The rk3288 has the ability to invert the polarity of the PWM.  Let's
enable that ability.  Note that this increases pwm_cells to 3 for
rk3288.
Signed-off-by: NDoug Anderson <dianders@chromium.org>
Reviewed-by: NCaesar Wang <caesar.wang@rock-chips.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The i.MX PWM version2 is embedded in several i.MX SoCs, such as i.MX27,
i.MX51 and i.MX6SL.  There is a 4-word (16 bit) sample FIFO in this IP.
Each FIFO slot determines the duty period of a PWM waveform in one full
cycle.  The IP spec mentions that we should not write a fourth sample
because the FIFO will become full and triggers a FIFO write error (FWE)
which will prevent the PWM from starting once it is enabled.  In order
to avoid any sample FIFO overflow issue, this patch clears all sample
FIFO by doing software reset in the configuration hook when the
controller is disabled or waits for a full PWM cycle to get a
relinquished FIFO slot when the controller is enabled and the FIFO is
fully loaded.

The FIFO overflow issue can be reproduced by the following commands on
the i.MX6SL EVK platform, assuming we use PWM2 for the debug LED which
is driven by the pin HSIC_STROBE and the maximal brightness is 255.

	echo 0   > /sys/class/leds/user/brightness
	echo 0   > /sys/class/leds/user/brightness
	echo 0   > /sys/class/leds/user/brightness
	echo 0   > /sys/class/leds/user/brightness
	echo 255 > /sys/class/leds/user/brightness

Here, FWE happens (PWMSR register reads 0x58) and the LED can not be
lighten.

Another way to reproduce the FIFO overflow issue is to run this script:

	while true;
		do echo 255 > /sys/class/leds/user/brightness;
	done

Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Shawn Guo <shawn.guo@freescale.com>
Cc: Lothar Waßmann <LW@KARO-electronics.de>
Cc: linux-pwm@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: NLiu Ying <Ying.Liu@freescale.com>
Acked-by: NShawn Guo <shawn.guo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This patch contains no logic change to cleanup indentation for register
definitions only.

Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Shawn Guo <shawn.guo@freescale.com>
Cc: Lothar Waßmann <LW@KARO-electronics.de>
Cc: linux-pwm@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: NLiu Ying <Ying.Liu@freescale.com>
Acked-by: NShawn Guo <shawn.guo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This patch adds missing parentheses around the argument of the macro
MX3_PWMCR_PRESCALER(x) to avoid any potential macro expansion issue.
Reported-by: NLothar Waßmann <LW@KARO-electronics.de>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Shawn Guo <shawn.guo@freescale.com>
Cc: Lothar Waßmann <LW@KARO-electronics.de>
Cc: linux-pwm@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: NLiu Ying <Ying.Liu@freescale.com>
Acked-by: NShawn Guo <shawn.guo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Since we cannot make sure the 'chip->npwm' will always be none zero here,
and then if either equal to zero, the kzalloc() will return ZERO_SIZE_PTR,
which equals to ((void *)16).

So this patch fix this with just doing the zero check before calling kzalloc().
Signed-off-by: NXiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

There is no sign of this IP block on non-x86 architectures and rather will not
be. Thus, make this explicit by applying a direct dependency to X86.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Use c99 initializers for structures.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@decl@
identifier i1,fld;
type T;
field list[n] fs;
@@

struct i1 {
 fs
 T fld;
 ...};

@bad@
identifier decl.i1,i2;
expression e;
initializer list[decl.n] is;
@@

struct i1 i2 = { is,
+ .fld = e
- e
 ,...};
// </smpl>
Signed-off-by: NJulia Lawall <Julia.Lawall@lip6.fr>
[thierry.reding: rebased and applied same fix for Braswell]
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

An x86 build seems to pull in the linux/io.h include indirectly. On
PowerPC that doesn't happen and the build breaks due to the readl() and
writel() functions not being declared. Fix this by explicitly including
linux/io.h.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Let's use managed functions for this driver.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The driver consists of core, PCI, and platform parts. It would be better
to split them into separate files.

The platform driver is now called pwm-lpss-platform. Thus, previously
set CONFIG_PWM_LPSS=m is not enough to build it. But we are on the safe
side since it seems no one from outside Intel is using it for now.

While here, move to use macros module_pci_driver() and
module_platform_driver().
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: NAlan Cox <alan@linux.intel.com>
[thierry.reding: change select to depends on PWM_LPSS, cleanup]
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This is pretty much the same as Baytrail PWM. Only difference is that the
input clock runs on different frequency.
Signed-off-by: NAlan Cox <alan@linux.intel.com>
Signed-off-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Commit 42fa98a9 ("pwm: fsl-ftm: Convert to direct regmap API usage")
introduced the following error when REGMAP_MMIO=n:

drivers/built-in.o: In function `fsl_pwm_probe':
>> pwm-fsl-ftm.c:(.text+0xd7d7): undefined reference to `devm_regmap_init_mmio_clk'

Select select REGMAP_MMIO in order to fix this error.
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The regmap core supports different endian modes for devices. This patch
convert to direct regmap API usage, preparing to support big endianness
for LS1 SoC.

Using the regmap framework it will be easy to support devices that only
differ in endianness with the same device driver.
Signed-off-by: NXiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This patch intends to prepare for converting to direct regmap API usage.
Signed-off-by: NXiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

If pwm_get() finds a look-up entry with a perfect match (both dev_id and
con_id match), the loop is aborted, and "p" still points to the correct
struct pwm_lookup.

If only an entry with a matching dev_id or con_id is found, the loop
terminates after traversing the whole list, and "p" now points to
arbitrary memory, not part of the pwm_lookup list.
Then pwm_set_period() and pwm_set_polarity() will set random values for
period resp. polarity.

To fix this, save period and polarity when finding a new best match,
just like is done for chip (for the provider) and index.

This fixes the LCD backlight on r8a7740/armadillo-legacy, which was fed
period 0 and polarity -1068821144 instead of 33333 resp. 1.

Fixes: 3796ce1d ("pwm: add period and polarity to struct pwm_lookup")
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Cc: stable@vger.kernel.org
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This patch added to support the PWM controller found on
RK3288 SoC.
Signed-off-by: NCaesar Wang <caesar.wang@rock-chips.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Removes the PWM period table. Instead the prescaler is computed
from the period value passed in the config() function.
Signed-off-by: NAjit Pal Singh <ajitpal.singh@st.com>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

ST PWM IP has a common enable/disable control for all the PWM
channels on a PWM cell. Disables PWM output on the PWM HW only
when disable is called for the last channel.
Signed-off-by: NAjit Pal Singh <ajitpal.singh@st.com>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

ST PWM IP shares the same clock prescaler across all the PWM
channels. Hence configuration requests which change the period
will affect all the channels. Do not allow period changes which
will stomp period settings of the already configured channels.
Signed-off-by: NAjit Pal Singh <ajitpal.singh@st.com>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This patch fixes the pwm driver to write the complete 8 bits of
the prescaler value to the PWM Control register.
Signed-off-by: NAjit Pal Singh <ajitpal.singh@st.com>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This driver supports all current STi platforms' PWM IPs.
Signed-off-by: NAjit Pal Singh <ajitpal.singh@st.com>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
[thierry.reding: rename module to pwm-sti, fix build breakage]
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The .config() hook imx_pwm_config() calls clk APIs like clk_prepare()
and clk_get_rate(), which might sleep, so we need to set can_sleep flag
on pwm_chip.
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>
Acked-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Unlike other Intel LPSS devices, the PWM does not have the
clock dividers or the gate. All we get from the clock is the
rate. Since PCI case uses the driver data to get the rate,
we can drop the clk and use the same data also in case of
ACPI. The frequency is the same.
Signed-off-by: NHeikki Krogerus <heikki.krogerus@linux.intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NChew, Chiau Ee <chiau.ee.chew@intel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The site-specific OOM messages are unnecessary, because they
duplicate the MM subsystem generic OOM message. The following
checkpatch warning is also removed.

  WARNING: Possible unnecessary 'out of memory' message
Signed-off-by: NJingoo Han <jg1.han@samsung.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

This commit adds a driver for the PWM controller found on Rockchip
RK29, RK30 and RK31 SoCs.
Signed-off-by: NBeniamino Galvani <b.galvani@gmail.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The Atmel PWM IP can be found on avr32 chips. This allows selecting and building
the driver on avr32.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Signed-off-by: NNicolas Ferre <nicolas.ferre@atmel.com>

The implementation of .config(), .enable() and .disable() operations in this
driver may sleep, thus set pwm_chip can_sleep flag.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Acked-by: NXiubo Li <Li.Xiubo@freescale.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Current code only works when pdev->id is 1. Fix it by passing correct
bit values to abx500_mask_and_set_register_interruptible().

Having DISABLE_PWM/ENABLE_PWM does not make the code more readable
because the bit values depend on pdev->id. Thus drop the DISABLE_PWM
and ENABLE_PWM defines.

This patch also removes an unnecessary return in ab8500_pwm_disable().
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Acked-by: NAlexandre BOURDIOL <alexandre.bourdiol@st.com>
Acked-by: NPhilippe Begnic <philippe.begnic@st.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

pwm_samsung_config() sets the manual update bit via a call to the
pwm_samsung_enable() function even when the channel is already running.
This causes noticable flicker on display if we try to change the
backlight brightness from minimum to maximum, continuously.

So, we remove the call to pwm_samsung_enable() from pwm_samsung_config
to avoid the flicker and this change doesn't harm normal working since
the pwm-backlight driver already calls pwm_samsung_enable() where
needed.
Signed-off-by: NAjay Kumar <ajaykumar.rs@samsung.com>
Reviewed-by: NTomasz Figa <t.figa@samsung.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Read/write through I2C can sleep, thus set pwm_chip can_sleep flag.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

atmel_pwm_config() calls clk_get_rate() which might sleep, so we need to
set pwm_chip can_sleep flag.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The .config() calls clk_get_rate() which might sleep, so we need to set
pwm_chip can_sleep flag.  Otherwise, we see the following warning when
using PWM driven heartbeat led.

WARNING: CPU: 0 PID: 0 at kernel/locking/mutex.c:856 mutex_trylock+0x184/0x1a4()
DEBUG_LOCKS_WARN_ON(in_interrupt())
Modules linked in:
CPU: 0 PID: 0 Comm: swapper Not tainted 3.14.0-rc5 #18
[<c0015420>] (unwind_backtrace) from [<c0012cb0>] (show_stack+0x10/0x14)
[<c0012cb0>] (show_stack) from [<c001daf8>] (warn_slowpath_common+0x6c/0x8c)
[<c001daf8>] (warn_slowpath_common) from [<c001dbac>] (warn_slowpath_fmt+0x30/0x40)
[<c001dbac>] (warn_slowpath_fmt) from [<c045df74>] (mutex_trylock+0x184/0x1a4)
[<c045df74>] (mutex_trylock) from [<c0360950>] (clk_prepare_lock+0xc/0xec)
[<c0360950>] (clk_prepare_lock) from [<c0362020>] (clk_get_rate+0xc/0x68)
[<c0362020>] (clk_get_rate) from [<c028d07c>] (mxs_pwm_config+0x20/0x198)
[<c028d07c>] (mxs_pwm_config) from [<c028bde8>] (pwm_config+0x60/0x70)
[<c028bde8>] (pwm_config) from [<c034b61c>] (__led_pwm_set+0x1c/0x3c)
[<c034b61c>] (__led_pwm_set) from [<c034bc3c>] (led_heartbeat_function+0x70/0x110)
[<c034bc3c>] (led_heartbeat_function) from [<c00292f0>] (call_timer_fn+0x7c/0x164)
[<c00292f0>] (call_timer_fn) from [<c00295c8>] (run_timer_softirq+0x1f0/0x260)
[<c00295c8>] (run_timer_softirq) from [<c002255c>] (__do_softirq+0xc4/0x2f0)
[<c002255c>] (__do_softirq) from [<c0022890>] (irq_exit+0xa4/0x10c)
[<c0022890>] (irq_exit) from [<c0010240>] (handle_IRQ+0x34/0x84)
[<c0010240>] (handle_IRQ) from [<c0013524>] (__irq_svc+0x44/0x54)
[<c0013524>] (__irq_svc) from [<c00107f8>] (arch_cpu_idle+0x40/0x48)
[<c00107f8>] (arch_cpu_idle) from [<c005deb8>] (cpu_startup_entry+0x70/0x198)
[<c005deb8>] (cpu_startup_entry) from [<c060aac8>] (start_kernel+0x2a8/0x2f8)
Reported-by: NStefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: NShawn Guo <shawn.guo@freescale.com>
Acked-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Tested-by: NStefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

These elementary functions should be inlined for fastest access. Also
fixes this warning as a side-effect (when no PM_SLEEP is selected):

drivers/pwm/pwm-tiehrpwm.c:141:12: warning: 'ehrpwm_read' defined but not used [-Wunused-function]
Signed-off-by: NWolfram Sang <wsa@sang-engineering.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Fixes following warnings on AM335X with no PM_SLEEP

drivers/pwm/pwm-tiehrpwm.c:534:13: warning: 'ehrpwm_pwm_save_context' defined but not used [-Wunused-function]
drivers/pwm/pwm-tiehrpwm.c:548:13: warning: 'ehrpwm_pwm_restore_context' defined but not used [-Wunused-function]
Signed-off-by: NWolfram Sang <wsa@sang-engineering.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

The struct is not used anymore and the polarity initialization will be
done using the PWM lookup table (or device tree).
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: NSimon Horman <horms+renesas@verge.net.au>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Add period and polarity members to struct pwm_lookup so that platforms
using the lookup table can be treated the same way as those using the
device tree.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Current twl6030_pwm_disable() implementation writes TWL6030_TOGGLE3_REG
twice, the second write sets TWL6030_PWMXEN bits so the PWM clock does
not disable.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Acked-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Fixes the following warnings reported by the 0-DAY kernel build testing
backend:

   drivers/pwm/pwm-lpss.c: In function 'pwm_lpss_probe_pci':
>> drivers/pwm/pwm-lpss.c:192:2: warning: passing argument 3 of 'pwm_lpss_probe' discards 'const' qualifier from pointer target type [enabled by default]
     lpwm = pwm_lpss_probe(&pdev->dev, &pdev->resource[0], info);
     ^
   drivers/pwm/pwm-lpss.c:130:30: note: expected 'struct pwm_lpss_boardinfo *' but argument is of type 'const struct pwm_lpss_boardinfo *'
    static struct pwm_lpss_chip *pwm_lpss_probe(struct device *dev,
                                 ^
>> drivers/pwm/pwm-lpss.c:143:28: sparse: incorrect type in return expression (different address spaces)
   drivers/pwm/pwm-lpss.c:143:28:    expected struct pwm_lpss_chip *
   drivers/pwm/pwm-lpss.c:143:28:    got void [noderef] <asn:2>*regs
>> drivers/pwm/pwm-lpss.c:192:63: sparse: incorrect type in argument 3 (different modifiers)
   drivers/pwm/pwm-lpss.c:192:63:    expected struct pwm_lpss_boardinfo *info
   drivers/pwm/pwm-lpss.c:192:63:    got struct pwm_lpss_boardinfo const *[assigned] info
   drivers/pwm/pwm-lpss.c: In function 'pwm_lpss_probe_pci':
   drivers/pwm/pwm-lpss.c:192:2: warning: passing argument 3 of 'pwm_lpss_probe' discards 'const' qualifier from pointer target type [enabled by default]
     lpwm = pwm_lpss_probe(&pdev->dev, &pdev->resource[0], info);
     ^
   drivers/pwm/pwm-lpss.c:130:30: note: expected 'struct pwm_lpss_boardinfo *' but argument is of type 'const struct pwm_lpss_boardinfo *'
    static struct pwm_lpss_chip *pwm_lpss_probe(struct device *dev,
                                 ^
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

pwmchip_add() returns zero on success and a negative value on error,
so the condition of the check must be inverted.
Signed-off-by: NBeniamino Galvani <b.galvani@gmail.com>
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>

Not all systems enumerate the PWM devices via ACPI. They can also be
exposed via the PCI interface.
Signed-off-by: NAlan Cox <alan@linux.intel.com>
Signed-off-by: NChew, Chiau Ee <chiau.ee.chew@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NThierry Reding <thierry.reding@gmail.com>