Android dex加载过程分析

啊啊啊啊啊啊啊，开始吧。

ClassLoader有两种，PathClassLoader和DexClassLoader

public class DexClassLoader extends BaseDexClassLoader {
 
    public DexClassLoader(String dexPath, String optimizedDirectory,
            String libraryPath, ClassLoader parent) {
        super(dexPath, new File(optimizedDirectory), libraryPath, parent);
    }
}

public class PathClassLoader extends BaseDexClassLoader {
 
    public PathClassLoader(String dexPath, ClassLoader parent) {
        super(dexPath, null, null, parent);
    }
 
    public PathClassLoader(String dexPath, String libraryPath,
            ClassLoader parent) {
        super(dexPath, null, libraryPath, parent);
    }
}

源码就这么点，我也是惊了个呆

不同之处就在于super（）函数的第二个参数

上次说到安装阶段已经生成了一个优化后的dex，所以PathClassLoader当中就是null，DexClassLoader则需要自行生成一个优化文件。

    public BaseDexClassLoader(String dexPath, File optimizedDirectory,
            String libraryPath, ClassLoader parent) {
        super(parent);
        this.pathList = new DexPathList(this, dexPath, libraryPath, optimizedDirectory);
    }

BaseDexClassLoader的构造函数

    public DexPathList(ClassLoader definingContext, String dexPath,
            String libraryPath, File optimizedDirectory) {
        if (definingContext == null) {
            throw new NullPointerException("definingContext == null");
        }

        if (dexPath == null) {
            throw new NullPointerException("dexPath == null");
        }

        if (optimizedDirectory != null) {
            if (!optimizedDirectory.exists())  {
                throw new IllegalArgumentException(
                        "optimizedDirectory doesn't exist: "
                        + optimizedDirectory);
            }

            if (!(optimizedDirectory.canRead()
                            && optimizedDirectory.canWrite())) {
                throw new IllegalArgumentException(
                        "optimizedDirectory not readable/writable: "
                        + optimizedDirectory);
            }
        }

        this.definingContext = definingContext;
        this.dexElements = makeDexElements(splitDexPath(dexPath), optimizedDirectory);
        this.nativeLibraryDirectories = splitLibraryPath(libraryPath);
    }

DexPathList的构造函数

先进行一些验证，然后关键是makeDexElements和splitLibraryPath函数

先看第一个makeDexElements

    /**
     * Makes an array of dex/resource path elements, one per element of
     * the given array.
     */
    private static Element[] makeDexElements(ArrayList<File> files,
            File optimizedDirectory) {
        ArrayList<Element> elements = new ArrayList<Element>();

        /*
         * Open all files and load the (direct or contained) dex files
         * up front.
         */
        for (File file : files) {
            File zip = null;
            DexFile dex = null;
            String name = file.getName();

            if (name.endsWith(DEX_SUFFIX)) {
                // Raw dex file (not inside a zip/jar).
                try {
                    dex = loadDexFile(file, optimizedDirectory);
                } catch (IOException ex) {
                    System.logE("Unable to load dex file: " + file, ex);
                }
            } else if (name.endsWith(APK_SUFFIX) || name.endsWith(JAR_SUFFIX)
                    || name.endsWith(ZIP_SUFFIX)) {
                zip = file;

                try {
                    dex = loadDexFile(file, optimizedDirectory);
                } catch (IOException ignored) {
                    /*
                     * IOException might get thrown "legitimately" by
                     * the DexFile constructor if the zip file turns
                     * out to be resource-only (that is, no
                     * classes.dex file in it). Safe to just ignore
                     * the exception here, and let dex == null.
                     */
                }
            } else if (file.isDirectory()) {
                // We support directories for looking up resources.
                // This is only useful for running libcore tests.
                elements.add(new Element(file, true, null, null));
            } else {
                System.logW("Unknown file type for: " + file);
            }

            if ((zip != null) || (dex != null)) {
                elements.add(new Element(file, false, zip, dex));
            }
        }

        return elements.toArray(new Element[elements.size()]);
    }

注释说的很清楚，产生一个dex或者resource元素的数组

如果文件以.dex、.apk、.jar和.zip结尾，都会调用loadDexFile，区别在于局部变量zip的赋值（具体看代码），如果路径是一个目录，则直接添加到array

前四种情况，都会在调用完loadDexFile后再添加

这里还有个小问题，makeDexElements函数的第一个参数，在调用时为splitDexPath(dexPath)

splitDexPath函数如下

    private static ArrayList<File> splitDexPath(String path) {
        return splitPaths(path, null, false);
    }

splitPaths函数如下

    private static ArrayList<File> splitPaths(String path1, String path2,
            boolean wantDirectories) {
        ArrayList<File> result = new ArrayList<File>();

        splitAndAdd(path1, wantDirectories, result);
        splitAndAdd(path2, wantDirectories, result);
        return result;
    }

最终的处理在splitAndAdd函数中

    private static void splitAndAdd(String searchPath, boolean directoriesOnly,
            ArrayList<File> resultList) {
        if (searchPath == null) {
            return;
        }
        for (String path : searchPath.split(":")) {
            try {
                StructStat sb = Libcore.os.stat(path);
                if (!directoriesOnly || S_ISDIR(sb.st_mode)) {
                    resultList.add(new File(path));
                }
            } catch (ErrnoException ignored) {
            }
        }
    }

回到主线任务，下面看loadDexFile函数

    private static DexFile loadDexFile(File file, File optimizedDirectory)
            throws IOException {
        if (optimizedDirectory == null) {
            return new DexFile(file);
        } else {
            String optimizedPath = optimizedPathFor(file, optimizedDirectory);
            return DexFile.loadDex(file.getPath(), optimizedPath, 0);
        }
    }

还记得吗，由于PathClassLoader的optimizedDirectory为null，所以这里调用的是new DexFile(file)

自然DexClassLoader调用的就是DexFile.loadDex(file.getPath(), optimizedPath, 0);
先看loadDex

    static public DexFile loadDex(String sourcePathName, String outputPathName,
        int flags) throws IOException {

        /*
         * TODO: we may want to cache previously-opened DexFile objects.
         * The cache would be synchronized with close().  This would help
         * us avoid mapping the same DEX more than once when an app
         * decided to open it multiple times.  In practice this may not
         * be a real issue.
         */
        return new DexFile(sourcePathName, outputPathName, flags);
    }

其实也就是DexFile的构造函数

    private DexFile(String sourceName, String outputName, int flags) throws IOException {
        if (outputName != null) {
            try {
                String parent = new File(outputName).getParent();
                if (Libcore.os.getuid() != Libcore.os.stat(parent).st_uid) {
                    throw new IllegalArgumentException("Optimized data directory " + parent
                            + " is not owned by the current user. Shared storage cannot protect"
                            + " your application from code injection attacks.");
                }
            } catch (ErrnoException ignored) {
                // assume we'll fail with a more contextual error later
            }
        }

        mCookie = openDexFile(sourceName, outputName, flags);
        mFileName = sourceName;
        guard.open("close");
        //System.out.println("DEX FILE cookie is " + mCookie);
    }

首先，如果是DexClassLoader，会先检查父目录的所有者。

无论是哪种ClassLoader，最终都会调用openDexFile

这是一个native函数，位于dalvik\vm\native\dalvik_system_DexFile.cpp

const DalvikNativeMethod dvm_dalvik_system_DexFile[] = {
    { "openDexFile",        "(Ljava/lang/String;Ljava/lang/String;I)I",
        Dalvik_dalvik_system_DexFile_openDexFile },
    { "openDexFile",        "([B)I",
        Dalvik_dalvik_system_DexFile_openDexFile_bytearray },
    { "closeDexFile",       "(I)V",
        Dalvik_dalvik_system_DexFile_closeDexFile },
    { "defineClass",        "(Ljava/lang/String;Ljava/lang/ClassLoader;I)Ljava/lang/Class;",
        Dalvik_dalvik_system_DexFile_defineClass },
    { "getClassNameList",   "(I)[Ljava/lang/String;",
        Dalvik_dalvik_system_DexFile_getClassNameList },
    { "isDexOptNeeded",     "(Ljava/lang/String;)Z",
        Dalvik_dalvik_system_DexFile_isDexOptNeeded },
    { NULL, NULL, NULL },
};

根据这个二维字符串数组，找到openDexFile所对应的函数

要注意函数参数，别找错

static void Dalvik_dalvik_system_DexFile_openDexFile(const u4* args,
    JValue* pResult)
{
    StringObject* sourceNameObj = (StringObject*) args[0];
    StringObject* outputNameObj = (StringObject*) args[1];
    DexOrJar* pDexOrJar = NULL;
    JarFile* pJarFile;
    RawDexFile* pRawDexFile;
    char* sourceName;
    char* outputName;

    if (sourceNameObj == NULL) {
        dvmThrowNullPointerException("sourceName == null");
        RETURN_VOID();
    }

    sourceName = dvmCreateCstrFromString(sourceNameObj);
    if (outputNameObj != NULL)
        outputName = dvmCreateCstrFromString(outputNameObj);
    else
        outputName = NULL;

    /*
     * We have to deal with the possibility that somebody might try to
     * open one of our bootstrap class DEX files.  The set of dependencies
     * will be different, and hence the results of optimization might be
     * different, which means we'd actually need to have two versions of
     * the optimized DEX: one that only knows about part of the boot class
     * path, and one that knows about everything in it.  The latter might
     * optimize field/method accesses based on a class that appeared later
     * in the class path.
     *
     * We can't let the user-defined class loader open it and start using
     * the classes, since the optimized form of the code skips some of
     * the method and field resolution that we would ordinarily do, and
     * we'd have the wrong semantics.
     *
     * We have to reject attempts to manually open a DEX fi