Re: [PATCH v6] x86/apic: limit irq affinity

[Yinghai Lu <yinghai@kernel.org>]

Ingo Molnar wrote:
> * Eric W. Biederman <ebiederm@xmission.com> wrote:
> 
>> Currently the irq code treats /proc/irq/N/smp_affinity as a strong 
>> hint on where we would like interrupts to be delivered, and we don't 
>> have good feedback from there to architecture specific code that knows 
>> what we really can do.  It is going to take some effort and some work 
>> to make that happen.
>>
>> I think the irq scheduler is the only scheduler (except for batch 
>> jobs) that we don't put in the kernel.  It seems to me that if we are 
>> going to go to all of the trouble to rewrite the generic code to 
>> better support irqbalance because we are having serious irqbalance 
>> problems, it will be less effort to suck irqbalance into the kernel 
>> along with everything else.
>>
>> I really think irqbalancing belongs in the kernel. [...]
> 
> Interesting. I've yet to see a solution that is maintainable and works 
> well, without putting too much policy into the kernel. Our previous 
> solutions didnt really work all that well.
> 
> What would your model be, and can it be implemented reasonably?

we already have dev numa node, so could just make irqblance to some smart to use
that device node corresponding for irq that is binding to the device.

YH