All the clocksource drivers's init function are now converted to return
an error code. CLOCKSOURCE_OF_DECLARE is no longer used as well as the
clksrc-of table.

Let's convert back the names:
 - CLOCKSOURCE_OF_DECLARE_RET => CLOCKSOURCE_OF_DECLARE
 - clksrc-of-ret              => clksrc-of
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

For exynos_mct and samsung_pwm_timer:
Acked-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>

For arch/arc:
Acked-by: NVineet Gupta <vgupta@synopsys.com>

For mediatek driver:
Acked-by: NMatthias Brugger <matthias.bgg@gmail.com>

For the Rockchip-part
Acked-by: NHeiko Stuebner <heiko@sntech.de>

For STi :
Acked-by: NPatrice Chotard <patrice.chotard@st.com>

For the mps2-timer.c and versatile.c changes:
Acked-by: NLiviu Dudau <Liviu.Dudau@arm.com>

For the OXNAS part :
Acked-by: NNeil Armstrong <narmstrong@baylibre.com>

For LPC32xx driver:
Acked-by: NSylvain Lemieux <slemieux.tyco@gmail.com>

For Broadcom Kona timer change:
Acked-by: NRay Jui <ray.jui@broadcom.com>

For Sun4i and Sun5i:
Acked-by: NChen-Yu Tsai <wens@csie.org>

For Meson6:
Acked-by: NCarlo Caione <carlo@caione.org>

For Keystone:
Acked-by: NSantosh Shilimkar <ssantosh@kernel.org>

For NPS:
Acked-by: NNoam Camus <noamca@mellanox.com>

For bcm2835:
Acked-by: NEric Anholt <eric@anholt.net>

The init functions do not return any error. They behave as the following:

  - panic, thus leading to a kernel crash while another timer may work and
       make the system boot up correctly

  or

  - print an error and let the caller unaware if the state of the system

Change that by converting the init functions to return an error conforming
to the CLOCKSOURCE_OF_RET prototype.

Proper error handling (rollback, errno value) will be changed later case
by case, thus this change just return back an error or success in the init
function.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

Allow the timer core to change the smp affinity of the broadcast timer
irq by setting CLOCK_EVT_FEAT_DYNIRQ flag. For this to work the timer
core needs to be told about the used irq.

This reduces interrupt pressure and wakeups on CPU0 as well as vastly
reducing the number of timer broadcast IPIs.
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

Commit 6dd74782 ("ARM: imx: move timer resources into a structure")
moved initialization parameters into a data structure, but neglected to set
the irq field in that data structure for non-DT boots. This causes the system
to hang if a non-DT boot is attempted.

Fixes: 6dd74782 ("ARM: imx: move timer resources into a structure")
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: http://lkml.kernel.org/r/1440066441-13930-1-git-send-email-linux@roeck-us.netSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

Migrate timer-imx-gpt driver to the new 'set-state' interface provided
by clockevents core, the earlier 'set-mode' interface is marked obsolete
now.

This also enables us to implement callbacks for new states of clockevent
devices, for example: ONESHOT_STOPPED.

Also drop:
- 'imx_timer.cem': It was caching the last state of the clockevent
  device. The same behavior can be achieved by using clockevents state
  helpers. These helpers are only required for oneshot mode as
  shutdown/resume wouldn't be done twice by the core.

- 'clock_event_mode_label': CLOCK_EVT_MODE_* shouldn't be used anymore
  by drivers. The prints are modified to print the set-state functions
  name now to debug the driver.

Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

The rework of the imx clocksource driver missed to add an entry for
imx27 which results in a boot failure on those machines.

Add the proper CLOCKSOURCE_OF_DECLARE() entry for imx27 and map it to
the imx21 init.

Fixes: bef11c88 'ARM: imx: initialize gpt device type for DT boot'
Signed-off-by: NPhilippe Reynes <tremyfr@gmail.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: daniel.lezcano@linaro.org
Cc: fabio.estevam@freescale.com
Cc: shawnguo@kernel.org
Cc: kernel@pengutronix.de
Link: http://lkml.kernel.org/r/1435439504-406-1-git-send-email-tremyfr@gmail.comSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

The include of <asm/mach/time.h> is not needed at all, and causes build
error in some cases.  Remove it.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

After the cleanup on imx timer driver, now it's ready to be moved into
drivers/clocksource/.  Let's do it.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>

With the cleanup done before, the platform specific headers now can be
removed.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

It splits irq enable/disable/acknowledge operations into device specific
functions as the hooks in imx_gpt_data, so that we can save the use of
timer_is_xxx() and cpu_is_xxx() checking in these irq functions.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

We now have pointer to imx_timer structure available where timer base
address is needed, so we can just kill global timer_base by using
imxtm->base instead.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

It defines offset of gpt registers TSTAT, TCN and TCMP per device
type in imx_gpt_data, so that these registers can be accessed in an
way without timer_is_v2() checking.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Since we now have imx_timer structure, it makes more sense to move those
clock event related variables into the structure, so that we can save
some global variables.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Set up .set_next_event hook via imx_gpt_data, so that we can save the
use of timer_is_v2().
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

It creates a gpt device speicific data structure and adds function hook
gpt_setup_tctl in there to set up gpt TCTL register.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Use different initialization function in CLOCKSOURCE_OF_DECLARE() to
initialize gpt device type for DT boot.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Define an enum for gpt timer device type in include/soc/imx/timer.h to
tell the gpt block differences among SoCs.  Update non-DT users (clock
drivers) to pass the device type.

As we now have include/soc/imx/timer.h, the declaration of
mxc_timer_init() is moved into there as the best fit.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Instead of passing around as individual argument, let's move timer
resources like irq and clocks together with base address into a data
structure, and pass pointer of the structure as argument to simplify
the function call interface.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Replace the __raw_readl/__raw_writel with readl_relaxed/writel_relaxed
which is endian-safe, as a step of moving the driver code into folder
drivers/clocksource.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

The comments were corrected as the following to reflect
the real situation of Freescale MXC timer IP block.
There are totally 4 version of the timer on Freescale i.MX SoCs.
Signed-off-by: NShenwei Wang <shenwei.wang@freescale.com>
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Pass physical address of timer block to mxc_timer_init() call, which in
turn does dynamic mapping within the function.  Thus, we can avoid using
static mapping in clock drivers.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

On i.MX6Q TO > 1.0, i.MX6DL and i.MX6SX, gpt per clock
can be from OSC instead of ipg_per, as ipg_per's rate
may be scaled when system enter low bus mode, to keep
system timer NOT drift, better to make gpt per clock
at fixed rate, here add support for gpt per clock to
be from OSC which is at fixed rate always.

There are some difference on this implementation of
gpt per clock source, see below for details:

i.MX6Q TO > 1.0: GPT_CR_CLKSRC, b'101 selects fix clock
    of OSC / 8 for gpt per clk;
i.MX6DL and i.MX6SX: GPT_CR_CLKSRC, b'101 selects OSC
    for gpt per clk, and we must enable GPT_CR_24MEM to
    enable OSC clk source for gpt per, GPT_PR_PRESCALER24M
    is for pre-scaling of this OSC clk, here set it to 8
    to make gpt per clk is 3MHz;
i.MX6SL: ipg_per can be from OSC directly, so no need to
    implement this new clk source for gpt per.
Signed-off-by: NAnson Huang <b20788@freescale.com>
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>

This patch uses clocksource_of_init() call for DT targets.
Signed-off-by: NAlexander Shiyan <shc_work@mail.ru>
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>

Base address for driver is global, there are no need to use
intermediate variable for it.
Signed-off-by: NAlexander Shiyan <shc_work@mail.ru>
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>

Use of_clk_get_by_name() for timer clocks for DT case.
This patch eliminates a lot of unneeded clk_register_clkdev()
calls for GPT.
Signed-off-by: NAlexander Shiyan <shc_work@mail.ru>
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>

Signed-off-by: NGilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>

The imx can support timer-based delays, so implement this.
Skips past jiffy calibration.
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

The 32 bit sched_clock interface now supports 64 bits. Upgrade to
the 64 bit function to allow us to remove the 32 bit registration
interface.

Cc: Sascha Hauer <kernel@pengutronix.de>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NKevin Hilman <khilman@linaro.org>

This flag is a NOOP since 2.6.35 and can be removed.

This is an update for 3.11 of a patch already sent for 3.10
Signed-off-by: NMichael Opdenacker <michael.opdenacker@free-electrons.com>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Nothing about the sched_clock implementation in the ARM port is
specific to the architecture. Generalize the code so that other
architectures can use it by selecting GENERIC_SCHED_CLOCK.
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
[jstultz: Merge minor collisions with other patches in my tree]
Signed-off-by: NJohn Stultz <john.stultz@linaro.org>

The return of v2_set_next_event() will lead to an infinite loop in
tick_handle_oneshot_broadcast() - "goto again;" with imx6q WAIT mode
(to be enabled).  This happens because when global event did not expire
any CPU local events, the broadcast device will be rearmed to a CPU
local next_event, which could be far away from now and result in a
max_delta_tick programming in set_next_event().

Fix the problem by detecting those next events with increments larger
than 0x7fffffff, and simply return zero in that case.

It leaves mx1_2_set_next_event() unchanged since only v2_set_next_event()
will be running with imx6q WAIT mode support.

Thanks Russell King for helping understand the problem.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

The clockevent core is able to figure out the best mult and shift,
calculate min_delta_ns and max_delta_ns, with the necessary info passed
into clockevents_config_and_register().  Use this combined configure
and register function where possible to make the codes less error prone
and gain some positive diff stat.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Cc: Nicolas Ferre <nicolas.ferre@atmel.com>
Reviewed-by: NAnton Vorontsov <cbouatmailru@gmail.com>
Acked-by: NKukjin Kim <kgene.kim@samsung.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Jason Cooper <jason@lakedaemon.net>
Tested-by: NRoland Stigge <stigge@antcom.de>
Acked-by: NEric Miao <eric.y.miao@gmail.com>
Cc: Haojian Zhuang <haojian.zhuang@gmail.com>
Cc: David Brown <davidb@codeaurora.org>
Tested-by: NTony Lindgren <tony@atomide.com>
Acked-by: NBarry Song <baohua.song@csr.com>
Tested-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NTony Prisk <linux@prisktech.co.nz>
Cc: Lennert Buytenhek <buytenh@wantstofly.org>
Cc: Wan ZongShun <mcuos.com@gmail.com>
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Reviewed-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NOlof Johansson <olof@lixom.net>

It moves a bunch of header files included in hardware.h and itself
from mach-imx/include/mach to mach-imx, and updates users to include
hardware.h rather than mach/hardware.h.  The files in mach-imx/devices
will need to include "../hardware.h".
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NArnd Bergmann <arnd@arndb.de>

Rename mach-imx/include/mach/common.h to mach-imx/common.h and update
all users to include common.h rather than mach/common.h.

It also removes an unneeded inclusion to common.h in
mach-imx/devices/devices.c.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NArnd Bergmann <arnd@arndb.de>

It's really unnecessary to have plat-mxc, and let's merge it into
mach-imx.  It's pretty much just a bunch of file renaming and
Kconfig/Makefile merge.

To make the change less invasive, we keep using Kconfig symbol
CONFIG_ARCH_MXC for mach-imx sub-architecture.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NArnd Bergmann <arnd@arndb.de>

This prevents an out-of-bounds access to the clock_event_mode_label when
debugging.
Signed-off-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

As the timer code now does a clk_get to get its clock we don't
need the struct clk argument anymore.
This also changes the alternative EPIT timer to do a clk_get.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

Contrary to the ipg clock the perclk rate is not changed or
gated in low power mode, so we choose perclk for gpt.

With the port to the common clock framework as a side effect
the timer used the rate returned from the peripheral clock
but the hardware was still programmed to use the ipg clock,
so this patch only changes the hardware to really use the
clock it already assumed.
Signed-off-by: NRichard Zhao <richard.zhao@freescale.com>
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

We used to pass the timer clock directly to mxc_timer_init. We
should instead request the correct clock. This is an intermediate
step: For now we request the clock in the timer code when NULL
is passed as clock.
Also, the gpt on some i.MX have an additional ipg clock which can
be gated. Request and enable this.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

It's for migrating to generic clk framework API.
Signed-off-by: NRichard Zhao <richard.zhao@linaro.org>
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>