Re: [for-linus][PATCH 01/10] tracing: Suppress sparse warnings triggered by is_signed_type()

[Linus Torvalds]

On Wed, Aug 24, 2022 at 4:28 PM Bart Van Assche <bvanassche@xxxxxxx> wrote:
>
> Can you take a look at the following report from the kernel test robot:
> https://lore.kernel.org/all/202208250433.EVz5pZkB-lkp@xxxxxxxxx/ ?
>
> Do I see correctly that gcc reports a new warning for the above
> definition of is_signed_type() with W=1? I'm not sure how to get rid of
> that new gcc warning without reintroducing a sparse warning.

Indeed. It looks like now gcc recognizes it as a NULL pointer, and
then special-cases that and warns for it.

Oh, how very annoying.

Let me think about this.

               Linus