Updated fine-grain locking patch for UNIX domain sockets

[Robert Watson]

On Mon, 3 Jul 2006, Robert Watson wrote:

> On Mon, 3 Jul 2006, David Xu wrote:
>
>> I found 5% performance decrease on dual P4, maybe P4 is quite bad when 
>> doing atomic operation. ;-) Thanks,
>
> When I've measured, generally, yes, P4 performance has been abysmal for 
> synchronization operations, both atomic operations and CPU-local interrupt 
> disabling, etc.
>
> I suspect rwlocks could use a bit of optimization in the contention case. 
> I've not dug into the code, so I'm not clear how they compare with respect 
> to adaptive behavior.

I ran some micro-benchmarks, and rwlocks don't perform substantially 
differently from sleep mutexes for uncontended operation -- I've not measured 
cost under contention, however.

Robert N M Watson
Computer Laboratory
University of Cambridge