diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..6214df7
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,59 @@
+# Built application files
+*.apk
+*.ap_
+
+# Files for the ART/Dalvik VM
+*.dex
+
+# Java class files
+*.class
+
+# Generated files
+bin/
+gen/
+out/
+
+# Gradle files
+.gradle/
+build/
+
+# Local configuration file (sdk path, etc)
+local.properties
+
+# Proguard folder generated by Eclipse
+proguard/
+
+# Log Files
+*.log
+
+# Android Studio Navigation editor temp files
+.navigation/
+
+# Android Studio captures folder
+captures/
+
+# IntelliJ
+*.iml
+.idea/
+
+# Keystore files
+# Uncomment the following line if you do not want to check your keystore files in.
+#*.jks
+
+# External native build folder generated in Android Studio 2.2 and later
+.externalNativeBuild
+
+# Google Services (e.g. APIs or Firebase)
+google-services.json
+
+# Freeline
+freeline.py
+freeline/
+freeline_project_description.json
+
+# fastlane
+fastlane/report.xml
+fastlane/Preview.html
+fastlane/screenshots
+fastlane/test_output
+fastlane/readme.md
diff --git a/.gitmodules b/.gitmodules
new file mode 100644
index 0000000..e69de29
diff --git a/app/.gitignore b/app/.gitignore
new file mode 100644
index 0000000..796b96d
--- /dev/null
+++ b/app/.gitignore
@@ -0,0 +1 @@
+/build
diff --git a/app/build.gradle b/app/build.gradle
new file mode 100644
index 0000000..672a34e
--- /dev/null
+++ b/app/build.gradle
@@ -0,0 +1,38 @@
+apply plugin: 'com.android.application'
+
+android {
+    compileSdkVersion 29
+    defaultConfig {
+        applicationId "com.example.yentencoin.yentenminer"
+        minSdkVersion 21
+        targetSdkVersion 29
+        versionCode 1
+        versionName "1.0"
+        testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
+    }
+    buildTypes {
+        release {
+            minifyEnabled false
+            proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'
+        }
+    }
+    flavorDimensions 'default'
+    productFlavors {
+        standard {
+        }
+        kitkat {
+            minSdkVersion 19
+        }
+    }
+}
+
+dependencies {
+    implementation project(':bitzenymininglibrary')
+    implementation fileTree(dir: 'libs', include: ['*.jar'])
+    implementation 'com.android.support:appcompat-v7:26.1.0'
+    implementation 'com.android.support.constraint:constraint-layout:1.0.2'
+    testImplementation 'junit:junit:4.12'
+    androidTestImplementation 'com.android.support.test:runner:1.0.1'
+    androidTestImplementation 'com.android.support.test.espresso:espresso-core:3.0.1'
+    androidTestImplementation 'com.microsoft.appcenter:espresso-test-extension:1.2'
+}
diff --git a/app/proguard-rules.pro b/app/proguard-rules.pro
new file mode 100644
index 0000000..f1b4245
--- /dev/null
+++ b/app/proguard-rules.pro
@@ -0,0 +1,21 @@
+# Add project specific ProGuard rules here.
+# You can control the set of applied configuration files using the
+# proguardFiles setting in build.gradle.
+#
+# For more details, see
+#   http://developer.android.com/guide/developing/tools/proguard.html
+
+# If your project uses WebView with JS, uncomment the following
+# and specify the fully qualified class name to the JavaScript interface
+# class:
+#-keepclassmembers class fqcn.of.javascript.interface.for.webview {
+#   public *;
+#}
+
+# Uncomment this to preserve the line number information for
+# debugging stack traces.
+#-keepattributes SourceFile,LineNumberTable
+
+# If you keep the line number information, uncomment this to
+# hide the original source file name.
+#-renamesourcefileattribute SourceFile
diff --git a/app/src/androidTest/java/com/example/ottylab/bitzenyminer/MainActivityTest.java b/app/src/androidTest/java/com/example/ottylab/bitzenyminer/MainActivityTest.java
new file mode 100644
index 0000000..867179e
--- /dev/null
+++ b/app/src/androidTest/java/com/example/ottylab/bitzenyminer/MainActivityTest.java
@@ -0,0 +1,104 @@
+package com.example.ottylab.bitzenyminer;
+
+
+import android.support.test.espresso.ViewInteraction;
+import android.support.test.rule.ActivityTestRule;
+import android.support.test.runner.AndroidJUnit4;
+import android.test.suitebuilder.annotation.LargeTest;
+import android.view.View;
+import android.view.ViewGroup;
+import android.view.ViewParent;
+
+import com.microsoft.appcenter.espresso.Factory;
+import com.microsoft.appcenter.espresso.ReportHelper;
+
+import org.hamcrest.Description;
+import org.hamcrest.Matcher;
+import org.hamcrest.TypeSafeMatcher;
+import org.junit.After;
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.runner.RunWith;
+
+import static android.support.test.espresso.Espresso.onView;
+import static android.support.test.espresso.action.ViewActions.click;
+import static android.support.test.espresso.assertion.ViewAssertions.matches;
+import static android.support.test.espresso.matcher.ViewMatchers.isDisplayed;
+import static android.support.test.espresso.matcher.ViewMatchers.withClassName;
+import static android.support.test.espresso.matcher.ViewMatchers.withId;
+import static android.support.test.espresso.matcher.ViewMatchers.withText;
+import static org.hamcrest.Matchers.allOf;
+import static org.hamcrest.Matchers.containsString;
+import static org.hamcrest.Matchers.is;
+
+@LargeTest
+@RunWith(AndroidJUnit4.class)
+public class MainActivityTest {
+
+    @Rule
+    public ActivityTestRule<MainActivity> mActivityTestRule = new ActivityTestRule<>(MainActivity.class);
+
+    @Rule
+    public ReportHelper reportHelper = Factory.getReportHelper();
+
+    @After
+    public void TearDown(){
+        reportHelper.label("Stopping App");
+    }
+
+    @Test
+    public void mainActivityTest() {
+        ViewInteraction appCompatCheckBox = onView(
+                allOf(withId(R.id.checkBoxBenchmark), withText("Benchmark"),
+                        childAtPosition(
+                                childAtPosition(
+                                        withClassName(is("android.widget.LinearLayout")),
+                                        4),
+                                1),
+                        isDisplayed()));
+        appCompatCheckBox.perform(click());
+
+        ViewInteraction appCompatButton = onView(
+                allOf(withId(R.id.buttonDrive), withText("Start"),
+                        childAtPosition(
+                                childAtPosition(
+                                        withClassName(is("android.widget.LinearLayout")),
+                                        4),
+                                0),
+                        isDisplayed()));
+        appCompatButton.perform(click());
+
+        try{
+            Thread.sleep(5000);
+        } catch (InterruptedException e){}
+
+        ViewInteraction textView = onView(
+                allOf(withId(R.id.textViewLog),
+                        childAtPosition(
+                                childAtPosition(
+                                        withId(android.R.id.content),
+                                        0),
+                                5),
+                        isDisplayed()));
+        textView.check(matches(withText(containsString("miner threads started, using 'yescrypt' algorithm."))));
+    }
+
+    private static Matcher<View> childAtPosition(
+            final Matcher<View> parentMatcher, final int position) {
+
+        return new TypeSafeMatcher<View>() {
+            @Override
+            public void describeTo(Description description) {
+                description.appendText("Child at position " + position + " in parent ");
+                parentMatcher.describeTo(description);
+            }
+
+            @Override
+            public boolean matchesSafely(View view) {
+                ViewParent parent = view.getParent();
+                return parent instanceof ViewGroup && parentMatcher.matches(parent)
+                        && view.equals(((ViewGroup) parent).getChildAt(position));
+            }
+        };
+    }
+}
diff --git a/app/src/main/AndroidManifest.xml b/app/src/main/AndroidManifest.xml
new file mode 100644
index 0000000..0e1c2c2
--- /dev/null
+++ b/app/src/main/AndroidManifest.xml
@@ -0,0 +1,21 @@
+<?xml version="1.0" encoding="utf-8"?>
+<manifest xmlns:android="http://schemas.android.com/apk/res/android"
+    package="com.example.ottylab.bitzenyminer">
+    <uses-permission android:name="android.permission.INTERNET" />
+
+    <application
+        android:allowBackup="true"
+        android:icon="@mipmap/ic_launcher"
+        android:label="@string/app_name"
+        android:supportsRtl="true"
+        android:theme="@style/AppTheme">
+        <activity android:name="com.example.ottylab.bitzenyminer.MainActivity">
+            <intent-filter>
+                <action android:name="android.intent.action.MAIN" />
+
+                <category android:name="android.intent.category.LAUNCHER" />
+            </intent-filter>
+        </activity>
+    </application>
+
+</manifest>
\ No newline at end of file
diff --git a/app/src/main/java/com/example/ottylab/bitzenyminer/MainActivity.java b/app/src/main/java/com/example/ottylab/bitzenyminer/MainActivity.java
new file mode 100644
index 0000000..1efdba1
--- /dev/null
+++ b/app/src/main/java/com/example/ottylab/bitzenyminer/MainActivity.java
@@ -0,0 +1,200 @@
+package com.example.ottylab.bitzenyminer;
+
+import android.content.SharedPreferences;
+import android.os.Build;
+import android.os.Handler;
+import android.os.Message;
+import android.preference.PreferenceManager;
+import android.support.v7.app.AppCompatActivity;
+import android.os.Bundle;
+import android.text.method.ScrollingMovementMethod;
+import android.util.Log;
+import android.view.View;
+import android.widget.AdapterView;
+import android.widget.ArrayAdapter;
+import android.widget.Button;
+import android.widget.CheckBox;
+import android.widget.EditText;
+import android.widget.Spinner;
+import android.widget.TextView;
+
+import com.example.ottylab.bitzenymininglibrary.BitZenyMiningLibrary;
+
+import java.lang.ref.WeakReference;
+import java.util.concurrent.BlockingQueue;
+import java.util.concurrent.LinkedBlockingQueue;
+
+public class MainActivity extends AppCompatActivity {
+    private static final String TAG = "BitZenyMiner";
+    private static final int LOG_LINES = 1000;
+
+    private BitZenyMiningLibrary miner;
+
+    private EditText editTextServer;
+    private EditText editTextUser;
+    private EditText editTextPassword;
+    private EditText editTextNThreads;
+    private Button buttonDrive;
+    private CheckBox checkBoxBenchmark;
+    private Spinner spinnerAlgorithm;
+    private TextView textViewLog;
+
+    private boolean running;
+    private BlockingQueue<String> logs = new LinkedBlockingQueue<>(LOG_LINES);
+
+    private static class JNICallbackHandler extends Handler {
+        private final WeakReference<MainActivity> activity;
+
+        public JNICallbackHandler(MainActivity activity) {
+            this.activity = new WeakReference<MainActivity>(activity);
+        }
+
+        @Override
+        public void handleMessage(Message msg) {
+            MainActivity activity = this.activity.get();
+            if (activity != null) {
+                String log = msg.getData().getString("log");
+                String logs = Utils.rotateStringQueue(activity.logs, log);
+                activity.textViewLog.setText(logs);
+                Log.d(TAG, log);
+            }
+        }
+    }
+
+    private static JNICallbackHandler sHandler;
+
+    @Override
+    protected void onCreate(Bundle savedInstanceState) {
+        super.onCreate(savedInstanceState);
+        setContentView(R.layout.activity_main);
+
+        showDeviceInfo();
+
+        sHandler = new JNICallbackHandler(this);
+        miner = new BitZenyMiningLibrary(sHandler);
+
+        editTextServer = (EditText) findViewById(R.id.editTextServer);
+        editTextServer.setOnFocusChangeListener(new View.OnFocusChangeListener() {
+            @Override
+            public void onFocusChange(View view, boolean b) {
+                storeSetting();
+            }
+        });
+
+        editTextUser = (EditText) findViewById(R.id.editTextUser);
+        editTextUser.setOnFocusChangeListener(new View.OnFocusChangeListener() {
+            @Override
+            public void onFocusChange(View view, boolean b) {
+                storeSetting();
+            }
+        });
+
+        editTextPassword= (EditText) findViewById(R.id.editTextPassword);
+        editTextPassword.setOnFocusChangeListener(new View.OnFocusChangeListener() {
+            @Override
+            public void onFocusChange(View view, boolean b) {
+                storeSetting();
+            }
+        });
+
+        editTextNThreads = (EditText) findViewById(R.id.editTextNThreads);
+        editTextNThreads.setOnFocusChangeListener(new View.OnFocusChangeListener() {
+            @Override
+            public void onFocusChange(View view, boolean b) {
+                storeSetting();
+            }
+        });
+
+        buttonDrive = (Button) findViewById(R.id.buttonDrive);
+        buttonDrive.setOnClickListener(new View.OnClickListener() {
+            @Override
+            public void onClick(View view) {
+                if (running) {
+                    Log.d(TAG, "Stop");
+                    miner.stopMining();
+                } else {
+                    Log.d(TAG, "Start");
+                    int n_threads = 0;
+                    try {
+                        n_threads = Integer.parseInt(editTextNThreads.getText().toString());
+                    } catch (NumberFormatException e){}
+
+                    BitZenyMiningLibrary.Algorithm algorithm =
+                            spinnerAlgorithm.getSelectedItemPosition() == 0 ?
+                                    BitZenyMiningLibrary.Algorithm.YESCRYPT : BitZenyMiningLibrary.Algorithm.YESPOWER;
+                    if (checkBoxBenchmark.isChecked()) {
+                        miner.startBenchmark(n_threads, algorithm);
+                    } else {
+                        miner.startMining(
+                            editTextServer.getText().toString(),
+                            editTextUser.getText().toString(),
+                            editTextPassword.getText().toString(),
+                            n_threads,
+                            algorithm);
+                    }
+                }
+
+                changeState(!running);
+                storeSetting();
+            }
+        });
+
+        checkBoxBenchmark = (CheckBox) findViewById(R.id.checkBoxBenchmark);
+        spinnerAlgorithm = (Spinner) findViewById(R.id.spinnerAlgorithm);
+        ArrayAdapter<CharSequence> adapter = ArrayAdapter.createFromResource(this, R.array.algorithms, android.R.layout.simple_spinner_item);
+        adapter.setDropDownViewResource(android.R.layout.simple_spinner_dropdown_item);
+        spinnerAlgorithm.setAdapter(adapter);
+
+        textViewLog = (TextView) findViewById(R.id.textViewLog);
+        textViewLog.setMovementMethod(new ScrollingMovementMethod());
+
+        restoreSetting();
+        changeState(miner.isMiningRunning());
+    }
+
+    private void changeState(boolean running) {
+        buttonDrive.setText(running ? "Stop" : "Start");
+        disableSetting(running);
+        this.running = running;
+    }
+
+    private void disableSetting(boolean running) {
+        editTextServer.setEnabled(!running);
+        editTextUser.setEnabled(!running);
+        editTextPassword.setEnabled(!running);
+        editTextNThreads.setEnabled(!running);
+        spinnerAlgorithm.setEnabled(!running);
+    }
+
+    private void storeSetting() {
+        SharedPreferences pref = PreferenceManager.getDefaultSharedPreferences(this);
+        SharedPreferences.Editor editor = pref.edit();
+        editor.putString("server", editTextServer.getText().toString());
+        editor.putString("user", editTextUser.getText().toString());
+        editor.putString("password", editTextPassword.getText().toString());
+        editor.putString("n_threads", editTextNThreads.getText().toString());
+        editor.putInt("algorithm", spinnerAlgorithm.getSelectedItemPosition());
+        editor.commit();
+    }
+
+    private void restoreSetting() {
+        SharedPreferences pref = PreferenceManager.getDefaultSharedPreferences(this);
+        editTextServer.setText(pref.getString("server", null));
+        editTextUser.setText(pref.getString("user", null));
+        editTextPassword.setText(pref.getString("password", null));
+        editTextNThreads.setText(pref.getString("n_threads", null));
+        spinnerAlgorithm.setSelection(pref.getInt("algorithm", 0));
+    }
+
+    private void showDeviceInfo() {
+        String[] keys = new String[]{ "os.arch", "os.name", "os.version" };
+        for (String key : keys) {
+            Log.d(TAG, key + ": " + System.getProperty(key));
+        }
+        Log.d(TAG, "CODE NAME: " + Build.VERSION.CODENAME);
+        Log.d(TAG, "SDK INT: " + Build.VERSION.SDK_INT);
+        Log.d(TAG, "MANUFACTURER: " + Build.MANUFACTURER);
+        Log.d(TAG, "MODEL: " + Build.MODEL);
+        Log.d(TAG, "PRODUCT: " + Build.PRODUCT);
+    }
+}
diff --git a/app/src/main/java/com/example/ottylab/bitzenyminer/Utils.java b/app/src/main/java/com/example/ottylab/bitzenyminer/Utils.java
new file mode 100644
index 0000000..07f8355
--- /dev/null
+++ b/app/src/main/java/com/example/ottylab/bitzenyminer/Utils.java
@@ -0,0 +1,26 @@
+package com.example.ottylab.bitzenyminer;
+
+import java.util.concurrent.BlockingQueue;
+
+/**
+ * Created by ochikage on 2018/02/18.
+ */
+
+public class Utils {
+    public static String rotateStringQueue(BlockingQueue<String> queue, String next) {
+        while (!queue.offer(next)) {
+            try {
+                queue.take();
+            } catch (InterruptedException e){
+                ;
+            }
+        }
+
+        StringBuilder logs = new StringBuilder();
+        for (String element: queue) {
+            logs.append(element);
+        }
+
+        return logs.toString();
+    }
+}
diff --git a/app/src/main/res/drawable-v24/ic_launcher_foreground.xml b/app/src/main/res/drawable-v24/ic_launcher_foreground.xml
new file mode 100644
index 0000000..c7bd21d
--- /dev/null
+++ b/app/src/main/res/drawable-v24/ic_launcher_foreground.xml
@@ -0,0 +1,34 @@
+<vector xmlns:android="http://schemas.android.com/apk/res/android"
+    xmlns:aapt="http://schemas.android.com/aapt"
+    android:width="108dp"
+    android:height="108dp"
+    android:viewportHeight="108"
+    android:viewportWidth="108">
+    <path
+        android:fillType="evenOdd"
+        android:pathData="M32,64C32,64 38.39,52.99 44.13,50.95C51.37,48.37 70.14,49.57 70.14,49.57L108.26,87.69L108,109.01L75.97,107.97L32,64Z"
+        android:strokeColor="#00000000"
+        android:strokeWidth="1">
+        <aapt:attr name="android:fillColor">
+            <gradient
+                android:endX="78.5885"
+                android:endY="90.9159"
+                android:startX="48.7653"
+                android:startY="61.0927"
+                android:type="linear">
+                <item
+                    android:color="#44000000"
+                    android:offset="0.0" />
+                <item
+                    android:color="#00000000"
+                    android:offset="1.0" />
+            </gradient>
+        </aapt:attr>
+    </path>
+    <path
+        android:fillColor="#FFFFFF"
+        android:fillType="nonZero"
+        android:pathData="M66.94,46.02L66.94,46.02C72.44,50.07 76,56.61 76,64L32,64C32,56.61 35.56,50.11 40.98,46.06L36.18,41.19C35.45,40.45 35.45,39.3 36.18,38.56C36.91,37.81 38.05,37.81 38.78,38.56L44.25,44.05C47.18,42.57 50.48,41.71 54,41.71C57.48,41.71 60.78,42.57 63.68,44.05L69.11,38.56C69.84,37.81 70.98,37.81 71.71,38.56C72.44,39.3 72.44,40.45 71.71,41.19L66.94,46.02ZM62.94,56.92C64.08,56.92 65,56.01 65,54.88C65,53.76 64.08,52.85 62.94,52.85C61.8,52.85 60.88,53.76 60.88,54.88C60.88,56.01 61.8,56.92 62.94,56.92ZM45.06,56.92C46.2,56.92 47.13,56.01 47.13,54.88C47.13,53.76 46.2,52.85 45.06,52.85C43.92,52.85 43,53.76 43,54.88C43,56.01 43.92,56.92 45.06,56.92Z"
+        android:strokeColor="#00000000"
+        android:strokeWidth="1" />
+</vector>
diff --git a/app/src/main/res/drawable/ic_launcher_background.xml b/app/src/main/res/drawable/ic_launcher_background.xml
new file mode 100644
index 0000000..d5fccc5
--- /dev/null
+++ b/app/src/main/res/drawable/ic_launcher_background.xml
@@ -0,0 +1,170 @@
+<?xml version="1.0" encoding="utf-8"?>
+<vector xmlns:android="http://schemas.android.com/apk/res/android"
+    android:width="108dp"
+    android:height="108dp"
+    android:viewportHeight="108"
+    android:viewportWidth="108">
+    <path
+        android:fillColor="#26A69A"
+        android:pathData="M0,0h108v108h-108z" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M9,0L9,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,0L19,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M29,0L29,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M39,0L39,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M49,0L49,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M59,0L59,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M69,0L69,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M79,0L79,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M89,0L89,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M99,0L99,108"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,9L108,9"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,19L108,19"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,29L108,29"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,39L108,39"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,49L108,49"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,59L108,59"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,69L108,69"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,79L108,79"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,89L108,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M0,99L108,99"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,29L89,29"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,39L89,39"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,49L89,49"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,59L89,59"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,69L89,69"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M19,79L89,79"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M29,19L29,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M39,19L39,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M49,19L49,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M59,19L59,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M69,19L69,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+    <path
+        android:fillColor="#00000000"
+        android:pathData="M79,19L79,89"
+        android:strokeColor="#33FFFFFF"
+        android:strokeWidth="0.8" />
+</vector>
diff --git a/app/src/main/res/layout/activity_main.xml b/app/src/main/res/layout/activity_main.xml
new file mode 100644
index 0000000..971ee29
--- /dev/null
+++ b/app/src/main/res/layout/activity_main.xml
@@ -0,0 +1,111 @@
+<?xml version="1.0" encoding="utf-8"?>
+<LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"
+    xmlns:app="http://schemas.android.com/apk/res-auto"
+    xmlns:tools="http://schemas.android.com/tools"
+    android:layout_width="match_parent"
+    android:layout_height="match_parent"
+    android:orientation="vertical"
+    tools:context="com.example.ottylab.bitzenyminer.MainActivity">
+
+    <EditText
+        android:id="@+id/editTextServer"
+        android:layout_width="match_parent"
+        android:layout_height="wrap_content"
+        android:layout_marginEnd="8dp"
+        android:layout_marginStart="8dp"
+        android:layout_marginTop="8dp"
+        android:ems="10"
+        android:hint="stratum+tcp://example.com:1234"
+        android:inputType="textPersonName"
+        app:layout_constraintEnd_toEndOf="parent"
+        app:layout_constraintStart_toStartOf="parent"
+        app:layout_constraintTop_toTopOf="parent" />
+
+    <EditText
+        android:id="@+id/editTextUser"
+        android:layout_width="match_parent"
+        android:layout_height="wrap_content"
+        android:layout_marginEnd="8dp"
+        android:layout_marginStart="8dp"
+        android:layout_marginTop="8dp"
+        android:ems="10"
+        android:hint="user.worker"
+        android:inputType="textPersonName"
+        app:layout_constraintEnd_toEndOf="parent"
+        app:layout_constraintStart_toStartOf="parent"
+        app:layout_constraintTop_toBottomOf="@+id/editTextServer" />
+
+    <EditText
+        android:id="@+id/editTextPassword"
+        android:layout_width="match_parent"
+        android:layout_height="wrap_content"
+        android:layout_marginEnd="8dp"
+        android:layout_marginStart="8dp"
+        android:layout_marginTop="8dp"
+        android:ems="10"
+        android:hint="password"
+        android:inputType="textPersonName"
+        app:layout_constraintEnd_toEndOf="parent"
+        app:layout_constraintStart_toStartOf="parent"
+        app:layout_constraintTop_toBottomOf="@+id/editTextUser" />
+
+    <EditText
+        android:id="@+id/editTextNThreads"
+        android:layout_width="match_parent"
+        android:layout_height="wrap_content"
+        android:layout_marginEnd="8dp"
+        android:layout_marginStart="8dp"
+        android:layout_marginTop="8dp"
+        android:ems="10"
+        android:hint="number of threads (auto: 0 or empty)"
+        android:inputType="number"
+        app:layout_constraintEnd_toEndOf="parent"
+        app:layout_constraintStart_toStartOf="parent"
+        app:layout_constraintTop_toBottomOf="@+id/editTextPassword" />
+
+    <LinearLayout
+        android:layout_width="match_parent"
+        android:layout_height="100dp"
+        android:orientation="horizontal">
+
+        <Button
+            android:id="@+id/buttonDrive"
+            android:layout_width="wrap_content"
+            android:layout_height="wrap_content"
+            android:layout_marginStart="16dp"
+            android:layout_marginTop="24dp"
+            android:text="Start"
+            app:layout_constraintStart_toStartOf="parent"
+            app:layout_constraintTop_toBottomOf="@+id/editTextPassword" />
+
+        <CheckBox
+            android:id="@+id/checkBoxBenchmark"
+            android:layout_width="106dp"
+            android:layout_height="44dp"
+            android:layout_marginStart="20dp"
+            android:layout_marginTop="24dp"
+            android:text="Benchmark"
+            app:layout_constraintLeft_toRightOf="@+id/editTextNThreads"
+            app:layout_constraintTop_toBottomOf="@+id/editTextPassword" />
+
+        <Spinner
+            android:id="@+id/spinnerAlgorithm"
+            android:layout_width="116dp"
+            android:layout_height="44dp"
+            android:layout_marginStart="20dp"
+            android:layout_marginTop="24dp" />
+    </LinearLayout>
+
+    <TextView
+        android:id="@+id/textViewLog"
+        android:layout_height="100dp"
+        android:layout_width="match_parent"
+        android:layout_weight="1"
+        android:layout_marginBottom="8dp"
+        android:layout_marginEnd="8dp"
+        android:layout_marginStart="8dp"
+        android:layout_marginTop="8dp"
+        android:ellipsize="marquee"
+        android:gravity="bottom"
+        android:scrollbars="vertical" />
+</LinearLayout>
diff --git a/app/src/main/res/mipmap-anydpi-v26/ic_launcher.xml b/app/src/main/res/mipmap-anydpi-v26/ic_launcher.xml
new file mode 100644
index 0000000..90f9580
--- /dev/null
+++ b/app/src/main/res/mipmap-anydpi-v26/ic_launcher.xml
@@ -0,0 +1,5 @@
+<?xml version="1.0" encoding="utf-8"?>
+<adaptive-icon xmlns:android="http://schemas.android.com/apk/res/android">
+  <background android:drawable="@mipmap/ic_launcher_adaptive_back"/>
+  <foreground android:drawable="@mipmap/ic_launcher_adaptive_fore"/>
+</adaptive-icon>
\ No newline at end of file
diff --git a/app/src/main/res/mipmap-hdpi/ic_launcher.png b/app/src/main/res/mipmap-hdpi/ic_launcher.png
new file mode 100644
index 0000000..df3494e
Binary files /dev/null and b/app/src/main/res/mipmap-hdpi/ic_launcher.png differ
diff --git a/app/src/main/res/mipmap-hdpi/ic_launcher_adaptive_back.png b/app/src/main/res/mipmap-hdpi/ic_launcher_adaptive_back.png
new file mode 100644
index 0000000..a7b3850
Binary files /dev/null and b/app/src/main/res/mipmap-hdpi/ic_launcher_adaptive_back.png differ
diff --git a/app/src/main/res/mipmap-hdpi/ic_launcher_adaptive_fore.png b/app/src/main/res/mipmap-hdpi/ic_launcher_adaptive_fore.png
new file mode 100644
index 0000000..7f68fa7
Binary files /dev/null and b/app/src/main/res/mipmap-hdpi/ic_launcher_adaptive_fore.png differ
diff --git a/app/src/main/res/mipmap-mdpi/ic_launcher.png b/app/src/main/res/mipmap-mdpi/ic_launcher.png
new file mode 100644
index 0000000..a3c6236
Binary files /dev/null and b/app/src/main/res/mipmap-mdpi/ic_launcher.png differ
diff --git a/app/src/main/res/mipmap-mdpi/ic_launcher_adaptive_back.png b/app/src/main/res/mipmap-mdpi/ic_launcher_adaptive_back.png
new file mode 100644
index 0000000..904f271
Binary files /dev/null and b/app/src/main/res/mipmap-mdpi/ic_launcher_adaptive_back.png differ
diff --git a/app/src/main/res/mipmap-mdpi/ic_launcher_adaptive_fore.png b/app/src/main/res/mipmap-mdpi/ic_launcher_adaptive_fore.png
new file mode 100644
index 0000000..c1a05fe
Binary files /dev/null and b/app/src/main/res/mipmap-mdpi/ic_launcher_adaptive_fore.png differ
diff --git a/app/src/main/res/mipmap-xhdpi/ic_launcher.png b/app/src/main/res/mipmap-xhdpi/ic_launcher.png
new file mode 100644
index 0000000..fbd8f2b
Binary files /dev/null and b/app/src/main/res/mipmap-xhdpi/ic_launcher.png differ
diff --git a/app/src/main/res/mipmap-xhdpi/ic_launcher_adaptive_back.png b/app/src/main/res/mipmap-xhdpi/ic_launcher_adaptive_back.png
new file mode 100644
index 0000000..95a6b1f
Binary files /dev/null and b/app/src/main/res/mipmap-xhdpi/ic_launcher_adaptive_back.png differ
diff --git a/app/src/main/res/mipmap-xhdpi/ic_launcher_adaptive_fore.png b/app/src/main/res/mipmap-xhdpi/ic_launcher_adaptive_fore.png
new file mode 100644
index 0000000..bf80a59
Binary files /dev/null and b/app/src/main/res/mipmap-xhdpi/ic_launcher_adaptive_fore.png differ
diff --git a/app/src/main/res/mipmap-xxhdpi/ic_launcher.png b/app/src/main/res/mipmap-xxhdpi/ic_launcher.png
new file mode 100644
index 0000000..96611f7
Binary files /dev/null and b/app/src/main/res/mipmap-xxhdpi/ic_launcher.png differ
diff --git a/app/src/main/res/mipmap-xxhdpi/ic_launcher_adaptive_back.png b/app/src/main/res/mipmap-xxhdpi/ic_launcher_adaptive_back.png
new file mode 100644
index 0000000..2b4543f
Binary files /dev/null and b/app/src/main/res/mipmap-xxhdpi/ic_launcher_adaptive_back.png differ
diff --git a/app/src/main/res/mipmap-xxhdpi/ic_launcher_adaptive_fore.png b/app/src/main/res/mipmap-xxhdpi/ic_launcher_adaptive_fore.png
new file mode 100644
index 0000000..b8b7b8c
Binary files /dev/null and b/app/src/main/res/mipmap-xxhdpi/ic_launcher_adaptive_fore.png differ
diff --git a/app/src/main/res/mipmap-xxxhdpi/ic_launcher.png b/app/src/main/res/mipmap-xxxhdpi/ic_launcher.png
new file mode 100644
index 0000000..c9d1827
Binary files /dev/null and b/app/src/main/res/mipmap-xxxhdpi/ic_launcher.png differ
diff --git a/app/src/main/res/mipmap-xxxhdpi/ic_launcher_adaptive_back.png b/app/src/main/res/mipmap-xxxhdpi/ic_launcher_adaptive_back.png
new file mode 100644
index 0000000..b28ec21
Binary files /dev/null and b/app/src/main/res/mipmap-xxxhdpi/ic_launcher_adaptive_back.png differ
diff --git a/app/src/main/res/mipmap-xxxhdpi/ic_launcher_adaptive_fore.png b/app/src/main/res/mipmap-xxxhdpi/ic_launcher_adaptive_fore.png
new file mode 100644
index 0000000..8cc386b
Binary files /dev/null and b/app/src/main/res/mipmap-xxxhdpi/ic_launcher_adaptive_fore.png differ
diff --git a/app/src/main/res/values/colors.xml b/app/src/main/res/values/colors.xml
new file mode 100644
index 0000000..3ab3e9c
--- /dev/null
+++ b/app/src/main/res/values/colors.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="utf-8"?>
+<resources>
+    <color name="colorPrimary">#3F51B5</color>
+    <color name="colorPrimaryDark">#303F9F</color>
+    <color name="colorAccent">#FF4081</color>
+</resources>
diff --git a/app/src/main/res/values/strings.xml b/app/src/main/res/values/strings.xml
new file mode 100644
index 0000000..2a5ceeb
--- /dev/null
+++ b/app/src/main/res/values/strings.xml
@@ -0,0 +1,7 @@
+<resources>
+    <string name="app_name">YentenMiner</string>
+    <string-array name="algorithms">
+        <item>yescrypt</item>
+        <item>yespower</item>
+    </string-array>
+</resources>
diff --git a/app/src/main/res/values/styles.xml b/app/src/main/res/values/styles.xml
new file mode 100644
index 0000000..5885930
--- /dev/null
+++ b/app/src/main/res/values/styles.xml
@@ -0,0 +1,11 @@
+<resources>
+
+    <!-- Base application theme. -->
+    <style name="AppTheme" parent="Theme.AppCompat.Light.DarkActionBar">
+        <!-- Customize your theme here. -->
+        <item name="colorPrimary">@color/colorPrimary</item>
+        <item name="colorPrimaryDark">@color/colorPrimaryDark</item>
+        <item name="colorAccent">@color/colorAccent</item>
+    </style>
+
+</resources>
diff --git a/app/src/test/java/com/example/ottylab/bitzenyminer/UtilsUnitTest.java b/app/src/test/java/com/example/ottylab/bitzenyminer/UtilsUnitTest.java
new file mode 100644
index 0000000..152756f
--- /dev/null
+++ b/app/src/test/java/com/example/ottylab/bitzenyminer/UtilsUnitTest.java
@@ -0,0 +1,34 @@
+package com.example.ottylab.bitzenyminer;
+
+
+import junit.framework.Assert;
+
+import org.junit.Test;
+
+import java.util.concurrent.BlockingQueue;
+import java.util.concurrent.LinkedBlockingQueue;
+
+import static org.junit.Assert.*;
+
+/**
+ * Example local unit test, which will execute on the development machine (host).
+ *
+ * @see <a href="http://d.android.com/tools/testing">Testing documentation</a>
+ */
+public class UtilsUnitTest {
+    @Test
+    public void testLogQueue() throws Exception {
+        final int CAPACITY = 10;
+
+        BlockingQueue<String> queue = new LinkedBlockingQueue<>(CAPACITY);
+        for (int i = 0; i < CAPACITY * 10; i++)  {
+            String output = Utils.rotateStringQueue(queue, String.valueOf(i));
+
+            int j = 0;
+            for (String element: queue) {
+                int expected = ((i < CAPACITY) ? 0 : i - CAPACITY + 1) + (j++);
+                Assert.assertEquals(String.valueOf(expected), element);
+            }
+        }
+    }
+}
\ No newline at end of file
diff --git a/bitzenymininglibrary/.gitignore b/bitzenymininglibrary/.gitignore
new file mode 100644
index 0000000..796b96d
--- /dev/null
+++ b/bitzenymininglibrary/.gitignore
@@ -0,0 +1 @@
+/build
diff --git a/bitzenymininglibrary/CMakeLists.txt b/bitzenymininglibrary/CMakeLists.txt
new file mode 100644
index 0000000..7dd6d05
--- /dev/null
+++ b/bitzenymininglibrary/CMakeLists.txt
@@ -0,0 +1,19 @@
+cmake_minimum_required(VERSION 3.4.1)
+
+# Include sub directory
+add_subdirectory(${CMAKE_SOURCE_DIR}/src/main/cpp/libcpuminer)
+
+add_library(
+    bitzenymininglibrary
+    SHARED
+    src/main/cpp/bitzenymininglibrary.cpp)
+
+find_library(
+    log-lib
+    log)
+
+target_link_libraries(
+    bitzenymininglibrary
+    ${log-lib}
+    cpuminer)
+
diff --git a/bitzenymininglibrary/build.gradle b/bitzenymininglibrary/build.gradle
new file mode 100644
index 0000000..c9e827f
--- /dev/null
+++ b/bitzenymininglibrary/build.gradle
@@ -0,0 +1,61 @@
+apply plugin: 'com.android.library'
+
+android {
+    compileSdkVersion 29
+    defaultConfig {
+        minSdkVersion 21
+        targetSdkVersion 29
+        versionCode 1
+        versionName "1.0"
+        testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
+        externalNativeBuild {
+            cmake {
+                cppFlags "-fexceptions"
+                arguments "-DTARGET=STANDARD"
+            }
+        }
+    }
+    buildTypes {
+        release {
+            minifyEnabled false
+            proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'
+        }
+    }
+    publishNonDefault true
+    flavorDimensions 'default'
+    productFlavors {
+        standard {
+            dimension 'default'
+            ndk {
+                abiFilters 'armeabi-v7a', 'arm64-v8a', 'x86', 'x86_64'
+            }
+        }
+        kitkat {
+            dimension 'default'
+            minSdkVersion 19
+            externalNativeBuild {
+                cmake {
+                    cFlags "-D__KITKAT__"
+                    arguments "-DTARGET=__KITKAT__"
+                }
+            }
+            ndk {
+                abiFilters 'armeabi-v7a', 'x86'
+            }
+        }
+    }
+    externalNativeBuild {
+        cmake {
+            path "CMakeLists.txt"
+        }
+    }
+
+}
+
+dependencies {
+    implementation fileTree(dir: 'libs', include: ['*.jar'])
+    implementation 'com.android.support:appcompat-v7:26.1.0'
+    testImplementation 'junit:junit:4.12'
+    androidTestImplementation 'com.android.support.test:runner:1.0.1'
+    androidTestImplementation 'com.android.support.test.espresso:espresso-core:3.0.1'
+}
diff --git a/bitzenymininglibrary/proguard-rules.pro b/bitzenymininglibrary/proguard-rules.pro
new file mode 100644
index 0000000..f1b4245
--- /dev/null
+++ b/bitzenymininglibrary/proguard-rules.pro
@@ -0,0 +1,21 @@
+# Add project specific ProGuard rules here.
+# You can control the set of applied configuration files using the
+# proguardFiles setting in build.gradle.
+#
+# For more details, see
+#   http://developer.android.com/guide/developing/tools/proguard.html
+
+# If your project uses WebView with JS, uncomment the following
+# and specify the fully qualified class name to the JavaScript interface
+# class:
+#-keepclassmembers class fqcn.of.javascript.interface.for.webview {
+#   public *;
+#}
+
+# Uncomment this to preserve the line number information for
+# debugging stack traces.
+#-keepattributes SourceFile,LineNumberTable
+
+# If you keep the line number information, uncomment this to
+# hide the original source file name.
+#-renamesourcefileattribute SourceFile
diff --git a/bitzenymininglibrary/src/androidTest/java/com/example/ottylab/bitzenymininglibrary/ExampleInstrumentedTest.java b/bitzenymininglibrary/src/androidTest/java/com/example/ottylab/bitzenymininglibrary/ExampleInstrumentedTest.java
new file mode 100644
index 0000000..39c5578
--- /dev/null
+++ b/bitzenymininglibrary/src/androidTest/java/com/example/ottylab/bitzenymininglibrary/ExampleInstrumentedTest.java
@@ -0,0 +1,26 @@
+package com.example.ottylab.bitzenymininglibrary;
+
+import android.content.Context;
+import android.support.test.InstrumentationRegistry;
+import android.support.test.runner.AndroidJUnit4;
+
+import org.junit.Test;
+import org.junit.runner.RunWith;
+
+import static org.junit.Assert.*;
+
+/**
+ * Instrumented test, which will execute on an Android device.
+ *
+ * @see <a href="http://d.android.com/tools/testing">Testing documentation</a>
+ */
+@RunWith(AndroidJUnit4.class)
+public class ExampleInstrumentedTest {
+    @Test
+    public void useAppContext() throws Exception {
+        // Context of the app under test.
+        Context appContext = InstrumentationRegistry.getTargetContext();
+
+        assertEquals("com.example.ottylab.bitzenymininglibrary.test", appContext.getPackageName());
+    }
+}
diff --git a/bitzenymininglibrary/src/main/AndroidManifest.xml b/bitzenymininglibrary/src/main/AndroidManifest.xml
new file mode 100644
index 0000000..9be64a2
--- /dev/null
+++ b/bitzenymininglibrary/src/main/AndroidManifest.xml
@@ -0,0 +1,2 @@
+<manifest xmlns:android="http://schemas.android.com/apk/res/android"
+    package="com.example.ottylab.bitzenymininglibrary" />
diff --git a/bitzenymininglibrary/src/main/cpp/bitzenymininglibrary.cpp b/bitzenymininglibrary/src/main/cpp/bitzenymininglibrary.cpp
new file mode 100644
index 0000000..8d864af
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/bitzenymininglibrary.cpp
@@ -0,0 +1,120 @@
+#include <jni.h>
+#include <string>
+#include <android/log.h>
+#include "libcpuminer/libcpuminer.h"
+
+#define Log(...) \
+    ((void)__android_log_print(ANDROID_LOG_INFO, "native: ", __VA_ARGS__))
+
+JavaVM *g_jvm;
+jclass g_class;
+
+extern "C"
+void vprintf_app(const char *format, va_list arg) {
+    char msg[1024]; //TODO: may cause buffer overflow
+    vsprintf(msg, format, arg);
+
+    JNIEnv *env;
+    int is_detach_requred = 0;
+
+    g_jvm->GetEnv(reinterpret_cast<void**>(&env), JNI_VERSION_1_6);
+    if (!env) {
+        g_jvm->AttachCurrentThread(&env, NULL);
+        is_detach_requred = 1;
+    }
+
+    jstring smsg = env->NewStringUTF(msg);
+    jmethodID mid = env->GetStaticMethodID(g_class, "output", "(Ljava/lang/String;)V");
+    env->CallStaticVoidMethod(g_class, mid, smsg);
+
+    if (is_detach_requred) {
+        g_jvm->DetachCurrentThread();
+    }
+}
+
+JNIEXPORT jint
+JNICALL
+JNI_OnLoad(JavaVM *vm, void *reserved) {
+    Log("JNI_OnLoad");
+    g_jvm = vm;
+
+    JNIEnv *env;
+    if (vm->GetEnv(reinterpret_cast<void**>(&env), JNI_VERSION_1_6) != JNI_OK) {
+        return -1;
+    }
+
+    jclass clazz = env->FindClass("com/example/ottylab/bitzenymininglibrary/BitZenyMiningLibrary");
+    g_class = (jclass) env->NewGlobalRef(clazz);
+
+    init(vprintf_app);
+
+    return JNI_VERSION_1_6;
+}
+
+extern "C"
+JNIEXPORT jint
+JNICALL
+Java_com_example_ottylab_bitzenymininglibrary_BitZenyMiningLibrary_initMining(
+        JNIEnv *env,
+        jobject /* this */) {
+    Log("initMining");
+    return init(vprintf_app);
+}
+
+extern "C"
+JNIEXPORT jboolean
+JNICALL
+Java_com_example_ottylab_bitzenymininglibrary_BitZenyMiningLibrary_isMiningRunning(
+        JNIEnv *env,
+        jobject /* this */) {
+    Log("isMiningRunning");
+    return (jboolean) is_running();
+}
+
+extern "C"
+JNIEXPORT jint
+JNICALL
+Java_com_example_ottylab_bitzenymininglibrary_BitZenyMiningLibrary_startMining(
+        JNIEnv *env,
+        jobject /* this */,
+        jstring url,
+        jstring user,
+        jstring password,
+        jint n_threads,
+        jint algo) {
+    Log("startMining");
+
+    const char *c_url = env->GetStringUTFChars(url, NULL);
+    const char *c_user = env->GetStringUTFChars(user, NULL);
+    const char *c_password = env->GetStringUTFChars(password, NULL);
+
+    int result = start(c_url, c_user, c_password, n_threads, algo);
+
+    env->ReleaseStringUTFChars(url, c_url);
+    env->ReleaseStringUTFChars(user, c_user);
+    env->ReleaseStringUTFChars(password, c_password);
+
+    return result;
+}
+extern "C"
+JNIEXPORT jint
+JNICALL
+Java_com_example_ottylab_bitzenymininglibrary_BitZenyMiningLibrary_startBenchmark(
+        JNIEnv *env,
+        jobject /* this */,
+        jint n_threads,
+        jint algo = 0) {
+    Log("startBenchmark");
+    return start(NULL, NULL, NULL, n_threads, algo);
+}
+
+
+extern "C"
+JNIEXPORT jint
+JNICALL
+Java_com_example_ottylab_bitzenymininglibrary_BitZenyMiningLibrary_stopMining(
+        JNIEnv *env,
+        jobject /* this */) {
+    Log("stopMining");
+    return stop();
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/.gitignore b/bitzenymininglibrary/src/main/cpp/libcpuminer/.gitignore
new file mode 100644
index 0000000..5d0d1ec
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/.gitignore
@@ -0,0 +1,29 @@
+
+minerd
+minerd.exe
+*.o
+
+autom4te.cache
+.deps
+
+Makefile
+Makefile.in
+INSTALL
+aclocal.m4
+configure
+configure.lineno
+depcomp
+missing
+install-sh
+stamp-h1
+compile
+config.log
+config.status
+config.status.lineno
+config.guess
+config.sub
+
+mingw32-config.cache
+
+deps/*
+!deps/build_*.sh
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/AUTHORS b/bitzenymininglibrary/src/main/cpp/libcpuminer/AUTHORS
new file mode 100644
index 0000000..984b32c
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/AUTHORS
@@ -0,0 +1,5 @@
+Jeff Garzik <jgarzik@pobox.com>
+
+ArtForz
+
+pooler <pooler@litecoinpool.org>
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/CMakeLists.txt b/bitzenymininglibrary/src/main/cpp/libcpuminer/CMakeLists.txt
new file mode 100644
index 0000000..221e3bf
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/CMakeLists.txt
@@ -0,0 +1,98 @@
+cmake_minimum_required(VERSION 3.4.1)
+
+enable_language(ASM)
+
+set(CMAKE_ASM_FLAGS "-g -O3")
+set(CMAKE_C_FLAGS " ${CMAKE_C_FLAGS} -g -O3")
+
+add_definitions(
+    -D__STDC_LIMIT_MACROS
+    -D__STDC_CONSTANT_MACROS
+    -D__linux
+)
+
+add_library(
+    cpuminer
+    STATIC
+
+    scrypt.c
+    scrypt-arm.S
+    scrypt-x64.S
+    scrypt-x86.S
+    sha2.c
+    sha2-arm.S
+    sha2-x64.S
+    sha2-x86-notextrel.S
+    util.c
+    yescrypt.c
+    yespower/yespower.c
+    yespower/sha256.c
+    libcpuminer.c
+)
+
+
+# Prebuilt libraries
+
+if (${TARGET} STREQUAL "__KITKAT__")
+    add_library(z STATIC IMPORTED)
+
+    set_target_properties(
+        z
+        PROPERTIES
+        IMPORTED_LOCATION ${CMAKE_SOURCE_DIR}/../distribution/zlib/lib19/${ANDROID_ABI}/libz.a
+    )
+
+    add_library(curl STATIC IMPORTED)
+
+    set_target_properties(
+        curl
+        PROPERTIES
+        IMPORTED_LOCATION ${CMAKE_SOURCE_DIR}/../distribution/curl/lib19/${ANDROID_ABI}/libcurl.a
+    )
+
+    add_library(jansson STATIC IMPORTED)
+
+    set_target_properties(
+        jansson
+        PROPERTIES
+        IMPORTED_LOCATION ${CMAKE_SOURCE_DIR}/../distribution/jansson/lib19/${ANDROID_ABI}/libjansson.a
+    )
+else()
+    add_library(z STATIC IMPORTED)
+
+    set_target_properties(
+        z
+        PROPERTIES
+        IMPORTED_LOCATION ${CMAKE_SOURCE_DIR}/../distribution/zlib/lib/${ANDROID_ABI}/libz.a
+    )
+
+    add_library(curl STATIC IMPORTED)
+
+    set_target_properties(
+        curl
+        PROPERTIES
+        IMPORTED_LOCATION ${CMAKE_SOURCE_DIR}/../distribution/curl/lib/${ANDROID_ABI}/libcurl.a
+    )
+
+    add_library(jansson STATIC IMPORTED)
+
+    set_target_properties(
+        jansson
+        PROPERTIES
+        IMPORTED_LOCATION ${CMAKE_SOURCE_DIR}/../distribution/jansson/lib/${ANDROID_ABI}/libjansson.a
+    )
+endif()
+
+target_include_directories(
+    cpuminer PRIVATE
+    ${CMAKE_SOURCE_DIR}/src/main/cpp/libcpuminer
+    ${CMAKE_SOURCE_DIR}/../distribution/zlib/include
+    ${CMAKE_SOURCE_DIR}/../distribution/curl/include
+    ${CMAKE_SOURCE_DIR}/../distribution/jansson/include)
+
+target_link_libraries(
+    cpuminer
+    curl
+    z
+    jansson
+)
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/COPYING b/bitzenymininglibrary/src/main/cpp/libcpuminer/COPYING
new file mode 100644
index 0000000..d60c31a
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/COPYING
@@ -0,0 +1,340 @@
+		    GNU GENERAL PUBLIC LICENSE
+		       Version 2, June 1991
+
+ Copyright (C) 1989, 1991 Free Software Foundation, Inc.
+     59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+			    Preamble
+
+  The licenses for most software are designed to take away your
+freedom to share and change it.  By contrast, the GNU General Public
+License is intended to guarantee your freedom to share and change free
+software--to make sure the software is free for all its users.  This
+General Public License applies to most of the Free Software
+Foundation's software and to any other program whose authors commit to
+using it.  (Some other Free Software Foundation software is covered by
+the GNU Library General Public License instead.)  You can apply it to
+your programs, too.
+
+  When we speak of free software, we are referring to freedom, not
+price.  Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+this service if you wish), that you receive source code or can get it
+if you want it, that you can change the software or use pieces of it
+in new free programs; and that you know you can do these things.
+
+  To protect your rights, we need to make restrictions that forbid
+anyone to deny you these rights or to ask you to surrender the rights.
+These restrictions translate to certain responsibilities for you if you
+distribute copies of the software, or if you modify it.
+
+  For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must give the recipients all the rights that
+you have.  You must make sure that they, too, receive or can get the
+source code.  And you must show them these terms so they know their
+rights.
+
+  We protect your rights with two steps: (1) copyright the software, and
+(2) offer you this license which gives you legal permission to copy,
+distribute and/or modify the software.
+
+  Also, for each author's protection and ours, we want to make certain
+that everyone understands that there is no warranty for this free
+software.  If the software is modified by someone else and passed on, we
+want its recipients to know that what they have is not the original, so
+that any problems introduced by others will not reflect on the original
+authors' reputations.
+
+  Finally, any free program is threatened constantly by software
+patents.  We wish to avoid the danger that redistributors of a free
+program will individually obtain patent licenses, in effect making the
+program proprietary.  To prevent this, we have made it clear that any
+patent must be licensed for everyone's free use or not licensed at all.
+
+  The precise terms and conditions for copying, distribution and
+modification follow.
+
+		    GNU GENERAL PUBLIC LICENSE
+   TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+
+  0. This License applies to any program or other work which contains
+a notice placed by the copyright holder saying it may be distributed
+under the terms of this General Public License.  The "Program", below,
+refers to any such program or work, and a "work based on the Program"
+means either the Program or any derivative work under copyright law:
+that is to say, a work containing the Program or a portion of it,
+either verbatim or with modifications and/or translated into another
+language.  (Hereinafter, translation is included without limitation in
+the term "modification".)  Each licensee is addressed as "you".
+
+Activities other than copying, distribution and modification are not
+covered by this License; they are outside its scope.  The act of
+running the Program is not restricted, and the output from the Program
+is covered only if its contents constitute a work based on the
+Program (independent of having been made by running the Program).
+Whether that is true depends on what the Program does.
+
+  1. You may copy and distribute verbatim copies of the Program's
+source code as you receive it, in any medium, provided that you
+conspicuously and appropriately publish on each copy an appropriate
+copyright notice and disclaimer of warranty; keep intact all the
+notices that refer to this License and to the absence of any warranty;
+and give any other recipients of the Program a copy of this License
+along with the Program.
+
+You may charge a fee for the physical act of transferring a copy, and
+you may at your option offer warranty protection in exchange for a fee.
+
+  2. You may modify your copy or copies of the Program or any portion
+of it, thus forming a work based on the Program, and copy and
+distribute such modifications or work under the terms of Section 1
+above, provided that you also meet all of these conditions:
+
+    a) You must cause the modified files to carry prominent notices
+    stating that you changed the files and the date of any change.
+
+    b) You must cause any work that you distribute or publish, that in
+    whole or in part contains or is derived from the Program or any
+    part thereof, to be licensed as a whole at no charge to all third
+    parties under the terms of this License.
+
+    c) If the modified program normally reads commands interactively
+    when run, you must cause it, when started running for such
+    interactive use in the most ordinary way, to print or display an
+    announcement including an appropriate copyright notice and a
+    notice that there is no warranty (or else, saying that you provide
+    a warranty) and that users may redistribute the program under
+    these conditions, and telling the user how to view a copy of this
+    License.  (Exception: if the Program itself is interactive but
+    does not normally print such an announcement, your work based on
+    the Program is not required to print an announcement.)
+
+These requirements apply to the modified work as a whole.  If
+identifiable sections of that work are not derived from the Program,
+and can be reasonably considered independent and separate works in
+themselves, then this License, and its terms, do not apply to those
+sections when you distribute them as separate works.  But when you
+distribute the same sections as part of a whole which is a work based
+on the Program, the distribution of the whole must be on the terms of
+this License, whose permissions for other licensees extend to the
+entire whole, and thus to each and every part regardless of who wrote it.
+
+Thus, it is not the intent of this section to claim rights or contest
+your rights to work written entirely by you; rather, the intent is to
+exercise the right to control the distribution of derivative or
+collective works based on the Program.
+
+In addition, mere aggregation of another work not based on the Program
+with the Program (or with a work based on the Program) on a volume of
+a storage or distribution medium does not bring the other work under
+the scope of this License.
+
+  3. You may copy and distribute the Program (or a work based on it,
+under Section 2) in object code or executable form under the terms of
+Sections 1 and 2 above provided that you also do one of the following:
+
+    a) Accompany it with the complete corresponding machine-readable
+    source code, which must be distributed under the terms of Sections
+    1 and 2 above on a medium customarily used for software interchange; or,
+
+    b) Accompany it with a written offer, valid for at least three
+    years, to give any third party, for a charge no more than your
+    cost of physically performing source distribution, a complete
+    machine-readable copy of the corresponding source code, to be
+    distributed under the terms of Sections 1 and 2 above on a medium
+    customarily used for software interchange; or,
+
+    c) Accompany it with the information you received as to the offer
+    to distribute corresponding source code.  (This alternative is
+    allowed only for noncommercial distribution and only if you
+    received the program in object code or executable form with such
+    an offer, in accord with Subsection b above.)
+
+The source code for a work means the preferred form of the work for
+making modifications to it.  For an executable work, complete source
+code means all the source code for all modules it contains, plus any
+associated interface definition files, plus the scripts used to
+control compilation and installation of the executable.  However, as a
+special exception, the source code distributed need not include
+anything that is normally distributed (in either source or binary
+form) with the major components (compiler, kernel, and so on) of the
+operating system on which the executable runs, unless that component
+itself accompanies the executable.
+
+If distribution of executable or object code is made by offering
+access to copy from a designated place, then offering equivalent
+access to copy the source code from the same place counts as
+distribution of the source code, even though third parties are not
+compelled to copy the source along with the object code.
+
+  4. You may not copy, modify, sublicense, or distribute the Program
+except as expressly provided under this License.  Any attempt
+otherwise to copy, modify, sublicense or distribute the Program is
+void, and will automatically terminate your rights under this License.
+However, parties who have received copies, or rights, from you under
+this License will not have their licenses terminated so long as such
+parties remain in full compliance.
+
+  5. You are not required to accept this License, since you have not
+signed it.  However, nothing else grants you permission to modify or
+distribute the Program or its derivative works.  These actions are
+prohibited by law if you do not accept this License.  Therefore, by
+modifying or distributing the Program (or any work based on the
+Program), you indicate your acceptance of this License to do so, and
+all its terms and conditions for copying, distributing or modifying
+the Program or works based on it.
+
+  6. Each time you redistribute the Program (or any work based on the
+Program), the recipient automatically receives a license from the
+original licensor to copy, distribute or modify the Program subject to
+these terms and conditions.  You may not impose any further
+restrictions on the recipients' exercise of the rights granted herein.
+You are not responsible for enforcing compliance by third parties to
+this License.
+
+  7. If, as a consequence of a court judgment or allegation of patent
+infringement or for any other reason (not limited to patent issues),
+conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License.  If you cannot
+distribute so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you
+may not distribute the Program at all.  For example, if a patent
+license would not permit royalty-free redistribution of the Program by
+all those who receive copies directly or indirectly through you, then
+the only way you could satisfy both it and this License would be to
+refrain entirely from distribution of the Program.
+
+If any portion of this section is held invalid or unenforceable under
+any particular circumstance, the balance of the section is intended to
+apply and the section as a whole is intended to apply in other
+circumstances.
+
+It is not the purpose of this section to induce you to infringe any
+patents or other property right claims or to contest validity of any
+such claims; this section has the sole purpose of protecting the
+integrity of the free software distribution system, which is
+implemented by public license practices.  Many people have made
+generous contributions to the wide range of software distributed
+through that system in reliance on consistent application of that
+system; it is up to the author/donor to decide if he or she is willing
+to distribute software through any other system and a licensee cannot
+impose that choice.
+
+This section is intended to make thoroughly clear what is believed to
+be a consequence of the rest of this License.
+
+  8. If the distribution and/or use of the Program is restricted in
+certain countries either by patents or by copyrighted interfaces, the
+original copyright holder who places the Program under this License
+may add an explicit geographical distribution limitation excluding
+those countries, so that distribution is permitted only in or among
+countries not thus excluded.  In such case, this License incorporates
+the limitation as if written in the body of this License.
+
+  9. The Free Software Foundation may publish revised and/or new versions
+of the General Public License from time to time.  Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+Each version is given a distinguishing version number.  If the Program
+specifies a version number of this License which applies to it and "any
+later version", you have the option of following the terms and conditions
+either of that version or of any later version published by the Free
+Software Foundation.  If the Program does not specify a version number of
+this License, you may choose any version ever published by the Free Software
+Foundation.
+
+  10. If you wish to incorporate parts of the Program into other free
+programs whose distribution conditions are different, write to the author
+to ask for permission.  For software which is copyrighted by the Free
+Software Foundation, write to the Free Software Foundation; we sometimes
+make exceptions for this.  Our decision will be guided by the two goals
+of preserving the free status of all derivatives of our free software and
+of promoting the sharing and reuse of software generally.
+
+			    NO WARRANTY
+
+  11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
+FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW.  EXCEPT WHEN
+OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
+PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
+OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.  THE ENTIRE RISK AS
+TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU.  SHOULD THE
+PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
+REPAIR OR CORRECTION.
+
+  12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
+REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
+INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
+OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
+TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
+YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
+PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGES.
+
+		     END OF TERMS AND CONDITIONS
+
+	    How to Apply These Terms to Your New Programs
+
+  If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+  To do so, attach the following notices to the program.  It is safest
+to attach them to the start of each source file to most effectively
+convey the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+    <one line to give the program's name and a brief idea of what it does.>
+    Copyright (C) <year>  <name of author>
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+
+
+Also add information on how to contact you by electronic and paper mail.
+
+If the program is interactive, make it output a short notice like this
+when it starts in an interactive mode:
+
+    Gnomovision version 69, Copyright (C) year  name of author
+    Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+    This is free software, and you are welcome to redistribute it
+    under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License.  Of course, the commands you use may
+be called something other than `show w' and `show c'; they could even be
+mouse-clicks or menu items--whatever suits your program.
+
+You should also get your employer (if you work as a programmer) or your
+school, if any, to sign a "copyright disclaimer" for the program, if
+necessary.  Here is a sample; alter the names:
+
+  Yoyodyne, Inc., hereby disclaims all copyright interest in the program
+  `Gnomovision' (which makes passes at compilers) written by James Hacker.
+
+  <signature of Ty Coon>, 1 April 1989
+  Ty Coon, President of Vice
+
+This General Public License does not permit incorporating your program into
+proprietary programs.  If your program is a subroutine library, you may
+consider it more useful to permit linking proprietary applications with the
+library.  If this is what you want to do, use the GNU Library General
+Public License instead of this License.
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/ChangeLog b/bitzenymininglibrary/src/main/cpp/libcpuminer/ChangeLog
new file mode 100644
index 0000000..326703b
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/ChangeLog
@@ -0,0 +1 @@
+See git repository ('git log') for full changelog.
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/Dockerfile b/bitzenymininglibrary/src/main/cpp/libcpuminer/Dockerfile
new file mode 100644
index 0000000..bce3c00
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/Dockerfile
@@ -0,0 +1,23 @@
+#
+# Dockerfile for cpuminer
+# usage: docker run creack/cpuminer --url xxxx --user xxxx --pass xxxx
+# ex: docker run creack/cpuminer --url stratum+tcp://ltc.pool.com:80 --user creack.worker1 --pass abcdef
+#
+#
+FROM ubuntu:16.04
+
+RUN apt-get update -qq
+RUN apt-get install -qqy automake
+RUN apt-get install -qqy libcurl4-openssl-dev
+RUN apt-get install -qqy git
+RUN apt-get install -qqy make
+RUN apt-get install -qqy build-essential
+
+RUN git clone https://github.com/BitzenyCoreDevelopers/cpuminer.git
+
+RUN cd cpuminer && ./autogen.sh
+RUN cd cpuminer && ./configure CFLAGS="-O3"
+RUN cd cpuminer && make
+
+WORKDIR /cpuminer
+ENTRYPOINT ["./minerd"]
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/LICENSE b/bitzenymininglibrary/src/main/cpp/libcpuminer/LICENSE
new file mode 100644
index 0000000..c43e8b0
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/LICENSE
@@ -0,0 +1,3 @@
+cpuminer is available under the terms of the GNU Public License version 2.
+
+See COPYING for details.
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/Makefile.am b/bitzenymininglibrary/src/main/cpp/libcpuminer/Makefile.am
new file mode 100644
index 0000000..2c4383a
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/Makefile.am
@@ -0,0 +1,38 @@
+
+if WANT_JANSSON
+JANSSON_INCLUDES= -I$(top_srcdir)/compat/jansson
+else
+JANSSON_INCLUDES=
+endif
+
+EXTRA_DIST	= example-cfg.json nomacro.pl
+
+SUBDIRS		= compat
+
+bin_PROGRAMS	= minerd
+
+dist_man_MANS	= minerd.1
+
+minerd_SOURCES	= elist.h miner.h compat.h \
+		  cpu-miner.c util.c \
+		  sha2.c scrypt.c yescrypt.c \
+      yespower/yespower.c yespower/sha256.c
+
+if USE_ASM
+if ARCH_x86
+minerd_SOURCES += sha2-x86.S scrypt-x86.S
+endif
+if ARCH_x86_64
+minerd_SOURCES += sha2-x64.S scrypt-x64.S
+endif
+if ARCH_ARM
+minerd_SOURCES += sha2-arm.S scrypt-arm.S
+endif
+if ARCH_PPC
+minerd_SOURCES += sha2-ppc.S scrypt-ppc.S
+endif
+endif
+minerd_LDFLAGS	= $(PTHREAD_FLAGS)
+minerd_LDADD	= @LIBCURL@ @JANSSON_LIBS@ @PTHREAD_LIBS@ @WS2_LIBS@
+minerd_CFLAGS	= -fno-strict-aliasing
+minerd_CPPFLAGS	= @LIBCURL_CPPFLAGS@ $(JANSSON_INCLUDES) $(PTHREAD_FLAGS)
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/NEWS b/bitzenymininglibrary/src/main/cpp/libcpuminer/NEWS
new file mode 100644
index 0000000..dd4e59e
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/NEWS
@@ -0,0 +1,303 @@
+Version 2.5.0 - Jun 22, 2017
+
+- Add Segwit support
+- Add support for 64-bit PowerPC
+
+Version 2.4.5 - Jun 10, 2016
+
+- Fix a memory leak affecting long polling
+
+Version 2.4.4 - Mar 24, 2016
+
+- Fix memory leaks affecting getblocktemplate
+
+Version 2.4.3 - Jan 24, 2016
+
+- Add support for the VIA PadLock Hash Engine on x86-64
+- Allow block version 4 when using getblocktemplate
+
+Version 2.4.2 - Apr 28, 2015
+
+- Add support for Stratum over TLS
+- Allow block version 3 when using getblocktemplate
+
+Version 2.4.1 - Feb 25, 2015
+
+- Add support for scrypt(N, 1, 1)
+- Add optimized PowerPC code for scrypt and SHA-256d
+- Allow use of getblocktemplate with merge-mined cryptocurrencies
+- Automatically switch to getwork if no payout address is provided
+- Fix CVE-2014-6251
+
+Version 2.4 - May 20, 2014
+
+- Add support for the getblocktemplate RPC method (BIP 22)
+- Allow tunnelling Stratum through HTTP proxies
+- Add a --no-redirect option to ignore redirection requests
+- Timeout for long polling is now disabled by default
+- Fix CPU affinity on Linux (kiyominer)
+- Add support for building under 64-bit Cygwin
+- Expand version information with build details
+
+Version 2.3.3 - Feb 27, 2014
+
+- The --url option is now mandatory
+- Do not switch to Stratum when using an HTTP proxy
+- Fix scheduling policy change on Linux (clbr)
+- Fix CPU affinity on FreeBSD (ache)
+- Compatibility fixes for various platforms, including Solaris 8
+  and old versions of OS X
+- A man page for minerd is now available
+
+Version 2.3.2 - Jul 10, 2013
+
+- Add optimizations for AVX2-capable x86-64 processors
+- Ensure that the output stream is flushed after every log message
+- Fix an undefined-behavior bug in the Stratum code
+
+Version 2.3.1 - Jun 18, 2013
+
+- Add a --cert option for specifying an SSL certificate (martinwguy)
+- Fix a bug that only made SHA-256d mining work at difficulty 1
+- Fix a couple of compatibility issues with some Stratum servers
+
+Version 2.3 - Jun 12, 2013
+
+- Add support for the Stratum mining protocol
+- Automatically switch to Stratum if the mining server supports
+  the X-Stratum extension, unless --no-stratum is used
+- Set CPU affinity on FreeBSD (lye)
+- Fix a bug in libcurl initialization (martinwguy)
+
+Version 2.2.3 - Aug 5, 2012
+
+- Add optimized ARM NEON code for scrypt and SHA-256d
+- Add a --benchmark option that allows offline testing
+- Support for the X-Reject-Reason extension
+
+Version 2.2.2 - Jun 7, 2012
+
+- Various performance improvements for x86 and x86-64
+- Optimize scrypt for ARMv5E and later processors
+- Set the priority of miner threads to idle on Windows
+- Add an option to start minerd as a daemon on POSIX systems
+
+Version 2.2.1 - May 2, 2012
+
+- Add optimized code for ARM processors
+- Support for building on NetBSD and OpenBSD
+- Various compatibility fixes for AIX (pontius)
+
+Version 2.2 - Apr 2, 2012
+
+- Add an optimized SHA-256d algorithm, with specialized code
+  for x86 and x86-64 and support for AVX and XOP instructions
+- Slight performance increase for scrypt on x86 and x86-64
+- The default timeout is now 270 seconds
+
+Version 2.1.5 - Mar 7, 2012
+
+- Add optimizations for AVX-capable x86-64 processors
+- Assume HTTP if no protocol is specified for the mining server
+- Fix MinGW compatibility issues and update build instructions
+- Add support for building on Solaris using gcc (pontius)
+
+Version 2.1.4 - Feb 28, 2012
+
+- Implement 4-way SHA-256 on x86-64
+- Add TCP keepalive to long polling connections
+- Support HTTP and SOCKS proxies via the --proxy option
+- Username and password are no longer mandatory
+- Add a script that makes assembly code compatible with old versions
+  of the GNU assembler that do not support macros
+
+Version 2.1.3 - Feb 12, 2012
+
+- Smart handling of long polling failures: switch to short scan time
+  if long polling fails, and only try to reactivate it if the server
+  continues to advertise the feature in HTTP headers
+- Add "X-Mining-Extensions: midstate" to HTTP headers (p2k)
+- Add support for the "submitold" extension, used by p2pool
+- It is now possible to specify username and password in the URL,
+  like this: http://username:password@host:port/
+- Add a --version option, and clean up --help output
+- Avoid division by zero when computing hash rates
+- Handle empty responses properly (TimothyA)
+- Eliminate the delay between starting threads
+
+Version 2.1.2 - Jan 26, 2012
+
+- Do not submit work that is known to be stale
+- Allow miner threads to ask for new work if the current one is at least
+  45 seconds old and long polling is enabled
+- Refresh work when long polling times out
+- Fix minor speed regression
+- Modify x86-64 code to make it compatible with older versions of binutils
+
+Version 2.1.1 - Jan 20, 2012
+
+- Handle network errors properly
+- Make scantime retargeting more accurate
+
+Version 2.1 - Jan 19, 2012
+
+- Share the same work among all threads
+- Do not ask for new work if the current one is not expired
+- Do not discard the work returned by long polling
+
+Version 2.0 - Jan 16, 2012
+
+- Change default port to 9332 for Litecoin and remove default credentials
+- Add 'scrypt' as the default algorithm and remove other algorithms (ArtForz)
+- Optimize scrypt for x86 and x86-64
+- Make scantime retargeting less granular (ArtForz)
+- Test the whole hash instead of just looking at the high 32 bits
+- Add configurable timeout, with a default of 180 seconds
+- Add share summary output (inlikeflynn)
+- Fix priority and CPU count detection on Windows
+- Fix parameters -u and -p, and add short options -o and -O
+
+Version 1.0.2 - Jun 13, 2011
+
+- Linux x86_64 optimisations - Con Kolivas
+- Optimise for x86_64 by default by using sse2_64 algo
+- Detects CPUs and sets number of threads accordingly
+- Uses CPU affinity for each thread where appropriate
+- Sets scheduling policy to lowest possible
+- Minor performance tweaks
+
+Version 1.0.1 - May 14, 2011
+
+- OSX support
+
+Version 1.0 - May 9, 2011
+
+- jansson 2.0 compatibility
+- correct off-by-one in date (month) display output
+- fix platform detection
+- improve yasm configure bits
+- support full URL, in X-Long-Polling header
+
+Version 0.8.1 - March 22, 2011
+
+- Make --user, --pass actually work
+
+- Add User-Agent HTTP header to requests, so that server operators may
+  more easily identify the miner client.
+
+- Fix minor bug in example JSON config file
+
+Version 0.8 - March 21, 2011
+
+- Support long polling: http://deepbit.net/longpolling.php
+
+- Adjust max workload based on scantime (default 5 seconds,
+  or 60 seconds for longpoll)
+
+- Standardize program output, and support syslog on Unix platforms
+
+- Suport --user/--pass options (and "user" and "pass" in config file),
+  as an alternative to the current --userpass
+
+Version 0.7.2 - March 14, 2011
+
+- Add port of ufasoft's sse2 assembly implementation (Linux only)
+  This is a substantial speed improvement on Intel CPUs.
+
+- Move all JSON-RPC I/O to separate thread.  This reduces the
+  number of HTTP connections from one-per-thread to one, reducing resource
+  usage on upstream bitcoind / pool server.
+
+Version 0.7.1 - March 2, 2011
+
+- Add support for JSON-format configuration file.  See example
+  file example-cfg.json.  Any long argument on the command line
+  may be stored in the config file.
+- Timestamp each solution found
+- Improve sha256_4way performance.  NOTE: This optimization makes
+  the 'hash' debug-print output for sha256_way incorrect.
+- Use __builtin_expect() intrinsic as compiler micro-optimization
+- Build on Intel compiler
+- HTTP library now follows HTTP redirects
+
+Version 0.7 - February 12, 2011
+
+- Re-use CURL object, thereby reuseing DNS cache and HTTP connections
+- Use bswap_32, if compiler intrinsic is not available
+- Disable full target validation (as opposed to simply H==0) for now
+
+Version 0.6.1 - February 4, 2011
+
+- Fully validate "hash < target", rather than simply stopping our scan
+  if the high 32 bits are 00000000.
+- Add --retry-pause, to set length of pause time between failure retries
+- Display proof-of-work hash and target, if -D (debug mode) enabled
+- Fix max-nonce auto-adjustment to actually work.  This means if your
+  scan takes longer than 5 seconds (--scantime), the miner will slowly
+  reduce the number of hashes you work on, before fetching a new work unit.
+
+Version 0.6 - January 29, 2011
+
+- Fetch new work unit, if scanhash takes longer than 5 seconds (--scantime)
+- BeeCee1's sha256 4way optimizations
+- lfm's byte swap optimization (improves via, cryptopp)
+- Fix non-working short options -q, -r
+
+Version 0.5 - December 28, 2010
+
+- Exit program, when all threads have exited
+- Improve JSON-RPC failure diagnostics and resilience
+- Add --quiet option, to disable hashmeter output.
+
+Version 0.3.3 - December 27, 2010
+
+- Critical fix for sha256_cryptopp 'cryptopp_asm' algo
+
+Version 0.3.2 - December 23, 2010
+
+- Critical fix for sha256_via
+
+Version 0.3.1 - December 19, 2010
+
+- Critical fix for sha256_via
+- Retry JSON-RPC failures (see --retry, under "minerd --help" output)
+
+Version 0.3 - December 18, 2010
+
+- Add crypto++ 32bit assembly implementation
+- show version upon 'minerd --help'
+- work around gcc 4.5.x bug that killed 4way performance
+
+Version 0.2.2 - December 6, 2010
+
+- VIA padlock implementation works now
+- Minor build and runtime fixes
+
+Version 0.2.1 - November 29, 2010
+
+- avoid buffer overflow when submitting solutions
+- add Crypto++ sha256 implementation (C only, ASM elided for now)
+- minor internal optimizations and cleanups
+
+Version 0.2 - November 27, 2010
+
+- Add script for building a Windows installer
+- improve hash performance (hashmeter) statistics
+- add tcatm 4way sha256 implementation
+- Add experimental VIA Padlock sha256 implementation
+
+Version 0.1.2 - November 26, 2010
+
+- many small cleanups and micro-optimizations
+- build win32 exe using mingw
+- RPC URL, username/password become command line arguments
+- remove unused OpenSSL dependency
+
+Version 0.1.1 - November 24, 2010
+
+- Do not build sha256_generic module separately from cpuminer.
+
+Version 0.1 - November 24, 2010
+
+- Initial release.
+
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/README b/bitzenymininglibrary/src/main/cpp/libcpuminer/README
new file mode 100644
index 0000000..b51af35
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/README
@@ -0,0 +1,87 @@
+This is a multi-threaded CPU miner for Litecoin and Bitcoin,
+fork of Jeff Garzik's reference cpuminer.
+
+License: GPLv2.  See COPYING for details.
+
+Downloads:  https://sourceforge.net/projects/cpuminer/files/
+Git tree:   https://github.com/pooler/cpuminer
+
+Dependencies:
+	libcurl			http://curl.haxx.se/libcurl/
+	jansson			http://www.digip.org/jansson/
+		(jansson is included in-tree)
+
+Basic *nix build instructions:
+	./autogen.sh	# only needed if building from git repo
+	./nomacro.pl	# only needed if building on Mac OS X or with Clang
+	./configure CFLAGS="-O3 -march=native -funroll-loops -fomit-frame-pointer" 
+	make
+
+Notes for AIX users:
+	* To build a 64-bit binary, export OBJECT_MODE=64
+	* GNU-style long options are not supported, but are accessible
+	  via configuration file
+
+Cross Compile for Windows 64bit, using Ubuntu
+
+	sudo apt-get install build-essential automake mingw-w64-x86-64-dev gcc-mingw-w64-x86-64
+	git clone https://github.com/BitzenyCoreDevelopers/cpuminer.git
+	cd cpuminer/deps
+	./build_win_deps.sh
+	cd ..
+	autoreconf -fi -I./deps/x86_64-w64-mingw32/share/aclocal
+	./configure --host=x86_64-w64-mingw32 \
+	CFLAGS="-O3 -I./deps/x86_64-w64-mingw32/include -std=c99 -DWIN32 -DCURL_STATICLIB -DPTW32_STATIC_LIB" \
+	--with-libcurl=deps/x86_64-w64-mingw32 LDFLAGS="-L./deps/x86_64-w64-mingw32/lib -static"
+	make
+
+TRY: -Ofast -flto -march=native
+
+
+Basic Windows build instructions, using MinGW:
+	Install MinGW and the MSYS Developer Tool Kit (http://www.mingw.org/)
+		* Make sure you have mstcpip.h in MinGW\include
+	If using MinGW-w64, install pthreads-w64
+	Install libcurl devel (http://curl.haxx.se/download.html)
+		* Make sure you have libcurl.m4 in MinGW\share\aclocal
+		* Make sure you have curl-config in MinGW\bin
+	In the MSYS shell, run:
+		./autogen.sh	# only needed if building from git repo
+		LIBCURL="-lcurldll" ./configure CFLAGS="-O3"
+		make
+
+Architecture-specific notes:
+	ARM:	No runtime CPU detection. The miner can take advantage
+		of some instructions specific to ARMv5E and later processors,
+		but the decision whether to use them is made at compile time,
+		based on compiler-defined macros.
+		To use NEON instructions, add "-mfpu=neon" to CFLAGS.
+	PowerPC: No runtime CPU detection.
+		To use AltiVec instructions, add "-maltivec" to CFLAGS.
+	x86:	The miner checks for SSE2 instructions support at runtime,
+		and uses them if they are available.
+	x86-64:	The miner can take advantage of AVX, AVX2 and XOP instructions,
+		but only if both the CPU and the operating system support them.
+		    * Linux supports AVX starting from kernel version 2.6.30.
+		    * FreeBSD supports AVX starting with 9.1-RELEASE.
+		    * Mac OS X added AVX support in the 10.6.8 update.
+		    * Windows supports AVX starting from Windows 7 SP1 and
+		      Windows Server 2008 R2 SP1.
+		The configure script outputs a warning if the assembler
+		doesn't support some instruction sets. In that case, the miner
+		can still be built, but unavailable optimizations are left off.
+		The miner uses the VIA Padlock Hash Engine where available.
+
+Usage instructions:  Run "minerd --help" to see options.
+
+Connecting through a proxy:  Use the --proxy option.
+To use a SOCKS proxy, add a socks4:// or socks5:// prefix to the proxy host.
+Protocols socks4a and socks5h, allowing remote name resolving, are also
+available since libcurl 7.18.0.
+If no protocol is specified, the proxy is assumed to be a HTTP proxy.
+When the --proxy option is not used, the program honors the http_proxy
+and all_proxy environment variables.
+
+Also many issues and FAQs are covered in the forum thread
+dedicated to this program,
+	https://bitcointalk.org/index.php?topic=55038.0
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/autogen.sh b/bitzenymininglibrary/src/main/cpp/libcpuminer/autogen.sh
new file mode 100644
index 0000000..989604a
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/autogen.sh
@@ -0,0 +1,12 @@
+#!/bin/sh
+
+# You need autoconf 2.5x, preferably 2.57 or later
+# You need automake 1.7 or later. 1.6 might work.
+
+set -e
+
+aclocal
+autoheader
+automake --gnu --add-missing --copy
+autoconf
+
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat.h
new file mode 100644
index 0000000..283fc9b
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat.h
@@ -0,0 +1,21 @@
+#ifndef __COMPAT_H__
+#define __COMPAT_H__
+
+#ifdef WIN32
+
+#include <windows.h>
+
+#define sleep(secs) Sleep((secs) * 1000)
+
+enum {
+	PRIO_PROCESS		= 0,
+};
+
+static inline int setpriority(int which, int who, int prio)
+{
+	return -!SetThreadPriority(GetCurrentThread(), THREAD_PRIORITY_IDLE);
+}
+
+#endif /* WIN32 */
+
+#endif /* __COMPAT_H__ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/Makefile.am b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/Makefile.am
new file mode 100644
index 0000000..77af3c5
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/Makefile.am
@@ -0,0 +1,7 @@
+
+if WANT_JANSSON
+SUBDIRS	= jansson
+else
+SUBDIRS	=
+endif
+
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/.gitignore b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/.gitignore
new file mode 100644
index 0000000..173737b
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/.gitignore
@@ -0,0 +1,3 @@
+
+libjansson.a
+
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/LICENSE b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/LICENSE
new file mode 100644
index 0000000..552b349
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/LICENSE
@@ -0,0 +1,19 @@
+Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/Makefile.am b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/Makefile.am
new file mode 100644
index 0000000..94a583f
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/Makefile.am
@@ -0,0 +1,18 @@
+
+noinst_LIBRARIES	= libjansson.a
+
+libjansson_a_SOURCES	= \
+			  config.h		\
+			  dump.c		\
+			  hashtable.c		\
+			  hashtable.h		\
+			  jansson.h		\
+			  jansson_private.h	\
+			  load.c		\
+			  strbuffer.c		\
+			  strbuffer.h		\
+			  utf.c			\
+			  utf.h			\
+			  util.h		\
+			  value.c
+
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/config.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/config.h
new file mode 100644
index 0000000..43858aa
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/config.h
@@ -0,0 +1,73 @@
+/* config.h.  Generated from config.h.in by configure.  */
+/* config.h.in.  Generated from configure.ac by autoheader.  */
+
+/* Define to 1 if you have the <dlfcn.h> header file. */
+#define HAVE_DLFCN_H 1
+
+/* Define to 1 if you have the <inttypes.h> header file. */
+#define HAVE_INTTYPES_H 1
+
+/* Define to 1 if you have the <memory.h> header file. */
+#define HAVE_MEMORY_H 1
+
+/* Define to 1 if you have the <stdint.h> header file. */
+#define HAVE_STDINT_H 1
+
+/* Define to 1 if you have the <stdlib.h> header file. */
+#define HAVE_STDLIB_H 1
+
+/* Define to 1 if you have the <strings.h> header file. */
+#define HAVE_STRINGS_H 1
+
+/* Define to 1 if you have the <string.h> header file. */
+#define HAVE_STRING_H 1
+
+/* Define to 1 if you have the <sys/stat.h> header file. */
+#define HAVE_SYS_STAT_H 1
+
+/* Define to 1 if you have the <sys/types.h> header file. */
+#define HAVE_SYS_TYPES_H 1
+
+/* Define to 1 if you have the <unistd.h> header file. */
+#define HAVE_UNISTD_H 1
+
+/* Define to the sub-directory in which libtool stores uninstalled libraries.
+   */
+#define LT_OBJDIR ".libs/"
+
+/* Name of package */
+#define PACKAGE "jansson"
+
+/* Define to the address where bug reports for this package should be sent. */
+#define PACKAGE_BUGREPORT "petri@digip.org"
+
+/* Define to the full name of this package. */
+#define PACKAGE_NAME "jansson"
+
+/* Define to the full name and version of this package. */
+#define PACKAGE_STRING "jansson 1.3"
+
+/* Define to the one symbol short name of this package. */
+#define PACKAGE_TARNAME "jansson"
+
+/* Define to the home page for this package. */
+#define PACKAGE_URL ""
+
+/* Define to the version of this package. */
+#define PACKAGE_VERSION "1.3"
+
+/* Define to 1 if you have the ANSI C header files. */
+#define STDC_HEADERS 1
+
+/* Version number of package */
+#define VERSION "1.3"
+
+/* Define to `__inline__' or `__inline' if that's what the C compiler
+   calls it, or to nothing if 'inline' is not supported under any name.  */
+#ifndef __cplusplus
+/* #undef inline */
+#endif
+
+/* Define to the type of a signed integer type of width exactly 32 bits if
+   such a type exists and the standard includes do not define it. */
+/* #undef int32_t */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/dump.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/dump.c
new file mode 100644
index 0000000..dc27fbd
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/dump.c
@@ -0,0 +1,460 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#define _GNU_SOURCE
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <assert.h>
+
+#include <jansson.h>
+#include "jansson_private.h"
+#include "strbuffer.h"
+#include "utf.h"
+
+#define MAX_INTEGER_STR_LENGTH  100
+#define MAX_REAL_STR_LENGTH     100
+
+typedef int (*dump_func)(const char *buffer, int size, void *data);
+
+struct string
+{
+    char *buffer;
+    int length;
+    int size;
+};
+
+static int dump_to_strbuffer(const char *buffer, int size, void *data)
+{
+    return strbuffer_append_bytes((strbuffer_t *)data, buffer, size);
+}
+
+static int dump_to_file(const char *buffer, int size, void *data)
+{
+    FILE *dest = (FILE *)data;
+    if(fwrite(buffer, size, 1, dest) != 1)
+        return -1;
+    return 0;
+}
+
+/* 256 spaces (the maximum indentation size) */
+static char whitespace[] = "                                                                                                                                                                                                                                                                ";
+
+static int dump_indent(unsigned long flags, int depth, int space, dump_func dump, void *data)
+{
+    if(JSON_INDENT(flags) > 0)
+    {
+        int i, ws_count = JSON_INDENT(flags);
+
+        if(dump("\n", 1, data))
+            return -1;
+
+        for(i = 0; i < depth; i++)
+        {
+            if(dump(whitespace, ws_count, data))
+                return -1;
+        }
+    }
+    else if(space && !(flags & JSON_COMPACT))
+    {
+        return dump(" ", 1, data);
+    }
+    return 0;
+}
+
+static int dump_string(const char *str, int ascii, dump_func dump, void *data)
+{
+    const char *pos, *end;
+    int32_t codepoint;
+
+    if(dump("\"", 1, data))
+        return -1;
+
+    end = pos = str;
+    while(1)
+    {
+        const char *text;
+        char seq[13];
+        int length;
+
+        while(*end)
+        {
+            end = utf8_iterate(pos, &codepoint);
+            if(!end)
+                return -1;
+
+            /* mandatory escape or control char */
+            if(codepoint == '\\' || codepoint == '"' || codepoint < 0x20)
+                break;
+
+            /* non-ASCII */
+            if(ascii && codepoint > 0x7F)
+                break;
+
+            pos = end;
+        }
+
+        if(pos != str) {
+            if(dump(str, pos - str, data))
+                return -1;
+        }
+
+        if(end == pos)
+            break;
+
+        /* handle \, ", and control codes */
+        length = 2;
+        switch(codepoint)
+        {
+            case '\\': text = "\\\\"; break;
+            case '\"': text = "\\\""; break;
+            case '\b': text = "\\b"; break;
+            case '\f': text = "\\f"; break;
+            case '\n': text = "\\n"; break;
+            case '\r': text = "\\r"; break;
+            case '\t': text = "\\t"; break;
+            default:
+            {
+                /* codepoint is in BMP */
+                if(codepoint < 0x10000)
+                {
+                    sprintf(seq, "\\u%04x", codepoint);
+                    length = 6;
+                }
+
+                /* not in BMP -> construct a UTF-16 surrogate pair */
+                else
+                {
+                    int32_t first, last;
+
+                    codepoint -= 0x10000;
+                    first = 0xD800 | ((codepoint & 0xffc00) >> 10);
+                    last = 0xDC00 | (codepoint & 0x003ff);
+
+                    sprintf(seq, "\\u%04x\\u%04x", first, last);
+                    length = 12;
+                }
+
+                text = seq;
+                break;
+            }
+        }
+
+        if(dump(text, length, data))
+            return -1;
+
+        str = pos = end;
+    }
+
+    return dump("\"", 1, data);
+}
+
+static int object_key_compare_keys(const void *key1, const void *key2)
+{
+    return strcmp((*(const object_key_t **)key1)->key,
+                  (*(const object_key_t **)key2)->key);
+}
+
+static int object_key_compare_serials(const void *key1, const void *key2)
+{
+    return (*(const object_key_t **)key1)->serial -
+           (*(const object_key_t **)key2)->serial;
+}
+
+static int do_dump(const json_t *json, unsigned long flags, int depth,
+                   dump_func dump, void *data)
+{
+    int ascii = flags & JSON_ENSURE_ASCII ? 1 : 0;
+
+    switch(json_typeof(json)) {
+        case JSON_NULL:
+            return dump("null", 4, data);
+
+        case JSON_TRUE:
+            return dump("true", 4, data);
+
+        case JSON_FALSE:
+            return dump("false", 5, data);
+
+        case JSON_INTEGER:
+        {
+            char buffer[MAX_INTEGER_STR_LENGTH];
+            int size;
+
+            size = snprintf(buffer, MAX_INTEGER_STR_LENGTH, "%d", json_integer_value(json));
+            if(size >= MAX_INTEGER_STR_LENGTH)
+                return -1;
+
+            return dump(buffer, size, data);
+        }
+
+        case JSON_REAL:
+        {
+            char buffer[MAX_REAL_STR_LENGTH];
+            int size;
+
+            size = snprintf(buffer, MAX_REAL_STR_LENGTH, "%.17g",
+                            json_real_value(json));
+            if(size >= MAX_REAL_STR_LENGTH)
+                return -1;
+
+            /* Make sure there's a dot or 'e' in the output. Otherwise
+               a real is converted to an integer when decoding */
+            if(strchr(buffer, '.') == NULL &&
+               strchr(buffer, 'e') == NULL)
+            {
+                if(size + 2 >= MAX_REAL_STR_LENGTH) {
+                    /* No space to append ".0" */
+                    return -1;
+                }
+                buffer[size] = '.';
+                buffer[size + 1] = '0';
+                size += 2;
+            }
+
+            return dump(buffer, size, data);
+        }
+
+        case JSON_STRING:
+            return dump_string(json_string_value(json), ascii, dump, data);
+
+        case JSON_ARRAY:
+        {
+            int i;
+            int n;
+            json_array_t *array;
+
+            /* detect circular references */
+            array = json_to_array(json);
+            if(array->visited)
+                goto array_error;
+            array->visited = 1;
+
+            n = json_array_size(json);
+
+            if(dump("[", 1, data))
+                goto array_error;
+            if(n == 0) {
+                array->visited = 0;
+                return dump("]", 1, data);
+            }
+            if(dump_indent(flags, depth + 1, 0, dump, data))
+                goto array_error;
+
+            for(i = 0; i < n; ++i) {
+                if(do_dump(json_array_get(json, i), flags, depth + 1,
+                           dump, data))
+                    goto array_error;
+
+                if(i < n - 1)
+                {
+                    if(dump(",", 1, data) ||
+                       dump_indent(flags, depth + 1, 1, dump, data))
+                        goto array_error;
+                }
+                else
+                {
+                    if(dump_indent(flags, depth, 0, dump, data))
+                        goto array_error;
+                }
+            }
+
+            array->visited = 0;
+            return dump("]", 1, data);
+
+        array_error:
+            array->visited = 0;
+            return -1;
+        }
+
+        case JSON_OBJECT:
+        {
+            json_object_t *object;
+            void *iter;
+            const char *separator;
+            int separator_length;
+
+            if(flags & JSON_COMPACT) {
+                separator = ":";
+                separator_length = 1;
+            }
+            else {
+                separator = ": ";
+                separator_length = 2;
+            }
+
+            /* detect circular references */
+            object = json_to_object(json);
+            if(object->visited)
+                goto object_error;
+            object->visited = 1;
+
+            iter = json_object_iter((json_t *)json);
+
+            if(dump("{", 1, data))
+                goto object_error;
+            if(!iter) {
+                object->visited = 0;
+                return dump("}", 1, data);
+            }
+            if(dump_indent(flags, depth + 1, 0, dump, data))
+                goto object_error;
+
+            if(flags & JSON_SORT_KEYS || flags & JSON_PRESERVE_ORDER)
+            {
+                const object_key_t **keys;
+                unsigned int size;
+                unsigned int i;
+                int (*cmp_func)(const void *, const void *);
+
+                size = json_object_size(json);
+                keys = malloc(size * sizeof(object_key_t *));
+                if(!keys)
+                    goto object_error;
+
+                i = 0;
+                while(iter)
+                {
+                    keys[i] = jsonp_object_iter_fullkey(iter);
+                    iter = json_object_iter_next((json_t *)json, iter);
+                    i++;
+                }
+                assert(i == size);
+
+                if(flags & JSON_SORT_KEYS)
+                    cmp_func = object_key_compare_keys;
+                else
+                    cmp_func = object_key_compare_serials;
+
+                qsort(keys, size, sizeof(object_key_t *), cmp_func);
+
+                for(i = 0; i < size; i++)
+                {
+                    const char *key;
+                    json_t *value;
+
+                    key = keys[i]->key;
+                    value = json_object_get(json, key);
+                    assert(value);
+
+                    dump_string(key, ascii, dump, data);
+                    if(dump(separator, separator_length, data) ||
+                       do_dump(value, flags, depth + 1, dump, data))
+                    {
+                        free(keys);
+                        goto object_error;
+                    }
+
+                    if(i < size - 1)
+                    {
+                        if(dump(",", 1, data) ||
+                           dump_indent(flags, depth + 1, 1, dump, data))
+                        {
+                            free(keys);
+                            goto object_error;
+                        }
+                    }
+                    else
+                    {
+                        if(dump_indent(flags, depth, 0, dump, data))
+                        {
+                            free(keys);
+                            goto object_error;
+                        }
+                    }
+                }
+
+                free(keys);
+            }
+            else
+            {
+                /* Don't sort keys */
+
+                while(iter)
+                {
+                    void *next = json_object_iter_next((json_t *)json, iter);
+
+                    dump_string(json_object_iter_key(iter), ascii, dump, data);
+                    if(dump(separator, separator_length, data) ||
+                       do_dump(json_object_iter_value(iter), flags, depth + 1,
+                               dump, data))
+                        goto object_error;
+
+                    if(next)
+                    {
+                        if(dump(",", 1, data) ||
+                           dump_indent(flags, depth + 1, 1, dump, data))
+                            goto object_error;
+                    }
+                    else
+                    {
+                        if(dump_indent(flags, depth, 0, dump, data))
+                            goto object_error;
+                    }
+
+                    iter = next;
+                }
+            }
+
+            object->visited = 0;
+            return dump("}", 1, data);
+
+        object_error:
+            object->visited = 0;
+            return -1;
+        }
+
+        default:
+            /* not reached */
+            return -1;
+    }
+}
+
+
+char *json_dumps(const json_t *json, unsigned long flags)
+{
+    strbuffer_t strbuff;
+    char *result;
+
+    if(!json_is_array(json) && !json_is_object(json))
+        return NULL;
+
+    if(strbuffer_init(&strbuff))
+        return NULL;
+
+    if(do_dump(json, flags, 0, dump_to_strbuffer, (void *)&strbuff)) {
+        strbuffer_close(&strbuff);
+        return NULL;
+    }
+
+    result = strdup(strbuffer_value(&strbuff));
+    strbuffer_close(&strbuff);
+
+    return result;
+}
+
+int json_dumpf(const json_t *json, FILE *output, unsigned long flags)
+{
+    if(!json_is_array(json) && !json_is_object(json))
+        return -1;
+
+    return do_dump(json, flags, 0, dump_to_file, (void *)output);
+}
+
+int json_dump_file(const json_t *json, const char *path, unsigned long flags)
+{
+    int result;
+
+    FILE *output = fopen(path, "w");
+    if(!output)
+        return -1;
+
+    result = json_dumpf(json, output, flags);
+
+    fclose(output);
+    return result;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/hashtable.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/hashtable.c
new file mode 100644
index 0000000..a312047
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/hashtable.c
@@ -0,0 +1,375 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * This library is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#include <config.h>
+
+#include <stdlib.h>
+#include "hashtable.h"
+
+typedef struct hashtable_list list_t;
+typedef struct hashtable_pair pair_t;
+typedef struct hashtable_bucket bucket_t;
+
+#define container_of(ptr_, type_, member_)                      \
+    ((type_ *)((char *)ptr_ - (size_t)&((type_ *)0)->member_))
+
+#define list_to_pair(list_)  container_of(list_, pair_t, list)
+
+static inline void list_init(list_t *list)
+{
+    list->next = list;
+    list->prev = list;
+}
+
+static inline void list_insert(list_t *list, list_t *node)
+{
+    node->next = list;
+    node->prev = list->prev;
+    list->prev->next = node;
+    list->prev = node;
+}
+
+static inline void list_remove(list_t *list)
+{
+    list->prev->next = list->next;
+    list->next->prev = list->prev;
+}
+
+static inline int bucket_is_empty(hashtable_t *hashtable, bucket_t *bucket)
+{
+    return bucket->first == &hashtable->list && bucket->first == bucket->last;
+}
+
+static void insert_to_bucket(hashtable_t *hashtable, bucket_t *bucket,
+                             list_t *list)
+{
+    if(bucket_is_empty(hashtable, bucket))
+    {
+        list_insert(&hashtable->list, list);
+        bucket->first = bucket->last = list;
+    }
+    else
+    {
+        list_insert(bucket->first, list);
+        bucket->first = list;
+    }
+}
+
+static unsigned int primes[] = {
+    5, 13, 23, 53, 97, 193, 389, 769, 1543, 3079, 6151, 12289, 24593,
+    49157, 98317, 196613, 393241, 786433, 1572869, 3145739, 6291469,
+    12582917, 25165843, 50331653, 100663319, 201326611, 402653189,
+    805306457, 1610612741
+};
+static const unsigned int num_primes = sizeof(primes) / sizeof(unsigned int);
+
+static inline unsigned int num_buckets(hashtable_t *hashtable)
+{
+    return primes[hashtable->num_buckets];
+}
+
+
+static pair_t *hashtable_find_pair(hashtable_t *hashtable, bucket_t *bucket,
+                                   const void *key, unsigned int hash)
+{
+    list_t *list;
+    pair_t *pair;
+
+    if(bucket_is_empty(hashtable, bucket))
+        return NULL;
+
+    list = bucket->first;
+    while(1)
+    {
+        pair = list_to_pair(list);
+        if(pair->hash == hash && hashtable->cmp_keys(pair->key, key))
+            return pair;
+
+        if(list == bucket->last)
+            break;
+
+        list = list->next;
+    }
+
+    return NULL;
+}
+
+/* returns 0 on success, -1 if key was not found */
+static int hashtable_do_del(hashtable_t *hashtable,
+                            const void *key, unsigned int hash)
+{
+    pair_t *pair;
+    bucket_t *bucket;
+    unsigned int index;
+
+    index = hash % num_buckets(hashtable);
+    bucket = &hashtable->buckets[index];
+
+    pair = hashtable_find_pair(hashtable, bucket, key, hash);
+    if(!pair)
+        return -1;
+
+    if(&pair->list == bucket->first && &pair->list == bucket->last)
+        bucket->first = bucket->last = &hashtable->list;
+
+    else if(&pair->list == bucket->first)
+        bucket->first = pair->list.next;
+
+    else if(&pair->list == bucket->last)
+        bucket->last = pair->list.prev;
+
+    list_remove(&pair->list);
+
+    if(hashtable->free_key)
+        hashtable->free_key(pair->key);
+    if(hashtable->free_value)
+        hashtable->free_value(pair->value);
+
+    free(pair);
+    hashtable->size--;
+
+    return 0;
+}
+
+static void hashtable_do_clear(hashtable_t *hashtable)
+{
+    list_t *list, *next;
+    pair_t *pair;
+
+    for(list = hashtable->list.next; list != &hashtable->list; list = next)
+    {
+        next = list->next;
+        pair = list_to_pair(list);
+        if(hashtable->free_key)
+            hashtable->free_key(pair->key);
+        if(hashtable->free_value)
+            hashtable->free_value(pair->value);
+        free(pair);
+    }
+}
+
+static int hashtable_do_rehash(hashtable_t *hashtable)
+{
+    list_t *list, *next;
+    pair_t *pair;
+    unsigned int i, index, new_size;
+
+    free(hashtable->buckets);
+
+    hashtable->num_buckets++;
+    new_size = num_buckets(hashtable);
+
+    hashtable->buckets = malloc(new_size * sizeof(bucket_t));
+    if(!hashtable->buckets)
+        return -1;
+
+    for(i = 0; i < num_buckets(hashtable); i++)
+    {
+        hashtable->buckets[i].first = hashtable->buckets[i].last =
+            &hashtable->list;
+    }
+
+    list = hashtable->list.next;
+    list_init(&hashtable->list);
+
+    for(; list != &hashtable->list; list = next) {
+        next = list->next;
+        pair = list_to_pair(list);
+        index = pair->hash % new_size;
+        insert_to_bucket(hashtable, &hashtable->buckets[index], &pair->list);
+    }
+
+    return 0;
+}
+
+
+hashtable_t *hashtable_create(key_hash_fn hash_key, key_cmp_fn cmp_keys,
+                              free_fn free_key, free_fn free_value)
+{
+    hashtable_t *hashtable = malloc(sizeof(hashtable_t));
+    if(!hashtable)
+        return NULL;
+
+    if(hashtable_init(hashtable, hash_key, cmp_keys, free_key, free_value))
+    {
+        free(hashtable);
+        return NULL;
+    }
+
+    return hashtable;
+}
+
+void hashtable_destroy(hashtable_t *hashtable)
+{
+    hashtable_close(hashtable);
+    free(hashtable);
+}
+
+int hashtable_init(hashtable_t *hashtable,
+                   key_hash_fn hash_key, key_cmp_fn cmp_keys,
+                   free_fn free_key, free_fn free_value)
+{
+    unsigned int i;
+
+    hashtable->size = 0;
+    hashtable->num_buckets = 0;  /* index to primes[] */
+    hashtable->buckets = malloc(num_buckets(hashtable) * sizeof(bucket_t));
+    if(!hashtable->buckets)
+        return -1;
+
+    list_init(&hashtable->list);
+
+    hashtable->hash_key = hash_key;
+    hashtable->cmp_keys = cmp_keys;
+    hashtable->free_key = free_key;
+    hashtable->free_value = free_value;
+
+    for(i = 0; i < num_buckets(hashtable); i++)
+    {
+        hashtable->buckets[i].first = hashtable->buckets[i].last =
+            &hashtable->list;
+    }
+
+    return 0;
+}
+
+void hashtable_close(hashtable_t *hashtable)
+{
+    hashtable_do_clear(hashtable);
+    free(hashtable->buckets);
+}
+
+int hashtable_set(hashtable_t *hashtable, void *key, void *value)
+{
+    pair_t *pair;
+    bucket_t *bucket;
+    unsigned int hash, index;
+
+    /* rehash if the load ratio exceeds 1 */
+    if(hashtable->size >= num_buckets(hashtable))
+        if(hashtable_do_rehash(hashtable))
+            return -1;
+
+    hash = hashtable->hash_key(key);
+    index = hash % num_buckets(hashtable);
+    bucket = &hashtable->buckets[index];
+    pair = hashtable_find_pair(hashtable, bucket, key, hash);
+
+    if(pair)
+    {
+        if(hashtable->free_key)
+            hashtable->free_key(key);
+        if(hashtable->free_value)
+            hashtable->free_value(pair->value);
+        pair->value = value;
+    }
+    else
+    {
+        pair = malloc(sizeof(pair_t));
+        if(!pair)
+            return -1;
+
+        pair->key = key;
+        pair->value = value;
+        pair->hash = hash;
+        list_init(&pair->list);
+
+        insert_to_bucket(hashtable, bucket, &pair->list);
+
+        hashtable->size++;
+    }
+    return 0;
+}
+
+void *hashtable_get(hashtable_t *hashtable, const void *key)
+{
+    pair_t *pair;
+    unsigned int hash;
+    bucket_t *bucket;
+
+    hash = hashtable->hash_key(key);
+    bucket = &hashtable->buckets[hash % num_buckets(hashtable)];
+
+    pair = hashtable_find_pair(hashtable, bucket, key, hash);
+    if(!pair)
+        return NULL;
+
+    return pair->value;
+}
+
+int hashtable_del(hashtable_t *hashtable, const void *key)
+{
+    unsigned int hash = hashtable->hash_key(key);
+    return hashtable_do_del(hashtable, key, hash);
+}
+
+void hashtable_clear(hashtable_t *hashtable)
+{
+    unsigned int i;
+
+    hashtable_do_clear(hashtable);
+
+    for(i = 0; i < num_buckets(hashtable); i++)
+    {
+        hashtable->buckets[i].first = hashtable->buckets[i].last =
+            &hashtable->list;
+    }
+
+    list_init(&hashtable->list);
+    hashtable->size = 0;
+}
+
+void *hashtable_iter(hashtable_t *hashtable)
+{
+    return hashtable_iter_next(hashtable, &hashtable->list);
+}
+
+void *hashtable_iter_at(hashtable_t *hashtable, const void *key)
+{
+    pair_t *pair;
+    unsigned int hash;
+    bucket_t *bucket;
+
+    hash = hashtable->hash_key(key);
+    bucket = &hashtable->buckets[hash % num_buckets(hashtable)];
+
+    pair = hashtable_find_pair(hashtable, bucket, key, hash);
+    if(!pair)
+        return NULL;
+
+    return &pair->list;
+}
+
+void *hashtable_iter_next(hashtable_t *hashtable, void *iter)
+{
+    list_t *list = (list_t *)iter;
+    if(list->next == &hashtable->list)
+        return NULL;
+    return list->next;
+}
+
+void *hashtable_iter_key(void *iter)
+{
+    pair_t *pair = list_to_pair((list_t *)iter);
+    return pair->key;
+}
+
+void *hashtable_iter_value(void *iter)
+{
+    pair_t *pair = list_to_pair((list_t *)iter);
+    return pair->value;
+}
+
+void hashtable_iter_set(hashtable_t *hashtable, void *iter, void *value)
+{
+    pair_t *pair = list_to_pair((list_t *)iter);
+
+    if(hashtable->free_value)
+        hashtable->free_value(pair->value);
+
+    pair->value = value;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/hashtable.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/hashtable.h
new file mode 100644
index 0000000..f03a769
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/hashtable.h
@@ -0,0 +1,207 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * This library is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef HASHTABLE_H
+#define HASHTABLE_H
+
+typedef unsigned int (*key_hash_fn)(const void *key);
+typedef int (*key_cmp_fn)(const void *key1, const void *key2);
+typedef void (*free_fn)(void *key);
+
+struct hashtable_list {
+    struct hashtable_list *prev;
+    struct hashtable_list *next;
+};
+
+struct hashtable_pair {
+    void *key;
+    void *value;
+    unsigned int hash;
+    struct hashtable_list list;
+};
+
+struct hashtable_bucket {
+    struct hashtable_list *first;
+    struct hashtable_list *last;
+};
+
+typedef struct hashtable {
+    unsigned int size;
+    struct hashtable_bucket *buckets;
+    unsigned int num_buckets;  /* index to primes[] */
+    struct hashtable_list list;
+
+    key_hash_fn hash_key;
+    key_cmp_fn cmp_keys;  /* returns non-zero for equal keys */
+    free_fn free_key;
+    free_fn free_value;
+} hashtable_t;
+
+/**
+ * hashtable_create - Create a hashtable object
+ *
+ * @hash_key: The key hashing function
+ * @cmp_keys: The key compare function. Returns non-zero for equal and
+ *     zero for unequal unequal keys
+ * @free_key: If non-NULL, called for a key that is no longer referenced.
+ * @free_value: If non-NULL, called for a value that is no longer referenced.
+ *
+ * Returns a new hashtable object that should be freed with
+ * hashtable_destroy when it's no longer used, or NULL on failure (out
+ * of memory).
+ */
+hashtable_t *hashtable_create(key_hash_fn hash_key, key_cmp_fn cmp_keys,
+                              free_fn free_key, free_fn free_value);
+
+/**
+ * hashtable_destroy - Destroy a hashtable object
+ *
+ * @hashtable: The hashtable
+ *
+ * Destroys a hashtable created with hashtable_create().
+ */
+void hashtable_destroy(hashtable_t *hashtable);
+
+/**
+ * hashtable_init - Initialize a hashtable object
+ *
+ * @hashtable: The (statically allocated) hashtable object
+ * @hash_key: The key hashing function
+ * @cmp_keys: The key compare function. Returns non-zero for equal and
+ *     zero for unequal unequal keys
+ * @free_key: If non-NULL, called for a key that is no longer referenced.
+ * @free_value: If non-NULL, called for a value that is no longer referenced.
+ *
+ * Initializes a statically allocated hashtable object. The object
+ * should be cleared with hashtable_close when it's no longer used.
+ *
+ * Returns 0 on success, -1 on error (out of memory).
+ */
+int hashtable_init(hashtable_t *hashtable,
+                   key_hash_fn hash_key, key_cmp_fn cmp_keys,
+                   free_fn free_key, free_fn free_value);
+
+/**
+ * hashtable_close - Release all resources used by a hashtable object
+ *
+ * @hashtable: The hashtable
+ *
+ * Destroys a statically allocated hashtable object.
+ */
+void hashtable_close(hashtable_t *hashtable);
+
+/**
+ * hashtable_set - Add/modify value in hashtable
+ *
+ * @hashtable: The hashtable object
+ * @key: The key
+ * @value: The value
+ *
+ * If a value with the given key already exists, its value is replaced
+ * with the new value.
+ *
+ * Key and value are "stealed" in the sense that hashtable frees them
+ * automatically when they are no longer used. The freeing is
+ * accomplished by calling free_key and free_value functions that were
+ * supplied to hashtable_new. In case one or both of the free
+ * functions is NULL, the corresponding item is not "stealed".
+ *
+ * Returns 0 on success, -1 on failure (out of memory).
+ */
+int hashtable_set(hashtable_t *hashtable, void *key, void *value);
+
+/**
+ * hashtable_get - Get a value associated with a key
+ *
+ * @hashtable: The hashtable object
+ * @key: The key
+ *
+ * Returns value if it is found, or NULL otherwise.
+ */
+void *hashtable_get(hashtable_t *hashtable, const void *key);
+
+/**
+ * hashtable_del - Remove a value from the hashtable
+ *
+ * @hashtable: The hashtable object
+ * @key: The key
+ *
+ * Returns 0 on success, or -1 if the key was not found.
+ */
+int hashtable_del(hashtable_t *hashtable, const void *key);
+
+/**
+ * hashtable_clear - Clear hashtable
+ *
+ * @hashtable: The hashtable object
+ *
+ * Removes all items from the hashtable.
+ */
+void hashtable_clear(hashtable_t *hashtable);
+
+/**
+ * hashtable_iter - Iterate over hashtable
+ *
+ * @hashtable: The hashtable object
+ *
+ * Returns an opaque iterator to the first element in the hashtable.
+ * The iterator should be passed to hashtable_iter_* functions.
+ * The hashtable items are not iterated over in any particular order.
+ *
+ * There's no need to free the iterator in any way. The iterator is
+ * valid as long as the item that is referenced by the iterator is not
+ * deleted. Other values may be added or deleted. In particular,
+ * hashtable_iter_next() may be called on an iterator, and after that
+ * the key/value pair pointed by the old iterator may be deleted.
+ */
+void *hashtable_iter(hashtable_t *hashtable);
+
+/**
+ * hashtable_iter_at - Return an iterator at a specific key
+ *
+ * @hashtable: The hashtable object
+ * @key: The key that the iterator should point to
+ *
+ * Like hashtable_iter() but returns an iterator pointing to a
+ * specific key.
+ */
+void *hashtable_iter_at(hashtable_t *hashtable, const void *key);
+
+/**
+ * hashtable_iter_next - Advance an iterator
+ *
+ * @hashtable: The hashtable object
+ * @iter: The iterator
+ *
+ * Returns a new iterator pointing to the next element in the
+ * hashtable or NULL if the whole hastable has been iterated over.
+ */
+void *hashtable_iter_next(hashtable_t *hashtable, void *iter);
+
+/**
+ * hashtable_iter_key - Retrieve the key pointed by an iterator
+ *
+ * @iter: The iterator
+ */
+void *hashtable_iter_key(void *iter);
+
+/**
+ * hashtable_iter_value - Retrieve the value pointed by an iterator
+ *
+ * @iter: The iterator
+ */
+void *hashtable_iter_value(void *iter);
+
+/**
+ * hashtable_iter_set - Set the value pointed by an iterator
+ *
+ * @iter: The iterator
+ * @value: The value to set
+ */
+void hashtable_iter_set(hashtable_t *hashtable, void *iter, void *value);
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/jansson.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/jansson.h
new file mode 100644
index 0000000..4c526fe
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/jansson.h
@@ -0,0 +1,191 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef JANSSON_H
+#define JANSSON_H
+
+#include <stdio.h>
+
+#ifndef __cplusplus
+#define JSON_INLINE inline
+#else
+#define JSON_INLINE inline
+extern "C" {
+#endif
+
+/* types */
+
+typedef enum {
+    JSON_OBJECT,
+    JSON_ARRAY,
+    JSON_STRING,
+    JSON_INTEGER,
+    JSON_REAL,
+    JSON_TRUE,
+    JSON_FALSE,
+    JSON_NULL
+} json_type;
+
+typedef struct {
+    json_type type;
+    unsigned long refcount;
+} json_t;
+
+#define json_typeof(json)      ((json)->type)
+#define json_is_object(json)   (json && json_typeof(json) == JSON_OBJECT)
+#define json_is_array(json)    (json && json_typeof(json) == JSON_ARRAY)
+#define json_is_string(json)   (json && json_typeof(json) == JSON_STRING)
+#define json_is_integer(json)  (json && json_typeof(json) == JSON_INTEGER)
+#define json_is_real(json)     (json && json_typeof(json) == JSON_REAL)
+#define json_is_number(json)   (json_is_integer(json) || json_is_real(json))
+#define json_is_true(json)     (json && json_typeof(json) == JSON_TRUE)
+#define json_is_false(json)    (json && json_typeof(json) == JSON_FALSE)
+#define json_is_boolean(json)  (json_is_true(json) || json_is_false(json))
+#define json_is_null(json)     (json && json_typeof(json) == JSON_NULL)
+
+/* construction, destruction, reference counting */
+
+json_t *json_object(void);
+json_t *json_array(void);
+json_t *json_string(const char *value);
+json_t *json_string_nocheck(const char *value);
+json_t *json_integer(int value);
+json_t *json_real(double value);
+json_t *json_true(void);
+json_t *json_false(void);
+json_t *json_null(void);
+
+static JSON_INLINE
+json_t *json_incref(json_t *json)
+{
+    if(json && json->refcount != (unsigned int)-1)
+        ++json->refcount;
+    return json;
+}
+
+/* do not call json_delete directly */
+void json_delete(json_t *json);
+
+static JSON_INLINE
+void json_decref(json_t *json)
+{
+    if(json && json->refcount != (unsigned int)-1 && --json->refcount == 0)
+        json_delete(json);
+}
+
+
+/* getters, setters, manipulation */
+
+unsigned int json_object_size(const json_t *object);
+json_t *json_object_get(const json_t *object, const char *key);
+int json_object_set_new(json_t *object, const char *key, json_t *value);
+int json_object_set_new_nocheck(json_t *object, const char *key, json_t *value);
+int json_object_del(json_t *object, const char *key);
+int json_object_clear(json_t *object);
+int json_object_update(json_t *object, json_t *other);
+void *json_object_iter(json_t *object);
+void *json_object_iter_at(json_t *object, const char *key);
+void *json_object_iter_next(json_t *object, void *iter);
+const char *json_object_iter_key(void *iter);
+json_t *json_object_iter_value(void *iter);
+int json_object_iter_set_new(json_t *object, void *iter, json_t *value);
+
+static JSON_INLINE
+int json_object_set(json_t *object, const char *key, json_t *value)
+{
+    return json_object_set_new(object, key, json_incref(value));
+}
+
+static JSON_INLINE
+int json_object_set_nocheck(json_t *object, const char *key, json_t *value)
+{
+    return json_object_set_new_nocheck(object, key, json_incref(value));
+}
+
+static inline
+int json_object_iter_set(json_t *object, void *iter, json_t *value)
+{
+    return json_object_iter_set_new(object, iter, json_incref(value));
+}
+
+unsigned int json_array_size(const json_t *array);
+json_t *json_array_get(const json_t *array, unsigned int index);
+int json_array_set_new(json_t *array, unsigned int index, json_t *value);
+int json_array_append_new(json_t *array, json_t *value);
+int json_array_insert_new(json_t *array, unsigned int index, json_t *value);
+int json_array_remove(json_t *array, unsigned int index);
+int json_array_clear(json_t *array);
+int json_array_extend(json_t *array, json_t *other);
+
+static JSON_INLINE
+int json_array_set(json_t *array, unsigned int index, json_t *value)
+{
+    return json_array_set_new(array, index, json_incref(value));
+}
+
+static JSON_INLINE
+int json_array_append(json_t *array, json_t *value)
+{
+    return json_array_append_new(array, json_incref(value));
+}
+
+static JSON_INLINE
+int json_array_insert(json_t *array, unsigned int index, json_t *value)
+{
+    return json_array_insert_new(array, index, json_incref(value));
+}
+
+const char *json_string_value(const json_t *string);
+int json_integer_value(const json_t *integer);
+double json_real_value(const json_t *real);
+double json_number_value(const json_t *json);
+
+int json_string_set(json_t *string, const char *value);
+int json_string_set_nocheck(json_t *string, const char *value);
+int json_integer_set(json_t *integer, int value);
+int json_real_set(json_t *real, double value);
+
+
+/* equality */
+
+int json_equal(json_t *value1, json_t *value2);
+
+
+/* copying */
+
+json_t *json_copy(json_t *value);
+json_t *json_deep_copy(json_t *value);
+
+
+/* loading, printing */
+
+#define JSON_ERROR_TEXT_LENGTH  160
+
+typedef struct {
+    char text[JSON_ERROR_TEXT_LENGTH];
+    int line;
+} json_error_t;
+
+json_t *json_loads(const char *input, json_error_t *error);
+json_t *json_loadf(FILE *input, json_error_t *error);
+json_t *json_load_file(const char *path, json_error_t *error);
+
+#define JSON_INDENT(n)      (n & 0xFF)
+#define JSON_COMPACT        0x100
+#define JSON_ENSURE_ASCII   0x200
+#define JSON_SORT_KEYS      0x400
+#define JSON_PRESERVE_ORDER 0x800
+
+char *json_dumps(const json_t *json, unsigned long flags);
+int json_dumpf(const json_t *json, FILE *output, unsigned long flags);
+int json_dump_file(const json_t *json, const char *path, unsigned long flags);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/jansson_private.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/jansson_private.h
new file mode 100644
index 0000000..4490702
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/jansson_private.h
@@ -0,0 +1,60 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef JANSSON_PRIVATE_H
+#define JANSSON_PRIVATE_H
+
+#include "jansson.h"
+#include "hashtable.h"
+
+#define container_of(ptr_, type_, member_)  \
+    ((type_ *)((char *)ptr_ - (size_t)&((type_ *)0)->member_))
+
+typedef struct {
+    json_t json;
+    hashtable_t hashtable;
+    unsigned long serial;
+    int visited;
+} json_object_t;
+
+typedef struct {
+    json_t json;
+    unsigned int size;
+    unsigned int entries;
+    json_t **table;
+    int visited;
+} json_array_t;
+
+typedef struct {
+    json_t json;
+    char *value;
+} json_string_t;
+
+typedef struct {
+    json_t json;
+    double value;
+} json_real_t;
+
+typedef struct {
+    json_t json;
+    int value;
+} json_integer_t;
+
+#define json_to_object(json_)  container_of(json_, json_object_t, json)
+#define json_to_array(json_)   container_of(json_, json_array_t, json)
+#define json_to_string(json_)  container_of(json_, json_string_t, json)
+#define json_to_real(json_)   container_of(json_, json_real_t, json)
+#define json_to_integer(json_) container_of(json_, json_integer_t, json)
+
+typedef struct {
+    unsigned long serial;
+    char key[];
+} object_key_t;
+
+const object_key_t *jsonp_object_iter_fullkey(void *iter);
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/load.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/load.c
new file mode 100644
index 0000000..d49a4da
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/load.c
@@ -0,0 +1,879 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#define _GNU_SOURCE
+#include <ctype.h>
+#include <errno.h>
+#include <limits.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <assert.h>
+
+#include <jansson.h>
+#include "jansson_private.h"
+#include "strbuffer.h"
+#include "utf.h"
+
+#define TOKEN_INVALID         -1
+#define TOKEN_EOF              0
+#define TOKEN_STRING         256
+#define TOKEN_INTEGER        257
+#define TOKEN_REAL           258
+#define TOKEN_TRUE           259
+#define TOKEN_FALSE          260
+#define TOKEN_NULL           261
+
+/* read one byte from stream, return EOF on end of file */
+typedef int (*get_func)(void *data);
+
+/* return non-zero if end of file has been reached */
+typedef int (*eof_func)(void *data);
+
+typedef struct {
+    get_func get;
+    eof_func eof;
+    void *data;
+    int stream_pos;
+    char buffer[5];
+    int buffer_pos;
+} stream_t;
+
+
+typedef struct {
+    stream_t stream;
+    strbuffer_t saved_text;
+    int token;
+    int line, column;
+    union {
+        char *string;
+        int integer;
+        double real;
+    } value;
+} lex_t;
+
+
+/*** error reporting ***/
+
+static void error_init(json_error_t *error)
+{
+    if(error)
+    {
+        error->text[0] = '\0';
+        error->line = -1;
+    }
+}
+
+static void error_set(json_error_t *error, const lex_t *lex,
+                      const char *msg, ...)
+{
+    va_list ap;
+    char text[JSON_ERROR_TEXT_LENGTH];
+
+    if(!error || error->text[0] != '\0') {
+        /* error already set */
+        return;
+    }
+
+    va_start(ap, msg);
+    vsnprintf(text, JSON_ERROR_TEXT_LENGTH, msg, ap);
+    va_end(ap);
+
+    if(lex)
+    {
+        const char *saved_text = strbuffer_value(&lex->saved_text);
+        error->line = lex->line;
+        if(saved_text && saved_text[0])
+        {
+            if(lex->saved_text.length <= 20) {
+                snprintf(error->text, JSON_ERROR_TEXT_LENGTH,
+                         "%s near '%s'", text, saved_text);
+            }
+            else
+                snprintf(error->text, JSON_ERROR_TEXT_LENGTH, "%s", text);
+        }
+        else
+        {
+            snprintf(error->text, JSON_ERROR_TEXT_LENGTH,
+                     "%s near end of file", text);
+        }
+    }
+    else
+    {
+        error->line = -1;
+        snprintf(error->text, JSON_ERROR_TEXT_LENGTH, "%s", text);
+    }
+}
+
+
+/*** lexical analyzer ***/
+
+static void
+stream_init(stream_t *stream, get_func get, eof_func eof, void *data)
+{
+    stream->get = get;
+    stream->eof = eof;
+    stream->data = data;
+    stream->stream_pos = 0;
+    stream->buffer[0] = '\0';
+    stream->buffer_pos = 0;
+}
+
+static char stream_get(stream_t *stream, json_error_t *error)
+{
+    char c;
+
+    if(!stream->buffer[stream->buffer_pos])
+    {
+        stream->buffer[0] = stream->get(stream->data);
+        stream->buffer_pos = 0;
+
+        c = stream->buffer[0];
+
+        if((unsigned char)c >= 0x80 && c != (char)EOF)
+        {
+            /* multi-byte UTF-8 sequence */
+            int i, count;
+
+            count = utf8_check_first(c);
+            if(!count)
+                goto out;
+
+            assert(count >= 2);
+
+            for(i = 1; i < count; i++)
+                stream->buffer[i] = stream->get(stream->data);
+
+            if(!utf8_check_full(stream->buffer, count, NULL))
+                goto out;
+
+            stream->stream_pos += count;
+            stream->buffer[count] = '\0';
+        }
+        else {
+            stream->buffer[1] = '\0';
+            stream->stream_pos++;
+        }
+    }
+
+    return stream->buffer[stream->buffer_pos++];
+
+out:
+    error_set(error, NULL, "unable to decode byte 0x%x at position %d",
+              (unsigned char)c, stream->stream_pos);
+
+    stream->buffer[0] = EOF;
+    stream->buffer[1] = '\0';
+    stream->buffer_pos = 1;
+
+    return EOF;
+}
+
+static void stream_unget(stream_t *stream, char c)
+{
+    assert(stream->buffer_pos > 0);
+    stream->buffer_pos--;
+    assert(stream->buffer[stream->buffer_pos] == c);
+}
+
+
+static int lex_get(lex_t *lex, json_error_t *error)
+{
+    return stream_get(&lex->stream, error);
+}
+
+static int lex_eof(lex_t *lex)
+{
+    return lex->stream.eof(lex->stream.data);
+}
+
+static void lex_save(lex_t *lex, char c)
+{
+    strbuffer_append_byte(&lex->saved_text, c);
+}
+
+static int lex_get_save(lex_t *lex, json_error_t *error)
+{
+    char c = stream_get(&lex->stream, error);
+    lex_save(lex, c);
+    return c;
+}
+
+static void lex_unget_unsave(lex_t *lex, char c)
+{
+    char d;
+    stream_unget(&lex->stream, c);
+    d = strbuffer_pop(&lex->saved_text);
+    assert(c == d);
+}
+
+static void lex_save_cached(lex_t *lex)
+{
+    while(lex->stream.buffer[lex->stream.buffer_pos] != '\0')
+    {
+        lex_save(lex, lex->stream.buffer[lex->stream.buffer_pos]);
+        lex->stream.buffer_pos++;
+    }
+}
+
+/* assumes that str points to 'u' plus at least 4 valid hex digits */
+static int32_t decode_unicode_escape(const char *str)
+{
+    int i;
+    int32_t value = 0;
+
+    assert(str[0] == 'u');
+
+    for(i = 1; i <= 4; i++) {
+        char c = str[i];
+        value <<= 4;
+        if(isdigit(c))
+            value += c - '0';
+        else if(islower(c))
+            value += c - 'a' + 10;
+        else if(isupper(c))
+            value += c - 'A' + 10;
+        else
+            assert(0);
+    }
+
+    return value;
+}
+
+static void lex_scan_string(lex_t *lex, json_error_t *error)
+{
+    char c;
+    const char *p;
+    char *t;
+    int i;
+
+    lex->value.string = NULL;
+    lex->token = TOKEN_INVALID;
+
+    c = lex_get_save(lex, error);
+
+    while(c != '"') {
+        if(c == (char)EOF) {
+            lex_unget_unsave(lex, c);
+            if(lex_eof(lex))
+                error_set(error, lex, "premature end of input");
+            goto out;
+        }
+
+        else if((unsigned char)c <= 0x1F) {
+            /* control character */
+            lex_unget_unsave(lex, c);
+            if(c == '\n')
+                error_set(error, lex, "unexpected newline", c);
+            else
+                error_set(error, lex, "control character 0x%x", c);
+            goto out;
+        }
+
+        else if(c == '\\') {
+            c = lex_get_save(lex, error);
+            if(c == 'u') {
+                c = lex_get_save(lex, error);
+                for(i = 0; i < 4; i++) {
+                    if(!isxdigit(c)) {
+                        lex_unget_unsave(lex, c);
+                        error_set(error, lex, "invalid escape");
+                        goto out;
+                    }
+                    c = lex_get_save(lex, error);
+                }
+            }
+            else if(c == '"' || c == '\\' || c == '/' || c == 'b' ||
+                    c == 'f' || c == 'n' || c == 'r' || c == 't')
+                c = lex_get_save(lex, error);
+            else {
+                lex_unget_unsave(lex, c);
+                error_set(error, lex, "invalid escape");
+                goto out;
+            }
+        }
+        else
+            c = lex_get_save(lex, error);
+    }
+
+    /* the actual value is at most of the same length as the source
+       string, because:
+         - shortcut escapes (e.g. "\t") (length 2) are converted to 1 byte
+         - a single \uXXXX escape (length 6) is converted to at most 3 bytes
+         - two \uXXXX escapes (length 12) forming an UTF-16 surrogate pair
+           are converted to 4 bytes
+    */
+    lex->value.string = malloc(lex->saved_text.length + 1);
+    if(!lex->value.string) {
+        /* this is not very nice, since TOKEN_INVALID is returned */
+        goto out;
+    }
+
+    /* the target */
+    t = lex->value.string;
+
+    /* + 1 to skip the " */
+    p = strbuffer_value(&lex->saved_text) + 1;
+
+    while(*p != '"') {
+        if(*p == '\\') {
+            p++;
+            if(*p == 'u') {
+                char buffer[4];
+                int length;
+                int32_t value;
+
+                value = decode_unicode_escape(p);
+                p += 5;
+
+                if(0xD800 <= value && value <= 0xDBFF) {
+                    /* surrogate pair */
+                    if(*p == '\\' && *(p + 1) == 'u') {
+                        int32_t value2 = decode_unicode_escape(++p);
+                        p += 5;
+
+                        if(0xDC00 <= value2 && value2 <= 0xDFFF) {
+                            /* valid second surrogate */
+                            value =
+                                ((value - 0xD800) << 10) +
+                                (value2 - 0xDC00) +
+                                0x10000;
+                        }
+                        else {
+                            /* invalid second surrogate */
+                            error_set(error, lex,
+                                      "invalid Unicode '\\u%04X\\u%04X'",
+                                      value, value2);
+                            goto out;
+                        }
+                    }
+                    else {
+                        /* no second surrogate */
+                        error_set(error, lex, "invalid Unicode '\\u%04X'",
+                                  value);
+                        goto out;
+                    }
+                }
+                else if(0xDC00 <= value && value <= 0xDFFF) {
+                    error_set(error, lex, "invalid Unicode '\\u%04X'", value);
+                    goto out;
+                }
+                else if(value == 0)
+                {
+                    error_set(error, lex, "\\u0000 is not allowed");
+                    goto out;
+                }
+
+                if(utf8_encode(value, buffer, &length))
+                    assert(0);
+
+                memcpy(t, buffer, length);
+                t += length;
+            }
+            else {
+                switch(*p) {
+                    case '"': case '\\': case '/':
+                        *t = *p; break;
+                    case 'b': *t = '\b'; break;
+                    case 'f': *t = '\f'; break;
+                    case 'n': *t = '\n'; break;
+                    case 'r': *t = '\r'; break;
+                    case 't': *t = '\t'; break;
+                    default: assert(0);
+                }
+                t++;
+                p++;
+            }
+        }
+        else
+            *(t++) = *(p++);
+    }
+    *t = '\0';
+    lex->token = TOKEN_STRING;
+    return;
+
+out:
+    free(lex->value.string);
+}
+
+static int lex_scan_number(lex_t *lex, char c, json_error_t *error)
+{
+    const char *saved_text;
+    char *end;
+    double value;
+
+    lex->token = TOKEN_INVALID;
+
+    if(c == '-')
+        c = lex_get_save(lex, error);
+
+    if(c == '0') {
+        c = lex_get_save(lex, error);
+        if(isdigit(c)) {
+            lex_unget_unsave(lex, c);
+            goto out;
+        }
+    }
+    else if(isdigit(c)) {
+        c = lex_get_save(lex, error);
+        while(isdigit(c))
+            c = lex_get_save(lex, error);
+    }
+    else {
+      lex_unget_unsave(lex, c);
+      goto out;
+    }
+
+    if(c != '.' && c != 'E' && c != 'e') {
+        long value;
+
+        lex_unget_unsave(lex, c);
+
+        saved_text = strbuffer_value(&lex->saved_text);
+        value = strtol(saved_text, &end, 10);
+        assert(end == saved_text + lex->saved_text.length);
+
+        if((value == LONG_MAX && errno == ERANGE) || value > INT_MAX) {
+            error_set(error, lex, "too big integer");
+            goto out;
+        }
+        else if((value == LONG_MIN && errno == ERANGE) || value < INT_MIN) {
+            error_set(error, lex, "too big negative integer");
+            goto out;
+        }
+
+        lex->token = TOKEN_INTEGER;
+        lex->value.integer = (int)value;
+        return 0;
+    }
+
+    if(c == '.') {
+        c = lex_get(lex, error);
+        if(!isdigit(c))
+            goto out;
+        lex_save(lex, c);
+
+        c = lex_get_save(lex, error);
+        while(isdigit(c))
+            c = lex_get_save(lex, error);
+    }
+
+    if(c == 'E' || c == 'e') {
+        c = lex_get_save(lex, error);
+        if(c == '+' || c == '-')
+            c = lex_get_save(lex, error);
+
+        if(!isdigit(c)) {
+            lex_unget_unsave(lex, c);
+            goto out;
+        }
+
+        c = lex_get_save(lex, error);
+        while(isdigit(c))
+            c = lex_get_save(lex, error);
+    }
+
+    lex_unget_unsave(lex, c);
+
+    saved_text = strbuffer_value(&lex->saved_text);
+    value = strtod(saved_text, &end);
+    assert(end == saved_text + lex->saved_text.length);
+
+    if(errno == ERANGE && value != 0) {
+        error_set(error, lex, "real number overflow");
+        goto out;
+    }
+
+    lex->token = TOKEN_REAL;
+    lex->value.real = value;
+    return 0;
+
+out:
+    return -1;
+}
+
+static int lex_scan(lex_t *lex, json_error_t *error)
+{
+    char c;
+
+    strbuffer_clear(&lex->saved_text);
+
+    if(lex->token == TOKEN_STRING) {
+        free(lex->value.string);
+        lex->value.string = NULL;
+    }
+
+    c = lex_get(lex, error);
+    while(c == ' ' || c == '\t' || c == '\n' || c == '\r')
+    {
+        if(c == '\n')
+            lex->line++;
+
+        c = lex_get(lex, error);
+    }
+
+    if(c == (char)EOF) {
+        if(lex_eof(lex))
+            lex->token = TOKEN_EOF;
+        else
+            lex->token = TOKEN_INVALID;
+        goto out;
+    }
+
+    lex_save(lex, c);
+
+    if(c == '{' || c == '}' || c == '[' || c == ']' || c == ':' || c == ',')
+        lex->token = c;
+
+    else if(c == '"')
+        lex_scan_string(lex, error);
+
+    else if(isdigit(c) || c == '-') {
+        if(lex_scan_number(lex, c, error))
+            goto out;
+    }
+
+    else if(isupper(c) || islower(c)) {
+        /* eat up the whole identifier for clearer error messages */
+        const char *saved_text;
+
+        c = lex_get_save(lex, error);
+        while(isupper(c) || islower(c))
+            c = lex_get_save(lex, error);
+        lex_unget_unsave(lex, c);
+
+        saved_text = strbuffer_value(&lex->saved_text);
+
+        if(strcmp(saved_text, "true") == 0)
+            lex->token = TOKEN_TRUE;
+        else if(strcmp(saved_text, "false") == 0)
+            lex->token = TOKEN_FALSE;
+        else if(strcmp(saved_text, "null") == 0)
+            lex->token = TOKEN_NULL;
+        else
+            lex->token = TOKEN_INVALID;
+    }
+
+    else {
+        /* save the rest of the input UTF-8 sequence to get an error
+           message of valid UTF-8 */
+        lex_save_cached(lex);
+        lex->token = TOKEN_INVALID;
+    }
+
+out:
+    return lex->token;
+}
+
+static char *lex_steal_string(lex_t *lex)
+{
+    char *result = NULL;
+    if(lex->token == TOKEN_STRING)
+    {
+        result = lex->value.string;
+        lex->value.string = NULL;
+    }
+    return result;
+}
+
+static int lex_init(lex_t *lex, get_func get, eof_func eof, void *data)
+{
+    stream_init(&lex->stream, get, eof, data);
+    if(strbuffer_init(&lex->saved_text))
+        return -1;
+
+    lex->token = TOKEN_INVALID;
+    lex->line = 1;
+
+    return 0;
+}
+
+static void lex_close(lex_t *lex)
+{
+    if(lex->token == TOKEN_STRING)
+        free(lex->value.string);
+    strbuffer_close(&lex->saved_text);
+}
+
+
+/*** parser ***/
+
+static json_t *parse_value(lex_t *lex, json_error_t *error);
+
+static json_t *parse_object(lex_t *lex, json_error_t *error)
+{
+    json_t *object = json_object();
+    if(!object)
+        return NULL;
+
+    lex_scan(lex, error);
+    if(lex->token == '}')
+        return object;
+
+    while(1) {
+        char *key;
+        json_t *value;
+
+        if(lex->token != TOKEN_STRING) {
+            error_set(error, lex, "string or '}' expected");
+            goto error;
+        }
+
+        key = lex_steal_string(lex);
+        if(!key)
+            return NULL;
+
+        lex_scan(lex, error);
+        if(lex->token != ':') {
+            free(key);
+            error_set(error, lex, "':' expected");
+            goto error;
+        }
+
+        lex_scan(lex, error);
+        value = parse_value(lex, error);
+        if(!value) {
+            free(key);
+            goto error;
+        }
+
+        if(json_object_set_nocheck(object, key, value)) {
+            free(key);
+            json_decref(value);
+            goto error;
+        }
+
+        json_decref(value);
+        free(key);
+
+        lex_scan(lex, error);
+        if(lex->token != ',')
+            break;
+
+        lex_scan(lex, error);
+    }
+
+    if(lex->token != '}') {
+        error_set(error, lex, "'}' expected");
+        goto error;
+    }
+
+    return object;
+
+error:
+    json_decref(object);
+    return NULL;
+}
+
+static json_t *parse_array(lex_t *lex, json_error_t *error)
+{
+    json_t *array = json_array();
+    if(!array)
+        return NULL;
+
+    lex_scan(lex, error);
+    if(lex->token == ']')
+        return array;
+
+    while(lex->token) {
+        json_t *elem = parse_value(lex, error);
+        if(!elem)
+            goto error;
+
+        if(json_array_append(array, elem)) {
+            json_decref(elem);
+            goto error;
+        }
+        json_decref(elem);
+
+        lex_scan(lex, error);
+        if(lex->token != ',')
+            break;
+
+        lex_scan(lex, error);
+    }
+
+    if(lex->token != ']') {
+        error_set(error, lex, "']' expected");
+        goto error;
+    }
+
+    return array;
+
+error:
+    json_decref(array);
+    return NULL;
+}
+
+static json_t *parse_value(lex_t *lex, json_error_t *error)
+{
+    json_t *json;
+
+    switch(lex->token) {
+        case TOKEN_STRING: {
+            json = json_string_nocheck(lex->value.string);
+            break;
+        }
+
+        case TOKEN_INTEGER: {
+            json = json_integer(lex->value.integer);
+            break;
+        }
+
+        case TOKEN_REAL: {
+            json = json_real(lex->value.real);
+            break;
+        }
+
+        case TOKEN_TRUE:
+            json = json_true();
+            break;
+
+        case TOKEN_FALSE:
+            json = json_false();
+            break;
+
+        case TOKEN_NULL:
+            json = json_null();
+            break;
+
+        case '{':
+            json = parse_object(lex, error);
+            break;
+
+        case '[':
+            json = parse_array(lex, error);
+            break;
+
+        case TOKEN_INVALID:
+            error_set(error, lex, "invalid token");
+            return NULL;
+
+        default:
+            error_set(error, lex, "unexpected token");
+            return NULL;
+    }
+
+    if(!json)
+        return NULL;
+
+    return json;
+}
+
+static json_t *parse_json(lex_t *lex, json_error_t *error)
+{
+    error_init(error);
+
+    lex_scan(lex, error);
+    if(lex->token != '[' && lex->token != '{') {
+        error_set(error, lex, "'[' or '{' expected");
+        return NULL;
+    }
+
+    return parse_value(lex, error);
+}
+
+typedef struct
+{
+    const char *data;
+    int pos;
+} string_data_t;
+
+static int string_get(void *data)
+{
+    char c;
+    string_data_t *stream = (string_data_t *)data;
+    c = stream->data[stream->pos];
+    if(c == '\0')
+        return EOF;
+    else
+    {
+        stream->pos++;
+        return c;
+    }
+}
+
+static int string_eof(void *data)
+{
+    string_data_t *stream = (string_data_t *)data;
+    return (stream->data[stream->pos] == '\0');
+}
+
+json_t *json_loads(const char *string, json_error_t *error)
+{
+    lex_t lex;
+    json_t *result;
+
+    string_data_t stream_data = {
+        .data = string,
+        .pos = 0
+    };
+
+    if(lex_init(&lex, string_get, string_eof, (void *)&stream_data))
+        return NULL;
+
+    result = parse_json(&lex, error);
+    if(!result)
+        goto out;
+
+    lex_scan(&lex, error);
+    if(lex.token != TOKEN_EOF) {
+        error_set(error, &lex, "end of file expected");
+        json_decref(result);
+        result = NULL;
+    }
+
+out:
+    lex_close(&lex);
+    return result;
+}
+
+json_t *json_loadf(FILE *input, json_error_t *error)
+{
+    lex_t lex;
+    json_t *result;
+
+    if(lex_init(&lex, (get_func)fgetc, (eof_func)feof, input))
+        return NULL;
+
+    result = parse_json(&lex, error);
+    if(!result)
+        goto out;
+
+    lex_scan(&lex, error);
+    if(lex.token != TOKEN_EOF) {
+        error_set(error, &lex, "end of file expected");
+        json_decref(result);
+        result = NULL;
+    }
+
+out:
+    lex_close(&lex);
+    return result;
+}
+
+json_t *json_load_file(const char *path, json_error_t *error)
+{
+    json_t *result;
+    FILE *fp;
+
+    error_init(error);
+
+    fp = fopen(path, "r");
+    if(!fp)
+    {
+        error_set(error, NULL, "unable to open %s: %s",
+                  path, strerror(errno));
+        return NULL;
+    }
+
+    result = json_loadf(fp, error);
+
+    fclose(fp);
+    return result;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/strbuffer.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/strbuffer.c
new file mode 100644
index 0000000..3496024
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/strbuffer.c
@@ -0,0 +1,95 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#define _GNU_SOURCE
+#include <stdlib.h>
+#include <string.h>
+#include "strbuffer.h"
+#include "util.h"
+
+#define STRBUFFER_MIN_SIZE  16
+#define STRBUFFER_FACTOR    2
+
+int strbuffer_init(strbuffer_t *strbuff)
+{
+    strbuff->size = STRBUFFER_MIN_SIZE;
+    strbuff->length = 0;
+
+    strbuff->value = malloc(strbuff->size);
+    if(!strbuff->value)
+        return -1;
+
+    /* initialize to empty */
+    strbuff->value[0] = '\0';
+    return 0;
+}
+
+void strbuffer_close(strbuffer_t *strbuff)
+{
+    free(strbuff->value);
+    strbuff->size = 0;
+    strbuff->length = 0;
+    strbuff->value = NULL;
+}
+
+void strbuffer_clear(strbuffer_t *strbuff)
+{
+    strbuff->length = 0;
+    strbuff->value[0] = '\0';
+}
+
+const char *strbuffer_value(const strbuffer_t *strbuff)
+{
+    return strbuff->value;
+}
+
+char *strbuffer_steal_value(strbuffer_t *strbuff)
+{
+    char *result = strbuff->value;
+    strbuffer_init(strbuff);
+    return result;
+}
+
+int strbuffer_append(strbuffer_t *strbuff, const char *string)
+{
+    return strbuffer_append_bytes(strbuff, string, strlen(string));
+}
+
+int strbuffer_append_byte(strbuffer_t *strbuff, char byte)
+{
+    return strbuffer_append_bytes(strbuff, &byte, 1);
+}
+
+int strbuffer_append_bytes(strbuffer_t *strbuff, const char *data, int size)
+{
+    if(strbuff->length + size >= strbuff->size)
+    {
+        strbuff->size = max(strbuff->size * STRBUFFER_FACTOR,
+                            strbuff->length + size + 1);
+
+        strbuff->value = realloc(strbuff->value, strbuff->size);
+        if(!strbuff->value)
+            return -1;
+    }
+
+    memcpy(strbuff->value + strbuff->length, data, size);
+    strbuff->length += size;
+    strbuff->value[strbuff->length] = '\0';
+
+    return 0;
+}
+
+char strbuffer_pop(strbuffer_t *strbuff)
+{
+    if(strbuff->length > 0) {
+        char c = strbuff->value[--strbuff->length];
+        strbuff->value[strbuff->length] = '\0';
+        return c;
+    }
+    else
+        return '\0';
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/strbuffer.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/strbuffer.h
new file mode 100644
index 0000000..f4c5f77
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/strbuffer.h
@@ -0,0 +1,31 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef STRBUFFER_H
+#define STRBUFFER_H
+
+typedef struct {
+    char *value;
+    int length;   /* bytes used */
+    int size;     /* bytes allocated */
+} strbuffer_t;
+
+int strbuffer_init(strbuffer_t *strbuff);
+void strbuffer_close(strbuffer_t *strbuff);
+
+void strbuffer_clear(strbuffer_t *strbuff);
+
+const char *strbuffer_value(const strbuffer_t *strbuff);
+char *strbuffer_steal_value(strbuffer_t *strbuff);
+
+int strbuffer_append(strbuffer_t *strbuff, const char *string);
+int strbuffer_append_byte(strbuffer_t *strbuff, char byte);
+int strbuffer_append_bytes(strbuffer_t *strbuff, const char *data, int size);
+
+char strbuffer_pop(strbuffer_t *strbuff);
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/utf.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/utf.c
new file mode 100644
index 0000000..92484d0
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/utf.c
@@ -0,0 +1,190 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#include <string.h>
+#include "utf.h"
+
+int utf8_encode(int32_t codepoint, char *buffer, int *size)
+{
+    if(codepoint < 0)
+        return -1;
+    else if(codepoint < 0x80)
+    {
+        buffer[0] = (char)codepoint;
+        *size = 1;
+    }
+    else if(codepoint < 0x800)
+    {
+        buffer[0] = 0xC0 + ((codepoint & 0x7C0) >> 6);
+        buffer[1] = 0x80 + ((codepoint & 0x03F));
+        *size = 2;
+    }
+    else if(codepoint < 0x10000)
+    {
+        buffer[0] = 0xE0 + ((codepoint & 0xF000) >> 12);
+        buffer[1] = 0x80 + ((codepoint & 0x0FC0) >> 6);
+        buffer[2] = 0x80 + ((codepoint & 0x003F));
+        *size = 3;
+    }
+    else if(codepoint <= 0x10FFFF)
+    {
+        buffer[0] = 0xF0 + ((codepoint & 0x1C0000) >> 18);
+        buffer[1] = 0x80 + ((codepoint & 0x03F000) >> 12);
+        buffer[2] = 0x80 + ((codepoint & 0x000FC0) >> 6);
+        buffer[3] = 0x80 + ((codepoint & 0x00003F));
+        *size = 4;
+    }
+    else
+        return -1;
+
+    return 0;
+}
+
+int utf8_check_first(char byte)
+{
+    unsigned char u = (unsigned char)byte;
+
+    if(u < 0x80)
+        return 1;
+
+    if(0x80 <= u && u <= 0xBF) {
+        /* second, third or fourth byte of a multi-byte
+           sequence, i.e. a "continuation byte" */
+        return 0;
+    }
+    else if(u == 0xC0 || u == 0xC1) {
+        /* overlong encoding of an ASCII byte */
+        return 0;
+    }
+    else if(0xC2 <= u && u <= 0xDF) {
+        /* 2-byte sequence */
+        return 2;
+    }
+
+    else if(0xE0 <= u && u <= 0xEF) {
+        /* 3-byte sequence */
+        return 3;
+    }
+    else if(0xF0 <= u && u <= 0xF4) {
+        /* 4-byte sequence */
+        return 4;
+    }
+    else { /* u >= 0xF5 */
+        /* Restricted (start of 4-, 5- or 6-byte sequence) or invalid
+           UTF-8 */
+        return 0;
+    }
+}
+
+int utf8_check_full(const char *buffer, int size, int32_t *codepoint)
+{
+    int i;
+    int32_t value = 0;
+    unsigned char u = (unsigned char)buffer[0];
+
+    if(size == 2)
+    {
+        value = u & 0x1F;
+    }
+    else if(size == 3)
+    {
+        value = u & 0xF;
+    }
+    else if(size == 4)
+    {
+        value = u & 0x7;
+    }
+    else
+        return 0;
+
+    for(i = 1; i < size; i++)
+    {
+        u = (unsigned char)buffer[i];
+
+        if(u < 0x80 || u > 0xBF) {
+            /* not a continuation byte */
+            return 0;
+        }
+
+        value = (value << 6) + (u & 0x3F);
+    }
+
+    if(value > 0x10FFFF) {
+        /* not in Unicode range */
+        return 0;
+    }
+
+    else if(0xD800 <= value && value <= 0xDFFF) {
+        /* invalid code point (UTF-16 surrogate halves) */
+        return 0;
+    }
+
+    else if((size == 2 && value < 0x80) ||
+            (size == 3 && value < 0x800) ||
+            (size == 4 && value < 0x10000)) {
+        /* overlong encoding */
+        return 0;
+    }
+
+    if(codepoint)
+        *codepoint = value;
+
+    return 1;
+}
+
+const char *utf8_iterate(const char *buffer, int32_t *codepoint)
+{
+    int count;
+    int32_t value;
+
+    if(!*buffer)
+        return buffer;
+
+    count = utf8_check_first(buffer[0]);
+    if(count <= 0)
+        return NULL;
+
+    if(count == 1)
+        value = (unsigned char)buffer[0];
+    else
+    {
+        if(!utf8_check_full(buffer, count, &value))
+            return NULL;
+    }
+
+    if(codepoint)
+        *codepoint = value;
+
+    return buffer + count;
+}
+
+int utf8_check_string(const char *string, int length)
+{
+    int i;
+
+    if(length == -1)
+        length = strlen(string);
+
+    for(i = 0; i < length; i++)
+    {
+        int count = utf8_check_first(string[i]);
+        if(count == 0)
+            return 0;
+        else if(count > 1)
+        {
+            if(i + count > length)
+                return 0;
+
+            if(!utf8_check_full(&string[i], count, NULL))
+                return 0;
+
+            i += count - 1;
+        }
+    }
+
+    return 1;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/utf.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/utf.h
new file mode 100644
index 0000000..d0ae6e9
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/utf.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef UTF_H
+#define UTF_H
+
+#include <config.h>
+
+#ifdef HAVE_INTTYPES_H
+/* inttypes.h includes stdint.h in a standard environment, so there's
+no need to include stdint.h separately. If inttypes.h doesn't define
+int32_t, it's defined in config.h. */
+#include <inttypes.h>
+#endif
+
+int utf8_encode(int codepoint, char *buffer, int *size);
+
+int utf8_check_first(char byte);
+int utf8_check_full(const char *buffer, int size, int32_t *codepoint);
+const char *utf8_iterate(const char *buffer, int32_t *codepoint);
+
+int utf8_check_string(const char *string, int length);
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/util.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/util.h
new file mode 100644
index 0000000..06a547b
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/util.h
@@ -0,0 +1,13 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef UTIL_H
+#define UTIL_H
+
+#define max(a, b)  ((a) > (b) ? (a) : (b))
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/value.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/value.c
new file mode 100644
index 0000000..e024fdb
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/compat/jansson/value.c
@@ -0,0 +1,976 @@
+/*
+ * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#define _GNU_SOURCE
+
+#include <config.h>
+
+#include <stdlib.h>
+#include <string.h>
+
+#include <jansson.h>
+#include "hashtable.h"
+#include "jansson_private.h"
+#include "utf.h"
+#include "util.h"
+
+
+static inline void json_init(json_t *json, json_type type)
+{
+    json->type = type;
+    json->refcount = 1;
+}
+
+
+/*** object ***/
+
+/* This macro just returns a pointer that's a few bytes backwards from
+   string. This makes it possible to pass a pointer to object_key_t
+   when only the string inside it is used, without actually creating
+   an object_key_t instance. */
+#define string_to_key(string)  container_of(string, object_key_t, key)
+
+static unsigned int hash_key(const void *ptr)
+{
+    const char *str = ((const object_key_t *)ptr)->key;
+
+    unsigned int hash = 5381;
+    unsigned int c;
+
+    while((c = (unsigned int)*str))
+    {
+        hash = ((hash << 5) + hash) + c;
+        str++;
+    }
+
+    return hash;
+}
+
+static int key_equal(const void *ptr1, const void *ptr2)
+{
+    return strcmp(((const object_key_t *)ptr1)->key,
+                  ((const object_key_t *)ptr2)->key) == 0;
+}
+
+static void value_decref(void *value)
+{
+    json_decref((json_t *)value);
+}
+
+json_t *json_object(void)
+{
+    json_object_t *object = malloc(sizeof(json_object_t));
+    if(!object)
+        return NULL;
+    json_init(&object->json, JSON_OBJECT);
+
+    if(hashtable_init(&object->hashtable, hash_key, key_equal,
+                      free, value_decref))
+    {
+        free(object);
+        return NULL;
+    }
+
+    object->serial = 0;
+    object->visited = 0;
+
+    return &object->json;
+}
+
+static void json_delete_object(json_object_t *object)
+{
+    hashtable_close(&object->hashtable);
+    free(object);
+}
+
+unsigned int json_object_size(const json_t *json)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json))
+        return -1;
+
+    object = json_to_object(json);
+    return object->hashtable.size;
+}
+
+json_t *json_object_get(const json_t *json, const char *key)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json))
+        return NULL;
+
+    object = json_to_object(json);
+    return hashtable_get(&object->hashtable, string_to_key(key));
+}
+
+int json_object_set_new_nocheck(json_t *json, const char *key, json_t *value)
+{
+    json_object_t *object;
+    object_key_t *k;
+
+    if(!key || !value)
+        return -1;
+
+    if(!json_is_object(json) || json == value)
+    {
+        json_decref(value);
+        return -1;
+    }
+    object = json_to_object(json);
+
+    k = malloc(sizeof(object_key_t) + strlen(key) + 1);
+    if(!k)
+        return -1;
+
+    k->serial = object->serial++;
+    strcpy(k->key, key);
+
+    if(hashtable_set(&object->hashtable, k, value))
+    {
+        json_decref(value);
+        return -1;
+    }
+
+    return 0;
+}
+
+int json_object_set_new(json_t *json, const char *key, json_t *value)
+{
+    if(!key || !utf8_check_string(key, -1))
+    {
+        json_decref(value);
+        return -1;
+    }
+
+    return json_object_set_new_nocheck(json, key, value);
+}
+
+int json_object_del(json_t *json, const char *key)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json))
+        return -1;
+
+    object = json_to_object(json);
+    return hashtable_del(&object->hashtable, string_to_key(key));
+}
+
+int json_object_clear(json_t *json)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json))
+        return -1;
+
+    object = json_to_object(json);
+    hashtable_clear(&object->hashtable);
+
+    return 0;
+}
+
+int json_object_update(json_t *object, json_t *other)
+{
+    void *iter;
+
+    if(!json_is_object(object) || !json_is_object(other))
+        return -1;
+
+    iter = json_object_iter(other);
+    while(iter) {
+        const char *key;
+        json_t *value;
+
+        key = json_object_iter_key(iter);
+        value = json_object_iter_value(iter);
+
+        if(json_object_set_nocheck(object, key, value))
+            return -1;
+
+        iter = json_object_iter_next(other, iter);
+    }
+
+    return 0;
+}
+
+void *json_object_iter(json_t *json)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json))
+        return NULL;
+
+    object = json_to_object(json);
+    return hashtable_iter(&object->hashtable);
+}
+
+void *json_object_iter_at(json_t *json, const char *key)
+{
+    json_object_t *object;
+
+    if(!key || !json_is_object(json))
+        return NULL;
+
+    object = json_to_object(json);
+    return hashtable_iter_at(&object->hashtable, string_to_key(key));
+}
+
+void *json_object_iter_next(json_t *json, void *iter)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json) || iter == NULL)
+        return NULL;
+
+    object = json_to_object(json);
+    return hashtable_iter_next(&object->hashtable, iter);
+}
+
+const object_key_t *jsonp_object_iter_fullkey(void *iter)
+{
+    if(!iter)
+        return NULL;
+
+    return hashtable_iter_key(iter);
+}
+
+const char *json_object_iter_key(void *iter)
+{
+    if(!iter)
+        return NULL;
+
+    return jsonp_object_iter_fullkey(iter)->key;
+}
+
+json_t *json_object_iter_value(void *iter)
+{
+    if(!iter)
+        return NULL;
+
+    return (json_t *)hashtable_iter_value(iter);
+}
+
+int json_object_iter_set_new(json_t *json, void *iter, json_t *value)
+{
+    json_object_t *object;
+
+    if(!json_is_object(json) || !iter || !value)
+        return -1;
+
+    object = json_to_object(json);
+    hashtable_iter_set(&object->hashtable, iter, value);
+
+    return 0;
+}
+
+static int json_object_equal(json_t *object1, json_t *object2)
+{
+    void *iter;
+
+    if(json_object_size(object1) != json_object_size(object2))
+        return 0;
+
+    iter = json_object_iter(object1);
+    while(iter)
+    {
+        const char *key;
+        json_t *value1, *value2;
+
+        key = json_object_iter_key(iter);
+        value1 = json_object_iter_value(iter);
+        value2 = json_object_get(object2, key);
+
+        if(!json_equal(value1, value2))
+            return 0;
+
+        iter = json_object_iter_next(object1, iter);
+    }
+
+    return 1;
+}
+
+static json_t *json_object_copy(json_t *object)
+{
+    json_t *result;
+    void *iter;
+
+    result = json_object();
+    if(!result)
+        return NULL;
+
+    iter = json_object_iter(object);
+    while(iter)
+    {
+        const char *key;
+        json_t *value;
+
+        key = json_object_iter_key(iter);
+        value = json_object_iter_value(iter);
+        json_object_set_nocheck(result, key, value);
+
+        iter = json_object_iter_next(object, iter);
+    }
+
+    return result;
+}
+
+static json_t *json_object_deep_copy(json_t *object)
+{
+    json_t *result;
+    void *iter;
+
+    result = json_object();
+    if(!result)
+        return NULL;
+
+    iter = json_object_iter(object);
+    while(iter)
+    {
+        const char *key;
+        json_t *value;
+
+        key = json_object_iter_key(iter);
+        value = json_object_iter_value(iter);
+        json_object_set_new_nocheck(result, key, json_deep_copy(value));
+
+        iter = json_object_iter_next(object, iter);
+    }
+
+    return result;
+}
+
+
+/*** array ***/
+
+json_t *json_array(void)
+{
+    json_array_t *array = malloc(sizeof(json_array_t));
+    if(!array)
+        return NULL;
+    json_init(&array->json, JSON_ARRAY);
+
+    array->entries = 0;
+    array->size = 8;
+
+    array->table = malloc(array->size * sizeof(json_t *));
+    if(!array->table) {
+        free(array);
+        return NULL;
+    }
+
+    array->visited = 0;
+
+    return &array->json;
+}
+
+static void json_delete_array(json_array_t *array)
+{
+    unsigned int i;
+
+    for(i = 0; i < array->entries; i++)
+        json_decref(array->table[i]);
+
+    free(array->table);
+    free(array);
+}
+
+unsigned int json_array_size(const json_t *json)
+{
+    if(!json_is_array(json))
+        return 0;
+
+    return json_to_array(json)->entries;
+}
+
+json_t *json_array_get(const json_t *json, unsigned int index)
+{
+    json_array_t *array;
+    if(!json_is_array(json))
+        return NULL;
+    array = json_to_array(json);
+
+    if(index >= array->entries)
+        return NULL;
+
+    return array->table[index];
+}
+
+int json_array_set_new(json_t *json, unsigned int index, json_t *value)
+{
+    json_array_t *array;
+
+    if(!value)
+        return -1;
+
+    if(!json_is_array(json) || json == value)
+    {
+        json_decref(value);
+        return -1;
+    }
+    array = json_to_array(json);
+
+    if(index >= array->entries)
+    {
+        json_decref(value);
+        return -1;
+    }
+
+    json_decref(array->table[index]);
+    array->table[index] = value;
+
+    return 0;
+}
+
+static void array_move(json_array_t *array, unsigned int dest,
+                       unsigned int src, unsigned int count)
+{
+    memmove(&array->table[dest], &array->table[src], count * sizeof(json_t *));
+}
+
+static void array_copy(json_t **dest, unsigned int dpos,
+                       json_t **src, unsigned int spos,
+                       unsigned int count)
+{
+    memcpy(&dest[dpos], &src[spos], count * sizeof(json_t *));
+}
+
+static json_t **json_array_grow(json_array_t *array,
+                                unsigned int amount,
+                                int copy)
+{
+    unsigned int new_size;
+    json_t **old_table, **new_table;
+
+    if(array->entries + amount <= array->size)
+        return array->table;
+
+    old_table = array->table;
+
+    new_size = max(array->size + amount, array->size * 2);
+    new_table = malloc(new_size * sizeof(json_t *));
+    if(!new_table)
+        return NULL;
+
+    array->size = new_size;
+    array->table = new_table;
+
+    if(copy) {
+        array_copy(array->table, 0, old_table, 0, array->entries);
+        free(old_table);
+        return array->table;
+    }
+
+    return old_table;
+}
+
+int json_array_append_new(json_t *json, json_t *value)
+{
+    json_array_t *array;
+
+    if(!value)
+        return -1;
+
+    if(!json_is_array(json) || json == value)
+    {
+        json_decref(value);
+        return -1;
+    }
+    array = json_to_array(json);
+
+    if(!json_array_grow(array, 1, 1)) {
+        json_decref(value);
+        return -1;
+    }
+
+    array->table[array->entries] = value;
+    array->entries++;
+
+    return 0;
+}
+
+int json_array_insert_new(json_t *json, unsigned int index, json_t *value)
+{
+    json_array_t *array;
+    json_t **old_table;
+
+    if(!value)
+        return -1;
+
+    if(!json_is_array(json) || json == value) {
+        json_decref(value);
+        return -1;
+    }
+    array = json_to_array(json);
+
+    if(index > array->entries) {
+        json_decref(value);
+        return -1;
+    }
+
+    old_table = json_array_grow(array, 1, 0);
+    if(!old_table) {
+        json_decref(value);
+        return -1;
+    }
+
+    if(old_table != array->table) {
+        array_copy(array->table, 0, old_table, 0, index);
+        array_copy(array->table, index + 1, old_table, index,
+                   array->entries - index);
+        free(old_table);
+    }
+    else
+        array_move(array, index + 1, index, array->entries - index);
+
+    array->table[index] = value;
+    array->entries++;
+
+    return 0;
+}
+
+int json_array_remove(json_t *json, unsigned int index)
+{
+    json_array_t *array;
+
+    if(!json_is_array(json))
+        return -1;
+    array = json_to_array(json);
+
+    if(index >= array->entries)
+        return -1;
+
+    json_decref(array->table[index]);
+
+    array_move(array, index, index + 1, array->entries - index);
+    array->entries--;
+
+    return 0;
+}
+
+int json_array_clear(json_t *json)
+{
+    json_array_t *array;
+    unsigned int i;
+
+    if(!json_is_array(json))
+        return -1;
+    array = json_to_array(json);
+
+    for(i = 0; i < array->entries; i++)
+        json_decref(array->table[i]);
+
+    array->entries = 0;
+    return 0;
+}
+
+int json_array_extend(json_t *json, json_t *other_json)
+{
+    json_array_t *array, *other;
+    unsigned int i;
+
+    if(!json_is_array(json) || !json_is_array(other_json))
+        return -1;
+    array = json_to_array(json);
+    other = json_to_array(other_json);
+
+    if(!json_array_grow(array, other->entries, 1))
+        return -1;
+
+    for(i = 0; i < other->entries; i++)
+        json_incref(other->table[i]);
+
+    array_copy(array->table, array->entries, other->table, 0, other->entries);
+
+    array->entries += other->entries;
+    return 0;
+}
+
+static int json_array_equal(json_t *array1, json_t *array2)
+{
+    unsigned int i, size;
+
+    size = json_array_size(array1);
+    if(size != json_array_size(array2))
+        return 0;
+
+    for(i = 0; i < size; i++)
+    {
+        json_t *value1, *value2;
+
+        value1 = json_array_get(array1, i);
+        value2 = json_array_get(array2, i);
+
+        if(!json_equal(value1, value2))
+            return 0;
+    }
+
+    return 1;
+}
+
+static json_t *json_array_copy(json_t *array)
+{
+    json_t *result;
+    unsigned int i;
+
+    result = json_array();
+    if(!result)
+        return NULL;
+
+    for(i = 0; i < json_array_size(array); i++)
+        json_array_append(result, json_array_get(array, i));
+
+    return result;
+}
+
+static json_t *json_array_deep_copy(json_t *array)
+{
+    json_t *result;
+    unsigned int i;
+
+    result = json_array();
+    if(!result)
+        return NULL;
+
+    for(i = 0; i < json_array_size(array); i++)
+        json_array_append_new(result, json_deep_copy(json_array_get(array, i)));
+
+    return result;
+}
+
+/*** string ***/
+
+json_t *json_string_nocheck(const char *value)
+{
+    json_string_t *string;
+
+    if(!value)
+        return NULL;
+
+    string = malloc(sizeof(json_string_t));
+    if(!string)
+        return NULL;
+    json_init(&string->json, JSON_STRING);
+
+    string->value = strdup(value);
+    if(!string->value) {
+        free(string);
+        return NULL;
+    }
+
+    return &string->json;
+}
+
+json_t *json_string(const char *value)
+{
+    if(!value || !utf8_check_string(value, -1))
+        return NULL;
+
+    return json_string_nocheck(value);
+}
+
+const char *json_string_value(const json_t *json)
+{
+    if(!json_is_string(json))
+        return NULL;
+
+    return json_to_string(json)->value;
+}
+
+int json_string_set_nocheck(json_t *json, const char *value)
+{
+    char *dup;
+    json_string_t *string;
+
+    dup = strdup(value);
+    if(!dup)
+        return -1;
+
+    string = json_to_string(json);
+    free(string->value);
+    string->value = dup;
+
+    return 0;
+}
+
+int json_string_set(json_t *json, const char *value)
+{
+    if(!value || !utf8_check_string(value, -1))
+        return -1;
+
+    return json_string_set_nocheck(json, value);
+}
+
+static void json_delete_string(json_string_t *string)
+{
+    free(string->value);
+    free(string);
+}
+
+static int json_string_equal(json_t *string1, json_t *string2)
+{
+    return strcmp(json_string_value(string1), json_string_value(string2)) == 0;
+}
+
+static json_t *json_string_copy(json_t *string)
+{
+    return json_string_nocheck(json_string_value(string));
+}
+
+
+/*** integer ***/
+
+json_t *json_integer(int value)
+{
+    json_integer_t *integer = malloc(sizeof(json_integer_t));
+    if(!integer)
+        return NULL;
+    json_init(&integer->json, JSON_INTEGER);
+
+    integer->value = value;
+    return &integer->json;
+}
+
+int json_integer_value(const json_t *json)
+{
+    if(!json_is_integer(json))
+        return 0;
+
+    return json_to_integer(json)->value;
+}
+
+int json_integer_set(json_t *json, int value)
+{
+    if(!json_is_integer(json))
+        return -1;
+
+    json_to_integer(json)->value = value;
+
+    return 0;
+}
+
+static void json_delete_integer(json_integer_t *integer)
+{
+    free(integer);
+}
+
+static int json_integer_equal(json_t *integer1, json_t *integer2)
+{
+    return json_integer_value(integer1) == json_integer_value(integer2);
+}
+
+static json_t *json_integer_copy(json_t *integer)
+{
+    return json_integer(json_integer_value(integer));
+}
+
+
+/*** real ***/
+
+json_t *json_real(double value)
+{
+    json_real_t *real = malloc(sizeof(json_real_t));
+    if(!real)
+        return NULL;
+    json_init(&real->json, JSON_REAL);
+
+    real->value = value;
+    return &real->json;
+}
+
+double json_real_value(const json_t *json)
+{
+    if(!json_is_real(json))
+        return 0;
+
+    return json_to_real(json)->value;
+}
+
+int json_real_set(json_t *json, double value)
+{
+    if(!json_is_real(json))
+        return 0;
+
+    json_to_real(json)->value = value;
+
+    return 0;
+}
+
+static void json_delete_real(json_real_t *real)
+{
+    free(real);
+}
+
+static int json_real_equal(json_t *real1, json_t *real2)
+{
+    return json_real_value(real1) == json_real_value(real2);
+}
+
+static json_t *json_real_copy(json_t *real)
+{
+    return json_real(json_real_value(real));
+}
+
+
+/*** number ***/
+
+double json_number_value(const json_t *json)
+{
+    if(json_is_integer(json))
+        return json_integer_value(json);
+    else if(json_is_real(json))
+        return json_real_value(json);
+    else
+        return 0.0;
+}
+
+
+/*** simple values ***/
+
+json_t *json_true(void)
+{
+    static json_t the_true = {
+        .type = JSON_TRUE,
+        .refcount = (unsigned int)-1
+    };
+    return &the_true;
+}
+
+
+json_t *json_false(void)
+{
+    static json_t the_false = {
+        .type = JSON_FALSE,
+        .refcount = (unsigned int)-1
+    };
+    return &the_false;
+}
+
+
+json_t *json_null(void)
+{
+    static json_t the_null = {
+        .type = JSON_NULL,
+        .refcount = (unsigned int)-1
+    };
+    return &the_null;
+}
+
+
+/*** deletion ***/
+
+void json_delete(json_t *json)
+{
+    if(json_is_object(json))
+        json_delete_object(json_to_object(json));
+
+    else if(json_is_array(json))
+        json_delete_array(json_to_array(json));
+
+    else if(json_is_string(json))
+        json_delete_string(json_to_string(json));
+
+    else if(json_is_integer(json))
+        json_delete_integer(json_to_integer(json));
+
+    else if(json_is_real(json))
+        json_delete_real(json_to_real(json));
+
+    /* json_delete is not called for true, false or null */
+}
+
+
+/*** equality ***/
+
+int json_equal(json_t *json1, json_t *json2)
+{
+    if(!json1 || !json2)
+        return 0;
+
+    if(json_typeof(json1) != json_typeof(json2))
+        return 0;
+
+    /* this covers true, false and null as they are singletons */
+    if(json1 == json2)
+        return 1;
+
+    if(json_is_object(json1))
+        return json_object_equal(json1, json2);
+
+    if(json_is_array(json1))
+        return json_array_equal(json1, json2);
+
+    if(json_is_string(json1))
+        return json_string_equal(json1, json2);
+
+    if(json_is_integer(json1))
+        return json_integer_equal(json1, json2);
+
+    if(json_is_real(json1))
+        return json_real_equal(json1, json2);
+
+    return 0;
+}
+
+
+/*** copying ***/
+
+json_t *json_copy(json_t *json)
+{
+    if(!json)
+        return NULL;
+
+    if(json_is_object(json))
+        return json_object_copy(json);
+
+    if(json_is_array(json))
+        return json_array_copy(json);
+
+    if(json_is_string(json))
+        return json_string_copy(json);
+
+    if(json_is_integer(json))
+        return json_integer_copy(json);
+
+    if(json_is_real(json))
+        return json_real_copy(json);
+
+    if(json_is_true(json) || json_is_false(json) || json_is_null(json))
+        return json;
+
+    return NULL;
+}
+
+json_t *json_deep_copy(json_t *json)
+{
+    if(!json)
+        return NULL;
+
+    if(json_is_object(json))
+        return json_object_deep_copy(json);
+
+    if(json_is_array(json))
+        return json_array_deep_copy(json);
+
+    /* for the rest of the types, deep copying doesn't differ from
+       shallow copying */
+
+    if(json_is_string(json))
+        return json_string_copy(json);
+
+    if(json_is_integer(json))
+        return json_integer_copy(json);
+
+    if(json_is_real(json))
+        return json_real_copy(json);
+
+    if(json_is_true(json) || json_is_false(json) || json_is_null(json))
+        return json;
+
+    return NULL;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/configure.ac b/bitzenymininglibrary/src/main/cpp/libcpuminer/configure.ac
new file mode 100644
index 0000000..38bb465
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/configure.ac
@@ -0,0 +1,136 @@
+AC_INIT([cpuminer], [2.5.0])
+
+AC_PREREQ([2.59c])
+AC_CANONICAL_SYSTEM
+AC_CONFIG_SRCDIR([cpu-miner.c])
+AM_INIT_AUTOMAKE([gnu])
+AC_CONFIG_HEADERS([cpuminer-config.h])
+
+dnl Make sure anyone changing configure.ac/Makefile.am has a clue
+AM_MAINTAINER_MODE
+
+dnl Checks for programs
+AC_PROG_CC_C99
+AC_PROG_GCC_TRADITIONAL
+AM_PROG_CC_C_O
+AM_PROG_AS
+AC_PROG_RANLIB
+
+dnl Checks for header files
+AC_HEADER_STDC
+AC_CHECK_HEADERS([sys/endian.h sys/param.h syslog.h])
+# sys/sysctl.h requires sys/types.h on FreeBSD
+# sys/sysctl.h requires sys/param.h on OpenBSD
+AC_CHECK_HEADERS([sys/sysctl.h], [], [],
+[#include <sys/types.h>
+#ifdef HAVE_SYS_PARAM_H
+#include <sys/param.h>
+#endif
+])
+
+AC_CHECK_DECLS([be32dec, le32dec, be32enc, le32enc], [], [],
+[AC_INCLUDES_DEFAULT
+#ifdef HAVE_SYS_ENDIAN_H
+#include <sys/endian.h>
+#endif
+])
+
+AC_FUNC_ALLOCA
+AC_CHECK_FUNCS([getopt_long])
+
+case $target in
+  i*86-*-*)
+    have_x86=true
+    ;;
+  x86_64-*-*|amd64-*-*)
+    have_x86_64=true
+    ;;
+  arm*-*-*)
+    have_arm=true
+    ;;
+  powerpc*-*-*)
+    have_ppc=true
+    ;;
+esac
+
+PTHREAD_FLAGS="-pthread"
+WS2_LIBS=""
+
+case $target in
+  *-*-mingw*)
+    have_win32=true
+    PTHREAD_FLAGS=""
+    WS2_LIBS="-lws2_32"
+    ;;
+esac
+
+AC_ARG_ENABLE([assembly],
+  AS_HELP_STRING([--disable-assembly], [disable assembly-language routines]))
+if test x$enable_assembly != xno; then
+  AC_DEFINE([USE_ASM], [1], [Define to 1 if assembly routines are wanted.])
+fi
+
+if test x$enable_assembly != xno -a x$have_x86_64 = xtrue
+then
+  AC_MSG_CHECKING(whether we can compile AVX code)
+  AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,[asm ("vmovdqa %ymm0, %ymm1");])],
+    AC_DEFINE(USE_AVX, 1, [Define to 1 if AVX assembly is available.])
+    AC_MSG_RESULT(yes)
+    AC_MSG_CHECKING(whether we can compile XOP code)
+    AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,[asm ("vprotd \$7, %xmm0, %xmm1");])],
+      AC_DEFINE(USE_XOP, 1, [Define to 1 if XOP assembly is available.])
+      AC_MSG_RESULT(yes)
+    ,
+      AC_MSG_RESULT(no)
+      AC_MSG_WARN([The assembler does not support the XOP instruction set.])
+    )
+    AC_MSG_CHECKING(whether we can compile AVX2 code)
+    AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,[asm ("vpaddd %ymm0, %ymm1, %ymm2");])],
+      AC_DEFINE(USE_AVX2, 1, [Define to 1 if AVX2 assembly is available.])
+      AC_MSG_RESULT(yes)
+    ,
+      AC_MSG_RESULT(no)
+      AC_MSG_WARN([The assembler does not support the AVX2 instruction set.])
+    )
+  ,
+    AC_MSG_RESULT(no)
+    AC_MSG_WARN([The assembler does not support the AVX instruction set.])
+  )
+fi
+
+AC_CHECK_LIB(jansson, json_loads, request_jansson=false, request_jansson=true)
+AC_CHECK_LIB([pthread], [pthread_create], PTHREAD_LIBS="-lpthread",
+  AC_CHECK_LIB([pthreadGC2], [pthread_create], PTHREAD_LIBS="-lpthreadGC2",
+    AC_CHECK_LIB([pthreadGC1], [pthread_create], PTHREAD_LIBS="-lpthreadGC1",
+      AC_CHECK_LIB([pthreadGC], [pthread_create], PTHREAD_LIBS="-lpthreadGC"
+))))
+
+AM_CONDITIONAL([WANT_JANSSON], [test x$request_jansson = xtrue])
+AM_CONDITIONAL([HAVE_WINDOWS], [test x$have_win32 = xtrue])
+AM_CONDITIONAL([USE_ASM], [test x$enable_assembly != xno])
+AM_CONDITIONAL([ARCH_x86], [test x$have_x86 = xtrue])
+AM_CONDITIONAL([ARCH_x86_64], [test x$have_x86_64 = xtrue])
+AM_CONDITIONAL([ARCH_ARM], [test x$have_arm = xtrue])
+AM_CONDITIONAL([ARCH_PPC], [test x$have_ppc = xtrue])
+
+if test x$request_jansson = xtrue
+then
+	JANSSON_LIBS="compat/jansson/libjansson.a"
+else
+	JANSSON_LIBS=-ljansson
+fi
+
+LIBCURL_CHECK_CONFIG(, 7.15.2, ,
+  [AC_MSG_ERROR([Missing required libcurl >= 7.15.2])])
+
+AC_SUBST(JANSSON_LIBS)
+AC_SUBST(PTHREAD_FLAGS)
+AC_SUBST(PTHREAD_LIBS)
+AC_SUBST(WS2_LIBS)
+
+AC_CONFIG_FILES([
+	Makefile
+	compat/Makefile
+	compat/jansson/Makefile
+	])
+AC_OUTPUT
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/cpu-miner.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/cpu-miner.c
new file mode 100644
index 0000000..2909a75
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/cpu-miner.c
@@ -0,0 +1,2035 @@
+/*
+ * Copyright 2010 Jeff Garzik
+ * Copyright 2012-2017 pooler
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+#define _GNU_SOURCE
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdbool.h>
+#include <inttypes.h>
+#include <unistd.h>
+#include <sys/time.h>
+#include <time.h>
+#ifdef WIN32
+#include <windows.h>
+#else
+#include <errno.h>
+#include <signal.h>
+#include <sys/resource.h>
+#if HAVE_SYS_SYSCTL_H
+#include <sys/types.h>
+#if HAVE_SYS_PARAM_H
+#include <sys/param.h>
+#endif
+#include <sys/sysctl.h>
+#endif
+#endif
+#include <jansson.h>
+#include <curl/curl.h>
+#include "compat.h"
+#include "miner.h"
+
+#define PROGRAM_NAME		"minerd"
+#define LP_SCANTIME		60
+
+#ifdef __linux /* Linux specific policy and affinity management */
+#include <sched.h>
+static inline void drop_policy(void)
+{
+	struct sched_param param;
+	param.sched_priority = 0;
+
+#ifdef SCHED_IDLE
+	if (unlikely(sched_setscheduler(0, SCHED_IDLE, &param) == -1))
+#endif
+#ifdef SCHED_BATCH
+		sched_setscheduler(0, SCHED_BATCH, &param);
+#endif
+}
+
+static inline void affine_to_cpu(int id, int cpu)
+{
+	cpu_set_t set;
+
+	CPU_ZERO(&set);
+	CPU_SET(cpu, &set);
+	sched_setaffinity(0, sizeof(set), &set);
+}
+#elif defined(__FreeBSD__) /* FreeBSD specific policy and affinity management */
+#include <sys/cpuset.h>
+static inline void drop_policy(void)
+{
+}
+
+static inline void affine_to_cpu(int id, int cpu)
+{
+	cpuset_t set;
+	CPU_ZERO(&set);
+	CPU_SET(cpu, &set);
+	cpuset_setaffinity(CPU_LEVEL_WHICH, CPU_WHICH_TID, -1, sizeof(cpuset_t), &set);
+}
+#else
+static inline void drop_policy(void)
+{
+}
+
+static inline void affine_to_cpu(int id, int cpu)
+{
+}
+#endif
+
+enum workio_commands {
+	WC_GET_WORK,
+	WC_SUBMIT_WORK,
+};
+
+struct workio_cmd {
+	enum workio_commands	cmd;
+	struct thr_info		*thr;
+	union {
+		struct work	*work;
+	} u;
+};
+
+enum algos {
+	ALGO_YESCRYPT,
+	ALGO_YESPOWER,
+	ALGO_SCRYPT,		/* scrypt(1024,1,1) */
+	ALGO_SHA256D,		/* SHA-256d */
+};
+
+static const char *algo_names[] = {
+	[ALGO_YESCRYPT]		= "yescrypt",
+	[ALGO_YESPOWER]		= "yespower",
+	[ALGO_SCRYPT]		= "scrypt",
+	[ALGO_SHA256D]		= "sha256d",
+};
+
+bool opt_debug = false;
+bool opt_protocol = false;
+static bool opt_benchmark = false;
+bool opt_redirect = true;
+bool want_longpoll = true;
+bool have_longpoll = false;
+bool have_gbt = true;
+bool allow_getwork = true;
+bool want_stratum = true;
+bool have_stratum = false;
+bool use_syslog = false;
+static bool opt_background = false;
+static bool opt_quiet = false;
+static int opt_retries = -1;
+static int opt_fail_pause = 30;
+int opt_timeout = 0;
+static int opt_scantime = 5;
+static enum algos opt_algo = ALGO_YESCRYPT;
+static int opt_scrypt_n = 1024;
+static int opt_n_threads;
+static int num_processors;
+static char *rpc_url;
+static char *rpc_userpass;
+static char *rpc_user, *rpc_pass;
+static int pk_script_size;
+static unsigned char pk_script[25];
+static char coinbase_sig[101] = "";
+char *opt_cert;
+char *opt_proxy;
+long opt_proxy_type;
+struct thr_info *thr_info;
+static int work_thr_id;
+int longpoll_thr_id = -1;
+int stratum_thr_id = -1;
+struct work_restart *work_restart = NULL;
+static struct stratum_ctx stratum;
+
+pthread_mutex_t applog_lock;
+static pthread_mutex_t stats_lock;
+
+static unsigned long accepted_count = 0L;
+static unsigned long rejected_count = 0L;
+static double *thr_hashrates;
+
+#ifdef HAVE_GETOPT_LONG
+#include <getopt.h>
+#else
+struct option {
+	const char *name;
+	int has_arg;
+	int *flag;
+	int val;
+};
+#endif
+
+static char const usage[] = "\
+Usage: " PROGRAM_NAME " [OPTIONS]\n\
+Options:\n\
+  -a, --algo=ALGO       specify the algorithm to use\n\
+                          yescrypt  yescrypt (default)\n\
+                          yespower  yespower 0.5\n\
+                          scrypt    scrypt(1024, 1, 1)\n\
+                          scrypt:N  scrypt(N, 1, 1)\n\
+                          sha256d   SHA-256d\n\
+  -o, --url=URL         URL of mining server\n\
+  -O, --userpass=U:P    username:password pair for mining server\n\
+  -u, --user=USERNAME   username for mining server\n\
+  -p, --pass=PASSWORD   password for mining server\n\
+      --cert=FILE       certificate for mining server using SSL\n\
+  -x, --proxy=[PROTOCOL://]HOST[:PORT]  connect through a proxy\n\
+  -t, --threads=N       number of miner threads (default: number of processors)\n\
+  -r, --retries=N       number of times to retry if a network call fails\n\
+                          (default: retry indefinitely)\n\
+  -R, --retry-pause=N   time to pause between retries, in seconds (default: 30)\n\
+  -T, --timeout=N       timeout for long polling, in seconds (default: none)\n\
+  -s, --scantime=N      upper bound on time spent scanning current work when\n\
+                          long polling is unavailable, in seconds (default: 5)\n\
+      --coinbase-addr=ADDR  payout address for solo mining\n\
+      --coinbase-sig=TEXT  data to insert in the coinbase when possible\n\
+      --no-longpoll     disable long polling support\n\
+      --no-getwork      disable getwork support\n\
+      --no-gbt          disable getblocktemplate support\n\
+      --no-stratum      disable X-Stratum support\n\
+      --no-redirect     ignore requests to change the URL of the mining server\n\
+  -q, --quiet           disable per-thread hashmeter output\n\
+  -D, --debug           enable debug output\n\
+  -P, --protocol-dump   verbose dump of protocol-level activities\n"
+#ifdef HAVE_SYSLOG_H
+"\
+  -S, --syslog          use system log for output messages\n"
+#endif
+#ifndef WIN32
+"\
+  -B, --background      run the miner in the background\n"
+#endif
+"\
+      --benchmark       run in offline benchmark mode\n\
+  -c, --config=FILE     load a JSON-format configuration file\n\
+  -V, --version         display version information and exit\n\
+  -h, --help            display this help text and exit\n\
+";
+
+static char const short_options[] =
+#ifndef WIN32
+	"B"
+#endif
+#ifdef HAVE_SYSLOG_H
+	"S"
+#endif
+	"a:c:Dhp:Px:qr:R:s:t:T:o:u:O:V";
+
+static struct option const options[] = {
+	{ "algo", 1, NULL, 'a' },
+#ifndef WIN32
+	{ "background", 0, NULL, 'B' },
+#endif
+	{ "benchmark", 0, NULL, 1005 },
+	{ "cert", 1, NULL, 1001 },
+	{ "coinbase-addr", 1, NULL, 1013 },
+	{ "coinbase-sig", 1, NULL, 1015 },
+	{ "config", 1, NULL, 'c' },
+	{ "debug", 0, NULL, 'D' },
+	{ "help", 0, NULL, 'h' },
+	{ "no-gbt", 0, NULL, 1011 },
+	{ "no-getwork", 0, NULL, 1010 },
+	{ "no-longpoll", 0, NULL, 1003 },
+	{ "no-redirect", 0, NULL, 1009 },
+	{ "no-stratum", 0, NULL, 1007 },
+	{ "pass", 1, NULL, 'p' },
+	{ "protocol-dump", 0, NULL, 'P' },
+	{ "proxy", 1, NULL, 'x' },
+	{ "quiet", 0, NULL, 'q' },
+	{ "retries", 1, NULL, 'r' },
+	{ "retry-pause", 1, NULL, 'R' },
+	{ "scantime", 1, NULL, 's' },
+#ifdef HAVE_SYSLOG_H
+	{ "syslog", 0, NULL, 'S' },
+#endif
+	{ "threads", 1, NULL, 't' },
+	{ "timeout", 1, NULL, 'T' },
+	{ "url", 1, NULL, 'o' },
+	{ "user", 1, NULL, 'u' },
+	{ "userpass", 1, NULL, 'O' },
+	{ "version", 0, NULL, 'V' },
+	{ 0, 0, 0, 0 }
+};
+
+struct work {
+	uint32_t data[32];
+	uint32_t target[8];
+
+	int height;
+	char *txs;
+	char *workid;
+
+	char *job_id;
+	size_t xnonce2_len;
+	unsigned char *xnonce2;
+};
+
+static struct work g_work;
+static time_t g_work_time;
+static pthread_mutex_t g_work_lock;
+static bool submit_old = false;
+static char *lp_id;
+
+static inline void work_free(struct work *w)
+{
+	free(w->txs);
+	free(w->workid);
+	free(w->job_id);
+	free(w->xnonce2);
+}
+
+static inline void work_copy(struct work *dest, const struct work *src)
+{
+	memcpy(dest, src, sizeof(struct work));
+	if (src->txs)
+		dest->txs = strdup(src->txs);
+	if (src->workid)
+		dest->workid = strdup(src->workid);
+	if (src->job_id)
+		dest->job_id = strdup(src->job_id);
+	if (src->xnonce2) {
+		dest->xnonce2 = malloc(src->xnonce2_len);
+		memcpy(dest->xnonce2, src->xnonce2, src->xnonce2_len);
+	}
+}
+
+static bool jobj_binary(const json_t *obj, const char *key,
+			void *buf, size_t buflen)
+{
+	const char *hexstr;
+	json_t *tmp;
+
+	tmp = json_object_get(obj, key);
+	if (unlikely(!tmp)) {
+		applog(LOG_ERR, "JSON key '%s' not found", key);
+		return false;
+	}
+	hexstr = json_string_value(tmp);
+	if (unlikely(!hexstr)) {
+		applog(LOG_ERR, "JSON key '%s' is not a string", key);
+		return false;
+	}
+	if (!hex2bin(buf, hexstr, buflen))
+		return false;
+
+	return true;
+}
+
+static bool work_decode(const json_t *val, struct work *work)
+{
+	int i;
+
+	if (unlikely(!jobj_binary(val, "data", work->data, sizeof(work->data)))) {
+		applog(LOG_ERR, "JSON invalid data");
+		goto err_out;
+	}
+	if (unlikely(!jobj_binary(val, "target", work->target, sizeof(work->target)))) {
+		applog(LOG_ERR, "JSON invalid target");
+		goto err_out;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(work->data); i++)
+		work->data[i] = le32dec(work->data + i);
+	for (i = 0; i < ARRAY_SIZE(work->target); i++)
+		work->target[i] = le32dec(work->target + i);
+
+	return true;
+
+err_out:
+	return false;
+}
+
+static bool gbt_work_decode(const json_t *val, struct work *work)
+{
+	int i, n;
+	uint32_t version, curtime, bits;
+	uint32_t prevhash[8];
+	uint32_t target[8];
+	int cbtx_size;
+	unsigned char *cbtx = NULL;
+	int tx_count, tx_size;
+	unsigned char txc_vi[9];
+	unsigned char (*merkle_tree)[32] = NULL;
+	bool coinbase_append = false;
+	bool submit_coinbase = false;
+	bool segwit = false;
+	json_t *tmp, *txa;
+	bool rc = false;
+
+	tmp = json_object_get(val, "rules");
+	if (tmp && json_is_array(tmp)) {
+		n = json_array_size(tmp);
+		for (i = 0; i < n; i++) {
+			const char *s = json_string_value(json_array_get(tmp, i));
+			if (!s)
+				continue;
+			if (!strcmp(s, "segwit") || !strcmp(s, "!segwit"))
+				segwit = true;
+		}
+	}
+
+	tmp = json_object_get(val, "mutable");
+	if (tmp && json_is_array(tmp)) {
+		n = json_array_size(tmp);
+		for (i = 0; i < n; i++) {
+			const char *s = json_string_value(json_array_get(tmp, i));
+			if (!s)
+				continue;
+			if (!strcmp(s, "coinbase/append"))
+				coinbase_append = true;
+			else if (!strcmp(s, "submit/coinbase"))
+				submit_coinbase = true;
+		}
+	}
+
+	tmp = json_object_get(val, "height");
+	if (!tmp || !json_is_integer(tmp)) {
+		applog(LOG_ERR, "JSON invalid height");
+		goto out;
+	}
+	work->height = json_integer_value(tmp);
+
+	tmp = json_object_get(val, "version");
+	if (!tmp || !json_is_integer(tmp)) {
+		applog(LOG_ERR, "JSON invalid version");
+		goto out;
+	}
+	version = json_integer_value(tmp);
+
+	if (unlikely(!jobj_binary(val, "previousblockhash", prevhash, sizeof(prevhash)))) {
+		applog(LOG_ERR, "JSON invalid previousblockhash");
+		goto out;
+	}
+
+	tmp = json_object_get(val, "curtime");
+	if (!tmp || !json_is_integer(tmp)) {
+		applog(LOG_ERR, "JSON invalid curtime");
+		goto out;
+	}
+	curtime = json_integer_value(tmp);
+
+	if (unlikely(!jobj_binary(val, "bits", &bits, sizeof(bits)))) {
+		applog(LOG_ERR, "JSON invalid bits");
+		goto out;
+	}
+
+	/* find count and size of transactions */
+	txa = json_object_get(val, "transactions");
+	if (!txa || !json_is_array(txa)) {
+		applog(LOG_ERR, "JSON invalid transactions");
+		goto out;
+	}
+	tx_count = json_array_size(txa);
+	tx_size = 0;
+	for (i = 0; i < tx_count; i++) {
+		const json_t *tx = json_array_get(txa, i);
+		const char *tx_hex = json_string_value(json_object_get(tx, "data"));
+		if (!tx_hex) {
+			applog(LOG_ERR, "JSON invalid transactions");
+			goto out;
+		}
+		tx_size += strlen(tx_hex) / 2;
+	}
+
+	/* build coinbase transaction */
+	tmp = json_object_get(val, "coinbasetxn");
+	if (tmp) {
+		const char *cbtx_hex = json_string_value(json_object_get(tmp, "data"));
+		cbtx_size = cbtx_hex ? strlen(cbtx_hex) / 2 : 0;
+		cbtx = malloc(cbtx_size + 100);
+		if (cbtx_size < 60 || !hex2bin(cbtx, cbtx_hex, cbtx_size)) {
+			applog(LOG_ERR, "JSON invalid coinbasetxn");
+			goto out;
+		}
+	} else {
+		int64_t cbvalue;
+		if (!pk_script_size) {
+			if (allow_getwork) {
+				applog(LOG_INFO, "No payout address provided, switching to getwork");
+				have_gbt = false;
+			} else
+				applog(LOG_ERR, "No payout address provided");
+			goto out;
+		}
+		tmp = json_object_get(val, "coinbasevalue");
+		if (!tmp || !json_is_number(tmp)) {
+			applog(LOG_ERR, "JSON invalid coinbasevalue");
+			goto out;
+		}
+		cbvalue = json_is_integer(tmp) ? json_integer_value(tmp) : json_number_value(tmp);
+		cbtx = malloc(256);
+		le32enc((uint32_t *)cbtx, 1); /* version */
+		cbtx[4] = 1; /* in-counter */
+		memset(cbtx+5, 0x00, 32); /* prev txout hash */
+		le32enc((uint32_t *)(cbtx+37), 0xffffffff); /* prev txout index */
+		cbtx_size = 43;
+		/* BIP 34: height in coinbase */
+		for (n = work->height; n; n >>= 8) {
+			cbtx[cbtx_size++] = n & 0xff;
+			if (n < 0x100 && n >= 0x80)
+				cbtx[cbtx_size++] = 0;
+		}
+		cbtx[42] = cbtx_size - 43;
+		cbtx[41] = cbtx_size - 42; /* scriptsig length */
+		le32enc((uint32_t *)(cbtx+cbtx_size), 0xffffffff); /* sequence */
+		cbtx_size += 4;
+		cbtx[cbtx_size++] = segwit ? 2 : 1; /* out-counter */
+		le32enc((uint32_t *)(cbtx+cbtx_size), (uint32_t)cbvalue); /* value */
+		le32enc((uint32_t *)(cbtx+cbtx_size+4), cbvalue >> 32);
+		cbtx_size += 8;
+		cbtx[cbtx_size++] = pk_script_size; /* txout-script length */
+		memcpy(cbtx+cbtx_size, pk_script, pk_script_size);
+		cbtx_size += pk_script_size;
+		if (segwit) {
+			unsigned char (*wtree)[32] = calloc(tx_count + 2, 32);
+			memset(cbtx+cbtx_size, 0, 8); /* value */
+			cbtx_size += 8;
+			cbtx[cbtx_size++] = 38; /* txout-script length */
+			cbtx[cbtx_size++] = 0x6a; /* txout-script */
+			cbtx[cbtx_size++] = 0x24;
+			cbtx[cbtx_size++] = 0xaa;
+			cbtx[cbtx_size++] = 0x21;
+			cbtx[cbtx_size++] = 0xa9;
+			cbtx[cbtx_size++] = 0xed;
+			for (i = 0; i < tx_count; i++) {
+				const json_t *tx = json_array_get(txa, i);
+				const json_t *hash = json_object_get(tx, "hash");
+				if (!hash || !hex2bin(wtree[1+i], json_string_value(hash), 32)) {
+					applog(LOG_ERR, "JSON invalid transaction hash");
+					free(wtree);
+					goto out;
+				}
+				memrev(wtree[1+i], 32);
+			}
+			n = tx_count + 1;
+			while (n > 1) {
+				if (n % 2)
+					memcpy(wtree[n], wtree[n-1], 32);
+				n = (n + 1) / 2;
+				for (i = 0; i < n; i++)
+					sha256d(wtree[i], wtree[2*i], 64);
+			}
+			memset(wtree[1], 0, 32);  /* witness reserved value = 0 */
+			sha256d(cbtx+cbtx_size, wtree[0], 64);
+			cbtx_size += 32;
+			free(wtree);
+		}
+		le32enc((uint32_t *)(cbtx+cbtx_size), 0); /* lock time */
+		cbtx_size += 4;
+		coinbase_append = true;
+	}
+	if (coinbase_append) {
+		unsigned char xsig[100];
+		int xsig_len = 0;
+		if (*coinbase_sig) {
+			n = strlen(coinbase_sig);
+			if (cbtx[41] + xsig_len + n <= 100) {
+				memcpy(xsig+xsig_len, coinbase_sig, n);
+				xsig_len += n;
+			} else {
+				applog(LOG_WARNING, "Signature does not fit in coinbase, skipping");
+			}
+		}
+		tmp = json_object_get(val, "coinbaseaux");
+		if (tmp && json_is_object(tmp)) {
+			void *iter = json_object_iter(tmp);
+			while (iter) {
+				unsigned char buf[100];
+				const char *s = json_string_value(json_object_iter_value(iter));
+				n = s ? strlen(s) / 2 : 0;
+				if (!s || n > 100 || !hex2bin(buf, s, n)) {
+					applog(LOG_ERR, "JSON invalid coinbaseaux");
+					break;
+				}
+				if (cbtx[41] + xsig_len + n <= 100) {
+					memcpy(xsig+xsig_len, buf, n);
+					xsig_len += n;
+				}
+				iter = json_object_iter_next(tmp, iter);
+			}
+		}
+		if (xsig_len) {
+			unsigned char *ssig_end = cbtx + 42 + cbtx[41];
+			int push_len = cbtx[41] + xsig_len < 76 ? 1 :
+			               cbtx[41] + 2 + xsig_len > 100 ? 0 : 2;
+			n = xsig_len + push_len;
+			memmove(ssig_end + n, ssig_end, cbtx_size - 42 - cbtx[41]);
+			cbtx[41] += n;
+			if (push_len == 2)
+				*(ssig_end++) = 0x4c; /* OP_PUSHDATA1 */
+			if (push_len)
+				*(ssig_end++) = xsig_len;
+			memcpy(ssig_end, xsig, xsig_len);
+			cbtx_size += n;
+		}
+	}
+
+	n = varint_encode(txc_vi, 1 + tx_count);
+	work->txs = malloc(2 * (n + cbtx_size + tx_size) + 1);
+	bin2hex(work->txs, txc_vi, n);
+	bin2hex(work->txs + 2*n, cbtx, cbtx_size);
+
+	/* generate merkle root */
+	merkle_tree = malloc(32 * ((1 + tx_count + 1) & ~1));
+	sha256d(merkle_tree[0], cbtx, cbtx_size);
+	for (i = 0; i < tx_count; i++) {
+		tmp = json_array_get(txa, i);
+		const char *tx_hex = json_string_value(json_object_get(tmp, "data"));
+		const int tx_size = tx_hex ? strlen(tx_hex) / 2 : 0;
+		if (segwit) {
+			const char *txid = json_string_value(json_object_get(tmp, "txid"));
+			if (!txid || !hex2bin(merkle_tree[1 + i], txid, 32)) {
+				applog(LOG_ERR, "JSON invalid transaction txid");
+				goto out;
+			}
+			memrev(merkle_tree[1 + i], 32);
+		} else {
+			unsigned char *tx = malloc(tx_size);
+			if (!tx_hex || !hex2bin(tx, tx_hex, tx_size)) {
+				applog(LOG_ERR, "JSON invalid transactions");
+				free(tx);
+				goto out;
+			}
+			sha256d(merkle_tree[1 + i], tx, tx_size);
+			free(tx);
+		}
+		if (!submit_coinbase)
+			strcat(work->txs, tx_hex);
+	}
+	n = 1 + tx_count;
+	while (n > 1) {
+		if (n % 2) {
+			memcpy(merkle_tree[n], merkle_tree[n-1], 32);
+			++n;
+		}
+		n /= 2;
+		for (i = 0; i < n; i++)
+			sha256d(merkle_tree[i], merkle_tree[2*i], 64);
+	}
+
+	/* assemble block header */
+	work->data[0] = swab32(version);
+	for (i = 0; i < 8; i++)
+		work->data[8 - i] = le32dec(prevhash + i);
+	for (i = 0; i < 8; i++)
+		work->data[9 + i] = be32dec((uint32_t *)merkle_tree[0] + i);
+	work->data[17] = swab32(curtime);
+	work->data[18] = le32dec(&bits);
+	memset(work->data + 19, 0x00, 52);
+	work->data[20] = 0x80000000;
+	work->data[31] = 0x00000280;
+
+	if (unlikely(!jobj_binary(val, "target", target, sizeof(target)))) {
+		applog(LOG_ERR, "JSON invalid target");
+		goto out;
+	}
+	for (i = 0; i < ARRAY_SIZE(work->target); i++)
+		work->target[7 - i] = be32dec(target + i);
+
+	tmp = json_object_get(val, "workid");
+	if (tmp) {
+		if (!json_is_string(tmp)) {
+			applog(LOG_ERR, "JSON invalid workid");
+			goto out;
+		}
+		work->workid = strdup(json_string_value(tmp));
+	}
+
+	/* Long polling */
+	tmp = json_object_get(val, "longpollid");
+	if (want_longpoll && json_is_string(tmp)) {
+		free(lp_id);
+		lp_id = strdup(json_string_value(tmp));
+		if (!have_longpoll) {
+			char *lp_uri;
+			tmp = json_object_get(val, "longpolluri");
+			lp_uri = strdup(json_is_string(tmp) ? json_string_value(tmp) : rpc_url);
+			have_longpoll = true;
+			tq_push(thr_info[longpoll_thr_id].q, lp_uri);
+		}
+	}
+
+	rc = true;
+
+out:
+	free(cbtx);
+	free(merkle_tree);
+	return rc;
+}
+
+static void share_result(int result, const char *reason)
+{
+	char s[345];
+	double hashrate;
+	int i;
+
+	hashrate = 0.;
+	pthread_mutex_lock(&stats_lock);
+	for (i = 0; i < opt_n_threads; i++)
+		hashrate += thr_hashrates[i];
+	result ? accepted_count++ : rejected_count++;
+	pthread_mutex_unlock(&stats_lock);
+
+	sprintf(s, hashrate >= 1e6 ? "%.0f" : "%.2f", 1e-3 * hashrate);
+	applog(LOG_INFO, "accepted: %lu/%lu (%.2f%%), %s khash/s %s",
+		   accepted_count,
+		   accepted_count + rejected_count,
+		   100. * accepted_count / (accepted_count + rejected_count),
+		   s,
+		   result ? "(yay!!!)" : "(booooo)");
+
+	if (opt_debug && reason)
+		applog(LOG_DEBUG, "DEBUG: reject reason: %s", reason);
+}
+
+static bool submit_upstream_work(CURL *curl, struct work *work)
+{
+	json_t *val, *res, *reason;
+	char data_str[2 * sizeof(work->data) + 1];
+	char s[345];
+	int i;
+	bool rc = false;
+
+	/* pass if the previous hash is not the current previous hash */
+	if (!submit_old && memcmp(work->data + 1, g_work.data + 1, 32)) {
+		if (opt_debug)
+			applog(LOG_DEBUG, "DEBUG: stale work detected, discarding");
+		return true;
+	}
+
+	if (have_stratum) {
+		uint32_t ntime, nonce;
+		char ntimestr[9], noncestr[9], *xnonce2str, *req;
+
+		le32enc(&ntime, work->data[17]);
+		le32enc(&nonce, work->data[19]);
+		bin2hex(ntimestr, (const unsigned char *)(&ntime), 4);
+		bin2hex(noncestr, (const unsigned char *)(&nonce), 4);
+		xnonce2str = abin2hex(work->xnonce2, work->xnonce2_len);
+		req = malloc(256 + strlen(rpc_user) + strlen(work->job_id) + 2 * work->xnonce2_len);
+		sprintf(req,
+			"{\"method\": \"mining.submit\", \"params\": [\"%s\", \"%s\", \"%s\", \"%s\", \"%s\"], \"id\":4}",
+			rpc_user, work->job_id, xnonce2str, ntimestr, noncestr);
+		free(xnonce2str);
+
+		rc = stratum_send_line(&stratum, req);
+		free(req);
+		if (unlikely(!rc)) {
+			applog(LOG_ERR, "submit_upstream_work stratum_send_line failed");
+			goto out;
+		}
+	} else if (work->txs) {
+		char *req;
+
+		for (i = 0; i < ARRAY_SIZE(work->data); i++)
+			be32enc(work->data + i, work->data[i]);
+		bin2hex(data_str, (unsigned char *)work->data, 80);
+		if (work->workid) {
+			char *params;
+			val = json_object();
+			json_object_set_new(val, "workid", json_string(work->workid));
+			params = json_dumps(val, 0);
+			json_decref(val);
+			req = malloc(128 + 2*80 + strlen(work->txs) + strlen(params));
+			sprintf(req,
+				"{\"method\": \"submitblock\", \"params\": [\"%s%s\", %s], \"id\":1}\r\n",
+				data_str, work->txs, params);
+			free(params);
+		} else {
+			req = malloc(128 + 2*80 + strlen(work->txs));
+			sprintf(req,
+				"{\"method\": \"submitblock\", \"params\": [\"%s%s\"], \"id\":1}\r\n",
+				data_str, work->txs);
+		}
+		val = json_rpc_call(curl, rpc_url, rpc_userpass, req, NULL, 0);
+		free(req);
+		if (unlikely(!val)) {
+			applog(LOG_ERR, "submit_upstream_work json_rpc_call failed");
+			goto out;
+		}
+
+		res = json_object_get(val, "result");
+		if (json_is_object(res)) {
+			char *res_str;
+			bool sumres = false;
+			void *iter = json_object_iter(res);
+			while (iter) {
+				if (json_is_null(json_object_iter_value(iter))) {
+					sumres = true;
+					break;
+				}
+				iter = json_object_iter_next(res, iter);
+			}
+			res_str = json_dumps(res, 0);
+			share_result(sumres, res_str);
+			free(res_str);
+		} else
+			share_result(json_is_null(res), json_string_value(res));
+
+		json_decref(val);
+	} else {
+		/* build hex string */
+		for (i = 0; i < ARRAY_SIZE(work->data); i++)
+			le32enc(work->data + i, work->data[i]);
+		bin2hex(data_str, (unsigned char *)work->data, sizeof(work->data));
+
+		/* build JSON-RPC request */
+		sprintf(s,
+			"{\"method\": \"getwork\", \"params\": [ \"%s\" ], \"id\":1}\r\n",
+			data_str);
+
+		/* issue JSON-RPC request */
+		val = json_rpc_call(curl, rpc_url, rpc_userpass, s, NULL, 0);
+		if (unlikely(!val)) {
+			applog(LOG_ERR, "submit_upstream_work json_rpc_call failed");
+			goto out;
+		}
+
+		res = json_object_get(val, "result");
+		reason = json_object_get(val, "reject-reason");
+		share_result(json_is_true(res), reason ? json_string_value(reason) : NULL);
+
+		json_decref(val);
+	}
+
+	rc = true;
+
+out:
+	return rc;
+}
+
+static const char *getwork_req =
+	"{\"method\": \"getwork\", \"params\": [], \"id\":0}\r\n";
+
+#define GBT_CAPABILITIES "[\"coinbasetxn\", \"coinbasevalue\", \"longpoll\", \"workid\"]"
+#define GBT_RULES "[\"segwit\"]"
+
+static const char *gbt_req =
+	"{\"method\": \"getblocktemplate\", \"params\": [{\"capabilities\": "
+	GBT_CAPABILITIES ", \"rules\": " GBT_RULES "}], \"id\":0}\r\n";
+static const char *gbt_lp_req =
+	"{\"method\": \"getblocktemplate\", \"params\": [{\"capabilities\": "
+	GBT_CAPABILITIES ", \"rules\": " GBT_RULES ", \"longpollid\": \"%s\"}], \"id\":0}\r\n";
+
+static bool get_upstream_work(CURL *curl, struct work *work)
+{
+	json_t *val;
+	int err;
+	bool rc;
+	struct timeval tv_start, tv_end, diff;
+
+start:
+	gettimeofday(&tv_start, NULL);
+	val = json_rpc_call(curl, rpc_url, rpc_userpass,
+			    have_gbt ? gbt_req : getwork_req,
+			    &err, have_gbt ? JSON_RPC_QUIET_404 : 0);
+	gettimeofday(&tv_end, NULL);
+
+	if (have_stratum) {
+		if (val)
+			json_decref(val);
+		return true;
+	}
+
+	if (!have_gbt && !allow_getwork) {
+		applog(LOG_ERR, "No usable protocol");
+		if (val)
+			json_decref(val);
+		return false;
+	}
+
+	if (have_gbt && allow_getwork && !val && err == CURLE_OK) {
+		applog(LOG_INFO, "getblocktemplate failed, falling back to getwork");
+		have_gbt = false;
+		goto start;
+	}
+
+	if (!val)
+		return false;
+
+	if (have_gbt) {
+		rc = gbt_work_decode(json_object_get(val, "result"), work);
+		if (!have_gbt) {
+			json_decref(val);
+			goto start;
+		}
+	} else
+		rc = work_decode(json_object_get(val, "result"), work);
+
+	if (opt_debug && rc) {
+		timeval_subtract(&diff, &tv_end, &tv_start);
+		applog(LOG_DEBUG, "DEBUG: got new work in %d ms",
+		       diff.tv_sec * 1000 + diff.tv_usec / 1000);
+	}
+
+	json_decref(val);
+
+	return rc;
+}
+
+static void workio_cmd_free(struct workio_cmd *wc)
+{
+	if (!wc)
+		return;
+
+	switch (wc->cmd) {
+	case WC_SUBMIT_WORK:
+		work_free(wc->u.work);
+		free(wc->u.work);
+		break;
+	default: /* do nothing */
+		break;
+	}
+
+	memset(wc, 0, sizeof(*wc));	/* poison */
+	free(wc);
+}
+
+static bool workio_get_work(struct workio_cmd *wc, CURL *curl)
+{
+	struct work *ret_work;
+	int failures = 0;
+
+	ret_work = calloc(1, sizeof(*ret_work));
+	if (!ret_work)
+		return false;
+
+	/* obtain new work from bitcoin via JSON-RPC */
+	while (!get_upstream_work(curl, ret_work)) {
+		if (unlikely((opt_retries >= 0) && (++failures > opt_retries))) {
+			applog(LOG_ERR, "json_rpc_call failed, terminating workio thread");
+			free(ret_work);
+			return false;
+		}
+
+		/* pause, then restart work-request loop */
+		applog(LOG_ERR, "json_rpc_call failed, retry after %d seconds",
+			opt_fail_pause);
+		sleep(opt_fail_pause);
+	}
+
+	/* send work to requesting thread */
+	if (!tq_push(wc->thr->q, ret_work))
+		free(ret_work);
+
+	return true;
+}
+
+static bool workio_submit_work(struct workio_cmd *wc, CURL *curl)
+{
+	int failures = 0;
+
+	/* submit solution to bitcoin via JSON-RPC */
+	while (!submit_upstream_work(curl, wc->u.work)) {
+		if (unlikely((opt_retries >= 0) && (++failures > opt_retries))) {
+			applog(LOG_ERR, "...terminating workio thread");
+			return false;
+		}
+
+		/* pause, then restart work-request loop */
+		applog(LOG_ERR, "...retry after %d seconds",
+			opt_fail_pause);
+		sleep(opt_fail_pause);
+	}
+
+	return true;
+}
+
+static void *workio_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	CURL *curl;
+	bool ok = true;
+
+	curl = curl_easy_init();
+	if (unlikely(!curl)) {
+		applog(LOG_ERR, "CURL initialization failed");
+		return NULL;
+	}
+
+	while (ok) {
+		struct workio_cmd *wc;
+
+		/* wait for workio_cmd sent to us, on our queue */
+		wc = tq_pop(mythr->q, NULL);
+		if (!wc) {
+			ok = false;
+			break;
+		}
+
+		/* process workio_cmd */
+		switch (wc->cmd) {
+		case WC_GET_WORK:
+			ok = workio_get_work(wc, curl);
+			break;
+		case WC_SUBMIT_WORK:
+			ok = workio_submit_work(wc, curl);
+			break;
+
+		default:		/* should never happen */
+			ok = false;
+			break;
+		}
+
+		workio_cmd_free(wc);
+	}
+
+	tq_freeze(mythr->q);
+	curl_easy_cleanup(curl);
+
+	return NULL;
+}
+
+static bool get_work(struct thr_info *thr, struct work *work)
+{
+	struct workio_cmd *wc;
+	struct work *work_heap;
+
+	if (opt_benchmark) {
+		memset(work->data, 0x55, 76);
+		work->data[17] = swab32(time(NULL));
+		memset(work->data + 19, 0x00, 52);
+		work->data[20] = 0x80000000;
+		work->data[31] = 0x00000280;
+		memset(work->target, 0x00, sizeof(work->target));
+		return true;
+	}
+
+	/* fill out work request message */
+	wc = calloc(1, sizeof(*wc));
+	if (!wc)
+		return false;
+
+	wc->cmd = WC_GET_WORK;
+	wc->thr = thr;
+
+	/* send work request to workio thread */
+	if (!tq_push(thr_info[work_thr_id].q, wc)) {
+		workio_cmd_free(wc);
+		return false;
+	}
+
+	/* wait for response, a unit of work */
+	work_heap = tq_pop(thr->q, NULL);
+	if (!work_heap)
+		return false;
+
+	/* copy returned work into storage provided by caller */
+	memcpy(work, work_heap, sizeof(*work));
+	free(work_heap);
+
+	return true;
+}
+
+static bool submit_work(struct thr_info *thr, const struct work *work_in)
+{
+	struct workio_cmd *wc;
+
+	/* fill out work request message */
+	wc = calloc(1, sizeof(*wc));
+	if (!wc)
+		return false;
+
+	wc->u.work = malloc(sizeof(*work_in));
+	if (!wc->u.work)
+		goto err_out;
+
+	wc->cmd = WC_SUBMIT_WORK;
+	wc->thr = thr;
+	work_copy(wc->u.work, work_in);
+
+	/* send solution to workio thread */
+	if (!tq_push(thr_info[work_thr_id].q, wc))
+		goto err_out;
+
+	return true;
+
+err_out:
+	workio_cmd_free(wc);
+	return false;
+}
+
+static void stratum_gen_work(struct stratum_ctx *sctx, struct work *work)
+{
+	unsigned char merkle_root[64];
+	int i;
+
+	pthread_mutex_lock(&sctx->work_lock);
+
+	free(work->job_id);
+	work->job_id = strdup(sctx->job.job_id);
+	work->xnonce2_len = sctx->xnonce2_size;
+	work->xnonce2 = realloc(work->xnonce2, sctx->xnonce2_size);
+	memcpy(work->xnonce2, sctx->job.xnonce2, sctx->xnonce2_size);
+
+	/* Generate merkle root */
+	sha256d(merkle_root, sctx->job.coinbase, sctx->job.coinbase_size);
+	for (i = 0; i < sctx->job.merkle_count; i++) {
+		memcpy(merkle_root + 32, sctx->job.merkle[i], 32);
+		sha256d(merkle_root, merkle_root, 64);
+	}
+
+	/* Increment extranonce2 */
+	for (i = 0; i < sctx->xnonce2_size && !++sctx->job.xnonce2[i]; i++);
+
+	/* Assemble block header */
+	memset(work->data, 0, 128);
+	work->data[0] = le32dec(sctx->job.version);
+	for (i = 0; i < 8; i++)
+		work->data[1 + i] = le32dec((uint32_t *)sctx->job.prevhash + i);
+	for (i = 0; i < 8; i++)
+		work->data[9 + i] = be32dec((uint32_t *)merkle_root + i);
+	work->data[17] = le32dec(sctx->job.ntime);
+	work->data[18] = le32dec(sctx->job.nbits);
+	work->data[20] = 0x80000000;
+	work->data[31] = 0x00000280;
+
+	pthread_mutex_unlock(&sctx->work_lock);
+
+	if (opt_debug) {
+		char *xnonce2str = abin2hex(work->xnonce2, work->xnonce2_len);
+		applog(LOG_DEBUG, "DEBUG: job_id='%s' extranonce2=%s ntime=%08x",
+		       work->job_id, xnonce2str, swab32(work->data[17]));
+		free(xnonce2str);
+	}
+
+	if (opt_algo == ALGO_SCRYPT || opt_algo == ALGO_YESCRYPT || opt_algo == ALGO_YESPOWER)
+		diff_to_target(work->target, sctx->job.diff / 65536.0);
+	else
+		diff_to_target(work->target, sctx->job.diff);
+}
+
+static void *miner_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	int thr_id = mythr->id;
+	struct work work = {{0}};
+	uint32_t max_nonce;
+	uint32_t end_nonce = 0xffffffffU / opt_n_threads * (thr_id + 1) - 0x20;
+	unsigned char *scratchbuf = NULL;
+	char s[16];
+	int i;
+
+	/* Set worker threads to nice 19 and then preferentially to SCHED_IDLE
+	 * and if that fails, then SCHED_BATCH. No need for this to be an
+	 * error if it fails */
+	if (!opt_benchmark) {
+		setpriority(PRIO_PROCESS, 0, 19);
+		drop_policy();
+	}
+
+	/* Cpu affinity only makes sense if the number of threads is a multiple
+	 * of the number of CPUs */
+	if (num_processors > 1 && opt_n_threads % num_processors == 0) {
+		if (!opt_quiet)
+			applog(LOG_INFO, "Binding thread %d to cpu %d",
+			       thr_id, thr_id % num_processors);
+		affine_to_cpu(thr_id, thr_id % num_processors);
+	}
+
+	if (opt_algo == ALGO_SCRYPT) {
+		scratchbuf = scrypt_buffer_alloc(opt_scrypt_n);
+		if (!scratchbuf) {
+			applog(LOG_ERR, "scrypt buffer allocation failed");
+			pthread_mutex_lock(&applog_lock);
+			exit(1);
+		}
+	}
+
+	while (1) {
+		unsigned long hashes_done;
+		struct timeval tv_start, tv_end, diff;
+		int64_t max64;
+		int rc;
+
+		if (have_stratum) {
+			while (time(NULL) >= g_work_time + 120)
+				sleep(1);
+			pthread_mutex_lock(&g_work_lock);
+			if (work.data[19] >= end_nonce && !memcmp(work.data, g_work.data, 76))
+				stratum_gen_work(&stratum, &g_work);
+		} else {
+			int min_scantime = have_longpoll ? LP_SCANTIME : opt_scantime;
+			/* obtain new work from internal workio thread */
+			pthread_mutex_lock(&g_work_lock);
+			if (!have_stratum &&
+			    (time(NULL) - g_work_time >= min_scantime ||
+			     work.data[19] >= end_nonce)) {
+				work_free(&g_work);
+				if (unlikely(!get_work(mythr, &g_work))) {
+					applog(LOG_ERR, "work retrieval failed, exiting "
+						"mining thread %d", mythr->id);
+					pthread_mutex_unlock(&g_work_lock);
+					goto out;
+				}
+				g_work_time = have_stratum ? 0 : time(NULL);
+			}
+			if (have_stratum) {
+				pthread_mutex_unlock(&g_work_lock);
+				continue;
+			}
+		}
+		if (memcmp(work.data, g_work.data, 76)) {
+			work_free(&work);
+			work_copy(&work, &g_work);
+			work.data[19] = 0xffffffffU / opt_n_threads * thr_id;
+		} else
+			work.data[19]++;
+		pthread_mutex_unlock(&g_work_lock);
+		work_restart[thr_id].restart = 0;
+
+		/* adjust max_nonce to meet target scan time */
+		if (have_stratum)
+			max64 = LP_SCANTIME;
+		else
+			max64 = g_work_time + (have_longpoll ? LP_SCANTIME : opt_scantime)
+			      - time(NULL);
+		max64 *= thr_hashrates[thr_id];
+		if (max64 <= 0) {
+			switch (opt_algo) {
+                        case ALGO_YESCRYPT:
+				max64 = 0x000fff;
+				break;
+                        case ALGO_YESPOWER:
+				max64 = 0x000fff;
+				break;
+			case ALGO_SCRYPT:
+				max64 = opt_scrypt_n < 16 ? 0x3ffff : 0x3fffff / opt_scrypt_n;
+				break;
+			case ALGO_SHA256D:
+				max64 = 0x1fffff;
+				break;
+			}
+		}
+		if (work.data[19] + max64 > end_nonce)
+			max_nonce = end_nonce;
+		else
+			max_nonce = work.data[19] + max64;
+
+		hashes_done = 0;
+		gettimeofday(&tv_start, NULL);
+
+		/* scan nonces for a proof-of-work hash */
+		switch (opt_algo) {
+		case ALGO_YESCRYPT:
+			rc = scanhash_yescrypt(thr_id, work.data, work.target,
+					       max_nonce, &hashes_done);
+			break;
+
+		case ALGO_YESPOWER:
+			rc = scanhash_yespower(thr_id, work.data, work.target,
+					       max_nonce, &hashes_done);
+			break;
+
+		case ALGO_SCRYPT:
+			rc = scanhash_scrypt(thr_id, work.data, scratchbuf, work.target,
+			                     max_nonce, &hashes_done, opt_scrypt_n);
+			break;
+
+		case ALGO_SHA256D:
+			rc = scanhash_sha256d(thr_id, work.data, work.target,
+			                      max_nonce, &hashes_done);
+			break;
+
+		default:
+			/* should never happen */
+			goto out;
+		}
+
+		/* record scanhash elapsed time */
+		gettimeofday(&tv_end, NULL);
+		timeval_subtract(&diff, &tv_end, &tv_start);
+		if (diff.tv_usec || diff.tv_sec) {
+			pthread_mutex_lock(&stats_lock);
+			thr_hashrates[thr_id] =
+				hashes_done / (diff.tv_sec + 1e-6 * diff.tv_usec);
+			pthread_mutex_unlock(&stats_lock);
+		}
+		if (!opt_quiet) {
+			sprintf(s, thr_hashrates[thr_id] >= 1e6 ? "%.0f" : "%.2f",
+				1e-3 * thr_hashrates[thr_id]);
+			applog(LOG_INFO, "thread %d: %lu hashes, %s khash/s",
+				thr_id, hashes_done, s);
+		}
+		if (opt_benchmark && thr_id == opt_n_threads - 1) {
+			double hashrate = 0.;
+			for (i = 0; i < opt_n_threads && thr_hashrates[i]; i++)
+				hashrate += thr_hashrates[i];
+			if (i == opt_n_threads) {
+				sprintf(s, hashrate >= 1e6 ? "%.0f" : "%.2f", 1e-3 * hashrate);
+				applog(LOG_INFO, "Total: %s khash/s", s);
+			}
+		}
+
+		/* if nonce found, submit work */
+		if (rc && !opt_benchmark && !submit_work(mythr, &work))
+			break;
+	}
+
+out:
+	tq_freeze(mythr->q);
+
+	return NULL;
+}
+
+static void restart_threads(void)
+{
+	int i;
+
+	for (i = 0; i < opt_n_threads; i++)
+		work_restart[i].restart = 1;
+}
+
+static void *longpoll_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	CURL *curl = NULL;
+	char *copy_start, *hdr_path = NULL, *lp_url = NULL;
+	bool need_slash = false;
+
+	curl = curl_easy_init();
+	if (unlikely(!curl)) {
+		applog(LOG_ERR, "CURL initialization failed");
+		goto out;
+	}
+
+start:
+	hdr_path = tq_pop(mythr->q, NULL);
+	if (!hdr_path)
+		goto out;
+
+	/* full URL */
+	if (strstr(hdr_path, "://")) {
+		lp_url = hdr_path;
+		hdr_path = NULL;
+	}
+
+	/* absolute path, on current server */
+	else {
+		copy_start = (*hdr_path == '/') ? (hdr_path + 1) : hdr_path;
+		if (rpc_url[strlen(rpc_url) - 1] != '/')
+			need_slash = true;
+
+		lp_url = malloc(strlen(rpc_url) + strlen(copy_start) + 2);
+		if (!lp_url)
+			goto out;
+
+		sprintf(lp_url, "%s%s%s", rpc_url, need_slash ? "/" : "", copy_start);
+	}
+
+	applog(LOG_INFO, "Long-polling activated for %s", lp_url);
+
+	while (1) {
+		json_t *val, *res, *soval;
+		char *req = NULL;
+		int err;
+
+		if (have_gbt) {
+			req = malloc(strlen(gbt_lp_req) + strlen(lp_id) + 1);
+			sprintf(req, gbt_lp_req, lp_id);
+		}
+		val = json_rpc_call(curl, lp_url, rpc_userpass,
+				    req ? req : getwork_req, &err,
+				    JSON_RPC_LONGPOLL);
+		free(req);
+		if (have_stratum) {
+			if (val)
+				json_decref(val);
+			goto out;
+		}
+		if (likely(val)) {
+			bool rc;
+			applog(LOG_INFO, "LONGPOLL pushed new work");
+			res = json_object_get(val, "result");
+			soval = json_object_get(res, "submitold");
+			submit_old = soval ? json_is_true(soval) : false;
+			pthread_mutex_lock(&g_work_lock);
+			work_free(&g_work);
+			if (have_gbt)
+				rc = gbt_work_decode(res, &g_work);
+			else
+				rc = work_decode(res, &g_work);
+			if (rc) {
+				time(&g_work_time);
+				restart_threads();
+			}
+			pthread_mutex_unlock(&g_work_lock);
+			json_decref(val);
+		} else {
+			pthread_mutex_lock(&g_work_lock);
+			g_work_time -= LP_SCANTIME;
+			pthread_mutex_unlock(&g_work_lock);
+			if (err == CURLE_OPERATION_TIMEDOUT) {
+				restart_threads();
+			} else {
+				have_longpoll = false;
+				restart_threads();
+				free(hdr_path);
+				free(lp_url);
+				lp_url = NULL;
+				sleep(opt_fail_pause);
+				goto start;
+			}
+		}
+	}
+
+out:
+	free(hdr_path);
+	free(lp_url);
+	tq_freeze(mythr->q);
+	if (curl)
+		curl_easy_cleanup(curl);
+
+	return NULL;
+}
+
+static bool stratum_handle_response(char *buf)
+{
+	json_t *val, *err_val, *res_val, *id_val;
+	json_error_t err;
+	bool ret = false;
+
+	val = JSON_LOADS(buf, &err);
+	if (!val) {
+		applog(LOG_INFO, "JSON decode failed(%d): %s", err.line, err.text);
+		goto out;
+	}
+
+	res_val = json_object_get(val, "result");
+	err_val = json_object_get(val, "error");
+	id_val = json_object_get(val, "id");
+
+	if (!id_val || json_is_null(id_val) || !res_val)
+		goto out;
+
+	share_result(json_is_true(res_val),
+		err_val ? json_string_value(json_array_get(err_val, 1)) : NULL);
+
+	ret = true;
+out:
+	if (val)
+		json_decref(val);
+
+	return ret;
+}
+
+static void *stratum_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	char *s;
+
+	stratum.url = tq_pop(mythr->q, NULL);
+	if (!stratum.url)
+		goto out;
+	applog(LOG_INFO, "Starting Stratum on %s", stratum.url);
+
+	while (1) {
+		int failures = 0;
+
+		while (!stratum.curl) {
+			pthread_mutex_lock(&g_work_lock);
+			g_work_time = 0;
+			pthread_mutex_unlock(&g_work_lock);
+			restart_threads();
+
+			if (!stratum_connect(&stratum, stratum.url) ||
+			    !stratum_subscribe(&stratum) ||
+			    !stratum_authorize(&stratum, rpc_user, rpc_pass)) {
+				stratum_disconnect(&stratum);
+				if (opt_retries >= 0 && ++failures > opt_retries) {
+					applog(LOG_ERR, "...terminating workio thread");
+					tq_push(thr_info[work_thr_id].q, NULL);
+					goto out;
+				}
+				applog(LOG_ERR, "...retry after %d seconds", opt_fail_pause);
+				sleep(opt_fail_pause);
+			}
+		}
+
+		if (stratum.job.job_id &&
+		    (!g_work_time || strcmp(stratum.job.job_id, g_work.job_id))) {
+			pthread_mutex_lock(&g_work_lock);
+			stratum_gen_work(&stratum, &g_work);
+			time(&g_work_time);
+			pthread_mutex_unlock(&g_work_lock);
+			if (stratum.job.clean) {
+				applog(LOG_INFO, "Stratum requested work restart");
+				restart_threads();
+			}
+		}
+
+		if (!stratum_socket_full(&stratum, 120)) {
+			applog(LOG_ERR, "Stratum connection timed out");
+			s = NULL;
+		} else
+			s = stratum_recv_line(&stratum);
+		if (!s) {
+			stratum_disconnect(&stratum);
+			applog(LOG_ERR, "Stratum connection interrupted");
+			continue;
+		}
+		if (!stratum_handle_method(&stratum, s))
+			stratum_handle_response(s);
+		free(s);
+	}
+
+out:
+	return NULL;
+}
+
+static void show_version_and_exit(void)
+{
+	printf(PACKAGE_STRING "\n built on " __DATE__ "\n features:"
+#if defined(USE_ASM) && defined(__i386__)
+		" i386"
+#endif
+#if defined(USE_ASM) && defined(__x86_64__)
+		" x86_64"
+		" PHE"
+#endif
+#if defined(USE_ASM) && (defined(__i386__) || defined(__x86_64__))
+		" SSE2"
+#endif
+#if defined(__x86_64__) && defined(USE_AVX)
+		" AVX"
+#endif
+#if defined(__x86_64__) && defined(USE_AVX2)
+		" AVX2"
+#endif
+#if defined(__x86_64__) && defined(USE_XOP)
+		" XOP"
+#endif
+#if defined(USE_ASM) && defined(__arm__) && defined(__APCS_32__)
+		" ARM"
+#if defined(__ARM_ARCH_5E__) || defined(__ARM_ARCH_5TE__) || \
+	defined(__ARM_ARCH_5TEJ__) || defined(__ARM_ARCH_6__) || \
+	defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || \
+	defined(__ARM_ARCH_6M__) || defined(__ARM_ARCH_6T2__) || \
+	defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || \
+	defined(__ARM_ARCH_7__) || \
+	defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || \
+	defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7EM__)
+		" ARMv5E"
+#endif
+#if defined(__ARM_NEON__)
+		" NEON"
+#endif
+#endif
+#if defined(USE_ASM) && (defined(__powerpc__) || defined(__ppc__) || defined(__PPC__))
+		" PowerPC"
+#if defined(__ALTIVEC__)
+		" AltiVec"
+#endif
+#endif
+		"\n");
+
+	printf("%s\n", curl_version());
+#ifdef JANSSON_VERSION
+	printf("libjansson %s\n", JANSSON_VERSION);
+#endif
+	exit(0);
+}
+
+static void show_usage_and_exit(int status)
+{
+	if (status)
+		fprintf(stderr, "Try `" PROGRAM_NAME " --help' for more information.\n");
+	else
+		printf(usage);
+	exit(status);
+}
+
+static void strhide(char *s)
+{
+	if (*s) *s++ = 'x';
+	while (*s) *s++ = '\0';
+}
+
+static void parse_config(json_t *config, char *pname, char *ref);
+
+static void parse_arg(int key, char *arg, char *pname)
+{
+	char *p;
+	int v, i;
+
+	switch(key) {
+	case 'a':
+		for (i = 0; i < ARRAY_SIZE(algo_names); i++) {
+			v = strlen(algo_names[i]);
+			if (!strncmp(arg, algo_names[i], v)) {
+				if (arg[v] == '\0') {
+					opt_algo = i;
+					break;
+				}
+				if (arg[v] == ':' && i == ALGO_SCRYPT) {
+					char *ep;
+					v = strtol(arg+v+1, &ep, 10);
+					if (*ep || v & (v-1) || v < 2)
+						continue;
+					opt_algo = i;
+					opt_scrypt_n = v;
+					break;
+				}
+			}
+		}
+		if (i == ARRAY_SIZE(algo_names)) {
+			fprintf(stderr, "%s: unknown algorithm -- '%s'\n",
+				pname, arg);
+			show_usage_and_exit(1);
+		}
+		break;
+	case 'B':
+		opt_background = true;
+		break;
+	case 'c': {
+		json_error_t err;
+		json_t *config = JSON_LOAD_FILE(arg, &err);
+		if (!json_is_object(config)) {
+			if (err.line < 0)
+				fprintf(stderr, "%s: %s\n", pname, err.text);
+			else
+				fprintf(stderr, "%s: %s:%d: %s\n",
+					pname, arg, err.line, err.text);
+			exit(1);
+		}
+		parse_config(config, pname, arg);
+		json_decref(config);
+		break;
+	}
+	case 'q':
+		opt_quiet = true;
+		break;
+	case 'D':
+		opt_debug = true;
+		break;
+	case 'p':
+		free(rpc_pass);
+		rpc_pass = strdup(arg);
+		strhide(arg);
+		break;
+	case 'P':
+		opt_protocol = true;
+		break;
+	case 'r':
+		v = atoi(arg);
+		if (v < -1 || v > 9999)	/* sanity check */
+			show_usage_and_exit(1);
+		opt_retries = v;
+		break;
+	case 'R':
+		v = atoi(arg);
+		if (v < 1 || v > 9999)	/* sanity check */
+			show_usage_and_exit(1);
+		opt_fail_pause = v;
+		break;
+	case 's':
+		v = atoi(arg);
+		if (v < 1 || v > 9999)	/* sanity check */
+			show_usage_and_exit(1);
+		opt_scantime = v;
+		break;
+	case 'T':
+		v = atoi(arg);
+		if (v < 1 || v > 99999)	/* sanity check */
+			show_usage_and_exit(1);
+		opt_timeout = v;
+		break;
+	case 't':
+		v = atoi(arg);
+		if (v < 1 || v > 9999)	/* sanity check */
+			show_usage_and_exit(1);
+		opt_n_threads = v;
+		break;
+	case 'u':
+		free(rpc_user);
+		rpc_user = strdup(arg);
+		break;
+	case 'o': {			/* --url */
+		char *ap, *hp;
+		ap = strstr(arg, "://");
+		ap = ap ? ap + 3 : arg;
+		hp = strrchr(arg, '@');
+		if (hp) {
+			*hp = '\0';
+			p = strchr(ap, ':');
+			if (p) {
+				free(rpc_userpass);
+				rpc_userpass = strdup(ap);
+				free(rpc_user);
+				rpc_user = calloc(p - ap + 1, 1);
+				strncpy(rpc_user, ap, p - ap);
+				free(rpc_pass);
+				rpc_pass = strdup(++p);
+				if (*p) *p++ = 'x';
+				v = strlen(hp + 1) + 1;
+				memmove(p + 1, hp + 1, v);
+				memset(p + v, 0, hp - p);
+				hp = p;
+			} else {
+				free(rpc_user);
+				rpc_user = strdup(ap);
+			}
+			*hp++ = '@';
+		} else
+			hp = ap;
+		if (ap != arg) {
+			if (strncasecmp(arg, "http://", 7) &&
+			    strncasecmp(arg, "https://", 8) &&
+			    strncasecmp(arg, "stratum+tcp://", 14) &&
+			    strncasecmp(arg, "stratum+tcps://", 15)) {
+				fprintf(stderr, "%s: unknown protocol -- '%s'\n",
+					pname, arg);
+				show_usage_and_exit(1);
+			}
+			free(rpc_url);
+			rpc_url = strdup(arg);
+			strcpy(rpc_url + (ap - arg), hp);
+		} else {
+			if (*hp == '\0' || *hp == '/') {
+				fprintf(stderr, "%s: invalid URL -- '%s'\n",
+					pname, arg);
+				show_usage_and_exit(1);
+			}
+			free(rpc_url);
+			rpc_url = malloc(strlen(hp) + 8);
+			sprintf(rpc_url, "http://%s", hp);
+		}
+		have_stratum = !opt_benchmark && !strncasecmp(rpc_url, "stratum", 7);
+		break;
+	}
+	case 'O':			/* --userpass */
+		p = strchr(arg, ':');
+		if (!p) {
+			fprintf(stderr, "%s: invalid username:password pair -- '%s'\n",
+				pname, arg);
+			show_usage_and_exit(1);
+		}
+		free(rpc_userpass);
+		rpc_userpass = strdup(arg);
+		free(rpc_user);
+		rpc_user = calloc(p - arg + 1, 1);
+		strncpy(rpc_user, arg, p - arg);
+		free(rpc_pass);
+		rpc_pass = strdup(++p);
+		strhide(p);
+		break;
+	case 'x':			/* --proxy */
+		if (!strncasecmp(arg, "socks4://", 9))
+			opt_proxy_type = CURLPROXY_SOCKS4;
+		else if (!strncasecmp(arg, "socks5://", 9))
+			opt_proxy_type = CURLPROXY_SOCKS5;
+#if LIBCURL_VERSION_NUM >= 0x071200
+		else if (!strncasecmp(arg, "socks4a://", 10))
+			opt_proxy_type = CURLPROXY_SOCKS4A;
+		else if (!strncasecmp(arg, "socks5h://", 10))
+			opt_proxy_type = CURLPROXY_SOCKS5_HOSTNAME;
+#endif
+		else
+			opt_proxy_type = CURLPROXY_HTTP;
+		free(opt_proxy);
+		opt_proxy = strdup(arg);
+		break;
+	case 1001:
+		free(opt_cert);
+		opt_cert = strdup(arg);
+		break;
+	case 1005:
+		opt_benchmark = true;
+		want_longpoll = false;
+		want_stratum = false;
+		have_stratum = false;
+		break;
+	case 1003:
+		want_longpoll = false;
+		break;
+	case 1007:
+		want_stratum = false;
+		break;
+	case 1009:
+		opt_redirect = false;
+		break;
+	case 1010:
+		allow_getwork = false;
+		break;
+	case 1011:
+		have_gbt = false;
+		break;
+	case 1013:			/* --coinbase-addr */
+		pk_script_size = address_to_script(pk_script, sizeof(pk_script), arg);
+		if (!pk_script_size) {
+			fprintf(stderr, "%s: invalid address -- '%s'\n",
+				pname, arg);
+			show_usage_and_exit(1);
+		}
+		break;
+	case 1015:			/* --coinbase-sig */
+		if (strlen(arg) + 1 > sizeof(coinbase_sig)) {
+			fprintf(stderr, "%s: coinbase signature too long\n", pname);
+			show_usage_and_exit(1);
+		}
+		strcpy(coinbase_sig, arg);
+		break;
+	case 'S':
+		use_syslog = true;
+		break;
+	case 'V':
+		show_version_and_exit();
+	case 'h':
+		show_usage_and_exit(0);
+	default:
+		show_usage_and_exit(1);
+	}
+}
+
+static void parse_config(json_t *config, char *pname, char *ref)
+{
+	int i;
+	char *s;
+	json_t *val;
+
+	for (i = 0; i < ARRAY_SIZE(options); i++) {
+		if (!options[i].name)
+			break;
+
+		val = json_object_get(config, options[i].name);
+		if (!val)
+			continue;
+
+		if (options[i].has_arg && json_is_string(val)) {
+			if (!strcmp(options[i].name, "config")) {
+				fprintf(stderr, "%s: %s: option '%s' not allowed here\n",
+					pname, ref, options[i].name);
+				exit(1);
+			}
+			s = strdup(json_string_value(val));
+			if (!s)
+				break;
+			parse_arg(options[i].val, s, pname);
+			free(s);
+		} else if (!options[i].has_arg && json_is_true(val)) {
+			parse_arg(options[i].val, "", pname);
+		} else {
+			fprintf(stderr, "%s: invalid argument for option '%s'\n",
+				pname, options[i].name);
+			exit(1);
+		}
+	}
+}
+
+static void parse_cmdline(int argc, char *argv[])
+{
+	int key;
+
+	while (1) {
+#if HAVE_GETOPT_LONG
+		key = getopt_long(argc, argv, short_options, options, NULL);
+#else
+		key = getopt(argc, argv, short_options);
+#endif
+		if (key < 0)
+			break;
+
+		parse_arg(key, optarg, argv[0]);
+	}
+	if (optind < argc) {
+		fprintf(stderr, "%s: unsupported non-option argument -- '%s'\n",
+			argv[0], argv[optind]);
+		show_usage_and_exit(1);
+	}
+}
+
+#ifndef WIN32
+static void signal_handler(int sig)
+{
+	switch (sig) {
+	case SIGHUP:
+		applog(LOG_INFO, "SIGHUP received");
+		break;
+	case SIGINT:
+		applog(LOG_INFO, "SIGINT received, exiting");
+		exit(0);
+		break;
+	case SIGTERM:
+		applog(LOG_INFO, "SIGTERM received, exiting");
+		exit(0);
+		break;
+	}
+}
+#endif
+
+int main(int argc, char *argv[])
+{
+	struct thr_info *thr;
+	long flags;
+	int i;
+
+	rpc_user = strdup("");
+	rpc_pass = strdup("");
+
+	/* parse command line */
+	parse_cmdline(argc, argv);
+
+	if (!opt_benchmark && !rpc_url) {
+		fprintf(stderr, "%s: no URL supplied\n", argv[0]);
+		show_usage_and_exit(1);
+	}
+
+	if (!rpc_userpass) {
+		rpc_userpass = malloc(strlen(rpc_user) + strlen(rpc_pass) + 2);
+		if (!rpc_userpass)
+			return 1;
+		sprintf(rpc_userpass, "%s:%s", rpc_user, rpc_pass);
+	}
+
+	pthread_mutex_init(&applog_lock, NULL);
+	pthread_mutex_init(&stats_lock, NULL);
+	pthread_mutex_init(&g_work_lock, NULL);
+	pthread_mutex_init(&stratum.sock_lock, NULL);
+	pthread_mutex_init(&stratum.work_lock, NULL);
+
+	flags = opt_benchmark || (strncasecmp(rpc_url, "https://", 8) &&
+	                          strncasecmp(rpc_url, "stratum+tcps://", 15))
+	      ? (CURL_GLOBAL_ALL & ~CURL_GLOBAL_SSL)
+	      : CURL_GLOBAL_ALL;
+	if (curl_global_init(flags)) {
+		applog(LOG_ERR, "CURL initialization failed");
+		return 1;
+	}
+
+#ifndef WIN32
+	if (opt_background) {
+		i = fork();
+		if (i < 0) exit(1);
+		if (i > 0) exit(0);
+		i = setsid();
+		if (i < 0)
+			applog(LOG_ERR, "setsid() failed (errno = %d)", errno);
+		i = chdir("/");
+		if (i < 0)
+			applog(LOG_ERR, "chdir() failed (errno = %d)", errno);
+		signal(SIGHUP, signal_handler);
+		signal(SIGINT, signal_handler);
+		signal(SIGTERM, signal_handler);
+	}
+#endif
+
+#if defined(WIN32)
+	SYSTEM_INFO sysinfo;
+	GetSystemInfo(&sysinfo);
+	num_processors = sysinfo.dwNumberOfProcessors;
+#elif defined(_SC_NPROCESSORS_CONF)
+	num_processors = sysconf(_SC_NPROCESSORS_CONF);
+#elif defined(CTL_HW) && defined(HW_NCPU)
+	int req[] = { CTL_HW, HW_NCPU };
+	size_t len = sizeof(num_processors);
+	sysctl(req, 2, &num_processors, &len, NULL, 0);
+#else
+	num_processors = 1;
+#endif
+	if (num_processors < 1)
+		num_processors = 1;
+	if (!opt_n_threads)
+		opt_n_threads = num_processors;
+
+#ifdef HAVE_SYSLOG_H
+	if (use_syslog)
+		openlog("cpuminer", LOG_PID, LOG_USER);
+#endif
+
+	work_restart = calloc(opt_n_threads, sizeof(*work_restart));
+	if (!work_restart)
+		return 1;
+
+	thr_info = calloc(opt_n_threads + 3, sizeof(*thr));
+	if (!thr_info)
+		return 1;
+
+	thr_hashrates = (double *) calloc(opt_n_threads, sizeof(double));
+	if (!thr_hashrates)
+		return 1;
+
+	/* init workio thread info */
+	work_thr_id = opt_n_threads;
+	thr = &thr_info[work_thr_id];
+	thr->id = work_thr_id;
+	thr->q = tq_new();
+	if (!thr->q)
+		return 1;
+
+	/* start work I/O thread */
+	if (pthread_create(&thr->pth, NULL, workio_thread, thr)) {
+		applog(LOG_ERR, "workio thread create failed");
+		return 1;
+	}
+
+	if (want_longpoll && !have_stratum) {
+		/* init longpoll thread info */
+		longpoll_thr_id = opt_n_threads + 1;
+		thr = &thr_info[longpoll_thr_id];
+		thr->id = longpoll_thr_id;
+		thr->q = tq_new();
+		if (!thr->q)
+			return 1;
+
+		/* start longpoll thread */
+		if (unlikely(pthread_create(&thr->pth, NULL, longpoll_thread, thr))) {
+			applog(LOG_ERR, "longpoll thread create failed");
+			return 1;
+		}
+	}
+	if (want_stratum) {
+		/* init stratum thread info */
+		stratum_thr_id = opt_n_threads + 2;
+		thr = &thr_info[stratum_thr_id];
+		thr->id = stratum_thr_id;
+		thr->q = tq_new();
+		if (!thr->q)
+			return 1;
+
+		/* start stratum thread */
+		if (unlikely(pthread_create(&thr->pth, NULL, stratum_thread, thr))) {
+			applog(LOG_ERR, "stratum thread create failed");
+			return 1;
+		}
+
+		if (have_stratum)
+			tq_push(thr_info[stratum_thr_id].q, strdup(rpc_url));
+	}
+
+	/* start mining threads */
+	for (i = 0; i < opt_n_threads; i++) {
+		thr = &thr_info[i];
+
+		thr->id = i;
+		thr->q = tq_new();
+		if (!thr->q)
+			return 1;
+
+		if (unlikely(pthread_create(&thr->pth, NULL, miner_thread, thr))) {
+			applog(LOG_ERR, "thread %d create failed", i);
+			return 1;
+		}
+	}
+
+	applog(LOG_INFO, "%d miner threads started, "
+		"using '%s' algorithm.",
+		opt_n_threads,
+		algo_names[opt_algo]);
+
+	/* main loop - simply wait for workio thread to exit */
+	pthread_join(thr_info[work_thr_id].pth, NULL);
+
+	applog(LOG_INFO, "workio thread dead, exiting.");
+
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/cpuminer-config.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/cpuminer-config.h
new file mode 100644
index 0000000..4279603
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/cpuminer-config.h
@@ -0,0 +1,196 @@
+/* cpuminer-config.h.  Generated from cpuminer-config.h.in by configure.  */
+/* cpuminer-config.h.in.  Generated from configure.ac by autoheader.  */
+
+/* Define to one of `_getb67', `GETB67', `getb67' for Cray-2 and Cray-YMP
+   systems. This function is required for `alloca.c' support on those systems.
+   */
+/* #undef CRAY_STACKSEG_END */
+
+/* Define to 1 if using `alloca.c'. */
+/* #undef C_ALLOCA */
+
+/* Define to 1 if you have `alloca', as a function or macro. */
+#define HAVE_ALLOCA 1
+
+/* Define to 1 if you have <alloca.h> and it should be used (not on Ultrix).
+   */
+#define HAVE_ALLOCA_H 1
+
+/* Define to 1 if you have the declaration of `be32dec', and to 0 if you
+   don't. */
+#define HAVE_DECL_BE32DEC 0
+
+/* Define to 1 if you have the declaration of `be32enc', and to 0 if you
+   don't. */
+#define HAVE_DECL_BE32ENC 0
+
+/* Define to 1 if you have the declaration of `le32dec', and to 0 if you
+   don't. */
+#define HAVE_DECL_LE32DEC 0
+
+/* Define to 1 if you have the declaration of `le32enc', and to 0 if you
+   don't. */
+#define HAVE_DECL_LE32ENC 0
+
+/* Define to 1 if you have the `getopt_long' function. */
+#define HAVE_GETOPT_LONG 1
+
+/* Define to 1 if you have the <inttypes.h> header file. */
+#define HAVE_INTTYPES_H 1
+
+/* Define to 1 if you have a functional curl library. */
+#define HAVE_LIBCURL 1
+
+/* Define to 1 if you have the <memory.h> header file. */
+#define HAVE_MEMORY_H 1
+
+/* Define to 1 if you have the <stdint.h> header file. */
+#define HAVE_STDINT_H 1
+
+/* Define to 1 if you have the <stdlib.h> header file. */
+#define HAVE_STDLIB_H 1
+
+/* Define to 1 if you have the <strings.h> header file. */
+#define HAVE_STRINGS_H 1
+
+/* Define to 1 if you have the <string.h> header file. */
+#define HAVE_STRING_H 1
+
+/* Define to 1 if you have the <syslog.h> header file. */
+#define HAVE_SYSLOG_H 0
+
+/* Define to 1 if you have the <sys/endian.h> header file. */
+#define HAVE_SYS_ENDIAN_H 1
+
+/* Define to 1 if you have the <sys/param.h> header file. */
+#define HAVE_SYS_PARAM_H 1
+
+/* Define to 1 if you have the <sys/stat.h> header file. */
+#define HAVE_SYS_STAT_H 1
+
+/* Define to 1 if you have the <sys/sysctl.h> header file. */
+/* #undef HAVE_SYS_SYSCTL_H */
+
+/* Define to 1 if you have the <sys/types.h> header file. */
+#define HAVE_SYS_TYPES_H 1
+
+/* Define to 1 if you have the <unistd.h> header file. */
+#define HAVE_UNISTD_H 1
+
+/* Defined if libcurl supports AsynchDNS */
+#define LIBCURL_FEATURE_ASYNCHDNS 1
+
+/* Defined if libcurl supports IDN */
+/* #undef LIBCURL_FEATURE_IDN */
+
+/* Defined if libcurl supports IPv6 */
+#define LIBCURL_FEATURE_IPV6 1
+
+/* Defined if libcurl supports KRB4 */
+/* #undef LIBCURL_FEATURE_KRB4 */
+
+/* Defined if libcurl supports libz */
+#define LIBCURL_FEATURE_LIBZ 1
+
+/* Defined if libcurl supports NTLM */
+/* #undef LIBCURL_FEATURE_NTLM */
+
+/* Defined if libcurl supports SSL */
+/* #undef LIBCURL_FEATURE_SSL */
+
+/* Defined if libcurl supports SSPI */
+/* #undef LIBCURL_FEATURE_SSPI */
+
+/* Defined if libcurl supports DICT */
+#define LIBCURL_PROTOCOL_DICT 1
+
+/* Defined if libcurl supports FILE */
+#define LIBCURL_PROTOCOL_FILE 1
+
+/* Defined if libcurl supports FTP */
+#define LIBCURL_PROTOCOL_FTP 1
+
+/* Defined if libcurl supports FTPS */
+/* #undef LIBCURL_PROTOCOL_FTPS */
+
+/* Defined if libcurl supports HTTP */
+#define LIBCURL_PROTOCOL_HTTP 1
+
+/* Defined if libcurl supports HTTPS */
+/* #undef LIBCURL_PROTOCOL_HTTPS */
+
+/* Defined if libcurl supports IMAP */
+#define LIBCURL_PROTOCOL_IMAP 1
+
+/* Defined if libcurl supports LDAP */
+/* #undef LIBCURL_PROTOCOL_LDAP */
+
+/* Defined if libcurl supports POP3 */
+#define LIBCURL_PROTOCOL_POP3 1
+
+/* Defined if libcurl supports RTSP */
+#define LIBCURL_PROTOCOL_RTSP 1
+
+/* Defined if libcurl supports SMTP */
+#define LIBCURL_PROTOCOL_SMTP 1
+
+/* Defined if libcurl supports TELNET */
+#define LIBCURL_PROTOCOL_TELNET 1
+
+/* Defined if libcurl supports TFTP */
+#define LIBCURL_PROTOCOL_TFTP 1
+
+/* Name of package */
+#define PACKAGE "cpuminer"
+
+/* Define to the address where bug reports for this package should be sent. */
+#define PACKAGE_BUGREPORT ""
+
+/* Define to the full name of this package. */
+#define PACKAGE_NAME "cpuminer"
+
+/* Define to the full name and version of this package. */
+#define PACKAGE_STRING "cpuminer 2.4"
+
+/* Define to the one symbol short name of this package. */
+#define PACKAGE_TARNAME "cpuminer"
+
+/* Define to the home page for this package. */
+#define PACKAGE_URL ""
+
+/* Define to the version of this package. */
+#define PACKAGE_VERSION "2.4"
+
+/* If using the C implementation of alloca, define if you know the
+   direction of stack growth for your system; otherwise it will be
+   automatically deduced at runtime.
+	STACK_DIRECTION > 0 => grows toward higher addresses
+	STACK_DIRECTION < 0 => grows toward lower addresses
+	STACK_DIRECTION = 0 => direction of growth unknown */
+/* #undef STACK_DIRECTION */
+
+/* Define to 1 if you have the ANSI C header files. */
+#define STDC_HEADERS 1
+
+/* Define to 1 if assembly routines are wanted. */
+#if !defined(__APPLE__)
+#define USE_ASM 1
+#endif // __APPLE__ or __i386__
+
+/* Define to 1 if AVX assembly is available. */
+/* #undef USE_AVX */
+
+/* Define to 1 if AVX2 assembly is available. */
+/* #undef USE_AVX2 */
+
+/* Define to 1 if XOP assembly is available. */
+/* #undef USE_XOP */
+
+/* Version number of package */
+#define VERSION "2.4"
+
+/* Define curl_free() as free() if our version of curl lacks curl_free. */
+/* #undef curl_free */
+
+/* Define to `unsigned int' if <sys/types.h> does not define. */
+/* #undef size_t */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/deps/build_win_deps.sh b/bitzenymininglibrary/src/main/cpp/libcpuminer/deps/build_win_deps.sh
new file mode 100644
index 0000000..5ef7ccf
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/deps/build_win_deps.sh
@@ -0,0 +1,36 @@
+#!/bin/bash
+set -e
+PREFIX=${PWD}/x86_64-w64-mingw32
+
+CURL_PACKAGE=curl-7.54.1
+CURL_PACKAGE_FILE=${CURL_PACKAGE}.tar.gz
+CURL_PACKAGE_FILE_SHA256=cd404b808b253512dafec4fed0fb2cc98370d818a7991826c3021984fc27f9d0
+CURL_CHECKSUM_FILE=${CURL_PACKAGE_FILE}.sha256
+
+PT_PACKAGE=pthreads-w32-2-9-1-release
+PT_PACKAGE_FILE=${PT_PACKAGE}.tar.gz
+PT_PACKAGE_FILE_SHA256=e6aca7aea8de33d9c8580bcb3a0ea3ec0a7ace4ba3f4e263ac7c7b66bc95fb4d
+PT_CHECKSUM_FILE=${PT_PACKAGE_FILE_SHA256}.sha256
+
+wget https://curl.haxx.se/download/$CURL_PACKAGE_FILE -O $CURL_PACKAGE_FILE
+echo "${CURL_PACKAGE_FILE_SHA256}  ${CURL_PACKAGE_FILE}" > $CURL_CHECKSUM_FILE
+sha256sum -c $CURL_CHECKSUM_FILE
+rm $CURL_CHECKSUM_FILE
+
+wget ftp://sourceware.org/pub/pthreads-win32/$PT_PACKAGE_FILE -O $PT_PACKAGE_FILE
+echo "${PT_PACKAGE_FILE_SHA256}  ${PT_PACKAGE_FILE}" > $PT_CHECKSUM_FILE
+sha256sum -c $PT_CHECKSUM_FILE
+rm $PT_CHECKSUM_FILE
+
+tar zxvf $CURL_PACKAGE_FILE
+tar zxvf $PT_PACKAGE_FILE
+
+cd $CURL_PACKAGE
+./configure --host=x86_64-w64-mingw32 --disable-shared --enable-static --with-winssl --prefix=$PREFIX
+make install
+
+cd ../${PT_PACKAGE}/
+cp config.h pthreads_win32_config.h
+make -f GNUmakefile CROSS="x86_64-w64-mingw32-" clean GC-static
+cp libpthreadGC2.a ${PREFIX}/lib/libpthread.a
+cp pthread.h semaphore.h sched.h ${PREFIX}/include
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/elist.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/elist.h
new file mode 100644
index 0000000..5703c39
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/elist.h
@@ -0,0 +1,254 @@
+#ifndef _LINUX_LIST_H
+#define _LINUX_LIST_H
+
+/*
+ * Simple doubly linked list implementation.
+ *
+ * Some of the internal functions ("__xxx") are useful when
+ * manipulating whole lists rather than single entries, as
+ * sometimes we already know the next/prev entries and we can
+ * generate better code by using them directly rather than
+ * using the generic single-entry routines.
+ */
+
+struct list_head {
+	struct list_head *next, *prev;
+};
+
+#define LIST_HEAD_INIT(name) { &(name), &(name) }
+
+#define LIST_HEAD(name) \
+	struct list_head name = LIST_HEAD_INIT(name)
+
+#define INIT_LIST_HEAD(ptr) do { \
+	(ptr)->next = (ptr); (ptr)->prev = (ptr); \
+} while (0)
+
+/*
+ * Insert a new entry between two known consecutive entries.
+ *
+ * This is only for internal list manipulation where we know
+ * the prev/next entries already!
+ */
+static inline void __list_add(struct list_head *new,
+			      struct list_head *prev,
+			      struct list_head *next)
+{
+	next->prev = new;
+	new->next = next;
+	new->prev = prev;
+	prev->next = new;
+}
+
+/**
+ * list_add - add a new entry
+ * @new: new entry to be added
+ * @head: list head to add it after
+ *
+ * Insert a new entry after the specified head.
+ * This is good for implementing stacks.
+ */
+static inline void list_add(struct list_head *new, struct list_head *head)
+{
+	__list_add(new, head, head->next);
+}
+
+/**
+ * list_add_tail - add a new entry
+ * @new: new entry to be added
+ * @head: list head to add it before
+ *
+ * Insert a new entry before the specified head.
+ * This is useful for implementing queues.
+ */
+static inline void list_add_tail(struct list_head *new, struct list_head *head)
+{
+	__list_add(new, head->prev, head);
+}
+
+/*
+ * Delete a list entry by making the prev/next entries
+ * point to each other.
+ *
+ * This is only for internal list manipulation where we know
+ * the prev/next entries already!
+ */
+static inline void __list_del(struct list_head *prev, struct list_head *next)
+{
+	next->prev = prev;
+	prev->next = next;
+}
+
+/**
+ * list_del - deletes entry from list.
+ * @entry: the element to delete from the list.
+ * Note: list_empty on entry does not return true after this, the entry is in an undefined state.
+ */
+static inline void list_del(struct list_head *entry)
+{
+	__list_del(entry->prev, entry->next);
+	entry->next = (void *) 0;
+	entry->prev = (void *) 0;
+}
+
+/**
+ * list_del_init - deletes entry from list and reinitialize it.
+ * @entry: the element to delete from the list.
+ */
+static inline void list_del_init(struct list_head *entry)
+{
+	__list_del(entry->prev, entry->next);
+	INIT_LIST_HEAD(entry);
+}
+
+/**
+ * list_move - delete from one list and add as another's head
+ * @list: the entry to move
+ * @head: the head that will precede our entry
+ */
+static inline void list_move(struct list_head *list, struct list_head *head)
+{
+        __list_del(list->prev, list->next);
+        list_add(list, head);
+}
+
+/**
+ * list_move_tail - delete from one list and add as another's tail
+ * @list: the entry to move
+ * @head: the head that will follow our entry
+ */
+static inline void list_move_tail(struct list_head *list,
+				  struct list_head *head)
+{
+        __list_del(list->prev, list->next);
+        list_add_tail(list, head);
+}
+
+/**
+ * list_empty - tests whether a list is empty
+ * @head: the list to test.
+ */
+static inline int list_empty(struct list_head *head)
+{
+	return head->next == head;
+}
+
+static inline void __list_splice(struct list_head *list,
+				 struct list_head *head)
+{
+	struct list_head *first = list->next;
+	struct list_head *last = list->prev;
+	struct list_head *at = head->next;
+
+	first->prev = head;
+	head->next = first;
+
+	last->next = at;
+	at->prev = last;
+}
+
+/**
+ * list_splice - join two lists
+ * @list: the new list to add.
+ * @head: the place to add it in the first list.
+ */
+static inline void list_splice(struct list_head *list, struct list_head *head)
+{
+	if (!list_empty(list))
+		__list_splice(list, head);
+}
+
+/**
+ * list_splice_init - join two lists and reinitialise the emptied list.
+ * @list: the new list to add.
+ * @head: the place to add it in the first list.
+ *
+ * The list at @list is reinitialised
+ */
+static inline void list_splice_init(struct list_head *list,
+				    struct list_head *head)
+{
+	if (!list_empty(list)) {
+		__list_splice(list, head);
+		INIT_LIST_HEAD(list);
+	}
+}
+
+/**
+ * list_entry - get the struct for this entry
+ * @ptr:	the &struct list_head pointer.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the list_struct within the struct.
+ */
+#define list_entry(ptr, type, member) \
+	((type *)((char *)(ptr)-(unsigned long)(&((type *)0)->member)))
+
+/**
+ * list_for_each	-	iterate over a list
+ * @pos:	the &struct list_head to use as a loop counter.
+ * @head:	the head for your list.
+ */
+#define list_for_each(pos, head) \
+	for (pos = (head)->next; pos != (head); \
+        	pos = pos->next)
+/**
+ * list_for_each_prev	-	iterate over a list backwards
+ * @pos:	the &struct list_head to use as a loop counter.
+ * @head:	the head for your list.
+ */
+#define list_for_each_prev(pos, head) \
+	for (pos = (head)->prev; pos != (head); \
+        	pos = pos->prev)
+
+/**
+ * list_for_each_safe	-	iterate over a list safe against removal of list entry
+ * @pos:	the &struct list_head to use as a loop counter.
+ * @n:		another &struct list_head to use as temporary storage
+ * @head:	the head for your list.
+ */
+#define list_for_each_safe(pos, n, head) \
+	for (pos = (head)->next, n = pos->next; pos != (head); \
+		pos = n, n = pos->next)
+
+/**
+ * list_for_each_entry	-	iterate over list of given type
+ * @pos:	the type * to use as a loop counter.
+ * @head:	the head for your list.
+ * @member:	the name of the list_struct within the struct.
+ * @type:	the type of the struct.
+ */
+#define list_for_each_entry(pos, head, member, type)			\
+	for (pos = list_entry((head)->next, type, member);	\
+	     &pos->member != (head); 					\
+	     pos = list_entry(pos->member.next, type, member))
+
+/**
+ * list_for_each_entry_safe - iterate over list of given type safe against removal of list entry
+ * @pos:	the type * to use as a loop counter.
+ * @n:		another type * to use as temporary storage
+ * @head:	the head for your list.
+ * @member:	the name of the list_struct within the struct.
+ * @type:	the type of the struct.
+ */
+#define list_for_each_entry_safe(pos, n, head, member, type)		\
+	for (pos = list_entry((head)->next, type, member),	\
+		n = list_entry(pos->member.next, type, member);	\
+	     &pos->member != (head); 					\
+	     pos = n, n = list_entry(n->member.next, type, member))
+
+/**
+ * list_for_each_entry_continue -       iterate over list of given type
+ *                      continuing after existing point
+ * @pos:        the type * to use as a loop counter.
+ * @head:       the head for your list.
+ * @member:     the name of the list_struct within the struct.
+ * @type:       the type of the struct.
+ */
+#define list_for_each_entry_continue(pos, head, member, type)		\
+	for (pos = list_entry(pos->member.next, type, member),	\
+		     prefetch(pos->member.next);			\
+	     &pos->member != (head);					\
+	     pos = list_entry(pos->member.next, type, member),	\
+		     prefetch(pos->member.next))
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/example-cfg.json b/bitzenymininglibrary/src/main/cpp/libcpuminer/example-cfg.json
new file mode 100644
index 0000000..228a66d
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/example-cfg.json
@@ -0,0 +1,13 @@
+{
+	"_comment1" : "Any long-format command line argument ",
+	"_comment2" : "may be used in this JSON configuration file",
+
+	"url" : "http://127.0.0.1:9332/",
+	"user" : "rpcuser",
+	"pass" : "rpcpass",
+
+	"algo" : "scrypt",
+	"threads" : "4",
+
+	"quiet" : true
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/libcpuminer.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/libcpuminer.c
new file mode 100644
index 0000000..33f8b26
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/libcpuminer.c
@@ -0,0 +1,1609 @@
+/*
+ * Copyright 2010 Jeff Garzik
+ * Copyright 2012-2017 pooler
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+#define _GNU_SOURCE
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdbool.h>
+#include <inttypes.h>
+#include <unistd.h>
+#include <sys/time.h>
+#include <time.h>
+
+#ifdef __KITKAT__
+#include <sys/limits.h>
+#endif // __KITKAT__
+
+#ifdef WIN32
+#include <windows.h>
+#else
+#include <errno.h>
+#include <signal.h>
+#include <sys/resource.h>
+#if HAVE_SYS_SYSCTL_H
+#include <sys/types.h>
+#if HAVE_SYS_PARAM_H
+#include <sys/param.h>
+#endif
+#include <linux/sysctl.h>
+#endif
+#endif
+#include <jansson.h>
+#include <curl/curl.h>
+#include "compat.h"
+#include "miner.h"
+
+#define PROGRAM_NAME		"minerd"
+#define LP_SCANTIME		60
+
+#ifdef __linux /* Linux specific policy and affinity management */
+#include <sched.h>
+static inline void drop_policy(void)
+{
+	struct sched_param param;
+	param.sched_priority = 0;
+
+#ifdef SCHED_IDLE
+	if (unlikely(sched_setscheduler(0, SCHED_IDLE, &param) == -1))
+#endif
+#ifdef SCHED_BATCH
+		sched_setscheduler(0, SCHED_BATCH, &param);
+#endif
+}
+
+#ifndef __KITKAT__
+#ifndef __APPLE__
+static inline void affine_to_cpu(int id, int cpu)
+{
+	cpu_set_t set;
+
+	CPU_ZERO(&set);
+	CPU_SET(cpu, &set);
+	sched_setaffinity(0, sizeof(set), &set);
+}
+#endif // __APPLE__
+#endif // __KITKAT__
+
+#elif defined(__FreeBSD__) /* FreeBSD specific policy and affinity management */
+#include <sys/cpuset.h>
+static inline void drop_policy(void)
+{
+}
+
+static inline void affine_to_cpu(int id, int cpu)
+{
+    cpuset_t set;
+    CPU_ZERO(&set);
+    CPU_SET(cpu, &set);
+    cpuset_setaffinity(CPU_LEVEL_WHICH, CPU_WHICH_TID, -1, sizeof(cpuset_t), &set);
+}
+#endif
+		
+enum workio_commands {
+	WC_GET_WORK,
+	WC_SUBMIT_WORK,
+};
+
+struct workio_cmd {
+	enum workio_commands	cmd;
+	struct thr_info		*thr;
+	union {
+		struct work	*work;
+	} u;
+};
+
+enum algos {
+	ALGO_YESCRYPT,
+	ALGO_YESPOWER,
+	ALGO_SCRYPT,		/* scrypt(1024,1,1) */
+	ALGO_SHA256D,		/* SHA-256d */
+};
+
+static const char *algo_names[] = {
+	[ALGO_YESCRYPT]		= "yescrypt",
+	[ALGO_YESPOWER]		= "yespower",
+	[ALGO_SCRYPT]		= "scrypt",
+	[ALGO_SHA256D]		= "sha256d",
+};
+
+bool opt_debug = false;
+bool opt_protocol = false;
+static bool opt_benchmark = false;
+bool opt_redirect = true;
+bool want_longpoll = true;
+bool have_longpoll = false;
+bool have_gbt = true;
+bool allow_getwork = true;
+bool want_stratum = true;
+bool have_stratum = false;
+bool use_syslog = false;
+static bool opt_quiet = false;
+static int opt_retries = -1;
+static int opt_fail_pause = 30;
+int opt_timeout = 0;
+static int opt_scantime = 5;
+static enum algos opt_algo = ALGO_YESCRYPT;
+static int opt_scrypt_n = 1024;
+static int opt_n_threads;
+static int num_processors;
+static char *rpc_url;
+static char *rpc_userpass;
+static char *rpc_user, *rpc_pass;
+static int pk_script_size;
+static unsigned char pk_script[25];
+static char coinbase_sig[101] = "";
+char *opt_cert;
+char *opt_proxy;
+long opt_proxy_type;
+struct thr_info *thr_info;
+static int work_thr_id;
+int longpoll_thr_id = -1;
+int stratum_thr_id = -1;
+struct work_restart *work_restart = NULL;
+static struct stratum_ctx stratum;
+
+pthread_mutex_t applog_lock;
+pthread_mutex_t run_lock;
+static pthread_mutex_t stats_lock;
+
+static unsigned long accepted_count = 0L;
+static unsigned long rejected_count = 0L;
+static double *thr_hashrates;
+
+extern void (*vprintf_cb)(const char *format, va_list arg);
+
+struct work {
+	uint32_t data[32];
+	uint32_t target[8];
+
+	int height;
+	char *txs;
+	char *workid;
+
+	char *job_id;
+	size_t xnonce2_len;
+	unsigned char *xnonce2;
+};
+
+static struct work g_work;
+static time_t g_work_time;
+static pthread_mutex_t g_work_lock;
+static bool submit_old = false;
+static char *lp_id;
+
+static inline void work_free(struct work *w)
+{
+	free(w->txs);
+	free(w->workid);
+	free(w->job_id);
+	free(w->xnonce2);
+}
+
+static inline void work_copy(struct work *dest, const struct work *src)
+{
+	memcpy(dest, src, sizeof(struct work));
+	if (src->txs)
+		dest->txs = strdup(src->txs);
+	if (src->workid)
+		dest->workid = strdup(src->workid);
+	if (src->job_id)
+		dest->job_id = strdup(src->job_id);
+	if (src->xnonce2) {
+		dest->xnonce2 = malloc(src->xnonce2_len);
+		memcpy(dest->xnonce2, src->xnonce2, src->xnonce2_len);
+	}
+}
+
+static bool jobj_binary(const json_t *obj, const char *key,
+			void *buf, size_t buflen)
+{
+	const char *hexstr;
+	json_t *tmp;
+
+	tmp = json_object_get(obj, key);
+	if (unlikely(!tmp)) {
+		applog(LOG_ERR, "JSON key '%s' not found", key);
+		return false;
+	}
+	hexstr = json_string_value(tmp);
+	if (unlikely(!hexstr)) {
+		applog(LOG_ERR, "JSON key '%s' is not a string", key);
+		return false;
+	}
+	if (!hex2bin(buf, hexstr, buflen))
+		return false;
+
+	return true;
+}
+
+static bool work_decode(const json_t *val, struct work *work)
+{
+	int i;
+
+	if (unlikely(!jobj_binary(val, "data", work->data, sizeof(work->data)))) {
+		applog(LOG_ERR, "JSON invalid data");
+		goto err_out;
+	}
+	if (unlikely(!jobj_binary(val, "target", work->target, sizeof(work->target)))) {
+		applog(LOG_ERR, "JSON invalid target");
+		goto err_out;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(work->data); i++)
+		work->data[i] = le32dec(work->data + i);
+	for (i = 0; i < ARRAY_SIZE(work->target); i++)
+		work->target[i] = le32dec(work->target + i);
+
+	return true;
+
+err_out:
+	return false;
+}
+
+static bool gbt_work_decode(const json_t *val, struct work *work)
+{
+	int i, n;
+	uint32_t version, curtime, bits;
+	uint32_t prevhash[8];
+	uint32_t target[8];
+	int cbtx_size;
+	unsigned char *cbtx = NULL;
+	int tx_count, tx_size;
+	unsigned char txc_vi[9];
+	unsigned char (*merkle_tree)[32] = NULL;
+	bool coinbase_append = false;
+	bool submit_coinbase = false;
+	bool segwit = false;
+	json_t *tmp, *txa;
+	bool rc = false;
+
+	tmp = json_object_get(val, "rules");
+	if (tmp && json_is_array(tmp)) {
+		n = json_array_size(tmp);
+		for (i = 0; i < n; i++) {
+			const char *s = json_string_value(json_array_get(tmp, i));
+			if (!s)
+				continue;
+			if (!strcmp(s, "segwit") || !strcmp(s, "!segwit"))
+				segwit = true;
+		}
+	}
+
+	tmp = json_object_get(val, "mutable");
+	if (tmp && json_is_array(tmp)) {
+		n = json_array_size(tmp);
+		for (i = 0; i < n; i++) {
+			const char *s = json_string_value(json_array_get(tmp, i));
+			if (!s)
+				continue;
+			if (!strcmp(s, "coinbase/append"))
+				coinbase_append = true;
+			else if (!strcmp(s, "submit/coinbase"))
+				submit_coinbase = true;
+		}
+	}
+
+	tmp = json_object_get(val, "height");
+	if (!tmp || !json_is_integer(tmp)) {
+		applog(LOG_ERR, "JSON invalid height");
+		goto out;
+	}
+	work->height = json_integer_value(tmp);
+
+	tmp = json_object_get(val, "version");
+	if (!tmp || !json_is_integer(tmp)) {
+		applog(LOG_ERR, "JSON invalid version");
+		goto out;
+	}
+	version = json_integer_value(tmp);
+
+	if (unlikely(!jobj_binary(val, "previousblockhash", prevhash, sizeof(prevhash)))) {
+		applog(LOG_ERR, "JSON invalid previousblockhash");
+		goto out;
+	}
+
+	tmp = json_object_get(val, "curtime");
+	if (!tmp || !json_is_integer(tmp)) {
+		applog(LOG_ERR, "JSON invalid curtime");
+		goto out;
+	}
+	curtime = json_integer_value(tmp);
+
+	if (unlikely(!jobj_binary(val, "bits", &bits, sizeof(bits)))) {
+		applog(LOG_ERR, "JSON invalid bits");
+		goto out;
+	}
+
+	/* find count and size of transactions */
+	txa = json_object_get(val, "transactions");
+	if (!txa || !json_is_array(txa)) {
+		applog(LOG_ERR, "JSON invalid transactions");
+		goto out;
+	}
+	tx_count = json_array_size(txa);
+	tx_size = 0;
+	for (i = 0; i < tx_count; i++) {
+		const json_t *tx = json_array_get(txa, i);
+		const char *tx_hex = json_string_value(json_object_get(tx, "data"));
+		if (!tx_hex) {
+			applog(LOG_ERR, "JSON invalid transactions");
+			goto out;
+		}
+		tx_size += strlen(tx_hex) / 2;
+	}
+
+	/* build coinbase transaction */
+	tmp = json_object_get(val, "coinbasetxn");
+	if (tmp) {
+		const char *cbtx_hex = json_string_value(json_object_get(tmp, "data"));
+		cbtx_size = cbtx_hex ? strlen(cbtx_hex) / 2 : 0;
+		cbtx = malloc(cbtx_size + 100);
+		if (cbtx_size < 60 || !hex2bin(cbtx, cbtx_hex, cbtx_size)) {
+			applog(LOG_ERR, "JSON invalid coinbasetxn");
+			goto out;
+		}
+	} else {
+		int64_t cbvalue;
+		if (!pk_script_size) {
+			if (allow_getwork) {
+				applog(LOG_INFO, "No payout address provided, switching to getwork");
+				have_gbt = false;
+			} else
+				applog(LOG_ERR, "No payout address provided");
+			goto out;
+		}
+		tmp = json_object_get(val, "coinbasevalue");
+		if (!tmp || !json_is_number(tmp)) {
+			applog(LOG_ERR, "JSON invalid coinbasevalue");
+			goto out;
+		}
+		cbvalue = json_is_integer(tmp) ? json_integer_value(tmp) : json_number_value(tmp);
+		cbtx = malloc(256);
+		le32enc((uint32_t *)cbtx, 1); /* version */
+		cbtx[4] = 1; /* in-counter */
+		memset(cbtx+5, 0x00, 32); /* prev txout hash */
+		le32enc((uint32_t *)(cbtx+37), 0xffffffff); /* prev txout index */
+		cbtx_size = 43;
+		/* BIP 34: height in coinbase */
+		for (n = work->height; n; n >>= 8) {
+			cbtx[cbtx_size++] = n & 0xff;
+			if (n < 0x100 && n >= 0x80)
+				cbtx[cbtx_size++] = 0;
+		}
+		cbtx[42] = cbtx_size - 43;
+		cbtx[41] = cbtx_size - 42; /* scriptsig length */
+		le32enc((uint32_t *)(cbtx+cbtx_size), 0xffffffff); /* sequence */
+		cbtx_size += 4;
+		cbtx[cbtx_size++] = segwit ? 2 : 1; /* out-counter */
+		le32enc((uint32_t *)(cbtx+cbtx_size), (uint32_t)cbvalue); /* value */
+		le32enc((uint32_t *)(cbtx+cbtx_size+4), cbvalue >> 32);
+		cbtx_size += 8;
+		cbtx[cbtx_size++] = pk_script_size; /* txout-script length */
+		memcpy(cbtx+cbtx_size, pk_script, pk_script_size);
+		cbtx_size += pk_script_size;
+		if (segwit) {
+			unsigned char (*wtree)[32] = calloc(tx_count + 2, 32);
+			memset(cbtx+cbtx_size, 0, 8); /* value */
+			cbtx_size += 8;
+			cbtx[cbtx_size++] = 38; /* txout-script length */
+			cbtx[cbtx_size++] = 0x6a; /* txout-script */
+			cbtx[cbtx_size++] = 0x24;
+			cbtx[cbtx_size++] = 0xaa;
+			cbtx[cbtx_size++] = 0x21;
+			cbtx[cbtx_size++] = 0xa9;
+			cbtx[cbtx_size++] = 0xed;
+			for (i = 0; i < tx_count; i++) {
+				const json_t *tx = json_array_get(txa, i);
+				const json_t *hash = json_object_get(tx, "hash");
+				if (!hash || !hex2bin(wtree[1+i], json_string_value(hash), 32)) {
+					applog(LOG_ERR, "JSON invalid transaction hash");
+					free(wtree);
+					goto out;
+				}
+				memrev(wtree[1+i], 32);
+			}
+			n = tx_count + 1;
+			while (n > 1) {
+				if (n % 2)
+					memcpy(wtree[n], wtree[n-1], 32);
+				n = (n + 1) / 2;
+				for (i = 0; i < n; i++)
+					sha256d(wtree[i], wtree[2*i], 64);
+			}
+			memset(wtree[1], 0, 32);  /* witness reserved value = 0 */
+			sha256d(cbtx+cbtx_size, wtree[0], 64);
+			cbtx_size += 32;
+			free(wtree);
+		}
+		le32enc((uint32_t *)(cbtx+cbtx_size), 0); /* lock time */
+		cbtx_size += 4;
+		coinbase_append = true;
+	}
+	if (coinbase_append) {
+		unsigned char xsig[100];
+		int xsig_len = 0;
+		if (*coinbase_sig) {
+			n = strlen(coinbase_sig);
+			if (cbtx[41] + xsig_len + n <= 100) {
+				memcpy(xsig+xsig_len, coinbase_sig, n);
+				xsig_len += n;
+			} else {
+				applog(LOG_WARNING, "Signature does not fit in coinbase, skipping");
+			}
+		}
+		tmp = json_object_get(val, "coinbaseaux");
+		if (tmp && json_is_object(tmp)) {
+			void *iter = json_object_iter(tmp);
+			while (iter) {
+				unsigned char buf[100];
+				const char *s = json_string_value(json_object_iter_value(iter));
+				n = s ? strlen(s) / 2 : 0;
+				if (!s || n > 100 || !hex2bin(buf, s, n)) {
+					applog(LOG_ERR, "JSON invalid coinbaseaux");
+					break;
+				}
+				if (cbtx[41] + xsig_len + n <= 100) {
+					memcpy(xsig+xsig_len, buf, n);
+					xsig_len += n;
+				}
+				iter = json_object_iter_next(tmp, iter);
+			}
+		}
+		if (xsig_len) {
+			unsigned char *ssig_end = cbtx + 42 + cbtx[41];
+			int push_len = cbtx[41] + xsig_len < 76 ? 1 :
+			               cbtx[41] + 2 + xsig_len > 100 ? 0 : 2;
+			n = xsig_len + push_len;
+			memmove(ssig_end + n, ssig_end, cbtx_size - 42 - cbtx[41]);
+			cbtx[41] += n;
+			if (push_len == 2)
+				*(ssig_end++) = 0x4c; /* OP_PUSHDATA1 */
+			if (push_len)
+				*(ssig_end++) = xsig_len;
+			memcpy(ssig_end, xsig, xsig_len);
+			cbtx_size += n;
+		}
+	}
+
+	n = varint_encode(txc_vi, 1 + tx_count);
+	work->txs = malloc(2 * (n + cbtx_size + tx_size) + 1);
+	bin2hex(work->txs, txc_vi, n);
+	bin2hex(work->txs + 2*n, cbtx, cbtx_size);
+
+	/* generate merkle root */
+	merkle_tree = malloc(32 * ((1 + tx_count + 1) & ~1));
+	sha256d(merkle_tree[0], cbtx, cbtx_size);
+	for (i = 0; i < tx_count; i++) {
+		tmp = json_array_get(txa, i);
+		const char *tx_hex = json_string_value(json_object_get(tmp, "data"));
+		const int tx_size = tx_hex ? strlen(tx_hex) / 2 : 0;
+		if (segwit) {
+			const char *txid = json_string_value(json_object_get(tmp, "txid"));
+			if (!txid || !hex2bin(merkle_tree[1 + i], txid, 32)) {
+				applog(LOG_ERR, "JSON invalid transaction txid");
+				goto out;
+			}
+			memrev(merkle_tree[1 + i], 32);
+		} else {
+			unsigned char *tx = malloc(tx_size);
+			if (!tx_hex || !hex2bin(tx, tx_hex, tx_size)) {
+				applog(LOG_ERR, "JSON invalid transactions");
+				free(tx);
+				goto out;
+			}
+			sha256d(merkle_tree[1 + i], tx, tx_size);
+			free(tx);
+		}
+		if (!submit_coinbase)
+			strcat(work->txs, tx_hex);
+	}
+	n = 1 + tx_count;
+	while (n > 1) {
+		if (n % 2) {
+			memcpy(merkle_tree[n], merkle_tree[n-1], 32);
+			++n;
+		}
+		n /= 2;
+		for (i = 0; i < n; i++)
+			sha256d(merkle_tree[i], merkle_tree[2*i], 64);
+	}
+
+	/* assemble block header */
+	work->data[0] = swab32(version);
+	for (i = 0; i < 8; i++)
+		work->data[8 - i] = le32dec(prevhash + i);
+	for (i = 0; i < 8; i++)
+		work->data[9 + i] = be32dec((uint32_t *)merkle_tree[0] + i);
+	work->data[17] = swab32(curtime);
+	work->data[18] = le32dec(&bits);
+	memset(work->data + 19, 0x00, 52);
+	work->data[20] = 0x80000000;
+	work->data[31] = 0x00000280;
+
+	if (unlikely(!jobj_binary(val, "target", target, sizeof(target)))) {
+		applog(LOG_ERR, "JSON invalid target");
+		goto out;
+	}
+	for (i = 0; i < ARRAY_SIZE(work->target); i++)
+		work->target[7 - i] = be32dec(target + i);
+
+	tmp = json_object_get(val, "workid");
+	if (tmp) {
+		if (!json_is_string(tmp)) {
+			applog(LOG_ERR, "JSON invalid workid");
+			goto out;
+		}
+		work->workid = strdup(json_string_value(tmp));
+	}
+
+	/* Long polling */
+	tmp = json_object_get(val, "longpollid");
+	if (want_longpoll && json_is_string(tmp)) {
+		free(lp_id);
+		lp_id = strdup(json_string_value(tmp));
+		if (!have_longpoll) {
+			char *lp_uri;
+			tmp = json_object_get(val, "longpolluri");
+			lp_uri = strdup(json_is_string(tmp) ? json_string_value(tmp) : rpc_url);
+			have_longpoll = true;
+			tq_push(thr_info[longpoll_thr_id].q, lp_uri);
+		}
+	}
+
+	rc = true;
+
+out:
+	free(cbtx);
+	free(merkle_tree);
+	return rc;
+}
+
+static void share_result(int result, const char *reason)
+{
+	char s[345];
+	double hashrate;
+	int i;
+
+	hashrate = 0.;
+	pthread_mutex_lock(&stats_lock);
+	for (i = 0; i < opt_n_threads; i++)
+		hashrate += thr_hashrates[i];
+	result ? accepted_count++ : rejected_count++;
+	pthread_mutex_unlock(&stats_lock);
+	
+	sprintf(s, hashrate >= 1e3 ? "%.0f" : "%.1f", hashrate);
+	applog(LOG_INFO, "accepted: %lu/%lu (%.2f%%), %s hash/s %s",
+		   accepted_count,
+		   accepted_count + rejected_count,
+		   100. * accepted_count / (accepted_count + rejected_count),
+		   s,
+		   result ? "(yay!!!)" : "(booooo)");
+
+	if (opt_debug && reason)
+		applog(LOG_DEBUG, "DEBUG: reject reason: %s", reason);
+}
+
+static bool submit_upstream_work(CURL *curl, struct work *work)
+{
+	json_t *val, *res, *reason;
+	char data_str[2 * sizeof(work->data) + 1];
+	char s[345];
+	int i;
+	bool rc = false;
+
+	/* pass if the previous hash is not the current previous hash */
+	if (!submit_old && memcmp(work->data + 1, g_work.data + 1, 32)) {
+		if (opt_debug)
+			applog(LOG_DEBUG, "DEBUG: stale work detected, discarding");
+		return true;
+	}
+
+	if (have_stratum) {
+		uint32_t ntime, nonce;
+		char ntimestr[9], noncestr[9], *xnonce2str, *req;
+
+		le32enc(&ntime, work->data[17]);
+		le32enc(&nonce, work->data[19]);
+		bin2hex(ntimestr, (const unsigned char *)(&ntime), 4);
+		bin2hex(noncestr, (const unsigned char *)(&nonce), 4);
+		xnonce2str = abin2hex(work->xnonce2, work->xnonce2_len);
+		req = malloc(256 + strlen(rpc_user) + strlen(work->job_id) + 2 * work->xnonce2_len);
+		sprintf(req,
+			"{\"method\": \"mining.submit\", \"params\": [\"%s\", \"%s\", \"%s\", \"%s\", \"%s\"], \"id\":4}",
+			rpc_user, work->job_id, xnonce2str, ntimestr, noncestr);
+		free(xnonce2str);
+
+		rc = stratum_send_line(&stratum, req);
+		free(req);
+		if (unlikely(!rc)) {
+			applog(LOG_ERR, "submit_upstream_work stratum_send_line failed");
+			goto out;
+		}
+	} else if (work->txs) {
+		char *req;
+
+		for (i = 0; i < ARRAY_SIZE(work->data); i++)
+			be32enc(work->data + i, work->data[i]);
+		bin2hex(data_str, (unsigned char *)work->data, 80);
+		if (work->workid) {
+			char *params;
+			val = json_object();
+			json_object_set_new(val, "workid", json_string(work->workid));
+			params = json_dumps(val, 0);
+			json_decref(val);
+			req = malloc(128 + 2*80 + strlen(work->txs) + strlen(params));
+			sprintf(req,
+				"{\"method\": \"submitblock\", \"params\": [\"%s%s\", %s], \"id\":1}\r\n",
+				data_str, work->txs, params);
+			free(params);
+		} else {
+			req = malloc(128 + 2*80 + strlen(work->txs));
+			sprintf(req,
+				"{\"method\": \"submitblock\", \"params\": [\"%s%s\"], \"id\":1}\r\n",
+				data_str, work->txs);
+		}
+		val = json_rpc_call(curl, rpc_url, rpc_userpass, req, NULL, 0);
+		free(req);
+		if (unlikely(!val)) {
+			applog(LOG_ERR, "submit_upstream_work json_rpc_call failed");
+			goto out;
+		}
+
+		res = json_object_get(val, "result");
+		if (json_is_object(res)) {
+			char *res_str;
+			bool sumres = false;
+			void *iter = json_object_iter(res);
+			while (iter) {
+				if (json_is_null(json_object_iter_value(iter))) {
+					sumres = true;
+					break;
+				}
+				iter = json_object_iter_next(res, iter);
+			}
+			res_str = json_dumps(res, 0);
+			share_result(sumres, res_str);
+			free(res_str);
+		} else
+			share_result(json_is_null(res), json_string_value(res));
+
+		json_decref(val);
+	} else {
+		/* build hex string */
+		for (i = 0; i < ARRAY_SIZE(work->data); i++)
+			le32enc(work->data + i, work->data[i]);
+		bin2hex(data_str, (unsigned char *)work->data, sizeof(work->data));
+
+		/* build JSON-RPC request */
+		sprintf(s,
+			"{\"method\": \"getwork\", \"params\": [ \"%s\" ], \"id\":1}\r\n",
+			data_str);
+
+		/* issue JSON-RPC request */
+		val = json_rpc_call(curl, rpc_url, rpc_userpass, s, NULL, 0);
+		if (unlikely(!val)) {
+			applog(LOG_ERR, "submit_upstream_work json_rpc_call failed");
+			goto out;
+		}
+
+		res = json_object_get(val, "result");
+		reason = json_object_get(val, "reject-reason");
+		share_result(json_is_true(res), reason ? json_string_value(reason) : NULL);
+
+		json_decref(val);
+	}
+
+	rc = true;
+
+out:
+	return rc;
+}
+
+static const char *getwork_req =
+	"{\"method\": \"getwork\", \"params\": [], \"id\":0}\r\n";
+
+#define GBT_CAPABILITIES "[\"coinbasetxn\", \"coinbasevalue\", \"longpoll\", \"workid\"]"
+#define GBT_RULES "[\"segwit\"]"
+
+static const char *gbt_req =
+	"{\"method\": \"getblocktemplate\", \"params\": [{\"capabilities\": "
+	GBT_CAPABILITIES ", \"rules\": " GBT_RULES "}], \"id\":0}\r\n";
+static const char *gbt_lp_req =
+	"{\"method\": \"getblocktemplate\", \"params\": [{\"capabilities\": "
+	GBT_CAPABILITIES ", \"rules\": " GBT_RULES ", \"longpollid\": \"%s\"}], \"id\":0}\r\n";
+
+static bool get_upstream_work(CURL *curl, struct work *work)
+{
+	json_t *val;
+	int err;
+	bool rc;
+	struct timeval tv_start, tv_end, diff;
+
+start:
+	gettimeofday(&tv_start, NULL);
+	val = json_rpc_call(curl, rpc_url, rpc_userpass,
+			    have_gbt ? gbt_req : getwork_req,
+			    &err, have_gbt ? JSON_RPC_QUIET_404 : 0);
+	gettimeofday(&tv_end, NULL);
+
+	if (have_stratum) {
+		if (val)
+			json_decref(val);
+		return true;
+	}
+
+	if (!have_gbt && !allow_getwork) {
+		applog(LOG_ERR, "No usable protocol");
+		if (val)
+			json_decref(val);
+		return false;
+	}
+
+	if (have_gbt && allow_getwork && !val && err == CURLE_OK) {
+		applog(LOG_INFO, "getblocktemplate failed, falling back to getwork");
+		have_gbt = false;
+		goto start;
+	}
+
+	if (!val)
+		return false;
+
+	if (have_gbt) {
+		rc = gbt_work_decode(json_object_get(val, "result"), work);
+		if (!have_gbt) {
+			json_decref(val);
+			goto start;
+		}
+	} else
+		rc = work_decode(json_object_get(val, "result"), work);
+
+	if (opt_debug && rc) {
+		timeval_subtract(&diff, &tv_end, &tv_start);
+		applog(LOG_DEBUG, "DEBUG: got new work in %d ms",
+		       diff.tv_sec * 1000 + diff.tv_usec / 1000);
+	}
+
+	json_decref(val);
+
+	return rc;
+}
+
+static void workio_cmd_free(struct workio_cmd *wc)
+{
+	if (!wc)
+		return;
+
+	switch (wc->cmd) {
+	case WC_SUBMIT_WORK:
+		work_free(wc->u.work);
+		free(wc->u.work);
+		break;
+	default: /* do nothing */
+		break;
+	}
+
+	memset(wc, 0, sizeof(*wc));	/* poison */
+	free(wc);
+}
+
+static bool workio_get_work(struct workio_cmd *wc, CURL *curl)
+{
+	struct work *ret_work;
+	int failures = 0;
+
+	ret_work = calloc(1, sizeof(*ret_work));
+	if (!ret_work)
+		return false;
+
+	/* obtain new work from bitcoin via JSON-RPC */
+	while (!get_upstream_work(curl, ret_work)) {
+		if (unlikely((opt_retries >= 0) && (++failures > opt_retries))) {
+			applog(LOG_ERR, "json_rpc_call failed, terminating workio thread");
+			free(ret_work);
+			return false;
+		}
+
+		/* pause, then restart work-request loop */
+		applog(LOG_ERR, "json_rpc_call failed, retry after %d seconds",
+			opt_fail_pause);
+		sleep(opt_fail_pause);
+	}
+
+	/* send work to requesting thread */
+	if (!tq_push(wc->thr->q, ret_work))
+		free(ret_work);
+
+	return true;
+}
+
+static bool workio_submit_work(struct workio_cmd *wc, CURL *curl)
+{
+	int failures = 0;
+
+	/* submit solution to bitcoin via JSON-RPC */
+	while (!submit_upstream_work(curl, wc->u.work)) {
+		if (unlikely((opt_retries >= 0) && (++failures > opt_retries))) {
+			applog(LOG_ERR, "...terminating workio thread");
+			return false;
+		}
+
+		/* pause, then restart work-request loop */
+		applog(LOG_ERR, "...retry after %d seconds",
+			opt_fail_pause);
+		sleep(opt_fail_pause);
+	}
+
+	return true;
+}
+
+static void *workio_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	CURL *curl;
+	bool ok = true;
+
+	curl = curl_easy_init();
+	if (unlikely(!curl)) {
+		applog(LOG_ERR, "CURL initialization failed");
+		return NULL;
+	}
+
+	while (ok && mythr->run) {
+		struct workio_cmd *wc;
+
+		/* wait for workio_cmd sent to us, on our queue */
+		wc = tq_pop(mythr->q, NULL);
+		if (!wc) {
+			ok = false;
+			break;
+		}
+
+		/* process workio_cmd */
+		switch (wc->cmd) {
+		case WC_GET_WORK:
+			ok = workio_get_work(wc, curl);
+			break;
+		case WC_SUBMIT_WORK:
+			ok = workio_submit_work(wc, curl);
+			break;
+
+		default:		/* should never happen */
+			ok = false;
+			break;
+		}
+
+		workio_cmd_free(wc);
+	}
+
+	tq_freeze(mythr->q);
+	curl_easy_cleanup(curl);
+	mythr->run = 0;
+
+	return NULL;
+}
+
+static bool get_work(struct thr_info *thr, struct work *work)
+{
+	struct workio_cmd *wc;
+	struct work *work_heap;
+
+	if (opt_benchmark) {
+		memset(work->data, 0x55, 76);
+		work->data[17] = swab32(time(NULL));
+		memset(work->data + 19, 0x00, 52);
+		work->data[20] = 0x80000000;
+		work->data[31] = 0x00000280;
+		memset(work->target, 0x00, sizeof(work->target));
+		return true;
+	}
+
+	/* fill out work request message */
+	wc = calloc(1, sizeof(*wc));
+	if (!wc)
+		return false;
+
+	wc->cmd = WC_GET_WORK;
+	wc->thr = thr;
+
+	/* send work request to workio thread */
+	if (!tq_push(thr_info[work_thr_id].q, wc)) {
+		workio_cmd_free(wc);
+		return false;
+	}
+
+	/* wait for response, a unit of work */
+	work_heap = tq_pop(thr->q, NULL);
+	if (!work_heap)
+		return false;
+
+	/* copy returned work into storage provided by caller */
+	memcpy(work, work_heap, sizeof(*work));
+	free(work_heap);
+
+	return true;
+}
+
+static bool submit_work(struct thr_info *thr, const struct work *work_in)
+{
+	struct workio_cmd *wc;
+	
+	/* fill out work request message */
+	wc = calloc(1, sizeof(*wc));
+	if (!wc)
+		return false;
+
+	wc->u.work = malloc(sizeof(*work_in));
+	if (!wc->u.work)
+		goto err_out;
+
+	wc->cmd = WC_SUBMIT_WORK;
+	wc->thr = thr;
+	work_copy(wc->u.work, work_in);
+
+	/* send solution to workio thread */
+	if (!tq_push(thr_info[work_thr_id].q, wc))
+		goto err_out;
+
+	return true;
+
+err_out:
+	workio_cmd_free(wc);
+	return false;
+}
+
+static void stratum_gen_work(struct stratum_ctx *sctx, struct work *work)
+{
+	unsigned char merkle_root[64];
+	int i;
+
+	pthread_mutex_lock(&sctx->work_lock);
+
+	free(work->job_id);
+	work->job_id = strdup(sctx->job.job_id);
+	work->xnonce2_len = sctx->xnonce2_size;
+	work->xnonce2 = realloc(work->xnonce2, sctx->xnonce2_size);
+	memcpy(work->xnonce2, sctx->job.xnonce2, sctx->xnonce2_size);
+
+	/* Generate merkle root */
+	sha256d(merkle_root, sctx->job.coinbase, sctx->job.coinbase_size);
+	for (i = 0; i < sctx->job.merkle_count; i++) {
+		memcpy(merkle_root + 32, sctx->job.merkle[i], 32);
+		sha256d(merkle_root, merkle_root, 64);
+	}
+	
+	/* Increment extranonce2 */
+	for (i = 0; i < sctx->xnonce2_size && !++sctx->job.xnonce2[i]; i++);
+
+	/* Assemble block header */
+	memset(work->data, 0, 128);
+	work->data[0] = le32dec(sctx->job.version);
+	for (i = 0; i < 8; i++)
+		work->data[1 + i] = le32dec((uint32_t *)sctx->job.prevhash + i);
+	for (i = 0; i < 8; i++)
+		work->data[9 + i] = be32dec((uint32_t *)merkle_root + i);
+	work->data[17] = le32dec(sctx->job.ntime);
+	work->data[18] = le32dec(sctx->job.nbits);
+	work->data[20] = 0x80000000;
+	work->data[31] = 0x00000280;
+
+	pthread_mutex_unlock(&sctx->work_lock);
+
+	if (opt_debug) {
+		char *xnonce2str = abin2hex(work->xnonce2, work->xnonce2_len);
+		applog(LOG_DEBUG, "DEBUG: job_id='%s' extranonce2=%s ntime=%08x",
+		       work->job_id, xnonce2str, swab32(work->data[17]));
+		free(xnonce2str);
+	}
+
+	if (opt_algo == ALGO_SCRYPT || opt_algo == ALGO_YESCRYPT || opt_algo == ALGO_YESPOWER)
+		diff_to_target(work->target, sctx->job.diff / 65536.0);
+	else
+		diff_to_target(work->target, sctx->job.diff);
+}
+
+static void *miner_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	int thr_id = mythr->id;
+	struct work work = {{0}};
+	uint32_t max_nonce;
+	uint32_t end_nonce = 0xffffffffU / opt_n_threads * (thr_id + 1) - 0x20;
+	unsigned char *scratchbuf = NULL;
+	char s[16];
+	int i;
+
+	/* Set worker threads to nice 19 and then preferentially to SCHED_IDLE
+	 * and if that fails, then SCHED_BATCH. No need for this to be an
+	 * error if it fails */
+	if (!opt_benchmark) {
+		setpriority(PRIO_PROCESS, 0, 19);
+#ifndef __APPLE__
+		drop_policy();
+#endif // __APPLE__
+	}
+
+	/* Cpu affinity only makes sense if the number of threads is a multiple
+	 * of the number of CPUs */
+#ifndef __KITKAT__
+#ifndef __APPLE__
+	if (num_processors > 1 && opt_n_threads % num_processors == 0) {
+		applog(LOG_INFO, "Binding thread %d to cpu %d",
+		       thr_id, thr_id % num_processors);
+		affine_to_cpu(thr_id, thr_id % num_processors);
+	}
+#endif // __APPLE__
+#endif // __KITKAT__
+	
+	if (opt_algo == ALGO_SCRYPT) {
+		scratchbuf = scrypt_buffer_alloc(opt_scrypt_n);
+		if (!scratchbuf) {
+			applog(LOG_ERR, "scrypt buffer allocation failed");
+			pthread_mutex_lock(&applog_lock);
+			exit(1);
+		}
+	}
+
+	while (mythr->run) {
+		unsigned long hashes_done;
+		struct timeval tv_start, tv_end, diff;
+		int64_t max64;
+		int rc;
+
+		if (have_stratum) {
+			while (time(NULL) >= g_work_time + 120)
+				sleep(1);
+			pthread_mutex_lock(&g_work_lock);
+			if (work.data[19] >= end_nonce && !memcmp(work.data, g_work.data, 76))
+				stratum_gen_work(&stratum, &g_work);
+		} else {
+			int min_scantime = have_longpoll ? LP_SCANTIME : opt_scantime;
+			/* obtain new work from internal workio thread */
+			pthread_mutex_lock(&g_work_lock);
+			if (!have_stratum &&
+			    (time(NULL) - g_work_time >= min_scantime ||
+			     work.data[19] >= end_nonce)) {
+				work_free(&g_work);
+				if (unlikely(!get_work(mythr, &g_work))) {
+					applog(LOG_ERR, "work retrieval failed, exiting "
+						"mining thread %d", mythr->id);
+					pthread_mutex_unlock(&g_work_lock);
+					goto out;
+				}
+				g_work_time = have_stratum ? 0 : time(NULL);
+			}
+			if (have_stratum) {
+				pthread_mutex_unlock(&g_work_lock);
+				continue;
+			}
+		}
+		if (memcmp(work.data, g_work.data, 76)) {
+			work_free(&work);
+			work_copy(&work, &g_work);
+			work.data[19] = 0xffffffffU / opt_n_threads * thr_id;
+		} else
+			work.data[19]++;
+		pthread_mutex_unlock(&g_work_lock);
+		work_restart[thr_id].restart = 0;
+		
+		/* adjust max_nonce to meet target scan time */
+		if (have_stratum)
+			max64 = LP_SCANTIME;
+		else
+			max64 = g_work_time + (have_longpoll ? LP_SCANTIME : opt_scantime)
+			      - time(NULL);
+		max64 *= thr_hashrates[thr_id];
+		if (max64 <= 0) {
+			switch (opt_algo) {
+			case ALGO_YESCRYPT:
+				max64 = 0x000fff;
+				break;
+			case ALGO_YESPOWER:
+				max64 = 0x000fff;
+				break;
+			case ALGO_SCRYPT:
+				max64 = opt_scrypt_n < 16 ? 0x3ffff : 0x3fffff / opt_scrypt_n;
+				break;
+			case ALGO_SHA256D:
+				max64 = 0x1fffff;
+				break;
+			}
+		}
+		if (work.data[19] + max64 > end_nonce)
+			max_nonce = end_nonce;
+		else
+			max_nonce = work.data[19] + max64;
+		
+		hashes_done = 0;
+		gettimeofday(&tv_start, NULL);
+
+		/* scan nonces for a proof-of-work hash */
+		switch (opt_algo) {
+		case ALGO_YESCRYPT:
+			rc = scanhash_yescrypt(thr_id, work.data, work.target,
+					       max_nonce, &hashes_done);
+			break;
+
+		case ALGO_YESPOWER:
+			rc = scanhash_yespower(thr_id, work.data, work.target,
+					       max_nonce, &hashes_done);
+			break;
+
+		case ALGO_SCRYPT:
+			rc = scanhash_scrypt(thr_id, work.data, scratchbuf, work.target,
+			                     max_nonce, &hashes_done, opt_scrypt_n);
+			break;
+
+		case ALGO_SHA256D:
+			rc = scanhash_sha256d(thr_id, work.data, work.target,
+			                      max_nonce, &hashes_done);
+			break;
+
+		default:
+			/* should never happen */
+			goto out;
+		}
+
+		/* record scanhash elapsed time */
+		gettimeofday(&tv_end, NULL);
+		timeval_subtract(&diff, &tv_end, &tv_start);
+		if (diff.tv_usec || diff.tv_sec) {
+			pthread_mutex_lock(&stats_lock);
+			thr_hashrates[thr_id] =
+				hashes_done / (diff.tv_sec + 1e-6 * diff.tv_usec);
+			pthread_mutex_unlock(&stats_lock);
+		}
+		if (!opt_quiet) {
+			sprintf(s, thr_hashrates[thr_id] >= 1e3 ? "%.0f" : "%.1f",
+				thr_hashrates[thr_id]);
+			applog(LOG_INFO, "thread %d: %lu hashes, %s hash/s",
+				thr_id, hashes_done, s);
+		}
+		if (opt_benchmark && thr_id == opt_n_threads - 1) {
+			double hashrate = 0.;
+			for (i = 0; i < opt_n_threads && thr_hashrates[i]; i++)
+				hashrate += thr_hashrates[i];
+			if (i == opt_n_threads) {
+				sprintf(s, hashrate >= 1e6 ? "%.0f" : "%.2f", 1e-3 * hashrate);
+				applog(LOG_INFO, "Total: %s khash/s", s);
+			}
+		}
+
+		/* if nonce found, submit work */
+		if (rc && !opt_benchmark && !submit_work(mythr, &work))
+			break;
+	}
+
+out:
+	tq_freeze(mythr->q);
+	mythr->run = 0;
+
+	return NULL;
+}
+
+static void restart_threads(void)
+{
+	int i;
+
+	for (i = 0; i < opt_n_threads; i++)
+		work_restart[i].restart = 1;
+}
+
+static void *longpoll_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	CURL *curl = NULL;
+	char *copy_start, *hdr_path = NULL, *lp_url = NULL;
+	bool need_slash = false;
+
+	curl = curl_easy_init();
+	if (unlikely(!curl)) {
+		applog(LOG_ERR, "CURL initialization failed");
+		goto out;
+	}
+
+start:
+	hdr_path = tq_pop(mythr->q, NULL);
+	if (!hdr_path)
+		goto out;
+
+	/* full URL */
+	if (strstr(hdr_path, "://")) {
+		lp_url = hdr_path;
+		hdr_path = NULL;
+	}
+	
+	/* absolute path, on current server */
+	else {
+		copy_start = (*hdr_path == '/') ? (hdr_path + 1) : hdr_path;
+		if (rpc_url[strlen(rpc_url) - 1] != '/')
+			need_slash = true;
+
+		lp_url = malloc(strlen(rpc_url) + strlen(copy_start) + 2);
+		if (!lp_url)
+			goto out;
+
+		sprintf(lp_url, "%s%s%s", rpc_url, need_slash ? "/" : "", copy_start);
+	}
+
+	applog(LOG_INFO, "Long-polling activated for %s", lp_url);
+
+	while (mythr->run) {
+		json_t *val, *res, *soval;
+		char *req = NULL;
+		int err;
+
+		if (have_gbt) {
+			req = malloc(strlen(gbt_lp_req) + strlen(lp_id) + 1);
+			sprintf(req, gbt_lp_req, lp_id);
+		}
+		val = json_rpc_call(curl, lp_url, rpc_userpass,
+				    req ? req : getwork_req, &err,
+				    JSON_RPC_LONGPOLL);
+		free(req);
+		if (have_stratum) {
+			if (val)
+				json_decref(val);
+			goto out;
+		}
+		if (likely(val)) {
+			bool rc;
+			applog(LOG_INFO, "LONGPOLL pushed new work");
+			res = json_object_get(val, "result");
+			soval = json_object_get(res, "submitold");
+			submit_old = soval ? json_is_true(soval) : false;
+			pthread_mutex_lock(&g_work_lock);
+			work_free(&g_work);
+			if (have_gbt)
+				rc = gbt_work_decode(res, &g_work);
+			else
+				rc = work_decode(res, &g_work);
+			if (rc) {
+				time(&g_work_time);
+				restart_threads();
+			}
+			pthread_mutex_unlock(&g_work_lock);
+			json_decref(val);
+		} else {
+			pthread_mutex_lock(&g_work_lock);
+			g_work_time -= LP_SCANTIME;
+			pthread_mutex_unlock(&g_work_lock);
+			if (err == CURLE_OPERATION_TIMEDOUT) {
+				restart_threads();
+			} else {
+				have_longpoll = false;
+				restart_threads();
+				free(hdr_path);
+				free(lp_url);
+				lp_url = NULL;
+				sleep(opt_fail_pause);
+				goto start;
+			}
+		}
+	}
+
+out:
+	free(hdr_path);
+	free(lp_url);
+	tq_freeze(mythr->q);
+	if (curl)
+		curl_easy_cleanup(curl);
+	mythr->run = 0;
+
+	return NULL;
+}
+
+static bool stratum_handle_response(char *buf)
+{
+	json_t *val, *err_val, *res_val, *id_val;
+	json_error_t err;
+	bool ret = false;
+
+	val = JSON_LOADS(buf, &err);
+	if (!val) {
+		applog(LOG_INFO, "JSON decode failed(%d): %s", err.line, err.text);
+		goto out;
+	}
+
+	res_val = json_object_get(val, "result");
+	err_val = json_object_get(val, "error");
+	id_val = json_object_get(val, "id");
+
+	if (!id_val || json_is_null(id_val) || !res_val)
+		goto out;
+
+	share_result(json_is_true(res_val),
+		err_val ? json_string_value(json_array_get(err_val, 1)) : NULL);
+
+	ret = true;
+out:
+	if (val)
+		json_decref(val);
+
+	return ret;
+}
+
+static void *stratum_thread(void *userdata)
+{
+	struct thr_info *mythr = userdata;
+	char *s;
+
+	stratum.url = tq_pop(mythr->q, NULL);
+	if (!stratum.url)
+		goto out;
+	applog(LOG_INFO, "Starting Stratum on %s", stratum.url);
+
+	while (mythr->run) {
+		int failures = 0;
+
+		while (!stratum.curl) {
+			pthread_mutex_lock(&g_work_lock);
+			g_work_time = 0;
+			pthread_mutex_unlock(&g_work_lock);
+			restart_threads();
+
+			if (!stratum_connect(&stratum, stratum.url) ||
+			    !stratum_subscribe(&stratum) ||
+			    !stratum_authorize(&stratum, rpc_user, rpc_pass)) {
+				stratum_disconnect(&stratum);
+				if (opt_retries >= 0 && ++failures > opt_retries) {
+					applog(LOG_ERR, "...terminating workio thread");
+					tq_push(thr_info[work_thr_id].q, NULL);
+					goto out;
+				}
+				applog(LOG_ERR, "...retry after %d seconds", opt_fail_pause);
+				sleep(opt_fail_pause);
+			}
+		}
+
+		if (stratum.job.job_id &&
+		    (!g_work_time || strcmp(stratum.job.job_id, g_work.job_id))) {
+			pthread_mutex_lock(&g_work_lock);
+			stratum_gen_work(&stratum, &g_work);
+			time(&g_work_time);
+			pthread_mutex_unlock(&g_work_lock);
+			if (stratum.job.clean) {
+				applog(LOG_INFO, "Stratum requested work restart");
+				restart_threads();
+			}
+		}
+		
+		if (!stratum_socket_full(&stratum, 120)) {
+			applog(LOG_ERR, "Stratum connection timed out");
+			s = NULL;
+		} else
+			s = stratum_recv_line(&stratum);
+		if (!s) {
+			stratum_disconnect(&stratum);
+			applog(LOG_ERR, "Stratum connection interrupted");
+			continue;
+		}
+		if (!stratum_handle_method(&stratum, s))
+			stratum_handle_response(s);
+		free(s);
+	}
+
+out:
+	mythr->run = 0;
+	return NULL;
+}
+
+#ifndef WIN32
+static void signal_handler(int sig)
+{
+	switch (sig) {
+	case SIGHUP:
+		applog(LOG_INFO, "SIGHUP received");
+		break;
+	case SIGINT:
+		applog(LOG_INFO, "SIGINT received, exiting");
+		exit(0);
+		break;
+	case SIGTERM:
+		applog(LOG_INFO, "SIGTERM received, exiting");
+		exit(0);
+		break;
+	}
+}
+#endif
+
+int init(void(*cb)(const char *format, va_list arg)) {
+	vprintf_cb = cb;
+	have_stratum = true;
+
+	pthread_mutex_init(&applog_lock, NULL);
+	pthread_mutex_init(&run_lock, NULL);
+	pthread_mutex_init(&stats_lock, NULL);
+	pthread_mutex_init(&g_work_lock, NULL);
+	pthread_mutex_init(&stratum.sock_lock, NULL);
+	pthread_mutex_init(&stratum.work_lock, NULL);
+
+#if defined(_SC_NPROCESSORS_CONF)
+	num_processors = sysconf(_SC_NPROCESSORS_CONF);
+#elif defined(CTL_HW) && defined(HW_NCPU)
+	int req[] = { CTL_HW, HW_NCPU };
+        size_t len = sizeof(num_processors);
+        sysctl(req, 2, &num_processors, &len, NULL, 0);
+#endif
+	if (num_processors < 1)
+		num_processors = 1;
+
+#ifdef HAVE_SYSLOG_H
+	if (use_syslog)
+		openlog("cpuminer", LOG_PID, LOG_USER);
+#endif
+
+	return 0;
+}
+
+int is_running() {
+	if (thr_info)  {
+		for (int i = 0; i < opt_n_threads + 3; i++) {
+			if (thr_info[i].run == 1) {
+				return 1;
+			}
+		}
+	}
+	return 0;
+}
+
+int start(const char *url, const char *user, const char *pass, const int n_threads, const int algo) {
+	struct thr_info *thr;
+	long flags;
+	int i;
+
+	if (url == NULL) {
+		opt_benchmark = true;
+		have_stratum = false;
+	} else {
+		rpc_url = strdup(url);
+		rpc_user = strdup(user);
+		rpc_pass = strdup(pass);
+	}
+
+	opt_n_threads = n_threads ? n_threads : num_processors;
+    opt_algo = algo;
+
+	pthread_mutex_lock(&run_lock);
+    if (is_running()) {
+		applog(LOG_INFO, "already running: thr_info[%d].pth=%lld", i, thr_info[i].pth);
+		pthread_mutex_unlock(&run_lock);
+		return 0;
+	}
+
+	flags = !opt_benchmark && strncmp(rpc_url, "https:", 6)
+			? (CURL_GLOBAL_ALL & ~CURL_GLOBAL_SSL)
+			: CURL_GLOBAL_ALL;
+	if (curl_global_init(flags)) {
+		applog(LOG_ERR, "CURL initialization failed");
+		return 1;
+	}
+
+	work_restart = calloc(opt_n_threads, sizeof(*work_restart));
+	if (!work_restart)
+		return 1;
+
+	thr_info = calloc(opt_n_threads + 3, sizeof(*thr));
+	if (!thr_info)
+		return 1;
+
+	thr_hashrates = (double *) calloc(opt_n_threads, sizeof(double));
+	if (!thr_hashrates)
+		return 1;
+
+	/* init workio thread info */
+	work_thr_id = opt_n_threads;
+	thr = &thr_info[work_thr_id];
+	thr->id = work_thr_id;
+	thr->q = tq_new();
+	thr->run = 1;
+	if (!thr->q)
+		return 1;
+
+	/* start work I/O thread */
+	if (pthread_create(&thr->pth, NULL, workio_thread, thr)) {
+		applog(LOG_ERR, "workio thread create failed");
+		return 1;
+	}
+
+	if (want_longpoll && !have_stratum) {
+		/* init longpoll thread info */
+		longpoll_thr_id = opt_n_threads + 1;
+		thr = &thr_info[longpoll_thr_id];
+		thr->id = longpoll_thr_id;
+		thr->q = tq_new();
+		thr->run = 1;
+		if (!thr->q)
+			return 1;
+
+		/* start longpoll thread */
+		if (unlikely(pthread_create(&thr->pth, NULL, longpoll_thread, thr))) {
+			applog(LOG_ERR, "longpoll thread create failed");
+			return 1;
+		}
+	}
+	if (want_stratum) {
+		/* init stratum thread info */
+		stratum_thr_id = opt_n_threads + 2;
+		thr = &thr_info[stratum_thr_id];
+		thr->id = stratum_thr_id;
+		thr->q = tq_new();
+		thr->run = 1;
+		if (!thr->q)
+			return 1;
+
+		/* start stratum thread */
+		if (unlikely(pthread_create(&thr->pth, NULL, stratum_thread, thr))) {
+			applog(LOG_ERR, "stratum thread create failed");
+			return 1;
+		}
+
+		if (have_stratum)
+			tq_push(thr_info[stratum_thr_id].q, strdup(rpc_url));
+	}
+
+	/* start mining threads */
+	for (i = 0; i < opt_n_threads; i++) {
+		thr = &thr_info[i];
+
+		thr->id = i;
+		thr->q = tq_new();
+		thr->run = 1;
+		if (!thr->q)
+			return 1;
+
+		if (unlikely(pthread_create(&thr->pth, NULL, miner_thread, thr))) {
+			applog(LOG_ERR, "thread %d create failed", i);
+			return 1;
+		}
+	}
+	pthread_mutex_unlock(&run_lock);
+
+	applog(LOG_INFO, "%d miner threads started, "
+				   "using '%s' algorithm.",
+		   opt_n_threads,
+		   algo_names[opt_algo]);
+
+	/* main loop - simply wait for workio thread to exit */
+	//pthread_join(thr_info[work_thr_id].pth, NULL);
+
+	//applog(LOG_INFO, "workio thread dead, exiting.");
+
+	return 0;
+}
+
+int stop() {
+	pthread_mutex_lock(&run_lock);
+    for (int i = 0; i < opt_n_threads + 3; i++) {
+		if (thr_info[i].run) {
+			applog(LOG_INFO, "stopping thr_info[%d].pth=%lld", i, thr_info[i].pth);
+			thr_info[i].run = 0;
+		}
+	}
+
+	restart_threads(); // force to exit immediately
+
+    int wait = 1;
+	while(wait) {
+		usleep(100000);
+		wait = 0;
+		for (int i = 0; i < opt_n_threads + 3; i++) {
+			if (thr_info[i].run) {
+				wait = 1;
+			}
+        }
+	}
+	pthread_mutex_unlock(&run_lock);
+
+    applog(LOG_INFO, "all threads stopped");
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/libcpuminer.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/libcpuminer.h
new file mode 100644
index 0000000..a821e22
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/libcpuminer.h
@@ -0,0 +1,15 @@
+#ifndef __LIB_CPU_MINER_H__
+#define __LIB_CPU_MINER_H__
+#ifdef __cplusplus
+extern "C"{
+#endif //__cplusplus
+
+int init(void(*cb)(const char *format, va_list arg));
+int is_running();
+int start(const char* url, const char* user, const char* pass, const int n_threads=0, const int algo=0);
+int stop();
+
+#ifdef __cplusplus
+}
+#endif //__cplusplus
+#endif //__LIB_CPU_MINER_H__
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/miner.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/miner.h
new file mode 100644
index 0000000..92578fa
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/miner.h
@@ -0,0 +1,282 @@
+//
+// ottylab update history <ottylab@gmail.com>
+//
+// 2018/02/17 @001 Add flag to control thread loops
+// 2018/04/02 @002 Suppot iOS
+//
+
+#ifndef __MINER_H__
+#define __MINER_H__
+
+#include "cpuminer-config.h"
+
+#include <stdbool.h>
+#include <inttypes.h>
+#include <sys/time.h>
+#include <pthread.h>
+#include <jansson.h>
+#include <curl/curl.h>
+
+#ifdef STDC_HEADERS
+# include <stdlib.h>
+# include <stddef.h>
+#else
+# ifdef HAVE_STDLIB_H
+#  include <stdlib.h>
+# endif
+#endif
+#ifdef HAVE_ALLOCA_H
+# include <alloca.h>
+#elif !defined alloca
+# ifdef __GNUC__
+#  define alloca __builtin_alloca
+# elif defined _AIX
+#  define alloca __alloca
+# elif defined _MSC_VER
+#  include <malloc.h>
+#  define alloca _alloca
+# elif !defined HAVE_ALLOCA
+#  ifdef  __cplusplus
+extern "C"
+#  endif
+void *alloca (size_t);
+# endif
+#endif
+
+#ifdef HAVE_SYSLOG_H
+#include <syslog.h>
+#else
+enum {
+	LOG_ERR,
+	LOG_WARNING,
+	LOG_NOTICE,
+	LOG_INFO,
+	LOG_DEBUG,
+};
+#endif
+
+#undef unlikely
+#undef likely
+#if defined(__GNUC__) && (__GNUC__ > 2) && defined(__OPTIMIZE__)
+#define unlikely(expr) (__builtin_expect(!!(expr), 0))
+#define likely(expr) (__builtin_expect(!!(expr), 1))
+#else
+#define unlikely(expr) (expr)
+#define likely(expr) (expr)
+#endif
+
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]))
+#endif
+
+#if ((__GNUC__ > 4) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3))
+#define WANT_BUILTIN_BSWAP
+#else
+#define bswap_32(x) ((((x) << 24) & 0xff000000u) | (((x) << 8) & 0x00ff0000u) \
+                   | (((x) >> 8) & 0x0000ff00u) | (((x) >> 24) & 0x000000ffu))
+#endif
+
+static inline uint32_t swab32(uint32_t v)
+{
+#ifdef WANT_BUILTIN_BSWAP
+	return __builtin_bswap32(v);
+#else
+	return bswap_32(v);
+#endif
+}
+
+#ifdef HAVE_SYS_ENDIAN_H
+#ifdef __APPLE__ //@002A
+#include <machine/endian.h> //@002A
+#else //@002A
+#include <sys/endian.h>
+#endif // __APPLE__ //@002A
+#endif
+
+#if !HAVE_DECL_BE32DEC
+static inline uint32_t be32dec(const void *pp)
+{
+	const uint8_t *p = (uint8_t const *)pp;
+	return ((uint32_t)(p[3]) + ((uint32_t)(p[2]) << 8) +
+	    ((uint32_t)(p[1]) << 16) + ((uint32_t)(p[0]) << 24));
+}
+#endif
+
+#if !HAVE_DECL_LE32DEC
+static inline uint32_t le32dec(const void *pp)
+{
+	const uint8_t *p = (uint8_t const *)pp;
+	return ((uint32_t)(p[0]) + ((uint32_t)(p[1]) << 8) +
+	    ((uint32_t)(p[2]) << 16) + ((uint32_t)(p[3]) << 24));
+}
+#endif
+
+#if !HAVE_DECL_BE32ENC
+static inline void be32enc(void *pp, uint32_t x)
+{
+	uint8_t *p = (uint8_t *)pp;
+	p[3] = x & 0xff;
+	p[2] = (x >> 8) & 0xff;
+	p[1] = (x >> 16) & 0xff;
+	p[0] = (x >> 24) & 0xff;
+}
+#endif
+
+#if !HAVE_DECL_LE32ENC
+static inline void le32enc(void *pp, uint32_t x)
+{
+	uint8_t *p = (uint8_t *)pp;
+	p[0] = x & 0xff;
+	p[1] = (x >> 8) & 0xff;
+	p[2] = (x >> 16) & 0xff;
+	p[3] = (x >> 24) & 0xff;
+}
+#endif
+
+#if JANSSON_MAJOR_VERSION >= 2
+#define JSON_LOADS(str, err_ptr) json_loads(str, 0, err_ptr)
+#define JSON_LOAD_FILE(path, err_ptr) json_load_file(path, 0, err_ptr)
+#else
+#define JSON_LOADS(str, err_ptr) json_loads(str, err_ptr)
+#define JSON_LOAD_FILE(path, err_ptr) json_load_file(path, err_ptr)
+#endif
+
+#define USER_AGENT PACKAGE_NAME "/" PACKAGE_VERSION
+
+void sha256_init(uint32_t *state);
+void sha256_transform(uint32_t *state, const uint32_t *block, int swap);
+void sha256d(unsigned char *hash, const unsigned char *data, int len);
+
+#ifdef USE_ASM
+#if defined(__ARM_NEON__) || defined(__ALTIVEC__) || defined(__i386__) || defined(__x86_64__)
+#define HAVE_SHA256_4WAY 1
+int sha256_use_4way();
+void sha256_init_4way(uint32_t *state);
+void sha256_transform_4way(uint32_t *state, const uint32_t *block, int swap);
+#endif
+#if defined(__x86_64__) && defined(USE_AVX2)
+#define HAVE_SHA256_8WAY 1
+int sha256_use_8way();
+void sha256_init_8way(uint32_t *state);
+void sha256_transform_8way(uint32_t *state, const uint32_t *block, int swap);
+#endif
+#endif
+
+extern int scanhash_yescrypt(int thr_id, uint32_t *pdata,
+        const uint32_t *ptarget, uint32_t max_nonce, unsigned long *hashes_done);
+
+extern int scanhash_yespower(int thr_id, uint32_t *pdata,
+        const uint32_t *ptarget, uint32_t max_nonce, unsigned long *hashes_done);
+
+extern int scanhash_sha256d(int thr_id, uint32_t *pdata,
+	const uint32_t *ptarget, uint32_t max_nonce, unsigned long *hashes_done);
+
+extern unsigned char *scrypt_buffer_alloc(int N);
+extern int scanhash_scrypt(int thr_id, uint32_t *pdata,
+	unsigned char *scratchbuf, const uint32_t *ptarget,
+	uint32_t max_nonce, unsigned long *hashes_done, int N);
+
+struct thr_info {
+	int		id;
+	pthread_t	pth;
+	struct thread_q	*q;
+    int run; //@001
+};
+
+struct work_restart {
+	volatile unsigned long	restart;
+	char			padding[128 - sizeof(unsigned long)];
+};
+
+extern bool opt_debug;
+extern bool opt_protocol;
+extern bool opt_redirect;
+extern int opt_timeout;
+extern bool want_longpoll;
+extern bool have_longpoll;
+extern bool have_gbt;
+extern bool allow_getwork;
+extern bool want_stratum;
+extern bool have_stratum;
+extern char *opt_cert;
+extern char *opt_proxy;
+extern long opt_proxy_type;
+extern bool use_syslog;
+extern pthread_mutex_t applog_lock;
+extern struct thr_info *thr_info;
+extern int longpoll_thr_id;
+extern int stratum_thr_id;
+extern struct work_restart *work_restart;
+
+#define JSON_RPC_LONGPOLL	(1 << 0)
+#define JSON_RPC_QUIET_404	(1 << 1)
+
+extern void applog(int prio, const char *fmt, ...);
+extern json_t *json_rpc_call(CURL *curl, const char *url, const char *userpass,
+	const char *rpc_req, int *curl_err, int flags);
+void memrev(unsigned char *p, size_t len);
+extern void bin2hex(char *s, const unsigned char *p, size_t len);
+extern char *abin2hex(const unsigned char *p, size_t len);
+extern bool hex2bin(unsigned char *p, const char *hexstr, size_t len);
+extern int varint_encode(unsigned char *p, uint64_t n);
+extern size_t address_to_script(unsigned char *out, size_t outsz, const char *addr);
+extern int timeval_subtract(struct timeval *result, struct timeval *x,
+	struct timeval *y);
+extern bool fulltest(const uint32_t *hash, const uint32_t *target);
+extern void diff_to_target(uint32_t *target, double diff);
+
+struct stratum_job {
+	char *job_id;
+	unsigned char prevhash[32];
+	size_t coinbase_size;
+	unsigned char *coinbase;
+	unsigned char *xnonce2;
+	int merkle_count;
+	unsigned char **merkle;
+	unsigned char version[4];
+	unsigned char nbits[4];
+	unsigned char ntime[4];
+	bool clean;
+	double diff;
+};
+
+struct stratum_ctx {
+	char *url;
+
+	CURL *curl;
+	char *curl_url;
+	char curl_err_str[CURL_ERROR_SIZE];
+	curl_socket_t sock;
+	size_t sockbuf_size;
+	char *sockbuf;
+	pthread_mutex_t sock_lock;
+
+	double next_diff;
+
+	char *session_id;
+	size_t xnonce1_size;
+	unsigned char *xnonce1;
+	size_t xnonce2_size;
+	struct stratum_job job;
+	pthread_mutex_t work_lock;
+};
+
+bool stratum_socket_full(struct stratum_ctx *sctx, int timeout);
+bool stratum_send_line(struct stratum_ctx *sctx, char *s);
+char *stratum_recv_line(struct stratum_ctx *sctx);
+bool stratum_connect(struct stratum_ctx *sctx, const char *url);
+void stratum_disconnect(struct stratum_ctx *sctx);
+bool stratum_subscribe(struct stratum_ctx *sctx);
+bool stratum_authorize(struct stratum_ctx *sctx, const char *user, const char *pass);
+bool stratum_handle_method(struct stratum_ctx *sctx, const char *s);
+
+struct thread_q;
+
+extern struct thread_q *tq_new(void);
+extern void tq_free(struct thread_q *tq);
+extern bool tq_push(struct thread_q *tq, void *data);
+extern void *tq_pop(struct thread_q *tq, const struct timespec *abstime);
+extern void tq_freeze(struct thread_q *tq);
+extern void tq_thaw(struct thread_q *tq);
+
+#endif /* __MINER_H__ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/minerd.1 b/bitzenymininglibrary/src/main/cpp/libcpuminer/minerd.1
new file mode 100644
index 0000000..05f5bf0
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/minerd.1
@@ -0,0 +1,225 @@
+.TH MINERD 1 "June 2017" "cpuminer 2.5.0"
+.SH NAME
+minerd \- CPU miner for Bitcoin and Litecoin
+.SH SYNOPSIS
+.B minerd
+[\fIOPTION\fR]...
+.SH DESCRIPTION
+.B minerd
+is a multi-threaded CPU miner for Bitcoin, Litecoin and other cryptocurrencies.
+It supports the getwork and getblocktemplate (BIP 22) methods,
+as well as the Stratum mining protocol.
+.PP
+In its normal mode of operation, \fBminerd\fR connects to a mining server
+(specified with the \fB\-o\fR option), receives work from it and starts hashing.
+As soon as a solution is found, it is submitted to the same mining server,
+which can accept or reject it.
+When using getwork or getblocktemplate,
+\fBminerd\fR can take advantage of long polling, if the server supports it;
+in any case, fresh work is fetched as needed.
+When using the Stratum protocol this is not possible,
+and the server is responsible for sending fresh work at least every minute;
+if it fails to do so,
+\fBminerd\fR may drop the connection and try reconnecting again.
+.PP
+By default, \fBminerd\fR writes all its messages to standard error.
+On systems that have a syslog, the \fB\-\-syslog\fR option can be used
+to write to it instead.
+.PP
+On start, the nice value of all miner threads is set to 19.
+On Linux, the scheduling policy is also changed to SCHED_IDLE,
+or to SCHED_BATCH if that fails.
+On multiprocessor systems, \fBminerd\fR
+automatically sets the CPU affinity of miner threads
+if the number of threads is a multiple of the number of processors.
+.SH EXAMPLES
+To connect to a Litecoin mining pool that provides a Stratum server
+at example.com on port 3333, authenticating as worker "foo" with password "bar":
+.PP
+.nf
+.RS
+minerd \-o stratum+tcp://example.com:3333 \-O foo:bar
+.RE
+.fi
+.PP
+To mine to a local Bitcoin testnet instance running on port 18332,
+authenticating with username "rpcuser" and password "rpcpass":
+.PP
+.nf
+.RS
+minerd \-a sha256d \-o http://localhost:18332 \-O rpcuser:rpcpass \\
+	\-\-coinbase\-addr=mpXwg4jMtRhuSpVq4xS3HFHmCmWp9NyGKt
+.RE
+.fi
+.PP
+To connect to a Litecoin P2Pool node running on my.server on port 9327,
+mining in the background and having output sent to the syslog facility,
+omitting the per-thread hashmeter output:
+.PP
+.nf
+.RS
+minerd \-BSq \-o http://my.server:9327
+.RE
+.fi
+.SH OPTIONS
+.TP
+\fB\-a\fR, \fB\-\-algo\fR=\fIALGORITHM\fR
+Set the hashing algorithm to use.
+Default is scrypt.
+Possible values are:
+.RS 11
+.TP 10
+.B scrypt
+scrypt(1024, 1, 1) (used by Litecoin)
+.TP
+.B scrypt:\fIN\fR
+scrypt(\fIN\fR, 1, 1) (\fIN\fR must be a power of 2 greater than 1)
+.TP
+.B sha256d
+SHA-256d (used by Bitcoin)
+.RE
+.TP
+\fB\-\-benchmark\fR
+Run in offline benchmark mode.
+.TP
+\fB\-B\fR, \fB\-\-background\fR
+Run in the background as a daemon.
+.TP
+\fB\-\-cert\fR=\fIFILE\fR
+Set an SSL certificate to use with the mining server.
+Only supported when using the HTTPS protocol.
+.TP
+\fB\-\-coinbase\-addr\fR=\fIADDRESS\fR
+Set a payout address for solo mining.
+This is only used in getblocktemplate mode,
+and only if the server does not provide a coinbase transaction.
+.TP
+\fB\-\-coinbase\-sig\fR=\fITEXT\fR
+Set a string to be included in the coinbase (if allowed by the server).
+This is only used in getblocktemplate mode.
+.TP
+\fB\-c\fR, \fB\-\-config\fR=\fIFILE\fR
+Load options from a configuration file.
+\fIFILE\fR must contain a JSON object
+mapping long options to their arguments (as strings),
+or to \fBtrue\fR if no argument is required.
+Sample configuration file:
+
+.nf
+	{
+		"url": "stratum+tcp://example.com:3333",
+		"userpass": "foo:bar",
+		"retry-pause": "10",
+		"quiet": true
+	}
+.fi
+.TP
+\fB\-D\fR, \fB\-\-debug\fR
+Enable debug output.
+.TP
+\fB\-h\fR, \fB\-\-help\fR
+Print a help message and exit.
+.TP
+\fB\-\-no\-gbt\fR
+Do not use the getblocktemplate RPC method.
+.TP
+\fB\-\-no\-getwork\fR
+Do not use the getwork RPC method.
+.TP
+\fB\-\-no\-longpoll\fR
+Do not use long polling.
+.TP
+\fB\-\-no\-redirect\fR
+Ignore requests from the server to switch to a different URL.
+.TP
+\fB\-\-no\-stratum\fR
+Do not switch to Stratum, even if the server advertises support for it.
+.TP
+\fB\-o\fR, \fB\-\-url\fR=[\fISCHEME\fR://][\fIUSERNAME\fR[:\fIPASSWORD\fR]@]\fIHOST\fR:\fIPORT\fR[/\fIPATH\fR]
+Set the URL of the mining server to connect to.
+Supported schemes are \fBhttp\fR, \fBhttps\fR, \fBstratum+tcp\fR
+and \fBstratum+tcps\fR.
+If no scheme is specified, http is assumed.
+Specifying a \fIPATH\fR is only supported for HTTP and HTTPS.
+Specifying credentials has the same effect as using the \fB\-O\fR option.
+
+By default, on HTTP and HTTPS,
+the miner tries to use the getblocktemplate RPC method,
+and falls back to using getwork if getblocktemplate is unavailable.
+This behavior can be modified by using the \fB\-\-no\-gbt\fR
+and \fB\-\-no\-getwork\fR options.
+.TP
+\fB\-O\fR, \fB\-\-userpass\fR=\fIUSERNAME\fR:\fIPASSWORD\fR
+Set the credentials to use for connecting to the mining server.
+Any value previously set with \fB\-u\fR or \fB\-p\fR is discarded.
+.TP
+\fB\-p\fR, \fB\-\-pass\fR=\fIPASSWORD\fR
+Set the password to use for connecting to the mining server.
+Any password previously set with \fB\-O\fR is discarded.
+.TP
+\fB\-P\fR, \fB\-\-protocol\-dump\fR
+Enable output of all protocol-level activities.
+.TP
+\fB\-q\fR, \fB\-\-quiet\fR
+Disable per-thread hashmeter output.
+.TP
+\fB\-r\fR, \fB\-\-retries\fR=\fIN\fR
+Set the maximum number of times to retry if a network call fails.
+If not specified, the miner will retry indefinitely.
+.TP
+\fB\-R\fR, \fB\-\-retry\-pause\fR=\fISECONDS\fR
+Set how long to wait between retries. Default is 30 seconds.
+.TP
+\fB\-s\fR, \fB\-\-scantime\fR=\fISECONDS\fR
+Set an upper bound on the time the miner can go without fetching fresh work.
+This setting has no effect in Stratum mode or when long polling is activated.
+Default is 5 seconds.
+.TP
+\fB\-S\fR, \fB\-\-syslog\fR
+Log to the syslog facility instead of standard error.
+.TP
+\fB\-t\fR, \fB\-\-threads\fR=\fIN\fR
+Set the number of miner threads.
+If not specified, the miner will try to detect the number of available processors
+and use that.
+.TP
+\fB\-T\fR, \fB\-\-timeout\fR=\fISECONDS\fR
+Set a timeout for long polling.
+.TP
+\fB\-u\fR, \fB\-\-user\fR=\fIUSERNAME\fR
+Set the username to use for connecting to the mining server.
+Any username previously set with \fB\-O\fR is discarded.
+.TP
+\fB\-V\fR, \fB\-\-version\fR
+Display version information and quit.
+.TP
+\fB\-x\fR, \fB\-\-proxy\fR=[\fISCHEME\fR://][\fIUSERNAME\fR:\fIPASSWORD\fR@]\fIHOST\fR:\fIPORT\fR
+Connect to the mining server through a proxy.
+Supported schemes are: \fBhttp\fR, \fBsocks4\fR, \fBsocks5\fR.
+Since libcurl 7.18.0, the following are also supported:
+\fBsocks4a\fR, \fBsocks5h\fR (SOCKS5 with remote name resolving).
+If no scheme is specified, the proxy is treated as an HTTP proxy.
+.SH ENVIRONMENT
+The following environment variables can be specified in lower case or upper case;
+the lower-case version has precedence. \fBhttp_proxy\fR is an exception
+as it is only available in lower case.
+.PP
+.RS
+.TP
+\fBhttp_proxy\fR [\fISCHEME\fR://]\fIHOST\fR:\fIPORT\fR
+Sets the proxy server to use for HTTP.
+.TP
+\fBHTTPS_PROXY\fR [\fISCHEME\fR://]\fIHOST\fR:\fIPORT\fR
+Sets the proxy server to use for HTTPS.
+.TP
+\fBALL_PROXY\fR [\fISCHEME\fR://]\fIHOST\fR:\fIPORT\fR
+Sets the proxy server to use if no protocol-specific proxy is set.
+.RE
+.PP
+Using an environment variable to set the proxy has the same effect as
+using the \fB\-x\fR option.
+.SH AUTHOR
+Most of the code in the current version of minerd was written by
+Pooler <pooler@litecoinpool.org> with contributions from others.
+
+The original minerd was written by Jeff Garzik <jeff@garzik.org>.
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/nomacro.pl b/bitzenymininglibrary/src/main/cpp/libcpuminer/nomacro.pl
new file mode 100644
index 0000000..17a28fd
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/nomacro.pl
@@ -0,0 +1,49 @@
+#!/usr/bin/perl
+# Copyright 2012, 2015 pooler@litecoinpool.org
+#
+# This program is free software; you can redistribute it and/or modify it
+# under the terms of the GNU General Public License as published by the Free
+# Software Foundation; either version 2 of the License, or (at your option)
+# any later version.  See COPYING for more details.
+#
+# nomacro.pl - expand assembler macros.
+
+use strict;
+
+foreach my $f (<*.S>) {
+	rename $f, "$f.orig" unless -e "$f.orig";
+	open FIN, "$f.orig";
+	open FOUT, ">$f";
+	my %macros = ();
+	my %m = ();
+	while (<FIN>) {
+		if (m/^\.macro\s+(\w+)\s*(.*)$/) {
+			$m{name} = $1;
+			@m{args} = [split /\s*,\s*/, $2];
+			$m{body} = "";
+			next;
+		}
+		if (m/^\.endm/) {
+			$macros{$m{name}} = {%m};
+			%m = ();
+			next;
+		}
+		for my $n (keys %macros) {
+			if (m/^\s*$n\b\s*(.*)$/) {
+				my @a = split /\s*,\s*/, $1;
+				$_ = $macros{$n}{body};
+				for my $i (0 .. $#{$macros{$n}{args}}) {
+					s/\\$macros{$n}{args}[$i]\b/$a[$i]/g;
+				}
+				last;
+			}
+		}
+		if (%m) {
+			$m{body} .= $_;
+			next;
+		}
+		print FOUT;
+	}
+	close FOUT;
+	close FIN;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/run-sha2-x86-notextrel.test.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/run-sha2-x86-notextrel.test.c
new file mode 100644
index 0000000..eb63315
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/run-sha2-x86-notextrel.test.c
@@ -0,0 +1,6 @@
+gcc -o original sha2-x86-notextrel-test.c sha2-x86.S
+gcc -o notextrel sha2-x86-notextrel-test.c sha2-x86-notextrel.S
+
+./original > original.log
+./notextrel > notextrel.log
+diff original.log notextrel.log
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-arm.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-arm.S
new file mode 100644
index 0000000..5be3b0e
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-arm.S
@@ -0,0 +1,1186 @@
+/*
+ * Copyright 2012, 2014 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(USE_ASM) && defined(__arm__) && defined(__APCS_32__)
+
+#if defined(__ARM_ARCH_5E__) || defined(__ARM_ARCH_5TE__) || \
+	defined(__ARM_ARCH_5TEJ__) || defined(__ARM_ARCH_6__) || \
+	defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || \
+	defined(__ARM_ARCH_6M__) || defined(__ARM_ARCH_6T2__) || \
+	defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__)
+#define __ARM_ARCH_5E_OR_6__
+#endif
+
+#if defined(__ARM_ARCH_5E_OR_6__) || defined(__ARM_ARCH_7__) || \
+	defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || \
+	defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7EM__)
+#define __ARM_ARCH_5E_OR_6_OR_7__
+#endif
+
+#ifdef __ARM_ARCH_5E_OR_6__
+
+.macro scrypt_shuffle
+	add	lr, r0, #9*4
+	ldmia	r0, {r2-r7}
+	ldmia	lr, {r2, r8-r12, lr}
+	str	r3, [r0, #5*4]
+	str	r5, [r0, #15*4]
+	str	r6, [r0, #12*4]
+	str	r7, [r0, #1*4]
+	ldr r5, [r0, #7*4]
+	str	r2, [r0, #13*4]
+	str	r8, [r0, #2*4]
+	strd	r4, [r0, #10*4]
+	str	r9, [r0, #7*4]
+	str	r10, [r0, #4*4]
+	str	r11, [r0, #9*4]
+	str	lr, [r0, #3*4]
+	
+	add	r2, r0, #64+0*4
+	add	lr, r0, #64+9*4
+	ldmia	r2, {r2-r7}
+	ldmia	lr, {r2, r8-r12, lr}
+	str	r3, [r0, #64+5*4]
+	str	r5, [r0, #64+15*4]
+	str	r6, [r0, #64+12*4]
+	str	r7, [r0, #64+1*4]
+	ldr r5, [r0, #64+7*4]
+	str	r2, [r0, #64+13*4]
+	str	r8, [r0, #64+2*4]
+	strd	r4, [r0, #64+10*4]
+	str	r9, [r0, #64+7*4]
+	str	r10, [r0, #64+4*4]
+	str	r11, [r0, #64+9*4]
+	str	lr, [r0, #64+3*4]
+.endm
+
+.macro salsa8_core_doubleround_body
+	add	r6, r2, r6
+	add	r7, r3, r7
+	eor	r10, r10, r6, ror #25
+	add	r6, r0, r4
+	eor	r11, r11, r7, ror #25
+	add	r7, r1, r5
+	strd	r10, [sp, #14*4]
+	eor	r12, r12, r6, ror #25
+	eor	lr, lr, r7, ror #25
+	
+	ldrd	r6, [sp, #10*4]
+	add	r2, r10, r2
+	add	r3, r11, r3
+	eor	r6, r6, r2, ror #23
+	add	r2, r12, r0
+	eor	r7, r7, r3, ror #23
+	add	r3, lr, r1
+	strd	r6, [sp, #10*4]
+	eor	r8, r8, r2, ror #23
+	eor	r9, r9, r3, ror #23
+	
+	ldrd	r2, [sp, #6*4]
+	add	r10, r6, r10
+	add	r11, r7, r11
+	eor	r2, r2, r10, ror #19
+	add	r10, r8, r12
+	eor	r3, r3, r11, ror #19
+	add	r11, r9, lr
+	eor	r4, r4, r10, ror #19
+	eor	r5, r5, r11, ror #19
+	
+	ldrd	r10, [sp, #2*4]
+	add	r6, r2, r6
+	add	r7, r3, r7
+	eor	r10, r10, r6, ror #14
+	add	r6, r4, r8
+	eor	r11, r11, r7, ror #14
+	add	r7, r5, r9
+	eor	r0, r0, r6, ror #14
+	eor	r1, r1, r7, ror #14
+	
+	
+	ldrd	r6, [sp, #14*4]
+	strd	r2, [sp, #6*4]
+	strd	r10, [sp, #2*4]
+	add	r6, r11, r6
+	add	r7, r0, r7
+	eor	r4, r4, r6, ror #25
+	add	r6, r1, r12
+	eor	r5, r5, r7, ror #25
+	add	r7, r10, lr
+	eor	r2, r2, r6, ror #25
+	eor	r3, r3, r7, ror #25
+	strd	r2, [sp, #6*4]
+	
+	add	r10, r3, r10
+	ldrd	r6, [sp, #10*4]
+	add	r11, r4, r11
+	eor	r8, r8, r10, ror #23
+	add	r10, r5, r0
+	eor	r9, r9, r11, ror #23
+	add	r11, r2, r1
+	eor	r6, r6, r10, ror #23
+	eor	r7, r7, r11, ror #23
+	strd	r6, [sp, #10*4]
+	
+	add	r2, r7, r2
+	ldrd	r10, [sp, #14*4]
+	add	r3, r8, r3
+	eor	r12, r12, r2, ror #19
+	add	r2, r9, r4
+	eor	lr, lr, r3, ror #19
+	add	r3, r6, r5
+	eor	r10, r10, r2, ror #19
+	eor	r11, r11, r3, ror #19
+	
+	ldrd	r2, [sp, #2*4]
+	add	r6, r11, r6
+	add	r7, r12, r7
+	eor	r0, r0, r6, ror #14
+	add	r6, lr, r8
+	eor	r1, r1, r7, ror #14
+	add	r7, r10, r9
+	eor	r2, r2, r6, ror #14
+	eor	r3, r3, r7, ror #14
+.endm
+
+.macro salsa8_core
+	ldmia	sp, {r0-r12, lr}
+	
+	ldrd	r10, [sp, #14*4]
+	salsa8_core_doubleround_body
+	ldrd	r6, [sp, #6*4]
+	strd	r2, [sp, #2*4]
+	strd	r10, [sp, #14*4]
+	salsa8_core_doubleround_body
+	ldrd	r6, [sp, #6*4]
+	strd	r2, [sp, #2*4]
+	strd	r10, [sp, #14*4]
+	salsa8_core_doubleround_body
+	ldrd	r6, [sp, #6*4]
+	strd	r2, [sp, #2*4]
+	strd	r10, [sp, #14*4]
+	salsa8_core_doubleround_body
+	
+	stmia	sp, {r0-r5}
+	strd	r8, [sp, #8*4]
+	str	r12, [sp, #12*4]
+	str	lr, [sp, #13*4]
+	strd	r10, [sp, #14*4]
+.endm
+
+#else
+
+.macro scrypt_shuffle
+.endm
+
+.macro salsa8_core_doubleround_body
+	ldr	r8, [sp, #8*4]
+	add	r11, r11, r10
+	ldr	lr, [sp, #13*4]
+	add	r12, r12, r3
+	eor	r2, r2, r11, ror #23
+	add	r11, r4, r0
+	eor	r7, r7, r12, ror #23
+	add	r12, r9, r5
+	str	r9, [sp, #9*4]
+	eor	r8, r8, r11, ror #23
+	str	r10, [sp, #14*4]
+	eor	lr, lr, r12, ror #23
+	
+	ldr	r11, [sp, #11*4]
+	add	r9, lr, r9
+	ldr	r12, [sp, #12*4]
+	add	r10, r2, r10
+	eor	r1, r1, r9, ror #19
+	add	r9, r7, r3
+	eor	r6, r6, r10, ror #19
+	add	r10, r8, r4
+	str	r8, [sp, #8*4]
+	eor	r11, r11, r9, ror #19
+	str	lr, [sp, #13*4]
+	eor	r12, r12, r10, ror #19
+	
+	ldr	r9, [sp, #10*4]
+	add	r8, r12, r8
+	ldr	r10, [sp, #15*4]
+	add	lr, r1, lr
+	eor	r0, r0, r8, ror #14
+	add	r8, r6, r2
+	eor	r5, r5, lr, ror #14
+	add	lr, r11, r7
+	eor	r9, r9, r8, ror #14
+	ldr	r8, [sp, #9*4]
+	eor	r10, r10, lr, ror #14
+	ldr	lr, [sp, #14*4]
+	
+	
+	add	r8, r9, r8
+	str	r9, [sp, #10*4]
+	add	lr, r10, lr
+	str	r10, [sp, #15*4]
+	eor	r11, r11, r8, ror #25
+	add	r8, r0, r3
+	eor	r12, r12, lr, ror #25
+	add	lr, r5, r4
+	eor	r1, r1, r8, ror #25
+	ldr	r8, [sp, #8*4]
+	eor	r6, r6, lr, ror #25
+	
+	add	r9, r11, r9
+	ldr	lr, [sp, #13*4]
+	add	r10, r12, r10
+	eor	r8, r8, r9, ror #23
+	add	r9, r1, r0
+	eor	lr, lr, r10, ror #23
+	add	r10, r6, r5
+	str	r11, [sp, #11*4]
+	eor	r2, r2, r9, ror #23
+	str	r12, [sp, #12*4]
+	eor	r7, r7, r10, ror #23
+	
+	ldr	r9, [sp, #9*4]
+	add	r11, r8, r11
+	ldr	r10, [sp, #14*4]
+	add	r12, lr, r12
+	eor	r9, r9, r11, ror #19
+	add	r11, r2, r1
+	eor	r10, r10, r12, ror #19
+	add	r12, r7, r6
+	str	r8, [sp, #8*4]
+	eor	r3, r3, r11, ror #19
+	str	lr, [sp, #13*4]
+	eor	r4, r4, r12, ror #19
+.endm
+
+.macro salsa8_core
+	ldmia	sp, {r0-r7}
+	
+	ldr	r12, [sp, #15*4]
+	ldr	r8, [sp, #11*4]
+	ldr	lr, [sp, #12*4]
+	
+	ldr	r9, [sp, #9*4]
+	add	r8, r8, r12
+	ldr	r11, [sp, #10*4]
+	add	lr, lr, r0
+	eor	r3, r3, r8, ror #25
+	add	r8, r5, r1
+	ldr	r10, [sp, #14*4]
+	eor	r4, r4, lr, ror #25
+	add	lr, r11, r6
+	eor	r9, r9, r8, ror #25
+	eor	r10, r10, lr, ror #25
+	
+	salsa8_core_doubleround_body
+	
+	ldr	r11, [sp, #10*4]
+	add	r8, r9, r8
+	ldr	r12, [sp, #15*4]
+	add	lr, r10, lr
+	eor	r11, r11, r8, ror #14
+	add	r8, r3, r2
+	eor	r12, r12, lr, ror #14
+	add	lr, r4, r7
+	eor	r0, r0, r8, ror #14
+	ldr	r8, [sp, #11*4]
+	eor	r5, r5, lr, ror #14
+	ldr	lr, [sp, #12*4]
+	
+	add	r8, r8, r12
+	str	r11, [sp, #10*4]
+	add	lr, lr, r0
+	str	r12, [sp, #15*4]
+	eor	r3, r3, r8, ror #25
+	add	r8, r5, r1
+	eor	r4, r4, lr, ror #25
+	add	lr, r11, r6
+	str	r9, [sp, #9*4]
+	eor	r9, r9, r8, ror #25
+	str	r10, [sp, #14*4]
+	eor	r10, r10, lr, ror #25
+	
+	salsa8_core_doubleround_body
+	
+	ldr	r11, [sp, #10*4]
+	add	r8, r9, r8
+	ldr	r12, [sp, #15*4]
+	add	lr, r10, lr
+	eor	r11, r11, r8, ror #14
+	add	r8, r3, r2
+	eor	r12, r12, lr, ror #14
+	add	lr, r4, r7
+	eor	r0, r0, r8, ror #14
+	ldr	r8, [sp, #11*4]
+	eor	r5, r5, lr, ror #14
+	ldr	lr, [sp, #12*4]
+	
+	add	r8, r8, r12
+	str	r11, [sp, #10*4]
+	add	lr, lr, r0
+	str	r12, [sp, #15*4]
+	eor	r3, r3, r8, ror #25
+	add	r8, r5, r1
+	eor	r4, r4, lr, ror #25
+	add	lr, r11, r6
+	str	r9, [sp, #9*4]
+	eor	r9, r9, r8, ror #25
+	str	r10, [sp, #14*4]
+	eor	r10, r10, lr, ror #25
+	
+	salsa8_core_doubleround_body
+	
+	ldr	r11, [sp, #10*4]
+	add	r8, r9, r8
+	ldr	r12, [sp, #15*4]
+	add	lr, r10, lr
+	eor	r11, r11, r8, ror #14
+	add	r8, r3, r2
+	eor	r12, r12, lr, ror #14
+	add	lr, r4, r7
+	eor	r0, r0, r8, ror #14
+	ldr	r8, [sp, #11*4]
+	eor	r5, r5, lr, ror #14
+	ldr	lr, [sp, #12*4]
+	
+	add	r8, r8, r12
+	str	r11, [sp, #10*4]
+	add	lr, lr, r0
+	str	r12, [sp, #15*4]
+	eor	r3, r3, r8, ror #25
+	add	r8, r5, r1
+	eor	r4, r4, lr, ror #25
+	add	lr, r11, r6
+	str	r9, [sp, #9*4]
+	eor	r9, r9, r8, ror #25
+	str	r10, [sp, #14*4]
+	eor	r10, r10, lr, ror #25
+	
+	salsa8_core_doubleround_body
+	
+	ldr	r11, [sp, #10*4]
+	add	r8, r9, r8
+	ldr	r12, [sp, #15*4]
+	add	lr, r10, lr
+	str	r9, [sp, #9*4]
+	eor	r11, r11, r8, ror #14
+	eor	r12, r12, lr, ror #14
+	add	r8, r3, r2
+	str	r10, [sp, #14*4]
+	add	lr, r4, r7
+	str	r11, [sp, #10*4]
+	eor	r0, r0, r8, ror #14
+	str	r12, [sp, #15*4]
+	eor	r5, r5, lr, ror #14
+	
+	stmia	sp, {r0-r7}
+.endm
+
+#endif
+
+
+.macro scrypt_core_macro1a_x4
+	ldmia	r0, {r4-r7}
+	ldmia	lr!, {r8-r11}
+	stmia	r1!, {r4-r7}
+	stmia	r3!, {r8-r11}
+	eor	r4, r4, r8
+	eor	r5, r5, r9
+	eor	r6, r6, r10
+	eor	r7, r7, r11
+	stmia	r0!, {r4-r7}
+	stmia	r12!, {r4-r7}
+.endm
+
+.macro scrypt_core_macro1b_x4
+	ldmia	r3!, {r8-r11}
+	ldmia	r2, {r4-r7}
+	eor	r8, r8, r4
+	eor	r9, r9, r5
+	eor	r10, r10, r6
+	eor	r11, r11, r7
+	ldmia	r0, {r4-r7}
+	stmia	r2!, {r8-r11}
+	eor	r4, r4, r8
+	eor	r5, r5, r9
+	eor	r6, r6, r10
+	eor	r7, r7, r11
+	ldmia	r1!, {r8-r11}
+	eor	r4, r4, r8
+	eor	r5, r5, r9
+	eor	r6, r6, r10
+	eor	r7, r7, r11
+	stmia	r0!, {r4-r7}
+	stmia	r12!, {r4-r7}
+.endm
+
+.macro scrypt_core_macro2_x4
+	ldmia	r12, {r4-r7}
+	ldmia	r0, {r8-r11}
+	add	r4, r4, r8
+	add	r5, r5, r9
+	add	r6, r6, r10
+	add	r7, r7, r11
+	stmia	r0!, {r4-r7}
+	ldmia	r2, {r8-r11}
+	eor	r4, r4, r8
+	eor	r5, r5, r9
+	eor	r6, r6, r10
+	eor	r7, r7, r11
+	stmia	r2!, {r4-r7}
+	stmia	r12!, {r4-r7}
+.endm
+
+.macro scrypt_core_macro3_x4
+	ldmia	r1!, {r4-r7}
+	ldmia	r0, {r8-r11}
+	add	r4, r4, r8
+	add	r5, r5, r9
+	add	r6, r6, r10
+	add	r7, r7, r11
+	stmia	r0!, {r4-r7}
+.endm
+
+.macro scrypt_core_macro3_x6
+	ldmia	r1!, {r2-r7}
+	ldmia	r0, {r8-r12, lr}
+	add	r2, r2, r8
+	add	r3, r3, r9
+	add	r4, r4, r10
+	add	r5, r5, r11
+	add	r6, r6, r12
+	add	r7, r7, lr
+	stmia	r0!, {r2-r7}
+.endm
+
+
+	.text
+	.code 32
+	.align 2
+	.globl scrypt_core
+	.globl _scrypt_core
+#ifdef __ELF__
+	.type scrypt_core, %function
+#endif
+scrypt_core:
+_scrypt_core:
+	stmfd	sp!, {r4-r11, lr}
+	mov	r12, sp
+	sub	sp, sp, #22*4
+	bic	sp, sp, #63
+	str	r12, [sp, #20*4]
+	str	r2, [sp, #21*4]
+	
+	scrypt_shuffle
+	
+	ldr	r2, [sp, #21*4]
+	str	r0, [sp, #16*4]
+	add	r12, r1, r2, lsl #7
+	str	r12, [sp, #18*4]
+scrypt_core_loop1:
+	add	lr, r0, #16*4
+	add	r3, r1, #16*4
+	mov	r12, sp
+	scrypt_core_macro1a_x4
+	scrypt_core_macro1a_x4
+	scrypt_core_macro1a_x4
+	scrypt_core_macro1a_x4
+	str	r1, [sp, #17*4]
+	
+	salsa8_core
+	
+	ldr	r0, [sp, #16*4]
+	mov	r12, sp
+	add	r2, r0, #16*4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	
+	salsa8_core
+	
+	ldr	r0, [sp, #16*4]
+	mov	r1, sp
+	add	r0, r0, #16*4
+	scrypt_core_macro3_x6
+	scrypt_core_macro3_x6
+	ldr	r3, [sp, #17*4]
+	ldr	r12, [sp, #18*4]
+	scrypt_core_macro3_x4
+	
+	add	r1, r3, #16*4
+	sub	r0, r0, #32*4
+	cmp	r1, r12
+	bne	scrypt_core_loop1
+	
+	ldr	r12, [sp, #21*4]
+	ldr	r4, [r0, #16*4]
+	sub	r2, r12, #1
+	str	r2, [sp, #21*4]
+	sub	r1, r1, r12, lsl #7
+	str	r1, [sp, #17*4]
+	and	r4, r4, r2
+	add	r1, r1, r4, lsl #7
+scrypt_core_loop2:
+	add	r2, r0, #16*4
+	add	r3, r1, #16*4
+	str	r12, [sp, #18*4]
+	mov	r12, sp
+#ifdef __ARM_ARCH_5E_OR_6_OR_7__
+	pld [r1, #24*4]
+	pld [r1, #8*4]
+#endif
+	scrypt_core_macro1b_x4
+	scrypt_core_macro1b_x4
+	scrypt_core_macro1b_x4
+	scrypt_core_macro1b_x4
+	
+	salsa8_core
+	
+	ldr	r0, [sp, #16*4]
+	mov	r12, sp
+	add	r2, r0, #16*4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	
+	salsa8_core
+	
+	ldr	r0, [sp, #16*4]
+	mov	r1, sp
+	ldr	r3, [sp, #17*4]
+	add	r0, r0, #16*4
+	ldr	r2, [sp, #21*4]
+	scrypt_core_macro3_x4
+	and	r4, r4, r2
+	add	r3, r3, r4, lsl #7
+	str	r3, [sp, #19*4]
+#ifdef __ARM_ARCH_5E_OR_6_OR_7__
+	pld	[r3, #16*4]
+	pld	[r3]
+#endif
+	scrypt_core_macro3_x6
+	scrypt_core_macro3_x6
+	
+	ldr	r12, [sp, #18*4]
+	sub	r0, r0, #32*4
+	ldr	r1, [sp, #19*4]
+	subs	r12, r12, #1
+	bne	scrypt_core_loop2
+	
+	scrypt_shuffle
+	
+	ldr	sp, [sp, #20*4]
+#ifdef __thumb__
+	ldmfd	sp!, {r4-r11, lr}
+	bx	lr
+#else
+	ldmfd	sp!, {r4-r11, pc}
+#endif
+
+
+#ifdef __ARM_NEON__
+
+.macro salsa8_core_3way_doubleround
+	ldrd	r6, [sp, #6*4]
+	vadd.u32	q4, q0, q1
+	add	r6, r2, r6
+	vadd.u32	q6, q8, q9
+	add	r7, r3, r7
+	vshl.u32	q5, q4, #7
+	eor	r10, r10, r6, ror #25
+	vshl.u32	q7, q6, #7
+	add	r6, r0, r4
+	vshr.u32	q4, q4, #32-7
+	eor	r11, r11, r7, ror #25
+	vshr.u32	q6, q6, #32-7
+	add	r7, r1, r5
+	veor.u32	q3, q3, q5
+	strd	r10, [sp, #14*4]
+	veor.u32	q11, q11, q7
+	eor	r12, r12, r6, ror #25
+	veor.u32	q3, q3, q4
+	eor	lr, lr, r7, ror #25
+	veor.u32	q11, q11, q6
+	
+	ldrd	r6, [sp, #10*4]
+	vadd.u32	q4, q3, q0
+	add	r2, r10, r2
+	vadd.u32	q6, q11, q8
+	add	r3, r11, r3
+	vshl.u32	q5, q4, #9
+	eor	r6, r6, r2, ror #23
+	vshl.u32	q7, q6, #9
+	add	r2, r12, r0
+	vshr.u32	q4, q4, #32-9
+	eor	r7, r7, r3, ror #23
+	vshr.u32	q6, q6, #32-9
+	add	r3, lr, r1
+	veor.u32	q2, q2, q5
+	strd	r6, [sp, #10*4]
+	veor.u32	q10, q10, q7
+	eor	r8, r8, r2, ror #23
+	veor.u32	q2, q2, q4
+	eor	r9, r9, r3, ror #23
+	veor.u32	q10, q10, q6
+	
+	ldrd	r2, [sp, #6*4]
+	vadd.u32	q4, q2, q3
+	add	r10, r6, r10
+	vadd.u32	q6, q10, q11
+	add	r11, r7, r11
+	vext.u32	q3, q3, q3, #3
+	eor	r2, r2, r10, ror #19
+	vshl.u32	q5, q4, #13
+	add	r10, r8, r12
+	vext.u32	q11, q11, q11, #3
+	eor	r3, r3, r11, ror #19
+	vshl.u32	q7, q6, #13
+	add	r11, r9, lr
+	vshr.u32	q4, q4, #32-13
+	eor	r4, r4, r10, ror #19
+	vshr.u32	q6, q6, #32-13
+	eor	r5, r5, r11, ror #19
+	veor.u32	q1, q1, q5
+	veor.u32	q9, q9, q7
+	veor.u32	q1, q1, q4
+	veor.u32	q9, q9, q6
+	
+	ldrd	r10, [sp, #2*4]
+	vadd.u32	q4, q1, q2
+	add	r6, r2, r6
+	vadd.u32	q6, q9, q10
+	add	r7, r3, r7
+	vswp.u32	d4, d5
+	eor	r10, r10, r6, ror #14
+	vshl.u32	q5, q4, #18
+	add	r6, r4, r8
+	vswp.u32	d20, d21
+	eor	r11, r11, r7, ror #14
+	vshl.u32	q7, q6, #18
+	add	r7, r5, r9
+	vshr.u32	q4, q4, #32-18
+	eor	r0, r0, r6, ror #14
+	vshr.u32	q6, q6, #32-18
+	eor	r1, r1, r7, ror #14
+	veor.u32	q0, q0, q5
+	ldrd	r6, [sp, #14*4]
+	veor.u32	q8, q8, q7
+	veor.u32	q0, q0, q4
+	veor.u32	q8, q8, q6
+	
+	
+	strd	r2, [sp, #6*4]
+	vadd.u32	q4, q0, q3
+	strd	r10, [sp, #2*4]
+	vadd.u32	q6, q8, q11
+	add	r6, r11, r6
+	vext.u32	q1, q1, q1, #1
+	add	r7, r0, r7
+	vshl.u32	q5, q4, #7
+	eor	r4, r4, r6, ror #25
+	vext.u32	q9, q9, q9, #1
+	add	r6, r1, r12
+	vshl.u32	q7, q6, #7
+	eor	r5, r5, r7, ror #25
+	vshr.u32	q4, q4, #32-7
+	add	r7, r10, lr
+	vshr.u32	q6, q6, #32-7
+	eor	r2, r2, r6, ror #25
+	veor.u32	q1, q1, q5
+	eor	r3, r3, r7, ror #25
+	veor.u32	q9, q9, q7
+	strd	r2, [sp, #6*4]
+	veor.u32	q1, q1, q4
+	veor.u32	q9, q9, q6
+	
+	add	r10, r3, r10
+	vadd.u32	q4, q1, q0
+	ldrd	r6, [sp, #10*4]
+	vadd.u32	q6, q9, q8
+	add	r11, r4, r11
+	vshl.u32	q5, q4, #9
+	eor	r8, r8, r10, ror #23
+	vshl.u32	q7, q6, #9
+	add	r10, r5, r0
+	vshr.u32	q4, q4, #32-9
+	eor	r9, r9, r11, ror #23
+	vshr.u32	q6, q6, #32-9
+	add	r11, r2, r1
+	veor.u32	q2, q2, q5
+	eor	r6, r6, r10, ror #23
+	veor.u32	q10, q10, q7
+	eor	r7, r7, r11, ror #23
+	veor.u32	q2, q2, q4
+	strd	r6, [sp, #10*4]
+	veor.u32	q10, q10, q6
+	
+	add	r2, r7, r2
+	vadd.u32	q4, q2, q1
+	ldrd	r10, [sp, #14*4]
+	vadd.u32	q6, q10, q9
+	add	r3, r8, r3
+	vext.u32	q1, q1, q1, #3
+	eor	r12, r12, r2, ror #19
+	vshl.u32	q5, q4, #13
+	add	r2, r9, r4
+	vext.u32	q9, q9, q9, #3
+	eor	lr, lr, r3, ror #19
+	vshl.u32	q7, q6, #13
+	add	r3, r6, r5
+	vshr.u32	q4, q4, #32-13
+	eor	r10, r10, r2, ror #19
+	vshr.u32	q6, q6, #32-13
+	eor	r11, r11, r3, ror #19
+	veor.u32	q3, q3, q5
+	veor.u32	q11, q11, q7
+	veor.u32	q3, q3, q4
+	veor.u32	q11, q11, q6
+	
+	ldrd	r2, [sp, #2*4]
+	vadd.u32	q4, q3, q2
+	add	r6, r11, r6
+	vadd.u32	q6, q11, q10
+	add	r7, r12, r7
+	vswp.u32	d4, d5
+	eor	r0, r0, r6, ror #14
+	vshl.u32	q5, q4, #18
+	add	r6, lr, r8
+	vswp.u32	d20, d21
+	eor	r1, r1, r7, ror #14
+	vshl.u32	q7, q6, #18
+	add	r7, r10, r9
+	vext.u32	q3, q3, q3, #1
+	eor	r2, r2, r6, ror #14
+	vshr.u32	q4, q4, #32-18
+	eor	r3, r3, r7, ror #14
+	vshr.u32	q6, q6, #32-18
+	strd	r2, [sp, #2*4]
+	vext.u32	q11, q11, q11, #1
+	strd	r10, [sp, #14*4]
+	veor.u32	q0, q0, q5
+	veor.u32	q8, q8, q7
+	veor.u32	q0, q0, q4
+	veor.u32	q8, q8, q6
+.endm
+
+.macro salsa8_core_3way
+	ldmia	sp, {r0-r12, lr}
+	ldrd	r10, [sp, #14*4]
+	salsa8_core_3way_doubleround
+	salsa8_core_3way_doubleround
+	salsa8_core_3way_doubleround
+	salsa8_core_3way_doubleround
+	stmia	sp, {r0-r5}
+	strd	r8, [sp, #8*4]
+	str	r12, [sp, #12*4]
+	str	lr, [sp, #13*4]
+.endm
+
+	.text
+	.code 32
+	.align 2
+	.globl scrypt_core_3way
+	.globl _scrypt_core_3way
+#ifdef __ELF__
+	.type scrypt_core_3way, %function
+#endif
+scrypt_core_3way:
+_scrypt_core_3way:
+	stmfd	sp!, {r4-r11, lr}
+	vpush	{q4-q7}
+	mov	r12, sp
+	sub	sp, sp, #24*16
+	bic	sp, sp, #63
+	str	r2, [sp, #4*16+3*4]
+	str	r12, [sp, #4*16+4*4]
+	
+	mov	r3, r0
+	vldmia	r3!, {q8-q15}
+	vmov.u64	q0, #0xffffffff
+	vmov.u32	q1, q8
+	vmov.u32	q2, q12
+	vbif.u32	q8, q9, q0
+	vbif.u32	q12, q13, q0
+	vbif.u32	q9, q10, q0
+	vbif.u32	q13, q14, q0
+	vbif.u32	q10, q11, q0
+	vbif.u32	q14, q15, q0
+	vbif.u32	q11, q1, q0
+	vbif.u32	q15, q2, q0
+	vldmia	r3!, {q0-q7}
+	vswp.u32	d17, d21
+	vswp.u32	d25, d29
+	vswp.u32	d18, d22
+	vswp.u32	d26, d30
+	vstmia	r0, {q8-q15}
+	vmov.u64	q8, #0xffffffff
+	vmov.u32	q9, q0
+	vmov.u32	q10, q4
+	vbif.u32	q0, q1, q8
+	vbif.u32	q4, q5, q8
+	vbif.u32	q1, q2, q8
+	vbif.u32	q5, q6, q8
+	vbif.u32	q2, q3, q8
+	vbif.u32	q6, q7, q8
+	vbif.u32	q3, q9, q8
+	vbif.u32	q7, q10, q8
+	vldmia	r3, {q8-q15}
+	vswp.u32	d1, d5
+	vswp.u32	d9, d13
+	vswp.u32	d2, d6
+	vswp.u32	d10, d14
+	add	r12, sp, #8*16
+	vstmia	r12!, {q0-q7}
+	vmov.u64	q0, #0xffffffff
+	vmov.u32	q1, q8
+	vmov.u32	q2, q12
+	vbif.u32	q8, q9, q0
+	vbif.u32	q12, q13, q0
+	vbif.u32	q9, q10, q0
+	vbif.u32	q13, q14, q0
+	vbif.u32	q10, q11, q0
+	vbif.u32	q14, q15, q0
+	vbif.u32	q11, q1, q0
+	vbif.u32	q15, q2, q0
+	vswp.u32	d17, d21
+	vswp.u32	d25, d29
+	vswp.u32	d18, d22
+	vswp.u32	d26, d30
+	vstmia	r12, {q8-q15}
+	
+	add	lr, sp, #128
+	vldmia	lr, {q0-q7}
+	add	r2, r1, r2, lsl #7
+	str	r0, [sp, #4*16+0*4]
+	str	r2, [sp, #4*16+2*4]
+scrypt_core_3way_loop1:
+	add	lr, r0, #16*4
+	add	r3, r1, #16*4
+	str	r1, [sp, #4*16+1*4]
+	mov	r12, sp
+	scrypt_core_macro1a_x4
+	scrypt_core_macro1a_x4
+	scrypt_core_macro1a_x4
+	ldr	r2, [sp, #4*16+3*4]
+	scrypt_core_macro1a_x4
+	sub	r1, r1, #4*16
+	
+	add	r1, r1, r2, lsl #7
+	vstmia	r1, {q0-q7}
+	add	r3, r1, r2, lsl #7
+	vstmia	r3, {q8-q15}
+	
+	add	lr, sp, #128
+	veor.u32	q0, q0, q4
+	veor.u32	q1, q1, q5
+	veor.u32	q2, q2, q6
+	veor.u32	q3, q3, q7
+	vstmia	lr, {q0-q3}
+	veor.u32	q8, q8, q12
+	veor.u32	q9, q9, q13
+	veor.u32	q10, q10, q14
+	veor.u32	q11, q11, q15
+	add	r12, sp, #256
+	vstmia	r12, {q8-q11}
+	
+	salsa8_core_3way
+	
+	ldr	r0, [sp, #4*16+0*4]
+	mov	r12, sp
+	add	r2, r0, #16*4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	
+	add	lr, sp, #128
+	vldmia	lr, {q4-q7}
+	vadd.u32	q4, q4, q0
+	vadd.u32	q5, q5, q1
+	vadd.u32	q6, q6, q2
+	vadd.u32	q7, q7, q3
+	add	r12, sp, #256
+	vldmia	r12, {q0-q3}
+	vstmia	lr, {q4-q7}
+	vadd.u32	q8, q8, q0
+	vadd.u32	q9, q9, q1
+	vadd.u32	q10, q10, q2
+	vadd.u32	q11, q11, q3
+	
+	add	r4, sp, #128+4*16
+	vldmia	r4, {q0-q3}
+	vstmia	r12, {q8-q11}
+	veor.u32	q0, q0, q4
+	veor.u32	q1, q1, q5
+	veor.u32	q2, q2, q6
+	veor.u32	q3, q3, q7
+	vstmia	r4, {q0-q3}
+	veor.u32	q8, q8, q12
+	veor.u32	q9, q9, q13
+	veor.u32	q10, q10, q14
+	veor.u32	q11, q11, q15
+	vmov	q12, q8
+	vmov	q13, q9
+	vmov	q14, q10
+	vmov	q15, q11
+	
+	salsa8_core_3way
+	
+	ldr	r0, [sp, #4*16+0*4]
+	mov	r1, sp
+	add	r0, r0, #16*4
+	scrypt_core_macro3_x6
+	scrypt_core_macro3_x6
+	scrypt_core_macro3_x4
+	sub	r0, r0, #8*16
+	
+	ldr	r1, [sp, #4*16+1*4]
+	ldr	r2, [sp, #4*16+2*4]
+	add	lr, sp, #128
+	add	r4, sp, #128+4*16
+	vldmia	r4, {q4-q7}
+	vadd.u32	q4, q4, q0
+	vadd.u32	q5, q5, q1
+	vadd.u32	q6, q6, q2
+	vadd.u32	q7, q7, q3
+	vstmia	r4, {q4-q7}
+	vldmia	lr, {q0-q3}
+	vadd.u32	q12, q12, q8
+	vadd.u32	q13, q13, q9
+	vadd.u32	q14, q14, q10
+	vadd.u32	q15, q15, q11
+	add	r12, sp, #256
+	vldmia	r12, {q8-q11}
+	
+	add	r1, r1, #8*16
+	cmp	r1, r2
+	bne	scrypt_core_3way_loop1
+	
+	ldr	r2, [sp, #4*16+3*4]
+	add	r5, sp, #256+4*16
+	vstmia	r5, {q12-q15}
+	
+	sub	r1, r1, r2, lsl #7
+	str	r1, [sp, #4*16+1*4]
+scrypt_core_3way_loop2:
+	str	r2, [sp, #4*16+2*4]
+	
+	ldr	r0, [sp, #4*16+0*4]
+	ldr	r1, [sp, #4*16+1*4]
+	ldr	r2, [sp, #4*16+3*4]
+	ldr	r4, [r0, #16*4]
+	sub	r2, r2, #1
+	and	r4, r4, r2
+	add	r1, r1, r4, lsl #7
+	add	r2, r0, #16*4
+	add	r3, r1, #16*4
+	mov	r12, sp
+	scrypt_core_macro1b_x4
+	scrypt_core_macro1b_x4
+	scrypt_core_macro1b_x4
+	scrypt_core_macro1b_x4
+	
+	ldr	r1, [sp, #4*16+1*4]
+	ldr	r2, [sp, #4*16+3*4]
+	add	r1, r1, r2, lsl #7
+	add	r3, r1, r2, lsl #7
+	sub	r2, r2, #1
+	vmov	r6, r7, d8
+	and	r6, r6, r2
+	add	r6, r1, r6, lsl #7
+	vmov	r7, r8, d24
+	add	lr, sp, #128
+	vldmia	lr, {q0-q3}
+	pld	[r6]
+	pld	[r6, #8*4]
+	pld	[r6, #16*4]
+	pld	[r6, #24*4]
+	vldmia	r6, {q8-q15}
+	and	r7, r7, r2
+	add	r7, r3, r7, lsl #7
+	veor.u32	q8, q8, q0
+	veor.u32	q9, q9, q1
+	veor.u32	q10, q10, q2
+	veor.u32	q11, q11, q3
+	pld	[r7]
+	pld	[r7, #8*4]
+	pld	[r7, #16*4]
+	pld	[r7, #24*4]
+	veor.u32	q12, q12, q4
+	veor.u32	q13, q13, q5
+	veor.u32	q14, q14, q6
+	veor.u32	q15, q15, q7
+	vldmia	r7, {q0-q7}
+	vstmia	lr, {q8-q15}
+	add	r12, sp, #256
+	vldmia	r12, {q8-q15}
+	veor.u32	q8, q8, q0
+	veor.u32	q9, q9, q1
+	veor.u32	q10, q10, q2
+	veor.u32	q11, q11, q3
+	veor.u32	q12, q12, q4
+	veor.u32	q13, q13, q5
+	veor.u32	q14, q14, q6
+	veor.u32	q15, q15, q7
+	
+	vldmia	lr, {q0-q7}
+	veor.u32	q0, q0, q4
+	veor.u32	q1, q1, q5
+	veor.u32	q2, q2, q6
+	veor.u32	q3, q3, q7
+	vstmia	lr, {q0-q3}
+	veor.u32	q8, q8, q12
+	veor.u32	q9, q9, q13
+	veor.u32	q10, q10, q14
+	veor.u32	q11, q11, q15
+	vstmia	r12, {q8-q15}
+	
+	salsa8_core_3way
+	
+	ldr	r0, [sp, #4*16+0*4]
+	mov	r12, sp
+	add	r2, r0, #16*4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	scrypt_core_macro2_x4
+	
+	add	lr, sp, #128
+	vldmia	lr, {q4-q7}
+	vadd.u32	q4, q4, q0
+	vadd.u32	q5, q5, q1
+	vadd.u32	q6, q6, q2
+	vadd.u32	q7, q7, q3
+	add	r12, sp, #256
+	vldmia	r12, {q12-q15}
+	vstmia	lr, {q4-q7}
+	vadd.u32	q12, q12, q8
+	vadd.u32	q13, q13, q9
+	vadd.u32	q14, q14, q10
+	vadd.u32	q15, q15, q11
+	
+	add	r4, sp, #128+4*16
+	vldmia	r4, {q0-q3}
+	vstmia	r12, {q12-q15}
+	veor.u32	q0, q0, q4
+	veor.u32	q1, q1, q5
+	veor.u32	q2, q2, q6
+	veor.u32	q3, q3, q7
+	add	r5, sp, #256+4*16
+	vldmia	r5, {q8-q11}
+	vstmia	r4, {q0-q3}
+	veor.u32	q8, q8, q12
+	veor.u32	q9, q9, q13
+	veor.u32	q10, q10, q14
+	veor.u32	q11, q11, q15
+	vmov	q12, q8
+	vmov	q13, q9
+	vmov	q14, q10
+	vmov	q15, q11
+	
+	salsa8_core_3way
+	
+	ldr	r0, [sp, #4*16+0*4]
+	ldr	r3, [sp, #4*16+1*4]
+	ldr	r2, [sp, #4*16+3*4]
+	mov	r1, sp
+	add	r0, r0, #16*4
+	sub	r2, r2, #1
+	scrypt_core_macro3_x4
+	and	r4, r4, r2
+	add	r3, r3, r4, lsl #7
+	pld	[r3, #16*4]
+	pld	[r3]
+	pld	[r3, #24*4]
+	pld	[r3, #8*4]
+	scrypt_core_macro3_x6
+	scrypt_core_macro3_x6
+	
+	add	lr, sp, #128
+	add	r4, sp, #128+4*16
+	vldmia	r4, {q4-q7}
+	vadd.u32	q4, q4, q0
+	vadd.u32	q5, q5, q1
+	vadd.u32	q6, q6, q2
+	vadd.u32	q7, q7, q3
+	vstmia	r4, {q4-q7}
+	vadd.u32	q12, q12, q8
+	vadd.u32	q13, q13, q9
+	vadd.u32	q14, q14, q10
+	vadd.u32	q15, q15, q11
+	add	r5, sp, #256+4*16
+	vstmia	r5, {q12-q15}
+	
+	ldr	r2, [sp, #4*16+2*4]
+	subs	r2, r2, #1
+	bne	scrypt_core_3way_loop2
+	
+	ldr	r0, [sp, #4*16+0*4]
+	vldmia	r0, {q8-q15}
+	vmov.u64	q0, #0xffffffff
+	vmov.u32	q1, q8
+	vmov.u32	q2, q12
+	vbif.u32	q8, q9, q0
+	vbif.u32	q12, q13, q0
+	vbif.u32	q9, q10, q0
+	vbif.u32	q13, q14, q0
+	vbif.u32	q10, q11, q0
+	vbif.u32	q14, q15, q0
+	vbif.u32	q11, q1, q0
+	vbif.u32	q15, q2, q0
+	add	r12, sp, #8*16
+	vldmia	r12!, {q0-q7}
+	vswp.u32	d17, d21
+	vswp.u32	d25, d29
+	vswp.u32	d18, d22
+	vswp.u32	d26, d30
+	vstmia	r0!, {q8-q15}
+	vmov.u64	q8, #0xffffffff
+	vmov.u32	q9, q0
+	vmov.u32	q10, q4
+	vbif.u32	q0, q1, q8
+	vbif.u32	q4, q5, q8
+	vbif.u32	q1, q2, q8
+	vbif.u32	q5, q6, q8
+	vbif.u32	q2, q3, q8
+	vbif.u32	q6, q7, q8
+	vbif.u32	q3, q9, q8
+	vbif.u32	q7, q10, q8
+	vldmia	r12, {q8-q15}
+	vswp.u32	d1, d5
+	vswp.u32	d9, d13
+	vswp.u32	d2, d6
+	vswp.u32	d10, d14
+	vstmia	r0!, {q0-q7}
+	vmov.u64	q0, #0xffffffff
+	vmov.u32	q1, q8
+	vmov.u32	q2, q12
+	vbif.u32	q8, q9, q0
+	vbif.u32	q12, q13, q0
+	vbif.u32	q9, q10, q0
+	vbif.u32	q13, q14, q0
+	vbif.u32	q10, q11, q0
+	vbif.u32	q14, q15, q0
+	vbif.u32	q11, q1, q0
+	vbif.u32	q15, q2, q0
+	vswp.u32	d17, d21
+	vswp.u32	d25, d29
+	vswp.u32	d18, d22
+	vswp.u32	d26, d30
+	vstmia	r0, {q8-q15}
+	
+	ldr	sp, [sp, #4*16+4*4]
+	vpop	{q4-q7}
+	ldmfd	sp!, {r4-r11, pc}
+
+#endif /* __ARM_NEON__ */
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-ppc.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-ppc.S
new file mode 100644
index 0000000..47ef643
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-ppc.S
@@ -0,0 +1,1148 @@
+/*
+ * Copyright 2014-2015 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(USE_ASM) && (defined(__powerpc__) || defined(__ppc__) || defined(__PPC__))
+
+#ifndef __APPLE__
+
+#define r0 0
+#define r1 1
+#define r2 2
+#define r3 3
+#define r4 4
+#define r5 5
+#define r6 6
+#define r7 7
+#define r8 8
+#define r9 9
+#define r10 10
+#define r11 11
+#define r12 12
+#define r13 13
+#define r14 14
+#define r15 15
+#define r16 16
+#define r17 17
+#define r18 18
+#define r19 19
+#define r20 20
+#define r21 21
+#define r22 22
+#define r23 23
+#define r24 24
+#define r25 25
+#define r26 26
+#define r27 27
+#define r28 28
+#define r29 29
+#define r30 30
+#define r31 31
+
+#ifdef __ALTIVEC__
+#define v0 0
+#define v1 1
+#define v2 2
+#define v3 3
+#define v4 4
+#define v5 5
+#define v6 6
+#define v7 7
+#define v8 8
+#define v9 9
+#define v10 10
+#define v11 11
+#define v12 12
+#define v13 13
+#define v14 14
+#define v15 15
+#define v16 16
+#define v17 17
+#define v18 18
+#define v19 19
+#define v20 20
+#define v21 21
+#define v22 22
+#define v23 23
+#define v24 24
+#define v25 25
+#define v26 26
+#define v27 27
+#define v28 28
+#define v29 29
+#define v30 30
+#define v31 31
+#endif
+
+#endif
+
+#if !(defined(__powerpc64__) || defined(__ppc64__) || defined(__PPC64__) || \
+	defined(__64BIT__) || defined(_LP64) || defined(__LP64__))
+#define ld lwz
+#define std stw
+#define stdu stwu
+#define stdux stwux
+#endif
+
+
+#ifdef __ALTIVEC__
+
+#ifdef __APPLE__
+	.machine ppc7400
+#endif
+
+.macro salsa8_core_doubleround
+	vadduwm	v4, v0, v1
+	vrlw	v4, v4, v16
+	vxor	v3, v3, v4
+	
+	vadduwm	v4, v3, v0
+	vrlw	v4, v4, v17
+	vxor	v2, v2, v4
+	
+	vadduwm	v4, v2, v3
+	vrlw	v4, v4, v18
+	vsldoi	v3, v3, v3, 12
+	vxor	v1, v1, v4
+	
+	vadduwm	v4, v1, v2
+	vrlw	v4, v4, v19
+	vsldoi	v1, v1, v1, 4
+	vxor	v0, v0, v4
+	
+	vadduwm	v4, v0, v3
+	vrlw	v4, v4, v16
+	vsldoi	v2, v2, v2, 8
+	vxor	v1, v1, v4
+	
+	vadduwm	v4, v1, v0
+	vrlw	v4, v4, v17
+	vxor	v2, v2, v4
+	
+	vadduwm	v4, v2, v1
+	vrlw	v4, v4, v18
+	vsldoi	v1, v1, v1, 12
+	vxor	v3, v3, v4
+	
+	vadduwm	v4, v3, v2
+	vrlw	v4, v4, v19
+	vsldoi	v3, v3, v3, 4
+	vxor	v0, v0, v4
+	vsldoi	v2, v2, v2, 8
+.endm
+
+.macro salsa8_core
+	salsa8_core_doubleround
+	salsa8_core_doubleround
+	salsa8_core_doubleround
+	salsa8_core_doubleround
+.endm
+
+#ifdef _AIX
+	.csect .text[PR]
+#else
+	.text
+#endif
+	.align 2
+	.globl scrypt_core
+	.globl _scrypt_core
+	.globl .scrypt_core
+#ifdef __ELF__
+	.type scrypt_core, %function
+#endif
+scrypt_core:
+_scrypt_core:
+.scrypt_core:
+	stdu	r1, -4*4(r1)
+	mfspr	r0, 256
+	std	r0, 2*4(r1)
+	oris	r0, r0, 0xffff
+	ori	r0, r0, 0xf000
+	mtspr	256, r0
+	
+	li	r6, 1*16
+	li	r7, 2*16
+	li	r8, 3*16
+	li	r9, 4*16
+	li	r10, 5*16
+	li	r11, 6*16
+	li	r12, 7*16
+	
+	lvx	v8, 0, r3
+	lvx	v9, r3, r6
+	lvx	v10, r3, r7
+	lvx	v11, r3, r8
+	lvx	v12, r3, r9
+	lvx	v13, r3, r10
+	lvx	v14, r3, r11
+	lvx	v15, r3, r12
+	
+	vxor	v0, v0, v0
+	vnor	v1, v0, v0
+	vsldoi	v2, v0, v1, 4
+	vsldoi	v3, v2, v0, 8
+	vor	v3, v3, v2
+	vsldoi	v1, v0, v1, 8
+	
+	vor	v4, v8, v8
+	vsel	v8, v8, v9, v3
+	vsel	v9, v9, v10, v3
+	vsel	v10, v10, v11, v3
+	vsel	v11, v11, v4, v3
+	vor	v4, v8, v8
+	vor	v5, v9, v9
+	vsel	v8, v8, v10, v1
+	vsel	v9, v11, v9, v1
+	vsel	v10, v10, v4, v1
+	vsel	v11, v5, v11, v1
+	
+	vor	v4, v12, v12
+	vsel	v12, v12, v13, v3
+	vsel	v13, v13, v14, v3
+	vsel	v14, v14, v15, v3
+	vsel	v15, v15, v4, v3
+	vor	v4, v12, v12
+	vor	v5, v13, v13
+	vsel	v12, v12, v14, v1
+	vsel	v13, v15, v13, v1
+	vsel	v14, v14, v4, v1
+	vsel	v15, v5, v15, v1
+	
+	vspltisw	v16, 7
+	vspltisw	v17, 9
+	vspltisw	v18, 13
+	vadduwm	v19, v17, v17
+	
+	mtctr	r5
+scrypt_core_loop1:
+	vxor	v8, v8, v12
+	stvx	v8, 0, r4
+	vxor	v9, v9, v13
+	stvx	v9, r4, r6
+	vxor	v10, v10, v14
+	stvx	v10, r4, r7
+	vxor	v11, v11, v15
+	stvx	v11, r4, r8
+	vor	v0, v8, v8
+	stvx	v12, r4, r9
+	vor	v1, v9, v9
+	stvx	v13, r4, r10
+	vor	v2, v10, v10
+	stvx	v14, r4, r11
+	vor	v3, v11, v11
+	stvx	v15, r4, r12
+	
+	salsa8_core
+	
+	vadduwm	v8, v8, v0
+	vadduwm	v9, v9, v1
+	vadduwm	v10, v10, v2
+	vadduwm	v11, v11, v3
+	
+	vxor	v12, v12, v8
+	vxor	v13, v13, v9
+	vxor	v14, v14, v10
+	vxor	v15, v15, v11
+	vor	v0, v12, v12
+	vor	v1, v13, v13
+	vor	v2, v14, v14
+	vor	v3, v15, v15
+	
+	salsa8_core
+	
+	vadduwm	v12, v12, v0
+	vadduwm	v13, v13, v1
+	vadduwm	v14, v14, v2
+	vadduwm	v15, v15, v3
+	
+	addi	r4, r4, 32*4
+	bdnz	scrypt_core_loop1
+	
+	stvx	v12, 0, r3
+	slwi	r6, r5, 7
+	subf	r4, r6, r4
+	mtctr	r5
+	addi	r5, r5, -1
+	addi	r7, r4, 1*16
+	addi	r8, r4, 2*16
+	addi	r9, r4, 3*16
+scrypt_core_loop2:
+	lwz	r6, 0(r3)
+	and	r6, r6, r5
+	slwi	r6, r6, 7
+	lvx	v0, r4, r6
+	vxor	v8, v8, v12
+	lvx	v1, r7, r6
+	vxor	v9, v9, v13
+	lvx	v2, r8, r6
+	vxor	v10, v10, v14
+	lvx	v3, r9, r6
+	vxor	v11, v11, v15
+	vxor	v0, v0, v8
+	vxor	v1, v1, v9
+	vxor	v2, v2, v10
+	vxor	v3, v3, v11
+	addi	r6, r6, 64
+	vor	v8, v0, v0
+	vor	v9, v1, v1
+	lvx	v5, r4, r6
+	vor	v10, v2, v2
+	lvx	v6, r7, r6
+	vor	v11, v3, v3
+	lvx	v7, r8, r6
+	
+	salsa8_core
+	
+	vadduwm	v8, v8, v0
+	lvx	v0, r9, r6
+	vadduwm	v9, v9, v1
+	vadduwm	v10, v10, v2
+	vadduwm	v11, v11, v3
+	
+	vxor	v12, v12, v5
+	vxor	v13, v13, v6
+	vxor	v14, v14, v7
+	vxor	v15, v15, v0
+	vxor	v12, v12, v8
+	vxor	v13, v13, v9
+	vxor	v14, v14, v10
+	vxor	v15, v15, v11
+	vor	v0, v12, v12
+	vor	v1, v13, v13
+	vor	v2, v14, v14
+	vor	v3, v15, v15
+	
+	salsa8_core
+	
+	vadduwm	v12, v12, v0
+	stvx	v12, 0, r3
+	vadduwm	v13, v13, v1
+	vadduwm	v14, v14, v2
+	vadduwm	v15, v15, v3
+	
+	bdnz	scrypt_core_loop2
+	
+	vxor	v0, v0, v0
+	vnor	v1, v0, v0
+	vsldoi	v2, v0, v1, 4
+	vsldoi	v3, v2, v0, 8
+	vor	v3, v3, v2
+	vsldoi	v1, v0, v1, 8
+	
+	vor	v4, v8, v8
+	vsel	v8, v8, v9, v3
+	vsel	v9, v9, v10, v3
+	vsel	v10, v10, v11, v3
+	vsel	v11, v11, v4, v3
+	vor	v4, v8, v8
+	vor	v5, v9, v9
+	vsel	v8, v8, v10, v1
+	vsel	v9, v11, v9, v1
+	vsel	v10, v10, v4, v1
+	vsel	v11, v5, v11, v1
+	
+	vor	v4, v12, v12
+	vsel	v12, v12, v13, v3
+	vsel	v13, v13, v14, v3
+	vsel	v14, v14, v15, v3
+	vsel	v15, v15, v4, v3
+	vor	v4, v12, v12
+	vor	v5, v13, v13
+	vsel	v12, v12, v14, v1
+	vsel	v13, v15, v13, v1
+	vsel	v14, v14, v4, v1
+	vsel	v15, v5, v15, v1
+	
+	li	r6, 1*16
+	li	r7, 2*16
+	li	r8, 3*16
+	li	r9, 4*16
+	
+	stvx	v8, 0, r3
+	stvx	v9, r3, r6
+	stvx	v10, r3, r7
+	stvx	v11, r3, r8
+	stvx	v12, r3, r9
+	stvx	v13, r3, r10
+	stvx	v14, r3, r11
+	stvx	v15, r3, r12
+	
+	ld	r0, 2*4(r1)
+	mtspr	256, r0
+	addi	r1, r1, 4*4
+	blr
+
+#else /* __ALTIVEC__ */
+
+.macro salsa8_core_doubleround
+	add	r0, r16, r28
+	add	r5, r21, r17
+	add	r6, r26, r22
+	add	r7, r31, r27
+	rotlwi	r0, r0, 7
+	rotlwi	r5, r5, 7
+	rotlwi	r6, r6, 7
+	rotlwi	r7, r7, 7
+	xor	r20, r20, r0
+	xor	r25, r25, r5
+	xor	r30, r30, r6
+	xor	r19, r19, r7
+	
+	add	r0, r20, r16
+	add	r5, r25, r21
+	add	r6, r30, r26
+	add	r7, r19, r31
+	rotlwi	r0, r0, 9
+	rotlwi	r5, r5, 9
+	rotlwi	r6, r6, 9
+	rotlwi	r7, r7, 9
+	xor	r24, r24, r0
+	xor	r29, r29, r5
+	xor	r18, r18, r6
+	xor	r23, r23, r7
+	
+	add	r0, r24, r20
+	add	r5, r29, r25
+	add	r6, r18, r30
+	add	r7, r23, r19
+	rotlwi	r0, r0, 13
+	rotlwi	r5, r5, 13
+	rotlwi	r6, r6, 13
+	rotlwi	r7, r7, 13
+	xor	r28, r28, r0
+	xor	r17, r17, r5
+	xor	r22, r22, r6
+	xor	r27, r27, r7
+	
+	add	r0, r28, r24
+	add	r5, r17, r29
+	add	r6, r22, r18
+	add	r7, r27, r23
+	rotlwi	r0, r0, 18
+	rotlwi	r5, r5, 18
+	rotlwi	r6, r6, 18
+	rotlwi	r7, r7, 18
+	xor	r16, r16, r0
+	xor	r21, r21, r5
+	xor	r26, r26, r6
+	xor	r31, r31, r7
+	
+	add	r0, r16, r19
+	add	r5, r21, r20
+	add	r6, r26, r25
+	add	r7, r31, r30
+	rotlwi	r0, r0, 7
+	rotlwi	r5, r5, 7
+	rotlwi	r6, r6, 7
+	rotlwi	r7, r7, 7
+	xor	r17, r17, r0
+	xor	r22, r22, r5
+	xor	r27, r27, r6
+	xor	r28, r28, r7
+	
+	add	r0, r17, r16
+	add	r5, r22, r21
+	add	r6, r27, r26
+	add	r7, r28, r31
+	rotlwi	r0, r0, 9
+	rotlwi	r5, r5, 9
+	rotlwi	r6, r6, 9
+	rotlwi	r7, r7, 9
+	xor	r18, r18, r0
+	xor	r23, r23, r5
+	xor	r24, r24, r6
+	xor	r29, r29, r7
+	
+	add	r0, r18, r17
+	add	r5, r23, r22
+	add	r6, r24, r27
+	add	r7, r29, r28
+	rotlwi	r0, r0, 13
+	rotlwi	r5, r5, 13
+	rotlwi	r6, r6, 13
+	rotlwi	r7, r7, 13
+	xor	r19, r19, r0
+	xor	r20, r20, r5
+	xor	r25, r25, r6
+	xor	r30, r30, r7
+	
+	add	r0, r19, r18
+	add	r5, r20, r23
+	add	r6, r25, r24
+	add	r7, r30, r29
+	rotlwi	r0, r0, 18
+	rotlwi	r5, r5, 18
+	rotlwi	r6, r6, 18
+	rotlwi	r7, r7, 18
+	xor	r16, r16, r0
+	xor	r21, r21, r5
+	xor	r26, r26, r6
+	xor	r31, r31, r7
+.endm
+
+.macro salsa8_core
+	salsa8_core_doubleround
+	salsa8_core_doubleround
+	salsa8_core_doubleround
+	salsa8_core_doubleround
+.endm
+
+#ifdef _AIX
+	.csect .text[PR]
+#else
+	.text
+#endif
+	.align 2
+	.globl scrypt_core
+	.globl _scrypt_core
+	.globl .scrypt_core
+#ifdef __ELF__
+	.type scrypt_core, %function
+#endif
+scrypt_core:
+_scrypt_core:
+.scrypt_core:
+	stdu	r1, -68*4(r1)
+	stw	r5, 2*4(r1)
+	std	r13, 4*4(r1)
+	std	r14, 6*4(r1)
+	std	r15, 8*4(r1)
+	std	r16, 10*4(r1)
+	std	r17, 12*4(r1)
+	std	r18, 14*4(r1)
+	std	r19, 16*4(r1)
+	std	r20, 18*4(r1)
+	std	r21, 20*4(r1)
+	std	r3, 22*4(r1)
+	std	r22, 48*4(r1)
+	std	r23, 50*4(r1)
+	std	r24, 52*4(r1)
+	std	r25, 54*4(r1)
+	std	r26, 56*4(r1)
+	std	r27, 58*4(r1)
+	std	r28, 60*4(r1)
+	std	r29, 62*4(r1)
+	std	r30, 64*4(r1)
+	std	r31, 66*4(r1)
+	
+	lwz	r16, 0*4(r3)
+	lwz	r17, 1*4(r3)
+	lwz	r18, 2*4(r3)
+	lwz	r19, 3*4(r3)
+	lwz	r20, 4*4(r3)
+	lwz	r21, 5*4(r3)
+	lwz	r22, 6*4(r3)
+	lwz	r23, 7*4(r3)
+	stw	r16, 24*4(r1)
+	stw	r17, 25*4(r1)
+	stw	r18, 26*4(r1)
+	stw	r19, 27*4(r1)
+	stw	r20, 28*4(r1)
+	stw	r21, 29*4(r1)
+	stw	r22, 30*4(r1)
+	stw	r23, 31*4(r1)
+	lwz	r24, 8*4(r3)
+	lwz	r25, 9*4(r3)
+	lwz	r26, 10*4(r3)
+	lwz	r27, 11*4(r3)
+	lwz	r28, 12*4(r3)
+	lwz	r29, 13*4(r3)
+	lwz	r30, 14*4(r3)
+	lwz	r31, 15*4(r3)
+	stw	r24, 32*4(r1)
+	stw	r25, 33*4(r1)
+	stw	r26, 34*4(r1)
+	stw	r27, 35*4(r1)
+	stw	r28, 36*4(r1)
+	stw	r29, 37*4(r1)
+	stw	r30, 38*4(r1)
+	stw	r31, 39*4(r1)
+	lwz	r16, 16*4(r3)
+	lwz	r17, 17*4(r3)
+	lwz	r18, 18*4(r3)
+	lwz	r19, 19*4(r3)
+	lwz	r20, 20*4(r3)
+	lwz	r21, 21*4(r3)
+	lwz	r22, 22*4(r3)
+	lwz	r23, 23*4(r3)
+	stw	r16, 40*4(r1)
+	stw	r17, 41*4(r1)
+	stw	r18, 42*4(r1)
+	stw	r19, 43*4(r1)
+	stw	r20, 44*4(r1)
+	stw	r21, 45*4(r1)
+	stw	r22, 46*4(r1)
+	stw	r23, 47*4(r1)
+	lwz	r8, 24*4(r3)
+	lwz	r9, 25*4(r3)
+	lwz	r10, 26*4(r3)
+	lwz	r11, 27*4(r3)
+	lwz	r12, 28*4(r3)
+	lwz	r13, 29*4(r3)
+	lwz	r14, 30*4(r3)
+	lwz	r15, 31*4(r3)
+	
+	mtctr	r5
+scrypt_core_loop1:
+	lwz	r16, 24*4(r1)
+	lwz	r17, 25*4(r1)
+	lwz	r18, 26*4(r1)
+	lwz	r19, 27*4(r1)
+	lwz	r20, 28*4(r1)
+	lwz	r21, 29*4(r1)
+	lwz	r22, 30*4(r1)
+	lwz	r23, 31*4(r1)
+	lwz	r24, 32*4(r1)
+	lwz	r25, 33*4(r1)
+	lwz	r26, 34*4(r1)
+	lwz	r27, 35*4(r1)
+	lwz	r28, 36*4(r1)
+	lwz	r29, 37*4(r1)
+	lwz	r30, 38*4(r1)
+	lwz	r31, 39*4(r1)
+	
+	lwz	r0, 40*4(r1)
+	lwz	r5, 41*4(r1)
+	lwz	r6, 42*4(r1)
+	lwz	r7, 43*4(r1)
+	xor	r16, r16, r0
+	xor	r17, r17, r5
+	xor	r18, r18, r6
+	xor	r19, r19, r7
+	stw	r16, 0*4(r4)
+	stw	r17, 1*4(r4)
+	stw	r18, 2*4(r4)
+	stw	r19, 3*4(r4)
+	stw	r0, 16*4(r4)
+	stw	r5, 17*4(r4)
+	stw	r6, 18*4(r4)
+	stw	r7, 19*4(r4)
+	
+	lwz	r0, 44*4(r1)
+	lwz	r5, 45*4(r1)
+	lwz	r6, 46*4(r1)
+	lwz	r7, 47*4(r1)
+	xor	r20, r20, r0
+	xor	r21, r21, r5
+	xor	r22, r22, r6
+	xor	r23, r23, r7
+	stw	r0, 20*4(r4)
+	stw	r5, 21*4(r4)
+	stw	r6, 22*4(r4)
+	stw	r7, 23*4(r4)
+	stw	r20, 4*4(r4)
+	stw	r21, 5*4(r4)
+	stw	r22, 6*4(r4)
+	stw	r23, 7*4(r4)
+	
+	xor	r24, r24, r8
+	xor	r25, r25, r9
+	xor	r26, r26, r10
+	xor	r27, r27, r11
+	xor	r28, r28, r12
+	xor	r29, r29, r13
+	xor	r30, r30, r14
+	xor	r31, r31, r15
+	stw	r24, 8*4(r4)
+	stw	r25, 9*4(r4)
+	stw	r26, 10*4(r4)
+	stw	r27, 11*4(r4)
+	stw	r28, 12*4(r4)
+	stw	r29, 13*4(r4)
+	stw	r30, 14*4(r4)
+	stw	r31, 15*4(r4)
+	stw	r8, 24*4(r4)
+	stw	r9, 25*4(r4)
+	stw	r10, 26*4(r4)
+	stw	r11, 27*4(r4)
+	stw	r12, 28*4(r4)
+	stw	r13, 29*4(r4)
+	stw	r14, 30*4(r4)
+	stw	r15, 31*4(r4)
+	
+	salsa8_core
+	
+	lwz	r0, 0*4(r4)
+	lwz	r5, 1*4(r4)
+	lwz	r6, 2*4(r4)
+	lwz	r7, 3*4(r4)
+	add	r16, r16, r0
+	add	r17, r17, r5
+	add	r18, r18, r6
+	add	r19, r19, r7
+	lwz	r0, 4*4(r4)
+	lwz	r5, 5*4(r4)
+	lwz	r6, 6*4(r4)
+	lwz	r7, 7*4(r4)
+	add	r20, r20, r0
+	add	r21, r21, r5
+	add	r22, r22, r6
+	add	r23, r23, r7
+	lwz	r0, 8*4(r4)
+	lwz	r5, 9*4(r4)
+	lwz	r6, 10*4(r4)
+	lwz	r7, 11*4(r4)
+	add	r24, r24, r0
+	add	r25, r25, r5
+	add	r26, r26, r6
+	add	r27, r27, r7
+	lwz	r0, 12*4(r4)
+	lwz	r5, 13*4(r4)
+	lwz	r6, 14*4(r4)
+	lwz	r7, 15*4(r4)
+	add	r28, r28, r0
+	add	r29, r29, r5
+	add	r30, r30, r6
+	add	r31, r31, r7
+	
+	stw	r16, 24*4(r1)
+	stw	r17, 25*4(r1)
+	stw	r18, 26*4(r1)
+	stw	r19, 27*4(r1)
+	stw	r20, 28*4(r1)
+	stw	r21, 29*4(r1)
+	stw	r22, 30*4(r1)
+	stw	r23, 31*4(r1)
+	stw	r24, 32*4(r1)
+	stw	r25, 33*4(r1)
+	stw	r26, 34*4(r1)
+	stw	r27, 35*4(r1)
+	stw	r28, 36*4(r1)
+	stw	r29, 37*4(r1)
+	stw	r30, 38*4(r1)
+	stw	r31, 39*4(r1)
+	
+	lwz	r0, 40*4(r1)
+	lwz	r5, 41*4(r1)
+	lwz	r6, 42*4(r1)
+	lwz	r7, 43*4(r1)
+	xor	r16, r16, r0
+	xor	r17, r17, r5
+	xor	r18, r18, r6
+	xor	r19, r19, r7
+	lwz	r0, 44*4(r1)
+	lwz	r5, 45*4(r1)
+	lwz	r6, 46*4(r1)
+	lwz	r7, 47*4(r1)
+	xor	r20, r20, r0
+	xor	r21, r21, r5
+	xor	r22, r22, r6
+	xor	r23, r23, r7
+	xor	r24, r24, r8
+	xor	r25, r25, r9
+	xor	r26, r26, r10
+	xor	r27, r27, r11
+	xor	r28, r28, r12
+	xor	r29, r29, r13
+	xor	r30, r30, r14
+	xor	r31, r31, r15
+	stw	r16, 40*4(r1)
+	stw	r17, 41*4(r1)
+	stw	r18, 42*4(r1)
+	stw	r19, 43*4(r1)
+	mr	r8, r24
+	mr	r9, r25
+	mr	r10, r26
+	mr	r11, r27
+	stw	r20, 44*4(r1)
+	stw	r21, 45*4(r1)
+	stw	r22, 46*4(r1)
+	stw	r23, 47*4(r1)
+	mr	r12, r28
+	mr	r13, r29
+	mr	r14, r30
+	mr	r15, r31
+	
+	salsa8_core
+	
+	lwz	r0, 40*4(r1)
+	lwz	r5, 41*4(r1)
+	lwz	r6, 42*4(r1)
+	lwz	r7, 43*4(r1)
+	add	r16, r16, r0
+	add	r17, r17, r5
+	add	r18, r18, r6
+	add	r19, r19, r7
+	lwz	r0, 44*4(r1)
+	lwz	r5, 45*4(r1)
+	lwz	r6, 46*4(r1)
+	lwz	r7, 47*4(r1)
+	add	r20, r20, r0
+	add	r21, r21, r5
+	add	r22, r22, r6
+	add	r23, r23, r7
+	add	r8, r8, r24
+	add	r9, r9, r25
+	add	r10, r10, r26
+	add	r11, r11, r27
+	stw	r16, 40*4(r1)
+	stw	r17, 41*4(r1)
+	stw	r18, 42*4(r1)
+	stw	r19, 43*4(r1)
+	add	r12, r12, r28
+	add	r13, r13, r29
+	add	r14, r14, r30
+	add	r15, r15, r31
+	stw	r20, 44*4(r1)
+	stw	r21, 45*4(r1)
+	stw	r22, 46*4(r1)
+	stw	r23, 47*4(r1)
+	
+	addi	r4, r4, 32*4
+	bdnz	scrypt_core_loop1
+	
+	lwz	r5, 2*4(r1)
+	slwi	r3, r5, 7
+	subf	r4, r3, r4
+	mtctr	r5
+	addi	r5, r5, -1
+	stw	r5, 2*4(r1)
+scrypt_core_loop2:
+	and	r3, r16, r5
+	slwi	r3, r3, 7
+	add	r3, r3, r4
+	mr	r0, r16
+	mr	r5, r17
+	mr	r6, r18
+	mr	r7, r19
+	lwz	r16, 24*4(r1)
+	lwz	r17, 25*4(r1)
+	lwz	r18, 26*4(r1)
+	lwz	r19, 27*4(r1)
+	lwz	r20, 28*4(r1)
+	lwz	r21, 29*4(r1)
+	lwz	r22, 30*4(r1)
+	lwz	r23, 31*4(r1)
+	lwz	r24, 32*4(r1)
+	lwz	r25, 33*4(r1)
+	lwz	r26, 34*4(r1)
+	lwz	r27, 35*4(r1)
+	lwz	r28, 36*4(r1)
+	lwz	r29, 37*4(r1)
+	lwz	r30, 38*4(r1)
+	lwz	r31, 39*4(r1)
+	xor	r16, r16, r0
+	xor	r17, r17, r5
+	xor	r18, r18, r6
+	xor	r19, r19, r7
+	lwz	r0, 44*4(r1)
+	lwz	r5, 45*4(r1)
+	lwz	r6, 46*4(r1)
+	lwz	r7, 47*4(r1)
+	xor	r20, r20, r0
+	xor	r21, r21, r5
+	xor	r22, r22, r6
+	xor	r23, r23, r7
+	xor	r24, r24, r8
+	xor	r25, r25, r9
+	xor	r26, r26, r10
+	xor	r27, r27, r11
+	xor	r28, r28, r12
+	xor	r29, r29, r13
+	xor	r30, r30, r14
+	xor	r31, r31, r15
+	
+	lwz	r0, 0*4(r3)
+	lwz	r5, 1*4(r3)
+	lwz	r6, 2*4(r3)
+	lwz	r7, 3*4(r3)
+	xor	r16, r16, r0
+	xor	r17, r17, r5
+	xor	r18, r18, r6
+	xor	r19, r19, r7
+	lwz	r0, 4*4(r3)
+	lwz	r5, 5*4(r3)
+	lwz	r6, 6*4(r3)
+	lwz	r7, 7*4(r3)
+	xor	r20, r20, r0
+	xor	r21, r21, r5
+	xor	r22, r22, r6
+	xor	r23, r23, r7
+	lwz	r0, 8*4(r3)
+	lwz	r5, 9*4(r3)
+	lwz	r6, 10*4(r3)
+	lwz	r7, 11*4(r3)
+	xor	r24, r24, r0
+	xor	r25, r25, r5
+	xor	r26, r26, r6
+	xor	r27, r27, r7
+	lwz	r0, 12*4(r3)
+	lwz	r5, 13*4(r3)
+	lwz	r6, 14*4(r3)
+	lwz	r7, 15*4(r3)
+	xor	r28, r28, r0
+	xor	r29, r29, r5
+	xor	r30, r30, r6
+	xor	r31, r31, r7
+	
+	stw	r16, 24*4(r1)
+	stw	r17, 25*4(r1)
+	stw	r18, 26*4(r1)
+	stw	r19, 27*4(r1)
+	stw	r20, 28*4(r1)
+	stw	r21, 29*4(r1)
+	stw	r22, 30*4(r1)
+	stw	r23, 31*4(r1)
+	stw	r24, 32*4(r1)
+	stw	r25, 33*4(r1)
+	stw	r26, 34*4(r1)
+	stw	r27, 35*4(r1)
+	stw	r28, 36*4(r1)
+	stw	r29, 37*4(r1)
+	stw	r30, 38*4(r1)
+	stw	r31, 39*4(r1)
+	
+	salsa8_core
+	
+	lwz	r0, 24*4(r1)
+	lwz	r5, 25*4(r1)
+	lwz	r6, 26*4(r1)
+	lwz	r7, 27*4(r1)
+	add	r16, r16, r0
+	add	r17, r17, r5
+	add	r18, r18, r6
+	add	r19, r19, r7
+	lwz	r0, 28*4(r1)
+	lwz	r5, 29*4(r1)
+	lwz	r6, 30*4(r1)
+	lwz	r7, 31*4(r1)
+	add	r20, r20, r0
+	add	r21, r21, r5
+	add	r22, r22, r6
+	add	r23, r23, r7
+	lwz	r0, 32*4(r1)
+	lwz	r5, 33*4(r1)
+	lwz	r6, 34*4(r1)
+	lwz	r7, 35*4(r1)
+	add	r24, r24, r0
+	add	r25, r25, r5
+	add	r26, r26, r6
+	add	r27, r27, r7
+	lwz	r0, 36*4(r1)
+	lwz	r5, 37*4(r1)
+	lwz	r6, 38*4(r1)
+	lwz	r7, 39*4(r1)
+	add	r28, r28, r0
+	add	r29, r29, r5
+	add	r30, r30, r6
+	add	r31, r31, r7
+	
+	stw	r16, 24*4(r1)
+	stw	r17, 25*4(r1)
+	stw	r18, 26*4(r1)
+	stw	r19, 27*4(r1)
+	stw	r20, 28*4(r1)
+	stw	r21, 29*4(r1)
+	stw	r22, 30*4(r1)
+	stw	r23, 31*4(r1)
+	stw	r24, 32*4(r1)
+	stw	r25, 33*4(r1)
+	stw	r26, 34*4(r1)
+	stw	r27, 35*4(r1)
+	stw	r28, 36*4(r1)
+	stw	r29, 37*4(r1)
+	stw	r30, 38*4(r1)
+	stw	r31, 39*4(r1)
+	
+	lwz	r0, 16*4(r3)
+	lwz	r5, 17*4(r3)
+	lwz	r6, 18*4(r3)
+	lwz	r7, 19*4(r3)
+	xor	r16, r16, r0
+	xor	r17, r17, r5
+	xor	r18, r18, r6
+	xor	r19, r19, r7
+	lwz	r0, 20*4(r3)
+	lwz	r5, 21*4(r3)
+	lwz	r6, 22*4(r3)
+	lwz	r7, 23*4(r3)
+	xor	r20, r20, r0
+	xor	r21, r21, r5
+	xor	r22, r22, r6
+	xor	r23, r23, r7
+	lwz	r0, 24*4(r3)
+	lwz	r5, 25*4(r3)
+	lwz	r6, 26*4(r3)
+	lwz	r7, 27*4(r3)
+	xor	r24, r24, r0
+	xor	r25, r25, r5
+	xor	r26, r26, r6
+	xor	r27, r27, r7
+	lwz	r0, 28*4(r3)
+	lwz	r5, 29*4(r3)
+	lwz	r6, 30*4(r3)
+	lwz	r7, 31*4(r3)
+	xor	r28, r28, r0
+	xor	r29, r29, r5
+	xor	r30, r30, r6
+	xor	r31, r31, r7
+	
+	lwz	r0, 40*4(r1)
+	lwz	r5, 41*4(r1)
+	lwz	r6, 42*4(r1)
+	lwz	r7, 43*4(r1)
+	xor	r16, r16, r0
+	xor	r17, r17, r5
+	xor	r18, r18, r6
+	xor	r19, r19, r7
+	lwz	r0, 44*4(r1)
+	lwz	r5, 45*4(r1)
+	lwz	r6, 46*4(r1)
+	lwz	r7, 47*4(r1)
+	xor	r20, r20, r0
+	xor	r21, r21, r5
+	xor	r22, r22, r6
+	xor	r23, r23, r7
+	xor	r24, r24, r8
+	xor	r25, r25, r9
+	xor	r26, r26, r10
+	xor	r27, r27, r11
+	xor	r28, r28, r12
+	xor	r29, r29, r13
+	xor	r30, r30, r14
+	xor	r31, r31, r15
+	stw	r16, 40*4(r1)
+	stw	r17, 41*4(r1)
+	stw	r18, 42*4(r1)
+	stw	r19, 43*4(r1)
+	mr	r8, r24
+	mr	r9, r25
+	mr	r10, r26
+	mr	r11, r27
+	stw	r20, 44*4(r1)
+	stw	r21, 45*4(r1)
+	stw	r22, 46*4(r1)
+	stw	r23, 47*4(r1)
+	mr	r12, r28
+	mr	r13, r29
+	mr	r14, r30
+	mr	r15, r31
+	
+	salsa8_core
+	
+	lwz	r0, 40*4(r1)
+	lwz	r5, 41*4(r1)
+	lwz	r6, 42*4(r1)
+	lwz	r7, 43*4(r1)
+	add	r16, r16, r0
+	add	r17, r17, r5
+	add	r18, r18, r6
+	add	r19, r19, r7
+	lwz	r0, 44*4(r1)
+	lwz	r5, 45*4(r1)
+	lwz	r6, 46*4(r1)
+	lwz	r7, 47*4(r1)
+	add	r20, r20, r0
+	add	r21, r21, r5
+	add	r22, r22, r6
+	add	r23, r23, r7
+	lwz	r5, 2*4(r1)
+	add	r8, r8, r24
+	add	r9, r9, r25
+	add	r10, r10, r26
+	add	r11, r11, r27
+	add	r12, r12, r28
+	add	r13, r13, r29
+	add	r14, r14, r30
+	add	r15, r15, r31
+	stw	r16, 40*4(r1)
+	stw	r17, 41*4(r1)
+	stw	r18, 42*4(r1)
+	stw	r19, 43*4(r1)
+	stw	r20, 44*4(r1)
+	stw	r21, 45*4(r1)
+	stw	r22, 46*4(r1)
+	stw	r23, 47*4(r1)
+	bdnz	scrypt_core_loop2
+	
+	ld	r3, 22*4(r1)
+	
+	lwz	r16, 24*4(r1)
+	lwz	r17, 25*4(r1)
+	lwz	r18, 26*4(r1)
+	lwz	r19, 27*4(r1)
+	lwz	r20, 28*4(r1)
+	lwz	r21, 29*4(r1)
+	lwz	r22, 30*4(r1)
+	lwz	r23, 31*4(r1)
+	stw	r16, 0*4(r3)
+	stw	r17, 1*4(r3)
+	stw	r18, 2*4(r3)
+	stw	r19, 3*4(r3)
+	stw	r20, 4*4(r3)
+	stw	r21, 5*4(r3)
+	stw	r22, 6*4(r3)
+	stw	r23, 7*4(r3)
+	lwz	r24, 32*4(r1)
+	lwz	r25, 33*4(r1)
+	lwz	r26, 34*4(r1)
+	lwz	r27, 35*4(r1)
+	lwz	r28, 36*4(r1)
+	lwz	r29, 37*4(r1)
+	lwz	r30, 38*4(r1)
+	lwz	r31, 39*4(r1)
+	stw	r24, 8*4(r3)
+	stw	r25, 9*4(r3)
+	stw	r26, 10*4(r3)
+	stw	r27, 11*4(r3)
+	stw	r28, 12*4(r3)
+	stw	r29, 13*4(r3)
+	stw	r30, 14*4(r3)
+	stw	r31, 15*4(r3)
+	lwz	r16, 40*4(r1)
+	lwz	r17, 41*4(r1)
+	lwz	r18, 42*4(r1)
+	lwz	r19, 43*4(r1)
+	lwz	r20, 44*4(r1)
+	lwz	r21, 45*4(r1)
+	lwz	r22, 46*4(r1)
+	lwz	r23, 47*4(r1)
+	stw	r16, 16*4(r3)
+	stw	r17, 17*4(r3)
+	stw	r18, 18*4(r3)
+	stw	r19, 19*4(r3)
+	stw	r20, 20*4(r3)
+	stw	r21, 21*4(r3)
+	stw	r22, 22*4(r3)
+	stw	r23, 23*4(r3)
+	stw	r8, 24*4(r3)
+	stw	r9, 25*4(r3)
+	stw	r10, 26*4(r3)
+	stw	r11, 27*4(r3)
+	stw	r12, 28*4(r3)
+	stw	r13, 29*4(r3)
+	stw	r14, 30*4(r3)
+	stw	r15, 31*4(r3)
+	
+	ld	r13, 4*4(r1)
+	ld	r14, 6*4(r1)
+	ld	r15, 8*4(r1)
+	ld	r16, 10*4(r1)
+	ld	r17, 12*4(r1)
+	ld	r18, 14*4(r1)
+	ld	r19, 16*4(r1)
+	ld	r20, 18*4(r1)
+	ld	r21, 20*4(r1)
+	ld	r22, 48*4(r1)
+	ld	r23, 50*4(r1)
+	ld	r24, 52*4(r1)
+	ld	r25, 54*4(r1)
+	ld	r26, 56*4(r1)
+	ld	r27, 58*4(r1)
+	ld	r28, 60*4(r1)
+	ld	r29, 62*4(r1)
+	ld	r30, 64*4(r1)
+	ld	r31, 66*4(r1)
+	addi	r1, r1, 68*4
+	blr
+
+#endif /* __ALTIVEC__ */
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-x64.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-x64.S
new file mode 100644
index 0000000..f9185d4
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-x64.S
@@ -0,0 +1,2907 @@
+/*
+ * Copyright 2011-2014 pooler@litecoinpool.org
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(__linux__) && defined(__ELF__)
+	.section .note.GNU-stack,"",%progbits
+#endif
+
+#if defined(USE_ASM) && defined(__x86_64__)
+
+	.text
+	.p2align 6
+	.globl scrypt_best_throughput
+	.globl _scrypt_best_throughput
+scrypt_best_throughput:
+_scrypt_best_throughput:
+	pushq	%rbx
+#if defined(USE_AVX2)
+	/* Check for AVX and OSXSAVE support */
+	movl	$1, %eax
+	cpuid
+	andl	$0x18000000, %ecx
+	cmpl	$0x18000000, %ecx
+	jne scrypt_best_throughput_no_avx2
+	/* Check for AVX2 support */
+	movl	$7, %eax
+	xorl	%ecx, %ecx
+	cpuid
+	andl	$0x00000020, %ebx
+	cmpl	$0x00000020, %ebx
+	jne scrypt_best_throughput_no_avx2
+	/* Check for XMM and YMM state support */
+	xorl	%ecx, %ecx
+	xgetbv
+	andl	$0x00000006, %eax
+	cmpl	$0x00000006, %eax
+	jne scrypt_best_throughput_no_avx2
+	movl	$6, %eax
+	jmp scrypt_best_throughput_exit
+scrypt_best_throughput_no_avx2:
+#endif
+	/* Check for AuthenticAMD */
+	xorq	%rax, %rax
+	cpuid
+	movl	$3, %eax
+	cmpl	$0x444d4163, %ecx
+	jne scrypt_best_throughput_not_amd
+	cmpl	$0x69746e65, %edx
+	jne scrypt_best_throughput_not_amd
+	cmpl	$0x68747541, %ebx
+	jne scrypt_best_throughput_not_amd
+	/* Check for AMD K8 or Bobcat */
+	movl	$1, %eax
+	cpuid
+	andl	$0x0ff00000, %eax
+	jz scrypt_best_throughput_one
+	cmpl	$0x00500000, %eax
+	je scrypt_best_throughput_one
+	movl	$3, %eax
+	jmp scrypt_best_throughput_exit
+scrypt_best_throughput_not_amd:
+	/* Check for GenuineIntel */
+	cmpl	$0x6c65746e, %ecx
+	jne scrypt_best_throughput_exit
+	cmpl	$0x49656e69, %edx
+	jne scrypt_best_throughput_exit
+	cmpl	$0x756e6547, %ebx
+	jne scrypt_best_throughput_exit
+	/* Check for Intel Atom */
+	movl	$1, %eax
+	cpuid
+	movl	%eax, %edx
+	andl	$0x0ff00f00, %eax
+	cmpl	$0x00000600, %eax
+	movl	$3, %eax
+	jnz scrypt_best_throughput_exit
+	andl	$0x000f00f0, %edx
+	cmpl	$0x000100c0, %edx
+	je scrypt_best_throughput_one
+	cmpl	$0x00020060, %edx
+	je scrypt_best_throughput_one
+	cmpl	$0x00030060, %edx
+	jne scrypt_best_throughput_exit
+scrypt_best_throughput_one:
+	movl	$1, %eax
+scrypt_best_throughput_exit:
+	popq	%rbx
+	ret
+	
+	
+.macro scrypt_shuffle src, so, dest, do
+	movl	\so+60(\src), %eax
+	movl	\so+44(\src), %ebx
+	movl	\so+28(\src), %ecx
+	movl	\so+12(\src), %edx
+	movl	%eax, \do+12(\dest)
+	movl	%ebx, \do+28(\dest)
+	movl	%ecx, \do+44(\dest)
+	movl	%edx, \do+60(\dest)
+	movl	\so+40(\src), %eax
+	movl	\so+8(\src), %ebx
+	movl	\so+48(\src), %ecx
+	movl	\so+16(\src), %edx
+	movl	%eax, \do+8(\dest)
+	movl	%ebx, \do+40(\dest)
+	movl	%ecx, \do+16(\dest)
+	movl	%edx, \do+48(\dest)
+	movl	\so+20(\src), %eax
+	movl	\so+4(\src), %ebx
+	movl	\so+52(\src), %ecx
+	movl	\so+36(\src), %edx
+	movl	%eax, \do+4(\dest)
+	movl	%ebx, \do+20(\dest)
+	movl	%ecx, \do+36(\dest)
+	movl	%edx, \do+52(\dest)
+	movl	\so+0(\src), %eax
+	movl	\so+24(\src), %ebx
+	movl	\so+32(\src), %ecx
+	movl	\so+56(\src), %edx
+	movl	%eax, \do+0(\dest)
+	movl	%ebx, \do+24(\dest)
+	movl	%ecx, \do+32(\dest)
+	movl	%edx, \do+56(\dest)
+.endm
+
+
+.macro salsa8_core_gen_doubleround
+	movq	72(%rsp), %r15
+	
+	leaq	(%r14, %rdx), %rbp
+	roll	$7, %ebp
+	xorl	%ebp, %r9d
+	leaq	(%rdi, %r15), %rbp
+	roll	$7, %ebp
+	xorl	%ebp, %r10d
+	leaq	(%rdx, %r9), %rbp
+	roll	$9, %ebp
+	xorl	%ebp, %r11d
+	leaq	(%r15, %r10), %rbp
+	roll	$9, %ebp
+	xorl	%ebp, %r13d
+	
+	leaq	(%r9, %r11), %rbp
+	roll	$13, %ebp
+	xorl	%ebp, %r14d
+	leaq	(%r10, %r13), %rbp
+	roll	$13, %ebp
+	xorl	%ebp, %edi
+	leaq	(%r11, %r14), %rbp
+	roll	$18, %ebp
+	xorl	%ebp, %edx
+	leaq	(%r13, %rdi), %rbp
+	roll	$18, %ebp
+	xorl	%ebp, %r15d
+	
+	movq	48(%rsp), %rbp
+	movq	%r15, 72(%rsp)
+	
+	leaq	(%rax, %rbp), %r15
+	roll	$7, %r15d
+	xorl	%r15d, %ebx
+	leaq	(%rbp, %rbx), %r15
+	roll	$9, %r15d
+	xorl	%r15d, %ecx
+	leaq	(%rbx, %rcx), %r15
+	roll	$13, %r15d
+	xorl	%r15d, %eax
+	leaq	(%rcx, %rax), %r15
+	roll	$18, %r15d
+	xorl	%r15d, %ebp
+	
+	movq	88(%rsp), %r15
+	movq	%rbp, 48(%rsp)
+	
+	leaq	(%r12, %r15), %rbp
+	roll	$7, %ebp
+	xorl	%ebp, %esi
+	leaq	(%r15, %rsi), %rbp
+	roll	$9, %ebp
+	xorl	%ebp, %r8d
+	leaq	(%rsi, %r8), %rbp
+	roll	$13, %ebp
+	xorl	%ebp, %r12d
+	leaq	(%r8, %r12), %rbp
+	roll	$18, %ebp
+	xorl	%ebp, %r15d
+	
+	movq	%r15, 88(%rsp)
+	movq	72(%rsp), %r15
+	
+	leaq	(%rsi, %rdx), %rbp
+	roll	$7, %ebp
+	xorl	%ebp, %edi
+	leaq	(%r9, %r15), %rbp
+	roll	$7, %ebp
+	xorl	%ebp, %eax
+	leaq	(%rdx, %rdi), %rbp
+	roll	$9, %ebp
+	xorl	%ebp, %ecx
+	leaq	(%r15, %rax), %rbp
+	roll	$9, %ebp
+	xorl	%ebp, %r8d
+	
+	leaq	(%rdi, %rcx), %rbp
+	roll	$13, %ebp
+	xorl	%ebp, %esi
+	leaq	(%rax, %r8), %rbp
+	roll	$13, %ebp
+	xorl	%ebp, %r9d
+	leaq	(%rcx, %rsi), %rbp
+	roll	$18, %ebp
+	xorl	%ebp, %edx
+	leaq	(%r8, %r9), %rbp
+	roll	$18, %ebp
+	xorl	%ebp, %r15d
+	
+	movq	48(%rsp), %rbp
+	movq	%r15, 72(%rsp)
+	
+	leaq	(%r10, %rbp), %r15
+	roll	$7, %r15d
+	xorl	%r15d, %r12d
+	leaq	(%rbp, %r12), %r15
+	roll	$9, %r15d
+	xorl	%r15d, %r11d
+	leaq	(%r12, %r11), %r15
+	roll	$13, %r15d
+	xorl	%r15d, %r10d
+	leaq	(%r11, %r10), %r15
+	roll	$18, %r15d
+	xorl	%r15d, %ebp
+	
+	movq	88(%rsp), %r15
+	movq	%rbp, 48(%rsp)
+	
+	leaq	(%rbx, %r15), %rbp
+	roll	$7, %ebp
+	xorl	%ebp, %r14d
+	leaq	(%r15, %r14), %rbp
+	roll	$9, %ebp
+	xorl	%ebp, %r13d
+	leaq	(%r14, %r13), %rbp
+	roll	$13, %ebp
+	xorl	%ebp, %ebx
+	leaq	(%r13, %rbx), %rbp
+	roll	$18, %ebp
+	xorl	%ebp, %r15d
+	
+	movq	%r15, 88(%rsp)
+.endm
+
+	.text
+	.p2align 6
+salsa8_core_gen:
+	/* 0: %rdx, %rdi, %rcx, %rsi */
+	movq	8(%rsp), %rdi
+	movq	%rdi, %rdx
+	shrq	$32, %rdi
+	movq	16(%rsp), %rsi
+	movq	%rsi, %rcx
+	shrq	$32, %rsi
+	/* 1: %r9, 72(%rsp), %rax, %r8 */
+	movq	24(%rsp), %r8
+	movq	%r8, %r9
+	shrq	$32, %r8
+	movq	%r8, 72(%rsp)
+	movq	32(%rsp), %r8
+	movq	%r8, %rax
+	shrq	$32, %r8
+	/* 2: %r11, %r10, 48(%rsp), %r12 */
+	movq	40(%rsp), %r10
+	movq	%r10, %r11
+	shrq	$32, %r10
+	movq	48(%rsp), %r12
+	/* movq	%r12, %r13 */
+	/* movq	%r13, 48(%rsp) */
+	shrq	$32, %r12
+	/* 3: %r14, %r13, %rbx, 88(%rsp) */
+	movq	56(%rsp), %r13
+	movq	%r13, %r14
+	shrq	$32, %r13
+	movq	64(%rsp), %r15
+	movq	%r15, %rbx
+	shrq	$32, %r15
+	movq	%r15, 88(%rsp)
+	
+	salsa8_core_gen_doubleround
+	salsa8_core_gen_doubleround
+	salsa8_core_gen_doubleround
+	salsa8_core_gen_doubleround
+	
+	shlq	$32, %rdi
+	xorq	%rdi, %rdx
+	movq	%rdx, 24(%rsp)
+	
+	shlq	$32, %rsi
+	xorq	%rsi, %rcx
+	movq	%rcx, 32(%rsp)
+	
+	movl	72(%rsp), %edi
+	shlq	$32, %rdi
+	xorq	%rdi, %r9
+	movq	%r9, 40(%rsp)
+	
+	movl	48(%rsp), %ebp
+	shlq	$32, %r8
+	xorq	%r8, %rax
+	movq	%rax, 48(%rsp)
+	
+	shlq	$32, %r10
+	xorq	%r10, %r11
+	movq	%r11, 56(%rsp)
+	
+	shlq	$32, %r12
+	xorq	%r12, %rbp
+	movq	%rbp, 64(%rsp)
+	
+	shlq	$32, %r13
+	xorq	%r13, %r14
+	movq	%r14, 72(%rsp)
+	
+	movdqa	24(%rsp), %xmm0
+	
+	shlq	$32, %r15
+	xorq	%r15, %rbx
+	movq	%rbx, 80(%rsp)
+	
+	movdqa	40(%rsp), %xmm1
+	movdqa	56(%rsp), %xmm2
+	movdqa	72(%rsp), %xmm3
+	
+	ret
+	
+	
+	.text
+	.p2align 6
+	.globl scrypt_core
+	.globl _scrypt_core
+scrypt_core:
+_scrypt_core:
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+#if defined(_WIN64) || defined(__CYGWIN__)
+	subq	$176, %rsp
+	movdqa	%xmm6, 8(%rsp)
+	movdqa	%xmm7, 24(%rsp)
+	movdqa	%xmm8, 40(%rsp)
+	movdqa	%xmm9, 56(%rsp)
+	movdqa	%xmm10, 72(%rsp)
+	movdqa	%xmm11, 88(%rsp)
+	movdqa	%xmm12, 104(%rsp)
+	movdqa	%xmm13, 120(%rsp)
+	movdqa	%xmm14, 136(%rsp)
+	movdqa	%xmm15, 152(%rsp)
+	pushq	%rdi
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+#else
+	movq	%rdx, %r8
+#endif
+
+.macro scrypt_core_cleanup
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	popq	%rdi
+	movdqa	8(%rsp), %xmm6
+	movdqa	24(%rsp), %xmm7
+	movdqa	40(%rsp), %xmm8
+	movdqa	56(%rsp), %xmm9
+	movdqa	72(%rsp), %xmm10
+	movdqa	88(%rsp), %xmm11
+	movdqa	104(%rsp), %xmm12
+	movdqa	120(%rsp), %xmm13
+	movdqa	136(%rsp), %xmm14
+	movdqa	152(%rsp), %xmm15
+	addq	$176, %rsp
+#endif
+	popq	%r15
+	popq	%r14
+	popq	%r13
+	popq	%r12
+	popq	%rbp
+	popq	%rbx
+.endm
+	
+	/* GenuineIntel processors have fast SIMD */
+	xorl	%eax, %eax
+	cpuid
+	cmpl	$0x6c65746e, %ecx
+	jne scrypt_core_gen
+	cmpl	$0x49656e69, %edx
+	jne scrypt_core_gen
+	cmpl	$0x756e6547, %ebx
+	je scrypt_core_xmm
+	
+	.p2align 6
+scrypt_core_gen:
+	subq	$136, %rsp
+	movdqa	0(%rdi), %xmm8
+	movdqa	16(%rdi), %xmm9
+	movdqa	32(%rdi), %xmm10
+	movdqa	48(%rdi), %xmm11
+	movdqa	64(%rdi), %xmm12
+	movdqa	80(%rdi), %xmm13
+	movdqa	96(%rdi), %xmm14
+	movdqa	112(%rdi), %xmm15
+	
+	movq	%r8, %rcx
+	shlq	$7, %rcx
+	addq	%rsi, %rcx
+	movq	%r8, 96(%rsp)
+	movq	%rdi, 104(%rsp)
+	movq	%rsi, 112(%rsp)
+	movq	%rcx, 120(%rsp)
+scrypt_core_gen_loop1:
+	movdqa	%xmm8, 0(%rsi)
+	movdqa	%xmm9, 16(%rsi)
+	movdqa	%xmm10, 32(%rsi)
+	movdqa	%xmm11, 48(%rsi)
+	movdqa	%xmm12, 64(%rsi)
+	movdqa	%xmm13, 80(%rsi)
+	movdqa	%xmm14, 96(%rsi)
+	movdqa	%xmm15, 112(%rsi)
+	
+	pxor	%xmm12, %xmm8
+	pxor	%xmm13, %xmm9
+	pxor	%xmm14, %xmm10
+	pxor	%xmm15, %xmm11
+	movdqa	%xmm8, 0(%rsp)
+	movdqa	%xmm9, 16(%rsp)
+	movdqa	%xmm10, 32(%rsp)
+	movdqa	%xmm11, 48(%rsp)
+	movq	%rsi, 128(%rsp)
+	call salsa8_core_gen
+	paddd	%xmm0, %xmm8
+	paddd	%xmm1, %xmm9
+	paddd	%xmm2, %xmm10
+	paddd	%xmm3, %xmm11
+	
+	pxor	%xmm8, %xmm12
+	pxor	%xmm9, %xmm13
+	pxor	%xmm10, %xmm14
+	pxor	%xmm11, %xmm15
+	movdqa	%xmm12, 0(%rsp)
+	movdqa	%xmm13, 16(%rsp)
+	movdqa	%xmm14, 32(%rsp)
+	movdqa	%xmm15, 48(%rsp)
+	call salsa8_core_gen
+	movq	128(%rsp), %rsi
+	paddd	%xmm0, %xmm12
+	paddd	%xmm1, %xmm13
+	paddd	%xmm2, %xmm14
+	paddd	%xmm3, %xmm15
+	
+	addq	$128, %rsi
+	movq	120(%rsp), %rcx
+	cmpq	%rcx, %rsi
+	jne scrypt_core_gen_loop1
+	
+	movq	96(%rsp), %r8
+	movq	%r8, %rcx
+	subl	$1, %r8d
+	movq	%r8, 96(%rsp)
+	movd	%xmm12, %edx
+scrypt_core_gen_loop2:
+	movq	112(%rsp), %rsi
+	andl	%r8d, %edx
+	shll	$7, %edx
+	addq	%rsi, %rdx
+	movdqa	0(%rdx), %xmm0
+	movdqa	16(%rdx), %xmm1
+	movdqa	32(%rdx), %xmm2
+	movdqa	48(%rdx), %xmm3
+	movdqa	64(%rdx), %xmm4
+	movdqa	80(%rdx), %xmm5
+	movdqa	96(%rdx), %xmm6
+	movdqa	112(%rdx), %xmm7
+	pxor	%xmm0, %xmm8
+	pxor	%xmm1, %xmm9
+	pxor	%xmm2, %xmm10
+	pxor	%xmm3, %xmm11
+	pxor	%xmm4, %xmm12
+	pxor	%xmm5, %xmm13
+	pxor	%xmm6, %xmm14
+	pxor	%xmm7, %xmm15
+	
+	pxor	%xmm12, %xmm8
+	pxor	%xmm13, %xmm9
+	pxor	%xmm14, %xmm10
+	pxor	%xmm15, %xmm11
+	movdqa	%xmm8, 0(%rsp)
+	movdqa	%xmm9, 16(%rsp)
+	movdqa	%xmm10, 32(%rsp)
+	movdqa	%xmm11, 48(%rsp)
+	movq	%rcx, 128(%rsp)
+	call salsa8_core_gen
+	paddd	%xmm0, %xmm8
+	paddd	%xmm1, %xmm9
+	paddd	%xmm2, %xmm10
+	paddd	%xmm3, %xmm11
+	
+	pxor	%xmm8, %xmm12
+	pxor	%xmm9, %xmm13
+	pxor	%xmm10, %xmm14
+	pxor	%xmm11, %xmm15
+	movdqa	%xmm12, 0(%rsp)
+	movdqa	%xmm13, 16(%rsp)
+	movdqa	%xmm14, 32(%rsp)
+	movdqa	%xmm15, 48(%rsp)
+	call salsa8_core_gen
+	movq	96(%rsp), %r8
+	movq	128(%rsp), %rcx
+	addl	0(%rsp), %edx
+	paddd	%xmm0, %xmm12
+	paddd	%xmm1, %xmm13
+	paddd	%xmm2, %xmm14
+	paddd	%xmm3, %xmm15
+	
+	subq	$1, %rcx
+	ja scrypt_core_gen_loop2
+	
+	movq	104(%rsp), %rdi
+	movdqa	%xmm8, 0(%rdi)
+	movdqa	%xmm9, 16(%rdi)
+	movdqa	%xmm10, 32(%rdi)
+	movdqa	%xmm11, 48(%rdi)
+	movdqa	%xmm12, 64(%rdi)
+	movdqa	%xmm13, 80(%rdi)
+	movdqa	%xmm14, 96(%rdi)
+	movdqa	%xmm15, 112(%rdi)
+	
+	addq	$136, %rsp
+	scrypt_core_cleanup
+	ret
+
+
+.macro salsa8_core_xmm_doubleround
+	movdqa	%xmm1, %xmm4
+	paddd	%xmm0, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$7, %xmm4
+	psrld	$25, %xmm5
+	pxor	%xmm4, %xmm3
+	movdqa	%xmm0, %xmm4
+	pxor	%xmm5, %xmm3
+	
+	paddd	%xmm3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$9, %xmm4
+	psrld	$23, %xmm5
+	pxor	%xmm4, %xmm2
+	movdqa	%xmm3, %xmm4
+	pxor	%xmm5, %xmm2
+	pshufd	$0x93, %xmm3, %xmm3
+	
+	paddd	%xmm2, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$13, %xmm4
+	psrld	$19, %xmm5
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm2, %xmm4
+	pxor	%xmm5, %xmm1
+	pshufd	$0x4e, %xmm2, %xmm2
+	
+	paddd	%xmm1, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$18, %xmm4
+	psrld	$14, %xmm5
+	pxor	%xmm4, %xmm0
+	movdqa	%xmm3, %xmm4
+	pxor	%xmm5, %xmm0
+	pshufd	$0x39, %xmm1, %xmm1
+	
+	paddd	%xmm0, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$7, %xmm4
+	psrld	$25, %xmm5
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm0, %xmm4
+	pxor	%xmm5, %xmm1
+	
+	paddd	%xmm1, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$9, %xmm4
+	psrld	$23, %xmm5
+	pxor	%xmm4, %xmm2
+	movdqa	%xmm1, %xmm4
+	pxor	%xmm5, %xmm2
+	pshufd	$0x93, %xmm1, %xmm1
+	
+	paddd	%xmm2, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$13, %xmm4
+	psrld	$19, %xmm5
+	pxor	%xmm4, %xmm3
+	movdqa	%xmm2, %xmm4
+	pxor	%xmm5, %xmm3
+	pshufd	$0x4e, %xmm2, %xmm2
+	
+	paddd	%xmm3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$18, %xmm4
+	psrld	$14, %xmm5
+	pxor	%xmm4, %xmm0
+	pshufd	$0x39, %xmm3, %xmm3
+	pxor	%xmm5, %xmm0
+.endm
+
+.macro salsa8_core_xmm
+	salsa8_core_xmm_doubleround
+	salsa8_core_xmm_doubleround
+	salsa8_core_xmm_doubleround
+	salsa8_core_xmm_doubleround
+.endm
+	
+	.p2align 6
+scrypt_core_xmm:
+	pcmpeqw	%xmm1, %xmm1
+	psrlq	$32, %xmm1
+	
+	movdqa	0(%rdi), %xmm8
+	movdqa	16(%rdi), %xmm11
+	movdqa	32(%rdi), %xmm10
+	movdqa	48(%rdi), %xmm9
+	movdqa	%xmm8, %xmm0
+	pxor	%xmm11, %xmm8
+	pand	%xmm1, %xmm8
+	pxor	%xmm11, %xmm8
+	pxor	%xmm10, %xmm11
+	pand	%xmm1, %xmm11
+	pxor	%xmm10, %xmm11
+	pxor	%xmm9, %xmm10
+	pand	%xmm1, %xmm10
+	pxor	%xmm9, %xmm10
+	pxor	%xmm0, %xmm9
+	pand	%xmm1, %xmm9
+	pxor	%xmm0, %xmm9
+	movdqa	%xmm8, %xmm0
+	pshufd	$0x4e, %xmm10, %xmm10
+	punpcklqdq	%xmm10, %xmm8
+	punpckhqdq	%xmm0, %xmm10
+	movdqa	%xmm11, %xmm0
+	pshufd	$0x4e, %xmm9, %xmm9
+	punpcklqdq	%xmm9, %xmm11
+	punpckhqdq	%xmm0, %xmm9
+	
+	movdqa	64(%rdi), %xmm12
+	movdqa	80(%rdi), %xmm15
+	movdqa	96(%rdi), %xmm14
+	movdqa	112(%rdi), %xmm13
+	movdqa	%xmm12, %xmm0
+	pxor	%xmm15, %xmm12
+	pand	%xmm1, %xmm12
+	pxor	%xmm15, %xmm12
+	pxor	%xmm14, %xmm15
+	pand	%xmm1, %xmm15
+	pxor	%xmm14, %xmm15
+	pxor	%xmm13, %xmm14
+	pand	%xmm1, %xmm14
+	pxor	%xmm13, %xmm14
+	pxor	%xmm0, %xmm13
+	pand	%xmm1, %xmm13
+	pxor	%xmm0, %xmm13
+	movdqa	%xmm12, %xmm0
+	pshufd	$0x4e, %xmm14, %xmm14
+	punpcklqdq	%xmm14, %xmm12
+	punpckhqdq	%xmm0, %xmm14
+	movdqa	%xmm15, %xmm0
+	pshufd	$0x4e, %xmm13, %xmm13
+	punpcklqdq	%xmm13, %xmm15
+	punpckhqdq	%xmm0, %xmm13
+	
+	movq	%rsi, %rdx
+	movq	%r8, %rcx
+	shlq	$7, %rcx
+	addq	%rsi, %rcx
+scrypt_core_xmm_loop1:
+	pxor	%xmm12, %xmm8
+	pxor	%xmm13, %xmm9
+	pxor	%xmm14, %xmm10
+	pxor	%xmm15, %xmm11
+	movdqa	%xmm8, 0(%rdx)
+	movdqa	%xmm9, 16(%rdx)
+	movdqa	%xmm10, 32(%rdx)
+	movdqa	%xmm11, 48(%rdx)
+	movdqa	%xmm12, 64(%rdx)
+	movdqa	%xmm13, 80(%rdx)
+	movdqa	%xmm14, 96(%rdx)
+	movdqa	%xmm15, 112(%rdx)
+	
+	movdqa	%xmm8, %xmm0
+	movdqa	%xmm9, %xmm1
+	movdqa	%xmm10, %xmm2
+	movdqa	%xmm11, %xmm3
+	salsa8_core_xmm
+	paddd	%xmm0, %xmm8
+	paddd	%xmm1, %xmm9
+	paddd	%xmm2, %xmm10
+	paddd	%xmm3, %xmm11
+	
+	pxor	%xmm8, %xmm12
+	pxor	%xmm9, %xmm13
+	pxor	%xmm10, %xmm14
+	pxor	%xmm11, %xmm15
+	movdqa	%xmm12, %xmm0
+	movdqa	%xmm13, %xmm1
+	movdqa	%xmm14, %xmm2
+	movdqa	%xmm15, %xmm3
+	salsa8_core_xmm
+	paddd	%xmm0, %xmm12
+	paddd	%xmm1, %xmm13
+	paddd	%xmm2, %xmm14
+	paddd	%xmm3, %xmm15
+	
+	addq	$128, %rdx
+	cmpq	%rcx, %rdx
+	jne scrypt_core_xmm_loop1
+	
+	movq	%r8, %rcx
+	subl	$1, %r8d
+scrypt_core_xmm_loop2:
+	movd	%xmm12, %edx
+	andl	%r8d, %edx
+	shll	$7, %edx
+	pxor	0(%rsi, %rdx), %xmm8
+	pxor	16(%rsi, %rdx), %xmm9
+	pxor	32(%rsi, %rdx), %xmm10
+	pxor	48(%rsi, %rdx), %xmm11
+	
+	pxor	%xmm12, %xmm8
+	pxor	%xmm13, %xmm9
+	pxor	%xmm14, %xmm10
+	pxor	%xmm15, %xmm11
+	movdqa	%xmm8, %xmm0
+	movdqa	%xmm9, %xmm1
+	movdqa	%xmm10, %xmm2
+	movdqa	%xmm11, %xmm3
+	salsa8_core_xmm
+	paddd	%xmm0, %xmm8
+	paddd	%xmm1, %xmm9
+	paddd	%xmm2, %xmm10
+	paddd	%xmm3, %xmm11
+	
+	pxor	64(%rsi, %rdx), %xmm12
+	pxor	80(%rsi, %rdx), %xmm13
+	pxor	96(%rsi, %rdx), %xmm14
+	pxor	112(%rsi, %rdx), %xmm15
+	pxor	%xmm8, %xmm12
+	pxor	%xmm9, %xmm13
+	pxor	%xmm10, %xmm14
+	pxor	%xmm11, %xmm15
+	movdqa	%xmm12, %xmm0
+	movdqa	%xmm13, %xmm1
+	movdqa	%xmm14, %xmm2
+	movdqa	%xmm15, %xmm3
+	salsa8_core_xmm
+	paddd	%xmm0, %xmm12
+	paddd	%xmm1, %xmm13
+	paddd	%xmm2, %xmm14
+	paddd	%xmm3, %xmm15
+	
+	subq	$1, %rcx
+	ja scrypt_core_xmm_loop2
+	
+	pcmpeqw	%xmm1, %xmm1
+	psrlq	$32, %xmm1
+	
+	movdqa	%xmm8, %xmm0
+	pxor	%xmm9, %xmm8
+	pand	%xmm1, %xmm8
+	pxor	%xmm9, %xmm8
+	pxor	%xmm10, %xmm9
+	pand	%xmm1, %xmm9
+	pxor	%xmm10, %xmm9
+	pxor	%xmm11, %xmm10
+	pand	%xmm1, %xmm10
+	pxor	%xmm11, %xmm10
+	pxor	%xmm0, %xmm11
+	pand	%xmm1, %xmm11
+	pxor	%xmm0, %xmm11
+	movdqa	%xmm8, %xmm0
+	pshufd	$0x4e, %xmm10, %xmm10
+	punpcklqdq	%xmm10, %xmm8
+	punpckhqdq	%xmm0, %xmm10
+	movdqa	%xmm9, %xmm0
+	pshufd	$0x4e, %xmm11, %xmm11
+	punpcklqdq	%xmm11, %xmm9
+	punpckhqdq	%xmm0, %xmm11
+	movdqa	%xmm8, 0(%rdi)
+	movdqa	%xmm11, 16(%rdi)
+	movdqa	%xmm10, 32(%rdi)
+	movdqa	%xmm9, 48(%rdi)
+	
+	movdqa	%xmm12, %xmm0
+	pxor	%xmm13, %xmm12
+	pand	%xmm1, %xmm12
+	pxor	%xmm13, %xmm12
+	pxor	%xmm14, %xmm13
+	pand	%xmm1, %xmm13
+	pxor	%xmm14, %xmm13
+	pxor	%xmm15, %xmm14
+	pand	%xmm1, %xmm14
+	pxor	%xmm15, %xmm14
+	pxor	%xmm0, %xmm15
+	pand	%xmm1, %xmm15
+	pxor	%xmm0, %xmm15
+	movdqa	%xmm12, %xmm0
+	pshufd	$0x4e, %xmm14, %xmm14
+	punpcklqdq	%xmm14, %xmm12
+	punpckhqdq	%xmm0, %xmm14
+	movdqa	%xmm13, %xmm0
+	pshufd	$0x4e, %xmm15, %xmm15
+	punpcklqdq	%xmm15, %xmm13
+	punpckhqdq	%xmm0, %xmm15
+	movdqa	%xmm12, 64(%rdi)
+	movdqa	%xmm15, 80(%rdi)
+	movdqa	%xmm14, 96(%rdi)
+	movdqa	%xmm13, 112(%rdi)
+	
+	scrypt_core_cleanup
+	ret
+	
+	
+#if defined(USE_AVX)
+.macro salsa8_core_3way_avx_doubleround
+	vpaddd	%xmm0, %xmm1, %xmm4
+	vpaddd	%xmm8, %xmm9, %xmm6
+	vpaddd	%xmm12, %xmm13, %xmm7
+	vpslld	$7, %xmm4, %xmm5
+	vpsrld	$25, %xmm4, %xmm4
+	vpxor	%xmm5, %xmm3, %xmm3
+	vpxor	%xmm4, %xmm3, %xmm3
+	vpslld	$7, %xmm6, %xmm5
+	vpsrld	$25, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm11, %xmm11
+	vpxor	%xmm6, %xmm11, %xmm11
+	vpslld	$7, %xmm7, %xmm5
+	vpsrld	$25, %xmm7, %xmm7
+	vpxor	%xmm5, %xmm15, %xmm15
+	vpxor	%xmm7, %xmm15, %xmm15
+	
+	vpaddd	%xmm3, %xmm0, %xmm4
+	vpaddd	%xmm11, %xmm8, %xmm6
+	vpaddd	%xmm15, %xmm12, %xmm7
+	vpslld	$9, %xmm4, %xmm5
+	vpsrld	$23, %xmm4, %xmm4
+	vpxor	%xmm5, %xmm2, %xmm2
+	vpxor	%xmm4, %xmm2, %xmm2
+	vpslld	$9, %xmm6, %xmm5
+	vpsrld	$23, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm10, %xmm10
+	vpxor	%xmm6, %xmm10, %xmm10
+	vpslld	$9, %xmm7, %xmm5
+	vpsrld	$23, %xmm7, %xmm7
+	vpxor	%xmm5, %xmm14, %xmm14
+	vpxor	%xmm7, %xmm14, %xmm14
+	
+	vpaddd	%xmm2, %xmm3, %xmm4
+	vpaddd	%xmm10, %xmm11, %xmm6
+	vpaddd	%xmm14, %xmm15, %xmm7
+	vpslld	$13, %xmm4, %xmm5
+	vpsrld	$19, %xmm4, %xmm4
+	vpshufd	$0x93, %xmm3, %xmm3
+	vpshufd	$0x93, %xmm11, %xmm11
+	vpshufd	$0x93, %xmm15, %xmm15
+	vpxor	%xmm5, %xmm1, %xmm1
+	vpxor	%xmm4, %xmm1, %xmm1
+	vpslld	$13, %xmm6, %xmm5
+	vpsrld	$19, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm9, %xmm9
+	vpxor	%xmm6, %xmm9, %xmm9
+	vpslld	$13, %xmm7, %xmm5
+	vpsrld	$19, %xmm7, %xmm7
+	vpxor	%xmm5, %xmm13, %xmm13
+	vpxor	%xmm7, %xmm13, %xmm13
+	
+	vpaddd	%xmm1, %xmm2, %xmm4
+	vpaddd	%xmm9, %xmm10, %xmm6
+	vpaddd	%xmm13, %xmm14, %xmm7
+	vpslld	$18, %xmm4, %xmm5
+	vpsrld	$14, %xmm4, %xmm4
+	vpshufd	$0x4e, %xmm2, %xmm2
+	vpshufd	$0x4e, %xmm10, %xmm10
+	vpshufd	$0x4e, %xmm14, %xmm14
+	vpxor	%xmm5, %xmm0, %xmm0
+	vpxor	%xmm4, %xmm0, %xmm0
+	vpslld	$18, %xmm6, %xmm5
+	vpsrld	$14, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm8, %xmm8
+	vpslld	$18, %xmm7, %xmm5
+	vpsrld	$14, %xmm7, %xmm7
+	vpxor	%xmm5, %xmm12, %xmm12
+	vpxor	%xmm7, %xmm12, %xmm12
+	
+	vpaddd	%xmm0, %xmm3, %xmm4
+	vpaddd	%xmm8, %xmm11, %xmm6
+	vpaddd	%xmm12, %xmm15, %xmm7
+	vpslld	$7, %xmm4, %xmm5
+	vpsrld	$25, %xmm4, %xmm4
+	vpshufd	$0x39, %xmm1, %xmm1
+	vpxor	%xmm5, %xmm1, %xmm1
+	vpxor	%xmm4, %xmm1, %xmm1
+	vpslld	$7, %xmm6, %xmm5
+	vpsrld	$25, %xmm6, %xmm6
+	vpshufd	$0x39, %xmm9, %xmm9
+	vpxor	%xmm5, %xmm9, %xmm9
+	vpxor	%xmm6, %xmm9, %xmm9
+	vpslld	$7, %xmm7, %xmm5
+	vpsrld	$25, %xmm7, %xmm7
+	vpshufd	$0x39, %xmm13, %xmm13
+	vpxor	%xmm5, %xmm13, %xmm13
+	vpxor	%xmm7, %xmm13, %xmm13
+	
+	vpaddd	%xmm1, %xmm0, %xmm4
+	vpaddd	%xmm9, %xmm8, %xmm6
+	vpaddd	%xmm13, %xmm12, %xmm7
+	vpslld	$9, %xmm4, %xmm5
+	vpsrld	$23, %xmm4, %xmm4
+	vpxor	%xmm5, %xmm2, %xmm2
+	vpxor	%xmm4, %xmm2, %xmm2
+	vpslld	$9, %xmm6, %xmm5
+	vpsrld	$23, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm10, %xmm10
+	vpxor	%xmm6, %xmm10, %xmm10
+	vpslld	$9, %xmm7, %xmm5
+	vpsrld	$23, %xmm7, %xmm7
+	vpxor	%xmm5, %xmm14, %xmm14
+	vpxor	%xmm7, %xmm14, %xmm14
+	
+	vpaddd	%xmm2, %xmm1, %xmm4
+	vpaddd	%xmm10, %xmm9, %xmm6
+	vpaddd	%xmm14, %xmm13, %xmm7
+	vpslld	$13, %xmm4, %xmm5
+	vpsrld	$19, %xmm4, %xmm4
+	vpshufd	$0x93, %xmm1, %xmm1
+	vpshufd	$0x93, %xmm9, %xmm9
+	vpshufd	$0x93, %xmm13, %xmm13
+	vpxor	%xmm5, %xmm3, %xmm3
+	vpxor	%xmm4, %xmm3, %xmm3
+	vpslld	$13, %xmm6, %xmm5
+	vpsrld	$19, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm11, %xmm11
+	vpxor	%xmm6, %xmm11, %xmm11
+	vpslld	$13, %xmm7, %xmm5
+	vpsrld	$19, %xmm7, %xmm7
+	vpxor	%xmm5, %xmm15, %xmm15
+	vpxor	%xmm7, %xmm15, %xmm15
+	
+	vpaddd	%xmm3, %xmm2, %xmm4
+	vpaddd	%xmm11, %xmm10, %xmm6
+	vpaddd	%xmm15, %xmm14, %xmm7
+	vpslld	$18, %xmm4, %xmm5
+	vpsrld	$14, %xmm4, %xmm4
+	vpshufd	$0x4e, %xmm2, %xmm2
+	vpshufd	$0x4e, %xmm10, %xmm10
+	vpxor	%xmm5, %xmm0, %xmm0
+	vpxor	%xmm4, %xmm0, %xmm0
+	vpslld	$18, %xmm6, %xmm5
+	vpsrld	$14, %xmm6, %xmm6
+	vpshufd	$0x4e, %xmm14, %xmm14
+	vpshufd	$0x39, %xmm11, %xmm11
+	vpxor	%xmm5, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm8, %xmm8
+	vpslld	$18, %xmm7, %xmm5
+	vpsrld	$14, %xmm7, %xmm7
+	vpshufd	$0x39, %xmm3, %xmm3
+	vpshufd	$0x39, %xmm15, %xmm15
+	vpxor	%xmm5, %xmm12, %xmm12
+	vpxor	%xmm7, %xmm12, %xmm12
+.endm
+
+.macro salsa8_core_3way_avx
+	salsa8_core_3way_avx_doubleround
+	salsa8_core_3way_avx_doubleround
+	salsa8_core_3way_avx_doubleround
+	salsa8_core_3way_avx_doubleround
+.endm
+#endif /* USE_AVX */
+	
+	.text
+	.p2align 6
+	.globl scrypt_core_3way
+	.globl _scrypt_core_3way
+scrypt_core_3way:
+_scrypt_core_3way:
+	pushq	%rbx
+	pushq	%rbp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	subq	$176, %rsp
+	movdqa	%xmm6, 8(%rsp)
+	movdqa	%xmm7, 24(%rsp)
+	movdqa	%xmm8, 40(%rsp)
+	movdqa	%xmm9, 56(%rsp)
+	movdqa	%xmm10, 72(%rsp)
+	movdqa	%xmm11, 88(%rsp)
+	movdqa	%xmm12, 104(%rsp)
+	movdqa	%xmm13, 120(%rsp)
+	movdqa	%xmm14, 136(%rsp)
+	movdqa	%xmm15, 152(%rsp)
+	pushq	%rdi
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+#else
+	movq	%rdx, %r8
+#endif
+	subq	$392, %rsp
+	
+.macro scrypt_core_3way_cleanup
+	addq	$392, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	popq	%rdi
+	movdqa	8(%rsp), %xmm6
+	movdqa	24(%rsp), %xmm7
+	movdqa	40(%rsp), %xmm8
+	movdqa	56(%rsp), %xmm9
+	movdqa	72(%rsp), %xmm10
+	movdqa	88(%rsp), %xmm11
+	movdqa	104(%rsp), %xmm12
+	movdqa	120(%rsp), %xmm13
+	movdqa	136(%rsp), %xmm14
+	movdqa	152(%rsp), %xmm15
+	addq	$176, %rsp
+#endif
+	popq	%rbp
+	popq	%rbx
+.endm
+	
+#if !defined(USE_AVX)
+	jmp scrypt_core_3way_xmm
+#else
+	/* Check for AVX and OSXSAVE support */
+	movl	$1, %eax
+	cpuid
+	andl	$0x18000000, %ecx
+	cmpl	$0x18000000, %ecx
+	jne scrypt_core_3way_xmm
+	/* Check for XMM and YMM state support */
+	xorl	%ecx, %ecx
+	xgetbv
+	andl	$0x00000006, %eax
+	cmpl	$0x00000006, %eax
+	jne scrypt_core_3way_xmm
+#if defined(USE_XOP)
+	/* Check for XOP support */
+	movl	$0x80000001, %eax
+	cpuid
+	andl	$0x00000800, %ecx
+	jnz scrypt_core_3way_xop
+#endif
+	
+scrypt_core_3way_avx:
+	scrypt_shuffle %rdi, 0, %rsp, 0
+	scrypt_shuffle %rdi, 64, %rsp, 64
+	scrypt_shuffle %rdi, 128, %rsp, 128
+	scrypt_shuffle %rdi, 192, %rsp, 192
+	scrypt_shuffle %rdi, 256, %rsp, 256
+	scrypt_shuffle %rdi, 320, %rsp, 320
+	
+	movdqa	64(%rsp), %xmm0
+	movdqa	80(%rsp), %xmm1
+	movdqa	96(%rsp), %xmm2
+	movdqa	112(%rsp), %xmm3
+	movdqa	128+64(%rsp), %xmm8
+	movdqa	128+80(%rsp), %xmm9
+	movdqa	128+96(%rsp), %xmm10
+	movdqa	128+112(%rsp), %xmm11
+	movdqa	256+64(%rsp), %xmm12
+	movdqa	256+80(%rsp), %xmm13
+	movdqa	256+96(%rsp), %xmm14
+	movdqa	256+112(%rsp), %xmm15
+	
+	movq	%rsi, %rbx
+	leaq	(%r8, %r8, 2), %rax
+	shlq	$7, %rax
+	addq	%rsi, %rax
+scrypt_core_3way_avx_loop1:
+	movdqa	%xmm0, 64(%rbx)
+	movdqa	%xmm1, 80(%rbx)
+	movdqa	%xmm2, 96(%rbx)
+	movdqa	%xmm3, 112(%rbx)
+	pxor	0(%rsp), %xmm0
+	pxor	16(%rsp), %xmm1
+	pxor	32(%rsp), %xmm2
+	pxor	48(%rsp), %xmm3
+	movdqa	%xmm8, 128+64(%rbx)
+	movdqa	%xmm9, 128+80(%rbx)
+	movdqa	%xmm10, 128+96(%rbx)
+	movdqa	%xmm11, 128+112(%rbx)
+	pxor	128+0(%rsp), %xmm8
+	pxor	128+16(%rsp), %xmm9
+	pxor	128+32(%rsp), %xmm10
+	pxor	128+48(%rsp), %xmm11
+	movdqa	%xmm12, 256+64(%rbx)
+	movdqa	%xmm13, 256+80(%rbx)
+	movdqa	%xmm14, 256+96(%rbx)
+	movdqa	%xmm15, 256+112(%rbx)
+	pxor	256+0(%rsp), %xmm12
+	pxor	256+16(%rsp), %xmm13
+	pxor	256+32(%rsp), %xmm14
+	pxor	256+48(%rsp), %xmm15
+	movdqa	%xmm0, 0(%rbx)
+	movdqa	%xmm1, 16(%rbx)
+	movdqa	%xmm2, 32(%rbx)
+	movdqa	%xmm3, 48(%rbx)
+	movdqa	%xmm8, 128+0(%rbx)
+	movdqa	%xmm9, 128+16(%rbx)
+	movdqa	%xmm10, 128+32(%rbx)
+	movdqa	%xmm11, 128+48(%rbx)
+	movdqa	%xmm12, 256+0(%rbx)
+	movdqa	%xmm13, 256+16(%rbx)
+	movdqa	%xmm14, 256+32(%rbx)
+	movdqa	%xmm15, 256+48(%rbx)
+	
+	salsa8_core_3way_avx
+	paddd	0(%rbx), %xmm0
+	paddd	16(%rbx), %xmm1
+	paddd	32(%rbx), %xmm2
+	paddd	48(%rbx), %xmm3
+	paddd	128+0(%rbx), %xmm8
+	paddd	128+16(%rbx), %xmm9
+	paddd	128+32(%rbx), %xmm10
+	paddd	128+48(%rbx), %xmm11
+	paddd	256+0(%rbx), %xmm12
+	paddd	256+16(%rbx), %xmm13
+	paddd	256+32(%rbx), %xmm14
+	paddd	256+48(%rbx), %xmm15
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	
+	pxor	64(%rbx), %xmm0
+	pxor	80(%rbx), %xmm1
+	pxor	96(%rbx), %xmm2
+	pxor	112(%rbx), %xmm3
+	pxor	128+64(%rbx), %xmm8
+	pxor	128+80(%rbx), %xmm9
+	pxor	128+96(%rbx), %xmm10
+	pxor	128+112(%rbx), %xmm11
+	pxor	256+64(%rbx), %xmm12
+	pxor	256+80(%rbx), %xmm13
+	pxor	256+96(%rbx), %xmm14
+	pxor	256+112(%rbx), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	salsa8_core_3way_avx
+	paddd	64(%rsp), %xmm0
+	paddd	80(%rsp), %xmm1
+	paddd	96(%rsp), %xmm2
+	paddd	112(%rsp), %xmm3
+	paddd	128+64(%rsp), %xmm8
+	paddd	128+80(%rsp), %xmm9
+	paddd	128+96(%rsp), %xmm10
+	paddd	128+112(%rsp), %xmm11
+	paddd	256+64(%rsp), %xmm12
+	paddd	256+80(%rsp), %xmm13
+	paddd	256+96(%rsp), %xmm14
+	paddd	256+112(%rsp), %xmm15
+	
+	addq	$3*128, %rbx
+	cmpq	%rax, %rbx
+	jne scrypt_core_3way_avx_loop1
+	
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	
+	movq	%r8, %rcx
+	subq	$1, %r8
+scrypt_core_3way_avx_loop2:
+	movd	%xmm0, %ebp
+	movd	%xmm8, %ebx
+	movd	%xmm12, %eax
+	pxor	0(%rsp), %xmm0
+	pxor	16(%rsp), %xmm1
+	pxor	32(%rsp), %xmm2
+	pxor	48(%rsp), %xmm3
+	pxor	128+0(%rsp), %xmm8
+	pxor	128+16(%rsp), %xmm9
+	pxor	128+32(%rsp), %xmm10
+	pxor	128+48(%rsp), %xmm11
+	pxor	256+0(%rsp), %xmm12
+	pxor	256+16(%rsp), %xmm13
+	pxor	256+32(%rsp), %xmm14
+	pxor	256+48(%rsp), %xmm15
+	andl	%r8d, %ebp
+	leaq	(%rbp, %rbp, 2), %rbp
+	shll	$7, %ebp
+	andl	%r8d, %ebx
+	leaq	1(%rbx, %rbx, 2), %rbx
+	shll	$7, %ebx
+	andl	%r8d, %eax
+	leaq	2(%rax, %rax, 2), %rax
+	shll	$7, %eax
+	pxor	0(%rsi, %rbp), %xmm0
+	pxor	16(%rsi, %rbp), %xmm1
+	pxor	32(%rsi, %rbp), %xmm2
+	pxor	48(%rsi, %rbp), %xmm3
+	pxor	0(%rsi, %rbx), %xmm8
+	pxor	16(%rsi, %rbx), %xmm9
+	pxor	32(%rsi, %rbx), %xmm10
+	pxor	48(%rsi, %rbx), %xmm11
+	pxor	0(%rsi, %rax), %xmm12
+	pxor	16(%rsi, %rax), %xmm13
+	pxor	32(%rsi, %rax), %xmm14
+	pxor	48(%rsi, %rax), %xmm15
+	
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	salsa8_core_3way_avx
+	paddd	0(%rsp), %xmm0
+	paddd	16(%rsp), %xmm1
+	paddd	32(%rsp), %xmm2
+	paddd	48(%rsp), %xmm3
+	paddd	128+0(%rsp), %xmm8
+	paddd	128+16(%rsp), %xmm9
+	paddd	128+32(%rsp), %xmm10
+	paddd	128+48(%rsp), %xmm11
+	paddd	256+0(%rsp), %xmm12
+	paddd	256+16(%rsp), %xmm13
+	paddd	256+32(%rsp), %xmm14
+	paddd	256+48(%rsp), %xmm15
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	
+	pxor	64(%rsi, %rbp), %xmm0
+	pxor	80(%rsi, %rbp), %xmm1
+	pxor	96(%rsi, %rbp), %xmm2
+	pxor	112(%rsi, %rbp), %xmm3
+	pxor	64(%rsi, %rbx), %xmm8
+	pxor	80(%rsi, %rbx), %xmm9
+	pxor	96(%rsi, %rbx), %xmm10
+	pxor	112(%rsi, %rbx), %xmm11
+	pxor	64(%rsi, %rax), %xmm12
+	pxor	80(%rsi, %rax), %xmm13
+	pxor	96(%rsi, %rax), %xmm14
+	pxor	112(%rsi, %rax), %xmm15
+	pxor	64(%rsp), %xmm0
+	pxor	80(%rsp), %xmm1
+	pxor	96(%rsp), %xmm2
+	pxor	112(%rsp), %xmm3
+	pxor	128+64(%rsp), %xmm8
+	pxor	128+80(%rsp), %xmm9
+	pxor	128+96(%rsp), %xmm10
+	pxor	128+112(%rsp), %xmm11
+	pxor	256+64(%rsp), %xmm12
+	pxor	256+80(%rsp), %xmm13
+	pxor	256+96(%rsp), %xmm14
+	pxor	256+112(%rsp), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	salsa8_core_3way_avx
+	paddd	64(%rsp), %xmm0
+	paddd	80(%rsp), %xmm1
+	paddd	96(%rsp), %xmm2
+	paddd	112(%rsp), %xmm3
+	paddd	128+64(%rsp), %xmm8
+	paddd	128+80(%rsp), %xmm9
+	paddd	128+96(%rsp), %xmm10
+	paddd	128+112(%rsp), %xmm11
+	paddd	256+64(%rsp), %xmm12
+	paddd	256+80(%rsp), %xmm13
+	paddd	256+96(%rsp), %xmm14
+	paddd	256+112(%rsp), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	
+	subq	$1, %rcx
+	ja scrypt_core_3way_avx_loop2
+	
+	scrypt_shuffle %rsp, 0, %rdi, 0
+	scrypt_shuffle %rsp, 64, %rdi, 64
+	scrypt_shuffle %rsp, 128, %rdi, 128
+	scrypt_shuffle %rsp, 192, %rdi, 192
+	scrypt_shuffle %rsp, 256, %rdi, 256
+	scrypt_shuffle %rsp, 320, %rdi, 320
+	
+	scrypt_core_3way_cleanup
+	ret
+
+#if defined(USE_XOP)
+.macro salsa8_core_3way_xop_doubleround
+	vpaddd	%xmm0, %xmm1, %xmm4
+	vpaddd	%xmm8, %xmm9, %xmm6
+	vpaddd	%xmm12, %xmm13, %xmm7
+	vprotd	$7, %xmm4, %xmm4
+	vprotd	$7, %xmm6, %xmm6
+	vprotd	$7, %xmm7, %xmm7
+	vpxor	%xmm4, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm11, %xmm11
+	vpxor	%xmm7, %xmm15, %xmm15
+	
+	vpaddd	%xmm3, %xmm0, %xmm4
+	vpaddd	%xmm11, %xmm8, %xmm6
+	vpaddd	%xmm15, %xmm12, %xmm7
+	vprotd	$9, %xmm4, %xmm4
+	vprotd	$9, %xmm6, %xmm6
+	vprotd	$9, %xmm7, %xmm7
+	vpxor	%xmm4, %xmm2, %xmm2
+	vpxor	%xmm6, %xmm10, %xmm10
+	vpxor	%xmm7, %xmm14, %xmm14
+	
+	vpaddd	%xmm2, %xmm3, %xmm4
+	vpaddd	%xmm10, %xmm11, %xmm6
+	vpaddd	%xmm14, %xmm15, %xmm7
+	vprotd	$13, %xmm4, %xmm4
+	vprotd	$13, %xmm6, %xmm6
+	vprotd	$13, %xmm7, %xmm7
+	vpshufd	$0x93, %xmm3, %xmm3
+	vpshufd	$0x93, %xmm11, %xmm11
+	vpshufd	$0x93, %xmm15, %xmm15
+	vpxor	%xmm4, %xmm1, %xmm1
+	vpxor	%xmm6, %xmm9, %xmm9
+	vpxor	%xmm7, %xmm13, %xmm13
+	
+	vpaddd	%xmm1, %xmm2, %xmm4
+	vpaddd	%xmm9, %xmm10, %xmm6
+	vpaddd	%xmm13, %xmm14, %xmm7
+	vprotd	$18, %xmm4, %xmm4
+	vprotd	$18, %xmm6, %xmm6
+	vprotd	$18, %xmm7, %xmm7
+	vpshufd	$0x4e, %xmm2, %xmm2
+	vpshufd	$0x4e, %xmm10, %xmm10
+	vpshufd	$0x4e, %xmm14, %xmm14
+	vpxor	%xmm6, %xmm8, %xmm8
+	vpxor	%xmm4, %xmm0, %xmm0
+	vpxor	%xmm7, %xmm12, %xmm12
+	
+	vpaddd	%xmm0, %xmm3, %xmm4
+	vpaddd	%xmm8, %xmm11, %xmm6
+	vpaddd	%xmm12, %xmm15, %xmm7
+	vprotd	$7, %xmm4, %xmm4
+	vprotd	$7, %xmm6, %xmm6
+	vprotd	$7, %xmm7, %xmm7
+	vpshufd	$0x39, %xmm1, %xmm1
+	vpshufd	$0x39, %xmm9, %xmm9
+	vpshufd	$0x39, %xmm13, %xmm13
+	vpxor	%xmm4, %xmm1, %xmm1
+	vpxor	%xmm6, %xmm9, %xmm9
+	vpxor	%xmm7, %xmm13, %xmm13
+	
+	vpaddd	%xmm1, %xmm0, %xmm4
+	vpaddd	%xmm9, %xmm8, %xmm6
+	vpaddd	%xmm13, %xmm12, %xmm7
+	vprotd	$9, %xmm4, %xmm4
+	vprotd	$9, %xmm6, %xmm6
+	vprotd	$9, %xmm7, %xmm7
+	vpxor	%xmm4, %xmm2, %xmm2
+	vpxor	%xmm6, %xmm10, %xmm10
+	vpxor	%xmm7, %xmm14, %xmm14
+	
+	vpaddd	%xmm2, %xmm1, %xmm4
+	vpaddd	%xmm10, %xmm9, %xmm6
+	vpaddd	%xmm14, %xmm13, %xmm7
+	vprotd	$13, %xmm4, %xmm4
+	vprotd	$13, %xmm6, %xmm6
+	vprotd	$13, %xmm7, %xmm7
+	vpshufd	$0x93, %xmm1, %xmm1
+	vpshufd	$0x93, %xmm9, %xmm9
+	vpshufd	$0x93, %xmm13, %xmm13
+	vpxor	%xmm4, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm11, %xmm11
+	vpxor	%xmm7, %xmm15, %xmm15
+	
+	vpaddd	%xmm3, %xmm2, %xmm4
+	vpaddd	%xmm11, %xmm10, %xmm6
+	vpaddd	%xmm15, %xmm14, %xmm7
+	vprotd	$18, %xmm4, %xmm4
+	vprotd	$18, %xmm6, %xmm6
+	vprotd	$18, %xmm7, %xmm7
+	vpshufd	$0x4e, %xmm2, %xmm2
+	vpshufd	$0x4e, %xmm10, %xmm10
+	vpshufd	$0x4e, %xmm14, %xmm14
+	vpxor	%xmm4, %xmm0, %xmm0
+	vpxor	%xmm6, %xmm8, %xmm8
+	vpxor	%xmm7, %xmm12, %xmm12
+	vpshufd	$0x39, %xmm3, %xmm3
+	vpshufd	$0x39, %xmm11, %xmm11
+	vpshufd	$0x39, %xmm15, %xmm15
+.endm
+
+.macro salsa8_core_3way_xop
+	salsa8_core_3way_xop_doubleround
+	salsa8_core_3way_xop_doubleround
+	salsa8_core_3way_xop_doubleround
+	salsa8_core_3way_xop_doubleround
+.endm
+	
+	.p2align 6
+scrypt_core_3way_xop:
+	scrypt_shuffle %rdi, 0, %rsp, 0
+	scrypt_shuffle %rdi, 64, %rsp, 64
+	scrypt_shuffle %rdi, 128, %rsp, 128
+	scrypt_shuffle %rdi, 192, %rsp, 192
+	scrypt_shuffle %rdi, 256, %rsp, 256
+	scrypt_shuffle %rdi, 320, %rsp, 320
+	
+	movdqa	64(%rsp), %xmm0
+	movdqa	80(%rsp), %xmm1
+	movdqa	96(%rsp), %xmm2
+	movdqa	112(%rsp), %xmm3
+	movdqa	128+64(%rsp), %xmm8
+	movdqa	128+80(%rsp), %xmm9
+	movdqa	128+96(%rsp), %xmm10
+	movdqa	128+112(%rsp), %xmm11
+	movdqa	256+64(%rsp), %xmm12
+	movdqa	256+80(%rsp), %xmm13
+	movdqa	256+96(%rsp), %xmm14
+	movdqa	256+112(%rsp), %xmm15
+	
+	movq	%rsi, %rbx
+	leaq	(%r8, %r8, 2), %rax
+	shlq	$7, %rax
+	addq	%rsi, %rax
+scrypt_core_3way_xop_loop1:
+	movdqa	%xmm0, 64(%rbx)
+	movdqa	%xmm1, 80(%rbx)
+	movdqa	%xmm2, 96(%rbx)
+	movdqa	%xmm3, 112(%rbx)
+	pxor	0(%rsp), %xmm0
+	pxor	16(%rsp), %xmm1
+	pxor	32(%rsp), %xmm2
+	pxor	48(%rsp), %xmm3
+	movdqa	%xmm8, 128+64(%rbx)
+	movdqa	%xmm9, 128+80(%rbx)
+	movdqa	%xmm10, 128+96(%rbx)
+	movdqa	%xmm11, 128+112(%rbx)
+	pxor	128+0(%rsp), %xmm8
+	pxor	128+16(%rsp), %xmm9
+	pxor	128+32(%rsp), %xmm10
+	pxor	128+48(%rsp), %xmm11
+	movdqa	%xmm12, 256+64(%rbx)
+	movdqa	%xmm13, 256+80(%rbx)
+	movdqa	%xmm14, 256+96(%rbx)
+	movdqa	%xmm15, 256+112(%rbx)
+	pxor	256+0(%rsp), %xmm12
+	pxor	256+16(%rsp), %xmm13
+	pxor	256+32(%rsp), %xmm14
+	pxor	256+48(%rsp), %xmm15
+	movdqa	%xmm0, 0(%rbx)
+	movdqa	%xmm1, 16(%rbx)
+	movdqa	%xmm2, 32(%rbx)
+	movdqa	%xmm3, 48(%rbx)
+	movdqa	%xmm8, 128+0(%rbx)
+	movdqa	%xmm9, 128+16(%rbx)
+	movdqa	%xmm10, 128+32(%rbx)
+	movdqa	%xmm11, 128+48(%rbx)
+	movdqa	%xmm12, 256+0(%rbx)
+	movdqa	%xmm13, 256+16(%rbx)
+	movdqa	%xmm14, 256+32(%rbx)
+	movdqa	%xmm15, 256+48(%rbx)
+	
+	salsa8_core_3way_xop
+	paddd	0(%rbx), %xmm0
+	paddd	16(%rbx), %xmm1
+	paddd	32(%rbx), %xmm2
+	paddd	48(%rbx), %xmm3
+	paddd	128+0(%rbx), %xmm8
+	paddd	128+16(%rbx), %xmm9
+	paddd	128+32(%rbx), %xmm10
+	paddd	128+48(%rbx), %xmm11
+	paddd	256+0(%rbx), %xmm12
+	paddd	256+16(%rbx), %xmm13
+	paddd	256+32(%rbx), %xmm14
+	paddd	256+48(%rbx), %xmm15
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	
+	pxor	64(%rbx), %xmm0
+	pxor	80(%rbx), %xmm1
+	pxor	96(%rbx), %xmm2
+	pxor	112(%rbx), %xmm3
+	pxor	128+64(%rbx), %xmm8
+	pxor	128+80(%rbx), %xmm9
+	pxor	128+96(%rbx), %xmm10
+	pxor	128+112(%rbx), %xmm11
+	pxor	256+64(%rbx), %xmm12
+	pxor	256+80(%rbx), %xmm13
+	pxor	256+96(%rbx), %xmm14
+	pxor	256+112(%rbx), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	salsa8_core_3way_xop
+	paddd	64(%rsp), %xmm0
+	paddd	80(%rsp), %xmm1
+	paddd	96(%rsp), %xmm2
+	paddd	112(%rsp), %xmm3
+	paddd	128+64(%rsp), %xmm8
+	paddd	128+80(%rsp), %xmm9
+	paddd	128+96(%rsp), %xmm10
+	paddd	128+112(%rsp), %xmm11
+	paddd	256+64(%rsp), %xmm12
+	paddd	256+80(%rsp), %xmm13
+	paddd	256+96(%rsp), %xmm14
+	paddd	256+112(%rsp), %xmm15
+	
+	addq	$3*128, %rbx
+	cmpq	%rax, %rbx
+	jne scrypt_core_3way_xop_loop1
+	
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	
+	movq	%r8, %rcx
+	subq	$1, %r8
+scrypt_core_3way_xop_loop2:
+	movd	%xmm0, %ebp
+	movd	%xmm8, %ebx
+	movd	%xmm12, %eax
+	pxor	0(%rsp), %xmm0
+	pxor	16(%rsp), %xmm1
+	pxor	32(%rsp), %xmm2
+	pxor	48(%rsp), %xmm3
+	pxor	128+0(%rsp), %xmm8
+	pxor	128+16(%rsp), %xmm9
+	pxor	128+32(%rsp), %xmm10
+	pxor	128+48(%rsp), %xmm11
+	pxor	256+0(%rsp), %xmm12
+	pxor	256+16(%rsp), %xmm13
+	pxor	256+32(%rsp), %xmm14
+	pxor	256+48(%rsp), %xmm15
+	andl	%r8d, %ebp
+	leaq	(%rbp, %rbp, 2), %rbp
+	shll	$7, %ebp
+	andl	%r8d, %ebx
+	leaq	1(%rbx, %rbx, 2), %rbx
+	shll	$7, %ebx
+	andl	%r8d, %eax
+	leaq	2(%rax, %rax, 2), %rax
+	shll	$7, %eax
+	pxor	0(%rsi, %rbp), %xmm0
+	pxor	16(%rsi, %rbp), %xmm1
+	pxor	32(%rsi, %rbp), %xmm2
+	pxor	48(%rsi, %rbp), %xmm3
+	pxor	0(%rsi, %rbx), %xmm8
+	pxor	16(%rsi, %rbx), %xmm9
+	pxor	32(%rsi, %rbx), %xmm10
+	pxor	48(%rsi, %rbx), %xmm11
+	pxor	0(%rsi, %rax), %xmm12
+	pxor	16(%rsi, %rax), %xmm13
+	pxor	32(%rsi, %rax), %xmm14
+	pxor	48(%rsi, %rax), %xmm15
+	
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	salsa8_core_3way_xop
+	paddd	0(%rsp), %xmm0
+	paddd	16(%rsp), %xmm1
+	paddd	32(%rsp), %xmm2
+	paddd	48(%rsp), %xmm3
+	paddd	128+0(%rsp), %xmm8
+	paddd	128+16(%rsp), %xmm9
+	paddd	128+32(%rsp), %xmm10
+	paddd	128+48(%rsp), %xmm11
+	paddd	256+0(%rsp), %xmm12
+	paddd	256+16(%rsp), %xmm13
+	paddd	256+32(%rsp), %xmm14
+	paddd	256+48(%rsp), %xmm15
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	
+	pxor	64(%rsi, %rbp), %xmm0
+	pxor	80(%rsi, %rbp), %xmm1
+	pxor	96(%rsi, %rbp), %xmm2
+	pxor	112(%rsi, %rbp), %xmm3
+	pxor	64(%rsi, %rbx), %xmm8
+	pxor	80(%rsi, %rbx), %xmm9
+	pxor	96(%rsi, %rbx), %xmm10
+	pxor	112(%rsi, %rbx), %xmm11
+	pxor	64(%rsi, %rax), %xmm12
+	pxor	80(%rsi, %rax), %xmm13
+	pxor	96(%rsi, %rax), %xmm14
+	pxor	112(%rsi, %rax), %xmm15
+	pxor	64(%rsp), %xmm0
+	pxor	80(%rsp), %xmm1
+	pxor	96(%rsp), %xmm2
+	pxor	112(%rsp), %xmm3
+	pxor	128+64(%rsp), %xmm8
+	pxor	128+80(%rsp), %xmm9
+	pxor	128+96(%rsp), %xmm10
+	pxor	128+112(%rsp), %xmm11
+	pxor	256+64(%rsp), %xmm12
+	pxor	256+80(%rsp), %xmm13
+	pxor	256+96(%rsp), %xmm14
+	pxor	256+112(%rsp), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	salsa8_core_3way_xop
+	paddd	64(%rsp), %xmm0
+	paddd	80(%rsp), %xmm1
+	paddd	96(%rsp), %xmm2
+	paddd	112(%rsp), %xmm3
+	paddd	128+64(%rsp), %xmm8
+	paddd	128+80(%rsp), %xmm9
+	paddd	128+96(%rsp), %xmm10
+	paddd	128+112(%rsp), %xmm11
+	paddd	256+64(%rsp), %xmm12
+	paddd	256+80(%rsp), %xmm13
+	paddd	256+96(%rsp), %xmm14
+	paddd	256+112(%rsp), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	
+	subq	$1, %rcx
+	ja scrypt_core_3way_xop_loop2
+	
+	scrypt_shuffle %rsp, 0, %rdi, 0
+	scrypt_shuffle %rsp, 64, %rdi, 64
+	scrypt_shuffle %rsp, 128, %rdi, 128
+	scrypt_shuffle %rsp, 192, %rdi, 192
+	scrypt_shuffle %rsp, 256, %rdi, 256
+	scrypt_shuffle %rsp, 320, %rdi, 320
+	
+	scrypt_core_3way_cleanup
+	ret
+#endif /* USE_XOP */
+#endif /* USE_AVX */
+	
+.macro salsa8_core_3way_xmm_doubleround
+	movdqa	%xmm1, %xmm4
+	movdqa	%xmm9, %xmm6
+	movdqa	%xmm13, %xmm7
+	paddd	%xmm0, %xmm4
+	paddd	%xmm8, %xmm6
+	paddd	%xmm12, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$7, %xmm4
+	psrld	$25, %xmm5
+	pxor	%xmm4, %xmm3
+	pxor	%xmm5, %xmm3
+	movdqa	%xmm0, %xmm4
+	movdqa	%xmm6, %xmm5
+	pslld	$7, %xmm6
+	psrld	$25, %xmm5
+	pxor	%xmm6, %xmm11
+	pxor	%xmm5, %xmm11
+	movdqa	%xmm8, %xmm6
+	movdqa	%xmm7, %xmm5
+	pslld	$7, %xmm7
+	psrld	$25, %xmm5
+	pxor	%xmm7, %xmm15
+	pxor	%xmm5, %xmm15
+	movdqa	%xmm12, %xmm7
+	
+	paddd	%xmm3, %xmm4
+	paddd	%xmm11, %xmm6
+	paddd	%xmm15, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$9, %xmm4
+	psrld	$23, %xmm5
+	pxor	%xmm4, %xmm2
+	movdqa	%xmm3, %xmm4
+	pshufd	$0x93, %xmm3, %xmm3
+	pxor	%xmm5, %xmm2
+	movdqa	%xmm6, %xmm5
+	pslld	$9, %xmm6
+	psrld	$23, %xmm5
+	pxor	%xmm6, %xmm10
+	movdqa	%xmm11, %xmm6
+	pshufd	$0x93, %xmm11, %xmm11
+	pxor	%xmm5, %xmm10
+	movdqa	%xmm7, %xmm5
+	pslld	$9, %xmm7
+	psrld	$23, %xmm5
+	pxor	%xmm7, %xmm14
+	movdqa	%xmm15, %xmm7
+	pxor	%xmm5, %xmm14
+	pshufd	$0x93, %xmm15, %xmm15
+	
+	paddd	%xmm2, %xmm4
+	paddd	%xmm10, %xmm6
+	paddd	%xmm14, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$13, %xmm4
+	psrld	$19, %xmm5
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm2, %xmm4
+	pshufd	$0x4e, %xmm2, %xmm2
+	pxor	%xmm5, %xmm1
+	movdqa	%xmm6, %xmm5
+	pslld	$13, %xmm6
+	psrld	$19, %xmm5
+	pxor	%xmm6, %xmm9
+	movdqa	%xmm10, %xmm6
+	pshufd	$0x4e, %xmm10, %xmm10
+	pxor	%xmm5, %xmm9
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm7
+	psrld	$19, %xmm5
+	pxor	%xmm7, %xmm13
+	movdqa	%xmm14, %xmm7
+	pshufd	$0x4e, %xmm14, %xmm14
+	pxor	%xmm5, %xmm13
+	
+	paddd	%xmm1, %xmm4
+	paddd	%xmm9, %xmm6
+	paddd	%xmm13, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$18, %xmm4
+	psrld	$14, %xmm5
+	pxor	%xmm4, %xmm0
+	pshufd	$0x39, %xmm1, %xmm1
+	pxor	%xmm5, %xmm0
+	movdqa	%xmm3, %xmm4
+	movdqa	%xmm6, %xmm5
+	pslld	$18, %xmm6
+	psrld	$14, %xmm5
+	pxor	%xmm6, %xmm8
+	pshufd	$0x39, %xmm9, %xmm9
+	pxor	%xmm5, %xmm8
+	movdqa	%xmm11, %xmm6
+	movdqa	%xmm7, %xmm5
+	pslld	$18, %xmm7
+	psrld	$14, %xmm5
+	pxor	%xmm7, %xmm12
+	movdqa	%xmm15, %xmm7
+	pxor	%xmm5, %xmm12
+	pshufd	$0x39, %xmm13, %xmm13
+	
+	paddd	%xmm0, %xmm4
+	paddd	%xmm8, %xmm6
+	paddd	%xmm12, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$7, %xmm4
+	psrld	$25, %xmm5
+	pxor	%xmm4, %xmm1
+	pxor	%xmm5, %xmm1
+	movdqa	%xmm0, %xmm4
+	movdqa	%xmm6, %xmm5
+	pslld	$7, %xmm6
+	psrld	$25, %xmm5
+	pxor	%xmm6, %xmm9
+	pxor	%xmm5, %xmm9
+	movdqa	%xmm8, %xmm6
+	movdqa	%xmm7, %xmm5
+	pslld	$7, %xmm7
+	psrld	$25, %xmm5
+	pxor	%xmm7, %xmm13
+	pxor	%xmm5, %xmm13
+	movdqa	%xmm12, %xmm7
+	
+	paddd	%xmm1, %xmm4
+	paddd	%xmm9, %xmm6
+	paddd	%xmm13, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$9, %xmm4
+	psrld	$23, %xmm5
+	pxor	%xmm4, %xmm2
+	movdqa	%xmm1, %xmm4
+	pshufd	$0x93, %xmm1, %xmm1
+	pxor	%xmm5, %xmm2
+	movdqa	%xmm6, %xmm5
+	pslld	$9, %xmm6
+	psrld	$23, %xmm5
+	pxor	%xmm6, %xmm10
+	movdqa	%xmm9, %xmm6
+	pshufd	$0x93, %xmm9, %xmm9
+	pxor	%xmm5, %xmm10
+	movdqa	%xmm7, %xmm5
+	pslld	$9, %xmm7
+	psrld	$23, %xmm5
+	pxor	%xmm7, %xmm14
+	movdqa	%xmm13, %xmm7
+	pshufd	$0x93, %xmm13, %xmm13
+	pxor	%xmm5, %xmm14
+	
+	paddd	%xmm2, %xmm4
+	paddd	%xmm10, %xmm6
+	paddd	%xmm14, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$13, %xmm4
+	psrld	$19, %xmm5
+	pxor	%xmm4, %xmm3
+	movdqa	%xmm2, %xmm4
+	pshufd	$0x4e, %xmm2, %xmm2
+	pxor	%xmm5, %xmm3
+	movdqa	%xmm6, %xmm5
+	pslld	$13, %xmm6
+	psrld	$19, %xmm5
+	pxor	%xmm6, %xmm11
+	movdqa	%xmm10, %xmm6
+	pshufd	$0x4e, %xmm10, %xmm10
+	pxor	%xmm5, %xmm11
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm7
+	psrld	$19, %xmm5
+	pxor	%xmm7, %xmm15
+	movdqa	%xmm14, %xmm7
+	pshufd	$0x4e, %xmm14, %xmm14
+	pxor	%xmm5, %xmm15
+	
+	paddd	%xmm3, %xmm4
+	paddd	%xmm11, %xmm6
+	paddd	%xmm15, %xmm7
+	movdqa	%xmm4, %xmm5
+	pslld	$18, %xmm4
+	psrld	$14, %xmm5
+	pxor	%xmm4, %xmm0
+	pshufd	$0x39, %xmm3, %xmm3
+	pxor	%xmm5, %xmm0
+	movdqa	%xmm6, %xmm5
+	pslld	$18, %xmm6
+	psrld	$14, %xmm5
+	pxor	%xmm6, %xmm8
+	pshufd	$0x39, %xmm11, %xmm11
+	pxor	%xmm5, %xmm8
+	movdqa	%xmm7, %xmm5
+	pslld	$18, %xmm7
+	psrld	$14, %xmm5
+	pxor	%xmm7, %xmm12
+	pshufd	$0x39, %xmm15, %xmm15
+	pxor	%xmm5, %xmm12
+.endm
+
+.macro salsa8_core_3way_xmm
+	salsa8_core_3way_xmm_doubleround
+	salsa8_core_3way_xmm_doubleround
+	salsa8_core_3way_xmm_doubleround
+	salsa8_core_3way_xmm_doubleround
+.endm
+	
+	.p2align 6
+scrypt_core_3way_xmm:
+	scrypt_shuffle %rdi, 0, %rsp, 0
+	scrypt_shuffle %rdi, 64, %rsp, 64
+	scrypt_shuffle %rdi, 128, %rsp, 128
+	scrypt_shuffle %rdi, 192, %rsp, 192
+	scrypt_shuffle %rdi, 256, %rsp, 256
+	scrypt_shuffle %rdi, 320, %rsp, 320
+	
+	movdqa	64(%rsp), %xmm0
+	movdqa	80(%rsp), %xmm1
+	movdqa	96(%rsp), %xmm2
+	movdqa	112(%rsp), %xmm3
+	movdqa	128+64(%rsp), %xmm8
+	movdqa	128+80(%rsp), %xmm9
+	movdqa	128+96(%rsp), %xmm10
+	movdqa	128+112(%rsp), %xmm11
+	movdqa	256+64(%rsp), %xmm12
+	movdqa	256+80(%rsp), %xmm13
+	movdqa	256+96(%rsp), %xmm14
+	movdqa	256+112(%rsp), %xmm15
+	
+	movq	%rsi, %rbx
+	leaq	(%r8, %r8, 2), %rax
+	shlq	$7, %rax
+	addq	%rsi, %rax
+scrypt_core_3way_xmm_loop1:
+	movdqa	%xmm0, 64(%rbx)
+	movdqa	%xmm1, 80(%rbx)
+	movdqa	%xmm2, 96(%rbx)
+	movdqa	%xmm3, 112(%rbx)
+	pxor	0(%rsp), %xmm0
+	pxor	16(%rsp), %xmm1
+	pxor	32(%rsp), %xmm2
+	pxor	48(%rsp), %xmm3
+	movdqa	%xmm8, 128+64(%rbx)
+	movdqa	%xmm9, 128+80(%rbx)
+	movdqa	%xmm10, 128+96(%rbx)
+	movdqa	%xmm11, 128+112(%rbx)
+	pxor	128+0(%rsp), %xmm8
+	pxor	128+16(%rsp), %xmm9
+	pxor	128+32(%rsp), %xmm10
+	pxor	128+48(%rsp), %xmm11
+	movdqa	%xmm12, 256+64(%rbx)
+	movdqa	%xmm13, 256+80(%rbx)
+	movdqa	%xmm14, 256+96(%rbx)
+	movdqa	%xmm15, 256+112(%rbx)
+	pxor	256+0(%rsp), %xmm12
+	pxor	256+16(%rsp), %xmm13
+	pxor	256+32(%rsp), %xmm14
+	pxor	256+48(%rsp), %xmm15
+	movdqa	%xmm0, 0(%rbx)
+	movdqa	%xmm1, 16(%rbx)
+	movdqa	%xmm2, 32(%rbx)
+	movdqa	%xmm3, 48(%rbx)
+	movdqa	%xmm8, 128+0(%rbx)
+	movdqa	%xmm9, 128+16(%rbx)
+	movdqa	%xmm10, 128+32(%rbx)
+	movdqa	%xmm11, 128+48(%rbx)
+	movdqa	%xmm12, 256+0(%rbx)
+	movdqa	%xmm13, 256+16(%rbx)
+	movdqa	%xmm14, 256+32(%rbx)
+	movdqa	%xmm15, 256+48(%rbx)
+	
+	salsa8_core_3way_xmm
+	paddd	0(%rbx), %xmm0
+	paddd	16(%rbx), %xmm1
+	paddd	32(%rbx), %xmm2
+	paddd	48(%rbx), %xmm3
+	paddd	128+0(%rbx), %xmm8
+	paddd	128+16(%rbx), %xmm9
+	paddd	128+32(%rbx), %xmm10
+	paddd	128+48(%rbx), %xmm11
+	paddd	256+0(%rbx), %xmm12
+	paddd	256+16(%rbx), %xmm13
+	paddd	256+32(%rbx), %xmm14
+	paddd	256+48(%rbx), %xmm15
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	
+	pxor	64(%rbx), %xmm0
+	pxor	80(%rbx), %xmm1
+	pxor	96(%rbx), %xmm2
+	pxor	112(%rbx), %xmm3
+	pxor	128+64(%rbx), %xmm8
+	pxor	128+80(%rbx), %xmm9
+	pxor	128+96(%rbx), %xmm10
+	pxor	128+112(%rbx), %xmm11
+	pxor	256+64(%rbx), %xmm12
+	pxor	256+80(%rbx), %xmm13
+	pxor	256+96(%rbx), %xmm14
+	pxor	256+112(%rbx), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	salsa8_core_3way_xmm
+	paddd	64(%rsp), %xmm0
+	paddd	80(%rsp), %xmm1
+	paddd	96(%rsp), %xmm2
+	paddd	112(%rsp), %xmm3
+	paddd	128+64(%rsp), %xmm8
+	paddd	128+80(%rsp), %xmm9
+	paddd	128+96(%rsp), %xmm10
+	paddd	128+112(%rsp), %xmm11
+	paddd	256+64(%rsp), %xmm12
+	paddd	256+80(%rsp), %xmm13
+	paddd	256+96(%rsp), %xmm14
+	paddd	256+112(%rsp), %xmm15
+	
+	addq	$3*128, %rbx
+	cmpq	%rax, %rbx
+	jne scrypt_core_3way_xmm_loop1
+	
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	
+	movq	%r8, %rcx
+	subq	$1, %r8
+scrypt_core_3way_xmm_loop2:
+	movd	%xmm0, %ebp
+	movd	%xmm8, %ebx
+	movd	%xmm12, %eax
+	pxor	0(%rsp), %xmm0
+	pxor	16(%rsp), %xmm1
+	pxor	32(%rsp), %xmm2
+	pxor	48(%rsp), %xmm3
+	pxor	128+0(%rsp), %xmm8
+	pxor	128+16(%rsp), %xmm9
+	pxor	128+32(%rsp), %xmm10
+	pxor	128+48(%rsp), %xmm11
+	pxor	256+0(%rsp), %xmm12
+	pxor	256+16(%rsp), %xmm13
+	pxor	256+32(%rsp), %xmm14
+	pxor	256+48(%rsp), %xmm15
+	andl	%r8d, %ebp
+	leaq	(%rbp, %rbp, 2), %rbp
+	shll	$7, %ebp
+	andl	%r8d, %ebx
+	leaq	1(%rbx, %rbx, 2), %rbx
+	shll	$7, %ebx
+	andl	%r8d, %eax
+	leaq	2(%rax, %rax, 2), %rax
+	shll	$7, %eax
+	pxor	0(%rsi, %rbp), %xmm0
+	pxor	16(%rsi, %rbp), %xmm1
+	pxor	32(%rsi, %rbp), %xmm2
+	pxor	48(%rsi, %rbp), %xmm3
+	pxor	0(%rsi, %rbx), %xmm8
+	pxor	16(%rsi, %rbx), %xmm9
+	pxor	32(%rsi, %rbx), %xmm10
+	pxor	48(%rsi, %rbx), %xmm11
+	pxor	0(%rsi, %rax), %xmm12
+	pxor	16(%rsi, %rax), %xmm13
+	pxor	32(%rsi, %rax), %xmm14
+	pxor	48(%rsi, %rax), %xmm15
+	
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	salsa8_core_3way_xmm
+	paddd	0(%rsp), %xmm0
+	paddd	16(%rsp), %xmm1
+	paddd	32(%rsp), %xmm2
+	paddd	48(%rsp), %xmm3
+	paddd	128+0(%rsp), %xmm8
+	paddd	128+16(%rsp), %xmm9
+	paddd	128+32(%rsp), %xmm10
+	paddd	128+48(%rsp), %xmm11
+	paddd	256+0(%rsp), %xmm12
+	paddd	256+16(%rsp), %xmm13
+	paddd	256+32(%rsp), %xmm14
+	paddd	256+48(%rsp), %xmm15
+	movdqa	%xmm0, 0(%rsp)
+	movdqa	%xmm1, 16(%rsp)
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm8, 128+0(%rsp)
+	movdqa	%xmm9, 128+16(%rsp)
+	movdqa	%xmm10, 128+32(%rsp)
+	movdqa	%xmm11, 128+48(%rsp)
+	movdqa	%xmm12, 256+0(%rsp)
+	movdqa	%xmm13, 256+16(%rsp)
+	movdqa	%xmm14, 256+32(%rsp)
+	movdqa	%xmm15, 256+48(%rsp)
+	
+	pxor	64(%rsi, %rbp), %xmm0
+	pxor	80(%rsi, %rbp), %xmm1
+	pxor	96(%rsi, %rbp), %xmm2
+	pxor	112(%rsi, %rbp), %xmm3
+	pxor	64(%rsi, %rbx), %xmm8
+	pxor	80(%rsi, %rbx), %xmm9
+	pxor	96(%rsi, %rbx), %xmm10
+	pxor	112(%rsi, %rbx), %xmm11
+	pxor	64(%rsi, %rax), %xmm12
+	pxor	80(%rsi, %rax), %xmm13
+	pxor	96(%rsi, %rax), %xmm14
+	pxor	112(%rsi, %rax), %xmm15
+	pxor	64(%rsp), %xmm0
+	pxor	80(%rsp), %xmm1
+	pxor	96(%rsp), %xmm2
+	pxor	112(%rsp), %xmm3
+	pxor	128+64(%rsp), %xmm8
+	pxor	128+80(%rsp), %xmm9
+	pxor	128+96(%rsp), %xmm10
+	pxor	128+112(%rsp), %xmm11
+	pxor	256+64(%rsp), %xmm12
+	pxor	256+80(%rsp), %xmm13
+	pxor	256+96(%rsp), %xmm14
+	pxor	256+112(%rsp), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	salsa8_core_3way_xmm
+	paddd	64(%rsp), %xmm0
+	paddd	80(%rsp), %xmm1
+	paddd	96(%rsp), %xmm2
+	paddd	112(%rsp), %xmm3
+	paddd	128+64(%rsp), %xmm8
+	paddd	128+80(%rsp), %xmm9
+	paddd	128+96(%rsp), %xmm10
+	paddd	128+112(%rsp), %xmm11
+	paddd	256+64(%rsp), %xmm12
+	paddd	256+80(%rsp), %xmm13
+	paddd	256+96(%rsp), %xmm14
+	paddd	256+112(%rsp), %xmm15
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm1, 80(%rsp)
+	movdqa	%xmm2, 96(%rsp)
+	movdqa	%xmm3, 112(%rsp)
+	movdqa	%xmm8, 128+64(%rsp)
+	movdqa	%xmm9, 128+80(%rsp)
+	movdqa	%xmm10, 128+96(%rsp)
+	movdqa	%xmm11, 128+112(%rsp)
+	movdqa	%xmm12, 256+64(%rsp)
+	movdqa	%xmm13, 256+80(%rsp)
+	movdqa	%xmm14, 256+96(%rsp)
+	movdqa	%xmm15, 256+112(%rsp)
+	
+	subq	$1, %rcx
+	ja scrypt_core_3way_xmm_loop2
+	
+	scrypt_shuffle %rsp, 0, %rdi, 0
+	scrypt_shuffle %rsp, 64, %rdi, 64
+	scrypt_shuffle %rsp, 128, %rdi, 128
+	scrypt_shuffle %rsp, 192, %rdi, 192
+	scrypt_shuffle %rsp, 256, %rdi, 256
+	scrypt_shuffle %rsp, 320, %rdi, 320
+	
+	scrypt_core_3way_cleanup
+	ret
+
+
+#if defined(USE_AVX2)
+
+.macro salsa8_core_6way_avx2_doubleround
+	vpaddd	%ymm0, %ymm1, %ymm4
+	vpaddd	%ymm8, %ymm9, %ymm6
+	vpaddd	%ymm12, %ymm13, %ymm7
+	vpslld	$7, %ymm4, %ymm5
+	vpsrld	$25, %ymm4, %ymm4
+	vpxor	%ymm5, %ymm3, %ymm3
+	vpxor	%ymm4, %ymm3, %ymm3
+	vpslld	$7, %ymm6, %ymm5
+	vpsrld	$25, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm11, %ymm11
+	vpxor	%ymm6, %ymm11, %ymm11
+	vpslld	$7, %ymm7, %ymm5
+	vpsrld	$25, %ymm7, %ymm7
+	vpxor	%ymm5, %ymm15, %ymm15
+	vpxor	%ymm7, %ymm15, %ymm15
+	
+	vpaddd	%ymm3, %ymm0, %ymm4
+	vpaddd	%ymm11, %ymm8, %ymm6
+	vpaddd	%ymm15, %ymm12, %ymm7
+	vpslld	$9, %ymm4, %ymm5
+	vpsrld	$23, %ymm4, %ymm4
+	vpxor	%ymm5, %ymm2, %ymm2
+	vpxor	%ymm4, %ymm2, %ymm2
+	vpslld	$9, %ymm6, %ymm5
+	vpsrld	$23, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm10, %ymm10
+	vpxor	%ymm6, %ymm10, %ymm10
+	vpslld	$9, %ymm7, %ymm5
+	vpsrld	$23, %ymm7, %ymm7
+	vpxor	%ymm5, %ymm14, %ymm14
+	vpxor	%ymm7, %ymm14, %ymm14
+	
+	vpaddd	%ymm2, %ymm3, %ymm4
+	vpaddd	%ymm10, %ymm11, %ymm6
+	vpaddd	%ymm14, %ymm15, %ymm7
+	vpslld	$13, %ymm4, %ymm5
+	vpsrld	$19, %ymm4, %ymm4
+	vpshufd	$0x93, %ymm3, %ymm3
+	vpshufd	$0x93, %ymm11, %ymm11
+	vpshufd	$0x93, %ymm15, %ymm15
+	vpxor	%ymm5, %ymm1, %ymm1
+	vpxor	%ymm4, %ymm1, %ymm1
+	vpslld	$13, %ymm6, %ymm5
+	vpsrld	$19, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm9, %ymm9
+	vpxor	%ymm6, %ymm9, %ymm9
+	vpslld	$13, %ymm7, %ymm5
+	vpsrld	$19, %ymm7, %ymm7
+	vpxor	%ymm5, %ymm13, %ymm13
+	vpxor	%ymm7, %ymm13, %ymm13
+	
+	vpaddd	%ymm1, %ymm2, %ymm4
+	vpaddd	%ymm9, %ymm10, %ymm6
+	vpaddd	%ymm13, %ymm14, %ymm7
+	vpslld	$18, %ymm4, %ymm5
+	vpsrld	$14, %ymm4, %ymm4
+	vpshufd	$0x4e, %ymm2, %ymm2
+	vpshufd	$0x4e, %ymm10, %ymm10
+	vpshufd	$0x4e, %ymm14, %ymm14
+	vpxor	%ymm5, %ymm0, %ymm0
+	vpxor	%ymm4, %ymm0, %ymm0
+	vpslld	$18, %ymm6, %ymm5
+	vpsrld	$14, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm8, %ymm8
+	vpxor	%ymm6, %ymm8, %ymm8
+	vpslld	$18, %ymm7, %ymm5
+	vpsrld	$14, %ymm7, %ymm7
+	vpxor	%ymm5, %ymm12, %ymm12
+	vpxor	%ymm7, %ymm12, %ymm12
+	
+	vpaddd	%ymm0, %ymm3, %ymm4
+	vpaddd	%ymm8, %ymm11, %ymm6
+	vpaddd	%ymm12, %ymm15, %ymm7
+	vpslld	$7, %ymm4, %ymm5
+	vpsrld	$25, %ymm4, %ymm4
+	vpshufd	$0x39, %ymm1, %ymm1
+	vpxor	%ymm5, %ymm1, %ymm1
+	vpxor	%ymm4, %ymm1, %ymm1
+	vpslld	$7, %ymm6, %ymm5
+	vpsrld	$25, %ymm6, %ymm6
+	vpshufd	$0x39, %ymm9, %ymm9
+	vpxor	%ymm5, %ymm9, %ymm9
+	vpxor	%ymm6, %ymm9, %ymm9
+	vpslld	$7, %ymm7, %ymm5
+	vpsrld	$25, %ymm7, %ymm7
+	vpshufd	$0x39, %ymm13, %ymm13
+	vpxor	%ymm5, %ymm13, %ymm13
+	vpxor	%ymm7, %ymm13, %ymm13
+	
+	vpaddd	%ymm1, %ymm0, %ymm4
+	vpaddd	%ymm9, %ymm8, %ymm6
+	vpaddd	%ymm13, %ymm12, %ymm7
+	vpslld	$9, %ymm4, %ymm5
+	vpsrld	$23, %ymm4, %ymm4
+	vpxor	%ymm5, %ymm2, %ymm2
+	vpxor	%ymm4, %ymm2, %ymm2
+	vpslld	$9, %ymm6, %ymm5
+	vpsrld	$23, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm10, %ymm10
+	vpxor	%ymm6, %ymm10, %ymm10
+	vpslld	$9, %ymm7, %ymm5
+	vpsrld	$23, %ymm7, %ymm7
+	vpxor	%ymm5, %ymm14, %ymm14
+	vpxor	%ymm7, %ymm14, %ymm14
+	
+	vpaddd	%ymm2, %ymm1, %ymm4
+	vpaddd	%ymm10, %ymm9, %ymm6
+	vpaddd	%ymm14, %ymm13, %ymm7
+	vpslld	$13, %ymm4, %ymm5
+	vpsrld	$19, %ymm4, %ymm4
+	vpshufd	$0x93, %ymm1, %ymm1
+	vpshufd	$0x93, %ymm9, %ymm9
+	vpshufd	$0x93, %ymm13, %ymm13
+	vpxor	%ymm5, %ymm3, %ymm3
+	vpxor	%ymm4, %ymm3, %ymm3
+	vpslld	$13, %ymm6, %ymm5
+	vpsrld	$19, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm11, %ymm11
+	vpxor	%ymm6, %ymm11, %ymm11
+	vpslld	$13, %ymm7, %ymm5
+	vpsrld	$19, %ymm7, %ymm7
+	vpxor	%ymm5, %ymm15, %ymm15
+	vpxor	%ymm7, %ymm15, %ymm15
+	
+	vpaddd	%ymm3, %ymm2, %ymm4
+	vpaddd	%ymm11, %ymm10, %ymm6
+	vpaddd	%ymm15, %ymm14, %ymm7
+	vpslld	$18, %ymm4, %ymm5
+	vpsrld	$14, %ymm4, %ymm4
+	vpshufd	$0x4e, %ymm2, %ymm2
+	vpshufd	$0x4e, %ymm10, %ymm10
+	vpxor	%ymm5, %ymm0, %ymm0
+	vpxor	%ymm4, %ymm0, %ymm0
+	vpslld	$18, %ymm6, %ymm5
+	vpsrld	$14, %ymm6, %ymm6
+	vpshufd	$0x4e, %ymm14, %ymm14
+	vpshufd	$0x39, %ymm11, %ymm11
+	vpxor	%ymm5, %ymm8, %ymm8
+	vpxor	%ymm6, %ymm8, %ymm8
+	vpslld	$18, %ymm7, %ymm5
+	vpsrld	$14, %ymm7, %ymm7
+	vpshufd	$0x39, %ymm3, %ymm3
+	vpshufd	$0x39, %ymm15, %ymm15
+	vpxor	%ymm5, %ymm12, %ymm12
+	vpxor	%ymm7, %ymm12, %ymm12
+.endm
+
+.macro salsa8_core_6way_avx2
+	salsa8_core_6way_avx2_doubleround
+	salsa8_core_6way_avx2_doubleround
+	salsa8_core_6way_avx2_doubleround
+	salsa8_core_6way_avx2_doubleround
+.endm
+	
+	.text
+	.p2align 6
+	.globl scrypt_core_6way
+	.globl _scrypt_core_6way
+scrypt_core_6way:
+_scrypt_core_6way:
+	pushq	%rbx
+	pushq	%rbp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	subq	$176, %rsp
+	vmovdqa	%xmm6, 8(%rsp)
+	vmovdqa	%xmm7, 24(%rsp)
+	vmovdqa	%xmm8, 40(%rsp)
+	vmovdqa	%xmm9, 56(%rsp)
+	vmovdqa	%xmm10, 72(%rsp)
+	vmovdqa	%xmm11, 88(%rsp)
+	vmovdqa	%xmm12, 104(%rsp)
+	vmovdqa	%xmm13, 120(%rsp)
+	vmovdqa	%xmm14, 136(%rsp)
+	vmovdqa	%xmm15, 152(%rsp)
+	pushq	%rdi
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+#else
+	movq	%rdx, %r8
+#endif
+	movq	%rsp, %rdx
+	subq	$768, %rsp
+	andq	$-128, %rsp
+	
+.macro scrypt_core_6way_cleanup
+	movq	%rdx, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	popq	%rdi
+	vmovdqa	8(%rsp), %xmm6
+	vmovdqa	24(%rsp), %xmm7
+	vmovdqa	40(%rsp), %xmm8
+	vmovdqa	56(%rsp), %xmm9
+	vmovdqa	72(%rsp), %xmm10
+	vmovdqa	88(%rsp), %xmm11
+	vmovdqa	104(%rsp), %xmm12
+	vmovdqa	120(%rsp), %xmm13
+	vmovdqa	136(%rsp), %xmm14
+	vmovdqa	152(%rsp), %xmm15
+	addq	$176, %rsp
+#endif
+	popq	%rbp
+	popq	%rbx
+.endm
+
+.macro scrypt_shuffle_pack2 src, so, dest, do
+	vmovdqa	\so+0*16(\src), %xmm0
+	vmovdqa	\so+1*16(\src), %xmm1
+	vmovdqa	\so+2*16(\src), %xmm2
+	vmovdqa	\so+3*16(\src), %xmm3
+	vinserti128	$1, \so+128+0*16(\src), %ymm0, %ymm0
+	vinserti128	$1, \so+128+1*16(\src), %ymm1, %ymm1
+	vinserti128	$1, \so+128+2*16(\src), %ymm2, %ymm2
+	vinserti128	$1, \so+128+3*16(\src), %ymm3, %ymm3
+	vpblendd	$0x33, %ymm0, %ymm2, %ymm4
+	vpblendd	$0xcc, %ymm1, %ymm3, %ymm5
+	vpblendd	$0x33, %ymm2, %ymm0, %ymm6
+	vpblendd	$0xcc, %ymm3, %ymm1, %ymm7
+	vpblendd	$0x55, %ymm7, %ymm6, %ymm3
+	vpblendd	$0x55, %ymm6, %ymm5, %ymm2
+	vpblendd	$0x55, %ymm5, %ymm4, %ymm1
+	vpblendd	$0x55, %ymm4, %ymm7, %ymm0
+	vmovdqa	%ymm0, \do+0*32(\dest)
+	vmovdqa	%ymm1, \do+1*32(\dest)
+	vmovdqa	%ymm2, \do+2*32(\dest)
+	vmovdqa	%ymm3, \do+3*32(\dest)
+.endm
+
+.macro scrypt_shuffle_unpack2 src, so, dest, do
+	vmovdqa	\so+0*32(\src), %ymm0
+	vmovdqa	\so+1*32(\src), %ymm1
+	vmovdqa	\so+2*32(\src), %ymm2
+	vmovdqa	\so+3*32(\src), %ymm3
+	vpblendd	$0x33, %ymm0, %ymm2, %ymm4
+	vpblendd	$0xcc, %ymm1, %ymm3, %ymm5
+	vpblendd	$0x33, %ymm2, %ymm0, %ymm6
+	vpblendd	$0xcc, %ymm3, %ymm1, %ymm7
+	vpblendd	$0x55, %ymm7, %ymm6, %ymm3
+	vpblendd	$0x55, %ymm6, %ymm5, %ymm2
+	vpblendd	$0x55, %ymm5, %ymm4, %ymm1
+	vpblendd	$0x55, %ymm4, %ymm7, %ymm0
+	vmovdqa	%xmm0, \do+0*16(\dest)
+	vmovdqa	%xmm1, \do+1*16(\dest)
+	vmovdqa	%xmm2, \do+2*16(\dest)
+	vmovdqa	%xmm3, \do+3*16(\dest)
+	vextracti128	$1, %ymm0, \do+128+0*16(\dest)
+	vextracti128	$1, %ymm1, \do+128+1*16(\dest)
+	vextracti128	$1, %ymm2, \do+128+2*16(\dest)
+	vextracti128	$1, %ymm3, \do+128+3*16(\dest)
+.endm
+	
+scrypt_core_6way_avx2:
+	scrypt_shuffle_pack2 %rdi, 0*256+0, %rsp, 0*128
+	scrypt_shuffle_pack2 %rdi, 0*256+64, %rsp, 1*128
+	scrypt_shuffle_pack2 %rdi, 1*256+0, %rsp, 2*128
+	scrypt_shuffle_pack2 %rdi, 1*256+64, %rsp, 3*128
+	scrypt_shuffle_pack2 %rdi, 2*256+0, %rsp, 4*128
+	scrypt_shuffle_pack2 %rdi, 2*256+64, %rsp, 5*128
+	
+	vmovdqa	0*256+4*32(%rsp), %ymm0
+	vmovdqa	0*256+5*32(%rsp), %ymm1
+	vmovdqa	0*256+6*32(%rsp), %ymm2
+	vmovdqa	0*256+7*32(%rsp), %ymm3
+	vmovdqa	1*256+4*32(%rsp), %ymm8
+	vmovdqa	1*256+5*32(%rsp), %ymm9
+	vmovdqa	1*256+6*32(%rsp), %ymm10
+	vmovdqa	1*256+7*32(%rsp), %ymm11
+	vmovdqa	2*256+4*32(%rsp), %ymm12
+	vmovdqa	2*256+5*32(%rsp), %ymm13
+	vmovdqa	2*256+6*32(%rsp), %ymm14
+	vmovdqa	2*256+7*32(%rsp), %ymm15
+	
+	movq	%rsi, %rbx
+	leaq	(%r8, %r8, 2), %rax
+	shlq	$8, %rax
+	addq	%rsi, %rax
+scrypt_core_6way_avx2_loop1:
+	vmovdqa	%ymm0, 0*256+4*32(%rbx)
+	vmovdqa	%ymm1, 0*256+5*32(%rbx)
+	vmovdqa	%ymm2, 0*256+6*32(%rbx)
+	vmovdqa	%ymm3, 0*256+7*32(%rbx)
+	vpxor	0*256+0*32(%rsp), %ymm0, %ymm0
+	vpxor	0*256+1*32(%rsp), %ymm1, %ymm1
+	vpxor	0*256+2*32(%rsp), %ymm2, %ymm2
+	vpxor	0*256+3*32(%rsp), %ymm3, %ymm3
+	vmovdqa	%ymm8, 1*256+4*32(%rbx)
+	vmovdqa	%ymm9, 1*256+5*32(%rbx)
+	vmovdqa	%ymm10, 1*256+6*32(%rbx)
+	vmovdqa	%ymm11, 1*256+7*32(%rbx)
+	vpxor	1*256+0*32(%rsp), %ymm8, %ymm8
+	vpxor	1*256+1*32(%rsp), %ymm9, %ymm9
+	vpxor	1*256+2*32(%rsp), %ymm10, %ymm10
+	vpxor	1*256+3*32(%rsp), %ymm11, %ymm11
+	vmovdqa	%ymm12, 2*256+4*32(%rbx)
+	vmovdqa	%ymm13, 2*256+5*32(%rbx)
+	vmovdqa	%ymm14, 2*256+6*32(%rbx)
+	vmovdqa	%ymm15, 2*256+7*32(%rbx)
+	vpxor	2*256+0*32(%rsp), %ymm12, %ymm12
+	vpxor	2*256+1*32(%rsp), %ymm13, %ymm13
+	vpxor	2*256+2*32(%rsp), %ymm14, %ymm14
+	vpxor	2*256+3*32(%rsp), %ymm15, %ymm15
+	vmovdqa	%ymm0, 0*256+0*32(%rbx)
+	vmovdqa	%ymm1, 0*256+1*32(%rbx)
+	vmovdqa	%ymm2, 0*256+2*32(%rbx)
+	vmovdqa	%ymm3, 0*256+3*32(%rbx)
+	vmovdqa	%ymm8, 1*256+0*32(%rbx)
+	vmovdqa	%ymm9, 1*256+1*32(%rbx)
+	vmovdqa	%ymm10, 1*256+2*32(%rbx)
+	vmovdqa	%ymm11, 1*256+3*32(%rbx)
+	vmovdqa	%ymm12, 2*256+0*32(%rbx)
+	vmovdqa	%ymm13, 2*256+1*32(%rbx)
+	vmovdqa	%ymm14, 2*256+2*32(%rbx)
+	vmovdqa	%ymm15, 2*256+3*32(%rbx)
+	
+	salsa8_core_6way_avx2
+	vpaddd	0*256+0*32(%rbx), %ymm0, %ymm0
+	vpaddd	0*256+1*32(%rbx), %ymm1, %ymm1
+	vpaddd	0*256+2*32(%rbx), %ymm2, %ymm2
+	vpaddd	0*256+3*32(%rbx), %ymm3, %ymm3
+	vpaddd	1*256+0*32(%rbx), %ymm8, %ymm8
+	vpaddd	1*256+1*32(%rbx), %ymm9, %ymm9
+	vpaddd	1*256+2*32(%rbx), %ymm10, %ymm10
+	vpaddd	1*256+3*32(%rbx), %ymm11, %ymm11
+	vpaddd	2*256+0*32(%rbx), %ymm12, %ymm12
+	vpaddd	2*256+1*32(%rbx), %ymm13, %ymm13
+	vpaddd	2*256+2*32(%rbx), %ymm14, %ymm14
+	vpaddd	2*256+3*32(%rbx), %ymm15, %ymm15
+	vmovdqa	%ymm0, 0*256+0*32(%rsp)
+	vmovdqa	%ymm1, 0*256+1*32(%rsp)
+	vmovdqa	%ymm2, 0*256+2*32(%rsp)
+	vmovdqa	%ymm3, 0*256+3*32(%rsp)
+	vmovdqa	%ymm8, 1*256+0*32(%rsp)
+	vmovdqa	%ymm9, 1*256+1*32(%rsp)
+	vmovdqa	%ymm10, 1*256+2*32(%rsp)
+	vmovdqa	%ymm11, 1*256+3*32(%rsp)
+	vmovdqa	%ymm12, 2*256+0*32(%rsp)
+	vmovdqa	%ymm13, 2*256+1*32(%rsp)
+	vmovdqa	%ymm14, 2*256+2*32(%rsp)
+	vmovdqa	%ymm15, 2*256+3*32(%rsp)
+	
+	vpxor	0*256+4*32(%rbx), %ymm0, %ymm0
+	vpxor	0*256+5*32(%rbx), %ymm1, %ymm1
+	vpxor	0*256+6*32(%rbx), %ymm2, %ymm2
+	vpxor	0*256+7*32(%rbx), %ymm3, %ymm3
+	vpxor	1*256+4*32(%rbx), %ymm8, %ymm8
+	vpxor	1*256+5*32(%rbx), %ymm9, %ymm9
+	vpxor	1*256+6*32(%rbx), %ymm10, %ymm10
+	vpxor	1*256+7*32(%rbx), %ymm11, %ymm11
+	vpxor	2*256+4*32(%rbx), %ymm12, %ymm12
+	vpxor	2*256+5*32(%rbx), %ymm13, %ymm13
+	vpxor	2*256+6*32(%rbx), %ymm14, %ymm14
+	vpxor	2*256+7*32(%rbx), %ymm15, %ymm15
+	vmovdqa	%ymm0, 0*256+4*32(%rsp)
+	vmovdqa	%ymm1, 0*256+5*32(%rsp)
+	vmovdqa	%ymm2, 0*256+6*32(%rsp)
+	vmovdqa	%ymm3, 0*256+7*32(%rsp)
+	vmovdqa	%ymm8, 1*256+4*32(%rsp)
+	vmovdqa	%ymm9, 1*256+5*32(%rsp)
+	vmovdqa	%ymm10, 1*256+6*32(%rsp)
+	vmovdqa	%ymm11, 1*256+7*32(%rsp)
+	vmovdqa	%ymm12, 2*256+4*32(%rsp)
+	vmovdqa	%ymm13, 2*256+5*32(%rsp)
+	vmovdqa	%ymm14, 2*256+6*32(%rsp)
+	vmovdqa	%ymm15, 2*256+7*32(%rsp)
+	salsa8_core_6way_avx2
+	vpaddd	0*256+4*32(%rsp), %ymm0, %ymm0
+	vpaddd	0*256+5*32(%rsp), %ymm1, %ymm1
+	vpaddd	0*256+6*32(%rsp), %ymm2, %ymm2
+	vpaddd	0*256+7*32(%rsp), %ymm3, %ymm3
+	vpaddd	1*256+4*32(%rsp), %ymm8, %ymm8
+	vpaddd	1*256+5*32(%rsp), %ymm9, %ymm9
+	vpaddd	1*256+6*32(%rsp), %ymm10, %ymm10
+	vpaddd	1*256+7*32(%rsp), %ymm11, %ymm11
+	vpaddd	2*256+4*32(%rsp), %ymm12, %ymm12
+	vpaddd	2*256+5*32(%rsp), %ymm13, %ymm13
+	vpaddd	2*256+6*32(%rsp), %ymm14, %ymm14
+	vpaddd	2*256+7*32(%rsp), %ymm15, %ymm15
+	
+	addq	$6*128, %rbx
+	cmpq	%rax, %rbx
+	jne scrypt_core_6way_avx2_loop1
+	
+	vmovdqa	%ymm0, 0*256+4*32(%rsp)
+	vmovdqa	%ymm1, 0*256+5*32(%rsp)
+	vmovdqa	%ymm2, 0*256+6*32(%rsp)
+	vmovdqa	%ymm3, 0*256+7*32(%rsp)
+	vmovdqa	%ymm8, 1*256+4*32(%rsp)
+	vmovdqa	%ymm9, 1*256+5*32(%rsp)
+	vmovdqa	%ymm10, 1*256+6*32(%rsp)
+	vmovdqa	%ymm11, 1*256+7*32(%rsp)
+	vmovdqa	%ymm12, 2*256+4*32(%rsp)
+	vmovdqa	%ymm13, 2*256+5*32(%rsp)
+	vmovdqa	%ymm14, 2*256+6*32(%rsp)
+	vmovdqa	%ymm15, 2*256+7*32(%rsp)
+	
+	movq	%r8, %rcx
+	leaq	-1(%r8), %r11
+scrypt_core_6way_avx2_loop2:
+	vmovd	%xmm0, %ebp
+	vmovd	%xmm8, %ebx
+	vmovd	%xmm12, %eax
+	vextracti128	$1, %ymm0, %xmm4
+	vextracti128	$1, %ymm8, %xmm5
+	vextracti128	$1, %ymm12, %xmm6
+	vmovd	%xmm4, %r8d
+	vmovd	%xmm5, %r9d
+	vmovd	%xmm6, %r10d
+	vpxor	0*256+0*32(%rsp), %ymm0, %ymm0
+	vpxor	0*256+1*32(%rsp), %ymm1, %ymm1
+	vpxor	0*256+2*32(%rsp), %ymm2, %ymm2
+	vpxor	0*256+3*32(%rsp), %ymm3, %ymm3
+	vpxor	1*256+0*32(%rsp), %ymm8, %ymm8
+	vpxor	1*256+1*32(%rsp), %ymm9, %ymm9
+	vpxor	1*256+2*32(%rsp), %ymm10, %ymm10
+	vpxor	1*256+3*32(%rsp), %ymm11, %ymm11
+	vpxor	2*256+0*32(%rsp), %ymm12, %ymm12
+	vpxor	2*256+1*32(%rsp), %ymm13, %ymm13
+	vpxor	2*256+2*32(%rsp), %ymm14, %ymm14
+	vpxor	2*256+3*32(%rsp), %ymm15, %ymm15
+	andl	%r11d, %ebp
+	leaq	0(%rbp, %rbp, 2), %rbp
+	shll	$8, %ebp
+	andl	%r11d, %ebx
+	leaq	1(%rbx, %rbx, 2), %rbx
+	shll	$8, %ebx
+	andl	%r11d, %eax
+	leaq	2(%rax, %rax, 2), %rax
+	shll	$8, %eax
+	andl	%r11d, %r8d
+	leaq	0(%r8, %r8, 2), %r8
+	shll	$8, %r8d
+	andl	%r11d, %r9d
+	leaq	1(%r9, %r9, 2), %r9
+	shll	$8, %r9d
+	andl	%r11d, %r10d
+	leaq	2(%r10, %r10, 2), %r10
+	shll	$8, %r10d
+	vmovdqa	0*32(%rsi, %rbp), %xmm4
+	vinserti128	$1, 0*32+16(%rsi, %r8), %ymm4, %ymm4
+	vmovdqa	1*32(%rsi, %rbp), %xmm5
+	vinserti128	$1, 1*32+16(%rsi, %r8), %ymm5, %ymm5
+	vmovdqa	2*32(%rsi, %rbp), %xmm6
+	vinserti128	$1, 2*32+16(%rsi, %r8), %ymm6, %ymm6
+	vmovdqa	3*32(%rsi, %rbp), %xmm7
+	vinserti128	$1, 3*32+16(%rsi, %r8), %ymm7, %ymm7
+	vpxor	%ymm4, %ymm0, %ymm0
+	vpxor	%ymm5, %ymm1, %ymm1
+	vpxor	%ymm6, %ymm2, %ymm2
+	vpxor	%ymm7, %ymm3, %ymm3
+	vmovdqa	0*32(%rsi, %rbx), %xmm4
+	vinserti128	$1, 0*32+16(%rsi, %r9), %ymm4, %ymm4
+	vmovdqa	1*32(%rsi, %rbx), %xmm5
+	vinserti128	$1, 1*32+16(%rsi, %r9), %ymm5, %ymm5
+	vmovdqa	2*32(%rsi, %rbx), %xmm6
+	vinserti128	$1, 2*32+16(%rsi, %r9), %ymm6, %ymm6
+	vmovdqa	3*32(%rsi, %rbx), %xmm7
+	vinserti128	$1, 3*32+16(%rsi, %r9), %ymm7, %ymm7
+	vpxor	%ymm4, %ymm8, %ymm8
+	vpxor	%ymm5, %ymm9, %ymm9
+	vpxor	%ymm6, %ymm10, %ymm10
+	vpxor	%ymm7, %ymm11, %ymm11
+	vmovdqa	0*32(%rsi, %rax), %xmm4
+	vinserti128	$1, 0*32+16(%rsi, %r10), %ymm4, %ymm4
+	vmovdqa	1*32(%rsi, %rax), %xmm5
+	vinserti128	$1, 1*32+16(%rsi, %r10), %ymm5, %ymm5
+	vmovdqa	2*32(%rsi, %rax), %xmm6
+	vinserti128	$1, 2*32+16(%rsi, %r10), %ymm6, %ymm6
+	vmovdqa	3*32(%rsi, %rax), %xmm7
+	vinserti128	$1, 3*32+16(%rsi, %r10), %ymm7, %ymm7
+	vpxor	%ymm4, %ymm12, %ymm12
+	vpxor	%ymm5, %ymm13, %ymm13
+	vpxor	%ymm6, %ymm14, %ymm14
+	vpxor	%ymm7, %ymm15, %ymm15
+	
+	vmovdqa	%ymm0, 0*256+0*32(%rsp)
+	vmovdqa	%ymm1, 0*256+1*32(%rsp)
+	vmovdqa	%ymm2, 0*256+2*32(%rsp)
+	vmovdqa	%ymm3, 0*256+3*32(%rsp)
+	vmovdqa	%ymm8, 1*256+0*32(%rsp)
+	vmovdqa	%ymm9, 1*256+1*32(%rsp)
+	vmovdqa	%ymm10, 1*256+2*32(%rsp)
+	vmovdqa	%ymm11, 1*256+3*32(%rsp)
+	vmovdqa	%ymm12, 2*256+0*32(%rsp)
+	vmovdqa	%ymm13, 2*256+1*32(%rsp)
+	vmovdqa	%ymm14, 2*256+2*32(%rsp)
+	vmovdqa	%ymm15, 2*256+3*32(%rsp)
+	salsa8_core_6way_avx2
+	vpaddd	0*256+0*32(%rsp), %ymm0, %ymm0
+	vpaddd	0*256+1*32(%rsp), %ymm1, %ymm1
+	vpaddd	0*256+2*32(%rsp), %ymm2, %ymm2
+	vpaddd	0*256+3*32(%rsp), %ymm3, %ymm3
+	vpaddd	1*256+0*32(%rsp), %ymm8, %ymm8
+	vpaddd	1*256+1*32(%rsp), %ymm9, %ymm9
+	vpaddd	1*256+2*32(%rsp), %ymm10, %ymm10
+	vpaddd	1*256+3*32(%rsp), %ymm11, %ymm11
+	vpaddd	2*256+0*32(%rsp), %ymm12, %ymm12
+	vpaddd	2*256+1*32(%rsp), %ymm13, %ymm13
+	vpaddd	2*256+2*32(%rsp), %ymm14, %ymm14
+	vpaddd	2*256+3*32(%rsp), %ymm15, %ymm15
+	vmovdqa	%ymm0, 0*256+0*32(%rsp)
+	vmovdqa	%ymm1, 0*256+1*32(%rsp)
+	vmovdqa	%ymm2, 0*256+2*32(%rsp)
+	vmovdqa	%ymm3, 0*256+3*32(%rsp)
+	vmovdqa	%ymm8, 1*256+0*32(%rsp)
+	vmovdqa	%ymm9, 1*256+1*32(%rsp)
+	vmovdqa	%ymm10, 1*256+2*32(%rsp)
+	vmovdqa	%ymm11, 1*256+3*32(%rsp)
+	vmovdqa	%ymm12, 2*256+0*32(%rsp)
+	vmovdqa	%ymm13, 2*256+1*32(%rsp)
+	vmovdqa	%ymm14, 2*256+2*32(%rsp)
+	vmovdqa	%ymm15, 2*256+3*32(%rsp)
+	
+	vmovdqa	4*32(%rsi, %rbp), %xmm4
+	vinserti128	$1, 4*32+16(%rsi, %r8), %ymm4, %ymm4
+	vmovdqa	5*32(%rsi, %rbp), %xmm5
+	vinserti128	$1, 5*32+16(%rsi, %r8), %ymm5, %ymm5
+	vmovdqa	6*32(%rsi, %rbp), %xmm6
+	vinserti128	$1, 6*32+16(%rsi, %r8), %ymm6, %ymm6
+	vmovdqa	7*32(%rsi, %rbp), %xmm7
+	vinserti128	$1, 7*32+16(%rsi, %r8), %ymm7, %ymm7
+	vpxor	%ymm4, %ymm0, %ymm0
+	vpxor	%ymm5, %ymm1, %ymm1
+	vpxor	%ymm6, %ymm2, %ymm2
+	vpxor	%ymm7, %ymm3, %ymm3
+	vmovdqa	4*32(%rsi, %rbx), %xmm4
+	vinserti128	$1, 4*32+16(%rsi, %r9), %ymm4, %ymm4
+	vmovdqa	5*32(%rsi, %rbx), %xmm5
+	vinserti128	$1, 5*32+16(%rsi, %r9), %ymm5, %ymm5
+	vmovdqa	6*32(%rsi, %rbx), %xmm6
+	vinserti128	$1, 6*32+16(%rsi, %r9), %ymm6, %ymm6
+	vmovdqa	7*32(%rsi, %rbx), %xmm7
+	vinserti128	$1, 7*32+16(%rsi, %r9), %ymm7, %ymm7
+	vpxor	%ymm4, %ymm8, %ymm8
+	vpxor	%ymm5, %ymm9, %ymm9
+	vpxor	%ymm6, %ymm10, %ymm10
+	vpxor	%ymm7, %ymm11, %ymm11
+	vmovdqa	4*32(%rsi, %rax), %xmm4
+	vinserti128	$1, 4*32+16(%rsi, %r10), %ymm4, %ymm4
+	vmovdqa	5*32(%rsi, %rax), %xmm5
+	vinserti128	$1, 5*32+16(%rsi, %r10), %ymm5, %ymm5
+	vmovdqa	6*32(%rsi, %rax), %xmm6
+	vinserti128	$1, 6*32+16(%rsi, %r10), %ymm6, %ymm6
+	vmovdqa	7*32(%rsi, %rax), %xmm7
+	vinserti128	$1, 7*32+16(%rsi, %r10), %ymm7, %ymm7
+	vpxor	%ymm4, %ymm12, %ymm12
+	vpxor	%ymm5, %ymm13, %ymm13
+	vpxor	%ymm6, %ymm14, %ymm14
+	vpxor	%ymm7, %ymm15, %ymm15
+	vpxor	0*256+4*32(%rsp), %ymm0, %ymm0
+	vpxor	0*256+5*32(%rsp), %ymm1, %ymm1
+	vpxor	0*256+6*32(%rsp), %ymm2, %ymm2
+	vpxor	0*256+7*32(%rsp), %ymm3, %ymm3
+	vpxor	1*256+4*32(%rsp), %ymm8, %ymm8
+	vpxor	1*256+5*32(%rsp), %ymm9, %ymm9
+	vpxor	1*256+6*32(%rsp), %ymm10, %ymm10
+	vpxor	1*256+7*32(%rsp), %ymm11, %ymm11
+	vpxor	2*256+4*32(%rsp), %ymm12, %ymm12
+	vpxor	2*256+5*32(%rsp), %ymm13, %ymm13
+	vpxor	2*256+6*32(%rsp), %ymm14, %ymm14
+	vpxor	2*256+7*32(%rsp), %ymm15, %ymm15
+	vmovdqa	%ymm0, 0*256+4*32(%rsp)
+	vmovdqa	%ymm1, 0*256+5*32(%rsp)
+	vmovdqa	%ymm2, 0*256+6*32(%rsp)
+	vmovdqa	%ymm3, 0*256+7*32(%rsp)
+	vmovdqa	%ymm8, 1*256+4*32(%rsp)
+	vmovdqa	%ymm9, 1*256+5*32(%rsp)
+	vmovdqa	%ymm10, 1*256+6*32(%rsp)
+	vmovdqa	%ymm11, 1*256+7*32(%rsp)
+	vmovdqa	%ymm12, 2*256+4*32(%rsp)
+	vmovdqa	%ymm13, 2*256+5*32(%rsp)
+	vmovdqa	%ymm14, 2*256+6*32(%rsp)
+	vmovdqa	%ymm15, 2*256+7*32(%rsp)
+	salsa8_core_6way_avx2
+	vpaddd	0*256+4*32(%rsp), %ymm0, %ymm0
+	vpaddd	0*256+5*32(%rsp), %ymm1, %ymm1
+	vpaddd	0*256+6*32(%rsp), %ymm2, %ymm2
+	vpaddd	0*256+7*32(%rsp), %ymm3, %ymm3
+	vpaddd	1*256+4*32(%rsp), %ymm8, %ymm8
+	vpaddd	1*256+5*32(%rsp), %ymm9, %ymm9
+	vpaddd	1*256+6*32(%rsp), %ymm10, %ymm10
+	vpaddd	1*256+7*32(%rsp), %ymm11, %ymm11
+	vpaddd	2*256+4*32(%rsp), %ymm12, %ymm12
+	vpaddd	2*256+5*32(%rsp), %ymm13, %ymm13
+	vpaddd	2*256+6*32(%rsp), %ymm14, %ymm14
+	vpaddd	2*256+7*32(%rsp), %ymm15, %ymm15
+	vmovdqa	%ymm0, 0*256+4*32(%rsp)
+	vmovdqa	%ymm1, 0*256+5*32(%rsp)
+	vmovdqa	%ymm2, 0*256+6*32(%rsp)
+	vmovdqa	%ymm3, 0*256+7*32(%rsp)
+	vmovdqa	%ymm8, 1*256+4*32(%rsp)
+	vmovdqa	%ymm9, 1*256+5*32(%rsp)
+	vmovdqa	%ymm10, 1*256+6*32(%rsp)
+	vmovdqa	%ymm11, 1*256+7*32(%rsp)
+	vmovdqa	%ymm12, 2*256+4*32(%rsp)
+	vmovdqa	%ymm13, 2*256+5*32(%rsp)
+	vmovdqa	%ymm14, 2*256+6*32(%rsp)
+	vmovdqa	%ymm15, 2*256+7*32(%rsp)
+	
+	subq	$1, %rcx
+	ja scrypt_core_6way_avx2_loop2
+	
+	scrypt_shuffle_unpack2 %rsp, 0*128, %rdi, 0*256+0
+	scrypt_shuffle_unpack2 %rsp, 1*128, %rdi, 0*256+64
+	scrypt_shuffle_unpack2 %rsp, 2*128, %rdi, 1*256+0
+	scrypt_shuffle_unpack2 %rsp, 3*128, %rdi, 1*256+64
+	scrypt_shuffle_unpack2 %rsp, 4*128, %rdi, 2*256+0
+	scrypt_shuffle_unpack2 %rsp, 5*128, %rdi, 2*256+64
+	
+	scrypt_core_6way_cleanup
+	ret
+
+#endif /* USE_AVX2 */
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-x86.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-x86.S
new file mode 100644
index 0000000..5ab7eda
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt-x86.S
@@ -0,0 +1,830 @@
+/*
+ * Copyright 2011-2012, 2014 pooler@litecoinpool.org
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(__linux__) && defined(__ELF__)
+	.section .note.GNU-stack,"",%progbits
+#endif
+
+#if defined(USE_ASM) && defined(__i386__)
+	
+.macro scrypt_shuffle src, so, dest, do
+	movl	\so+60(\src), %eax
+	movl	\so+44(\src), %ebx
+	movl	\so+28(\src), %ecx
+	movl	\so+12(\src), %edx
+	movl	%eax, \do+12(\dest)
+	movl	%ebx, \do+28(\dest)
+	movl	%ecx, \do+44(\dest)
+	movl	%edx, \do+60(\dest)
+	movl	\so+40(\src), %eax
+	movl	\so+8(\src), %ebx
+	movl	\so+48(\src), %ecx
+	movl	\so+16(\src), %edx
+	movl	%eax, \do+8(\dest)
+	movl	%ebx, \do+40(\dest)
+	movl	%ecx, \do+16(\dest)
+	movl	%edx, \do+48(\dest)
+	movl	\so+20(\src), %eax
+	movl	\so+4(\src), %ebx
+	movl	\so+52(\src), %ecx
+	movl	\so+36(\src), %edx
+	movl	%eax, \do+4(\dest)
+	movl	%ebx, \do+20(\dest)
+	movl	%ecx, \do+36(\dest)
+	movl	%edx, \do+52(\dest)
+	movl	\so+0(\src), %eax
+	movl	\so+24(\src), %ebx
+	movl	\so+32(\src), %ecx
+	movl	\so+56(\src), %edx
+	movl	%eax, \do+0(\dest)
+	movl	%ebx, \do+24(\dest)
+	movl	%ecx, \do+32(\dest)
+	movl	%edx, \do+56(\dest)
+.endm
+
+.macro salsa8_core_gen_quadround
+	movl	52(%esp), %ecx
+	movl	4(%esp), %edx
+	movl	20(%esp), %ebx
+	movl	8(%esp), %esi
+	leal	(%ecx, %edx), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 4(%esp)
+	movl	36(%esp), %edi
+	leal	(%edx, %ebx), %ebp
+	roll	$9, %ebp
+	xorl	%ebp, %edi
+	movl	24(%esp), %ebp
+	movl	%edi, 8(%esp)
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	40(%esp), %ebx
+	movl	%ecx, 20(%esp)
+	addl	%edi, %ecx
+	roll	$18, %ecx
+	leal	(%esi, %ebp), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 24(%esp)
+	movl	56(%esp), %edi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %edi
+	movl	%edi, 36(%esp)
+	movl	28(%esp), %ecx
+	movl	%edx, 28(%esp)
+	movl	44(%esp), %edx
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %esi
+	movl	60(%esp), %ebx
+	movl	%esi, 40(%esp)
+	addl	%edi, %esi
+	roll	$18, %esi
+	leal	(%ecx, %edx), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 44(%esp)
+	movl	12(%esp), %edi
+	xorl	%esi, %ebp
+	leal	(%edx, %ebx), %esi
+	roll	$9, %esi
+	xorl	%esi, %edi
+	movl	%edi, 12(%esp)
+	movl	48(%esp), %esi
+	movl	%ebp, 48(%esp)
+	movl	64(%esp), %ebp
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	16(%esp), %ebx
+	movl	%ecx, 16(%esp)
+	addl	%edi, %ecx
+	roll	$18, %ecx
+	leal	(%esi, %ebp), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	32(%esp), %edi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %edi
+	movl	%edi, 32(%esp)
+	movl	%ebx, %ecx
+	movl	%edx, 52(%esp)
+	movl	28(%esp), %edx
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %esi
+	movl	40(%esp), %ebx
+	movl	%esi, 28(%esp)
+	addl	%edi, %esi
+	roll	$18, %esi
+	leal	(%ecx, %edx), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 40(%esp)
+	movl	12(%esp), %edi
+	xorl	%esi, %ebp
+	leal	(%edx, %ebx), %esi
+	roll	$9, %esi
+	xorl	%esi, %edi
+	movl	%edi, 12(%esp)
+	movl	4(%esp), %esi
+	movl	%ebp, 4(%esp)
+	movl	48(%esp), %ebp
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	16(%esp), %ebx
+	movl	%ecx, 16(%esp)
+	addl	%edi, %ecx
+	roll	$18, %ecx
+	leal	(%esi, %ebp), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 48(%esp)
+	movl	32(%esp), %edi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %edi
+	movl	%edi, 32(%esp)
+	movl	24(%esp), %ecx
+	movl	%edx, 24(%esp)
+	movl	52(%esp), %edx
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %esi
+	movl	28(%esp), %ebx
+	movl	%esi, 28(%esp)
+	addl	%edi, %esi
+	roll	$18, %esi
+	leal	(%ecx, %edx), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 52(%esp)
+	movl	8(%esp), %edi
+	xorl	%esi, %ebp
+	leal	(%edx, %ebx), %esi
+	roll	$9, %esi
+	xorl	%esi, %edi
+	movl	%edi, 8(%esp)
+	movl	44(%esp), %esi
+	movl	%ebp, 44(%esp)
+	movl	4(%esp), %ebp
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	20(%esp), %ebx
+	movl	%ecx, 4(%esp)
+	addl	%edi, %ecx
+	roll	$18, %ecx
+	leal	(%esi, %ebp), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	36(%esp), %edi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %edi
+	movl	%edi, 20(%esp)
+	movl	%ebx, %ecx
+	movl	%edx, 36(%esp)
+	movl	24(%esp), %edx
+	addl	%edi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %esi
+	movl	28(%esp), %ebx
+	movl	%esi, 24(%esp)
+	addl	%edi, %esi
+	roll	$18, %esi
+	leal	(%ecx, %edx), %edi
+	roll	$7, %edi
+	xorl	%edi, %ebx
+	movl	%ebx, 28(%esp)
+	xorl	%esi, %ebp
+	movl	8(%esp), %esi
+	leal	(%edx, %ebx), %edi
+	roll	$9, %edi
+	xorl	%edi, %esi
+	movl	40(%esp), %edi
+	movl	%ebp, 8(%esp)
+	movl	44(%esp), %ebp
+	movl	%esi, 40(%esp)
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	4(%esp), %ebx
+	movl	%ecx, 44(%esp)
+	addl	%esi, %ecx
+	roll	$18, %ecx
+	leal	(%edi, %ebp), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	%ebx, 4(%esp)
+	movl	20(%esp), %esi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %esi
+	movl	%esi, 56(%esp)
+	movl	48(%esp), %ecx
+	movl	%edx, 20(%esp)
+	movl	36(%esp), %edx
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %edi
+	movl	24(%esp), %ebx
+	movl	%edi, 24(%esp)
+	addl	%esi, %edi
+	roll	$18, %edi
+	leal	(%ecx, %edx), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	%ebx, 60(%esp)
+	movl	12(%esp), %esi
+	xorl	%edi, %ebp
+	leal	(%edx, %ebx), %edi
+	roll	$9, %edi
+	xorl	%edi, %esi
+	movl	%esi, 12(%esp)
+	movl	52(%esp), %edi
+	movl	%ebp, 36(%esp)
+	movl	8(%esp), %ebp
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	16(%esp), %ebx
+	movl	%ecx, 16(%esp)
+	addl	%esi, %ecx
+	roll	$18, %ecx
+	leal	(%edi, %ebp), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	32(%esp), %esi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %esi
+	movl	%esi, 32(%esp)
+	movl	%ebx, %ecx
+	movl	%edx, 48(%esp)
+	movl	20(%esp), %edx
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %edi
+	movl	24(%esp), %ebx
+	movl	%edi, 20(%esp)
+	addl	%esi, %edi
+	roll	$18, %edi
+	leal	(%ecx, %edx), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	%ebx, 8(%esp)
+	movl	12(%esp), %esi
+	xorl	%edi, %ebp
+	leal	(%edx, %ebx), %edi
+	roll	$9, %edi
+	xorl	%edi, %esi
+	movl	%esi, 12(%esp)
+	movl	28(%esp), %edi
+	movl	%ebp, 52(%esp)
+	movl	36(%esp), %ebp
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	16(%esp), %ebx
+	movl	%ecx, 16(%esp)
+	addl	%esi, %ecx
+	roll	$18, %ecx
+	leal	(%edi, %ebp), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	%ebx, 28(%esp)
+	movl	32(%esp), %esi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %esi
+	movl	%esi, 32(%esp)
+	movl	4(%esp), %ecx
+	movl	%edx, 4(%esp)
+	movl	48(%esp), %edx
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %edi
+	movl	20(%esp), %ebx
+	movl	%edi, 20(%esp)
+	addl	%esi, %edi
+	roll	$18, %edi
+	leal	(%ecx, %edx), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	%ebx, 48(%esp)
+	movl	40(%esp), %esi
+	xorl	%edi, %ebp
+	leal	(%edx, %ebx), %edi
+	roll	$9, %edi
+	xorl	%edi, %esi
+	movl	%esi, 36(%esp)
+	movl	60(%esp), %edi
+	movl	%ebp, 24(%esp)
+	movl	52(%esp), %ebp
+	addl	%esi, %ebx
+	roll	$13, %ebx
+	xorl	%ebx, %ecx
+	movl	44(%esp), %ebx
+	movl	%ecx, 40(%esp)
+	addl	%esi, %ecx
+	roll	$18, %ecx
+	leal	(%edi, %ebp), %esi
+	roll	$7, %esi
+	xorl	%esi, %ebx
+	movl	%ebx, 52(%esp)
+	movl	56(%esp), %esi
+	xorl	%ecx, %edx
+	leal	(%ebp, %ebx), %ecx
+	roll	$9, %ecx
+	xorl	%ecx, %esi
+	movl	%esi, 56(%esp)
+	addl	%esi, %ebx
+	movl	%edx, 44(%esp)
+	roll	$13, %ebx
+	xorl	%ebx, %edi
+	movl	%edi, 60(%esp)
+	addl	%esi, %edi
+	roll	$18, %edi
+	xorl	%edi, %ebp
+	movl	%ebp, 64(%esp)
+.endm
+
+	.text
+	.p2align 5
+salsa8_core_gen:
+	salsa8_core_gen_quadround
+	salsa8_core_gen_quadround
+	ret
+	
+	
+	.text
+	.p2align 5
+	.globl scrypt_core
+	.globl _scrypt_core
+scrypt_core:
+_scrypt_core:
+	pushl	%ebx
+	pushl	%ebp
+	pushl	%edi
+	pushl	%esi
+	
+	/* Check for SSE2 availability */
+	movl	$1, %eax
+	cpuid
+	andl	$0x04000000, %edx
+	jnz scrypt_core_sse2
+	
+scrypt_core_gen:
+	movl	20(%esp), %edi
+	movl	24(%esp), %esi
+	movl	28(%esp), %ecx
+	subl	$72, %esp
+	
+.macro scrypt_core_macro1a p, q
+	movl	\p(%edi), %eax
+	movl	\q(%edi), %edx
+	movl	%eax, \p(%esi)
+	movl	%edx, \q(%esi)
+	xorl	%edx, %eax
+	movl	%eax, \p(%edi)
+	movl	%eax, \p(%esp)
+.endm
+	
+.macro scrypt_core_macro1b p, q
+	movl	\p(%edi), %eax
+	xorl	\p(%esi, %edx), %eax
+	movl	\q(%edi), %ebx
+	xorl	\q(%esi, %edx), %ebx
+	movl	%ebx, \q(%edi)
+	xorl	%ebx, %eax
+	movl	%eax, \p(%edi)
+	movl	%eax, \p(%esp)
+.endm
+	
+.macro scrypt_core_macro2 p, q
+	movl	\p(%esp), %eax
+	addl	\p(%edi), %eax
+	movl	%eax, \p(%edi)
+	xorl	\q(%edi), %eax
+	movl	%eax, \q(%edi)
+	movl	%eax, \p(%esp)
+.endm
+	
+.macro scrypt_core_macro3 p, q
+	movl	\p(%esp), %eax
+	addl	\q(%edi), %eax
+	movl	%eax, \q(%edi)
+.endm
+	
+	shll	$7, %ecx
+	addl	%esi, %ecx
+scrypt_core_gen_loop1:
+	movl	%esi, 64(%esp)
+	movl	%ecx, 68(%esp)
+	
+	scrypt_core_macro1a	0, 64
+	scrypt_core_macro1a	4, 68
+	scrypt_core_macro1a	8, 72
+	scrypt_core_macro1a	12, 76
+	scrypt_core_macro1a	16, 80
+	scrypt_core_macro1a	20, 84
+	scrypt_core_macro1a	24, 88
+	scrypt_core_macro1a	28, 92
+	scrypt_core_macro1a	32, 96
+	scrypt_core_macro1a	36, 100
+	scrypt_core_macro1a	40, 104
+	scrypt_core_macro1a	44, 108
+	scrypt_core_macro1a	48, 112
+	scrypt_core_macro1a	52, 116
+	scrypt_core_macro1a	56, 120
+	scrypt_core_macro1a	60, 124
+	
+	call salsa8_core_gen
+	
+	movl	92(%esp), %edi
+	scrypt_core_macro2	0, 64
+	scrypt_core_macro2	4, 68
+	scrypt_core_macro2	8, 72
+	scrypt_core_macro2	12, 76
+	scrypt_core_macro2	16, 80
+	scrypt_core_macro2	20, 84
+	scrypt_core_macro2	24, 88
+	scrypt_core_macro2	28, 92
+	scrypt_core_macro2	32, 96
+	scrypt_core_macro2	36, 100
+	scrypt_core_macro2	40, 104
+	scrypt_core_macro2	44, 108
+	scrypt_core_macro2	48, 112
+	scrypt_core_macro2	52, 116
+	scrypt_core_macro2	56, 120
+	scrypt_core_macro2	60, 124
+	
+	call salsa8_core_gen
+	
+	movl	92(%esp), %edi
+	scrypt_core_macro3	0, 64
+	scrypt_core_macro3	4, 68
+	scrypt_core_macro3	8, 72
+	scrypt_core_macro3	12, 76
+	scrypt_core_macro3	16, 80
+	scrypt_core_macro3	20, 84
+	scrypt_core_macro3	24, 88
+	scrypt_core_macro3	28, 92
+	scrypt_core_macro3	32, 96
+	scrypt_core_macro3	36, 100
+	scrypt_core_macro3	40, 104
+	scrypt_core_macro3	44, 108
+	scrypt_core_macro3	48, 112
+	scrypt_core_macro3	52, 116
+	scrypt_core_macro3	56, 120
+	scrypt_core_macro3	60, 124
+	
+	movl	64(%esp), %esi
+	movl	68(%esp), %ecx
+	addl	$128, %esi
+	cmpl	%ecx, %esi
+	jne scrypt_core_gen_loop1
+
+	movl	96(%esp), %esi
+	movl	100(%esp), %ecx
+	movl	%ecx, %eax
+	subl	$1, %eax
+	movl	%eax, 100(%esp)
+scrypt_core_gen_loop2:
+	movl	%ecx, 68(%esp)
+	
+	movl	64(%edi), %edx
+	andl	100(%esp), %edx
+	shll	$7, %edx
+	
+	scrypt_core_macro1b	0, 64
+	scrypt_core_macro1b	4, 68
+	scrypt_core_macro1b	8, 72
+	scrypt_core_macro1b	12, 76
+	scrypt_core_macro1b	16, 80
+	scrypt_core_macro1b	20, 84
+	scrypt_core_macro1b	24, 88
+	scrypt_core_macro1b	28, 92
+	scrypt_core_macro1b	32, 96
+	scrypt_core_macro1b	36, 100
+	scrypt_core_macro1b	40, 104
+	scrypt_core_macro1b	44, 108
+	scrypt_core_macro1b	48, 112
+	scrypt_core_macro1b	52, 116
+	scrypt_core_macro1b	56, 120
+	scrypt_core_macro1b	60, 124
+	
+	call salsa8_core_gen
+	
+	movl	92(%esp), %edi
+	scrypt_core_macro2	0, 64
+	scrypt_core_macro2	4, 68
+	scrypt_core_macro2	8, 72
+	scrypt_core_macro2	12, 76
+	scrypt_core_macro2	16, 80
+	scrypt_core_macro2	20, 84
+	scrypt_core_macro2	24, 88
+	scrypt_core_macro2	28, 92
+	scrypt_core_macro2	32, 96
+	scrypt_core_macro2	36, 100
+	scrypt_core_macro2	40, 104
+	scrypt_core_macro2	44, 108
+	scrypt_core_macro2	48, 112
+	scrypt_core_macro2	52, 116
+	scrypt_core_macro2	56, 120
+	scrypt_core_macro2	60, 124
+	
+	call salsa8_core_gen
+	
+	movl	92(%esp), %edi
+	movl	96(%esp), %esi
+	scrypt_core_macro3	0, 64
+	scrypt_core_macro3	4, 68
+	scrypt_core_macro3	8, 72
+	scrypt_core_macro3	12, 76
+	scrypt_core_macro3	16, 80
+	scrypt_core_macro3	20, 84
+	scrypt_core_macro3	24, 88
+	scrypt_core_macro3	28, 92
+	scrypt_core_macro3	32, 96
+	scrypt_core_macro3	36, 100
+	scrypt_core_macro3	40, 104
+	scrypt_core_macro3	44, 108
+	scrypt_core_macro3	48, 112
+	scrypt_core_macro3	52, 116
+	scrypt_core_macro3	56, 120
+	scrypt_core_macro3	60, 124
+	
+	movl	68(%esp), %ecx
+	subl	$1, %ecx
+	ja scrypt_core_gen_loop2
+	
+	addl	$72, %esp
+	popl	%esi
+	popl	%edi
+	popl	%ebp
+	popl	%ebx
+	ret
+
+
+.macro salsa8_core_sse2_doubleround
+	movdqa	%xmm1, %xmm4
+	paddd	%xmm0, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$7, %xmm4
+	psrld	$25, %xmm5
+	pxor	%xmm4, %xmm3
+	movdqa	%xmm0, %xmm4
+	pxor	%xmm5, %xmm3
+	
+	paddd	%xmm3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$9, %xmm4
+	psrld	$23, %xmm5
+	pxor	%xmm4, %xmm2
+	movdqa	%xmm3, %xmm4
+	pxor	%xmm5, %xmm2
+	pshufd	$0x93, %xmm3, %xmm3
+	
+	paddd	%xmm2, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$13, %xmm4
+	psrld	$19, %xmm5
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm2, %xmm4
+	pxor	%xmm5, %xmm1
+	pshufd	$0x4e, %xmm2, %xmm2
+	
+	paddd	%xmm1, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$18, %xmm4
+	psrld	$14, %xmm5
+	pxor	%xmm4, %xmm0
+	movdqa	%xmm3, %xmm4
+	pxor	%xmm5, %xmm0
+	pshufd	$0x39, %xmm1, %xmm1
+	
+	paddd	%xmm0, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$7, %xmm4
+	psrld	$25, %xmm5
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm0, %xmm4
+	pxor	%xmm5, %xmm1
+	
+	paddd	%xmm1, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$9, %xmm4
+	psrld	$23, %xmm5
+	pxor	%xmm4, %xmm2
+	movdqa	%xmm1, %xmm4
+	pxor	%xmm5, %xmm2
+	pshufd	$0x93, %xmm1, %xmm1
+	
+	paddd	%xmm2, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$13, %xmm4
+	psrld	$19, %xmm5
+	pxor	%xmm4, %xmm3
+	movdqa	%xmm2, %xmm4
+	pxor	%xmm5, %xmm3
+	pshufd	$0x4e, %xmm2, %xmm2
+	
+	paddd	%xmm3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$18, %xmm4
+	psrld	$14, %xmm5
+	pxor	%xmm4, %xmm0
+	pshufd	$0x39, %xmm3, %xmm3
+	pxor	%xmm5, %xmm0
+.endm
+
+.macro salsa8_core_sse2
+	salsa8_core_sse2_doubleround
+	salsa8_core_sse2_doubleround
+	salsa8_core_sse2_doubleround
+	salsa8_core_sse2_doubleround
+.endm
+	
+	.p2align 5
+scrypt_core_sse2:
+	movl	20(%esp), %edi
+	movl	24(%esp), %esi
+	movl	%esp, %ebp
+	subl	$128, %esp
+	andl	$-16, %esp
+	
+	scrypt_shuffle %edi, 0, %esp, 0
+	scrypt_shuffle %edi, 64, %esp, 64
+	
+	movdqa	96(%esp), %xmm6
+	movdqa	112(%esp), %xmm7
+	
+	movl	%esi, %edx
+	movl	28(%ebp), %ecx
+	shll	$7, %ecx
+	addl	%esi, %ecx
+scrypt_core_sse2_loop1:
+	movdqa	0(%esp), %xmm0
+	movdqa	16(%esp), %xmm1
+	movdqa	32(%esp), %xmm2
+	movdqa	48(%esp), %xmm3
+	movdqa	64(%esp), %xmm4
+	movdqa	80(%esp), %xmm5
+	pxor	%xmm4, %xmm0
+	pxor	%xmm5, %xmm1
+	movdqa	%xmm0, 0(%edx)
+	movdqa	%xmm1, 16(%edx)
+	pxor	%xmm6, %xmm2
+	pxor	%xmm7, %xmm3
+	movdqa	%xmm2, 32(%edx)
+	movdqa	%xmm3, 48(%edx)
+	movdqa	%xmm4, 64(%edx)
+	movdqa	%xmm5, 80(%edx)
+	movdqa	%xmm6, 96(%edx)
+	movdqa	%xmm7, 112(%edx)
+	
+	salsa8_core_sse2
+	paddd	0(%edx), %xmm0
+	paddd	16(%edx), %xmm1
+	paddd	32(%edx), %xmm2
+	paddd	48(%edx), %xmm3
+	movdqa	%xmm0, 0(%esp)
+	movdqa	%xmm1, 16(%esp)
+	movdqa	%xmm2, 32(%esp)
+	movdqa	%xmm3, 48(%esp)
+	
+	pxor	64(%esp), %xmm0
+	pxor	80(%esp), %xmm1
+	pxor	%xmm6, %xmm2
+	pxor	%xmm7, %xmm3
+	movdqa	%xmm0, 64(%esp)
+	movdqa	%xmm1, 80(%esp)
+	movdqa	%xmm2, %xmm6
+	movdqa	%xmm3, %xmm7
+	salsa8_core_sse2
+	paddd	64(%esp), %xmm0
+	paddd	80(%esp), %xmm1
+	paddd	%xmm2, %xmm6
+	paddd	%xmm3, %xmm7
+	movdqa	%xmm0, 64(%esp)
+	movdqa	%xmm1, 80(%esp)
+	
+	addl	$128, %edx
+	cmpl	%ecx, %edx
+	jne scrypt_core_sse2_loop1
+	
+	movdqa	64(%esp), %xmm4
+	movdqa	80(%esp), %xmm5
+	
+	movl	28(%ebp), %ecx
+	movl	%ecx, %eax
+	subl	$1, %eax
+scrypt_core_sse2_loop2:
+	movd	%xmm4, %edx
+	movdqa	0(%esp), %xmm0
+	movdqa	16(%esp), %xmm1
+	movdqa	32(%esp), %xmm2
+	movdqa	48(%esp), %xmm3
+	andl	%eax, %edx
+	shll	$7, %edx
+	pxor	0(%esi, %edx), %xmm0
+	pxor	16(%esi, %edx), %xmm1
+	pxor	32(%esi, %edx), %xmm2
+	pxor	48(%esi, %edx), %xmm3
+	
+	pxor	%xmm4, %xmm0
+	pxor	%xmm5, %xmm1
+	movdqa	%xmm0, 0(%esp)
+	movdqa	%xmm1, 16(%esp)
+	pxor	%xmm6, %xmm2
+	pxor	%xmm7, %xmm3
+	movdqa	%xmm2, 32(%esp)
+	movdqa	%xmm3, 48(%esp)
+	salsa8_core_sse2
+	paddd	0(%esp), %xmm0
+	paddd	16(%esp), %xmm1
+	paddd	32(%esp), %xmm2
+	paddd	48(%esp), %xmm3
+	movdqa	%xmm0, 0(%esp)
+	movdqa	%xmm1, 16(%esp)
+	movdqa	%xmm2, 32(%esp)
+	movdqa	%xmm3, 48(%esp)
+	
+	pxor	64(%esi, %edx), %xmm0
+	pxor	80(%esi, %edx), %xmm1
+	pxor	96(%esi, %edx), %xmm2
+	pxor	112(%esi, %edx), %xmm3
+	pxor	64(%esp), %xmm0
+	pxor	80(%esp), %xmm1
+	pxor	%xmm6, %xmm2
+	pxor	%xmm7, %xmm3
+	movdqa	%xmm0, 64(%esp)
+	movdqa	%xmm1, 80(%esp)
+	movdqa	%xmm2, %xmm6
+	movdqa	%xmm3, %xmm7
+	salsa8_core_sse2
+	paddd	64(%esp), %xmm0
+	paddd	80(%esp), %xmm1
+	paddd	%xmm2, %xmm6
+	paddd	%xmm3, %xmm7
+	movdqa	%xmm0, %xmm4
+	movdqa	%xmm1, %xmm5
+	movdqa	%xmm0, 64(%esp)
+	movdqa	%xmm1, 80(%esp)
+	
+	subl	$1, %ecx
+	ja scrypt_core_sse2_loop2
+	
+	movdqa	%xmm6, 96(%esp)
+	movdqa	%xmm7, 112(%esp)
+	
+	scrypt_shuffle %esp, 0, %edi, 0
+	scrypt_shuffle %esp, 64, %edi, 64
+	
+	movl	%ebp, %esp
+	popl	%esi
+	popl	%edi
+	popl	%ebp
+	popl	%ebx
+	ret
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt.c
new file mode 100644
index 0000000..b8dce87
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/scrypt.c
@@ -0,0 +1,762 @@
+/*
+ * Copyright 2009 Colin Percival, 2011 ArtForz, 2011-2014 pooler
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ */
+
+#include "cpuminer-config.h"
+#include "miner.h"
+
+#include <stdlib.h>
+#include <string.h>
+#include <inttypes.h>
+
+static const uint32_t keypad[12] = {
+	0x80000000, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x00000280
+};
+static const uint32_t innerpad[11] = {
+	0x80000000, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x000004a0
+};
+static const uint32_t outerpad[8] = {
+	0x80000000, 0, 0, 0, 0, 0, 0, 0x00000300
+};
+static const uint32_t finalblk[16] = {
+	0x00000001, 0x80000000, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x00000620
+};
+
+static inline void HMAC_SHA256_80_init(const uint32_t *key,
+	uint32_t *tstate, uint32_t *ostate)
+{
+	uint32_t ihash[8];
+	uint32_t pad[16];
+	int i;
+
+	/* tstate is assumed to contain the midstate of key */
+	memcpy(pad, key + 16, 16);
+	memcpy(pad + 4, keypad, 48);
+	sha256_transform(tstate, pad, 0);
+	memcpy(ihash, tstate, 32);
+
+	sha256_init(ostate);
+	for (i = 0; i < 8; i++)
+		pad[i] = ihash[i] ^ 0x5c5c5c5c;
+	for (; i < 16; i++)
+		pad[i] = 0x5c5c5c5c;
+	sha256_transform(ostate, pad, 0);
+
+	sha256_init(tstate);
+	for (i = 0; i < 8; i++)
+		pad[i] = ihash[i] ^ 0x36363636;
+	for (; i < 16; i++)
+		pad[i] = 0x36363636;
+	sha256_transform(tstate, pad, 0);
+}
+
+static inline void PBKDF2_SHA256_80_128(const uint32_t *tstate,
+	const uint32_t *ostate, const uint32_t *salt, uint32_t *output)
+{
+	uint32_t istate[8], ostate2[8];
+	uint32_t ibuf[16], obuf[16];
+	int i, j;
+
+	memcpy(istate, tstate, 32);
+	sha256_transform(istate, salt, 0);
+	
+	memcpy(ibuf, salt + 16, 16);
+	memcpy(ibuf + 5, innerpad, 44);
+	memcpy(obuf + 8, outerpad, 32);
+
+	for (i = 0; i < 4; i++) {
+		memcpy(obuf, istate, 32);
+		ibuf[4] = i + 1;
+		sha256_transform(obuf, ibuf, 0);
+
+		memcpy(ostate2, ostate, 32);
+		sha256_transform(ostate2, obuf, 0);
+		for (j = 0; j < 8; j++)
+			output[8 * i + j] = swab32(ostate2[j]);
+	}
+}
+
+static inline void PBKDF2_SHA256_128_32(uint32_t *tstate, uint32_t *ostate,
+	const uint32_t *salt, uint32_t *output)
+{
+	uint32_t buf[16];
+	int i;
+	
+	sha256_transform(tstate, salt, 1);
+	sha256_transform(tstate, salt + 16, 1);
+	sha256_transform(tstate, finalblk, 0);
+	memcpy(buf, tstate, 32);
+	memcpy(buf + 8, outerpad, 32);
+
+	sha256_transform(ostate, buf, 0);
+	for (i = 0; i < 8; i++)
+		output[i] = swab32(ostate[i]);
+}
+
+
+#ifdef HAVE_SHA256_4WAY
+
+static const uint32_t keypad_4way[4 * 12] = {
+	0x80000000, 0x80000000, 0x80000000, 0x80000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000280, 0x00000280, 0x00000280, 0x00000280
+};
+static const uint32_t innerpad_4way[4 * 11] = {
+	0x80000000, 0x80000000, 0x80000000, 0x80000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x000004a0, 0x000004a0, 0x000004a0, 0x000004a0
+};
+static const uint32_t outerpad_4way[4 * 8] = {
+	0x80000000, 0x80000000, 0x80000000, 0x80000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000300, 0x00000300, 0x00000300, 0x00000300
+};
+static const uint32_t finalblk_4way[4 * 16] __attribute__((aligned(16))) = {
+	0x00000001, 0x00000001, 0x00000001, 0x00000001,
+	0x80000000, 0x80000000, 0x80000000, 0x80000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000620, 0x00000620, 0x00000620, 0x00000620
+};
+
+static inline void HMAC_SHA256_80_init_4way(const uint32_t *key,
+	uint32_t *tstate, uint32_t *ostate)
+{
+	uint32_t ihash[4 * 8] __attribute__((aligned(16)));
+	uint32_t pad[4 * 16] __attribute__((aligned(16)));
+	int i;
+
+	/* tstate is assumed to contain the midstate of key */
+	memcpy(pad, key + 4 * 16, 4 * 16);
+	memcpy(pad + 4 * 4, keypad_4way, 4 * 48);
+	sha256_transform_4way(tstate, pad, 0);
+	memcpy(ihash, tstate, 4 * 32);
+
+	sha256_init_4way(ostate);
+	for (i = 0; i < 4 * 8; i++)
+		pad[i] = ihash[i] ^ 0x5c5c5c5c;
+	for (; i < 4 * 16; i++)
+		pad[i] = 0x5c5c5c5c;
+	sha256_transform_4way(ostate, pad, 0);
+
+	sha256_init_4way(tstate);
+	for (i = 0; i < 4 * 8; i++)
+		pad[i] = ihash[i] ^ 0x36363636;
+	for (; i < 4 * 16; i++)
+		pad[i] = 0x36363636;
+	sha256_transform_4way(tstate, pad, 0);
+}
+
+static inline void PBKDF2_SHA256_80_128_4way(const uint32_t *tstate,
+	const uint32_t *ostate, const uint32_t *salt, uint32_t *output)
+{
+	uint32_t istate[4 * 8] __attribute__((aligned(16)));
+	uint32_t ostate2[4 * 8] __attribute__((aligned(16)));
+	uint32_t ibuf[4 * 16] __attribute__((aligned(16)));
+	uint32_t obuf[4 * 16] __attribute__((aligned(16)));
+	int i, j;
+
+	memcpy(istate, tstate, 4 * 32);
+	sha256_transform_4way(istate, salt, 0);
+	
+	memcpy(ibuf, salt + 4 * 16, 4 * 16);
+	memcpy(ibuf + 4 * 5, innerpad_4way, 4 * 44);
+	memcpy(obuf + 4 * 8, outerpad_4way, 4 * 32);
+
+	for (i = 0; i < 4; i++) {
+		memcpy(obuf, istate, 4 * 32);
+		ibuf[4 * 4 + 0] = i + 1;
+		ibuf[4 * 4 + 1] = i + 1;
+		ibuf[4 * 4 + 2] = i + 1;
+		ibuf[4 * 4 + 3] = i + 1;
+		sha256_transform_4way(obuf, ibuf, 0);
+
+		memcpy(ostate2, ostate, 4 * 32);
+		sha256_transform_4way(ostate2, obuf, 0);
+		for (j = 0; j < 4 * 8; j++)
+			output[4 * 8 * i + j] = swab32(ostate2[j]);
+	}
+}
+
+static inline void PBKDF2_SHA256_128_32_4way(uint32_t *tstate,
+	uint32_t *ostate, const uint32_t *salt, uint32_t *output)
+{
+	uint32_t buf[4 * 16] __attribute__((aligned(16)));
+	int i;
+	
+	sha256_transform_4way(tstate, salt, 1);
+	sha256_transform_4way(tstate, salt + 4 * 16, 1);
+	sha256_transform_4way(tstate, finalblk_4way, 0);
+	memcpy(buf, tstate, 4 * 32);
+	memcpy(buf + 4 * 8, outerpad_4way, 4 * 32);
+
+	sha256_transform_4way(ostate, buf, 0);
+	for (i = 0; i < 4 * 8; i++)
+		output[i] = swab32(ostate[i]);
+}
+
+#endif /* HAVE_SHA256_4WAY */
+
+
+#ifdef HAVE_SHA256_8WAY
+
+static const uint32_t finalblk_8way[8 * 16] __attribute__((aligned(32))) = {
+	0x00000001, 0x00000001, 0x00000001, 0x00000001, 0x00000001, 0x00000001, 0x00000001, 0x00000001,
+	0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000620, 0x00000620, 0x00000620, 0x00000620, 0x00000620, 0x00000620, 0x00000620, 0x00000620
+};
+
+static inline void HMAC_SHA256_80_init_8way(const uint32_t *key,
+	uint32_t *tstate, uint32_t *ostate)
+{
+	uint32_t ihash[8 * 8] __attribute__((aligned(32)));
+	uint32_t pad[8 * 16] __attribute__((aligned(32)));
+	int i;
+	
+	/* tstate is assumed to contain the midstate of key */
+	memcpy(pad, key + 8 * 16, 8 * 16);
+	for (i = 0; i < 8; i++)
+		pad[8 * 4 + i] = 0x80000000;
+	memset(pad + 8 * 5, 0x00, 8 * 40);
+	for (i = 0; i < 8; i++)
+		pad[8 * 15 + i] = 0x00000280;
+	sha256_transform_8way(tstate, pad, 0);
+	memcpy(ihash, tstate, 8 * 32);
+	
+	sha256_init_8way(ostate);
+	for (i = 0; i < 8 * 8; i++)
+		pad[i] = ihash[i] ^ 0x5c5c5c5c;
+	for (; i < 8 * 16; i++)
+		pad[i] = 0x5c5c5c5c;
+	sha256_transform_8way(ostate, pad, 0);
+	
+	sha256_init_8way(tstate);
+	for (i = 0; i < 8 * 8; i++)
+		pad[i] = ihash[i] ^ 0x36363636;
+	for (; i < 8 * 16; i++)
+		pad[i] = 0x36363636;
+	sha256_transform_8way(tstate, pad, 0);
+}
+
+static inline void PBKDF2_SHA256_80_128_8way(const uint32_t *tstate,
+	const uint32_t *ostate, const uint32_t *salt, uint32_t *output)
+{
+	uint32_t istate[8 * 8] __attribute__((aligned(32)));
+	uint32_t ostate2[8 * 8] __attribute__((aligned(32)));
+	uint32_t ibuf[8 * 16] __attribute__((aligned(32)));
+	uint32_t obuf[8 * 16] __attribute__((aligned(32)));
+	int i, j;
+	
+	memcpy(istate, tstate, 8 * 32);
+	sha256_transform_8way(istate, salt, 0);
+	
+	memcpy(ibuf, salt + 8 * 16, 8 * 16);
+	for (i = 0; i < 8; i++)
+		ibuf[8 * 5 + i] = 0x80000000;
+	memset(ibuf + 8 * 6, 0x00, 8 * 36);
+	for (i = 0; i < 8; i++)
+		ibuf[8 * 15 + i] = 0x000004a0;
+	
+	for (i = 0; i < 8; i++)
+		obuf[8 * 8 + i] = 0x80000000;
+	memset(obuf + 8 * 9, 0x00, 8 * 24);
+	for (i = 0; i < 8; i++)
+		obuf[8 * 15 + i] = 0x00000300;
+	
+	for (i = 0; i < 4; i++) {
+		memcpy(obuf, istate, 8 * 32);
+		ibuf[8 * 4 + 0] = i + 1;
+		ibuf[8 * 4 + 1] = i + 1;
+		ibuf[8 * 4 + 2] = i + 1;
+		ibuf[8 * 4 + 3] = i + 1;
+		ibuf[8 * 4 + 4] = i + 1;
+		ibuf[8 * 4 + 5] = i + 1;
+		ibuf[8 * 4 + 6] = i + 1;
+		ibuf[8 * 4 + 7] = i + 1;
+		sha256_transform_8way(obuf, ibuf, 0);
+		
+		memcpy(ostate2, ostate, 8 * 32);
+		sha256_transform_8way(ostate2, obuf, 0);
+		for (j = 0; j < 8 * 8; j++)
+			output[8 * 8 * i + j] = swab32(ostate2[j]);
+	}
+}
+
+static inline void PBKDF2_SHA256_128_32_8way(uint32_t *tstate,
+	uint32_t *ostate, const uint32_t *salt, uint32_t *output)
+{
+	uint32_t buf[8 * 16] __attribute__((aligned(32)));
+	int i;
+	
+	sha256_transform_8way(tstate, salt, 1);
+	sha256_transform_8way(tstate, salt + 8 * 16, 1);
+	sha256_transform_8way(tstate, finalblk_8way, 0);
+	
+	memcpy(buf, tstate, 8 * 32);
+	for (i = 0; i < 8; i++)
+		buf[8 * 8 + i] = 0x80000000;
+	memset(buf + 8 * 9, 0x00, 8 * 24);
+	for (i = 0; i < 8; i++)
+		buf[8 * 15 + i] = 0x00000300;
+	sha256_transform_8way(ostate, buf, 0);
+	
+	for (i = 0; i < 8 * 8; i++)
+		output[i] = swab32(ostate[i]);
+}
+
+#endif /* HAVE_SHA256_8WAY */
+
+
+#if defined(USE_ASM) && defined(__x86_64__)
+
+#define SCRYPT_MAX_WAYS 12
+#define HAVE_SCRYPT_3WAY 1
+int scrypt_best_throughput();
+void scrypt_core(uint32_t *X, uint32_t *V, int N);
+void scrypt_core_3way(uint32_t *X, uint32_t *V, int N);
+#if defined(USE_AVX2)
+#undef SCRYPT_MAX_WAYS
+#define SCRYPT_MAX_WAYS 24
+#define HAVE_SCRYPT_6WAY 1
+void scrypt_core_6way(uint32_t *X, uint32_t *V, int N);
+#endif
+
+#elif defined(USE_ASM) && defined(__i386__)
+
+#define SCRYPT_MAX_WAYS 4
+#define scrypt_best_throughput() 1
+void scrypt_core(uint32_t *X, uint32_t *V, int N);
+
+#elif defined(USE_ASM) && defined(__arm__) && defined(__APCS_32__)
+
+void scrypt_core(uint32_t *X, uint32_t *V, int N);
+#if defined(__ARM_NEON__)
+#undef HAVE_SHA256_4WAY
+#define SCRYPT_MAX_WAYS 3
+#define HAVE_SCRYPT_3WAY 1
+#define scrypt_best_throughput() 3
+void scrypt_core_3way(uint32_t *X, uint32_t *V, int N);
+#endif
+
+#elif defined(USE_ASM) && (defined(__powerpc__) || defined(__ppc__) || defined(__PPC__))
+
+#define SCRYPT_MAX_WAYS 4
+#define scrypt_best_throughput() 1
+void scrypt_core(uint32_t *X, uint32_t *V, int N);
+
+#else
+
+static inline void xor_salsa8(uint32_t B[16], const uint32_t Bx[16])
+{
+	uint32_t x00,x01,x02,x03,x04,x05,x06,x07,x08,x09,x10,x11,x12,x13,x14,x15;
+	int i;
+
+	x00 = (B[ 0] ^= Bx[ 0]);
+	x01 = (B[ 1] ^= Bx[ 1]);
+	x02 = (B[ 2] ^= Bx[ 2]);
+	x03 = (B[ 3] ^= Bx[ 3]);
+	x04 = (B[ 4] ^= Bx[ 4]);
+	x05 = (B[ 5] ^= Bx[ 5]);
+	x06 = (B[ 6] ^= Bx[ 6]);
+	x07 = (B[ 7] ^= Bx[ 7]);
+	x08 = (B[ 8] ^= Bx[ 8]);
+	x09 = (B[ 9] ^= Bx[ 9]);
+	x10 = (B[10] ^= Bx[10]);
+	x11 = (B[11] ^= Bx[11]);
+	x12 = (B[12] ^= Bx[12]);
+	x13 = (B[13] ^= Bx[13]);
+	x14 = (B[14] ^= Bx[14]);
+	x15 = (B[15] ^= Bx[15]);
+	for (i = 0; i < 8; i += 2) {
+#define R(a, b) (((a) << (b)) | ((a) >> (32 - (b))))
+		/* Operate on columns. */
+		x04 ^= R(x00+x12, 7);	x09 ^= R(x05+x01, 7);
+		x14 ^= R(x10+x06, 7);	x03 ^= R(x15+x11, 7);
+		
+		x08 ^= R(x04+x00, 9);	x13 ^= R(x09+x05, 9);
+		x02 ^= R(x14+x10, 9);	x07 ^= R(x03+x15, 9);
+		
+		x12 ^= R(x08+x04,13);	x01 ^= R(x13+x09,13);
+		x06 ^= R(x02+x14,13);	x11 ^= R(x07+x03,13);
+		
+		x00 ^= R(x12+x08,18);	x05 ^= R(x01+x13,18);
+		x10 ^= R(x06+x02,18);	x15 ^= R(x11+x07,18);
+		
+		/* Operate on rows. */
+		x01 ^= R(x00+x03, 7);	x06 ^= R(x05+x04, 7);
+		x11 ^= R(x10+x09, 7);	x12 ^= R(x15+x14, 7);
+		
+		x02 ^= R(x01+x00, 9);	x07 ^= R(x06+x05, 9);
+		x08 ^= R(x11+x10, 9);	x13 ^= R(x12+x15, 9);
+		
+		x03 ^= R(x02+x01,13);	x04 ^= R(x07+x06,13);
+		x09 ^= R(x08+x11,13);	x14 ^= R(x13+x12,13);
+		
+		x00 ^= R(x03+x02,18);	x05 ^= R(x04+x07,18);
+		x10 ^= R(x09+x08,18);	x15 ^= R(x14+x13,18);
+#undef R
+	}
+	B[ 0] += x00;
+	B[ 1] += x01;
+	B[ 2] += x02;
+	B[ 3] += x03;
+	B[ 4] += x04;
+	B[ 5] += x05;
+	B[ 6] += x06;
+	B[ 7] += x07;
+	B[ 8] += x08;
+	B[ 9] += x09;
+	B[10] += x10;
+	B[11] += x11;
+	B[12] += x12;
+	B[13] += x13;
+	B[14] += x14;
+	B[15] += x15;
+}
+
+static inline void scrypt_core(uint32_t *X, uint32_t *V, int N)
+{
+	uint32_t i, j, k;
+	
+	for (i = 0; i < N; i++) {
+		memcpy(&V[i * 32], X, 128);
+		xor_salsa8(&X[0], &X[16]);
+		xor_salsa8(&X[16], &X[0]);
+	}
+	for (i = 0; i < N; i++) {
+		j = 32 * (X[16] & (N - 1));
+		for (k = 0; k < 32; k++)
+			X[k] ^= V[j + k];
+		xor_salsa8(&X[0], &X[16]);
+		xor_salsa8(&X[16], &X[0]);
+	}
+}
+
+#endif
+
+#ifndef SCRYPT_MAX_WAYS
+#define SCRYPT_MAX_WAYS 1
+#define scrypt_best_throughput() 1
+#endif
+
+unsigned char *scrypt_buffer_alloc(int N)
+{
+	return malloc((size_t)N * SCRYPT_MAX_WAYS * 128 + 63);
+}
+
+static void scrypt_1024_1_1_256(const uint32_t *input, uint32_t *output,
+	uint32_t *midstate, unsigned char *scratchpad, int N)
+{
+	uint32_t tstate[8], ostate[8];
+	uint32_t X[32] __attribute__((aligned(128)));
+	uint32_t *V;
+	
+	V = (uint32_t *)(((uintptr_t)(scratchpad) + 63) & ~ (uintptr_t)(63));
+
+	memcpy(tstate, midstate, 32);
+	HMAC_SHA256_80_init(input, tstate, ostate);
+	PBKDF2_SHA256_80_128(tstate, ostate, input, X);
+
+	scrypt_core(X, V, N);
+
+	PBKDF2_SHA256_128_32(tstate, ostate, X, output);
+}
+
+#ifdef HAVE_SHA256_4WAY
+static void scrypt_1024_1_1_256_4way(const uint32_t *input,
+	uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
+{
+	uint32_t tstate[4 * 8] __attribute__((aligned(128)));
+	uint32_t ostate[4 * 8] __attribute__((aligned(128)));
+	uint32_t W[4 * 32] __attribute__((aligned(128)));
+	uint32_t X[4 * 32] __attribute__((aligned(128)));
+	uint32_t *V;
+	int i, k;
+	
+	V = (uint32_t *)(((uintptr_t)(scratchpad) + 63) & ~ (uintptr_t)(63));
+
+	for (i = 0; i < 20; i++)
+		for (k = 0; k < 4; k++)
+			W[4 * i + k] = input[k * 20 + i];
+	for (i = 0; i < 8; i++)
+		for (k = 0; k < 4; k++)
+			tstate[4 * i + k] = midstate[i];
+	HMAC_SHA256_80_init_4way(W, tstate, ostate);
+	PBKDF2_SHA256_80_128_4way(tstate, ostate, W, W);
+	for (i = 0; i < 32; i++)
+		for (k = 0; k < 4; k++)
+			X[k * 32 + i] = W[4 * i + k];
+	scrypt_core(X + 0 * 32, V, N);
+	scrypt_core(X + 1 * 32, V, N);
+	scrypt_core(X + 2 * 32, V, N);
+	scrypt_core(X + 3 * 32, V, N);
+	for (i = 0; i < 32; i++)
+		for (k = 0; k < 4; k++)
+			W[4 * i + k] = X[k * 32 + i];
+	PBKDF2_SHA256_128_32_4way(tstate, ostate, W, W);
+	for (i = 0; i < 8; i++)
+		for (k = 0; k < 4; k++)
+			output[k * 8 + i] = W[4 * i + k];
+}
+#endif /* HAVE_SHA256_4WAY */
+
+#ifdef HAVE_SCRYPT_3WAY
+
+static void scrypt_1024_1_1_256_3way(const uint32_t *input,
+	uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
+{
+	uint32_t tstate[3 * 8], ostate[3 * 8];
+	uint32_t X[3 * 32] __attribute__((aligned(64)));
+	uint32_t *V;
+	
+	V = (uint32_t *)(((uintptr_t)(scratchpad) + 63) & ~ (uintptr_t)(63));
+
+	memcpy(tstate +  0, midstate, 32);
+	memcpy(tstate +  8, midstate, 32);
+	memcpy(tstate + 16, midstate, 32);
+	HMAC_SHA256_80_init(input +  0, tstate +  0, ostate +  0);
+	HMAC_SHA256_80_init(input + 20, tstate +  8, ostate +  8);
+	HMAC_SHA256_80_init(input + 40, tstate + 16, ostate + 16);
+	PBKDF2_SHA256_80_128(tstate +  0, ostate +  0, input +  0, X +  0);
+	PBKDF2_SHA256_80_128(tstate +  8, ostate +  8, input + 20, X + 32);
+	PBKDF2_SHA256_80_128(tstate + 16, ostate + 16, input + 40, X + 64);
+
+	scrypt_core_3way(X, V, N);
+
+	PBKDF2_SHA256_128_32(tstate +  0, ostate +  0, X +  0, output +  0);
+	PBKDF2_SHA256_128_32(tstate +  8, ostate +  8, X + 32, output +  8);
+	PBKDF2_SHA256_128_32(tstate + 16, ostate + 16, X + 64, output + 16);
+}
+
+#ifdef HAVE_SHA256_4WAY
+static void scrypt_1024_1_1_256_12way(const uint32_t *input,
+	uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
+{
+	uint32_t tstate[12 * 8] __attribute__((aligned(128)));
+	uint32_t ostate[12 * 8] __attribute__((aligned(128)));
+	uint32_t W[12 * 32] __attribute__((aligned(128)));
+	uint32_t X[12 * 32] __attribute__((aligned(128)));
+	uint32_t *V;
+	int i, j, k;
+	
+	V = (uint32_t *)(((uintptr_t)(scratchpad) + 63) & ~ (uintptr_t)(63));
+
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 20; i++)
+			for (k = 0; k < 4; k++)
+				W[128 * j + 4 * i + k] = input[80 * j + k * 20 + i];
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 8; i++)
+			for (k = 0; k < 4; k++)
+				tstate[32 * j + 4 * i + k] = midstate[i];
+	HMAC_SHA256_80_init_4way(W +   0, tstate +  0, ostate +  0);
+	HMAC_SHA256_80_init_4way(W + 128, tstate + 32, ostate + 32);
+	HMAC_SHA256_80_init_4way(W + 256, tstate + 64, ostate + 64);
+	PBKDF2_SHA256_80_128_4way(tstate +  0, ostate +  0, W +   0, W +   0);
+	PBKDF2_SHA256_80_128_4way(tstate + 32, ostate + 32, W + 128, W + 128);
+	PBKDF2_SHA256_80_128_4way(tstate + 64, ostate + 64, W + 256, W + 256);
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 32; i++)
+			for (k = 0; k < 4; k++)
+				X[128 * j + k * 32 + i] = W[128 * j + 4 * i + k];
+	scrypt_core_3way(X + 0 * 96, V, N);
+	scrypt_core_3way(X + 1 * 96, V, N);
+	scrypt_core_3way(X + 2 * 96, V, N);
+	scrypt_core_3way(X + 3 * 96, V, N);
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 32; i++)
+			for (k = 0; k < 4; k++)
+				W[128 * j + 4 * i + k] = X[128 * j + k * 32 + i];
+	PBKDF2_SHA256_128_32_4way(tstate +  0, ostate +  0, W +   0, W +   0);
+	PBKDF2_SHA256_128_32_4way(tstate + 32, ostate + 32, W + 128, W + 128);
+	PBKDF2_SHA256_128_32_4way(tstate + 64, ostate + 64, W + 256, W + 256);
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 8; i++)
+			for (k = 0; k < 4; k++)
+				output[32 * j + k * 8 + i] = W[128 * j + 4 * i + k];
+}
+#endif /* HAVE_SHA256_4WAY */
+
+#endif /* HAVE_SCRYPT_3WAY */
+
+#ifdef HAVE_SCRYPT_6WAY
+static void scrypt_1024_1_1_256_24way(const uint32_t *input,
+	uint32_t *output, uint32_t *midstate, unsigned char *scratchpad, int N)
+{
+	uint32_t tstate[24 * 8] __attribute__((aligned(128)));
+	uint32_t ostate[24 * 8] __attribute__((aligned(128)));
+	uint32_t W[24 * 32] __attribute__((aligned(128)));
+	uint32_t X[24 * 32] __attribute__((aligned(128)));
+	uint32_t *V;
+	int i, j, k;
+	
+	V = (uint32_t *)(((uintptr_t)(scratchpad) + 63) & ~ (uintptr_t)(63));
+	
+	for (j = 0; j < 3; j++) 
+		for (i = 0; i < 20; i++)
+			for (k = 0; k < 8; k++)
+				W[8 * 32 * j + 8 * i + k] = input[8 * 20 * j + k * 20 + i];
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 8; i++)
+			for (k = 0; k < 8; k++)
+				tstate[8 * 8 * j + 8 * i + k] = midstate[i];
+	HMAC_SHA256_80_init_8way(W +   0, tstate +   0, ostate +   0);
+	HMAC_SHA256_80_init_8way(W + 256, tstate +  64, ostate +  64);
+	HMAC_SHA256_80_init_8way(W + 512, tstate + 128, ostate + 128);
+	PBKDF2_SHA256_80_128_8way(tstate +   0, ostate +   0, W +   0, W +   0);
+	PBKDF2_SHA256_80_128_8way(tstate +  64, ostate +  64, W + 256, W + 256);
+	PBKDF2_SHA256_80_128_8way(tstate + 128, ostate + 128, W + 512, W + 512);
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 32; i++)
+			for (k = 0; k < 8; k++)
+				X[8 * 32 * j + k * 32 + i] = W[8 * 32 * j + 8 * i + k];
+	scrypt_core_6way(X + 0 * 32, V, N);
+	scrypt_core_6way(X + 6 * 32, V, N);
+	scrypt_core_6way(X + 12 * 32, V, N);
+	scrypt_core_6way(X + 18 * 32, V, N);
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 32; i++)
+			for (k = 0; k < 8; k++)
+				W[8 * 32 * j + 8 * i + k] = X[8 * 32 * j + k * 32 + i];
+	PBKDF2_SHA256_128_32_8way(tstate +   0, ostate +   0, W +   0, W +   0);
+	PBKDF2_SHA256_128_32_8way(tstate +  64, ostate +  64, W + 256, W + 256);
+	PBKDF2_SHA256_128_32_8way(tstate + 128, ostate + 128, W + 512, W + 512);
+	for (j = 0; j < 3; j++)
+		for (i = 0; i < 8; i++)
+			for (k = 0; k < 8; k++)
+				output[8 * 8 * j + k * 8 + i] = W[8 * 32 * j + 8 * i + k];
+}
+#endif /* HAVE_SCRYPT_6WAY */
+
+int scanhash_scrypt(int thr_id, uint32_t *pdata,
+	unsigned char *scratchbuf, const uint32_t *ptarget,
+	uint32_t max_nonce, unsigned long *hashes_done, int N)
+{
+	uint32_t data[SCRYPT_MAX_WAYS * 20], hash[SCRYPT_MAX_WAYS * 8];
+	uint32_t midstate[8];
+	uint32_t n = pdata[19] - 1;
+	const uint32_t Htarg = ptarget[7];
+	int throughput = scrypt_best_throughput();
+	int i;
+	
+#ifdef HAVE_SHA256_4WAY
+	if (sha256_use_4way())
+		throughput *= 4;
+#endif
+	
+	for (i = 0; i < throughput; i++)
+		memcpy(data + i * 20, pdata, 80);
+	
+	sha256_init(midstate);
+	sha256_transform(midstate, data, 0);
+	
+	do {
+		for (i = 0; i < throughput; i++)
+			data[i * 20 + 19] = ++n;
+		
+#if defined(HAVE_SHA256_4WAY)
+		if (throughput == 4)
+			scrypt_1024_1_1_256_4way(data, hash, midstate, scratchbuf, N);
+		else
+#endif
+#if defined(HAVE_SCRYPT_3WAY) && defined(HAVE_SHA256_4WAY)
+		if (throughput == 12)
+			scrypt_1024_1_1_256_12way(data, hash, midstate, scratchbuf, N);
+		else
+#endif
+#if defined(HAVE_SCRYPT_6WAY)
+		if (throughput == 24)
+			scrypt_1024_1_1_256_24way(data, hash, midstate, scratchbuf, N);
+		else
+#endif
+#if defined(HAVE_SCRYPT_3WAY)
+		if (throughput == 3)
+			scrypt_1024_1_1_256_3way(data, hash, midstate, scratchbuf, N);
+		else
+#endif
+		scrypt_1024_1_1_256(data, hash, midstate, scratchbuf, N);
+		
+		for (i = 0; i < throughput; i++) {
+			if (hash[i * 8 + 7] <= Htarg && fulltest(hash + i * 8, ptarget)) {
+				*hashes_done = n - pdata[19] + 1;
+				pdata[19] = data[i * 20 + 19];
+				return 1;
+			}
+		}
+	} while (n < max_nonce && !work_restart[thr_id].restart);
+	
+	*hashes_done = n - pdata[19] + 1;
+	pdata[19] = n;
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-arm.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-arm.S
new file mode 100644
index 0000000..bd7fdc5
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-arm.S
@@ -0,0 +1,1583 @@
+/*
+ * Copyright 2012 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(USE_ASM) && defined(__arm__) && defined(__APCS_32__)
+
+.macro sha256_k
+	.align 2
+	.long 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5
+	.long 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5
+	.long 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3
+	.long 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174
+	.long 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc
+	.long 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da
+	.long 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7
+	.long 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967
+	.long 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13
+	.long 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85
+	.long 0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3
+	.long 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070
+	.long 0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5
+	.long 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3
+	.long 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208
+	.long 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
+.endm
+
+.macro sha256_extend_doubleround_core i, rw, ra, rb, ry, rz
+	mov	r12, \ry, ror #17
+	add	r11, r11, \ra
+	eor	r12, r12, \ry, ror #19
+	mov	\ra, lr, ror #7
+	eor	r12, r12, \ry, lsr #10
+	eor	\ra, \ra, lr, ror #18
+	add	r12, r12, r11
+	ldr	r11, [\rw, #(\i+2)*4]
+	eor	\ra, \ra, lr, lsr #3
+	add	\ra, \ra, r12
+
+	mov	r12, \rz, ror #17
+	str	\ra, [\rw, #(\i+16)*4]
+	add	lr, lr, \rb
+	eor	r12, r12, \rz, ror #19
+	mov	\rb, r11, ror #7
+	eor	r12, r12, \rz, lsr #10
+	eor	\rb, \rb, r11, ror #18
+	add	lr, lr, r12
+	eor	\rb, \rb, r11, lsr #3
+	add	\rb, \rb, lr
+.endm
+
+.macro sha256_extend_doubleround_head i, rw, ra, rb, ry, rz
+	ldr	lr, [\rw, #(\i+1)*4]
+	sha256_extend_doubleround_core \i, \rw, \ra, \rb, \ry, \rz
+	ldr	lr, [\rw, #(\i+3)*4]
+.endm
+
+.macro sha256_extend_doubleround_body i, rw, ra, rb, ry, rz
+	str	\rz, [\rw, #(\i+15)*4]
+	sha256_extend_doubleround_core \i, \rw, \ra, \rb, \ry, \rz
+	ldr	lr, [\rw, #(\i+3)*4]
+.endm
+
+.macro sha256_extend_doubleround_foot i, rw, ra, rb, ry, rz
+	str	\rz, [\rw, #(\i+15)*4]
+	sha256_extend_doubleround_core \i, \rw, \ra, \rb, \ry, \rz
+	str	\rb, [\rw, #(\i+17)*4]
+.endm
+
+.macro sha256_main_round i, ka, rw, ra, rb, rc, rd, re, rf, rg, rh
+	ldr	r12, [\rw, #(\i)*4]
+	and	r3, \rf, \re
+	bic	lr, \rg, \re
+	orr	lr, lr, r3
+	ldr	r3, \ka + (\i)*4
+	add	\rh, \rh, lr
+	eor	lr, \re, \re, ror #5
+	add	\rh, \rh, r12
+	eor	lr, lr, \re, ror #19
+	add	\rh, \rh, r3
+	eor	r3, \ra, \rb
+	add	\rh, \rh, lr, ror #6
+
+	and	r3, r3, \rc
+	eor	r12, \ra, \ra, ror #11
+	and	lr, \ra, \rb
+	eor	r12, r12, \ra, ror #20
+	eor	lr, lr, r3
+	add	r3, \rh, lr
+	add	\rh, \rh, \rd
+	add	\rd, r3, r12, ror #2
+.endm
+
+.macro sha256_main_quadround i, ka, rw
+	sha256_main_round \i+0, \ka, \rw, r4, r5, r6, r7, r8, r9, r10, r11
+	sha256_main_round \i+1, \ka, \rw, r7, r4, r5, r6, r11, r8, r9, r10
+	sha256_main_round \i+2, \ka, \rw, r6, r7, r4, r5, r10, r11, r8, r9
+	sha256_main_round \i+3, \ka, \rw, r5, r6, r7, r4, r9, r10, r11, r8
+.endm
+
+
+	.text
+	.code 32
+	.align 2
+	.globl sha256_transform
+	.globl _sha256_transform
+#ifdef __ELF__
+	.type sha256_transform, %function
+#endif
+sha256_transform:
+_sha256_transform:
+	stmfd	sp!, {r4-r11, lr}
+	cmp	r2, #0
+	sub	sp, sp, #64*4
+	bne	sha256_transform_swap
+	
+	ldmia	r1!, {r4-r11}
+	stmia	sp, {r4-r11}
+	add	r3, sp, #8*4
+	ldmia	r1, {r4-r11}
+	stmia	r3, {r4-r11}
+	b	sha256_transform_extend
+
+.macro bswap rd, rn
+	eor	r12, \rn, \rn, ror #16
+	bic	r12, r12, #0x00ff0000
+	mov	\rd, \rn, ror #8
+	eor	\rd, \rd, r12, lsr #8
+.endm
+
+sha256_transform_swap:
+	ldmia	r1!, {r4-r11}
+	bswap	r4, r4
+	bswap	r5, r5
+	bswap	r6, r6
+	bswap	r7, r7
+	bswap	r8, r8
+	bswap	r9, r9
+	bswap	r10, r10
+	bswap	r11, r11
+	stmia	sp, {r4-r11}
+	add	r3, sp, #8*4
+	ldmia	r1, {r4-r11}
+	bswap	r4, r4
+	bswap	r5, r5
+	bswap	r6, r6
+	bswap	r7, r7
+	bswap	r8, r8
+	bswap	r9, r9
+	bswap	r10, r10
+	bswap	r11, r11
+	stmia	r3, {r4-r11}
+	
+sha256_transform_extend:
+	add	r12, sp, #9*4
+	ldr	r11, [sp, #0*4]
+	ldmia	r12, {r4-r10}
+	sha256_extend_doubleround_head  0, sp, r4, r5, r9, r10
+	sha256_extend_doubleround_body  2, sp, r6, r7, r4, r5
+	sha256_extend_doubleround_body  4, sp, r8, r9, r6, r7
+	sha256_extend_doubleround_body  6, sp, r10, r4, r8, r9
+	sha256_extend_doubleround_body  8, sp, r5, r6, r10, r4
+	sha256_extend_doubleround_body 10, sp, r7, r8, r5, r6
+	sha256_extend_doubleround_body 12, sp, r9, r10, r7, r8
+	sha256_extend_doubleround_body 14, sp, r4, r5, r9, r10
+	sha256_extend_doubleround_body 16, sp, r6, r7, r4, r5
+	sha256_extend_doubleround_body 18, sp, r8, r9, r6, r7
+	sha256_extend_doubleround_body 20, sp, r10, r4, r8, r9
+	sha256_extend_doubleround_body 22, sp, r5, r6, r10, r4
+	sha256_extend_doubleround_body 24, sp, r7, r8, r5, r6
+	sha256_extend_doubleround_body 26, sp, r9, r10, r7, r8
+	sha256_extend_doubleround_body 28, sp, r4, r5, r9, r10
+	sha256_extend_doubleround_body 30, sp, r6, r7, r4, r5
+	sha256_extend_doubleround_body 32, sp, r8, r9, r6, r7
+	sha256_extend_doubleround_body 34, sp, r10, r4, r8, r9
+	sha256_extend_doubleround_body 36, sp, r5, r6, r10, r4
+	sha256_extend_doubleround_body 38, sp, r7, r8, r5, r6
+	sha256_extend_doubleround_body 40, sp, r9, r10, r7, r8
+	sha256_extend_doubleround_body 42, sp, r4, r5, r9, r10
+	sha256_extend_doubleround_body 44, sp, r6, r7, r4, r5
+	sha256_extend_doubleround_foot 46, sp, r8, r9, r6, r7
+	
+	ldmia	r0, {r4-r11}
+	sha256_main_quadround  0, sha256_transform_k, sp
+	sha256_main_quadround  4, sha256_transform_k, sp
+	sha256_main_quadround  8, sha256_transform_k, sp
+	sha256_main_quadround 12, sha256_transform_k, sp
+	sha256_main_quadround 16, sha256_transform_k, sp
+	sha256_main_quadround 20, sha256_transform_k, sp
+	sha256_main_quadround 24, sha256_transform_k, sp
+	sha256_main_quadround 28, sha256_transform_k, sp
+	b	sha256_transform_k_over
+sha256_transform_k:
+	sha256_k
+sha256_transform_k_over:
+	sha256_main_quadround 32, sha256_transform_k, sp
+	sha256_main_quadround 36, sha256_transform_k, sp
+	sha256_main_quadround 40, sha256_transform_k, sp
+	sha256_main_quadround 44, sha256_transform_k, sp
+	sha256_main_quadround 48, sha256_transform_k, sp
+	sha256_main_quadround 52, sha256_transform_k, sp
+	sha256_main_quadround 56, sha256_transform_k, sp
+	sha256_main_quadround 60, sha256_transform_k, sp
+	
+	ldmia	r0, {r1, r2, r3, r12}
+	add	r4, r4, r1
+	add	r5, r5, r2
+	add	r6, r6, r3
+	add	r7, r7, r12
+	stmia	r0!, {r4-r7}
+	ldmia	r0, {r1, r2, r3, r12}
+	add	r8, r8, r1
+	add	r9, r9, r2
+	add	r10, r10, r3
+	add	r11, r11, r12
+	stmia	r0, {r8-r11}
+	
+	add	sp, sp, #64*4
+#ifdef __thumb__
+	ldmfd	sp!, {r4-r11, lr}
+	bx	lr
+#else
+	ldmfd	sp!, {r4-r11, pc}
+#endif
+
+
+	.text
+	.code 32
+	.align 2
+	.globl sha256d_ms
+	.globl _sha256d_ms
+#ifdef __ELF__
+	.type sha256d_ms, %function
+#endif
+sha256d_ms:
+_sha256d_ms:
+	stmfd	sp!, {r4-r11, lr}
+	sub	sp, sp, #64*4
+	
+	cmp	r0, r0
+	
+	ldr	lr, [r1, #3*4]
+	ldr	r6, [r1, #18*4]
+	ldr	r7, [r1, #19*4]
+	
+	mov	r12, lr, ror #7
+	str	r6, [sp, #18*4]
+	eor	r12, r12, lr, ror #18
+	str	r7, [sp, #19*4]
+	eor	r12, r12, lr, lsr #3
+	ldr	r8, [r1, #20*4]
+	add	r6, r6, r12
+	ldr	r10, [r1, #22*4]
+	add	r7, r7, lr
+	str	r6, [r1, #18*4]
+	
+	mov	r12, r6, ror #17
+	str	r7, [r1, #19*4]
+	eor	r12, r12, r6, ror #19
+	str	r8, [sp, #20*4]
+	eor	r12, r12, r6, lsr #10
+	ldr	r4, [r1, #23*4]
+	add	r8, r8, r12
+	ldr	r5, [r1, #24*4]
+	
+	mov	r9, r7, ror #17
+	str	r8, [r1, #20*4]
+	eor	r9, r9, r7, ror #19
+	str	r10, [sp, #21*4]
+	eor	r9, r9, r7, lsr #10
+	str	r4, [sp, #22*4]
+	
+	mov	r12, r8, ror #17
+	str	r9, [r1, #21*4]
+	eor	r12, r12, r8, ror #19
+	str	r5, [sp, #23*4]
+	eor	r12, r12, r8, lsr #10
+	mov	lr, r9, ror #17
+	add	r10, r10, r12
+	ldr	r11, [r1, #30*4]
+	
+	eor	lr, lr, r9, ror #19
+	str	r10, [r1, #22*4]
+	eor	lr, lr, r9, lsr #10
+	str	r11, [sp, #24*4]
+	add	r4, r4, lr
+	
+	mov	r12, r10, ror #17
+	str	r4, [r1, #23*4]
+	eor	r12, r12, r10, ror #19
+	mov	lr, r4, ror #17
+	eor	r12, r12, r10, lsr #10
+	eor	lr, lr, r4, ror #19
+	add	r5, r5, r12
+	eor	lr, lr, r4, lsr #10
+	str	r5, [r1, #24*4]
+	add	r6, r6, lr
+	
+	mov	r12, r5, ror #17
+	str	r6, [r1, #25*4]
+	eor	r12, r12, r5, ror #19
+	mov	lr, r6, ror #17
+	eor	r12, r12, r5, lsr #10
+	eor	lr, lr, r6, ror #19
+	add	r7, r7, r12
+	eor	lr, lr, r6, lsr #10
+	str	r7, [r1, #26*4]
+	add	r8, r8, lr
+	
+	mov	r12, r7, ror #17
+	str	r8, [r1, #27*4]
+	eor	r12, r12, r7, ror #19
+	mov	lr, r8, ror #17
+	eor	r12, r12, r7, lsr #10
+	eor	lr, lr, r8, ror #19
+	add	r9, r9, r12
+	eor	lr, lr, r8, lsr #10
+	str	r9, [r1, #28*4]
+	add	r10, r10, lr
+	
+	ldr	lr, [r1, #31*4]
+	mov	r12, r9, ror #17
+	str	r10, [r1, #29*4]
+	eor	r12, r12, r9, ror #19
+	str	lr, [sp, #25*4]
+	eor	r12, r12, r9, lsr #10
+	add	r11, r11, r12
+	add	r5, r5, lr
+	mov	r12, r10, ror #17
+	add	r4, r4, r11
+	
+	ldr	r11, [r1, #16*4]
+	eor	r12, r12, r10, ror #19
+	str	r4, [r1, #30*4]
+	eor	r12, r12, r10, lsr #10
+	add	r5, r5, r12
+	ldr	lr, [r1, #17*4]
+	
+sha256d_ms_extend_loop2:
+	sha256_extend_doubleround_body 16, r1, r6, r7, r4, r5
+	sha256_extend_doubleround_body 18, r1, r8, r9, r6, r7
+	sha256_extend_doubleround_body 20, r1, r10, r4, r8, r9
+	sha256_extend_doubleround_body 22, r1, r5, r6, r10, r4
+	sha256_extend_doubleround_body 24, r1, r7, r8, r5, r6
+	sha256_extend_doubleround_body 26, r1, r9, r10, r7, r8
+	sha256_extend_doubleround_body 28, r1, r4, r5, r9, r10
+	sha256_extend_doubleround_body 30, r1, r6, r7, r4, r5
+	sha256_extend_doubleround_body 32, r1, r8, r9, r6, r7
+	sha256_extend_doubleround_body 34, r1, r10, r4, r8, r9
+	sha256_extend_doubleround_body 36, r1, r5, r6, r10, r4
+	sha256_extend_doubleround_body 38, r1, r7, r8, r5, r6
+	sha256_extend_doubleround_body 40, r1, r9, r10, r7, r8
+	sha256_extend_doubleround_body 42, r1, r4, r5, r9, r10
+	bne	sha256d_ms_extend_coda2
+	sha256_extend_doubleround_body 44, r1, r6, r7, r4, r5
+	sha256_extend_doubleround_foot 46, r1, r8, r9, r6, r7
+	
+	ldr	r4,  [r3, #0*4]
+	ldr	r9,  [r3, #1*4]
+	ldr	r10, [r3, #2*4]
+	ldr	r11, [r3, #3*4]
+	ldr	r8,  [r3, #4*4]
+	ldr	r5,  [r3, #5*4]
+	ldr	r6,  [r3, #6*4]
+	ldr	r7,  [r3, #7*4]
+	b	sha256d_ms_main_loop1
+	
+sha256d_ms_main_loop2:
+	sha256_main_round  0, sha256d_ms_k, r1, r4, r5, r6, r7, r8, r9, r10, r11
+	sha256_main_round  1, sha256d_ms_k, r1, r7, r4, r5, r6, r11, r8, r9, r10
+	sha256_main_round  2, sha256d_ms_k, r1, r6, r7, r4, r5, r10, r11, r8, r9
+sha256d_ms_main_loop1:
+	sha256_main_round  3, sha256d_ms_k, r1, r5, r6, r7, r4, r9, r10, r11, r8
+	sha256_main_quadround  4, sha256d_ms_k, r1
+	sha256_main_quadround  8, sha256d_ms_k, r1
+	sha256_main_quadround 12, sha256d_ms_k, r1
+	sha256_main_quadround 16, sha256d_ms_k, r1
+	sha256_main_quadround 20, sha256d_ms_k, r1
+	sha256_main_quadround 24, sha256d_ms_k, r1
+	sha256_main_quadround 28, sha256d_ms_k, r1
+	b	sha256d_ms_k_over
+sha256d_ms_k:
+	sha256_k
+sha256d_ms_k_over:
+	sha256_main_quadround 32, sha256d_ms_k, r1
+	sha256_main_quadround 36, sha256d_ms_k, r1
+	sha256_main_quadround 40, sha256d_ms_k, r1
+	sha256_main_quadround 44, sha256d_ms_k, r1
+	sha256_main_quadround 48, sha256d_ms_k, r1
+	sha256_main_quadround 52, sha256d_ms_k, r1
+	sha256_main_round 56, sha256d_ms_k, r1, r4, r5, r6, r7, r8, r9, r10, r11
+	bne	sha256d_ms_finish
+	sha256_main_round 57, sha256d_ms_k, r1, r7, r4, r5, r6, r11, r8, r9, r10
+	sha256_main_round 58, sha256d_ms_k, r1, r6, r7, r4, r5, r10, r11, r8, r9
+	sha256_main_round 59, sha256d_ms_k, r1, r5, r6, r7, r4, r9, r10, r11, r8
+	sha256_main_quadround 60, sha256d_ms_k, r1
+	
+	ldmia	r2!, {r3, r12, lr}
+	add	r4, r4, r3
+	add	r5, r5, r12
+	add	r6, r6, lr
+	stmia	sp, {r4-r6}
+	ldmia	r2, {r3, r4, r5, r6, r12}
+	add	lr, sp, #3*4
+	add	r7, r7, r3
+	add	r8, r8, r4
+	add	r9, r9, r5
+	add	r10, r10, r6
+	add	r11, r11, r12
+	add	r12, sp, #18*4
+	stmia	lr!, {r7-r11}
+	
+	ldmia	r12, {r4-r11}
+	str	r4,  [r1, #18*4]
+	str	r5,  [r1, #19*4]
+	str	r6,  [r1, #20*4]
+	str	r7,  [r1, #22*4]
+	str	r8,  [r1, #23*4]
+	str	r9,  [r1, #24*4]
+	str	r10, [r1, #30*4]
+	str	r11, [r1, #31*4]
+	
+	mov	r3,  #0x80000000
+	mov	r4,  #0
+	mov	r5,  #0
+	mov	r6,  #0
+	mov	r7,  #0
+	mov	r8,  #0
+	mov	r9,  #0
+	mov	r10, #0x00000100
+	stmia	lr, {r3-r10}
+	
+	ldr	lr, [sp, #1*4]
+	movs	r1, sp
+	ldr	r4, [sp, #0*4]
+	
+	ldr	r11, [sp, #2*4]
+	mov	r12, lr, ror #7
+	eor	r12, r12, lr, ror #18
+	add	r5, lr, #0x00a00000
+	eor	r12, r12, lr, lsr #3
+	mov	lr, r11, ror #7
+	add	r4, r4, r12
+	eor	lr, lr, r11, ror #18
+	str	r4, [sp, #16*4]
+	eor	lr, lr, r11, lsr #3
+	mov	r12, r4, ror #17
+	add	r5, r5, lr
+	ldr	lr, [sp, #3*4]
+	
+	str	r5, [sp, #17*4]
+	eor	r12, r12, r4, ror #19
+	mov	r6, lr, ror #7
+	eor	r12, r12, r4, lsr #10
+	eor	r6, r6, lr, ror #18
+	add	r11, r11, r12
+	eor	r6, r6, lr, lsr #3
+	mov	r12, r5, ror #17
+	add	r6, r6, r11
+	ldr	r11, [sp, #4*4]
+	
+	str	r6, [sp, #18*4]
+	eor	r12, r12, r5, ror #19
+	mov	r7, r11, ror #7
+	eor	r12, r12, r5, lsr #10
+	eor	r7, r7, r11, ror #18
+	add	lr, lr, r12
+	eor	r7, r7, r11, lsr #3
+	mov	r12, r6, ror #17
+	add	r7, r7, lr
+	ldr	lr, [sp, #5*4]
+	
+	str	r7, [sp, #19*4]
+	eor	r12, r12, r6, ror #19
+	mov	r8, lr, ror #7
+	eor	r12, r12, r6, lsr #10
+	eor	r8, r8, lr, ror #18
+	add	r11, r11, r12
+	eor	r8, r8, lr, lsr #3
+	mov	r12, r7, ror #17
+	add	r8, r8, r11
+	ldr	r11, [sp, #6*4]
+	
+	str	r8, [sp, #20*4]
+	eor	r12, r12, r7, ror #19
+	mov	r9, r11, ror #7
+	eor	r12, r12, r7, lsr #10
+	eor	r9, r9, r11, ror #18
+	add	lr, lr, r12
+	eor	r9, r9, r11, lsr #3
+	mov	r12, r8, ror #17
+	add	r9, r9, lr
+	ldr	lr, [sp, #7*4]
+	
+	str	r9, [sp, #21*4]
+	eor	r12, r12, r8, ror #19
+	mov	r10, lr, ror #7
+	eor	r12, r12, r8, lsr #10
+	eor	r10, r10, lr, ror #18
+	add	r11, r11, r12
+	eor	r10, r10, lr, lsr #3
+	mov	r12, r9, ror #17
+	add	r11, r11, #0x00000100
+	add	lr, lr, r4
+	add	r10, r10, r11
+	
+	eor	r12, r12, r9, ror #19
+	str	r10, [sp, #22*4]
+	add	lr, lr, #0x11000000
+	eor	r12, r12, r9, lsr #10
+	add	lr, lr, r12
+	mov	r12, r10, ror #17
+	add	r4, lr, #0x00002000
+	eor	r12, r12, r10, ror #19
+	str	r4, [sp, #23*4]
+	add	r5, r5, #0x80000000
+	eor	r12, r12, r10, lsr #10
+	add	r5, r5, r12
+
+	mov	r12, r4, ror #17
+	str	r5, [sp, #24*4]
+	eor	r12, r12, r4, ror #19
+	mov	r11, r5, ror #17
+	eor	r12, r12, r4, lsr #10
+	eor	r11, r11, r5, ror #19
+	add	r6, r6, r12
+	eor	r11, r11, r5, lsr #10
+	str	r6, [sp, #25*4]
+	add	r7, r7, r11
+	
+	mov	r12, r6, ror #17
+	str	r7, [sp, #26*4]
+	eor	r12, r12, r6, ror #19
+	mov	r11, r7, ror #17
+	eor	r12, r12, r6, lsr #10
+	eor	r11, r11, r7, ror #19
+	add	r8, r8, r12
+	eor	r11, r11, r7, lsr #10
+	str	r8, [sp, #27*4]
+	add	r9, r9, r11
+	
+	mov	lr, r8, ror #17
+	mov	r12, r9, ror #17
+	str	r9, [sp, #28*4]
+	add	r4, r4, #0x00400000
+	eor	lr, lr, r8, ror #19
+	eor	r12, r12, r9, ror #19
+	eor	lr, lr, r8, lsr #10
+	eor	r12, r12, r9, lsr #10
+	add	r4, r4, #0x00000022
+	add	r10, r10, lr
+	add	r4, r4, r12
+	ldr	r11, [sp, #16*4]
+	
+	add	r5, r5, #0x00000100
+	str	r4, [sp, #30*4]
+	mov	lr, r11, ror #7
+	str	r10, [sp, #29*4]
+	mov	r12, r10, ror #17
+	eor	lr, lr, r11, ror #18
+	eor	r12, r12, r10, ror #19
+	eor	lr, lr, r11, lsr #3
+	eor	r12, r12, r10, lsr #10
+	add	r5, r5, lr
+	ldr	lr, [r1, #17*4]
+	add	r5, r5, r12
+	
+	b	sha256d_ms_extend_loop2
+	
+sha256d_ms_extend_coda2:
+	str	r5, [r1, #(44+15)*4]
+	mov	r12, r4, ror #17
+	add	r11, r11, r6
+	mov	r6, lr, ror #7
+	eor	r12, r12, r4, ror #19
+	eor	r6, r6, lr, ror #18
+	eor	r12, r12, r4, lsr #10
+	eor	r6, r6, lr, lsr #3
+	add	r12, r12, r11
+	add	r6, r6, r12
+	str	r6, [r1, #(44+16)*4]
+	
+	adr	r2, sha256d_ms_h
+	ldmia	r2, {r4-r11}
+	b	sha256d_ms_main_loop2
+
+sha256d_ms_h:
+	.long 0x6a09e667, 0xbb67ae85, 0x3c6ef372, 0xa54ff53a
+	.long 0x510e527f, 0x9b05688c, 0x1f83d9ab, 0x5be0cd19
+
+.macro sha256_main_round_red i, ka, rw, rd, re, rf, rg, rh
+	ldr	r12, [\rw, #(\i)*4]
+	and	r3, \rf, \re
+	bic	lr, \rg, \re
+	add	\rh, \rh, \rd
+	orr	lr, lr, r3
+	ldr	r3, \ka + (\i)*4
+	add	\rh, \rh, lr
+	eor	lr, \re, \re, ror #5
+	add	\rh, \rh, r12
+	eor	lr, lr, \re, ror #19
+	add	\rh, \rh, r3
+	add	\rh, \rh, lr, ror #6
+.endm
+	
+sha256d_ms_finish:
+	sha256_main_round_red 57, sha256d_ms_k, r1, r6, r11, r8, r9, r10
+	sha256_main_round_red 58, sha256d_ms_k, r1, r5, r10, r11, r8, r9
+	sha256_main_round_red 59, sha256d_ms_k, r1, r4, r9, r10, r11, r8
+	ldr	r5, [r2, #7*4]
+	sha256_main_round_red 60, sha256d_ms_k, r1, r7, r8, r9, r10, r11
+	
+	add	r11, r11, r5
+	str	r11, [r0, #7*4]
+	
+	add	sp, sp, #64*4
+#ifdef __thumb__
+	ldmfd	sp!, {r4-r11, lr}
+	bx	lr
+#else
+	ldmfd	sp!, {r4-r11, pc}
+#endif
+
+
+#ifdef __ARM_NEON__
+
+	.text
+	.code 32
+	.align 2
+	.globl sha256_init_4way
+	.globl _sha256_init_4way
+#ifdef __ELF__
+	.type sha256_init_4way, %function
+#endif
+sha256_init_4way:
+_sha256_init_4way:
+	adr	r12, sha256_4h
+	vldmia	r12, {q8-q15}
+	vstmia	r0, {q8-q15}
+	bx	lr
+	.align 4
+sha256_4h:
+	.long 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667
+	.long 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85
+	.long 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372
+	.long 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a
+	.long 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f
+	.long 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c
+	.long 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab
+	.long 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19
+
+.macro sha256_4k
+	.long 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98
+	.long 0x71374491, 0x71374491, 0x71374491, 0x71374491
+	.long 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf
+	.long 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5
+	.long 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b
+	.long 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1
+	.long 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4
+	.long 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5
+	.long 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98
+	.long 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01
+	.long 0x243185be, 0x243185be, 0x243185be, 0x243185be
+	.long 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3
+	.long 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74
+	.long 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe
+	.long 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7
+	.long 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174
+	.long 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1
+	.long 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786
+	.long 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6
+	.long 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc
+	.long 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f
+	.long 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa
+	.long 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc
+	.long 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da
+	.long 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152
+	.long 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d
+	.long 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8
+	.long 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7
+	.long 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3
+	.long 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147
+	.long 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351
+	.long 0x14292967, 0x14292967, 0x14292967, 0x14292967
+	.long 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85
+	.long 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138
+	.long 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc
+	.long 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13
+	.long 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354
+	.long 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb
+	.long 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e
+	.long 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85
+	.long 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1
+	.long 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b
+	.long 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70
+	.long 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3
+	.long 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819
+	.long 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624
+	.long 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585
+	.long 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070
+	.long 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116
+	.long 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08
+	.long 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c
+	.long 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5
+	.long 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3
+	.long 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a
+	.long 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f
+	.long 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3
+	.long 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee
+	.long 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f
+	.long 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814
+	.long 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208
+	.long 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa
+	.long 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb
+	.long 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7
+	.long 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2
+.endm
+
+.macro sha256_4way_extend_doubleround_core i, rr, rw, ra, rb, ry, rz
+	vadd.u32	q5, q5, \ra
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, \ry, #19
+	vshl.u32	q1, \ry, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	\ra, q6, #7
+	vshl.u32	q0, q6, #32-7
+	veor.u32	q4, q4, q1
+	veor.u32	\ra, \ra, q0
+	vshr.u32	q1, \ry, #10
+	vshr.u32	q0, q6, #18
+	veor.u32	q4, q4, q1
+	veor.u32	\ra, \ra, q0
+	vshl.u32	q1, q6, #32-18
+	vshr.u32	q0, q6, #3
+	veor.u32	\ra, \ra, q1
+	vadd.u32	q4, q4, q5
+	veor.u32	\ra, \ra, q0
+	vld1.u32	{q5}, [\rr]!
+	vadd.u32	\ra, \ra, q4
+
+	vshr.u32	q4, \rz, #17
+	vshl.u32	q0, \rz, #32-17
+	vadd.u32	q6, q6, \rb
+	vst1.u32	{\ra}, [\rw]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, \rz, #19
+	vshl.u32	q1, \rz, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	\rb, q5, #7
+	veor.u32	q4, q4, q1
+	vshl.u32	q0, q5, #32-7
+	vshr.u32	q1, \rz, #10
+	veor.u32	\rb, \rb, q0
+	vshr.u32	q0, q5, #18
+	veor.u32	q4, q4, q1
+	veor.u32	\rb, \rb, q0
+	vshl.u32	q1, q5, #32-18
+	vshr.u32	q0, q5, #3
+	veor.u32	\rb, \rb, q1
+	vadd.u32	q1, q6, q4
+	veor.u32	\rb, \rb, q0
+.endm
+
+.macro sha256_4way_extend_doubleround_head i, rr, rw, ra, rb, ry, rz
+	vld1.u32	{q6}, [\rr]!
+	vshr.u32	q4, \ry, #17
+	vshl.u32	q0, \ry, #32-17
+	sha256_4way_extend_doubleround_core \i, \rr, \rw, \ra, \rb, \ry, \rz
+	vld1.u32	{q6}, [\rr]!
+	vadd.u32	\rb, \rb, q1
+.endm
+
+.macro sha256_4way_extend_doubleround_body i, rr, rw, ra, rb, ry, rz
+	vshr.u32	q4, \ry, #17
+	vshl.u32	q0, \ry, #32-17
+	vst1.u32	{\rz}, [\rw]!
+	sha256_4way_extend_doubleround_core \i, \rr, \rw, \ra, \rb, \ry, \rz
+	vld1.u32	{q6}, [\rr]!
+	vadd.u32	\rb, \rb, q1
+.endm
+
+.macro sha256_4way_extend_doubleround_foot i, rr, rw, ra, rb, ry, rz
+	vshr.u32	q4, \ry, #17
+	vshl.u32	q0, \ry, #32-17
+	vst1.u32	{\rz}, [\rw]!
+	sha256_4way_extend_doubleround_core \i, \rr, \rw, \ra, \rb, \ry, \rz
+	vadd.u32	\rb, \rb, q1
+	vst1.u32	{\rb}, [\rw]!
+.endm
+
+.macro sha256_4way_main_round i, rk, rw, ra, rb, rc, rd, re, rf, rg, rh
+	vld1.u32	{q8}, [\rw]!
+	vand.u32	q9, \rf, \re
+	vbic.u32	q10, \rg, \re
+	vshr.u32	q11, \re, #5
+	vorr.u32	q10, q10, q9
+	vld1.u32	{q9}, [\rk]!
+	vadd.u32	\rh, \rh, q10
+	vshl.u32	q12, \re, #32-5
+	veor.u32	q10, \re, q11
+	vshr.u32	q11, \re, #19
+	veor.u32	q10, q10, q12
+	vshl.u32	q12, \re, #32-19
+	veor.u32	q10, q10, q11
+	vadd.u32	\rh, \rh, q8
+	veor.u32	q10, q10, q12
+	vadd.u32	\rh, \rh, q9
+	veor.u32	q9, \ra, \rb
+	vshr.u32	q11, q10, #6
+	vshl.u32	q13, q10, #32-6
+	vadd.u32	\rh, \rh, q11
+
+	vshr.u32	q11, \ra, #11
+	vshl.u32	q12, \ra, #32-11
+	veor.u32	q8, \ra, q11
+	vand.u32	q10, \ra, \rb
+	veor.u32	q8, q8, q12
+	vshr.u32	q11, \ra, #20
+	vshl.u32	q12, \ra, #32-20
+	veor.u32	q8, q8, q11
+	vand.u32	q9, q9, \rc
+	veor.u32	q8, q8, q12
+	vadd.u32	\rh, \rh, q13
+	veor.u32	q10, q10, q9
+	vshr.u32	q11, q8, #2
+	vshl.u32	q12, q8, #32-2
+	vadd.u32	q9, \rh, q10
+	vadd.u32	q12, q12, q11
+	vadd.u32	\rh, \rh, \rd
+	vadd.u32	\rd, q9, q12
+.endm
+
+.macro sha256_4way_main_quadround i, rk, rw
+	sha256_4way_main_round \i+0, \rk, \rw, q0, q1, q2, q3, q4, q5, q6, q7
+	sha256_4way_main_round \i+1, \rk, \rw, q3, q0, q1, q2, q7, q4, q5, q6
+	sha256_4way_main_round \i+2, \rk, \rw, q2, q3, q0, q1, q6, q7, q4, q5
+	sha256_4way_main_round \i+3, \rk, \rw, q1, q2, q3, q0, q5, q6, q7, q4
+.endm
+
+
+	.text
+	.code 32
+	.align 2
+	.globl sha256_transform_4way
+	.globl _sha256_transform_4way
+#ifdef __ELF__
+	.type sha256_transform_4way, %function
+#endif
+sha256_transform_4way:
+_sha256_transform_4way:
+	stmfd	sp!, {r4, lr}
+	vpush	{q4-q7}
+	mov	r12, sp
+	sub	sp, sp, #64*16
+	bic	sp, sp, #63
+	cmp	r2, #0
+	bne	sha256_transform_4way_swap
+	
+	vldmia	r1!, {q0-q7}
+	vstmia	sp, {q0-q7}
+	add	r3, sp, #8*16
+	vldmia	r1, {q8-q15}
+	vstmia	r3, {q8-q15}
+	b	sha256_transform_4way_extend
+
+sha256_transform_4way_swap:
+	vldmia	r1!, {q0-q7}
+	vrev32.8	q0, q0
+	vrev32.8	q1, q1
+	vrev32.8	q2, q2
+	vrev32.8	q3, q3
+	vldmia	r1, {q8-q15}
+	vrev32.8	q4, q4
+	vrev32.8	q5, q5
+	vrev32.8	q6, q6
+	vrev32.8	q7, q7
+	vstmia	sp, {q0-q7}
+	vrev32.8	q8, q8
+	vrev32.8	q9, q9
+	vrev32.8	q10, q10
+	vrev32.8	q11, q11
+	vrev32.8	q12, q12
+	vrev32.8	q13, q13
+	vrev32.8	q14, q14
+	vrev32.8	q15, q15
+	add	r3, sp, #8*16
+	vstmia	r3, {q8-q15}
+	
+sha256_transform_4way_extend:
+	add	r1, sp, #1*16
+	add	r2, sp, #16*16
+	vmov.u32	q5, q0
+	sha256_4way_extend_doubleround_head  0, r1, r2,  q9, q10, q14, q15
+	sha256_4way_extend_doubleround_body  2, r1, r2, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_body  4, r1, r2, q13, q14, q11, q12
+	sha256_4way_extend_doubleround_body  6, r1, r2, q15,  q9, q13, q14
+	sha256_4way_extend_doubleround_body  8, r1, r2, q10, q11, q15,  q9
+	sha256_4way_extend_doubleround_body 10, r1, r2, q12, q13, q10, q11
+	sha256_4way_extend_doubleround_body 12, r1, r2, q14, q15, q12, q13
+	sha256_4way_extend_doubleround_body 14, r1, r2,  q9, q10, q14, q15
+	sha256_4way_extend_doubleround_body 16, r1, r2, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_body 18, r1, r2, q13, q14, q11, q12
+	sha256_4way_extend_doubleround_body 20, r1, r2, q15,  q9, q13, q14
+	sha256_4way_extend_doubleround_body 22, r1, r2, q10, q11, q15,  q9
+	sha256_4way_extend_doubleround_body 24, r1, r2, q12, q13, q10, q11
+	sha256_4way_extend_doubleround_body 26, r1, r2, q14, q15, q12, q13
+	sha256_4way_extend_doubleround_body 28, r1, r2,  q9, q10, q14, q15
+	sha256_4way_extend_doubleround_body 30, r1, r2, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_body 32, r1, r2, q13, q14, q11, q12
+	sha256_4way_extend_doubleround_body 34, r1, r2, q15,  q9, q13, q14
+	sha256_4way_extend_doubleround_body 36, r1, r2, q10, q11, q15,  q9
+	sha256_4way_extend_doubleround_body 38, r1, r2, q12, q13, q10, q11
+	sha256_4way_extend_doubleround_body 40, r1, r2, q14, q15, q12, q13
+	sha256_4way_extend_doubleround_body 42, r1, r2,  q9, q10, q14, q15
+	sha256_4way_extend_doubleround_body 44, r1, r2, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_foot 46, r1, r2, q13, q14, q11, q12
+	
+	vldmia	r0, {q0-q7}
+	adr	r4, sha256_transform_4way_4k
+	b	sha256_transform_4way_4k_over
+	.align 4
+sha256_transform_4way_4k:
+	sha256_4k
+sha256_transform_4way_4k_over:
+	sha256_4way_main_quadround  0, r4, sp
+	sha256_4way_main_quadround  4, r4, sp
+	sha256_4way_main_quadround  8, r4, sp
+	sha256_4way_main_quadround 12, r4, sp
+	sha256_4way_main_quadround 16, r4, sp
+	sha256_4way_main_quadround 20, r4, sp
+	sha256_4way_main_quadround 24, r4, sp
+	sha256_4way_main_quadround 28, r4, sp
+	sha256_4way_main_quadround 32, r4, sp
+	sha256_4way_main_quadround 36, r4, sp
+	sha256_4way_main_quadround 40, r4, sp
+	sha256_4way_main_quadround 44, r4, sp
+	sha256_4way_main_quadround 48, r4, sp
+	sha256_4way_main_quadround 52, r4, sp
+	sha256_4way_main_quadround 56, r4, sp
+	sha256_4way_main_quadround 60, r4, sp
+	
+	vldmia	r0, {q8-q15}
+	vadd.u32	q0, q0, q8
+	vadd.u32	q1, q1, q9
+	vadd.u32	q2, q2, q10
+	vadd.u32	q3, q3, q11
+	vadd.u32	q4, q4, q12
+	vadd.u32	q5, q5, q13
+	vadd.u32	q6, q6, q14
+	vadd.u32	q7, q7, q15
+	vstmia	r0, {q0-q7}
+	
+	mov	sp, r12
+	vpop	{q4-q7}
+	ldmfd	sp!, {r4, pc}
+	
+
+	.text
+	.code 32
+	.align 2
+	.globl sha256d_ms_4way
+	.globl _sha256d_ms_4way
+#ifdef __ELF__
+	.type sha256d_ms_4way, %function
+#endif
+sha256d_ms_4way:
+_sha256d_ms_4way:
+	stmfd	sp!, {r4, lr}
+	vpush	{q4-q7}
+	mov	r12, sp
+	sub	sp, sp, #64*16
+	bic	sp, sp, #63
+	
+	add	r4, r1, #3*16
+	vld1.u32	{q6}, [r4]!
+	add	r1, r1, #18*16
+	vldmia	r1, {q11-q13}
+	cmp	r0, r0
+	
+	vshr.u32	q10, q6, #7
+	vshl.u32	q0, q6, #32-7
+	vshr.u32	q1, q6, #18
+	veor.u32	q10, q10, q0
+	vshl.u32	q0, q6, #32-18
+	veor.u32	q10, q10, q1
+	vshr.u32	q1, q6, #3
+	veor.u32	q10, q10, q0
+	vstmia	sp!, {q11-q13}
+	veor.u32	q4, q10, q1
+	vadd.u32	q12, q12, q6
+	vadd.u32	q11, q11, q4
+	
+	vshr.u32	q14, q12, #17
+	vshr.u32	q4, q11, #17
+	vshl.u32	q0, q11, #32-17
+	vst1.u32	{q11}, [r1]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q11, #19
+	vshl.u32	q1, q11, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q12}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q11, #10
+	vshl.u32	q0, q12, #32-17
+	veor.u32	q4, q4, q1
+	veor.u32	q14, q14, q0
+	vadd.u32	q13, q13, q4
+	vshr.u32	q0, q12, #19
+	vshl.u32	q1, q12, #32-19
+	veor.u32	q14, q14, q0
+	vst1.u32	{q13}, [r1]!
+	veor.u32	q14, q14, q1
+	vshr.u32	q1, q12, #10
+	
+	vshr.u32	q4, q13, #17
+	vshl.u32	q0, q13, #32-17
+	veor.u32	q14, q14, q1
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q13, #19
+	vshl.u32	q1, q13, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q14}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q13, #10
+	vld1.u32	{q15}, [r1]
+	veor.u32	q4, q4, q1
+	vst1.u32	{q15}, [sp]!
+	vadd.u32	q15, q15, q4
+	vshr.u32	q4, q14, #17
+	vshl.u32	q0, q14, #32-17
+	vshl.u32	q1, q14, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q14, #19
+	vst1.u32	{q15}, [r1]!
+	veor.u32	q4, q4, q0
+	vld1.u32	{q9}, [r1]
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q14, #10
+	vst1.u32	{q9}, [sp]!
+	veor.u32	q5, q4, q1
+	
+	vshr.u32	q4, q15, #17
+	vadd.u32	q9, q9, q5
+	vshl.u32	q0, q15, #32-17
+	vshl.u32	q1, q15, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q15, #19
+	vst1.u32	{q9}, [r1]!
+	veor.u32	q4, q4, q0
+	vld1.u32	{q10}, [r1]
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q15, #10
+	vst1.u32	{q10}, [sp]!
+	veor.u32	q4, q4, q1
+	vshl.u32	q0, q9, #32-17
+	vadd.u32	q10, q10, q4
+	vshr.u32	q4, q9, #17
+	vshl.u32	q1, q9, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q9, #19
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q9, #10
+	veor.u32	q4, q4, q0
+	vst1.u32	{q10}, [r1]!
+	veor.u32	q5, q4, q1
+	
+	vshr.u32	q4, q10, #17
+	vshl.u32	q0, q10, #32-17
+	vadd.u32	q11, q11, q5
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q10, #19
+	vshl.u32	q1, q10, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q11}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q10, #10
+	vshl.u32	q0, q11, #32-17
+	veor.u32	q2, q4, q1
+	vshr.u32	q4, q11, #17
+	vadd.u32	q12, q12, q2
+	vshl.u32	q1, q11, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q11, #19
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q11, #10
+	veor.u32	q4, q4, q0
+	vst1.u32	{q12}, [r1]!
+	veor.u32	q5, q4, q1
+	
+	vshr.u32	q4, q12, #17
+	vshl.u32	q0, q12, #32-17
+	vadd.u32	q13, q13, q5
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q12, #19
+	vshl.u32	q1, q12, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q13}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q12, #10
+	vshl.u32	q0, q13, #32-17
+	veor.u32	q2, q4, q1
+	vshr.u32	q4, q13, #17
+	vadd.u32	q14, q14, q2
+	vshl.u32	q1, q13, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q13, #19
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q13, #10
+	veor.u32	q4, q4, q0
+	vst1.u32	{q14}, [r1]!
+	veor.u32	q5, q4, q1
+	add	r4, r4, #12*16
+	
+	vshr.u32	q4, q14, #17
+	vshl.u32	q0, q14, #32-17
+	vadd.u32	q15, q15, q5
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q14, #19
+	vshl.u32	q1, q14, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q15}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q14, #10
+	vld1.u32	{q2}, [r1]
+	veor.u32	q4, q4, q1
+	vshl.u32	q0, q15, #32-17
+	vadd.u32	q9, q9, q4
+	vst1.u32	{q2}, [sp]!
+	vadd.u32	q9, q9, q2
+	vshr.u32	q4, q15, #17
+	vshr.u32	q2, q15, #19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q9}, [r1]!
+	vshl.u32	q1, q15, #32-19
+	veor.u32	q4, q4, q2
+	vshr.u32	q0, q15, #10
+	veor.u32	q4, q4, q1
+	vld1.u32	{q5-q6}, [r4]!
+	veor.u32	q4, q4, q0
+	vld1.u32	{q2}, [r1]
+	vadd.u32	q10, q10, q4
+	vst1.u32	{q2}, [sp]!
+	vadd.u32	q10, q10, q2
+	
+	sub	sp, sp, #8*16
+	
+sha256d_ms_4way_extend_loop2:
+	sha256_4way_extend_doubleround_body 16, r4, r1, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_body 18, r4, r1, q13, q14, q11, q12
+	sha256_4way_extend_doubleround_body 20, r4, r1, q15,  q9, q13, q14
+	sha256_4way_extend_doubleround_body 22, r4, r1, q10, q11, q15,  q9
+	sha256_4way_extend_doubleround_body 24, r4, r1, q12, q13, q10, q11
+	sha256_4way_extend_doubleround_body 26, r4, r1, q14, q15, q12, q13
+	sha256_4way_extend_doubleround_body 28, r4, r1,  q9, q10, q14, q15
+	sha256_4way_extend_doubleround_body 30, r4, r1, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_body 32, r4, r1, q13, q14, q11, q12
+	sha256_4way_extend_doubleround_body 34, r4, r1, q15,  q9, q13, q14
+	sha256_4way_extend_doubleround_body 36, r4, r1, q10, q11, q15,  q9
+	sha256_4way_extend_doubleround_body 38, r4, r1, q12, q13, q10, q11
+	sha256_4way_extend_doubleround_body 40, r4, r1, q14, q15, q12, q13
+	sha256_4way_extend_doubleround_body 42, r4, r1,  q9, q10, q14, q15
+	sha256_4way_extend_doubleround_body 44, r4, r1, q11, q12,  q9, q10
+	sha256_4way_extend_doubleround_foot 46, r4, r1, q13, q14, q11, q12
+	bne	sha256d_ms_4way_extend_coda2
+	
+	vldmia	r3!, {q4-q7}
+	vldmia	r3, {q0-q3}
+	vswp	q0, q4
+	adr	r3, sha256d_ms_4way_4k+3*16
+	sub r1, r1, #(64-3)*16
+	b	sha256d_ms_4way_main_loop1
+	
+	.align 4
+sha256d_ms_4way_4k:
+	sha256_4k
+	
+sha256d_ms_4way_main_loop2:
+	sha256_4way_main_round  0, r3, r1, q0, q1, q2, q3, q4, q5, q6, q7
+	sha256_4way_main_round  1, r3, r1, q3, q0, q1, q2, q7, q4, q5, q6
+	sha256_4way_main_round  2, r3, r1, q2, q3, q0, q1, q6, q7, q4, q5
+sha256d_ms_4way_main_loop1:
+	sha256_4way_main_round  3, r3, r1, q1, q2, q3, q0, q5, q6, q7, q4
+	sha256_4way_main_quadround  4, r3, r1
+	sha256_4way_main_quadround  8, r3, r1
+	sha256_4way_main_quadround 12, r3, r1
+	sha256_4way_main_quadround 16, r3, r1
+	sha256_4way_main_quadround 20, r3, r1
+	sha256_4way_main_quadround 24, r3, r1
+	sha256_4way_main_quadround 28, r3, r1
+	sha256_4way_main_quadround 32, r3, r1
+	sha256_4way_main_quadround 36, r3, r1
+	sha256_4way_main_quadround 40, r3, r1
+	sha256_4way_main_quadround 44, r3, r1
+	sha256_4way_main_quadround 48, r3, r1
+	sha256_4way_main_quadround 52, r3, r1
+	sha256_4way_main_round 56, r3, r1, q0, q1, q2, q3, q4, q5, q6, q7
+	bne	sha256d_ms_4way_finish
+	sha256_4way_main_round 57, r3, r1, q3, q0, q1, q2, q7, q4, q5, q6
+	sha256_4way_main_round 58, r3, r1, q2, q3, q0, q1, q6, q7, q4, q5
+	sha256_4way_main_round 59, r3, r1, q1, q2, q3, q0, q5, q6, q7, q4
+	sha256_4way_main_quadround 60, r3, r1
+	
+	vldmia	r2, {q8-q15}
+	vadd.u32	q0, q0, q8
+	vadd.u32	q1, q1, q9
+	vadd.u32	q2, q2, q10
+	vadd.u32	q3, q3, q11
+	vadd.u32	q4, q4, q12
+	vadd.u32	q5, q5, q13
+	vadd.u32	q6, q6, q14
+	vadd.u32	q7, q7, q15
+	
+	vldmia	sp, {q8-q15}
+	sub	r1, r1, #(64-18)*16
+	vstmia	r1, {q8-q10}
+	add	r1, r1, #4*16
+	vstmia	r1, {q11-q13}
+	add	r1, r1, #8*16
+	vstmia	r1, {q14-q15}
+	
+	vstmia	sp, {q0-q7}
+	vmov.u32	q8,  #0x80000000
+	vmov.u32	q9,  #0
+	vmov.u32	q10, #0
+	vmov.u32	q11, #0
+	vmov.u32	q12, #0
+	vmov.u32	q13, #0
+	vmov.u32	q14, #0
+	vmov.u32	q15, #0x00000100
+	add	r1, sp, #8*16
+	vstmia	r1!, {q8-q15}
+	adds	r4, sp, #2*16
+	
+	vshr.u32	q9, q1, #7
+	vshl.u32	q2, q1, #32-7
+	vshr.u32	q4, q1, #18
+	veor.u32	q9, q9, q2
+	vshl.u32	q3, q1, #32-18
+	veor.u32	q9, q9, q4
+	vshr.u32	q2, q1, #3
+	veor.u32	q9, q9, q3
+	vld1.u32	{q5}, [r4]!
+	veor.u32	q9, q9, q2
+	vmov.u32	q7, #0x00a00000
+	vadd.u32	q9, q9, q0
+	vshr.u32	q10, q5, #7
+	vshl.u32	q0, q5, #32-7
+	vshl.u32	q3, q5, #32-18
+	veor.u32	q10, q10, q0
+	vshr.u32	q0, q5, #18
+	veor.u32	q10, q10, q3
+	vst1.u32	{q9}, [r1]!
+	vadd.u32	q3, q1, q7
+	veor.u32	q10, q10, q0
+	vshr.u32	q0, q5, #3
+	vld1.u32	{q6}, [r4]!
+	veor.u32	q10, q10, q0
+	
+	vshr.u32	q4, q9, #17
+	vshl.u32	q0, q9, #32-17
+	vadd.u32	q10, q10, q3
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q9, #19
+	vshl.u32	q1, q9, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q11, q6, #7
+	vshl.u32	q0, q6, #32-7
+	veor.u32	q4, q4, q1
+	veor.u32	q11, q11, q0
+	vshr.u32	q1, q9, #10
+	vshr.u32	q0, q6, #18
+	veor.u32	q4, q4, q1
+	veor.u32	q11, q11, q0
+	vshl.u32	q1, q6, #32-18
+	vshr.u32	q0, q6, #3
+	veor.u32	q11, q11, q1
+	vadd.u32	q4, q4, q5
+	veor.u32	q11, q11, q0
+	vld1.u32	{q5}, [r4]!
+	vadd.u32	q11, q11, q4
+	vshr.u32	q4, q10, #17
+	vshl.u32	q0, q10, #32-17
+	vst1.u32	{q10}, [r1]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q10, #19
+	vshl.u32	q1, q10, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q12, q5, #7
+	veor.u32	q4, q4, q1
+	vshl.u32	q0, q5, #32-7
+	vshr.u32	q1, q10, #10
+	veor.u32	q12, q12, q0
+	vshr.u32	q0, q5, #18
+	veor.u32	q4, q4, q1
+	veor.u32	q12, q12, q0
+	vshl.u32	q1, q5, #32-18
+	vst1.u32	{q11}, [r1]!
+	veor.u32	q12, q12, q1
+	vshr.u32	q0, q5, #3
+	vadd.u32	q1, q6, q4
+	veor.u32	q12, q12, q0
+	
+	vshr.u32	q4, q11, #17
+	vshl.u32	q0, q11, #32-17
+	vadd.u32	q12, q12, q1
+	vld1.u32	{q6}, [r4]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q11, #19
+	vshl.u32	q1, q11, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q13, q6, #7
+	vshl.u32	q0, q6, #32-7
+	veor.u32	q4, q4, q1
+	veor.u32	q13, q13, q0
+	vshr.u32	q1, q11, #10
+	vshr.u32	q0, q6, #18
+	veor.u32	q4, q4, q1
+	veor.u32	q13, q13, q0
+	vshl.u32	q1, q6, #32-18
+	vshr.u32	q0, q6, #3
+	veor.u32	q13, q13, q1
+	vadd.u32	q4, q4, q5
+	veor.u32	q13, q13, q0
+	vld1.u32	{q5}, [r4]!
+	vadd.u32	q13, q13, q4
+	vshr.u32	q4, q12, #17
+	vshl.u32	q0, q12, #32-17
+	vst1.u32	{q12}, [r1]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q12, #19
+	vshl.u32	q1, q12, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q14, q5, #7
+	veor.u32	q4, q4, q1
+	vshl.u32	q0, q5, #32-7
+	vshr.u32	q1, q12, #10
+	veor.u32	q14, q14, q0
+	vshr.u32	q0, q5, #18
+	veor.u32	q4, q4, q1
+	veor.u32	q14, q14, q0
+	vshl.u32	q1, q5, #32-18
+	vst1.u32	{q13}, [r1]!
+	veor.u32	q14, q14, q1
+	vshr.u32	q0, q5, #3
+	vadd.u32	q1, q6, q4
+	veor.u32	q14, q14, q0
+	
+	vshr.u32	q4, q13, #17
+	vshl.u32	q0, q13, #32-17
+	vadd.u32	q14, q14, q1
+	vld1.u32	{q6}, [r4]!
+	vadd.u32	q5, q5, q15
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q13, #19
+	vshl.u32	q1, q13, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q15, q6, #7
+	vshl.u32	q0, q6, #32-7
+	veor.u32	q4, q4, q1
+	veor.u32	q15, q15, q0
+	vshr.u32	q1, q13, #10
+	vshr.u32	q0, q6, #18
+	veor.u32	q4, q4, q1
+	veor.u32	q15, q15, q0
+	vshl.u32	q1, q6, #32-18
+	vshr.u32	q0, q6, #3
+	veor.u32	q15, q15, q1
+	vadd.u32	q4, q4, q5
+	veor.u32	q15, q15, q0
+	vmov.u32	q5, #0x80000000
+	vadd.u32	q15, q15, q4
+	vshr.u32	q4, q14, #17
+	vshl.u32	q0, q14, #32-17
+	vadd.u32	q6, q6, q9
+	vst1.u32	{q14}, [r1]!
+	vmov.u32	q7, #0x11000000
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q14, #19
+	vshl.u32	q1, q14, #32-19
+	vadd.u32	q6, q6, q7
+	vmov.u32	q2, #0x00002000
+	veor.u32	q4, q4, q0
+	vst1.u32	{q15}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q14, #10
+	vadd.u32	q6, q6, q2
+	veor.u32	q1, q4, q1
+	add	r4, r4, #8*16
+	
+	vshr.u32	q4, q15, #17
+	vshl.u32	q0, q15, #32-17
+	vadd.u32	q9, q6, q1
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q15, #19
+	vshl.u32	q1, q15, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q9}, [r1]!
+	vadd.u32	q5, q5, q10
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q15, #10
+	vshl.u32	q0, q9, #32-17
+	veor.u32	q10, q4, q1
+	vshr.u32	q4, q9, #17
+	vadd.u32	q10, q10, q5
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q9, #19
+	vshl.u32	q1, q9, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q9, #10
+	veor.u32	q4, q4, q1
+	vst1.u32	{q10}, [r1]!
+	veor.u32	q1, q4, q0
+	
+	vshr.u32	q4, q10, #17
+	vshl.u32	q0, q10, #32-17
+	vadd.u32	q11, q11, q1
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q10, #19
+	vshl.u32	q1, q10, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q11}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q10, #10
+	vshl.u32	q0, q11, #32-17
+	veor.u32	q1, q4, q1
+	vshr.u32	q4, q11, #17
+	vadd.u32	q12, q12, q1
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q11, #19
+	vshl.u32	q1, q11, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q11, #10
+	veor.u32	q4, q4, q1
+	vst1.u32	{q12}, [r1]!
+	veor.u32	q1, q4, q0
+	
+	vshr.u32	q4, q12, #17
+	vshl.u32	q0, q12, #32-17
+	vadd.u32	q13, q13, q1
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q12, #19
+	vshl.u32	q1, q12, #32-19
+	veor.u32	q4, q4, q0
+	vst1.u32	{q13}, [r1]!
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q12, #10
+	vshl.u32	q0, q13, #32-17
+	veor.u32	q1, q4, q1
+	vshr.u32	q4, q13, #17
+	vadd.u32	q14, q14, q1
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q13, #19
+	vshl.u32	q1, q13, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q13, #10
+	veor.u32	q4, q4, q1
+	vst1.u32	{q14}, [r1]!
+	veor.u32	q4, q4, q0
+	vmov.u32	q6, #0x00000100
+	vadd.u32	q15, q15, q4
+	
+	vshr.u32	q4, q14, #17
+	vshl.u32	q0, q14, #32-17
+	vmov.u32	q7, #0x00400000
+	vst1.u32	{q15}, [r1]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q14, #19
+	vshl.u32	q1, q14, #32-19
+	veor.u32	q4, q4, q0
+	vadd.u32	q9, q9, q7
+	veor.u32	q4, q4, q1
+	vshr.u32	q1, q14, #10
+	vmov.u32	q2, #0x00000022
+	veor.u32	q4, q4, q1
+	vadd.u32	q9, q9, q2
+	vld1.u32	{q5}, [r4]!
+	vadd.u32	q9, q9, q4
+	vshr.u32	q4, q15, #17
+	vshl.u32	q0, q15, #32-17
+	vadd.u32	q6, q6, q10
+	vst1.u32	{q9}, [r1]!
+	veor.u32	q4, q4, q0
+	vshr.u32	q0, q15, #19
+	vshl.u32	q1, q15, #32-19
+	veor.u32	q4, q4, q0
+	vshr.u32	q10, q5, #7
+	veor.u32	q4, q4, q1
+	vshl.u32	q0, q5, #32-7
+	vshr.u32	q1, q15, #10
+	veor.u32	q10, q10, q0
+	vshr.u32	q0, q5, #18
+	veor.u32	q4, q4, q1
+	veor.u32	q10, q10, q0
+	vshl.u32	q1, q5, #32-18
+	vshr.u32	q0, q5, #3
+	veor.u32	q10, q10, q1
+	vadd.u32	q1, q6, q4
+	veor.u32	q10, q10, q0
+	vld1.u32	{q6}, [r4]!
+	vadd.u32	q10, q10, q1
+	
+	b	sha256d_ms_4way_extend_loop2
+	
+	.align 4
+sha256d_ms_4way_4h:
+	.long 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667
+	.long 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85
+	.long 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372
+	.long 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a
+	.long 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f
+	.long 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c
+	.long 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab
+	.long 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19
+	
+sha256d_ms_4way_extend_coda2:
+	adr	r4, sha256d_ms_4way_4h
+	mov	r1, sp
+	vldmia	r4, {q0-q7}
+	vmov.u32	q15, q7
+	sub	r3, r3, #64*16
+	b	sha256d_ms_4way_main_loop2
+
+.macro sha256_4way_main_round_red i, rk, rw, rd, re, rf, rg, rh
+	vld1.u32	{q8}, [\rw]!
+	vand.u32	q9, \rf, \re
+	vbic.u32	q10, \rg, \re
+	vshr.u32	q11, \re, #5
+	vorr.u32	q10, q10, q9
+	vshl.u32	q12, \re, #32-5
+	vadd.u32	\rh, \rh, q10
+	veor.u32	q10, \re, q11
+	vshr.u32	q11, \re, #19
+	veor.u32	q10, q10, q12
+	vshl.u32	q12, \re, #32-19
+	veor.u32	q10, q10, q11
+	vadd.u32	\rh, \rh, q8
+	veor.u32	q10, q10, q12
+	vld1.u32	{q9}, [\rk]!
+	vadd.u32	\rh, \rh, \rd
+	vshr.u32	q11, q10, #6
+	vadd.u32	\rh, \rh, q9
+	vshl.u32	q13, q10, #32-6
+	vadd.u32	\rh, \rh, q11
+	vadd.u32	\rh, \rh, q13
+.endm
+
+sha256d_ms_4way_finish:
+	sha256_4way_main_round_red 57, r3, r1, q2, q7, q4, q5, q6
+	sha256_4way_main_round_red 58, r3, r1, q1, q6, q7, q4, q5
+	sha256_4way_main_round_red 59, r3, r1, q0, q5, q6, q7, q4
+	sha256_4way_main_round_red 60, r3, r1, q3, q4, q5, q6, q7
+	
+	vadd.u32	q7, q7, q15
+	add	r0, r0, #7*16
+	vst1.u32	{q7}, [r0]
+	
+	mov	sp, r12
+	vpop	{q4-q7}
+	ldmfd	sp!, {r4, pc}
+
+
+	.text
+	.code 32
+	.align 2
+	.globl sha256_use_4way
+	.globl _sha256_use_4way
+#ifdef __ELF__
+	.type sha256_use_4way, %function
+#endif
+sha256_use_4way:
+_sha256_use_4way:
+	mov	r0, #1
+	bx	lr
+
+#endif /* __ARM_NEON__ */
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-ppc.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-ppc.S
new file mode 100644
index 0000000..a0b60d2
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-ppc.S
@@ -0,0 +1,2007 @@
+/*
+ * Copyright 2014-2015 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(USE_ASM) && (defined(__powerpc__) || defined(__ppc__) || defined(__PPC__))
+
+#ifdef __APPLE__
+
+#define HI(name) ha16(name)
+#define LO(name) lo16(name)
+
+#else
+
+#define HI(name) name@ha
+#define LO(name) name@l
+
+#define r0 0
+#define r1 1
+#define r2 2
+#define r3 3
+#define r4 4
+#define r5 5
+#define r6 6
+#define r7 7
+#define r8 8
+#define r9 9
+#define r10 10
+#define r11 11
+#define r12 12
+#define r13 13
+#define r14 14
+#define r15 15
+#define r16 16
+#define r17 17
+#define r18 18
+#define r19 19
+#define r20 20
+#define r21 21
+#define r22 22
+#define r23 23
+#define r24 24
+#define r25 25
+#define r26 26
+#define r27 27
+#define r28 28
+#define r29 29
+#define r30 30
+#define r31 31
+
+#ifdef __ALTIVEC__
+#define v0 0
+#define v1 1
+#define v2 2
+#define v3 3
+#define v4 4
+#define v5 5
+#define v6 6
+#define v7 7
+#define v8 8
+#define v9 9
+#define v10 10
+#define v11 11
+#define v12 12
+#define v13 13
+#define v14 14
+#define v15 15
+#define v16 16
+#define v17 17
+#define v18 18
+#define v19 19
+#define v20 20
+#define v21 21
+#define v22 22
+#define v23 23
+#define v24 24
+#define v25 25
+#define v26 26
+#define v27 27
+#define v28 28
+#define v29 29
+#define v30 30
+#define v31 31
+#endif
+
+#endif
+
+#if !(defined(__powerpc64__) || defined(__ppc64__) || defined(__PPC64__) || \
+	defined(__64BIT__) || defined(_LP64) || defined(__LP64__))
+#define ld lwz
+#define std stw
+#define stdu stwu
+#define stdux stwux
+#endif
+
+
+#ifdef _AIX
+	.csect .text[RO]
+#else
+	.data
+#endif
+	.align 2
+sha256_h:
+	.long 0x6a09e667, 0xbb67ae85, 0x3c6ef372, 0xa54ff53a
+	.long 0x510e527f, 0x9b05688c, 0x1f83d9ab, 0x5be0cd19
+
+	.align 2
+sha256_k:
+	.long 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5
+	.long 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5
+	.long 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3
+	.long 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174
+	.long 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc
+	.long 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da
+	.long 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7
+	.long 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967
+	.long 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13
+	.long 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85
+	.long 0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3
+	.long 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070
+	.long 0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5
+	.long 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3
+	.long 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208
+	.long 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
+
+#ifdef _AIX
+	.toc
+T.sha256_h:
+	.tc sha256_h[TC], sha256_h
+T.sha256_k:
+	.tc sha256_k[TC], sha256_k
+#endif
+
+
+.macro sha256_extend_doubleround i, rw, wo, ra, rb, ry, rz
+	lwz	r14, \wo+(\i+1)*4(\rw)
+	rotrwi	r12, \ry, 17
+	rotrwi	r13, \ry, 19
+	add	r11, r11, \ra
+	xor	r12, r12, r13
+	srwi	r13, \ry, 10
+	rotrwi	\ra, r14, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r14, 18
+	add	r12, r12, r11
+	xor	\ra, \ra, r13
+	srwi	r13, r14, 3
+	lwz	r11, \wo+(\i+2)*4(\rw)
+	xor	\ra, \ra, r13
+	rotrwi	r13, \rz, 19
+	add	\ra, \ra, r12
+
+	rotrwi	r12, \rz, 17
+	add	r14, r14, \rb
+	xor	r12, r12, r13
+	srwi	r13, \rz, 10
+	rotrwi	\rb, r11, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r11, 18
+	stw	\ra, \wo+(\i+16)*4(\rw)
+	xor	\rb, \rb, r13
+	srwi	r13, r11, 3
+	add	r14, r14, r12
+	xor	\rb, \rb, r13
+	add	\rb, \rb, r14
+	stw	\rb, \wo+(\i+17)*4(\rw)
+.endm
+
+
+.macro sha256_main_round i, rk, rw, wo, ra, rb, rc, rd, re, rf, rg, rh
+	lwz	r12, \wo+(\i)*4(\rw)
+	and	r13, \rf, \re
+	andc	r14, \rg, \re
+	lwz	r15, (\i)*4(\rk)
+	or	r14, r14, r13
+	rotrwi	r13, \re, 5
+	add	\rh, \rh, r14
+	xor	r14, \re, r13
+	rotrwi	r13, \re, 19
+	add	\rh, \rh, r12
+	xor	r14, r14, r13
+	add	\rh, \rh, r15
+	rotrwi	r13, r14, 6
+	xor	r15, \ra, \rb
+	add	\rh, \rh, r13
+
+	rotrwi	r13, \ra, 11
+	and	r15, r15, \rc
+	xor	r12, \ra, r13
+	rotrwi	r13, \ra, 20
+	and	r14, \ra, \rb
+	xor	r12, r12, r13
+	xor	r14, r14, r15
+	rotrwi	r13, r12, 2
+	add	r15, \rh, r14
+	add	\rh, \rh, \rd
+	add	\rd, r15, r13
+.endm
+
+.macro sha256_main_quadround i, rk, rw, wo
+	sha256_main_round \i+0, \rk, \rw, \wo, r4, r5, r6, r7, r8, r9, r10, r11
+	sha256_main_round \i+1, \rk, \rw, \wo, r7, r4, r5, r6, r11, r8, r9, r10
+	sha256_main_round \i+2, \rk, \rw, \wo, r6, r7, r4, r5, r10, r11, r8, r9
+	sha256_main_round \i+3, \rk, \rw, \wo, r5, r6, r7, r4, r9, r10, r11, r8
+.endm
+
+
+#ifdef _AIX
+	.csect .text[PR]
+#else
+	.text
+#endif
+	.align 2
+	.globl sha256_transform
+	.globl _sha256_transform
+	.globl .sha256_transform
+#ifdef __ELF__
+	.type sha256_transform, %function
+#endif
+sha256_transform:
+_sha256_transform:
+.sha256_transform:
+	stdu	r1, -76*4(r1)
+	cmpwi	0, r5, 0
+	std	r13, 2*4(r1)
+	std	r14, 4*4(r1)
+	std	r15, 6*4(r1)
+	std	r16, 72*4(r1)
+	
+	bne	0, sha256_transform_swap
+	
+	lwz	r11, 0*4(r4)
+	lwz	r14, 1*4(r4)
+	lwz	r15, 2*4(r4)
+	lwz	r7, 3*4(r4)
+	lwz	r8, 4*4(r4)
+	lwz	r9, 5*4(r4)
+	lwz	r10, 6*4(r4)
+	lwz	r0, 7*4(r4)
+	lwz	r12, 8*4(r4)
+	lwz	r13, 9*4(r4)
+	lwz	r5, 10*4(r4)
+	lwz	r6, 11*4(r4)
+	stw	r11, 8*4+0*4(r1)
+	stw	r14, 8*4+1*4(r1)
+	stw	r15, 8*4+2*4(r1)
+	stw	r7, 8*4+3*4(r1)
+	stw	r8, 8*4+4*4(r1)
+	stw	r9, 8*4+5*4(r1)
+	stw	r10, 8*4+6*4(r1)
+	stw	r0, 8*4+7*4(r1)
+	stw	r12, 8*4+8*4(r1)
+	stw	r13, 8*4+9*4(r1)
+	stw	r5, 8*4+10*4(r1)
+	stw	r6, 8*4+11*4(r1)
+	lwz	r7, 12*4(r4)
+	lwz	r8, 13*4(r4)
+	lwz	r9, 14*4(r4)
+	lwz	r10, 15*4(r4)
+	mr	r4, r13
+	stw	r7, 8*4+12*4(r1)
+	stw	r8, 8*4+13*4(r1)
+	stw	r9, 8*4+14*4(r1)
+	stw	r10, 8*4+15*4(r1)
+	b	sha256_transform_extend
+	
+sha256_transform_swap:
+	li	r13, 1*4
+	li	r14, 2*4
+	li	r15, 3*4
+	lwbrx	r11, 0, r4
+	lwbrx	r7, r4, r13
+	lwbrx	r8, r4, r14
+	lwbrx	r9, r4, r15
+	addi	r4, r4, 4*4
+	stw	r11, 8*4+0*4(r1)
+	stw	r7, 8*4+1*4(r1)
+	stw	r8, 8*4+2*4(r1)
+	stw	r9, 8*4+3*4(r1)
+	lwbrx	r7, 0, r4
+	lwbrx	r8, r4, r13
+	lwbrx	r9, r4, r14
+	lwbrx	r10, r4, r15
+	addi	r4, r4, 4*4
+	stw	r7, 8*4+4*4(r1)
+	stw	r8, 8*4+5*4(r1)
+	stw	r9, 8*4+6*4(r1)
+	stw	r10, 8*4+7*4(r1)
+	lwbrx	r8, 0, r4
+	lwbrx	r12, r4, r13
+	lwbrx	r5, r4, r14
+	lwbrx	r6, r4, r15
+	addi	r4, r4, 4*4
+	stw	r8, 8*4+8*4(r1)
+	stw	r12, 8*4+9*4(r1)
+	stw	r5, 8*4+10*4(r1)
+	stw	r6, 8*4+11*4(r1)
+	lwbrx	r7, 0, r4
+	lwbrx	r8, r4, r13
+	lwbrx	r9, r4, r14
+	lwbrx	r10, r4, r15
+	mr	r4, r12
+	stw	r7, 8*4+12*4(r1)
+	stw	r8, 8*4+13*4(r1)
+	stw	r9, 8*4+14*4(r1)
+	stw	r10, 8*4+15*4(r1)
+	
+sha256_transform_extend:
+	sha256_extend_doubleround  0, r1, 8*4, r4, r5, r9, r10
+	sha256_extend_doubleround  2, r1, 8*4, r6, r7, r4, r5
+	sha256_extend_doubleround  4, r1, 8*4, r8, r9, r6, r7
+	sha256_extend_doubleround  6, r1, 8*4, r10, r4, r8, r9
+	sha256_extend_doubleround  8, r1, 8*4, r5, r6, r10, r4
+	sha256_extend_doubleround 10, r1, 8*4, r7, r8, r5, r6
+	sha256_extend_doubleround 12, r1, 8*4, r9, r10, r7, r8
+	sha256_extend_doubleround 14, r1, 8*4, r4, r5, r9, r10
+	sha256_extend_doubleround 16, r1, 8*4, r6, r7, r4, r5
+	sha256_extend_doubleround 18, r1, 8*4, r8, r9, r6, r7
+	sha256_extend_doubleround 20, r1, 8*4, r10, r4, r8, r9
+	sha256_extend_doubleround 22, r1, 8*4, r5, r6, r10, r4
+	sha256_extend_doubleround 24, r1, 8*4, r7, r8, r5, r6
+	sha256_extend_doubleround 26, r1, 8*4, r9, r10, r7, r8
+	sha256_extend_doubleround 28, r1, 8*4, r4, r5, r9, r10
+	sha256_extend_doubleround 30, r1, 8*4, r6, r7, r4, r5
+	sha256_extend_doubleround 32, r1, 8*4, r8, r9, r6, r7
+	sha256_extend_doubleround 34, r1, 8*4, r10, r4, r8, r9
+	sha256_extend_doubleround 36, r1, 8*4, r5, r6, r10, r4
+	sha256_extend_doubleround 38, r1, 8*4, r7, r8, r5, r6
+	sha256_extend_doubleround 40, r1, 8*4, r9, r10, r7, r8
+	sha256_extend_doubleround 42, r1, 8*4, r4, r5, r9, r10
+	sha256_extend_doubleround 44, r1, 8*4, r6, r7, r4, r5
+	sha256_extend_doubleround 46, r1, 8*4, r8, r9, r6, r7
+	
+	lwz	r4, 0*4(r3)
+	lwz	r5, 1*4(r3)
+	lwz	r6, 2*4(r3)
+	lwz	r7, 3*4(r3)
+	lwz	r8, 4*4(r3)
+	lwz	r9, 5*4(r3)
+	lwz	r10, 6*4(r3)
+	lwz	r11, 7*4(r3)
+#ifdef _AIX
+	ld	r16, T.sha256_k(r2)
+#else
+	lis	r16, HI(sha256_k)
+	addi	r16, r16, LO(sha256_k)
+#endif
+	sha256_main_quadround  0, r16, r1, 8*4
+	sha256_main_quadround  4, r16, r1, 8*4
+	sha256_main_quadround  8, r16, r1, 8*4
+	sha256_main_quadround 12, r16, r1, 8*4
+	sha256_main_quadround 16, r16, r1, 8*4
+	sha256_main_quadround 20, r16, r1, 8*4
+	sha256_main_quadround 24, r16, r1, 8*4
+	sha256_main_quadround 28, r16, r1, 8*4
+	sha256_main_quadround 32, r16, r1, 8*4
+	sha256_main_quadround 36, r16, r1, 8*4
+	sha256_main_quadround 40, r16, r1, 8*4
+	sha256_main_quadround 44, r16, r1, 8*4
+	sha256_main_quadround 48, r16, r1, 8*4
+	sha256_main_quadround 52, r16, r1, 8*4
+	sha256_main_quadround 56, r16, r1, 8*4
+	sha256_main_quadround 60, r16, r1, 8*4
+	
+	lwz	r12, 0*4(r3)
+	lwz	r13, 1*4(r3)
+	lwz	r14, 2*4(r3)
+	lwz	r15, 3*4(r3)
+	add	r4, r4, r12
+	add	r5, r5, r13
+	add	r6, r6, r14
+	add	r7, r7, r15
+	stw	r4, 0*4(r3)
+	stw	r5, 1*4(r3)
+	stw	r6, 2*4(r3)
+	stw	r7, 3*4(r3)
+	lwz	r12, 4*4(r3)
+	lwz	r13, 5*4(r3)
+	lwz	r14, 6*4(r3)
+	lwz	r15, 7*4(r3)
+	add	r8, r8, r12
+	add	r9, r9, r13
+	add	r10, r10, r14
+	add	r11, r11, r15
+	stw	r8, 4*4(r3)
+	stw	r9, 5*4(r3)
+	stw	r10, 6*4(r3)
+	stw	r11, 7*4(r3)
+	
+	ld	r13, 2*4(r1)
+	ld	r14, 4*4(r1)
+	ld	r15, 6*4(r1)
+	ld	r16, 72*4(r1)
+	addi	r1, r1, 76*4
+	blr
+
+
+	.align 2
+	.globl sha256d_ms
+	.globl _sha256d_ms
+	.globl .sha256d_ms
+#ifdef __ELF__
+	.type sha256d_ms, %function
+#endif
+sha256d_ms:
+_sha256d_ms:
+.sha256d_ms:
+	stdu	r1, -80*4(r1)
+	std	r13, 2*4(r1)
+	std	r14, 4*4(r1)
+	std	r15, 6*4(r1)
+	std	r16, 72*4(r1)
+	std	r17, 74*4(r1)
+	std	r18, 76*4(r1)
+	
+	mr	r17, r4
+	mr	r18, r5
+	mr	r16, r6
+	
+	lwz	r14, 3*4(r17)
+	lwz	r6, 18*4(r17)
+	lwz	r7, 19*4(r17)
+	
+	rotrwi	r12, r14, 7
+	rotrwi	r13, r14, 18
+	stw	r6, 8*4+18*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r14, 3
+	stw	r7, 8*4+19*4(r1)
+	xor	r12, r12, r13
+	lwz	r8, 20*4(r17)
+	add	r6, r6, r12
+	lwz	r10, 22*4(r17)
+	add	r7, r7, r14
+	stw	r6, 18*4(r17)
+	
+	rotrwi	r12, r6, 17
+	rotrwi	r13, r6, 19
+	stw	r7, 19*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r6, 10
+	stw	r8, 8*4+20*4(r1)
+	xor	r12, r12, r13
+	lwz	r4, 23*4(r17)
+	add	r8, r8, r12
+	lwz	r5, 24*4(r17)
+	
+	rotrwi	r9, r7, 17
+	rotrwi	r13, r7, 19
+	stw	r8, 20*4(r17)
+	xor	r9, r9, r13
+	srwi	r13, r7, 10
+	stw	r10, 8*4+21*4(r1)
+	xor	r9, r9, r13
+	stw	r4, 8*4+22*4(r1)
+	
+	rotrwi	r12, r8, 17
+	rotrwi	r13, r8, 19
+	stw	r9, 21*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r8, 10
+	stw	r5, 8*4+23*4(r1)
+	xor	r12, r12, r13
+	rotrwi	r14, r9, 17
+	rotrwi	r13, r9, 19
+	add	r10, r10, r12
+	lwz	r11, 30*4(r17)
+	
+	xor	r14, r14, r13
+	srwi	r13, r9, 10
+	stw	r10, 22*4(r17)
+	xor	r14, r14, r13
+	stw	r11, 8*4+24*4(r1)
+	add	r4, r4, r14
+	
+	rotrwi	r12, r10, 17
+	rotrwi	r13, r10, 19
+	stw	r4, 23*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r10, 10
+	rotrwi	r14, r4, 17
+	xor	r12, r12, r13
+	rotrwi	r13, r4, 19
+	xor	r14, r14, r13
+	srwi	r13, r4, 10
+	add	r5, r5, r12
+	xor	r14, r14, r13
+	stw	r5, 24*4(r17)
+	add	r6, r6, r14
+	
+	rotrwi	r12, r5, 17
+	rotrwi	r13, r5, 19
+	stw	r6, 25*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r5, 10
+	rotrwi	r14, r6, 17
+	xor	r12, r12, r13
+	rotrwi	r13, r6, 19
+	xor	r14, r14, r13
+	srwi	r13, r6, 10
+	add	r7, r7, r12
+	xor	r14, r14, r13
+	stw	r7, 26*4(r17)
+	add	r8, r8, r14
+	
+	rotrwi	r12, r7, 17
+	rotrwi	r13, r7, 19
+	stw	r8, 27*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r7, 10
+	rotrwi	r14, r8, 17
+	xor	r12, r12, r13
+	rotrwi	r13, r8, 19
+	xor	r14, r14, r13
+	srwi	r13, r8, 10
+	add	r9, r9, r12
+	xor	r14, r14, r13
+	stw	r9, 28*4(r17)
+	add	r10, r10, r14
+	
+	lwz	r14, 31*4(r17)
+	rotrwi	r12, r9, 17
+	rotrwi	r13, r9, 19
+	stw	r10, 29*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r9, 10
+	stw	r14, 8*4+25*4(r1)
+	xor	r12, r12, r13
+	add	r11, r11, r12
+	add	r5, r5, r14
+	rotrwi	r12, r10, 17
+	rotrwi	r13, r10, 19
+	add	r4, r4, r11
+	
+	lwz	r11, 16*4(r17)
+	xor	r12, r12, r13
+	srwi	r13, r10, 10
+	stw	r4, 30*4(r17)
+	xor	r12, r12, r13
+	add	r5, r5, r12
+	stw	r5, 31*4(r17)
+	
+	sha256_extend_doubleround 16, r17, 0, r6, r7, r4, r5
+	sha256_extend_doubleround 18, r17, 0, r8, r9, r6, r7
+	sha256_extend_doubleround 20, r17, 0, r10, r4, r8, r9
+	sha256_extend_doubleround 22, r17, 0, r5, r6, r10, r4
+	sha256_extend_doubleround 24, r17, 0, r7, r8, r5, r6
+	sha256_extend_doubleround 26, r17, 0, r9, r10, r7, r8
+	sha256_extend_doubleround 28, r17, 0, r4, r5, r9, r10
+	sha256_extend_doubleround 30, r17, 0, r6, r7, r4, r5
+	sha256_extend_doubleround 32, r17, 0, r8, r9, r6, r7
+	sha256_extend_doubleround 34, r17, 0, r10, r4, r8, r9
+	sha256_extend_doubleround 36, r17, 0, r5, r6, r10, r4
+	sha256_extend_doubleround 38, r17, 0, r7, r8, r5, r6
+	sha256_extend_doubleround 40, r17, 0, r9, r10, r7, r8
+	sha256_extend_doubleround 42, r17, 0, r4, r5, r9, r10
+	sha256_extend_doubleround 44, r17, 0, r6, r7, r4, r5
+	sha256_extend_doubleround 46, r17, 0, r8, r9, r6, r7
+	
+	lwz	r4,  0*4(r16)
+	lwz	r9,  1*4(r16)
+	lwz	r10, 2*4(r16)
+	lwz	r11, 3*4(r16)
+	lwz	r8,  4*4(r16)
+	lwz	r5,  5*4(r16)
+	lwz	r6,  6*4(r16)
+	lwz	r7,  7*4(r16)
+#ifdef _AIX
+	ld	r16, T.sha256_k(r2)
+#else
+	lis	r16, HI(sha256_k)
+	addi	r16, r16, LO(sha256_k)
+#endif
+	
+	sha256_main_round  3, r16, r17, 0, r5, r6, r7, r4, r9, r10, r11, r8
+	sha256_main_quadround  4, r16, r17, 0
+	sha256_main_quadround  8, r16, r17, 0
+	sha256_main_quadround 12, r16, r17, 0
+	sha256_main_quadround 16, r16, r17, 0
+	sha256_main_quadround 20, r16, r17, 0
+	sha256_main_quadround 24, r16, r17, 0
+	sha256_main_quadround 28, r16, r17, 0
+	sha256_main_quadround 32, r16, r17, 0
+	sha256_main_quadround 36, r16, r17, 0
+	sha256_main_quadround 40, r16, r17, 0
+	sha256_main_quadround 44, r16, r17, 0
+	sha256_main_quadround 48, r16, r17, 0
+	sha256_main_quadround 52, r16, r17, 0
+	sha256_main_quadround 56, r16, r17, 0
+	sha256_main_quadround 60, r16, r17, 0
+	
+	lwz	r12, 0*4(r18)
+	lwz	r13, 1*4(r18)
+	lwz	r14, 2*4(r18)
+	lwz	r15, 3*4(r18)
+	add	r4, r4, r12
+	add	r5, r5, r13
+	add	r6, r6, r14
+	add	r7, r7, r15
+	stw	r4, 8*4+0*4(r1)
+	stw	r5, 8*4+1*4(r1)
+	stw	r6, 8*4+2*4(r1)
+	stw	r7, 8*4+3*4(r1)
+	lwz	r12, 4*4(r18)
+	lwz	r13, 5*4(r18)
+	lwz	r14, 6*4(r18)
+	lwz	r15, 7*4(r18)
+	add	r8, r8, r12
+	add	r9, r9, r13
+	add	r10, r10, r14
+	add	r11, r11, r15
+	stw	r8, 8*4+4*4(r1)
+	stw	r9, 8*4+5*4(r1)
+	stw	r10, 8*4+6*4(r1)
+	stw	r11, 8*4+7*4(r1)
+
+	lwz	r4, 8*4+18*4(r1)
+	lwz	r5, 8*4+19*4(r1)
+	lwz	r6, 8*4+20*4(r1)
+	lwz	r7, 8*4+21*4(r1)
+	lwz	r8, 8*4+22*4(r1)
+	lwz	r9, 8*4+23*4(r1)
+	lwz	r10, 8*4+24*4(r1)
+	lwz	r11, 8*4+25*4(r1)
+	stw	r4,  18*4(r17)
+	stw	r5,  19*4(r17)
+	stw	r6,  20*4(r17)
+	stw	r7,  22*4(r17)
+	stw	r8,  23*4(r17)
+	stw	r9,  24*4(r17)
+	stw	r10, 30*4(r17)
+	stw	r11, 31*4(r17)
+	
+	lis	r8, 0x8000
+	li	r9,  0
+	li	r10, 0x0100
+	
+	lwz	r14, 8*4+1*4(r1)
+	lwz	r4, 8*4+0*4(r1)
+	
+	lwz	r11, 8*4+2*4(r1)
+	rotrwi	r12, r14, 7
+	rotrwi	r13, r14, 18
+	
+	stw	r8, 8*4+8*4(r1)
+	stw	r9, 8*4+9*4(r1)
+	stw	r9, 8*4+10*4(r1)
+	stw	r9, 8*4+11*4(r1)
+	stw	r9, 8*4+12*4(r1)
+	stw	r9, 8*4+13*4(r1)
+	stw	r9, 8*4+14*4(r1)
+	stw	r10, 8*4+15*4(r1)
+	
+	xor	r12, r12, r13
+	srwi	r13, r14, 3
+	addis	r5, r14, 0x00a0
+	xor	r12, r12, r13
+	rotrwi	r14, r11, 7
+	rotrwi	r13, r11, 18
+	add	r4, r4, r12
+	xor	r14, r14, r13
+	srwi	r13, r11, 3
+	stw	r4, 8*4+16*4(r1)
+	xor	r14, r14, r13
+	rotrwi	r12, r4, 17
+	rotrwi	r13, r4, 19
+	add	r5, r5, r14
+	lwz	r14, 8*4+3*4(r1)
+	
+	stw	r5, 8*4+17*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r4, 10
+	rotrwi	r6, r14, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r14, 18
+	xor	r6, r6, r13
+	srwi	r13, r14, 3
+	add	r11, r11, r12
+	xor	r6, r6, r13
+	rotrwi	r12, r5, 17
+	rotrwi	r13, r5, 19
+	add	r6, r6, r11
+	lwz	r11, 8*4+4*4(r1)
+	
+	stw	r6, 8*4+18*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r5, 10
+	rotrwi	r7, r11, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r11, 18
+	xor	r7, r7, r13
+	srwi	r13, r11, 3
+	add	r14, r14, r12
+	xor	r7, r7, r13
+	rotrwi	r12, r6, 17
+	rotrwi	r13, r6, 19
+	add	r7, r7, r14
+	lwz	r14, 8*4+5*4(r1)
+	
+	stw	r7, 8*4+19*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r6, 10
+	rotrwi	r8, r14, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r14, 18
+	xor	r8, r8, r13
+	srwi	r13, r14, 3
+	add	r11, r11, r12
+	xor	r8, r8, r13
+	rotrwi	r12, r7, 17
+	rotrwi	r13, r7, 19
+	add	r8, r8, r11
+	lwz	r11, 8*4+6*4(r1)
+	
+	stw	r8, 8*4+20*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r7, 10
+	rotrwi	r9, r11, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r11, 18
+	xor	r9, r9, r13
+	srwi	r13, r11, 3
+	add	r14, r14, r12
+	xor	r9, r9, r13
+	rotrwi	r12, r8, 17
+	rotrwi	r13, r8, 19
+	add	r9, r9, r14
+	lwz	r14, 8*4+7*4(r1)
+	
+	stw	r9, 8*4+21*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r8, 10
+	rotrwi	r10, r14, 7
+	xor	r12, r12, r13
+	rotrwi	r13, r14, 18
+	xor	r10, r10, r13
+	srwi	r13, r14, 3
+	add	r11, r11, r12
+	xor	r10, r10, r13
+	rotrwi	r12, r9, 17
+	rotrwi	r13, r9, 19
+	addi	r11, r11, 0x0100
+	add	r14, r14, r4
+	add	r10, r10, r11
+	
+	xor	r12, r12, r13
+	srwi	r13, r9, 10
+	stw	r10, 8*4+22*4(r1)
+	addis	r14, r14, 0x1100
+	xor	r12, r12, r13
+	add	r14, r14, r12
+	rotrwi	r12, r10, 17
+	rotrwi	r13, r10, 19
+	addi	r4, r14, 0x2000
+	xor	r12, r12, r13
+	srwi	r13, r10, 10
+	stw	r4, 8*4+23*4(r1)
+	addis	r5, r5, 0x8000
+	xor	r12, r12, r13
+	add	r5, r5, r12
+
+	rotrwi	r12, r4, 17
+	rotrwi	r13, r4, 19
+	stw	r5, 8*4+24*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r4, 10
+	rotrwi	r11, r5, 17
+	xor	r12, r12, r13
+	rotrwi	r13, r5, 19
+	xor	r11, r11, r13
+	srwi	r13, r5, 10
+	add	r6, r6, r12
+	xor	r11, r11, r13
+	stw	r6, 8*4+25*4(r1)
+	add	r7, r7, r11
+	
+	rotrwi	r12, r6, 17
+	rotrwi	r13, r6, 19
+	stw	r7, 8*4+26*4(r1)
+	xor	r12, r12, r13
+	srwi	r13, r6, 10
+	rotrwi	r11, r7, 17
+	xor	r12, r12, r13
+	rotrwi	r13, r7, 19
+	xor	r11, r11, r13
+	srwi	r13, r7, 10
+	add	r8, r8, r12
+	xor	r11, r11, r13
+	stw	r8, 8*4+27*4(r1)
+	add	r9, r9, r11
+	
+	rotrwi	r14, r8, 17
+	rotrwi	r13, r8, 19
+	rotrwi	r12, r9, 17
+	stw	r9, 8*4+28*4(r1)
+	addis	r4, r4, 0x0040
+	xor	r14, r14, r13
+	rotrwi	r13, r9, 19
+	xor	r12, r12, r13
+	srwi	r13, r8, 10
+	xor	r14, r14, r13
+	srwi	r13, r9, 10
+	xor	r12, r12, r13
+	addi	r4, r4, 0x0022
+	add	r10, r10, r14
+	add	r4, r4, r12
+	lwz	r11, 8*4+16*4(r1)
+	
+	addi	r5, r5, 0x0100
+	stw	r4, 8*4+30*4(r1)
+	rotrwi	r14, r11, 7
+	stw	r10, 8*4+29*4(r1)
+	rotrwi	r13, r11, 18
+	rotrwi	r12, r10, 17
+	xor	r14, r14, r13
+	rotrwi	r13, r10, 19
+	xor	r12, r12, r13
+	srwi	r13, r11, 3
+	xor	r14, r14, r13
+	srwi	r13, r10, 10
+	xor	r12, r12, r13
+	add	r5, r5, r14
+	add	r5, r5, r12
+	stw	r5, 8*4+31*4(r1)
+	
+	sha256_extend_doubleround 16, r1, 8*4, r6, r7, r4, r5
+	sha256_extend_doubleround 18, r1, 8*4, r8, r9, r6, r7
+	sha256_extend_doubleround 20, r1, 8*4, r10, r4, r8, r9
+	sha256_extend_doubleround 22, r1, 8*4, r5, r6, r10, r4
+	sha256_extend_doubleround 24, r1, 8*4, r7, r8, r5, r6
+	sha256_extend_doubleround 26, r1, 8*4, r9, r10, r7, r8
+	sha256_extend_doubleround 28, r1, 8*4, r4, r5, r9, r10
+	sha256_extend_doubleround 30, r1, 8*4, r6, r7, r4, r5
+	sha256_extend_doubleround 32, r1, 8*4, r8, r9, r6, r7
+	sha256_extend_doubleround 34, r1, 8*4, r10, r4, r8, r9
+	sha256_extend_doubleround 36, r1, 8*4, r5, r6, r10, r4
+	sha256_extend_doubleround 38, r1, 8*4, r7, r8, r5, r6
+	sha256_extend_doubleround 40, r1, 8*4, r9, r10, r7, r8
+	sha256_extend_doubleround 42, r1, 8*4, r4, r5, r9, r10
+	
+#ifdef _AIX
+	ld	r18, T.sha256_h(r2)
+#else
+	lis	r18, HI(sha256_h)
+	addi	r18, r18, LO(sha256_h)
+#endif
+	
+	lwz	r14, 8*4+(44+1)*4(r1)
+	rotrwi	r12, r4, 17
+	rotrwi	r13, r4, 19
+	add	r15, r11, r6
+	rotrwi	r6, r14, 7
+	rotrwi	r11, r14, 18
+	xor	r12, r12, r13
+	xor	r6, r6, r11
+	
+	lwz	r8, 4*4(r18)
+	lwz	r9, 5*4(r18)
+	lwz	r10, 6*4(r18)
+	lwz	r11, 7*4(r18)
+	
+	srwi	r13, r4, 10
+	srwi	r14, r14, 3
+	xor	r12, r12, r13
+	xor	r6, r6, r14
+	add	r12, r12, r15
+	add	r6, r6, r12
+	stw	r6, 8*4+(44+16)*4(r1)
+	
+	lwz	r4, 0*4(r18)
+	lwz	r5, 1*4(r18)
+	lwz	r6, 2*4(r18)
+	lwz	r7, 3*4(r18)
+	
+	sha256_main_quadround  0, r16, r1, 8*4
+	sha256_main_quadround  4, r16, r1, 8*4
+	sha256_main_quadround  8, r16, r1, 8*4
+	sha256_main_quadround 12, r16, r1, 8*4
+	sha256_main_quadround 16, r16, r1, 8*4
+	sha256_main_quadround 20, r16, r1, 8*4
+	sha256_main_quadround 24, r16, r1, 8*4
+	sha256_main_quadround 28, r16, r1, 8*4
+	sha256_main_quadround 32, r16, r1, 8*4
+	sha256_main_quadround 36, r16, r1, 8*4
+	sha256_main_quadround 40, r16, r1, 8*4
+	sha256_main_quadround 44, r16, r1, 8*4
+	sha256_main_quadround 48, r16, r1, 8*4
+	sha256_main_quadround 52, r16, r1, 8*4
+	sha256_main_round 56, r16, r1, 8*4, r4, r5, r6, r7, r8, r9, r10, r11
+
+.macro sha256_main_round_red i, rk, rw, wo, rd, re, rf, rg, rh
+	lwz	r12, \wo+(\i)*4(\rw)
+	and	r15, \rf, \re
+	andc	r14, \rg, \re
+	add	\rh, \rh, \rd
+	or	r14, r14, r15
+	lwz	r15, (\i)*4(\rk)
+	rotrwi	r13, \re, 5
+	add	\rh, \rh, r14
+	xor	r14, \re, r13
+	rotrwi	r13, \re, 19
+	add	\rh, \rh, r12
+	xor	r14, r14, r13
+	add	\rh, \rh, r15
+	rotrwi	r13, r14, 6
+	add	\rh, \rh, r13
+.endm
+	
+	sha256_main_round_red 57, r16, r1, 8*4, r6, r11, r8, r9, r10
+	sha256_main_round_red 58, r16, r1, 8*4, r5, r10, r11, r8, r9
+	sha256_main_round_red 59, r16, r1, 8*4, r4, r9, r10, r11, r8
+	lwz	r5, 7*4(r18)
+	sha256_main_round_red 60, r16, r1, 8*4, r7, r8, r9, r10, r11
+	
+	add	r11, r11, r5
+	stw	r11, 7*4(r3)
+	
+	ld	r13, 2*4(r1)
+	ld	r14, 4*4(r1)
+	ld	r15, 6*4(r1)
+	ld	r16, 72*4(r1)
+	ld	r17, 74*4(r1)
+	ld	r18, 76*4(r1)
+	addi	r1, r1, 80*4
+	blr
+
+
+#ifdef __ALTIVEC__
+
+#ifdef __APPLE__
+	.machine ppc7400
+#endif
+
+#ifdef _AIX
+	.csect .text[RO]
+#else
+	.data
+#endif
+	.align 4
+sha256_4h:
+	.long 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667
+	.long 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85
+	.long 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372
+	.long 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a
+	.long 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f
+	.long 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c
+	.long 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab
+	.long 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19
+
+	.align 4
+sha256_4k:
+	.long 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98
+	.long 0x71374491, 0x71374491, 0x71374491, 0x71374491
+	.long 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf
+	.long 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5
+	.long 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b
+	.long 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1
+	.long 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4
+	.long 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5
+	.long 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98
+	.long 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01
+	.long 0x243185be, 0x243185be, 0x243185be, 0x243185be
+	.long 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3
+	.long 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74
+	.long 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe
+	.long 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7
+	.long 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174
+	.long 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1
+	.long 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786
+	.long 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6
+	.long 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc
+	.long 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f
+	.long 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa
+	.long 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc
+	.long 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da
+	.long 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152
+	.long 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d
+	.long 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8
+	.long 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7
+	.long 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3
+	.long 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147
+	.long 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351
+	.long 0x14292967, 0x14292967, 0x14292967, 0x14292967
+	.long 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85
+	.long 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138
+	.long 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc
+	.long 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13
+	.long 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354
+	.long 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb
+	.long 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e
+	.long 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85
+	.long 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1
+	.long 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b
+	.long 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70
+	.long 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3
+	.long 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819
+	.long 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624
+	.long 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585
+	.long 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070
+	.long 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116
+	.long 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08
+	.long 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c
+	.long 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5
+	.long 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3
+	.long 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a
+	.long 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f
+	.long 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3
+	.long 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee
+	.long 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f
+	.long 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814
+	.long 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208
+	.long 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa
+	.long 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb
+	.long 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7
+	.long 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2
+
+	.align 4
+sha256d_4preext2:
+	.long 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000
+	.long 0x11002000, 0x11002000, 0x11002000, 0x11002000
+	.long 0x80000000, 0x80000000, 0x80000000, 0x80000000
+	.long 0x00400022, 0x00400022, 0x00400022, 0x00400022
+
+	.align 4
+br_perm:
+	.long 0x03020100, 0x07060504, 0x0b0a0908, 0x0f0e0d0c
+
+#ifdef _AIX
+	.toc
+T.sha256_4h:
+	.tc sha256_4h[TC], sha256_4h
+T.sha256_4k:
+	.tc sha256_4k[TC], sha256_4k
+T.sha256d_4preext2:
+	.tc sha256d_4preext2[TC], sha256d_4preext2
+T.br_perm:
+	.tc br_perm[TC], br_perm
+#endif
+
+
+.macro sha256_4way_extend_setup
+	vspltisw	v0, 10
+	vspltisw	v1, -7
+	vspltisw	v16, 3
+	vspltisw	v17, 15
+	vspltisw	v18, 14
+	vspltisw	v19, 13
+.endm
+
+.macro sha256_4way_extend_doubleround i, rw, va, vb, vy, vz
+	lvx	v14, \rw, r7
+	vrlw	v12, \vy, v17
+	vrlw	v13, \vy, v19
+	vadduwm	v11, v11, \va
+	vxor	v12, v12, v13
+	vsrw	v13, \vy, v0
+	vrlw	\va, v14, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v14, v18
+	vadduwm	v12, v12, v11
+	vxor	\va, \va, v13
+	vsrw	v13, v14, v16
+	lvx	v11, \rw, r8
+	vxor	\va, \va, v13
+	vrlw	v13, \vz, v19
+	vadduwm	\va, \va, v12
+
+	vrlw	v12, \vz, v17
+	vadduwm	v14, v14, \vb
+	vxor	v12, v12, v13
+	vsrw	v13, \vz, v0
+	vrlw	\vb, v11, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v11, v18
+	stvx	\va, \rw, r10
+	vxor	\vb, \vb, v13
+	vsrw	v13, v11, v16
+	vadduwm	v14, v14, v12
+	vxor	\vb, \vb, v13
+	vadduwm	\vb, \vb, v14
+	stvx	\vb, \rw, r11
+	addi	\rw, \rw, 2*16
+.endm
+
+
+.macro sha256_4way_main_setup
+	vspltisw	v2, 12
+	vspltisw	v3, -5
+	vspltisw	v16, -6
+	vspltisw	v17, -11
+	vspltisw	v18, -2
+.endm
+
+.macro sha256_4way_main_round i, rk, rw, va, vb, vc, vd, ve, vf, vg, vh
+	li	r6, (\i)*16
+	lvx	v12, \rw, r6
+	vand	v13, \vf, \ve
+	vandc	v14, \vg, \ve
+	lvx	v15, \rk, r6
+	vor	v14, v14, v13
+	vrlw	v13, \ve, v3
+	vadduwm	\vh, \vh, v14
+	vxor	v14, \ve, v13
+	vrlw	v13, \ve, v19
+	vadduwm	\vh, \vh, v12
+	vxor	v14, v14, v13
+	vadduwm	\vh, \vh, v15
+	vrlw	v13, v14, v16
+	vxor	v15, \va, \vb
+	vadduwm	\vh, \vh, v13
+
+	vrlw	v13, \va, v17
+	vand	v15, v15, \vc
+	vxor	v12, \va, v13
+	vrlw	v13, \va, v2
+	vand	v14, \va, \vb
+	vxor	v12, v12, v13
+	vxor	v14, v14, v15
+	vrlw	v13, v12, v18
+	vadduwm	v15, \vh, v14
+	vadduwm	\vh, \vh, \vd
+	vadduwm	\vd, v15, v13
+.endm
+
+.macro sha256_4way_main_quadround i, rk, rw
+	sha256_4way_main_round \i+0, \rk, \rw, v4, v5, v6, v7, v8, v9, v10, v11
+	sha256_4way_main_round \i+1, \rk, \rw, v7, v4, v5, v6, v11, v8, v9, v10
+	sha256_4way_main_round \i+2, \rk, \rw, v6, v7, v4, v5, v10, v11, v8, v9
+	sha256_4way_main_round \i+3, \rk, \rw, v5, v6, v7, v4, v9, v10, v11, v8
+.endm
+
+
+#ifdef _AIX
+	.csect .text[PR]
+#else
+	.text
+#endif
+	.align 2
+	.globl sha256_init_4way
+	.globl _sha256_init_4way
+	.globl .sha256_init_4way
+#ifdef __ELF__
+	.type sha256_init_4way, %function
+#endif
+sha256_init_4way:
+_sha256_init_4way:
+.sha256_init_4way:
+	mfspr	r0, 256
+	oris	r12, r0, 0xff00
+	mtspr	256, r12
+	
+#ifdef _AIX
+	ld	r4, T.sha256_4h(r2)
+#else
+	lis	r4, HI(sha256_4h)
+	addi	r4, r4, LO(sha256_4h)
+#endif
+	li	r5, 1*16
+	li	r6, 2*16
+	li	r7, 3*16
+	li	r8, 4*16
+	li	r9, 5*16
+	li	r10, 6*16
+	li	r11, 7*16
+	lvx	v0, 0, r4
+	lvx	v1, r4, r5
+	lvx	v2, r4, r6
+	lvx	v3, r4, r7
+	lvx	v4, r4, r8
+	lvx	v5, r4, r9
+	lvx	v6, r4, r10
+	lvx	v7, r4, r11
+	stvx	v0, 0, r3
+	stvx	v1, r3, r5
+	stvx	v2, r3, r6
+	stvx	v3, r3, r7
+	stvx	v4, r3, r8
+	stvx	v5, r3, r9
+	stvx	v6, r3, r10
+	stvx	v7, r3, r11
+	
+	mtspr	256, r0
+	blr
+
+
+	.align 2
+	.globl sha256_transform_4way
+	.globl _sha256_transform_4way
+	.globl .sha256_transform_4way
+#ifdef __ELF__
+	.type sha256_transform_4way, %function
+#endif
+sha256_transform_4way:
+_sha256_transform_4way:
+.sha256_transform_4way:
+	mfspr	r0, 256
+	oris	r12, r0, 0xffff
+	ori	r12, r12, 0xf000
+	mtspr	256, r12
+	
+	andi.	r6, r1, 15
+	cmpwi	0, r5, 0
+	li	r7, -(4*4+64*16)
+	subf	r6, r6, r7
+	stdux	r1, r1, r6
+
+	li	r7, 1*16
+	li	r8, 2*16
+	li	r9, 3*16
+	li	r10, 4*16
+	li	r11, 5*16
+	li	r12, 6*16
+	li	r6, 7*16
+	
+	bne	0, sha256_transform_4way_swap
+	
+	lvx	v11, 0, r4
+	lvx	v1, r4, r7
+	lvx	v2, r4, r8
+	lvx	v3, r4, r9
+	lvx	v4, r4, r10
+	lvx	v5, r4, r11
+	lvx	v6, r4, r12
+	lvx	v7, r4, r6
+	addi	r5, r1, 4*4
+	stvx	v11, 0, r5
+	stvx	v1, r5, r7
+	stvx	v2, r5, r8
+	stvx	v3, r5, r9
+	stvx	v4, r5, r10
+	stvx	v5, r5, r11
+	stvx	v6, r5, r12
+	stvx	v7, r5, r6
+	addi	r4, r4, 8*16
+	lvx	v0, 0, r4
+	lvx	v4, r4, r7
+	lvx	v5, r4, r8
+	lvx	v6, r4, r9
+	lvx	v7, r4, r10
+	lvx	v8, r4, r11
+	lvx	v9, r4, r12
+	lvx	v10, r4, r6
+	addi	r4, r1, 4*4+8*16
+	stvx	v0, 0, r4
+	stvx	v4, r4, r7
+	stvx	v5, r4, r8
+	stvx	v6, r4, r9
+	stvx	v7, r4, r10
+	stvx	v8, r4, r11
+	stvx	v9, r4, r12
+	stvx	v10, r4, r6
+	b	sha256_transform_4way_extend
+
+sha256_transform_4way_swap:
+#ifdef _AIX
+	ld	r5, T.br_perm(r2)
+#else
+	lis	r5, HI(br_perm)
+	addi	r5, r5, LO(br_perm)
+#endif
+	lvx	v19, 0, r5
+	
+	lvx	v11, 0, r4
+	lvx	v1, r4, r7
+	lvx	v2, r4, r8
+	lvx	v3, r4, r9
+	lvx	v4, r4, r10
+	lvx	v5, r4, r11
+	lvx	v6, r4, r12
+	lvx	v7, r4, r6
+	vperm	v11, v11, v11, v19
+	vperm	v1, v1, v1, v19
+	vperm	v2, v2, v2, v19
+	vperm	v3, v3, v3, v19
+	vperm	v4, v4, v4, v19
+	vperm	v5, v5, v5, v19
+	vperm	v6, v6, v6, v19
+	vperm	v7, v7, v7, v19
+	addi	r5, r1, 4*4
+	stvx	v11, 0, r5
+	stvx	v1, r5, r7
+	stvx	v2, r5, r8
+	stvx	v3, r5, r9
+	stvx	v4, r5, r10
+	stvx	v5, r5, r11
+	stvx	v6, r5, r12
+	stvx	v7, r5, r6
+	addi	r4, r4, 8*16
+	lvx	v0, 0, r4
+	lvx	v4, r4, r7
+	lvx	v5, r4, r8
+	lvx	v6, r4, r9
+	lvx	v7, r4, r10
+	lvx	v8, r4, r11
+	lvx	v9, r4, r12
+	lvx	v10, r4, r6
+	vperm	v0, v0, v0, v19
+	vperm	v4, v4, v4, v19
+	vperm	v5, v5, v5, v19
+	vperm	v6, v6, v6, v19
+	vperm	v7, v7, v7, v19
+	vperm	v8, v8, v8, v19
+	vperm	v9, v9, v9, v19
+	vperm	v10, v10, v10, v19
+	addi	r4, r1, 4*4+8*16
+	stvx	v0, 0, r4
+	stvx	v4, r4, r7
+	stvx	v5, r4, r8
+	stvx	v6, r4, r9
+	stvx	v7, r4, r10
+	stvx	v8, r4, r11
+	stvx	v9, r4, r12
+	stvx	v10, r4, r6
+	
+sha256_transform_4way_extend:
+	li	r10, 16*16
+	li	r11, 17*16
+	sha256_4way_extend_setup
+	sha256_4way_extend_doubleround  0, r5, v4, v5, v9, v10
+	sha256_4way_extend_doubleround  2, r5, v6, v7, v4, v5
+	sha256_4way_extend_doubleround  4, r5, v8, v9, v6, v7
+	sha256_4way_extend_doubleround  6, r5, v10, v4, v8, v9
+	sha256_4way_extend_doubleround  8, r5, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 10, r5, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 12, r5, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 14, r5, v4, v5, v9, v10
+	sha256_4way_extend_doubleround 16, r5, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 18, r5, v8, v9, v6, v7
+	sha256_4way_extend_doubleround 20, r5, v10, v4, v8, v9
+	sha256_4way_extend_doubleround 22, r5, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 24, r5, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 26, r5, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 28, r5, v4, v5, v9, v10
+	sha256_4way_extend_doubleround 30, r5, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 32, r5, v8, v9, v6, v7
+	sha256_4way_extend_doubleround 34, r5, v10, v4, v8, v9
+	sha256_4way_extend_doubleround 36, r5, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 38, r5, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 40, r5, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 42, r5, v4, v5, v9, v10
+	sha256_4way_extend_doubleround 44, r5, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 46, r5, v8, v9, v6, v7
+	
+	addi	r11, r3, 4*16
+	lvx	v4, 0, r3
+	lvx	v5, r3, r7
+	lvx	v6, r3, r8
+	lvx	v7, r3, r9
+	lvx	v8, 0, r11
+	lvx	v9, r11, r7
+	lvx	v10, r11, r8
+	lvx	v11, r11, r9
+#ifdef _AIX
+	ld	r12, T.sha256_4k(r2)
+#else
+	lis	r12, HI(sha256_4k)
+	addi	r12, r12, LO(sha256_4k)
+#endif
+	addi	r5, r1, 4*4
+	sha256_4way_main_setup
+	sha256_4way_main_quadround  0, r12, r5
+	sha256_4way_main_quadround  4, r12, r5
+	sha256_4way_main_quadround  8, r12, r5
+	sha256_4way_main_quadround 12, r12, r5
+	sha256_4way_main_quadround 16, r12, r5
+	sha256_4way_main_quadround 20, r12, r5
+	sha256_4way_main_quadround 24, r12, r5
+	sha256_4way_main_quadround 28, r12, r5
+	sha256_4way_main_quadround 32, r12, r5
+	sha256_4way_main_quadround 36, r12, r5
+	sha256_4way_main_quadround 40, r12, r5
+	sha256_4way_main_quadround 44, r12, r5
+	sha256_4way_main_quadround 48, r12, r5
+	sha256_4way_main_quadround 52, r12, r5
+	sha256_4way_main_quadround 56, r12, r5
+	sha256_4way_main_quadround 60, r12, r5
+	
+	lvx	v12, 0, r3
+	lvx	v13, r3, r7
+	lvx	v14, r3, r8
+	lvx	v15, r3, r9
+	lvx	v16, 0, r11
+	lvx	v17, r11, r7
+	lvx	v18, r11, r8
+	lvx	v19, r11, r9
+	vadduwm	v4, v4, v12
+	vadduwm	v5, v5, v13
+	vadduwm	v6, v6, v14
+	vadduwm	v7, v7, v15
+	vadduwm	v8, v8, v16
+	vadduwm	v9, v9, v17
+	vadduwm	v10, v10, v18
+	vadduwm	v11, v11, v19
+	stvx	v4, 0, r3
+	stvx	v5, r3, r7
+	stvx	v6, r3, r8
+	stvx	v7, r3, r9
+	stvx	v8, 0, r11
+	stvx	v9, r11, r7
+	stvx	v10, r11, r8
+	stvx	v11, r11, r9
+	
+	ld	r1, 0(r1)
+	mtspr	256, r0
+	blr
+
+
+	.align 2
+	.globl sha256d_ms_4way
+	.globl _sha256d_ms_4way
+	.globl .sha256d_ms_4way
+#ifdef __ELF__
+	.type sha256d_ms_4way, %function
+#endif
+sha256d_ms_4way:
+_sha256d_ms_4way:
+.sha256d_ms_4way:
+	mfspr	r0, 256
+	oris	r12, r0, 0xffff
+	ori	r12, r12, 0xf000
+	mtspr	256, r12
+	
+	andi.	r12, r1, 15
+	li	r11, -(4*4+64*16)
+	subf	r12, r12, r11
+	stdux	r1, r1, r12
+	
+	li	r7, 1*16
+	li	r8, 2*16
+	li	r9, 3*16
+	li	r10, 16*16
+	li	r11, 17*16
+	
+	sha256_4way_extend_setup
+	
+	addi	r4, r4, 2*16
+	addi	r12, r1, 4*4+18*16
+	lvx	v14, r4, r7
+	lvx	v6, r4, r10
+	lvx	v7, r4, r11
+	
+	vrlw	v12, v14, v1
+	vrlw	v13, v14, v18
+	stvx	v6, 0, r12
+	vxor	v12, v12, v13
+	vsrw	v13, v14, v16
+	stvx	v7, r12, r7
+	vxor	v12, v12, v13
+	vadduwm	v6, v6, v12
+	vadduwm	v7, v7, v14
+	stvx	v6, r4, r10
+	
+	vrlw	v12, v6, v17
+	vrlw	v13, v6, v19
+	stvx	v7, r4, r11
+	addi	r4, r4, 18*16
+	lvx	v8, 0, r4
+	vxor	v12, v12, v13
+	vsrw	v13, v6, v0
+	stvx	v8, r12, r8
+	vxor	v12, v12, v13
+	vadduwm	v8, v8, v12
+	
+	vrlw	v9, v7, v17
+	vrlw	v13, v7, v19
+	stvx	v8, 0, r4
+	vxor	v9, v9, v13
+	vsrw	v13, v7, v0
+	vxor	v9, v9, v13
+	
+	vrlw	v12, v8, v17
+	vrlw	v13, v8, v19
+	stvx	v9, r4, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v8, v0
+	lvx	v10, r4, r8
+	lvx	v4, r4, r9
+	vxor	v12, v12, v13
+	stvx	v10, r12, r9
+	addi	r12, r12, 4*16
+	stvx	v4, 0, r12
+	vrlw	v14, v9, v17
+	vrlw	v13, v9, v19
+	vadduwm	v10, v10, v12
+	
+	vxor	v14, v14, v13
+	vsrw	v13, v9, v0
+	stvx	v10, r4, r8
+	vxor	v14, v14, v13
+	vadduwm	v4, v4, v14
+	
+	vrlw	v12, v10, v17
+	vrlw	v13, v10, v19
+	stvx	v4, r4, r9
+	vxor	v12, v12, v13
+	vsrw	v13, v10, v0
+	vrlw	v14, v4, v17
+	vxor	v12, v12, v13
+	vrlw	v13, v4, v19
+	addi	r4, r4, 4*16
+	lvx	v5, 0, r4
+	vxor	v14, v14, v13
+	stvx	v5, r12, r7
+	vsrw	v13, v4, v0
+	vadduwm	v5, v5, v12
+	vxor	v14, v14, v13
+	stvx	v5, 0, r4
+	vadduwm	v6, v6, v14
+	
+	vrlw	v12, v5, v17
+	vrlw	v13, v5, v19
+	stvx	v6, r4, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v5, v0
+	vrlw	v14, v6, v17
+	vxor	v12, v12, v13
+	vrlw	v13, v6, v19
+	vxor	v14, v14, v13
+	vsrw	v13, v6, v0
+	vadduwm	v7, v7, v12
+	vxor	v14, v14, v13
+	stvx	v7, r4, r8
+	vadduwm	v8, v8, v14
+	
+	vrlw	v12, v7, v17
+	vrlw	v13, v7, v19
+	stvx	v8, r4, r9
+	vxor	v12, v12, v13
+	vsrw	v13, v7, v0
+	vrlw	v14, v8, v17
+	vxor	v12, v12, v13
+	vrlw	v13, v8, v19
+	vxor	v14, v14, v13
+	vsrw	v13, v8, v0
+	vadduwm	v9, v9, v12
+	vxor	v14, v14, v13
+	addi	r4, r4, 4*16
+	stvx	v9, 0, r4
+	vadduwm	v10, v10, v14
+	
+	vrlw	v12, v9, v17
+	vrlw	v13, v9, v19
+	stvx	v10, r4, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v9, v0
+	lvx	v11, r4, r8
+	lvx	v14, r4, r9
+	stvx	v11, r12, r8
+	stvx	v14, r12, r9
+	vxor	v12, v12, v13
+	vadduwm	v11, v11, v12
+	vadduwm	v5, v5, v14
+	vrlw	v12, v10, v17
+	vrlw	v13, v10, v19
+	vadduwm	v4, v4, v11
+	
+	vxor	v12, v12, v13
+	vsrw	v13, v10, v0
+	stvx	v4, r4, r8
+	vxor	v12, v12, v13
+	vadduwm	v5, v5, v12
+	stvx	v5, r4, r9
+	addi	r4, r4, -12*16
+	lvx	v11, 0, r4
+	
+	sha256_4way_extend_doubleround 16, r4, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 18, r4, v8, v9, v6, v7
+	sha256_4way_extend_doubleround 20, r4, v10, v4, v8, v9
+	sha256_4way_extend_doubleround 22, r4, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 24, r4, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 26, r4, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 28, r4, v4, v5, v9, v10
+	sha256_4way_extend_doubleround 30, r4, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 32, r4, v8, v9, v6, v7
+	sha256_4way_extend_doubleround 34, r4, v10, v4, v8, v9
+	sha256_4way_extend_doubleround 36, r4, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 38, r4, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 40, r4, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 42, r4, v4, v5, v9, v10
+	sha256_4way_extend_doubleround 44, r4, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 46, r4, v8, v9, v6, v7
+	addi	r4, r4, -48*16
+	
+	lvx	v4, 0, r6
+	lvx	v9, r6, r7
+	lvx	v10, r6, r8
+	lvx	v11, r6, r9
+	addi	r12, r6, 4*16
+	lvx	v8, 0, r12
+	lvx	v5, r12, r7
+	lvx	v6, r12, r8
+	lvx	v7, r12, r9
+#ifdef _AIX
+	ld	r12, T.sha256_4k(r2)
+#else
+	lis	r12, HI(sha256_4k)
+	addi	r12, r12, LO(sha256_4k)
+#endif
+	sha256_4way_main_setup
+	sha256_4way_main_round  3, r12, r4, v5, v6, v7, v4, v9, v10, v11, v8
+	sha256_4way_main_quadround  4, r12, r4
+	sha256_4way_main_quadround  8, r12, r4
+	sha256_4way_main_quadround 12, r12, r4
+	sha256_4way_main_quadround 16, r12, r4
+	sha256_4way_main_quadround 20, r12, r4
+	sha256_4way_main_quadround 24, r12, r4
+	sha256_4way_main_quadround 28, r12, r4
+	sha256_4way_main_quadround 32, r12, r4
+	sha256_4way_main_quadround 36, r12, r4
+	sha256_4way_main_quadround 40, r12, r4
+	sha256_4way_main_quadround 44, r12, r4
+	sha256_4way_main_quadround 48, r12, r4
+	sha256_4way_main_quadround 52, r12, r4
+	sha256_4way_main_quadround 56, r12, r4
+	sha256_4way_main_quadround 60, r12, r4
+	
+	lvx	v12, 0, r5
+	lvx	v13, r5, r7
+	lvx	v14, r5, r8
+	lvx	v15, r5, r9
+	addi	r12, r5, 4*16
+	lvx	v16, 0, r12
+	lvx	v17, r12, r7
+	lvx	v18, r12, r8
+	lvx	v19, r12, r9
+	vadduwm	v4, v4, v12
+	vadduwm	v5, v5, v13
+	vadduwm	v6, v6, v14
+	vadduwm	v7, v7, v15
+	vadduwm	v8, v8, v16
+	vadduwm	v9, v9, v17
+	vadduwm	v10, v10, v18
+	vadduwm	v11, v11, v19
+	addi	r12, r1, 4*4
+	stvx	v4, 0, r12
+	stvx	v5, r12, r7
+	stvx	v6, r12, r8
+	stvx	v7, r12, r9
+	addi	r12, r12, 4*16
+	stvx	v8, 0, r12
+	stvx	v9, r12, r7
+	stvx	v10, r12, r8
+	stvx	v11, r12, r9
+	
+	addi	r12, r1, 4*4+18*16
+	lvx	v4, 0, r12
+	lvx	v5, r12, r7
+	lvx	v6, r12, r8
+	lvx	v7, r12, r9
+	addi	r12, r12, 4*16
+	lvx	v8, 0, r12
+	lvx	v9, r12, r7
+	lvx	v10, r12, r8
+	lvx	v11, r12, r9
+	addi	r12, r4, 18*16
+	stvx	v4, 0, r12
+	stvx	v5, r12, r7
+	stvx	v6, r12, r8
+	addi	r12, r4, 22*16
+	stvx	v7, 0, r12
+	stvx	v8, r12, r7
+	stvx	v9, r12, r8
+	addi	r12, r4, 30*16
+	stvx	v10, 0, r12
+	stvx	v11, r12, r7
+	
+	addi	r4, r1, 4*4
+	
+	sha256_4way_extend_setup
+	
+#ifdef _AIX
+	ld	r12, T.sha256d_4preext2(r2)
+#else
+	lis	r12, HI(sha256d_4preext2)
+	addi	r12, r12, LO(sha256d_4preext2)
+#endif
+	lvx	v2, 0, r12
+	
+	vxor	v9, v9, v9
+	vspltisw	v3, 1
+	lvx	v4, r12, r8
+	vsldoi	v3, v3, v3, 1
+	addi	r5, r1, 4*4+8*16
+	stvx	v4, 0, r5
+	stvx	v9, r5, r7
+	stvx	v9, r5, r8
+	stvx	v9, r5, r9
+	addi	r5, r5, 4*16
+	stvx	v9, 0, r5
+	stvx	v9, r5, r7
+	stvx	v9, r5, r8
+	stvx	v3, r5, r9
+	
+	lvx	v4, 0, r4
+	lvx	v14, r4, r7
+	
+	lvx	v11, r4, r8
+	vrlw	v12, v14, v1
+	vrlw	v13, v14, v18
+	
+	vxor	v12, v12, v13
+	vsrw	v13, v14, v16
+	vadduwm	v5, v14, v2
+	vxor	v12, v12, v13
+	vrlw	v14, v11, v1
+	vrlw	v13, v11, v18
+	vadduwm	v4, v4, v12
+	vxor	v14, v14, v13
+	vsrw	v13, v11, v16
+	stvx	v4, r4, r10
+	vxor	v14, v14, v13
+	vrlw	v12, v4, v17
+	vrlw	v13, v4, v19
+	vadduwm	v5, v5, v14
+	
+	stvx	v5, r4, r11
+	addi	r4, r4, 2*16
+	lvx	v14, r4, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v4, v0
+	vrlw	v6, v14, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v14, v18
+	vxor	v6, v6, v13
+	vsrw	v13, v14, v16
+	vadduwm	v11, v11, v12
+	vxor	v6, v6, v13
+	vrlw	v12, v5, v17
+	vrlw	v13, v5, v19
+	vadduwm	v6, v6, v11
+	lvx	v11, r4, r8
+	
+	stvx	v6, r4, r10
+	vxor	v12, v12, v13
+	vsrw	v13, v5, v0
+	vrlw	v7, v11, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v11, v18
+	vxor	v7, v7, v13
+	vsrw	v13, v11, v16
+	vadduwm	v14, v14, v12
+	vxor	v7, v7, v13
+	vrlw	v12, v6, v17
+	vrlw	v13, v6, v19
+	vadduwm	v7, v7, v14
+	
+	stvx	v7, r4, r11
+	addi	r4, r4, 2*16
+	lvx	v14, r4, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v6, v0
+	vrlw	v8, v14, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v14, v18
+	vxor	v8, v8, v13
+	vsrw	v13, v14, v16
+	vadduwm	v11, v11, v12
+	vxor	v8, v8, v13
+	vrlw	v12, v7, v17
+	vrlw	v13, v7, v19
+	vadduwm	v8, v8, v11
+	lvx	v11, r4, r8
+	
+	stvx	v8, r4, r10
+	vxor	v12, v12, v13
+	vsrw	v13, v7, v0
+	vrlw	v9, v11, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v11, v18
+	vxor	v9, v9, v13
+	vsrw	v13, v11, v16
+	vadduwm	v14, v14, v12
+	vxor	v9, v9, v13
+	vrlw	v12, v8, v17
+	vrlw	v13, v8, v19
+	vadduwm	v9, v9, v14
+	
+	stvx	v9, r4, r11
+	addi	r4, r4, 2*16
+	lvx	v14, r4, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v8, v0
+	vrlw	v10, v14, v1
+	vxor	v12, v12, v13
+	vrlw	v13, v14, v18
+	vxor	v10, v10, v13
+	vsrw	v13, v14, v16
+	vadduwm	v11, v11, v12
+	vxor	v10, v10, v13
+	vrlw	v12, v9, v17
+	vrlw	v13, v9, v19
+	vadduwm	v11, v11, v3
+	vadduwm	v14, v14, v4
+	vadduwm	v10, v10, v11
+	
+	lvx	v2, r12, r7
+	vxor	v12, v12, v13
+	vsrw	v13, v9, v0
+	stvx	v10, r4, r10
+	vxor	v12, v12, v13
+	vadduwm	v14, v14, v12
+	vrlw	v12, v10, v17
+	vrlw	v13, v10, v19
+	vadduwm	v4, v14, v2
+	lvx	v2, r12, r8
+	vxor	v12, v12, v13
+	vsrw	v13, v10, v0
+	stvx	v4, r4, r11
+	vadduwm	v5, v5, v2
+	vxor	v12, v12, v13
+	vadduwm	v5, v5, v12
+
+	vrlw	v12, v4, v17
+	vrlw	v13, v4, v19
+	addi	r4, r4, 2*16
+	stvx	v5, r4, r10
+	vxor	v12, v12, v13
+	vsrw	v13, v4, v0
+	vrlw	v11, v5, v17
+	vxor	v12, v12, v13
+	vrlw	v13, v5, v19
+	vxor	v11, v11, v13
+	vsrw	v13, v5, v0
+	vadduwm	v6, v6, v12
+	vxor	v11, v11, v13
+	stvx	v6, r4, r11
+	vadduwm	v7, v7, v11
+	
+	vrlw	v12, v6, v17
+	vrlw	v13, v6, v19
+	addi	r4, r4, 2*16
+	stvx	v7, r4, r10
+	vxor	v12, v12, v13
+	vsrw	v13, v6, v0
+	vrlw	v11, v7, v17
+	vxor	v12, v12, v13
+	vrlw	v13, v7, v19
+	vxor	v11, v11, v13
+	vsrw	v13, v7, v0
+	vadduwm	v8, v8, v12
+	vxor	v11, v11, v13
+	stvx	v8, r4, r11
+	vadduwm	v9, v9, v11
+	
+	lvx	v2, r12, r9
+	vrlw	v14, v8, v17
+	vrlw	v13, v8, v19
+	vrlw	v12, v9, v17
+	addi	r4, r4, 2*16
+	stvx	v9, r4, r10
+	vxor	v14, v14, v13
+	vrlw	v13, v9, v19
+	vxor	v12, v12, v13
+	vsrw	v13, v8, v0
+	vxor	v14, v14, v13
+	vsrw	v13, v9, v0
+	vxor	v12, v12, v13
+	vadduwm	v4, v4, v2
+	vadduwm	v10, v10, v14
+	vadduwm	v4, v4, v12
+	stvx	v10, r4, r11
+	addi	r4, r4, 2*16
+	lvx	v11, r4, r8
+	
+	vadduwm	v5, v5, v3
+	stvx	v4, r4, r10
+	vrlw	v14, v11, v1
+	vrlw	v13, v11, v18
+	vrlw	v12, v10, v17
+	vxor	v14, v14, v13
+	vrlw	v13, v10, v19
+	vxor	v12, v12, v13
+	vsrw	v13, v11, v16
+	vxor	v14, v14, v13
+	vsrw	v13, v10, v0
+	vxor	v12, v12, v13
+	vadduwm	v5, v5, v14
+	vadduwm	v5, v5, v12
+	stvx	v5, r4, r11
+	addi	r4, r4, 2*16
+	
+	sha256_4way_extend_doubleround 16, r4, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 18, r4, v8, v9, v6, v7
+	sha256_4way_extend_doubleround 20, r4, v10, v4, v8, v9
+	sha256_4way_extend_doubleround 22, r4, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 24, r4, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 26, r4, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 28, r4, v4, v5, v9, v10
+	sha256_4way_extend_doubleround 30, r4, v6, v7, v4, v5
+	sha256_4way_extend_doubleround 32, r4, v8, v9, v6, v7
+	sha256_4way_extend_doubleround 34, r4, v10, v4, v8, v9
+	sha256_4way_extend_doubleround 36, r4, v5, v6, v10, v4
+	sha256_4way_extend_doubleround 38, r4, v7, v8, v5, v6
+	sha256_4way_extend_doubleround 40, r4, v9, v10, v7, v8
+	sha256_4way_extend_doubleround 42, r4, v4, v5, v9, v10
+	
+	lvx	v14, r4, r7
+	vrlw	v12, v4, v17
+	vrlw	v13, v4, v19
+	vadduwm	v15, v11, v6
+	vrlw	v6, v14, v1
+	vrlw	v11, v14, v18
+	vxor	v12, v12, v13
+	vxor	v6, v6, v11
+	vsrw	v13, v4, v0
+	vsrw	v14, v14, v16
+	vxor	v12, v12, v13
+	vxor	v6, v6, v14
+	vadduwm	v12, v12, v15
+	vadduwm	v6, v6, v12
+	stvx	v6, r4, r10
+	addi	r4, r4, -44*16
+	
+#ifdef _AIX
+	ld	r5, T.sha256_4h(r2)
+#else
+	lis	r5, HI(sha256_4h)
+	addi	r5, r5, LO(sha256_4h)
+#endif
+	lvx	v4, 0, r5
+	lvx	v5, r5, r7
+	lvx	v6, r5, r8
+	lvx	v7, r5, r9
+	addi	r12, r5, 4*16
+	lvx	v8, 0, r12
+	lvx	v9, r12, r7
+	lvx	v10, r12, r8
+	lvx	v11, r12, r9
+#ifdef _AIX
+	ld	r12, T.sha256_4k(r2)
+#else
+	lis	r12, HI(sha256_4k)
+	addi	r12, r12, LO(sha256_4k)
+#endif
+	sha256_4way_main_setup
+	sha256_4way_main_quadround  0, r12, r4
+	sha256_4way_main_quadround  4, r12, r4
+	sha256_4way_main_quadround  8, r12, r4
+	sha256_4way_main_quadround 12, r12, r4
+	sha256_4way_main_quadround 16, r12, r4
+	sha256_4way_main_quadround 20, r12, r4
+	sha256_4way_main_quadround 24, r12, r4
+	sha256_4way_main_quadround 28, r12, r4
+	sha256_4way_main_quadround 32, r12, r4
+	sha256_4way_main_quadround 36, r12, r4
+	sha256_4way_main_quadround 40, r12, r4
+	sha256_4way_main_quadround 44, r12, r4
+	sha256_4way_main_quadround 48, r12, r4
+	sha256_4way_main_quadround 52, r12, r4
+	sha256_4way_main_round 56, r12, r4, v4, v5, v6, v7, v8, v9, v10, v11
+
+.macro sha256_4way_main_round_red i, rk, rw, vd, ve, vf, vg, vh
+	li	r6, (\i)*16
+	vand	v15, \vf, \ve
+	vandc	v14, \vg, \ve
+	lvx	v12, \rw, r6
+	vadduwm	\vh, \vh, \vd
+	vor	v14, v14, v15
+	lvx	v15, \rk, r6
+	vrlw	v13, \ve, v3
+	vadduwm	\vh, \vh, v14
+	vxor	v14, \ve, v13
+	vrlw	v13, \ve, v19
+	vadduwm	\vh, \vh, v12
+	vxor	v14, v14, v13
+	vadduwm	\vh, \vh, v15
+	vrlw	v13, v14, v16
+	vadduwm	\vh, \vh, v13
+.endm
+
+	sha256_4way_main_round_red 57, r12, r4, v6, v11, v8, v9, v10
+	sha256_4way_main_round_red 58, r12, r4, v5, v10, v11, v8, v9
+	sha256_4way_main_round_red 59, r12, r4, v4, v9, v10, v11, v8
+	sha256_4way_main_round_red 60, r12, r4, v7, v8, v9, v10, v11
+	
+	li	r12, 7*16
+	lvx	v19, r5, r12
+	vadduwm	v11, v11, v19
+	stvx	v11, r3, r12
+	
+	ld	r1, 0(r1)
+	mtspr	256, r0
+	blr
+
+
+	.align 2
+	.globl sha256_use_4way
+	.globl _sha256_use_4way
+	.globl .sha256_use_4way
+#ifdef __ELF__
+	.type sha256_use_4way, %function
+#endif
+sha256_use_4way:
+_sha256_use_4way:
+.sha256_use_4way:
+	li	r3, 1
+	blr
+
+#endif /* __ALTIVEC__ */
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x64.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x64.S
new file mode 100644
index 0000000..770d3ba
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x64.S
@@ -0,0 +1,4222 @@
+/*
+ * Copyright 2012-2015 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(__linux__) && defined(__ELF__)
+	.section .note.GNU-stack,"",%progbits
+#endif
+
+#if defined(USE_ASM) && defined(__x86_64__)
+
+	.data
+	.p2align 4
+sha256_h:
+	.long 0x6a09e667, 0xbb67ae85, 0x3c6ef372, 0xa54ff53a
+	.long 0x510e527f, 0x9b05688c, 0x1f83d9ab, 0x5be0cd19
+
+	.data
+	.p2align 6
+sha256_k:
+	.long 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5
+	.long 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5
+	.long 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3
+	.long 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174
+	.long 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc
+	.long 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da
+	.long 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7
+	.long 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967
+	.long 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13
+	.long 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85
+	.long 0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3
+	.long 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070
+	.long 0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5
+	.long 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3
+	.long 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208
+	.long 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
+
+bswap_xmm_mask:
+	.long 0x00010203, 0x04050607, 0x08090a0b, 0x0c0d0e0f
+
+
+.macro sha256_mixed_quadround ra, rb, rc, rd, re, rf, rg, rh, x0, x1, x2, x3
+	movdqa	\x3, %xmm4
+	movl	\re, %eax
+	movdqa	\x2, %xmm6
+	rorl	$(25-11), %eax
+	movl	\ra, %ebx
+	pslldq	$12, %xmm4
+	rorl	$(22-13), %ebx
+	psrldq	$4, %xmm6
+	xorl	\re, %eax
+	movl	\rf, %ecx
+	rorl	$(11-6), %eax
+	pxor	%xmm6, %xmm4
+	movdqa	\x1, %xmm5
+	xorl	\ra, %ebx
+	xorl	\rg, %ecx
+	xorl	\re, %eax
+	paddd	\x0, %xmm4
+	movdqa	\x0, %xmm7
+	andl	\re, %ecx
+	rorl	$(13-2), %ebx
+	xorl	\ra, %ebx
+	pslldq	$12, %xmm5
+	psrldq	$4, %xmm7
+	rorl	$6, %eax
+	xorl	\rg, %ecx
+	pxor	%xmm7, %xmm5
+	rorl	$2, %ebx
+	addl	%eax, %ecx
+	addl	(%rsp) , %ecx
+	movdqa	%xmm5, %xmm6
+	movl	\ra, %eax
+	addl	%ecx, \rh
+	movl	\ra, %ecx
+	movdqa	%xmm5, %xmm7
+	orl	\rc, %eax
+	addl	\rh, \rd
+	andl	\rc, %ecx
+	pslld	$(32-7), %xmm5
+	psrld	$7, %xmm6
+	andl	\rb, %eax
+	addl	%ebx, \rh
+	orl	%ecx, %eax
+	por	%xmm6, %xmm5
+	addl	%eax, \rh
+	
+	movl	\rd, %eax
+	movdqa	%xmm7, %xmm6
+	movl	\rh, %ebx
+	rorl	$(25-11), %eax
+	xorl	\rd, %eax
+	movdqa	%xmm7, %xmm8
+	movl	\re, %ecx
+	rorl	$(22-13), %ebx
+	xorl	\rh, %ebx
+	pslld	$(32-18), %xmm7
+	rorl	$(11-6), %eax
+	xorl	\rf, %ecx
+	rorl	$(13-2), %ebx
+	psrld	$18, %xmm6
+	xorl	\rd, %eax
+	andl	\rd, %ecx
+	rorl	$6, %eax
+	pxor	%xmm7, %xmm5
+	xorl	\rh, %ebx
+	xorl	\rf, %ecx
+	psrld	$3, %xmm8
+	addl	%eax, %ecx
+	addl	1*4(%rsp), %ecx
+	rorl	$2, %ebx
+	pxor	%xmm6, %xmm5
+	movl	\rh, %eax
+	addl	%ecx, \rg
+	movl	\rh, %ecx
+	pxor	%xmm8, %xmm5
+	orl	\rb, %eax
+	addl	\rg, \rc
+	andl	\rb, %ecx
+	pshufd	$0xfa, \x3, %xmm6
+	andl	\ra, %eax
+	addl	%ebx, \rg
+	paddd	%xmm5, %xmm4
+	orl	%ecx, %eax
+	addl	%eax, \rg
+	
+	movl	\rc, %eax
+	movdqa	%xmm6, %xmm7
+	movl	\rg, %ebx
+	rorl	$(25-11), %eax
+	xorl	\rc, %eax
+	movdqa	%xmm6, %xmm8
+	rorl	$(22-13), %ebx
+	movl	\rd, %ecx
+	xorl	\rg, %ebx
+	psrlq	$17, %xmm6
+	psrlq	$19, %xmm7
+	rorl	$(11-6), %eax
+	xorl	\re, %ecx
+	xorl	\rc, %eax
+	psrld	$10, %xmm8
+	pxor	%xmm7, %xmm6
+	andl	\rc, %ecx
+	rorl	$(13-2), %ebx
+	xorl	\rg, %ebx
+	pxor	%xmm6, %xmm8
+	xorl	\re, %ecx
+	rorl	$6, %eax
+	addl	%eax, %ecx
+	pshufd	$0x8f, %xmm8, %xmm8
+	rorl	$2, %ebx
+	addl	2*4(%rsp), %ecx
+	movl	\rg, %eax
+	psrldq	$8, %xmm8
+	addl	%ecx, \rf
+	movl	\rg, %ecx
+	orl	\ra, %eax
+	paddd	%xmm8, %xmm4
+	addl	\rf, \rb
+	andl	\ra, %ecx
+	andl	\rh, %eax
+	pshufd	$0x50, %xmm4, %xmm6
+	addl	%ebx, \rf
+	orl	%ecx, %eax
+	addl	%eax, \rf
+	
+	movdqa	%xmm6, %xmm7
+	movl	\rb, %eax
+	rorl	$(25-11), %eax
+	movl	\rf, %ebx
+	movdqa	%xmm6, \x0
+	rorl	$(22-13), %ebx
+	xorl	\rb, %eax
+	movl	\rc, %ecx
+	psrlq	$17, %xmm6
+	rorl	$(11-6), %eax
+	xorl	\rf, %ebx
+	xorl	\rd, %ecx
+	psrlq	$19, %xmm7
+	xorl	\rb, %eax
+	andl	\rb, %ecx
+	rorl	$(13-2), %ebx
+	psrld	$10, \x0
+	xorl	\rf, %ebx
+	rorl	$6, %eax
+	pxor	%xmm7, %xmm6
+	xorl	\rd, %ecx
+	rorl	$2, %ebx
+	addl	%eax, %ecx
+	pxor	%xmm6, \x0
+	addl	3*4(%rsp), %ecx
+	movl	\rf, %eax
+	addl	%ecx, \re
+	pshufd	$0xf8, \x0, \x0
+	movl	\rf, %ecx
+	orl	\rh, %eax
+	addl	\re, \ra
+	pslldq	$8, \x0
+	andl	\rh, %ecx
+	andl	\rg, %eax
+	paddd	%xmm4, \x0
+	addl	%ebx, \re
+	orl	%ecx, %eax
+	addl	%eax, \re
+.endm
+
+.macro sha256_main_round i, ra, rb, rc, rd, re, rf, rg, rh
+	movl	\re, %eax
+	rorl	$(25-11), %eax
+	movl	\ra, %ebx
+	xorl	\re, %eax
+	rorl	$(22-13), %ebx
+	movl	\rf, %ecx
+	xorl	\ra, %ebx
+	rorl	$(11-6), %eax
+	xorl	\rg, %ecx
+	xorl	\re, %eax
+	rorl	$(13-2), %ebx
+	andl	\re, %ecx
+	xorl	\ra, %ebx
+	rorl	$6, %eax
+	xorl	\rg, %ecx
+	addl	%eax, %ecx
+	rorl	$2, %ebx
+	addl	\i*4(%rsp), %ecx
+	movl	\ra, %eax
+	addl	%ecx, \rh
+	movl	\ra, %ecx
+	orl	\rc, %eax
+	addl	\rh, \rd
+	andl	\rc, %ecx
+	andl	\rb, %eax
+	addl	%ebx, \rh
+	orl	%ecx, %eax
+	addl	%eax, \rh
+.endm
+
+
+	.text
+	.p2align 6
+sha256_transform_sse2:
+	pushq	%rbx
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	pushq	%rsi
+	subq	$5*16, %rsp
+	movdqa	%xmm6, 1*16(%rsp)
+	movdqa	%xmm7, 2*16(%rsp)
+	movdqa	%xmm8, 3*16(%rsp)
+	movdqa	%xmm9, 4*16(%rsp)
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+#else
+	subq	$16, %rsp
+#endif
+	
+	movl	0*4(%rdi), %r8d
+	movl	1*4(%rdi), %r9d
+	movl	2*4(%rdi), %r10d
+	movl	3*4(%rdi), %r11d
+	movl	4*4(%rdi), %r12d
+	movl	5*4(%rdi), %r13d
+	movl	6*4(%rdi), %r14d
+	movl	7*4(%rdi), %r15d
+	
+	testq	%rdx, %rdx
+	jnz sha256_transform_sse2_swap
+	
+	movdqu	0*16(%rsi), %xmm0
+	movdqu	1*16(%rsi), %xmm1
+	movdqu	2*16(%rsi), %xmm2
+	movdqu	3*16(%rsi), %xmm3
+	jmp sha256_transform_sse2_core
+	
+sha256_transform_sse2_swap:
+	movdqu	0*16(%rsi), %xmm0
+	movdqu	1*16(%rsi), %xmm1
+	movdqu	2*16(%rsi), %xmm2
+	movdqu	3*16(%rsi), %xmm3
+	pshuflw	$0xb1, %xmm0, %xmm0
+	pshuflw	$0xb1, %xmm1, %xmm1
+	pshuflw	$0xb1, %xmm2, %xmm2
+	pshuflw	$0xb1, %xmm3, %xmm3
+	pshufhw	$0xb1, %xmm0, %xmm0
+	pshufhw	$0xb1, %xmm1, %xmm1
+	pshufhw	$0xb1, %xmm2, %xmm2
+	pshufhw	$0xb1, %xmm3, %xmm3
+	movdqa	%xmm0, %xmm4
+	movdqa	%xmm1, %xmm5
+	movdqa	%xmm2, %xmm6
+	movdqa	%xmm3, %xmm7
+	psrlw	$8, %xmm4
+	psrlw	$8, %xmm5
+	psrlw	$8, %xmm6
+	psrlw	$8, %xmm7
+	psllw	$8, %xmm0
+	psllw	$8, %xmm1
+	psllw	$8, %xmm2
+	psllw	$8, %xmm3
+	pxor	%xmm4, %xmm0
+	pxor	%xmm5, %xmm1
+	pxor	%xmm6, %xmm2
+	pxor	%xmm7, %xmm3
+	
+sha256_transform_sse2_core:
+	leaq	sha256_k(%rip), %rdx
+	movq	$48, %rsi
+	.p2align 4
+sha256_transform_sse2_loop:
+	movdqa	0*16(%rdx), %xmm9
+	paddd	%xmm0, %xmm9
+	movdqa	%xmm9, (%rsp)
+	sha256_mixed_quadround %r8d, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d, %xmm0, %xmm1, %xmm2, %xmm3
+	movdqa	1*16(%rdx), %xmm9
+	paddd	%xmm1, %xmm9
+	movdqa	%xmm9, (%rsp)
+	sha256_mixed_quadround %r12d, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d, %xmm1, %xmm2, %xmm3, %xmm0
+	movdqa	2*16(%rdx), %xmm9
+	paddd	%xmm2, %xmm9
+	movdqa	%xmm9, (%rsp)
+	sha256_mixed_quadround %r8d, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d, %xmm2, %xmm3, %xmm0, %xmm1
+	movdqa	3*16(%rdx), %xmm9
+	paddd	%xmm3, %xmm9
+	movdqa	%xmm9, (%rsp)
+	addq	$4*16, %rdx
+	sha256_mixed_quadround %r12d, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d, %xmm3, %xmm0, %xmm1, %xmm2
+	
+	subq	$16, %rsi
+	jne sha256_transform_sse2_loop
+	
+	paddd	0*16(%rdx), %xmm0
+	movdqa	%xmm0, (%rsp)
+	sha256_main_round 0, %r8d, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d
+	sha256_main_round 1, %r15d, %r8d, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d
+	sha256_main_round 2, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d, %r12d, %r13d
+	sha256_main_round 3, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d, %r12d
+	paddd	1*16(%rdx), %xmm1
+	movdqa	%xmm1, (%rsp)
+	sha256_main_round 0, %r12d, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d
+	sha256_main_round 1, %r11d, %r12d, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d
+	sha256_main_round 2, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d, %r8d, %r9d
+	sha256_main_round 3, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d, %r8d
+	paddd	2*16(%rdx), %xmm2
+	movdqa	%xmm2, (%rsp)
+	sha256_main_round 0, %r8d, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d
+	sha256_main_round 1, %r15d, %r8d, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d
+	sha256_main_round 2, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d, %r12d, %r13d
+	sha256_main_round 3, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d, %r12d
+	paddd	3*16(%rdx), %xmm3
+	movdqa	%xmm3, (%rsp)
+	sha256_main_round 0, %r12d, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d, %r11d
+	sha256_main_round 1, %r11d, %r12d, %r13d, %r14d, %r15d, %r8d, %r9d, %r10d
+	sha256_main_round 2, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d, %r8d, %r9d
+	sha256_main_round 3, %r9d, %r10d, %r11d, %r12d, %r13d, %r14d, %r15d, %r8d
+	
+	addl	%r8d, 0*4(%rdi)
+	addl	%r9d, 1*4(%rdi)
+	addl	%r10d, 2*4(%rdi)
+	addl	%r11d, 3*4(%rdi)
+	addl	%r12d, 4*4(%rdi)
+	addl	%r13d, 5*4(%rdi)
+	addl	%r14d, 6*4(%rdi)
+	addl	%r15d, 7*4(%rdi)
+	
+#if defined(_WIN64) || defined(__CYGWIN__)
+	movdqa	1*16(%rsp), %xmm6
+	movdqa	2*16(%rsp), %xmm7
+	movdqa	3*16(%rsp), %xmm8
+	movdqa	4*16(%rsp), %xmm9
+	addq	$5*16, %rsp
+	popq	%rsi
+	popq	%rdi
+#else
+	addq	$16, %rsp
+#endif
+	popq	%r15
+	popq	%r14
+	popq	%r13
+	popq	%r12
+	popq	%rbx
+	ret
+
+
+	.text
+	.p2align 6
+sha256_transform_phe:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+#endif
+	movq	%rsp, %r8
+	subq	$64, %rsp
+	andq	$-64, %rsp
+	
+	testq	%rdx, %rdx
+	jnz sha256_transform_phe_noswap
+	
+	movl	0*4(%rsi), %eax
+	movl	1*4(%rsi), %ecx
+	movl	2*4(%rsi), %edx
+	movl	3*4(%rsi), %r9d
+	bswapl	%eax
+	bswapl	%ecx
+	bswapl	%edx
+	bswapl	%r9d
+	movl	%eax, 0*4(%rsp)
+	movl	%ecx, 1*4(%rsp)
+	movl	%edx, 2*4(%rsp)
+	movl	%r9d, 3*4(%rsp)
+	movl	4*4(%rsi), %eax
+	movl	5*4(%rsi), %ecx
+	movl	6*4(%rsi), %edx
+	movl	7*4(%rsi), %r9d
+	bswapl	%eax
+	bswapl	%ecx
+	bswapl	%edx
+	bswapl	%r9d
+	movl	%eax, 4*4(%rsp)
+	movl	%ecx, 5*4(%rsp)
+	movl	%edx, 6*4(%rsp)
+	movl	%r9d, 7*4(%rsp)
+	
+	movdqu	2*16(%rsi), %xmm0
+	movdqu	3*16(%rsi), %xmm2
+	pshuflw	$0xb1, %xmm0, %xmm0
+	pshuflw	$0xb1, %xmm2, %xmm2
+	pshufhw	$0xb1, %xmm0, %xmm0
+	pshufhw	$0xb1, %xmm2, %xmm2
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm2, %xmm3
+	psrlw	$8, %xmm1
+	psrlw	$8, %xmm3
+	psllw	$8, %xmm0
+	psllw	$8, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm3, %xmm2
+	movdqa	%xmm0, 2*16(%rsp)
+	movdqa	%xmm2, 3*16(%rsp)
+	
+	jmp sha256_transform_phe_core
+	
+sha256_transform_phe_noswap:
+	movdqu	0*16(%rsi), %xmm0
+	movdqu	1*16(%rsi), %xmm1
+	movdqu	2*16(%rsi), %xmm2
+	movdqu	3*16(%rsi), %xmm3
+	movdqa	%xmm0, 0*16(%rsp)
+	movdqa	%xmm1, 1*16(%rsp)
+	movdqa	%xmm2, 2*16(%rsp)
+	movdqa	%xmm3, 3*16(%rsp)
+	
+sha256_transform_phe_core:
+	movq	%rsp, %rsi
+	movq	$-1, %rax
+	movq	$1, %rcx
+	/* rep xsha256 */
+	.byte 0xf3, 0x0f, 0xa6, 0xd0
+	
+	movq	%r8, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	popq	%rdi
+#endif
+	ret
+	
+	
+	.data
+	.p2align 3
+sha256_transform_addr:
+	.quad sha256_transform_sse2
+	
+	.text
+	.p2align 3
+	.globl sha256_transform
+	.globl _sha256_transform
+sha256_transform:
+_sha256_transform:
+	jmp *sha256_transform_addr(%rip)
+
+
+	.text
+	.p2align 6
+	.globl sha256d_ms
+	.globl _sha256d_ms
+sha256d_ms:
+_sha256d_ms:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+#endif
+	movq	%rsp, %r8
+	subq	$32, %rsp
+	andq	$-32, %rsp
+	
+	movdqa	0*16(%rdx), %xmm0
+	movdqa	1*16(%rdx), %xmm1
+	movdqa	%xmm0, 0*16(%rdi)
+	movdqa	%xmm1, 1*16(%rdi)
+	
+	movl	0*4(%rsi), %eax
+	movl	1*4(%rsi), %ecx
+	movl	2*4(%rsi), %edx
+	movl	3*4(%rsi), %r9d
+	bswapl	%eax
+	bswapl	%ecx
+	bswapl	%edx
+	bswapl	%r9d
+	movl	%eax, 0*4(%rsp)
+	movl	%ecx, 1*4(%rsp)
+	movl	%edx, 2*4(%rsp)
+	movl	%r9d, 3*4(%rsp)
+	
+	movq	%rsp, %rsi
+	movl	$64, %eax
+	movl	$80, %ecx
+	/* rep xsha256 */
+	.byte 0xf3, 0x0f, 0xa6, 0xd0
+	
+	movdqa	bswap_xmm_mask(%rip), %xmm1
+	movdqa	0*16(%rdi), %xmm0
+	movdqa	1*16(%rdi), %xmm2
+	pshufb	%xmm1, %xmm0
+	pshufb	%xmm1, %xmm2
+	movdqa	%xmm0, 0*16(%rsp)
+	movdqa	%xmm2, 1*16(%rsp)
+	
+	movdqa	sha256_h+0*16(%rip), %xmm0
+	movdqa	sha256_h+1*16(%rip), %xmm1
+	movdqa	%xmm0, 0*16(%rdi)
+	movdqa	%xmm1, 1*16(%rdi)
+	
+	movq	%rsp, %rsi
+	xorq	%rax, %rax
+	movl	$32, %ecx
+	/* rep xsha256 */
+	.byte 0xf3, 0x0f, 0xa6, 0xd0
+	
+	movq	%r8, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	popq	%rdi
+#endif
+	ret
+
+
+	.data
+	.p2align 7
+sha256_4h:
+	.long 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667
+	.long 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85
+	.long 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372
+	.long 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a
+	.long 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f
+	.long 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c
+	.long 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab
+	.long 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19
+
+	.data
+	.p2align 7
+sha256_4k:
+	.long 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98
+	.long 0x71374491, 0x71374491, 0x71374491, 0x71374491
+	.long 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf
+	.long 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5
+	.long 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b
+	.long 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1
+	.long 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4
+	.long 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5
+	.long 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98
+	.long 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01
+	.long 0x243185be, 0x243185be, 0x243185be, 0x243185be
+	.long 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3
+	.long 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74
+	.long 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe
+	.long 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7
+	.long 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174
+	.long 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1
+	.long 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786
+	.long 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6
+	.long 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc
+	.long 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f
+	.long 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa
+	.long 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc
+	.long 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da
+	.long 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152
+	.long 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d
+	.long 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8
+	.long 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7
+	.long 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3
+	.long 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147
+	.long 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351
+	.long 0x14292967, 0x14292967, 0x14292967, 0x14292967
+	.long 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85
+	.long 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138
+	.long 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc
+	.long 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13
+	.long 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354
+	.long 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb
+	.long 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e
+	.long 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85
+	.long 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1
+	.long 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b
+	.long 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70
+	.long 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3
+	.long 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819
+	.long 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624
+	.long 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585
+	.long 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070
+	.long 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116
+	.long 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08
+	.long 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c
+	.long 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5
+	.long 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3
+	.long 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a
+	.long 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f
+	.long 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3
+	.long 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee
+	.long 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f
+	.long 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814
+	.long 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208
+	.long 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa
+	.long 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb
+	.long 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7
+	.long 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2
+
+	.data
+	.p2align 6
+sha256d_4preext2_17:
+	.long 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000
+sha256d_4preext2_23:
+	.long 0x11002000, 0x11002000, 0x11002000, 0x11002000
+sha256d_4preext2_24:
+	.long 0x80000000, 0x80000000, 0x80000000, 0x80000000
+sha256d_4preext2_30:
+	.long 0x00400022, 0x00400022, 0x00400022, 0x00400022
+
+
+#ifdef USE_AVX2
+
+	.data
+	.p2align 7
+sha256_8h:
+	.long 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667
+	.long 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85
+	.long 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372
+	.long 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a
+	.long 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f
+	.long 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c
+	.long 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab
+	.long 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19
+
+	.data
+	.p2align 7
+sha256_8k:
+	.long 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98
+	.long 0x71374491, 0x71374491, 0x71374491, 0x71374491, 0x71374491, 0x71374491, 0x71374491, 0x71374491
+	.long 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf
+	.long 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5
+	.long 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b
+	.long 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1
+	.long 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4
+	.long 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5
+	.long 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98
+	.long 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01
+	.long 0x243185be, 0x243185be, 0x243185be, 0x243185be, 0x243185be, 0x243185be, 0x243185be, 0x243185be
+	.long 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3
+	.long 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74
+	.long 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe
+	.long 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7
+	.long 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174
+	.long 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1
+	.long 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786
+	.long 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6
+	.long 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc
+	.long 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f
+	.long 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa
+	.long 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc
+	.long 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da
+	.long 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152
+	.long 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d
+	.long 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8
+	.long 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7
+	.long 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3
+	.long 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147
+	.long 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351
+	.long 0x14292967, 0x14292967, 0x14292967, 0x14292967, 0x14292967, 0x14292967, 0x14292967, 0x14292967
+	.long 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85
+	.long 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138
+	.long 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc
+	.long 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13
+	.long 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354
+	.long 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb
+	.long 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e
+	.long 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85
+	.long 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1
+	.long 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b
+	.long 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70
+	.long 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3
+	.long 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819
+	.long 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624
+	.long 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585
+	.long 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070
+	.long 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116
+	.long 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08
+	.long 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c
+	.long 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5
+	.long 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3
+	.long 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a
+	.long 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f
+	.long 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3
+	.long 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee
+	.long 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f
+	.long 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814
+	.long 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208
+	.long 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa
+	.long 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb
+	.long 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7
+	.long 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2
+
+	.data
+	.p2align 6
+sha256d_8preext2_17:
+	.long 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000
+sha256d_8preext2_23:
+	.long 0x11002000, 0x11002000, 0x11002000, 0x11002000, 0x11002000, 0x11002000, 0x11002000, 0x11002000
+sha256d_8preext2_24:
+	.long 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000
+sha256d_8preext2_30:
+	.long 0x00400022, 0x00400022, 0x00400022, 0x00400022, 0x00400022, 0x00400022, 0x00400022, 0x00400022
+
+#endif /* USE_AVX2 */
+
+
+	.text
+	.p2align 6
+	.globl sha256_init_4way
+	.globl _sha256_init_4way
+sha256_init_4way:
+_sha256_init_4way:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	movq	%rcx, %rdi
+#endif
+	movdqa	sha256_4h+0(%rip), %xmm0
+	movdqa	sha256_4h+16(%rip), %xmm1
+	movdqa	sha256_4h+32(%rip), %xmm2
+	movdqa	sha256_4h+48(%rip), %xmm3
+	movdqu	%xmm0, 0(%rdi)
+	movdqu	%xmm1, 16(%rdi)
+	movdqu	%xmm2, 32(%rdi)
+	movdqu	%xmm3, 48(%rdi)
+	movdqa	sha256_4h+64(%rip), %xmm0
+	movdqa	sha256_4h+80(%rip), %xmm1
+	movdqa	sha256_4h+96(%rip), %xmm2
+	movdqa	sha256_4h+112(%rip), %xmm3
+	movdqu	%xmm0, 64(%rdi)
+	movdqu	%xmm1, 80(%rdi)
+	movdqu	%xmm2, 96(%rdi)
+	movdqu	%xmm3, 112(%rdi)
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rdi
+#endif
+	ret
+
+
+#ifdef USE_AVX2
+	.text
+	.p2align 6
+	.globl sha256_init_8way
+	.globl _sha256_init_8way
+sha256_init_8way:
+_sha256_init_8way:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	movq	%rcx, %rdi
+#endif
+	vpbroadcastd	sha256_4h+0(%rip), %ymm0
+	vpbroadcastd	sha256_4h+16(%rip), %ymm1
+	vpbroadcastd	sha256_4h+32(%rip), %ymm2
+	vpbroadcastd	sha256_4h+48(%rip), %ymm3
+	vmovdqu	%ymm0, 0*32(%rdi)
+	vmovdqu	%ymm1, 1*32(%rdi)
+	vmovdqu	%ymm2, 2*32(%rdi)
+	vmovdqu	%ymm3, 3*32(%rdi)
+	vpbroadcastd	sha256_4h+64(%rip), %ymm0
+	vpbroadcastd	sha256_4h+80(%rip), %ymm1
+	vpbroadcastd	sha256_4h+96(%rip), %ymm2
+	vpbroadcastd	sha256_4h+112(%rip), %ymm3
+	vmovdqu	%ymm0, 4*32(%rdi)
+	vmovdqu	%ymm1, 5*32(%rdi)
+	vmovdqu	%ymm2, 6*32(%rdi)
+	vmovdqu	%ymm3, 7*32(%rdi)
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rdi
+#endif
+	ret
+#endif /* USE_AVX2 */
+
+
+.macro sha256_sse2_extend_round i
+	movdqa	(\i-15)*16(%rax), %xmm0
+	movdqa	%xmm0, %xmm2
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	(\i-16)*16(%rax), %xmm0
+	paddd	(\i-7)*16(%rax), %xmm0
+
+	movdqa	%xmm3, %xmm2
+	psrld	$10, %xmm3
+	pslld	$13, %xmm2
+	movdqa	%xmm3, %xmm1
+	psrld	$7, %xmm1
+	pxor	%xmm1, %xmm3
+	pxor	%xmm2, %xmm3
+	psrld	$2, %xmm1
+	pslld	$2, %xmm2
+	pxor	%xmm1, %xmm3
+	pxor	%xmm2, %xmm3
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, \i*16(%rax)
+.endm
+
+.macro sha256_sse2_extend_doubleround i
+	movdqa	(\i-15)*16(%rax), %xmm0
+	movdqa	(\i-14)*16(%rax), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+
+	paddd	(\i-16)*16(%rax), %xmm0
+	paddd	(\i-15)*16(%rax), %xmm4
+
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+
+	paddd	(\i-7)*16(%rax), %xmm0
+	paddd	(\i-6)*16(%rax), %xmm4
+
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, \i*16(%rax)
+	movdqa	%xmm7, (\i+1)*16(%rax)
+.endm
+
+.macro sha256_sse2_main_round i
+	movdqa	16*(\i)(%rax), %xmm6
+
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%rsp), %xmm2
+	pandn	%xmm2, %xmm1
+	paddd	32(%rsp), %xmm6
+
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	0(%rsp), %xmm2
+	movdqa	%xmm2, 16(%rsp)
+
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%rsp)
+
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	paddd	16*(\i)(%rcx), %xmm6
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm2, %xmm0
+	pxor	%xmm1, %xmm0
+	movdqa	%xmm5, %xmm1
+	paddd	%xmm0, %xmm6
+
+	movdqa	%xmm3, %xmm0
+	movdqa	%xmm4, %xmm3
+	movdqa	%xmm4, %xmm2
+	paddd	%xmm6, %xmm0
+	pand	%xmm5, %xmm2
+	pand	%xmm7, %xmm1
+	pand	%xmm7, %xmm4
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm5, %xmm4
+	movdqa	%xmm7, %xmm5
+	pxor	%xmm2, %xmm1
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm7, %xmm2
+	psrld	$2, %xmm7
+	movdqa	%xmm7, %xmm1
+	pslld	$10, %xmm2
+	psrld	$11, %xmm1
+	pxor	%xmm2, %xmm7
+	pslld	$9, %xmm2
+	pxor	%xmm1, %xmm7
+	psrld	$9, %xmm1
+	pxor	%xmm2, %xmm7
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm7
+	pxor	%xmm2, %xmm7
+	paddd	%xmm6, %xmm7
+.endm
+
+.macro sha256_sse2_main_quadround i
+	sha256_sse2_main_round \i+0
+	sha256_sse2_main_round \i+1
+	sha256_sse2_main_round \i+2
+	sha256_sse2_main_round \i+3
+.endm
+
+
+#if defined(USE_AVX)
+
+.macro sha256_avx_extend_round i
+	vmovdqa	(\i-15)*16(%rax), %xmm0
+	vpslld	$14, %xmm0, %xmm2
+	vpsrld	$3, %xmm0, %xmm0
+	vpsrld	$4, %xmm0, %xmm1
+	vpxor	%xmm1, %xmm0, %xmm0
+	vpxor	%xmm2, %xmm0, %xmm0
+	vpsrld	$11, %xmm1, %xmm1
+	vpslld	$11, %xmm2, %xmm2
+	vpxor	%xmm1, %xmm0, %xmm0
+	vpxor	%xmm2, %xmm0, %xmm0
+	vpaddd	(\i-16)*16(%rax), %xmm0, %xmm0
+	vpaddd	(\i-7)*16(%rax), %xmm0, %xmm0
+
+	vpslld	$13, %xmm3, %xmm2
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$7, %xmm3, %xmm1
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpsrld	$2, %xmm1, %xmm1
+	vpslld	$2, %xmm2, %xmm2
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vmovdqa	%xmm3, \i*16(%rax)
+.endm
+
+.macro sha256_avx_extend_doubleround i
+	vmovdqa	(\i-15)*16(%rax), %xmm0
+	vmovdqa	(\i-14)*16(%rax), %xmm4
+	vpslld	$14, %xmm0, %xmm2
+	vpslld	$14, %xmm4, %xmm6
+	vpsrld	$3, %xmm0, %xmm8
+	vpsrld	$3, %xmm4, %xmm4
+	vpsrld	$7, %xmm0, %xmm1
+	vpsrld	$4, %xmm4, %xmm5
+	vpxor	%xmm1, %xmm8, %xmm8
+	vpxor	%xmm5, %xmm4, %xmm4
+	vpsrld	$11, %xmm1, %xmm1
+	vpsrld	$11, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpslld	$11, %xmm2, %xmm2
+	vpslld	$11, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm8, %xmm8
+	vpxor	%xmm5, %xmm4, %xmm4
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm4, %xmm4
+
+	vpaddd	%xmm0, %xmm4, %xmm4
+	vpaddd	(\i-16)*16(%rax), %xmm8, %xmm0
+
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+
+	vpaddd	(\i-7)*16(%rax), %xmm0, %xmm0
+	vpaddd	(\i-6)*16(%rax), %xmm4, %xmm4
+
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, \i*16(%rax)
+	vmovdqa	%xmm7, (\i+1)*16(%rax)
+.endm
+
+.macro sha256_avx_main_round i, r0, r1, r2, r3, r4, r5, r6, r7
+	vpaddd	16*(\i)(%rax), \r0, %xmm6
+	vpaddd	16*(\i)(%rcx), %xmm6, %xmm6
+
+	vpandn	\r1, \r3, %xmm1
+	vpand	\r3, \r2, %xmm2
+	vpxor	%xmm2, %xmm1, %xmm1
+	vpaddd	%xmm1, %xmm6, %xmm6
+
+	vpslld	$7, \r3, %xmm1
+	vpsrld	$6, \r3, \r0
+	vpsrld	$5, \r0, %xmm2
+	vpxor	%xmm1, \r0, \r0
+	vpxor	%xmm2, \r0, \r0
+	vpslld	$14, %xmm1, %xmm1
+	vpsrld	$14, %xmm2, %xmm2
+	vpxor	%xmm1, \r0, \r0
+	vpxor	%xmm2, \r0, \r0
+	vpslld	$5, %xmm1, %xmm1
+	vpxor	%xmm1, \r0, \r0
+	vpaddd	\r0, %xmm6, %xmm6
+	vpaddd	%xmm6, \r4, \r0
+
+	vpand	\r6, \r5, %xmm2
+	vpand	\r7, \r5, \r4
+	vpand	\r7, \r6, %xmm1
+	vpxor	\r4, %xmm1, %xmm1
+	vpxor	%xmm2, %xmm1, %xmm1
+	vpaddd	%xmm1, %xmm6, %xmm6
+
+	vpslld	$10, \r7, %xmm2
+	vpsrld	$2, \r7, \r4
+	vpsrld	$11, \r4, %xmm1
+	vpxor	%xmm2, \r4, \r4
+	vpxor	%xmm1, \r4, \r4
+	vpslld	$9, %xmm2, %xmm2
+	vpsrld	$9, %xmm1, %xmm1
+	vpxor	%xmm2, \r4, \r4
+	vpxor	%xmm1, \r4, \r4
+	vpslld	$11, %xmm2, %xmm2
+	vpxor	%xmm2, \r4, \r4
+	vpaddd	%xmm6, \r4, \r4
+.endm
+
+.macro sha256_avx_main_quadround i
+	sha256_avx_main_round \i+0, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3, %xmm4, %xmm5, %xmm7
+	sha256_avx_main_round \i+1, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4, %xmm5, %xmm7, %xmm3
+	sha256_avx_main_round \i+2, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5, %xmm7, %xmm3, %xmm4
+	sha256_avx_main_round \i+3, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7, %xmm3, %xmm4, %xmm5
+.endm
+
+#endif /* USE_AVX */
+
+
+#if defined(USE_AVX2)
+
+.macro sha256_avx2_extend_round i
+	vmovdqa	(\i-15)*32(%rax), %ymm0
+	vpslld	$14, %ymm0, %ymm2
+	vpsrld	$3, %ymm0, %ymm0
+	vpsrld	$4, %ymm0, %ymm1
+	vpxor	%ymm1, %ymm0, %ymm0
+	vpxor	%ymm2, %ymm0, %ymm0
+	vpsrld	$11, %ymm1, %ymm1
+	vpslld	$11, %ymm2, %ymm2
+	vpxor	%ymm1, %ymm0, %ymm0
+	vpxor	%ymm2, %ymm0, %ymm0
+	vpaddd	(\i-16)*32(%rax), %ymm0, %ymm0
+	vpaddd	(\i-7)*32(%rax), %ymm0, %ymm0
+
+	vpslld	$13, %ymm3, %ymm2
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$7, %ymm3, %ymm1
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpsrld	$2, %ymm1, %ymm1
+	vpslld	$2, %ymm2, %ymm2
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vmovdqa	%ymm3, \i*32(%rax)
+.endm
+
+.macro sha256_avx2_extend_doubleround i
+	vmovdqa	(\i-15)*32(%rax), %ymm0
+	vmovdqa	(\i-14)*32(%rax), %ymm4
+	vpslld	$14, %ymm0, %ymm2
+	vpslld	$14, %ymm4, %ymm6
+	vpsrld	$3, %ymm0, %ymm8
+	vpsrld	$3, %ymm4, %ymm4
+	vpsrld	$7, %ymm0, %ymm1
+	vpsrld	$4, %ymm4, %ymm5
+	vpxor	%ymm1, %ymm8, %ymm8
+	vpxor	%ymm5, %ymm4, %ymm4
+	vpsrld	$11, %ymm1, %ymm1
+	vpsrld	$11, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm8, %ymm8
+	vpxor	%ymm6, %ymm4, %ymm4
+	vpslld	$11, %ymm2, %ymm2
+	vpslld	$11, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm8, %ymm8
+	vpxor	%ymm5, %ymm4, %ymm4
+	vpxor	%ymm2, %ymm8, %ymm8
+	vpxor	%ymm6, %ymm4, %ymm4
+
+	vpaddd	%ymm0, %ymm4, %ymm4
+	vpaddd	(\i-16)*32(%rax), %ymm8, %ymm0
+
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+
+	vpaddd	(\i-7)*32(%rax), %ymm0, %ymm0
+	vpaddd	(\i-6)*32(%rax), %ymm4, %ymm4
+
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vpaddd	%ymm4, %ymm7, %ymm7
+	vmovdqa	%ymm3, \i*32(%rax)
+	vmovdqa	%ymm7, (\i+1)*32(%rax)
+.endm
+
+.macro sha256_avx2_main_round i, r0, r1, r2, r3, r4, r5, r6, r7
+	vpaddd	32*(\i)(%rax), \r0, %ymm6
+	vpaddd	32*(\i)(%rcx), %ymm6, %ymm6
+
+	vpandn	\r1, \r3, %ymm1
+	vpand	\r3, \r2, %ymm2
+	vpxor	%ymm2, %ymm1, %ymm1
+	vpaddd	%ymm1, %ymm6, %ymm6
+
+	vpslld	$7, \r3, %ymm1
+	vpsrld	$6, \r3, \r0
+	vpsrld	$5, \r0, %ymm2
+	vpxor	%ymm1, \r0, \r0
+	vpxor	%ymm2, \r0, \r0
+	vpslld	$14, %ymm1, %ymm1
+	vpsrld	$14, %ymm2, %ymm2
+	vpxor	%ymm1, \r0, \r0
+	vpxor	%ymm2, \r0, \r0
+	vpslld	$5, %ymm1, %ymm1
+	vpxor	%ymm1, \r0, \r0
+	vpaddd	\r0, %ymm6, %ymm6
+	vpaddd	%ymm6, \r4, \r0
+
+	vpand	\r6, \r5, %ymm2
+	vpand	\r7, \r5, \r4
+	vpand	\r7, \r6, %ymm1
+	vpxor	\r4, %ymm1, %ymm1
+	vpxor	%ymm2, %ymm1, %ymm1
+	vpaddd	%ymm1, %ymm6, %ymm6
+
+	vpslld	$10, \r7, %ymm2
+	vpsrld	$2, \r7, \r4
+	vpsrld	$11, \r4, %ymm1
+	vpxor	%ymm2, \r4, \r4
+	vpxor	%ymm1, \r4, \r4
+	vpslld	$9, %ymm2, %ymm2
+	vpsrld	$9, %ymm1, %ymm1
+	vpxor	%ymm2, \r4, \r4
+	vpxor	%ymm1, \r4, \r4
+	vpslld	$11, %ymm2, %ymm2
+	vpxor	%ymm2, \r4, \r4
+	vpaddd	%ymm6, \r4, \r4
+.endm
+
+.macro sha256_avx2_main_quadround i
+	sha256_avx2_main_round \i+0, %ymm10, %ymm9, %ymm8, %ymm0, %ymm3, %ymm4, %ymm5, %ymm7
+	sha256_avx2_main_round \i+1, %ymm9, %ymm8, %ymm0, %ymm10, %ymm4, %ymm5, %ymm7, %ymm3
+	sha256_avx2_main_round \i+2, %ymm8, %ymm0, %ymm10, %ymm9, %ymm5, %ymm7, %ymm3, %ymm4
+	sha256_avx2_main_round \i+3, %ymm0, %ymm10, %ymm9, %ymm8, %ymm7, %ymm3, %ymm4, %ymm5
+.endm
+
+#endif /* USE_AVX2 */
+
+
+#if defined(USE_XOP)
+
+.macro sha256_xop_extend_round i
+	vmovdqa	(\i-15)*16(%rax), %xmm0
+	vprotd	$25, %xmm0, %xmm1
+	vprotd	$14, %xmm0, %xmm2
+	vpsrld	$3, %xmm0, %xmm0
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm2, %xmm0, %xmm0
+
+	vpaddd	(\i-16)*16(%rax), %xmm0, %xmm0
+	vpaddd	(\i-7)*16(%rax), %xmm0, %xmm0
+
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$13, %xmm3, %xmm2
+	vpsrld	$10, %xmm3, %xmm3
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vmovdqa	%xmm3, \i*16(%rax)
+.endm
+
+.macro sha256_xop_extend_doubleround i
+	vmovdqa	(\i-15)*16(%rax), %xmm0
+	vmovdqa	(\i-14)*16(%rax), %xmm4
+	vprotd	$25, %xmm0, %xmm1
+	vprotd	$25, %xmm4, %xmm5
+	vprotd	$14, %xmm0, %xmm2
+	vprotd	$14, %xmm4, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$3, %xmm0, %xmm0
+	vpsrld	$3, %xmm4, %xmm4
+	vpxor	%xmm2, %xmm0, %xmm0
+	vpxor	%xmm6, %xmm4, %xmm4
+
+	vpaddd	(\i-16)*16(%rax), %xmm0, %xmm0
+	vpaddd	(\i-15)*16(%rax), %xmm4, %xmm4
+
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+
+	vpaddd	(\i-7)*16(%rax), %xmm0, %xmm0
+	vpaddd	(\i-6)*16(%rax), %xmm4, %xmm4
+
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, \i*16(%rax)
+	vmovdqa	%xmm7, (\i+1)*16(%rax)
+.endm
+	
+.macro sha256_xop_main_round i, r0, r1, r2, r3, r4, r5, r6, r7
+	vpaddd	16*(\i)(%rax), \r0, %xmm6
+	vpaddd	16*(\i)(%rcx), %xmm6, %xmm6
+
+	vpandn	\r1, \r3, %xmm1
+	vpand	\r3, \r2, %xmm2
+	vpxor	%xmm2, %xmm1, %xmm1
+	vpaddd	%xmm1, %xmm6, %xmm6
+
+	vprotd	$26, \r3, %xmm1
+	vprotd	$21, \r3, %xmm2
+	vpxor	%xmm1, %xmm2, %xmm2
+	vprotd	$7, \r3, \r0
+	vpxor	%xmm2, \r0, \r0
+	vpaddd	\r0, %xmm6, %xmm6
+	vpaddd	%xmm6, \r4, \r0
+
+	vpand	\r6, \r5, %xmm2
+	vpand	\r7, \r5, \r4
+	vpand	\r7, \r6, %xmm1
+	vpxor	\r4, %xmm1, %xmm1
+	vpxor	%xmm2, %xmm1, %xmm1
+	vpaddd	%xmm1, %xmm6, %xmm6
+
+	vprotd	$30, \r7, %xmm1
+	vprotd	$19, \r7, %xmm2
+	vpxor	%xmm1, %xmm2, %xmm2
+	vprotd	$10, \r7, \r4
+	vpxor	%xmm2, \r4, \r4
+	vpaddd	%xmm6, \r4, \r4
+.endm
+
+.macro sha256_xop_main_quadround i
+	sha256_xop_main_round \i+0, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3, %xmm4, %xmm5, %xmm7
+	sha256_xop_main_round \i+1, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4, %xmm5, %xmm7, %xmm3
+	sha256_xop_main_round \i+2, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5, %xmm7, %xmm3, %xmm4
+	sha256_xop_main_round \i+3, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7, %xmm3, %xmm4, %xmm5
+.endm
+
+#endif /* USE_XOP */
+
+
+	.text
+	.p2align 6
+sha256_transform_4way_core_sse2:
+	leaq	256(%rsp), %rcx
+	leaq	48*16(%rcx), %rax
+	movdqa	-2*16(%rcx), %xmm3
+	movdqa	-1*16(%rcx), %xmm7
+sha256_transform_4way_sse2_extend_loop:
+	movdqa	-15*16(%rcx), %xmm0
+	movdqa	-14*16(%rcx), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+
+	paddd	-16*16(%rcx), %xmm0
+	paddd	-15*16(%rcx), %xmm4
+
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+
+	paddd	-7*16(%rcx), %xmm0
+	paddd	-6*16(%rcx), %xmm4
+
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, (%rcx)
+	movdqa	%xmm7, 16(%rcx)
+	addq	$2*16, %rcx
+	cmpq	%rcx, %rax
+	jne sha256_transform_4way_sse2_extend_loop
+	
+	movdqu	0(%rdi), %xmm7
+	movdqu	16(%rdi), %xmm5
+	movdqu	32(%rdi), %xmm4
+	movdqu	48(%rdi), %xmm3
+	movdqu	64(%rdi), %xmm0
+	movdqu	80(%rdi), %xmm8
+	movdqu	96(%rdi), %xmm9
+	movdqu	112(%rdi), %xmm10
+	
+	leaq	sha256_4k(%rip), %rcx
+	xorq	%rax, %rax
+sha256_transform_4way_sse2_main_loop:
+	movdqa	(%rsp, %rax), %xmm6
+	paddd	(%rcx, %rax), %xmm6
+	paddd	%xmm10, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm9, %xmm2
+	pandn	%xmm2, %xmm1
+
+	movdqa	%xmm2, %xmm10
+	movdqa	%xmm8, %xmm2
+	movdqa	%xmm2, %xmm9
+
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, %xmm8
+
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm1, %xmm0
+	paddd	%xmm0, %xmm6
+
+	movdqa	%xmm3, %xmm0
+	paddd	%xmm6, %xmm0
+
+	movdqa	%xmm5, %xmm1
+	movdqa	%xmm4, %xmm3
+	movdqa	%xmm4, %xmm2
+	pand	%xmm5, %xmm2
+	pand	%xmm7, %xmm4
+	pand	%xmm7, %xmm1
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm5, %xmm4
+	movdqa	%xmm7, %xmm5
+	pxor	%xmm2, %xmm1
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm7, %xmm2
+	psrld	$2, %xmm7
+	movdqa	%xmm7, %xmm1
+	pslld	$10, %xmm2
+	psrld	$11, %xmm1
+	pxor	%xmm2, %xmm7
+	pxor	%xmm1, %xmm7
+	pslld	$9, %xmm2
+	psrld	$9, %xmm1
+	pxor	%xmm2, %xmm7
+	pxor	%xmm1, %xmm7
+	pslld	$11, %xmm2
+	pxor	%xmm2, %xmm7
+	paddd	%xmm6, %xmm7
+	
+	addq	$16, %rax
+	cmpq	$16*64, %rax
+	jne sha256_transform_4way_sse2_main_loop
+	jmp sha256_transform_4way_finish
+
+
+#if defined(USE_AVX)
+	.text
+	.p2align 6
+sha256_transform_4way_core_avx:
+	leaq	256(%rsp), %rax
+	movdqa	-2*16(%rax), %xmm3
+	movdqa	-1*16(%rax), %xmm7
+	sha256_avx_extend_doubleround 0
+	sha256_avx_extend_doubleround 2
+	sha256_avx_extend_doubleround 4
+	sha256_avx_extend_doubleround 6
+	sha256_avx_extend_doubleround 8
+	sha256_avx_extend_doubleround 10
+	sha256_avx_extend_doubleround 12
+	sha256_avx_extend_doubleround 14
+	sha256_avx_extend_doubleround 16
+	sha256_avx_extend_doubleround 18
+	sha256_avx_extend_doubleround 20
+	sha256_avx_extend_doubleround 22
+	sha256_avx_extend_doubleround 24
+	sha256_avx_extend_doubleround 26
+	sha256_avx_extend_doubleround 28
+	sha256_avx_extend_doubleround 30
+	sha256_avx_extend_doubleround 32
+	sha256_avx_extend_doubleround 34
+	sha256_avx_extend_doubleround 36
+	sha256_avx_extend_doubleround 38
+	sha256_avx_extend_doubleround 40
+	sha256_avx_extend_doubleround 42
+	sha256_avx_extend_doubleround 44
+	sha256_avx_extend_doubleround 46
+	movdqu	0(%rdi), %xmm7
+	movdqu	16(%rdi), %xmm5
+	movdqu	32(%rdi), %xmm4
+	movdqu	48(%rdi), %xmm3
+	movdqu	64(%rdi), %xmm0
+	movdqu	80(%rdi), %xmm8
+	movdqu	96(%rdi), %xmm9
+	movdqu	112(%rdi), %xmm10
+	movq	%rsp, %rax
+	leaq	sha256_4k(%rip), %rcx
+	sha256_avx_main_quadround 0
+	sha256_avx_main_quadround 4
+	sha256_avx_main_quadround 8
+	sha256_avx_main_quadround 12
+	sha256_avx_main_quadround 16
+	sha256_avx_main_quadround 20
+	sha256_avx_main_quadround 24
+	sha256_avx_main_quadround 28
+	sha256_avx_main_quadround 32
+	sha256_avx_main_quadround 36
+	sha256_avx_main_quadround 40
+	sha256_avx_main_quadround 44
+	sha256_avx_main_quadround 48
+	sha256_avx_main_quadround 52
+	sha256_avx_main_quadround 56
+	sha256_avx_main_quadround 60
+	jmp sha256_transform_4way_finish
+#endif /* USE_AVX */
+
+
+#if defined(USE_XOP)
+	.text
+	.p2align 6
+sha256_transform_4way_core_xop:
+	leaq	256(%rsp), %rax
+	movdqa	-2*16(%rax), %xmm3
+	movdqa	-1*16(%rax), %xmm7
+	sha256_xop_extend_doubleround 0
+	sha256_xop_extend_doubleround 2
+	sha256_xop_extend_doubleround 4
+	sha256_xop_extend_doubleround 6
+	sha256_xop_extend_doubleround 8
+	sha256_xop_extend_doubleround 10
+	sha256_xop_extend_doubleround 12
+	sha256_xop_extend_doubleround 14
+	sha256_xop_extend_doubleround 16
+	sha256_xop_extend_doubleround 18
+	sha256_xop_extend_doubleround 20
+	sha256_xop_extend_doubleround 22
+	sha256_xop_extend_doubleround 24
+	sha256_xop_extend_doubleround 26
+	sha256_xop_extend_doubleround 28
+	sha256_xop_extend_doubleround 30
+	sha256_xop_extend_doubleround 32
+	sha256_xop_extend_doubleround 34
+	sha256_xop_extend_doubleround 36
+	sha256_xop_extend_doubleround 38
+	sha256_xop_extend_doubleround 40
+	sha256_xop_extend_doubleround 42
+	sha256_xop_extend_doubleround 44
+	sha256_xop_extend_doubleround 46
+	movdqu	0(%rdi), %xmm7
+	movdqu	16(%rdi), %xmm5
+	movdqu	32(%rdi), %xmm4
+	movdqu	48(%rdi), %xmm3
+	movdqu	64(%rdi), %xmm0
+	movdqu	80(%rdi), %xmm8
+	movdqu	96(%rdi), %xmm9
+	movdqu	112(%rdi), %xmm10
+	movq	%rsp, %rax
+	leaq	sha256_4k(%rip), %rcx
+	sha256_xop_main_quadround 0
+	sha256_xop_main_quadround 4
+	sha256_xop_main_quadround 8
+	sha256_xop_main_quadround 12
+	sha256_xop_main_quadround 16
+	sha256_xop_main_quadround 20
+	sha256_xop_main_quadround 24
+	sha256_xop_main_quadround 28
+	sha256_xop_main_quadround 32
+	sha256_xop_main_quadround 36
+	sha256_xop_main_quadround 40
+	sha256_xop_main_quadround 44
+	sha256_xop_main_quadround 48
+	sha256_xop_main_quadround 52
+	sha256_xop_main_quadround 56
+	sha256_xop_main_quadround 60
+	jmp sha256_transform_4way_finish
+#endif /* USE_XOP */
+
+
+	.data
+	.p2align 3
+sha256_transform_4way_core_addr:
+	.quad 0x0
+
+.macro p2bswap_rsi_rsp i
+	movdqu	\i*16(%rsi), %xmm0
+	movdqu	(\i+1)*16(%rsi), %xmm2
+	pshuflw	$0xb1, %xmm0, %xmm0
+	pshuflw	$0xb1, %xmm2, %xmm2
+	pshufhw	$0xb1, %xmm0, %xmm0
+	pshufhw	$0xb1, %xmm2, %xmm2
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm2, %xmm3
+	psrlw	$8, %xmm1
+	psrlw	$8, %xmm3
+	psllw	$8, %xmm0
+	psllw	$8, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm3, %xmm2
+	movdqa	%xmm0, \i*16(%rsp)
+	movdqa	%xmm2, (\i+1)*16(%rsp)
+.endm
+	
+	.text
+	.p2align 6
+	.globl sha256_transform_4way
+	.globl _sha256_transform_4way
+sha256_transform_4way:
+_sha256_transform_4way:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	subq	$96, %rsp
+	movdqa	%xmm6, 0(%rsp)
+	movdqa	%xmm7, 16(%rsp)
+	movdqa	%xmm8, 32(%rsp)
+	movdqa	%xmm9, 48(%rsp)
+	movdqa	%xmm10, 64(%rsp)
+	movdqa	%xmm11, 80(%rsp)
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+#endif
+	movq	%rsp, %r8
+	subq	$1032, %rsp
+	andq	$-128, %rsp
+	
+	testq	%rdx, %rdx
+	jnz sha256_transform_4way_swap
+	
+	movdqu	0*16(%rsi), %xmm0
+	movdqu	1*16(%rsi), %xmm1
+	movdqu	2*16(%rsi), %xmm2
+	movdqu	3*16(%rsi), %xmm3
+	movdqu	4*16(%rsi), %xmm4
+	movdqu	5*16(%rsi), %xmm5
+	movdqu	6*16(%rsi), %xmm6
+	movdqu	7*16(%rsi), %xmm7
+	movdqa	%xmm0, 0*16(%rsp)
+	movdqa	%xmm1, 1*16(%rsp)
+	movdqa	%xmm2, 2*16(%rsp)
+	movdqa	%xmm3, 3*16(%rsp)
+	movdqa	%xmm4, 4*16(%rsp)
+	movdqa	%xmm5, 5*16(%rsp)
+	movdqa	%xmm6, 6*16(%rsp)
+	movdqa	%xmm7, 7*16(%rsp)
+	movdqu	8*16(%rsi), %xmm0
+	movdqu	9*16(%rsi), %xmm1
+	movdqu	10*16(%rsi), %xmm2
+	movdqu	11*16(%rsi), %xmm3
+	movdqu	12*16(%rsi), %xmm4
+	movdqu	13*16(%rsi), %xmm5
+	movdqu	14*16(%rsi), %xmm6
+	movdqu	15*16(%rsi), %xmm7
+	movdqa	%xmm0, 8*16(%rsp)
+	movdqa	%xmm1, 9*16(%rsp)
+	movdqa	%xmm2, 10*16(%rsp)
+	movdqa	%xmm3, 11*16(%rsp)
+	movdqa	%xmm4, 12*16(%rsp)
+	movdqa	%xmm5, 13*16(%rsp)
+	movdqa	%xmm6, 14*16(%rsp)
+	movdqa	%xmm7, 15*16(%rsp)
+	jmp *sha256_transform_4way_core_addr(%rip)
+	
+	.p2align 6
+sha256_transform_4way_swap:
+	p2bswap_rsi_rsp 0
+	p2bswap_rsi_rsp 2
+	p2bswap_rsi_rsp 4
+	p2bswap_rsi_rsp 6
+	p2bswap_rsi_rsp 8
+	p2bswap_rsi_rsp 10
+	p2bswap_rsi_rsp 12
+	p2bswap_rsi_rsp 14
+	jmp *sha256_transform_4way_core_addr(%rip)
+	
+	.p2align 6
+sha256_transform_4way_finish:
+	movdqu	0(%rdi), %xmm2
+	movdqu	16(%rdi), %xmm6
+	movdqu	32(%rdi), %xmm11
+	movdqu	48(%rdi), %xmm1
+	paddd	%xmm2, %xmm7
+	paddd	%xmm6, %xmm5
+	paddd	%xmm11, %xmm4
+	paddd	%xmm1, %xmm3
+	movdqu	64(%rdi), %xmm2
+	movdqu	80(%rdi), %xmm6
+	movdqu	96(%rdi), %xmm11
+	movdqu	112(%rdi), %xmm1
+	paddd	%xmm2, %xmm0
+	paddd	%xmm6, %xmm8
+	paddd	%xmm11, %xmm9
+	paddd	%xmm1, %xmm10
+	
+	movdqu	%xmm7, 0(%rdi)
+	movdqu	%xmm5, 16(%rdi)
+	movdqu	%xmm4, 32(%rdi)
+	movdqu	%xmm3, 48(%rdi)
+	movdqu	%xmm0, 64(%rdi)
+	movdqu	%xmm8, 80(%rdi)
+	movdqu	%xmm9, 96(%rdi)
+	movdqu	%xmm10, 112(%rdi)
+	
+	movq	%r8, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	movdqa	0(%rsp), %xmm6
+	movdqa	16(%rsp), %xmm7
+	movdqa	32(%rsp), %xmm8
+	movdqa	48(%rsp), %xmm9
+	movdqa	64(%rsp), %xmm10
+	movdqa	80(%rsp), %xmm11
+	addq	$96, %rsp
+	popq	%rdi
+#endif
+	ret
+
+
+#ifdef USE_AVX2
+
+	.text
+	.p2align 6
+sha256_transform_8way_core_avx2:
+	leaq	8*64(%rsp), %rax
+	vmovdqa	-2*32(%rax), %ymm3
+	vmovdqa	-1*32(%rax), %ymm7
+	sha256_avx2_extend_doubleround 0
+	sha256_avx2_extend_doubleround 2
+	sha256_avx2_extend_doubleround 4
+	sha256_avx2_extend_doubleround 6
+	sha256_avx2_extend_doubleround 8
+	sha256_avx2_extend_doubleround 10
+	sha256_avx2_extend_doubleround 12
+	sha256_avx2_extend_doubleround 14
+	sha256_avx2_extend_doubleround 16
+	sha256_avx2_extend_doubleround 18
+	sha256_avx2_extend_doubleround 20
+	sha256_avx2_extend_doubleround 22
+	sha256_avx2_extend_doubleround 24
+	sha256_avx2_extend_doubleround 26
+	sha256_avx2_extend_doubleround 28
+	sha256_avx2_extend_doubleround 30
+	sha256_avx2_extend_doubleround 32
+	sha256_avx2_extend_doubleround 34
+	sha256_avx2_extend_doubleround 36
+	sha256_avx2_extend_doubleround 38
+	sha256_avx2_extend_doubleround 40
+	sha256_avx2_extend_doubleround 42
+	sha256_avx2_extend_doubleround 44
+	sha256_avx2_extend_doubleround 46
+	vmovdqu	0*32(%rdi), %ymm7
+	vmovdqu	1*32(%rdi), %ymm5
+	vmovdqu	2*32(%rdi), %ymm4
+	vmovdqu	3*32(%rdi), %ymm3
+	vmovdqu	4*32(%rdi), %ymm0
+	vmovdqu	5*32(%rdi), %ymm8
+	vmovdqu	6*32(%rdi), %ymm9
+	vmovdqu	7*32(%rdi), %ymm10
+	movq	%rsp, %rax
+	leaq	sha256_8k(%rip), %rcx
+	sha256_avx2_main_quadround 0
+	sha256_avx2_main_quadround 4
+	sha256_avx2_main_quadround 8
+	sha256_avx2_main_quadround 12
+	sha256_avx2_main_quadround 16
+	sha256_avx2_main_quadround 20
+	sha256_avx2_main_quadround 24
+	sha256_avx2_main_quadround 28
+	sha256_avx2_main_quadround 32
+	sha256_avx2_main_quadround 36
+	sha256_avx2_main_quadround 40
+	sha256_avx2_main_quadround 44
+	sha256_avx2_main_quadround 48
+	sha256_avx2_main_quadround 52
+	sha256_avx2_main_quadround 56
+	sha256_avx2_main_quadround 60
+	jmp sha256_transform_8way_finish
+
+.macro p2bswap_avx2_rsi_rsp i
+	vmovdqu	\i*32(%rsi), %ymm0
+	vmovdqu	(\i+1)*32(%rsi), %ymm2
+	vpshuflw	$0xb1, %ymm0, %ymm0
+	vpshuflw	$0xb1, %ymm2, %ymm2
+	vpshufhw	$0xb1, %ymm0, %ymm0
+	vpshufhw	$0xb1, %ymm2, %ymm2
+	vpsrlw	$8, %ymm0, %ymm1
+	vpsrlw	$8, %ymm2, %ymm3
+	vpsllw	$8, %ymm0, %ymm0
+	vpsllw	$8, %ymm2, %ymm2
+	vpxor	%ymm1, %ymm0, %ymm0
+	vpxor	%ymm3, %ymm2, %ymm2
+	vmovdqa	%ymm0, \i*32(%rsp)
+	vmovdqa	%ymm2, (\i+1)*32(%rsp)
+.endm
+	
+	.text
+	.p2align 6
+	.globl sha256_transform_8way
+	.globl _sha256_transform_8way
+sha256_transform_8way:
+_sha256_transform_8way:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	subq	$96, %rsp
+	vmovdqa	%xmm6, 0(%rsp)
+	vmovdqa	%xmm7, 16(%rsp)
+	vmovdqa	%xmm8, 32(%rsp)
+	vmovdqa	%xmm9, 48(%rsp)
+	vmovdqa	%xmm10, 64(%rsp)
+	vmovdqa	%xmm11, 80(%rsp)
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+#endif
+	movq	%rsp, %r8
+	subq	$64*32, %rsp
+	andq	$-128, %rsp
+	
+	testq	%rdx, %rdx
+	jnz sha256_transform_8way_swap
+	
+	vmovdqu	0*32(%rsi), %ymm0
+	vmovdqu	1*32(%rsi), %ymm1
+	vmovdqu	2*32(%rsi), %ymm2
+	vmovdqu	3*32(%rsi), %ymm3
+	vmovdqu	4*32(%rsi), %ymm4
+	vmovdqu	5*32(%rsi), %ymm5
+	vmovdqu	6*32(%rsi), %ymm6
+	vmovdqu	7*32(%rsi), %ymm7
+	vmovdqa	%ymm0, 0*32(%rsp)
+	vmovdqa	%ymm1, 1*32(%rsp)
+	vmovdqa	%ymm2, 2*32(%rsp)
+	vmovdqa	%ymm3, 3*32(%rsp)
+	vmovdqa	%ymm4, 4*32(%rsp)
+	vmovdqa	%ymm5, 5*32(%rsp)
+	vmovdqa	%ymm6, 6*32(%rsp)
+	vmovdqa	%ymm7, 7*32(%rsp)
+	vmovdqu	8*32(%rsi), %ymm0
+	vmovdqu	9*32(%rsi), %ymm1
+	vmovdqu	10*32(%rsi), %ymm2
+	vmovdqu	11*32(%rsi), %ymm3
+	vmovdqu	12*32(%rsi), %ymm4
+	vmovdqu	13*32(%rsi), %ymm5
+	vmovdqu	14*32(%rsi), %ymm6
+	vmovdqu	15*32(%rsi), %ymm7
+	vmovdqa	%ymm0, 8*32(%rsp)
+	vmovdqa	%ymm1, 9*32(%rsp)
+	vmovdqa	%ymm2, 10*32(%rsp)
+	vmovdqa	%ymm3, 11*32(%rsp)
+	vmovdqa	%ymm4, 12*32(%rsp)
+	vmovdqa	%ymm5, 13*32(%rsp)
+	vmovdqa	%ymm6, 14*32(%rsp)
+	vmovdqa	%ymm7, 15*32(%rsp)
+	jmp sha256_transform_8way_core_avx2
+	
+	.p2align 6
+sha256_transform_8way_swap:
+	p2bswap_avx2_rsi_rsp 0
+	p2bswap_avx2_rsi_rsp 2
+	p2bswap_avx2_rsi_rsp 4
+	p2bswap_avx2_rsi_rsp 6
+	p2bswap_avx2_rsi_rsp 8
+	p2bswap_avx2_rsi_rsp 10
+	p2bswap_avx2_rsi_rsp 12
+	p2bswap_avx2_rsi_rsp 14
+	jmp sha256_transform_8way_core_avx2
+	
+	.p2align 6
+sha256_transform_8way_finish:
+	vmovdqu	0*32(%rdi), %ymm2
+	vmovdqu	1*32(%rdi), %ymm6
+	vmovdqu	2*32(%rdi), %ymm11
+	vmovdqu	3*32(%rdi), %ymm1
+	vpaddd	%ymm2, %ymm7, %ymm7
+	vpaddd	%ymm6, %ymm5, %ymm5
+	vpaddd	%ymm11, %ymm4, %ymm4
+	vpaddd	%ymm1, %ymm3, %ymm3
+	vmovdqu	4*32(%rdi), %ymm2
+	vmovdqu	5*32(%rdi), %ymm6
+	vmovdqu	6*32(%rdi), %ymm11
+	vmovdqu	7*32(%rdi), %ymm1
+	vpaddd	%ymm2, %ymm0, %ymm0
+	vpaddd	%ymm6, %ymm8, %ymm8
+	vpaddd	%ymm11, %ymm9, %ymm9
+	vpaddd	%ymm1, %ymm10, %ymm10
+	
+	vmovdqu	%ymm7, 0*32(%rdi)
+	vmovdqu	%ymm5, 1*32(%rdi)
+	vmovdqu	%ymm4, 2*32(%rdi)
+	vmovdqu	%ymm3, 3*32(%rdi)
+	vmovdqu	%ymm0, 4*32(%rdi)
+	vmovdqu	%ymm8, 5*32(%rdi)
+	vmovdqu	%ymm9, 6*32(%rdi)
+	vmovdqu	%ymm10, 7*32(%rdi)
+	
+	movq	%r8, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	vmovdqa	0(%rsp), %xmm6
+	vmovdqa	16(%rsp), %xmm7
+	vmovdqa	32(%rsp), %xmm8
+	vmovdqa	48(%rsp), %xmm9
+	vmovdqa	64(%rsp), %xmm10
+	vmovdqa	80(%rsp), %xmm11
+	addq	$96, %rsp
+	popq	%rdi
+#endif
+	ret
+
+#endif /* USE_AVX2 */
+	
+	
+	.data
+	.p2align 3
+sha256d_ms_4way_addr:
+	.quad 0x0
+	
+	.text
+	.p2align 6
+	.globl sha256d_ms_4way
+	.globl _sha256d_ms_4way
+sha256d_ms_4way:
+_sha256d_ms_4way:
+	jmp *sha256d_ms_4way_addr(%rip)
+	
+	
+	.p2align 6
+sha256d_ms_4way_sse2:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	subq	$32, %rsp
+	movdqa	%xmm6, 0(%rsp)
+	movdqa	%xmm7, 16(%rsp)
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+	movq	%r9, %rcx
+#endif
+	subq	$8+67*16, %rsp
+	
+	leaq	256(%rsi), %rax
+	
+sha256d_ms_4way_sse2_extend_loop1:
+	movdqa	3*16(%rsi), %xmm0
+	movdqa	2*16(%rax), %xmm3
+	movdqa	3*16(%rax), %xmm7
+	movdqa	%xmm3, 5*16(%rsp)
+	movdqa	%xmm7, 6*16(%rsp)
+	movdqa	%xmm0, %xmm2
+	paddd	%xmm0, %xmm7
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, 2*16(%rax)
+	movdqa	%xmm7, 3*16(%rax)
+	
+	movdqa	4*16(%rax), %xmm0
+	movdqa	%xmm0, 7*16(%rsp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, 4*16(%rax)
+	movdqa	%xmm7, 5*16(%rax)
+	
+	movdqa	6*16(%rax), %xmm0
+	movdqa	7*16(%rax), %xmm4
+	movdqa	%xmm0, 9*16(%rsp)
+	movdqa	%xmm4, 10*16(%rsp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 6*16(%rax)
+	movdqa	%xmm7, 7*16(%rax)
+	
+	movdqa	8*16(%rax), %xmm0
+	movdqa	2*16(%rax), %xmm4
+	movdqa	%xmm0, 11*16(%rsp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 8*16(%rax)
+	movdqa	%xmm7, 9*16(%rax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	3*16(%rax), %xmm3
+	paddd	4*16(%rax), %xmm7
+	movdqa	%xmm3, 10*16(%rax)
+	movdqa	%xmm7, 11*16(%rax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	5*16(%rax), %xmm3
+	paddd	6*16(%rax), %xmm7
+	movdqa	%xmm3, 12*16(%rax)
+	movdqa	%xmm7, 13*16(%rax)
+	
+	movdqa	14*16(%rax), %xmm0
+	movdqa	15*16(%rax), %xmm4
+	movdqa	%xmm0, 17*16(%rsp)
+	movdqa	%xmm4, 18*16(%rsp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	7*16(%rax), %xmm0
+	paddd	8*16(%rax), %xmm4
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 14*16(%rax)
+	movdqa	%xmm7, 15*16(%rax)
+	
+sha256d_ms_4way_sse2_extend_loop2:
+	sha256_sse2_extend_doubleround 16
+	sha256_sse2_extend_doubleround 18
+	sha256_sse2_extend_doubleround 20
+	sha256_sse2_extend_doubleround 22
+	sha256_sse2_extend_doubleround 24
+	sha256_sse2_extend_doubleround 26
+	sha256_sse2_extend_doubleround 28
+	sha256_sse2_extend_doubleround 30
+	sha256_sse2_extend_doubleround 32
+	sha256_sse2_extend_doubleround 34
+	sha256_sse2_extend_doubleround 36
+	sha256_sse2_extend_doubleround 38
+	sha256_sse2_extend_doubleround 40
+	sha256_sse2_extend_doubleround 42
+	jz sha256d_ms_4way_sse2_extend_coda2
+	sha256_sse2_extend_doubleround 44
+	sha256_sse2_extend_doubleround 46
+	
+	movdqa	0(%rcx), %xmm3
+	movdqa	16(%rcx), %xmm0
+	movdqa	32(%rcx), %xmm1
+	movdqa	48(%rcx), %xmm2
+	movdqa	64(%rcx), %xmm6
+	movdqa	80(%rcx), %xmm7
+	movdqa	96(%rcx), %xmm5
+	movdqa	112(%rcx), %xmm4
+	movdqa	%xmm1, 0(%rsp)
+	movdqa	%xmm2, 16(%rsp)
+	movdqa	%xmm6, 32(%rsp)
+	
+	movq	%rsi, %rax
+	leaq	sha256_4k(%rip), %rcx
+	jmp sha256d_ms_4way_sse2_main_loop1
+	
+sha256d_ms_4way_sse2_main_loop2:
+	sha256_sse2_main_round 0
+	sha256_sse2_main_round 1
+	sha256_sse2_main_round 2
+sha256d_ms_4way_sse2_main_loop1:
+	sha256_sse2_main_round 3
+	sha256_sse2_main_quadround 4
+	sha256_sse2_main_quadround 8
+	sha256_sse2_main_quadround 12
+	sha256_sse2_main_quadround 16
+	sha256_sse2_main_quadround 20
+	sha256_sse2_main_quadround 24
+	sha256_sse2_main_quadround 28
+	sha256_sse2_main_quadround 32
+	sha256_sse2_main_quadround 36
+	sha256_sse2_main_quadround 40
+	sha256_sse2_main_quadround 44
+	sha256_sse2_main_quadround 48
+	sha256_sse2_main_quadround 52
+	sha256_sse2_main_round 56
+	jz sha256d_ms_4way_sse2_finish
+	sha256_sse2_main_round 57
+	sha256_sse2_main_round 58
+	sha256_sse2_main_round 59
+	sha256_sse2_main_quadround 60
+	
+	movdqa	5*16(%rsp), %xmm1
+	movdqa	6*16(%rsp), %xmm2
+	movdqa	7*16(%rsp), %xmm6
+	movdqa	%xmm1, 18*16(%rsi)
+	movdqa	%xmm2, 19*16(%rsi)
+	movdqa	%xmm6, 20*16(%rsi)
+	movdqa	9*16(%rsp), %xmm1
+	movdqa	10*16(%rsp), %xmm2
+	movdqa	11*16(%rsp), %xmm6
+	movdqa	%xmm1, 22*16(%rsi)
+	movdqa	%xmm2, 23*16(%rsi)
+	movdqa	%xmm6, 24*16(%rsi)
+	movdqa	17*16(%rsp), %xmm1
+	movdqa	18*16(%rsp), %xmm2
+	movdqa	%xmm1, 30*16(%rsi)
+	movdqa	%xmm2, 31*16(%rsi)
+	
+	movdqa	0(%rsp), %xmm1
+	movdqa	16(%rsp), %xmm2
+	movdqa	32(%rsp), %xmm6
+	paddd	0(%rdx), %xmm7
+	paddd	16(%rdx), %xmm5
+	paddd	32(%rdx), %xmm4
+	paddd	48(%rdx), %xmm3
+	paddd	64(%rdx), %xmm0
+	paddd	80(%rdx), %xmm1
+	paddd	96(%rdx), %xmm2
+	paddd	112(%rdx), %xmm6
+	
+	movdqa	%xmm7, 48+0(%rsp)
+	movdqa	%xmm5, 48+16(%rsp)
+	movdqa	%xmm4, 48+32(%rsp)
+	movdqa	%xmm3, 48+48(%rsp)
+	movdqa	%xmm0, 48+64(%rsp)
+	movdqa	%xmm1, 48+80(%rsp)
+	movdqa	%xmm2, 48+96(%rsp)
+	movdqa	%xmm6, 48+112(%rsp)
+	
+	pxor	%xmm0, %xmm0
+	movq	$0x8000000000000100, %rax
+	movd	%rax, %xmm1
+	pshufd	$0x55, %xmm1, %xmm2
+	pshufd	$0x00, %xmm1, %xmm1
+	movdqa	%xmm2, 48+128(%rsp)
+	movdqa	%xmm0, 48+144(%rsp)
+	movdqa	%xmm0, 48+160(%rsp)
+	movdqa	%xmm0, 48+176(%rsp)
+	movdqa	%xmm0, 48+192(%rsp)
+	movdqa	%xmm0, 48+208(%rsp)
+	movdqa	%xmm0, 48+224(%rsp)
+	movdqa	%xmm1, 48+240(%rsp)
+	
+	leaq	19*16(%rsp), %rax
+	cmpq	%rax, %rax
+	
+	movdqa	-15*16(%rax), %xmm0
+	movdqa	-14*16(%rax), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	paddd	-16*16(%rax), %xmm0
+	paddd	-15*16(%rax), %xmm4
+	paddd	sha256d_4preext2_17(%rip), %xmm4
+	movdqa	%xmm0, %xmm3
+	movdqa	%xmm4, %xmm7
+	movdqa	%xmm3, 0*16(%rax)
+	movdqa	%xmm7, 1*16(%rax)
+	
+	sha256_sse2_extend_doubleround 2
+	sha256_sse2_extend_doubleround 4
+	
+	movdqa	-9*16(%rax), %xmm0
+	movdqa	sha256d_4preext2_23(%rip), %xmm4
+	movdqa	%xmm0, %xmm2
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	-10*16(%rax), %xmm0
+	paddd	-9*16(%rax), %xmm4
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	-1*16(%rax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	paddd	0*16(%rax), %xmm4
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 6*16(%rax)
+	movdqa	%xmm7, 7*16(%rax)
+	
+	movdqa	sha256d_4preext2_24(%rip), %xmm0
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	1*16(%rax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	2*16(%rax), %xmm7
+	movdqa	%xmm3, 8*16(%rax)
+	movdqa	%xmm7, 9*16(%rax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	3*16(%rax), %xmm3
+	paddd	4*16(%rax), %xmm7
+	movdqa	%xmm3, 10*16(%rax)
+	movdqa	%xmm7, 11*16(%rax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	5*16(%rax), %xmm3
+	paddd	6*16(%rax), %xmm7
+	movdqa	%xmm3, 12*16(%rax)
+	movdqa	%xmm7, 13*16(%rax)
+	
+	movdqa	sha256d_4preext2_30(%rip), %xmm0
+	movdqa	0*16(%rax), %xmm4
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm6
+	psrld	$4, %xmm5
+	pxor	%xmm5, %xmm4
+	pxor	%xmm6, %xmm4
+	psrld	$11, %xmm5
+	pslld	$11, %xmm6
+	pxor	%xmm5, %xmm4
+	pxor	%xmm6, %xmm4
+	paddd	-1*16(%rax), %xmm4
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	7*16(%rax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	paddd	8*16(%rax), %xmm4
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 14*16(%rax)
+	movdqa	%xmm7, 15*16(%rax)
+	
+	jmp sha256d_ms_4way_sse2_extend_loop2
+	
+sha256d_ms_4way_sse2_extend_coda2:
+	sha256_sse2_extend_round 44
+	
+	movdqa	sha256_4h+0(%rip), %xmm7
+	movdqa	sha256_4h+16(%rip), %xmm5
+	movdqa	sha256_4h+32(%rip), %xmm4
+	movdqa	sha256_4h+48(%rip), %xmm3
+	movdqa	sha256_4h+64(%rip), %xmm0
+	movdqa	sha256_4h+80(%rip), %xmm1
+	movdqa	sha256_4h+96(%rip), %xmm2
+	movdqa	sha256_4h+112(%rip), %xmm6
+	movdqa	%xmm1, 0(%rsp)
+	movdqa	%xmm2, 16(%rsp)
+	movdqa	%xmm6, 32(%rsp)
+	
+	leaq	48(%rsp), %rax
+	leaq	sha256_4k(%rip), %rcx
+	jmp sha256d_ms_4way_sse2_main_loop2
+
+.macro sha256_sse2_main_round_red i, r7
+	movdqa	16*\i(%rax), %xmm6
+	paddd	16*\i(%rcx), %xmm6
+	paddd	32(%rsp), %xmm6
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%rsp), %xmm2
+	paddd	\r7, %xmm6
+	pandn	%xmm2, %xmm1
+	movdqa	%xmm2, 32(%rsp)
+	movdqa	0(%rsp), %xmm2
+	movdqa	%xmm2, 16(%rsp)
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%rsp)
+	paddd	%xmm1, %xmm6
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm1, %xmm0
+	paddd	%xmm6, %xmm0
+.endm
+
+sha256d_ms_4way_sse2_finish:
+	sha256_sse2_main_round_red 57, %xmm3
+	sha256_sse2_main_round_red 58, %xmm4
+	sha256_sse2_main_round_red 59, %xmm5
+	sha256_sse2_main_round_red 60, %xmm7
+	
+	paddd	sha256_4h+112(%rip), %xmm0
+	movdqa	%xmm0, 112(%rdi)
+	
+	addq	$8+67*16, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	movdqa	0(%rsp), %xmm6
+	movdqa	16(%rsp), %xmm7
+	addq	$32, %rsp
+	popq	%rdi
+#endif
+	ret
+	
+	
+#if defined(USE_AVX)
+	
+	.p2align 6
+sha256d_ms_4way_avx:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	subq	$80, %rsp
+	movdqa	%xmm6, 0(%rsp)
+	movdqa	%xmm7, 16(%rsp)
+	movdqa	%xmm8, 32(%rsp)
+	movdqa	%xmm9, 48(%rsp)
+	movdqa	%xmm10, 64(%rsp)
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+	movq	%r9, %rcx
+#endif
+	subq	$1032, %rsp
+	
+	leaq	256(%rsi), %rax
+	
+sha256d_ms_4way_avx_extend_loop1:
+	vmovdqa	3*16(%rsi), %xmm0
+	vmovdqa	2*16(%rax), %xmm3
+	vmovdqa	3*16(%rax), %xmm7
+	vmovdqa	%xmm3, 2*16(%rsp)
+	vmovdqa	%xmm7, 3*16(%rsp)
+	vpaddd	%xmm0, %xmm7, %xmm7
+	vpslld	$14, %xmm0, %xmm2
+	vpsrld	$3, %xmm0, %xmm0
+	vpsrld	$4, %xmm0, %xmm1
+	vpxor	%xmm1, %xmm0, %xmm0
+	vpxor	%xmm2, %xmm0, %xmm0
+	vpsrld	$11, %xmm1, %xmm1
+	vpslld	$11, %xmm2, %xmm2
+	vpxor	%xmm1, %xmm0, %xmm0
+	vpxor	%xmm2, %xmm0, %xmm0
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vmovdqa	%xmm3, 2*16(%rax)
+	vmovdqa	%xmm7, 3*16(%rax)
+	
+	vmovdqa	4*16(%rax), %xmm0
+	vmovdqa	%xmm0, 4*16(%rsp)
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vmovdqa	%xmm3, 4*16(%rax)
+	vmovdqa	%xmm7, 5*16(%rax)
+	
+	vmovdqa	6*16(%rax), %xmm0
+	vmovdqa	7*16(%rax), %xmm4
+	vmovdqa	%xmm0, 6*16(%rsp)
+	vmovdqa	%xmm4, 7*16(%rsp)
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 6*16(%rax)
+	vmovdqa	%xmm7, 7*16(%rax)
+	
+	vmovdqa	8*16(%rax), %xmm0
+	vmovdqa	2*16(%rax), %xmm4
+	vmovdqa	%xmm0, 8*16(%rsp)
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 8*16(%rax)
+	vmovdqa	%xmm7, 9*16(%rax)
+	
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	3*16(%rax), %xmm3, %xmm3
+	vpaddd	4*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 10*16(%rax)
+	vmovdqa	%xmm7, 11*16(%rax)
+	
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	5*16(%rax), %xmm3, %xmm3
+	vpaddd	6*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 12*16(%rax)
+	vmovdqa	%xmm7, 13*16(%rax)
+	
+	vmovdqa	14*16(%rax), %xmm0
+	vmovdqa	15*16(%rax), %xmm4
+	vmovdqa	%xmm0, 14*16(%rsp)
+	vmovdqa	%xmm4, 15*16(%rsp)
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpaddd	7*16(%rax), %xmm0, %xmm0
+	vpaddd	8*16(%rax), %xmm4, %xmm4
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 14*16(%rax)
+	vmovdqa	%xmm7, 15*16(%rax)
+	
+sha256d_ms_4way_avx_extend_loop2:
+	sha256_avx_extend_doubleround 16
+	sha256_avx_extend_doubleround 18
+	sha256_avx_extend_doubleround 20
+	sha256_avx_extend_doubleround 22
+	sha256_avx_extend_doubleround 24
+	sha256_avx_extend_doubleround 26
+	sha256_avx_extend_doubleround 28
+	sha256_avx_extend_doubleround 30
+	sha256_avx_extend_doubleround 32
+	sha256_avx_extend_doubleround 34
+	sha256_avx_extend_doubleround 36
+	sha256_avx_extend_doubleround 38
+	sha256_avx_extend_doubleround 40
+	sha256_avx_extend_doubleround 42
+	jz sha256d_ms_4way_avx_extend_coda2
+	sha256_avx_extend_doubleround 44
+	sha256_avx_extend_doubleround 46
+	
+	movdqa	0(%rcx), %xmm7
+	movdqa	16(%rcx), %xmm8
+	movdqa	32(%rcx), %xmm9
+	movdqa	48(%rcx), %xmm10
+	movdqa	64(%rcx), %xmm0
+	movdqa	80(%rcx), %xmm5
+	movdqa	96(%rcx), %xmm4
+	movdqa	112(%rcx), %xmm3
+	
+	movq	%rsi, %rax
+	leaq	sha256_4k(%rip), %rcx
+	jmp sha256d_ms_4way_avx_main_loop1
+	
+sha256d_ms_4way_avx_main_loop2:
+	sha256_avx_main_round 0, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3, %xmm4, %xmm5, %xmm7
+	sha256_avx_main_round 1, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4, %xmm5, %xmm7, %xmm3
+	sha256_avx_main_round 2, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5, %xmm7, %xmm3, %xmm4
+sha256d_ms_4way_avx_main_loop1:
+	sha256_avx_main_round 3, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7, %xmm3, %xmm4, %xmm5
+	sha256_avx_main_quadround 4
+	sha256_avx_main_quadround 8
+	sha256_avx_main_quadround 12
+	sha256_avx_main_quadround 16
+	sha256_avx_main_quadround 20
+	sha256_avx_main_quadround 24
+	sha256_avx_main_quadround 28
+	sha256_avx_main_quadround 32
+	sha256_avx_main_quadround 36
+	sha256_avx_main_quadround 40
+	sha256_avx_main_quadround 44
+	sha256_avx_main_quadround 48
+	sha256_avx_main_quadround 52
+	sha256_avx_main_round 56, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3, %xmm4, %xmm5, %xmm7
+	jz sha256d_ms_4way_avx_finish
+	sha256_avx_main_round 57, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4, %xmm5, %xmm7, %xmm3
+	sha256_avx_main_round 58, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5, %xmm7, %xmm3, %xmm4
+	sha256_avx_main_round 59, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7, %xmm3, %xmm4, %xmm5
+	sha256_avx_main_quadround 60
+	
+	movdqa	2*16(%rsp), %xmm1
+	movdqa	3*16(%rsp), %xmm2
+	movdqa	4*16(%rsp), %xmm6
+	movdqa	%xmm1, 18*16(%rsi)
+	movdqa	%xmm2, 19*16(%rsi)
+	movdqa	%xmm6, 20*16(%rsi)
+	movdqa	6*16(%rsp), %xmm1
+	movdqa	7*16(%rsp), %xmm2
+	movdqa	8*16(%rsp), %xmm6
+	movdqa	%xmm1, 22*16(%rsi)
+	movdqa	%xmm2, 23*16(%rsi)
+	movdqa	%xmm6, 24*16(%rsi)
+	movdqa	14*16(%rsp), %xmm1
+	movdqa	15*16(%rsp), %xmm2
+	movdqa	%xmm1, 30*16(%rsi)
+	movdqa	%xmm2, 31*16(%rsi)
+	
+	paddd	0(%rdx), %xmm7
+	paddd	16(%rdx), %xmm5
+	paddd	32(%rdx), %xmm4
+	paddd	48(%rdx), %xmm3
+	paddd	64(%rdx), %xmm0
+	paddd	80(%rdx), %xmm8
+	paddd	96(%rdx), %xmm9
+	paddd	112(%rdx), %xmm10
+	
+	movdqa	%xmm7, 0(%rsp)
+	movdqa	%xmm5, 16(%rsp)
+	movdqa	%xmm4, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm8, 80(%rsp)
+	movdqa	%xmm9, 96(%rsp)
+	movdqa	%xmm10, 112(%rsp)
+	
+	pxor	%xmm0, %xmm0
+	movq	$0x8000000000000100, %rax
+	movd	%rax, %xmm1
+	pshufd	$0x55, %xmm1, %xmm2
+	pshufd	$0x00, %xmm1, %xmm1
+	movdqa	%xmm2, 128(%rsp)
+	movdqa	%xmm0, 144(%rsp)
+	movdqa	%xmm0, 160(%rsp)
+	movdqa	%xmm0, 176(%rsp)
+	movdqa	%xmm0, 192(%rsp)
+	movdqa	%xmm0, 208(%rsp)
+	movdqa	%xmm0, 224(%rsp)
+	movdqa	%xmm1, 240(%rsp)
+	
+	leaq	256(%rsp), %rax
+	cmpq	%rax, %rax
+	
+	vmovdqa	-15*16(%rax), %xmm0
+	vmovdqa	-14*16(%rax), %xmm4
+	vpslld	$14, %xmm0, %xmm2
+	vpslld	$14, %xmm4, %xmm6
+	vpsrld	$3, %xmm0, %xmm8
+	vpsrld	$3, %xmm4, %xmm4
+	vpsrld	$7, %xmm0, %xmm1
+	vpsrld	$4, %xmm4, %xmm5
+	vpxor	%xmm1, %xmm8, %xmm8
+	vpxor	%xmm5, %xmm4, %xmm4
+	vpsrld	$11, %xmm1, %xmm1
+	vpsrld	$11, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpslld	$11, %xmm2, %xmm2
+	vpslld	$11, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm8, %xmm8
+	vpxor	%xmm5, %xmm4, %xmm4
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpaddd	%xmm0, %xmm4, %xmm4
+	vpaddd	-16*16(%rax), %xmm8, %xmm3
+	vpaddd	sha256d_4preext2_17(%rip), %xmm4, %xmm7
+	vmovdqa	%xmm3, 0*16(%rax)
+	vmovdqa	%xmm7, 1*16(%rax)
+	
+	sha256_avx_extend_doubleround 2
+	sha256_avx_extend_doubleround 4
+	
+	vmovdqa	-9*16(%rax), %xmm0
+	vpslld	$14, %xmm0, %xmm2
+	vpsrld	$3, %xmm0, %xmm8
+	vpsrld	$7, %xmm0, %xmm1
+	vpxor	%xmm1, %xmm8, %xmm8
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpsrld	$11, %xmm1, %xmm1
+	vpslld	$11, %xmm2, %xmm2
+	vpxor	%xmm1, %xmm8, %xmm8
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpaddd	sha256d_4preext2_23(%rip), %xmm0, %xmm4
+	vpaddd	-10*16(%rax), %xmm8, %xmm0
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpaddd	-1*16(%rax), %xmm0, %xmm0
+	vpaddd	0*16(%rax), %xmm4, %xmm4
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 6*16(%rax)
+	vmovdqa	%xmm7, 7*16(%rax)
+	
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	sha256d_4preext2_24(%rip), %xmm3, %xmm3
+	vpaddd	1*16(%rax), %xmm3, %xmm3
+	vpaddd	2*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 8*16(%rax)
+	vmovdqa	%xmm7, 9*16(%rax)
+	
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	3*16(%rax), %xmm3, %xmm3
+	vpaddd	4*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 10*16(%rax)
+	vmovdqa	%xmm7, 11*16(%rax)
+	
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	5*16(%rax), %xmm3, %xmm3
+	vpaddd	6*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 12*16(%rax)
+	vmovdqa	%xmm7, 13*16(%rax)
+	
+	vmovdqa	sha256d_4preext2_30(%rip), %xmm0
+	vmovdqa	0*16(%rax), %xmm4
+	vpslld	$14, %xmm4, %xmm6
+	vpsrld	$3, %xmm4, %xmm4
+	vpsrld	$4, %xmm4, %xmm5
+	vpxor	%xmm5, %xmm4, %xmm4
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpsrld	$11, %xmm5, %xmm5
+	vpslld	$11, %xmm6, %xmm6
+	vpxor	%xmm5, %xmm4, %xmm4
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpaddd	-1*16(%rax), %xmm4, %xmm4
+	vpslld	$13, %xmm3, %xmm2
+	vpslld	$13, %xmm7, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpaddd	7*16(%rax), %xmm0, %xmm0
+	vpaddd	8*16(%rax), %xmm4, %xmm4
+	vpsrld	$7, %xmm3, %xmm1
+	vpsrld	$7, %xmm7, %xmm5
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpsrld	$2, %xmm1, %xmm1
+	vpsrld	$2, %xmm5, %xmm5
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpslld	$2, %xmm2, %xmm2
+	vpslld	$2, %xmm6, %xmm6
+	vpxor	%xmm1, %xmm3, %xmm3
+	vpxor	%xmm5, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 14*16(%rax)
+	vmovdqa	%xmm7, 15*16(%rax)
+	
+	jmp sha256d_ms_4way_avx_extend_loop2
+	
+sha256d_ms_4way_avx_extend_coda2:
+	sha256_avx_extend_round 44
+	
+	movdqa	sha256_4h+0(%rip), %xmm7
+	movdqa	sha256_4h+16(%rip), %xmm5
+	movdqa	sha256_4h+32(%rip), %xmm4
+	movdqa	sha256_4h+48(%rip), %xmm3
+	movdqa	sha256_4h+64(%rip), %xmm0
+	movdqa	sha256_4h+80(%rip), %xmm8
+	movdqa	sha256_4h+96(%rip), %xmm9
+	movdqa	sha256_4h+112(%rip), %xmm10
+	
+	movq	%rsp, %rax
+	leaq	sha256_4k(%rip), %rcx
+	jmp sha256d_ms_4way_avx_main_loop2
+
+.macro sha256_avx_main_round_red i, r0, r1, r2, r3, r4
+	vpaddd	16*\i(%rax), \r0, %xmm6
+	vpaddd	16*\i(%rcx), %xmm6, %xmm6
+	vpandn	\r1, \r3, %xmm1
+	vpand	\r3, \r2, %xmm2
+	vpxor	%xmm2, %xmm1, %xmm1
+	vpaddd	%xmm1, %xmm6, %xmm6
+	vpslld	$7, \r3, %xmm1
+	vpsrld	$6, \r3, \r0
+	vpsrld	$5, \r0, %xmm2
+	vpxor	%xmm1, \r0, \r0
+	vpxor	%xmm2, \r0, \r0
+	vpslld	$14, %xmm1, %xmm1
+	vpsrld	$14, %xmm2, %xmm2
+	vpxor	%xmm1, \r0, \r0
+	vpxor	%xmm2, \r0, \r0
+	vpslld	$5, %xmm1, %xmm1
+	vpxor	%xmm1, \r0, \r0
+	vpaddd	\r0, %xmm6, %xmm6
+	vpaddd	%xmm6, \r4, \r0
+.endm
+
+sha256d_ms_4way_avx_finish:
+	sha256_avx_main_round_red 57, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4
+	sha256_avx_main_round_red 58, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5
+	sha256_avx_main_round_red 59, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7
+	sha256_avx_main_round_red 60, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3
+	
+	paddd	sha256_4h+112(%rip), %xmm10
+	movdqa	%xmm10, 112(%rdi)
+	
+	addq	$1032, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	movdqa	0(%rsp), %xmm6
+	movdqa	16(%rsp), %xmm7
+	movdqa	32(%rsp), %xmm8
+	movdqa	48(%rsp), %xmm9
+	movdqa	64(%rsp), %xmm10
+	addq	$80, %rsp
+	popq	%rdi
+#endif
+	ret
+	
+#endif /* USE_AVX */
+	
+	
+#if defined(USE_XOP)
+	
+	.p2align 6
+sha256d_ms_4way_xop:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	subq	$80, %rsp
+	movdqa	%xmm6, 0(%rsp)
+	movdqa	%xmm7, 16(%rsp)
+	movdqa	%xmm8, 32(%rsp)
+	movdqa	%xmm9, 48(%rsp)
+	movdqa	%xmm10, 64(%rsp)
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+	movq	%r9, %rcx
+#endif
+	subq	$1032, %rsp
+	
+	leaq	256(%rsi), %rax
+	
+sha256d_ms_4way_xop_extend_loop1:
+	vmovdqa	3*16(%rsi), %xmm0
+	vmovdqa	2*16(%rax), %xmm3
+	vmovdqa	3*16(%rax), %xmm7
+	vmovdqa	%xmm3, 2*16(%rsp)
+	vmovdqa	%xmm7, 3*16(%rsp)
+	vpaddd	%xmm0, %xmm7, %xmm7
+	vprotd	$25, %xmm0, %xmm1
+	vprotd	$14, %xmm0, %xmm2
+	vpsrld	$3, %xmm0, %xmm0
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm2, %xmm0, %xmm0
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vmovdqa	%xmm3, 2*16(%rax)
+	vmovdqa	%xmm7, 3*16(%rax)
+	
+	vmovdqa	4*16(%rax), %xmm0
+	vmovdqa	%xmm0, 4*16(%rsp)
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vmovdqa	%xmm3, 4*16(%rax)
+	vmovdqa	%xmm7, 5*16(%rax)
+	
+	vmovdqa	6*16(%rax), %xmm0
+	vmovdqa	7*16(%rax), %xmm4
+	vmovdqa	%xmm0, 6*16(%rsp)
+	vmovdqa	%xmm4, 7*16(%rsp)
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 6*16(%rax)
+	vmovdqa	%xmm7, 7*16(%rax)
+	
+	vmovdqa	8*16(%rax), %xmm0
+	vmovdqa	2*16(%rax), %xmm4
+	vmovdqa	%xmm0, 8*16(%rsp)
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 8*16(%rax)
+	vmovdqa	%xmm7, 9*16(%rax)
+	
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	3*16(%rax), %xmm3, %xmm3
+	vpaddd	4*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 10*16(%rax)
+	vmovdqa	%xmm7, 11*16(%rax)
+	
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	5*16(%rax), %xmm3, %xmm3
+	vpaddd	6*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 12*16(%rax)
+	vmovdqa	%xmm7, 13*16(%rax)
+	
+	vmovdqa	14*16(%rax), %xmm0
+	vmovdqa	15*16(%rax), %xmm4
+	vmovdqa	%xmm0, 14*16(%rsp)
+	vmovdqa	%xmm4, 15*16(%rsp)
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpaddd	7*16(%rax), %xmm0, %xmm0
+	vpaddd	8*16(%rax), %xmm4, %xmm4
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 14*16(%rax)
+	vmovdqa	%xmm7, 15*16(%rax)
+	
+sha256d_ms_4way_xop_extend_loop2:
+	sha256_xop_extend_doubleround 16
+	sha256_xop_extend_doubleround 18
+	sha256_xop_extend_doubleround 20
+	sha256_xop_extend_doubleround 22
+	sha256_xop_extend_doubleround 24
+	sha256_xop_extend_doubleround 26
+	sha256_xop_extend_doubleround 28
+	sha256_xop_extend_doubleround 30
+	sha256_xop_extend_doubleround 32
+	sha256_xop_extend_doubleround 34
+	sha256_xop_extend_doubleround 36
+	sha256_xop_extend_doubleround 38
+	sha256_xop_extend_doubleround 40
+	sha256_xop_extend_doubleround 42
+	jz sha256d_ms_4way_xop_extend_coda2
+	sha256_xop_extend_doubleround 44
+	sha256_xop_extend_doubleround 46
+	
+	movdqa	0(%rcx), %xmm7
+	movdqa	16(%rcx), %xmm8
+	movdqa	32(%rcx), %xmm9
+	movdqa	48(%rcx), %xmm10
+	movdqa	64(%rcx), %xmm0
+	movdqa	80(%rcx), %xmm5
+	movdqa	96(%rcx), %xmm4
+	movdqa	112(%rcx), %xmm3
+	
+	movq	%rsi, %rax
+	leaq	sha256_4k(%rip), %rcx
+	jmp sha256d_ms_4way_xop_main_loop1
+	
+sha256d_ms_4way_xop_main_loop2:
+	sha256_xop_main_round 0, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3, %xmm4, %xmm5, %xmm7
+	sha256_xop_main_round 1, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4, %xmm5, %xmm7, %xmm3
+	sha256_xop_main_round 2, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5, %xmm7, %xmm3, %xmm4
+sha256d_ms_4way_xop_main_loop1:
+	sha256_xop_main_round 3, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7, %xmm3, %xmm4, %xmm5
+	sha256_xop_main_quadround 4
+	sha256_xop_main_quadround 8
+	sha256_xop_main_quadround 12
+	sha256_xop_main_quadround 16
+	sha256_xop_main_quadround 20
+	sha256_xop_main_quadround 24
+	sha256_xop_main_quadround 28
+	sha256_xop_main_quadround 32
+	sha256_xop_main_quadround 36
+	sha256_xop_main_quadround 40
+	sha256_xop_main_quadround 44
+	sha256_xop_main_quadround 48
+	sha256_xop_main_quadround 52
+	sha256_xop_main_round 56, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3, %xmm4, %xmm5, %xmm7
+	jz sha256d_ms_4way_xop_finish
+	sha256_xop_main_round 57, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4, %xmm5, %xmm7, %xmm3
+	sha256_xop_main_round 58, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5, %xmm7, %xmm3, %xmm4
+	sha256_xop_main_round 59, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7, %xmm3, %xmm4, %xmm5
+	sha256_xop_main_quadround 60
+	
+	movdqa	2*16(%rsp), %xmm1
+	movdqa	3*16(%rsp), %xmm2
+	movdqa	4*16(%rsp), %xmm6
+	movdqa	%xmm1, 18*16(%rsi)
+	movdqa	%xmm2, 19*16(%rsi)
+	movdqa	%xmm6, 20*16(%rsi)
+	movdqa	6*16(%rsp), %xmm1
+	movdqa	7*16(%rsp), %xmm2
+	movdqa	8*16(%rsp), %xmm6
+	movdqa	%xmm1, 22*16(%rsi)
+	movdqa	%xmm2, 23*16(%rsi)
+	movdqa	%xmm6, 24*16(%rsi)
+	movdqa	14*16(%rsp), %xmm1
+	movdqa	15*16(%rsp), %xmm2
+	movdqa	%xmm1, 30*16(%rsi)
+	movdqa	%xmm2, 31*16(%rsi)
+	
+	paddd	0(%rdx), %xmm7
+	paddd	16(%rdx), %xmm5
+	paddd	32(%rdx), %xmm4
+	paddd	48(%rdx), %xmm3
+	paddd	64(%rdx), %xmm0
+	paddd	80(%rdx), %xmm8
+	paddd	96(%rdx), %xmm9
+	paddd	112(%rdx), %xmm10
+	
+	movdqa	%xmm7, 0(%rsp)
+	movdqa	%xmm5, 16(%rsp)
+	movdqa	%xmm4, 32(%rsp)
+	movdqa	%xmm3, 48(%rsp)
+	movdqa	%xmm0, 64(%rsp)
+	movdqa	%xmm8, 80(%rsp)
+	movdqa	%xmm9, 96(%rsp)
+	movdqa	%xmm10, 112(%rsp)
+	
+	pxor	%xmm0, %xmm0
+	movq	$0x8000000000000100, %rax
+	movd	%rax, %xmm1
+	pshufd	$0x55, %xmm1, %xmm2
+	pshufd	$0x00, %xmm1, %xmm1
+	movdqa	%xmm2, 128(%rsp)
+	movdqa	%xmm0, 144(%rsp)
+	movdqa	%xmm0, 160(%rsp)
+	movdqa	%xmm0, 176(%rsp)
+	movdqa	%xmm0, 192(%rsp)
+	movdqa	%xmm0, 208(%rsp)
+	movdqa	%xmm0, 224(%rsp)
+	movdqa	%xmm1, 240(%rsp)
+	
+	leaq	256(%rsp), %rax
+	cmpq	%rax, %rax
+	
+	vmovdqa	-15*16(%rax), %xmm0
+	vmovdqa	-14*16(%rax), %xmm4
+	vprotd	$25, %xmm0, %xmm1
+	vprotd	$25, %xmm4, %xmm5
+	vprotd	$14, %xmm0, %xmm2
+	vprotd	$14, %xmm4, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$3, %xmm0, %xmm8
+	vpsrld	$3, %xmm4, %xmm4
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpaddd	%xmm0, %xmm4, %xmm4
+	vpaddd	-16*16(%rax), %xmm8, %xmm3
+	vpaddd	sha256d_4preext2_17(%rip), %xmm4, %xmm7
+	vmovdqa	%xmm3, 0*16(%rax)
+	vmovdqa	%xmm7, 1*16(%rax)
+	
+	sha256_xop_extend_doubleround 2
+	sha256_xop_extend_doubleround 4
+	
+	vmovdqa	-9*16(%rax), %xmm0
+	vprotd	$25, %xmm0, %xmm1
+	vprotd	$14, %xmm0, %xmm2
+	vpsrld	$3, %xmm0, %xmm8
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm2, %xmm8, %xmm8
+	vpaddd	sha256d_4preext2_23(%rip), %xmm0, %xmm4
+	vpaddd	-10*16(%rax), %xmm8, %xmm0
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpaddd	-1*16(%rax), %xmm0, %xmm0
+	vpaddd	0*16(%rax), %xmm4, %xmm4
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 6*16(%rax)
+	vmovdqa	%xmm7, 7*16(%rax)
+	
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	sha256d_4preext2_24(%rip), %xmm3, %xmm3
+	vpaddd	1*16(%rax), %xmm3, %xmm3
+	vpaddd	2*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 8*16(%rax)
+	vmovdqa	%xmm7, 9*16(%rax)
+	
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	3*16(%rax), %xmm3, %xmm3
+	vpaddd	4*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 10*16(%rax)
+	vmovdqa	%xmm7, 11*16(%rax)
+	
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	5*16(%rax), %xmm3, %xmm3
+	vpaddd	6*16(%rax), %xmm7, %xmm7
+	vmovdqa	%xmm3, 12*16(%rax)
+	vmovdqa	%xmm7, 13*16(%rax)
+	
+	vmovdqa	sha256d_4preext2_30(%rip), %xmm0
+	vmovdqa	0*16(%rax), %xmm4
+	vprotd	$25, %xmm4, %xmm5
+	vprotd	$14, %xmm4, %xmm6
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpsrld	$3, %xmm4, %xmm4
+	vpxor	%xmm6, %xmm4, %xmm4
+	vpaddd	-1*16(%rax), %xmm4, %xmm4
+	vprotd	$15, %xmm3, %xmm1
+	vprotd	$15, %xmm7, %xmm5
+	vprotd	$13, %xmm3, %xmm2
+	vprotd	$13, %xmm7, %xmm6
+	vpxor	%xmm1, %xmm2, %xmm2
+	vpxor	%xmm5, %xmm6, %xmm6
+	vpaddd	7*16(%rax), %xmm0, %xmm0
+	vpaddd	8*16(%rax), %xmm4, %xmm4
+	vpsrld	$10, %xmm3, %xmm3
+	vpsrld	$10, %xmm7, %xmm7
+	vpxor	%xmm2, %xmm3, %xmm3
+	vpxor	%xmm6, %xmm7, %xmm7
+	vpaddd	%xmm0, %xmm3, %xmm3
+	vpaddd	%xmm4, %xmm7, %xmm7
+	vmovdqa	%xmm3, 14*16(%rax)
+	vmovdqa	%xmm7, 15*16(%rax)
+	
+	jmp sha256d_ms_4way_xop_extend_loop2
+	
+sha256d_ms_4way_xop_extend_coda2:
+	sha256_xop_extend_round 44
+	
+	movdqa	sha256_4h+0(%rip), %xmm7
+	movdqa	sha256_4h+16(%rip), %xmm5
+	movdqa	sha256_4h+32(%rip), %xmm4
+	movdqa	sha256_4h+48(%rip), %xmm3
+	movdqa	sha256_4h+64(%rip), %xmm0
+	movdqa	sha256_4h+80(%rip), %xmm8
+	movdqa	sha256_4h+96(%rip), %xmm9
+	movdqa	sha256_4h+112(%rip), %xmm10
+	
+	movq	%rsp, %rax
+	leaq	sha256_4k(%rip), %rcx
+	jmp sha256d_ms_4way_xop_main_loop2
+
+.macro sha256_xop_main_round_red i, r0, r1, r2, r3, r4
+	vpaddd	16*\i(%rax), \r0, %xmm6
+	vpaddd	16*\i(%rcx), %xmm6, %xmm6
+	vpandn	\r1, \r3, %xmm1
+	vpand	\r3, \r2, %xmm2
+	vpxor	%xmm2, %xmm1, %xmm1
+	vpaddd	%xmm1, %xmm6, %xmm6
+	vprotd	$26, \r3, %xmm1
+	vprotd	$21, \r3, %xmm2
+	vpxor	%xmm1, %xmm2, %xmm2
+	vprotd	$7, \r3, \r0
+	vpxor	%xmm2, \r0, \r0
+	vpaddd	\r0, %xmm6, %xmm6
+	vpaddd	%xmm6, \r4, \r0
+.endm
+
+sha256d_ms_4way_xop_finish:
+	sha256_xop_main_round_red 57, %xmm9, %xmm8, %xmm0, %xmm10, %xmm4
+	sha256_xop_main_round_red 58, %xmm8, %xmm0, %xmm10, %xmm9, %xmm5
+	sha256_xop_main_round_red 59, %xmm0, %xmm10, %xmm9, %xmm8, %xmm7
+	sha256_xop_main_round_red 60, %xmm10, %xmm9, %xmm8, %xmm0, %xmm3
+	
+	paddd	sha256_4h+112(%rip), %xmm10
+	movdqa	%xmm10, 112(%rdi)
+	
+	addq	$1032, %rsp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	movdqa	0(%rsp), %xmm6
+	movdqa	16(%rsp), %xmm7
+	movdqa	32(%rsp), %xmm8
+	movdqa	48(%rsp), %xmm9
+	movdqa	64(%rsp), %xmm10
+	addq	$80, %rsp
+	popq	%rdi
+#endif
+	ret
+	
+#endif /* USE_XOP */
+
+
+	.text
+	.p2align 6
+	.globl sha256_use_4way
+	.globl _sha256_use_4way
+sha256_use_4way:
+_sha256_use_4way:
+	pushq	%rbx
+	pushq	%rcx
+	pushq	%rdx
+	
+	/* Check for VIA PadLock Hash Engine */
+	movl	$0xc0000000, %eax
+	cpuid
+	cmpl	$0xc0000001, %eax
+	jb	sha256_use_4way_no_phe
+	movl	$0xc0000001, %eax
+	cpuid
+	andl	$0x00000c00, %edx
+	cmpl	$0x00000c00, %edx
+	jne	sha256_use_4way_no_phe
+	leaq	sha256_transform_phe(%rip), %rdx
+	movq	%rdx, sha256_transform_addr(%rip)
+	xorl	%eax, %eax
+	jmp	sha256_use_4way_exit
+sha256_use_4way_no_phe:
+#if defined(USE_AVX)
+	/* Check for AVX and OSXSAVE support */
+	movl	$1, %eax
+	cpuid
+	andl	$0x18000000, %ecx
+	cmpl	$0x18000000, %ecx
+	jne sha256_use_4way_base
+	/* Check for XMM and YMM state support */
+	xorl	%ecx, %ecx
+	xgetbv
+	andl	$0x00000006, %eax
+	cmpl	$0x00000006, %eax
+	jne sha256_use_4way_base
+#if defined(USE_XOP)
+	/* Check for XOP support */
+	movl	$0x80000001, %eax
+	cpuid
+	andl	$0x00000800, %ecx
+	jz sha256_use_4way_avx
+	
+sha256_use_4way_xop:
+	leaq	sha256d_ms_4way_xop(%rip), %rcx
+	leaq	sha256_transform_4way_core_xop(%rip), %rdx
+	jmp sha256_use_4way_done
+#endif /* USE_XOP */
+	
+sha256_use_4way_avx:
+	leaq	sha256d_ms_4way_avx(%rip), %rcx
+	leaq	sha256_transform_4way_core_avx(%rip), %rdx
+	jmp sha256_use_4way_done
+#endif /* USE_AVX */
+	
+sha256_use_4way_base:
+	leaq	sha256d_ms_4way_sse2(%rip), %rcx
+	leaq	sha256_transform_4way_core_sse2(%rip), %rdx
+	
+sha256_use_4way_done:
+	movq	%rcx, sha256d_ms_4way_addr(%rip)
+	movq	%rdx, sha256_transform_4way_core_addr(%rip)
+	movl	$1, %eax
+sha256_use_4way_exit:
+	popq	%rdx
+	popq	%rcx
+	popq	%rbx
+	ret
+
+
+#if defined(USE_AVX2)
+
+	.text
+	.p2align 6
+	.globl sha256d_ms_8way
+	.globl _sha256d_ms_8way
+sha256d_ms_8way:
+_sha256d_ms_8way:
+sha256d_ms_8way_avx2:
+#if defined(_WIN64) || defined(__CYGWIN__)
+	pushq	%rdi
+	subq	$80, %rsp
+	vmovdqa	%xmm6, 0(%rsp)
+	vmovdqa	%xmm7, 16(%rsp)
+	vmovdqa	%xmm8, 32(%rsp)
+	vmovdqa	%xmm9, 48(%rsp)
+	vmovdqa	%xmm10, 64(%rsp)
+	pushq	%rsi
+	movq	%rcx, %rdi
+	movq	%rdx, %rsi
+	movq	%r8, %rdx
+	movq	%r9, %rcx
+#endif
+	pushq	%rbp
+	movq	%rsp, %rbp
+	subq	$64*32, %rsp
+	andq	$-128, %rsp
+	
+	leaq	16*32(%rsi), %rax
+	
+sha256d_ms_8way_avx2_extend_loop1:
+	vmovdqa	3*32(%rsi), %ymm0
+	vmovdqa	2*32(%rax), %ymm3
+	vmovdqa	3*32(%rax), %ymm7
+	vmovdqa	%ymm3, 2*32(%rsp)
+	vmovdqa	%ymm7, 3*32(%rsp)
+	vpaddd	%ymm0, %ymm7, %ymm7
+	vpslld	$14, %ymm0, %ymm2
+	vpsrld	$3, %ymm0, %ymm0
+	vpsrld	$4, %ymm0, %ymm1
+	vpxor	%ymm1, %ymm0, %ymm0
+	vpxor	%ymm2, %ymm0, %ymm0
+	vpsrld	$11, %ymm1, %ymm1
+	vpslld	$11, %ymm2, %ymm2
+	vpxor	%ymm1, %ymm0, %ymm0
+	vpxor	%ymm2, %ymm0, %ymm0
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vmovdqa	%ymm3, 2*32(%rax)
+	vmovdqa	%ymm7, 3*32(%rax)
+	
+	vmovdqa	4*32(%rax), %ymm0
+	vmovdqa	%ymm0, 4*32(%rsp)
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vmovdqa	%ymm3, 4*32(%rax)
+	vmovdqa	%ymm7, 5*32(%rax)
+	
+	vmovdqa	6*32(%rax), %ymm0
+	vmovdqa	7*32(%rax), %ymm4
+	vmovdqa	%ymm0, 6*32(%rsp)
+	vmovdqa	%ymm4, 7*32(%rsp)
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vpaddd	%ymm4, %ymm7, %ymm7
+	vmovdqa	%ymm3, 6*32(%rax)
+	vmovdqa	%ymm7, 7*32(%rax)
+	
+	vmovdqa	8*32(%rax), %ymm0
+	vmovdqa	2*32(%rax), %ymm4
+	vmovdqa	%ymm0, 8*32(%rsp)
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vpaddd	%ymm4, %ymm7, %ymm7
+	vmovdqa	%ymm3, 8*32(%rax)
+	vmovdqa	%ymm7, 9*32(%rax)
+	
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	3*32(%rax), %ymm3, %ymm3
+	vpaddd	4*32(%rax), %ymm7, %ymm7
+	vmovdqa	%ymm3, 10*32(%rax)
+	vmovdqa	%ymm7, 11*32(%rax)
+	
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	5*32(%rax), %ymm3, %ymm3
+	vpaddd	6*32(%rax), %ymm7, %ymm7
+	vmovdqa	%ymm3, 12*32(%rax)
+	vmovdqa	%ymm7, 13*32(%rax)
+	
+	vmovdqa	14*32(%rax), %ymm0
+	vmovdqa	15*32(%rax), %ymm4
+	vmovdqa	%ymm0, 14*32(%rsp)
+	vmovdqa	%ymm4, 15*32(%rsp)
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpaddd	7*32(%rax), %ymm0, %ymm0
+	vpaddd	8*32(%rax), %ymm4, %ymm4
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vpaddd	%ymm4, %ymm7, %ymm7
+	vmovdqa	%ymm3, 14*32(%rax)
+	vmovdqa	%ymm7, 15*32(%rax)
+	
+sha256d_ms_8way_avx2_extend_loop2:
+	sha256_avx2_extend_doubleround 16
+	sha256_avx2_extend_doubleround 18
+	sha256_avx2_extend_doubleround 20
+	sha256_avx2_extend_doubleround 22
+	sha256_avx2_extend_doubleround 24
+	sha256_avx2_extend_doubleround 26
+	sha256_avx2_extend_doubleround 28
+	sha256_avx2_extend_doubleround 30
+	sha256_avx2_extend_doubleround 32
+	sha256_avx2_extend_doubleround 34
+	sha256_avx2_extend_doubleround 36
+	sha256_avx2_extend_doubleround 38
+	sha256_avx2_extend_doubleround 40
+	sha256_avx2_extend_doubleround 42
+	jz sha256d_ms_8way_avx2_extend_coda2
+	sha256_avx2_extend_doubleround 44
+	sha256_avx2_extend_doubleround 46
+	
+	vmovdqa	0(%rcx), %ymm7
+	vmovdqa	32(%rcx), %ymm8
+	vmovdqa	64(%rcx), %ymm9
+	vmovdqa	96(%rcx), %ymm10
+	vmovdqa	128(%rcx), %ymm0
+	vmovdqa	160(%rcx), %ymm5
+	vmovdqa	192(%rcx), %ymm4
+	vmovdqa	224(%rcx), %ymm3
+	
+	movq	%rsi, %rax
+	leaq	sha256_8k(%rip), %rcx
+	jmp sha256d_ms_8way_avx2_main_loop1
+	
+sha256d_ms_8way_avx2_main_loop2:
+	sha256_avx2_main_round 0, %ymm10, %ymm9, %ymm8, %ymm0, %ymm3, %ymm4, %ymm5, %ymm7
+	sha256_avx2_main_round 1, %ymm9, %ymm8, %ymm0, %ymm10, %ymm4, %ymm5, %ymm7, %ymm3
+	sha256_avx2_main_round 2, %ymm8, %ymm0, %ymm10, %ymm9, %ymm5, %ymm7, %ymm3, %ymm4
+sha256d_ms_8way_avx2_main_loop1:
+	sha256_avx2_main_round 3, %ymm0, %ymm10, %ymm9, %ymm8, %ymm7, %ymm3, %ymm4, %ymm5
+	sha256_avx2_main_quadround 4
+	sha256_avx2_main_quadround 8
+	sha256_avx2_main_quadround 12
+	sha256_avx2_main_quadround 16
+	sha256_avx2_main_quadround 20
+	sha256_avx2_main_quadround 24
+	sha256_avx2_main_quadround 28
+	sha256_avx2_main_quadround 32
+	sha256_avx2_main_quadround 36
+	sha256_avx2_main_quadround 40
+	sha256_avx2_main_quadround 44
+	sha256_avx2_main_quadround 48
+	sha256_avx2_main_quadround 52
+	sha256_avx2_main_round 56, %ymm10, %ymm9, %ymm8, %ymm0, %ymm3, %ymm4, %ymm5, %ymm7
+	jz sha256d_ms_8way_avx2_finish
+	sha256_avx2_main_round 57, %ymm9, %ymm8, %ymm0, %ymm10, %ymm4, %ymm5, %ymm7, %ymm3
+	sha256_avx2_main_round 58, %ymm8, %ymm0, %ymm10, %ymm9, %ymm5, %ymm7, %ymm3, %ymm4
+	sha256_avx2_main_round 59, %ymm0, %ymm10, %ymm9, %ymm8, %ymm7, %ymm3, %ymm4, %ymm5
+	sha256_avx2_main_quadround 60
+	
+	vmovdqa	2*32(%rsp), %ymm1
+	vmovdqa	3*32(%rsp), %ymm2
+	vmovdqa	4*32(%rsp), %ymm6
+	vmovdqa	%ymm1, 18*32(%rsi)
+	vmovdqa	%ymm2, 19*32(%rsi)
+	vmovdqa	%ymm6, 20*32(%rsi)
+	vmovdqa	6*32(%rsp), %ymm1
+	vmovdqa	7*32(%rsp), %ymm2
+	vmovdqa	8*32(%rsp), %ymm6
+	vmovdqa	%ymm1, 22*32(%rsi)
+	vmovdqa	%ymm2, 23*32(%rsi)
+	vmovdqa	%ymm6, 24*32(%rsi)
+	vmovdqa	14*32(%rsp), %ymm1
+	vmovdqa	15*32(%rsp), %ymm2
+	vmovdqa	%ymm1, 30*32(%rsi)
+	vmovdqa	%ymm2, 31*32(%rsi)
+	
+	vpaddd	0(%rdx), %ymm7, %ymm7
+	vpaddd	32(%rdx), %ymm5, %ymm5
+	vpaddd	64(%rdx), %ymm4, %ymm4
+	vpaddd	96(%rdx), %ymm3, %ymm3
+	vpaddd	128(%rdx), %ymm0, %ymm0
+	vpaddd	160(%rdx), %ymm8, %ymm8
+	vpaddd	192(%rdx), %ymm9, %ymm9
+	vpaddd	224(%rdx), %ymm10, %ymm10
+	
+	vmovdqa	%ymm7, 0(%rsp)
+	vmovdqa	%ymm5, 32(%rsp)
+	vmovdqa	%ymm4, 64(%rsp)
+	vmovdqa	%ymm3, 96(%rsp)
+	vmovdqa	%ymm0, 128(%rsp)
+	vmovdqa	%ymm8, 160(%rsp)
+	vmovdqa	%ymm9, 192(%rsp)
+	vmovdqa	%ymm10, 224(%rsp)
+	
+	vpxor	%ymm0, %ymm0, %ymm0
+	movq	$0x8000000000000100, %rax
+	vmovd	%rax, %xmm1
+	vinserti128	$1, %xmm1, %ymm1, %ymm1
+	vpshufd	$0x55, %ymm1, %ymm2
+	vpshufd	$0x00, %ymm1, %ymm1
+	vmovdqa	%ymm2, 8*32(%rsp)
+	vmovdqa	%ymm0, 9*32(%rsp)
+	vmovdqa	%ymm0, 10*32(%rsp)
+	vmovdqa	%ymm0, 11*32(%rsp)
+	vmovdqa	%ymm0, 12*32(%rsp)
+	vmovdqa	%ymm0, 13*32(%rsp)
+	vmovdqa	%ymm0, 14*32(%rsp)
+	vmovdqa	%ymm1, 15*32(%rsp)
+	
+	leaq	16*32(%rsp), %rax
+	cmpq	%rax, %rax
+	
+	vmovdqa	-15*32(%rax), %ymm0
+	vmovdqa	-14*32(%rax), %ymm4
+	vpslld	$14, %ymm0, %ymm2
+	vpslld	$14, %ymm4, %ymm6
+	vpsrld	$3, %ymm0, %ymm8
+	vpsrld	$3, %ymm4, %ymm4
+	vpsrld	$7, %ymm0, %ymm1
+	vpsrld	$4, %ymm4, %ymm5
+	vpxor	%ymm1, %ymm8, %ymm8
+	vpxor	%ymm5, %ymm4, %ymm4
+	vpsrld	$11, %ymm1, %ymm1
+	vpsrld	$11, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm8, %ymm8
+	vpxor	%ymm6, %ymm4, %ymm4
+	vpslld	$11, %ymm2, %ymm2
+	vpslld	$11, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm8, %ymm8
+	vpxor	%ymm5, %ymm4, %ymm4
+	vpxor	%ymm2, %ymm8, %ymm8
+	vpxor	%ymm6, %ymm4, %ymm4
+	vpaddd	%ymm0, %ymm4, %ymm4
+	vpaddd	-16*32(%rax), %ymm8, %ymm3
+	vpaddd	sha256d_8preext2_17(%rip), %ymm4, %ymm7
+	vmovdqa	%ymm3, 0*32(%rax)
+	vmovdqa	%ymm7, 1*32(%rax)
+	
+	sha256_avx2_extend_doubleround 2
+	sha256_avx2_extend_doubleround 4
+	
+	vmovdqa	-9*32(%rax), %ymm0
+	vpslld	$14, %ymm0, %ymm2
+	vpsrld	$3, %ymm0, %ymm8
+	vpsrld	$7, %ymm0, %ymm1
+	vpxor	%ymm1, %ymm8, %ymm8
+	vpxor	%ymm2, %ymm8, %ymm8
+	vpsrld	$11, %ymm1, %ymm1
+	vpslld	$11, %ymm2, %ymm2
+	vpxor	%ymm1, %ymm8, %ymm8
+	vpxor	%ymm2, %ymm8, %ymm8
+	vpaddd	sha256d_8preext2_23(%rip), %ymm0, %ymm4
+	vpaddd	-10*32(%rax), %ymm8, %ymm0
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpaddd	-1*32(%rax), %ymm0, %ymm0
+	vpaddd	0*32(%rax), %ymm4, %ymm4
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vpaddd	%ymm4, %ymm7, %ymm7
+	vmovdqa	%ymm3, 6*32(%rax)
+	vmovdqa	%ymm7, 7*32(%rax)
+	
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	sha256d_8preext2_24(%rip), %ymm3, %ymm3
+	vpaddd	1*32(%rax), %ymm3, %ymm3
+	vpaddd	2*32(%rax), %ymm7, %ymm7
+	vmovdqa	%ymm3, 8*32(%rax)
+	vmovdqa	%ymm7, 9*32(%rax)
+	
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	3*32(%rax), %ymm3, %ymm3
+	vpaddd	4*32(%rax), %ymm7, %ymm7
+	vmovdqa	%ymm3, 10*32(%rax)
+	vmovdqa	%ymm7, 11*32(%rax)
+	
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	5*32(%rax), %ymm3, %ymm3
+	vpaddd	6*32(%rax), %ymm7, %ymm7
+	vmovdqa	%ymm3, 12*32(%rax)
+	vmovdqa	%ymm7, 13*32(%rax)
+	
+	vmovdqa	sha256d_8preext2_30(%rip), %ymm0
+	vmovdqa	0*32(%rax), %ymm4
+	vpslld	$14, %ymm4, %ymm6
+	vpsrld	$3, %ymm4, %ymm4
+	vpsrld	$4, %ymm4, %ymm5
+	vpxor	%ymm5, %ymm4, %ymm4
+	vpxor	%ymm6, %ymm4, %ymm4
+	vpsrld	$11, %ymm5, %ymm5
+	vpslld	$11, %ymm6, %ymm6
+	vpxor	%ymm5, %ymm4, %ymm4
+	vpxor	%ymm6, %ymm4, %ymm4
+	vpaddd	-1*32(%rax), %ymm4, %ymm4
+	vpslld	$13, %ymm3, %ymm2
+	vpslld	$13, %ymm7, %ymm6
+	vpsrld	$10, %ymm3, %ymm3
+	vpsrld	$10, %ymm7, %ymm7
+	vpaddd	7*32(%rax), %ymm0, %ymm0
+	vpaddd	8*32(%rax), %ymm4, %ymm4
+	vpsrld	$7, %ymm3, %ymm1
+	vpsrld	$7, %ymm7, %ymm5
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpsrld	$2, %ymm1, %ymm1
+	vpsrld	$2, %ymm5, %ymm5
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpslld	$2, %ymm2, %ymm2
+	vpslld	$2, %ymm6, %ymm6
+	vpxor	%ymm1, %ymm3, %ymm3
+	vpxor	%ymm5, %ymm7, %ymm7
+	vpxor	%ymm2, %ymm3, %ymm3
+	vpxor	%ymm6, %ymm7, %ymm7
+	vpaddd	%ymm0, %ymm3, %ymm3
+	vpaddd	%ymm4, %ymm7, %ymm7
+	vmovdqa	%ymm3, 14*32(%rax)
+	vmovdqa	%ymm7, 15*32(%rax)
+	
+	jmp sha256d_ms_8way_avx2_extend_loop2
+	
+sha256d_ms_8way_avx2_extend_coda2:
+	sha256_avx2_extend_round 44
+	
+	vmovdqa	sha256_8h+0(%rip), %ymm7
+	vmovdqa	sha256_8h+32(%rip), %ymm5
+	vmovdqa	sha256_8h+64(%rip), %ymm4
+	vmovdqa	sha256_8h+96(%rip), %ymm3
+	vmovdqa	sha256_8h+128(%rip), %ymm0
+	vmovdqa	sha256_8h+160(%rip), %ymm8
+	vmovdqa	sha256_8h+192(%rip), %ymm9
+	vmovdqa	sha256_8h+224(%rip), %ymm10
+	
+	movq	%rsp, %rax
+	leaq	sha256_8k(%rip), %rcx
+	jmp sha256d_ms_8way_avx2_main_loop2
+
+.macro sha256_avx2_main_round_red i, r0, r1, r2, r3, r4
+	vpaddd	32*\i(%rax), \r0, %ymm6
+	vpaddd	32*\i(%rcx), %ymm6, %ymm6
+	vpandn	\r1, \r3, %ymm1
+	vpand	\r3, \r2, %ymm2
+	vpxor	%ymm2, %ymm1, %ymm1
+	vpaddd	%ymm1, %ymm6, %ymm6
+	vpslld	$7, \r3, %ymm1
+	vpsrld	$6, \r3, \r0
+	vpsrld	$5, \r0, %ymm2
+	vpxor	%ymm1, \r0, \r0
+	vpxor	%ymm2, \r0, \r0
+	vpslld	$14, %ymm1, %ymm1
+	vpsrld	$14, %ymm2, %ymm2
+	vpxor	%ymm1, \r0, \r0
+	vpxor	%ymm2, \r0, \r0
+	vpslld	$5, %ymm1, %ymm1
+	vpxor	%ymm1, \r0, \r0
+	vpaddd	\r0, %ymm6, %ymm6
+	vpaddd	%ymm6, \r4, \r0
+.endm
+
+sha256d_ms_8way_avx2_finish:
+	sha256_avx2_main_round_red 57, %ymm9, %ymm8, %ymm0, %ymm10, %ymm4
+	sha256_avx2_main_round_red 58, %ymm8, %ymm0, %ymm10, %ymm9, %ymm5
+	sha256_avx2_main_round_red 59, %ymm0, %ymm10, %ymm9, %ymm8, %ymm7
+	sha256_avx2_main_round_red 60, %ymm10, %ymm9, %ymm8, %ymm0, %ymm3
+	
+	vpaddd	sha256_8h+224(%rip), %ymm10, %ymm10
+	vmovdqa	%ymm10, 224(%rdi)
+	
+	movq	%rbp, %rsp
+	popq	%rbp
+#if defined(_WIN64) || defined(__CYGWIN__)
+	popq	%rsi
+	vmovdqa	0(%rsp), %xmm6
+	vmovdqa	16(%rsp), %xmm7
+	vmovdqa	32(%rsp), %xmm8
+	vmovdqa	48(%rsp), %xmm9
+	vmovdqa	64(%rsp), %xmm10
+	addq	$80, %rsp
+	popq	%rdi
+#endif
+	ret
+
+
+	.text
+	.p2align 6
+	.globl sha256_use_8way
+	.globl _sha256_use_8way
+sha256_use_8way:
+_sha256_use_8way:
+	pushq	%rbx
+	
+	/* Check for AVX and OSXSAVE support */
+	movl	$1, %eax
+	cpuid
+	andl	$0x18000000, %ecx
+	cmpl	$0x18000000, %ecx
+	jne sha256_use_8way_no
+	/* Check for AVX2 support */
+	movl	$7, %eax
+	xorl	%ecx, %ecx
+	cpuid
+	andl	$0x00000020, %ebx
+	cmpl	$0x00000020, %ebx
+	jne sha256_use_8way_no
+	/* Check for XMM and YMM state support */
+	xorl	%ecx, %ecx
+	xgetbv
+	andl	$0x00000006, %eax
+	cmpl	$0x00000006, %eax
+	jne sha256_use_8way_no
+	
+sha256_use_8way_yes:
+	movl	$1, %eax
+	jmp sha256_use_8way_done
+	
+sha256_use_8way_no:
+	xorl	%eax, %eax
+	
+sha256_use_8way_done:
+	popq	%rbx
+	ret
+
+#endif /* USE_AVX2 */
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86-notextrel-test.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86-notextrel-test.c
new file mode 100644
index 0000000..8dc415a
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86-notextrel-test.c
@@ -0,0 +1,81 @@
+#include<stdio.h>
+#include<stdint.h>
+#include<stdlib.h>
+
+#define STATE_NUM 32 
+
+int sha256_use_4way();
+void sha256_init_4way(uint32_t *state);
+void sha256_transform_4way(uint32_t *state, const uint32_t *block, int swap);
+void sha256d_ms_4way(uint32_t *hash,  uint32_t *data, const uint32_t *midstate, const uint32_t *prehash);
+
+void test_sha256_use_4way() {
+    printf("%d\n", sha256_use_4way());
+}
+
+void test_sha256_init_4way() {
+    uint32_t state[STATE_NUM];
+    sha256_init_4way(state);
+
+    for(int i = 0; i < STATE_NUM; i++) {
+        printf("%08x", state[i]);
+    }
+
+    printf("\n");
+}
+
+void test_sha256_transform_4way() {
+    uint32_t state[STATE_NUM] __attribute__((aligned(128))) = { 0 };
+    uint32_t block[STATE_NUM] __attribute__((aligned(128))) = { 0 };
+
+    for (int i = 0; i < STATE_NUM; i++) {
+        state[i] = rand();
+        block[i] = rand();
+    }
+
+    sha256_transform_4way(state, block, 0);
+
+    for(int i = 0; i < STATE_NUM; i++) {
+        printf("%08x ", state[i]);
+    }
+    printf("\n");
+}
+
+void test_sha256d_ms_4way() {
+	uint32_t hash[STATE_NUM] __attribute__((aligned(32))) = { 0 };
+	uint32_t data[8 * STATE_NUM] __attribute__((aligned(128))) = { 0 };
+	uint32_t midstate[STATE_NUM] __attribute__((aligned(32))) = { 0 };
+	uint32_t prehash[STATE_NUM] __attribute__((aligned(32))) = { 0 };
+
+    for (int i = 0; i < STATE_NUM; i++) {
+        hash[i] = rand();
+        midstate[i] = rand();
+        prehash[i] = rand();
+    }
+
+    for (int i = 0; i < 8 * STATE_NUM; i++) {
+        data[i] = rand();
+    }
+
+    sha256d_ms_4way(hash, data, midstate, prehash);
+
+    for(int i = 0; i < STATE_NUM; i++) {
+        printf("%08x ", hash[i]);
+    }
+    printf("\n");
+
+    for(int i = 0; i < 8 * STATE_NUM; i++) {
+        printf("%08x ", data[i]);
+    }
+    printf("\n");
+}
+
+int main() {
+    test_sha256_use_4way();
+    test_sha256_init_4way();
+    for (int i = 0; i < 100; i++) {
+        test_sha256_transform_4way();
+        test_sha256d_ms_4way();
+    }
+    return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86-notextrel.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86-notextrel.S
new file mode 100644
index 0000000..992721e
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86-notextrel.S
@@ -0,0 +1,1240 @@
+/*
+ * Copyright 2012 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(__linux__) && defined(__ELF__)
+	.section .note.GNU-stack,"",%progbits
+#endif
+
+#if defined(USE_ASM) && defined(__i386__)
+
+
+.macro sha256_4h
+	pushl $0x5be0cd19; pushl $0x5be0cd19; pushl $0x5be0cd19; pushl $0x5be0cd19
+	pushl $0x1f83d9ab; pushl $0x1f83d9ab; pushl $0x1f83d9ab; pushl $0x1f83d9ab
+	pushl $0x9b05688c; pushl $0x9b05688c; pushl $0x9b05688c; pushl $0x9b05688c
+	pushl $0x510e527f; pushl $0x510e527f; pushl $0x510e527f; pushl $0x510e527f
+	pushl $0xa54ff53a; pushl $0xa54ff53a; pushl $0xa54ff53a; pushl $0xa54ff53a
+	pushl $0x3c6ef372; pushl $0x3c6ef372; pushl $0x3c6ef372; pushl $0x3c6ef372
+	pushl $0xbb67ae85; pushl $0xbb67ae85; pushl $0xbb67ae85; pushl $0xbb67ae85
+	pushl $0x6a09e667; pushl $0x6a09e667; pushl $0x6a09e667; pushl $0x6a09e667
+.endm
+
+.macro sha256_4k
+    pushl $0xc67178f2; pushl $0xc67178f2; pushl $0xc67178f2; pushl $0xc67178f2
+    pushl $0xbef9a3f7; pushl $0xbef9a3f7; pushl $0xbef9a3f7; pushl $0xbef9a3f7
+    pushl $0xa4506ceb; pushl $0xa4506ceb; pushl $0xa4506ceb; pushl $0xa4506ceb
+    pushl $0x90befffa; pushl $0x90befffa; pushl $0x90befffa; pushl $0x90befffa
+    pushl $0x8cc70208; pushl $0x8cc70208; pushl $0x8cc70208; pushl $0x8cc70208
+    pushl $0x84c87814; pushl $0x84c87814; pushl $0x84c87814; pushl $0x84c87814
+    pushl $0x78a5636f; pushl $0x78a5636f; pushl $0x78a5636f; pushl $0x78a5636f
+    pushl $0x748f82ee; pushl $0x748f82ee; pushl $0x748f82ee; pushl $0x748f82ee
+    pushl $0x682e6ff3; pushl $0x682e6ff3; pushl $0x682e6ff3; pushl $0x682e6ff3
+    pushl $0x5b9cca4f; pushl $0x5b9cca4f; pushl $0x5b9cca4f; pushl $0x5b9cca4f
+    pushl $0x4ed8aa4a; pushl $0x4ed8aa4a; pushl $0x4ed8aa4a; pushl $0x4ed8aa4a
+    pushl $0x391c0cb3; pushl $0x391c0cb3; pushl $0x391c0cb3; pushl $0x391c0cb3
+    pushl $0x34b0bcb5; pushl $0x34b0bcb5; pushl $0x34b0bcb5; pushl $0x34b0bcb5
+    pushl $0x2748774c; pushl $0x2748774c; pushl $0x2748774c; pushl $0x2748774c
+    pushl $0x1e376c08; pushl $0x1e376c08; pushl $0x1e376c08; pushl $0x1e376c08
+    pushl $0x19a4c116; pushl $0x19a4c116; pushl $0x19a4c116; pushl $0x19a4c116
+    pushl $0x106aa070; pushl $0x106aa070; pushl $0x106aa070; pushl $0x106aa070
+    pushl $0xf40e3585; pushl $0xf40e3585; pushl $0xf40e3585; pushl $0xf40e3585
+    pushl $0xd6990624; pushl $0xd6990624; pushl $0xd6990624; pushl $0xd6990624
+    pushl $0xd192e819; pushl $0xd192e819; pushl $0xd192e819; pushl $0xd192e819
+    pushl $0xc76c51a3; pushl $0xc76c51a3; pushl $0xc76c51a3; pushl $0xc76c51a3
+    pushl $0xc24b8b70; pushl $0xc24b8b70; pushl $0xc24b8b70; pushl $0xc24b8b70
+    pushl $0xa81a664b; pushl $0xa81a664b; pushl $0xa81a664b; pushl $0xa81a664b
+    pushl $0xa2bfe8a1; pushl $0xa2bfe8a1; pushl $0xa2bfe8a1; pushl $0xa2bfe8a1
+    pushl $0x92722c85; pushl $0x92722c85; pushl $0x92722c85; pushl $0x92722c85
+    pushl $0x81c2c92e; pushl $0x81c2c92e; pushl $0x81c2c92e; pushl $0x81c2c92e
+    pushl $0x766a0abb; pushl $0x766a0abb; pushl $0x766a0abb; pushl $0x766a0abb
+    pushl $0x650a7354; pushl $0x650a7354; pushl $0x650a7354; pushl $0x650a7354
+    pushl $0x53380d13; pushl $0x53380d13; pushl $0x53380d13; pushl $0x53380d13
+    pushl $0x4d2c6dfc; pushl $0x4d2c6dfc; pushl $0x4d2c6dfc; pushl $0x4d2c6dfc
+    pushl $0x2e1b2138; pushl $0x2e1b2138; pushl $0x2e1b2138; pushl $0x2e1b2138
+    pushl $0x27b70a85; pushl $0x27b70a85; pushl $0x27b70a85; pushl $0x27b70a85
+    pushl $0x14292967; pushl $0x14292967; pushl $0x14292967; pushl $0x14292967
+    pushl $0x06ca6351; pushl $0x06ca6351; pushl $0x06ca6351; pushl $0x06ca6351
+    pushl $0xd5a79147; pushl $0xd5a79147; pushl $0xd5a79147; pushl $0xd5a79147
+    pushl $0xc6e00bf3; pushl $0xc6e00bf3; pushl $0xc6e00bf3; pushl $0xc6e00bf3
+    pushl $0xbf597fc7; pushl $0xbf597fc7; pushl $0xbf597fc7; pushl $0xbf597fc7
+    pushl $0xb00327c8; pushl $0xb00327c8; pushl $0xb00327c8; pushl $0xb00327c8
+    pushl $0xa831c66d; pushl $0xa831c66d; pushl $0xa831c66d; pushl $0xa831c66d
+    pushl $0x983e5152; pushl $0x983e5152; pushl $0x983e5152; pushl $0x983e5152
+    pushl $0x76f988da; pushl $0x76f988da; pushl $0x76f988da; pushl $0x76f988da
+    pushl $0x5cb0a9dc; pushl $0x5cb0a9dc; pushl $0x5cb0a9dc; pushl $0x5cb0a9dc
+    pushl $0x4a7484aa; pushl $0x4a7484aa; pushl $0x4a7484aa; pushl $0x4a7484aa
+    pushl $0x2de92c6f; pushl $0x2de92c6f; pushl $0x2de92c6f; pushl $0x2de92c6f
+    pushl $0x240ca1cc; pushl $0x240ca1cc; pushl $0x240ca1cc; pushl $0x240ca1cc
+    pushl $0x0fc19dc6; pushl $0x0fc19dc6; pushl $0x0fc19dc6; pushl $0x0fc19dc6
+    pushl $0xefbe4786; pushl $0xefbe4786; pushl $0xefbe4786; pushl $0xefbe4786
+    pushl $0xe49b69c1; pushl $0xe49b69c1; pushl $0xe49b69c1; pushl $0xe49b69c1
+    pushl $0xc19bf174; pushl $0xc19bf174; pushl $0xc19bf174; pushl $0xc19bf174
+    pushl $0x9bdc06a7; pushl $0x9bdc06a7; pushl $0x9bdc06a7; pushl $0x9bdc06a7
+    pushl $0x80deb1fe; pushl $0x80deb1fe; pushl $0x80deb1fe; pushl $0x80deb1fe
+    pushl $0x72be5d74; pushl $0x72be5d74; pushl $0x72be5d74; pushl $0x72be5d74
+    pushl $0x550c7dc3; pushl $0x550c7dc3; pushl $0x550c7dc3; pushl $0x550c7dc3
+    pushl $0x243185be; pushl $0x243185be; pushl $0x243185be; pushl $0x243185be
+    pushl $0x12835b01; pushl $0x12835b01; pushl $0x12835b01; pushl $0x12835b01
+    pushl $0xd807aa98; pushl $0xd807aa98; pushl $0xd807aa98; pushl $0xd807aa98
+    pushl $0xab1c5ed5; pushl $0xab1c5ed5; pushl $0xab1c5ed5; pushl $0xab1c5ed5
+    pushl $0x923f82a4; pushl $0x923f82a4; pushl $0x923f82a4; pushl $0x923f82a4
+    pushl $0x59f111f1; pushl $0x59f111f1; pushl $0x59f111f1; pushl $0x59f111f1
+    pushl $0x3956c25b; pushl $0x3956c25b; pushl $0x3956c25b; pushl $0x3956c25b
+    pushl $0xe9b5dba5; pushl $0xe9b5dba5; pushl $0xe9b5dba5; pushl $0xe9b5dba5
+    pushl $0xb5c0fbcf; pushl $0xb5c0fbcf; pushl $0xb5c0fbcf; pushl $0xb5c0fbcf
+    pushl $0x71374491; pushl $0x71374491; pushl $0x71374491; pushl $0x71374491
+    pushl $0x428a2f98; pushl $0x428a2f98; pushl $0x428a2f98; pushl $0x428a2f98
+.endm
+
+.macro sha256d_4preext2_15
+	pushl $0x00000100; pushl $0x00000100; pushl $0x00000100; pushl $0x00000100
+.endm
+.macro sha256d_4preext2_17
+	pushl $0x00a00000; pushl $0x00a00000; pushl $0x00a00000; pushl $0x00a00000
+.endm
+.macro sha256d_4preext2_23
+	pushl $0x11002000; pushl $0x11002000; pushl $0x11002000; pushl $0x11002000
+.endm
+.macro sha256d_4preext2_24
+	pushl $0x80000000; pushl $0x80000000; pushl $0x80000000; pushl $0x80000000
+.endm
+.macro sha256d_4preext2_30
+	pushl $0x00400022; pushl $0x00400022; pushl $0x00400022; pushl $0x00400022
+.endm
+
+
+	.text
+	.p2align 5
+	.globl sha256_init_4way
+	.globl _sha256_init_4way
+sha256_init_4way:
+_sha256_init_4way:
+	movl	4(%esp), %edx
+    pushl   %ebp
+    movl    %esp, %ebp
+    and     $0xfffffff0, %esp
+    subl    $128, %esp
+    pushl   %ebp
+    subl    $124, %esp
+    sha256_4h
+    movl    %esp, %ebp
+	movdqa	(%ebp), %xmm0    #sha256_4h
+	movdqa	16(%ebp), %xmm1  #sha256_4h
+	movdqa	32(%ebp), %xmm2  #sha256_4h
+	movdqa	48(%ebp), %xmm3  #sha256_4h
+	movdqu	%xmm0, 0(%edx)
+	movdqu	%xmm1, 16(%edx)
+	movdqu	%xmm2, 32(%edx)
+	movdqu	%xmm3, 48(%edx)
+	movdqa	64(%ebp), %xmm0  #sha256_4h
+	movdqa	80(%ebp), %xmm1  #sha256_4h
+	movdqa	96(%ebp), %xmm2  #sha256_4h
+	movdqa	112(%ebp), %xmm3 #sha256_4h
+	movdqu	%xmm0, 64(%edx)
+	movdqu	%xmm1, 80(%edx)
+	movdqu	%xmm2, 96(%edx)
+	movdqu	%xmm3, 112(%edx)
+    movl    %ebp, %esp
+    addl    $128, %esp
+    addl    $124, %esp
+    popl    %ebp
+	movl	%ebp, %esp
+    popl    %ebp
+	ret
+
+
+.macro sha256_sse2_extend_round i
+	movdqa	(\i-15)*16(%eax), %xmm0
+	movdqa	%xmm0, %xmm2
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	(\i-16)*16(%eax), %xmm0
+	paddd	(\i-7)*16(%eax), %xmm0
+
+	movdqa	%xmm3, %xmm2
+	psrld	$10, %xmm3
+	pslld	$13, %xmm2
+	movdqa	%xmm3, %xmm1
+	psrld	$7, %xmm1
+	pxor	%xmm1, %xmm3
+	pxor	%xmm2, %xmm3
+	psrld	$2, %xmm1
+	pslld	$2, %xmm2
+	pxor	%xmm1, %xmm3
+	pxor	%xmm2, %xmm3
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, \i*16(%eax)
+.endm
+
+.macro sha256_sse2_extend_doubleround i
+	movdqa	(\i-15)*16(%eax), %xmm0
+	movdqa	(\i-14)*16(%eax), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+
+	paddd	(\i-16)*16(%eax), %xmm0
+	paddd	(\i-15)*16(%eax), %xmm4
+
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+
+	paddd	(\i-7)*16(%eax), %xmm0
+	paddd	(\i-6)*16(%eax), %xmm4
+
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, \i*16(%eax)
+	movdqa	%xmm7, (\i+1)*16(%eax)
+.endm
+
+.macro sha256_sse2_main_round i
+	movdqa	16*(\i)(%eax), %xmm6
+
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%esp), %xmm2
+	pandn	%xmm2, %xmm1
+	paddd	32(%esp), %xmm6
+
+	movdqa	%xmm2, 32(%esp)
+	movdqa	0(%esp), %xmm2
+	movdqa	%xmm2, 16(%esp)
+
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%esp)
+
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	paddd	(128+16*(\i))(%ebp), %xmm6 #sha256_4k
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm2, %xmm0
+	pxor	%xmm1, %xmm0
+	movdqa	%xmm5, %xmm1
+	paddd	%xmm0, %xmm6
+
+	movdqa	%xmm3, %xmm0
+	movdqa	%xmm4, %xmm3
+	movdqa	%xmm4, %xmm2
+	paddd	%xmm6, %xmm0
+	pand	%xmm5, %xmm2
+	pand	%xmm7, %xmm1
+	pand	%xmm7, %xmm4
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm5, %xmm4
+	movdqa	%xmm7, %xmm5
+	pxor	%xmm2, %xmm1
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm7, %xmm2
+	psrld	$2, %xmm7
+	movdqa	%xmm7, %xmm1
+	pslld	$10, %xmm2
+	psrld	$11, %xmm1
+	pxor	%xmm2, %xmm7
+	pslld	$9, %xmm2
+	pxor	%xmm1, %xmm7
+	psrld	$9, %xmm1
+	pxor	%xmm2, %xmm7
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm7
+	pxor	%xmm2, %xmm7
+	paddd	%xmm6, %xmm7
+.endm
+
+.macro sha256_sse2_main_quadround i
+	sha256_sse2_main_round \i+0
+	sha256_sse2_main_round \i+1
+	sha256_sse2_main_round \i+2
+	sha256_sse2_main_round \i+3
+.endm
+
+
+.macro p2bswap_esi_esp i
+	movdqu	\i*16(%esi), %xmm0
+	movdqu	(\i+1)*16(%esi), %xmm2
+	pshuflw	$0xb1, %xmm0, %xmm0
+	pshuflw	$0xb1, %xmm2, %xmm2
+	pshufhw	$0xb1, %xmm0, %xmm0
+	pshufhw	$0xb1, %xmm2, %xmm2
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm2, %xmm3
+	psrlw	$8, %xmm1
+	psrlw	$8, %xmm3
+	psllw	$8, %xmm0
+	psllw	$8, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm3, %xmm2
+	movdqa	%xmm0, (\i+3)*16(%esp)
+	movdqa	%xmm2, (\i+4)*16(%esp)
+.endm
+
+	.text
+	.p2align 5
+	.globl sha256_transform_4way
+	.globl _sha256_transform_4way
+sha256_transform_4way:
+_sha256_transform_4way:
+	pushl	%edi
+	pushl	%esi
+	movl	12(%esp), %edi
+	movl	16(%esp), %esi
+	movl	20(%esp), %ecx
+	movl	%esp, %edx
+
+    pushl   %ebp
+    movl    %esp, %ebp
+    and     $0xfffffff0, %esp
+    subl    $128, %esp
+    pushl   %ebp
+    subl    $124, %esp
+    sha256_4k
+    movl    %esp, %ebp
+
+	subl	$67*16, %esp
+	andl	$-128, %esp
+	
+	testl	%ecx, %ecx
+	jnz sha256_transform_4way_swap
+	
+	movdqu	0*16(%esi), %xmm0
+	movdqu	1*16(%esi), %xmm1
+	movdqu	2*16(%esi), %xmm2
+	movdqu	3*16(%esi), %xmm3
+	movdqu	4*16(%esi), %xmm4
+	movdqu	5*16(%esi), %xmm5
+	movdqu	6*16(%esi), %xmm6
+	movdqu	7*16(%esi), %xmm7
+	movdqa	%xmm0, 3*16(%esp)
+	movdqa	%xmm1, 4*16(%esp)
+	movdqa	%xmm2, 5*16(%esp)
+	movdqa	%xmm3, 6*16(%esp)
+	movdqa	%xmm4, 7*16(%esp)
+	movdqa	%xmm5, 8*16(%esp)
+	movdqa	%xmm6, 9*16(%esp)
+	movdqa	%xmm7, 10*16(%esp)
+	movdqu	8*16(%esi), %xmm0
+	movdqu	9*16(%esi), %xmm1
+	movdqu	10*16(%esi), %xmm2
+	movdqu	11*16(%esi), %xmm3
+	movdqu	12*16(%esi), %xmm4
+	movdqu	13*16(%esi), %xmm5
+	movdqu	14*16(%esi), %xmm6
+	movdqu	15*16(%esi), %xmm7
+	movdqa	%xmm0, 11*16(%esp)
+	movdqa	%xmm1, 12*16(%esp)
+	movdqa	%xmm2, 13*16(%esp)
+	movdqa	%xmm3, 14*16(%esp)
+	movdqa	%xmm4, 15*16(%esp)
+	movdqa	%xmm5, 16*16(%esp)
+	movdqa	%xmm6, 17*16(%esp)
+	movdqa	%xmm7, 18*16(%esp)
+	jmp sha256_transform_4way_extend
+	
+	.p2align 5
+sha256_transform_4way_swap:
+	p2bswap_esi_esp 0
+	p2bswap_esi_esp 2
+	p2bswap_esi_esp 4
+	p2bswap_esi_esp 6
+	p2bswap_esi_esp 8
+	p2bswap_esi_esp 10
+	p2bswap_esi_esp 12
+	p2bswap_esi_esp 14
+	
+sha256_transform_4way_extend:
+	leal	19*16(%esp), %ecx
+	leal	48*16(%ecx), %eax
+	movdqa	-2*16(%ecx), %xmm3
+	movdqa	-1*16(%ecx), %xmm7
+sha256_transform_4way_extend_loop:
+	movdqa	-15*16(%ecx), %xmm0
+	movdqa	-14*16(%ecx), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+
+	paddd	-16*16(%ecx), %xmm0
+	paddd	-15*16(%ecx), %xmm4
+
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+
+	paddd	-7*16(%ecx), %xmm0
+	paddd	-6*16(%ecx), %xmm4
+
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, (%ecx)
+	movdqa	%xmm7, 16(%ecx)
+	addl	$2*16, %ecx
+	cmpl	%ecx, %eax
+	jne sha256_transform_4way_extend_loop
+	
+	movdqu	0(%edi), %xmm7
+	movdqu	16(%edi), %xmm5
+	movdqu	32(%edi), %xmm4
+	movdqu	48(%edi), %xmm3
+	movdqu	64(%edi), %xmm0
+	movdqu	80(%edi), %xmm1
+	movdqu	96(%edi), %xmm2
+	movdqu	112(%edi), %xmm6
+	movdqa	%xmm1, 0(%esp)
+	movdqa	%xmm2, 16(%esp)
+	movdqa	%xmm6, 32(%esp)
+	
+	xorl	%eax, %eax
+sha256_transform_4way_main_loop:
+	movdqa	3*16(%esp, %eax), %xmm6
+	paddd	(%ebp, %eax), %xmm6 #sha256_4k
+	paddd	32(%esp), %xmm6
+
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%esp), %xmm2
+	pandn	%xmm2, %xmm1
+
+	movdqa	%xmm2, 32(%esp)
+	movdqa	0(%esp), %xmm2
+	movdqa	%xmm2, 16(%esp)
+
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%esp)
+
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm1, %xmm0
+	paddd	%xmm0, %xmm6
+
+	movdqa	%xmm3, %xmm0
+	paddd	%xmm6, %xmm0
+
+	movdqa	%xmm5, %xmm1
+	movdqa	%xmm4, %xmm3
+	movdqa	%xmm4, %xmm2
+	pand	%xmm5, %xmm2
+	pand	%xmm7, %xmm4
+	pand	%xmm7, %xmm1
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm5, %xmm4
+	movdqa	%xmm7, %xmm5
+	pxor	%xmm2, %xmm1
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm7, %xmm2
+	psrld	$2, %xmm7
+	movdqa	%xmm7, %xmm1
+	pslld	$10, %xmm2
+	psrld	$11, %xmm1
+	pxor	%xmm2, %xmm7
+	pxor	%xmm1, %xmm7
+	pslld	$9, %xmm2
+	psrld	$9, %xmm1
+	pxor	%xmm2, %xmm7
+	pxor	%xmm1, %xmm7
+	pslld	$11, %xmm2
+	pxor	%xmm2, %xmm7
+	paddd	%xmm6, %xmm7
+	
+	addl	$16, %eax
+	cmpl	$16*64, %eax
+	jne sha256_transform_4way_main_loop
+	
+	movdqu	0(%edi), %xmm1
+	movdqu	16(%edi), %xmm2
+	paddd	%xmm1, %xmm7
+	paddd	%xmm2, %xmm5
+	movdqu	32(%edi), %xmm1
+	movdqu	48(%edi), %xmm2
+	paddd	%xmm1, %xmm4
+	paddd	%xmm2, %xmm3
+	
+	movdqu	%xmm7, 0(%edi)
+	movdqu	%xmm5, 16(%edi)
+	movdqu	%xmm4, 32(%edi)
+	movdqu	%xmm3, 48(%edi)
+	
+	movdqu	64(%edi), %xmm1
+	movdqu	80(%edi), %xmm2
+	movdqu	96(%edi), %xmm6
+	movdqu	112(%edi), %xmm7
+	paddd	%xmm1, %xmm0
+	paddd	0(%esp), %xmm2
+	paddd	16(%esp), %xmm6
+	paddd	32(%esp), %xmm7
+	
+	movdqu	%xmm0, 64(%edi)
+	movdqu	%xmm2, 80(%edi)
+	movdqu	%xmm6, 96(%edi)
+	movdqu	%xmm7, 112(%edi)
+
+    movl    %ebp, %esp
+    addl    $1148, %esp
+    popl    %ebp
+    movl    %ebp, %esp
+    popl    %ebp
+	
+	movl	%edx, %esp
+	popl	%esi
+	popl	%edi
+	ret
+
+
+	.text
+	.p2align 5
+	.globl sha256d_ms_4way
+	.globl _sha256d_ms_4way
+sha256d_ms_4way:
+_sha256d_ms_4way:
+	pushl	%edi
+	pushl	%esi
+	pushl	%ebp
+	movl	16(%esp), %edi
+	movl	20(%esp), %esi
+	movl	24(%esp), %edx
+	movl	28(%esp), %ecx
+    movl    %esp, %ebp
+    and     $0xfffffff0, %esp
+    subl    $128, %esp
+    pushl   %ebp
+    subl    $124, %esp
+    sha256d_4preext2_30
+    sha256d_4preext2_24
+    sha256d_4preext2_23
+    sha256d_4preext2_17
+    sha256d_4preext2_15
+    sha256_4k
+    sha256_4h
+    movl    %esp, %ebp
+	subl	$67*16, %esp
+	andl	$-128, %esp
+	
+	leal	256(%esi), %eax
+	
+sha256d_ms_4way_extend_loop1:
+	movdqa	3*16(%esi), %xmm0
+	movdqa	2*16(%eax), %xmm3
+	movdqa	3*16(%eax), %xmm7
+	movdqa	%xmm3, 5*16(%esp)
+	movdqa	%xmm7, 6*16(%esp)
+	movdqa	%xmm0, %xmm2
+	paddd	%xmm0, %xmm7
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, 2*16(%eax)
+	movdqa	%xmm7, 3*16(%eax)
+	
+	movdqa	4*16(%eax), %xmm0
+	movdqa	%xmm0, 7*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, 4*16(%eax)
+	movdqa	%xmm7, 5*16(%eax)
+	
+	movdqa	6*16(%eax), %xmm0
+	movdqa	7*16(%eax), %xmm4
+	movdqa	%xmm0, 9*16(%esp)
+	movdqa	%xmm4, 10*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 6*16(%eax)
+	movdqa	%xmm7, 7*16(%eax)
+	
+	movdqa	8*16(%eax), %xmm0
+	movdqa	2*16(%eax), %xmm4
+	movdqa	%xmm0, 11*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 8*16(%eax)
+	movdqa	%xmm7, 9*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	3*16(%eax), %xmm3
+	paddd	4*16(%eax), %xmm7
+	movdqa	%xmm3, 10*16(%eax)
+	movdqa	%xmm7, 11*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	5*16(%eax), %xmm3
+	paddd	6*16(%eax), %xmm7
+	movdqa	%xmm3, 12*16(%eax)
+	movdqa	%xmm7, 13*16(%eax)
+	
+	movdqa	14*16(%eax), %xmm0
+	movdqa	15*16(%eax), %xmm4
+	movdqa	%xmm0, 17*16(%esp)
+	movdqa	%xmm4, 18*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	7*16(%eax), %xmm0
+	paddd	8*16(%eax), %xmm4
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 14*16(%eax)
+	movdqa	%xmm7, 15*16(%eax)
+	
+sha256d_ms_4way_extend_loop2:
+	sha256_sse2_extend_doubleround 16
+	sha256_sse2_extend_doubleround 18
+	sha256_sse2_extend_doubleround 20
+	sha256_sse2_extend_doubleround 22
+	sha256_sse2_extend_doubleround 24
+	sha256_sse2_extend_doubleround 26
+	sha256_sse2_extend_doubleround 28
+	sha256_sse2_extend_doubleround 30
+	sha256_sse2_extend_doubleround 32
+	sha256_sse2_extend_doubleround 34
+	sha256_sse2_extend_doubleround 36
+	sha256_sse2_extend_doubleround 38
+	sha256_sse2_extend_doubleround 40
+	sha256_sse2_extend_doubleround 42
+	jz sha256d_ms_4way_extend_coda2
+	sha256_sse2_extend_doubleround 44
+	sha256_sse2_extend_doubleround 46
+	
+	movdqa	0(%ecx), %xmm3
+	movdqa	16(%ecx), %xmm0
+	movdqa	32(%ecx), %xmm1
+	movdqa	48(%ecx), %xmm2
+	movdqa	64(%ecx), %xmm6
+	movdqa	80(%ecx), %xmm7
+	movdqa	96(%ecx), %xmm5
+	movdqa	112(%ecx), %xmm4
+	movdqa	%xmm1, 0(%esp)
+	movdqa	%xmm2, 16(%esp)
+	movdqa	%xmm6, 32(%esp)
+	
+	movl	%esi, %eax
+	jmp sha256d_ms_4way_main_loop1
+	
+sha256d_ms_4way_main_loop2:
+	sha256_sse2_main_round 0
+	sha256_sse2_main_round 1
+	sha256_sse2_main_round 2
+sha256d_ms_4way_main_loop1:
+	sha256_sse2_main_round 3
+	sha256_sse2_main_quadround 4
+	sha256_sse2_main_quadround 8
+	sha256_sse2_main_quadround 12
+	sha256_sse2_main_quadround 16
+	sha256_sse2_main_quadround 20
+	sha256_sse2_main_quadround 24
+	sha256_sse2_main_quadround 28
+	sha256_sse2_main_quadround 32
+	sha256_sse2_main_quadround 36
+	sha256_sse2_main_quadround 40
+	sha256_sse2_main_quadround 44
+	sha256_sse2_main_quadround 48
+	sha256_sse2_main_quadround 52
+	sha256_sse2_main_round 56
+	jz sha256d_ms_4way_finish
+	sha256_sse2_main_round 57
+	sha256_sse2_main_round 58
+	sha256_sse2_main_round 59
+	sha256_sse2_main_quadround 60
+	
+	movdqa	5*16(%esp), %xmm1
+	movdqa	6*16(%esp), %xmm2
+	movdqa	7*16(%esp), %xmm6
+	movdqa	%xmm1, 18*16(%esi)
+	movdqa	%xmm2, 19*16(%esi)
+	movdqa	%xmm6, 20*16(%esi)
+	movdqa	9*16(%esp), %xmm1
+	movdqa	10*16(%esp), %xmm2
+	movdqa	11*16(%esp), %xmm6
+	movdqa	%xmm1, 22*16(%esi)
+	movdqa	%xmm2, 23*16(%esi)
+	movdqa	%xmm6, 24*16(%esi)
+	movdqa	17*16(%esp), %xmm1
+	movdqa	18*16(%esp), %xmm2
+	movdqa	%xmm1, 30*16(%esi)
+	movdqa	%xmm2, 31*16(%esi)
+	
+	movdqa	0(%esp), %xmm1
+	movdqa	16(%esp), %xmm2
+	movdqa	32(%esp), %xmm6
+	paddd	0(%edx), %xmm7
+	paddd	16(%edx), %xmm5
+	paddd	32(%edx), %xmm4
+	paddd	48(%edx), %xmm3
+	paddd	64(%edx), %xmm0
+	paddd	80(%edx), %xmm1
+	paddd	96(%edx), %xmm2
+	paddd	112(%edx), %xmm6
+	
+	movdqa	%xmm7, 48+0(%esp)
+	movdqa	%xmm5, 48+16(%esp)
+	movdqa	%xmm4, 48+32(%esp)
+	movdqa	%xmm3, 48+48(%esp)
+	movdqa	%xmm0, 48+64(%esp)
+	movdqa	%xmm1, 48+80(%esp)
+	movdqa	%xmm2, 48+96(%esp)
+	movdqa	%xmm6, 48+112(%esp)
+	
+	movdqa	1152(%ebp), %xmm1 #sha256d_4preext2_15
+	movdqa	1200(%ebp), %xmm2 #sha256d_4preext2_24
+	pxor	%xmm0, %xmm0
+	movdqa	%xmm2, 48+128(%esp)
+	movdqa	%xmm0, 48+144(%esp)
+	movdqa	%xmm0, 48+160(%esp)
+	movdqa	%xmm0, 48+176(%esp)
+	movdqa	%xmm0, 48+192(%esp)
+	movdqa	%xmm0, 48+208(%esp)
+	movdqa	%xmm0, 48+224(%esp)
+	movdqa	%xmm1, 48+240(%esp)
+	
+	leal	19*16(%esp), %eax
+	cmpl	%eax, %eax
+	
+	movdqa	-15*16(%eax), %xmm0
+	movdqa	-14*16(%eax), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	paddd	-16*16(%eax), %xmm0
+	paddd	-15*16(%eax), %xmm4
+	paddd	1168(%ebp), %xmm4 #sha256d_4preext2_17
+	movdqa	%xmm0, %xmm3
+	movdqa	%xmm4, %xmm7
+	movdqa	%xmm3, 0*16(%eax)
+	movdqa	%xmm7, 1*16(%eax)
+	
+	sha256_sse2_extend_doubleround 2
+	sha256_sse2_extend_doubleround 4
+	
+	movdqa	-9*16(%eax), %xmm0
+	movdqa	1184(%ebp), %xmm4 #sha256d_4preext2_23
+	movdqa	%xmm0, %xmm2
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	-10*16(%eax), %xmm0
+	paddd	-9*16(%eax), %xmm4
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	-1*16(%eax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	paddd	0*16(%eax), %xmm4
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 6*16(%eax)
+	movdqa	%xmm7, 7*16(%eax)
+	
+	movdqa	1200(%ebp), %xmm0 #sha256d_4preext2_24
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	1*16(%eax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	2*16(%eax), %xmm7
+	movdqa	%xmm3, 8*16(%eax)
+	movdqa	%xmm7, 9*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	3*16(%eax), %xmm3
+	paddd	4*16(%eax), %xmm7
+	movdqa	%xmm3, 10*16(%eax)
+	movdqa	%xmm7, 11*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	5*16(%eax), %xmm3
+	paddd	6*16(%eax), %xmm7
+	movdqa	%xmm3, 12*16(%eax)
+	movdqa	%xmm7, 13*16(%eax)
+	
+	movdqa	1216(%ebp), %xmm0 #sha256d_4preext2_30
+	movdqa	0*16(%eax), %xmm4
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm6
+	psrld	$4, %xmm5
+	pxor	%xmm5, %xmm4
+	pxor	%xmm6, %xmm4
+	psrld	$11, %xmm5
+	pslld	$11, %xmm6
+	pxor	%xmm5, %xmm4
+	pxor	%xmm6, %xmm4
+	paddd	-1*16(%eax), %xmm4
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	7*16(%eax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	paddd	8*16(%eax), %xmm4
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 14*16(%eax)
+	movdqa	%xmm7, 15*16(%eax)
+	
+	jmp sha256d_ms_4way_extend_loop2
+	
+sha256d_ms_4way_extend_coda2:
+	sha256_sse2_extend_round 44
+	
+	movdqa	0(%ebp), %xmm7   #sha256_4h
+	movdqa	16(%ebp), %xmm5  #sha256_4h
+	movdqa	32(%ebp), %xmm4  #sha256_4h
+	movdqa	48(%ebp), %xmm3  #sha256_4h
+	movdqa	64(%ebp), %xmm0  #sha256_4h
+	movdqa	80(%ebp), %xmm1  #sha256_4h
+	movdqa	96(%ebp), %xmm2  #sha256_4h
+	movdqa	112(%ebp), %xmm6 #sha256_4h
+	movdqa	%xmm1, 0(%esp)
+	movdqa	%xmm2, 16(%esp)
+	movdqa	%xmm6, 32(%esp)
+	
+	leal	48(%esp), %eax
+	jmp sha256d_ms_4way_main_loop2
+
+.macro sha256_sse2_main_round_red i, r7
+	movdqa	16*(\i)(%eax), %xmm6
+	paddd	(128+16*(\i))(%ebp), %xmm6 #sha256_4k
+	paddd	32(%esp), %xmm6
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%esp), %xmm2
+	paddd	\r7, %xmm6
+	pandn	%xmm2, %xmm1
+	movdqa	%xmm2, 32(%esp)
+	movdqa	0(%esp), %xmm2
+	movdqa	%xmm2, 16(%esp)
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%esp)
+	paddd	%xmm1, %xmm6
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm1, %xmm0
+	paddd	%xmm6, %xmm0
+.endm
+
+sha256d_ms_4way_finish:
+	sha256_sse2_main_round_red 57, %xmm3
+	sha256_sse2_main_round_red 58, %xmm4
+	sha256_sse2_main_round_red 59, %xmm5
+	sha256_sse2_main_round_red 60, %xmm7
+	
+	paddd	112(%ebp), %xmm0  #sha256_4h
+	movdqa	%xmm0, 112(%edi)
+	
+    movl    %ebp, %esp
+    addl    $1356, %esp
+    popl    %ebp
+	movl	%ebp, %esp
+	popl	%ebp
+	popl	%esi
+	popl	%edi
+	ret
+
+
+	.text
+	.p2align 5
+	.globl sha256_use_4way
+	.globl _sha256_use_4way
+sha256_use_4way:
+_sha256_use_4way:
+	pushl	%ebx
+	
+	/* Check for SSE2 availability */
+	movl	$1, %eax
+	cpuid
+	andl	$0x04000000, %edx
+	jnz sha256_use_4way_sse2
+	xorl	%eax, %eax
+	popl	%ebx
+	ret
+	
+sha256_use_4way_sse2:
+	movl	$1, %eax
+	popl	%ebx
+	ret
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86.S b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86.S
new file mode 100644
index 0000000..e2eb112
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2-x86.S
@@ -0,0 +1,1193 @@
+/*
+ * Copyright 2012 pooler@litecoinpool.org
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+
+#if defined(__linux__) && defined(__ELF__)
+	.section .note.GNU-stack,"",%progbits
+#endif
+
+#if defined(USE_ASM) && defined(__i386__)
+
+	.data
+	.p2align 7
+sha256_4h:
+	.long 0x6a09e667, 0x6a09e667, 0x6a09e667, 0x6a09e667
+	.long 0xbb67ae85, 0xbb67ae85, 0xbb67ae85, 0xbb67ae85
+	.long 0x3c6ef372, 0x3c6ef372, 0x3c6ef372, 0x3c6ef372
+	.long 0xa54ff53a, 0xa54ff53a, 0xa54ff53a, 0xa54ff53a
+	.long 0x510e527f, 0x510e527f, 0x510e527f, 0x510e527f
+	.long 0x9b05688c, 0x9b05688c, 0x9b05688c, 0x9b05688c
+	.long 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab, 0x1f83d9ab
+	.long 0x5be0cd19, 0x5be0cd19, 0x5be0cd19, 0x5be0cd19
+
+	.data
+	.p2align 7
+sha256_4k:
+	.long 0x428a2f98, 0x428a2f98, 0x428a2f98, 0x428a2f98
+	.long 0x71374491, 0x71374491, 0x71374491, 0x71374491
+	.long 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf, 0xb5c0fbcf
+	.long 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5, 0xe9b5dba5
+	.long 0x3956c25b, 0x3956c25b, 0x3956c25b, 0x3956c25b
+	.long 0x59f111f1, 0x59f111f1, 0x59f111f1, 0x59f111f1
+	.long 0x923f82a4, 0x923f82a4, 0x923f82a4, 0x923f82a4
+	.long 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5, 0xab1c5ed5
+	.long 0xd807aa98, 0xd807aa98, 0xd807aa98, 0xd807aa98
+	.long 0x12835b01, 0x12835b01, 0x12835b01, 0x12835b01
+	.long 0x243185be, 0x243185be, 0x243185be, 0x243185be
+	.long 0x550c7dc3, 0x550c7dc3, 0x550c7dc3, 0x550c7dc3
+	.long 0x72be5d74, 0x72be5d74, 0x72be5d74, 0x72be5d74
+	.long 0x80deb1fe, 0x80deb1fe, 0x80deb1fe, 0x80deb1fe
+	.long 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7, 0x9bdc06a7
+	.long 0xc19bf174, 0xc19bf174, 0xc19bf174, 0xc19bf174
+	.long 0xe49b69c1, 0xe49b69c1, 0xe49b69c1, 0xe49b69c1
+	.long 0xefbe4786, 0xefbe4786, 0xefbe4786, 0xefbe4786
+	.long 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6, 0x0fc19dc6
+	.long 0x240ca1cc, 0x240ca1cc, 0x240ca1cc, 0x240ca1cc
+	.long 0x2de92c6f, 0x2de92c6f, 0x2de92c6f, 0x2de92c6f
+	.long 0x4a7484aa, 0x4a7484aa, 0x4a7484aa, 0x4a7484aa
+	.long 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc, 0x5cb0a9dc
+	.long 0x76f988da, 0x76f988da, 0x76f988da, 0x76f988da
+	.long 0x983e5152, 0x983e5152, 0x983e5152, 0x983e5152
+	.long 0xa831c66d, 0xa831c66d, 0xa831c66d, 0xa831c66d
+	.long 0xb00327c8, 0xb00327c8, 0xb00327c8, 0xb00327c8
+	.long 0xbf597fc7, 0xbf597fc7, 0xbf597fc7, 0xbf597fc7
+	.long 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3, 0xc6e00bf3
+	.long 0xd5a79147, 0xd5a79147, 0xd5a79147, 0xd5a79147
+	.long 0x06ca6351, 0x06ca6351, 0x06ca6351, 0x06ca6351
+	.long 0x14292967, 0x14292967, 0x14292967, 0x14292967
+	.long 0x27b70a85, 0x27b70a85, 0x27b70a85, 0x27b70a85
+	.long 0x2e1b2138, 0x2e1b2138, 0x2e1b2138, 0x2e1b2138
+	.long 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc, 0x4d2c6dfc
+	.long 0x53380d13, 0x53380d13, 0x53380d13, 0x53380d13
+	.long 0x650a7354, 0x650a7354, 0x650a7354, 0x650a7354
+	.long 0x766a0abb, 0x766a0abb, 0x766a0abb, 0x766a0abb
+	.long 0x81c2c92e, 0x81c2c92e, 0x81c2c92e, 0x81c2c92e
+	.long 0x92722c85, 0x92722c85, 0x92722c85, 0x92722c85
+	.long 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1, 0xa2bfe8a1
+	.long 0xa81a664b, 0xa81a664b, 0xa81a664b, 0xa81a664b
+	.long 0xc24b8b70, 0xc24b8b70, 0xc24b8b70, 0xc24b8b70
+	.long 0xc76c51a3, 0xc76c51a3, 0xc76c51a3, 0xc76c51a3
+	.long 0xd192e819, 0xd192e819, 0xd192e819, 0xd192e819
+	.long 0xd6990624, 0xd6990624, 0xd6990624, 0xd6990624
+	.long 0xf40e3585, 0xf40e3585, 0xf40e3585, 0xf40e3585
+	.long 0x106aa070, 0x106aa070, 0x106aa070, 0x106aa070
+	.long 0x19a4c116, 0x19a4c116, 0x19a4c116, 0x19a4c116
+	.long 0x1e376c08, 0x1e376c08, 0x1e376c08, 0x1e376c08
+	.long 0x2748774c, 0x2748774c, 0x2748774c, 0x2748774c
+	.long 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5, 0x34b0bcb5
+	.long 0x391c0cb3, 0x391c0cb3, 0x391c0cb3, 0x391c0cb3
+	.long 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a, 0x4ed8aa4a
+	.long 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f, 0x5b9cca4f
+	.long 0x682e6ff3, 0x682e6ff3, 0x682e6ff3, 0x682e6ff3
+	.long 0x748f82ee, 0x748f82ee, 0x748f82ee, 0x748f82ee
+	.long 0x78a5636f, 0x78a5636f, 0x78a5636f, 0x78a5636f
+	.long 0x84c87814, 0x84c87814, 0x84c87814, 0x84c87814
+	.long 0x8cc70208, 0x8cc70208, 0x8cc70208, 0x8cc70208
+	.long 0x90befffa, 0x90befffa, 0x90befffa, 0x90befffa
+	.long 0xa4506ceb, 0xa4506ceb, 0xa4506ceb, 0xa4506ceb
+	.long 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7, 0xbef9a3f7
+	.long 0xc67178f2, 0xc67178f2, 0xc67178f2, 0xc67178f2
+
+	.data
+	.p2align 6
+sha256d_4preext2_15:
+	.long 0x00000100, 0x00000100, 0x00000100, 0x00000100
+sha256d_4preext2_17:
+	.long 0x00a00000, 0x00a00000, 0x00a00000, 0x00a00000
+sha256d_4preext2_23:
+	.long 0x11002000, 0x11002000, 0x11002000, 0x11002000
+sha256d_4preext2_24:
+	.long 0x80000000, 0x80000000, 0x80000000, 0x80000000
+sha256d_4preext2_30:
+	.long 0x00400022, 0x00400022, 0x00400022, 0x00400022
+
+
+	.text
+	.p2align 5
+	.globl sha256_init_4way
+	.globl _sha256_init_4way
+sha256_init_4way:
+_sha256_init_4way:
+	movl	4(%esp), %edx
+	movdqa	sha256_4h+0, %xmm0
+	movdqa	sha256_4h+16, %xmm1
+	movdqa	sha256_4h+32, %xmm2
+	movdqa	sha256_4h+48, %xmm3
+	movdqu	%xmm0, 0(%edx)
+	movdqu	%xmm1, 16(%edx)
+	movdqu	%xmm2, 32(%edx)
+	movdqu	%xmm3, 48(%edx)
+	movdqa	sha256_4h+64, %xmm0
+	movdqa	sha256_4h+80, %xmm1
+	movdqa	sha256_4h+96, %xmm2
+	movdqa	sha256_4h+112, %xmm3
+	movdqu	%xmm0, 64(%edx)
+	movdqu	%xmm1, 80(%edx)
+	movdqu	%xmm2, 96(%edx)
+	movdqu	%xmm3, 112(%edx)
+	ret
+
+
+.macro sha256_sse2_extend_round i
+	movdqa	(\i-15)*16(%eax), %xmm0
+	movdqa	%xmm0, %xmm2
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	(\i-16)*16(%eax), %xmm0
+	paddd	(\i-7)*16(%eax), %xmm0
+
+	movdqa	%xmm3, %xmm2
+	psrld	$10, %xmm3
+	pslld	$13, %xmm2
+	movdqa	%xmm3, %xmm1
+	psrld	$7, %xmm1
+	pxor	%xmm1, %xmm3
+	pxor	%xmm2, %xmm3
+	psrld	$2, %xmm1
+	pslld	$2, %xmm2
+	pxor	%xmm1, %xmm3
+	pxor	%xmm2, %xmm3
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, \i*16(%eax)
+.endm
+
+.macro sha256_sse2_extend_doubleround i
+	movdqa	(\i-15)*16(%eax), %xmm0
+	movdqa	(\i-14)*16(%eax), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+
+	paddd	(\i-16)*16(%eax), %xmm0
+	paddd	(\i-15)*16(%eax), %xmm4
+
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+
+	paddd	(\i-7)*16(%eax), %xmm0
+	paddd	(\i-6)*16(%eax), %xmm4
+
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, \i*16(%eax)
+	movdqa	%xmm7, (\i+1)*16(%eax)
+.endm
+
+.macro sha256_sse2_main_round i
+	movdqa	16*(\i)(%eax), %xmm6
+
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%esp), %xmm2
+	pandn	%xmm2, %xmm1
+	paddd	32(%esp), %xmm6
+
+	movdqa	%xmm2, 32(%esp)
+	movdqa	0(%esp), %xmm2
+	movdqa	%xmm2, 16(%esp)
+
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%esp)
+
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	paddd	16*(\i)+sha256_4k, %xmm6
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm2, %xmm0
+	pxor	%xmm1, %xmm0
+	movdqa	%xmm5, %xmm1
+	paddd	%xmm0, %xmm6
+
+	movdqa	%xmm3, %xmm0
+	movdqa	%xmm4, %xmm3
+	movdqa	%xmm4, %xmm2
+	paddd	%xmm6, %xmm0
+	pand	%xmm5, %xmm2
+	pand	%xmm7, %xmm1
+	pand	%xmm7, %xmm4
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm5, %xmm4
+	movdqa	%xmm7, %xmm5
+	pxor	%xmm2, %xmm1
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm7, %xmm2
+	psrld	$2, %xmm7
+	movdqa	%xmm7, %xmm1
+	pslld	$10, %xmm2
+	psrld	$11, %xmm1
+	pxor	%xmm2, %xmm7
+	pslld	$9, %xmm2
+	pxor	%xmm1, %xmm7
+	psrld	$9, %xmm1
+	pxor	%xmm2, %xmm7
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm7
+	pxor	%xmm2, %xmm7
+	paddd	%xmm6, %xmm7
+.endm
+
+.macro sha256_sse2_main_quadround i
+	sha256_sse2_main_round \i+0
+	sha256_sse2_main_round \i+1
+	sha256_sse2_main_round \i+2
+	sha256_sse2_main_round \i+3
+.endm
+
+
+.macro p2bswap_esi_esp i
+	movdqu	\i*16(%esi), %xmm0
+	movdqu	(\i+1)*16(%esi), %xmm2
+	pshuflw	$0xb1, %xmm0, %xmm0
+	pshuflw	$0xb1, %xmm2, %xmm2
+	pshufhw	$0xb1, %xmm0, %xmm0
+	pshufhw	$0xb1, %xmm2, %xmm2
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm2, %xmm3
+	psrlw	$8, %xmm1
+	psrlw	$8, %xmm3
+	psllw	$8, %xmm0
+	psllw	$8, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm3, %xmm2
+	movdqa	%xmm0, (\i+3)*16(%esp)
+	movdqa	%xmm2, (\i+4)*16(%esp)
+.endm
+
+	.text
+	.p2align 5
+	.globl sha256_transform_4way
+	.globl _sha256_transform_4way
+sha256_transform_4way:
+_sha256_transform_4way:
+	pushl	%edi
+	pushl	%esi
+	movl	12(%esp), %edi
+	movl	16(%esp), %esi
+	movl	20(%esp), %ecx
+	movl	%esp, %edx
+	subl	$67*16, %esp
+	andl	$-128, %esp
+	
+	testl	%ecx, %ecx
+	jnz sha256_transform_4way_swap
+	
+	movdqu	0*16(%esi), %xmm0
+	movdqu	1*16(%esi), %xmm1
+	movdqu	2*16(%esi), %xmm2
+	movdqu	3*16(%esi), %xmm3
+	movdqu	4*16(%esi), %xmm4
+	movdqu	5*16(%esi), %xmm5
+	movdqu	6*16(%esi), %xmm6
+	movdqu	7*16(%esi), %xmm7
+	movdqa	%xmm0, 3*16(%esp)
+	movdqa	%xmm1, 4*16(%esp)
+	movdqa	%xmm2, 5*16(%esp)
+	movdqa	%xmm3, 6*16(%esp)
+	movdqa	%xmm4, 7*16(%esp)
+	movdqa	%xmm5, 8*16(%esp)
+	movdqa	%xmm6, 9*16(%esp)
+	movdqa	%xmm7, 10*16(%esp)
+	movdqu	8*16(%esi), %xmm0
+	movdqu	9*16(%esi), %xmm1
+	movdqu	10*16(%esi), %xmm2
+	movdqu	11*16(%esi), %xmm3
+	movdqu	12*16(%esi), %xmm4
+	movdqu	13*16(%esi), %xmm5
+	movdqu	14*16(%esi), %xmm6
+	movdqu	15*16(%esi), %xmm7
+	movdqa	%xmm0, 11*16(%esp)
+	movdqa	%xmm1, 12*16(%esp)
+	movdqa	%xmm2, 13*16(%esp)
+	movdqa	%xmm3, 14*16(%esp)
+	movdqa	%xmm4, 15*16(%esp)
+	movdqa	%xmm5, 16*16(%esp)
+	movdqa	%xmm6, 17*16(%esp)
+	movdqa	%xmm7, 18*16(%esp)
+	jmp sha256_transform_4way_extend
+	
+	.p2align 5
+sha256_transform_4way_swap:
+	p2bswap_esi_esp 0
+	p2bswap_esi_esp 2
+	p2bswap_esi_esp 4
+	p2bswap_esi_esp 6
+	p2bswap_esi_esp 8
+	p2bswap_esi_esp 10
+	p2bswap_esi_esp 12
+	p2bswap_esi_esp 14
+	
+sha256_transform_4way_extend:
+	leal	19*16(%esp), %ecx
+	leal	48*16(%ecx), %eax
+	movdqa	-2*16(%ecx), %xmm3
+	movdqa	-1*16(%ecx), %xmm7
+sha256_transform_4way_extend_loop:
+	movdqa	-15*16(%ecx), %xmm0
+	movdqa	-14*16(%ecx), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+
+	paddd	-16*16(%ecx), %xmm0
+	paddd	-15*16(%ecx), %xmm4
+
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+
+	paddd	-7*16(%ecx), %xmm0
+	paddd	-6*16(%ecx), %xmm4
+
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, (%ecx)
+	movdqa	%xmm7, 16(%ecx)
+	addl	$2*16, %ecx
+	cmpl	%ecx, %eax
+	jne sha256_transform_4way_extend_loop
+	
+	movdqu	0(%edi), %xmm7
+	movdqu	16(%edi), %xmm5
+	movdqu	32(%edi), %xmm4
+	movdqu	48(%edi), %xmm3
+	movdqu	64(%edi), %xmm0
+	movdqu	80(%edi), %xmm1
+	movdqu	96(%edi), %xmm2
+	movdqu	112(%edi), %xmm6
+	movdqa	%xmm1, 0(%esp)
+	movdqa	%xmm2, 16(%esp)
+	movdqa	%xmm6, 32(%esp)
+	
+	xorl	%eax, %eax
+sha256_transform_4way_main_loop:
+	movdqa	3*16(%esp, %eax), %xmm6
+	paddd	sha256_4k(%eax), %xmm6
+	paddd	32(%esp), %xmm6
+
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%esp), %xmm2
+	pandn	%xmm2, %xmm1
+
+	movdqa	%xmm2, 32(%esp)
+	movdqa	0(%esp), %xmm2
+	movdqa	%xmm2, 16(%esp)
+
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%esp)
+
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm1, %xmm0
+	paddd	%xmm0, %xmm6
+
+	movdqa	%xmm3, %xmm0
+	paddd	%xmm6, %xmm0
+
+	movdqa	%xmm5, %xmm1
+	movdqa	%xmm4, %xmm3
+	movdqa	%xmm4, %xmm2
+	pand	%xmm5, %xmm2
+	pand	%xmm7, %xmm4
+	pand	%xmm7, %xmm1
+	pxor	%xmm4, %xmm1
+	movdqa	%xmm5, %xmm4
+	movdqa	%xmm7, %xmm5
+	pxor	%xmm2, %xmm1
+	paddd	%xmm1, %xmm6
+
+	movdqa	%xmm7, %xmm2
+	psrld	$2, %xmm7
+	movdqa	%xmm7, %xmm1
+	pslld	$10, %xmm2
+	psrld	$11, %xmm1
+	pxor	%xmm2, %xmm7
+	pxor	%xmm1, %xmm7
+	pslld	$9, %xmm2
+	psrld	$9, %xmm1
+	pxor	%xmm2, %xmm7
+	pxor	%xmm1, %xmm7
+	pslld	$11, %xmm2
+	pxor	%xmm2, %xmm7
+	paddd	%xmm6, %xmm7
+	
+	addl	$16, %eax
+	cmpl	$16*64, %eax
+	jne sha256_transform_4way_main_loop
+	
+	movdqu	0(%edi), %xmm1
+	movdqu	16(%edi), %xmm2
+	paddd	%xmm1, %xmm7
+	paddd	%xmm2, %xmm5
+	movdqu	32(%edi), %xmm1
+	movdqu	48(%edi), %xmm2
+	paddd	%xmm1, %xmm4
+	paddd	%xmm2, %xmm3
+	
+	movdqu	%xmm7, 0(%edi)
+	movdqu	%xmm5, 16(%edi)
+	movdqu	%xmm4, 32(%edi)
+	movdqu	%xmm3, 48(%edi)
+	
+	movdqu	64(%edi), %xmm1
+	movdqu	80(%edi), %xmm2
+	movdqu	96(%edi), %xmm6
+	movdqu	112(%edi), %xmm7
+	paddd	%xmm1, %xmm0
+	paddd	0(%esp), %xmm2
+	paddd	16(%esp), %xmm6
+	paddd	32(%esp), %xmm7
+	
+	movdqu	%xmm0, 64(%edi)
+	movdqu	%xmm2, 80(%edi)
+	movdqu	%xmm6, 96(%edi)
+	movdqu	%xmm7, 112(%edi)
+	
+	movl	%edx, %esp
+	popl	%esi
+	popl	%edi
+	ret
+
+
+	.text
+	.p2align 5
+	.globl sha256d_ms_4way
+	.globl _sha256d_ms_4way
+sha256d_ms_4way:
+_sha256d_ms_4way:
+	pushl	%edi
+	pushl	%esi
+	pushl	%ebp
+	movl	16(%esp), %edi
+	movl	20(%esp), %esi
+	movl	24(%esp), %edx
+	movl	28(%esp), %ecx
+	movl	%esp, %ebp
+	subl	$67*16, %esp
+	andl	$-128, %esp
+	
+	leal	256(%esi), %eax
+	
+sha256d_ms_4way_extend_loop1:
+	movdqa	3*16(%esi), %xmm0
+	movdqa	2*16(%eax), %xmm3
+	movdqa	3*16(%eax), %xmm7
+	movdqa	%xmm3, 5*16(%esp)
+	movdqa	%xmm7, 6*16(%esp)
+	movdqa	%xmm0, %xmm2
+	paddd	%xmm0, %xmm7
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, 2*16(%eax)
+	movdqa	%xmm7, 3*16(%eax)
+	
+	movdqa	4*16(%eax), %xmm0
+	movdqa	%xmm0, 7*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	movdqa	%xmm3, 4*16(%eax)
+	movdqa	%xmm7, 5*16(%eax)
+	
+	movdqa	6*16(%eax), %xmm0
+	movdqa	7*16(%eax), %xmm4
+	movdqa	%xmm0, 9*16(%esp)
+	movdqa	%xmm4, 10*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 6*16(%eax)
+	movdqa	%xmm7, 7*16(%eax)
+	
+	movdqa	8*16(%eax), %xmm0
+	movdqa	2*16(%eax), %xmm4
+	movdqa	%xmm0, 11*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 8*16(%eax)
+	movdqa	%xmm7, 9*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	3*16(%eax), %xmm3
+	paddd	4*16(%eax), %xmm7
+	movdqa	%xmm3, 10*16(%eax)
+	movdqa	%xmm7, 11*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	5*16(%eax), %xmm3
+	paddd	6*16(%eax), %xmm7
+	movdqa	%xmm3, 12*16(%eax)
+	movdqa	%xmm7, 13*16(%eax)
+	
+	movdqa	14*16(%eax), %xmm0
+	movdqa	15*16(%eax), %xmm4
+	movdqa	%xmm0, 17*16(%esp)
+	movdqa	%xmm4, 18*16(%esp)
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	7*16(%eax), %xmm0
+	paddd	8*16(%eax), %xmm4
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 14*16(%eax)
+	movdqa	%xmm7, 15*16(%eax)
+	
+sha256d_ms_4way_extend_loop2:
+	sha256_sse2_extend_doubleround 16
+	sha256_sse2_extend_doubleround 18
+	sha256_sse2_extend_doubleround 20
+	sha256_sse2_extend_doubleround 22
+	sha256_sse2_extend_doubleround 24
+	sha256_sse2_extend_doubleround 26
+	sha256_sse2_extend_doubleround 28
+	sha256_sse2_extend_doubleround 30
+	sha256_sse2_extend_doubleround 32
+	sha256_sse2_extend_doubleround 34
+	sha256_sse2_extend_doubleround 36
+	sha256_sse2_extend_doubleround 38
+	sha256_sse2_extend_doubleround 40
+	sha256_sse2_extend_doubleround 42
+	jz sha256d_ms_4way_extend_coda2
+	sha256_sse2_extend_doubleround 44
+	sha256_sse2_extend_doubleround 46
+	
+	movdqa	0(%ecx), %xmm3
+	movdqa	16(%ecx), %xmm0
+	movdqa	32(%ecx), %xmm1
+	movdqa	48(%ecx), %xmm2
+	movdqa	64(%ecx), %xmm6
+	movdqa	80(%ecx), %xmm7
+	movdqa	96(%ecx), %xmm5
+	movdqa	112(%ecx), %xmm4
+	movdqa	%xmm1, 0(%esp)
+	movdqa	%xmm2, 16(%esp)
+	movdqa	%xmm6, 32(%esp)
+	
+	movl	%esi, %eax
+	jmp sha256d_ms_4way_main_loop1
+	
+sha256d_ms_4way_main_loop2:
+	sha256_sse2_main_round 0
+	sha256_sse2_main_round 1
+	sha256_sse2_main_round 2
+sha256d_ms_4way_main_loop1:
+	sha256_sse2_main_round 3
+	sha256_sse2_main_quadround 4
+	sha256_sse2_main_quadround 8
+	sha256_sse2_main_quadround 12
+	sha256_sse2_main_quadround 16
+	sha256_sse2_main_quadround 20
+	sha256_sse2_main_quadround 24
+	sha256_sse2_main_quadround 28
+	sha256_sse2_main_quadround 32
+	sha256_sse2_main_quadround 36
+	sha256_sse2_main_quadround 40
+	sha256_sse2_main_quadround 44
+	sha256_sse2_main_quadround 48
+	sha256_sse2_main_quadround 52
+	sha256_sse2_main_round 56
+	jz sha256d_ms_4way_finish
+	sha256_sse2_main_round 57
+	sha256_sse2_main_round 58
+	sha256_sse2_main_round 59
+	sha256_sse2_main_quadround 60
+	
+	movdqa	5*16(%esp), %xmm1
+	movdqa	6*16(%esp), %xmm2
+	movdqa	7*16(%esp), %xmm6
+	movdqa	%xmm1, 18*16(%esi)
+	movdqa	%xmm2, 19*16(%esi)
+	movdqa	%xmm6, 20*16(%esi)
+	movdqa	9*16(%esp), %xmm1
+	movdqa	10*16(%esp), %xmm2
+	movdqa	11*16(%esp), %xmm6
+	movdqa	%xmm1, 22*16(%esi)
+	movdqa	%xmm2, 23*16(%esi)
+	movdqa	%xmm6, 24*16(%esi)
+	movdqa	17*16(%esp), %xmm1
+	movdqa	18*16(%esp), %xmm2
+	movdqa	%xmm1, 30*16(%esi)
+	movdqa	%xmm2, 31*16(%esi)
+	
+	movdqa	0(%esp), %xmm1
+	movdqa	16(%esp), %xmm2
+	movdqa	32(%esp), %xmm6
+	paddd	0(%edx), %xmm7
+	paddd	16(%edx), %xmm5
+	paddd	32(%edx), %xmm4
+	paddd	48(%edx), %xmm3
+	paddd	64(%edx), %xmm0
+	paddd	80(%edx), %xmm1
+	paddd	96(%edx), %xmm2
+	paddd	112(%edx), %xmm6
+	
+	movdqa	%xmm7, 48+0(%esp)
+	movdqa	%xmm5, 48+16(%esp)
+	movdqa	%xmm4, 48+32(%esp)
+	movdqa	%xmm3, 48+48(%esp)
+	movdqa	%xmm0, 48+64(%esp)
+	movdqa	%xmm1, 48+80(%esp)
+	movdqa	%xmm2, 48+96(%esp)
+	movdqa	%xmm6, 48+112(%esp)
+	
+	movdqa	sha256d_4preext2_15, %xmm1
+	movdqa	sha256d_4preext2_24, %xmm2
+	pxor	%xmm0, %xmm0
+	movdqa	%xmm2, 48+128(%esp)
+	movdqa	%xmm0, 48+144(%esp)
+	movdqa	%xmm0, 48+160(%esp)
+	movdqa	%xmm0, 48+176(%esp)
+	movdqa	%xmm0, 48+192(%esp)
+	movdqa	%xmm0, 48+208(%esp)
+	movdqa	%xmm0, 48+224(%esp)
+	movdqa	%xmm1, 48+240(%esp)
+	
+	leal	19*16(%esp), %eax
+	cmpl	%eax, %eax
+	
+	movdqa	-15*16(%eax), %xmm0
+	movdqa	-14*16(%eax), %xmm4
+	movdqa	%xmm0, %xmm2
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm0
+	psrld	$3, %xmm4
+	movdqa	%xmm0, %xmm1
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm2
+	pslld	$14, %xmm6
+	psrld	$4, %xmm1
+	psrld	$4, %xmm5
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	psrld	$11, %xmm1
+	psrld	$11, %xmm5
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	pslld	$11, %xmm2
+	pslld	$11, %xmm6
+	pxor	%xmm1, %xmm0
+	pxor	%xmm5, %xmm4
+	pxor	%xmm2, %xmm0
+	pxor	%xmm6, %xmm4
+	paddd	-16*16(%eax), %xmm0
+	paddd	-15*16(%eax), %xmm4
+	paddd	sha256d_4preext2_17, %xmm4
+	movdqa	%xmm0, %xmm3
+	movdqa	%xmm4, %xmm7
+	movdqa	%xmm3, 0*16(%eax)
+	movdqa	%xmm7, 1*16(%eax)
+	
+	sha256_sse2_extend_doubleround 2
+	sha256_sse2_extend_doubleround 4
+	
+	movdqa	-9*16(%eax), %xmm0
+	movdqa	sha256d_4preext2_23, %xmm4
+	movdqa	%xmm0, %xmm2
+	psrld	$3, %xmm0
+	movdqa	%xmm0, %xmm1
+	pslld	$14, %xmm2
+	psrld	$4, %xmm1
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	psrld	$11, %xmm1
+	pslld	$11, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	paddd	-10*16(%eax), %xmm0
+	paddd	-9*16(%eax), %xmm4
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	-1*16(%eax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	paddd	0*16(%eax), %xmm4
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 6*16(%eax)
+	movdqa	%xmm7, 7*16(%eax)
+	
+	movdqa	sha256d_4preext2_24, %xmm0
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	1*16(%eax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	2*16(%eax), %xmm7
+	movdqa	%xmm3, 8*16(%eax)
+	movdqa	%xmm7, 9*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	3*16(%eax), %xmm3
+	paddd	4*16(%eax), %xmm7
+	movdqa	%xmm3, 10*16(%eax)
+	movdqa	%xmm7, 11*16(%eax)
+	
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	5*16(%eax), %xmm3
+	paddd	6*16(%eax), %xmm7
+	movdqa	%xmm3, 12*16(%eax)
+	movdqa	%xmm7, 13*16(%eax)
+	
+	movdqa	sha256d_4preext2_30, %xmm0
+	movdqa	0*16(%eax), %xmm4
+	movdqa	%xmm4, %xmm6
+	psrld	$3, %xmm4
+	movdqa	%xmm4, %xmm5
+	pslld	$14, %xmm6
+	psrld	$4, %xmm5
+	pxor	%xmm5, %xmm4
+	pxor	%xmm6, %xmm4
+	psrld	$11, %xmm5
+	pslld	$11, %xmm6
+	pxor	%xmm5, %xmm4
+	pxor	%xmm6, %xmm4
+	paddd	-1*16(%eax), %xmm4
+	movdqa	%xmm3, %xmm2
+	movdqa	%xmm7, %xmm6
+	psrld	$10, %xmm3
+	psrld	$10, %xmm7
+	movdqa	%xmm3, %xmm1
+	movdqa	%xmm7, %xmm5
+	paddd	7*16(%eax), %xmm0
+	pslld	$13, %xmm2
+	pslld	$13, %xmm6
+	psrld	$7, %xmm1
+	psrld	$7, %xmm5
+	paddd	8*16(%eax), %xmm4
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	psrld	$2, %xmm1
+	psrld	$2, %xmm5
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	pslld	$2, %xmm2
+	pslld	$2, %xmm6
+	pxor	%xmm1, %xmm3
+	pxor	%xmm5, %xmm7
+	pxor	%xmm2, %xmm3
+	pxor	%xmm6, %xmm7
+	paddd	%xmm0, %xmm3
+	paddd	%xmm4, %xmm7
+	movdqa	%xmm3, 14*16(%eax)
+	movdqa	%xmm7, 15*16(%eax)
+	
+	jmp sha256d_ms_4way_extend_loop2
+	
+sha256d_ms_4way_extend_coda2:
+	sha256_sse2_extend_round 44
+	
+	movdqa	sha256_4h+0, %xmm7
+	movdqa	sha256_4h+16, %xmm5
+	movdqa	sha256_4h+32, %xmm4
+	movdqa	sha256_4h+48, %xmm3
+	movdqa	sha256_4h+64, %xmm0
+	movdqa	sha256_4h+80, %xmm1
+	movdqa	sha256_4h+96, %xmm2
+	movdqa	sha256_4h+112, %xmm6
+	movdqa	%xmm1, 0(%esp)
+	movdqa	%xmm2, 16(%esp)
+	movdqa	%xmm6, 32(%esp)
+	
+	leal	48(%esp), %eax
+	jmp sha256d_ms_4way_main_loop2
+
+.macro sha256_sse2_main_round_red i, r7
+	movdqa	16*(\i)(%eax), %xmm6
+	paddd	16*(\i)+sha256_4k, %xmm6
+	paddd	32(%esp), %xmm6
+	movdqa	%xmm0, %xmm1
+	movdqa	16(%esp), %xmm2
+	paddd	\r7, %xmm6
+	pandn	%xmm2, %xmm1
+	movdqa	%xmm2, 32(%esp)
+	movdqa	0(%esp), %xmm2
+	movdqa	%xmm2, 16(%esp)
+	pand	%xmm0, %xmm2
+	pxor	%xmm2, %xmm1
+	movdqa	%xmm0, 0(%esp)
+	paddd	%xmm1, %xmm6
+	movdqa	%xmm0, %xmm1
+	psrld	$6, %xmm0
+	movdqa	%xmm0, %xmm2
+	pslld	$7, %xmm1
+	psrld	$5, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$14, %xmm1
+	psrld	$14, %xmm2
+	pxor	%xmm1, %xmm0
+	pxor	%xmm2, %xmm0
+	pslld	$5, %xmm1
+	pxor	%xmm1, %xmm0
+	paddd	%xmm6, %xmm0
+.endm
+
+sha256d_ms_4way_finish:
+	sha256_sse2_main_round_red 57, %xmm3
+	sha256_sse2_main_round_red 58, %xmm4
+	sha256_sse2_main_round_red 59, %xmm5
+	sha256_sse2_main_round_red 60, %xmm7
+	
+	paddd	sha256_4h+112, %xmm0
+	movdqa	%xmm0, 112(%edi)
+	
+	movl	%ebp, %esp
+	popl	%ebp
+	popl	%esi
+	popl	%edi
+	ret
+
+
+	.text
+	.p2align 5
+	.globl sha256_use_4way
+	.globl _sha256_use_4way
+sha256_use_4way:
+_sha256_use_4way:
+	pushl	%ebx
+	
+	/* Check for SSE2 availability */
+	movl	$1, %eax
+	cpuid
+	andl	$0x04000000, %edx
+	jnz sha256_use_4way_sse2
+	xorl	%eax, %eax
+	popl	%ebx
+	ret
+	
+sha256_use_4way_sse2:
+	movl	$1, %eax
+	popl	%ebx
+	ret
+
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2.c
new file mode 100644
index 0000000..9447abb
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha2.c
@@ -0,0 +1,633 @@
+/*
+ * Copyright 2011 ArtForz
+ * Copyright 2011-2013 pooler
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+
+#include "cpuminer-config.h"
+#include "miner.h"
+
+#include <string.h>
+#include <inttypes.h>
+
+#if defined(USE_ASM) && \
+	(defined(__x86_64__) || \
+	 (defined(__arm__) && defined(__APCS_32__)) || \
+	 (defined(__powerpc__) || defined(__ppc__) || defined(__PPC__)))
+#define EXTERN_SHA256
+#endif
+
+static const uint32_t sha256_h[8] = {
+	0x6a09e667, 0xbb67ae85, 0x3c6ef372, 0xa54ff53a,
+	0x510e527f, 0x9b05688c, 0x1f83d9ab, 0x5be0cd19
+};
+
+static const uint32_t sha256_k[64] = {
+	0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5,
+	0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
+	0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3,
+	0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
+	0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
+	0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
+	0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
+	0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
+	0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13,
+	0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
+	0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3,
+	0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
+	0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5,
+	0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
+	0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
+	0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
+};
+
+void sha256_init(uint32_t *state)
+{
+	memcpy(state, sha256_h, 32);
+}
+
+/* Elementary functions used by SHA256 */
+#define Ch(x, y, z)     ((x & (y ^ z)) ^ z)
+#define Maj(x, y, z)    ((x & (y | z)) | (y & z))
+#define ROTR(x, n)      ((x >> n) | (x << (32 - n)))
+#define S0(x)           (ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
+#define S1(x)           (ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
+#define s0(x)           (ROTR(x, 7) ^ ROTR(x, 18) ^ (x >> 3))
+#define s1(x)           (ROTR(x, 17) ^ ROTR(x, 19) ^ (x >> 10))
+
+/* SHA256 round function */
+#define RND(a, b, c, d, e, f, g, h, k) \
+	do { \
+		t0 = h + S1(e) + Ch(e, f, g) + k; \
+		t1 = S0(a) + Maj(a, b, c); \
+		d += t0; \
+		h  = t0 + t1; \
+	} while (0)
+
+/* Adjusted round function for rotating state */
+#define RNDr(S, W, i) \
+	RND(S[(64 - i) % 8], S[(65 - i) % 8], \
+	    S[(66 - i) % 8], S[(67 - i) % 8], \
+	    S[(68 - i) % 8], S[(69 - i) % 8], \
+	    S[(70 - i) % 8], S[(71 - i) % 8], \
+	    W[i] + sha256_k[i])
+
+#ifndef EXTERN_SHA256
+
+/*
+ * SHA256 block compression function.  The 256-bit state is transformed via
+ * the 512-bit input block to produce a new state.
+ */
+void sha256_transform(uint32_t *state, const uint32_t *block, int swap)
+{
+	uint32_t W[64];
+	uint32_t S[8];
+	uint32_t t0, t1;
+	int i;
+
+	/* 1. Prepare message schedule W. */
+	if (swap) {
+		for (i = 0; i < 16; i++)
+			W[i] = swab32(block[i]);
+	} else
+		memcpy(W, block, 64);
+	for (i = 16; i < 64; i += 2) {
+		W[i]   = s1(W[i - 2]) + W[i - 7] + s0(W[i - 15]) + W[i - 16];
+		W[i+1] = s1(W[i - 1]) + W[i - 6] + s0(W[i - 14]) + W[i - 15];
+	}
+
+	/* 2. Initialize working variables. */
+	memcpy(S, state, 32);
+
+	/* 3. Mix. */
+	RNDr(S, W,  0);
+	RNDr(S, W,  1);
+	RNDr(S, W,  2);
+	RNDr(S, W,  3);
+	RNDr(S, W,  4);
+	RNDr(S, W,  5);
+	RNDr(S, W,  6);
+	RNDr(S, W,  7);
+	RNDr(S, W,  8);
+	RNDr(S, W,  9);
+	RNDr(S, W, 10);
+	RNDr(S, W, 11);
+	RNDr(S, W, 12);
+	RNDr(S, W, 13);
+	RNDr(S, W, 14);
+	RNDr(S, W, 15);
+	RNDr(S, W, 16);
+	RNDr(S, W, 17);
+	RNDr(S, W, 18);
+	RNDr(S, W, 19);
+	RNDr(S, W, 20);
+	RNDr(S, W, 21);
+	RNDr(S, W, 22);
+	RNDr(S, W, 23);
+	RNDr(S, W, 24);
+	RNDr(S, W, 25);
+	RNDr(S, W, 26);
+	RNDr(S, W, 27);
+	RNDr(S, W, 28);
+	RNDr(S, W, 29);
+	RNDr(S, W, 30);
+	RNDr(S, W, 31);
+	RNDr(S, W, 32);
+	RNDr(S, W, 33);
+	RNDr(S, W, 34);
+	RNDr(S, W, 35);
+	RNDr(S, W, 36);
+	RNDr(S, W, 37);
+	RNDr(S, W, 38);
+	RNDr(S, W, 39);
+	RNDr(S, W, 40);
+	RNDr(S, W, 41);
+	RNDr(S, W, 42);
+	RNDr(S, W, 43);
+	RNDr(S, W, 44);
+	RNDr(S, W, 45);
+	RNDr(S, W, 46);
+	RNDr(S, W, 47);
+	RNDr(S, W, 48);
+	RNDr(S, W, 49);
+	RNDr(S, W, 50);
+	RNDr(S, W, 51);
+	RNDr(S, W, 52);
+	RNDr(S, W, 53);
+	RNDr(S, W, 54);
+	RNDr(S, W, 55);
+	RNDr(S, W, 56);
+	RNDr(S, W, 57);
+	RNDr(S, W, 58);
+	RNDr(S, W, 59);
+	RNDr(S, W, 60);
+	RNDr(S, W, 61);
+	RNDr(S, W, 62);
+	RNDr(S, W, 63);
+
+	/* 4. Mix local working variables into global state */
+	for (i = 0; i < 8; i++)
+		state[i] += S[i];
+}
+
+#endif /* EXTERN_SHA256 */
+
+
+static const uint32_t sha256d_hash1[16] = {
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000000,
+	0x80000000, 0x00000000, 0x00000000, 0x00000000,
+	0x00000000, 0x00000000, 0x00000000, 0x00000100
+};
+
+static void sha256d_80_swap(uint32_t *hash, const uint32_t *data)
+{
+	uint32_t S[16];
+	int i;
+
+	sha256_init(S);
+	sha256_transform(S, data, 0);
+	sha256_transform(S, data + 16, 0);
+	memcpy(S + 8, sha256d_hash1 + 8, 32);
+	sha256_init(hash);
+	sha256_transform(hash, S, 0);
+	for (i = 0; i < 8; i++)
+		hash[i] = swab32(hash[i]);
+}
+
+void sha256d(unsigned char *hash, const unsigned char *data, int len)
+{
+	uint32_t S[16], T[16];
+	int i, r;
+
+	sha256_init(S);
+	for (r = len; r > -9; r -= 64) {
+		if (r < 64)
+			memset(T, 0, 64);
+		memcpy(T, data + len - r, r > 64 ? 64 : (r < 0 ? 0 : r));
+		if (r >= 0 && r < 64)
+			((unsigned char *)T)[r] = 0x80;
+		for (i = 0; i < 16; i++)
+			T[i] = be32dec(T + i);
+		if (r < 56)
+			T[15] = 8 * len;
+		sha256_transform(S, T, 0);
+	}
+	memcpy(S + 8, sha256d_hash1 + 8, 32);
+	sha256_init(T);
+	sha256_transform(T, S, 0);
+	for (i = 0; i < 8; i++)
+		be32enc((uint32_t *)hash + i, T[i]);
+}
+
+static inline void sha256d_preextend(uint32_t *W)
+{
+	W[16] = s1(W[14]) + W[ 9] + s0(W[ 1]) + W[ 0];
+	W[17] = s1(W[15]) + W[10] + s0(W[ 2]) + W[ 1];
+	W[18] = s1(W[16]) + W[11]             + W[ 2];
+	W[19] = s1(W[17]) + W[12] + s0(W[ 4]);
+	W[20] =             W[13] + s0(W[ 5]) + W[ 4];
+	W[21] =             W[14] + s0(W[ 6]) + W[ 5];
+	W[22] =             W[15] + s0(W[ 7]) + W[ 6];
+	W[23] =             W[16] + s0(W[ 8]) + W[ 7];
+	W[24] =             W[17] + s0(W[ 9]) + W[ 8];
+	W[25] =                     s0(W[10]) + W[ 9];
+	W[26] =                     s0(W[11]) + W[10];
+	W[27] =                     s0(W[12]) + W[11];
+	W[28] =                     s0(W[13]) + W[12];
+	W[29] =                     s0(W[14]) + W[13];
+	W[30] =                     s0(W[15]) + W[14];
+	W[31] =                     s0(W[16]) + W[15];
+}
+
+static inline void sha256d_prehash(uint32_t *S, const uint32_t *W)
+{
+	uint32_t t0, t1;
+	RNDr(S, W, 0);
+	RNDr(S, W, 1);
+	RNDr(S, W, 2);
+}
+
+#ifdef EXTERN_SHA256
+
+void sha256d_ms(uint32_t *hash, uint32_t *W,
+	const uint32_t *midstate, const uint32_t *prehash);
+
+#else
+
+static inline void sha256d_ms(uint32_t *hash, uint32_t *W,
+	const uint32_t *midstate, const uint32_t *prehash)
+{
+	uint32_t S[64];
+	uint32_t t0, t1;
+	int i;
+
+	S[18] = W[18];
+	S[19] = W[19];
+	S[20] = W[20];
+	S[22] = W[22];
+	S[23] = W[23];
+	S[24] = W[24];
+	S[30] = W[30];
+	S[31] = W[31];
+
+	W[18] += s0(W[3]);
+	W[19] += W[3];
+	W[20] += s1(W[18]);
+	W[21]  = s1(W[19]);
+	W[22] += s1(W[20]);
+	W[23] += s1(W[21]);
+	W[24] += s1(W[22]);
+	W[25]  = s1(W[23]) + W[18];
+	W[26]  = s1(W[24]) + W[19];
+	W[27]  = s1(W[25]) + W[20];
+	W[28]  = s1(W[26]) + W[21];
+	W[29]  = s1(W[27]) + W[22];
+	W[30] += s1(W[28]) + W[23];
+	W[31] += s1(W[29]) + W[24];
+	for (i = 32; i < 64; i += 2) {
+		W[i]   = s1(W[i - 2]) + W[i - 7] + s0(W[i - 15]) + W[i - 16];
+		W[i+1] = s1(W[i - 1]) + W[i - 6] + s0(W[i - 14]) + W[i - 15];
+	}
+
+	memcpy(S, prehash, 32);
+
+	RNDr(S, W,  3);
+	RNDr(S, W,  4);
+	RNDr(S, W,  5);
+	RNDr(S, W,  6);
+	RNDr(S, W,  7);
+	RNDr(S, W,  8);
+	RNDr(S, W,  9);
+	RNDr(S, W, 10);
+	RNDr(S, W, 11);
+	RNDr(S, W, 12);
+	RNDr(S, W, 13);
+	RNDr(S, W, 14);
+	RNDr(S, W, 15);
+	RNDr(S, W, 16);
+	RNDr(S, W, 17);
+	RNDr(S, W, 18);
+	RNDr(S, W, 19);
+	RNDr(S, W, 20);
+	RNDr(S, W, 21);
+	RNDr(S, W, 22);
+	RNDr(S, W, 23);
+	RNDr(S, W, 24);
+	RNDr(S, W, 25);
+	RNDr(S, W, 26);
+	RNDr(S, W, 27);
+	RNDr(S, W, 28);
+	RNDr(S, W, 29);
+	RNDr(S, W, 30);
+	RNDr(S, W, 31);
+	RNDr(S, W, 32);
+	RNDr(S, W, 33);
+	RNDr(S, W, 34);
+	RNDr(S, W, 35);
+	RNDr(S, W, 36);
+	RNDr(S, W, 37);
+	RNDr(S, W, 38);
+	RNDr(S, W, 39);
+	RNDr(S, W, 40);
+	RNDr(S, W, 41);
+	RNDr(S, W, 42);
+	RNDr(S, W, 43);
+	RNDr(S, W, 44);
+	RNDr(S, W, 45);
+	RNDr(S, W, 46);
+	RNDr(S, W, 47);
+	RNDr(S, W, 48);
+	RNDr(S, W, 49);
+	RNDr(S, W, 50);
+	RNDr(S, W, 51);
+	RNDr(S, W, 52);
+	RNDr(S, W, 53);
+	RNDr(S, W, 54);
+	RNDr(S, W, 55);
+	RNDr(S, W, 56);
+	RNDr(S, W, 57);
+	RNDr(S, W, 58);
+	RNDr(S, W, 59);
+	RNDr(S, W, 60);
+	RNDr(S, W, 61);
+	RNDr(S, W, 62);
+	RNDr(S, W, 63);
+
+	for (i = 0; i < 8; i++)
+		S[i] += midstate[i];
+	
+	W[18] = S[18];
+	W[19] = S[19];
+	W[20] = S[20];
+	W[22] = S[22];
+	W[23] = S[23];
+	W[24] = S[24];
+	W[30] = S[30];
+	W[31] = S[31];
+	
+	memcpy(S + 8, sha256d_hash1 + 8, 32);
+	S[16] = s1(sha256d_hash1[14]) + sha256d_hash1[ 9] + s0(S[ 1]) + S[ 0];
+	S[17] = s1(sha256d_hash1[15]) + sha256d_hash1[10] + s0(S[ 2]) + S[ 1];
+	S[18] = s1(S[16]) + sha256d_hash1[11] + s0(S[ 3]) + S[ 2];
+	S[19] = s1(S[17]) + sha256d_hash1[12] + s0(S[ 4]) + S[ 3];
+	S[20] = s1(S[18]) + sha256d_hash1[13] + s0(S[ 5]) + S[ 4];
+	S[21] = s1(S[19]) + sha256d_hash1[14] + s0(S[ 6]) + S[ 5];
+	S[22] = s1(S[20]) + sha256d_hash1[15] + s0(S[ 7]) + S[ 6];
+	S[23] = s1(S[21]) + S[16] + s0(sha256d_hash1[ 8]) + S[ 7];
+	S[24] = s1(S[22]) + S[17] + s0(sha256d_hash1[ 9]) + sha256d_hash1[ 8];
+	S[25] = s1(S[23]) + S[18] + s0(sha256d_hash1[10]) + sha256d_hash1[ 9];
+	S[26] = s1(S[24]) + S[19] + s0(sha256d_hash1[11]) + sha256d_hash1[10];
+	S[27] = s1(S[25]) + S[20] + s0(sha256d_hash1[12]) + sha256d_hash1[11];
+	S[28] = s1(S[26]) + S[21] + s0(sha256d_hash1[13]) + sha256d_hash1[12];
+	S[29] = s1(S[27]) + S[22] + s0(sha256d_hash1[14]) + sha256d_hash1[13];
+	S[30] = s1(S[28]) + S[23] + s0(sha256d_hash1[15]) + sha256d_hash1[14];
+	S[31] = s1(S[29]) + S[24] + s0(S[16])             + sha256d_hash1[15];
+	for (i = 32; i < 60; i += 2) {
+		S[i]   = s1(S[i - 2]) + S[i - 7] + s0(S[i - 15]) + S[i - 16];
+		S[i+1] = s1(S[i - 1]) + S[i - 6] + s0(S[i - 14]) + S[i - 15];
+	}
+	S[60] = s1(S[58]) + S[53] + s0(S[45]) + S[44];
+
+	sha256_init(hash);
+
+	RNDr(hash, S,  0);
+	RNDr(hash, S,  1);
+	RNDr(hash, S,  2);
+	RNDr(hash, S,  3);
+	RNDr(hash, S,  4);
+	RNDr(hash, S,  5);
+	RNDr(hash, S,  6);
+	RNDr(hash, S,  7);
+	RNDr(hash, S,  8);
+	RNDr(hash, S,  9);
+	RNDr(hash, S, 10);
+	RNDr(hash, S, 11);
+	RNDr(hash, S, 12);
+	RNDr(hash, S, 13);
+	RNDr(hash, S, 14);
+	RNDr(hash, S, 15);
+	RNDr(hash, S, 16);
+	RNDr(hash, S, 17);
+	RNDr(hash, S, 18);
+	RNDr(hash, S, 19);
+	RNDr(hash, S, 20);
+	RNDr(hash, S, 21);
+	RNDr(hash, S, 22);
+	RNDr(hash, S, 23);
+	RNDr(hash, S, 24);
+	RNDr(hash, S, 25);
+	RNDr(hash, S, 26);
+	RNDr(hash, S, 27);
+	RNDr(hash, S, 28);
+	RNDr(hash, S, 29);
+	RNDr(hash, S, 30);
+	RNDr(hash, S, 31);
+	RNDr(hash, S, 32);
+	RNDr(hash, S, 33);
+	RNDr(hash, S, 34);
+	RNDr(hash, S, 35);
+	RNDr(hash, S, 36);
+	RNDr(hash, S, 37);
+	RNDr(hash, S, 38);
+	RNDr(hash, S, 39);
+	RNDr(hash, S, 40);
+	RNDr(hash, S, 41);
+	RNDr(hash, S, 42);
+	RNDr(hash, S, 43);
+	RNDr(hash, S, 44);
+	RNDr(hash, S, 45);
+	RNDr(hash, S, 46);
+	RNDr(hash, S, 47);
+	RNDr(hash, S, 48);
+	RNDr(hash, S, 49);
+	RNDr(hash, S, 50);
+	RNDr(hash, S, 51);
+	RNDr(hash, S, 52);
+	RNDr(hash, S, 53);
+	RNDr(hash, S, 54);
+	RNDr(hash, S, 55);
+	RNDr(hash, S, 56);
+	
+	hash[2] += hash[6] + S1(hash[3]) + Ch(hash[3], hash[4], hash[5])
+	         + S[57] + sha256_k[57];
+	hash[1] += hash[5] + S1(hash[2]) + Ch(hash[2], hash[3], hash[4])
+	         + S[58] + sha256_k[58];
+	hash[0] += hash[4] + S1(hash[1]) + Ch(hash[1], hash[2], hash[3])
+	         + S[59] + sha256_k[59];
+	hash[7] += hash[3] + S1(hash[0]) + Ch(hash[0], hash[1], hash[2])
+	         + S[60] + sha256_k[60]
+	         + sha256_h[7];
+}
+
+#endif /* EXTERN_SHA256 */
+
+#ifdef HAVE_SHA256_4WAY
+
+void sha256d_ms_4way(uint32_t *hash,  uint32_t *data,
+	const uint32_t *midstate, const uint32_t *prehash);
+
+static inline int scanhash_sha256d_4way(int thr_id, uint32_t *pdata,
+	const uint32_t *ptarget, uint32_t max_nonce, unsigned long *hashes_done)
+{
+	uint32_t data[4 * 64] __attribute__((aligned(128)));
+	uint32_t hash[4 * 8] __attribute__((aligned(32)));
+	uint32_t midstate[4 * 8] __attribute__((aligned(32)));
+	uint32_t prehash[4 * 8] __attribute__((aligned(32)));
+	uint32_t n = pdata[19] - 1;
+	const uint32_t first_nonce = pdata[19];
+	const uint32_t Htarg = ptarget[7];
+	int i, j;
+	
+	memcpy(data, pdata + 16, 64);
+	sha256d_preextend(data);
+	for (i = 31; i >= 0; i--)
+		for (j = 0; j < 4; j++)
+			data[i * 4 + j] = data[i];
+	
+	sha256_init(midstate);
+	sha256_transform(midstate, pdata, 0);
+	memcpy(prehash, midstate, 32);
+	sha256d_prehash(prehash, pdata + 16);
+	for (i = 7; i >= 0; i--) {
+		for (j = 0; j < 4; j++) {
+			midstate[i * 4 + j] = midstate[i];
+			prehash[i * 4 + j] = prehash[i];
+		}
+	}
+	
+	do {
+		for (i = 0; i < 4; i++)
+			data[4 * 3 + i] = ++n;
+		
+		sha256d_ms_4way(hash, data, midstate, prehash);
+		
+		for (i = 0; i < 4; i++) {
+			if (swab32(hash[4 * 7 + i]) <= Htarg) {
+				pdata[19] = data[4 * 3 + i];
+				sha256d_80_swap(hash, pdata);
+				if (fulltest(hash, ptarget)) {
+					*hashes_done = n - first_nonce + 1;
+					return 1;
+				}
+			}
+		}
+	} while (n < max_nonce && !work_restart[thr_id].restart);
+	
+	*hashes_done = n - first_nonce + 1;
+	pdata[19] = n;
+	return 0;
+}
+
+#endif /* HAVE_SHA256_4WAY */
+
+#ifdef HAVE_SHA256_8WAY
+
+void sha256d_ms_8way(uint32_t *hash,  uint32_t *data,
+	const uint32_t *midstate, const uint32_t *prehash);
+
+static inline int scanhash_sha256d_8way(int thr_id, uint32_t *pdata,
+	const uint32_t *ptarget, uint32_t max_nonce, unsigned long *hashes_done)
+{
+	uint32_t data[8 * 64] __attribute__((aligned(128)));
+	uint32_t hash[8 * 8] __attribute__((aligned(32)));
+	uint32_t midstate[8 * 8] __attribute__((aligned(32)));
+	uint32_t prehash[8 * 8] __attribute__((aligned(32)));
+	uint32_t n = pdata[19] - 1;
+	const uint32_t first_nonce = pdata[19];
+	const uint32_t Htarg = ptarget[7];
+	int i, j;
+	
+	memcpy(data, pdata + 16, 64);
+	sha256d_preextend(data);
+	for (i = 31; i >= 0; i--)
+		for (j = 0; j < 8; j++)
+			data[i * 8 + j] = data[i];
+	
+	sha256_init(midstate);
+	sha256_transform(midstate, pdata, 0);
+	memcpy(prehash, midstate, 32);
+	sha256d_prehash(prehash, pdata + 16);
+	for (i = 7; i >= 0; i--) {
+		for (j = 0; j < 8; j++) {
+			midstate[i * 8 + j] = midstate[i];
+			prehash[i * 8 + j] = prehash[i];
+		}
+	}
+	
+	do {
+		for (i = 0; i < 8; i++)
+			data[8 * 3 + i] = ++n;
+		
+		sha256d_ms_8way(hash, data, midstate, prehash);
+		
+		for (i = 0; i < 8; i++) {
+			if (swab32(hash[8 * 7 + i]) <= Htarg) {
+				pdata[19] = data[8 * 3 + i];
+				sha256d_80_swap(hash, pdata);
+				if (fulltest(hash, ptarget)) {
+					*hashes_done = n - first_nonce + 1;
+					return 1;
+				}
+			}
+		}
+	} while (n < max_nonce && !work_restart[thr_id].restart);
+	
+	*hashes_done = n - first_nonce + 1;
+	pdata[19] = n;
+	return 0;
+}
+
+#endif /* HAVE_SHA256_8WAY */
+
+int scanhash_sha256d(int thr_id, uint32_t *pdata, const uint32_t *ptarget,
+	uint32_t max_nonce, unsigned long *hashes_done)
+{
+	uint32_t data[64] __attribute__((aligned(128)));
+	uint32_t hash[8] __attribute__((aligned(32)));
+	uint32_t midstate[8] __attribute__((aligned(32)));
+	uint32_t prehash[8] __attribute__((aligned(32)));
+	uint32_t n = pdata[19] - 1;
+	const uint32_t first_nonce = pdata[19];
+	const uint32_t Htarg = ptarget[7];
+	
+#ifdef HAVE_SHA256_8WAY
+	if (sha256_use_8way())
+		return scanhash_sha256d_8way(thr_id, pdata, ptarget,
+			max_nonce, hashes_done);
+#endif
+#ifdef HAVE_SHA256_4WAY
+	if (sha256_use_4way())
+		return scanhash_sha256d_4way(thr_id, pdata, ptarget,
+			max_nonce, hashes_done);
+#endif
+	
+	memcpy(data, pdata + 16, 64);
+	sha256d_preextend(data);
+	
+	sha256_init(midstate);
+	sha256_transform(midstate, pdata, 0);
+	memcpy(prehash, midstate, 32);
+	sha256d_prehash(prehash, pdata + 16);
+	
+	do {
+		data[3] = ++n;
+		sha256d_ms(hash, data, midstate, prehash);
+		if (swab32(hash[7]) <= Htarg) {
+			pdata[19] = data[3];
+			sha256d_80_swap(hash, pdata);
+			if (fulltest(hash, ptarget)) {
+				*hashes_done = n - first_nonce + 1;
+				return 1;
+			}
+		}
+	} while (n < max_nonce && !work_restart[thr_id].restart);
+	
+	*hashes_done = n - first_nonce + 1;
+	pdata[19] = n;
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha256.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha256.c
new file mode 100644
index 0000000..f9036a4
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha256.c
@@ -0,0 +1,411 @@
+/*-
+ * Copyright 2005,2007,2009 Colin Percival
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <sys/types.h>
+
+#include <stdint.h>
+#include <string.h>
+
+#include "sysendian.h"
+
+#include "sha256.h"
+
+/*
+ * Encode a length len/4 vector of (uint32_t) into a length len vector of
+ * (unsigned char) in big-endian form.  Assumes len is a multiple of 4.
+ */
+static void
+be32enc_vect(unsigned char *dst, const uint32_t *src, size_t len)
+{
+	size_t i;
+
+	for (i = 0; i < len / 4; i++)
+		be32enc(dst + i * 4, src[i]);
+}
+
+/*
+ * Decode a big-endian length len vector of (unsigned char) into a length
+ * len/4 vector of (uint32_t).  Assumes len is a multiple of 4.
+ */
+static void
+be32dec_vect(uint32_t *dst, const unsigned char *src, size_t len)
+{
+	size_t i;
+
+	for (i = 0; i < len / 4; i++)
+		dst[i] = be32dec(src + i * 4);
+}
+
+/* Elementary functions used by SHA256 */
+#define Ch(x, y, z)	((x & (y ^ z)) ^ z)
+#define Maj(x, y, z)	((x & (y | z)) | (y & z))
+#define SHR(x, n)	(x >> n)
+#define ROTR(x, n)	((x >> n) | (x << (32 - n)))
+#define S0(x)		(ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
+#define S1(x)		(ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
+#define s0(x)		(ROTR(x, 7) ^ ROTR(x, 18) ^ SHR(x, 3))
+#define s1(x)		(ROTR(x, 17) ^ ROTR(x, 19) ^ SHR(x, 10))
+
+/* SHA256 round function */
+#define RND(a, b, c, d, e, f, g, h, k)			\
+	t0 = h + S1(e) + Ch(e, f, g) + k;		\
+	t1 = S0(a) + Maj(a, b, c);			\
+	d += t0;					\
+	h  = t0 + t1;
+
+/* Adjusted round function for rotating state */
+#define RNDr(S, W, i, k)			\
+	RND(S[(64 - i) % 8], S[(65 - i) % 8],	\
+	    S[(66 - i) % 8], S[(67 - i) % 8],	\
+	    S[(68 - i) % 8], S[(69 - i) % 8],	\
+	    S[(70 - i) % 8], S[(71 - i) % 8],	\
+	    W[i] + k)
+
+/*
+ * SHA256 block compression function.  The 256-bit state is transformed via
+ * the 512-bit input block to produce a new state.
+ */
+static void
+SHA256_Transform(uint32_t * state, const unsigned char block[64])
+{
+	uint32_t W[64];
+	uint32_t S[8];
+	uint32_t t0, t1;
+	int i;
+
+	/* 1. Prepare message schedule W. */
+	be32dec_vect(W, block, 64);
+	for (i = 16; i < 64; i++)
+		W[i] = s1(W[i - 2]) + W[i - 7] + s0(W[i - 15]) + W[i - 16];
+
+	/* 2. Initialize working variables. */
+	memcpy(S, state, 32);
+
+	/* 3. Mix. */
+	RNDr(S, W, 0, 0x428a2f98);
+	RNDr(S, W, 1, 0x71374491);
+	RNDr(S, W, 2, 0xb5c0fbcf);
+	RNDr(S, W, 3, 0xe9b5dba5);
+	RNDr(S, W, 4, 0x3956c25b);
+	RNDr(S, W, 5, 0x59f111f1);
+	RNDr(S, W, 6, 0x923f82a4);
+	RNDr(S, W, 7, 0xab1c5ed5);
+	RNDr(S, W, 8, 0xd807aa98);
+	RNDr(S, W, 9, 0x12835b01);
+	RNDr(S, W, 10, 0x243185be);
+	RNDr(S, W, 11, 0x550c7dc3);
+	RNDr(S, W, 12, 0x72be5d74);
+	RNDr(S, W, 13, 0x80deb1fe);
+	RNDr(S, W, 14, 0x9bdc06a7);
+	RNDr(S, W, 15, 0xc19bf174);
+	RNDr(S, W, 16, 0xe49b69c1);
+	RNDr(S, W, 17, 0xefbe4786);
+	RNDr(S, W, 18, 0x0fc19dc6);
+	RNDr(S, W, 19, 0x240ca1cc);
+	RNDr(S, W, 20, 0x2de92c6f);
+	RNDr(S, W, 21, 0x4a7484aa);
+	RNDr(S, W, 22, 0x5cb0a9dc);
+	RNDr(S, W, 23, 0x76f988da);
+	RNDr(S, W, 24, 0x983e5152);
+	RNDr(S, W, 25, 0xa831c66d);
+	RNDr(S, W, 26, 0xb00327c8);
+	RNDr(S, W, 27, 0xbf597fc7);
+	RNDr(S, W, 28, 0xc6e00bf3);
+	RNDr(S, W, 29, 0xd5a79147);
+	RNDr(S, W, 30, 0x06ca6351);
+	RNDr(S, W, 31, 0x14292967);
+	RNDr(S, W, 32, 0x27b70a85);
+	RNDr(S, W, 33, 0x2e1b2138);
+	RNDr(S, W, 34, 0x4d2c6dfc);
+	RNDr(S, W, 35, 0x53380d13);
+	RNDr(S, W, 36, 0x650a7354);
+	RNDr(S, W, 37, 0x766a0abb);
+	RNDr(S, W, 38, 0x81c2c92e);
+	RNDr(S, W, 39, 0x92722c85);
+	RNDr(S, W, 40, 0xa2bfe8a1);
+	RNDr(S, W, 41, 0xa81a664b);
+	RNDr(S, W, 42, 0xc24b8b70);
+	RNDr(S, W, 43, 0xc76c51a3);
+	RNDr(S, W, 44, 0xd192e819);
+	RNDr(S, W, 45, 0xd6990624);
+	RNDr(S, W, 46, 0xf40e3585);
+	RNDr(S, W, 47, 0x106aa070);
+	RNDr(S, W, 48, 0x19a4c116);
+	RNDr(S, W, 49, 0x1e376c08);
+	RNDr(S, W, 50, 0x2748774c);
+	RNDr(S, W, 51, 0x34b0bcb5);
+	RNDr(S, W, 52, 0x391c0cb3);
+	RNDr(S, W, 53, 0x4ed8aa4a);
+	RNDr(S, W, 54, 0x5b9cca4f);
+	RNDr(S, W, 55, 0x682e6ff3);
+	RNDr(S, W, 56, 0x748f82ee);
+	RNDr(S, W, 57, 0x78a5636f);
+	RNDr(S, W, 58, 0x84c87814);
+	RNDr(S, W, 59, 0x8cc70208);
+	RNDr(S, W, 60, 0x90befffa);
+	RNDr(S, W, 61, 0xa4506ceb);
+	RNDr(S, W, 62, 0xbef9a3f7);
+	RNDr(S, W, 63, 0xc67178f2);
+
+	/* 4. Mix local working variables into global state */
+	for (i = 0; i < 8; i++)
+		state[i] += S[i];
+
+	/* Clean the stack. */
+	memset(W, 0, 256);
+	memset(S, 0, 32);
+	t0 = t1 = 0;
+}
+
+static unsigned char PAD[64] = {
+	0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+
+/* Add padding and terminating bit-count. */
+static void
+SHA256_Pad(SHA256_CTX * ctx)
+{
+	unsigned char len[8];
+	uint32_t r, plen;
+
+	/*
+	 * Convert length to a vector of bytes -- we do this now rather
+	 * than later because the length will change after we pad.
+	 */
+	be32enc_vect(len, ctx->count, 8);
+
+	/* Add 1--64 bytes so that the resulting length is 56 mod 64 */
+	r = (ctx->count[1] >> 3) & 0x3f;
+	plen = (r < 56) ? (56 - r) : (120 - r);
+	SHA256_Update(ctx, PAD, (size_t)plen);
+
+	/* Add the terminating bit-count */
+	SHA256_Update(ctx, len, 8);
+}
+
+/* SHA-256 initialization.  Begins a SHA-256 operation. */
+static void
+SHA256_Init(SHA256_CTX * ctx)
+{
+
+	/* Zero bits processed so far */
+	ctx->count[0] = ctx->count[1] = 0;
+
+	/* Magic initialization constants */
+	ctx->state[0] = 0x6A09E667;
+	ctx->state[1] = 0xBB67AE85;
+	ctx->state[2] = 0x3C6EF372;
+	ctx->state[3] = 0xA54FF53A;
+	ctx->state[4] = 0x510E527F;
+	ctx->state[5] = 0x9B05688C;
+	ctx->state[6] = 0x1F83D9AB;
+	ctx->state[7] = 0x5BE0CD19;
+}
+
+/* Add bytes into the hash */
+static void
+SHA256_Update(SHA256_CTX * ctx, const void *in, size_t len)
+{
+	uint32_t bitlen[2];
+	uint32_t r;
+	const unsigned char *src = in;
+
+	/* Number of bytes left in the buffer from previous updates */
+	r = (ctx->count[1] >> 3) & 0x3f;
+
+	/* Convert the length into a number of bits */
+	bitlen[1] = ((uint32_t)len) << 3;
+	bitlen[0] = (uint32_t)(len >> 29);
+
+	/* Update number of bits */
+	if ((ctx->count[1] += bitlen[1]) < bitlen[1])
+		ctx->count[0]++;
+	ctx->count[0] += bitlen[0];
+
+	/* Handle the case where we don't need to perform any transforms */
+	if (len < 64 - r) {
+		memcpy(&ctx->buf[r], src, len);
+		return;
+	}
+
+	/* Finish the current block */
+	memcpy(&ctx->buf[r], src, 64 - r);
+	SHA256_Transform(ctx->state, ctx->buf);
+	src += 64 - r;
+	len -= 64 - r;
+
+	/* Perform complete blocks */
+	while (len >= 64) {
+		SHA256_Transform(ctx->state, src);
+		src += 64;
+		len -= 64;
+	}
+
+	/* Copy left over data into buffer */
+	memcpy(ctx->buf, src, len);
+}
+
+/*
+ * SHA-256 finalization.  Pads the input data, exports the hash value,
+ * and clears the context state.
+ */
+static void
+SHA256_Final(unsigned char digest[32], SHA256_CTX * ctx)
+{
+
+	/* Add padding */
+	SHA256_Pad(ctx);
+
+	/* Write the hash */
+	be32enc_vect(digest, ctx->state, 32);
+
+	/* Clear the context state */
+	memset((void *)ctx, 0, sizeof(*ctx));
+}
+
+/* Initialize an HMAC-SHA256 operation with the given key. */
+static void
+HMAC_SHA256_Init(HMAC_SHA256_CTX * ctx, const void * _K, size_t Klen)
+{
+	unsigned char pad[64];
+	unsigned char khash[32];
+	const unsigned char * K = _K;
+	size_t i;
+
+	/* If Klen > 64, the key is really SHA256(K). */
+	if (Klen > 64) {
+		SHA256_Init(&ctx->ictx);
+		SHA256_Update(&ctx->ictx, K, Klen);
+		SHA256_Final(khash, &ctx->ictx);
+		K = khash;
+		Klen = 32;
+	}
+
+	/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
+	SHA256_Init(&ctx->ictx);
+	memset(pad, 0x36, 64);
+	for (i = 0; i < Klen; i++)
+		pad[i] ^= K[i];
+	SHA256_Update(&ctx->ictx, pad, 64);
+
+	/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
+	SHA256_Init(&ctx->octx);
+	memset(pad, 0x5c, 64);
+	for (i = 0; i < Klen; i++)
+		pad[i] ^= K[i];
+	SHA256_Update(&ctx->octx, pad, 64);
+
+	/* Clean the stack. */
+	memset(khash, 0, 32);
+}
+
+/* Add bytes to the HMAC-SHA256 operation. */
+static void
+HMAC_SHA256_Update(HMAC_SHA256_CTX * ctx, const void *in, size_t len)
+{
+
+	/* Feed data to the inner SHA256 operation. */
+	SHA256_Update(&ctx->ictx, in, len);
+}
+
+/* Finish an HMAC-SHA256 operation. */
+static void
+HMAC_SHA256_Final(unsigned char digest[32], HMAC_SHA256_CTX * ctx)
+{
+	unsigned char ihash[32];
+
+	/* Finish the inner SHA256 operation. */
+	SHA256_Final(ihash, &ctx->ictx);
+
+	/* Feed the inner hash to the outer SHA256 operation. */
+	SHA256_Update(&ctx->octx, ihash, 32);
+
+	/* Finish the outer SHA256 operation. */
+	SHA256_Final(digest, &ctx->octx);
+
+	/* Clean the stack. */
+	memset(ihash, 0, 32);
+}
+
+/**
+ * PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
+ * Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
+ * write the output to buf.  The value dkLen must be at most 32 * (2^32 - 1).
+ */
+static void
+PBKDF2_SHA256(const uint8_t * passwd, size_t passwdlen, const uint8_t * salt,
+    size_t saltlen, uint64_t c, uint8_t * buf, size_t dkLen)
+{
+	HMAC_SHA256_CTX PShctx, hctx;
+	size_t i;
+	uint8_t ivec[4];
+	uint8_t U[32];
+	uint8_t T[32];
+	uint64_t j;
+	int k;
+	size_t clen;
+
+	/* Compute HMAC state after processing P and S. */
+	HMAC_SHA256_Init(&PShctx, passwd, passwdlen);
+	HMAC_SHA256_Update(&PShctx, salt, saltlen);
+
+	/* Iterate through the blocks. */
+	for (i = 0; i * 32 < dkLen; i++) {
+		/* Generate INT(i + 1). */
+		be32enc(ivec, (uint32_t)(i + 1));
+
+		/* Compute U_1 = PRF(P, S || INT(i)). */
+		memcpy(&hctx, &PShctx, sizeof(HMAC_SHA256_CTX));
+		HMAC_SHA256_Update(&hctx, ivec, 4);
+		HMAC_SHA256_Final(U, &hctx);
+
+		/* T_i = U_1 ... */
+		memcpy(T, U, 32);
+
+		for (j = 2; j <= c; j++) {
+			/* Compute U_j. */
+			HMAC_SHA256_Init(&hctx, passwd, passwdlen);
+			HMAC_SHA256_Update(&hctx, U, 32);
+			HMAC_SHA256_Final(U, &hctx);
+
+			/* ... xor U_j ... */
+			for (k = 0; k < 32; k++)
+				T[k] ^= U[k];
+		}
+
+		/* Copy as many bytes as necessary into buf. */
+		clen = dkLen - i * 32;
+		if (clen > 32)
+			clen = 32;
+		memcpy(&buf[i * 32], T, clen);
+	}
+
+	/* Clean PShctx, since we never called _Final on it. */
+	memset(&PShctx, 0, sizeof(HMAC_SHA256_CTX));
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sha256.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha256.h
new file mode 100644
index 0000000..6c18338
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sha256.h
@@ -0,0 +1,62 @@
+/*-
+ * Copyright 2005,2007,2009 Colin Percival
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * $FreeBSD: src/lib/libmd/sha256.h,v 1.2 2006/01/17 15:35:56 phk Exp $
+ */
+
+#ifndef _SHA256_H_
+#define _SHA256_H_
+
+#include <sys/types.h>
+
+#include <stdint.h>
+
+typedef struct SHA256Context {
+	uint32_t state[8];
+	uint32_t count[2];
+	unsigned char buf[64];
+} SHA256_CTX;
+
+typedef struct HMAC_SHA256Context {
+	SHA256_CTX ictx;
+	SHA256_CTX octx;
+} HMAC_SHA256_CTX;
+
+static void	SHA256_Init(SHA256_CTX *);
+static void	SHA256_Update(SHA256_CTX *, const void *, size_t);
+static void	SHA256_Final(unsigned char [32], SHA256_CTX *);
+static void	HMAC_SHA256_Init(HMAC_SHA256_CTX *, const void *, size_t);
+static void	HMAC_SHA256_Update(HMAC_SHA256_CTX *, const void *, size_t);
+static void	HMAC_SHA256_Final(unsigned char [32], HMAC_SHA256_CTX *);
+
+/**
+ * PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
+ * Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
+ * write the output to buf.  The value dkLen must be at most 32 * (2^32 - 1).
+ */
+static void	PBKDF2_SHA256(const uint8_t *, size_t, const uint8_t *, size_t,
+    uint64_t, uint8_t *, size_t);
+
+#endif /* !_SHA256_H_ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/sysendian.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/sysendian.h
new file mode 100644
index 0000000..0495194
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/sysendian.h
@@ -0,0 +1,138 @@
+/*-
+ * Copyright 2007-2009 Colin Percival
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ */
+#ifndef _SYSENDIAN_H_
+#define _SYSENDIAN_H_
+
+/* If we don't have be64enc, the <sys/endian.h> we have isn't usable. */
+#if !HAVE_DECL_BE64ENC
+#undef HAVE_SYS_ENDIAN_H
+#endif
+
+#ifdef HAVE_SYS_ENDIAN_H
+
+#include <sys/endian.h>
+
+#else
+
+#include <stdint.h>
+
+static inline uint32_t
+be32dec(const void *pp)
+{
+	const uint8_t *p = (uint8_t const *)pp;
+
+	return ((uint32_t)(p[3]) + ((uint32_t)(p[2]) << 8) +
+	    ((uint32_t)(p[1]) << 16) + ((uint32_t)(p[0]) << 24));
+}
+
+static inline void
+be32enc(void *pp, uint32_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[3] = x & 0xff;
+	p[2] = (x >> 8) & 0xff;
+	p[1] = (x >> 16) & 0xff;
+	p[0] = (x >> 24) & 0xff;
+}
+
+static inline uint64_t
+be64dec(const void *pp)
+{
+	const uint8_t *p = (uint8_t const *)pp;
+
+	return ((uint64_t)(p[7]) + ((uint64_t)(p[6]) << 8) +
+	    ((uint64_t)(p[5]) << 16) + ((uint64_t)(p[4]) << 24) +
+	    ((uint64_t)(p[3]) << 32) + ((uint64_t)(p[2]) << 40) +
+	    ((uint64_t)(p[1]) << 48) + ((uint64_t)(p[0]) << 56));
+}
+
+static inline void
+be64enc(void *pp, uint64_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[7] = x & 0xff;
+	p[6] = (x >> 8) & 0xff;
+	p[5] = (x >> 16) & 0xff;
+	p[4] = (x >> 24) & 0xff;
+	p[3] = (x >> 32) & 0xff;
+	p[2] = (x >> 40) & 0xff;
+	p[1] = (x >> 48) & 0xff;
+	p[0] = (x >> 56) & 0xff;
+}
+
+static inline uint32_t
+le32dec(const void *pp)
+{
+	const uint8_t *p = (uint8_t const *)pp;
+
+	return ((uint32_t)(p[0]) + ((uint32_t)(p[1]) << 8) +
+	    ((uint32_t)(p[2]) << 16) + ((uint32_t)(p[3]) << 24));
+}
+
+static inline void
+le32enc(void *pp, uint32_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[0] = x & 0xff;
+	p[1] = (x >> 8) & 0xff;
+	p[2] = (x >> 16) & 0xff;
+	p[3] = (x >> 24) & 0xff;
+}
+
+static inline uint64_t
+le64dec(const void *pp)
+{
+	const uint8_t *p = (uint8_t const *)pp;
+
+	return ((uint64_t)(p[0]) + ((uint64_t)(p[1]) << 8) +
+	    ((uint64_t)(p[2]) << 16) + ((uint64_t)(p[3]) << 24) +
+	    ((uint64_t)(p[4]) << 32) + ((uint64_t)(p[5]) << 40) +
+	    ((uint64_t)(p[6]) << 48) + ((uint64_t)(p[7]) << 56));
+}
+
+static inline void
+le64enc(void *pp, uint64_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[0] = x & 0xff;
+	p[1] = (x >> 8) & 0xff;
+	p[2] = (x >> 16) & 0xff;
+	p[3] = (x >> 24) & 0xff;
+	p[4] = (x >> 32) & 0xff;
+	p[5] = (x >> 40) & 0xff;
+	p[6] = (x >> 48) & 0xff;
+	p[7] = (x >> 56) & 0xff;
+}
+#endif /* !HAVE_SYS_ENDIAN_H */
+
+#endif /* !_SYSENDIAN_H_ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/util.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/util.c
new file mode 100644
index 0000000..0c6dc35
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/util.c
@@ -0,0 +1,1592 @@
+/*
+ * Copyright 2010 Jeff Garzik
+ * Copyright 2012 Luke Dashjr
+ * Copyright 2012-2017 pooler
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.  See COPYING for more details.
+ */
+//
+//
+// ottylab update history <ottylab@gmail.com>
+//
+// 2018/02/17 @001 Support Android logcat
+// 2018/02/17 @002 Log with callback
+// 2018/08/20 @003 User default Log when callback is null
+//
+
+#define _GNU_SOURCE
+#include "cpuminer-config.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <ctype.h>
+#include <stdarg.h>
+#include <string.h>
+#include <stdbool.h>
+#include <inttypes.h>
+#include <limits.h>
+#include <errno.h>
+#include <unistd.h>
+#include <jansson.h>
+#include <curl/curl.h>
+#include <time.h>
+#if defined(WIN32)
+#include <winsock2.h>
+#include <mstcpip.h>
+#else
+#include <sys/socket.h>
+#include <netinet/in.h>
+#include <netinet/tcp.h>
+#endif
+#include "compat.h"
+#include "miner.h"
+#include "elist.h"
+//@001A start
+#ifdef ANDROID
+#include <android/log.h>
+#endif
+//@001A end
+void (*vprintf_cb)(const char *format, va_list arg); //@002A
+
+struct data_buffer {
+	void		*buf;
+	size_t		len;
+};
+
+struct upload_buffer {
+	const void	*buf;
+	size_t		len;
+	size_t		pos;
+};
+
+struct header_info {
+	char		*lp_path;
+	char		*reason;
+	char		*stratum_url;
+};
+
+struct tq_ent {
+	void			*data;
+	struct list_head	q_node;
+};
+
+struct thread_q {
+	struct list_head	q;
+
+	bool frozen;
+
+	pthread_mutex_t		mutex;
+	pthread_cond_t		cond;
+};
+
+void applog(int prio, const char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+
+#ifdef HAVE_SYSLOG_H
+	if (use_syslog) {
+		va_list ap2;
+		char *buf;
+		int len;
+		
+		va_copy(ap2, ap);
+		len = vsnprintf(NULL, 0, fmt, ap2) + 1;
+		va_end(ap2);
+		buf = alloca(len);
+		if (vsnprintf(buf, len, fmt, ap) >= 0)
+			syslog(prio, "%s", buf);
+	}
+#else
+//@001A Start
+#ifdef ANDROID
+	if (!vprintf_cb) { //@002A
+		int android_prio ;
+		switch(prio) {
+			case LOG_ERR:
+				android_prio = ANDROID_LOG_ERROR;
+				break;
+			case LOG_WARNING:
+				android_prio = ANDROID_LOG_WARN;
+				break;
+			case LOG_NOTICE:
+				android_prio = ANDROID_LOG_INFO;
+				break;
+			case LOG_INFO:
+				android_prio = ANDROID_LOG_INFO;
+				break;
+			case LOG_DEBUG:
+				android_prio = ANDROID_LOG_DEBUG;
+				break;
+			default:
+				android_prio = ANDROID_LOG_UNKNOWN;
+				break;
+		}
+		__android_log_vprint(android_prio, "libcpuminer", fmt, ap);
+	}
+#else
+	if (0) { }
+#endif
+//@001A End
+	//if (0) { }//@001D
+#endif
+	else {
+		char *f;
+		int len;
+		time_t now;
+		struct tm tm, *tm_p;
+
+		time(&now);
+
+		pthread_mutex_lock(&applog_lock);
+		tm_p = localtime(&now);
+		memcpy(&tm, tm_p, sizeof(tm));
+		pthread_mutex_unlock(&applog_lock);
+
+		len = 40 + strlen(fmt) + 2;
+		f = alloca(len);
+		sprintf(f, "[%d-%02d-%02d %02d:%02d:%02d] %s\n",
+			tm.tm_year + 1900,
+			tm.tm_mon + 1,
+			tm.tm_mday,
+			tm.tm_hour,
+			tm.tm_min,
+			tm.tm_sec,
+			fmt);
+		pthread_mutex_lock(&applog_lock);
+		//@003A start
+		if (vprintf_cb) {
+			vprintf_cb(f, ap); //@002A
+		} else {
+			vfprintf(stderr, f, ap);	/* atomic write to stderr */
+		}
+		//@003A end
+		fflush(stderr);
+		pthread_mutex_unlock(&applog_lock);
+	}
+	va_end(ap);
+}
+
+/* Modify the representation of integer numbers which would cause an overflow
+ * so that they are treated as floating-point numbers.
+ * This is a hack to overcome the limitations of some versions of Jansson. */
+static char *hack_json_numbers(const char *in)
+{
+	char *out;
+	int i, off, intoff;
+	bool in_str, in_int;
+
+	out = calloc(2 * strlen(in) + 1, 1);
+	if (!out)
+		return NULL;
+	off = intoff = 0;
+	in_str = in_int = false;
+	for (i = 0; in[i]; i++) {
+		char c = in[i];
+		if (c == '"') {
+			in_str = !in_str;
+		} else if (c == '\\') {
+			out[off++] = c;
+			if (!in[++i])
+				break;
+		} else if (!in_str && !in_int && isdigit(c)) {
+			intoff = off;
+			in_int = true;
+		} else if (in_int && !isdigit(c)) {
+			if (c != '.' && c != 'e' && c != 'E' && c != '+' && c != '-') {
+				in_int = false;
+				if (off - intoff > 4) {
+					char *end;
+#if JSON_INTEGER_IS_LONG_LONG
+					errno = 0;
+					strtoll(out + intoff, &end, 10);
+					if (!*end && errno == ERANGE) {
+#else
+					long l;
+					errno = 0;
+					l = strtol(out + intoff, &end, 10);
+					if (!*end && (errno == ERANGE || l > INT_MAX)) {
+#endif
+						out[off++] = '.';
+						out[off++] = '0';
+					}
+				}
+			}
+		}
+		out[off++] = in[i];
+	}
+	return out;
+}
+
+static void databuf_free(struct data_buffer *db)
+{
+	if (!db)
+		return;
+
+	free(db->buf);
+
+	memset(db, 0, sizeof(*db));
+}
+
+static size_t all_data_cb(const void *ptr, size_t size, size_t nmemb,
+			  void *user_data)
+{
+	struct data_buffer *db = user_data;
+	size_t len = size * nmemb;
+	size_t oldlen, newlen;
+	void *newmem;
+	static const unsigned char zero = 0;
+
+	oldlen = db->len;
+	newlen = oldlen + len;
+
+	newmem = realloc(db->buf, newlen + 1);
+	if (!newmem)
+		return 0;
+
+	db->buf = newmem;
+	db->len = newlen;
+	memcpy(db->buf + oldlen, ptr, len);
+	memcpy(db->buf + newlen, &zero, 1);	/* null terminate */
+
+	return len;
+}
+
+static size_t upload_data_cb(void *ptr, size_t size, size_t nmemb,
+			     void *user_data)
+{
+	struct upload_buffer *ub = user_data;
+	int len = size * nmemb;
+
+	if (len > ub->len - ub->pos)
+		len = ub->len - ub->pos;
+
+	if (len) {
+		memcpy(ptr, ub->buf + ub->pos, len);
+		ub->pos += len;
+	}
+
+	return len;
+}
+
+#if LIBCURL_VERSION_NUM >= 0x071200
+static int seek_data_cb(void *user_data, curl_off_t offset, int origin)
+{
+	struct upload_buffer *ub = user_data;
+	
+	switch (origin) {
+	case SEEK_SET:
+		ub->pos = offset;
+		break;
+	case SEEK_CUR:
+		ub->pos += offset;
+		break;
+	case SEEK_END:
+		ub->pos = ub->len + offset;
+		break;
+	default:
+		return 1; /* CURL_SEEKFUNC_FAIL */
+	}
+
+	return 0; /* CURL_SEEKFUNC_OK */
+}
+#endif
+
+static size_t resp_hdr_cb(void *ptr, size_t size, size_t nmemb, void *user_data)
+{
+	struct header_info *hi = user_data;
+	size_t remlen, slen, ptrlen = size * nmemb;
+	char *rem, *val = NULL, *key = NULL;
+	void *tmp;
+
+	val = calloc(1, ptrlen);
+	key = calloc(1, ptrlen);
+	if (!key || !val)
+		goto out;
+
+	tmp = memchr(ptr, ':', ptrlen);
+	if (!tmp || (tmp == ptr))	/* skip empty keys / blanks */
+		goto out;
+	slen = tmp - ptr;
+	if ((slen + 1) == ptrlen)	/* skip key w/ no value */
+		goto out;
+	memcpy(key, ptr, slen);		/* store & nul term key */
+	key[slen] = 0;
+
+	rem = ptr + slen + 1;		/* trim value's leading whitespace */
+	remlen = ptrlen - slen - 1;
+	while ((remlen > 0) && (isspace(*rem))) {
+		remlen--;
+		rem++;
+	}
+
+	memcpy(val, rem, remlen);	/* store value, trim trailing ws */
+	val[remlen] = 0;
+	while ((*val) && (isspace(val[strlen(val) - 1]))) {
+		val[strlen(val) - 1] = 0;
+	}
+	if (!*val)			/* skip blank value */
+		goto out;
+
+	if (!strcasecmp("X-Long-Polling", key)) {
+		hi->lp_path = val;	/* steal memory reference */
+		val = NULL;
+	}
+
+	if (!strcasecmp("X-Reject-Reason", key)) {
+		hi->reason = val;	/* steal memory reference */
+		val = NULL;
+	}
+
+	if (!strcasecmp("X-Stratum", key)) {
+		hi->stratum_url = val;	/* steal memory reference */
+		val = NULL;
+	}
+
+out:
+	free(key);
+	free(val);
+	return ptrlen;
+}
+
+#if LIBCURL_VERSION_NUM >= 0x070f06
+static int sockopt_keepalive_cb(void *userdata, curl_socket_t fd,
+	curlsocktype purpose)
+{
+	int keepalive = 1;
+	int tcp_keepcnt = 3;
+	int tcp_keepidle = 50;
+	int tcp_keepintvl = 50;
+
+#ifndef WIN32
+	if (unlikely(setsockopt(fd, SOL_SOCKET, SO_KEEPALIVE, &keepalive,
+		sizeof(keepalive))))
+		return 1;
+#ifdef __linux
+	if (unlikely(setsockopt(fd, SOL_TCP, TCP_KEEPCNT,
+		&tcp_keepcnt, sizeof(tcp_keepcnt))))
+		return 1;
+	if (unlikely(setsockopt(fd, SOL_TCP, TCP_KEEPIDLE,
+		&tcp_keepidle, sizeof(tcp_keepidle))))
+		return 1;
+	if (unlikely(setsockopt(fd, SOL_TCP, TCP_KEEPINTVL,
+		&tcp_keepintvl, sizeof(tcp_keepintvl))))
+		return 1;
+#endif /* __linux */
+#ifdef __APPLE_CC__
+	if (unlikely(setsockopt(fd, IPPROTO_TCP, TCP_KEEPALIVE,
+		&tcp_keepintvl, sizeof(tcp_keepintvl))))
+		return 1;
+#endif /* __APPLE_CC__ */
+#else /* WIN32 */
+	struct tcp_keepalive vals;
+	vals.onoff = 1;
+	vals.keepalivetime = tcp_keepidle * 1000;
+	vals.keepaliveinterval = tcp_keepintvl * 1000;
+	DWORD outputBytes;
+	if (unlikely(WSAIoctl(fd, SIO_KEEPALIVE_VALS, &vals, sizeof(vals),
+		NULL, 0, &outputBytes, NULL, NULL)))
+		return 1;
+#endif /* WIN32 */
+
+	return 0;
+}
+#endif
+
+json_t *json_rpc_call(CURL *curl, const char *url,
+		      const char *userpass, const char *rpc_req,
+		      int *curl_err, int flags)
+{
+	json_t *val, *err_val, *res_val;
+	int rc;
+	long http_rc;
+	struct data_buffer all_data = {0};
+	struct upload_buffer upload_data;
+	char *json_buf;
+	json_error_t err;
+	struct curl_slist *headers = NULL;
+	char len_hdr[64];
+	char curl_err_str[CURL_ERROR_SIZE];
+	long timeout = (flags & JSON_RPC_LONGPOLL) ? opt_timeout : 30;
+	struct header_info hi = {0};
+
+	/* it is assumed that 'curl' is freshly [re]initialized at this pt */
+
+	if (opt_protocol)
+		curl_easy_setopt(curl, CURLOPT_VERBOSE, 1);
+	curl_easy_setopt(curl, CURLOPT_URL, url);
+	if (opt_cert)
+		curl_easy_setopt(curl, CURLOPT_CAINFO, opt_cert);
+	curl_easy_setopt(curl, CURLOPT_ENCODING, "");
+	curl_easy_setopt(curl, CURLOPT_FAILONERROR, 1);
+	curl_easy_setopt(curl, CURLOPT_NOSIGNAL, 1);
+	curl_easy_setopt(curl, CURLOPT_TCP_NODELAY, 1);
+	curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, all_data_cb);
+	curl_easy_setopt(curl, CURLOPT_WRITEDATA, &all_data);
+	curl_easy_setopt(curl, CURLOPT_READFUNCTION, upload_data_cb);
+	curl_easy_setopt(curl, CURLOPT_READDATA, &upload_data);
+#if LIBCURL_VERSION_NUM >= 0x071200
+	curl_easy_setopt(curl, CURLOPT_SEEKFUNCTION, &seek_data_cb);
+	curl_easy_setopt(curl, CURLOPT_SEEKDATA, &upload_data);
+#endif
+	curl_easy_setopt(curl, CURLOPT_ERRORBUFFER, curl_err_str);
+	if (opt_redirect)
+		curl_easy_setopt(curl, CURLOPT_FOLLOWLOCATION, 1);
+	curl_easy_setopt(curl, CURLOPT_TIMEOUT, timeout);
+	curl_easy_setopt(curl, CURLOPT_HEADERFUNCTION, resp_hdr_cb);
+	curl_easy_setopt(curl, CURLOPT_HEADERDATA, &hi);
+	if (opt_proxy) {
+		curl_easy_setopt(curl, CURLOPT_PROXY, opt_proxy);
+		curl_easy_setopt(curl, CURLOPT_PROXYTYPE, opt_proxy_type);
+	}
+	if (userpass) {
+		curl_easy_setopt(curl, CURLOPT_USERPWD, userpass);
+		curl_easy_setopt(curl, CURLOPT_HTTPAUTH, CURLAUTH_BASIC);
+	}
+#if LIBCURL_VERSION_NUM >= 0x070f06
+	if (flags & JSON_RPC_LONGPOLL)
+		curl_easy_setopt(curl, CURLOPT_SOCKOPTFUNCTION, sockopt_keepalive_cb);
+#endif
+	curl_easy_setopt(curl, CURLOPT_POST, 1);
+
+	if (opt_protocol)
+		applog(LOG_DEBUG, "JSON protocol request:\n%s\n", rpc_req);
+
+	upload_data.buf = rpc_req;
+	upload_data.len = strlen(rpc_req);
+	upload_data.pos = 0;
+	sprintf(len_hdr, "Content-Length: %lu",
+		(unsigned long) upload_data.len);
+
+	headers = curl_slist_append(headers, "Content-Type: application/json");
+	headers = curl_slist_append(headers, len_hdr);
+	headers = curl_slist_append(headers, "User-Agent: " USER_AGENT);
+	headers = curl_slist_append(headers, "X-Mining-Extensions: midstate");
+	headers = curl_slist_append(headers, "Accept:"); /* disable Accept hdr*/
+	headers = curl_slist_append(headers, "Expect:"); /* disable Expect hdr*/
+
+	curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
+
+	rc = curl_easy_perform(curl);
+	if (curl_err != NULL)
+		*curl_err = rc;
+	if (rc) {
+		curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &http_rc);
+		if (!((flags & JSON_RPC_LONGPOLL) && rc == CURLE_OPERATION_TIMEDOUT) &&
+		    !((flags & JSON_RPC_QUIET_404) && http_rc == 404))
+			applog(LOG_ERR, "HTTP request failed: %s", curl_err_str);
+		if (curl_err && (flags & JSON_RPC_QUIET_404) && http_rc == 404)
+			*curl_err = CURLE_OK;
+		goto err_out;
+	}
+
+	/* If X-Stratum was found, activate Stratum */
+	if (want_stratum && hi.stratum_url &&
+	    !strncasecmp(hi.stratum_url, "stratum+tcp://", 14)) {
+		have_stratum = true;
+		tq_push(thr_info[stratum_thr_id].q, hi.stratum_url);
+		hi.stratum_url = NULL;
+	}
+
+	/* If X-Long-Polling was found, activate long polling */
+	if (!have_longpoll && want_longpoll && hi.lp_path && !have_gbt &&
+	    allow_getwork && !have_stratum) {
+		have_longpoll = true;
+		tq_push(thr_info[longpoll_thr_id].q, hi.lp_path);
+		hi.lp_path = NULL;
+	}
+
+	if (!all_data.buf) {
+		applog(LOG_ERR, "Empty data received in json_rpc_call.");
+		goto err_out;
+	}
+
+	json_buf = hack_json_numbers(all_data.buf);
+	errno = 0; /* needed for Jansson < 2.1 */
+	val = JSON_LOADS(json_buf, &err);
+	free(json_buf);
+	if (!val) {
+		applog(LOG_ERR, "JSON decode failed(%d): %s", err.line, err.text);
+		goto err_out;
+	}
+
+	if (opt_protocol) {
+		char *s = json_dumps(val, JSON_INDENT(3));
+		applog(LOG_DEBUG, "JSON protocol response:\n%s", s);
+		free(s);
+	}
+
+	/* JSON-RPC valid response returns a 'result' and a null 'error'. */
+	res_val = json_object_get(val, "result");
+	err_val = json_object_get(val, "error");
+
+	if (!res_val || (err_val && !json_is_null(err_val))) {
+		char *s;
+
+		if (err_val)
+			s = json_dumps(err_val, JSON_INDENT(3));
+		else
+			s = strdup("(unknown reason)");
+
+		applog(LOG_ERR, "JSON-RPC call failed: %s", s);
+
+		free(s);
+
+		goto err_out;
+	}
+
+	if (hi.reason)
+		json_object_set_new(val, "reject-reason", json_string(hi.reason));
+
+	databuf_free(&all_data);
+	curl_slist_free_all(headers);
+	curl_easy_reset(curl);
+	return val;
+
+err_out:
+	free(hi.lp_path);
+	free(hi.reason);
+	free(hi.stratum_url);
+	databuf_free(&all_data);
+	curl_slist_free_all(headers);
+	curl_easy_reset(curl);
+	return NULL;
+}
+
+void memrev(unsigned char *p, size_t len)
+{
+	unsigned char c, *q;
+	for (q = p + len - 1; p < q; p++, q--) {
+		c = *p;
+		*p = *q;
+		*q = c;
+	}
+}
+
+void bin2hex(char *s, const unsigned char *p, size_t len)
+{
+	int i;
+	for (i = 0; i < len; i++)
+		sprintf(s + (i * 2), "%02x", (unsigned int) p[i]);
+}
+
+char *abin2hex(const unsigned char *p, size_t len)
+{
+	char *s = malloc((len * 2) + 1);
+	if (!s)
+		return NULL;
+	bin2hex(s, p, len);
+	return s;
+}
+
+bool hex2bin(unsigned char *p, const char *hexstr, size_t len)
+{
+	char hex_byte[3];
+	char *ep;
+
+	hex_byte[2] = '\0';
+
+	while (*hexstr && len) {
+		if (!hexstr[1]) {
+			applog(LOG_ERR, "hex2bin str truncated");
+			return false;
+		}
+		hex_byte[0] = hexstr[0];
+		hex_byte[1] = hexstr[1];
+		*p = (unsigned char) strtol(hex_byte, &ep, 16);
+		if (*ep) {
+			applog(LOG_ERR, "hex2bin failed on '%s'", hex_byte);
+			return false;
+		}
+		p++;
+		hexstr += 2;
+		len--;
+	}
+
+	return (len == 0 && *hexstr == 0) ? true : false;
+}
+
+int varint_encode(unsigned char *p, uint64_t n)
+{
+	int i;
+	if (n < 0xfd) {
+		p[0] = n;
+		return 1;
+	}
+	if (n <= 0xffff) {
+		p[0] = 0xfd;
+		p[1] = n & 0xff;
+		p[2] = n >> 8;
+		return 3;
+	}
+	if (n <= 0xffffffff) {
+		p[0] = 0xfe;
+		for (i = 1; i < 5; i++) {
+			p[i] = n & 0xff;
+			n >>= 8;
+		}
+		return 5;
+	}
+	p[0] = 0xff;
+	for (i = 1; i < 9; i++) {
+		p[i] = n & 0xff;
+		n >>= 8;
+	}
+	return 9;
+}
+
+static const char b58digits[] = "123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz";
+
+static bool b58dec(unsigned char *bin, size_t binsz, const char *b58)
+{
+	size_t i, j;
+	uint64_t t;
+	uint32_t c;
+	uint32_t *outi;
+	size_t outisz = (binsz + 3) / 4;
+	int rem = binsz % 4;
+	uint32_t remmask = 0xffffffff << (8 * rem);
+	size_t b58sz = strlen(b58);
+	bool rc = false;
+
+	outi = calloc(outisz, sizeof(*outi));
+
+	for (i = 0; i < b58sz; ++i) {
+		for (c = 0; b58digits[c] != b58[i]; c++)
+			if (!b58digits[c])
+				goto out;
+		for (j = outisz; j--; ) {
+			t = (uint64_t)outi[j] * 58 + c;
+			c = t >> 32;
+			outi[j] = t & 0xffffffff;
+		}
+		if (c || outi[0] & remmask)
+			goto out;
+	}
+
+	j = 0;
+	switch (rem) {
+		case 3:
+			*(bin++) = (outi[0] >> 16) & 0xff;
+		case 2:
+			*(bin++) = (outi[0] >> 8) & 0xff;
+		case 1:
+			*(bin++) = outi[0] & 0xff;
+			++j;
+		default:
+			break;
+	}
+	for (; j < outisz; ++j) {
+		be32enc((uint32_t *)bin, outi[j]);
+		bin += sizeof(uint32_t);
+	}
+
+	rc = true;
+out:
+	free(outi);
+	return rc;
+}
+
+static int b58check(unsigned char *bin, size_t binsz, const char *b58)
+{
+	unsigned char buf[32];
+	int i;
+
+	sha256d(buf, bin, binsz - 4);
+	if (memcmp(&bin[binsz - 4], buf, 4))
+		return -1;
+
+	/* Check number of zeros is correct AFTER verifying checksum
+	 * (to avoid possibility of accessing the string beyond the end) */
+	for (i = 0; bin[i] == '\0' && b58[i] == '1'; ++i);
+	if (bin[i] == '\0' || b58[i] == '1')
+		return -3;
+
+	return bin[0];
+}
+
+size_t address_to_script(unsigned char *out, size_t outsz, const char *addr)
+{
+	unsigned char addrbin[25];
+	int addrver;
+	size_t rv;
+
+	if (!b58dec(addrbin, sizeof(addrbin), addr))
+		return 0;
+	addrver = b58check(addrbin, sizeof(addrbin), addr);
+	if (addrver < 0)
+		return 0;
+	switch (addrver) {
+		case 5:    /* Bitcoin script hash */
+		case 196:  /* Testnet script hash */
+			if (outsz < (rv = 23))
+				return rv;
+			out[ 0] = 0xa9;  /* OP_HASH160 */
+			out[ 1] = 0x14;  /* push 20 bytes */
+			memcpy(&out[2], &addrbin[1], 20);
+			out[22] = 0x87;  /* OP_EQUAL */
+			return rv;
+		default:
+			if (outsz < (rv = 25))
+				return rv;
+			out[ 0] = 0x76;  /* OP_DUP */
+			out[ 1] = 0xa9;  /* OP_HASH160 */
+			out[ 2] = 0x14;  /* push 20 bytes */
+			memcpy(&out[3], &addrbin[1], 20);
+			out[23] = 0x88;  /* OP_EQUALVERIFY */
+			out[24] = 0xac;  /* OP_CHECKSIG */
+			return rv;
+	}
+}
+
+/* Subtract the `struct timeval' values X and Y,
+   storing the result in RESULT.
+   Return 1 if the difference is negative, otherwise 0.  */
+int timeval_subtract(struct timeval *result, struct timeval *x,
+	struct timeval *y)
+{
+	/* Perform the carry for the later subtraction by updating Y. */
+	if (x->tv_usec < y->tv_usec) {
+		int nsec = (y->tv_usec - x->tv_usec) / 1000000 + 1;
+		y->tv_usec -= 1000000 * nsec;
+		y->tv_sec += nsec;
+	}
+	if (x->tv_usec - y->tv_usec > 1000000) {
+		int nsec = (x->tv_usec - y->tv_usec) / 1000000;
+		y->tv_usec += 1000000 * nsec;
+		y->tv_sec -= nsec;
+	}
+
+	/* Compute the time remaining to wait.
+	 * `tv_usec' is certainly positive. */
+	result->tv_sec = x->tv_sec - y->tv_sec;
+	result->tv_usec = x->tv_usec - y->tv_usec;
+
+	/* Return 1 if result is negative. */
+	return x->tv_sec < y->tv_sec;
+}
+
+bool fulltest(const uint32_t *hash, const uint32_t *target)
+{
+	int i;
+	bool rc = true;
+	
+	for (i = 7; i >= 0; i--) {
+		if (hash[i] > target[i]) {
+			rc = false;
+			break;
+		}
+		if (hash[i] < target[i]) {
+			rc = true;
+			break;
+		}
+	}
+
+	if (opt_debug) {
+		uint32_t hash_be[8], target_be[8];
+		char hash_str[65], target_str[65];
+		
+		for (i = 0; i < 8; i++) {
+			be32enc(hash_be + i, hash[7 - i]);
+			be32enc(target_be + i, target[7 - i]);
+		}
+		bin2hex(hash_str, (unsigned char *)hash_be, 32);
+		bin2hex(target_str, (unsigned char *)target_be, 32);
+
+		applog(LOG_DEBUG, "DEBUG: %s\nHash:   %s\nTarget: %s",
+			rc ? "hash <= target"
+			   : "hash > target (false positive)",
+			hash_str,
+			target_str);
+	}
+
+	return rc;
+}
+
+void diff_to_target(uint32_t *target, double diff)
+{
+	uint64_t m;
+	int k;
+	
+	for (k = 6; k > 0 && diff > 1.0; k--)
+		diff /= 4294967296.0;
+	m = 4294901760.0 / diff;
+	if (m == 0 && k == 6)
+		memset(target, 0xff, 32);
+	else {
+		memset(target, 0, 32);
+		target[k] = (uint32_t)m;
+		target[k + 1] = (uint32_t)(m >> 32);
+	}
+}
+
+#ifdef WIN32
+#define socket_blocks() (WSAGetLastError() == WSAEWOULDBLOCK)
+#else
+#define socket_blocks() (errno == EAGAIN || errno == EWOULDBLOCK)
+#endif
+
+static bool send_line(struct stratum_ctx *sctx, char *s)
+{
+	ssize_t len, sent = 0;
+	
+	len = strlen(s);
+	s[len++] = '\n';
+
+	while (len > 0) {
+		struct timeval timeout = {0, 0};
+		ssize_t n;
+		fd_set wd;
+
+		FD_ZERO(&wd);
+		FD_SET(sctx->sock, &wd);
+		if (select(sctx->sock + 1, NULL, &wd, NULL, &timeout) < 1)
+			return false;
+#if LIBCURL_VERSION_NUM >= 0x071202
+		CURLcode rc = curl_easy_send(sctx->curl, s + sent, len, (size_t *)&n);
+		if (rc != CURLE_OK) {
+			if (rc != CURLE_AGAIN)
+#else
+		n = send(sctx->sock, s + sent, len, 0);
+		if (n < 0) {
+			if (!socket_blocks())
+#endif
+				return false;
+			n = 0;
+		}
+		sent += n;
+		len -= n;
+	}
+
+	return true;
+}
+
+bool stratum_send_line(struct stratum_ctx *sctx, char *s)
+{
+	bool ret = false;
+
+	if (opt_protocol)
+		applog(LOG_DEBUG, "> %s", s);
+
+	pthread_mutex_lock(&sctx->sock_lock);
+	ret = send_line(sctx, s);
+	pthread_mutex_unlock(&sctx->sock_lock);
+
+	return ret;
+}
+
+static bool socket_full(curl_socket_t sock, int timeout)
+{
+	struct timeval tv;
+	fd_set rd;
+
+	FD_ZERO(&rd);
+	FD_SET(sock, &rd);
+	tv.tv_sec = timeout;
+	tv.tv_usec = 0;
+	if (select(sock + 1, &rd, NULL, NULL, &tv) > 0)
+		return true;
+	return false;
+}
+
+bool stratum_socket_full(struct stratum_ctx *sctx, int timeout)
+{
+	return strlen(sctx->sockbuf) || socket_full(sctx->sock, timeout);
+}
+
+#define RBUFSIZE 2048
+#define RECVSIZE (RBUFSIZE - 4)
+
+static void stratum_buffer_append(struct stratum_ctx *sctx, const char *s)
+{
+	size_t old, new;
+
+	old = strlen(sctx->sockbuf);
+	new = old + strlen(s) + 1;
+	if (new >= sctx->sockbuf_size) {
+		sctx->sockbuf_size = new + (RBUFSIZE - (new % RBUFSIZE));
+		sctx->sockbuf = realloc(sctx->sockbuf, sctx->sockbuf_size);
+	}
+	strcpy(sctx->sockbuf + old, s);
+}
+
+char *stratum_recv_line(struct stratum_ctx *sctx)
+{
+	ssize_t len, buflen;
+	char *tok, *sret = NULL;
+
+	if (!strstr(sctx->sockbuf, "\n")) {
+		bool ret = true;
+		time_t rstart;
+
+		time(&rstart);
+		if (!socket_full(sctx->sock, 60)) {
+			applog(LOG_ERR, "stratum_recv_line timed out");
+			goto out;
+		}
+		do {
+			char s[RBUFSIZE];
+			ssize_t n;
+
+			memset(s, 0, RBUFSIZE);
+#if LIBCURL_VERSION_NUM >= 0x071202
+			CURLcode rc = curl_easy_recv(sctx->curl, s, RECVSIZE, (size_t *)&n);
+			if (rc == CURLE_OK && !n) {
+				ret = false;
+				break;
+			}
+			if (rc != CURLE_OK) {
+				if (rc != CURLE_AGAIN || !socket_full(sctx->sock, 1)) {
+#else
+			n = recv(sctx->sock, s, RECVSIZE, 0);
+			if (!n) {
+				ret = false;
+				break;
+			}
+			if (n < 0) {
+				if (!socket_blocks() || !socket_full(sctx->sock, 1)) {
+#endif
+					ret = false;
+					break;
+				}
+			} else
+				stratum_buffer_append(sctx, s);
+		} while (time(NULL) - rstart < 60 && !strstr(sctx->sockbuf, "\n"));
+
+		if (!ret) {
+			applog(LOG_ERR, "stratum_recv_line failed");
+			goto out;
+		}
+	}
+
+	buflen = strlen(sctx->sockbuf);
+	tok = strtok(sctx->sockbuf, "\n");
+	if (!tok) {
+		applog(LOG_ERR, "stratum_recv_line failed to parse a newline-terminated string");
+		goto out;
+	}
+	sret = strdup(tok);
+	len = strlen(sret);
+
+	if (buflen > len + 1)
+		memmove(sctx->sockbuf, sctx->sockbuf + len + 1, buflen - len + 1);
+	else
+		sctx->sockbuf[0] = '\0';
+
+out:
+	if (sret && opt_protocol)
+		applog(LOG_DEBUG, "< %s", sret);
+	return sret;
+}
+
+#if LIBCURL_VERSION_NUM >= 0x071101
+static curl_socket_t opensocket_grab_cb(void *clientp, curlsocktype purpose,
+	struct curl_sockaddr *addr)
+{
+	curl_socket_t *sock = clientp;
+	*sock = socket(addr->family, addr->socktype, addr->protocol);
+	return *sock;
+}
+#endif
+
+bool stratum_connect(struct stratum_ctx *sctx, const char *url)
+{
+	CURL *curl;
+	int rc;
+
+	pthread_mutex_lock(&sctx->sock_lock);
+	if (sctx->curl)
+		curl_easy_cleanup(sctx->curl);
+	sctx->curl = curl_easy_init();
+	if (!sctx->curl) {
+		applog(LOG_ERR, "CURL initialization failed");
+		pthread_mutex_unlock(&sctx->sock_lock);
+		return false;
+	}
+	curl = sctx->curl;
+	if (!sctx->sockbuf) {
+		sctx->sockbuf = calloc(RBUFSIZE, 1);
+		sctx->sockbuf_size = RBUFSIZE;
+	}
+	sctx->sockbuf[0] = '\0';
+	pthread_mutex_unlock(&sctx->sock_lock);
+
+	if (url != sctx->url) {
+		free(sctx->url);
+		sctx->url = strdup(url);
+	}
+	free(sctx->curl_url);
+	sctx->curl_url = malloc(strlen(url));
+	sprintf(sctx->curl_url, "http%s", url + 11);
+
+	if (opt_protocol)
+		curl_easy_setopt(curl, CURLOPT_VERBOSE, 1);
+	curl_easy_setopt(curl, CURLOPT_URL, sctx->curl_url);
+	if (opt_cert)
+		curl_easy_setopt(curl, CURLOPT_CAINFO, opt_cert);
+	curl_easy_setopt(curl, CURLOPT_FRESH_CONNECT, 1);
+	curl_easy_setopt(curl, CURLOPT_CONNECTTIMEOUT, 30);
+	curl_easy_setopt(curl, CURLOPT_ERRORBUFFER, sctx->curl_err_str);
+	curl_easy_setopt(curl, CURLOPT_NOSIGNAL, 1);
+	curl_easy_setopt(curl, CURLOPT_TCP_NODELAY, 1);
+	if (opt_proxy) {
+		curl_easy_setopt(curl, CURLOPT_PROXY, opt_proxy);
+		curl_easy_setopt(curl, CURLOPT_PROXYTYPE, opt_proxy_type);
+	}
+	curl_easy_setopt(curl, CURLOPT_HTTPPROXYTUNNEL, 1);
+#if LIBCURL_VERSION_NUM >= 0x070f06
+	curl_easy_setopt(curl, CURLOPT_SOCKOPTFUNCTION, sockopt_keepalive_cb);
+#endif
+#if LIBCURL_VERSION_NUM >= 0x071101
+	curl_easy_setopt(curl, CURLOPT_OPENSOCKETFUNCTION, opensocket_grab_cb);
+	curl_easy_setopt(curl, CURLOPT_OPENSOCKETDATA, &sctx->sock);
+#endif
+	curl_easy_setopt(curl, CURLOPT_CONNECT_ONLY, 1);
+
+	rc = curl_easy_perform(curl);
+	if (rc) {
+		applog(LOG_ERR, "Stratum connection failed: %s", sctx->curl_err_str);
+		curl_easy_cleanup(curl);
+		sctx->curl = NULL;
+		return false;
+	}
+
+#if LIBCURL_VERSION_NUM < 0x071101
+	/* CURLINFO_LASTSOCKET is broken on Win64; only use it as a last resort */
+	curl_easy_getinfo(curl, CURLINFO_LASTSOCKET, (long *)&sctx->sock);
+#endif
+
+	return true;
+}
+
+void stratum_disconnect(struct stratum_ctx *sctx)
+{
+	pthread_mutex_lock(&sctx->sock_lock);
+	if (sctx->curl) {
+		curl_easy_cleanup(sctx->curl);
+		sctx->curl = NULL;
+		sctx->sockbuf[0] = '\0';
+	}
+	pthread_mutex_unlock(&sctx->sock_lock);
+}
+
+static const char *get_stratum_session_id(json_t *val)
+{
+	json_t *arr_val;
+	int i, n;
+
+	arr_val = json_array_get(val, 0);
+	if (!arr_val || !json_is_array(arr_val))
+		return NULL;
+	n = json_array_size(arr_val);
+	for (i = 0; i < n; i++) {
+		const char *notify;
+		json_t *arr = json_array_get(arr_val, i);
+
+		if (!arr || !json_is_array(arr))
+			break;
+		notify = json_string_value(json_array_get(arr, 0));
+		if (!notify)
+			continue;
+		if (!strcasecmp(notify, "mining.notify"))
+			return json_string_value(json_array_get(arr, 1));
+	}
+	return NULL;
+}
+
+bool stratum_subscribe(struct stratum_ctx *sctx)
+{
+	char *s, *sret = NULL;
+	const char *sid, *xnonce1;
+	int xn2_size;
+	json_t *val = NULL, *res_val, *err_val;
+	json_error_t err;
+	bool ret = false, retry = false;
+
+start:
+	s = malloc(128 + (sctx->session_id ? strlen(sctx->session_id) : 0));
+	if (retry)
+		sprintf(s, "{\"id\": 1, \"method\": \"mining.subscribe\", \"params\": []}");
+	else if (sctx->session_id)
+		sprintf(s, "{\"id\": 1, \"method\": \"mining.subscribe\", \"params\": [\"" USER_AGENT "\", \"%s\"]}", sctx->session_id);
+	else
+		sprintf(s, "{\"id\": 1, \"method\": \"mining.subscribe\", \"params\": [\"" USER_AGENT "\"]}");
+
+	if (!stratum_send_line(sctx, s)) {
+		applog(LOG_ERR, "stratum_subscribe send failed");
+		goto out;
+	}
+
+	if (!socket_full(sctx->sock, 30)) {
+		applog(LOG_ERR, "stratum_subscribe timed out");
+		goto out;
+	}
+
+	sret = stratum_recv_line(sctx);
+	if (!sret)
+		goto out;
+
+	val = JSON_LOADS(sret, &err);
+	free(sret);
+	if (!val) {
+		applog(LOG_ERR, "JSON decode failed(%d): %s", err.line, err.text);
+		goto out;
+	}
+
+	res_val = json_object_get(val, "result");
+	err_val = json_object_get(val, "error");
+
+	if (!res_val || json_is_null(res_val) ||
+	    (err_val && !json_is_null(err_val))) {
+		if (opt_debug || retry) {
+			free(s);
+			if (err_val)
+				s = json_dumps(err_val, JSON_INDENT(3));
+			else
+				s = strdup("(unknown reason)");
+			applog(LOG_ERR, "JSON-RPC call failed: %s", s);
+		}
+		goto out;
+	}
+
+	sid = get_stratum_session_id(res_val);
+	if (opt_debug && !sid)
+		applog(LOG_DEBUG, "Failed to get Stratum session id");
+	xnonce1 = json_string_value(json_array_get(res_val, 1));
+	if (!xnonce1) {
+		applog(LOG_ERR, "Failed to get extranonce1");
+		goto out;
+	}
+	xn2_size = json_integer_value(json_array_get(res_val, 2));
+	if (!xn2_size) {
+		applog(LOG_ERR, "Failed to get extranonce2_size");
+		goto out;
+	}
+	if (xn2_size < 0 || xn2_size > 100) {
+		applog(LOG_ERR, "Invalid value of extranonce2_size");
+		goto out;
+	}
+
+	pthread_mutex_lock(&sctx->work_lock);
+	free(sctx->session_id);
+	free(sctx->xnonce1);
+	sctx->session_id = sid ? strdup(sid) : NULL;
+	sctx->xnonce1_size = strlen(xnonce1) / 2;
+	sctx->xnonce1 = malloc(sctx->xnonce1_size);
+	hex2bin(sctx->xnonce1, xnonce1, sctx->xnonce1_size);
+	sctx->xnonce2_size = xn2_size;
+	sctx->next_diff = 1.0;
+	pthread_mutex_unlock(&sctx->work_lock);
+
+	if (opt_debug && sid)
+		applog(LOG_DEBUG, "Stratum session id: %s", sctx->session_id);
+
+	ret = true;
+
+out:
+	free(s);
+	if (val)
+		json_decref(val);
+
+	if (!ret) {
+		if (sret && !retry) {
+			retry = true;
+			goto start;
+		}
+	}
+
+	return ret;
+}
+
+bool stratum_authorize(struct stratum_ctx *sctx, const char *user, const char *pass)
+{
+	json_t *val = NULL, *res_val, *err_val;
+	char *s, *sret;
+	json_error_t err;
+	bool ret = false;
+
+	s = malloc(80 + strlen(user) + strlen(pass));
+	sprintf(s, "{\"id\": 2, \"method\": \"mining.authorize\", \"params\": [\"%s\", \"%s\"]}",
+	        user, pass);
+
+	if (!stratum_send_line(sctx, s))
+		goto out;
+
+	while (1) {
+		sret = stratum_recv_line(sctx);
+		if (!sret)
+			goto out;
+		if (!stratum_handle_method(sctx, sret))
+			break;
+		free(sret);
+	}
+
+	val = JSON_LOADS(sret, &err);
+	free(sret);
+	if (!val) {
+		applog(LOG_ERR, "JSON decode failed(%d): %s", err.line, err.text);
+		goto out;
+	}
+
+	res_val = json_object_get(val, "result");
+	err_val = json_object_get(val, "error");
+
+	if (!res_val || json_is_false(res_val) ||
+	    (err_val && !json_is_null(err_val)))  {
+		applog(LOG_ERR, "Stratum authentication failed");
+		goto out;
+	}
+
+	ret = true;
+
+out:
+	free(s);
+	if (val)
+		json_decref(val);
+
+	return ret;
+}
+
+static bool stratum_notify(struct stratum_ctx *sctx, json_t *params)
+{
+	const char *job_id, *prevhash, *coinb1, *coinb2, *version, *nbits, *ntime;
+	size_t coinb1_size, coinb2_size;
+	bool clean, ret = false;
+	int merkle_count, i;
+	json_t *merkle_arr;
+	unsigned char **merkle;
+
+	job_id = json_string_value(json_array_get(params, 0));
+	prevhash = json_string_value(json_array_get(params, 1));
+	coinb1 = json_string_value(json_array_get(params, 2));
+	coinb2 = json_string_value(json_array_get(params, 3));
+	merkle_arr = json_array_get(params, 4);
+	if (!merkle_arr || !json_is_array(merkle_arr))
+		goto out;
+	merkle_count = json_array_size(merkle_arr);
+	version = json_string_value(json_array_get(params, 5));
+	nbits = json_string_value(json_array_get(params, 6));
+	ntime = json_string_value(json_array_get(params, 7));
+	clean = json_is_true(json_array_get(params, 8));
+
+	if (!job_id || !prevhash || !coinb1 || !coinb2 || !version || !nbits || !ntime ||
+	    strlen(prevhash) != 64 || strlen(version) != 8 ||
+	    strlen(nbits) != 8 || strlen(ntime) != 8) {
+		applog(LOG_ERR, "Stratum notify: invalid parameters");
+		goto out;
+	}
+	merkle = malloc(merkle_count * sizeof(char *));
+	for (i = 0; i < merkle_count; i++) {
+		const char *s = json_string_value(json_array_get(merkle_arr, i));
+		if (!s || strlen(s) != 64) {
+			while (i--)
+				free(merkle[i]);
+			free(merkle);
+			applog(LOG_ERR, "Stratum notify: invalid Merkle branch");
+			goto out;
+		}
+		merkle[i] = malloc(32);
+		hex2bin(merkle[i], s, 32);
+	}
+
+	pthread_mutex_lock(&sctx->work_lock);
+
+	coinb1_size = strlen(coinb1) / 2;
+	coinb2_size = strlen(coinb2) / 2;
+	sctx->job.coinbase_size = coinb1_size + sctx->xnonce1_size +
+	                          sctx->xnonce2_size + coinb2_size;
+	sctx->job.coinbase = realloc(sctx->job.coinbase, sctx->job.coinbase_size);
+	sctx->job.xnonce2 = sctx->job.coinbase + coinb1_size + sctx->xnonce1_size;
+	hex2bin(sctx->job.coinbase, coinb1, coinb1_size);
+	memcpy(sctx->job.coinbase + coinb1_size, sctx->xnonce1, sctx->xnonce1_size);
+	if (!sctx->job.job_id || strcmp(sctx->job.job_id, job_id))
+		memset(sctx->job.xnonce2, 0, sctx->xnonce2_size);
+	hex2bin(sctx->job.xnonce2 + sctx->xnonce2_size, coinb2, coinb2_size);
+
+	free(sctx->job.job_id);
+	sctx->job.job_id = strdup(job_id);
+	hex2bin(sctx->job.prevhash, prevhash, 32);
+
+	for (i = 0; i < sctx->job.merkle_count; i++)
+		free(sctx->job.merkle[i]);
+	free(sctx->job.merkle);
+	sctx->job.merkle = merkle;
+	sctx->job.merkle_count = merkle_count;
+
+	hex2bin(sctx->job.version, version, 4);
+	hex2bin(sctx->job.nbits, nbits, 4);
+	hex2bin(sctx->job.ntime, ntime, 4);
+	sctx->job.clean = clean;
+
+	sctx->job.diff = sctx->next_diff;
+
+	pthread_mutex_unlock(&sctx->work_lock);
+
+	ret = true;
+
+out:
+	return ret;
+}
+
+static bool stratum_set_difficulty(struct stratum_ctx *sctx, json_t *params)
+{
+	double diff;
+
+	diff = json_number_value(json_array_get(params, 0));
+	if (diff == 0)
+		return false;
+
+	pthread_mutex_lock(&sctx->work_lock);
+	sctx->next_diff = diff;
+	pthread_mutex_unlock(&sctx->work_lock);
+
+	if (opt_debug)
+		applog(LOG_DEBUG, "Stratum difficulty set to %g", diff);
+
+	return true;
+}
+
+static bool stratum_reconnect(struct stratum_ctx *sctx, json_t *params)
+{
+	json_t *port_val;
+	char *url;
+	const char *host;
+	int port;
+
+	host = json_string_value(json_array_get(params, 0));
+	port_val = json_array_get(params, 1);
+	if (json_is_string(port_val))
+		port = atoi(json_string_value(port_val));
+	else
+		port = json_integer_value(port_val);
+	if (!host || !port)
+		return false;
+
+	url = malloc(32 + strlen(host));
+	strncpy(url, sctx->url, 15);
+	sprintf(strstr(url, "://") + 3, "%s:%d", host, port);
+
+	if (!opt_redirect) {
+		applog(LOG_INFO, "Ignoring request to reconnect to %s", url);
+		free(url);
+		return true;
+	}
+
+	applog(LOG_NOTICE, "Server requested reconnection to %s", url);
+
+	free(sctx->url);
+	sctx->url = url;
+	stratum_disconnect(sctx);
+
+	return true;
+}
+
+static bool stratum_get_version(struct stratum_ctx *sctx, json_t *id)
+{
+	char *s;
+	json_t *val;
+	bool ret;
+	
+	if (!id || json_is_null(id))
+		return false;
+
+	val = json_object();
+	json_object_set(val, "id", id);
+	json_object_set_new(val, "error", json_null());
+	json_object_set_new(val, "result", json_string(USER_AGENT));
+	s = json_dumps(val, 0);
+	ret = stratum_send_line(sctx, s);
+	json_decref(val);
+	free(s);
+
+	return ret;
+}
+
+static bool stratum_show_message(struct stratum_ctx *sctx, json_t *id, json_t *params)
+{
+	char *s;
+	json_t *val;
+	bool ret;
+
+	val = json_array_get(params, 0);
+	if (val)
+		applog(LOG_NOTICE, "MESSAGE FROM SERVER: %s", json_string_value(val));
+	
+	if (!id || json_is_null(id))
+		return true;
+
+	val = json_object();
+	json_object_set(val, "id", id);
+	json_object_set_new(val, "error", json_null());
+	json_object_set_new(val, "result", json_true());
+	s = json_dumps(val, 0);
+	ret = stratum_send_line(sctx, s);
+	json_decref(val);
+	free(s);
+
+	return ret;
+}
+
+bool stratum_handle_method(struct stratum_ctx *sctx, const char *s)
+{
+	json_t *val, *id, *params;
+	json_error_t err;
+	const char *method;
+	bool ret = false;
+
+	val = JSON_LOADS(s, &err);
+	if (!val) {
+		applog(LOG_ERR, "JSON decode failed(%d): %s", err.line, err.text);
+		goto out;
+	}
+
+	method = json_string_value(json_object_get(val, "method"));
+	if (!method)
+		goto out;
+	id = json_object_get(val, "id");
+	params = json_object_get(val, "params");
+
+	if (!strcasecmp(method, "mining.notify")) {
+		ret = stratum_notify(sctx, params);
+		goto out;
+	}
+	if (!strcasecmp(method, "mining.set_difficulty")) {
+		ret = stratum_set_difficulty(sctx, params);
+		goto out;
+	}
+	if (!strcasecmp(method, "client.reconnect")) {
+		ret = stratum_reconnect(sctx, params);
+		goto out;
+	}
+	if (!strcasecmp(method, "client.get_version")) {
+		ret = stratum_get_version(sctx, id);
+		goto out;
+	}
+	if (!strcasecmp(method, "client.show_message")) {
+		ret = stratum_show_message(sctx, id, params);
+		goto out;
+	}
+
+out:
+	if (val)
+		json_decref(val);
+
+	return ret;
+}
+
+struct thread_q *tq_new(void)
+{
+	struct thread_q *tq;
+
+	tq = calloc(1, sizeof(*tq));
+	if (!tq)
+		return NULL;
+
+	INIT_LIST_HEAD(&tq->q);
+	pthread_mutex_init(&tq->mutex, NULL);
+	pthread_cond_init(&tq->cond, NULL);
+
+	return tq;
+}
+
+void tq_free(struct thread_q *tq)
+{
+	struct tq_ent *ent, *iter;
+
+	if (!tq)
+		return;
+
+	list_for_each_entry_safe(ent, iter, &tq->q, q_node, struct tq_ent) {
+		list_del(&ent->q_node);
+		free(ent);
+	}
+
+	pthread_cond_destroy(&tq->cond);
+	pthread_mutex_destroy(&tq->mutex);
+
+	memset(tq, 0, sizeof(*tq));	/* poison */
+	free(tq);
+}
+
+static void tq_freezethaw(struct thread_q *tq, bool frozen)
+{
+	pthread_mutex_lock(&tq->mutex);
+
+	tq->frozen = frozen;
+
+	pthread_cond_signal(&tq->cond);
+	pthread_mutex_unlock(&tq->mutex);
+}
+
+void tq_freeze(struct thread_q *tq)
+{
+	tq_freezethaw(tq, true);
+}
+
+void tq_thaw(struct thread_q *tq)
+{
+	tq_freezethaw(tq, false);
+}
+
+bool tq_push(struct thread_q *tq, void *data)
+{
+	struct tq_ent *ent;
+	bool rc = true;
+
+	ent = calloc(1, sizeof(*ent));
+	if (!ent)
+		return false;
+
+	ent->data = data;
+	INIT_LIST_HEAD(&ent->q_node);
+
+	pthread_mutex_lock(&tq->mutex);
+
+	if (!tq->frozen) {
+		list_add_tail(&ent->q_node, &tq->q);
+	} else {
+		free(ent);
+		rc = false;
+	}
+
+	pthread_cond_signal(&tq->cond);
+	pthread_mutex_unlock(&tq->mutex);
+
+	return rc;
+}
+
+void *tq_pop(struct thread_q *tq, const struct timespec *abstime)
+{
+	struct tq_ent *ent;
+	void *rval = NULL;
+	int rc;
+
+	pthread_mutex_lock(&tq->mutex);
+
+	if (!list_empty(&tq->q))
+		goto pop;
+
+	if (abstime)
+		rc = pthread_cond_timedwait(&tq->cond, &tq->mutex, abstime);
+	else
+		rc = pthread_cond_wait(&tq->cond, &tq->mutex);
+	if (rc)
+		goto out;
+	if (list_empty(&tq->q))
+		goto out;
+
+pop:
+	ent = list_entry(tq->q.next, struct tq_ent, q_node);
+	rval = ent->data;
+
+	list_del(&ent->q_node);
+	free(ent);
+
+out:
+	pthread_mutex_unlock(&tq->mutex);
+	return rval;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-best.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-best.c
new file mode 100644
index 0000000..4e83621
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-best.c
@@ -0,0 +1,5 @@
+#ifdef __SSE2__
+#include "yescrypt-simd.c"
+#else
+#include "yescrypt-opt.c"
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-opt.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-opt.c
new file mode 100644
index 0000000..5313450
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-opt.c
@@ -0,0 +1,962 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2013,2014 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ */
+//
+//
+// ottylab update history <ottylab@gmail.com>
+//
+// 2018/03/20 @001 Support KitKat
+//
+
+#ifdef __i386__
+#warning "This implementation does not use SIMD, and thus it runs a lot slower than the SIMD-enabled implementation. Enable at least SSE2 in the C compiler and use yescrypt-best.c instead unless you're building this SIMD-less implementation on purpose (portability to older CPUs or testing)."
+#elif defined(__x86_64__)
+#warning "This implementation does not use SIMD, and thus it runs a lot slower than the SIMD-enabled implementation. Use yescrypt-best.c instead unless you're building this SIMD-less implementation on purpose (for testing only)."
+#endif
+
+#include <errno.h>
+#include <stdint.h>
+#include <stdlib.h>
+
+//@001 start
+#ifdef __KITKAT__
+#include <sys/limits.h>
+#endif // __KITKAT__
+//@001 end
+
+#include "sha256.h"
+#include "sysendian.h"
+
+#include "yescrypt.h"
+
+#include "yescrypt-platform.c"
+
+static inline void
+blkcpy(uint64_t * dest, const uint64_t * src, size_t count)
+{
+	do {
+		*dest++ = *src++; *dest++ = *src++;
+		*dest++ = *src++; *dest++ = *src++;
+	} while (count -= 4);
+}
+
+static inline void
+blkxor(uint64_t * dest, const uint64_t * src, size_t count)
+{
+	do {
+		*dest++ ^= *src++; *dest++ ^= *src++;
+		*dest++ ^= *src++; *dest++ ^= *src++;
+	} while (count -= 4);
+}
+
+typedef union {
+	uint32_t w[16];
+	uint64_t d[8];
+} salsa20_blk_t;
+
+static inline void
+salsa20_simd_shuffle(const salsa20_blk_t * Bin, salsa20_blk_t * Bout)
+{
+#define COMBINE(out, in1, in2) \
+	Bout->d[out] = Bin->w[in1 * 2] | ((uint64_t)Bin->w[in2 * 2 + 1] << 32);
+	COMBINE(0, 0, 2)
+	COMBINE(1, 5, 7)
+	COMBINE(2, 2, 4)
+	COMBINE(3, 7, 1)
+	COMBINE(4, 4, 6)
+	COMBINE(5, 1, 3)
+	COMBINE(6, 6, 0)
+	COMBINE(7, 3, 5)
+#undef COMBINE
+}
+
+static inline void
+salsa20_simd_unshuffle(const salsa20_blk_t * Bin, salsa20_blk_t * Bout)
+{
+#define COMBINE(out, in1, in2) \
+	Bout->w[out * 2] = Bin->d[in1]; \
+	Bout->w[out * 2 + 1] = Bin->d[in2] >> 32;
+	COMBINE(0, 0, 6)
+	COMBINE(1, 5, 3)
+	COMBINE(2, 2, 0)
+	COMBINE(3, 7, 5)
+	COMBINE(4, 4, 2)
+	COMBINE(5, 1, 7)
+	COMBINE(6, 6, 4)
+	COMBINE(7, 3, 1)
+#undef COMBINE
+}
+
+/**
+ * salsa20_8(B):
+ * Apply the salsa20/8 core to the provided block.
+ */
+static void
+salsa20_8(uint64_t B[8])
+{
+	size_t i;
+	salsa20_blk_t X;
+#define x X.w
+
+	salsa20_simd_unshuffle((const salsa20_blk_t *)B, &X);
+
+	for (i = 0; i < 8; i += 2) {
+#define R(a,b) (((a) << (b)) | ((a) >> (32 - (b))))
+		/* Operate on columns */
+		x[ 4] ^= R(x[ 0]+x[12], 7);  x[ 8] ^= R(x[ 4]+x[ 0], 9);
+		x[12] ^= R(x[ 8]+x[ 4],13);  x[ 0] ^= R(x[12]+x[ 8],18);
+
+		x[ 9] ^= R(x[ 5]+x[ 1], 7);  x[13] ^= R(x[ 9]+x[ 5], 9);
+		x[ 1] ^= R(x[13]+x[ 9],13);  x[ 5] ^= R(x[ 1]+x[13],18);
+
+		x[14] ^= R(x[10]+x[ 6], 7);  x[ 2] ^= R(x[14]+x[10], 9);
+		x[ 6] ^= R(x[ 2]+x[14],13);  x[10] ^= R(x[ 6]+x[ 2],18);
+
+		x[ 3] ^= R(x[15]+x[11], 7);  x[ 7] ^= R(x[ 3]+x[15], 9);
+		x[11] ^= R(x[ 7]+x[ 3],13);  x[15] ^= R(x[11]+x[ 7],18);
+
+		/* Operate on rows */
+		x[ 1] ^= R(x[ 0]+x[ 3], 7);  x[ 2] ^= R(x[ 1]+x[ 0], 9);
+		x[ 3] ^= R(x[ 2]+x[ 1],13);  x[ 0] ^= R(x[ 3]+x[ 2],18);
+
+		x[ 6] ^= R(x[ 5]+x[ 4], 7);  x[ 7] ^= R(x[ 6]+x[ 5], 9);
+		x[ 4] ^= R(x[ 7]+x[ 6],13);  x[ 5] ^= R(x[ 4]+x[ 7],18);
+
+		x[11] ^= R(x[10]+x[ 9], 7);  x[ 8] ^= R(x[11]+x[10], 9);
+		x[ 9] ^= R(x[ 8]+x[11],13);  x[10] ^= R(x[ 9]+x[ 8],18);
+
+		x[12] ^= R(x[15]+x[14], 7);  x[13] ^= R(x[12]+x[15], 9);
+		x[14] ^= R(x[13]+x[12],13);  x[15] ^= R(x[14]+x[13],18);
+#undef R
+	}
+#undef x
+
+	{
+		salsa20_blk_t Y;
+		salsa20_simd_shuffle(&X, &Y);
+		for (i = 0; i < 16; i += 4) {
+			((salsa20_blk_t *)B)->w[i] += Y.w[i];
+			((salsa20_blk_t *)B)->w[i + 1] += Y.w[i + 1];
+			((salsa20_blk_t *)B)->w[i + 2] += Y.w[i + 2];
+			((salsa20_blk_t *)B)->w[i + 3] += Y.w[i + 3];
+		}
+	}
+}
+
+/**
+ * blockmix_salsa8(Bin, Bout, X, r):
+ * Compute Bout = BlockMix_{salsa20/8, r}(Bin).  The input Bin must be 128r
+ * bytes in length; the output Bout must also be the same size.  The
+ * temporary space X must be 64 bytes.
+ */
+static void
+blockmix_salsa8(const uint64_t * Bin, uint64_t * Bout, uint64_t * X, size_t r)
+{
+	size_t i;
+
+	/* 1: X <-- B_{2r - 1} */
+	blkcpy(X, &Bin[(2 * r - 1) * 8], 8);
+
+	/* 2: for i = 0 to 2r - 1 do */
+	for (i = 0; i < 2 * r; i += 2) {
+		/* 3: X <-- H(X \xor B_i) */
+		blkxor(X, &Bin[i * 8], 8);
+		salsa20_8(X);
+
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		blkcpy(&Bout[i * 4], X, 8);
+
+		/* 3: X <-- H(X \xor B_i) */
+		blkxor(X, &Bin[i * 8 + 8], 8);
+		salsa20_8(X);
+
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		blkcpy(&Bout[i * 4 + r * 8], X, 8);
+	}
+}
+
+/* These are tunable */
+#define S_BITS 8
+#define S_SIMD 2
+#define S_P 4
+#define S_ROUNDS 6
+
+/* Number of S-boxes.  Not tunable, hard-coded in a few places. */
+#define S_N 2
+
+/* Derived values.  Not tunable on their own. */
+#define S_SIZE1 (1 << S_BITS)
+#define S_MASK ((S_SIZE1 - 1) * S_SIMD * 8)
+#define S_MASK2 (((uint64_t)S_MASK << 32) | S_MASK)
+#define S_SIZE_ALL (S_N * S_SIZE1 * S_SIMD)
+#define S_P_SIZE (S_P * S_SIMD)
+#define S_MIN_R ((S_P * S_SIMD + 15) / 16)
+
+/**
+ * pwxform(B):
+ * Transform the provided block using the provided S-boxes.
+ */
+static void
+block_pwxform(uint64_t * B, const uint64_t * S)
+{
+	uint64_t (*X)[S_SIMD] = (uint64_t (*)[S_SIMD])B;
+	const uint8_t *S0 = (const uint8_t *)S;
+	const uint8_t *S1 = (const uint8_t *)(S + S_SIZE1 * S_SIMD);
+	size_t i, j;
+#if S_SIMD > 2
+	size_t k;
+#endif
+
+	for (j = 0; j < S_P; j++) {
+		uint64_t *Xj = X[j];
+		uint64_t x0 = Xj[0];
+#if S_SIMD > 1
+		uint64_t x1 = Xj[1];
+#endif
+
+		for (i = 0; i < S_ROUNDS; i++) {
+			uint64_t x = x0 & S_MASK2;
+			const uint64_t *p0, *p1;
+
+			p0 = (const uint64_t *)(S0 + (uint32_t)x);
+			p1 = (const uint64_t *)(S1 + (x >> 32));
+
+			x0 = (uint64_t)(x0 >> 32) * (uint32_t)x0;
+			x0 += p0[0];
+			x0 ^= p1[0];
+
+#if S_SIMD > 1
+			x1 = (uint64_t)(x1 >> 32) * (uint32_t)x1;
+			x1 += p0[1];
+			x1 ^= p1[1];
+#endif
+
+#if S_SIMD > 2
+			for (k = 2; k < S_SIMD; k++) {
+				x = Xj[k];
+
+				x = (uint64_t)(x >> 32) * (uint32_t)x;
+				x += p0[k];
+				x ^= p1[k];
+
+				Xj[k] = x;
+			}
+#endif
+		}
+
+		Xj[0] = x0;
+#if S_SIMD > 1
+		Xj[1] = x1;
+#endif
+	}
+}
+
+/**
+ * blockmix_pwxform(Bin, Bout, S, r):
+ * Compute Bout = BlockMix_pwxform{salsa20/8, S, r}(Bin).  The input Bin must
+ * be 128r bytes in length; the output Bout must also be the same size.
+ *
+ * S lacks const qualifier to match blockmix_salsa8()'s prototype, which we
+ * need to refer to both functions via the same function pointers.
+ */
+static void
+blockmix_pwxform(const uint64_t * Bin, uint64_t * Bout, uint64_t * S, size_t r)
+{
+	size_t r1, r2, i;
+
+	/* Convert 128-byte blocks to (S_P_SIZE * 64-bit) blocks */
+	r1 = r * 128 / (S_P_SIZE * 8);
+
+	/* X <-- B_{r1 - 1} */
+	blkcpy(Bout, &Bin[(r1 - 1) * S_P_SIZE], S_P_SIZE);
+
+	/* X <-- X \xor B_i */
+	blkxor(Bout, Bin, S_P_SIZE);
+
+	/* X <-- H'(X) */
+	/* B'_i <-- X */
+	block_pwxform(Bout, S);
+
+	/* for i = 0 to r1 - 1 do */
+	for (i = 1; i < r1; i++) {
+		/* X <-- X \xor B_i */
+		blkcpy(&Bout[i * S_P_SIZE], &Bout[(i - 1) * S_P_SIZE],
+		    S_P_SIZE);
+		blkxor(&Bout[i * S_P_SIZE], &Bin[i * S_P_SIZE], S_P_SIZE);
+
+		/* X <-- H'(X) */
+		/* B'_i <-- X */
+		block_pwxform(&Bout[i * S_P_SIZE], S);
+	}
+
+	/* Handle partial blocks */
+	if (i * S_P_SIZE < r * 16)
+		blkcpy(&Bout[i * S_P_SIZE], &Bin[i * S_P_SIZE],
+		    r * 16 - i * S_P_SIZE);
+
+	i = (r1 - 1) * S_P_SIZE / 8;
+	/* Convert 128-byte blocks to 64-byte blocks */
+	r2 = r * 2;
+
+	/* B'_i <-- H(B'_i) */
+	salsa20_8(&Bout[i * 8]);
+	i++;
+
+	for (; i < r2; i++) {
+		/* B'_i <-- H(B'_i \xor B'_{i-1}) */
+		blkxor(&Bout[i * 8], &Bout[(i - 1) * 8], 8);
+		salsa20_8(&Bout[i * 8]);
+	}
+}
+
+/**
+ * integerify(B, r):
+ * Return the result of parsing B_{2r-1} as a little-endian integer.
+ */
+static inline uint64_t
+integerify(const uint64_t * B, size_t r)
+{
+/*
+ * Our 64-bit words are in host byte order, and word 6 holds the second 32-bit
+ * word of B_{2r-1} due to SIMD shuffling.  The 64-bit value we return is also
+ * in host byte order, as it should be.
+ */
+	const uint64_t * X = &B[(2 * r - 1) * 8];
+	uint32_t lo = X[0];
+	uint32_t hi = X[6] >> 32;
+	return ((uint64_t)hi << 32) + lo;
+}
+
+/**
+ * smix1(B, r, N, flags, V, NROM, shared, XY, S):
+ * Compute first loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 256r + 64 bytes in length.  The value N must be even and
+ * no smaller than 2.
+ */
+static void
+smix1(uint64_t * B, size_t r, uint64_t N, yescrypt_flags_t flags,
+    uint64_t * V, uint64_t NROM, const yescrypt_shared_t * shared,
+    uint64_t * XY, uint64_t * S)
+{
+	void (*blockmix)(const uint64_t *, uint64_t *, uint64_t *, size_t) =
+	    (S ? blockmix_pwxform : blockmix_salsa8);
+	const uint64_t * VROM = shared->shared1.aligned;
+	uint32_t VROM_mask = shared->mask1;
+	size_t s = 16 * r;
+	uint64_t * X = V;
+	uint64_t * Y = &XY[s];
+	uint64_t * Z = S ? S : &XY[2 * s];
+	uint64_t n, i, j;
+	size_t k;
+
+	/* 1: X <-- B */
+	/* 3: V_i <-- X */
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = (const salsa20_blk_t *)&B[i * 8];
+		salsa20_blk_t *tmp = (salsa20_blk_t *)Y;
+		salsa20_blk_t *dst = (salsa20_blk_t *)&X[i * 8];
+		for (k = 0; k < 16; k++)
+			tmp->w[k] = le32dec(&src->w[k]);
+		salsa20_simd_shuffle(tmp, dst);
+	}
+
+	/* 4: X <-- H(X) */
+	/* 3: V_i <-- X */
+	blockmix(X, Y, Z, r);
+	blkcpy(&V[s], Y, s);
+
+	X = XY;
+
+	if (NROM && (VROM_mask & 1)) {
+		if ((1 & VROM_mask) == 1) {
+			/* j <-- Integerify(X) mod NROM */
+			j = integerify(Y, r) & (NROM - 1);
+
+			/* X <-- H(X \xor VROM_j) */
+			blkxor(Y, &VROM[j * s], s);
+		}
+
+		blockmix(Y, X, Z, r);
+
+		/* 2: for i = 0 to N - 1 do */
+		for (n = 1, i = 2; i < N; i += 2) {
+			/* 3: V_i <-- X */
+			blkcpy(&V[i * s], X, s);
+
+			if ((i & (i - 1)) == 0)
+				n <<= 1;
+
+			/* j <-- Wrap(Integerify(X), i) */
+			j = integerify(X, r) & (n - 1);
+			j += i - n;
+
+			/* X <-- X \xor V_j */
+			blkxor(X, &V[j * s], s);
+
+			/* 4: X <-- H(X) */
+			blockmix(X, Y, Z, r);
+
+			/* 3: V_i <-- X */
+			blkcpy(&V[(i + 1) * s], Y, s);
+
+			j = integerify(Y, r);
+			if (((i + 1) & VROM_mask) == 1) {
+				/* j <-- Integerify(X) mod NROM */
+				j &= NROM - 1;
+
+				/* X <-- H(X \xor VROM_j) */
+				blkxor(Y, &VROM[j * s], s);
+			} else {
+				/* j <-- Wrap(Integerify(X), i) */
+				j &= n - 1;
+				j += i + 1 - n;
+
+				/* X <-- H(X \xor V_j) */
+				blkxor(Y, &V[j * s], s);
+			}
+
+			blockmix(Y, X, Z, r);
+		}
+	} else {
+		yescrypt_flags_t rw = flags & YESCRYPT_RW;
+
+		/* 4: X <-- H(X) */
+		blockmix(Y, X, Z, r);
+
+		/* 2: for i = 0 to N - 1 do */
+		for (n = 1, i = 2; i < N; i += 2) {
+			/* 3: V_i <-- X */
+			blkcpy(&V[i * s], X, s);
+
+			if (rw) {
+				if ((i & (i - 1)) == 0)
+					n <<= 1;
+
+				/* j <-- Wrap(Integerify(X), i) */
+				j = integerify(X, r) & (n - 1);
+				j += i - n;
+
+				/* X <-- X \xor V_j */
+				blkxor(X, &V[j * s], s);
+			}
+
+			/* 4: X <-- H(X) */
+			blockmix(X, Y, Z, r);
+
+			/* 3: V_i <-- X */
+			blkcpy(&V[(i + 1) * s], Y, s);
+
+			if (rw) {
+				/* j <-- Wrap(Integerify(X), i) */
+				j = integerify(Y, r) & (n - 1);
+				j += (i + 1) - n;
+
+				/* X <-- X \xor V_j */
+				blkxor(Y, &V[j * s], s);
+			}
+
+			/* 4: X <-- H(X) */
+			blockmix(Y, X, Z, r);
+		}
+	}
+
+	/* B' <-- X */
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = (const salsa20_blk_t *)&X[i * 8];
+		salsa20_blk_t *tmp = (salsa20_blk_t *)Y;
+		salsa20_blk_t *dst = (salsa20_blk_t *)&B[i * 8];
+		for (k = 0; k < 16; k++)
+			le32enc(&tmp->w[k], src->w[k]);
+		salsa20_simd_unshuffle(tmp, dst);
+	}
+}
+
+/**
+ * smix2(B, r, N, Nloop, flags, V, NROM, shared, XY, S):
+ * Compute second loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 256r + 64 bytes in length.  The value N must be a
+ * power of 2 greater than 1.  The value Nloop must be even.
+ */
+static void
+smix2(uint64_t * B, size_t r, uint64_t N, uint64_t Nloop,
+    yescrypt_flags_t flags,
+    uint64_t * V, uint64_t NROM, const yescrypt_shared_t * shared,
+    uint64_t * XY, uint64_t * S)
+{
+	void (*blockmix)(const uint64_t *, uint64_t *, uint64_t *, size_t) =
+	    (S ? blockmix_pwxform : blockmix_salsa8);
+	const uint64_t * VROM = shared->shared1.aligned;
+	uint32_t VROM_mask = shared->mask1 | 1;
+	size_t s = 16 * r;
+	yescrypt_flags_t rw = flags & YESCRYPT_RW;
+	uint64_t * X = XY;
+	uint64_t * Y = &XY[s];
+	uint64_t * Z = S ? S : &XY[2 * s];
+	uint64_t i, j;
+	size_t k;
+
+	if (Nloop == 0)
+		return;
+
+	/* X <-- B' */
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = (const salsa20_blk_t *)&B[i * 8];
+		salsa20_blk_t *tmp = (salsa20_blk_t *)Y;
+		salsa20_blk_t *dst = (salsa20_blk_t *)&X[i * 8];
+		for (k = 0; k < 16; k++)
+			tmp->w[k] = le32dec(&src->w[k]);
+		salsa20_simd_shuffle(tmp, dst);
+	}
+
+	if (NROM) {
+		/* 6: for i = 0 to N - 1 do */
+		for (i = 0; i < Nloop; i += 2) {
+			/* 7: j <-- Integerify(X) mod N */
+			j = integerify(X, r) & (N - 1);
+
+			/* 8: X <-- H(X \xor V_j) */
+			blkxor(X, &V[j * s], s);
+			/* V_j <-- Xprev \xor V_j */
+			if (rw)
+				blkcpy(&V[j * s], X, s);
+			blockmix(X, Y, Z, r);
+
+			j = integerify(Y, r);
+			if (((i + 1) & VROM_mask) == 1) {
+				/* j <-- Integerify(X) mod NROM */
+				j &= NROM - 1;
+
+				/* X <-- H(X \xor VROM_j) */
+				blkxor(Y, &VROM[j * s], s);
+			} else {
+				/* 7: j <-- Integerify(X) mod N */
+				j &= N - 1;
+
+				/* 8: X <-- H(X \xor V_j) */
+				blkxor(Y, &V[j * s], s);
+				/* V_j <-- Xprev \xor V_j */
+				if (rw)
+					blkcpy(&V[j * s], Y, s);
+			}
+
+			blockmix(Y, X, Z, r);
+		}
+	} else {
+		/* 6: for i = 0 to N - 1 do */
+		i = Nloop / 2;
+		do {
+			/* 7: j <-- Integerify(X) mod N */
+			j = integerify(X, r) & (N - 1);
+
+			/* 8: X <-- H(X \xor V_j) */
+			blkxor(X, &V[j * s], s);
+			/* V_j <-- Xprev \xor V_j */
+			if (rw)
+				blkcpy(&V[j * s], X, s);
+			blockmix(X, Y, Z, r);
+
+			/* 7: j <-- Integerify(X) mod N */
+			j = integerify(Y, r) & (N - 1);
+
+			/* 8: X <-- H(X \xor V_j) */
+			blkxor(Y, &V[j * s], s);
+			/* V_j <-- Xprev \xor V_j */
+			if (rw)
+				blkcpy(&V[j * s], Y, s);
+			blockmix(Y, X, Z, r);
+		} while (--i);
+	}
+
+	/* 10: B' <-- X */
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = (const salsa20_blk_t *)&X[i * 8];
+		salsa20_blk_t *tmp = (salsa20_blk_t *)Y;
+		salsa20_blk_t *dst = (salsa20_blk_t *)&B[i * 8];
+		for (k = 0; k < 16; k++)
+			le32enc(&tmp->w[k], src->w[k]);
+		salsa20_simd_unshuffle(tmp, dst);
+	}
+}
+
+/**
+ * p2floor(x):
+ * Largest power of 2 not greater than argument.
+ */
+static uint64_t
+p2floor(uint64_t x)
+{
+	uint64_t y;
+	while ((y = x & (x - 1)))
+		x = y;
+	return x;
+}
+
+/**
+ * smix(B, r, N, p, t, flags, V, NROM, shared, XY, S):
+ * Compute B = SMix_r(B, N).  The input B must be 128rp bytes in length; the
+ * temporary storage V must be 128rN bytes in length; the temporary storage
+ * XY must be 256r+64 or (256r+64)*p bytes in length (the larger size is
+ * required with OpenMP-enabled builds).  The value N must be a power of 2
+ * greater than 1.
+ */
+static void
+smix(uint64_t * B, size_t r, uint64_t N, uint32_t p, uint32_t t,
+    yescrypt_flags_t flags,
+    uint64_t * V, uint64_t NROM, const yescrypt_shared_t * shared,
+    uint64_t * XY, uint64_t * S)
+{
+	size_t s = 16 * r;
+	uint64_t Nchunk = N / p, Nloop_all, Nloop_rw;
+	uint32_t i;
+
+	Nloop_all = Nchunk;
+	if (flags & YESCRYPT_RW) {
+		if (t <= 1) {
+			if (t)
+				Nloop_all *= 2; /* 2/3 */
+			Nloop_all = (Nloop_all + 2) / 3; /* 1/3, round up */
+		} else {
+			Nloop_all *= t - 1;
+		}
+	} else if (t) {
+		if (t == 1)
+			Nloop_all += (Nloop_all + 1) / 2; /* 1.5, round up */
+		Nloop_all *= t;
+	}
+
+	Nloop_rw = 0;
+	if (flags & __YESCRYPT_INIT_SHARED)
+		Nloop_rw = Nloop_all;
+	else if (flags & YESCRYPT_RW)
+		Nloop_rw = Nloop_all / p;
+
+	Nchunk &= ~(uint64_t)1; /* round down to even */
+	Nloop_all++; Nloop_all &= ~(uint64_t)1; /* round up to even */
+	Nloop_rw &= ~(uint64_t)1; /* round down to even */
+
+#ifdef _OPENMP
+#pragma omp parallel if (p > 1) default(none) private(i) shared(B, r, N, p, flags, V, NROM, shared, XY, S, s, Nchunk, Nloop_all, Nloop_rw)
+	{
+#pragma omp for
+#endif
+	for (i = 0; i < p; i++) {
+		uint64_t Vchunk = i * Nchunk;
+		uint64_t * Bp = &B[i * s];
+		uint64_t * Vp = &V[Vchunk * s];
+#ifdef _OPENMP
+		uint64_t * XYp = &XY[i * (2 * s + 8)];
+#else
+		uint64_t * XYp = XY;
+#endif
+		uint64_t Np = (i < p - 1) ? Nchunk : (N - Vchunk);
+		uint64_t * Sp = S ? &S[i * S_SIZE_ALL] : S;
+		if (Sp)
+			smix1(Bp, 1, S_SIZE_ALL / 16,
+			    flags & ~YESCRYPT_PWXFORM,
+			    Sp, NROM, shared, XYp, NULL);
+		if (!(flags & __YESCRYPT_INIT_SHARED_2))
+			smix1(Bp, r, Np, flags, Vp, NROM, shared, XYp, Sp);
+		smix2(Bp, r, p2floor(Np), Nloop_rw, flags, Vp,
+		    NROM, shared, XYp, Sp);
+	}
+
+	if (Nloop_all > Nloop_rw) {
+#ifdef _OPENMP
+#pragma omp for
+#endif
+		for (i = 0; i < p; i++) {
+			uint64_t * Bp = &B[i * s];
+#ifdef _OPENMP
+			uint64_t * XYp = &XY[i * (2 * s + 8)];
+#else
+			uint64_t * XYp = XY;
+#endif
+			uint64_t * Sp = S ? &S[i * S_SIZE_ALL] : S;
+			smix2(Bp, r, N, Nloop_all - Nloop_rw,
+			    flags & ~YESCRYPT_RW, V, NROM, shared, XYp, Sp);
+		}
+	}
+#ifdef _OPENMP
+	}
+#endif
+}
+
+/**
+ * yescrypt_kdf(shared, local, passwd, passwdlen, salt, saltlen,
+ *     N, r, p, t, flags, buf, buflen):
+ * Compute scrypt(passwd[0 .. passwdlen - 1], salt[0 .. saltlen - 1], N, r,
+ * p, buflen), or a revision of scrypt as requested by flags and shared, and
+ * write the result into buf.  The parameters r, p, and buflen must satisfy
+ * r * p < 2^30 and buflen <= (2^32 - 1) * 32.  The parameter N must be a power
+ * of 2 greater than 1.
+ *
+ * t controls computation time while not affecting peak memory usage.  shared
+ * and flags may request special modes as described in yescrypt.h.  local is
+ * the thread-local data structure, allowing to preserve and reuse a memory
+ * allocation across calls, thereby reducing its overhead.
+ *
+ * Return 0 on success; or -1 on error.
+ */
+static int
+yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
+    const uint8_t * passwd, size_t passwdlen,
+    const uint8_t * salt, size_t saltlen,
+    uint64_t N, uint32_t r, uint32_t p, uint32_t t, yescrypt_flags_t flags,
+    uint8_t * buf, size_t buflen)
+{
+	yescrypt_region_t tmp;
+	uint64_t NROM;
+	size_t B_size, V_size, XY_size, need;
+	uint64_t * B, * V, * XY, * S;
+	uint64_t sha256[4];
+
+	/*
+	 * YESCRYPT_PARALLEL_SMIX is a no-op at p = 1 for its intended purpose,
+	 * so don't let it have side-effects.  Without this adjustment, it'd
+	 * enable the SHA-256 password pre-hashing and output post-hashing,
+	 * because any deviation from classic scrypt implies those.
+	 */
+	if (p == 1)
+		flags &= ~YESCRYPT_PARALLEL_SMIX;
+
+	/* Sanity-check parameters */
+	if (flags & ~YESCRYPT_KNOWN_FLAGS) {
+		errno = EINVAL;
+		return -1;
+	}
+#if SIZE_MAX > UINT32_MAX
+	if (buflen > (((uint64_t)(1) << 32) - 1) * 32) {
+		errno = EFBIG;
+		return -1;
+	}
+#endif
+	if ((uint64_t)(r) * (uint64_t)(p) >= (1 << 30)) {
+		errno = EFBIG;
+		return -1;
+	}
+	if (((N & (N - 1)) != 0) || (N <= 1) || (r < 1) || (p < 1)) {
+		errno = EINVAL;
+		return -1;
+	}
+	if ((flags & YESCRYPT_PARALLEL_SMIX) && (N / p <= 1)) {
+		errno = EINVAL;
+		return -1;
+	}
+#if S_MIN_R > 1
+	if ((flags & YESCRYPT_PWXFORM) && (r < S_MIN_R)) {
+		errno = EINVAL;
+		return -1;
+	}
+#endif
+	if ((p > SIZE_MAX / ((size_t)256 * r + 64)) ||
+#if SIZE_MAX / 256 <= UINT32_MAX
+	    (r > SIZE_MAX / 256) ||
+#endif
+	    (N > SIZE_MAX / 128 / r)) {
+		errno = ENOMEM;
+		return -1;
+	}
+	if (N > UINT64_MAX / ((uint64_t)t + 1)) {
+		errno = EFBIG;
+		return -1;
+	}
+#ifdef _OPENMP
+	if (!(flags & YESCRYPT_PARALLEL_SMIX) &&
+	    (N > SIZE_MAX / 128 / (r * p))) {
+		errno = ENOMEM;
+		return -1;
+	}
+#endif
+	if ((flags & YESCRYPT_PWXFORM) &&
+#ifndef _OPENMP
+	    (flags & YESCRYPT_PARALLEL_SMIX) &&
+#endif
+	    p > SIZE_MAX / (S_SIZE_ALL * sizeof(*S))) {
+		errno = ENOMEM;
+		return -1;
+	}
+
+	NROM = 0;
+	if (shared->shared1.aligned) {
+		NROM = shared->shared1.aligned_size / ((size_t)128 * r);
+		if (((NROM & (NROM - 1)) != 0) || (NROM <= 1) ||
+		    !(flags & YESCRYPT_RW)) {
+			errno = EINVAL;
+			return -1;
+		}
+	}
+
+	/* Allocate memory */
+	V = NULL;
+	V_size = (size_t)128 * r * N;
+#ifdef _OPENMP
+	if (!(flags & YESCRYPT_PARALLEL_SMIX))
+		V_size *= p;
+#endif
+	need = V_size;
+	if (flags & __YESCRYPT_INIT_SHARED) {
+		if (local->aligned_size < need) {
+			if (local->base || local->aligned ||
+			    local->base_size || local->aligned_size) {
+				errno = EINVAL;
+				return -1;
+			}
+			if (!alloc_region(local, need))
+				return -1;
+		}
+		V = (uint64_t *)local->aligned;
+		need = 0;
+	}
+	B_size = (size_t)128 * r * p;
+	need += B_size;
+	if (need < B_size) {
+		errno = ENOMEM;
+		return -1;
+	}
+	XY_size = (size_t)256 * r + 64;
+#ifdef _OPENMP
+	XY_size *= p;
+#endif
+	need += XY_size;
+	if (need < XY_size) {
+		errno = ENOMEM;
+		return -1;
+	}
+	if (flags & YESCRYPT_PWXFORM) {
+		size_t S_size = S_SIZE_ALL * sizeof(*S);
+#ifdef _OPENMP
+		S_size *= p;
+#else
+		if (flags & YESCRYPT_PARALLEL_SMIX)
+			S_size *= p;
+#endif
+		need += S_size;
+		if (need < S_size) {
+			errno = ENOMEM;
+			return -1;
+		}
+	}
+	if (flags & __YESCRYPT_INIT_SHARED) {
+		if (!alloc_region(&tmp, need))
+			return -1;
+		B = (uint64_t *)tmp.aligned;
+		XY = (uint64_t *)((uint8_t *)B + B_size);
+	} else {
+		init_region(&tmp);
+		if (local->aligned_size < need) {
+			if (free_region(local))
+				return -1;
+			if (!alloc_region(local, need))
+				return -1;
+		}
+		B = (uint64_t *)local->aligned;
+		V = (uint64_t *)((uint8_t *)B + B_size);
+		XY = (uint64_t *)((uint8_t *)V + V_size);
+	}
+	S = NULL;
+	if (flags & YESCRYPT_PWXFORM)
+		S = (uint64_t *)((uint8_t *)XY + XY_size);
+
+	if (t || flags) {
+		SHA256_CTX ctx;
+		SHA256_Init(&ctx);
+		SHA256_Update(&ctx, passwd, passwdlen);
+		SHA256_Final((uint8_t *)sha256, &ctx);
+		passwd = (uint8_t *)sha256;
+		passwdlen = sizeof(sha256);
+	}
+
+	/* 1: (B_0 ... B_{p-1}) <-- PBKDF2(P, S, 1, p * MFLen) */
+	PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, 1,
+	    (uint8_t *)B, B_size);
+
+	if (t || flags)
+		blkcpy(sha256, B, sizeof(sha256) / sizeof(sha256[0]));
+
+	if (p == 1 || (flags & YESCRYPT_PARALLEL_SMIX)) {
+		smix(B, r, N, p, t, flags, V, NROM, shared, XY, S);
+	} else {
+		uint32_t i;
+
+		/* 2: for i = 0 to p - 1 do */
+#ifdef _OPENMP
+#pragma omp parallel for default(none) private(i) shared(B, r, N, p, t, flags, V, NROM, shared, XY, S)
+#endif
+		for (i = 0; i < p; i++) {
+			/* 3: B_i <-- MF(B_i, N) */
+#ifdef _OPENMP
+			smix(&B[(size_t)16 * r * i], r, N, 1, t, flags,
+			    &V[(size_t)16 * r * i * N],
+			    NROM, shared,
+			    &XY[((size_t)32 * r + 8) * i],
+			    S ? &S[S_SIZE_ALL * i] : S);
+#else
+			smix(&B[(size_t)16 * r * i], r, N, 1, t, flags, V,
+			    NROM, shared, XY, S);
+#endif
+		}
+	}
+
+	/* 5: DK <-- PBKDF2(P, B, 1, dkLen) */
+	PBKDF2_SHA256(passwd, passwdlen, (uint8_t *)B, B_size, 1, buf, buflen);
+
+	/*
+	 * Except when computing classic scrypt, allow all computation so far
+	 * to be performed on the client.  The final steps below match those of
+	 * SCRAM (RFC 5802), so that an extension of SCRAM (with the steps so
+	 * far in place of SCRAM's use of PBKDF2 and with SHA-256 in place of
+	 * SCRAM's use of SHA-1) would be usable with yescrypt hashes.
+	 */
+	if ((t || flags) && buflen == sizeof(sha256)) {
+		/* Compute ClientKey */
+		{
+			HMAC_SHA256_CTX ctx;
+			HMAC_SHA256_Init(&ctx, buf, buflen);
+			HMAC_SHA256_Update(&ctx, "Client Key", 10);
+			HMAC_SHA256_Final((uint8_t *)sha256, &ctx);
+		}
+		/* Compute StoredKey */
+		{
+			SHA256_CTX ctx;
+			SHA256_Init(&ctx);
+			SHA256_Update(&ctx, (uint8_t *)sha256, sizeof(sha256));
+			SHA256_Final(buf, &ctx);
+		}
+	}
+
+	if (free_region(&tmp))
+		return -1;
+
+	/* Success! */
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-platform.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-platform.c
new file mode 100644
index 0000000..f8583ca
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-platform.c
@@ -0,0 +1,191 @@
+/*-
+ * Copyright 2013,2014 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+//#include <sys/mman.h>
+
+#define HUGEPAGE_THRESHOLD		(12 * 1024 * 1024)
+
+#ifdef __x86_64__
+#define HUGEPAGE_SIZE			(2 * 1024 * 1024)
+#else
+#undef HUGEPAGE_SIZE
+#endif
+
+static void *
+alloc_region(yescrypt_region_t * region, size_t size)
+{
+	size_t base_size = size;
+	uint8_t * base, * aligned;
+#ifdef MAP_ANON
+	int flags =
+#ifdef MAP_NOCORE
+	    MAP_NOCORE |
+#endif
+	    MAP_ANON | MAP_PRIVATE;
+#if defined(MAP_HUGETLB) && defined(HUGEPAGE_SIZE)
+	size_t new_size = size;
+	const size_t hugepage_mask = (size_t)HUGEPAGE_SIZE - 1;
+	if (size >= HUGEPAGE_THRESHOLD && size + hugepage_mask >= size) {
+		flags |= MAP_HUGETLB;
+/*
+ * Linux's munmap() fails on MAP_HUGETLB mappings if size is not a multiple of
+ * huge page size, so let's round up to huge page size here.
+ */
+		new_size = size + hugepage_mask;
+		new_size &= ~hugepage_mask;
+	}
+	base = mmap(NULL, new_size, PROT_READ | PROT_WRITE, flags, -1, 0);
+	if (base != MAP_FAILED) {
+		base_size = new_size;
+	} else
+	if (flags & MAP_HUGETLB) {
+		flags &= ~MAP_HUGETLB;
+		base = mmap(NULL, size, PROT_READ | PROT_WRITE, flags, -1, 0);
+	}
+
+#else
+	base = mmap(NULL, size, PROT_READ | PROT_WRITE, flags, -1, 0);
+#endif
+	if (base == MAP_FAILED)
+		base = NULL;
+	aligned = base;
+#elif defined(HAVE_POSIX_MEMALIGN)
+	if ((errno = posix_memalign((void **)&base, 64, size)) != 0)
+		base = NULL;
+	aligned = base;
+#else
+	base = aligned = NULL;
+	if (size + 63 < size) {
+		errno = ENOMEM;
+	} else if ((base = malloc(size + 63)) != NULL) {
+		aligned = base + 63;
+		aligned -= (uintptr_t)aligned & 63;
+	}
+#endif
+	region->base = base;
+	region->aligned = aligned;
+	region->base_size = base ? base_size : 0;
+	region->aligned_size = base ? size : 0;
+	return aligned;
+}
+
+static inline void
+init_region(yescrypt_region_t * region)
+{
+	region->base = region->aligned = NULL;
+	region->base_size = region->aligned_size = 0;
+}
+
+static int
+free_region(yescrypt_region_t * region)
+{
+	if (region->base) {
+#ifdef MAP_ANON
+		if (munmap(region->base, region->base_size))
+			return -1;
+#else
+		free(region->base);
+#endif
+	}
+	init_region(region);
+	return 0;
+}
+
+static int
+yescrypt_init_shared(yescrypt_shared_t * shared,
+    const uint8_t * param, size_t paramlen,
+    uint64_t N, uint32_t r, uint32_t p,
+    yescrypt_init_shared_flags_t flags, uint32_t mask,
+    uint8_t * buf, size_t buflen)
+{
+	yescrypt_shared1_t * shared1 = &shared->shared1;
+	yescrypt_shared_t dummy, half1, half2;
+	uint8_t salt[32];
+
+	if (flags & YESCRYPT_SHARED_PREALLOCATED) {
+		if (!shared1->aligned || !shared1->aligned_size)
+			return -1;
+	} else {
+		init_region(shared1);
+	}
+	shared->mask1 = 1;
+	if (!param && !paramlen && !N && !r && !p && !buf && !buflen)
+		return 0;
+
+	init_region(&dummy.shared1);
+	dummy.mask1 = 1;
+	if (yescrypt_kdf(&dummy, shared1,
+	    param, paramlen, NULL, 0, N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
+	    salt, sizeof(salt)))
+		goto out;
+
+	half1 = half2 = *shared;
+	half1.shared1.aligned_size /= 2;
+	half2.shared1.aligned += half1.shared1.aligned_size;
+	half2.shared1.aligned_size = half1.shared1.aligned_size;
+	N /= 2;
+
+	if (p > 1 && yescrypt_kdf(&half1, &half2.shared1,
+	    param, paramlen, salt, sizeof(salt), N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_2,
+	    salt, sizeof(salt)))
+		goto out;
+
+	if (yescrypt_kdf(&half2, &half1.shared1,
+	    param, paramlen, salt, sizeof(salt), N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
+	    salt, sizeof(salt)))
+		goto out;
+
+	if (yescrypt_kdf(&half1, &half2.shared1,
+	    param, paramlen, salt, sizeof(salt), N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
+	    buf, buflen))
+		goto out;
+
+	shared->mask1 = mask;
+
+	return 0;
+
+out:
+	if (!(flags & YESCRYPT_SHARED_PREALLOCATED))
+		free_region(shared1);
+	return -1;
+}
+
+static int
+yescrypt_free_shared(yescrypt_shared_t * shared)
+{
+	return free_region(&shared->shared1);
+}
+
+static int
+yescrypt_init_local(yescrypt_local_t * local)
+{
+	init_region(local);
+	return 0;
+}
+
+static int
+yescrypt_free_local(yescrypt_local_t * local)
+{
+	return free_region(local);
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-ref.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-ref.c
new file mode 100644
index 0000000..9785734
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-ref.c
@@ -0,0 +1,758 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2013,2014 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ *
+ * This is the reference implementation.  Its purpose is to provide a simple
+ * human- and machine-readable specification that implementations intended
+ * for actual use should be tested against.  It is deliberately mostly not
+ * optimized, and it is not meant to be used in production.  Instead, use
+ * yescrypt-best.c or one of the source files included from there.
+ */
+
+#warning "This reference implementation is deliberately mostly not optimized. Use yescrypt-best.c instead unless you're testing (against) the reference implementation on purpose."
+
+#include <errno.h>
+#include <stdint.h>
+#include <stdlib.h>
+
+#include "sha256.h"
+#include "sysendian.h"
+
+#include "yescrypt.h"
+
+static void
+blkcpy(uint32_t * dest, const uint32_t * src, size_t count)
+{
+	do {
+		*dest++ = *src++;
+	} while (--count);
+}
+
+static void
+blkxor(uint32_t * dest, const uint32_t * src, size_t count)
+{
+	do {
+		*dest++ ^= *src++;
+	} while (--count);
+}
+
+/**
+ * salsa20_8(B):
+ * Apply the salsa20/8 core to the provided block.
+ */
+static void
+salsa20_8(uint32_t B[16])
+{
+	uint32_t x[16];
+	size_t i;
+
+	/* Mimic SIMD shuffling */
+	for (i = 0; i < 16; i++)
+		x[i * 5 % 16] = B[i];
+
+	for (i = 0; i < 8; i += 2) {
+#define R(a,b) (((a) << (b)) | ((a) >> (32 - (b))))
+		/* Operate on columns */
+		x[ 4] ^= R(x[ 0]+x[12], 7);  x[ 8] ^= R(x[ 4]+x[ 0], 9);
+		x[12] ^= R(x[ 8]+x[ 4],13);  x[ 0] ^= R(x[12]+x[ 8],18);
+
+		x[ 9] ^= R(x[ 5]+x[ 1], 7);  x[13] ^= R(x[ 9]+x[ 5], 9);
+		x[ 1] ^= R(x[13]+x[ 9],13);  x[ 5] ^= R(x[ 1]+x[13],18);
+
+		x[14] ^= R(x[10]+x[ 6], 7);  x[ 2] ^= R(x[14]+x[10], 9);
+		x[ 6] ^= R(x[ 2]+x[14],13);  x[10] ^= R(x[ 6]+x[ 2],18);
+
+		x[ 3] ^= R(x[15]+x[11], 7);  x[ 7] ^= R(x[ 3]+x[15], 9);
+		x[11] ^= R(x[ 7]+x[ 3],13);  x[15] ^= R(x[11]+x[ 7],18);
+
+		/* Operate on rows */
+		x[ 1] ^= R(x[ 0]+x[ 3], 7);  x[ 2] ^= R(x[ 1]+x[ 0], 9);
+		x[ 3] ^= R(x[ 2]+x[ 1],13);  x[ 0] ^= R(x[ 3]+x[ 2],18);
+
+		x[ 6] ^= R(x[ 5]+x[ 4], 7);  x[ 7] ^= R(x[ 6]+x[ 5], 9);
+		x[ 4] ^= R(x[ 7]+x[ 6],13);  x[ 5] ^= R(x[ 4]+x[ 7],18);
+
+		x[11] ^= R(x[10]+x[ 9], 7);  x[ 8] ^= R(x[11]+x[10], 9);
+		x[ 9] ^= R(x[ 8]+x[11],13);  x[10] ^= R(x[ 9]+x[ 8],18);
+
+		x[12] ^= R(x[15]+x[14], 7);  x[13] ^= R(x[12]+x[15], 9);
+		x[14] ^= R(x[13]+x[12],13);  x[15] ^= R(x[14]+x[13],18);
+#undef R
+	}
+
+	/* Mimic SIMD shuffling */
+	for (i = 0; i < 16; i++)
+		B[i] += x[i * 5 % 16];
+}
+
+/**
+ * blockmix_salsa8(B, Y, r):
+ * Compute B = BlockMix_{salsa20/8, r}(B).  The input B must be 128r bytes in
+ * length; the temporary space Y must also be the same size.
+ */
+static void
+blockmix_salsa8(uint32_t * B, uint32_t * Y, size_t r)
+{
+	uint32_t X[16];
+	size_t i;
+
+	/* 1: X <-- B_{2r - 1} */
+	blkcpy(X, &B[(2 * r - 1) * 16], 16);
+
+	/* 2: for i = 0 to 2r - 1 do */
+	for (i = 0; i < 2 * r; i++) {
+		/* 3: X <-- H(X \xor B_i) */
+		blkxor(X, &B[i * 16], 16);
+		salsa20_8(X);
+
+		/* 4: Y_i <-- X */
+		blkcpy(&Y[i * 16], X, 16);
+	}
+
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	for (i = 0; i < r; i++)
+		blkcpy(&B[i * 16], &Y[(i * 2) * 16], 16);
+	for (i = 0; i < r; i++)
+		blkcpy(&B[(i + r) * 16], &Y[(i * 2 + 1) * 16], 16);
+}
+
+/* These are tunable */
+#define S_BITS 8
+#define S_SIMD 2
+#define S_P 4
+#define S_ROUNDS 6
+
+/* Number of S-boxes.  Not tunable, hard-coded in a few places. */
+#define S_N 2
+
+/* Derived values.  Not tunable on their own. */
+#define S_SIZE1 (1 << S_BITS)
+#define S_MASK ((S_SIZE1 - 1) * S_SIMD * 8)
+#define S_SIZE_ALL (S_N * S_SIZE1 * S_SIMD * 2)
+#define S_P_SIZE (S_P * S_SIMD * 2)
+#define S_MIN_R ((S_P * S_SIMD + 15) / 16)
+
+/**
+ * pwxform(B):
+ * Transform the provided block using the provided S-boxes.
+ */
+static void
+block_pwxform(uint32_t * B, const uint32_t * S)
+{
+	uint32_t (*X)[S_SIMD][2] = (uint32_t (*)[S_SIMD][2])B;
+	const uint32_t (*S0)[2] = (const uint32_t (*)[2])S;
+	const uint32_t (*S1)[2] = S0 + S_SIZE1 * S_SIMD;
+	size_t i, j, k;
+
+	for (i = 0; i < S_ROUNDS; i++) {
+		for (j = 0; j < S_P; j++) {
+			uint32_t xl = X[j][0][0];
+			uint32_t xh = X[j][0][1];
+			const uint32_t (*p0)[2], (*p1)[2];
+
+			p0 = S0 + (xl & S_MASK) / sizeof(*S0);
+			p1 = S1 + (xh & S_MASK) / sizeof(*S1);
+
+			for (k = 0; k < S_SIMD; k++) {
+				uint64_t x, s0, s1;
+
+				s0 = ((uint64_t)p0[k][1] << 32) + p0[k][0];
+				s1 = ((uint64_t)p1[k][1] << 32) + p1[k][0];
+
+				xl = X[j][k][0];
+				xh = X[j][k][1];
+
+				x = (uint64_t)xh * xl;
+				x += s0;
+				x ^= s1;
+
+				X[j][k][0] = x;
+				X[j][k][1] = x >> 32;
+			}
+		}
+	}
+}
+
+/**
+ * blockmix_pwxform(B, Y, S, r):
+ * Compute B = BlockMix_pwxform{salsa20/8, S, r}(B).  The input B must be 128r
+ * bytes in length; the temporary space Y must be at least S_P_SIZE*4 bytes.
+ */
+static void
+blockmix_pwxform(uint32_t * B, uint32_t * Y, const uint32_t * S, size_t r)
+{
+	size_t r1, r2, i;
+
+	/* Convert 128-byte blocks to (S_P_SIZE * 32-bit) blocks */
+	r1 = r * 128 / (S_P_SIZE * 4);
+
+	/* X <-- B_{r1 - 1} */
+	blkcpy(Y, &B[(r1 - 1) * S_P_SIZE], S_P_SIZE);
+
+	/* for i = 0 to r1 - 1 do */
+	for (i = 0; i < r1; i++) {
+		/* X <-- X \xor B_i */
+		blkxor(Y, &B[i * S_P_SIZE], S_P_SIZE);
+
+		/* X <-- H'(X) */
+		block_pwxform(Y, S);
+
+		/* B'_i <-- X */
+		blkcpy(&B[i * S_P_SIZE], Y, S_P_SIZE);
+	}
+
+	i = (r1 - 1) * S_P_SIZE / 16;
+	/* Convert 128-byte blocks to 64-byte blocks */
+	r2 = r * 2;
+
+	/* B_i <-- H(B_i) */
+	salsa20_8(&B[i * 16]);
+	i++;
+
+	for (; i < r2; i++) {
+		/* B_i <-- H(B_i \xor B_{i-1}) */
+		blkxor(&B[i * 16], &B[(i - 1) * 16], 16);
+		salsa20_8(&B[i * 16]);
+	}
+}
+
+/**
+ * integerify(B, r):
+ * Return the result of parsing B_{2r-1} as a little-endian integer.
+ */
+static uint64_t
+integerify(const uint32_t * B, size_t r)
+{
+/*
+ * Our 32-bit words are in host byte order, and word 13 is the second word of
+ * B_{2r-1} due to SIMD shuffling.  The 64-bit value we return is also in host
+ * byte order, as it should be.
+ */
+	const uint32_t * X = &B[(2 * r - 1) * 16];
+	return ((uint64_t)X[13] << 32) + X[0];
+}
+
+/**
+ * p2floor(x):
+ * Largest power of 2 not greater than argument.
+ */
+static uint64_t
+p2floor(uint64_t x)
+{
+	uint64_t y;
+	while ((y = x & (x - 1)))
+		x = y;
+	return x;
+}
+
+/**
+ * wrap(x, i):
+ * Wrap x to the range 0 to i-1.
+ */
+static uint64_t
+wrap(uint64_t x, uint64_t i)
+{
+	uint64_t n = p2floor(i);
+	return (x & (n - 1)) + (i - n);
+}
+
+/**
+ * smix1(B, r, N, flags, V, NROM, shared, XY, S):
+ * Compute first loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 256r bytes in length.
+ */
+static void
+smix1(uint32_t * B, size_t r, uint64_t N, yescrypt_flags_t flags,
+    uint32_t * V, uint64_t NROM, const yescrypt_shared_t * shared,
+    uint32_t * XY, uint32_t * S)
+{
+	const uint32_t * VROM = shared->shared1.aligned;
+	uint32_t VROM_mask = NROM ? shared->mask1 : 0;
+	size_t s = 32 * r;
+	uint32_t * X = XY;
+	uint32_t * Y = &XY[s];
+	uint64_t i, j;
+	size_t k;
+
+	/* 1: X <-- B */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			X[k * 16 + i] = le32dec(&B[k * 16 + (i * 5 % 16)]);
+
+	/* 2: for i = 0 to N - 1 do */
+	for (i = 0; i < N; i++) {
+		/* 3: V_i <-- X */
+		blkcpy(&V[i * s], X, s);
+
+		if ((i & VROM_mask) == 1) {
+			/* j <-- Integerify(X) mod NROM */
+			j = integerify(X, r) & (NROM - 1);
+
+			/* X <-- H(X \xor VROM_j) */
+			blkxor(X, &VROM[j * s], s);
+		} else if ((flags & YESCRYPT_RW) && i > 1) {
+			/* j <-- Wrap(Integerify(X), i) */
+			j = wrap(integerify(X, r), i);
+
+			/* X <-- X \xor V_j */
+			blkxor(X, &V[j * s], s);
+		}
+
+		/* 4: X <-- H(X) */
+		if (S)
+			blockmix_pwxform(X, Y, S, r);
+		else
+			blockmix_salsa8(X, Y, r);
+	}
+
+	/* B' <-- X */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			le32enc(&B[k * 16 + (i * 5 % 16)], X[k * 16 + i]);
+}
+
+/**
+ * smix2(B, r, N, Nloop, flags, V, NROM, shared, XY, S):
+ * Compute second loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 256r bytes in length.  The value N must be a power of 2
+ * greater than 1.
+ */
+static void
+smix2(uint32_t * B, size_t r, uint64_t N, uint64_t Nloop,
+    yescrypt_flags_t flags, uint32_t * V, uint64_t NROM,
+    const yescrypt_shared_t * shared, uint32_t * XY, uint32_t * S)
+{
+	const uint32_t * VROM = shared->shared1.aligned;
+	uint32_t VROM_mask = NROM ? (shared->mask1 | 1) : 0;
+	size_t s = 32 * r;
+	uint32_t * X = XY;
+	uint32_t * Y = &XY[s];
+	uint64_t i, j;
+	size_t k;
+
+	/* X <-- B */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			X[k * 16 + i] = le32dec(&B[k * 16 + (i * 5 % 16)]);
+
+	/* 6: for i = 0 to N - 1 do */
+	for (i = 0; i < Nloop; i++) {
+		if ((i & VROM_mask) == 1) {
+			/* j <-- Integerify(X) mod NROM */
+			j = integerify(X, r) & (NROM - 1);
+
+			/* X <-- H(X \xor VROM_j) */
+			blkxor(X, &VROM[j * s], s);
+		} else {
+			/* 7: j <-- Integerify(X) mod N */
+			j = integerify(X, r) & (N - 1);
+
+			/* 8.1: X <-- X \xor V_j */
+			blkxor(X, &V[j * s], s);
+			/* V_j <-- X */
+			if (flags & YESCRYPT_RW)
+				blkcpy(&V[j * s], X, s);
+		}
+
+		/* 8.2: X <-- H(X) */
+		if (S)
+			blockmix_pwxform(X, Y, S, r);
+		else
+			blockmix_salsa8(X, Y, r);
+	}
+
+	/* 10: B' <-- X */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			le32enc(&B[k * 16 + (i * 5 % 16)], X[k * 16 + i]);
+}
+
+/**
+ * smix(B, r, N, p, t, flags, V, NROM, shared, XY, S):
+ * Compute B = SMix_r(B, N).  The input B must be 128rp bytes in length; the
+ * temporary storage V must be 128rN bytes in length; the temporary storage
+ * XY must be 256r bytes in length.  The value N must be a power of 2 greater
+ * than 1.
+ */
+static void
+smix(uint32_t * B, size_t r, uint64_t N, uint32_t p, uint32_t t,
+    yescrypt_flags_t flags,
+    uint32_t * V, uint64_t NROM, const yescrypt_shared_t * shared,
+    uint32_t * XY, uint32_t * S)
+{
+	size_t s = 32 * r;
+	uint64_t Vchunk = 0, Nchunk = N / p, Nloop_all, Nloop_rw;
+	uint32_t i;
+
+	Nloop_all = Nchunk;
+	if (flags & YESCRYPT_RW) {
+		if (t <= 1) {
+			if (t)
+				Nloop_all *= 2; /* 2/3 */
+			Nloop_all = (Nloop_all + 2) / 3; /* 1/3, round up */
+		} else {
+			Nloop_all *= t - 1;
+		}
+	} else if (t) {
+		if (t == 1)
+			Nloop_all += (Nloop_all + 1) / 2; /* 1.5, round up */
+		Nloop_all *= t;
+	}
+
+	Nloop_rw = 0;
+	if (flags & __YESCRYPT_INIT_SHARED)
+		Nloop_rw = Nloop_all;
+	else if (flags & YESCRYPT_RW)
+		Nloop_rw = Nloop_all / p;
+
+	Nchunk &= ~(uint64_t)1; /* round down to even */
+	Nloop_all++; Nloop_all &= ~(uint64_t)1; /* round up to even */
+	Nloop_rw &= ~(uint64_t)1; /* round down to even */
+
+	for (i = 0; i < p; i++) {
+		uint32_t * Bp = &B[i * s];
+		uint32_t * Vp = &V[Vchunk * s];
+		uint64_t Np = (i < p - 1) ? Nchunk : (N - Vchunk);
+		uint32_t * Sp = S ? &S[i * S_SIZE_ALL] : S;
+		if (Sp)
+			smix1(Bp, 1, S_SIZE_ALL / 32,
+			    flags & ~YESCRYPT_PWXFORM,
+			    Sp, NROM, shared, XY, NULL);
+		if (!(flags & __YESCRYPT_INIT_SHARED_2))
+			smix1(Bp, r, Np, flags, Vp, NROM, shared, XY, Sp);
+		smix2(Bp, r, p2floor(Np), Nloop_rw, flags, Vp,
+		    NROM, shared, XY, Sp);
+		Vchunk += Nchunk;
+	}
+
+	for (i = 0; i < p; i++) {
+		uint32_t * Bp = &B[i * s];
+		uint32_t * Sp = S ? &S[i * S_SIZE_ALL] : NULL;
+		smix2(Bp, r, N, Nloop_all - Nloop_rw, flags & ~YESCRYPT_RW,
+		    V, NROM, shared, XY, Sp);
+	}
+}
+
+/**
+ * yescrypt_kdf(shared, local, passwd, passwdlen, salt, saltlen,
+ *     N, r, p, t, flags, buf, buflen):
+ * Compute scrypt(passwd[0 .. passwdlen - 1], salt[0 .. saltlen - 1], N, r,
+ * p, buflen), or a revision of scrypt as requested by flags and shared, and
+ * write the result into buf.  The parameters r, p, and buflen must satisfy
+ * r * p < 2^30 and buflen <= (2^32 - 1) * 32.  The parameter N must be a power
+ * of 2 greater than 1.
+ *
+ * t controls computation time while not affecting peak memory usage.  shared
+ * and flags may request special modes as described in yescrypt.h.  local is
+ * the thread-local data structure, allowing optimized implementations to
+ * preserve and reuse a memory allocation across calls, thereby reducing its
+ * overhead (this reference implementation does not make that optimization).
+ *
+ * Return 0 on success; or -1 on error.
+ */
+int
+yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
+    const uint8_t * passwd, size_t passwdlen,
+    const uint8_t * salt, size_t saltlen,
+    uint64_t N, uint32_t r, uint32_t p, uint32_t t, yescrypt_flags_t flags,
+    uint8_t * buf, size_t buflen)
+{
+	int retval = -1;
+	uint64_t NROM;
+	size_t B_size, V_size;
+	uint32_t * B, * V, * XY, * S;
+	uint32_t sha256[8];
+
+	/*
+	 * YESCRYPT_PARALLEL_SMIX is a no-op at p = 1 for its intended purpose,
+	 * so don't let it have side-effects.  Without this adjustment, it'd
+	 * enable the SHA-256 password pre-hashing and output post-hashing,
+	 * because any deviation from classic scrypt implies those.
+	 */
+	if (p == 1)
+		flags &= ~YESCRYPT_PARALLEL_SMIX;
+
+	/* Sanity-check parameters */
+	if (flags & ~YESCRYPT_KNOWN_FLAGS) {
+		errno = EINVAL;
+		return -1;
+	}
+#if SIZE_MAX > UINT32_MAX
+	if (buflen > (((uint64_t)(1) << 32) - 1) * 32) {
+		errno = EFBIG;
+		return -1;
+	}
+#endif
+	if ((uint64_t)(r) * (uint64_t)(p) >= (1 << 30)) {
+		errno = EFBIG;
+		return -1;
+	}
+	if (((N & (N - 1)) != 0) || (N <= 1) || (r < 1) || (p < 1)) {
+		errno = EINVAL;
+		return -1;
+	}
+	if ((flags & YESCRYPT_PARALLEL_SMIX) && (N / p <= 1)) {
+		errno = EINVAL;
+		return -1;
+	}
+#if S_MIN_R > 1
+	if ((flags & YESCRYPT_PWXFORM) && (r < S_MIN_R)) {
+		errno = EINVAL;
+		return -1;
+	}
+#endif
+	if ((r > SIZE_MAX / 128 / p) ||
+#if SIZE_MAX / 256 <= UINT32_MAX
+	    (r > SIZE_MAX / 256) ||
+#endif
+	    (N > SIZE_MAX / 128 / r)) {
+		errno = ENOMEM;
+		return -1;
+	}
+	if (N > UINT64_MAX / ((uint64_t)t + 1)) {
+		errno = EFBIG;
+		return -1;
+	}
+	if (((flags & (YESCRYPT_PWXFORM | YESCRYPT_PARALLEL_SMIX)) ==
+	    (YESCRYPT_PWXFORM | YESCRYPT_PARALLEL_SMIX)) &&
+	    p > SIZE_MAX / (S_SIZE_ALL * sizeof(*S))) {
+		errno = ENOMEM;
+		return -1;
+	}
+
+	NROM = 0;
+	if (shared->shared1.aligned) {
+		NROM = shared->shared1.aligned_size / ((size_t)128 * r);
+/*
+ * This implementation could support NROM without YESCRYPT_RW as well, but we
+ * currently don't want to make such support available so that it can be safely
+ * excluded from optimized implementations (where it'd require extra code).
+ */
+		if (((NROM & (NROM - 1)) != 0) || (NROM <= 1) ||
+		    !(flags & YESCRYPT_RW)) {
+			errno = EINVAL;
+			return -1;
+		}
+	}
+
+	/* Allocate memory */
+	V_size = (size_t)128 * r * N;
+	if (flags & __YESCRYPT_INIT_SHARED) {
+		V = (uint32_t *)local->aligned;
+		if (local->aligned_size < V_size) {
+			if (local->base || local->aligned ||
+			    local->base_size || local->aligned_size) {
+				errno = EINVAL;
+				return -1;
+			}
+			if ((V = malloc(V_size)) == NULL)
+				return -1;
+			local->base = local->aligned = V;
+			local->base_size = local->aligned_size = V_size;
+		}
+	} else {
+		if ((V = malloc(V_size)) == NULL)
+			return -1;
+	}
+	B_size = (size_t)128 * r * p;
+	if ((B = malloc(B_size)) == NULL)
+		goto free_V;
+	if ((XY = malloc((size_t)256 * r)) == NULL)
+		goto free_B;
+	S = NULL;
+	if (flags & YESCRYPT_PWXFORM) {
+		size_t S_size = S_SIZE_ALL * sizeof(*S);
+		if (flags & YESCRYPT_PARALLEL_SMIX)
+			S_size *= p;
+		if ((S = malloc(S_size)) == NULL)
+			goto free_XY;
+	}
+
+	if (t || flags) {
+		SHA256_CTX ctx;
+		SHA256_Init(&ctx);
+		SHA256_Update(&ctx, passwd, passwdlen);
+		SHA256_Final((uint8_t *)sha256, &ctx);
+		passwd = (uint8_t *)sha256;
+		passwdlen = sizeof(sha256);
+	}
+
+	/* 1: (B_0 ... B_{p-1}) <-- PBKDF2(P, S, 1, p * MFLen) */
+	PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, 1,
+	    (uint8_t *)B, B_size);
+
+	if (t || flags)
+		blkcpy(sha256, B, sizeof(sha256) / sizeof(sha256[0]));
+
+	if (flags & YESCRYPT_PARALLEL_SMIX) {
+		smix(B, r, N, p, t, flags, V, NROM, shared, XY, S);
+	} else {
+		uint32_t i;
+
+		/* 2: for i = 0 to p - 1 do */
+		for (i = 0; i < p; i++) {
+			/* 3: B_i <-- MF(B_i, N) */
+			smix(&B[(size_t)32 * r * i], r, N, 1, t, flags, V,
+			    NROM, shared, XY, S);
+		}
+	}
+
+	/* 5: DK <-- PBKDF2(P, B, 1, dkLen) */
+	PBKDF2_SHA256(passwd, passwdlen, (uint8_t *)B, B_size, 1, buf, buflen);
+
+	/*
+	 * Except when computing classic scrypt, allow all computation so far
+	 * to be performed on the client.  The final steps below match those of
+	 * SCRAM (RFC 5802), so that an extension of SCRAM (with the steps so
+	 * far in place of SCRAM's use of PBKDF2 and with SHA-256 in place of
+	 * SCRAM's use of SHA-1) would be usable with yescrypt hashes.
+	 */
+	if ((t || flags) && buflen == sizeof(sha256)) {
+		/* Compute ClientKey */
+		{
+			HMAC_SHA256_CTX ctx;
+			HMAC_SHA256_Init(&ctx, buf, buflen);
+			HMAC_SHA256_Update(&ctx, "Client Key", 10);
+			HMAC_SHA256_Final((uint8_t *)sha256, &ctx);
+		}
+		/* Compute StoredKey */
+		{
+			SHA256_CTX ctx;
+			SHA256_Init(&ctx);
+			SHA256_Update(&ctx, (uint8_t *)sha256, sizeof(sha256));
+			SHA256_Final(buf, &ctx);
+		}
+	}
+
+	/* Success! */
+	retval = 0;
+
+	/* Free memory */
+	free(S);
+free_XY:
+	free(XY);
+free_B:
+	free(B);
+free_V:
+	if (!(flags & __YESCRYPT_INIT_SHARED))
+		free(V);
+
+	return retval;
+}
+
+int
+yescrypt_init_shared(yescrypt_shared_t * shared,
+    const uint8_t * param, size_t paramlen,
+    uint64_t N, uint32_t r, uint32_t p,
+    yescrypt_init_shared_flags_t flags, uint32_t mask,
+    uint8_t * buf, size_t buflen)
+{
+	yescrypt_shared1_t * shared1 = &shared->shared1;
+	yescrypt_shared_t dummy, half1, half2;
+	uint8_t salt[32];
+
+	if (flags & YESCRYPT_SHARED_PREALLOCATED) {
+		if (!shared1->aligned || !shared1->aligned_size)
+			return -1;
+	} else {
+		shared1->base = shared1->aligned = NULL;
+		shared1->base_size = shared1->aligned_size = 0;
+	}
+	shared->mask1 = 1;
+	if (!param && !paramlen && !N && !r && !p && !buf && !buflen)
+		return 0;
+
+	dummy.shared1.base = dummy.shared1.aligned = NULL;
+	dummy.shared1.base_size = dummy.shared1.aligned_size = 0;
+	dummy.mask1 = 1;
+	if (yescrypt_kdf(&dummy, shared1,
+	    param, paramlen, NULL, 0, N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
+	    salt, sizeof(salt)))
+		goto out;
+
+	half1 = half2 = *shared;
+	half1.shared1.aligned_size /= 2;
+	half2.shared1.aligned += half1.shared1.aligned_size;
+	half2.shared1.aligned_size = half1.shared1.aligned_size;
+	N /= 2;
+
+	if (p > 1 && yescrypt_kdf(&half1, &half2.shared1,
+	    param, paramlen, salt, sizeof(salt), N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_2,
+	    salt, sizeof(salt)))
+		goto out;
+
+	if (yescrypt_kdf(&half2, &half1.shared1,
+	    param, paramlen, salt, sizeof(salt), N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
+	    salt, sizeof(salt)))
+		goto out;
+
+	if (yescrypt_kdf(&half1, &half2.shared1,
+	    param, paramlen, salt, sizeof(salt), N, r, p, 0,
+	    YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | __YESCRYPT_INIT_SHARED_1,
+	    buf, buflen))
+		goto out;
+
+	shared->mask1 = mask;
+
+	return 0;
+
+out:
+	if (!(flags & YESCRYPT_SHARED_PREALLOCATED))
+		free(shared1->base);
+	return -1;
+}
+
+int
+yescrypt_free_shared(yescrypt_shared_t * shared)
+{
+	free(shared->shared1.base);
+	shared->shared1.base = shared->shared1.aligned = NULL;
+	shared->shared1.base_size = shared->shared1.aligned_size = 0;
+	return 0;
+}
+
+int
+yescrypt_init_local(yescrypt_local_t * local)
+{
+/* The reference implementation doesn't use the local structure */
+	local->base = local->aligned = NULL;
+	local->base_size = local->aligned_size = 0;
+	return 0;
+}
+
+int
+yescrypt_free_local(yescrypt_local_t * local)
+{
+/* The reference implementation frees its memory in yescrypt_kdf() */
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-simd.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-simd.c
new file mode 100644
index 0000000..354b54a
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt-simd.c
@@ -0,0 +1,1386 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2012-2014 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ */
+//
+//
+// ottylab update history <ottylab@gmail.com>
+//
+// 2018/03/20 @001 Support KitKat
+//
+
+/*
+ * On 64-bit, enabling SSE4.1 helps our pwxform code indirectly, via avoiding
+ * gcc bug 54349 (fixed for gcc 4.9+).  On 32-bit, it's of direct help.  AVX
+ * and XOP are of further help either way.
+ */
+#ifndef __SSE4_1__
+#warning "Consider enabling SSE4.1, AVX, or XOP in the C compiler for significantly better performance"
+#endif
+
+#include <emmintrin.h>
+#ifdef __XOP__
+#include <x86intrin.h>
+#endif
+
+#include <errno.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+
+//@001 start
+#ifdef __KITKAT__
+#include <sys/limits.h>
+#endif // __KITKAT__
+//@001 end
+
+#include "sha256.h"
+#include "sysendian.h"
+
+#include "yescrypt.h"
+
+#include "yescrypt-platform.c"
+
+#if __STDC_VERSION__ >= 199901L
+/* have restrict */
+#elif defined(__GNUC__)
+#define restrict __restrict
+#else
+#define restrict
+#endif
+
+#define PREFETCH(x, hint) _mm_prefetch((const char *)(x), (hint));
+#define PREFETCH_OUT(x, hint) /* disabled */
+
+#ifdef __XOP__
+#define ARX(out, in1, in2, s) \
+	out = _mm_xor_si128(out, _mm_roti_epi32(_mm_add_epi32(in1, in2), s));
+#else
+#define ARX(out, in1, in2, s) \
+	{ \
+		__m128i T = _mm_add_epi32(in1, in2); \
+		out = _mm_xor_si128(out, _mm_slli_epi32(T, s)); \
+		out = _mm_xor_si128(out, _mm_srli_epi32(T, 32-s)); \
+	}
+#endif
+
+#define SALSA20_2ROUNDS \
+	/* Operate on "columns" */ \
+	ARX(X1, X0, X3, 7) \
+	ARX(X2, X1, X0, 9) \
+	ARX(X3, X2, X1, 13) \
+	ARX(X0, X3, X2, 18) \
+\
+	/* Rearrange data */ \
+	X1 = _mm_shuffle_epi32(X1, 0x93); \
+	X2 = _mm_shuffle_epi32(X2, 0x4E); \
+	X3 = _mm_shuffle_epi32(X3, 0x39); \
+\
+	/* Operate on "rows" */ \
+	ARX(X3, X0, X1, 7) \
+	ARX(X2, X3, X0, 9) \
+	ARX(X1, X2, X3, 13) \
+	ARX(X0, X1, X2, 18) \
+\
+	/* Rearrange data */ \
+	X1 = _mm_shuffle_epi32(X1, 0x39); \
+	X2 = _mm_shuffle_epi32(X2, 0x4E); \
+	X3 = _mm_shuffle_epi32(X3, 0x93);
+
+/**
+ * Apply the salsa20/8 core to the block provided in (X0 ... X3).
+ */
+#define SALSA20_8_BASE(maybe_decl, out) \
+	{ \
+		maybe_decl Y0 = X0; \
+		maybe_decl Y1 = X1; \
+		maybe_decl Y2 = X2; \
+		maybe_decl Y3 = X3; \
+		SALSA20_2ROUNDS \
+		SALSA20_2ROUNDS \
+		SALSA20_2ROUNDS \
+		SALSA20_2ROUNDS \
+		(out)[0] = X0 = _mm_add_epi32(X0, Y0); \
+		(out)[1] = X1 = _mm_add_epi32(X1, Y1); \
+		(out)[2] = X2 = _mm_add_epi32(X2, Y2); \
+		(out)[3] = X3 = _mm_add_epi32(X3, Y3); \
+	}
+#define SALSA20_8(out) \
+	SALSA20_8_BASE(__m128i, out)
+
+/**
+ * Apply the salsa20/8 core to the block provided in (X0 ... X3) ^ (Z0 ... Z3).
+ */
+#define SALSA20_8_XOR_ANY(maybe_decl, Z0, Z1, Z2, Z3, out) \
+	X0 = _mm_xor_si128(X0, Z0); \
+	X1 = _mm_xor_si128(X1, Z1); \
+	X2 = _mm_xor_si128(X2, Z2); \
+	X3 = _mm_xor_si128(X3, Z3); \
+	SALSA20_8_BASE(maybe_decl, out)
+
+#define SALSA20_8_XOR_MEM(in, out) \
+	SALSA20_8_XOR_ANY(__m128i, (in)[0], (in)[1], (in)[2], (in)[3], out)
+
+#define SALSA20_8_XOR_REG(out) \
+	SALSA20_8_XOR_ANY(/* empty */, Y0, Y1, Y2, Y3, out)
+
+typedef union {
+	uint32_t w[16];
+	__m128i q[4];
+} salsa20_blk_t;
+
+/**
+ * blockmix_salsa8(Bin, Bout, r):
+ * Compute Bout = BlockMix_{salsa20/8, r}(Bin).  The input Bin must be 128r
+ * bytes in length; the output Bout must also be the same size.
+ */
+static inline void
+blockmix_salsa8(const salsa20_blk_t *restrict Bin,
+    salsa20_blk_t *restrict Bout, size_t r)
+{
+	__m128i X0, X1, X2, X3;
+	size_t i;
+
+	r--;
+	PREFETCH(&Bin[r * 2 + 1], _MM_HINT_T0)
+	for (i = 0; i < r; i++) {
+		PREFETCH(&Bin[i * 2], _MM_HINT_T0)
+		PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+		PREFETCH(&Bin[i * 2 + 1], _MM_HINT_T0)
+		PREFETCH_OUT(&Bout[r + 1 + i], _MM_HINT_T0)
+	}
+	PREFETCH(&Bin[r * 2], _MM_HINT_T0)
+	PREFETCH_OUT(&Bout[r], _MM_HINT_T0)
+	PREFETCH_OUT(&Bout[r * 2 + 1], _MM_HINT_T0)
+
+	/* 1: X <-- B_{2r - 1} */
+	X0 = Bin[r * 2 + 1].q[0];
+	X1 = Bin[r * 2 + 1].q[1];
+	X2 = Bin[r * 2 + 1].q[2];
+	X3 = Bin[r * 2 + 1].q[3];
+
+	/* 3: X <-- H(X \xor B_i) */
+	/* 4: Y_i <-- X */
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	SALSA20_8_XOR_MEM(Bin[0].q, Bout[0].q)
+
+	/* 2: for i = 0 to 2r - 1 do */
+	for (i = 0; i < r;) {
+		/* 3: X <-- H(X \xor B_i) */
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		SALSA20_8_XOR_MEM(Bin[i * 2 + 1].q, Bout[r + 1 + i].q)
+
+		i++;
+
+		/* 3: X <-- H(X \xor B_i) */
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		SALSA20_8_XOR_MEM(Bin[i * 2].q, Bout[i].q)
+	}
+
+	/* 3: X <-- H(X \xor B_i) */
+	/* 4: Y_i <-- X */
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	SALSA20_8_XOR_MEM(Bin[r * 2 + 1].q, Bout[r * 2 + 1].q)
+}
+
+/*
+ * (V)PSRLDQ and (V)PSHUFD have higher throughput than (V)PSRLQ on some CPUs
+ * starting with Sandy Bridge.  Additionally, PSHUFD uses separate source and
+ * destination registers, whereas the shifts would require an extra move
+ * instruction for our code when building without AVX.  Unfortunately, PSHUFD
+ * is much slower on Conroe (4 cycles latency vs. 1 cycle latency for PSRLQ)
+ * and somewhat slower on some non-Intel CPUs (luckily not including AMD
+ * Bulldozer and Piledriver).  Since for many other CPUs using (V)PSHUFD is a
+ * win in terms of throughput or/and not needing a move instruction, we
+ * currently use it despite of the higher latency on some older CPUs.  As an
+ * alternative, the #if below may be patched to only enable use of (V)PSHUFD
+ * when building with SSE4.1 or newer, which is not available on older CPUs
+ * where this instruction has higher latency.
+ */
+#if 1
+#define HI32(X) \
+	_mm_shuffle_epi32((X), _MM_SHUFFLE(2,3,0,1))
+#elif 0
+#define HI32(X) \
+	_mm_srli_si128((X), 4)
+#else
+#define HI32(X) \
+	_mm_srli_epi64((X), 32)
+#endif
+
+#if defined(__x86_64__) && (defined(__ICC) || defined(__llvm__))
+/* Intel's name, also supported by recent gcc */
+#define EXTRACT64(X) _mm_cvtsi128_si64(X)
+#elif defined(__x86_64__) && !defined(_MSC_VER) && !defined(__OPEN64__)
+/* gcc got the 'x' name earlier than non-'x', MSVC and Open64 had bugs */
+#define EXTRACT64(X) _mm_cvtsi128_si64x(X)
+#elif defined(__x86_64__) && defined(__SSE4_1__)
+/* No known bugs for this intrinsic */
+#include <smmintrin.h>
+#define EXTRACT64(X) _mm_extract_epi64((X), 0)
+#elif defined(__SSE4_1__)
+/* 32-bit */
+#include <smmintrin.h>
+#if 0
+/* This is currently unused by the code below, which instead uses these two
+ * intrinsics explicitly when (!defined(__x86_64__) && defined(__SSE4_1__)) */
+#define EXTRACT64(X) \
+	((uint64_t)(uint32_t)_mm_cvtsi128_si32(X) | \
+	((uint64_t)(uint32_t)_mm_extract_epi32((X), 1) << 32))
+#endif
+#else
+/* 32-bit or compilers with known past bugs in _mm_cvtsi128_si64*() */
+#define EXTRACT64(X) \
+	((uint64_t)(uint32_t)_mm_cvtsi128_si32(X) | \
+	((uint64_t)(uint32_t)_mm_cvtsi128_si32(HI32(X)) << 32))
+#endif
+
+/* This is tunable */
+#define S_BITS 8
+
+/* Not tunable in this implementation, hard-coded in a few places */
+#define S_SIMD 2
+#define S_P 4
+
+/* Number of S-boxes.  Not tunable by design, hard-coded in a few places. */
+#define S_N 2
+
+/* Derived values.  Not tunable except via S_BITS above. */
+#define S_SIZE1 (1 << S_BITS)
+#define S_MASK ((S_SIZE1 - 1) * S_SIMD * 8)
+#define S_MASK2 (((uint64_t)S_MASK << 32) | S_MASK)
+#define S_SIZE_ALL (S_N * S_SIZE1 * S_SIMD * 8)
+
+#if !defined(__x86_64__) && defined(__SSE4_1__)
+/* 32-bit with SSE4.1 */
+#define PWXFORM_X_T __m128i
+#define PWXFORM_SIMD(X, x, s0, s1) \
+	x = _mm_and_si128(X, _mm_set1_epi64x(S_MASK2)); \
+	s0 = *(const __m128i *)(S0 + (uint32_t)_mm_cvtsi128_si32(x)); \
+	s1 = *(const __m128i *)(S1 + (uint32_t)_mm_extract_epi32(x, 1)); \
+	X = _mm_mul_epu32(HI32(X), X); \
+	X = _mm_add_epi64(X, s0); \
+	X = _mm_xor_si128(X, s1);
+#else
+/* 64-bit, or 32-bit without SSE4.1 */
+#define PWXFORM_X_T uint64_t
+#define PWXFORM_SIMD(X, x, s0, s1) \
+	x = EXTRACT64(X) & S_MASK2; \
+	s0 = *(const __m128i *)(S0 + (uint32_t)x); \
+	s1 = *(const __m128i *)(S1 + (x >> 32)); \
+	X = _mm_mul_epu32(HI32(X), X); \
+	X = _mm_add_epi64(X, s0); \
+	X = _mm_xor_si128(X, s1);
+#endif
+
+#define PWXFORM_ROUND \
+	PWXFORM_SIMD(X0, x0, s00, s01) \
+	PWXFORM_SIMD(X1, x1, s10, s11) \
+	PWXFORM_SIMD(X2, x2, s20, s21) \
+	PWXFORM_SIMD(X3, x3, s30, s31)
+
+#define PWXFORM \
+	{ \
+		PWXFORM_X_T x0, x1, x2, x3; \
+		__m128i s00, s01, s10, s11, s20, s21, s30, s31; \
+		PWXFORM_ROUND PWXFORM_ROUND \
+		PWXFORM_ROUND PWXFORM_ROUND \
+		PWXFORM_ROUND PWXFORM_ROUND \
+	}
+
+#define XOR4(in) \
+	X0 = _mm_xor_si128(X0, (in)[0]); \
+	X1 = _mm_xor_si128(X1, (in)[1]); \
+	X2 = _mm_xor_si128(X2, (in)[2]); \
+	X3 = _mm_xor_si128(X3, (in)[3]);
+
+#define OUT(out) \
+	(out)[0] = X0; \
+	(out)[1] = X1; \
+	(out)[2] = X2; \
+	(out)[3] = X3;
+
+/**
+ * blockmix_pwxform(Bin, Bout, r, S):
+ * Compute Bout = BlockMix_pwxform{salsa20/8, r, S}(Bin).  The input Bin must
+ * be 128r bytes in length; the output Bout must also be the same size.
+ */
+static void
+blockmix(const salsa20_blk_t *restrict Bin, salsa20_blk_t *restrict Bout,
+    size_t r, const __m128i *restrict S)
+{
+	const uint8_t * S0, * S1;
+	__m128i X0, X1, X2, X3;
+	size_t i;
+
+	if (!S) {
+		blockmix_salsa8(Bin, Bout, r);
+		return;
+	}
+
+	S0 = (const uint8_t *)S;
+	S1 = (const uint8_t *)S + S_SIZE_ALL / 2;
+
+	/* Convert 128-byte blocks to 64-byte blocks */
+	r *= 2;
+
+	r--;
+	PREFETCH(&Bin[r], _MM_HINT_T0)
+	for (i = 0; i < r; i++) {
+		PREFETCH(&Bin[i], _MM_HINT_T0)
+		PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+	}
+	PREFETCH_OUT(&Bout[r], _MM_HINT_T0)
+
+	/* X <-- B_{r1 - 1} */
+	X0 = Bin[r].q[0];
+	X1 = Bin[r].q[1];
+	X2 = Bin[r].q[2];
+	X3 = Bin[r].q[3];
+
+	/* for i = 0 to r1 - 1 do */
+	for (i = 0; i < r; i++) {
+		/* X <-- H'(X \xor B_i) */
+		XOR4(Bin[i].q)
+		PWXFORM
+		/* B'_i <-- X */
+		OUT(Bout[i].q)
+	}
+
+	/* Last iteration of the loop above */
+	XOR4(Bin[i].q)
+	PWXFORM
+
+	/* B'_i <-- H(B'_i) */
+	SALSA20_8(Bout[i].q)
+}
+
+#define XOR4_2(in1, in2) \
+	X0 = _mm_xor_si128((in1)[0], (in2)[0]); \
+	X1 = _mm_xor_si128((in1)[1], (in2)[1]); \
+	X2 = _mm_xor_si128((in1)[2], (in2)[2]); \
+	X3 = _mm_xor_si128((in1)[3], (in2)[3]);
+
+static inline uint32_t
+blockmix_salsa8_xor(const salsa20_blk_t *restrict Bin1,
+    const salsa20_blk_t *restrict Bin2, salsa20_blk_t *restrict Bout,
+    size_t r, int Bin2_in_ROM)
+{
+	__m128i X0, X1, X2, X3;
+	size_t i;
+
+	r--;
+	if (Bin2_in_ROM) {
+		PREFETCH(&Bin2[r * 2 + 1], _MM_HINT_NTA)
+		PREFETCH(&Bin1[r * 2 + 1], _MM_HINT_T0)
+		for (i = 0; i < r; i++) {
+			PREFETCH(&Bin2[i * 2], _MM_HINT_NTA)
+			PREFETCH(&Bin1[i * 2], _MM_HINT_T0)
+			PREFETCH(&Bin2[i * 2 + 1], _MM_HINT_NTA)
+			PREFETCH(&Bin1[i * 2 + 1], _MM_HINT_T0)
+			PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+			PREFETCH_OUT(&Bout[r + 1 + i], _MM_HINT_T0)
+		}
+		PREFETCH(&Bin2[r * 2], _MM_HINT_T0)
+	} else {
+		PREFETCH(&Bin2[r * 2 + 1], _MM_HINT_T0)
+		PREFETCH(&Bin1[r * 2 + 1], _MM_HINT_T0)
+		for (i = 0; i < r; i++) {
+			PREFETCH(&Bin2[i * 2], _MM_HINT_T0)
+			PREFETCH(&Bin1[i * 2], _MM_HINT_T0)
+			PREFETCH(&Bin2[i * 2 + 1], _MM_HINT_T0)
+			PREFETCH(&Bin1[i * 2 + 1], _MM_HINT_T0)
+			PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+			PREFETCH_OUT(&Bout[r + 1 + i], _MM_HINT_T0)
+		}
+		PREFETCH(&Bin2[r * 2], _MM_HINT_T0)
+	}
+	PREFETCH(&Bin1[r * 2], _MM_HINT_T0)
+	PREFETCH_OUT(&Bout[r], _MM_HINT_T0)
+	PREFETCH_OUT(&Bout[r * 2 + 1], _MM_HINT_T0)
+
+	/* 1: X <-- B_{2r - 1} */
+	XOR4_2(Bin1[r * 2 + 1].q, Bin2[r * 2 + 1].q)
+
+	/* 3: X <-- H(X \xor B_i) */
+	/* 4: Y_i <-- X */
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	XOR4(Bin1[0].q)
+	SALSA20_8_XOR_MEM(Bin2[0].q, Bout[0].q)
+
+	/* 2: for i = 0 to 2r - 1 do */
+	for (i = 0; i < r;) {
+		/* 3: X <-- H(X \xor B_i) */
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		XOR4(Bin1[i * 2 + 1].q)
+		SALSA20_8_XOR_MEM(Bin2[i * 2 + 1].q, Bout[r + 1 + i].q)
+
+		i++;
+
+		/* 3: X <-- H(X \xor B_i) */
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		XOR4(Bin1[i * 2].q)
+		SALSA20_8_XOR_MEM(Bin2[i * 2].q, Bout[i].q)
+	}
+
+	/* 3: X <-- H(X \xor B_i) */
+	/* 4: Y_i <-- X */
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	XOR4(Bin1[r * 2 + 1].q)
+	SALSA20_8_XOR_MEM(Bin2[r * 2 + 1].q, Bout[r * 2 + 1].q)
+
+	return _mm_cvtsi128_si32(X0);
+}
+
+static uint32_t
+blockmix_xor(const salsa20_blk_t *restrict Bin1,
+    const salsa20_blk_t *restrict Bin2, salsa20_blk_t *restrict Bout,
+    size_t r, int Bin2_in_ROM, const __m128i *restrict S)
+{
+	const uint8_t * S0, * S1;
+	__m128i X0, X1, X2, X3;
+	size_t i;
+
+	if (!S)
+		return blockmix_salsa8_xor(Bin1, Bin2, Bout, r, Bin2_in_ROM);
+
+	S0 = (const uint8_t *)S;
+	S1 = (const uint8_t *)S + S_SIZE_ALL / 2;
+
+	/* Convert 128-byte blocks to 64-byte blocks */
+	r *= 2;
+
+	r--;
+	if (Bin2_in_ROM) {
+		PREFETCH(&Bin2[r], _MM_HINT_NTA)
+		PREFETCH(&Bin1[r], _MM_HINT_T0)
+		for (i = 0; i < r; i++) {
+			PREFETCH(&Bin2[i], _MM_HINT_NTA)
+			PREFETCH(&Bin1[i], _MM_HINT_T0)
+			PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+		}
+	} else {
+		PREFETCH(&Bin2[r], _MM_HINT_T0)
+		PREFETCH(&Bin1[r], _MM_HINT_T0)
+		for (i = 0; i < r; i++) {
+			PREFETCH(&Bin2[i], _MM_HINT_T0)
+			PREFETCH(&Bin1[i], _MM_HINT_T0)
+			PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+		}
+	}
+	PREFETCH_OUT(&Bout[r], _MM_HINT_T0);
+
+	/* X <-- B_{r1 - 1} */
+	XOR4_2(Bin1[r].q, Bin2[r].q)
+
+	/* for i = 0 to r1 - 1 do */
+	for (i = 0; i < r; i++) {
+		/* X <-- H'(X \xor B_i) */
+		XOR4(Bin1[i].q)
+		XOR4(Bin2[i].q)
+		PWXFORM
+		/* B'_i <-- X */
+		OUT(Bout[i].q)
+	}
+
+	/* Last iteration of the loop above */
+	XOR4(Bin1[i].q)
+	XOR4(Bin2[i].q)
+	PWXFORM
+
+	/* B'_i <-- H(B'_i) */
+	SALSA20_8(Bout[i].q)
+
+	return _mm_cvtsi128_si32(X0);
+}
+
+#undef XOR4
+#define XOR4(in, out) \
+	(out)[0] = Y0 = _mm_xor_si128((in)[0], (out)[0]); \
+	(out)[1] = Y1 = _mm_xor_si128((in)[1], (out)[1]); \
+	(out)[2] = Y2 = _mm_xor_si128((in)[2], (out)[2]); \
+	(out)[3] = Y3 = _mm_xor_si128((in)[3], (out)[3]);
+
+static inline uint32_t
+blockmix_salsa8_xor_save(const salsa20_blk_t *restrict Bin1,
+    salsa20_blk_t *restrict Bin2, salsa20_blk_t *restrict Bout,
+    size_t r)
+{
+	__m128i X0, X1, X2, X3, Y0, Y1, Y2, Y3;
+	size_t i;
+
+	r--;
+	PREFETCH(&Bin2[r * 2 + 1], _MM_HINT_T0)
+	PREFETCH(&Bin1[r * 2 + 1], _MM_HINT_T0)
+	for (i = 0; i < r; i++) {
+		PREFETCH(&Bin2[i * 2], _MM_HINT_T0)
+		PREFETCH(&Bin1[i * 2], _MM_HINT_T0)
+		PREFETCH(&Bin2[i * 2 + 1], _MM_HINT_T0)
+		PREFETCH(&Bin1[i * 2 + 1], _MM_HINT_T0)
+		PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+		PREFETCH_OUT(&Bout[r + 1 + i], _MM_HINT_T0)
+	}
+	PREFETCH(&Bin2[r * 2], _MM_HINT_T0)
+	PREFETCH(&Bin1[r * 2], _MM_HINT_T0)
+	PREFETCH_OUT(&Bout[r], _MM_HINT_T0)
+	PREFETCH_OUT(&Bout[r * 2 + 1], _MM_HINT_T0)
+
+	/* 1: X <-- B_{2r - 1} */
+	XOR4_2(Bin1[r * 2 + 1].q, Bin2[r * 2 + 1].q)
+
+	/* 3: X <-- H(X \xor B_i) */
+	/* 4: Y_i <-- X */
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	XOR4(Bin1[0].q, Bin2[0].q)
+	SALSA20_8_XOR_REG(Bout[0].q)
+
+	/* 2: for i = 0 to 2r - 1 do */
+	for (i = 0; i < r;) {
+		/* 3: X <-- H(X \xor B_i) */
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		XOR4(Bin1[i * 2 + 1].q, Bin2[i * 2 + 1].q)
+		SALSA20_8_XOR_REG(Bout[r + 1 + i].q)
+
+		i++;
+
+		/* 3: X <-- H(X \xor B_i) */
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		XOR4(Bin1[i * 2].q, Bin2[i * 2].q)
+		SALSA20_8_XOR_REG(Bout[i].q)
+	}
+
+	/* 3: X <-- H(X \xor B_i) */
+	/* 4: Y_i <-- X */
+	/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+	XOR4(Bin1[r * 2 + 1].q, Bin2[r * 2 + 1].q)
+	SALSA20_8_XOR_REG(Bout[r * 2 + 1].q)
+
+	return _mm_cvtsi128_si32(X0);
+}
+
+#define XOR4_Y \
+	X0 = _mm_xor_si128(X0, Y0); \
+	X1 = _mm_xor_si128(X1, Y1); \
+	X2 = _mm_xor_si128(X2, Y2); \
+	X3 = _mm_xor_si128(X3, Y3);
+
+static uint32_t
+blockmix_xor_save(const salsa20_blk_t *restrict Bin1,
+    salsa20_blk_t *restrict Bin2, salsa20_blk_t *restrict Bout,
+    size_t r, const __m128i *restrict S)
+{
+	const uint8_t * S0, * S1;
+	__m128i X0, X1, X2, X3, Y0, Y1, Y2, Y3;
+	size_t i;
+
+	if (!S)
+		return blockmix_salsa8_xor_save(Bin1, Bin2, Bout, r);
+
+	S0 = (const uint8_t *)S;
+	S1 = (const uint8_t *)S + S_SIZE_ALL / 2;
+
+	/* Convert 128-byte blocks to 64-byte blocks */
+	r *= 2;
+
+	r--;
+	PREFETCH(&Bin2[r], _MM_HINT_T0)
+	PREFETCH(&Bin1[r], _MM_HINT_T0)
+	for (i = 0; i < r; i++) {
+		PREFETCH(&Bin2[i], _MM_HINT_T0)
+		PREFETCH(&Bin1[i], _MM_HINT_T0)
+		PREFETCH_OUT(&Bout[i], _MM_HINT_T0)
+	}
+	PREFETCH_OUT(&Bout[r], _MM_HINT_T0);
+
+	/* X <-- B_{r1 - 1} */
+	XOR4_2(Bin1[r].q, Bin2[r].q)
+
+	/* for i = 0 to r1 - 1 do */
+	for (i = 0; i < r; i++) {
+		XOR4(Bin1[i].q, Bin2[i].q)
+		/* X <-- H'(X \xor B_i) */
+		XOR4_Y
+		PWXFORM
+		/* B'_i <-- X */
+		OUT(Bout[i].q)
+	}
+
+	/* Last iteration of the loop above */
+	XOR4(Bin1[i].q, Bin2[i].q)
+	XOR4_Y
+	PWXFORM
+
+	/* B'_i <-- H(B'_i) */
+	SALSA20_8(Bout[i].q)
+
+	return _mm_cvtsi128_si32(X0);
+}
+
+#undef ARX
+#undef SALSA20_2ROUNDS
+#undef SALSA20_8
+#undef SALSA20_8_XOR_ANY
+#undef SALSA20_8_XOR_MEM
+#undef SALSA20_8_XOR_REG
+#undef PWXFORM_SIMD_1
+#undef PWXFORM_SIMD_2
+#undef PWXFORM_ROUND
+#undef PWXFORM
+#undef OUT
+#undef XOR4
+#undef XOR4_2
+#undef XOR4_Y
+
+/**
+ * integerify(B, r):
+ * Return the result of parsing B_{2r-1} as a little-endian integer.
+ */
+static inline uint32_t
+integerify(const salsa20_blk_t * B, size_t r)
+{
+	return B[2 * r - 1].w[0];
+}
+
+/**
+ * smix1(B, r, N, flags, V, NROM, shared, XY, S):
+ * Compute first loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 128r bytes in length.  The value N must be even and no
+ * smaller than 2.  The array V must be aligned to a multiple of 64 bytes, and
+ * arrays B and XY to a multiple of at least 16 bytes (aligning them to 64
+ * bytes as well saves cache lines, but might result in cache bank conflicts).
+ */
+static void
+smix1(uint8_t * B, size_t r, uint32_t N, yescrypt_flags_t flags,
+    salsa20_blk_t * V, uint32_t NROM, const yescrypt_shared_t * shared,
+    salsa20_blk_t * XY, void * S)
+{
+	const salsa20_blk_t * VROM = shared->shared1.aligned;
+	uint32_t VROM_mask = shared->mask1;
+	size_t s = 2 * r;
+	salsa20_blk_t * X = V, * Y;
+	uint32_t i, j;
+	size_t k;
+
+	/* 1: X <-- B */
+	/* 3: V_i <-- X */
+	for (k = 0; k < 2 * r; k++) {
+		for (i = 0; i < 16; i++) {
+			X[k].w[i] = le32dec(&B[(k * 16 + (i * 5 % 16)) * 4]);
+		}
+	}
+
+	if (NROM && (VROM_mask & 1)) {
+		uint32_t n;
+		salsa20_blk_t * V_n;
+		const salsa20_blk_t * V_j;
+
+		/* 4: X <-- H(X) */
+		/* 3: V_i <-- X */
+		Y = &V[s];
+		blockmix(X, Y, r, S);
+
+		X = &V[2 * s];
+		if ((1 & VROM_mask) == 1) {
+			/* j <-- Integerify(X) mod NROM */
+			j = integerify(Y, r) & (NROM - 1);
+			V_j = &VROM[j * s];
+
+			/* X <-- H(X \xor VROM_j) */
+			j = blockmix_xor(Y, V_j, X, r, 1, S);
+		} else {
+			/* X <-- H(X) */
+			blockmix(Y, X, r, S);
+			j = integerify(X, r);
+		}
+
+		for (n = 2; n < N; n <<= 1) {
+			uint32_t m = (n < N / 2) ? n : (N - 1 - n);
+
+			V_n = &V[n * s];
+
+			/* 2: for i = 0 to N - 1 do */
+			for (i = 1; i < m; i += 2) {
+				/* j <-- Wrap(Integerify(X), i) */
+				j &= n - 1;
+				j += i - 1;
+				V_j = &V[j * s];
+
+				/* X <-- X \xor V_j */
+				/* 4: X <-- H(X) */
+				/* 3: V_i <-- X */
+				Y = &V_n[i * s];
+				j = blockmix_xor(X, V_j, Y, r, 0, S);
+
+				if (((n + i) & VROM_mask) == 1) {
+					/* j <-- Integerify(X) mod NROM */
+					j &= NROM - 1;
+					V_j = &VROM[j * s];
+				} else {
+					/* j <-- Wrap(Integerify(X), i) */
+					j &= n - 1;
+					j += i;
+					V_j = &V[j * s];
+				}
+
+				/* X <-- H(X \xor VROM_j) */
+				X = &V_n[(i + 1) * s];
+				j = blockmix_xor(Y, V_j, X, r, 1, S);
+			}
+		}
+
+		n >>= 1;
+
+		/* j <-- Wrap(Integerify(X), i) */
+		j &= n - 1;
+		j += N - 2 - n;
+		V_j = &V[j * s];
+
+		/* X <-- X \xor V_j */
+		/* 4: X <-- H(X) */
+		/* 3: V_i <-- X */
+		Y = &V[(N - 1) * s];
+		j = blockmix_xor(X, V_j, Y, r, 0, S);
+
+		if (((N - 1) & VROM_mask) == 1) {
+			/* j <-- Integerify(X) mod NROM */
+			j &= NROM - 1;
+			V_j = &VROM[j * s];
+		} else {
+			/* j <-- Wrap(Integerify(X), i) */
+			j &= n - 1;
+			j += N - 1 - n;
+			V_j = &V[j * s];
+		}
+
+		/* X <-- X \xor V_j */
+		/* 4: X <-- H(X) */
+		X = XY;
+		blockmix_xor(Y, V_j, X, r, 1, S);
+	} else if (flags & YESCRYPT_RW) {
+		uint32_t n;
+		salsa20_blk_t * V_n, * V_j;
+
+		/* 4: X <-- H(X) */
+		/* 3: V_i <-- X */
+		Y = &V[s];
+		blockmix(X, Y, r, S);
+
+		/* 4: X <-- H(X) */
+		/* 3: V_i <-- X */
+		X = &V[2 * s];
+		blockmix(Y, X, r, S);
+		j = integerify(X, r);
+
+		for (n = 2; n < N; n <<= 1) {
+			uint32_t m = (n < N / 2) ? n : (N - 1 - n);
+
+			V_n = &V[n * s];
+
+			/* 2: for i = 0 to N - 1 do */
+			for (i = 1; i < m; i += 2) {
+				Y = &V_n[i * s];
+
+				/* j <-- Wrap(Integerify(X), i) */
+				j &= n - 1;
+				j += i - 1;
+				V_j = &V[j * s];
+
+				/* X <-- X \xor V_j */
+				/* 4: X <-- H(X) */
+				/* 3: V_i <-- X */
+				j = blockmix_xor(X, V_j, Y, r, 0, S);
+
+				/* j <-- Wrap(Integerify(X), i) */
+				j &= n - 1;
+				j += i;
+				V_j = &V[j * s];
+
+				/* X <-- X \xor V_j */
+				/* 4: X <-- H(X) */
+				/* 3: V_i <-- X */
+				X = &V_n[(i + 1) * s];
+				j = blockmix_xor(Y, V_j, X, r, 0, S);
+			}
+		}
+
+		n >>= 1;
+
+		/* j <-- Wrap(Integerify(X), i) */
+		j &= n - 1;
+		j += N - 2 - n;
+		V_j = &V[j * s];
+
+		/* X <-- X \xor V_j */
+		/* 4: X <-- H(X) */
+		/* 3: V_i <-- X */
+		Y = &V[(N - 1) * s];
+		j = blockmix_xor(X, V_j, Y, r, 0, S);
+
+		/* j <-- Wrap(Integerify(X), i) */
+		j &= n - 1;
+		j += N - 1 - n;
+		V_j = &V[j * s];
+
+		/* X <-- X \xor V_j */
+		/* 4: X <-- H(X) */
+		X = XY;
+		blockmix_xor(Y, V_j, X, r, 0, S);
+	} else {
+		/* 2: for i = 0 to N - 1 do */
+		for (i = 1; i < N - 1; i += 2) {
+			/* 4: X <-- H(X) */
+			/* 3: V_i <-- X */
+			Y = &V[i * s];
+			blockmix(X, Y, r, S);
+
+			/* 4: X <-- H(X) */
+			/* 3: V_i <-- X */
+			X = &V[(i + 1) * s];
+			blockmix(Y, X, r, S);
+		}
+
+		/* 4: X <-- H(X) */
+		/* 3: V_i <-- X */
+		Y = &V[i * s];
+		blockmix(X, Y, r, S);
+
+		/* 4: X <-- H(X) */
+		X = XY;
+		blockmix(Y, X, r, S);
+	}
+
+	/* B' <-- X */
+	for (k = 0; k < 2 * r; k++) {
+		for (i = 0; i < 16; i++) {
+			le32enc(&B[(k * 16 + (i * 5 % 16)) * 4], X[k].w[i]);
+		}
+	}
+}
+
+/**
+ * smix2(B, r, N, Nloop, flags, V, NROM, shared, XY, S):
+ * Compute second loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 256r bytes in length.  The value N must be a power of 2
+ * greater than 1.  The value Nloop must be even.  The array V must be aligned
+ * to a multiple of 64 bytes, and arrays B and XY to a multiple of at least 16
+ * bytes (aligning them to 64 bytes as well saves cache lines, but might result
+ * in cache bank conflicts).
+ */
+static void
+smix2(uint8_t * B, size_t r, uint32_t N, uint64_t Nloop,
+    yescrypt_flags_t flags, salsa20_blk_t * V, uint32_t NROM,
+    const yescrypt_shared_t * shared, salsa20_blk_t * XY, void * S)
+{
+	const salsa20_blk_t * VROM = shared->shared1.aligned;
+	uint32_t VROM_mask = shared->mask1;
+	size_t s = 2 * r;
+	salsa20_blk_t * X = XY, * Y = &XY[s];
+	uint64_t i;
+	uint32_t j;
+	size_t k;
+
+	if (Nloop == 0)
+		return;
+
+	/* X <-- B' */
+	/* 3: V_i <-- X */
+	for (k = 0; k < 2 * r; k++) {
+		for (i = 0; i < 16; i++) {
+			X[k].w[i] = le32dec(&B[(k * 16 + (i * 5 % 16)) * 4]);
+		}
+	}
+
+	i = Nloop / 2;
+
+	/* 7: j <-- Integerify(X) mod N */
+	j = integerify(X, r) & (N - 1);
+
+/*
+ * Normally, NROM implies YESCRYPT_RW, but we check for these separately
+ * because YESCRYPT_PARALLEL_SMIX resets YESCRYPT_RW for the smix2() calls
+ * operating on the entire V.
+ */
+	if (NROM && (flags & YESCRYPT_RW)) {
+		/* 6: for i = 0 to N - 1 do */
+		for (i = 0; i < Nloop; i += 2) {
+			salsa20_blk_t * V_j = &V[j * s];
+
+			/* 8: X <-- H(X \xor V_j) */
+			/* V_j <-- Xprev \xor V_j */
+			/* j <-- Integerify(X) mod NROM */
+			j = blockmix_xor_save(X, V_j, Y, r, S);
+
+			if (((i + 1) & VROM_mask) == 1) {
+				const salsa20_blk_t * VROM_j;
+
+				j &= NROM - 1;
+				VROM_j = &VROM[j * s];
+
+				/* X <-- H(X \xor VROM_j) */
+				/* 7: j <-- Integerify(X) mod N */
+				j = blockmix_xor(Y, VROM_j, X, r, 1, S);
+			} else {
+				j &= N - 1;
+				V_j = &V[j * s];
+
+				/* 8: X <-- H(X \xor V_j) */
+				/* V_j <-- Xprev \xor V_j */
+				/* j <-- Integerify(X) mod NROM */
+				j = blockmix_xor_save(Y, V_j, X, r, S);
+			}
+			j &= N - 1;
+			V_j = &V[j * s];
+		}
+	} else if (NROM) {
+		/* 6: for i = 0 to N - 1 do */
+		for (i = 0; i < Nloop; i += 2) {
+			const salsa20_blk_t * V_j = &V[j * s];
+
+			/* 8: X <-- H(X \xor V_j) */
+			/* V_j <-- Xprev \xor V_j */
+			/* j <-- Integerify(X) mod NROM */
+			j = blockmix_xor(X, V_j, Y, r, 0, S);
+
+			if (((i + 1) & VROM_mask) == 1) {
+				j &= NROM - 1;
+				V_j = &VROM[j * s];
+			} else {
+				j &= N - 1;
+				V_j = &V[j * s];
+			}
+
+			/* X <-- H(X \xor VROM_j) */
+			/* 7: j <-- Integerify(X) mod N */
+			j = blockmix_xor(Y, V_j, X, r, 1, S);
+			j &= N - 1;
+			V_j = &V[j * s];
+		}
+	} else if (flags & YESCRYPT_RW) {
+		/* 6: for i = 0 to N - 1 do */
+		do {
+			salsa20_blk_t * V_j = &V[j * s];
+
+			/* 8: X <-- H(X \xor V_j) */
+			/* V_j <-- Xprev \xor V_j */
+			/* 7: j <-- Integerify(X) mod N */
+			j = blockmix_xor_save(X, V_j, Y, r, S);
+			j &= N - 1;
+			V_j = &V[j * s];
+
+			/* 8: X <-- H(X \xor V_j) */
+			/* V_j <-- Xprev \xor V_j */
+			/* 7: j <-- Integerify(X) mod N */
+			j = blockmix_xor_save(Y, V_j, X, r, S);
+			j &= N - 1;
+		} while (--i);
+	} else {
+		/* 6: for i = 0 to N - 1 do */
+		do {
+			const salsa20_blk_t * V_j = &V[j * s];
+
+			/* 8: X <-- H(X \xor V_j) */
+			/* 7: j <-- Integerify(X) mod N */
+			j = blockmix_xor(X, V_j, Y, r, 0, S);
+			j &= N - 1;
+			V_j = &V[j * s];
+
+			/* 8: X <-- H(X \xor V_j) */
+			/* 7: j <-- Integerify(X) mod N */
+			j = blockmix_xor(Y, V_j, X, r, 0, S);
+			j &= N - 1;
+		} while (--i);
+	}
+
+	/* 10: B' <-- X */
+	for (k = 0; k < 2 * r; k++) {
+		for (i = 0; i < 16; i++) {
+			le32enc(&B[(k * 16 + (i * 5 % 16)) * 4], X[k].w[i]);
+		}
+	}
+}
+
+/**
+ * p2floor(x):
+ * Largest power of 2 not greater than argument.
+ */
+static uint64_t
+p2floor(uint64_t x)
+{
+	uint64_t y;
+	while ((y = x & (x - 1)))
+		x = y;
+	return x;
+}
+
+/**
+ * smix(B, r, N, p, t, flags, V, NROM, shared, XY, S):
+ * Compute B = SMix_r(B, N).  The input B must be 128rp bytes in length; the
+ * temporary storage V must be 128rN bytes in length; the temporary storage XY
+ * must be 256r or 256rp bytes in length (the larger size is required with
+ * OpenMP-enabled builds).  The value N must be a power of 2 greater than 1.
+ * The array V must be aligned to a multiple of 64 bytes, and arrays B and
+ * XY to a multiple of at least 16 bytes (aligning them to 64 bytes as well
+ * saves cache lines and helps avoid false sharing in OpenMP-enabled builds
+ * when p > 1, but it might also result in cache bank conflicts).
+ */
+static void
+smix(uint8_t * B, size_t r, uint32_t N, uint32_t p, uint32_t t,
+    yescrypt_flags_t flags,
+    salsa20_blk_t * V, uint32_t NROM, const yescrypt_shared_t * shared,
+    salsa20_blk_t * XY, void * S)
+{
+	size_t s = 2 * r;
+	uint32_t Nchunk = N / p;
+	uint64_t Nloop_all, Nloop_rw;
+	uint32_t i;
+
+	Nloop_all = Nchunk;
+	if (flags & YESCRYPT_RW) {
+		if (t <= 1) {
+			if (t)
+				Nloop_all *= 2; /* 2/3 */
+			Nloop_all = (Nloop_all + 2) / 3; /* 1/3, round up */
+		} else {
+			Nloop_all *= t - 1;
+		}
+	} else if (t) {
+		if (t == 1)
+			Nloop_all += (Nloop_all + 1) / 2; /* 1.5, round up */
+		Nloop_all *= t;
+	}
+
+	Nloop_rw = 0;
+	if (flags & __YESCRYPT_INIT_SHARED)
+		Nloop_rw = Nloop_all;
+	else if (flags & YESCRYPT_RW)
+		Nloop_rw = Nloop_all / p;
+
+	Nchunk &= ~(uint32_t)1; /* round down to even */
+	Nloop_all++; Nloop_all &= ~(uint64_t)1; /* round up to even */
+	Nloop_rw &= ~(uint64_t)1; /* round down to even */
+
+#ifdef _OPENMP
+#pragma omp parallel if (p > 1) default(none) private(i) shared(B, r, N, p, flags, V, NROM, shared, XY, S, s, Nchunk, Nloop_all, Nloop_rw)
+	{
+#pragma omp for
+#endif
+	for (i = 0; i < p; i++) {
+		uint32_t Vchunk = i * Nchunk;
+		uint8_t * Bp = &B[128 * r * i];
+		salsa20_blk_t * Vp = &V[Vchunk * s];
+#ifdef _OPENMP
+		salsa20_blk_t * XYp = &XY[i * (2 * s)];
+#else
+		salsa20_blk_t * XYp = XY;
+#endif
+		uint32_t Np = (i < p - 1) ? Nchunk : (N - Vchunk);
+		void * Sp = S ? ((uint8_t *)S + i * S_SIZE_ALL) : S;
+		if (Sp)
+			smix1(Bp, 1, S_SIZE_ALL / 128,
+			    flags & ~YESCRYPT_PWXFORM,
+			    Sp, NROM, shared, XYp, NULL);
+		if (!(flags & __YESCRYPT_INIT_SHARED_2))
+			smix1(Bp, r, Np, flags, Vp, NROM, shared, XYp, Sp);
+		smix2(Bp, r, p2floor(Np), Nloop_rw, flags, Vp,
+		    NROM, shared, XYp, Sp);
+	}
+
+	if (Nloop_all > Nloop_rw) {
+#ifdef _OPENMP
+#pragma omp for
+#endif
+		for (i = 0; i < p; i++) {
+			uint8_t * Bp = &B[128 * r * i];
+#ifdef _OPENMP
+			salsa20_blk_t * XYp = &XY[i * (2 * s)];
+#else
+			salsa20_blk_t * XYp = XY;
+#endif
+			void * Sp = S ? ((uint8_t *)S + i * S_SIZE_ALL) : S;
+			smix2(Bp, r, N, Nloop_all - Nloop_rw,
+			    flags & ~YESCRYPT_RW, V, NROM, shared, XYp, Sp);
+		}
+	}
+#ifdef _OPENMP
+	}
+#endif
+}
+
+/**
+ * yescrypt_kdf(shared, local, passwd, passwdlen, salt, saltlen,
+ *     N, r, p, t, flags, buf, buflen):
+ * Compute scrypt(passwd[0 .. passwdlen - 1], salt[0 .. saltlen - 1], N, r,
+ * p, buflen), or a revision of scrypt as requested by flags and shared, and
+ * write the result into buf.  The parameters r, p, and buflen must satisfy
+ * r * p < 2^30 and buflen <= (2^32 - 1) * 32.  The parameter N must be a power
+ * of 2 greater than 1.  (This optimized implementation currently additionally
+ * limits N to the range from 8 to 2^31, but other implementation might not.)
+ *
+ * t controls computation time while not affecting peak memory usage.  shared
+ * and flags may request special modes as described in yescrypt.h.  local is
+ * the thread-local data structure, allowing to preserve and reuse a memory
+ * allocation across calls, thereby reducing its overhead.
+ *
+ * Return 0 on success; or -1 on error.
+ */
+static int
+yescrypt_kdf(const yescrypt_shared_t * shared, yescrypt_local_t * local,
+    const uint8_t * passwd, size_t passwdlen,
+    const uint8_t * salt, size_t saltlen,
+    uint64_t N, uint32_t r, uint32_t p, uint32_t t, yescrypt_flags_t flags,
+    uint8_t * buf, size_t buflen)
+{
+	yescrypt_region_t tmp;
+	uint64_t NROM;
+	size_t B_size, V_size, XY_size, need;
+	uint8_t * B, * S;
+	salsa20_blk_t * V, * XY;
+	uint8_t sha256[32];
+
+	/*
+	 * YESCRYPT_PARALLEL_SMIX is a no-op at p = 1 for its intended purpose,
+	 * so don't let it have side-effects.  Without this adjustment, it'd
+	 * enable the SHA-256 password pre-hashing and output post-hashing,
+	 * because any deviation from classic scrypt implies those.
+	 */
+	if (p == 1)
+		flags &= ~YESCRYPT_PARALLEL_SMIX;
+
+	/* Sanity-check parameters */
+	if (flags & ~YESCRYPT_KNOWN_FLAGS) {
+		errno = EINVAL;
+		return -1;
+	}
+#if SIZE_MAX > UINT32_MAX
+	if (buflen > (((uint64_t)(1) << 32) - 1) * 32) {
+		errno = EFBIG;
+		return -1;
+	}
+#endif
+	if ((uint64_t)(r) * (uint64_t)(p) >= (1 << 30)) {
+		errno = EFBIG;
+		return -1;
+	}
+	if (N > UINT32_MAX) {
+		errno = EFBIG;
+		return -1;
+	}
+	if (((N & (N - 1)) != 0) || (N <= 7) || (r < 1) || (p < 1)) {
+		errno = EINVAL;
+		return -1;
+	}
+	if ((flags & YESCRYPT_PARALLEL_SMIX) && (N / p <= 7)) {
+		errno = EINVAL;
+		return -1;
+	}
+	if ((r > SIZE_MAX / 256 / p) ||
+	    (N > SIZE_MAX / 128 / r)) {
+		errno = ENOMEM;
+		return -1;
+	}
+#ifdef _OPENMP
+	if (!(flags & YESCRYPT_PARALLEL_SMIX) &&
+	    (N > SIZE_MAX / 128 / (r * p))) {
+		errno = ENOMEM;
+		return -1;
+	}
+#endif
+	if ((flags & YESCRYPT_PWXFORM) &&
+#ifndef _OPENMP
+	    (flags & YESCRYPT_PARALLEL_SMIX) &&
+#endif
+	    p > SIZE_MAX / S_SIZE_ALL) {
+		errno = ENOMEM;
+		return -1;
+	}
+
+	NROM = 0;
+	if (shared->shared1.aligned) {
+		NROM = shared->shared1.aligned_size / ((size_t)128 * r);
+		if (NROM > UINT32_MAX) {
+			errno = EFBIG;
+			return -1;
+		}
+		if (((NROM & (NROM - 1)) != 0) || (NROM <= 7) ||
+		    !(flags & YESCRYPT_RW)) {
+			errno = EINVAL;
+			return -1;
+		}
+	}
+
+	/* Allocate memory */
+	V = NULL;
+	V_size = (size_t)128 * r * N;
+#ifdef _OPENMP
+	if (!(flags & YESCRYPT_PARALLEL_SMIX))
+		V_size *= p;
+#endif
+	need = V_size;
+	if (flags & __YESCRYPT_INIT_SHARED) {
+		if (local->aligned_size < need) {
+			if (local->base || local->aligned ||
+			    local->base_size || local->aligned_size) {
+				errno = EINVAL;
+				return -1;
+			}
+			if (!alloc_region(local, need))
+				return -1;
+		}
+		V = (salsa20_blk_t *)local->aligned;
+		need = 0;
+	}
+	B_size = (size_t)128 * r * p;
+	need += B_size;
+	if (need < B_size) {
+		errno = ENOMEM;
+		return -1;
+	}
+	XY_size = (size_t)256 * r;
+#ifdef _OPENMP
+	XY_size *= p;
+#endif
+	need += XY_size;
+	if (need < XY_size) {
+		errno = ENOMEM;
+		return -1;
+	}
+	if (flags & YESCRYPT_PWXFORM) {
+		size_t S_size = S_SIZE_ALL;
+#ifdef _OPENMP
+		S_size *= p;
+#else
+		if (flags & YESCRYPT_PARALLEL_SMIX)
+			S_size *= p;
+#endif
+		need += S_size;
+		if (need < S_size) {
+			errno = ENOMEM;
+			return -1;
+		}
+	}
+	if (flags & __YESCRYPT_INIT_SHARED) {
+		if (!alloc_region(&tmp, need))
+			return -1;
+		B = (uint8_t *)tmp.aligned;
+		XY = (salsa20_blk_t *)((uint8_t *)B + B_size);
+	} else {
+		init_region(&tmp);
+		if (local->aligned_size < need) {
+			if (free_region(local))
+				return -1;
+			if (!alloc_region(local, need))
+				return -1;
+		}
+		B = (uint8_t *)local->aligned;
+		V = (salsa20_blk_t *)((uint8_t *)B + B_size);
+		XY = (salsa20_blk_t *)((uint8_t *)V + V_size);
+	}
+	S = NULL;
+	if (flags & YESCRYPT_PWXFORM)
+		S = (uint8_t *)XY + XY_size;
+
+	if (t || flags) {
+		SHA256_CTX ctx;
+		SHA256_Init(&ctx);
+		SHA256_Update(&ctx, passwd, passwdlen);
+		SHA256_Final(sha256, &ctx);
+		passwd = sha256;
+		passwdlen = sizeof(sha256);
+	}
+
+	/* 1: (B_0 ... B_{p-1}) <-- PBKDF2(P, S, 1, p * MFLen) */
+	PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, 1, B, B_size);
+
+	if (t || flags)
+		memcpy(sha256, B, sizeof(sha256));
+
+	if (p == 1 || (flags & YESCRYPT_PARALLEL_SMIX)) {
+		smix(B, r, N, p, t, flags, V, NROM, shared, XY, S);
+	} else {
+		uint32_t i;
+
+		/* 2: for i = 0 to p - 1 do */
+#ifdef _OPENMP
+#pragma omp parallel for default(none) private(i) shared(B, r, N, p, t, flags, V, NROM, shared, XY, S)
+#endif
+		for (i = 0; i < p; i++) {
+			/* 3: B_i <-- MF(B_i, N) */
+#ifdef _OPENMP
+			smix(&B[(size_t)128 * r * i], r, N, 1, t, flags,
+			    &V[(size_t)2 * r * i * N],
+			    NROM, shared,
+			    &XY[(size_t)4 * r * i],
+			    S ? &S[S_SIZE_ALL * i] : S);
+#else
+			smix(&B[(size_t)128 * r * i], r, N, 1, t, flags, V,
+			    NROM, shared, XY, S);
+#endif
+		}
+	}
+
+	/* 5: DK <-- PBKDF2(P, B, 1, dkLen) */
+	PBKDF2_SHA256(passwd, passwdlen, B, B_size, 1, buf, buflen);
+
+	/*
+	 * Except when computing classic scrypt, allow all computation so far
+	 * to be performed on the client.  The final steps below match those of
+	 * SCRAM (RFC 5802), so that an extension of SCRAM (with the steps so
+	 * far in place of SCRAM's use of PBKDF2 and with SHA-256 in place of
+	 * SCRAM's use of SHA-1) would be usable with yescrypt hashes.
+	 */
+	if ((t || flags) && buflen == sizeof(sha256)) {
+		/* Compute ClientKey */
+		{
+			HMAC_SHA256_CTX ctx;
+			HMAC_SHA256_Init(&ctx, buf, buflen);
+			HMAC_SHA256_Update(&ctx, "Client Key", 10);
+			HMAC_SHA256_Final(sha256, &ctx);
+		}
+		/* Compute StoredKey */
+		{
+			SHA256_CTX ctx;
+			SHA256_Init(&ctx);
+			SHA256_Update(&ctx, sha256, sizeof(sha256));
+			SHA256_Final(buf, &ctx);
+		}
+	}
+
+	if (free_region(&tmp))
+		return -1;
+
+	/* Success! */
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt.c
new file mode 100644
index 0000000..10546ce
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt.c
@@ -0,0 +1,115 @@
+/*-
+ * Copyright 2014 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+#include "yescrypt.h"
+#include "sha256.c"
+#include "yescrypt-best.c"
+
+#define YESCRYPT_N 2048
+#define YESCRYPT_R 8
+#define YESCRYPT_P 1
+#define YESCRYPT_T 0
+#define YESCRYPT_FLAGS (YESCRYPT_RW | YESCRYPT_PWXFORM)
+
+static int yescrypt_bitzeny(const uint8_t *passwd, size_t passwdlen,
+                            const uint8_t *salt, size_t saltlen,
+                            uint8_t *buf, size_t buflen)
+{
+    static __thread int initialized = 0;
+    static __thread yescrypt_shared_t shared;
+    static __thread yescrypt_local_t local;
+    int retval;
+    if (!initialized) {
+        /* "shared" could in fact be shared, but it's simpler to keep it private
+         * along with "local".  It's dummy and tiny anyway. */
+        if (yescrypt_init_shared(&shared, NULL, 0,
+                                 0, 0, 0, YESCRYPT_SHARED_DEFAULTS, 0, NULL, 0))
+            return -1;
+        if (yescrypt_init_local(&local)) {
+            yescrypt_free_shared(&shared);
+            return -1;
+        }
+        initialized = 1;
+    }
+    retval = yescrypt_kdf(&shared, &local, passwd, passwdlen, salt, saltlen,
+                          YESCRYPT_N, YESCRYPT_R, YESCRYPT_P, YESCRYPT_T,
+                          YESCRYPT_FLAGS, buf, buflen);
+#if 0
+    if (yescrypt_free_local(&local)) {
+        yescrypt_free_shared(&shared);
+        return -1;
+    }
+    if (yescrypt_free_shared(&shared))
+        return -1;
+    initialized = 0;
+#endif
+    if (retval < 0) {
+        yescrypt_free_local(&local);
+        yescrypt_free_shared(&shared);
+    }
+    return retval;
+}
+
+static void yescrypt_hash(const char *input, char *output)
+{
+    yescrypt_bitzeny((const uint8_t *) input, 80,
+                     (const uint8_t *) input, 80,
+                     (uint8_t *) output, 32);
+}
+
+#include <stdbool.h>
+struct work_restart {
+	volatile unsigned long	restart;
+	char			padding[128 - sizeof(unsigned long)];
+};
+
+extern struct work_restart *work_restart;
+extern bool fulltest(const uint32_t *hash, const uint32_t *target);
+
+static int pretest(const uint32_t *hash, const uint32_t *target)
+{
+	return hash[7] < target[7];
+}
+
+int scanhash_yescrypt(int thr_id, uint32_t *pdata, const uint32_t *ptarget,
+		      uint32_t max_nonce, unsigned long *hashes_done)
+{
+	uint32_t data[20] __attribute__((aligned(128)));
+	uint32_t hash[8] __attribute__((aligned(32)));
+	uint32_t n = pdata[19] - 1;
+	const uint32_t first_nonce = pdata[19];
+
+	for (int i = 0; i < 20; i++) {
+		be32enc(&data[i], pdata[i]);
+	}
+	do {
+		be32enc(&data[19], ++n);
+		yescrypt_hash((char *)data, (char *)hash);
+		if (pretest(hash, ptarget) && fulltest(hash, ptarget)) {
+			pdata[19] = n;
+			*hashes_done = n - first_nonce + 1;
+			return 1;
+		}
+	} while (n < max_nonce && !work_restart[thr_id].restart);
+	
+	*hashes_done = n - first_nonce + 1;
+	pdata[19] = n;
+	return 0;
+}
+
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt.h
new file mode 100644
index 0000000..ab4467b
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yescrypt.h
@@ -0,0 +1,274 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2013,2014 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ */
+#ifndef _YESCRYPT_H_
+#define _YESCRYPT_H_
+
+#include <stdint.h>
+#include <stdlib.h> /* for size_t */
+
+/**
+ * Internal type used by the memory allocator.  Please do not use it directly.
+ * Use yescrypt_shared_t and yescrypt_local_t as appropriate instead, since
+ * they might differ from each other in a future version.
+ */
+typedef struct {
+	void * base, * aligned;
+	size_t base_size, aligned_size;
+} yescrypt_region_t;
+
+/**
+ * Types for shared (ROM) and thread-local (RAM) data structures.
+ */
+typedef yescrypt_region_t yescrypt_shared1_t;
+typedef struct {
+	yescrypt_shared1_t shared1;
+	uint32_t mask1;
+} yescrypt_shared_t;
+typedef yescrypt_region_t yescrypt_local_t;
+
+/**
+ * Possible values for yescrypt_init_shared()'s flags argument.
+ */
+typedef enum {
+	YESCRYPT_SHARED_DEFAULTS = 0,
+	YESCRYPT_SHARED_PREALLOCATED = 0x100
+} yescrypt_init_shared_flags_t;
+
+/**
+ * Possible values for the flags argument of yescrypt_kdf(),
+ * yescrypt_gensalt_r(), yescrypt_gensalt().  These may be OR'ed together,
+ * except that YESCRYPT_WORM and YESCRYPT_RW are mutually exclusive.
+ * Please refer to the description of yescrypt_kdf() below for the meaning of
+ * these flags.
+ */
+typedef enum {
+/* public */
+	YESCRYPT_WORM = 0,
+	YESCRYPT_RW = 1,
+	YESCRYPT_PARALLEL_SMIX = 2,
+	YESCRYPT_PWXFORM = 4,
+/* private */
+	__YESCRYPT_INIT_SHARED_1 = 0x10000,
+	__YESCRYPT_INIT_SHARED_2 = 0x20000,
+	__YESCRYPT_INIT_SHARED = 0x30000
+} yescrypt_flags_t;
+
+#define YESCRYPT_KNOWN_FLAGS \
+	(YESCRYPT_RW | YESCRYPT_PARALLEL_SMIX | YESCRYPT_PWXFORM | \
+	__YESCRYPT_INIT_SHARED)
+
+/**
+ * yescrypt_init_shared(shared, param, paramlen, N, r, p, flags, mask,
+ *     buf, buflen):
+ * Optionally allocate memory for and initialize the shared (ROM) data
+ * structure.  The parameters N, r, and p must satisfy the same conditions as
+ * with crypto_scrypt().  param and paramlen specify a local parameter with
+ * which the ROM is seeded.  If buf is not NULL, then it is used to return
+ * buflen bytes of message digest for the initialized ROM (the caller may use
+ * this to verify that the ROM has been computed in the same way that it was on
+ * a previous run).
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * If bit YESCRYPT_SHARED_PREALLOCATED in flags is set, then memory for the
+ * ROM is assumed to have been preallocated by the caller, with
+ * shared->shared1.aligned being the start address of the ROM and
+ * shared->shared1.aligned_size being its size (which must be consistent with
+ * N, r, and p).  This may be used e.g. when the ROM is to be placed in a SysV
+ * shared memory segment allocated by the caller.
+ *
+ * mask controls the frequency of ROM accesses by yescrypt_kdf().  Normally it
+ * should be set to 1, to interleave RAM and ROM accesses, which works well
+ * when both regions reside in the machine's RAM anyway.  Other values may be
+ * used e.g. when the ROM is memory-mapped from a disk file.  Recommended mask
+ * values are powers of 2 minus 1 or minus 2.  Here's the effect of some mask
+ * values:
+ * mask	value	ROM accesses in SMix 1st loop	ROM accesses in SMix 2nd loop
+ *	0		0				1/2
+ *	1		1/2				1/2
+ *	2		0				1/4
+ *	3		1/4				1/4
+ *	6		0				1/8
+ *	7		1/8				1/8
+ *	14		0				1/16
+ *	15		1/16				1/16
+ *	1022		0				1/1024
+ *	1023		1/1024				1/1024
+ *
+ * Actual computation of the ROM contents may be avoided, if you don't intend
+ * to use a ROM but need a dummy shared structure, by calling this function
+ * with NULL, 0, 0, 0, 0, YESCRYPT_SHARED_DEFAULTS, 0, NULL, 0 for the
+ * arguments starting with param and on.
+ *
+ * MT-safe as long as shared is local to the thread.
+ */
+static int yescrypt_init_shared(yescrypt_shared_t * __shared,
+    const uint8_t * __param, size_t __paramlen,
+    uint64_t __N, uint32_t __r, uint32_t __p,
+    yescrypt_init_shared_flags_t __flags, uint32_t __mask,
+    uint8_t * __buf, size_t __buflen);
+
+/**
+ * yescrypt_free_shared(shared):
+ * Free memory that had been allocated with yescrypt_init_shared().
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * MT-safe as long as shared is local to the thread.
+ */
+static int yescrypt_free_shared(yescrypt_shared_t * __shared);
+
+/**
+ * yescrypt_init_local(local):
+ * Initialize the thread-local (RAM) data structure.  Actual memory allocation
+ * is currently fully postponed until a call to yescrypt_kdf() or yescrypt_r().
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * MT-safe as long as local is local to the thread.
+ */
+static int yescrypt_init_local(yescrypt_local_t * __local);
+
+/**
+ * yescrypt_free_local(local):
+ * Free memory that may have been allocated for an initialized thread-local
+ * (RAM) data structure.
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * MT-safe as long as local is local to the thread.
+ */
+static int yescrypt_free_local(yescrypt_local_t * __local);
+
+/**
+ * yescrypt_kdf(shared, local, passwd, passwdlen, salt, saltlen,
+ *     N, r, p, t, flags, buf, buflen):
+ * Compute scrypt(passwd[0 .. passwdlen - 1], salt[0 .. saltlen - 1], N, r,
+ * p, buflen), or a revision of scrypt as requested by flags and shared, and
+ * write the result into buf.  The parameters N, r, p, and buflen must satisfy
+ * the same conditions as with crypto_scrypt().  t controls computation time
+ * while not affecting peak memory usage.  shared and flags may request
+ * special modes as described below.  local is the thread-local data
+ * structure, allowing to preserve and reuse a memory allocation across calls,
+ * thereby reducing its overhead.
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * t controls computation time.  t = 0 is optimal in terms of achieving the
+ * highest area-time for ASIC attackers.  Thus, higher computation time, if
+ * affordable, is best achieved by increasing N rather than by increasing t.
+ * However, if the higher memory usage (which goes along with higher N) is not
+ * affordable, or if fine-tuning of the time is needed (recall that N must be a
+ * power of 2), then t = 1 or above may be used to increase time while staying
+ * at the same peak memory usage.  t = 1 increases the time by 25% and
+ * decreases the normalized area-time to 96% of optimal.  (Of course, in
+ * absolute terms the area-time increases with higher t.  It's just that it
+ * would increase slightly more with higher N*r rather than with higher t.)
+ * t = 2 increases the time by another 20% and decreases the normalized
+ * area-time to 89% of optimal.  Thus, these two values are reasonable to use
+ * for fine-tuning.  Values of t higher than 2 result in further increase in
+ * time while reducing the efficiency much further (e.g., down to around 50% of
+ * optimal for t = 5, which runs 3 to 4 times slower than t = 0, with exact
+ * numbers varying by the flags settings).
+ *
+ * Classic scrypt is available by setting t = 0 and flags to YESCRYPT_WORM and
+ * passing a dummy shared structure (see the description of
+ * yescrypt_init_shared() above for how to produce one).  In this mode, the
+ * thread-local memory region (RAM) is first sequentially written to and then
+ * randomly read from.  This algorithm is friendly towards time-memory
+ * tradeoffs (TMTO), available both to defenders (albeit not in this
+ * implementation) and to attackers.
+ *
+ * Setting YESCRYPT_RW adds extra random reads and writes to the thread-local
+ * memory region (RAM), which makes TMTO a lot less efficient.  This may be
+ * used to slow down the kinds of attackers who would otherwise benefit from
+ * classic scrypt's efficient TMTO.  Since classic scrypt's TMTO allows not
+ * only for the tradeoff, but also for a decrease of attacker's area-time (by
+ * up to a constant factor), setting YESCRYPT_RW substantially increases the
+ * cost of attacks in area-time terms as well.  Yet another benefit of it is
+ * that optimal area-time is reached at an earlier time than with classic
+ * scrypt, and t = 0 actually corresponds to this earlier completion time,
+ * resulting in quicker hash computations (and thus in higher request rate
+ * capacity).  Due to these properties, YESCRYPT_RW should almost always be
+ * set, except when compatibility with classic scrypt or TMTO-friendliness are
+ * desired.
+ *
+ * YESCRYPT_PARALLEL_SMIX moves parallelism that is present with p > 1 to a
+ * lower level as compared to where it is in classic scrypt.  This reduces
+ * flexibility for efficient computation (for both attackers and defenders) by
+ * requiring that, short of resorting to TMTO, the full amount of memory be
+ * allocated as needed for the specified p, regardless of whether that
+ * parallelism is actually being fully made use of or not.  (For comparison, a
+ * single instance of classic scrypt may be computed in less memory without any
+ * CPU time overhead, but in more real time, by not making full use of the
+ * parallelism.)  This may be desirable when the defender has enough memory
+ * with sufficiently low latency and high bandwidth for efficient full parallel
+ * execution, yet the required memory size is high enough that some likely
+ * attackers might end up being forced to choose between using higher latency
+ * memory than they could use otherwise (waiting for data longer) or using TMTO
+ * (waiting for data more times per one hash computation).  The area-time cost
+ * for other kinds of attackers (who would use the same memory type and TMTO
+ * factor or no TMTO either way) remains roughly the same, given the same
+ * running time for the defender.  In the TMTO-friendly YESCRYPT_WORM mode, as
+ * long as the defender has enough memory that is just as fast as the smaller
+ * per-thread regions would be, doesn't expect to ever need greater
+ * flexibility (except possibly via TMTO), and doesn't need backwards
+ * compatibility with classic scrypt, there are no other serious drawbacks to
+ * this setting.  In the YESCRYPT_RW mode, which is meant to discourage TMTO,
+ * this new approach to parallelization makes TMTO less inefficient.  (This is
+ * an unfortunate side-effect of avoiding some random writes, as we have to in
+ * order to allow for parallel threads to access a common memory region without
+ * synchronization overhead.)  Thus, in this mode this setting poses an extra
+ * tradeoff of its own (higher area-time cost for a subset of attackers vs.
+ * better TMTO resistance).  Setting YESCRYPT_PARALLEL_SMIX also changes the
+ * way the running time is to be controlled from N*r*p (for classic scrypt) to
+ * N*r (in this modification).  All of this applies only when p > 1.  For
+ * p = 1, this setting is a no-op.
+ *
+ * Passing a real shared structure, with ROM contents previously computed by
+ * yescrypt_init_shared(), enables the use of ROM and requires YESCRYPT_RW for
+ * the thread-local RAM region.  In order to allow for initialization of the
+ * ROM to be split into a separate program, the shared->shared1.aligned and
+ * shared->shared1.aligned_size fields may be set by the caller of
+ * yescrypt_kdf() manually rather than with yescrypt_init_shared().
+ *
+ * local must be initialized with yescrypt_init_local().
+ *
+ * MT-safe as long as local and buf are local to the thread.
+ */
+static int yescrypt_kdf(const yescrypt_shared_t * __shared,
+    yescrypt_local_t * __local,
+    const uint8_t * __passwd, size_t __passwdlen,
+    const uint8_t * __salt, size_t __saltlen,
+    uint64_t __N, uint32_t __r, uint32_t __p, uint32_t __t,
+    yescrypt_flags_t __flags,
+    uint8_t * __buf, size_t __buflen);
+
+#endif /* !_YESCRYPT_H_ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/.dirstamp b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/.dirstamp
new file mode 100644
index 0000000..e69de29
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/PERFORMANCE b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/PERFORMANCE
new file mode 100644
index 0000000..99cddd0
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/PERFORMANCE
@@ -0,0 +1,95 @@
+Included with yespower is the "benchmark" program, which is built by
+simply invoking "make".  When invoked without parameters, it tests
+yespower 0.5 at N = 2048, r = 8, which appears to be the lowest setting
+in use by existing cryptocurrencies.  On an i7-4770K with 4x DDR3-1600
+(on two memory channels) running CentOS 7 for x86-64 (and built with
+CentOS 7's default version of gcc) and with thread affinity set, this
+reports between 3700 and 3800 hashes per second for both SSE2 and AVX
+builds, e.g.:
+
+$ GOMP_CPU_AFFINITY=0-7 OMP_NUM_THREADS=4 ./benchmark
+version=0.5 N=2048 r=8
+Will use 2048.00 KiB RAM
+a5 9f ec 4c 4f dd a1 6e 3b 14 05 ad da 66 d5 25 b6 8e 7c ad fc fe 6a c0 66 c7 ad 11 8c d8 05 90
+Benchmarking 1 thread ...
+1018 H/s real, 1018 H/s virtual (2047 hashes in 2.01 seconds)
+Benchmarking 4 threads ...
+3773 H/s real, 950 H/s virtual (8188 hashes in 2.17 seconds)
+min 0.984 ms, avg 1.052 ms, max 1.074 ms
+
+Running 8 threads (to match the logical rather than the physical CPU
+core count) results in very slightly worse performance on this system,
+but this might be the other way around on another and/or with other
+parameters.  Upgrading to yespower 1.0, performance at these parameters
+improves to almost 4000 hashes per second:
+
+$ GOMP_CPU_AFFINITY=0-7 OMP_NUM_THREADS=4 ./benchmark 10
+version=1.0 N=2048 r=8
+Will use 2048.00 KiB RAM
+d0 78 cd d4 cf 3f 5a a8 4e 3c 4a 58 66 29 81 d8 2d 27 e5 67 36 37 c4 be 77 63 61 32 24 c1 8a 93
+Benchmarking 1 thread ...
+1080 H/s real, 1080 H/s virtual (4095 hashes in 3.79 seconds)
+Benchmarking 4 threads ...
+3995 H/s real, 1011 H/s virtual (16380 hashes in 4.10 seconds)
+min 0.923 ms, avg 0.989 ms, max 1.137 ms
+
+Running 8 threads results in substantial slowdown with this new version
+(to between 3200 and 3400 hashes per second) because of cache thrashing.
+
+For higher settings such as those achieving 8 MiB instead of the 2 MiB
+above, this system performs at around 800 hashes per second for yespower
+0.5 and at around 830 hashes per second for yespower 1.0:
+
+$ GOMP_CPU_AFFINITY=0-7 OMP_NUM_THREADS=4 ./benchmark 5 2048 32
+version=0.5 N=2048 r=32
+Will use 8192.00 KiB RAM
+56 0a 89 1b 5c a2 e1 c6 36 11 1a 9f f7 c8 94 a5 d0 a2 60 2f 43 fd cf a5 94 9b 95 e2 2f e4 46 1e
+Benchmarking 1 thread ...
+265 H/s real, 265 H/s virtual (1023 hashes in 3.85 seconds)
+Benchmarking 4 threads ...
+803 H/s real, 200 H/s virtual (4092 hashes in 5.09 seconds)
+min 4.924 ms, avg 4.980 ms, max 5.074 ms
+
+$ GOMP_CPU_AFFINITY=0-7 OMP_NUM_THREADS=4 ./benchmark 10 2048 32
+version=1.0 N=2048 r=32
+Will use 8192.00 KiB RAM
+f7 69 26 ae 4a dc 56 53 90 2f f0 22 78 ea aa 39 eb 99 84 11 ac 3e a6 24 2e 19 6d fb c4 3d 68 25
+Benchmarking 1 thread ...
+275 H/s real, 275 H/s virtual (1023 hashes in 3.71 seconds)
+Benchmarking 4 threads ...
+831 H/s real, 209 H/s virtual (4092 hashes in 4.92 seconds)
+min 3.614 ms, avg 4.769 ms, max 5.011 ms
+
+Again, running 8 threads results in a slowdown, albeit not as bad as can
+be seen for lower settings.
+
+On x86(-64), the following code versions may reasonably be built: SSE2,
+AVX, and XOP.  (There's no reason to build for AVX2 and higher, which is
+unsuitable for and thus unused by current yespower anyway.  There's also
+no reason to build yespower as-is for SSE4, although there's a disabled
+by default 32-bit specific SSE4 code version that may be re-enabled and
+given a try if someone is so inclined; it may perform slightly slower or
+slightly faster across different systems.)
+
+yescrypt and especially yespower 1.0 have been designed to fit the SSE2
+instruction set almost perfectly, so there's very little benefit from
+the AVX and XOP builds, yet even at yespower 1.0 there may be
+performance differences between SSE2, AVX, and XOP builds within 2% or
+so (and it is unclear which is the fastest on a given system until
+tested, except that where XOP is supported it is almost always faster
+than AVX).
+
+Proper setting of thread affinities to run exactly one thread per
+physical CPU core is non-trivial.  In the above examples, it so happened
+that the first 4 logical CPU numbers corresponded to different physical
+cores, but this won't always be the case.  This can vary even between
+apparently similar systems.  On Linux, the mapping of logical CPUs to
+physical cores may be obtained from /proc/cpuinfo (on x86[-64] and MIC)
+or sysfs, which an optimized implementation of e.g. a cryptocurrency
+miner could use.  If you do not bother obtaining this information from
+the operating system, you might be better off not setting thread
+affinities at all (in order to avoid the risk of doing this incorrectly,
+which would have a greater negative performance impact) and/or running
+as many threads as there are logical CPUs.  Also, there's no certainty
+whether different and future CPUs will run yespower faster using one or
+maybe more threads per physical core.
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/README b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/README
new file mode 100644
index 0000000..59e565d
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/README
@@ -0,0 +1,192 @@
+	What is yespower?
+
+yespower is a proof-of-work (PoW) focused fork of yescrypt.  While
+yescrypt is a password-based key derivation function (KDF) and password
+hashing scheme, and thus is meant for processing passwords, yespower is
+meant for processing trial inputs such as block headers (including
+nonces) in PoW-based blockchains.
+
+On its own, yespower isn't a complete proof-of-work system.  Rather, in
+the blockchain use case, yespower's return value is meant to be checked
+for being numerically no greater than the blockchain's current target
+(which is related to mining difficulty) or else the proof attempt
+(yespower invocation) is to be repeated (with a different nonce) until
+the condition is finally met (allowing a new block to be mined).  This
+process isn't specific to yespower and isn't part of yespower itself
+(rather, it is similar in many PoW-based blockchains and is to be
+defined and implemented externally to yespower) and thus isn't described
+in here any further.
+
+
+	Why or why not yespower?
+
+Different proof-of-work schemes in existence vary in many aspects,
+including in friendliness to different types of hardware.  There's
+demand for all sorts of hardware (un)friendliness in those - for
+different use cases and by different communities.
+
+yespower in particular is designed to be CPU-friendly, GPU-unfriendly,
+and FPGA/ASIC-neutral.  In other words, it's meant to be relatively
+efficient to compute on current CPUs and relatively inefficient on
+current GPUs.  Unfortunately, being GPU-unfriendly also means that
+eventual FPGA and ASIC implementations will only compete with CPUs, and
+at least ASICs will win over the CPUs (FPGAs might not because of this
+market's peculiarities - large FPGAs are even more "over-priced" than
+large CPUs are), albeit by far not to the extent they did e.g. for
+Bitcoin and Litecoin.
+
+There's a lot of talk about "ASIC resistance".  What is (or should be)
+meant by that is limiting the advantage of specialized ASICs.  While
+limiting the advantage at KDF to e.g. 10x and at password hashing to
+e.g. 100x (talking orders of magnitude here, in whatever terms) may be
+considered "ASIC resistant" (as compared to e.g. 100,000x we'd have
+without trying), similar improvement factors are practically not "ASIC
+resistant" for cryptocurrency mining where they can make all the
+difference between CPU mining being profitable and not.  There might
+also exist in-between PoW use cases where moderate ASIC advantage is OK,
+such as with non-cryptocurrency and/or private/permissioned blockchains.
+
+Thus, current yespower may be considered either a short-term choice
+(valid until one of its uses provides sufficient perceived incentive to
+likely result in specialized ASICs) or a deliberate choice of a pro-CPU,
+anti-GPU, moderately-pro-ASIC PoW scheme.  It is also possible to
+respond to known improvements in future GPUs/implementations and/or to
+ASICs with new versions of yespower that users would need to switch to.
+
+
+	yespower versions.
+
+yespower includes optimized and specialized re-implementation of the
+obsolete yescrypt 0.5 (based off its first submission to Password
+Hashing Competition back in 2014) now re-released as yespower 0.5, and
+brand new proof-of-work specific variation known as yespower 1.0.
+
+yespower 0.5 is intended as a compatible upgrade for cryptocurrencies
+that already use yescrypt 0.5 (providing a few percent speedup), and
+yespower 1.0 may be used as a further upgrade or a new choice of PoW by
+those and other cryptocurrencies and other projects.
+
+There are many significant differences between yespower 0.5 and 1.0
+under the hood, but the main user visible difference is yespower 1.0
+greatly improving on GPU-unfriendliness in light of improvements seen in
+modern GPUs (up to and including NVIDIA Volta) and GPU implementations
+of yescrypt 0.5.  This is achieved mostly through greater use of CPUs'
+L2 cache.
+
+The version of algorithm to use is requested through parameters,
+allowing for both algorithms to co-exist in client and miner
+implementations (such as in preparation for a cryptocurrency hard-fork
+and/or supporting multiple cryptocurrencies in one program).
+
+
+	Parameter selection.
+
+For new uses of yespower, set the requested version to the highest
+supported, and set N*r to the highest you can reasonably afford in terms
+of proof verification time (which might in turn be determined by desired
+share rate per mining pool server), using one of the following options:
+
+1 MiB: N = 1024, r = 8
+2 MiB: N = 2048, r = 8
+4 MiB: N = 1024, r = 32
+8 MiB: N = 2048, r = 32
+16 MiB: N = 4096, r = 32
+
+and so on for higher N keeping r=32.
+
+You may also set the personalization string to your liking, but that is
+not required (you can set its pointer to NULL and its length to 0).  Its
+support is provided mostly for compatibility with existing modifications
+of yescrypt 0.5.
+
+
+	Performance.
+
+Please refer to PERFORMANCE for some benchmarks and performance tuning.
+
+
+	How to test yespower for proper operation.
+
+On a Unix-like system, invoke "make check".  This will build and run a
+program called "tests", and check its output against the supplied file
+TESTS-OK.  If everything matches, the final line of output should be the
+word "PASSED".
+
+We do most of our testing on Linux systems with gcc.  The supplied
+Makefile assumes that you use gcc.
+
+
+	Alternate code versions and make targets.
+
+Two implementations of yespower are included: reference and optimized.
+By default, the optimized implementation is built.  Internally, the
+optimized implementation uses conditional compilation to choose between
+usage of various SIMD instruction sets where supported and scalar code.
+
+The reference implementation is unoptimized and is very slow, but it has
+simpler and shorter source code.  Its purpose is to provide a simple
+human- and machine-readable specification that implementations intended
+for actual use should be tested against.  It is deliberately mostly not
+optimized, and it is not meant to be used in production.
+
+Similarly to "make check", there's "make check-ref" to build and test
+the reference implementation.  There's also "make ref" to build the
+reference implementation and have the "benchmark" program use it.
+
+"make clean" may need to be run between making different builds.
+
+
+	How to integrate yespower in a program.
+
+Although yespower.h provides several functions, chances are that you
+will only need to use yespower_tls().  Please see the comment on this
+function in yespower.h and its example usage in tests.c and benchmark.c,
+including parameter sets requesting yescrypt 0.5 as used by certain
+existing cryptocurrencies.
+
+To integrate yespower in an altcoin based on Bitcoin Core, you might
+invoke yespower_tls() from either a maybe-new (depending on where you
+fork from) CBlockHeader::GetPoWHash() (and invoke that where PoW is
+needed like e.g. Litecoin does for scrypt) or CBlockHeader::GetHash()
+(and implement caching for its return value like e.g. YACoin does for
+scrypt).  Further detail on this (generating new genesis blocks, etc.)
+is not yespower-specific and thus is not provided here.  Just like (and
+even more so than) yespower itself, this guidance is provided as-is and
+without guarantee of being correct and safe to follow.  You're supposed
+to know what you're doing.
+
+
+	Credits.
+
+scrypt has been designed by Colin Percival.  yescrypt and yespower have
+been designed by Solar Designer building upon scrypt.
+
+The following other people and projects have also indirectly helped make
+yespower what it is:
+
+ - Bill Cox
+ - Rich Felker
+ - Anthony Ferrara
+ - Christian Forler
+ - Taylor Hornby
+ - Dmitry Khovratovich
+ - Samuel Neves
+ - Marcos Simplicio
+ - Ken T Takusagawa
+ - Jakob Wenzel
+ - Christian Winnerlein
+
+ - DARPA Cyber Fast Track
+ - Password Hashing Competition
+
+
+	Contact info.
+
+First, please check the yespower homepage for new versions, etc.:
+
+    http://www.openwall.com/yespower/
+
+If you have anything valuable to add or a non-trivial question to ask,
+you may contact the maintainer of yespower at:
+
+    Solar Designer <solar at openwall.com>
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/TESTS-OK b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/TESTS-OK
new file mode 100644
index 0000000..5413e85
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/TESTS-OK
@@ -0,0 +1,16 @@
+yespower(5, 2048, 8, "Client Key") = a5 9f ec 4c 4f dd a1 6e 3b 14 05 ad da 66 d5 25 b6 8e 7c ad fc fe 6a c0 66 c7 ad 11 8c d8 05 90
+yespower(5, 2048, 8, BSTY) = 5e a2 b2 95 6a 9e ac e3 0a 32 37 ff 1d 44 1e de e1 dc 25 aa b8 f0 ea 15 c1 21 65 f8 3a 7b c2 65
+yespower(5, 4096, 16, "Client Key") = 92 7e 72 d0 de d3 d8 04 75 47 3f 40 f1 74 3c 67 28 9d 45 3d 52 42 d4 f5 5a f4 e3 25 e0 66 99 c5
+yespower(5, 4096, 24, "Jagaricoin") = 0e 13 66 97 32 11 e7 fe a8 ad 9d 81 98 9c 84 a2 54 d9 68 c9 d3 33 dd 8f f0 99 32 4f 38 61 1e 04
+yespower(5, 4096, 32, "WaviBanana") = 3a e0 5a bb 3c 5c f6 f7 54 15 a9 25 54 c9 8d 50 e3 8e c9 55 2c fa 78 37 36 16 f4 80 b2 4e 55 9f
+yespower(5, 2048, 32, "Client Key") = 56 0a 89 1b 5c a2 e1 c6 36 11 1a 9f f7 c8 94 a5 d0 a2 60 2f 43 fd cf a5 94 9b 95 e2 2f e4 46 1e
+yespower(5, 1024, 32, "Client Key") = 2a 79 e5 3d 1b e6 66 9b c5 56 cc c4 17 bc e3 d2 2a 74 a2 32 f5 6b 8e 1d 39 b4 57 92 67 5d e1 08
+yespower(5, 2048, 8, NULL) = 5e cb d8 e8 d7 c9 0b ae d4 bb f8 91 6a 12 25 dc c3 c6 5f 5c 91 65 ba e8 1c dd e3 cf fa d1 28 e8
+yespower(10, 2048, 8, NULL) = 69 e0 e8 95 b3 df 7a ee b8 37 d7 1f e1 99 e9 d3 4f 7e c4 6e cb ca 7a 2c 43 08 e5 18 57 ae 9b 46
+yespower(10, 4096, 16, NULL) = 33 fb 8f 06 38 24 a4 a0 20 f6 3d ca 53 5f 5c a6 6a b5 57 64 68 c7 5d 1c ca ac 75 42 f7 64 95 ac
+yespower(10, 4096, 32, NULL) = 77 1a ee fd a8 fe 79 a0 82 5b c7 f2 ae e1 62 ab 55 78 57 46 39 ff c6 ca 37 23 cc 18 e5 e3 e2 85
+yespower(10, 2048, 32, NULL) = d5 ef b8 13 cd 26 3e 9b 34 54 01 30 23 3c bb c6 a9 21 fb ff 34 31 e5 ec 1a 1a bd e2 ae a6 ff 4d
+yespower(10, 1024, 32, NULL) = 50 1b 79 2d b4 2e 38 8f 6e 7d 45 3c 95 d0 3a 12 a3 60 16 a5 15 4a 68 83 90 dd c6 09 a4 0c 67 99
+yespower(10, 1024, 32, "personality test") = 1f 02 69 ac f5 65 c4 9a dc 0e f9 b8 f2 6a b3 80 8c dc 38 39 4a 25 4f dd ee dc c3 aa cf f6 ad 9d
+XOR of yespower(5, ...) = 44 6c 17 2d fd 2b 39 64 d9 34 08 a3 bf 55 9e c7 42 31 7a 2c f2 b2 48 7c ec 41 17 e8 3d 7d 96 b0
+XOR of yespower(10, ...) = a9 fe c9 66 53 bd 25 ea 18 24 f3 f9 cc 96 c5 98 5f 19 1d 1d ca ee d3 e9 d0 52 f8 ac 76 bc bb da
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/benchmark.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/benchmark.c
new file mode 100644
index 0000000..4842dfa
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/benchmark.c
@@ -0,0 +1,262 @@
+/*-
+ * Copyright 2013-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <stdio.h>
+#include <stdlib.h> /* for atoi() */
+#include <string.h>
+#include <unistd.h>
+#include <time.h>
+#include <sys/times.h>
+#include <sched.h>
+
+#include "yespower.h"
+
+#ifdef _OPENMP
+#include <omp.h>
+
+#define NSAVE 1000
+
+static uint64_t time_us(void)
+{
+	struct timespec t;
+#ifdef CLOCK_MONOTONIC_RAW
+	if (clock_gettime(CLOCK_MONOTONIC_RAW, &t))
+		return 0;
+#else
+	if (clock_gettime(CLOCK_MONOTONIC, &t))
+		return 0;
+#endif
+	return 1 + (uint64_t)t.tv_sec * 1000000 + t.tv_nsec / 1000;
+}
+#endif
+
+int main(int argc, const char * const *argv)
+{
+	yespower_params_t params = {
+		.version = YESPOWER_0_5,
+		.N = 2048,
+		.r = 8,
+		.pers = (const uint8_t *)"Client Key",
+		.perslen = 10
+	};
+
+	if (argc > 1)
+		params.version = atoi(argv[1]);
+	if (argc > 2)
+		params.N = atoi(argv[2]);
+	if (argc > 3)
+		params.r = atoi(argv[3]);
+
+	printf("version=%.1f N=%u r=%u\n",
+	    params.version * 0.1, params.N, params.r);
+
+	printf("Will use %.2f KiB RAM\n", 0.125 * params.N * params.r);
+
+	static __thread union {
+		uint8_t u8[80];
+		uint32_t u32[20];
+	} src;
+	yespower_binary_t dst;
+	unsigned int i;
+
+	for (i = 0; i < sizeof(src); i++)
+		src.u8[i] = i * 3;
+
+	if (yespower_tls(src.u8, sizeof(src), &params, &dst)) {
+		puts("FAILED");
+		return 1;
+	}
+
+	for (i = 0; i < sizeof(dst); i++)
+		printf("%02x%c", dst.uc[i], i < sizeof(dst) - 1 ? ' ' : '\n');
+
+	puts("Benchmarking 1 thread ...");
+
+	clock_t clk_tck = sysconf(_SC_CLK_TCK);
+	struct tms start_tms, end_tms;
+	clock_t start = times(&start_tms), end;
+	unsigned int n;
+	unsigned long long count;
+#ifdef _OPENMP
+	yespower_binary_t save[NSAVE];
+	unsigned int nsave = 0;
+#endif
+	uint32_t seed = start * 1812433253U;
+
+	n = 1;
+	count = 0;
+	do {
+		for (i = 0; i < n; i++) {
+			yespower_binary_t *p = &dst;
+#ifdef _OPENMP
+			if (nsave < NSAVE)
+				p = &save[nsave++];
+#endif
+			src.u32[19] = seed + (count + i);
+			if (yespower_tls(src.u8, sizeof(src), &params, p)) {
+				puts("FAILED");
+				return 1;
+			}
+		}
+		count += n;
+
+		end = times(&end_tms);
+		n <<= 1;
+	} while (end - start < clk_tck * 2);
+
+	clock_t start_v = start_tms.tms_utime + start_tms.tms_stime +
+	    start_tms.tms_cutime + start_tms.tms_cstime;
+	clock_t end_v = end_tms.tms_utime + end_tms.tms_stime +
+	    end_tms.tms_cutime + end_tms.tms_cstime;
+
+	printf("%llu H/s real, %llu H/s virtual "
+	    "(%llu hashes in %.2f seconds)\n",
+	    count * clk_tck / (end - start),
+	    count * clk_tck / (end_v - start_v),
+	    count, (double)(end - start) / clk_tck);
+
+	for (i = 0; i < nsave; i++) {
+		unsigned int j;
+		for (j = i + 1; j < nsave; j++) {
+			unsigned int k = 8;
+			if (!memcmp(&save[i], &save[j], k)) {
+				printf("%u-byte collision(s) detected\n", k);
+				i = nsave; break;
+			}
+		}
+	}
+
+#ifdef _OPENMP
+	unsigned int nt = omp_get_max_threads();
+
+	printf("Benchmarking %u thread%s ...\n",
+	    nt, nt == 1 ? "" : "s");
+
+	typedef struct {
+		uint64_t min, max, total;
+	} thread_data_s;
+	union {
+		thread_data_s s;
+		uint8_t cachelines[2][64]; /* avoid false sharing */
+	} thread_data[nt]; /* tricky to align this when on stack */
+
+	unsigned int t;
+	for (t = 0; t < nt; t++) {
+		thread_data_s *td = &thread_data[t].s;
+		td->min = ~(uint64_t)0; td->max = 0; td->total = 0;
+	}
+
+	unsigned long long count1 = count, count_restart = 0;
+
+	if (!geteuid()) {
+		puts("Running as root, so trying to set SCHED_RR");
+#pragma omp parallel
+		{
+			struct sched_param param = { .sched_priority = 1 };
+			if (sched_setscheduler(getpid(), SCHED_RR, &param))
+				perror("sched_setscheduler");
+		}
+	}
+
+	start = times(&start_tms);
+
+	n = count * omp_get_max_threads();
+	count = 0;
+	do {
+#pragma omp parallel for default(none) copyin(src) private(i, dst) shared(n, thread_data, params, seed, count, save, nsave)
+		for (i = 0; i < n; i++) {
+			unsigned int j = count + i;
+
+			src.u32[19] = seed + j;
+
+			uint64_t start1 = time_us();
+
+			if (yespower_tls(src.u8, sizeof(src), &params, &dst)) {
+#pragma omp critical
+				puts("FAILED");
+			}
+
+			uint64_t end1 = time_us();
+			if (end1 < start1)
+				end1 = start1;
+			uint64_t diff1 = end1 - start1;
+
+			thread_data_s *td = &thread_data[omp_get_thread_num()].s;
+			td->total += diff1;
+			if (diff1 < td->min)
+				td->min = diff1;
+			if (diff1 > td->max)
+				td->max = diff1;
+
+#ifdef _OPENMP
+			if (j < nsave && memcmp(&save[j], &dst, sizeof(dst))) {
+#pragma omp critical
+				printf("Mismatch at %u\n", j);
+			}
+#endif
+		}
+
+		count += n;
+		if ((count - n) < count1 && count >= count1) {
+/* Disregard our repeat of single thread's results (could be partially cached
+ * by same core, but OTOH other cores not yet warmed up to full clock rate). */
+			start = times(&start_tms);
+			count_restart = count;
+			for (t = 0; t < nt; t++) {
+				thread_data_s *td = &thread_data[t].s;
+				td->min = ~(uint64_t)0; td->max = 0; td->total = 0;
+			}
+		} else {
+			n <<= 1;
+		}
+
+		end = times(&end_tms);
+	} while (end - start < clk_tck);
+
+	if (!count_restart)
+		puts("Didn't reach single-thread's hash count");
+	count -= count_restart;
+
+	start_v = start_tms.tms_utime + start_tms.tms_stime +
+	    start_tms.tms_cutime + start_tms.tms_cstime;
+	end_v = end_tms.tms_utime + end_tms.tms_stime +
+	    end_tms.tms_cutime + end_tms.tms_cstime;
+
+	printf("%llu H/s real, %llu H/s virtual "
+	    "(%llu hashes in %.2f seconds)\n",
+	    count * clk_tck / (end - start),
+	    count * clk_tck / (end_v - start_v),
+	    count, (double)(end - start) / clk_tck);
+
+	uint64_t min = ~(uint64_t)0, max = 0, total = 0;
+	for (t = 0; t < nt; t++) {
+		thread_data_s *td = &thread_data[t].s;
+		total += td->total;
+		if (td->min < min)
+			min = td->min;
+		if (td->max > max)
+			max = td->max;
+	}
+	printf("min %.3f ms, avg %.3f ms, max %.3f ms\n",
+		min / 1000.0, total / 1000.0 / count, max / 1000.0);
+#endif
+
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/insecure_memzero.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/insecure_memzero.h
new file mode 100644
index 0000000..5a0ba75
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/insecure_memzero.h
@@ -0,0 +1 @@
+#define insecure_memzero(buf, len) /* empty */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sha256.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sha256.c
new file mode 100644
index 0000000..ee2d5d2
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sha256.c
@@ -0,0 +1,646 @@
+/*-
+ * Copyright 2005-2016 Colin Percival
+ * Copyright 2016-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <assert.h>
+#include <stdint.h>
+#include <string.h>
+
+#include "insecure_memzero.h"
+#include "sysendian.h"
+
+#include "sha256.h"
+
+#ifdef __ICC
+/* Miscompile with icc 14.0.0 (at least), so don't use restrict there */
+#define restrict
+#elif __STDC_VERSION__ >= 199901L
+/* Have restrict */
+#elif defined(__GNUC__)
+#define restrict __restrict
+#else
+#define restrict
+#endif
+
+/*
+ * Encode a length len*2 vector of (uint32_t) into a length len*8 vector of
+ * (uint8_t) in big-endian form.
+ */
+static void
+be32enc_vect(uint8_t * dst, const uint32_t * src, size_t len)
+{
+
+	/* Encode vector, two words at a time. */
+	do {
+		be32enc(&dst[0], src[0]);
+		be32enc(&dst[4], src[1]);
+		src += 2;
+		dst += 8;
+	} while (--len);
+}
+
+/*
+ * Decode a big-endian length len*8 vector of (uint8_t) into a length
+ * len*2 vector of (uint32_t).
+ */
+static void
+be32dec_vect(uint32_t * dst, const uint8_t * src, size_t len)
+{
+
+	/* Decode vector, two words at a time. */
+	do {
+		dst[0] = be32dec(&src[0]);
+		dst[1] = be32dec(&src[4]);
+		src += 8;
+		dst += 2;
+	} while (--len);
+}
+
+/* SHA256 round constants. */
+static const uint32_t Krnd[64] = {
+	0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5,
+	0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
+	0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3,
+	0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
+	0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
+	0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
+	0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
+	0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
+	0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13,
+	0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
+	0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3,
+	0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
+	0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5,
+	0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
+	0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
+	0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
+};
+
+/* Elementary functions used by SHA256 */
+#define Ch(x, y, z)	((x & (y ^ z)) ^ z)
+#define Maj(x, y, z)	((x & (y | z)) | (y & z))
+#define SHR(x, n)	(x >> n)
+#define ROTR(x, n)	((x >> n) | (x << (32 - n)))
+#define S0(x)		(ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
+#define S1(x)		(ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
+#define s0(x)		(ROTR(x, 7) ^ ROTR(x, 18) ^ SHR(x, 3))
+#define s1(x)		(ROTR(x, 17) ^ ROTR(x, 19) ^ SHR(x, 10))
+
+/* SHA256 round function */
+#define RND(a, b, c, d, e, f, g, h, k)			\
+	h += S1(e) + Ch(e, f, g) + k;			\
+	d += h;						\
+	h += S0(a) + Maj(a, b, c);
+
+/* Adjusted round function for rotating state */
+#define RNDr(S, W, i, ii)			\
+	RND(S[(64 - i) % 8], S[(65 - i) % 8],	\
+	    S[(66 - i) % 8], S[(67 - i) % 8],	\
+	    S[(68 - i) % 8], S[(69 - i) % 8],	\
+	    S[(70 - i) % 8], S[(71 - i) % 8],	\
+	    W[i + ii] + Krnd[i + ii])
+
+/* Message schedule computation */
+#define MSCH(W, ii, i)				\
+	W[i + ii + 16] = s1(W[i + ii + 14]) + W[i + ii + 9] + s0(W[i + ii + 1]) + W[i + ii]
+
+/*
+ * SHA256 block compression function.  The 256-bit state is transformed via
+ * the 512-bit input block to produce a new state.
+ */
+static void
+SHA256_Transform(uint32_t state[static restrict 8],
+    const uint8_t block[static restrict 64],
+    uint32_t W[static restrict 64], uint32_t S[static restrict 8])
+{
+	int i;
+
+	/* 1. Prepare the first part of the message schedule W. */
+	be32dec_vect(W, block, 8);
+
+	/* 2. Initialize working variables. */
+	memcpy(S, state, 32);
+
+	/* 3. Mix. */
+	for (i = 0; i < 64; i += 16) {
+		RNDr(S, W, 0, i);
+		RNDr(S, W, 1, i);
+		RNDr(S, W, 2, i);
+		RNDr(S, W, 3, i);
+		RNDr(S, W, 4, i);
+		RNDr(S, W, 5, i);
+		RNDr(S, W, 6, i);
+		RNDr(S, W, 7, i);
+		RNDr(S, W, 8, i);
+		RNDr(S, W, 9, i);
+		RNDr(S, W, 10, i);
+		RNDr(S, W, 11, i);
+		RNDr(S, W, 12, i);
+		RNDr(S, W, 13, i);
+		RNDr(S, W, 14, i);
+		RNDr(S, W, 15, i);
+
+		if (i == 48)
+			break;
+		MSCH(W, 0, i);
+		MSCH(W, 1, i);
+		MSCH(W, 2, i);
+		MSCH(W, 3, i);
+		MSCH(W, 4, i);
+		MSCH(W, 5, i);
+		MSCH(W, 6, i);
+		MSCH(W, 7, i);
+		MSCH(W, 8, i);
+		MSCH(W, 9, i);
+		MSCH(W, 10, i);
+		MSCH(W, 11, i);
+		MSCH(W, 12, i);
+		MSCH(W, 13, i);
+		MSCH(W, 14, i);
+		MSCH(W, 15, i);
+	}
+
+	/* 4. Mix local working variables into global state. */
+	state[0] += S[0];
+	state[1] += S[1];
+	state[2] += S[2];
+	state[3] += S[3];
+	state[4] += S[4];
+	state[5] += S[5];
+	state[6] += S[6];
+	state[7] += S[7];
+}
+
+static const uint8_t PAD[64] = {
+	0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+
+/* Add padding and terminating bit-count. */
+static void
+SHA256_Pad(SHA256_CTX * ctx, uint32_t tmp32[static restrict 72])
+{
+	size_t r;
+
+	/* Figure out how many bytes we have buffered. */
+	r = (ctx->count >> 3) & 0x3f;
+
+	/* Pad to 56 mod 64, transforming if we finish a block en route. */
+	if (r < 56) {
+		/* Pad to 56 mod 64. */
+		memcpy(&ctx->buf[r], PAD, 56 - r);
+	} else {
+		/* Finish the current block and mix. */
+		memcpy(&ctx->buf[r], PAD, 64 - r);
+		SHA256_Transform(ctx->state, ctx->buf, &tmp32[0], &tmp32[64]);
+
+		/* The start of the final block is all zeroes. */
+		memset(&ctx->buf[0], 0, 56);
+	}
+
+	/* Add the terminating bit-count. */
+	be64enc(&ctx->buf[56], ctx->count);
+
+	/* Mix in the final block. */
+	SHA256_Transform(ctx->state, ctx->buf, &tmp32[0], &tmp32[64]);
+}
+
+/* Magic initialization constants. */
+static const uint32_t initial_state[8] = {
+	0x6A09E667, 0xBB67AE85, 0x3C6EF372, 0xA54FF53A,
+	0x510E527F, 0x9B05688C, 0x1F83D9AB, 0x5BE0CD19
+};
+
+/**
+ * SHA256_Init(ctx):
+ * Initialize the SHA256 context ${ctx}.
+ */
+void
+SHA256_Init(SHA256_CTX * ctx)
+{
+
+	/* Zero bits processed so far. */
+	ctx->count = 0;
+
+	/* Initialize state. */
+	memcpy(ctx->state, initial_state, sizeof(initial_state));
+}
+
+/**
+ * SHA256_Update(ctx, in, len):
+ * Input ${len} bytes from ${in} into the SHA256 context ${ctx}.
+ */
+static void
+_SHA256_Update(SHA256_CTX * ctx, const void * in, size_t len,
+    uint32_t tmp32[static restrict 72])
+{
+	uint32_t r;
+	const uint8_t * src = in;
+
+	/* Return immediately if we have nothing to do. */
+	if (len == 0)
+		return;
+
+	/* Number of bytes left in the buffer from previous updates. */
+	r = (ctx->count >> 3) & 0x3f;
+
+	/* Update number of bits. */
+	ctx->count += (uint64_t)(len) << 3;
+
+	/* Handle the case where we don't need to perform any transforms. */
+	if (len < 64 - r) {
+		memcpy(&ctx->buf[r], src, len);
+		return;
+	}
+
+	/* Finish the current block. */
+	memcpy(&ctx->buf[r], src, 64 - r);
+	SHA256_Transform(ctx->state, ctx->buf, &tmp32[0], &tmp32[64]);
+	src += 64 - r;
+	len -= 64 - r;
+
+	/* Perform complete blocks. */
+	while (len >= 64) {
+		SHA256_Transform(ctx->state, src, &tmp32[0], &tmp32[64]);
+		src += 64;
+		len -= 64;
+	}
+
+	/* Copy left over data into buffer. */
+	memcpy(ctx->buf, src, len);
+}
+
+/* Wrapper function for intermediate-values sanitization. */
+void
+SHA256_Update(SHA256_CTX * ctx, const void * in, size_t len)
+{
+	uint32_t tmp32[72];
+
+	/* Call the real function. */
+	_SHA256_Update(ctx, in, len, tmp32);
+
+	/* Clean the stack. */
+	insecure_memzero(tmp32, 288);
+}
+
+/**
+ * SHA256_Final(digest, ctx):
+ * Output the SHA256 hash of the data input to the context ${ctx} into the
+ * buffer ${digest}.
+ */
+static void
+_SHA256_Final(uint8_t digest[32], SHA256_CTX * ctx,
+    uint32_t tmp32[static restrict 72])
+{
+
+	/* Add padding. */
+	SHA256_Pad(ctx, tmp32);
+
+	/* Write the hash. */
+	be32enc_vect(digest, ctx->state, 4);
+}
+
+/* Wrapper function for intermediate-values sanitization. */
+void
+SHA256_Final(uint8_t digest[32], SHA256_CTX * ctx)
+{
+	uint32_t tmp32[72];
+
+	/* Call the real function. */
+	_SHA256_Final(digest, ctx, tmp32);
+
+	/* Clear the context state. */
+	insecure_memzero(ctx, sizeof(SHA256_CTX));
+
+	/* Clean the stack. */
+	insecure_memzero(tmp32, 288);
+}
+
+/**
+ * SHA256_Buf(in, len, digest):
+ * Compute the SHA256 hash of ${len} bytes from ${in} and write it to ${digest}.
+ */
+void
+SHA256_Buf(const void * in, size_t len, uint8_t digest[32])
+{
+	SHA256_CTX ctx;
+	uint32_t tmp32[72];
+
+	SHA256_Init(&ctx);
+	_SHA256_Update(&ctx, in, len, tmp32);
+	_SHA256_Final(digest, &ctx, tmp32);
+
+	/* Clean the stack. */
+	insecure_memzero(&ctx, sizeof(SHA256_CTX));
+	insecure_memzero(tmp32, 288);
+}
+
+/**
+ * HMAC_SHA256_Init(ctx, K, Klen):
+ * Initialize the HMAC-SHA256 context ${ctx} with ${Klen} bytes of key from
+ * ${K}.
+ */
+static void
+_HMAC_SHA256_Init(HMAC_SHA256_CTX * ctx, const void * _K, size_t Klen,
+    uint32_t tmp32[static restrict 72], uint8_t pad[static restrict 64],
+    uint8_t khash[static restrict 32])
+{
+	const uint8_t * K = _K;
+	size_t i;
+
+	/* If Klen > 64, the key is really SHA256(K). */
+	if (Klen > 64) {
+		SHA256_Init(&ctx->ictx);
+		_SHA256_Update(&ctx->ictx, K, Klen, tmp32);
+		_SHA256_Final(khash, &ctx->ictx, tmp32);
+		K = khash;
+		Klen = 32;
+	}
+
+	/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
+	SHA256_Init(&ctx->ictx);
+	memset(pad, 0x36, 64);
+	for (i = 0; i < Klen; i++)
+		pad[i] ^= K[i];
+	_SHA256_Update(&ctx->ictx, pad, 64, tmp32);
+
+	/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
+	SHA256_Init(&ctx->octx);
+	memset(pad, 0x5c, 64);
+	for (i = 0; i < Klen; i++)
+		pad[i] ^= K[i];
+	_SHA256_Update(&ctx->octx, pad, 64, tmp32);
+}
+
+/* Wrapper function for intermediate-values sanitization. */
+void
+HMAC_SHA256_Init(HMAC_SHA256_CTX * ctx, const void * _K, size_t Klen)
+{
+	uint32_t tmp32[72];
+	uint8_t pad[64];
+	uint8_t khash[32];
+
+	/* Call the real function. */
+	_HMAC_SHA256_Init(ctx, _K, Klen, tmp32, pad, khash);
+
+	/* Clean the stack. */
+	insecure_memzero(tmp32, 288);
+	insecure_memzero(khash, 32);
+	insecure_memzero(pad, 64);
+}
+
+/**
+ * HMAC_SHA256_Update(ctx, in, len):
+ * Input ${len} bytes from ${in} into the HMAC-SHA256 context ${ctx}.
+ */
+static void
+_HMAC_SHA256_Update(HMAC_SHA256_CTX * ctx, const void * in, size_t len,
+    uint32_t tmp32[static restrict 72])
+{
+
+	/* Feed data to the inner SHA256 operation. */
+	_SHA256_Update(&ctx->ictx, in, len, tmp32);
+}
+
+/* Wrapper function for intermediate-values sanitization. */
+void
+HMAC_SHA256_Update(HMAC_SHA256_CTX * ctx, const void * in, size_t len)
+{
+	uint32_t tmp32[72];
+
+	/* Call the real function. */
+	_HMAC_SHA256_Update(ctx, in, len, tmp32);
+
+	/* Clean the stack. */
+	insecure_memzero(tmp32, 288);
+}
+
+/**
+ * HMAC_SHA256_Final(digest, ctx):
+ * Output the HMAC-SHA256 of the data input to the context ${ctx} into the
+ * buffer ${digest}.
+ */
+static void
+_HMAC_SHA256_Final(uint8_t digest[32], HMAC_SHA256_CTX * ctx,
+    uint32_t tmp32[static restrict 72], uint8_t ihash[static restrict 32])
+{
+
+	/* Finish the inner SHA256 operation. */
+	_SHA256_Final(ihash, &ctx->ictx, tmp32);
+
+	/* Feed the inner hash to the outer SHA256 operation. */
+	_SHA256_Update(&ctx->octx, ihash, 32, tmp32);
+
+	/* Finish the outer SHA256 operation. */
+	_SHA256_Final(digest, &ctx->octx, tmp32);
+}
+
+/* Wrapper function for intermediate-values sanitization. */
+void
+HMAC_SHA256_Final(uint8_t digest[32], HMAC_SHA256_CTX * ctx)
+{
+	uint32_t tmp32[72];
+	uint8_t ihash[32];
+
+	/* Call the real function. */
+	_HMAC_SHA256_Final(digest, ctx, tmp32, ihash);
+
+	/* Clean the stack. */
+	insecure_memzero(tmp32, 288);
+	insecure_memzero(ihash, 32);
+}
+
+/**
+ * HMAC_SHA256_Buf(K, Klen, in, len, digest):
+ * Compute the HMAC-SHA256 of ${len} bytes from ${in} using the key ${K} of
+ * length ${Klen}, and write the result to ${digest}.
+ */
+void
+HMAC_SHA256_Buf(const void * K, size_t Klen, const void * in, size_t len,
+    uint8_t digest[32])
+{
+	HMAC_SHA256_CTX ctx;
+	uint32_t tmp32[72];
+	uint8_t tmp8[96];
+
+	_HMAC_SHA256_Init(&ctx, K, Klen, tmp32, &tmp8[0], &tmp8[64]);
+	_HMAC_SHA256_Update(&ctx, in, len, tmp32);
+	_HMAC_SHA256_Final(digest, &ctx, tmp32, &tmp8[0]);
+
+	/* Clean the stack. */
+	insecure_memzero(&ctx, sizeof(HMAC_SHA256_CTX));
+	insecure_memzero(tmp32, 288);
+	insecure_memzero(tmp8, 96);
+}
+
+/* Add padding and terminating bit-count, but don't invoke Transform yet. */
+static int
+SHA256_Pad_Almost(SHA256_CTX * ctx, uint8_t len[static restrict 8],
+    uint32_t tmp32[static restrict 72])
+{
+	uint32_t r;
+
+	r = (ctx->count >> 3) & 0x3f;
+	if (r >= 56)
+		return -1;
+
+	/*
+	 * Convert length to a vector of bytes -- we do this now rather
+	 * than later because the length will change after we pad.
+	 */
+	be64enc(len, ctx->count);
+
+	/* Add 1--56 bytes so that the resulting length is 56 mod 64. */
+	_SHA256_Update(ctx, PAD, 56 - r, tmp32);
+
+	/* Add the terminating bit-count. */
+	ctx->buf[63] = len[7];
+	_SHA256_Update(ctx, len, 7, tmp32);
+
+	return 0;
+}
+
+/**
+ * PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
+ * Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
+ * write the output to buf.  The value dkLen must be at most 32 * (2^32 - 1).
+ */
+void
+PBKDF2_SHA256(const uint8_t * passwd, size_t passwdlen, const uint8_t * salt,
+    size_t saltlen, uint64_t c, uint8_t * buf, size_t dkLen)
+{
+	HMAC_SHA256_CTX Phctx, PShctx, hctx;
+	uint32_t tmp32[72];
+	union {
+		uint8_t tmp8[96];
+		uint32_t state[8];
+	} u;
+	size_t i;
+	uint8_t ivec[4];
+	uint8_t U[32];
+	uint8_t T[32];
+	uint64_t j;
+	int k;
+	size_t clen;
+
+	/* Sanity-check. */
+	assert(dkLen <= 32 * (size_t)(UINT32_MAX));
+
+	if (c == 1 && (dkLen & 31) == 0 && (saltlen & 63) <= 51) {
+		uint32_t oldcount;
+		uint8_t * ivecp;
+
+		/* Compute HMAC state after processing P and S. */
+		_HMAC_SHA256_Init(&hctx, passwd, passwdlen,
+		    tmp32, &u.tmp8[0], &u.tmp8[64]);
+		_HMAC_SHA256_Update(&hctx, salt, saltlen, tmp32);
+
+		/* Prepare ictx padding. */
+		oldcount = hctx.ictx.count & (0x3f << 3);
+		_HMAC_SHA256_Update(&hctx, "\0\0\0", 4, tmp32);
+		if ((hctx.ictx.count & (0x3f << 3)) < oldcount ||
+		    SHA256_Pad_Almost(&hctx.ictx, u.tmp8, tmp32))
+			goto generic; /* Can't happen due to saltlen check */
+		ivecp = hctx.ictx.buf + (oldcount >> 3);
+
+		/* Prepare octx padding. */
+		hctx.octx.count += 32 << 3;
+		SHA256_Pad_Almost(&hctx.octx, u.tmp8, tmp32);
+
+		/* Iterate through the blocks. */
+		for (i = 0; i * 32 < dkLen; i++) {
+			/* Generate INT(i + 1). */
+			be32enc(ivecp, (uint32_t)(i + 1));
+
+			/* Compute U_1 = PRF(P, S || INT(i)). */
+			memcpy(u.state, hctx.ictx.state, sizeof(u.state));
+			SHA256_Transform(u.state, hctx.ictx.buf,
+			    &tmp32[0], &tmp32[64]);
+			be32enc_vect(hctx.octx.buf, u.state, 4);
+			memcpy(u.state, hctx.octx.state, sizeof(u.state));
+			SHA256_Transform(u.state, hctx.octx.buf,
+			    &tmp32[0], &tmp32[64]);
+			be32enc_vect(&buf[i * 32], u.state, 4);
+		}
+
+		goto cleanup;
+	}
+
+generic:
+	/* Compute HMAC state after processing P. */
+	_HMAC_SHA256_Init(&Phctx, passwd, passwdlen,
+	    tmp32, &u.tmp8[0], &u.tmp8[64]);
+
+	/* Compute HMAC state after processing P and S. */
+	memcpy(&PShctx, &Phctx, sizeof(HMAC_SHA256_CTX));
+	_HMAC_SHA256_Update(&PShctx, salt, saltlen, tmp32);
+
+	/* Iterate through the blocks. */
+	for (i = 0; i * 32 < dkLen; i++) {
+		/* Generate INT(i + 1). */
+		be32enc(ivec, (uint32_t)(i + 1));
+
+		/* Compute U_1 = PRF(P, S || INT(i)). */
+		memcpy(&hctx, &PShctx, sizeof(HMAC_SHA256_CTX));
+		_HMAC_SHA256_Update(&hctx, ivec, 4, tmp32);
+		_HMAC_SHA256_Final(T, &hctx, tmp32, u.tmp8);
+
+		if (c > 1) {
+			/* T_i = U_1 ... */
+			memcpy(U, T, 32);
+
+			for (j = 2; j <= c; j++) {
+				/* Compute U_j. */
+				memcpy(&hctx, &Phctx, sizeof(HMAC_SHA256_CTX));
+				_HMAC_SHA256_Update(&hctx, U, 32, tmp32);
+				_HMAC_SHA256_Final(U, &hctx, tmp32, u.tmp8);
+
+				/* ... xor U_j ... */
+				for (k = 0; k < 32; k++)
+					T[k] ^= U[k];
+			}
+		}
+
+		/* Copy as many bytes as necessary into buf. */
+		clen = dkLen - i * 32;
+		if (clen > 32)
+			clen = 32;
+		memcpy(&buf[i * 32], T, clen);
+	}
+
+	/* Clean the stack. */
+	insecure_memzero(&Phctx, sizeof(HMAC_SHA256_CTX));
+	insecure_memzero(&PShctx, sizeof(HMAC_SHA256_CTX));
+	insecure_memzero(U, 32);
+	insecure_memzero(T, 32);
+
+cleanup:
+	insecure_memzero(&hctx, sizeof(HMAC_SHA256_CTX));
+	insecure_memzero(tmp32, 288);
+	insecure_memzero(&u, sizeof(u));
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sha256.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sha256.h
new file mode 100644
index 0000000..6210502
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sha256.h
@@ -0,0 +1,129 @@
+/*-
+ * Copyright 2005-2016 Colin Percival
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#ifndef _SHA256_H_
+#define _SHA256_H_
+
+#include <stddef.h>
+#include <stdint.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/*
+ * Use #defines in order to avoid namespace collisions with anyone else's
+ * SHA256 code (e.g., the code in OpenSSL).
+ */
+#define SHA256_Init libcperciva_SHA256_Init
+#define SHA256_Update libcperciva_SHA256_Update
+#define SHA256_Final libcperciva_SHA256_Final
+#define SHA256_Buf libcperciva_SHA256_Buf
+#define SHA256_CTX libcperciva_SHA256_CTX
+#define HMAC_SHA256_Init libcperciva_HMAC_SHA256_Init
+#define HMAC_SHA256_Update libcperciva_HMAC_SHA256_Update
+#define HMAC_SHA256_Final libcperciva_HMAC_SHA256_Final
+#define HMAC_SHA256_Buf libcperciva_HMAC_SHA256_Buf
+#define HMAC_SHA256_CTX libcperciva_HMAC_SHA256_CTX
+
+/* Context structure for SHA256 operations. */
+typedef struct {
+	uint32_t state[8];
+	uint64_t count;
+	uint8_t buf[64];
+} SHA256_CTX;
+
+/**
+ * SHA256_Init(ctx):
+ * Initialize the SHA256 context ${ctx}.
+ */
+void SHA256_Init(SHA256_CTX *);
+
+/**
+ * SHA256_Update(ctx, in, len):
+ * Input ${len} bytes from ${in} into the SHA256 context ${ctx}.
+ */
+void SHA256_Update(SHA256_CTX *, const void *, size_t);
+
+/**
+ * SHA256_Final(digest, ctx):
+ * Output the SHA256 hash of the data input to the context ${ctx} into the
+ * buffer ${digest}.
+ */
+void SHA256_Final(uint8_t[32], SHA256_CTX *);
+
+/**
+ * SHA256_Buf(in, len, digest):
+ * Compute the SHA256 hash of ${len} bytes from ${in} and write it to ${digest}.
+ */
+void SHA256_Buf(const void *, size_t, uint8_t[32]);
+
+/* Context structure for HMAC-SHA256 operations. */
+typedef struct {
+	SHA256_CTX ictx;
+	SHA256_CTX octx;
+} HMAC_SHA256_CTX;
+
+/**
+ * HMAC_SHA256_Init(ctx, K, Klen):
+ * Initialize the HMAC-SHA256 context ${ctx} with ${Klen} bytes of key from
+ * ${K}.
+ */
+void HMAC_SHA256_Init(HMAC_SHA256_CTX *, const void *, size_t);
+
+/**
+ * HMAC_SHA256_Update(ctx, in, len):
+ * Input ${len} bytes from ${in} into the HMAC-SHA256 context ${ctx}.
+ */
+void HMAC_SHA256_Update(HMAC_SHA256_CTX *, const void *, size_t);
+
+/**
+ * HMAC_SHA256_Final(digest, ctx):
+ * Output the HMAC-SHA256 of the data input to the context ${ctx} into the
+ * buffer ${digest}.
+ */
+void HMAC_SHA256_Final(uint8_t[32], HMAC_SHA256_CTX *);
+
+/**
+ * HMAC_SHA256_Buf(K, Klen, in, len, digest):
+ * Compute the HMAC-SHA256 of ${len} bytes from ${in} using the key ${K} of
+ * length ${Klen}, and write the result to ${digest}.
+ */
+void HMAC_SHA256_Buf(const void *, size_t, const void *, size_t, uint8_t[32]);
+
+/**
+ * PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
+ * Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
+ * write the output to buf.  The value dkLen must be at most 32 * (2^32 - 1).
+ */
+void PBKDF2_SHA256(const uint8_t *, size_t, const uint8_t *, size_t,
+    uint64_t, uint8_t *, size_t);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* !_SHA256_H_ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sysendian.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sysendian.h
new file mode 100644
index 0000000..52c1fe7
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/sysendian.h
@@ -0,0 +1,94 @@
+/*-
+ * Copyright 2007-2014 Colin Percival
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#ifndef _SYSENDIAN_H_
+#define _SYSENDIAN_H_
+
+#include <stdint.h>
+
+/* Avoid namespace collisions with BSD <sys/endian.h>. */
+#define be32dec libcperciva_be32dec
+#define be32enc libcperciva_be32enc
+#define be64enc libcperciva_be64enc
+#define le32dec libcperciva_le32dec
+#define le32enc libcperciva_le32enc
+
+static inline uint32_t
+be32dec(const void * pp)
+{
+	const uint8_t * p = (uint8_t const *)pp;
+
+	return ((uint32_t)(p[3]) + ((uint32_t)(p[2]) << 8) +
+	    ((uint32_t)(p[1]) << 16) + ((uint32_t)(p[0]) << 24));
+}
+
+static inline void
+be32enc(void * pp, uint32_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[3] = x & 0xff;
+	p[2] = (x >> 8) & 0xff;
+	p[1] = (x >> 16) & 0xff;
+	p[0] = (x >> 24) & 0xff;
+}
+
+static inline void
+be64enc(void * pp, uint64_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[7] = x & 0xff;
+	p[6] = (x >> 8) & 0xff;
+	p[5] = (x >> 16) & 0xff;
+	p[4] = (x >> 24) & 0xff;
+	p[3] = (x >> 32) & 0xff;
+	p[2] = (x >> 40) & 0xff;
+	p[1] = (x >> 48) & 0xff;
+	p[0] = (x >> 56) & 0xff;
+}
+
+static inline uint32_t
+le32dec(const void * pp)
+{
+	const uint8_t * p = (uint8_t const *)pp;
+
+	return ((uint32_t)(p[0]) + ((uint32_t)(p[1]) << 8) +
+	    ((uint32_t)(p[2]) << 16) + ((uint32_t)(p[3]) << 24));
+}
+
+static inline void
+le32enc(void * pp, uint32_t x)
+{
+	uint8_t * p = (uint8_t *)pp;
+
+	p[0] = x & 0xff;
+	p[1] = (x >> 8) & 0xff;
+	p[2] = (x >> 16) & 0xff;
+	p[3] = (x >> 24) & 0xff;
+}
+
+#endif /* !_SYSENDIAN_H_ */
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/tests.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/tests.c
new file mode 100644
index 0000000..3c394a8
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/tests.c
@@ -0,0 +1,182 @@
+/*-
+ * Copyright 2013-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <stdio.h>
+#include <string.h>
+
+#include "yespower.h"
+
+#undef TEST_PBKDF2_SHA256
+
+#ifdef TEST_PBKDF2_SHA256
+#include <assert.h>
+
+#include "sha256.h"
+
+static void print_PBKDF2_SHA256_raw(const char *passwd, size_t passwdlen,
+    const char *salt, size_t saltlen, uint64_t c, size_t dkLen)
+{
+	uint8_t dk[64];
+	size_t i;
+
+	assert(dkLen <= sizeof(dk));
+
+	/* XXX This prints the strings truncated at first NUL */
+	printf("PBKDF2_SHA256(\"%s\", \"%s\", %llu, %llu) = ",
+	    passwd, salt, (unsigned long long)c, (unsigned long long)dkLen);
+
+	PBKDF2_SHA256((const uint8_t *) passwd, passwdlen,
+	    (const uint8_t *) salt, saltlen, c, dk, dkLen);
+
+	for (i = 0; i < dkLen; i++)
+		printf("%02x%c", dk[i], i < dkLen - 1 ? ' ' : '\n');
+}
+
+static void print_PBKDF2_SHA256(const char *passwd,
+    const char *salt, uint64_t c, size_t dkLen)
+{
+	print_PBKDF2_SHA256_raw(passwd, strlen(passwd), salt, strlen(salt), c,
+	    dkLen);
+}
+#endif
+
+static const char *pers_bsty_magic = "BSTY";
+
+static void print_yespower(yespower_version_t version, uint32_t N, uint32_t r,
+    const char *pers)
+{
+	yespower_params_t params = {
+		.version = version,
+		.N = N,
+		.r = r,
+		.pers = (const uint8_t *)pers,
+		.perslen = pers ? strlen(pers) : 0
+	};
+	uint8_t src[80];
+	yespower_binary_t dst;
+	size_t i;
+
+	const char *q = (pers && pers != pers_bsty_magic) ? "\"": "";
+	printf("yespower(%u, %u, %u, %s%s%s) = ", (unsigned int)version, N, r,
+	    q, pers ? pers : "NULL", q);
+
+	for (i = 0; i < sizeof(src); i++)
+		src[i] = i * 3;
+
+	if (pers == pers_bsty_magic) {
+		params.pers = src;
+		params.perslen = sizeof(src);
+	}
+
+	if (yespower_tls(src, sizeof(src), &params, &dst)) {
+		puts("FAILED");
+		return;
+	}
+
+	for (i = 0; i < sizeof(dst); i++)
+		printf("%02x%c", dst.uc[i], i < sizeof(dst) - 1 ? ' ' : '\n');
+}
+
+static void print_yespower_loop(yespower_version_t version, const char *pers)
+{
+	uint32_t N, r;
+	uint8_t src[80];
+	yespower_binary_t dst, xor = {{0}};
+	size_t i;
+
+	printf("XOR of yespower(%u, ...) = ", (unsigned int)version);
+
+	for (i = 0; i < sizeof(src); i++)
+		src[i] = i * 3;
+
+	for (N = 1024; N <= 4096; N <<= 1) {
+		for (r = 8; r <= 32; r++) {
+			yespower_params_t params = {
+				.version = version,
+				.N = N,
+				.r = r,
+				.pers = (const uint8_t *)pers,
+				.perslen = pers ? strlen(pers) : 0
+			};
+			if (yespower_tls(src, sizeof(src), &params, &dst)) {
+				puts("FAILED");
+				return;
+			}
+			for (i = 0; i < sizeof(xor); i++)
+				xor.uc[i] ^= dst.uc[i];
+		}
+	}
+
+	for (i = 0; i < sizeof(xor); i++)
+		printf("%02x%c", xor.uc[i], i < sizeof(xor) - 1 ? ' ' : '\n');
+}
+
+int main(void)
+{
+	setvbuf(stdout, NULL, _IOLBF, 0);
+
+#ifdef TEST_PBKDF2_SHA256
+	print_PBKDF2_SHA256("password", "salt", 1, 20);
+	print_PBKDF2_SHA256("password", "salt", 2, 20);
+	print_PBKDF2_SHA256("password", "salt", 4096, 20);
+	print_PBKDF2_SHA256("password", "salt", 16777216, 20);
+	print_PBKDF2_SHA256("passwordPASSWORDpassword",
+	    "saltSALTsaltSALTsaltSALTsaltSALTsalt", 4096, 25);
+	print_PBKDF2_SHA256_raw("pass\0word", 9, "sa\0lt", 5, 4096, 16);
+#if 0
+	print_PBKDF2_SHA256("password", "salt", 1, 32);
+	print_PBKDF2_SHA256("password", "salt", 2, 32);
+	print_PBKDF2_SHA256("password", "salt", 4096, 32);
+	print_PBKDF2_SHA256("password", "salt", 16777216, 32);
+	print_PBKDF2_SHA256("passwordPASSWORDpassword",
+	    "saltSALTsaltSALTsaltSALTsaltSALTsalt", 4096, 40);
+	print_PBKDF2_SHA256("password", "salt", 4096, 16);
+	print_PBKDF2_SHA256("password", "salt", 1, 20);
+	print_PBKDF2_SHA256("password", "salt", 2, 20);
+	print_PBKDF2_SHA256("password", "salt", 4096, 20);
+	print_PBKDF2_SHA256("password", "salt", 16777216, 20);
+	print_PBKDF2_SHA256("password", "salt", 4096, 25);
+	print_PBKDF2_SHA256("password", "salt", 4096, 16);
+#endif
+#endif
+
+	print_yespower(YESPOWER_0_5, 2048, 8, "Client Key"); /* yescrypt 0.5 */
+	print_yespower(YESPOWER_0_5, 2048, 8, pers_bsty_magic); /* BSTY */
+	print_yespower(YESPOWER_0_5, 4096, 16, "Client Key"); /* Cryply */
+	print_yespower(YESPOWER_0_5, 4096, 24, "Jagaricoin");
+	print_yespower(YESPOWER_0_5, 4096, 32, "WaviBanana");
+	print_yespower(YESPOWER_0_5, 2048, 32, "Client Key");
+	print_yespower(YESPOWER_0_5, 1024, 32, "Client Key");
+
+	print_yespower(YESPOWER_0_5, 2048, 8, NULL); /* no personality */
+
+	print_yespower(YESPOWER_1_0, 2048, 8, NULL);
+	print_yespower(YESPOWER_1_0, 4096, 16, NULL);
+	print_yespower(YESPOWER_1_0, 4096, 32, NULL);
+	print_yespower(YESPOWER_1_0, 2048, 32, NULL);
+	print_yespower(YESPOWER_1_0, 1024, 32, NULL);
+
+	print_yespower(YESPOWER_1_0, 1024, 32, "personality test");
+
+	print_yespower_loop(YESPOWER_0_5, "Client Key");
+	print_yespower_loop(YESPOWER_1_0, NULL);
+
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-opt.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-opt.c
new file mode 100644
index 0000000..3159800
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-opt.c
@@ -0,0 +1,1147 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2012-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ *
+ * This is a proof-of-work focused fork of yescrypt, including optimized and
+ * cut-down implementation of the obsolete yescrypt 0.5 (based off its first
+ * submission to PHC back in 2014) and a new proof-of-work specific variation
+ * known as yespower 1.0.  The former is intended as an upgrade for
+ * cryptocurrencies that already use yescrypt 0.5 and the latter may be used
+ * as a further upgrade (hard fork) by those and other cryptocurrencies.  The
+ * version of algorithm to use is requested through parameters, allowing for
+ * both algorithms to co-exist in client and miner implementations (such as in
+ * preparation for a hard-fork).
+ */
+
+#ifndef _YESPOWER_OPT_C_PASS_
+#define _YESPOWER_OPT_C_PASS_ 1
+#endif
+
+#if _YESPOWER_OPT_C_PASS_ == 1
+/*
+ * AVX and especially XOP speed up Salsa20 a lot, but needlessly result in
+ * extra instruction prefixes for pwxform (which we make more use of).  While
+ * no slowdown from the prefixes is generally observed on AMD CPUs supporting
+ * XOP, some slowdown is sometimes observed on Intel CPUs with AVX.
+ */
+#ifdef __XOP__
+#warning "Note: XOP is enabled.  That's great."
+#elif defined(__AVX__)
+#warning "Note: AVX is enabled.  That's OK."
+#elif defined(__SSE2__)
+#warning "Note: AVX and XOP are not enabled.  That's OK."
+#elif defined(__x86_64__) || defined(__i386__)
+#warning "SSE2 not enabled.  Expect poor performance."
+#else
+#warning "Note: building generic code for non-x86.  That's OK."
+#endif
+
+/*
+ * The SSE4 code version has fewer instructions than the generic SSE2 version,
+ * but all of the instructions are SIMD, thereby wasting the scalar execution
+ * units.  Thus, the generic SSE2 version below actually runs faster on some
+ * CPUs due to its balanced mix of SIMD and scalar instructions.
+ */
+#undef USE_SSE4_FOR_32BIT
+
+#ifdef __SSE2__
+/*
+ * GCC before 4.9 would by default unnecessarily use store/load (without
+ * SSE4.1) or (V)PEXTR (with SSE4.1 or AVX) instead of simply (V)MOV.
+ * This was tracked as GCC bug 54349.
+ * "-mtune=corei7" works around this, but is only supported for GCC 4.6+.
+ * We use inline asm for pre-4.6 GCC, further down this file.
+ */
+#if __GNUC__ == 4 && __GNUC_MINOR__ >= 6 && __GNUC_MINOR__ < 9 && \
+    !defined(__clang__) && !defined(__ICC)
+#pragma GCC target ("tune=corei7")
+#endif
+#include <emmintrin.h>
+#ifdef __XOP__
+#include <x86intrin.h>
+#endif
+#elif defined(__SSE__)
+#include <xmmintrin.h>
+#endif
+
+#include <errno.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "insecure_memzero.h"
+#include "sha256.h"
+#include "sysendian.h"
+
+#include "yespower.h"
+
+#include "yespower-platform.c"
+
+#if __STDC_VERSION__ >= 199901L
+/* Have restrict */
+#elif defined(__GNUC__)
+#define restrict __restrict
+#else
+#define restrict
+#endif
+
+#ifdef __GNUC__
+#define unlikely(exp) __builtin_expect(exp, 0)
+#else
+#define unlikely(exp) (exp)
+#endif
+
+#ifdef __SSE__
+#define PREFETCH(x, hint) _mm_prefetch((const char *)(x), (hint));
+#else
+#undef PREFETCH
+#endif
+
+typedef union {
+	uint32_t w[16];
+	uint64_t d[8];
+#ifdef __SSE2__
+	__m128i q[4];
+#endif
+} salsa20_blk_t;
+
+static inline void salsa20_simd_shuffle(const salsa20_blk_t *Bin,
+    salsa20_blk_t *Bout)
+{
+#define COMBINE(out, in1, in2) \
+	Bout->d[out] = Bin->w[in1 * 2] | ((uint64_t)Bin->w[in2 * 2 + 1] << 32);
+	COMBINE(0, 0, 2)
+	COMBINE(1, 5, 7)
+	COMBINE(2, 2, 4)
+	COMBINE(3, 7, 1)
+	COMBINE(4, 4, 6)
+	COMBINE(5, 1, 3)
+	COMBINE(6, 6, 0)
+	COMBINE(7, 3, 5)
+#undef COMBINE
+}
+
+static inline void salsa20_simd_unshuffle(const salsa20_blk_t *Bin,
+    salsa20_blk_t *Bout)
+{
+#define UNCOMBINE(out, in1, in2) \
+	Bout->w[out * 2] = Bin->d[in1]; \
+	Bout->w[out * 2 + 1] = Bin->d[in2] >> 32;
+	UNCOMBINE(0, 0, 6)
+	UNCOMBINE(1, 5, 3)
+	UNCOMBINE(2, 2, 0)
+	UNCOMBINE(3, 7, 5)
+	UNCOMBINE(4, 4, 2)
+	UNCOMBINE(5, 1, 7)
+	UNCOMBINE(6, 6, 4)
+	UNCOMBINE(7, 3, 1)
+#undef UNCOMBINE
+}
+
+#ifdef __SSE2__
+#define DECL_X \
+	__m128i X0, X1, X2, X3;
+#define DECL_Y \
+	__m128i Y0, Y1, Y2, Y3;
+#define READ_X(in) \
+	X0 = (in).q[0]; X1 = (in).q[1]; X2 = (in).q[2]; X3 = (in).q[3];
+#define WRITE_X(out) \
+	(out).q[0] = X0; (out).q[1] = X1; (out).q[2] = X2; (out).q[3] = X3;
+
+#ifdef __XOP__
+#define ARX(out, in1, in2, s) \
+	out = _mm_xor_si128(out, _mm_roti_epi32(_mm_add_epi32(in1, in2), s));
+#else
+#define ARX(out, in1, in2, s) { \
+	__m128i tmp = _mm_add_epi32(in1, in2); \
+	out = _mm_xor_si128(out, _mm_slli_epi32(tmp, s)); \
+	out = _mm_xor_si128(out, _mm_srli_epi32(tmp, 32 - s)); \
+}
+#endif
+
+#define SALSA20_2ROUNDS \
+	/* Operate on "columns" */ \
+	ARX(X1, X0, X3, 7) \
+	ARX(X2, X1, X0, 9) \
+	ARX(X3, X2, X1, 13) \
+	ARX(X0, X3, X2, 18) \
+	/* Rearrange data */ \
+	X1 = _mm_shuffle_epi32(X1, 0x93); \
+	X2 = _mm_shuffle_epi32(X2, 0x4E); \
+	X3 = _mm_shuffle_epi32(X3, 0x39); \
+	/* Operate on "rows" */ \
+	ARX(X3, X0, X1, 7) \
+	ARX(X2, X3, X0, 9) \
+	ARX(X1, X2, X3, 13) \
+	ARX(X0, X1, X2, 18) \
+	/* Rearrange data */ \
+	X1 = _mm_shuffle_epi32(X1, 0x39); \
+	X2 = _mm_shuffle_epi32(X2, 0x4E); \
+	X3 = _mm_shuffle_epi32(X3, 0x93);
+
+/**
+ * Apply the Salsa20 core to the block provided in (X0 ... X3).
+ */
+#define SALSA20_wrapper(out, rounds) { \
+	__m128i Z0 = X0, Z1 = X1, Z2 = X2, Z3 = X3; \
+	rounds \
+	(out).q[0] = X0 = _mm_add_epi32(X0, Z0); \
+	(out).q[1] = X1 = _mm_add_epi32(X1, Z1); \
+	(out).q[2] = X2 = _mm_add_epi32(X2, Z2); \
+	(out).q[3] = X3 = _mm_add_epi32(X3, Z3); \
+}
+
+/**
+ * Apply the Salsa20/2 core to the block provided in X.
+ */
+#define SALSA20_2(out) \
+	SALSA20_wrapper(out, SALSA20_2ROUNDS)
+
+#define SALSA20_8ROUNDS \
+	SALSA20_2ROUNDS SALSA20_2ROUNDS SALSA20_2ROUNDS SALSA20_2ROUNDS
+
+/**
+ * Apply the Salsa20/8 core to the block provided in X.
+ */
+#define SALSA20_8(out) \
+	SALSA20_wrapper(out, SALSA20_8ROUNDS)
+
+#define XOR_X(in) \
+	X0 = _mm_xor_si128(X0, (in).q[0]); \
+	X1 = _mm_xor_si128(X1, (in).q[1]); \
+	X2 = _mm_xor_si128(X2, (in).q[2]); \
+	X3 = _mm_xor_si128(X3, (in).q[3]);
+
+#define XOR_X_2(in1, in2) \
+	X0 = _mm_xor_si128((in1).q[0], (in2).q[0]); \
+	X1 = _mm_xor_si128((in1).q[1], (in2).q[1]); \
+	X2 = _mm_xor_si128((in1).q[2], (in2).q[2]); \
+	X3 = _mm_xor_si128((in1).q[3], (in2).q[3]);
+
+#define XOR_X_WRITE_XOR_Y_2(out, in) \
+	(out).q[0] = Y0 = _mm_xor_si128((out).q[0], (in).q[0]); \
+	(out).q[1] = Y1 = _mm_xor_si128((out).q[1], (in).q[1]); \
+	(out).q[2] = Y2 = _mm_xor_si128((out).q[2], (in).q[2]); \
+	(out).q[3] = Y3 = _mm_xor_si128((out).q[3], (in).q[3]); \
+	X0 = _mm_xor_si128(X0, Y0); \
+	X1 = _mm_xor_si128(X1, Y1); \
+	X2 = _mm_xor_si128(X2, Y2); \
+	X3 = _mm_xor_si128(X3, Y3);
+
+#define INTEGERIFY _mm_cvtsi128_si32(X0)
+
+#else /* !defined(__SSE2__) */
+
+#define DECL_X \
+	salsa20_blk_t X;
+#define DECL_Y \
+	salsa20_blk_t Y;
+
+#define COPY(out, in) \
+	(out).d[0] = (in).d[0]; \
+	(out).d[1] = (in).d[1]; \
+	(out).d[2] = (in).d[2]; \
+	(out).d[3] = (in).d[3]; \
+	(out).d[4] = (in).d[4]; \
+	(out).d[5] = (in).d[5]; \
+	(out).d[6] = (in).d[6]; \
+	(out).d[7] = (in).d[7];
+
+#define READ_X(in) COPY(X, in)
+#define WRITE_X(out) COPY(out, X)
+
+/**
+ * salsa20(B):
+ * Apply the Salsa20 core to the provided block.
+ */
+static inline void salsa20(salsa20_blk_t *restrict B,
+    salsa20_blk_t *restrict Bout, uint32_t doublerounds)
+{
+	salsa20_blk_t X;
+#define x X.w
+
+	salsa20_simd_unshuffle(B, &X);
+
+	do {
+#define R(a,b) (((a) << (b)) | ((a) >> (32 - (b))))
+		/* Operate on columns */
+		x[ 4] ^= R(x[ 0]+x[12], 7);  x[ 8] ^= R(x[ 4]+x[ 0], 9);
+		x[12] ^= R(x[ 8]+x[ 4],13);  x[ 0] ^= R(x[12]+x[ 8],18);
+
+		x[ 9] ^= R(x[ 5]+x[ 1], 7);  x[13] ^= R(x[ 9]+x[ 5], 9);
+		x[ 1] ^= R(x[13]+x[ 9],13);  x[ 5] ^= R(x[ 1]+x[13],18);
+
+		x[14] ^= R(x[10]+x[ 6], 7);  x[ 2] ^= R(x[14]+x[10], 9);
+		x[ 6] ^= R(x[ 2]+x[14],13);  x[10] ^= R(x[ 6]+x[ 2],18);
+
+		x[ 3] ^= R(x[15]+x[11], 7);  x[ 7] ^= R(x[ 3]+x[15], 9);
+		x[11] ^= R(x[ 7]+x[ 3],13);  x[15] ^= R(x[11]+x[ 7],18);
+
+		/* Operate on rows */
+		x[ 1] ^= R(x[ 0]+x[ 3], 7);  x[ 2] ^= R(x[ 1]+x[ 0], 9);
+		x[ 3] ^= R(x[ 2]+x[ 1],13);  x[ 0] ^= R(x[ 3]+x[ 2],18);
+
+		x[ 6] ^= R(x[ 5]+x[ 4], 7);  x[ 7] ^= R(x[ 6]+x[ 5], 9);
+		x[ 4] ^= R(x[ 7]+x[ 6],13);  x[ 5] ^= R(x[ 4]+x[ 7],18);
+
+		x[11] ^= R(x[10]+x[ 9], 7);  x[ 8] ^= R(x[11]+x[10], 9);
+		x[ 9] ^= R(x[ 8]+x[11],13);  x[10] ^= R(x[ 9]+x[ 8],18);
+
+		x[12] ^= R(x[15]+x[14], 7);  x[13] ^= R(x[12]+x[15], 9);
+		x[14] ^= R(x[13]+x[12],13);  x[15] ^= R(x[14]+x[13],18);
+#undef R
+	} while (--doublerounds);
+#undef x
+
+	{
+		uint32_t i;
+		salsa20_simd_shuffle(&X, Bout);
+		for (i = 0; i < 16; i += 4) {
+			B->w[i] = Bout->w[i] += B->w[i];
+			B->w[i + 1] = Bout->w[i + 1] += B->w[i + 1];
+			B->w[i + 2] = Bout->w[i + 2] += B->w[i + 2];
+			B->w[i + 3] = Bout->w[i + 3] += B->w[i + 3];
+		}
+	}
+}
+
+/**
+ * Apply the Salsa20/2 core to the block provided in X.
+ */
+#define SALSA20_2(out) \
+	salsa20(&X, &out, 1);
+
+/**
+ * Apply the Salsa20/8 core to the block provided in X.
+ */
+#define SALSA20_8(out) \
+	salsa20(&X, &out, 4);
+
+#define XOR(out, in1, in2) \
+	(out).d[0] = (in1).d[0] ^ (in2).d[0]; \
+	(out).d[1] = (in1).d[1] ^ (in2).d[1]; \
+	(out).d[2] = (in1).d[2] ^ (in2).d[2]; \
+	(out).d[3] = (in1).d[3] ^ (in2).d[3]; \
+	(out).d[4] = (in1).d[4] ^ (in2).d[4]; \
+	(out).d[5] = (in1).d[5] ^ (in2).d[5]; \
+	(out).d[6] = (in1).d[6] ^ (in2).d[6]; \
+	(out).d[7] = (in1).d[7] ^ (in2).d[7];
+
+#define XOR_X(in) XOR(X, X, in)
+#define XOR_X_2(in1, in2) XOR(X, in1, in2)
+#define XOR_X_WRITE_XOR_Y_2(out, in) \
+	XOR(Y, out, in) \
+	COPY(out, Y) \
+	XOR(X, X, Y)
+
+#define INTEGERIFY (uint32_t)X.d[0]
+#endif
+
+/**
+ * Apply the Salsa20 core to the block provided in X ^ in.
+ */
+#define SALSA20_XOR_MEM(in, out) \
+	XOR_X(in) \
+	SALSA20(out)
+
+#define SALSA20 SALSA20_8
+#else /* pass 2 */
+#undef SALSA20
+#define SALSA20 SALSA20_2
+#endif
+
+/**
+ * blockmix_salsa(Bin, Bout):
+ * Compute Bout = BlockMix_{salsa20, 1}(Bin).  The input Bin must be 128
+ * bytes in length; the output Bout must also be the same size.
+ */
+static inline void blockmix_salsa(const salsa20_blk_t *restrict Bin,
+    salsa20_blk_t *restrict Bout)
+{
+	DECL_X
+
+	READ_X(Bin[1])
+	SALSA20_XOR_MEM(Bin[0], Bout[0])
+	SALSA20_XOR_MEM(Bin[1], Bout[1])
+}
+
+static inline uint32_t blockmix_salsa_xor(const salsa20_blk_t *restrict Bin1,
+    const salsa20_blk_t *restrict Bin2, salsa20_blk_t *restrict Bout)
+{
+	DECL_X
+
+	XOR_X_2(Bin1[1], Bin2[1])
+	XOR_X(Bin1[0])
+	SALSA20_XOR_MEM(Bin2[0], Bout[0])
+	XOR_X(Bin1[1])
+	SALSA20_XOR_MEM(Bin2[1], Bout[1])
+
+	return INTEGERIFY;
+}
+
+#if _YESPOWER_OPT_C_PASS_ == 1
+/* This is tunable, but it is part of what defines a yespower version */
+/* Version 0.5 */
+#define Swidth_0_5 8
+/* Version 1.0 */
+#define Swidth_1_0 11
+
+/* Not tunable in this implementation, hard-coded in a few places */
+#define PWXsimple 2
+#define PWXgather 4
+
+/* Derived value.  Not tunable on its own. */
+#define PWXbytes (PWXgather * PWXsimple * 8)
+
+/* (Maybe-)runtime derived values.  Not tunable on their own. */
+#define Swidth_to_Sbytes1(Swidth) ((1 << (Swidth)) * PWXsimple * 8)
+#define Swidth_to_Smask(Swidth) (((1 << (Swidth)) - 1) * PWXsimple * 8)
+#define Smask_to_Smask2(Smask) (((uint64_t)(Smask) << 32) | (Smask))
+
+/* These should be compile-time derived */
+#define Smask2_0_5 Smask_to_Smask2(Swidth_to_Smask(Swidth_0_5))
+#define Smask2_1_0 Smask_to_Smask2(Swidth_to_Smask(Swidth_1_0))
+
+typedef struct {
+	uint8_t *S0, *S1, *S2;
+	size_t w;
+	uint32_t Sbytes;
+} pwxform_ctx_t;
+
+#define DECL_SMASK2REG /* empty */
+#define MAYBE_MEMORY_BARRIER /* empty */
+
+#ifdef __SSE2__
+/*
+ * (V)PSRLDQ and (V)PSHUFD have higher throughput than (V)PSRLQ on some CPUs
+ * starting with Sandy Bridge.  Additionally, PSHUFD uses separate source and
+ * destination registers, whereas the shifts would require an extra move
+ * instruction for our code when building without AVX.  Unfortunately, PSHUFD
+ * is much slower on Conroe (4 cycles latency vs. 1 cycle latency for PSRLQ)
+ * and somewhat slower on some non-Intel CPUs (luckily not including AMD
+ * Bulldozer and Piledriver).
+ */
+#ifdef __AVX__
+#define HI32(X) \
+	_mm_srli_si128((X), 4)
+#elif 1 /* As an option, check for __SSE4_1__ here not to hurt Conroe */
+#define HI32(X) \
+	_mm_shuffle_epi32((X), _MM_SHUFFLE(2,3,0,1))
+#else
+#define HI32(X) \
+	_mm_srli_epi64((X), 32)
+#endif
+
+#if defined(__x86_64__) && \
+    __GNUC__ == 4 && __GNUC_MINOR__ < 6 && !defined(__ICC)
+#ifdef __AVX__
+#define MOVQ "vmovq"
+#else
+/* "movq" would be more correct, but "movd" is supported by older binutils
+ * due to an error in AMD's spec for x86-64. */
+#define MOVQ "movd"
+#endif
+#define EXTRACT64(X) ({ \
+	uint64_t result; \
+	__asm__(MOVQ " %1, %0" : "=r" (result) : "x" (X)); \
+	result; \
+})
+#elif defined(__x86_64__) && !defined(_MSC_VER) && !defined(__OPEN64__)
+/* MSVC and Open64 had bugs */
+#define EXTRACT64(X) _mm_cvtsi128_si64(X)
+#elif defined(__x86_64__) && defined(__SSE4_1__)
+/* No known bugs for this intrinsic */
+#include <smmintrin.h>
+#define EXTRACT64(X) _mm_extract_epi64((X), 0)
+#elif defined(USE_SSE4_FOR_32BIT) && defined(__SSE4_1__)
+/* 32-bit */
+#include <smmintrin.h>
+#if 0
+/* This is currently unused by the code below, which instead uses these two
+ * intrinsics explicitly when (!defined(__x86_64__) && defined(__SSE4_1__)) */
+#define EXTRACT64(X) \
+	((uint64_t)(uint32_t)_mm_cvtsi128_si32(X) | \
+	((uint64_t)(uint32_t)_mm_extract_epi32((X), 1) << 32))
+#endif
+#else
+/* 32-bit or compilers with known past bugs in _mm_cvtsi128_si64() */
+#define EXTRACT64(X) \
+	((uint64_t)(uint32_t)_mm_cvtsi128_si32(X) | \
+	((uint64_t)(uint32_t)_mm_cvtsi128_si32(HI32(X)) << 32))
+#endif
+
+#if defined(__x86_64__) && (defined(__AVX__) || !defined(__GNUC__))
+/* 64-bit with AVX */
+/* Force use of 64-bit AND instead of two 32-bit ANDs */
+#undef DECL_SMASK2REG
+#if defined(__GNUC__) && !defined(__ICC)
+#define DECL_SMASK2REG uint64_t Smask2reg = Smask2;
+/* Force use of lower-numbered registers to reduce number of prefixes, relying
+ * on out-of-order execution and register renaming. */
+#define FORCE_REGALLOC_1 \
+	__asm__("" : "=a" (x), "+d" (Smask2reg), "+S" (S0), "+D" (S1));
+#define FORCE_REGALLOC_2 \
+	__asm__("" : : "c" (lo));
+#else
+static volatile uint64_t Smask2var = Smask2;
+#define DECL_SMASK2REG uint64_t Smask2reg = Smask2var;
+#define FORCE_REGALLOC_1 /* empty */
+#define FORCE_REGALLOC_2 /* empty */
+#endif
+#define PWXFORM_SIMD(X) { \
+	uint64_t x; \
+	FORCE_REGALLOC_1 \
+	uint32_t lo = x = EXTRACT64(X) & Smask2reg; \
+	FORCE_REGALLOC_2 \
+	uint32_t hi = x >> 32; \
+	X = _mm_mul_epu32(HI32(X), X); \
+	X = _mm_add_epi64(X, *(__m128i *)(S0 + lo)); \
+	X = _mm_xor_si128(X, *(__m128i *)(S1 + hi)); \
+}
+#elif defined(__x86_64__)
+/* 64-bit without AVX.  This relies on out-of-order execution and register
+ * renaming.  It may actually be fastest on CPUs with AVX(2) as well - e.g.,
+ * it runs great on Haswell. */
+#warning "Note: using x86-64 inline assembly for pwxform.  That's great."
+#undef MAYBE_MEMORY_BARRIER
+#define MAYBE_MEMORY_BARRIER \
+	__asm__("" : : : "memory");
+#define PWXFORM_SIMD(X) { \
+	__m128i H; \
+	__asm__( \
+	    "movd %0, %%rax\n\t" \
+	    "pshufd $0xb1, %0, %1\n\t" \
+	    "andq %2, %%rax\n\t" \
+	    "pmuludq %1, %0\n\t" \
+	    "movl %%eax, %%ecx\n\t" \
+	    "shrq $0x20, %%rax\n\t" \
+	    "paddq (%3,%%rcx), %0\n\t" \
+	    "pxor (%4,%%rax), %0\n\t" \
+	    : "+x" (X), "=x" (H) \
+	    : "d" (Smask2), "S" (S0), "D" (S1) \
+	    : "cc", "ax", "cx"); \
+}
+#elif defined(USE_SSE4_FOR_32BIT) && defined(__SSE4_1__)
+/* 32-bit with SSE4.1 */
+#define PWXFORM_SIMD(X) { \
+	__m128i x = _mm_and_si128(X, _mm_set1_epi64x(Smask2)); \
+	__m128i s0 = *(__m128i *)(S0 + (uint32_t)_mm_cvtsi128_si32(x)); \
+	__m128i s1 = *(__m128i *)(S1 + (uint32_t)_mm_extract_epi32(x, 1)); \
+	X = _mm_mul_epu32(HI32(X), X); \
+	X = _mm_add_epi64(X, s0); \
+	X = _mm_xor_si128(X, s1); \
+}
+#else
+/* 32-bit without SSE4.1 */
+#define PWXFORM_SIMD(X) { \
+	uint64_t x = EXTRACT64(X) & Smask2; \
+	__m128i s0 = *(__m128i *)(S0 + (uint32_t)x); \
+	__m128i s1 = *(__m128i *)(S1 + (x >> 32)); \
+	X = _mm_mul_epu32(HI32(X), X); \
+	X = _mm_add_epi64(X, s0); \
+	X = _mm_xor_si128(X, s1); \
+}
+#endif
+
+#define PWXFORM_SIMD_WRITE(X, Sw) \
+	PWXFORM_SIMD(X) \
+	MAYBE_MEMORY_BARRIER \
+	*(__m128i *)(Sw + w) = X; \
+	MAYBE_MEMORY_BARRIER
+
+#define PWXFORM_ROUND \
+	PWXFORM_SIMD(X0) \
+	PWXFORM_SIMD(X1) \
+	PWXFORM_SIMD(X2) \
+	PWXFORM_SIMD(X3)
+
+#define PWXFORM_ROUND_WRITE4 \
+	PWXFORM_SIMD_WRITE(X0, S0) \
+	PWXFORM_SIMD_WRITE(X1, S1) \
+	w += 16; \
+	PWXFORM_SIMD_WRITE(X2, S0) \
+	PWXFORM_SIMD_WRITE(X3, S1) \
+	w += 16;
+
+#define PWXFORM_ROUND_WRITE2 \
+	PWXFORM_SIMD_WRITE(X0, S0) \
+	PWXFORM_SIMD_WRITE(X1, S1) \
+	w += 16; \
+	PWXFORM_SIMD(X2) \
+	PWXFORM_SIMD(X3)
+
+#else /* !defined(__SSE2__) */
+
+#define PWXFORM_SIMD(x0, x1) { \
+	uint64_t x = x0 & Smask2; \
+	uint64_t *p0 = (uint64_t *)(S0 + (uint32_t)x); \
+	uint64_t *p1 = (uint64_t *)(S1 + (x >> 32)); \
+	x0 = ((x0 >> 32) * (uint32_t)x0 + p0[0]) ^ p1[0]; \
+	x1 = ((x1 >> 32) * (uint32_t)x1 + p0[1]) ^ p1[1]; \
+}
+
+#define PWXFORM_SIMD_WRITE(x0, x1, Sw) \
+	PWXFORM_SIMD(x0, x1) \
+	((uint64_t *)(Sw + w))[0] = x0; \
+	((uint64_t *)(Sw + w))[1] = x1;
+
+#define PWXFORM_ROUND \
+	PWXFORM_SIMD(X.d[0], X.d[1]) \
+	PWXFORM_SIMD(X.d[2], X.d[3]) \
+	PWXFORM_SIMD(X.d[4], X.d[5]) \
+	PWXFORM_SIMD(X.d[6], X.d[7])
+
+#define PWXFORM_ROUND_WRITE4 \
+	PWXFORM_SIMD_WRITE(X.d[0], X.d[1], S0) \
+	PWXFORM_SIMD_WRITE(X.d[2], X.d[3], S1) \
+	w += 16; \
+	PWXFORM_SIMD_WRITE(X.d[4], X.d[5], S0) \
+	PWXFORM_SIMD_WRITE(X.d[6], X.d[7], S1) \
+	w += 16;
+
+#define PWXFORM_ROUND_WRITE2 \
+	PWXFORM_SIMD_WRITE(X.d[0], X.d[1], S0) \
+	PWXFORM_SIMD_WRITE(X.d[2], X.d[3], S1) \
+	w += 16; \
+	PWXFORM_SIMD(X.d[4], X.d[5]) \
+	PWXFORM_SIMD(X.d[6], X.d[7])
+#endif
+
+#define PWXFORM \
+	PWXFORM_ROUND PWXFORM_ROUND PWXFORM_ROUND \
+	PWXFORM_ROUND PWXFORM_ROUND PWXFORM_ROUND
+
+#define Smask2 Smask2_0_5
+
+#else /* pass 2 */
+
+#undef PWXFORM
+#define PWXFORM \
+	PWXFORM_ROUND_WRITE4 PWXFORM_ROUND_WRITE2 PWXFORM_ROUND_WRITE2 \
+	w &= Smask2; \
+	{ \
+		uint8_t *Stmp = S2; \
+		S2 = S1; \
+		S1 = S0; \
+		S0 = Stmp; \
+	}
+
+#undef Smask2
+#define Smask2 Smask2_1_0
+
+#endif
+
+/**
+ * blockmix_pwxform(Bin, Bout, r, S):
+ * Compute Bout = BlockMix_pwxform{salsa20, r, S}(Bin).  The input Bin must
+ * be 128r bytes in length; the output Bout must also be the same size.
+ */
+static void blockmix(const salsa20_blk_t *restrict Bin,
+    salsa20_blk_t *restrict Bout, size_t r, pwxform_ctx_t *restrict ctx)
+{
+	if (unlikely(!ctx)) {
+		blockmix_salsa(Bin, Bout);
+		return;
+	}
+
+	uint8_t *S0 = ctx->S0, *S1 = ctx->S1;
+#if _YESPOWER_OPT_C_PASS_ > 1
+	uint8_t *S2 = ctx->S2;
+	size_t w = ctx->w;
+#endif
+	size_t i;
+	DECL_X
+
+	/* Convert count of 128-byte blocks to max index of 64-byte block */
+	r = r * 2 - 1;
+
+	READ_X(Bin[r])
+
+	DECL_SMASK2REG
+
+	i = 0;
+	do {
+		XOR_X(Bin[i])
+		PWXFORM
+		if (unlikely(i >= r))
+			break;
+		WRITE_X(Bout[i])
+		i++;
+	} while (1);
+
+#if _YESPOWER_OPT_C_PASS_ > 1
+	ctx->S0 = S0; ctx->S1 = S1; ctx->S2 = S2;
+	ctx->w = w;
+#endif
+
+	SALSA20(Bout[i])
+}
+
+static uint32_t blockmix_xor(const salsa20_blk_t *restrict Bin1,
+    const salsa20_blk_t *restrict Bin2, salsa20_blk_t *restrict Bout,
+    size_t r, pwxform_ctx_t *restrict ctx)
+{
+	if (unlikely(!ctx))
+		return blockmix_salsa_xor(Bin1, Bin2, Bout);
+
+	uint8_t *S0 = ctx->S0, *S1 = ctx->S1;
+#if _YESPOWER_OPT_C_PASS_ > 1
+	uint8_t *S2 = ctx->S2;
+	size_t w = ctx->w;
+#endif
+	size_t i;
+	DECL_X
+
+	/* Convert count of 128-byte blocks to max index of 64-byte block */
+	r = r * 2 - 1;
+
+#ifdef PREFETCH
+	PREFETCH(&Bin2[r], _MM_HINT_T0)
+	for (i = 0; i < r; i++) {
+		PREFETCH(&Bin2[i], _MM_HINT_T0)
+	}
+#endif
+
+	XOR_X_2(Bin1[r], Bin2[r])
+
+	DECL_SMASK2REG
+
+	i = 0;
+	r--;
+	do {
+		XOR_X(Bin1[i])
+		XOR_X(Bin2[i])
+		PWXFORM
+		WRITE_X(Bout[i])
+
+		XOR_X(Bin1[i + 1])
+		XOR_X(Bin2[i + 1])
+		PWXFORM
+
+		if (unlikely(i >= r))
+			break;
+
+		WRITE_X(Bout[i + 1])
+
+		i += 2;
+	} while (1);
+	i++;
+
+#if _YESPOWER_OPT_C_PASS_ > 1
+	ctx->S0 = S0; ctx->S1 = S1; ctx->S2 = S2;
+	ctx->w = w;
+#endif
+
+	SALSA20(Bout[i])
+
+	return INTEGERIFY;
+}
+
+static uint32_t blockmix_xor_save(salsa20_blk_t *restrict Bin1out,
+    salsa20_blk_t *restrict Bin2,
+    size_t r, pwxform_ctx_t *restrict ctx)
+{
+	uint8_t *S0 = ctx->S0, *S1 = ctx->S1;
+#if _YESPOWER_OPT_C_PASS_ > 1
+	uint8_t *S2 = ctx->S2;
+	size_t w = ctx->w;
+#endif
+	size_t i;
+	DECL_X
+	DECL_Y
+
+	/* Convert count of 128-byte blocks to max index of 64-byte block */
+	r = r * 2 - 1;
+
+#ifdef PREFETCH
+	PREFETCH(&Bin2[r], _MM_HINT_T0)
+	for (i = 0; i < r; i++) {
+		PREFETCH(&Bin2[i], _MM_HINT_T0)
+	}
+#endif
+
+	XOR_X_2(Bin1out[r], Bin2[r])
+
+	DECL_SMASK2REG
+
+	i = 0;
+	r--;
+	do {
+		XOR_X_WRITE_XOR_Y_2(Bin2[i], Bin1out[i])
+		PWXFORM
+		WRITE_X(Bin1out[i])
+
+		XOR_X_WRITE_XOR_Y_2(Bin2[i + 1], Bin1out[i + 1])
+		PWXFORM
+
+		if (unlikely(i >= r))
+			break;
+
+		WRITE_X(Bin1out[i + 1])
+
+		i += 2;
+	} while (1);
+	i++;
+
+#if _YESPOWER_OPT_C_PASS_ > 1
+	ctx->S0 = S0; ctx->S1 = S1; ctx->S2 = S2;
+	ctx->w = w;
+#endif
+
+	SALSA20(Bin1out[i])
+
+	return INTEGERIFY;
+}
+
+#if _YESPOWER_OPT_C_PASS_ == 1
+/**
+ * integerify(B, r):
+ * Return the result of parsing B_{2r-1} as a little-endian integer.
+ */
+static inline uint32_t integerify(const salsa20_blk_t *B, size_t r)
+{
+/*
+ * Our 64-bit words are in host byte order, which is why we don't just read
+ * w[0] here (would be wrong on big-endian).  Also, our 32-bit words are
+ * SIMD-shuffled, but we only care about the least significant 32 bits anyway.
+ */
+	return (uint32_t)B[2 * r - 1].d[0];
+}
+#endif
+
+/**
+ * smix1(B, r, N, V, XY, S):
+ * Compute first loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 128r+64 bytes in length.  N must be even and at least 4.
+ * The array V must be aligned to a multiple of 64 bytes, and arrays B and XY
+ * to a multiple of at least 16 bytes.
+ */
+static void smix1(uint8_t *B, size_t r, uint32_t N,
+    salsa20_blk_t *V, salsa20_blk_t *XY, pwxform_ctx_t *ctx)
+{
+	size_t s = 2 * r;
+	salsa20_blk_t *X = V, *Y = &V[s], *V_j;
+	uint32_t i, j, n;
+
+#if _YESPOWER_OPT_C_PASS_ == 1
+	for (i = 0; i < 2 * r; i++) {
+#else
+	for (i = 0; i < 2; i++) {
+#endif
+		const salsa20_blk_t *src = (salsa20_blk_t *)&B[i * 64];
+		salsa20_blk_t *tmp = Y;
+		salsa20_blk_t *dst = &X[i];
+		size_t k;
+		for (k = 0; k < 16; k++)
+			tmp->w[k] = le32dec(&src->w[k]);
+		salsa20_simd_shuffle(tmp, dst);
+	}
+
+#if _YESPOWER_OPT_C_PASS_ > 1
+	for (i = 1; i < r; i++)
+		blockmix(&X[(i - 1) * 2], &X[i * 2], 1, ctx);
+#endif
+
+	blockmix(X, Y, r, ctx);
+	X = Y + s;
+	blockmix(Y, X, r, ctx);
+	j = integerify(X, r);
+
+	for (n = 2; n < N; n <<= 1) {
+		uint32_t m = (n < N / 2) ? n : (N - 1 - n);
+		for (i = 1; i < m; i += 2) {
+			Y = X + s;
+			j &= n - 1;
+			j += i - 1;
+			V_j = &V[j * s];
+			j = blockmix_xor(X, V_j, Y, r, ctx);
+			j &= n - 1;
+			j += i;
+			V_j = &V[j * s];
+			X = Y + s;
+			j = blockmix_xor(Y, V_j, X, r, ctx);
+		}
+	}
+	n >>= 1;
+
+	j &= n - 1;
+	j += N - 2 - n;
+	V_j = &V[j * s];
+	Y = X + s;
+	j = blockmix_xor(X, V_j, Y, r, ctx);
+	j &= n - 1;
+	j += N - 1 - n;
+	V_j = &V[j * s];
+	blockmix_xor(Y, V_j, XY, r, ctx);
+
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = &XY[i];
+		salsa20_blk_t *tmp = &XY[s];
+		salsa20_blk_t *dst = (salsa20_blk_t *)&B[i * 64];
+		size_t k;
+		for (k = 0; k < 16; k++)
+			le32enc(&tmp->w[k], src->w[k]);
+		salsa20_simd_unshuffle(tmp, dst);
+	}
+}
+
+/**
+ * smix2(B, r, N, Nloop, V, XY, S):
+ * Compute second loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage XY must be 256r bytes in length.  N must be a power of 2 and at
+ * least 2.  Nloop must be even.  The array V must be aligned to a multiple of
+ * 64 bytes, and arrays B and XY to a multiple of at least 16 bytes.
+ */
+static void smix2(uint8_t *B, size_t r, uint32_t N, uint32_t Nloop,
+    salsa20_blk_t *V, salsa20_blk_t *XY, pwxform_ctx_t *ctx)
+{
+	size_t s = 2 * r;
+	salsa20_blk_t *X = XY, *Y = &XY[s];
+	uint32_t i, j;
+
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = (salsa20_blk_t *)&B[i * 64];
+		salsa20_blk_t *tmp = Y;
+		salsa20_blk_t *dst = &X[i];
+		size_t k;
+		for (k = 0; k < 16; k++)
+			tmp->w[k] = le32dec(&src->w[k]);
+		salsa20_simd_shuffle(tmp, dst);
+	}
+
+	j = integerify(X, r) & (N - 1);
+
+#if _YESPOWER_OPT_C_PASS_ == 1
+	if (Nloop > 2) {
+#endif
+		do {
+			salsa20_blk_t *V_j = &V[j * s];
+			j = blockmix_xor_save(X, V_j, r, ctx) & (N - 1);
+			V_j = &V[j * s];
+			j = blockmix_xor_save(X, V_j, r, ctx) & (N - 1);
+		} while (Nloop -= 2);
+#if _YESPOWER_OPT_C_PASS_ == 1
+	} else {
+		do {
+			const salsa20_blk_t * V_j = &V[j * s];
+			j = blockmix_xor(X, V_j, Y, r, ctx) & (N - 1);
+			V_j = &V[j * s];
+			j = blockmix_xor(Y, V_j, X, r, ctx) & (N - 1);
+		} while (Nloop -= 2);
+	}
+#endif
+
+	for (i = 0; i < 2 * r; i++) {
+		const salsa20_blk_t *src = &X[i];
+		salsa20_blk_t *tmp = Y;
+		salsa20_blk_t *dst = (salsa20_blk_t *)&B[i * 64];
+		size_t k;
+		for (k = 0; k < 16; k++)
+			le32enc(&tmp->w[k], src->w[k]);
+		salsa20_simd_unshuffle(tmp, dst);
+	}
+}
+
+/**
+ * smix(B, r, N, V, XY, S):
+ * Compute B = SMix_r(B, N).  The input B must be 128rp bytes in length; the
+ * temporary storage V must be 128rN bytes in length; the temporary storage
+ * XY must be 256r bytes in length.  N must be a power of 2 and at least 16.
+ * The array V must be aligned to a multiple of 64 bytes, and arrays B and XY
+ * to a multiple of at least 16 bytes (aligning them to 64 bytes as well saves
+ * cache lines, but it might also result in cache bank conflicts).
+ */
+static void smix(uint8_t *B, size_t r, uint32_t N,
+    salsa20_blk_t *V, salsa20_blk_t *XY, pwxform_ctx_t *ctx)
+{
+#if _YESPOWER_OPT_C_PASS_ == 1
+	uint32_t Nloop_all = (N + 2) / 3; /* 1/3, round up */
+	uint32_t Nloop_rw = Nloop_all;
+
+	Nloop_all++; Nloop_all &= ~(uint32_t)1; /* round up to even */
+	Nloop_rw &= ~(uint32_t)1; /* round down to even */
+#else
+	uint32_t Nloop_rw = (N + 2) / 3; /* 1/3, round up */
+	Nloop_rw++; Nloop_rw &= ~(uint32_t)1; /* round up to even */
+#endif
+
+	smix1(B, 1, ctx->Sbytes / 128, (salsa20_blk_t *)ctx->S0, XY, NULL);
+	smix1(B, r, N, V, XY, ctx);
+	smix2(B, r, N, Nloop_rw /* must be > 2 */, V, XY, ctx);
+#if _YESPOWER_OPT_C_PASS_ == 1
+	if (Nloop_all > Nloop_rw)
+		smix2(B, r, N, 2, V, XY, ctx);
+#endif
+}
+
+#if _YESPOWER_OPT_C_PASS_ == 1
+#undef _YESPOWER_OPT_C_PASS_
+#define _YESPOWER_OPT_C_PASS_ 2
+#define blockmix_salsa blockmix_salsa_1_0
+#define blockmix_salsa_xor blockmix_salsa_xor_1_0
+#define blockmix blockmix_1_0
+#define blockmix_xor blockmix_xor_1_0
+#define blockmix_xor_save blockmix_xor_save_1_0
+#define smix1 smix1_1_0
+#define smix2 smix2_1_0
+#define smix smix_1_0
+#include "yespower-opt.c"
+#undef smix
+
+/**
+ * yespower(local, src, srclen, params, dst):
+ * Compute yespower(src[0 .. srclen - 1], N, r), to be checked for "< target".
+ * local is the thread-local data structure, allowing to preserve and reuse a
+ * memory allocation across calls, thereby reducing its overhead.
+ *
+ * Return 0 on success; or -1 on error.
+ */
+int yespower(yespower_local_t *local,
+    const uint8_t *src, size_t srclen,
+    const yespower_params_t *params,
+    yespower_binary_t *dst)
+{
+	yespower_version_t version = params->version;
+	uint32_t N = params->N;
+	uint32_t r = params->r;
+	const uint8_t *pers = params->pers;
+	size_t perslen = params->perslen;
+	uint32_t Swidth;
+	size_t B_size, V_size, XY_size, need;
+	uint8_t *B, *S;
+	salsa20_blk_t *V, *XY;
+	pwxform_ctx_t ctx;
+	uint8_t sha256[32];
+
+	/* Sanity-check parameters */
+	if ((version != YESPOWER_0_5 && version != YESPOWER_1_0) ||
+	    N < 1024 || N > 512 * 1024 || r < 8 || r > 32 ||
+	    (N & (N - 1)) != 0 ||
+	    (!pers && perslen)) {
+		errno = EINVAL;
+		return -1;
+	}
+
+	/* Allocate memory */
+	B_size = (size_t)128 * r;
+	V_size = B_size * N;
+	if (version == YESPOWER_0_5) {
+		XY_size = B_size * 2;
+		Swidth = Swidth_0_5;
+		ctx.Sbytes = 2 * Swidth_to_Sbytes1(Swidth);
+	} else {
+		XY_size = B_size + 64;
+		Swidth = Swidth_1_0;
+		ctx.Sbytes = 3 * Swidth_to_Sbytes1(Swidth);
+	}
+	need = B_size + V_size + XY_size + ctx.Sbytes;
+	if (local->aligned_size < need) {
+		if (free_region(local))
+			return -1;
+		if (!alloc_region(local, need))
+			return -1;
+	}
+	B = (uint8_t *)local->aligned;
+	V = (salsa20_blk_t *)((uint8_t *)B + B_size);
+	XY = (salsa20_blk_t *)((uint8_t *)V + V_size);
+	S = (uint8_t *)XY + XY_size;
+	ctx.S0 = S;
+	ctx.S1 = S + Swidth_to_Sbytes1(Swidth);
+
+	SHA256_Buf(src, srclen, sha256);
+
+	if (version == YESPOWER_0_5) {
+		PBKDF2_SHA256(sha256, sizeof(sha256), src, srclen, 1,
+		    B, B_size);
+		memcpy(sha256, B, sizeof(sha256));
+		smix(B, r, N, V, XY, &ctx);
+		PBKDF2_SHA256(sha256, sizeof(sha256), B, B_size, 1,
+		    (uint8_t *)dst, sizeof(*dst));
+
+		if (pers) {
+			HMAC_SHA256_Buf(dst, sizeof(*dst), pers, perslen,
+			    sha256);
+			SHA256_Buf(sha256, sizeof(sha256), (uint8_t *)dst);
+		}
+	} else {
+		ctx.S2 = S + 2 * Swidth_to_Sbytes1(Swidth);
+		ctx.w = 0;
+
+		if (pers) {
+			src = pers;
+			srclen = perslen;
+		} else {
+			srclen = 0;
+		}
+
+		PBKDF2_SHA256(sha256, sizeof(sha256), src, srclen, 1, B, 128);
+		memcpy(sha256, B, sizeof(sha256));
+		smix_1_0(B, r, N, V, XY, &ctx);
+		HMAC_SHA256_Buf(B + B_size - 64, 64,
+		    sha256, sizeof(sha256), (uint8_t *)dst);
+	}
+
+	/* Success! */
+	return 0;
+}
+
+/**
+ * yespower_tls(src, srclen, params, dst):
+ * Compute yespower(src[0 .. srclen - 1], N, r), to be checked for "< target".
+ * The memory allocation is maintained internally using thread-local storage.
+ *
+ * Return 0 on success; or -1 on error.
+ */
+int yespower_tls(const uint8_t *src, size_t srclen,
+    const yespower_params_t *params, yespower_binary_t *dst)
+{
+	static __thread int initialized = 0;
+	static __thread yespower_local_t local;
+
+	if (!initialized) {
+		if (yespower_init_local(&local))
+			return -1;
+		initialized = 1;
+	}
+
+	return yespower(&local, src, srclen, params, dst);
+}
+
+int yespower_init_local(yespower_local_t *local)
+{
+	init_region(local);
+	return 0;
+}
+
+int yespower_free_local(yespower_local_t *local)
+{
+	return free_region(local);
+}
+#endif
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-platform.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-platform.c
new file mode 100644
index 0000000..2b1a03f
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-platform.c
@@ -0,0 +1,107 @@
+/*-
+ * Copyright 2013-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#ifdef __unix__
+#include <sys/mman.h>
+#endif
+
+#define HUGEPAGE_THRESHOLD		(12 * 1024 * 1024)
+
+#ifdef __x86_64__
+#define HUGEPAGE_SIZE			(2 * 1024 * 1024)
+#else
+#undef HUGEPAGE_SIZE
+#endif
+
+static void *alloc_region(yespower_region_t *region, size_t size)
+{
+	size_t base_size = size;
+	uint8_t *base, *aligned;
+#ifdef MAP_ANON
+	int flags =
+#ifdef MAP_NOCORE
+	    MAP_NOCORE |
+#endif
+	    MAP_ANON | MAP_PRIVATE;
+#if defined(MAP_HUGETLB) && defined(HUGEPAGE_SIZE)
+	size_t new_size = size;
+	const size_t hugepage_mask = (size_t)HUGEPAGE_SIZE - 1;
+	if (size >= HUGEPAGE_THRESHOLD && size + hugepage_mask >= size) {
+		flags |= MAP_HUGETLB;
+/*
+ * Linux's munmap() fails on MAP_HUGETLB mappings if size is not a multiple of
+ * huge page size, so let's round up to huge page size here.
+ */
+		new_size = size + hugepage_mask;
+		new_size &= ~hugepage_mask;
+	}
+	base = mmap(NULL, new_size, PROT_READ | PROT_WRITE, flags, -1, 0);
+	if (base != MAP_FAILED) {
+		base_size = new_size;
+	} else if (flags & MAP_HUGETLB) {
+		flags &= ~MAP_HUGETLB;
+		base = mmap(NULL, size, PROT_READ | PROT_WRITE, flags, -1, 0);
+	}
+
+#else
+	base = mmap(NULL, size, PROT_READ | PROT_WRITE, flags, -1, 0);
+#endif
+	if (base == MAP_FAILED)
+		base = NULL;
+	aligned = base;
+#elif defined(HAVE_POSIX_MEMALIGN)
+	if ((errno = posix_memalign((void **)&base, 64, size)) != 0)
+		base = NULL;
+	aligned = base;
+#else
+	base = aligned = NULL;
+	if (size + 63 < size) {
+		errno = ENOMEM;
+	} else if ((base = malloc(size + 63)) != NULL) {
+		aligned = base + 63;
+		aligned -= (uintptr_t)aligned & 63;
+	}
+#endif
+	region->base = base;
+	region->aligned = aligned;
+	region->base_size = base ? base_size : 0;
+	region->aligned_size = base ? size : 0;
+	return aligned;
+}
+
+static inline void init_region(yespower_region_t *region)
+{
+	region->base = region->aligned = NULL;
+	region->base_size = region->aligned_size = 0;
+}
+
+static int free_region(yespower_region_t *region)
+{
+	if (region->base) {
+#ifdef MAP_ANON
+		if (munmap(region->base, region->base_size))
+			return -1;
+#else
+		free(region->base);
+#endif
+	}
+	init_region(region);
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-ref.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-ref.c
new file mode 100644
index 0000000..29c03d2
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower-ref.c
@@ -0,0 +1,580 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2013-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ *
+ * This is a proof-of-work focused fork of yescrypt, including reference and
+ * cut-down implementation of the obsolete yescrypt 0.5 (based off its first
+ * submission to PHC back in 2014) and a new proof-of-work specific variation
+ * known as yespower 1.0.  The former is intended as an upgrade for
+ * cryptocurrencies that already use yescrypt 0.5 and the latter may be used
+ * as a further upgrade (hard fork) by those and other cryptocurrencies.  The
+ * version of algorithm to use is requested through parameters, allowing for
+ * both algorithms to co-exist in client and miner implementations (such as in
+ * preparation for a hard-fork).
+ *
+ * This is the reference implementation.  Its purpose is to provide a simple
+ * human- and machine-readable specification that implementations intended
+ * for actual use should be tested against.  It is deliberately mostly not
+ * optimized, and it is not meant to be used in production.  Instead, use
+ * yespower-opt.c.
+ */
+
+#warning "This reference implementation is deliberately mostly not optimized. Use yespower-opt.c instead unless you're testing (against) the reference implementation on purpose."
+
+#include <errno.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "sha256.h"
+#include "sysendian.h"
+
+#include "yespower.h"
+
+static void blkcpy(uint32_t *dst, const uint32_t *src, size_t count)
+{
+	do {
+		*dst++ = *src++;
+	} while (--count);
+}
+
+static void blkxor(uint32_t *dst, const uint32_t *src, size_t count)
+{
+	do {
+		*dst++ ^= *src++;
+	} while (--count);
+}
+
+/**
+ * salsa20(B):
+ * Apply the Salsa20 core to the provided block.
+ */
+static void salsa20(uint32_t B[16], uint32_t rounds)
+{
+	uint32_t x[16];
+	size_t i;
+
+	/* SIMD unshuffle */
+	for (i = 0; i < 16; i++)
+		x[i * 5 % 16] = B[i];
+
+	for (i = 0; i < rounds; i += 2) {
+#define R(a,b) (((a) << (b)) | ((a) >> (32 - (b))))
+		/* Operate on columns */
+		x[ 4] ^= R(x[ 0]+x[12], 7);  x[ 8] ^= R(x[ 4]+x[ 0], 9);
+		x[12] ^= R(x[ 8]+x[ 4],13);  x[ 0] ^= R(x[12]+x[ 8],18);
+
+		x[ 9] ^= R(x[ 5]+x[ 1], 7);  x[13] ^= R(x[ 9]+x[ 5], 9);
+		x[ 1] ^= R(x[13]+x[ 9],13);  x[ 5] ^= R(x[ 1]+x[13],18);
+
+		x[14] ^= R(x[10]+x[ 6], 7);  x[ 2] ^= R(x[14]+x[10], 9);
+		x[ 6] ^= R(x[ 2]+x[14],13);  x[10] ^= R(x[ 6]+x[ 2],18);
+
+		x[ 3] ^= R(x[15]+x[11], 7);  x[ 7] ^= R(x[ 3]+x[15], 9);
+		x[11] ^= R(x[ 7]+x[ 3],13);  x[15] ^= R(x[11]+x[ 7],18);
+
+		/* Operate on rows */
+		x[ 1] ^= R(x[ 0]+x[ 3], 7);  x[ 2] ^= R(x[ 1]+x[ 0], 9);
+		x[ 3] ^= R(x[ 2]+x[ 1],13);  x[ 0] ^= R(x[ 3]+x[ 2],18);
+
+		x[ 6] ^= R(x[ 5]+x[ 4], 7);  x[ 7] ^= R(x[ 6]+x[ 5], 9);
+		x[ 4] ^= R(x[ 7]+x[ 6],13);  x[ 5] ^= R(x[ 4]+x[ 7],18);
+
+		x[11] ^= R(x[10]+x[ 9], 7);  x[ 8] ^= R(x[11]+x[10], 9);
+		x[ 9] ^= R(x[ 8]+x[11],13);  x[10] ^= R(x[ 9]+x[ 8],18);
+
+		x[12] ^= R(x[15]+x[14], 7);  x[13] ^= R(x[12]+x[15], 9);
+		x[14] ^= R(x[13]+x[12],13);  x[15] ^= R(x[14]+x[13],18);
+#undef R
+	}
+
+	/* SIMD shuffle */
+	for (i = 0; i < 16; i++)
+		B[i] += x[i * 5 % 16];
+}
+
+/**
+ * blockmix_salsa(B):
+ * Compute B = BlockMix_{salsa20, 1}(B).  The input B must be 128 bytes in
+ * length.
+ */
+static void blockmix_salsa(uint32_t *B, uint32_t rounds)
+{
+	uint32_t X[16];
+	size_t i;
+
+	/* 1: X <-- B_{2r - 1} */
+	blkcpy(X, &B[16], 16);
+
+	/* 2: for i = 0 to 2r - 1 do */
+	for (i = 0; i < 2; i++) {
+		/* 3: X <-- H(X xor B_i) */
+		blkxor(X, &B[i * 16], 16);
+		salsa20(X, rounds);
+
+		/* 4: Y_i <-- X */
+		/* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
+		blkcpy(&B[i * 16], X, 16);
+	}
+}
+
+/*
+ * These are tunable, but they must meet certain constraints and are part of
+ * what defines a yespower version.
+ */
+#define PWXsimple 2
+#define PWXgather 4
+/* Version 0.5 */
+#define PWXrounds_0_5 6
+#define Swidth_0_5 8
+/* Version 1.0 */
+#define PWXrounds_1_0 3
+#define Swidth_1_0 11
+
+/* Derived values.  Not tunable on their own. */
+#define PWXbytes (PWXgather * PWXsimple * 8)
+#define PWXwords (PWXbytes / sizeof(uint32_t))
+#define rmin ((PWXbytes + 127) / 128)
+
+/* Runtime derived values.  Not tunable on their own. */
+#define Swidth_to_Sbytes1(Swidth) ((1 << Swidth) * PWXsimple * 8)
+#define Swidth_to_Smask(Swidth) (((1 << Swidth) - 1) * PWXsimple * 8)
+
+typedef struct {
+	yespower_version_t version;
+	uint32_t salsa20_rounds;
+	uint32_t PWXrounds, Swidth, Sbytes, Smask;
+	uint32_t *S;
+	uint32_t (*S0)[2], (*S1)[2], (*S2)[2];
+	size_t w;
+} pwxform_ctx_t;
+
+/**
+ * pwxform(B):
+ * Transform the provided block using the provided S-boxes.
+ */
+static void pwxform(uint32_t *B, pwxform_ctx_t *ctx)
+{
+	uint32_t (*X)[PWXsimple][2] = (uint32_t (*)[PWXsimple][2])B;
+	uint32_t (*S0)[2] = ctx->S0, (*S1)[2] = ctx->S1, (*S2)[2] = ctx->S2;
+	uint32_t Smask = ctx->Smask;
+	size_t w = ctx->w;
+	size_t i, j, k;
+
+	/* 1: for i = 0 to PWXrounds - 1 do */
+	for (i = 0; i < ctx->PWXrounds; i++) {
+		/* 2: for j = 0 to PWXgather - 1 do */
+		for (j = 0; j < PWXgather; j++) {
+			uint32_t xl = X[j][0][0];
+			uint32_t xh = X[j][0][1];
+			uint32_t (*p0)[2], (*p1)[2];
+
+			/* 3: p0 <-- (lo(B_{j,0}) & Smask) / (PWXsimple * 8) */
+			p0 = S0 + (xl & Smask) / sizeof(*S0);
+			/* 4: p1 <-- (hi(B_{j,0}) & Smask) / (PWXsimple * 8) */
+			p1 = S1 + (xh & Smask) / sizeof(*S1);
+
+			/* 5: for k = 0 to PWXsimple - 1 do */
+			for (k = 0; k < PWXsimple; k++) {
+				uint64_t x, s0, s1;
+
+				/* 6: B_{j,k} <-- (hi(B_{j,k}) * lo(B_{j,k}) + S0_{p0,k}) xor S1_{p1,k} */
+				s0 = ((uint64_t)p0[k][1] << 32) + p0[k][0];
+				s1 = ((uint64_t)p1[k][1] << 32) + p1[k][0];
+
+				xl = X[j][k][0];
+				xh = X[j][k][1];
+
+				x = (uint64_t)xh * xl;
+				x += s0;
+				x ^= s1;
+
+				X[j][k][0] = x;
+				X[j][k][1] = x >> 32;
+			}
+
+			if (ctx->version != YESPOWER_0_5 &&
+			    (i == 0 || j < PWXgather / 2)) {
+				if (j & 1) {
+					for (k = 0; k < PWXsimple; k++) {
+						S1[w][0] = X[j][k][0];
+						S1[w][1] = X[j][k][1];
+						w++;
+					}
+				} else {
+					for (k = 0; k < PWXsimple; k++) {
+						S0[w + k][0] = X[j][k][0];
+						S0[w + k][1] = X[j][k][1];
+					}
+				}
+			}
+		}
+	}
+
+	if (ctx->version != YESPOWER_0_5) {
+		/* 14: (S0, S1, S2) <-- (S2, S0, S1) */
+		ctx->S0 = S2;
+		ctx->S1 = S0;
+		ctx->S2 = S1;
+		/* 15: w <-- w mod 2^Swidth */
+		ctx->w = w & ((1 << ctx->Swidth) * PWXsimple - 1);
+	}
+}
+
+/**
+ * blockmix_pwxform(B, ctx, r):
+ * Compute B = BlockMix_pwxform{salsa20, ctx, r}(B).  The input B must be
+ * 128r bytes in length.
+ */
+static void blockmix_pwxform(uint32_t *B, pwxform_ctx_t *ctx, size_t r)
+{
+	uint32_t X[PWXwords];
+	size_t r1, i;
+
+	/* Convert 128-byte blocks to PWXbytes blocks */
+	/* 1: r_1 <-- 128r / PWXbytes */
+	r1 = 128 * r / PWXbytes;
+
+	/* 2: X <-- B'_{r_1 - 1} */
+	blkcpy(X, &B[(r1 - 1) * PWXwords], PWXwords);
+
+	/* 3: for i = 0 to r_1 - 1 do */
+	for (i = 0; i < r1; i++) {
+		/* 4: if r_1 > 1 */
+		if (r1 > 1) {
+			/* 5: X <-- X xor B'_i */
+			blkxor(X, &B[i * PWXwords], PWXwords);
+		}
+
+		/* 7: X <-- pwxform(X) */
+		pwxform(X, ctx);
+
+		/* 8: B'_i <-- X */
+		blkcpy(&B[i * PWXwords], X, PWXwords);
+	}
+
+	/* 10: i <-- floor((r_1 - 1) * PWXbytes / 64) */
+	i = (r1 - 1) * PWXbytes / 64;
+
+	/* 11: B_i <-- H(B_i) */
+	salsa20(&B[i * 16], ctx->salsa20_rounds);
+
+#if 1 /* No-op with our current pwxform settings, but do it to make sure */
+	/* 12: for i = i + 1 to 2r - 1 do */
+	for (i++; i < 2 * r; i++) {
+		/* 13: B_i <-- H(B_i xor B_{i-1}) */
+		blkxor(&B[i * 16], &B[(i - 1) * 16], 16);
+		salsa20(&B[i * 16], ctx->salsa20_rounds);
+	}
+#endif
+}
+
+/**
+ * integerify(B, r):
+ * Return the result of parsing B_{2r-1} as a little-endian integer.
+ */
+static uint32_t integerify(const uint32_t *B, size_t r)
+{
+/*
+ * Our 32-bit words are in host byte order.  Also, they are SIMD-shuffled, but
+ * we only care about the least significant 32 bits anyway.
+ */
+	const uint32_t *X = &B[(2 * r - 1) * 16];
+	return X[0];
+}
+
+/**
+ * p2floor(x):
+ * Largest power of 2 not greater than argument.
+ */
+static uint32_t p2floor(uint32_t x)
+{
+	uint32_t y;
+	while ((y = x & (x - 1)))
+		x = y;
+	return x;
+}
+
+/**
+ * wrap(x, i):
+ * Wrap x to the range 0 to i-1.
+ */
+static uint32_t wrap(uint32_t x, uint32_t i)
+{
+	uint32_t n = p2floor(i);
+	return (x & (n - 1)) + (i - n);
+}
+
+/**
+ * smix1(B, r, N, V, X, ctx):
+ * Compute first loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage X must be 128r bytes in length.
+ */
+static void smix1(uint32_t *B, size_t r, uint32_t N,
+    uint32_t *V, uint32_t *X, pwxform_ctx_t *ctx)
+{
+	size_t s = 32 * r;
+	uint32_t i, j;
+	size_t k;
+
+	/* 1: X <-- B */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			X[k * 16 + i] = le32dec(&B[k * 16 + (i * 5 % 16)]);
+
+	if (ctx->version != YESPOWER_0_5) {
+		for (k = 1; k < r; k++) {
+			blkcpy(&X[k * 32], &X[(k - 1) * 32], 32);
+			blockmix_pwxform(&X[k * 32], ctx, 1);
+		}
+	}
+
+	/* 2: for i = 0 to N - 1 do */
+	for (i = 0; i < N; i++) {
+		/* 3: V_i <-- X */
+		blkcpy(&V[i * s], X, s);
+
+		if (i > 1) {
+			/* j <-- Wrap(Integerify(X), i) */
+			j = wrap(integerify(X, r), i);
+
+			/* X <-- X xor V_j */
+			blkxor(X, &V[j * s], s);
+		}
+
+		/* 4: X <-- H(X) */
+		if (V != ctx->S)
+			blockmix_pwxform(X, ctx, r);
+		else
+			blockmix_salsa(X, ctx->salsa20_rounds);
+	}
+
+	/* B' <-- X */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			le32enc(&B[k * 16 + (i * 5 % 16)], X[k * 16 + i]);
+}
+
+/**
+ * smix2(B, r, N, Nloop, V, X, ctx):
+ * Compute second loop of B = SMix_r(B, N).  The input B must be 128r bytes in
+ * length; the temporary storage V must be 128rN bytes in length; the temporary
+ * storage X must be 128r bytes in length.  The value N must be a power of 2
+ * greater than 1.
+ */
+static void smix2(uint32_t *B, size_t r, uint32_t N, uint32_t Nloop,
+    uint32_t *V, uint32_t *X, pwxform_ctx_t *ctx)
+{
+	size_t s = 32 * r;
+	uint32_t i, j;
+	size_t k;
+
+	/* X <-- B */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			X[k * 16 + i] = le32dec(&B[k * 16 + (i * 5 % 16)]);
+
+	/* 6: for i = 0 to N - 1 do */
+	for (i = 0; i < Nloop; i++) {
+		/* 7: j <-- Integerify(X) mod N */
+		j = integerify(X, r) & (N - 1);
+
+		/* 8.1: X <-- X xor V_j */
+		blkxor(X, &V[j * s], s);
+		/* V_j <-- X */
+		if (Nloop != 2)
+			blkcpy(&V[j * s], X, s);
+
+		/* 8.2: X <-- H(X) */
+		blockmix_pwxform(X, ctx, r);
+	}
+
+	/* 10: B' <-- X */
+	for (k = 0; k < 2 * r; k++)
+		for (i = 0; i < 16; i++)
+			le32enc(&B[k * 16 + (i * 5 % 16)], X[k * 16 + i]);
+}
+
+/**
+ * smix(B, r, N, p, t, V, X, ctx):
+ * Compute B = SMix_r(B, N).  The input B must be 128rp bytes in length; the
+ * temporary storage V must be 128rN bytes in length; the temporary storage
+ * X must be 128r bytes in length.  The value N must be a power of 2 and at
+ * least 16.
+ */
+static void smix(uint32_t *B, size_t r, uint32_t N,
+    uint32_t *V, uint32_t *X, pwxform_ctx_t *ctx)
+{
+	uint32_t Nloop_all = (N + 2) / 3; /* 1/3, round up */
+	uint32_t Nloop_rw = Nloop_all;
+
+	Nloop_all++; Nloop_all &= ~(uint32_t)1; /* round up to even */
+	if (ctx->version == YESPOWER_0_5) {
+		Nloop_rw &= ~(uint32_t)1; /* round down to even */
+	} else {
+		Nloop_rw++; Nloop_rw &= ~(uint32_t)1; /* round up to even */
+	}
+
+	smix1(B, 1, ctx->Sbytes / 128, ctx->S, X, ctx);
+	smix1(B, r, N, V, X, ctx);
+	smix2(B, r, N, Nloop_rw /* must be > 2 */, V, X, ctx);
+	smix2(B, r, N, Nloop_all - Nloop_rw /* 0 or 2 */, V, X, ctx);
+}
+
+/**
+ * yespower(local, src, srclen, params, dst):
+ * Compute yespower(src[0 .. srclen - 1], N, r), to be checked for "< target".
+ *
+ * Return 0 on success; or -1 on error.
+ */
+int yespower(yespower_local_t *local,
+    const uint8_t *src, size_t srclen,
+    const yespower_params_t *params, yespower_binary_t *dst)
+{
+	yespower_version_t version = params->version;
+	uint32_t N = params->N;
+	uint32_t r = params->r;
+	const uint8_t *pers = params->pers;
+	size_t perslen = params->perslen;
+	int retval = -1;
+	size_t B_size, V_size;
+	uint32_t *B, *V, *X, *S;
+	pwxform_ctx_t ctx;
+	uint32_t sha256[8];
+
+	/* Sanity-check parameters */
+	if ((version != YESPOWER_0_5 && version != YESPOWER_1_0) ||
+	    N < 1024 || N > 512 * 1024 || r < 8 || r > 32 ||
+	    (N & (N - 1)) != 0 || r < rmin ||
+	    (!pers && perslen)) {
+		errno = EINVAL;
+		return -1;
+	}
+
+	/* Allocate memory */
+	B_size = (size_t)128 * r;
+	V_size = B_size * N;
+	if ((V = malloc(V_size)) == NULL)
+		return -1;
+	if ((B = malloc(B_size)) == NULL)
+		goto free_V;
+	if ((X = malloc(B_size)) == NULL)
+		goto free_B;
+	ctx.version = version;
+	if (version == YESPOWER_0_5) {
+		ctx.salsa20_rounds = 8;
+		ctx.PWXrounds = PWXrounds_0_5;
+		ctx.Swidth = Swidth_0_5;
+		ctx.Sbytes = 2 * Swidth_to_Sbytes1(ctx.Swidth);
+	} else {
+		ctx.salsa20_rounds = 2;
+		ctx.PWXrounds = PWXrounds_1_0;
+		ctx.Swidth = Swidth_1_0;
+		ctx.Sbytes = 3 * Swidth_to_Sbytes1(ctx.Swidth);
+	}
+	if ((S = malloc(ctx.Sbytes)) == NULL)
+		goto free_X;
+	ctx.S = S;
+	ctx.S0 = (uint32_t (*)[2])S;
+	ctx.S1 = ctx.S0 + (1 << ctx.Swidth) * PWXsimple;
+	ctx.S2 = ctx.S1 + (1 << ctx.Swidth) * PWXsimple;
+	ctx.Smask = Swidth_to_Smask(ctx.Swidth);
+	ctx.w = 0;
+
+	SHA256_Buf(src, srclen, (uint8_t *)sha256);
+
+	if (version != YESPOWER_0_5) {
+		if (pers) {
+			src = pers;
+			srclen = perslen;
+		} else {
+			srclen = 0;
+		}
+	}
+
+	/* 1: (B_0 ... B_{p-1}) <-- PBKDF2(P, S, 1, p * MFLen) */
+	PBKDF2_SHA256((uint8_t *)sha256, sizeof(sha256),
+	    src, srclen, 1, (uint8_t *)B, B_size);
+
+	blkcpy(sha256, B, sizeof(sha256) / sizeof(sha256[0]));
+
+	/* 3: B_i <-- MF(B_i, N) */
+	smix(B, r, N, V, X, &ctx);
+
+	if (version == YESPOWER_0_5) {
+		/* 5: DK <-- PBKDF2(P, B, 1, dkLen) */
+		PBKDF2_SHA256((uint8_t *)sha256, sizeof(sha256),
+		    (uint8_t *)B, B_size, 1, (uint8_t *)dst, sizeof(*dst));
+
+		if (pers) {
+			HMAC_SHA256_Buf(dst, sizeof(*dst), pers, perslen,
+			    (uint8_t *)sha256);
+			SHA256_Buf(sha256, sizeof(sha256), (uint8_t *)dst);
+		}
+	} else {
+		HMAC_SHA256_Buf((uint8_t *)B + B_size - 64, 64,
+		    sha256, sizeof(sha256), (uint8_t *)dst);
+	}
+
+	/* Success! */
+	retval = 0;
+
+	/* Free memory */
+	free(S);
+free_X:
+	free(X);
+free_B:
+	free(B);
+free_V:
+	free(V);
+
+	return retval;
+}
+
+int yespower_tls(const uint8_t *src, size_t srclen,
+    const yespower_params_t *params, yespower_binary_t *dst)
+{
+/* The reference implementation doesn't use thread-local storage */
+	return yespower(NULL, src, srclen, params, dst);
+}
+
+int yespower_init_local(yespower_local_t *local)
+{
+/* The reference implementation doesn't use the local structure */
+	local->base = local->aligned = NULL;
+	local->base_size = local->aligned_size = 0;
+	return 0;
+}
+
+int yespower_free_local(yespower_local_t *local)
+{
+/* The reference implementation frees its memory in yespower() */
+	(void)local; /* unused */
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower.c b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower.c
new file mode 100644
index 0000000..3b8d083
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower.c
@@ -0,0 +1,109 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2012-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ *
+ * This is a proof-of-work focused fork of yescrypt, including optimized and
+ * cut-down implementation of the obsolete yescrypt 0.5 (based off its first
+ * submission to PHC back in 2014) and a new proof-of-work specific variation
+ * known as yespower 1.0.  The former is intended as an upgrade for
+ * cryptocurrencies that already use yescrypt 0.5 and the latter may be used
+ * as a further upgrade (hard fork) by those and other cryptocurrencies.  The
+ * version of algorithm to use is requested through parameters, allowing for
+ * both algorithms to co-exist in client and miner implementations (such as in
+ * preparation for a hard-fork).
+ */
+ 
+#include "miner.h"
+#include "yespower.h"
+#include "yespower-opt.c"
+
+static const yespower_params_t yespower_BITZENY = {YESPOWER_1_0, 4096, 16, NULL, 0};
+
+void yespower_hash( const char *input, char *output, uint32_t len )
+{
+  yespower_tls( (yespower_binary_t*)input, len,
+                &yespower_BITZENY,
+                (yespower_binary_t*)output );
+}
+
+extern struct work_restart *work_restart;
+
+extern bool fulltest(const uint32_t *hash, const uint32_t *target);
+
+static int pretest(const uint32_t *hash, const uint32_t *target)
+{
+	return hash[7] < target[7];
+}
+
+int scanhash_yespower(int thr_id, uint32_t *pdata,
+	const uint32_t *ptarget,
+	uint32_t max_nonce, unsigned long *hashes_done)
+{
+	static const yespower_params_t params = {
+		.version = YESPOWER_1_0,
+		.N = 4096,
+		.r = 16,
+		.pers = NULL,
+		.perslen = 0
+	};
+	union {
+		uint8_t u8[8];
+		uint32_t u32[20];
+	} data;
+	union {
+		yespower_binary_t yb;
+		uint32_t u32[7];
+	} hash;
+	uint32_t n = pdata[19] - 1;
+	const uint32_t Htarg = ptarget[7];
+	int i;
+
+	for (i = 0; i < 19; i++)
+		be32enc(&data.u32[i], pdata[i]);
+
+	do {
+		be32enc(&data.u32[19], ++n);
+
+		if (yespower_tls(data.u8, 80, &params, &hash.yb))
+			abort();
+
+		if (le32dec(&hash.u32[7]) <= Htarg) {
+			for (i = 0; i < 7; i++)
+				hash.u32[i] = le32dec(&hash.u32[i]);
+			if (fulltest(hash.u32, ptarget)) {
+				*hashes_done = n - pdata[19] + 1;
+				pdata[19] = n;
+				return 1;
+			}
+		}
+	} while (n < max_nonce && !work_restart[thr_id].restart);
+
+	*hashes_done = n - pdata[19] + 1;
+	pdata[19] = n;
+	return 0;
+}
diff --git a/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower.h b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower.h
new file mode 100644
index 0000000..b388d72
--- /dev/null
+++ b/bitzenymininglibrary/src/main/cpp/libcpuminer/yespower/yespower.h
@@ -0,0 +1,130 @@
+/*-
+ * Copyright 2009 Colin Percival
+ * Copyright 2013-2018 Alexander Peslyak
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * This file was originally written by Colin Percival as part of the Tarsnap
+ * online backup system.
+ */
+#ifndef _YESPOWER_H_
+#define _YESPOWER_H_
+
+#include <stdint.h>
+#include <stdlib.h> /* for size_t */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ * Internal type used by the memory allocator.  Please do not use it directly.
+ * Use yespower_local_t instead.
+ */
+typedef struct {
+	void *base, *aligned;
+	size_t base_size, aligned_size;
+} yespower_region_t;
+
+/**
+ * Type for thread-local (RAM) data structure.
+ */
+typedef yespower_region_t yespower_local_t;
+
+/*
+ * Type for yespower algorithm version numbers.
+ */
+typedef enum { YESPOWER_0_5 = 5, YESPOWER_1_0 = 10 } yespower_version_t;
+
+/**
+ * yespower parameters combined into one struct.
+ */
+typedef struct {
+	yespower_version_t version;
+	uint32_t N, r;
+	const uint8_t *pers;
+	size_t perslen;
+} yespower_params_t;
+
+/**
+ * A 256-bit yespower hash.
+ */
+typedef struct {
+	unsigned char uc[32];
+} yespower_binary_t;
+
+/**
+ * yespower_init_local(local):
+ * Initialize the thread-local (RAM) data structure.  Actual memory allocation
+ * is currently fully postponed until a call to yespower().
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * MT-safe as long as local is local to the thread.
+ */
+extern int yespower_init_local(yespower_local_t *local);
+
+/**
+ * yespower_free_local(local):
+ * Free memory that may have been allocated for an initialized thread-local
+ * (RAM) data structure.
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * MT-safe as long as local is local to the thread.
+ */
+extern int yespower_free_local(yespower_local_t *local);
+
+/**
+ * yespower(local, src, srclen, params, dst):
+ * Compute yespower(src[0 .. srclen - 1], N, r), to be checked for "< target".
+ * local is the thread-local data structure, allowing to preserve and reuse a
+ * memory allocation across calls, thereby reducing processing overhead.
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * local must be initialized with yespower_init_local().
+ *
+ * MT-safe as long as local and dst are local to the thread.
+ */
+extern int yespower(yespower_local_t *local,
+    const uint8_t *src, size_t srclen,
+    const yespower_params_t *params, yespower_binary_t *dst);
+
+/**
+ * yespower_tls(src, srclen, params, dst):
+ * Compute yespower(src[0 .. srclen - 1], N, r), to be checked for "< target".
+ * The memory allocation is maintained internally using thread-local storage.
+ *
+ * Return 0 on success; or -1 on error.
+ *
+ * MT-safe as long as dst is local to the thread.
+ */
+extern int yespower_tls(const uint8_t *src, size_t srclen,
+    const yespower_params_t *params, yespower_binary_t *dst);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* !_YESPOWER_H_ */
diff --git a/bitzenymininglibrary/src/main/java/com/example/ottylab/bitzenymininglibrary/BitZenyMiningLibrary.java b/bitzenymininglibrary/src/main/java/com/example/ottylab/bitzenymininglibrary/BitZenyMiningLibrary.java
new file mode 100644
index 0000000..cddd5d8
--- /dev/null
+++ b/bitzenymininglibrary/src/main/java/com/example/ottylab/bitzenymininglibrary/BitZenyMiningLibrary.java
@@ -0,0 +1,63 @@
+package com.example.ottylab.bitzenymininglibrary;
+
+import android.os.Bundle;
+import android.os.Handler;
+import android.os.Message;
+
+public class BitZenyMiningLibrary {
+    private static final String TAG = "BitZenyMiningLibrary";
+
+    static {
+        System.loadLibrary("bitzenymininglibrary");
+    }
+
+    public enum Algorithm {
+        YESCRYPT,
+        YESPOWER,
+    }
+
+    private static Handler sHandler;
+
+    public BitZenyMiningLibrary() {}
+    public BitZenyMiningLibrary(Handler handler) {
+        sHandler = handler;
+    }
+    public int startMining(String url, String user, String password, int n_threads, Algorithm algo) {
+        switch (algo) {
+            case YESCRYPT:
+                return startMining(url, user, password, n_threads, 0);
+            case YESPOWER:
+                return startMining(url, user, password, n_threads, 1);
+            default:
+                return -1;
+        }
+    }
+
+    public int startBenchmark(int n_threads, Algorithm algo) {
+        switch (algo) {
+            case YESCRYPT:
+                return startBenchmark(n_threads, 0);
+            case YESPOWER:
+                return startBenchmark(n_threads, 1);
+            default:
+                return -1;
+        }
+    }
+
+    private static void output(String message) {
+        // Generate message
+        Message msg = new Message();
+        Bundle bundle = new Bundle();
+        bundle.putString("log", message);
+        msg.setData(bundle);
+
+        if (sHandler != null) {
+            sHandler.sendMessage(msg);
+        }
+    }
+
+    public native boolean isMiningRunning();
+    public native int stopMining();
+    private native int startMining(String url, String user, String password, int n_threads, int algo);
+    private native int startBenchmark(int n_threads, int algo);
+}
\ No newline at end of file
diff --git a/bitzenymininglibrary/src/main/res/values/strings.xml b/bitzenymininglibrary/src/main/res/values/strings.xml
new file mode 100644
index 0000000..117db75
--- /dev/null
+++ b/bitzenymininglibrary/src/main/res/values/strings.xml
@@ -0,0 +1,3 @@
+<resources>
+    <string name="app_name">BitzenyMiningLibrary</string>
+</resources>
diff --git a/bitzenymininglibrary/src/test/java/com/example/ottylab/bitzenymininglibrary/ExampleUnitTest.java b/bitzenymininglibrary/src/test/java/com/example/ottylab/bitzenymininglibrary/ExampleUnitTest.java
new file mode 100644
index 0000000..0021344
--- /dev/null
+++ b/bitzenymininglibrary/src/test/java/com/example/ottylab/bitzenymininglibrary/ExampleUnitTest.java
@@ -0,0 +1,17 @@
+package com.example.ottylab.bitzenymininglibrary;
+
+import org.junit.Test;
+
+import static org.junit.Assert.*;
+
+/**
+ * Example local unit test, which will execute on the development machine (host).
+ *
+ * @see <a href="http://d.android.com/tools/testing">Testing documentation</a>
+ */
+public class ExampleUnitTest {
+    @Test
+    public void addition_isCorrect() throws Exception {
+        assertEquals(4, 2 + 2);
+    }
+}
\ No newline at end of file
diff --git a/build.gradle b/build.gradle
new file mode 100644
index 0000000..ac470f7
--- /dev/null
+++ b/build.gradle
@@ -0,0 +1,27 @@
+// Top-level build file where you can add configuration options common to all sub-projects/modules.
+
+buildscript {
+    
+    repositories {
+        google()
+        jcenter()
+    }
+    dependencies {
+        classpath 'com.android.tools.build:gradle:3.6.1'
+        
+
+        // NOTE: Do not place your application dependencies here; they belong
+        // in the individual module build.gradle files
+    }
+}
+
+allprojects {
+    repositories {
+        google()
+        jcenter()
+    }
+}
+
+task clean(type: Delete) {
+    delete rootProject.buildDir
+}
diff --git a/distribution/curl/include/curl/curl.h b/distribution/curl/include/curl/curl.h
new file mode 100644
index 0000000..9ad5c20
--- /dev/null
+++ b/distribution/curl/include/curl/curl.h
@@ -0,0 +1,2748 @@
+#ifndef __CURL_CURL_H
+#define __CURL_CURL_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+
+/*
+ * If you have libcurl problems, all docs and details are found here:
+ *   https://curl.haxx.se/libcurl/
+ *
+ * curl-library mailing list subscription and unsubscription web interface:
+ *   https://cool.haxx.se/mailman/listinfo/curl-library/
+ */
+
+#ifdef CURL_NO_OLDIES
+#define CURL_STRICTER
+#endif
+
+#include "curlver.h"         /* libcurl version defines   */
+#include "system.h"          /* determine things run-time */
+
+/*
+ * Define WIN32 when build target is Win32 API
+ */
+
+#if (defined(_WIN32) || defined(__WIN32__)) && \
+     !defined(WIN32) && !defined(__SYMBIAN32__)
+#define WIN32
+#endif
+
+#include <stdio.h>
+#include <limits.h>
+
+#if defined(__FreeBSD__) && (__FreeBSD__ >= 2)
+/* Needed for __FreeBSD_version symbol definition */
+#include <osreldate.h>
+#endif
+
+/* The include stuff here below is mainly for time_t! */
+#include <sys/types.h>
+#include <time.h>
+
+#if defined(WIN32) && !defined(_WIN32_WCE) && !defined(__CYGWIN__)
+#if !(defined(_WINSOCKAPI_) || defined(_WINSOCK_H) || \
+      defined(__LWIP_OPT_H__) || defined(LWIP_HDR_OPT_H))
+/* The check above prevents the winsock2 inclusion if winsock.h already was
+   included, since they can't co-exist without problems */
+#include <winsock2.h>
+#include <ws2tcpip.h>
+#endif
+#endif
+
+/* HP-UX systems version 9, 10 and 11 lack sys/select.h and so does oldish
+   libc5-based Linux systems. Only include it on systems that are known to
+   require it! */
+#if defined(_AIX) || defined(__NOVELL_LIBC__) || defined(__NetBSD__) || \
+    defined(__minix) || defined(__SYMBIAN32__) || defined(__INTEGRITY) || \
+    defined(ANDROID) || defined(__ANDROID__) || defined(__OpenBSD__) || \
+    defined(__CYGWIN__) || \
+   (defined(__FreeBSD_version) && (__FreeBSD_version < 800000))
+#include <sys/select.h>
+#endif
+
+#if !defined(WIN32) && !defined(_WIN32_WCE)
+#include <sys/socket.h>
+#endif
+
+#if !defined(WIN32) && !defined(__WATCOMC__) && !defined(__VXWORKS__)
+#include <sys/time.h>
+#endif
+
+#ifdef __BEOS__
+#include <support/SupportDefs.h>
+#endif
+
+#ifdef  __cplusplus
+extern "C" {
+#endif
+
+#if defined(BUILDING_LIBCURL) || defined(CURL_STRICTER)
+typedef struct Curl_easy CURL;
+typedef struct Curl_share CURLSH;
+#else
+typedef void CURL;
+typedef void CURLSH;
+#endif
+
+/*
+ * libcurl external API function linkage decorations.
+ */
+
+#ifdef CURL_STATICLIB
+#  define CURL_EXTERN
+#elif defined(WIN32) || defined(_WIN32) || defined(__SYMBIAN32__)
+#  if defined(BUILDING_LIBCURL)
+#    define CURL_EXTERN  __declspec(dllexport)
+#  else
+#    define CURL_EXTERN  __declspec(dllimport)
+#  endif
+#elif defined(BUILDING_LIBCURL) && defined(CURL_HIDDEN_SYMBOLS)
+#  define CURL_EXTERN CURL_EXTERN_SYMBOL
+#else
+#  define CURL_EXTERN
+#endif
+
+#ifndef curl_socket_typedef
+/* socket typedef */
+#if defined(WIN32) && !defined(__LWIP_OPT_H__) && !defined(LWIP_HDR_OPT_H)
+typedef SOCKET curl_socket_t;
+#define CURL_SOCKET_BAD INVALID_SOCKET
+#else
+typedef int curl_socket_t;
+#define CURL_SOCKET_BAD -1
+#endif
+#define curl_socket_typedef
+#endif /* curl_socket_typedef */
+
+/* enum for the different supported SSL backends */
+typedef enum {
+  CURLSSLBACKEND_NONE = 0,
+  CURLSSLBACKEND_OPENSSL = 1,
+  CURLSSLBACKEND_GNUTLS = 2,
+  CURLSSLBACKEND_NSS = 3,
+  CURLSSLBACKEND_OBSOLETE4 = 4,  /* Was QSOSSL. */
+  CURLSSLBACKEND_GSKIT = 5,
+  CURLSSLBACKEND_POLARSSL = 6,
+  CURLSSLBACKEND_WOLFSSL = 7,
+  CURLSSLBACKEND_SCHANNEL = 8,
+  CURLSSLBACKEND_DARWINSSL = 9,
+  CURLSSLBACKEND_AXTLS = 10,
+  CURLSSLBACKEND_MBEDTLS = 11
+} curl_sslbackend;
+
+/* aliases for library clones and renames */
+#define CURLSSLBACKEND_LIBRESSL CURLSSLBACKEND_OPENSSL
+#define CURLSSLBACKEND_BORINGSSL CURLSSLBACKEND_OPENSSL
+#define CURLSSLBACKEND_CYASSL CURLSSLBACKEND_WOLFSSL
+
+struct curl_httppost {
+  struct curl_httppost *next;       /* next entry in the list */
+  char *name;                       /* pointer to allocated name */
+  long namelength;                  /* length of name length */
+  char *contents;                   /* pointer to allocated data contents */
+  long contentslength;              /* length of contents field, see also
+                                       CURL_HTTPPOST_LARGE */
+  char *buffer;                     /* pointer to allocated buffer contents */
+  long bufferlength;                /* length of buffer field */
+  char *contenttype;                /* Content-Type */
+  struct curl_slist *contentheader; /* list of extra headers for this form */
+  struct curl_httppost *more;       /* if one field name has more than one
+                                       file, this link should link to following
+                                       files */
+  long flags;                       /* as defined below */
+
+/* specified content is a file name */
+#define CURL_HTTPPOST_FILENAME (1<<0)
+/* specified content is a file name */
+#define CURL_HTTPPOST_READFILE (1<<1)
+/* name is only stored pointer do not free in formfree */
+#define CURL_HTTPPOST_PTRNAME (1<<2)
+/* contents is only stored pointer do not free in formfree */
+#define CURL_HTTPPOST_PTRCONTENTS (1<<3)
+/* upload file from buffer */
+#define CURL_HTTPPOST_BUFFER (1<<4)
+/* upload file from pointer contents */
+#define CURL_HTTPPOST_PTRBUFFER (1<<5)
+/* upload file contents by using the regular read callback to get the data and
+   pass the given pointer as custom pointer */
+#define CURL_HTTPPOST_CALLBACK (1<<6)
+/* use size in 'contentlen', added in 7.46.0 */
+#define CURL_HTTPPOST_LARGE (1<<7)
+
+  char *showfilename;               /* The file name to show. If not set, the
+                                       actual file name will be used (if this
+                                       is a file part) */
+  void *userp;                      /* custom pointer used for
+                                       HTTPPOST_CALLBACK posts */
+  curl_off_t contentlen;            /* alternative length of contents
+                                       field. Used if CURL_HTTPPOST_LARGE is
+                                       set. Added in 7.46.0 */
+};
+
+/* This is the CURLOPT_PROGRESSFUNCTION callback proto. It is now considered
+   deprecated but was the only choice up until 7.31.0 */
+typedef int (*curl_progress_callback)(void *clientp,
+                                      double dltotal,
+                                      double dlnow,
+                                      double ultotal,
+                                      double ulnow);
+
+/* This is the CURLOPT_XFERINFOFUNCTION callback proto. It was introduced in
+   7.32.0, it avoids floating point and provides more detailed information. */
+typedef int (*curl_xferinfo_callback)(void *clientp,
+                                      curl_off_t dltotal,
+                                      curl_off_t dlnow,
+                                      curl_off_t ultotal,
+                                      curl_off_t ulnow);
+
+#ifndef CURL_MAX_READ_SIZE
+  /* The maximum receive buffer size configurable via CURLOPT_BUFFERSIZE. */
+#define CURL_MAX_READ_SIZE 524288
+#endif
+
+#ifndef CURL_MAX_WRITE_SIZE
+  /* Tests have proven that 20K is a very bad buffer size for uploads on
+     Windows, while 16K for some odd reason performed a lot better.
+     We do the ifndef check to allow this value to easier be changed at build
+     time for those who feel adventurous. The practical minimum is about
+     400 bytes since libcurl uses a buffer of this size as a scratch area
+     (unrelated to network send operations). */
+#define CURL_MAX_WRITE_SIZE 16384
+#endif
+
+#ifndef CURL_MAX_HTTP_HEADER
+/* The only reason to have a max limit for this is to avoid the risk of a bad
+   server feeding libcurl with a never-ending header that will cause reallocs
+   infinitely */
+#define CURL_MAX_HTTP_HEADER (100*1024)
+#endif
+
+/* This is a magic return code for the write callback that, when returned,
+   will signal libcurl to pause receiving on the current transfer. */
+#define CURL_WRITEFUNC_PAUSE 0x10000001
+
+typedef size_t (*curl_write_callback)(char *buffer,
+                                      size_t size,
+                                      size_t nitems,
+                                      void *outstream);
+
+
+
+/* enumeration of file types */
+typedef enum {
+  CURLFILETYPE_FILE = 0,
+  CURLFILETYPE_DIRECTORY,
+  CURLFILETYPE_SYMLINK,
+  CURLFILETYPE_DEVICE_BLOCK,
+  CURLFILETYPE_DEVICE_CHAR,
+  CURLFILETYPE_NAMEDPIPE,
+  CURLFILETYPE_SOCKET,
+  CURLFILETYPE_DOOR, /* is possible only on Sun Solaris now */
+
+  CURLFILETYPE_UNKNOWN /* should never occur */
+} curlfiletype;
+
+#define CURLFINFOFLAG_KNOWN_FILENAME    (1<<0)
+#define CURLFINFOFLAG_KNOWN_FILETYPE    (1<<1)
+#define CURLFINFOFLAG_KNOWN_TIME        (1<<2)
+#define CURLFINFOFLAG_KNOWN_PERM        (1<<3)
+#define CURLFINFOFLAG_KNOWN_UID         (1<<4)
+#define CURLFINFOFLAG_KNOWN_GID         (1<<5)
+#define CURLFINFOFLAG_KNOWN_SIZE        (1<<6)
+#define CURLFINFOFLAG_KNOWN_HLINKCOUNT  (1<<7)
+
+/* Content of this structure depends on information which is known and is
+   achievable (e.g. by FTP LIST parsing). Please see the url_easy_setopt(3) man
+   page for callbacks returning this structure -- some fields are mandatory,
+   some others are optional. The FLAG field has special meaning. */
+struct curl_fileinfo {
+  char *filename;
+  curlfiletype filetype;
+  time_t time;
+  unsigned int perm;
+  int uid;
+  int gid;
+  curl_off_t size;
+  long int hardlinks;
+
+  struct {
+    /* If some of these fields is not NULL, it is a pointer to b_data. */
+    char *time;
+    char *perm;
+    char *user;
+    char *group;
+    char *target; /* pointer to the target filename of a symlink */
+  } strings;
+
+  unsigned int flags;
+
+  /* used internally */
+  char *b_data;
+  size_t b_size;
+  size_t b_used;
+};
+
+/* return codes for CURLOPT_CHUNK_BGN_FUNCTION */
+#define CURL_CHUNK_BGN_FUNC_OK      0
+#define CURL_CHUNK_BGN_FUNC_FAIL    1 /* tell the lib to end the task */
+#define CURL_CHUNK_BGN_FUNC_SKIP    2 /* skip this chunk over */
+
+/* if splitting of data transfer is enabled, this callback is called before
+   download of an individual chunk started. Note that parameter "remains" works
+   only for FTP wildcard downloading (for now), otherwise is not used */
+typedef long (*curl_chunk_bgn_callback)(const void *transfer_info,
+                                        void *ptr,
+                                        int remains);
+
+/* return codes for CURLOPT_CHUNK_END_FUNCTION */
+#define CURL_CHUNK_END_FUNC_OK      0
+#define CURL_CHUNK_END_FUNC_FAIL    1 /* tell the lib to end the task */
+
+/* If splitting of data transfer is enabled this callback is called after
+   download of an individual chunk finished.
+   Note! After this callback was set then it have to be called FOR ALL chunks.
+   Even if downloading of this chunk was skipped in CHUNK_BGN_FUNC.
+   This is the reason why we don't need "transfer_info" parameter in this
+   callback and we are not interested in "remains" parameter too. */
+typedef long (*curl_chunk_end_callback)(void *ptr);
+
+/* return codes for FNMATCHFUNCTION */
+#define CURL_FNMATCHFUNC_MATCH    0 /* string corresponds to the pattern */
+#define CURL_FNMATCHFUNC_NOMATCH  1 /* pattern doesn't match the string */
+#define CURL_FNMATCHFUNC_FAIL     2 /* an error occurred */
+
+/* callback type for wildcard downloading pattern matching. If the
+   string matches the pattern, return CURL_FNMATCHFUNC_MATCH value, etc. */
+typedef int (*curl_fnmatch_callback)(void *ptr,
+                                     const char *pattern,
+                                     const char *string);
+
+/* These are the return codes for the seek callbacks */
+#define CURL_SEEKFUNC_OK       0
+#define CURL_SEEKFUNC_FAIL     1 /* fail the entire transfer */
+#define CURL_SEEKFUNC_CANTSEEK 2 /* tell libcurl seeking can't be done, so
+                                    libcurl might try other means instead */
+typedef int (*curl_seek_callback)(void *instream,
+                                  curl_off_t offset,
+                                  int origin); /* 'whence' */
+
+/* This is a return code for the read callback that, when returned, will
+   signal libcurl to immediately abort the current transfer. */
+#define CURL_READFUNC_ABORT 0x10000000
+/* This is a return code for the read callback that, when returned, will
+   signal libcurl to pause sending data on the current transfer. */
+#define CURL_READFUNC_PAUSE 0x10000001
+
+typedef size_t (*curl_read_callback)(char *buffer,
+                                      size_t size,
+                                      size_t nitems,
+                                      void *instream);
+
+typedef enum {
+  CURLSOCKTYPE_IPCXN,  /* socket created for a specific IP connection */
+  CURLSOCKTYPE_ACCEPT, /* socket created by accept() call */
+  CURLSOCKTYPE_LAST    /* never use */
+} curlsocktype;
+
+/* The return code from the sockopt_callback can signal information back
+   to libcurl: */
+#define CURL_SOCKOPT_OK 0
+#define CURL_SOCKOPT_ERROR 1 /* causes libcurl to abort and return
+                                CURLE_ABORTED_BY_CALLBACK */
+#define CURL_SOCKOPT_ALREADY_CONNECTED 2
+
+typedef int (*curl_sockopt_callback)(void *clientp,
+                                     curl_socket_t curlfd,
+                                     curlsocktype purpose);
+
+struct curl_sockaddr {
+  int family;
+  int socktype;
+  int protocol;
+  unsigned int addrlen; /* addrlen was a socklen_t type before 7.18.0 but it
+                           turned really ugly and painful on the systems that
+                           lack this type */
+  struct sockaddr addr;
+};
+
+typedef curl_socket_t
+(*curl_opensocket_callback)(void *clientp,
+                            curlsocktype purpose,
+                            struct curl_sockaddr *address);
+
+typedef int
+(*curl_closesocket_callback)(void *clientp, curl_socket_t item);
+
+typedef enum {
+  CURLIOE_OK,            /* I/O operation successful */
+  CURLIOE_UNKNOWNCMD,    /* command was unknown to callback */
+  CURLIOE_FAILRESTART,   /* failed to restart the read */
+  CURLIOE_LAST           /* never use */
+} curlioerr;
+
+typedef enum {
+  CURLIOCMD_NOP,         /* no operation */
+  CURLIOCMD_RESTARTREAD, /* restart the read stream from start */
+  CURLIOCMD_LAST         /* never use */
+} curliocmd;
+
+typedef curlioerr (*curl_ioctl_callback)(CURL *handle,
+                                         int cmd,
+                                         void *clientp);
+
+#ifndef CURL_DID_MEMORY_FUNC_TYPEDEFS
+/*
+ * The following typedef's are signatures of malloc, free, realloc, strdup and
+ * calloc respectively.  Function pointers of these types can be passed to the
+ * curl_global_init_mem() function to set user defined memory management
+ * callback routines.
+ */
+typedef void *(*curl_malloc_callback)(size_t size);
+typedef void (*curl_free_callback)(void *ptr);
+typedef void *(*curl_realloc_callback)(void *ptr, size_t size);
+typedef char *(*curl_strdup_callback)(const char *str);
+typedef void *(*curl_calloc_callback)(size_t nmemb, size_t size);
+
+#define CURL_DID_MEMORY_FUNC_TYPEDEFS
+#endif
+
+/* the kind of data that is passed to information_callback*/
+typedef enum {
+  CURLINFO_TEXT = 0,
+  CURLINFO_HEADER_IN,    /* 1 */
+  CURLINFO_HEADER_OUT,   /* 2 */
+  CURLINFO_DATA_IN,      /* 3 */
+  CURLINFO_DATA_OUT,     /* 4 */
+  CURLINFO_SSL_DATA_IN,  /* 5 */
+  CURLINFO_SSL_DATA_OUT, /* 6 */
+  CURLINFO_END
+} curl_infotype;
+
+typedef int (*curl_debug_callback)
+       (CURL *handle,      /* the handle/transfer this concerns */
+        curl_infotype type, /* what kind of data */
+        char *data,        /* points to the data */
+        size_t size,       /* size of the data pointed to */
+        void *userptr);    /* whatever the user please */
+
+/* All possible error codes from all sorts of curl functions. Future versions
+   may return other values, stay prepared.
+
+   Always add new return codes last. Never *EVER* remove any. The return
+   codes must remain the same!
+ */
+
+typedef enum {
+  CURLE_OK = 0,
+  CURLE_UNSUPPORTED_PROTOCOL,    /* 1 */
+  CURLE_FAILED_INIT,             /* 2 */
+  CURLE_URL_MALFORMAT,           /* 3 */
+  CURLE_NOT_BUILT_IN,            /* 4 - [was obsoleted in August 2007 for
+                                    7.17.0, reused in April 2011 for 7.21.5] */
+  CURLE_COULDNT_RESOLVE_PROXY,   /* 5 */
+  CURLE_COULDNT_RESOLVE_HOST,    /* 6 */
+  CURLE_COULDNT_CONNECT,         /* 7 */
+  CURLE_WEIRD_SERVER_REPLY,      /* 8 */
+  CURLE_REMOTE_ACCESS_DENIED,    /* 9 a service was denied by the server
+                                    due to lack of access - when login fails
+                                    this is not returned. */
+  CURLE_FTP_ACCEPT_FAILED,       /* 10 - [was obsoleted in April 2006 for
+                                    7.15.4, reused in Dec 2011 for 7.24.0]*/
+  CURLE_FTP_WEIRD_PASS_REPLY,    /* 11 */
+  CURLE_FTP_ACCEPT_TIMEOUT,      /* 12 - timeout occurred accepting server
+                                    [was obsoleted in August 2007 for 7.17.0,
+                                    reused in Dec 2011 for 7.24.0]*/
+  CURLE_FTP_WEIRD_PASV_REPLY,    /* 13 */
+  CURLE_FTP_WEIRD_227_FORMAT,    /* 14 */
+  CURLE_FTP_CANT_GET_HOST,       /* 15 */
+  CURLE_HTTP2,                   /* 16 - A problem in the http2 framing layer.
+                                    [was obsoleted in August 2007 for 7.17.0,
+                                    reused in July 2014 for 7.38.0] */
+  CURLE_FTP_COULDNT_SET_TYPE,    /* 17 */
+  CURLE_PARTIAL_FILE,            /* 18 */
+  CURLE_FTP_COULDNT_RETR_FILE,   /* 19 */
+  CURLE_OBSOLETE20,              /* 20 - NOT USED */
+  CURLE_QUOTE_ERROR,             /* 21 - quote command failure */
+  CURLE_HTTP_RETURNED_ERROR,     /* 22 */
+  CURLE_WRITE_ERROR,             /* 23 */
+  CURLE_OBSOLETE24,              /* 24 - NOT USED */
+  CURLE_UPLOAD_FAILED,           /* 25 - failed upload "command" */
+  CURLE_READ_ERROR,              /* 26 - couldn't open/read from file */
+  CURLE_OUT_OF_MEMORY,           /* 27 */
+  /* Note: CURLE_OUT_OF_MEMORY may sometimes indicate a conversion error
+           instead of a memory allocation error if CURL_DOES_CONVERSIONS
+           is defined
+  */
+  CURLE_OPERATION_TIMEDOUT,      /* 28 - the timeout time was reached */
+  CURLE_OBSOLETE29,              /* 29 - NOT USED */
+  CURLE_FTP_PORT_FAILED,         /* 30 - FTP PORT operation failed */
+  CURLE_FTP_COULDNT_USE_REST,    /* 31 - the REST command failed */
+  CURLE_OBSOLETE32,              /* 32 - NOT USED */
+  CURLE_RANGE_ERROR,             /* 33 - RANGE "command" didn't work */
+  CURLE_HTTP_POST_ERROR,         /* 34 */
+  CURLE_SSL_CONNECT_ERROR,       /* 35 - wrong when connecting with SSL */
+  CURLE_BAD_DOWNLOAD_RESUME,     /* 36 - couldn't resume download */
+  CURLE_FILE_COULDNT_READ_FILE,  /* 37 */
+  CURLE_LDAP_CANNOT_BIND,        /* 38 */
+  CURLE_LDAP_SEARCH_FAILED,      /* 39 */
+  CURLE_OBSOLETE40,              /* 40 - NOT USED */
+  CURLE_FUNCTION_NOT_FOUND,      /* 41 - NOT USED starting with 7.53.0 */
+  CURLE_ABORTED_BY_CALLBACK,     /* 42 */
+  CURLE_BAD_FUNCTION_ARGUMENT,   /* 43 */
+  CURLE_OBSOLETE44,              /* 44 - NOT USED */
+  CURLE_INTERFACE_FAILED,        /* 45 - CURLOPT_INTERFACE failed */
+  CURLE_OBSOLETE46,              /* 46 - NOT USED */
+  CURLE_TOO_MANY_REDIRECTS,      /* 47 - catch endless re-direct loops */
+  CURLE_UNKNOWN_OPTION,          /* 48 - User specified an unknown option */
+  CURLE_TELNET_OPTION_SYNTAX,    /* 49 - Malformed telnet option */
+  CURLE_OBSOLETE50,              /* 50 - NOT USED */
+  CURLE_PEER_FAILED_VERIFICATION, /* 51 - peer's certificate or fingerprint
+                                     wasn't verified fine */
+  CURLE_GOT_NOTHING,             /* 52 - when this is a specific error */
+  CURLE_SSL_ENGINE_NOTFOUND,     /* 53 - SSL crypto engine not found */
+  CURLE_SSL_ENGINE_SETFAILED,    /* 54 - can not set SSL crypto engine as
+                                    default */
+  CURLE_SEND_ERROR,              /* 55 - failed sending network data */
+  CURLE_RECV_ERROR,              /* 56 - failure in receiving network data */
+  CURLE_OBSOLETE57,              /* 57 - NOT IN USE */
+  CURLE_SSL_CERTPROBLEM,         /* 58 - problem with the local certificate */
+  CURLE_SSL_CIPHER,              /* 59 - couldn't use specified cipher */
+  CURLE_SSL_CACERT,              /* 60 - problem with the CA cert (path?) */
+  CURLE_BAD_CONTENT_ENCODING,    /* 61 - Unrecognized/bad encoding */
+  CURLE_LDAP_INVALID_URL,        /* 62 - Invalid LDAP URL */
+  CURLE_FILESIZE_EXCEEDED,       /* 63 - Maximum file size exceeded */
+  CURLE_USE_SSL_FAILED,          /* 64 - Requested FTP SSL level failed */
+  CURLE_SEND_FAIL_REWIND,        /* 65 - Sending the data requires a rewind
+                                    that failed */
+  CURLE_SSL_ENGINE_INITFAILED,   /* 66 - failed to initialise ENGINE */
+  CURLE_LOGIN_DENIED,            /* 67 - user, password or similar was not
+                                    accepted and we failed to login */
+  CURLE_TFTP_NOTFOUND,           /* 68 - file not found on server */
+  CURLE_TFTP_PERM,               /* 69 - permission problem on server */
+  CURLE_REMOTE_DISK_FULL,        /* 70 - out of disk space on server */
+  CURLE_TFTP_ILLEGAL,            /* 71 - Illegal TFTP operation */
+  CURLE_TFTP_UNKNOWNID,          /* 72 - Unknown transfer ID */
+  CURLE_REMOTE_FILE_EXISTS,      /* 73 - File already exists */
+  CURLE_TFTP_NOSUCHUSER,         /* 74 - No such user */
+  CURLE_CONV_FAILED,             /* 75 - conversion failed */
+  CURLE_CONV_REQD,               /* 76 - caller must register conversion
+                                    callbacks using curl_easy_setopt options
+                                    CURLOPT_CONV_FROM_NETWORK_FUNCTION,
+                                    CURLOPT_CONV_TO_NETWORK_FUNCTION, and
+                                    CURLOPT_CONV_FROM_UTF8_FUNCTION */
+  CURLE_SSL_CACERT_BADFILE,      /* 77 - could not load CACERT file, missing
+                                    or wrong format */
+  CURLE_REMOTE_FILE_NOT_FOUND,   /* 78 - remote file not found */
+  CURLE_SSH,                     /* 79 - error from the SSH layer, somewhat
+                                    generic so the error message will be of
+                                    interest when this has happened */
+
+  CURLE_SSL_SHUTDOWN_FAILED,     /* 80 - Failed to shut down the SSL
+                                    connection */
+  CURLE_AGAIN,                   /* 81 - socket is not ready for send/recv,
+                                    wait till it's ready and try again (Added
+                                    in 7.18.2) */
+  CURLE_SSL_CRL_BADFILE,         /* 82 - could not load CRL file, missing or
+                                    wrong format (Added in 7.19.0) */
+  CURLE_SSL_ISSUER_ERROR,        /* 83 - Issuer check failed.  (Added in
+                                    7.19.0) */
+  CURLE_FTP_PRET_FAILED,         /* 84 - a PRET command failed */
+  CURLE_RTSP_CSEQ_ERROR,         /* 85 - mismatch of RTSP CSeq numbers */
+  CURLE_RTSP_SESSION_ERROR,      /* 86 - mismatch of RTSP Session Ids */
+  CURLE_FTP_BAD_FILE_LIST,       /* 87 - unable to parse FTP file list */
+  CURLE_CHUNK_FAILED,            /* 88 - chunk callback reported error */
+  CURLE_NO_CONNECTION_AVAILABLE, /* 89 - No connection available, the
+                                    session will be queued */
+  CURLE_SSL_PINNEDPUBKEYNOTMATCH, /* 90 - specified pinned public key did not
+                                     match */
+  CURLE_SSL_INVALIDCERTSTATUS,   /* 91 - invalid certificate status */
+  CURLE_HTTP2_STREAM,            /* 92 - stream error in HTTP/2 framing layer
+                                    */
+  CURL_LAST /* never use! */
+} CURLcode;
+
+#ifndef CURL_NO_OLDIES /* define this to test if your app builds with all
+                          the obsolete stuff removed! */
+
+/* Previously obsolete error code re-used in 7.38.0 */
+#define CURLE_OBSOLETE16 CURLE_HTTP2
+
+/* Previously obsolete error codes re-used in 7.24.0 */
+#define CURLE_OBSOLETE10 CURLE_FTP_ACCEPT_FAILED
+#define CURLE_OBSOLETE12 CURLE_FTP_ACCEPT_TIMEOUT
+
+/*  compatibility with older names */
+#define CURLOPT_ENCODING CURLOPT_ACCEPT_ENCODING
+#define CURLE_FTP_WEIRD_SERVER_REPLY CURLE_WEIRD_SERVER_REPLY
+
+/* The following were added in 7.21.5, April 2011 */
+#define CURLE_UNKNOWN_TELNET_OPTION CURLE_UNKNOWN_OPTION
+
+/* The following were added in 7.17.1 */
+/* These are scheduled to disappear by 2009 */
+#define CURLE_SSL_PEER_CERTIFICATE CURLE_PEER_FAILED_VERIFICATION
+
+/* The following were added in 7.17.0 */
+/* These are scheduled to disappear by 2009 */
+#define CURLE_OBSOLETE CURLE_OBSOLETE50 /* no one should be using this! */
+#define CURLE_BAD_PASSWORD_ENTERED CURLE_OBSOLETE46
+#define CURLE_BAD_CALLING_ORDER CURLE_OBSOLETE44
+#define CURLE_FTP_USER_PASSWORD_INCORRECT CURLE_OBSOLETE10
+#define CURLE_FTP_CANT_RECONNECT CURLE_OBSOLETE16
+#define CURLE_FTP_COULDNT_GET_SIZE CURLE_OBSOLETE32
+#define CURLE_FTP_COULDNT_SET_ASCII CURLE_OBSOLETE29
+#define CURLE_FTP_WEIRD_USER_REPLY CURLE_OBSOLETE12
+#define CURLE_FTP_WRITE_ERROR CURLE_OBSOLETE20
+#define CURLE_LIBRARY_NOT_FOUND CURLE_OBSOLETE40
+#define CURLE_MALFORMAT_USER CURLE_OBSOLETE24
+#define CURLE_SHARE_IN_USE CURLE_OBSOLETE57
+#define CURLE_URL_MALFORMAT_USER CURLE_NOT_BUILT_IN
+
+#define CURLE_FTP_ACCESS_DENIED CURLE_REMOTE_ACCESS_DENIED
+#define CURLE_FTP_COULDNT_SET_BINARY CURLE_FTP_COULDNT_SET_TYPE
+#define CURLE_FTP_QUOTE_ERROR CURLE_QUOTE_ERROR
+#define CURLE_TFTP_DISKFULL CURLE_REMOTE_DISK_FULL
+#define CURLE_TFTP_EXISTS CURLE_REMOTE_FILE_EXISTS
+#define CURLE_HTTP_RANGE_ERROR CURLE_RANGE_ERROR
+#define CURLE_FTP_SSL_FAILED CURLE_USE_SSL_FAILED
+
+/* The following were added earlier */
+
+#define CURLE_OPERATION_TIMEOUTED CURLE_OPERATION_TIMEDOUT
+
+#define CURLE_HTTP_NOT_FOUND CURLE_HTTP_RETURNED_ERROR
+#define CURLE_HTTP_PORT_FAILED CURLE_INTERFACE_FAILED
+#define CURLE_FTP_COULDNT_STOR_FILE CURLE_UPLOAD_FAILED
+
+#define CURLE_FTP_PARTIAL_FILE CURLE_PARTIAL_FILE
+#define CURLE_FTP_BAD_DOWNLOAD_RESUME CURLE_BAD_DOWNLOAD_RESUME
+
+/* This was the error code 50 in 7.7.3 and a few earlier versions, this
+   is no longer used by libcurl but is instead #defined here only to not
+   make programs break */
+#define CURLE_ALREADY_COMPLETE 99999
+
+/* Provide defines for really old option names */
+#define CURLOPT_FILE CURLOPT_WRITEDATA /* name changed in 7.9.7 */
+#define CURLOPT_INFILE CURLOPT_READDATA /* name changed in 7.9.7 */
+#define CURLOPT_WRITEHEADER CURLOPT_HEADERDATA
+
+/* Since long deprecated options with no code in the lib that does anything
+   with them. */
+#define CURLOPT_WRITEINFO CURLOPT_OBSOLETE40
+#define CURLOPT_CLOSEPOLICY CURLOPT_OBSOLETE72
+
+#endif /*!CURL_NO_OLDIES*/
+
+/* This prototype applies to all conversion callbacks */
+typedef CURLcode (*curl_conv_callback)(char *buffer, size_t length);
+
+typedef CURLcode (*curl_ssl_ctx_callback)(CURL *curl,    /* easy handle */
+                                          void *ssl_ctx, /* actually an
+                                                            OpenSSL SSL_CTX */
+                                          void *userptr);
+
+typedef enum {
+  CURLPROXY_HTTP = 0,   /* added in 7.10, new in 7.19.4 default is to use
+                           CONNECT HTTP/1.1 */
+  CURLPROXY_HTTP_1_0 = 1,   /* added in 7.19.4, force to use CONNECT
+                               HTTP/1.0  */
+  CURLPROXY_HTTPS = 2, /* added in 7.52.0 */
+  CURLPROXY_SOCKS4 = 4, /* support added in 7.15.2, enum existed already
+                           in 7.10 */
+  CURLPROXY_SOCKS5 = 5, /* added in 7.10 */
+  CURLPROXY_SOCKS4A = 6, /* added in 7.18.0 */
+  CURLPROXY_SOCKS5_HOSTNAME = 7 /* Use the SOCKS5 protocol but pass along the
+                                   host name rather than the IP address. added
+                                   in 7.18.0 */
+} curl_proxytype;  /* this enum was added in 7.10 */
+
+/*
+ * Bitmasks for CURLOPT_HTTPAUTH and CURLOPT_PROXYAUTH options:
+ *
+ * CURLAUTH_NONE         - No HTTP authentication
+ * CURLAUTH_BASIC        - HTTP Basic authentication (default)
+ * CURLAUTH_DIGEST       - HTTP Digest authentication
+ * CURLAUTH_NEGOTIATE    - HTTP Negotiate (SPNEGO) authentication
+ * CURLAUTH_GSSNEGOTIATE - Alias for CURLAUTH_NEGOTIATE (deprecated)
+ * CURLAUTH_NTLM         - HTTP NTLM authentication
+ * CURLAUTH_DIGEST_IE    - HTTP Digest authentication with IE flavour
+ * CURLAUTH_NTLM_WB      - HTTP NTLM authentication delegated to winbind helper
+ * CURLAUTH_ONLY         - Use together with a single other type to force no
+ *                         authentication or just that single type
+ * CURLAUTH_ANY          - All fine types set
+ * CURLAUTH_ANYSAFE      - All fine types except Basic
+ */
+
+#define CURLAUTH_NONE         ((unsigned long)0)
+#define CURLAUTH_BASIC        (((unsigned long)1)<<0)
+#define CURLAUTH_DIGEST       (((unsigned long)1)<<1)
+#define CURLAUTH_NEGOTIATE    (((unsigned long)1)<<2)
+/* Deprecated since the advent of CURLAUTH_NEGOTIATE */
+#define CURLAUTH_GSSNEGOTIATE CURLAUTH_NEGOTIATE
+/* Used for CURLOPT_SOCKS5_AUTH to stay terminologically correct */
+#define CURLAUTH_GSSAPI CURLAUTH_NEGOTIATE
+#define CURLAUTH_NTLM         (((unsigned long)1)<<3)
+#define CURLAUTH_DIGEST_IE    (((unsigned long)1)<<4)
+#define CURLAUTH_NTLM_WB      (((unsigned long)1)<<5)
+#define CURLAUTH_ONLY         (((unsigned long)1)<<31)
+#define CURLAUTH_ANY          (~CURLAUTH_DIGEST_IE)
+#define CURLAUTH_ANYSAFE      (~(CURLAUTH_BASIC|CURLAUTH_DIGEST_IE))
+
+#define CURLSSH_AUTH_ANY       ~0     /* all types supported by the server */
+#define CURLSSH_AUTH_NONE      0      /* none allowed, silly but complete */
+#define CURLSSH_AUTH_PUBLICKEY (1<<0) /* public/private key files */
+#define CURLSSH_AUTH_PASSWORD  (1<<1) /* password */
+#define CURLSSH_AUTH_HOST      (1<<2) /* host key files */
+#define CURLSSH_AUTH_KEYBOARD  (1<<3) /* keyboard interactive */
+#define CURLSSH_AUTH_AGENT     (1<<4) /* agent (ssh-agent, pageant...) */
+#define CURLSSH_AUTH_DEFAULT CURLSSH_AUTH_ANY
+
+#define CURLGSSAPI_DELEGATION_NONE        0      /* no delegation (default) */
+#define CURLGSSAPI_DELEGATION_POLICY_FLAG (1<<0) /* if permitted by policy */
+#define CURLGSSAPI_DELEGATION_FLAG        (1<<1) /* delegate always */
+
+#define CURL_ERROR_SIZE 256
+
+enum curl_khtype {
+  CURLKHTYPE_UNKNOWN,
+  CURLKHTYPE_RSA1,
+  CURLKHTYPE_RSA,
+  CURLKHTYPE_DSS
+};
+
+struct curl_khkey {
+  const char *key; /* points to a zero-terminated string encoded with base64
+                      if len is zero, otherwise to the "raw" data */
+  size_t len;
+  enum curl_khtype keytype;
+};
+
+/* this is the set of return values expected from the curl_sshkeycallback
+   callback */
+enum curl_khstat {
+  CURLKHSTAT_FINE_ADD_TO_FILE,
+  CURLKHSTAT_FINE,
+  CURLKHSTAT_REJECT, /* reject the connection, return an error */
+  CURLKHSTAT_DEFER,  /* do not accept it, but we can't answer right now so
+                        this causes a CURLE_DEFER error but otherwise the
+                        connection will be left intact etc */
+  CURLKHSTAT_LAST    /* not for use, only a marker for last-in-list */
+};
+
+/* this is the set of status codes pass in to the callback */
+enum curl_khmatch {
+  CURLKHMATCH_OK,       /* match */
+  CURLKHMATCH_MISMATCH, /* host found, key mismatch! */
+  CURLKHMATCH_MISSING,  /* no matching host/key found */
+  CURLKHMATCH_LAST      /* not for use, only a marker for last-in-list */
+};
+
+typedef int
+  (*curl_sshkeycallback) (CURL *easy,     /* easy handle */
+                          const struct curl_khkey *knownkey, /* known */
+                          const struct curl_khkey *foundkey, /* found */
+                          enum curl_khmatch, /* libcurl's view on the keys */
+                          void *clientp); /* custom pointer passed from app */
+
+/* parameter for the CURLOPT_USE_SSL option */
+typedef enum {
+  CURLUSESSL_NONE,    /* do not attempt to use SSL */
+  CURLUSESSL_TRY,     /* try using SSL, proceed anyway otherwise */
+  CURLUSESSL_CONTROL, /* SSL for the control connection or fail */
+  CURLUSESSL_ALL,     /* SSL for all communication or fail */
+  CURLUSESSL_LAST     /* not an option, never use */
+} curl_usessl;
+
+/* Definition of bits for the CURLOPT_SSL_OPTIONS argument: */
+
+/* - ALLOW_BEAST tells libcurl to allow the BEAST SSL vulnerability in the
+   name of improving interoperability with older servers. Some SSL libraries
+   have introduced work-arounds for this flaw but those work-arounds sometimes
+   make the SSL communication fail. To regain functionality with those broken
+   servers, a user can this way allow the vulnerability back. */
+#define CURLSSLOPT_ALLOW_BEAST (1<<0)
+
+/* - NO_REVOKE tells libcurl to disable certificate revocation checks for those
+   SSL backends where such behavior is present. */
+#define CURLSSLOPT_NO_REVOKE (1<<1)
+
+#ifndef CURL_NO_OLDIES /* define this to test if your app builds with all
+                          the obsolete stuff removed! */
+
+/* Backwards compatibility with older names */
+/* These are scheduled to disappear by 2009 */
+
+#define CURLFTPSSL_NONE CURLUSESSL_NONE
+#define CURLFTPSSL_TRY CURLUSESSL_TRY
+#define CURLFTPSSL_CONTROL CURLUSESSL_CONTROL
+#define CURLFTPSSL_ALL CURLUSESSL_ALL
+#define CURLFTPSSL_LAST CURLUSESSL_LAST
+#define curl_ftpssl curl_usessl
+#endif /*!CURL_NO_OLDIES*/
+
+/* parameter for the CURLOPT_FTP_SSL_CCC option */
+typedef enum {
+  CURLFTPSSL_CCC_NONE,    /* do not send CCC */
+  CURLFTPSSL_CCC_PASSIVE, /* Let the server initiate the shutdown */
+  CURLFTPSSL_CCC_ACTIVE,  /* Initiate the shutdown */
+  CURLFTPSSL_CCC_LAST     /* not an option, never use */
+} curl_ftpccc;
+
+/* parameter for the CURLOPT_FTPSSLAUTH option */
+typedef enum {
+  CURLFTPAUTH_DEFAULT, /* let libcurl decide */
+  CURLFTPAUTH_SSL,     /* use "AUTH SSL" */
+  CURLFTPAUTH_TLS,     /* use "AUTH TLS" */
+  CURLFTPAUTH_LAST /* not an option, never use */
+} curl_ftpauth;
+
+/* parameter for the CURLOPT_FTP_CREATE_MISSING_DIRS option */
+typedef enum {
+  CURLFTP_CREATE_DIR_NONE,  /* do NOT create missing dirs! */
+  CURLFTP_CREATE_DIR,       /* (FTP/SFTP) if CWD fails, try MKD and then CWD
+                               again if MKD succeeded, for SFTP this does
+                               similar magic */
+  CURLFTP_CREATE_DIR_RETRY, /* (FTP only) if CWD fails, try MKD and then CWD
+                               again even if MKD failed! */
+  CURLFTP_CREATE_DIR_LAST   /* not an option, never use */
+} curl_ftpcreatedir;
+
+/* parameter for the CURLOPT_FTP_FILEMETHOD option */
+typedef enum {
+  CURLFTPMETHOD_DEFAULT,   /* let libcurl pick */
+  CURLFTPMETHOD_MULTICWD,  /* single CWD operation for each path part */
+  CURLFTPMETHOD_NOCWD,     /* no CWD at all */
+  CURLFTPMETHOD_SINGLECWD, /* one CWD to full dir, then work on file */
+  CURLFTPMETHOD_LAST       /* not an option, never use */
+} curl_ftpmethod;
+
+/* bitmask defines for CURLOPT_HEADEROPT */
+#define CURLHEADER_UNIFIED  0
+#define CURLHEADER_SEPARATE (1<<0)
+
+/* CURLPROTO_ defines are for the CURLOPT_*PROTOCOLS options */
+#define CURLPROTO_HTTP   (1<<0)
+#define CURLPROTO_HTTPS  (1<<1)
+#define CURLPROTO_FTP    (1<<2)
+#define CURLPROTO_FTPS   (1<<3)
+#define CURLPROTO_SCP    (1<<4)
+#define CURLPROTO_SFTP   (1<<5)
+#define CURLPROTO_TELNET (1<<6)
+#define CURLPROTO_LDAP   (1<<7)
+#define CURLPROTO_LDAPS  (1<<8)
+#define CURLPROTO_DICT   (1<<9)
+#define CURLPROTO_FILE   (1<<10)
+#define CURLPROTO_TFTP   (1<<11)
+#define CURLPROTO_IMAP   (1<<12)
+#define CURLPROTO_IMAPS  (1<<13)
+#define CURLPROTO_POP3   (1<<14)
+#define CURLPROTO_POP3S  (1<<15)
+#define CURLPROTO_SMTP   (1<<16)
+#define CURLPROTO_SMTPS  (1<<17)
+#define CURLPROTO_RTSP   (1<<18)
+#define CURLPROTO_RTMP   (1<<19)
+#define CURLPROTO_RTMPT  (1<<20)
+#define CURLPROTO_RTMPE  (1<<21)
+#define CURLPROTO_RTMPTE (1<<22)
+#define CURLPROTO_RTMPS  (1<<23)
+#define CURLPROTO_RTMPTS (1<<24)
+#define CURLPROTO_GOPHER (1<<25)
+#define CURLPROTO_SMB    (1<<26)
+#define CURLPROTO_SMBS   (1<<27)
+#define CURLPROTO_ALL    (~0) /* enable everything */
+
+/* long may be 32 or 64 bits, but we should never depend on anything else
+   but 32 */
+#define CURLOPTTYPE_LONG          0
+#define CURLOPTTYPE_OBJECTPOINT   10000
+#define CURLOPTTYPE_STRINGPOINT   10000
+#define CURLOPTTYPE_FUNCTIONPOINT 20000
+#define CURLOPTTYPE_OFF_T         30000
+
+/* *STRINGPOINT is an alias for OBJECTPOINT to allow tools to extract the
+   string options from the header file */
+
+/* name is uppercase CURLOPT_<name>,
+   type is one of the defined CURLOPTTYPE_<type>
+   number is unique identifier */
+#ifdef CINIT
+#undef CINIT
+#endif
+
+#ifdef CURL_ISOCPP
+#define CINIT(na,t,nu) CURLOPT_ ## na = CURLOPTTYPE_ ## t + nu
+#else
+/* The macro "##" is ISO C, we assume pre-ISO C doesn't support it. */
+#define LONG          CURLOPTTYPE_LONG
+#define OBJECTPOINT   CURLOPTTYPE_OBJECTPOINT
+#define STRINGPOINT   CURLOPTTYPE_OBJECTPOINT
+#define FUNCTIONPOINT CURLOPTTYPE_FUNCTIONPOINT
+#define OFF_T         CURLOPTTYPE_OFF_T
+#define CINIT(name,type,number) CURLOPT_/**/name = type + number
+#endif
+
+/*
+ * This macro-mania below setups the CURLOPT_[what] enum, to be used with
+ * curl_easy_setopt(). The first argument in the CINIT() macro is the [what]
+ * word.
+ */
+
+typedef enum {
+  /* This is the FILE * or void * the regular output should be written to. */
+  CINIT(WRITEDATA, OBJECTPOINT, 1),
+
+  /* The full URL to get/put */
+  CINIT(URL, STRINGPOINT, 2),
+
+  /* Port number to connect to, if other than default. */
+  CINIT(PORT, LONG, 3),
+
+  /* Name of proxy to use. */
+  CINIT(PROXY, STRINGPOINT, 4),
+
+  /* "user:password;options" to use when fetching. */
+  CINIT(USERPWD, STRINGPOINT, 5),
+
+  /* "user:password" to use with proxy. */
+  CINIT(PROXYUSERPWD, STRINGPOINT, 6),
+
+  /* Range to get, specified as an ASCII string. */
+  CINIT(RANGE, STRINGPOINT, 7),
+
+  /* not used */
+
+  /* Specified file stream to upload from (use as input): */
+  CINIT(READDATA, OBJECTPOINT, 9),
+
+  /* Buffer to receive error messages in, must be at least CURL_ERROR_SIZE
+   * bytes big. If this is not used, error messages go to stderr instead: */
+  CINIT(ERRORBUFFER, OBJECTPOINT, 10),
+
+  /* Function that will be called to store the output (instead of fwrite). The
+   * parameters will use fwrite() syntax, make sure to follow them. */
+  CINIT(WRITEFUNCTION, FUNCTIONPOINT, 11),
+
+  /* Function that will be called to read the input (instead of fread). The
+   * parameters will use fread() syntax, make sure to follow them. */
+  CINIT(READFUNCTION, FUNCTIONPOINT, 12),
+
+  /* Time-out the read operation after this amount of seconds */
+  CINIT(TIMEOUT, LONG, 13),
+
+  /* If the CURLOPT_INFILE is used, this can be used to inform libcurl about
+   * how large the file being sent really is. That allows better error
+   * checking and better verifies that the upload was successful. -1 means
+   * unknown size.
+   *
+   * For large file support, there is also a _LARGE version of the key
+   * which takes an off_t type, allowing platforms with larger off_t
+   * sizes to handle larger files.  See below for INFILESIZE_LARGE.
+   */
+  CINIT(INFILESIZE, LONG, 14),
+
+  /* POST static input fields. */
+  CINIT(POSTFIELDS, OBJECTPOINT, 15),
+
+  /* Set the referrer page (needed by some CGIs) */
+  CINIT(REFERER, STRINGPOINT, 16),
+
+  /* Set the FTP PORT string (interface name, named or numerical IP address)
+     Use i.e '-' to use default address. */
+  CINIT(FTPPORT, STRINGPOINT, 17),
+
+  /* Set the User-Agent string (examined by some CGIs) */
+  CINIT(USERAGENT, STRINGPOINT, 18),
+
+  /* If the download receives less than "low speed limit" bytes/second
+   * during "low speed time" seconds, the operations is aborted.
+   * You could i.e if you have a pretty high speed connection, abort if
+   * it is less than 2000 bytes/sec during 20 seconds.
+   */
+
+  /* Set the "low speed limit" */
+  CINIT(LOW_SPEED_LIMIT, LONG, 19),
+
+  /* Set the "low speed time" */
+  CINIT(LOW_SPEED_TIME, LONG, 20),
+
+  /* Set the continuation offset.
+   *
+   * Note there is also a _LARGE version of this key which uses
+   * off_t types, allowing for large file offsets on platforms which
+   * use larger-than-32-bit off_t's.  Look below for RESUME_FROM_LARGE.
+   */
+  CINIT(RESUME_FROM, LONG, 21),
+
+  /* Set cookie in request: */
+  CINIT(COOKIE, STRINGPOINT, 22),
+
+  /* This points to a linked list of headers, struct curl_slist kind. This
+     list is also used for RTSP (in spite of its name) */
+  CINIT(HTTPHEADER, OBJECTPOINT, 23),
+
+  /* This points to a linked list of post entries, struct curl_httppost */
+  CINIT(HTTPPOST, OBJECTPOINT, 24),
+
+  /* name of the file keeping your private SSL-certificate */
+  CINIT(SSLCERT, STRINGPOINT, 25),
+
+  /* password for the SSL or SSH private key */
+  CINIT(KEYPASSWD, STRINGPOINT, 26),
+
+  /* send TYPE parameter? */
+  CINIT(CRLF, LONG, 27),
+
+  /* send linked-list of QUOTE commands */
+  CINIT(QUOTE, OBJECTPOINT, 28),
+
+  /* send FILE * or void * to store headers to, if you use a callback it
+     is simply passed to the callback unmodified */
+  CINIT(HEADERDATA, OBJECTPOINT, 29),
+
+  /* point to a file to read the initial cookies from, also enables
+     "cookie awareness" */
+  CINIT(COOKIEFILE, STRINGPOINT, 31),
+
+  /* What version to specifically try to use.
+     See CURL_SSLVERSION defines below. */
+  CINIT(SSLVERSION, LONG, 32),
+
+  /* What kind of HTTP time condition to use, see defines */
+  CINIT(TIMECONDITION, LONG, 33),
+
+  /* Time to use with the above condition. Specified in number of seconds
+     since 1 Jan 1970 */
+  CINIT(TIMEVALUE, LONG, 34),
+
+  /* 35 = OBSOLETE */
+
+  /* Custom request, for customizing the get command like
+     HTTP: DELETE, TRACE and others
+     FTP: to use a different list command
+     */
+  CINIT(CUSTOMREQUEST, STRINGPOINT, 36),
+
+  /* FILE handle to use instead of stderr */
+  CINIT(STDERR, OBJECTPOINT, 37),
+
+  /* 38 is not used */
+
+  /* send linked-list of post-transfer QUOTE commands */
+  CINIT(POSTQUOTE, OBJECTPOINT, 39),
+
+  CINIT(OBSOLETE40, OBJECTPOINT, 40), /* OBSOLETE, do not use! */
+
+  CINIT(VERBOSE, LONG, 41),      /* talk a lot */
+  CINIT(HEADER, LONG, 42),       /* throw the header out too */
+  CINIT(NOPROGRESS, LONG, 43),   /* shut off the progress meter */
+  CINIT(NOBODY, LONG, 44),       /* use HEAD to get http document */
+  CINIT(FAILONERROR, LONG, 45),  /* no output on http error codes >= 400 */
+  CINIT(UPLOAD, LONG, 46),       /* this is an upload */
+  CINIT(POST, LONG, 47),         /* HTTP POST method */
+  CINIT(DIRLISTONLY, LONG, 48),  /* bare names when listing directories */
+
+  CINIT(APPEND, LONG, 50),       /* Append instead of overwrite on upload! */
+
+  /* Specify whether to read the user+password from the .netrc or the URL.
+   * This must be one of the CURL_NETRC_* enums below. */
+  CINIT(NETRC, LONG, 51),
+
+  CINIT(FOLLOWLOCATION, LONG, 52),  /* use Location: Luke! */
+
+  CINIT(TRANSFERTEXT, LONG, 53), /* transfer data in text/ASCII format */
+  CINIT(PUT, LONG, 54),          /* HTTP PUT */
+
+  /* 55 = OBSOLETE */
+
+  /* DEPRECATED
+   * Function that will be called instead of the internal progress display
+   * function. This function should be defined as the curl_progress_callback
+   * prototype defines. */
+  CINIT(PROGRESSFUNCTION, FUNCTIONPOINT, 56),
+
+  /* Data passed to the CURLOPT_PROGRESSFUNCTION and CURLOPT_XFERINFOFUNCTION
+     callbacks */
+  CINIT(PROGRESSDATA, OBJECTPOINT, 57),
+#define CURLOPT_XFERINFODATA CURLOPT_PROGRESSDATA
+
+  /* We want the referrer field set automatically when following locations */
+  CINIT(AUTOREFERER, LONG, 58),
+
+  /* Port of the proxy, can be set in the proxy string as well with:
+     "[host]:[port]" */
+  CINIT(PROXYPORT, LONG, 59),
+
+  /* size of the POST input data, if strlen() is not good to use */
+  CINIT(POSTFIELDSIZE, LONG, 60),
+
+  /* tunnel non-http operations through a HTTP proxy */
+  CINIT(HTTPPROXYTUNNEL, LONG, 61),
+
+  /* Set the interface string to use as outgoing network interface */
+  CINIT(INTERFACE, STRINGPOINT, 62),
+
+  /* Set the krb4/5 security level, this also enables krb4/5 awareness.  This
+   * is a string, 'clear', 'safe', 'confidential' or 'private'.  If the string
+   * is set but doesn't match one of these, 'private' will be used.  */
+  CINIT(KRBLEVEL, STRINGPOINT, 63),
+
+  /* Set if we should verify the peer in ssl handshake, set 1 to verify. */
+  CINIT(SSL_VERIFYPEER, LONG, 64),
+
+  /* The CApath or CAfile used to validate the peer certificate
+     this option is used only if SSL_VERIFYPEER is true */
+  CINIT(CAINFO, STRINGPOINT, 65),
+
+  /* 66 = OBSOLETE */
+  /* 67 = OBSOLETE */
+
+  /* Maximum number of http redirects to follow */
+  CINIT(MAXREDIRS, LONG, 68),
+
+  /* Pass a long set to 1 to get the date of the requested document (if
+     possible)! Pass a zero to shut it off. */
+  CINIT(FILETIME, LONG, 69),
+
+  /* This points to a linked list of telnet options */
+  CINIT(TELNETOPTIONS, OBJECTPOINT, 70),
+
+  /* Max amount of cached alive connections */
+  CINIT(MAXCONNECTS, LONG, 71),
+
+  CINIT(OBSOLETE72, LONG, 72), /* OBSOLETE, do not use! */
+
+  /* 73 = OBSOLETE */
+
+  /* Set to explicitly use a new connection for the upcoming transfer.
+     Do not use this unless you're absolutely sure of this, as it makes the
+     operation slower and is less friendly for the network. */
+  CINIT(FRESH_CONNECT, LONG, 74),
+
+  /* Set to explicitly forbid the upcoming transfer's connection to be re-used
+     when done. Do not use this unless you're absolutely sure of this, as it
+     makes the operation slower and is less friendly for the network. */
+  CINIT(FORBID_REUSE, LONG, 75),
+
+  /* Set to a file name that contains random data for libcurl to use to
+     seed the random engine when doing SSL connects. */
+  CINIT(RANDOM_FILE, STRINGPOINT, 76),
+
+  /* Set to the Entropy Gathering Daemon socket pathname */
+  CINIT(EGDSOCKET, STRINGPOINT, 77),
+
+  /* Time-out connect operations after this amount of seconds, if connects are
+     OK within this time, then fine... This only aborts the connect phase. */
+  CINIT(CONNECTTIMEOUT, LONG, 78),
+
+  /* Function that will be called to store headers (instead of fwrite). The
+   * parameters will use fwrite() syntax, make sure to follow them. */
+  CINIT(HEADERFUNCTION, FUNCTIONPOINT, 79),
+
+  /* Set this to force the HTTP request to get back to GET. Only really usable
+     if POST, PUT or a custom request have been used first.
+   */
+  CINIT(HTTPGET, LONG, 80),
+
+  /* Set if we should verify the Common name from the peer certificate in ssl
+   * handshake, set 1 to check existence, 2 to ensure that it matches the
+   * provided hostname. */
+  CINIT(SSL_VERIFYHOST, LONG, 81),
+
+  /* Specify which file name to write all known cookies in after completed
+     operation. Set file name to "-" (dash) to make it go to stdout. */
+  CINIT(COOKIEJAR, STRINGPOINT, 82),
+
+  /* Specify which SSL ciphers to use */
+  CINIT(SSL_CIPHER_LIST, STRINGPOINT, 83),
+
+  /* Specify which HTTP version to use! This must be set to one of the
+     CURL_HTTP_VERSION* enums set below. */
+  CINIT(HTTP_VERSION, LONG, 84),
+
+  /* Specifically switch on or off the FTP engine's use of the EPSV command. By
+     default, that one will always be attempted before the more traditional
+     PASV command. */
+  CINIT(FTP_USE_EPSV, LONG, 85),
+
+  /* type of the file keeping your SSL-certificate ("DER", "PEM", "ENG") */
+  CINIT(SSLCERTTYPE, STRINGPOINT, 86),
+
+  /* name of the file keeping your private SSL-key */
+  CINIT(SSLKEY, STRINGPOINT, 87),
+
+  /* type of the file keeping your private SSL-key ("DER", "PEM", "ENG") */
+  CINIT(SSLKEYTYPE, STRINGPOINT, 88),
+
+  /* crypto engine for the SSL-sub system */
+  CINIT(SSLENGINE, STRINGPOINT, 89),
+
+  /* set the crypto engine for the SSL-sub system as default
+     the param has no meaning...
+   */
+  CINIT(SSLENGINE_DEFAULT, LONG, 90),
+
+  /* Non-zero value means to use the global dns cache */
+  CINIT(DNS_USE_GLOBAL_CACHE, LONG, 91), /* DEPRECATED, do not use! */
+
+  /* DNS cache timeout */
+  CINIT(DNS_CACHE_TIMEOUT, LONG, 92),
+
+  /* send linked-list of pre-transfer QUOTE commands */
+  CINIT(PREQUOTE, OBJECTPOINT, 93),
+
+  /* set the debug function */
+  CINIT(DEBUGFUNCTION, FUNCTIONPOINT, 94),
+
+  /* set the data for the debug function */
+  CINIT(DEBUGDATA, OBJECTPOINT, 95),
+
+  /* mark this as start of a cookie session */
+  CINIT(COOKIESESSION, LONG, 96),
+
+  /* The CApath directory used to validate the peer certificate
+     this option is used only if SSL_VERIFYPEER is true */
+  CINIT(CAPATH, STRINGPOINT, 97),
+
+  /* Instruct libcurl to use a smaller receive buffer */
+  CINIT(BUFFERSIZE, LONG, 98),
+
+  /* Instruct libcurl to not use any signal/alarm handlers, even when using
+     timeouts. This option is useful for multi-threaded applications.
+     See libcurl-the-guide for more background information. */
+  CINIT(NOSIGNAL, LONG, 99),
+
+  /* Provide a CURLShare for mutexing non-ts data */
+  CINIT(SHARE, OBJECTPOINT, 100),
+
+  /* indicates type of proxy. accepted values are CURLPROXY_HTTP (default),
+     CURLPROXY_HTTPS, CURLPROXY_SOCKS4, CURLPROXY_SOCKS4A and
+     CURLPROXY_SOCKS5. */
+  CINIT(PROXYTYPE, LONG, 101),
+
+  /* Set the Accept-Encoding string. Use this to tell a server you would like
+     the response to be compressed. Before 7.21.6, this was known as
+     CURLOPT_ENCODING */
+  CINIT(ACCEPT_ENCODING, STRINGPOINT, 102),
+
+  /* Set pointer to private data */
+  CINIT(PRIVATE, OBJECTPOINT, 103),
+
+  /* Set aliases for HTTP 200 in the HTTP Response header */
+  CINIT(HTTP200ALIASES, OBJECTPOINT, 104),
+
+  /* Continue to send authentication (user+password) when following locations,
+     even when hostname changed. This can potentially send off the name
+     and password to whatever host the server decides. */
+  CINIT(UNRESTRICTED_AUTH, LONG, 105),
+
+  /* Specifically switch on or off the FTP engine's use of the EPRT command (
+     it also disables the LPRT attempt). By default, those ones will always be
+     attempted before the good old traditional PORT command. */
+  CINIT(FTP_USE_EPRT, LONG, 106),
+
+  /* Set this to a bitmask value to enable the particular authentications
+     methods you like. Use this in combination with CURLOPT_USERPWD.
+     Note that setting multiple bits may cause extra network round-trips. */
+  CINIT(HTTPAUTH, LONG, 107),
+
+  /* Set the ssl context callback function, currently only for OpenSSL ssl_ctx
+     in second argument. The function must be matching the
+     curl_ssl_ctx_callback proto. */
+  CINIT(SSL_CTX_FUNCTION, FUNCTIONPOINT, 108),
+
+  /* Set the userdata for the ssl context callback function's third
+     argument */
+  CINIT(SSL_CTX_DATA, OBJECTPOINT, 109),
+
+  /* FTP Option that causes missing dirs to be created on the remote server.
+     In 7.19.4 we introduced the convenience enums for this option using the
+     CURLFTP_CREATE_DIR prefix.
+  */
+  CINIT(FTP_CREATE_MISSING_DIRS, LONG, 110),
+
+  /* Set this to a bitmask value to enable the particular authentications
+     methods you like. Use this in combination with CURLOPT_PROXYUSERPWD.
+     Note that setting multiple bits may cause extra network round-trips. */
+  CINIT(PROXYAUTH, LONG, 111),
+
+  /* FTP option that changes the timeout, in seconds, associated with
+     getting a response.  This is different from transfer timeout time and
+     essentially places a demand on the FTP server to acknowledge commands
+     in a timely manner. */
+  CINIT(FTP_RESPONSE_TIMEOUT, LONG, 112),
+#define CURLOPT_SERVER_RESPONSE_TIMEOUT CURLOPT_FTP_RESPONSE_TIMEOUT
+
+  /* Set this option to one of the CURL_IPRESOLVE_* defines (see below) to
+     tell libcurl to resolve names to those IP versions only. This only has
+     affect on systems with support for more than one, i.e IPv4 _and_ IPv6. */
+  CINIT(IPRESOLVE, LONG, 113),
+
+  /* Set this option to limit the size of a file that will be downloaded from
+     an HTTP or FTP server.
+
+     Note there is also _LARGE version which adds large file support for
+     platforms which have larger off_t sizes.  See MAXFILESIZE_LARGE below. */
+  CINIT(MAXFILESIZE, LONG, 114),
+
+  /* See the comment for INFILESIZE above, but in short, specifies
+   * the size of the file being uploaded.  -1 means unknown.
+   */
+  CINIT(INFILESIZE_LARGE, OFF_T, 115),
+
+  /* Sets the continuation offset.  There is also a LONG version of this;
+   * look above for RESUME_FROM.
+   */
+  CINIT(RESUME_FROM_LARGE, OFF_T, 116),
+
+  /* Sets the maximum size of data that will be downloaded from
+   * an HTTP or FTP server.  See MAXFILESIZE above for the LONG version.
+   */
+  CINIT(MAXFILESIZE_LARGE, OFF_T, 117),
+
+  /* Set this option to the file name of your .netrc file you want libcurl
+     to parse (using the CURLOPT_NETRC option). If not set, libcurl will do
+     a poor attempt to find the user's home directory and check for a .netrc
+     file in there. */
+  CINIT(NETRC_FILE, STRINGPOINT, 118),
+
+  /* Enable SSL/TLS for FTP, pick one of:
+     CURLUSESSL_TRY     - try using SSL, proceed anyway otherwise
+     CURLUSESSL_CONTROL - SSL for the control connection or fail
+     CURLUSESSL_ALL     - SSL for all communication or fail
+  */
+  CINIT(USE_SSL, LONG, 119),
+
+  /* The _LARGE version of the standard POSTFIELDSIZE option */
+  CINIT(POSTFIELDSIZE_LARGE, OFF_T, 120),
+
+  /* Enable/disable the TCP Nagle algorithm */
+  CINIT(TCP_NODELAY, LONG, 121),
+
+  /* 122 OBSOLETE, used in 7.12.3. Gone in 7.13.0 */
+  /* 123 OBSOLETE. Gone in 7.16.0 */
+  /* 124 OBSOLETE, used in 7.12.3. Gone in 7.13.0 */
+  /* 125 OBSOLETE, used in 7.12.3. Gone in 7.13.0 */
+  /* 126 OBSOLETE, used in 7.12.3. Gone in 7.13.0 */
+  /* 127 OBSOLETE. Gone in 7.16.0 */
+  /* 128 OBSOLETE. Gone in 7.16.0 */
+
+  /* When FTP over SSL/TLS is selected (with CURLOPT_USE_SSL), this option
+     can be used to change libcurl's default action which is to first try
+     "AUTH SSL" and then "AUTH TLS" in this order, and proceed when a OK
+     response has been received.
+
+     Available parameters are:
+     CURLFTPAUTH_DEFAULT - let libcurl decide
+     CURLFTPAUTH_SSL     - try "AUTH SSL" first, then TLS
+     CURLFTPAUTH_TLS     - try "AUTH TLS" first, then SSL
+  */
+  CINIT(FTPSSLAUTH, LONG, 129),
+
+  CINIT(IOCTLFUNCTION, FUNCTIONPOINT, 130),
+  CINIT(IOCTLDATA, OBJECTPOINT, 131),
+
+  /* 132 OBSOLETE. Gone in 7.16.0 */
+  /* 133 OBSOLETE. Gone in 7.16.0 */
+
+  /* zero terminated string for pass on to the FTP server when asked for
+     "account" info */
+  CINIT(FTP_ACCOUNT, STRINGPOINT, 134),
+
+  /* feed cookie into cookie engine */
+  CINIT(COOKIELIST, STRINGPOINT, 135),
+
+  /* ignore Content-Length */
+  CINIT(IGNORE_CONTENT_LENGTH, LONG, 136),
+
+  /* Set to non-zero to skip the IP address received in a 227 PASV FTP server
+     response. Typically used for FTP-SSL purposes but is not restricted to
+     that. libcurl will then instead use the same IP address it used for the
+     control connection. */
+  CINIT(FTP_SKIP_PASV_IP, LONG, 137),
+
+  /* Select "file method" to use when doing FTP, see the curl_ftpmethod
+     above. */
+  CINIT(FTP_FILEMETHOD, LONG, 138),
+
+  /* Local port number to bind the socket to */
+  CINIT(LOCALPORT, LONG, 139),
+
+  /* Number of ports to try, including the first one set with LOCALPORT.
+     Thus, setting it to 1 will make no additional attempts but the first.
+  */
+  CINIT(LOCALPORTRANGE, LONG, 140),
+
+  /* no transfer, set up connection and let application use the socket by
+     extracting it with CURLINFO_LASTSOCKET */
+  CINIT(CONNECT_ONLY, LONG, 141),
+
+  /* Function that will be called to convert from the
+     network encoding (instead of using the iconv calls in libcurl) */
+  CINIT(CONV_FROM_NETWORK_FUNCTION, FUNCTIONPOINT, 142),
+
+  /* Function that will be called to convert to the
+     network encoding (instead of using the iconv calls in libcurl) */
+  CINIT(CONV_TO_NETWORK_FUNCTION, FUNCTIONPOINT, 143),
+
+  /* Function that will be called to convert from UTF8
+     (instead of using the iconv calls in libcurl)
+     Note that this is used only for SSL certificate processing */
+  CINIT(CONV_FROM_UTF8_FUNCTION, FUNCTIONPOINT, 144),
+
+  /* if the connection proceeds too quickly then need to slow it down */
+  /* limit-rate: maximum number of bytes per second to send or receive */
+  CINIT(MAX_SEND_SPEED_LARGE, OFF_T, 145),
+  CINIT(MAX_RECV_SPEED_LARGE, OFF_T, 146),
+
+  /* Pointer to command string to send if USER/PASS fails. */
+  CINIT(FTP_ALTERNATIVE_TO_USER, STRINGPOINT, 147),
+
+  /* callback function for setting socket options */
+  CINIT(SOCKOPTFUNCTION, FUNCTIONPOINT, 148),
+  CINIT(SOCKOPTDATA, OBJECTPOINT, 149),
+
+  /* set to 0 to disable session ID re-use for this transfer, default is
+     enabled (== 1) */
+  CINIT(SSL_SESSIONID_CACHE, LONG, 150),
+
+  /* allowed SSH authentication methods */
+  CINIT(SSH_AUTH_TYPES, LONG, 151),
+
+  /* Used by scp/sftp to do public/private key authentication */
+  CINIT(SSH_PUBLIC_KEYFILE, STRINGPOINT, 152),
+  CINIT(SSH_PRIVATE_KEYFILE, STRINGPOINT, 153),
+
+  /* Send CCC (Clear Command Channel) after authentication */
+  CINIT(FTP_SSL_CCC, LONG, 154),
+
+  /* Same as TIMEOUT and CONNECTTIMEOUT, but with ms resolution */
+  CINIT(TIMEOUT_MS, LONG, 155),
+  CINIT(CONNECTTIMEOUT_MS, LONG, 156),
+
+  /* set to zero to disable the libcurl's decoding and thus pass the raw body
+     data to the application even when it is encoded/compressed */
+  CINIT(HTTP_TRANSFER_DECODING, LONG, 157),
+  CINIT(HTTP_CONTENT_DECODING, LONG, 158),
+
+  /* Permission used when creating new files and directories on the remote
+     server for protocols that support it, SFTP/SCP/FILE */
+  CINIT(NEW_FILE_PERMS, LONG, 159),
+  CINIT(NEW_DIRECTORY_PERMS, LONG, 160),
+
+  /* Set the behaviour of POST when redirecting. Values must be set to one
+     of CURL_REDIR* defines below. This used to be called CURLOPT_POST301 */
+  CINIT(POSTREDIR, LONG, 161),
+
+  /* used by scp/sftp to verify the host's public key */
+  CINIT(SSH_HOST_PUBLIC_KEY_MD5, STRINGPOINT, 162),
+
+  /* Callback function for opening socket (instead of socket(2)). Optionally,
+     callback is able change the address or refuse to connect returning
+     CURL_SOCKET_BAD.  The callback should have type
+     curl_opensocket_callback */
+  CINIT(OPENSOCKETFUNCTION, FUNCTIONPOINT, 163),
+  CINIT(OPENSOCKETDATA, OBJECTPOINT, 164),
+
+  /* POST volatile input fields. */
+  CINIT(COPYPOSTFIELDS, OBJECTPOINT, 165),
+
+  /* set transfer mode (;type=<a|i>) when doing FTP via an HTTP proxy */
+  CINIT(PROXY_TRANSFER_MODE, LONG, 166),
+
+  /* Callback function for seeking in the input stream */
+  CINIT(SEEKFUNCTION, FUNCTIONPOINT, 167),
+  CINIT(SEEKDATA, OBJECTPOINT, 168),
+
+  /* CRL file */
+  CINIT(CRLFILE, STRINGPOINT, 169),
+
+  /* Issuer certificate */
+  CINIT(ISSUERCERT, STRINGPOINT, 170),
+
+  /* (IPv6) Address scope */
+  CINIT(ADDRESS_SCOPE, LONG, 171),
+
+  /* Collect certificate chain info and allow it to get retrievable with
+     CURLINFO_CERTINFO after the transfer is complete. */
+  CINIT(CERTINFO, LONG, 172),
+
+  /* "name" and "pwd" to use when fetching. */
+  CINIT(USERNAME, STRINGPOINT, 173),
+  CINIT(PASSWORD, STRINGPOINT, 174),
+
+    /* "name" and "pwd" to use with Proxy when fetching. */
+  CINIT(PROXYUSERNAME, STRINGPOINT, 175),
+  CINIT(PROXYPASSWORD, STRINGPOINT, 176),
+
+  /* Comma separated list of hostnames defining no-proxy zones. These should
+     match both hostnames directly, and hostnames within a domain. For
+     example, local.com will match local.com and www.local.com, but NOT
+     notlocal.com or www.notlocal.com. For compatibility with other
+     implementations of this, .local.com will be considered to be the same as
+     local.com. A single * is the only valid wildcard, and effectively
+     disables the use of proxy. */
+  CINIT(NOPROXY, STRINGPOINT, 177),
+
+  /* block size for TFTP transfers */
+  CINIT(TFTP_BLKSIZE, LONG, 178),
+
+  /* Socks Service */
+  CINIT(SOCKS5_GSSAPI_SERVICE, STRINGPOINT, 179), /* DEPRECATED, do not use! */
+
+  /* Socks Service */
+  CINIT(SOCKS5_GSSAPI_NEC, LONG, 180),
+
+  /* set the bitmask for the protocols that are allowed to be used for the
+     transfer, which thus helps the app which takes URLs from users or other
+     external inputs and want to restrict what protocol(s) to deal
+     with. Defaults to CURLPROTO_ALL. */
+  CINIT(PROTOCOLS, LONG, 181),
+
+  /* set the bitmask for the protocols that libcurl is allowed to follow to,
+     as a subset of the CURLOPT_PROTOCOLS ones. That means the protocol needs
+     to be set in both bitmasks to be allowed to get redirected to. Defaults
+     to all protocols except FILE and SCP. */
+  CINIT(REDIR_PROTOCOLS, LONG, 182),
+
+  /* set the SSH knownhost file name to use */
+  CINIT(SSH_KNOWNHOSTS, STRINGPOINT, 183),
+
+  /* set the SSH host key callback, must point to a curl_sshkeycallback
+     function */
+  CINIT(SSH_KEYFUNCTION, FUNCTIONPOINT, 184),
+
+  /* set the SSH host key callback custom pointer */
+  CINIT(SSH_KEYDATA, OBJECTPOINT, 185),
+
+  /* set the SMTP mail originator */
+  CINIT(MAIL_FROM, STRINGPOINT, 186),
+
+  /* set the list of SMTP mail receiver(s) */
+  CINIT(MAIL_RCPT, OBJECTPOINT, 187),
+
+  /* FTP: send PRET before PASV */
+  CINIT(FTP_USE_PRET, LONG, 188),
+
+  /* RTSP request method (OPTIONS, SETUP, PLAY, etc...) */
+  CINIT(RTSP_REQUEST, LONG, 189),
+
+  /* The RTSP session identifier */
+  CINIT(RTSP_SESSION_ID, STRINGPOINT, 190),
+
+  /* The RTSP stream URI */
+  CINIT(RTSP_STREAM_URI, STRINGPOINT, 191),
+
+  /* The Transport: header to use in RTSP requests */
+  CINIT(RTSP_TRANSPORT, STRINGPOINT, 192),
+
+  /* Manually initialize the client RTSP CSeq for this handle */
+  CINIT(RTSP_CLIENT_CSEQ, LONG, 193),
+
+  /* Manually initialize the server RTSP CSeq for this handle */
+  CINIT(RTSP_SERVER_CSEQ, LONG, 194),
+
+  /* The stream to pass to INTERLEAVEFUNCTION. */
+  CINIT(INTERLEAVEDATA, OBJECTPOINT, 195),
+
+  /* Let the application define a custom write method for RTP data */
+  CINIT(INTERLEAVEFUNCTION, FUNCTIONPOINT, 196),
+
+  /* Turn on wildcard matching */
+  CINIT(WILDCARDMATCH, LONG, 197),
+
+  /* Directory matching callback called before downloading of an
+     individual file (chunk) started */
+  CINIT(CHUNK_BGN_FUNCTION, FUNCTIONPOINT, 198),
+
+  /* Directory matching callback called after the file (chunk)
+     was downloaded, or skipped */
+  CINIT(CHUNK_END_FUNCTION, FUNCTIONPOINT, 199),
+
+  /* Change match (fnmatch-like) callback for wildcard matching */
+  CINIT(FNMATCH_FUNCTION, FUNCTIONPOINT, 200),
+
+  /* Let the application define custom chunk data pointer */
+  CINIT(CHUNK_DATA, OBJECTPOINT, 201),
+
+  /* FNMATCH_FUNCTION user pointer */
+  CINIT(FNMATCH_DATA, OBJECTPOINT, 202),
+
+  /* send linked-list of name:port:address sets */
+  CINIT(RESOLVE, OBJECTPOINT, 203),
+
+  /* Set a username for authenticated TLS */
+  CINIT(TLSAUTH_USERNAME, STRINGPOINT, 204),
+
+  /* Set a password for authenticated TLS */
+  CINIT(TLSAUTH_PASSWORD, STRINGPOINT, 205),
+
+  /* Set authentication type for authenticated TLS */
+  CINIT(TLSAUTH_TYPE, STRINGPOINT, 206),
+
+  /* Set to 1 to enable the "TE:" header in HTTP requests to ask for
+     compressed transfer-encoded responses. Set to 0 to disable the use of TE:
+     in outgoing requests. The current default is 0, but it might change in a
+     future libcurl release.
+
+     libcurl will ask for the compressed methods it knows of, and if that
+     isn't any, it will not ask for transfer-encoding at all even if this
+     option is set to 1.
+
+  */
+  CINIT(TRANSFER_ENCODING, LONG, 207),
+
+  /* Callback function for closing socket (instead of close(2)). The callback
+     should have type curl_closesocket_callback */
+  CINIT(CLOSESOCKETFUNCTION, FUNCTIONPOINT, 208),
+  CINIT(CLOSESOCKETDATA, OBJECTPOINT, 209),
+
+  /* allow GSSAPI credential delegation */
+  CINIT(GSSAPI_DELEGATION, LONG, 210),
+
+  /* Set the name servers to use for DNS resolution */
+  CINIT(DNS_SERVERS, STRINGPOINT, 211),
+
+  /* Time-out accept operations (currently for FTP only) after this amount
+     of milliseconds. */
+  CINIT(ACCEPTTIMEOUT_MS, LONG, 212),
+
+  /* Set TCP keepalive */
+  CINIT(TCP_KEEPALIVE, LONG, 213),
+
+  /* non-universal keepalive knobs (Linux, AIX, HP-UX, more) */
+  CINIT(TCP_KEEPIDLE, LONG, 214),
+  CINIT(TCP_KEEPINTVL, LONG, 215),
+
+  /* Enable/disable specific SSL features with a bitmask, see CURLSSLOPT_* */
+  CINIT(SSL_OPTIONS, LONG, 216),
+
+  /* Set the SMTP auth originator */
+  CINIT(MAIL_AUTH, STRINGPOINT, 217),
+
+  /* Enable/disable SASL initial response */
+  CINIT(SASL_IR, LONG, 218),
+
+  /* Function that will be called instead of the internal progress display
+   * function. This function should be defined as the curl_xferinfo_callback
+   * prototype defines. (Deprecates CURLOPT_PROGRESSFUNCTION) */
+  CINIT(XFERINFOFUNCTION, FUNCTIONPOINT, 219),
+
+  /* The XOAUTH2 bearer token */
+  CINIT(XOAUTH2_BEARER, STRINGPOINT, 220),
+
+  /* Set the interface string to use as outgoing network
+   * interface for DNS requests.
+   * Only supported by the c-ares DNS backend */
+  CINIT(DNS_INTERFACE, STRINGPOINT, 221),
+
+  /* Set the local IPv4 address to use for outgoing DNS requests.
+   * Only supported by the c-ares DNS backend */
+  CINIT(DNS_LOCAL_IP4, STRINGPOINT, 222),
+
+  /* Set the local IPv4 address to use for outgoing DNS requests.
+   * Only supported by the c-ares DNS backend */
+  CINIT(DNS_LOCAL_IP6, STRINGPOINT, 223),
+
+  /* Set authentication options directly */
+  CINIT(LOGIN_OPTIONS, STRINGPOINT, 224),
+
+  /* Enable/disable TLS NPN extension (http2 over ssl might fail without) */
+  CINIT(SSL_ENABLE_NPN, LONG, 225),
+
+  /* Enable/disable TLS ALPN extension (http2 over ssl might fail without) */
+  CINIT(SSL_ENABLE_ALPN, LONG, 226),
+
+  /* Time to wait for a response to a HTTP request containing an
+   * Expect: 100-continue header before sending the data anyway. */
+  CINIT(EXPECT_100_TIMEOUT_MS, LONG, 227),
+
+  /* This points to a linked list of headers used for proxy requests only,
+     struct curl_slist kind */
+  CINIT(PROXYHEADER, OBJECTPOINT, 228),
+
+  /* Pass in a bitmask of "header options" */
+  CINIT(HEADEROPT, LONG, 229),
+
+  /* The public key in DER form used to validate the peer public key
+     this option is used only if SSL_VERIFYPEER is true */
+  CINIT(PINNEDPUBLICKEY, STRINGPOINT, 230),
+
+  /* Path to Unix domain socket */
+  CINIT(UNIX_SOCKET_PATH, STRINGPOINT, 231),
+
+  /* Set if we should verify the certificate status. */
+  CINIT(SSL_VERIFYSTATUS, LONG, 232),
+
+  /* Set if we should enable TLS false start. */
+  CINIT(SSL_FALSESTART, LONG, 233),
+
+  /* Do not squash dot-dot sequences */
+  CINIT(PATH_AS_IS, LONG, 234),
+
+  /* Proxy Service Name */
+  CINIT(PROXY_SERVICE_NAME, STRINGPOINT, 235),
+
+  /* Service Name */
+  CINIT(SERVICE_NAME, STRINGPOINT, 236),
+
+  /* Wait/don't wait for pipe/mutex to clarify */
+  CINIT(PIPEWAIT, LONG, 237),
+
+  /* Set the protocol used when curl is given a URL without a protocol */
+  CINIT(DEFAULT_PROTOCOL, STRINGPOINT, 238),
+
+  /* Set stream weight, 1 - 256 (default is 16) */
+  CINIT(STREAM_WEIGHT, LONG, 239),
+
+  /* Set stream dependency on another CURL handle */
+  CINIT(STREAM_DEPENDS, OBJECTPOINT, 240),
+
+  /* Set E-xclusive stream dependency on another CURL handle */
+  CINIT(STREAM_DEPENDS_E, OBJECTPOINT, 241),
+
+  /* Do not send any tftp option requests to the server */
+  CINIT(TFTP_NO_OPTIONS, LONG, 242),
+
+  /* Linked-list of host:port:connect-to-host:connect-to-port,
+     overrides the URL's host:port (only for the network layer) */
+  CINIT(CONNECT_TO, OBJECTPOINT, 243),
+
+  /* Set TCP Fast Open */
+  CINIT(TCP_FASTOPEN, LONG, 244),
+
+  /* Continue to send data if the server responds early with an
+   * HTTP status code >= 300 */
+  CINIT(KEEP_SENDING_ON_ERROR, LONG, 245),
+
+  /* The CApath or CAfile used to validate the proxy certificate
+     this option is used only if PROXY_SSL_VERIFYPEER is true */
+  CINIT(PROXY_CAINFO, STRINGPOINT, 246),
+
+  /* The CApath directory used to validate the proxy certificate
+     this option is used only if PROXY_SSL_VERIFYPEER is true */
+  CINIT(PROXY_CAPATH, STRINGPOINT, 247),
+
+  /* Set if we should verify the proxy in ssl handshake,
+     set 1 to verify. */
+  CINIT(PROXY_SSL_VERIFYPEER, LONG, 248),
+
+  /* Set if we should verify the Common name from the proxy certificate in ssl
+   * handshake, set 1 to check existence, 2 to ensure that it matches
+   * the provided hostname. */
+  CINIT(PROXY_SSL_VERIFYHOST, LONG, 249),
+
+  /* What version to specifically try to use for proxy.
+     See CURL_SSLVERSION defines below. */
+  CINIT(PROXY_SSLVERSION, LONG, 250),
+
+  /* Set a username for authenticated TLS for proxy */
+  CINIT(PROXY_TLSAUTH_USERNAME, STRINGPOINT, 251),
+
+  /* Set a password for authenticated TLS for proxy */
+  CINIT(PROXY_TLSAUTH_PASSWORD, STRINGPOINT, 252),
+
+  /* Set authentication type for authenticated TLS for proxy */
+  CINIT(PROXY_TLSAUTH_TYPE, STRINGPOINT, 253),
+
+  /* name of the file keeping your private SSL-certificate for proxy */
+  CINIT(PROXY_SSLCERT, STRINGPOINT, 254),
+
+  /* type of the file keeping your SSL-certificate ("DER", "PEM", "ENG") for
+     proxy */
+  CINIT(PROXY_SSLCERTTYPE, STRINGPOINT, 255),
+
+  /* name of the file keeping your private SSL-key for proxy */
+  CINIT(PROXY_SSLKEY, STRINGPOINT, 256),
+
+  /* type of the file keeping your private SSL-key ("DER", "PEM", "ENG") for
+     proxy */
+  CINIT(PROXY_SSLKEYTYPE, STRINGPOINT, 257),
+
+  /* password for the SSL private key for proxy */
+  CINIT(PROXY_KEYPASSWD, STRINGPOINT, 258),
+
+  /* Specify which SSL ciphers to use for proxy */
+  CINIT(PROXY_SSL_CIPHER_LIST, STRINGPOINT, 259),
+
+  /* CRL file for proxy */
+  CINIT(PROXY_CRLFILE, STRINGPOINT, 260),
+
+  /* Enable/disable specific SSL features with a bitmask for proxy, see
+     CURLSSLOPT_* */
+  CINIT(PROXY_SSL_OPTIONS, LONG, 261),
+
+  /* Name of pre proxy to use. */
+  CINIT(PRE_PROXY, STRINGPOINT, 262),
+
+  /* The public key in DER form used to validate the proxy public key
+     this option is used only if PROXY_SSL_VERIFYPEER is true */
+  CINIT(PROXY_PINNEDPUBLICKEY, STRINGPOINT, 263),
+
+  /* Path to an abstract Unix domain socket */
+  CINIT(ABSTRACT_UNIX_SOCKET, STRINGPOINT, 264),
+
+  /* Suppress proxy CONNECT response headers from user callbacks */
+  CINIT(SUPPRESS_CONNECT_HEADERS, LONG, 265),
+
+  /* The request target, instead of extracted from the URL */
+  CINIT(REQUEST_TARGET, STRINGPOINT, 266),
+
+  /* bitmask of allowed auth methods for connections to SOCKS5 proxies */
+  CINIT(SOCKS5_AUTH, LONG, 267),
+
+  /* Enable/disable SSH compression */
+  CINIT(SSH_COMPRESSION, LONG, 268),
+
+  /* Post MIME data. */
+  CINIT(MIMEPOST, OBJECTPOINT, 269),
+
+  CURLOPT_LASTENTRY /* the last unused */
+} CURLoption;
+
+#ifndef CURL_NO_OLDIES /* define this to test if your app builds with all
+                          the obsolete stuff removed! */
+
+/* Backwards compatibility with older names */
+/* These are scheduled to disappear by 2011 */
+
+/* This was added in version 7.19.1 */
+#define CURLOPT_POST301 CURLOPT_POSTREDIR
+
+/* These are scheduled to disappear by 2009 */
+
+/* The following were added in 7.17.0 */
+#define CURLOPT_SSLKEYPASSWD CURLOPT_KEYPASSWD
+#define CURLOPT_FTPAPPEND CURLOPT_APPEND
+#define CURLOPT_FTPLISTONLY CURLOPT_DIRLISTONLY
+#define CURLOPT_FTP_SSL CURLOPT_USE_SSL
+
+/* The following were added earlier */
+
+#define CURLOPT_SSLCERTPASSWD CURLOPT_KEYPASSWD
+#define CURLOPT_KRB4LEVEL CURLOPT_KRBLEVEL
+
+#else
+/* This is set if CURL_NO_OLDIES is defined at compile-time */
+#undef CURLOPT_DNS_USE_GLOBAL_CACHE /* soon obsolete */
+#endif
+
+
+  /* Below here follows defines for the CURLOPT_IPRESOLVE option. If a host
+     name resolves addresses using more than one IP protocol version, this
+     option might be handy to force libcurl to use a specific IP version. */
+#define CURL_IPRESOLVE_WHATEVER 0 /* default, resolves addresses to all IP
+                                     versions that your system allows */
+#define CURL_IPRESOLVE_V4       1 /* resolve to IPv4 addresses */
+#define CURL_IPRESOLVE_V6       2 /* resolve to IPv6 addresses */
+
+  /* three convenient "aliases" that follow the name scheme better */
+#define CURLOPT_RTSPHEADER CURLOPT_HTTPHEADER
+
+  /* These enums are for use with the CURLOPT_HTTP_VERSION option. */
+enum {
+  CURL_HTTP_VERSION_NONE, /* setting this means we don't care, and that we'd
+                             like the library to choose the best possible
+                             for us! */
+  CURL_HTTP_VERSION_1_0,  /* please use HTTP 1.0 in the request */
+  CURL_HTTP_VERSION_1_1,  /* please use HTTP 1.1 in the request */
+  CURL_HTTP_VERSION_2_0,  /* please use HTTP 2 in the request */
+  CURL_HTTP_VERSION_2TLS, /* use version 2 for HTTPS, version 1.1 for HTTP */
+  CURL_HTTP_VERSION_2_PRIOR_KNOWLEDGE,  /* please use HTTP 2 without HTTP/1.1
+                                           Upgrade */
+
+  CURL_HTTP_VERSION_LAST /* *ILLEGAL* http version */
+};
+
+/* Convenience definition simple because the name of the version is HTTP/2 and
+   not 2.0. The 2_0 version of the enum name was set while the version was
+   still planned to be 2.0 and we stick to it for compatibility. */
+#define CURL_HTTP_VERSION_2 CURL_HTTP_VERSION_2_0
+
+/*
+ * Public API enums for RTSP requests
+ */
+enum {
+    CURL_RTSPREQ_NONE, /* first in list */
+    CURL_RTSPREQ_OPTIONS,
+    CURL_RTSPREQ_DESCRIBE,
+    CURL_RTSPREQ_ANNOUNCE,
+    CURL_RTSPREQ_SETUP,
+    CURL_RTSPREQ_PLAY,
+    CURL_RTSPREQ_PAUSE,
+    CURL_RTSPREQ_TEARDOWN,
+    CURL_RTSPREQ_GET_PARAMETER,
+    CURL_RTSPREQ_SET_PARAMETER,
+    CURL_RTSPREQ_RECORD,
+    CURL_RTSPREQ_RECEIVE,
+    CURL_RTSPREQ_LAST /* last in list */
+};
+
+  /* These enums are for use with the CURLOPT_NETRC option. */
+enum CURL_NETRC_OPTION {
+  CURL_NETRC_IGNORED,     /* The .netrc will never be read.
+                           * This is the default. */
+  CURL_NETRC_OPTIONAL,    /* A user:password in the URL will be preferred
+                           * to one in the .netrc. */
+  CURL_NETRC_REQUIRED,    /* A user:password in the URL will be ignored.
+                           * Unless one is set programmatically, the .netrc
+                           * will be queried. */
+  CURL_NETRC_LAST
+};
+
+enum {
+  CURL_SSLVERSION_DEFAULT,
+  CURL_SSLVERSION_TLSv1, /* TLS 1.x */
+  CURL_SSLVERSION_SSLv2,
+  CURL_SSLVERSION_SSLv3,
+  CURL_SSLVERSION_TLSv1_0,
+  CURL_SSLVERSION_TLSv1_1,
+  CURL_SSLVERSION_TLSv1_2,
+  CURL_SSLVERSION_TLSv1_3,
+
+  CURL_SSLVERSION_LAST /* never use, keep last */
+};
+
+enum {
+  CURL_SSLVERSION_MAX_NONE =     0,
+  CURL_SSLVERSION_MAX_DEFAULT =  (CURL_SSLVERSION_TLSv1   << 16),
+  CURL_SSLVERSION_MAX_TLSv1_0 =  (CURL_SSLVERSION_TLSv1_0 << 16),
+  CURL_SSLVERSION_MAX_TLSv1_1 =  (CURL_SSLVERSION_TLSv1_1 << 16),
+  CURL_SSLVERSION_MAX_TLSv1_2 =  (CURL_SSLVERSION_TLSv1_2 << 16),
+  CURL_SSLVERSION_MAX_TLSv1_3 =  (CURL_SSLVERSION_TLSv1_3 << 16),
+
+  /* never use, keep last */
+  CURL_SSLVERSION_MAX_LAST =     (CURL_SSLVERSION_LAST    << 16)
+};
+
+enum CURL_TLSAUTH {
+  CURL_TLSAUTH_NONE,
+  CURL_TLSAUTH_SRP,
+  CURL_TLSAUTH_LAST /* never use, keep last */
+};
+
+/* symbols to use with CURLOPT_POSTREDIR.
+   CURL_REDIR_POST_301, CURL_REDIR_POST_302 and CURL_REDIR_POST_303
+   can be bitwise ORed so that CURL_REDIR_POST_301 | CURL_REDIR_POST_302
+   | CURL_REDIR_POST_303 == CURL_REDIR_POST_ALL */
+
+#define CURL_REDIR_GET_ALL  0
+#define CURL_REDIR_POST_301 1
+#define CURL_REDIR_POST_302 2
+#define CURL_REDIR_POST_303 4
+#define CURL_REDIR_POST_ALL \
+    (CURL_REDIR_POST_301|CURL_REDIR_POST_302|CURL_REDIR_POST_303)
+
+typedef enum {
+  CURL_TIMECOND_NONE,
+
+  CURL_TIMECOND_IFMODSINCE,
+  CURL_TIMECOND_IFUNMODSINCE,
+  CURL_TIMECOND_LASTMOD,
+
+  CURL_TIMECOND_LAST
+} curl_TimeCond;
+
+/* Special size_t value signaling a zero-terminated string. */
+#define CURL_ZERO_TERMINATED ((size_t) -1)
+
+/* curl_strequal() and curl_strnequal() are subject for removal in a future
+   release */
+CURL_EXTERN int curl_strequal(const char *s1, const char *s2);
+CURL_EXTERN int curl_strnequal(const char *s1, const char *s2, size_t n);
+
+/* Mime/form handling support. */
+typedef struct curl_mime_s      curl_mime;      /* Mime context. */
+typedef struct curl_mimepart_s  curl_mimepart;  /* Mime part context. */
+
+/*
+ * NAME curl_mime_init()
+ *
+ * DESCRIPTION
+ *
+ * Create a mime context and return its handle. The easy parameter is the
+ * target handle.
+ */
+CURL_EXTERN curl_mime *curl_mime_init(CURL *easy);
+
+/*
+ * NAME curl_mime_free()
+ *
+ * DESCRIPTION
+ *
+ * release a mime handle and its substructures.
+ */
+CURL_EXTERN void curl_mime_free(curl_mime *mime);
+
+/*
+ * NAME curl_mime_addpart()
+ *
+ * DESCRIPTION
+ *
+ * Append a new empty part to the given mime context and return a handle to
+ * the created part.
+ */
+CURL_EXTERN curl_mimepart *curl_mime_addpart(curl_mime *mime);
+
+/*
+ * NAME curl_mime_name()
+ *
+ * DESCRIPTION
+ *
+ * Set mime/form part name.
+ */
+CURL_EXTERN CURLcode curl_mime_name(curl_mimepart *part, const char *name);
+
+/*
+ * NAME curl_mime_filename()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part remote file name.
+ */
+CURL_EXTERN CURLcode curl_mime_filename(curl_mimepart *part,
+                                        const char *filename);
+
+/*
+ * NAME curl_mime_type()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part type.
+ */
+CURL_EXTERN CURLcode curl_mime_type(curl_mimepart *part, const char *mimetype);
+
+/*
+ * NAME curl_mime_encoder()
+ *
+ * DESCRIPTION
+ *
+ * Set mime data transfer encoder.
+ */
+CURL_EXTERN CURLcode curl_mime_encoder(curl_mimepart *part,
+                                       const char *encoding);
+
+/*
+ * NAME curl_mime_data()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part data source from memory data,
+ */
+CURL_EXTERN CURLcode curl_mime_data(curl_mimepart *part,
+                                    const char *data, size_t datasize);
+
+/*
+ * NAME curl_mime_filedata()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part data source from named file.
+ */
+CURL_EXTERN CURLcode curl_mime_filedata(curl_mimepart *part,
+                                        const char *filename);
+
+/*
+ * NAME curl_mime_data_cb()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part data source from callback function.
+ */
+CURL_EXTERN CURLcode curl_mime_data_cb(curl_mimepart *part,
+                                       curl_off_t datasize,
+                                       curl_read_callback readfunc,
+                                       curl_seek_callback seekfunc,
+                                       curl_free_callback freefunc,
+                                       void *arg);
+
+/*
+ * NAME curl_mime_subparts()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part data source from subparts.
+ */
+CURL_EXTERN CURLcode curl_mime_subparts(curl_mimepart *part,
+                                        curl_mime *subparts);
+/*
+ * NAME curl_mime_headers()
+ *
+ * DESCRIPTION
+ *
+ * Set mime part headers.
+ */
+CURL_EXTERN CURLcode curl_mime_headers(curl_mimepart *part,
+                                       struct curl_slist *headers,
+                                       int take_ownership);
+
+/* Old form API. */
+/* name is uppercase CURLFORM_<name> */
+#ifdef CFINIT
+#undef CFINIT
+#endif
+
+#ifdef CURL_ISOCPP
+#define CFINIT(name) CURLFORM_ ## name
+#else
+/* The macro "##" is ISO C, we assume pre-ISO C doesn't support it. */
+#define CFINIT(name) CURLFORM_/**/name
+#endif
+
+typedef enum {
+  CFINIT(NOTHING),        /********* the first one is unused ************/
+
+  /*  */
+  CFINIT(COPYNAME),
+  CFINIT(PTRNAME),
+  CFINIT(NAMELENGTH),
+  CFINIT(COPYCONTENTS),
+  CFINIT(PTRCONTENTS),
+  CFINIT(CONTENTSLENGTH),
+  CFINIT(FILECONTENT),
+  CFINIT(ARRAY),
+  CFINIT(OBSOLETE),
+  CFINIT(FILE),
+
+  CFINIT(BUFFER),
+  CFINIT(BUFFERPTR),
+  CFINIT(BUFFERLENGTH),
+
+  CFINIT(CONTENTTYPE),
+  CFINIT(CONTENTHEADER),
+  CFINIT(FILENAME),
+  CFINIT(END),
+  CFINIT(OBSOLETE2),
+
+  CFINIT(STREAM),
+  CFINIT(CONTENTLEN), /* added in 7.46.0, provide a curl_off_t length */
+
+  CURLFORM_LASTENTRY /* the last unused */
+} CURLformoption;
+
+#undef CFINIT /* done */
+
+/* structure to be used as parameter for CURLFORM_ARRAY */
+struct curl_forms {
+  CURLformoption option;
+  const char     *value;
+};
+
+/* use this for multipart formpost building */
+/* Returns code for curl_formadd()
+ *
+ * Returns:
+ * CURL_FORMADD_OK             on success
+ * CURL_FORMADD_MEMORY         if the FormInfo allocation fails
+ * CURL_FORMADD_OPTION_TWICE   if one option is given twice for one Form
+ * CURL_FORMADD_NULL           if a null pointer was given for a char
+ * CURL_FORMADD_MEMORY         if the allocation of a FormInfo struct failed
+ * CURL_FORMADD_UNKNOWN_OPTION if an unknown option was used
+ * CURL_FORMADD_INCOMPLETE     if the some FormInfo is not complete (or error)
+ * CURL_FORMADD_MEMORY         if a curl_httppost struct cannot be allocated
+ * CURL_FORMADD_MEMORY         if some allocation for string copying failed.
+ * CURL_FORMADD_ILLEGAL_ARRAY  if an illegal option is used in an array
+ *
+ ***************************************************************************/
+typedef enum {
+  CURL_FORMADD_OK, /* first, no error */
+
+  CURL_FORMADD_MEMORY,
+  CURL_FORMADD_OPTION_TWICE,
+  CURL_FORMADD_NULL,
+  CURL_FORMADD_UNKNOWN_OPTION,
+  CURL_FORMADD_INCOMPLETE,
+  CURL_FORMADD_ILLEGAL_ARRAY,
+  CURL_FORMADD_DISABLED, /* libcurl was built with this disabled */
+
+  CURL_FORMADD_LAST /* last */
+} CURLFORMcode;
+
+/*
+ * NAME curl_formadd()
+ *
+ * DESCRIPTION
+ *
+ * Pretty advanced function for building multi-part formposts. Each invoke
+ * adds one part that together construct a full post. Then use
+ * CURLOPT_HTTPPOST to send it off to libcurl.
+ */
+CURL_EXTERN CURLFORMcode curl_formadd(struct curl_httppost **httppost,
+                                      struct curl_httppost **last_post,
+                                      ...);
+
+/*
+ * callback function for curl_formget()
+ * The void *arg pointer will be the one passed as second argument to
+ *   curl_formget().
+ * The character buffer passed to it must not be freed.
+ * Should return the buffer length passed to it as the argument "len" on
+ *   success.
+ */
+typedef size_t (*curl_formget_callback)(void *arg, const char *buf,
+                                        size_t len);
+
+/*
+ * NAME curl_formget()
+ *
+ * DESCRIPTION
+ *
+ * Serialize a curl_httppost struct built with curl_formadd().
+ * Accepts a void pointer as second argument which will be passed to
+ * the curl_formget_callback function.
+ * Returns 0 on success.
+ */
+CURL_EXTERN int curl_formget(struct curl_httppost *form, void *arg,
+                             curl_formget_callback append);
+/*
+ * NAME curl_formfree()
+ *
+ * DESCRIPTION
+ *
+ * Free a multipart formpost previously built with curl_formadd().
+ */
+CURL_EXTERN void curl_formfree(struct curl_httppost *form);
+
+/*
+ * NAME curl_getenv()
+ *
+ * DESCRIPTION
+ *
+ * Returns a malloc()'ed string that MUST be curl_free()ed after usage is
+ * complete. DEPRECATED - see lib/README.curlx
+ */
+CURL_EXTERN char *curl_getenv(const char *variable);
+
+/*
+ * NAME curl_version()
+ *
+ * DESCRIPTION
+ *
+ * Returns a static ascii string of the libcurl version.
+ */
+CURL_EXTERN char *curl_version(void);
+
+/*
+ * NAME curl_easy_escape()
+ *
+ * DESCRIPTION
+ *
+ * Escapes URL strings (converts all letters consider illegal in URLs to their
+ * %XX versions). This function returns a new allocated string or NULL if an
+ * error occurred.
+ */
+CURL_EXTERN char *curl_easy_escape(CURL *handle,
+                                   const char *string,
+                                   int length);
+
+/* the previous version: */
+CURL_EXTERN char *curl_escape(const char *string,
+                              int length);
+
+
+/*
+ * NAME curl_easy_unescape()
+ *
+ * DESCRIPTION
+ *
+ * Unescapes URL encoding in strings (converts all %XX codes to their 8bit
+ * versions). This function returns a new allocated string or NULL if an error
+ * occurred.
+ * Conversion Note: On non-ASCII platforms the ASCII %XX codes are
+ * converted into the host encoding.
+ */
+CURL_EXTERN char *curl_easy_unescape(CURL *handle,
+                                     const char *string,
+                                     int length,
+                                     int *outlength);
+
+/* the previous version */
+CURL_EXTERN char *curl_unescape(const char *string,
+                                int length);
+
+/*
+ * NAME curl_free()
+ *
+ * DESCRIPTION
+ *
+ * Provided for de-allocation in the same translation unit that did the
+ * allocation. Added in libcurl 7.10
+ */
+CURL_EXTERN void curl_free(void *p);
+
+/*
+ * NAME curl_global_init()
+ *
+ * DESCRIPTION
+ *
+ * curl_global_init() should be invoked exactly once for each application that
+ * uses libcurl and before any call of other libcurl functions.
+ *
+ * This function is not thread-safe!
+ */
+CURL_EXTERN CURLcode curl_global_init(long flags);
+
+/*
+ * NAME curl_global_init_mem()
+ *
+ * DESCRIPTION
+ *
+ * curl_global_init() or curl_global_init_mem() should be invoked exactly once
+ * for each application that uses libcurl.  This function can be used to
+ * initialize libcurl and set user defined memory management callback
+ * functions.  Users can implement memory management routines to check for
+ * memory leaks, check for mis-use of the curl library etc.  User registered
+ * callback routines with be invoked by this library instead of the system
+ * memory management routines like malloc, free etc.
+ */
+CURL_EXTERN CURLcode curl_global_init_mem(long flags,
+                                          curl_malloc_callback m,
+                                          curl_free_callback f,
+                                          curl_realloc_callback r,
+                                          curl_strdup_callback s,
+                                          curl_calloc_callback c);
+
+/*
+ * NAME curl_global_cleanup()
+ *
+ * DESCRIPTION
+ *
+ * curl_global_cleanup() should be invoked exactly once for each application
+ * that uses libcurl
+ */
+CURL_EXTERN void curl_global_cleanup(void);
+
+/* linked-list structure for the CURLOPT_QUOTE option (and other) */
+struct curl_slist {
+  char *data;
+  struct curl_slist *next;
+};
+
+/*
+ * NAME curl_global_sslset()
+ *
+ * DESCRIPTION
+ *
+ * When built with multiple SSL backends, curl_global_sslset() allows to
+ * choose one. This function can only be called once, and it must be called
+ * *before* curl_global_init().
+ *
+ * The backend can be identified by the id (e.g. CURLSSLBACKEND_OPENSSL). The
+ * backend can also be specified via the name parameter (passing -1 as id).
+ * If both id and name are specified, the name will be ignored. If neither id
+ * nor name are specified, the function will fail with
+ * CURLSSLSET_UNKNOWN_BACKEND and set the "avail" pointer to the
+ * NULL-terminated list of available backends.
+ *
+ * Upon success, the function returns CURLSSLSET_OK.
+ *
+ * If the specified SSL backend is not available, the function returns
+ * CURLSSLSET_UNKNOWN_BACKEND and sets the "avail" pointer to a NULL-terminated
+ * list of available SSL backends.
+ *
+ * The SSL backend can be set only once. If it has already been set, a
+ * subsequent attempt to change it will result in a CURLSSLSET_TOO_LATE.
+ */
+
+typedef struct {
+  curl_sslbackend id;
+  const char *name;
+} curl_ssl_backend;
+
+typedef enum {
+  CURLSSLSET_OK = 0,
+  CURLSSLSET_UNKNOWN_BACKEND,
+  CURLSSLSET_TOO_LATE,
+  CURLSSLSET_NO_BACKENDS /* libcurl was built without any SSL support */
+} CURLsslset;
+
+CURL_EXTERN CURLsslset curl_global_sslset(curl_sslbackend id, const char *name,
+                                          const curl_ssl_backend ***avail);
+
+/*
+ * NAME curl_slist_append()
+ *
+ * DESCRIPTION
+ *
+ * Appends a string to a linked list. If no list exists, it will be created
+ * first. Returns the new list, after appending.
+ */
+CURL_EXTERN struct curl_slist *curl_slist_append(struct curl_slist *,
+                                                 const char *);
+
+/*
+ * NAME curl_slist_free_all()
+ *
+ * DESCRIPTION
+ *
+ * free a previously built curl_slist.
+ */
+CURL_EXTERN void curl_slist_free_all(struct curl_slist *);
+
+/*
+ * NAME curl_getdate()
+ *
+ * DESCRIPTION
+ *
+ * Returns the time, in seconds since 1 Jan 1970 of the time string given in
+ * the first argument. The time argument in the second parameter is unused
+ * and should be set to NULL.
+ */
+CURL_EXTERN time_t curl_getdate(const char *p, const time_t *unused);
+
+/* info about the certificate chain, only for OpenSSL builds. Asked
+   for with CURLOPT_CERTINFO / CURLINFO_CERTINFO */
+struct curl_certinfo {
+  int num_of_certs;             /* number of certificates with information */
+  struct curl_slist **certinfo; /* for each index in this array, there's a
+                                   linked list with textual information in the
+                                   format "name: value" */
+};
+
+/* Information about the SSL library used and the respective internal SSL
+   handle, which can be used to obtain further information regarding the
+   connection. Asked for with CURLINFO_TLS_SSL_PTR or CURLINFO_TLS_SESSION. */
+struct curl_tlssessioninfo {
+  curl_sslbackend backend;
+  void *internals;
+};
+
+#define CURLINFO_STRING   0x100000
+#define CURLINFO_LONG     0x200000
+#define CURLINFO_DOUBLE   0x300000
+#define CURLINFO_SLIST    0x400000
+#define CURLINFO_PTR      0x400000 /* same as SLIST */
+#define CURLINFO_SOCKET   0x500000
+#define CURLINFO_OFF_T    0x600000
+#define CURLINFO_MASK     0x0fffff
+#define CURLINFO_TYPEMASK 0xf00000
+
+typedef enum {
+  CURLINFO_NONE, /* first, never use this */
+  CURLINFO_EFFECTIVE_URL    = CURLINFO_STRING + 1,
+  CURLINFO_RESPONSE_CODE    = CURLINFO_LONG   + 2,
+  CURLINFO_TOTAL_TIME       = CURLINFO_DOUBLE + 3,
+  CURLINFO_NAMELOOKUP_TIME  = CURLINFO_DOUBLE + 4,
+  CURLINFO_CONNECT_TIME     = CURLINFO_DOUBLE + 5,
+  CURLINFO_PRETRANSFER_TIME = CURLINFO_DOUBLE + 6,
+  CURLINFO_SIZE_UPLOAD      = CURLINFO_DOUBLE + 7,
+  CURLINFO_SIZE_UPLOAD_T    = CURLINFO_OFF_T  + 7,
+  CURLINFO_SIZE_DOWNLOAD    = CURLINFO_DOUBLE + 8,
+  CURLINFO_SIZE_DOWNLOAD_T  = CURLINFO_OFF_T  + 8,
+  CURLINFO_SPEED_DOWNLOAD   = CURLINFO_DOUBLE + 9,
+  CURLINFO_SPEED_DOWNLOAD_T = CURLINFO_OFF_T  + 9,
+  CURLINFO_SPEED_UPLOAD     = CURLINFO_DOUBLE + 10,
+  CURLINFO_SPEED_UPLOAD_T   = CURLINFO_OFF_T  + 10,
+  CURLINFO_HEADER_SIZE      = CURLINFO_LONG   + 11,
+  CURLINFO_REQUEST_SIZE     = CURLINFO_LONG   + 12,
+  CURLINFO_SSL_VERIFYRESULT = CURLINFO_LONG   + 13,
+  CURLINFO_FILETIME         = CURLINFO_LONG   + 14,
+  CURLINFO_CONTENT_LENGTH_DOWNLOAD   = CURLINFO_DOUBLE + 15,
+  CURLINFO_CONTENT_LENGTH_DOWNLOAD_T = CURLINFO_OFF_T  + 15,
+  CURLINFO_CONTENT_LENGTH_UPLOAD     = CURLINFO_DOUBLE + 16,
+  CURLINFO_CONTENT_LENGTH_UPLOAD_T   = CURLINFO_OFF_T  + 16,
+  CURLINFO_STARTTRANSFER_TIME = CURLINFO_DOUBLE + 17,
+  CURLINFO_CONTENT_TYPE     = CURLINFO_STRING + 18,
+  CURLINFO_REDIRECT_TIME    = CURLINFO_DOUBLE + 19,
+  CURLINFO_REDIRECT_COUNT   = CURLINFO_LONG   + 20,
+  CURLINFO_PRIVATE          = CURLINFO_STRING + 21,
+  CURLINFO_HTTP_CONNECTCODE = CURLINFO_LONG   + 22,
+  CURLINFO_HTTPAUTH_AVAIL   = CURLINFO_LONG   + 23,
+  CURLINFO_PROXYAUTH_AVAIL  = CURLINFO_LONG   + 24,
+  CURLINFO_OS_ERRNO         = CURLINFO_LONG   + 25,
+  CURLINFO_NUM_CONNECTS     = CURLINFO_LONG   + 26,
+  CURLINFO_SSL_ENGINES      = CURLINFO_SLIST  + 27,
+  CURLINFO_COOKIELIST       = CURLINFO_SLIST  + 28,
+  CURLINFO_LASTSOCKET       = CURLINFO_LONG   + 29,
+  CURLINFO_FTP_ENTRY_PATH   = CURLINFO_STRING + 30,
+  CURLINFO_REDIRECT_URL     = CURLINFO_STRING + 31,
+  CURLINFO_PRIMARY_IP       = CURLINFO_STRING + 32,
+  CURLINFO_APPCONNECT_TIME  = CURLINFO_DOUBLE + 33,
+  CURLINFO_CERTINFO         = CURLINFO_PTR    + 34,
+  CURLINFO_CONDITION_UNMET  = CURLINFO_LONG   + 35,
+  CURLINFO_RTSP_SESSION_ID  = CURLINFO_STRING + 36,
+  CURLINFO_RTSP_CLIENT_CSEQ = CURLINFO_LONG   + 37,
+  CURLINFO_RTSP_SERVER_CSEQ = CURLINFO_LONG   + 38,
+  CURLINFO_RTSP_CSEQ_RECV   = CURLINFO_LONG   + 39,
+  CURLINFO_PRIMARY_PORT     = CURLINFO_LONG   + 40,
+  CURLINFO_LOCAL_IP         = CURLINFO_STRING + 41,
+  CURLINFO_LOCAL_PORT       = CURLINFO_LONG   + 42,
+  CURLINFO_TLS_SESSION      = CURLINFO_PTR    + 43,
+  CURLINFO_ACTIVESOCKET     = CURLINFO_SOCKET + 44,
+  CURLINFO_TLS_SSL_PTR      = CURLINFO_PTR    + 45,
+  CURLINFO_HTTP_VERSION     = CURLINFO_LONG   + 46,
+  CURLINFO_PROXY_SSL_VERIFYRESULT = CURLINFO_LONG + 47,
+  CURLINFO_PROTOCOL         = CURLINFO_LONG   + 48,
+  CURLINFO_SCHEME           = CURLINFO_STRING + 49,
+  /* Fill in new entries below here! */
+
+  CURLINFO_LASTONE          = 49
+} CURLINFO;
+
+/* CURLINFO_RESPONSE_CODE is the new name for the option previously known as
+   CURLINFO_HTTP_CODE */
+#define CURLINFO_HTTP_CODE CURLINFO_RESPONSE_CODE
+
+typedef enum {
+  CURLCLOSEPOLICY_NONE, /* first, never use this */
+
+  CURLCLOSEPOLICY_OLDEST,
+  CURLCLOSEPOLICY_LEAST_RECENTLY_USED,
+  CURLCLOSEPOLICY_LEAST_TRAFFIC,
+  CURLCLOSEPOLICY_SLOWEST,
+  CURLCLOSEPOLICY_CALLBACK,
+
+  CURLCLOSEPOLICY_LAST /* last, never use this */
+} curl_closepolicy;
+
+#define CURL_GLOBAL_SSL (1<<0) /* no purpose since since 7.57.0 */
+#define CURL_GLOBAL_WIN32 (1<<1)
+#define CURL_GLOBAL_ALL (CURL_GLOBAL_SSL|CURL_GLOBAL_WIN32)
+#define CURL_GLOBAL_NOTHING 0
+#define CURL_GLOBAL_DEFAULT CURL_GLOBAL_ALL
+#define CURL_GLOBAL_ACK_EINTR (1<<2)
+
+
+/*****************************************************************************
+ * Setup defines, protos etc for the sharing stuff.
+ */
+
+/* Different data locks for a single share */
+typedef enum {
+  CURL_LOCK_DATA_NONE = 0,
+  /*  CURL_LOCK_DATA_SHARE is used internally to say that
+   *  the locking is just made to change the internal state of the share
+   *  itself.
+   */
+  CURL_LOCK_DATA_SHARE,
+  CURL_LOCK_DATA_COOKIE,
+  CURL_LOCK_DATA_DNS,
+  CURL_LOCK_DATA_SSL_SESSION,
+  CURL_LOCK_DATA_CONNECT,
+  CURL_LOCK_DATA_LAST
+} curl_lock_data;
+
+/* Different lock access types */
+typedef enum {
+  CURL_LOCK_ACCESS_NONE = 0,   /* unspecified action */
+  CURL_LOCK_ACCESS_SHARED = 1, /* for read perhaps */
+  CURL_LOCK_ACCESS_SINGLE = 2, /* for write perhaps */
+  CURL_LOCK_ACCESS_LAST        /* never use */
+} curl_lock_access;
+
+typedef void (*curl_lock_function)(CURL *handle,
+                                   curl_lock_data data,
+                                   curl_lock_access locktype,
+                                   void *userptr);
+typedef void (*curl_unlock_function)(CURL *handle,
+                                     curl_lock_data data,
+                                     void *userptr);
+
+
+typedef enum {
+  CURLSHE_OK,  /* all is fine */
+  CURLSHE_BAD_OPTION, /* 1 */
+  CURLSHE_IN_USE,     /* 2 */
+  CURLSHE_INVALID,    /* 3 */
+  CURLSHE_NOMEM,      /* 4 out of memory */
+  CURLSHE_NOT_BUILT_IN, /* 5 feature not present in lib */
+  CURLSHE_LAST        /* never use */
+} CURLSHcode;
+
+typedef enum {
+  CURLSHOPT_NONE,  /* don't use */
+  CURLSHOPT_SHARE,   /* specify a data type to share */
+  CURLSHOPT_UNSHARE, /* specify which data type to stop sharing */
+  CURLSHOPT_LOCKFUNC,   /* pass in a 'curl_lock_function' pointer */
+  CURLSHOPT_UNLOCKFUNC, /* pass in a 'curl_unlock_function' pointer */
+  CURLSHOPT_USERDATA,   /* pass in a user data pointer used in the lock/unlock
+                           callback functions */
+  CURLSHOPT_LAST  /* never use */
+} CURLSHoption;
+
+CURL_EXTERN CURLSH *curl_share_init(void);
+CURL_EXTERN CURLSHcode curl_share_setopt(CURLSH *, CURLSHoption option, ...);
+CURL_EXTERN CURLSHcode curl_share_cleanup(CURLSH *);
+
+/****************************************************************************
+ * Structures for querying information about the curl library at runtime.
+ */
+
+typedef enum {
+  CURLVERSION_FIRST,
+  CURLVERSION_SECOND,
+  CURLVERSION_THIRD,
+  CURLVERSION_FOURTH,
+  CURLVERSION_FIFTH,
+  CURLVERSION_LAST /* never actually use this */
+} CURLversion;
+
+/* The 'CURLVERSION_NOW' is the symbolic name meant to be used by
+   basically all programs ever that want to get version information. It is
+   meant to be a built-in version number for what kind of struct the caller
+   expects. If the struct ever changes, we redefine the NOW to another enum
+   from above. */
+#define CURLVERSION_NOW CURLVERSION_FIFTH
+
+typedef struct {
+  CURLversion age;          /* age of the returned struct */
+  const char *version;      /* LIBCURL_VERSION */
+  unsigned int version_num; /* LIBCURL_VERSION_NUM */
+  const char *host;         /* OS/host/cpu/machine when configured */
+  int features;             /* bitmask, see defines below */
+  const char *ssl_version;  /* human readable string */
+  long ssl_version_num;     /* not used anymore, always 0 */
+  const char *libz_version; /* human readable string */
+  /* protocols is terminated by an entry with a NULL protoname */
+  const char * const *protocols;
+
+  /* The fields below this were added in CURLVERSION_SECOND */
+  const char *ares;
+  int ares_num;
+
+  /* This field was added in CURLVERSION_THIRD */
+  const char *libidn;
+
+  /* These field were added in CURLVERSION_FOURTH */
+
+  /* Same as '_libiconv_version' if built with HAVE_ICONV */
+  int iconv_ver_num;
+
+  const char *libssh_version; /* human readable string */
+
+  /* These fields were added in CURLVERSION_FIFTH */
+
+  unsigned int brotli_ver_num; /* Numeric Brotli version
+                                  (MAJOR << 24) | (MINOR << 12) | PATCH */
+  const char *brotli_version; /* human readable string. */
+
+} curl_version_info_data;
+
+#define CURL_VERSION_IPV6         (1<<0)  /* IPv6-enabled */
+#define CURL_VERSION_KERBEROS4    (1<<1)  /* Kerberos V4 auth is supported
+                                             (deprecated) */
+#define CURL_VERSION_SSL          (1<<2)  /* SSL options are present */
+#define CURL_VERSION_LIBZ         (1<<3)  /* libz features are present */
+#define CURL_VERSION_NTLM         (1<<4)  /* NTLM auth is supported */
+#define CURL_VERSION_GSSNEGOTIATE (1<<5)  /* Negotiate auth is supported
+                                             (deprecated) */
+#define CURL_VERSION_DEBUG        (1<<6)  /* Built with debug capabilities */
+#define CURL_VERSION_ASYNCHDNS    (1<<7)  /* Asynchronous DNS resolves */
+#define CURL_VERSION_SPNEGO       (1<<8)  /* SPNEGO auth is supported */
+#define CURL_VERSION_LARGEFILE    (1<<9)  /* Supports files larger than 2GB */
+#define CURL_VERSION_IDN          (1<<10) /* Internationized Domain Names are
+                                             supported */
+#define CURL_VERSION_SSPI         (1<<11) /* Built against Windows SSPI */
+#define CURL_VERSION_CONV         (1<<12) /* Character conversions supported */
+#define CURL_VERSION_CURLDEBUG    (1<<13) /* Debug memory tracking supported */
+#define CURL_VERSION_TLSAUTH_SRP  (1<<14) /* TLS-SRP auth is supported */
+#define CURL_VERSION_NTLM_WB      (1<<15) /* NTLM delegation to winbind helper
+                                             is supported */
+#define CURL_VERSION_HTTP2        (1<<16) /* HTTP2 support built-in */
+#define CURL_VERSION_GSSAPI       (1<<17) /* Built against a GSS-API library */
+#define CURL_VERSION_KERBEROS5    (1<<18) /* Kerberos V5 auth is supported */
+#define CURL_VERSION_UNIX_SOCKETS (1<<19) /* Unix domain sockets support */
+#define CURL_VERSION_PSL          (1<<20) /* Mozilla's Public Suffix List, used
+                                             for cookie domain verification */
+#define CURL_VERSION_HTTPS_PROXY  (1<<21) /* HTTPS-proxy support built-in */
+#define CURL_VERSION_MULTI_SSL    (1<<22) /* Multiple SSL backends available */
+#define CURL_VERSION_BROTLI       (1<<23) /* Brotli features are present. */
+
+ /*
+ * NAME curl_version_info()
+ *
+ * DESCRIPTION
+ *
+ * This function returns a pointer to a static copy of the version info
+ * struct. See above.
+ */
+CURL_EXTERN curl_version_info_data *curl_version_info(CURLversion);
+
+/*
+ * NAME curl_easy_strerror()
+ *
+ * DESCRIPTION
+ *
+ * The curl_easy_strerror function may be used to turn a CURLcode value
+ * into the equivalent human readable error string.  This is useful
+ * for printing meaningful error messages.
+ */
+CURL_EXTERN const char *curl_easy_strerror(CURLcode);
+
+/*
+ * NAME curl_share_strerror()
+ *
+ * DESCRIPTION
+ *
+ * The curl_share_strerror function may be used to turn a CURLSHcode value
+ * into the equivalent human readable error string.  This is useful
+ * for printing meaningful error messages.
+ */
+CURL_EXTERN const char *curl_share_strerror(CURLSHcode);
+
+/*
+ * NAME curl_easy_pause()
+ *
+ * DESCRIPTION
+ *
+ * The curl_easy_pause function pauses or unpauses transfers. Select the new
+ * state by setting the bitmask, use the convenience defines below.
+ *
+ */
+CURL_EXTERN CURLcode curl_easy_pause(CURL *handle, int bitmask);
+
+#define CURLPAUSE_RECV      (1<<0)
+#define CURLPAUSE_RECV_CONT (0)
+
+#define CURLPAUSE_SEND      (1<<2)
+#define CURLPAUSE_SEND_CONT (0)
+
+#define CURLPAUSE_ALL       (CURLPAUSE_RECV|CURLPAUSE_SEND)
+#define CURLPAUSE_CONT      (CURLPAUSE_RECV_CONT|CURLPAUSE_SEND_CONT)
+
+#ifdef  __cplusplus
+}
+#endif
+
+/* unfortunately, the easy.h and multi.h include files need options and info
+  stuff before they can be included! */
+#include "easy.h" /* nothing in curl is fun without the easy stuff */
+#include "multi.h"
+
+/* the typechecker doesn't work in C++ (yet) */
+#if defined(__GNUC__) && defined(__GNUC_MINOR__) && \
+    ((__GNUC__ > 4) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3)) && \
+    !defined(__cplusplus) && !defined(CURL_DISABLE_TYPECHECK)
+#include "typecheck-gcc.h"
+#else
+#if defined(__STDC__) && (__STDC__ >= 1)
+/* This preprocessor magic that replaces a call with the exact same call is
+   only done to make sure application authors pass exactly three arguments
+   to these functions. */
+#define curl_easy_setopt(handle,opt,param) curl_easy_setopt(handle,opt,param)
+#define curl_easy_getinfo(handle,info,arg) curl_easy_getinfo(handle,info,arg)
+#define curl_share_setopt(share,opt,param) curl_share_setopt(share,opt,param)
+#define curl_multi_setopt(handle,opt,param) curl_multi_setopt(handle,opt,param)
+#endif /* __STDC__ >= 1 */
+#endif /* gcc >= 4.3 && !__cplusplus */
+
+#endif /* __CURL_CURL_H */
diff --git a/distribution/curl/include/curl/curlver.h b/distribution/curl/include/curl/curlver.h
new file mode 100644
index 0000000..30b1463
--- /dev/null
+++ b/distribution/curl/include/curl/curlver.h
@@ -0,0 +1,77 @@
+#ifndef __CURL_CURLVER_H
+#define __CURL_CURLVER_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+
+/* This header file contains nothing but libcurl version info, generated by
+   a script at release-time. This was made its own header file in 7.11.2 */
+
+/* This is the global package copyright */
+#define LIBCURL_COPYRIGHT "1996 - 2017 Daniel Stenberg, <daniel@haxx.se>."
+
+/* This is the version number of the libcurl package from which this header
+   file origins: */
+#define LIBCURL_VERSION "7.57.0"
+
+/* The numeric version number is also available "in parts" by using these
+   defines: */
+#define LIBCURL_VERSION_MAJOR 7
+#define LIBCURL_VERSION_MINOR 57
+#define LIBCURL_VERSION_PATCH 0
+
+/* This is the numeric version of the libcurl version number, meant for easier
+   parsing and comparions by programs. The LIBCURL_VERSION_NUM define will
+   always follow this syntax:
+
+         0xXXYYZZ
+
+   Where XX, YY and ZZ are the main version, release and patch numbers in
+   hexadecimal (using 8 bits each). All three numbers are always represented
+   using two digits.  1.2 would appear as "0x010200" while version 9.11.7
+   appears as "0x090b07".
+
+   This 6-digit (24 bits) hexadecimal number does not show pre-release number,
+   and it is always a greater number in a more recent release. It makes
+   comparisons with greater than and less than work.
+
+   Note: This define is the full hex number and _does not_ use the
+   CURL_VERSION_BITS() macro since curl's own configure script greps for it
+   and needs it to contain the full number.
+*/
+#define LIBCURL_VERSION_NUM 0x073900
+
+/*
+ * This is the date and time when the full source package was created. The
+ * timestamp is not stored in git, as the timestamp is properly set in the
+ * tarballs by the maketgz script.
+ *
+ * The format of the date follows this template:
+ *
+ * "2007-11-23"
+ */
+#define LIBCURL_TIMESTAMP "2017-11-29"
+
+#define CURL_VERSION_BITS(x,y,z) ((x)<<16|(y)<<8|z)
+#define CURL_AT_LEAST_VERSION(x,y,z) \
+  (LIBCURL_VERSION_NUM >= CURL_VERSION_BITS(x, y, z))
+
+#endif /* __CURL_CURLVER_H */
diff --git a/distribution/curl/include/curl/easy.h b/distribution/curl/include/curl/easy.h
new file mode 100644
index 0000000..752c504
--- /dev/null
+++ b/distribution/curl/include/curl/easy.h
@@ -0,0 +1,102 @@
+#ifndef __CURL_EASY_H
+#define __CURL_EASY_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2016, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+#ifdef  __cplusplus
+extern "C" {
+#endif
+
+CURL_EXTERN CURL *curl_easy_init(void);
+CURL_EXTERN CURLcode curl_easy_setopt(CURL *curl, CURLoption option, ...);
+CURL_EXTERN CURLcode curl_easy_perform(CURL *curl);
+CURL_EXTERN void curl_easy_cleanup(CURL *curl);
+
+/*
+ * NAME curl_easy_getinfo()
+ *
+ * DESCRIPTION
+ *
+ * Request internal information from the curl session with this function.  The
+ * third argument MUST be a pointer to a long, a pointer to a char * or a
+ * pointer to a double (as the documentation describes elsewhere).  The data
+ * pointed to will be filled in accordingly and can be relied upon only if the
+ * function returns CURLE_OK.  This function is intended to get used *AFTER* a
+ * performed transfer, all results from this function are undefined until the
+ * transfer is completed.
+ */
+CURL_EXTERN CURLcode curl_easy_getinfo(CURL *curl, CURLINFO info, ...);
+
+
+/*
+ * NAME curl_easy_duphandle()
+ *
+ * DESCRIPTION
+ *
+ * Creates a new curl session handle with the same options set for the handle
+ * passed in. Duplicating a handle could only be a matter of cloning data and
+ * options, internal state info and things like persistent connections cannot
+ * be transferred. It is useful in multithreaded applications when you can run
+ * curl_easy_duphandle() for each new thread to avoid a series of identical
+ * curl_easy_setopt() invokes in every thread.
+ */
+CURL_EXTERN CURL *curl_easy_duphandle(CURL *curl);
+
+/*
+ * NAME curl_easy_reset()
+ *
+ * DESCRIPTION
+ *
+ * Re-initializes a CURL handle to the default values. This puts back the
+ * handle to the same state as it was in when it was just created.
+ *
+ * It does keep: live connections, the Session ID cache, the DNS cache and the
+ * cookies.
+ */
+CURL_EXTERN void curl_easy_reset(CURL *curl);
+
+/*
+ * NAME curl_easy_recv()
+ *
+ * DESCRIPTION
+ *
+ * Receives data from the connected socket. Use after successful
+ * curl_easy_perform() with CURLOPT_CONNECT_ONLY option.
+ */
+CURL_EXTERN CURLcode curl_easy_recv(CURL *curl, void *buffer, size_t buflen,
+                                    size_t *n);
+
+/*
+ * NAME curl_easy_send()
+ *
+ * DESCRIPTION
+ *
+ * Sends data over the connected socket. Use after successful
+ * curl_easy_perform() with CURLOPT_CONNECT_ONLY option.
+ */
+CURL_EXTERN CURLcode curl_easy_send(CURL *curl, const void *buffer,
+                                    size_t buflen, size_t *n);
+
+#ifdef  __cplusplus
+}
+#endif
+
+#endif
diff --git a/distribution/curl/include/curl/mprintf.h b/distribution/curl/include/curl/mprintf.h
new file mode 100644
index 0000000..e20f546
--- /dev/null
+++ b/distribution/curl/include/curl/mprintf.h
@@ -0,0 +1,50 @@
+#ifndef __CURL_MPRINTF_H
+#define __CURL_MPRINTF_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2016, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+
+#include <stdarg.h>
+#include <stdio.h> /* needed for FILE */
+#include "curl.h"  /* for CURL_EXTERN */
+
+#ifdef  __cplusplus
+extern "C" {
+#endif
+
+CURL_EXTERN int curl_mprintf(const char *format, ...);
+CURL_EXTERN int curl_mfprintf(FILE *fd, const char *format, ...);
+CURL_EXTERN int curl_msprintf(char *buffer, const char *format, ...);
+CURL_EXTERN int curl_msnprintf(char *buffer, size_t maxlength,
+                               const char *format, ...);
+CURL_EXTERN int curl_mvprintf(const char *format, va_list args);
+CURL_EXTERN int curl_mvfprintf(FILE *fd, const char *format, va_list args);
+CURL_EXTERN int curl_mvsprintf(char *buffer, const char *format, va_list args);
+CURL_EXTERN int curl_mvsnprintf(char *buffer, size_t maxlength,
+                                const char *format, va_list args);
+CURL_EXTERN char *curl_maprintf(const char *format, ...);
+CURL_EXTERN char *curl_mvaprintf(const char *format, va_list args);
+
+#ifdef  __cplusplus
+}
+#endif
+
+#endif /* __CURL_MPRINTF_H */
diff --git a/distribution/curl/include/curl/multi.h b/distribution/curl/include/curl/multi.h
new file mode 100644
index 0000000..911c91d
--- /dev/null
+++ b/distribution/curl/include/curl/multi.h
@@ -0,0 +1,439 @@
+#ifndef __CURL_MULTI_H
+#define __CURL_MULTI_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+/*
+  This is an "external" header file. Don't give away any internals here!
+
+  GOALS
+
+  o Enable a "pull" interface. The application that uses libcurl decides where
+    and when to ask libcurl to get/send data.
+
+  o Enable multiple simultaneous transfers in the same thread without making it
+    complicated for the application.
+
+  o Enable the application to select() on its own file descriptors and curl's
+    file descriptors simultaneous easily.
+
+*/
+
+/*
+ * This header file should not really need to include "curl.h" since curl.h
+ * itself includes this file and we expect user applications to do #include
+ * <curl/curl.h> without the need for especially including multi.h.
+ *
+ * For some reason we added this include here at one point, and rather than to
+ * break existing (wrongly written) libcurl applications, we leave it as-is
+ * but with this warning attached.
+ */
+#include "curl.h"
+
+#ifdef  __cplusplus
+extern "C" {
+#endif
+
+#if defined(BUILDING_LIBCURL) || defined(CURL_STRICTER)
+typedef struct Curl_multi CURLM;
+#else
+typedef void CURLM;
+#endif
+
+typedef enum {
+  CURLM_CALL_MULTI_PERFORM = -1, /* please call curl_multi_perform() or
+                                    curl_multi_socket*() soon */
+  CURLM_OK,
+  CURLM_BAD_HANDLE,      /* the passed-in handle is not a valid CURLM handle */
+  CURLM_BAD_EASY_HANDLE, /* an easy handle was not good/valid */
+  CURLM_OUT_OF_MEMORY,   /* if you ever get this, you're in deep sh*t */
+  CURLM_INTERNAL_ERROR,  /* this is a libcurl bug */
+  CURLM_BAD_SOCKET,      /* the passed in socket argument did not match */
+  CURLM_UNKNOWN_OPTION,  /* curl_multi_setopt() with unsupported option */
+  CURLM_ADDED_ALREADY,   /* an easy handle already added to a multi handle was
+                            attempted to get added - again */
+  CURLM_LAST
+} CURLMcode;
+
+/* just to make code nicer when using curl_multi_socket() you can now check
+   for CURLM_CALL_MULTI_SOCKET too in the same style it works for
+   curl_multi_perform() and CURLM_CALL_MULTI_PERFORM */
+#define CURLM_CALL_MULTI_SOCKET CURLM_CALL_MULTI_PERFORM
+
+/* bitmask bits for CURLMOPT_PIPELINING */
+#define CURLPIPE_NOTHING   0L
+#define CURLPIPE_HTTP1     1L
+#define CURLPIPE_MULTIPLEX 2L
+
+typedef enum {
+  CURLMSG_NONE, /* first, not used */
+  CURLMSG_DONE, /* This easy handle has completed. 'result' contains
+                   the CURLcode of the transfer */
+  CURLMSG_LAST /* last, not used */
+} CURLMSG;
+
+struct CURLMsg {
+  CURLMSG msg;       /* what this message means */
+  CURL *easy_handle; /* the handle it concerns */
+  union {
+    void *whatever;    /* message-specific data */
+    CURLcode result;   /* return code for transfer */
+  } data;
+};
+typedef struct CURLMsg CURLMsg;
+
+/* Based on poll(2) structure and values.
+ * We don't use pollfd and POLL* constants explicitly
+ * to cover platforms without poll(). */
+#define CURL_WAIT_POLLIN    0x0001
+#define CURL_WAIT_POLLPRI   0x0002
+#define CURL_WAIT_POLLOUT   0x0004
+
+struct curl_waitfd {
+  curl_socket_t fd;
+  short events;
+  short revents; /* not supported yet */
+};
+
+/*
+ * Name:    curl_multi_init()
+ *
+ * Desc:    inititalize multi-style curl usage
+ *
+ * Returns: a new CURLM handle to use in all 'curl_multi' functions.
+ */
+CURL_EXTERN CURLM *curl_multi_init(void);
+
+/*
+ * Name:    curl_multi_add_handle()
+ *
+ * Desc:    add a standard curl handle to the multi stack
+ *
+ * Returns: CURLMcode type, general multi error code.
+ */
+CURL_EXTERN CURLMcode curl_multi_add_handle(CURLM *multi_handle,
+                                            CURL *curl_handle);
+
+ /*
+  * Name:    curl_multi_remove_handle()
+  *
+  * Desc:    removes a curl handle from the multi stack again
+  *
+  * Returns: CURLMcode type, general multi error code.
+  */
+CURL_EXTERN CURLMcode curl_multi_remove_handle(CURLM *multi_handle,
+                                               CURL *curl_handle);
+
+ /*
+  * Name:    curl_multi_fdset()
+  *
+  * Desc:    Ask curl for its fd_set sets. The app can use these to select() or
+  *          poll() on. We want curl_multi_perform() called as soon as one of
+  *          them are ready.
+  *
+  * Returns: CURLMcode type, general multi error code.
+  */
+CURL_EXTERN CURLMcode curl_multi_fdset(CURLM *multi_handle,
+                                       fd_set *read_fd_set,
+                                       fd_set *write_fd_set,
+                                       fd_set *exc_fd_set,
+                                       int *max_fd);
+
+/*
+ * Name:     curl_multi_wait()
+ *
+ * Desc:     Poll on all fds within a CURLM set as well as any
+ *           additional fds passed to the function.
+ *
+ * Returns:  CURLMcode type, general multi error code.
+ */
+CURL_EXTERN CURLMcode curl_multi_wait(CURLM *multi_handle,
+                                      struct curl_waitfd extra_fds[],
+                                      unsigned int extra_nfds,
+                                      int timeout_ms,
+                                      int *ret);
+
+ /*
+  * Name:    curl_multi_perform()
+  *
+  * Desc:    When the app thinks there's data available for curl it calls this
+  *          function to read/write whatever there is right now. This returns
+  *          as soon as the reads and writes are done. This function does not
+  *          require that there actually is data available for reading or that
+  *          data can be written, it can be called just in case. It returns
+  *          the number of handles that still transfer data in the second
+  *          argument's integer-pointer.
+  *
+  * Returns: CURLMcode type, general multi error code. *NOTE* that this only
+  *          returns errors etc regarding the whole multi stack. There might
+  *          still have occurred problems on invidual transfers even when this
+  *          returns OK.
+  */
+CURL_EXTERN CURLMcode curl_multi_perform(CURLM *multi_handle,
+                                         int *running_handles);
+
+ /*
+  * Name:    curl_multi_cleanup()
+  *
+  * Desc:    Cleans up and removes a whole multi stack. It does not free or
+  *          touch any individual easy handles in any way. We need to define
+  *          in what state those handles will be if this function is called
+  *          in the middle of a transfer.
+  *
+  * Returns: CURLMcode type, general multi error code.
+  */
+CURL_EXTERN CURLMcode curl_multi_cleanup(CURLM *multi_handle);
+
+/*
+ * Name:    curl_multi_info_read()
+ *
+ * Desc:    Ask the multi handle if there's any messages/informationals from
+ *          the individual transfers. Messages include informationals such as
+ *          error code from the transfer or just the fact that a transfer is
+ *          completed. More details on these should be written down as well.
+ *
+ *          Repeated calls to this function will return a new struct each
+ *          time, until a special "end of msgs" struct is returned as a signal
+ *          that there is no more to get at this point.
+ *
+ *          The data the returned pointer points to will not survive calling
+ *          curl_multi_cleanup().
+ *
+ *          The 'CURLMsg' struct is meant to be very simple and only contain
+ *          very basic information. If more involved information is wanted,
+ *          we will provide the particular "transfer handle" in that struct
+ *          and that should/could/would be used in subsequent
+ *          curl_easy_getinfo() calls (or similar). The point being that we
+ *          must never expose complex structs to applications, as then we'll
+ *          undoubtably get backwards compatibility problems in the future.
+ *
+ * Returns: A pointer to a filled-in struct, or NULL if it failed or ran out
+ *          of structs. It also writes the number of messages left in the
+ *          queue (after this read) in the integer the second argument points
+ *          to.
+ */
+CURL_EXTERN CURLMsg *curl_multi_info_read(CURLM *multi_handle,
+                                          int *msgs_in_queue);
+
+/*
+ * Name:    curl_multi_strerror()
+ *
+ * Desc:    The curl_multi_strerror function may be used to turn a CURLMcode
+ *          value into the equivalent human readable error string.  This is
+ *          useful for printing meaningful error messages.
+ *
+ * Returns: A pointer to a zero-terminated error message.
+ */
+CURL_EXTERN const char *curl_multi_strerror(CURLMcode);
+
+/*
+ * Name:    curl_multi_socket() and
+ *          curl_multi_socket_all()
+ *
+ * Desc:    An alternative version of curl_multi_perform() that allows the
+ *          application to pass in one of the file descriptors that have been
+ *          detected to have "action" on them and let libcurl perform.
+ *          See man page for details.
+ */
+#define CURL_POLL_NONE   0
+#define CURL_POLL_IN     1
+#define CURL_POLL_OUT    2
+#define CURL_POLL_INOUT  3
+#define CURL_POLL_REMOVE 4
+
+#define CURL_SOCKET_TIMEOUT CURL_SOCKET_BAD
+
+#define CURL_CSELECT_IN   0x01
+#define CURL_CSELECT_OUT  0x02
+#define CURL_CSELECT_ERR  0x04
+
+typedef int (*curl_socket_callback)(CURL *easy,      /* easy handle */
+                                    curl_socket_t s, /* socket */
+                                    int what,        /* see above */
+                                    void *userp,     /* private callback
+                                                        pointer */
+                                    void *socketp);  /* private socket
+                                                        pointer */
+/*
+ * Name:    curl_multi_timer_callback
+ *
+ * Desc:    Called by libcurl whenever the library detects a change in the
+ *          maximum number of milliseconds the app is allowed to wait before
+ *          curl_multi_socket() or curl_multi_perform() must be called
+ *          (to allow libcurl's timed events to take place).
+ *
+ * Returns: The callback should return zero.
+ */
+typedef int (*curl_multi_timer_callback)(CURLM *multi,    /* multi handle */
+                                         long timeout_ms, /* see above */
+                                         void *userp);    /* private callback
+                                                             pointer */
+
+CURL_EXTERN CURLMcode curl_multi_socket(CURLM *multi_handle, curl_socket_t s,
+                                        int *running_handles);
+
+CURL_EXTERN CURLMcode curl_multi_socket_action(CURLM *multi_handle,
+                                               curl_socket_t s,
+                                               int ev_bitmask,
+                                               int *running_handles);
+
+CURL_EXTERN CURLMcode curl_multi_socket_all(CURLM *multi_handle,
+                                            int *running_handles);
+
+#ifndef CURL_ALLOW_OLD_MULTI_SOCKET
+/* This macro below was added in 7.16.3 to push users who recompile to use
+   the new curl_multi_socket_action() instead of the old curl_multi_socket()
+*/
+#define curl_multi_socket(x,y,z) curl_multi_socket_action(x,y,0,z)
+#endif
+
+/*
+ * Name:    curl_multi_timeout()
+ *
+ * Desc:    Returns the maximum number of milliseconds the app is allowed to
+ *          wait before curl_multi_socket() or curl_multi_perform() must be
+ *          called (to allow libcurl's timed events to take place).
+ *
+ * Returns: CURLM error code.
+ */
+CURL_EXTERN CURLMcode curl_multi_timeout(CURLM *multi_handle,
+                                         long *milliseconds);
+
+#undef CINIT /* re-using the same name as in curl.h */
+
+#ifdef CURL_ISOCPP
+#define CINIT(name,type,num) CURLMOPT_ ## name = CURLOPTTYPE_ ## type + num
+#else
+/* The macro "##" is ISO C, we assume pre-ISO C doesn't support it. */
+#define LONG          CURLOPTTYPE_LONG
+#define OBJECTPOINT   CURLOPTTYPE_OBJECTPOINT
+#define FUNCTIONPOINT CURLOPTTYPE_FUNCTIONPOINT
+#define OFF_T         CURLOPTTYPE_OFF_T
+#define CINIT(name,type,number) CURLMOPT_/**/name = type + number
+#endif
+
+typedef enum {
+  /* This is the socket callback function pointer */
+  CINIT(SOCKETFUNCTION, FUNCTIONPOINT, 1),
+
+  /* This is the argument passed to the socket callback */
+  CINIT(SOCKETDATA, OBJECTPOINT, 2),
+
+    /* set to 1 to enable pipelining for this multi handle */
+  CINIT(PIPELINING, LONG, 3),
+
+   /* This is the timer callback function pointer */
+  CINIT(TIMERFUNCTION, FUNCTIONPOINT, 4),
+
+  /* This is the argument passed to the timer callback */
+  CINIT(TIMERDATA, OBJECTPOINT, 5),
+
+  /* maximum number of entries in the connection cache */
+  CINIT(MAXCONNECTS, LONG, 6),
+
+  /* maximum number of (pipelining) connections to one host */
+  CINIT(MAX_HOST_CONNECTIONS, LONG, 7),
+
+  /* maximum number of requests in a pipeline */
+  CINIT(MAX_PIPELINE_LENGTH, LONG, 8),
+
+  /* a connection with a content-length longer than this
+     will not be considered for pipelining */
+  CINIT(CONTENT_LENGTH_PENALTY_SIZE, OFF_T, 9),
+
+  /* a connection with a chunk length longer than this
+     will not be considered for pipelining */
+  CINIT(CHUNK_LENGTH_PENALTY_SIZE, OFF_T, 10),
+
+  /* a list of site names(+port) that are blacklisted from
+     pipelining */
+  CINIT(PIPELINING_SITE_BL, OBJECTPOINT, 11),
+
+  /* a list of server types that are blacklisted from
+     pipelining */
+  CINIT(PIPELINING_SERVER_BL, OBJECTPOINT, 12),
+
+  /* maximum number of open connections in total */
+  CINIT(MAX_TOTAL_CONNECTIONS, LONG, 13),
+
+   /* This is the server push callback function pointer */
+  CINIT(PUSHFUNCTION, FUNCTIONPOINT, 14),
+
+  /* This is the argument passed to the server push callback */
+  CINIT(PUSHDATA, OBJECTPOINT, 15),
+
+  CURLMOPT_LASTENTRY /* the last unused */
+} CURLMoption;
+
+
+/*
+ * Name:    curl_multi_setopt()
+ *
+ * Desc:    Sets options for the multi handle.
+ *
+ * Returns: CURLM error code.
+ */
+CURL_EXTERN CURLMcode curl_multi_setopt(CURLM *multi_handle,
+                                        CURLMoption option, ...);
+
+
+/*
+ * Name:    curl_multi_assign()
+ *
+ * Desc:    This function sets an association in the multi handle between the
+ *          given socket and a private pointer of the application. This is
+ *          (only) useful for curl_multi_socket uses.
+ *
+ * Returns: CURLM error code.
+ */
+CURL_EXTERN CURLMcode curl_multi_assign(CURLM *multi_handle,
+                                        curl_socket_t sockfd, void *sockp);
+
+
+/*
+ * Name: curl_push_callback
+ *
+ * Desc: This callback gets called when a new stream is being pushed by the
+ *       server. It approves or denies the new stream.
+ *
+ * Returns: CURL_PUSH_OK or CURL_PUSH_DENY.
+ */
+#define CURL_PUSH_OK   0
+#define CURL_PUSH_DENY 1
+
+struct curl_pushheaders;  /* forward declaration only */
+
+CURL_EXTERN char *curl_pushheader_bynum(struct curl_pushheaders *h,
+                                        size_t num);
+CURL_EXTERN char *curl_pushheader_byname(struct curl_pushheaders *h,
+                                         const char *name);
+
+typedef int (*curl_push_callback)(CURL *parent,
+                                  CURL *easy,
+                                  size_t num_headers,
+                                  struct curl_pushheaders *headers,
+                                  void *userp);
+
+#ifdef __cplusplus
+} /* end of extern "C" */
+#endif
+
+#endif
diff --git a/distribution/curl/include/curl/stdcheaders.h b/distribution/curl/include/curl/stdcheaders.h
new file mode 100644
index 0000000..027b6f4
--- /dev/null
+++ b/distribution/curl/include/curl/stdcheaders.h
@@ -0,0 +1,33 @@
+#ifndef __STDC_HEADERS_H
+#define __STDC_HEADERS_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2016, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+
+#include <sys/types.h>
+
+size_t fread(void *, size_t, size_t, FILE *);
+size_t fwrite(const void *, size_t, size_t, FILE *);
+
+int strcasecmp(const char *, const char *);
+int strncasecmp(const char *, const char *, size_t);
+
+#endif /* __STDC_HEADERS_H */
diff --git a/distribution/curl/include/curl/system.h b/distribution/curl/include/curl/system.h
new file mode 100644
index 0000000..39dae75
--- /dev/null
+++ b/distribution/curl/include/curl/system.h
@@ -0,0 +1,471 @@
+#ifndef __CURL_SYSTEM_H
+#define __CURL_SYSTEM_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+
+/*
+ * Try to keep one section per platform, compiler and architecture, otherwise,
+ * if an existing section is reused for a different one and later on the
+ * original is adjusted, probably the piggybacking one can be adversely
+ * changed.
+ *
+ * In order to differentiate between platforms/compilers/architectures use
+ * only compiler built in predefined preprocessor symbols.
+ *
+ * curl_off_t
+ * ----------
+ *
+ * For any given platform/compiler curl_off_t must be typedef'ed to a 64-bit
+ * wide signed integral data type. The width of this data type must remain
+ * constant and independent of any possible large file support settings.
+ *
+ * As an exception to the above, curl_off_t shall be typedef'ed to a 32-bit
+ * wide signed integral data type if there is no 64-bit type.
+ *
+ * As a general rule, curl_off_t shall not be mapped to off_t. This rule shall
+ * only be violated if off_t is the only 64-bit data type available and the
+ * size of off_t is independent of large file support settings. Keep your
+ * build on the safe side avoiding an off_t gating.  If you have a 64-bit
+ * off_t then take for sure that another 64-bit data type exists, dig deeper
+ * and you will find it.
+ *
+ */
+
+#if defined(__DJGPP__) || defined(__GO32__)
+#  if defined(__DJGPP__) && (__DJGPP__ > 1)
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  else
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__SALFORDC__)
+#  define CURL_TYPEOF_CURL_OFF_T     long
+#  define CURL_FORMAT_CURL_OFF_T     "ld"
+#  define CURL_FORMAT_CURL_OFF_TU    "lu"
+#  define CURL_SUFFIX_CURL_OFF_T     L
+#  define CURL_SUFFIX_CURL_OFF_TU    UL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__BORLANDC__)
+#  if (__BORLANDC__ < 0x520)
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  else
+#    define CURL_TYPEOF_CURL_OFF_T     __int64
+#    define CURL_FORMAT_CURL_OFF_T     "I64d"
+#    define CURL_FORMAT_CURL_OFF_TU    "I64u"
+#    define CURL_SUFFIX_CURL_OFF_T     i64
+#    define CURL_SUFFIX_CURL_OFF_TU    ui64
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__TURBOC__)
+#  define CURL_TYPEOF_CURL_OFF_T     long
+#  define CURL_FORMAT_CURL_OFF_T     "ld"
+#  define CURL_FORMAT_CURL_OFF_TU    "lu"
+#  define CURL_SUFFIX_CURL_OFF_T     L
+#  define CURL_SUFFIX_CURL_OFF_TU    UL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__WATCOMC__)
+#  if defined(__386__)
+#    define CURL_TYPEOF_CURL_OFF_T     __int64
+#    define CURL_FORMAT_CURL_OFF_T     "I64d"
+#    define CURL_FORMAT_CURL_OFF_TU    "I64u"
+#    define CURL_SUFFIX_CURL_OFF_T     i64
+#    define CURL_SUFFIX_CURL_OFF_TU    ui64
+#  else
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__POCC__)
+#  if (__POCC__ < 280)
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  elif defined(_MSC_VER)
+#    define CURL_TYPEOF_CURL_OFF_T     __int64
+#    define CURL_FORMAT_CURL_OFF_T     "I64d"
+#    define CURL_FORMAT_CURL_OFF_TU    "I64u"
+#    define CURL_SUFFIX_CURL_OFF_T     i64
+#    define CURL_SUFFIX_CURL_OFF_TU    ui64
+#  else
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__LCC__)
+#  define CURL_TYPEOF_CURL_OFF_T     long
+#  define CURL_FORMAT_CURL_OFF_T     "ld"
+#  define CURL_FORMAT_CURL_OFF_TU    "lu"
+#  define CURL_SUFFIX_CURL_OFF_T     L
+#  define CURL_SUFFIX_CURL_OFF_TU    UL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__SYMBIAN32__)
+#  if defined(__EABI__)  /* Treat all ARM compilers equally */
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  elif defined(__CW32__)
+#    pragma longlong on
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  elif defined(__VC32__)
+#    define CURL_TYPEOF_CURL_OFF_T     __int64
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T unsigned int
+
+#elif defined(__MWERKS__)
+#  define CURL_TYPEOF_CURL_OFF_T     long long
+#  define CURL_FORMAT_CURL_OFF_T     "lld"
+#  define CURL_FORMAT_CURL_OFF_TU    "llu"
+#  define CURL_SUFFIX_CURL_OFF_T     LL
+#  define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(_WIN32_WCE)
+#  define CURL_TYPEOF_CURL_OFF_T     __int64
+#  define CURL_FORMAT_CURL_OFF_T     "I64d"
+#  define CURL_FORMAT_CURL_OFF_TU    "I64u"
+#  define CURL_SUFFIX_CURL_OFF_T     i64
+#  define CURL_SUFFIX_CURL_OFF_TU    ui64
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__MINGW32__)
+#  define CURL_TYPEOF_CURL_OFF_T     long long
+#  define CURL_FORMAT_CURL_OFF_T     "I64d"
+#  define CURL_FORMAT_CURL_OFF_TU    "I64u"
+#  define CURL_SUFFIX_CURL_OFF_T     LL
+#  define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#  define CURL_PULL_SYS_TYPES_H      1
+#  define CURL_PULL_WS2TCPIP_H       1
+
+#elif defined(__VMS)
+#  if defined(__VAX)
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  else
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T unsigned int
+
+#elif defined(__OS400__)
+#  if defined(__ILEC400__)
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#    define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#    define CURL_PULL_SYS_TYPES_H      1
+#    define CURL_PULL_SYS_SOCKET_H     1
+#  endif
+
+#elif defined(__MVS__)
+#  if defined(__IBMC__) || defined(__IBMCPP__)
+#    if defined(_ILP32)
+#    elif defined(_LP64)
+#    endif
+#    if defined(_LONG_LONG)
+#      define CURL_TYPEOF_CURL_OFF_T     long long
+#      define CURL_FORMAT_CURL_OFF_T     "lld"
+#      define CURL_FORMAT_CURL_OFF_TU    "llu"
+#      define CURL_SUFFIX_CURL_OFF_T     LL
+#      define CURL_SUFFIX_CURL_OFF_TU    ULL
+#    elif defined(_LP64)
+#      define CURL_TYPEOF_CURL_OFF_T     long
+#      define CURL_FORMAT_CURL_OFF_T     "ld"
+#      define CURL_FORMAT_CURL_OFF_TU    "lu"
+#      define CURL_SUFFIX_CURL_OFF_T     L
+#      define CURL_SUFFIX_CURL_OFF_TU    UL
+#    else
+#      define CURL_TYPEOF_CURL_OFF_T     long
+#      define CURL_FORMAT_CURL_OFF_T     "ld"
+#      define CURL_FORMAT_CURL_OFF_TU    "lu"
+#      define CURL_SUFFIX_CURL_OFF_T     L
+#      define CURL_SUFFIX_CURL_OFF_TU    UL
+#    endif
+#    define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#    define CURL_PULL_SYS_TYPES_H      1
+#    define CURL_PULL_SYS_SOCKET_H     1
+#  endif
+
+#elif defined(__370__)
+#  if defined(__IBMC__) || defined(__IBMCPP__)
+#    if defined(_ILP32)
+#    elif defined(_LP64)
+#    endif
+#    if defined(_LONG_LONG)
+#      define CURL_TYPEOF_CURL_OFF_T     long long
+#      define CURL_FORMAT_CURL_OFF_T     "lld"
+#      define CURL_FORMAT_CURL_OFF_TU    "llu"
+#      define CURL_SUFFIX_CURL_OFF_T     LL
+#      define CURL_SUFFIX_CURL_OFF_TU    ULL
+#    elif defined(_LP64)
+#      define CURL_TYPEOF_CURL_OFF_T     long
+#      define CURL_FORMAT_CURL_OFF_T     "ld"
+#      define CURL_FORMAT_CURL_OFF_TU    "lu"
+#      define CURL_SUFFIX_CURL_OFF_T     L
+#      define CURL_SUFFIX_CURL_OFF_TU    UL
+#    else
+#      define CURL_TYPEOF_CURL_OFF_T     long
+#      define CURL_FORMAT_CURL_OFF_T     "ld"
+#      define CURL_FORMAT_CURL_OFF_TU    "lu"
+#      define CURL_SUFFIX_CURL_OFF_T     L
+#      define CURL_SUFFIX_CURL_OFF_TU    UL
+#    endif
+#    define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#    define CURL_PULL_SYS_TYPES_H      1
+#    define CURL_PULL_SYS_SOCKET_H     1
+#  endif
+
+#elif defined(TPF)
+#  define CURL_TYPEOF_CURL_OFF_T     long
+#  define CURL_FORMAT_CURL_OFF_T     "ld"
+#  define CURL_FORMAT_CURL_OFF_TU    "lu"
+#  define CURL_SUFFIX_CURL_OFF_T     L
+#  define CURL_SUFFIX_CURL_OFF_TU    UL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+#elif defined(__TINYC__) /* also known as tcc */
+
+#  define CURL_TYPEOF_CURL_OFF_T     long long
+#  define CURL_FORMAT_CURL_OFF_T     "lld"
+#  define CURL_FORMAT_CURL_OFF_TU    "llu"
+#  define CURL_SUFFIX_CURL_OFF_T     LL
+#  define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#  define CURL_PULL_SYS_TYPES_H      1
+#  define CURL_PULL_SYS_SOCKET_H     1
+
+#elif defined(__SUNPRO_C) /* Oracle Solaris Studio */
+#  if !defined(__LP64) && (defined(__ILP32) ||                          \
+                           defined(__i386) || defined(__sparcv8))
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  elif defined(__LP64) || \
+        defined(__amd64) || defined(__sparcv9)
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#  define CURL_PULL_SYS_TYPES_H      1
+#  define CURL_PULL_SYS_SOCKET_H     1
+
+/* ===================================== */
+/*    KEEP MSVC THE PENULTIMATE ENTRY    */
+/* ===================================== */
+
+#elif defined(_MSC_VER)
+#  if (_MSC_VER >= 900) && (_INTEGRAL_MAX_BITS >= 64)
+#    define CURL_TYPEOF_CURL_OFF_T     __int64
+#    define CURL_FORMAT_CURL_OFF_T     "I64d"
+#    define CURL_FORMAT_CURL_OFF_TU    "I64u"
+#    define CURL_SUFFIX_CURL_OFF_T     i64
+#    define CURL_SUFFIX_CURL_OFF_TU    ui64
+#  else
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T int
+
+/* ===================================== */
+/*    KEEP GENERIC GCC THE LAST ENTRY    */
+/* ===================================== */
+
+#elif defined(__GNUC__)
+#  if !defined(__LP64__) &&                                             \
+  (defined(__ILP32__) || defined(__i386__) || defined(__hppa__) ||      \
+   defined(__ppc__) || defined(__powerpc__) || defined(__arm__) ||      \
+   defined(__sparc__) || defined(__mips__) || defined(__sh__) ||        \
+   defined(__XTENSA__) ||                                               \
+   (defined(__SIZEOF_LONG__) && __SIZEOF_LONG__ == 4))
+#    define CURL_TYPEOF_CURL_OFF_T     long long
+#    define CURL_FORMAT_CURL_OFF_T     "lld"
+#    define CURL_FORMAT_CURL_OFF_TU    "llu"
+#    define CURL_SUFFIX_CURL_OFF_T     LL
+#    define CURL_SUFFIX_CURL_OFF_TU    ULL
+#  elif defined(__LP64__) || \
+        defined(__x86_64__) || defined(__ppc64__) || defined(__sparc64__) || \
+        (defined(__SIZEOF_LONG__) && __SIZEOF_LONG__ == 8)
+#    define CURL_TYPEOF_CURL_OFF_T     long
+#    define CURL_FORMAT_CURL_OFF_T     "ld"
+#    define CURL_FORMAT_CURL_OFF_TU    "lu"
+#    define CURL_SUFFIX_CURL_OFF_T     L
+#    define CURL_SUFFIX_CURL_OFF_TU    UL
+#  endif
+#  define CURL_TYPEOF_CURL_SOCKLEN_T socklen_t
+#  define CURL_PULL_SYS_TYPES_H      1
+#  define CURL_PULL_SYS_SOCKET_H     1
+
+#else
+/* generic "safe guess" on old 32 bit style */
+# define CURL_TYPEOF_CURL_OFF_T     long
+# define CURL_FORMAT_CURL_OFF_T     "ld"
+# define CURL_FORMAT_CURL_OFF_TU    "lu"
+# define CURL_SUFFIX_CURL_OFF_T     L
+# define CURL_SUFFIX_CURL_OFF_TU    UL
+# define CURL_TYPEOF_CURL_SOCKLEN_T int
+#endif
+
+#ifdef _AIX
+/* AIX needs <sys/poll.h> */
+#define CURL_PULL_SYS_POLL_H
+#endif
+
+
+/* CURL_PULL_WS2TCPIP_H is defined above when inclusion of header file  */
+/* ws2tcpip.h is required here to properly make type definitions below. */
+#ifdef CURL_PULL_WS2TCPIP_H
+#  include <winsock2.h>
+#  include <windows.h>
+#  include <ws2tcpip.h>
+#endif
+
+/* CURL_PULL_SYS_TYPES_H is defined above when inclusion of header file  */
+/* sys/types.h is required here to properly make type definitions below. */
+#ifdef CURL_PULL_SYS_TYPES_H
+#  include <sys/types.h>
+#endif
+
+/* CURL_PULL_SYS_SOCKET_H is defined above when inclusion of header file  */
+/* sys/socket.h is required here to properly make type definitions below. */
+#ifdef CURL_PULL_SYS_SOCKET_H
+#  include <sys/socket.h>
+#endif
+
+/* CURL_PULL_SYS_POLL_H is defined above when inclusion of header file    */
+/* sys/poll.h is required here to properly make type definitions below.   */
+#ifdef CURL_PULL_SYS_POLL_H
+#  include <sys/poll.h>
+#endif
+
+/* Data type definition of curl_socklen_t. */
+#ifdef CURL_TYPEOF_CURL_SOCKLEN_T
+  typedef CURL_TYPEOF_CURL_SOCKLEN_T curl_socklen_t;
+#endif
+
+/* Data type definition of curl_off_t. */
+
+#ifdef CURL_TYPEOF_CURL_OFF_T
+  typedef CURL_TYPEOF_CURL_OFF_T curl_off_t;
+#endif
+
+/*
+ * CURL_ISOCPP and CURL_OFF_T_C definitions are done here in order to allow
+ * these to be visible and exported by the external libcurl interface API,
+ * while also making them visible to the library internals, simply including
+ * curl_setup.h, without actually needing to include curl.h internally.
+ * If some day this section would grow big enough, all this should be moved
+ * to its own header file.
+ */
+
+/*
+ * Figure out if we can use the ## preprocessor operator, which is supported
+ * by ISO/ANSI C and C++. Some compilers support it without setting __STDC__
+ * or  __cplusplus so we need to carefully check for them too.
+ */
+
+#if defined(__STDC__) || defined(_MSC_VER) || defined(__cplusplus) || \
+  defined(__HP_aCC) || defined(__BORLANDC__) || defined(__LCC__) || \
+  defined(__POCC__) || defined(__SALFORDC__) || defined(__HIGHC__) || \
+  defined(__ILEC400__)
+  /* This compiler is believed to have an ISO compatible preprocessor */
+#define CURL_ISOCPP
+#else
+  /* This compiler is believed NOT to have an ISO compatible preprocessor */
+#undef CURL_ISOCPP
+#endif
+
+/*
+ * Macros for minimum-width signed and unsigned curl_off_t integer constants.
+ */
+
+#if defined(__BORLANDC__) && (__BORLANDC__ == 0x0551)
+#  define __CURL_OFF_T_C_HLPR2(x) x
+#  define __CURL_OFF_T_C_HLPR1(x) __CURL_OFF_T_C_HLPR2(x)
+#  define CURL_OFF_T_C(Val)  __CURL_OFF_T_C_HLPR1(Val) ## \
+                             __CURL_OFF_T_C_HLPR1(CURL_SUFFIX_CURL_OFF_T)
+#  define CURL_OFF_TU_C(Val) __CURL_OFF_T_C_HLPR1(Val) ## \
+                             __CURL_OFF_T_C_HLPR1(CURL_SUFFIX_CURL_OFF_TU)
+#else
+#  ifdef CURL_ISOCPP
+#    define __CURL_OFF_T_C_HLPR2(Val,Suffix) Val ## Suffix
+#  else
+#    define __CURL_OFF_T_C_HLPR2(Val,Suffix) Val/**/Suffix
+#  endif
+#  define __CURL_OFF_T_C_HLPR1(Val,Suffix) __CURL_OFF_T_C_HLPR2(Val,Suffix)
+#  define CURL_OFF_T_C(Val)  __CURL_OFF_T_C_HLPR1(Val,CURL_SUFFIX_CURL_OFF_T)
+#  define CURL_OFF_TU_C(Val) __CURL_OFF_T_C_HLPR1(Val,CURL_SUFFIX_CURL_OFF_TU)
+#endif
+
+#endif /* __CURL_SYSTEM_H */
diff --git a/distribution/curl/include/curl/typecheck-gcc.h b/distribution/curl/include/curl/typecheck-gcc.h
new file mode 100644
index 0000000..10c74c7
--- /dev/null
+++ b/distribution/curl/include/curl/typecheck-gcc.h
@@ -0,0 +1,683 @@
+#ifndef __CURL_TYPECHECK_GCC_H
+#define __CURL_TYPECHECK_GCC_H
+/***************************************************************************
+ *                                  _   _ ____  _
+ *  Project                     ___| | | |  _ \| |
+ *                             / __| | | | |_) | |
+ *                            | (__| |_| |  _ <| |___
+ *                             \___|\___/|_| \_\_____|
+ *
+ * Copyright (C) 1998 - 2017, Daniel Stenberg, <daniel@haxx.se>, et al.
+ *
+ * This software is licensed as described in the file COPYING, which
+ * you should have received as part of this distribution. The terms
+ * are also available at https://curl.haxx.se/docs/copyright.html.
+ *
+ * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+ * copies of the Software, and permit persons to whom the Software is
+ * furnished to do so, under the terms of the COPYING file.
+ *
+ * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+ * KIND, either express or implied.
+ *
+ ***************************************************************************/
+
+/* wraps curl_easy_setopt() with typechecking */
+
+/* To add a new kind of warning, add an
+ *   if(_curl_is_sometype_option(_curl_opt))
+ *     if(!_curl_is_sometype(value))
+ *       _curl_easy_setopt_err_sometype();
+ * block and define _curl_is_sometype_option, _curl_is_sometype and
+ * _curl_easy_setopt_err_sometype below
+ *
+ * NOTE: We use two nested 'if' statements here instead of the && operator, in
+ *       order to work around gcc bug #32061.  It affects only gcc 4.3.x/4.4.x
+ *       when compiling with -Wlogical-op.
+ *
+ * To add an option that uses the same type as an existing option, you'll just
+ * need to extend the appropriate _curl_*_option macro
+ */
+#define curl_easy_setopt(handle, option, value)                               \
+__extension__ ({                                                              \
+  __typeof__(option) _curl_opt = option;                                     \
+  if(__builtin_constant_p(_curl_opt)) {                                       \
+    if(_curl_is_long_option(_curl_opt))                                       \
+      if(!_curl_is_long(value))                                               \
+        _curl_easy_setopt_err_long();                                         \
+    if(_curl_is_off_t_option(_curl_opt))                                      \
+      if(!_curl_is_off_t(value))                                              \
+        _curl_easy_setopt_err_curl_off_t();                                   \
+    if(_curl_is_string_option(_curl_opt))                                     \
+      if(!_curl_is_string(value))                                             \
+        _curl_easy_setopt_err_string();                                       \
+    if(_curl_is_write_cb_option(_curl_opt))                                   \
+      if(!_curl_is_write_cb(value))                                           \
+        _curl_easy_setopt_err_write_callback();                               \
+    if((_curl_opt) == CURLOPT_READFUNCTION)                                   \
+      if(!_curl_is_read_cb(value))                                            \
+        _curl_easy_setopt_err_read_cb();                                      \
+    if((_curl_opt) == CURLOPT_IOCTLFUNCTION)                                  \
+      if(!_curl_is_ioctl_cb(value))                                           \
+        _curl_easy_setopt_err_ioctl_cb();                                     \
+    if((_curl_opt) == CURLOPT_SOCKOPTFUNCTION)                                \
+      if(!_curl_is_sockopt_cb(value))                                         \
+        _curl_easy_setopt_err_sockopt_cb();                                   \
+    if((_curl_opt) == CURLOPT_OPENSOCKETFUNCTION)                             \
+      if(!_curl_is_opensocket_cb(value))                                      \
+        _curl_easy_setopt_err_opensocket_cb();                                \
+    if((_curl_opt) == CURLOPT_PROGRESSFUNCTION)                               \
+      if(!_curl_is_progress_cb(value))                                        \
+        _curl_easy_setopt_err_progress_cb();                                  \
+    if((_curl_opt) == CURLOPT_DEBUGFUNCTION)                                  \
+      if(!_curl_is_debug_cb(value))                                           \
+        _curl_easy_setopt_err_debug_cb();                                     \
+    if((_curl_opt) == CURLOPT_SSL_CTX_FUNCTION)                               \
+      if(!_curl_is_ssl_ctx_cb(value))                                         \
+        _curl_easy_setopt_err_ssl_ctx_cb();                                   \
+    if(_curl_is_conv_cb_option(_curl_opt))                                    \
+      if(!_curl_is_conv_cb(value))                                            \
+        _curl_easy_setopt_err_conv_cb();                                      \
+    if((_curl_opt) == CURLOPT_SEEKFUNCTION)                                   \
+      if(!_curl_is_seek_cb(value))                                            \
+        _curl_easy_setopt_err_seek_cb();                                      \
+    if(_curl_is_cb_data_option(_curl_opt))                                    \
+      if(!_curl_is_cb_data(value))                                            \
+        _curl_easy_setopt_err_cb_data();                                      \
+    if((_curl_opt) == CURLOPT_ERRORBUFFER)                                    \
+      if(!_curl_is_error_buffer(value))                                       \
+        _curl_easy_setopt_err_error_buffer();                                 \
+    if((_curl_opt) == CURLOPT_STDERR)                                         \
+      if(!_curl_is_FILE(value))                                               \
+        _curl_easy_setopt_err_FILE();                                         \
+    if(_curl_is_postfields_option(_curl_opt))                                 \
+      if(!_curl_is_postfields(value))                                         \
+        _curl_easy_setopt_err_postfields();                                   \
+    if((_curl_opt) == CURLOPT_HTTPPOST)                                       \
+      if(!_curl_is_arr((value), struct curl_httppost))                        \
+        _curl_easy_setopt_err_curl_httpost();                                 \
+    if((_curl_opt) == CURLOPT_MIMEPOST)                                       \
+      if(!_curl_is_ptr((value), curl_mime))                                   \
+        _curl_easy_setopt_err_curl_mimepost();                                \
+    if(_curl_is_slist_option(_curl_opt))                                      \
+      if(!_curl_is_arr((value), struct curl_slist))                           \
+        _curl_easy_setopt_err_curl_slist();                                   \
+    if((_curl_opt) == CURLOPT_SHARE)                                          \
+      if(!_curl_is_ptr((value), CURLSH))                                      \
+        _curl_easy_setopt_err_CURLSH();                                       \
+  }                                                                           \
+  curl_easy_setopt(handle, _curl_opt, value);                                 \
+})
+
+/* wraps curl_easy_getinfo() with typechecking */
+/* FIXME: don't allow const pointers */
+#define curl_easy_getinfo(handle, info, arg)                                  \
+__extension__ ({                                                              \
+  __typeof__(info) _curl_info = info;                                         \
+  if(__builtin_constant_p(_curl_info)) {                                      \
+    if(_curl_is_string_info(_curl_info))                                      \
+      if(!_curl_is_arr((arg), char *))                                        \
+        _curl_easy_getinfo_err_string();                                      \
+    if(_curl_is_long_info(_curl_info))                                        \
+      if(!_curl_is_arr((arg), long))                                          \
+        _curl_easy_getinfo_err_long();                                        \
+    if(_curl_is_double_info(_curl_info))                                      \
+      if(!_curl_is_arr((arg), double))                                        \
+        _curl_easy_getinfo_err_double();                                      \
+    if(_curl_is_slist_info(_curl_info))                                       \
+      if(!_curl_is_arr((arg), struct curl_slist *))                           \
+        _curl_easy_getinfo_err_curl_slist();                                  \
+    if(_curl_is_tlssessioninfo_info(_curl_info))                              \
+      if(!_curl_is_arr((arg), struct curl_tlssessioninfo *))                  \
+        _curl_easy_getinfo_err_curl_tlssesssioninfo();                        \
+    if(_curl_is_certinfo_info(_curl_info))                                    \
+      if(!_curl_is_arr((arg), struct curl_certinfo *))                        \
+        _curl_easy_getinfo_err_curl_certinfo();                               \
+    if(_curl_is_socket_info(_curl_info))                                      \
+      if(!_curl_is_arr((arg), curl_socket_t))                                 \
+        _curl_easy_getinfo_err_curl_socket();                                 \
+    if(_curl_is_off_t_info(_curl_info))                                       \
+      if(!_curl_is_arr((arg), curl_off_t))                                    \
+        _curl_easy_getinfo_err_curl_off_t();                                  \
+  }                                                                           \
+  curl_easy_getinfo(handle, _curl_info, arg);                                 \
+})
+
+/* TODO: typechecking for curl_share_setopt() and curl_multi_setopt(),
+ * for now just make sure that the functions are called with three
+ * arguments
+ */
+#define curl_share_setopt(share,opt,param) curl_share_setopt(share,opt,param)
+#define curl_multi_setopt(handle,opt,param) curl_multi_setopt(handle,opt,param)
+
+
+/* the actual warnings, triggered by calling the _curl_easy_setopt_err*
+ * functions */
+
+/* To define a new warning, use _CURL_WARNING(identifier, "message") */
+#define _CURL_WARNING(id, message)                                            \
+  static void __attribute__((__warning__(message)))                           \
+  __attribute__((__unused__)) __attribute__((__noinline__))                   \
+  id(void) { __asm__(""); }
+
+_CURL_WARNING(_curl_easy_setopt_err_long,
+  "curl_easy_setopt expects a long argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_curl_off_t,
+  "curl_easy_setopt expects a curl_off_t argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_string,
+              "curl_easy_setopt expects a "
+              "string ('char *' or char[]) argument for this option"
+  )
+_CURL_WARNING(_curl_easy_setopt_err_write_callback,
+  "curl_easy_setopt expects a curl_write_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_read_cb,
+  "curl_easy_setopt expects a curl_read_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_ioctl_cb,
+  "curl_easy_setopt expects a curl_ioctl_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_sockopt_cb,
+  "curl_easy_setopt expects a curl_sockopt_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_opensocket_cb,
+              "curl_easy_setopt expects a "
+              "curl_opensocket_callback argument for this option"
+  )
+_CURL_WARNING(_curl_easy_setopt_err_progress_cb,
+  "curl_easy_setopt expects a curl_progress_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_debug_cb,
+  "curl_easy_setopt expects a curl_debug_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_ssl_ctx_cb,
+  "curl_easy_setopt expects a curl_ssl_ctx_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_conv_cb,
+  "curl_easy_setopt expects a curl_conv_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_seek_cb,
+  "curl_easy_setopt expects a curl_seek_callback argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_cb_data,
+              "curl_easy_setopt expects a "
+              "private data pointer as argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_error_buffer,
+              "curl_easy_setopt expects a "
+              "char buffer of CURL_ERROR_SIZE as argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_FILE,
+  "curl_easy_setopt expects a 'FILE *' argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_postfields,
+  "curl_easy_setopt expects a 'void *' or 'char *' argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_curl_httpost,
+              "curl_easy_setopt expects a 'struct curl_httppost *' "
+              "argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_curl_mimepost,
+              "curl_easy_setopt expects a 'curl_mime *' "
+              "argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_curl_slist,
+  "curl_easy_setopt expects a 'struct curl_slist *' argument for this option")
+_CURL_WARNING(_curl_easy_setopt_err_CURLSH,
+  "curl_easy_setopt expects a CURLSH* argument for this option")
+
+_CURL_WARNING(_curl_easy_getinfo_err_string,
+  "curl_easy_getinfo expects a pointer to 'char *' for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_long,
+  "curl_easy_getinfo expects a pointer to long for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_double,
+  "curl_easy_getinfo expects a pointer to double for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_curl_slist,
+  "curl_easy_getinfo expects a pointer to 'struct curl_slist *' for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_curl_tlssesssioninfo,
+              "curl_easy_getinfo expects a pointer to "
+              "'struct curl_tlssessioninfo *' for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_curl_certinfo,
+              "curl_easy_getinfo expects a pointer to "
+              "'struct curl_certinfo *' for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_curl_socket,
+  "curl_easy_getinfo expects a pointer to curl_socket_t for this info")
+_CURL_WARNING(_curl_easy_getinfo_err_curl_off_t,
+  "curl_easy_getinfo expects a pointer to curl_off_t for this info")
+
+/* groups of curl_easy_setops options that take the same type of argument */
+
+/* To add a new option to one of the groups, just add
+ *   (option) == CURLOPT_SOMETHING
+ * to the or-expression. If the option takes a long or curl_off_t, you don't
+ * have to do anything
+ */
+
+/* evaluates to true if option takes a long argument */
+#define _curl_is_long_option(option)                                          \
+  (0 < (option) && (option) < CURLOPTTYPE_OBJECTPOINT)
+
+#define _curl_is_off_t_option(option)                                         \
+  ((option) > CURLOPTTYPE_OFF_T)
+
+/* evaluates to true if option takes a char* argument */
+#define _curl_is_string_option(option)                                        \
+  ((option) == CURLOPT_ABSTRACT_UNIX_SOCKET ||                                \
+   (option) == CURLOPT_ACCEPT_ENCODING ||                                     \
+   (option) == CURLOPT_CAINFO ||                                              \
+   (option) == CURLOPT_CAPATH ||                                              \
+   (option) == CURLOPT_COOKIE ||                                              \
+   (option) == CURLOPT_COOKIEFILE ||                                          \
+   (option) == CURLOPT_COOKIEJAR ||                                           \
+   (option) == CURLOPT_COOKIELIST ||                                          \
+   (option) == CURLOPT_CRLFILE ||                                             \
+   (option) == CURLOPT_CUSTOMREQUEST ||                                       \
+   (option) == CURLOPT_DEFAULT_PROTOCOL ||                                    \
+   (option) == CURLOPT_DNS_INTERFACE ||                                       \
+   (option) == CURLOPT_DNS_LOCAL_IP4 ||                                       \
+   (option) == CURLOPT_DNS_LOCAL_IP6 ||                                       \
+   (option) == CURLOPT_DNS_SERVERS ||                                         \
+   (option) == CURLOPT_EGDSOCKET ||                                           \
+   (option) == CURLOPT_FTPPORT ||                                             \
+   (option) == CURLOPT_FTP_ACCOUNT ||                                         \
+   (option) == CURLOPT_FTP_ALTERNATIVE_TO_USER ||                             \
+   (option) == CURLOPT_INTERFACE ||                                           \
+   (option) == CURLOPT_ISSUERCERT ||                                          \
+   (option) == CURLOPT_KEYPASSWD ||                                           \
+   (option) == CURLOPT_KRBLEVEL ||                                            \
+   (option) == CURLOPT_LOGIN_OPTIONS ||                                       \
+   (option) == CURLOPT_MAIL_AUTH ||                                           \
+   (option) == CURLOPT_MAIL_FROM ||                                           \
+   (option) == CURLOPT_NETRC_FILE ||                                          \
+   (option) == CURLOPT_NOPROXY ||                                             \
+   (option) == CURLOPT_PASSWORD ||                                            \
+   (option) == CURLOPT_PINNEDPUBLICKEY ||                                     \
+   (option) == CURLOPT_PRE_PROXY ||                                           \
+   (option) == CURLOPT_PROXY ||                                               \
+   (option) == CURLOPT_PROXYPASSWORD ||                                       \
+   (option) == CURLOPT_PROXYUSERNAME ||                                       \
+   (option) == CURLOPT_PROXYUSERPWD ||                                        \
+   (option) == CURLOPT_PROXY_CAINFO ||                                        \
+   (option) == CURLOPT_PROXY_CAPATH ||                                        \
+   (option) == CURLOPT_PROXY_CRLFILE ||                                       \
+   (option) == CURLOPT_PROXY_KEYPASSWD ||                                     \
+   (option) == CURLOPT_PROXY_PINNEDPUBLICKEY ||                               \
+   (option) == CURLOPT_PROXY_SERVICE_NAME ||                                  \
+   (option) == CURLOPT_PROXY_SSLCERT ||                                       \
+   (option) == CURLOPT_PROXY_SSLCERTTYPE ||                                   \
+   (option) == CURLOPT_PROXY_SSLKEY ||                                        \
+   (option) == CURLOPT_PROXY_SSLKEYTYPE ||                                    \
+   (option) == CURLOPT_PROXY_SSL_CIPHER_LIST ||                               \
+   (option) == CURLOPT_PROXY_TLSAUTH_PASSWORD ||                              \
+   (option) == CURLOPT_PROXY_TLSAUTH_USERNAME ||                              \
+   (option) == CURLOPT_PROXY_TLSAUTH_TYPE ||                                  \
+   (option) == CURLOPT_RANDOM_FILE ||                                         \
+   (option) == CURLOPT_RANGE ||                                               \
+   (option) == CURLOPT_REFERER ||                                             \
+   (option) == CURLOPT_RTSP_SESSION_ID ||                                     \
+   (option) == CURLOPT_RTSP_STREAM_URI ||                                     \
+   (option) == CURLOPT_RTSP_TRANSPORT ||                                      \
+   (option) == CURLOPT_SERVICE_NAME ||                                        \
+   (option) == CURLOPT_SOCKS5_GSSAPI_SERVICE ||                               \
+   (option) == CURLOPT_SSH_HOST_PUBLIC_KEY_MD5 ||                             \
+   (option) == CURLOPT_SSH_KNOWNHOSTS ||                                      \
+   (option) == CURLOPT_SSH_PRIVATE_KEYFILE ||                                 \
+   (option) == CURLOPT_SSH_PUBLIC_KEYFILE ||                                  \
+   (option) == CURLOPT_SSLCERT ||                                             \
+   (option) == CURLOPT_SSLCERTTYPE ||                                         \
+   (option) == CURLOPT_SSLENGINE ||                                           \
+   (option) == CURLOPT_SSLKEY ||                                              \
+   (option) == CURLOPT_SSLKEYTYPE ||                                          \
+   (option) == CURLOPT_SSL_CIPHER_LIST ||                                     \
+   (option) == CURLOPT_TLSAUTH_PASSWORD ||                                    \
+   (option) == CURLOPT_TLSAUTH_TYPE ||                                        \
+   (option) == CURLOPT_TLSAUTH_USERNAME ||                                    \
+   (option) == CURLOPT_UNIX_SOCKET_PATH ||                                    \
+   (option) == CURLOPT_URL ||                                                 \
+   (option) == CURLOPT_USERAGENT ||                                           \
+   (option) == CURLOPT_USERNAME ||                                            \
+   (option) == CURLOPT_USERPWD ||                                             \
+   (option) == CURLOPT_XOAUTH2_BEARER ||                                      \
+   0)
+
+/* evaluates to true if option takes a curl_write_callback argument */
+#define _curl_is_write_cb_option(option)                                      \
+  ((option) == CURLOPT_HEADERFUNCTION ||                                      \
+   (option) == CURLOPT_WRITEFUNCTION)
+
+/* evaluates to true if option takes a curl_conv_callback argument */
+#define _curl_is_conv_cb_option(option)                                       \
+  ((option) == CURLOPT_CONV_TO_NETWORK_FUNCTION ||                            \
+   (option) == CURLOPT_CONV_FROM_NETWORK_FUNCTION ||                          \
+   (option) == CURLOPT_CONV_FROM_UTF8_FUNCTION)
+
+/* evaluates to true if option takes a data argument to pass to a callback */
+#define _curl_is_cb_data_option(option)                                       \
+  ((option) == CURLOPT_CHUNK_DATA ||                                          \
+   (option) == CURLOPT_CLOSESOCKETDATA ||                                     \
+   (option) == CURLOPT_DEBUGDATA ||                                           \
+   (option) == CURLOPT_FNMATCH_DATA ||                                        \
+   (option) == CURLOPT_HEADERDATA ||                                          \
+   (option) == CURLOPT_INTERLEAVEDATA ||                                      \
+   (option) == CURLOPT_IOCTLDATA ||                                           \
+   (option) == CURLOPT_OPENSOCKETDATA ||                                      \
+   (option) == CURLOPT_PRIVATE ||                                             \
+   (option) == CURLOPT_PROGRESSDATA ||                                        \
+   (option) == CURLOPT_READDATA ||                                            \
+   (option) == CURLOPT_SEEKDATA ||                                            \
+   (option) == CURLOPT_SOCKOPTDATA ||                                         \
+   (option) == CURLOPT_SSH_KEYDATA ||                                         \
+   (option) == CURLOPT_SSL_CTX_DATA ||                                        \
+   (option) == CURLOPT_WRITEDATA ||                                           \
+   0)
+
+/* evaluates to true if option takes a POST data argument (void* or char*) */
+#define _curl_is_postfields_option(option)                                    \
+  ((option) == CURLOPT_POSTFIELDS ||                                          \
+   (option) == CURLOPT_COPYPOSTFIELDS ||                                      \
+   0)
+
+/* evaluates to true if option takes a struct curl_slist * argument */
+#define _curl_is_slist_option(option)                                         \
+  ((option) == CURLOPT_HTTP200ALIASES ||                                      \
+   (option) == CURLOPT_HTTPHEADER ||                                          \
+   (option) == CURLOPT_MAIL_RCPT ||                                           \
+   (option) == CURLOPT_POSTQUOTE ||                                           \
+   (option) == CURLOPT_PREQUOTE ||                                            \
+   (option) == CURLOPT_PROXYHEADER ||                                         \
+   (option) == CURLOPT_QUOTE ||                                               \
+   (option) == CURLOPT_RESOLVE ||                                             \
+   (option) == CURLOPT_TELNETOPTIONS ||                                       \
+   0)
+
+/* groups of curl_easy_getinfo infos that take the same type of argument */
+
+/* evaluates to true if info expects a pointer to char * argument */
+#define _curl_is_string_info(info)                                            \
+  (CURLINFO_STRING < (info) && (info) < CURLINFO_LONG)
+
+/* evaluates to true if info expects a pointer to long argument */
+#define _curl_is_long_info(info)                                              \
+  (CURLINFO_LONG < (info) && (info) < CURLINFO_DOUBLE)
+
+/* evaluates to true if info expects a pointer to double argument */
+#define _curl_is_double_info(info)                                            \
+  (CURLINFO_DOUBLE < (info) && (info) < CURLINFO_SLIST)
+
+/* true if info expects a pointer to struct curl_slist * argument */
+#define _curl_is_slist_info(info)                                       \
+  (((info) == CURLINFO_SSL_ENGINES) || ((info) == CURLINFO_COOKIELIST))
+
+/* true if info expects a pointer to struct curl_tlssessioninfo * argument */
+#define _curl_is_tlssessioninfo_info(info)                              \
+  (((info) == CURLINFO_TLS_SSL_PTR) || ((info) == CURLINFO_TLS_SESSION))
+
+/* true if info expects a pointer to struct curl_certinfo * argument */
+#define _curl_is_certinfo_info(info) ((info) == CURLINFO_CERTINFO)
+
+/* true if info expects a pointer to struct curl_socket_t argument */
+#define _curl_is_socket_info(info)                                            \
+  (CURLINFO_SOCKET < (info) && (info) < CURLINFO_OFF_T)
+
+/* true if info expects a pointer to curl_off_t argument */
+#define _curl_is_off_t_info(info)                                             \
+  (CURLINFO_OFF_T < (info))
+
+
+/* typecheck helpers -- check whether given expression has requested type*/
+
+/* For pointers, you can use the _curl_is_ptr/_curl_is_arr macros,
+ * otherwise define a new macro. Search for __builtin_types_compatible_p
+ * in the GCC manual.
+ * NOTE: these macros MUST NOT EVALUATE their arguments! The argument is
+ * the actual expression passed to the curl_easy_setopt macro. This
+ * means that you can only apply the sizeof and __typeof__ operators, no
+ * == or whatsoever.
+ */
+
+/* XXX: should evaluate to true iff expr is a pointer */
+#define _curl_is_any_ptr(expr)                                                \
+  (sizeof(expr) == sizeof(void *))
+
+/* evaluates to true if expr is NULL */
+/* XXX: must not evaluate expr, so this check is not accurate */
+#define _curl_is_NULL(expr)                                                   \
+  (__builtin_types_compatible_p(__typeof__(expr), __typeof__(NULL)))
+
+/* evaluates to true if expr is type*, const type* or NULL */
+#define _curl_is_ptr(expr, type)                                              \
+  (_curl_is_NULL(expr) ||                                                     \
+   __builtin_types_compatible_p(__typeof__(expr), type *) ||                  \
+   __builtin_types_compatible_p(__typeof__(expr), const type *))
+
+/* evaluates to true if expr is one of type[], type*, NULL or const type* */
+#define _curl_is_arr(expr, type)                                              \
+  (_curl_is_ptr((expr), type) ||                                              \
+   __builtin_types_compatible_p(__typeof__(expr), type []))
+
+/* evaluates to true if expr is a string */
+#define _curl_is_string(expr)                                                 \
+  (_curl_is_arr((expr), char) ||                                              \
+   _curl_is_arr((expr), signed char) ||                                       \
+   _curl_is_arr((expr), unsigned char))
+
+/* evaluates to true if expr is a long (no matter the signedness)
+ * XXX: for now, int is also accepted (and therefore short and char, which
+ * are promoted to int when passed to a variadic function) */
+#define _curl_is_long(expr)                                                   \
+  (__builtin_types_compatible_p(__typeof__(expr), long) ||                    \
+   __builtin_types_compatible_p(__typeof__(expr), signed long) ||             \
+   __builtin_types_compatible_p(__typeof__(expr), unsigned long) ||           \
+   __builtin_types_compatible_p(__typeof__(expr), int) ||                     \
+   __builtin_types_compatible_p(__typeof__(expr), signed int) ||              \
+   __builtin_types_compatible_p(__typeof__(expr), unsigned int) ||            \
+   __builtin_types_compatible_p(__typeof__(expr), short) ||                   \
+   __builtin_types_compatible_p(__typeof__(expr), signed short) ||            \
+   __builtin_types_compatible_p(__typeof__(expr), unsigned short) ||          \
+   __builtin_types_compatible_p(__typeof__(expr), char) ||                    \
+   __builtin_types_compatible_p(__typeof__(expr), signed char) ||             \
+   __builtin_types_compatible_p(__typeof__(expr), unsigned char))
+
+/* evaluates to true if expr is of type curl_off_t */
+#define _curl_is_off_t(expr)                                                  \
+  (__builtin_types_compatible_p(__typeof__(expr), curl_off_t))
+
+/* evaluates to true if expr is abuffer suitable for CURLOPT_ERRORBUFFER */
+/* XXX: also check size of an char[] array? */
+#define _curl_is_error_buffer(expr)                                           \
+  (_curl_is_NULL(expr) ||                                                     \
+   __builtin_types_compatible_p(__typeof__(expr), char *) ||                  \
+   __builtin_types_compatible_p(__typeof__(expr), char[]))
+
+/* evaluates to true if expr is of type (const) void* or (const) FILE* */
+#if 0
+#define _curl_is_cb_data(expr)                                                \
+  (_curl_is_ptr((expr), void) ||                                              \
+   _curl_is_ptr((expr), FILE))
+#else /* be less strict */
+#define _curl_is_cb_data(expr)                                                \
+  _curl_is_any_ptr(expr)
+#endif
+
+/* evaluates to true if expr is of type FILE* */
+#define _curl_is_FILE(expr)                                             \
+  (_curl_is_NULL(expr) ||                                              \
+   (__builtin_types_compatible_p(__typeof__(expr), FILE *)))
+
+/* evaluates to true if expr can be passed as POST data (void* or char*) */
+#define _curl_is_postfields(expr)                                             \
+  (_curl_is_ptr((expr), void) ||                                              \
+   _curl_is_arr((expr), char))
+
+/* FIXME: the whole callback checking is messy...
+ * The idea is to tolerate char vs. void and const vs. not const
+ * pointers in arguments at least
+ */
+/* helper: __builtin_types_compatible_p distinguishes between functions and
+ * function pointers, hide it */
+#define _curl_callback_compatible(func, type)                                 \
+  (__builtin_types_compatible_p(__typeof__(func), type) ||                    \
+   __builtin_types_compatible_p(__typeof__(func) *, type))
+
+/* evaluates to true if expr is of type curl_read_callback or "similar" */
+#define _curl_is_read_cb(expr)                                          \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), __typeof__(fread) *) ||                  \
+   _curl_callback_compatible((expr), curl_read_callback) ||                   \
+   _curl_callback_compatible((expr), _curl_read_callback1) ||                 \
+   _curl_callback_compatible((expr), _curl_read_callback2) ||                 \
+   _curl_callback_compatible((expr), _curl_read_callback3) ||                 \
+   _curl_callback_compatible((expr), _curl_read_callback4) ||                 \
+   _curl_callback_compatible((expr), _curl_read_callback5) ||                 \
+   _curl_callback_compatible((expr), _curl_read_callback6))
+typedef size_t (*_curl_read_callback1)(char *, size_t, size_t, void *);
+typedef size_t (*_curl_read_callback2)(char *, size_t, size_t, const void *);
+typedef size_t (*_curl_read_callback3)(char *, size_t, size_t, FILE *);
+typedef size_t (*_curl_read_callback4)(void *, size_t, size_t, void *);
+typedef size_t (*_curl_read_callback5)(void *, size_t, size_t, const void *);
+typedef size_t (*_curl_read_callback6)(void *, size_t, size_t, FILE *);
+
+/* evaluates to true if expr is of type curl_write_callback or "similar" */
+#define _curl_is_write_cb(expr)                                               \
+  (_curl_is_read_cb(expr) ||                                            \
+   _curl_callback_compatible((expr), __typeof__(fwrite) *) ||                 \
+   _curl_callback_compatible((expr), curl_write_callback) ||                  \
+   _curl_callback_compatible((expr), _curl_write_callback1) ||                \
+   _curl_callback_compatible((expr), _curl_write_callback2) ||                \
+   _curl_callback_compatible((expr), _curl_write_callback3) ||                \
+   _curl_callback_compatible((expr), _curl_write_callback4) ||                \
+   _curl_callback_compatible((expr), _curl_write_callback5) ||                \
+   _curl_callback_compatible((expr), _curl_write_callback6))
+typedef size_t (*_curl_write_callback1)(const char *, size_t, size_t, void *);
+typedef size_t (*_curl_write_callback2)(const char *, size_t, size_t,
+                                       const void *);
+typedef size_t (*_curl_write_callback3)(const char *, size_t, size_t, FILE *);
+typedef size_t (*_curl_write_callback4)(const void *, size_t, size_t, void *);
+typedef size_t (*_curl_write_callback5)(const void *, size_t, size_t,
+                                       const void *);
+typedef size_t (*_curl_write_callback6)(const void *, size_t, size_t, FILE *);
+
+/* evaluates to true if expr is of type curl_ioctl_callback or "similar" */
+#define _curl_is_ioctl_cb(expr)                                         \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_ioctl_callback) ||                  \
+   _curl_callback_compatible((expr), _curl_ioctl_callback1) ||                \
+   _curl_callback_compatible((expr), _curl_ioctl_callback2) ||                \
+   _curl_callback_compatible((expr), _curl_ioctl_callback3) ||                \
+   _curl_callback_compatible((expr), _curl_ioctl_callback4))
+typedef curlioerr (*_curl_ioctl_callback1)(CURL *, int, void *);
+typedef curlioerr (*_curl_ioctl_callback2)(CURL *, int, const void *);
+typedef curlioerr (*_curl_ioctl_callback3)(CURL *, curliocmd, void *);
+typedef curlioerr (*_curl_ioctl_callback4)(CURL *, curliocmd, const void *);
+
+/* evaluates to true if expr is of type curl_sockopt_callback or "similar" */
+#define _curl_is_sockopt_cb(expr)                                       \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_sockopt_callback) ||                \
+   _curl_callback_compatible((expr), _curl_sockopt_callback1) ||              \
+   _curl_callback_compatible((expr), _curl_sockopt_callback2))
+typedef int (*_curl_sockopt_callback1)(void *, curl_socket_t, curlsocktype);
+typedef int (*_curl_sockopt_callback2)(const void *, curl_socket_t,
+                                      curlsocktype);
+
+/* evaluates to true if expr is of type curl_opensocket_callback or
+   "similar" */
+#define _curl_is_opensocket_cb(expr)                                    \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_opensocket_callback) ||             \
+   _curl_callback_compatible((expr), _curl_opensocket_callback1) ||           \
+   _curl_callback_compatible((expr), _curl_opensocket_callback2) ||           \
+   _curl_callback_compatible((expr), _curl_opensocket_callback3) ||           \
+   _curl_callback_compatible((expr), _curl_opensocket_callback4))
+typedef curl_socket_t (*_curl_opensocket_callback1)
+  (void *, curlsocktype, struct curl_sockaddr *);
+typedef curl_socket_t (*_curl_opensocket_callback2)
+  (void *, curlsocktype, const struct curl_sockaddr *);
+typedef curl_socket_t (*_curl_opensocket_callback3)
+  (const void *, curlsocktype, struct curl_sockaddr *);
+typedef curl_socket_t (*_curl_opensocket_callback4)
+  (const void *, curlsocktype, const struct curl_sockaddr *);
+
+/* evaluates to true if expr is of type curl_progress_callback or "similar" */
+#define _curl_is_progress_cb(expr)                                      \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_progress_callback) ||               \
+   _curl_callback_compatible((expr), _curl_progress_callback1) ||             \
+   _curl_callback_compatible((expr), _curl_progress_callback2))
+typedef int (*_curl_progress_callback1)(void *,
+    double, double, double, double);
+typedef int (*_curl_progress_callback2)(const void *,
+    double, double, double, double);
+
+/* evaluates to true if expr is of type curl_debug_callback or "similar" */
+#define _curl_is_debug_cb(expr)                                         \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_debug_callback) ||                  \
+   _curl_callback_compatible((expr), _curl_debug_callback1) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback2) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback3) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback4) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback5) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback6) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback7) ||                \
+   _curl_callback_compatible((expr), _curl_debug_callback8))
+typedef int (*_curl_debug_callback1) (CURL *,
+    curl_infotype, char *, size_t, void *);
+typedef int (*_curl_debug_callback2) (CURL *,
+    curl_infotype, char *, size_t, const void *);
+typedef int (*_curl_debug_callback3) (CURL *,
+    curl_infotype, const char *, size_t, void *);
+typedef int (*_curl_debug_callback4) (CURL *,
+    curl_infotype, const char *, size_t, const void *);
+typedef int (*_curl_debug_callback5) (CURL *,
+    curl_infotype, unsigned char *, size_t, void *);
+typedef int (*_curl_debug_callback6) (CURL *,
+    curl_infotype, unsigned char *, size_t, const void *);
+typedef int (*_curl_debug_callback7) (CURL *,
+    curl_infotype, const unsigned char *, size_t, void *);
+typedef int (*_curl_debug_callback8) (CURL *,
+    curl_infotype, const unsigned char *, size_t, const void *);
+
+/* evaluates to true if expr is of type curl_ssl_ctx_callback or "similar" */
+/* this is getting even messier... */
+#define _curl_is_ssl_ctx_cb(expr)                                       \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_ssl_ctx_callback) ||                \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback1) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback2) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback3) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback4) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback5) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback6) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback7) ||              \
+   _curl_callback_compatible((expr), _curl_ssl_ctx_callback8))
+typedef CURLcode (*_curl_ssl_ctx_callback1)(CURL *, void *, void *);
+typedef CURLcode (*_curl_ssl_ctx_callback2)(CURL *, void *, const void *);
+typedef CURLcode (*_curl_ssl_ctx_callback3)(CURL *, const void *, void *);
+typedef CURLcode (*_curl_ssl_ctx_callback4)(CURL *, const void *,
+                                            const void *);
+#ifdef HEADER_SSL_H
+/* hack: if we included OpenSSL's ssl.h, we know about SSL_CTX
+ * this will of course break if we're included before OpenSSL headers...
+ */
+typedef CURLcode (*_curl_ssl_ctx_callback5)(CURL *, SSL_CTX, void *);
+typedef CURLcode (*_curl_ssl_ctx_callback6)(CURL *, SSL_CTX, const void *);
+typedef CURLcode (*_curl_ssl_ctx_callback7)(CURL *, const SSL_CTX, void *);
+typedef CURLcode (*_curl_ssl_ctx_callback8)(CURL *, const SSL_CTX,
+                                           const void *);
+#else
+typedef _curl_ssl_ctx_callback1 _curl_ssl_ctx_callback5;
+typedef _curl_ssl_ctx_callback1 _curl_ssl_ctx_callback6;
+typedef _curl_ssl_ctx_callback1 _curl_ssl_ctx_callback7;
+typedef _curl_ssl_ctx_callback1 _curl_ssl_ctx_callback8;
+#endif
+
+/* evaluates to true if expr is of type curl_conv_callback or "similar" */
+#define _curl_is_conv_cb(expr)                                          \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_conv_callback) ||                   \
+   _curl_callback_compatible((expr), _curl_conv_callback1) ||                 \
+   _curl_callback_compatible((expr), _curl_conv_callback2) ||                 \
+   _curl_callback_compatible((expr), _curl_conv_callback3) ||                 \
+   _curl_callback_compatible((expr), _curl_conv_callback4))
+typedef CURLcode (*_curl_conv_callback1)(char *, size_t length);
+typedef CURLcode (*_curl_conv_callback2)(const char *, size_t length);
+typedef CURLcode (*_curl_conv_callback3)(void *, size_t length);
+typedef CURLcode (*_curl_conv_callback4)(const void *, size_t length);
+
+/* evaluates to true if expr is of type curl_seek_callback or "similar" */
+#define _curl_is_seek_cb(expr)                                          \
+  (_curl_is_NULL(expr) ||                                                     \
+   _curl_callback_compatible((expr), curl_seek_callback) ||                   \
+   _curl_callback_compatible((expr), _curl_seek_callback1) ||                 \
+   _curl_callback_compatible((expr), _curl_seek_callback2))
+typedef CURLcode (*_curl_seek_callback1)(void *, curl_off_t, int);
+typedef CURLcode (*_curl_seek_callback2)(const void *, curl_off_t, int);
+
+
+#endif /* __CURL_TYPECHECK_GCC_H */
diff --git a/distribution/curl/lib/arm64-v8a/libcurl.a b/distribution/curl/lib/arm64-v8a/libcurl.a
new file mode 100644
index 0000000..a46eab3
Binary files /dev/null and b/distribution/curl/lib/arm64-v8a/libcurl.a differ
diff --git a/distribution/curl/lib/armeabi-v7a/libcurl.a b/distribution/curl/lib/armeabi-v7a/libcurl.a
new file mode 100644
index 0000000..b36a50b
Binary files /dev/null and b/distribution/curl/lib/armeabi-v7a/libcurl.a differ
diff --git a/distribution/curl/lib/x86/libcurl.a b/distribution/curl/lib/x86/libcurl.a
new file mode 100644
index 0000000..1e74f26
Binary files /dev/null and b/distribution/curl/lib/x86/libcurl.a differ
diff --git a/distribution/curl/lib/x86_64/libcurl.a b/distribution/curl/lib/x86_64/libcurl.a
new file mode 100644
index 0000000..cb96aa2
Binary files /dev/null and b/distribution/curl/lib/x86_64/libcurl.a differ
diff --git a/distribution/curl/lib19/armeabi-v7a/libcurl.a b/distribution/curl/lib19/armeabi-v7a/libcurl.a
new file mode 100644
index 0000000..d45ce5a
Binary files /dev/null and b/distribution/curl/lib19/armeabi-v7a/libcurl.a differ
diff --git a/distribution/curl/lib19/x86/libcurl.a b/distribution/curl/lib19/x86/libcurl.a
new file mode 100644
index 0000000..2909674
Binary files /dev/null and b/distribution/curl/lib19/x86/libcurl.a differ
diff --git a/distribution/jansson/include/jansson.h b/distribution/jansson/include/jansson.h
new file mode 100644
index 0000000..a5927bd
--- /dev/null
+++ b/distribution/jansson/include/jansson.h
@@ -0,0 +1,315 @@
+/*
+ * Copyright (c) 2009-2016 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#ifndef JANSSON_H
+#define JANSSON_H
+
+#include <stdio.h>
+#include <stdlib.h>  /* for size_t */
+#include <stdarg.h>
+
+#include "jansson_config.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* version */
+
+#define JANSSON_MAJOR_VERSION  2
+#define JANSSON_MINOR_VERSION  10
+#define JANSSON_MICRO_VERSION  0
+
+/* Micro version is omitted if it's 0 */
+#define JANSSON_VERSION  "2.10"
+
+/* Version as a 3-byte hex number, e.g. 0x010201 == 1.2.1. Use this
+   for numeric comparisons, e.g. #if JANSSON_VERSION_HEX >= ... */
+#define JANSSON_VERSION_HEX  ((JANSSON_MAJOR_VERSION << 16) |   \
+                              (JANSSON_MINOR_VERSION << 8)  |   \
+                              (JANSSON_MICRO_VERSION << 0))
+
+
+/* types */
+
+typedef enum {
+    JSON_OBJECT,
+    JSON_ARRAY,
+    JSON_STRING,
+    JSON_INTEGER,
+    JSON_REAL,
+    JSON_TRUE,
+    JSON_FALSE,
+    JSON_NULL
+} json_type;
+
+typedef struct json_t {
+    json_type type;
+    size_t refcount;
+} json_t;
+
+#ifndef JANSSON_USING_CMAKE /* disabled if using cmake */
+#if JSON_INTEGER_IS_LONG_LONG
+#ifdef _WIN32
+#define JSON_INTEGER_FORMAT "I64d"
+#else
+#define JSON_INTEGER_FORMAT "lld"
+#endif
+typedef long long json_int_t;
+#else
+#define JSON_INTEGER_FORMAT "ld"
+typedef long json_int_t;
+#endif /* JSON_INTEGER_IS_LONG_LONG */
+#endif
+
+#define json_typeof(json)      ((json)->type)
+#define json_is_object(json)   ((json) && json_typeof(json) == JSON_OBJECT)
+#define json_is_array(json)    ((json) && json_typeof(json) == JSON_ARRAY)
+#define json_is_string(json)   ((json) && json_typeof(json) == JSON_STRING)
+#define json_is_integer(json)  ((json) && json_typeof(json) == JSON_INTEGER)
+#define json_is_real(json)     ((json) && json_typeof(json) == JSON_REAL)
+#define json_is_number(json)   (json_is_integer(json) || json_is_real(json))
+#define json_is_true(json)     ((json) && json_typeof(json) == JSON_TRUE)
+#define json_is_false(json)    ((json) && json_typeof(json) == JSON_FALSE)
+#define json_boolean_value     json_is_true
+#define json_is_boolean(json)  (json_is_true(json) || json_is_false(json))
+#define json_is_null(json)     ((json) && json_typeof(json) == JSON_NULL)
+
+/* construction, destruction, reference counting */
+
+json_t *json_object(void);
+json_t *json_array(void);
+json_t *json_string(const char *value);
+json_t *json_stringn(const char *value, size_t len);
+json_t *json_string_nocheck(const char *value);
+json_t *json_stringn_nocheck(const char *value, size_t len);
+json_t *json_integer(json_int_t value);
+json_t *json_real(double value);
+json_t *json_true(void);
+json_t *json_false(void);
+#define json_boolean(val)      ((val) ? json_true() : json_false())
+json_t *json_null(void);
+
+static JSON_INLINE
+json_t *json_incref(json_t *json)
+{
+    if(json && json->refcount != (size_t)-1)
+        ++json->refcount;
+    return json;
+}
+
+/* do not call json_delete directly */
+void json_delete(json_t *json);
+
+static JSON_INLINE
+void json_decref(json_t *json)
+{
+    if(json && json->refcount != (size_t)-1 && --json->refcount == 0)
+        json_delete(json);
+}
+
+#if defined(__GNUC__) || defined(__clang__)
+static JSON_INLINE
+void json_decrefp(json_t **json)
+{
+    if(json) {
+        json_decref(*json);
+	*json = NULL;
+    }
+}
+
+#define json_auto_t json_t __attribute__((cleanup(json_decrefp)))
+#endif
+
+
+/* error reporting */
+
+#define JSON_ERROR_TEXT_LENGTH    160
+#define JSON_ERROR_SOURCE_LENGTH   80
+
+typedef struct {
+    int line;
+    int column;
+    int position;
+    char source[JSON_ERROR_SOURCE_LENGTH];
+    char text[JSON_ERROR_TEXT_LENGTH];
+} json_error_t;
+
+
+/* getters, setters, manipulation */
+
+void json_object_seed(size_t seed);
+size_t json_object_size(const json_t *object);
+json_t *json_object_get(const json_t *object, const char *key);
+int json_object_set_new(json_t *object, const char *key, json_t *value);
+int json_object_set_new_nocheck(json_t *object, const char *key, json_t *value);
+int json_object_del(json_t *object, const char *key);
+int json_object_clear(json_t *object);
+int json_object_update(json_t *object, json_t *other);
+int json_object_update_existing(json_t *object, json_t *other);
+int json_object_update_missing(json_t *object, json_t *other);
+void *json_object_iter(json_t *object);
+void *json_object_iter_at(json_t *object, const char *key);
+void *json_object_key_to_iter(const char *key);
+void *json_object_iter_next(json_t *object, void *iter);
+const char *json_object_iter_key(void *iter);
+json_t *json_object_iter_value(void *iter);
+int json_object_iter_set_new(json_t *object, void *iter, json_t *value);
+
+#define json_object_foreach(object, key, value) \
+    for(key = json_object_iter_key(json_object_iter(object)); \
+        key && (value = json_object_iter_value(json_object_key_to_iter(key))); \
+        key = json_object_iter_key(json_object_iter_next(object, json_object_key_to_iter(key))))
+
+#define json_object_foreach_safe(object, n, key, value)     \
+    for(key = json_object_iter_key(json_object_iter(object)), \
+            n = json_object_iter_next(object, json_object_key_to_iter(key)); \
+        key && (value = json_object_iter_value(json_object_key_to_iter(key))); \
+        key = json_object_iter_key(n), \
+            n = json_object_iter_next(object, json_object_key_to_iter(key)))
+
+#define json_array_foreach(array, index, value) \
+	for(index = 0; \
+		index < json_array_size(array) && (value = json_array_get(array, index)); \
+		index++)
+
+static JSON_INLINE
+int json_object_set(json_t *object, const char *key, json_t *value)
+{
+    return json_object_set_new(object, key, json_incref(value));
+}
+
+static JSON_INLINE
+int json_object_set_nocheck(json_t *object, const char *key, json_t *value)
+{
+    return json_object_set_new_nocheck(object, key, json_incref(value));
+}
+
+static JSON_INLINE
+int json_object_iter_set(json_t *object, void *iter, json_t *value)
+{
+    return json_object_iter_set_new(object, iter, json_incref(value));
+}
+
+size_t json_array_size(const json_t *array);
+json_t *json_array_get(const json_t *array, size_t index);
+int json_array_set_new(json_t *array, size_t index, json_t *value);
+int json_array_append_new(json_t *array, json_t *value);
+int json_array_insert_new(json_t *array, size_t index, json_t *value);
+int json_array_remove(json_t *array, size_t index);
+int json_array_clear(json_t *array);
+int json_array_extend(json_t *array, json_t *other);
+
+static JSON_INLINE
+int json_array_set(json_t *array, size_t ind, json_t *value)
+{
+    return json_array_set_new(array, ind, json_incref(value));
+}
+
+static JSON_INLINE
+int json_array_append(json_t *array, json_t *value)
+{
+    return json_array_append_new(array, json_incref(value));
+}
+
+static JSON_INLINE
+int json_array_insert(json_t *array, size_t ind, json_t *value)
+{
+    return json_array_insert_new(array, ind, json_incref(value));
+}
+
+const char *json_string_value(const json_t *string);
+size_t json_string_length(const json_t *string);
+json_int_t json_integer_value(const json_t *integer);
+double json_real_value(const json_t *real);
+double json_number_value(const json_t *json);
+
+int json_string_set(json_t *string, const char *value);
+int json_string_setn(json_t *string, const char *value, size_t len);
+int json_string_set_nocheck(json_t *string, const char *value);
+int json_string_setn_nocheck(json_t *string, const char *value, size_t len);
+int json_integer_set(json_t *integer, json_int_t value);
+int json_real_set(json_t *real, double value);
+
+/* pack, unpack */
+
+json_t *json_pack(const char *fmt, ...);
+json_t *json_pack_ex(json_error_t *error, size_t flags, const char *fmt, ...);
+json_t *json_vpack_ex(json_error_t *error, size_t flags, const char *fmt, va_list ap);
+
+#define JSON_VALIDATE_ONLY  0x1
+#define JSON_STRICT         0x2
+
+int json_unpack(json_t *root, const char *fmt, ...);
+int json_unpack_ex(json_t *root, json_error_t *error, size_t flags, const char *fmt, ...);
+int json_vunpack_ex(json_t *root, json_error_t *error, size_t flags, const char *fmt, va_list ap);
+
+
+/* equality */
+
+int json_equal(json_t *value1, json_t *value2);
+
+
+/* copying */
+
+json_t *json_copy(json_t *value);
+json_t *json_deep_copy(const json_t *value);
+
+
+/* decoding */
+
+#define JSON_REJECT_DUPLICATES  0x1
+#define JSON_DISABLE_EOF_CHECK  0x2
+#define JSON_DECODE_ANY         0x4
+#define JSON_DECODE_INT_AS_REAL 0x8
+#define JSON_ALLOW_NUL          0x10
+
+typedef size_t (*json_load_callback_t)(void *buffer, size_t buflen, void *data);
+
+json_t *json_loads(const char *input, size_t flags, json_error_t *error);
+json_t *json_loadb(const char *buffer, size_t buflen, size_t flags, json_error_t *error);
+json_t *json_loadf(FILE *input, size_t flags, json_error_t *error);
+json_t *json_loadfd(int input, size_t flags, json_error_t *error);
+json_t *json_load_file(const char *path, size_t flags, json_error_t *error);
+json_t *json_load_callback(json_load_callback_t callback, void *data, size_t flags, json_error_t *error);
+
+
+/* encoding */
+
+#define JSON_MAX_INDENT         0x1F
+#define JSON_INDENT(n)          ((n) & JSON_MAX_INDENT)
+#define JSON_COMPACT            0x20
+#define JSON_ENSURE_ASCII       0x40
+#define JSON_SORT_KEYS          0x80
+#define JSON_PRESERVE_ORDER     0x100
+#define JSON_ENCODE_ANY         0x200
+#define JSON_ESCAPE_SLASH       0x400
+#define JSON_REAL_PRECISION(n)  (((n) & 0x1F) << 11)
+#define JSON_EMBED              0x10000
+
+typedef int (*json_dump_callback_t)(const char *buffer, size_t size, void *data);
+
+char *json_dumps(const json_t *json, size_t flags);
+size_t json_dumpb(const json_t *json, char *buffer, size_t size, size_t flags);
+int json_dumpf(const json_t *json, FILE *output, size_t flags);
+int json_dumpfd(const json_t *json, int output, size_t flags);
+int json_dump_file(const json_t *json, const char *path, size_t flags);
+int json_dump_callback(const json_t *json, json_dump_callback_t callback, void *data, size_t flags);
+
+/* custom memory allocation */
+
+typedef void *(*json_malloc_t)(size_t);
+typedef void (*json_free_t)(void *);
+
+void json_set_alloc_funcs(json_malloc_t malloc_fn, json_free_t free_fn);
+void json_get_alloc_funcs(json_malloc_t *malloc_fn, json_free_t *free_fn);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/distribution/jansson/include/jansson_config.h b/distribution/jansson/include/jansson_config.h
new file mode 100644
index 0000000..f1a5ddd
--- /dev/null
+++ b/distribution/jansson/include/jansson_config.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright (c) 2010-2016 Petri Lehtinen <petri@digip.org>
+ *
+ * Jansson is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ *
+ *
+ * This file specifies a part of the site-specific configuration for
+ * Jansson, namely those things that affect the public API in
+ * jansson.h.
+ *
+ * The configure script copies this file to jansson_config.h and
+ * replaces @var@ substitutions by values that fit your system. If you
+ * cannot run the configure script, you can do the value substitution
+ * by hand.
+ */
+
+#ifndef JANSSON_CONFIG_H
+#define JANSSON_CONFIG_H
+
+/* If your compiler supports the inline keyword in C, JSON_INLINE is
+   defined to `inline', otherwise empty. In C++, the inline is always
+   supported. */
+#ifdef __cplusplus
+#define JSON_INLINE inline
+#else
+#define JSON_INLINE inline
+#endif
+
+/* If your compiler supports the `long long` type and the strtoll()
+   library function, JSON_INTEGER_IS_LONG_LONG is defined to 1,
+   otherwise to 0. */
+#define JSON_INTEGER_IS_LONG_LONG 1
+
+/* If locale.h and localeconv() are available, define to 1,
+   otherwise to 0. */
+#define JSON_HAVE_LOCALECONV 1
+
+/* Maximum recursion depth for parsing JSON input.
+   This limits the depth of e.g. array-within-array constructions. */
+#define JSON_PARSER_MAX_DEPTH 2048
+
+#endif
diff --git a/distribution/jansson/lib/arm64-v8a/libjansson.a b/distribution/jansson/lib/arm64-v8a/libjansson.a
new file mode 100644
index 0000000..5edfabf
Binary files /dev/null and b/distribution/jansson/lib/arm64-v8a/libjansson.a differ
diff --git a/distribution/jansson/lib/armeabi-v7a/libjansson.a b/distribution/jansson/lib/armeabi-v7a/libjansson.a
new file mode 100644
index 0000000..80b0ecd
Binary files /dev/null and b/distribution/jansson/lib/armeabi-v7a/libjansson.a differ
diff --git a/distribution/jansson/lib/x86/libjansson.a b/distribution/jansson/lib/x86/libjansson.a
new file mode 100644
index 0000000..7a49242
Binary files /dev/null and b/distribution/jansson/lib/x86/libjansson.a differ
diff --git a/distribution/jansson/lib/x86_64/libjansson.a b/distribution/jansson/lib/x86_64/libjansson.a
new file mode 100644
index 0000000..b5fedab
Binary files /dev/null and b/distribution/jansson/lib/x86_64/libjansson.a differ
diff --git a/distribution/jansson/lib19/armeabi-v7a/libjansson.a b/distribution/jansson/lib19/armeabi-v7a/libjansson.a
new file mode 100644
index 0000000..e07b951
Binary files /dev/null and b/distribution/jansson/lib19/armeabi-v7a/libjansson.a differ
diff --git a/distribution/jansson/lib19/x86/libjansson.a b/distribution/jansson/lib19/x86/libjansson.a
new file mode 100644
index 0000000..37c0517
Binary files /dev/null and b/distribution/jansson/lib19/x86/libjansson.a differ
diff --git a/distribution/zlib/include/zconf.h b/distribution/zlib/include/zconf.h
new file mode 100644
index 0000000..77398c1
--- /dev/null
+++ b/distribution/zlib/include/zconf.h
@@ -0,0 +1,534 @@
+/* zconf.h -- configuration of the zlib compression library
+ * Copyright (C) 1995-2016 Jean-loup Gailly, Mark Adler
+ * For conditions of distribution and use, see copyright notice in zlib.h
+ */
+
+/* @(#) $Id$ */
+
+#ifndef ZCONF_H
+#define ZCONF_H
+
+/*
+ * If you *really* need a unique prefix for all types and library functions,
+ * compile with -DZ_PREFIX. The "standard" zlib should be compiled without it.
+ * Even better than compiling with -DZ_PREFIX would be to use configure to set
+ * this permanently in zconf.h using "./configure --zprefix".
+ */
+#ifdef Z_PREFIX     /* may be set to #if 1 by ./configure */
+#  define Z_PREFIX_SET
+
+/* all linked symbols and init macros */
+#  define _dist_code            z__dist_code
+#  define _length_code          z__length_code
+#  define _tr_align             z__tr_align
+#  define _tr_flush_bits        z__tr_flush_bits
+#  define _tr_flush_block       z__tr_flush_block
+#  define _tr_init              z__tr_init
+#  define _tr_stored_block      z__tr_stored_block
+#  define _tr_tally             z__tr_tally
+#  define adler32               z_adler32
+#  define adler32_combine       z_adler32_combine
+#  define adler32_combine64     z_adler32_combine64
+#  define adler32_z             z_adler32_z
+#  ifndef Z_SOLO
+#    define compress              z_compress
+#    define compress2             z_compress2
+#    define compressBound         z_compressBound
+#  endif
+#  define crc32                 z_crc32
+#  define crc32_combine         z_crc32_combine
+#  define crc32_combine64       z_crc32_combine64
+#  define crc32_z               z_crc32_z
+#  define deflate               z_deflate
+#  define deflateBound          z_deflateBound
+#  define deflateCopy           z_deflateCopy
+#  define deflateEnd            z_deflateEnd
+#  define deflateGetDictionary  z_deflateGetDictionary
+#  define deflateInit           z_deflateInit
+#  define deflateInit2          z_deflateInit2
+#  define deflateInit2_         z_deflateInit2_
+#  define deflateInit_          z_deflateInit_
+#  define deflateParams         z_deflateParams
+#  define deflatePending        z_deflatePending
+#  define deflatePrime          z_deflatePrime
+#  define deflateReset          z_deflateReset
+#  define deflateResetKeep      z_deflateResetKeep
+#  define deflateSetDictionary  z_deflateSetDictionary
+#  define deflateSetHeader      z_deflateSetHeader
+#  define deflateTune           z_deflateTune
+#  define deflate_copyright     z_deflate_copyright
+#  define get_crc_table         z_get_crc_table
+#  ifndef Z_SOLO
+#    define gz_error              z_gz_error
+#    define gz_intmax             z_gz_intmax
+#    define gz_strwinerror        z_gz_strwinerror
+#    define gzbuffer              z_gzbuffer
+#    define gzclearerr            z_gzclearerr
+#    define gzclose               z_gzclose
+#    define gzclose_r             z_gzclose_r
+#    define gzclose_w             z_gzclose_w
+#    define gzdirect              z_gzdirect
+#    define gzdopen               z_gzdopen
+#    define gzeof                 z_gzeof
+#    define gzerror               z_gzerror
+#    define gzflush               z_gzflush
+#    define gzfread               z_gzfread
+#    define gzfwrite              z_gzfwrite
+#    define gzgetc                z_gzgetc
+#    define gzgetc_               z_gzgetc_
+#    define gzgets                z_gzgets
+#    define gzoffset              z_gzoffset
+#    define gzoffset64            z_gzoffset64
+#    define gzopen                z_gzopen
+#    define gzopen64              z_gzopen64
+#    ifdef _WIN32
+#      define gzopen_w              z_gzopen_w
+#    endif
+#    define gzprintf              z_gzprintf
+#    define gzputc                z_gzputc
+#    define gzputs                z_gzputs
+#    define gzread                z_gzread
+#    define gzrewind              z_gzrewind
+#    define gzseek                z_gzseek
+#    define gzseek64              z_gzseek64
+#    define gzsetparams           z_gzsetparams
+#    define gztell                z_gztell
+#    define gztell64              z_gztell64
+#    define gzungetc              z_gzungetc
+#    define gzvprintf             z_gzvprintf
+#    define gzwrite               z_gzwrite
+#  endif
+#  define inflate               z_inflate
+#  define inflateBack           z_inflateBack
+#  define inflateBackEnd        z_inflateBackEnd
+#  define inflateBackInit       z_inflateBackInit
+#  define inflateBackInit_      z_inflateBackInit_
+#  define inflateCodesUsed      z_inflateCodesUsed
+#  define inflateCopy           z_inflateCopy
+#  define inflateEnd            z_inflateEnd
+#  define inflateGetDictionary  z_inflateGetDictionary
+#  define inflateGetHeader      z_inflateGetHeader
+#  define inflateInit           z_inflateInit
+#  define inflateInit2          z_inflateInit2
+#  define inflateInit2_         z_inflateInit2_
+#  define inflateInit_          z_inflateInit_
+#  define inflateMark           z_inflateMark
+#  define inflatePrime          z_inflatePrime
+#  define inflateReset          z_inflateReset
+#  define inflateReset2         z_inflateReset2
+#  define inflateResetKeep      z_inflateResetKeep
+#  define inflateSetDictionary  z_inflateSetDictionary
+#  define inflateSync           z_inflateSync
+#  define inflateSyncPoint      z_inflateSyncPoint
+#  define inflateUndermine      z_inflateUndermine
+#  define inflateValidate       z_inflateValidate
+#  define inflate_copyright     z_inflate_copyright
+#  define inflate_fast          z_inflate_fast
+#  define inflate_table         z_inflate_table
+#  ifndef Z_SOLO
+#    define uncompress            z_uncompress
+#    define uncompress2           z_uncompress2
+#  endif
+#  define zError                z_zError
+#  ifndef Z_SOLO
+#    define zcalloc               z_zcalloc
+#    define zcfree                z_zcfree
+#  endif
+#  define zlibCompileFlags      z_zlibCompileFlags
+#  define zlibVersion           z_zlibVersion
+
+/* all zlib typedefs in zlib.h and zconf.h */
+#  define Byte                  z_Byte
+#  define Bytef                 z_Bytef
+#  define alloc_func            z_alloc_func
+#  define charf                 z_charf
+#  define free_func             z_free_func
+#  ifndef Z_SOLO
+#    define gzFile                z_gzFile
+#  endif
+#  define gz_header             z_gz_header
+#  define gz_headerp            z_gz_headerp
+#  define in_func               z_in_func
+#  define intf                  z_intf
+#  define out_func              z_out_func
+#  define uInt                  z_uInt
+#  define uIntf                 z_uIntf
+#  define uLong                 z_uLong
+#  define uLongf                z_uLongf
+#  define voidp                 z_voidp
+#  define voidpc                z_voidpc
+#  define voidpf                z_voidpf
+
+/* all zlib structs in zlib.h and zconf.h */
+#  define gz_header_s           z_gz_header_s
+#  define internal_state        z_internal_state
+
+#endif
+
+#if defined(__MSDOS__) && !defined(MSDOS)
+#  define MSDOS
+#endif
+#if (defined(OS_2) || defined(__OS2__)) && !defined(OS2)
+#  define OS2
+#endif
+#if defined(_WINDOWS) && !defined(WINDOWS)
+#  define WINDOWS
+#endif
+#if defined(_WIN32) || defined(_WIN32_WCE) || defined(__WIN32__)
+#  ifndef WIN32
+#    define WIN32
+#  endif
+#endif
+#if (defined(MSDOS) || defined(OS2) || defined(WINDOWS)) && !defined(WIN32)
+#  if !defined(__GNUC__) && !defined(__FLAT__) && !defined(__386__)
+#    ifndef SYS16BIT
+#      define SYS16BIT
+#    endif
+#  endif
+#endif
+
+/*
+ * Compile with -DMAXSEG_64K if the alloc function cannot allocate more
+ * than 64k bytes at a time (needed on systems with 16-bit int).
+ */
+#ifdef SYS16BIT
+#  define MAXSEG_64K
+#endif
+#ifdef MSDOS
+#  define UNALIGNED_OK
+#endif
+
+#ifdef __STDC_VERSION__
+#  ifndef STDC
+#    define STDC
+#  endif
+#  if __STDC_VERSION__ >= 199901L
+#    ifndef STDC99
+#      define STDC99
+#    endif
+#  endif
+#endif
+#if !defined(STDC) && (defined(__STDC__) || defined(__cplusplus))
+#  define STDC
+#endif
+#if !defined(STDC) && (defined(__GNUC__) || defined(__BORLANDC__))
+#  define STDC
+#endif
+#if !defined(STDC) && (defined(MSDOS) || defined(WINDOWS) || defined(WIN32))
+#  define STDC
+#endif
+#if !defined(STDC) && (defined(OS2) || defined(__HOS_AIX__))
+#  define STDC
+#endif
+
+#if defined(__OS400__) && !defined(STDC)    /* iSeries (formerly AS/400). */
+#  define STDC
+#endif
+
+#ifndef STDC
+#  ifndef const /* cannot use !defined(STDC) && !defined(const) on Mac */
+#    define const       /* note: need a more gentle solution here */
+#  endif
+#endif
+
+#if defined(ZLIB_CONST) && !defined(z_const)
+#  define z_const const
+#else
+#  define z_const
+#endif
+
+#ifdef Z_SOLO
+   typedef unsigned long z_size_t;
+#else
+#  define z_longlong long long
+#  if defined(NO_SIZE_T)
+     typedef unsigned NO_SIZE_T z_size_t;
+#  elif defined(STDC)
+#    include <stddef.h>
+     typedef size_t z_size_t;
+#  else
+     typedef unsigned long z_size_t;
+#  endif
+#  undef z_longlong
+#endif
+
+/* Maximum value for memLevel in deflateInit2 */
+#ifndef MAX_MEM_LEVEL
+#  ifdef MAXSEG_64K
+#    define MAX_MEM_LEVEL 8
+#  else
+#    define MAX_MEM_LEVEL 9
+#  endif
+#endif
+
+/* Maximum value for windowBits in deflateInit2 and inflateInit2.
+ * WARNING: reducing MAX_WBITS makes minigzip unable to extract .gz files
+ * created by gzip. (Files created by minigzip can still be extracted by
+ * gzip.)
+ */
+#ifndef MAX_WBITS
+#  define MAX_WBITS   15 /* 32K LZ77 window */
+#endif
+
+/* The memory requirements for deflate are (in bytes):
+            (1 << (windowBits+2)) +  (1 << (memLevel+9))
+ that is: 128K for windowBits=15  +  128K for memLevel = 8  (default values)
+ plus a few kilobytes for small objects. For example, if you want to reduce
+ the default memory requirements from 256K to 128K, compile with
+     make CFLAGS="-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7"
+ Of course this will generally degrade compression (there's no free lunch).
+
+   The memory requirements for inflate are (in bytes) 1 << windowBits
+ that is, 32K for windowBits=15 (default value) plus about 7 kilobytes
+ for small objects.
+*/
+
+                        /* Type declarations */
+
+#ifndef OF /* function prototypes */
+#  ifdef STDC
+#    define OF(args)  args
+#  else
+#    define OF(args)  ()
+#  endif
+#endif
+
+#ifndef Z_ARG /* function prototypes for stdarg */
+#  if defined(STDC) || defined(Z_HAVE_STDARG_H)
+#    define Z_ARG(args)  args
+#  else
+#    define Z_ARG(args)  ()
+#  endif
+#endif
+
+/* The following definitions for FAR are needed only for MSDOS mixed
+ * model programming (small or medium model with some far allocations).
+ * This was tested only with MSC; for other MSDOS compilers you may have
+ * to define NO_MEMCPY in zutil.h.  If you don't need the mixed model,
+ * just define FAR to be empty.
+ */
+#ifdef SYS16BIT
+#  if defined(M_I86SM) || defined(M_I86MM)
+     /* MSC small or medium model */
+#    define SMALL_MEDIUM
+#    ifdef _MSC_VER
+#      define FAR _far
+#    else
+#      define FAR far
+#    endif
+#  endif
+#  if (defined(__SMALL__) || defined(__MEDIUM__))
+     /* Turbo C small or medium model */
+#    define SMALL_MEDIUM
+#    ifdef __BORLANDC__
+#      define FAR _far
+#    else
+#      define FAR far
+#    endif
+#  endif
+#endif
+
+#if defined(WINDOWS) || defined(WIN32)
+   /* If building or using zlib as a DLL, define ZLIB_DLL.
+    * This is not mandatory, but it offers a little performance increase.
+    */
+#  ifdef ZLIB_DLL
+#    if defined(WIN32) && (!defined(__BORLANDC__) || (__BORLANDC__ >= 0x500))
+#      ifdef ZLIB_INTERNAL
+#        define ZEXTERN extern __declspec(dllexport)
+#      else
+#        define ZEXTERN extern __declspec(dllimport)
+#      endif
+#    endif
+#  endif  /* ZLIB_DLL */
+   /* If building or using zlib with the WINAPI/WINAPIV calling convention,
+    * define ZLIB_WINAPI.
+    * Caution: the standard ZLIB1.DLL is NOT compiled using ZLIB_WINAPI.
+    */
+#  ifdef ZLIB_WINAPI
+#    ifdef FAR
+#      undef FAR
+#    endif
+#    include <windows.h>
+     /* No need for _export, use ZLIB.DEF instead. */
+     /* For complete Windows compatibility, use WINAPI, not __stdcall. */
+#    define ZEXPORT WINAPI
+#    ifdef WIN32
+#      define ZEXPORTVA WINAPIV
+#    else
+#      define ZEXPORTVA FAR CDECL
+#    endif
+#  endif
+#endif
+
+#if defined (__BEOS__)
+#  ifdef ZLIB_DLL
+#    ifdef ZLIB_INTERNAL
+#      define ZEXPORT   __declspec(dllexport)
+#      define ZEXPORTVA __declspec(dllexport)
+#    else
+#      define ZEXPORT   __declspec(dllimport)
+#      define ZEXPORTVA __declspec(dllimport)
+#    endif
+#  endif
+#endif
+
+#ifndef ZEXTERN
+#  define ZEXTERN extern
+#endif
+#ifndef ZEXPORT
+#  define ZEXPORT
+#endif
+#ifndef ZEXPORTVA
+#  define ZEXPORTVA
+#endif
+
+#ifndef FAR
+#  define FAR
+#endif
+
+#if !defined(__MACTYPES__)
+typedef unsigned char  Byte;  /* 8 bits */
+#endif
+typedef unsigned int   uInt;  /* 16 bits or more */
+typedef unsigned long  uLong; /* 32 bits or more */
+
+#ifdef SMALL_MEDIUM
+   /* Borland C/C++ and some old MSC versions ignore FAR inside typedef */
+#  define Bytef Byte FAR
+#else
+   typedef Byte  FAR Bytef;
+#endif
+typedef char  FAR charf;
+typedef int   FAR intf;
+typedef uInt  FAR uIntf;
+typedef uLong FAR uLongf;
+
+#ifdef STDC
+   typedef void const *voidpc;
+   typedef void FAR   *voidpf;
+   typedef void       *voidp;
+#else
+   typedef Byte const *voidpc;
+   typedef Byte FAR   *voidpf;
+   typedef Byte       *voidp;
+#endif
+
+#if !defined(Z_U4) && !defined(Z_SOLO) && defined(STDC)
+#  include <limits.h>
+#  if (UINT_MAX == 0xffffffffUL)
+#    define Z_U4 unsigned
+#  elif (ULONG_MAX == 0xffffffffUL)
+#    define Z_U4 unsigned long
+#  elif (USHRT_MAX == 0xffffffffUL)
+#    define Z_U4 unsigned short
+#  endif
+#endif
+
+#ifdef Z_U4
+   typedef Z_U4 z_crc_t;
+#else
+   typedef unsigned long z_crc_t;
+#endif
+
+#if 1    /* was set to #if 1 by ./configure */
+#  define Z_HAVE_UNISTD_H
+#endif
+
+#if 1    /* was set to #if 1 by ./configure */
+#  define Z_HAVE_STDARG_H
+#endif
+
+#ifdef STDC
+#  ifndef Z_SOLO
+#    include <sys/types.h>      /* for off_t */
+#  endif
+#endif
+
+#if defined(STDC) || defined(Z_HAVE_STDARG_H)
+#  ifndef Z_SOLO
+#    include <stdarg.h>         /* for va_list */
+#  endif
+#endif
+
+#ifdef _WIN32
+#  ifndef Z_SOLO
+#    include <stddef.h>         /* for wchar_t */
+#  endif
+#endif
+
+/* a little trick to accommodate both "#define _LARGEFILE64_SOURCE" and
+ * "#define _LARGEFILE64_SOURCE 1" as requesting 64-bit operations, (even
+ * though the former does not conform to the LFS document), but considering
+ * both "#undef _LARGEFILE64_SOURCE" and "#define _LARGEFILE64_SOURCE 0" as
+ * equivalently requesting no 64-bit operations
+ */
+#if defined(_LARGEFILE64_SOURCE) && -_LARGEFILE64_SOURCE - -1 == 1
+#  undef _LARGEFILE64_SOURCE
+#endif
+
+#if defined(__WATCOMC__) && !defined(Z_HAVE_UNISTD_H)
+#  define Z_HAVE_UNISTD_H
+#endif
+#ifndef Z_SOLO
+#  if defined(Z_HAVE_UNISTD_H) || defined(_LARGEFILE64_SOURCE)
+#    include <unistd.h>         /* for SEEK_*, off_t, and _LFS64_LARGEFILE */
+#    ifdef VMS
+#      include <unixio.h>       /* for off_t */
+#    endif
+#    ifndef z_off_t
+#      define z_off_t off_t
+#    endif
+#  endif
+#endif
+
+#if defined(_LFS64_LARGEFILE) && _LFS64_LARGEFILE-0
+#  define Z_LFS64
+#endif
+
+#if defined(_LARGEFILE64_SOURCE) && defined(Z_LFS64)
+#  define Z_LARGE64
+#endif
+
+#if defined(_FILE_OFFSET_BITS) && _FILE_OFFSET_BITS-0 == 64 && defined(Z_LFS64)
+#  define Z_WANT64
+#endif
+
+#if !defined(SEEK_SET) && !defined(Z_SOLO)
+#  define SEEK_SET        0       /* Seek from beginning of file.  */
+#  define SEEK_CUR        1       /* Seek from current position.  */
+#  define SEEK_END        2       /* Set file pointer to EOF plus "offset" */
+#endif
+
+#ifndef z_off_t
+#  define z_off_t long
+#endif
+
+#if !defined(_WIN32) && defined(Z_LARGE64)
+#  define z_off64_t off64_t
+#else
+#  if defined(_WIN32) && !defined(__GNUC__) && !defined(Z_SOLO)
+#    define z_off64_t __int64
+#  else
+#    define z_off64_t z_off_t
+#  endif
+#endif
+
+/* MVS linker does not support external names larger than 8 bytes */
+#if defined(__MVS__)
+  #pragma map(deflateInit_,"DEIN")
+  #pragma map(deflateInit2_,"DEIN2")
+  #pragma map(deflateEnd,"DEEND")
+  #pragma map(deflateBound,"DEBND")
+  #pragma map(inflateInit_,"ININ")
+  #pragma map(inflateInit2_,"ININ2")
+  #pragma map(inflateEnd,"INEND")
+  #pragma map(inflateSync,"INSY")
+  #pragma map(inflateSetDictionary,"INSEDI")
+  #pragma map(compressBound,"CMBND")
+  #pragma map(inflate_table,"INTABL")
+  #pragma map(inflate_fast,"INFA")
+  #pragma map(inflate_copyright,"INCOPY")
+#endif
+
+#endif /* ZCONF_H */
diff --git a/distribution/zlib/include/zlib.h b/distribution/zlib/include/zlib.h
new file mode 100644
index 0000000..f09cdaf
--- /dev/null
+++ b/distribution/zlib/include/zlib.h
@@ -0,0 +1,1912 @@
+/* zlib.h -- interface of the 'zlib' general purpose compression library
+  version 1.2.11, January 15th, 2017
+
+  Copyright (C) 1995-2017 Jean-loup Gailly and Mark Adler
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  Jean-loup Gailly        Mark Adler
+  jloup@gzip.org          madler@alumni.caltech.edu
+
+
+  The data format used by the zlib library is described by RFCs (Request for
+  Comments) 1950 to 1952 in the files http://tools.ietf.org/html/rfc1950
+  (zlib format), rfc1951 (deflate format) and rfc1952 (gzip format).
+*/
+
+#ifndef ZLIB_H
+#define ZLIB_H
+
+#include "zconf.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define ZLIB_VERSION "1.2.11"
+#define ZLIB_VERNUM 0x12b0
+#define ZLIB_VER_MAJOR 1
+#define ZLIB_VER_MINOR 2
+#define ZLIB_VER_REVISION 11
+#define ZLIB_VER_SUBREVISION 0
+
+/*
+    The 'zlib' compression library provides in-memory compression and
+  decompression functions, including integrity checks of the uncompressed data.
+  This version of the library supports only one compression method (deflation)
+  but other algorithms will be added later and will have the same stream
+  interface.
+
+    Compression can be done in a single step if the buffers are large enough,
+  or can be done by repeated calls of the compression function.  In the latter
+  case, the application must provide more input and/or consume the output
+  (providing more output space) before each call.
+
+    The compressed data format used by default by the in-memory functions is
+  the zlib format, which is a zlib wrapper documented in RFC 1950, wrapped
+  around a deflate stream, which is itself documented in RFC 1951.
+
+    The library also supports reading and writing files in gzip (.gz) format
+  with an interface similar to that of stdio using the functions that start
+  with "gz".  The gzip format is different from the zlib format.  gzip is a
+  gzip wrapper, documented in RFC 1952, wrapped around a deflate stream.
+
+    This library can optionally read and write gzip and raw deflate streams in
+  memory as well.
+
+    The zlib format was designed to be compact and fast for use in memory
+  and on communications channels.  The gzip format was designed for single-
+  file compression on file systems, has a larger header than zlib to maintain
+  directory information, and uses a different, slower check method than zlib.
+
+    The library does not install any signal handler.  The decoder checks
+  the consistency of the compressed data, so the library should never crash
+  even in the case of corrupted input.
+*/
+
+typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size));
+typedef void   (*free_func)  OF((voidpf opaque, voidpf address));
+
+struct internal_state;
+
+typedef struct z_stream_s {
+    z_const Bytef *next_in;     /* next input byte */
+    uInt     avail_in;  /* number of bytes available at next_in */
+    uLong    total_in;  /* total number of input bytes read so far */
+
+    Bytef    *next_out; /* next output byte will go here */
+    uInt     avail_out; /* remaining free space at next_out */
+    uLong    total_out; /* total number of bytes output so far */
+
+    z_const char *msg;  /* last error message, NULL if no error */
+    struct internal_state FAR *state; /* not visible by applications */
+
+    alloc_func zalloc;  /* used to allocate the internal state */
+    free_func  zfree;   /* used to free the internal state */
+    voidpf     opaque;  /* private data object passed to zalloc and zfree */
+
+    int     data_type;  /* best guess about the data type: binary or text
+                           for deflate, or the decoding state for inflate */
+    uLong   adler;      /* Adler-32 or CRC-32 value of the uncompressed data */
+    uLong   reserved;   /* reserved for future use */
+} z_stream;
+
+typedef z_stream FAR *z_streamp;
+
+/*
+     gzip header information passed to and from zlib routines.  See RFC 1952
+  for more details on the meanings of these fields.
+*/
+typedef struct gz_header_s {
+    int     text;       /* true if compressed data believed to be text */
+    uLong   time;       /* modification time */
+    int     xflags;     /* extra flags (not used when writing a gzip file) */
+    int     os;         /* operating system */
+    Bytef   *extra;     /* pointer to extra field or Z_NULL if none */
+    uInt    extra_len;  /* extra field length (valid if extra != Z_NULL) */
+    uInt    extra_max;  /* space at extra (only when reading header) */
+    Bytef   *name;      /* pointer to zero-terminated file name or Z_NULL */
+    uInt    name_max;   /* space at name (only when reading header) */
+    Bytef   *comment;   /* pointer to zero-terminated comment or Z_NULL */
+    uInt    comm_max;   /* space at comment (only when reading header) */
+    int     hcrc;       /* true if there was or will be a header crc */
+    int     done;       /* true when done reading gzip header (not used
+                           when writing a gzip file) */
+} gz_header;
+
+typedef gz_header FAR *gz_headerp;
+
+/*
+     The application must update next_in and avail_in when avail_in has dropped
+   to zero.  It must update next_out and avail_out when avail_out has dropped
+   to zero.  The application must initialize zalloc, zfree and opaque before
+   calling the init function.  All other fields are set by the compression
+   library and must not be updated by the application.
+
+     The opaque value provided by the application will be passed as the first
+   parameter for calls of zalloc and zfree.  This can be useful for custom
+   memory management.  The compression library attaches no meaning to the
+   opaque value.
+
+     zalloc must return Z_NULL if there is not enough memory for the object.
+   If zlib is used in a multi-threaded application, zalloc and zfree must be
+   thread safe.  In that case, zlib is thread-safe.  When zalloc and zfree are
+   Z_NULL on entry to the initialization function, they are set to internal
+   routines that use the standard library functions malloc() and free().
+
+     On 16-bit systems, the functions zalloc and zfree must be able to allocate
+   exactly 65536 bytes, but will not be required to allocate more than this if
+   the symbol MAXSEG_64K is defined (see zconf.h).  WARNING: On MSDOS, pointers
+   returned by zalloc for objects of exactly 65536 bytes *must* have their
+   offset normalized to zero.  The default allocation function provided by this
+   library ensures this (see zutil.c).  To reduce memory requirements and avoid
+   any allocation of 64K objects, at the expense of compression ratio, compile
+   the library with -DMAX_WBITS=14 (see zconf.h).
+
+     The fields total_in and total_out can be used for statistics or progress
+   reports.  After compression, total_in holds the total size of the
+   uncompressed data and may be saved for use by the decompressor (particularly
+   if the decompressor wants to decompress everything in a single step).
+*/
+
+                        /* constants */
+
+#define Z_NO_FLUSH      0
+#define Z_PARTIAL_FLUSH 1
+#define Z_SYNC_FLUSH    2
+#define Z_FULL_FLUSH    3
+#define Z_FINISH        4
+#define Z_BLOCK         5
+#define Z_TREES         6
+/* Allowed flush values; see deflate() and inflate() below for details */
+
+#define Z_OK            0
+#define Z_STREAM_END    1
+#define Z_NEED_DICT     2
+#define Z_ERRNO        (-1)
+#define Z_STREAM_ERROR (-2)
+#define Z_DATA_ERROR   (-3)
+#define Z_MEM_ERROR    (-4)
+#define Z_BUF_ERROR    (-5)
+#define Z_VERSION_ERROR (-6)
+/* Return codes for the compression/decompression functions. Negative values
+ * are errors, positive values are used for special but normal events.
+ */
+
+#define Z_NO_COMPRESSION         0
+#define Z_BEST_SPEED             1
+#define Z_BEST_COMPRESSION       9
+#define Z_DEFAULT_COMPRESSION  (-1)
+/* compression levels */
+
+#define Z_FILTERED            1
+#define Z_HUFFMAN_ONLY        2
+#define Z_RLE                 3
+#define Z_FIXED               4
+#define Z_DEFAULT_STRATEGY    0
+/* compression strategy; see deflateInit2() below for details */
+
+#define Z_BINARY   0
+#define Z_TEXT     1
+#define Z_ASCII    Z_TEXT   /* for compatibility with 1.2.2 and earlier */
+#define Z_UNKNOWN  2
+/* Possible values of the data_type field for deflate() */
+
+#define Z_DEFLATED   8
+/* The deflate compression method (the only one supported in this version) */
+
+#define Z_NULL  0  /* for initializing zalloc, zfree, opaque */
+
+#define zlib_version zlibVersion()
+/* for compatibility with versions < 1.0.2 */
+
+
+                        /* basic functions */
+
+ZEXTERN const char * ZEXPORT zlibVersion OF((void));
+/* The application can compare zlibVersion and ZLIB_VERSION for consistency.
+   If the first character differs, the library code actually used is not
+   compatible with the zlib.h header file used by the application.  This check
+   is automatically made by deflateInit and inflateInit.
+ */
+
+/*
+ZEXTERN int ZEXPORT deflateInit OF((z_streamp strm, int level));
+
+     Initializes the internal stream state for compression.  The fields
+   zalloc, zfree and opaque must be initialized before by the caller.  If
+   zalloc and zfree are set to Z_NULL, deflateInit updates them to use default
+   allocation functions.
+
+     The compression level must be Z_DEFAULT_COMPRESSION, or between 0 and 9:
+   1 gives best speed, 9 gives best compression, 0 gives no compression at all
+   (the input data is simply copied a block at a time).  Z_DEFAULT_COMPRESSION
+   requests a default compromise between speed and compression (currently
+   equivalent to level 6).
+
+     deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_STREAM_ERROR if level is not a valid compression level, or
+   Z_VERSION_ERROR if the zlib library version (zlib_version) is incompatible
+   with the version assumed by the caller (ZLIB_VERSION).  msg is set to null
+   if there is no error message.  deflateInit does not perform any compression:
+   this will be done by deflate().
+*/
+
+
+ZEXTERN int ZEXPORT deflate OF((z_streamp strm, int flush));
+/*
+    deflate compresses as much data as possible, and stops when the input
+  buffer becomes empty or the output buffer becomes full.  It may introduce
+  some output latency (reading input without producing any output) except when
+  forced to flush.
+
+    The detailed semantics are as follows.  deflate performs one or both of the
+  following actions:
+
+  - Compress more input starting at next_in and update next_in and avail_in
+    accordingly.  If not all input can be processed (because there is not
+    enough room in the output buffer), next_in and avail_in are updated and
+    processing will resume at this point for the next call of deflate().
+
+  - Generate more output starting at next_out and update next_out and avail_out
+    accordingly.  This action is forced if the parameter flush is non zero.
+    Forcing flush frequently degrades the compression ratio, so this parameter
+    should be set only when necessary.  Some output may be provided even if
+    flush is zero.
+
+    Before the call of deflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming more
+  output, and updating avail_in or avail_out accordingly; avail_out should
+  never be zero before the call.  The application can consume the compressed
+  output when it wants, for example when the output buffer is full (avail_out
+  == 0), or after each call of deflate().  If deflate returns Z_OK and with
+  zero avail_out, it must be called again after making room in the output
+  buffer because there might be more output pending. See deflatePending(),
+  which can be used if desired to determine whether or not there is more ouput
+  in that case.
+
+    Normally the parameter flush is set to Z_NO_FLUSH, which allows deflate to
+  decide how much data to accumulate before producing output, in order to
+  maximize compression.
+
+    If the parameter flush is set to Z_SYNC_FLUSH, all pending output is
+  flushed to the output buffer and the output is aligned on a byte boundary, so
+  that the decompressor can get all input data available so far.  (In
+  particular avail_in is zero after the call if enough output space has been
+  provided before the call.) Flushing may degrade compression for some
+  compression algorithms and so it should be used only when necessary.  This
+  completes the current deflate block and follows it with an empty stored block
+  that is three bits plus filler bits to the next byte, followed by four bytes
+  (00 00 ff ff).
+
+    If flush is set to Z_PARTIAL_FLUSH, all pending output is flushed to the
+  output buffer, but the output is not aligned to a byte boundary.  All of the
+  input data so far will be available to the decompressor, as for Z_SYNC_FLUSH.
+  This completes the current deflate block and follows it with an empty fixed
+  codes block that is 10 bits long.  This assures that enough bytes are output
+  in order for the decompressor to finish the block before the empty fixed
+  codes block.
+
+    If flush is set to Z_BLOCK, a deflate block is completed and emitted, as
+  for Z_SYNC_FLUSH, but the output is not aligned on a byte boundary, and up to
+  seven bits of the current block are held to be written as the next byte after
+  the next deflate block is completed.  In this case, the decompressor may not
+  be provided enough bits at this point in order to complete decompression of
+  the data provided so far to the compressor.  It may need to wait for the next
+  block to be emitted.  This is for advanced applications that need to control
+  the emission of deflate blocks.
+
+    If flush is set to Z_FULL_FLUSH, all output is flushed as with
+  Z_SYNC_FLUSH, and the compression state is reset so that decompression can
+  restart from this point if previous compressed data has been damaged or if
+  random access is desired.  Using Z_FULL_FLUSH too often can seriously degrade
+  compression.
+
+    If deflate returns with avail_out == 0, this function must be called again
+  with the same value of the flush parameter and more output space (updated
+  avail_out), until the flush is complete (deflate returns with non-zero
+  avail_out).  In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that
+  avail_out is greater than six to avoid repeated flush markers due to
+  avail_out == 0 on return.
+
+    If the parameter flush is set to Z_FINISH, pending input is processed,
+  pending output is flushed and deflate returns with Z_STREAM_END if there was
+  enough output space.  If deflate returns with Z_OK or Z_BUF_ERROR, this
+  function must be called again with Z_FINISH and more output space (updated
+  avail_out) but no more input data, until it returns with Z_STREAM_END or an
+  error.  After deflate has returned Z_STREAM_END, the only possible operations
+  on the stream are deflateReset or deflateEnd.
+
+    Z_FINISH can be used in the first deflate call after deflateInit if all the
+  compression is to be done in a single step.  In order to complete in one
+  call, avail_out must be at least the value returned by deflateBound (see
+  below).  Then deflate is guaranteed to return Z_STREAM_END.  If not enough
+  output space is provided, deflate will not return Z_STREAM_END, and it must
+  be called again as described above.
+
+    deflate() sets strm->adler to the Adler-32 checksum of all input read
+  so far (that is, total_in bytes).  If a gzip stream is being generated, then
+  strm->adler will be the CRC-32 checksum of the input read so far.  (See
+  deflateInit2 below.)
+
+    deflate() may update strm->data_type if it can make a good guess about
+  the input data type (Z_BINARY or Z_TEXT).  If in doubt, the data is
+  considered binary.  This field is only for information purposes and does not
+  affect the compression algorithm in any manner.
+
+    deflate() returns Z_OK if some progress has been made (more input
+  processed or more output produced), Z_STREAM_END if all input has been
+  consumed and all output has been produced (only when flush is set to
+  Z_FINISH), Z_STREAM_ERROR if the stream state was inconsistent (for example
+  if next_in or next_out was Z_NULL or the state was inadvertently written over
+  by the application), or Z_BUF_ERROR if no progress is possible (for example
+  avail_in or avail_out was zero).  Note that Z_BUF_ERROR is not fatal, and
+  deflate() can be called again with more input and more output space to
+  continue compressing.
+*/
+
+
+ZEXTERN int ZEXPORT deflateEnd OF((z_streamp strm));
+/*
+     All dynamically allocated data structures for this stream are freed.
+   This function discards any unprocessed input and does not flush any pending
+   output.
+
+     deflateEnd returns Z_OK if success, Z_STREAM_ERROR if the
+   stream state was inconsistent, Z_DATA_ERROR if the stream was freed
+   prematurely (some input or output was discarded).  In the error case, msg
+   may be set but then points to a static string (which must not be
+   deallocated).
+*/
+
+
+/*
+ZEXTERN int ZEXPORT inflateInit OF((z_streamp strm));
+
+     Initializes the internal stream state for decompression.  The fields
+   next_in, avail_in, zalloc, zfree and opaque must be initialized before by
+   the caller.  In the current version of inflate, the provided input is not
+   read or consumed.  The allocation of a sliding window will be deferred to
+   the first call of inflate (if the decompression does not complete on the
+   first call).  If zalloc and zfree are set to Z_NULL, inflateInit updates
+   them to use default allocation functions.
+
+     inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_VERSION_ERROR if the zlib library version is incompatible with the
+   version assumed by the caller, or Z_STREAM_ERROR if the parameters are
+   invalid, such as a null pointer to the structure.  msg is set to null if
+   there is no error message.  inflateInit does not perform any decompression.
+   Actual decompression will be done by inflate().  So next_in, and avail_in,
+   next_out, and avail_out are unused and unchanged.  The current
+   implementation of inflateInit() does not process any header information --
+   that is deferred until inflate() is called.
+*/
+
+
+ZEXTERN int ZEXPORT inflate OF((z_streamp strm, int flush));
+/*
+    inflate decompresses as much data as possible, and stops when the input
+  buffer becomes empty or the output buffer becomes full.  It may introduce
+  some output latency (reading input without producing any output) except when
+  forced to flush.
+
+  The detailed semantics are as follows.  inflate performs one or both of the
+  following actions:
+
+  - Decompress more input starting at next_in and update next_in and avail_in
+    accordingly.  If not all input can be processed (because there is not
+    enough room in the output buffer), then next_in and avail_in are updated
+    accordingly, and processing will resume at this point for the next call of
+    inflate().
+
+  - Generate more output starting at next_out and update next_out and avail_out
+    accordingly.  inflate() provides as much output as possible, until there is
+    no more input data or no more space in the output buffer (see below about
+    the flush parameter).
+
+    Before the call of inflate(), the application should ensure that at least
+  one of the actions is possible, by providing more input and/or consuming more
+  output, and updating the next_* and avail_* values accordingly.  If the
+  caller of inflate() does not provide both available input and available
+  output space, it is possible that there will be no progress made.  The
+  application can consume the uncompressed output when it wants, for example
+  when the output buffer is full (avail_out == 0), or after each call of
+  inflate().  If inflate returns Z_OK and with zero avail_out, it must be
+  called again after making room in the output buffer because there might be
+  more output pending.
+
+    The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH, Z_FINISH,
+  Z_BLOCK, or Z_TREES.  Z_SYNC_FLUSH requests that inflate() flush as much
+  output as possible to the output buffer.  Z_BLOCK requests that inflate()
+  stop if and when it gets to the next deflate block boundary.  When decoding
+  the zlib or gzip format, this will cause inflate() to return immediately
+  after the header and before the first block.  When doing a raw inflate,
+  inflate() will go ahead and process the first block, and will return when it
+  gets to the end of that block, or when it runs out of data.
+
+    The Z_BLOCK option assists in appending to or combining deflate streams.
+  To assist in this, on return inflate() always sets strm->data_type to the
+  number of unused bits in the last byte taken from strm->next_in, plus 64 if
+  inflate() is currently decoding the last block in the deflate stream, plus
+  128 if inflate() returned immediately after decoding an end-of-block code or
+  decoding the complete header up to just before the first byte of the deflate
+  stream.  The end-of-block will not be indicated until all of the uncompressed
+  data from that block has been written to strm->next_out.  The number of
+  unused bits may in general be greater than seven, except when bit 7 of
+  data_type is set, in which case the number of unused bits will be less than
+  eight.  data_type is set as noted here every time inflate() returns for all
+  flush options, and so can be used to determine the amount of currently
+  consumed input in bits.
+
+    The Z_TREES option behaves as Z_BLOCK does, but it also returns when the
+  end of each deflate block header is reached, before any actual data in that
+  block is decoded.  This allows the caller to determine the length of the
+  deflate block header for later use in random access within a deflate block.
+  256 is added to the value of strm->data_type when inflate() returns
+  immediately after reaching the end of the deflate block header.
+
+    inflate() should normally be called until it returns Z_STREAM_END or an
+  error.  However if all decompression is to be performed in a single step (a
+  single call of inflate), the parameter flush should be set to Z_FINISH.  In
+  this case all pending input is processed and all pending output is flushed;
+  avail_out must be large enough to hold all of the uncompressed data for the
+  operation to complete.  (The size of the uncompressed data may have been
+  saved by the compressor for this purpose.)  The use of Z_FINISH is not
+  required to perform an inflation in one step.  However it may be used to
+  inform inflate that a faster approach can be used for the single inflate()
+  call.  Z_FINISH also informs inflate to not maintain a sliding window if the
+  stream completes, which reduces inflate's memory footprint.  If the stream
+  does not complete, either because not all of the stream is provided or not
+  enough output space is provided, then a sliding window will be allocated and
+  inflate() can be called again to continue the operation as if Z_NO_FLUSH had
+  been used.
+
+     In this implementation, inflate() always flushes as much output as
+  possible to the output buffer, and always uses the faster approach on the
+  first call.  So the effects of the flush parameter in this implementation are
+  on the return value of inflate() as noted below, when inflate() returns early
+  when Z_BLOCK or Z_TREES is used, and when inflate() avoids the allocation of
+  memory for a sliding window when Z_FINISH is used.
+
+     If a preset dictionary is needed after this call (see inflateSetDictionary
+  below), inflate sets strm->adler to the Adler-32 checksum of the dictionary
+  chosen by the compressor and returns Z_NEED_DICT; otherwise it sets
+  strm->adler to the Adler-32 checksum of all output produced so far (that is,
+  total_out bytes) and returns Z_OK, Z_STREAM_END or an error code as described
+  below.  At the end of the stream, inflate() checks that its computed Adler-32
+  checksum is equal to that saved by the compressor and returns Z_STREAM_END
+  only if the checksum is correct.
+
+    inflate() can decompress and check either zlib-wrapped or gzip-wrapped
+  deflate data.  The header type is detected automatically, if requested when
+  initializing with inflateInit2().  Any information contained in the gzip
+  header is not retained unless inflateGetHeader() is used.  When processing
+  gzip-wrapped deflate data, strm->adler32 is set to the CRC-32 of the output
+  produced so far.  The CRC-32 is checked against the gzip trailer, as is the
+  uncompressed length, modulo 2^32.
+
+    inflate() returns Z_OK if some progress has been made (more input processed
+  or more output produced), Z_STREAM_END if the end of the compressed data has
+  been reached and all uncompressed output has been produced, Z_NEED_DICT if a
+  preset dictionary is needed at this point, Z_DATA_ERROR if the input data was
+  corrupted (input stream not conforming to the zlib format or incorrect check
+  value, in which case strm->msg points to a string with a more specific
+  error), Z_STREAM_ERROR if the stream structure was inconsistent (for example
+  next_in or next_out was Z_NULL, or the state was inadvertently written over
+  by the application), Z_MEM_ERROR if there was not enough memory, Z_BUF_ERROR
+  if no progress was possible or if there was not enough room in the output
+  buffer when Z_FINISH is used.  Note that Z_BUF_ERROR is not fatal, and
+  inflate() can be called again with more input and more output space to
+  continue decompressing.  If Z_DATA_ERROR is returned, the application may
+  then call inflateSync() to look for a good compression block if a partial
+  recovery of the data is to be attempted.
+*/
+
+
+ZEXTERN int ZEXPORT inflateEnd OF((z_streamp strm));
+/*
+     All dynamically allocated data structures for this stream are freed.
+   This function discards any unprocessed input and does not flush any pending
+   output.
+
+     inflateEnd returns Z_OK if success, or Z_STREAM_ERROR if the stream state
+   was inconsistent.
+*/
+
+
+                        /* Advanced functions */
+
+/*
+    The following functions are needed only in some special applications.
+*/
+
+/*
+ZEXTERN int ZEXPORT deflateInit2 OF((z_streamp strm,
+                                     int  level,
+                                     int  method,
+                                     int  windowBits,
+                                     int  memLevel,
+                                     int  strategy));
+
+     This is another version of deflateInit with more compression options.  The
+   fields next_in, zalloc, zfree and opaque must be initialized before by the
+   caller.
+
+     The method parameter is the compression method.  It must be Z_DEFLATED in
+   this version of the library.
+
+     The windowBits parameter is the base two logarithm of the window size
+   (the size of the history buffer).  It should be in the range 8..15 for this
+   version of the library.  Larger values of this parameter result in better
+   compression at the expense of memory usage.  The default value is 15 if
+   deflateInit is used instead.
+
+     For the current implementation of deflate(), a windowBits value of 8 (a
+   window size of 256 bytes) is not supported.  As a result, a request for 8
+   will result in 9 (a 512-byte window).  In that case, providing 8 to
+   inflateInit2() will result in an error when the zlib header with 9 is
+   checked against the initialization of inflate().  The remedy is to not use 8
+   with deflateInit2() with this initialization, or at least in that case use 9
+   with inflateInit2().
+
+     windowBits can also be -8..-15 for raw deflate.  In this case, -windowBits
+   determines the window size.  deflate() will then generate raw deflate data
+   with no zlib header or trailer, and will not compute a check value.
+
+     windowBits can also be greater than 15 for optional gzip encoding.  Add
+   16 to windowBits to write a simple gzip header and trailer around the
+   compressed data instead of a zlib wrapper.  The gzip header will have no
+   file name, no extra data, no comment, no modification time (set to zero), no
+   header crc, and the operating system will be set to the appropriate value,
+   if the operating system was determined at compile time.  If a gzip stream is
+   being written, strm->adler is a CRC-32 instead of an Adler-32.
+
+     For raw deflate or gzip encoding, a request for a 256-byte window is
+   rejected as invalid, since only the zlib header provides a means of
+   transmitting the window size to the decompressor.
+
+     The memLevel parameter specifies how much memory should be allocated
+   for the internal compression state.  memLevel=1 uses minimum memory but is
+   slow and reduces compression ratio; memLevel=9 uses maximum memory for
+   optimal speed.  The default value is 8.  See zconf.h for total memory usage
+   as a function of windowBits and memLevel.
+
+     The strategy parameter is used to tune the compression algorithm.  Use the
+   value Z_DEFAULT_STRATEGY for normal data, Z_FILTERED for data produced by a
+   filter (or predictor), Z_HUFFMAN_ONLY to force Huffman encoding only (no
+   string match), or Z_RLE to limit match distances to one (run-length
+   encoding).  Filtered data consists mostly of small values with a somewhat
+   random distribution.  In this case, the compression algorithm is tuned to
+   compress them better.  The effect of Z_FILTERED is to force more Huffman
+   coding and less string matching; it is somewhat intermediate between
+   Z_DEFAULT_STRATEGY and Z_HUFFMAN_ONLY.  Z_RLE is designed to be almost as
+   fast as Z_HUFFMAN_ONLY, but give better compression for PNG image data.  The
+   strategy parameter only affects the compression ratio but not the
+   correctness of the compressed output even if it is not set appropriately.
+   Z_FIXED prevents the use of dynamic Huffman codes, allowing for a simpler
+   decoder for special applications.
+
+     deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_STREAM_ERROR if any parameter is invalid (such as an invalid
+   method), or Z_VERSION_ERROR if the zlib library version (zlib_version) is
+   incompatible with the version assumed by the caller (ZLIB_VERSION).  msg is
+   set to null if there is no error message.  deflateInit2 does not perform any
+   compression: this will be done by deflate().
+*/
+
+ZEXTERN int ZEXPORT deflateSetDictionary OF((z_streamp strm,
+                                             const Bytef *dictionary,
+                                             uInt  dictLength));
+/*
+     Initializes the compression dictionary from the given byte sequence
+   without producing any compressed output.  When using the zlib format, this
+   function must be called immediately after deflateInit, deflateInit2 or
+   deflateReset, and before any call of deflate.  When doing raw deflate, this
+   function must be called either before any call of deflate, or immediately
+   after the completion of a deflate block, i.e. after all input has been
+   consumed and all output has been delivered when using any of the flush
+   options Z_BLOCK, Z_PARTIAL_FLUSH, Z_SYNC_FLUSH, or Z_FULL_FLUSH.  The
+   compressor and decompressor must use exactly the same dictionary (see
+   inflateSetDictionary).
+
+     The dictionary should consist of strings (byte sequences) that are likely
+   to be encountered later in the data to be compressed, with the most commonly
+   used strings preferably put towards the end of the dictionary.  Using a
+   dictionary is most useful when the data to be compressed is short and can be
+   predicted with good accuracy; the data can then be compressed better than
+   with the default empty dictionary.
+
+     Depending on the size of the compression data structures selected by
+   deflateInit or deflateInit2, a part of the dictionary may in effect be
+   discarded, for example if the dictionary is larger than the window size
+   provided in deflateInit or deflateInit2.  Thus the strings most likely to be
+   useful should be put at the end of the dictionary, not at the front.  In
+   addition, the current implementation of deflate will use at most the window
+   size minus 262 bytes of the provided dictionary.
+
+     Upon return of this function, strm->adler is set to the Adler-32 value
+   of the dictionary; the decompressor may later use this value to determine
+   which dictionary has been used by the compressor.  (The Adler-32 value
+   applies to the whole dictionary even if only a subset of the dictionary is
+   actually used by the compressor.) If a raw deflate was requested, then the
+   Adler-32 value is not computed and strm->adler is not set.
+
+     deflateSetDictionary returns Z_OK if success, or Z_STREAM_ERROR if a
+   parameter is invalid (e.g.  dictionary being Z_NULL) or the stream state is
+   inconsistent (for example if deflate has already been called for this stream
+   or if not at a block boundary for raw deflate).  deflateSetDictionary does
+   not perform any compression: this will be done by deflate().
+*/
+
+ZEXTERN int ZEXPORT deflateGetDictionary OF((z_streamp strm,
+                                             Bytef *dictionary,
+                                             uInt  *dictLength));
+/*
+     Returns the sliding dictionary being maintained by deflate.  dictLength is
+   set to the number of bytes in the dictionary, and that many bytes are copied
+   to dictionary.  dictionary must have enough space, where 32768 bytes is
+   always enough.  If deflateGetDictionary() is called with dictionary equal to
+   Z_NULL, then only the dictionary length is returned, and nothing is copied.
+   Similary, if dictLength is Z_NULL, then it is not set.
+
+     deflateGetDictionary() may return a length less than the window size, even
+   when more than the window size in input has been provided. It may return up
+   to 258 bytes less in that case, due to how zlib's implementation of deflate
+   manages the sliding window and lookahead for matches, where matches can be
+   up to 258 bytes long. If the application needs the last window-size bytes of
+   input, then that would need to be saved by the application outside of zlib.
+
+     deflateGetDictionary returns Z_OK on success, or Z_STREAM_ERROR if the
+   stream state is inconsistent.
+*/
+
+ZEXTERN int ZEXPORT deflateCopy OF((z_streamp dest,
+                                    z_streamp source));
+/*
+     Sets the destination stream as a complete copy of the source stream.
+
+     This function can be useful when several compression strategies will be
+   tried, for example when there are several ways of pre-processing the input
+   data with a filter.  The streams that will be discarded should then be freed
+   by calling deflateEnd.  Note that deflateCopy duplicates the internal
+   compression state which can be quite large, so this strategy is slow and can
+   consume lots of memory.
+
+     deflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not
+   enough memory, Z_STREAM_ERROR if the source stream state was inconsistent
+   (such as zalloc being Z_NULL).  msg is left unchanged in both source and
+   destination.
+*/
+
+ZEXTERN int ZEXPORT deflateReset OF((z_streamp strm));
+/*
+     This function is equivalent to deflateEnd followed by deflateInit, but
+   does not free and reallocate the internal compression state.  The stream
+   will leave the compression level and any other attributes that may have been
+   set unchanged.
+
+     deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL).
+*/
+
+ZEXTERN int ZEXPORT deflateParams OF((z_streamp strm,
+                                      int level,
+                                      int strategy));
+/*
+     Dynamically update the compression level and compression strategy.  The
+   interpretation of level and strategy is as in deflateInit2().  This can be
+   used to switch between compression and straight copy of the input data, or
+   to switch to a different kind of input data requiring a different strategy.
+   If the compression approach (which is a function of the level) or the
+   strategy is changed, and if any input has been consumed in a previous
+   deflate() call, then the input available so far is compressed with the old
+   level and strategy using deflate(strm, Z_BLOCK).  There are three approaches
+   for the compression levels 0, 1..3, and 4..9 respectively.  The new level
+   and strategy will take effect at the next call of deflate().
+
+     If a deflate(strm, Z_BLOCK) is performed by deflateParams(), and it does
+   not have enough output space to complete, then the parameter change will not
+   take effect.  In this case, deflateParams() can be called again with the
+   same parameters and more output space to try again.
+
+     In order to assure a change in the parameters on the first try, the
+   deflate stream should be flushed using deflate() with Z_BLOCK or other flush
+   request until strm.avail_out is not zero, before calling deflateParams().
+   Then no more input data should be provided before the deflateParams() call.
+   If this is done, the old level and strategy will be applied to the data
+   compressed before deflateParams(), and the new level and strategy will be
+   applied to the the data compressed after deflateParams().
+
+     deflateParams returns Z_OK on success, Z_STREAM_ERROR if the source stream
+   state was inconsistent or if a parameter was invalid, or Z_BUF_ERROR if
+   there was not enough output space to complete the compression of the
+   available input data before a change in the strategy or approach.  Note that
+   in the case of a Z_BUF_ERROR, the parameters are not changed.  A return
+   value of Z_BUF_ERROR is not fatal, in which case deflateParams() can be
+   retried with more output space.
+*/
+
+ZEXTERN int ZEXPORT deflateTune OF((z_streamp strm,
+                                    int good_length,
+                                    int max_lazy,
+                                    int nice_length,
+                                    int max_chain));
+/*
+     Fine tune deflate's internal compression parameters.  This should only be
+   used by someone who understands the algorithm used by zlib's deflate for
+   searching for the best matching string, and even then only by the most
+   fanatic optimizer trying to squeeze out the last compressed bit for their
+   specific input data.  Read the deflate.c source code for the meaning of the
+   max_lazy, good_length, nice_length, and max_chain parameters.
+
+     deflateTune() can be called after deflateInit() or deflateInit2(), and
+   returns Z_OK on success, or Z_STREAM_ERROR for an invalid deflate stream.
+ */
+
+ZEXTERN uLong ZEXPORT deflateBound OF((z_streamp strm,
+                                       uLong sourceLen));
+/*
+     deflateBound() returns an upper bound on the compressed size after
+   deflation of sourceLen bytes.  It must be called after deflateInit() or
+   deflateInit2(), and after deflateSetHeader(), if used.  This would be used
+   to allocate an output buffer for deflation in a single pass, and so would be
+   called before deflate().  If that first deflate() call is provided the
+   sourceLen input bytes, an output buffer allocated to the size returned by
+   deflateBound(), and the flush value Z_FINISH, then deflate() is guaranteed
+   to return Z_STREAM_END.  Note that it is possible for the compressed size to
+   be larger than the value returned by deflateBound() if flush options other
+   than Z_FINISH or Z_NO_FLUSH are used.
+*/
+
+ZEXTERN int ZEXPORT deflatePending OF((z_streamp strm,
+                                       unsigned *pending,
+                                       int *bits));
+/*
+     deflatePending() returns the number of bytes and bits of output that have
+   been generated, but not yet provided in the available output.  The bytes not
+   provided would be due to the available output space having being consumed.
+   The number of bits of output not provided are between 0 and 7, where they
+   await more bits to join them in order to fill out a full byte.  If pending
+   or bits are Z_NULL, then those values are not set.
+
+     deflatePending returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent.
+ */
+
+ZEXTERN int ZEXPORT deflatePrime OF((z_streamp strm,
+                                     int bits,
+                                     int value));
+/*
+     deflatePrime() inserts bits in the deflate output stream.  The intent
+   is that this function is used to start off the deflate output with the bits
+   leftover from a previous deflate stream when appending to it.  As such, this
+   function can only be used for raw deflate, and must be used before the first
+   deflate() call after a deflateInit2() or deflateReset().  bits must be less
+   than or equal to 16, and that many of the least significant bits of value
+   will be inserted in the output.
+
+     deflatePrime returns Z_OK if success, Z_BUF_ERROR if there was not enough
+   room in the internal buffer to insert the bits, or Z_STREAM_ERROR if the
+   source stream state was inconsistent.
+*/
+
+ZEXTERN int ZEXPORT deflateSetHeader OF((z_streamp strm,
+                                         gz_headerp head));
+/*
+     deflateSetHeader() provides gzip header information for when a gzip
+   stream is requested by deflateInit2().  deflateSetHeader() may be called
+   after deflateInit2() or deflateReset() and before the first call of
+   deflate().  The text, time, os, extra field, name, and comment information
+   in the provided gz_header structure are written to the gzip header (xflag is
+   ignored -- the extra flags are set according to the compression level).  The
+   caller must assure that, if not Z_NULL, name and comment are terminated with
+   a zero byte, and that if extra is not Z_NULL, that extra_len bytes are
+   available there.  If hcrc is true, a gzip header crc is included.  Note that
+   the current versions of the command-line version of gzip (up through version
+   1.3.x) do not support header crc's, and will report that it is a "multi-part
+   gzip file" and give up.
+
+     If deflateSetHeader is not used, the default gzip header has text false,
+   the time set to zero, and os set to 255, with no extra, name, or comment
+   fields.  The gzip header is returned to the default state by deflateReset().
+
+     deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent.
+*/
+
+/*
+ZEXTERN int ZEXPORT inflateInit2 OF((z_streamp strm,
+                                     int  windowBits));
+
+     This is another version of inflateInit with an extra parameter.  The
+   fields next_in, avail_in, zalloc, zfree and opaque must be initialized
+   before by the caller.
+
+     The windowBits parameter is the base two logarithm of the maximum window
+   size (the size of the history buffer).  It should be in the range 8..15 for
+   this version of the library.  The default value is 15 if inflateInit is used
+   instead.  windowBits must be greater than or equal to the windowBits value
+   provided to deflateInit2() while compressing, or it must be equal to 15 if
+   deflateInit2() was not used.  If a compressed stream with a larger window
+   size is given as input, inflate() will return with the error code
+   Z_DATA_ERROR instead of trying to allocate a larger window.
+
+     windowBits can also be zero to request that inflate use the window size in
+   the zlib header of the compressed stream.
+
+     windowBits can also be -8..-15 for raw inflate.  In this case, -windowBits
+   determines the window size.  inflate() will then process raw deflate data,
+   not looking for a zlib or gzip header, not generating a check value, and not
+   looking for any check values for comparison at the end of the stream.  This
+   is for use with other formats that use the deflate compressed data format
+   such as zip.  Those formats provide their own check values.  If a custom
+   format is developed using the raw deflate format for compressed data, it is
+   recommended that a check value such as an Adler-32 or a CRC-32 be applied to
+   the uncompressed data as is done in the zlib, gzip, and zip formats.  For
+   most applications, the zlib format should be used as is.  Note that comments
+   above on the use in deflateInit2() applies to the magnitude of windowBits.
+
+     windowBits can also be greater than 15 for optional gzip decoding.  Add
+   32 to windowBits to enable zlib and gzip decoding with automatic header
+   detection, or add 16 to decode only the gzip format (the zlib format will
+   return a Z_DATA_ERROR).  If a gzip stream is being decoded, strm->adler is a
+   CRC-32 instead of an Adler-32.  Unlike the gunzip utility and gzread() (see
+   below), inflate() will not automatically decode concatenated gzip streams.
+   inflate() will return Z_STREAM_END at the end of the gzip stream.  The state
+   would need to be reset to continue decoding a subsequent gzip stream.
+
+     inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_VERSION_ERROR if the zlib library version is incompatible with the
+   version assumed by the caller, or Z_STREAM_ERROR if the parameters are
+   invalid, such as a null pointer to the structure.  msg is set to null if
+   there is no error message.  inflateInit2 does not perform any decompression
+   apart from possibly reading the zlib header if present: actual decompression
+   will be done by inflate().  (So next_in and avail_in may be modified, but
+   next_out and avail_out are unused and unchanged.) The current implementation
+   of inflateInit2() does not process any header information -- that is
+   deferred until inflate() is called.
+*/
+
+ZEXTERN int ZEXPORT inflateSetDictionary OF((z_streamp strm,
+                                             const Bytef *dictionary,
+                                             uInt  dictLength));
+/*
+     Initializes the decompression dictionary from the given uncompressed byte
+   sequence.  This function must be called immediately after a call of inflate,
+   if that call returned Z_NEED_DICT.  The dictionary chosen by the compressor
+   can be determined from the Adler-32 value returned by that call of inflate.
+   The compressor and decompressor must use exactly the same dictionary (see
+   deflateSetDictionary).  For raw inflate, this function can be called at any
+   time to set the dictionary.  If the provided dictionary is smaller than the
+   window and there is already data in the window, then the provided dictionary
+   will amend what's there.  The application must insure that the dictionary
+   that was used for compression is provided.
+
+     inflateSetDictionary returns Z_OK if success, Z_STREAM_ERROR if a
+   parameter is invalid (e.g.  dictionary being Z_NULL) or the stream state is
+   inconsistent, Z_DATA_ERROR if the given dictionary doesn't match the
+   expected one (incorrect Adler-32 value).  inflateSetDictionary does not
+   perform any decompression: this will be done by subsequent calls of
+   inflate().
+*/
+
+ZEXTERN int ZEXPORT inflateGetDictionary OF((z_streamp strm,
+                                             Bytef *dictionary,
+                                             uInt  *dictLength));
+/*
+     Returns the sliding dictionary being maintained by inflate.  dictLength is
+   set to the number of bytes in the dictionary, and that many bytes are copied
+   to dictionary.  dictionary must have enough space, where 32768 bytes is
+   always enough.  If inflateGetDictionary() is called with dictionary equal to
+   Z_NULL, then only the dictionary length is returned, and nothing is copied.
+   Similary, if dictLength is Z_NULL, then it is not set.
+
+     inflateGetDictionary returns Z_OK on success, or Z_STREAM_ERROR if the
+   stream state is inconsistent.
+*/
+
+ZEXTERN int ZEXPORT inflateSync OF((z_streamp strm));
+/*
+     Skips invalid compressed data until a possible full flush point (see above
+   for the description of deflate with Z_FULL_FLUSH) can be found, or until all
+   available input is skipped.  No output is provided.
+
+     inflateSync searches for a 00 00 FF FF pattern in the compressed data.
+   All full flush points have this pattern, but not all occurrences of this
+   pattern are full flush points.
+
+     inflateSync returns Z_OK if a possible full flush point has been found,
+   Z_BUF_ERROR if no more input was provided, Z_DATA_ERROR if no flush point
+   has been found, or Z_STREAM_ERROR if the stream structure was inconsistent.
+   In the success case, the application may save the current current value of
+   total_in which indicates where valid compressed data was found.  In the
+   error case, the application may repeatedly call inflateSync, providing more
+   input each time, until success or end of the input data.
+*/
+
+ZEXTERN int ZEXPORT inflateCopy OF((z_streamp dest,
+                                    z_streamp source));
+/*
+     Sets the destination stream as a complete copy of the source stream.
+
+     This function can be useful when randomly accessing a large stream.  The
+   first pass through the stream can periodically record the inflate state,
+   allowing restarting inflate at those points when randomly accessing the
+   stream.
+
+     inflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not
+   enough memory, Z_STREAM_ERROR if the source stream state was inconsistent
+   (such as zalloc being Z_NULL).  msg is left unchanged in both source and
+   destination.
+*/
+
+ZEXTERN int ZEXPORT inflateReset OF((z_streamp strm));
+/*
+     This function is equivalent to inflateEnd followed by inflateInit,
+   but does not free and reallocate the internal decompression state.  The
+   stream will keep attributes that may have been set by inflateInit2.
+
+     inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL).
+*/
+
+ZEXTERN int ZEXPORT inflateReset2 OF((z_streamp strm,
+                                      int windowBits));
+/*
+     This function is the same as inflateReset, but it also permits changing
+   the wrap and window size requests.  The windowBits parameter is interpreted
+   the same as it is for inflateInit2.  If the window size is changed, then the
+   memory allocated for the window is freed, and the window will be reallocated
+   by inflate() if needed.
+
+     inflateReset2 returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent (such as zalloc or state being Z_NULL), or if
+   the windowBits parameter is invalid.
+*/
+
+ZEXTERN int ZEXPORT inflatePrime OF((z_streamp strm,
+                                     int bits,
+                                     int value));
+/*
+     This function inserts bits in the inflate input stream.  The intent is
+   that this function is used to start inflating at a bit position in the
+   middle of a byte.  The provided bits will be used before any bytes are used
+   from next_in.  This function should only be used with raw inflate, and
+   should be used before the first inflate() call after inflateInit2() or
+   inflateReset().  bits must be less than or equal to 16, and that many of the
+   least significant bits of value will be inserted in the input.
+
+     If bits is negative, then the input stream bit buffer is emptied.  Then
+   inflatePrime() can be called again to put bits in the buffer.  This is used
+   to clear out bits leftover after feeding inflate a block description prior
+   to feeding inflate codes.
+
+     inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent.
+*/
+
+ZEXTERN long ZEXPORT inflateMark OF((z_streamp strm));
+/*
+     This function returns two values, one in the lower 16 bits of the return
+   value, and the other in the remaining upper bits, obtained by shifting the
+   return value down 16 bits.  If the upper value is -1 and the lower value is
+   zero, then inflate() is currently decoding information outside of a block.
+   If the upper value is -1 and the lower value is non-zero, then inflate is in
+   the middle of a stored block, with the lower value equaling the number of
+   bytes from the input remaining to copy.  If the upper value is not -1, then
+   it is the number of bits back from the current bit position in the input of
+   the code (literal or length/distance pair) currently being processed.  In
+   that case the lower value is the number of bytes already emitted for that
+   code.
+
+     A code is being processed if inflate is waiting for more input to complete
+   decoding of the code, or if it has completed decoding but is waiting for
+   more output space to write the literal or match data.
+
+     inflateMark() is used to mark locations in the input data for random
+   access, which may be at bit positions, and to note those cases where the
+   output of a code may span boundaries of random access blocks.  The current
+   location in the input stream can be determined from avail_in and data_type
+   as noted in the description for the Z_BLOCK flush parameter for inflate.
+
+     inflateMark returns the value noted above, or -65536 if the provided
+   source stream state was inconsistent.
+*/
+
+ZEXTERN int ZEXPORT inflateGetHeader OF((z_streamp strm,
+                                         gz_headerp head));
+/*
+     inflateGetHeader() requests that gzip header information be stored in the
+   provided gz_header structure.  inflateGetHeader() may be called after
+   inflateInit2() or inflateReset(), and before the first call of inflate().
+   As inflate() processes the gzip stream, head->done is zero until the header
+   is completed, at which time head->done is set to one.  If a zlib stream is
+   being decoded, then head->done is set to -1 to indicate that there will be
+   no gzip header information forthcoming.  Note that Z_BLOCK or Z_TREES can be
+   used to force inflate() to return immediately after header processing is
+   complete and before any actual data is decompressed.
+
+     The text, time, xflags, and os fields are filled in with the gzip header
+   contents.  hcrc is set to true if there is a header CRC.  (The header CRC
+   was valid if done is set to one.) If extra is not Z_NULL, then extra_max
+   contains the maximum number of bytes to write to extra.  Once done is true,
+   extra_len contains the actual extra field length, and extra contains the
+   extra field, or that field truncated if extra_max is less than extra_len.
+   If name is not Z_NULL, then up to name_max characters are written there,
+   terminated with a zero unless the length is greater than name_max.  If
+   comment is not Z_NULL, then up to comm_max characters are written there,
+   terminated with a zero unless the length is greater than comm_max.  When any
+   of extra, name, or comment are not Z_NULL and the respective field is not
+   present in the header, then that field is set to Z_NULL to signal its
+   absence.  This allows the use of deflateSetHeader() with the returned
+   structure to duplicate the header.  However if those fields are set to
+   allocated memory, then the application will need to save those pointers
+   elsewhere so that they can be eventually freed.
+
+     If inflateGetHeader is not used, then the header information is simply
+   discarded.  The header is always checked for validity, including the header
+   CRC if present.  inflateReset() will reset the process to discard the header
+   information.  The application would need to call inflateGetHeader() again to
+   retrieve the header from the next gzip stream.
+
+     inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source
+   stream state was inconsistent.
+*/
+
+/*
+ZEXTERN int ZEXPORT inflateBackInit OF((z_streamp strm, int windowBits,
+                                        unsigned char FAR *window));
+
+     Initialize the internal stream state for decompression using inflateBack()
+   calls.  The fields zalloc, zfree and opaque in strm must be initialized
+   before the call.  If zalloc and zfree are Z_NULL, then the default library-
+   derived memory allocation routines are used.  windowBits is the base two
+   logarithm of the window size, in the range 8..15.  window is a caller
+   supplied buffer of that size.  Except for special applications where it is
+   assured that deflate was used with small window sizes, windowBits must be 15
+   and a 32K byte window must be supplied to be able to decompress general
+   deflate streams.
+
+     See inflateBack() for the usage of these routines.
+
+     inflateBackInit will return Z_OK on success, Z_STREAM_ERROR if any of
+   the parameters are invalid, Z_MEM_ERROR if the internal state could not be
+   allocated, or Z_VERSION_ERROR if the version of the library does not match
+   the version of the header file.
+*/
+
+typedef unsigned (*in_func) OF((void FAR *,
+                                z_const unsigned char FAR * FAR *));
+typedef int (*out_func) OF((void FAR *, unsigned char FAR *, unsigned));
+
+ZEXTERN int ZEXPORT inflateBack OF((z_streamp strm,
+                                    in_func in, void FAR *in_desc,
+                                    out_func out, void FAR *out_desc));
+/*
+     inflateBack() does a raw inflate with a single call using a call-back
+   interface for input and output.  This is potentially more efficient than
+   inflate() for file i/o applications, in that it avoids copying between the
+   output and the sliding window by simply making the window itself the output
+   buffer.  inflate() can be faster on modern CPUs when used with large
+   buffers.  inflateBack() trusts the application to not change the output
+   buffer passed by the output function, at least until inflateBack() returns.
+
+     inflateBackInit() must be called first to allocate the internal state
+   and to initialize the state with the user-provided window buffer.
+   inflateBack() may then be used multiple times to inflate a complete, raw
+   deflate stream with each call.  inflateBackEnd() is then called to free the
+   allocated state.
+
+     A raw deflate stream is one with no zlib or gzip header or trailer.
+   This routine would normally be used in a utility that reads zip or gzip
+   files and writes out uncompressed files.  The utility would decode the
+   header and process the trailer on its own, hence this routine expects only
+   the raw deflate stream to decompress.  This is different from the default
+   behavior of inflate(), which expects a zlib header and trailer around the
+   deflate stream.
+
+     inflateBack() uses two subroutines supplied by the caller that are then
+   called by inflateBack() for input and output.  inflateBack() calls those
+   routines until it reads a complete deflate stream and writes out all of the
+   uncompressed data, or until it encounters an error.  The function's
+   parameters and return types are defined above in the in_func and out_func
+   typedefs.  inflateBack() will call in(in_desc, &buf) which should return the
+   number of bytes of provided input, and a pointer to that input in buf.  If
+   there is no input available, in() must return zero -- buf is ignored in that
+   case -- and inflateBack() will return a buffer error.  inflateBack() will
+   call out(out_desc, buf, len) to write the uncompressed data buf[0..len-1].
+   out() should return zero on success, or non-zero on failure.  If out()
+   returns non-zero, inflateBack() will return with an error.  Neither in() nor
+   out() are permitted to change the contents of the window provided to
+   inflateBackInit(), which is also the buffer that out() uses to write from.
+   The length written by out() will be at most the window size.  Any non-zero
+   amount of input may be provided by in().
+
+     For convenience, inflateBack() can be provided input on the first call by
+   setting strm->next_in and strm->avail_in.  If that input is exhausted, then
+   in() will be called.  Therefore strm->next_in must be initialized before
+   calling inflateBack().  If strm->next_in is Z_NULL, then in() will be called
+   immediately for input.  If strm->next_in is not Z_NULL, then strm->avail_in
+   must also be initialized, and then if strm->avail_in is not zero, input will
+   initially be taken from strm->next_in[0 ..  strm->avail_in - 1].
+
+     The in_desc and out_desc parameters of inflateBack() is passed as the
+   first parameter of in() and out() respectively when they are called.  These
+   descriptors can be optionally used to pass any information that the caller-
+   supplied in() and out() functions need to do their job.
+
+     On return, inflateBack() will set strm->next_in and strm->avail_in to
+   pass back any unused input that was provided by the last in() call.  The
+   return values of inflateBack() can be Z_STREAM_END on success, Z_BUF_ERROR
+   if in() or out() returned an error, Z_DATA_ERROR if there was a format error
+   in the deflate stream (in which case strm->msg is set to indicate the nature
+   of the error), or Z_STREAM_ERROR if the stream was not properly initialized.
+   In the case of Z_BUF_ERROR, an input or output error can be distinguished
+   using strm->next_in which will be Z_NULL only if in() returned an error.  If
+   strm->next_in is not Z_NULL, then the Z_BUF_ERROR was due to out() returning
+   non-zero.  (in() will always be called before out(), so strm->next_in is
+   assured to be defined if out() returns non-zero.)  Note that inflateBack()
+   cannot return Z_OK.
+*/
+
+ZEXTERN int ZEXPORT inflateBackEnd OF((z_streamp strm));
+/*
+     All memory allocated by inflateBackInit() is freed.
+
+     inflateBackEnd() returns Z_OK on success, or Z_STREAM_ERROR if the stream
+   state was inconsistent.
+*/
+
+ZEXTERN uLong ZEXPORT zlibCompileFlags OF((void));
+/* Return flags indicating compile-time options.
+
+    Type sizes, two bits each, 00 = 16 bits, 01 = 32, 10 = 64, 11 = other:
+     1.0: size of uInt
+     3.2: size of uLong
+     5.4: size of voidpf (pointer)
+     7.6: size of z_off_t
+
+    Compiler, assembler, and debug options:
+     8: ZLIB_DEBUG
+     9: ASMV or ASMINF -- use ASM code
+     10: ZLIB_WINAPI -- exported functions use the WINAPI calling convention
+     11: 0 (reserved)
+
+    One-time table building (smaller code, but not thread-safe if true):
+     12: BUILDFIXED -- build static block decoding tables when needed
+     13: DYNAMIC_CRC_TABLE -- build CRC calculation tables when needed
+     14,15: 0 (reserved)
+
+    Library content (indicates missing functionality):
+     16: NO_GZCOMPRESS -- gz* functions cannot compress (to avoid linking
+                          deflate code when not needed)
+     17: NO_GZIP -- deflate can't write gzip streams, and inflate can't detect
+                    and decode gzip streams (to avoid linking crc code)
+     18-19: 0 (reserved)
+
+    Operation variations (changes in library functionality):
+     20: PKZIP_BUG_WORKAROUND -- slightly more permissive inflate
+     21: FASTEST -- deflate algorithm with only one, lowest compression level
+     22,23: 0 (reserved)
+
+    The sprintf variant used by gzprintf (zero is best):
+     24: 0 = vs*, 1 = s* -- 1 means limited to 20 arguments after the format
+     25: 0 = *nprintf, 1 = *printf -- 1 means gzprintf() not secure!
+     26: 0 = returns value, 1 = void -- 1 means inferred string length returned
+
+    Remainder:
+     27-31: 0 (reserved)
+ */
+
+#ifndef Z_SOLO
+
+                        /* utility functions */
+
+/*
+     The following utility functions are implemented on top of the basic
+   stream-oriented functions.  To simplify the interface, some default options
+   are assumed (compression level and memory usage, standard memory allocation
+   functions).  The source code of these utility functions can be modified if
+   you need special options.
+*/
+
+ZEXTERN int ZEXPORT compress OF((Bytef *dest,   uLongf *destLen,
+                                 const Bytef *source, uLong sourceLen));
+/*
+     Compresses the source buffer into the destination buffer.  sourceLen is
+   the byte length of the source buffer.  Upon entry, destLen is the total size
+   of the destination buffer, which must be at least the value returned by
+   compressBound(sourceLen).  Upon exit, destLen is the actual size of the
+   compressed data.  compress() is equivalent to compress2() with a level
+   parameter of Z_DEFAULT_COMPRESSION.
+
+     compress returns Z_OK if success, Z_MEM_ERROR if there was not
+   enough memory, Z_BUF_ERROR if there was not enough room in the output
+   buffer.
+*/
+
+ZEXTERN int ZEXPORT compress2 OF((Bytef *dest,   uLongf *destLen,
+                                  const Bytef *source, uLong sourceLen,
+                                  int level));
+/*
+     Compresses the source buffer into the destination buffer.  The level
+   parameter has the same meaning as in deflateInit.  sourceLen is the byte
+   length of the source buffer.  Upon entry, destLen is the total size of the
+   destination buffer, which must be at least the value returned by
+   compressBound(sourceLen).  Upon exit, destLen is the actual size of the
+   compressed data.
+
+     compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough
+   memory, Z_BUF_ERROR if there was not enough room in the output buffer,
+   Z_STREAM_ERROR if the level parameter is invalid.
+*/
+
+ZEXTERN uLong ZEXPORT compressBound OF((uLong sourceLen));
+/*
+     compressBound() returns an upper bound on the compressed size after
+   compress() or compress2() on sourceLen bytes.  It would be used before a
+   compress() or compress2() call to allocate the destination buffer.
+*/
+
+ZEXTERN int ZEXPORT uncompress OF((Bytef *dest,   uLongf *destLen,
+                                   const Bytef *source, uLong sourceLen));
+/*
+     Decompresses the source buffer into the destination buffer.  sourceLen is
+   the byte length of the source buffer.  Upon entry, destLen is the total size
+   of the destination buffer, which must be large enough to hold the entire
+   uncompressed data.  (The size of the uncompressed data must have been saved
+   previously by the compressor and transmitted to the decompressor by some
+   mechanism outside the scope of this compression library.) Upon exit, destLen
+   is the actual size of the uncompressed data.
+
+     uncompress returns Z_OK if success, Z_MEM_ERROR if there was not
+   enough memory, Z_BUF_ERROR if there was not enough room in the output
+   buffer, or Z_DATA_ERROR if the input data was corrupted or incomplete.  In
+   the case where there is not enough room, uncompress() will fill the output
+   buffer with the uncompressed data up to that point.
+*/
+
+ZEXTERN int ZEXPORT uncompress2 OF((Bytef *dest,   uLongf *destLen,
+                                    const Bytef *source, uLong *sourceLen));
+/*
+     Same as uncompress, except that sourceLen is a pointer, where the
+   length of the source is *sourceLen.  On return, *sourceLen is the number of
+   source bytes consumed.
+*/
+
+                        /* gzip file access functions */
+
+/*
+     This library supports reading and writing files in gzip (.gz) format with
+   an interface similar to that of stdio, using the functions that start with
+   "gz".  The gzip format is different from the zlib format.  gzip is a gzip
+   wrapper, documented in RFC 1952, wrapped around a deflate stream.
+*/
+
+typedef struct gzFile_s *gzFile;    /* semi-opaque gzip file descriptor */
+
+/*
+ZEXTERN gzFile ZEXPORT gzopen OF((const char *path, const char *mode));
+
+     Opens a gzip (.gz) file for reading or writing.  The mode parameter is as
+   in fopen ("rb" or "wb") but can also include a compression level ("wb9") or
+   a strategy: 'f' for filtered data as in "wb6f", 'h' for Huffman-only
+   compression as in "wb1h", 'R' for run-length encoding as in "wb1R", or 'F'
+   for fixed code compression as in "wb9F".  (See the description of
+   deflateInit2 for more information about the strategy parameter.)  'T' will
+   request transparent writing or appending with no compression and not using
+   the gzip format.
+
+     "a" can be used instead of "w" to request that the gzip stream that will
+   be written be appended to the file.  "+" will result in an error, since
+   reading and writing to the same gzip file is not supported.  The addition of
+   "x" when writing will create the file exclusively, which fails if the file
+   already exists.  On systems that support it, the addition of "e" when
+   reading or writing will set the flag to close the file on an execve() call.
+
+     These functions, as well as gzip, will read and decode a sequence of gzip
+   streams in a file.  The append function of gzopen() can be used to create
+   such a file.  (Also see gzflush() for another way to do this.)  When
+   appending, gzopen does not test whether the file begins with a gzip stream,
+   nor does it look for the end of the gzip streams to begin appending.  gzopen
+   will simply append a gzip stream to the existing file.
+
+     gzopen can be used to read a file which is not in gzip format; in this
+   case gzread will directly read from the file without decompression.  When
+   reading, this will be detected automatically by looking for the magic two-
+   byte gzip header.
+
+     gzopen returns NULL if the file could not be opened, if there was
+   insufficient memory to allocate the gzFile state, or if an invalid mode was
+   specified (an 'r', 'w', or 'a' was not provided, or '+' was provided).
+   errno can be checked to determine if the reason gzopen failed was that the
+   file could not be opened.
+*/
+
+ZEXTERN gzFile ZEXPORT gzdopen OF((int fd, const char *mode));
+/*
+     gzdopen associates a gzFile with the file descriptor fd.  File descriptors
+   are obtained from calls like open, dup, creat, pipe or fileno (if the file
+   has been previously opened with fopen).  The mode parameter is as in gzopen.
+
+     The next call of gzclose on the returned gzFile will also close the file
+   descriptor fd, just like fclose(fdopen(fd, mode)) closes the file descriptor
+   fd.  If you want to keep fd open, use fd = dup(fd_keep); gz = gzdopen(fd,
+   mode);.  The duplicated descriptor should be saved to avoid a leak, since
+   gzdopen does not close fd if it fails.  If you are using fileno() to get the
+   file descriptor from a FILE *, then you will have to use dup() to avoid
+   double-close()ing the file descriptor.  Both gzclose() and fclose() will
+   close the associated file descriptor, so they need to have different file
+   descriptors.
+
+     gzdopen returns NULL if there was insufficient memory to allocate the
+   gzFile state, if an invalid mode was specified (an 'r', 'w', or 'a' was not
+   provided, or '+' was provided), or if fd is -1.  The file descriptor is not
+   used until the next gz* read, write, seek, or close operation, so gzdopen
+   will not detect if fd is invalid (unless fd is -1).
+*/
+
+ZEXTERN int ZEXPORT gzbuffer OF((gzFile file, unsigned size));
+/*
+     Set the internal buffer size used by this library's functions.  The
+   default buffer size is 8192 bytes.  This function must be called after
+   gzopen() or gzdopen(), and before any other calls that read or write the
+   file.  The buffer memory allocation is always deferred to the first read or
+   write.  Three times that size in buffer space is allocated.  A larger buffer
+   size of, for example, 64K or 128K bytes will noticeably increase the speed
+   of decompression (reading).
+
+     The new buffer size also affects the maximum length for gzprintf().
+
+     gzbuffer() returns 0 on success, or -1 on failure, such as being called
+   too late.
+*/
+
+ZEXTERN int ZEXPORT gzsetparams OF((gzFile file, int level, int strategy));
+/*
+     Dynamically update the compression level or strategy.  See the description
+   of deflateInit2 for the meaning of these parameters.  Previously provided
+   data is flushed before the parameter change.
+
+     gzsetparams returns Z_OK if success, Z_STREAM_ERROR if the file was not
+   opened for writing, Z_ERRNO if there is an error writing the flushed data,
+   or Z_MEM_ERROR if there is a memory allocation error.
+*/
+
+ZEXTERN int ZEXPORT gzread OF((gzFile file, voidp buf, unsigned len));
+/*
+     Reads the given number of uncompressed bytes from the compressed file.  If
+   the input file is not in gzip format, gzread copies the given number of
+   bytes into the buffer directly from the file.
+
+     After reaching the end of a gzip stream in the input, gzread will continue
+   to read, looking for another gzip stream.  Any number of gzip streams may be
+   concatenated in the input file, and will all be decompressed by gzread().
+   If something other than a gzip stream is encountered after a gzip stream,
+   that remaining trailing garbage is ignored (and no error is returned).
+
+     gzread can be used to read a gzip file that is being concurrently written.
+   Upon reaching the end of the input, gzread will return with the available
+   data.  If the error code returned by gzerror is Z_OK or Z_BUF_ERROR, then
+   gzclearerr can be used to clear the end of file indicator in order to permit
+   gzread to be tried again.  Z_OK indicates that a gzip stream was completed
+   on the last gzread.  Z_BUF_ERROR indicates that the input file ended in the
+   middle of a gzip stream.  Note that gzread does not return -1 in the event
+   of an incomplete gzip stream.  This error is deferred until gzclose(), which
+   will return Z_BUF_ERROR if the last gzread ended in the middle of a gzip
+   stream.  Alternatively, gzerror can be used before gzclose to detect this
+   case.
+
+     gzread returns the number of uncompressed bytes actually read, less than
+   len for end of file, or -1 for error.  If len is too large to fit in an int,
+   then nothing is read, -1 is returned, and the error state is set to
+   Z_STREAM_ERROR.
+*/
+
+ZEXTERN z_size_t ZEXPORT gzfread OF((voidp buf, z_size_t size, z_size_t nitems,
+                                     gzFile file));
+/*
+     Read up to nitems items of size size from file to buf, otherwise operating
+   as gzread() does.  This duplicates the interface of stdio's fread(), with
+   size_t request and return types.  If the library defines size_t, then
+   z_size_t is identical to size_t.  If not, then z_size_t is an unsigned
+   integer type that can contain a pointer.
+
+     gzfread() returns the number of full items read of size size, or zero if
+   the end of the file was reached and a full item could not be read, or if
+   there was an error.  gzerror() must be consulted if zero is returned in
+   order to determine if there was an error.  If the multiplication of size and
+   nitems overflows, i.e. the product does not fit in a z_size_t, then nothing
+   is read, zero is returned, and the error state is set to Z_STREAM_ERROR.
+
+     In the event that the end of file is reached and only a partial item is
+   available at the end, i.e. the remaining uncompressed data length is not a
+   multiple of size, then the final partial item is nevetheless read into buf
+   and the end-of-file flag is set.  The length of the partial item read is not
+   provided, but could be inferred from the result of gztell().  This behavior
+   is the same as the behavior of fread() implementations in common libraries,
+   but it prevents the direct use of gzfread() to read a concurrently written
+   file, reseting and retrying on end-of-file, when size is not 1.
+*/
+
+ZEXTERN int ZEXPORT gzwrite OF((gzFile file,
+                                voidpc buf, unsigned len));
+/*
+     Writes the given number of uncompressed bytes into the compressed file.
+   gzwrite returns the number of uncompressed bytes written or 0 in case of
+   error.
+*/
+
+ZEXTERN z_size_t ZEXPORT gzfwrite OF((voidpc buf, z_size_t size,
+                                      z_size_t nitems, gzFile file));
+/*
+     gzfwrite() writes nitems items of size size from buf to file, duplicating
+   the interface of stdio's fwrite(), with size_t request and return types.  If
+   the library defines size_t, then z_size_t is identical to size_t.  If not,
+   then z_size_t is an unsigned integer type that can contain a pointer.
+
+     gzfwrite() returns the number of full items written of size size, or zero
+   if there was an error.  If the multiplication of size and nitems overflows,
+   i.e. the product does not fit in a z_size_t, then nothing is written, zero
+   is returned, and the error state is set to Z_STREAM_ERROR.
+*/
+
+ZEXTERN int ZEXPORTVA gzprintf Z_ARG((gzFile file, const char *format, ...));
+/*
+     Converts, formats, and writes the arguments to the compressed file under
+   control of the format string, as in fprintf.  gzprintf returns the number of
+   uncompressed bytes actually written, or a negative zlib error code in case
+   of error.  The number of uncompressed bytes written is limited to 8191, or
+   one less than the buffer size given to gzbuffer().  The caller should assure
+   that this limit is not exceeded.  If it is exceeded, then gzprintf() will
+   return an error (0) with nothing written.  In this case, there may also be a
+   buffer overflow with unpredictable consequences, which is possible only if
+   zlib was compiled with the insecure functions sprintf() or vsprintf()
+   because the secure snprintf() or vsnprintf() functions were not available.
+   This can be determined using zlibCompileFlags().
+*/
+
+ZEXTERN int ZEXPORT gzputs OF((gzFile file, const char *s));
+/*
+     Writes the given null-terminated string to the compressed file, excluding
+   the terminating null character.
+
+     gzputs returns the number of characters written, or -1 in case of error.
+*/
+
+ZEXTERN char * ZEXPORT gzgets OF((gzFile file, char *buf, int len));
+/*
+     Reads bytes from the compressed file until len-1 characters are read, or a
+   newline character is read and transferred to buf, or an end-of-file
+   condition is encountered.  If any characters are read or if len == 1, the
+   string is terminated with a null character.  If no characters are read due
+   to an end-of-file or len < 1, then the buffer is left untouched.
+
+     gzgets returns buf which is a null-terminated string, or it returns NULL
+   for end-of-file or in case of error.  If there was an error, the contents at
+   buf are indeterminate.
+*/
+
+ZEXTERN int ZEXPORT gzputc OF((gzFile file, int c));
+/*
+     Writes c, converted to an unsigned char, into the compressed file.  gzputc
+   returns the value that was written, or -1 in case of error.
+*/
+
+ZEXTERN int ZEXPORT gzgetc OF((gzFile file));
+/*
+     Reads one byte from the compressed file.  gzgetc returns this byte or -1
+   in case of end of file or error.  This is implemented as a macro for speed.
+   As such, it does not do all of the checking the other functions do.  I.e.
+   it does not check to see if file is NULL, nor whether the structure file
+   points to has been clobbered or not.
+*/
+
+ZEXTERN int ZEXPORT gzungetc OF((int c, gzFile file));
+/*
+     Push one character back onto the stream to be read as the first character
+   on the next read.  At least one character of push-back is allowed.
+   gzungetc() returns the character pushed, or -1 on failure.  gzungetc() will
+   fail if c is -1, and may fail if a character has been pushed but not read
+   yet.  If gzungetc is used immediately after gzopen or gzdopen, at least the
+   output buffer size of pushed characters is allowed.  (See gzbuffer above.)
+   The pushed character will be discarded if the stream is repositioned with
+   gzseek() or gzrewind().
+*/
+
+ZEXTERN int ZEXPORT gzflush OF((gzFile file, int flush));
+/*
+     Flushes all pending output into the compressed file.  The parameter flush
+   is as in the deflate() function.  The return value is the zlib error number
+   (see function gzerror below).  gzflush is only permitted when writing.
+
+     If the flush parameter is Z_FINISH, the remaining data is written and the
+   gzip stream is completed in the output.  If gzwrite() is called again, a new
+   gzip stream will be started in the output.  gzread() is able to read such
+   concatenated gzip streams.
+
+     gzflush should be called only when strictly necessary because it will
+   degrade compression if called too often.
+*/
+
+/*
+ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile file,
+                                   z_off_t offset, int whence));
+
+     Sets the starting position for the next gzread or gzwrite on the given
+   compressed file.  The offset represents a number of bytes in the
+   uncompressed data stream.  The whence parameter is defined as in lseek(2);
+   the value SEEK_END is not supported.
+
+     If the file is opened for reading, this function is emulated but can be
+   extremely slow.  If the file is opened for writing, only forward seeks are
+   supported; gzseek then compresses a sequence of zeroes up to the new
+   starting position.
+
+     gzseek returns the resulting offset location as measured in bytes from
+   the beginning of the uncompressed stream, or -1 in case of error, in
+   particular if the file is opened for writing and the new starting position
+   would be before the current position.
+*/
+
+ZEXTERN int ZEXPORT    gzrewind OF((gzFile file));
+/*
+     Rewinds the given file. This function is supported only for reading.
+
+     gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET)
+*/
+
+/*
+ZEXTERN z_off_t ZEXPORT    gztell OF((gzFile file));
+
+     Returns the starting position for the next gzread or gzwrite on the given
+   compressed file.  This position represents a number of bytes in the
+   uncompressed data stream, and is zero when starting, even if appending or
+   reading a gzip stream from the middle of a file using gzdopen().
+
+     gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR)
+*/
+
+/*
+ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile file));
+
+     Returns the current offset in the file being read or written.  This offset
+   includes the count of bytes that precede the gzip stream, for example when
+   appending or when using gzdopen() for reading.  When reading, the offset
+   does not include as yet unused buffered input.  This information can be used
+   for a progress indicator.  On error, gzoffset() returns -1.
+*/
+
+ZEXTERN int ZEXPORT gzeof OF((gzFile file));
+/*
+     Returns true (1) if the end-of-file indicator has been set while reading,
+   false (0) otherwise.  Note that the end-of-file indicator is set only if the
+   read tried to go past the end of the input, but came up short.  Therefore,
+   just like feof(), gzeof() may return false even if there is no more data to
+   read, in the event that the last read request was for the exact number of
+   bytes remaining in the input file.  This will happen if the input file size
+   is an exact multiple of the buffer size.
+
+     If gzeof() returns true, then the read functions will return no more data,
+   unless the end-of-file indicator is reset by gzclearerr() and the input file
+   has grown since the previous end of file was detected.
+*/
+
+ZEXTERN int ZEXPORT gzdirect OF((gzFile file));
+/*
+     Returns true (1) if file is being copied directly while reading, or false
+   (0) if file is a gzip stream being decompressed.
+
+     If the input file is empty, gzdirect() will return true, since the input
+   does not contain a gzip stream.
+
+     If gzdirect() is used immediately after gzopen() or gzdopen() it will
+   cause buffers to be allocated to allow reading the file to determine if it
+   is a gzip file.  Therefore if gzbuffer() is used, it should be called before
+   gzdirect().
+
+     When writing, gzdirect() returns true (1) if transparent writing was
+   requested ("wT" for the gzopen() mode), or false (0) otherwise.  (Note:
+   gzdirect() is not needed when writing.  Transparent writing must be
+   explicitly requested, so the application already knows the answer.  When
+   linking statically, using gzdirect() will include all of the zlib code for
+   gzip file reading and decompression, which may not be desired.)
+*/
+
+ZEXTERN int ZEXPORT    gzclose OF((gzFile file));
+/*
+     Flushes all pending output if necessary, closes the compressed file and
+   deallocates the (de)compression state.  Note that once file is closed, you
+   cannot call gzerror with file, since its structures have been deallocated.
+   gzclose must not be called more than once on the same file, just as free
+   must not be called more than once on the same allocation.
+
+     gzclose will return Z_STREAM_ERROR if file is not valid, Z_ERRNO on a
+   file operation error, Z_MEM_ERROR if out of memory, Z_BUF_ERROR if the
+   last read ended in the middle of a gzip stream, or Z_OK on success.
+*/
+
+ZEXTERN int ZEXPORT gzclose_r OF((gzFile file));
+ZEXTERN int ZEXPORT gzclose_w OF((gzFile file));
+/*
+     Same as gzclose(), but gzclose_r() is only for use when reading, and
+   gzclose_w() is only for use when writing or appending.  The advantage to
+   using these instead of gzclose() is that they avoid linking in zlib
+   compression or decompression code that is not used when only reading or only
+   writing respectively.  If gzclose() is used, then both compression and
+   decompression code will be included the application when linking to a static
+   zlib library.
+*/
+
+ZEXTERN const char * ZEXPORT gzerror OF((gzFile file, int *errnum));
+/*
+     Returns the error message for the last error which occurred on the given
+   compressed file.  errnum is set to zlib error number.  If an error occurred
+   in the file system and not in the compression library, errnum is set to
+   Z_ERRNO and the application may consult errno to get the exact error code.
+
+     The application must not modify the returned string.  Future calls to
+   this function may invalidate the previously returned string.  If file is
+   closed, then the string previously returned by gzerror will no longer be
+   available.
+
+     gzerror() should be used to distinguish errors from end-of-file for those
+   functions above that do not distinguish those cases in their return values.
+*/
+
+ZEXTERN void ZEXPORT gzclearerr OF((gzFile file));
+/*
+     Clears the error and end-of-file flags for file.  This is analogous to the
+   clearerr() function in stdio.  This is useful for continuing to read a gzip
+   file that is being written concurrently.
+*/
+
+#endif /* !Z_SOLO */
+
+                        /* checksum functions */
+
+/*
+     These functions are not related to compression but are exported
+   anyway because they might be useful in applications using the compression
+   library.
+*/
+
+ZEXTERN uLong ZEXPORT adler32 OF((uLong adler, const Bytef *buf, uInt len));
+/*
+     Update a running Adler-32 checksum with the bytes buf[0..len-1] and
+   return the updated checksum.  If buf is Z_NULL, this function returns the
+   required initial value for the checksum.
+
+     An Adler-32 checksum is almost as reliable as a CRC-32 but can be computed
+   much faster.
+
+   Usage example:
+
+     uLong adler = adler32(0L, Z_NULL, 0);
+
+     while (read_buffer(buffer, length) != EOF) {
+       adler = adler32(adler, buffer, length);
+     }
+     if (adler != original_adler) error();
+*/
+
+ZEXTERN uLong ZEXPORT adler32_z OF((uLong adler, const Bytef *buf,
+                                    z_size_t len));
+/*
+     Same as adler32(), but with a size_t length.
+*/
+
+/*
+ZEXTERN uLong ZEXPORT adler32_combine OF((uLong adler1, uLong adler2,
+                                          z_off_t len2));
+
+     Combine two Adler-32 checksums into one.  For two sequences of bytes, seq1
+   and seq2 with lengths len1 and len2, Adler-32 checksums were calculated for
+   each, adler1 and adler2.  adler32_combine() returns the Adler-32 checksum of
+   seq1 and seq2 concatenated, requiring only adler1, adler2, and len2.  Note
+   that the z_off_t type (like off_t) is a signed integer.  If len2 is
+   negative, the result has no meaning or utility.
+*/
+
+ZEXTERN uLong ZEXPORT crc32   OF((uLong crc, const Bytef *buf, uInt len));
+/*
+     Update a running CRC-32 with the bytes buf[0..len-1] and return the
+   updated CRC-32.  If buf is Z_NULL, this function returns the required
+   initial value for the crc.  Pre- and post-conditioning (one's complement) is
+   performed within this function so it shouldn't be done by the application.
+
+   Usage example:
+
+     uLong crc = crc32(0L, Z_NULL, 0);
+
+     while (read_buffer(buffer, length) != EOF) {
+       crc = crc32(crc, buffer, length);
+     }
+     if (crc != original_crc) error();
+*/
+
+ZEXTERN uLong ZEXPORT crc32_z OF((uLong adler, const Bytef *buf,
+                                  z_size_t len));
+/*
+     Same as crc32(), but with a size_t length.
+*/
+
+/*
+ZEXTERN uLong ZEXPORT crc32_combine OF((uLong crc1, uLong crc2, z_off_t len2));
+
+     Combine two CRC-32 check values into one.  For two sequences of bytes,
+   seq1 and seq2 with lengths len1 and len2, CRC-32 check values were
+   calculated for each, crc1 and crc2.  crc32_combine() returns the CRC-32
+   check value of seq1 and seq2 concatenated, requiring only crc1, crc2, and
+   len2.
+*/
+
+
+                        /* various hacks, don't look :) */
+
+/* deflateInit and inflateInit are macros to allow checking the zlib version
+ * and the compiler's view of z_stream:
+ */
+ZEXTERN int ZEXPORT deflateInit_ OF((z_streamp strm, int level,
+                                     const char *version, int stream_size));
+ZEXTERN int ZEXPORT inflateInit_ OF((z_streamp strm,
+                                     const char *version, int stream_size));
+ZEXTERN int ZEXPORT deflateInit2_ OF((z_streamp strm, int  level, int  method,
+                                      int windowBits, int memLevel,
+                                      int strategy, const char *version,
+                                      int stream_size));
+ZEXTERN int ZEXPORT inflateInit2_ OF((z_streamp strm, int  windowBits,
+                                      const char *version, int stream_size));
+ZEXTERN int ZEXPORT inflateBackInit_ OF((z_streamp strm, int windowBits,
+                                         unsigned char FAR *window,
+                                         const char *version,
+                                         int stream_size));
+#ifdef Z_PREFIX_SET
+#  define z_deflateInit(strm, level) \
+          deflateInit_((strm), (level), ZLIB_VERSION, (int)sizeof(z_stream))
+#  define z_inflateInit(strm) \
+          inflateInit_((strm), ZLIB_VERSION, (int)sizeof(z_stream))
+#  define z_deflateInit2(strm, level, method, windowBits, memLevel, strategy) \
+          deflateInit2_((strm),(level),(method),(windowBits),(memLevel),\
+                        (strategy), ZLIB_VERSION, (int)sizeof(z_stream))
+#  define z_inflateInit2(strm, windowBits) \
+          inflateInit2_((strm), (windowBits), ZLIB_VERSION, \
+                        (int)sizeof(z_stream))
+#  define z_inflateBackInit(strm, windowBits, window) \
+          inflateBackInit_((strm), (windowBits), (window), \
+                           ZLIB_VERSION, (int)sizeof(z_stream))
+#else
+#  define deflateInit(strm, level) \
+          deflateInit_((strm), (level), ZLIB_VERSION, (int)sizeof(z_stream))
+#  define inflateInit(strm) \
+          inflateInit_((strm), ZLIB_VERSION, (int)sizeof(z_stream))
+#  define deflateInit2(strm, level, method, windowBits, memLevel, strategy) \
+          deflateInit2_((strm),(level),(method),(windowBits),(memLevel),\
+                        (strategy), ZLIB_VERSION, (int)sizeof(z_stream))
+#  define inflateInit2(strm, windowBits) \
+          inflateInit2_((strm), (windowBits), ZLIB_VERSION, \
+                        (int)sizeof(z_stream))
+#  define inflateBackInit(strm, windowBits, window) \
+          inflateBackInit_((strm), (windowBits), (window), \
+                           ZLIB_VERSION, (int)sizeof(z_stream))
+#endif
+
+#ifndef Z_SOLO
+
+/* gzgetc() macro and its supporting function and exposed data structure.  Note
+ * that the real internal state is much larger than the exposed structure.
+ * This abbreviated structure exposes just enough for the gzgetc() macro.  The
+ * user should not mess with these exposed elements, since their names or
+ * behavior could change in the future, perhaps even capriciously.  They can
+ * only be used by the gzgetc() macro.  You have been warned.
+ */
+struct gzFile_s {
+    unsigned have;
+    unsigned char *next;
+    z_off64_t pos;
+};
+ZEXTERN int ZEXPORT gzgetc_ OF((gzFile file));  /* backward compatibility */
+#ifdef Z_PREFIX_SET
+#  undef z_gzgetc
+#  define z_gzgetc(g) \
+          ((g)->have ? ((g)->have--, (g)->pos++, *((g)->next)++) : (gzgetc)(g))
+#else
+#  define gzgetc(g) \
+          ((g)->have ? ((g)->have--, (g)->pos++, *((g)->next)++) : (gzgetc)(g))
+#endif
+
+/* provide 64-bit offset functions if _LARGEFILE64_SOURCE defined, and/or
+ * change the regular functions to 64 bits if _FILE_OFFSET_BITS is 64 (if
+ * both are true, the application gets the *64 functions, and the regular
+ * functions are changed to 64 bits) -- in case these are set on systems
+ * without large file support, _LFS64_LARGEFILE must also be true
+ */
+#ifdef Z_LARGE64
+   ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *));
+   ZEXTERN z_off64_t ZEXPORT gzseek64 OF((gzFile, z_off64_t, int));
+   ZEXTERN z_off64_t ZEXPORT gztell64 OF((gzFile));
+   ZEXTERN z_off64_t ZEXPORT gzoffset64 OF((gzFile));
+   ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off64_t));
+   ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off64_t));
+#endif
+
+#if !defined(ZLIB_INTERNAL) && defined(Z_WANT64)
+#  ifdef Z_PREFIX_SET
+#    define z_gzopen z_gzopen64
+#    define z_gzseek z_gzseek64
+#    define z_gztell z_gztell64
+#    define z_gzoffset z_gzoffset64
+#    define z_adler32_combine z_adler32_combine64
+#    define z_crc32_combine z_crc32_combine64
+#  else
+#    define gzopen gzopen64
+#    define gzseek gzseek64
+#    define gztell gztell64
+#    define gzoffset gzoffset64
+#    define adler32_combine adler32_combine64
+#    define crc32_combine crc32_combine64
+#  endif
+#  ifndef Z_LARGE64
+     ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *));
+     ZEXTERN z_off_t ZEXPORT gzseek64 OF((gzFile, z_off_t, int));
+     ZEXTERN z_off_t ZEXPORT gztell64 OF((gzFile));
+     ZEXTERN z_off_t ZEXPORT gzoffset64 OF((gzFile));
+     ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t));
+     ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t));
+#  endif
+#else
+   ZEXTERN gzFile ZEXPORT gzopen OF((const char *, const char *));
+   ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile, z_off_t, int));
+   ZEXTERN z_off_t ZEXPORT gztell OF((gzFile));
+   ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile));
+   ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t));
+   ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t));
+#endif
+
+#else /* Z_SOLO */
+
+   ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t));
+   ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t));
+
+#endif /* !Z_SOLO */
+
+/* undocumented functions */
+ZEXTERN const char   * ZEXPORT zError           OF((int));
+ZEXTERN int            ZEXPORT inflateSyncPoint OF((z_streamp));
+ZEXTERN const z_crc_t FAR * ZEXPORT get_crc_table    OF((void));
+ZEXTERN int            ZEXPORT inflateUndermine OF((z_streamp, int));
+ZEXTERN int            ZEXPORT inflateValidate OF((z_streamp, int));
+ZEXTERN unsigned long  ZEXPORT inflateCodesUsed OF ((z_streamp));
+ZEXTERN int            ZEXPORT inflateResetKeep OF((z_streamp));
+ZEXTERN int            ZEXPORT deflateResetKeep OF((z_streamp));
+#if (defined(_WIN32) || defined(__CYGWIN__)) && !defined(Z_SOLO)
+ZEXTERN gzFile         ZEXPORT gzopen_w OF((const wchar_t *path,
+                                            const char *mode));
+#endif
+#if defined(STDC) || defined(Z_HAVE_STDARG_H)
+#  ifndef Z_SOLO
+ZEXTERN int            ZEXPORTVA gzvprintf Z_ARG((gzFile file,
+                                                  const char *format,
+                                                  va_list va));
+#  endif
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* ZLIB_H */
diff --git a/distribution/zlib/lib/arm64-v8a/libz.a b/distribution/zlib/lib/arm64-v8a/libz.a
new file mode 100644
index 0000000..035371c
Binary files /dev/null and b/distribution/zlib/lib/arm64-v8a/libz.a differ
diff --git a/distribution/zlib/lib/armeabi-v7a/libz.a b/distribution/zlib/lib/armeabi-v7a/libz.a
new file mode 100644
index 0000000..1ab33d5
Binary files /dev/null and b/distribution/zlib/lib/armeabi-v7a/libz.a differ
diff --git a/distribution/zlib/lib/x86/libz.a b/distribution/zlib/lib/x86/libz.a
new file mode 100644
index 0000000..9b825ed
Binary files /dev/null and b/distribution/zlib/lib/x86/libz.a differ
diff --git a/distribution/zlib/lib/x86_64/libz.a b/distribution/zlib/lib/x86_64/libz.a
new file mode 100644
index 0000000..2792d1c
Binary files /dev/null and b/distribution/zlib/lib/x86_64/libz.a differ
diff --git a/distribution/zlib/lib19/armeabi-v7a/libz.a b/distribution/zlib/lib19/armeabi-v7a/libz.a
new file mode 100644
index 0000000..462e087
Binary files /dev/null and b/distribution/zlib/lib19/armeabi-v7a/libz.a differ
diff --git a/distribution/zlib/lib19/x86/libz.a b/distribution/zlib/lib19/x86/libz.a
new file mode 100644
index 0000000..8be1284
Binary files /dev/null and b/distribution/zlib/lib19/x86/libz.a differ
diff --git a/gradle.properties b/gradle.properties
new file mode 100644
index 0000000..aac7c9b
--- /dev/null
+++ b/gradle.properties
@@ -0,0 +1,17 @@
+# Project-wide Gradle settings.
+
+# IDE (e.g. Android Studio) users:
+# Gradle settings configured through the IDE *will override*
+# any settings specified in this file.
+
+# For more details on how to configure your build environment visit
+# http://www.gradle.org/docs/current/userguide/build_environment.html
+
+# Specifies the JVM arguments used for the daemon process.
+# The setting is particularly useful for tweaking memory settings.
+org.gradle.jvmargs=-Xmx1536m
+
+# When configured, Gradle will run in incubating parallel mode.
+# This option should only be used with decoupled projects. More details, visit
+# http://www.gradle.org/docs/current/userguide/multi_project_builds.html#sec:decoupled_projects
+# org.gradle.parallel=true
diff --git a/gradle/wrapper/gradle-wrapper.jar b/gradle/wrapper/gradle-wrapper.jar
new file mode 100644
index 0000000..13372ae
Binary files /dev/null and b/gradle/wrapper/gradle-wrapper.jar differ
diff --git a/gradle/wrapper/gradle-wrapper.properties b/gradle/wrapper/gradle-wrapper.properties
new file mode 100644
index 0000000..815a4fe
--- /dev/null
+++ b/gradle/wrapper/gradle-wrapper.properties
@@ -0,0 +1,6 @@
+#Tue Feb 22 00:48:31 MSK 2022
+distributionBase=GRADLE_USER_HOME
+distributionUrl=https\://services.gradle.org/distributions/gradle-5.6.4-bin.zip
+distributionPath=wrapper/dists
+zipStorePath=wrapper/dists
+zipStoreBase=GRADLE_USER_HOME
diff --git a/gradlew b/gradlew
new file mode 100644
index 0000000..9d82f78
--- /dev/null
+++ b/gradlew
@@ -0,0 +1,160 @@
+#!/usr/bin/env bash
+
+##############################################################################
+##
+##  Gradle start up script for UN*X
+##
+##############################################################################
+
+# Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
+DEFAULT_JVM_OPTS=""
+
+APP_NAME="Gradle"
+APP_BASE_NAME=`basename "$0"`
+
+# Use the maximum available, or set MAX_FD != -1 to use that value.
+MAX_FD="maximum"
+
+warn ( ) {
+    echo "$*"
+}
+
+die ( ) {
+    echo
+    echo "$*"
+    echo
+    exit 1
+}
+
+# OS specific support (must be 'true' or 'false').
+cygwin=false
+msys=false
+darwin=false
+case "`uname`" in
+  CYGWIN* )
+    cygwin=true
+    ;;
+  Darwin* )
+    darwin=true
+    ;;
+  MINGW* )
+    msys=true
+    ;;
+esac
+
+# Attempt to set APP_HOME
+# Resolve links: $0 may be a link
+PRG="$0"
+# Need this for relative symlinks.
+while [ -h "$PRG" ] ; do
+    ls=`ls -ld "$PRG"`
+    link=`expr "$ls" : '.*-> \(.*\)$'`
+    if expr "$link" : '/.*' > /dev/null; then
+        PRG="$link"
+    else
+        PRG=`dirname "$PRG"`"/$link"
+    fi
+done
+SAVED="`pwd`"
+cd "`dirname \"$PRG\"`/" >/dev/null
+APP_HOME="`pwd -P`"
+cd "$SAVED" >/dev/null
+
+CLASSPATH=$APP_HOME/gradle/wrapper/gradle-wrapper.jar
+
+# Determine the Java command to use to start the JVM.
+if [ -n "$JAVA_HOME" ] ; then
+    if [ -x "$JAVA_HOME/jre/sh/java" ] ; then
+        # IBM's JDK on AIX uses strange locations for the executables
+        JAVACMD="$JAVA_HOME/jre/sh/java"
+    else
+        JAVACMD="$JAVA_HOME/bin/java"
+    fi
+    if [ ! -x "$JAVACMD" ] ; then
+        die "ERROR: JAVA_HOME is set to an invalid directory: $JAVA_HOME
+
+Please set the JAVA_HOME variable in your environment to match the
+location of your Java installation."
+    fi
+else
+    JAVACMD="java"
+    which java >/dev/null 2>&1 || die "ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
+
+Please set the JAVA_HOME variable in your environment to match the
+location of your Java installation."
+fi
+
+# Increase the maximum file descriptors if we can.
+if [ "$cygwin" = "false" -a "$darwin" = "false" ] ; then
+    MAX_FD_LIMIT=`ulimit -H -n`
+    if [ $? -eq 0 ] ; then
+        if [ "$MAX_FD" = "maximum" -o "$MAX_FD" = "max" ] ; then
+            MAX_FD="$MAX_FD_LIMIT"
+        fi
+        ulimit -n $MAX_FD
+        if [ $? -ne 0 ] ; then
+            warn "Could not set maximum file descriptor limit: $MAX_FD"
+        fi
+    else
+        warn "Could not query maximum file descriptor limit: $MAX_FD_LIMIT"
+    fi
+fi
+
+# For Darwin, add options to specify how the application appears in the dock
+if $darwin; then
+    GRADLE_OPTS="$GRADLE_OPTS \"-Xdock:name=$APP_NAME\" \"-Xdock:icon=$APP_HOME/media/gradle.icns\""
+fi
+
+# For Cygwin, switch paths to Windows format before running java
+if $cygwin ; then
+    APP_HOME=`cygpath --path --mixed "$APP_HOME"`
+    CLASSPATH=`cygpath --path --mixed "$CLASSPATH"`
+    JAVACMD=`cygpath --unix "$JAVACMD"`
+
+    # We build the pattern for arguments to be converted via cygpath
+    ROOTDIRSRAW=`find -L / -maxdepth 1 -mindepth 1 -type d 2>/dev/null`
+    SEP=""
+    for dir in $ROOTDIRSRAW ; do
+        ROOTDIRS="$ROOTDIRS$SEP$dir"
+        SEP="|"
+    done
+    OURCYGPATTERN="(^($ROOTDIRS))"
+    # Add a user-defined pattern to the cygpath arguments
+    if [ "$GRADLE_CYGPATTERN" != "" ] ; then
+        OURCYGPATTERN="$OURCYGPATTERN|($GRADLE_CYGPATTERN)"
+    fi
+    # Now convert the arguments - kludge to limit ourselves to /bin/sh
+    i=0
+    for arg in "$@" ; do
+        CHECK=`echo "$arg"|egrep -c "$OURCYGPATTERN" -`
+        CHECK2=`echo "$arg"|egrep -c "^-"`                                 ### Determine if an option
+
+        if [ $CHECK -ne 0 ] && [ $CHECK2 -eq 0 ] ; then                    ### Added a condition
+            eval `echo args$i`=`cygpath --path --ignore --mixed "$arg"`
+        else
+            eval `echo args$i`="\"$arg\""
+        fi
+        i=$((i+1))
+    done
+    case $i in
+        (0) set -- ;;
+        (1) set -- "$args0" ;;
+        (2) set -- "$args0" "$args1" ;;
+        (3) set -- "$args0" "$args1" "$args2" ;;
+        (4) set -- "$args0" "$args1" "$args2" "$args3" ;;
+        (5) set -- "$args0" "$args1" "$args2" "$args3" "$args4" ;;
+        (6) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" ;;
+        (7) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" ;;
+        (8) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" "$args7" ;;
+        (9) set -- "$args0" "$args1" "$args2" "$args3" "$args4" "$args5" "$args6" "$args7" "$args8" ;;
+    esac
+fi
+
+# Split up the JVM_OPTS And GRADLE_OPTS values into an array, following the shell quoting and substitution rules
+function splitJvmOpts() {
+    JVM_OPTS=("$@")
+}
+eval splitJvmOpts $DEFAULT_JVM_OPTS $JAVA_OPTS $GRADLE_OPTS
+JVM_OPTS[${#JVM_OPTS[*]}]="-Dorg.gradle.appname=$APP_BASE_NAME"
+
+exec "$JAVACMD" "${JVM_OPTS[@]}" -classpath "$CLASSPATH" org.gradle.wrapper.GradleWrapperMain "$@"
diff --git a/gradlew.bat b/gradlew.bat
new file mode 100644
index 0000000..8a0b282
--- /dev/null
+++ b/gradlew.bat
@@ -0,0 +1,90 @@
+@if "%DEBUG%" == "" @echo off
+@rem ##########################################################################
+@rem
+@rem  Gradle startup script for Windows
+@rem
+@rem ##########################################################################
+
+@rem Set local scope for the variables with windows NT shell
+if "%OS%"=="Windows_NT" setlocal
+
+@rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
+set DEFAULT_JVM_OPTS=
+
+set DIRNAME=%~dp0
+if "%DIRNAME%" == "" set DIRNAME=.
+set APP_BASE_NAME=%~n0
+set APP_HOME=%DIRNAME%
+
+@rem Find java.exe
+if defined JAVA_HOME goto findJavaFromJavaHome
+
+set JAVA_EXE=java.exe
+%JAVA_EXE% -version >NUL 2>&1
+if "%ERRORLEVEL%" == "0" goto init
+
+echo.
+echo ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
+echo.
+echo Please set the JAVA_HOME variable in your environment to match the
+echo location of your Java installation.
+
+goto fail
+
+:findJavaFromJavaHome
+set JAVA_HOME=%JAVA_HOME:"=%
+set JAVA_EXE=%JAVA_HOME%/bin/java.exe
+
+if exist "%JAVA_EXE%" goto init
+
+echo.
+echo ERROR: JAVA_HOME is set to an invalid directory: %JAVA_HOME%
+echo.
+echo Please set the JAVA_HOME variable in your environment to match the
+echo location of your Java installation.
+
+goto fail
+
+:init
+@rem Get command-line arguments, handling Windowz variants
+
+if not "%OS%" == "Windows_NT" goto win9xME_args
+if "%@eval[2+2]" == "4" goto 4NT_args
+
+:win9xME_args
+@rem Slurp the command line arguments.
+set CMD_LINE_ARGS=
+set _SKIP=2
+
+:win9xME_args_slurp
+if "x%~1" == "x" goto execute
+
+set CMD_LINE_ARGS=%*
+goto execute
+
+:4NT_args
+@rem Get arguments from the 4NT Shell from JP Software
+set CMD_LINE_ARGS=%$
+
+:execute
+@rem Setup the command line
+
+set CLASSPATH=%APP_HOME%\gradle\wrapper\gradle-wrapper.jar
+
+@rem Execute Gradle
+"%JAVA_EXE%" %DEFAULT_JVM_OPTS% %JAVA_OPTS% %GRADLE_OPTS% "-Dorg.gradle.appname=%APP_BASE_NAME%" -classpath "%CLASSPATH%" org.gradle.wrapper.GradleWrapperMain %CMD_LINE_ARGS%
+
+:end
+@rem End local scope for the variables with windows NT shell
+if "%ERRORLEVEL%"=="0" goto mainEnd
+
+:fail
+rem Set variable GRADLE_EXIT_CONSOLE if you need the _script_ return code instead of
+rem the _cmd.exe /c_ return code!
+if  not "" == "%GRADLE_EXIT_CONSOLE%" exit 1
+exit /b 1
+
+:mainEnd
+if "%OS%"=="Windows_NT" endlocal
+
+:omega
diff --git a/img/01.png b/img/01.png
new file mode 100644
index 0000000..6711175
Binary files /dev/null and b/img/01.png differ
diff --git a/img/02.png b/img/02.png
new file mode 100644
index 0000000..73dcaf3
Binary files /dev/null and b/img/02.png differ
diff --git a/img/ytn_and_miner_screen.jpg b/img/ytn_and_miner_screen.jpg
new file mode 100644
index 0000000..01f1016
Binary files /dev/null and b/img/ytn_and_miner_screen.jpg differ
diff --git a/settings.gradle b/settings.gradle
new file mode 100644
index 0000000..5cd7c48
--- /dev/null
+++ b/settings.gradle
@@ -0,0 +1 @@
+include ':app', ':bitzenymininglibrary'