Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

1 files changed, 1159 insertions, 0 deletions

diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
new file mode 100644
index 000000000..96626f806
--- /dev/null
+++ b/drivers/rtc/interface.c
@@ -0,0 +1,1159 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * RTC subsystem, interface functions
+ *
+ * Copyright (C) 2005 Tower Technologies
+ * Author: Alessandro Zummo <a.zummo@towertech.it>
+ *
+ * based on arch/arm/common/rtctime.c
+ */
+
+#include <linux/rtc.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/log2.h>
+#include <linux/workqueue.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/rtc.h>
+
+static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer);
+static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer);
+
+static void rtc_add_offset(struct rtc_device *rtc, struct rtc_time *tm)
+{
+	time64_t secs;
+
+	if (!rtc->offset_secs)
+		return;
+
+	secs = rtc_tm_to_time64(tm);
+
+	/*
+	 * Since the reading time values from RTC device are always in the RTC
+	 * original valid range, but we need to skip the overlapped region
+	 * between expanded range and original range, which is no need to add
+	 * the offset.
+	 */
+	if ((rtc->start_secs > rtc->range_min && secs >= rtc->start_secs) ||
+	    (rtc->start_secs < rtc->range_min &&
+	     secs <= (rtc->start_secs + rtc->range_max - rtc->range_min)))
+		return;
+
+	rtc_time64_to_tm(secs + rtc->offset_secs, tm);
+}
+
+static void rtc_subtract_offset(struct rtc_device *rtc, struct rtc_time *tm)
+{
+	time64_t secs;
+
+	if (!rtc->offset_secs)
+		return;
+
+	secs = rtc_tm_to_time64(tm);
+
+	/*
+	 * If the setting time values are in the valid range of RTC hardware
+	 * device, then no need to subtract the offset when setting time to RTC
+	 * device. Otherwise we need to subtract the offset to make the time
+	 * values are valid for RTC hardware device.
+	 */
+	if (secs >= rtc->range_min && secs <= rtc->range_max)
+		return;
+
+	rtc_time64_to_tm(secs - rtc->offset_secs, tm);
+}
+
+static int rtc_valid_range(struct rtc_device *rtc, struct rtc_time *tm)
+{
+	if (rtc->range_min != rtc->range_max) {
+		time64_t time = rtc_tm_to_time64(tm);
+		time64_t range_min = rtc->set_start_time ? rtc->start_secs :
+			rtc->range_min;
+		timeu64_t range_max = rtc->set_start_time ?
+			(rtc->start_secs + rtc->range_max - rtc->range_min) :
+			rtc->range_max;
+
+		if (time < range_min || time > range_max)
+			return -ERANGE;
+	}
+
+	return 0;
+}
+
+static int __rtc_read_time(struct rtc_device *rtc, struct rtc_time *tm)
+{
+	int err;
+
+	if (!rtc->ops) {
+		err = -ENODEV;
+	} else if (!rtc->ops->read_time) {
+		err = -EINVAL;
+	} else {
+		memset(tm, 0, sizeof(struct rtc_time));
+		err = rtc->ops->read_time(rtc->dev.parent, tm);
+		if (err < 0) {
+			dev_dbg(&rtc->dev, "read_time: fail to read: %d\n",
+				err);
+			return err;
+		}
+
+		rtc_add_offset(rtc, tm);
+
+		err = rtc_valid_tm(tm);
+		if (err < 0)
+			dev_dbg(&rtc->dev, "read_time: rtc_time isn't valid\n");
+	}
+	return err;
+}
+
+int rtc_read_time(struct rtc_device *rtc, struct rtc_time *tm)
+{
+	int err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	err = __rtc_read_time(rtc, tm);
+	mutex_unlock(&rtc->ops_lock);
+
+	trace_rtc_read_time(rtc_tm_to_time64(tm), err);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_read_time);
+
+int rtc_set_time(struct rtc_device *rtc, struct rtc_time *tm)
+{
+	int err, uie;
+
+	err = rtc_valid_tm(tm);
+	if (err != 0)
+		return err;
+
+	err = rtc_valid_range(rtc, tm);
+	if (err)
+		return err;
+
+	rtc_subtract_offset(rtc, tm);
+
+#ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
+	uie = rtc->uie_rtctimer.enabled || rtc->uie_irq_active;
+#else
+	uie = rtc->uie_rtctimer.enabled;
+#endif
+	if (uie) {
+		err = rtc_update_irq_enable(rtc, 0);
+		if (err)
+			return err;
+	}
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	if (!rtc->ops)
+		err = -ENODEV;
+	else if (rtc->ops->set_time)
+		err = rtc->ops->set_time(rtc->dev.parent, tm);
+	else
+		err = -EINVAL;
+
+	pm_stay_awake(rtc->dev.parent);
+	mutex_unlock(&rtc->ops_lock);
+	/* A timer might have just expired */
+	schedule_work(&rtc->irqwork);
+
+	if (uie) {
+		err = rtc_update_irq_enable(rtc, 1);
+		if (err)
+			return err;
+	}
+
+	trace_rtc_set_time(rtc_tm_to_time64(tm), err);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_set_time);
+
+static int rtc_read_alarm_internal(struct rtc_device *rtc,
+				   struct rtc_wkalrm *alarm)
+{
+	int err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	if (!rtc->ops) {
+		err = -ENODEV;
+	} else if (!test_bit(RTC_FEATURE_ALARM, rtc->features) || !rtc->ops->read_alarm) {
+		err = -EINVAL;
+	} else {
+		alarm->enabled = 0;
+		alarm->pending = 0;
+		alarm->time.tm_sec = -1;
+		alarm->time.tm_min = -1;
+		alarm->time.tm_hour = -1;
+		alarm->time.tm_mday = -1;
+		alarm->time.tm_mon = -1;
+		alarm->time.tm_year = -1;
+		alarm->time.tm_wday = -1;
+		alarm->time.tm_yday = -1;
+		alarm->time.tm_isdst = -1;
+		err = rtc->ops->read_alarm(rtc->dev.parent, alarm);
+	}
+
+	mutex_unlock(&rtc->ops_lock);
+
+	trace_rtc_read_alarm(err?0:rtc_tm_to_time64(&alarm->time), err);
+	return err;
+}
+
+int __rtc_read_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	int err;
+	struct rtc_time before, now;
+	int first_time = 1;
+	time64_t t_now, t_alm;
+	enum { none, day, month, year } missing = none;
+	unsigned int days;
+
+	/* The lower level RTC driver may return -1 in some fields,
+	 * creating invalid alarm->time values, for reasons like:
+	 *
+	 *   - The hardware may not be capable of filling them in;
+	 *     many alarms match only on time-of-day fields, not
+	 *     day/month/year calendar data.
+	 *
+	 *   - Some hardware uses illegal values as "wildcard" match
+	 *     values, which non-Linux firmware (like a BIOS) may try
+	 *     to set up as e.g. "alarm 15 minutes after each hour".
+	 *     Linux uses only oneshot alarms.
+	 *
+	 * When we see that here, we deal with it by using values from
+	 * a current RTC timestamp for any missing (-1) values.  The
+	 * RTC driver prevents "periodic alarm" modes.
+	 *
+	 * But this can be racey, because some fields of the RTC timestamp
+	 * may have wrapped in the interval since we read the RTC alarm,
+	 * which would lead to us inserting inconsistent values in place
+	 * of the -1 fields.
+	 *
+	 * Reading the alarm and timestamp in the reverse sequence
+	 * would have the same race condition, and not solve the issue.
+	 *
+	 * So, we must first read the RTC timestamp,
+	 * then read the RTC alarm value,
+	 * and then read a second RTC timestamp.
+	 *
+	 * If any fields of the second timestamp have changed
+	 * when compared with the first timestamp, then we know
+	 * our timestamp may be inconsistent with that used by
+	 * the low-level rtc_read_alarm_internal() function.
+	 *
+	 * So, when the two timestamps disagree, we just loop and do
+	 * the process again to get a fully consistent set of values.
+	 *
+	 * This could all instead be done in the lower level driver,
+	 * but since more than one lower level RTC implementation needs it,
+	 * then it's probably best to do it here instead of there..
+	 */
+
+	/* Get the "before" timestamp */
+	err = rtc_read_time(rtc, &before);
+	if (err < 0)
+		return err;
+	do {
+		if (!first_time)
+			memcpy(&before, &now, sizeof(struct rtc_time));
+		first_time = 0;
+
+		/* get the RTC alarm values, which may be incomplete */
+		err = rtc_read_alarm_internal(rtc, alarm);
+		if (err)
+			return err;
+
+		/* full-function RTCs won't have such missing fields */
+		err = rtc_valid_tm(&alarm->time);
+		if (!err)
+			goto done;
+
+		/* get the "after" timestamp, to detect wrapped fields */
+		err = rtc_read_time(rtc, &now);
+		if (err < 0)
+			return err;
+
+		/* note that tm_sec is a "don't care" value here: */
+	} while (before.tm_min  != now.tm_min ||
+		 before.tm_hour != now.tm_hour ||
+		 before.tm_mon  != now.tm_mon ||
+		 before.tm_year != now.tm_year);
+
+	/* Fill in the missing alarm fields using the timestamp; we
+	 * know there's at least one since alarm->time is invalid.
+	 */
+	if (alarm->time.tm_sec == -1)
+		alarm->time.tm_sec = now.tm_sec;
+	if (alarm->time.tm_min == -1)
+		alarm->time.tm_min = now.tm_min;
+	if (alarm->time.tm_hour == -1)
+		alarm->time.tm_hour = now.tm_hour;
+
+	/* For simplicity, only support date rollover for now */
+	if (alarm->time.tm_mday < 1 || alarm->time.tm_mday > 31) {
+		alarm->time.tm_mday = now.tm_mday;
+		missing = day;
+	}
+	if ((unsigned int)alarm->time.tm_mon >= 12) {
+		alarm->time.tm_mon = now.tm_mon;
+		if (missing == none)
+			missing = month;
+	}
+	if (alarm->time.tm_year == -1) {
+		alarm->time.tm_year = now.tm_year;
+		if (missing == none)
+			missing = year;
+	}
+
+	/* Can't proceed if alarm is still invalid after replacing
+	 * missing fields.
+	 */
+	err = rtc_valid_tm(&alarm->time);
+	if (err)
+		goto done;
+
+	/* with luck, no rollover is needed */
+	t_now = rtc_tm_to_time64(&now);
+	t_alm = rtc_tm_to_time64(&alarm->time);
+	if (t_now < t_alm)
+		goto done;
+
+	switch (missing) {
+	/* 24 hour rollover ... if it's now 10am Monday, an alarm that
+	 * that will trigger at 5am will do so at 5am Tuesday, which
+	 * could also be in the next month or year.  This is a common
+	 * case, especially for PCs.
+	 */
+	case day:
+		dev_dbg(&rtc->dev, "alarm rollover: %s\n", "day");
+		t_alm += 24 * 60 * 60;
+		rtc_time64_to_tm(t_alm, &alarm->time);
+		break;
+
+	/* Month rollover ... if it's the 31th, an alarm on the 3rd will
+	 * be next month.  An alarm matching on the 30th, 29th, or 28th
+	 * may end up in the month after that!  Many newer PCs support
+	 * this type of alarm.
+	 */
+	case month:
+		dev_dbg(&rtc->dev, "alarm rollover: %s\n", "month");
+		do {
+			if (alarm->time.tm_mon < 11) {
+				alarm->time.tm_mon++;
+			} else {
+				alarm->time.tm_mon = 0;
+				alarm->time.tm_year++;
+			}
+			days = rtc_month_days(alarm->time.tm_mon,
+					      alarm->time.tm_year);
+		} while (days < alarm->time.tm_mday);
+		break;
+
+	/* Year rollover ... easy except for leap years! */
+	case year:
+		dev_dbg(&rtc->dev, "alarm rollover: %s\n", "year");
+		do {
+			alarm->time.tm_year++;
+		} while (!is_leap_year(alarm->time.tm_year + 1900) &&
+			 rtc_valid_tm(&alarm->time) != 0);
+		break;
+
+	default:
+		dev_warn(&rtc->dev, "alarm rollover not handled\n");
+	}
+
+	err = rtc_valid_tm(&alarm->time);
+
+done:
+	if (err && alarm->enabled)
+		dev_warn(&rtc->dev, "invalid alarm value: %ptR\n",
+			 &alarm->time);
+	else
+		rtc_add_offset(rtc, &alarm->time);
+
+	return err;
+}
+
+/**
+ * rtc_read_next_alarm - read the next expiring alarm
+ * @rtc: RTC device
+ * @alarm: storage for the alarm information
+ *
+ * Read the next expiring alarm from the RTC timerqueue. This returns
+ * the alarm that will actually fire next, which may be different from
+ * rtc_read_alarm() if multiple timers are queued (e.g., alarmtimer
+ * and wakealarm sysfs both active).
+ *
+ * Returns: 0 on success, -ENOENT if no alarm is pending, or other error.
+ */
+int rtc_read_next_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	struct timerqueue_node *next;
+	int err;
+
+	if (!rtc || !alarm)
+		return -EINVAL;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	next = timerqueue_getnext(&rtc->timerqueue);
+	if (!next) {
+		err = -ENOENT;
+		goto unlock;
+	}
+
+	memset(alarm, 0, sizeof(struct rtc_wkalrm));
+	alarm->time = rtc_ktime_to_tm(next->expires);
+	alarm->enabled = 1;
+
+unlock:
+	mutex_unlock(&rtc->ops_lock);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_read_next_alarm);
+
+int rtc_read_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	int err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+	if (!rtc->ops) {
+		err = -ENODEV;
+	} else if (!test_bit(RTC_FEATURE_ALARM, rtc->features)) {
+		err = -EINVAL;
+	} else {
+		memset(alarm, 0, sizeof(struct rtc_wkalrm));
+		alarm->enabled = rtc->aie_timer.enabled;
+		alarm->time = rtc_ktime_to_tm(rtc->aie_timer.node.expires);
+	}
+	mutex_unlock(&rtc->ops_lock);
+
+	trace_rtc_read_alarm(rtc_tm_to_time64(&alarm->time), err);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_read_alarm);
+
+static int __rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	struct rtc_time tm;
+	time64_t now, scheduled;
+	int err;
+
+	err = rtc_valid_tm(&alarm->time);
+	if (err)
+		return err;
+
+	scheduled = rtc_tm_to_time64(&alarm->time);
+
+	/* Make sure we're not setting alarms in the past */
+	err = __rtc_read_time(rtc, &tm);
+	if (err)
+		return err;
+	now = rtc_tm_to_time64(&tm);
+
+	if (scheduled <= now)
+		return -ETIME;
+	/*
+	 * XXX - We just checked to make sure the alarm time is not
+	 * in the past, but there is still a race window where if
+	 * the is alarm set for the next second and the second ticks
+	 * over right here, before we set the alarm.
+	 */
+
+	rtc_subtract_offset(rtc, &alarm->time);
+
+	if (!rtc->ops)
+		err = -ENODEV;
+	else if (!test_bit(RTC_FEATURE_ALARM, rtc->features))
+		err = -EINVAL;
+	else
+		err = rtc->ops->set_alarm(rtc->dev.parent, alarm);
+
+	/*
+	 * Check for potential race described above. If the waiting for next
+	 * second, and the second just ticked since the check above, either
+	 *
+	 * 1) It ticked after the alarm was set, and an alarm irq should be
+	 *    generated.
+	 *
+	 * 2) It ticked before the alarm was set, and alarm irq most likely will
+	 * not be generated.
+	 *
+	 * While we cannot easily check for which of these two scenarios we
+	 * are in, we can return -ETIME to signal that the timer has already
+	 * expired, which is true in both cases.
+	 */
+	if (!err && (scheduled - now) <= 1) {
+		err = __rtc_read_time(rtc, &tm);
+		if (err)
+			return err;
+		now = rtc_tm_to_time64(&tm);
+		if (scheduled <= now)
+			return -ETIME;
+	}
+
+	trace_rtc_set_alarm(rtc_tm_to_time64(&alarm->time), err);
+	return err;
+}
+
+int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	ktime_t alarm_time;
+	int err;
+
+	if (!rtc->ops)
+		return -ENODEV;
+	else if (!test_bit(RTC_FEATURE_ALARM, rtc->features))
+		return -EINVAL;
+
+	err = rtc_valid_tm(&alarm->time);
+	if (err != 0)
+		return err;
+
+	err = rtc_valid_range(rtc, &alarm->time);
+	if (err)
+		return err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+	if (rtc->aie_timer.enabled)
+		rtc_timer_remove(rtc, &rtc->aie_timer);
+
+	alarm_time = rtc_tm_to_ktime(alarm->time);
+	/*
+	 * Round down so we never miss a deadline, checking for past deadline is
+	 * done in __rtc_set_alarm
+	 */
+	if (test_bit(RTC_FEATURE_ALARM_RES_MINUTE, rtc->features))
+		alarm_time = ktime_sub_ns(alarm_time, (u64)alarm->time.tm_sec * NSEC_PER_SEC);
+
+	rtc->aie_timer.node.expires = alarm_time;
+	rtc->aie_timer.period = 0;
+	if (alarm->enabled)
+		err = rtc_timer_enqueue(rtc, &rtc->aie_timer);
+
+	mutex_unlock(&rtc->ops_lock);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_set_alarm);
+
+/* Called once per device from rtc_device_register */
+int rtc_initialize_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	int err;
+	struct rtc_time now;
+
+	err = rtc_valid_tm(&alarm->time);
+	if (err != 0)
+		return err;
+
+	err = rtc_read_time(rtc, &now);
+	if (err)
+		return err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	rtc->aie_timer.node.expires = rtc_tm_to_ktime(alarm->time);
+	rtc->aie_timer.period = 0;
+
+	/* Alarm has to be enabled & in the future for us to enqueue it */
+	if (alarm->enabled && (rtc_tm_to_ktime(now) <
+			 rtc->aie_timer.node.expires)) {
+		rtc->aie_timer.enabled = 1;
+		timerqueue_add(&rtc->timerqueue, &rtc->aie_timer.node);
+		trace_rtc_timer_enqueue(&rtc->aie_timer);
+	}
+	mutex_unlock(&rtc->ops_lock);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_initialize_alarm);
+
+int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled)
+{
+	int err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	if (rtc->aie_timer.enabled != enabled) {
+		if (enabled)
+			err = rtc_timer_enqueue(rtc, &rtc->aie_timer);
+		else
+			rtc_timer_remove(rtc, &rtc->aie_timer);
+	}
+
+	if (err)
+		/* nothing */;
+	else if (!rtc->ops)
+		err = -ENODEV;
+	else if (!test_bit(RTC_FEATURE_ALARM, rtc->features) || !rtc->ops->alarm_irq_enable)
+		err = -EINVAL;
+	else
+		err = rtc->ops->alarm_irq_enable(rtc->dev.parent, enabled);
+
+	mutex_unlock(&rtc->ops_lock);
+
+	trace_rtc_alarm_irq_enable(enabled, err);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_alarm_irq_enable);
+
+int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled)
+{
+	int err;
+
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+#ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
+	if (enabled == 0 && rtc->uie_irq_active) {
+		mutex_unlock(&rtc->ops_lock);
+		return rtc_dev_update_irq_enable_emul(rtc, 0);
+	}
+#endif
+	/* make sure we're changing state */
+	if (rtc->uie_rtctimer.enabled == enabled)
+		goto out;
+
+	if (!test_bit(RTC_FEATURE_UPDATE_INTERRUPT, rtc->features) ||
+	    !test_bit(RTC_FEATURE_ALARM, rtc->features)) {
+		mutex_unlock(&rtc->ops_lock);
+#ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
+		return rtc_dev_update_irq_enable_emul(rtc, enabled);
+#else
+		return -EINVAL;
+#endif
+	}
+
+	if (enabled) {
+		struct rtc_time tm;
+		ktime_t now, onesec;
+
+		err = __rtc_read_time(rtc, &tm);
+		if (err)
+			goto out;
+		onesec = ktime_set(1, 0);
+		now = rtc_tm_to_ktime(tm);
+		rtc->uie_rtctimer.node.expires = ktime_add(now, onesec);
+		rtc->uie_rtctimer.period = ktime_set(1, 0);
+		err = rtc_timer_enqueue(rtc, &rtc->uie_rtctimer);
+		if (!err && rtc->ops && rtc->ops->alarm_irq_enable)
+			err = rtc->ops->alarm_irq_enable(rtc->dev.parent, 1);
+		if (err)
+			goto out;
+	} else {
+		rtc_timer_remove(rtc, &rtc->uie_rtctimer);
+	}
+
+out:
+	mutex_unlock(&rtc->ops_lock);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_update_irq_enable);
+
+/**
+ * rtc_handle_legacy_irq - AIE, UIE and PIE event hook
+ * @rtc: pointer to the rtc device
+ * @num: number of occurrence of the event
+ * @mode: type of the event, RTC_AF, RTC_UF or RTC_PF
+ *
+ * This function is called when an AIE, UIE or PIE mode interrupt
+ * has occurred (or been emulated).
+ *
+ */
+void rtc_handle_legacy_irq(struct rtc_device *rtc, int num, int mode)
+{
+	unsigned long flags;
+
+	/* mark one irq of the appropriate mode */
+	spin_lock_irqsave(&rtc->irq_lock, flags);
+	rtc->irq_data = (rtc->irq_data + (num << 8)) | (RTC_IRQF | mode);
+	spin_unlock_irqrestore(&rtc->irq_lock, flags);
+
+	wake_up_interruptible(&rtc->irq_queue);
+	kill_fasync(&rtc->async_queue, SIGIO, POLL_IN);
+}
+
+/**
+ * rtc_aie_update_irq - AIE mode rtctimer hook
+ * @rtc: pointer to the rtc_device
+ *
+ * This functions is called when the aie_timer expires.
+ */
+void rtc_aie_update_irq(struct rtc_device *rtc)
+{
+	rtc_handle_legacy_irq(rtc, 1, RTC_AF);
+}
+
+/**
+ * rtc_uie_update_irq - UIE mode rtctimer hook
+ * @rtc: pointer to the rtc_device
+ *
+ * This functions is called when the uie_timer expires.
+ */
+void rtc_uie_update_irq(struct rtc_device *rtc)
+{
+	rtc_handle_legacy_irq(rtc, 1,  RTC_UF);
+}
+
+/**
+ * rtc_pie_update_irq - PIE mode hrtimer hook
+ * @timer: pointer to the pie mode hrtimer
+ *
+ * This function is used to emulate PIE mode interrupts
+ * using an hrtimer. This function is called when the periodic
+ * hrtimer expires.
+ */
+enum hrtimer_restart rtc_pie_update_irq(struct hrtimer *timer)
+{
+	struct rtc_device *rtc;
+	ktime_t period;
+	u64 count;
+
+	rtc = container_of(timer, struct rtc_device, pie_timer);
+
+	period = NSEC_PER_SEC / rtc->irq_freq;
+	count = hrtimer_forward_now(timer, period);
+
+	rtc_handle_legacy_irq(rtc, count, RTC_PF);
+
+	return HRTIMER_RESTART;
+}
+
+/**
+ * rtc_update_irq - Triggered when a RTC interrupt occurs.
+ * @rtc: the rtc device
+ * @num: how many irqs are being reported (usually one)
+ * @events: mask of RTC_IRQF with one or more of RTC_PF, RTC_AF, RTC_UF
+ * Context: any
+ */
+void rtc_update_irq(struct rtc_device *rtc,
+		    unsigned long num, unsigned long events)
+{
+	if (IS_ERR_OR_NULL(rtc))
+		return;
+
+	pm_stay_awake(rtc->dev.parent);
+	schedule_work(&rtc->irqwork);
+}
+EXPORT_SYMBOL_GPL(rtc_update_irq);
+
+struct rtc_device *rtc_class_open(const char *name)
+{
+	struct device *dev;
+	struct rtc_device *rtc = NULL;
+
+	dev = class_find_device_by_name(&rtc_class, name);
+	if (dev)
+		rtc = to_rtc_device(dev);
+
+	if (rtc) {
+		if (!try_module_get(rtc->owner)) {
+			put_device(dev);
+			rtc = NULL;
+		}
+	}
+
+	return rtc;
+}
+EXPORT_SYMBOL_GPL(rtc_class_open);
+
+void rtc_class_close(struct rtc_device *rtc)
+{
+	module_put(rtc->owner);
+	put_device(&rtc->dev);
+}
+EXPORT_SYMBOL_GPL(rtc_class_close);
+
+static int rtc_update_hrtimer(struct rtc_device *rtc, int enabled)
+{
+	/*
+	 * We always cancel the timer here first, because otherwise
+	 * we could run into BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
+	 * when we manage to start the timer before the callback
+	 * returns HRTIMER_RESTART.
+	 *
+	 * We cannot use hrtimer_cancel() here as a running callback
+	 * could be blocked on rtc->irq_task_lock and hrtimer_cancel()
+	 * would spin forever.
+	 */
+	if (hrtimer_try_to_cancel(&rtc->pie_timer) < 0)
+		return -1;
+
+	if (enabled) {
+		ktime_t period = NSEC_PER_SEC / rtc->irq_freq;
+
+		hrtimer_start(&rtc->pie_timer, period, HRTIMER_MODE_REL);
+	}
+	return 0;
+}
+
+/**
+ * rtc_irq_set_state - enable/disable 2^N Hz periodic IRQs
+ * @rtc: the rtc device
+ * @enabled: true to enable periodic IRQs
+ * Context: any
+ *
+ * Note that rtc_irq_set_freq() should previously have been used to
+ * specify the desired frequency of periodic IRQ.
+ */
+int rtc_irq_set_state(struct rtc_device *rtc, int enabled)
+{
+	int err = 0;
+
+	while (rtc_update_hrtimer(rtc, enabled) < 0)
+		cpu_relax();
+
+	rtc->pie_enabled = enabled;
+
+	trace_rtc_irq_set_state(enabled, err);
+	return err;
+}
+
+/**
+ * rtc_irq_set_freq - set 2^N Hz periodic IRQ frequency for IRQ
+ * @rtc: the rtc device
+ * @freq: positive frequency
+ * Context: any
+ *
+ * Note that rtc_irq_set_state() is used to enable or disable the
+ * periodic IRQs.
+ */
+int rtc_irq_set_freq(struct rtc_device *rtc, int freq)
+{
+	int err = 0;
+
+	if (freq <= 0 || freq > RTC_MAX_FREQ)
+		return -EINVAL;
+
+	rtc->irq_freq = freq;
+	while (rtc->pie_enabled && rtc_update_hrtimer(rtc, 1) < 0)
+		cpu_relax();
+
+	trace_rtc_irq_set_freq(freq, err);
+	return err;
+}
+
+/**
+ * rtc_timer_enqueue - Adds a rtc_timer to the rtc_device timerqueue
+ * @rtc: rtc device
+ * @timer: timer being added.
+ *
+ * Enqueues a timer onto the rtc devices timerqueue and sets
+ * the next alarm event appropriately.
+ *
+ * Sets the enabled bit on the added timer.
+ *
+ * Must hold ops_lock for proper serialization of timerqueue
+ */
+static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer)
+{
+	struct timerqueue_node *next = timerqueue_getnext(&rtc->timerqueue);
+	struct rtc_time tm;
+	ktime_t now;
+	int err;
+
+	err = __rtc_read_time(rtc, &tm);
+	if (err)
+		return err;
+
+	timer->enabled = 1;
+	now = rtc_tm_to_ktime(tm);
+
+	/* Skip over expired timers */
+	while (next) {
+		if (next->expires >= now)
+			break;
+		next = timerqueue_iterate_next(next);
+	}
+
+	timerqueue_add(&rtc->timerqueue, &timer->node);
+	trace_rtc_timer_enqueue(timer);
+	if (!next || ktime_before(timer->node.expires, next->expires)) {
+		struct rtc_wkalrm alarm;
+
+		alarm.time = rtc_ktime_to_tm(timer->node.expires);
+		alarm.enabled = 1;
+		err = __rtc_set_alarm(rtc, &alarm);
+		if (err == -ETIME) {
+			pm_stay_awake(rtc->dev.parent);
+			schedule_work(&rtc->irqwork);
+		} else if (err) {
+			timerqueue_del(&rtc->timerqueue, &timer->node);
+			trace_rtc_timer_dequeue(timer);
+			timer->enabled = 0;
+			return err;
+		}
+	}
+	return 0;
+}
+
+static void rtc_alarm_disable(struct rtc_device *rtc)
+{
+	if (!rtc->ops || !test_bit(RTC_FEATURE_ALARM, rtc->features) || !rtc->ops->alarm_irq_enable)
+		return;
+
+	rtc->ops->alarm_irq_enable(rtc->dev.parent, false);
+	trace_rtc_alarm_irq_enable(0, 0);
+}
+
+/**
+ * rtc_timer_remove - Removes a rtc_timer from the rtc_device timerqueue
+ * @rtc: rtc device
+ * @timer: timer being removed.
+ *
+ * Removes a timer onto the rtc devices timerqueue and sets
+ * the next alarm event appropriately.
+ *
+ * Clears the enabled bit on the removed timer.
+ *
+ * Must hold ops_lock for proper serialization of timerqueue
+ */
+static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer)
+{
+	struct timerqueue_node *next = timerqueue_getnext(&rtc->timerqueue);
+
+	timerqueue_del(&rtc->timerqueue, &timer->node);
+	trace_rtc_timer_dequeue(timer);
+	timer->enabled = 0;
+	if (next == &timer->node) {
+		struct rtc_wkalrm alarm;
+		int err;
+
+		next = timerqueue_getnext(&rtc->timerqueue);
+		if (!next) {
+			rtc_alarm_disable(rtc);
+			return;
+		}
+		alarm.time = rtc_ktime_to_tm(next->expires);
+		alarm.enabled = 1;
+		err = __rtc_set_alarm(rtc, &alarm);
+		if (err == -ETIME) {
+			pm_stay_awake(rtc->dev.parent);
+			schedule_work(&rtc->irqwork);
+		}
+	}
+}
+
+/**
+ * rtc_timer_do_work - Expires rtc timers
+ * @work: work item
+ *
+ * Expires rtc timers. Reprograms next alarm event if needed.
+ * Called via worktask.
+ *
+ * Serializes access to timerqueue via ops_lock mutex
+ */
+void rtc_timer_do_work(struct work_struct *work)
+{
+	struct rtc_timer *timer;
+	struct timerqueue_node *next;
+	ktime_t now;
+	struct rtc_time tm;
+	int err;
+
+	struct rtc_device *rtc =
+		container_of(work, struct rtc_device, irqwork);
+
+	mutex_lock(&rtc->ops_lock);
+again:
+	err = __rtc_read_time(rtc, &tm);
+	if (err) {
+		mutex_unlock(&rtc->ops_lock);
+		return;
+	}
+	now = rtc_tm_to_ktime(tm);
+	while ((next = timerqueue_getnext(&rtc->timerqueue))) {
+		if (next->expires > now)
+			break;
+
+		/* expire timer */
+		timer = container_of(next, struct rtc_timer, node);
+		timerqueue_del(&rtc->timerqueue, &timer->node);
+		trace_rtc_timer_dequeue(timer);
+		timer->enabled = 0;
+		if (timer->func)
+			timer->func(timer->rtc);
+
+		trace_rtc_timer_fired(timer);
+		/* Re-add/fwd periodic timers */
+		if (ktime_to_ns(timer->period)) {
+			timer->node.expires = ktime_add(timer->node.expires,
+							timer->period);
+			timer->enabled = 1;
+			timerqueue_add(&rtc->timerqueue, &timer->node);
+			trace_rtc_timer_enqueue(timer);
+		}
+	}
+
+	/* Set next alarm */
+	if (next) {
+		struct rtc_wkalrm alarm;
+		int err;
+		int retry = 3;
+
+		alarm.time = rtc_ktime_to_tm(next->expires);
+		alarm.enabled = 1;
+reprogram:
+		err = __rtc_set_alarm(rtc, &alarm);
+		if (err == -ETIME) {
+			goto again;
+		} else if (err) {
+			if (retry-- > 0)
+				goto reprogram;
+
+			timer = container_of(next, struct rtc_timer, node);
+			timerqueue_del(&rtc->timerqueue, &timer->node);
+			trace_rtc_timer_dequeue(timer);
+			timer->enabled = 0;
+			dev_err(&rtc->dev, "__rtc_set_alarm: err=%d\n", err);
+			goto again;
+		}
+	} else {
+		rtc_alarm_disable(rtc);
+	}
+
+	pm_relax(rtc->dev.parent);
+	mutex_unlock(&rtc->ops_lock);
+}
+
+/* rtc_timer_init - Initializes an rtc_timer
+ * @timer: timer to be intiialized
+ * @f: function pointer to be called when timer fires
+ * @rtc: pointer to the rtc_device
+ *
+ * Kernel interface to initializing an rtc_timer.
+ */
+void rtc_timer_init(struct rtc_timer *timer, void (*f)(struct rtc_device *r),
+		    struct rtc_device *rtc)
+{
+	timerqueue_init(&timer->node);
+	timer->enabled = 0;
+	timer->func = f;
+	timer->rtc = rtc;
+}
+
+/* rtc_timer_start - Sets an rtc_timer to fire in the future
+ * @ rtc: rtc device to be used
+ * @ timer: timer being set
+ * @ expires: time at which to expire the timer
+ * @ period: period that the timer will recur
+ *
+ * Kernel interface to set an rtc_timer
+ */
+int rtc_timer_start(struct rtc_device *rtc, struct rtc_timer *timer,
+		    ktime_t expires, ktime_t period)
+{
+	int ret = 0;
+
+	mutex_lock(&rtc->ops_lock);
+	if (timer->enabled)
+		rtc_timer_remove(rtc, timer);
+
+	timer->node.expires = expires;
+	timer->period = period;
+
+	ret = rtc_timer_enqueue(rtc, timer);
+
+	mutex_unlock(&rtc->ops_lock);
+	return ret;
+}
+
+/* rtc_timer_cancel - Stops an rtc_timer
+ * @ rtc: rtc device to be used
+ * @ timer: timer being set
+ *
+ * Kernel interface to cancel an rtc_timer
+ */
+void rtc_timer_cancel(struct rtc_device *rtc, struct rtc_timer *timer)
+{
+	mutex_lock(&rtc->ops_lock);
+	if (timer->enabled)
+		rtc_timer_remove(rtc, timer);
+	mutex_unlock(&rtc->ops_lock);
+}
+
+/**
+ * rtc_read_offset - Read the amount of rtc offset in parts per billion
+ * @rtc: rtc device to be used
+ * @offset: the offset in parts per billion
+ *
+ * see below for details.
+ *
+ * Kernel interface to read rtc clock offset
+ * Returns 0 on success, or a negative number on error.
+ * If read_offset() is not implemented for the rtc, return -EINVAL
+ */
+int rtc_read_offset(struct rtc_device *rtc, long *offset)
+{
+	int ret;
+
+	if (!rtc->ops)
+		return -ENODEV;
+
+	if (!rtc->ops->read_offset)
+		return -EINVAL;
+
+	mutex_lock(&rtc->ops_lock);
+	ret = rtc->ops->read_offset(rtc->dev.parent, offset);
+	mutex_unlock(&rtc->ops_lock);
+
+	trace_rtc_read_offset(*offset, ret);
+	return ret;
+}
+
+/**
+ * rtc_set_offset - Adjusts the duration of the average second
+ * @rtc: rtc device to be used
+ * @offset: the offset in parts per billion
+ *
+ * Some rtc's allow an adjustment to the average duration of a second
+ * to compensate for differences in the actual clock rate due to temperature,
+ * the crystal, capacitor, etc.
+ *
+ * The adjustment applied is as follows:
+ *   t = t0 * (1 + offset * 1e-9)
+ * where t0 is the measured length of 1 RTC second with offset = 0
+ *
+ * Kernel interface to adjust an rtc clock offset.
+ * Return 0 on success, or a negative number on error.
+ * If the rtc offset is not setable (or not implemented), return -EINVAL
+ */
+int rtc_set_offset(struct rtc_device *rtc, long offset)
+{
+	int ret;
+
+	if (!rtc->ops)
+		return -ENODEV;
+
+	if (!rtc->ops->set_offset)
+		return -EINVAL;
+
+	mutex_lock(&rtc->ops_lock);
+	ret = rtc->ops->set_offset(rtc->dev.parent, offset);
+	mutex_unlock(&rtc->ops_lock);
+
+	trace_rtc_set_offset(offset, ret);
+	return ret;
+}