Re: [PATCH] i2c: muxes: Fix return value check in mule_i2c_mux_probe()

[Farouk Bouabid <farouk.bouabid@cherry.de>]

Hi,

On 04.11.24 11:56, Wolfram Sang wrote:
> On Thu, Oct 31, 2024 at 01:09:17PM +0100, Andi Shyti wrote:
>> Hi Yang,
>>
>> On Sat, Oct 26, 2024 at 11:09:42AM +0800, Yang Yingliang wrote:
>>> From: Yang Yingliang <yangyingliang@huawei.com>
>>>
>>> If dev_get_regmap() fails, it returns NULL pointer not ERR_PTR(),
>>> replace IS_ERR() with NULL pointer check, and return -ENODEV.
>>>
>>> Fixes: d0f8e97866bf ("i2c: muxes: add support for tsd,mule-i2c multiplexer")
>>> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
>> merged to i2c/i2c-host-fixes.
> I'd like an ack from Farouk here. In general, -ENODEV makes the driver
> core sliently fail (ok, we have a printout but still). But not getting
> the regmap sounds like a real error to me. I'd suggest -ENOENT or
> something. But maybe this mux is a special case, Farouk should know.
>

I think the -ENODEV is good enough since we already log the error.

Thanks

Best regards