Re: [PATCH] dt-bindings: Revert "dt-bindings: soc: grf: add naneng combo phy register compatible"

[Rob Herring]

On Wed, Mar 2, 2022 at 11:25 AM Peter Geis <pgwipeout@xxxxxxxxx> wrote:
>
> On Wed, Mar 2, 2022 at 12:14 PM Vinod Koul <vkoul@xxxxxxxxxx> wrote:
> >
> > On 02-03-22, 11:04, Rob Herring wrote:
> > > On Wed, Mar 2, 2022 at 8:34 AM Vinod Koul <vkoul@xxxxxxxxxx> wrote:
> > > >
> > > > This reverts commit b3df807e1fb0 ("dt-bindings: soc: grf: add naneng
> > > > combo phy register compatible") as that was wrongly merged, so better to
> > > > drop the wrong patch
> > > >
> > > > Signed-off-by: Vinod Koul <vkoul@xxxxxxxxxx>
> > > > ---
> > > > I am applying this to phy-next to fix the issue
> > >
> > > Reverting will just cause a different warning that it is undocumented.
> >
> > Right, but a patch for that would fix that
> >
> > > The fix in the other thread won't apply either if you revert.
> >
> > It is not applying for me, so that needs to be updated anyways..
>
> It seems phy-next has fallen out of sync with -next.
> It's missing this patch:
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/Documentation/devicetree/bindings/soc/rockchip/grf.yaml?h=next-20220302&id=7dbb47d64acf4aac131a2aaade726913aa62abe7

That is not how things work. linux-next is a tree that no one can
apply patches to (in the worst case like this one). It's useful for
integration testing and a shortcut for getting a maintainer's tree,
but should not be the basis for patches to the lists. You should
generally use the last rc1 or a maintainer's tree when there is a
known dependency. Using a stable base means 'git am -3' works and the
merge tools work rather than git just failing to apply anything.

Rob