linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] kprobes/x86: Return correct length in __copy_instruction()
@ 2015-03-09 11:19 Eugene Shatokhin
  2015-03-10  2:59 ` Masami Hiramatsu
  0 siblings, 1 reply; 6+ messages in thread
From: Eugene Shatokhin @ 2015-03-09 11:19 UTC (permalink / raw)
  To: Masami Hiramatsu; +Cc: LKML, Eugene Shatokhin

On x86-64, __copy_instruction() always returns 0 (error) if the
instruction uses %rip-relative addressing. This is because
kernel_insn_init() is called the second time for 'insn' instance
in such cases and sets all its fields to 0.

Because of this, trying to place a Kprobe on such instruction will
fail, register_kprobe() will return -EINVAL.

This patch fixes the problem.

Signed-off-by: Eugene Shatokhin <eugene.shatokhin@rosalab.ru>
---
 arch/x86/kernel/kprobes/core.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
index 4e3d5a9..03189d8 100644
--- a/arch/x86/kernel/kprobes/core.c
+++ b/arch/x86/kernel/kprobes/core.c
@@ -354,6 +354,7 @@ int __copy_instruction(u8 *dest, u8 *src)
 {
 	struct insn insn;
 	kprobe_opcode_t buf[MAX_INSN_SIZE];
+	int length;
 	unsigned long recovered_insn =
 		recover_probed_instruction(buf, (unsigned long)src);
 
@@ -361,16 +362,18 @@ int __copy_instruction(u8 *dest, u8 *src)
 		return 0;
 	kernel_insn_init(&insn, (void *)recovered_insn, MAX_INSN_SIZE);
 	insn_get_length(&insn);
+	length = insn.length;
+
 	/* Another subsystem puts a breakpoint, failed to recover */
 	if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION)
 		return 0;
-	memcpy(dest, insn.kaddr, insn.length);
+	memcpy(dest, insn.kaddr, length);
 
 #ifdef CONFIG_X86_64
 	if (insn_rip_relative(&insn)) {
 		s64 newdisp;
 		u8 *disp;
-		kernel_insn_init(&insn, dest, insn.length);
+		kernel_insn_init(&insn, dest, length);
 		insn_get_displacement(&insn);
 		/*
 		 * The copied instruction uses the %rip-relative addressing
@@ -394,7 +397,7 @@ int __copy_instruction(u8 *dest, u8 *src)
 		*(s32 *) disp = (s32) newdisp;
 	}
 #endif
-	return insn.length;
+	return length;
 }
 
 static int arch_copy_kprobe(struct kprobe *p)
-- 
1.7.11.3


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] kprobes/x86: Return correct length in __copy_instruction()
  2015-03-09 11:19 [PATCH] kprobes/x86: Return correct length in __copy_instruction() Eugene Shatokhin
@ 2015-03-10  2:59 ` Masami Hiramatsu
  2015-03-16 15:03   ` Ingo Molnar
  0 siblings, 1 reply; 6+ messages in thread
From: Masami Hiramatsu @ 2015-03-10  2:59 UTC (permalink / raw)
  To: Eugene Shatokhin, Ingo Molnar; +Cc: LKML, Ingo Molnar

(2015/03/09 20:19), Eugene Shatokhin wrote:
> On x86-64, __copy_instruction() always returns 0 (error) if the
> instruction uses %rip-relative addressing. This is because
> kernel_insn_init() is called the second time for 'insn' instance
> in such cases and sets all its fields to 0.
> 
> Because of this, trying to place a Kprobe on such instruction will
> fail, register_kprobe() will return -EINVAL.

Oops, good catch!

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

Thank you!

> 
> This patch fixes the problem.
> 
> Signed-off-by: Eugene Shatokhin <eugene.shatokhin@rosalab.ru>
> ---
>  arch/x86/kernel/kprobes/core.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
> index 4e3d5a9..03189d8 100644
> --- a/arch/x86/kernel/kprobes/core.c
> +++ b/arch/x86/kernel/kprobes/core.c
> @@ -354,6 +354,7 @@ int __copy_instruction(u8 *dest, u8 *src)
>  {
>  	struct insn insn;
>  	kprobe_opcode_t buf[MAX_INSN_SIZE];
> +	int length;
>  	unsigned long recovered_insn =
>  		recover_probed_instruction(buf, (unsigned long)src);
>  
> @@ -361,16 +362,18 @@ int __copy_instruction(u8 *dest, u8 *src)
>  		return 0;
>  	kernel_insn_init(&insn, (void *)recovered_insn, MAX_INSN_SIZE);
>  	insn_get_length(&insn);
> +	length = insn.length;
> +
>  	/* Another subsystem puts a breakpoint, failed to recover */
>  	if (insn.opcode.bytes[0] == BREAKPOINT_INSTRUCTION)
>  		return 0;
> -	memcpy(dest, insn.kaddr, insn.length);
> +	memcpy(dest, insn.kaddr, length);
>  
>  #ifdef CONFIG_X86_64
>  	if (insn_rip_relative(&insn)) {
>  		s64 newdisp;
>  		u8 *disp;
> -		kernel_insn_init(&insn, dest, insn.length);
> +		kernel_insn_init(&insn, dest, length);
>  		insn_get_displacement(&insn);
>  		/*
>  		 * The copied instruction uses the %rip-relative addressing
> @@ -394,7 +397,7 @@ int __copy_instruction(u8 *dest, u8 *src)
>  		*(s32 *) disp = (s32) newdisp;
>  	}
>  #endif
> -	return insn.length;
> +	return length;
>  }
>  
>  static int arch_copy_kprobe(struct kprobe *p)
> 


-- 
Masami HIRAMATSU
Software Platform Research Dept. Linux Technology Research Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: masami.hiramatsu.pt@hitachi.com



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] kprobes/x86: Return correct length in __copy_instruction()
  2015-03-10  2:59 ` Masami Hiramatsu
@ 2015-03-16 15:03   ` Ingo Molnar
  2015-03-17  9:45     ` Masami Hiramatsu
  0 siblings, 1 reply; 6+ messages in thread
From: Ingo Molnar @ 2015-03-16 15:03 UTC (permalink / raw)
  To: Masami Hiramatsu; +Cc: Eugene Shatokhin, Ingo Molnar, LKML, Ingo Molnar


* Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> wrote:

> (2015/03/09 20:19), Eugene Shatokhin wrote:
> > On x86-64, __copy_instruction() always returns 0 (error) if the
> > instruction uses %rip-relative addressing. This is because
> > kernel_insn_init() is called the second time for 'insn' instance
> > in such cases and sets all its fields to 0.
> > 
> > Because of this, trying to place a Kprobe on such instruction will
> > fail, register_kprobe() will return -EINVAL.
> 
> Oops, good catch!
> 
> Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>

Mind resending it with your SOB, as I wasn't Cc:-ed on the original 
submission?

Thanks,

	Ingo

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] kprobes/x86: Return correct length in __copy_instruction()
  2015-03-16 15:03   ` Ingo Molnar
@ 2015-03-17  9:45     ` Masami Hiramatsu
  2015-04-27 11:03       ` Eugene Shatokhin
  0 siblings, 1 reply; 6+ messages in thread
From: Masami Hiramatsu @ 2015-03-17  9:45 UTC (permalink / raw)
  To: Ingo Molnar; +Cc: Eugene Shatokhin, Ingo Molnar, LKML, Ingo Molnar

(2015/03/17 0:03), Ingo Molnar wrote:
> 
> * Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> wrote:
> 
>> (2015/03/09 20:19), Eugene Shatokhin wrote:
>>> On x86-64, __copy_instruction() always returns 0 (error) if the
>>> instruction uses %rip-relative addressing. This is because
>>> kernel_insn_init() is called the second time for 'insn' instance
>>> in such cases and sets all its fields to 0.
>>>
>>> Because of this, trying to place a Kprobe on such instruction will
>>> fail, register_kprobe() will return -EINVAL.
>>
>> Oops, good catch!
>>
>> Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
> 
> Mind resending it with your SOB, as I wasn't Cc:-ed on the original 
> submission?

Yes, I'll do :)

Thank you,

> 
> Thanks,
> 
> 	Ingo
> 


-- 
Masami HIRAMATSU
Software Platform Research Dept. Linux Technology Research Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: masami.hiramatsu.pt@hitachi.com



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] kprobes/x86: Return correct length in __copy_instruction()
  2015-03-17  9:45     ` Masami Hiramatsu
@ 2015-04-27 11:03       ` Eugene Shatokhin
  2015-05-11 13:47         ` Ingo Molnar
  0 siblings, 1 reply; 6+ messages in thread
From: Eugene Shatokhin @ 2015-04-27 11:03 UTC (permalink / raw)
  To: Masami Hiramatsu, Ingo Molnar; +Cc: Ingo Molnar, LKML, Ingo Molnar

Hi,

Now that the patch is in mainline (commit 
c80e5c0c23ce2282476fdc64c4b5e3d3a40723fd) and kernel 4.1-rc1 is out, do 
you mind if I send the backports of that patch to -stable?

Regards,
Eugene

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] kprobes/x86: Return correct length in __copy_instruction()
  2015-04-27 11:03       ` Eugene Shatokhin
@ 2015-05-11 13:47         ` Ingo Molnar
  0 siblings, 0 replies; 6+ messages in thread
From: Ingo Molnar @ 2015-05-11 13:47 UTC (permalink / raw)
  To: Eugene Shatokhin; +Cc: Masami Hiramatsu, Ingo Molnar, LKML, Ingo Molnar


* Eugene Shatokhin <eugene.shatokhin@rosalab.ru> wrote:

> Hi,
> 
> Now that the patch is in mainline (commit
> c80e5c0c23ce2282476fdc64c4b5e3d3a40723fd) and kernel 4.1-rc1 is out, do you
> mind if I send the backports of that patch to -stable?

No objections from me!

Thanks,

	Ingo

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2015-05-11 13:48 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-03-09 11:19 [PATCH] kprobes/x86: Return correct length in __copy_instruction() Eugene Shatokhin
2015-03-10  2:59 ` Masami Hiramatsu
2015-03-16 15:03   ` Ingo Molnar
2015-03-17  9:45     ` Masami Hiramatsu
2015-04-27 11:03       ` Eugene Shatokhin
2015-05-11 13:47         ` Ingo Molnar

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).