1 files changed, 58 insertions, 61 deletions

diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h
index 806160c44b17..7da3c81c3f59 100644
--- a/kernel/rcu/tasks.h
+++ b/kernel/rcu/tasks.h
@@ -197,6 +197,7 @@ static int __noreturn rcu_tasks_kthread(void *arg)
 	 * This loop is terminated by the system going down.  ;-)
 	 */
 	for (;;) {
+		set_tasks_gp_state(rtp, RTGS_WAIT_CBS);
 
 		/* Pick up any new callbacks. */
 		raw_spin_lock_irqsave(&rtp->cbs_lock, flags);
@@ -236,8 +237,6 @@ static int __noreturn rcu_tasks_kthread(void *arg)
 		}
 		/* Paranoid sleep to keep this from entering a tight loop */
 		schedule_timeout_idle(rtp->gp_sleep);
-
-		set_tasks_gp_state(rtp, RTGS_WAIT_CBS);
 	}
 }
 
@@ -369,7 +368,7 @@ static void rcu_tasks_wait_gp(struct rcu_tasks *rtp)
 ////////////////////////////////////////////////////////////////////////
 //
 // Simple variant of RCU whose quiescent states are voluntary context
-// switch, cond_resched_rcu_qs(), user-space execution, and idle.
+// switch, cond_resched_tasks_rcu_qs(), user-space execution, and idle.
 // As such, grace periods can take one good long time.  There are no
 // read-side primitives similar to rcu_read_lock() and rcu_read_unlock()
 // because this implementation is intended to get the system into a safe
@@ -540,7 +539,7 @@ DEFINE_RCU_TASKS(rcu_tasks, rcu_tasks_wait_gp, call_rcu_tasks, "RCU Tasks");
  * period elapses, in other words after all currently executing RCU
  * read-side critical sections have completed. call_rcu_tasks() assumes
  * that the read-side critical sections end at a voluntary context
- * switch (not a preemption!), cond_resched_rcu_qs(), entry into idle,
+ * switch (not a preemption!), cond_resched_tasks_rcu_qs(), entry into idle,
  * or transition to usermode execution.  As such, there are no read-side
  * primitives analogous to rcu_read_lock() and rcu_read_unlock() because
  * this primitive is intended to determine that all tasks have passed
@@ -678,11 +677,11 @@ DEFINE_RCU_TASKS(rcu_tasks_rude, rcu_tasks_rude_wait_gp, call_rcu_tasks_rude,
  * period elapses, in other words after all currently executing RCU
  * read-side critical sections have completed. call_rcu_tasks_rude()
  * assumes that the read-side critical sections end at context switch,
- * cond_resched_rcu_qs(), or transition to usermode execution.  As such,
- * there are no read-side primitives analogous to rcu_read_lock() and
- * rcu_read_unlock() because this primitive is intended to determine
- * that all tasks have passed through a safe state, not so much for
- * data-structure synchronization.
+ * cond_resched_tasks_rcu_qs(), or transition to usermode execution (as
+ * usermode execution is schedulable). As such, there are no read-side
+ * primitives analogous to rcu_read_lock() and rcu_read_unlock() because
+ * this primitive is intended to determine that all tasks have passed
+ * through a safe state, not so much for data-structure synchronization.
  *
  * See the description of call_rcu() for more detailed information on
  * memory ordering guarantees.
@@ -700,8 +699,8 @@ EXPORT_SYMBOL_GPL(call_rcu_tasks_rude);
  * grace period has elapsed, in other words after all currently
  * executing rcu-tasks read-side critical sections have elapsed.  These
  * read-side critical sections are delimited by calls to schedule(),
- * cond_resched_tasks_rcu_qs(), userspace execution, and (in theory,
- * anyway) cond_resched().
+ * cond_resched_tasks_rcu_qs(), userspace execution (which is a schedulable
+ * context), and (in theory, anyway) cond_resched().
  *
  * This is a very specialized primitive, intended only for a few uses in
  * tracing and other situations requiring manipulation of function preambles
@@ -758,7 +757,7 @@ EXPORT_SYMBOL_GPL(show_rcu_tasks_rude_gp_kthread);
 // 2.	Protects code in the idle loop, exception entry/exit, and
 //	CPU-hotplug code paths, similar to the capabilities of SRCU.
 //
-// 3.	Avoids expensive read-side instruction, having overhead similar
+// 3.	Avoids expensive read-side instructions, having overhead similar
 //	to that of Preemptible RCU.
 //
 // There are of course downsides.  The grace-period code can send IPIs to
@@ -848,7 +847,7 @@ static void rcu_read_unlock_iw(struct irq_work *iwp)
 static DEFINE_IRQ_WORK(rcu_tasks_trace_iw, rcu_read_unlock_iw);
 
 /* If we are the last reader, wake up the grace-period kthread. */
-void rcu_read_unlock_trace_special(struct task_struct *t, int nesting)
+void rcu_read_unlock_trace_special(struct task_struct *t)
 {
 	int nq = READ_ONCE(t->trc_reader_special.b.need_qs);
 
@@ -858,7 +857,7 @@ void rcu_read_unlock_trace_special(struct task_struct *t, int nesting)
 	// Update .need_qs before ->trc_reader_nesting for irq/NMI handlers.
 	if (nq)
 		WRITE_ONCE(t->trc_reader_special.b.need_qs, false);
-	WRITE_ONCE(t->trc_reader_nesting, nesting);
+	WRITE_ONCE(t->trc_reader_nesting, 0);
 	if (nq && atomic_dec_and_test(&trc_n_readers_need_end))
 		irq_work_queue(&rcu_tasks_trace_iw);
 }
@@ -890,32 +889,24 @@ static void trc_read_check_handler(void *t_in)
 
 	// If the task is no longer running on this CPU, leave.
 	if (unlikely(texp != t)) {
-		if (WARN_ON_ONCE(atomic_dec_and_test(&trc_n_readers_need_end)))
-			wake_up(&trc_wait);
 		goto reset_ipi; // Already on holdout list, so will check later.
 	}
 
 	// If the task is not in a read-side critical section, and
 	// if this is the last reader, awaken the grace-period kthread.
 	if (likely(!READ_ONCE(t->trc_reader_nesting))) {
-		if (WARN_ON_ONCE(atomic_dec_and_test(&trc_n_readers_need_end)))
-			wake_up(&trc_wait);
-		// Mark as checked after decrement to avoid false
-		// positives on the above WARN_ON_ONCE().
 		WRITE_ONCE(t->trc_reader_checked, true);
 		goto reset_ipi;
 	}
 	// If we are racing with an rcu_read_unlock_trace(), try again later.
-	if (unlikely(READ_ONCE(t->trc_reader_nesting) < 0)) {
-		if (WARN_ON_ONCE(atomic_dec_and_test(&trc_n_readers_need_end)))
-			wake_up(&trc_wait);
+	if (unlikely(READ_ONCE(t->trc_reader_nesting) < 0))
 		goto reset_ipi;
-	}
 	WRITE_ONCE(t->trc_reader_checked, true);
 
 	// Get here if the task is in a read-side critical section.  Set
 	// its state so that it will awaken the grace-period kthread upon
 	// exit from that critical section.
+	atomic_inc(&trc_n_readers_need_end); // One more to wait on.
 	WARN_ON_ONCE(READ_ONCE(t->trc_reader_special.b.need_qs));
 	WRITE_ONCE(t->trc_reader_special.b.need_qs, true);
 
@@ -928,10 +919,10 @@ reset_ipi:
 }
 
 /* Callback function for scheduler to check locked-down task.  */
-static bool trc_inspect_reader(struct task_struct *t, void *arg)
+static int trc_inspect_reader(struct task_struct *t, void *arg)
 {
 	int cpu = task_cpu(t);
-	bool in_qs = false;
+	int nesting;
 	bool ofl = cpu_is_offline(cpu);
 
 	if (task_curr(t)) {
@@ -939,7 +930,7 @@ static bool trc_inspect_reader(struct task_struct *t, void *arg)
 
 		// If no chance of heavyweight readers, do it the hard way.
 		if (!ofl && !IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB))
-			return false;
+			return -EINVAL;
 
 		// If heavyweight readers are enabled on the remote task,
 		// we can inspect its state despite its currently running.
@@ -947,22 +938,22 @@ static bool trc_inspect_reader(struct task_struct *t, void *arg)
 		n_heavy_reader_attempts++;
 		if (!ofl && // Check for "running" idle tasks on offline CPUs.
 		    !rcu_dynticks_zero_in_eqs(cpu, &t->trc_reader_nesting))
-			return false; // No quiescent state, do it the hard way.
+			return -EINVAL; // No quiescent state, do it the hard way.
 		n_heavy_reader_updates++;
 		if (ofl)
 			n_heavy_reader_ofl_updates++;
-		in_qs = true;
+		nesting = 0;
 	} else {
 		// The task is not running, so C-language access is safe.
-		in_qs = likely(!t->trc_reader_nesting);
+		nesting = t->trc_reader_nesting;
 	}
 
-	// Mark as checked so that the grace-period kthread will
-	// remove it from the holdout list.
-	t->trc_reader_checked = true;
-
-	if (in_qs)
-		return true;  // Already in quiescent state, done!!!
+	// If not exiting a read-side critical section, mark as checked
+	// so that the grace-period kthread will remove it from the
+	// holdout list.
+	t->trc_reader_checked = nesting >= 0;
+	if (nesting <= 0)
+		return nesting ? -EINVAL : 0;  // If in QS, done, otherwise try again later.
 
 	// The task is in a read-side critical section, so set up its
 	// state so that it will awaken the grace-period kthread upon exit
@@ -970,7 +961,7 @@ static bool trc_inspect_reader(struct task_struct *t, void *arg)
 	atomic_inc(&trc_n_readers_need_end); // One more to wait on.
 	WARN_ON_ONCE(READ_ONCE(t->trc_reader_special.b.need_qs));
 	WRITE_ONCE(t->trc_reader_special.b.need_qs, true);
-	return true;
+	return 0;
 }
 
 /* Attempt to extract the state for the specified task. */
@@ -992,7 +983,7 @@ static void trc_wait_for_one_reader(struct task_struct *t,
 
 	// Attempt to nail down the task for inspection.
 	get_task_struct(t);
-	if (try_invoke_on_locked_down_task(t, trc_inspect_reader, NULL)) {
+	if (!task_call_func(t, trc_inspect_reader, NULL)) {
 		put_task_struct(t);
 		return;
 	}
@@ -1000,7 +991,7 @@ static void trc_wait_for_one_reader(struct task_struct *t,
 
 	// If this task is not yet on the holdout list, then we are in
 	// an RCU read-side critical section.  Otherwise, the invocation of
-	// rcu_add_holdout() that added it to the list did the necessary
+	// trc_add_holdout() that added it to the list did the necessary
 	// get_task_struct().  Either way, the task cannot be freed out
 	// from under this code.
 
@@ -1015,21 +1006,17 @@ static void trc_wait_for_one_reader(struct task_struct *t,
 		if (per_cpu(trc_ipi_to_cpu, cpu) || t->trc_ipi_to_cpu >= 0)
 			return;
 
-		atomic_inc(&trc_n_readers_need_end);
 		per_cpu(trc_ipi_to_cpu, cpu) = true;
 		t->trc_ipi_to_cpu = cpu;
 		rcu_tasks_trace.n_ipis++;
-		if (smp_call_function_single(cpu,
-					     trc_read_check_handler, t, 0)) {
+		if (smp_call_function_single(cpu, trc_read_check_handler, t, 0)) {
 			// Just in case there is some other reason for
 			// failure than the target CPU being offline.
+			WARN_ONCE(1, "%s():  smp_call_function_single() failed for CPU: %d\n",
+				  __func__, cpu);
 			rcu_tasks_trace.n_ipis_fails++;
 			per_cpu(trc_ipi_to_cpu, cpu) = false;
-			t->trc_ipi_to_cpu = cpu;
-			if (atomic_dec_and_test(&trc_n_readers_need_end)) {
-				WARN_ON_ONCE(1);
-				wake_up(&trc_wait);
-			}
+			t->trc_ipi_to_cpu = -1;
 		}
 	}
 }
@@ -1099,9 +1086,9 @@ static void show_stalled_task_trace(struct task_struct *t, bool *firstreport)
 	cpu = task_cpu(t);
 	pr_alert("P%d: %c%c%c nesting: %d%c cpu: %d\n",
 		 t->pid,
-		 ".I"[READ_ONCE(t->trc_ipi_to_cpu) > 0],
+		 ".I"[READ_ONCE(t->trc_ipi_to_cpu) >= 0],
 		 ".i"[is_idle_task(t)],
-		 ".N"[cpu > 0 && tick_nohz_full_cpu(cpu)],
+		 ".N"[cpu >= 0 && tick_nohz_full_cpu(cpu)],
 		 READ_ONCE(t->trc_reader_nesting),
 		 " N"[!!READ_ONCE(t->trc_reader_special.b.need_qs)],
 		 cpu);
@@ -1144,20 +1131,34 @@ static void check_all_holdout_tasks_trace(struct list_head *hop,
 	cpus_read_unlock();
 
 	if (needreport) {
-		if (firstreport)
+		if (*firstreport)
 			pr_err("INFO: rcu_tasks_trace detected stalls? (Late IPI?)\n");
 		show_stalled_ipi_trace();
 	}
 }
 
+static void rcu_tasks_trace_empty_fn(void *unused)
+{
+}
+
 /* Wait for grace period to complete and provide ordering. */
 static void rcu_tasks_trace_postgp(struct rcu_tasks *rtp)
 {
+	int cpu;
 	bool firstreport;
 	struct task_struct *g, *t;
 	LIST_HEAD(holdouts);
 	long ret;
 
+	// Wait for any lingering IPI handlers to complete.  Note that
+	// if a CPU has gone offline or transitioned to userspace in the
+	// meantime, all IPI handlers should have been drained beforehand.
+	// Yes, this assumes that CPUs process IPIs in order.  If that ever
+	// changes, there will need to be a recheck and/or timed wait.
+	for_each_online_cpu(cpu)
+		if (smp_load_acquire(per_cpu_ptr(&trc_ipi_to_cpu, cpu)))
+			smp_call_function_single(cpu, rcu_tasks_trace_empty_fn, NULL, 1);
+
 	// Remove the safety count.
 	smp_mb__before_atomic();  // Order vs. earlier atomics
 	atomic_dec(&trc_n_readers_need_end);
@@ -1200,7 +1201,7 @@ static void exit_tasks_rcu_finish_trace(struct task_struct *t)
 	WARN_ON_ONCE(READ_ONCE(t->trc_reader_nesting));
 	WRITE_ONCE(t->trc_reader_nesting, 0);
 	if (WARN_ON_ONCE(READ_ONCE(t->trc_reader_special.b.need_qs)))
-		rcu_read_unlock_trace_special(t, 0);
+		rcu_read_unlock_trace_special(t);
 }
 
 /**
@@ -1208,15 +1209,11 @@ static void exit_tasks_rcu_finish_trace(struct task_struct *t)
  * @rhp: structure to be used for queueing the RCU updates.
  * @func: actual callback function to be invoked after the grace period
  *
- * The callback function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed. call_rcu_tasks_trace()
- * assumes that the read-side critical sections end at context switch,
- * cond_resched_rcu_qs(), or transition to usermode execution.  As such,
- * there are no read-side primitives analogous to rcu_read_lock() and
- * rcu_read_unlock() because this primitive is intended to determine
- * that all tasks have passed through a safe state, not so much for
- * data-structure synchronization.
+ * The callback function will be invoked some time after a trace rcu-tasks
+ * grace period elapses, in other words after all currently executing
+ * trace rcu-tasks read-side critical sections have completed. These
+ * read-side critical sections are delimited by calls to rcu_read_lock_trace()
+ * and rcu_read_unlock_trace().
  *
  * See the description of call_rcu() for more detailed information on
  * memory ordering guarantees.
@@ -1232,7 +1229,7 @@ EXPORT_SYMBOL_GPL(call_rcu_tasks_trace);
  *
  * Control will return to the caller some time after a trace rcu-tasks
  * grace period has elapsed, in other words after all currently executing
- * rcu-tasks read-side critical sections have elapsed.  These read-side
+ * trace rcu-tasks read-side critical sections have elapsed. These read-side
  * critical sections are delimited by calls to rcu_read_lock_trace()
  * and rcu_read_unlock_trace().
  *