[PATCH v7 1/2] mtd: nand: add sunxi NAND flash controller support

[boris.brezillon@free-electrons.com (Boris Brezillon)]

Hi Brian,

On Tue, 28 Oct 2014 11:13:11 -0700
Brian Norris <computersforpeace@gmail.com> wrote:

> Hi Boris,
> 
> On Tue, Oct 21, 2014 at 03:08:41PM +0200, Boris Brezillon wrote:
> > +static int sunxi_nand_hw_ecc_ctrl_init(struct mtd_info *mtd,
> > +				       struct nand_ecc_ctrl *ecc,
> > +				       struct device_node *np)
> > +{
> > +	struct nand_ecclayout *layout;
> > +	int i, j;
> > +	int ret;
> > +
> > +	ret = sunxi_nand_hw_common_ecc_ctrl_init(mtd, ecc, np);
> > +	if (ret)
> > +		return ret;
> > +
> > +	ecc->read_page = sunxi_nfc_hw_ecc_read_page;
> > +	ecc->write_page = sunxi_nfc_hw_ecc_write_page;
> > +	layout = ecc->layout;
> > +
> > +	for (i = 0; i < ecc->steps; i++) {
> 
> Hmm, did you retest this since changing this to ecc->steps? I think this
> won't work, since ecc->steps is only initialized in nand_scan_tail(),
> which comes after this. I only recommended the change for the
> {read,write}_page() type of functions, not the initialization functions.

Oh, crap!
Indeed, I only checked that the NAND was properly detected, but didn't
test all the features.
Anyway, I'm glad you found this issue before applying the series.

> 
> [snip the rest]
> 
> So I'd suggest the following additional patch, and otherwise, the series
> is fine with me. If you give your ack, I can just squash this in and
> apply it to my tree:

I'll test it tomorrow (and this time I'll test everything ;-)).

Thanks,

Boris

-- 
Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com