<div class="gmail_quote">2010/10/13 Jean-Michel Hautbois <span dir="ltr"><<a href="mailto:jhautbois@gmail.com">jhautbois@gmail.com</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="gmail_quote">2010/10/12 Stephen Hemminger <span dir="ltr"><<a href="mailto:shemminger@linux-foundation.org" target="_blank">shemminger@linux-foundation.org</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="im">
<div><div></div><div>> Thus, my jitter is 47µs.<br>
> This may seem low, but this is about two to three times what I expect and I<br>
> cannot see why a bridge which is only forwarding a frame would have jitter ?<br>
> If you have any idea, everything is interesting for me.<br>
> I thought maybe to skb_clone ?<br>
<br>
</div></div></div><div class="im">You probably have NAPI and/or interrupt coalescing setting<br>
enabled on one or both of the devices.<br>
<font color="#888888"><br>
--<br>
</font></div></blockquote></div><br>Hi Stephen,<br><br>I have NAPI on my ethernet device, at least.<br>I will check for interrupt coalescing.<br>AFAIK, there is a skb_clone followed by a kfree in netif_receive_skb (the frame coming from eth0 to br0).<br>
There is a jitter in this call of "freeing memory". I am looking at the path, in order to see why there is a clone.<br><br>BR,<br><font color="#888888">JM<br>
</font></blockquote></div><br>I have to tell I am working on a mpc52xx and am currently only interested in the eth->br path for the moment.<br>JM<br>