go语言创建新进程过程详解 （os.StartProcess源码分析）

Posted 2020-06-17

关于如何使用go语言实现新进程的创建和进程间通信，我在网上找了不少的资料，但是始终未能发现让自己满意的答案，因此我打算自己来分析这部分源代码，然后善加利用，并且分享给大家，期望大家能从中获得启发。

首先我们来看一段代码

proc, _ := os.StartProcess(name, args, attr)

if err != nil {

fmt.Println(err)

}

_, err = proc.Wait()

if err != nil {

fmt.Println(err)

}

我们来看看这个os.StartProcess里面到底做了什么东西？ 而 proc.Wait()又做了什么？跟我一起深入进去吧。

// StartProcess starts a new process with the program, arguments and attributes

// specified by name, argv and attr.

//

// StartProcess is a low-level interface. The os/exec package provides

// higher-level interfaces.

//

// If there is an error, it will be of type *PathError.

func StartProcess(name string, argv []string, attr *ProcAttr) (*Process, error) {

return startProcess(name, argv, attr)

}

注释是说，这个函数依照提供三个参数来实现开启新进程的操作，它是一个低级接口，而os/exec包装提供高级接口。如果这里出现报错，应该会是一个指针型路径错误。

下一步我们探究startProcess是什么？

func startProcess(name string, argv []string, attr *ProcAttr) (p *Process, err error) {

sysattr := &syscall.ProcAttr{

Dir: attr.Dir,

Env: attr.Env,

Sys: attr.Sys,

}

for _, f := range attr.Files {

sysattr.Files = append(sysattr.Files, f.Fd())

}

pid, h, e := syscall.StartProcess(name, argv, sysattr)

if e != nil {

return nil, &PathError{"fork/exec", name, e}

}

return newProcess(pid, h), nil

}

首先我们看到sysattr被赋予一个 &syscall.ProcAttr指针，这个syscall里面的ProcAttr是什么结构呢，要先理解它，这样有助于理解我们利用它来启动后面的syscall

// ProcAttr holds the attributes that will be applied to a new process

// started by StartProcess.

type ProcAttr struct {

// If Dir is non-empty, the child changes into the directory before

// creating the process.

Dir string

// If Env is non-nil, it gives the environment variables for the

// new process in the form returned by Environ.

// If it is nil, the result of Environ will be used.

Env []string

// Files specifies the open files inherited by the new process.  The

// first three entries correspond to standard input, standard output, and

// standard error.  An implementation may support additional entries,

// depending on the underlying operating system.  A nil entry corresponds

// to that file being closed when the process starts.

Files []*File

// Operating system-specific process creation attributes.

// Note that setting this field means that your program

// may not execute properly or even compile on some

// operating systems.

Sys *syscall.SysProcAttr

}

第一句简单明了，说明了ProcAttr结构中包含了我们启动进程过程中使用的多项属性值，

1）Dir是目录的意思，相当于新进程的工作目录，如果配置了就会跳转目录。

2）Env是指新的进程的环境变量列表。

3）Files前三项对应标准输入，标准输出和标准错误输出。每个实现可以支持其他条目，如果传入的条目是nil，该进程启动时，file就是关闭的。

4）最后一个*syscall.SysProcAttr就是系统属性，不过作者也提醒道有些参数在跨平台过程中有可能不起作用。

下面我们看下*syscall.SysProcAttr结构。

type SysProcAttr struct {

Chroot     string      // Chroot.

Credential *Credential // Credential.

Ptrace     bool        // Enable tracing.

Setsid     bool        // Create session.

Setpgid    bool        // Set process group ID to new pid (SYSV setpgrp)

Setctty    bool        // Set controlling terminal to fd 0

Noctty     bool        // Detach fd 0 from controlling terminal

}

// Credential holds user and group identities to be assumed

// by a child process started by StartProcess.

type Credential struct {

Uid    uint32   // User ID.

Gid    uint32   // Group ID.

Groups []uint32 // Supplementary group IDs.

}

可以看到这里面所涉及到的属性。（部分属性跨平台不起作用）

1）Chroot      

2) Credential包括uid\gid\groups设定

3）一些bool属性，参与设定新进程的使用过程。

Ptrace     是否允许tracing

Setsid     是否开启sid

Setpgid    是否设定组id给新进程

Setctty    是否可以使用终端访问

Noctty     将终端和fd0 进行分离。

OK，现在我们了解了这么多之后，还是谁去看看前面的代码吧。如下：

func startProcess(name string, argv []string, attr *ProcAttr) (p *Process, err error) {

sysattr := &syscall.ProcAttr{

Dir: attr.Dir,

Env: attr.Env,

Sys: attr.Sys,

}

for _, f := range attr.Files {

sysattr.Files = append(sysattr.Files, f.Fd())

}

pid, h, e := syscall.StartProcess(name, argv, sysattr)

if e != nil {

return nil, &PathError{"fork/exec", name, e}

}

return newProcess(pid, h), nil

}

继续看startProcess

sysattr := &syscall.ProcAttr{

Dir: attr.Dir,

Env: attr.Env,

Sys: attr.Sys,

}

Dir工作目录，Env环境变量、Sys 内容被赋予了sysattr 。

for _, f := range attr.Files {

sysattr.Files = append(sysattr.Files, f.Fd())

}

文件Files属性被安排加入到sysattr中，这样我们就把attr *ProcAttr参数的整体内容都赋予了sysattr ，下面看如何利用这个sysattr

pid, h, e := syscall.StartProcess(name, argv, sysattr) sysattr作为第三项参数传入了新的 

syscall.StartProcess(name, argv, sysattr)

注意：这里我们注意到一个问题，看看我们期初的代码

proc, _ := os.StartProcess(name, args, attr)

if err != nil {

fmt.Println(err)

}

这一行代码和我们的期初的代码是多么相像啊，于是我们明白调用os的StartProcess就是调用syscall.StartProcess，因此我们明白，syscall.StartProcess属于底层调用。os.StartProcess是上层调用。os.StartProces只是在syscall.StartProcess外面包装了一层而已，因此，我们明白，当我们想新创建一个进程的时候，只要参数都已经输入完毕，我们既可以使用os.StartProcess来实现，也可以使用syscall.StartProcess来实现。只不过需要注意的是，两者返回的对象不相同。

怎么个不相同呢？

我们看到了os.StartProcess 返回的是return newProcess(pid, h), nil， 而

syscall.StartProcess返回的是pid, h, e 

也就是说os.StartProcess 返回的是syscall.StartProcess返回值对pid和h的包装的结果。

// Process stores the information about a process created by StartProcess.

type Process struct {

Pid    int

handle uintptr

isdone uint32 // process has been successfully waited on, non zero if true

}

func newProcess(pid int, handle uintptr) *Process {

p := &Process{Pid: pid, handle: handle}

runtime.SetFinalizer(p, (*Process).Release)

return p

}

通过观察这个包装的过程我们明白，之所以返回这个结果的目的是为了处理一些程序在进行时过程中的问题。下面我们就得了解下程序运行时的概念。

runtime.SetFinalizer(p, (*Process).Release)这一行在做什么呢？

这部分就是难点了，如果理解了这部分就会了解程序为什么包装了这一层，它的目的何在。

下面则是一大段英文。我门来试着理解一下。

// SetFinalizer sets the finalizer associated with x to f.

// When the garbage collector finds an unreachable block

// with an associated finalizer, it clears the association and runs

// f(x) in a separate goroutine.  This makes x reachable again, but

// now without an associated finalizer.  Assuming that SetFinalizer

// is not called again, the next time the garbage collector sees

// that x is unreachable, it will free x.

//

// SetFinalizer(x, nil) clears any finalizer associated with x.

//

// The argument x must be a pointer to an object allocated by

// calling new or by taking the address of a composite literal.

// The argument f must be a function that takes a single argument

// to which x‘s type can be assigned, and can have arbitrary ignored return

// values. If either of these is not true, SetFinalizer aborts the

// program.

//

// Finalizers are run in dependency order: if A points at B, both have

// finalizers, and they are otherwise unreachable, only the finalizer

// for A runs; once A is freed, the finalizer for B can run.

// If a cyclic structure includes a block with a finalizer, that

// cycle is not guaranteed to be garbage collected and the finalizer

// is not guaranteed to run, because there is no ordering that

// respects the dependencies.

//

// The finalizer for x is scheduled to run at some arbitrary time after

// x becomes unreachable.

// There is no guarantee that finalizers will run before a program exits,

// so typically they are useful only for releasing non-memory resources

// associated with an object during a long-running program.

// For example, an os.File object could use a finalizer to close the

// associated operating system file descriptor when a program discards

// an os.File without calling Close, but it would be a mistake

// to depend on a finalizer to flush an in-memory I/O buffer such as a

// bufio.Writer, because the buffer would not be flushed at program exit.

//

// It is not guaranteed that a finalizer will run if the size of *x is

// zero bytes.

//

// It is not guaranteed that a finalizer will run for objects allocated

// in initializers for package-level variables. Such objects may be

// linker-allocated, not heap-allocated.

//

// A single goroutine runs all finalizers for a program, sequentially.

// If a finalizer must run for a long time, it should do so by starting

// a new goroutine.

我这里不是想照抄英文，只是为了文章的完整性，我们来看看它说了什么吧。首先我们理解一下，这部分代码是在垃圾回收部分的，因此我理解我们已经接近了更加底层，而且我有预感，这个runtime.SetFinalizer(p, (*Process).Release)应该是为这个新启动的进程设定了一个垃圾回收机制，也就是说我们如何回收已经完结的进程或者运行过程中的进程所产生的一系列垃圾数据。

本文出自 “技术老男人” 博客，请务必保留此出处http://allragedbody.blog.51cto.com/1352220/1747146