Re: [TOMOYO #5 18/18] LSM expansion for TOMOYO Linux.

[Paul Moore <paul.moore@hp.com>]

On Friday 16 November 2007 10:45:32 pm Tetsuo Handa wrote:
> Paul Moore wrote:
> > I might be missing something here, but why do you need to do a skb_peek()
> > again?  You already have the skb and the sock, just do the unlink.
>
> The skb might be already dequeued by other thread while I slept inside
> security_post_recv_datagram().

Okay, well if that is the case I think you are going to have another problem 
in that you could end up throwing away skbs that haven't been through your 
security_post_recv_datagram() hook because you _always_ throw away the result 
of the second skb_peek().  Once again, if I'm wrong please correct me.

> > Second, why not move the 'no_peek' code to just before 'no_packet'?
>
> Oh, I didn't notice I can insert here. Now I can also move the rest code
> like
>
> |		error = security_post_recv_datagram(sk, skb, flags);
> |		if (error)
> |                       goto force_dequeue;
> |
> |         } while (!wait_for_packet(sk, err, &timeo));

Where did the 'if (skb) return skb;' code go?  Don't you need to do you LSM 
call before you return the skb?

-- 
paul moore
linux security @ hp