[PATCH] security: keyring: remove redundant NULL check

[PATCH] security: keyring: remove redundant NULL check

[Davidlohr Bueso]

From: Davidlohr Bueso <dave@gnu.org>
Date: Sun, 23 Jan 2011 22:01:20 -0300

Signed-off-by: Davidlohr Bueso <dave@gnu.org>
---
 security/keys/keyring.c |    9 ++++-----
 1 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/security/keys/keyring.c b/security/keys/keyring.c
index d37f713..cb8110b 100644
--- a/security/keys/keyring.c
+++ b/security/keys/keyring.c
@@ -866,11 +866,10 @@ void __key_link_end(struct key *keyring, struct key_type *type,
 	if (type == &key_type_keyring)
 		up_write(&keyring_serialise_link_sem);
 
-	if (prealloc) {
-		kfree(prealloc);
-		key_payload_reserve(keyring,
-				    keyring->datalen - KEYQUOTA_LINK_BYTES);
-	}
+	kfree(prealloc);
+	key_payload_reserve(keyring,
+			keyring->datalen - KEYQUOTA_LINK_BYTES);
+
 	up_write(&keyring->sem);
 }
 
-- 
1.7.1

Re: [PATCH] security: keyring: remove redundant NULL check

[David Howells]

Davidlohr Bueso <dave@gnu.org> wrote:

> -	if (prealloc) {
> -		kfree(prealloc);
> -		key_payload_reserve(keyring,
> -				    keyring->datalen - KEYQUOTA_LINK_BYTES);
> -	}
> +	kfree(prealloc);
> +	key_payload_reserve(keyring,
> +			keyring->datalen - KEYQUOTA_LINK_BYTES);
> +

This is not correct, since you'd be removing the quota allocation for the new
link in the case that __key_link() was called (and cleared *_prealloc).

However, the upstream code isn't correct either, since the preallocation may
not have extended the reservation at all (if it decided to displace a key
rather than adding a key).

David