kvm tools: Implement virtio net TSO/UFO support

This patch bumps host to guest tcp bandwidth from

1060 Mib/s to 1760 Mib/s,

and guest to host tcp bandwidth from

 342 Mib/s to  619 Mib/s.

*************************
Without TSO and UFO
*************************
(guest <- host)
root@sid1:~# iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 192.168.33.15 port 5001 connected with 192.168.33.2 port 38733
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-10.0 sec  1.23 GBytes  1.06 Gbits/sec
^Croot@sid1:~# iperf -s -u
------------------------------------------------------------
Server listening on UDP port 5001
Receiving 1470 byte datagrams
UDP buffer size:   110 KByte (default)
------------------------------------------------------------
[  3] local 192.168.33.15 port 5001 connected with 192.168.33.2 port 54933
[ ID] Interval       Transfer     Bandwidth       Jitter   Lost/Total Datagrams
[  3]  0.0-10.0 sec  1.25 MBytes  1.05 Mbits/sec  0.030 ms    0/  893 (0%)

(guest to host)
root@sid1:~# iperf -c host
------------------------------------------------------------
Client connecting to host, TCP port 5001
TCP window size: 16.0 KByte (default)
------------------------------------------------------------
[  3] local 192.168.33.15 port 42197 connected with 192.168.33.2 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-10.0 sec    408 MBytes    342 Mbits/sec
root@sid1:~# iperf -c host -u
------------------------------------------------------------
Client connecting to host, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size:   110 KByte (default)
------------------------------------------------------------
[  3] local 192.168.33.15 port 56176 connected with 192.168.33.2 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-10.0 sec  1.25 MBytes  1.05 Mbits/sec
[  3] Sent 893 datagrams
[  3] Server Report:
[  3]  0.0-10.0 sec  1.25 MBytes  1.05 Mbits/sec  0.012 ms    0/  893 (0%)

*************************
With TSO and UFO
*************************

(guest <- host)
root@sid1:~# iperf  -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 192.168.33.15 port 5001 connected with 192.168.33.2 port 42767
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-10.0 sec  2.05 GBytes  1.76 Gbits/sec
root@sid1:~# iperf  -s -u
------------------------------------------------------------
Server listening on UDP port 5001
Receiving 1470 byte datagrams
UDP buffer size:   110 KByte (default)
------------------------------------------------------------
[  3] local 192.168.33.15 port 5001 connected with 192.168.33.2 port 35049
[ ID] Interval       Transfer     Bandwidth       Jitter   Lost/Total Datagrams
[  3]  0.0-10.0 sec  1.25 MBytes  1.05 Mbits/sec  0.031 ms    0/  893 (0%)

(guest -> host)
asias@hj:~$ iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 192.168.33.2 port 5001 connected with 192.168.33.15 port 60868
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-10.0 sec   738 MBytes   619 Mbits/sec
asias@hj:~$ iperf -s -u
------------------------------------------------------------
Server listening on UDP port 5001
Receiving 1470 byte datagrams
UDP buffer size:  112 KByte (default)
------------------------------------------------------------
[  3] local 192.168.33.2 port 5001 connected with 192.168.33.15 port 40602
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3]  0.0-10.0 sec  1.25 MBytes  1.05 Mbits/sec   0.030 ms    0/  893 (0%)

Signed-off-by: Asias He <asias.hejun@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
This commit is contained in:
Asias He
2015-06-01 16:39:42 +01:00
committed by Will Deacon
parent d39cefd257
commit 246c834771
+32 -20
View File
@@ -57,8 +57,14 @@ static struct net_device net_device = {
.mac = {0x00, 0x11, 0x22, 0x33, 0x44, 0x55},
.status = VIRTIO_NET_S_LINK_UP,
},
.host_features = 1UL << VIRTIO_NET_F_MAC,
.host_features = 1UL << VIRTIO_NET_F_MAC |
1UL << VIRTIO_NET_F_CSUM |
1UL << VIRTIO_NET_F_HOST_UFO |
1UL << VIRTIO_NET_F_HOST_TSO4 |
1UL << VIRTIO_NET_F_HOST_TSO6 |
1UL << VIRTIO_NET_F_GUEST_UFO |
1UL << VIRTIO_NET_F_GUEST_TSO4 |
1UL << VIRTIO_NET_F_GUEST_TSO6,
};
static void *virtio_net_rx_thread(void *p)
@@ -81,13 +87,8 @@ static void *virtio_net_rx_thread(void *p)
while (virt_queue__available(vq)) {
head = virt_queue__get_iov(vq, iov, &out, &in, self);
/* We do not specify GSO or CSUM features, So we can ignore virtio_net_hdr */
len = readv(net_device.tap_fd, iov + 1, in - 1);
/* However, We have to tell guest we have write the virtio_net_hdr */
virt_queue__set_used_elem(vq, head, sizeof(struct virtio_net_hdr) + len);
len = readv(net_device.tap_fd, iov, in);
virt_queue__set_used_elem(vq, head, len);
/* We should interrupt guest right now, otherwise latency is huge. */
kvm__irq_line(self, VIRTIO_NET_IRQ, 1);
}
@@ -119,7 +120,7 @@ static void *virtio_net_tx_thread(void *p)
while (virt_queue__available(vq)) {
head = virt_queue__get_iov(vq, iov, &out, &in, self);
len = writev(net_device.tap_fd, iov + 1, out - 1);
len = writev(net_device.tap_fd, iov, out);
virt_queue__set_used_elem(vq, head, len);
}
@@ -280,23 +281,22 @@ static struct pci_device_header virtio_net_pci_device = {
static bool virtio_net__tap_init(const struct virtio_net_parameters *params)
{
struct ifreq ifr;
int sock = socket(AF_INET, SOCK_STREAM, 0);
int i, pid, status;
int i, pid, status, offload, hdr_len;
struct sockaddr_in sin = {0};
struct ifreq ifr;
for (i = 0 ; i < 6 ; i++)
net_device.net_config.mac[i] = params->guest_mac[i];
net_device.tap_fd = open("/dev/net/tun", O_RDWR);
if (net_device.tap_fd < 0) {
warning("Unable to open /dev/net/tun\n");
warning("Unable to open /dev/net/tun");
goto fail;
}
memset(&ifr, 0, sizeof(ifr));
ifr.ifr_flags = IFF_TAP | IFF_NO_PI;
ifr.ifr_flags = IFF_TAP | IFF_NO_PI | IFF_VNET_HDR;
if (ioctl(net_device.tap_fd, TUNSETIFF, &ifr) < 0) {
warning("Config tap device error. Are you root?");
goto fail;
@@ -304,7 +304,22 @@ static bool virtio_net__tap_init(const struct virtio_net_parameters *params)
strncpy(net_device.tap_name, ifr.ifr_name, sizeof(net_device.tap_name));
ioctl(net_device.tap_fd, TUNSETNOCSUM, 1);
if (ioctl(net_device.tap_fd, TUNSETNOCSUM, 1) < 0) {
warning("Config tap device TUNSETNOCSUM error");
goto fail;
}
hdr_len = sizeof(struct virtio_net_hdr);
if (ioctl(net_device.tap_fd, TUNSETVNETHDRSZ, &hdr_len) < 0) {
warning("Config tap device TUNSETVNETHDRSZ error");
goto fail;
}
offload = TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | TUN_F_UFO;
if (ioctl(net_device.tap_fd, TUNSETOFFLOAD, offload) < 0) {
warning("Config tap device TUNSETOFFLOAD error");
goto fail;
}
if (strcmp(params->script, "none")) {
pid = fork();
@@ -320,15 +335,12 @@ static bool virtio_net__tap_init(const struct virtio_net_parameters *params)
}
} else {
memset(&ifr, 0, sizeof(ifr));
strncpy(ifr.ifr_name, net_device.tap_name, sizeof(net_device.tap_name));
sin.sin_addr.s_addr = inet_addr(params->host_ip);
memcpy(&(ifr.ifr_addr), &sin, sizeof(ifr.ifr_addr));
ifr.ifr_addr.sa_family = AF_INET;
if (ioctl(sock, SIOCSIFADDR, &ifr) < 0) {
warning("Can not set ip address on tap device");
warning("Could not set ip address on tap device");
goto fail;
}
}