On Wed, Mar 27, 2002 at 11:02:09PM -0500, Kunal Trivedi wrote:
> > limitations of interrupt driven I/O. Interrupts are expensive, and when
> > you're pushing 100Mbit of data, that's 100k Interrupts/sec. Eventually,
> > interrupts are so frequent that the machine is not able to do anything
> > useful between them and the interrupt handlers consume all of the CPU's
> > time. I suspect this is what is happening. That would explain why the
> > throughput goes down when you pound the machine harder (more interrupts
> > consume more time, and the ability of the machine to process packets
> > decreases).
>
> yes, we thought about it but then i found that cpu is not
> reaching its max. utilization. During these experiments, cpu
> was used 17%. It was idle for 83%.
Did you just look at 'top' for determining this? Time spent in
interrupt context isn't counted in that figure :-)
Try Andrew Morton's cyclesoak.
http://www.zip.com.au/~akpm/linux/#zc
cheers,
Lennert
_______________________________________________
Bridge mailing list
[EMAIL PROTECTED]
http://www.math.leidenuniv.nl/mailman/listinfo/bridge
