regmap: Merge up v6.4-rc3

Merge up v6.4-rc3 to get fixes which make my CI more stable.

6 files changed, 167 insertions, 56 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 68baa8194d9f..db016e418931 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -10150,8 +10150,20 @@ void perf_tp_event(u16 event_type, u64 count, void *record, int entry_size,
 	perf_trace_buf_update(record, event_type);
 
 	hlist_for_each_entry_rcu(event, head, hlist_entry) {
-		if (perf_tp_event_match(event, &data, regs))
+		if (perf_tp_event_match(event, &data, regs)) {
 			perf_swevent_event(event, count, &data, regs);
+
+			/*
+			 * Here use the same on-stack perf_sample_data,
+			 * some members in data are event-specific and
+			 * need to be re-computed for different sweveents.
+			 * Re-initialize data->sample_flags safely to avoid
+			 * the problem that next event skips preparing data
+			 * because data->sample_flags is set.
+			 */
+			perf_sample_data_init(&data, 0, 0);
+			perf_sample_save_raw_data(&data, &raw);
+		}
 	}
 
 	/*
diff --git a/kernel/locking/rwsem.c b/kernel/locking/rwsem.c
index acb5a50309a1..9eabd585ce7a 100644
--- a/kernel/locking/rwsem.c
+++ b/kernel/locking/rwsem.c
@@ -1240,7 +1240,7 @@ static struct rw_semaphore *rwsem_downgrade_wake(struct rw_semaphore *sem)
 /*
  * lock for reading
  */
-static inline int __down_read_common(struct rw_semaphore *sem, int state)
+static __always_inline int __down_read_common(struct rw_semaphore *sem, int state)
 {
 	int ret = 0;
 	long count;
@@ -1258,17 +1258,17 @@ out:
 	return ret;
 }
 
-static inline void __down_read(struct rw_semaphore *sem)
+static __always_inline void __down_read(struct rw_semaphore *sem)
 {
 	__down_read_common(sem, TASK_UNINTERRUPTIBLE);
 }
 
-static inline int __down_read_interruptible(struct rw_semaphore *sem)
+static __always_inline int __down_read_interruptible(struct rw_semaphore *sem)
 {
 	return __down_read_common(sem, TASK_INTERRUPTIBLE);
 }
 
-static inline int __down_read_killable(struct rw_semaphore *sem)
+static __always_inline int __down_read_killable(struct rw_semaphore *sem)
 {
 	return __down_read_common(sem, TASK_KILLABLE);
 }
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 944c3ae39861..a68d1276bab0 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -11492,7 +11492,7 @@ void call_trace_sched_update_nr_running(struct rq *rq, int count)
 
 #ifdef CONFIG_SCHED_MM_CID
 
-/**
+/*
  * @cid_lock: Guarantee forward-progress of cid allocation.
  *
  * Concurrency ID allocation within a bitmap is mostly lock-free. The cid_lock
@@ -11501,7 +11501,7 @@ void call_trace_sched_update_nr_running(struct rq *rq, int count)
  */
 DEFINE_RAW_SPINLOCK(cid_lock);
 
-/**
+/*
  * @use_cid_lock: Select cid allocation behavior: lock-free vs spinlock.
  *
  * When @use_cid_lock is 0, the cid allocation is lock-free. When contention is
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 93bf2b4e47e5..771d1e040303 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -35,14 +35,15 @@ static __cacheline_aligned_in_smp DEFINE_RAW_SPINLOCK(tick_broadcast_lock);
 #ifdef CONFIG_TICK_ONESHOT
 static DEFINE_PER_CPU(struct clock_event_device *, tick_oneshot_wakeup_device);
 
-static void tick_broadcast_setup_oneshot(struct clock_event_device *bc);
+static void tick_broadcast_setup_oneshot(struct clock_event_device *bc, bool from_periodic);
 static void tick_broadcast_clear_oneshot(int cpu);
 static void tick_resume_broadcast_oneshot(struct clock_event_device *bc);
 # ifdef CONFIG_HOTPLUG_CPU
 static void tick_broadcast_oneshot_offline(unsigned int cpu);
 # endif
 #else
-static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { BUG(); }
+static inline void
+tick_broadcast_setup_oneshot(struct clock_event_device *bc, bool from_periodic) { BUG(); }
 static inline void tick_broadcast_clear_oneshot(int cpu) { }
 static inline void tick_resume_broadcast_oneshot(struct clock_event_device *bc) { }
 # ifdef CONFIG_HOTPLUG_CPU
@@ -264,7 +265,7 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 		if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
 			tick_broadcast_start_periodic(bc);
 		else
-			tick_broadcast_setup_oneshot(bc);
+			tick_broadcast_setup_oneshot(bc, false);
 		ret = 1;
 	} else {
 		/*
@@ -500,7 +501,7 @@ void tick_broadcast_control(enum tick_broadcast_mode mode)
 			if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
 				tick_broadcast_start_periodic(bc);
 			else
-				tick_broadcast_setup_oneshot(bc);
+				tick_broadcast_setup_oneshot(bc, false);
 		}
 	}
 out:
@@ -1020,48 +1021,101 @@ static inline ktime_t tick_get_next_period(void)
 /**
  * tick_broadcast_setup_oneshot - setup the broadcast device
  */
-static void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
+static void tick_broadcast_setup_oneshot(struct clock_event_device *bc,
+					 bool from_periodic)
 {
 	int cpu = smp_processor_id();
+	ktime_t nexttick = 0;
 
 	if (!bc)
 		return;
 
-	/* Set it up only once ! */
-	if (bc->event_handler != tick_handle_oneshot_broadcast) {
-		int was_periodic = clockevent_state_periodic(bc);
-
-		bc->event_handler = tick_handle_oneshot_broadcast;
-
+	/*
+	 * When the broadcast device was switched to oneshot by the first
+	 * CPU handling the NOHZ change, the other CPUs will reach this
+	 * code via hrtimer_run_queues() -> tick_check_oneshot_change()
+	 * too. Set up the broadcast device only once!
+	 */
+	if (bc->event_handler == tick_handle_oneshot_broadcast) {
 		/*
-		 * We must be careful here. There might be other CPUs
-		 * waiting for periodic broadcast. We need to set the
-		 * oneshot_mask bits for those and program the
-		 * broadcast device to fire.
+		 * The CPU which switched from periodic to oneshot mode
+		 * set the broadcast oneshot bit for all other CPUs which
+		 * are in the general (periodic) broadcast mask to ensure
+		 * that CPUs which wait for the periodic broadcast are
+		 * woken up.
+		 *
+		 * Clear the bit for the local CPU as the set bit would
+		 * prevent the first tick_broadcast_enter() after this CPU
+		 * switched to oneshot state to program the broadcast
+		 * device.
+		 *
+		 * This code can also be reached via tick_broadcast_control(),
+		 * but this cannot avoid the tick_broadcast_clear_oneshot()
+		 * as that would break the periodic to oneshot transition of
+		 * secondary CPUs. But that's harmless as the below only
+		 * clears already cleared bits.
 		 */
+		tick_broadcast_clear_oneshot(cpu);
+		return;
+	}
+
+
+	bc->event_handler = tick_handle_oneshot_broadcast;
+	bc->next_event = KTIME_MAX;
+
+	/*
+	 * When the tick mode is switched from periodic to oneshot it must
+	 * be ensured that CPUs which are waiting for periodic broadcast
+	 * get their wake-up at the next tick.  This is achieved by ORing
+	 * tick_broadcast_mask into tick_broadcast_oneshot_mask.
+	 *
+	 * For other callers, e.g. broadcast device replacement,
+	 * tick_broadcast_oneshot_mask must not be touched as this would
+	 * set bits for CPUs which are already NOHZ, but not idle. Their
+	 * next tick_broadcast_enter() would observe the bit set and fail
+	 * to update the expiry time and the broadcast event device.
+	 */
+	if (from_periodic) {
 		cpumask_copy(tmpmask, tick_broadcast_mask);
+		/* Remove the local CPU as it is obviously not idle */
 		cpumask_clear_cpu(cpu, tmpmask);
-		cpumask_or(tick_broadcast_oneshot_mask,
-			   tick_broadcast_oneshot_mask, tmpmask);
+		cpumask_or(tick_broadcast_oneshot_mask, tick_broadcast_oneshot_mask, tmpmask);
 
-		if (was_periodic && !cpumask_empty(tmpmask)) {
-			ktime_t nextevt = tick_get_next_period();
+		/*
+		 * Ensure that the oneshot broadcast handler will wake the
+		 * CPUs which are still waiting for periodic broadcast.
+		 */
+		nexttick = tick_get_next_period();
+		tick_broadcast_init_next_event(tmpmask, nexttick);
 
-			clockevents_switch_state(bc, CLOCK_EVT_STATE_ONESHOT);
-			tick_broadcast_init_next_event(tmpmask, nextevt);
-			tick_broadcast_set_event(bc, cpu, nextevt);
-		} else
-			bc->next_event = KTIME_MAX;
-	} else {
 		/*
-		 * The first cpu which switches to oneshot mode sets
-		 * the bit for all other cpus which are in the general
-		 * (periodic) broadcast mask. So the bit is set and
-		 * would prevent the first broadcast enter after this
-		 * to program the bc device.
+		 * If the underlying broadcast clock event device is
+		 * already in oneshot state, then there is nothing to do.
+		 * The device was already armed for the next tick
+		 * in tick_handle_broadcast_periodic()
 		 */
-		tick_broadcast_clear_oneshot(cpu);
+		if (clockevent_state_oneshot(bc))
+			return;
 	}
+
+	/*
+	 * When switching from periodic to oneshot mode arm the broadcast
+	 * device for the next tick.
+	 *
+	 * If the broadcast device has been replaced in oneshot mode and
+	 * the oneshot broadcast mask is not empty, then arm it to expire
+	 * immediately in order to reevaluate the next expiring timer.
+	 * @nexttick is 0 and therefore in the past which will cause the
+	 * clockevent code to force an event.
+	 *
+	 * For both cases the programming can be avoided when the oneshot
+	 * broadcast mask is empty.
+	 *
+	 * tick_broadcast_set_event() implicitly switches the broadcast
+	 * device to oneshot state.
+	 */
+	if (!cpumask_empty(tick_broadcast_oneshot_mask))
+		tick_broadcast_set_event(bc, cpu, nexttick);
 }
 
 /*
@@ -1070,14 +1124,16 @@ static void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
 void tick_broadcast_switch_to_oneshot(void)
 {
 	struct clock_event_device *bc;
+	enum tick_device_mode oldmode;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
 
+	oldmode = tick_broadcast_device.mode;
 	tick_broadcast_device.mode = TICKDEV_MODE_ONESHOT;
 	bc = tick_broadcast_device.evtdev;
 	if (bc)
-		tick_broadcast_setup_oneshot(bc);
+		tick_broadcast_setup_oneshot(bc, oldmode == TICKDEV_MODE_PERIODIC);
 
 	raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
 }
diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c
index 9abb3905bc8e..18d36842faf5 100644
--- a/kernel/trace/fprobe.c
+++ b/kernel/trace/fprobe.c
@@ -17,36 +17,30 @@
 struct fprobe_rethook_node {
 	struct rethook_node node;
 	unsigned long entry_ip;
+	unsigned long entry_parent_ip;
 	char data[];
 };
 
-static void fprobe_handler(unsigned long ip, unsigned long parent_ip,
-			   struct ftrace_ops *ops, struct ftrace_regs *fregs)
+static inline void __fprobe_handler(unsigned long ip, unsigned long parent_ip,
+			struct ftrace_ops *ops, struct ftrace_regs *fregs)
 {
 	struct fprobe_rethook_node *fpr;
 	struct rethook_node *rh = NULL;
 	struct fprobe *fp;
 	void *entry_data = NULL;
-	int bit, ret;
+	int ret = 0;
 
 	fp = container_of(ops, struct fprobe, ops);
-	if (fprobe_disabled(fp))
-		return;
-
-	bit = ftrace_test_recursion_trylock(ip, parent_ip);
-	if (bit < 0) {
-		fp->nmissed++;
-		return;
-	}
 
 	if (fp->exit_handler) {
 		rh = rethook_try_get(fp->rethook);
 		if (!rh) {
 			fp->nmissed++;
-			goto out;
+			return;
 		}
 		fpr = container_of(rh, struct fprobe_rethook_node, node);
 		fpr->entry_ip = ip;
+		fpr->entry_parent_ip = parent_ip;
 		if (fp->entry_data_size)
 			entry_data = fpr->data;
 	}
@@ -61,23 +55,60 @@ static void fprobe_handler(unsigned long ip, unsigned long parent_ip,
 		else
 			rethook_hook(rh, ftrace_get_regs(fregs), true);
 	}
-out:
+}
+
+static void fprobe_handler(unsigned long ip, unsigned long parent_ip,
+		struct ftrace_ops *ops, struct ftrace_regs *fregs)
+{
+	struct fprobe *fp;
+	int bit;
+
+	fp = container_of(ops, struct fprobe, ops);
+	if (fprobe_disabled(fp))
+		return;
+
+	/* recursion detection has to go before any traceable function and
+	 * all functions before this point should be marked as notrace
+	 */
+	bit = ftrace_test_recursion_trylock(ip, parent_ip);
+	if (bit < 0) {
+		fp->nmissed++;
+		return;
+	}
+	__fprobe_handler(ip, parent_ip, ops, fregs);
 	ftrace_test_recursion_unlock(bit);
+
 }
 NOKPROBE_SYMBOL(fprobe_handler);
 
 static void fprobe_kprobe_handler(unsigned long ip, unsigned long parent_ip,
 				  struct ftrace_ops *ops, struct ftrace_regs *fregs)
 {
-	struct fprobe *fp = container_of(ops, struct fprobe, ops);
+	struct fprobe *fp;
+	int bit;
+
+	fp = container_of(ops, struct fprobe, ops);
+	if (fprobe_disabled(fp))
+		return;
+
+	/* recursion detection has to go before any traceable function and
+	 * all functions called before this point should be marked as notrace
+	 */
+	bit = ftrace_test_recursion_trylock(ip, parent_ip);
+	if (bit < 0) {
+		fp->nmissed++;
+		return;
+	}
 
 	if (unlikely(kprobe_running())) {
 		fp->nmissed++;
 		return;
 	}
+
 	kprobe_busy_begin();
-	fprobe_handler(ip, parent_ip, ops, fregs);
+	__fprobe_handler(ip, parent_ip, ops, fregs);
 	kprobe_busy_end();
+	ftrace_test_recursion_unlock(bit);
 }
 
 static void fprobe_exit_handler(struct rethook_node *rh, void *data,
@@ -85,14 +116,26 @@ static void fprobe_exit_handler(struct rethook_node *rh, void *data,
 {
 	struct fprobe *fp = (struct fprobe *)data;
 	struct fprobe_rethook_node *fpr;
+	int bit;
 
 	if (!fp || fprobe_disabled(fp))
 		return;
 
 	fpr = container_of(rh, struct fprobe_rethook_node, node);
 
+	/*
+	 * we need to assure no calls to traceable functions in-between the
+	 * end of fprobe_handler and the beginning of fprobe_exit_handler.
+	 */
+	bit = ftrace_test_recursion_trylock(fpr->entry_ip, fpr->entry_parent_ip);
+	if (bit < 0) {
+		fp->nmissed++;
+		return;
+	}
+
 	fp->exit_handler(fp, fpr->entry_ip, regs,
 			 fp->entry_data_size ? (void *)fpr->data : NULL);
+	ftrace_test_recursion_unlock(bit);
 }
 NOKPROBE_SYMBOL(fprobe_exit_handler);
 
diff --git a/kernel/trace/rethook.c b/kernel/trace/rethook.c
index 32c3dfdb4d6a..60f6cb2b486b 100644
--- a/kernel/trace/rethook.c
+++ b/kernel/trace/rethook.c
@@ -288,7 +288,7 @@ unsigned long rethook_trampoline_handler(struct pt_regs *regs,
 	 * These loops must be protected from rethook_free_rcu() because those
 	 * are accessing 'rhn->rethook'.
 	 */
-	preempt_disable();
+	preempt_disable_notrace();
 
 	/*
 	 * Run the handler on the shadow stack. Do not unlink the list here because
@@ -321,7 +321,7 @@ unsigned long rethook_trampoline_handler(struct pt_regs *regs,
 		first = first->next;
 		rethook_recycle(rhn);
 	}
-	preempt_enable();
+	preempt_enable_notrace();
 
 	return correct_ret_addr;
 }