From: Lothar Rubusch <l.rubusch@gmail.com>
To: lars@metafoo.de, Michael.Hennerich@analog.com, jic23@kernel.org,
dlechner@baylibre.com, nuno.sa@analog.com, andy@kernel.org,
corbet@lwn.net
Cc: linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-doc@vger.kernel.org, eraretuya@gmail.com,
l.rubusch@gmail.com
Subject: [PATCH v9 00/11] iio: accel: adxl345: add interrupt based sensor events
Date: Tue, 10 Jun 2025 21:59:22 +0000 [thread overview]
Message-ID: <20250610215933.84795-1-l.rubusch@gmail.com> (raw)
Add several interrupt based sensor detection events:
- refactoring and fixes
- activity
- inactivity
- documentation
In brief, this implementation uses the inactivity registers for inactivity
events with a period >1s. For a configured period <1s the free-fall
registers will be used. The related features, such as setting the link-bit,
auto-sleep, the AC-coupling events, etc. are turned on/off accordingly. For
instance, activity and inactivity enabled imply a link-bit set. Disabling
one, removes the link-bit. Setting e.g. activity AC with inactivity DC, sets
the link bit and so on.
Due to several items found for the ADXL313, I tried to apply those refacs
here as well. I'm still unsure, if I should rathe wrap them up in one big
refac patch or keep them separated by the items as I tried here.
By the number of possible combinations of configurations and their
verification, as additionally the refactorings - I implemented an automated
verification by a labgrid/pytest setup with over 300 test steps and hardware
in the loop. Hopefully the setup is mostly correct. Further I extended the
codespell githook by an ispell check (which gave better results).
Let's see how good that works out. Somehow I'm sure you guys will (still)
find something...
Signed-off-by: Lothar Rubusch <l.rubusch@gmail.com>
---
v8 -> v9:
- githook: apply codespell checks, extend by spellcheck using ispell
- refac: apply `regmap_assing_bits()` in several places
- refac: remove ADXL345_POWER_CTL_STANDBY in adxl345.h not needed anymore
- refac: rename `ADXL345_ACT_INACT_DC/AC` to `ADXL345_COUPLING_DC/AC`
- refac: remove variable irq from `struct adxl345_state`
- refac: apply expressions, such as MILLI or MICRO from linux/units.h
- refac: apply (missing) scaling factor 62.5mg/LSB to tap detection
- change `IIO_EV_TYPE_MAG_REFERENCED` to `IIO_EV_TYPE_MAG_ADAPTIVE`
- `adxl345_fifo_transfer()`: eliminate variable for expression count / 2
- `adxl345_is_act_inact_ac()`: make return boolean, or negative error
- make activity enable cover x, y and z axis together, while signals come
on particular axis
- `adxl345_read_mag_value()`: separate `MAG` and `MAG_ADAPTIVE` event
value read and write function to reduce redundant code
- `adxl345_read_mag_config()`: separate `MAG` and `MAG_ADAPTIVE` event
config read and write functions to reduce redundant code
- `adxl345_set_act_inact_en()`: move linkbit detection out to
`adxl345_set_act_inact_linkbit()`
- `adxl345_set_act_inact_en()`: fix unsetting register INT ENABLE at
disabling feature(s)
- apply scaling factor 62.5mg/LSB to activity/inactivity
- apply scaling factor 62.5mg/LSB to activity AC/inactivity AC
- drop dedicated freefall patch
- add patch: fix missing scale factor for tap detection
- add patch: make irq a function local variable
- add patch: simplify measure enable
- add patch: simplify interrupt mapping
- add patch: simplify FIFO reading
- add patch: replace magic numbers by unit expressions
- rename `adxl345_set_inact_time_s()` to `adxl345_set_inact_time()`
- `adxl345_set_inact_time()`: implement inactivity using inactivity
register (period 1s or higher), freefall register (below 1s) or let it
setup a period based on given ODR (0s provided)
- doc: update documentation
v7 -> v8:
- activity/inactivity are MAG events
- separate AC coupled activity/inactivity events as MAG_REFERENCED events,
since AC coupling introduces a (some kind of) reference relation
- since freefall and inactivity (DC coupled) are then actually identical,
this results in a challenging situation for the freefall patch. Thus,
the freefall patch is moved to end of this series (before documentation)
- freefall: provide separate sysfs handles to configure and enable freefall
- documentation: update sections on activity/inactivity, freefall, event
names and examples
v6 -> v7:
- freefall: add a virtual channel, replace OR'ing the axis by AND'ing them
- inactivity: add a virtual channel, replace OR'ing the axis by AND'ing them
v5 -> v6:
- replace bool axis_en for tap and activity/inactivity
- apply freefall bit mask
- change `measure_en` to use `regmap_update_bits()` for POWER_CTL register
- fix comments and update documentation, particularly on inactivity time
v4 -> v5:
- read_config_value() and write_config_value() now use direct returns,
in case of a failure, measurement stays turned off
- fifo evaluation returns 0 in case of success
- axis enum turned into three different set of defines for tap, act and inact
- turn the suppress bit into a separate define macro
- variable naming, generally use axis_ctrl for similar variables
v3 -> v4:
- rename patch "add double tap suppress bit" to
"set the tap suppress bit permanently" to make it more comprehensive
- added patch "cleanup regmap return values"
- added patch "introduce adxl345_push_event function", as a solution
to the return value problem, group all int_stat evaluating pieces
in the same function
- tap, act and inact axis enabling are using now regmap cache
- activity enable depending on individual axis now, as the sensor offers
such feature
- inactivity enable depending on individual axis now, as the sensor offers
such feature
- fix bug in previous patch: separate axis direction in interrupt handler
sharing the same variable for tap and activity, if tap and activity
enabled together
- refac of the direction identification of previous patch: only read
act/tap axis once now in interrupt handler if both is enabled
- fix bug in previous patch: return value of pushed event in interrupt
handler
- several cleanups
v2 -> v3:
- generally introduction of regmap cache for all directly stored 8-bit
values, specification of volatile regs and cleanup
- moving thresholds, unchanged values and flags to regmap cache, in
consequence removal of corresponding member values of the state
instance
- removal of intio and int_map member fields due to regmap cache, thus
split of set_interrupts() patches in two parts
- rework documentation
- rework of ac-bit comment
v1 -> v2:
- implementation of all events (but tap2 suppress bit) by means IIO ABI
- add sample frequency / ODR configuration
- add g ranges configuration
- add activity/inactivity using auto-sleep and powersave
- add dynamic adjustment of default values for
activity/inactivity thresholds and time for inactivity based on ODR
and g range (can be overwritten)
- add sensor documentation
---
Lothar Rubusch (11):
iio: accel: adxl345: apply scale factor to tap threshold
iio: accel: adxl345: make data struct variable irq function local
iio: accel: adxl345: simplify measure enable
iio: accel: adxl345: simplify interrupt mapping
iio: accel: adxl345: simplify reading the FIFO
iio: accel: adxl345: replace magic numbers by unit expressions
iio: accel: adxl345: add activity event feature
iio: accel: adxl345: add inactivity feature
iio: accel: adxl345: add coupling detection for activity/inactivity
iio: accel: adxl345: extend inactivity time for less than 1s
docs: iio: add documentation for adxl345 driver
Documentation/iio/adxl345.rst | 429 +++++++++++++++++
Documentation/iio/index.rst | 1 +
drivers/iio/accel/adxl345.h | 1 -
drivers/iio/accel/adxl345_core.c | 793 +++++++++++++++++++++++++++++--
4 files changed, 1195 insertions(+), 29 deletions(-)
create mode 100644 Documentation/iio/adxl345.rst
--
2.39.5
next reply other threads:[~2025-06-10 21:59 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-10 21:59 Lothar Rubusch [this message]
2025-06-10 21:59 ` [PATCH v9 01/11] iio: accel: adxl345: apply scale factor to tap threshold Lothar Rubusch
2025-06-14 13:42 ` Jonathan Cameron
2025-06-15 22:20 ` Lothar Rubusch
2025-06-21 16:55 ` Jonathan Cameron
2025-06-21 18:14 ` Lothar Rubusch
2025-06-10 21:59 ` [PATCH v9 02/11] iio: accel: adxl345: make data struct variable irq function local Lothar Rubusch
2025-06-14 13:43 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 03/11] iio: accel: adxl345: simplify measure enable Lothar Rubusch
2025-06-14 13:45 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 04/11] iio: accel: adxl345: simplify interrupt mapping Lothar Rubusch
2025-06-11 15:20 ` Andy Shevchenko
2025-06-10 21:59 ` [PATCH v9 05/11] iio: accel: adxl345: simplify reading the FIFO Lothar Rubusch
2025-06-14 13:47 ` Jonathan Cameron
2025-06-15 22:14 ` Lothar Rubusch
2025-06-21 16:58 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 06/11] iio: accel: adxl345: replace magic numbers by unit expressions Lothar Rubusch
2025-06-14 13:49 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 07/11] iio: accel: adxl345: add activity event feature Lothar Rubusch
2025-06-12 11:51 ` Andy Shevchenko
2025-06-21 18:06 ` Lothar Rubusch
2025-06-21 19:55 ` Andy Shevchenko
2025-06-10 21:59 ` [PATCH v9 08/11] iio: accel: adxl345: add inactivity feature Lothar Rubusch
2025-06-12 12:15 ` Andy Shevchenko
2025-06-14 13:55 ` Jonathan Cameron
2025-06-14 19:02 ` Andy Shevchenko
2025-06-21 18:53 ` Lothar Rubusch
2025-06-21 19:24 ` Andy Shevchenko
2025-06-21 19:28 ` Andy Shevchenko
2025-06-14 14:00 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 09/11] iio: accel: adxl345: add coupling detection for activity/inactivity Lothar Rubusch
2025-06-14 14:04 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 10/11] iio: accel: adxl345: extend inactivity time for less than 1s Lothar Rubusch
2025-06-14 14:10 ` Jonathan Cameron
2025-06-10 21:59 ` [PATCH v9 11/11] docs: iio: add documentation for adxl345 driver Lothar Rubusch
2025-06-14 14:17 ` Jonathan Cameron
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250610215933.84795-1-l.rubusch@gmail.com \
--to=l.rubusch@gmail.com \
--cc=Michael.Hennerich@analog.com \
--cc=andy@kernel.org \
--cc=corbet@lwn.net \
--cc=dlechner@baylibre.com \
--cc=eraretuya@gmail.com \
--cc=jic23@kernel.org \
--cc=lars@metafoo.de \
--cc=linux-doc@vger.kernel.org \
--cc=linux-iio@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=nuno.sa@analog.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).