[CRIU] [PATCH 4/4] x86/compat: provide optimized memcpy_x86

[Pavel Emelyanov]

On 11/22/2016 02:42 PM, Dmitry Safonov wrote:
> Implementation for PIE's builtin_memcpy located at
> criu/arch/x86/include/asm/string.h
> 
> Signed-off-by: Dmitry Safonov <dsafonov at virtuozzo.com>
> ---
>  criu/arch/x86/compat/memcpy-compat.c | 25 +++++++++++++++++++++++++
>  1 file changed, 25 insertions(+)
>  create mode 100644 criu/arch/x86/compat/memcpy-compat.c
> 
> diff --git a/criu/arch/x86/compat/memcpy-compat.c b/criu/arch/x86/compat/memcpy-compat.c
> new file mode 100644
> index 000000000000..793c27dcaa7d
> --- /dev/null
> +++ b/criu/arch/x86/compat/memcpy-compat.c

Another memcpy for pie? Shouldn't they all be in compel?

> @@ -0,0 +1,25 @@
> +#include <stddef.h>
> +
> +/*
> + * This provides an optimized implementation of memcpy, and a simplified
> + * implementation of memset and memmove. These are used here because the
> + * standard kernel runtime versions are not yet available and we don't
> + * trust the gcc built-in implementations as they may do unexpected things
> + * (e.g. FPU ops) in the minimal decompression stub execution environment.
> + *
> + * From Linux kernel boot helpers: arch/x86/boot/compressed/string.c
> + */
> +
> +void *memcpy_x86(void *dest, const void *src, size_t n)
> +{
> +	int d0, d1, d2;
> +	asm volatile(
> +		"rep ; movsl\n\t"
> +		"movl %4,%%ecx\n\t"
> +		"rep ; movsb\n\t"
> +		: "=&c" (d0), "=&D" (d1), "=&S" (d2)
> +		: "0" (n >> 2), "g" (n & 3), "1" (dest), "2" (src)
> +		: "memory");
> +
> +	return dest;
> +}
>