[PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe

[PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe

[Masami Hiramatsu]

 Fix a bug in regiseter_kretprobe() which does not check
 rp->kp.symbol_name == NULL before calling kprobe_lookup_name.

 For maintainability, this introduces kprobe_addr helper function which
 resolves addr field. It is used by register_kprobe and register_kretprobe.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
CC: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
CC: Jim Keniston <jkenisto@us.ibm.com>
---
 kernel/kprobes.c |   43 ++++++++++++++++++++++++++-----------------
 1 file changed, 26 insertions(+), 17 deletions(-)

Index: 2.6.25-rc2-mm1/kernel/kprobes.c
===================================================================
--- 2.6.25-rc2-mm1.orig/kernel/kprobes.c	2008-02-18 11:37:18.000000000 -0500
+++ 2.6.25-rc2-mm1/kernel/kprobes.c	2008-02-18 11:37:18.000000000 -0500
@@ -498,27 +498,36 @@ static int __kprobes in_kprobes_function
 	return 0;
 }

+/*
+ * If we have a symbol_name argument, look it up and add the offset field
+ * to it. This way, we can specify a relative address to a symbol.
+ */
+static kprobe_opcode_t __kprobes *kprobe_addr(struct kprobe *p)
+{
+	kprobe_opcode_t *addr = p->addr;
+	if (p->symbol_name) {
+		if (addr)
+			return NULL;
+		kprobe_lookup_name(p->symbol_name, addr);
+	}
+
+	if (!addr)
+		return NULL;
+	return (kprobe_opcode_t *)(((char *)addr) + p->offset);
+}
+
 static int __kprobes __register_kprobe(struct kprobe *p,
 	unsigned long called_from)
 {
 	int ret = 0;
 	struct kprobe *old_p;
 	struct module *probed_mod;
+	kprobe_opcode_t *addr;

-	/*
-	 * If we have a symbol_name argument look it up,
-	 * and add it to the address.  That way the addr
-	 * field can either be global or relative to a symbol.
-	 */
-	if (p->symbol_name) {
-		if (p->addr)
-			return -EINVAL;
-		kprobe_lookup_name(p->symbol_name, p->addr);
-	}
-
-	if (!p->addr)
+	addr = kprobe_addr(p);
+	if (!addr)
 		return -EINVAL;
-	p->addr = (kprobe_opcode_t *)(((char *)p->addr)+ p->offset);
+	p->addr = addr;

 	if (!kernel_text_address((unsigned long) p->addr) ||
 	    in_kprobes_functions((unsigned long) p->addr))
@@ -722,12 +731,12 @@ int __kprobes register_kretprobe(struct
 	int ret = 0;
 	struct kretprobe_instance *inst;
 	int i;
-	void *addr = rp->kp.addr;
+	void *addr;

 	if (kretprobe_blacklist_size) {
-		if (addr == NULL)
-			kprobe_lookup_name(rp->kp.symbol_name, addr);
-		addr += rp->kp.offset;
+		addr = kprobe_addr(&rp->kp);
+		if (!addr)
+			return -EINVAL;

 		for (i = 0; kretprobe_blacklist[i].name != NULL; i++) {
 			if (kretprobe_blacklist[i].addr == addr)
-- 
Masami Hiramatsu

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

e-mail: mhiramat@redhat.com

Re: [PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe

[Andrew Morton]

On Mon, 03 Mar 2008 17:26:36 -0500
Masami Hiramatsu <mhiramat@redhat.com> wrote:
>

This:

> Subject: [PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe 

implies (to me) that the patch fixes a bug which is only in -mm.  But
afacit this is a bugfix against mainline, yes?

Re: [PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe

[Masami Hiramatsu]

Andrew Morton wrote:
> On Mon, 03 Mar 2008 17:26:36 -0500
> Masami Hiramatsu <mhiramat@redhat.com> wrote:
> 
> This:
> 
>> Subject: [PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe 
> 
> implies (to me) that the patch fixes a bug which is only in -mm.  But
> afacit this is a bugfix against mainline, yes?

Yes, that bug is also in mainline...

Thank you,


-- 
Masami Hiramatsu

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

e-mail: mhiramat@redhat.com

Re: [PATCH][2.6.25-rc2-mm1] fix a null pointer bug in register_kretprobe

[Ananth N Mavinakayanahalli]

On Mon, Mar 03, 2008 at 05:26:36PM -0500, Masami Hiramatsu wrote:
> X-Enigmail-Version: 0.95.6
> Content-Type: text/plain; charset=ISO-8859-1
> Content-Transfer-Encoding: 7bit
> X-Scanned-By: MIMEDefang 2.58 on 172.16.52.254
> 
>  Fix a bug in regiseter_kretprobe() which does not check
>  rp->kp.symbol_name == NULL before calling kprobe_lookup_name.
> 
>  For maintainability, this introduces kprobe_addr helper function which
>  resolves addr field. It is used by register_kprobe and register_kretprobe.
> 
> Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
> CC: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> CC: Jim Keniston <jkenisto@us.ibm.com>

Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>

Thanks Masami!

> ---
>  kernel/kprobes.c |   43 ++++++++++++++++++++++++++-----------------
>  1 file changed, 26 insertions(+), 17 deletions(-)
> 
> Index: 2.6.25-rc2-mm1/kernel/kprobes.c
> ===================================================================
> --- 2.6.25-rc2-mm1.orig/kernel/kprobes.c	2008-02-18 11:37:18.000000000 -0500
> +++ 2.6.25-rc2-mm1/kernel/kprobes.c	2008-02-18 11:37:18.000000000 -0500
> @@ -498,27 +498,36 @@ static int __kprobes in_kprobes_function
>  	return 0;
>  }
> 
> +/*
> + * If we have a symbol_name argument, look it up and add the offset field
> + * to it. This way, we can specify a relative address to a symbol.
> + */
> +static kprobe_opcode_t __kprobes *kprobe_addr(struct kprobe *p)
> +{
> +	kprobe_opcode_t *addr = p->addr;
> +	if (p->symbol_name) {
> +		if (addr)
> +			return NULL;
> +		kprobe_lookup_name(p->symbol_name, addr);
> +	}
> +
> +	if (!addr)
> +		return NULL;
> +	return (kprobe_opcode_t *)(((char *)addr) + p->offset);
> +}
> +
>  static int __kprobes __register_kprobe(struct kprobe *p,
>  	unsigned long called_from)
>  {
>  	int ret = 0;
>  	struct kprobe *old_p;
>  	struct module *probed_mod;
> +	kprobe_opcode_t *addr;
> 
> -	/*
> -	 * If we have a symbol_name argument look it up,
> -	 * and add it to the address.  That way the addr
> -	 * field can either be global or relative to a symbol.
> -	 */
> -	if (p->symbol_name) {
> -		if (p->addr)
> -			return -EINVAL;
> -		kprobe_lookup_name(p->symbol_name, p->addr);
> -	}
> -
> -	if (!p->addr)
> +	addr = kprobe_addr(p);
> +	if (!addr)
>  		return -EINVAL;
> -	p->addr = (kprobe_opcode_t *)(((char *)p->addr)+ p->offset);
> +	p->addr = addr;
> 
>  	if (!kernel_text_address((unsigned long) p->addr) ||
>  	    in_kprobes_functions((unsigned long) p->addr))
> @@ -722,12 +731,12 @@ int __kprobes register_kretprobe(struct
>  	int ret = 0;
>  	struct kretprobe_instance *inst;
>  	int i;
> -	void *addr = rp->kp.addr;
> +	void *addr;
> 
>  	if (kretprobe_blacklist_size) {
> -		if (addr == NULL)
> -			kprobe_lookup_name(rp->kp.symbol_name, addr);
> -		addr += rp->kp.offset;
> +		addr = kprobe_addr(&rp->kp);
> +		if (!addr)
> +			return -EINVAL;
> 
>  		for (i = 0; kretprobe_blacklist[i].name != NULL; i++) {
>  			if (kretprobe_blacklist[i].addr == addr)
> -- 
> Masami Hiramatsu
> 
> Software Engineer
> Hitachi Computer Products (America) Inc.
> Software Solutions Division
> 
> e-mail: mhiramat@redhat.com
>