Re: [PATCH 1/4] hwmon: (pmbus) Add get_error_flags support to regulator ops

From: Zev Weiss
Date: Thu Feb 17 2022 - 18:47:19 EST

On Thu, Feb 17, 2022 at 10:11:32AM PST, Guenter Roeck wrote:
On 2/17/22 02:44, Zev Weiss wrote:
The various PMBus status bits don't all map perfectly to the more
limited set of REGULATOR_ERROR_* flags, but there's a reasonable
number where they correspond well enough.

Signed-off-by: Zev Weiss <zev@xxxxxxxxxxxxxxxxx>
---
drivers/hwmon/pmbus/pmbus_core.c | 97 ++++++++++++++++++++++++++++++++
1 file changed, 97 insertions(+)

diff --git a/drivers/hwmon/pmbus/pmbus_core.c b/drivers/hwmon/pmbus/pmbus_core.c
index 776ee2237be2..a274e8e524a5 100644
--- a/drivers/hwmon/pmbus/pmbus_core.c
+++ b/drivers/hwmon/pmbus/pmbus_core.c
@@ -2417,10 +2417,107 @@ static int pmbus_regulator_disable(struct regulator_dev *rdev)
return _pmbus_regulator_on_off(rdev, 0);
}
+/* A PMBus status flag and the corresponding REGULATOR_ERROR_* flag */
+struct pmbus_regulator_status_assoc {
+ int pflag, rflag;
+};
+
+/* PMBus->regulator bit mappings for a PMBus status register */
+struct pmbus_regulator_status_category {
+ int func;
+ int reg;
+ const struct pmbus_regulator_status_assoc *bits; /* zero-terminated */
+};
+
+static const struct pmbus_regulator_status_category pmbus_regulator_flag_map[] = {
+ {
+ .func = PMBUS_HAVE_STATUS_VOUT,
+ .reg = PMBUS_STATUS_VOUT,
+ .bits = (const struct pmbus_regulator_status_assoc[]) {
+ { PB_VOLTAGE_UV_WARNING, REGULATOR_ERROR_UNDER_VOLTAGE_WARN },
+ { PB_VOLTAGE_UV_FAULT, REGULATOR_ERROR_UNDER_VOLTAGE },
+ { PB_VOLTAGE_OV_WARNING, REGULATOR_ERROR_OVER_VOLTAGE_WARN },
+ { PB_VOLTAGE_OV_FAULT, REGULATOR_ERROR_REGULATION_OUT },
+ { },
+ },
+ }, {
+ .func = PMBUS_HAVE_STATUS_IOUT,
+ .reg = PMBUS_STATUS_IOUT,
+ .bits = (const struct pmbus_regulator_status_assoc[]) {
+ { PB_IOUT_OC_WARNING, REGULATOR_ERROR_OVER_CURRENT_WARN },
+ { PB_IOUT_OC_FAULT, REGULATOR_ERROR_OVER_CURRENT },
+ { PB_IOUT_OC_LV_FAULT, REGULATOR_ERROR_OVER_CURRENT },
+ { },
+ },
+ }, {
+ .func = PMBUS_HAVE_STATUS_TEMP,
+ .reg = PMBUS_STATUS_TEMPERATURE,
+ .bits = (const struct pmbus_regulator_status_assoc[]) {
+ { PB_TEMP_OT_WARNING, REGULATOR_ERROR_OVER_TEMP_WARN },
+ { PB_TEMP_OT_FAULT, REGULATOR_ERROR_OVER_TEMP },
+ { },
+ },
+ },
+};
+
+static int pmbus_regulator_get_error_flags(struct regulator_dev *rdev, unsigned int *flags)
+{
+ int i, status, statusreg;
+ const struct pmbus_regulator_status_category *cat;
+ const struct pmbus_regulator_status_assoc *bit;
+ struct device *dev = rdev_get_dev(rdev);
+ struct i2c_client *client = to_i2c_client(dev->parent);
+ struct pmbus_data *data = i2c_get_clientdata(client);
+ u8 page = rdev_get_id(rdev);
+ int func = data->info->func[page];
+
+ *flags = 0;
+
+ for (i = 0; i < ARRAY_SIZE(pmbus_regulator_flag_map); i++) {
+ cat = &pmbus_regulator_flag_map[i];
+ if (!(func & cat->func))
+ continue;
+
+ status = pmbus_read_byte_data(client, page, cat->reg);
+ if (status < 0)
+ return status;
+
+ for (bit = cat->bits; bit->pflag; bit++) {
+ if (status & bit->pflag)
+ *flags |= bit->rflag;
+ }
+ }
+
+ /*
+ * Map what bits of STATUS_{WORD,BYTE} we can to REGULATOR_ERROR_*
+ * bits. Some of the other bits are tempting (especially for cases
+ * where we don't have the relevant PMBUS_HAVE_STATUS_*
+ * functionality), but there's an unfortunate ambiguity in that
+ * they're defined as indicating a fault *or* a warning, so we can't
+ * easily determine whether to report REGULATOR_ERROR_<foo> or
+ * REGULATOR_ERROR_<foo>_WARN.
+ */
+ statusreg = data->has_status_word ? PMBUS_STATUS_WORD : PMBUS_STATUS_BYTE;
+ status = pmbus_get_status(client, page, statusreg);
+

pmbus_get_status() calls data->read_status if PMBUS_STATUS_WORD is provided
as parameter, and data->read_status is set to pmbus_read_status_byte()
if reading the word status is not supported. Given that, why not just call
pmbus_get_status(client, page, PMBUS_STATUS_WORD) ?

Good point, I'll change it to do that instead. (And send v2 separately from the power-efuse driver patches.)


+ if (status < 0)
+ return status;
+
+ if (pmbus_regulator_is_enabled(rdev) && (status & PB_STATUS_OFF))
+ *flags |= REGULATOR_ERROR_FAIL;
+ if (status & PB_STATUS_IOUT_OC)
+ *flags |= REGULATOR_ERROR_OVER_CURRENT;

If the current status register is supported, this effectively means that
an overcurrent warning is always reported as both REGULATOR_ERROR_OVER_CURRENT
and REGULATOR_ERROR_OVER_CURRENT_WARN. Is that intentional ?


No, but I don't think (by my reading of the spec) that's what would happen?

I'm looking at table 16 ("STATUS_WORD Message Contents") in section 17.2 ("STATUS_WORD") of Part II of revision 1.3.1 of the PMBus spec, which says that bit 4 of the low byte (PB_STATUS_IOUT_OC) indicates an output overcurrent fault, not a warning (in contrast to most of the other bits, which may indicate either).


+ if (status & PB_STATUS_VOUT_OV)
+ *flags |= REGULATOR_ERROR_REGULATION_OUT;

Same for voltage.

Likewise, PB_STATUS_VOUT_OV is specified as indicating a fault, not a warning.

On the other side, temperature limit violations are not
reported at all unless the temperature status register exists.
That seems to be a bit inconsistent to me.


Right -- that's because PB_STATUS_TEMPERATURE is one of the "fault or warning" bits (unlike VOUT_OV and IOUT_OC), and hence it's an ambiguous case as described in the comment before the pmbus_get_status() call.

It's certainly not ideal, but it seemed like the best approach I could see given the semantics of the available flags -- I'm open to other possibilities though if there's something else that would work better.

Thanks for the review,
Zev