linux/drivers/power/supply/sbs-battery.c

1295 lines
33 KiB
C
Raw Normal View History

treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 157 Based on 3 normalized pattern(s): this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version [author] [kishon] [vijay] [abraham] [i] [kishon]@[ti] [com] this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version [author] [graeme] [gregory] [gg]@[slimlogic] [co] [uk] [author] [kishon] [vijay] [abraham] [i] [kishon]@[ti] [com] [based] [on] [twl6030]_[usb] [c] [author] [hema] [hk] [hemahk]@[ti] [com] this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details extracted by the scancode license scanner the SPDX license identifier GPL-2.0-or-later has been chosen to replace the boilerplate/reference in 1105 file(s). Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Allison Randal <allison@lohutok.net> Reviewed-by: Richard Fontana <rfontana@redhat.com> Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Cc: linux-spdx@vger.kernel.org Link: https://lkml.kernel.org/r/20190527070033.202006027@linutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-05-27 06:55:06 +00:00
// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Gas Gauge driver for SBS Compliant Batteries
*
* Copyright (c) 2010, NVIDIA Corporation.
*/
#include <linux/bits.h>
#include <linux/delay.h>
#include <linux/devm-helpers.h>
#include <linux/err.h>
#include <linux/gpio/consumer.h>
#include <linux/i2c.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/property.h>
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
#include <linux/of_device.h>
#include <linux/power/sbs-battery.h>
#include <linux/power_supply.h>
#include <linux/slab.h>
#include <linux/stat.h>
enum {
REG_MANUFACTURER_DATA,
REG_BATTERY_MODE,
REG_TEMPERATURE,
REG_VOLTAGE,
REG_CURRENT_NOW,
REG_CURRENT_AVG,
REG_MAX_ERR,
REG_CAPACITY,
REG_TIME_TO_EMPTY_NOW,
REG_TIME_TO_EMPTY_AVG,
REG_TIME_TO_FULL_AVG,
REG_STATUS,
REG_CAPACITY_LEVEL,
REG_CYCLE_COUNT,
REG_SERIAL_NUMBER,
REG_REMAINING_CAPACITY,
REG_REMAINING_CAPACITY_CHARGE,
REG_FULL_CHARGE_CAPACITY,
REG_FULL_CHARGE_CAPACITY_CHARGE,
REG_DESIGN_CAPACITY,
REG_DESIGN_CAPACITY_CHARGE,
REG_DESIGN_VOLTAGE_MIN,
REG_DESIGN_VOLTAGE_MAX,
REG_CHEMISTRY,
REG_MANUFACTURER,
REG_MODEL_NAME,
REG_CHARGE_CURRENT,
REG_CHARGE_VOLTAGE,
};
#define REG_ADDR_SPEC_INFO 0x1A
#define SPEC_INFO_VERSION_MASK GENMASK(7, 4)
#define SPEC_INFO_VERSION_SHIFT 4
#define SBS_VERSION_1_0 1
#define SBS_VERSION_1_1 2
#define SBS_VERSION_1_1_WITH_PEC 3
#define REG_ADDR_MANUFACTURE_DATE 0x1B
/* Battery Mode defines */
#define BATTERY_MODE_OFFSET 0x03
#define BATTERY_MODE_CAPACITY_MASK BIT(15)
enum sbs_capacity_mode {
CAPACITY_MODE_AMPS = 0,
CAPACITY_MODE_WATTS = BATTERY_MODE_CAPACITY_MASK
};
#define BATTERY_MODE_CHARGER_MASK (1<<14)
/* manufacturer access defines */
#define MANUFACTURER_ACCESS_STATUS 0x0006
#define MANUFACTURER_ACCESS_SLEEP 0x0011
/* battery status value bits */
#define BATTERY_INITIALIZED 0x80
#define BATTERY_DISCHARGING 0x40
#define BATTERY_FULL_CHARGED 0x20
#define BATTERY_FULL_DISCHARGED 0x10
/* min_value and max_value are only valid for numerical data */
#define SBS_DATA(_psp, _addr, _min_value, _max_value) { \
.psp = _psp, \
.addr = _addr, \
.min_value = _min_value, \
.max_value = _max_value, \
}
static const struct chip_data {
enum power_supply_property psp;
u8 addr;
int min_value;
int max_value;
} sbs_data[] = {
[REG_MANUFACTURER_DATA] =
SBS_DATA(POWER_SUPPLY_PROP_PRESENT, 0x00, 0, 65535),
[REG_BATTERY_MODE] =
SBS_DATA(-1, 0x03, 0, 65535),
[REG_TEMPERATURE] =
SBS_DATA(POWER_SUPPLY_PROP_TEMP, 0x08, 0, 65535),
[REG_VOLTAGE] =
SBS_DATA(POWER_SUPPLY_PROP_VOLTAGE_NOW, 0x09, 0, 65535),
[REG_CURRENT_NOW] =
SBS_DATA(POWER_SUPPLY_PROP_CURRENT_NOW, 0x0A, -32768, 32767),
[REG_CURRENT_AVG] =
SBS_DATA(POWER_SUPPLY_PROP_CURRENT_AVG, 0x0B, -32768, 32767),
[REG_MAX_ERR] =
SBS_DATA(POWER_SUPPLY_PROP_CAPACITY_ERROR_MARGIN, 0x0c, 0, 100),
[REG_CAPACITY] =
SBS_DATA(POWER_SUPPLY_PROP_CAPACITY, 0x0D, 0, 100),
[REG_REMAINING_CAPACITY] =
SBS_DATA(POWER_SUPPLY_PROP_ENERGY_NOW, 0x0F, 0, 65535),
[REG_REMAINING_CAPACITY_CHARGE] =
SBS_DATA(POWER_SUPPLY_PROP_CHARGE_NOW, 0x0F, 0, 65535),
[REG_FULL_CHARGE_CAPACITY] =
SBS_DATA(POWER_SUPPLY_PROP_ENERGY_FULL, 0x10, 0, 65535),
[REG_FULL_CHARGE_CAPACITY_CHARGE] =
SBS_DATA(POWER_SUPPLY_PROP_CHARGE_FULL, 0x10, 0, 65535),
[REG_TIME_TO_EMPTY_NOW] =
SBS_DATA(POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW, 0x11, 0, 65535),
[REG_TIME_TO_EMPTY_AVG] =
SBS_DATA(POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG, 0x12, 0, 65535),
[REG_TIME_TO_FULL_AVG] =
SBS_DATA(POWER_SUPPLY_PROP_TIME_TO_FULL_AVG, 0x13, 0, 65535),
[REG_CHARGE_CURRENT] =
SBS_DATA(POWER_SUPPLY_PROP_CONSTANT_CHARGE_CURRENT_MAX, 0x14, 0, 65535),
[REG_CHARGE_VOLTAGE] =
SBS_DATA(POWER_SUPPLY_PROP_CONSTANT_CHARGE_VOLTAGE_MAX, 0x15, 0, 65535),
[REG_STATUS] =
SBS_DATA(POWER_SUPPLY_PROP_STATUS, 0x16, 0, 65535),
[REG_CAPACITY_LEVEL] =
SBS_DATA(POWER_SUPPLY_PROP_CAPACITY_LEVEL, 0x16, 0, 65535),
[REG_CYCLE_COUNT] =
SBS_DATA(POWER_SUPPLY_PROP_CYCLE_COUNT, 0x17, 0, 65535),
[REG_DESIGN_CAPACITY] =
SBS_DATA(POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN, 0x18, 0, 65535),
[REG_DESIGN_CAPACITY_CHARGE] =
SBS_DATA(POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN, 0x18, 0, 65535),
[REG_DESIGN_VOLTAGE_MIN] =
SBS_DATA(POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN, 0x19, 0, 65535),
[REG_DESIGN_VOLTAGE_MAX] =
SBS_DATA(POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN, 0x19, 0, 65535),
[REG_SERIAL_NUMBER] =
SBS_DATA(POWER_SUPPLY_PROP_SERIAL_NUMBER, 0x1C, 0, 65535),
/* Properties of type `const char *' */
[REG_MANUFACTURER] =
SBS_DATA(POWER_SUPPLY_PROP_MANUFACTURER, 0x20, 0, 65535),
[REG_MODEL_NAME] =
SBS_DATA(POWER_SUPPLY_PROP_MODEL_NAME, 0x21, 0, 65535),
[REG_CHEMISTRY] =
SBS_DATA(POWER_SUPPLY_PROP_TECHNOLOGY, 0x22, 0, 65535)
};
static const enum power_supply_property sbs_properties[] = {
POWER_SUPPLY_PROP_STATUS,
POWER_SUPPLY_PROP_CAPACITY_LEVEL,
POWER_SUPPLY_PROP_HEALTH,
POWER_SUPPLY_PROP_PRESENT,
POWER_SUPPLY_PROP_TECHNOLOGY,
POWER_SUPPLY_PROP_CYCLE_COUNT,
POWER_SUPPLY_PROP_VOLTAGE_NOW,
POWER_SUPPLY_PROP_CURRENT_NOW,
POWER_SUPPLY_PROP_CURRENT_AVG,
POWER_SUPPLY_PROP_CAPACITY,
POWER_SUPPLY_PROP_CAPACITY_ERROR_MARGIN,
POWER_SUPPLY_PROP_TEMP,
POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW,
POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG,
POWER_SUPPLY_PROP_TIME_TO_FULL_AVG,
POWER_SUPPLY_PROP_SERIAL_NUMBER,
POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN,
POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN,
POWER_SUPPLY_PROP_ENERGY_NOW,
POWER_SUPPLY_PROP_ENERGY_FULL,
POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN,
POWER_SUPPLY_PROP_CHARGE_NOW,
POWER_SUPPLY_PROP_CHARGE_FULL,
POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN,
POWER_SUPPLY_PROP_CONSTANT_CHARGE_CURRENT_MAX,
POWER_SUPPLY_PROP_CONSTANT_CHARGE_VOLTAGE_MAX,
POWER_SUPPLY_PROP_MANUFACTURE_YEAR,
POWER_SUPPLY_PROP_MANUFACTURE_MONTH,
POWER_SUPPLY_PROP_MANUFACTURE_DAY,
/* Properties of type `const char *' */
POWER_SUPPLY_PROP_MANUFACTURER,
POWER_SUPPLY_PROP_MODEL_NAME
};
/* Supports special manufacturer commands from TI BQ20Z65 and BQ20Z75 IC. */
#define SBS_FLAGS_TI_BQ20ZX5 BIT(0)
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
static const enum power_supply_property string_properties[] = {
POWER_SUPPLY_PROP_TECHNOLOGY,
POWER_SUPPLY_PROP_MANUFACTURER,
POWER_SUPPLY_PROP_MODEL_NAME,
};
#define NR_STRING_BUFFERS ARRAY_SIZE(string_properties)
struct sbs_info {
struct i2c_client *client;
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
struct power_supply *power_supply;
bool is_present;
struct gpio_desc *gpio_detect;
bool charger_broadcasts;
int last_state;
int poll_time;
u32 i2c_retry_count;
u32 poll_retry_count;
struct delayed_work work;
struct mutex mode_lock;
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
u32 flags;
int technology;
char strings[NR_STRING_BUFFERS][I2C_SMBUS_BLOCK_MAX + 1];
};
static char *sbs_get_string_buf(struct sbs_info *chip,
enum power_supply_property psp)
{
int i = 0;
for (i = 0; i < NR_STRING_BUFFERS; i++)
if (string_properties[i] == psp)
return chip->strings[i];
return ERR_PTR(-EINVAL);
}
static void sbs_invalidate_cached_props(struct sbs_info *chip)
{
int i = 0;
chip->technology = -1;
for (i = 0; i < NR_STRING_BUFFERS; i++)
chip->strings[i][0] = 0;
}
sbs-battery: add option to always register battery Commit a22b41a31e53 ("sbs-battery: Probe should try talking to the device") introduced a step in probing the SBS battery, that tries to talk to the device before actually registering it, saying: this driver doesn't actually try talking to the device at probe time, so if it's incorrectly configured in the device tree or platform data (or if the battery has been removed from the system), then probe will succeed and every access will sit there and time out. The end result is a possibly laggy system that thinks it has a battery but can never read status, which isn't very useful. Which is of course reasonable. However, it is also very well possible for a device to boot up on wall-power and be connected to a battery later on. This is a scenario that the driver supported before said patch was applied, and even easily achieved by booting up with the battery attached and removing it later on. sbs-battery's 'present' sysfs file can be used to determine if the device is available or not. So with automated device detection lacking for now, in some cases it is possible that one wants to register a battery, even if none are attached at the moment. To facilitate this, add a module parameter that can be used to configure forced loading module loading time. If set, the battery will always be registered without checking the sanity of the connection. Signed-off-by: Frans Klaver <frans.klaver@xsens.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-06-10 12:16:56 +00:00
static bool force_load;
static int sbs_read_word_data(struct i2c_client *client, u8 address);
static int sbs_write_word_data(struct i2c_client *client, u8 address, u16 value);
static void sbs_disable_charger_broadcasts(struct sbs_info *chip)
{
int val = sbs_read_word_data(chip->client, BATTERY_MODE_OFFSET);
if (val < 0)
goto exit;
val |= BATTERY_MODE_CHARGER_MASK;
val = sbs_write_word_data(chip->client, BATTERY_MODE_OFFSET, val);
exit:
if (val < 0)
dev_err(&chip->client->dev,
"Failed to disable charger broadcasting: %d\n", val);
else
dev_dbg(&chip->client->dev, "%s\n", __func__);
}
static int sbs_update_presence(struct sbs_info *chip, bool is_present)
{
struct i2c_client *client = chip->client;
int retries = chip->i2c_retry_count;
s32 ret = 0;
u8 version;
if (chip->is_present == is_present)
return 0;
if (!is_present) {
chip->is_present = false;
/* Disable PEC when no device is present */
client->flags &= ~I2C_CLIENT_PEC;
sbs_invalidate_cached_props(chip);
return 0;
}
/* Check if device supports packet error checking and use it */
while (retries > 0) {
ret = i2c_smbus_read_word_data(client, REG_ADDR_SPEC_INFO);
if (ret >= 0)
break;
/*
* Some batteries trigger the detection pin before the
* I2C bus is properly connected. This works around the
* issue.
*/
msleep(100);
retries--;
}
if (ret < 0) {
dev_dbg(&client->dev, "failed to read spec info: %d\n", ret);
/* fallback to old behaviour */
client->flags &= ~I2C_CLIENT_PEC;
chip->is_present = true;
return ret;
}
version = (ret & SPEC_INFO_VERSION_MASK) >> SPEC_INFO_VERSION_SHIFT;
if (version == SBS_VERSION_1_1_WITH_PEC)
client->flags |= I2C_CLIENT_PEC;
else
client->flags &= ~I2C_CLIENT_PEC;
if (of_device_is_compatible(client->dev.parent->of_node, "google,cros-ec-i2c-tunnel")
&& client->flags & I2C_CLIENT_PEC) {
dev_info(&client->dev, "Disabling PEC because of broken Cros-EC implementation\n");
client->flags &= ~I2C_CLIENT_PEC;
}
dev_dbg(&client->dev, "PEC: %s\n", (client->flags & I2C_CLIENT_PEC) ?
"enabled" : "disabled");
if (!chip->is_present && is_present && !chip->charger_broadcasts)
sbs_disable_charger_broadcasts(chip);
chip->is_present = true;
return 0;
}
static int sbs_read_word_data(struct i2c_client *client, u8 address)
{
struct sbs_info *chip = i2c_get_clientdata(client);
int retries = chip->i2c_retry_count;
s32 ret = 0;
while (retries > 0) {
ret = i2c_smbus_read_word_data(client, address);
if (ret >= 0)
break;
retries--;
}
if (ret < 0) {
dev_dbg(&client->dev,
"%s: i2c read at address 0x%x failed\n",
__func__, address);
return ret;
}
return ret;
}
static int sbs_read_string_data_fallback(struct i2c_client *client, u8 address, char *values)
{
struct sbs_info *chip = i2c_get_clientdata(client);
s32 ret = 0, block_length = 0;
int retries_length, retries_block;
u8 block_buffer[I2C_SMBUS_BLOCK_MAX + 1];
retries_length = chip->i2c_retry_count;
retries_block = chip->i2c_retry_count;
dev_warn_once(&client->dev, "I2C adapter does not support I2C_FUNC_SMBUS_READ_BLOCK_DATA.\n"
"Fallback method does not support PEC.\n");
/* Adapter needs to support these two functions */
if (!i2c_check_functionality(client->adapter,
I2C_FUNC_SMBUS_BYTE_DATA |
I2C_FUNC_SMBUS_I2C_BLOCK)){
return -ENODEV;
}
/* Get the length of block data */
while (retries_length > 0) {
ret = i2c_smbus_read_byte_data(client, address);
if (ret >= 0)
break;
retries_length--;
}
if (ret < 0) {
dev_dbg(&client->dev,
"%s: i2c read at address 0x%x failed\n",
__func__, address);
return ret;
}
/* block_length does not include NULL terminator */
block_length = ret;
if (block_length > I2C_SMBUS_BLOCK_MAX) {
dev_err(&client->dev,
"%s: Returned block_length is longer than 0x%x\n",
__func__, I2C_SMBUS_BLOCK_MAX);
return -EINVAL;
}
/* Get the block data */
while (retries_block > 0) {
ret = i2c_smbus_read_i2c_block_data(
client, address,
block_length + 1, block_buffer);
if (ret >= 0)
break;
retries_block--;
}
if (ret < 0) {
dev_dbg(&client->dev,
"%s: i2c read at address 0x%x failed\n",
__func__, address);
return ret;
}
/* block_buffer[0] == block_length */
memcpy(values, block_buffer + 1, block_length);
values[block_length] = '\0';
return ret;
}
static int sbs_read_string_data(struct i2c_client *client, u8 address, char *values)
{
struct sbs_info *chip = i2c_get_clientdata(client);
int retries = chip->i2c_retry_count;
int ret = 0;
if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_READ_BLOCK_DATA)) {
bool pec = client->flags & I2C_CLIENT_PEC;
client->flags &= ~I2C_CLIENT_PEC;
ret = sbs_read_string_data_fallback(client, address, values);
if (pec)
client->flags |= I2C_CLIENT_PEC;
return ret;
}
while (retries > 0) {
ret = i2c_smbus_read_block_data(client, address, values);
if (ret >= 0)
break;
retries--;
}
if (ret < 0) {
dev_dbg(&client->dev, "failed to read block 0x%x: %d\n", address, ret);
return ret;
}
/* add string termination */
values[ret] = '\0';
return ret;
}
static int sbs_write_word_data(struct i2c_client *client, u8 address,
u16 value)
{
struct sbs_info *chip = i2c_get_clientdata(client);
int retries = chip->i2c_retry_count;
s32 ret = 0;
while (retries > 0) {
ret = i2c_smbus_write_word_data(client, address, value);
if (ret >= 0)
break;
retries--;
}
if (ret < 0) {
dev_dbg(&client->dev,
"%s: i2c write to address 0x%x failed\n",
__func__, address);
return ret;
}
return 0;
}
static int sbs_status_correct(struct i2c_client *client, int *intval)
{
int ret;
ret = sbs_read_word_data(client, sbs_data[REG_CURRENT_NOW].addr);
if (ret < 0)
return ret;
ret = (s16)ret;
/* Not drawing current -> not charging (i.e. idle) */
if (*intval != POWER_SUPPLY_STATUS_FULL && ret == 0)
*intval = POWER_SUPPLY_STATUS_NOT_CHARGING;
if (*intval == POWER_SUPPLY_STATUS_FULL) {
/* Drawing or providing current when full */
if (ret > 0)
*intval = POWER_SUPPLY_STATUS_CHARGING;
else if (ret < 0)
*intval = POWER_SUPPLY_STATUS_DISCHARGING;
}
return 0;
}
static bool sbs_bat_needs_calibration(struct i2c_client *client)
{
int ret;
ret = sbs_read_word_data(client, sbs_data[REG_BATTERY_MODE].addr);
if (ret < 0)
return false;
return !!(ret & BIT(7));
}
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
static int sbs_get_ti_battery_presence_and_health(
struct i2c_client *client, enum power_supply_property psp,
union power_supply_propval *val)
{
s32 ret;
/*
* Write to ManufacturerAccess with ManufacturerAccess command
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
* and then read the status.
*/
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
ret = sbs_write_word_data(client, sbs_data[REG_MANUFACTURER_DATA].addr,
MANUFACTURER_ACCESS_STATUS);
if (ret < 0) {
if (psp == POWER_SUPPLY_PROP_PRESENT)
val->intval = 0; /* battery removed */
return ret;
}
ret = sbs_read_word_data(client, sbs_data[REG_MANUFACTURER_DATA].addr);
if (ret < 0) {
if (psp == POWER_SUPPLY_PROP_PRESENT)
val->intval = 0; /* battery removed */
return ret;
}
if (ret < sbs_data[REG_MANUFACTURER_DATA].min_value ||
ret > sbs_data[REG_MANUFACTURER_DATA].max_value) {
val->intval = 0;
return 0;
}
/* Mask the upper nibble of 2nd byte and
* lower byte of response then
* shift the result by 8 to get status*/
ret &= 0x0F00;
ret >>= 8;
if (psp == POWER_SUPPLY_PROP_PRESENT) {
if (ret == 0x0F)
/* battery removed */
val->intval = 0;
else
val->intval = 1;
} else if (psp == POWER_SUPPLY_PROP_HEALTH) {
if (ret == 0x09)
val->intval = POWER_SUPPLY_HEALTH_UNSPEC_FAILURE;
else if (ret == 0x0B)
val->intval = POWER_SUPPLY_HEALTH_OVERHEAT;
else if (ret == 0x0C)
val->intval = POWER_SUPPLY_HEALTH_DEAD;
else if (sbs_bat_needs_calibration(client))
val->intval = POWER_SUPPLY_HEALTH_CALIBRATION_REQUIRED;
else
val->intval = POWER_SUPPLY_HEALTH_GOOD;
}
return 0;
}
static int sbs_get_battery_presence_and_health(
struct i2c_client *client, enum power_supply_property psp,
union power_supply_propval *val)
{
struct sbs_info *chip = i2c_get_clientdata(client);
int ret;
if (chip->flags & SBS_FLAGS_TI_BQ20ZX5)
return sbs_get_ti_battery_presence_and_health(client, psp, val);
/* Dummy command; if it succeeds, battery is present. */
ret = sbs_read_word_data(client, sbs_data[REG_STATUS].addr);
if (ret < 0) { /* battery not present*/
if (psp == POWER_SUPPLY_PROP_PRESENT) {
val->intval = 0;
return 0;
}
return ret;
}
if (psp == POWER_SUPPLY_PROP_PRESENT)
val->intval = 1; /* battery present */
else { /* POWER_SUPPLY_PROP_HEALTH */
if (sbs_bat_needs_calibration(client)) {
val->intval = POWER_SUPPLY_HEALTH_CALIBRATION_REQUIRED;
} else {
/* SBS spec doesn't have a general health command. */
val->intval = POWER_SUPPLY_HEALTH_UNKNOWN;
}
}
return 0;
}
static int sbs_get_battery_property(struct i2c_client *client,
int reg_offset, enum power_supply_property psp,
union power_supply_propval *val)
{
struct sbs_info *chip = i2c_get_clientdata(client);
s32 ret;
ret = sbs_read_word_data(client, sbs_data[reg_offset].addr);
if (ret < 0)
return ret;
/* returned values are 16 bit */
if (sbs_data[reg_offset].min_value < 0)
ret = (s16)ret;
if (ret >= sbs_data[reg_offset].min_value &&
ret <= sbs_data[reg_offset].max_value) {
val->intval = ret;
if (psp == POWER_SUPPLY_PROP_CAPACITY_LEVEL) {
if (!(ret & BATTERY_INITIALIZED))
val->intval =
POWER_SUPPLY_CAPACITY_LEVEL_UNKNOWN;
else if (ret & BATTERY_FULL_CHARGED)
val->intval =
POWER_SUPPLY_CAPACITY_LEVEL_FULL;
else if (ret & BATTERY_FULL_DISCHARGED)
val->intval =
POWER_SUPPLY_CAPACITY_LEVEL_CRITICAL;
else
val->intval =
POWER_SUPPLY_CAPACITY_LEVEL_NORMAL;
return 0;
} else if (psp != POWER_SUPPLY_PROP_STATUS) {
return 0;
}
if (ret & BATTERY_FULL_CHARGED)
val->intval = POWER_SUPPLY_STATUS_FULL;
else if (ret & BATTERY_DISCHARGING)
val->intval = POWER_SUPPLY_STATUS_DISCHARGING;
else
val->intval = POWER_SUPPLY_STATUS_CHARGING;
sbs_status_correct(client, &val->intval);
if (chip->poll_time == 0)
chip->last_state = val->intval;
else if (chip->last_state != val->intval) {
cancel_delayed_work_sync(&chip->work);
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
power_supply_changed(chip->power_supply);
chip->poll_time = 0;
}
} else {
if (psp == POWER_SUPPLY_PROP_STATUS)
val->intval = POWER_SUPPLY_STATUS_UNKNOWN;
else if (psp == POWER_SUPPLY_PROP_CAPACITY)
/* sbs spec says that this can be >100 %
* even if max value is 100 %
*/
val->intval = min(ret, 100);
else
val->intval = 0;
}
return 0;
}
static int sbs_get_property_index(struct i2c_client *client,
enum power_supply_property psp)
{
int count;
for (count = 0; count < ARRAY_SIZE(sbs_data); count++)
if (psp == sbs_data[count].psp)
return count;
dev_warn(&client->dev,
"%s: Invalid Property - %d\n", __func__, psp);
return -EINVAL;
}
static const char *sbs_get_constant_string(struct sbs_info *chip,
enum power_supply_property psp)
{
int ret;
char *buf;
u8 addr;
buf = sbs_get_string_buf(chip, psp);
if (IS_ERR(buf))
return buf;
if (!buf[0]) {
ret = sbs_get_property_index(chip->client, psp);
if (ret < 0)
return ERR_PTR(ret);
addr = sbs_data[ret].addr;
ret = sbs_read_string_data(chip->client, addr, buf);
if (ret < 0)
return ERR_PTR(ret);
}
return buf;
}
static void sbs_unit_adjustment(struct i2c_client *client,
enum power_supply_property psp, union power_supply_propval *val)
{
#define BASE_UNIT_CONVERSION 1000
#define BATTERY_MODE_CAP_MULT_WATT (10 * BASE_UNIT_CONVERSION)
#define TIME_UNIT_CONVERSION 60
#define TEMP_KELVIN_TO_CELSIUS 2731
switch (psp) {
case POWER_SUPPLY_PROP_ENERGY_NOW:
case POWER_SUPPLY_PROP_ENERGY_FULL:
case POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN:
/* sbs provides energy in units of 10mWh.
* Convert to µWh
*/
val->intval *= BATTERY_MODE_CAP_MULT_WATT;
break;
case POWER_SUPPLY_PROP_VOLTAGE_NOW:
case POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN:
case POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN:
case POWER_SUPPLY_PROP_CURRENT_NOW:
case POWER_SUPPLY_PROP_CURRENT_AVG:
case POWER_SUPPLY_PROP_CHARGE_NOW:
case POWER_SUPPLY_PROP_CONSTANT_CHARGE_CURRENT_MAX:
case POWER_SUPPLY_PROP_CONSTANT_CHARGE_VOLTAGE_MAX:
case POWER_SUPPLY_PROP_CHARGE_FULL:
case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN:
val->intval *= BASE_UNIT_CONVERSION;
break;
case POWER_SUPPLY_PROP_TEMP:
/* sbs provides battery temperature in 0.1K
* so convert it to 0.1°C
*/
val->intval -= TEMP_KELVIN_TO_CELSIUS;
break;
case POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW:
case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
case POWER_SUPPLY_PROP_TIME_TO_FULL_AVG:
/* sbs provides time to empty and time to full in minutes.
* Convert to seconds
*/
val->intval *= TIME_UNIT_CONVERSION;
break;
default:
dev_dbg(&client->dev,
"%s: no need for unit conversion %d\n", __func__, psp);
}
}
static enum sbs_capacity_mode sbs_set_capacity_mode(struct i2c_client *client,
enum sbs_capacity_mode mode)
{
int ret, original_val;
original_val = sbs_read_word_data(client, BATTERY_MODE_OFFSET);
if (original_val < 0)
return original_val;
if ((original_val & BATTERY_MODE_CAPACITY_MASK) == mode)
return mode;
if (mode == CAPACITY_MODE_AMPS)
ret = original_val & ~BATTERY_MODE_CAPACITY_MASK;
else
ret = original_val | BATTERY_MODE_CAPACITY_MASK;
ret = sbs_write_word_data(client, BATTERY_MODE_OFFSET, ret);
if (ret < 0)
return ret;
usleep_range(1000, 2000);
return original_val & BATTERY_MODE_CAPACITY_MASK;
}
static int sbs_get_battery_capacity(struct i2c_client *client,
int reg_offset, enum power_supply_property psp,
union power_supply_propval *val)
{
s32 ret;
enum sbs_capacity_mode mode = CAPACITY_MODE_WATTS;
if (power_supply_is_amp_property(psp))
mode = CAPACITY_MODE_AMPS;
mode = sbs_set_capacity_mode(client, mode);
if ((int)mode < 0)
return mode;
ret = sbs_read_word_data(client, sbs_data[reg_offset].addr);
if (ret < 0)
return ret;
val->intval = ret;
ret = sbs_set_capacity_mode(client, mode);
if (ret < 0)
return ret;
return 0;
}
static char sbs_serial[5];
static int sbs_get_battery_serial_number(struct i2c_client *client,
union power_supply_propval *val)
{
int ret;
ret = sbs_read_word_data(client, sbs_data[REG_SERIAL_NUMBER].addr);
if (ret < 0)
return ret;
sprintf(sbs_serial, "%04x", ret);
val->strval = sbs_serial;
return 0;
}
static int sbs_get_chemistry(struct sbs_info *chip,
union power_supply_propval *val)
{
const char *chemistry;
if (chip->technology != -1) {
val->intval = chip->technology;
return 0;
}
chemistry = sbs_get_constant_string(chip, POWER_SUPPLY_PROP_TECHNOLOGY);
if (IS_ERR(chemistry))
return PTR_ERR(chemistry);
if (!strncasecmp(chemistry, "LION", 4))
chip->technology = POWER_SUPPLY_TECHNOLOGY_LION;
else if (!strncasecmp(chemistry, "LiP", 3))
chip->technology = POWER_SUPPLY_TECHNOLOGY_LIPO;
else if (!strncasecmp(chemistry, "NiCd", 4))
chip->technology = POWER_SUPPLY_TECHNOLOGY_NiCd;
else if (!strncasecmp(chemistry, "NiMH", 4))
chip->technology = POWER_SUPPLY_TECHNOLOGY_NiMH;
else
chip->technology = POWER_SUPPLY_TECHNOLOGY_UNKNOWN;
if (chip->technology == POWER_SUPPLY_TECHNOLOGY_UNKNOWN)
dev_warn(&chip->client->dev, "Unknown chemistry: %s\n", chemistry);
val->intval = chip->technology;
return 0;
}
static int sbs_get_battery_manufacture_date(struct i2c_client *client,
enum power_supply_property psp,
union power_supply_propval *val)
{
int ret;
u16 day, month, year;
ret = sbs_read_word_data(client, REG_ADDR_MANUFACTURE_DATE);
if (ret < 0)
return ret;
day = ret & GENMASK(4, 0);
month = (ret & GENMASK(8, 5)) >> 5;
year = ((ret & GENMASK(15, 9)) >> 9) + 1980;
switch (psp) {
case POWER_SUPPLY_PROP_MANUFACTURE_YEAR:
val->intval = year;
break;
case POWER_SUPPLY_PROP_MANUFACTURE_MONTH:
val->intval = month;
break;
case POWER_SUPPLY_PROP_MANUFACTURE_DAY:
val->intval = day;
break;
default:
return -EINVAL;
}
return 0;
}
static int sbs_get_property(struct power_supply *psy,
enum power_supply_property psp,
union power_supply_propval *val)
{
int ret = 0;
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
struct sbs_info *chip = power_supply_get_drvdata(psy);
struct i2c_client *client = chip->client;
const char *str;
if (chip->gpio_detect) {
ret = gpiod_get_value_cansleep(chip->gpio_detect);
if (ret < 0)
return ret;
if (psp == POWER_SUPPLY_PROP_PRESENT) {
val->intval = ret;
sbs_update_presence(chip, ret);
return 0;
}
if (ret == 0)
return -ENODATA;
}
switch (psp) {
case POWER_SUPPLY_PROP_PRESENT:
case POWER_SUPPLY_PROP_HEALTH:
ret = sbs_get_battery_presence_and_health(client, psp, val);
/* this can only be true if no gpio is used */
if (psp == POWER_SUPPLY_PROP_PRESENT)
return 0;
break;
case POWER_SUPPLY_PROP_TECHNOLOGY:
ret = sbs_get_chemistry(chip, val);
if (ret < 0)
break;
goto done; /* don't trigger power_supply_changed()! */
case POWER_SUPPLY_PROP_ENERGY_NOW:
case POWER_SUPPLY_PROP_ENERGY_FULL:
case POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN:
case POWER_SUPPLY_PROP_CHARGE_NOW:
case POWER_SUPPLY_PROP_CHARGE_FULL:
case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN:
ret = sbs_get_property_index(client, psp);
if (ret < 0)
break;
/* sbs_get_battery_capacity() will change the battery mode
* temporarily to read the requested attribute. Ensure we stay
* in the desired mode for the duration of the attribute read.
*/
mutex_lock(&chip->mode_lock);
ret = sbs_get_battery_capacity(client, ret, psp, val);
mutex_unlock(&chip->mode_lock);
break;
case POWER_SUPPLY_PROP_SERIAL_NUMBER:
ret = sbs_get_battery_serial_number(client, val);
break;
case POWER_SUPPLY_PROP_STATUS:
case POWER_SUPPLY_PROP_CAPACITY_LEVEL:
case POWER_SUPPLY_PROP_CYCLE_COUNT:
case POWER_SUPPLY_PROP_VOLTAGE_NOW:
case POWER_SUPPLY_PROP_CURRENT_NOW:
case POWER_SUPPLY_PROP_CURRENT_AVG:
case POWER_SUPPLY_PROP_TEMP:
case POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW:
case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
case POWER_SUPPLY_PROP_TIME_TO_FULL_AVG:
case POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN:
case POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN:
case POWER_SUPPLY_PROP_CONSTANT_CHARGE_CURRENT_MAX:
case POWER_SUPPLY_PROP_CONSTANT_CHARGE_VOLTAGE_MAX:
case POWER_SUPPLY_PROP_CAPACITY:
case POWER_SUPPLY_PROP_CAPACITY_ERROR_MARGIN:
ret = sbs_get_property_index(client, psp);
if (ret < 0)
break;
ret = sbs_get_battery_property(client, ret, psp, val);
break;
case POWER_SUPPLY_PROP_MODEL_NAME:
case POWER_SUPPLY_PROP_MANUFACTURER:
str = sbs_get_constant_string(chip, psp);
if (IS_ERR(str))
ret = PTR_ERR(str);
else
val->strval = str;
break;
case POWER_SUPPLY_PROP_MANUFACTURE_YEAR:
case POWER_SUPPLY_PROP_MANUFACTURE_MONTH:
case POWER_SUPPLY_PROP_MANUFACTURE_DAY:
ret = sbs_get_battery_manufacture_date(client, psp, val);
break;
default:
dev_err(&client->dev,
"%s: INVALID property\n", __func__);
return -EINVAL;
}
if (!chip->gpio_detect && chip->is_present != (ret >= 0)) {
bool old_present = chip->is_present;
union power_supply_propval val;
int err = sbs_get_battery_presence_and_health(
client, POWER_SUPPLY_PROP_PRESENT, &val);
sbs_update_presence(chip, !err && val.intval);
if (old_present != chip->is_present)
power_supply_changed(chip->power_supply);
}
done:
if (!ret) {
/* Convert units to match requirements for power supply class */
sbs_unit_adjustment(client, psp, val);
dev_dbg(&client->dev,
"%s: property = %d, value = %x\n", __func__,
psp, val->intval);
} else if (!chip->is_present) {
/* battery not present, so return NODATA for properties */
ret = -ENODATA;
}
return ret;
}
static void sbs_supply_changed(struct sbs_info *chip)
{
struct power_supply *battery = chip->power_supply;
int ret;
ret = gpiod_get_value_cansleep(chip->gpio_detect);
if (ret < 0)
return;
sbs_update_presence(chip, ret);
power_supply_changed(battery);
}
static irqreturn_t sbs_irq(int irq, void *devid)
{
sbs_supply_changed(devid);
return IRQ_HANDLED;
}
static void sbs_alert(struct i2c_client *client, enum i2c_alert_protocol prot,
unsigned int data)
{
sbs_supply_changed(i2c_get_clientdata(client));
}
static void sbs_external_power_changed(struct power_supply *psy)
{
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
struct sbs_info *chip = power_supply_get_drvdata(psy);
/* cancel outstanding work */
cancel_delayed_work_sync(&chip->work);
schedule_delayed_work(&chip->work, HZ);
chip->poll_time = chip->poll_retry_count;
}
static void sbs_delayed_work(struct work_struct *work)
{
struct sbs_info *chip;
s32 ret;
chip = container_of(work, struct sbs_info, work.work);
ret = sbs_read_word_data(chip->client, sbs_data[REG_STATUS].addr);
/* if the read failed, give up on this work */
if (ret < 0) {
chip->poll_time = 0;
return;
}
if (ret & BATTERY_FULL_CHARGED)
ret = POWER_SUPPLY_STATUS_FULL;
else if (ret & BATTERY_DISCHARGING)
ret = POWER_SUPPLY_STATUS_DISCHARGING;
else
ret = POWER_SUPPLY_STATUS_CHARGING;
sbs_status_correct(chip->client, &ret);
if (chip->last_state != ret) {
chip->poll_time = 0;
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
power_supply_changed(chip->power_supply);
return;
}
if (chip->poll_time > 0) {
schedule_delayed_work(&chip->work, HZ);
chip->poll_time--;
return;
}
}
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
static const struct power_supply_desc sbs_default_desc = {
.type = POWER_SUPPLY_TYPE_BATTERY,
.properties = sbs_properties,
.num_properties = ARRAY_SIZE(sbs_properties),
.get_property = sbs_get_property,
.external_power_changed = sbs_external_power_changed,
};
static int sbs_probe(struct i2c_client *client)
{
struct sbs_info *chip;
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
struct power_supply_desc *sbs_desc;
struct sbs_platform_data *pdata = client->dev.platform_data;
struct power_supply_config psy_cfg = {};
int rc;
int irq;
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
sbs_desc = devm_kmemdup(&client->dev, &sbs_default_desc,
sizeof(*sbs_desc), GFP_KERNEL);
if (!sbs_desc)
return -ENOMEM;
sbs_desc->name = devm_kasprintf(&client->dev, GFP_KERNEL, "sbs-%s",
dev_name(&client->dev));
if (!sbs_desc->name)
return -ENOMEM;
chip = devm_kzalloc(&client->dev, sizeof(struct sbs_info), GFP_KERNEL);
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
if (!chip)
return -ENOMEM;
chip->flags = (u32)(uintptr_t)device_get_match_data(&client->dev);
chip->client = client;
psy_cfg.of_node = client->dev.of_node;
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
psy_cfg.drv_data = chip;
chip->last_state = POWER_SUPPLY_STATUS_UNKNOWN;
sbs_invalidate_cached_props(chip);
mutex_init(&chip->mode_lock);
/* use pdata if available, fall back to DT properties,
* or hardcoded defaults if not
*/
rc = device_property_read_u32(&client->dev, "sbs,i2c-retry-count",
&chip->i2c_retry_count);
if (rc)
chip->i2c_retry_count = 0;
rc = device_property_read_u32(&client->dev, "sbs,poll-retry-count",
&chip->poll_retry_count);
if (rc)
chip->poll_retry_count = 0;
if (pdata) {
chip->poll_retry_count = pdata->poll_retry_count;
chip->i2c_retry_count = pdata->i2c_retry_count;
}
chip->i2c_retry_count = chip->i2c_retry_count + 1;
chip->charger_broadcasts = !device_property_read_bool(&client->dev,
"sbs,disable-charger-broadcasts");
chip->gpio_detect = devm_gpiod_get_optional(&client->dev,
"sbs,battery-detect", GPIOD_IN);
if (IS_ERR(chip->gpio_detect))
return dev_err_probe(&client->dev, PTR_ERR(chip->gpio_detect),
"Failed to get gpio\n");
i2c_set_clientdata(client, chip);
if (!chip->gpio_detect)
goto skip_gpio;
irq = gpiod_to_irq(chip->gpio_detect);
if (irq <= 0) {
dev_warn(&client->dev, "Failed to get gpio as irq: %d\n", irq);
goto skip_gpio;
}
rc = devm_request_threaded_irq(&client->dev, irq, NULL, sbs_irq,
IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
dev_name(&client->dev), chip);
if (rc) {
dev_warn(&client->dev, "Failed to request irq: %d\n", rc);
goto skip_gpio;
}
skip_gpio:
/*
sbs-battery: add option to always register battery Commit a22b41a31e53 ("sbs-battery: Probe should try talking to the device") introduced a step in probing the SBS battery, that tries to talk to the device before actually registering it, saying: this driver doesn't actually try talking to the device at probe time, so if it's incorrectly configured in the device tree or platform data (or if the battery has been removed from the system), then probe will succeed and every access will sit there and time out. The end result is a possibly laggy system that thinks it has a battery but can never read status, which isn't very useful. Which is of course reasonable. However, it is also very well possible for a device to boot up on wall-power and be connected to a battery later on. This is a scenario that the driver supported before said patch was applied, and even easily achieved by booting up with the battery attached and removing it later on. sbs-battery's 'present' sysfs file can be used to determine if the device is available or not. So with automated device detection lacking for now, in some cases it is possible that one wants to register a battery, even if none are attached at the moment. To facilitate this, add a module parameter that can be used to configure forced loading module loading time. If set, the battery will always be registered without checking the sanity of the connection. Signed-off-by: Frans Klaver <frans.klaver@xsens.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-06-10 12:16:56 +00:00
* Before we register, we might need to make sure we can actually talk
* to the battery.
*/
if (!(force_load || chip->gpio_detect)) {
union power_supply_propval val;
sbs-battery: add option to always register battery Commit a22b41a31e53 ("sbs-battery: Probe should try talking to the device") introduced a step in probing the SBS battery, that tries to talk to the device before actually registering it, saying: this driver doesn't actually try talking to the device at probe time, so if it's incorrectly configured in the device tree or platform data (or if the battery has been removed from the system), then probe will succeed and every access will sit there and time out. The end result is a possibly laggy system that thinks it has a battery but can never read status, which isn't very useful. Which is of course reasonable. However, it is also very well possible for a device to boot up on wall-power and be connected to a battery later on. This is a scenario that the driver supported before said patch was applied, and even easily achieved by booting up with the battery attached and removing it later on. sbs-battery's 'present' sysfs file can be used to determine if the device is available or not. So with automated device detection lacking for now, in some cases it is possible that one wants to register a battery, even if none are attached at the moment. To facilitate this, add a module parameter that can be used to configure forced loading module loading time. If set, the battery will always be registered without checking the sanity of the connection. Signed-off-by: Frans Klaver <frans.klaver@xsens.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-06-10 12:16:56 +00:00
rc = sbs_get_battery_presence_and_health(
client, POWER_SUPPLY_PROP_PRESENT, &val);
if (rc < 0 || !val.intval)
return dev_err_probe(&client->dev, -ENODEV,
"Failed to get present status\n");
}
rc = devm_delayed_work_autocancel(&client->dev, &chip->work,
sbs_delayed_work);
if (rc)
return rc;
chip->power_supply = devm_power_supply_register(&client->dev, sbs_desc,
power_supply: Change ownership from driver to core Change the ownership of power_supply structure from each driver implementing the class to the power supply core. The patch changes power_supply_register() function thus all drivers implementing power supply class are adjusted. Each driver provides the implementation of power supply. However it should not be the owner of power supply class instance because it is exposed by core to other subsystems with power_supply_get_by_name(). These other subsystems have no knowledge when the driver will unregister the power supply. This leads to several issues when driver is unbound - mostly because user of power supply accesses freed memory. Instead let the core own the instance of struct 'power_supply'. Other users of this power supply will still access valid memory because it will be freed when device reference count reaches 0. Currently this means "it will leak" but power_supply_put() call in next patches will solve it. This solves invalid memory references in following race condition scenario: Thread 1: charger manager Thread 2: power supply driver, used by charger manager THREAD 1 (charger manager) THREAD 2 (power supply driver) ========================== ============================== psy = power_supply_get_by_name() Driver unbind, .remove power_supply_unregister() Device fully removed psy->get_property() The 'get_property' call is executed in invalid context because the driver was unbound and struct 'power_supply' memory was freed. This could be observed easily with charger manager driver (here compiled with max17040 fuel gauge): $ cat /sys/devices/virtual/power_supply/cm-battery/capacity & $ echo "1-0036" > /sys/bus/i2c/drivers/max17040/unbind [ 55.725123] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 55.732584] pgd = d98d4000 [ 55.734060] [00000000] *pgd=5afa2831, *pte=00000000, *ppte=00000000 [ 55.740318] Internal error: Oops: 80000007 [#1] PREEMPT SMP ARM [ 55.746210] Modules linked in: [ 55.749259] CPU: 1 PID: 2936 Comm: cat Tainted: G W 3.19.0-rc1-next-20141226-00048-gf79f475f3c44-dirty #1496 [ 55.760190] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 55.766270] task: d9b76f00 ti: daf54000 task.ti: daf54000 [ 55.771647] PC is at 0x0 [ 55.774182] LR is at charger_get_property+0x2f4/0x36c [ 55.779201] pc : [<00000000>] lr : [<c034b0b4>] psr: 60000013 [ 55.779201] sp : daf55e90 ip : 00000003 fp : 00000000 [ 55.790657] r10: 00000000 r9 : c06e2878 r8 : d9b26c68 [ 55.795865] r7 : dad81610 r6 : daec7410 r5 : daf55ebc r4 : 00000000 [ 55.802367] r3 : 00000000 r2 : daf55ebc r1 : 0000002a r0 : d9b26c68 [ 55.808879] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 55.815994] Control: 10c5387d Table: 598d406a DAC: 00000015 [ 55.821723] Process cat (pid: 2936, stack limit = 0xdaf54210) [ 55.827451] Stack: (0xdaf55e90 to 0xdaf56000) [ 55.831795] 5e80: 60000013 c01459c4 0000002a c06f8ef8 [ 55.839956] 5ea0: db651000 c06f8ef8 daebac00 c04cb668 daebac08 c0346864 00000000 c01459c4 [ 55.848115] 5ec0: d99eaa80 c06f8ef8 00000fff 00001000 db651000 c027f25c c027f240 d99eaa80 [ 55.856274] 5ee0: d9a06c00 c0146218 daf55f18 00001000 d99eaa80 db4c18c0 00000001 00000001 [ 55.864468] 5f00: daf55f80 c0144c78 c0144c54 c0107f90 00015000 d99eaab0 00000000 00000000 [ 55.872603] 5f20: 000051c7 00000000 db4c18c0 c04a9370 00015000 00001000 daf55f80 00001000 [ 55.880763] 5f40: daf54000 00015000 00000000 c00e53dc db4c18c0 c00e548c 0000000d 00008124 [ 55.888937] 5f60: 00000001 00000000 00000000 db4c18c0 db4c18c0 00001000 00015000 c00e5550 [ 55.897099] 5f80: 00000000 00000000 00001000 00001000 00015000 00000003 00000003 c000f364 [ 55.905239] 5fa0: 00000000 c000f1a0 00001000 00015000 00000003 00015000 00001000 0001333c [ 55.913399] 5fc0: 00001000 00015000 00000003 00000003 00000002 00000000 00000000 00000000 [ 55.921560] 5fe0: 7fffe000 be999850 0000a225 b6f3c19c 60000010 00000003 00000000 00000000 [ 55.929744] [<c034b0b4>] (charger_get_property) from [<c0346864>] (power_supply_show_property+0x48/0x20c) [ 55.939286] [<c0346864>] (power_supply_show_property) from [<c027f25c>] (dev_attr_show+0x1c/0x48) [ 55.948130] [<c027f25c>] (dev_attr_show) from [<c0146218>] (sysfs_kf_seq_show+0x84/0x104) [ 55.956298] [<c0146218>] (sysfs_kf_seq_show) from [<c0144c78>] (kernfs_seq_show+0x24/0x28) [ 55.964536] [<c0144c78>] (kernfs_seq_show) from [<c0107f90>] (seq_read+0x1b0/0x484) [ 55.972172] [<c0107f90>] (seq_read) from [<c00e53dc>] (__vfs_read+0x18/0x4c) [ 55.979188] [<c00e53dc>] (__vfs_read) from [<c00e548c>] (vfs_read+0x7c/0x100) [ 55.986304] [<c00e548c>] (vfs_read) from [<c00e5550>] (SyS_read+0x40/0x8c) [ 55.993164] [<c00e5550>] (SyS_read) from [<c000f1a0>] (ret_fast_syscall+0x0/0x48) [ 56.000626] Code: bad PC value [ 56.011652] ---[ end trace 7b64343fbdae8ef1 ]--- Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> [for the nvec part] Reviewed-by: Marc Dietrich <marvin24@gmx.de> [for compal-laptop.c] Acked-by: Darren Hart <dvhart@linux.intel.com> [for the mfd part] Acked-by: Lee Jones <lee.jones@linaro.org> [for the hid part] Acked-by: Jiri Kosina <jkosina@suse.cz> [for the acpi part] Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-03-12 07:44:11 +00:00
&psy_cfg);
if (IS_ERR(chip->power_supply))
return dev_err_probe(&client->dev, PTR_ERR(chip->power_supply),
"Failed to register power supply\n");
dev_info(&client->dev,
"%s: battery gas gauge device registered\n", client->name);
return 0;
}
#if defined CONFIG_PM_SLEEP
static int sbs_suspend(struct device *dev)
{
struct i2c_client *client = to_i2c_client(dev);
struct sbs_info *chip = i2c_get_clientdata(client);
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
int ret;
if (chip->poll_time > 0)
cancel_delayed_work_sync(&chip->work);
if (chip->flags & SBS_FLAGS_TI_BQ20ZX5) {
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
/* Write to manufacturer access with sleep command. */
ret = sbs_write_word_data(client,
sbs_data[REG_MANUFACTURER_DATA].addr,
MANUFACTURER_ACCESS_SLEEP);
if (chip->is_present && ret < 0)
return ret;
}
return 0;
}
static SIMPLE_DEV_PM_OPS(sbs_pm_ops, sbs_suspend, NULL);
#define SBS_PM_OPS (&sbs_pm_ops)
#else
#define SBS_PM_OPS NULL
#endif
static const struct i2c_device_id sbs_id[] = {
{ "bq20z65", 0 },
{ "bq20z75", 0 },
{ "sbs-battery", 1 },
{}
};
MODULE_DEVICE_TABLE(i2c, sbs_id);
static const struct of_device_id sbs_dt_ids[] = {
{ .compatible = "sbs,sbs-battery" },
{
.compatible = "ti,bq20z65",
.data = (void *)SBS_FLAGS_TI_BQ20ZX5,
},
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
{
.compatible = "ti,bq20z75",
.data = (void *)SBS_FLAGS_TI_BQ20ZX5,
power: supply: sbs-battery: don't assume MANUFACTURER_DATA formats This driver was originally submitted for the TI BQ20Z75 battery IC (commit a7640bfa10c5 ("power_supply: Add driver for TI BQ20Z75 gas gauge IC")) and later renamed to express generic SBS support. While it's mostly true that this driver implemented a standard SBS command set, it takes liberties with the REG_MANUFACTURER_DATA register. This register is specified in the SBS spec, but it doesn't make any mention of what its actual contents are. We've sort of noticed this optionality previously, with commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional"), where we found that some batteries NAK writes to this register. What this really means is that so far, we've just been lucky that most batteries have either been compatible with the TI chip, or else at least haven't reported highly-unexpected values. For instance, one battery I have here seems to report either 0x0000 or 0x0100 to the MANUFACTURER_ACCESS_STATUS command -- while this seems to match either Wake Up (bits[11:8] = 0000b) or Normal Discharge (bits[11:8] = 0001b) status for the TI part [1], they don't seem to actually correspond to real states (for instance, I never see 0101b = Charge, even when charging). On other batteries, I'm getting apparently random data in return, which means that occasionally, we interpret this as "battery not present" or "battery is not healthy". All in all, it seems to be a really bad idea to make assumptions about REG_MANUFACTURER_DATA, unless we already know what battery we're using. Therefore, this patch reimplements the "present" and "health" checks to the following on most SBS batteries: 1. HEALTH: report "unknown" -- I couldn't find a standard SBS command that gives us much useful here 2. PRESENT: just send a REG_STATUS command; if it succeeds, then the battery is present Also, we stop sending MANUFACTURER_ACCESS_SLEEP to non-TI parts. I have no proof that this is useful and supported. If someone explicitly provided a 'ti,bq20z75' compatible property, then we continue to use the existing TI command behaviors, and we effectively revert commit 17c6d3979e5b ("sbs-battery: make writes to ManufacturerAccess optional") to again make these commands required. [1] http://www.ti.com/lit/er/sluu265a/sluu265a.pdf Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Guenter Roeck <linux@roeck-us.net> Acked-by: Rhyland Klein <rklein@nvidia.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
2018-06-12 20:20:41 +00:00
},
{ }
};
MODULE_DEVICE_TABLE(of, sbs_dt_ids);
static struct i2c_driver sbs_battery_driver = {
.probe_new = sbs_probe,
.alert = sbs_alert,
.id_table = sbs_id,
.driver = {
.name = "sbs-battery",
.of_match_table = sbs_dt_ids,
.pm = SBS_PM_OPS,
},
};
module_i2c_driver(sbs_battery_driver);
MODULE_DESCRIPTION("SBS battery monitor driver");
MODULE_LICENSE("GPL");
sbs-battery: add option to always register battery Commit a22b41a31e53 ("sbs-battery: Probe should try talking to the device") introduced a step in probing the SBS battery, that tries to talk to the device before actually registering it, saying: this driver doesn't actually try talking to the device at probe time, so if it's incorrectly configured in the device tree or platform data (or if the battery has been removed from the system), then probe will succeed and every access will sit there and time out. The end result is a possibly laggy system that thinks it has a battery but can never read status, which isn't very useful. Which is of course reasonable. However, it is also very well possible for a device to boot up on wall-power and be connected to a battery later on. This is a scenario that the driver supported before said patch was applied, and even easily achieved by booting up with the battery attached and removing it later on. sbs-battery's 'present' sysfs file can be used to determine if the device is available or not. So with automated device detection lacking for now, in some cases it is possible that one wants to register a battery, even if none are attached at the moment. To facilitate this, add a module parameter that can be used to configure forced loading module loading time. If set, the battery will always be registered without checking the sanity of the connection. Signed-off-by: Frans Klaver <frans.klaver@xsens.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-06-10 12:16:56 +00:00
module_param(force_load, bool, 0444);
sbs-battery: add option to always register battery Commit a22b41a31e53 ("sbs-battery: Probe should try talking to the device") introduced a step in probing the SBS battery, that tries to talk to the device before actually registering it, saying: this driver doesn't actually try talking to the device at probe time, so if it's incorrectly configured in the device tree or platform data (or if the battery has been removed from the system), then probe will succeed and every access will sit there and time out. The end result is a possibly laggy system that thinks it has a battery but can never read status, which isn't very useful. Which is of course reasonable. However, it is also very well possible for a device to boot up on wall-power and be connected to a battery later on. This is a scenario that the driver supported before said patch was applied, and even easily achieved by booting up with the battery attached and removing it later on. sbs-battery's 'present' sysfs file can be used to determine if the device is available or not. So with automated device detection lacking for now, in some cases it is possible that one wants to register a battery, even if none are attached at the moment. To facilitate this, add a module parameter that can be used to configure forced loading module loading time. If set, the battery will always be registered without checking the sanity of the connection. Signed-off-by: Frans Klaver <frans.klaver@xsens.com> Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-06-10 12:16:56 +00:00
MODULE_PARM_DESC(force_load,
"Attempt to load the driver even if no battery is connected");