tcp/dccp: get rid of central timewait timer
Using a timer wheel for timewait sockets was nice ~15 years ago when
memory was expensive and machines had a single processor.
This does not scale, code is ugly and source of huge latencies
(Typically 30 ms have been seen, cpus spinning on death_lock spinlock.)
We can afford to use an extra 64 bytes per timewait sock and spread
timewait load to all cpus to have better behavior.
Tested:
On following test, /proc/sys/net/ipv4/tcp_tw_recycle is set to 1
on the target (lpaa24)
Before patch :
lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
419594
lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
437171
While test is running, we can observe 25 or even 33 ms latencies.
lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23
...
1000 packets transmitted, 1000 received, 0% packet loss, time 20601ms
rtt min/avg/max/mdev = 0.020/0.217/25.771/1.535 ms, pipe 2
lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23
...
1000 packets transmitted, 1000 received, 0% packet loss, time 20702ms
rtt min/avg/max/mdev = 0.019/0.183/33.761/1.441 ms, pipe 2
After patch :
About 90% increase of throughput :
lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
810442
lpaa23:~# ./super_netperf 200 -H lpaa24 -t TCP_CC -l 60 -- -p0,0
800992
And latencies are kept to minimal values during this load, even
if network utilization is 90% higher :
lpaa24:~# ping -c 1000 -i 0.02 -qn lpaa23
...
1000 packets transmitted, 1000 received, 0% packet loss, time 19991ms
rtt min/avg/max/mdev = 0.023/0.064/0.360/0.042 ms
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index 332f7d6..5f56666 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -27,28 +27,16 @@
struct inet_timewait_death_row dccp_death_row = {
.sysctl_max_tw_buckets = NR_FILE * 2,
- .period = DCCP_TIMEWAIT_LEN / INET_TWDR_TWKILL_SLOTS,
- .death_lock = __SPIN_LOCK_UNLOCKED(dccp_death_row.death_lock),
.hashinfo = &dccp_hashinfo,
- .tw_timer = TIMER_INITIALIZER(inet_twdr_hangman, 0,
- (unsigned long)&dccp_death_row),
- .twkill_work = __WORK_INITIALIZER(dccp_death_row.twkill_work,
- inet_twdr_twkill_work),
-/* Short-time timewait calendar */
-
- .twcal_hand = -1,
- .twcal_timer = TIMER_INITIALIZER(inet_twdr_twcal_tick, 0,
- (unsigned long)&dccp_death_row),
};
EXPORT_SYMBOL_GPL(dccp_death_row);
void dccp_time_wait(struct sock *sk, int state, int timeo)
{
- struct inet_timewait_sock *tw = NULL;
+ struct inet_timewait_sock *tw;
- if (dccp_death_row.tw_count < dccp_death_row.sysctl_max_tw_buckets)
- tw = inet_twsk_alloc(sk, state);
+ tw = inet_twsk_alloc(sk, &dccp_death_row, state);
if (tw != NULL) {
const struct inet_connection_sock *icsk = inet_csk(sk);
@@ -71,8 +59,7 @@
if (state == DCCP_TIME_WAIT)
timeo = DCCP_TIMEWAIT_LEN;
- inet_twsk_schedule(tw, &dccp_death_row, timeo,
- DCCP_TIMEWAIT_LEN);
+ inet_twsk_schedule(tw, timeo);
inet_twsk_put(tw);
} else {
/* Sorry, if we're out of memory, just CLOSE this