The intel_pmc_core driver is mostly used as a debugging driver for Intel
platforms that support SLPS0 (S0ix). But the driver may also be used to
communicate actions to the PMC in order to ensure transition to SLPS0 on
some systems and architectures. As such the driver should be built on all
platforms it supports. Indicate this in the Kconfig. Also update the list
of supported features.
Signed-off-by: NDavid E. Box <david.e.box@linux.intel.com>
Suggested-by: NMario Limonciello <mario.limonciello@dell.com>
Link: https://lore.kernel.org/r/20210319201844.3305399-1-david.e.box@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Fixes off-by-one bugs in the macro assignments for the crashlog control
bits. Was initially tested on emulation but bug revealed after testing on
silicon.

Fixes: 5ef9998c ("platform/x86: Intel PMT Crashlog capability driver")
Signed-off-by: NDavid E. Box <david.e.box@linux.intel.com>
Link: https://lore.kernel.org/r/20210317024455.3071477-2-david.e.box@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Initialize the struct resource in intel_pmt_dev_register to zero to avoid a
fault should the char *name field be non-zero.
Signed-off-by: NDavid E. Box <david.e.box@linux.intel.com>
Link: https://lore.kernel.org/r/20210317024455.3071477-1-david.e.box@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Stop reporting SW_DOCK events because this breaks suspend-on-lid-close.

SW_DOCK should only be reported for docking stations, but all the DSDTs in
my DSDT collection which use the intel-vbtn code, always seem to use this
for 2-in-1s / convertibles and set SW_DOCK=1 when in laptop-mode (in tandem
with setting SW_TABLET_MODE=0).

This causes userspace to think the laptop is docked to a port-replicator
and to disable suspend-on-lid-close, which is undesirable.

Map the dock events to KEY_IGNORE to avoid this broken SW_DOCK reporting.

Note this may theoretically cause us to stop reporting SW_DOCK on some
device where the 0xCA and 0xCB intel-vbtn events are actually used for
reporting docking to a classic docking-station / port-replicator but
I'm not aware of any such devices.

Also the most important thing is that we only report SW_DOCK when it
reliably reports being docked to a classic docking-station without any
false positives, which clearly is not the case here. If there is a
chance of reporting false positives then it is better to not report
SW_DOCK at all.

Cc: stable@vger.kernel.org
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321163513.72328-1-hdegoede@redhat.com

Cleanup create_attributes_level_sysfs_files():

1. There is no need to call sysfs_remove_file() on error, sysman_init()
will already call release_attributes_data() on failure which already does
this.

2. There is no need for the pr_debug() calls sysfs_create_file() should
never fail and if it does it will already complain about the problem
itself.

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-8-hdegoede@redhat.com

When either the attributes or the password interface is not found, then
unregister the 2 wmi drivers again and return -ENODEV from sysman_init().

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Reported-by: NAlexander Naumann <alexandernaumann@gmx.de>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-7-hdegoede@redhat.com

Cleanup sysman_init() error-exit handling:

1. There is no need for the fail_reset_bios and fail_authentication_kset
   eror-exit cases, these can be handled by release_attributes_data()

2. Rename all the labels from fail_what_failed, to err_what_to_cleanup
   this is the usual way to name these and avoids the need to rename
   them when extra steps are added.

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-6-hdegoede@redhat.com

platform/x86: dell-wmi-sysman: Fix release_attributes_data() getting called twice on init_bios_attributes() failure

All calls of init_bios_attributes() will result in a
goto fail_create_group if they fail, which calls
release_attributes_data().

So there is no need to call release_attributes_data() from
init_bios_attributes() on failure itself.

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-5-hdegoede@redhat.com

During some of the error-exit paths it is possible that
release_attributes_data() will get called multiple times,
which results in exit_foo_attributes() getting called multiple
times.

Make it safe to call exit_foo_attributes() multiple times,
avoiding double-free()s in this case.

Note that release_attributes_data() really should only be called
once during error-exit paths. This will be fixed in a separate patch
and it is good to have the exit_foo_attributes() functions modified
this way regardless.

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-4-hdegoede@redhat.com

It is possible for release_attributes_data() to get called when the
main_dir_kset has not been created yet, move the removal of the bios-reset
sysfs attr to under a if (main_dir_kset) check to avoid a NULL pointer
deref.

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Reported-by: NAlexander Naumann <alexandernaumann@gmx.de>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-3-hdegoede@redhat.com

On some system the WMI GUIDs used by dell-wmi-sysman are present but there
are no enum type attributes, this causes init_bios_attributes() to return
-ENODEV, after which sysman_init() does a "goto fail_create_group" and then
calls release_attributes_data().

release_attributes_data() calls kset_unregister(wmi_priv.main_dir_kset);
but before this commit it was missing a "wmi_priv.main_dir_kset = NULL;"
statement; and after calling release_attributes_data() the sysman_init()
error handling does this:

        if (wmi_priv.main_dir_kset) {
                kset_unregister(wmi_priv.main_dir_kset);
                wmi_priv.main_dir_kset = NULL;
        }

Which causes a second kset_unregister(wmi_priv.main_dir_kset), leading to
a double-free, which causes a crash.

Add the missing "wmi_priv.main_dir_kset = NULL;" statement to
release_attributes_data() to fix this double-free crash.

Fixes: e8a60aa7 ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems")
Cc: Divya Bharathi <Divya_Bharathi@dell.com>
Cc: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321115901.35072-2-hdegoede@redhat.com

Testing has shown that setting /sys/firmware/acpi/platform_profile to
"balanced" when /sys/bus/platform/devices/thinkpad_acpi/dytc_lapmode
reports 1, causes dytc_lapmode to get reset to 0 and then it becomes
stuck at 0 for aprox. 30 minutes even if the laptop is used on a lap.

Disabling CQL (when enabled) before issuing the DYTC_CMD_RESET to get
back to balanced mode and re-enabling it afterwards again, like the
code already does when switching to low-power / performance mode fixes
this.

Fixes: c3bfcd4c ("platform/x86: thinkpad_acpi: Add platform profile support")
Tested-by: NMark Pearson <markpearson@lenovo.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210321113108.7069-1-hdegoede@redhat.com

On many recent ThinkPad laptops, there's a new LED next to the ESC key,
that indicates the FnLock status.
When the Fn+ESC combo is pressed, FnLock is toggled, which causes the
Media Key functionality to change, making it so that the media keys
either perform their media key function, or function as an F-key by
default. The Fn key can be used the access the alternate function at any
time.

With the current linux kernel, the LED doens't change state if you press
the Fn+ESC key combo. However, the media key functionality *does*
change. This is annoying, since the LED will stay on if it was on during
bootup, and it makes it hard to keep track what the current state of the
FnLock is.

This patch calls an ACPI function, that gets the current media key
state, when the Fn+ESC key combo is pressed. Through testing it was
discovered that this function causes the LED to update correctly to
reflect the current state when this function is called.

The relevant ACPI calls are the following:
\_SB_.PCI0.LPC0.EC0_.HKEY.GMKS: Get media key state, returns 0x603 if the FnLock mode is enabled, and 0x602 if it's disabled.
\_SB_.PCI0.LPC0.EC0_.HKEY.SMKS: Set media key state, sending a 1 will enable FnLock mode, and a 0 will disable it.

Relevant discussion:
https://bugzilla.kernel.org/show_bug.cgi?id=207841
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1881015Signed-off-by: NEsteve Varela Colominas <esteve.varela@gmail.com>
Link: https://lore.kernel.org/r/20210315195823.23212-1-esteve.varela@gmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Lenovo platforms with DYTC versions earlier than version 5 don't set
the lapmode interface correctly, causing issues with thermald on
older platforms.

Add checking to only create the dytc_lapmode interface for version
5 and later.

Fixes: 1ac09656 ("platform/x86: thinkpad_acpi: Add palm sensor support")
Signed-off-by: NMark Pearson <markpearson@lenovo.com>
Link: https://lore.kernel.org/r/20210311174843.3161-1-markpearson@lenovo.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Like a few other system the Lenovo ThinkPad X1 Tablet Gen 2 miss the
HEBC method, which prevent the power button from working. Add a quirk
to enable the button array on this system family and fix the power
button.
Signed-off-by: NAlban Bedel <albeu@free.fr>
Tested-by: NAlexander Kobel <a-kobel@a-kobel.de>
Link: https://lore.kernel.org/r/20210222141559.3775-1-albeu@free.frSigned-off-by: NHans de Goede <hdegoede@redhat.com>

The ACPI_PLATFORM_PROFILE option essentially provides a library and not
really an independent module. Thus it seems to be more user-friendly to
hide this option and simply make drivers depending on it select it.
Suggested-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NMaximilian Luz <luzmaximilian@gmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

SFI-based platforms are gone. So does this framework.

This removes mention of SFI through the drivers and other code as well.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Increasing the virtual timeout time to account for scenarios
that may require more time, like DisplayPort Multi-Stream Transport
(DP MST), where the disconnect time can be extended longer than
usual.

The recommended timeout range is 5-10 seconds, of which
we will take the lower bound.
Signed-off-by: NCasey Bowman <casey.g.bowman@intel.com>
Acked-by: NHeikki Krogerus <heikki.krogerus@linux.intel.com>
Acked-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210210192041.17022-1-casey.g.bowman@intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Neither original structure nor platform_data is declared with const.
Drop mistakenly added const when assing platform_data.

Fixes: a507e5d9 ("platform/x86: intel_scu_wdt: Get rid of custom x86 model comparison")
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20210204150508.62659-1-andriy.shevchenko@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Neither original structure nor platform_data is declared with const.
Drop mistakenly added const when assing platform_data.

Fixes: a507e5d9 ("platform/x86: intel_scu_wdt: Get rid of custom x86 model comparison")
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20210204150508.62659-1-andriy.shevchenko@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

LED class support is needed by the ideapad-laptop module to compile after
the referenced commit. Add missing NEW_LEDS and LEDS_CLASS to Kconfig.

Fixes: 503325f8 ("platform/x86: ideapad-laptop: add keyboard backlight control support")
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210204142010.356675-1-pobrn@protonmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

ACPI_EXCEPTION is only intended for internal use by the ACPICA code.

ACPI_EXCEPTION was being used here to change the apci_status code into
something human-readable, use acpi_format_exception() for this instead,
which is the proper way to do this outside of ACPICA.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210204140205.268344-1-hdegoede@redhat.com

With the #if IS_ENABLED(CONFIG_ACPI_PLATFORM_PROFILE), we get the following
errors when thinkpad_acpi is builtin while CONFIG_ACPI_PLATFORM_PROFILE=m :

drivers/platform/x86/thinkpad_acpi.c:10186: undefined reference to `platform_profile_notify'
drivers/platform/x86/thinkpad_acpi.c:10226: undefined reference to `platform_profile_register'
drivers/platform/x86/thinkpad_acpi.c:10246: undefined reference to `platform_profile_remove'

This could be fixed by changing the IS_ENABLED to IS_REACHABLE, but
I believe that it is better to just switch to using depends on.

Using depends on ensures that platform-profile support is always
available when thinkpad_acpi is build, hopefully leading to less
confusing bug-reports about it sometimes not working.

Cc: Mark Pearson <markpearson@lenovo.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210204140158.268289-2-hdegoede@redhat.com

Some of the new dytc handling functions are not marked static, even though
they are only used internally.

Mark these static, fixing the following compiler warnings:

drivers/platform/x86/thinkpad_acpi.c:10081:5: warning: no previous prototype for 'dytc_profile_get' [-Wmissing-prototypes]
drivers/platform/x86/thinkpad_acpi.c:10095:5: warning: no previous prototype for 'dytc_cql_command' [-Wmissing-prototypes]
drivers/platform/x86/thinkpad_acpi.c:10133:5: warning: no previous prototype for 'dytc_profile_set' [-Wmissing-prototypes]

Cc: Mark Pearson <markpearson@lenovo.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210204140158.268289-1-hdegoede@redhat.com

Explicitly check the status rather then relying on output.pointer staying
NULL on an error.
Reported-by: NMaximilian Luz <luzmaximilian@gmail.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210204121931.131617-1-hdegoede@redhat.com

A user without a Dell system doesn't need to pick any of these
drivers.

Users with a Dell system can enable this submenu and all drivers
behind it will be enabled.
Suggested-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NMario Limonciello <mario.limonciello@dell.com>
Link: https://lore.kernel.org/r/20210203195832.2950605-1-mario.limonciello@dell.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Certain models have a so-called "always on USB charging" feature,
which enables USB charging even when the computer is turned off
or suspended, and which may be controlled/queried using the
SALS/HALS ACPI methods. Expose this functionality via a new device
attribute (usb_charging). Tested on: Lenovo YOGA 520-14IKB 80X8
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-28-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

On certain models it is possible to control/query the
keyboard backlight via the SALS/HALS ACPI methods. Add
support for that, and register an LED class device to expose
this functionality. Tested on: Lenovo YOGA 520-14IKB 80X8
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-27-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Consumers can determine if the value of an attribute changed much more
easily if changes are broadcast using sysfs_notify(), so utilize it.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-26-pobrn@protonmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Fix all checkpatch warnings. Reorder variable definitions from
longest to shortest. Add more whitespaces for better readability.
Rename variables named `ret` to `err` where appropriate. Reorder
sysfs attributes show/store callbacks and the `ideapad_attributes`
array in lexicographic order. And other minor formatting changes.
No significant functional changes are intended.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-25-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

The touchpad can be enabled/disabled via this attribute on a Lenovo
Yoga 520-14IKB. Allow writing as it provides legitimate functionality.

This reverts commit 7f363145.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-24-pobrn@protonmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Minor formatting changes. Use seq_puts() where possible.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-23-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Remove conservation mode reporting since it is already reported via
the appropriate device attribute, and its state can be deduced from
the value of GBMD. Add the return value of the GBMD and HALS ACPI
methods to the output. Use seq_puts() where possible.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-22-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Bit 30 of _CFG is set if the device has a touchpad, check that
in is_visible() for the touchpad attribute. Show 'touchpad', if
supported, in the list of capabilities in the 'cfg' debugfs file.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-21-pobrn@protonmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Bit 9 in the return value of the HALS ACPI method is set if
Fn-lock is supported. Change ideapad_check_features() to check it.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-20-pobrn@protonmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Store the supported features in the driver private
data, and modify the is_visible() callback to use it,
and create ideapad_check_features() to populate it.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-19-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Create dedicated helper functions for accessing the main ACPI methods:
GBMD, SMBC, HALS, SALS; and utilize them. Use `unsigned long` consistently
in every ACPI helper wherever possible. Change names to better express
purpose. Do not assign values to output parameters in case of failure.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-18-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Group and rename constants depending on which ACPI interface
they pertain to, and rename CFG_X constants to CFG_CAP_X.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-17-pobrn@protonmail.comSigned-off-by: NHans de Goede <hdegoede@redhat.com>

Do not handle zero length buffer separately. Use kstrtouint() instead
of sscanf(). Use kstrtobool() in store_ideapad_cam(). These
introduce minor ABI changes, but it is expected that no users rely
on the previous behavior. Thus the change is deemed justifed.
Additionally, use `!!` to convert to `int` and use the "%d" format
specifier in sysfs_emit() for boolean-like attributes.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-16-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Consumers can differentiate an error from a successful read much more
easily if the read() call fails with an appropriate errno instead of
returning a magic string like "-1". This introduces an ABI change, but
not many users are expected to be relying on the previous behavior,
and this change makes this module conforming to the standard behavior
that sysfs attribute show/store callbacks return an appropriate
errno in case of failure. Thus the ABI breakage is deemed justified.
Signed-off-by: NBarnabás Pőcze <pobrn@protonmail.com>
Link: https://lore.kernel.org/r/20210203215403.290792-15-pobrn@protonmail.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>