Golang 源码分析

2017-05-10

Golang 函数调用规约

汇编相关的寄存器

函数内存布局

几个重要的汇编

// mcall switches from the g to the g0 stack and invokes fn(g),
// where g is the goroutine that made the call.
// mcall saves g's current PC/SP in g->sched so that it can be restored later.
// It is up to fn to arrange for that later execution, typically by recording
// g in a data structure, causing something to call ready(g) later.
// mcall returns to the original goroutine g later, when g has been rescheduled.
// fn must not return at all; typically it ends by calling schedule, to let the m
// run other goroutines.
//
// mcall can only be called from g stacks (not g0, not gsignal).
//
// This must NOT be go:noescape: if fn is a stack-allocated closure,
// fn puts g on a run queue, and g executes before fn returns, the
// closure will be invalidated while it is still executing.
func mcall(fn func(*g))

mcall 会从 g 栈切换到 g0 调用函数 fn(g) g 就是当前的 goroutine。 mcall 会保存 g 的当前 PC/SP 到 g->sched，所以 g 可以在随后重新被加载。 fn 必需不能有返回值，一般情况下这个由调用 schedule 结束，这样可以让 m 运行其他 goroutines。 也就是说 mcall 使得当前 goroutines 放弃当前的运行。

mcall 只能在 g 栈被调用，不能在 g0，gsingal 栈调用

// 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 //保存 fn 函数指针
	
	get_tls(CX)
	MOVQ	g(CX), AX	// save state in g->sched
	MOVQ	0(SP), BX	// caller's PC
	MOVQ	BX, (g_sched+gobuf_pc)(AX)
	LEAQ	fn+0(FP), BX	// caller's SP
	MOVQ	BX, (g_sched+gobuf_sp)(AX)
	MOVQ	AX, (g_sched+gobuf_g)(AX)
	MOVQ	BP, (g_sched+gobuf_bp)(AX)

	// switch to m->g0 & its stack, call fn
	MOVQ	g(CX), BX
	MOVQ	g_m(BX), BX
	MOVQ	m_g0(BX), SI
	CMPQ	SI, AX	// if g == m->g0 call badmcall
	JNE	3(PC)
	MOVQ	$runtime·badmcall(SB), AX
	JMP	AX
	MOVQ	SI, g(CX)	// g = m->g0
	MOVQ	(g_sched+gobuf_sp)(SI), SP	// sp = m->g0->sched.sp
	PUSHQ	AX      //切换栈指针后，将 g 参数压如栈
	MOVQ	DI, DX
	MOVQ	0(DI), DI
	CALL	DI      //调用函数 fn
	POPQ	AX
	MOVQ	$runtime·badmcall2(SB), AX
	JMP	AX
	RET

// systemstack runs fn on a system stack.
// If systemstack is called from the per-OS-thread (g0) stack, or
// if systemstack is called from the signal handling (gsignal) stack,
// systemstack calls fn directly and returns.
// Otherwise, systemstack is being called from the limited stack
// of an ordinary goroutine. In this case, systemstack switches
// to the per-OS-thread stack, calls fn, and switches back.
// It is common to use a func literal as the argument, in order
// to share inputs and outputs with the code around the call
// to system stack:
//
//	... set up y ...
//	systemstack(func() {
//		x = bigcall(y)
//	})
//	... use x ...
//
//go:noescape
func systemstack(fn func())

systemstack 运行 fn 在系统栈上，如果 systemstack 是在 g0 栈或者 gsignal 栈上，直接调用 fn 函数然后返回。 否则，systemstack 将切换到 g0 栈上调用 fn 然后切换回来。

// func systemstack(fn func())
TEXT runtime·systemstack(SB), NOSPLIT, $0-8
	MOVQ	fn+0(FP), DI	// DI = fn
	get_tls(CX)
	MOVQ	g(CX), AX	// AX = g
	MOVQ	g_m(AX), BX	// BX = m

	MOVQ	m_gsignal(BX), DX	// DX = gsignal
	CMPQ	AX, DX
	JEQ	noswitch

	MOVQ	m_g0(BX), DX	// DX = g0
	CMPQ	AX, DX
	JEQ	noswitch

	MOVQ	m_curg(BX), R8
	CMPQ	AX, R8
	JEQ	switch
	
	// Bad: g is not gsignal, not g0, not curg. What is it?
	MOVQ	$runtime·badsystemstack(SB), AX
	CALL	AX

switch:
	// save our state in g->sched. Pretend to
	// be systemstack_switch if the G stack is scanned.
	MOVQ	$runtime·systemstack_switch(SB), SI
	MOVQ	SI, (g_sched+gobuf_pc)(AX)
	MOVQ	SP, (g_sched+gobuf_sp)(AX)
	MOVQ	AX, (g_sched+gobuf_g)(AX)
	MOVQ	BP, (g_sched+gobuf_bp)(AX)

	// switch to g0
	MOVQ	DX, g(CX) //保存 g0 栈到 tls 中
	MOVQ	(g_sched+gobuf_sp)(DX), BX
	// make it look like mstart called systemstack on g0, to stop traceback
	SUBQ	$8, BX
	MOVQ	$runtime·mstart(SB), DX
	MOVQ	DX, 0(BX)
	MOVQ	BX, SP

	// call target function
	MOVQ	DI, DX
	MOVQ	0(DI), DI
	CALL	DI

	// switch back to g
	get_tls(CX)
	MOVQ	g(CX), AX
	MOVQ	g_m(AX), BX
	MOVQ	m_curg(BX), AX
	MOVQ	AX, g(CX)
	MOVQ	(g_sched+gobuf_sp)(AX), SP
	MOVQ	$0, (g_sched+gobuf_sp)(AX)
	RET

noswitch:
	// already on m stack, just call directly
	MOVQ	DI, DX
	MOVQ	0(DI), DI
	CALL	DI
	RET

// goexit is the return stub at the top of every goroutine call stack.
// Each goroutine stack is constructed as if goexit called the
// goroutine's entry point function, so that when the entry point
// function returns, it will return to goexit, which will call goexit1
// to perform the actual exit.
//
// This function must never be called directly. Call goexit1 instead.
// gentraceback assumes that goexit terminates the stack. A direct
// call on the stack will cause gentraceback to stop walking the stack
// prematurely and if there are leftover stack barriers it may panic.
func goexit(neverCallThisFunction)

goexit 返回 goroutines 的栈的顶部。每一个 goroutines 栈构造时，会压入栈顶。 当 goroutines 返回时 goexit 实际上会执行。

goexit 不能直接调用。必需使用 goexit1 代替。

// 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

// getcallerpc returns the program counter (PC) of its caller's caller.
// getcallersp returns the stack pointer (SP) of its caller's caller.
// For both, the argp must be a pointer to the caller's first function argument.
// The implementation may or may not use argp, depending on
// the architecture.
//
// For example:
//
//	func f(arg1, arg2, arg3 int) {
//		pc := getcallerpc(unsafe.Pointer(&arg1))
//		sp := getcallersp(unsafe.Pointer(&arg1))
//	}
//
// These two lines find the PC and SP immediately following
// the call to f (where f will return).
//
// The call to getcallerpc and getcallersp must be done in the
// frame being asked about. It would not be correct for f to pass &arg1
// to another function g and let g call getcallerpc/getcallersp.
// The call inside g might return information about g's caller or
// information about f's caller or complete garbage.
//
// The result of getcallersp is correct at the time of the return,
// but it may be invalidated by any subsequent call to a function
// that might relocate the stack in order to grow or shrink it.
// A general rule is that the result of getcallersp should be used
// immediately and can only be passed to nosplit functions.

//go:noescape
func getcallerpc(argp unsafe.Pointer) uintptr

//go:noescape
func getcallersp(argp unsafe.Pointer) uintptr

func gogo(buf *gobuf)

// void gogo(Gobuf*)
// restore state from Gobuf; longjmp
TEXT runtime·gogo(SB), NOSPLIT, $0-8
	MOVQ	buf+0(FP), BX		// gobuf
	MOVQ	gobuf_g(BX), DX
	MOVQ	0(DX), CX		// make sure g != nil
	get_tls(CX)
	MOVQ	DX, g(CX)
	MOVQ	gobuf_sp(BX), SP	// restore SP
	MOVQ	gobuf_ret(BX), AX
	MOVQ	gobuf_ctxt(BX), DX
	MOVQ	gobuf_bp(BX), BP
	MOVQ	$0, gobuf_sp(BX)	// clear to help garbage collector
	MOVQ	$0, gobuf_ret(BX)
	MOVQ	$0, gobuf_ctxt(BX)
	MOVQ	$0, gobuf_bp(BX)
	MOVQ	gobuf_pc(BX), BX
	JMP	BX

gogo 函数在 m 上切换栈到 g 上，执行函数。 注意这里最后使用到 JMP , 那么函数在最后到 return 后 pc 是啥？ 我们可以查看下构造 goroutines 是栈上到布局

// Create a new g running fn with siz bytes of arguments.
// Put it on the queue of g's waiting to run.
// The compiler turns a go statement into a call to this.
// Cannot split the stack because it assumes that the arguments
// are available sequentially after &fn; they would not be
// copied if a stack split occurred.
//go:nosplit
func newproc(siz int32, fn *funcval) {
	argp := add(unsafe.Pointer(&fn), sys.PtrSize)
	pc := getcallerpc(unsafe.Pointer(&siz))
	systemstack(func() {
		newproc1(fn, (*uint8)(argp), siz, 0, pc)
	})
}

// Create a new g running fn with narg bytes of arguments starting
// at argp and returning nret bytes of results.  callerpc is the
// address of the go statement that created this. The new g is put
// on the queue of g's waiting to run.
func newproc1(fn *funcval, argp *uint8, narg int32, nret int32, callerpc uintptr) *g {
	_g_ := getg()

	if fn == nil {
		_g_.m.throwing = -1 // do not dump full stacks
		throw("go of nil func value")
	}
	_g_.m.locks++ // disable preemption because it can be holding p in a local var
	siz := narg + nret
	siz = (siz + 7) &^ 7

	// We could allocate a larger initial stack if necessary.
	// Not worth it: this is almost always an error.
	// 4*sizeof(uintreg): extra space added below
	// sizeof(uintreg): caller's LR (arm) or return address (x86, in gostartcall).
	if siz >= _StackMin-4*sys.RegSize-sys.RegSize {
		throw("newproc: function arguments too large for new goroutine")
	}

	_p_ := _g_.m.p.ptr()
	newg := gfget(_p_)
	if newg == nil {
		newg = malg(_StackMin)
		casgstatus(newg, _Gidle, _Gdead)
		newg.gcRescan = -1
		allgadd(newg) // publishes with a g->status of Gdead so GC scanner doesn't look at uninitialized stack.
	}
	if newg.stack.hi == 0 {
		throw("newproc1: newg missing stack")
	}

	if readgstatus(newg) != _Gdead {
		throw("newproc1: new g is not Gdead")
	}

	totalSize := 4*sys.RegSize + uintptr(siz) + sys.MinFrameSize // extra space in case of reads slightly beyond frame
	totalSize += -totalSize & (sys.SpAlign - 1)                  // align to spAlign
	sp := newg.stack.hi - totalSize
	spArg := sp
	if usesLR {
		// caller's LR
		*(*unsafe.Pointer)(unsafe.Pointer(sp)) = nil
		prepGoExitFrame(sp)
		spArg += sys.MinFrameSize
	}
	memmove(unsafe.Pointer(spArg), unsafe.Pointer(argp), uintptr(narg))

	memclr(unsafe.Pointer(&newg.sched), unsafe.Sizeof(newg.sched))
	newg.sched.sp = sp
	newg.stktopsp = sp
	newg.sched.pc = funcPC(goexit) + sys.PCQuantum // +PCQuantum so that previous instruction is in same function
	newg.sched.g = guintptr(unsafe.Pointer(newg))
	gostartcallfn(&newg.sched, fn)
	newg.gopc = callerpc
	newg.startpc = fn.fn
	if isSystemGoroutine(newg) {
		atomic.Xadd(&sched.ngsys, +1)
	}
	// The stack is dirty from the argument frame, so queue it for
	// scanning. Do this before setting it to runnable so we still
	// own the G. If we're recycling a G, it may already be on the
	// rescan list.
	if newg.gcRescan == -1 {
		queueRescan(newg)
	} else {
		// The recycled G is already on the rescan list. Just
		// mark the stack dirty.
		newg.gcscanvalid = false
	}
	casgstatus(newg, _Gdead, _Grunnable)

	if _p_.goidcache == _p_.goidcacheend {
		// Sched.goidgen is the last allocated id,
		// this batch must be [sched.goidgen+1, sched.goidgen+GoidCacheBatch].
		// At startup sched.goidgen=0, so main goroutine receives goid=1.
		_p_.goidcache = atomic.Xadd64(&sched.goidgen, _GoidCacheBatch)
		_p_.goidcache -= _GoidCacheBatch - 1
		_p_.goidcacheend = _p_.goidcache + _GoidCacheBatch
	}
	newg.goid = int64(_p_.goidcache)
	_p_.goidcache++
	if raceenabled {
		newg.racectx = racegostart(callerpc)
	}
	if trace.enabled {
		traceGoCreate(newg, newg.startpc)
	}
	runqput(_p_, newg, true)

	if atomic.Load(&sched.npidle) != 0 && atomic.Load(&sched.nmspinning) == 0 && unsafe.Pointer(fn.fn) != unsafe.Pointer(funcPC(main)) { // TODO: fast atomic
		wakep()
	}
	_g_.m.locks--
	if _g_.m.locks == 0 && _g_.preempt { // restore the preemption request in case we've cleared it in newstack
		_g_.stackguard0 = stackPreempt
	}
	return newg
}

// adjust Gobuf as it if executed a call to fn with context ctxt
// and then did an immediate gosave.
func gostartcall(buf *gobuf, fn, ctxt unsafe.Pointer) {
	sp := buf.sp
	if sys.RegSize > sys.PtrSize {
		sp -= sys.PtrSize
		*(*uintptr)(unsafe.Pointer(sp)) = 0
	}
	sp -= sys.PtrSize
	*(*uintptr)(unsafe.Pointer(sp)) = buf.pc
	buf.sp = sp
	buf.pc = uintptr(fn)
	buf.ctxt = ctxt
}

func gosave(buf *gobuf)

// Save state of caller into g->sched. Smashes R8, R9.
TEXT gosave<>(SB),NOSPLIT,$0
	get_tls(R8)
	MOVQ	g(R8), R8
	MOVQ	0(SP), R9
	MOVQ	R9, (g_sched+gobuf_pc)(R8)
	LEAQ	8(SP), R9
	MOVQ	R9, (g_sched+gobuf_sp)(R8)
	MOVQ	$0, (g_sched+gobuf_ret)(R8)
	MOVQ	$0, (g_sched+gobuf_ctxt)(R8)
	MOVQ	BP, (g_sched+gobuf_bp)(R8)
	RET

Goroutine 需要解决的问题

Metadata 组织

syscall 和 network 的问题

goroutine 之间的数据同步问题

##

参考