Thread (182 messages) flat view 182 messages, 27 authors, 2008-08-01

Re: [regression] nf_iterate(), BUG: unable to handle kernel NULL pointer dereference

From: Pekka Enberg <hidden>
Date: 2008-07-24 13:23:49
Also in: lkml
Subsystem: memory management, memory management - core, netfilter, networking [general], slab allocator, the rest · Maintainers: Andrew Morton, David Hildenbrand, Pablo Neira Ayuso, Florian Westphal, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Vlastimil Babka, Harry Yoo, Linus Torvalds

On Thu, 2008-07-24 at 14:49 +0200, Patrick McHardy wrote:
Pekka Enberg wrote:
quoted
On Thu, Jul 24, 2008 at 3:03 PM, Patrick McHardy [off-list ref] wrote:
quoted
Ingo Molnar wrote:
quoted
* Ingo Molnar [off-list ref] wrote:
quoted
here's a new type of crash a -tip testbox triggered today:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000000
IP: [<0000000000000000>] 0x0
...
Call Trace:
 [<ffffffff80777f25>] ? ipv4_confirm+0x8d/0x122
 [<ffffffff80733107>] nf_iterate+0x43/0xa5
Does reverting 31d8519c fix this?
Hmm, why do you think it's related? It looks like elem->hook() is a
NULL pointer but my patch shouldn't make any difference here...
No, its already in ipv4_confirm, so its most likely helper->help()
thats NULL, which is contained in an extension.

The reason why I think its this patch is (quoting from an old
email that I never got a response to):
Oh, I'm really sorry about that.
---
Your patch introduced a use-after-free and double-free.
krealloc() frees the old pointer, but it is still used
for the ->move operations, then freed again.

To fix this I think we need a __krealloc() that doesn't
free the old memory, especially since it must not be
freed immediately because it may still be used in a RCU
read side (see the last part in the patch attached to
this mail (based on a kernel without your patch)).
Agreed. Something like this, perhaps?

[PATCH] netfilter: fix double-free and use-after free

As suggested by Patrick McHardy, introduce a __krealloc() that doesn't
free the original buffer to fix a double-free and use-after-free bug
introduced by me in netfilter that uses RCU.

Reported-by: Patrick McHardy <redacted>
Signed-off-by: Pekka Enberg <redacted>
---
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 9aa90a6..be6f1d4 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -96,6 +96,7 @@ int kmem_ptr_validate(struct kmem_cache *cachep, const void *ptr);
 /*
  * Common kmalloc functions provided by all allocators
  */
+void * __must_check __krealloc(const void *, size_t, gfp_t);
 void * __must_check krealloc(const void *, size_t, gfp_t);
 void kfree(const void *);
 size_t ksize(const void *);
diff --git a/mm/util.c b/mm/util.c
index 8f18683..6ef9e99 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -68,25 +68,22 @@ void *kmemdup(const void *src, size_t len, gfp_t gfp)
 EXPORT_SYMBOL(kmemdup);
 
 /**
- * krealloc - reallocate memory. The contents will remain unchanged.
+ * __krealloc - like krealloc() but don't free @p.
  * @p: object to reallocate memory for.
  * @new_size: how many bytes of memory are required.
  * @flags: the type of memory to allocate.
  *
- * The contents of the object pointed to are preserved up to the
- * lesser of the new and old sizes.  If @p is %NULL, krealloc()
- * behaves exactly like kmalloc().  If @size is 0 and @p is not a
- * %NULL pointer, the object pointed to is freed.
+ * This function is like krealloc() except it never frees the originally
+ * allocated buffer. Use this if you don't want to free the buffer immediately
+ * like, for example, with RCU.
  */
-void *krealloc(const void *p, size_t new_size, gfp_t flags)
+void *__krealloc(const void *p, size_t new_size, gfp_t flags)
 {
 	void *ret;
 	size_t ks = 0;
 
-	if (unlikely(!new_size)) {
-		kfree(p);
+	if (unlikely(!new_size))
 		return ZERO_SIZE_PTR;
-	}
 
 	if (p)
 		ks = ksize(p);
@@ -95,10 +92,37 @@ void *krealloc(const void *p, size_t new_size, gfp_t flags)
 		return (void *)p;
 
 	ret = kmalloc_track_caller(new_size, flags);
-	if (ret && p) {
+	if (ret && p)
 		memcpy(ret, p, ks);
+
+	return ret;
+}
+EXPORT_SYMBOL(__krealloc);
+
+/**
+ * krealloc - reallocate memory. The contents will remain unchanged.
+ * @p: object to reallocate memory for.
+ * @new_size: how many bytes of memory are required.
+ * @flags: the type of memory to allocate.
+ *
+ * The contents of the object pointed to are preserved up to the
+ * lesser of the new and old sizes.  If @p is %NULL, krealloc()
+ * behaves exactly like kmalloc().  If @size is 0 and @p is not a
+ * %NULL pointer, the object pointed to is freed.
+ */
+void *krealloc(const void *p, size_t new_size, gfp_t flags)
+{
+	void *ret;
+
+	if (unlikely(!new_size)) {
 		kfree(p);
+		return ZERO_SIZE_PTR;
 	}
+
+	ret = __krealloc(p, new_size, flags);
+	if (ret && p != ret)
+		kfree(p);
+
 	return ret;
 }
 EXPORT_SYMBOL(krealloc);
diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index 3469bc7..c956ef7 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -95,7 +95,7 @@ void *__nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp)
 	newlen = newoff + t->len;
 	rcu_read_unlock();
 
-	new = krealloc(ct->ext, newlen, gfp);
+	new = __krealloc(ct->ext, newlen, gfp);
 	if (!new)
 		return NULL;
 


Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help