From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Bellion and Thomas Heinz Subject: Re: [ANNOUNCE] nf-hipac v0.8 released Date: Wed, 2 Jul 2003 15:48:19 +0200 Sender: netdev-bounce@oss.sgi.com Message-ID: <200307021548.19989.nf@hipac.org> References: <200307021426.56138.nf@hipac.org> <3F02D964.7050301@draigBrady.com> Reply-To: nf@hipac.org Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Cc: linux-kernel@vger.kernel.org, netdev@oss.sgi.com Return-path: To: P@draigbrady.com In-Reply-To: <3F02D964.7050301@draigBrady.com> Content-Disposition: inline Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org Hi P=E1draig > > Since real world network traffic always consists of a lot of differen= t > > sized packets taking maximum sized packets is very euphemistic. 1450 = byte > > packets at 950 Mbit/s correspond to approx. 80,000 packets/sec. > > We are really interested in how our algorithm performs at higher pack= et > > rates. Our performance tests are based on 100 Mbit hardware so we cou= dn't > > test with more than approx. 80,000 packets/sec even with minimum size= d > > packets. > > Interrupt latency is the problem here. You'll require napi et. al > to get over this hump. Yes we know, but with 128 byte frame size you can archieve a packet rate = of at=20 most 97,656 packets/sec (in theory) on 100 Mbit hardware. We don't think = this=20 few more packets would have changed the results fundamentally, so it's=20 probably not worth it on 100 Mbit. Certainly you are right, that napi is required on gigabit to saturate the= link=20 with small sized packets.=20 > Cool. The same sort of test with ordinary netfilter that > I did showed it could only handle around 125 rules at this > packet rate on a 1.4GHz PIII, e1000 @ 100Mb/s. > > # ./readprofile -m /boot/System.map | sort -nr | head -30 > 6779 total 0.0047 > 4441 default_idle 69.3906 > 787 handle_IRQ_event 7.0268 > 589 ip_packet_match 1.6733 > 433 ipt_do_table 0.6294 > 106 eth_type_trans 0.5521 > [...] What do you want to show with this profile? Most of the time is spend in = the=20 idle loop and in icq handling and only a few percentage in ip_packet_matc= h=20 and ipt_do_table, so we don't quite get how this matches your statement=20 above. Could you explain this in a few words? Regards, +-----------------------+----------------------+ | Michael Bellion | Thomas Heinz | | | | +-----------------------+----------------------+