goroutine退出过程 #

// The main goroutine.
func main() {
	g := getg()
	//...
 
    //main包 init函数，递归的调用import包中定义的init函数
    fn := main_init // make an indirect call, as the linker doesn't know the address of the main package when laying down the runtime
    fn()
 
    //...
     
    //调用main.main函数(用户定义的main函数):进行间接调用是因为链接器在放置运行时不知道主包的地址
    fn = main_main // make an indirect call, as the linker doesn't know the address of the main package when laying down the runtime
    fn()
     
    //...
 
    //系统API:exit函数,退出进程
    exit(0)
     
	for { 
		var x *int32
		*x = 0 // 无效指针代码，会导致程序退出
	}
}

// main.go
package main

import "time"

// the function's body is empty
func add(x, y int64) int64

func main() {
    go add(2, 3)

    time.Sleep(time.Minute)
}

//  add_amd.s
TEXT ·add(SB),$0-24
	MOVQ x+0(FP), BX
	MOVQ y+8(FP), BP
	ADDQ BP, BX
	MOVQ BX, ret+16(FP)
	RET

// The top-most function running on a goroutine
// returns to goexit+PCQuantum.
TEXT runtime·goexit(SB),NOSPLIT,$0-0
	BYTE	$0x90	// NOP
	CALL	runtime·goexit1(SB)	// does not return
	// traceback from goexit1 must hit code range of goexit
	BYTE	$0x90	// NOP


// Finishes execution of the current goroutine.
func goexit1() {
	if raceenabled {   //忽略
		racegoend()
	}
	if trace.enabled {   //忽略
		traceGoEnd()
	}
	mcall(goexit0)
}

// func mcall(fn func(*g))
// Switch to m->g0's stack, call fn(g).
// Fn must never return. It should gogo(&g->sched)
// to keep running g.
TEXT runtime·mcall(SB), NOSPLIT, $0-8
	MOVQ	fn+0(FP), DI //参数

	get_tls(CX)
	MOVQ	g(CX), AX // save state in gN->sched
	MOVQ	0(SP), BX // caller's PC   -->看下方的图
	MOVQ	BX, (g_sched+gobuf_pc)(AX) //保存caller's pc到正在运行的gN.sched.pc
	LEAQ	fn+0(FP), BX // caller's SP
	MOVQ	BX, (g_sched+gobuf_sp)(AX) //保存caller's sp到正在运行的gN.sched.sp
	MOVQ	AX, (g_sched+gobuf_g)(AX) //保存gN到正在运行的gN.sched.g
	MOVQ	BP, (g_sched+gobuf_bp)(AX) //保存bp到正在运行的gN.sched.bp

	// switch to m->g0 & its stack, call fn
	MOVQ	g(CX), BX // bx=gN
	MOVQ	g_m(BX), BX // bx=gN.m
	MOVQ	m_g0(BX), SI // si=gN.m.g0
	CMPQ	SI, AX // if g == m->g0 call badmcall; 这个gN不能等于g0, g0应该是用户调度用的.
	JNE	3(PC)
	MOVQ	$runtime·badmcall(SB), AX
	JMP	AX
	MOVQ	SI, g(CX)	// g = m->g0; 就是把m.tls[0](TLS)的值从gN的地址换为g0的地址,这样线程通过fs寄存器能找到g0继而找到m   -----------here
	MOVQ	(g_sched+gobuf_sp)(SI), SP // sp = m->g0->sched.sp,把g0的寄存器SP恢复到真实的SP   ---------------here
	PUSHQ	AX //gN压栈,作为后面call的参数
	MOVQ	DI, DX //dx = di(fn函数结构体)
	MOVQ	0(DI), DI //所以这里是取真正的fn
	CALL	DI //开始调用fn
	POPQ	AX
	MOVQ	$runtime·badmcall2(SB), AX
	JMP	AX
	RET

	MOVQ DI, DX        # dx = di(fn函数)
	MOVQ 0(DI), DI     # 所以这里是取真正的fn

type funcval struct {
    fn uintptr
    // variable-size, fn-specific data here
}

// goexit continuation on g0.
func goexit0(gp *g) {
	_g_ := getg() // g0

	casgstatus(gp, _Grunning, _Gdead) // 修改gN的状态
	if isSystemGoroutine(gp, false) {
		atomic.Xadd(&sched.ngsys, -1)
	}
	gp.m = nil
	locked := gp.lockedm != 0
	gp.lockedm = 0
	_g_.m.lockedg = 0
	gp.paniconfault = false
	gp._defer = nil // should be true already but just in case.
	gp._panic = nil // non-nil for Goexit during panic. points at stack-allocated data.
	gp.writebuf = nil
	gp.waitreason = 0
	gp.param = nil
	gp.labels = nil
	gp.timer = nil

    //...

	// Note that gp's stack scan is now "valid" because it has no
	// stack.
	gp.gcscanvalid = true
	dropg() //dropg函数解除g和m之间的关系，其实就是设置g->m = nil, m->currg = nil.

    //...

    gfput(_g_.m.p.ptr(), gp) //放在gfree列表中,如果本地列表太长，则将一个批次转移到全局列表中.
    
    //...

	schedule()
}