diff mbox series

[v2,14/15] drivers: thermal: tsens: Create function to return sign-extended temperature

Message ID 64a3d07ebe5c4cfb4643d91f5f6605e8a4ffa48b.1566907161.git.amit.kucheria@linaro.org (mailing list archive)
State Superseded, archived
Headers show
Series thermal: qcom: tsens: Add interrupt support | expand

Commit Message

Amit Kucheria Aug. 27, 2019, 12:14 p.m. UTC
Hide the details of how to convert values read from TSENS HW to mCelsius
behind a function. All versions of the IP can be supported as a result.

Signed-off-by: Amit Kucheria <amit.kucheria@linaro.org>
---
 drivers/thermal/qcom/tsens-common.c | 50 +++++++++++++++++++++--------
 drivers/thermal/qcom/tsens.h        |  2 ++
 2 files changed, 38 insertions(+), 14 deletions(-)

Comments

Stephen Boyd Aug. 28, 2019, 12:38 a.m. UTC | #1
Quoting Amit Kucheria (2019-08-27 05:14:10)
> diff --git a/drivers/thermal/qcom/tsens-common.c b/drivers/thermal/qcom/tsens-common.c
> index ea2c46cc6a66a..06b44cfd5eab9 100644
> --- a/drivers/thermal/qcom/tsens-common.c
> +++ b/drivers/thermal/qcom/tsens-common.c
> @@ -84,13 +84,43 @@ static inline int code_to_degc(u32 adc_code, const struct tsens_sensor *s)
>         return degc;
>  }
>  
> +/**
> + * tsens_hw_to_mC - Return sign-extended temperature in mCelsius.
> + * @s:     Pointer to sensor struct

sensor? This isn't golang!

> + * @field: Index into regmap_field array pointing to temperature data
> + *
> + * This function handles temperature returned in ADC code or deciCelsius
> + * depending on IP version.
> + *
> + * Return: Temperature in milliCelsius on success, a negative errno will
> + * be returned in error cases
> + */
> +static int tsens_hw_to_mC(struct tsens_sensor *s, int field)
> +{
> +       struct tsens_priv *priv = s->priv;
> +       u32 temp = 0;
> +       int ret;
> +
> +       ret = regmap_field_read(priv->rf[field], &temp);
> +       if (ret)
> +               return ret;
> +
> +       if (priv->feat->adc) {
> +               /* Convert temperature from ADC code to milliCelsius */

Nitpick: Move this comment above the if and drop the braces.

> +               return code_to_degc(temp, s) * 1000;
> +       }
> +
> +       /* deciCelsius -> milliCelsius along with sign extension */
> +       return sign_extend32(temp, priv->tempres) * 100;
> +}
> +
>  int get_temp_tsens_valid(struct tsens_sensor *s, int *temp)
>  {
>         struct tsens_priv *priv = s->priv;
>         int hw_id = s->hw_id;
>         u32 temp_idx = LAST_TEMP_0 + hw_id;
>         u32 valid_idx = VALID_0 + hw_id;
> -       u32 last_temp = 0, valid, mask;
> +       u32 valid;
>         int ret;
>  
>         ret = regmap_field_read(priv->rf[valid_idx], &valid);
> @@ -310,6 +328,10 @@ int __init init_common(struct tsens_priv *priv)
>                         goto err_put_device;
>                 }
>         }
> +
> +       /* Save away resolution of signed temperature value for this IP */
> +       priv->tempres = priv->fields[LAST_TEMP_0].msb - priv->fields[LAST_TEMP_0].lsb;
> +

Why not just calculate this in the function that uses it? Is there a
reason to stash it away in the struct?

>         for (i = 0, j = VALID_0; i < priv->feat->max_sensors; i++, j++) {
>                 priv->rf[j] = devm_regmap_field_alloc(dev, priv->tm_map,
>                                                       priv->fields[j]);
Stephen Boyd Aug. 28, 2019, 3:44 p.m. UTC | #2
Quoting Amit Kucheria (2019-08-28 03:35:28)
> (Resending, replied only to Stephen by mistake)
> 
> On Wed, Aug 28, 2019 at 6:08 AM Stephen Boyd <swboyd@chromium.org> wrote:
> >
> > Quoting Amit Kucheria (2019-08-27 05:14:10)
> > > @@ -310,6 +328,10 @@ int __init init_common(struct tsens_priv *priv)
> > >                         goto err_put_device;
> > >                 }
> > >         }
> > > +
> > > +       /* Save away resolution of signed temperature value for this IP */
> > > +       priv->tempres = priv->fields[LAST_TEMP_0].msb - priv->fields
> [LAST_TEMP_0].lsb;
> > > +
> >
> > Why not just calculate this in the function that uses it? Is there a
> > reason to stash it away in the struct?
> 
> To avoid recalculating in an often-called function. It doesn't change for an IP
> version.
> 
> We can't make it static either inside that function since the initializer isn't
> constant.
> 

This sounds like a super micro optimization. It's a couple derefs and a
subtraction. If it isn't used anywhere else please just move it into the
function where it's used.
diff mbox series

Patch

diff --git a/drivers/thermal/qcom/tsens-common.c b/drivers/thermal/qcom/tsens-common.c
index ea2c46cc6a66a..06b44cfd5eab9 100644
--- a/drivers/thermal/qcom/tsens-common.c
+++ b/drivers/thermal/qcom/tsens-common.c
@@ -84,13 +84,43 @@  static inline int code_to_degc(u32 adc_code, const struct tsens_sensor *s)
 	return degc;
 }
 
+/**
+ * tsens_hw_to_mC - Return sign-extended temperature in mCelsius.
+ * @s:     Pointer to sensor struct
+ * @field: Index into regmap_field array pointing to temperature data
+ *
+ * This function handles temperature returned in ADC code or deciCelsius
+ * depending on IP version.
+ *
+ * Return: Temperature in milliCelsius on success, a negative errno will
+ * be returned in error cases
+ */
+static int tsens_hw_to_mC(struct tsens_sensor *s, int field)
+{
+	struct tsens_priv *priv = s->priv;
+	u32 temp = 0;
+	int ret;
+
+	ret = regmap_field_read(priv->rf[field], &temp);
+	if (ret)
+		return ret;
+
+	if (priv->feat->adc) {
+		/* Convert temperature from ADC code to milliCelsius */
+		return code_to_degc(temp, s) * 1000;
+	}
+
+	/* deciCelsius -> milliCelsius along with sign extension */
+	return sign_extend32(temp, priv->tempres) * 100;
+}
+
 int get_temp_tsens_valid(struct tsens_sensor *s, int *temp)
 {
 	struct tsens_priv *priv = s->priv;
 	int hw_id = s->hw_id;
 	u32 temp_idx = LAST_TEMP_0 + hw_id;
 	u32 valid_idx = VALID_0 + hw_id;
-	u32 last_temp = 0, valid, mask;
+	u32 valid;
 	int ret;
 
 	ret = regmap_field_read(priv->rf[valid_idx], &valid);
@@ -108,19 +138,7 @@  int get_temp_tsens_valid(struct tsens_sensor *s, int *temp)
 	}
 
 	/* Valid bit is set, OK to read the temperature */
-	ret = regmap_field_read(priv->rf[temp_idx], &last_temp);
-	if (ret)
-		return ret;
-
-	if (priv->feat->adc) {
-		/* Convert temperature from ADC code to milliCelsius */
-		*temp = code_to_degc(last_temp, s) * 1000;
-	} else {
-		mask = GENMASK(priv->fields[LAST_TEMP_0].msb,
-			       priv->fields[LAST_TEMP_0].lsb);
-		/* Convert temperature from deciCelsius to milliCelsius */
-		*temp = sign_extend32(last_temp, fls(mask) - 1) * 100;
-	}
+	*temp = tsens_hw_to_mC(s, temp_idx);
 
 	return 0;
 }
@@ -310,6 +328,10 @@  int __init init_common(struct tsens_priv *priv)
 			goto err_put_device;
 		}
 	}
+
+	/* Save away resolution of signed temperature value for this IP */
+	priv->tempres = priv->fields[LAST_TEMP_0].msb - priv->fields[LAST_TEMP_0].lsb;
+
 	for (i = 0, j = VALID_0; i < priv->feat->max_sensors; i++, j++) {
 		priv->rf[j] = devm_regmap_field_alloc(dev, priv->tm_map,
 						      priv->fields[j]);
diff --git a/drivers/thermal/qcom/tsens.h b/drivers/thermal/qcom/tsens.h
index e1d6af71b2b9a..409395e436f32 100644
--- a/drivers/thermal/qcom/tsens.h
+++ b/drivers/thermal/qcom/tsens.h
@@ -283,6 +283,7 @@  struct tsens_context {
  * struct tsens_priv - private data for each instance of the tsens IP
  * @dev: pointer to struct device
  * @num_sensors: number of sensors enabled on this device
+ * @tempres: number of bits used to represent signed temperature (resolution)
  * @tm_map: pointer to TM register address space
  * @srot_map: pointer to SROT register address space
  * @tm_offset: deal with old device trees that don't address TM and SROT
@@ -299,6 +300,7 @@  struct tsens_context {
 struct tsens_priv {
 	struct device			*dev;
 	u32				num_sensors;
+	u32				tempres;
 	struct regmap			*tm_map;
 	struct regmap			*srot_map;
 	u32				tm_offset;