[Qemu-devel] [PATCH 2/2] cpus: be more paranoid in avoiding deadlocks

[Paolo Bonzini <pbonzini@redhat.com>]

For good measure, ensure that the following sequence:

   thread 1 calls qemu_mutex_lock_iothread
   thread 2 calls qemu_mutex_lock_iothread
   VCPU thread are created
   VCPU thread enters execution loop

results in the VCPU threads letting the other two threads run
and obeying iothread_requesting_mutex even if the VCPUs are
not halted.  To do this, check iothread_requesting_mutex
before execution starts.

I could not reproduce this problem; it was only found by
inspection.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 cpus.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/cpus.c b/cpus.c
index 83c078e..0fac143 100644
--- a/cpus.c
+++ b/cpus.c
@@ -1025,6 +1025,9 @@ static void *qemu_tcg_cpu_thread_fn(void *arg)
         }
     }
 
+    /* process any pending work */
+    exit_request = 1;
+
     while (1) {
         tcg_exec_all();
 
@@ -1115,10 +1118,11 @@ bool qemu_in_vcpu_thread(void)
 
 void qemu_mutex_lock_iothread(void)
 {
+    atomic_inc(&iothread_requesting_mutex);
     if (!tcg_enabled() || !first_cpu) {
         qemu_mutex_lock(&qemu_global_mutex);
+        atomic_dec(&iothread_requesting_mutex);
     } else {
-        atomic_inc(&iothread_requesting_mutex);
         if (qemu_mutex_trylock(&qemu_global_mutex)) {
             qemu_cpu_kick_thread(first_cpu);
             qemu_mutex_lock(&qemu_global_mutex);
-- 
2.3.0