| Message ID | 20240726-dev-iio-masklength-private3-v1-0-82913fc0fb87@analog.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | iio: make masklength __private (final round) | expand |
On Fri, 26 Jul 2024 10:22:52 +0200 Nuno Sa via B4 Relay <devnull+nuno.sa.analog.com@kernel.org> wrote: > Hi Jonathan, > > Here it goes the final conversion round. On the testing branch a git > grep: > > git grep "\->masklength" > > Gives no more users so hopefully I did it right :). > > Not entirely related with this but I'll mention it anyways :). Looking > at the other [INTERN] fields, I see: > > * active_scan_mask: which seems to be pretty similar to masklength; > * scan_timestamp: Has some apparently valid users; > * scan_bytes: Also has a few users; > * trig: may need a bit more inspection. > > The first 3 look like valid RO fields that drivers are allowed to > access. I mean, for scan_bytes I guess drivers could calculate that > based on the active_scan_mask but I think that's just not worth it given > the core already did it for us. > > So I wonder if a similar work would be valuable? scan_timestamp and > scan_bytes would be pretty trivial but for active_scan_mask we do have > lots of users (but also trivial I think). > > For Trig, most of the writes are drivers doing > > indio_dev->trig = iio_trigger_get(data->trig); > > So we either assume it's public or remove that line from drivers. Though > that could break userland apps relying on some trigger to be > automatically set after device probe. We can anyways have another helper > to do that for drivers (not sure this field is accessed in some > fastpath). > > Also need to better check if there are any readers for it. > > Anyways, it would be nice to get your inputs on the last [INTERNAL] fields > in struct iio_dev > > Also want to mention that I deliberately removed some mailing lists > given by get_maintainers as this is fairly simple and so I don't think > the "spam" is worth it :). I applied these about 10 days ago but for some reason didn't say that! So these are applied if anyone was wondering. Jonathan > > --- > Nuno Sa (23): > iio: accel: adxl380: make use of iio_get_masklength() > iio: adc: max1118: make use of iio_for_each_active_channel() > iio: adc: max1118: make use of iio_for_each_active_channel() > iio: adc: mcp3911: make use of iio_for_each_active_channel() > iio: adc: mt6360-adc: make use of iio_for_each_active_channel() > iio: adc: rockchip_saradc: make use of iio_for_each_active_channel() > iio: adc: rtq6056: make use of iio_for_each_active_channel() > iio: adc: stm32-adc: make use of iio_get_masklength() > iio: adc: stm32-dfsdm-adc: make use of iio_get_masklength() > iio: adc: ti-adc0832: make use of iio_for_each_active_channel() > iio: adc: ti-adc084s021: make use of iio_for_each_active_channel() > iio: adc: ti-ads1015: make use of iio_get_masklength() > iio: adc: ti-ads1119: make use of iio_get_masklength() > iio: adc: ti-ads1298: make use of iio_for_each_active_channel() > iio: adc: ti-adc12138: make use of iio_for_each_active_channel() > iio: adc: ti-ads124s08: make use of iio_for_each_active_channel() > iio: adc: ti-ads131e08: make use of iio_for_each_active_channel() > iio: adc: ti-ads8688: make use of iio_for_each_active_channel() > iio: adc: vf610_adc: make use of iio_get_masklength() > iio: adc: xilinx-xadc-core: use new '.masklength' accessors > iio: common: cros_ec_sensors_core: use new '.masklength' accessors > staging: iio: impedance-analyzer: ad5933: make use of iio_get_masklength() > iio: core: annotate masklength as __private > > drivers/iio/accel/adxl380.c | 2 +- > drivers/iio/adc/ina2xx-adc.c | 3 +-- > drivers/iio/adc/max1118.c | 3 +-- > drivers/iio/adc/mcp3911.c | 2 +- > drivers/iio/adc/mt6360-adc.c | 2 +- > drivers/iio/adc/rockchip_saradc.c | 2 +- > drivers/iio/adc/rtq6056.c | 2 +- > drivers/iio/adc/stm32-adc.c | 4 ++-- > drivers/iio/adc/stm32-dfsdm-adc.c | 2 +- > drivers/iio/adc/ti-adc0832.c | 3 +-- > drivers/iio/adc/ti-adc084s021.c | 3 +-- > drivers/iio/adc/ti-adc12138.c | 3 +-- > drivers/iio/adc/ti-ads1015.c | 2 +- > drivers/iio/adc/ti-ads1119.c | 4 ++-- > drivers/iio/adc/ti-ads124s08.c | 3 +-- > drivers/iio/adc/ti-ads1298.c | 3 +-- > drivers/iio/adc/ti-ads131e08.c | 2 +- > drivers/iio/adc/ti-ads8688.c | 4 +--- > drivers/iio/adc/vf610_adc.c | 2 +- > drivers/iio/adc/xilinx-xadc-core.c | 5 ++--- > drivers/iio/common/cros_ec_sensors/cros_ec_sensors_core.c | 8 +++----- > drivers/iio/industrialio-buffer.c | 2 +- > drivers/staging/iio/impedance-analyzer/ad5933.c | 5 +++-- > include/linux/iio/iio.h | 4 ++-- > 24 files changed, 32 insertions(+), 43 deletions(-) > --- > base-commit: 472438c7e0e2261c6737a8321f46ef176eef1c8f > change-id: 20240724-dev-iio-masklength-private3-7a2d819acc7d > -- > > Thanks! > - Nuno Sá > >
Hi Jonathan, Here it goes the final conversion round. On the testing branch a git grep: git grep "\->masklength" Gives no more users so hopefully I did it right :). Not entirely related with this but I'll mention it anyways :). Looking at the other [INTERN] fields, I see: * active_scan_mask: which seems to be pretty similar to masklength; * scan_timestamp: Has some apparently valid users; * scan_bytes: Also has a few users; * trig: may need a bit more inspection. The first 3 look like valid RO fields that drivers are allowed to access. I mean, for scan_bytes I guess drivers could calculate that based on the active_scan_mask but I think that's just not worth it given the core already did it for us. So I wonder if a similar work would be valuable? scan_timestamp and scan_bytes would be pretty trivial but for active_scan_mask we do have lots of users (but also trivial I think). For Trig, most of the writes are drivers doing indio_dev->trig = iio_trigger_get(data->trig); So we either assume it's public or remove that line from drivers. Though that could break userland apps relying on some trigger to be automatically set after device probe. We can anyways have another helper to do that for drivers (not sure this field is accessed in some fastpath). Also need to better check if there are any readers for it. Anyways, it would be nice to get your inputs on the last [INTERNAL] fields in struct iio_dev Also want to mention that I deliberately removed some mailing lists given by get_maintainers as this is fairly simple and so I don't think the "spam" is worth it :). --- Nuno Sa (23): iio: accel: adxl380: make use of iio_get_masklength() iio: adc: max1118: make use of iio_for_each_active_channel() iio: adc: max1118: make use of iio_for_each_active_channel() iio: adc: mcp3911: make use of iio_for_each_active_channel() iio: adc: mt6360-adc: make use of iio_for_each_active_channel() iio: adc: rockchip_saradc: make use of iio_for_each_active_channel() iio: adc: rtq6056: make use of iio_for_each_active_channel() iio: adc: stm32-adc: make use of iio_get_masklength() iio: adc: stm32-dfsdm-adc: make use of iio_get_masklength() iio: adc: ti-adc0832: make use of iio_for_each_active_channel() iio: adc: ti-adc084s021: make use of iio_for_each_active_channel() iio: adc: ti-ads1015: make use of iio_get_masklength() iio: adc: ti-ads1119: make use of iio_get_masklength() iio: adc: ti-ads1298: make use of iio_for_each_active_channel() iio: adc: ti-adc12138: make use of iio_for_each_active_channel() iio: adc: ti-ads124s08: make use of iio_for_each_active_channel() iio: adc: ti-ads131e08: make use of iio_for_each_active_channel() iio: adc: ti-ads8688: make use of iio_for_each_active_channel() iio: adc: vf610_adc: make use of iio_get_masklength() iio: adc: xilinx-xadc-core: use new '.masklength' accessors iio: common: cros_ec_sensors_core: use new '.masklength' accessors staging: iio: impedance-analyzer: ad5933: make use of iio_get_masklength() iio: core: annotate masklength as __private drivers/iio/accel/adxl380.c | 2 +- drivers/iio/adc/ina2xx-adc.c | 3 +-- drivers/iio/adc/max1118.c | 3 +-- drivers/iio/adc/mcp3911.c | 2 +- drivers/iio/adc/mt6360-adc.c | 2 +- drivers/iio/adc/rockchip_saradc.c | 2 +- drivers/iio/adc/rtq6056.c | 2 +- drivers/iio/adc/stm32-adc.c | 4 ++-- drivers/iio/adc/stm32-dfsdm-adc.c | 2 +- drivers/iio/adc/ti-adc0832.c | 3 +-- drivers/iio/adc/ti-adc084s021.c | 3 +-- drivers/iio/adc/ti-adc12138.c | 3 +-- drivers/iio/adc/ti-ads1015.c | 2 +- drivers/iio/adc/ti-ads1119.c | 4 ++-- drivers/iio/adc/ti-ads124s08.c | 3 +-- drivers/iio/adc/ti-ads1298.c | 3 +-- drivers/iio/adc/ti-ads131e08.c | 2 +- drivers/iio/adc/ti-ads8688.c | 4 +--- drivers/iio/adc/vf610_adc.c | 2 +- drivers/iio/adc/xilinx-xadc-core.c | 5 ++--- drivers/iio/common/cros_ec_sensors/cros_ec_sensors_core.c | 8 +++----- drivers/iio/industrialio-buffer.c | 2 +- drivers/staging/iio/impedance-analyzer/ad5933.c | 5 +++-- include/linux/iio/iio.h | 4 ++-- 24 files changed, 32 insertions(+), 43 deletions(-) --- base-commit: 472438c7e0e2261c6737a8321f46ef176eef1c8f change-id: 20240724-dev-iio-masklength-private3-7a2d819acc7d -- Thanks! - Nuno Sá