Fix flakyness in TestWatchpointMultipleThreads

[labath]

Summary

the inferior in the test deliberately does not lock a mutex when accessing the watched variable.
The reason for that is unclear as, based on the logs, the original intention of the test was to
check whether watchpoints get propagated to newly created threads, which should work fine even
with a mutex. Furthermore, in the unlikely event (which I have still observed happening from time
to time) that two threads do manage the execute the "critical section" simultaneously, the test
will fail, as it is expecting the watchpoint "hit count" to be 1, but in this case it will be 2.

Given this, I have simply chose to lock the mutex always, so that we have more predictible
behavior. Watchpoints being hit simultaneously is still (and correctly!) tested by
TestConcurrentEvents.

packages/Python/lldbsuite/test/functionalities/watchpoint/multiple_threads/TestWatchpointMultipleThreads.py

def hello_multiple_threads(self):
 
         # We should be stopped again due to the breakpoint.
         # The stop reason of the thread should be breakpoint.
         self.expect("thread list", STOPPED_DUE_TO_BREAKPOINT,
             substrs = ['stopped',
                        'stop reason = breakpoint'])
 
         # Now let's set a write-type watchpoint for variable 'g_val'.
-        # The main.cpp, by design, misbehaves by not following the agreed upon
-        # protocol of using a mutex while accessing the global pool and by not
-        # writing to the variable.
         self.expect("watchpoint set variable -w write g_val", WATCHPOINT_CREATED,
             substrs = ['Watchpoint created', 'size = 4', 'type = w'])
 
         # Use the '-v' option to do verbose listing of the watchpoint.
         # The hit count should be 0 initially.
         self.expect("watchpoint list -v",
             substrs = ['hit_count = 0'])
 
def hello_multiple_threads_wp_set_and_then_delete(self):
 
         # We should be stopped again due to the breakpoint.
         # The stop reason of the thread should be breakpoint.
         self.expect("thread list", STOPPED_DUE_TO_BREAKPOINT,
             substrs = ['stopped',
                        'stop reason = breakpoint'])
 
         # Now let's set a write-type watchpoint for variable 'g_val'.
-        # The main.cpp, by design, misbehaves by not following the agreed upon
-        # protocol of using a mutex while accessing the global pool and by not
-        # writing to the variable.
         self.expect("watchpoint set variable -w write g_val", WATCHPOINT_CREATED,
             substrs = ['Watchpoint created', 'size = 4', 'type = w'])
 
         # Use the '-v' option to do verbose listing of the watchpoint.
         # The hit count should be 0 initially.
         self.expect("watchpoint list -v",
             substrs = ['hit_count = 0'])
 

packages/Python/lldbsuite/test/functionalities/watchpoint/multiple_threads/main.cpp

 
 uint32_t g_val = 0;
 
 
 uint32_t
 access_pool (bool flag = false)
 {
     static std::mutex g_access_mutex;
-    if (!flag)
-        g_access_mutex.lock();
 
     uint32_t old_val = g_val;
     if (flag)
     {
         printf("changing g_val to %d...\n", old_val + 1);
         g_val = old_val + 1;
     }
 
-    if (!flag)
-        g_access_mutex.unlock();
     return g_val;
 }
 
 void
 thread_func (uint32_t thread_index)
 {
-    // Break here in order to allow the thread
-    // to inherit the global watchpoint state.
-    printf ("%s (thread index = %u) startng...\n", __FUNCTION__, thread_index);
+    printf ("%s (thread index = %u) starting...\n", __FUNCTION__, thread_index);
 
     uint32_t count = 0;
     uint32_t val;
     while (count++ < 15)
     {
         // random micro second sleep from zero to 3 seconds
         int usec = g_distribution(g_random_engine);
         printf ("%s (thread = %u) doing a usleep (%d)...\n", __FUNCTION__, thread_index, usec);