The asus_wmi_get_devstate_simple() returns 0-1 on success.  In theory
according to static checkers, it can return either -EIO or -ENODEV on
failure.  Currently the error handling code only handles -ENODEV and
-EIO is treated as success.  Let's make it handle the -EIO error as
well.

It's possible that it can't actually return -EIO and this patch is not
needed but in that case this patch is harmless and silences a static
checker warning so it's still worth it.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Commit d80d1341 ("i2c: designware: Move common probe code into
i2c_dw_probe()") caused the I2C adapter lookup code here to fail for PCI
enumerated i2c-designware because commit changed the adapter name but
didn't update it here.

Fix the I2C adapter lookup by using the "Synopsys DesignWare I2C adapter"
name.
Reported-by: NJeremiah Mahler <jmmahler@gmail.com>
Fixes: d80d1341 ("i2c: designware: Move common probe code into i2c_dw_probe()")
Signed-off-by: NJarkko Nikula <jarkko.nikula@linux.intel.com>
Tested-by: NJeremiah Mahler <jmmahler@gmail.com>
Signed-off-by: NOlof Johansson <olof@lixom.net>

Like some of the other Yoga models the Lenovo Yoga 900 does not have a
hw rfkill switch, and trying to read the hw rfkill switch through the
ideapad module causes it to always reported blocking breaking wifi.

This commit adds the Lenovo Yoga 900 to the no_hw_rfkill dmi list, fixing
the wifi breakage.

BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1275490
Cc: stable@vger.kernel.org
Reported-and-tested-by: NKevin Fenzi <kevin@scrye.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

This changes the entry to the whitelist of machines that do not have
a physical rfkill switch. Unfortunately, the Yoga 3 generation seems
to use upper-case letters for the YOGA 3 Pro-1370, while it uses normal
capitalization for its Yoga 3 1170 and 1470 siblings.

In order to catch all variants of the Yoga 3, I'm changing both
the entry for the 1470 (using "Yoga" as the name) and the entry for
the Pro 1370 (using all-caps "YOGA") to not match the exact model number
but only the generation. This way, the 1170 and 1470 models share one
entry, but if the firmware changes from one format to the other, it will
still work.

The second entry for Yoga 2 Pro that was recently added for some
reason ended up not being added in alphanumeric order, and I'm
moving the Yoga 3 1470 entry down while making the change, so they
are sorted more logically.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The ideapad-laptop handles most special keys on various Lenovo Laptops
including the Yoga line. Unfortunately, the Yoga 3 11/13/14 models have
one important exception, which is the Fn-ESC combination.

On other Lenovo Laptops, this is FnLock, which switches the function keys
between the primary (Mute, Vol down, Vol up, ...) and the secondary (F1,
F2, F3, ...) behavior. On the new machines, FnLock is only available
through BIOS setup (possibly through a yet-to-be-implemented feature
in this driver) but not through Fn-ESC, but instead the ESC key itself
switched between ESC and a "Paper Display" app for Windows.

Unfortunately, that means that you can never have both ESC *and* the
function keys working at the same time without needing to press Fn on
one of them.
As pointed out in the official Lenovo Forum by dozens of users, this
makes the machine rather useless for any serious work [1].

I have now studied the ACPI DSDT one more time and found the event
that is generated for the ESC key. Unlike all other key events on this
machine, it is actually a WMI, while the other ones are read from the
embedded controller.

I am now installing a WMI notifier that uses the event number from the
WMI subsystem as the scancode. The only event number generated here is
'128', and that fits in nicely with the two existing ranges of scancodes
used by the EC: 0-15 for the 16-bit VPCCMD_R_VPC register, 16-17 for
the VPCCMD_R_NOVO register and 64-67 for VPCCMD_R_SPECIAL_BUTTONS.

The only sane way to handle this button (in absence of the Windows Paper
Display driver) seems to be to have it emit KEY_ESC, so that is what
I use as the default. Should any user ever want to overwrite the default,
they can install their own keymap.

To ensure that we can still build the driver without adding a CONFIG_WMI
dependency, all new code is enclosed in #ifdef.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

[1] https://forums.lenovo.com/t5/Lenovo-Yoga-Series-Notebooks/YOGA-3-14-How-to-reclaim-my-Esc-key-and-permanently-disable/td-p/2070816Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The thinkpad_acpi driver currently emits error messages on unsupported
brightness interfaces, giving the impression that someone will implement
those. However, this error is spit out on nearly every thinkpad in
production since 2 years now. Furthermore, the backlight interfaces on
those devices are supported by the i915 driver just fine.

Downgrade the error message to a normal pr_info() and stop telling people
to report it to IBM.
Signed-off-by: NDavid Herrmann <dh.herrmann@gmail.com>
Acked-by: NHenrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Commit 53147b6c ("toshiba_acpi: Fix
hotkeys registration on some toshiba models") fixed an issue on some
laptops regarding hotkeys registration, however, if failed to address
the initialization of the hotkey_event_type variable, and thus, it can
lead to potential unwanted effects as the variable is being checked.

This patch initializes such variable to avoid such unwanted effects.

Cc: <stable@vger.kernel.org> # 4.1+
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

This patch solves:

on x86_64:

when CONFIG_ACPI is not enabled:

../drivers/mfd/intel_soc_pmic_bxtwc.c: In function 'bxtwc_probe':
../drivers/mfd/intel_soc_pmic_bxtwc.c:342:2:
    error: implicit declaration of function 'acpi_evaluate_integer' [-Werror=implicit-function-declaration]
        status = acpi_evaluate_integer(handle, "_HRV", NULL, &hrv);
        ^
Reported-by: NRandy Dunlap <rdunlap@infradead.org>
Signed-off-by: NLee Jones <lee.jones@linaro.org>

vga_switcheroo_client_ops has always been declared const since its
introduction with 26ec685f ("vga_switcheroo: Introduce struct
vga_switcheroo_client_ops").

Do so for vga_switcheroo_handler as well.

 drivers/gpu/drm/amd/amdgpu/amdgpu.ko:
   6 .rodata       00009888
- 19 .data         00001f00
+ 19 .data         00001ee0
 drivers/gpu/drm/nouveau/nouveau.ko:
   6 .rodata       000460b8
  17 .data         00018fe0
 drivers/gpu/drm/radeon/radeon.ko:
-  7 .rodata       00030944
+  7 .rodata       00030964
- 21 .data         0000d6a0
+ 21 .data         0000d678
 drivers/platform/x86/apple-gmux.ko:
-  7 .rodata       00000140
+  7 .rodata       00000160
- 11 .data         000000e0
+ 11 .data         000000b8

Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Darren Hart <dvhart@linux.intel.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: NLukas Wunner <lukas@wunner.de>
Reviewed-by: Christian König <christian.koenig@amd.com>.
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Protect the dev member assignment in ->remove() since user may potentially call
unbind from a sysfs even if the driver is built-in. The latter might be racy
with ongoing SCU communication.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Eliminate some boilerplate code by using module_pci_driver() instead of
init/exit, moving the salient bits from init into probe.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Switch the code to use struct device * instead of struct pci_dev * since there
is no reason to access PCI related features in the driver.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

As much as possible propagate a pointer to struct intel_scu_ipc_dev.

There is no functional change.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The error handling is broken right now since it leaves resources unfreed.
Convert the code to use managed resources to fix the error handling.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

These are not implementations of default architecture code but helpers
for drivers. Move them to the place they belong to.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Acked-by: NDarren Hart <dvhart@linux.intel.com>
Acked-by: NHitoshi Mitake <mitake.hitoshi@lab.ntt.co.jp>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

The macros KELVIN_TO_CELSIUS and CELSIUS_TO_KELVIN actually convert
between deciKelvins and Celsius, so rename them to reflect that. While
at it, use a statement expression in DECI_KELVIN_TO_CELSIUS to prevent
expanding the argument multiple times and get rid of a few casts.
Signed-off-by: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: NDarren Hart <dvhart@linux.intel.com>
Signed-off-by: NZhang Rui <rui.zhang@intel.com>

Some EC implementations include a small nvram space used to store
verified boot context data. This patch offers a way to expose this
data to userspace.
Reviewed-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NEmilio López <emilio.lopez@collabora.co.uk>
Signed-off-by: NOlof Johansson <olof@lixom.net>

Function led_rgb_store() contains some direct returns in error cases that
leak the already allocated cros_ec_command message structure. Make sure
that 'msg' is freed in all exit paths. Detected by Coverity CID 1309666.
Signed-off-by: NChristian Engelmayer <cengelma@gmx.at>
Signed-off-by: NOlof Johansson <olof@lixom.net>

The allocated cros_ec_command message structure is not freed in function
sequence_store(). Make sure that 'msg' is freed in all exit paths.
Detected by Coverity CID 1309667.
Signed-off-by: NChristian Engelmayer <cengelma@gmx.at>
Signed-off-by: NOlof Johansson <olof@lixom.net>

With Chrome running on 64-bit ARM devices, add ARM64 to the list of
supported architectures.
Signed-off-by: NThierry Reding <treding@nvidia.com>
[olof; Fixed up due to addition of COMPILE_TEST]
Signed-off-by: NOlof Johansson <olof@lixom.net>

If the cros_ec_dev driver is built as a module, modalias information is
not filled so the module is not autoloaded. Add a platform device table
and use the MODULE_DEVICE_TABLE() macro to export that information in
the module so user-space can match the modalias uevent and autoload it.
Signed-off-by: NJavier Martinez Canillas <javier.martinez@collabora.co.uk>
Signed-off-by: NOlof Johansson <olof@lixom.net>

Since the verion of ACPI in Google BIOS does not enumerate the devices
in the LPC bus, the cros_ec_lpc driver resorts to DMI data to check if
a system is supported by the driver and autoload if built as a module.

Add information about the Google Pixel 2 to the DMI device table.
Signed-off-by: NJavier Martinez Canillas <javier.martinez@collabora.co.uk>
Signed-off-by: NOlof Johansson <olof@lixom.net>

Commit 6db07b63 ("mfd: cros_ec: Check result code from EC messages")
added a common cros_ec_check_result() function that can be used to check
the ec_msg->result for errors and warns about them.

Use the existing function instead of duplicating same check in the driver.
Signed-off-by: NJavier Martinez Canillas <javier.martinez@collabora.co.uk>
Reviewed-by: NGwendal Grignou <gwendal@chromium.org>
Tested-by: NGwendal Grignou <gwendal@chromium.org>
Signed-off-by: NOlof Johansson <olof@lixom.net>

The ChromeOS EC LPC and chardev drivers depend on CROS_EC_PROTO but
MFD_CROS_EC select CROS_EC_PROTO instead. Mixing select and depends
on is bad practice as it may lead to circular Kconfig dependencies.

Since the platform devices that are matched with these drivers are
registered by the ChromeOS EC mfd driver, they really depend on
MFD_CROS_EC. And because this config option selects CROS_EC_PROTO,
that dependency is met as well. So make the drivers to depend on
MFD_CROS_EC instead of CROS_EC_PROTO.
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NOlof Johansson <olof@lixom.net>

This reverts commit d12bbcd3 ("platform/chrome: Don't make
CHROME_PLATFORMS depends on X86 || ARM") since it was found to
not be the correct fix for the MFD_CROS_EC config unmet direct
dependencies warning.

The correct solution was to add the needed dependencies to the
MFD_CROS_EC symbol. Besides the revert, this patch extends the
CHROME_PLATFORMS symbol dependencies and adds || COMPILE_TEST
to allow drivers to have build coverage on other architectures.
Suggested-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NOlof Johansson <olof@lixom.net>

Since v3.0 kernel, acer-wmi driver auto detects internal 3G device and
provides the threeg rfkill interface. So the sysfs interface of 3G is
no longer necessary now.

The 7b8aca65 patch added kernel information log to remind to user space
for the threeg sysfs interface will be removed in 2012, then b58b9ffc patch
updated the interface removing time to 2014.

I think the 3 years lead time is enough for user space application to use
rfkill instead of sysfs interface to control 3G device.

This patch removes code about threeg sysfs interface. And it also removes the
unused interface sysfs that exposes which ACPI-WMI method used by acer-wmi
driver on the machine. The information is already exposed by the acer-wmi
initial log.

Cc: Carlos Corbacho <carlos@strangeworlds.co.uk>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
Cc: Dmitry Torokhov <dtor@mail.ru>
Cc: Corentin Chary <corentincj@iksaif.net>
Cc: Martin Kepplinger <martink@posteo.de>
Cc: Darren Hart <dvhart@infradead.org>
Signed-off-by: NLee, Chun-Yi <jlee@suse.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The %*ph specifier allows to dump small buffers in hex format. Let's use it
instead of passing direct values via stack.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NPaul Fox <pgf@laptop.org>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Toshiba laptops that feature WMI events for hotkeys were left unsupported
by the toshiba_acpi driver, however, commit a88bc06e ("toshiba_acpi:
Avoid registering input device on WMI event laptops") added hardware
support for such laptops, but the hotkeys are not handled there.

This driver adds support for hotkey monitoring on certain Toshiba laptops
that manage the hotkeys via WMI events instead of the Toshiba
Configuration Interface (TCI).

The toshiba_acpi driver and this one can co-exist, as this only takes
care of hotkeys, while the proper takes care of hardware related stuff.

Currently the driver is under the EXPERIMENTAL flag, as the keymap
and the notify function are incomplete (due to lack of hardware to test).
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

sony_nv_hotkeys_decode can return a negative value. real_ev is a u32 variable.
The check for real_ev > 0 is incorrect.

Use an intermediate ret variable.

The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1].

[1]: http://permalink.gmane.org/gmane.linux.kernel/2046107Signed-off-by: NAndrzej Hajda <a.hajda@samsung.com>
[dvhart: clarify commit msg, drop superfluous else block]
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should
be left enabled so as to allow them to work as expected during the
suspend-resume cycle, but doesn't guarantee that it will wake the system
from a suspended state, enable_irq_wake is recommended to be used for
the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly
introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc: Darren Hart <dvhart@infradead.org>
Cc: platform-driver-x86@vger.kernel.org
Signed-off-by: NSudeep Holla <sudeep.holla@arm.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Add charge control limit to the power supply subsystem of the Compal
platform driver.

This apparently was present in the original driver by Cezary Jackiewicz
at http://eko.one.pl/index.php?page=compal-laptop but it seems to have been
overlooked.

The Kconfig description is updated to reflect this addition. It now also
mentions the hwmon interface that was already present.
Signed-off-by: NRoald Frederickx <roald.frederickx@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Afters suspend/resume cycle with closed lid the kbd backlight level
is lost. This patch will will restore this value to last known level.
Signed-off-by: NOleksij Rempel <linux@rempel-privat.de>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The function toshiba_acpi_setup_keyboard currently has two u32 variables
used to store the Hotkey Event Type and the result of the HCI_SYSTEM_EVENT
query.

This patch removes those two variables, as we already have a global
variable named "hotkey_event_type" and the result of the HCI_SYSTEM_EVENT
query can be checked directly from the function.
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

This patch adds the 0x prefix to the values printed by such function, the
values are already being printed in hex, but without the prefix, causing
confusion, even though the file under Documentation/ABI clearly states
that hey are hex values.

Simply add the 0x prefix to avoid such confusion.
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The driver currently uses the hotkey enabling value of 0x09 to enable
hotkey events, but windows uses a different value (0x01).

All Toshiba laptops accept the following "hotkey" parameters:
0x01 - Enable hotkey and system events.
0x03 - Enable system events only.
0x09 - Enable hotkey events only.
0x0b - Disable (hotkey and system) events.

This patch changes the default hotkey enabling value from 0x09 to 0x01,
enabling both the hotkey and system events.
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Currently the driver has two functions enabling hotkeys support,
but these two functions can be merged into one.

This patch merges these two functions, moving some checks to the
*enable_hotkeys function, simplifying code in the process.
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Do not write initialize magic on systems that do not have
feature query 0xb. Fixes Bug #82451.

Redefine FEATURE_QUERY to align with 0xb and FEATURE2 with 0xd
for code clearity.

Add a new test function, hp_wmi_bios_2008_later() & simplify
hp_wmi_bios_2009_later(), which fixes a bug in cases where
an improper value is returned. Probably also fixes Bug #69131.

Add missing __init tag.
Signed-off-by: NKyle Evans <kvans32@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Commit a2b3471b ("toshiba_acpi: Use the Hotkey Event Type function
for keymap choosing") changed the *setup_keyboard function to query for
the Hotkey Event Type to help choose the correct keymap, but turns out
that here are certain Toshiba models out there not implementing this
feature, and thus, failing to continue the input device registration and
leaving such laptops without hotkey support.

This patch changes such check, and instead of returning an error if
the Hotkey Event Type is not present, we simply inform userspace about it,
changing the message printed from err to notice, making the function
responsible for registering the input device to continue.

This issue was found on a Toshiba Portege Z30-B, but there might be
some other models out there affected by this regression as well.

Cc: <stable@vger.kernel.org> # 4.1+
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

Commit e1a949c1 ("toshiba_acpi: Refactor *{get, set} functions return
value") made changes on the return type of the HCI/SCI functions, but a
typo on the USB Sleep and Music code is always reporting non existent
support for such feature.

This patch corrects the typo, changing an assignment to a comparison,
making the laptops with actual support for such feature to work again.
Signed-off-by: NAzael Avalos <coproscefalo@gmail.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The work performed by wmi_gtoa is equivalent to simply sprintf(out,
"%pUL", in), so one could replace its body by this. However, most
users feed the result directly as a %s argument to some other function
which also understands the %p extensions (they all ultimately use
vsnprintf), so we can eliminate some stack buffers and quite a bit of
code by just using %pUL directly.

In wmi_dev_uevent I'm not sure whether there's room for a
nul-terminator in env->buf, so I've just replaced wmi_gtoa with the
equivalent sprintf call.
Signed-off-by: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>