Messages in this thread | | | Subject | Re: [RFC PATCH 2/2] macvtap: TX zero copy between guest and host kernel | From | Shirley Ma <> | Date | Wed, 29 Sep 2010 07:31:29 -0700 |
| |
On Wed, 2010-09-29 at 10:16 +0200, Michael S. Tsirkin wrote: > If you look at dev_hard_start_xmit you will see a call > to skb_orphan_try which often calls the skb destructor. > So I suspect this is almost equivalent to your original patch, > and has the same correctness issue.
I forgot to mention, in skb_orphan_try, I still kept the skb->sk reference to sk, I assign skb->sk to NULL when the DMA done.
Thanks Shirley
| |