-net socket,connect/listen does not work in 2.7.0

Using 2.7.0 release on Debian Sid. What I did: start one VM with:
----
/home/pierre/build/qemu/build/x86_64-softmmu/qemu-system-x86_64 \
-smp 4 \
-cpu Nehalem \
-soundhw ac97 \
-k fr \
-localtime \
-enable-kvm \
-m 4099 \
-drive file=/mnt/virtualMachines/qemu/lfs-7.10-porg.qcow2,cache=writeback \
-cdrom /mnt/virtualMachines/qemu/grub-img.iso \
-boot order=c,once=d,menu=on \
-vga std \
-serial mon:stdio \
-net nic,vlan=0,model=e1000,macaddr=52:54:00:12:34:58 \
-net user,vlan=0,hostfwd=tcp::2223-10.0.2.9:22 \
-net nic,vlan=1,model=e1000,macaddr=52:54:00:12:34:56 \
-net socket,vlan=1,listen=:4321
----
Start another one with:
----
/home/pierre/build/qemu/build/x86_64-softmmu/qemu-system-x86_64 \
-smp 4 \
-cpu Nehalem \
-soundhw ac97 \
-k fr \
-localtime \
-enable-kvm \
-m 4099 \
-drive file=/mnt/virtualMachines/qemu/lfs-7.10-october.qcow2,cache=writeback \
-cdrom /mnt/virtualMachines/qemu/grub-img.iso \
-boot order=c \
-serial mon:stdio \
-vga std \
-net nic,vlan=0,model=e1000,macaddr=52:54:00:12:34:57 \
-net socket,vlan=0,connect=localhost:4321
----
The network settings of the first machine are:
----
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 52:54:00:12:34:58 brd ff:ff:ff:ff:ff:ff
3: enp0s4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master br0 state UP group default qlen 1000
    link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff
4: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff
    inet 10.0.2.9/24 brd 10.0.2.255 scope global br0
       valid_lft forever preferred_lft forever
----
The network settings on the second machine are:
----
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 52:54:00:12:34:57 brd ff:ff:ff:ff:ff:ff
    inet 10.0.2.10/24 brd 10.0.2.255 scope global enp0s3
       valid_lft forever preferred_lft forever
----
typing "ping -c 1 10.0.2.10" on the first machine returns:
----
PING 10.0.2.10 (10.0.2.10): 56 data bytes
92 bytes from virtuallfs (10.0.2.9): Destination Host Unreachable
--- 10.0.2.10 ping statistics ---
1 packets transmitted, 0 packets received, 100% packet loss
----
and something similar when typing "ping -c 1 10.0.2.9" on the second machine.

This very same setting works as expected in version 2.6.0. I could bisect, and the offending commit is 16a3df403b1:
----
commit 16a3df403b10c4ac347159e39005fd520b2648bb
Author: Zhang Chen <email address hidden>
Date:   Fri May 13 15:35:19 2016 +0800

    net/net: Add SocketReadState for reuse codes
    
    This function is from net/socket.c, move it to net.c and net.h.
    Add SocketReadState to make others reuse net_fill_rstate().
    suggestion from jason.
    
    v4:
     - move 'rs->finalize = finalize' to rs_init()
    
    v3:
     - remove SocketReadState init callback
     - put finalize callback to net_fill_rstate()
    
    v2:
     - rename ReadState to SocketReadState
     - add SocketReadState init and finalize callback
    
    v1:
     - init patch
    
    Signed-off-by: Zhang Chen <email address hidden>
    Signed-off-by: Li Zhijian <email address hidden>
    Signed-off-by: Wen Congyang <email address hidden>
    Signed-off-by: Jason Wang <email address hidden>
----

BTW, the systems on both VM are built from http://www.linuxfromscratch.org. But I do not think this is important, since I could do the bisect. Of course, I'll be happy to try other VMs, if you point me to some.