This content has been marked as final. Show 1 reply
I'm benchmarking RTSJ memory allocation and GC and am a little surprised by the results.
Basically, the bench consists of the following loop:
String s = new String;
long before =
s = Integer.toString(j);
This code is supplemented with calls to System.nanoTime() and runs either directly from main() or from a RealtimeThread. The bench is ran on Solaris x86.
When running the test using the "regular" JVM, s = Integer.toString(j); performs