platform/kernel/linux-rpi.git
2 years agopower: supply: da9150-fg: Remove unnecessary print function dev_err()
Yang Li [Tue, 1 Mar 2022 08:16:19 +0000 (16:16 +0800)]
power: supply: da9150-fg: Remove unnecessary print function dev_err()

The print function dev_err() is redundant because
platform_get_irq_byname() already prints an error.

Eliminate the follow coccicheck warning:
./drivers/power/supply/da9150-fg.c:524:2-9: line 524 is redundant
because platform_get_irq() already prints an error

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Reviewed-by: Adam Thomson <Adam.Thomson.Opensource@diasemi.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: fix a handful of spelling mistakes
Colin Ian King [Wed, 2 Mar 2022 13:08:32 +0000 (13:08 +0000)]
power: supply: ab8500: fix a handful of spelling mistakes

There are a few spelling mistakes in comments and in a dev_err
error message. Fix them.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_fg: Account for line impedance
Linus Walleij [Tue, 1 Mar 2022 12:42:54 +0000 (13:42 +0100)]
power: supply: ab8500_fg: Account for line impedance

We add the line impedance to the inner resistance determined
from the battery voltage or other means to improve the
capacity estimations.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agodt-bindings: power: supply: ab8500_fg: Add line impedance
Linus Walleij [Tue, 1 Mar 2022 12:42:53 +0000 (13:42 +0100)]
dt-bindings: power: supply: ab8500_fg: Add line impedance

To improve the inner resistance measurement of the battery we need
to account for the line impedance of the connector to the battery.

Cc: devicetree@vger.kernel.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp20x_usb_power: fix platform_get_irq.cocci warnings
Yihao Han [Wed, 2 Mar 2022 02:34:51 +0000 (18:34 -0800)]
power: supply: axp20x_usb_power: fix platform_get_irq.cocci warnings

Remove dev_err() messages after platform_get_irq*() failures.
platform_get_irq() already prints an error.

Generated by: scripts/coccinelle/api/platform_get_irq.cocci

Signed-off-by: Yihao Han <hanyihao@vivo.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp20x_ac_power: fix platform_get_irq.cocci warning
Yihao Han [Thu, 3 Mar 2022 12:27:19 +0000 (04:27 -0800)]
power: supply: axp20x_ac_power: fix platform_get_irq.cocci warning

Remove dev_err() messages after platform_get_irq*() failures.
platform_get_irq() already prints an error.

Generated by: scripts/coccinelle/api/platform_get_irq.cocci

Signed-off-by: Yihao Han <hanyihao@vivo.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: wm8350-power: Add missing free in free_charger_irq
Jiasheng Jiang [Thu, 3 Mar 2022 09:43:22 +0000 (17:43 +0800)]
power: supply: wm8350-power: Add missing free in free_charger_irq

In free_charger_irq(), there is no free for 'WM8350_IRQ_CHG_FAST_RDY'.
Therefore, it should be better to add it in order to avoid the memory leak.

Fixes: 14431aa0c5a4 ("power_supply: Add support for WM8350 PMU")
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: wm8350-power: Handle error for wm8350_register_irq
Jiasheng Jiang [Fri, 4 Mar 2022 01:57:51 +0000 (09:57 +0800)]
power: supply: wm8350-power: Handle error for wm8350_register_irq

As the potential failure of the wm8350_register_irq(),
it should be better to check it and return error if fails.
Also, use 'free_' in order to avoid same code.

Fixes: 14431aa0c5a4 ("power_supply: Add support for WM8350 PMU")
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: Static data for Samsung batteries
Linus Walleij [Fri, 25 Feb 2022 23:28:00 +0000 (00:28 +0100)]
power: supply: Static data for Samsung batteries

If we detect a Samsung SDI battery, we return a static
struct power_supply_battery_info and avoid looking further.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_fg: Use VBAT-to-Ri if possible
Linus Walleij [Fri, 25 Feb 2022 23:27:59 +0000 (00:27 +0100)]
power: supply: ab8500_fg: Use VBAT-to-Ri if possible

Augment the AB8500 fuel gauge to use the VBAT-to-Ri method of
estimating the internal resistance if possible. Else fall back
to using the temperature-to-Ri or just the default Ri.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: Support VBAT-to-Ri lookup tables
Linus Walleij [Fri, 25 Feb 2022 23:27:58 +0000 (00:27 +0100)]
power: supply: Support VBAT-to-Ri lookup tables

In Samsung devices, the method used to compensate for temperature,
age, load etc is by way of VBAT to Ri tables, which correlates the
battery voltage under load (VBAT) to an internal resistance (Ri).

Using this Ri and a measurement of the current out of the battery
(IBAT) the open circuit voltage (OCV) can be calculated as:

  OCV = VBAT - (Ri * IBAT)

The details are described in comments to struct
power_supply_battery_info in the commit.

Since not all batteries supply this VBAT-to-Ri data, the fallback
method to use the temperature-to-Ri lookup table can also be used
as a fallback.

Add two helper functions to check if we have the tables needed for
using power_supply_vbat2ri() or power_supply_temp2resist_simple()
respectively, so capacity estimation code can choose which one
to employ.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Standardize BTI resistance
Linus Walleij [Fri, 25 Feb 2022 23:27:57 +0000 (00:27 +0100)]
power: supply: ab8500: Standardize BTI resistance

The Battery Type Indicator (BTI) resistor is a resistor mounted
between a special terminal on the battery and ground. By sending
a fixed current (such as 7mA) through this resistor and measuring
the voltage over it, the resistance can be determined, and this
verifies the battery type.

Typical side view of the battery:

  o     o     o
 GND   BTI   +3.8V

Typical example of the electrical layout:

  +3.8 V   BTI
    |       |
    | +     |
 _______   [ ] 7kOhm
   ___      |
    |       |
    |       |
   GND     GND

By verifying this resistance before attempting to charge the
battery we add an additional level of security.

In some systems this is used for plug-and-play of batteries with
different capacity. In other cases, this is merely used to verify
that the right type of battery is connected, if several batteries
have the same physical shape and can be plugged into the same
slot. Sometimes this is just a surplus security mechanism.

Nokia and Samsung among many other vendors are known to use these
BTI resistors.

Add the BTI properties to struct power_supply_battery_info and
switch the AB8500 charger code over to using it.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Standardize alert mode charging
Linus Walleij [Fri, 25 Feb 2022 23:27:56 +0000 (00:27 +0100)]
power: supply: ab8500: Standardize alert mode charging

The AB8500 code is using a special current and voltage setting
when the battery is in "alert mode", i.e. when it is starting
to go outside normal operating conditions so it is too
cold or too hot. This makes sense as a way for the charging
algorithm to deal with hostile environments.

Add the needed members to the struct power_supply_battery_info,
and switch the AB8500 charging code over to using this.

Reviewed-by: Matti Vaittineen <matti.vaittinen@fi.rohmeurope.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Standardize maintenance charging
Linus Walleij [Fri, 25 Feb 2022 23:27:55 +0000 (00:27 +0100)]
power: supply: ab8500: Standardize maintenance charging

Maintenance charging is the phase of keeping up the charge
after the battery has charged fully using CC/CV charging.

This can be done in many successive phases and is usually
done with a slightly lower constant voltage than CV, and
a slightly lower allowed current.

Add an array of maintenance charging points each with a
current, voltage and safety timer, and add helper functions
to use these. Migrate the AB8500 code over.

This is used in several Samsung products using the AB8500
and these batteries and their complete parameters will
be added later as full examples, but the default battery
in the AB8500 code serves as a reasonable example so far.

Reviewed-by: Matti Vaittinen <matti.vaittinen@fi.rohmeurope.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Delay applying charge_type changes when OTG 5V Vbus...
Hans de Goede [Sat, 12 Feb 2022 16:48:17 +0000 (17:48 +0100)]
power: supply: bq24190_charger: Delay applying charge_type changes when OTG 5V Vbus boost is on

Recently userspace has started switching power_supply class devices with
a charge_type psy-property between fast and trickle charge mode, see:
https://gitlab.freedesktop.org/hadess/power-profiles-daemon/-/issues/85

Before this patch bq24190_charger_set_charge_type() would unconditionally
write charging or none to the BQ24190_REG_POC_CHG_CONFIG bits, replacing
the otg setting of those bits when the OTG 5V Vbus boost converter was on,
turning the 5V Vbus off, removing the power from any attached peripherals.

This fixes this by keeping track of otg_vbus_enabled and the requested
charger_type settings and when otg_vbus_enabled is true, delay applying
the charger_type until the 5V boost converter is turned off.

Cc: Bastien Nocera <hadess@hadess.net>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Fix bq24190_vbus_is_enabled() wrong false return
Hans de Goede [Sat, 12 Feb 2022 16:48:16 +0000 (17:48 +0100)]
power: supply: bq24190_charger: Fix bq24190_vbus_is_enabled() wrong false return

The datasheet says that the BQ24190_REG_POC_CHG_CONFIG bits can
have a value of either 10(0x2) or 11(0x3) for OTG (5V boost regulator)
mode.

Sofar bq24190_vbus_is_enabled() was only checking for 10 but some BIOS-es
uses 11 when enabling the regulator at boot.

Make bq24190_vbus_is_enabled() also check for 11 so that it does not
wrongly returns false when the bits are set to 11.

Fixes: 66b6bef2c4e0 ("power: supply: bq24190_charger: Export 5V boost converter as regulator")
Cc: Bastien Nocera <hadess@hadess.net>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Use acpi_quirk_skip_acpi_ac_and_battery()
Hans de Goede [Thu, 24 Feb 2022 22:28:05 +0000 (23:28 +0100)]
power: supply: axp288_fuel_gauge: Use acpi_quirk_skip_acpi_ac_and_battery()

Normally the native AXP288 fg/charger drivers are preferred but one some
devices the ACPI drivers should be used instead.

The ACPI battery/ac drivers use the acpi_quirk_skip_acpi_ac_and_battery()
helper to determine if they should skip loading because native fuel-gauge/
charger drivers like the AXP288 drivers will be used.

The new acpi_quirk_skip_acpi_ac_and_battery() helper includes a list of
exceptions for boards where the ACPI drivers should be used instead.

Use this new helper to avoid loading on such boards. Note this requires
adding a Kconfig dependency on ACPI, this is not a problem because ACPI
should be enabled on all boards with an AXP288 PMIC anyways.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_charger: Use acpi_quirk_skip_acpi_ac_and_battery()
Hans de Goede [Thu, 24 Feb 2022 22:28:04 +0000 (23:28 +0100)]
power: supply: axp288_charger: Use acpi_quirk_skip_acpi_ac_and_battery()

Normally the native AXP288 fg/charger drivers are preferred but one some
devices the ACPI drivers should be used instead.

The ACPI battery/ac drivers use the acpi_quirk_skip_acpi_ac_and_battery()
helper to determine if they should skip loading because native fuel-gauge/
charger drivers like the AXP288 drivers will be used.

The new acpi_quirk_skip_acpi_ac_and_battery() helper includes a list of
exceptions for boards where the ACPI drivers should be used instead.

Use this new helper to avoid loading on such boards. Note this requires
adding a Kconfig dependency on ACPI, this is not a problem because ACPI
should be enabled on all boards with an AXP288 PMIC anyways.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: max8997_charger: Use devm_work_autocancel()
Christophe JAILLET [Sun, 13 Feb 2022 11:19:31 +0000 (12:19 +0100)]
power: supply: max8997_charger: Use devm_work_autocancel()

Use devm_work_autocancel() instead of hand writing it.
It saves a few lines of code.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: max14656: Use devm_work_autocancel()
Christophe JAILLET [Sun, 13 Feb 2022 11:54:56 +0000 (12:54 +0100)]
power: supply: max14656: Use devm_work_autocancel()

Use devm_delayed_work_autocancel() instead of hand writing it.
It saves a few lines of code.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: da9150-fg: Use devm_delayed_work_autocancel()
Christophe JAILLET [Sun, 13 Feb 2022 17:55:05 +0000 (18:55 +0100)]
power: supply: da9150-fg: Use devm_delayed_work_autocancel()

This driver only uses managed resources, except for the delayed work, if
it is used.

Use devm_delayed_work_autocancel() to also manage the delayed work.
The error handling path of the probe and the remove function can both be
removed.

This saves a few lines of code.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Adam Thomson <Adam.Thomson.Opensource@diasemi.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: sbs-charger: Don't cancel work that is not initialized
Christophe JAILLET [Sun, 13 Feb 2022 17:07:03 +0000 (18:07 +0100)]
power: supply: sbs-charger: Don't cancel work that is not initialized

This driver can use an interrupt or polling in order get the charger's
status.

When using polling, a delayed work is used.

However, the remove() function unconditionally call
cancel_delayed_work_sync(), even if the delayed work is not used and is not
initialized.

In order to fix it, use devm_delayed_work_autocancel() and remove the now
useless remove() function.

Fixes: feb583e37f8a ("power: supply: add sbs-charger driver")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoABI: testing: sysfs-class-power: Adds "Long Life" entry
Ricardo Rivera-Matos [Tue, 15 Feb 2022 00:07:58 +0000 (18:07 -0600)]
ABI: testing: sysfs-class-power: Adds "Long Life" entry

Adds a brief desciption of the "Long Life" charge type and adds "Long Life"
to the list of valid values.

Signed-off-by: Ricardo Rivera-Matos <rriveram@opensource.cirrus.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25980: Implements POWER_SUPPLY_CHARGE_TYPE_BYPASS
Ricardo Rivera-Matos [Tue, 15 Feb 2022 00:07:57 +0000 (18:07 -0600)]
power: supply: bq25980: Implements POWER_SUPPLY_CHARGE_TYPE_BYPASS

This patch remaps the bypass operation from POWER_SUPPLY_CHARGE_TYPE_FAST
to POWER_SUPPLY_CHARGE_TYPE_BYPASS.

Signed-off-by: Ricardo Rivera-Matos <rriveram@opensource.cirrus.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: Introduces bypass charging property
Ricardo Rivera-Matos [Tue, 15 Feb 2022 00:07:56 +0000 (18:07 -0600)]
power: supply: Introduces bypass charging property

Adds a POWER_SUPPLY_CHARGE_TYPE_BYPASS option to the POWER_SUPPLY_PROP_CHARGE_TYPE
property to facilitate bypass charging operation.

In bypass charging operation, the charger bypasses the charging path around the
integrated converter allowing for a "smart" wall adaptor to perform the power
conversion externally.

This operational mode is critical for the USB PPS standard of power adaptors and is
becoming a common feature in modern charging ICs such as:

- BQ25980
- BQ25975
- BQ25960
- LN8000
- LN8410

Signed-off-by: Ricardo Rivera-Matos <rriveram@opensource.cirrus.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: max17042_battery: Use devm_work_autocancel()
Christophe JAILLET [Fri, 18 Feb 2022 17:37:37 +0000 (18:37 +0100)]
power: supply: max17042_battery: Use devm_work_autocancel()

Use devm_work_autocancel() instead of hand-writing it.
This saves a few lines of code.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: Use an rbtree rather than flat register cache
Mark Brown [Tue, 22 Feb 2022 21:43:31 +0000 (21:43 +0000)]
power: supply: Use an rbtree rather than flat register cache

The smb347 has a very sparse register map (the maximum register is 0x3f but
less than 10% of the possible registers appear to be defined) and doesn't
have any hardware defaults specified so the sparser data structure of an
rbtree is a better fit for it's needs than a flat cache. Since it uses I2C
for the control interface there is no performance concern with the slightly
more involved code so let's convert it.

This will mean we avoid any issues created by assuming that any previously
unaccessed registers hold a value that doesn't match what's in the hardware
(eg, an _update_bits() suppressing a write).

Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288-charger: Set Vhold to 4.4V
Hans de Goede [Tue, 8 Feb 2022 12:51:47 +0000 (13:51 +0100)]
power: supply: axp288-charger: Set Vhold to 4.4V

The AXP288's recommended and factory default Vhold value (minimum
input voltage below which the input current draw will be reduced)
is 4.4V. This lines up with other charger IC's such as the TI
bq2419x/bq2429x series which use 4.36V or 4.44V.

For some reason some BIOS-es initialize Vhold to 4.6V or even 4.7V
which combined with the typical voltage drop over typically low
wire gauge micro-USB cables leads to the input-current getting
capped below 1A (with a 2A capable dedicated charger) based on Vhold.

This leads to slow charging, or even to the device slowly discharging
if the device is in heavy use.

As the Linux AXP288 drivers use the builtin BC1.2 charger detection
and send the input-current-limit according to the detected charger
there really is no reason not to use the recommended 4.4V Vhold.

Set Vhold to 4.4V to fix the slow charging issue on various devices.

There is one exception, the special-case of the HP X2 2-in-1s which
combine this BC1.2 capable PMIC with a Type-C port and a 5V/3A factory
provided charger with a Type-C plug which does not do BC1.2. These
have their input-current-limit hardcoded to 3A (like under Windows)
and use a higher Vhold on purpose to limit the current when used
with other chargers. To avoid touching Vhold on these HP X2 laptops
the code setting Vhold is added to an else branch of the if checking
for these models.

Note this also fixes the sofar unused VBUS_ISPOUT_VHOLD_SET_MASK
define, which was wrong.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Remove unneeded variable
Changcheng Deng [Fri, 7 Jan 2022 02:26:02 +0000 (02:26 +0000)]
power: supply: ab8500: Remove unneeded variable

Remove unneeded variable used to store return value.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Changcheng Deng <deng.changcheng@zte.com.cn>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ltc2941: clean up error messages
Michał Mirosław [Tue, 11 Jan 2022 02:32:41 +0000 (03:32 +0100)]
power: supply: ltc2941: clean up error messages

Replace dev_err() with dev_err_probe() in probe() and extend register
access failure messages. dev_err()s in _reset() are removed as they
are redundant: register access wrappers already log the error.

Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ltc2941: simplify Qlsb calculation
Michał Mirosław [Tue, 11 Jan 2022 02:32:40 +0000 (03:32 +0100)]
power: supply: ltc2941: simplify Qlsb calculation

Replace two divisions with a subtraction+shift for a small code size
improvement and less brackets.

Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_charger: Fix spelling typo
Hong Peng [Wed, 19 Jan 2022 03:53:01 +0000 (11:53 +0800)]
power: supply: ab8500_charger: Fix spelling typo

fix the comment typo: "interrupts", "structcure"

Signed-off-by: Hong Peng <elicec@foxmail.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Fix memory leak in ab8500_fg_sysfs_init
Miaoqian Lin [Mon, 24 Jan 2022 13:13:46 +0000 (13:13 +0000)]
power: supply: ab8500: Fix memory leak in ab8500_fg_sysfs_init

kobject_init_and_add() takes reference even when it fails.
According to the doc of kobject_init_and_add():

   If this function returns an error, kobject_put() must be called to
   properly clean up the memory associated with the object.

Fix memory leak by calling kobject_put().

Fixes: 8c0984e5a753 ("power: move power supply drivers to power/supply")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: PCHG: Use MKBP for device event handling
Daisuke Nojiri [Wed, 26 Jan 2022 18:04:10 +0000 (10:04 -0800)]
power: supply: PCHG: Use MKBP for device event handling

This change makes the PCHG driver receive device events through
MKBP protocol since CrOS EC switched to deliver all peripheral
charge events to the MKBP protocol. This will unify PCHG event
handling on X86 and ARM.

Signed-off-by: Daisuke Nojiri <dnojiri@chromium.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_charger: Fix VBAT interval check
Linus Walleij [Sat, 29 Jan 2022 00:49:25 +0000 (01:49 +0100)]
power: supply: ab8500_charger: Fix VBAT interval check

When using USB charging, the AB8500 charger is periodically
checking VBAT for a threshold at 3.8V.

This crashes badly, as the class_for_each_device() was passed
the wrong argument. I think this has maybe worked by chance
in the past because of how the structs were arranged but it
is leading to crashes now.

Fix this up and also switch to using microvolts for the
voltages like the rest of the code.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_charger: Restrict ADC retrieveal
Linus Walleij [Sat, 29 Jan 2022 00:49:24 +0000 (01:49 +0100)]
power: supply: ab8500_charger: Restrict ADC retrieveal

The AB8505 only has two ADC channels: the voltage and current
provided from VBUS (USB). It does not support AC charging at all.
Make sure we don't try to retrieve the non-existing channels.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_chargalg: Drop enable/disable sysfs
Linus Walleij [Sat, 29 Jan 2022 00:49:23 +0000 (01:49 +0100)]
power: supply: ab8500_chargalg: Drop enable/disable sysfs

There is a sysfs ABI to enable/disable charging of different
types (AC/USB).

Since we don't have any userspace for this code, this sits
unused and it is not used on production products either.

Drop this code.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_chargalg: Drop charging step
Linus Walleij [Sat, 29 Jan 2022 00:49:22 +0000 (01:49 +0100)]
power: supply: ab8500_chargalg: Drop charging step

There is a sysfs ABI to change the "charging step" of the
charger i.e. limit how much we charge from userspace.

Since we don't have any userspace for this code, this sits
unused and it is not used on production products either.

Drop this code.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_fg: Drop useless parameter
Linus Walleij [Sat, 29 Jan 2022 00:49:21 +0000 (01:49 +0100)]
power: supply: ab8500_fg: Drop useless parameter

All calls to ab8500_fg_calc_cap_discharge_voltage() require
compensation and pass true as the second argument so just drop
this argument.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_fg: Safeguard compensated voltage
Linus Walleij [Sat, 29 Jan 2022 00:49:20 +0000 (01:49 +0100)]
power: supply: ab8500_fg: Safeguard compensated voltage

In some cases when the platform is dissapating more than
500mA the voltage measurements and compensation will become
instable. Add a parameter to bail out of the voltage
measurement if this happens.

This code was found in a Samsung vendor tree.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_fg: Break out load compensated voltage
Linus Walleij [Sat, 29 Jan 2022 00:49:19 +0000 (01:49 +0100)]
power: supply: ab8500_fg: Break out load compensated voltage

Break out the part of the function providing the load compensated
capacity that provides the load compensated voltage and use
that to get the load compensated capacity.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500_fg: Break loop for measurement
Linus Walleij [Sat, 29 Jan 2022 00:49:18 +0000 (01:49 +0100)]
power: supply: ab8500_fg: Break loop for measurement

In the Samsung code tree we find that it can happen that this
measurement loop goes on for a long time, and it seems like a
good idea to break it after 70 iterations if it goes on for
too long.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Integrate thermal zone
Linus Walleij [Sat, 29 Jan 2022 00:49:17 +0000 (01:49 +0100)]
power: supply: ab8500: Integrate thermal zone

Instead of providing our own homebrewn thermal measurement
code for an NTC and passing tables, we put the NTC thermistor
into the device tree, create a passive thermal zone, and poll
this thermal zone for the temperature.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Swap max and overvoltage
Linus Walleij [Sat, 29 Jan 2022 00:49:16 +0000 (01:49 +0100)]
power: supply: ab8500: Swap max and overvoltage

We should terminate charging when we reach the voltage_max_design_uv
not overvoltage_limit_uv, this is an abuse of that struct member.

The overvoltage limit is actually not configurable on the AB8500,
it is fixed to 4.75 V so drop a comment about that in the code.

Fixes: 2a5f41830aad ("power: supply: ab8500: Standardize voltages")
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ab8500: Drop BATCTRL thermal mode
Linus Walleij [Sat, 29 Jan 2022 00:49:15 +0000 (01:49 +0100)]
power: supply: ab8500: Drop BATCTRL thermal mode

The BATCTRL mode reads the temperature of the battery by
enabling a certain probing current (7-20 mA) and then measure
the voltage of the NTC mounted inside the battery.

None of the AB8500 product or the reference designs use this
mode. What we use is the so-called BATTEMP mode which enables
an internal 230 kOhm pull-up to 1.8 V to the external NTC on
pin BatTemp (N16) and then measures the voltage over the NTC
using the ADC:

        1.8V (VTVOUT)
         |
        [ ] 230 kOhm
         |
BatTemp  +---------------- ADC
Pin N16  | _
         |/
        [/] NTC
       _/|
         |
        GND

Cut out the BATCTRL code to clear the forest and stop
maintaining code we can never test.

The current inducing method is still used to probe for the
battery type using the internal BTI (battery type indicator)
on the BatCtrl (C3) pin in a separate code path.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: core: Use device_property_string_array_count()
Andy Shevchenko [Wed, 2 Feb 2022 18:23:25 +0000 (20:23 +0200)]
power: supply: core: Use device_property_string_array_count()

Use device_property_string_array_count() to get number of strings
in a string array property.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: rt9455: Don't pass an error code in remove callback
Uwe Kleine-König [Sat, 5 Feb 2022 18:35:12 +0000 (19:35 +0100)]
power: supply: rt9455: Don't pass an error code in remove callback

When ret is not zero there were already one or two error messages emitted
about a problem (because rt9455_register_reset() emits a message in most
cases then). Passing on that error code to the i2c core only results in
another error message. Suppress that by returning zero unconditionally.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: core: Simplify hwmon memory allocation
Christophe JAILLET [Sat, 5 Feb 2022 16:40:57 +0000 (17:40 +0100)]
power: supply: core: Simplify hwmon memory allocation

Use devm_bitmap_zalloc() instead of hand writing it.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: ug3105_battery: Add driver for uPI uG3105 battery monitor
Hans de Goede [Sun, 6 Feb 2022 18:35:43 +0000 (19:35 +0100)]
power: supply: ug3105_battery: Add driver for uPI uG3105 battery monitor

Add a new battery driver for the uPI uG3105 battery monitor.

Note the uG3105 is not a full-featured autonomous fuel-gauge. Instead it
is expected to be use in combination with some always on microcontroller
reading its coulomb-counter before it can wrap (must be read every 400
seconds!).

Since Linux does not monitor coulomb-counter changes while the device is
off or suspended, the coulomb counter is not used atm.

So far this driver is only used on x86/ACPI (non devicetree) devs
(also note there is no of_match table). Therefor there is no devicetree
bindings documentation for this driver's "upisemi,rsns-microohm" property
since this is not used in actual devicetree files and the dt bindings
maintainers have requested properties with no actual dt users to
_not_ be added to the dt bindings.

The property's name has been chosen so that it should not need to be
changed if/when devicetree enumeration support gets added later, as it
mirrors "maxim,rsns-microohm" from the "maxim,max17042" bindings.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Disallow ccc_ireg and cvc_vreg to be higher then...
Hans de Goede [Sun, 6 Feb 2022 18:35:42 +0000 (19:35 +0100)]
power: supply: bq24190_charger: Disallow ccc_ireg and cvc_vreg to be higher then the fwnode values

If the fwnode data as parsed by power_supply_get_battery_info() provides
max values for ccc_ireg and cvc_vreg then do not allow the user to later
set these to higher values then those specified by the firmware,
otherwise the battery might get damaged.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Program charger with fwnode supplied ccc_ireg and...
Hans de Goede [Sun, 6 Feb 2022 18:35:41 +0000 (19:35 +0100)]
power: supply: bq24190_charger: Program charger with fwnode supplied ccc_ireg and cvc_vreg

So far the bq24190_charger driver has been relying on either the chips
default constant_charge_current_max_ua and constant_charge_voltage_max_uv
values, or on the BIOS or bootloader to program these for us.

This does not happen on all boards, causing e.g. the wrong (too low)
values to be used on Lenovo Yoga Tablet 2 830F/L and 1050F/L tablets.

If power_supply_get_battery_info() provides us with values for these
settings, then program the charger accordingly.

And if the user later overrides these values then save the user-values
so that these will be restored after a suspend/resume.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Store ichg-max and vreg-max in bq24190_dev_info
Hans de Goede [Sun, 6 Feb 2022 18:35:40 +0000 (19:35 +0100)]
power: supply: bq24190_charger: Store ichg-max and vreg-max in bq24190_dev_info

Store ichg-max and vreg-max in bq24190_dev_info once from
bq24190_get_config() and drop the bq24190_charger_get_current_max() and
bq24190_charger_get_voltage_max() helpers.

This is a preparation patch for honoring the
constant_charge_current_max_ua and constant_charge_voltage_max_uv
values from power_supply_get_battery_info().

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Always call power_supply_get_battery_info()
Hans de Goede [Sun, 6 Feb 2022 18:35:39 +0000 (19:35 +0100)]
power: supply: bq24190_charger: Always call power_supply_get_battery_info()

power_supply_get_battery_info() now also supports getting battery_info
on boards not using dt/of. Remove the of_node check. If neither of nor
other battery-info is present the function will fail making this change
a no-op in that case.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq24190_charger: Turn off 5V boost regulator on shutdown
Hans de Goede [Sun, 6 Feb 2022 18:35:38 +0000 (19:35 +0100)]
power: supply: bq24190_charger: Turn off 5V boost regulator on shutdown

Turn off the 5V boost regulator on shutdown, there are 3 reasons for
doing this:

1. It drains he battery if left on
2. If left on the device will not charge when plugged into a charger
3. If left on and the powered peripheral attached to a Type-C port is
   removed before the next boot, then the Type-C port-controller will
   see VBus being present while nothing is attached confusing the
   TCPM state-machine.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: core: Add support for generic fwnodes to power_supply_get_battery_info()
Hans de Goede [Sun, 6 Feb 2022 18:35:37 +0000 (19:35 +0100)]
power: supply: core: Add support for generic fwnodes to power_supply_get_battery_info()

Add support to power_supply_get_battery_info() to read the properties from
other fwnode types such as swnodes added by platform code on x86 devices.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: core: Use fwnode_property_*() in power_supply_get_battery_info()
Hans de Goede [Sun, 6 Feb 2022 18:35:36 +0000 (19:35 +0100)]
power: supply: core: Use fwnode_property_*() in power_supply_get_battery_info()

Switch power_supply_get_battery_info() over to use the generic
fwnode_property_*() property read functions. This is a preparation patch
for adding support for reading properties from other fwnode types such
as swnode properties added by platform code on x86 devices.

Note the parsing of the 2d matrix "ocv-capacity-table-%d" and
"resistance-temp-table" properties is not converted since this depends on
the raw of_get_property() accessor function of which there is no
fwnode_property_*() equivalent AFAICT. This means that these properties
will not be supported in swnodes for now.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Fix spelling mistake "resisitor" -> "resistor"
Colin Ian King [Wed, 2 Feb 2022 09:12:46 +0000 (09:12 +0000)]
power: supply: axp288_fuel_gauge: Fix spelling mistake "resisitor" -> "resistor"

There is a spelling mistake in a MODULE_PARM_DESC description. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoMerge tag 'psy-extcon-i2c-mfd-for-v5.18-signed' into psy-next
Sebastian Reichel [Tue, 1 Feb 2022 14:33:24 +0000 (15:33 +0100)]
Merge tag 'psy-extcon-i2c-mfd-for-v5.18-signed' into psy-next

Immutable branch between power-supply, mfd, i2c and extcon for for 5.18

This immutable branch fixes the charger setup on Xiaomi Mi Pad 2 and
Lenovo Yogabook, which requires updates to multiple drivers throughout
the tree.

Signed-off-by: Sebastian Reichel <sre@kernel.org>
2 years agoextcon: intel-cht-wc: Report RID_A for ACA adapters
Hans de Goede [Tue, 1 Feb 2022 13:07:06 +0000 (14:07 +0100)]
extcon: intel-cht-wc: Report RID_A for ACA adapters

Make cht_wc_extcon_get_id() report RID_A for ACA adapters, instead of
reporting ID_FLOAT.

According to the spec. we should read the USB-ID pin ADC value
to determine the resistance of the used pull-down resister and
then return RID_A / RID_B / RID_C based on this. But all "Accessory
Charger Adapter"s (ACAs) which users can actually buy always use
a combination of a charging port with one or more USB-A ports, so
they should always use a resistor indicating RID_A. But the spec
is hard to read / badly-worded so some of them actually indicate
they are a RID_B ACA even though they clearly are a RID_A ACA.

To workaround this simply always return INTEL_USB_RID_A, which
matches all the ACAs which users can actually buy.

Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoextcon: intel-cht-wc: Add support for registering a power_supply class-device
Hans de Goede [Tue, 1 Feb 2022 13:07:05 +0000 (14:07 +0100)]
extcon: intel-cht-wc: Add support for registering a power_supply class-device

The bq25890 used on the Yogabook YB1-X90 / -X91 models relies on
the extcon-driver's BC-1.2 charger detection, and the bq25890 driver
expect this info to be  available through a parent power_supply
class-device which models the detected charger (idem to how the Type-C
TCPM code registers a power_supply classdev for the connected charger).

Add support for registering the power_supply class-device expected
by this setup.

Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoextcon: intel-cht-wc: Refactor cht_wc_extcon_get_charger()
Hans de Goede [Tue, 1 Feb 2022 13:07:04 +0000 (14:07 +0100)]
extcon: intel-cht-wc: Refactor cht_wc_extcon_get_charger()

This is a preparation patch for adding support for registering
a power_supply class device.

Setting usbsrc to "CHT_WC_USBSRC_TYPE_SDP << CHT_WC_USBSRC_TYPE_SHIFT"
will make the following switch-case return EXTCON_CHG_USB_SDP
just as before, so there is no functional change.

Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoextcon: intel-cht-wc: Support devs with Micro-B / USB-2 only Type-C connectors
Hans de Goede [Tue, 1 Feb 2022 13:07:03 +0000 (14:07 +0100)]
extcon: intel-cht-wc: Support devs with Micro-B / USB-2 only Type-C connectors

So far the extcon-intel-cht-wc code has only been tested on devices with
a Type-C connector with USB-PD, USB3 (superspeed) and DP-altmode support
through a FUSB302 Type-C controller.

Some devices with the intel-cht-wc PMIC however come with an USB-micro-B
connector, or an USB-2 only Type-C connector without USB-PD.

Which device-model we are running on can be identified with the new
cht_wc_model intel_soc_pmic field. On models without a Type-C controller
the extcon code must control the Vbus 5V boost converter and the USB role
switch depending on the detected cable-type.

Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoextcon: intel-cht-wc: Use new cht_wc_model intel_soc_pmic field
Hans de Goede [Tue, 1 Feb 2022 13:07:02 +0000 (14:07 +0100)]
extcon: intel-cht-wc: Use new cht_wc_model intel_soc_pmic field

The CHT_WC_VBUS_GPIO_CTLO GPIO actually driving an external 5V Vboost
converter for Vbus depends on the board on which the Cherry Trail -
Whiskey Cove PMIC is actually used.

Since the information about the exact PMIC setup is necessary in other
places too, struct intel_soc_pmic now has a new cht_wc_model field
indicating the board model.

Only poke the CHT_WC_VBUS_GPIO_CTLO GPIO if this new field is set to
INTEL_CHT_WC_GPD_WIN_POCKET, which indicates the Type-C (with PD and
DP-altmode) setup used on the GPD pocket and GPD win; and on which
this GPIO actually controls an external 5V Vboost converter.

Acked-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoi2c: cht-wc: Make charger i2c-client instantiation board/device-model specific
Hans de Goede [Tue, 1 Feb 2022 13:07:01 +0000 (14:07 +0100)]
i2c: cht-wc: Make charger i2c-client instantiation board/device-model specific

The i2c-controller on the Cherry Trail - Whiskey Cove PMIC is special
in that it is always connected to the I2C charger IC of the board on
which the PMIC is used; and the charger IC is not described in ACPI,
so the i2c-cht-wc code needs to instantiate an i2c-client for it itself.

So far this was hardcoded to instantiate an i2c-client for the
bq24292i, with all properties, etc. set to match how this charger
is used on the GPD win and GPD pocket devices.

There is a rudimentary check to make sure the ACPI tables are at least
somewhat as expected, but this is far from accurate, leading to
a wrong i2c-client being instantiated for the charger on some boards.

Switch to the new DMI based intel_cht_wc_get_model() helper which is
exported by the MFD driver for the CHT Whiskey Cove PMIC to help PMIC
cell drivers like the i2c-cht-wc code reliably detect which board
they are running on.

And add board_info for the charger ICs as found on the other 2 known
boards with a Whisky Cove PMIC.

This has been tested on all 3 known boards.

Acked-by: Wolfram Sang <wsa@kernel.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agomfd: intel_soc_pmic_chtwc: Add cht_wc_model data to struct intel_soc_pmic
Hans de Goede [Tue, 1 Feb 2022 13:07:00 +0000 (14:07 +0100)]
mfd: intel_soc_pmic_chtwc: Add cht_wc_model data to struct intel_soc_pmic

Tablet / laptop designs using an Intel Cherry Trail x86 main SoC with
an Intel Whiskey Cove PMIC do not use a single standard setup for
the charger, fuel-gauge and other chips surrounding the PMIC /
charging+data USB port.

Unlike what is normal on x86 this diversity in designs is not handled
by the ACPI tables. On 2 of the 3 known designs there are no standard
(PNP0C0A) ACPI battery devices and on the 3th design the ACPI battery
device does not work under Linux due to it requiring non-standard
and undocumented ACPI behavior.

So to make things work under Linux we use native charger and fuel-gauge
drivers on these devices, re-using the native drivers used on ARM boards
with the same charger / fuel-gauge ICs.

This requires various MFD-cell drivers for the CHT-WC PMIC cells to
know which model they are exactly running on so that they can e.g.
instantiate an I2C-client for the right model charger-IC (the charger
is connected to an I2C-controller which is part of the PMIC).

Rather then duplicating DMI-id matching to check which model we are
running on in each MFD-cell driver, add a check for this to the
shared drivers/mfd/intel_soc_pmic_chtwc.c code by using a
DMI table for all 3 known models:

1. The GPD Win and GPD Pocket mini-laptops, these are really 2 models
but the Pocket re-uses the GPD Win's design in a different housing:

The WC PMIC is connected to a TI BQ24292i charger, paired with
a Maxim MAX17047 fuelgauge + a FUSB302 USB Type-C Controller +
a PI3USB30532 USB switch, for a fully functional Type-C port.

2. The Xiaomi Mi Pad 2:

The WC PMIC is connected to a TI BQ25890 charger, paired with
a TI BQ27520 fuelgauge, using the TI BQ25890 for BC1.2 charger type
detection, for a USB-2 only Type-C port without PD.

3. The Lenovo Yoga Book YB1-X90 / Lenovo Yoga Book YB1-X91 series:

The WC PMIC is connected to a TI BQ25892 charger, paired with
a TI BQ27542 fuelgauge, using the WC PMIC for BC1.2 charger type
detection and using the BQ25892's Mediatek Pump Express+ (1.0)
support to enable charging with up to 12V through a micro-USB port.

Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Use the devm_regmap_field_bulk_alloc() helper
Hans de Goede [Tue, 1 Feb 2022 13:06:59 +0000 (14:06 +0100)]
power: supply: bq25890: Use the devm_regmap_field_bulk_alloc() helper

Use the devm_regmap_field_bulk_alloc() helper function instead of
open-coding this ourselves.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Support higher charging voltages through Pump Express+ protocol
Yauhen Kharuzhy [Tue, 1 Feb 2022 13:06:58 +0000 (14:06 +0100)]
power: supply: bq25890: Support higher charging voltages through Pump Express+ protocol

Add a "linux,pump-express-vbus-max" property which indicates if the Pump
Express+ protocol should be used to increase the charging protocol.

If this new property is set and a DCP charger is detected then request
a higher charging voltage through the Pump Express+ protocol.

So far this new property is only used on x86/ACPI (non devicetree) devs,
IOW it is not used in actual devicetree files. The devicetree-bindings
maintainers have requested properties like these to not be added to the
devicetree-bindings, so the new property is deliberately not added
to the existing devicetree-bindings.

Changes by Hans de Goede:
- Port to my bq25890 patch-series + various cleanups
- Make behavior configurable through a new "linux,pump-express-vbus-max"
  device-property
- Sleep 1 second before re-checking the Vbus voltage after requesting
  it to be raised, to ensure that the ADC has time to sampled the new Vbus
- Add VBUSV bq25890_tables[] entry and use it in bq25890_get_vbus_voltage()
- Tweak commit message

Signed-off-by: Yauhen Kharuzhy <jekhor@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Co-developed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: On the bq25892 set the IINLIM based on external charger detection
Hans de Goede [Tue, 1 Feb 2022 13:06:57 +0000 (14:06 +0100)]
power: supply: bq25890: On the bq25892 set the IINLIM based on external charger detection

The bq25892 does not have builtin charger-type detection like the bq25980,
there might be some external charger detection capability, which will be
modelled as a power_supply class-device supplying the bq25892.

Use the usb_type property value from the supplier psy-device to set the
input-current-limit (when available).

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Add support for registering the Vbus boost converter as a...
Hans de Goede [Tue, 1 Feb 2022 13:06:56 +0000 (14:06 +0100)]
power: supply: bq25890: Add support for registering the Vbus boost converter as a regulator

The bq25890_charger code supports enabling/disabling the boost converter
based on usb-phy notifications. But the usb-phy framework is not used on
all boards/platforms. At support for registering the Vbus boost converter
as a standard regulator when there is no usb-phy on the board.

Also add support for providing regulator_init_data through platform_data
for use on boards where device-tree is not used and the platform code must
thus provide the regulator_init_data.

Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Add bq25890_set_otg_cfg() helper
Hans de Goede [Tue, 1 Feb 2022 13:06:55 +0000 (14:06 +0100)]
power: supply: bq25890: Add bq25890_set_otg_cfg() helper

Add a bq25890_set_otg_cfg() helper function, this is a preparation
patch for adding regulator support.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Drop dev->platform_data == NULL check
Hans de Goede [Tue, 1 Feb 2022 13:06:54 +0000 (14:06 +0100)]
power: supply: bq25890: Drop dev->platform_data == NULL check

Drop the "if (!dev->platform_data)" check, this seems to be an attempt
for allowing loading the driver on devices without devicetree stemming
from the initial commit of the driver (with the presumed intention being
the "return -ENODEV" else branch getting replaced with something else).

With the new "linux,skip-init" and "linux,read-back-settings" properties
the driver can actually supports devices without devicetree and this
check no longer makes sense.

While at it, also switch to dev_err_probe(), which is already used in
various other places in the driver.

Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Enable charging on boards where we skip reset
Hans de Goede [Tue, 1 Feb 2022 13:06:53 +0000 (14:06 +0100)]
power: supply: bq25890: Enable charging on boards where we skip reset

On boards where the "linux,skip-reset" boolean property is set we don't
reset the charger; and on some boards where the fw takes care of
initalizition F_CHG_CFG is set to 0 before handing control over to the OS.

Explicitly set F_CHG_CFG to 1 on boards where we don't reset the charger,
so that charging is always enabled on these boards, like it is always
enabled on boards where we do reset the charger.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Add support to read back the settings from the chip
Hans de Goede [Tue, 1 Feb 2022 13:06:52 +0000 (14:06 +0100)]
power: supply: bq25890: Add support to read back the settings from the chip

On most x86/ACPI devices there is no devicetree to supply the necessary
init-data. Instead the firmware already fully initializes the bq25890
charger at boot. To support this, add support for reading back the
settings from the chip through a new "linux,read-back-settings" boolean.

So far this new property is only used on x86/ACPI (non devicetree) devs,
IOW it is not used in actual devicetree files. The devicetree-bindings
maintainers have requested properties like these to not be added to the
devicetree-bindings, so the new property is deliberately not added
to the existing devicetree-bindings.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Add support to skip reset at probe() / remove()
Hans de Goede [Tue, 1 Feb 2022 13:06:51 +0000 (14:06 +0100)]
power: supply: bq25890: Add support to skip reset at probe() / remove()

On most x86/ACPI devices the firmware already fully initializes
the bq25890 charger at boot, in this case it is best to not reset
it at probe() time.

At support for a new "linux,skip-reset" boolean property to support this.
So far this new property is only used on x86/ACPI (non devicetree) devs,
IOW it is not used in actual devicetree files. The devicetree-bindings
maintainers have requested properties like these to not be added to the
devicetree-bindings, so the new property is deliberately not added
to the existing devicetree-bindings.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Add a bq25890_rw_init_data() helper
Hans de Goede [Tue, 1 Feb 2022 13:06:50 +0000 (14:06 +0100)]
power: supply: bq25890: Add a bq25890_rw_init_data() helper

On most x86/ACPI devices there is no devicetree to supply the necessary
init-data. Instead the firmware already fully initializes the bq25890
charger at boot.

Factor out the current code to write all the init_data from devicetree
into a new bq25890_rw_init_data() helper which can both write the data
to the charger (the current behavior) as well as read it back from
the charger into the init_data struct.

This is a preparation patch for adding support for x86/ACPI device's
where the init_data must be read back from the bq25890 charger.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Reduce reported CONSTANT_CHARGE_CURRENT_MAX for low temperatures
Yauhen Kharuzhy [Tue, 1 Feb 2022 13:06:49 +0000 (14:06 +0100)]
power: supply: bq25890: Reduce reported CONSTANT_CHARGE_CURRENT_MAX for low temperatures

Take into account possible current reduction due to low-temperature when
reading POWER_SUPPLY_PROP_CONSTANT_CHARGE_CURRENT_MAX. As described in
the datasheet in cool (0-20° Celcius) conditions the current limit is
decreased to 20% or 50% of ICHG field value depended on JEITA_ISET field.

Also add NTC_FAULT field value to the debug message in
bq25890_get_chip_state().

Changed by Hans de Goede:
- Fix reading F_CHG_FAULT instead of F_NTC_FIELD for state->ntc_fault
- Only read JEITA_ISET field if necessary
- Tweak commit message a bit

Signed-off-by: Yauhen Kharuzhy <jekhor@gmail.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Co-developed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: bq25890: Rename IILIM field to IINLIM
Yauhen Kharuzhy [Tue, 1 Feb 2022 13:06:48 +0000 (14:06 +0100)]
power: supply: bq25890: Rename IILIM field to IINLIM

Rename the Input Current Limit field in the REG00 from IILIM to IINLIM
accordingly with the bq2589x datasheet. This is just cosmetical change
to reduce confusion.

Signed-off-by: Yauhen Kharuzhy <jekhor@gmail.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: core: Refactor power_supply_set_input_current_limit_from_supplier()
Hans de Goede [Tue, 1 Feb 2022 13:06:47 +0000 (14:06 +0100)]
power: supply: core: Refactor power_supply_set_input_current_limit_from_supplier()

Some (USB) charger ICs have variants with USB D+ and D- pins to do their
own builtin charger-type detection, like e.g. the bq24190 and bq25890 and
also variants which lack this functionality, e.g. the bq24192 and bq25892.

In case the charger-type; and thus the input-current-limit detection is
done outside the charger IC then we need some way to communicate this to
the charger IC. In the past extcon was used for this, but if the external
detection does e.g. full USB PD negotiation then the extcon cable-types do
not convey enough information.

For these setups it was decided to model the external charging "brick"
and the parameters negotiated with it as a power_supply class-device
itself; and power_supply_set_input_current_limit_from_supplier() was
introduced to allow drivers to get the input-current-limit this way.

But in some cases psy drivers may want to know other properties, e.g. the
bq25892 can do "quick-charge" negotiation by pulsing its current draw,
but this should only be done if the usb_type psy-property of its supplier
is set to DCP (and device-properties indicate the board allows higher
voltages).

Instead of adding extra helper functions for each property which
a psy-driver wants to query from its supplier, refactor
power_supply_set_input_current_limit_from_supplier() into a
more generic power_supply_get_property_from_supplier() function.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Add a no_current_sense_res module_param
Hans de Goede [Thu, 6 Jan 2022 11:06:08 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Add a no_current_sense_res module_param

Some boards with an AXP288 fuel-gauge appear to have a broken (approx.
2 milli-ohm instead of 10) current sense resistor.

This makes the coulomb-counter part of the fuel-gauge useless, but the
OCV based capacity reporting is still working. Add a no_current_sense_res
module_param to disable use of the coulomb-counter using parts of the
fuel-gauge to allow users to work around this.

Note this is a module parameter and not done through DMI quirks, since
this seems to be a defect on some boards, not something which all boards
of the same model share.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Take lock before updating the valid flag
Hans de Goede [Thu, 6 Jan 2022 11:06:07 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Take lock before updating the valid flag

The valid flag is protected by the mutex, so code clearing it
should take the mutex before cleating the valid flag.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Refactor IRQ initialization
Hans de Goede [Thu, 6 Jan 2022 11:06:06 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Refactor IRQ initialization

Refactor the IRQ initialization code:

* Move the looking up of the vIRQs to the beginning of probe(), failing
  probe early if this fails
* Do the actual requesting of IRQs inline in probe() and properly abort
  probe() on errors
* Use devm_request_threaded_irq(), completing the conversion of probe() to
  only use devm managed resources and remove the remove() driver function.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Use devm_power_supply_register()
Hans de Goede [Thu, 6 Jan 2022 11:06:05 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Use devm_power_supply_register()

Use devm_power_supply_register() instead of
power_supply_register().

Note as a side-effect this changes the release order so that now
first the IRQs get free-ed and then the psy gets unregistered.
This is actually a bug-fix since this fixes the IRQ possibly trying
to reference the unregistered psy.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Use devm_add_action_or_reset() for iio chan release
Hans de Goede [Thu, 6 Jan 2022 11:06:04 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Use devm_add_action_or_reset() for iio chan release

An existing comment already mentions we: "cannot use devm_iio_channel_get
because x86 systems lack the device<->channel maps which iio_channel_get
will try to use when passed a non NULL device pointer".

Work around this by registering a devm action to free the iio-channels.
This is a step on the way to fully converting the probe() function to
only use devm managed resources.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Add axp288_fuel_gauge_read_initial_regs()
Hans de Goede [Thu, 6 Jan 2022 11:06:03 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Add axp288_fuel_gauge_read_initial_regs()

Refactor probe a bit, introducing a new
axp288_fuel_gauge_read_initial_regs() helper. This replaces a whole
bunch of gotos and removes the unblock_punit_i2c_access label.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp288_fuel_gauge: Add dev helper var to probe()
Hans de Goede [Thu, 6 Jan 2022 11:06:02 +0000 (12:06 +0100)]
power: supply: axp288_fuel_gauge: Add dev helper var to probe()

Add a dev local variable to probe() as shortcut for &pdev->dev, this is
a preparation change for making more use of devm managed resources.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: cpcap-battery: Add battery type auto detection for mapphone devices
Carl Philipp Klemm [Wed, 29 Dec 2021 00:08:53 +0000 (01:08 +0100)]
power: supply: cpcap-battery: Add battery type auto detection for mapphone devices

This allows cpcap-battery to detect whitch battery is inserted, HW4X or
BW8X for xt875 and EB41 for xt894 by examining the battery nvmem. If no
known battery is detected sane defaults are used.

Signed-off-by: Carl Philipp Klemm <philipp@uvos.xyz>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: axp20x_battery: properly report current when discharging
Evgeny Boger [Wed, 12 Jan 2022 08:47:27 +0000 (11:47 +0300)]
power: supply: axp20x_battery: properly report current when discharging

As stated in [1], negative current values are used for discharging
batteries.

AXP PMICs internally have two different ADC channels for shunt current
measurement: one used during charging and one during discharging.
The values reported by these ADCs are unsigned.
While the driver properly selects ADC channel to get the data from,
it doesn't apply negative sign when reporting discharging current.

[1] Documentation/ABI/testing/sysfs-class-power

Signed-off-by: Evgeny Boger <boger@wirenboard.com>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: reset: gemini-poweroff: Fix IRQ check in gemini_poweroff_probe
Miaoqian Lin [Thu, 20 Jan 2022 10:46:54 +0000 (10:46 +0000)]
power: reset: gemini-poweroff: Fix IRQ check in gemini_poweroff_probe

platform_get_irq() returns negative error number instead 0 on failure.
And the doc of platform_get_irq() provides a usage example:

    int irq = platform_get_irq(pdev, 0);
    if (irq < 0)
        return irq;

Fix the check of return value to catch errors correctly.

Fixes: f7a388d6cd1c ("power: reset: Add a driver for the Gemini poweroff")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: mp2629_charger: use platform_get_irq()
Sergey Shtylyov [Thu, 27 Jan 2022 19:44:25 +0000 (22:44 +0300)]
power: supply: mp2629_charger: use platform_get_irq()

Calling platform_get_irq_optional() doesn't make sense if you then bail out
on any error it returns.  Switch to calling platform_get_irq() instead and
remove dev_err() call as platform_get_irq() already curses loudly on error.

Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agopower: supply: cros_usbpd: Use struct_size() helper in kzalloc()
Gustavo A. R. Silva [Tue, 25 Jan 2022 17:52:45 +0000 (11:52 -0600)]
power: supply: cros_usbpd: Use struct_size() helper in kzalloc()

Make use of the struct_size() helper instead of an open-coded version,
in order to avoid any potential type mistakes or integer overflows that,
in the worst scenario, could lead to heap overflows.

Also, address the following sparse warnings:
drivers/power/supply/cros_usbpd-charger.c:107:23: warning: using sizeof on a flexible structure

Link: https://github.com/KSPP/linux/issues/174
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2 years agoLinux 5.17-rc1
Linus Torvalds [Sun, 23 Jan 2022 08:12:53 +0000 (10:12 +0200)]
Linux 5.17-rc1

2 years agoMerge tag 'perf-tools-for-v5.17-2022-01-22' of git://git.kernel.org/pub/scm/linux...
Linus Torvalds [Sun, 23 Jan 2022 06:14:21 +0000 (08:14 +0200)]
Merge tag 'perf-tools-for-v5.17-2022-01-22' of git://git./linux/kernel/git/acme/linux

Pull more perf tools updates from Arnaldo Carvalho de Melo:

 - Fix printing 'phys_addr' in 'perf script'.

 - Fix failure to add events with 'perf probe' in ppc64 due to not
   removing leading dot (ppc64 ABIv1).

 - Fix cpu_map__item() python binding building.

 - Support event alias in form foo-bar-baz, add pmu-events and
   parse-event tests for it.

 - No need to setup affinities when starting a workload or attaching to
   a pid.

 - Use path__join() to compose a path instead of ad-hoc snprintf()
   equivalent.

 - Override attr->sample_period for non-libpfm4 events.

 - Use libperf cpumap APIs instead of accessing the internal state
   directly.

 - Sync x86 arch prctl headers and files changed by the new
   set_mempolicy_home_node syscall with the kernel sources.

 - Remove duplicate include in cpumap.h.

 - Remove redundant err variable.

* tag 'perf-tools-for-v5.17-2022-01-22' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf tools: Remove redundant err variable
  perf test: Add parse-events test for aliases with hyphens
  perf test: Add pmu-events test for aliases with hyphens
  perf parse-events: Support event alias in form foo-bar-baz
  perf evsel: Override attr->sample_period for non-libpfm4 events
  perf cpumap: Remove duplicate include in cpumap.h
  perf cpumap: Migrate to libperf cpumap api
  perf python: Fix cpu_map__item() building
  perf script: Fix printing 'phys_addr' failure issue
  tools headers UAPI: Sync files changed by new set_mempolicy_home_node syscall
  tools headers UAPI: Sync x86 arch prctl headers with the kernel sources
  perf machine: Use path__join() to compose a path instead of snprintf(dir, '/', filename)
  perf evlist: No need to setup affinities when disabling events for pid targets
  perf evlist: No need to setup affinities when enabling events for pid targets
  perf stat: No need to setup affinities when starting a workload
  perf affinity: Allow passing a NULL arg to affinity__cleanup()
  perf probe: Fix ppc64 'perf probe add events failed' case

2 years agoMerge tag 'trace-v5.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt...
Linus Torvalds [Sun, 23 Jan 2022 06:07:02 +0000 (08:07 +0200)]
Merge tag 'trace-v5.17-3' of git://git./linux/kernel/git/rostedt/linux-trace

Pull ftrace fix from Steven Rostedt:
 "Fix s390 breakage from sorting mcount tables.

  The latest merge of the tracing tree sorts the mcount table at build
  time. But s390 appears to do things differently (like always) and
  replaces the sorted table back to the original unsorted one. As the
  ftrace algorithm depends on it being sorted, bad things happen when it
  is not, and s390 experienced those bad things.

  Add a new config to tell the boot if the mcount table is sorted or
  not, and allow s390 to opt out of it"

* tag 'trace-v5.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  ftrace: Fix assuming build time sort works for s390

2 years agoftrace: Fix assuming build time sort works for s390
Steven Rostedt (Google) [Sat, 22 Jan 2022 14:17:10 +0000 (09:17 -0500)]
ftrace: Fix assuming build time sort works for s390

To speed up the boot process, as mcount_loc needs to be sorted for ftrace
to work properly, sorting it at build time is more efficient than boot up
and can save milliseconds of time. Unfortunately, this change broke s390
as it will modify the mcount_loc location after the sorting takes place
and will put back the unsorted locations. Since the sorting is skipped at
boot up if it is believed that it was sorted at run time, ftrace can crash
as its algorithms are dependent on the list being sorted.

Add a new config BUILDTIME_MCOUNT_SORT that is set when
BUILDTIME_TABLE_SORT but not if S390 is set. Use this config to determine
if sorting should take place at boot up.

Link: https://lore.kernel.org/all/yt9dee51ctfn.fsf@linux.ibm.com/
Fixes: 72b3942a173c ("scripts: ftrace - move the sort-processing in ftrace_init")
Reported-by: Sven Schnelle <svens@linux.ibm.com>
Tested-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
2 years agoMerge tag 'kbuild-fixes-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/masah...
Linus Torvalds [Sun, 23 Jan 2022 04:32:29 +0000 (06:32 +0200)]
Merge tag 'kbuild-fixes-v5.17' of git://git./linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - Bring include/uapi/linux/nfc.h into the UAPI compile-test coverage

 - Revert the workaround of CONFIG_CC_IMPLICIT_FALLTHROUGH

 - Fix build errors in certs/Makefile

* tag 'kbuild-fixes-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  certs: Fix build error when CONFIG_MODULE_SIG_KEY is empty
  certs: Fix build error when CONFIG_MODULE_SIG_KEY is PKCS#11 URI
  Revert "Makefile: Do not quote value for CONFIG_CC_IMPLICIT_FALLTHROUGH"
  usr/include/Makefile: add linux/nfc.h to the compile-test coverage

2 years agoMerge tag 'bitmap-5.17-rc1' of git://github.com/norov/linux
Linus Torvalds [Sun, 23 Jan 2022 04:20:44 +0000 (06:20 +0200)]
Merge tag 'bitmap-5.17-rc1' of git://github.com/norov/linux

Pull bitmap updates from Yury Norov:

 - introduce for_each_set_bitrange()

 - use find_first_*_bit() instead of find_next_*_bit() where possible

 - unify for_each_bit() macros

* tag 'bitmap-5.17-rc1' of git://github.com/norov/linux:
  vsprintf: rework bitmap_list_string
  lib: bitmap: add performance test for bitmap_print_to_pagebuf
  bitmap: unify find_bit operations
  mm/percpu: micro-optimize pcpu_is_populated()
  Replace for_each_*_bit_from() with for_each_*_bit() where appropriate
  find: micro-optimize for_each_{set,clear}_bit()
  include/linux: move for_each_bit() macros from bitops.h to find.h
  cpumask: replace cpumask_next_* with cpumask_first_* where appropriate
  tools: sync tools/bitmap with mother linux
  all: replace find_next{,_zero}_bit with find_first{,_zero}_bit where appropriate
  cpumask: use find_first_and_bit()
  lib: add find_first_and_bit()
  arch: remove GENERIC_FIND_FIRST_BIT entirely
  include: move find.h from asm_generic to linux
  bitops: move find_bit_*_le functions from le.h to find.h
  bitops: protect find_first_{,zero}_bit properly

2 years agoperf tools: Remove redundant err variable
Minghao Chi [Wed, 12 Jan 2022 08:01:09 +0000 (08:01 +0000)]
perf tools: Remove redundant err variable

Return value from perf_event__process_tracing_data() directly instead
of taking this in another redundant variable.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lore.kernel.org/lkml/20220112080109.666800-1-chi.minghao@zte.com.cn
Signed-off-by: CGEL ZTE <cgel.zte@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf test: Add parse-events test for aliases with hyphens
John Garry [Mon, 17 Jan 2022 15:10:15 +0000 (23:10 +0800)]
perf test: Add parse-events test for aliases with hyphens

Add a test which allows us to test parsing an event alias with hyphens.

Since these events typically do not exist on most host systems, add the
alias to the fake pmu.

Function perf_pmu__test_parse_init() has terms added to match known test
aliases.

Signed-off-by: John Garry <john.garry@huawei.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Liu <liuqi115@huawei.com>
Cc: Shaokun Zhang <zhangshaokun@hisilicon.com>
Cc: linuxarm@huawei.com
Link: https://lore.kernel.org/r/1642432215-234089-4-git-send-email-john.garry@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf test: Add pmu-events test for aliases with hyphens
John Garry [Mon, 17 Jan 2022 15:10:14 +0000 (23:10 +0800)]
perf test: Add pmu-events test for aliases with hyphens

Add a test for aliases with hyphens in the name to ensure that the
pmu-events tables are as expects. There should be no reason why these sort
of aliases would be treated differently, but no harm in checking.

Signed-off-by: John Garry <john.garry@huawei.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Liu <liuqi115@huawei.com>
Cc: Shaokun Zhang <zhangshaokun@hisilicon.com>
Cc: linuxarm@huawei.com
Link: https://lore.kernel.org/r/1642432215-234089-3-git-send-email-john.garry@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf parse-events: Support event alias in form foo-bar-baz
John Garry [Mon, 17 Jan 2022 15:10:13 +0000 (23:10 +0800)]
perf parse-events: Support event alias in form foo-bar-baz

Event aliasing for events whose name in the form foo-bar-baz is not
supported, while foo-bar, foo_bar_baz, and other combinations are, i.e.
two hyphens are not supported.

The HiSilicon D06 platform has events in such form:

  $ ./perf list sdir-home-migrate

  List of pre-defined events (to be used in -e):

  uncore hha:
    sdir-home-migrate
   [Unit: hisi_sccl,hha]

  $ sudo ./perf stat -e sdir-home-migrate
  event syntax error: 'sdir-home-migrate'
                          \___ parser error
  Run 'perf list' for a list of valid events

   Usage: perf stat [<options>] [<command>]

   -e, --event <event>event selector. use 'perf list' to list available events

To support, add an extra PMU event symbol type for "baz", and add a new
rule in the bison file.

Signed-off-by: John Garry <john.garry@huawei.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qi Liu <liuqi115@huawei.com>
Cc: Shaokun Zhang <zhangshaokun@hisilicon.com>
Cc: linuxarm@huawei.com
Link: https://lore.kernel.org/r/1642432215-234089-2-git-send-email-john.garry@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>