Re: igb(4) Pondering a bind to cpu patch

2012-05-10 Thread Sean Bruno
On Tue, 2012-05-08 at 21:36 -0700, Sean Bruno wrote: > On Thu, 2012-05-03 at 15:33 -0700, Sean Bruno wrote: > > > > On Wed, 2012-04-25 at 12:30 -0700, Sean Bruno wrote: > > > On Wed, 2012-04-25 at 06:32 -0700, John Baldwin wrote: > > > > CPU IDs are not guaranteed to be dense. However, you can us

Re: igb(4) Pondering a bind to cpu patch

2012-05-08 Thread Sean Bruno
On Thu, 2012-05-03 at 15:33 -0700, Sean Bruno wrote: > > On Wed, 2012-04-25 at 12:30 -0700, Sean Bruno wrote: > > On Wed, 2012-04-25 at 06:32 -0700, John Baldwin wrote: > > > CPU IDs are not guaranteed to be dense. However, you can use > > > CPU_FIRST() and > > > CPU_NEXT() with your static glob

Re: igb(4) Pondering a bind to cpu patch

2012-05-03 Thread Sean Bruno
On Wed, 2012-04-25 at 12:30 -0700, Sean Bruno wrote: > On Wed, 2012-04-25 at 06:32 -0700, John Baldwin wrote: > > CPU IDs are not guaranteed to be dense. However, you can use > > CPU_FIRST() and > > CPU_NEXT() with your static global instead. > > > Ah, does CPU_NEXT() reset to 0 when it reache

Re: igb(4) Pondering a bind to cpu patch

2012-04-25 Thread John Baldwin
On Wednesday, April 25, 2012 3:30:25 pm Sean Bruno wrote: > On Wed, 2012-04-25 at 06:32 -0700, John Baldwin wrote: > > CPU IDs are not guaranteed to be dense. However, you can use > > CPU_FIRST() and > > CPU_NEXT() with your static global instead. > > > Ah, does CPU_NEXT() reset to 0 when it rea

Re: igb(4) Pondering a bind to cpu patch

2012-04-25 Thread Sean Bruno
On Wed, 2012-04-25 at 06:32 -0700, John Baldwin wrote: > CPU IDs are not guaranteed to be dense. However, you can use > CPU_FIRST() and > CPU_NEXT() with your static global instead. > Ah, does CPU_NEXT() reset to 0 when it reaches the end of its list of CPUs? > OTOH, if igb were to just leave

Re: igb(4) Pondering a bind to cpu patch

2012-04-25 Thread John Baldwin
On Tuesday, April 24, 2012 8:11:07 pm Sean Bruno wrote: > http://people.freebsd.org/~sbruno/if_igb.c.txt > > Scenario I've just seen: > > 8 core machine > 2 igb(4) interfaces > set num_queues=4 > > igb0:0 --> cpu0 > igb0:1 --> cpu1 > igb0:2 --> cpu2 > igb0:3 --> cpu3 > > igb1:0 --> cpu0 > igb1:

Re: igb(4) Pondering a bind to cpu patch

2012-04-24 Thread Juli Mallett
On Tue, Apr 24, 2012 at 17:11, Sean Bruno wrote: > http://people.freebsd.org/~sbruno/if_igb.c.txt > > 8 core machine > 2 igb(4) interfaces > set num_queues=4 > > igb0:0 --> cpu0 > [...] > igb1:0 --> cpu0 > [...] > > I suspect, that we need a static global to keep track of what cpu last > was last