Re: [PATCH] uprobes: fix scratch register selection for rip-relative fixups

[Oleg Nesterov]

On 05/01, Denys Vlasenko wrote:
>
> Before this patch, instructions such as div, mul,
> shifts with count in CL, cmpxchg are mishandled.

Thanks. I'll try to read this patch tomorrow, but you do know that I can't
review (or even understand ;) the change in riprel_analyze().

As for other changes, perhaps we can cleanup them later somehow, but I am
not going to discuss this now. I agree that the fix should be as simple
as possible.

Only on nit,

> -#define UPROBE_FIX_RIP_AX	0x08
> -#define UPROBE_FIX_RIP_CX	0x10
> +#define UPROBE_FIX_RIP_SI	0x08
> +#define UPROBE_FIX_RIP_DI	0x10
> +#define UPROBE_FIX_RIP_BX	0x20
> +#define UPROBE_FIX_RIP_ALL	0x38

OK, but please define UPROBE_FIX_RIP_ALL as

	#define UPROBE_FIX_RIP_ALL	\
		(UPROBE_FIX_RIP_SI | UPROBE_FIX_RIP_DI | UPROBE_FIX_RIP_BX)

Please do not bother to do this right now, lets wait for review.

Oleg.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/