carl-mastrangelo

#include <stdlib.h>

void * jmalloc(size_t size) {
  return malloc(size);
}

void jfree(void * ptr) {
  free(ptr);
}

carl-mastrangelo

Before

---------------------------------------------------------------------------------------------------------------------------
Benchmark                                                                                    Time           CPU Iterations
---------------------------------------------------------------------------------------------------------------------------
BM_HpackEncoderInitDestroy                                                                 198 ns        198 ns    3535074
BM_HpackEncoderEncodeDeadline                                                              129 ns        129 ns    5219422 framing_bytes/iter:9 header_bytes/iter:6 
BM_HpackEncoderEncodeHeader<EmptyBatch>/0/16384                                             31 ns         31 ns   22189800 framing_bytes/iter:9 header_bytes/iter:0 
BM_HpackEncoderEncodeHeader<EmptyBatch>/1/16384                                             31 ns         31 ns   22162902 framing_bytes/iter:9 header_bytes/iter:0 
BM_HpackEncoderEncodeHeader

carl-mastrangelo

# JMH version: 1.21
# VM version: JDK 13-ea, OpenJDK 64-Bit Server VM, 13-ea+22
# VM invoker: /home/carl/Downloads/jdk-13/bin/java
# VM options: -XX:+UnlockDiagnosticVMOptions -XX:+UnlockExperimentalVMOptions -XX:+EnableJVMCI -XX:+UseJVMCICompiler -XX:+LogCompilation -XX:LogFile=/tmp/blah.txt -XX:+PrintAssembly -XX:+PrintInterpreter -XX:+PrintNMethods -XX:+PrintNativeNMethods -XX:+PrintSignatureHandlers -XX:+PrintAdapterHandlers -XX:+PrintStubCode -XX:+PrintCompilation -XX:+PrintInlining -XX:+TraceClassLoading -XX:PrintAssemblyOptions=syntax -XX:PrintAssemblyOptions=intel
# Warmup: 10 iterations, 10 s each
# Measurement: 5 iterations, 1 s each
# Timeout: 10 min per iteration
# Threads: 1 thread, will synchronize iterations
# Benchmark mode: Average time, time/op

carl-mastrangelo

C2:

# JMH version: 1.21
# VM version: JDK 11.0.2, Java HotSpot(TM) 64-Bit Server VM, 11.0.2+9-LTS
# VM invoker: /home/carl/bin/jdk-11.0.2/bin/java
# VM options: -XX:+UnlockDiagnosticVMOptions -XX:+UnlockExperimentalVMOptions -XX:+LogCompilation -XX:LogFile=/tmp/blah.txt -XX:+PrintAssembly -XX:+PrintInterpreter -XX:+PrintNMethods -XX:+PrintNativeNMethods -XX:+PrintSignatureHandlers -XX:+PrintAdapterHandlers -XX:+PrintStubCode -XX:+PrintCompilation -XX:+PrintInlining -XX:+TraceClassLoading -XX:PrintAssemblyOptions=syntax -XX:PrintAssemblyOptions=intel
# Warmup: 10 iterations, 1 s each
# Measurement: 5 iterations, 1 s each
# Timeout: 10 min per iteration
# Threads: 1 thread, will synchronize iterations

carl-mastrangelo

package test;


import java.lang.invoke.MethodHandles;
import java.lang.invoke.VarHandle;
import java.util.concurrent.atomic.AtomicLong;
import org.openjdk.jcstress.annotations.Actor;
import org.openjdk.jcstress.annotations.Expect;
import org.openjdk.jcstress.annotations.JCStressTest;
import org.openjdk.jcstress.annotations.Outcome;

carl-mastrangelo

  public class DelayedForwardingClientCall<ReqT, RespT>
      extends ForwardingClientCall<ReqT, RespT>{

    private final ClientCall<ReqT, RespT> delegate;
    private final Queue<Runnable> calls = new ArrayDeque<Runnable>();
    private boolean started;

    protected DelayedForwardingClientCall(ClientCall<ReqT, RespT> delegate) {
      this.delegate = delegate;

carl-mastrangelo

openssl pkcs8 -inform PEM -nocrypt -topk8 -in $INFILE -outform PEM -out $OUTFILE

carl-mastrangelo

// https://play.golang.org/p/auAxrJQ1b1
package main

import (
	"fmt"
	"reflect"
	"unsafe"
)

type rtype struct {

carl-mastrangelo

/*


Benchmark                                                                      Mode      Cnt        Score    Error  Units
AtomicBenchmark.compareAndSet                                                sample  1665797      493.817 ±  2.426  ns/op
AtomicBenchmark.compareAndSet:compareAndSet·p0.00                            sample                25.000           ns/op
AtomicBenchmark.compareAndSet:compareAndSet·p0.50                            sample               274.000           ns/op
AtomicBenchmark.compareAndSet:compareAndSet·p0.90                            sample               950.000           ns/op
AtomicBenchmark.compareAndSet:compareAndSet·p0.95                            sample              1606.000           ns/op
AtomicBenchmark.compareAndSet:compareAndSet·p0.99                            sample              3760.000           ns/op

carl-mastrangelo

import random
import pprint

def run(tries):
  count = 0
  for i in range(tries):
     if random.randrange(1<<count) <= 0:
       count += 1
  return count