linux-kernel - Re: [PATCH] [v3] arch: arm: boot: dts: Create HPE GXP Device Tree

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAK8P3a1HuTMt5WmOw3aQcLwGZd_WQ5nsKZ2HFDPQ0CHFvKsygA@mail.gmail.com>
Date:   Wed, 16 Feb 2022 17:50:47 +0100
From:   Arnd Bergmann <arnd@...db.de>
To:     "Hawkins, Nick" <nick.hawkins@....com>
Cc:     Arnd Bergmann <arnd@...db.de>,
        "Verdun, Jean-Marie" <verdun@....com>,
        Olof Johansson <olof@...om.net>, SoC Team <soc@...nel.org>,
        Rob Herring <robh+dt@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        DTML <devicetree@...r.kernel.org>, Joel Stanley <joel@....id.au>
Subject: Re: [PATCH] [v3] arch: arm: boot: dts: Create HPE GXP Device Tree

On Wed, Feb 16, 2022 at 5:29 PM Hawkins, Nick <nick.hawkins@....com> wrote:
> > +GXP ARM ARCHITECTURE
> > +M:     Nick Hawkins <nick.hawkins@....com>
> > +M:     Jean-Marie <verdun@....com>
>
> It looks like you are missing the family name here.
>
> NH: Do you mean I need to have an HPE in front of it?

No, I meant a 'Verdun' after Jean-Marie ;-)

> ...
> > diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
> > index 235ad559acb2..a96b4d5b7f68 100644
> > --- a/arch/arm/boot/dts/Makefile
> > +++ b/arch/arm/boot/dts/Makefile
> > @@ -1549,3 +1549,5 @@ dtb-$(CONFIG_ARCH_ASPEED) += \
> >         aspeed-bmc-vegman-n110.dtb \
> >         aspeed-bmc-vegman-rx20.dtb \
> >         aspeed-bmc-vegman-sx20.dtb
> > +dtb-$(CONFIG_ARCH_HPE_GXP) += \
> > +       hpe-bmc-dl360gen10.dtb
>
> This Kconfig symbol does not yet exist
>
> NH: Should I include the definition of this Kconfig symbol in this patch or just not modify the makefile when doing device tree changes?

I would start with a patch for the bindings, or one per binding, then
the Kconfig
patch, then the dts file and finally the MAINTAINERS entry.

The order is not super important, but it helps to do it in a way that one patch
builds on the previous one in the series, rather than having a single big patch,
or separate small patches that are not in a series.

>
> > +
> > +/ {
> > +       #address-cells = <1>;
> > +       #size-cells = <1>;
> > +       compatible = "hpe,gxp";
>
> No board specific compatible string? Also, this value is not documented anywhere.
>
> NH: I was uncertain if this needed to be documented or not, checkpatch did not notice it. Where would you recommend I should put this in documentation? I was considering Documentation/devicetree/bindings/soc/hpe/

Documentation/devicetree/bindings/arm/hpe.yaml would be best.
The bindings/soc/ directory is not for the SoC as a whole but more
for components on the SoC that are either for identifying the chip
(as in struct soc_device in Linux)  or for things that don't fit anywhere
else but are specific to one soc.

> ...
>
> > +
> > +               watchdog: watchdog@...00090 {
> > +                       compatible = "hpe,gxp-wdt";
> > +                       reg = <0xc0000090 0x02>, <0xc0000096 0x01>;
> > +               };
>
> As mentioned in a previous review, it would be helpful to have the drivers and bindings together in the same series for easier review.
>
> NH: Just to confirm, I need the driver source and the new DTS file in one review? I also need to have a review to add files: arch/arm/configs/gxp_defconfig, arch/arm/mach-hpe/gxp.c should those be included as well? I am trying to not create too massive of a patch. I could also create completely separate patches for these items so that reviewers could refer to it.

Yes, I think it's best to have all patches in a larger series for the
initial review. The defconfig
patch is usually separate, but the gxp.c file can be in the same patch
as maintainers entry
and the Kconfig and Makefile changes (outside of dts).

Try to keep each patch in the series self-contained, e.g. don't
separate a .c file from the Makefile
and Kconfig change, but don't group things into one patch that are not
directly related, e.g.
the dts file.

> ...
>
> To a lesser degree, the same is true for the uart.
>
> Please check the devicetree files in order to validate the bindings. In this case, you should get a warning about the 'ehci@' name being non-standard. The normal name is usb@
>
> NH: Which tool are you using to validate bindings? I did not receive any warnings but I did receive several errors which I corrected.

You can run "make dtbs W=1" to get most of the useful checks enabled.

> > +
> > +               vrom@...00000 {
> > +                       compatible = "mtd-ram";
>
> Same thing here, "vrom" is clearly not a standard name.
>
> NH: VROM is a Virtual ROM device that we use to create a memory mapping of a SPI flash content copy that is served to a host system. How should I rename it?

Not sure, I don't remember seeing this case so far. Adding Joel to Cc,
maybe he knows of
another BMC that does the same thing.

         Arnd