On Thu, Sep 25, 2014 at 08:25:33AM +0100, Martin Townsend wrote: > Hi Alex, > > I'm not keen on handling kfree_skb inside of lowpan_process_data for reasons already stated in a previous email. I'll have a rethink to see if there's another way of handling this. One idea I'm investigating is pskb_expand_head, this doesn't create a new sk_buff but adds room to the head and/or tail which seems perfect for decompression. There must only be a reference count of 1, so we would need to make a copy if this is the case before passing to decompression, I can't think why there would be more than 1 reference as the packet will only have passed through the 802.15.4 layer and only 6LoWPAN can decompress it. If it's cloned for monitoring then this would be a problem. > Any thoughts on this? > For me it doesn't matter if kfree_skb is inside of lowpan_process_data or not. Go ahead and send fixes for this, I will only review if there are still issues which are not solved by your patch. Don't know what you exactly mean with "monitoring" now. - Alex -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
