Ydb stable 22-4-4322.4.43

x-stable-origin-commit: 8d49d46cc834835bf3e50870516acd7376a63bcf

1 files changed, 6244 insertions, 0 deletions

diff --git a/contrib/go/_std_1.18/src/runtime/proc.go b/contrib/go/_std_1.18/src/runtime/proc.go
new file mode 100644
index 0000000000..b997a467ba
--- /dev/null
+++ b/contrib/go/_std_1.18/src/runtime/proc.go
@@ -0,0 +1,6244 @@
+// Copyright 2014 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package runtime
+
+import (
+	"internal/abi"
+	"internal/cpu"
+	"internal/goarch"
+	"runtime/internal/atomic"
+	"runtime/internal/sys"
+	"unsafe"
+)
+
+// set using cmd/go/internal/modload.ModInfoProg
+var modinfo string
+
+// Goroutine scheduler
+// The scheduler's job is to distribute ready-to-run goroutines over worker threads.
+//
+// The main concepts are:
+// G - goroutine.
+// M - worker thread, or machine.
+// P - processor, a resource that is required to execute Go code.
+//     M must have an associated P to execute Go code, however it can be
+//     blocked or in a syscall w/o an associated P.
+//
+// Design doc at https://golang.org/s/go11sched.
+
+// Worker thread parking/unparking.
+// We need to balance between keeping enough running worker threads to utilize
+// available hardware parallelism and parking excessive running worker threads
+// to conserve CPU resources and power. This is not simple for two reasons:
+// (1) scheduler state is intentionally distributed (in particular, per-P work
+// queues), so it is not possible to compute global predicates on fast paths;
+// (2) for optimal thread management we would need to know the future (don't park
+// a worker thread when a new goroutine will be readied in near future).
+//
+// Three rejected approaches that would work badly:
+// 1. Centralize all scheduler state (would inhibit scalability).
+// 2. Direct goroutine handoff. That is, when we ready a new goroutine and there
+//    is a spare P, unpark a thread and handoff it the thread and the goroutine.
+//    This would lead to thread state thrashing, as the thread that readied the
+//    goroutine can be out of work the very next moment, we will need to park it.
+//    Also, it would destroy locality of computation as we want to preserve
+//    dependent goroutines on the same thread; and introduce additional latency.
+// 3. Unpark an additional thread whenever we ready a goroutine and there is an
+//    idle P, but don't do handoff. This would lead to excessive thread parking/
+//    unparking as the additional threads will instantly park without discovering
+//    any work to do.
+//
+// The current approach:
+//
+// This approach applies to three primary sources of potential work: readying a
+// goroutine, new/modified-earlier timers, and idle-priority GC. See below for
+// additional details.
+//
+// We unpark an additional thread when we submit work if (this is wakep()):
+// 1. There is an idle P, and
+// 2. There are no "spinning" worker threads.
+//
+// A worker thread is considered spinning if it is out of local work and did
+// not find work in the global run queue or netpoller; the spinning state is
+// denoted in m.spinning and in sched.nmspinning. Threads unparked this way are
+// also considered spinning; we don't do goroutine handoff so such threads are
+// out of work initially. Spinning threads spin on looking for work in per-P
+// run queues and timer heaps or from the GC before parking. If a spinning
+// thread finds work it takes itself out of the spinning state and proceeds to
+// execution. If it does not find work it takes itself out of the spinning
+// state and then parks.
+//
+// If there is at least one spinning thread (sched.nmspinning>1), we don't
+// unpark new threads when submitting work. To compensate for that, if the last
+// spinning thread finds work and stops spinning, it must unpark a new spinning
+// thread.  This approach smooths out unjustified spikes of thread unparking,
+// but at the same time guarantees eventual maximal CPU parallelism
+// utilization.
+//
+// The main implementation complication is that we need to be very careful
+// during spinning->non-spinning thread transition. This transition can race
+// with submission of new work, and either one part or another needs to unpark
+// another worker thread. If they both fail to do that, we can end up with
+// semi-persistent CPU underutilization.
+//
+// The general pattern for submission is:
+// 1. Submit work to the local run queue, timer heap, or GC state.
+// 2. #StoreLoad-style memory barrier.
+// 3. Check sched.nmspinning.
+//
+// The general pattern for spinning->non-spinning transition is:
+// 1. Decrement nmspinning.
+// 2. #StoreLoad-style memory barrier.
+// 3. Check all per-P work queues and GC for new work.
+//
+// Note that all this complexity does not apply to global run queue as we are
+// not sloppy about thread unparking when submitting to global queue. Also see
+// comments for nmspinning manipulation.
+//
+// How these different sources of work behave varies, though it doesn't affect
+// the synchronization approach:
+// * Ready goroutine: this is an obvious source of work; the goroutine is
+//   immediately ready and must run on some thread eventually.
+// * New/modified-earlier timer: The current timer implementation (see time.go)
+//   uses netpoll in a thread with no work available to wait for the soonest
+//   timer. If there is no thread waiting, we want a new spinning thread to go
+//   wait.
+// * Idle-priority GC: The GC wakes a stopped idle thread to contribute to
+//   background GC work (note: currently disabled per golang.org/issue/19112).
+//   Also see golang.org/issue/44313, as this should be extended to all GC
+//   workers.
+
+var (
+	m0           m
+	g0           g
+	mcache0      *mcache
+	raceprocctx0 uintptr
+)
+
+//go:linkname runtime_inittask runtime..inittask
+var runtime_inittask initTask
+
+//go:linkname main_inittask main..inittask
+var main_inittask initTask
+
+// main_init_done is a signal used by cgocallbackg that initialization
+// has been completed. It is made before _cgo_notify_runtime_init_done,
+// so all cgo calls can rely on it existing. When main_init is complete,
+// it is closed, meaning cgocallbackg can reliably receive from it.
+var main_init_done chan bool
+
+//go:linkname main_main main.main
+func main_main()
+
+// mainStarted indicates that the main M has started.
+var mainStarted bool
+
+// runtimeInitTime is the nanotime() at which the runtime started.
+var runtimeInitTime int64
+
+// Value to use for signal mask for newly created M's.
+var initSigmask sigset
+
+// The main goroutine.
+func main() {
+	g := getg()
+
+	// Racectx of m0->g0 is used only as the parent of the main goroutine.
+	// It must not be used for anything else.
+	g.m.g0.racectx = 0
+
+	// Max stack size is 1 GB on 64-bit, 250 MB on 32-bit.
+	// Using decimal instead of binary GB and MB because
+	// they look nicer in the stack overflow failure message.
+	if goarch.PtrSize == 8 {
+		maxstacksize = 1000000000
+	} else {
+		maxstacksize = 250000000
+	}
+
+	// An upper limit for max stack size. Used to avoid random crashes
+	// after calling SetMaxStack and trying to allocate a stack that is too big,
+	// since stackalloc works with 32-bit sizes.
+	maxstackceiling = 2 * maxstacksize
+
+	// Allow newproc to start new Ms.
+	mainStarted = true
+
+	if GOARCH != "wasm" { // no threads on wasm yet, so no sysmon
+		systemstack(func() {
+			newm(sysmon, nil, -1)
+		})
+	}
+
+	// Lock the main goroutine onto this, the main OS thread,
+	// during initialization. Most programs won't care, but a few
+	// do require certain calls to be made by the main thread.
+	// Those can arrange for main.main to run in the main thread
+	// by calling runtime.LockOSThread during initialization
+	// to preserve the lock.
+	lockOSThread()
+
+	if g.m != &m0 {
+		throw("runtime.main not on m0")
+	}
+
+	// Record when the world started.
+	// Must be before doInit for tracing init.
+	runtimeInitTime = nanotime()
+	if runtimeInitTime == 0 {
+		throw("nanotime returning zero")
+	}
+
+	if debug.inittrace != 0 {
+		inittrace.id = getg().goid
+		inittrace.active = true
+	}
+
+	doInit(&runtime_inittask) // Must be before defer.
+
+	// Defer unlock so that runtime.Goexit during init does the unlock too.
+	needUnlock := true
+	defer func() {
+		if needUnlock {
+			unlockOSThread()
+		}
+	}()
+
+	gcenable()
+
+	main_init_done = make(chan bool)
+	if iscgo {
+		if _cgo_thread_start == nil {
+			throw("_cgo_thread_start missing")
+		}
+		if GOOS != "windows" {
+			if _cgo_setenv == nil {
+				throw("_cgo_setenv missing")
+			}
+			if _cgo_unsetenv == nil {
+				throw("_cgo_unsetenv missing")
+			}
+		}
+		if _cgo_notify_runtime_init_done == nil {
+			throw("_cgo_notify_runtime_init_done missing")
+		}
+		// Start the template thread in case we enter Go from
+		// a C-created thread and need to create a new thread.
+		startTemplateThread()
+		cgocall(_cgo_notify_runtime_init_done, nil)
+	}
+
+	doInit(&main_inittask)
+
+	// Disable init tracing after main init done to avoid overhead
+	// of collecting statistics in malloc and newproc
+	inittrace.active = false
+
+	close(main_init_done)
+
+	needUnlock = false
+	unlockOSThread()
+
+	if isarchive || islibrary {
+		// A program compiled with -buildmode=c-archive or c-shared
+		// has a main, but it is not executed.
+		return
+	}
+	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()
+	if raceenabled {
+		racefini()
+	}
+
+	// Make racy client program work: if panicking on
+	// another goroutine at the same time as main returns,
+	// let the other goroutine finish printing the panic trace.
+	// Once it does, it will exit. See issues 3934 and 20018.
+	if atomic.Load(&runningPanicDefers) != 0 {
+		// Running deferred functions should not take long.
+		for c := 0; c < 1000; c++ {
+			if atomic.Load(&runningPanicDefers) == 0 {
+				break
+			}
+			Gosched()
+		}
+	}
+	if atomic.Load(&panicking) != 0 {
+		gopark(nil, nil, waitReasonPanicWait, traceEvGoStop, 1)
+	}
+
+	exit(0)
+	for {
+		var x *int32
+		*x = 0
+	}
+}
+
+// os_beforeExit is called from os.Exit(0).
+//go:linkname os_beforeExit os.runtime_beforeExit
+func os_beforeExit() {
+	if raceenabled {
+		racefini()
+	}
+}
+
+// start forcegc helper goroutine
+func init() {
+	go forcegchelper()
+}
+
+func forcegchelper() {
+	forcegc.g = getg()
+	lockInit(&forcegc.lock, lockRankForcegc)
+	for {
+		lock(&forcegc.lock)
+		if forcegc.idle != 0 {
+			throw("forcegc: phase error")
+		}
+		atomic.Store(&forcegc.idle, 1)
+		goparkunlock(&forcegc.lock, waitReasonForceGCIdle, traceEvGoBlock, 1)
+		// this goroutine is explicitly resumed by sysmon
+		if debug.gctrace > 0 {
+			println("GC forced")
+		}
+		// Time-triggered, fully concurrent.
+		gcStart(gcTrigger{kind: gcTriggerTime, now: nanotime()})
+	}
+}
+
+//go:nosplit
+
+// Gosched yields the processor, allowing other goroutines to run. It does not
+// suspend the current goroutine, so execution resumes automatically.
+func Gosched() {
+	checkTimeouts()
+	mcall(gosched_m)
+}
+
+// goschedguarded yields the processor like gosched, but also checks
+// for forbidden states and opts out of the yield in those cases.
+//go:nosplit
+func goschedguarded() {
+	mcall(goschedguarded_m)
+}
+
+// Puts the current goroutine into a waiting state and calls unlockf on the
+// system stack.
+//
+// If unlockf returns false, the goroutine is resumed.
+//
+// unlockf must not access this G's stack, as it may be moved between
+// the call to gopark and the call to unlockf.
+//
+// Note that because unlockf is called after putting the G into a waiting
+// state, the G may have already been readied by the time unlockf is called
+// unless there is external synchronization preventing the G from being
+// readied. If unlockf returns false, it must guarantee that the G cannot be
+// externally readied.
+//
+// Reason explains why the goroutine has been parked. It is displayed in stack
+// traces and heap dumps. Reasons should be unique and descriptive. Do not
+// re-use reasons, add new ones.
+func gopark(unlockf func(*g, unsafe.Pointer) bool, lock unsafe.Pointer, reason waitReason, traceEv byte, traceskip int) {
+	if reason != waitReasonSleep {
+		checkTimeouts() // timeouts may expire while two goroutines keep the scheduler busy
+	}
+	mp := acquirem()
+	gp := mp.curg
+	status := readgstatus(gp)
+	if status != _Grunning && status != _Gscanrunning {
+		throw("gopark: bad g status")
+	}
+	mp.waitlock = lock
+	mp.waitunlockf = unlockf
+	gp.waitreason = reason
+	mp.waittraceev = traceEv
+	mp.waittraceskip = traceskip
+	releasem(mp)
+	// can't do anything that might move the G between Ms here.
+	mcall(park_m)
+}
+
+// Puts the current goroutine into a waiting state and unlocks the lock.
+// The goroutine can be made runnable again by calling goready(gp).
+func goparkunlock(lock *mutex, reason waitReason, traceEv byte, traceskip int) {
+	gopark(parkunlock_c, unsafe.Pointer(lock), reason, traceEv, traceskip)
+}
+
+func goready(gp *g, traceskip int) {
+	systemstack(func() {
+		ready(gp, traceskip, true)
+	})
+}
+
+//go:nosplit
+func acquireSudog() *sudog {
+	// Delicate dance: the semaphore implementation calls
+	// acquireSudog, acquireSudog calls new(sudog),
+	// new calls malloc, malloc can call the garbage collector,
+	// and the garbage collector calls the semaphore implementation
+	// in stopTheWorld.
+	// Break the cycle by doing acquirem/releasem around new(sudog).
+	// The acquirem/releasem increments m.locks during new(sudog),
+	// which keeps the garbage collector from being invoked.
+	mp := acquirem()
+	pp := mp.p.ptr()
+	if len(pp.sudogcache) == 0 {
+		lock(&sched.sudoglock)
+		// First, try to grab a batch from central cache.
+		for len(pp.sudogcache) < cap(pp.sudogcache)/2 && sched.sudogcache != nil {
+			s := sched.sudogcache
+			sched.sudogcache = s.next
+			s.next = nil
+			pp.sudogcache = append(pp.sudogcache, s)
+		}
+		unlock(&sched.sudoglock)
+		// If the central cache is empty, allocate a new one.
+		if len(pp.sudogcache) == 0 {
+			pp.sudogcache = append(pp.sudogcache, new(sudog))
+		}
+	}
+	n := len(pp.sudogcache)
+	s := pp.sudogcache[n-1]
+	pp.sudogcache[n-1] = nil
+	pp.sudogcache = pp.sudogcache[:n-1]
+	if s.elem != nil {
+		throw("acquireSudog: found s.elem != nil in cache")
+	}
+	releasem(mp)
+	return s
+}
+
+//go:nosplit
+func releaseSudog(s *sudog) {
+	if s.elem != nil {
+		throw("runtime: sudog with non-nil elem")
+	}
+	if s.isSelect {
+		throw("runtime: sudog with non-false isSelect")
+	}
+	if s.next != nil {
+		throw("runtime: sudog with non-nil next")
+	}
+	if s.prev != nil {
+		throw("runtime: sudog with non-nil prev")
+	}
+	if s.waitlink != nil {
+		throw("runtime: sudog with non-nil waitlink")
+	}
+	if s.c != nil {
+		throw("runtime: sudog with non-nil c")
+	}
+	gp := getg()
+	if gp.param != nil {
+		throw("runtime: releaseSudog with non-nil gp.param")
+	}
+	mp := acquirem() // avoid rescheduling to another P
+	pp := mp.p.ptr()
+	if len(pp.sudogcache) == cap(pp.sudogcache) {
+		// Transfer half of local cache to the central cache.
+		var first, last *sudog
+		for len(pp.sudogcache) > cap(pp.sudogcache)/2 {
+			n := len(pp.sudogcache)
+			p := pp.sudogcache[n-1]
+			pp.sudogcache[n-1] = nil
+			pp.sudogcache = pp.sudogcache[:n-1]
+			if first == nil {
+				first = p
+			} else {
+				last.next = p
+			}
+			last = p
+		}
+		lock(&sched.sudoglock)
+		last.next = sched.sudogcache
+		sched.sudogcache = first
+		unlock(&sched.sudoglock)
+	}
+	pp.sudogcache = append(pp.sudogcache, s)
+	releasem(mp)
+}
+
+// called from assembly
+func badmcall(fn func(*g)) {
+	throw("runtime: mcall called on m->g0 stack")
+}
+
+func badmcall2(fn func(*g)) {
+	throw("runtime: mcall function returned")
+}
+
+func badreflectcall() {
+	panic(plainError("arg size to reflect.call more than 1GB"))
+}
+
+var badmorestackg0Msg = "fatal: morestack on g0\n"
+
+//go:nosplit
+//go:nowritebarrierrec
+func badmorestackg0() {
+	sp := stringStructOf(&badmorestackg0Msg)
+	write(2, sp.str, int32(sp.len))
+}
+
+var badmorestackgsignalMsg = "fatal: morestack on gsignal\n"
+
+//go:nosplit
+//go:nowritebarrierrec
+func badmorestackgsignal() {
+	sp := stringStructOf(&badmorestackgsignalMsg)
+	write(2, sp.str, int32(sp.len))
+}
+
+//go:nosplit
+func badctxt() {
+	throw("ctxt != 0")
+}
+
+func lockedOSThread() bool {
+	gp := getg()
+	return gp.lockedm != 0 && gp.m.lockedg != 0
+}
+
+var (
+	// allgs contains all Gs ever created (including dead Gs), and thus
+	// never shrinks.
+	//
+	// Access via the slice is protected by allglock or stop-the-world.
+	// Readers that cannot take the lock may (carefully!) use the atomic
+	// variables below.
+	allglock mutex
+	allgs    []*g
+
+	// allglen and allgptr are atomic variables that contain len(allgs) and
+	// &allgs[0] respectively. Proper ordering depends on totally-ordered
+	// loads and stores. Writes are protected by allglock.
+	//
+	// allgptr is updated before allglen. Readers should read allglen
+	// before allgptr to ensure that allglen is always <= len(allgptr). New
+	// Gs appended during the race can be missed. For a consistent view of
+	// all Gs, allglock must be held.
+	//
+	// allgptr copies should always be stored as a concrete type or
+	// unsafe.Pointer, not uintptr, to ensure that GC can still reach it
+	// even if it points to a stale array.
+	allglen uintptr
+	allgptr **g
+)
+
+func allgadd(gp *g) {
+	if readgstatus(gp) == _Gidle {
+		throw("allgadd: bad status Gidle")
+	}
+
+	lock(&allglock)
+	allgs = append(allgs, gp)
+	if &allgs[0] != allgptr {
+		atomicstorep(unsafe.Pointer(&allgptr), unsafe.Pointer(&allgs[0]))
+	}
+	atomic.Storeuintptr(&allglen, uintptr(len(allgs)))
+	unlock(&allglock)
+}
+
+// allGsSnapshot returns a snapshot of the slice of all Gs.
+//
+// The world must be stopped or allglock must be held.
+func allGsSnapshot() []*g {
+	assertWorldStoppedOrLockHeld(&allglock)
+
+	// Because the world is stopped or allglock is held, allgadd
+	// cannot happen concurrently with this. allgs grows
+	// monotonically and existing entries never change, so we can
+	// simply return a copy of the slice header. For added safety,
+	// we trim everything past len because that can still change.
+	return allgs[:len(allgs):len(allgs)]
+}
+
+// atomicAllG returns &allgs[0] and len(allgs) for use with atomicAllGIndex.
+func atomicAllG() (**g, uintptr) {
+	length := atomic.Loaduintptr(&allglen)
+	ptr := (**g)(atomic.Loadp(unsafe.Pointer(&allgptr)))
+	return ptr, length
+}
+
+// atomicAllGIndex returns ptr[i] with the allgptr returned from atomicAllG.
+func atomicAllGIndex(ptr **g, i uintptr) *g {
+	return *(**g)(add(unsafe.Pointer(ptr), i*goarch.PtrSize))
+}
+
+// forEachG calls fn on every G from allgs.
+//
+// forEachG takes a lock to exclude concurrent addition of new Gs.
+func forEachG(fn func(gp *g)) {
+	lock(&allglock)
+	for _, gp := range allgs {
+		fn(gp)
+	}
+	unlock(&allglock)
+}
+
+// forEachGRace calls fn on every G from allgs.
+//
+// forEachGRace avoids locking, but does not exclude addition of new Gs during
+// execution, which may be missed.
+func forEachGRace(fn func(gp *g)) {
+	ptr, length := atomicAllG()
+	for i := uintptr(0); i < length; i++ {
+		gp := atomicAllGIndex(ptr, i)
+		fn(gp)
+	}
+	return
+}
+
+const (
+	// Number of goroutine ids to grab from sched.goidgen to local per-P cache at once.
+	// 16 seems to provide enough amortization, but other than that it's mostly arbitrary number.
+	_GoidCacheBatch = 16
+)
+
+// cpuinit extracts the environment variable GODEBUG from the environment on
+// Unix-like operating systems and calls internal/cpu.Initialize.
+func cpuinit() {
+	const prefix = "GODEBUG="
+	var env string
+
+	switch GOOS {
+	case "aix", "darwin", "ios", "dragonfly", "freebsd", "netbsd", "openbsd", "illumos", "solaris", "linux":
+		cpu.DebugOptions = true
+
+		// Similar to goenv_unix but extracts the environment value for
+		// GODEBUG directly.
+		// TODO(moehrmann): remove when general goenvs() can be called before cpuinit()
+		n := int32(0)
+		for argv_index(argv, argc+1+n) != nil {
+			n++
+		}
+
+		for i := int32(0); i < n; i++ {
+			p := argv_index(argv, argc+1+i)
+			s := *(*string)(unsafe.Pointer(&stringStruct{unsafe.Pointer(p), findnull(p)}))
+
+			if hasPrefix(s, prefix) {
+				env = gostring(p)[len(prefix):]
+				break
+			}
+		}
+	}
+
+	cpu.Initialize(env)
+
+	// Support cpu feature variables are used in code generated by the compiler
+	// to guard execution of instructions that can not be assumed to be always supported.
+	switch GOARCH {
+	case "386", "amd64":
+		x86HasPOPCNT = cpu.X86.HasPOPCNT
+		x86HasSSE41 = cpu.X86.HasSSE41
+		x86HasFMA = cpu.X86.HasFMA
+
+	case "arm":
+		armHasVFPv4 = cpu.ARM.HasVFPv4
+
+	case "arm64":
+		arm64HasATOMICS = cpu.ARM64.HasATOMICS
+	}
+}
+
+// The bootstrap sequence is:
+//
+//	call osinit
+//	call schedinit
+//	make & queue new G
+//	call runtime·mstart
+//
+// The new G calls runtime·main.
+func schedinit() {
+	lockInit(&sched.lock, lockRankSched)
+	lockInit(&sched.sysmonlock, lockRankSysmon)
+	lockInit(&sched.deferlock, lockRankDefer)
+	lockInit(&sched.sudoglock, lockRankSudog)
+	lockInit(&deadlock, lockRankDeadlock)
+	lockInit(&paniclk, lockRankPanic)
+	lockInit(&allglock, lockRankAllg)
+	lockInit(&allpLock, lockRankAllp)
+	lockInit(&reflectOffs.lock, lockRankReflectOffs)
+	lockInit(&finlock, lockRankFin)
+	lockInit(&trace.bufLock, lockRankTraceBuf)
+	lockInit(&trace.stringsLock, lockRankTraceStrings)
+	lockInit(&trace.lock, lockRankTrace)
+	lockInit(&cpuprof.lock, lockRankCpuprof)
+	lockInit(&trace.stackTab.lock, lockRankTraceStackTab)
+	// Enforce that this lock is always a leaf lock.
+	// All of this lock's critical sections should be
+	// extremely short.
+	lockInit(&memstats.heapStats.noPLock, lockRankLeafRank)
+
+	// raceinit must be the first call to race detector.
+	// In particular, it must be done before mallocinit below calls racemapshadow.
+	_g_ := getg()
+	if raceenabled {
+		_g_.racectx, raceprocctx0 = raceinit()
+	}
+
+	sched.maxmcount = 10000
+
+	// The world starts stopped.
+	worldStopped()
+
+	moduledataverify()
+	stackinit()
+	mallocinit()
+	cpuinit()      // must run before alginit
+	alginit()      // maps, hash, fastrand must not be used before this call
+	fastrandinit() // must run before mcommoninit
+	mcommoninit(_g_.m, -1)
+	modulesinit()   // provides activeModules
+	typelinksinit() // uses maps, activeModules
+	itabsinit()     // uses activeModules
+	stkobjinit()    // must run before GC starts
+
+	sigsave(&_g_.m.sigmask)
+	initSigmask = _g_.m.sigmask
+
+	if offset := unsafe.Offsetof(sched.timeToRun); offset%8 != 0 {
+		println(offset)
+		throw("sched.timeToRun not aligned to 8 bytes")
+	}
+
+	goargs()
+	goenvs()
+	parsedebugvars()
+	gcinit()
+
+	lock(&sched.lock)
+	sched.lastpoll = uint64(nanotime())
+	procs := ncpu
+	if n, ok := atoi32(gogetenv("GOMAXPROCS")); ok && n > 0 {
+		procs = n
+	}
+	if procresize(procs) != nil {
+		throw("unknown runnable goroutine during bootstrap")
+	}
+	unlock(&sched.lock)
+
+	// World is effectively started now, as P's can run.
+	worldStarted()
+
+	// For cgocheck > 1, we turn on the write barrier at all times
+	// and check all pointer writes. We can't do this until after
+	// procresize because the write barrier needs a P.
+	if debug.cgocheck > 1 {
+		writeBarrier.cgo = true
+		writeBarrier.enabled = true
+		for _, p := range allp {
+			p.wbBuf.reset()
+		}
+	}
+
+	if buildVersion == "" {
+		// Condition should never trigger. This code just serves
+		// to ensure runtime·buildVersion is kept in the resulting binary.
+		buildVersion = "unknown"
+	}
+	if len(modinfo) == 1 {
+		// Condition should never trigger. This code just serves
+		// to ensure runtime·modinfo is kept in the resulting binary.
+		modinfo = ""
+	}
+}
+
+func dumpgstatus(gp *g) {
+	_g_ := getg()
+	print("runtime: gp: gp=", gp, ", goid=", gp.goid, ", gp->atomicstatus=", readgstatus(gp), "\n")
+	print("runtime:  g:  g=", _g_, ", goid=", _g_.goid, ",  g->atomicstatus=", readgstatus(_g_), "\n")
+}
+
+// sched.lock must be held.
+func checkmcount() {
+	assertLockHeld(&sched.lock)
+
+	if mcount() > sched.maxmcount {
+		print("runtime: program exceeds ", sched.maxmcount, "-thread limit\n")
+		throw("thread exhaustion")
+	}
+}
+
+// mReserveID returns the next ID to use for a new m. This new m is immediately
+// considered 'running' by checkdead.
+//
+// sched.lock must be held.
+func mReserveID() int64 {
+	assertLockHeld(&sched.lock)
+
+	if sched.mnext+1 < sched.mnext {
+		throw("runtime: thread ID overflow")
+	}
+	id := sched.mnext
+	sched.mnext++
+	checkmcount()
+	return id
+}
+
+// Pre-allocated ID may be passed as 'id', or omitted by passing -1.
+func mcommoninit(mp *m, id int64) {
+	_g_ := getg()
+
+	// g0 stack won't make sense for user (and is not necessary unwindable).
+	if _g_ != _g_.m.g0 {
+		callers(1, mp.createstack[:])
+	}
+
+	lock(&sched.lock)
+
+	if id >= 0 {
+		mp.id = id
+	} else {
+		mp.id = mReserveID()
+	}
+
+	lo := uint32(int64Hash(uint64(mp.id), fastrandseed))
+	hi := uint32(int64Hash(uint64(cputicks()), ^fastrandseed))
+	if lo|hi == 0 {
+		hi = 1
+	}
+	// Same behavior as for 1.17.
+	// TODO: Simplify ths.
+	if goarch.BigEndian {
+		mp.fastrand = uint64(lo)<<32 | uint64(hi)
+	} else {
+		mp.fastrand = uint64(hi)<<32 | uint64(lo)
+	}
+
+	mpreinit(mp)
+	if mp.gsignal != nil {
+		mp.gsignal.stackguard1 = mp.gsignal.stack.lo + _StackGuard
+	}
+
+	// Add to allm so garbage collector doesn't free g->m
+	// when it is just in a register or thread-local storage.
+	mp.alllink = allm
+
+	// NumCgoCall() iterates over allm w/o schedlock,
+	// so we need to publish it safely.
+	atomicstorep(unsafe.Pointer(&allm), unsafe.Pointer(mp))
+	unlock(&sched.lock)
+
+	// Allocate memory to hold a cgo traceback if the cgo call crashes.
+	if iscgo || GOOS == "solaris" || GOOS == "illumos" || GOOS == "windows" {
+		mp.cgoCallers = new(cgoCallers)
+	}
+}
+
+var fastrandseed uintptr
+
+func fastrandinit() {
+	s := (*[unsafe.Sizeof(fastrandseed)]byte)(unsafe.Pointer(&fastrandseed))[:]
+	getRandomData(s)
+}
+
+// Mark gp ready to run.
+func ready(gp *g, traceskip int, next bool) {
+	if trace.enabled {
+		traceGoUnpark(gp, traceskip)
+	}
+
+	status := readgstatus(gp)
+
+	// Mark runnable.
+	_g_ := getg()
+	mp := acquirem() // disable preemption because it can be holding p in a local var
+	if status&^_Gscan != _Gwaiting {
+		dumpgstatus(gp)
+		throw("bad g->status in ready")
+	}
+
+	// status is Gwaiting or Gscanwaiting, make Grunnable and put on runq
+	casgstatus(gp, _Gwaiting, _Grunnable)
+	runqput(_g_.m.p.ptr(), gp, next)
+	wakep()
+	releasem(mp)
+}
+
+// freezeStopWait is a large value that freezetheworld sets
+// sched.stopwait to in order to request that all Gs permanently stop.
+const freezeStopWait = 0x7fffffff
+
+// freezing is set to non-zero if the runtime is trying to freeze the
+// world.
+var freezing uint32
+
+// Similar to stopTheWorld but best-effort and can be called several times.
+// There is no reverse operation, used during crashing.
+// This function must not lock any mutexes.
+func freezetheworld() {
+	atomic.Store(&freezing, 1)
+	// stopwait and preemption requests can be lost
+	// due to races with concurrently executing threads,
+	// so try several times
+	for i := 0; i < 5; i++ {
+		// this should tell the scheduler to not start any new goroutines
+		sched.stopwait = freezeStopWait
+		atomic.Store(&sched.gcwaiting, 1)
+		// this should stop running goroutines
+		if !preemptall() {
+			break // no running goroutines
+		}
+		usleep(1000)
+	}
+	// to be sure
+	usleep(1000)
+	preemptall()
+	usleep(1000)
+}
+
+// All reads and writes of g's status go through readgstatus, casgstatus
+// castogscanstatus, casfrom_Gscanstatus.
+//go:nosplit
+func readgstatus(gp *g) uint32 {
+	return atomic.Load(&gp.atomicstatus)
+}
+
+// The Gscanstatuses are acting like locks and this releases them.
+// If it proves to be a performance hit we should be able to make these
+// simple atomic stores but for now we are going to throw if
+// we see an inconsistent state.
+func casfrom_Gscanstatus(gp *g, oldval, newval uint32) {
+	success := false
+
+	// Check that transition is valid.
+	switch oldval {
+	default:
+		print("runtime: casfrom_Gscanstatus bad oldval gp=", gp, ", oldval=", hex(oldval), ", newval=", hex(newval), "\n")
+		dumpgstatus(gp)
+		throw("casfrom_Gscanstatus:top gp->status is not in scan state")
+	case _Gscanrunnable,
+		_Gscanwaiting,
+		_Gscanrunning,
+		_Gscansyscall,
+		_Gscanpreempted:
+		if newval == oldval&^_Gscan {
+			success = atomic.Cas(&gp.atomicstatus, oldval, newval)
+		}
+	}
+	if !success {
+		print("runtime: casfrom_Gscanstatus failed gp=", gp, ", oldval=", hex(oldval), ", newval=", hex(newval), "\n")
+		dumpgstatus(gp)
+		throw("casfrom_Gscanstatus: gp->status is not in scan state")
+	}
+	releaseLockRank(lockRankGscan)
+}
+
+// This will return false if the gp is not in the expected status and the cas fails.
+// This acts like a lock acquire while the casfromgstatus acts like a lock release.
+func castogscanstatus(gp *g, oldval, newval uint32) bool {
+	switch oldval {
+	case _Grunnable,
+		_Grunning,
+		_Gwaiting,
+		_Gsyscall:
+		if newval == oldval|_Gscan {
+			r := atomic.Cas(&gp.atomicstatus, oldval, newval)
+			if r {
+				acquireLockRank(lockRankGscan)
+			}
+			return r
+
+		}
+	}
+	print("runtime: castogscanstatus oldval=", hex(oldval), " newval=", hex(newval), "\n")
+	throw("castogscanstatus")
+	panic("not reached")
+}
+
+// If asked to move to or from a Gscanstatus this will throw. Use the castogscanstatus
+// and casfrom_Gscanstatus instead.
+// casgstatus will loop if the g->atomicstatus is in a Gscan status until the routine that
+// put it in the Gscan state is finished.
+//go:nosplit
+func casgstatus(gp *g, oldval, newval uint32) {
+	if (oldval&_Gscan != 0) || (newval&_Gscan != 0) || oldval == newval {
+		systemstack(func() {
+			print("runtime: casgstatus: oldval=", hex(oldval), " newval=", hex(newval), "\n")
+			throw("casgstatus: bad incoming values")
+		})
+	}
+
+	acquireLockRank(lockRankGscan)
+	releaseLockRank(lockRankGscan)
+
+	// See https://golang.org/cl/21503 for justification of the yield delay.
+	const yieldDelay = 5 * 1000
+	var nextYield int64
+
+	// loop if gp->atomicstatus is in a scan state giving
+	// GC time to finish and change the state to oldval.
+	for i := 0; !atomic.Cas(&gp.atomicstatus, oldval, newval); i++ {
+		if oldval == _Gwaiting && gp.atomicstatus == _Grunnable {
+			throw("casgstatus: waiting for Gwaiting but is Grunnable")
+		}
+		if i == 0 {
+			nextYield = nanotime() + yieldDelay
+		}
+		if nanotime() < nextYield {
+			for x := 0; x < 10 && gp.atomicstatus != oldval; x++ {
+				procyield(1)
+			}
+		} else {
+			osyield()
+			nextYield = nanotime() + yieldDelay/2
+		}
+	}
+
+	// Handle tracking for scheduling latencies.
+	if oldval == _Grunning {
+		// Track every 8th time a goroutine transitions out of running.
+		if gp.trackingSeq%gTrackingPeriod == 0 {
+			gp.tracking = true
+		}
+		gp.trackingSeq++
+	}
+	if gp.tracking {
+		if oldval == _Grunnable {
+			// We transitioned out of runnable, so measure how much
+			// time we spent in this state and add it to
+			// runnableTime.
+			now := nanotime()
+			gp.runnableTime += now - gp.runnableStamp
+			gp.runnableStamp = 0
+		}
+		if newval == _Grunnable {
+			// We just transitioned into runnable, so record what
+			// time that happened.
+			now := nanotime()
+			gp.runnableStamp = now
+		} else if newval == _Grunning {
+			// We're transitioning into running, so turn off
+			// tracking and record how much time we spent in
+			// runnable.
+			gp.tracking = false
+			sched.timeToRun.record(gp.runnableTime)
+			gp.runnableTime = 0
+		}
+	}
+}
+
+// casgstatus(gp, oldstatus, Gcopystack), assuming oldstatus is Gwaiting or Grunnable.
+// Returns old status. Cannot call casgstatus directly, because we are racing with an
+// async wakeup that might come in from netpoll. If we see Gwaiting from the readgstatus,
+// it might have become Grunnable by the time we get to the cas. If we called casgstatus,
+// it would loop waiting for the status to go back to Gwaiting, which it never will.
+//go:nosplit
+func casgcopystack(gp *g) uint32 {
+	for {
+		oldstatus := readgstatus(gp) &^ _Gscan
+		if oldstatus != _Gwaiting && oldstatus != _Grunnable {
+			throw("copystack: bad status, not Gwaiting or Grunnable")
+		}
+		if atomic.Cas(&gp.atomicstatus, oldstatus, _Gcopystack) {
+			return oldstatus
+		}
+	}
+}
+
+// casGToPreemptScan transitions gp from _Grunning to _Gscan|_Gpreempted.
+//
+// TODO(austin): This is the only status operation that both changes
+// the status and locks the _Gscan bit. Rethink this.
+func casGToPreemptScan(gp *g, old, new uint32) {
+	if old != _Grunning || new != _Gscan|_Gpreempted {
+		throw("bad g transition")
+	}
+	acquireLockRank(lockRankGscan)
+	for !atomic.Cas(&gp.atomicstatus, _Grunning, _Gscan|_Gpreempted) {
+	}
+}
+
+// casGFromPreempted attempts to transition gp from _Gpreempted to
+// _Gwaiting. If successful, the caller is responsible for
+// re-scheduling gp.
+func casGFromPreempted(gp *g, old, new uint32) bool {
+	if old != _Gpreempted || new != _Gwaiting {
+		throw("bad g transition")
+	}
+	return atomic.Cas(&gp.atomicstatus, _Gpreempted, _Gwaiting)
+}
+
+// stopTheWorld stops all P's from executing goroutines, interrupting
+// all goroutines at GC safe points and records reason as the reason
+// for the stop. On return, only the current goroutine's P is running.
+// stopTheWorld must not be called from a system stack and the caller
+// must not hold worldsema. The caller must call startTheWorld when
+// other P's should resume execution.
+//
+// stopTheWorld is safe for multiple goroutines to call at the
+// same time. Each will execute its own stop, and the stops will
+// be serialized.
+//
+// This is also used by routines that do stack dumps. If the system is
+// in panic or being exited, this may not reliably stop all
+// goroutines.
+func stopTheWorld(reason string) {
+	semacquire(&worldsema)
+	gp := getg()
+	gp.m.preemptoff = reason
+	systemstack(func() {
+		// Mark the goroutine which called stopTheWorld preemptible so its
+		// stack may be scanned.
+		// This lets a mark worker scan us while we try to stop the world
+		// since otherwise we could get in a mutual preemption deadlock.
+		// We must not modify anything on the G stack because a stack shrink
+		// may occur. A stack shrink is otherwise OK though because in order
+		// to return from this function (and to leave the system stack) we
+		// must have preempted all goroutines, including any attempting
+		// to scan our stack, in which case, any stack shrinking will
+		// have already completed by the time we exit.
+		casgstatus(gp, _Grunning, _Gwaiting)
+		stopTheWorldWithSema()
+		casgstatus(gp, _Gwaiting, _Grunning)
+	})
+}
+
+// startTheWorld undoes the effects of stopTheWorld.
+func startTheWorld() {
+	systemstack(func() { startTheWorldWithSema(false) })
+
+	// worldsema must be held over startTheWorldWithSema to ensure
+	// gomaxprocs cannot change while worldsema is held.
+	//
+	// Release worldsema with direct handoff to the next waiter, but
+	// acquirem so that semrelease1 doesn't try to yield our time.
+	//
+	// Otherwise if e.g. ReadMemStats is being called in a loop,
+	// it might stomp on other attempts to stop the world, such as
+	// for starting or ending GC. The operation this blocks is
+	// so heavy-weight that we should just try to be as fair as
+	// possible here.
+	//
+	// We don't want to just allow us to get preempted between now
+	// and releasing the semaphore because then we keep everyone
+	// (including, for example, GCs) waiting longer.
+	mp := acquirem()
+	mp.preemptoff = ""
+	semrelease1(&worldsema, true, 0)
+	releasem(mp)
+}
+
+// stopTheWorldGC has the same effect as stopTheWorld, but blocks
+// until the GC is not running. It also blocks a GC from starting
+// until startTheWorldGC is called.
+func stopTheWorldGC(reason string) {
+	semacquire(&gcsema)
+	stopTheWorld(reason)
+}
+
+// startTheWorldGC undoes the effects of stopTheWorldGC.
+func startTheWorldGC() {
+	startTheWorld()
+	semrelease(&gcsema)
+}
+
+// Holding worldsema grants an M the right to try to stop the world.
+var worldsema uint32 = 1
+
+// Holding gcsema grants the M the right to block a GC, and blocks
+// until the current GC is done. In particular, it prevents gomaxprocs
+// from changing concurrently.
+//
+// TODO(mknyszek): Once gomaxprocs and the execution tracer can handle
+// being changed/enabled during a GC, remove this.
+var gcsema uint32 = 1
+
+// stopTheWorldWithSema is the core implementation of stopTheWorld.
+// The caller is responsible for acquiring worldsema and disabling
+// preemption first and then should stopTheWorldWithSema on the system
+// stack:
+//
+//	semacquire(&worldsema, 0)
+//	m.preemptoff = "reason"
+//	systemstack(stopTheWorldWithSema)
+//
+// When finished, the caller must either call startTheWorld or undo
+// these three operations separately:
+//
+//	m.preemptoff = ""
+//	systemstack(startTheWorldWithSema)
+//	semrelease(&worldsema)
+//
+// It is allowed to acquire worldsema once and then execute multiple
+// startTheWorldWithSema/stopTheWorldWithSema pairs.
+// Other P's are able to execute between successive calls to
+// startTheWorldWithSema and stopTheWorldWithSema.
+// Holding worldsema causes any other goroutines invoking
+// stopTheWorld to block.
+func stopTheWorldWithSema() {
+	_g_ := getg()
+
+	// If we hold a lock, then we won't be able to stop another M
+	// that is blocked trying to acquire the lock.
+	if _g_.m.locks > 0 {
+		throw("stopTheWorld: holding locks")
+	}
+
+	lock(&sched.lock)
+	sched.stopwait = gomaxprocs
+	atomic.Store(&sched.gcwaiting, 1)
+	preemptall()
+	// stop current P
+	_g_.m.p.ptr().status = _Pgcstop // Pgcstop is only diagnostic.
+	sched.stopwait--
+	// try to retake all P's in Psyscall status
+	for _, p := range allp {
+		s := p.status
+		if s == _Psyscall && atomic.Cas(&p.status, s, _Pgcstop) {
+			if trace.enabled {
+				traceGoSysBlock(p)
+				traceProcStop(p)
+			}
+			p.syscalltick++
+			sched.stopwait--
+		}
+	}
+	// stop idle P's
+	for {
+		p := pidleget()
+		if p == nil {
+			break
+		}
+		p.status = _Pgcstop
+		sched.stopwait--
+	}
+	wait := sched.stopwait > 0
+	unlock(&sched.lock)
+
+	// wait for remaining P's to stop voluntarily
+	if wait {
+		for {
+			// wait for 100us, then try to re-preempt in case of any races
+			if notetsleep(&sched.stopnote, 100*1000) {
+				noteclear(&sched.stopnote)
+				break
+			}
+			preemptall()
+		}
+	}
+
+	// sanity checks
+	bad := ""
+	if sched.stopwait != 0 {
+		bad = "stopTheWorld: not stopped (stopwait != 0)"
+	} else {
+		for _, p := range allp {
+			if p.status != _Pgcstop {
+				bad = "stopTheWorld: not stopped (status != _Pgcstop)"
+			}
+		}
+	}
+	if atomic.Load(&freezing) != 0 {
+		// Some other thread is panicking. This can cause the
+		// sanity checks above to fail if the panic happens in
+		// the signal handler on a stopped thread. Either way,
+		// we should halt this thread.
+		lock(&deadlock)
+		lock(&deadlock)
+	}
+	if bad != "" {
+		throw(bad)
+	}
+
+	worldStopped()
+}
+
+func startTheWorldWithSema(emitTraceEvent bool) int64 {
+	assertWorldStopped()
+
+	mp := acquirem() // disable preemption because it can be holding p in a local var
+	if netpollinited() {
+		list := netpoll(0) // non-blocking
+		injectglist(&list)
+	}
+	lock(&sched.lock)
+
+	procs := gomaxprocs
+	if newprocs != 0 {
+		procs = newprocs
+		newprocs = 0
+	}
+	p1 := procresize(procs)
+	sched.gcwaiting = 0
+	if sched.sysmonwait != 0 {
+		sched.sysmonwait = 0
+		notewakeup(&sched.sysmonnote)
+	}
+	unlock(&sched.lock)
+
+	worldStarted()
+
+	for p1 != nil {
+		p := p1
+		p1 = p1.link.ptr()
+		if p.m != 0 {
+			mp := p.m.ptr()
+			p.m = 0
+			if mp.nextp != 0 {
+				throw("startTheWorld: inconsistent mp->nextp")
+			}
+			mp.nextp.set(p)
+			notewakeup(&mp.park)
+		} else {
+			// Start M to run P.  Do not start another M below.
+			newm(nil, p, -1)
+		}
+	}
+
+	// Capture start-the-world time before doing clean-up tasks.
+	startTime := nanotime()
+	if emitTraceEvent {
+		traceGCSTWDone()
+	}
+
+	// Wakeup an additional proc in case we have excessive runnable goroutines
+	// in local queues or in the global queue. If we don't, the proc will park itself.
+	// If we have lots of excessive work, resetspinning will unpark additional procs as necessary.
+	wakep()
+
+	releasem(mp)
+
+	return startTime
+}
+
+// usesLibcall indicates whether this runtime performs system calls
+// via libcall.
+func usesLibcall() bool {
+	switch GOOS {
+	case "aix", "darwin", "illumos", "ios", "solaris", "windows":
+		return true
+	case "openbsd":
+		return GOARCH == "386" || GOARCH == "amd64" || GOARCH == "arm" || GOARCH == "arm64"
+	}
+	return false
+}
+
+// mStackIsSystemAllocated indicates whether this runtime starts on a
+// system-allocated stack.
+func mStackIsSystemAllocated() bool {
+	switch GOOS {
+	case "aix", "darwin", "plan9", "illumos", "ios", "solaris", "windows":
+		return true
+	case "openbsd":
+		switch GOARCH {
+		case "386", "amd64", "arm", "arm64":
+			return true
+		}
+	}
+	return false
+}
+
+// mstart is the entry-point for new Ms.
+// It is written in assembly, uses ABI0, is marked TOPFRAME, and calls mstart0.
+func mstart()
+
+// mstart0 is the Go entry-point for new Ms.
+// This must not split the stack because we may not even have stack
+// bounds set up yet.
+//
+// May run during STW (because it doesn't have a P yet), so write
+// barriers are not allowed.
+//
+//go:nosplit
+//go:nowritebarrierrec
+func mstart0() {
+	_g_ := getg()
+
+	osStack := _g_.stack.lo == 0
+	if osStack {
+		// Initialize stack bounds from system stack.
+		// Cgo may have left stack size in stack.hi.
+		// minit may update the stack bounds.
+		//
+		// Note: these bounds may not be very accurate.
+		// We set hi to &size, but there are things above
+		// it. The 1024 is supposed to compensate this,
+		// but is somewhat arbitrary.
+		size := _g_.stack.hi
+		if size == 0 {
+			size = 8192 * sys.StackGuardMultiplier
+		}
+		_g_.stack.hi = uintptr(noescape(unsafe.Pointer(&size)))
+		_g_.stack.lo = _g_.stack.hi - size + 1024
+	}
+	// Initialize stack guard so that we can start calling regular
+	// Go code.
+	_g_.stackguard0 = _g_.stack.lo + _StackGuard
+	// This is the g0, so we can also call go:systemstack
+	// functions, which check stackguard1.
+	_g_.stackguard1 = _g_.stackguard0
+	mstart1()
+
+	// Exit this thread.
+	if mStackIsSystemAllocated() {
+		// Windows, Solaris, illumos, Darwin, AIX and Plan 9 always system-allocate
+		// the stack, but put it in _g_.stack before mstart,
+		// so the logic above hasn't set osStack yet.
+		osStack = true
+	}
+	mexit(osStack)
+}
+
+// The go:noinline is to guarantee the getcallerpc/getcallersp below are safe,
+// so that we can set up g0.sched to return to the call of mstart1 above.
+//go:noinline
+func mstart1() {
+	_g_ := getg()
+
+	if _g_ != _g_.m.g0 {
+		throw("bad runtime·mstart")
+	}
+
+	// Set up m.g0.sched as a label returning to just
+	// after the mstart1 call in mstart0 above, for use by goexit0 and mcall.
+	// We're never coming back to mstart1 after we call schedule,
+	// so other calls can reuse the current frame.
+	// And goexit0 does a gogo that needs to return from mstart1
+	// and let mstart0 exit the thread.
+	_g_.sched.g = guintptr(unsafe.Pointer(_g_))
+	_g_.sched.pc = getcallerpc()
+	_g_.sched.sp = getcallersp()
+
+	asminit()
+	minit()
+
+	// Install signal handlers; after minit so that minit can
+	// prepare the thread to be able to handle the signals.
+	if _g_.m == &m0 {
+		mstartm0()
+	}
+
+	if fn := _g_.m.mstartfn; fn != nil {
+		fn()
+	}
+
+	if _g_.m != &m0 {
+		acquirep(_g_.m.nextp.ptr())
+		_g_.m.nextp = 0
+	}
+	schedule()
+}
+
+// mstartm0 implements part of mstart1 that only runs on the m0.
+//
+// Write barriers are allowed here because we know the GC can't be
+// running yet, so they'll be no-ops.
+//
+//go:yeswritebarrierrec
+func mstartm0() {
+	// Create an extra M for callbacks on threads not created by Go.
+	// An extra M is also needed on Windows for callbacks created by
+	// syscall.NewCallback. See issue #6751 for details.
+	if (iscgo || GOOS == "windows") && !cgoHasExtraM {
+		cgoHasExtraM = true
+		newextram()
+	}
+	initsig(false)
+}
+
+// mPark causes a thread to park itself, returning once woken.
+//go:nosplit
+func mPark() {
+	gp := getg()
+	notesleep(&gp.m.park)
+	noteclear(&gp.m.park)
+}
+
+// mexit tears down and exits the current thread.
+//
+// Don't call this directly to exit the thread, since it must run at
+// the top of the thread stack. Instead, use gogo(&_g_.m.g0.sched) to
+// unwind the stack to the point that exits the thread.
+//
+// It is entered with m.p != nil, so write barriers are allowed. It
+// will release the P before exiting.
+//
+//go:yeswritebarrierrec
+func mexit(osStack bool) {
+	g := getg()
+	m := g.m
+
+	if m == &m0 {
+		// This is the main thread. Just wedge it.
+		//
+		// On Linux, exiting the main thread puts the process
+		// into a non-waitable zombie state. On Plan 9,
+		// exiting the main thread unblocks wait even though
+		// other threads are still running. On Solaris we can
+		// neither exitThread nor return from mstart. Other
+		// bad things probably happen on other platforms.
+		//
+		// We could try to clean up this M more before wedging
+		// it, but that complicates signal handling.
+		handoffp(releasep())
+		lock(&sched.lock)
+		sched.nmfreed++
+		checkdead()
+		unlock(&sched.lock)
+		mPark()
+		throw("locked m0 woke up")
+	}
+
+	sigblock(true)
+	unminit()
+
+	// Free the gsignal stack.
+	if m.gsignal != nil {
+		stackfree(m.gsignal.stack)
+		// On some platforms, when calling into VDSO (e.g. nanotime)
+		// we store our g on the gsignal stack, if there is one.
+		// Now the stack is freed, unlink it from the m, so we
+		// won't write to it when calling VDSO code.
+		m.gsignal = nil
+	}
+
+	// Remove m from allm.
+	lock(&sched.lock)
+	for pprev := &allm; *pprev != nil; pprev = &(*pprev).alllink {
+		if *pprev == m {
+			*pprev = m.alllink
+			goto found
+		}
+	}
+	throw("m not found in allm")
+found:
+	if !osStack {
+		// Delay reaping m until it's done with the stack.
+		//
+		// If this is using an OS stack, the OS will free it
+		// so there's no need for reaping.
+		atomic.Store(&m.freeWait, 1)
+		// Put m on the free list, though it will not be reaped until
+		// freeWait is 0. Note that the free list must not be linked
+		// through alllink because some functions walk allm without
+		// locking, so may be using alllink.
+		m.freelink = sched.freem
+		sched.freem = m
+	}
+	unlock(&sched.lock)
+
+	atomic.Xadd64(&ncgocall, int64(m.ncgocall))
+
+	// Release the P.
+	handoffp(releasep())
+	// After this point we must not have write barriers.
+
+	// Invoke the deadlock detector. This must happen after
+	// handoffp because it may have started a new M to take our
+	// P's work.
+	lock(&sched.lock)
+	sched.nmfreed++
+	checkdead()
+	unlock(&sched.lock)
+
+	if GOOS == "darwin" || GOOS == "ios" {
+		// Make sure pendingPreemptSignals is correct when an M exits.
+		// For #41702.
+		if atomic.Load(&m.signalPending) != 0 {
+			atomic.Xadd(&pendingPreemptSignals, -1)
+		}
+	}
+
+	// Destroy all allocated resources. After this is called, we may no
+	// longer take any locks.
+	mdestroy(m)
+
+	if osStack {
+		// Return from mstart and let the system thread
+		// library free the g0 stack and terminate the thread.
+		return
+	}
+
+	// mstart is the thread's entry point, so there's nothing to
+	// return to. Exit the thread directly. exitThread will clear
+	// m.freeWait when it's done with the stack and the m can be
+	// reaped.
+	exitThread(&m.freeWait)
+}
+
+// forEachP calls fn(p) for every P p when p reaches a GC safe point.
+// If a P is currently executing code, this will bring the P to a GC
+// safe point and execute fn on that P. If the P is not executing code
+// (it is idle or in a syscall), this will call fn(p) directly while
+// preventing the P from exiting its state. This does not ensure that
+// fn will run on every CPU executing Go code, but it acts as a global
+// memory barrier. GC uses this as a "ragged barrier."
+//
+// The caller must hold worldsema.
+//
+//go:systemstack
+func forEachP(fn func(*p)) {
+	mp := acquirem()
+	_p_ := getg().m.p.ptr()
+
+	lock(&sched.lock)
+	if sched.safePointWait != 0 {
+		throw("forEachP: sched.safePointWait != 0")
+	}
+	sched.safePointWait = gomaxprocs - 1
+	sched.safePointFn = fn
+
+	// Ask all Ps to run the safe point function.
+	for _, p := range allp {
+		if p != _p_ {
+			atomic.Store(&p.runSafePointFn, 1)
+		}
+	}
+	preemptall()
+
+	// Any P entering _Pidle or _Psyscall from now on will observe
+	// p.runSafePointFn == 1 and will call runSafePointFn when
+	// changing its status to _Pidle/_Psyscall.
+
+	// Run safe point function for all idle Ps. sched.pidle will
+	// not change because we hold sched.lock.
+	for p := sched.pidle.ptr(); p != nil; p = p.link.ptr() {
+		if atomic.Cas(&p.runSafePointFn, 1, 0) {
+			fn(p)
+			sched.safePointWait--
+		}
+	}
+
+	wait := sched.safePointWait > 0
+	unlock(&sched.lock)
+
+	// Run fn for the current P.
+	fn(_p_)
+
+	// Force Ps currently in _Psyscall into _Pidle and hand them
+	// off to induce safe point function execution.
+	for _, p := range allp {
+		s := p.status
+		if s == _Psyscall && p.runSafePointFn == 1 && atomic.Cas(&p.status, s, _Pidle) {
+			if trace.enabled {
+				traceGoSysBlock(p)
+				traceProcStop(p)
+			}
+			p.syscalltick++
+			handoffp(p)
+		}
+	}
+
+	// Wait for remaining Ps to run fn.
+	if wait {
+		for {
+			// Wait for 100us, then try to re-preempt in
+			// case of any races.
+			//
+			// Requires system stack.
+			if notetsleep(&sched.safePointNote, 100*1000) {
+				noteclear(&sched.safePointNote)
+				break
+			}
+			preemptall()
+		}
+	}
+	if sched.safePointWait != 0 {
+		throw("forEachP: not done")
+	}
+	for _, p := range allp {
+		if p.runSafePointFn != 0 {
+			throw("forEachP: P did not run fn")
+		}
+	}
+
+	lock(&sched.lock)
+	sched.safePointFn = nil
+	unlock(&sched.lock)
+	releasem(mp)
+}
+
+// runSafePointFn runs the safe point function, if any, for this P.
+// This should be called like
+//
+//     if getg().m.p.runSafePointFn != 0 {
+//         runSafePointFn()
+//     }
+//
+// runSafePointFn must be checked on any transition in to _Pidle or
+// _Psyscall to avoid a race where forEachP sees that the P is running
+// just before the P goes into _Pidle/_Psyscall and neither forEachP
+// nor the P run the safe-point function.
+func runSafePointFn() {
+	p := getg().m.p.ptr()
+	// Resolve the race between forEachP running the safe-point
+	// function on this P's behalf and this P running the
+	// safe-point function directly.
+	if !atomic.Cas(&p.runSafePointFn, 1, 0) {
+		return
+	}
+	sched.safePointFn(p)
+	lock(&sched.lock)
+	sched.safePointWait--
+	if sched.safePointWait == 0 {
+		notewakeup(&sched.safePointNote)
+	}
+	unlock(&sched.lock)
+}
+
+// When running with cgo, we call _cgo_thread_start
+// to start threads for us so that we can play nicely with
+// foreign code.
+var cgoThreadStart unsafe.Pointer
+
+type cgothreadstart struct {
+	g   guintptr
+	tls *uint64
+	fn  unsafe.Pointer
+}
+
+// Allocate a new m unassociated with any thread.
+// Can use p for allocation context if needed.
+// fn is recorded as the new m's m.mstartfn.
+// id is optional pre-allocated m ID. Omit by passing -1.
+//
+// This function is allowed to have write barriers even if the caller
+// isn't because it borrows _p_.
+//
+//go:yeswritebarrierrec
+func allocm(_p_ *p, fn func(), id int64) *m {
+	allocmLock.rlock()
+
+	// The caller owns _p_, but we may borrow (i.e., acquirep) it. We must
+	// disable preemption to ensure it is not stolen, which would make the
+	// caller lose ownership.
+	acquirem()
+
+	_g_ := getg()
+	if _g_.m.p == 0 {
+		acquirep(_p_) // temporarily borrow p for mallocs in this function
+	}
+
+	// Release the free M list. We need to do this somewhere and
+	// this may free up a stack we can use.
+	if sched.freem != nil {
+		lock(&sched.lock)
+		var newList *m
+		for freem := sched.freem; freem != nil; {
+			if freem.freeWait != 0 {
+				next := freem.freelink
+				freem.freelink = newList
+				newList = freem
+				freem = next
+				continue
+			}
+			// stackfree must be on the system stack, but allocm is
+			// reachable off the system stack transitively from
+			// startm.
+			systemstack(func() {
+				stackfree(freem.g0.stack)
+			})
+			freem = freem.freelink
+		}
+		sched.freem = newList
+		unlock(&sched.lock)
+	}
+
+	mp := new(m)
+	mp.mstartfn = fn
+	mcommoninit(mp, id)
+
+	// In case of cgo or Solaris or illumos or Darwin, pthread_create will make us a stack.
+	// Windows and Plan 9 will layout sched stack on OS stack.
+	if iscgo || mStackIsSystemAllocated() {
+		mp.g0 = malg(-1)
+	} else {
+		mp.g0 = malg(8192 * sys.StackGuardMultiplier)
+	}
+	mp.g0.m = mp
+
+	if _p_ == _g_.m.p.ptr() {
+		releasep()
+	}
+
+	releasem(_g_.m)
+	allocmLock.runlock()
+	return mp
+}
+
+// needm is called when a cgo callback happens on a
+// thread without an m (a thread not created by Go).
+// In this case, needm is expected to find an m to use
+// and return with m, g initialized correctly.
+// Since m and g are not set now (likely nil, but see below)
+// needm is limited in what routines it can call. In particular
+// it can only call nosplit functions (textflag 7) and cannot
+// do any scheduling that requires an m.
+//
+// In order to avoid needing heavy lifting here, we adopt
+// the following strategy: there is a stack of available m's
+// that can be stolen. Using compare-and-swap
+// to pop from the stack has ABA races, so we simulate
+// a lock by doing an exchange (via Casuintptr) to steal the stack
+// head and replace the top pointer with MLOCKED (1).
+// This serves as a simple spin lock that we can use even
+// without an m. The thread that locks the stack in this way
+// unlocks the stack by storing a valid stack head pointer.
+//
+// In order to make sure that there is always an m structure
+// available to be stolen, we maintain the invariant that there
+// is always one more than needed. At the beginning of the
+// program (if cgo is in use) the list is seeded with a single m.
+// If needm finds that it has taken the last m off the list, its job
+// is - once it has installed its own m so that it can do things like
+// allocate memory - to create a spare m and put it on the list.
+//
+// Each of these extra m's also has a g0 and a curg that are
+// pressed into service as the scheduling stack and current
+// goroutine for the duration of the cgo callback.
+//
+// When the callback is done with the m, it calls dropm to
+// put the m back on the list.
+//go:nosplit
+func needm() {
+	if (iscgo || GOOS == "windows") && !cgoHasExtraM {
+		// Can happen if C/C++ code calls Go from a global ctor.
+		// Can also happen on Windows if a global ctor uses a
+		// callback created by syscall.NewCallback. See issue #6751
+		// for details.
+		//
+		// Can not throw, because scheduler is not initialized yet.
+		write(2, unsafe.Pointer(&earlycgocallback[0]), int32(len(earlycgocallback)))
+		exit(1)
+	}
+
+	// Save and block signals before getting an M.
+	// The signal handler may call needm itself,
+	// and we must avoid a deadlock. Also, once g is installed,
+	// any incoming signals will try to execute,
+	// but we won't have the sigaltstack settings and other data
+	// set up appropriately until the end of minit, which will
+	// unblock the signals. This is the same dance as when
+	// starting a new m to run Go code via newosproc.
+	var sigmask sigset
+	sigsave(&sigmask)
+	sigblock(false)
+
+	// Lock extra list, take head, unlock popped list.
+	// nilokay=false is safe here because of the invariant above,
+	// that the extra list always contains or will soon contain
+	// at least one m.
+	mp := lockextra(false)
+
+	// Set needextram when we've just emptied the list,
+	// so that the eventual call into cgocallbackg will
+	// allocate a new m for the extra list. We delay the
+	// allocation until then so that it can be done
+	// after exitsyscall makes sure it is okay to be
+	// running at all (that is, there's no garbage collection
+	// running right now).
+	mp.needextram = mp.schedlink == 0
+	extraMCount--
+	unlockextra(mp.schedlink.ptr())
+
+	// Store the original signal mask for use by minit.
+	mp.sigmask = sigmask
+
+	// Install TLS on some platforms (previously setg
+	// would do this if necessary).
+	osSetupTLS(mp)
+
+	// Install g (= m->g0) and set the stack bounds
+	// to match the current stack. We don't actually know
+	// how big the stack is, like we don't know how big any
+	// scheduling stack is, but we assume there's at least 32 kB,
+	// which is more than enough for us.
+	setg(mp.g0)
+	_g_ := getg()
+	_g_.stack.hi = getcallersp() + 1024
+	_g_.stack.lo = getcallersp() - 32*1024
+	_g_.stackguard0 = _g_.stack.lo + _StackGuard
+
+	// Initialize this thread to use the m.
+	asminit()
+	minit()
+
+	// mp.curg is now a real goroutine.
+	casgstatus(mp.curg, _Gdead, _Gsyscall)
+	atomic.Xadd(&sched.ngsys, -1)
+}
+
+var earlycgocallback = []byte("fatal error: cgo callback before cgo call\n")
+
+// newextram allocates m's and puts them on the extra list.
+// It is called with a working local m, so that it can do things
+// like call schedlock and allocate.
+func newextram() {
+	c := atomic.Xchg(&extraMWaiters, 0)
+	if c > 0 {
+		for i := uint32(0); i < c; i++ {
+			oneNewExtraM()
+		}
+	} else {
+		// Make sure there is at least one extra M.
+		mp := lockextra(true)
+		unlockextra(mp)
+		if mp == nil {
+			oneNewExtraM()
+		}
+	}
+}
+
+// oneNewExtraM allocates an m and puts it on the extra list.
+func oneNewExtraM() {
+	// Create extra goroutine locked to extra m.
+	// The goroutine is the context in which the cgo callback will run.
+	// The sched.pc will never be returned to, but setting it to
+	// goexit makes clear to the traceback routines where
+	// the goroutine stack ends.
+	mp := allocm(nil, nil, -1)
+	gp := malg(4096)
+	gp.sched.pc = abi.FuncPCABI0(goexit) + sys.PCQuantum
+	gp.sched.sp = gp.stack.hi
+	gp.sched.sp -= 4 * goarch.PtrSize // extra space in case of reads slightly beyond frame
+	gp.sched.lr = 0
+	gp.sched.g = guintptr(unsafe.Pointer(gp))
+	gp.syscallpc = gp.sched.pc
+	gp.syscallsp = gp.sched.sp
+	gp.stktopsp = gp.sched.sp
+	// malg returns status as _Gidle. Change to _Gdead before
+	// adding to allg where GC can see it. We use _Gdead to hide
+	// this from tracebacks and stack scans since it isn't a
+	// "real" goroutine until needm grabs it.
+	casgstatus(gp, _Gidle, _Gdead)
+	gp.m = mp
+	mp.curg = gp
+	mp.lockedInt++
+	mp.lockedg.set(gp)
+	gp.lockedm.set(mp)
+	gp.goid = int64(atomic.Xadd64(&sched.goidgen, 1))
+	if raceenabled {
+		gp.racectx = racegostart(abi.FuncPCABIInternal(newextram) + sys.PCQuantum)
+	}
+	// put on allg for garbage collector
+	allgadd(gp)
+
+	// gp is now on the allg list, but we don't want it to be
+	// counted by gcount. It would be more "proper" to increment
+	// sched.ngfree, but that requires locking. Incrementing ngsys
+	// has the same effect.
+	atomic.Xadd(&sched.ngsys, +1)
+
+	// Add m to the extra list.
+	mnext := lockextra(true)
+	mp.schedlink.set(mnext)
+	extraMCount++
+	unlockextra(mp)
+}
+
+// dropm is called when a cgo callback has called needm but is now
+// done with the callback and returning back into the non-Go thread.
+// It puts the current m back onto the extra list.
+//
+// The main expense here is the call to signalstack to release the
+// m's signal stack, and then the call to needm on the next callback
+// from this thread. It is tempting to try to save the m for next time,
+// which would eliminate both these costs, but there might not be
+// a next time: the current thread (which Go does not control) might exit.
+// If we saved the m for that thread, there would be an m leak each time
+// such a thread exited. Instead, we acquire and release an m on each
+// call. These should typically not be scheduling operations, just a few
+// atomics, so the cost should be small.
+//
+// TODO(rsc): An alternative would be to allocate a dummy pthread per-thread
+// variable using pthread_key_create. Unlike the pthread keys we already use
+// on OS X, this dummy key would never be read by Go code. It would exist
+// only so that we could register at thread-exit-time destructor.
+// That destructor would put the m back onto the extra list.
+// This is purely a performance optimization. The current version,
+// in which dropm happens on each cgo call, is still correct too.
+// We may have to keep the current version on systems with cgo
+// but without pthreads, like Windows.
+func dropm() {
+	// Clear m and g, and return m to the extra list.
+	// After the call to setg we can only call nosplit functions
+	// with no pointer manipulation.
+	mp := getg().m
+
+	// Return mp.curg to dead state.
+	casgstatus(mp.curg, _Gsyscall, _Gdead)
+	mp.curg.preemptStop = false
+	atomic.Xadd(&sched.ngsys, +1)
+
+	// Block signals before unminit.
+	// Unminit unregisters the signal handling stack (but needs g on some systems).
+	// Setg(nil) clears g, which is the signal handler's cue not to run Go handlers.
+	// It's important not to try to handle a signal between those two steps.
+	sigmask := mp.sigmask
+	sigblock(false)
+	unminit()
+
+	mnext := lockextra(true)
+	extraMCount++
+	mp.schedlink.set(mnext)
+
+	setg(nil)
+
+	// Commit the release of mp.
+	unlockextra(mp)
+
+	msigrestore(sigmask)
+}
+
+// A helper function for EnsureDropM.
+func getm() uintptr {
+	return uintptr(unsafe.Pointer(getg().m))
+}
+
+var extram uintptr
+var extraMCount uint32 // Protected by lockextra
+var extraMWaiters uint32
+
+// lockextra locks the extra list and returns the list head.
+// The caller must unlock the list by storing a new list head
+// to extram. If nilokay is true, then lockextra will
+// return a nil list head if that's what it finds. If nilokay is false,
+// lockextra will keep waiting until the list head is no longer nil.
+//go:nosplit
+func lockextra(nilokay bool) *m {
+	const locked = 1
+
+	incr := false
+	for {
+		old := atomic.Loaduintptr(&extram)
+		if old == locked {
+			osyield_no_g()
+			continue
+		}
+		if old == 0 && !nilokay {
+			if !incr {
+				// Add 1 to the number of threads
+				// waiting for an M.
+				// This is cleared by newextram.
+				atomic.Xadd(&extraMWaiters, 1)
+				incr = true
+			}
+			usleep_no_g(1)
+			continue
+		}
+		if atomic.Casuintptr(&extram, old, locked) {
+			return (*m)(unsafe.Pointer(old))
+		}
+		osyield_no_g()
+		continue
+	}
+}
+
+//go:nosplit
+func unlockextra(mp *m) {
+	atomic.Storeuintptr(&extram, uintptr(unsafe.Pointer(mp)))
+}
+
+var (
+	// allocmLock is locked for read when creating new Ms in allocm and their
+	// addition to allm. Thus acquiring this lock for write blocks the
+	// creation of new Ms.
+	allocmLock rwmutex
+
+	// execLock serializes exec and clone to avoid bugs or unspecified
+	// behaviour around exec'ing while creating/destroying threads. See
+	// issue #19546.
+	execLock rwmutex
+)
+
+// newmHandoff contains a list of m structures that need new OS threads.
+// This is used by newm in situations where newm itself can't safely
+// start an OS thread.
+var newmHandoff struct {
+	lock mutex
+
+	// newm points to a list of M structures that need new OS
+	// threads. The list is linked through m.schedlink.
+	newm muintptr
+
+	// waiting indicates that wake needs to be notified when an m
+	// is put on the list.
+	waiting bool
+	wake    note
+
+	// haveTemplateThread indicates that the templateThread has
+	// been started. This is not protected by lock. Use cas to set
+	// to 1.
+	haveTemplateThread uint32
+}
+
+// Create a new m. It will start off with a call to fn, or else the scheduler.
+// fn needs to be static and not a heap allocated closure.
+// May run with m.p==nil, so write barriers are not allowed.
+//
+// id is optional pre-allocated m ID. Omit by passing -1.
+//go:nowritebarrierrec
+func newm(fn func(), _p_ *p, id int64) {
+	// allocm adds a new M to allm, but they do not start until created by
+	// the OS in newm1 or the template thread.
+	//
+	// doAllThreadsSyscall requires that every M in allm will eventually
+	// start and be signal-able, even with a STW.
+	//
+	// Disable preemption here until we start the thread to ensure that
+	// newm is not preempted between allocm and starting the new thread,
+	// ensuring that anything added to allm is guaranteed to eventually
+	// start.
+	acquirem()
+
+	mp := allocm(_p_, fn, id)
+	mp.nextp.set(_p_)
+	mp.sigmask = initSigmask
+	if gp := getg(); gp != nil && gp.m != nil && (gp.m.lockedExt != 0 || gp.m.incgo) && GOOS != "plan9" {
+		// We're on a locked M or a thread that may have been
+		// started by C. The kernel state of this thread may
+		// be strange (the user may have locked it for that
+		// purpose). We don't want to clone that into another
+		// thread. Instead, ask a known-good thread to create
+		// the thread for us.
+		//
+		// This is disabled on Plan 9. See golang.org/issue/22227.
+		//
+		// TODO: This may be unnecessary on Windows, which
+		// doesn't model thread creation off fork.
+		lock(&newmHandoff.lock)
+		if newmHandoff.haveTemplateThread == 0 {
+			throw("on a locked thread with no template thread")
+		}
+		mp.schedlink = newmHandoff.newm
+		newmHandoff.newm.set(mp)
+		if newmHandoff.waiting {
+			newmHandoff.waiting = false
+			notewakeup(&newmHandoff.wake)
+		}
+		unlock(&newmHandoff.lock)
+		// The M has not started yet, but the template thread does not
+		// participate in STW, so it will always process queued Ms and
+		// it is safe to releasem.
+		releasem(getg().m)
+		return
+	}
+	newm1(mp)
+	releasem(getg().m)
+}
+
+func newm1(mp *m) {
+	if iscgo {
+		var ts cgothreadstart
+		if _cgo_thread_start == nil {
+			throw("_cgo_thread_start missing")
+		}
+		ts.g.set(mp.g0)
+		ts.tls = (*uint64)(unsafe.Pointer(&mp.tls[0]))
+		ts.fn = unsafe.Pointer(abi.FuncPCABI0(mstart))
+		if msanenabled {
+			msanwrite(unsafe.Pointer(&ts), unsafe.Sizeof(ts))
+		}
+		if asanenabled {
+			asanwrite(unsafe.Pointer(&ts), unsafe.Sizeof(ts))
+		}
+		execLock.rlock() // Prevent process clone.
+		asmcgocall(_cgo_thread_start, unsafe.Pointer(&ts))
+		execLock.runlock()
+		return
+	}
+	execLock.rlock() // Prevent process clone.
+	newosproc(mp)
+	execLock.runlock()
+}
+
+// startTemplateThread starts the template thread if it is not already
+// running.
+//
+// The calling thread must itself be in a known-good state.
+func startTemplateThread() {
+	if GOARCH == "wasm" { // no threads on wasm yet
+		return
+	}
+
+	// Disable preemption to guarantee that the template thread will be
+	// created before a park once haveTemplateThread is set.
+	mp := acquirem()
+	if !atomic.Cas(&newmHandoff.haveTemplateThread, 0, 1) {
+		releasem(mp)
+		return
+	}
+	newm(templateThread, nil, -1)
+	releasem(mp)
+}
+
+// templateThread is a thread in a known-good state that exists solely
+// to start new threads in known-good states when the calling thread
+// may not be in a good state.
+//
+// Many programs never need this, so templateThread is started lazily
+// when we first enter a state that might lead to running on a thread
+// in an unknown state.
+//
+// templateThread runs on an M without a P, so it must not have write
+// barriers.
+//
+//go:nowritebarrierrec
+func templateThread() {
+	lock(&sched.lock)
+	sched.nmsys++
+	checkdead()
+	unlock(&sched.lock)
+
+	for {
+		lock(&newmHandoff.lock)
+		for newmHandoff.newm != 0 {
+			newm := newmHandoff.newm.ptr()
+			newmHandoff.newm = 0
+			unlock(&newmHandoff.lock)
+			for newm != nil {
+				next := newm.schedlink.ptr()
+				newm.schedlink = 0
+				newm1(newm)
+				newm = next
+			}
+			lock(&newmHandoff.lock)
+		}
+		newmHandoff.waiting = true
+		noteclear(&newmHandoff.wake)
+		unlock(&newmHandoff.lock)
+		notesleep(&newmHandoff.wake)
+	}
+}
+
+// Stops execution of the current m until new work is available.
+// Returns with acquired P.
+func stopm() {
+	_g_ := getg()
+
+	if _g_.m.locks != 0 {
+		throw("stopm holding locks")
+	}
+	if _g_.m.p != 0 {
+		throw("stopm holding p")
+	}
+	if _g_.m.spinning {
+		throw("stopm spinning")
+	}
+
+	lock(&sched.lock)
+	mput(_g_.m)
+	unlock(&sched.lock)
+	mPark()
+	acquirep(_g_.m.nextp.ptr())
+	_g_.m.nextp = 0
+}
+
+func mspinning() {
+	// startm's caller incremented nmspinning. Set the new M's spinning.
+	getg().m.spinning = true
+}
+
+// Schedules some M to run the p (creates an M if necessary).
+// If p==nil, tries to get an idle P, if no idle P's does nothing.
+// May run with m.p==nil, so write barriers are not allowed.
+// If spinning is set, the caller has incremented nmspinning and startm will
+// either decrement nmspinning or set m.spinning in the newly started M.
+//
+// Callers passing a non-nil P must call from a non-preemptible context. See
+// comment on acquirem below.
+//
+// Must not have write barriers because this may be called without a P.
+//go:nowritebarrierrec
+func startm(_p_ *p, spinning bool) {
+	// Disable preemption.
+	//
+	// Every owned P must have an owner that will eventually stop it in the
+	// event of a GC stop request. startm takes transient ownership of a P
+	// (either from argument or pidleget below) and transfers ownership to
+	// a started M, which will be responsible for performing the stop.
+	//
+	// Preemption must be disabled during this transient ownership,
+	// otherwise the P this is running on may enter GC stop while still
+	// holding the transient P, leaving that P in limbo and deadlocking the
+	// STW.
+	//
+	// Callers passing a non-nil P must already be in non-preemptible
+	// context, otherwise such preemption could occur on function entry to
+	// startm. Callers passing a nil P may be preemptible, so we must
+	// disable preemption before acquiring a P from pidleget below.
+	mp := acquirem()
+	lock(&sched.lock)
+	if _p_ == nil {
+		_p_ = pidleget()
+		if _p_ == nil {
+			unlock(&sched.lock)
+			if spinning {
+				// The caller incremented nmspinning, but there are no idle Ps,
+				// so it's okay to just undo the increment and give up.
+				if int32(atomic.Xadd(&sched.nmspinning, -1)) < 0 {
+					throw("startm: negative nmspinning")
+				}
+			}
+			releasem(mp)
+			return
+		}
+	}
+	nmp := mget()
+	if nmp == nil {
+		// No M is available, we must drop sched.lock and call newm.
+		// However, we already own a P to assign to the M.
+		//
+		// Once sched.lock is released, another G (e.g., in a syscall),
+		// could find no idle P while checkdead finds a runnable G but
+		// no running M's because this new M hasn't started yet, thus
+		// throwing in an apparent deadlock.
+		//
+		// Avoid this situation by pre-allocating the ID for the new M,
+		// thus marking it as 'running' before we drop sched.lock. This
+		// new M will eventually run the scheduler to execute any
+		// queued G's.
+		id := mReserveID()
+		unlock(&sched.lock)
+
+		var fn func()
+		if spinning {
+			// The caller incremented nmspinning, so set m.spinning in the new M.
+			fn = mspinning
+		}
+		newm(fn, _p_, id)
+		// Ownership transfer of _p_ committed by start in newm.
+		// Preemption is now safe.
+		releasem(mp)
+		return
+	}
+	unlock(&sched.lock)
+	if nmp.spinning {
+		throw("startm: m is spinning")
+	}
+	if nmp.nextp != 0 {
+		throw("startm: m has p")
+	}
+	if spinning && !runqempty(_p_) {
+		throw("startm: p has runnable gs")
+	}
+	// The caller incremented nmspinning, so set m.spinning in the new M.
+	nmp.spinning = spinning
+	nmp.nextp.set(_p_)
+	notewakeup(&nmp.park)
+	// Ownership transfer of _p_ committed by wakeup. Preemption is now
+	// safe.
+	releasem(mp)
+}
+
+// Hands off P from syscall or locked M.
+// Always runs without a P, so write barriers are not allowed.
+//go:nowritebarrierrec
+func handoffp(_p_ *p) {
+	// handoffp must start an M in any situation where
+	// findrunnable would return a G to run on _p_.
+
+	// if it has local work, start it straight away
+	if !runqempty(_p_) || sched.runqsize != 0 {
+		startm(_p_, false)
+		return
+	}
+	// if it has GC work, start it straight away
+	if gcBlackenEnabled != 0 && gcMarkWorkAvailable(_p_) {
+		startm(_p_, false)
+		return
+	}
+	// no local work, check that there are no spinning/idle M's,
+	// otherwise our help is not required
+	if atomic.Load(&sched.nmspinning)+atomic.Load(&sched.npidle) == 0 && atomic.Cas(&sched.nmspinning, 0, 1) { // TODO: fast atomic
+		startm(_p_, true)
+		return
+	}
+	lock(&sched.lock)
+	if sched.gcwaiting != 0 {
+		_p_.status = _Pgcstop
+		sched.stopwait--
+		if sched.stopwait == 0 {
+			notewakeup(&sched.stopnote)
+		}
+		unlock(&sched.lock)
+		return
+	}
+	if _p_.runSafePointFn != 0 && atomic.Cas(&_p_.runSafePointFn, 1, 0) {
+		sched.safePointFn(_p_)
+		sched.safePointWait--
+		if sched.safePointWait == 0 {
+			notewakeup(&sched.safePointNote)
+		}
+	}
+	if sched.runqsize != 0 {
+		unlock(&sched.lock)
+		startm(_p_, false)
+		return
+	}
+	// If this is the last running P and nobody is polling network,
+	// need to wakeup another M to poll network.
+	if sched.npidle == uint32(gomaxprocs-1) && atomic.Load64(&sched.lastpoll) != 0 {
+		unlock(&sched.lock)
+		startm(_p_, false)
+		return
+	}
+
+	// The scheduler lock cannot be held when calling wakeNetPoller below
+	// because wakeNetPoller may call wakep which may call startm.
+	when := nobarrierWakeTime(_p_)
+	pidleput(_p_)
+	unlock(&sched.lock)
+
+	if when != 0 {
+		wakeNetPoller(when)
+	}
+}
+
+// Tries to add one more P to execute G's.
+// Called when a G is made runnable (newproc, ready).
+func wakep() {
+	if atomic.Load(&sched.npidle) == 0 {
+		return
+	}
+	// be conservative about spinning threads
+	if atomic.Load(&sched.nmspinning) != 0 || !atomic.Cas(&sched.nmspinning, 0, 1) {
+		return
+	}
+	startm(nil, true)
+}
+
+// Stops execution of the current m that is locked to a g until the g is runnable again.
+// Returns with acquired P.
+func stoplockedm() {
+	_g_ := getg()
+
+	if _g_.m.lockedg == 0 || _g_.m.lockedg.ptr().lockedm.ptr() != _g_.m {
+		throw("stoplockedm: inconsistent locking")
+	}
+	if _g_.m.p != 0 {
+		// Schedule another M to run this p.
+		_p_ := releasep()
+		handoffp(_p_)
+	}
+	incidlelocked(1)
+	// Wait until another thread schedules lockedg again.
+	mPark()
+	status := readgstatus(_g_.m.lockedg.ptr())
+	if status&^_Gscan != _Grunnable {
+		print("runtime:stoplockedm: lockedg (atomicstatus=", status, ") is not Grunnable or Gscanrunnable\n")
+		dumpgstatus(_g_.m.lockedg.ptr())
+		throw("stoplockedm: not runnable")
+	}
+	acquirep(_g_.m.nextp.ptr())
+	_g_.m.nextp = 0
+}
+
+// Schedules the locked m to run the locked gp.
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func startlockedm(gp *g) {
+	_g_ := getg()
+
+	mp := gp.lockedm.ptr()
+	if mp == _g_.m {
+		throw("startlockedm: locked to me")
+	}
+	if mp.nextp != 0 {
+		throw("startlockedm: m has p")
+	}
+	// directly handoff current P to the locked m
+	incidlelocked(-1)
+	_p_ := releasep()
+	mp.nextp.set(_p_)
+	notewakeup(&mp.park)
+	stopm()
+}
+
+// Stops the current m for stopTheWorld.
+// Returns when the world is restarted.
+func gcstopm() {
+	_g_ := getg()
+
+	if sched.gcwaiting == 0 {
+		throw("gcstopm: not waiting for gc")
+	}
+	if _g_.m.spinning {
+		_g_.m.spinning = false
+		// OK to just drop nmspinning here,
+		// startTheWorld will unpark threads as necessary.
+		if int32(atomic.Xadd(&sched.nmspinning, -1)) < 0 {
+			throw("gcstopm: negative nmspinning")
+		}
+	}
+	_p_ := releasep()
+	lock(&sched.lock)
+	_p_.status = _Pgcstop
+	sched.stopwait--
+	if sched.stopwait == 0 {
+		notewakeup(&sched.stopnote)
+	}
+	unlock(&sched.lock)
+	stopm()
+}
+
+// Schedules gp to run on the current M.
+// If inheritTime is true, gp inherits the remaining time in the
+// current time slice. Otherwise, it starts a new time slice.
+// Never returns.
+//
+// Write barriers are allowed because this is called immediately after
+// acquiring a P in several places.
+//
+//go:yeswritebarrierrec
+func execute(gp *g, inheritTime bool) {
+	_g_ := getg()
+
+	// Assign gp.m before entering _Grunning so running Gs have an
+	// M.
+	_g_.m.curg = gp
+	gp.m = _g_.m
+	casgstatus(gp, _Grunnable, _Grunning)
+	gp.waitsince = 0
+	gp.preempt = false
+	gp.stackguard0 = gp.stack.lo + _StackGuard
+	if !inheritTime {
+		_g_.m.p.ptr().schedtick++
+	}
+
+	// Check whether the profiler needs to be turned on or off.
+	hz := sched.profilehz
+	if _g_.m.profilehz != hz {
+		setThreadCPUProfiler(hz)
+	}
+
+	if trace.enabled {
+		// GoSysExit has to happen when we have a P, but before GoStart.
+		// So we emit it here.
+		if gp.syscallsp != 0 && gp.sysblocktraced {
+			traceGoSysExit(gp.sysexitticks)
+		}
+		traceGoStart()
+	}
+
+	gogo(&gp.sched)
+}
+
+// Finds a runnable goroutine to execute.
+// Tries to steal from other P's, get g from local or global queue, poll network.
+func findrunnable() (gp *g, inheritTime bool) {
+	_g_ := getg()
+
+	// The conditions here and in handoffp must agree: if
+	// findrunnable would return a G to run, handoffp must start
+	// an M.
+
+top:
+	_p_ := _g_.m.p.ptr()
+	if sched.gcwaiting != 0 {
+		gcstopm()
+		goto top
+	}
+	if _p_.runSafePointFn != 0 {
+		runSafePointFn()
+	}
+
+	now, pollUntil, _ := checkTimers(_p_, 0)
+
+	if fingwait && fingwake {
+		if gp := wakefing(); gp != nil {
+			ready(gp, 0, true)
+		}
+	}
+	if *cgo_yield != nil {
+		asmcgocall(*cgo_yield, nil)
+	}
+
+	// local runq
+	if gp, inheritTime := runqget(_p_); gp != nil {
+		return gp, inheritTime
+	}
+
+	// global runq
+	if sched.runqsize != 0 {
+		lock(&sched.lock)
+		gp := globrunqget(_p_, 0)
+		unlock(&sched.lock)
+		if gp != nil {
+			return gp, false
+		}
+	}
+
+	// Poll network.
+	// This netpoll is only an optimization before we resort to stealing.
+	// We can safely skip it if there are no waiters or a thread is blocked
+	// in netpoll already. If there is any kind of logical race with that
+	// blocked thread (e.g. it has already returned from netpoll, but does
+	// not set lastpoll yet), this thread will do blocking netpoll below
+	// anyway.
+	if netpollinited() && atomic.Load(&netpollWaiters) > 0 && atomic.Load64(&sched.lastpoll) != 0 {
+		if list := netpoll(0); !list.empty() { // non-blocking
+			gp := list.pop()
+			injectglist(&list)
+			casgstatus(gp, _Gwaiting, _Grunnable)
+			if trace.enabled {
+				traceGoUnpark(gp, 0)
+			}
+			return gp, false
+		}
+	}
+
+	// Spinning Ms: steal work from other Ps.
+	//
+	// Limit the number of spinning Ms to half the number of busy Ps.
+	// This is necessary to prevent excessive CPU consumption when
+	// GOMAXPROCS>>1 but the program parallelism is low.
+	procs := uint32(gomaxprocs)
+	if _g_.m.spinning || 2*atomic.Load(&sched.nmspinning) < procs-atomic.Load(&sched.npidle) {
+		if !_g_.m.spinning {
+			_g_.m.spinning = true
+			atomic.Xadd(&sched.nmspinning, 1)
+		}
+
+		gp, inheritTime, tnow, w, newWork := stealWork(now)
+		now = tnow
+		if gp != nil {
+			// Successfully stole.
+			return gp, inheritTime
+		}
+		if newWork {
+			// There may be new timer or GC work; restart to
+			// discover.
+			goto top
+		}
+		if w != 0 && (pollUntil == 0 || w < pollUntil) {
+			// Earlier timer to wait for.
+			pollUntil = w
+		}
+	}
+
+	// We have nothing to do.
+	//
+	// If we're in the GC mark phase, can safely scan and blacken objects,
+	// and have work to do, run idle-time marking rather than give up the
+	// P.
+	if gcBlackenEnabled != 0 && gcMarkWorkAvailable(_p_) {
+		node := (*gcBgMarkWorkerNode)(gcBgMarkWorkerPool.pop())
+		if node != nil {
+			_p_.gcMarkWorkerMode = gcMarkWorkerIdleMode
+			gp := node.gp.ptr()
+			casgstatus(gp, _Gwaiting, _Grunnable)
+			if trace.enabled {
+				traceGoUnpark(gp, 0)
+			}
+			return gp, false
+		}
+	}
+
+	// wasm only:
+	// If a callback returned and no other goroutine is awake,
+	// then wake event handler goroutine which pauses execution
+	// until a callback was triggered.
+	gp, otherReady := beforeIdle(now, pollUntil)
+	if gp != nil {
+		casgstatus(gp, _Gwaiting, _Grunnable)
+		if trace.enabled {
+			traceGoUnpark(gp, 0)
+		}
+		return gp, false
+	}
+	if otherReady {
+		goto top
+	}
+
+	// Before we drop our P, make a snapshot of the allp slice,
+	// which can change underfoot once we no longer block
+	// safe-points. We don't need to snapshot the contents because
+	// everything up to cap(allp) is immutable.
+	allpSnapshot := allp
+	// Also snapshot masks. Value changes are OK, but we can't allow
+	// len to change out from under us.
+	idlepMaskSnapshot := idlepMask
+	timerpMaskSnapshot := timerpMask
+
+	// return P and block
+	lock(&sched.lock)
+	if sched.gcwaiting != 0 || _p_.runSafePointFn != 0 {
+		unlock(&sched.lock)
+		goto top
+	}
+	if sched.runqsize != 0 {
+		gp := globrunqget(_p_, 0)
+		unlock(&sched.lock)
+		return gp, false
+	}
+	if releasep() != _p_ {
+		throw("findrunnable: wrong p")
+	}
+	pidleput(_p_)
+	unlock(&sched.lock)
+
+	// Delicate dance: thread transitions from spinning to non-spinning
+	// state, potentially concurrently with submission of new work. We must
+	// drop nmspinning first and then check all sources again (with
+	// #StoreLoad memory barrier in between). If we do it the other way
+	// around, another thread can submit work after we've checked all
+	// sources but before we drop nmspinning; as a result nobody will
+	// unpark a thread to run the work.
+	//
+	// This applies to the following sources of work:
+	//
+	// * Goroutines added to a per-P run queue.
+	// * New/modified-earlier timers on a per-P timer heap.
+	// * Idle-priority GC work (barring golang.org/issue/19112).
+	//
+	// If we discover new work below, we need to restore m.spinning as a signal
+	// for resetspinning to unpark a new worker thread (because there can be more
+	// than one starving goroutine). However, if after discovering new work
+	// we also observe no idle Ps it is OK to skip unparking a new worker
+	// thread: the system is fully loaded so no spinning threads are required.
+	// Also see "Worker thread parking/unparking" comment at the top of the file.
+	wasSpinning := _g_.m.spinning
+	if _g_.m.spinning {
+		_g_.m.spinning = false
+		if int32(atomic.Xadd(&sched.nmspinning, -1)) < 0 {
+			throw("findrunnable: negative nmspinning")
+		}
+
+		// Note the for correctness, only the last M transitioning from
+		// spinning to non-spinning must perform these rechecks to
+		// ensure no missed work. We are performing it on every M that
+		// transitions as a conservative change to monitor effects on
+		// latency. See golang.org/issue/43997.
+
+		// Check all runqueues once again.
+		_p_ = checkRunqsNoP(allpSnapshot, idlepMaskSnapshot)
+		if _p_ != nil {
+			acquirep(_p_)
+			_g_.m.spinning = true
+			atomic.Xadd(&sched.nmspinning, 1)
+			goto top
+		}
+
+		// Check for idle-priority GC work again.
+		_p_, gp = checkIdleGCNoP()
+		if _p_ != nil {
+			acquirep(_p_)
+			_g_.m.spinning = true
+			atomic.Xadd(&sched.nmspinning, 1)
+
+			// Run the idle worker.
+			_p_.gcMarkWorkerMode = gcMarkWorkerIdleMode
+			casgstatus(gp, _Gwaiting, _Grunnable)
+			if trace.enabled {
+				traceGoUnpark(gp, 0)
+			}
+			return gp, false
+		}
+
+		// Finally, check for timer creation or expiry concurrently with
+		// transitioning from spinning to non-spinning.
+		//
+		// Note that we cannot use checkTimers here because it calls
+		// adjusttimers which may need to allocate memory, and that isn't
+		// allowed when we don't have an active P.
+		pollUntil = checkTimersNoP(allpSnapshot, timerpMaskSnapshot, pollUntil)
+	}
+
+	// Poll network until next timer.
+	if netpollinited() && (atomic.Load(&netpollWaiters) > 0 || pollUntil != 0) && atomic.Xchg64(&sched.lastpoll, 0) != 0 {
+		atomic.Store64(&sched.pollUntil, uint64(pollUntil))
+		if _g_.m.p != 0 {
+			throw("findrunnable: netpoll with p")
+		}
+		if _g_.m.spinning {
+			throw("findrunnable: netpoll with spinning")
+		}
+		delay := int64(-1)
+		if pollUntil != 0 {
+			if now == 0 {
+				now = nanotime()
+			}
+			delay = pollUntil - now
+			if delay < 0 {
+				delay = 0
+			}
+		}
+		if faketime != 0 {
+			// When using fake time, just poll.
+			delay = 0
+		}
+		list := netpoll(delay) // block until new work is available
+		atomic.Store64(&sched.pollUntil, 0)
+		atomic.Store64(&sched.lastpoll, uint64(nanotime()))
+		if faketime != 0 && list.empty() {
+			// Using fake time and nothing is ready; stop M.
+			// When all M's stop, checkdead will call timejump.
+			stopm()
+			goto top
+		}
+		lock(&sched.lock)
+		_p_ = pidleget()
+		unlock(&sched.lock)
+		if _p_ == nil {
+			injectglist(&list)
+		} else {
+			acquirep(_p_)
+			if !list.empty() {
+				gp := list.pop()
+				injectglist(&list)
+				casgstatus(gp, _Gwaiting, _Grunnable)
+				if trace.enabled {
+					traceGoUnpark(gp, 0)
+				}
+				return gp, false
+			}
+			if wasSpinning {
+				_g_.m.spinning = true
+				atomic.Xadd(&sched.nmspinning, 1)
+			}
+			goto top
+		}
+	} else if pollUntil != 0 && netpollinited() {
+		pollerPollUntil := int64(atomic.Load64(&sched.pollUntil))
+		if pollerPollUntil == 0 || pollerPollUntil > pollUntil {
+			netpollBreak()
+		}
+	}
+	stopm()
+	goto top
+}
+
+// pollWork reports whether there is non-background work this P could
+// be doing. This is a fairly lightweight check to be used for
+// background work loops, like idle GC. It checks a subset of the
+// conditions checked by the actual scheduler.
+func pollWork() bool {
+	if sched.runqsize != 0 {
+		return true
+	}
+	p := getg().m.p.ptr()
+	if !runqempty(p) {
+		return true
+	}
+	if netpollinited() && atomic.Load(&netpollWaiters) > 0 && sched.lastpoll != 0 {
+		if list := netpoll(0); !list.empty() {
+			injectglist(&list)
+			return true
+		}
+	}
+	return false
+}
+
+// stealWork attempts to steal a runnable goroutine or timer from any P.
+//
+// If newWork is true, new work may have been readied.
+//
+// If now is not 0 it is the current time. stealWork returns the passed time or
+// the current time if now was passed as 0.
+func stealWork(now int64) (gp *g, inheritTime bool, rnow, pollUntil int64, newWork bool) {
+	pp := getg().m.p.ptr()
+
+	ranTimer := false
+
+	const stealTries = 4
+	for i := 0; i < stealTries; i++ {
+		stealTimersOrRunNextG := i == stealTries-1
+
+		for enum := stealOrder.start(fastrand()); !enum.done(); enum.next() {
+			if sched.gcwaiting != 0 {
+				// GC work may be available.
+				return nil, false, now, pollUntil, true
+			}
+			p2 := allp[enum.position()]
+			if pp == p2 {
+				continue
+			}
+
+			// Steal timers from p2. This call to checkTimers is the only place
+			// where we might hold a lock on a different P's timers. We do this
+			// once on the last pass before checking runnext because stealing
+			// from the other P's runnext should be the last resort, so if there
+			// are timers to steal do that first.
+			//
+			// We only check timers on one of the stealing iterations because
+			// the time stored in now doesn't change in this loop and checking
+			// the timers for each P more than once with the same value of now
+			// is probably a waste of time.
+			//
+			// timerpMask tells us whether the P may have timers at all. If it
+			// can't, no need to check at all.
+			if stealTimersOrRunNextG && timerpMask.read(enum.position()) {
+				tnow, w, ran := checkTimers(p2, now)
+				now = tnow
+				if w != 0 && (pollUntil == 0 || w < pollUntil) {
+					pollUntil = w
+				}
+				if ran {
+					// Running the timers may have
+					// made an arbitrary number of G's
+					// ready and added them to this P's
+					// local run queue. That invalidates
+					// the assumption of runqsteal
+					// that it always has room to add
+					// stolen G's. So check now if there
+					// is a local G to run.
+					if gp, inheritTime := runqget(pp); gp != nil {
+						return gp, inheritTime, now, pollUntil, ranTimer
+					}
+					ranTimer = true
+				}
+			}
+
+			// Don't bother to attempt to steal if p2 is idle.
+			if !idlepMask.read(enum.position()) {
+				if gp := runqsteal(pp, p2, stealTimersOrRunNextG); gp != nil {
+					return gp, false, now, pollUntil, ranTimer
+				}
+			}
+		}
+	}
+
+	// No goroutines found to steal. Regardless, running a timer may have
+	// made some goroutine ready that we missed. Indicate the next timer to
+	// wait for.
+	return nil, false, now, pollUntil, ranTimer
+}
+
+// Check all Ps for a runnable G to steal.
+//
+// On entry we have no P. If a G is available to steal and a P is available,
+// the P is returned which the caller should acquire and attempt to steal the
+// work to.
+func checkRunqsNoP(allpSnapshot []*p, idlepMaskSnapshot pMask) *p {
+	for id, p2 := range allpSnapshot {
+		if !idlepMaskSnapshot.read(uint32(id)) && !runqempty(p2) {
+			lock(&sched.lock)
+			pp := pidleget()
+			unlock(&sched.lock)
+			if pp != nil {
+				return pp
+			}
+
+			// Can't get a P, don't bother checking remaining Ps.
+			break
+		}
+	}
+
+	return nil
+}
+
+// Check all Ps for a timer expiring sooner than pollUntil.
+//
+// Returns updated pollUntil value.
+func checkTimersNoP(allpSnapshot []*p, timerpMaskSnapshot pMask, pollUntil int64) int64 {
+	for id, p2 := range allpSnapshot {
+		if timerpMaskSnapshot.read(uint32(id)) {
+			w := nobarrierWakeTime(p2)
+			if w != 0 && (pollUntil == 0 || w < pollUntil) {
+				pollUntil = w
+			}
+		}
+	}
+
+	return pollUntil
+}
+
+// Check for idle-priority GC, without a P on entry.
+//
+// If some GC work, a P, and a worker G are all available, the P and G will be
+// returned. The returned P has not been wired yet.
+func checkIdleGCNoP() (*p, *g) {
+	// N.B. Since we have no P, gcBlackenEnabled may change at any time; we
+	// must check again after acquiring a P.
+	if atomic.Load(&gcBlackenEnabled) == 0 {
+		return nil, nil
+	}
+	if !gcMarkWorkAvailable(nil) {
+		return nil, nil
+	}
+
+	// Work is available; we can start an idle GC worker only if there is
+	// an available P and available worker G.
+	//
+	// We can attempt to acquire these in either order, though both have
+	// synchronization concerns (see below). Workers are almost always
+	// available (see comment in findRunnableGCWorker for the one case
+	// there may be none). Since we're slightly less likely to find a P,
+	// check for that first.
+	//
+	// Synchronization: note that we must hold sched.lock until we are
+	// committed to keeping it. Otherwise we cannot put the unnecessary P
+	// back in sched.pidle without performing the full set of idle
+	// transition checks.
+	//
+	// If we were to check gcBgMarkWorkerPool first, we must somehow handle
+	// the assumption in gcControllerState.findRunnableGCWorker that an
+	// empty gcBgMarkWorkerPool is only possible if gcMarkDone is running.
+	lock(&sched.lock)
+	pp := pidleget()
+	if pp == nil {
+		unlock(&sched.lock)
+		return nil, nil
+	}
+
+	// Now that we own a P, gcBlackenEnabled can't change (as it requires
+	// STW).
+	if gcBlackenEnabled == 0 {
+		pidleput(pp)
+		unlock(&sched.lock)
+		return nil, nil
+	}
+
+	node := (*gcBgMarkWorkerNode)(gcBgMarkWorkerPool.pop())
+	if node == nil {
+		pidleput(pp)
+		unlock(&sched.lock)
+		return nil, nil
+	}
+
+	unlock(&sched.lock)
+
+	return pp, node.gp.ptr()
+}
+
+// wakeNetPoller wakes up the thread sleeping in the network poller if it isn't
+// going to wake up before the when argument; or it wakes an idle P to service
+// timers and the network poller if there isn't one already.
+func wakeNetPoller(when int64) {
+	if atomic.Load64(&sched.lastpoll) == 0 {
+		// In findrunnable we ensure that when polling the pollUntil
+		// field is either zero or the time to which the current
+		// poll is expected to run. This can have a spurious wakeup
+		// but should never miss a wakeup.
+		pollerPollUntil := int64(atomic.Load64(&sched.pollUntil))
+		if pollerPollUntil == 0 || pollerPollUntil > when {
+			netpollBreak()
+		}
+	} else {
+		// There are no threads in the network poller, try to get
+		// one there so it can handle new timers.
+		if GOOS != "plan9" { // Temporary workaround - see issue #42303.
+			wakep()
+		}
+	}
+}
+
+func resetspinning() {
+	_g_ := getg()
+	if !_g_.m.spinning {
+		throw("resetspinning: not a spinning m")
+	}
+	_g_.m.spinning = false
+	nmspinning := atomic.Xadd(&sched.nmspinning, -1)
+	if int32(nmspinning) < 0 {
+		throw("findrunnable: negative nmspinning")
+	}
+	// M wakeup policy is deliberately somewhat conservative, so check if we
+	// need to wakeup another P here. See "Worker thread parking/unparking"
+	// comment at the top of the file for details.
+	wakep()
+}
+
+// injectglist adds each runnable G on the list to some run queue,
+// and clears glist. If there is no current P, they are added to the
+// global queue, and up to npidle M's are started to run them.
+// Otherwise, for each idle P, this adds a G to the global queue
+// and starts an M. Any remaining G's are added to the current P's
+// local run queue.
+// This may temporarily acquire sched.lock.
+// Can run concurrently with GC.
+func injectglist(glist *gList) {
+	if glist.empty() {
+		return
+	}
+	if trace.enabled {
+		for gp := glist.head.ptr(); gp != nil; gp = gp.schedlink.ptr() {
+			traceGoUnpark(gp, 0)
+		}
+	}
+
+	// Mark all the goroutines as runnable before we put them
+	// on the run queues.
+	head := glist.head.ptr()
+	var tail *g
+	qsize := 0
+	for gp := head; gp != nil; gp = gp.schedlink.ptr() {
+		tail = gp
+		qsize++
+		casgstatus(gp, _Gwaiting, _Grunnable)
+	}
+
+	// Turn the gList into a gQueue.
+	var q gQueue
+	q.head.set(head)
+	q.tail.set(tail)
+	*glist = gList{}
+
+	startIdle := func(n int) {
+		for ; n != 0 && sched.npidle != 0; n-- {
+			startm(nil, false)
+		}
+	}
+
+	pp := getg().m.p.ptr()
+	if pp == nil {
+		lock(&sched.lock)
+		globrunqputbatch(&q, int32(qsize))
+		unlock(&sched.lock)
+		startIdle(qsize)
+		return
+	}
+
+	npidle := int(atomic.Load(&sched.npidle))
+	var globq gQueue
+	var n int
+	for n = 0; n < npidle && !q.empty(); n++ {
+		g := q.pop()
+		globq.pushBack(g)
+	}
+	if n > 0 {
+		lock(&sched.lock)
+		globrunqputbatch(&globq, int32(n))
+		unlock(&sched.lock)
+		startIdle(n)
+		qsize -= n
+	}
+
+	if !q.empty() {
+		runqputbatch(pp, &q, qsize)
+	}
+}
+
+// One round of scheduler: find a runnable goroutine and execute it.
+// Never returns.
+func schedule() {
+	_g_ := getg()
+
+	if _g_.m.locks != 0 {
+		throw("schedule: holding locks")
+	}
+
+	if _g_.m.lockedg != 0 {
+		stoplockedm()
+		execute(_g_.m.lockedg.ptr(), false) // Never returns.
+	}
+
+	// We should not schedule away from a g that is executing a cgo call,
+	// since the cgo call is using the m's g0 stack.
+	if _g_.m.incgo {
+		throw("schedule: in cgo")
+	}
+
+top:
+	pp := _g_.m.p.ptr()
+	pp.preempt = false
+
+	if sched.gcwaiting != 0 {
+		gcstopm()
+		goto top
+	}
+	if pp.runSafePointFn != 0 {
+		runSafePointFn()
+	}
+
+	// Sanity check: if we are spinning, the run queue should be empty.
+	// Check this before calling checkTimers, as that might call
+	// goready to put a ready goroutine on the local run queue.
+	if _g_.m.spinning && (pp.runnext != 0 || pp.runqhead != pp.runqtail) {
+		throw("schedule: spinning with local work")
+	}
+
+	checkTimers(pp, 0)
+
+	var gp *g
+	var inheritTime bool
+
+	// Normal goroutines will check for need to wakeP in ready,
+	// but GCworkers and tracereaders will not, so the check must
+	// be done here instead.
+	tryWakeP := false
+	if trace.enabled || trace.shutdown {
+		gp = traceReader()
+		if gp != nil {
+			casgstatus(gp, _Gwaiting, _Grunnable)
+			traceGoUnpark(gp, 0)
+			tryWakeP = true
+		}
+	}
+	if gp == nil && gcBlackenEnabled != 0 {
+		gp = gcController.findRunnableGCWorker(_g_.m.p.ptr())
+		if gp != nil {
+			tryWakeP = true
+		}
+	}
+	if gp == nil {
+		// Check the global runnable queue once in a while to ensure fairness.
+		// Otherwise two goroutines can completely occupy the local runqueue
+		// by constantly respawning each other.
+		if _g_.m.p.ptr().schedtick%61 == 0 && sched.runqsize > 0 {
+			lock(&sched.lock)
+			gp = globrunqget(_g_.m.p.ptr(), 1)
+			unlock(&sched.lock)
+		}
+	}
+	if gp == nil {
+		gp, inheritTime = runqget(_g_.m.p.ptr())
+		// We can see gp != nil here even if the M is spinning,
+		// if checkTimers added a local goroutine via goready.
+	}
+	if gp == nil {
+		gp, inheritTime = findrunnable() // blocks until work is available
+	}
+
+	// This thread is going to run a goroutine and is not spinning anymore,
+	// so if it was marked as spinning we need to reset it now and potentially
+	// start a new spinning M.
+	if _g_.m.spinning {
+		resetspinning()
+	}
+
+	if sched.disable.user && !schedEnabled(gp) {
+		// Scheduling of this goroutine is disabled. Put it on
+		// the list of pending runnable goroutines for when we
+		// re-enable user scheduling and look again.
+		lock(&sched.lock)
+		if schedEnabled(gp) {
+			// Something re-enabled scheduling while we
+			// were acquiring the lock.
+			unlock(&sched.lock)
+		} else {
+			sched.disable.runnable.pushBack(gp)
+			sched.disable.n++
+			unlock(&sched.lock)
+			goto top
+		}
+	}
+
+	// If about to schedule a not-normal goroutine (a GCworker or tracereader),
+	// wake a P if there is one.
+	if tryWakeP {
+		wakep()
+	}
+	if gp.lockedm != 0 {
+		// Hands off own p to the locked m,
+		// then blocks waiting for a new p.
+		startlockedm(gp)
+		goto top
+	}
+
+	execute(gp, inheritTime)
+}
+
+// dropg removes the association between m and the current goroutine m->curg (gp for short).
+// Typically a caller sets gp's status away from Grunning and then
+// immediately calls dropg to finish the job. The caller is also responsible
+// for arranging that gp will be restarted using ready at an
+// appropriate time. After calling dropg and arranging for gp to be
+// readied later, the caller can do other work but eventually should
+// call schedule to restart the scheduling of goroutines on this m.
+func dropg() {
+	_g_ := getg()
+
+	setMNoWB(&_g_.m.curg.m, nil)
+	setGNoWB(&_g_.m.curg, nil)
+}
+
+// checkTimers runs any timers for the P that are ready.
+// If now is not 0 it is the current time.
+// It returns the passed time or the current time if now was passed as 0.
+// and the time when the next timer should run or 0 if there is no next timer,
+// and reports whether it ran any timers.
+// If the time when the next timer should run is not 0,
+// it is always larger than the returned time.
+// We pass now in and out to avoid extra calls of nanotime.
+//go:yeswritebarrierrec
+func checkTimers(pp *p, now int64) (rnow, pollUntil int64, ran bool) {
+	// If it's not yet time for the first timer, or the first adjusted
+	// timer, then there is nothing to do.
+	next := int64(atomic.Load64(&pp.timer0When))
+	nextAdj := int64(atomic.Load64(&pp.timerModifiedEarliest))
+	if next == 0 || (nextAdj != 0 && nextAdj < next) {
+		next = nextAdj
+	}
+
+	if next == 0 {
+		// No timers to run or adjust.
+		return now, 0, false
+	}
+
+	if now == 0 {
+		now = nanotime()
+	}
+	if now < next {
+		// Next timer is not ready to run, but keep going
+		// if we would clear deleted timers.
+		// This corresponds to the condition below where
+		// we decide whether to call clearDeletedTimers.
+		if pp != getg().m.p.ptr() || int(atomic.Load(&pp.deletedTimers)) <= int(atomic.Load(&pp.numTimers)/4) {
+			return now, next, false
+		}
+	}
+
+	lock(&pp.timersLock)
+
+	if len(pp.timers) > 0 {
+		adjusttimers(pp, now)
+		for len(pp.timers) > 0 {
+			// Note that runtimer may temporarily unlock
+			// pp.timersLock.
+			if tw := runtimer(pp, now); tw != 0 {
+				if tw > 0 {
+					pollUntil = tw
+				}
+				break
+			}
+			ran = true
+		}
+	}
+
+	// If this is the local P, and there are a lot of deleted timers,
+	// clear them out. We only do this for the local P to reduce
+	// lock contention on timersLock.
+	if pp == getg().m.p.ptr() && int(atomic.Load(&pp.deletedTimers)) > len(pp.timers)/4 {
+		clearDeletedTimers(pp)
+	}
+
+	unlock(&pp.timersLock)
+
+	return now, pollUntil, ran
+}
+
+func parkunlock_c(gp *g, lock unsafe.Pointer) bool {
+	unlock((*mutex)(lock))
+	return true
+}
+
+// park continuation on g0.
+func park_m(gp *g) {
+	_g_ := getg()
+
+	if trace.enabled {
+		traceGoPark(_g_.m.waittraceev, _g_.m.waittraceskip)
+	}
+
+	casgstatus(gp, _Grunning, _Gwaiting)
+	dropg()
+
+	if fn := _g_.m.waitunlockf; fn != nil {
+		ok := fn(gp, _g_.m.waitlock)
+		_g_.m.waitunlockf = nil
+		_g_.m.waitlock = nil
+		if !ok {
+			if trace.enabled {
+				traceGoUnpark(gp, 2)
+			}
+			casgstatus(gp, _Gwaiting, _Grunnable)
+			execute(gp, true) // Schedule it back, never returns.
+		}
+	}
+	schedule()
+}
+
+func goschedImpl(gp *g) {
+	status := readgstatus(gp)
+	if status&^_Gscan != _Grunning {
+		dumpgstatus(gp)
+		throw("bad g status")
+	}
+	casgstatus(gp, _Grunning, _Grunnable)
+	dropg()
+	lock(&sched.lock)
+	globrunqput(gp)
+	unlock(&sched.lock)
+
+	schedule()
+}
+
+// Gosched continuation on g0.
+func gosched_m(gp *g) {
+	if trace.enabled {
+		traceGoSched()
+	}
+	goschedImpl(gp)
+}
+
+// goschedguarded is a forbidden-states-avoided version of gosched_m
+func goschedguarded_m(gp *g) {
+
+	if !canPreemptM(gp.m) {
+		gogo(&gp.sched) // never return
+	}
+
+	if trace.enabled {
+		traceGoSched()
+	}
+	goschedImpl(gp)
+}
+
+func gopreempt_m(gp *g) {
+	if trace.enabled {
+		traceGoPreempt()
+	}
+	goschedImpl(gp)
+}
+
+// preemptPark parks gp and puts it in _Gpreempted.
+//
+//go:systemstack
+func preemptPark(gp *g) {
+	if trace.enabled {
+		traceGoPark(traceEvGoBlock, 0)
+	}
+	status := readgstatus(gp)
+	if status&^_Gscan != _Grunning {
+		dumpgstatus(gp)
+		throw("bad g status")
+	}
+	gp.waitreason = waitReasonPreempted
+
+	if gp.asyncSafePoint {
+		// Double-check that async preemption does not
+		// happen in SPWRITE assembly functions.
+		// isAsyncSafePoint must exclude this case.
+		f := findfunc(gp.sched.pc)
+		if !f.valid() {
+			throw("preempt at unknown pc")
+		}
+		if f.flag&funcFlag_SPWRITE != 0 {
+			println("runtime: unexpected SPWRITE function", funcname(f), "in async preempt")
+			throw("preempt SPWRITE")
+		}
+	}
+
+	// Transition from _Grunning to _Gscan|_Gpreempted. We can't
+	// be in _Grunning when we dropg because then we'd be running
+	// without an M, but the moment we're in _Gpreempted,
+	// something could claim this G before we've fully cleaned it
+	// up. Hence, we set the scan bit to lock down further
+	// transitions until we can dropg.
+	casGToPreemptScan(gp, _Grunning, _Gscan|_Gpreempted)
+	dropg()
+	casfrom_Gscanstatus(gp, _Gscan|_Gpreempted, _Gpreempted)
+	schedule()
+}
+
+// goyield is like Gosched, but it:
+// - emits a GoPreempt trace event instead of a GoSched trace event
+// - puts the current G on the runq of the current P instead of the globrunq
+func goyield() {
+	checkTimeouts()
+	mcall(goyield_m)
+}
+
+func goyield_m(gp *g) {
+	if trace.enabled {
+		traceGoPreempt()
+	}
+	pp := gp.m.p.ptr()
+	casgstatus(gp, _Grunning, _Grunnable)
+	dropg()
+	runqput(pp, gp, false)
+	schedule()
+}
+
+// Finishes execution of the current goroutine.
+func goexit1() {
+	if raceenabled {
+		racegoend()
+	}
+	if trace.enabled {
+		traceGoEnd()
+	}
+	mcall(goexit0)
+}
+
+// goexit continuation on g0.
+func goexit0(gp *g) {
+	_g_ := getg()
+	_p_ := _g_.m.p.ptr()
+
+	casgstatus(gp, _Grunning, _Gdead)
+	gcController.addScannableStack(_p_, -int64(gp.stack.hi-gp.stack.lo))
+	if isSystemGoroutine(gp, false) {
+		atomic.Xadd(&sched.ngsys, -1)
+	}
+	gp.m = nil
+	locked := gp.lockedm != 0
+	gp.lockedm = 0
+	_g_.m.lockedg = 0
+	gp.preemptStop = false
+	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
+
+	if gcBlackenEnabled != 0 && gp.gcAssistBytes > 0 {
+		// Flush assist credit to the global pool. This gives
+		// better information to pacing if the application is
+		// rapidly creating an exiting goroutines.
+		assistWorkPerByte := gcController.assistWorkPerByte.Load()
+		scanCredit := int64(assistWorkPerByte * float64(gp.gcAssistBytes))
+		atomic.Xaddint64(&gcController.bgScanCredit, scanCredit)
+		gp.gcAssistBytes = 0
+	}
+
+	dropg()
+
+	if GOARCH == "wasm" { // no threads yet on wasm
+		gfput(_p_, gp)
+		schedule() // never returns
+	}
+
+	if _g_.m.lockedInt != 0 {
+		print("invalid m->lockedInt = ", _g_.m.lockedInt, "\n")
+		throw("internal lockOSThread error")
+	}
+	gfput(_p_, gp)
+	if locked {
+		// The goroutine may have locked this thread because
+		// it put it in an unusual kernel state. Kill it
+		// rather than returning it to the thread pool.
+
+		// Return to mstart, which will release the P and exit
+		// the thread.
+		if GOOS != "plan9" { // See golang.org/issue/22227.
+			gogo(&_g_.m.g0.sched)
+		} else {
+			// Clear lockedExt on plan9 since we may end up re-using
+			// this thread.
+			_g_.m.lockedExt = 0
+		}
+	}
+	schedule()
+}
+
+// save updates getg().sched to refer to pc and sp so that a following
+// gogo will restore pc and sp.
+//
+// save must not have write barriers because invoking a write barrier
+// can clobber getg().sched.
+//
+//go:nosplit
+//go:nowritebarrierrec
+func save(pc, sp uintptr) {
+	_g_ := getg()
+
+	if _g_ == _g_.m.g0 || _g_ == _g_.m.gsignal {
+		// m.g0.sched is special and must describe the context
+		// for exiting the thread. mstart1 writes to it directly.
+		// m.gsignal.sched should not be used at all.
+		// This check makes sure save calls do not accidentally
+		// run in contexts where they'd write to system g's.
+		throw("save on system g not allowed")
+	}
+
+	_g_.sched.pc = pc
+	_g_.sched.sp = sp
+	_g_.sched.lr = 0
+	_g_.sched.ret = 0
+	// We need to ensure ctxt is zero, but can't have a write
+	// barrier here. However, it should always already be zero.
+	// Assert that.
+	if _g_.sched.ctxt != nil {
+		badctxt()
+	}
+}
+
+// The goroutine g is about to enter a system call.
+// Record that it's not using the cpu anymore.
+// This is called only from the go syscall library and cgocall,
+// not from the low-level system calls used by the runtime.
+//
+// Entersyscall cannot split the stack: the save must
+// make g->sched refer to the caller's stack segment, because
+// entersyscall is going to return immediately after.
+//
+// Nothing entersyscall calls can split the stack either.
+// We cannot safely move the stack during an active call to syscall,
+// because we do not know which of the uintptr arguments are
+// really pointers (back into the stack).
+// In practice, this means that we make the fast path run through
+// entersyscall doing no-split things, and the slow path has to use systemstack
+// to run bigger things on the system stack.
+//
+// reentersyscall is the entry point used by cgo callbacks, where explicitly
+// saved SP and PC are restored. This is needed when exitsyscall will be called
+// from a function further up in the call stack than the parent, as g->syscallsp
+// must always point to a valid stack frame. entersyscall below is the normal
+// entry point for syscalls, which obtains the SP and PC from the caller.
+//
+// Syscall tracing:
+// At the start of a syscall we emit traceGoSysCall to capture the stack trace.
+// If the syscall does not block, that is it, we do not emit any other events.
+// If the syscall blocks (that is, P is retaken), retaker emits traceGoSysBlock;
+// when syscall returns we emit traceGoSysExit and when the goroutine starts running
+// (potentially instantly, if exitsyscallfast returns true) we emit traceGoStart.
+// To ensure that traceGoSysExit is emitted strictly after traceGoSysBlock,
+// we remember current value of syscalltick in m (_g_.m.syscalltick = _g_.m.p.ptr().syscalltick),
+// whoever emits traceGoSysBlock increments p.syscalltick afterwards;
+// and we wait for the increment before emitting traceGoSysExit.
+// Note that the increment is done even if tracing is not enabled,
+// because tracing can be enabled in the middle of syscall. We don't want the wait to hang.
+//
+//go:nosplit
+func reentersyscall(pc, sp uintptr) {
+	_g_ := getg()
+
+	// Disable preemption because during this function g is in Gsyscall status,
+	// but can have inconsistent g->sched, do not let GC observe it.
+	_g_.m.locks++
+
+	// Entersyscall must not call any function that might split/grow the stack.
+	// (See details in comment above.)
+	// Catch calls that might, by replacing the stack guard with something that
+	// will trip any stack check and leaving a flag to tell newstack to die.
+	_g_.stackguard0 = stackPreempt
+	_g_.throwsplit = true
+
+	// Leave SP around for GC and traceback.
+	save(pc, sp)
+	_g_.syscallsp = sp
+	_g_.syscallpc = pc
+	casgstatus(_g_, _Grunning, _Gsyscall)
+	if _g_.syscallsp < _g_.stack.lo || _g_.stack.hi < _g_.syscallsp {
+		systemstack(func() {
+			print("entersyscall inconsistent ", hex(_g_.syscallsp), " [", hex(_g_.stack.lo), ",", hex(_g_.stack.hi), "]\n")
+			throw("entersyscall")
+		})
+	}
+
+	if trace.enabled {
+		systemstack(traceGoSysCall)
+		// systemstack itself clobbers g.sched.{pc,sp} and we might
+		// need them later when the G is genuinely blocked in a
+		// syscall
+		save(pc, sp)
+	}
+
+	if atomic.Load(&sched.sysmonwait) != 0 {
+		systemstack(entersyscall_sysmon)
+		save(pc, sp)
+	}
+
+	if _g_.m.p.ptr().runSafePointFn != 0 {
+		// runSafePointFn may stack split if run on this stack
+		systemstack(runSafePointFn)
+		save(pc, sp)
+	}
+
+	_g_.m.syscalltick = _g_.m.p.ptr().syscalltick
+	_g_.sysblocktraced = true
+	pp := _g_.m.p.ptr()
+	pp.m = 0
+	_g_.m.oldp.set(pp)
+	_g_.m.p = 0
+	atomic.Store(&pp.status, _Psyscall)
+	if sched.gcwaiting != 0 {
+		systemstack(entersyscall_gcwait)
+		save(pc, sp)
+	}
+
+	_g_.m.locks--
+}
+
+// Standard syscall entry used by the go syscall library and normal cgo calls.
+//
+// This is exported via linkname to assembly in the syscall package.
+//
+//go:nosplit
+//go:linkname entersyscall
+func entersyscall() {
+	reentersyscall(getcallerpc(), getcallersp())
+}
+
+func entersyscall_sysmon() {
+	lock(&sched.lock)
+	if atomic.Load(&sched.sysmonwait) != 0 {
+		atomic.Store(&sched.sysmonwait, 0)
+		notewakeup(&sched.sysmonnote)
+	}
+	unlock(&sched.lock)
+}
+
+func entersyscall_gcwait() {
+	_g_ := getg()
+	_p_ := _g_.m.oldp.ptr()
+
+	lock(&sched.lock)
+	if sched.stopwait > 0 && atomic.Cas(&_p_.status, _Psyscall, _Pgcstop) {
+		if trace.enabled {
+			traceGoSysBlock(_p_)
+			traceProcStop(_p_)
+		}
+		_p_.syscalltick++
+		if sched.stopwait--; sched.stopwait == 0 {
+			notewakeup(&sched.stopnote)
+		}
+	}
+	unlock(&sched.lock)
+}
+
+// The same as entersyscall(), but with a hint that the syscall is blocking.
+//go:nosplit
+func entersyscallblock() {
+	_g_ := getg()
+
+	_g_.m.locks++ // see comment in entersyscall
+	_g_.throwsplit = true
+	_g_.stackguard0 = stackPreempt // see comment in entersyscall
+	_g_.m.syscalltick = _g_.m.p.ptr().syscalltick
+	_g_.sysblocktraced = true
+	_g_.m.p.ptr().syscalltick++
+
+	// Leave SP around for GC and traceback.
+	pc := getcallerpc()
+	sp := getcallersp()
+	save(pc, sp)
+	_g_.syscallsp = _g_.sched.sp
+	_g_.syscallpc = _g_.sched.pc
+	if _g_.syscallsp < _g_.stack.lo || _g_.stack.hi < _g_.syscallsp {
+		sp1 := sp
+		sp2 := _g_.sched.sp
+		sp3 := _g_.syscallsp
+		systemstack(func() {
+			print("entersyscallblock inconsistent ", hex(sp1), " ", hex(sp2), " ", hex(sp3), " [", hex(_g_.stack.lo), ",", hex(_g_.stack.hi), "]\n")
+			throw("entersyscallblock")
+		})
+	}
+	casgstatus(_g_, _Grunning, _Gsyscall)
+	if _g_.syscallsp < _g_.stack.lo || _g_.stack.hi < _g_.syscallsp {
+		systemstack(func() {
+			print("entersyscallblock inconsistent ", hex(sp), " ", hex(_g_.sched.sp), " ", hex(_g_.syscallsp), " [", hex(_g_.stack.lo), ",", hex(_g_.stack.hi), "]\n")
+			throw("entersyscallblock")
+		})
+	}
+
+	systemstack(entersyscallblock_handoff)
+
+	// Resave for traceback during blocked call.
+	save(getcallerpc(), getcallersp())
+
+	_g_.m.locks--
+}
+
+func entersyscallblock_handoff() {
+	if trace.enabled {
+		traceGoSysCall()
+		traceGoSysBlock(getg().m.p.ptr())
+	}
+	handoffp(releasep())
+}
+
+// The goroutine g exited its system call.
+// Arrange for it to run on a cpu again.
+// This is called only from the go syscall library, not
+// from the low-level system calls used by the runtime.
+//
+// Write barriers are not allowed because our P may have been stolen.
+//
+// This is exported via linkname to assembly in the syscall package.
+//
+//go:nosplit
+//go:nowritebarrierrec
+//go:linkname exitsyscall
+func exitsyscall() {
+	_g_ := getg()
+
+	_g_.m.locks++ // see comment in entersyscall
+	if getcallersp() > _g_.syscallsp {
+		throw("exitsyscall: syscall frame is no longer valid")
+	}
+
+	_g_.waitsince = 0
+	oldp := _g_.m.oldp.ptr()
+	_g_.m.oldp = 0
+	if exitsyscallfast(oldp) {
+		if trace.enabled {
+			if oldp != _g_.m.p.ptr() || _g_.m.syscalltick != _g_.m.p.ptr().syscalltick {
+				systemstack(traceGoStart)
+			}
+		}
+		// There's a cpu for us, so we can run.
+		_g_.m.p.ptr().syscalltick++
+		// We need to cas the status and scan before resuming...
+		casgstatus(_g_, _Gsyscall, _Grunning)
+
+		// Garbage collector isn't running (since we are),
+		// so okay to clear syscallsp.
+		_g_.syscallsp = 0
+		_g_.m.locks--
+		if _g_.preempt {
+			// restore the preemption request in case we've cleared it in newstack
+			_g_.stackguard0 = stackPreempt
+		} else {
+			// otherwise restore the real _StackGuard, we've spoiled it in entersyscall/entersyscallblock
+			_g_.stackguard0 = _g_.stack.lo + _StackGuard
+		}
+		_g_.throwsplit = false
+
+		if sched.disable.user && !schedEnabled(_g_) {
+			// Scheduling of this goroutine is disabled.
+			Gosched()
+		}
+
+		return
+	}
+
+	_g_.sysexitticks = 0
+	if trace.enabled {
+		// Wait till traceGoSysBlock event is emitted.
+		// This ensures consistency of the trace (the goroutine is started after it is blocked).
+		for oldp != nil && oldp.syscalltick == _g_.m.syscalltick {
+			osyield()
+		}
+		// We can't trace syscall exit right now because we don't have a P.
+		// Tracing code can invoke write barriers that cannot run without a P.
+		// So instead we remember the syscall exit time and emit the event
+		// in execute when we have a P.
+		_g_.sysexitticks = cputicks()
+	}
+
+	_g_.m.locks--
+
+	// Call the scheduler.
+	mcall(exitsyscall0)
+
+	// Scheduler returned, so we're allowed to run now.
+	// Delete the syscallsp information that we left for
+	// the garbage collector during the system call.
+	// Must wait until now because until gosched returns
+	// we don't know for sure that the garbage collector
+	// is not running.
+	_g_.syscallsp = 0
+	_g_.m.p.ptr().syscalltick++
+	_g_.throwsplit = false
+}
+
+//go:nosplit
+func exitsyscallfast(oldp *p) bool {
+	_g_ := getg()
+
+	// Freezetheworld sets stopwait but does not retake P's.
+	if sched.stopwait == freezeStopWait {
+		return false
+	}
+
+	// Try to re-acquire the last P.
+	if oldp != nil && oldp.status == _Psyscall && atomic.Cas(&oldp.status, _Psyscall, _Pidle) {
+		// There's a cpu for us, so we can run.
+		wirep(oldp)
+		exitsyscallfast_reacquired()
+		return true
+	}
+
+	// Try to get any other idle P.
+	if sched.pidle != 0 {
+		var ok bool
+		systemstack(func() {
+			ok = exitsyscallfast_pidle()
+			if ok && trace.enabled {
+				if oldp != nil {
+					// Wait till traceGoSysBlock event is emitted.
+					// This ensures consistency of the trace (the goroutine is started after it is blocked).
+					for oldp.syscalltick == _g_.m.syscalltick {
+						osyield()
+					}
+				}
+				traceGoSysExit(0)
+			}
+		})
+		if ok {
+			return true
+		}
+	}
+	return false
+}
+
+// exitsyscallfast_reacquired is the exitsyscall path on which this G
+// has successfully reacquired the P it was running on before the
+// syscall.
+//
+//go:nosplit
+func exitsyscallfast_reacquired() {
+	_g_ := getg()
+	if _g_.m.syscalltick != _g_.m.p.ptr().syscalltick {
+		if trace.enabled {
+			// The p was retaken and then enter into syscall again (since _g_.m.syscalltick has changed).
+			// traceGoSysBlock for this syscall was already emitted,
+			// but here we effectively retake the p from the new syscall running on the same p.
+			systemstack(func() {
+				// Denote blocking of the new syscall.
+				traceGoSysBlock(_g_.m.p.ptr())
+				// Denote completion of the current syscall.
+				traceGoSysExit(0)
+			})
+		}
+		_g_.m.p.ptr().syscalltick++
+	}
+}
+
+func exitsyscallfast_pidle() bool {
+	lock(&sched.lock)
+	_p_ := pidleget()
+	if _p_ != nil && atomic.Load(&sched.sysmonwait) != 0 {
+		atomic.Store(&sched.sysmonwait, 0)
+		notewakeup(&sched.sysmonnote)
+	}
+	unlock(&sched.lock)
+	if _p_ != nil {
+		acquirep(_p_)
+		return true
+	}
+	return false
+}
+
+// exitsyscall slow path on g0.
+// Failed to acquire P, enqueue gp as runnable.
+//
+// Called via mcall, so gp is the calling g from this M.
+//
+//go:nowritebarrierrec
+func exitsyscall0(gp *g) {
+	casgstatus(gp, _Gsyscall, _Grunnable)
+	dropg()
+	lock(&sched.lock)
+	var _p_ *p
+	if schedEnabled(gp) {
+		_p_ = pidleget()
+	}
+	var locked bool
+	if _p_ == nil {
+		globrunqput(gp)
+
+		// Below, we stoplockedm if gp is locked. globrunqput releases
+		// ownership of gp, so we must check if gp is locked prior to
+		// committing the release by unlocking sched.lock, otherwise we
+		// could race with another M transitioning gp from unlocked to
+		// locked.
+		locked = gp.lockedm != 0
+	} else if atomic.Load(&sched.sysmonwait) != 0 {
+		atomic.Store(&sched.sysmonwait, 0)
+		notewakeup(&sched.sysmonnote)
+	}
+	unlock(&sched.lock)
+	if _p_ != nil {
+		acquirep(_p_)
+		execute(gp, false) // Never returns.
+	}
+	if locked {
+		// Wait until another thread schedules gp and so m again.
+		//
+		// N.B. lockedm must be this M, as this g was running on this M
+		// before entersyscall.
+		stoplockedm()
+		execute(gp, false) // Never returns.
+	}
+	stopm()
+	schedule() // Never returns.
+}
+
+// Called from syscall package before fork.
+//go:linkname syscall_runtime_BeforeFork syscall.runtime_BeforeFork
+//go:nosplit
+func syscall_runtime_BeforeFork() {
+	gp := getg().m.curg
+
+	// Block signals during a fork, so that the child does not run
+	// a signal handler before exec if a signal is sent to the process
+	// group. See issue #18600.
+	gp.m.locks++
+	sigsave(&gp.m.sigmask)
+	sigblock(false)
+
+	// This function is called before fork in syscall package.
+	// Code between fork and exec must not allocate memory nor even try to grow stack.
+	// Here we spoil g->_StackGuard to reliably detect any attempts to grow stack.
+	// runtime_AfterFork will undo this in parent process, but not in child.
+	gp.stackguard0 = stackFork
+}
+
+// Called from syscall package after fork in parent.
+//go:linkname syscall_runtime_AfterFork syscall.runtime_AfterFork
+//go:nosplit
+func syscall_runtime_AfterFork() {
+	gp := getg().m.curg
+
+	// See the comments in beforefork.
+	gp.stackguard0 = gp.stack.lo + _StackGuard
+
+	msigrestore(gp.m.sigmask)
+
+	gp.m.locks--
+}
+
+// inForkedChild is true while manipulating signals in the child process.
+// This is used to avoid calling libc functions in case we are using vfork.
+var inForkedChild bool
+
+// Called from syscall package after fork in child.
+// It resets non-sigignored signals to the default handler, and
+// restores the signal mask in preparation for the exec.
+//
+// Because this might be called during a vfork, and therefore may be
+// temporarily sharing address space with the parent process, this must
+// not change any global variables or calling into C code that may do so.
+//
+//go:linkname syscall_runtime_AfterForkInChild syscall.runtime_AfterForkInChild
+//go:nosplit
+//go:nowritebarrierrec
+func syscall_runtime_AfterForkInChild() {
+	// It's OK to change the global variable inForkedChild here
+	// because we are going to change it back. There is no race here,
+	// because if we are sharing address space with the parent process,
+	// then the parent process can not be running concurrently.
+	inForkedChild = true
+
+	clearSignalHandlers()
+
+	// When we are the child we are the only thread running,
+	// so we know that nothing else has changed gp.m.sigmask.
+	msigrestore(getg().m.sigmask)
+
+	inForkedChild = false
+}
+
+// pendingPreemptSignals is the number of preemption signals
+// that have been sent but not received. This is only used on Darwin.
+// For #41702.
+var pendingPreemptSignals uint32
+
+// Called from syscall package before Exec.
+//go:linkname syscall_runtime_BeforeExec syscall.runtime_BeforeExec
+func syscall_runtime_BeforeExec() {
+	// Prevent thread creation during exec.
+	execLock.lock()
+
+	// On Darwin, wait for all pending preemption signals to
+	// be received. See issue #41702.
+	if GOOS == "darwin" || GOOS == "ios" {
+		for int32(atomic.Load(&pendingPreemptSignals)) > 0 {
+			osyield()
+		}
+	}
+}
+
+// Called from syscall package after Exec.
+//go:linkname syscall_runtime_AfterExec syscall.runtime_AfterExec
+func syscall_runtime_AfterExec() {
+	execLock.unlock()
+}
+
+// Allocate a new g, with a stack big enough for stacksize bytes.
+func malg(stacksize int32) *g {
+	newg := new(g)
+	if stacksize >= 0 {
+		stacksize = round2(_StackSystem + stacksize)
+		systemstack(func() {
+			newg.stack = stackalloc(uint32(stacksize))
+		})
+		newg.stackguard0 = newg.stack.lo + _StackGuard
+		newg.stackguard1 = ^uintptr(0)
+		// Clear the bottom word of the stack. We record g
+		// there on gsignal stack during VDSO on ARM and ARM64.
+		*(*uintptr)(unsafe.Pointer(newg.stack.lo)) = 0
+	}
+	return newg
+}
+
+// Create a new g running fn.
+// Put it on the queue of g's waiting to run.
+// The compiler turns a go statement into a call to this.
+func newproc(fn *funcval) {
+	gp := getg()
+	pc := getcallerpc()
+	systemstack(func() {
+		newg := newproc1(fn, gp, pc)
+
+		_p_ := getg().m.p.ptr()
+		runqput(_p_, newg, true)
+
+		if mainStarted {
+			wakep()
+		}
+	})
+}
+
+// Create a new g in state _Grunnable, starting at fn. callerpc is the
+// address of the go statement that created this. The caller is responsible
+// for adding the new g to the scheduler.
+func newproc1(fn *funcval, callergp *g, callerpc uintptr) *g {
+	_g_ := getg()
+
+	if fn == nil {
+		_g_.m.throwing = -1 // do not dump full stacks
+		throw("go of nil func value")
+	}
+	acquirem() // disable preemption because it can be holding p in a local var
+
+	_p_ := _g_.m.p.ptr()
+	newg := gfget(_p_)
+	if newg == nil {
+		newg = malg(_StackMin)
+		casgstatus(newg, _Gidle, _Gdead)
+		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 := uintptr(4*goarch.PtrSize + sys.MinFrameSize) // extra space in case of reads slightly beyond frame
+	totalSize = alignUp(totalSize, sys.StackAlign)
+	sp := newg.stack.hi - totalSize
+	spArg := sp
+	if usesLR {
+		// caller's LR
+		*(*uintptr)(unsafe.Pointer(sp)) = 0
+		prepGoExitFrame(sp)
+		spArg += sys.MinFrameSize
+	}
+
+	memclrNoHeapPointers(unsafe.Pointer(&newg.sched), unsafe.Sizeof(newg.sched))
+	newg.sched.sp = sp
+	newg.stktopsp = sp
+	newg.sched.pc = abi.FuncPCABI0(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.ancestors = saveAncestors(callergp)
+	newg.startpc = fn.fn
+	if isSystemGoroutine(newg, false) {
+		atomic.Xadd(&sched.ngsys, +1)
+	} else {
+		// Only user goroutines inherit pprof labels.
+		if _g_.m.curg != nil {
+			newg.labels = _g_.m.curg.labels
+		}
+	}
+	// Track initial transition?
+	newg.trackingSeq = uint8(fastrand())
+	if newg.trackingSeq%gTrackingPeriod == 0 {
+		newg.tracking = true
+	}
+	casgstatus(newg, _Gdead, _Grunnable)
+	gcController.addScannableStack(_p_, int64(newg.stack.hi-newg.stack.lo))
+
+	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)
+	}
+	releasem(_g_.m)
+
+	return newg
+}
+
+// saveAncestors copies previous ancestors of the given caller g and
+// includes infor for the current caller into a new set of tracebacks for
+// a g being created.
+func saveAncestors(callergp *g) *[]ancestorInfo {
+	// Copy all prior info, except for the root goroutine (goid 0).
+	if debug.tracebackancestors <= 0 || callergp.goid == 0 {
+		return nil
+	}
+	var callerAncestors []ancestorInfo
+	if callergp.ancestors != nil {
+		callerAncestors = *callergp.ancestors
+	}
+	n := int32(len(callerAncestors)) + 1
+	if n > debug.tracebackancestors {
+		n = debug.tracebackancestors
+	}
+	ancestors := make([]ancestorInfo, n)
+	copy(ancestors[1:], callerAncestors)
+
+	var pcs [_TracebackMaxFrames]uintptr
+	npcs := gcallers(callergp, 0, pcs[:])
+	ipcs := make([]uintptr, npcs)
+	copy(ipcs, pcs[:])
+	ancestors[0] = ancestorInfo{
+		pcs:  ipcs,
+		goid: callergp.goid,
+		gopc: callergp.gopc,
+	}
+
+	ancestorsp := new([]ancestorInfo)
+	*ancestorsp = ancestors
+	return ancestorsp
+}
+
+// Put on gfree list.
+// If local list is too long, transfer a batch to the global list.
+func gfput(_p_ *p, gp *g) {
+	if readgstatus(gp) != _Gdead {
+		throw("gfput: bad status (not Gdead)")
+	}
+
+	stksize := gp.stack.hi - gp.stack.lo
+
+	if stksize != _FixedStack {
+		// non-standard stack size - free it.
+		stackfree(gp.stack)
+		gp.stack.lo = 0
+		gp.stack.hi = 0
+		gp.stackguard0 = 0
+	}
+
+	_p_.gFree.push(gp)
+	_p_.gFree.n++
+	if _p_.gFree.n >= 64 {
+		var (
+			inc      int32
+			stackQ   gQueue
+			noStackQ gQueue
+		)
+		for _p_.gFree.n >= 32 {
+			gp = _p_.gFree.pop()
+			_p_.gFree.n--
+			if gp.stack.lo == 0 {
+				noStackQ.push(gp)
+			} else {
+				stackQ.push(gp)
+			}
+			inc++
+		}
+		lock(&sched.gFree.lock)
+		sched.gFree.noStack.pushAll(noStackQ)
+		sched.gFree.stack.pushAll(stackQ)
+		sched.gFree.n += inc
+		unlock(&sched.gFree.lock)
+	}
+}
+
+// Get from gfree list.
+// If local list is empty, grab a batch from global list.
+func gfget(_p_ *p) *g {
+retry:
+	if _p_.gFree.empty() && (!sched.gFree.stack.empty() || !sched.gFree.noStack.empty()) {
+		lock(&sched.gFree.lock)
+		// Move a batch of free Gs to the P.
+		for _p_.gFree.n < 32 {
+			// Prefer Gs with stacks.
+			gp := sched.gFree.stack.pop()
+			if gp == nil {
+				gp = sched.gFree.noStack.pop()
+				if gp == nil {
+					break
+				}
+			}
+			sched.gFree.n--
+			_p_.gFree.push(gp)
+			_p_.gFree.n++
+		}
+		unlock(&sched.gFree.lock)
+		goto retry
+	}
+	gp := _p_.gFree.pop()
+	if gp == nil {
+		return nil
+	}
+	_p_.gFree.n--
+	if gp.stack.lo == 0 {
+		// Stack was deallocated in gfput. Allocate a new one.
+		systemstack(func() {
+			gp.stack = stackalloc(_FixedStack)
+		})
+		gp.stackguard0 = gp.stack.lo + _StackGuard
+	} else {
+		if raceenabled {
+			racemalloc(unsafe.Pointer(gp.stack.lo), gp.stack.hi-gp.stack.lo)
+		}
+		if msanenabled {
+			msanmalloc(unsafe.Pointer(gp.stack.lo), gp.stack.hi-gp.stack.lo)
+		}
+		if asanenabled {
+			asanunpoison(unsafe.Pointer(gp.stack.lo), gp.stack.hi-gp.stack.lo)
+		}
+	}
+	return gp
+}
+
+// Purge all cached G's from gfree list to the global list.
+func gfpurge(_p_ *p) {
+	var (
+		inc      int32
+		stackQ   gQueue
+		noStackQ gQueue
+	)
+	for !_p_.gFree.empty() {
+		gp := _p_.gFree.pop()
+		_p_.gFree.n--
+		if gp.stack.lo == 0 {
+			noStackQ.push(gp)
+		} else {
+			stackQ.push(gp)
+		}
+		inc++
+	}
+	lock(&sched.gFree.lock)
+	sched.gFree.noStack.pushAll(noStackQ)
+	sched.gFree.stack.pushAll(stackQ)
+	sched.gFree.n += inc
+	unlock(&sched.gFree.lock)
+}
+
+// Breakpoint executes a breakpoint trap.
+func Breakpoint() {
+	breakpoint()
+}
+
+// dolockOSThread is called by LockOSThread and lockOSThread below
+// after they modify m.locked. Do not allow preemption during this call,
+// or else the m might be different in this function than in the caller.
+//go:nosplit
+func dolockOSThread() {
+	if GOARCH == "wasm" {
+		return // no threads on wasm yet
+	}
+	_g_ := getg()
+	_g_.m.lockedg.set(_g_)
+	_g_.lockedm.set(_g_.m)
+}
+
+//go:nosplit
+
+// LockOSThread wires the calling goroutine to its current operating system thread.
+// The calling goroutine will always execute in that thread,
+// and no other goroutine will execute in it,
+// until the calling goroutine has made as many calls to
+// UnlockOSThread as to LockOSThread.
+// If the calling goroutine exits without unlocking the thread,
+// the thread will be terminated.
+//
+// All init functions are run on the startup thread. Calling LockOSThread
+// from an init function will cause the main function to be invoked on
+// that thread.
+//
+// A goroutine should call LockOSThread before calling OS services or
+// non-Go library functions that depend on per-thread state.
+func LockOSThread() {
+	if atomic.Load(&newmHandoff.haveTemplateThread) == 0 && GOOS != "plan9" {
+		// If we need to start a new thread from the locked
+		// thread, we need the template thread. Start it now
+		// while we're in a known-good state.
+		startTemplateThread()
+	}
+	_g_ := getg()
+	_g_.m.lockedExt++
+	if _g_.m.lockedExt == 0 {
+		_g_.m.lockedExt--
+		panic("LockOSThread nesting overflow")
+	}
+	dolockOSThread()
+}
+
+//go:nosplit
+func lockOSThread() {
+	getg().m.lockedInt++
+	dolockOSThread()
+}
+
+// dounlockOSThread is called by UnlockOSThread and unlockOSThread below
+// after they update m->locked. Do not allow preemption during this call,
+// or else the m might be in different in this function than in the caller.
+//go:nosplit
+func dounlockOSThread() {
+	if GOARCH == "wasm" {
+		return // no threads on wasm yet
+	}
+	_g_ := getg()
+	if _g_.m.lockedInt != 0 || _g_.m.lockedExt != 0 {
+		return
+	}
+	_g_.m.lockedg = 0
+	_g_.lockedm = 0
+}
+
+//go:nosplit
+
+// UnlockOSThread undoes an earlier call to LockOSThread.
+// If this drops the number of active LockOSThread calls on the
+// calling goroutine to zero, it unwires the calling goroutine from
+// its fixed operating system thread.
+// If there are no active LockOSThread calls, this is a no-op.
+//
+// Before calling UnlockOSThread, the caller must ensure that the OS
+// thread is suitable for running other goroutines. If the caller made
+// any permanent changes to the state of the thread that would affect
+// other goroutines, it should not call this function and thus leave
+// the goroutine locked to the OS thread until the goroutine (and
+// hence the thread) exits.
+func UnlockOSThread() {
+	_g_ := getg()
+	if _g_.m.lockedExt == 0 {
+		return
+	}
+	_g_.m.lockedExt--
+	dounlockOSThread()
+}
+
+//go:nosplit
+func unlockOSThread() {
+	_g_ := getg()
+	if _g_.m.lockedInt == 0 {
+		systemstack(badunlockosthread)
+	}
+	_g_.m.lockedInt--
+	dounlockOSThread()
+}
+
+func badunlockosthread() {
+	throw("runtime: internal error: misuse of lockOSThread/unlockOSThread")
+}
+
+func gcount() int32 {
+	n := int32(atomic.Loaduintptr(&allglen)) - sched.gFree.n - int32(atomic.Load(&sched.ngsys))
+	for _, _p_ := range allp {
+		n -= _p_.gFree.n
+	}
+
+	// All these variables can be changed concurrently, so the result can be inconsistent.
+	// But at least the current goroutine is running.
+	if n < 1 {
+		n = 1
+	}
+	return n
+}
+
+func mcount() int32 {
+	return int32(sched.mnext - sched.nmfreed)
+}
+
+var prof struct {
+	signalLock uint32
+	hz         int32
+}
+
+func _System()                    { _System() }
+func _ExternalCode()              { _ExternalCode() }
+func _LostExternalCode()          { _LostExternalCode() }
+func _GC()                        { _GC() }
+func _LostSIGPROFDuringAtomic64() { _LostSIGPROFDuringAtomic64() }
+func _VDSO()                      { _VDSO() }
+
+// Called if we receive a SIGPROF signal.
+// Called by the signal handler, may run during STW.
+//go:nowritebarrierrec
+func sigprof(pc, sp, lr uintptr, gp *g, mp *m) {
+	if prof.hz == 0 {
+		return
+	}
+
+	// If mp.profilehz is 0, then profiling is not enabled for this thread.
+	// We must check this to avoid a deadlock between setcpuprofilerate
+	// and the call to cpuprof.add, below.
+	if mp != nil && mp.profilehz == 0 {
+		return
+	}
+
+	// On mips{,le}/arm, 64bit atomics are emulated with spinlocks, in
+	// runtime/internal/atomic. If SIGPROF arrives while the program is inside
+	// the critical section, it creates a deadlock (when writing the sample).
+	// As a workaround, create a counter of SIGPROFs while in critical section
+	// to store the count, and pass it to sigprof.add() later when SIGPROF is
+	// received from somewhere else (with _LostSIGPROFDuringAtomic64 as pc).
+	if GOARCH == "mips" || GOARCH == "mipsle" || GOARCH == "arm" {
+		if f := findfunc(pc); f.valid() {
+			if hasPrefix(funcname(f), "runtime/internal/atomic") {
+				cpuprof.lostAtomic++
+				return
+			}
+		}
+		if GOARCH == "arm" && goarm < 7 && GOOS == "linux" && pc&0xffff0000 == 0xffff0000 {
+			// runtime/internal/atomic functions call into kernel
+			// helpers on arm < 7. See
+			// runtime/internal/atomic/sys_linux_arm.s.
+			cpuprof.lostAtomic++
+			return
+		}
+	}
+
+	// Profiling runs concurrently with GC, so it must not allocate.
+	// Set a trap in case the code does allocate.
+	// Note that on windows, one thread takes profiles of all the
+	// other threads, so mp is usually not getg().m.
+	// In fact mp may not even be stopped.
+	// See golang.org/issue/17165.
+	getg().m.mallocing++
+
+	var stk [maxCPUProfStack]uintptr
+	n := 0
+	if mp.ncgo > 0 && mp.curg != nil && mp.curg.syscallpc != 0 && mp.curg.syscallsp != 0 {
+		cgoOff := 0
+		// Check cgoCallersUse to make sure that we are not
+		// interrupting other code that is fiddling with
+		// cgoCallers.  We are running in a signal handler
+		// with all signals blocked, so we don't have to worry
+		// about any other code interrupting us.
+		if atomic.Load(&mp.cgoCallersUse) == 0 && mp.cgoCallers != nil && mp.cgoCallers[0] != 0 {
+			for cgoOff < len(mp.cgoCallers) && mp.cgoCallers[cgoOff] != 0 {
+				cgoOff++
+			}
+			copy(stk[:], mp.cgoCallers[:cgoOff])
+			mp.cgoCallers[0] = 0
+		}
+
+		// Collect Go stack that leads to the cgo call.
+		n = gentraceback(mp.curg.syscallpc, mp.curg.syscallsp, 0, mp.curg, 0, &stk[cgoOff], len(stk)-cgoOff, nil, nil, 0)
+		if n > 0 {
+			n += cgoOff
+		}
+	} else {
+		n = gentraceback(pc, sp, lr, gp, 0, &stk[0], len(stk), nil, nil, _TraceTrap|_TraceJumpStack)
+	}
+
+	if n <= 0 {
+		// Normal traceback is impossible or has failed.
+		// See if it falls into several common cases.
+		n = 0
+		if usesLibcall() && mp.libcallg != 0 && mp.libcallpc != 0 && mp.libcallsp != 0 {
+			// Libcall, i.e. runtime syscall on windows.
+			// Collect Go stack that leads to the call.
+			n = gentraceback(mp.libcallpc, mp.libcallsp, 0, mp.libcallg.ptr(), 0, &stk[0], len(stk), nil, nil, 0)
+		}
+		if n == 0 && mp != nil && mp.vdsoSP != 0 {
+			n = gentraceback(mp.vdsoPC, mp.vdsoSP, 0, gp, 0, &stk[0], len(stk), nil, nil, _TraceTrap|_TraceJumpStack)
+		}
+		if n == 0 {
+			// If all of the above has failed, account it against abstract "System" or "GC".
+			n = 2
+			if inVDSOPage(pc) {
+				pc = abi.FuncPCABIInternal(_VDSO) + sys.PCQuantum
+			} else if pc > firstmoduledata.etext {
+				// "ExternalCode" is better than "etext".
+				pc = abi.FuncPCABIInternal(_ExternalCode) + sys.PCQuantum
+			}
+			stk[0] = pc
+			if mp.preemptoff != "" {
+				stk[1] = abi.FuncPCABIInternal(_GC) + sys.PCQuantum
+			} else {
+				stk[1] = abi.FuncPCABIInternal(_System) + sys.PCQuantum
+			}
+		}
+	}
+
+	if prof.hz != 0 {
+		// Note: it can happen on Windows that we interrupted a system thread
+		// with no g, so gp could nil. The other nil checks are done out of
+		// caution, but not expected to be nil in practice.
+		var tagPtr *unsafe.Pointer
+		if gp != nil && gp.m != nil && gp.m.curg != nil {
+			tagPtr = &gp.m.curg.labels
+		}
+		cpuprof.add(tagPtr, stk[:n])
+	}
+	getg().m.mallocing--
+}
+
+// setcpuprofilerate sets the CPU profiling rate to hz times per second.
+// If hz <= 0, setcpuprofilerate turns off CPU profiling.
+func setcpuprofilerate(hz int32) {
+	// Force sane arguments.
+	if hz < 0 {
+		hz = 0
+	}
+
+	// Disable preemption, otherwise we can be rescheduled to another thread
+	// that has profiling enabled.
+	_g_ := getg()
+	_g_.m.locks++
+
+	// Stop profiler on this thread so that it is safe to lock prof.
+	// if a profiling signal came in while we had prof locked,
+	// it would deadlock.
+	setThreadCPUProfiler(0)
+
+	for !atomic.Cas(&prof.signalLock, 0, 1) {
+		osyield()
+	}
+	if prof.hz != hz {
+		setProcessCPUProfiler(hz)
+		prof.hz = hz
+	}
+	atomic.Store(&prof.signalLock, 0)
+
+	lock(&sched.lock)
+	sched.profilehz = hz
+	unlock(&sched.lock)
+
+	if hz != 0 {
+		setThreadCPUProfiler(hz)
+	}
+
+	_g_.m.locks--
+}
+
+// init initializes pp, which may be a freshly allocated p or a
+// previously destroyed p, and transitions it to status _Pgcstop.
+func (pp *p) init(id int32) {
+	pp.id = id
+	pp.status = _Pgcstop
+	pp.sudogcache = pp.sudogbuf[:0]
+	pp.deferpool = pp.deferpoolbuf[:0]
+	pp.wbBuf.reset()
+	if pp.mcache == nil {
+		if id == 0 {
+			if mcache0 == nil {
+				throw("missing mcache?")
+			}
+			// Use the bootstrap mcache0. Only one P will get
+			// mcache0: the one with ID 0.
+			pp.mcache = mcache0
+		} else {
+			pp.mcache = allocmcache()
+		}
+	}
+	if raceenabled && pp.raceprocctx == 0 {
+		if id == 0 {
+			pp.raceprocctx = raceprocctx0
+			raceprocctx0 = 0 // bootstrap
+		} else {
+			pp.raceprocctx = raceproccreate()
+		}
+	}
+	lockInit(&pp.timersLock, lockRankTimers)
+
+	// This P may get timers when it starts running. Set the mask here
+	// since the P may not go through pidleget (notably P 0 on startup).
+	timerpMask.set(id)
+	// Similarly, we may not go through pidleget before this P starts
+	// running if it is P 0 on startup.
+	idlepMask.clear(id)
+}
+
+// destroy releases all of the resources associated with pp and
+// transitions it to status _Pdead.
+//
+// sched.lock must be held and the world must be stopped.
+func (pp *p) destroy() {
+	assertLockHeld(&sched.lock)
+	assertWorldStopped()
+
+	// Move all runnable goroutines to the global queue
+	for pp.runqhead != pp.runqtail {
+		// Pop from tail of local queue
+		pp.runqtail--
+		gp := pp.runq[pp.runqtail%uint32(len(pp.runq))].ptr()
+		// Push onto head of global queue
+		globrunqputhead(gp)
+	}
+	if pp.runnext != 0 {
+		globrunqputhead(pp.runnext.ptr())
+		pp.runnext = 0
+	}
+	if len(pp.timers) > 0 {
+		plocal := getg().m.p.ptr()
+		// The world is stopped, but we acquire timersLock to
+		// protect against sysmon calling timeSleepUntil.
+		// This is the only case where we hold the timersLock of
+		// more than one P, so there are no deadlock concerns.
+		lock(&plocal.timersLock)
+		lock(&pp.timersLock)
+		moveTimers(plocal, pp.timers)
+		pp.timers = nil
+		pp.numTimers = 0
+		pp.deletedTimers = 0
+		atomic.Store64(&pp.timer0When, 0)
+		unlock(&pp.timersLock)
+		unlock(&plocal.timersLock)
+	}
+	// Flush p's write barrier buffer.
+	if gcphase != _GCoff {
+		wbBufFlush1(pp)
+		pp.gcw.dispose()
+	}
+	for i := range pp.sudogbuf {
+		pp.sudogbuf[i] = nil
+	}
+	pp.sudogcache = pp.sudogbuf[:0]
+	for j := range pp.deferpoolbuf {
+		pp.deferpoolbuf[j] = nil
+	}
+	pp.deferpool = pp.deferpoolbuf[:0]
+	systemstack(func() {
+		for i := 0; i < pp.mspancache.len; i++ {
+			// Safe to call since the world is stopped.
+			mheap_.spanalloc.free(unsafe.Pointer(pp.mspancache.buf[i]))
+		}
+		pp.mspancache.len = 0
+		lock(&mheap_.lock)
+		pp.pcache.flush(&mheap_.pages)
+		unlock(&mheap_.lock)
+	})
+	freemcache(pp.mcache)
+	pp.mcache = nil
+	gfpurge(pp)
+	traceProcFree(pp)
+	if raceenabled {
+		if pp.timerRaceCtx != 0 {
+			// The race detector code uses a callback to fetch
+			// the proc context, so arrange for that callback
+			// to see the right thing.
+			// This hack only works because we are the only
+			// thread running.
+			mp := getg().m
+			phold := mp.p.ptr()
+			mp.p.set(pp)
+
+			racectxend(pp.timerRaceCtx)
+			pp.timerRaceCtx = 0
+
+			mp.p.set(phold)
+		}
+		raceprocdestroy(pp.raceprocctx)
+		pp.raceprocctx = 0
+	}
+	pp.gcAssistTime = 0
+	pp.status = _Pdead
+}
+
+// Change number of processors.
+//
+// sched.lock must be held, and the world must be stopped.
+//
+// gcworkbufs must not be being modified by either the GC or the write barrier
+// code, so the GC must not be running if the number of Ps actually changes.
+//
+// Returns list of Ps with local work, they need to be scheduled by the caller.
+func procresize(nprocs int32) *p {
+	assertLockHeld(&sched.lock)
+	assertWorldStopped()
+
+	old := gomaxprocs
+	if old < 0 || nprocs <= 0 {
+		throw("procresize: invalid arg")
+	}
+	if trace.enabled {
+		traceGomaxprocs(nprocs)
+	}
+
+	// update statistics
+	now := nanotime()
+	if sched.procresizetime != 0 {
+		sched.totaltime += int64(old) * (now - sched.procresizetime)
+	}
+	sched.procresizetime = now
+
+	maskWords := (nprocs + 31) / 32
+
+	// Grow allp if necessary.
+	if nprocs > int32(len(allp)) {
+		// Synchronize with retake, which could be running
+		// concurrently since it doesn't run on a P.
+		lock(&allpLock)
+		if nprocs <= int32(cap(allp)) {
+			allp = allp[:nprocs]
+		} else {
+			nallp := make([]*p, nprocs)
+			// Copy everything up to allp's cap so we
+			// never lose old allocated Ps.
+			copy(nallp, allp[:cap(allp)])
+			allp = nallp
+		}
+
+		if maskWords <= int32(cap(idlepMask)) {
+			idlepMask = idlepMask[:maskWords]
+			timerpMask = timerpMask[:maskWords]
+		} else {
+			nidlepMask := make([]uint32, maskWords)
+			// No need to copy beyond len, old Ps are irrelevant.
+			copy(nidlepMask, idlepMask)
+			idlepMask = nidlepMask
+
+			ntimerpMask := make([]uint32, maskWords)
+			copy(ntimerpMask, timerpMask)
+			timerpMask = ntimerpMask
+		}
+		unlock(&allpLock)
+	}
+
+	// initialize new P's
+	for i := old; i < nprocs; i++ {
+		pp := allp[i]
+		if pp == nil {
+			pp = new(p)
+		}
+		pp.init(i)
+		atomicstorep(unsafe.Pointer(&allp[i]), unsafe.Pointer(pp))
+	}
+
+	_g_ := getg()
+	if _g_.m.p != 0 && _g_.m.p.ptr().id < nprocs {
+		// continue to use the current P
+		_g_.m.p.ptr().status = _Prunning
+		_g_.m.p.ptr().mcache.prepareForSweep()
+	} else {
+		// release the current P and acquire allp[0].
+		//
+		// We must do this before destroying our current P
+		// because p.destroy itself has write barriers, so we
+		// need to do that from a valid P.
+		if _g_.m.p != 0 {
+			if trace.enabled {
+				// Pretend that we were descheduled
+				// and then scheduled again to keep
+				// the trace sane.
+				traceGoSched()
+				traceProcStop(_g_.m.p.ptr())
+			}
+			_g_.m.p.ptr().m = 0
+		}
+		_g_.m.p = 0
+		p := allp[0]
+		p.m = 0
+		p.status = _Pidle
+		acquirep(p)
+		if trace.enabled {
+			traceGoStart()
+		}
+	}
+
+	// g.m.p is now set, so we no longer need mcache0 for bootstrapping.
+	mcache0 = nil
+
+	// release resources from unused P's
+	for i := nprocs; i < old; i++ {
+		p := allp[i]
+		p.destroy()
+		// can't free P itself because it can be referenced by an M in syscall
+	}
+
+	// Trim allp.
+	if int32(len(allp)) != nprocs {
+		lock(&allpLock)
+		allp = allp[:nprocs]
+		idlepMask = idlepMask[:maskWords]
+		timerpMask = timerpMask[:maskWords]
+		unlock(&allpLock)
+	}
+
+	var runnablePs *p
+	for i := nprocs - 1; i >= 0; i-- {
+		p := allp[i]
+		if _g_.m.p.ptr() == p {
+			continue
+		}
+		p.status = _Pidle
+		if runqempty(p) {
+			pidleput(p)
+		} else {
+			p.m.set(mget())
+			p.link.set(runnablePs)
+			runnablePs = p
+		}
+	}
+	stealOrder.reset(uint32(nprocs))
+	var int32p *int32 = &gomaxprocs // make compiler check that gomaxprocs is an int32
+	atomic.Store((*uint32)(unsafe.Pointer(int32p)), uint32(nprocs))
+	return runnablePs
+}
+
+// Associate p and the current m.
+//
+// This function is allowed to have write barriers even if the caller
+// isn't because it immediately acquires _p_.
+//
+//go:yeswritebarrierrec
+func acquirep(_p_ *p) {
+	// Do the part that isn't allowed to have write barriers.
+	wirep(_p_)
+
+	// Have p; write barriers now allowed.
+
+	// Perform deferred mcache flush before this P can allocate
+	// from a potentially stale mcache.
+	_p_.mcache.prepareForSweep()
+
+	if trace.enabled {
+		traceProcStart()
+	}
+}
+
+// wirep is the first step of acquirep, which actually associates the
+// current M to _p_. This is broken out so we can disallow write
+// barriers for this part, since we don't yet have a P.
+//
+//go:nowritebarrierrec
+//go:nosplit
+func wirep(_p_ *p) {
+	_g_ := getg()
+
+	if _g_.m.p != 0 {
+		throw("wirep: already in go")
+	}
+	if _p_.m != 0 || _p_.status != _Pidle {
+		id := int64(0)
+		if _p_.m != 0 {
+			id = _p_.m.ptr().id
+		}
+		print("wirep: p->m=", _p_.m, "(", id, ") p->status=", _p_.status, "\n")
+		throw("wirep: invalid p state")
+	}
+	_g_.m.p.set(_p_)
+	_p_.m.set(_g_.m)
+	_p_.status = _Prunning
+}
+
+// Disassociate p and the current m.
+func releasep() *p {
+	_g_ := getg()
+
+	if _g_.m.p == 0 {
+		throw("releasep: invalid arg")
+	}
+	_p_ := _g_.m.p.ptr()
+	if _p_.m.ptr() != _g_.m || _p_.status != _Prunning {
+		print("releasep: m=", _g_.m, " m->p=", _g_.m.p.ptr(), " p->m=", hex(_p_.m), " p->status=", _p_.status, "\n")
+		throw("releasep: invalid p state")
+	}
+	if trace.enabled {
+		traceProcStop(_g_.m.p.ptr())
+	}
+	_g_.m.p = 0
+	_p_.m = 0
+	_p_.status = _Pidle
+	return _p_
+}
+
+func incidlelocked(v int32) {
+	lock(&sched.lock)
+	sched.nmidlelocked += v
+	if v > 0 {
+		checkdead()
+	}
+	unlock(&sched.lock)
+}
+
+// Check for deadlock situation.
+// The check is based on number of running M's, if 0 -> deadlock.
+// sched.lock must be held.
+func checkdead() {
+	assertLockHeld(&sched.lock)
+
+	// For -buildmode=c-shared or -buildmode=c-archive it's OK if
+	// there are no running goroutines. The calling program is
+	// assumed to be running.
+	if islibrary || isarchive {
+		return
+	}
+
+	// If we are dying because of a signal caught on an already idle thread,
+	// freezetheworld will cause all running threads to block.
+	// And runtime will essentially enter into deadlock state,
+	// except that there is a thread that will call exit soon.
+	if panicking > 0 {
+		return
+	}
+
+	// If we are not running under cgo, but we have an extra M then account
+	// for it. (It is possible to have an extra M on Windows without cgo to
+	// accommodate callbacks created by syscall.NewCallback. See issue #6751
+	// for details.)
+	var run0 int32
+	if !iscgo && cgoHasExtraM {
+		mp := lockextra(true)
+		haveExtraM := extraMCount > 0
+		unlockextra(mp)
+		if haveExtraM {
+			run0 = 1
+		}
+	}
+
+	run := mcount() - sched.nmidle - sched.nmidlelocked - sched.nmsys
+	if run > run0 {
+		return
+	}
+	if run < 0 {
+		print("runtime: checkdead: nmidle=", sched.nmidle, " nmidlelocked=", sched.nmidlelocked, " mcount=", mcount(), " nmsys=", sched.nmsys, "\n")
+		throw("checkdead: inconsistent counts")
+	}
+
+	grunning := 0
+	forEachG(func(gp *g) {
+		if isSystemGoroutine(gp, false) {
+			return
+		}
+		s := readgstatus(gp)
+		switch s &^ _Gscan {
+		case _Gwaiting,
+			_Gpreempted:
+			grunning++
+		case _Grunnable,
+			_Grunning,
+			_Gsyscall:
+			print("runtime: checkdead: find g ", gp.goid, " in status ", s, "\n")
+			throw("checkdead: runnable g")
+		}
+	})
+	if grunning == 0 { // possible if main goroutine calls runtime·Goexit()
+		unlock(&sched.lock) // unlock so that GODEBUG=scheddetail=1 doesn't hang
+		throw("no goroutines (main called runtime.Goexit) - deadlock!")
+	}
+
+	// Maybe jump time forward for playground.
+	if faketime != 0 {
+		when, _p_ := timeSleepUntil()
+		if _p_ != nil {
+			faketime = when
+			for pp := &sched.pidle; *pp != 0; pp = &(*pp).ptr().link {
+				if (*pp).ptr() == _p_ {
+					*pp = _p_.link
+					break
+				}
+			}
+			mp := mget()
+			if mp == nil {
+				// There should always be a free M since
+				// nothing is running.
+				throw("checkdead: no m for timer")
+			}
+			mp.nextp.set(_p_)
+			notewakeup(&mp.park)
+			return
+		}
+	}
+
+	// There are no goroutines running, so we can look at the P's.
+	for _, _p_ := range allp {
+		if len(_p_.timers) > 0 {
+			return
+		}
+	}
+
+	getg().m.throwing = -1 // do not dump full stacks
+	unlock(&sched.lock)    // unlock so that GODEBUG=scheddetail=1 doesn't hang
+	throw("all goroutines are asleep - deadlock!")
+}
+
+// forcegcperiod is the maximum time in nanoseconds between garbage
+// collections. If we go this long without a garbage collection, one
+// is forced to run.
+//
+// This is a variable for testing purposes. It normally doesn't change.
+var forcegcperiod int64 = 2 * 60 * 1e9
+
+// needSysmonWorkaround is true if the workaround for
+// golang.org/issue/42515 is needed on NetBSD.
+var needSysmonWorkaround bool = false
+
+// Always runs without a P, so write barriers are not allowed.
+//
+//go:nowritebarrierrec
+func sysmon() {
+	lock(&sched.lock)
+	sched.nmsys++
+	checkdead()
+	unlock(&sched.lock)
+
+	lasttrace := int64(0)
+	idle := 0 // how many cycles in succession we had not wokeup somebody
+	delay := uint32(0)
+
+	for {
+		if idle == 0 { // start with 20us sleep...
+			delay = 20
+		} else if idle > 50 { // start doubling the sleep after 1ms...
+			delay *= 2
+		}
+		if delay > 10*1000 { // up to 10ms
+			delay = 10 * 1000
+		}
+		usleep(delay)
+
+		// sysmon should not enter deep sleep if schedtrace is enabled so that
+		// it can print that information at the right time.
+		//
+		// It should also not enter deep sleep if there are any active P's so
+		// that it can retake P's from syscalls, preempt long running G's, and
+		// poll the network if all P's are busy for long stretches.
+		//
+		// It should wakeup from deep sleep if any P's become active either due
+		// to exiting a syscall or waking up due to a timer expiring so that it
+		// can resume performing those duties. If it wakes from a syscall it
+		// resets idle and delay as a bet that since it had retaken a P from a
+		// syscall before, it may need to do it again shortly after the
+		// application starts work again. It does not reset idle when waking
+		// from a timer to avoid adding system load to applications that spend
+		// most of their time sleeping.
+		now := nanotime()
+		if debug.schedtrace <= 0 && (sched.gcwaiting != 0 || atomic.Load(&sched.npidle) == uint32(gomaxprocs)) {
+			lock(&sched.lock)
+			if atomic.Load(&sched.gcwaiting) != 0 || atomic.Load(&sched.npidle) == uint32(gomaxprocs) {
+				syscallWake := false
+				next, _ := timeSleepUntil()
+				if next > now {
+					atomic.Store(&sched.sysmonwait, 1)
+					unlock(&sched.lock)
+					// Make wake-up period small enough
+					// for the sampling to be correct.
+					sleep := forcegcperiod / 2
+					if next-now < sleep {
+						sleep = next - now
+					}
+					shouldRelax := sleep >= osRelaxMinNS
+					if shouldRelax {
+						osRelax(true)
+					}
+					syscallWake = notetsleep(&sched.sysmonnote, sleep)
+					if shouldRelax {
+						osRelax(false)
+					}
+					lock(&sched.lock)
+					atomic.Store(&sched.sysmonwait, 0)
+					noteclear(&sched.sysmonnote)
+				}
+				if syscallWake {
+					idle = 0
+					delay = 20
+				}
+			}
+			unlock(&sched.lock)
+		}
+
+		lock(&sched.sysmonlock)
+		// Update now in case we blocked on sysmonnote or spent a long time
+		// blocked on schedlock or sysmonlock above.
+		now = nanotime()
+
+		// trigger libc interceptors if needed
+		if *cgo_yield != nil {
+			asmcgocall(*cgo_yield, nil)
+		}
+		// poll network if not polled for more than 10ms
+		lastpoll := int64(atomic.Load64(&sched.lastpoll))
+		if netpollinited() && lastpoll != 0 && lastpoll+10*1000*1000 < now {
+			atomic.Cas64(&sched.lastpoll, uint64(lastpoll), uint64(now))
+			list := netpoll(0) // non-blocking - returns list of goroutines
+			if !list.empty() {
+				// Need to decrement number of idle locked M's
+				// (pretending that one more is running) before injectglist.
+				// Otherwise it can lead to the following situation:
+				// injectglist grabs all P's but before it starts M's to run the P's,
+				// another M returns from syscall, finishes running its G,
+				// observes that there is no work to do and no other running M's
+				// and reports deadlock.
+				incidlelocked(-1)
+				injectglist(&list)
+				incidlelocked(1)
+			}
+		}
+		if GOOS == "netbsd" && needSysmonWorkaround {
+			// netpoll is responsible for waiting for timer
+			// expiration, so we typically don't have to worry
+			// about starting an M to service timers. (Note that
+			// sleep for timeSleepUntil above simply ensures sysmon
+			// starts running again when that timer expiration may
+			// cause Go code to run again).
+			//
+			// However, netbsd has a kernel bug that sometimes
+			// misses netpollBreak wake-ups, which can lead to
+			// unbounded delays servicing timers. If we detect this
+			// overrun, then startm to get something to handle the
+			// timer.
+			//
+			// See issue 42515 and
+			// https://gnats.netbsd.org/cgi-bin/query-pr-single.pl?number=50094.
+			if next, _ := timeSleepUntil(); next < now {
+				startm(nil, false)
+			}
+		}
+		if atomic.Load(&scavenge.sysmonWake) != 0 {
+			// Kick the scavenger awake if someone requested it.
+			wakeScavenger()
+		}
+		// retake P's blocked in syscalls
+		// and preempt long running G's
+		if retake(now) != 0 {
+			idle = 0
+		} else {
+			idle++
+		}
+		// check if we need to force a GC
+		if t := (gcTrigger{kind: gcTriggerTime, now: now}); t.test() && atomic.Load(&forcegc.idle) != 0 {
+			lock(&forcegc.lock)
+			forcegc.idle = 0
+			var list gList
+			list.push(forcegc.g)
+			injectglist(&list)
+			unlock(&forcegc.lock)
+		}
+		if debug.schedtrace > 0 && lasttrace+int64(debug.schedtrace)*1000000 <= now {
+			lasttrace = now
+			schedtrace(debug.scheddetail > 0)
+		}
+		unlock(&sched.sysmonlock)
+	}
+}
+
+type sysmontick struct {
+	schedtick   uint32
+	schedwhen   int64
+	syscalltick uint32
+	syscallwhen int64
+}
+
+// forcePreemptNS is the time slice given to a G before it is
+// preempted.
+const forcePreemptNS = 10 * 1000 * 1000 // 10ms
+
+func retake(now int64) uint32 {
+	n := 0
+	// Prevent allp slice changes. This lock will be completely
+	// uncontended unless we're already stopping the world.
+	lock(&allpLock)
+	// We can't use a range loop over allp because we may
+	// temporarily drop the allpLock. Hence, we need to re-fetch
+	// allp each time around the loop.
+	for i := 0; i < len(allp); i++ {
+		_p_ := allp[i]
+		if _p_ == nil {
+			// This can happen if procresize has grown
+			// allp but not yet created new Ps.
+			continue
+		}
+		pd := &_p_.sysmontick
+		s := _p_.status
+		sysretake := false
+		if s == _Prunning || s == _Psyscall {
+			// Preempt G if it's running for too long.
+			t := int64(_p_.schedtick)
+			if int64(pd.schedtick) != t {
+				pd.schedtick = uint32(t)
+				pd.schedwhen = now
+			} else if pd.schedwhen+forcePreemptNS <= now {
+				preemptone(_p_)
+				// In case of syscall, preemptone() doesn't
+				// work, because there is no M wired to P.
+				sysretake = true
+			}
+		}
+		if s == _Psyscall {
+			// Retake P from syscall if it's there for more than 1 sysmon tick (at least 20us).
+			t := int64(_p_.syscalltick)
+			if !sysretake && int64(pd.syscalltick) != t {
+				pd.syscalltick = uint32(t)
+				pd.syscallwhen = now
+				continue
+			}
+			// On the one hand we don't want to retake Ps if there is no other work to do,
+			// but on the other hand we want to retake them eventually
+			// because they can prevent the sysmon thread from deep sleep.
+			if runqempty(_p_) && atomic.Load(&sched.nmspinning)+atomic.Load(&sched.npidle) > 0 && pd.syscallwhen+10*1000*1000 > now {
+				continue
+			}
+			// Drop allpLock so we can take sched.lock.
+			unlock(&allpLock)
+			// Need to decrement number of idle locked M's
+			// (pretending that one more is running) before the CAS.
+			// Otherwise the M from which we retake can exit the syscall,
+			// increment nmidle and report deadlock.
+			incidlelocked(-1)
+			if atomic.Cas(&_p_.status, s, _Pidle) {
+				if trace.enabled {
+					traceGoSysBlock(_p_)
+					traceProcStop(_p_)
+				}
+				n++
+				_p_.syscalltick++
+				handoffp(_p_)
+			}
+			incidlelocked(1)
+			lock(&allpLock)
+		}
+	}
+	unlock(&allpLock)
+	return uint32(n)
+}
+
+// Tell all goroutines that they have been preempted and they should stop.
+// This function is purely best-effort. It can fail to inform a goroutine if a
+// processor just started running it.
+// No locks need to be held.
+// Returns true if preemption request was issued to at least one goroutine.
+func preemptall() bool {
+	res := false
+	for _, _p_ := range allp {
+		if _p_.status != _Prunning {
+			continue
+		}
+		if preemptone(_p_) {
+			res = true
+		}
+	}
+	return res
+}
+
+// Tell the goroutine running on processor P to stop.
+// This function is purely best-effort. It can incorrectly fail to inform the
+// goroutine. It can inform the wrong goroutine. Even if it informs the
+// correct goroutine, that goroutine might ignore the request if it is
+// simultaneously executing newstack.
+// No lock needs to be held.
+// Returns true if preemption request was issued.
+// The actual preemption will happen at some point in the future
+// and will be indicated by the gp->status no longer being
+// Grunning
+func preemptone(_p_ *p) bool {
+	mp := _p_.m.ptr()
+	if mp == nil || mp == getg().m {
+		return false
+	}
+	gp := mp.curg
+	if gp == nil || gp == mp.g0 {
+		return false
+	}
+
+	gp.preempt = true
+
+	// Every call in a goroutine checks for stack overflow by
+	// comparing the current stack pointer to gp->stackguard0.
+	// Setting gp->stackguard0 to StackPreempt folds
+	// preemption into the normal stack overflow check.
+	gp.stackguard0 = stackPreempt
+
+	// Request an async preemption of this P.
+	if preemptMSupported && debug.asyncpreemptoff == 0 {
+		_p_.preempt = true
+		preemptM(mp)
+	}
+
+	return true
+}
+
+var starttime int64
+
+func schedtrace(detailed bool) {
+	now := nanotime()
+	if starttime == 0 {
+		starttime = now
+	}
+
+	lock(&sched.lock)
+	print("SCHED ", (now-starttime)/1e6, "ms: gomaxprocs=", gomaxprocs, " idleprocs=", sched.npidle, " threads=", mcount(), " spinningthreads=", sched.nmspinning, " idlethreads=", sched.nmidle, " runqueue=", sched.runqsize)
+	if detailed {
+		print(" gcwaiting=", sched.gcwaiting, " nmidlelocked=", sched.nmidlelocked, " stopwait=", sched.stopwait, " sysmonwait=", sched.sysmonwait, "\n")
+	}
+	// We must be careful while reading data from P's, M's and G's.
+	// Even if we hold schedlock, most data can be changed concurrently.
+	// E.g. (p->m ? p->m->id : -1) can crash if p->m changes from non-nil to nil.
+	for i, _p_ := range allp {
+		mp := _p_.m.ptr()
+		h := atomic.Load(&_p_.runqhead)
+		t := atomic.Load(&_p_.runqtail)
+		if detailed {
+			id := int64(-1)
+			if mp != nil {
+				id = mp.id
+			}
+			print("  P", i, ": status=", _p_.status, " schedtick=", _p_.schedtick, " syscalltick=", _p_.syscalltick, " m=", id, " runqsize=", t-h, " gfreecnt=", _p_.gFree.n, " timerslen=", len(_p_.timers), "\n")
+		} else {
+			// In non-detailed mode format lengths of per-P run queues as:
+			// [len1 len2 len3 len4]
+			print(" ")
+			if i == 0 {
+				print("[")
+			}
+			print(t - h)
+			if i == len(allp)-1 {
+				print("]\n")
+			}
+		}
+	}
+
+	if !detailed {
+		unlock(&sched.lock)
+		return
+	}
+
+	for mp := allm; mp != nil; mp = mp.alllink {
+		_p_ := mp.p.ptr()
+		gp := mp.curg
+		lockedg := mp.lockedg.ptr()
+		id1 := int32(-1)
+		if _p_ != nil {
+			id1 = _p_.id
+		}
+		id2 := int64(-1)
+		if gp != nil {
+			id2 = gp.goid
+		}
+		id3 := int64(-1)
+		if lockedg != nil {
+			id3 = lockedg.goid
+		}
+		print("  M", mp.id, ": p=", id1, " curg=", id2, " mallocing=", mp.mallocing, " throwing=", mp.throwing, " preemptoff=", mp.preemptoff, ""+" locks=", mp.locks, " dying=", mp.dying, " spinning=", mp.spinning, " blocked=", mp.blocked, " lockedg=", id3, "\n")
+	}
+
+	forEachG(func(gp *g) {
+		mp := gp.m
+		lockedm := gp.lockedm.ptr()
+		id1 := int64(-1)
+		if mp != nil {
+			id1 = mp.id
+		}
+		id2 := int64(-1)
+		if lockedm != nil {
+			id2 = lockedm.id
+		}
+		print("  G", gp.goid, ": status=", readgstatus(gp), "(", gp.waitreason.String(), ") m=", id1, " lockedm=", id2, "\n")
+	})
+	unlock(&sched.lock)
+}
+
+// schedEnableUser enables or disables the scheduling of user
+// goroutines.
+//
+// This does not stop already running user goroutines, so the caller
+// should first stop the world when disabling user goroutines.
+func schedEnableUser(enable bool) {
+	lock(&sched.lock)
+	if sched.disable.user == !enable {
+		unlock(&sched.lock)
+		return
+	}
+	sched.disable.user = !enable
+	if enable {
+		n := sched.disable.n
+		sched.disable.n = 0
+		globrunqputbatch(&sched.disable.runnable, n)
+		unlock(&sched.lock)
+		for ; n != 0 && sched.npidle != 0; n-- {
+			startm(nil, false)
+		}
+	} else {
+		unlock(&sched.lock)
+	}
+}
+
+// schedEnabled reports whether gp should be scheduled. It returns
+// false is scheduling of gp is disabled.
+//
+// sched.lock must be held.
+func schedEnabled(gp *g) bool {
+	assertLockHeld(&sched.lock)
+
+	if sched.disable.user {
+		return isSystemGoroutine(gp, true)
+	}
+	return true
+}
+
+// Put mp on midle list.
+// sched.lock must be held.
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func mput(mp *m) {
+	assertLockHeld(&sched.lock)
+
+	mp.schedlink = sched.midle
+	sched.midle.set(mp)
+	sched.nmidle++
+	checkdead()
+}
+
+// Try to get an m from midle list.
+// sched.lock must be held.
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func mget() *m {
+	assertLockHeld(&sched.lock)
+
+	mp := sched.midle.ptr()
+	if mp != nil {
+		sched.midle = mp.schedlink
+		sched.nmidle--
+	}
+	return mp
+}
+
+// Put gp on the global runnable queue.
+// sched.lock must be held.
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func globrunqput(gp *g) {
+	assertLockHeld(&sched.lock)
+
+	sched.runq.pushBack(gp)
+	sched.runqsize++
+}
+
+// Put gp at the head of the global runnable queue.
+// sched.lock must be held.
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func globrunqputhead(gp *g) {
+	assertLockHeld(&sched.lock)
+
+	sched.runq.push(gp)
+	sched.runqsize++
+}
+
+// Put a batch of runnable goroutines on the global runnable queue.
+// This clears *batch.
+// sched.lock must be held.
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func globrunqputbatch(batch *gQueue, n int32) {
+	assertLockHeld(&sched.lock)
+
+	sched.runq.pushBackAll(*batch)
+	sched.runqsize += n
+	*batch = gQueue{}
+}
+
+// Try get a batch of G's from the global runnable queue.
+// sched.lock must be held.
+func globrunqget(_p_ *p, max int32) *g {
+	assertLockHeld(&sched.lock)
+
+	if sched.runqsize == 0 {
+		return nil
+	}
+
+	n := sched.runqsize/gomaxprocs + 1
+	if n > sched.runqsize {
+		n = sched.runqsize
+	}
+	if max > 0 && n > max {
+		n = max
+	}
+	if n > int32(len(_p_.runq))/2 {
+		n = int32(len(_p_.runq)) / 2
+	}
+
+	sched.runqsize -= n
+
+	gp := sched.runq.pop()
+	n--
+	for ; n > 0; n-- {
+		gp1 := sched.runq.pop()
+		runqput(_p_, gp1, false)
+	}
+	return gp
+}
+
+// pMask is an atomic bitstring with one bit per P.
+type pMask []uint32
+
+// read returns true if P id's bit is set.
+func (p pMask) read(id uint32) bool {
+	word := id / 32
+	mask := uint32(1) << (id % 32)
+	return (atomic.Load(&p[word]) & mask) != 0
+}
+
+// set sets P id's bit.
+func (p pMask) set(id int32) {
+	word := id / 32
+	mask := uint32(1) << (id % 32)
+	atomic.Or(&p[word], mask)
+}
+
+// clear clears P id's bit.
+func (p pMask) clear(id int32) {
+	word := id / 32
+	mask := uint32(1) << (id % 32)
+	atomic.And(&p[word], ^mask)
+}
+
+// updateTimerPMask clears pp's timer mask if it has no timers on its heap.
+//
+// Ideally, the timer mask would be kept immediately consistent on any timer
+// operations. Unfortunately, updating a shared global data structure in the
+// timer hot path adds too much overhead in applications frequently switching
+// between no timers and some timers.
+//
+// As a compromise, the timer mask is updated only on pidleget / pidleput. A
+// running P (returned by pidleget) may add a timer at any time, so its mask
+// must be set. An idle P (passed to pidleput) cannot add new timers while
+// idle, so if it has no timers at that time, its mask may be cleared.
+//
+// Thus, we get the following effects on timer-stealing in findrunnable:
+//
+// * Idle Ps with no timers when they go idle are never checked in findrunnable
+//   (for work- or timer-stealing; this is the ideal case).
+// * Running Ps must always be checked.
+// * Idle Ps whose timers are stolen must continue to be checked until they run
+//   again, even after timer expiration.
+//
+// When the P starts running again, the mask should be set, as a timer may be
+// added at any time.
+//
+// TODO(prattmic): Additional targeted updates may improve the above cases.
+// e.g., updating the mask when stealing a timer.
+func updateTimerPMask(pp *p) {
+	if atomic.Load(&pp.numTimers) > 0 {
+		return
+	}
+
+	// Looks like there are no timers, however another P may transiently
+	// decrement numTimers when handling a timerModified timer in
+	// checkTimers. We must take timersLock to serialize with these changes.
+	lock(&pp.timersLock)
+	if atomic.Load(&pp.numTimers) == 0 {
+		timerpMask.clear(pp.id)
+	}
+	unlock(&pp.timersLock)
+}
+
+// pidleput puts p to on the _Pidle list.
+//
+// This releases ownership of p. Once sched.lock is released it is no longer
+// safe to use p.
+//
+// sched.lock must be held.
+//
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func pidleput(_p_ *p) {
+	assertLockHeld(&sched.lock)
+
+	if !runqempty(_p_) {
+		throw("pidleput: P has non-empty run queue")
+	}
+	updateTimerPMask(_p_) // clear if there are no timers.
+	idlepMask.set(_p_.id)
+	_p_.link = sched.pidle
+	sched.pidle.set(_p_)
+	atomic.Xadd(&sched.npidle, 1) // TODO: fast atomic
+}
+
+// pidleget tries to get a p from the _Pidle list, acquiring ownership.
+//
+// sched.lock must be held.
+//
+// May run during STW, so write barriers are not allowed.
+//go:nowritebarrierrec
+func pidleget() *p {
+	assertLockHeld(&sched.lock)
+
+	_p_ := sched.pidle.ptr()
+	if _p_ != nil {
+		// Timer may get added at any time now.
+		timerpMask.set(_p_.id)
+		idlepMask.clear(_p_.id)
+		sched.pidle = _p_.link
+		atomic.Xadd(&sched.npidle, -1) // TODO: fast atomic
+	}
+	return _p_
+}
+
+// runqempty reports whether _p_ has no Gs on its local run queue.
+// It never returns true spuriously.
+func runqempty(_p_ *p) bool {
+	// Defend against a race where 1) _p_ has G1 in runqnext but runqhead == runqtail,
+	// 2) runqput on _p_ kicks G1 to the runq, 3) runqget on _p_ empties runqnext.
+	// Simply observing that runqhead == runqtail and then observing that runqnext == nil
+	// does not mean the queue is empty.
+	for {
+		head := atomic.Load(&_p_.runqhead)
+		tail := atomic.Load(&_p_.runqtail)
+		runnext := atomic.Loaduintptr((*uintptr)(unsafe.Pointer(&_p_.runnext)))
+		if tail == atomic.Load(&_p_.runqtail) {
+			return head == tail && runnext == 0
+		}
+	}
+}
+
+// To shake out latent assumptions about scheduling order,
+// we introduce some randomness into scheduling decisions
+// when running with the race detector.
+// The need for this was made obvious by changing the
+// (deterministic) scheduling order in Go 1.5 and breaking
+// many poorly-written tests.
+// With the randomness here, as long as the tests pass
+// consistently with -race, they shouldn't have latent scheduling
+// assumptions.
+const randomizeScheduler = raceenabled
+
+// runqput tries to put g on the local runnable queue.
+// If next is false, runqput adds g to the tail of the runnable queue.
+// If next is true, runqput puts g in the _p_.runnext slot.
+// If the run queue is full, runnext puts g on the global queue.
+// Executed only by the owner P.
+func runqput(_p_ *p, gp *g, next bool) {
+	if randomizeScheduler && next && fastrandn(2) == 0 {
+		next = false
+	}
+
+	if next {
+	retryNext:
+		oldnext := _p_.runnext
+		if !_p_.runnext.cas(oldnext, guintptr(unsafe.Pointer(gp))) {
+			goto retryNext
+		}
+		if oldnext == 0 {
+			return
+		}
+		// Kick the old runnext out to the regular run queue.
+		gp = oldnext.ptr()
+	}
+
+retry:
+	h := atomic.LoadAcq(&_p_.runqhead) // load-acquire, synchronize with consumers
+	t := _p_.runqtail
+	if t-h < uint32(len(_p_.runq)) {
+		_p_.runq[t%uint32(len(_p_.runq))].set(gp)
+		atomic.StoreRel(&_p_.runqtail, t+1) // store-release, makes the item available for consumption
+		return
+	}
+	if runqputslow(_p_, gp, h, t) {
+		return
+	}
+	// the queue is not full, now the put above must succeed
+	goto retry
+}
+
+// Put g and a batch of work from local runnable queue on global queue.
+// Executed only by the owner P.
+func runqputslow(_p_ *p, gp *g, h, t uint32) bool {
+	var batch [len(_p_.runq)/2 + 1]*g
+
+	// First, grab a batch from local queue.
+	n := t - h
+	n = n / 2
+	if n != uint32(len(_p_.runq)/2) {
+		throw("runqputslow: queue is not full")
+	}
+	for i := uint32(0); i < n; i++ {
+		batch[i] = _p_.runq[(h+i)%uint32(len(_p_.runq))].ptr()
+	}
+	if !atomic.CasRel(&_p_.runqhead, h, h+n) { // cas-release, commits consume
+		return false
+	}
+	batch[n] = gp
+
+	if randomizeScheduler {
+		for i := uint32(1); i <= n; i++ {
+			j := fastrandn(i + 1)
+			batch[i], batch[j] = batch[j], batch[i]
+		}
+	}
+
+	// Link the goroutines.
+	for i := uint32(0); i < n; i++ {
+		batch[i].schedlink.set(batch[i+1])
+	}
+	var q gQueue
+	q.head.set(batch[0])
+	q.tail.set(batch[n])
+
+	// Now put the batch on global queue.
+	lock(&sched.lock)
+	globrunqputbatch(&q, int32(n+1))
+	unlock(&sched.lock)
+	return true
+}
+
+// runqputbatch tries to put all the G's on q on the local runnable queue.
+// If the queue is full, they are put on the global queue; in that case
+// this will temporarily acquire the scheduler lock.
+// Executed only by the owner P.
+func runqputbatch(pp *p, q *gQueue, qsize int) {
+	h := atomic.LoadAcq(&pp.runqhead)
+	t := pp.runqtail
+	n := uint32(0)
+	for !q.empty() && t-h < uint32(len(pp.runq)) {
+		gp := q.pop()
+		pp.runq[t%uint32(len(pp.runq))].set(gp)
+		t++
+		n++
+	}
+	qsize -= int(n)
+
+	if randomizeScheduler {
+		off := func(o uint32) uint32 {
+			return (pp.runqtail + o) % uint32(len(pp.runq))
+		}
+		for i := uint32(1); i < n; i++ {
+			j := fastrandn(i + 1)
+			pp.runq[off(i)], pp.runq[off(j)] = pp.runq[off(j)], pp.runq[off(i)]
+		}
+	}
+
+	atomic.StoreRel(&pp.runqtail, t)
+	if !q.empty() {
+		lock(&sched.lock)
+		globrunqputbatch(q, int32(qsize))
+		unlock(&sched.lock)
+	}
+}
+
+// Get g from local runnable queue.
+// If inheritTime is true, gp should inherit the remaining time in the
+// current time slice. Otherwise, it should start a new time slice.
+// Executed only by the owner P.
+func runqget(_p_ *p) (gp *g, inheritTime bool) {
+	// If there's a runnext, it's the next G to run.
+	next := _p_.runnext
+	// If the runnext is non-0 and the CAS fails, it could only have been stolen by another P,
+	// because other Ps can race to set runnext to 0, but only the current P can set it to non-0.
+	// Hence, there's no need to retry this CAS if it falls.
+	if next != 0 && _p_.runnext.cas(next, 0) {
+		return next.ptr(), true
+	}
+
+	for {
+		h := atomic.LoadAcq(&_p_.runqhead) // load-acquire, synchronize with other consumers
+		t := _p_.runqtail
+		if t == h {
+			return nil, false
+		}
+		gp := _p_.runq[h%uint32(len(_p_.runq))].ptr()
+		if atomic.CasRel(&_p_.runqhead, h, h+1) { // cas-release, commits consume
+			return gp, false
+		}
+	}
+}
+
+// runqdrain drains the local runnable queue of _p_ and returns all goroutines in it.
+// Executed only by the owner P.
+func runqdrain(_p_ *p) (drainQ gQueue, n uint32) {
+	oldNext := _p_.runnext
+	if oldNext != 0 && _p_.runnext.cas(oldNext, 0) {
+		drainQ.pushBack(oldNext.ptr())
+		n++
+	}
+
+retry:
+	h := atomic.LoadAcq(&_p_.runqhead) // load-acquire, synchronize with other consumers
+	t := _p_.runqtail
+	qn := t - h
+	if qn == 0 {
+		return
+	}
+	if qn > uint32(len(_p_.runq)) { // read inconsistent h and t
+		goto retry
+	}
+
+	if !atomic.CasRel(&_p_.runqhead, h, h+qn) { // cas-release, commits consume
+		goto retry
+	}
+
+	// We've inverted the order in which it gets G's from the local P's runnable queue
+	// and then advances the head pointer because we don't want to mess up the statuses of G's
+	// while runqdrain() and runqsteal() are running in parallel.
+	// Thus we should advance the head pointer before draining the local P into a gQueue,
+	// so that we can update any gp.schedlink only after we take the full ownership of G,
+	// meanwhile, other P's can't access to all G's in local P's runnable queue and steal them.
+	// See https://groups.google.com/g/golang-dev/c/0pTKxEKhHSc/m/6Q85QjdVBQAJ for more details.
+	for i := uint32(0); i < qn; i++ {
+		gp := _p_.runq[(h+i)%uint32(len(_p_.runq))].ptr()
+		drainQ.pushBack(gp)
+		n++
+	}
+	return
+}
+
+// Grabs a batch of goroutines from _p_'s runnable queue into batch.
+// Batch is a ring buffer starting at batchHead.
+// Returns number of grabbed goroutines.
+// Can be executed by any P.
+func runqgrab(_p_ *p, batch *[256]guintptr, batchHead uint32, stealRunNextG bool) uint32 {
+	for {
+		h := atomic.LoadAcq(&_p_.runqhead) // load-acquire, synchronize with other consumers
+		t := atomic.LoadAcq(&_p_.runqtail) // load-acquire, synchronize with the producer
+		n := t - h
+		n = n - n/2
+		if n == 0 {
+			if stealRunNextG {
+				// Try to steal from _p_.runnext.
+				if next := _p_.runnext; next != 0 {
+					if _p_.status == _Prunning {
+						// Sleep to ensure that _p_ isn't about to run the g
+						// we are about to steal.
+						// The important use case here is when the g running
+						// on _p_ ready()s another g and then almost
+						// immediately blocks. Instead of stealing runnext
+						// in this window, back off to give _p_ a chance to
+						// schedule runnext. This will avoid thrashing gs
+						// between different Ps.
+						// A sync chan send/recv takes ~50ns as of time of
+						// writing, so 3us gives ~50x overshoot.
+						if GOOS != "windows" {
+							usleep(3)
+						} else {
+							// On windows system timer granularity is
+							// 1-15ms, which is way too much for this
+							// optimization. So just yield.
+							osyield()
+						}
+					}
+					if !_p_.runnext.cas(next, 0) {
+						continue
+					}
+					batch[batchHead%uint32(len(batch))] = next
+					return 1
+				}
+			}
+			return 0
+		}
+		if n > uint32(len(_p_.runq)/2) { // read inconsistent h and t
+			continue
+		}
+		for i := uint32(0); i < n; i++ {
+			g := _p_.runq[(h+i)%uint32(len(_p_.runq))]
+			batch[(batchHead+i)%uint32(len(batch))] = g
+		}
+		if atomic.CasRel(&_p_.runqhead, h, h+n) { // cas-release, commits consume
+			return n
+		}
+	}
+}
+
+// Steal half of elements from local runnable queue of p2
+// and put onto local runnable queue of p.
+// Returns one of the stolen elements (or nil if failed).
+func runqsteal(_p_, p2 *p, stealRunNextG bool) *g {
+	t := _p_.runqtail
+	n := runqgrab(p2, &_p_.runq, t, stealRunNextG)
+	if n == 0 {
+		return nil
+	}
+	n--
+	gp := _p_.runq[(t+n)%uint32(len(_p_.runq))].ptr()
+	if n == 0 {
+		return gp
+	}
+	h := atomic.LoadAcq(&_p_.runqhead) // load-acquire, synchronize with consumers
+	if t-h+n >= uint32(len(_p_.runq)) {
+		throw("runqsteal: runq overflow")
+	}
+	atomic.StoreRel(&_p_.runqtail, t+n) // store-release, makes the item available for consumption
+	return gp
+}
+
+// A gQueue is a dequeue of Gs linked through g.schedlink. A G can only
+// be on one gQueue or gList at a time.
+type gQueue struct {
+	head guintptr
+	tail guintptr
+}
+
+// empty reports whether q is empty.
+func (q *gQueue) empty() bool {
+	return q.head == 0
+}
+
+// push adds gp to the head of q.
+func (q *gQueue) push(gp *g) {
+	gp.schedlink = q.head
+	q.head.set(gp)
+	if q.tail == 0 {
+		q.tail.set(gp)
+	}
+}
+
+// pushBack adds gp to the tail of q.
+func (q *gQueue) pushBack(gp *g) {
+	gp.schedlink = 0
+	if q.tail != 0 {
+		q.tail.ptr().schedlink.set(gp)
+	} else {
+		q.head.set(gp)
+	}
+	q.tail.set(gp)
+}
+
+// pushBackAll adds all Gs in q2 to the tail of q. After this q2 must
+// not be used.
+func (q *gQueue) pushBackAll(q2 gQueue) {
+	if q2.tail == 0 {
+		return
+	}
+	q2.tail.ptr().schedlink = 0
+	if q.tail != 0 {
+		q.tail.ptr().schedlink = q2.head
+	} else {
+		q.head = q2.head
+	}
+	q.tail = q2.tail
+}
+
+// pop removes and returns the head of queue q. It returns nil if
+// q is empty.
+func (q *gQueue) pop() *g {
+	gp := q.head.ptr()
+	if gp != nil {
+		q.head = gp.schedlink
+		if q.head == 0 {
+			q.tail = 0
+		}
+	}
+	return gp
+}
+
+// popList takes all Gs in q and returns them as a gList.
+func (q *gQueue) popList() gList {
+	stack := gList{q.head}
+	*q = gQueue{}
+	return stack
+}
+
+// A gList is a list of Gs linked through g.schedlink. A G can only be
+// on one gQueue or gList at a time.
+type gList struct {
+	head guintptr
+}
+
+// empty reports whether l is empty.
+func (l *gList) empty() bool {
+	return l.head == 0
+}
+
+// push adds gp to the head of l.
+func (l *gList) push(gp *g) {
+	gp.schedlink = l.head
+	l.head.set(gp)
+}
+
+// pushAll prepends all Gs in q to l.
+func (l *gList) pushAll(q gQueue) {
+	if !q.empty() {
+		q.tail.ptr().schedlink = l.head
+		l.head = q.head
+	}
+}
+
+// pop removes and returns the head of l. If l is empty, it returns nil.
+func (l *gList) pop() *g {
+	gp := l.head.ptr()
+	if gp != nil {
+		l.head = gp.schedlink
+	}
+	return gp
+}
+
+//go:linkname setMaxThreads runtime/debug.setMaxThreads
+func setMaxThreads(in int) (out int) {
+	lock(&sched.lock)
+	out = int(sched.maxmcount)
+	if in > 0x7fffffff { // MaxInt32
+		sched.maxmcount = 0x7fffffff
+	} else {
+		sched.maxmcount = int32(in)
+	}
+	checkmcount()
+	unlock(&sched.lock)
+	return
+}
+
+//go:nosplit
+func procPin() int {
+	_g_ := getg()
+	mp := _g_.m
+
+	mp.locks++
+	return int(mp.p.ptr().id)
+}
+
+//go:nosplit
+func procUnpin() {
+	_g_ := getg()
+	_g_.m.locks--
+}
+
+//go:linkname sync_runtime_procPin sync.runtime_procPin
+//go:nosplit
+func sync_runtime_procPin() int {
+	return procPin()
+}
+
+//go:linkname sync_runtime_procUnpin sync.runtime_procUnpin
+//go:nosplit
+func sync_runtime_procUnpin() {
+	procUnpin()
+}
+
+//go:linkname sync_atomic_runtime_procPin sync/atomic.runtime_procPin
+//go:nosplit
+func sync_atomic_runtime_procPin() int {
+	return procPin()
+}
+
+//go:linkname sync_atomic_runtime_procUnpin sync/atomic.runtime_procUnpin
+//go:nosplit
+func sync_atomic_runtime_procUnpin() {
+	procUnpin()
+}
+
+// Active spinning for sync.Mutex.
+//go:linkname sync_runtime_canSpin sync.runtime_canSpin
+//go:nosplit
+func sync_runtime_canSpin(i int) bool {
+	// sync.Mutex is cooperative, so we are conservative with spinning.
+	// Spin only few times and only if running on a multicore machine and
+	// GOMAXPROCS>1 and there is at least one other running P and local runq is empty.
+	// As opposed to runtime mutex we don't do passive spinning here,
+	// because there can be work on global runq or on other Ps.
+	if i >= active_spin || ncpu <= 1 || gomaxprocs <= int32(sched.npidle+sched.nmspinning)+1 {
+		return false
+	}
+	if p := getg().m.p.ptr(); !runqempty(p) {
+		return false
+	}
+	return true
+}
+
+//go:linkname sync_runtime_doSpin sync.runtime_doSpin
+//go:nosplit
+func sync_runtime_doSpin() {
+	procyield(active_spin_cnt)
+}
+
+var stealOrder randomOrder
+
+// randomOrder/randomEnum are helper types for randomized work stealing.
+// They allow to enumerate all Ps in different pseudo-random orders without repetitions.
+// The algorithm is based on the fact that if we have X such that X and GOMAXPROCS
+// are coprime, then a sequences of (i + X) % GOMAXPROCS gives the required enumeration.
+type randomOrder struct {
+	count    uint32
+	coprimes []uint32
+}
+
+type randomEnum struct {
+	i     uint32
+	count uint32
+	pos   uint32
+	inc   uint32
+}
+
+func (ord *randomOrder) reset(count uint32) {
+	ord.count = count
+	ord.coprimes = ord.coprimes[:0]
+	for i := uint32(1); i <= count; i++ {
+		if gcd(i, count) == 1 {
+			ord.coprimes = append(ord.coprimes, i)
+		}
+	}
+}
+
+func (ord *randomOrder) start(i uint32) randomEnum {
+	return randomEnum{
+		count: ord.count,
+		pos:   i % ord.count,
+		inc:   ord.coprimes[i%uint32(len(ord.coprimes))],
+	}
+}
+
+func (enum *randomEnum) done() bool {
+	return enum.i == enum.count
+}
+
+func (enum *randomEnum) next() {
+	enum.i++
+	enum.pos = (enum.pos + enum.inc) % enum.count
+}
+
+func (enum *randomEnum) position() uint32 {
+	return enum.pos
+}
+
+func gcd(a, b uint32) uint32 {
+	for b != 0 {
+		a, b = b, a%b
+	}
+	return a
+}
+
+// An initTask represents the set of initializations that need to be done for a package.
+// Keep in sync with ../../test/initempty.go:initTask
+type initTask struct {
+	// TODO: pack the first 3 fields more tightly?
+	state uintptr // 0 = uninitialized, 1 = in progress, 2 = done
+	ndeps uintptr
+	nfns  uintptr
+	// followed by ndeps instances of an *initTask, one per package depended on
+	// followed by nfns pcs, one per init function to run
+}
+
+// inittrace stores statistics for init functions which are
+// updated by malloc and newproc when active is true.
+var inittrace tracestat
+
+type tracestat struct {
+	active bool   // init tracing activation status
+	id     int64  // init goroutine id
+	allocs uint64 // heap allocations
+	bytes  uint64 // heap allocated bytes
+}
+
+func doInit(t *initTask) {
+	switch t.state {
+	case 2: // fully initialized
+		return
+	case 1: // initialization in progress
+		throw("recursive call during initialization - linker skew")
+	default: // not initialized yet
+		t.state = 1 // initialization in progress
+
+		for i := uintptr(0); i < t.ndeps; i++ {
+			p := add(unsafe.Pointer(t), (3+i)*goarch.PtrSize)
+			t2 := *(**initTask)(p)
+			doInit(t2)
+		}
+
+		if t.nfns == 0 {
+			t.state = 2 // initialization done
+			return
+		}
+
+		var (
+			start  int64
+			before tracestat
+		)
+
+		if inittrace.active {
+			start = nanotime()
+			// Load stats non-atomically since tracinit is updated only by this init goroutine.
+			before = inittrace
+		}
+
+		firstFunc := add(unsafe.Pointer(t), (3+t.ndeps)*goarch.PtrSize)
+		for i := uintptr(0); i < t.nfns; i++ {
+			p := add(firstFunc, i*goarch.PtrSize)
+			f := *(*func())(unsafe.Pointer(&p))
+			f()
+		}
+
+		if inittrace.active {
+			end := nanotime()
+			// Load stats non-atomically since tracinit is updated only by this init goroutine.
+			after := inittrace
+
+			f := *(*func())(unsafe.Pointer(&firstFunc))
+			pkg := funcpkgpath(findfunc(abi.FuncPCABIInternal(f)))
+
+			var sbuf [24]byte
+			print("init ", pkg, " @")
+			print(string(fmtNSAsMS(sbuf[:], uint64(start-runtimeInitTime))), " ms, ")
+			print(string(fmtNSAsMS(sbuf[:], uint64(end-start))), " ms clock, ")
+			print(string(itoa(sbuf[:], after.bytes-before.bytes)), " bytes, ")
+			print(string(itoa(sbuf[:], after.allocs-before.allocs)), " allocs")
+			print("\n")
+		}
+
+		t.state = 2 // initialization done
+	}
+}