Re: dev.ix.0.queueX.interrupt_rate

2015-08-26 Thread Gerrit Kühn
On Tue, 25 Aug 2015 19:45:08 +0300 Slawa Olhovchenkov s...@zxy.spb.ru wrote about Re: dev.ix.0.queueX.interrupt_rate: SO For discover poor network performance you need: [...] I am already through this, see https://lists.freebsd.org/pipermail/freebsd-net/2015-June/042536.html I just wanted

Re: dev.ix.0.queueX.interrupt_rate

2015-08-26 Thread Slawa Olhovchenkov
On Wed, Aug 26, 2015 at 08:31:14AM +0200, Gerrit Kuhn wrote: On Tue, 25 Aug 2015 19:45:08 +0300 Slawa Olhovchenkov s...@zxy.spb.ru wrote about Re: dev.ix.0.queueX.interrupt_rate: SO For discover poor network performance you need: [...] I am already through this, see https

Re: dev.ix.0.queueX.interrupt_rate

2015-08-25 Thread Slawa Olhovchenkov
On Tue, Aug 25, 2015 at 04:04:41PM +0200, Gerrit Kuhn wrote: On Tue, 25 Aug 2015 07:55:49 -0400 (EDT) Rick Macklem rmack...@uoguelph.ca wrote about Re: dev.ix.0.queueX.interrupt_rate: RM If you have tso enabled, you could try this patch: RM https://reviews.freebsd.org/D3477 RM RM

Re: dev.ix.0.queueX.interrupt_rate

2015-08-25 Thread Gerrit Kühn
On Tue, 25 Aug 2015 07:55:49 -0400 (EDT) Rick Macklem rmack...@uoguelph.ca wrote about Re: dev.ix.0.queueX.interrupt_rate: RM If you have tso enabled, you could try this patch: RM https://reviews.freebsd.org/D3477 RM RM If TSO is disabled, then we don't have an explanation for poor NFS RM

Re: dev.ix.0.queueX.interrupt_rate

2015-08-25 Thread Gerrit Kühn
On Mon, 24 Aug 2015 22:29:26 +0300 Slawa Olhovchenkov s...@zxy.spb.ru wrote about dev.ix.0.queueX.interrupt_rate: SO Last -stable, no tuning. Is this normal? From 10.2-rel (and still having severe performance issues with NFS as reported before): dev.ix.0.queue7.interrupt_rate: 31250 dev.ix.0

Re: dev.ix.0.queueX.interrupt_rate

2015-08-25 Thread Rick Macklem
Gerritt Kuhn wrote: On Mon, 24 Aug 2015 22:29:26 +0300 Slawa Olhovchenkov s...@zxy.spb.ru wrote about dev.ix.0.queueX.interrupt_rate: SO Last -stable, no tuning. Is this normal? From 10.2-rel (and still having severe performance issues with NFS as reported before): dev.ix.0.queue7

dev.ix.0.queueX.interrupt_rate

2015-08-24 Thread Slawa Olhovchenkov
I have '82599ES 10-Gigabit SFI/SFP+ Network Connection' and see strange: # sysctl dev.ix.0 | grep interrupt_rate dev.ix.0.queue7.interrupt_rate: 50 dev.ix.0.queue6.interrupt_rate: 50 dev.ix.0.queue5.interrupt_rate: 31250 dev.ix.0.queue4.interrupt_rate: 31250

Re: dev.ix.0.queueX.interrupt_rate

2015-08-24 Thread Eric Joyner
What's strange about it? The interrupt rate only changes if traffic goes out on the queue, and so if whatever applications you use don't utilize the core the queue is bound to, then the interrupt rate won't change. Or are you confused about the units? I know ixl(4) uses usecs instead of Hz, so

Re: dev.ix.0.queueX.interrupt_rate

2015-08-24 Thread Slawa Olhovchenkov
On Mon, Aug 24, 2015 at 10:13:05PM +, Eric Joyner wrote: What's strange about it? The interrupt rate only changes if traffic goes out on the queue, and so if whatever applications you use don't utilize the core the queue is bound to, then the interrupt rate won't change. Or are you

Re: dev.ix.0.queueX.interrupt_rate

2015-08-24 Thread Babak Farrokhi
You may want to disable interrupt moderation and manually configure interrupt rate per queue if you have high pps. I have had a lot of headaches with ix driver and fixed it by manually setting interrupt rates. sysctl hw.ix.enable_aim=0 — Babak On 25 Aug 2015, at 2:47, Slawa Olhovchenkov