[alsa-devel] [PATCH 2/3] Blackfin I2S(TDM) CPU DAI driver

[Mark Brown]

On Thu, Jul 16, 2009 at 04:00:06PM +0800, Barry Song wrote:
> Signed-off-by: Barry Song <21cnbao at gmail.com>
> Even though TDM mode can be as part of I2S DAI, but there
> are so much difference in configuration and data flow,
> it's very ugly to integrate I2S and TDM into a module.

It'll be a day or two before I have the bandwidth to properly review the
code here but one thing that springs to mind reading the commit message: 
is the 2 channel I2S case not just a special case of the TDM mode?  Is
it just that the extra work required for being able to handle TDM has a
big runtime cost in buffer allocation or something and so treating it as
a special case of TDM introduces unacceptable overheads.  I'd just like
to better understand what the problems are here, I'm perfectly willing
to believe that this is a sensible solution.