[alsa-devel] [PATCHv2.1-CA0132 HDA Codec 1/2] Add DSP loader code to CA0132 codec

[Takashi Iwai]

At Fri, 10 Aug 2012 23:17:51 -0700,
Ian Minett wrote:
> 
> From: Ian Minett <ian_minett at creativelabs.com>
> 
> Hi Takashi,
> Great, it's good to hear that you are able to assist with adding the DSP 
> loader modifications to hda_intel.c.
> 
> I've attached a draft patch as you requested. Please let us know if you have 
> any questions concerning it. The first patch is the previous update to add DSP 
> features to patch_ca0132, then the second is the draft patch, to add the 
> new DMA loader calls.

Well, for cleanness, I think we should start rather from scratch:

- The first patch to add a new hda_bus_ops for DSP transfer;
  it won't touch patch_ca0132.c but only the core part (hda_intel.c,
  hda_codec.[ch])

- Add the DSP transfer to patch_ca0132.c;
  at this point, just do DMA transfer but no new mixers etc won't be
  added

- Add new features to patch_ca0132.c

At the early development stage, the first two items can be a single
patch.  We can split to two when applied to the upstream.  But, mixing
all changes in a single patch is pretty bad.
So the point is to concentrate only on DMA transfer stuff at first.
Once after confirmed it's working, continue to the new DSP features.

> We're also currently ascertaining the license information to go along with the 
> firmware binary release.

Good.  This is really mandatory for upstreaming.

> The project team would like to know, beyond the outstanding items mentioned 
> above, do you see any other potential issues with the updates to the codec that
> could possibly slow the submission process? We're starting to get rather close 
> to our target date.
> Also, once the codec has made it into the ALSA tree, how long do you estimate 
> it would take before it is upstreamed to the Linux kernel? If you could 
> provide any info on the timeframe for this process, we'd greatly appreciate it.

Heh, this depends on how hard we work :)
If it can be quickly done, it'll be merged even in 3.7 kernel.


Takashi