On Fri, Apr 07, 2006 at 06:03:04PM +0800, Yi Yang (yang.y.yi@xxxxxxxxx) wrote:I'm considering how to free it, because cloned skbs share data with original skb, so this case is special,
No, I also use clone to send skb, so they should have the same overhead.Can you explain why there is such a big difference between netlink_unicast and netlink_broadcast?Netlink broadcast clones skbs, while unicasting requires the whole new
one.
I missed that.
After rereading fsevent_send_to_process() I do not see how original skb
is freed though.
>from running cpu, not from any other cpus.How does keventd know about your own structures?Btw, you need some rebalancing of the per-cpu queues, probably inDoes keventd not do it? if so, keventd should be modified.
keventd, since CPUs can go offline and your messages will stuck foreve
there.
You have an per-cpu object, but your keventd function gets object