java - Why JVM does't execute the code? -

i'm testing branch prediction, code this:

//first loop int sortedsum = 0; long sortedstarttime = system.nanotime(); (int = 0; < sortedarray.length; i++) {     if (sortedarray[i] < 0) {         sortedsum += sortedarray[i];     } } long sortedtime = system.nanotime() - sortedstarttime;  //second loop int randomsum = 0; long randomstarttime = system.nanotime(); (int = 0; < randomarray.length; i++) {     if (randomarray[i] < 0) {         randomsum += randomarray[i];     } } long randomtime = system.nanotime() - randomstarttime;  system.out.println("random time: " + randomtime); system.out.println("sorted time: " + sortedtime); 

it's simple print time each loop.

note sortedsum , randomsum assigned in loop never accessed.

we don't talk branch prediction here, output result. output is:

random time: 32 sorted time: 1595942 

then, put second loop before first(first randomarray loop, second sortedarray). output is:

random time: 1541919 sorted time: 40 

it seems jvm doesn't execute second loop. decompile class file, decompiler doesn't erase anything.

why? , why jvm execute first loop?

p.s. environment is:

ubuntu 12.04 lts java version "1.7.0_11" java(tm) se runtime environment (build 1.7.0_11-b21) java hotspot(tm) 64-bit server vm (build 23.6-b04, mixed mode) 

when runtime code enough triggers whole method optimised (in background)

this means if have loop iterates on 10k times , doesn't anything, timing how long takes detect loop doesn't thing , replace it.

in case of second loop, method has been optimise away, in either case.

i suggest have outer loop runs test 3 times , should see mean.

the reason can drop loop is;

• and don't use value calculated, calculatesum can dropped
• then array access isn't needed
• then loop isn't need.