<html><body>
<p><tt>Roland Dreier <rdreier@cisco.com> wrote on 10/19/2006 09:10:35 PM:<br>
> It's entirely possible that my implementation of the missing event<br>
> hint in ehca is wrong. I just guessed based on how poll CQ is<br>
> implemented -- if the consumer requests a hint about missing events,<br>
> then I lock the CQ and check if its empty after requesting<br>
> notification.<br>
> <br>
> I looked over my code again, and I don't see anything obviously wrong,<br>
> but it's quite possible I made a mistake that I just can't see right<br>
> now (like reversing a truth value somewhere). Someone who knows how<br>
> ehca works might be able to spot the error.<br>
> <br>
> - R.<br>
</tt><br>
<tt>The oprofile data (with your napi + this hack patch) looks good, it reduced cpu utilization significantly. (I was wrong about cpu utilization.) I will talk with ehca team regarding this missing event hint patch on ehca.</tt><br>
<br>
thanks<br>
Shirley Ma<br>
IBM Linux Technology Center<br>
</body></html>