Skip to contentSkip to conversation

regulator: twl: Fix the formula to calculate vsel and voltage for twl6030ldo

PATCH Applied Axel Lin 3 participants 7 messages Jul 9, 2012 – Jul 16, 2012

Subsystems (2)
lore
Share & tools
lore

Axel Lin

author
#0 lore raw

PATCH 1/2 regulator: twl: Fix the formula to calculate vsel and voltage for twl6030ldo

In twl6030ldo_set_voltage, current code use below formula to calculate vsel:

        vsel = (min_uV/1000 - 1000)/100 + 1;
This is worng because when min_uV is 1000000 uV, vsel is 1.
It should be 0 in this case.
Fix it by change the equation to: (This equation is common for linear mapping)
        vsel = DIV_ROUND_UP(min_uV - rdev->desc->min_uV, rdev->desc->uV_step);

In twl6030ldo_get_voltage, current code use below formula to calculate voltage:

        mV = 1000mv + 100mv * (vsel - 1)
This is worng because when vsel is 0, mV is 900mV. Note the min_uV is 1000mV.
Fix it by change the equation to: (This equation is common for linear mapping)
        return rdev->desc->min_uV + vsel * rdev->desc->uV_step;
---
 drivers/regulator/twl-regulator.c |   17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)
drivers/regulator/twl-regulator.c +4 -13
index 8f0bd56..bb51dec 100644
@@ -569,30 +569,21 @@ twl6030ldo_set_voltage(struct regulator_dev *rdev, int min_uV, int max_uV,
 	if ((min_uV/1000 < info->min_mV) || (max_uV/1000 > info->max_mV))
 		return -EDOM;
 
-	/*
-	 * Use the below formula to calculate vsel
-	 * mV = 1000mv + 100mv * (vsel - 1)
-	 */
-	vsel = (min_uV/1000 - 1000)/100 + 1;
+	vsel = DIV_ROUND_UP(min_uV - rdev->desc->min_uV, rdev->desc->uV_step);
 	*selector = vsel;
-	return twlreg_write(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE, vsel);
 
+	return twlreg_write(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE, vsel);
 }
 
 static int twl6030ldo_get_voltage(struct regulator_dev *rdev)
 {
 	struct twlreg_info	*info = rdev_get_drvdata(rdev);
-	int		vsel = twlreg_read(info, TWL_MODULE_PM_RECEIVER,
-								VREG_VOLTAGE);
+	int vsel = twlreg_read(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE);
 
 	if (vsel < 0)
 		return vsel;
 
-	/*
-	 * Use the below formula to calculate vsel
-	 * mV = 1000mv + 100mv * (vsel - 1)
-	 */
-	return (1000 + (100 * (vsel - 1))) * 1000;
+	return rdev->desc->min_uV + vsel * rdev->desc->uV_step;
 }
 
 static struct regulator_ops twl6030ldo_ops = {
Signature · 1 line
1.7.9.5
author
#1 lore raw

PATCH RESEND 2/2 regulator: twl: Convert twl6030ldo_ops to [get|set]_voltage_sel

---
Sorry, I CCed wrong guys. So here is the resend.
Axel

 drivers/regulator/twl-regulator.c |   24 +++++++-----------------
 1 file changed, 7 insertions(+), 17 deletions(-)
drivers/regulator/twl-regulator.c +7 -17
index bb51dec..de99b78 100644
@@ -560,37 +560,27 @@ static struct regulator_ops twl6030coresmps_ops = {
 };
 
 static int
-twl6030ldo_set_voltage(struct regulator_dev *rdev, int min_uV, int max_uV,
-		       unsigned *selector)
+twl6030ldo_set_voltage_sel(struct regulator_dev *rdev, unsigned selector)
 {
 	struct twlreg_info	*info = rdev_get_drvdata(rdev);
-	int			vsel;
 
-	if ((min_uV/1000 < info->min_mV) || (max_uV/1000 > info->max_mV))
-		return -EDOM;
-
-	vsel = DIV_ROUND_UP(min_uV - rdev->desc->min_uV, rdev->desc->uV_step);
-	*selector = vsel;
-
-	return twlreg_write(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE, vsel);
+	return twlreg_write(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE,
+			    selector);
 }
 
-static int twl6030ldo_get_voltage(struct regulator_dev *rdev)
+static int twl6030ldo_get_voltage_sel(struct regulator_dev *rdev)
 {
 	struct twlreg_info	*info = rdev_get_drvdata(rdev);
 	int vsel = twlreg_read(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE);
 
-	if (vsel < 0)
-		return vsel;
-
-	return rdev->desc->min_uV + vsel * rdev->desc->uV_step;
+	return vsel;
 }
 
 static struct regulator_ops twl6030ldo_ops = {
 	.list_voltage	= regulator_list_voltage_linear,
 
-	.set_voltage	= twl6030ldo_set_voltage,
-	.get_voltage	= twl6030ldo_get_voltage,
+	.set_voltage_sel = twl6030ldo_set_voltage_sel,
+	.get_voltage_sel = twl6030ldo_get_voltage_sel,
 
 	.enable		= twl6030reg_enable,
 	.disable	= twl6030reg_disable,
Signature · 1 line
1.7.9.5
author
#2 lore raw
In reply to Axel Lin

於 一,2012-07-09 於 11:22 +0800,Axel Lin 提到:

Quoted 7 lines

In twl6030ldo_set_voltage, current code use below formula to calculate vsel:

        vsel = (min_uV/1000 - 1000)/100 + 1;
This is worng because when min_uV is 1000000 uV, vsel is 1.
It should be 0 in this case.
Fix it by change the equation to: (This equation is common for linear mapping)
        vsel = DIV_ROUND_UP(min_uV - rdev->desc->min_uV, rdev->desc->uV_step);
Quoted 5 lines

In twl6030ldo_get_voltage, current code use below formula to calculate voltage:

        mV = 1000mv + 100mv * (vsel - 1)
This is worng because when vsel is 0, mV is 900mV. Note the min_uV is 1000mV.
Fix it by change the equation to: (This equation is common for linear mapping)
        return rdev->desc->min_uV + vsel * rdev->desc->uV_step;

While I'm thinking I need to rework this patch so that it doesn't use rdev->desc->min_uV and rdev->desc->uV_step and then can be applied to current Linus' tree.

But while I am tracking back to commit 3e3d3be79c

Author: Rajendra Nayak <[email protected]>
Date:   Thu Apr 22 14:18:32 2010 +0530

twl6030: regulator: Remove vsel tables and use formula for
calculation

All twl6030 regulators can be programmed from 1.0v to 3.3v with 100mV steps.
The below formula can be used to calculate the vsel values to be programmed in the VREG_VOLTAGE registers.

Voltage(in mV) = 1000mv + 100mv * (vsel - 1)

Ex: if vsel = 0x9, mV = 1000 + 100 * (9 -1) = 1800mV.

This patch removes all existing VSEL tables for twl6030 adjustable regulators and just uses the formula directly for vsel calculations after verifing they fall in the allowed range.

I found a problem that before commit 3e3d3be79c, the voltage tables were not linear mapping. So why we can convert these voltage mapping table to Voltage(in mV) = 1000mv + 100mv * (vsel - 1)?

Did I miss something?

Regards,
Axel

#3 lore raw
Quoted 3 lines · Axel Lin

I found a problem that before commit 3e3d3be79c, the voltage tables were not linear mapping. So why we can convert these voltage mapping table to Voltage(in mV) = 1000mv + 100mv * (vsel - 1)?

Guys, it's pretty concerning that there's been no response at all to this. Looking at the changes Axel's statement above is accurate for at least VAUX1 and VMMC, I didn't check all the regulators, so either the original code was buggy or there's an issue that needs fixing in the current code.

Applied
#6 lore raw
In reply to Axel Lin
Quoted 3 lines · Axel Lin

In twl6030ldo_set_voltage, current code use below formula to calculate vsel:

        vsel = (min_uV/1000 - 1000)/100 + 1;

This is worng because when min_uV is 1000000 uV, vsel is 1.

Applied both, thanks - even if we need to fix some of these to be using tables again this set of changes at least maintains the status quo.

#4 lore raw
In reply to Axel Lin

Hi Axel,

My apologies for the delay in responding to this one.

Quoted 5 lines · Axel Lin

於 一,2012-07-09 於 11:22 +0800,Axel Lin 提到:

Quoted 4 lines

In twl6030ldo_set_voltage, current code use below formula to calculate vsel:

         vsel = (min_uV/1000 - 1000)/100 + 1;

This is worng because when min_uV is 1000000 uV, vsel is 1. It should be 0 in this case.

Why? Do you know of any documentation which states this?
I am referring to the twl6030 Top Functional spec version 1.18 and Table - 223 "LDO Output Voltage Selection Code" clearly states that the vsel should be 1 when voltage expected is 1000000 uV.

Also refer to equation 3 "LDO Output Voltage Selection Code" in the same document and you will see this mentioned..
Absolute Voltage value = 1.0V + 0.1V * (binary value – 00000001)

Quoted 33 lines
Quoted 8 lines

Fix it by change the equation to: (This equation is common for linear mapping)

         vsel = DIV_ROUND_UP(min_uV - rdev->desc->min_uV, rdev->desc->uV_step);

In twl6030ldo_get_voltage, current code use below formula to calculate voltage:

         mV = 1000mv + 100mv * (vsel - 1)
This is worng because when vsel is 0, mV is 900mV. Note the min_uV is 1000mV.
Fix it by change the equation to: (This equation is common for linear mapping)
         return rdev->desc->min_uV + vsel * rdev->desc->uV_step;

While I'm thinking I need to rework this patch so that it doesn't use rdev->desc->min_uV and rdev->desc->uV_step and then can be applied to current Linus' tree.

But while I am tracking back to commit 3e3d3be79c

Author: Rajendra Nayak<[email protected]>
Date:   Thu Apr 22 14:18:32 2010 +0530

twl6030: regulator: Remove vsel tables and use formula for
calculation

All twl6030 regulators can be programmed from 1.0v to 3.3v with 100mV steps.
The below formula can be used to calculate the vsel values to be programmed in the VREG_VOLTAGE registers.

Voltage(in mV) = 1000mv + 100mv * (vsel - 1)

Ex: if vsel = 0x9, mV = 1000 + 100 * (9 -1) = 1800mV.

This patch removes all existing VSEL tables for twl6030 adjustable regulators and just uses the formula directly for vsel calculations after verifing they fall in the allowed range.

Quoted 7 lines

I found a problem that before commit 3e3d3be79c, the voltage tables were not linear mapping. So why we can convert these voltage mapping table to Voltage(in mV) = 1000mv + 100mv * (vsel - 1)?

Did I miss something?

All voltage tables before commit '3e3d3be79c' for twl6030 regulators were clearly wrong. They assumed similarity with twl4030 regulators which was not right.

regards,
Rajendra

Quoted 5 lines

Regards,
Axel

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

author
#5 lore raw
Quoted 9 lines · Axel Lin
Quoted 5 lines

於 一,2012-07-09 於 11:22 +0800,Axel Lin 提到:

Quoted 4 lines

In twl6030ldo_set_voltage, current code use below formula to calculate vsel:

         vsel = (min_uV/1000 - 1000)/100 + 1;

This is worng because when min_uV is 1000000 uV, vsel is 1. It should be 0 in this case.

Why? Do you know of any documentation which states this?

I double check with the datasheet again now.
You are right in this. So we cannot use linear mapping for twl6030ldo_ops here.
I'm going to send a patch to fix it, I'd appreciate if someone can review and test it.

Quoted 9 lines
Quoted 5 lines

I found a problem that before commit 3e3d3be79c, the voltage tables were not linear mapping. So why we can convert these voltage mapping table to Voltage(in mV) = 1000mv + 100mv * (vsel - 1)?

Did I miss something?

All voltage tables before commit '3e3d3be79c' for twl6030 regulators were clearly wrong. They assumed similarity with twl4030 regulators which was not right.

Good to know that.
Thanks,
Axel

Quoted 10 lines

regards,
Rajendra

Quoted 5 lines

Regards,
Axel

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html