Re: [PATCH 27/36] tty: propagate u8 data to tty_operations::write()

[Jiri Slaby <jirislaby@kernel.org>]

Hi,

On 11. 08. 23, 13:52, Alexander Gordeev wrote:
> On Thu, Aug 10, 2023 at 11:15:01AM +0200, Jiri Slaby (SUSE) wrote:
> 
> Hi Jiri,
> 
>> Data are now typed as u8. Propagate this change to
>> tty_operations::write().
> ...
>>   drivers/s390/char/con3215.c            | 3 +--
>>   drivers/s390/char/con3270.c            | 3 +--
>>   drivers/s390/char/sclp_tty.c           | 2 +-
>>   drivers/s390/char/sclp_vt220.c         | 2 +-
> ...
>> diff --git a/drivers/s390/char/con3215.c b/drivers/s390/char/con3215.c
>> index a1fef666c9b0..16b6f430dfd3 100644
>> --- a/drivers/s390/char/con3215.c
>> +++ b/drivers/s390/char/con3215.c
>> @@ -1021,8 +1021,7 @@ static unsigned int tty3215_write_room(struct tty_struct *tty)
>>   /*
>>    * String write routine for 3215 ttys
>>    */
>> -static int tty3215_write(struct tty_struct *tty,
>> -			 const unsigned char *buf, int count)
>> +static int tty3215_write(struct tty_struct *tty, const u8 *buf, int count)
>>   {
>>   	handle_write(tty->driver_data, buf, count);
>>   	return count;
>> diff --git a/drivers/s390/char/con3270.c b/drivers/s390/char/con3270.c
>> index d9983550062d..123524bff734 100644
>> --- a/drivers/s390/char/con3270.c
>> +++ b/drivers/s390/char/con3270.c
>> @@ -1803,8 +1803,7 @@ static void tty3270_do_write(struct tty3270 *tp, struct tty_struct *tty,
>>   /*
>>    * String write routine for 3270 ttys
>>    */
>> -static int tty3270_write(struct tty_struct *tty,
>> -			 const unsigned char *buf, int count)
>> +static int tty3270_write(struct tty_struct *tty, const u8 *buf, int count)
>>   {
>>   	struct tty3270 *tp;
>>   
>> diff --git a/drivers/s390/char/sclp_tty.c b/drivers/s390/char/sclp_tty.c
>> index 971fbb52740b..cc0f6a97124e 100644
>> --- a/drivers/s390/char/sclp_tty.c
>> +++ b/drivers/s390/char/sclp_tty.c
>> @@ -230,7 +230,7 @@ static int sclp_tty_write_string(const unsigned char *str, int count, int may_fa
>>    * routine will return the number of characters actually accepted for writing.
>>    */
>>   static int
>> -sclp_tty_write(struct tty_struct *tty, const unsigned char *buf, int count)
>> +sclp_tty_write(struct tty_struct *tty, const u8 *buf, int count)
>>   {
>>   	if (sclp_tty_chars_count > 0) {
>>   		sclp_tty_write_string(sclp_tty_chars, sclp_tty_chars_count, 0);
>> diff --git a/drivers/s390/char/sclp_vt220.c b/drivers/s390/char/sclp_vt220.c
>> index a32f34a1c6d2..44974d801c1e 100644
>> --- a/drivers/s390/char/sclp_vt220.c
>> +++ b/drivers/s390/char/sclp_vt220.c
>> @@ -463,7 +463,7 @@ __sclp_vt220_write(const unsigned char *buf, int count, int do_schedule,
>>    * number of characters actually accepted for writing.
>>    */
>>   static int
>> -sclp_vt220_write(struct tty_struct *tty, const unsigned char *buf, int count)
>> +sclp_vt220_write(struct tty_struct *tty, const u8 *buf, int count)
>>   {
>>   	return __sclp_vt220_write(buf, count, 1, 0, 1);
>>   }
> ...
> 
> In this and following patch it is only ::write() (and other) callbacks
> switched to u8, while the called internal functions stay with old types.
> 
> Do you plan to follow with part II etc, or it is archs/drivers that
> are expected to sort it out eventually?

Yes, I would say something like part V. would do the job -- at least 
that's my plan. Slowly propagating from the tty core to leafs (drivers).

> (I speak for s390, but the other affected archs are in the same boat I guess).


thanks,
-- 
js
suse labs