Re: [PATCH v5 02/15] drm/edid: Break out reading block 0 of the EDID

[Geert Uytterhoeven]

Hi Doug,

On Mon, Oct 4, 2021 at 6:26 PM Doug Anderson <dianders@xxxxxxxxxxxx> wrote:
> On Mon, Oct 4, 2021 at 8:42 AM Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> > > -       if ((edid = kmalloc(EDID_LENGTH, GFP_KERNEL)) == NULL)
> > > +       edid = (u8 *)drm_do_get_edid_base_block(get_edid_block, data,
> > > +                                               &connector->edid_corrupt,
> > > +                                               &connector->null_edid_counter);
> > > +       if (IS_ERR_OR_NULL(edid)) {
> > > +               if (IS_ERR(edid))
> >
> > So edid is an error code, not a valid pointer...
> >
> > > +                       connector_bad_edid(connector, edid, 1);
> >
> > ... while connector_bad_edid() expects edid to be a valid pointer,
> > causing a crash:
> >
> > Unable to handle kernel NULL pointer dereference at virtual address
>
> Sigh. Thanks for the report and analysis. I guess I don't have any
> displays reporting invalid EDIDs to test with. Hopefully this will
> help:

It doesn't happen all the time.  Looks like my EDID is only invalid after
a reset needed to resolve an s2ram crash in the adv7511 driver...

> https://lore.kernel.org/r/20211004092100.1.Ic90a5ebd44c75db963112be167a03cc96f9fb249@changeid/

Thanks for the quick fix!

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds