Do not touch keyboard backlight unless explicitly passed a module
parameter.  In this way we won't make wrong assumptions about what are
good default values since they actually are different from model to
model.

The only side effect is that we won't know what is the current value
until set via the sysfs attributes.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NDaniel Serpell <daniel.serpell@gmail.com>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

Return -1 in the error handling case instead of 0, as done elsewhere in
this function.
Signed-off-by: NWei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

It is quite some time that this one has been deprecated.
Get rid of it.

Should some really important user be overseen, it may be reverted and
the userspace program worked on first, but it is time to do something
to get rid of this old stuff...
Signed-off-by: NThomas Renninger <trenn@suse.de>
Acked-by: NMatthew Garrett <matthew.garrett@nebula.com>
Acked-by: NHenrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

SVS151290S series uses handle 0x0163 for keyboard backlight and 0x015B for the graphics switch.
Signed-off-by: NArthur Wirski <awirski@gmail.com>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

Found with a network device in QEMU/KVM guest not working anymore.

Bisected to commit c13085e5
    ACPICA: Resource Mgr: Prevent infinite loops in resource walks

That commit will check acpi_resource length strictly which causes
acpi_set_current_resources to return failure and IRQ for PCI
devices is not set properly.

Set length for all those TYPE_END_TAG acpi_resources.

[rjw: Changelog]
Bisected-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

We need to decrement "i" first because the current "i" was not allocated
succesfully.  Also we should go free the way down to zero to avoid a
leak.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

Z series and other recent models have 0x14? for lid and keyboard
backlight.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

The call to handlers 0x124 and 0x135 (rfkill control) seems to take a
bitmask to control various states of the device. For our rfkill we need
a fully on/off. SVZ1311Z9R/X's LTE modem needs more bits up.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=47751Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>
Cc: stable@kernel.org

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

The second argument of ACPI driver .remove() operation is only used
by the ACPI processor driver and the value passed to that driver
through it is always available from the given struct acpi_device
object's removal_type field.  For this reason, the second ACPI driver
.remove() argument is in fact useless, so drop it.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NJiang Liu <jiang.liu@huawei.com>
Acked-by: NToshi Kani <toshi.kani@hp.com>
Acked-by: NYinghai Lu <yinghai@kernel.org>

SN06 in some cases returns an Integer instead of a buffer. While the
code handling the return value was trying to cope with the difference,
the memcpy call was not making any difference between the two types of
acpi_object union. This regression was introduced in 3.5.
While there also rework the return value logic to improve readability.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=48671
Cc: <stable@vger.kernel.org>
Cc: Fabrizio Narni <shibotto@gmail.com>
Cc: <mus.svz@gmail.com>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <matthew.garrett@nebula.com>

According to compiler warnings, quite some suspend/resume functions
in platform x86 drivers are not used for CONFIG_PM_SLEEP unset, so
add #ifdefs to prevent them from being built in that case.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>

Make the sony-laptop driver define its PM callbacks through
a struct dev_pm_ops object rather than by using legacy PM hooks
in struct acpi_device_ops.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>

None of the drivers implementing the ACPI device suspend callback
uses the pm_message_t argument of it, so this argument may be dropped
entirely from that callback.  This will simplify switching the ACPI
bus type to PM handling based on struct dev_pm_ops.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>

Since bab7084c, find_snc_handle
returns -EINVAL, not -1.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

This needs to be signed to handle negative error codes.
Remove a redundant check, read_limits is always called with a valid
handle.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

We made this an unsigned long and it causes a bug on 64 bit big endian
systems when we try to pass the value to sony_nc_int_call().

Also value has to be signed because validate() returns negative error
codes.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

SNC needs input devices so better have those ready before starting
handle events.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Some models offer the option to store the limits on the battery
(firmware?).
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Some Vaios come with both integrated and discrete graphics, plus a
switch for choosing one of the two. When the switch position is changed,
a notification is generated.
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

This setting is stored in the EC and available across reboots.

[malattia@linux.it: group function specific variables in a struct, use
kstrtoul]
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Also make the initialization function return a value for consistency
with all the other setup functions.
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Add support for handle 0x0143 (Vaio SA/SB/SC, CA/CB) and rework the code
to be hable to support different handles for the keyboard backlight
function.

[malattia@linux.it: group function specific variables in a struct]
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

A few models offer the chance to set whether to resume from S3 and/or S4
when opening the lid.

[malattia@linux.it: create three sysfs files for S3/4/5 rather than
using a single one accepting a bitmask. Support S5 since the DSDT
exports it. Use a struct to hold all the related values, caching of the
current status value rather than re-reading all the time in the sysfs
show function.]
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

All handles must be greater than 0, also return more meaningful error
codes on invalid conditions.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

[malattia@linux.it: support string based profiles names]
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Allows limiting the maximum battery charge level to a selectable value
(100%, 80% and 50%).

[malattia@linux.it: group function specific variables in a struct, use
kstrtoul. Allow 0 to 100 values into sysfs files rounding to the actual
limit.]
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Loop through the list of SNC handles and run the proper initialization
function for each of the known handles. Also return void in function
where we are not checking the return value anyway.
For notifications we also know which handle is linked to the event and
the code becomes simpler to read with a switch rather than using
convoluted ifs.

[malattia@linux.it: Code reworked to merge the initialization code
improvements and the notify callback changes. Modified the code paths to
allow easier error exit paths.  Also fixed some missing break statements
and spelling mistakes.]
Signed-off-by: NMarco Chiappero <marco@absence.it>
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

This avoids surprises like echoing "enable" into a sysfs file and
finding that the feature was actually disabled.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

All calls into the SNC device methods have zero or one arguments and
return an integer or a buffer (some functions go as far as returning an
integer _or_ a buffer depending on the input parameter...).
This allows simplifying a couple of code paths and prepares the field
for other users of functions returning buffers.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

The goto target location would still try to free a buffer that was
never allocated.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

The SNC device on recent Vaio laptops also stores the soft status and
leaves it available after reboot. Use it and always set the last soft
and hard status on module load.

[malattia@linux.it: patch taken from a largely modified sony-laptop.c,
smaller modifications to the original code to simplify it]
Signed-off-by: NMattia Dongili <malattia@linux.it>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>

Fix scancodes returned by driver to match scancodes used to remap keys.

(Before the patch FN/E returned scancode 0x1B, but to remap scancode
0x14 had to be used).

The scancodes returned by the sony-laptop driver for function keys did not
match the scancodes used to remap keys.  Also, since the scancode was sent
to the input subsystem after the mapped keysym the /lib/udev/keymap
utility was confused about which scancode to report for which keysym.

This patch fixes the driver so the correct scancode is shown for each
key.  It also adds to the documentation a description of where to find
the scancodes.
Signed-off-by: NJohn Hughes <john@calva.com>
Acked-by: NDmitry Torokhov <dtor@mail.ru>
Signed-off-by: NMatthew Garrett <mjg@redhat.com>