[U-Boot] [PATCH v1 3/5] dm: serial: Add setparity

[Patrice CHOTARD]

Hi Simon

On 05/18/2018 03:37 AM, Simon Glass wrote:
> Hi Patrick,
> 
> On 17 May 2018 at 06:50, Patrice Chotard <patrice.chotard at st.com> wrote:
>> From: Patrick Delaunay <patrick.delaunay at st.com>
>>
>> Implements serial setparity ops to allow uart parity change.
>> It allows to select ODD, EVEN or NONE parity.
>>
>> Signed-off-by: Patrick Delaunay <patrick.delaunay at st.com>
>> Signed-off-by: Patrice Chotard <patrice.chotard at st.com>
>> ---
>>
>>   include/serial.h | 16 ++++++++++++++++
>>   1 file changed, 16 insertions(+)
>>
>> diff --git a/include/serial.h b/include/serial.h
>> index 384df94ed0b3..b9ef6d91c9c5 100644
>> --- a/include/serial.h
>> +++ b/include/serial.h
>> @@ -67,6 +67,12 @@ extern int usbtty_tstc(void);
>>
>>   struct udevice;
>>
>> +enum serial_par {
>> +       SERIAL_PAR_NONE,
>> +       SERIAL_PAR_ODD,
>> +       SERIAL_PAR_EVEN
>> +};
>> +
>>   /**
>>    * struct struct dm_serial_ops - Driver model serial operations
>>    *
>> @@ -143,6 +149,16 @@ struct dm_serial_ops {
>>           */
>>          int (*loop)(struct udevice *dev, int on);
>>   #endif
>> +       /**
>> +        * setparity() - Set up the parity
>> +        *
>> +        * Set up a new parity for this device.
>> +        *
>> +        * @dev: Device pointer
>> +        * @parity: parity to use
>> +        * @return 0 if OK, -ve on error
>> +        */
>> +       int (*setparity)(struct udevice *dev, enum serial_par parity);
> 
> To me it seems that changing parity while in the middle of operation
> might be tricky. I suppose this follows along with setbrg() so fair
> enough.
> 
> But I worry about adding more operations here. The next thing to come
> is presumably the length (7 bits, 8 bits, ...). Perhaps we should have
> a more generic setconfig() which change change speed, format and
> parity all at once? For format and parity, we could have a 'default'
> parameter value.

Ok that make sense, i will resubmit a v2

> 
> Also, there should be a corresponding function in serial-uclass.c and
> ideal a call from some sandbox test (although I see at present we
> don't have test/dm/serial.c)

Ok

Patrice

> 
> Regards,
> Simon
>