[openib-general] [PATCH] for review to timeout send MADs

[Roland Dreier]

> Any idea on where the cutoff for huge is or is this likely to be a
> matter of experience ? Is it at least 2 ? What about 4 ?

Depends on the number of CPUs and the workload.  The single workqueue
design starts being inefficient when you start getting idle time because every
workqueue thread is asleep waiting for an HCA to do something (like a modify
QP firmware command).  If there are other HCAs ready to do useful work they
won't be able to get to it and throughput will be lower than it would be with a
thread per HCA port.

 - R.