Re: [PATCH] reset: remove the reset-tn48m driver

[Randy Dunlap]

Hi,

On 1/12/26 3:51 AM, Krzysztof Kozlowski wrote:
> On Mon, Jan 12, 2026 at 10:16:44AM +0100, Robert Marko wrote:
>> On Mon, Jan 12, 2026 at 8:26 AM Krzysztof Kozlowski <krzk@xxxxxxxxxx> wrote:
>>>
>>> On 12/01/2026 07:49, Randy Dunlap wrote:
>>>> The reset-tn48m driver is useless without the simple-mfd parent, which
>>>> is not present in the kernel tree, so remove it and references to it.
>>>
>>> I don't understand that sentence. simple-mfd is present in the kernel
>>> tree, so the condition is clearly false.
>>>
>>> Driver has proper instantiation mechanism, thus all out of tree users
>>> can properly use it and there is no ever requirement to have in-tree
>>> DTS. Otherwise you would remove half of the kernel drivers...
>>
>> The simple-mfd driver itself exists, but all of the TN48M drivers
>> depend on a symbol
>> that sadly, was reverted in [1], and thus you cannot build them, nor
>> are they instantiated without it.
> 
> Ah, so missing is MFD_TN48M_CPLD. You should just say that driver is
> impossible to build outside of compile testing, thus it is unusable.
> simple-mfd is rather confusing in this context, because itself exists.
> 
>>
>> I am sad that was the case as the drivers work, and they were planned
>> for expansion for more
>> switch models but it ended up in this limbo state.
>>
>> [1] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20260109&id=540e6a8114d0
>>
>> Regards,
>> Robert
>>
>>>
>>>>
>>>> Fixes: 5cd3921d16b6 ("reset: Add Delta TN48M CPLD reset controller")
>>>
>>> No bug to be fixed. Or at least not explained yet...
> 
> In that case I would say fixed commit is what you pointed:
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20260109&id=540e6a8114d0

OK, if there is another version.
Hopefully there will just be a patch for the simple-mfd parent.

thanks.
-- 
~Randy