Merge "Samsung Device Tree updates and improvements for v4.7" from Krzysztof Kozlowski:

1. Enable accelerated AES (Security SubSystem) on Exynos4412-based boards.
2. Enable HDMI CEC on Exynos4412-based Odroid.
3. Add regulator supplies for eMMC/SD on Odroid XU3/XU4.
4. Fix DTC unit name warnings.
5. Merge topic branch of new Artik5 board.

* tag 'samsung-dt-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux:
  ARM: dts: s5p: Fix DTC unit name warnings in SMDKv210 board
  ARM: dts: exynos: Fix DTC unit name warnings in Exynos5440
  ARM: dts: exynos: Fix DTC unit name warnings in SMDK5420
  ARM: dts: exynos: Fix DTC unit name warnings in Peach Pit
  ARM: dts: exynos: Fix DTC unit name warnings in Exynos542x
  ARM: dts: exynos: Fix DTC unit name warnings in Exynos5250
  ARM: dts: exynos: Fix DTC unit name warnings in Exynos4x12
  ARM: dts: exynos: Fix DTC unit name warnings in Trats2 board
  ARM: dts: exynos: Fix DTC unit name warnings in Exynos4
  ARM: dts: exynos: Fix DTC unit name warnings in Exynos3250
  ARM: dts: exynos: Fix DTC unit name warnings in cros-adc-thermistors
  ARM: dts: exynos: Add eMMC and SD regulator supplies to Odroid XU3/XU4
  ARM: dts: exynos: Enable the HDMI CEC device on Exynos4412 Odroid boards
  ARM: dts: exynos: Add node for the HDMI CEC device to exynos4
  ARM: dts: exynos: Add HDMI CEC pin definition to exynos4 pinctrl
  ARM: dts: exynos: Enable SSS on Odroid X/X2/U3 family
  ARM: dts: exynos: Enable SSS on Trats2
  ARM: dts: exynos: Add Security SubSystem node to Exynos4

Merge tag 'samsung-dt-exynos3250-artik5-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux into next/dt

Merge "Topic branch for Device Tree changes for Exynos 3250 for v4.7" from Krzysztof Kozlowski:

Merge necessary new clocks from Sylwester (used by new board) and add support
for Exynos3250-based Artik5 board.

* tag 'samsung-dt-exynos3250-artik5-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux:
  ARM: dts: exynos: Add MSHC2 DT node for SD card for exynos3250-artik5-eval board
  ARM: dts: exynos: Add exynos3250-artik5 dtsi file for ARTIK5 module
  ARM: dts: exynos: Add MSHC2 DT node for Exynos3250 SoC
  ARM: dts: exynos: Add UART2 DT node for Exynos3250 SoC
  ARM: dts: exynos: Add initial gpio setting of MMC2 device for exynos3250-monk
  ARM: dts: exynos: Add initial pin configuration for exynos3250-rinato
  clk: samsung: exynos3250: Add MMC2 clock
  clk: samsung: exynos3250: Add UART2 clock
  dt-bindings: Add the clock id of UART2 and MMC2 for Exynos3250

Merge tag 'v4.7-rockchip-dts32-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into next/dt

A lot display-controller nodes for DSI and the Analogix DP on rk3288
as well as general display+hdmi support on rk3036. With the Analogix
DP support, Veyron Chromeboks can now finally use their internal
display.

Other than this big improvement we have thermal support on the rk3228,
a long time missing binding document for the General Register Files
block, better operating points for Veyron devices and a bunch of fixes
with parts stemming from warnings that new dtc version can generate.

* tag 'v4.7-rockchip-dts32-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip

: (27 commits)
  ARM: dts: rockchip: move rk3036 memory definition to board files
  ARM: dts: rockchip: enable the eDP on rk3288 veyron devices
  ARM: dts: rockchip: simple panel and backlight supplies on veyron boards
  ARM: dts: rockchip: override edp hpd handling on veyron-pinky and speedy
  ARM: dts: rockchip: add rk3288-veyron-minnie backlight and panel settings
  ARM: dts: rockchip: add rk3288-veyron-jaq backlight and panel overrides
  ARM: dts: rockchip: add core rk3288-veyron backlight and panel nodes
  ARM: dts: rockchip: add startup delay to rk3288-veyron panel-regulators
  ARM: dts: rockchip: move edp-hpd pin definition into common location
  ARM: dts: rockchip: add rk3288 displayport controller node
  ARM: dts: rockchip: add rk3288 edp-phy node
  ARM: dts: rockchip: add missing unitname to cpu_leakage efuse
  ARM: dts: rockchip: drop unneeded properties from mipi node
  ARM: dts: rockchip: clean up gpio-keys nodes
  ARM: dts: rockchip: fix missing usbphy unit-names
  ARM: dts: rockchip: fix rk3288 power-domain unit names
  ARM: dts: rockchip: update rk3288-veyron cpu operating points
  ARM: dts: rockchip: remove broken-cd from emmc and sdio
  ARM: dts: rockchip: enable the tsadc for rk3228 evb
  ARM: dts: rockchip: add the thermal main info found on rk3228
  ...

Signed-off-by: Olof Johansson <olof@lixom.net>

Merge tag 'socfpga_dts_for_v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/dinguyen/linux into next/dt

SoCFPGA DTS updates for v4.7
- Update SD/MMC node for Arria10
- Update Arria10 with clock and interrupt fields for DMA
- Remove 'phy-addr' from stmmac node
- Remove ethernet node from Cyclone5 DTSI
- Add LEDs/KEYs/SWs support on Sockit
- Add L2 and OCRAM EDAC dts entries
- Add reset control for USB

* tag 'socfpga_dts_for_v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/dinguyen/linux

:
  ARM: dts: socfpga: add reset control for USB
  ARM: dts: socfpga: Add Altera Arria10 OCRAM EDAC devicetree entry
  ARM: dts: socfpga: Add Altera Arria10 L2 Cache EDAC devicetree entry
  ARM: dts: socfpga: Add support for HPS KEYs/SWs on SoCKit
  ARM: dts: socfpga: Add support for HPS LEDs on SoCKit
  ARM: dts: socfpga: Drop gmac0 from CV dtsi
  ARM: dts: socfpga: Drop phy-addr OF property from CV dtsi
  ARM: dts: socfpga: Add missing clock and interrupt fields for Arria10 DMA
  ARM: dts: socfpga: add the clk-phase property for sd/mmc clock
  ARM: dts: socfpga: add cap-sd-highspeed for SD/MMC node

Signed-off-by: Olof Johansson <olof@lixom.net>

Device Tree additions for LPC18xx platform
 - CREG clock controller
 - Real Time Clock (RTC)
 - Analog peripherals (ADC/DAC)
 - Warning fixes for the new dtc compiler

With the CREG clock controller in place it is now possible
to enable the internal RTC on LPC18xx/43xx platforms. The
analog peripherals (ADC/DAC) has also been added here and
enabled on both the EA4357 dev kit and Hitex eval board.

In addition to the new entries there are a fixes for the
DT warnings generated by the new dtc.

* tag 'lpc18xx_dts_for_4.7' of https://github.com/manabian/linux-lpc

:
  dt-bindings: phy-lpc18xx-usb-otg: remove unit address from binding
  ARM: dts: lpc4350-hitex-eval: fix unit name warnings from dtc
  ARM: dts: lpc4357-ea4357: fix unit name warnings from dtc
  ARM: dts: lpc18xx: remove unit addresses from creg childs
  ARM: dts: armv7-m: add unit name to interrupt-controller
  ARM: dts: lpc4350-hitex-eval: add adc1
  ARM: dts: lpc4357-ea4357: add dac
  ARM: dts: lpc4357-ea4357: add adc0
  ARM: dts: lpc18xx: add dac node
  ARM: dts: lpc18xx: add adc nodes
  ARM: dts: lpc18xx: add rtc node
  ARM: dts: lpc18xx: add creg-clk node

Signed-off-by: Olof Johansson <olof@lixom.net>

Merge tag 'versatile-dts-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-integrator into next/dt

Versatile DTS changes, baseline for the v4.7 series:
- Add CLCD panel nodes to PB1176 and PB11MPCore
- Add a DT binding blurb for the Versatile IB2 syscon
- Add DTS files for the (QEMU supported) RealView EB
  boards in all variants.
- Add DTS files for the (QEMU supported) RealView PBA8
  and PBX-A9 board variants.

* tag 'versatile-dts-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-integrator

:
  ARM: dts: realview: DT support for the PBA8 and PBX-A9
  ARM: dts: realview: support all the RealView EB board variants
  ARM: dts: realview: PB1176: define a standard VGA panel
  ARM: dts: realview: PB11MPCore: define a standard VGA panel
  Documentation/DT: add blurb for IB2 syscon to Versatile

Signed-off-by: Olof Johansson <olof@lixom.net>

Merge tag 'renesas-cleanup-for-v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into next/dt

Renesas ARM Based SoC Cleanup for v4.7

* Remove unnecessary clock-output-names properties from DT
* Use generic pinctrl properties in DT

* tag 'renesas-cleanup-for-v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas

: (26 commits)
  ARM: dts: sh73a0: Remove unnecessary clock-output-names properties
  ARM: dts: r8a73a4: Remove unnecessary clock-output-names properties
  ARM: dts: lager: Remove unnecessary clock-output-names properties
  ARM: dts: porter: Remove unnecessary clock-output-names properties
  ARM: dts: koelsch: Remove unnecessary clock-output-names properties
  ARM: dts: gose: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7794: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7793: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7791: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7779: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7778: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7740: Remove unnecessary clock-output-names properties
  ARM: dts: r7s72100: Remove unnecessary clock-output-names properties
  ARM: dts: r8a7790: Remove unnecessary clock-output-names properties
  ARM: dts: kzm9d: use generic pinctrl properties
  ARM: dts: kzm9g: use generic pinctrl properties
  ARM: dts: silk: use generic pinctrl properties
  ARM: dts: alt: use generic pinctrl properties
  ARM: dts: gose: use generic pinctrl properties
  ARM: dts: porter: use generic pinctrl properties
  ...

Signed-off-by: Olof Johansson <olof@lixom.net>

Merge tag 'renesas-arm64-dt-for-v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into next/dt

Renesas ARM64 Based SoC DT Updates for v4.7

* Use USB3.0 fallback compatibility string in DT for r8a7795 SoC
* Add CAN support to DT for r8a7795 SoC

* tag 'renesas-arm64-dt-for-v4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas

:
  arm64: dts: r8a7795: Use USB3.0 fallback compatibility string
  arm64: dts: r8a7795: Add CAN support
  arm64: dts: r8a7795: Add CAN external clock support

Signed-off-by: Olof Johansson <olof@lixom.net>

First DT batch for 4.7, additions for sama5d2 SoC:
- chipid node to identify the SoC
- SFR node (Special Function Registers)
- LCD controller's node

* tag 'at91-dt' of git://git.kernel.org/pub/scm/linux/kernel/git/nferre/linux-at91

:
  ARM: dts: at91: sama5d2: add LCD controller
  ARM: dts: at91: sama5d2: add chipid node
  ARM: dts: at91: sama5d2: add SFR node

Signed-off-by: Olof Johansson <olof@lixom.net>

Fix following DTC warnings in SMDKv210 board:

Warning (unit_address_vs_reg): Node /soc/fimd@f8000000/display-timings/timing@0 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in Exynos5440 boards:

Warning (unit_address_vs_reg): Node /pinctrl has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /rtc has a reg or ranges property, but no unit name

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in Exynos5420 SMDK5420:

Warning (unit_address_vs_reg): Node /dp-controller@145B0000/display-timings/timing@0 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in Exynos5420 Peach Pit:

Warning (unit_address_vs_reg): Node /dp-controller@145B0000/ports/port@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c@12CD0000/lvds-bridge@48/ports/port@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c@12CD0000/lvds-bridge@48/ports/port@1 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in all Exynos542x/5800 boards:

Warning (unit_address_vs_reg): Node /video-phy@10040728 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /video-phy@10040714 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /usb@12000000 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /usb@12000000/dwc3 has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /usb@12400000 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /usb@12400000/dwc3 has a reg or ranges property, but no unit name

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in all Exynos5250 boards:

Warning (unit_address_vs_reg): Node /dp-controller@145B0000/display-timings/timing@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /usb@12000000 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /usb@12000000/dwc3 has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /hdmi has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /mixer has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /video-phy@10040720 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /fixed-regulator@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /fixed-regulator@1 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /fixed-regulator@2 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c@12C70000/trackpad has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /i2c@12CD0000/lvds-bridge@20/ports/port@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c@12CD0000/lvds-bridge@20/ports/port@1 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c-arbitrator/i2c@0/embedded-controller has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /i2c-arbitrator/i2c@0/power-regulator has a reg or ranges property, but no unit name
Warning (unit_address_vs_reg): Node /i2c@12CA0000/embedded-controller has a reg or ranges property, but no unit name

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in Exynos4x12 boards:

Warning (unit_address_vs_reg): Node /camera/fimc-is@12000000/pmu has a reg or ranges property, but no unit name

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in Trats2 board:

Warning (unit_address_vs_reg): Node /i2c-gpio-1/max77693@66/regulators/ESAFEOUT1@1 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c-gpio-1/max77693@66/regulators/ESAFEOUT2@2 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /i2c-gpio-1/max77693@66/regulators/CHARGER@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /thermistor-ap@0 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /thermistor-battery@1 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in all Exynos4 boards:

Warning (unit_address_vs_reg): Node /soc/video-phy@10020710 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in Exynos3250 boards:

Warning (unit_address_vs_reg): Node /soc/video-phy@10020710 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Fix following DTC warnings in cros-adc-thermistors:

Warning (unit_address_vs_reg): Node /adc@12D10000/ncp15wb473@3 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /adc@12D10000/ncp15wb473@4 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /adc@12D10000/ncp15wb473@5 has a unit name, but no reg property
Warning (unit_address_vs_reg): Node /adc@12D10000/ncp15wb473@6 has a unit name, but no reg property

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Add vmmc and vqmmc supplies from MF circuit sheets for eMMC and SD on
odroid XU3 and XU4 to avoid warnings:

dwmmc_exynos 12200000.mmc: Looking up vmmc-supply property in node /mmc@12200000 failed

Also remove their always_on properties so the regulators could be
disabled when not used.

Signed-off-by: Markus Reichl <m.reichl@fivetechno.de>
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>

Topic branch for Device Tree changes for Exynos 3250 for v4.7:

Merge necessary new clocks from Sylwester (used by new board) and add support
for Exynos3250-based Artik5 board.

Add the resets property for the 2 USB controllers.

Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

Add the device tree entries needed to support the Altera On-Chip
RAM EDAC on the Arria10 chip.

Signed-off-by: Thor Thayer <tthayer@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

Add the device tree entries needed to support the Altera L2
cache EDAC on the Arria10 chip.

Signed-off-by: Thor Thayer <tthayer@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

Add support for the keys and flip-switches on the SoCFPGA SoCkit board.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

Add support for the blue LEDs on the SoCFPGA SoCkit board.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

The socfpga_cyclone5.dtsi is included by all DTS files which describe boards
using the Cyclone V SoC. The Cyclone V SoC has two ethernet controllers and
different boards use none, one or both of them.

The /soc/ethernet@ff702000/{} node in socfpga_cyclone5.dtsi unconditionaly
enabled gmac0 interface, which is clearly wrong for those boards which use
gmac1 interface instead.

This patch removes the entire /soc/ethernet@ff702000/{} node from the
socfpga_cyclone5.dtsi file. This is correct, since all of the board which
include this file also have correct gmac0 or gmac1 node present in them.
Minor correction had to be done to EBV SoCrates, which didn't define PHY
mode explicitly, but inherited it from the socfpga_cyclone5.dtsi .

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

The phy-addr property of stmmac is deprecated and the stmmac driver
does not use it either. On the contrary, the driver will warn if
this property is defined. Remove it.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

The PL330 DMA driver will not load on Arria10 without devicetree entries
for clocks and clock_names.  This patch adds those entries.  It also adds
the ninth interrupt, which is required for error detection.

Signed-off-by: Graham Moore <grmoore@opensource.altera.com>
Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

The CIU clock for the SD/MMC should be the sdmmc_clk and not the
sdmmc_free_clk. Also, add the correct phase shift the sdmmc_clk.

Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

Enable SD highspeed support for the SoCFPGA Arria10 devkit.

Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com>

Pull ARM fixes from Russell King:
 "A couple of small fixes, and wiring up the new syscalls which appeared
  during the merge window"

* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
  ARM: 8550/1: protect idiv patching against undefined gcc behavior
  ARM: wire up preadv2 and pwritev2 syscalls
  ARM: SMP enable of cache maintanence broadcast

Pull MMC fixes from Ulf Hansson:
 "Here are a couple of mmc fixes intended for v4.6 rc3:

  MMC host:
   - sdhci: Fix regression setting power on Trats2 board
   - sdhci-pci: Add support and PCI IDs for more Broxton host controllers"

* tag 'mmc-v4.6-rc1' of git://git.linaro.org/people/ulf.hansson/mmc:
  mmc: sdhci-pci: Add support and PCI IDs for more Broxton host controllers
  mmc: sdhci: Fix regression setting power on Trats2 board

Pull i2c fixes from Wolfram Sang:
 "Some bugfixes from I2C:

   - fix a uevent triggered boot problem by removing a useless debug
     print

   - fix sysfs-attributes of the new i2c-demux-pinctrl driver to follow
     standard kernel behaviour

   - fix a potential division-by-zero error (needed two takes)"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: jz4780: really prevent potential division by zero
  Revert "i2c: jz4780: prevent potential division by zero"
  i2c: jz4780: prevent potential division by zero
  i2c: mux: demux-pinctrl: Update docs to new sysfs-attributes
  i2c: mux: demux-pinctrl: Clean up sysfs attributes
  i2c: prevent endless uevent loop with CONFIG_I2C_DEBUG_CORE

This reverts commit 1028b55b

.

It's broken: it makes ext4 return an error at an invalid point, causing
the readdir wrappers to write the the position of the last successful
directory entry into the position field, which means that the next
readdir will now return that last successful entry _again_.

You can only return fatal errors (that terminate the readdir directory
walk) from within the filesystem readdir functions, the "normal" errors
(that happen when the readdir buffer fills up, for example) happen in
the iterorator where we know the position of the actual failing entry.

I do have a very different patch that does the "signal_pending()"
handling inside the iterator function where it is allowable, but while
that one passes all the sanity checks, I screwed up something like four
times while emailing it out, so I'm not going to commit it today.

So my track record is not good enough, and the stars will have to align
better before that one gets committed.  And it would be good to get some
review too, of course, since celestial alignments are always an iffy
debugging model.

IOW, let's just revert the commit that caused the problem for now.

Reported-by: Greg Thelen <gthelen@google.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull parisc fixes from Helge Deller:
 "Since commit 0de79858 ("parisc: Use generic extable search and
  sort routines") module loading is boken on parisc, because the parisc
  module loader wasn't prepared for the new R_PARISC_PCREL32 relocations.

  In addition, due to that breakage, Mikulas Patocka noticed that
  handling exceptions from modules probably never worked on parisc.  It
  was just masked by the fact that exceptions from modules don't happen
  during normal use.

  This patch series fixes those issues and survives the tests of the
  lib/test_user_copy kernel module test.  Some patches are tagged for
  stable"

* 'parisc-4.6-3' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Update comment regarding relative extable support
  parisc: Unbreak handling exceptions from kernel modules
  parisc: Fix kernel crash with reversed copy_from_user()
  parisc: Avoid function pointers for kernel exception routines
  parisc: Handle R_PARISC_PCREL32 relocations in kernel modules

Pull libnvdimm fixes from Dan Williams:
 "Three fixes, the first two are tagged for -stable:

   - The ndctl utility/library gained expanded unit tests illuminating a
     long standing bug in the libnvdimm SMART data retrieval
     implementation.

     It has been broken since its initial implementation, now fixed.

   - Another one line fix for the detection of stale info blocks.

     Without this change userspace can get into a situation where it is
     unable to reconfigure a namespace.

   - Fix the badblock initialization path in the presence of the new (in
     v4.6-rc1) section alignment workarounds.

     Without this change badblocks will be reported at the wrong offset.

  These have received a build success report from the kbuild robot and
  have appeared in -next with no reported issues"

* 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  libnvdimm, pfn: fix nvdimm_namespace_add_poison() vs section alignment
  libnvdimm, pfn: fix uuid validation
  libnvdimm: fix smart data retrieval

Pull GPIO fixes from Linus Walleij:
 "Here is a set of four GPIO fixes.  The two fixes to the core are
  serious as they are regressing minor architectures.

  Core fixes:

   - Defer GPIO device setup until after gpiolib is initialized.

     It turns out that a few very tightly integrated GPIO platform
     drivers initialize so early (befor core_initcall()) so that the
     gpiolib isn't even initialized itself.  That limits what the
     library can do, and we cannot reference uninitialized fields until
     later.

     Defer some of the initialization until right after the gpiolib is
     initialized in these (rare) cases.

   - As a consequence: do not use devm_* resources when allocating the
     states in the initial set-up of the gpiochip.

  Driver fixes:

   - In ACPI retrieveal: ignore GpioInt when looking for output GPIOs.

   - Fix legacy builds on the PXA without a backing pin controller.

   - Use correct datatype on pca953x register writes"

* tag 'gpio-v4.6-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
  gpio: pca953x: Use correct u16 value for register word write
  gpiolib: Defer gpio device setup until after gpiolib initialization
  gpiolib: Do not use devm functions when registering gpio chip
  gpio: pxa: fix legacy non pinctrl aware builds
  gpio / ACPI: ignore GpioInt() GPIOs when requesting GPIO_OUT_*