device: distribute crypto work as slice of elements

After reducing UDP stack traversal overhead via GSO and GRO,
runtime.chanrecv() began to account for a high percentage (20% in one
environment) of perf samples during a throughput benchmark. The
individual packet channel ops with the crypto goroutines was the primary
contributor to this overhead.

Updating these channels to pass vectors, which the device package
already handles at its ends, reduced this overhead substantially, and
improved throughput.

The iperf3 results below demonstrate the effect of this commit between
two Linux computers with i5-12400 CPUs. There is roughly ~13us of round
trip latency between them.

The first result is with UDP GSO and GRO, and with single element
channels.

Starting Test: protocol: TCP, 1 streams, 131072 byte blocks
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-10.00  sec  12.3 GBytes  10.6 Gbits/sec  232   3.15 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
Test Complete. Summary Results:
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  12.3 GBytes  10.6 Gbits/sec  232   sender
[  5]   0.00-10.04  sec  12.3 GBytes  10.6 Gbits/sec        receiver

The second result is with channels updated to pass a slice of
elements.

Starting Test: protocol: TCP, 1 streams, 131072 byte blocks
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-10.00  sec  13.2 GBytes  11.3 Gbits/sec  182   3.15 MBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
Test Complete. Summary Results:
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  13.2 GBytes  11.3 Gbits/sec  182   sender
[  5]   0.00-10.04  sec  13.2 GBytes  11.3 Gbits/sec        receiver

Reviewed-by: Adrian Dewhurst <adrian@tailscale.com>
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

1 files changed, 29 insertions, 29 deletions

diff --git a/device/send.go b/device/send.go
index cd8a2a0..e838c4e 100644
--- a/device/send.go
+++ b/device/send.go
@@ -385,9 +385,7 @@ top:
 			// add to parallel and sequential queue
 			if peer.isRunning.Load() {
 				peer.queue.outbound.c <- elems
-				for _, elem := range *elems {
-					peer.device.queue.encryption.c <- elem
-				}
+				peer.device.queue.encryption.c <- elems
 			} else {
 				for _, elem := range *elems {
 					peer.device.PutMessageBuffer(elem.buffer)
@@ -447,32 +445,34 @@ func (device *Device) RoutineEncryption(id int) {
 	defer device.log.Verbosef("Routine: encryption worker %d - stopped", id)
 	device.log.Verbosef("Routine: encryption worker %d - started", id)
 
-	for elem := range device.queue.encryption.c {
-		// populate header fields
-		header := elem.buffer[:MessageTransportHeaderSize]
-
-		fieldType := header[0:4]
-		fieldReceiver := header[4:8]
-		fieldNonce := header[8:16]
-
-		binary.LittleEndian.PutUint32(fieldType, MessageTransportType)
-		binary.LittleEndian.PutUint32(fieldReceiver, elem.keypair.remoteIndex)
-		binary.LittleEndian.PutUint64(fieldNonce, elem.nonce)
-
-		// pad content to multiple of 16
-		paddingSize := calculatePaddingSize(len(elem.packet), int(device.tun.mtu.Load()))
-		elem.packet = append(elem.packet, paddingZeros[:paddingSize]...)
-
-		// encrypt content and release to consumer
-
-		binary.LittleEndian.PutUint64(nonce[4:], elem.nonce)
-		elem.packet = elem.keypair.send.Seal(
-			header,
-			nonce[:],
-			elem.packet,
-			nil,
-		)
-		elem.Unlock()
+	for elems := range device.queue.encryption.c {
+		for _, elem := range *elems {
+			// populate header fields
+			header := elem.buffer[:MessageTransportHeaderSize]
+
+			fieldType := header[0:4]
+			fieldReceiver := header[4:8]
+			fieldNonce := header[8:16]
+
+			binary.LittleEndian.PutUint32(fieldType, MessageTransportType)
+			binary.LittleEndian.PutUint32(fieldReceiver, elem.keypair.remoteIndex)
+			binary.LittleEndian.PutUint64(fieldNonce, elem.nonce)
+
+			// pad content to multiple of 16
+			paddingSize := calculatePaddingSize(len(elem.packet), int(device.tun.mtu.Load()))
+			elem.packet = append(elem.packet, paddingZeros[:paddingSize]...)
+
+			// encrypt content and release to consumer
+
+			binary.LittleEndian.PutUint64(nonce[4:], elem.nonce)
+			elem.packet = elem.keypair.send.Seal(
+				header,
+				nonce[:],
+				elem.packet,
+				nil,
+			)
+			elem.Unlock()
+		}
 	}
 }