[PATCH 3/4] mtd: spi-nor-core: Rework default_init() to take flash_parameter

[Takahiro Kuwano]

On 4/15/2024 4:27 PM, Tudor Ambarus wrote:
> 
> 
> On 4/15/24 08:09, Takahiro Kuwano wrote:
>> Hi Tudor,
> 
> Hi!
> 
>>
>> On 4/15/2024 3:47 PM, Tudor Ambarus wrote:
>>>
>>>
>>> On 4/15/24 05:33, tkuw584924 at gmail.com wrote:
>>>> From: Takahiro Kuwano <Takahiro.Kuwano at infineon.com>
>>>>
>>>> default_init() fixup hook should be used to initialize flash parameters
>>>> when its information is not provided in SFDP. To support that case, it
>>>> needs to take flash_parameter structure like as other hooks.
>>>>
>>>> Signed-off-by: Takahiro Kuwano <Takahiro.Kuwano at infineon.com>
>>>> ---
>>>
>>> I'd like to get rid of the default_init hook, let's not extend it if
>>> possible. Can you use the late_init hook instead?
>>>
>> It looks easy to migrate from default_init to late_init so I will do it.
>> Could you provide the links to related discussion in Linux MTD side so that
>> I can summarize it in commit message?
>>
> 
> I can't, I don't remember if I brought this up or when, but I can
> explain why.
> 
> default_init() is wrong, it contributes to the maze of initializing
> flash parameters. We'd like to get rid of it because the flash
> parameters that it initializes are not really used at SFDP parsing time,
> thus they can be initialized later.
> 
> Ideally we want SFDP to initialize all the flash parameters. If (when)
> SFDP tables are wrong, we fix them with the post_sfdp/bfpt hooks, to
> emphasize that SFDP is indeed wrong. When there are parameters that are
> not covered by SFDP, we initialize them in late_init() - these
> parameters have nothing to do with SFDP and they are not needed earlier.
> With this we'll have a clearer view of who initializes what.
> 
> Feel free to use this in the commit message if you think it helps.
> Cheers,
> ta

Of course it helps!