[PATCH v10] dmaengine: Add MOXA ART DMA engine driver

[mark.rutland@arm.com (Mark Rutland)]

On Tue, Oct 08, 2013 at 10:53:36AM +0100, Jonas Jensen wrote:
> On 7 October 2013 17:12, Mark Rutland <mark.rutland@arm.com> wrote:
> > Sorry I didn't notice this previously, but "moxa" isn't in
> > Documentation/devicetree/bindings/vendor-prefixes.txt (despite several
> > bindings using it). Could you cook up a separate patch to add an entry
> > for Moxa, please?
> 
> Yes, I'll submit a separate patch.

Cheers.

> 
> > Also, given the SoC is called "ART" it's a shame that we're calling this
> > "moxa,moxart-dma" rather than "moxa,art-dma". We already have precedent
> > for "moxart" in bindings though, so changing that's likely to lead to
> > more problems.
> 
> Sorry about that, I think the "moxart" contraction was suggested and
> has been sticky ever since.
> 
> It's at least a little appropriate because the physical chip text
> reads "MOXA ART" (photo):
> 
> https://lh4.googleusercontent.com/-A-2FXDrObU8/UMcMc_K2vEI/AAAAAAAABwg/ldaLZ7ps1P4/w1331-h998-no/UC-7112-LX-picture4.jpg
> 
> Currently three drivers in linux-next use the name with accompanying
> device tree bindings.
> Considering the amount of patches required, can we keep the name, please?

Yeah, I think we have to keep it. It's not objectively wrong, and we
have other contractions (e.g. vexpress) in bindings. It just looks a bit
more odd than the others due to the repetition of "moxa". There's no
benefit to be had changing it now.

> 
> > Sorry for yet more pendantry, but could we instead have:
> >
> >  - interrupts: Should contain an interrupt-specifier for the sole
> >                interrupt generated by the device.
> 
> Fixed in v11.

Sounds good.

Cheers,
Mark.