从源码分析 Android dexClassLoader 加载机制原理

 一直想了解  Android 动态加载类的原理，今天抽空分析了下DexClassLoader 源码

  DexClassLoader 是加载包含classes.dex文件的jar文件或者apk文件；   通过构造函数发现需要一个应用私有的，可写的目录去缓存优化的classes。可以用使用File dexoutputDir = context.getDir(“dex”,0);创建一个这样的目录，不要使用外部缓存，以保护你的应用被代码注入。

其源码如下：

public classDexClassLoaderextendsBaseDexClassLoader {

37    /**38     * Creates a {@code DexClassLoader} that finds interpreted and native39     * code.  Interpreted classes are found in a set of DEX files contained40     * in Jar or APK files.41     *42     * <p>The path lists are separated using the character specified by the43     * {@code path.separator} system property, which defaults to {@code :}.44     *45     * @param dexPath the list of jar/apk files containing classes and46     *     resources, delimited by {@code File.pathSeparator}, which47     *     defaults to {@code ":"} on Android48     * @param optimizedDirectory directory where optimized dex files49     *     should be written; must not be {@code null}50     * @param libraryPath the list of directories containing native51     *     libraries, delimited by {@code File.pathSeparator}; may be52     *     {@code null}53     * @param parent the parent class loader54     */55    public DexClassLoader(String dexPath, String optimizedDirectory,56            String libraryPath, ClassLoader parent) {57        super(dexPath, new File(optimizedDirectory), libraryPath, parent);58    }59}60

再解释下几个构造函数参数的意义：

dexpath为jar或apk文件目录。

optimizedDirectory为优化dex缓存目录。

libraryPath包含native lib的目录路径。

parent父类加载器。

然后执行的是父类的构造函数：

super(dexPath, new File(optimizedDirectory), libraryPath, parent);

BaseDexClassLoader 的构造函数如下：

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

第一句调用的还是父类的构造函数，也就是ClassLoader的构造函数：

protected ClassLoader(ClassLoader parentLoader) {        this(parentLoader, false);    }    /*     * constructor for the BootClassLoader which needs parent to be null.     */    ClassLoader(ClassLoader parentLoader, boolean nullAllowed) {       if (parentLoader == null && !nullAllowed) {            throw new NullPointerException(“parentLoader == null && !nullAllowed”);      }      parent = parentLoader;}

该构造函数把传进来的父类加载器赋给了私有变量parent。

再来看第二句：

this.pathList = new DexPathList(this, dexPath, libraryPath, optimizedDirectory);

pathList为该类的私有成员变量，类型为DexPathList，进入到DexPathList函数：

/**78     * Constructs an instance.79     *80     * @param definingContext the context in which any as-yet unresolved81     * classes should be defined82     * @param dexPath list of dex/resource path elements, separated by83     * {@code File.pathSeparator}84     * @param libraryPath list of native library directory path elements,85     * separated by {@code File.pathSeparator}86     * @param optimizedDirectory directory where optimized {@code .dex} files87     * should be found and written to, or {@code null} to use the default88     * system directory for same89     */90    public DexPathList(ClassLoader definingContext, String dexPath,91            String libraryPath, File optimizedDirectory) {9293        if (definingContext == null) {94            throw new NullPointerException("definingContext == null");95        }9697        if (dexPath == null) {98            throw new NullPointerException("dexPath == null");99        }100101        if (optimizedDirectory != null) {102            if (!optimizedDirectory.exists())  {103                throw new IllegalArgumentException(104                        "optimizedDirectory doesn't exist: "105                        + optimizedDirectory);106            }107108            if (!(optimizedDirectory.canRead()109                            && optimizedDirectory.canWrite())) {110                throw new IllegalArgumentException(111                        "optimizedDirectory not readable/writable: "112                        + optimizedDirectory);113            }114        }115116        this.definingContext = definingContext;117118        ArrayList<IOException> suppressedExceptions = new ArrayList<IOException>();

119        // save dexPath for BaseDexClassLoader120        this.dexElements = makePathElements(splitDexPath(dexPath), optimizedDirectory,1        suppressedExceptions);122123        // Native libraries may exist in both the system and124        // application library paths, and we use this search order:125        //126        //   1. This class loader's library path for application libraries (libraryPath):127        //   1.1. Native library directories128        //   1.2. Path to libraries in apk-files129        //   2. The VM's library path from the system property for system libraries130        //      also known as java.library.path131        //132        // This order was reversed prior to Gingerbread; see http://b/2933456.133        this.nativeLibraryDirectories = splitPaths(libraryPath, false);134        this.systemNativeLibraryDirectories =135                splitPaths(System.getProperty("java.library.path"), true);136        List<File> allNativeLibraryDirectories = new ArrayList<>(nativeLibraryDirectories);137        allNativeLibraryDirectories.addAll(systemNativeLibraryDirectories);138139        this.nativeLibraryPathElements = makePathElements(allNativeLibraryDirectories, null,140                                                          suppressedExceptions);141142        if (suppressedExceptions.size() > 0) {143            this.dexElementsSuppressedExceptions =144                suppressedExceptions.toArray(new IOException[suppressedExceptions.size()]);145        } else {146            dexElementsSuppressedExceptions = null;147        }148    }149

前面是一些对于传入参数的验证，然后调用了makeDexElements。

private static Element[] makeDexElements(ArrayList<File> files, File optimizedDirectory,                                             ArrayList<IOException> suppressedExceptions) {ArrayList<Element> elements = new ArrayList<Element>();        for (File file : files) {            File zip = null;            DexFile dex = null;            String name = file.getName();            if (name.endsWith(DEX_SUFFIX)) {               //dex文件处理                // 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)) {   //apk，jar，zip文件处理                zip = file;                try {                    dex = loadDexFile(file, optimizedDirectory);                } catch (IOException suppressed) {                    suppressedExceptions.add(suppressed);                }            } else if (file.isDirectory()) {                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()]);    }}

不管是dex文件，还是apk文件最终加载的都是loadDexFile，跟进这个函数：

如果optimizedDirectory为null就会调用openDexFile(fileName, null, 0);加载文件。否则调用DexFile.loadDex(file.getPath(), optimizedPath, 0);而这个函数也只是直接调用new DexFile(sourcePathName, outputPathName, flags);里面调用的也是openDexFile(sourceName, outputName, flags);所以最后都是调用openDexFile，跟进这个函数：

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);        }}

private static int openDexFile(String sourceName, String outputName,        int flags) throws IOException {        return openDexFileNative(new File(sourceName).getCanonicalPath(),                                 (outputName == null) ? null : new File(outputName).getCanonicalPath(),                                 flags);}

而这个函数调用的是so的openDexFileNative这个函数。打开成功则返回一个cookie。

接下来就是分析native函数的实现部分了。

———-openDexFileNative———-

static void Dalvik_dalvik_system_DexFile_openDexFileNative(const u4* args,JValue* pResult){    ……………if (hasDexExtension(sourceName)            && dvmRawDexFileOpen(sourceName, outputName, &pRawDexFile, false) == 0) {        ALOGV(“Opening DEX file ‘%s’ (DEX)”, sourceName);        pDexOrJar = (DexOrJar*) malloc(sizeof(DexOrJar));        pDexOrJar->isDex = true;        pDexOrJar->pRawDexFile = pRawDexFile;        pDexOrJar->pDexMemory = NULL;    } else if (dvmJarFileOpen(sourceName, outputName, &pJarFile, false) == 0) {        ALOGV(“Opening DEX file ‘%s’ (Jar)”, sourceName);        pDexOrJar = (DexOrJar*) malloc(sizeof(DexOrJar));        pDexOrJar->isDex = false;        pDexOrJar->pJarFile = pJarFile;        pDexOrJar->pDexMemory = NULL;    } else {        ALOGV(“Unable to open DEX file ‘%s’”, sourceName);        dvmThrowIOException(“unable to open DEX file”);    }    ……………}

这里会根据是否为dex文件或者包含classes.dex文件的jar，分别调用函数dvmRawDexFileOpen和dvmJarFileOpen来处理，最终返回一个DexOrJar的结构。

首先来看dvmRawDexFileOpen函数的处理：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104

int dvmRawDexFileOpen(const char* fileName, const char* odexOutputName,
    RawDexFile** ppRawDexFile, bool isBootstrap)
{
    .................
    dexFd = open(fileName, O_RDONLY);
    if (dexFd < 0) goto bail;

    /* If we fork/exec into dexopt, don't let it inherit the open fd. */
    dvmSetCloseOnExec(dexFd);

    //校验前8个字节的magic是否正确，然后把校验和保存到adler32
    if (verifyMagicAndGetAdler32(dexFd, &adler32) < 0) {
        ALOGE("Error with header for %s", fileName);
        goto bail;
    }
    //得到文件修改时间以及文件大小
   if (getModTimeAndSize(dexFd, &modTime, &fileSize) < 0) {
        ALOGE("Error with stat for %s", fileName);
        goto bail;
    }
    .................
    //调用函数dexOptCreateEmptyHeader,构造了一个DexOptHeader结构体，写入fd并返回
    optFd = dvmOpenCachedDexFile(fileName, cachedName, modTime,
        adler32, isBootstrap, &newFile, /*createIfMissing=*/true);

    if (optFd < 0) {
        ALOGI("Unable to open or create cache for %s (%s)",
                fileName, cachedName);
        goto bail;
    }
    locked = true;

       //如果成功生了opt头
    if (newFile) {
        u8 startWhen, copyWhen, endWhen;
        bool result;
       off_t dexOffset;

        dexOffset = lseek(optFd, 0, SEEK_CUR);
        result = (dexOffset > 0);

        if (result) {
            startWhen = dvmGetRelativeTimeUsec();
            // 将dex文件中的内容写入文件的当前位置，也就是从dexOffset的偏移处开始写
            result = copyFileToFile(optFd, dexFd, fileSize) == 0;
            copyWhen = dvmGetRelativeTimeUsec();
        }

        if (result) {
            //对dex文件进行优化
            result = dvmOptimizeDexFile(optFd, dexOffset, fileSize,
                fileName, modTime, adler32, isBootstrap);
        }

        if (!result) {
            ALOGE("Unable to extract+optimize DEX from '%s'", fileName);
            goto bail;
        }

        endWhen = dvmGetRelativeTimeUsec();
        ALOGD("DEX prep '%s': copy in %dms, rewrite %dms",
            fileName,
            (int) (copyWhen - startWhen) / 1000,
            (int) (endWhen - copyWhen) / 1000);
    }

     //dvmDexFileOpenFromFd这个函数最主要在这里干了两件事情
     // 1.将优化后得dex文件（也就是odex文件）通过mmap映射到内存中，并通过mprotect修改它的映射内存为只读权限
     // 2.将映射为只读的这块dex数据中的内容全部提取到DexFile这个数据结构中去
    if (dvmDexFileOpenFromFd(optFd, &pDvmDex) != 0) {
        ALOGI("Unable to map cached %s", fileName);
        goto bail;
    }

    if (locked) {
        /* unlock the fd */
       if (!dvmUnlockCachedDexFile(optFd)) {
            /* uh oh -- this process needs to exit or we'll wedge the system */
            ALOGE("Unable to unlock DEX file");
            goto bail;
        }
        locked = false;
    }

    ALOGV("Successfully opened '%s'", fileName);
    //填充结构体 RawDexFile
    *ppRawDexFile = (RawDexFile*) calloc(1, sizeof(RawDexFile));
    (*ppRawDexFile)->cacheFileName = cachedName;
   (*ppRawDexFile)->pDvmDex = pDvmDex;
    cachedName = NULL;      // don't free it below
    result = 0;

bail:
    free(cachedName);
    if (dexFd >= 0) {
        close(dexFd);
    }
    if (optFd >= 0) {
        if (locked)
            (void) dvmUnlockCachedDexFile(optFd);
        close(optFd);
    }
    return result;
}

最后成功的话，填充RawDexFile。

dvmJarFileOpen的代码处理也是差不多的。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77

int dvmJarFileOpen(const char* fileName, const char* odexOutputName,
    JarFile** ppJarFile, bool isBootstrap)
{
    ...
    ...
    ...
    //调用函数dexZipOpenArchive来打开zip文件，并缓存到系统内存里
    if (dexZipOpenArchive(fileName, &archive) != 0)
        goto bail;
    archiveOpen = true;
    ...
    //这行代码设置当执行完成后，关闭这个文件句柄
    dvmSetCloseOnExec(dexZipGetArchiveFd(&archive));
    ...
    //优先处理已经优化了的Dex文件
    fd = openAlternateSuffix(fileName, "odex", O_RDONLY, &cachedName);
    ...
    //从压缩包里找到Dex文件，然后打开这个文件
    entry = dexZipFindEntry(&archive, kDexInJarName);
    ...
    //把未经过优化的Dex文件进行优化处理，并输出到指定的文件
    if (odexOutputName == NULL) {
                cachedName = dexOptGenerateCacheFileName(fileName,
                                kDexInJarName);
    }
    ...
    //创建缓存的优化文件
    fd = dvmOpenCachedDexFile(fileName, cachedName,
                    dexGetZipEntryModTime(&archive, entry),
                    dexGetZipEntryCrc32(&archive, entry),
                    isBootstrap, &newFile, /*createIfMissing=*/true);
    ...
    //调用函数dexZipExtractEntryToFile从压缩包里解压文件出来
    if (result) {
                    startWhen = dvmGetRelativeTimeUsec();
                    result = dexZipExtractEntryToFile(&archive, entry, fd) == 0;
                    extractWhen = dvmGetRelativeTimeUsec();
                 }
    ...
    //调用函数dvmOptimizeDexFile对Ｄex文件进行优化处理
    if (result) {
                    result = dvmOptimizeDexFile(fd, dexOffset,
                                dexGetZipEntryUncompLen(&archive, entry),
                                fileName,
                                dexGetZipEntryModTime(&archive, entry),
                                dexGetZipEntryCrc32(&archive, entry),
                                isBootstrap);
                }
    ...
    //调用函数dvmDexFileOpenFromFd来缓存dex文件
    //并分析文件的内容。比如标记是否优化的文件，通过签名检查Dex文件是否合法
    if (dvmDexFileOpenFromFd(fd, &pDvmDex) != 0) {
        ALOGI("Unable to map %s in %s", kDexInJarName, fileName);
        goto bail;
    }
    ...
    //保存文件到缓存里，标记这个文件句柄已经保存到缓存
    if (locked) {
        /* unlock the fd */
        if (!dvmUnlockCachedDexFile(fd)) {
            /* uh oh -- this process needs to exit or we'll wedge the system */
            ALOGE("Unable to unlock DEX file");
            goto bail;
        }
        locked = false;
    }
    ...
     //设置一些相关信息返回前面的函数处理。
    *ppJarFile = (JarFile*) calloc(1, sizeof(JarFile));
    (*ppJarFile)->archive = archive;
    (*ppJarFile)->cacheFileName = cachedName;
    (*ppJarFile)->pDvmDex = pDvmDex;
    cachedName = NULL;      // don't free it below
    result = 0;
    ...

}

最后成功的话，填充JarFile。