[Intel-gfx] Problems with HDMI audio on Intel DG45FC motherboard

[Wu Fengguang]

On Mon, Oct 26, 2009 at 04:35:55PM +0800, David Härdeman wrote:
> On Tue, October 20, 2009 23:43, Shane W wrote:
> > I know this has come up before but can't we not start/stop
> > the infoframe if the sample format on the new track is the
> > same as the old. I mean playing an album, you're gonna get
> > 44100, 2ch s16le 95% of the time so that would atleast
> > cause the problem to appear much less.
> 
> Looking at the HDMI spec, the only case where the infoframe is likely to
> change is where you change the number of channels (all other fields are
> set to zero - refer to stream header), so the infoframe is likely to still
> be valid even if you first play a 44.1kHz, 2ch s16le clip followed by a
> 96kHz, 2ch s32le clip (for example).
> 
> And I'm guessing this approach might fix the problems I'm seeing as
> well...I'll try writing a test patch.

Would you try if the hdmi-intel.patch posted in the previous email to
Shane fix your problem? It will only stop/start infoframe transmission
on the first clip.

> Fengguang, note that if you do change the sizeof(ai) call in
> hdmi_fill_audio_infoframe that I mentioned, it seems that a 14-byte struct
> will be written to a 13-byte DIP buffer and I'm not sure if that will
> cause the packet byte index to wrap to zero (meaning that the first byte
> of the buffer will be overwritten with the last byte of the struct). Still
> not sure why the buffer is 13 bytes though...

The byte index wraps on 32:

[  866.179041] ALSA sound/pci/hda/patch_intelhdmi.c:342: HDMI: DIP GP[0] buf reported size=13, written=32

The spec also makes that clear.

Thanks,
Fengguang