Re: [PATCH v2 1/2] Provide READ_ONCE_NOCHECK()

[Ingo Molnar <mingo@kernel.org>]

* Andrey Ryabinin <aryabinin@virtuozzo.com> wrote:

> +#define __READ_ONCE_SIZE						\
> +({									\
> +	switch (size) {							\
> +	case 1: *(__u8 *)res = *(volatile __u8 *)p; break;		\
> +	case 2: *(__u16 *)res = *(volatile __u16 *)p; break;		\
> +	case 4: *(__u32 *)res = *(volatile __u32 *)p; break;		\
> +	case 8: *(__u64 *)res = *(volatile __u64 *)p; break;		\
> +	default:							\
> +		barrier();						\
> +		__builtin_memcpy((void *)res, (const void *)p, size);	\
> +		barrier();						\
> +	}								\
> +})
> +
> +static __always_inline
> +void __read_once_size_check(const volatile void *p, void *res, int size)
>  {
> +	__READ_ONCE_SIZE;
>  }

> +#ifdef CONFIG_KASAN
> +static __no_sanitize_address __maybe_unused
> +void __read_once_size_nocheck(const volatile void *p, void *res, int size)
> +{
> +	__READ_ONCE_SIZE;
> +}
> +#else
> +static __always_inline __alias(__read_once_size_check)
> +void __read_once_size_nocheck(const volatile void *p, void *res, int size);
> +#endif

> +#define __READ_ONCE(x, check)				\
> +({								\
> +	union { typeof(x) __val; char __c[1]; } __u;		\
> +	__read_once_size##check(&(x), __u.__c, sizeof(x));	\
> +	__u.__val;						\
> +})
> +#define READ_ONCE(x) __READ_ONCE(x, _check)
> +#define READ_ONCE_NOCHECK(x) __READ_ONCE(x, _nocheck)

So all this code is quite convoluted, and as the changelog explains it's done to 
work around a GCC inlining + no-kasan bug.

But please explain this in the code as well, in a comment, so future generations 
are not kept wondering why these relatively simple wrappers are coded in such an 
ugly and roundabout fashion.

Thanks,

	Ingo