[concurrency-interest] The very best CAS loop

Peter Levart peter.levart at gmail.com
Wed Sep 28 12:34:36 EDT 2016



On 09/28/2016 05:19 PM, Andrew Haley wrote:
> On 28/09/16 15:29, Peter Levart wrote:
>> Yes, but that's an entirely different benchmark. This just tests raw
>> speed of a single thread in its own cache-line.
> Well, exactly, that's the idea.  I wanted to make sure that we really
> had false sharing in the test case; I can do that by making sure that
> when there is no false sharing things really are much faster.
>
> Andrew.
>

So the updateFn was the same? Hm, then I really must see the results of 
this on your AArch64:

http://cr.openjdk.java.net/~plevart/misc/GetAndUpdate/GetAndUpdateBench3.java
http://cr.openjdk.java.net/~plevart/misc/GetAndUpdate/benchmarks.jar

Regards, Peter

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.oswego.edu/pipermail/concurrency-interest/attachments/20160928/7255eec9/attachment.html>


More information about the Concurrency-interest mailing list