pktgen: avoid expensive set_current_state() call in loop

Avoid calling set_current_state() inside the busy-loop in
pktgen_thread_worker().  In case of pkt_dev->delay, then it is still
used/enabled in pktgen_xmit() via the spin() call.

The set_current_state(TASK_INTERRUPTIBLE) uses a xchg, which implicit
is LOCK prefixed.  I've measured the asm LOCK operation to take approx
8ns on this E5-2630 CPU.  Performance increase corrolate with this
measurement.

Performance data with CLONE_SKB==100000, rx-usecs=30:
 (single CPU performance, ixgbe 10Gbit/s, E5-2630)
 * Prev:  5454050 pps --> 183.35ns (1/5454050*10^9)
 * Now:   5684009 pps --> 175.93ns (1/5684009*10^9)
 * Diff:  +229959 pps -->  -7.42ns

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Jesper Dangaard Brouer 2014-06-26 13:16:49 +02:00 committed by David S. Miller
parent 9ceb87fcea
commit baac167b70
1 changed files with 3 additions and 6 deletions

View File

@ -3407,10 +3407,10 @@ static int pktgen_thread_worker(void *arg)
pr_debug("starting pktgen/%d: pid=%d\n", cpu, task_pid_nr(current)); pr_debug("starting pktgen/%d: pid=%d\n", cpu, task_pid_nr(current));
set_current_state(TASK_INTERRUPTIBLE);
set_freezable(); set_freezable();
__set_current_state(TASK_RUNNING);
while (!kthread_should_stop()) { while (!kthread_should_stop()) {
pkt_dev = next_to_run(t); pkt_dev = next_to_run(t);
@ -3424,8 +3424,6 @@ static int pktgen_thread_worker(void *arg)
continue; continue;
} }
__set_current_state(TASK_RUNNING);
if (likely(pkt_dev)) { if (likely(pkt_dev)) {
pktgen_xmit(pkt_dev); pktgen_xmit(pkt_dev);
@ -3456,9 +3454,8 @@ static int pktgen_thread_worker(void *arg)
} }
try_to_freeze(); try_to_freeze();
set_current_state(TASK_INTERRUPTIBLE);
} }
set_current_state(TASK_INTERRUPTIBLE);
pr_debug("%s stopping all device\n", t->tsk->comm); pr_debug("%s stopping all device\n", t->tsk->comm);
pktgen_stop(t); pktgen_stop(t);