Bug 1321299

Description of problem:
If use vhostuser to boot a guest, then migration fails with error message displayed.

Version-Release number of selected component (if applicable):
guest and host kernel:
3.10.0-364.rt56.241.el7.x86_64
host qemu:
qemu-kvm-rhev-2.5.0-3.el7.x86_64
ovs-dpdk:
openvswitch-dpdk-2.5.0-3.el7.x86_64

How reproducible:
100%

Steps to Reproduce:
1. setup vhostuser environment with script below:
#!/bin/sh

HUGETLB_PATH="/mnt/hugetlbfs/"
eth_model="X540-AT2"

# mount hugetlbfs
if [ ! -d "$HUGETLB_PATH" ]; then
	mkdir $HUGETLB_PATH
fi

PAGESIZE_1G=`cat /proc/cmdline | grep "default_hugepagesz=1G"`
if [ ! -z "$PAGESIZE_1G" ]; then
	mount -t hugetlbfs hugetlbfs $HUGETLB_PATH -o size=16G
else
	PAGESIZE_2M=`cat /proc/cmdline | grep "default_hugepagesz=2M"`
	if [ ! -z "$PAGESIZE_2M" ]; then
		mount -t  hugetlbfs -o pagesize=2048K none $HUGETLB_PATH
	else
		echo "please add hugepage setting in /boot/grub2/grub.cfg first."	
		exit
	fi
fi

# load/unload modules
for kmod in openvswitch vfio vfio-pci; do
	if lsmod | grep -q $kmod; then
	echo "not loading $kmod (already loaded)"
else
	if modprobe -v $kmod; then
		echo "loaded $kmod module"
	else
		echo "Failed to load $kmmod module, exiting"
		exit 1
	fi
fi
done

# bind the devices to dpdk module
dpdk_nics=`lspci -D  | grep $eth_model | awk '{print $1}'`
echo DPDK adapters: $dpdk_nics
dpdk_nic_kmod=vfio-pci

for nic in $dpdk_nics; do
	dpdk_nic_bind.py -b $dpdk_nic_kmod $nic
done
dpdk_nic_bind.py --status

# completely remove old ovs configuration
echo "remove old ovs configuration"
killall ovs-vswitchd
killall ovsdb-server
sleep 3
rm -rf /var/run/openvswitch/*
rm -rf /etc/openvswitch/*db*
rm -rf /var/log/openvswitch/*

# start database server
echo "start new ovs"
export DB_SOCK="/var/run/openvswitch/db.sock"
mkdir -p /var/run/openvswitch
mkdir -p /etc/openvswitch
/bin/ovsdb-tool create /etc/openvswitch/conf.db /usr/share/openvswitch/vswitch.ovsschema
rm -rf /dev/usvhost-1
/sbin/ovsdb-server -v --remote=punix:$DB_SOCK \
    --remote=db:Open_vSwitch,Open_vSwitch,manager_options \
    --pidfile --detach || exit 1

# start ovs
echo "openvswitch init"
/bin/ovs-vsctl --no-wait init
screen -dmS ovs sudo su -g qemu -c "umask 002; /sbin/ovs-vswitchd --dpdk -c 0x1 -n 3 --socket-mem 1024,1024 -- unix:$DB_SOCK --pidfile --mlockall --log-file=/var/log/openvswitch/ovs-vswitchd.log"

# configure the bridge
echo "configure the bridge0"
/bin/ovs-vsctl --if-exists del-br ovsbr0
/bin/ovs-vsctl add-br ovsbr0 -- set bridge ovsbr0 datapath_type=netdev
/bin/ovs-vsctl add-port ovsbr0 dpdk0 -- set Interface dpdk0 type=dpdk
/bin/ovs-vsctl add-port ovsbr0 vhost-user1 -- set Interface vhost-user1 type=dpdkvhostuser

echo "configure the bridge1"
/bin/ovs-vsctl --if-exists del-br ovsbr1
/bin/ovs-vsctl add-br ovsbr1 -- set bridge ovsbr1 datapath_type=netdev
/bin/ovs-vsctl add-port ovsbr1 dpdk1 -- set Interface dpdk1 type=dpdk
/bin/ovs-vsctl add-port ovsbr1 vhost-user2 -- set Interface vhost-user2 type=dpdkvhostuser

/bin/ovs-vsctl show

echo "configure flowtable for bridge0"
/bin/ovs-ofctl del-flows ovsbr0
/bin/ovs-ofctl add-flow ovsbr0 "in_port=1,idle_timeout=0 actions=output:2"
/bin/ovs-ofctl add-flow ovsbr0 "in_port=2,idle_timeout=0 actions=output:1"

echo "configure flowtable for bridge1"
/bin/ovs-ofctl del-flows ovsbr1
/bin/ovs-ofctl add-flow ovsbr1 "in_port=1,idle_timeout=0 actions=output:2"
/bin/ovs-ofctl add-flow ovsbr1 "in_port=2,idle_timeout=0 actions=output:1"


2. boot a guest in src host
/usr/libexec/qemu-kvm -name rhel7.2-rt-355 -machine pc-i440fx-rhel7.2.0 -cpu host,+tsc-deadline,pmu=off -m 4096 -realtime mlock=off -smp 4,sockets=4,cores=1,threads=1 \

-object memory-backend-file,id=ram-node0,prealloc=yes,mem-path=/mnt/hugetlbfs,share=on,size=4096M -numa node,nodeid=0,cpus=0-3,memdev=ram-node0 \

 -drive file=/home/rhel7.2-rt-355.qcow2,if=none,id=drive-virtio-disk0,format=qcow2,snapshot=off -device virtio-blk-pci,drive=drive-virtio-disk0,id=virtio-disk0 \

-netdev tap,id=hostnet0,vhost=on -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:a1:d0:5f \

-chardev socket,id=charnet1,path=/var/run/openvswitch/vhost-user1 -netdev type=vhost-user,id=hostnet1,chardev=charnet1 -device virtio-net-pci,netdev=hostnet1,id=net1,mac=52:54:00:b2:02:01 -chardev socket,id=charnet2,path=/var/run/openvswitch/vhost-user2 -netdev type=vhost-user,id=hostnet2,chardev=charnet2 -device virtio-net-pci,netdev=hostnet2,id=net2,mac=52:54:00:b2:02:02 \

-monitor stdio -device qxl-vga,id=video0 -serial unix:/tmp/console,server,nowait -vnc :1 -spice port=5900,disable-ticketing

3. boot a guest in the same host (as dst host), with 'incoming'

4. do local migration, failed
(qemu) migrate -d tcp:10.73.64.240:4444



Actual results:
Error message displayed after step 4 is executed:
migrate: Migration disabled: vhost-user backend lacks VHOST_USER_PROTOCOL_F_LOG_SHMFD feature.

Expected results:
migration works without error

Additional info:
I've tested migration with upstream dpdk and ovs, it works fine.
I'm wondering if downstream gonna support migration or not. If not, please feel free to close this bug.