Re: [tip: x86/pasid] x86/asm: Carve out a generic movdir64b() helper for general usage

From: Peter Zijlstra
Date: Wed Oct 07 2020 - 13:10:06 EST


On Wed, Oct 07, 2020 at 04:14:49PM -0000, tip-bot2 for Dave Jiang wrote:
> diff --git a/arch/x86/include/asm/special_insns.h b/arch/x86/include/asm/special_insns.h
> index 59a3e13..d4baa0e 100644
> --- a/arch/x86/include/asm/special_insns.h
> +++ b/arch/x86/include/asm/special_insns.h
> @@ -234,6 +234,28 @@ static inline void clwb(volatile void *__p)
>
> #define nop() asm volatile ("nop")
>
> +/* The dst parameter must be 64-bytes aligned */
> +static inline void movdir64b(void *dst, const void *src)
> +{
> + const struct { char _[64]; } *__src = src;
> + struct { char _[64]; } *__dst = dst;
> +
> + /*
> + * MOVDIR64B %(rdx), rax.

(%rdx), %rax, surely? Also, that's a horrible convention, but I suppose
(%rdx), (%rax) was out?

> + *
> + * Both __src and __dst must be memory constraints in order to tell the
> + * compiler that no other memory accesses should be reordered around
> + * this one.
> + *
> + * Also, both must be supplied as lvalues because this tells
> + * the compiler what the object is (its size) the instruction accesses.
> + * I.e., not the pointers but what they point to, thus the deref'ing '*'.

Can we pretty please get a binutils version that knows about this
instruction, such that we know when we can get rid of the silly .byte
encoded mess?

> + */
> + asm volatile(".byte 0x66, 0x0f, 0x38, 0xf8, 0x02"
> + : "+m" (*__dst)
> + : "m" (*__src), "a" (__dst), "d" (__src));
> +}
> +
> #endif /* __KERNEL__ */
>
> #endif /* _ASM_X86_SPECIAL_INSNS_H */