Skip to contentSkip to conversation

livepatch: Fix usage of kobject_init_and_add()

PATCH In review Tobin C. Harding 5 participants 14 messages Apr 30, 2019 Live patching

lore
Share & tools
lore
#0 lore raw

cover livepatch: Fix usage of kobject_init_and_add()

Hi,

Currently there are a few places in kernel/livepatch/ which do not correctly use kobject_init_and_add().

An error return from kobject_init_and_add() requires a call to kobject_put().

The cleanup function after a successful call to kobject_init_and_add() is kobject_del().

This set is part of an effort to check/fix all callsites of kobject_init_and_add().

This set fixes all callsites under kernel/livepatch/

thanks,
Tobin.

Tobin C. Harding (2):
livepatch: Fix kobject memleak
livepatch: Use correct kobject cleanup function

 kernel/livepatch/core.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)
Signature · 1 line
2.21.0
#1 lore raw

PATCH 1/2 livepatch: Fix kobject memleak

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

Add call to kobject_put() in error path of kobject_init_and_add().

---
 kernel/livepatch/core.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
kernel/livepatch/core.c +9 -3
index eb0ee10a1981..98a7bec41faa 100644
@@ -727,7 +727,9 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func)
 	ret = kobject_init_and_add(&func->kobj, &klp_ktype_func,
 				   &obj->kobj, "%s,%lu", func->old_name,
 				   func->old_sympos ? func->old_sympos : 1);
-	if (!ret)
+	if (ret)
+		kobject_put(&func->kobj);
+	else
 		func->kobj_added = true;
 
 	return ret;
@@ -803,8 +805,10 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj)
 	name = klp_is_module(obj) ? obj->name : "vmlinux";
 	ret = kobject_init_and_add(&obj->kobj, &klp_ktype_object,
 				   &patch->kobj, "%s", name);
-	if (ret)
+	if (ret) {
+		kobject_put(&func->kobj);
 		return ret;
+	}
 	obj->kobj_added = true;
 
 	klp_for_each_func(obj, func) {
@@ -862,8 +866,10 @@ static int klp_init_patch(struct klp_patch *patch)
 
 	ret = kobject_init_and_add(&patch->kobj, &klp_ktype_patch,
 				   klp_root_kobj, "%s", patch->mod->name);
-	if (ret)
+	if (ret) {
+		kobject_put(&func->kobj);
 		return ret;
+	}
 	patch->kobj_added = true;
 
 	if (patch->replace) {
Signature · 1 line
2.21.0
maintainer
#3 lore raw

Re: [PATCH 1/2] livepatch: Fix kobject memleak

Quoted 9 lines · Greg Kroah-Hartman
Quoted 6 lines · Tobin C. Harding

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

Add call to kobject_put() in error path of kobject_init_and_add().

Well, it does not even compile...

Quoted 26 lines · Tobin C. Harding

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

Add call to kobject_put() in error path of kobject_init_and_add().

---
 kernel/livepatch/core.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
kernel/livepatch/core.c +3 -1
index eb0ee10a1981..98a7bec41faa 100644
@@ -727,7 +727,9 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func)
 	ret = kobject_init_and_add(&func->kobj, &klp_ktype_func,
 				   &obj->kobj, "%s,%lu", func->old_name,
 				   func->old_sympos ? func->old_sympos : 1);
-	if (!ret)
+	if (ret)
+		kobject_put(&func->kobj);
+	else
 		func->kobj_added = true;
 
 	return ret;

kernel/livepatch/core.c static int klp_init_object(struct klp_patch *patch, struct klp_object *obj) · line 809

Quoted 6 lines
kernel/livepatch/core.c +2 -1
@@ -803,8 +805,10 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj)
 	name = klp_is_module(obj) ? obj->name : "vmlinux";
 	ret = kobject_init_and_add(&obj->kobj, &klp_ktype_object,
 				   &patch->kobj, "%s", name);
-	if (ret)
+	if (ret) {
+		kobject_put(&func->kobj);

kobject_put(&obj->kobj); I suppose.

kernel/livepatch/core.c static int klp_init_patch(struct klp_patch *patch) · line 870

Quoted 12 lines
kernel/livepatch/core.c +3 -1
 		return ret;
+	}
 	obj->kobj_added = true;
 
 	klp_for_each_func(obj, func) {
@@ -862,8 +866,10 @@ static int klp_init_patch(struct klp_patch *patch)
 
 	ret = kobject_init_and_add(&patch->kobj, &klp_ktype_patch,
 				   klp_root_kobj, "%s", patch->mod->name);
-	if (ret)
+	if (ret) {
+		kobject_put(&func->kobj);

kobject_put(&patch->kobj);

Thanks,
Miroslav

#4 lore raw

Re: [PATCH 1/2] livepatch: Fix kobject memleak

Quoted 6 lines · Miroslav Benes
Quoted 4 lines · Greg Kroah-Hartman
Quoted 3 lines · Tobin C. Harding

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

Add call to kobject_put() in error path of kobject_init_and_add().

Well, it does not even compile...

The idea is correct :)

Quoted 26 lines · Tobin C. Harding
Quoted 25 lines

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

Add call to kobject_put() in error path of kobject_init_and_add().

---
 kernel/livepatch/core.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
kernel/livepatch/core.c +3 -1
index eb0ee10a1981..98a7bec41faa 100644
@@ -727,7 +727,9 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func)
 	ret = kobject_init_and_add(&func->kobj, &klp_ktype_func,
 				   &obj->kobj, "%s,%lu", func->old_name,
 				   func->old_sympos ? func->old_sympos : 1);
-	if (!ret)
+	if (ret)
+		kobject_put(&func->kobj);
+	else
 		func->kobj_added = true;
 
Quoted 10 lines

kernel/livepatch/core.c static int klp_init_object(struct klp_patch *patch, struct klp_object *obj) · line 809

Quoted 8 lines
kernel/livepatch/core.c +2 -1
 	return ret;
@@ -803,8 +805,10 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj)
 	name = klp_is_module(obj) ? obj->name : "vmlinux";
 	ret = kobject_init_and_add(&obj->kobj, &klp_ktype_object,
 				   &patch->kobj, "%s", name);
-	if (ret)
+	if (ret) {
+		kobject_put(&func->kobj);

kobject_put(&obj->kobj); I suppose.

Yeah, that makes it better, sorry for not catching that in the review :(

greg k-h

#5 lore raw

Re: [PATCH 1/2] livepatch: Fix kobject memleak

In reply to Miroslav Benes
Quoted 6 lines · Miroslav Benes
Quoted 4 lines · Greg Kroah-Hartman
Quoted 3 lines · Tobin C. Harding

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

Add call to kobject_put() in error path of kobject_init_and_add().

Well, it does not even compile...

My apologies, I did compile this but obviously I don't know how to configure the kernel.

Thanks for the review.

Tobin

maintainer
#6 lore raw

Re: [PATCH 1/2] livepatch: Fix kobject memleak

In reply to Tobin C. Harding
Quoted 2 lines · Tobin C. Harding

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

I see, the ref count is always initialized to 1 via:

+ kobject_init_and_add()
+ kobject_init()
+ kobject_init_internal()
+ kref_init()

Quoted 10 lines
---
 kernel/livepatch/core.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
kernel/livepatch/core.c +0 -0
index eb0ee10a1981..98a7bec41faa 100644

kernel/livepatch/core.c static int klp_init_func(struct klp_object *obj, struct klp_func *func) · line 733

Quoted 8 lines
kernel/livepatch/core.c +3 -1
@@ -727,7 +727,9 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func)
 	ret = kobject_init_and_add(&func->kobj, &klp_ktype_func,
 				   &obj->kobj, "%s,%lu", func->old_name,
 				   func->old_sympos ? func->old_sympos : 1);
-	if (!ret)
+	if (ret)
+		kobject_put(&func->kobj);
+	else
 		func->kobj_added = true;

We could actually get rid of the custom kobj_added. Intead, we could check for kobj->state_initialized in the various klp_free* functions.

Best Regards,
Petr

#7 lore raw

Re: [PATCH 1/2] livepatch: Fix kobject memleak

Quoted 17 lines · Petr Mladek
Quoted 2 lines · Tobin C. Harding

Currently error return from kobject_init_and_add() is not followed by a call to kobject_put(). This means there is a memory leak.

I see, the ref count is always initialized to 1 via:

+ kobject_init_and_add()
+ kobject_init()
+ kobject_init_internal()
+ kref_init()

Quoted 5 lines
---
 kernel/livepatch/core.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
Quoted 16 lines

kernel/livepatch/core.c static int klp_init_func(struct klp_object *obj, struct klp_func *func) · line 733

Quoted 13 lines
kernel/livepatch/core.c +3 -1
index eb0ee10a1981..98a7bec41faa 100644
@@ -727,7 +727,9 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func)
 	ret = kobject_init_and_add(&func->kobj, &klp_ktype_func,
 				   &obj->kobj, "%s,%lu", func->old_name,
 				   func->old_sympos ? func->old_sympos : 1);
-	if (!ret)
+	if (ret)
+		kobject_put(&func->kobj);
+	else
 		func->kobj_added = true;

We could actually get rid of the custom kobj_added. Intead, we could check for kobj->state_initialized in the various klp_free* functions.

Why do you need to care about this at all anyway? The kobject can handle it's own lifetime just fine (that's what it is there for), why do you need to also track it?

thanks,

greg k-h

#8 lore raw

PATCH 2/2 livepatch: Use correct kobject cleanup function

In reply to Tobin C. Harding

The correct cleanup function after a call to kobject_init_and_add() has succeeded is kobject_del() _not_ kobject_put(). kobject_del() calls kobject_put().

Use correct cleanup function when removing a kobject.

---
 kernel/livepatch/core.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
kernel/livepatch/core.c +3 -5
index 98a7bec41faa..4cce6bb6e073 100644
@@ -589,9 +589,8 @@ static void __klp_free_funcs(struct klp_object *obj, bool nops_only)
 
 		list_del(&func->node);
 
-		/* Might be called from klp_init_patch() error path. */
 		if (func->kobj_added) {
-			kobject_put(&func->kobj);
+			kobject_del(&func->kobj);
 		} else if (func->nop) {
 			klp_free_func_nop(func);
 		}
@@ -625,9 +624,8 @@ static void __klp_free_objects(struct klp_patch *patch, bool nops_only)
 
 		list_del(&obj->node);
 
-		/* Might be called from klp_init_patch() error path. */
 		if (obj->kobj_added) {
-			kobject_put(&obj->kobj);
+			kobject_del(&obj->kobj);
 		} else if (obj->dynamic) {
 			klp_free_object_dynamic(obj);
 		}
@@ -676,7 +674,7 @@ static void klp_free_patch_finish(struct klp_patch *patch)
 	 * cannot get enabled again.
 	 */
 	if (patch->kobj_added) {
-		kobject_put(&patch->kobj);
+		kobject_del(&patch->kobj);
 		wait_for_completion(&patch->finish);
 	}
 
Signature · 1 line
2.21.0
maintainer
#10 lore raw

Re: [PATCH 2/2] livepatch: Use correct kobject cleanup function

In reply to Tobin C. Harding
Quoted 16 lines · Tobin C. Harding

The correct cleanup function after a call to kobject_init_and_add() has succeeded is kobject_del() _not_ kobject_put(). kobject_del() calls kobject_put().

Use correct cleanup function when removing a kobject.

---
 kernel/livepatch/core.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
kernel/livepatch/core.c +0 -0
index 98a7bec41faa..4cce6bb6e073 100644

kernel/livepatch/core.c static void __klp_free_funcs(struct klp_object *obj, bool nops_only) · line 592

Quoted 4 lines
kernel/livepatch/core.c +0 -1
@@ -589,9 +589,8 @@ static void __klp_free_funcs(struct klp_object *obj, bool nops_only)
 
 		list_del(&func->node);
 
-		/* Might be called from klp_init_patch() error path. */

Could you leave the comment as is? If I am not mistaken, it is still valid. func->kobj_added check is here exactly because the function may be called as mentioned.

One could argue that the comment is not so important, but the change does not belong to the patch anyway in my opinion.

kernel/livepatch/core.c static void __klp_free_objects(struct klp_patch *patch, bool nops_only) · line 628

Quoted 11 lines
kernel/livepatch/core.c +1 -2
 		if (func->kobj_added) {
-			kobject_put(&func->kobj);
+			kobject_del(&func->kobj);
 		} else if (func->nop) {
 			klp_free_func_nop(func);
 		}
@@ -625,9 +624,8 @@ static void __klp_free_objects(struct klp_patch *patch, bool nops_only)
 
 		list_del(&obj->node);
 
-		/* Might be called from klp_init_patch() error path. */

Same here.

Quoted 7 lines
kernel/livepatch/core.c +1 -1
 		if (obj->kobj_added) {
-			kobject_put(&obj->kobj);
+			kobject_del(&obj->kobj);
 		} else if (obj->dynamic) {
 			klp_free_object_dynamic(obj);
 		}

kernel/livepatch/core.c static void klp_free_patch_finish(struct klp_patch *patch) · line 679

Quoted 7 lines
kernel/livepatch/core.c +1 -1
@@ -676,7 +674,7 @@ static void klp_free_patch_finish(struct klp_patch *patch)
 	 * cannot get enabled again.
 	 */
 	if (patch->kobj_added) {
-		kobject_put(&patch->kobj);
+		kobject_del(&patch->kobj);
 		wait_for_completion(&patch->finish);
 	}

Miroslav

#11 lore raw

Re: [PATCH 2/2] livepatch: Use correct kobject cleanup function

Quoted 19 lines · Miroslav Benes
Quoted 17 lines · Tobin C. Harding

The correct cleanup function after a call to kobject_init_and_add() has succeeded is kobject_del() _not_ kobject_put(). kobject_del() calls kobject_put().

Use correct cleanup function when removing a kobject.

---
 kernel/livepatch/core.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
kernel/livepatch/core.c +0 -0
index 98a7bec41faa..4cce6bb6e073 100644
@@ -589,9 +589,8 @@ static void __klp_free_funcs(struct klp_object *obj, bool nops_only)
 
Quoted 7 lines

kernel/livepatch/core.c static void __klp_free_funcs(struct klp_object *obj, bool nops_only) · line 592

Quoted 3 lines
kernel/livepatch/core.c +0 -1
 		list_del(&func->node);
 
-		/* Might be called from klp_init_patch() error path. */

Could you leave the comment as is? If I am not mistaken, it is still valid. func->kobj_added check is here exactly because the function may be called as mentioned.

Will put it back in for you on v2

thanks,
Tobin.

maintainer
#12 lore raw

Re: [PATCH 2/2] livepatch: Use correct kobject cleanup function

In reply to Tobin C. Harding
Quoted 3 lines · Tobin C. Harding

The correct cleanup function after a call to kobject_init_and_add() has succeeded is kobject_del() _not_ kobject_put(). kobject_del() calls kobject_put().

Really? I see only kobject_put(kobj->parent) in kobject_del. It decreases a reference of the _parent_ object and not the given one.

Also the section "Kobject removal" in Documentation/kobject.txt says that kobject_del() is for two-stage removal. kobject_put() still needs to get called at a later time.

IMHO, this patch causes that kobject_put() would never get called.

That said, we could probably make the removal a bit cleaner by using kobject_del() in klp_free_patch_start() and kobject_put() in klp_free_patch_finish(). But I have to think more about it.

Best Regards,
Petr

#13 lore raw

Re: [PATCH 2/2] livepatch: Use correct kobject cleanup function

Quoted 8 lines · Petr Mladek
Quoted 3 lines · Tobin C. Harding

The correct cleanup function after a call to kobject_init_and_add() has succeeded is kobject_del() _not_ kobject_put(). kobject_del() calls kobject_put().

Really? I see only kobject_put(kobj->parent) in kobject_del. It decreases a reference of the _parent_ object and not the given one.

Thanks Petr, you are right. I misread kobject_del(). The story thickens, so we need to call kobject_del() AND kobject_put().

Quoted 3 lines

Also the section "Kobject removal" in Documentation/kobject.txt says that kobject_del() is for two-stage removal. kobject_put() still needs to get called at a later time.

Is this call sequence above what is meant by 'two-stage removal', I didn't really understand that bit of the docs (and I almost always just assume docs are stale and take them as a hint only :)

Quoted 1 line

IMHO, this patch causes that kobject_put() would never get called.

I'll do a v2 of this one and re-check all the patches on this I've already sent (including the docs ones).

Quoted 4 lines

That said, we could probably make the removal a bit cleaner by using kobject_del() in klp_free_patch_start() and kobject_put() in klp_free_patch_finish(). But I have to think more about it.

Noted, thanks for your review.

Tobin