Re: [PATCH v8 1/1] gpio: add sloppy logic analyzer using polling

From: Andy Shevchenko
Date: Fri Jun 02 2023 - 10:07:24 EST


On Fri, Jun 02, 2023 at 08:57:36AM +0200, Geert Uytterhoeven wrote:
> On Fri, Jun 2, 2023 at 8:51 AM Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> > On Thu, Jun 1, 2023 at 11:40 PM <andy.shevchenko@xxxxxxxxx> wrote:
> > > Tue, Mar 29, 2022 at 11:11:26AM +0200, Wolfram Sang kirjoitti:

...

> > > However, I haven't checked if it's pure theoretical issue with the current code
> > > base of debugfs or a potential problem. Easy fix is to check an error code and
> >
> > I think debugfs_create_dir() can only fail reasonably due to OOM.
>
> Oops, you were talking about NULL, not an error code.
> I don't think that can ever happen.

That's looks correct, I don't see (looking briefly into the fs/debugfs) that
NULL is possible to be returned. Wondering, why the heck we (used to) have
a lot of IS_ERR_OR_NULL() checks.

> And if I did miss something, it would crash when dereferencing a NULL
> pointer in d_really_is_positive() (as called in start_creating())...

I don't see how, but I see a big comment on top of `if (!parent)`.

--
With Best Regards,
Andy Shevchenko