<br><font size=2><tt>"Michael S. Tsirkin" <mst@mellanox.co.il>
wrote on 05/09/2006 11:44:52 AM:<br>
<br>
> Quoting r. Shirley Ma <xma@us.ibm.com>:<br>
> > According to some results from different resouces, NAPI only
gives<br>
> 3%-10% performance improvement on single CQ.<br>
> <br>
> When you say performance you mean bandwidth.<br>
> But I think it should improve the CPU utilization on RX side significantly.<br>
> If it does, that an important metric as well.<br>
<br>
No, CPU utilization wasn't reduced. When you use single CQ, NAPI polls
on both RX/TX.</tt></font>
<br>
<br><font size=2><tt>> > I am trying a simple NAPI patch on splitting
CQ now to see how <br>
> much performance there.<br>
> <br>
> What are you using for a benchmark?<br>
> <br>
> -- <br>
> MST</tt></font>
<br>
<br><font size=2 face="sans-serif">netperf, iperf, mpstat, netpipe, oprofiling,
what's your suggestion?</font>
<br><font size=2 face="sans-serif"><br>
Thanks<br>
Shirley Ma<br>
IBM Linux Technology Center<br>
15300 SW Koll Parkway<br>
Beaverton, OR 97006-6063<br>
Phone(Fax): (503) 578-7638<br>
<br>
</font>
<br>
<br><font size=2><tt><br>
</tt></font>