netstack/tcp: software segmentation offload

Right now, we send each tcp packet separately, we call one system
call per-packet. This patch allows to generate multiple tcp packets
and send them by sendmmsg.

The arguable part of this CL is a way how to handle multiple headers.
This CL adds the next field to the Prepandable buffer.

Nginx test results:

Server Software:        nginx/1.15.9
Server Hostname:        10.138.0.2
Server Port:            8080

Document Path:          /10m.txt
Document Length:        10485760 bytes

w/o gso:
Concurrency Level:      5
Time taken for tests:   5.491 seconds
Complete requests:      100
Failed requests:        0
Total transferred:      1048600200 bytes
HTML transferred:       1048576000 bytes
Requests per second:    18.21 [#/sec] (mean)
Time per request:       274.525 [ms] (mean)
Time per request:       54.905 [ms] (mean, across all concurrent requests)
Transfer rate:          186508.03 [Kbytes/sec] received

sw-gso:

Concurrency Level:      5
Time taken for tests:   3.852 seconds
Complete requests:      100
Failed requests:        0
Total transferred:      1048600200 bytes
HTML transferred:       1048576000 bytes
Requests per second:    25.96 [#/sec] (mean)
Time per request:       192.576 [ms] (mean)
Time per request:       38.515 [ms] (mean, across all concurrent requests)
Transfer rate:          265874.92 [Kbytes/sec] received

w/o gso:
$ ./tcp_benchmark --client --duration 15  --ideal
[SUM]  0.0-15.1 sec  2.20 GBytes  1.25 Gbits/sec

software gso:
$ tcp_benchmark --client --duration 15  --ideal --gso $((1<<16)) --swgso
[SUM]  0.0-15.1 sec  3.99 GBytes  2.26 Gbits/sec

PiperOrigin-RevId: 276112677

2 files changed, 19 insertions, 0 deletions

diff --git a/pkg/tcpip/link/waitable/waitable.go b/pkg/tcpip/link/waitable/waitable.go
index e9b50b802..a04fc1062 100644
--- a/pkg/tcpip/link/waitable/waitable.go
+++ b/pkg/tcpip/link/waitable/waitable.go
@@ -109,6 +109,19 @@ func (e *Endpoint) WritePacket(r *stack.Route, gso *stack.GSO, hdr buffer.Prepen
 	return err
 }
 
+// WritePackets implements stack.LinkEndpoint.WritePackets. It is called by
+// higher-level protocols to write packets. It only forwards packets to the
+// lower endpoint if Wait or WaitWrite haven't been called.
+func (e *Endpoint) WritePackets(r *stack.Route, gso *stack.GSO, hdrs []stack.PacketDescriptor, payload buffer.VectorisedView, protocol tcpip.NetworkProtocolNumber) (int, *tcpip.Error) {
+	if !e.writeGate.Enter() {
+		return len(hdrs), nil
+	}
+
+	n, err := e.lower.WritePackets(r, gso, hdrs, payload, protocol)
+	e.writeGate.Leave()
+	return n, err
+}
+
 // WriteRawPacket implements stack.LinkEndpoint.WriteRawPacket.
 func (e *Endpoint) WriteRawPacket(packet buffer.VectorisedView) *tcpip.Error {
 	if !e.writeGate.Enter() {
diff --git a/pkg/tcpip/link/waitable/waitable_test.go b/pkg/tcpip/link/waitable/waitable_test.go
index f03bbceec..5f0f8fa2d 100644
--- a/pkg/tcpip/link/waitable/waitable_test.go
+++ b/pkg/tcpip/link/waitable/waitable_test.go
@@ -70,6 +70,12 @@ func (e *countedEndpoint) WritePacket(r *stack.Route, _ *stack.GSO, hdr buffer.P
 	return nil
 }
 
+// WritePackets implements stack.LinkEndpoint.WritePackets.
+func (e *countedEndpoint) WritePackets(r *stack.Route, _ *stack.GSO, hdrs []stack.PacketDescriptor, payload buffer.VectorisedView, protocol tcpip.NetworkProtocolNumber) (int, *tcpip.Error) {
+	e.writeCount += len(hdrs)
+	return len(hdrs), nil
+}
+
 func (e *countedEndpoint) WriteRawPacket(packet buffer.VectorisedView) *tcpip.Error {
 	e.writeCount++
 	return nil