On Wed, 2007-08-15 at 21:48 -0700, Michael Wu wrote: > On Wednesday 15 August 2007 07:49, Johannes Berg wrote: > > Is there a possibility of a race condition here? If the interface is > > brought down right after this check the SKB could be copied to that > > interface after the skb queue has been flushed. > > > Yeah. This can be avoided by holding a write lock on the subif lock while > flushing the skb queue. Not sure how much I like doing that, but it should > work. Hmm. Not entirely sure what you mean. johannes
Attachment:
signature.asc
Description: This is a digitally signed message part
