- 10 Apr, 2022 10 commits
-
-
Miquel Raynal authored
In order to later move this variable within the opaque structure, let's create a helper for accessing it in read-only mode. This helper will be exposed to device drivers and kept accessible for the few that could need it. The write access to this variable however should be fully reserved to the core so in a second step we will hide this variable into the opaque structure. Cc: Eugen Hristev <eugen.hristev@microchip.com> Cc: Nicolas Ferre <nicolas.ferre@microchip.com> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com> Cc: Ludovic Desroches <ludovic.desroches@microchip.com> Signed-off-by:
Miquel Raynal <miquel.raynal@bootlin.com> Link: https://lore.kernel.org/r/20220207143840.707510-11-miquel.raynal@bootlin.comSigned-off-by:
Jonathan Cameron <Jonathan.Cameron@huawei.com>
-
Miquel Raynal authored
As we are going to hide the currentmode inside the opaque structure, this helper would soon need to call a non-inline function which would simply drop the benefit of having the helper defined inline in a header. One alternative is to move this helper in the core as there is no more interest in defining it inline in a header. We will pay the minor cost either way. Let's do like the iio_device_id() helper which also refers to the opaque structure and gets defined in the core. Suggested-by:
Jonathan Cameron <jic23@kernel.org> Signed-off-by:
-
Miquel Raynal authored
The st_sensors_core driver hardcodes the content of the iio_device_claim_direct_mode() and iio_device_release_direct_mode() helpers. Let's get rid of this handcrafted implementation and use the proper core helpers instead. Additionally, this lowers the tab level (which is always good) and prevents the use of the ->currentmode variable which is not supposed to be used like this anyway. Cc: Denis Ciocca <denis.ciocca@st.com> Signed-off-by:
-
Miquel Raynal authored
An odr_lock has been introduced to protect local accesses to the odr internal cache and ensure the cached value always reflected the actual value. Using the mlock() for this purpose is no longer needed, so let's drop these extra mutex_lock/unlock() calls. Suggested-by:
-
Miquel Raynal authored
Right now the (framework) mlock lock is (ab)used for multiple purposes: 1- protecting concurrent accesses over the odr local cache 2- avoid changing samplig frequency whilst buffer is running Let's start by handling situation #1 with a local lock. Suggested-by:
-
Miquel Raynal authored
The use of a lock there seems pointless. Besides preventing to read these information from userspace while buffers are enabled (which is not supposed to happen), it only protect read accesses over static const values, which are never supposed to be written anyway. Drop these lock calls. Suggested-by:
-
Miquel Raynal authored
As there is no cleanup to do, let's return as early as possible in the various ST sensor drivers _write_raw() callback functions. There is no functional change. Suggested-by:
-
Miquel Raynal authored
This is an internal variable of the core, let's use the iio_buffer_enabled() helper which is exported for the following purpose: telling if the current mode is a buffered mode, which is precisely what this driver looks for. Cc: Olivier Moysan <olivier.moysan@foss.st.com> Cc: Fabrice Gasnier <fabrice.gasnier@foss.st.com> Signed-off-by:
Fabrice Gasnier <fabrice.gasnier@foss.st.com> Link: https://lore.kernel.org/r/20220207143840.707510-4-miquel.raynal@bootlin.comSigned-off-by:
-
Miquel Raynal authored
This is an internal variable for the core, here it is set to a "default" value by the driver in order to later be able to perform checks against it. None of this is needed because this check actually cares about the buffers being enabled or not. So it is an unproper side-channel access to the information "are the buffers enabled?", returned officially by the iio_buffer_enabled() helper. Use this helper instead. Cc: Song Qiang <songqiang1304521@gmail.com> Signed-off-by:
-
Miquel Raynal authored
Let's provide more details about these two variables because their understanding may not be straightforward for someone not used to the IIO subsystem internal logic. The different modes will soon be also be more documented for the same reason. Signed-off-by:
-
- 04 Apr, 2022 30 commits
-
-
Marek Vasut authored
Replace sysfs attributes with read_avail() callback. This also permits removal of ads1115_info, since the scale attribute tables are now part of chip data. Signed-off-by:
Marek Vasut <marex@denx.de> Cc: Daniel Baluta <daniel.baluta@nxp.com> Reviewed-by:
Andy Shevchenko <andy@kernel.org> Link: https://lore.kernel.org/r/20220328194725.149150-10-marex@denx.deSigned-off-by:
-
Marek Vasut authored
Instead of storing only data_rate in private data, store pointer to the whole chip data and use the data_rate from chip data throughout the driver. No functional change. This is done in preparation for switching to read_avail(). Signed-off-by:
-
Marek Vasut authored
Replace chip type enumeration in match data with pointer to static constant structure which contains all the different chip properties in one place, and then replace handling of chip type in probe() with simple copy of fields in the new match data structure into struct iio_dev. This reduces code and increases static data. Signed-off-by:
-
Marek Vasut authored
Add compile-time static_assert wrapper to verify that shifted realbits fit into storagebits. The macro is implemented in a more generic way so it can be used to verify other values if required. Signed-off-by:
-
Marek Vasut authored
Add support for TI TLA2024 ADC. This chip is compatible with ADS1015 except it has no comparator in it, hence the comparator configuration bits are missing in Configuration Register and the Hi_Thresh/Lo_Thresh registers are missing as well and so is event support. Signed-off-by:
-
Marek Vasut authored
Pass event_spec and num_event_specs to ADS1015_V_CHAN and ADS1015_V_DIFF_CHAN macros, to make it possible to pass no event_spec at all for chips which have no comparator and thus no events. No functional change. Signed-off-by:
-
Marek Vasut authored
These macros differ only in the number of valid bits of each ADC sample and the shift of those bits, i.e. ADS1015 is 12bit ADC shifted by 4 left, ADS1115 is 16bit ADC shifted by 0. No functional change. Signed-off-by:
-
Marek Vasut authored
Switch the driver from code implementing test whether a regmap register is writeable to static const tables describing the test. No functional change. Signed-off-by:
-
Marek Vasut authored
Add ti,tla2024 compatible string. This device is compatible with ADS1015 except it has no on-chip comparator. Acked-by:
Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com> Signed-off-by:
-
Marek Vasut authored
Add missing ti,ads1115 compatible string. This compatible string is supported by the Linux kernel driver and the ads1015 is a 12bit ADC while ads1115 is 16bit ADC. Add the missing compatible string. Acked-by:
-
Marek Vasut authored
Add runtime check to verify whether storagebits are at least as big as shifted realbits. This should help spot broken drivers which may set realbits + shift above storagebits. Signed-off-by:
Nuno Sá <nuno.sa@analog.com> Reviewed-by:
-
Paul Lemmermann authored
Changed the leading spaces to tabs, in accordance with kernel coding conventions, and removed trailing comma. Found with checkpatch. Signed-off-by:
Paul Lemmermann <thepaulodoom@thepaulodoom.com> Link: https://lore.kernel.org/r/YkInN6SL7pP2f5Sf@hp-amd-paulSigned-off-by:
-
Paul Lemmermann authored
Cleaning up code. Found with checkpatch. Signed-off-by:
-
Jakob Koschel authored
To move the list iterator variable into the list_for_each_entry_*() macro in the future it should be avoided to use the list iterator variable after the loop body. To *never* use the list iterator variable after the loop it was concluded to use a separate iterator variable instead of a found boolean [1]. This removes the need to use a found variable and simply checking if the variable was set, can determine if the break/goto was hit. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1] Signed-off-by:
Jakob Koschel <jakobkoschel@gmail.com> Reviewed-by:
-
Jakob Koschel authored
To move the list iterator variable into the list_for_each_entry_*() macro in the future it should be avoided to use the list iterator variable after the loop body. To *never* use the list iterator variable after the loop it was concluded to use a separate iterator variable instead of a found boolean [1]. This removes the need to use a found variable and simply checking if the variable was set, can determine if the break/goto was hit. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1] Signed-off-by:
-
Jakob Koschel authored
In preparation to limit the scope of the list iterator variable to the list traversal loop, use a dedicated pointer to iterate through the list [1]. Since that variable should not be used past the loop iteration, a separate variable is used to 'remember the current location within the loop'. To either continue iterating from that position or start a new iteration (if the previous iteration was complete) list_prepare_entry() is used. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1] Signed-off-by:
-
H. Nikolaus Schaller authored
Although technically checking for ADC values below 0 is correct, because they are outside of the calibration values, there is usually noise which spuriously fills the console log with error messages if calculated input voltage gets close to 0V. Ignore small negative calculated values, but clamp them to 0. Signed-off-by:
H. Nikolaus Schaller <hns@goldelico.com> Link: https://lore.kernel.org/r/1cee45bfc3fa2ab59dcc17242fb52468035360a1.1646743982.git.hns@goldelico.comSigned-off-by:
-
Alexandru Tachici authored
Add sequencer support for AD7192. Signed-off-by:
Alexandru Tachici <alexandru.tachici@analog.com> Link: https://lore.kernel.org/r/20220322105029.86389-7-alexandru.tachici@analog.comSigned-off-by:
-
Alexandru Tachici authored
Add sequencer support for AD7124. Signed-off-by:
-
Lars-Peter Clausen authored
Some sigma-delta chips support sampling of multiple channels in continuous mode. When the operating with more than one channel enabled, the channel sequencer cycles through the enabled channels in sequential order, from first channel to the last one. If a channel is disabled, it is skipped by the sequencer. If more than one channel is used in continuous mode, instruct the device to append the status to the SPI transfer (1 extra byte) every time we receive a sample. All sigma-delta chips possessing a sampling sequencer have this ability. Inside the status register there will be the number of the converted channel. In this way, even if the CPU won't keep up with the sampling rate, it won't send to userspace wrong channel samples. When multiple channels are enabled in continuous mode, the device needs to perform a measurement on all slots before we can push to userspace the sample. If, during sequencing and data reading, a channel measurement is lost, a desync occurred. In this case, ad_sigma_delta drops the incomplete sample and waits for the device to send the measurement on the first active slot. Co-developed-by:
Lars-Peter Clausen <lars@metafoo.de> Link: https://lore.kernel.org/r/20220322105029.86389-5-alexandru.tachici@analog.comSigned-off-by:
-
Alexandru Tachici authored
The callback .set_channel cannot be used to enable multiple channels at once, only one is allowed simultaneously. By adding an update_scan_mode callback, every time the continuous mode is activated, channels will be enabled/disabled accordingly. Signed-off-by:
-
Alexandru Tachici authored
The callback .set_channel cannot be used to enable and disable multiple channels at once as they are presented in the active_scan_mask. By adding an update_scan_mode callback, every time the continuous mode is activated, channels will be enabled/disabled accordingly. Signed-off-by:
-
Alexandru Tachici authored
The 24 bits data is stored in 32 bits in BE. There is no need to shift it. This confuses user-space apps. Fixes: b3af341b ("iio: adc: Add ad7124 support") Signed-off-by:
-
Michael Srba authored
The difference between the ICM-20608-D and the other ICM-20608 variants is the addition of a DMP (Digital Motion Processor) core. This difference is deemed substantial enough to change the WHOAMI register value. Since this driver doesn't currently acknowledge the exisence of something like a DMP core, simply copy ICM-20608 except for the aforementioned WHOAMI register. Signed-off-by:
Michael Srba <Michael.Srba@seznam.cz> Acked-by:
Jean-Baptiste Maneyrol <jean-baptiste.maneyrol@tdk.com> Link: https://lore.kernel.org/r/20220323121550.16096-3-michael.srba@seznam.czSigned-off-by:
-
Michael Srba authored
ICM-20608-D differs from the other ICM-20608 variants by having a DMP (Digital Motion Processor) core tacked on. Despite having a different WHOAMI register, this variant is completely interchangeable with the other ICM-20608 variants by simply pretending the DMP core doesn't exist. Signed-off-by:
Rob Herring <robh@kernel.org> Reviewed-by:
Krzysztof Kozlowski <krzk@kernel.org> Link: https://lore.kernel.org/r/20220323121550.16096-2-michael.srba@seznam.czSigned-off-by:
-
Lorenzo Bianconi authored
Using my kernel.org email for iio devices. Signed-off-by:
Lorenzo Bianconi <lorenzo@kernel.org> Link: https://lore.kernel.org/r/c75f0e1c683b6ad0333d5504fb00c6f1b14dd1bf.1648114531.git.lorenzo@kernel.orgSigned-off-by:
-
Andy Shevchenko authored
Convert the module to be property provider agnostic and allow it to be used on non-OF platforms. The conversion slightly changes the logic behind property reading for the configuration values. Original code allocates just as much memory as needed. Then for each separate 32- or 64-bit value it reads it from the property and converts to a raw one which will be fed to the sensor. In the new code we allocate the amount of memory needed to retrieve all values at once from the property and then convert them as required. Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by:
-
Andy Shevchenko authored
There are several, possibly copied'n'pasted, places in the functions, where OF node is put and error returned directly, while the common exit point exists. Unify all these cases to use a single error path. Suggested-by:
-
Andy Shevchenko authored
In a couple of messages the constants, which have their definitions, are hard coded into the message text. Unhardcode them. While at it, add a trailing \n where it's currently missing. Signed-off-by:
-
Xiaoke Wang authored
kstrdup() is also a memory allocation-related function, it returns NULL when some memory errors happen. So it is better to check the return value of it so to catch the memory error in time. Besides, there should have a kfree() to clear up the allocation if we get a failure later in this function to prevent memory leak. Signed-off-by:
Xiaoke Wang <xkernel.wang@foxmail.com> Link: https://lore.kernel.org/r/tencent_C920CFCC33B9CC1C63141FE1334A39FF8508@qq.comSigned-off-by:
-
