Date: Sun, 13 Sep 2009 06:01:24 -0700 (PDT) From: Barney Cordoba <barney_cordoba@yahoo.com> To: Mike Tancsa <mike@sentex.net>, Jack Vogel <jfvogel@gmail.com> Cc: freebsd-net@freebsd.org Subject: Re: em driver input errors Message-ID: <883232.31223.qm@web63902.mail.re1.yahoo.com> In-Reply-To: <132674.73051.qm@web63901.mail.re1.yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
=0A=0A--- On Sun, 9/13/09, Barney Cordoba <barney_cordoba@yahoo.com> wrote:= =0A=0A> From: Barney Cordoba <barney_cordoba@yahoo.com>=0A> Subject: Re: em= driver input errors=0A> To: "Mike Tancsa" <mike@sentex.net>, "Jack Vogel" = <jfvogel@gmail.com>=0A> Cc: freebsd-net@freebsd.org=0A> Date: Sunday, Septe= mber 13, 2009, 8:40 AM=0A> =0A> =0A> --- On Sun, 9/13/09, Barney Cordoba <b= arney_cordoba@yahoo.com>=0A> wrote:=0A> =0A> > From: Barney Cordoba <barney= _cordoba@yahoo.com>=0A> > Subject: Re: em driver input errors=0A> > To: "Mi= ke Tancsa" <mike@sentex.net>,=0A> "Jack Vogel" <jfvogel@gmail.com>=0A> > Cc= : freebsd-net@freebsd.org=0A> > Date: Sunday, September 13, 2009, 8:30 AM= =0A> > =0A> > =0A> > --- On Fri, 9/11/09, Jack Vogel <jfvogel@gmail.com>=0A= > > wrote:=0A> > =0A> > > From: Jack Vogel <jfvogel@gmail.com>=0A> > > Subj= ect: Re: em driver input errors=0A> > > To: "Mike Tancsa" <mike@sentex.net>= =0A> > > Cc: "Barney Cordoba" <barney_cordoba@yahoo.com>,=0A> > freebsd-net= @freebsd.org=0A> > > Date: Friday, September 11, 2009, 12:38 PM=0A> > > Gla= d to hear this.=0A> > > =0A> > > Jack=0A> > > =0A> > > =0A> > > On Fri, Sep= 11, 2009 at 4:46 AM,=0A> > > Mike Tancsa <mike@sentex.net>=0A> > > wrote:= =0A> > > =0A> > > At 11:28 AM 9/9/2009, Mike Tancsa wrote:=0A> > > =0A> > >= =0A> > > At 11:17 AM 9/9/2009, Mike Tancsa wrote:=0A> > > =0A> > > =0A> > = > The board is an intel=0A> > > =0A> > > =0A> > > =0A> > > http://www.intel= ..com/support/motherboards/server/s3000ah/=0A> > > =0A> > > =0A> > > =0A> > = > Not sure if its wired as PCI-X or just a 32bit=0A> bus.=0A> > =A0I am=0A>= > > just popping in an em pcie nic to see if that=0A> makes a=0A> > > diff= erence. =A0I have an igb as well as bge I can=0A> try=0A> > later.=0A> > > = =0A> > > =0A> > > =0A> > > =0A> > > OK, now there is=0A> > > =0A> > > =0A> = > > =0A> > > em0@pci0:5:0:0: class=3D0x020000 card=3D0xa01f8086=0A> > > chi= p=3D0x10d38086 rev=3D0x00 hdr=3D0x00=0A> > > =0A> > >=A0 =A0 =A0vendor =A0 = =A0 =3D 'Intel Corporation'=0A> > > =0A> > >=A0 =A0 =A0class =A0 =A0 =A0=3D= network=0A> > > =0A> > >=A0 =A0 =A0subclass =A0 =3D ethernet=0A> > > =0A> = > >=A0 =A0 =A0cap 01[c8] =3D powerspec 2 =A0supports D0 D3=0A> > =A0current= =0A> > > D0=0A> > > =0A> > >=A0 =A0 =A0cap 05[d0] =3D MSI supports 1 messag= e, 64=0A> > bit=0A> > > =0A> > >=A0 =A0 =A0cap 10[e0] =3D PCI-Express 1 end= point max=0A> > data 128(256)=0A> > > link x1(x1)=0A> > > =0A> > >=A0 =A0 = =A0cap 11[a0] =3D MSI-X supports 5 messages in=0A> > map 0x1c=0A> > > enabl= ed=0A> > > =0A> > > =0A> > > =0A> > > =0A> > > =0A> > > =0A> > > =0A> > > = =0A> > > =0A> > > =0A> > > OK, much better. =A0Two nights in a row without= =0A> errors,=0A> > and=0A> > > Friday AM has a lot of level0 dumps. =A0Perh= aps as=0A> you=0A> > > speculated, the onboard NICs were wired to a=0A> slo= wer=0A> > bus...=0A> > > =A0The pcie 1x hasnt shown any errors yet.=0A> > >= =0A> > > =0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: Excessive= =0A> > collisions =3D=0A> > > 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 ke= rnel: em0: Sequence=0A> errors =3D=0A> > 0=0A> > > =0A> > > Sep 11 00:01:00= backup3 kernel: em0: Defer count=0A> =3D 0=0A> > > =0A> > > Sep 11 00:01:0= 0 backup3 kernel: em0: Missed=0A> Packets =3D=0A> > 0=0A> > > =0A> > > Sep = 11 00:01:00 backup3 kernel: em0: Receive No=0A> > Buffers =3D=0A> > > 0=0A>= > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: Receive=0A> Length=0A> >= Errors=0A> > > =3D 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0:= Receive=0A> errors =3D=0A> > 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 ke= rnel: em0: Crc errors =3D=0A> 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 ke= rnel: em0: Alignment=0A> errors=0A> > =3D 0=0A> > > =0A> > > Sep 11 00:01:0= 0 backup3 kernel: em0:=0A> > Collision/Carrier=0A> > > extension errors =3D= 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: RX overruns=0A> = =3D 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: watchdog=0A> ti= meouts=0A> > =3D 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: RX= MSIX IRQ=0A> =3D=0A> > 50004846=0A> > > TX MSIX IRQ =3D 40678847 LINK MSIX= IRQ =3D 1=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: XON Rcvd = =3D=0A> 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: XON Xmtd = =3D=0A> 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: XOFF Rcvd = =3D=0A> 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: XOFF Xmtd = =3D=0A> 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: Good Packet= s=0A> Rcvd=0A> > =3D=0A> > > 73064815=0A> > > =0A> > > Sep 11 00:01:00 back= up3 kernel: em0: Good Packets=0A> Xmtd=0A> > =3D=0A> > > 52479296=0A> > > = =0A> > > Sep 11 00:01:00 backup3 kernel: em0: TSO Contexts=0A> Xmtd=0A> > = =3D 0=0A> > > =0A> > > Sep 11 00:01:00 backup3 kernel: em0: TSO=0A> Context= s=0A> > Failed =3D=0A> > > 0=0A> > > =0A> > > =0A> > > =0A> > > =0A> > > = =0A> > > Name =A0 =A0Mtu Network =A0 =A0 =A0 Address =A0 =A0 =A0=0A> =A0 = =A0=0A> > =A0=0A> > > =A0Ipkts Ierrs =A0 =A0Opkts Oerrs =A0Coll=0A> > > =0A= > > > em0 =A0 =A01500 <Link#1> =A0 =A0=0A> > =A000:1b:21:3f:62:72=0A> > > 1= 93269942 =A0 =A0 0 133269168 =A0 =A0 0 =A0 =A0 0=0A> > > =0A> > > em0 =A0 = =A01500 10.45.129.132 10.45.129.133 =A0 =A0=0A> =A0 =A0=0A> > =A0=0A> > > = =A00 =A0 =A0 - =A0 =A0 =A0 =A00 =A0 =A0 - =A0 =A0 -=0A> > > =0A> > > em1 = =A0 =A01500 <Link#2> =A0 =A0=0A> =A000:15:17:57:31:8a=0A> > =A0=0A> > > =A0= =A0 =A00 =A0 =A0 0 =A0 =A0 =A0 =A00 =A0 =A0 0 =A0 =A0 0=0A> > > =0A> > > e= m1 =A0 =A01500 10.45.129.128 10.45.129.129 =A0 =A0=0A> =A0 =A0=0A> > =A0=0A= > > > =A00 =A0 =A0 - =A0 =A0 =A0 =A00 =A0 =A0 - =A0 =A0 -=0A> > > =0A> > > = em2* =A0 1500 <Link#3> =A0 =A0=0A> =A000:15:17:57:31:8b=0A> > =A0 =A0=0A> >= > =A0 =A00 =A0 =A0 0 =A0 =A0 =A0 =A00 =A0 =A0 0 =A0 =A0=0A> > > 0=0A> > > = =0A> > > =0A> > > =0A> > >=A0 =A0 =A0 =A0 =A0---Mike=0A> > > =0A> > =0A> > = Intel really shouldn't let MB manufacturers market=0A> dual=0A> > gigabit = =0A> > systems with 32bit controllers. The NICs aren't=0A> intended to=0A> = > be=0A> > used that way, and it makes them look bad, when its=0A> really= =0A> > the fault=0A> > of the MB manufacturer for cheaping out on the=0A> d= esign.=0A> > =0A> > Barney=0A> > =0A> =0A> I take that back. I see now that= Intel is the MB=0A> manufacturer, =0A> which is really outrageous and irre= sponsible. Jack, feel=0A> free to=0A> pass that on.=0A> =0A> Barney=0A> =0A= =0AOk, I spoke too soon. In reviewing the spec, the main controller is =0Ap= ciE, and the second controller clearly isn't designed to be used as=0Aa loa= ded port. Im not quite sure why the 2nd nic is on there. This=0AMB seems mo= re like a science project with both pciE and pciX controllers=0Aon it. But = the spec clearly says that the PCIx bus is 32big/33Mhz, so=0Athere's really= no excuse for not knowing.=0A=0ABarney=0A=0A=0A
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?883232.31223.qm>