On 06/16/2015 07:10 PM, Alexei Starovoitov wrote:
...
Ideally we would allow a blend of tracing and networking programs,
then the best solution would be one or two stable tracepoints in
networking stack where skb is visible and receiving/transmitting task
is also visible, then skb->len and task->pid together would give nice
foundation for accurate stats.
I think combining both seems interesting anyway, we need to find
a way to make this gluing of both worlds easy to use, though. It's
certainly interesting for stats/diagnostics, but one wouldn't be
able to use the current/future skb eBPF helpers from {cls,act}_bpf
in that context.
--
To unsubscribe from this list: send the line "unsubscribe linux-api" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
