[PATCH v2] bootm: Fix bmi->images pointer not initialized in some cases

[Simon Glass]

Hi Tom,

On Fri, 23 May 2025 at 08:08, Tom Rini <trini at konsulko.com> wrote:
>
> On Fri, May 23, 2025 at 02:48:15PM +0100, Simon Glass wrote:
> > Hi Benjamin,
> >
> > On Tue, 20 May 2025 at 22:35, Benjamin ROBIN <dev at benjarobin.fr> wrote:
> > >
> > > When building with only bootz command, without bootm, images pointer
> > > inside bootm_info structure is not initialized. And since this structure
> > > is stored in stack, the generated error is kind of random, but most of
> > > the time this will generate: "ramdisk - allocation error".
> > >
> > > Also, after analysis, this problem could occur with the command booti,
> > > if the command bootm is disabled.
> > >
> > > Currently bootm_init() is called by: do_bootz(), do_bootm(), do_booti()
> > > and by do_stm32prog(). And all of these commands execute bootm_run_states()
> > > which access the images pointer stored into bootm_info structure.
> > >
> > > So, to fix this issue, just do the assignment unconditionally.
> > >
> > > Fixes: c2211ff65136 ("bootm: Add more fields to bootm_info")
> > > Signed-off-by: Benjamin ROBIN <dev at benjarobin.fr>
> > > ---
> > >  boot/bootm.c | 3 +--
> > >  1 file changed, 1 insertion(+), 2 deletions(-)
> > >
> > > diff --git a/boot/bootm.c b/boot/bootm.c
> > > index f6aa32746b7e..108ca7fb472e 100644
> > > --- a/boot/bootm.c
> > > +++ b/boot/bootm.c
> > > @@ -1169,8 +1169,7 @@ void bootm_init(struct bootm_info *bmi)
> > >  {
> > >         memset(bmi, '\0', sizeof(struct bootm_info));
> > >         bmi->boot_progress = true;
> > > -       if (IS_ENABLED(CONFIG_CMD_BOOTM))
> > > -               bmi->images = &images;
> > > +       bmi->images = &images;
> >
> > In my tree I have:
> >
> > if (IS_ENABLED(CONFIG_CMD_BOOTM) || IS_ENABLED(CONFIG_CMD_BOOTZ) ||
> >     IS_ENABLED(CONFIG_CMD_BOOTI) || IS_ENABLED(CONFIG_PXE_UTILS))
> >
> > but it seems that the few boards that didn't include boot/bootm.c have
> > been dropped, so your patch is fine.
>
> Yes, looking at the code, we should have always been doing the
> assignment as there's a few places where we reference bmi->images and
> not the global images, FYI.

My intent is to eventually drop the global, but that is a ways off.

Regards,
Simon