Re: [PATCH v4 07/19] mtd: rawnand: Compare the actual timing values

[Boris Brezillon <boris.brezillon@xxxxxxxxxxxxx>]

On Mon, 25 May 2020 19:42:27 +0200
Miquel Raynal <miquel.raynal@xxxxxxxxxxx> wrote:

> Avoid relying just on the default timing mode to discriminate if the
> data interface must be restored. Do a memcmp() instead.

Maybe you should explain why you do that.

> 
> Signed-off-by: Miquel Raynal <miquel.raynal@xxxxxxxxxxx>

Reviewed-by: Boris Brezillon <boris.brezillon@xxxxxxxxxxxxx>

> ---
>  drivers/mtd/nand/raw/nand_base.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c
> index 14f1359a60b8..7567c973964b 100644
> --- a/drivers/mtd/nand/raw/nand_base.c
> +++ b/drivers/mtd/nand/raw/nand_base.c
> @@ -2512,7 +2512,8 @@ int nand_reset(struct nand_chip *chip, int chipnr)
>  	 * nand_setup_data_interface() uses ->set/get_features() which would
>  	 * fail anyway as the parameter page is not available yet.
>  	 */
> -	if (!chip->onfi_timing_mode_default)
> +	if (!memcmp(&chip->data_interface, &saved_data_intf,
> +		    sizeof(saved_data_intf)))
>  		return 0;
>  
>  	chip->data_interface = saved_data_intf;

We should probably have the data_interface object allocated at some
point, and play with pointers instead of copying the data around.
That's possible now that you've patched all drivers to use nand_scan()
and control the cleanup path.

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/