Re: [PATCH] cmd: pxe_utils: Check fdtcontroladdr in label_boot

[Ramon Fried <rfried.dev@gmail.com>]

On Sun, Oct 24, 2021 at 10:56 PM Simon Glass <sjg@chromium.org> wrote:
>
> Hi,
>
> On Fri, 15 Oct 2021 at 05:21, Andre Przywara <andre.przywara@arm.com> wrote:
> >
> > On Thu, 14 Oct 2021 12:24:45 -0600
> > Simon Glass <sjg@chromium.org> wrote:
> >
> > Hi Simon,
> >
> > > On Thu, 14 Oct 2021 at 02:40, Peter Hoyes <peter.hoyes@arm.com> wrote:
> > > >
> > > > From: Peter Hoyes <Peter.Hoyes@arm.com>
> > > >
> > > > If using OF_CONTROL, fdtcontroladdr is set to the fdt used to configure
> > > > U-Boot. When using PXE, if no fdt is defined in the menu file, and
> > > > there is no fdt at fdt_addr, add fall back on fdtcontroladdr too.
> > > >
> > > > We are developing board support for the Armv8r64 FVP using
> > > > config_distro_bootcmd. We are also using OF_BOARD and would like the
> > > > PXE boot option to default to the fdt provided by board_fdt_blob_setup.
> > > >
> > > > Signed-off-by: Peter Hoyes <Peter.Hoyes@arm.com>
> > > > ---
> > > >  cmd/pxe_utils.c | 8 +++++++-
> > > >  1 file changed, 7 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/cmd/pxe_utils.c b/cmd/pxe_utils.c
> > > > index 067c24e5ff..8f8e69ca97 100644
> > > > --- a/cmd/pxe_utils.c
> > > > +++ b/cmd/pxe_utils.c
> > > > @@ -556,7 +556,10 @@ static int label_boot(struct cmd_tbl *cmdtp, struct pxe_label *label)
> > > >          * Scenario 2: If there is an fdt_addr specified, pass it along to
> > > >          * bootm, and adjust argc appropriately.
> > > >          *
> > > > -        * Scenario 3: fdt blob is not available.
> > > > +        * Scenario 3: If there is an fdtcontroladdr specified, pass it along to
> > > > +        * bootm, and adjust argc appropriately.
> > > > +        *
> > > > +        * Scenario 4: fdt blob is not available.
> > > >          */
> > > >         bootm_argv[3] = env_get("fdt_addr_r");
> > > >
> > > > @@ -646,6 +649,9 @@ static int label_boot(struct cmd_tbl *cmdtp, struct pxe_label *label)
> > > >         if (!bootm_argv[3])
> > > >                 bootm_argv[3] = env_get("fdt_addr");
> > > >
> > > > +       if (!bootm_argv[3])
> > > > +               bootm_argv[3] = env_get("fdtcontroladdr");
> > > > +
> > > >         if (bootm_argv[3]) {
> > > >                 if (!bootm_argv[2])
> > > >                         bootm_argv[2] = "-";
> > > > --
> > > > 2.25.1
> > > >
> > >
> > > This is a bit strange as fdtcontroladdr is the control dtb that U-Boot
> > > uses.
> >
> > Yes, that is the idea. This DTB should be as good as any, and this is the
> > last check before giving up on DTs completely, so "U-Boot's DTB" (actually
> > there is no such thing) is better than none at all. In fact EFI boot does
> > the very same thing:
> > https://source.denx.de/u-boot/u-boot/-/blob/master/include/config_distro_bootcmd.h#L144-L148
> >
> > > Can you not set fdt_addr to that, if you need to? Is it not
> > > already set to that?
> >
> > According to doc/README.distro, fdt_addr is reserved for DTBs residing at
> > a (fixed?) ROM address. We don't really know yet where the DT comes from,
> > the cleanest seems to be to be passed down from TF-A, in which case this
> > would definitely not be ROM. board_fdt_blob_setup() knows this address,
> > and this is what will be written to $fdtcontroladdr.
> >
> > FWIW, this patch looks good to me.
>
> OK.
>
> But note that OF_BOARD is going to become an option, not something
> that let's us leave out the DT. Board vendors must provide a DT in
> U-Boot. So long as that is understood:
>
> Reviewed-by: Simon Glass <sjg@chromium.org>
>
> Regards,
> Simom
applied to u-boot-net/next
Thanks,
Ramon.