On 09/20/2017 12:12 PM, Kazunori Ogata wrote:
By the way, I'm running the benchmark on POWER8 machine. Removing volatile won't change the performance on x86.
I would like to compare different variants of optimization using final fields. I won't compare it with current code that uses volatile. I think that overhead that is caused by fences imposed on POWER8 by final fields (does POWER8 use any memory ordering instructions on the read side for final fields?) can be offset with better overall algorithm.
Regards, Peter
