Re: [PATCH 18/20] arch: dts: Fix EHCI/OHCI DT nodes name

[Serge Semin]

On Wed, Oct 14, 2020 at 11:41:17AM -0700, Florian Fainelli wrote:
> On 10/14/20 11:11 AM, Serge Semin wrote:
> > On Wed, Oct 14, 2020 at 11:00:45AM -0700, Florian Fainelli wrote:
> >> On 10/14/20 3:14 AM, Serge Semin wrote:
> >>> In accordance with the Generic EHCI/OHCI bindings the corresponding node
> >>> name is suppose to comply with the Generic USB HCD DT schema, which
> >>> requires the USB nodes to have the name acceptable by the regexp:
> >>> "^usb(@.*)?" . Let's fix the DTS files, which have the nodes defined with
> >>> incompatible names.
> >>>
> >>> Signed-off-by: Serge Semin <Sergey.Semin@xxxxxxxxxxxxxxxxxxxx>
> >>>
> >>> ---
> >>>
> >>> Please, test the patch out to make sure it doesn't brake the dependent DTS
> >>> files. I did only a manual grepping of the possible nodes dependencies.
> >>
> > 
> >> Not sure how you envisioned these change to be picked up, but you may
> >> need to split these changes between ARM/ARM64, MIPS and PowerPC at
> >> least. And within ARM/ARM64 you will most likely have to split according
> >> to the various SoC maintainers.
> > 
> > Hmm, I don't really know how it's going to be done in this case, but there must
> > be a way to get the cross-platform patches picked up in general. For
> > instance, see the patches like:
> > 714acdbd1c94 arch: rename copy_thread_tls() back to copy_thread()
> > 140c8180eb7c arch: remove HAVE_COPY_THREAD_TLS
> > They touched the files from different files, but still have been merged in.
> 

> That situation is different, when a new facility is added and someone
> has gone through the work of adding support for all architectures (or
> nearly all of them), you want them to be merged in a way that limits
> merge conflicts with other architecture changes.
> 
> Here you are fixing warnings, and each file you touch can clearly be
> independently change from other files in the series without causing
> merge conflicts. You are however creating the potential for merge
> conflicts with other changes that the various SoC maintainers have
> queued up.
> 
> > Maybe I should have copied these three patches to the "linux-arch@xxxxxxxxxxxxxxx"
> > list or some other mailing list...
> 
> Maybe Rob can queue them through his device tree repository, with the
> ack of the various SoC maintainers...

That's what I hoped for in the first place. But AFAICS now Rob does the splitting
of his patches himself, while the repo is used either for the
Documentation/devicetree/ patches or for the Rob'es own work. So it seems to me
I'll have to split the series up and resubmit... (

Hope I am wrong. So, @Rob, will you be able to merge this and the next two patches
in via your repo or you'd rather suggest for me to split it up and resubmit?

-Sergey

> -- 
> Florian