On Thu, Oct 08, 2020 at 06:05:39PM +0300, Andy Shevchenko wrote:

> My main concern is what was the idea behind? Does it mean we support optional
> DMA in such case? If now, why not to return an error code directly?

...

> no idea, the code has been this way since 2013
> (5eda87b890f867b098e5566b5543642851e8b9c3)

That's "ASoC: dmaengine: support deferred probe for DMA channels" for
those playing at home.  It's been that way since before then since
previously we ignored errors entirely.

> It's worth asking the question on the mailing list, I don't know if this is a
> bug or a feature.

I'm fairly sure it's intentional for systems with limited DMA channels
available but ICBW, it's obviously been quite some time.  In retrospect
a comment explaining the decision would have been a good idea.