[PATCH v2] ARM: dts: aspeed: Add Facebook Wedge100 BMC
joel at jms.id.au
Fri Aug 2 15:25:43 AEST 2019
On Fri, 2 Aug 2019 at 05:20, Tao Ren <taoren at fb.com> wrote:
> On 8/1/19, 9:21 PM, "Joel Stanley" <joel at jms.id.au> wrote:
> > On Fri, 2 Aug 2019 at 04:10, Tao Ren <taoren at fb.com> wrote:
> >> Add initial version of device tree for Facebook Wedge100 AST2400 BMC
> >> platform.
> >> Signed-off-by: Tao Ren <taoren at fb.com>
> >> Reviewed-by: Andrew Jeffery <andrew at aj.id.au>
> >> ---
> >> Changes in v2:
> >> - remove "debug" from bootargs.
> > Thanks. I applied wedge40 and then this one fails to apply due to
> > conflicts in the Makefile. Next time you have two patches, send them
> > as a series they apply one atop the other.
> I thought about asking you if I should send them as a series although they are logically independent patches..
> Sorry about that and I will do so for future patches.
> > The naming of these two files suggests they come from a family. I
> > noticed there's very minor differences, a pca9548 switch and the use
> > of a watchdog.
> > Are these device trees complete? If yes, do you think it's worthwhile
> > to have a common wedge description in eg.
> > aspeed-bmc-facebook-wedge.dtsi, and put the unique description in
> > respective dts board files?
> > The upside of this is reduced duplication.
> > If you have a reason not to, then that is okay and we can leave it as
> > you submitted them.
> Thank you for the suggestion. I'm also considering moving common stuff into "dtsi" file, but let me take care of it in a separate patch, mainly because:
> 1) I have one more BMC platform (galaxy100) which is also similar to wedge.
> I haven't started the platform, but once I have galaxy100 device tree ready, it would be easier for me to extract common part.
> 2) the device tree is not complete yet.
> For example, all the i2c devices are still created from userspace.
> I'm trying to move the logic from userspace to device tree but I haven't decided what to do with those cpld/fpga devices.
I've applied both of these to the aspeed tree for 5.4.
More information about the Linux-aspeed