[PATCH 02/30] mtd: spi-nor: winbond: Normalize names

[Miquel Raynal]

On 22/07/2026 at 15:07:46 +02, "Michael Walle" <mwalle at kernel.org> wrote:

> Hi Miquel,
>
> On Wed Jul 22, 2026 at 2:52 PM CEST, Miquel Raynal wrote:
>>>> As a result, W25Q01NWxxIQ is irrelevant and would better be named
>>>> W25Q01NW-Q, W25Q02JV is irrelevant and would better be named W25Q02JV-M,
>>>> etc.
>>>
>>> OTOH if you try to search for that name on the internet/datasheets
>>> you don't find anything, while the xx somehow indicates that
>>> something is missing here.
>>
>> The datasheet is often the same for N/Q/M variants, a table close to the
>> end mentions the differences. The intermediate letters have no useful
>> meaning for us and IIRC they number may vary. The meanging is:
>>
>>         - W: Winbond
>>         - 25Q: SPI NOR family
>>         - 02J: Density
>>         - V: Voltage level
>>         - Useless information for software support such as the package
>>           size or some ratings which have no impact on the software, the
>>           absence of lead, etc.
>>         - And then there are subtleties such as features available or
>>           not, enabled by default or not, which can be disabled or not,
>>           that is carried by the last letter.
>
> I get that, but still, that part name "W25Q01NW-Q" simply doesn't
> exist. I've argued with Tudor years ago, since he wanted me to use a
> similar naming (w25q32jwm). I mean I don't care to much. It's like
> I've said, you won't find anything about that on the web, as the
> name is made up by us :)

I get the same results with the two names. I know it is not
fully deterministic since I must have a history in searching for SPI NOR
datasheets on the internet, but both copy-paste in Google search engine
return very similar results.

"w25q32jwm" looks like an exact name, whereas with the minus sign it
indicates a potential break in the name, no? The problem with the 'x' is
that you don't know how many you need.