| Age | Commit message (Collapse) | Author |
|---|
|
commit f8cd222feb82ecd82dcf610fcc15186f55f9c2b5 upstream.
One of a class of bugs pointed out by Lars in a recent review.
iio_push_to_buffers_with_timestamp assumes the buffer used is aligned
to the size of the timestamp (8 bytes). This is not guaranteed in
this driver which uses a 32 byte array of smaller elements on the stack.
As Lars also noted this anti pattern can involve a leak of data to
userspace and that indeed can happen here. We close both issues by
moving to a suitable structure in the iio_priv() data with alignment
explicitly requested. This data is allocated with kzalloc so no
data can leak apart from previous readings. The explicit alignment
isn't technically needed here, but it reduced fragility and avoids
cut and paste into drivers where it will be needed.
If we want this in older stables will need manual backport due to
driver reworks.
Fixes: c43a102e67db ("iio: ina2xx: add support for TI INA2xx Power Monitors")
Reported-by: Lars-Peter Clausen <lars@metafoo.de>
Cc: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Cc: Marc Titinger <mtitinger@baylibre.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: <Stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
When the buffer is enabled for ina2xx driver, a dedicated kthread is
invoked to capture mesurement data. When the buffer is disabled, the
kthread is stopped.
However if the kthread gets register access errors, it immediately exits
and when the malfunctional buffer is disabled, the stale task_struct
pointer is accessed as there is no kthread to be stopped.
A similar issue in the usbip driver is prevented by kthread_get_run and
kthread_stop_put helpers by increasing usage count of the task_struct.
This change applies the same solution.
Cc: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Cc: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Fixes: c43a102e67db ("iio: ina2xx: add support for TI INA2xx Power Monitors")
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Currently, the registers are read out once per conversion interval. If
the reading is delayed as the conversion has not yet finished, this extra
time is treated as being part of the readout, although it should delay
the start of the poll interval. This results in the interval starting
slightly earlier in each iteration, until all time between reads is
spent polling the status registers instead of sleeping.
To fix this, the delay has to account for the state of the conversion
ready flag. Whenever the conversion is already finished, schedule the next
read on the regular interval, otherwise schedule it one interval after the
flag bit has been set.
Split the work function in two functions, one for the status poll and one
for reading the values, to be able to note down the time when the flag
bit is raised.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
As the timestamp is no longer (ab-)used to measure the function run time,
it can be taken at the correct time, i.e. when the conversion has finished.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The iio timestamp clock is user selectable and may be non-monotonic. Also,
only part of the acquisition time is measured, thus the delay was longer
than intended.
Use a monotonic timestamp to track the time for the next poll iteration.
The timestamp is advanced by the sampling interval each iteration. In case
the conversion overrruns the register readout (i.e. fast sampling combined
with a slow bus), one or multiple samples will be dropped.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Calibration register is used for calculating current register in
hardware according to datasheet:
current = shunt_volt * calib_register / 2048 (ina 226)
current = shunt_volt * calib_register / 4096 (ina 219)
Fix calib_register value to 2048 for ina226 and 4096 for ina 219 in
order to avoid truncation error and provide best precision allowed
by shunt_voltage measurement. Make current scale value follow changes
of shunt_resistor from sysfs as calib_register value is now fixed.
Power_lsb value should also follow shunt_resistor changes as stated in
datasheet:
power_lsb = 25 * current_lsb (ina 226)
power_lsb = 20 * current_lsb (ina 219)
This is a part of the patchset: https://lkml.org/lkml/2017/11/22/394
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Reviewed-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The conversion time can be up to 16 seconds (8 ms per channel, 2 channels,
1024 times averaging).
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Although the datasheet states the CNVR flag is cleared by reading the
BUS_VOLTAGE register, it is actually cleared by reading any of the
voltage/current/power registers.
The behaviour has been confirmed by TI support:
http://e2e.ti.com/support/amplifiers/current-shunt-monitors/f/931/p/647053/2378282
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The timestamp is inserted into the buffer after the sample data by
iio_push_to_buffers_with_timestamp, document the space requirement for
the timestamp.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
iio_push_to_buffers_with_timestamp expects a void pointer, so the cast
is both unnecessary and misleading.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Reducing shunt and bus voltage range improves the accuracy, so allow
altering the default settings.
Both settings are exposed as gain values. While for the shunt voltage
this is straightforward, the bus range settings of 32V (default) and 16V
are mapped to gain values of 1 resp. 2, to provide a uniform API to
userspace.
As the gain settings are incorporated into the raw values by the sensor
itself, adjusting of the scale attributes is not necessary.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
While the config uses the physical value corresponding to the LSB
for both the power and the bus voltage register, the shunt voltage is
specified as parts of 1 mV. Use the LSB physical value for all registers.
No functional change.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Lower bits of the INA219/220 bus voltage register are conversion
status flags, properly shift the value.
When reading via IIO buffer, the value is passed on unaltered,
shifting is the responsibility of the user.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
According to the ABI documentation, the shunt resistor value should be
specificied in Ohm. As this is also used/documented for the MAX9611,
use the same for the INA2xx driver.
This poses an ABI break for anyone actually altering the shunt value
through the sysfs interface, it does not alter the default value nor
a value set from the devicetree.
Minor change: Fix comment, 1mA is 10^-3A.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next
Jonathan writes:
Round one of new device support, features and cleanup for IIO in the 4.15 cycle.
Note there is a misc driver drop in here given we have support
in IIO and the feeling is no one will care.
A large part of this series is a boiler plate removal series avoiding
the need to explicitly provide THIS_MODULE in various locations.
It's very dull but touches all drivers.
New device support
* ad5446
- add ids to support compatible parts DAC081S101, DAC101S101,
DAC121S101.
- add the dac7512 id and drop the misc driver as feeling is no
one is using it (was introduced for a board that is long obsolete)
* mt6577
- add bindings for mt2712 which is fully compatible with other
supported parts.
* st_pressure
- add support for LPS33HW and LPS35HW with bindings (ids mostly).
New features
* ccs811
- Add support for the data ready trigger.
* mma8452
- remove artifical restriction on supporting multiple event types
at the same time.
* tcs3472
- support out of threshold events
Core and tree wide cleanup
* Use macro magic to remove the need to provide THIS_MODULE as part of
struct iio_info or struct iio_trigger_ops. This is similar to
work done in a number of other subsystems (e.g. i2c, spi).
All drivers are fixed and then the fields in these structures are
removed.
This will cause build failures for out of tree drivers and any
new drivers that cross with this work going into the kernel.
Note mostly done with a coccinelle patch, included in the series
on the mailing list but not merged as the fields no longer exist
in the structures so the any hold outs will cause a build failure.
Cleanups
* ads1015
- avoid writing config register when it doesn't change.
- add 10% to conversion wait time as it seems it is sometimes
a little small.
* ade7753
- replace use of core mlock with a local lock. This is part of a
long term effort to make the use of mlock opaque and single
purpose.
* ade7759
- expand the use of buf_lock to cover previous mlock cases. This
is a slightly nicer solution to the same issue as in ade7753.
* cros_ec
- drop an unused variable
* inv_mpu6050
- add a missing break in a switch for consistency - not actual
bug,
- make some local arrays static to save on object code size.
* max5481
- drop manual setting of the spi module owner as handled by the
spi core.
* max5487
- drop manual setting of the spi module owner as handled by the
spi core.
* max9611
- drop explicit setting of the i2c module owner as handled by
the i2c core.
* mcp320x
- speed up reads on single channel devices,
- drop unused of_device_id data elements,
- document the struct mcp320x,
- improve binding docs to reflect restrictions on spi setup and
to make it explicit that the reference regulator is needed.
* mma8452
- symbolic to octal permissions,
- unsigned to unsigned int.
* st_lsm6dsx
- avoid setting odr values multiple times,
- drop config of LIR as it is only ever set to the existing
defaults,
- drop rounding configuration as it only ever matches the defaults.
* ti-ads8688
- drop manual setting of the spi module owner as handled by the
spi core.
* tsl2x7x
- constify the i2c_device_id,
- cleanup limit checks to avoid static checker warnings (and generally
have nicer code).
|
|
We want the staging and iio fixes in here to handle the merge issues.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The equivalent of both of these are now done via macro magic when
the relevant register calls are made. The actual structure
elements will shortly go away.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Lars-Peter Clausen <lars@metafoo.de>
|
|
Proper support for the INA219 lowered the minimum sampling period from
2*140us to 2*84us. Subtracting 200us later leads to an underflow and
an almost infinite udelay later.
Using a signed int for the sampling period provides sufficient range
(at most 2*8640*1024us), but catches the underflow when comparing with
buffer_us.
Fixes: 18edac2e22f4 ("iio: adc: Fix integration time/averaging for INA219/220")
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
While the INA226 has a conversion ready flag (CVRF) in the R/W Mask/Enable
register with read-to-clear semantics, the corresponding bit of the INA219
(CNVR) is part of the bus voltage register. The flag is cleared by reading
the power register.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
INA226/230/231 has integration times per voltage channel and common
averaging setting for both channels, while the INA219/220 only has a
combined integration time/averaging setting per channel.
Only expose the averaging attribute for the INA226, and expose the correct
integration times for the INA219.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
Flags for shared channel attributes should be set on all channels of a
channel set. I.e. SAMP_FREQUENCY and OVERSAMPLING_RATIO are set on the
in_voltage{0,1} channels, thus should be set on in_power, in_current.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
Commit 7906dd52c5a0 ("iio: ina2xx: Fix whitespace and re-order code")
changed the register number of the MASK_ENABLE register from 0x06 to the
value equivalent GENMASK(2,1), although its no mask.
Also fix a typo (INA2_6_6 instead of INA2_2_6), and use the datasheet
name ("Mask/Enable") for the register number define.
Fix bad indentation for channel attributes.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Acked-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
The driver doesn't have a struct of_device_id table but supported devices
are registered via Device Trees. This is working on the assumption that a
I2C device registered via OF will always match a legacy I2C device ID and
that the MODALIAS reported will always be of the form i2c:<device>.
But this could change in the future so the correct approach is to have an
OF device ID table if the devices are registered via OF.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
As a precursor to splitting buffer.h, lets make sure all drivers
include the relevant headers rather than relying on picking them
up from kfifo_buf.h.
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Reviewed-by: Lars-Peter Clausen <lars@metafoo.de>
|
|
commit 1961bce76452 "iio: ina2xx: Remove trace_printk debug
statements" removed the code that used the chip->prev_ns field.
This patch cleans it up further by removing the unused field
and assignments.
Signed-off-by: Alison Schofield <amsfield22@gmail.com>
Cc: Daniel Baluta <daniel.baluta@gmail.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
Adding missing indio_dev->dev.of_node references to allow iio consumers
to access the device channels.
Signed-off-by: Matt Ranostay <mranostay@gmail.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
Adds a new per-device sysfs attribute "current_timestamp_clock" to allow
userspace to select a particular POSIX clock for buffered samples and
events timestamping.
Following clocks, as listed in clock_gettime(2), are supported:
CLOCK_REALTIME, CLOCK_MONOTONIC, CLOCK_MONOTONIC_RAW,
CLOCK_REALTIME_COARSE, CLOCK_MONOTONIC_COARSE, CLOCK_BOOTTIME and
CLOCK_TAI.
Signed-off-by: Gregor Boirie <gregor.boirie@parrot.com>
Acked-by: Sanchayan Maity <maitysanchayan@gmail.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
The scale would result in uV instead of expected mV.
Mostly cosmetic, since the value of 'Power' was computed OK.
Signed-off-by: Marc Titinger <marc.titinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
The user (or an init script) may setup RShunt via sysfs after the
driver was initialized, for instance based on the EEPROM contents
of a modular probe. The calibration register must be set accordingly.
Signed-off-by: Marc Titinger <marc.titinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
These are generally for devlopment use only, remove these
from performance-critical code, convert to dev_dbg elswhere.
Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
Group of probably overly rigorous whitespace and code cleanups.
- Alphabetize includes
- Assign to variables in the order they are defined
- Alignment issues
- Group alike statements together
- Use helper macros
Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
This was extracted from a reposting of the driver after it had been applied
to the IIO tree. I have fast tracked it as the driver will be in 4.5 and
it would be nice to fix this trivial issue before it is.
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
POWER and CURRENT were swapped out in the buffer:
was current2 and power3, correct order is power2 and current3.
Signed-off-by: Marc Titinger <mtitinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
PID PPID USER STAT VSZ %VSZ %CPU COMMAND
144 2 root DW 0 0% 33% [ina226:1-8800us]
141 2 root DW 0 0% 25% [ina226:0-8800us]
40 2 root SW 0 0% 15% [irq/156-4802a00]
147 2 root DW 0 0% 7% [ina226:2-8800us]
145 1 root S 1236 0% 6% dd if /dev/iio:device1 of /dev/null
148 1 root S 1236 0% 4% dd if /dev/iio:device2 of /dev/null
149 137 root R 1244 0% 3% top -d 1
142 1 root S 1236 0% 2% dd if /dev/iio:device0 of /dev/null
Signed-off-by: Marc Titinger <mtitinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
Different probe modules use different resistor values. The front-end
application may read a probe ID (from eeprom) and set the shunt value
accordingly.
Signed-off-by: Marc Titinger <mtitinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
This can lead to repeated or skipped samples depending on the clock beat
between the capture thread and the chip sampling clock, but will also spare
reading/waiting for the Capture Ready Flag and improve the available i2c
bandwidth for reading measurements.
Output of iio_info:
...snip...
4 device-specific attributes found:
attr 0: in_oversampling_ratio value: 4
attr 1: in_allow_async_readout value: 0
attr 2: integration_time_available value: 140 204 332 588 1100 2116...
attr 3: in_sampling_frequency value: 114
Signed-off-by: Marc Titinger <mtitinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
|
in SOFTWARE buffer mode, a kthread will capture the active scan_elements
into a kfifo, then compute the remaining time until the next capture tick
and do an active wait (udelay).
This will produce a stream of up to fours channels plus a 64bits
timestamps (ns).
Tested with ina226, on BeagleBoneBlack.
Datasheet: http://www.ti.com/lit/gpn/ina226
Signed-off-by: Marc Titinger <mtitinger@baylibre.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
|
Jonathan Cameron
Akinobu Mita
Stefan Brüns
Maciej Purski
Greg Kroah-Hartman
Jonathan Cameron
Javier Martinez Canillas
Alison Schofield
Matt Ranostay
Gregor Boirie
Marc Titinger
Andrew F. Davis