Re: Problems with jffs2 rootfs using 2.6.28-rc3

[Tony Lindgren <tony@atomide.com>]

* Steve Sakoman <sakoman@gmail.com> [081107 14:20]:
> On Fri, Nov 7, 2008 at 1:30 PM, David Brownell <david-b@pacbell.net> wrote:
> > On Friday 07 November 2008, Steve Sakoman wrote:
> >> On Fri, Nov 7, 2008 at 12:46 PM, David Brownell <david-b@pacbell.net> wrote:
> >> > On Friday 07 November 2008, Steve Sakoman wrote:
> >> >> I'm having little success.  It used to work back in 2.6.26 timeframe
> >> >> but seems to have broken sometime during 2.6.27.
> >> >
> >> > There's a patch floating around making the omap2 NAND driver
> >> > use readsw not readsl, which might explain that problem...
> >>
> >> Do you recall where you saw that patch?  Here or linux-mtd?  My
> >> initial search efforts weren't successful.
> >
> > On linux-omap.  Or just look at the driver to see the
> > readsl()... it's a one-liner.
> 
> Yes, that one line change did indeed fix the issue!
> 
> Is that patch currently in the queue somewhere?

Grr, this should be sent via the MTD list! But the driver is not there,
so pushing Juha's patch to l-o today.

Tony

> For the record:
> 
> diff --git a/drivers/mtd/nand/omap2.c b/drivers/mtd/nand/omap2.c
> index 2ede116..d18a8c9 100644
> --- a/drivers/mtd/nand/omap2.c
> +++ b/drivers/mtd/nand/omap2.c
> @@ -196,7 +196,7 @@ static void omap_read_buf16(struct mtd_info *mtd,
> u_char *buf, int len)
>  {
>  	struct nand_chip *nand = mtd->priv;
> 
> -	__raw_readsl(nand->IO_ADDR_R, buf, len / 2);
> +	readsw(nand->IO_ADDR_R, buf, len / 2);
>  }
> 
>  /*