summaryrefslogtreecommitdiffstats
path: root/src/runtime/netpoll_kqueue.go
blob: 3af45e689261639053c8e9f2ffe07f142a1e72ef (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
// Copyright 2013 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.

//go:build darwin || dragonfly || freebsd || netbsd || openbsd

package runtime

// Integrated network poller (kqueue-based implementation).

import (
	"internal/goarch"
	"runtime/internal/atomic"
	"unsafe"
)

var (
	kq int32 = -1

	netpollBreakRd, netpollBreakWr uintptr // for netpollBreak

	netpollWakeSig atomic.Uint32 // used to avoid duplicate calls of netpollBreak
)

func netpollinit() {
	kq = kqueue()
	if kq < 0 {
		println("runtime: kqueue failed with", -kq)
		throw("runtime: netpollinit failed")
	}
	closeonexec(kq)
	r, w, errno := nonblockingPipe()
	if errno != 0 {
		println("runtime: pipe failed with", -errno)
		throw("runtime: pipe failed")
	}
	ev := keventt{
		filter: _EVFILT_READ,
		flags:  _EV_ADD,
	}
	*(*uintptr)(unsafe.Pointer(&ev.ident)) = uintptr(r)
	n := kevent(kq, &ev, 1, nil, 0, nil)
	if n < 0 {
		println("runtime: kevent failed with", -n)
		throw("runtime: kevent failed")
	}
	netpollBreakRd = uintptr(r)
	netpollBreakWr = uintptr(w)
}

func netpollIsPollDescriptor(fd uintptr) bool {
	return fd == uintptr(kq) || fd == netpollBreakRd || fd == netpollBreakWr
}

func netpollopen(fd uintptr, pd *pollDesc) int32 {
	// Arm both EVFILT_READ and EVFILT_WRITE in edge-triggered mode (EV_CLEAR)
	// for the whole fd lifetime. The notifications are automatically unregistered
	// when fd is closed.
	var ev [2]keventt
	*(*uintptr)(unsafe.Pointer(&ev[0].ident)) = fd
	ev[0].filter = _EVFILT_READ
	ev[0].flags = _EV_ADD | _EV_CLEAR
	ev[0].fflags = 0
	ev[0].data = 0

	if goarch.PtrSize == 4 {
		// We only have a pointer-sized field to store into,
		// so on a 32-bit system we get no sequence protection.
		// TODO(iant): If we notice any problems we could at least
		// steal the low-order 2 bits for a tiny sequence number.
		ev[0].udata = (*byte)(unsafe.Pointer(pd))
	} else {
		tp := taggedPointerPack(unsafe.Pointer(pd), pd.fdseq.Load())
		ev[0].udata = (*byte)(unsafe.Pointer(uintptr(tp)))
	}
	ev[1] = ev[0]
	ev[1].filter = _EVFILT_WRITE
	n := kevent(kq, &ev[0], 2, nil, 0, nil)
	if n < 0 {
		return -n
	}
	return 0
}

func netpollclose(fd uintptr) int32 {
	// Don't need to unregister because calling close()
	// on fd will remove any kevents that reference the descriptor.
	return 0
}

func netpollarm(pd *pollDesc, mode int) {
	throw("runtime: unused")
}

// netpollBreak interrupts a kevent.
func netpollBreak() {
	// Failing to cas indicates there is an in-flight wakeup, so we're done here.
	if !netpollWakeSig.CompareAndSwap(0, 1) {
		return
	}

	for {
		var b byte
		n := write(netpollBreakWr, unsafe.Pointer(&b), 1)
		if n == 1 || n == -_EAGAIN {
			break
		}
		if n == -_EINTR {
			continue
		}
		println("runtime: netpollBreak write failed with", -n)
		throw("runtime: netpollBreak write failed")
	}
}

// netpoll checks for ready network connections.
// Returns list of goroutines that become runnable.
// delay < 0: blocks indefinitely
// delay == 0: does not block, just polls
// delay > 0: block for up to that many nanoseconds
func netpoll(delay int64) gList {
	if kq == -1 {
		return gList{}
	}
	var tp *timespec
	var ts timespec
	if delay < 0 {
		tp = nil
	} else if delay == 0 {
		tp = &ts
	} else {
		ts.setNsec(delay)
		if ts.tv_sec > 1e6 {
			// Darwin returns EINVAL if the sleep time is too long.
			ts.tv_sec = 1e6
		}
		tp = &ts
	}
	var events [64]keventt
retry:
	n := kevent(kq, nil, 0, &events[0], int32(len(events)), tp)
	if n < 0 {
		if n != -_EINTR {
			println("runtime: kevent on fd", kq, "failed with", -n)
			throw("runtime: netpoll failed")
		}
		// If a timed sleep was interrupted, just return to
		// recalculate how long we should sleep now.
		if delay > 0 {
			return gList{}
		}
		goto retry
	}
	var toRun gList
	for i := 0; i < int(n); i++ {
		ev := &events[i]

		if uintptr(ev.ident) == netpollBreakRd {
			if ev.filter != _EVFILT_READ {
				println("runtime: netpoll: break fd ready for", ev.filter)
				throw("runtime: netpoll: break fd ready for something unexpected")
			}
			if delay != 0 {
				// netpollBreak could be picked up by a
				// nonblocking poll. Only read the byte
				// if blocking.
				var tmp [16]byte
				read(int32(netpollBreakRd), noescape(unsafe.Pointer(&tmp[0])), int32(len(tmp)))
				netpollWakeSig.Store(0)
			}
			continue
		}

		var mode int32
		switch ev.filter {
		case _EVFILT_READ:
			mode += 'r'

			// On some systems when the read end of a pipe
			// is closed the write end will not get a
			// _EVFILT_WRITE event, but will get a
			// _EVFILT_READ event with EV_EOF set.
			// Note that setting 'w' here just means that we
			// will wake up a goroutine waiting to write;
			// that goroutine will try the write again,
			// and the appropriate thing will happen based
			// on what that write returns (success, EPIPE, EAGAIN).
			if ev.flags&_EV_EOF != 0 {
				mode += 'w'
			}
		case _EVFILT_WRITE:
			mode += 'w'
		}
		if mode != 0 {
			var pd *pollDesc
			var tag uintptr
			if goarch.PtrSize == 4 {
				// No sequence protection on 32-bit systems.
				// See netpollopen for details.
				pd = (*pollDesc)(unsafe.Pointer(ev.udata))
				tag = 0
			} else {
				tp := taggedPointer(uintptr(unsafe.Pointer(ev.udata)))
				pd = (*pollDesc)(tp.pointer())
				tag = tp.tag()
				if pd.fdseq.Load() != tag {
					continue
				}
			}
			pd.setEventErr(ev.flags == _EV_ERROR, tag)
			netpollready(&toRun, pd, mode)
		}
	}
	return toRun
}