Re: [PATCH mm] kfence: fix printk format for ptrdiff_t

[Christophe Leroy]

Le 16/03/2021 à 16:33, Segher Boessenkool a écrit :
> On Tue, Mar 16, 2021 at 09:32:32AM +0100, Christophe Leroy wrote:
> > +segher
>
> I cannot see through the wood of #defines here, sorry.
>
> > Still a problem.
> >
> > I don't understand, gcc bug ?
>
> Rule #1: If you do not understand what is happening, it is not a
> compiler bug.  I'm not saying that it isn't, just that it is much more
> likely something else.
>
> > The offending argument is 'const ptrdiff_t object_index'
> >
> > We have:
> >
> > arch/powerpc/include/uapi/asm/posix_types.h:typedef long	
> > __kernel_ptrdiff_t;
>
> So this is a 64-bit build.

No it's 32 bits. The code in posix-types.h is

#ifdef __powerpc64__
...
#else
...
typedef long		__kernel_ptrdiff_t;


> > include/linux/types.h:typedef __kernel_ptrdiff_t	ptrdiff_t;
> >
> > And get:
> >
> >    CC      mm/kfence/report.o
> > In file included from ./include/linux/printk.h:7,
> >                   from ./include/linux/kernel.h:16,
> >                   from mm/kfence/report.c:10:
> > mm/kfence/report.c: In function 'kfence_report_error':
> > ./include/linux/kern_levels.h:5:18: warning: format '%td' expects argument
> > of type 'ptrdiff_t', but argument 6 has type 'long int' [-Wformat=]
>
> This is declared as
>          const ptrdiff_t object_index = meta ? meta - kfence_metadata : -1;
> so maybe something with that goes wrong?  What happens if you delete the
> (useless) "const" here?

No change.

Christophe