[PATCH v2 0/2] usb: roles: intel: Use static mode by default

[PATCH v2 0/2] usb: roles: intel: Use static mode by default

[Saranya Gopal]

Hi Hans and Heikki,

I have addressed the review comments of Hans and resubmitting this
patch series of enabling static DRD mode for role switch in all
platforms except Cherrytrail.

Saranya Gopal (2):
  xhci-ext-caps.c: Add property to disable Intel SW switch
  usb: roles: intel: Enable static DRD mode for role switch

 drivers/usb/host/xhci-ext-caps.c               | 18 +++++++++++++++++
 drivers/usb/roles/intel-xhci-usb-role-switch.c | 28 +++++++++++++++++++++++---
 2 files changed, 43 insertions(+), 3 deletions(-)

-- 
1.9.1

[PATCH v2 1/2] xhci-ext-caps.c: Add property to disable Intel SW switch

[Saranya Gopal]

In platforms like Cherrytrail, 'SW switch enable' bit
should not be enabled for role switch. This patch
adds a property to Intel USB Role Switch platform driver
to denote that SW switch should be disabled in
Cherrytrail devices.

Signed-off-by: Saranya Gopal <saranya.gopal@intel.com>
Signed-off-by: Balaji Manoharan <m.balaji@intel.com>
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
---
 drivers/usb/host/xhci-ext-caps.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/usb/host/xhci-ext-caps.c b/drivers/usb/host/xhci-ext-caps.c
index 399113f..f498160 100644
--- a/drivers/usb/host/xhci-ext-caps.c
+++ b/drivers/usb/host/xhci-ext-caps.c
@@ -6,11 +6,20 @@
  */
 
 #include <linux/platform_device.h>
+#include <linux/property.h>
+#include <linux/pci.h>
 #include "xhci.h"
 
 #define USB_SW_DRV_NAME		"intel_xhci_usb_sw"
 #define USB_SW_RESOURCE_SIZE	0x400
 
+#define PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI	0x22b5
+
+static const struct property_entry role_switch_props[] = {
+	PROPERTY_ENTRY_BOOL("sw_switch_disable"),
+	{},
+};
+
 static void xhci_intel_unregister_pdev(void *arg)
 {
 	platform_device_unregister(arg);
@@ -21,6 +30,7 @@ static int xhci_create_intel_xhci_sw_pdev(struct xhci_hcd *xhci, u32 cap_offset)
 	struct usb_hcd *hcd = xhci_to_hcd(xhci);
 	struct device *dev = hcd->self.controller;
 	struct platform_device *pdev;
+	struct pci_dev *pci = to_pci_dev(dev);
 	struct resource	res = { 0, };
 	int ret;
 
@@ -43,6 +53,14 @@ static int xhci_create_intel_xhci_sw_pdev(struct xhci_hcd *xhci, u32 cap_offset)
 		return ret;
 	}
 
+	if (pci->device == PCI_DEVICE_ID_INTEL_CHERRYVIEW_XHCI) {
+		ret = platform_device_add_properties(pdev, role_switch_props);
+		if (ret) {
+			dev_err(dev, "failed to register device properties\n");
+			return ret;
+		}
+	}
+
 	pdev->dev.parent = dev;
 
 	ret = platform_device_add(pdev);
-- 
1.9.1

Re: [PATCH v2 1/2] xhci-ext-caps.c: Add property to disable Intel SW switch

[Heikki Krogerus]

On Wed, Aug 28, 2019 at 08:38:51PM +0530, Saranya Gopal wrote:
> In platforms like Cherrytrail, 'SW switch enable' bit
> should not be enabled for role switch. This patch
> adds a property to Intel USB Role Switch platform driver
> to denote that SW switch should be disabled in
> Cherrytrail devices.
> 
> Signed-off-by: Saranya Gopal <saranya.gopal@intel.com>
> Signed-off-by: Balaji Manoharan <m.balaji@intel.com>
> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>

You can't use my SoB anymore. With this patch you should probable put
here something like "Suggested-by: Heikki..." instead.

thanks,

-- 
heikki

[PATCH v2 2/2] usb: roles: intel: Enable static DRD mode for role switch

[Saranya Gopal]

Enable static DRD mode in Intel platforms which guarantees
successful role switch all the time. This fixes issues like
software role switch failure after cold boot and issue with
role switch when USB 3.0 cable is used. But, do not enable
static DRD mode for Cherrytrail devices which rely on firmware
for role switch.

Signed-off-by: Saranya Gopal <saranya.gopal@intel.com>
Signed-off-by: Balaji Manoharan <m.balaji@intel.com>
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
---
 drivers/usb/roles/intel-xhci-usb-role-switch.c | 28 +++++++++++++++++++++++---
 1 file changed, 25 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/roles/intel-xhci-usb-role-switch.c b/drivers/usb/roles/intel-xhci-usb-role-switch.c
index 277de96..a903bef 100644
--- a/drivers/usb/roles/intel-xhci-usb-role-switch.c
+++ b/drivers/usb/roles/intel-xhci-usb-role-switch.c
@@ -19,6 +19,7 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/property.h>
 #include <linux/usb/role.h>
 
 /* register definition */
@@ -26,6 +27,12 @@
 #define SW_VBUS_VALID			BIT(24)
 #define SW_IDPIN_EN			BIT(21)
 #define SW_IDPIN			BIT(20)
+#define SW_SWITCH_EN			BIT(16)
+
+#define DRD_CONFIG_DYNAMIC		0
+#define DRD_CONFIG_STATIC_HOST		1
+#define DRD_CONFIG_STATIC_DEVICE	2
+#define DRD_CONFIG_MASK			3
 
 #define DUAL_ROLE_CFG1			0x6c
 #define HOST_MODE			BIT(29)
@@ -37,6 +44,7 @@
 struct intel_xhci_usb_data {
 	struct usb_role_switch *role_sw;
 	void __iomem *base;
+	bool enable_sw_switch;
 };
 
 static int intel_xhci_usb_set_role(struct device *dev, enum usb_role role)
@@ -44,7 +52,7 @@ static int intel_xhci_usb_set_role(struct device *dev, enum usb_role role)
 	struct intel_xhci_usb_data *data = dev_get_drvdata(dev);
 	unsigned long timeout;
 	acpi_status status;
-	u32 glk, val;
+	u32 glk, val, drd_config;
 
 	/*
 	 * On many CHT devices ACPI event (_AEI) handlers read / modify /
@@ -59,24 +67,35 @@ static int intel_xhci_usb_set_role(struct device *dev, enum usb_role role)
 
 	pm_runtime_get_sync(dev);
 
-	/* Set idpin value as requested */
+	/*
+	 * Set idpin value as requested.
+	 * Since some devices rely on firmware setting DRD_CONFIG and
+	 * SW_SWITCH_EN bits to be zero for role switch,
+	 * do not set these bits for those devices.
+	 */
 	val = readl(data->base + DUAL_ROLE_CFG0);
 	switch (role) {
 	case USB_ROLE_NONE:
 		val |= SW_IDPIN;
 		val &= ~SW_VBUS_VALID;
+		drd_config = DRD_CONFIG_DYNAMIC;
 		break;
 	case USB_ROLE_HOST:
 		val &= ~SW_IDPIN;
 		val &= ~SW_VBUS_VALID;
+		drd_config = DRD_CONFIG_STATIC_HOST;
 		break;
 	case USB_ROLE_DEVICE:
 		val |= SW_IDPIN;
 		val |= SW_VBUS_VALID;
+		drd_config = DRD_CONFIG_STATIC_DEVICE;
 		break;
 	}
 	val |= SW_IDPIN_EN;
-
+	if (data->enable_sw_switch) {
+		val &= ~DRD_CONFIG_MASK;
+		val |= SW_SWITCH_EN | drd_config;
+	}
 	writel(val, data->base + DUAL_ROLE_CFG0);
 
 	acpi_release_global_lock(glk);
@@ -147,6 +166,9 @@ static int intel_xhci_usb_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, data);
 
+	data->enable_sw_switch = !device_property_read_bool(dev,
+						"sw_switch_disable");
+
 	data->role_sw = usb_role_switch_register(dev, &sw_desc);
 	if (IS_ERR(data->role_sw))
 		return PTR_ERR(data->role_sw);
-- 
1.9.1

Re: [PATCH v2 2/2] usb: roles: intel: Enable static DRD mode for role switch

[Heikki Krogerus]

On Wed, Aug 28, 2019 at 08:38:52PM +0530, Saranya Gopal wrote:
> Enable static DRD mode in Intel platforms which guarantees
> successful role switch all the time. This fixes issues like
> software role switch failure after cold boot and issue with
> role switch when USB 3.0 cable is used. But, do not enable
> static DRD mode for Cherrytrail devices which rely on firmware
> for role switch.
> 
> Signed-off-by: Saranya Gopal <saranya.gopal@intel.com>
> Signed-off-by: Balaji Manoharan <m.balaji@intel.com>
> Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>

Ditto. Can't use my SoB anymore.

Note. Since I only proposed using the device property, in this patch
you should also not use my "Suggested-by" tag. So I think it is OK to
use it in the first patch of this series, but not in this second patch.

thanks,

-- 
heikki

RE: [PATCH v2 2/2] usb: roles: intel: Enable static DRD mode for role switch

[Gopal, Saranya]

> On Wed, Aug 28, 2019 at 08:38:52PM +0530, Saranya Gopal wrote:
> > Enable static DRD mode in Intel platforms which guarantees
> > successful role switch all the time. This fixes issues like
> > software role switch failure after cold boot and issue with
> > role switch when USB 3.0 cable is used. But, do not enable
> > static DRD mode for Cherrytrail devices which rely on firmware
> > for role switch.
> >
> > Signed-off-by: Saranya Gopal <saranya.gopal@intel.com>
> > Signed-off-by: Balaji Manoharan <m.balaji@intel.com>
> > Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
> 
> Ditto. Can't use my SoB anymore.
> 
> Note. Since I only proposed using the device property, in this patch
> you should also not use my "Suggested-by" tag. So I think it is OK to
> use it in the first patch of this series, but not in this second patch.

Sure, let me remove the tag and resend the patches.
> 
> thanks,
> 
> --
> heikki