Factor out the RTC initialization from the platform device specific
parts in order to share the RTC device ops with other drivers.
Specifically, it will be shared with rtc-pxa driver.
Signed-off-by: NRob Herring <robh@kernel.org>
Cc: Robert Jarzmik <robert.jarzmik@free.fr>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Cc: rtc-linux@googlegroups.com
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

If ds3232 work on some platform that is not implementing
irq_set_wake, ds3232 will get a WARNING trace in resume.
So fix ds3232->suspended state to false when irq_set_irq_wake
return error.

WARNING: CPU: 0 PID: 729 at kernel/irq/manage.c:604 irq_set_irq_wake+0x4b/0x8c()
Unbalanced IRQ 201 wake disable
Modules linked in:
CPU: 0 PID: 729 Comm: sh Not tainted 3.12.19-rt30+ #25
[<800107d9>] (unwind_backtrace+0x1/0x88) from [<8000e4ef>] (show_stack+0xb/0xc)
[<8000e4ef>] (show_stack+0xb/0xc) from [<802b5fa9>] (dump_stack+0x4d/0x60)
[<802b5fa9>] (dump_stack+0x4d/0x60) from [<800186dd>] (warn_slowpath_common+0x45/0x64)
[<800186dd>] (warn_slowpath_common+0x45/0x64) from [<80018717>] (warn_slowpath_fmt+0x1b/0x24)
[<80018717>] (warn_slowpath_fmt+0x1b/0x24) from [<8003a8d3>] (irq_set_irq_wake+0x4b/0x8c)
[<8003a8d3>] (irq_set_irq_wake+0x4b/0x8c) from [<80204fcb>] (ds3232_resume+0x2d/0x36)
[<80204fcb>] (ds3232_resume+0x2d/0x36) from [<801954c7>] (dpm_run_callback.isra.13+0xb/0x28)
[<801954c7>] (dpm_run_callback.isra.13+0xb/0x28) from [<80195b1b>] (device_resume+0x7b/0xa2)
[<80195b1b>] (device_resume+0x7b/0xa2) from [<80195f0f>] (dpm_resume+0xbb/0x19c)
[<80195f0f>] (dpm_resume+0xbb/0x19c) from [<801960d9>] (dpm_resume_end+0x9/0x12)
[<801960d9>] (dpm_resume_end+0x9/0x12) from [<80037e1d>] (suspend_devices_and_enter+0x17d/0x1d0)
[<80037e1d>] (suspend_devices_and_enter+0x17d/0x1d0) from [<80037ee1>] (pm_suspend+0x71/0x128)
[<80037ee1>] (pm_suspend+0x71/0x128) from [<80037449>] (state_store+0x6d/0x80)
[<80037449>] (state_store+0x6d/0x80) from [<800af4d5>] (sysfs_write_file+0x9f/0xde)
[<800af4d5>] (sysfs_write_file+0x9f/0xde) from [<8007a437>] (vfs_write+0x7b/0x104)
[<8007a437>] (vfs_write+0x7b/0x104) from [<8007a7f7>] (SyS_write+0x27/0x48)
[<8007a7f7>] (SyS_write+0x27/0x48) from [<8000c121>] (ret_fast_syscall+0x1/0x44)
Signed-off-by: NWang Dongsheng <dongsheng.wang@freescale.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

It's unnecessary the code that assigns info->rtc_clk to NULL in
s3c_rtc_remove.
Signed-off-by: NJoonyoung Shim <jy0922.shim@samsung.com>
Reviewed-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

It's missed to call clk_unprepare() about info->rtc_src_clk in
s3c_rtc_remove and to call clk_disable_unprepare about info->rtc_clk in
error routine of s3c_rtc_probe.
Signed-off-by: NJoonyoung Shim <jy0922.shim@samsung.com>
Reviewed-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The clock enable/disable codes for alarm have been removed from
commit 24e14554 ("drivers/rtc/rtc-s3c.c: delete duplicate clock
control") and the clocks are disabled even if alarm is set, so alarm
interrupt can't happen.

The s3c_rtc_setaie function can be called several times with 'enabled'
argument having same value, so it needs to check whether clocks are
enabled or not.
Signed-off-by: NJoonyoung Shim <jy0922.shim@samsung.com>
Cc: <stable@vger.kernel.org> # v4.1
Reviewed-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Definitions from linux/platform_data/atmel.h are not used, remove the
include.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

This driver is using device tree but is not including of.h
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

According to the Armada38x functional errata FE-3124064, writing to
the RTC TIME register may fail. As a workaround, after writing to RTC
TIME register, issue a dummy write of 0x0 twice to the RTC Status
register.  This is the updated implementation of the Errata that
eliminates the need of the long 100ms delay during the RTC set time
procedure.

[gregory.clement@free-electrons.com]: removed the mutex and use the
spinlock again
Signed-off-by: NNadav Haklai <nadavh@marvell.com>
Reviewed-by: NNeta Zur Hershkovits <neta@marvell.com>
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Check time validity when reading time as this is when we need to know.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

RX8025_BIT_CTRL2_CTFG was set to 0 only when it was already 0.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

irq_freq is already initialized to 1 in rtc_device_register()
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Wait for the user to set the time to reset the validity bits. Until then,
the time may be invalid.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

rx8025_init_client is modifying ctrl[0] and writing it to RX8025_REG_CTRL2
but ctrl[0] is actually RX8025_REG_CTRL1.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Instead of bailing out, disable alarms and continue when
devm_request_threaded_irq() fails. This allows to still provide some
functionality.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Remove useless error messages, at that point, the user already knows
something went wrong but will not be able to do anything about it anyway.
It is also highly unlikely that some registers are readable/writable
but not some other ones.

Also, transform rx8025_read_reg to be more resemblant to
i2c_smbus_read_byte_data()
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Stop setting the time to epoch when it is invalid. The proper way to handle
that is to return an error when it is invalid instead of returning an
incorrect value.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The datasheet specifies that transfer mode must be 0 for write and either
0x4 (simplified read) or 0 (standard read). 0x8 is not specified, use
standard mode.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The hardware is only capable of handling dates between 2000 and 2099,
enforce that.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Use BIT() instead of hand coding.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

It is useless to print a message when probe fails as the user is already
aware that it failed.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Use devm_request_threaded_irq() so it is not necessary to call free_irq()
explicitly.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The driver currently emulates the concept of threaded IRQ using a
workqueue, switch to threaded IRQ instead.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Remove the useless includes and order the remaining ones alphabetically.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c

Spinlock acquisition/release is moved out of the loop body to get
atomic states of NVRAM reading and writing operations.
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c

The change enables burst mode of access to SRAM for any read()/write()
operations, it is worth to mention that this may influence on
userspace, for instance prior to the change

  read(fd, buf, 1);
  read(fd, buf + 1, 1);

and

  read(fd, buf, 2);

sequences of syscalls over DS1511's sysfs "nvram" fd led to different
DS1511 state changes and/or buf content, if some userspace applications
are written specifically for DS1511 and exploit this strange
"feature", they may be impacted.

Also the change corrects NVRAM size accessible to userspace from 255
bytes to 256 bytes.
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller size in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The change removes redundant sysfs binary file boundary checks, since
this task is already done on caller side in fs/sysfs/file.c
Signed-off-by: NVladimir Zapolskiy <vz@mleia.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

If century field is supported by the RTC CMOS device, then we should use
it and then do not consider years greater that 169 as an error.

For information, the year field of the rtc_time structure contains the
value to add to 1970 to obtain the current year.

This was a hack to be able to support years for 1970 to 2069.
This patch remains compatible with this implementation.
Signed-off-by: NSylvain Chouleur <sylvain.chouleur@intel.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Commit dca1a4b5 ("clk: at91: keep slow clk enabled to prevent system
hang") added a workaround for the slow clock as it is not properly handled
by its users.

Get and use the slow clock as it is necessary for the at91rm9200 rtc.
Acked-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

drivers/rtc/rtc-gemini.c:151:1-3: WARNING: PTR_ERR_OR_ZERO can be used

 Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR

Generated by: scripts/coccinelle/api/ptr_ret.cocci
Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

IS_ERR(_OR_NULL) already contain an 'unlikely' compiler flag and there
is no need to do that again from its callers. Drop it.

gemini driver was using likely() for a failure case while the rtc driver
is getting registered. That looks wrong and it should really be
unlikely. But because we are killing all the unlikely() flags, lets kill
that too.
Signed-off-by: NViresh Kumar <viresh.kumar@linaro.org>
Acked-by: NHans Ulli Kroll <ulli.kroll@googlemail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Sort included headers alphabetically.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>

See help for clk_get_rate(): "obtain the current clock rate (in Hz) for a
clock source. This is only valid once the clock source has been enabled."

It currently returns the correct value but that may not stay that way.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>