mbox series

[0/4] ARM: dts: hummingboard2: fixes and improvements

Message ID cover.1548861664.git.baruch@tkos.co.il (mailing list archive)
Headers show
Series ARM: dts: hummingboard2: fixes and improvements | expand

Message

Baruch Siach Jan. 30, 2019, 4:18 p.m. UTC
The first two patches fix UART on the optional mikroBUS header.

The last two patches add optional CAN and RS485 peripherals. I sent these 
patches before, but they were not applied at the time:

  https://patchwork.kernel.org/patch/10528011/
  https://patchwork.kernel.org/patch/10528013/

Both patches are rebased here on top of the UART fixes.

Baruch Siach (4):
  ARM: dts: hummingboard2: fix uart3 Rx pin conf
  ARM: dts: hummingboard2: add uart3 pinmux for board rev 1.4
  ARM: dts: hummingboard2: describe optional rs485
  ARM: dts: hummingboard2: describe optional can bus

 arch/arm/boot/dts/imx6qdl-hummingboard2.dtsi | 42 +++++++++++++++++++-
 1 file changed, 41 insertions(+), 1 deletion(-)

Comments

Jon Nettleton Jan. 31, 2019, 6:42 a.m. UTC | #1
On Wed, Jan 30, 2019 at 5:18 PM Baruch Siach <baruch@tkos.co.il> wrote:
>
> The first two patches fix UART on the optional mikroBUS header.
>
> The last two patches add optional CAN and RS485 peripherals. I sent these
> patches before, but they were not applied at the time:
>
>   https://patchwork.kernel.org/patch/10528011/
>   https://patchwork.kernel.org/patch/10528013/
>
> Both patches are rebased here on top of the UART fixes.
>
> Baruch Siach (4):
>   ARM: dts: hummingboard2: fix uart3 Rx pin conf
>   ARM: dts: hummingboard2: add uart3 pinmux for board rev 1.4
>   ARM: dts: hummingboard2: describe optional rs485
>   ARM: dts: hummingboard2: describe optional can bus
>
>  arch/arm/boot/dts/imx6qdl-hummingboard2.dtsi | 42 +++++++++++++++++++-
>  1 file changed, 41 insertions(+), 1 deletion(-)
>

Thanks Baruch,

I would like to note here that I am unhappy with that the only
solution for configuration options is to add comments mentioning
conflicting device-tree entries.  While it does at least provide a
reference for users wanting this functionality it is a terrible
solution.  The only way to manage this properly would be to support
device-tree overlay fragments in the kernel-tree. I would much rather
have an overlays subdirectory infrastructure or some other mechanism
where these options could be properly written and applied to a "base"
dtb.

-Jon