Peter Maydell <peter.mayd...@linaro.org> writes: > On 4 April 2013 17:59, Anthony Liguori <aligu...@us.ibm.com> wrote: >> So I think this is a long way of saying: >> >> Reviewed-by: Anthony Liguori <aligu...@us.ibm.com> > > Any chance we could update the commit message to include > this more authoritative analysis?
Yes, please do. I'm also not sure that just always dropping the lock is the best strategy either. I think a simple counter with a nice comment that explains why we need to periodically drop the lock is a better solution as it gives a way to experiment with the right value to ensure fairness between the threads without having excessing lock acquisition/release. Regards, Anthony Liguori > > thanks > -- PMM
