[Beowulf] Bernhard Kuhn's real time interrupt patch

[Patrick Geoffray]

Hi Andrew,

Andrew Piskorski wrote:
> So I was wondering, has anyone tried that?  I believe the low-latency
> OS-bypass NICs are already doing some fancy mix of polling and waiting
> for slow Linux interrupts, but shouldn't Kuhn's rtirq-patch still
> help?

I don't think this patch improves the interrupt latency or the interrupt 
overhead when there is no contention. Context switching is the metric 
you want to improve.

With OS-bypass NICs, interrupts are rare and there is usually no contention.

Patrick
-- 

Patrick Geoffray
Myricom, Inc.
http://www.myri.com