Date: Thu, 28 Apr 2011 10:13:39 -0400 From: Adam Stylinski <kungfujesus06@gmail.com> To: Steven Hartland <killing@multiplay.co.uk> Cc: freebsd-net@freebsd.org Subject: Re: em0 performance subpar Message-ID: <20110428141339.GD2800@ossumpossum.geop.uc.edu> In-Reply-To: <B04EE072E7F04402AC289EFBF55E1D05@multiplay.co.uk> References: <20110428072946.GA11391@zephyr.adamsnet> <4DB965D8.7090906@sentex.net> <B04EE072E7F04402AC289EFBF55E1D05@multiplay.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
--NtwzykIc2mflq5ck Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Apr 28, 2011 at 02:52:59PM +0100, Steven Hartland wrote: > Running em's here we regularly see them hitting pretty much line rate > although there are a lot of different em's >=20 > Here we have the following under 8.0+ > em0@pci0:6:0:0: class=3D0x020000 card=3D0x000015d9 chip=3D0x10968086 rev= =3D0x01 hdr=3D0x00 > vendor =3D 'Intel Corporation' > device =3D 'Intel PRO/1000 EB (Intel PRO/1000 EB)' > class =3D network > subclass =3D ethernet >=20 > You don't say which OS version your running? >=20 > Regards > Steve >=20 > ----- Original Message -----=20 > From: "Mike Tancsa" <mike@sentex.net> > To: "Adam Stylinski" <kungfujesus06@gmail.com> > Cc: <freebsd-net@freebsd.org> > Sent: Thursday, April 28, 2011 2:04 PM > Subject: Re: em0 performance subpar >=20 >=20 > > On 4/28/2011 3:29 AM, Adam Stylinski wrote: > >> Hello, > >> > >> I have an intel gigabit network adapter (the 1000 GT w/chipset 82541PI= ) which performs poorly in Freebsd compared to the same=20 > >> card in Linux. I've tried this card in two different freebsd boxes an= d for whatever reason I get poor transmit performance.=20 > >> I've done all of the tweaking specified in just about every guide out = there (the usual TCP window scaling, larger nmbclusters,=20 > >> delayed acks, etc) and still I get only around 600mbps. I'm using jum= bo frames, with an MTU of 9000. I'm testing this with=20 > >> iperf. While I realize that this may not be the most realistic test, = linux hosts with the same card can achieve 995Mbit/s to=20 > >> another host running this. When the Freebsd box is the server, Linux = hosts can transmit to it at around 800 something Mbit/s.=20 > >> I've increased the transmit descriptors as specified in the if_em man = page, and while that gave me 20 or 30 more mbit/s, my=20 > >> transmit performance is still below normal. > >> > >> sysctl stats report that the card is trigger a lot of tx_desc_fail2: > >> dev.em.0.tx_desc_fail2: 3431 > > > > Try the tests using the tools in /usr/src/tools/tools/netperf to > > generate / test udp traffic. > > > > Perhaps give the driver from HEAD a try. There are a few fixes to it. I > > back ported it to RELENG_8, but it should work on 8.2R as well. > > > > http://www.tancsa.com/em-723.tgz > > > > what does pciconf -lvc for your em NIC show ? > > also, vmstat -i > > > > ---Mike > > > > > > > > --=20 > > ------------------- > > Mike Tancsa, tel +1 519 651 3400 > > Sentex Communications, mike@sentex.net > > Providing Internet services since 1994 www.sentex.net > > Cambridge, Ontario Canada http://www.tancsa.com/ > > _______________________________________________ > > freebsd-net@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-net > > To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org" > >=20 >=20 >=20 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > This e.mail is private and confidential between Multiplay (UK) Ltd. and t= he person or entity to whom it is addressed. In the event of misdirection, = the recipient is prohibited from using, copying, printing or otherwise diss= eminating it or any information contained in it.=20 >=20 > In the event of misdirection, illegible or incomplete transmission please= telephone +44 845 868 1337 > or return the E.mail to postmaster@multiplay.co.uk. >=20 I am running FreeBSD 8.2-RELEASE on one of the boxes with the issues, and t= he other box which has the same issue is 8-STABLE. The issue doesn't seem = isolated to one box though and like I said, Linux seems to not have this is= sue. --=20 Adam Stylinski PGP Key: http://pohl.ececs.uc.edu/~adam/publickey.pub Blog: http://technicallyliving.blogspot.com --NtwzykIc2mflq5ck Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (FreeBSD) iQIcBAEBAgAGBQJNuXYSAAoJED6sRHE6TvmnVyYP/1hIzq7S7JMvNPPqbt2eN+yN cN9e4qKnrjRpui5fbufI4WPlTGzKCtN69vhgkXq1k5oKTF9We48UN+eWiLmi4QP+ tGbyHmA6GQMLhohq8iPOiPLJnigIyk5kXxlIMlA5Wf6toKwmQHugtUv1FLHX3ml2 pYO+eZ7xkqHTjy2QwtBkq4TqiSAxi6WG5HMPA5pBEHe0TOgPyGDq4mNK0FXpama4 lOSLpnV/wCSXIscCfJ+HrcVIsCfgWpsC0sjroowm8jSmjuXQweOn0hHvut8puD54 9zeRlLiy6WLhP3BQo2ByJ9FBMJpoDjoh08Bm4/2uXUgcg5aFByNR6YTjuEtl+bfU 8FLKMg0LIEmexgYfWRSJ9RZ8eHmVCmI3bh+5bGaxcYfGRFYd/DLWaeNaosdQs5Sa e7CmaQaHhylIG/NsagQJo+pcbxtLF+4sY7atEiijidS0Yb1DnZlFpgU/G/l/PFKm 6J4S8wPq0COmZgOWJHcIJvHOXxKYTY5jwDL5fhlmwjHQgKGh6ZLF+HSAR9UmRr/X YZGbTpONcc9oZ2A/LSQi3zPNNCCZpMIbEyH/CpWfBgYIyujdJpf/nQ/0PhZmZa9l 1T2FduEjtBNdlz+bQULmxnKtf57sf+ocK76iqJ3tXfSZiLTaKkEOWCfmc9sbRFsO JbyLK9BMcHZ0rYWZELtn =/b2b -----END PGP SIGNATURE----- --NtwzykIc2mflq5ck--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110428141339.GD2800>