Re: [PATCH 1/2] kprobes/x86: boost: Fix checking if there is enough room for a jump

[Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>]

On 2015/06/02 1:32, Eugene Shatokhin wrote:
> Kprobes' "boost" feature allows to avoid single-stepping in some cases,
> along with its overhead. It needs a relative jump placed in the insn
> slot right after the instruction. So the length of the instruction plus
> 5 (the length of the near relative unconditional jump) must not exceed
> the length of the slot.
> 
> However, the code currently checks if that sum is strictly less than
> the length of the slot. So "boost" cannot be used for the instructions
> of 10 bytes in size (with MAX_INSN_SIZE == 15), like
> "movabs $0x45525f4b434f4c43,%rax" (48 b8 43 4c 4f 43 4b 5f 52 45),
> "movl $0x1,0xf8dd(%rip)"          (c7 05 dd f8 00 00 01 00 00 00), etc.
> 
> This patch fixes that conditional.

Looks good to me,

Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>

Thanks!

> 
> Signed-off-by: Eugene Shatokhin <eugene.shatokhin@rosalab.ru>
> ---
>  arch/x86/kernel/kprobes/core.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
> index 1deffe6..0a42b76 100644
> --- a/arch/x86/kernel/kprobes/core.c
> +++ b/arch/x86/kernel/kprobes/core.c
> @@ -881,7 +881,7 @@ static void resume_execution(struct kprobe *p, struct pt_regs *regs,
>  
>  	if (p->ainsn.boostable == 0) {
>  		if ((regs->ip > copy_ip) &&
> -		    (regs->ip - copy_ip) + 5 < MAX_INSN_SIZE) {
> +		    (regs->ip - copy_ip) + 5 <= MAX_INSN_SIZE) {
>  			/*
>  			 * These instructions can be executed directly if it
>  			 * jumps back to correct address.
> 


-- 
Masami HIRAMATSU
Linux Technology Research Center, System Productivity Research Dept.
Center for Technology Innovation - Systems Engineering
Hitachi, Ltd., Research & Development Group
E-mail: masami.hiramatsu.pt@hitachi.com