From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Borkmann Subject: Re: [PATCH v2 net-next] pkt_sched: fq: Fair Queue packet scheduler Date: Wed, 04 Sep 2013 13:59:27 +0200 Message-ID: <5227209F.4060708@redhat.com> References: <1377816595.8277.54.camel@edumazet-glaptop> <5226C4A0.6040709@redhat.com> <1378274376.7360.82.camel@edumazet-glaptop> <5226D39C.9070401@redhat.com> <1378290638.7360.85.camel@edumazet-glaptop> <1378294029.7360.92.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Jason Wang , David Miller , netdev , Yuchung Cheng , Neal Cardwell , "Michael S. Tsirkin" To: Eric Dumazet Return-path: Received: from mx1.redhat.com ([209.132.183.28]:35039 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757249Ab3IDL7f (ORCPT ); Wed, 4 Sep 2013 07:59:35 -0400 In-Reply-To: <1378294029.7360.92.camel@edumazet-glaptop> Sender: netdev-owner@vger.kernel.org List-ID: On 09/04/2013 01:27 PM, Eric Dumazet wrote: > On Wed, 2013-09-04 at 03:30 -0700, Eric Dumazet wrote: >> On Wed, 2013-09-04 at 14:30 +0800, Jason Wang wrote: >> >>>> And tcpdump would certainly help ;) >>> >>> See attachment. >>> >> >> Nothing obvious on tcpdump (only that lot of frames are missing) >> >> 1) Are you capturing part of the payload only (like tcpdump -s 128) >> >> 2) What is the setup. >> >> 3) tc -s -d qdisc > > If you use FQ in the guest, then it could be that high resolution timers > have high latency ? Probably they internally switch to a lower resolution clock event source if there's no hardware support available: The [source event] management layer provides interfaces for hrtimers to implement high resolution timers [...] [and it] supports these more advanced functions only when appropriate clock event sources have been registered, otherwise the traditional periodic tick based behaviour is retained. [1] [1] https://www.kernel.org/doc/ols/2006/ols2006v1-pages-333-346.pdf > So FQ arms short timers, but effective duration could be much longer. > > Here I get a smooth latency of up to ~3 us > > lpq83:~# ./netperf -H lpq84 ; ./tc -s -d qd ; dmesg | tail -n1 > MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lpq84.prod.google.com () port 0 AF_INET > Recv Send Send > Socket Socket Message Elapsed > Size Size Size Time Throughput > bytes bytes bytes secs. 10^6bits/sec > > 87380 16384 16384 10.00 9410.82 > qdisc fq 8005: dev eth0 root refcnt 32 limit 10000p flow_limit 100p buckets 1024 quantum 3028 initial_quantum 15140 > Sent 50545633991 bytes 33385894 pkt (dropped 0, overlimits 0 requeues 19) > rate 9258Mbit 764335pps backlog 0b 0p requeues 19 > 117 flow, 115 inactive, 0 throttled > 0 gc, 0 highprio, 0 retrans, 96861 throttled, 0 flows_plimit > [ 572.551664] latency = 3035 ns > > > What do you get with this debugging patch ? > > diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c > index 32ad015..c1312a0 100644 > --- a/net/sched/sch_fq.c > +++ b/net/sched/sch_fq.c > @@ -103,6 +103,7 @@ struct fq_sched_data { > u64 stat_internal_packets; > u64 stat_tcp_retrans; > u64 stat_throttled; > + s64 slatency; > u64 stat_flows_plimit; > u64 stat_pkts_too_long; > u64 stat_allocation_errors; > @@ -393,6 +394,7 @@ static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch) > static void fq_check_throttled(struct fq_sched_data *q, u64 now) > { > struct rb_node *p; > + bool first = true; > > if (q->time_next_delayed_flow > now) > return; > @@ -405,6 +407,13 @@ static void fq_check_throttled(struct fq_sched_data *q, u64 now) > q->time_next_delayed_flow = f->time_next_packet; > break; > } > + if (first) { > + s64 delay = now - f->time_next_packet; > + > + first = false; > + delay -= q->slatency >> 3; > + q->slatency += delay; > + } > rb_erase(p, &q->delayed); > q->throttled_flows--; > fq_flow_add_tail(&q->old_flows, f); > @@ -711,6 +720,7 @@ static int fq_dump(struct Qdisc *sch, struct sk_buff *skb) > if (opts == NULL) > goto nla_put_failure; > > + pr_err("latency = %lld ns\n", q->slatency >> 3); > if (nla_put_u32(skb, TCA_FQ_PLIMIT, sch->limit) || > nla_put_u32(skb, TCA_FQ_FLOW_PLIMIT, q->flow_plimit) || > nla_put_u32(skb, TCA_FQ_QUANTUM, q->quantum) || > > > -- > To unsubscribe from this list: send the line "unsubscribe netdev" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html >