On Thu, 2014-07-03 at 15:48 +0200, Bart Van Assche wrote: > Reduce completion queue lock contention by polling for multiple > work completions at once. Limit the number of poll cycles per > completion notification to preserve fairness if multiple verbs > applications use the same port or if multiple IB interrupts have > been mapped to the same CPU core.
Do you have performance/CPU utilization numbers for this change? When we looked at this back in the 2.6.38 time-frame, there were several performance anomalies with this -- and an block-iopoll conversion -- and IIRC, neither was a win. It's possible that the issues were specific to the drivers and that newer hardware is better, but it'd be good to see the numbers. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html