kvm tools: Move virtio drivers under virtio directory

This patch moves the virtio drivers under virtio directory.

Suggested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Pekka Enberg <penberg@kern

kvm tools: Move virtio drivers under virtio directory

This patch moves the virtio drivers under virtio directory.

Suggested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

Revert "kvm tools: Use threadpool for virtio-net"

This reverts commit a37089da817ce7aad9789aeb9fc09b68e088ad9a.

kvm tools: Use threadpool for virtio-net

virtio-net has been converted to use the threadpool. This is very similar to
the change done in virtio-blk, only here we had 2 queues to handle.

Signed-off

kvm tools: Use threadpool for virtio-net

virtio-net has been converted to use the threadpool. This is very similar to
the change done in virtio-blk, only here we had 2 queues to handle.

Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

kvm tools: Use non shared pin/irqs for virtio devices

There is no need for shared IRQs for virtio devices.

Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Pekka Enberg <penberg@k

kvm tools: Use non shared pin/irqs for virtio devices

There is no need for shared IRQs for virtio devices.

Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

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.

*******

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>

show more ...

kvm tools: Setup bridged network by a script

Use original hardcode network by default.

#./kvm run ... -n virtio --tapscript=./util/kvm-ifup-vbr0
# brctl show
bridge name bridge id

kvm tools: Setup bridged network by a script

Use original hardcode network by default.

#./kvm run ... -n virtio --tapscript=./util/kvm-ifup-vbr0
# brctl show
bridge name bridge id STP enabled interfaces
vbr0 8000.e272c7c391f4 no tap0
guest)# ifconfig eth6
eth6 Link encap:Ethernet HWaddr 00:11:22:33:44:55
inet addr:192.168.33.192 Bcast:192.168.33.255 Mask:255.255.255.0
inet6 addr: fe80::211:22ff:fe33:4455/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:22 errors:0 dropped:0 overruns:0 frame:0
TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:3725 (3.6 KiB) TX bytes:852 (852.0 b)
guest)# ping amosk.info
PING amosk.info (69.175.108.82) 56(84) bytes of data.
64 bytes from nurpulat.uz (69.175.108.82): icmp_seq=1 ttl=43 time=306 ms

Changes from v1:
- rebased to latest tree
- replace system() by execv()

Signed-off-by: Amos Kong <kongjianjun@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

kvm tools: Add option to specify guest MAC

Add --guest-mac to specify the MAC of the guest.

Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

kvm tools: Enable network by default

Enable virtio networking by default, warn if user doesn't have tun/tap interface or not enough permissions to start it.

Signed-off-by: Sasha Levin <levinsasha92

kvm tools: Enable network by default

Enable virtio networking by default, warn if user doesn't have tun/tap interface or not enough permissions to start it.

Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

kvm tools: Organize net parameters into struct

Move network configuration parameters into a struct.
The amount of network parameters will be rather large, so better do it early.

Signed-off-by: Sash

kvm tools: Organize net parameters into struct

Move network configuration parameters into a struct.
The amount of network parameters will be rather large, so better do it early.

Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

kvm tools: Make host side IP configurable

Add --host-ip-addr parameter to allow changing the host-side IP address.
Add a networking group to the cmdline menu.

Reviewed-by: Asias He <asias.hejun@gma

kvm tools: Make host side IP configurable

Add --host-ip-addr parameter to allow changing the host-side IP address.
Add a networking group to the cmdline menu.

Reviewed-by: Asias He <asias.hejun@gmail.com>
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

kvm tools: Set up tun interface using ioctls

Use ioctls to assign IP address and bring interface up instead of using ifconfig.
Not breaking aliasing rules this time.

Reviewed-by: Asias He <asias.he

kvm tools: Set up tun interface using ioctls

Use ioctls to assign IP address and bring interface up instead of using ifconfig.
Not breaking aliasing rules this time.

Reviewed-by: Asias He <asias.hejun@gmail.com>
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...

kvm tools: Implement virtio network device

This patch implement virtio network device.
Use '-n virtio or --network=virtio' to enable it.

The current implementation uses tap which needs root privile

kvm tools: Implement virtio network device

This patch implement virtio network device.
Use '-n virtio or --network=virtio' to enable it.

The current implementation uses tap which needs root privileges to create a
virtual network device (tap0) on host side. Actually, what we need is
CAP_NET_ADMIN.

The host side tap0 is set to 192.168.33.2/24.
You need to configure the guest side eth0 to any ip address in
192.168.33.0/24.

Here are some scp performance test for differenct implementations:
None of rx and tx as thread:
guest to host 3.2MB/s
host to guest 3.1MB/s

Only rx as thread:
guest to host 14.7MB/s
host to guest 33.4MB/s

Both rx and tx as thread(This patch works this way):
guest to host 19.8MB/s
host to guest 32.5MB/s

Signed-off-by: Asias He <asias.hejun@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>

show more ...