dma_pte_free_level() has an off-by-one error when checking whether a pte
is completely covered by a range.  Take for example the case of
attempting to free pfn 0x0 - 0x1ff, ie.  512 entries covering the first
2M superpage.

The level_size() is 0x200 and we test:

  static void dma_pte_free_level(...
	...

	if (!(0 > 0 || 0x1ff < 0 + 0x200)) {
		...
	}

Clearly the 2nd test is true, which means we fail to take the branch to
clear and free the pagetable entry.  As a result, we're leaking
pagetables and failing to install new pages over the range.

This was found with a PCI device assigned to a QEMU guest using vfio-pci
without a VGA device present.  The first 1M of guest address space is
mapped with various combinations of 4K pages, but eventually the range
is entirely freed and replaced with a 2M contiguous mapping.
intel-iommu errors out with something like:

  ERROR: DMA PTE for vPFN 0x0 already set (to 5c2b8003 not 849c00083)

In this case 5c2b8003 is the pointer to the previous leaf page that was
neither freed nor cleared and 849c00083 is the superpage entry that
we're trying to replace it with.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

We usually rely on the fact that struct members not specified in the
initializer are set to NULL.  So do that with fsnotify function pointers
as well.
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Eric Paris <eparis@parisplace.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

After removing event structure creation from the generic layer there is
no reason for separate .should_send_event and .handle_event callbacks.
So just remove the first one.
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Eric Paris <eparis@parisplace.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Currently fsnotify framework creates one event structure for each
notification event and links this event into all interested notification
groups.  This is done so that we save memory when several notification
groups are interested in the event.  However the need for event
structure shared between inotify & fanotify bloats the event structure
so the result is often higher memory consumption.

Another problem is that fsnotify framework keeps path references with
outstanding events so that fanotify can return open file descriptors
with its events.  This has the undesirable effect that filesystem cannot
be unmounted while there are outstanding events - a regression for
inotify compared to a situation before it was converted to fsnotify
framework.  For fanotify this problem is hard to avoid and users of
fanotify should kind of expect this behavior when they ask for file
descriptors from notified files.

This patch changes fsnotify and its users to create separate event
structure for each group.  This allows for much simpler code (~400 lines
removed by this patch) and also smaller event structures.  For example
on 64-bit system original struct fsnotify_event consumes 120 bytes, plus
additional space for file name, additional 24 bytes for second and each
subsequent group linking the event, and additional 32 bytes for each
inotify group for private data.  After the conversion inotify event
consumes 48 bytes plus space for file name which is considerably less
memory unless file names are long and there are several groups
interested in the events (both of which are uncommon).  Fanotify event
fits in 56 bytes after the conversion (fanotify doesn't care about file
names so its events don't have to have it allocated).  A win unless
there are four or more fanotify groups interested in the event.

The conversion also solves the problem with unmount when only inotify is
used as we don't have to grab path references for inotify events.

[hughd@google.com: fanotify: fix corruption preventing startup]
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Eric Paris <eparis@parisplace.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Rounding of name length when passing it to userspace was done in several
places.  Provide a function to do it and use it in all places.
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Eric Paris <eparis@parisplace.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Record actively mapped pages and provide an api for asserting a given
page is dma inactive before execution proceeds.  Placing
debug_dma_assert_idle() in cow_user_page() flagged the violation of the
dma-api in the NET_DMA implementation (see commit 77873803 "net_dma:
mark broken").

The implementation includes the capability to count, in a limited way,
repeat mappings of the same page that occur without an intervening
unmap.  This 'overlap' counter is limited to the few bits of tag space
in a radix tree.  This mechanism is added to mitigate false negative
cases where, for example, a page is dma mapped twice and
debug_dma_assert_idle() is called after the page is un-mapped once.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Vinod Koul <vinod.koul@intel.com>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: James Bottomley <JBottomley@Parallels.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull battery updates from Dmitry Eremin-Solenikov:
 "I'm picking up power supply maintainership from Anton Vorontov.  Could
  you please pull battery-2.6 git tree changes prepared for the v3.14
  release.

  Highlights:

   - Power supply notifier

   - Several drivers gained DT support

   - Added Maxim 14577 driver

   - Change of maintainer"

* tag 'for-v3.14' of git://git.infradead.org/battery-2.6:
  MAINTAINERS: Pick up power supply maintainership
  max17042_battery: Add IRQF_ONESHOT flag to use default irq handler
  gpio-charger: Support wakeup events
  power_supply: Add charger support for Maxim 14577
  dt: Binding documentation for isp1704 charger
  isp1704_charger: Add DT support
  charger-manager: of_cm_parse_desc() should be static
  bq2415x_charger: Add DT support
  power_supply: Add power_supply_get_by_phandle
  bq2415x_charger: Use power_supply notifier for automode
  power: reset: Add as3722 power-off driver
  mfd: AS3722: Add dt node properties for system power controller
  charger-manager: Support deivce tree in charger manager driver
  charger-manager: Modify the way of checking battery's temperature
  power_supply: Add power_supply notifier

Pull MFD changes from Lee Jones:
 "New drivers
   - Samsung Maxim 14577; Micro USB, Regulator, IRQ Controller and
     Battery Charger
   - TI/National Semiconductor LP3943 I2C GPIO Expander and PWM
     Generator

  Existing driver adaptions
   - Expansion of Wolfson Arizona DSP and High-Pass filter controls
   - TI TWL6040 default Regmap support and Regcache addition/bypass
   - Some nice Smatch catch fixes
   - Conversion of TI OMAP-USB and TI TWL6030 to endian neutralness
   - ChromeOS EC timing (delay) adaptions and added dependency on OF
   - Many constifications of 'struct {mfd_cell,regmap_irq,et.al}'
   - Watchdog support added for NVIDIA AS3722
   - Convert functions to static in TI AM335x
   - Realigned previously defeated functionality in TI AM335x
   - IIO ADC-TSC concurrency dead-lock/timeout resolution
   - Addition of Power Management and Clock support for Samsung core
   - DEFINE_PCI_DEVICE_TABLE macro removal from MFD Subsystem
   - Greater use of irqdomain functionality in ST-E AB8500
   - Removal of 'include/linux/mfd/abx500/ab8500-gpio.h'
   - Wolfson WM831x PMIC Power Management changes s/poweroff/shutdown/
   - Device Tree documentation added for TI/Nat Semi LP3943
   - Version detection and voltage tables for TI TPS6586x PMIC devices
   - Simplification of Freescale MC13XXX (de-)initialisation routines
   - Clean-up and simplification of the Realtek parent driver
   - Added support for RTL8402 Realtek PCI-Express card reader
   - Resource leak fix for Maxim 77686
   - Possible suspend BUG() fix in OMAP USB TLL
   - Support for new Wolfson WM5110 Revision (D)
   - Testing of automatic assignment of of_node in mfd_add_device()
   - Reversion of the above when it started to cause issues
   - Remove legacy Platform Data from;
              TI TWL Core, Qualcomm SSBI and ST-E ABx500 Pinctrl
   - Clean-ups; tabbing issues, function name changes, 'drvdata = NULL'
              removal, unused uninitialised warning mitigation, error
              message clarity, removal of redundant/duplicate checks,
              licensing (GPL -> GPL2), coding consistency, duplicate
              function declaration, ret checks, commit corrections,
              redundant of_match_ptr() helper removal, spelling,
              #if-deffery removal and header guards name changes"

* tag 'mfd-3.14-1' of git://git.linaro.org/people/ljones/mfd: (78 commits)
  mfd: wm5110: Add register patch for rev D chip
  mfd: omap-usb-tll: Don't hold lock during pm_runtime_get/put_sync()
  gpio: lp3943: Remove redundant of_match_ptr helper
  mfd: sta2x11-mfd: Use named constants for pci_power_t values
  Documentation: mfd: Fix LDO index in s2mps11.txt
  mfd: Cleanup mfd-mcp-sa11x0.h header
  mfd: max8997: Use "IS_ENABLED(CONFIG_OF)" for DT code.
  mfd: twl6030: Fix endianness problem in IRQ handler
  mfd: sec-core: Add cells for S5M8767-clocks
  mfd: max14577: Remove redundant of_match_ptr helper
  mfd: twl6040: Fix sparse non static symbol warning
  mfd: Revert "mfd: Always assign of_node in mfd_add_device()"
  mfd: rtsx: Fix sparse non static symbol warning
  mfd: max77693: Set proper maximum register for MUIC regmap
  mfd: max77686: Fix regmap resource leak on driver remove
  mfd: Represent correct filenames in file headers
  mfd: rtsx: Add support for card reader rtl8402
  mfd: rtsx: Add set pull control macro and simplify rtl8411
  mfd: max8997: Enforce mfd_add_devices() return value check
  mfd: mc13xxx: Simplify probe() & remove()
  ...

Pull sound updates from Takashi Iwai:
 "It was holiday season, so no wonder that there are little changes in
  framework level, although diffstat shows quite many changes spreaded
  over sound/* directories.  Most of changes are cleanups, code
  refactoring and fixes.

  Some highlights:
   - Removal of OSS sleep_on usages by Arnd
   - Simplified memalloc helper codes, drop obsoleted features; now it's
     built into PCM driver instead of an individual module
   - Warn if PCM buffer preallocation fails, which will show page
     allocation issues more clearly
   - Compress offload API updates for sample rates by Vinod
   - PCM glitch workaround on ctxfi emu20k1 by Sarah
   - Drop cs46xx DSP blobs, using firmware loader now
   - USB-audio quitks for Plantronics Gamecom 780, Creative VF0420, and
     Focusrite Saffire 6

  HD-audio specifics:
   - Standardize Kconfigs of HD-audio codec drivers; now "make
     localmodconfig" recognizes configs properly (finally!)
   - Parallel PM implementation by Mengdong
   - BayleyBay/ValleyView2 board fixups
   - Broadwell audio support
   - Runtime PM improvement (PantherPoint, etc)
   - Quirks: Dell subwooer, Gigabyte mobo jack detection oddity, Dell
     AiO click noise fixes, Dell headset mic fixes, etc
   - Automatic bind with HDMI codec parser without generic parser
   - More AD codec fixes (since 3.12 regression) including the automatic
     stereo mix support
   - Common Thinkpad ACPI helper for Realtek and Conexant codecs

  ASoC specifics:
   - Update to the generic DMA code to support deferred probe and
     managed resources
   - New drivers for BCM2835 (used in Raspberry Pi), Tegra with MAX98090
     and Analog Devices AXI I2S and S/PDIF controller IPs
   - Device tree support for the simple card, max98090 and cs42l52
   - Conversion of the Samsung drivers to native dmaengine, making them
     multiplatform compatible and hopefully helping keep them more
     modern and up to date.
   - More regmap conversions, including a very welcome one for twl6040
     from Peter Ujfalusi
   - A big overhaul of the DaVinci drivers also from Peter Ujfalusi
   - Lots of DMA updates from Lars-Peter
   - Improvements to the constraints handling code from Lars-Peter
   - A very helpful conversion of the TWL4030 driver to regmap from Peter
   - A new driver for the Freescale ESAI controller from Nicolin Chen
   - Conversion of some of the drivers to use params_width()
   - Extensions to DPCM for use with compressed audio from Liam"

* tag 'sound-3.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (396 commits)
  ASoC: dapm: Fix double prefix addition
  ASoC: compress: Add suport for DPCM into compressed audio
  ASoC: DPCM: make some DPCM API calls non static for compressed usage
  ASoC: core: Fix possible NULL pointer dereference of pcm->config
  ALSA: hda - add headset mic detect quirks for some Dell machines
  ASoC: tlv320aic32x4: Fix regmap range_min
  ASoC: core: Return -ENOTSUPP from set_sysclk() if no operation provided
  ASoC: dapm: Change prototype of soc_widget_read
  ASoC: samsung: Remove SND_DMAENGINE_PCM_FLAG_NO_RESIDUE flag
  ASoC: axi-{spdif,i2s}: Remove SND_DMAENGINE_PCM_FLAG_NO_RESIDUE flag
  ASoC: generic-dmaengine-pcm: Check DMA residue granularity
  ASoC: generic-dmaengine-pcm: Check NO_RESIDUE flag at runtime
  dma: pl330: Set residue_granularity
  dma: Indicate residue granularity in dma_slave_caps
  ASoC: simple-card: fix one bug to writing to the platform data
  ASoC: pcm: Use snd_pcm_rate_mask_intersect() helper
  ALSA: Add helper function for intersecting two rate masks
  ASoC: s6000: Don't mix SNDRV_PCM_RATE_CONTINUOUS with specific rates
  ASoC: fsl: Don't mix SNDRV_PCM_RATE_CONTINUOUS with specific rates
  ASoC: pcm: Properly initialize hw->rate_max
  ...

Pull bulk pin control changes from Linus Walleij:
 "This has been queued and tested for a while.  Lots of action here,
  like in the GPIO tree, embedded stuff like this is really hot now it
  seems.  Details in the signed tag.  I'm especially happy about the
  Qualcomm driver as it is used in such a huge subset of mobile handsets
  out there, and these platforms in general need better upstream support

   - New driver for the Qualcomm TLMM pin controller and its msm8x74
     subdriver.

   - New driver for the Broadcom Capri BCM281xx SoC.

   - New subdriver for the imx25 pin controller.

   - New subdriver for the Tegra124 pin controller.

   - Lock GPIO lines as IRQs for select combined pin control and GPIO
     drivers for baytrail and sirf.

   - Some semi-big refactorings and extenstions to the sirf driver.

   - Lots of patching, cleanup and fixing in the Renesas "PFC" driver
     and associated subdrivers as usual.  It is settling down a little
     bit now it seems.

   - Minor fixes and incremental updates here and there as usual"

* tag 'pinctrl-v3.14-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl: (72 commits)
  pinctrl: sunxi: Honor GPIO output initial vaules
  pinctrl: capri: add dependency on OF
  ARM: bcm11351: Enable pinctrl for Broadcom Capri SoCs
  ARM: pinctrl: Add Broadcom Capri pinctrl driver
  pinctrl: Add pinctrl binding for Broadcom Capri SoCs
  pinctrl: Add void * to pinctrl_pin_desc
  pinctrl: st: Fix a typo in probe
  pinctrl: Fix some typos and grammar issues in the documentation
  pinctrl: sirf: lock IRQs when starting them
  pinctrl: sirf: put gpio interrupt pin into input status automatically
  pinctrl: sirf: use only one irq_domain for the whole device node
  pinctrl: single: fix infinite loop caused by bad mask
  pinctrl: single: fix pcs_disable with bits_per_mux
  pinctrl: single: fix DT bindings documentation
  pinctrl: as3722: Set pin to output mode for some function
  pinctrl: sirf: add pin group for USP0 with only RX or TX frame sync
  pinctrl: sirf: fix the pins of sdmmc5 connected with TriG
  pinctrl: sirf: add lost usp1_uart_nostreamctrl group for atlas6
  pinctrl: sunxi: Add Allwinner A20 clock output pin functions
  pinctrl/lantiq: fix typo
  ...

Pull GPIO tree bulk changes from Linus Walleij:
 "A big set this merge window, as we have much going on in this
  subsystem.  The changes to other subsystems (notably a slew of ARM
  machines as I am doing away with their custom APIs) have all been
  ACKed to the extent possible.

  Major changes this time:

   - Some core improvements and cleanups to the new GPIO descriptor API.
     This seems to be working now so we can start the exodus to this
     API, moving gradually away from the global GPIO numberspace.

   - Incremental improvements to the ACPI GPIO core, and move the few
     GPIO ACPI clients we have to the GPIO descriptor API right *now*
     before we go any further.  We actually managed to contain this
     *before* we started to litter the kernel with yet another hackish
     global numberspace for the ACPI GPIOs, which is a big win.

   - The RFkill GPIO driver and all platforms using it have been
     migrated to use the GPIO descriptors rather than fixed number
     assignments.  Tegra machine has been migrated as part of this.

   - New drivers for MOXA ART, Xtensa GPIO32 and SMSC SCH311x.  Those
     should be really good examples of how I expect a nice GPIO driver
     to look these days.

   - Do away with custom GPIO implementations on a major part of the ARM
     machines: ks8695, lpc32xx, mv78xx0.  Make a first step towards the
     same in the horribly convoluted Samsung S3C include forest.  We
     expect to continue to clean this up as we move forward.

   - Flag GPIO lines used for IRQ on adnp, bcm-kona, em, intel-mid and
     lynxpoint.

     This makes the GPIOlib core aware that a certain GPIO line is used
     for IRQs and can then enforce some semantics such as disallowing a
     GPIO line marked as in use for IRQ to be switched to output mode.

   - Drop all use of irq_set_chip_and_handler_name().  The name provided
     in these cases were just unhelpful tags like "mux" or "demux".

   - Extend the MCP23s08 driver to handle interrupts.

   - Minor incremental improvements for rcar, lynxpoint, em 74x164 and
     msm drivers.

   - Some non-urgent bug fixes here and there, duplicate #includes and
     that usual kind of cleanups"

Fix up broken Kconfig file manually to make this all compile.

* tag 'gpio-v3.14-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio: (71 commits)
  gpio: mcp23s08: fix casting caused build warning
  gpio: mcp23s08: depend on OF_GPIO
  gpio: mcp23s08: Add irq functionality for i2c chips
  ARM: S5P[v210|c100|64x0]: Fix build error
  gpio: pxa: clamp gpio get value to [0,1]
  ARM: s3c24xx: explicit dependency on <plat/gpio-cfg.h>
  ARM: S3C[24|64]xx: move includes back under <mach/> scope
  Documentation / ACPI: update to GPIO descriptor API
  gpio / ACPI: get rid of acpi_gpio.h
  gpio / ACPI: register to ACPI events automatically
  mmc: sdhci-acpi: convert to use GPIO descriptor API
  ARM: s3c24xx: fix build error
  gpio: f7188x: set can_sleep attribute
  gpio: samsung: Update documentation
  gpio: samsung: Remove hardware.h inclusion
  gpio: xtensa: depend on HAVE_XTENSA_GPIO32
  gpio: clps711x: Enable driver compilation with COMPILE_TEST
  gpio: clps711x: Use of_match_ptr()
  net: rfkill: gpio: convert to descriptor-based GPIO interface
  leds: s3c24xx: Fix build failure
  ...

Pull i2c updates from Wolfram Sang:
 "For 3.14, the I2C subsystem has the following to offer:

   - new drivers for Renesas RIIC and RobotFuzz OSIF
   - driver cleanups & improvements & bugfixes

  Pretty standard stuff this time, I'd say.  There is more complex stuff
  coming up, but I didn't have the bandwidth between the years to pull
  it in for this release.  Sadly"

* 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux: (26 commits)
  i2c: s3c2410: fix quirk usage for 64-bit
  i2c: pnx: Use devm_*() functions
  i2c: at91: add a new compatibility string for the at91sam9261
  i2c-ismt: support I2C_SMBUS_I2C_BLOCK_DATA transaction type
  i2c: Add bus driver for for OSIF USB i2c device.
  i2c: i2c-tiny-usb: Remove RobotFuzz USB vendor:product ID
  i2c: designware: remove HAVE_CLK build dependecy
  Documentation: i2c: Remove obsolete example
  i2c: nomadik: remove platform data header
  i2c: nomadik: auto-calculate slave setup time
  i2c: viperboard: remove superfluous assignment
  i2c: xilinx: Use devm_* functions
  i2c: xilinx: Do not enable irq before irq handler
  i2c: xilinx: Fix i2c checkpatch warnings
  i2c: at91: document clock properties
  i2c: isch: Use devm_request_region()
  i2c: viperboard: Use devm_kzalloc() functions
  i2c: imx: propagate irq error code in probe
  i2c: s3c2410: dont need CPU_FREQ transitions for exynos series
  i2c: s3c2410: Add polling mode support
  ...

Pull security layer updates from James Morris:
 "Changes for this kernel include maintenance updates for Smack, SELinux
  (and several networking fixes), IMA and TPM"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (39 commits)
  SELinux: Fix memory leak upon loading policy
  tpm/tpm-sysfs: active_show() can be static
  tpm: tpm_tis: Fix compile problems with CONFIG_PM_SLEEP/CONFIG_PNP
  tpm: Make tpm-dev allocate a per-file structure
  tpm: Use the ops structure instead of a copy in tpm_vendor_specific
  tpm: Create a tpm_class_ops structure and use it in the drivers
  tpm: Pull all driver sysfs code into tpm-sysfs.c
  tpm: Move sysfs functions from tpm-interface to tpm-sysfs
  tpm: Pull everything related to /dev/tpmX into tpm-dev.c
  char: tpm: nuvoton: remove unused variable
  tpm: MAINTAINERS: Cleanup TPM Maintainers file
  tpm/tpm_i2c_atmel: fix coccinelle warnings
  tpm/tpm_ibmvtpm: fix unreachable code warning (smatch warning)
  tpm/tpm_i2c_stm_st33: Check return code of get_burstcount
  tpm/tpm_ppi: Check return value of acpi_get_name
  tpm/tpm_ppi: Do not compare strcmp(a,b) == -1
  ima: remove unneeded size_limit argument from ima_eventdigest_init_common()
  ima: update IMA-templates.txt documentation
  ima: pass HASH_ALGO__LAST as hash algo in ima_eventdigest_init()
  ima: change the default hash algorithm to SHA1 in ima_eventdigest_ng_init()
  ...

Evaluation of revision D of WM5110 suggests updates to the register
patch for optimal performance. For the sake of clarity rev C of the chip
does not require a register patch.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

pm_runtime_get/put_sync() can sleep so don't hold spinlock while
calling them.

This patch prevents a BUG() during system suspend when
CONFIG_DEBUG_ATOMIC_SLEEP is enabled.

Bug is present in Kernel versions v3.9 onwards.
Reported-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Roger Quadros <rogerq@ti.com>
Tested-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

'lp3943_gpio_of_match' is always compiled in. Hence the
helper macro is not needed.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

If nothing more than to improve code readability.

The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@
expression pdev;
@@

pci_set_power_state(pdev,
- 0
+ PCI_D0
 )
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

LDO indices start from 1. Fix the example appropriately.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

Commit a1fd844c ("ARM: sa1100: move platform_data definitions")
moved the file to the current location but forgot to remove the pointer
to its previous location. Clean it up. While at it also change the header
file protection macros appropriately.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

Instead of "#if define CONFIG_OF" use "IS_ENABLED(CONFIG_OF)"
option for DT code to avoid if-deffery in code.
Signed-off-by: Manish Badarkhe <badarkhe.manish@gmail.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The current TWL 6030 IRQ handler assumes little endianness.
This change makes it endian-neutral.
Signed-off-by: Danke Xie <d.xie@sta.samsung.com>
Signed-off-by: Taras Kondratiuk <taras.kondratiuk@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

S5M8767 chip has 3 crystal oscillators running at 32KHz. These are
supported by s2mps11-clk driver.
Signed-off-by: Tushar Behera <tushar.behera@linaro.org>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

'max14577_dt_match' is always compiled in. Hence the helper macro
is not needed.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

Fixes the following sparse warning:

drivers/mfd/twl6040.c:89:20: warning:
 symbol 'twl6040_patch' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

This reverts commit 68044bee13770918e0b28dd44aa98c889ec7558f.

We've had confirmed reports of this patch causing unforeseen issues
with existing MFD users. It has been agreed by the original author
and myself that reversion is the best solution.
Acked-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

Fixes the following sparse warning:

drivers/mfd/rtl8411.c:473:6: warning:
 symbol 'rtl8411_init_common_params' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The MUIC block in max77693 has different I2C address than PMIC. The
driver allocated two regmaps: for PMIC and MUIC. However it used the
same regmap_config (with max_register field) for both regmaps. Actual
maximum address of register for MUIC is different than for PMIC.

Define another regmap_config for MUIC with proper max_register value.
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The regmap used by max77686 MFD driver was not freed with regmap_exit()
on driver exit. This lead to leak of resources.

Replace regmap_init_i2c() call in driver probe with initialization of
managed register map so the regmap will be properly freed by the device
management code.

Cc: stable@vger.kernel.org
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The original author(s) probably copy/pasted these headers from the
existing public header files.
Signed-off-by: Laszlo Papp <lpapp@kde.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

rtl8402 is much like rtl8411, so just add it to rtl8411.c
Signed-off-by: Micky Ching <micky_ching@realsil.com.cn>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

Add set pull control macro to reduce code for setting pull control, and
use a common init function to reduce code for rtl8411.c. So this patch
is used to just simplify code.
Signed-off-by: Micky Ching <micky_ching@realsil.com.cn>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The original author provided a random return value check which is
redundant and seemingly floating. This patch not only relocates
the check so it is more clearly associated with the invokation of
mfd_add_devices(), but provides a store for the error value. We
also print a meaningful message on error before returning.
Signed-off-by: Laszlo Papp <lpapp@kde.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

This patch simplifies probe() and remove() functions by moving
some initialisation code out from the I2C/SPI init() and exit()
functions and into the core driver.
Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

Symbol MFD_MC13783 always selected by MFD_MC13XXX, so no need
to keep additional symbol.
Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

mc13xxx_get_flags() declaration given twice.
This patch removes this duplicate.
Signed-off-by: Alexander Shiyan <shc_work@mail.ru>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

This is a read-only data structure.
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

In preparation for passing a const pointer directly to
ssbi_write() from the regmap APIs.
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The ssbi driver assumes that the device is DT based. Remove the
platform data structs that will never be used and hide the enum
in the only C file that uses it.
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

mfd_add_device() assigns .of_node in the device objects it creates only
if the mfd_cell for the device has the .of_compatible field set and the
DT node for the top-level MFD device contains a child whose compatible
property matches the cell's .of_compatible field.

This leaves .of_node unset in many cases. When this happens, entries in
the DT /aliases property which refer to the top-level MFD DT node will
never match the MFD child devices, hence causing the requested alias not
to be honored.

Solve this by setting each MFD child device's .of_node equal to the top-
level MFD device's .of_node field in the cases where it would otherwise
remain unset.

The first use-case for this will be aliases for the TPS6586x's RTC
device.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>

The EC has specific timing it requires. Add support for an optional delay
after raising CS to fix timing issues. This is configurable based on
a DT property "google,cros-ec-spi-msg-delay".

If this property isn't set, then no delay will be added. However, if set
it will cause a delay equal to the value passed to it to be inserted at
the end of a transaction.
Signed-off-by: Rhyland Klein <rklein@nvidia.com>
Reviewed-by: Bernie Thompson <bhthompson@chromium.org>
Reviewed-by: Andrew Bresticker <abrestic@chromium.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>