Merge tag 'backport/v3.14.24-ltsi-rc1/sh-drivers-fixes-20141126' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/rcar-snd-to-next-20141121' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/rcar-snd-to-v3.18-rc6' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/gpio-20141117' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/irq-renesas-intc-irqpin-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/codecs-20141124' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sata-rcar-to-v3.18-rc5' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Conflicts:
Documentation/devicetree/bindings/ata/sata_rcar.txt

Merge tag 'backport/v3.14.24-ltsi-rc1/phy-rcar-gen2-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/phy-rcar-gen2-usb-to-v3.15' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/usb-20141113' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/pci-rcar-gen2-to-v3.16' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'v3.14.25' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

This is the 3.14.25 stable release

Conflicts:
Makefile

Merge tag 'backport/v3.14.24-ltsi-rc1/pm-domains-20141126' into backport/v3.14.24-ltsi-rc1/sh-drivers

PM / domains: Add late_initcall to disable unused PM domains

As default behavior let genpd at late init try to disable the unused
PM domains.

Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Kevin Hilman <khilman@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
(cherry picked from commit 2fe71dcdfd10d3f71bb559609a8ceda9f76b1e2c)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: Document SoC-specific bindings

The documentation only mentioned the generic fallback compatible
property.
Add the missing SoC-specific compatible properties, which are already in
use.

Also drop a bogus 0x unit-address prefix while we're at it.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 56ba98acc398883324c0e70dc8aee1dc53eb2331)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: Add Volume Ramp support

This patch adds Volume Ramp to Renesas sound driver.

amixer set "DVC Out" 100%
amixer set "DVC Out Ramp Up Rate"   "0.125 dB/64 steps"
amixer set "DVC Out Ramp Down Rate" "0.125 dB/512 steps"
amixer set "DVC Out Ramp" on
aplay xxx.wav &
amixer set "DVC Out"  80%  // Volume Down
amixer set "DVC Out" 100%  // Volume Up

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 3539cacff2031f6d47881c5f3a4932b0ad5ec224)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: fallback to PIO mode if DMA mode was failed

Current Renesas R-Car sound driver probe will be failed
if it try to use DMA mode and it couldn't use for some reasons.
But PIO mode works even though in such case.
This patch try to fallback to PIO mode if DMA mode probing was failed.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit d3a768233243b5892a9c74b85896b9e8c017b259)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: tidyup debug message format and timing

Current Renesas R-Car sound driver debug message
is using random format
(ex "ssi0: xxx" / "SSI0 xxx" / "ssi[0]: xxx")
and confusable timing
("xxx probe failed" and "xxx probed" are shown in same time)
This patch fixes these

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 30cc4faf703955cd5cd07da489bd817ae43e3fec)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: enable enumerated DVC valume settings

DVC controls some digital volume features.
Volume Ramp is listed as "XX dB / YY steps",
and this enumerated settings are easy for users.
This patch adds rsnd_dvc_cfg_e and care about
enumerated settings.
Compiler will report like below at this point,
but, it will be removed if Volume Ramp was supported.
  warning: '_rsnd_dvc_pcm_new_e' defined but not used

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 018342976ce971944dd4d9309f75e86382079a2b)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: enable single DVC valume settings

DVC controls some digital volume features.
Some of them requests values for "each channels",
but, some of them requests values for "feature".
And, Volume Ramp has "feature" settings.
This patch adds rsnd_dvc_cfg_s and care about
single settings.
Compiler will report like below at this point,
but, it will be removed if Volume Ramp was supported.
  warning: '_rsnd_dvc_pcm_new_s' defined but not used

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit ab2e479667507329475c8ef93d61f3dbe654c3c2)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: enable multiple DVC valume settings

DVC controls some digital volume features.
Some of them requests values for "each channels",
but, some of them requests values for "feature".
Current dvc.c is supporting Mute/Volume,
and these have "each channels" settings.
This patch adds rsnd_dvc_cfg_m and care about
multiple settings for each channels.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit ec14af91a03f7d68b2a72bec20be2ab583d3f63a)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: move DVC_DVUER settings under rsnd_dvc_volume_update()

We need to Enable/Disable DVC_DVUER register if we set
DVCp_ZCMCR, DVCp_VRCTR, DVCp_VRPDR, DVCp_VRDBR,
DVCp_VOL0R, DVCp_VOL1R, DVCp_VOL2R, DVCp_VOL3R,
DVCp_VOL4R, DVCp_VOL5R, DVCp_VOL6R, DVCp_VOL7R
and, these are controlled under rsnd_dvc_volume_update().
This patch moves DVC_DVUER settings to it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 140bab8961eb4047070b46a6dd50ec87496e0cde)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: control DVC_DVUCR under rsnd_dvc_volume_update()

rsnd_dvc_volume_update() is main function to control
DVC feature like Digital Volume / Mute / Ramp etc.
DVC_DVUCR should be controlled under this function.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 1c5d1c988302f324ac396ac13461d59d091be605)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: add struct rsnd_dvc_cfg and control DVC settings

DVC can control Digital Volume / Mute / Volume Ramp etc,
and these uses different max value.
Current driver is using fixed max value for each settings.
This patch adds new struct rsnd_dvc_cfg, and control these.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 92b9a6991b2e3a4ccf5ffc956730d36835d53a79)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: tidyup RSND_DVC_VOLUME_NUM to RSND_DVC_CHANNELS

RSND_DVC_VOLUME_NUM means DVC channel number.
This patch tidyups this un-understandable naming

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit 9960ce97432bdb1defc76ed80ac19e37e8778bc6)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: tidyup debug information when read/write

b8c637864a6904a9ba8e0df556d5bdf9f26b2c54
(ASoC: rsnd: use regmap_mmio instead of original regmap bus)
added regmap_mmio support on Renesas R-Car sound driver.
Then, debug information of register read/write
indicates regmap index, not register address.
This is a little bit confusable information.
This patch tidyup debug message, and added regmap debug hint
on comment area.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
(cherry picked from commit ace0eb1e91a75b84b1be3d610b79509a5bd94df1)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: rsnd: remove unsupported PAUSE flag

R-Car sound doesn't support PAUSE.
Remove SNDRV_PCM_INFO_PAUSE flags from snd_pcm_hardware info

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org
(cherry picked from commit 706c66213e5e623e23f521b1acbd8171af7a3549)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

gpio: add flags argument to gpiod_get*() functions

The huge majority of GPIOs have their direction and initial value set
right after being obtained by one of the gpiod_get() functions. The
integer GPIO API had gpio_request_one() that took a convenience flags
parameter allowing to specify an direction and value applied to the
returned GPIO. This feature greatly simplifies client code and ensures
errors are always handled properly.

A similar feature has been requested for the gpiod API. Since setting
the direction of a GPIO is so often the very next action done after
obtaining its descriptor, we prefer to extend the existing functions
instead of introducing new functions that would raise the
number of gpiod getters to 16 (!).

The drawback of this approach is that all gpiod clients need to be
updated. To limit the pain, temporary macros are introduced that allow
gpiod_get*() to be called with or without the extra flags argument. They
will be removed once all consumer code has been updated.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Mark Brown <broonie@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit 39b2bbe3d715cf5013b5c48695ccdd25bd3bf120)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

gpio: include linux/bug.h in interface header

Today's linux-next kernel started showing build errors for the
use of WARN_ON in linux/gpio/consumer.h:

In file included from drivers/video/backlight/pwm_bl.c:13:0:
include/linux/gpio/consumer.h: In function 'gpiod_put':
include/linux/gpio/consumer.h:81:2: error: implicit declaration of function 'WARN_ON' [-Werror=implicit-function-declaration]

It's not clear why this never happened before, but this patch
fixes it by including the header that contains the defintion
of this macro.

Signed-off-by: Arnd Bergmann <arnd@arnd.de>
Acked-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit cdf86cd233207ed992a647f0b9d42c60735756e7)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

gpio: Add helpers for optional GPIOs

Introduce gpiod_get_optional() and gpiod_get_index_optional() helpers
that make it easier for drivers to handle optional GPIOs.

Currently in order to handle optional GPIOs, a driver needs to special
case error handling for -ENOENT, such as this:

gpio = gpiod_get(dev, "foo");
if (IS_ERR(gpio)) {
if (PTR_ERR(gpio) != -ENOENT)
return PTR_ERR(gpio);

gpio = NULL;
}

if (gpio) {
/* set up GPIO */
}

With these new helpers the above is reduced to:

gpio = gpiod_get_optional(dev, "foo");
if (IS_ERR(gpio))
return PTR_ERR(gpio);

if (gpio) {
/* set up GPIO */
}

While at it, device-managed variants of these functions are also
provided.

Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit 29a1f2333e07bbbecb920cc78fd035fe8f53207a)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

gpio: Add missing device-managed documentation

Add the GPIO-related device-managed functions to the list of functions
in Documentation/driver-model/devres.txt.

Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit f9748ef13b6a7dfe9fe9bcae8b3a8cbc10c1e188)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

gpiolib: add gpiochip_get_desc() driver function

Some drivers dealing with a gpio_chip might need to act on its
descriptors directly; one example is pinctrl drivers that need to lock a
GPIO for being used as IRQ using gpiod_lock_as_irq().

This patch exports a gpiochip_get_desc() function that returns the
GPIO descriptor at the requested index. It also sweeps the
gpio_to_chip() function out of the consumer interface since any holder
of a gpio_chip reference can manipulate its GPIOs way beyond what a
consumer should be allowed to do.

As a result, gpio_chip is not visible anymore to simple GPIO consumers.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Jean-Jacques Hiblot <jjhiblot@traphandler.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit bb1e88ccb771492ac908ac295ec135efa1d53093)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

gpio: make gpiod_direction_output take a logical value

The documentation was not clear about whether
gpio_direction_output should take a logical value or the physical
level on the output line, i.e. whether the ACTIVE_LOW status
would be taken into account.

This converts gpiod_direction_output to use the logical level
and adds a new gpiod_direction_output_raw for the raw value.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit ef70bbe1aaa612f75360e5df5952fddec50b7ca9)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

irqchip: renesas-intc-irqpin: Add minimal runtime PM support

This is just enough to let pm_clk_*() enable the functional clock, and
manage it for suspend/resume, if present.
Before, it was assumed enabled by the bootloader or reset state.

To prevent the clock from being disabled while the module is needed for
wake-up, implement irq_chip.irq_set_wake(), which increments/decrements
the clock's enable_count when needed.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lkml.kernel.org/r/1410527720-18061-3-git-send-email-geert+renesas@glider.be
Signed-off-by: Jason Cooper <jason@lakedaemon.net>
(cherry picked from commit 705bc96c2c15313c0677607f6e81800f4d2b4534)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

irqchip: renesas-intc-irqpin: Add helper variable dev = &pdev->dev

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lkml.kernel.org/r/1410527720-18061-2-git-send-email-geert+renesas@glider.be
Signed-off-by: Jason Cooper <jason@lakedaemon.net>
(cherry picked from commit 36845f1b54edc3237626d48270bbf3bc26d45f4f)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

irqchip: renesas-intc-irqpin: Document SoC-specific bindings

The documentation only mentioned the generic fallback compatible property.
Add the missing SoC-specific compatible properties, which are already in
use.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Link: https://lkml.kernel.org/r/1409212800-4957-1-git-send-email-geert+renesas@glider.be
Signed-off-by: Jason Cooper <jason@lakedaemon.net>
(cherry picked from commit 9857b069b27abbc3d3b2c9f474e40d8b2b55a228)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

irqchip: renesas-intc-irqpin: Add suspend-to-RAM wake up support

Set the ->irq_enable() and ->irq_disable() methods to NULL to enable
lazy disable of interrupts, and set IRQCHIP_MASK_ON_SUSPEND to tell the
core that only IRQs marked as wake-ups need to stay enabled during
suspend-to-RAM.

This makes wake-up by gpio-keys from suspend-to-RAM work on
r8a7740/Armadillo.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lkml.kernel.org/r/1408546172-22484-1-git-send-email-geert+renesas@glider.be
Signed-off-by: Jason Cooper <jason@lakedaemon.net>
(cherry picked from commit 1c36d42c4ffee9e38e122ed822dd0a545a3a86da)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ASoC: codecs: Make OF supported CODECs visible in Kconfig

Now that we have a generic card driver we can't rely on the card driver
selecting the CODECs for us so make the CODECs that can be enabled with
OF directly selectable in Kconfig. For the platforms not using OF it's
not clear that we don't still want to have some board specific selection
since the kernel needs to contain code to register the devices; ACPI could
provide this from firmware does not yet support any kind of generic card.

It may also be desirable to hide these if OF is not enabled to reduce
noise.

Signed-off-by: Mark Brown <broonie@linaro.org>
(cherry picked from commit 1291e14175e6b83efe1464f32189acb21bc4be09)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ata: sata_rcar: Add r8a7793 device support

Signed-off-by: Koji Matsuoka <koji.matsuoka.xm@renesas.com>
Signed-off-by: Yoshihiro Kaneko <ykaneko0929@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
(cherry picked from commit e35b98849f2530bb77f8fe649b3eaa1489ff9d33)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

ata: sata_rcar: Disable DIPM mode for r8a7790 ES1

Unlike other SATA R-Car r8a7790 controllers the r8a7790 ES1 SATA R-Car
controller needs to be run with DIPM disabled.

Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: stable@vger.kernel.org
(cherry picked from commit aa1cf25887099bba68f1f3879c0d394e08b8779f)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

phy: Renesas R-Car Gen2 PHY driver

This PHY, though formally being a part of Renesas USBHS controller, contains the
UGCTRL2 register that controls multiplexing of the USB ports (Renesas calls them
channels) to the different USB controllers: channel 0 can be connected to either
PCI EHCI/OHCI or USBHS controllers, channel 2 can be connected to PCI EHCI/OHCI
or xHCI controllers.

This is a new driver for this USB PHY currently already supported under drivers/
usb/phy/. The reason for writing the new driver was the requirement that the
multiplexing  of USB channels to the controller be dynamic, depending on what
USB drivers  are loaded,  rather than static as provided by the old driver. The
infrastructure provided by drivers/phy/phy-core.c  seems to fit that purpose
ideally. The new driver only  supports device tree probing  for now.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
(cherry picked from commit 1233f59f745b237d85f12aa9cf12ffab469f322d)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
drivers/phy/Kconfig
drivers/phy/Makefile

usb: phy: rcar-gen2-usb: always use 'dev' variable in probe() method

The probe() method has the 'dev' local variable declared and used but strangely
not in all cases where it should be...

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
(cherry picked from commit 7241a21a7178eb327ca2c8f2ffc325b10e1e05bd)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

usb: hcd: add generic PHY support

Add the generic PHY support, analogous to the USB PHY support. Intended it to be
used with the PCI EHCI/OHCI drivers and the xHCI platform driver.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 0043325495222139daa0696db736f67658dc7770)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

usb: rename phy to usb_phy in HCD

The USB PHY member of the HCD structure is renamed to 'usb_phy' and
modifications are done in all drivers accessing it.
This is in preparation to adding the generic PHY support.

Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
[Sergei: added missing 'drivers/usb/misc/lvstest.c' file, resolved rejects,
updated changelog.]
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 3d46e73dfdb840f460e5b06416965d132570ec33)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
drivers/usb/misc/lvstest.c

usb: rename usb_port device objects

The current port name "portX" is ambiguous.  Before adding more port
messages rename ports to "<hub-device-name>-portX"

This is an ABI change, but the suspicion is that it will go unnoticed as
the port power control implementation has been broken since its
introduction.  If however, someone was relying on the old name we can
add sysfs links from the old name to the new name.

Additionally, it unifies/simplifies port dev_printk messages and modifies
instances of:
dev_XXX(hub->intfdev, ..."port %d"...
dev_XXX(&hdev->dev, ..."port%d"...
into:
dev_XXX(&port_dev->dev, ...

Now that the names are unique usb_port devices it would be nice if they
could be included in /sys/bus/usb.  However, it turns out that this
breaks 'lsusb -t'.  For now, create a dummy port driver so that print
messages are prefixed "usb 1-1-port3" rather than the
subsystem-ambiguous " 1-1-port3".

Finally, it corrects an odd usage of sscanf("port%d") in usb-acpi.c.

Suggested-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit d99f6b41308779244662109a9c2bad09a82e8ac6)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

usb: disable port power control if not supported in wHubCharacteristics

A hub indicates whether it supports per-port power control via the
wHubCharacteristics field in its descriptor.  If it is not supported
a hub will still emulate ClearPortPower(PORT_POWER) requests by
stopping the link state machine.  However, since this does not save
power do not bother suspending.

This also consolidates support checks into a
hub_is_port_power_switchable() helper.

Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 9262c19d14c433a6a1ba25c3ff897cb89e412309)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

usb: Reset USB-3 devices on USB-3 link bounce

On disconnect USB3 protocol ports transit from U0 to SS.Inactive to Rx.Detect,
on a recoverable error, the port stays in SS.Inactive and we recover from it by
doing a warm-reset (through usb_device_reset if we have a udev for the port).

If this really is a disconnect we may end up trying the warm-reset anyways,
since khubd may run before the SS.Inactive to Rx.Detect transition, or it
may get skipped if the transition to Rx.Detect happens before khubd gets run.

With a loose connector, or in the case which actually led me to debugging this
bad ACPI firmware toggling Vbus off and on in quick succession, the port
may transition from Rx.Detect to U0 again before khubd gets run. In this case
the device state is unknown really, but khubd happily goes into the resuscitate
an existing device path, and the device driver never gets notified about the
device state being messed up.

If the above scenario happens with a streams using device, as soon as an urb
is submitted to an endpoint with streams, the following appears in dmesg:

ERROR Transfer event for disabled endpoint or incorrect stream ring
@0000000036807420 00000000 00000000 04000000 04078000

Notice how the TRB address is all zeros. I've seen this both on Intel
Pantherpoint and Nec xhci hosts.

Luckily we can detect the U0 to SS.Inactive to Rx.Detect to U0 all having
happened before khubd runs case since the C_LINK_STATE bit gets set in the
portchange bits on the U0 -> SS.Inactive change. This bit will also be set on
suspend / resume, but then it gets cleared by port_hub_init before khubd runs.

So if the C_LINK_STATE bit is set and a warm-reset is not needed, iow the port
is not still in SS.Inactive, and the port still has a connection, then the
device needs to be reset to put it back in a known state.

I've verified that doing the device reset also fixes the transfer event with
all zeros address issue.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
(cherry picked from commit a82b76f7fa6154e8ab2d8071842a3e38b9c0d0ff)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

usb: chipidea: add TPL support for targeted hosts

For OTG and Embedded hosts, they may need TPL (Targeted Peripheral List)
for usb certification and other vender specific requirements, the
platform can tell chipidea core driver if it supports tpl through DT
or platform data.

Signed-off-by: Peter Chen <peter.chen@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit f6a9ff07832a9d30d457e976e6233b4351cd4cdf)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Add gen2 device tree support

Add device tree probing support to the 'pci-rcar-gen2' driver.

[Sergei: numerous fixes/cleanups/additions]
[bhelgaas: whitespace fix]
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Arnd Bergmann <arnd@arndb.de>
(cherry picked from commit d47b62f4b1c0374517ccbdf3ee4a7cceead58905)

Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Use new OF interrupt mapping when possible

Use new OF interrupt mapping (of_irq_parse_and_map_pci()) when possible.
This is the recommended method of doing the IRQ mapping.  For old
devicetrees we fall back to the previous practice.

This allows interrupts to be remapped across bridges.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
(cherry picked from commit f8f2fe7355fb04dd129d49ac0ad440beb44f0f79)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Break out window size handling

Break out the hard coded window size code to allow dynamic setup. The
window size is still left at 1GiB but with this patch changing window size
is easy for testing.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
(cherry picked from commit 33966fd9b5bcae325a283d7d46156bab25bc162b)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Register each instance independently

Convert the code to allow per-device probe() like other device drivers.
This also delays driver registration due to change from subsys_initcall()
to regular module_platform_driver().

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
(cherry picked from commit 546cadda3575153bdd0af4febfe958cc5945f95a)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Fix bridge logic configuration accesses

The bridge logic at slot 0 only supports reads up to 0x40 and the rest of
the PCI configuration space for this slot is marked as reserved in the
manual.

Trying a read from offset 0x100 is producing an error from the bridge. With
error interrupts enabled, the following is printed:

  pci-rcar-gen2 ee0d0000.pci: error irq: status 00000014

Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
(cherry picked from commit e64a2a973e179613489a0ddb1bb550dff108164e)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Add error interrupt handling

Add option to enable interrupts to report any errors from the AHB-PCI
bridge to help find any issues with the bridge when in use.

Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
(cherry picked from commit 80a595d941a2ce2953e6c77d822d8a4c3d8fa5a0)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

PCI: rcar: Check platform_get_irq() return code

The current code does not check the return from platform_get_irq() so add
an error check and return if this call does fail.

Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
(cherry picked from commit ed65b78881c713b41051310780f94bde3c010db9)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

phy: core: Fix of_phy_provider_lookup to return PHY provider for sub node

Fixed of_phy_provider_lookup to return 'phy_provider' if _of_phy_get
passes the node pointer of the sub-node of phy provider node. This is
needed when phy provider implements multiple PHYs and each PHY is
modelled as the sub-node of PHY provider device node.

Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
Acked-by: Lee Jones <lee.jones@linaro.org>
(cherry picked from commit 2a4c37016ca96e413cd352985d3a0db8cfb7716c)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

phy: fix kernel oops in phy_lookup()

The kernel oopses in phy_lookup() due to 'phy->init_data' being NULL if we
register PHYs from a device tree probing driver and then call phy_get() on a
device that has no representation in the device tree (e.g. a PCI device).
Checking the pointer before dereferening it and skipping an interation if
it's NULL prevents this kernel oops.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 743bb387a1edbf1ebbba6cf77c1af3e488886c39)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

phy: core: Let node ptr of PHY point to PHY and not of PHY provider

In case of multi-phy PHY providers, each PHY should be modeled as a sub
node of the PHY provider. Then each PHY will have a different node pointer
(node pointer of sub node) than that of PHY provider. Added this provision
in the PHY core.
Also fixed all drivers to use the updated API.

Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>
Acked-by: Lee Jones <lee.jones@linaro.org>
(cherry picked from commit f0ed817638b59aa927f1f7e9564dd8796b18dc4f)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
drivers/phy/phy-berlin-sata.c
drivers/phy/phy-exynos5-usbdrd.c
drivers/phy/phy-exynos5250-sata.c
drivers/phy/phy-hix5hd2-sata.c
drivers/phy/phy-qcom-apq8064-sata.c
drivers/phy/phy-samsung-usb2.c
drivers/phy/phy-sun4i-usb.c
drivers/phy/phy-ti-pipe3.c
drivers/phy/phy-xgene.c

Merge tag 'backport/v3.14.24-ltsi-rc1/rcar-thermal-to-v3.18-rc4' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/i2c-sh-mobile-to-v3.18-rc4' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/xhci-plat-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-mobile-sdhi-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/pm-runtime-20141105' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/mmc-20141105' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/usbhs-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-tmu-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-sci-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-pfc-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-mtu2-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-msiof-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-drivers-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/sh-cmt-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/rspi-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/rcar-snd-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/rcar-du-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Conflicts:
drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c

Merge tag 'backport/v3.14.24-ltsi-rc1/of-20141105' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/pinctrl-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/mach-shmobile-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/i2c-rcar-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/da9210-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

Merge tag 'backport/v3.14.24-ltsi-rc1/cpg-to-v3.18-rc1' into backport/v3.14.24-ltsi-rc1+v3.14.25/snapshot-merge.wip

drm/rcar-du: Add OF support

Implement support for the R-Car DU DT bindings in the rcar-du DRM
driver.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
(cherry picked from commit 96c026911890ceacee238da00a0b140ad634cc43)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

drm/rcar-du: Use struct videomode in platform data

In preparation for DT support where panel timings will be described by a
DRM-agnostic video mode, replace the struct drm_mode_modeinfo instance
in the panel platform data with a struct videomode.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
(cherry picked from commit 1d46fea7d091f9dc2d4fd3fcb9f0117ca288f9a5)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

drm/rcar-du: Update copyright notice

The "Renesas Corporation" listed in the copyright notice doesn't exist.
Replace it with "Renesas Electronics Corporation" and update the
copyright years.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
(cherry picked from commit 36d50464e05f498fa4024270e091b306af5de898)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

spi: rspi: Add support for new R-Car Gen2 SoCs

Add support for QSPI in:
  - r8a7792 (R-Car V2H)
  - r8a7793 (R-Car M2-N)
  - r8a7794 (R-Car E2)

r8a7791 is now called "R-Car M2-W".

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@linaro.org>
(cherry picked from commit 6089af775a4b4ea5237648de650dda548cb841c2)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
Documentation/devicetree/bindings/spi/spi-rspi.txt

spi: rspi: Add DT support to DMA setup

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@linaro.org>
(cherry picked from commit e825b8dd2b363e9134006fb141825518a11b2bf4)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
Documentation/devicetree/bindings/spi/spi-rspi.txt

spi: rspi: Configure DMA slave bus width to 8 bit

The new Renesas R-Car Gen2 DMA Controller driver (rcar-dmac) requires
explicit configuration of the DMA slave bus width.
Hardcode the DMA transfer size to 1 byte for both directions, as that's
the only supported configuration (16-bit DMA support was removed in
commit 9c5de2c1754c2bb3c69c4d7bf0d0edc0a61d8232 ("spi: rspi: Remove
unused 16-bit DMA support")).

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@linaro.org>
(cherry picked from commit a30b95a7d81cfc3442beb5a9635f22b19c97bbfc)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

spi: rspi: Use devm_kasprintf()

Use the devm_kasprintf() helper function instead of open coding
error-prone buffer handling and string formatting.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@linaro.org>
(cherry picked from commit 43937455c9bf294cffc2f25c4a4d0a1b9bbd88e6)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

spi: rspi: Remove unneeded semicolon

Introduced by commit 426ef76dd8a394a0e04d096941cd9acb49539a3e ("spi:
rspi: Add DT support").

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@linaro.org>
(cherry picked from commit d64b472678e17ef9afb251577f4f544793483fa9)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

devres: Improve devm_kasprintf()/kvasprintf() support

  - Add devm_kasprintf()/kvasprintf(), introduced by commit
    75f2a4ead5d5890ada9c2663a70fb58613c0d9f2 ("devres: Add
    devm_kasprintf and devm_kvasprintf API"), to
    Documentation/driver-model/devres.txt,
  - Improve kernel doc: the string is not an existing formatted string,
    but is formatted into the newly-allocated buffer,
  - Add a __printf() annotation to devm_kasprintf(), so the compiler
    will verify the format string argument types.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit bef59c5024be687ef4f228915af9961307a779ab)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
Documentation/driver-model/devres.txt
drivers/base/devres.c

devres: Add devm_kasprintf and devm_kvasprintf API

devm_kasprintf() and devm_kvasprintf() are the managed counterparts
for kasprintf() and kvasprintf().

Signed-off-by: Himangi Saraogi <himangi774@gmail.com>
Acked-by: Julia Lawall <julia.lawall@lip6.fr>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 75f2a4ead5d5890ada9c2663a70fb58613c0d9f2)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Conflicts:
drivers/base/devres.c

mmc: tmio: prevent endless loop in tmio_mmc_set_clock()

I spent a couple of days with the driver just hanging due to me forgetting to
specify the external crystal frequency,  so that clk_get_rate() returned 0 and
thus the loop in tmio_mmc_set_clock() never ended. I don't think that's an
acceptable behavior, so I suggest that the minimum frequency is checked for 0
in tmio_mmc_host_probe().

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Acked-by: Ian Molton <ian.molton@codethink.co.uk>
Cc: stable@vger.kernel.org
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit bb98d9d1d2e791f368295c9c09c9a03fe3ac0e62)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: tmio: enable odd number size access

Current tmio is using sd_ctrl_read16/write16_rep()
for data transfer.
It works if transfer size was even number,
but, last 1 byte will be ignored if
transfer size was odd number.
This patch adds new tmio_mmc_transfer_data()
and solve this issue.

Tested-by: Shinobu Uehara <shinobu.uehara.xc@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit b9bd7ff8069a7125da897de657da421c497d4f15)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: use .multi_io_quirk on sh_mobile

Now, sh_mobile_sdhi can use .multi_io_quirk callback
instead of MMC_CAP2_NO_MULTI_READ flags.
let's use it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit 8b4efe2f58b1699a57f6c30d17345b801ab4213b)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: use .multi_io_quirk on tmio_mmc

Now, tmio_mmc can use .multi_io_quirk callback
instead of MMC_CAP2_NO_MULTI_READ flags.
let's use it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit bbf0208d39121bd8873b032459cb2b5f35e14593)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: tmio-mmc: Add support for SDHI on new R-Car Gen2 SoCs

- r8a7792 (R-Car V2H)
  - r8a7793 (R-Car M2-N)
  - r8a7794 (R-Car E2)

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit a638640335d86549da2b7d38f82d584294333512)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: tmio: add actual clock support as option

Some controller is supporting actual clock on SD_CLK_CTRL :: DIV[7:0].
Renesas SH-Mobile SDHI doesn't support,
but, Renesas R-Car SDHI supports it.
This patch adds new TMIO_MMC_CLK_ACTUAL flag for it.

[Kuninori Morimoto: tidyuped for upstreaming]

Tested-by: Nguyen Xuan Nui <nx-nui@jinso.co.jp>
Tested-by: Hiep Cao Minh <cm-hiep@jinso.co.jp>
Signed-off-by: Shinobu Uehara <shinobu.uehara.xc@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit da29fe2bf573f0ae56fdc2e790387cb73fc8c6f8)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: tmio: remove SCLKEN bit setting from tmio_mmc_set_clock()

TMIO clock is set via tmio_mmc_set_clock() -> tmio_mmc_clk_start(),
and SCLKEN bit will be set on tmio_mmc_clk_start().
It is not needed on tmio_mmc_set_clock() function.
The required clock setting will not be able to set
in some clocks without this patch.

Tested-by: Nguyen Xuan Nui <nx-nui@jinso.co.jp>
Tested-by: Hiep Cao Minh <cm-hiep@jinso.co.jp>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit b30d6958a014812158d9729a3e0b6d448f6e8968)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: tmio: remove Renesas specific #ifdef

This patch adds new TMIO_MMC_HAVE_CTL_DMA_REG flag,
and remove Renesas specific #ifdef from tmio driver

Tested-by: Nguyen Xuan Nui <nx-nui@jinso.co.jp>
Tested-by: Hiep Cao Minh <cm-hiep@jinso.co.jp>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit e85dd04ea8c8d32ba8eae278959d28df34338e9d)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

mmc: tmio: check ILL_FUNC instead of CBSY

Some controllers need to check SD bus status when writing data.
Then, it checks ILL_FUNC bit on SD_INFO2 register,
and this method is controlled via TMIO_MMC_HAS_IDLE_WAIT flags.
Same method is required on tmio_mmc_data_irq() which will
be called after writing data.

Current driver is checking CBSY bit for this purpose,
but, some controllers doesn't have CBSY bit.
This patch checks ILL_FUNC bit instead of CBSY bit
if it has TMIO_MMC_HAS_IDLE_WAIT flags

[Kuninori Morimoto: tidyuped for upstreaming]

Tested-by: Nguyen Xuan Nui <nx-nui@jinso.co.jp>
Tested-by: Hiep Cao Minh <cm-hiep@jinso.co.jp>
Signed-off-by: Shinobu Uehara <shinobu.uehara.xc@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
(cherry picked from commit 81e888daab027e05130b0459e5d8e0b42dac4323)
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>