Hi Stefan,
>
> (Added Niel to Cc)
>
> On 05.02.21 10:55, Holger Brunck wrote:
> > The pram variable is evaluated in common/board_r.c as a unsigned long
> > and not as a string using a hex formataion. This incorrect setting
> > leads to a wrong calculation of preserved RAM for u-boot. This is
> > fixed now. The pram variable is now a unsigned long specifying the
> > amount of kB of preserved RAM.
> >
> > Signed-off-by: Holger Brunck <holger.bru...@hitachi-powergrids.com>
> > CC: Stefan Roese <s...@denx.de>
> > ---
> > board/keymile/common/common.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/board/keymile/common/common.c
> > b/board/keymile/common/common.c index df507e2..7343d47 100644
> > --- a/board/keymile/common/common.c
> > +++ b/board/keymile/common/common.c
> > @@ -60,7 +60,7 @@ int set_km_env(void)
> > strict_strtoul(p, 16, &rootfssize);
> > pram = (rootfssize + CONFIG_KM_RESERVED_PRAM +
> CONFIG_KM_PHRAM +
> > CONFIG_KM_PNVRAM) / 0x400;
> > - sprintf((char *)buf, "0x%x", pram);
> > + sprintf((char *)buf, "%u", pram);
> > env_set("pram", (char *)buf);
>
> I've commented to Niel as well before:
>
> Why don't you switch to using a different env_set_foo() API instead?
> Like env_set_ulong() in this case? Or env_set_hex() in the HEX case.
> This could also be done for some other env_set cases in this file as well to
> reduce code size and complexity.
>
You are right. And I just saw now that Niel is already doing this in
[v2,2/3] keymile: common: update to set_env_hex(), fix "pram" radix
which is still under review.
So this patch can be abandoned.
Best regards
Holger
