[PATCH v4 4/4] arm64: lto: Strengthen READ_ONCE() to acquire when CONFIG_LTO=y

[Mark Rutland]

On Tue, Nov 03, 2020 at 12:17:21PM +0000, Will Deacon wrote:
> When building with LTO, there is an increased risk of the compiler
> converting an address dependency headed by a READ_ONCE() invocation
> into a control dependency and consequently allowing for harmful
> reordering by the CPU.
> 
> Ensure that such transformations are harmless by overriding the generic
> READ_ONCE() definition with one that provides acquire semantics when
> building with LTO.
> 
> Acked-by: Peter Zijlstra (Intel) <peterz at infradead.org>
> Signed-off-by: Will Deacon <will at kernel.org>

[...]

Could we add a note above __READ_ONCE() along the lines of the commit
message, e.g.

/*
 * With LTO a compiler might convert an address dependency headed by a
 * READ_ONCE() into a control dependency, allowing for harmful
 * reordering by the CPU.
 *
 * To prevent this, upgrade READ_OONCE() to provide acquire semantics
 * when building with LTO.
 */

Either way:

Acked-by: Mark Rutland <mark.rutland at arm.com>

Mark

> +#define __READ_ONCE(x)							\
> +({									\
> +	typeof(&(x)) __x = &(x);					\
> +	int atomic = 1;							\
> +	union { __unqual_scalar_typeof(*__x) __val; char __c[1]; } __u;	\
> +	switch (sizeof(x)) {						\
> +	case 1:								\
> +		asm volatile(__LOAD_RCPC(b, %w0, %1)			\
> +			: "=r" (*(__u8 *)__u.__c)			\
> +			: "Q" (*__x) : "memory");			\
> +		break;							\
> +	case 2:								\
> +		asm volatile(__LOAD_RCPC(h, %w0, %1)			\
> +			: "=r" (*(__u16 *)__u.__c)			\
> +			: "Q" (*__x) : "memory");			\
> +		break;							\
> +	case 4:								\
> +		asm volatile(__LOAD_RCPC(, %w0, %1)			\
> +			: "=r" (*(__u32 *)__u.__c)			\
> +			: "Q" (*__x) : "memory");			\
> +		break;							\
> +	case 8:								\
> +		asm volatile(__LOAD_RCPC(, %0, %1)			\
> +			: "=r" (*(__u64 *)__u.__c)			\
> +			: "Q" (*__x) : "memory");			\
> +		break;							\
> +	default:							\
> +		atomic = 0;						\
> +	}								\
> +	atomic ? (typeof(*__x))__u.__val : (*(volatile typeof(__x))__x);\
> +})