On 2/12/23 05:47, Ondrej Zary wrote:
> On Saturday 11 February 2023 20:11:06 Sergey Shtylyov wrote:
>> On 2/11/23 5:42 PM, Ondrej Zary wrote:
>>
>>> Only bpck driver uses devtype but it never gets set in pata_parport.
>>> Remove it.
>>>
>>> Signed-off-by: Ondrej Zary <linux@xxxxxxx>
>>> ---
>>> drivers/ata/pata_parport/bpck.c | 2 +-
>>> include/linux/pata_parport.h | 3 ---
>>> 2 files changed, 1 insertion(+), 4 deletions(-)
>>>
>>> diff --git a/drivers/ata/pata_parport/bpck.c b/drivers/ata/pata_parport/bpck.c
>>> index b9174cf8863c..451a068fe28a 100644
>>> --- a/drivers/ata/pata_parport/bpck.c
>>> +++ b/drivers/ata/pata_parport/bpck.c
>>> @@ -241,7 +241,7 @@ static void bpck_connect ( PIA *pi )
>>>
>>> WR(5,8);
>>>
>>> - if (pi->devtype == PI_PCD) {
>>> + if (1 /*pi->devtype == PI_PCD*/) { /* FIXME */
>>> WR(0x46,0x10); /* fiddle with ESS logic ??? */
>>
>> Why not drop this entire *if* stmt?
>
> I decided to keep it (for now) as a marker of a possible bug. I currently don't have HW to test this driver.
Then leave that if as-is and only add a comment detailing what needs to be
done (rather than just "FIXME"). This "if (1)" is just too odd and will
likely trigger code checker warnings.
>
>>
>>> WR(0x4c,0x38);
>>> WR(0x4d,0x88);
>> [...]
>>
>> MBR, Sergey
>>
>
>
--
Damien Le Moal
Western Digital Research
