Re: [PATCH v2 5/9] mtd: nand: atmel: Report PMECC failures as errors

[Romain Izard]

2017-09-18 12:00 GMT+02:00 Boris Brezillon <boris.brezillon@xxxxxxxxxxxxxxxxxx>:
> Hi Romain,
>
> On Fri, 15 Sep 2017 16:04:07 +0200
> Romain Izard <romain.izard.pro@xxxxxxxxx> wrote:
>
>> It is not normal for the PMECC to fail when trying to fix ECC errors.
>> Report these cases as errors.
>
> I'm not sure we want to have ECC error messages at this level. ECC
> errors are rather unusual but not impossible, and sometimes it's even
> not a real error (I'm thinking of bitflips in erased pages for
> example, which are not necessarily detected/fixed in hardware).
>
> If we decide to print error messages when unfixable bitflips are
> detected, it should be done in the nand-controller driver (somewhere
> along those lines [1]).
>
> Regards,
>
> Boris
>
> [1]http://elixir.free-electrons.com/linux/latest/source/drivers/mtd/nand/atmel/nand-controller.c#L827
>
>>
>> Signed-off-by: Romain Izard <romain.izard.pro@xxxxxxxxx>
>> ---
>>  drivers/mtd/nand/atmel/pmecc.c | 4 +++-
>>  1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/mtd/nand/atmel/pmecc.c b/drivers/mtd/nand/atmel/pmecc.c
>> index 8d1208f38025..2a23f1ff945f 100644
>> --- a/drivers/mtd/nand/atmel/pmecc.c
>> +++ b/drivers/mtd/nand/atmel/pmecc.c
>> @@ -687,6 +687,8 @@ static int atmel_pmecc_err_location(struct atmel_pmecc_user *user)
>>        * Number of roots does not match the degree of smu
>>        * unable to correct error.
>>        */
>> +     dev_err(pmecc->dev,
>> +             "PMECC: Impossible to calculate error location.\n");
>>       return -EBADMSG;
>>  }
>>
>> @@ -729,7 +731,7 @@ int atmel_pmecc_correct_sector(struct atmel_pmecc_user *user, int sector,
>>                       ptr = ecc + byte - sectorsize;
>>                       area = "ECC";
>>               } else {
>> -                     dev_dbg(pmecc->dev,
>> +                     dev_err(pmecc->dev,
>>                               "Invalid errpos value (%d, max is %d)\n",
>>                               errpos, (sectorsize + eccbytes) * 8);
>>                       return -EINVAL;
>

Ok, I will drop this patch.