Skip to contentSkip to conversation

usb: chipidea: imx: make over-current polarity configurable for i.MX25

RFC PATCH New Uwe Kleine-König 2 participants 2 messages Oct 24, 2017 – Nov 9, 2017

Subsystems (4)
Device tree bindings USB Arm i.MX and Layerscape Chipidea USB high speed dual role controller
lore
Share & tools
lore

Uwe Kleine-König

author
#0 lore raw

RFC PATCH 1/1 usb: chipidea: imx: make over-current polarity configurable for i.MX25

In commit 9dba516ed282 ("usb: chipidea: imx: set over current polarity per dts setting") i.MX6 and i.MX7 learned the dt property "over-current-active-high". The difference compared to i.MX25 is that on the latter the default in hardware and the driver is already active-high. For that reason the over-current-active-high property is added to the imx25 template to keep the default behaviour.

The obvious downside is that device trees that were compiled before this change don't have that property and so when run with a kernel including this change suddenly use active low logic. This could only be circumvented with another property "over-current-active-low" and different default handling for imx6/7 vs imx25 if none (or both) of the properties are present. For simplicity I stick to the easier and better understandable alternative at the cost of dt incompatibility.

Signed-off-by Uwe Kleine-K?nig <[email protected]>

---
Hello,

is it ok to break compatibility here?

Best regards
Uwe

 arch/arm/boot/dts/imx25.dtsi       |  2 ++
 drivers/usb/chipidea/usbmisc_imx.c | 15 +++++++++++++--
 2 files changed, 15 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx25.dtsi +2 -0
index 09ce8b81fafa..bb0054603846 100644
@@ -565,6 +565,7 @@
 				fsl,usbphy = <&usbphy0>;
 				phy_type = "utmi";
 				dr_mode = "otg";
+				over-current-active-high;
 				status = "disabled";
 			};
 

@@ -576,6 +577,7 @@
 				clock-names = "ipg", "ahb", "per";
 				fsl,usbmisc = <&usbmisc 1>;
 				fsl,usbphy = <&usbphy1>;
+				over-current-active-high;
 				status = "disabled";
 			};
drivers/usb/chipidea/usbmisc_imx.c +13 -2
index 9f4a0185dd60..6f3d0341357c 100644
@@ -125,16 +125,27 @@ static int usbmisc_imx25_init(struct imx_usbmisc_data *data)
 		val = readl(usbmisc->base);
 		val &= ~(MX25_OTG_SIC_MASK | MX25_OTG_PP_BIT);
 		val |= (MX25_EHCI_INTERFACE_DIFF_UNI & MX25_EHCI_INTERFACE_MASK) << MX25_OTG_SIC_SHIFT;
-		val |= (MX25_OTG_PM_BIT | MX25_OTG_OCPOL_BIT);
+		val |= MX25_OTG_PM_BIT;
+
+		if (data->oc_polarity)
+			val |= MX25_OTG_OCPOL_BIT;
+		else
+			val &= ~MX25_OTG_OCPOL_BIT;
+
 		writel(val, usbmisc->base);
 		break;
 	case 1:
 		val = readl(usbmisc->base);
 		val &= ~(MX25_H1_SIC_MASK | MX25_H1_PP_BIT |  MX25_H1_IPPUE_UP_BIT);
 		val |= (MX25_EHCI_INTERFACE_SINGLE_UNI & MX25_EHCI_INTERFACE_MASK) << MX25_H1_SIC_SHIFT;
-		val |= (MX25_H1_PM_BIT | MX25_H1_OCPOL_BIT | MX25_H1_TLL_BIT |
+		val |= (MX25_H1_PM_BIT | MX25_H1_TLL_BIT |
 			MX25_H1_USBTE_BIT | MX25_H1_IPPUE_DOWN_BIT);
 
+		if (data->oc_polarity)
+			val |= MX25_H1_OCPOL_BIT;
+		else
+			val &= ~MX25_H1_OCPOL_BIT;
+
 		writel(val, usbmisc->base);
 
 		break;
Signature · 1 line
2.11.0
#1 lore raw

RFC PATCH 1/1 usb: chipidea: imx: make over-current polarity configurable for i.MX25

Quoted 7 lines · Uwe Kleine-K?nig

In commit 9dba516ed282 ("usb: chipidea: imx: set over current polarity per dts setting") i.MX6 and i.MX7 learned the dt property "over-current-active-high". The difference compared to i.MX25 is that on the latter the default in hardware and the driver is already active-high. For that reason the over-current-active-high property is added to the imx25 template to keep the default behaviour.

Quoted 7 lines

The obvious downside is that device trees that were compiled before this change don't have that property and so when run with a kernel including this change suddenly use active low logic. This could only be circumvented with another property "over-current-active-low" and different default handling for imx6/7 vs imx25 if none (or both) of the properties are present. For simplicity I stick to the easier and better understandable alternative at the cost of dt incompatibility.

Sorry to reply late, in order to keep compatibility, my suggestion is add another property "over-current-active-low" and change imx25 related code accordingly.

Peter

Quoted 76 lines
Signed-off-by Uwe Kleine-K?nig <[email protected]>

---
Hello,

is it ok to break compatibility here?

Best regards
Uwe

 arch/arm/boot/dts/imx25.dtsi       |  2 ++
 drivers/usb/chipidea/usbmisc_imx.c | 15 +++++++++++++--
 2 files changed, 15 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx25.dtsi +2 -0
index 09ce8b81fafa..bb0054603846 100644
@@ -565,6 +565,7 @@
 				fsl,usbphy = <&usbphy0>;
 				phy_type = "utmi";
 				dr_mode = "otg";
+				over-current-active-high;
 				status = "disabled";
 			};
 

@@ -576,6 +577,7 @@
 				clock-names = "ipg", "ahb", "per";
 				fsl,usbmisc = <&usbmisc 1>;
 				fsl,usbphy = <&usbphy1>;
+				over-current-active-high;
 				status = "disabled";
 			};
drivers/usb/chipidea/usbmisc_imx.c +13 -2
index 9f4a0185dd60..6f3d0341357c 100644
@@ -125,16 +125,27 @@ static int usbmisc_imx25_init(struct imx_usbmisc_data *data)
 		val = readl(usbmisc->base);
 		val &= ~(MX25_OTG_SIC_MASK | MX25_OTG_PP_BIT);
 		val |= (MX25_EHCI_INTERFACE_DIFF_UNI & MX25_EHCI_INTERFACE_MASK) << MX25_OTG_SIC_SHIFT;
-		val |= (MX25_OTG_PM_BIT | MX25_OTG_OCPOL_BIT);
+		val |= MX25_OTG_PM_BIT;
+
+		if (data->oc_polarity)
+			val |= MX25_OTG_OCPOL_BIT;
+		else
+			val &= ~MX25_OTG_OCPOL_BIT;
+
 		writel(val, usbmisc->base);
 		break;
 	case 1:
 		val = readl(usbmisc->base);
 		val &= ~(MX25_H1_SIC_MASK | MX25_H1_PP_BIT |  MX25_H1_IPPUE_UP_BIT);
 		val |= (MX25_EHCI_INTERFACE_SINGLE_UNI & MX25_EHCI_INTERFACE_MASK) << MX25_H1_SIC_SHIFT;
-		val |= (MX25_H1_PM_BIT | MX25_H1_OCPOL_BIT | MX25_H1_TLL_BIT |
+		val |= (MX25_H1_PM_BIT | MX25_H1_TLL_BIT |
 			MX25_H1_USBTE_BIT | MX25_H1_IPPUE_DOWN_BIT);
 
+		if (data->oc_polarity)
+			val |= MX25_H1_OCPOL_BIT;
+		else
+			val &= ~MX25_H1_OCPOL_BIT;
+
 		writel(val, usbmisc->base);
 
 		break;
Signature · 7 lines
2.11.0


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Signature · 2 lines
Best Regards,
Peter Chen