Re: [PATCH 6/6] serial: imx: do not set RS485 enabled if it is not supported

[Lino Sanfilippo <LinoSanfilippo@gmx.de>]

Hi,

On 29.09.23 08:39, Uwe Kleine-König wrote:
> On Fri, Sep 29, 2023 at 12:12:46AM +0200, Lino Sanfilippo wrote:
>> From: Lino Sanfilippo <l.sanfilippo@kunbus.com>
>>
>> If the imx driver cannot support RS485 it sets the UARTS rs485_supported
>> structure to NULL. But it still calls uart_get_rs485_mode() which may set
>> the RS485_ENABLED flag.
>> The flag however is evaluated by the serial core in uart_configure_port()
>
> I wonder if this is the code location where this problem should be
> addressed. Or alternatively don't let uart_get_rs485_mode() set
> RS485_ENABLED (and other flags) if rs485_supported doesn't suggest that
> this works?
>

This is indeed the better solution. Especially since the check is then in the serial
core and all RS485 supporting drivers benefit from it. I will change this for v2, thanks!


>> [...]
>>
>> Signed-off-by: Lino Sanfilippo <l.sanfilippo@kunbus.com>
>
> I don't know how picky Greg is here, but formally you missed to add an
> S-o-b line for the sender of this patch (i.e. you with your gmx
> address).
>

Hm, until now there have never been complaints about this. Is this really an issue? Of
course I can also S-o-b with my gmx address but adding two S-o-b's for the same person seems
a bit odd to me...

BR,
Lino

> Best regards
> Uwe
>