2019-03-14 06:02:48 +08:00
|
|
|
// SPDX-License-Identifier: GPL-2.0
|
2006-03-27 17:16:37 +08:00
|
|
|
/*
|
|
|
|
* RTC subsystem, base class
|
|
|
|
*
|
|
|
|
* Copyright (C) 2005 Tower Technologies
|
|
|
|
* Author: Alessandro Zummo <a.zummo@towertech.it>
|
|
|
|
*
|
|
|
|
* class skeleton from drivers/hwmon/hwmon.c
|
2019-03-14 06:02:48 +08:00
|
|
|
*/
|
2006-03-27 17:16:37 +08:00
|
|
|
|
2013-02-22 08:45:23 +08:00
|
|
|
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
|
|
|
|
|
2006-03-27 17:16:37 +08:00
|
|
|
#include <linux/module.h>
|
2014-01-24 07:55:14 +08:00
|
|
|
#include <linux/of.h>
|
2006-03-27 17:16:37 +08:00
|
|
|
#include <linux/rtc.h>
|
|
|
|
#include <linux/kdev_t.h>
|
|
|
|
#include <linux/idr.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
|
|
|
#include <linux/slab.h>
|
RTC: Rework RTC code to use timerqueue for events
This patch reworks a large portion of the generic RTC code
to in-effect virtualize the rtc interrupt code.
The current RTC interface is very much a raw hardware interface.
Via the proc, /dev/, or sysfs interfaces, applciations can set
the hardware to trigger interrupts in one of three modes:
AIE: Alarm interrupt
UIE: Update interrupt (ie: once per second)
PIE: Periodic interrupt (sub-second irqs)
The problem with this interface is that it limits the RTC hardware
so it can only be used by one application at a time.
The purpose of this patch is to extend the RTC code so that we can
multiplex multiple applications event needs onto a single RTC device.
This is done by utilizing the timerqueue infrastructure to manage
a list of events, which cause the RTC hardware to be programmed
to fire an interrupt for the next event in the list.
In order to preserve the functionality of the exsting proc,/dev/ and
sysfs interfaces, we emulate the different interrupt modes as follows:
AIE: We create a rtc_timer dedicated to AIE mode interrupts. There is
only one per device, so we don't change existing interface semantics.
UIE: Again, a dedicated rtc_timer, set for periodic mode, is used
to emulate UIE interrupts. Again, only one per device.
PIE: Since PIE mode interrupts fire faster then the RTC's clock read
granularity, we emulate PIE mode interrupts using a hrtimer. Again,
one per device.
With this patch, the rtctest.c application in Documentation/rtc.txt
passes fine on x86 hardware. However, there may very well still be
bugs, so greatly I'd appreciate any feedback or testing!
Signed-off-by: John Stultz <john.stultz@linaro.org>
LKML Reference: <1290136329-18291-4-git-send-email-john.stultz@linaro.org>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Richard Cochran <richardcochran@gmail.com>
2010-09-24 06:07:34 +08:00
|
|
|
#include <linux/workqueue.h>
|
2006-03-27 17:16:37 +08:00
|
|
|
|
2007-05-08 15:33:27 +08:00
|
|
|
#include "rtc-core.h"
|
|
|
|
|
2011-11-03 04:37:49 +08:00
|
|
|
static DEFINE_IDA(rtc_ida);
|
2006-03-27 17:16:37 +08:00
|
|
|
struct class *rtc_class;
|
|
|
|
|
2007-05-08 15:33:40 +08:00
|
|
|
static void rtc_device_release(struct device *dev)
|
2006-03-27 17:16:37 +08:00
|
|
|
{
|
2007-05-08 15:33:40 +08:00
|
|
|
struct rtc_device *rtc = to_rtc_device(dev);
|
2019-03-20 19:59:09 +08:00
|
|
|
|
2011-11-03 04:37:49 +08:00
|
|
|
ida_simple_remove(&rtc_ida, rtc->id);
|
2006-03-27 17:16:37 +08:00
|
|
|
kfree(rtc);
|
|
|
|
}
|
|
|
|
|
2012-10-05 08:14:12 +08:00
|
|
|
#ifdef CONFIG_RTC_HCTOSYS_DEVICE
|
|
|
|
/* Result of the last RTC to system clock attempt. */
|
|
|
|
int rtc_hctosys_ret = -ENODEV;
|
|
|
|
#endif
|
2007-05-08 15:33:42 +08:00
|
|
|
|
2013-07-04 06:07:59 +08:00
|
|
|
#if defined(CONFIG_PM_SLEEP) && defined(CONFIG_RTC_HCTOSYS_DEVICE)
|
2007-05-08 15:33:42 +08:00
|
|
|
/*
|
|
|
|
* On suspend(), measure the delta between one RTC and the
|
|
|
|
* system's wall clock; restore it on resume().
|
|
|
|
*/
|
|
|
|
|
2014-11-21 13:06:12 +08:00
|
|
|
static struct timespec64 old_rtc, old_system, old_delta;
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
|
2013-07-04 06:07:59 +08:00
|
|
|
static int rtc_suspend(struct device *dev)
|
2007-05-08 15:33:42 +08:00
|
|
|
{
|
|
|
|
struct rtc_device *rtc = to_rtc_device(dev);
|
|
|
|
struct rtc_time tm;
|
2014-11-21 13:06:12 +08:00
|
|
|
struct timespec64 delta, delta_delta;
|
2014-08-09 05:20:24 +08:00
|
|
|
int err;
|
2013-01-16 00:09:48 +08:00
|
|
|
|
2015-04-02 11:34:38 +08:00
|
|
|
if (timekeeping_rtc_skipsuspend())
|
2013-01-16 00:09:48 +08:00
|
|
|
return 0;
|
|
|
|
|
2009-01-07 06:42:11 +08:00
|
|
|
if (strcmp(dev_name(&rtc->dev), CONFIG_RTC_HCTOSYS_DEVICE) != 0)
|
2007-05-08 15:33:42 +08:00
|
|
|
return 0;
|
|
|
|
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
/* snapshot the current RTC and system time at suspend*/
|
2014-08-09 05:20:24 +08:00
|
|
|
err = rtc_read_time(rtc, &tm);
|
|
|
|
if (err < 0) {
|
|
|
|
pr_debug("%s: fail to read rtc time\n", dev_name(&rtc->dev));
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-07-11 20:47:12 +08:00
|
|
|
ktime_get_real_ts64(&old_system);
|
2014-11-21 13:06:12 +08:00
|
|
|
old_rtc.tv_sec = rtc_tm_to_time64(&tm);
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* To avoid drift caused by repeated suspend/resumes,
|
|
|
|
* which each can add ~1 second drift error,
|
|
|
|
* try to compensate so the difference in system time
|
|
|
|
* and rtc time stays close to constant.
|
|
|
|
*/
|
2014-11-21 13:06:12 +08:00
|
|
|
delta = timespec64_sub(old_system, old_rtc);
|
|
|
|
delta_delta = timespec64_sub(delta, old_delta);
|
2011-11-23 10:24:51 +08:00
|
|
|
if (delta_delta.tv_sec < -2 || delta_delta.tv_sec >= 2) {
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
/*
|
|
|
|
* if delta_delta is too large, assume time correction
|
2019-03-20 19:59:09 +08:00
|
|
|
* has occurred and set old_delta to the current delta.
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
*/
|
|
|
|
old_delta = delta;
|
|
|
|
} else {
|
|
|
|
/* Otherwise try to adjust old_system to compensate */
|
2014-11-21 13:06:12 +08:00
|
|
|
old_system = timespec64_sub(old_system, delta_delta);
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
}
|
2007-05-08 15:33:42 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int rtc_resume(struct device *dev)
|
|
|
|
{
|
|
|
|
struct rtc_device *rtc = to_rtc_device(dev);
|
|
|
|
struct rtc_time tm;
|
2014-11-21 13:06:12 +08:00
|
|
|
struct timespec64 new_system, new_rtc;
|
|
|
|
struct timespec64 sleep_time;
|
2014-08-09 05:20:24 +08:00
|
|
|
int err;
|
2007-05-08 15:33:42 +08:00
|
|
|
|
2015-04-02 11:34:38 +08:00
|
|
|
if (timekeeping_rtc_skipresume())
|
2013-01-16 00:09:48 +08:00
|
|
|
return 0;
|
|
|
|
|
2012-10-05 08:14:12 +08:00
|
|
|
rtc_hctosys_ret = -ENODEV;
|
2009-01-07 06:42:11 +08:00
|
|
|
if (strcmp(dev_name(&rtc->dev), CONFIG_RTC_HCTOSYS_DEVICE) != 0)
|
2007-05-08 15:33:42 +08:00
|
|
|
return 0;
|
|
|
|
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
/* snapshot the current rtc and system time at resume */
|
2018-07-11 20:47:12 +08:00
|
|
|
ktime_get_real_ts64(&new_system);
|
2014-08-09 05:20:24 +08:00
|
|
|
err = rtc_read_time(rtc, &tm);
|
|
|
|
if (err < 0) {
|
|
|
|
pr_debug("%s: fail to read rtc time\n", dev_name(&rtc->dev));
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-11-21 13:06:12 +08:00
|
|
|
new_rtc.tv_sec = rtc_tm_to_time64(&tm);
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
new_rtc.tv_nsec = 0;
|
|
|
|
|
2011-11-23 10:24:51 +08:00
|
|
|
if (new_rtc.tv_sec < old_rtc.tv_sec) {
|
|
|
|
pr_debug("%s: time travel!\n", dev_name(&rtc->dev));
|
2007-05-08 15:33:42 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
/* calculate the RTC time delta (sleep time)*/
|
2014-11-21 13:06:12 +08:00
|
|
|
sleep_time = timespec64_sub(new_rtc, old_rtc);
|
rtc: Avoid accumulating time drift in suspend/resume
Because the RTC interface is only a second granular interface,
each time we read from the RTC for suspend/resume, we introduce a
half second (on average) of error.
In order to avoid this error accumulating as the system is suspended
over and over, this patch measures the time delta between the RTC
and the system CLOCK_REALTIME.
If the delta is less then 2 seconds from the last suspend, we compensate
by using the previous time delta (keeping it close). If it is larger
then 2 seconds, we assume the clock was set or has been changed, so we
do no correction and update the delta.
Note: If NTP is running, ths could seem to "fight" with the NTP corrected
time, where as if the system time was off by 1 second, and NTP slewed the
value in, a suspend/resume cycle could undo this correction, by trying to
restore the previous offset from the RTC. However, without this patch,
since each read could cause almost a full second worth of error, its
possible to get almost 2 seconds of error just from the suspend/resume
cycle alone, so this about equal to any offset added by the compensation.
Further on systems that suspend/resume frequently, this should keep time
closer then NTP could compensate for if the errors were allowed to
accumulate.
Credits to Arve Hjønnevåg for suggesting this solution.
This patch also improves some of the variable names and adds more clear
comments.
CC: Arve Hjønnevåg <arve@android.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
2011-05-28 02:33:18 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Since these RTC suspend/resume handlers are not called
|
|
|
|
* at the very end of suspend or the start of resume,
|
|
|
|
* some run-time may pass on either sides of the sleep time
|
|
|
|
* so subtract kernel run-time between rtc_suspend to rtc_resume
|
|
|
|
* to keep things accurate.
|
|
|
|
*/
|
2014-11-21 13:06:12 +08:00
|
|
|
sleep_time = timespec64_sub(sleep_time,
|
2019-03-20 19:59:09 +08:00
|
|
|
timespec64_sub(new_system, old_system));
|
2007-05-08 15:33:42 +08:00
|
|
|
|
2011-11-23 10:24:51 +08:00
|
|
|
if (sleep_time.tv_sec >= 0)
|
2014-11-21 13:06:12 +08:00
|
|
|
timekeeping_inject_sleeptime64(&sleep_time);
|
2012-10-05 08:14:12 +08:00
|
|
|
rtc_hctosys_ret = 0;
|
2007-05-08 15:33:42 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-07-04 06:07:59 +08:00
|
|
|
static SIMPLE_DEV_PM_OPS(rtc_class_dev_pm_ops, rtc_suspend, rtc_resume);
|
|
|
|
#define RTC_CLASS_DEV_PM_OPS (&rtc_class_dev_pm_ops)
|
2007-05-08 15:33:42 +08:00
|
|
|
#else
|
2013-07-04 06:07:59 +08:00
|
|
|
#define RTC_CLASS_DEV_PM_OPS NULL
|
2007-05-08 15:33:42 +08:00
|
|
|
#endif
|
|
|
|
|
2017-07-06 17:42:00 +08:00
|
|
|
/* Ensure the caller will set the id before releasing the device */
|
2017-07-06 17:41:58 +08:00
|
|
|
static struct rtc_device *rtc_allocate_device(void)
|
|
|
|
{
|
|
|
|
struct rtc_device *rtc;
|
|
|
|
|
|
|
|
rtc = kzalloc(sizeof(*rtc), GFP_KERNEL);
|
|
|
|
if (!rtc)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
device_initialize(&rtc->dev);
|
|
|
|
|
2017-10-14 01:54:33 +08:00
|
|
|
/* Drivers can revise this default after allocating the device. */
|
|
|
|
rtc->set_offset_nsec = NSEC_PER_SEC / 2;
|
|
|
|
|
2017-07-06 17:41:58 +08:00
|
|
|
rtc->irq_freq = 1;
|
|
|
|
rtc->max_user_freq = 64;
|
|
|
|
rtc->dev.class = rtc_class;
|
|
|
|
rtc->dev.groups = rtc_get_dev_attribute_groups();
|
|
|
|
rtc->dev.release = rtc_device_release;
|
|
|
|
|
|
|
|
mutex_init(&rtc->ops_lock);
|
|
|
|
spin_lock_init(&rtc->irq_lock);
|
|
|
|
init_waitqueue_head(&rtc->irq_queue);
|
|
|
|
|
|
|
|
/* Init timerqueue */
|
|
|
|
timerqueue_init_head(&rtc->timerqueue);
|
|
|
|
INIT_WORK(&rtc->irqwork, rtc_timer_do_work);
|
|
|
|
/* Init aie timer */
|
2018-12-19 05:11:26 +08:00
|
|
|
rtc_timer_init(&rtc->aie_timer, rtc_aie_update_irq, rtc);
|
2017-07-06 17:41:58 +08:00
|
|
|
/* Init uie timer */
|
2018-12-19 05:11:26 +08:00
|
|
|
rtc_timer_init(&rtc->uie_rtctimer, rtc_uie_update_irq, rtc);
|
2017-07-06 17:41:58 +08:00
|
|
|
/* Init pie timer */
|
|
|
|
hrtimer_init(&rtc->pie_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
|
|
|
|
rtc->pie_timer.function = rtc_pie_update_irq;
|
|
|
|
rtc->pie_enabled = 0;
|
|
|
|
|
|
|
|
return rtc;
|
|
|
|
}
|
2007-05-08 15:33:42 +08:00
|
|
|
|
2017-07-06 17:41:59 +08:00
|
|
|
static int rtc_device_get_id(struct device *dev)
|
|
|
|
{
|
|
|
|
int of_id = -1, id = -1;
|
|
|
|
|
|
|
|
if (dev->of_node)
|
|
|
|
of_id = of_alias_get_id(dev->of_node, "rtc");
|
|
|
|
else if (dev->parent && dev->parent->of_node)
|
|
|
|
of_id = of_alias_get_id(dev->parent->of_node, "rtc");
|
|
|
|
|
|
|
|
if (of_id >= 0) {
|
|
|
|
id = ida_simple_get(&rtc_ida, of_id, of_id + 1, GFP_KERNEL);
|
|
|
|
if (id < 0)
|
|
|
|
dev_warn(dev, "/aliases ID %d not available\n", of_id);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (id < 0)
|
|
|
|
id = ida_simple_get(&rtc_ida, 0, 0, GFP_KERNEL);
|
|
|
|
|
|
|
|
return id;
|
|
|
|
}
|
|
|
|
|
rtc: Add one offset seconds to expand RTC range
From our investigation for all RTC drivers, 1 driver will be expired before
year 2017, 7 drivers will be expired before year 2038, 23 drivers will be
expired before year 2069, 72 drivers will be expired before 2100 and 104
drivers will be expired before 2106. Especially for these early expired
drivers, we need to expand the RTC range to make the RTC can still work
after the expired year.
So we can expand the RTC range by adding one offset to the time when reading
from hardware, and subtracting it when writing back. For example, if you have
an RTC that can do 100 years, and currently is configured to be based in
Jan 1 1970, so it can represents times from 1970 to 2069. Then if you change
the start year from 1970 to 2000, which means it can represents times from
2000 to 2099. By adding or subtracting the offset produced by moving the wrap
point, all times between 1970 and 1999 from RTC hardware could get interpreted
as times from 2070 to 2099, but the interpretation of dates between 2000 and
2069 would not change.
Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
2018-01-08 14:04:50 +08:00
|
|
|
static void rtc_device_get_offset(struct rtc_device *rtc)
|
|
|
|
{
|
|
|
|
time64_t range_secs;
|
|
|
|
u32 start_year;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If RTC driver did not implement the range of RTC hardware device,
|
|
|
|
* then we can not expand the RTC range by adding or subtracting one
|
|
|
|
* offset.
|
|
|
|
*/
|
|
|
|
if (rtc->range_min == rtc->range_max)
|
|
|
|
return;
|
|
|
|
|
|
|
|
ret = device_property_read_u32(rtc->dev.parent, "start-year",
|
|
|
|
&start_year);
|
|
|
|
if (!ret) {
|
|
|
|
rtc->start_secs = mktime64(start_year, 1, 1, 0, 0, 0);
|
|
|
|
rtc->set_start_time = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If user did not implement the start time for RTC driver, then no
|
|
|
|
* need to expand the RTC range.
|
|
|
|
*/
|
|
|
|
if (!rtc->set_start_time)
|
|
|
|
return;
|
|
|
|
|
|
|
|
range_secs = rtc->range_max - rtc->range_min + 1;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the start_secs is larger than the maximum seconds (rtc->range_max)
|
|
|
|
* supported by RTC hardware or the maximum seconds of new expanded
|
|
|
|
* range (start_secs + rtc->range_max - rtc->range_min) is less than
|
|
|
|
* rtc->range_min, which means the minimum seconds (rtc->range_min) of
|
|
|
|
* RTC hardware will be mapped to start_secs by adding one offset, so
|
|
|
|
* the offset seconds calculation formula should be:
|
|
|
|
* rtc->offset_secs = rtc->start_secs - rtc->range_min;
|
|
|
|
*
|
|
|
|
* If the start_secs is larger than the minimum seconds (rtc->range_min)
|
|
|
|
* supported by RTC hardware, then there is one region is overlapped
|
|
|
|
* between the original RTC hardware range and the new expanded range,
|
|
|
|
* and this overlapped region do not need to be mapped into the new
|
|
|
|
* expanded range due to it is valid for RTC device. So the minimum
|
|
|
|
* seconds of RTC hardware (rtc->range_min) should be mapped to
|
|
|
|
* rtc->range_max + 1, then the offset seconds formula should be:
|
|
|
|
* rtc->offset_secs = rtc->range_max - rtc->range_min + 1;
|
|
|
|
*
|
|
|
|
* If the start_secs is less than the minimum seconds (rtc->range_min),
|
|
|
|
* which is similar to case 2. So the start_secs should be mapped to
|
|
|
|
* start_secs + rtc->range_max - rtc->range_min + 1, then the
|
|
|
|
* offset seconds formula should be:
|
|
|
|
* rtc->offset_secs = -(rtc->range_max - rtc->range_min + 1);
|
|
|
|
*
|
|
|
|
* Otherwise the offset seconds should be 0.
|
|
|
|
*/
|
|
|
|
if (rtc->start_secs > rtc->range_max ||
|
|
|
|
rtc->start_secs + range_secs - 1 < rtc->range_min)
|
|
|
|
rtc->offset_secs = rtc->start_secs - rtc->range_min;
|
|
|
|
else if (rtc->start_secs > rtc->range_min)
|
|
|
|
rtc->offset_secs = range_secs;
|
|
|
|
else if (rtc->start_secs < rtc->range_min)
|
|
|
|
rtc->offset_secs = -range_secs;
|
|
|
|
else
|
|
|
|
rtc->offset_secs = 0;
|
|
|
|
}
|
|
|
|
|
2006-03-27 17:16:37 +08:00
|
|
|
/**
|
|
|
|
* rtc_device_unregister - removes the previously registered RTC class device
|
|
|
|
*
|
|
|
|
* @rtc: the RTC class device to destroy
|
|
|
|
*/
|
2018-09-13 04:22:45 +08:00
|
|
|
static void rtc_device_unregister(struct rtc_device *rtc)
|
2006-03-27 17:16:37 +08:00
|
|
|
{
|
2015-07-21 07:02:50 +08:00
|
|
|
mutex_lock(&rtc->ops_lock);
|
|
|
|
/*
|
|
|
|
* Remove innards of this RTC, then disable it, before
|
|
|
|
* letting any rtc_class_open() users access it again
|
|
|
|
*/
|
|
|
|
rtc_proc_del_device(rtc);
|
2017-03-18 02:48:21 +08:00
|
|
|
cdev_device_del(&rtc->char_dev, &rtc->dev);
|
2015-07-21 07:02:50 +08:00
|
|
|
rtc->ops = NULL;
|
|
|
|
mutex_unlock(&rtc->ops_lock);
|
|
|
|
put_device(&rtc->dev);
|
2006-03-27 17:16:37 +08:00
|
|
|
}
|
|
|
|
|
2017-07-06 17:42:00 +08:00
|
|
|
static void devm_rtc_release_device(struct device *dev, void *res)
|
|
|
|
{
|
|
|
|
struct rtc_device *rtc = *(struct rtc_device **)res;
|
|
|
|
|
2018-03-01 04:58:02 +08:00
|
|
|
rtc_nvmem_unregister(rtc);
|
|
|
|
|
2017-07-06 17:42:00 +08:00
|
|
|
if (rtc->registered)
|
|
|
|
rtc_device_unregister(rtc);
|
|
|
|
else
|
|
|
|
put_device(&rtc->dev);
|
|
|
|
}
|
|
|
|
|
|
|
|
struct rtc_device *devm_rtc_allocate_device(struct device *dev)
|
|
|
|
{
|
|
|
|
struct rtc_device **ptr, *rtc;
|
|
|
|
int id, err;
|
|
|
|
|
|
|
|
id = rtc_device_get_id(dev);
|
|
|
|
if (id < 0)
|
|
|
|
return ERR_PTR(id);
|
|
|
|
|
|
|
|
ptr = devres_alloc(devm_rtc_release_device, sizeof(*ptr), GFP_KERNEL);
|
|
|
|
if (!ptr) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto exit_ida;
|
|
|
|
}
|
|
|
|
|
|
|
|
rtc = rtc_allocate_device();
|
|
|
|
if (!rtc) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto exit_devres;
|
|
|
|
}
|
|
|
|
|
|
|
|
*ptr = rtc;
|
|
|
|
devres_add(dev, ptr);
|
|
|
|
|
|
|
|
rtc->id = id;
|
|
|
|
rtc->dev.parent = dev;
|
|
|
|
dev_set_name(&rtc->dev, "rtc%d", id);
|
|
|
|
|
|
|
|
return rtc;
|
|
|
|
|
|
|
|
exit_devres:
|
|
|
|
devres_free(ptr);
|
|
|
|
exit_ida:
|
|
|
|
ida_simple_remove(&rtc_ida, id);
|
|
|
|
return ERR_PTR(err);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(devm_rtc_allocate_device);
|
|
|
|
|
|
|
|
int __rtc_register_device(struct module *owner, struct rtc_device *rtc)
|
|
|
|
{
|
|
|
|
struct rtc_wkalrm alrm;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
if (!rtc->ops)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
rtc->owner = owner;
|
rtc: Add one offset seconds to expand RTC range
From our investigation for all RTC drivers, 1 driver will be expired before
year 2017, 7 drivers will be expired before year 2038, 23 drivers will be
expired before year 2069, 72 drivers will be expired before 2100 and 104
drivers will be expired before 2106. Especially for these early expired
drivers, we need to expand the RTC range to make the RTC can still work
after the expired year.
So we can expand the RTC range by adding one offset to the time when reading
from hardware, and subtracting it when writing back. For example, if you have
an RTC that can do 100 years, and currently is configured to be based in
Jan 1 1970, so it can represents times from 1970 to 2069. Then if you change
the start year from 1970 to 2000, which means it can represents times from
2000 to 2099. By adding or subtracting the offset produced by moving the wrap
point, all times between 1970 and 1999 from RTC hardware could get interpreted
as times from 2070 to 2099, but the interpretation of dates between 2000 and
2069 would not change.
Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
2018-01-08 14:04:50 +08:00
|
|
|
rtc_device_get_offset(rtc);
|
2017-07-06 17:42:00 +08:00
|
|
|
|
|
|
|
/* Check to see if there is an ALARM already set in hw */
|
|
|
|
err = __rtc_read_alarm(rtc, &alrm);
|
|
|
|
if (!err && !rtc_valid_tm(&alrm.time))
|
|
|
|
rtc_initialize_alarm(rtc, &alrm);
|
|
|
|
|
|
|
|
rtc_dev_prepare(rtc);
|
|
|
|
|
|
|
|
err = cdev_device_add(&rtc->char_dev, &rtc->dev);
|
|
|
|
if (err)
|
|
|
|
dev_warn(rtc->dev.parent, "failed to add char device %d:%d\n",
|
|
|
|
MAJOR(rtc->dev.devt), rtc->id);
|
|
|
|
else
|
|
|
|
dev_dbg(rtc->dev.parent, "char device (%d:%d)\n",
|
|
|
|
MAJOR(rtc->dev.devt), rtc->id);
|
|
|
|
|
|
|
|
rtc_proc_add_device(rtc);
|
|
|
|
|
|
|
|
rtc->registered = true;
|
|
|
|
dev_info(rtc->dev.parent, "registered as %s\n",
|
|
|
|
dev_name(&rtc->dev));
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(__rtc_register_device);
|
|
|
|
|
2018-12-19 05:15:58 +08:00
|
|
|
/**
|
|
|
|
* devm_rtc_device_register - resource managed rtc_device_register()
|
|
|
|
* @dev: the device to register
|
|
|
|
* @name: the name of the device (unused)
|
|
|
|
* @ops: the rtc operations structure
|
|
|
|
* @owner: the module owner
|
|
|
|
*
|
|
|
|
* @return a struct rtc on success, or an ERR_PTR on error
|
|
|
|
*
|
|
|
|
* Managed rtc_device_register(). The rtc_device returned from this function
|
|
|
|
* are automatically freed on driver detach.
|
|
|
|
* This function is deprecated, use devm_rtc_allocate_device and
|
|
|
|
* rtc_register_device instead
|
|
|
|
*/
|
|
|
|
struct rtc_device *devm_rtc_device_register(struct device *dev,
|
2019-03-20 19:59:09 +08:00
|
|
|
const char *name,
|
|
|
|
const struct rtc_class_ops *ops,
|
|
|
|
struct module *owner)
|
2018-12-19 05:15:58 +08:00
|
|
|
{
|
|
|
|
struct rtc_device *rtc;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
rtc = devm_rtc_allocate_device(dev);
|
|
|
|
if (IS_ERR(rtc))
|
|
|
|
return rtc;
|
|
|
|
|
|
|
|
rtc->ops = ops;
|
|
|
|
|
|
|
|
err = __rtc_register_device(owner, rtc);
|
|
|
|
if (err)
|
|
|
|
return ERR_PTR(err);
|
|
|
|
|
|
|
|
return rtc;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(devm_rtc_device_register);
|
|
|
|
|
2006-03-27 17:16:37 +08:00
|
|
|
static int __init rtc_init(void)
|
|
|
|
{
|
|
|
|
rtc_class = class_create(THIS_MODULE, "rtc");
|
|
|
|
if (IS_ERR(rtc_class)) {
|
2013-02-22 08:45:23 +08:00
|
|
|
pr_err("couldn't create class\n");
|
2006-03-27 17:16:37 +08:00
|
|
|
return PTR_ERR(rtc_class);
|
|
|
|
}
|
2013-07-04 06:07:59 +08:00
|
|
|
rtc_class->pm = RTC_CLASS_DEV_PM_OPS;
|
2007-05-08 15:33:27 +08:00
|
|
|
rtc_dev_init();
|
2006-03-27 17:16:37 +08:00
|
|
|
return 0;
|
|
|
|
}
|
2006-10-01 14:28:15 +08:00
|
|
|
subsys_initcall(rtc_init);
|