Re: [PATCH v5 1/5] mtd: spi-nor: core: add manufacturer flags

[Esben Haabendal <esben@geanix.com>]

Erez <erezgeva2@gmail.com> writes:

> On Thu, 26 Sept 2024 at 09:46, Esben Haabendal <esben@geanix.com> wrote:
>>
>> Erez <erezgeva2@gmail.com> writes:
>>
>> > On Mon, 23 Sept 2024 at 18:19, Michael Walle <mwalle@kernel.org> wrote:
>> >>
>> >> > > > I would gladly remove the obsolete mx25l12805d.
>> >> > > Why? I don't see any need for that.
>> >> > Maybe because we do not want compatibility table?
>> >>
>> >> I don't get this? Anyway, we do not remove support for older
>> >> flashes for no reason.
>> >
>> > I did not insist, you asked.
>> > Macronix stopped selling these chips 15 year ago.
>> > How long do you want to support old chips?
>>
>> It is not unusual for embedded products to have a support span of more
>> than 20 years. And chips such as these flashes might not be entirely new
>> when the product is introduced. So dropping support for SPI-NOR flashes
>> that are newer than 25-30 years is definitely a risk. Somebody out there
>> might not be able to upgrade to latest kernel versions anymore, which is
>> not a position we should put anyone in. With the increasing pressure to
>> upgrade product for better security, we definitely should not make it
>> more difficult to run newer kernel versions than absolutely necessary.
>
> I do not insist. Nor send any patch in this direction.

I did not say or imply that you did any such thing.

You asked an open question, and I gave my response. Nothing more,
nothing less.

> Each project can define the extent of backward compatibility.
> In terms of compilers, linkers and tools, i.e. build environment.
> In terms of standards like the C standard we use.
> In terms of network protocols.
> And also what Hardware do we support.
>
> There is no harm in asking where the boundaries are.
> All projects move their boundaries all the time.
> The Linux kernel is no exception.