Re: [PATCH 10/82] locking/atomic/x86: Silence intentional wrapping addition

[Mark Rutland]

On Mon, Jan 22, 2024 at 04:26:45PM -0800, Kees Cook wrote:
> Annotate atomic_add_return() to avoid signed overflow instrumentation.
> It is expected to wrap around.
> 
> Cc: Will Deacon <will@xxxxxxxxxx>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Cc: Boqun Feng <boqun.feng@xxxxxxxxx>
> Cc: Mark Rutland <mark.rutland@xxxxxxx>
> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Borislav Petkov <bp@xxxxxxxxx>
> Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
> Cc: x86@xxxxxxxxxx
> Cc: "H. Peter Anvin" <hpa@xxxxxxxxx>
> Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx>
> ---
>  arch/x86/include/asm/atomic.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/include/asm/atomic.h b/arch/x86/include/asm/atomic.h
> index 55a55ec04350..4120cdd87da8 100644
> --- a/arch/x86/include/asm/atomic.h
> +++ b/arch/x86/include/asm/atomic.h
> @@ -80,7 +80,7 @@ static __always_inline bool arch_atomic_add_negative(int i, atomic_t *v)
>  }
>  #define arch_atomic_add_negative arch_atomic_add_negative
>  
> -static __always_inline int arch_atomic_add_return(int i, atomic_t *v)
> +static __always_inline __signed_wrap int arch_atomic_add_return(int i, atomic_t *v)
>  {
>  	return i + xadd(&v->counter, i);
>  }

I think that here (and in the arm64 patch) it'd be better to use add_wrap() on
the specific statement, i.e. have:

static __always_inline int arch_atomic_add_return(int i, atomic_t *v)
{
	return add_wrap(i, xadd(&v->counter, i));
}

.. since otherwise the annotation could applly to the '+' or something else
(e.g. if the 'xadd() part is a special macro), and the annotation might
unexpectedly hide things if we add other statements here in future.

Mark.

> -- 
> 2.34.1
>