From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [rfc/rft][patch] should use scheduler sync hint in tcp_prequeue()? Date: Tue, 02 Mar 2010 12:27:53 +0100 Message-ID: <1267529273.2964.111.camel@edumazet-laptop> References: <1267522901.25906.101.camel@marge.simson.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev , LKML , Ingo Molnar To: Mike Galbraith Return-path: In-Reply-To: <1267522901.25906.101.camel@marge.simson.net> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Le mardi 02 mars 2010 =C3=A0 10:41 +0100, Mike Galbraith a =C3=A9crit : > Greetings network land. >=20 > The reason for this query is that wake_affine() fails if there is one > and only one task on a runqueue to encourage tasks spreading out, whi= ch > increases cpu utilization. However, for tasks which are communicatin= g > at high frequency, the cost of the resulting cache misses, should > partners land in non-shared caches, is horrible to behold. My Q6600 = has > shared caches, which may or may not be hit IFF something perturbs the > system, and bounces partner to the right core. That won't happen on = a > box with no shared caches of course, and even with shared caches > available, the pain is highly visible in the TCP numbers below.=20 >=20 > The sync hint tells wake_affine() that the waker is likely going to > sleep soonish, so it subtracts the waker from the load imbalance > calculation, allowing the partner task to be awakened affine. In the > shared cache available case, that is also an enabler that the task be > placed in an idle shared cache, which can increase throughput quite a > bit (see .31 vs .33 AF UNIX), or may cost a bit if there is little to= no > execution overlap (see pipe). >=20 > Now, I _could_ change wake_affine() to globally succeed in the one ta= sk > case, but am loath to do so because that very well may upset delicate > load balancing apple cart. I think it's much safer to target the spo= t > that I know hurts like hell. Thoughts? >=20 > diff --git a/include/net/tcp.h b/include/net/tcp.h > index 34f5cc2..ba3fc64 100644 > --- a/include/net/tcp.h > +++ b/include/net/tcp.h > @@ -939,7 +939,7 @@ static inline int tcp_prequeue(struct sock *sk, s= truct sk_buff *skb) > =20 > tp->ucopy.memory =3D 0; > } else if (skb_queue_len(&tp->ucopy.prequeue) =3D=3D 1) { > - wake_up_interruptible_poll(sk->sk_sleep, > + wake_up_interruptible_sync_poll(sk->sk_sleep, > POLLIN | POLLRDNORM | POLLRDBAND); > if (!inet_csk_ack_scheduled(sk)) > inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, >=20 I suspect this discussion is more a lkml topic but anyway... This wake_up_interruptible_sync_poll() change might be good for loopbac= k communications (and pleases tbench), but is it desirable for regular multi flows NIC traffic ?=20 Ingo probably can answer to this question, since he changed sock_def_readable() (and others) in commit 6f3d09291b498299 I suspect he missed tcp_prequeue() case, maybe not... sched, net: socket wakeups are sync 'sync' wakeups are a hint towards the scheduler that (certain) networking related wakeups likely create coupling between tasks.