tests: Handle test retries through the same queue

This removes the separate rerun step from the parallel-vm.py processing
and instead, simply requeues the failed test cases into the same queue
that is used for the initial run. This is simpler and more efficient
since reruns start as soon as any VM is ready for processing them
instead of having to wait for all VMs to complete the first round.
Furthermore, this allows VMs to be stopped sooner when no more test
cases remain and that is helpful especially with the time travel patches
that make the wait-for-next-test step in the VM use all available CPU.

Signed-off-by: Jouni Malinen <j@w1.fi>

tests/hwsim/vm/parallel-vm.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 #
 # Parallel VM test case executor
-# Copyright (c) 2014-2018, Jouni Malinen <j@w1.fi>
+# Copyright (c) 2014-2019, Jouni Malinen <j@w1.fi>
 #
 # This software may be distributed under the terms of the BSD license.
 # See README for more details.
@@ -85,9 +85,10 @@ def get_failed(vm):
         failed += vm[i]['failed']
     return failed
 
-def vm_read_stdout(vm, i):
+def vm_read_stdout(vm, i, test_queue):
     global total_started, total_passed, total_failed, total_skipped
     global rerun_failures
+    global first_run_failures
 
     ready = False
     try:
@@ -125,6 +126,15 @@ def vm_read_stdout(vm, i):
                 name = vals[1]
             logger.debug("VM[%d] test case failed: %s" % (i, name))
             vm['failed'].append(name)
+            if name != vm['current_name']:
+                logger.info("VM[%d] test result mismatch: %s (expected %s)" % (i, name, vm['current_name']))
+            else:
+                count = vm['current_count']
+                if count == 0:
+                    first_run_failures.append(name)
+                if rerun_failures and count < 1:
+                    logger.debug("Requeue test case %s" % name)
+                    test_queue.append((name, vm['current_count'] + 1))
         elif line.startswith("NOT-FOUND"):
             ready = True
             total_failed += 1
@@ -151,9 +161,11 @@ def show_progress(scr):
     global tests
     global first_run_failures
     global total_started, total_passed, total_failed, total_skipped
+    global rerun_failures
 
     total_tests = len(tests)
     logger.info("Total tests: %d" % total_tests)
+    test_queue = [(t, 0) for t in tests]
 
     scr.leaveok(1)
     scr.addstr(0, 0, "Parallel test execution status", curses.A_BOLD)
@@ -164,19 +176,11 @@ def show_progress(scr):
     scr.addstr(num_servers + 1, 20, "TOTAL={} STARTED=0 PASS=0 FAIL=0 SKIP=0".format(total_tests))
     scr.refresh()
 
-    completed_first_pass = False
-    rerun_tests = []
-
     while True:
         running = False
-        first_running = False
         updated = False
 
-        for i in range(0, num_servers):
+        for i in range(num_servers):
-            if completed_first_pass:
-                continue
-            if vm[i]['first_run_done']:
-                continue
             if not vm[i]['proc']:
                 continue
             if vm[i]['proc'].poll() is not None:
@@ -188,72 +192,43 @@ def show_progress(scr):
                     if "Kernel panic" in f.read():
                         scr.addstr("kernel panic")
                         logger.info("VM[%d] kernel panic" % i)
-                    else:
+                        updated = True
+                    if test_queue:
                         scr.addstr("unexpected exit")
                         logger.info("VM[%d] unexpected exit" % i)
                         updated = True
                 continue
 
             running = True
-            first_running = True
             try:
                 err = vm[i]['proc'].stderr.read()
                 if err != None:
                     err = err.decode()
                     vm[i]['err'] += err
-                    logger.debug("VM[%d] stderr.read[%s]" % (i, err))
+                    logger.info("VM[%d] stderr.read[%s]" % (i, err))
             except IOError as e:
                 if e.errno != errno.EAGAIN:
                     raise
 
-            if vm_read_stdout(vm[i], i):
+            if vm_read_stdout(vm[i], i, test_queue):
                 scr.move(i + 1, 10)
                 scr.clrtoeol()
                 updated = True
-                if not tests:
+                if not test_queue:
-                    vm[i]['first_run_done'] = True
+                    vm[i]['proc'].stdin.write(b'\n')
-                    scr.addstr("completed first round")
+                    vm[i]['proc'].stdin.flush()
-                    logger.info("VM[%d] completed first round" % i)
+                    scr.addstr("shutting down")
+                    logger.info("VM[%d] shutting down" % i)
                     continue
                 else:
-                    name = tests.pop(0)
+                    (name, count) = test_queue.pop(0)
+                    vm[i]['current_name'] = name
+                    vm[i]['current_count'] = count
                     vm[i]['proc'].stdin.write(name.encode() + b'\n')
                     vm[i]['proc'].stdin.flush()
                     scr.addstr(name)
                     logger.debug("VM[%d] start test %s" % (i, name))
 
-        if not first_running and not completed_first_pass:
-            logger.info("First round of testing completed")
-            if tests:
-                logger.info("Unexpected test cases remaining from first round: " + str(tests))
-                raise Exception("Unexpected test cases remaining from first round")
-            completed_first_pass = True
-            for name in get_failed(vm):
-                if rerun_failures:
-                    rerun_tests.append(name)
-                first_run_failures.append(name)
-
-        for i in range(num_servers):
-            if not completed_first_pass:
-                continue
-            if not vm[i]['proc']:
-                continue
-            if vm[i]['proc'].poll() is not None:
-                vm[i]['proc'] = None
-                scr.move(i + 1, 10)
-                scr.clrtoeol()
-                log = '{}/{}.srv.{}/console'.format(dir, timestamp, i + 1)
-                with open(log, 'r') as f:
-                    if "Kernel panic" in f.read():
-                        scr.addstr("kernel panic")
-                        logger.info("VM[%d] kernel panic" % i)
-                    else:
-                        scr.addstr("completed run")
-                        logger.info("VM[%d] completed run" % i)
-                updated = True
-                continue
-
-            running = True
             try:
                 err = vm[i]['proc'].stderr.read()
                 if err != None:
@@ -264,28 +239,6 @@ def show_progress(scr):
                 if e.errno != errno.EAGAIN:
                     raise
 
-            ready = False
-            if vm[i]['first_run_done']:
-                vm[i]['first_run_done'] = False
-                ready = True
-            else:
-                ready = vm_read_stdout(vm[i], i)
-            if ready:
-                scr.move(i + 1, 10)
-                scr.clrtoeol()
-                updated = True
-                if not rerun_tests:
-                    vm[i]['proc'].stdin.write(b'\n')
-                    vm[i]['proc'].stdin.flush()
-                    scr.addstr("shutting down")
-                    logger.info("VM[%d] shutting down" % i)
-                else:
-                    name = rerun_tests.pop(0)
-                    vm[i]['proc'].stdin.write(name.encode() + b'\n')
-                    vm[i]['proc'].stdin.flush()
-                    scr.addstr(name + "(*)")
-                    logger.debug("VM[%d] start test %s (*)" % (i, name))
-
         if not running:
             break
 
@@ -309,19 +262,19 @@ def show_progress(scr):
                     scr.addstr(f)
                     scr.addstr(' ')
 
-            scr.move(0, 35)
-            scr.clrtoeol()
-            if rerun_tests:
-                scr.addstr("(RETRY FAILED %d)" % len(rerun_tests))
-            elif rerun_failures:
-                pass
-            elif first_run_failures:
-                scr.addstr("(RETRY FAILED)")
-
             scr.refresh()
 
         time.sleep(0.25)
 
+    for i in range(num_servers):
+        if not vm[i]['proc']:
+            continue
+        vm[i]['proc'] = None
+        scr.move(i + 1, 10)
+        scr.clrtoeol()
+        scr.addstr("still running")
+        logger.info("VM[%d] still running" % i)
+
     scr.refresh()
     time.sleep(0.3)
 
@@ -455,7 +408,6 @@ def main():
         if args.telnet:
             cmd += ['--telnet', str(args.telnet + i)]
         vm[i] = {}
-        vm[i]['first_run_done'] = False
         vm[i]['proc'] = subprocess.Popen(cmd,
                                          stdin=subprocess.PIPE,
                                          stdout=subprocess.PIPE,
@@ -537,7 +489,7 @@ def main():
     print("Logs: " + dir + '/' + str(timestamp))
     logger.info("Logs: " + dir + '/' + str(timestamp))
 
-    for i in range(0, num_servers):
+    for i in range(num_servers):
         if len(vm[i]['pending']) > 0:
             logger.info("Unprocessed stdout from VM[%d]: '%s'" %
                         (i, vm[i]['pending']))