On Fri, May 22, 2020 at 2:16 AM Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> wrote:
> On Wed, 20 May 2020 07:56:36 +0000 Joel Stanley <joel@xxxxxxxxx> wrote:
> > On Mon, 11 May 2020 at 15:19, Rob Herring <robh+dt@xxxxxxxxxx> wrote:
> > > On Wed, 6 May 2020 at 23:13, Joel Stanley <joel@xxxxxxxxx> wrote:
> > > > > These are IPMB nodes with the SLAVE_ADDRESS bit set:
> > > > >
> > > > > +&i2c5 {
> > > > > + //Host3 IPMB bus
> > > > > + status = "okay";
> > > > > + multi-master;
> > > > > + ipmb5@10 {
> > > > > + compatible = "ipmb-dev";
> > > > > + reg = <(0x10 | I2C_OWN_SLAVE_ADDRESS)>;
> > > > > + i2c-protocol;
> > > > > + };
> > > > >
> > > > > This is a correct entry, so dtc should not warn about it.
> > > >
> > > > I sent a patch for dtc here:
> > > > https://lore.kernel.org/lkml/20200508063904.60162-1-joel@xxxxxxxxx/
> > >
> > > Patches for dtc need to be against upstream dtc. There's already a
> > > similar patch posted for it which I commented on and never saw a
> > > respin.
> >
> > Can I suggest some instructions in scsripts/dtc explaining that you
> > don't take patches in the kernel tree for this code?
> >
> > I've sent the patch so it applies to the dtc tree. It would be good to
> > see that change propagate over to -next as others have reported this
> > warning.
>
> These warnings now appear in the arm-soc tree.
Right, I also saw them earlier.
Joel, have you sent your patch to David Gibson for integration into
upstream dtc?
I don't know who sent the other patch, but as long as one of them
gets merged, I'd hope we can pull that into kernel as well.
Arnd
