[PATCH v9 4/6] Documentation: DT: PL011: hi6220: add compatible string for Hisilicon designed UART

[Linus Walleij]

On Tue, Jun 2, 2015 at 11:13 AM, Marc Zyngier <marc.zyngier at arm.com> wrote:
> On 02/06/15 09:59, Linus Walleij wrote:

>> enum amba_vendor {
>>         AMBA_VENDOR_ARM = 0x41,
>> +       AMBA_VENDOR_HISILICON = 0x48,
>>
>> Then patch drivers/tty/serial/amba_pl011.c to add vendor_hisilicon
>> and a match table for 0x00048011 just like everyone else.
>
> That feels weird. This amba_vendor enum is not under control of the DT
> author, nor the kernel. This is a set of codes that are managed by a
> third party (probably ARM). What if some company with a name starting
> with 'H' (Hilarious Inc?) comes up with some actual HW and ends up
> conflicting with the above?

Sorry, as indicated by
commit db4fa45ed3182d8206af241811dfc99369ffa849
"spi: pl022: Add support for chip select extension"
bits [6:0] should be the JEP106 manufacturer identity code.

Not that I know how you get such a code. Hardware which is
modified and does not set the JEP106 ID code is buggy I
guess.

Anyways, the pattern we have to handle this is to use that ID
code.

Yours,
Linus Walleij