From mboxrd@z Thu Jan 1 00:00:00 1970 From: Antonio Almeida Subject: Re: [PATCH iproute2] Re: HTB accuracy for high speed Date: Tue, 19 May 2009 11:55:43 +0100 Message-ID: <298f5c050905190355r1594b51ex7a7f3d452cc9c059@mail.gmail.com> References: <20090517201528.GA8552@ami.dom.local> <20090518065629.GA6006@ff.dom.local> <298f5c050905180954m791c14eaxe1f4b2c92f952a2f@mail.gmail.com> <20090518175352.GB2755@ami.dom.local> <298f5c050905181123o55c35e74n24cd4ee2c07d8fca@mail.gmail.com> <20090518183221.GC2755@ami.dom.local> <298f5c050905181156w2254fcb4p889d67e26fad6e6@mail.gmail.com> <20090518190541.GA3047@ami.dom.local> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Stephen Hemminger , netdev@vger.kernel.org, kaber@trash.net, davem@davemloft.net, devik@cdi.cz, Eric Dumazet To: Jarek Poplawski Return-path: Received: from yx-out-2324.google.com ([74.125.44.29]:1828 "EHLO yx-out-2324.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750849AbZESKzo convert rfc822-to-8bit (ORCPT ); Tue, 19 May 2009 06:55:44 -0400 Received: by yx-out-2324.google.com with SMTP id 3so2347354yxj.1 for ; Tue, 19 May 2009 03:55:45 -0700 (PDT) In-Reply-To: <20090518190541.GA3047@ami.dom.local> Sender: netdev-owner@vger.kernel.org List-ID: Doesn't seem to make any diference seting HTB_HYSTERESIS to 0. Here're the values using #define HTB_HYSTERESIS 0 800 bytes: class htb 1:108 parent 1:10 leaf 108: prio 7 quantum 1514 rate 555000Kbit ceil 555000Kbit burst 70901b/8 mpu 0b overhead 0b cburst 70901b/8 mpu 0b overhead 0b level 0 Sent 9773257752 bytes 12277962 pkt (dropped 6292541, overlimits 0 requ= eues 0) rate 621796Kbit 97644pps backlog 0b 127p requeues 0 lended: 12277835 borrowed: 0 giants: 0 tokens: -7 ctokens: -7 850 bytes: class htb 1:108 parent 1:10 leaf 108: prio 7 quantum 1514 rate 555000Kbit ceil 555000Kbit burst 70901b/8 mpu 0b overhead 0b cburst 70901b/8 mpu 0b overhead 0b level 0 Sent 18225005732 bytes 22409017 pkt (dropped 11937269, overlimits 0 re= queues 0) rate 600890Kbit 88796pps backlog 0b 43p requeues 0 lended: 22408974 borrowed: 0 giants: 0 tokens: -2 ctokens: -2 900 bytes: class htb 1:108 parent 1:10 leaf 108: prio 7 quantum 1514 rate 555000Kbit ceil 555000Kbit burst 70901b/8 mpu 0b overhead 0b cburst 70901b/8 mpu 0b overhead 0b level 0 Sent 29790867368 bytes 35400708 pkt (dropped 18399726, overlimits 0 re= queues 0) rate 636361Kbit 88779pps backlog 0b 127p requeues 0 lended: 35400581 borrowed: 0 giants: 0 tokens: -2 ctokens: -2 Antonio Almeida On Mon, May 18, 2009 at 8:05 PM, Jarek Poplawski wr= ote: > On Mon, May 18, 2009 at 07:56:12PM +0100, Antonio Almeida wrote: >> Precise measurements: >> >> 800 bytes: >> class htb 1:108 parent 1:10 leaf 108: prio 7 quantum 1514 rate >> 555000Kbit ceil 555000Kbit burst 70901b/8 mpu 0b overhead 0b cburst >> 70901b/8 mpu 0b overhead 0b level 0 >> =A0Sent 46793626324 bytes 57771194 pkt (dropped 29920019, overlimits= 0 requeues 0) >> =A0rate 621714Kbit 97631pps backlog 0b 126p requeues 0 >> =A0lended: 57771068 borrowed: 0 giants: 0 >> =A0tokens: -8 ctokens: -8 >> >> >> 850 bytes: >> class htb 1:108 parent 1:10 leaf 108: prio 7 quantum 1514 rate >> 555000Kbit ceil 555000Kbit burst 70901b/8 mpu 0b overhead 0b cburst >> 70901b/8 mpu 0b overhead 0b level 0 >> =A0Sent 63422144616 bytes 77714246 pkt (dropped 41012275, overlimits= 0 requeues 0) >> =A0rate 600699Kbit 88756pps backlog 0b 127p requeues 0 >> =A0lended: 77714119 borrowed: 0 giants: 0 >> =A0tokens: -11 ctokens: -11 >> >> >> 900 bytes: >> class htb 1:108 parent 1:10 leaf 108: prio 7 quantum 1514 rate >> 555000Kbit ceil 555000Kbit burst 70901b/8 mpu 0b overhead 0b cburst >> 70901b/8 mpu 0b overhead 0b level 0 >> =A0Sent 76868403562 bytes 92835297 pkt (dropped 48565133, overlimits= 0 requeues 0) >> =A0rate 636195Kbit 88755pps backlog 0b 126p requeues 0 >> =A0lended: 92835171 borrowed: 0 giants: 0 >> =A0tokens: -7 ctokens: -7 >> >> >> If you need more values you're free to ask. > > Since you're so kind... :-) There is a line in net/sched/sch_htb.c: > > #define HTB_HYSTERESIS 1 =A0 =A0 =A0 =A0/* whether to use mode hyster= esis for speedup */ > > Could you change 1 to 0, and repeat these tests above after recompili= ng? > > More thanks, > Jarek P. >