Re: [PATCH 4.19.y-cip 01/12] arm64: dts: renesas: r8a774a1-hihope-rzg2m[-ex/-ex-idk-1110wr]: Rename HiHope RZ/G2M boards


Biju Das <biju.das.jz@...>
 

Hi Pavel,

Thanks for the feedback.

Subject: Re: [PATCH 4.19.y-cip 01/12] arm64: dts: renesas: r8a774a1-hihope-
rzg2m[-ex/-ex-idk-1110wr]: Rename HiHope RZ/G2M boards

Hi!

Ok, so this is anti-social:

-dtb-$(CONFIG_ARCH_R8A774A1) += r8a774a1-hihope-rzg2m.dtb
+dtb-$(CONFIG_ARCH_R8A774A1) += r8a774a1-hihope-rzg2m-rev2.dtb
This renames dts away, but at the end of series, new dts is created
with the r8a774a1-hihope-rzg2m.dts name, but this time it is for rev4 (not
rev2) board.

So... people with rev2 boards and existing build script will get rev4 dts..
without any error.
Yes, that is true. But if you agree, We need to use our latest/greatest
SoC/board as the main SoC/Board and rest are with explicit revision in
dts/dtb .
I believe it would be better to always use explicit revision in these cases;
anything else is fairly confusing, as kernel update breaks your setup (aka a
regresssion).

Now, if -rev2 had few copies, and people really have -rev4, there will not be
much breakage. But imagine if rare -rev5 is released in future...

But mainline already made the choice. I'd just prefer not making same choice
in future.

Now, what to do here? One way would be to apply a series, but add a
README.cip file explaining incompatible change (and any future stuff people
need to know).
OK. Either README.cip or " by including more information on the reference hardware and test information on CIP wiki." as suggested by Nobuhiro.
Please let us know, How do we want to proceed?

Cheers,
Biju


Renesas Electronics Europe GmbH, Geschaeftsfuehrer/President: Carsten Jauch, Sitz der Gesellschaft/Registered office: Duesseldorf, Arcadiastrasse 10, 40472 Duesseldorf, Germany, Handelsregister/Commercial Register: Duesseldorf, HRB 3708 USt-IDNr./Tax identification no.: DE 119353406 WEEE-Reg.-Nr./WEEE reg. no.: DE 14978647

Join cip-dev@lists.cip-project.org to automatically receive all group messages.