运行基准测试时出现异常

Question

我只是想编写简单的基准 -

public class IntegerCompareBenchmark extends SimpleBenchmark {

        private Integer left;
        private Integer right;

        @Override protected void setUp() {
            left = 100;
            right = 200;
        }

        public int timeIntsCompare(int reps) {
            int val = 0;
            for (int i = 0; i < reps; i++) {
                val += Ints.compare(left, right);
            }
            return val;
        }

        public int timeIntegerCompare(int reps) {
            int val = 0;
            for (int i = 0; i < reps; i++) {
                val += left.compareTo(right);
            }
            return val;
        }


    public static void main(String[] args) throws Exception {
        Runner.main(IntegerCompareBenchmark.class, args);
    }
}

但它失败了，但出现了异常 -

Failed to execute java -cp com.google.caliper.InProcessRunner --warmupMillis 3000 --runMillis 1000 --measurementType TIME --marker //ZxJ/ -Dbenchmark=IntsCompare com.poc.IntegerCompareBenchmark
starting Scenario{vm=java, trial=0, benchmark=IntsCompare}
[caliper] [starting warmup]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
[caliper] [starting measured section]
[caliper] [done measured section]
Error: Doing 2x as much work didn't take 2x as much time! Is the JIT optimizing away the body of your benchmark?

Usage: Runner [OPTIONS...] <benchmark>

  <benchmark>: a benchmark class or suite

OPTIONS

  -D<param>=<value>: fix a benchmark parameter to a given value.
        Multiple values can be supplied by separating them with the
        delimiter specified in the --delimiter argument.

        For example: "-Dfoo=bar,baz,bat"

        "benchmark" is a special parameter that can be used to specify
        which benchmark methods to run. For example, if a benchmark has
        the method "timeFoo", it can be run alone by using
        "-Dbenchmark=Foo". "benchmark" also accepts a delimiter
        separated list of methods to run.

  -J<param>=<value>: set a JVM argument to the given value.
        Multiple values can be supplied by separating them with the
        delimiter specified in the --delimiter argument.

        For example: "-JmemoryMax=-Xmx32M,-Xmx512M"

  --delimiter <delimiter>: character or string to use as a delimiter
        for parameter and vm values.
        Default: ","

  --warmupMillis <millis>: duration to warmup each benchmark

  --runMillis <millis>: duration to execute each benchmark

  --captureVmLog: record the VM's just-in-time compiler and GC logs.
        This may slow down or break benchmark display tools.

  --measureMemory: measure the number of allocations done and the amount of
        memory used by invocations of the benchmark.
        Default: off

  --vm <vm>: executable to test benchmark on. Multiple VMs may be passed
        in as a list separated by the delimiter specified in the
        --delimiter argument.

  --timeUnit <unit>: unit of time to use for result. Depends on the units
        defined in the benchmark's getTimeUnitNames() method, if defined.
        Default Options: ns, us, ms, s

  --instanceUnit <unit>: unit to use for allocation instances result.
        Depends on the units defined in the benchmark's
        getInstanceUnitNames() method, if defined.
        Default Options: instances, K instances, M instances, B instances

  --memoryUnit <unit>: unit to use for allocation memory size result.
        Depends on the units defined in the benchmark's
        getMemoryUnitNames() method, if defined.
        Default Options: B, KB, MB, GB

  --saveResults <file/dir>: write results to this file or directory

  --printScore: if present, also display an aggregate score for this run,
        where higher is better. This number has no particular meaning,
        but can be compared to scores from other runs that use the exact
        same arguments.

  --uploadResults <file/dir>: upload this file or directory of files
        to the web app. This argument ends Caliper early and is thus
        incompatible with all other arguments.

  --debug: run without measurement for use with debugger or profiling.

  --debug-reps: fixed number of reps to run with --debug.
        Default: "1000"
[caliper] [scenarios finished]

An exception was thrown from the benchmark code.
com.google.caliper.ConfigurationException: Failed to execute java -cp <class path files> com.google.caliper.InProcessRunner --warmupMillis 3000 --runMillis 1000 --measurementType TIME --marker //ZxJ/ -Dbenchmark=IntsCompare com.poc.IntegerCompareBenchmark
    at com.google.caliper.Runner.measure(Runner.java:309)
    at com.google.caliper.Runner.runScenario(Runner.java:229)
    at com.google.caliper.Runner.runOutOfProcess(Runner.java:378)
    at com.google.caliper.Runner.run(Runner.java:97)
    at com.google.caliper.Runner.main(Runner.java:423)
    at com.google.caliper.Runner.main(Runner.java:440)
    at com.poc.IntegerCompareBenchmark.main(IntegerCompareBenchmark.java:45)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
    at java.lang.reflect.Method.invoke(Method.java:597)
    at com.intellij.rt.execution.application.AppMain.main(AppMain.java:120)

Process finished with exit code 1

我的其他基准，运行得很好..我在这里做错了什么吗？

Answer 1

看起来您的虚拟机的 JIT 将循环体优化为空。考虑使用 left 和 right 的旋转值？ （也许基于 val？）