[PATCH 2/2] mtd/nand: Fix IFC driver to support 2K NAND page

[Scott Wood]

On 12/28/2011 10:59 PM, Prabhakar Kushwaha wrote:
> 1) OOB area should be updated irrespective of NAND page size. Earlier it was
> updated only for 512byte NAND page.
> 
> 2) During OOB update fbcr should be equal to OOB size.
> 
> Signed-off-by: Poonam Aggrwal <poonam.aggrwal at freescale.com>
> Signed-off-by: Prabhakar Kushwaha <prabhakar at freescale.com>
> ---
>  git://git.kernel.org/pub/scm/linux/kernel/git/galak/powerpc.git (branch next)

The IFC driver hasn't been merged into that tree that I can see.

>  Tested on P1010RDB
> 
>  drivers/mtd/nand/fsl_ifc_nand.c |   20 ++++++++------------
>  1 files changed, 8 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c
> index 2df7206..2c02168 100644
> --- a/drivers/mtd/nand/fsl_ifc_nand.c
> +++ b/drivers/mtd/nand/fsl_ifc_nand.c
> @@ -439,20 +439,16 @@ static void fsl_ifc_cmdfunc(struct mtd_info *mtd, unsigned int command,
>  			out_be32(&ifc->ifc_nand.nand_fir1,
>  				 (IFC_FIR_OP_CW1 << IFC_NAND_FIR1_OP5_SHIFT));
>  
> -			if (column >= mtd->writesize) {
> -				/* OOB area --> READOOB */
> -				column -= mtd->writesize;
> -				nand_fcr0 |= NAND_CMD_READOOB <<
> -						IFC_NAND_FCR0_CMD0_SHIFT;
> -				ifc_nand_ctrl->oob = 1;
> -			} else if (column < 256)
> +			if (column < 256)
>  				/* First 256 bytes --> READ0 */
>  				nand_fcr0 |=
>  				NAND_CMD_READ0 << IFC_NAND_FCR0_CMD0_SHIFT;
> -			else
> -				/* Second 256 bytes --> READ1 */
> -				nand_fcr0 |=
> -				NAND_CMD_READ1 << IFC_NAND_FCR0_CMD0_SHIFT;
> +		}
> +
> +		if (column >= mtd->writesize) {
> +			/* OOB area --> READOOB */
> +			column -= mtd->writesize;
> +			ifc_nand_ctrl->oob = 1;
>  		}

Where is NAND_CMD_READOOB going to be set in the small-page case?

The small-page code should read something like:

	if (column >= mtd->writesize) {
		nand_fcr0 |=
			NAND_CMD_READOOB << IFC_NAND_FCR0_CMD0_SHIFT;
	} else {
		nand_fcr0 |=
			NAND_CMD_READ0 << IFC_NAND_FCR0_CMD0_SHIFT;
	}

It looks like we can get rid of ctrl->column, BTW.

-Scott