Speed Impact experiment on GCJ

Rui Wang Rui.Wang@newcastle.ac.uk
Wed Feb 15 01:41:00 GMT 2006


Having compiled the code with -O3 flag, it runs much faster compare to previous one. Thanks. 

However, I wonder why we need optimization to make native machine code running faster than Java bytecode? In my experiment, the code is no more complicated than a HelloWrold program. It has no features that will mess things up like dynamic class loading. Therefore, the translation from Java to native code should be straightforward, isn't ? 

>We can't figure out why either, becasue your test uses the files
>task.test, servers.d, and result.d, none of which you provided.

Sorry, I forgot to attach the input files. You can create a task.test file with the following information in it:

200
300
400
500
These are the number of requests. 

For servers.d, it tells that how many objects need to be created. so it should contains something like:

1
2
3

This will create three objects. You will not need result.d since it is the output file. 
Thank you. 
Rui 



More information about the Java mailing list