RE: [PATCH v3 6/6] riscv: dts: starfive: Add initial StarFive JH8100 device tree

From: JeeHeng Sia
Date: Wed Dec 13 2023 - 19:35:02 EST




> -----Original Message-----
> From: Emil Renner Berthing <emil.renner.berthing@xxxxxxxxxxxxx>
> Sent: Wednesday, December 13, 2023 8:39 PM
> To: Emil Renner Berthing <emil.renner.berthing@xxxxxxxxxxxxx>; JeeHeng Sia <jeeheng.sia@xxxxxxxxxxxxxxxx>; kernel@xxxxxxxx;
> robh+dt@xxxxxxxxxx; krzysztof.kozlowski+dt@xxxxxxxxxx; krzk@xxxxxxxxxx; conor+dt@xxxxxxxxxx; paul.walmsley@xxxxxxxxxx;
> palmer@xxxxxxxxxxx; aou@xxxxxxxxxxxxxxxxx; daniel.lezcano@xxxxxxxxxx; tglx@xxxxxxxxxxxxx; conor@xxxxxxxxxx;
> anup@xxxxxxxxxxxxxx; gregkh@xxxxxxxxxxxxxxxxxxx; jirislaby@xxxxxxxxxx; michal.simek@xxxxxxx; Michael Zhu
> <michael.zhu@xxxxxxxxxxxxxxxx>; drew@xxxxxxxxxxxxxxx
> Cc: devicetree@xxxxxxxxxxxxxxx; linux-riscv@xxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Leyfoon Tan
> <leyfoon.tan@xxxxxxxxxxxxxxxx>
> Subject: Re: [PATCH v3 6/6] riscv: dts: starfive: Add initial StarFive JH8100 device tree
>
> Emil Renner Berthing wrote:
> > Sia Jee Heng wrote:
> > > Add initial device tree for the StarFive JH8100 RISC-V SoC.
> > >
> > > Signed-off-by: Sia Jee Heng <jeeheng.sia@xxxxxxxxxxxxxxxx>
> > > Reviewed-by: Ley Foon Tan <leyfoon.tan@xxxxxxxxxxxxxxxx>
> >
> > Looks good to me, thanks.
> >
> > Acked-by: Emil Renner Berthing <emil.renner.berthing@xxxxxxxxxxxxx>
>
> I just learned that this JH8100 is not actually a real SoC yet but just an FPGA
> implementation, and no indication that that's actually a product that will be
> mass produced. Hence a lot of details may change before it becomes a real
> SoC/product people can buy, so let's not add this device tree yet before
> everything is set in silicon.
>
> Please consider my Acked-by abeve revoked.
>
> Sia Jee Heng: With that said I still think it's super awesome that you're
> beginning upstreaming support for your new SoCs early. I wish more SoC vendors
> would follow your example.
Hi Emil, I am not sure what you mean. We verified on FPGA & Emulator, and the logic
is pretty much close to the real silicon. I did mention that in the cover letter as well.
I am new to Linux, so I am wondering if there is a Linux upstream guideline mentioning
that pre-silicon software is not allowed to upstream? Hope there is an updated Linux
upstream guideline that benefit other vendors.
>
> /Emil