linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH -tip 4/6 V4] x86: kprobes checks safeness of insertion address.
@ 2009-04-02 17:24 Masami Hiramatsu
  2009-04-03  5:18 ` Ananth N Mavinakayanahalli
  0 siblings, 1 reply; 6+ messages in thread
From: Masami Hiramatsu @ 2009-04-02 17:24 UTC (permalink / raw)
  To: Ingo Molnar, Ananth N Mavinakayanahalli, Jim Keniston
  Cc: LKML, systemtap-ml, kvm, Andi Kleen, Andrew Morton,
	Arnaldo Carvalho de Melo, Steven Rostedt, Frederic Weisbecker

Ensure safeness of inserting kprobes by checking whether the specified
address is at the first byte of a instruction. This is done by decoding
probed function from its head to the probe point.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
---

 arch/x86/kernel/kprobes.c |   51 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 51 insertions(+), 0 deletions(-)


diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 7b5169d..39c79cc 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -48,12 +48,14 @@
 #include <linux/preempt.h>
 #include <linux/module.h>
 #include <linux/kdebug.h>
+#include <linux/kallsyms.h>

 #include <asm/cacheflush.h>
 #include <asm/desc.h>
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 #include <asm/alternative.h>
+#include <asm/insn.h>

 void jprobe_return_end(void);

@@ -244,6 +246,53 @@ retry:
 	}
 }

+/* Recover original instruction */
+static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
+{
+	struct kprobe *kp;
+	kp = get_kprobe((void *)addr);
+	if (!kp)
+		return -EINVAL;
+
+	/* Don't use p->ainsn.insn; which will be modified by fix_riprel */
+	memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	buf[0] = kp->opcode;
+	return 0;
+}
+
+/* Dummy buffers for lookup_symbol_attrs */
+static char __dummy_buf[KSYM_NAME_LEN];
+
+/* Check whether the address can be probed */
+static int __kprobes can_probe(unsigned long paddr)
+{
+	int ret;
+	unsigned long addr, offset = 0;
+	struct insn insn;
+	kprobe_opcode_t buf[MAX_INSN_SIZE];
+
+	/* Lookup symbol including addr */
+	if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
+		return 0;
+
+	/* Decode instructions */
+	addr = paddr - offset;
+	while (addr < paddr) {
+		insn_init_kernel(&insn, (void *)addr);
+		insn_get_opcode(&insn);
+		if (OPCODE1(&insn) == BREAKPOINT_INSTRUCTION) {
+			ret = recover_probed_instruction(buf, addr);
+			if (ret)
+				return 0;
+			insn_init_kernel(&insn, buf);
+		}
+		insn_get_length(&insn);
+		addr += insn.length;
+	}
+
+	return (addr == paddr);
+}
+
 /*
  * Returns non-zero if opcode modifies the interrupt flag.
  */
@@ -359,6 +408,8 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p)

 int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
+	if (!can_probe((unsigned long)p->addr))
+		return -EILSEQ;
 	/* insn: must be on special executable page on x86. */
 	p->ainsn.insn = get_insn_slot();
 	if (!p->ainsn.insn)
-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com


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

* Re: [PATCH -tip 4/6 V4] x86: kprobes checks safeness of insertion address.
  2009-04-02 17:24 [PATCH -tip 4/6 V4] x86: kprobes checks safeness of insertion address Masami Hiramatsu
@ 2009-04-03  5:18 ` Ananth N Mavinakayanahalli
  2009-04-03 16:02   ` [PATCH -tip 4/6 V4.1] " Masami Hiramatsu
  0 siblings, 1 reply; 6+ messages in thread
From: Ananth N Mavinakayanahalli @ 2009-04-03  5:18 UTC (permalink / raw)
  To: Masami Hiramatsu
  Cc: Ingo Molnar, Jim Keniston, LKML, systemtap-ml, kvm, Andi Kleen,
	Andrew Morton, Arnaldo Carvalho de Melo, Steven Rostedt,
	Frederic Weisbecker

On Thu, Apr 02, 2009 at 01:24:57PM -0400, Masami Hiramatsu wrote:
 
> +/* Recover original instruction */
> +static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
> +{
> +	struct kprobe *kp;
> +	kp = get_kprobe((void *)addr);
> +	if (!kp)
> +		return -EINVAL;
> +
> +	/* Don't use p->ainsn.insn; which will be modified by fix_riprel */
> +	memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
> +	buf[0] = kp->opcode;
> +	return 0;
> +}
> +
> +/* Dummy buffers for lookup_symbol_attrs */
> +static char __dummy_buf[KSYM_NAME_LEN];
> +
> +/* Check whether the address can be probed */
> +static int __kprobes can_probe(unsigned long paddr)

A better description would've been "Check if paddr is at an instruction
boundary". Otherwise looks good.

Ananth

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

* [PATCH -tip 4/6 V4.1] x86: kprobes checks safeness of insertion address.
  2009-04-03  5:18 ` Ananth N Mavinakayanahalli
@ 2009-04-03 16:02   ` Masami Hiramatsu
  2009-04-03 17:28     ` Jim Keniston
  0 siblings, 1 reply; 6+ messages in thread
From: Masami Hiramatsu @ 2009-04-03 16:02 UTC (permalink / raw)
  To: ananth
  Cc: Ingo Molnar, Jim Keniston, LKML, systemtap-ml, kvm, Andi Kleen,
	Andrew Morton, Arnaldo Carvalho de Melo, Steven Rostedt,
	Frederic Weisbecker

Ensure safeness of inserting kprobes by checking whether the specified
address is at the first byte of a instruction. This is done by decoding
probed function from its head to the probe point.

changes from v4:
 - change a comment according to Ananth's suggestion.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
---

 arch/x86/kernel/kprobes.c |   51 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 51 insertions(+), 0 deletions(-)


diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 7b5169d..e4f36e7 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -48,12 +48,14 @@
 #include <linux/preempt.h>
 #include <linux/module.h>
 #include <linux/kdebug.h>
+#include <linux/kallsyms.h>

 #include <asm/cacheflush.h>
 #include <asm/desc.h>
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 #include <asm/alternative.h>
+#include <asm/insn.h>

 void jprobe_return_end(void);

@@ -244,6 +246,53 @@ retry:
 	}
 }

+/* Recover original instruction */
+static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
+{
+	struct kprobe *kp;
+	kp = get_kprobe((void *)addr);
+	if (!kp)
+		return -EINVAL;
+
+	/* Don't use p->ainsn.insn; which will be modified by fix_riprel */
+	memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	buf[0] = kp->opcode;
+	return 0;
+}
+
+/* Dummy buffers for lookup_symbol_attrs */
+static char __dummy_buf[KSYM_NAME_LEN];
+
+/* Check if paddr is at an instruction boundary */
+static int __kprobes can_probe(unsigned long paddr)
+{
+	int ret;
+	unsigned long addr, offset = 0;
+	struct insn insn;
+	kprobe_opcode_t buf[MAX_INSN_SIZE];
+
+	/* Lookup symbol including addr */
+	if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
+		return 0;
+
+	/* Decode instructions */
+	addr = paddr - offset;
+	while (addr < paddr) {
+		insn_init_kernel(&insn, (void *)addr);
+		insn_get_opcode(&insn);
+		if (OPCODE1(&insn) == BREAKPOINT_INSTRUCTION) {
+			ret = recover_probed_instruction(buf, addr);
+			if (ret)
+				return 0;
+			insn_init_kernel(&insn, buf);
+		}
+		insn_get_length(&insn);
+		addr += insn.length;
+	}
+
+	return (addr == paddr);
+}
+
 /*
  * Returns non-zero if opcode modifies the interrupt flag.
  */
@@ -359,6 +408,8 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p)

 int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
+	if (!can_probe((unsigned long)p->addr))
+		return -EILSEQ;
 	/* insn: must be on special executable page on x86. */
 	p->ainsn.insn = get_insn_slot();
 	if (!p->ainsn.insn)
-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com


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

* Re: [PATCH -tip 4/6 V4.1] x86: kprobes checks safeness of insertion address.
  2009-04-03 16:02   ` [PATCH -tip 4/6 V4.1] " Masami Hiramatsu
@ 2009-04-03 17:28     ` Jim Keniston
  2009-04-03 19:20       ` Masami Hiramatsu
  0 siblings, 1 reply; 6+ messages in thread
From: Jim Keniston @ 2009-04-03 17:28 UTC (permalink / raw)
  To: Masami Hiramatsu
  Cc: ananth, Ingo Molnar, LKML, systemtap-ml, kvm, Andi Kleen,
	Andrew Morton, Arnaldo Carvalho de Melo, Steven Rostedt,
	Frederic Weisbecker

On Fri, 2009-04-03 at 12:02 -0400, Masami Hiramatsu wrote:
> Ensure safeness of inserting kprobes by checking whether the specified
> address is at the first byte of a instruction. This is done by decoding
> probed function from its head to the probe point.
> 
> changes from v4:
>  - change a comment according to Ananth's suggestion.
> 
> Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> Cc: Jim Keniston <jkenisto@us.ibm.com>
> Cc: Ingo Molnar <mingo@elte.hu>
> ---
> 
>  arch/x86/kernel/kprobes.c |   51 +++++++++++++++++++++++++++++++++++++++++++++
>  1 files changed, 51 insertions(+), 0 deletions(-)
> 
> 
> diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
...
> 
> +/* Recover original instruction */

/* Recover the probed instruction at addr for further analysis. */
See below.

> +static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
> +{
> +	struct kprobe *kp;
> +	kp = get_kprobe((void *)addr);
> +	if (!kp)
> +		return -EINVAL;
> +
> +	/* Don't use p->ainsn.insn; which will be modified by fix_riprel */

fix_riprel doesn't affect the instruction's length, which is what
concerns this patch.  But we want this function to be useful for
unforeseen uses as well, so I like the code you have.  Just consider the
suggested comment changes.

	/*
	 * Don't use p->ainsn.insn, which could be modified -- e.g.,
	 * by fix_riprel().
	 */

> +	memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
> +	buf[0] = kp->opcode;
> +	return 0;
> +}

Jim Keniston


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

* Re: [PATCH -tip 4/6 V4.1] x86: kprobes checks safeness of insertion address.
  2009-04-03 17:28     ` Jim Keniston
@ 2009-04-03 19:20       ` Masami Hiramatsu
  2009-04-03 21:21         ` [PATCH -tip 4/6 V4.2] " Masami Hiramatsu
  0 siblings, 1 reply; 6+ messages in thread
From: Masami Hiramatsu @ 2009-04-03 19:20 UTC (permalink / raw)
  To: Jim Keniston
  Cc: ananth, Ingo Molnar, LKML, systemtap-ml, kvm, Andi Kleen,
	Andrew Morton, Arnaldo Carvalho de Melo, Steven Rostedt,
	Frederic Weisbecker

Jim Keniston wrote:
> On Fri, 2009-04-03 at 12:02 -0400, Masami Hiramatsu wrote:
>> Ensure safeness of inserting kprobes by checking whether the specified
>> address is at the first byte of a instruction. This is done by decoding
>> probed function from its head to the probe point.
>>
>> changes from v4:
>>  - change a comment according to Ananth's suggestion.
>>
>> Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
>> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
>> Cc: Jim Keniston <jkenisto@us.ibm.com>
>> Cc: Ingo Molnar <mingo@elte.hu>
>> ---
>>
>>  arch/x86/kernel/kprobes.c |   51 +++++++++++++++++++++++++++++++++++++++++++++
>>  1 files changed, 51 insertions(+), 0 deletions(-)
>>
>>
>> diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
> ...
>> +/* Recover original instruction */
> 
> /* Recover the probed instruction at addr for further analysis. */
> See below.

Sure.

> 
>> +static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
>> +{
>> +	struct kprobe *kp;
>> +	kp = get_kprobe((void *)addr);
>> +	if (!kp)
>> +		return -EINVAL;
>> +
>> +	/* Don't use p->ainsn.insn; which will be modified by fix_riprel */
> 
> fix_riprel doesn't affect the instruction's length, which is what
> concerns this patch.  But we want this function to be useful for
> unforeseen uses as well, so I like the code you have.  Just consider the
> suggested comment changes.
> 
> 	/*
> 	 * Don't use p->ainsn.insn, which could be modified -- e.g.,
> 	 * by fix_riprel().
> 	 */

Thanks, I'll update comments then!

> 
>> +	memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
>> +	buf[0] = kp->opcode;
>> +	return 0;
>> +}
> 
> Jim Keniston
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com


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

* [PATCH -tip 4/6 V4.2] x86: kprobes checks safeness of insertion address.
  2009-04-03 19:20       ` Masami Hiramatsu
@ 2009-04-03 21:21         ` Masami Hiramatsu
  0 siblings, 0 replies; 6+ messages in thread
From: Masami Hiramatsu @ 2009-04-03 21:21 UTC (permalink / raw)
  To: Jim Keniston
  Cc: ananth, Ingo Molnar, LKML, systemtap-ml, kvm, Andi Kleen,
	Andrew Morton, Arnaldo Carvalho de Melo, Steven Rostedt,
	Frederic Weisbecker

x86: kprobes checks safeness of insertion address.

From: Masami Hiramatsu <mhiramat@redhat.com>

Ensure safeness of inserting kprobes by checking whether the specified
address is at the first byte of a instruction. This is done by decoding
probed function from its head to the probe point.

changes from v4.1:
 - update comments according to Jim's suggestion.
 - s/lookup_symbol_attrs/kallsyms_lookup/ in a comment.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
---

 arch/x86/kernel/kprobes.c |   54 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 54 insertions(+), 0 deletions(-)


diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 7b5169d..3d5e85f 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -48,12 +48,14 @@
 #include <linux/preempt.h>
 #include <linux/module.h>
 #include <linux/kdebug.h>
+#include <linux/kallsyms.h>

 #include <asm/cacheflush.h>
 #include <asm/desc.h>
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 #include <asm/alternative.h>
+#include <asm/insn.h>

 void jprobe_return_end(void);

@@ -244,6 +246,56 @@ retry:
 	}
 }

+/* Recover the probed instruction at addr for further analysis. */
+static int recover_probed_instruction(kprobe_opcode_t *buf, unsigned long addr)
+{
+	struct kprobe *kp;
+	kp = get_kprobe((void *)addr);
+	if (!kp)
+		return -EINVAL;
+
+	/*
+	 * Don't use p->ainsn.insn, which could be modified -- e.g.,
+	 * by fix_riprel().
+	 */
+	memcpy(buf, kp->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+	buf[0] = kp->opcode;
+	return 0;
+}
+
+/* Dummy buffers for kallsyms_lookup */
+static char __dummy_buf[KSYM_NAME_LEN];
+
+/* Check if paddr is at an instruction boundary */
+static int __kprobes can_probe(unsigned long paddr)
+{
+	int ret;
+	unsigned long addr, offset = 0;
+	struct insn insn;
+	kprobe_opcode_t buf[MAX_INSN_SIZE];
+
+	/* Lookup symbol including addr */
+	if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
+		return 0;
+
+	/* Decode instructions */
+	addr = paddr - offset;
+	while (addr < paddr) {
+		insn_init_kernel(&insn, (void *)addr);
+		insn_get_opcode(&insn);
+		if (OPCODE1(&insn) == BREAKPOINT_INSTRUCTION) {
+			ret = recover_probed_instruction(buf, addr);
+			if (ret)
+				return 0;
+			insn_init_kernel(&insn, buf);
+		}
+		insn_get_length(&insn);
+		addr += insn.length;
+	}
+
+	return (addr == paddr);
+}
+
 /*
  * Returns non-zero if opcode modifies the interrupt flag.
  */
@@ -359,6 +411,8 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p)

 int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
+	if (!can_probe((unsigned long)p->addr))
+		return -EILSEQ;
 	/* insn: must be on special executable page on x86. */
 	p->ainsn.insn = get_insn_slot();
 	if (!p->ainsn.insn)

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com


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

end of thread, other threads:[~2009-04-03 21:22 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-04-02 17:24 [PATCH -tip 4/6 V4] x86: kprobes checks safeness of insertion address Masami Hiramatsu
2009-04-03  5:18 ` Ananth N Mavinakayanahalli
2009-04-03 16:02   ` [PATCH -tip 4/6 V4.1] " Masami Hiramatsu
2009-04-03 17:28     ` Jim Keniston
2009-04-03 19:20       ` Masami Hiramatsu
2009-04-03 21:21         ` [PATCH -tip 4/6 V4.2] " Masami Hiramatsu

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).