Controller> mem
> 0xfe8f4000-0xfe8f7fff irq 16 at device 20.2 on pci0
> 
> later:
> hdac0: <ATI SB600 High Definition Audio Controller> mem
> 0xfe8f4000-0xfe8f7fff irq 16 at device 20.2 on pci0
> hdac1: <ATI RS780 High Definition Audio Controller> mem
> 0xfeae8000-0xfeaebfff irq 19 at device 5.1 on pci1
> 
> Any ideas?

jhb is really the guru here... but my guess is that the first is the
initial probe.  If you notice the second sequence is in bus order, which
I'm guessing is the order that they were attached in.

robert.

-- 
Robert Noland <rnoland at FreeBSD.org>
FreeBSD
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: This is a digitally signed message part
Url : http://lists.freebsd.org/pipermail/freebsd-current/attachments/20090408/9633e31a/attachment.pgp