Thread (17 messages) 17 messages, 4 authors, 2022-08-29

Re: [PATCH v5 6/7] iio: add MEMSensing MSA311 3-axis accelerometer driver

From: Andy Shevchenko <hidden>
Date: 2022-08-12 22:35:22
Also in: linux-iio, lkml

On Fri, Aug 12, 2022 at 7:52 PM Dmitry Rokosov [off-list ref] wrote:
MSA311 is a tri-axial, low-g accelerometer with I2C digital output for
sensitivity consumer applications. It has dynamic user-selectable full
scales range of +-2g/+-4g/+-8g/+-16g and allows acceleration measurements
with output data rates from 1Hz to 1000Hz.
Spec: https://cdn-shop.adafruit.com/product-files/5309/MSA311-V1.1-ENG.pdf
Can you use Datasheet: tag below (just before your SoB tag)?
This driver supports following MSA311 features:
    - IIO interface
    - Different power modes: NORMAL and SUSPEND (using pm_runtime)
    - ODR (Output Data Rate) selection
    - Scale and samp_freq selection
    - IIO triggered buffer, IIO reg access
    - NEW_DATA interrupt + trigger

Below features to be done:
    - Motion Events: ACTIVE, TAP, ORIENT, FREEFALL
    - Low Power mode
...
+static const struct {
+       int val;
+       int val2;
+} msa311_fs_table[] = {
+       {0, 9580}, {0, 19160}, {0, 38320}, {0, 76641}
+};
At least you may deduplicate the type definition for these data structures, like

struct iio_float {
  int integer;
  int fract;
};

...
+               dev_err(dev,
+                       "cannot set odr %u.%luHz, not available in %s mode\n",
+                       msa311_odr_table[odr].val,
+                       msa311_odr_table[odr].val2 / MILLIHZ_PER_HZ,
Logically it's MICROHZ_PER_MILLIHZ.
+                       msa311_pwr_modes[pwr_mode]);
...
+       freq_uhz = msa311_odr_table[odr].val * MICROHZ_PER_HZ +
+                  msa311_odr_table[odr].val2;
+       wait_ms = (MICROHZ_PER_HZ / freq_uhz) * MSEC_PER_SEC;
On the contrary this seems correct.

...
+       err = iio_device_claim_direct_mode(indio_dev);
+       if (err)
+               return err;
+
+       mutex_lock(&msa311->lock);
+       err = msa311_get_axis(msa311, chan, &axis);
+       mutex_unlock(&msa311->lock);
+
+       iio_device_release_direct_mode(indio_dev);
+
+       if (err) {
+               dev_err(dev, "cannot get axis %s (%pe)\n",
+                       chan->datasheet_name, ERR_PTR(err));
+               return err;
+       }
+
+       pm_runtime_mark_last_busy(dev);
+       pm_runtime_put_autosuspend(dev);
All error cases above miss the PM restoration. Is it on purpose?
Otherwise fix it here and check everywhere else.

...
+       used = scnprintf(msa311->chip_name, sizeof(msa311->chip_name),
+                        "msa311-%hhx", partid);
+       msa311->chip_name[used] = '\0';
What is this for exactly?

...
+       /* Disable all interrupts by default */
+       err = regmap_write(msa311->regs, MSA311_INT_SET_0_REG, 0);
+       if (err)
+               return dev_err_probe(dev, err,
+                                    "cannot disable set0 interrupts\n");
+
+       err = regmap_write(msa311->regs, MSA311_INT_SET_1_REG, 0);
+       if (err)
+               return dev_err_probe(dev, err,
+                                    "cannot disable set1 interrupts\n");
Wondering if the above can be combined to bulk write.

...
+       /* Unmap all INT1 interrupts by default */
+       err = regmap_write(msa311->regs, MSA311_INT_MAP_0_REG, 0);
+       if (err)
+               return dev_err_probe(dev, err,
+                                    "failed to unmap map0 interrupts\n");
+
+       err = regmap_write(msa311->regs, MSA311_INT_MAP_1_REG, 0);
+       if (err)
+               return dev_err_probe(dev, err,
+                                    "failed to unmap map1 interrupts\n");
Ditto.

...
+       /* Keep going without interrupts if no initialized I2C irq */
IRQ

...
+                                    "cannot allocate newdata trigger\n");
In case you wish to make them shorter, you can use "can't".

...
+                                    "can't register newdata trigger\n");
...esp. if you already do this in the other message, but not in all.
So, some consistency would be good to have.

...

I avoided commenting on the points that were not addressed and you
explained why.

-- 
With Best Regards,
Andy Shevchenko
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help