Why -O0 and -O2 have the same performance？

[rppp]

./cling -O0 '#include <stdio.h> ' '#include <time.h>' 'int fib(int n){if(n<=1){return 1;}return fib(n-1)+fib(n-2);} auto start=clock();printf("%d\n",fib(40));printf("%f\n",(clock()-start)/(double)CLOCKS_PER_SEC);'

./cling -O2 '#include <stdio.h> ' '#include <time.h>' 'int fib(int n){if(n<=1){return 1;}return fib(n-1)+fib(n-2);} auto start=clock();printf("%d\n",fib(40));printf("%f\n",(clock()-start)/(double)CLOCKS_PER_SEC);'

Can I use the “-force-interpreter” of LLVM/lli option?

[rppp]

In my experience, it seems that the performance of cling is the same as that of g++ -O0, so cling uses JIT -O0.