On Thu, 2006-09-21 at 19:31 -0700, harish wrote: > How did the CPU utilizations compare for the NAPI vs. no NAPI case? > What are your thoughts on what bottleneck you are hitting? >
The CPU utilization reported by netperf is not accurate since it is not reported on a per cpu basis and I don't have one number to reliably describe utilization. I think the bottleneck is that the CPU that handles the softirq that handles the rx sk_buffs is 100% utilized and it dictates the limit. _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
