[tools] Make killing of processes more robust

We spawn individual tests in their own shell, and then just kill that
shell later. This often leaves the tests running (see linked bugs).
By spawning the shell in its own new process group, we can just kill
that whole process group later, which seems to work reliably for hanging
tests.

R=machenbach@chromium.org

Bug: v8:8292, v8:8700
Change-Id: I6e38467d687cc0b395467d4b377644de7700f066
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2274634Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#68634}

tools/testrunner/local/command.py

@@ -200,13 +200,17 @@ class PosixCommand(BaseCommand):
         stderr=subprocess.PIPE,
         env=self._get_env(),
         shell=True,
+        # Make the new shell create its own process group. This allows to kill
+        # all spawned processes reliably (https://crbug.com/v8/8292).
+        preexec_fn=os.setsid,
       )
     except Exception as e:
       sys.stderr.write('Error executing: %s\n' % self)
       raise e
 
   def _kill_process(self, process):
-    process.kill()
+    # Kill the whole process group (PID == GPID after setsid).
+    os.killpg(process.pid, signal.SIGKILL)
 
 
 def taskkill_windows(process, verbose=False, force=True):