Age | Commit message (Collapse) | Author |
|
git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next
Jonathan writes:
Second set of new device support, features and minor fixes for IIO in the 5.6 cycle
Just a small set this time.
As we are very near the merge window, I've rolled a few fixes in here
rather than adding noise just before release. A short delay here will
do little harm.
New device support
* adis16480
- Add support for adis16490. After earlier rework this is simple ID plus
chip info.
Features
* kxcjk1013
- mount matrix support.
* lsm_6dsx
- mount matrix support.
Cleanups / minor or late breaking fixes
* ad7124
- add support to ad-sigma-delta and use it in this driver to allow
the the interrupt type to be IRQF_TRIGGER_LOW unlike most other devices
using this framework.
* adis
- use delay structure now available in SPI to handle transfer delays
- introduce a timeouts structure to allow support of new devices
* ak8975
- drop platform data support. No one is using it and it adds complexity.
- use device_get_match_data rather than open coding much the same thing.
* dht11
- drop meaningless todo
* at91-samad2_adc
- switch to dma_request_chan
* altas-sensor
- add a helper function to compute number of channels. Needed for new device
support that is under review.
* bma400
- add a lower bound check on scale.
* inv_mpu6050
- add support for temperature data in the fifos for all chips.
- support an odd situation where a board supports only interrupt triggering
on both edges.
* st_lsm6dsx
- check and handle potential error return.
* st_sensors
- fix some values for the LSM9DS0 which is ever so slightly different from
other devices using the same whoami value.
- switch over to generic functions from dt ones, avoiding need for separate
ACPI support.
* stm32-adc
- switch to dma_request_chan
- suppress an error print in deferred probe case.
* stm32-dac
- drop private data structure element for reset controller as only used in
probe.
- reflect more cleanly that the reset controller is optional whilst ensuring
that if is specified any errors are caught.
* stm32-dfsdm
- switch to dma_request_chan
- fix missing application of formatting to single conversions.
- ensure the sampling rate is updated when the oversampling ratio is changed.
* tag 'iio-for-5.6b' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (29 commits)
iio: dac: stm32-dac: better handle reset controller failures
iio: dac: stm32-dac: use reset controller only at probe time
dt-bindings: iio: accel: kxcjk1013: Document mount-matrix property
iio: accel: kxcjk1013: Support orientation matrix
iio: imu: st_lsm6dsx: add mount matrix support
iio: adc: stm32-adc: don't print an error on probe deferral
dt-bindings: iio: adis16480: add compatible entry for ADIS16490
iio: imu: adis16480: Add support for ADIS16490
iio: accel: bma400: prevent setting accel scale too low
iio: imu/mpu6050: support dual-edge IRQ
iio: imu: inv_mpu6050: add fifo temperature data support
iio: magnetometer: ak8975: Convert to use device_get_match_data()
iio: magnetometer: ak8975: Get rid of platform data
iio: adc: ad7124: Set IRQ type to falling
iio: adc: ad-sigma-delta: Allow custom IRQ flags
iio: imu: adis: use new `delay` structure for SPI transfer delays
iio: adc: stm32-dfsdm: adapt sampling rate to oversampling ratio
iio: adc: stm32-dfsdm: fix single conversion
iio: st_sensors: Make use of device properties
iio: st_sensors: Drop redundant parameter from st_sensors_of_name_probe()
...
|
|
Use devm_reset_control_get_optional_exclusive() instead of
devm_reset_control_get_exclusive() as reset controller is optional.
Nevertheless if reset controller is expected but reports an
error, propagate the error code to the caller. In such case
a nice error trace is emitted unless we're deferring the probe
operation.
Signed-off-by: Etienne Carriere <etienne.carriere@st.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
This change removes the reset controller reference from the local
DAC instance since it is used only at probe time.
Signed-off-by: Etienne Carriere <etienne.carriere@st.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
The generic IIO mount-matrix property conveys physical orientation of the
hardware chip.
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Hardware could be physically mounted in any possible direction and
userpspace needs to be aware of the mounting orientation in order to
process sensor's data correctly. In particular this helps iio-sensor-proxy
to report display's orientation properly on a phone/tablet devices.
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Allow to read the mount-matrix device tree property and provide the
mount_matrix file for userspace to read.
Signed-off-by: Martin Kepplinger <martin.kepplinger@puri.sm>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Do not print an error trace when deferring probe for some resource.
Signed-off-by: Etienne Carriere <etienne.carriere@st.com>
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
This change adds an entry for ADIS16490 in the list of compatible devices
defined in the dt-bindings of the adis16480 driver.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
The ADIS16490 is part of the same family with ADIS16495 and ADIS16497,
the main difference is the temperature, accelerometer and gyroscope
scales.
Datasheet:
Link: https://www.analog.com/media/en/technical-documentation/data-sheets/adis16490.pdf
Signed-off-by: Stefan Popa <stefan.popa@analog.com>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
This puts an upper bound on "val2" but it also needs to have a lower
bound (BMA400_SCALE_MIN).
Fixes: 465c811f1f20 ("iio: accel: Add driver for the BMA400")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Make mpu6050 usable on platforms which provide only any-edge interrupts.
One example of this kind of platform is AT91SAM9G45
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Acked-by: Jean-Baptiste Maneyrol <jmaneyrol@invensense.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Add support of temperature data in fifo for all chips.
Enable unification of scan elements for icm20602.
Add macros for generating scan elements.
Signed-off-by: Jean-Baptiste Maneyrol <jmaneyrol@invensense.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Convert to use device_get_match_data() instead of open coded variant.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Since IIO framework supports device property API and driver has been moved
already to the use of GPIO descriptors the logical continuation is to
get rid of platform data completely. We are on the safe side here since
there are no users of it in the kernel.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Ad7124 data-sheet specifies that the falling edge
of the DOUT line should be used for an interrupt.
The current irq flag (IRQF_TRIGGER_LOW) used will
cause unwanted behaviour. When enabling the interrupt
it will fire once because the DOUT line is already low.
This will make the driver to read an unfinished conversion
from the chip.
This patch sets the irq type to the one specified in
the data-sheet.
Signed-off-by: Alexandru Tachici <alexandru.tachici@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Before this patch the ad_sigma_delta implementation hardcoded
the irq trigger type to low, assuming that all Sigma-Delta ADCs
have the same interrupt-type.
This patch allows all drivers using the ad_sigma_delta layer to set the
irq trigger type to the one specified in the datasheet.
Signed-off-by: Alexandru Tachici <alexandru.tachici@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
|
|
Some work has been done :)
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-66-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Usage of the "buf_loc" variable does not simplify the function.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-65-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_tx_queue_get_num_queued() can take advantage of BIT() instead of
maintaining one variable for a counter and another for a mask.
In add, wfx_tx_queue_get_num_queued() has no real reason to return a
size_t instead of an int.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-64-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Since enum action has now only two cases, it can be dropped. Then
hif_handle_tx_data() can be simplified.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-63-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Obviously, the value of wvif->state as no reason to be wrong. At least,
if it the case, dropping the frame is probably not the bast thing to do.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-62-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The identifier do_probe is unused since "staging: wfx: remove workaround
to send probe requests"
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-61-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
In current code, packet_id is derivated from mac802.11 packet sequence
number, but this number is only unique for a station. It is not
sufficient.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-60-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_ps_notify_sta() is too complex compared to the task it do.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-59-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
When mac80211 ask for a frame to be sent after a DTIM, driver should:
1. Update TIM with multicast bit set (using update_ie). This function
can be called whenever.
2. Keep buffered all frames marked "after dtim"
3. When it receive a suspend_resume indication (see
wfx_suspend_resume_mc()), send all the buffered frames. This
indication is sent by the firmware 4ms before the dtim.
4. If one of the frames returns status "REQUEUE", it means that the
DTIM period was ended before to be able to send the frame.
5. When all the buffered frames were sent or if DTIM period was ended,
driver should update the TIM with multicast bit reset.
All the mess with the asynchronous works can be dropped.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-58-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Indeed, it is not necessary to pass whole hif_ind_suspend_resume_tx to
wfx_suspend_resume_mc().
In add, the structure hif_ind_suspend_resume_tx come from hardware API.
It is not intended to be manipulated in upper layers of the driver.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-57-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
At the beginning, firmware could send suspend_resume indication to
notify that a station wake up or sleep down. However, mac80211 already
handles power save status of stations and this behavior has been removed
from the firmware. So now, when suspend_resume indication is received,
it is always to notify that a DTIM is about to be sent.
So, it is possible to simply wfx_suspend_resume().
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-56-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_ps_notify() is called for any changes in the TIM. However,
association ID 0 is a very special case that should be handled
independently.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-55-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
pspoll_mask is here to send data buffered in driver. But since station
is marked buffered, TIM for this station is 1 and mac80211 will call
sta_notify when a ps-poll is received. So pspoll_mask is useless and
sta_alseep_mask is sufficient.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-54-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_tx_get_tid() was used as a wrapper around ieee80211_get_tid(). It
did sometime return WFX_MAX_TID to ask to upper layers to not include
the frame in "buffered" counter. The objective of this behavior is not
clear, but tests has shown that wfx_tx_get_tid() can be replaced by
ieee80211_get_tid() without any regressions.
BTW, it is not necessary to save the tid in tx_rpiv since it can be
retrieved from the 802.11 header.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-53-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Just for sanity.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-52-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The "link-id" is a slot number provided to the chip. A link-id is
allocated to every station associated with the chip (mainly when the
chip is in AP mode). It is more or less the same thing than the
association ID, but it is limited to 14 values.
Firmware uses the link-id to track the power save status of the
stations.
The current code try to associate a link-id as soon as data are
exchanged with station. It is far easier to rely on sta_add() and
sta_remove().
Until now the value WFX_LINK_ID_NO_ASSOC, was only used when no more
link-id was available. Now, we also use this value for
not-yet-associated stations (that was its primary behavior).
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-51-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Argument provided to wfx_set_tim_impl() is always wvif->aid0_bit_set and
there is no reason to provide another argument.
Also rename wfx_set_tim_impl() into wfx_update_tim() to reflect the new
behavior.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-50-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Indeed, sta was used after call to rcu_unlock()
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-49-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_tx_confirm_cb() retrieves the station associated with a frame using
the MAC address from the 802.11 header. In the other side wfx_tx()
retrieves the station using sta field from the ieee80211_tx_control
argument.
In wfx_cqm_bssloss_sm(), wfx_tx() was called directly without valid sta
field, but with a valid MAC address in 802.11 header. So there the
processing of this packet was unbalanced and may produce weird bugs.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-48-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
It simplify the code if field buffered is hosted in the struct sta_priv
instead of in the struct wfx_link_entry. More globally, struct
wfx_link_entry has no real reasons to exist and should be dropped soon.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-47-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
It seems that purpose of "early_data" was to prevent sending data to
mac80211 before station was completely associated. It is a useless
precaution.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-46-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
It is recommended to use function time_*() to compare jiffies.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-45-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
If RTS threshold is 0, it currently disables RTS. It should mean
"enabled for every frames".
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-44-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_conf_tx() is called for each queue. On every call, the function
updates UAPSD mask and PM mode for all queues. It is a pity since the
UAPSD configuration very rarely changes and it makes exchanges between
the host and the chip more difficult to track.
This patch avoid to update UAPSD and Power Mode in most usual cases.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-43-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
When multiple vif are in use (the only supported configuration is one
station and one AP), the driver force power save flag on station.
This behavior allows the station to leave the station channel and make
its business on AP channel.
However, this has a big impact on station performances (especially since
only legacy PS is supported).
When both vifs use the same channel, it is not necessary to keep this
restriction. This greatly improve station performances.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-42-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
mac80211 and the device are both able to control power-save per vif.
But, the current code retrieve power-save from wfx_config(). So, it does
not allow to setup power-save independently for each vif. Driver just
has to rely on wfx_bss_info_changed() instead of wfx_config().
wfx_config() has nothing to do anymore, but we keep it since it is
mandatory for mac80211.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-41-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_scan_complete() do nothing with argument hif_ind_scan_cmpl. In add,
hif_ind_scan_cmpl come from hardware API and is not expected to be used
with upper layers of the driver.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-40-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
wfx_update_filtering() has no reason to instantiate a struct
hif_mib_bcn_filter_enable. Drop it and simplify the code.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-39-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
HIF functions return only serious errors (OOM or device freeze). The
current handling of errors in wfx_set_mcast_filter() does not bring
anything. Finally it may disturb the developer more than it helps.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-38-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The structure hif_mib_config_data_filter come from hardware
API. It is not intended to be manipulated in upper layers of the driver.
In add, current code for hif_set_config_data_filter() is too dumb. It
should pack data with hardware representation instead of leaving all
work to the caller.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-37-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The structure hif_mib_mac_addr_data_frame_condition come from hardware
API. It is not intended to be manipulated in upper layers of the driver.
In add, current code for hif_set_mac_addr_condition() is too dumb. It
should pack data with hardware representation instead of leaving all
work to the caller.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-36-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The structure hif_mib_set_data_filtering come from hardware API. It is
not intended to be manipulated in upper layers of the driver.
In add, current code for hif_set_data_filtering() is too dumb. It should
pack data with hardware representation instead of leaving all work to
the caller.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-35-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The field "default_filter" was not obvious.
In add, explicitly declare that fields default_filter and enable are
booleans.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-34-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The current API defines bitfields. It is not very convenient. Prefer to
use bitmasks.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20200115135338.14374-33-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|