Re: [alsa-devel] [PATCH 1/2] ASoC: ux500_pcm: Differentiate between pdata and DT initialisation

[Lars-Peter Clausen <lars@metafoo.de>]

On 12/03/2013 01:15 PM, Lars-Peter Clausen wrote:
> On 12/03/2013 01:04 PM, Lee Jones wrote:
>> On Tue, 03 Dec 2013, Lars-Peter Clausen wrote:
>>
>>> On 12/03/2013 11:27 AM, Lee Jones wrote:
>>>> If booting with full DT support (i.e. DMA too, the last piece of the
>>>> puzzle), then we don't need to use the compatible_request_channel call
>>>> back or require some of the historical bumph which probably isn't
>>>> required by a platform data start-up now either. This will also be
>>>> ripped out in upcoming commits.
>>>>
>>>> Cc: alsa-devel@alsa-project.org
>>>> Acked-by: Linus Walleij <linus.walleij@linaro.org>
>>>> Signed-off-by: Lee Jones <lee.jones@linaro.org>
>>>
>>> Strictly speaking you do not need the second config, but well considering
>>> that this will all hopefully be removed soon anyway it should be fine.
>>
>> That's true it will, but why don't we need the second config? I won't
>> want the compat function to be called in the DT case.
> 
> It will only be called if it was not possible to request the DMA channel
> from the DT.

The logic used to request the channel is basically this (in pseudo code).

chan = NULL.

if (of_node && !(flags & NO_DT))
	chan = dma_request_slave_channel(...)

if (!chan && (flags & COMPAT))
	chan = dma_request_channel(compat_filter_fn, compat_filter_data);