added new cancel-reduced action thread termination method

We now manage to cancel threads that block inside a retry loop to
terminate without the need to cancel the thread. Avoiding cancellation
helps keep the system complexity minimal and thus provides for better
stability. This also solves some issues with improper shutdown when
inside an action retry loop.

2 files changed, 44 insertions, 0 deletions

diff --git a/runtime/wti.c b/runtime/wti.c
index 44a27b3e..2dfc2d3f 100644
--- a/runtime/wti.c
+++ b/runtime/wti.c
@@ -109,6 +109,29 @@ wtiSetState(wti_t *pThis, sbool bNewVal)
 }
 
 
+/* advise all workers to start by interrupting them. That should unblock all srSleep()
+ * calls.
+ */
+rsRetVal
+wtiWakeupThrd(wti_t *pThis)
+{
+	DEFiRet;
+
+	ISOBJ_TYPE_assert(pThis, wti);
+
+
+	if(wtiGetState(pThis)) {
+		/* we first try the cooperative "cancel" interface */
+		pthread_kill(pThis->thrdID, SIGTTIN);
+		dbgprintf("sent SIGTTIN to worker thread %u, giving it a chance to terminate\n", (unsigned) pThis->thrdID);
+		srSleep(0, 10000);
+		dbgprintf("cooperative worker termination failed, using cancellation...\n");
+	}
+
+	RETiRet;
+}
+
+
 /* Cancel the thread. If the thread is not running. But it is save and legal to
  * call wtiCancelThrd() in such situations. This function only returns when the
  * thread has terminated. Else we may get race conditions all over the code...
@@ -125,7 +148,16 @@ wtiCancelThrd(wti_t *pThis)
 
 	ISOBJ_TYPE_assert(pThis, wti);
 
+
 	if(wtiGetState(pThis)) {
+		/* we first try the cooperative "cancel" interface */
+#if 0
+		pthread_kill(pThis->thrdID, SIGTTIN);
+		dbgprintf("sent SIGTTIN to worker thread %u, giving it a chance to terminate\n", (unsigned) pThis->thrdID);
+		srSleep(0, 10000);
+		dbgprintf("cooperative worker termination failed, using cancellation...\n");
+#endif
+
 		dbgoprint((obj_t*) pThis, "canceling worker thread\n");
 		pthread_cancel(pThis->thrdID);
 		/* now wait until the thread terminates... */
diff --git a/runtime/wtp.c b/runtime/wtp.c
index 51fab191..65155efc 100644
--- a/runtime/wtp.c
+++ b/runtime/wtp.c
@@ -220,6 +220,7 @@ wtpShutdownAll(wtp_t *pThis, wtpState_t tShutdownCmd, struct timespec *ptTimeout
 {
 	DEFiRet;
 	int bTimedOut;
+	int i;
 
 	ISOBJ_TYPE_assert(pThis, wtp);
 
@@ -234,6 +235,11 @@ wtpShutdownAll(wtp_t *pThis, wtpState_t tShutdownCmd, struct timespec *ptTimeout
 	pthread_cleanup_push(mutexCancelCleanup, &pThis->mutWtp);
 	bTimedOut = 0;
 	while(pThis->iCurNumWrkThrd > 0 && !bTimedOut) {
+
+for(i = 0 ; i < pThis->iNumWorkerThreads ; ++i) {
+	wtiWakeupThrd(pThis->pWrkr[i]);
+}
+
 		DBGPRINTF("%s: waiting %ldms on worker thread termination, %d still running\n",
 			   wtpGetDbgHdr(pThis), timeoutVal(ptTimeout),
 			   ATOMIC_FETCH_32BIT(&pThis->iCurNumWrkThrd, &pThis->mutCurNumWrkThrd));
@@ -348,9 +354,15 @@ wtpWorker(void *arg) /* the arg is actually a wti object, even though we are in
 	pThis = pWti->pWtp;
 	ISOBJ_TYPE_assert(pThis, wtp);
 
+	/* block all signals */
 	sigfillset(&sigSet);
 	pthread_sigmask(SIG_BLOCK, &sigSet, NULL);
 
+	/* but ignore SIGTTN, which we (ab)use to signal the thread to shutdown -- rgerhards, 2009-07-20 */
+	sigemptyset(&sigSet);
+	sigaddset(&sigSet, SIGTTIN);
+	pthread_sigmask(SIG_UNBLOCK, &sigSet, NULL);
+
 #	if HAVE_PRCTL && defined PR_SET_NAME
 	/* set thread name - we ignore if the call fails, has no harsh consequences... */
 	pszDbgHdr = wtpGetDbgHdr(pThis);