Date: Sun, 3 Jan 2010 09:00:05 -0800 (PST) From: Barney Cordoba <barney_cordoba@yahoo.com> To: Mike Tancsa <mike@sentex.net>, =?iso-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de> Cc: freebsd-net@freebsd.org Subject: Re: igb interrupt moderation Message-ID: <863054.14157.qm@web63908.mail.re1.yahoo.com> In-Reply-To: <1ABE4BCD-135C-44E0-80D8-52DFAE17B3D3@lurchi.franken.de>
next in thread | previous in thread | raw e-mail | index | archive | help
=0A=0A--- On Sun, 1/3/10, Michael T=FCxen <Michael.Tuexen@lurchi.franken.de= > wrote:=0A=0A> From: Michael T=FCxen <Michael.Tuexen@lurchi.franken.de>=0A= > Subject: Re: igb interrupt moderation=0A> To: "Mike Tancsa" <mike@sentex.= net>=0A> Cc: "Barney Cordoba" <barney_cordoba@yahoo.com>, jfvogel@gmail.com= , freebsd-net@freebsd.org=0A> Date: Sunday, January 3, 2010, 11:38 AM=0A> O= n Jan 3, 2010, at 5:23 PM, Mike=0A> Tancsa wrote:=0A> =0A> > At 11:13 AM 1/= 3/2010, Michael T=FCxen wrote:=0A> >> >=0A> >> > Just a separate datapoint = about this driver,=0A> unless I apply=0A> >> >=0A> >> > http://people.freeb= sd.org/~yongari/igb/igb.buf.patch6=0A> >> >=0A> >> > the driver is not real= ly usable for me in=0A> RELENG_8 on the dual port version of the card=0A> >= > Could you elaborate on what you mean by "not=0A> really usable"?=0A> > = =0A> > =0A> > Hi,=0A> >=A0 =A0 =A0 =A0 Some link state issues=0A> (getting = confused about what port is up), problems at high=0A> packet rates.=A0 I do= nt have this card in production, but=0A> in my test environment it was much= more stable on RELENG_8=0A> with the above patch in that I was not able to= wedge the=0A> box.=A0 pps rates were pretty ok on a low end i7 as=0A> well= .=0A> Thanks for the information. I'll give it a try. I have a=0A> problem = when I flood=0A> a system with SCTP INITs. The system under attack becomes= =0A> completely unresponsive=0A> on the console. However, it continues to s= end INIT-ACKs=0A> back. After the last=0A> commit from Jack it recovers aft= er the attack. Not yet sure=0A> what is going on.=0A> Using the em driver d= oes not have the problem. However,=0A> when using the em=0A> driver only on= e core is fully used, when using the igb=0A> driver both cores are fully=0A= > used. Unfortunately I do not have a more than dual core=0A> machine avail= able for=0A> this testing...=0A=0ATry em and lower the interrupt moderation= to something like 500 (about=0A100 packets per int is good). The latency i= sn't going to be noticable and=0Ayou'll see your cpu burden reduced quite a= bit. =0A=0AAre you using a single NIC on a server, or do you have a firewa= ll or=0Abridge?=0A=0ABarney=0A=0A=0A=0A=0A
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?863054.14157.qm>