| From: |
| Nuno Sá <nuno.sa-AT-analog.com> |
| To: |
| <linux-iio-AT-vger.kernel.org>, <devicetree-AT-vger.kernel.org> |
| Subject: |
| [PATCH v4 0/3] Add support for LTC2688 |
| Date: |
| Fri, 25 Feb 2022 14:01:26 +0100 |
| Message-ID: |
| <20220225130129.69-1-nuno.sa@analog.com> |
| Cc: |
| Rob Herring <robh+dt-AT-kernel.org>, Jonathan Cameron <jic23-AT-kernel.org>, Lars-Peter Clausen <lars-AT-metafoo.de>, Michael Hennerich <Michael.Hennerich-AT-analog.com> |
| Archive-link: |
| Article |
The ABI defined for this driver has some subtleties that were previously
discussed in this RFC [1]. This might not be the final state but,
hopefully, we are close to it:
toggle mode channels:
* out_voltageY_toggle_en
* out_voltageY_raw0
* out_voltageY_raw1
* out_voltageY_symbol
dither mode channels:
* out_voltageY_dither_en
* out_voltageY_dither_raw
* out_voltageY_dither_raw_available
* out_voltageY_dither_offset
* out_voltageY_dither_frequency
* out_voltageY_dither_frequency_available
* out_voltageY_dither_phase
* out_voltageY_dither_phase_available
Default channels won't have any of the above ABIs. A channel is toggle
capable if the devicetree 'adi,toggle-mode' flag is set. For dither, the
assumption is more silent. If 'adi,toggle-mode' is not given and a
channel is associated with a TGPx pin through 'adi,toggle-dither-input',
then the channel is assumed to be dither capable (there's no point in
having a dither capable channel without an input clock).
changes in v2:
ltc2688:
* Use local buffer for regmap read. Do not assume that reg is part of
larger buffer;
* Renamed GPIO to "clr" so that is consistent with the datasheet;
* Renamed 'mask' and 'm' to info. 'mask' is a thing from the past;
* Removed 'LTC2688_CHAN_TOGGLE()' and defined to static ext_info arrays;
* Use 'regmap_set_bits' to set external ref;
* Use FIELD_{PREP|GET} for dither amplitude and channel calibbias where
only 13bits are used;
* Use 'regmap_write()' instead of update_bits for channels settings;
* Init 'val' at the beginning of the channel configuration loop
(and drop mask);
* Comment 'ltc2688_reg_writable()' to account for the special condition;
* Kmemdup default channels so that it can be safely changed per probed
device;
* Replace extended info multiplexer functions by individual functions;
* Use raw0 ABI for toggle channels;
* Use dedicated offset ABI for dither channels;
* Misc changes (spell fixes, blank lines...);
* Have a clock property per channel. Note that we this I moved to OF
since we now have to use 'devm_get_clk_from_child()' which is using
device_node. Note that I could use 'to_of_node()' but mixing of.h and
property.h does not feel like a good idea.
ABI:
* Added out_voltageY_raw0 ABI for toggle mode;
* Added out_voltageY_dither_offset.
Bindings:
* Use standard microvolt unit;
* Added constrains for adi,output-range-microvolt and removed negative
values from the dts example;
* Moved clocks to the channel object;
* Dropped clock-names;
* Add a dependency between 'adi,toggle-dither-input' and 'clocks'.
Changes in v3:
ltc2688:
* Fix mismatch between functions and function pointers detected by kernel
test bot;
* Always use if (ret) when ret > 0 has no meaning;
* Rename ltc2688_bulk_disable -> ltc2688_disable_regulators;
* Report dither phase in radians rather than degrees.
ABI:
* Specify units for dither_phase and dither_freqency;
* Say why its useful to have dither_en and toggle_en;
* Combine out_voltageY_raw0 and out_voltageY_raw1;
* Fix some description issues in out_voltageY_raw{0|1} and
out_voltageY_symbol.
Bindings:
* Remove mentions to ABI (linux specifix);
* Slightly rephrased VREF and adi,toggle-dither-input properties and
suggested.
changes in v4:
ltc2688:
* Use reg_size + val_size instead of plain 3 in regmap;
* Use out_unlock instead of unlock in goto labels;
* Add comma to LTC2688_CHANNEL(), ltc2688_regmap_bus and
ltc2688_regmap_bus;
* Use __clear_bit() instead of clear_bit();
* Flip the logic in vref regulator so that error condition is handled
first;
* Change to device API. With this, we need to_of_node()
for devm_get_clk_from_child().
ABI:
* Update kernel version.
Bindings:
* Add Rob's Rb tag.
[1]: https://marc.info/?l=linux-iio&m=163662843603265&w=2
Nuno Sá (3):
iio: dac: add support for ltc2688
iio: ABI: add ABI file for the LTC2688 DAC
dt-bindings: iio: Add ltc2688 documentation
.../ABI/testing/sysfs-bus-iio-dac-ltc2688 | 86 ++
.../bindings/iio/dac/adi,ltc2688.yaml | 146 +++
MAINTAINERS | 9 +
drivers/iio/dac/Kconfig | 11 +
drivers/iio/dac/Makefile | 1 +
drivers/iio/dac/ltc2688.c | 1071 +++++++++++++++++
6 files changed, 1324 insertions(+)
create mode 100644 Documentation/ABI/testing/sysfs-bus-iio-dac-ltc2688
create mode 100644 Documentation/devicetree/bindings/iio/dac/adi,ltc2688.yaml
create mode 100644 drivers/iio/dac/ltc2688.c
--
2.34.1