Compare commits

..
Author SHA1 Message Date
JackDoan fd29ebdacd use a trampoline to get to the spicy _x functions 2026-09-25 13:01:50 -05:00
JackDoan 2ff4852551 split tun setup and tun IO 2026-09-25 12:26:07 -05:00
Caleb JasikandClaude Opus 5.5 10e1924489 udp/darwin: batch sends with sendmsg_x
WriteBatch hands up to 64 datagrams to one sendmsg_x call instead of
one sendto each. The socket is unconnected, so xnu still runs every
message through sendit, per-send MAC checks included; the saving is the
syscall crossings.

xnu's per-message loop stops at the first message that fails. For
EWOULDBLOCK, ENOBUFS and EMSGSIZE it reports how many went out, or the
errno alone when none did; for any other error it returns only the
errno, with the messages before it already sent. So each call carries
one run of datagrams to a single destination, and nothing is resent
after the kernel may have taken it. A datagram that fails on its own
account (a full interface queue, an oversize packet) goes to sendto,
which delivers or reports it, so it costs only itself, as without
batching; a refused destination costs its run. ENOSYS/EOPNOTSUPP, or
EPERM/EINVAL on a datagram sendto then delivers, switch the socket
back to sendto for good.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-24 16:10:26 -05:00
Caleb JasikandClaude Opus 5.5 4d191f805f overlay/darwin: drain several tun packets per read wakeup
The darwin tun returned one packet per Read, and listenIn flushes its
UDP send batch after every Read, so each outbound packet went to the
underlay as its own batch. tunQueue.Read now keeps calling readv until
the utun would block, up to tunReadBatch packets, so a burst reaches the
send batch together.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-24 12:44:00 -05:00
13 changed files with 930 additions and 371 deletions

No files matched your search

+5
View File
@@ -0,0 +1,5 @@
// Package msgx calls xnu's private batched socket calls, sendmsg_x and recvmsg_x, through libSystem.
//
// The calls are not public API, so the package never links them: it looks them up with dlsym on first use.
// The package is darwin-only, and on iOS Available is always false, so App Store builds keep to public API.
package msgx
+16
View File
@@ -0,0 +1,16 @@
package msgx
import "golang.org/x/sys/unix"
// Hdr mirrors xnu's struct msghdr_x (bsd/sys/socket_private.h), which the SDK doesn't ship.
// recvmsg_x reports each message's length in Datalen rather than in its return value.
type Hdr struct {
Name *byte
Namelen uint32
Iov *unix.Iovec
Iovlen int32
Control *byte
Controllen uint32
Flags int32
Datalen uint64
}
+77
View File
@@ -0,0 +1,77 @@
//go:build !ios
package msgx
import (
"sync"
"syscall"
"unsafe"
)
// dlsym is public libSystem API, imported the way x/sys/unix imports its libSystem stubs. The local name is
// distinct from runtime/race's dlsym import.
//
//go:cgo_import_dynamic msgx_dlsym dlsym "/usr/lib/libSystem.B.dylib"
// dlsymTrampolineAddr is the address of msgx_darwin.s's jump to dlsym.
var dlsymTrampolineAddr uintptr
// syscall6X calls the C function at fn on the system stack and returns errno when it returns a 64-bit -1.
// It is how x/sys/unix calls libSystem, and the runtime keeps it linkable for that.
//
//go:linkname syscall6X syscall.syscall6X
func syscall6X(fn, a1, a2, a3, a4, a5, a6 uintptr) (r1, r2 uintptr, err syscall.Errno)
// rtldDefault is dlfcn.h's RTLD_DEFAULT, ((void *)-2): search every image loaded in the process.
const rtldDefault = ^uintptr(1)
var (
sendmsgXName = []byte("sendmsg_x\x00")
recvmsgXName = []byte("recvmsg_x\x00")
)
type funcs struct {
sendmsgX uintptr
recvmsgX uintptr
}
var resolve = sync.OnceValue(func() funcs {
return funcs{sendmsgX: lookup(sendmsgXName), recvmsgX: lookup(recvmsgXName)}
})
// lookup returns the address of the NUL-terminated C symbol name, or 0 when no loaded image exports it.
func lookup(name []byte) uintptr {
p, _, _ := syscall6X(dlsymTrampolineAddr, rtldDefault, uintptr(unsafe.Pointer(&name[0])), 0, 0, 0, 0)
return p
}
// Available reports whether libSystem exports both sendmsg_x and recvmsg_x. When it is false, Send and Recv
// return ENOSYS.
func Available() bool {
f := resolve()
return f.sendmsgX != 0 && f.recvmsgX != 0
}
// Send passes hdrs to sendmsg_x on fd and returns how many messages the kernel took.
func Send(fd uintptr, hdrs []Hdr, flags int) (int, syscall.Errno) {
return call(resolve().sendmsgX, fd, hdrs, flags)
}
// Recv passes hdrs to recvmsg_x on fd and returns how many messages it filled.
func Recv(fd uintptr, hdrs []Hdr, flags int) (int, syscall.Errno) {
return call(resolve().recvmsgX, fd, hdrs, flags)
}
func call(fn, fd uintptr, hdrs []Hdr, flags int) (int, syscall.Errno) {
if fn == 0 {
return 0, syscall.ENOSYS
}
if len(hdrs) == 0 {
return 0, syscall.EINVAL
}
r, _, errno := syscall6X(fn, fd, uintptr(unsafe.Pointer(&hdrs[0])), uintptr(len(hdrs)), uintptr(flags), 0, 0)
if errno != 0 {
return 0, errno
}
return int(r), 0
}
+9
View File
@@ -0,0 +1,9 @@
//go:build !ios
#include "textflag.h"
// The same trampoline shape x/sys/unix uses for its libSystem stubs; it assembles for amd64 and arm64 alike.
TEXT msgx_dlsym_trampoline<>(SB),NOSPLIT,$0-0
JMP msgx_dlsym(SB)
GLOBL ·dlsymTrampolineAddr(SB), RODATA, $8
DATA ·dlsymTrampolineAddr(SB)/8, $msgx_dlsym_trampoline<>(SB)
+69
View File
@@ -0,0 +1,69 @@
//go:build !ios
package msgx
import (
"bytes"
"syscall"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"golang.org/x/sys/unix"
)
// TestAvailable pins that dlsym finds both calls in this macOS's libSystem, so the batched paths are live.
func TestAvailable(t *testing.T) {
assert.True(t, Available())
}
// TestSendRecvRoundTrip pins the libSystem calling path end to end: a batch goes out through sendmsg_x in one
// call and comes back through recvmsg_x in one call, each message whole, in order, with Datalen set.
func TestSendRecvRoundTrip(t *testing.T) {
fds, err := unix.Socketpair(unix.AF_UNIX, unix.SOCK_DGRAM, 0)
require.NoError(t, err)
t.Cleanup(func() {
_ = unix.Close(fds[0])
_ = unix.Close(fds[1])
})
// A dropped datagram fails the receive instead of hanging it.
require.NoError(t, unix.SetsockoptTimeval(fds[1], unix.SOL_SOCKET, unix.SO_RCVTIMEO, &unix.Timeval{Sec: 2}))
const n = 5
var want [n][]byte
var iovs [n]unix.Iovec
hdrs := make([]Hdr, n)
for i := range n {
want[i] = bytes.Repeat([]byte{byte(i + 1)}, 10+i)
iovs[i] = unix.Iovec{Base: &want[i][0]}
iovs[i].SetLen(len(want[i]))
hdrs[i] = Hdr{Iov: &iovs[i], Iovlen: 1}
}
sent, errno := Send(uintptr(fds[0]), hdrs, 0)
require.Zero(t, errno)
require.Equal(t, n, sent)
var bufs [n + 1][64]byte
var riovs [n + 1]unix.Iovec
rhdrs := make([]Hdr, n+1)
for i := range rhdrs {
riovs[i] = unix.Iovec{Base: &bufs[i][0]}
riovs[i].SetLen(len(bufs[i]))
rhdrs[i] = Hdr{Iov: &riovs[i], Iovlen: 1}
}
got, errno := Recv(uintptr(fds[1]), rhdrs, 0)
require.Zero(t, errno)
require.Equal(t, n, got)
for i := range n {
l := int(rhdrs[i].Datalen)
assert.Equal(t, want[i], bufs[i][:l], "message %d", i)
}
}
// TestEmptyBatch pins that an empty batch is refused before it reaches the kernel.
func TestEmptyBatch(t *testing.T) {
_, errno := Send(0, nil, 0)
assert.Equal(t, syscall.EINVAL, errno)
_, errno = Recv(0, nil, 0)
assert.Equal(t, syscall.EINVAL, errno)
}
+12
View File
@@ -0,0 +1,12 @@
package msgx
import "syscall"
// Available is always false on iOS, so App Store builds never look up private symbols.
func Available() bool { return false }
// Send always returns ENOSYS on iOS.
func Send(uintptr, []Hdr, int) (int, syscall.Errno) { return 0, syscall.ENOSYS }
// Recv always returns ENOSYS on iOS.
func Recv(uintptr, []Hdr, int) (int, syscall.Errno) { return 0, syscall.ENOSYS }
+2 -17
View File
@@ -15,13 +15,9 @@ type singleQueue struct {
// NewSingleQueue wraps a one-datagram-per-Read ReadWriteCloser (a legacy tun device) into a Queue. // NewSingleQueue wraps a one-datagram-per-Read ReadWriteCloser (a legacy tun device) into a Queue.
// bufSize is the per-queue read scratch size and must be at least the largest datagram the source can return. // bufSize is the per-queue read scratch size and must be at least the largest datagram the source can return.
// Close closes rwc. The Queue implements BatchWriter when rwc does. // Close closes rwc.
func NewSingleQueue(rwc io.ReadWriteCloser, bufSize int) Queue { func NewSingleQueue(rwc io.ReadWriteCloser, bufSize int) Queue {
q := &singleQueue{rw: rwc, closer: rwc, buf: make([]byte, bufSize)} return &singleQueue{rw: rwc, closer: rwc, buf: make([]byte, bufSize)}
if bw, ok := rwc.(BatchWriter); ok {
return &batchSingleQueue{singleQueue: q, bw: bw}
}
return q
} }
// NewSingleQueueNoClose is NewSingleQueue for a source owned by someone else, // NewSingleQueueNoClose is NewSingleQueue for a source owned by someone else,
@@ -51,14 +47,3 @@ func (q *singleQueue) Close() error {
} }
return q.closer.Close() return q.closer.Close()
} }
// batchSingleQueue is a singleQueue whose source also implements BatchWriter.
// It is a separate type so a plain singleQueue never advertises WriteBatch.
type batchSingleQueue struct {
*singleQueue
bw BatchWriter
}
func (q *batchSingleQueue) WriteBatch(pkts [][]byte) error {
return q.bw.WriteBatch(pkts)
}
+328
View File
@@ -0,0 +1,328 @@
//go:build darwin && !ios
package tio
import (
"fmt"
"log/slog"
"os"
"sync"
"sync/atomic"
"syscall"
_ "unsafe" // for go:linkname
"github.com/slackhq/nebula/internal/msgx"
"golang.org/x/sys/unix"
)
// tunMaxPacket is the largest packet utun can return at any device MTU.
const tunMaxPacket = 65535
// tunReadBatch is the most packets one Utun.Read returns.
const tunReadBatch = 64
// tunReadArena is Utun's receive buffer. Draining stops once less than tunMaxPacket of it is left,
// so every readv has room for the largest packet utun can return.
const tunReadArena = 4 * tunMaxPacket
// tunWriteBatch is how many packets one sendmsg_x call may carry.
const tunWriteBatch = 64
// Utun is a darwin utun control socket as a Queue. Every packet crosses the socket behind a 4 byte
// address family prefix, which Read strips and the writes add.
//
// The fd lives in the runtime poller, so Close wakes a Read parked on it; darwin needs no QueueSet
// shutdown fd.
type Utun struct {
f *os.File
l *slog.Logger
// readBuf and readRet are Read's scratch, owned by the single reader.
readBuf []byte
readRet [tunReadBatch]Packet
// noSendmsgX is set when libSystem lacks sendmsg_x or the kernel refuses it; WriteBatch then writes one packet at a time.
noSendmsgX atomic.Bool
batchMu sync.Mutex
batch tunBatch
}
// tunBatch is WriteBatch's scratch, guarded by Utun.batchMu. The kernel reads every entry in place.
type tunBatch struct {
heads [tunWriteBatch][4]byte
iovs [tunWriteBatch][2]unix.Iovec
hdrs [tunWriteBatch]msgx.Hdr
pkts [tunWriteBatch][]byte
}
// NewUtun wraps a utun control socket fd and makes it non-blocking.
// On failure it does NOT close fd: the caller owns fd until NewUtun succeeds, after which Close closes it.
func NewUtun(fd int, l *slog.Logger) (*Utun, error) {
// Before os.NewFile, which only hands a non-blocking fd to the runtime poller.
if err := unix.SetNonblock(fd, true); err != nil {
return nil, fmt.Errorf("failed to set the tun fd to non-blocking mode: %w", err)
}
u := &Utun{
f: os.NewFile(uintptr(fd), "utun"),
l: l,
readBuf: make([]byte, tunReadArena),
}
u.noSendmsgX.Store(!msgx.Available())
return u, nil
}
// tunWritev and tunReadv are linkname'd to x/sys/unix's libc-routed writev/readv stubs so the
// calls go through libSystem's pinned trampoline. A raw syscall.Syscall(SYS_WRITEV/SYS_READV, ...)
// on darwin/arm64 emits an SVC #0x80 trap (see $GOROOT/src/syscall/asm_darwin_arm64.s), the path
// Apple keeps warning they will eventually disallow. We pull the low-level stubs instead of calling
// unix.Writev/unix.Readv because those take [][]byte and rebuild the []Iovec every call, which
// heap-allocates the header; linkname'ing the stubs lets us hand them our own stack-allocated
// iovecs. See golang/go#78049.
//go:linkname tunWritev golang.org/x/sys/unix.writev
//go:noescape
func tunWritev(fd int, iovecs []unix.Iovec) (n int, err error)
//go:linkname tunReadv golang.org/x/sys/unix.readv
//go:noescape
func tunReadv(fd int, iovecs []unix.Iovec) (n int, err error)
// Read waits for the utun to become readable, then reads packets until it would block, up to
// tunReadBatch, so the caller encrypts and sends them as one batch rather than one per wakeup. Each
// packet is read with its AF prefix scattered away, so the payload lands directly in readBuf.
// An error after at least one packet is dropped in favor of returning those packets; a persistent
// one recurs on the next Read.
func (u *Utun) Read() ([]Packet, error) {
rc, err := u.f.SyscallConn()
if err != nil {
return nil, err
}
var head [4]byte
n, off := 0, 0
var callErr error
err = rc.Read(func(fd uintptr) bool {
for n < tunReadBatch && len(u.readBuf)-off >= tunMaxPacket {
iovecs := [2]unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &u.readBuf[off], Len: uint64(len(u.readBuf) - off)},
}
l, e := tunReadv(int(fd), iovecs[:])
if e != nil {
if errno, ok := e.(syscall.Errno); ok && errno.Temporary() {
// Park on the poller only while there is nothing to hand back.
return n > 0
}
callErr = e
return true
}
if l < 4 {
// A datagram too short to carry the AF prefix, or end of file; stop rather than spin on it.
return true
}
end := off + l - 4
u.readRet[n] = Packet{Bytes: u.readBuf[off:end:end]}
n++
off = end
}
return true
})
if n > 0 {
return u.readRet[:n], nil
}
if err != nil {
return nil, err
}
return nil, callErr
}
// Write pushes one IP packet onto the utun device. Safe for concurrent use:
// the AF prefix and iovecs are per-call stack state, and the fd write itself
// serializes on the runtime's fd mutex (see the Queue contract in tio.go).
func (u *Utun) Write(from []byte) (int, error) {
if len(from) == 0 {
return 0, syscall.EIO
}
var head [4]byte
af, err := tunAF(from)
if err != nil {
return 0, err
}
head[3] = af
// Grab rc as a local so the compiler can devirtualize the call and keep the closure on the stack.
rc, err := u.f.SyscallConn()
if err != nil {
return 0, err
}
var n int
var callErr error
err = rc.Write(func(fd uintptr) bool {
iovecs := []unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &from[0], Len: uint64(len(from))},
}
n, callErr = tunWritev(int(fd), iovecs)
// Type-assert to syscall.Errno so the EAGAIN/EWOULDBLOCK/EINTR check doesn't box the errno
// constants into error interfaces on every call.
if errno, ok := callErr.(syscall.Errno); ok && errno.Temporary() {
return false
}
return true
})
if err != nil {
return 0, err
}
if callErr != nil {
return 0, callErr
}
return n - 4, nil
}
// tunAF returns the utun address-family prefix byte for an IP packet.
func tunAF(pkt []byte) (byte, error) {
switch pkt[0] >> 4 {
case 4:
return syscall.AF_INET, nil
case 6:
return syscall.AF_INET6, nil
default:
return 0, fmt.Errorf("unable to determine IP version from packet")
}
}
// WriteBatch writes pkts to the utun device with sendmsg_x, xnu's private batched sendmsg, up to
// tunWriteBatch packets per syscall. The kernel still hands each packet to utun on its own, so this
// saves syscalls, not per-packet kernel work. Safe for concurrent use.
//
// An empty packet or one with no IP version is skipped and reported, as Write would. Any other
// error from sendmsg_x means the kernel took an unknown prefix of that call and dropped the rest, so
// it is reported rather than retried, which could deliver packets twice.
//
// Under mbuf exhaustion, sendmsg_x silently drops packets that writev would have delivered: the
// kernel counts each packet as sent once it is copied in, and when a later allocation fails it
// frees that prefix unsent and returns its length as a short count, which is indistinguishable
// from a real one.
func (u *Utun) WriteBatch(pkts [][]byte) error {
if u.noSendmsgX.Load() {
return u.writeEach(pkts)
}
u.batchMu.Lock()
defer u.batchMu.Unlock()
b := &u.batch
var firstErr error
for len(pkts) > 0 {
n := 0
for len(pkts) > 0 && n < tunWriteBatch {
p := pkts[0]
pkts = pkts[1:]
var af byte
err := error(syscall.EIO)
if len(p) > 0 {
af, err = tunAF(p)
}
if err != nil {
if firstErr == nil {
firstErr = err
}
continue
}
b.heads[n] = [4]byte{3: af}
b.iovs[n] = [2]unix.Iovec{
{Base: &b.heads[n][0], Len: 4},
{Base: &p[0], Len: uint64(len(p))},
}
b.hdrs[n] = msgx.Hdr{Iov: &b.iovs[n][0], Iovlen: 2}
b.pkts[n] = p
n++
}
if n == 0 {
continue
}
for off := 0; off < n; {
sent, err := u.sendmsgX(off, n)
switch {
case err == nil && sent > 0:
// The kernel stops early without an error on a packet larger than the socket's send
// buffer, or when sending a later packet fails with ENOBUFS; resend the remainder.
off += sent
case err == nil:
if werr := u.writeEach(b.pkts[off:n]); werr != nil && firstErr == nil {
firstErr = werr
}
off = n
case err == unix.EMSGSIZE:
// Only the first packet being larger than the socket's send buffer fails the whole call,
// before the kernel takes any; Write reports that one and the rest go out in the next call.
if werr := u.writeEach(b.pkts[off : off+1]); werr != nil && firstErr == nil {
firstErr = werr
}
off++
case err == unix.ENOSYS || err == unix.EPERM || err == unix.EOPNOTSUPP:
// sendmsg_x is private API: fall back to writev if a kernel or sandbox refuses it.
if u.noSendmsgX.CompareAndSwap(false, true) {
u.l.Warn("sendmsg_x unavailable on the tun device, writing one packet per syscall", "error", err)
}
if werr := u.writeEach(b.pkts[off:n]); werr != nil && firstErr == nil {
firstErr = werr
}
clear(b.iovs[:n])
clear(b.pkts[:n])
if werr := u.writeEach(pkts); werr != nil && firstErr == nil {
firstErr = werr
}
return firstErr
default:
if firstErr == nil {
firstErr = err
}
off = n
}
}
clear(b.iovs[:n])
clear(b.pkts[:n])
}
return firstErr
}
// sendmsgX hands entries off through n-1 of u.batch to sendmsg_x and returns how many the kernel took.
func (u *Utun) sendmsgX(off, n int) (int, error) {
rc, err := u.f.SyscallConn()
if err != nil {
return 0, err
}
var sent int
var errno syscall.Errno
err = rc.Write(func(fd uintptr) bool {
sent, errno = msgx.Send(fd, u.batch.hdrs[off:n], 0)
// sendmsg_x reports EAGAIN only when it took nothing; a partial batch returns its count.
return !errno.Temporary()
})
if err != nil {
return 0, err
}
if errno != 0 {
return 0, errno
}
return sent, nil
}
// writeEach writes pkts one at a time with Write.
func (u *Utun) writeEach(pkts [][]byte) error {
var firstErr error
for _, p := range pkts {
if _, err := u.Write(p); err != nil && firstErr == nil {
firstErr = err
}
}
return firstErr
}
func (u *Utun) Close() error {
return u.f.Close()
}
@@ -1,37 +1,42 @@
//go:build !ios && !e2e_testing //go:build darwin && !ios
package overlay package tio
import ( import (
"bytes" "bytes"
"os" "log/slog"
"syscall" "syscall"
"testing" "testing"
"github.com/slackhq/nebula/test"
"github.com/stretchr/testify/assert" "github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require" "github.com/stretchr/testify/require"
"golang.org/x/sys/unix" "golang.org/x/sys/unix"
) )
// newSocketpairTun returns a tun writing to one end of a connected AF_UNIX datagram socketpair, which // newSocketpairUtun returns a Utun over one end of a connected AF_UNIX datagram socketpair, which
// takes sendmsg_x the way a utun control socket does, and the other end's fd for reading back. // takes sendmsg_x the way a utun control socket does, and the other end's fd for the test to use.
// sndbuf is the writer's send buffer, which caps the largest datagram it accepts. // sndbuf is the Utun's send buffer, which caps the largest datagram it accepts.
func newSocketpairTun(t *testing.T, sndbuf int) (*tun, int) { func newSocketpairUtun(t *testing.T, sndbuf int) (*Utun, int) {
t.Helper() t.Helper()
fds, err := unix.Socketpair(unix.AF_UNIX, unix.SOCK_DGRAM, 0) fds, err := unix.Socketpair(unix.AF_UNIX, unix.SOCK_DGRAM, 0)
require.NoError(t, err) require.NoError(t, err)
require.NoError(t, unix.SetsockoptInt(fds[0], unix.SOL_SOCKET, unix.SO_SNDBUF, sndbuf)) require.NoError(t, unix.SetsockoptInt(fds[0], unix.SOL_SOCKET, unix.SO_SNDBUF, sndbuf))
// An AF_UNIX datagram send is bounded by the receiver's buffer, so size both ends' for the tests.
require.NoError(t, unix.SetsockoptInt(fds[0], unix.SOL_SOCKET, unix.SO_RCVBUF, 1<<20))
require.NoError(t, unix.SetsockoptInt(fds[1], unix.SOL_SOCKET, unix.SO_RCVBUF, 1<<20)) require.NoError(t, unix.SetsockoptInt(fds[1], unix.SOL_SOCKET, unix.SO_RCVBUF, 1<<20))
// A dropped datagram fails the read-back instead of hanging it. // A dropped datagram fails the read-back instead of hanging it.
require.NoError(t, unix.SetsockoptTimeval(fds[1], unix.SOL_SOCKET, unix.SO_RCVTIMEO, &unix.Timeval{Sec: 2})) require.NoError(t, unix.SetsockoptTimeval(fds[1], unix.SOL_SOCKET, unix.SO_RCVTIMEO, &unix.Timeval{Sec: 2}))
require.NoError(t, unix.SetNonblock(fds[0], true)) u, err := NewUtun(fds[0], slog.New(slog.DiscardHandler))
f := os.NewFile(uintptr(fds[0]), "socketpair") if err != nil {
_ = unix.Close(fds[0])
_ = unix.Close(fds[1])
t.Fatal(err)
}
t.Cleanup(func() { t.Cleanup(func() {
_ = f.Close() _ = u.Close()
_ = unix.Close(fds[1]) _ = unix.Close(fds[1])
}) })
return &tun{f: f, l: test.NewLogger()}, fds[1] return u, fds[1]
} }
// testTunPkt returns packet i of size bytes, alternating IPv4 and IPv6, and the datagram the utun // testTunPkt returns packet i of size bytes, alternating IPv4 and IPv6, and the datagram the utun
@@ -75,12 +80,12 @@ func forEachTunWritePath(t *testing.T, fn func(t *testing.T, fallback bool)) {
} }
} }
// TestTunWriteBatch pins that WriteBatch delivers every packet whole, in order and behind the right // TestUtunWriteBatch pins that WriteBatch delivers every packet whole, in order and behind the right
// utun AF prefix, across several sendmsg_x calls, and returns nil. // utun AF prefix, across several sendmsg_x calls, and returns nil.
func TestTunWriteBatch(t *testing.T) { func TestUtunWriteBatch(t *testing.T) {
forEachTunWritePath(t, func(t *testing.T, fallback bool) { forEachTunWritePath(t, func(t *testing.T, fallback bool) {
tn, r := newSocketpairTun(t, 1<<20) u, r := newSocketpairUtun(t, 1<<20)
tn.noSendmsgX.Store(fallback) u.noSendmsgX.Store(fallback)
var pkts, want [][]byte var pkts, want [][]byte
for i := range 3*tunWriteBatch + 5 { for i := range 3*tunWriteBatch + 5 {
@@ -88,18 +93,18 @@ func TestTunWriteBatch(t *testing.T) {
pkts = append(pkts, p) pkts = append(pkts, p)
want = append(want, w) want = append(want, w)
} }
require.NoError(t, tn.WriteBatch(pkts)) require.NoError(t, u.WriteBatch(pkts))
assert.Equal(t, fallback, tn.noSendmsgX.Load()) assert.Equal(t, fallback, u.noSendmsgX.Load())
readTunBack(t, r, want) readTunBack(t, r, want)
}) })
} }
// TestTunWriteBatchSkipsInvalid pins that an empty packet or one with no IP version is skipped // TestUtunWriteBatchSkipsInvalid pins that an empty packet or one with no IP version is skipped
// without shifting the packets after it, and that the first such error is the one reported. // without shifting the packets after it, and that the first such error is the one reported.
func TestTunWriteBatchSkipsInvalid(t *testing.T) { func TestUtunWriteBatchSkipsInvalid(t *testing.T) {
forEachTunWritePath(t, func(t *testing.T, fallback bool) { forEachTunWritePath(t, func(t *testing.T, fallback bool) {
tn, r := newSocketpairTun(t, 1<<20) u, r := newSocketpairUtun(t, 1<<20)
tn.noSendmsgX.Store(fallback) u.noSendmsgX.Store(fallback)
var pkts, want [][]byte var pkts, want [][]byte
for i := range 20 { for i := range 20 {
@@ -109,16 +114,16 @@ func TestTunWriteBatchSkipsInvalid(t *testing.T) {
} }
pkts = append(pkts[:10], append([][]byte{{0x10, 1, 2}, {}}, pkts[10:]...)...) pkts = append(pkts[:10], append([][]byte{{0x10, 1, 2}, {}}, pkts[10:]...)...)
err := tn.WriteBatch(pkts) err := u.WriteBatch(pkts)
assert.ErrorContains(t, err, "IP version") assert.ErrorContains(t, err, "IP version")
readTunBack(t, r, want) readTunBack(t, r, want)
}) })
} }
// TestTunWriteBatchOversize pins that a packet larger than the socket's send buffer drops only // TestUtunWriteBatchOversize pins that a packet larger than the socket's send buffer drops only
// itself and is reported, whether sendmsg_x reaches it mid-call (a short count) or first (EMSGSIZE // itself and is reported, whether sendmsg_x reaches it mid-call (a short count) or first (EMSGSIZE
// for the whole call). The send buffer is 1024 bytes and the oversize packet 2000. // for the whole call). The send buffer is 1024 bytes and the oversize packet 2000.
func TestTunWriteBatchOversize(t *testing.T) { func TestUtunWriteBatchOversize(t *testing.T) {
for _, tc := range []struct { for _, tc := range []struct {
name string name string
at int at int
@@ -128,8 +133,8 @@ func TestTunWriteBatchOversize(t *testing.T) {
} { } {
t.Run(tc.name, func(t *testing.T) { t.Run(tc.name, func(t *testing.T) {
forEachTunWritePath(t, func(t *testing.T, fallback bool) { forEachTunWritePath(t, func(t *testing.T, fallback bool) {
tn, r := newSocketpairTun(t, 1024) u, r := newSocketpairUtun(t, 1024)
tn.noSendmsgX.Store(fallback) u.noSendmsgX.Store(fallback)
var pkts, want [][]byte var pkts, want [][]byte
for i := range 2*tunWriteBatch + 10 { for i := range 2*tunWriteBatch + 10 {
@@ -143,11 +148,51 @@ func TestTunWriteBatchOversize(t *testing.T) {
want = append(want, w) want = append(want, w)
} }
} }
err := tn.WriteBatch(pkts) err := u.WriteBatch(pkts)
assert.ErrorIs(t, err, unix.EMSGSIZE) assert.ErrorIs(t, err, unix.EMSGSIZE)
assert.Equal(t, fallback, tn.noSendmsgX.Load()) assert.Equal(t, fallback, u.noSendmsgX.Load())
readTunBack(t, r, want) readTunBack(t, r, want)
}) })
}) })
} }
} }
// TestUtunReadDrains pins that Read hands back every queued packet, AF prefix stripped and in
// order, capped at tunReadBatch per call, and doesn't wait for more once the queue is empty.
func TestUtunReadDrains(t *testing.T) {
u, w := newSocketpairUtun(t, 1<<20)
total := tunReadBatch + 3
var want [][]byte
for i := range total {
pkt, wire := testTunPkt(i, 60+i)
_, err := unix.Write(w, wire)
require.NoError(t, err)
want = append(want, pkt)
}
var got [][]byte
for _, size := range []int{tunReadBatch, 3} {
pkts, err := u.Read()
require.NoError(t, err)
require.Len(t, pkts, size)
for _, p := range pkts {
got = append(got, p.Clone().Bytes)
}
}
assert.Equal(t, want, got)
}
// TestUtunReadClipsPackets pins that each returned packet's capacity ends at its own bytes, so
// appending to one can't overwrite the next.
func TestUtunReadClipsPackets(t *testing.T) {
u, w := newSocketpairUtun(t, 1<<20)
for i := range 2 {
_, wire := testTunPkt(i, 100)
_, err := unix.Write(w, wire)
require.NoError(t, err)
}
pkts, err := u.Read()
require.NoError(t, err)
require.Len(t, pkts, 2)
assert.Equal(t, len(pkts[0].Bytes), cap(pkts[0].Bytes))
}
+12 -282
View File
@@ -7,10 +7,7 @@ import (
"fmt" "fmt"
"log/slog" "log/slog"
"net/netip" "net/netip"
"os"
"sync"
"sync/atomic" "sync/atomic"
"syscall"
"unsafe" "unsafe"
"github.com/gaissmai/bart" "github.com/gaissmai/bart"
@@ -23,7 +20,7 @@ import (
) )
type tun struct { type tun struct {
f *os.File queue *tio.Utun
Device string Device string
vpnNetworks []netip.Prefix vpnNetworks []netip.Prefix
DefaultMTU int DefaultMTU int
@@ -34,35 +31,6 @@ type tun struct {
// and routes for it. NEPacketTunnelProvider on darwin does this. // and routes for it. NEPacketTunnelProvider on darwin does this.
hostOwned bool hostOwned bool
l *slog.Logger l *slog.Logger
// noSendmsgX is set once sendmsg_x is refused; WriteBatch then writes one packet at a time.
noSendmsgX atomic.Bool
batchMu sync.Mutex
batch tunBatch
}
// tunWriteBatch is how many packets one sendmsg_x call may carry.
const tunWriteBatch = 64
// tunBatch is WriteBatch's scratch, guarded by tun.batchMu. The kernel reads every entry in place.
type tunBatch struct {
heads [tunWriteBatch][4]byte
iovs [tunWriteBatch][2]unix.Iovec
hdrs [tunWriteBatch]msghdrX
pkts [tunWriteBatch][]byte
}
// msghdrX mirrors xnu's struct msghdr_x (bsd/sys/socket_private.h).
// Keep in sync with msghdrX in udp/udp_darwin.go.
type msghdrX struct {
Name *byte
Namelen uint32
Iov *unix.Iovec
Iovlen int32
Control *byte
Controllen uint32
Flags int32
Datalen uint64
} }
type ifReq struct { type ifReq struct {
@@ -147,13 +115,13 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*t
return nil, fmt.Errorf("failed to retrieve tun name: %w", err) return nil, fmt.Errorf("failed to retrieve tun name: %w", err)
} }
err = unix.SetNonblock(fd, true) queue, err := tio.NewUtun(fd, l)
if err != nil { if err != nil {
return nil, fmt.Errorf("SetNonblock: %v", err) return nil, err
} }
t := &tun{ t := &tun{
f: os.NewFile(uintptr(fd), ""), queue: queue,
Device: name, Device: name,
vpnNetworks: vpnNetworks, vpnNetworks: vpnNetworks,
DefaultMTU: c.GetInt("tun.mtu", DefaultMTU), DefaultMTU: c.GetInt("tun.mtu", DefaultMTU),
@@ -186,15 +154,15 @@ func (t *tun) deviceBytes() (o [16]byte) {
// network extension is handed its device. Everything about moving packets is shared with newTun, // network extension is handed its device. Everything about moving packets is shared with newTun,
// only the setup differs: the host owns addressing and routing here. // only the setup differs: the host owns addressing and routing here.
func newTunFromFd(c *config.C, l *slog.Logger, deviceFd int, vpnNetworks []netip.Prefix) (*tun, error) { func newTunFromFd(c *config.C, l *slog.Logger, deviceFd int, vpnNetworks []netip.Prefix) (*tun, error) {
if err := unix.SetNonblock(deviceFd, true); err != nil { queue, err := tio.NewUtun(deviceFd, l)
if err != nil {
// We own the fd from the moment it is handed to us // We own the fd from the moment it is handed to us
_ = unix.Close(deviceFd) _ = unix.Close(deviceFd)
return nil, fmt.Errorf("failed to set the tun fd to non-blocking mode: %w", err) return nil, err
} }
file := os.NewFile(uintptr(deviceFd), "/dev/tun")
t := &tun{ t := &tun{
f: file, queue: queue,
Device: utunNameFromFd(deviceFd), Device: utunNameFromFd(deviceFd),
vpnNetworks: vpnNetworks, vpnNetworks: vpnNetworks,
DefaultMTU: c.GetInt("tun.mtu", DefaultMTU), DefaultMTU: c.GetInt("tun.mtu", DefaultMTU),
@@ -203,7 +171,7 @@ func newTunFromFd(c *config.C, l *slog.Logger, deviceFd int, vpnNetworks []netip
} }
if err := t.reload(c, true); err != nil { if err := t.reload(c, true); err != nil {
_ = file.Close() _ = queue.Close()
return nil, err return nil, err
} }
@@ -227,8 +195,8 @@ func utunNameFromFd(fd int) string {
} }
func (t *tun) Close() error { func (t *tun) Close() error {
if t.f != nil { if t.queue != nil {
return t.f.Close() return t.queue.Close()
} }
return nil return nil
} }
@@ -586,244 +554,6 @@ func delRoute(prefix netip.Prefix, gateway netroute.Addr) error {
return nil return nil
} }
// tunWritev and tunReadv are linkname'd to x/sys/unix's libc-routed writev/readv stubs so the
// calls go through libSystem's pinned trampoline. A raw syscall.Syscall(SYS_WRITEV/SYS_READV, ...)
// on darwin/arm64 emits an SVC #0x80 trap (see $GOROOT/src/syscall/asm_darwin_arm64.s), the path
// Apple keeps warning they will eventually disallow. We pull the low-level stubs instead of calling
// unix.Writev/unix.Readv because those take [][]byte and rebuild the []Iovec every call, which
// heap-allocates the header; linkname'ing the stubs lets us hand them our own stack-allocated
// iovecs. See golang/go#78049.
//go:linkname tunWritev golang.org/x/sys/unix.writev
//go:noescape
func tunWritev(fd int, iovecs []unix.Iovec) (n int, err error)
//go:linkname tunReadv golang.org/x/sys/unix.readv
//go:noescape
func tunReadv(fd int, iovecs []unix.Iovec) (n int, err error)
// Read pulls one IP packet off the utun device, scattering the 4 byte protocol header away from
// the packet so the payload lands directly in to.
func (t *tun) Read(to []byte) (int, error) {
var head [4]byte
rc, err := t.f.SyscallConn()
if err != nil {
return 0, err
}
var n int
var callErr error
err = rc.Read(func(fd uintptr) bool {
iovecs := []unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &to[0], Len: uint64(len(to))},
}
n, callErr = tunReadv(int(fd), iovecs)
if errno, ok := callErr.(syscall.Errno); ok && errno.Temporary() {
return false
}
return true
})
if err != nil {
return 0, err
}
if callErr != nil {
return 0, callErr
}
if n < 4 {
return 0, nil
}
return n - 4, nil
}
// Write pushes one IP packet onto the utun device. Safe for concurrent use:
// the AF prefix and iovecs are per-call stack state, and the fd write itself
// serializes on the runtime's fd mutex (see the Queue contract in tio.go).
func (t *tun) Write(from []byte) (int, error) {
if len(from) == 0 {
return 0, syscall.EIO
}
var head [4]byte
af, err := tunAF(from)
if err != nil {
return 0, err
}
head[3] = af
// Grab rc as a local so the compiler can devirtualize the call and keep the closure on the stack.
rc, err := t.f.SyscallConn()
if err != nil {
return 0, err
}
var n int
var callErr error
err = rc.Write(func(fd uintptr) bool {
iovecs := []unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &from[0], Len: uint64(len(from))},
}
n, callErr = tunWritev(int(fd), iovecs)
// Type-assert to syscall.Errno so the EAGAIN/EWOULDBLOCK/EINTR check doesn't box the errno
// constants into error interfaces on every call.
if errno, ok := callErr.(syscall.Errno); ok && errno.Temporary() {
return false
}
return true
})
if err != nil {
return 0, err
}
if callErr != nil {
return 0, callErr
}
return n - 4, nil
}
// tunAF returns the utun address-family prefix byte for an IP packet.
func tunAF(pkt []byte) (byte, error) {
switch pkt[0] >> 4 {
case 4:
return syscall.AF_INET, nil
case 6:
return syscall.AF_INET6, nil
default:
return 0, fmt.Errorf("unable to determine IP version from packet")
}
}
// WriteBatch writes pkts to the utun device with sendmsg_x, xnu's private batched sendmsg, up to
// tunWriteBatch packets per syscall. The kernel still hands each packet to utun on its own, so this
// saves syscalls, not per-packet kernel work. Safe for concurrent use.
//
// An empty packet or one with no IP version is skipped and reported, as Write would. Any other
// error from sendmsg_x means the kernel took an unknown prefix of that call and dropped the rest, so
// it is reported rather than retried, which could deliver packets twice.
//
// Under mbuf exhaustion, sendmsg_x silently drops packets that writev would have delivered: the
// kernel counts each packet as sent once it is copied in, and when a later allocation fails it
// frees that prefix unsent and returns its length as a short count, which is indistinguishable
// from a real one.
func (t *tun) WriteBatch(pkts [][]byte) error {
if t.noSendmsgX.Load() {
return t.writeEach(pkts)
}
t.batchMu.Lock()
defer t.batchMu.Unlock()
b := &t.batch
var firstErr error
for len(pkts) > 0 {
n := 0
for len(pkts) > 0 && n < tunWriteBatch {
p := pkts[0]
pkts = pkts[1:]
var af byte
err := error(syscall.EIO)
if len(p) > 0 {
af, err = tunAF(p)
}
if err != nil {
if firstErr == nil {
firstErr = err
}
continue
}
b.heads[n] = [4]byte{3: af}
b.iovs[n] = [2]unix.Iovec{
{Base: &b.heads[n][0], Len: 4},
{Base: &p[0], Len: uint64(len(p))},
}
b.hdrs[n] = msghdrX{Iov: &b.iovs[n][0], Iovlen: 2}
b.pkts[n] = p
n++
}
if n == 0 {
continue
}
for off := 0; off < n; {
sent, err := t.sendmsgX(off, n)
switch {
case err == nil && sent > 0:
// The kernel stops early without an error on a packet larger than the socket's send
// buffer, or when sending a later packet fails with ENOBUFS; resend the remainder.
off += sent
case err == nil:
if werr := t.writeEach(b.pkts[off:n]); werr != nil && firstErr == nil {
firstErr = werr
}
off = n
case err == unix.EMSGSIZE:
// Only the first packet being larger than the socket's send buffer fails the whole call,
// before the kernel takes any; Write reports that one and the rest go out in the next call.
if werr := t.writeEach(b.pkts[off : off+1]); werr != nil && firstErr == nil {
firstErr = werr
}
off++
case err == unix.ENOSYS || err == unix.EPERM || err == unix.EOPNOTSUPP:
// sendmsg_x is private API: fall back to writev if a kernel or sandbox refuses it.
if t.noSendmsgX.CompareAndSwap(false, true) {
t.l.Warn("sendmsg_x unavailable on the tun device, writing one packet per syscall", "error", err)
}
if werr := t.writeEach(b.pkts[off:n]); werr != nil && firstErr == nil {
firstErr = werr
}
clear(b.iovs[:n])
clear(b.pkts[:n])
if werr := t.writeEach(pkts); werr != nil && firstErr == nil {
firstErr = werr
}
return firstErr
default:
if firstErr == nil {
firstErr = err
}
off = n
}
}
clear(b.iovs[:n])
clear(b.pkts[:n])
}
return firstErr
}
// sendmsgX hands entries off through n-1 of t.batch to sendmsg_x and returns how many the kernel took.
func (t *tun) sendmsgX(off, n int) (int, error) {
rc, err := t.f.SyscallConn()
if err != nil {
return 0, err
}
var sent uintptr
var errno syscall.Errno
err = rc.Write(func(fd uintptr) bool {
sent, _, errno = unix.Syscall6(unix.SYS_SENDMSG_X, fd, uintptr(unsafe.Pointer(&t.batch.hdrs[off])), uintptr(n-off), 0, 0, 0)
// sendmsg_x reports EAGAIN only when it took nothing; a partial batch returns its count.
return !errno.Temporary()
})
if err != nil {
return 0, err
}
if errno != 0 {
return 0, errno
}
return int(sent), nil
}
// writeEach writes pkts one at a time with Write.
func (t *tun) writeEach(pkts [][]byte) error {
var firstErr error
for _, p := range pkts {
if _, err := t.Write(p); err != nil && firstErr == nil {
firstErr = err
}
}
return firstErr
}
func (t *tun) Networks() []netip.Prefix { func (t *tun) Networks() []netip.Prefix {
return t.vpnNetworks return t.vpnNetworks
} }
@@ -833,5 +563,5 @@ func (t *tun) Name() string {
} }
func (t *tun) Queues(int) ([]tio.Queue, error) { func (t *tun) Queues(int) ([]tio.Queue, error) {
return []tio.Queue{tio.NewSingleQueue(t, defaultBatchBufSize)}, nil return []tio.Queue{t.queue}, nil
} }
+163 -30
View File
@@ -10,10 +10,12 @@ import (
"log/slog" "log/slog"
"net" "net"
"net/netip" "net/netip"
"sync/atomic"
"syscall" "syscall"
"unsafe" "unsafe"
"github.com/slackhq/nebula/config" "github.com/slackhq/nebula/config"
"github.com/slackhq/nebula/internal/msgx"
"golang.org/x/sys/unix" "golang.org/x/sys/unix"
) )
@@ -22,6 +24,9 @@ type StdConn struct {
isV4 bool isV4 bool
sysFd uintptr sysFd uintptr
l *slog.Logger l *slog.Logger
noSendmsgX atomic.Bool
sb *sendScratch
} }
var _ Conn = &StdConn{} var _ Conn = &StdConn{}
@@ -35,6 +40,7 @@ func NewListener(l *slog.Logger, s Settings) (Conn, error) {
if uc, ok := pc.(*net.UDPConn); ok { if uc, ok := pc.(*net.UDPConn); ok {
c := &StdConn{UDPConn: uc, l: l} c := &StdConn{UDPConn: uc, l: l}
c.noSendmsgX.Store(!msgx.Available())
rc, err := uc.SyscallConn() rc, err := uc.SyscallConn()
if err != nil { if err != nil {
@@ -139,11 +145,138 @@ func (u *StdConn) WriteTo(b []byte, ap netip.AddrPort) error {
} }
} }
// sendBatch is how many datagrams one sendmsg_x call may carry.
const sendBatch = 64
// sendScratch is WriteBatch's scratch. The kernel reads every entry in place.
type sendScratch struct {
names [sendBatch]unix.RawSockaddrInet6
iovs [sendBatch]unix.Iovec
hdrs [sendBatch]msgx.Hdr
// idx maps each entry back to its index in WriteBatch's bufs.
idx [sendBatch]int
}
// WriteBatch sends bufs with sendmsg_x, darwin's private batched sendmsg, so one syscall carries up to
// sendBatch datagrams. On this unconnected socket xnu still runs each message through sendit, so the
// saving is the syscall crossings, not the per-send policy checks.
//
// xnu's per-message loop stops at the first message that fails. For the errors in sentNothing it reports
// how many went out; for any other it returns only the errno, with the messages before it already sent.
// Each call therefore carries one run of datagrams to the same destination: a destination the kernel
// refuses fails on its first datagram, so a failed call costs that run and nothing else, and no datagram
// is ever resent after the kernel may have taken it.
//
// TODO: WriteTo maps EWOULDBLOCK to an error, so a full send buffer
// silently drops the rest of a burst (linux blocks instead). Poll for
// writability on EAGAIN before giving up on the remainder.
func (u *StdConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, error) { func (u *StdConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, error) {
// An un-sendable destination costs its own packet, never the ones behind it in the batch. if u.noSendmsgX.Load() {
// TODO: WriteTo maps EWOULDBLOCK to an error, so a full send buffer return u.writeEach(bufs, addrs), nil
// silently drops the rest of a burst (linux blocks instead). Poll for }
// writability on EAGAIN before giving up on the remainder. if u.sb == nil {
u.sb = new(sendScratch)
}
b := u.sb
written := 0
for i := 0; i < len(bufs); {
n := 0
for ; i < len(bufs) && n < sendBatch; i++ {
namelen, err := u.putSockaddr(&b.names[n], addrs[i])
if err != nil {
u.l.Debug("failed to write packet in batch", "udpAddr", addrs[i], "error", err)
continue
}
b.iovs[n] = unix.Iovec{}
if len(bufs[i]) > 0 {
b.iovs[n].Base = &bufs[i][0]
b.iovs[n].SetLen(len(bufs[i]))
}
b.hdrs[n] = msgx.Hdr{Name: (*byte)(unsafe.Pointer(&b.names[n])), Namelen: namelen, Iov: &b.iovs[n], Iovlen: 1}
b.idx[n] = i
n++
}
for off := 0; off < n; {
dst := addrs[b.idx[off]]
end := off + 1
for end < n && addrs[b.idx[end]] == dst {
end++
}
sent, errno := u.sendmsgX(off, end)
switch {
case errno == 0 && sent > 0:
written += sent
off += sent
continue
case sentNothing(errno):
// The datagram at off failed on its own account. sendto delivers or reports it, as without
// batching, and the rest of the run goes back to sendmsg_x.
j := b.idx[off]
written += u.writeEach(bufs[j:j+1], addrs[j:j+1])
off++
continue
case errno == 0:
case errno == unix.ENOSYS || errno == unix.EOPNOTSUPP:
u.fallBack(errno)
case errno == unix.EPERM || errno == unix.EINVAL:
// Either can refuse one destination, but a datagram that sendto delivers was refused by
// sendmsg_x itself, as a sandbox or an older kernel would.
j := b.idx[off]
if u.writeEach(bufs[j:j+1], addrs[j:j+1]) == 0 {
off = end
continue
}
written++
off++
u.fallBack(errno)
default:
u.l.Debug("failed to write packets in batch", "udpAddr", dst, "packets", end-off, "error", errno)
off = end
continue
}
// The kernel took nothing from this call, so the rest can go out one datagram at a time.
for _, j := range b.idx[off:n] {
written += u.writeEach(bufs[j:j+1], addrs[j:j+1])
}
off = n
if u.noSendmsgX.Load() {
clear(b.iovs[:n])
return written + u.writeEach(bufs[i:], addrs[i:]), nil
}
}
clear(b.iovs[:n])
}
return written, nil
}
// sendmsgX hands entries off through end-1 of u.sb to sendmsg_x and returns how many the kernel took.
// The socket is non-blocking, so a full send buffer returns EAGAIN rather than waiting.
func (u *StdConn) sendmsgX(off, end int) (int, syscall.Errno) {
for {
sent, errno := msgx.Send(u.sysFd, u.sb.hdrs[off:end], 0)
if errno == unix.EINTR {
continue
}
return sent, errno
}
}
// sentNothing reports whether sendmsg_x failing with errno means no datagram went out: once any has, xnu
// turns these errors into a short count instead.
func sentNothing(errno syscall.Errno) bool {
return errno == unix.EWOULDBLOCK || errno == unix.ENOBUFS || errno == unix.EMSGSIZE
}
// fallBack switches WriteBatch to one sendto per datagram for the life of the socket.
func (u *StdConn) fallBack(errno syscall.Errno) {
if u.noSendmsgX.CompareAndSwap(false, true) {
u.l.Warn("sendmsg_x unavailable, sending one datagram per syscall", "error", errno)
}
}
// writeEach sends bufs one at a time with WriteTo and returns how many were written.
func (u *StdConn) writeEach(bufs [][]byte, addrs []netip.AddrPort) int {
written := 0 written := 0
for i, b := range bufs { for i, b := range bufs {
if err := u.WriteTo(b, addrs[i]); err == nil { if err := u.WriteTo(b, addrs[i]); err == nil {
@@ -152,7 +285,23 @@ func (u *StdConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, error)
u.l.Debug("failed to write packet in batch", "udpAddr", addrs[i], "error", err) u.l.Debug("failed to write packet in batch", "udpAddr", addrs[i], "error", err)
} }
} }
return written, nil return written
}
// putSockaddr writes ap into sa in the socket's address family, as WriteTo would, and returns its length.
func (u *StdConn) putSockaddr(sa *unix.RawSockaddrInet6, ap netip.AddrPort) (uint32, error) {
if u.isV4 {
if ap.Addr().Is6() {
return 0, ErrInvalidIPv6RemoteForSocket
}
sa4 := (*unix.RawSockaddrInet4)(unsafe.Pointer(sa))
*sa4 = unix.RawSockaddrInet4{Family: unix.AF_INET, Addr: ap.Addr().As4()}
binary.BigEndian.PutUint16((*[2]byte)(unsafe.Pointer(&sa4.Port))[:], ap.Port())
return unix.SizeofSockaddrInet4, nil
}
*sa = unix.RawSockaddrInet6{Family: unix.AF_INET6, Addr: ap.Addr().As16()}
binary.BigEndian.PutUint16((*[2]byte)(unsafe.Pointer(&sa.Port))[:], ap.Port())
return unix.SizeofSockaddrInet6, nil
} }
func (u *StdConn) LocalAddr() (netip.AddrPort, error) { func (u *StdConn) LocalAddr() (netip.AddrPort, error) {
@@ -183,24 +332,12 @@ func NewUDPStatsEmitter(udpConns []Conn) func() {
// recvBatch is how many datagrams one recvmsg_x call may return. // recvBatch is how many datagrams one recvmsg_x call may return.
const recvBatch = 64 const recvBatch = 64
// msghdrX mirrors xnu's struct msghdr_x (bsd/sys/socket_private.h). recvmsg_x reports each
// datagram's length in Datalen rather than in its return value.
// Keep in sync with msghdrX in overlay/tun_darwin.go.
type msghdrX struct {
Name *byte
Namelen uint32
Iov *unix.Iovec
Iovlen int32
Control *byte
Controllen uint32
Flags int32
Datalen uint64
}
// ListenOut drains the socket with recvmsg_x, darwin's private batched recvmsg, so one syscall reads // ListenOut drains the socket with recvmsg_x, darwin's private batched recvmsg, so one syscall reads
// up to recvBatch datagrams instead of one. Sends stay on sendto: xnu batches sendmsg_x through // up to recvBatch datagrams instead of one.
// the stack only on connected sockets, and on this unconnected one it loops sendit per message.
func (u *StdConn) ListenOut(r EncReader, flush func()) error { func (u *StdConn) ListenOut(r EncReader, flush func()) error {
if !msgx.Available() {
return u.listenOutSingle(r, flush)
}
rc, err := u.UDPConn.SyscallConn() rc, err := u.UDPConn.SyscallConn()
if err != nil { if err != nil {
return err return err
@@ -209,7 +346,7 @@ func (u *StdConn) ListenOut(r EncReader, flush func()) error {
bufs := make([][]byte, recvBatch) bufs := make([][]byte, recvBatch)
names := make([]unix.RawSockaddrInet6, recvBatch) names := make([]unix.RawSockaddrInet6, recvBatch)
iovs := make([]unix.Iovec, recvBatch) iovs := make([]unix.Iovec, recvBatch)
hdrs := make([]msghdrX, recvBatch) hdrs := make([]msgx.Hdr, recvBatch)
for i := range hdrs { for i := range hdrs {
bufs[i] = make([]byte, MTU) bufs[i] = make([]byte, MTU)
iovs[i].Base = &bufs[i][0] iovs[i].Base = &bufs[i][0]
@@ -228,12 +365,8 @@ func (u *StdConn) ListenOut(r EncReader, flush func()) error {
hdrs[i].Flags = 0 hdrs[i].Flags = 0
hdrs[i].Datalen = 0 hdrs[i].Datalen = 0
} }
r0, _, e := unix.Syscall6(unix.SYS_RECVMSG_X, fd, uintptr(unsafe.Pointer(&hdrs[0])), recvBatch, 0, 0, 0) n, errno = msgx.Recv(fd, hdrs, 0)
if e == unix.EAGAIN { return errno != unix.EAGAIN
return false
}
n, errno = int(r0), e
return true
}) })
if err != nil { if err != nil {
if errors.Is(err, net.ErrClosed) { if errors.Is(err, net.ErrClosed) {
@@ -243,7 +376,7 @@ func (u *StdConn) ListenOut(r EncReader, flush func()) error {
continue continue
} }
if errno == unix.ENOSYS || errno == unix.EPERM || errno == unix.EOPNOTSUPP { if errno == unix.ENOSYS || errno == unix.EPERM || errno == unix.EOPNOTSUPP {
// recvmsg_x is private, so a kernel or sandbox (such as an iOS extension's) may refuse it. // recvmsg_x is private, so a kernel or sandbox may refuse it.
u.l.Warn("recvmsg_x unavailable, reading one datagram per syscall", "error", errno) u.l.Warn("recvmsg_x unavailable, reading one datagram per syscall", "error", errno)
return u.listenOutSingle(r, flush) return u.listenOutSingle(r, flush)
} }
@@ -273,7 +406,7 @@ func (u *StdConn) ListenOut(r EncReader, flush func()) error {
// checkMsghdrX rejects a recvmsg_x result that intact msghdr_x entries can't produce, which is how a change // checkMsghdrX rejects a recvmsg_x result that intact msghdr_x entries can't produce, which is how a change
// to xnu's private struct layout would show up. // to xnu's private struct layout would show up.
func checkMsghdrX(hdrs []msghdrX, names []unix.RawSockaddrInet6, n int) error { func checkMsghdrX(hdrs []msgx.Hdr, names []unix.RawSockaddrInet6, n int) error {
if n > len(hdrs) { if n > len(hdrs) {
return fmt.Errorf("returned %d datagrams for %d headers", n, len(hdrs)) return fmt.Errorf("returned %d datagrams for %d headers", n, len(hdrs))
} }
+13 -13
View File
@@ -14,6 +14,7 @@ import (
"time" "time"
"unsafe" "unsafe"
"github.com/slackhq/nebula/internal/msgx"
"github.com/stretchr/testify/assert" "github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require" "github.com/stretchr/testify/require"
"golang.org/x/sys/unix" "golang.org/x/sys/unix"
@@ -81,7 +82,7 @@ func TestListenOutBatchesBurst(t *testing.T) {
} }
} }
// msghdrX round-trips through xnu's recvmsg_x on a loopback socket, v4 and v6. The kernel reads Name, Namelen, // msgx.Hdr round-trips through xnu's recvmsg_x on a loopback socket, v4 and v6. The kernel reads Name, Namelen,
// Iov and Iovlen from each entry and writes Namelen, Flags and Datalen back, so a field at the wrong offset or a // Iov and Iovlen from each entry and writes Namelen, Flags and Datalen back, so a field at the wrong offset or a
// wrong stride between entries shows up as a wrong length, sender, or payload. // wrong stride between entries shows up as a wrong length, sender, or payload.
// TestListenOutSingleFallback pins the path ListenOut takes when the kernel refuses recvmsg_x. // TestListenOutSingleFallback pins the path ListenOut takes when the kernel refuses recvmsg_x.
@@ -134,8 +135,8 @@ func TestListenOutSingleFallback(t *testing.T) {
} }
func TestCheckMsghdrX(t *testing.T) { func TestCheckMsghdrX(t *testing.T) {
valid := func() ([]msghdrX, []unix.RawSockaddrInet6) { valid := func() ([]msgx.Hdr, []unix.RawSockaddrInet6) {
hdrs := make([]msghdrX, 2) hdrs := make([]msgx.Hdr, 2)
names := make([]unix.RawSockaddrInet6, 2) names := make([]unix.RawSockaddrInet6, 2)
hdrs[0].Namelen, hdrs[0].Datalen = unix.SizeofSockaddrInet4, 1 hdrs[0].Namelen, hdrs[0].Datalen = unix.SizeofSockaddrInet4, 1
names[0].Family = unix.AF_INET names[0].Family = unix.AF_INET
@@ -151,12 +152,12 @@ func TestCheckMsghdrX(t *testing.T) {
for _, tc := range []struct { for _, tc := range []struct {
name string name string
n int n int
mutate func([]msghdrX, []unix.RawSockaddrInet6) mutate func([]msgx.Hdr, []unix.RawSockaddrInet6)
}{ }{
{"count past headers", 3, func([]msghdrX, []unix.RawSockaddrInet6) {}}, {"count past headers", 3, func([]msgx.Hdr, []unix.RawSockaddrInet6) {}},
{"datalen past buffer", 2, func(h []msghdrX, _ []unix.RawSockaddrInet6) { h[1].Datalen = MTU + 1 }}, {"datalen past buffer", 2, func(h []msgx.Hdr, _ []unix.RawSockaddrInet6) { h[1].Datalen = MTU + 1 }},
{"unknown family", 2, func(_ []msghdrX, a []unix.RawSockaddrInet6) { a[0].Family = unix.AF_UNIX }}, {"unknown family", 2, func(_ []msgx.Hdr, a []unix.RawSockaddrInet6) { a[0].Family = unix.AF_UNIX }},
{"family and length disagree", 2, func(h []msghdrX, _ []unix.RawSockaddrInet6) { h[0].Namelen = unix.SizeofSockaddrInet6 }}, {"family and length disagree", 2, func(h []msgx.Hdr, _ []unix.RawSockaddrInet6) { h[0].Namelen = unix.SizeofSockaddrInet6 }},
} { } {
t.Run(tc.name, func(t *testing.T) { t.Run(tc.name, func(t *testing.T) {
hdrs, names := valid() hdrs, names := valid()
@@ -238,7 +239,7 @@ func recvmsgXRoundTrip(t *testing.T, v6 bool, lens []int, nh int, settle time.Du
bufs := make([][]byte, nh) bufs := make([][]byte, nh)
names := make([]unix.RawSockaddrInet6, nh) names := make([]unix.RawSockaddrInet6, nh)
iovs := make([]unix.Iovec, nh) iovs := make([]unix.Iovec, nh)
hdrs := make([]msghdrX, nh) hdrs := make([]msgx.Hdr, nh)
for i := range hdrs { for i := range hdrs {
bufs[i] = make([]byte, MTU) bufs[i] = make([]byte, MTU)
iovs[i].Base = &bufs[i][0] iovs[i].Base = &bufs[i][0]
@@ -253,7 +254,7 @@ func recvmsgXRoundTrip(t *testing.T, v6 bool, lens []int, nh int, settle time.Du
for next := 0; next < len(lens); { for next := 0; next < len(lens); {
for i := range hdrs { for i := range hdrs {
clear(bufs[i]) clear(bufs[i])
hdrs[i] = msghdrX{ hdrs[i] = msgx.Hdr{
Name: (*byte)(unsafe.Pointer(&names[i])), Name: (*byte)(unsafe.Pointer(&names[i])),
Namelen: unix.SizeofSockaddrInet6, Namelen: unix.SizeofSockaddrInet6,
Iov: &iovs[i], Iov: &iovs[i],
@@ -265,9 +266,8 @@ func recvmsgXRoundTrip(t *testing.T, v6 bool, lens []int, nh int, settle time.Du
var n int var n int
var errno syscall.Errno var errno syscall.Errno
require.NoError(t, rc.Read(func(fd uintptr) bool { require.NoError(t, rc.Read(func(fd uintptr) bool {
r0, _, e := unix.Syscall6(unix.SYS_RECVMSG_X, fd, uintptr(unsafe.Pointer(&hdrs[0])), uintptr(nh), 0, 0, 0) n, errno = msgx.Recv(fd, hdrs[:nh], 0)
n, errno = int(r0), e return errno != unix.EAGAIN
return e != unix.EAGAIN
}), "received %d of %d datagrams", next, len(lens)) }), "received %d of %d datagrams", next, len(lens))
require.Zero(t, errno) require.Zero(t, errno)
require.Positive(t, n) require.Positive(t, n)
+150
View File
@@ -0,0 +1,150 @@
//go:build !e2e_testing
package udp
import (
"fmt"
"log/slog"
"net"
"net/netip"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// recvAll reads n datagrams from pc, or fails the test after a second without one.
func recvAll(t *testing.T, pc net.PacketConn, n int) []string {
t.Helper()
var got []string
buf := make([]byte, 64)
for len(got) < n {
require.NoError(t, pc.SetReadDeadline(time.Now().Add(time.Second)))
l, _, err := pc.ReadFrom(buf)
require.NoError(t, err, "after %d of %d datagrams", len(got), n)
got = append(got, string(buf[:l]))
}
return got
}
// A batch longer than sendBatch reaches each peer whole and in order through sendmsg_x, without falling
// back to sendto: a run to one peer that spans two calls, then datagrams alternating between peers.
func TestWriteBatchSendsEachToItsPeer(t *testing.T) {
for _, tc := range []struct{ name, listen string }{
{"v4", "127.0.0.1"},
{"v6", "::1"},
} {
t.Run(tc.name, func(t *testing.T) {
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.AddrPortFrom(netip.MustParseAddr(tc.listen), 0)})
require.NoError(t, err)
defer c.Close()
var peers [2]net.PacketConn
var dsts [2]netip.AddrPort
for i := range peers {
peers[i], err = net.ListenPacket("udp", net.JoinHostPort(tc.listen, "0"))
require.NoError(t, err)
defer peers[i].Close()
dsts[i] = netip.MustParseAddrPort(peers[i].LocalAddr().String())
}
const total = sendBatch*2 + 5
bufs := make([][]byte, total)
addrs := make([]netip.AddrPort, total)
var want [2][]string
for i := range bufs {
bufs[i] = fmt.Appendf(nil, "pkt-%03d", i)
p := i % 2
if i < sendBatch+6 {
p = 0
}
addrs[i] = dsts[p]
want[p] = append(want[p], string(bufs[i]))
}
n, err := c.WriteBatch(bufs, addrs)
require.NoError(t, err)
assert.Equal(t, total, n)
assert.False(t, c.(*StdConn).noSendmsgX.Load(), "fell back to sendto")
for i := range peers {
assert.Equal(t, want[i], recvAll(t, peers[i], len(want[i])))
}
})
}
}
// Destinations that can't be sent to cost only their own packets: an IPv6 peer on an IPv4 socket is
// skipped before the call, and a broadcast address without SO_BROADCAST is refused by the kernel
// mid-batch. Neither stops the packets around it, delivers any twice, or turns sendmsg_x off.
func TestWriteBatchSkipsUnsendableDestinations(t *testing.T) {
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0")})
require.NoError(t, err)
defer c.Close()
peer, err := net.ListenPacket("udp", "127.0.0.1:0")
require.NoError(t, err)
defer peer.Close()
good := netip.MustParseAddrPort(peer.LocalAddr().String())
bcast := netip.AddrPortFrom(netip.MustParseAddr("255.255.255.255"), good.Port())
v6 := netip.MustParseAddrPort("[::1]:4242")
addrs := []netip.AddrPort{good, bcast, bcast, good, v6, good, bcast}
bufs := make([][]byte, len(addrs))
for i := range bufs {
bufs[i] = fmt.Appendf(nil, "pkt-%d", i)
}
n, err := c.WriteBatch(bufs, addrs)
require.NoError(t, err)
assert.Equal(t, 3, n)
assert.False(t, c.(*StdConn).noSendmsgX.Load(), "fell back to sendto")
assert.Equal(t, []string{"pkt-0", "pkt-3", "pkt-5"}, recvAll(t, peer, 3))
require.NoError(t, peer.SetReadDeadline(time.Now().Add(100*time.Millisecond)))
_, _, err = peer.ReadFrom(make([]byte, 64))
assert.Error(t, err, "a datagram arrived twice")
}
// A datagram the kernel refuses as its own fault (EMSGSIZE here, as ENOBUFS would be under a full
// interface queue) costs only itself, whether it fails after others in the same call (a short count) or
// first (the errno alone): the rest of its run still arrives, in order, and sendmsg_x stays on.
func TestWriteBatchLosesOnlyTheFailedDatagram(t *testing.T) {
for _, tc := range []struct {
name string
big int
}{
{"mid-call", 2},
{"first", 0},
} {
t.Run(tc.name, func(t *testing.T) {
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0")})
require.NoError(t, err)
defer c.Close()
peer, err := net.ListenPacket("udp", "127.0.0.1:0")
require.NoError(t, err)
defer peer.Close()
dst := netip.MustParseAddrPort(peer.LocalAddr().String())
bufs := make([][]byte, 5)
addrs := make([]netip.AddrPort, len(bufs))
var want []string
for i := range bufs {
addrs[i] = dst
if i == tc.big {
// Past the 65535-byte IP datagram limit, so the kernel can only refuse it.
bufs[i] = make([]byte, 70000)
continue
}
bufs[i] = fmt.Appendf(nil, "pkt-%d", i)
want = append(want, string(bufs[i]))
}
n, err := c.WriteBatch(bufs, addrs)
require.NoError(t, err)
assert.Equal(t, len(want), n)
assert.False(t, c.(*StdConn).noSendmsgX.Load(), "fell back to sendto")
assert.Equal(t, want, recvAll(t, peer, len(want)))
})
}
}