mfence on i686 with volatile?
Tim.Bell at Sun.COM
Wed Nov 18 00:22:32 PST 2009
David Holmes - Sun Microsystems wrote:
> Is this with -client or -server ?
Also - are you running your tests on a single-CPU machine? I recall HotSpot
detects that and will elide any n-way synchronization measures if so...
> For x86 we should see a StoreLoad barrier after the assignment to v2:
> For -server printOptoAssembly shows our "mfence" replacement a lock:addl
> (which is the implementation of membar-volatile().
> [ Got to go get hsdis myself to test printAssembly directly ]
More information about the hotspot-dev