xfrm: Fix deletion of offloaded SAs on failure.

klassert · klassert · commit c5d4d7d83165 · 2017-09-11T10:36:41.000+02:00
When we off load a SA, it gets pushed to the NIC before we can add it. In case of a failure, we don't delete this SA from the NIC. Fix this by calling xfrm_dev_state_delete on failure. Fixes: d77e38e ("xfrm: Add an IPsec hardware offloading API") Reported-by: Shannon Nelson <shannon.nelson@oracle.com> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
@@ -657,6 +657,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 	if (err < 0) {
 		x->km.state = XFRM_STATE_DEAD;
+		xfrm_dev_state_delete(x);
 		__xfrm_state_put(x);
 		goto out;
 	}

Original file line number	Diff line number	Diff line change
`@@ -657,6 +657,7 @@ static int xfrm_add_sa(struct sk_buff skb, struct nlmsghdr nlh,`
`657`	`657`
`658`	`658`	`if (err < 0) {`
`659`	`659`	`x->km.state = XFRM_STATE_DEAD;`
	`660`	`+ xfrm_dev_state_delete(x);`
`660`	`661`	`__xfrm_state_put(x);`
`661`	`662`	`goto out;`
`662`	`663`	`}`