Re: [PATCH v2 1/2] Move ralink-gdma to its own directory

[George Hilliard <thirtythreeforty@gmail.com>]

On Thu, Jan 10, 2019, 6:21 PM NeilBrown <neil@brown.name wrote:
>
> On Thu, Jan 10 2019, thirtythreeforty@gmail.com wrote:
>
> > From: George Hilliard <thirtythreeforty@gmail.com>
> >
> > This is in preparation to allow it and the mt7621-dma drivers to be
> > built separately.  They are completely independent pieces of software,
> > and the Kconfig specifies very different requirements.
> >
> > Cc: linux-kernel@vger.kernel.org
> > Cc: devel@driverdev.osuosl.org
> > Cc: Neil Brown <neil@brown.name>
> > Signed-off-by: George Hilliard <thirtythreeforty@gmail.com>
>
> Hi,
>  thanks for taking an interest in these drivers.
>  I original submitted this code because I though I needed it for my
>  mt7621 hardware, but I've subsequently realized that neither of these
>  dma drivers are used in this hardware.
>  Consequently I cannot test any changes you make.
>  But maybe you can - which would be excellent!
>
>  So this is just letting you and Greg know that despite my stated
>  interest, I cannot actually review or test this.
>
> Thanks,
> NeilBrown
>
>

Thanks for the heads up. Honestly I am not sure to what extent I can
test code changes either, at least with the DMA driver. I'm working
with the MT7688, and official docs for it and its cousin the MT7628
are pretty sparse, so I'm currently not even certain that the ralink
gdma driver works for my SoC.

Onion lists these drivers in their OpenWRT device trees' compatible
strings, so they're related. The SPI driver works out of the box at
least. The MMC driver wants to work but needs debugging. All the
drivers need to be better documented. etc. I hope I can improve them,
and I'll make sure I test any actual kernel code I change!

George