[PATCH v2] ARM: dts: aspeed: Add IBM P11 BMC boards
Andrew Jeffery
andrew at codeconstruct.com.au
Fri Apr 26 10:03:32 AEST 2024
On Thu, 2024-04-25 at 16:41 -0500, Eddie James wrote:
> On 4/15/24 19:02, Andrew Jeffery wrote:
> > On Mon, 2024-04-15 at 10:49 -0500, Eddie James wrote:
> > > Add the Blueridge and Fuji BMC systems.
> > >
> > > Signed-off-by: Eddie James <eajames at linux.ibm.com>
> > > ---
> > > Changes since v1:
> > > - Change "ody" to "odyssey"
> > >
> > > arch/arm/boot/dts/aspeed/Makefile | 2 +
> > > .../dts/aspeed/aspeed-bmc-ibm-blueridge.dts | 1752 +++++++
> > > .../boot/dts/aspeed/aspeed-bmc-ibm-fuji.dts | 4016 +++++++++++++++++
> > > .../arm/boot/dts/aspeed/ibm-power11-quad.dtsi | 1704 +++++++
> > > 4 files changed, 7474 insertions(+)
> > A few points:
> >
> > - Altogether this patch is pretty large. Can we do one for each of the
> > FSI topology and the platforms?
> >
> > - The only list to which you've sent the patch is linux-aspeed, which
> > doesn't appear to be archived on lore. Can you please add at least one
> > that is? There are a few that should probably be included regardless.
> >
> > - I don't plan to merge the patch until it has had review from ideally
> > more than one other IBMer
> >
> > On the review front:
> >
> > It looks like you still need to update
> > Documentation/devicetree/bindings/arm/aspeed/aspeed.yaml with the new
> > platform compatible strings.
> >
> > Also, can you please run
> >
> > ```
> > make CHECK_DTBS=y aspeed/aspeed-bmc-ibm-{blueridge,fuji}.dtb
> > ```
> >
> > and address the warnings? This is best done by first applying the patch
> > on top of my tree below. It goes some way to eliminating many of the
> > warnings currently produced for the Aspeed BMC SoCs:
> >
> > https://github.com/amboar/linux/tree/dt-warnings/all
>
>
> Thanks, I've sent v3. One patch has turned into 14 to address the
> warnings...
>
Super work, thanks for that.
> There were still many but they're almost all from
> aspeed-g6.dtsi.
>
This is based on top of my WIP series fixing a bunch of warnings, or on
top of an upstream tag for submission? Admittedly I didn't fix them all
but I did clean up a decent chunk of what was there.
> The only remaining ones from Blueridge/Fuji that I saw
> were for undocumented compatibles for max31785 and pca955x.
>
Fantastic.
> Those
> bindings exist but are not in yaml format so the tool doesn't pick them
> up. I'm sure those can be addressed at another time...
Yep, sounds good.
Thanks for your efforts.
Andrew
More information about the Linux-aspeed
mailing list