Description of problem:
running x11perf -all within a Spiced (with QXL) F15 guest, sometimes the guest is stuck. In the stdio I see:
flush_display_commands: update timeout

I could not reproduce it with upstream code, but it may be just a race somewhere.

There are two threads which seems to be spinning and competing.


Some gdb output of the two stuck threads:

(gdb) bt
#0  0x0000003be1e0e4ed in read () at ../sysdeps/unix/syscall-template.S:82
#1  0x0000003cbfe1271a in read (__nbytes=4, __buf=0x7fc09cee89ec, __fd=14) at /usr/include/bits/unistd.h:45
#2  receive_data (n=4, in_buf=0x7fc09cee89ec, fd=14) at red_worker.h:131
#3  read_message (message=0x7fc09cee89ec, fd=14) at red_worker.h:144
#4  qxl_worker_update_area (qxl_worker=<optimized out>, surface_id=0, qxl_area=<optimized out>, qxl_dirty_rects=0x0, num_dirty_rects=0, clear_dirty_region=0) at red_dispatcher.c:227
#5  0x00000000005db1a4 in ioport_write (opaque=0x3612be0, addr=<optimized out>, val=0) at /usr/src/debug/qemu-kvm-0.14.0/hw/qxl.c:991
#6  0x000000000042f4df in kvm_handle_io (count=1, size=1, direction=1, data=<optimized out>, port=49218) at /usr/src/debug/qemu-kvm-0.14.0/kvm-all.c:818
#7  kvm_run (env=0x31e6b10) at /usr/src/debug/qemu-kvm-0.14.0/qemu-kvm.c:617
#8  0x0000000000430a39 in kvm_cpu_exec (env=<optimized out>) at /usr/src/debug/qemu-kvm-0.14.0/qemu-kvm.c:1233
#9  0x0000000000430cc1 in kvm_main_loop_cpu (env=0x31e6b10) at /usr/src/debug/qemu-kvm-0.14.0/qemu-kvm.c:1419
#10 ap_main_loop (_env=0x31e6b10) at /usr/src/debug/qemu-kvm-0.14.0/qemu-kvm.c:1466
#11 0x0000003be1e07cd1 in start_thread (arg=0x7fc09cee9700) at pthread_create.c:305
#12 0x0000003be1adfd3d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115
(gdb) thread 14
[Switching to thread 14 (Thread 0x7fc04eff8700 (LWP 21478))]
#0  0x0000003be1e0e48d in write () at ../sysdeps/unix/syscall-template.S:82
82	T_PSEUDO (SYSCALL_SYMBOL, SYSCALL_NAME, SYSCALL_NARGS)
(gdb) bt
#0  0x0000003be1e0e48d in write () at ../sysdeps/unix/syscall-template.S:82
#1  0x0000003cbfe1d9c2 in send_data (n=4, in_buf=<optimized out>, fd=15) at red_worker.h:110
#2  write_message (fd=15, message=<optimized out>) at red_worker.h:123
#3  0x0000003cbfe2f2d5 in handle_dev_update (worker=0x7fc04ee23890) at red_worker.c:9738
#4  handle_dev_input (listener=0x7fc04ee23890, events=<optimized out>) at red_worker.c:9936
#5  0x0000003cbfe30385 in red_worker_main (arg=<optimized out>) at red_worker.c:10286
#6  0x0000003be1e07cd1 in start_thread (arg=0x7fc04eff8700) at pthread_create.c:305
#7  0x0000003be1adfd3d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115

Version-Release number of selected component (if applicable):
[ykaul@ykaul wireshark]$ rpm -qa |egrep "spice|qemu|kvm"
qemu-common-0.14.0-7.fc15.x86_64
spice-gtk-0.5-6.fc15.x86_64
qemu-img-0.14.0-7.fc15.x86_64
qemu-debuginfo-0.14.0-7.fc15.x86_64
spice-gtk3-0.5-6.fc15.x86_64
spice-server-0.8.1-1.fc15.x86_64
spice-gtk-python-0.5-6.fc15.x86_64
qemu-system-x86-0.14.0-7.fc15.x86_64
qemu-kvm-0.14.0-7.fc15.x86_64
spice-debuginfo-0.8.1-1.fc15.x86_64
gpxe-roms-qemu-1.0.1-4.fc15.noarch
spice-glib-0.5-6.fc15.x86_64

How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info: