[RFC PATCH v2 03/13] ARM: mach-at91: cpus/cpu node dts updates
Jean-Christophe PLAGNIOL-VILLARD
plagnioj at jcrosoft.com
Wed Apr 24 05:52:59 EST 2013
On 14:53 Tue 23 Apr , Lorenzo Pieralisi wrote:
> On Tue, Apr 23, 2013 at 02:11:46PM +0100, Rob Herring wrote:
> > On 04/22/2013 10:27 AM, Lorenzo Pieralisi wrote:
> > > This patch updates the in-kernel dts files according to the latest cpus
> > > and cpu bindings updates for ARM.
> > >
> > > Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi at arm.com>
> > > ---
> > > arch/arm/boot/dts/at91sam9260.dtsi | 2 +-
> > > arch/arm/boot/dts/at91sam9263.dtsi | 2 +-
> > > arch/arm/boot/dts/at91sam9g45.dtsi | 2 +-
> > > arch/arm/boot/dts/at91sam9n12.dtsi | 2 +-
> > > 4 files changed, 4 insertions(+), 4 deletions(-)
> > >
> > > diff --git a/arch/arm/boot/dts/at91sam9260.dtsi b/arch/arm/boot/dts/at91sam9260.dtsi
> > > index cb7bcc5..2e9de85 100644
> > > --- a/arch/arm/boot/dts/at91sam9260.dtsi
> > > +++ b/arch/arm/boot/dts/at91sam9260.dtsi
> > > @@ -33,7 +33,7 @@
> > > };
> > > cpus {
> > > cpu at 0 {
> > > - compatible = "arm,arm926ejs";
> > > + compatible = "arm,arm926";
> >
> > I don't understand why you are doing this. If this does not match the
> > documentation, fix the documentation. We can't continue on changing dts
> > files without reqard to breaking compatibility.
>
> IMHO compatibility is already broken. There are a number of dts in the
> kernel missing cpus and cpu nodes, others with cpu nodes missing
> device_type = "cpu", missing cpu nodes compatible properties and the list
> goes on and on. Those files got merged in the kernel before bindings were
> properly defined for ARM so at that point in time the only reference was the
> ePAPR and still, it was not followed (eg my broken patch above fails to add
> device_type = "cpu" to the cpu node, should I change the documentation (ePAPR)
> to make the dts above compliant ? I do not think so, I reckon we should fix
> all dts and force them to comply with the ePAPR and the in-kernel bindings).
>
> If we do not set in stone the bindings and draw a line now, this stuff will
> go wild, it is already in a state that I do not like much.
>
> The reason we are patching the compatible property above is to avoid having
> compatible properties containing suffixes for CPUs, we do not deem that
> necessary, see:
>
> http://lists.infradead.org/pipermail/linux-arm-kernel/2013-January/145305.html
>
> That's just my opinion, open to change it to find a proper solution to this
> issue as long as we make progress.
I do not agree when you set the compatible you need to be preceise the cpu is
a arm926ejs not a arm926
Best Regards,
J/
>
> Thanks for the review,
> Lorenzo
>
> >
> > Rob
> >
> > > };
> > > };
> > >
> > > diff --git a/arch/arm/boot/dts/at91sam9263.dtsi b/arch/arm/boot/dts/at91sam9263.dtsi
> > > index 271d4de..25c4725 100644
> > > --- a/arch/arm/boot/dts/at91sam9263.dtsi
> > > +++ b/arch/arm/boot/dts/at91sam9263.dtsi
> > > @@ -30,7 +30,7 @@
> > > };
> > > cpus {
> > > cpu at 0 {
> > > - compatible = "arm,arm926ejs";
> > > + compatible = "arm,arm926";
> > > };
> > > };
> > >
> > > diff --git a/arch/arm/boot/dts/at91sam9g45.dtsi b/arch/arm/boot/dts/at91sam9g45.dtsi
> > > index 6b1d4ca..cf647d1 100644
> > > --- a/arch/arm/boot/dts/at91sam9g45.dtsi
> > > +++ b/arch/arm/boot/dts/at91sam9g45.dtsi
> > > @@ -36,7 +36,7 @@
> > > };
> > > cpus {
> > > cpu at 0 {
> > > - compatible = "arm,arm926ejs";
> > > + compatible = "arm,arm926";
> > > };
> > > };
> > >
> > > diff --git a/arch/arm/boot/dts/at91sam9n12.dtsi b/arch/arm/boot/dts/at91sam9n12.dtsi
> > > index 7750f98..d531ae3 100644
> > > --- a/arch/arm/boot/dts/at91sam9n12.dtsi
> > > +++ b/arch/arm/boot/dts/at91sam9n12.dtsi
> > > @@ -32,7 +32,7 @@
> > > };
> > > cpus {
> > > cpu at 0 {
> > > - compatible = "arm,arm926ejs";
> > > + compatible = "arm,arm926";
> > > };
> > > };
> > >
> > >
> >
> >
>
> _______________________________________________
> devicetree-discuss mailing list
> devicetree-discuss at lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/devicetree-discuss
More information about the devicetree-discuss
mailing list