Re: Package conflicts, breaks, and problematic upgrades... with irqbalance
On Tue, Feb 27, 2018 at 07:14:38PM +0000, Zach Marano wrote:
> irqbalance is supposed to spread interrupts across all CPU's but in GCE
> (and from what I recall KVM as well) it ends up assigning all interrupts to
> CPU0. It may be considered broken except that it was originally meant for
> bare metal hardware and therefore- maybe its not actually broken in a
> virtual environment?
Sure it is broken if it does no sensible things. If it can't do
anything, it should be a no-op and do nothing.
> With irqbalance on a 32 core VM:
> for i in $(seq 0 300); do grep . /proc/irq/$i/smp_affinity /dev/null
> 2>/dev/null; done
> /proc/irq/0/smp_affinity:ffffffff
> /proc/irq/1/smp_affinity:ffffffff
> /proc/irq/2/smp_affinity:ffffffff
> /proc/irq/3/smp_affinity:ffffffff
This looks for me like it just does not manage to balance anything.
However, I just started a system on GCE with 64 cores, installed
irqbalance and are greeted with this:
/proc/irq/28/smp_affinity:00000000,00010000
/proc/irq/29/smp_affinity:04000000,00000000
/proc/irq/30/smp_affinity:00004000,00000000
/proc/irq/31/smp_affinity:00000000,00001000
/proc/irq/32/smp_affinity:00000000,00000002
/proc/irq/33/smp_affinity:00000000,00100000
/proc/irq/34/smp_affinity:00000000,20000000
/proc/irq/35/smp_affinity:00000000,02000000
/proc/irq/36/smp_affinity:80000000,00000000
So it does something.
Bastian
--
Each kiss is as the first.
-- Miramanee, Kirk's wife, "The Paradise Syndrome",
stardate 4842.6
Reply to: