[Intel-gfx] Query on Intel IGP and audio over displayport

[Mikael Öhman]

On Mon, May 17, 2010 at 3:16 AM, Zhenyu Wang <zhenyuw at linux.intel.com>
wrote:
>
> On 2010.05.15 03:46:35 +0200, Mikael Öhman wrote:
> > Hello.
> > I was wondering if it was possible to send audio over displayport using
a Core
> > i5 661 (Clarkdale) IGP.
> > I have seen some charts indicating that the hardware actually supports
it, but
> > I'm not even sure if I should ask here or perhaps to some ALSA mailing
list.
> > On google I have only found things related to DP->HMDI with audio, which
I
> > don't need.
>
> Yeah, we're looking into that actually. And current difficult is how to
> identify the DP sink device really has audio support. One is to use EDID
> CEA ext 'basic audio' feature, and another I've tried is to parse DPCD
> downstream device type. Unfortunately the DP monitors we have now don't
> have sane indication for them all, so we don't have good hardware to
verify
> this (also tried windows driver but still failed with DP audio on it).
>
> Maybe we should just use EDID to determine whether enable audio or not.

Xorg.log do contain some EDID output. I don't know what tool i should use
otherwise. get-edid gives me some errors (and it tries to parse my second
monitor only, I need to disconnect it when i get home). It doesn't give me
any more information anyway.

(II) intel(0): EDID for output DP3
(II) intel(0): Manufacturer: NEC  Model: 677a  Serial#: 0
(II) intel(0): Year: 2009  Week: 49
(II) intel(0): EDID Version: 1.4
(II) intel(0): Digital Display Input
(II) intel(0): 8 bits per channel
(II) intel(0): Digital interface is DisplayPort
snip
(II) intel(0): Monitor name: EA231WMi
(II) intel(0): Serial No: 9Z104746NB
(II) intel(0): Number of EDID sections to follow: 1
(II) intel(0): EDID (in hex):
(II) intel(0): 00ffffffffffff0038a37a6700000000
(II) intel(0): 31130104a5331d78e2f805a4554d9d25
(II) intel(0): 125054bfef8081008140818081c09500
(II) intel(0): 9040b300a9c0023a801871382d40582c
(II) intel(0): 4500fe1f1100001e000000fd00384c1f
(II) intel(0): 5311000a202020202020000000fc0045
(II) intel(0): 41323331574d690a20202020000000ff
(II) intel(0): 00395a3130343734364e420a202001a5

If i read the specification for EDID correctly, 01 would specify 1 extension
block. The extension block doesn't seem to printed in Xorg.0.log


>
> >
> > aplay only lists
> > card 0: Intel [HDA Intel], device 0: ALC889 Analog [ALC889 Analog]
> > card 0: Intel [HDA Intel], device 1: ALC889 Digital [ALC889 Digital]
> > card 0: Intel [HDA Intel], device 3: INTEL HDMI 0 [INTEL HDMI 0]
> > card 0: Intel [HDA Intel], device 7: INTEL HDMI 1 [INTEL HDMI 1]
> > and naturally I've tried them all, with no sound.
> >
> > I'm using Debian unstable, and xserver-xorg-video-intel 2.11
(everything else
> > also uptodate with Sid if that matters.)
> >
> >
>
> You may try to force 'has_audio' to true in intel_dp.c, and try with HDA
> HDMI codec. Let me know if that'll work for you, and what's your monitor
model?
>

I have a NEC EA231WMi
I just accidentally found that my monitor had the option to select sound
from DP (and I can't test it with any other equipment).
I will test it be hardcoding has_audio=true tonight.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20100517/6cf8432e/attachment.html>