[POLLING] strange interrupt/system load

[Barney Cordoba]

--- On Mon, 9/14/09, rihad <rihad at mail.ru> wrote:

> From: rihad <rihad at mail.ru>
> Subject: Re: [POLLING] strange interrupt/system load
> To: "Barney Cordoba" <barney_cordoba at yahoo.com>
> Cc: freebsd-net at FreeBSD.org
> Date: Monday, September 14, 2009, 7:23 AM
> Barney Cordoba wrote:
> > 
> >> Without polling (current load around 190-200
> mbit/s, around
> >> 24-26 kpps):
> >>
> >> top:
> >> CPU:  0.0% user,  0.0% nice,  8.4%
> >> system,  0.0% interrupt, 91.6% idle
> >>
> >> Interrupts/s: 18322 total
> >> 28 mpt0 irq16
> >> 1999 cpu0: time
> >> 6906 em0 irq256
> >> 3392 em1 irq257
> >> 1999 cpu1: time
> >> 1999 cpu2: time
> >> 1999 cpu3: time
> > 
> > You really need to look at the taskq usage as
> averaging on a 4 core
> CPU:  0.0% user,  0..0% nice, 10.0% system, 
> 0.0% interrupt, 90.0% idle
>     27 root        1
> -68    -     0K   
> 16K -      1 137:47 40.28% em0 taskq
>     28 root        1
> -68    -     0K   
> 16K -      2   5:05 
> 0.88% em1 taskq
> 
> > You'll do a lot better setting your ITR to 2000 or so.
> You really don't
> > need an interrupt every 4 packets at those traffic
> levels.
> 
> Sorry, how would I do that? And how do I find the current
> ITR value?
> 
I made mine a sysctl long ago, so I'm not sure what the current state
of em is. It used to be a macro MAX_INTS_PER_SEC

Barney