Re: [PATCH 3/8] mtd: rawnand: mtk: Add DT property mtk,fdm-ecc-size

[xiaolei li <xiaolei.li-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>]

Hi Boris,

On Wed, 2018-04-11 at 21:13 +0200, Boris Brezillon wrote:
> On Wed, 11 Apr 2018 11:41:53 +0800
> Xiaolei Li <xiaolei.li-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org> wrote:
> 
> > For some MTK NAND chips, BootROM may access more than one byte
> > ECC protected FDM data, but now we fix ECC protected FDM byte as 1.
> > This will make some chips be failed to boot up.
> > 
> > With this DT property setting, different MTK NAND chips with the same
> > NAND controller IP can work well.
> > 
> > Signed-off-by: Xiaolei Li <xiaolei.li-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
> > ---
> >  Documentation/devicetree/bindings/mtd/mtk-nand.txt |  6 ++++++
> >  drivers/mtd/nand/raw/mtk_nand.c                    | 25 ++++++++++++++++------
> >  2 files changed, 24 insertions(+), 7 deletions(-)
> > 
> > diff --git a/Documentation/devicetree/bindings/mtd/mtk-nand.txt b/Documentation/devicetree/bindings/mtd/mtk-nand.txt
> > index ef786568..a8e4136 100644
> > --- a/Documentation/devicetree/bindings/mtd/mtk-nand.txt
> > +++ b/Documentation/devicetree/bindings/mtd/mtk-nand.txt
> > @@ -47,6 +47,12 @@ Children nodes properties:
> >  - reg:			Chip Select Signal, default 0.
> >  			Set as reg = <0>, <1> when need 2 CS.
> >  Optional:
> > +- mtk,fdm-ecc-size:	Integer representing ECC protected FDM bytes.
> > +			Should be in the range [1,8], if not present 1.
> > +			On some MTK NAND chips, BootROM may access more than
> > +			one byte ECC protected FDM data. Different MTK chips
> > +			with the same NAND controller IP will work well with
> > +			this properity setting.
> 
> Is this something that changes on a per-SoC basis, or can a specific
> SoC have a different behavior depending on the version of the BootROM
> it embeds (that would be quite tricky to deal with since that would
> mean having different dts if you start using newer revisions of the same
> SoC).
> 
> If it's on a per-SoC basis, I'd recommend defining new compatibles
> instead of adding this property.
> 
Thanks for your advice. I will change to add new compatibles for this
setting next patch.

Thanks,
Xiaolei
>