[PATCH v3 07/10] mtd: IFC NAND: utilize oob_required parameter

[Scott Wood]

On 04/30/2012 02:08 PM, Brian Norris wrote:
> On Mon, Apr 30, 2012 at 9:43 AM, Scott Wood <scottwood at freescale.com> wrote:
>> On 04/27/2012 08:29 PM, Brian Norris wrote:
>>> @@ -717,7 +718,8 @@ static void fsl_ifc_write_page(struct mtd_info *mtd, struct nand_chip *chip,
>>>                              const uint8_t *buf, int oob_required)
>>>  {
>>>       fsl_ifc_write_buf(mtd, buf, mtd->writesize);
>>> -     fsl_ifc_write_buf(mtd, chip->oob_poi, mtd->oobsize);
>>> +     if (oob_required)
>>> +             fsl_ifc_write_buf(mtd, chip->oob_poi, mtd->oobsize);
>>>  }
>>
>> This will result in writing junk to the non-ECC OOB bytes as opposed to
>> leaving it alone.
> 
> Then I'll drop the write_page change from this patch.
> 
> Is the read_page change sane?

It should be harmless.

> Did you review the (misspelled) eLBC patch? (patch 06/10)

No, that one wasn't CCed to me (I should probably get around to
subscribing to linux-mtd...).  It looks like the same situation as IFC.

-Scott