A question about bytecodes + unsigned load performance ./. add performace

Ulf Zibis Ulf.Zibis at gmx.de
Fri Jan 30 15:43:41 PST 2009


Am 21.01.2009 19:11, Christian Thalinger schrieb:
> On Tue, 2009-01-20 at 22:12 +0100, Christian Thalinger wrote:
>   
>> On Tue, 2009-01-20 at 11:22 -0800, Tom Rodriguez wrote:
>>     
>>> I think on an out of order machine that much of this simply gets  
>>> hidden, especially when there's a lot of load and store traffic.  Try  
>>> measuring a loop without stores, maybe something that simply sums the   
>>> results.  
>>>       
>> Good idea.  I will try that tomorrow.
>>     
>
> You were absolutely right.  Simply summing up the bytes as unsigned
> values is 22% faster.
>
> I hope I can try it on a SPARC box soon.
>
> -- Christian
>
>   
Hi,

I have experienced differences, if benchmark test exceeds CPU L1-cache 
or not.

Try:
https://java-nio-charset-enhanced.dev.java.net/source/browse/java-nio-charset-enhanced/trunk/test/DecoderBenchmark.java?rev=607&view=markup

-Ulf





More information about the hotspot-compiler-dev mailing list