linux_dsm_epyc7002/drivers/iio
Fabrice Gasnier 7f75591fc5 iio: core: fix a possible circular locking dependency
This fixes a possible circular locking dependency detected warning seen
with:
- CONFIG_PROVE_LOCKING=y
- consumer/provider IIO devices (ex: "voltage-divider" consumer of "adc")

When using the IIO consumer interface, e.g. iio_channel_get(), the consumer
device will likely call iio_read_channel_raw() or similar that rely on
'info_exist_lock' mutex.

typically:
...
	mutex_lock(&chan->indio_dev->info_exist_lock);
	if (chan->indio_dev->info == NULL) {
		ret = -ENODEV;
		goto err_unlock;
	}
	ret = do_some_ops()
err_unlock:
	mutex_unlock(&chan->indio_dev->info_exist_lock);
	return ret;
...

Same mutex is also hold in iio_device_unregister().

The following deadlock warning happens when:
- the consumer device has called an API like iio_read_channel_raw()
  at least once.
- the consumer driver is unregistered, removed (unbind from sysfs)

======================================================
WARNING: possible circular locking dependency detected
4.19.24 #577 Not tainted
------------------------------------------------------
sh/372 is trying to acquire lock:
(kn->count#30){++++}, at: kernfs_remove_by_name_ns+0x3c/0x84

but task is already holding lock:
(&dev->info_exist_lock){+.+.}, at: iio_device_unregister+0x18/0x60

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&dev->info_exist_lock){+.+.}:
       __mutex_lock+0x70/0xa3c
       mutex_lock_nested+0x1c/0x24
       iio_read_channel_raw+0x1c/0x60
       iio_read_channel_info+0xa8/0xb0
       dev_attr_show+0x1c/0x48
       sysfs_kf_seq_show+0x84/0xec
       seq_read+0x154/0x528
       __vfs_read+0x2c/0x15c
       vfs_read+0x8c/0x110
       ksys_read+0x4c/0xac
       ret_fast_syscall+0x0/0x28
       0xbedefb60

-> #0 (kn->count#30){++++}:
       lock_acquire+0xd8/0x268
       __kernfs_remove+0x288/0x374
       kernfs_remove_by_name_ns+0x3c/0x84
       remove_files+0x34/0x78
       sysfs_remove_group+0x40/0x9c
       sysfs_remove_groups+0x24/0x34
       device_remove_attrs+0x38/0x64
       device_del+0x11c/0x360
       cdev_device_del+0x14/0x2c
       iio_device_unregister+0x24/0x60
       release_nodes+0x1bc/0x200
       device_release_driver_internal+0x1a0/0x230
       unbind_store+0x80/0x130
       kernfs_fop_write+0x100/0x1e4
       __vfs_write+0x2c/0x160
       vfs_write+0xa4/0x17c
       ksys_write+0x4c/0xac
       ret_fast_syscall+0x0/0x28
       0xbe906840

other info that might help us debug this:

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&dev->info_exist_lock);
                               lock(kn->count#30);
                               lock(&dev->info_exist_lock);
  lock(kn->count#30);

 *** DEADLOCK ***
...

cdev_device_del() can be called without holding the lock. It should be safe
as info_exist_lock prevents kernelspace consumers to use the exported
routines during/after provider removal. cdev_device_del() is for userspace.

Help to reproduce:
See example: Documentation/devicetree/bindings/iio/afe/voltage-divider.txt
sysv {
	compatible = "voltage-divider";
	io-channels = <&adc 0>;
	output-ohms = <22>;
	full-ohms = <222>;
};

First, go to iio:deviceX for the "voltage-divider", do one read:
$ cd /sys/bus/iio/devices/iio:deviceX
$ cat in_voltage0_raw

Then, unbind the consumer driver. It triggers above deadlock warning.
$ cd /sys/bus/platform/drivers/iio-rescale/
$ echo sysv > unbind

Note I don't actually expect stable will pick this up all the
way back into IIO being in staging, but if's probably valid that
far back.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Fixes: ac917a8111 ("staging:iio:core set the iio_dev.info pointer to null on unregister")
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
2019-03-31 11:11:31 +01:00
..
accel io: accel: kxcjk1013: restore the range after resume. 2019-03-09 17:30:14 +00:00
adc iio: ad_sigma_delta: select channel when reading register 2019-03-24 17:51:03 +00:00
afe iio: afe: rescale: new driver 2018-04-28 18:03:22 +01:00
amplifiers drivers: iio: Update MODULE AUTHOR email address 2018-08-19 17:23:39 +01:00
buffer iio:kfifo_buf: check for uint overflow 2018-03-30 11:16:12 +01:00
chemical iio: pms7003: select IIO_TRIGGERED_BUFFER 2019-03-16 16:34:56 +00:00
common iio: cros_ec: Fix the maths for gyro scale calculation 2019-03-16 16:08:55 +00:00
counter iio: 104-quad-8: Provide defines for magic numbers 2018-06-10 11:52:30 +01:00
dac iio: dac: mcp4725: add missing powerdown bits in store eeprom 2019-03-16 15:15:31 +00:00
dummy iio:dummy: Add extra paragraphs on Kconfig 2018-03-03 15:20:16 +00:00
frequency iio: ad9523: fix a missing check of return value 2019-01-05 16:10:50 +00:00
gyro iio: gyro: mpu3050: fix chip ID reading 2019-03-09 16:55:06 +00:00
health iio: health: max30102: Mark expected switch fall-throughs 2018-08-19 17:21:35 +01:00
humidity iio/hid-sensors: Fix IIO_CHAN_INFO_RAW returning wrong values for signed numbers 2018-11-16 11:42:12 +00:00
imu iio:bmi160: use if (ret) instead of if (ret < 0) 2019-02-09 15:25:25 +00:00
light iio: light: add driver for MAX44009 2019-02-02 15:47:46 +00:00
magnetometer iio: magnetometer: mag3110: add vdd/vddio regulator operation support 2019-01-12 19:00:23 +00:00
multiplexer iio: multiplexer: switch to SPDX license identifier 2018-08-20 19:28:26 +01:00
orientation iio/hid-sensors: Fix IIO_CHAN_INFO_RAW returning wrong values for signed numbers 2018-11-16 11:42:12 +00:00
potentiometer iio: potentiometer: Add driver for Microchip MCP41xxx/42xxx 2018-11-25 12:35:26 +00:00
potentiostat iio: potentiostat: lmp91000: add LMP91002 support 2018-05-12 11:02:44 +01:00
pressure iio:st_pressure:initial lps22hh sensor support 2019-01-19 17:32:41 +00:00
proximity iio: proximity: Add driver support for ST's VL53L0X ToF ranging sensor. 2018-09-22 16:42:56 +01:00
resolver staging:iio:ad2s90: Move out of staging 2018-12-01 15:40:39 +00:00
temperature iio/hid-sensors: Fix IIO_CHAN_INFO_RAW returning wrong values for signed numbers 2018-11-16 11:42:12 +00:00
trigger drivers: iio: Update MODULE AUTHOR email address 2018-08-19 17:23:39 +01:00
iio_core_trigger.h
iio_core.h the rest of drivers/*: annotate ->poll() instances 2017-11-28 11:06:58 -05:00
industrialio-buffer.c iio: Fix scan mask selection 2019-03-09 16:55:06 +00:00
industrialio-configfs.c iio: make function argument and some structures const 2017-10-19 16:15:24 +02:00
industrialio-core.c iio: core: fix a possible circular locking dependency 2019-03-31 11:11:31 +01:00
industrialio-event.c vfs: do bulk POLL* -> EPOLL* replacement 2018-02-11 14:34:03 -08:00
industrialio-sw-device.c iio: make function argument and some structures const 2017-10-19 16:15:24 +02:00
industrialio-sw-trigger.c iio: make function argument and some structures const 2017-10-19 16:15:24 +02:00
industrialio-trigger.c iio:trigger: Remove necessity to have a trig->ops structure. 2017-08-22 21:20:28 +01:00
industrialio-triggered-event.c
inkern.c treewide: kzalloc() -> kcalloc() 2018-06-12 16:19:22 -07:00
Kconfig staging: iio: ad2s1200: Move driver out of staging 2018-05-20 12:17:48 +01:00
Makefile staging: iio: ad2s1200: Move driver out of staging 2018-05-20 12:17:48 +01:00