[PATCH v4 0/4] phy: rcar-gen3-usb2: add support for r8a77995

[PATCH v4 0/4] phy: rcar-gen3-usb2: add support for r8a77995

[Yoshihiro Shimoda]

This patch set is based on the latest phy.git / next branch
(the commit id = 415060b21f318e009d865b4bcbf8f220ebc36964)

After this patch set is applied, a usb 2.0 host node that is combined
with usb 2.0 peripheral needs 'dr_mode = "otg";' property.

Changes from v3:
 - Use enum phy_mode in patch 2.
 - Remove "can_role_swap" parameter and revise the commit log in patch 2
   because a case of "can_role_swap = true" and "has_otg = false" is
   not supported for role swap at the moment.
 - Changes the name of "has_otg" to "has_otg_pins" in patch 3.
 - Use of_device_get_match_data() instead of of_match_device()
 - Add "Reviewed-by:" into patch No.1, 2 and 4. (Simon-san, Thanks!)
  - Since patch No.3 is changed in big at v4, I dropped his "Reviewed-by".

Changes from v2:
 - Revise the commit log ("SoCs" is not third-person singular present).

Changes from v1:
 - Revise typo "wronly" to "wrongly".
 - Remove RCAR_GEN3_PHY_HAS_DEDICATED_PINS from generic gen3 entry in
   patch 3/4
 - Remove the driver change from patch 4/4.
 - Revise the commit log of patch 4/4.


Yoshihiro Shimoda (4):
  phy: rcar-gen3-usb2: check dr_mode for otg mode
  phy: rcar-gen3-usb2: cleanup the role_{store,show}()
  phy: rcar-gen3-usb2: add SoC-specific parameter for dedicated pins
  phy: rcar-gen3-usb2: add binding for r8a77995

 .../devicetree/bindings/phy/rcar-gen3-phy-usb2.txt |  2 +
 drivers/phy/renesas/phy-rcar-gen3-usb2.c           | 95 +++++++++++++++-------
 2 files changed, 68 insertions(+), 29 deletions(-)

-- 
1.9.1

[PATCH v4 0/4] phy: rcar-gen3-usb2: add support for r8a77995

[Yoshihiro Shimoda]

This patch set is based on the latest phy.git / next branch
(the commit id = 415060b21f318e009d865b4bcbf8f220ebc36964)

After this patch set is applied, a usb 2.0 host node that is combined
with usb 2.0 peripheral needs 'dr_mode = "otg";' property.

Changes from v3:
 - Use enum phy_mode in patch 2.
 - Remove "can_role_swap" parameter and revise the commit log in patch 2
   because a case of "can_role_swap = true" and "has_otg = false" is
   not supported for role swap at the moment.
 - Changes the name of "has_otg" to "has_otg_pins" in patch 3.
 - Use of_device_get_match_data() instead of of_match_device()
 - Add "Reviewed-by:" into patch No.1, 2 and 4. (Simon-san, Thanks!)
  - Since patch No.3 is changed in big at v4, I dropped his "Reviewed-by".

Changes from v2:
 - Revise the commit log ("SoCs" is not third-person singular present).

Changes from v1:
 - Revise typo "wronly" to "wrongly".
 - Remove RCAR_GEN3_PHY_HAS_DEDICATED_PINS from generic gen3 entry in
   patch 3/4
 - Remove the driver change from patch 4/4.
 - Revise the commit log of patch 4/4.


Yoshihiro Shimoda (4):
  phy: rcar-gen3-usb2: check dr_mode for otg mode
  phy: rcar-gen3-usb2: cleanup the role_{store,show}()
  phy: rcar-gen3-usb2: add SoC-specific parameter for dedicated pins
  phy: rcar-gen3-usb2: add binding for r8a77995

 .../devicetree/bindings/phy/rcar-gen3-phy-usb2.txt |  2 +
 drivers/phy/renesas/phy-rcar-gen3-usb2.c           | 95 +++++++++++++++-------
 2 files changed, 68 insertions(+), 29 deletions(-)

-- 
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH v4 1/4] phy: rcar-gen3-usb2: check dr_mode for otg mode

[Yoshihiro Shimoda]

The previous code assumed a channel has otg capability if a channel
has interrupt property. But, it is not good because:
 - Battery charging feature also needs interrupt property.
 - Some R-Car Gen3 SoCs (e.g. R-Car D3) don't have OTG capability.

So, this patch checks whether usb 2.0 host node has dr_mode property or
not. If it has 'dr_mode = "otg";', this driver enables otg capability.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
index 54c3429..e00e99a 100644
--- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
+++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
@@ -1,7 +1,7 @@
 /*
  * Renesas R-Car Gen3 for USB2.0 PHY driver
  *
- * Copyright (C) 2015 Renesas Electronics Corporation
+ * Copyright (C) 2015-2017 Renesas Electronics Corporation
  *
  * This is based on the phy-rcar-gen2 driver:
  * Copyright (C) 2014 Renesas Solutions Corp.
@@ -22,6 +22,7 @@
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/regulator/consumer.h>
+#include <linux/usb/of.h>
 #include <linux/workqueue.h>
 
 /******* USB2.0 Host registers (original offset is +0x200) *******/
@@ -415,13 +416,16 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 	/* call request_irq for OTG */
 	irq = platform_get_irq(pdev, 0);
 	if (irq >= 0) {
-		int ret;
-
 		INIT_WORK(&channel->work, rcar_gen3_phy_usb2_work);
 		irq = devm_request_irq(dev, irq, rcar_gen3_phy_usb2_irq,
 				       IRQF_SHARED, dev_name(dev), channel);
 		if (irq < 0)
 			dev_err(dev, "No irq handler (%d)\n", irq);
+	}
+
+	if (of_usb_get_dr_mode_by_phy(dev->of_node, 0) == USB_DR_MODE_OTG) {
+		int ret;
+
 		channel->has_otg = true;
 		channel->extcon = devm_extcon_dev_allocate(dev,
 							rcar_gen3_phy_cable);
-- 
1.9.1

[PATCH v4 2/4] phy: rcar-gen3-usb2: cleanup the role_{store,show}()

[Yoshihiro Shimoda]

This patch cleanups the role_{store,show}() and replaces the local
"bool" for host/device mode selection with the "enum phy_mode" in
the role_store().

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 58 ++++++++++++++++++++++----------
 1 file changed, 40 insertions(+), 18 deletions(-)

diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
index e00e99a..7619468 100644
--- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
+++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
@@ -219,14 +219,10 @@ static bool rcar_gen3_is_host(struct rcar_gen3_chan *ch)
 	return !(readl(ch->base + USB2_COMMCTRL) & USB2_COMMCTRL_OTG_PERI);
 }
 
-static ssize_t role_store(struct device *dev, struct device_attribute *attr,
-			  const char *buf, size_t count)
+static ssize_t rcar_gen3_otg_change_role(struct rcar_gen3_chan *ch,
+					 enum phy_mode mode)
 {
-	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
-	bool is_b_device, is_host, new_mode_is_host;
-
-	if (!ch->has_otg || !ch->phy->init_count)
-		return -EIO;
+	bool is_b_device, is_host;
 
 	/*
 	 * is_b_device: true is B-Device. false is A-Device.
@@ -234,18 +230,13 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
 	 */
 	is_b_device = rcar_gen3_check_id(ch);
 	is_host = rcar_gen3_is_host(ch);
-	if (!strncmp(buf, "host", strlen("host")))
-		new_mode_is_host = true;
-	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
-		new_mode_is_host = false;
-	else
-		return -EINVAL;
 
 	/* If current and new mode is the same, this returns the error */
-	if (is_host == new_mode_is_host)
+	if ((is_host && mode == PHY_MODE_USB_HOST) ||
+	    (!is_host && mode == PHY_MODE_USB_DEVICE))
 		return -EINVAL;
 
-	if (new_mode_is_host) {		/* And is_host must be false */
+	if (mode == PHY_MODE_USB_HOST) { /* And is_host must be false */
 		if (!is_b_device)	/* A-Peripheral */
 			rcar_gen3_init_from_a_peri_to_a_host(ch);
 		else			/* B-Peripheral */
@@ -257,6 +248,32 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
 			rcar_gen3_init_for_peri(ch);
 	}
 
+	return 0;
+}
+
+static ssize_t role_store(struct device *dev, struct device_attribute *attr,
+			  const char *buf, size_t count)
+{
+	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
+	enum phy_mode mode;
+	ssize_t ret = -EIO;
+
+	if (!ch->phy->init_count)
+		return -EIO;
+
+	if (!strncmp(buf, "host", strlen("host")))
+		mode = PHY_MODE_USB_HOST;
+	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
+		mode = PHY_MODE_USB_DEVICE;
+	else
+		return -EINVAL;
+
+	if (ch->has_otg)
+		ret = rcar_gen3_otg_change_role(ch, mode);
+
+	if (ret < 0)
+		return ret;
+
 	return count;
 }
 
@@ -264,12 +281,17 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
 			 char *buf)
 {
 	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
+	bool is_host;
 
-	if (!ch->has_otg || !ch->phy->init_count)
+	if (!ch->phy->init_count)
+		return -EIO;
+
+	if (ch->has_otg)
+		is_host = rcar_gen3_is_host(ch);
+	else
 		return -EIO;
 
-	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
-							    "peripheral");
+	return sprintf(buf, "%s\n", is_host ? "host" : "peripheral");
 }
 static DEVICE_ATTR_RW(role);
 
-- 
1.9.1

[PATCH v4 3/4] phy: rcar-gen3-usb2: add SoC-specific parameter for dedicated pins

[Yoshihiro Shimoda]

This patch adds SoC-specific parameter to avoid reading/writing
specific registers wrongly if this driver runs on a SoC which doesn't
have dedicated pins (e.g. R-Car D3). This patch also changes the
value "has_otg" to "has_otg_pins" for slightly easier reading of
the code.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 31 +++++++++++++++++++++----------
 1 file changed, 21 insertions(+), 10 deletions(-)

diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
index 7619468..d097539 100644
--- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
+++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
@@ -18,6 +18,7 @@
 #include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
+#include <linux/of_device.h>
 #include <linux/phy/phy.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
@@ -80,6 +81,8 @@
 #define USB2_ADPCTRL_IDPULLUP		BIT(5)	/* 1 = ID sampling is enabled */
 #define USB2_ADPCTRL_DRVVBUS		BIT(4)
 
+#define RCAR_GEN3_PHY_HAS_DEDICATED_PINS	1
+
 struct rcar_gen3_chan {
 	void __iomem *base;
 	struct extcon_dev *extcon;
@@ -87,7 +90,7 @@ struct rcar_gen3_chan {
 	struct regulator *vbus;
 	struct work_struct work;
 	bool extcon_host;
-	bool has_otg;
+	bool has_otg_pins;
 };
 
 static void rcar_gen3_phy_usb2_work(struct work_struct *work)
@@ -268,7 +271,7 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
 	else
 		return -EINVAL;
 
-	if (ch->has_otg)
+	if (ch->has_otg_pins)
 		ret = rcar_gen3_otg_change_role(ch, mode);
 
 	if (ret < 0)
@@ -286,7 +289,7 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
 	if (!ch->phy->init_count)
 		return -EIO;
 
-	if (ch->has_otg)
+	if (ch->has_otg_pins)
 		is_host = rcar_gen3_is_host(ch);
 	else
 		return -EIO;
@@ -326,7 +329,7 @@ static int rcar_gen3_phy_usb2_init(struct phy *p)
 	writel(USB2_OC_TIMSET_INIT, usb2_base + USB2_OC_TIMSET);
 
 	/* Initialize otg part */
-	if (channel->has_otg)
+	if (channel->has_otg_pins)
 		rcar_gen3_init_otg(channel);
 
 	return 0;
@@ -400,9 +403,17 @@ static irqreturn_t rcar_gen3_phy_usb2_irq(int irq, void *_ch)
 }
 
 static const struct of_device_id rcar_gen3_phy_usb2_match_table[] = {
-	{ .compatible = "renesas,usb2-phy-r8a7795" },
-	{ .compatible = "renesas,usb2-phy-r8a7796" },
-	{ .compatible = "renesas,rcar-gen3-usb2-phy" },
+	{
+		.compatible = "renesas,usb2-phy-r8a7795",
+		.data = (void *)RCAR_GEN3_PHY_HAS_DEDICATED_PINS,
+	},
+	{
+		.compatible = "renesas,usb2-phy-r8a7796",
+		.data = (void *)RCAR_GEN3_PHY_HAS_DEDICATED_PINS,
+	},
+	{
+		.compatible = "renesas,rcar-gen3-usb2-phy",
+	},
 	{ }
 };
 MODULE_DEVICE_TABLE(of, rcar_gen3_phy_usb2_match_table);
@@ -448,7 +459,7 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 	if (of_usb_get_dr_mode_by_phy(dev->of_node, 0) == USB_DR_MODE_OTG) {
 		int ret;
 
-		channel->has_otg = true;
+		channel->has_otg_pins = (uintptr_t)of_device_get_match_data(dev);
 		channel->extcon = devm_extcon_dev_allocate(dev,
 							rcar_gen3_phy_cable);
 		if (IS_ERR(channel->extcon))
@@ -490,7 +501,7 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 		dev_err(dev, "Failed to register PHY provider\n");
 		ret = PTR_ERR(provider);
 		goto error;
-	} else if (channel->has_otg) {
+	} else if (channel->has_otg_pins) {
 		int ret;
 
 		ret = device_create_file(dev, &dev_attr_role);
@@ -510,7 +521,7 @@ static int rcar_gen3_phy_usb2_remove(struct platform_device *pdev)
 {
 	struct rcar_gen3_chan *channel = platform_get_drvdata(pdev);
 
-	if (channel->has_otg)
+	if (channel->has_otg_pins)
 		device_remove_file(&pdev->dev, &dev_attr_role);
 
 	pm_runtime_disable(&pdev->dev);
-- 
1.9.1

[PATCH v4 4/4] phy: rcar-gen3-usb2: add binding for r8a77995

[Yoshihiro Shimoda]

This patch adds binding for r8a77995 (R-Car D3). Since r8a77995 doesn't
have dedicated pins (ID, VBUS), this will match against the generic
fallback on R-Car D3.

For now, this driver doesn't support usb role swap for r8a77995.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
---
 Documentation/devicetree/bindings/phy/rcar-gen3-phy-usb2.txt | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/phy/rcar-gen3-phy-usb2.txt b/Documentation/devicetree/bindings/phy/rcar-gen3-phy-usb2.txt
index ace9cce..99b651b 100644
--- a/Documentation/devicetree/bindings/phy/rcar-gen3-phy-usb2.txt
+++ b/Documentation/devicetree/bindings/phy/rcar-gen3-phy-usb2.txt
@@ -8,6 +8,8 @@ Required properties:
 	      SoC.
 	      "renesas,usb2-phy-r8a7796" if the device is a part of an R8A7796
 	      SoC.
+	      "renesas,usb2-phy-r8a77995" if the device is a part of an
+	      R8A77995 SoC.
 	      "renesas,rcar-gen3-usb2-phy" for a generic R-Car Gen3 compatible device.
 
 	      When compatible with the generic version, nodes must list the
-- 
1.9.1

Re: [PATCH v4 2/4] phy: rcar-gen3-usb2: cleanup the role_{store,show}()

[Simon Horman]

On Tue, Oct 10, 2017 at 02:58:28PM +0900, Yoshihiro Shimoda wrote:
> This patch cleanups the role_{store,show}() and replaces the local
> "bool" for host/device mode selection with the "enum phy_mode" in
> the role_store().
> 
> Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> ---
>  drivers/phy/renesas/phy-rcar-gen3-usb2.c | 58 ++++++++++++++++++++++----------
>  1 file changed, 40 insertions(+), 18 deletions(-)
> 
> diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> index e00e99a..7619468 100644
> --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> @@ -219,14 +219,10 @@ static bool rcar_gen3_is_host(struct rcar_gen3_chan *ch)
>  	return !(readl(ch->base + USB2_COMMCTRL) & USB2_COMMCTRL_OTG_PERI);
>  }
>  
> -static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> -			  const char *buf, size_t count)
> +static ssize_t rcar_gen3_otg_change_role(struct rcar_gen3_chan *ch,
> +					 enum phy_mode mode)
>  {
> -	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> -	bool is_b_device, is_host, new_mode_is_host;
> -
> -	if (!ch->has_otg || !ch->phy->init_count)
> -		return -EIO;
> +	bool is_b_device, is_host;
>  
>  	/*
>  	 * is_b_device: true is B-Device. false is A-Device.
> @@ -234,18 +230,13 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
>  	 */
>  	is_b_device = rcar_gen3_check_id(ch);
>  	is_host = rcar_gen3_is_host(ch);
> -	if (!strncmp(buf, "host", strlen("host")))
> -		new_mode_is_host = true;
> -	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> -		new_mode_is_host = false;
> -	else
> -		return -EINVAL;
>  
>  	/* If current and new mode is the same, this returns the error */
> -	if (is_host == new_mode_is_host)
> +	if ((is_host && mode == PHY_MODE_USB_HOST) ||
> +	    (!is_host && mode == PHY_MODE_USB_DEVICE))

Maybe it is not worth the effort, but it seems that if
rcar_gen3_is_host returned enum phy_mode then the above
could be changed into a single comparison.

>  		return -EINVAL;
>  
> -	if (new_mode_is_host) {		/* And is_host must be false */
> +	if (mode == PHY_MODE_USB_HOST) { /* And is_host must be false */
>  		if (!is_b_device)	/* A-Peripheral */
>  			rcar_gen3_init_from_a_peri_to_a_host(ch);
>  		else			/* B-Peripheral */
> @@ -257,6 +248,32 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
>  			rcar_gen3_init_for_peri(ch);
>  	}
>  
> +	return 0;
> +}
> +
> +static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> +			  const char *buf, size_t count)
> +{
> +	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> +	enum phy_mode mode;
> +	ssize_t ret = -EIO;
> +
> +	if (!ch->phy->init_count)
> +		return -EIO;
> +
> +	if (!strncmp(buf, "host", strlen("host")))
> +		mode = PHY_MODE_USB_HOST;
> +	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> +		mode = PHY_MODE_USB_DEVICE;
> +	else
> +		return -EINVAL;
> +
> +	if (ch->has_otg)
> +		ret = rcar_gen3_otg_change_role(ch, mode);
> +
> +	if (ret < 0)
> +		return ret;
> +
>  	return count;
>  }
>  
> @@ -264,12 +281,17 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
>  			 char *buf)
>  {
>  	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> +	bool is_host;
>  
> -	if (!ch->has_otg || !ch->phy->init_count)
> +	if (!ch->phy->init_count)
> +		return -EIO;
> +
> +	if (ch->has_otg)
> +		is_host = rcar_gen3_is_host(ch);
> +	else
>  		return -EIO;
>  
> -	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
> -							    "peripheral");
> +	return sprintf(buf, "%s\n", is_host ? "host" : "peripheral");
>  }

I'm not sure why the above hunk is necessary.
They function seems logically the same before and after.

>  static DEVICE_ATTR_RW(role);
>  
> -- 
> 1.9.1
> 

Re: [PATCH v4 2/4] phy: rcar-gen3-usb2: cleanup the role_{store,show}()

[Simon Horman]

On Tue, Oct 10, 2017 at 02:58:28PM +0900, Yoshihiro Shimoda wrote:
> This patch cleanups the role_{store,show}() and replaces the local
> "bool" for host/device mode selection with the "enum phy_mode" in
> the role_store().
> 
> Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh-zM6kxYcvzFBBDgjK7y7TUQ@public.gmane.org>
> ---
>  drivers/phy/renesas/phy-rcar-gen3-usb2.c | 58 ++++++++++++++++++++++----------
>  1 file changed, 40 insertions(+), 18 deletions(-)
> 
> diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> index e00e99a..7619468 100644
> --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> @@ -219,14 +219,10 @@ static bool rcar_gen3_is_host(struct rcar_gen3_chan *ch)
>  	return !(readl(ch->base + USB2_COMMCTRL) & USB2_COMMCTRL_OTG_PERI);
>  }
>  
> -static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> -			  const char *buf, size_t count)
> +static ssize_t rcar_gen3_otg_change_role(struct rcar_gen3_chan *ch,
> +					 enum phy_mode mode)
>  {
> -	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> -	bool is_b_device, is_host, new_mode_is_host;
> -
> -	if (!ch->has_otg || !ch->phy->init_count)
> -		return -EIO;
> +	bool is_b_device, is_host;
>  
>  	/*
>  	 * is_b_device: true is B-Device. false is A-Device.
> @@ -234,18 +230,13 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
>  	 */
>  	is_b_device = rcar_gen3_check_id(ch);
>  	is_host = rcar_gen3_is_host(ch);
> -	if (!strncmp(buf, "host", strlen("host")))
> -		new_mode_is_host = true;
> -	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> -		new_mode_is_host = false;
> -	else
> -		return -EINVAL;
>  
>  	/* If current and new mode is the same, this returns the error */
> -	if (is_host == new_mode_is_host)
> +	if ((is_host && mode == PHY_MODE_USB_HOST) ||
> +	    (!is_host && mode == PHY_MODE_USB_DEVICE))

Maybe it is not worth the effort, but it seems that if
rcar_gen3_is_host returned enum phy_mode then the above
could be changed into a single comparison.

>  		return -EINVAL;
>  
> -	if (new_mode_is_host) {		/* And is_host must be false */
> +	if (mode == PHY_MODE_USB_HOST) { /* And is_host must be false */
>  		if (!is_b_device)	/* A-Peripheral */
>  			rcar_gen3_init_from_a_peri_to_a_host(ch);
>  		else			/* B-Peripheral */
> @@ -257,6 +248,32 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
>  			rcar_gen3_init_for_peri(ch);
>  	}
>  
> +	return 0;
> +}
> +
> +static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> +			  const char *buf, size_t count)
> +{
> +	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> +	enum phy_mode mode;
> +	ssize_t ret = -EIO;
> +
> +	if (!ch->phy->init_count)
> +		return -EIO;
> +
> +	if (!strncmp(buf, "host", strlen("host")))
> +		mode = PHY_MODE_USB_HOST;
> +	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> +		mode = PHY_MODE_USB_DEVICE;
> +	else
> +		return -EINVAL;
> +
> +	if (ch->has_otg)
> +		ret = rcar_gen3_otg_change_role(ch, mode);
> +
> +	if (ret < 0)
> +		return ret;
> +
>  	return count;
>  }
>  
> @@ -264,12 +281,17 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
>  			 char *buf)
>  {
>  	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> +	bool is_host;
>  
> -	if (!ch->has_otg || !ch->phy->init_count)
> +	if (!ch->phy->init_count)
> +		return -EIO;
> +
> +	if (ch->has_otg)
> +		is_host = rcar_gen3_is_host(ch);
> +	else
>  		return -EIO;
>  
> -	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
> -							    "peripheral");
> +	return sprintf(buf, "%s\n", is_host ? "host" : "peripheral");
>  }

I'm not sure why the above hunk is necessary.
They function seems logically the same before and after.

>  static DEVICE_ATTR_RW(role);
>  
> -- 
> 1.9.1
> 
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: [PATCH v4 2/4] phy: rcar-gen3-usb2: cleanup the role_{store,show}()

[Yoshihiro Shimoda]

Hi Simon-san,

Thank you for the comments!

> From: Simon Horman, Sent: Tuesday, October 10, 2017 4:46 PM
> 
> On Tue, Oct 10, 2017 at 02:58:28PM +0900, Yoshihiro Shimoda wrote:
> > This patch cleanups the role_{store,show}() and replaces the local
> > "bool" for host/device mode selection with the "enum phy_mode" in
> > the role_store().
> >
> > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> > ---
> >  drivers/phy/renesas/phy-rcar-gen3-usb2.c | 58 ++++++++++++++++++++++----------
> >  1 file changed, 40 insertions(+), 18 deletions(-)
> >
> > diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > index e00e99a..7619468 100644
> > --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > @@ -219,14 +219,10 @@ static bool rcar_gen3_is_host(struct rcar_gen3_chan *ch)
> >  	return !(readl(ch->base + USB2_COMMCTRL) & USB2_COMMCTRL_OTG_PERI);
> >  }
> >
> > -static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> > -			  const char *buf, size_t count)
> > +static ssize_t rcar_gen3_otg_change_role(struct rcar_gen3_chan *ch,
> > +					 enum phy_mode mode)
> >  {
> > -	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > -	bool is_b_device, is_host, new_mode_is_host;
> > -
> > -	if (!ch->has_otg || !ch->phy->init_count)
> > -		return -EIO;
> > +	bool is_b_device, is_host;
> >
> >  	/*
> >  	 * is_b_device: true is B-Device. false is A-Device.
> > @@ -234,18 +230,13 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> >  	 */
> >  	is_b_device = rcar_gen3_check_id(ch);
> >  	is_host = rcar_gen3_is_host(ch);
> > -	if (!strncmp(buf, "host", strlen("host")))
> > -		new_mode_is_host = true;
> > -	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> > -		new_mode_is_host = false;
> > -	else
> > -		return -EINVAL;
> >
> >  	/* If current and new mode is the same, this returns the error */
> > -	if (is_host == new_mode_is_host)
> > +	if ((is_host && mode == PHY_MODE_USB_HOST) ||
> > +	    (!is_host && mode == PHY_MODE_USB_DEVICE))
> 
> Maybe it is not worth the effort, but it seems that if
> rcar_gen3_is_host returned enum phy_mode then the above
> could be changed into a single comparison.

It's a nice idea! Since role_show() below should not be changed,
I will add rcar_gen3_is_phy_mode_host() and changed the above code
into a single comparison.

> >  		return -EINVAL;
> >
> > -	if (new_mode_is_host) {		/* And is_host must be false */
> > +	if (mode == PHY_MODE_USB_HOST) { /* And is_host must be false */
> >  		if (!is_b_device)	/* A-Peripheral */
> >  			rcar_gen3_init_from_a_peri_to_a_host(ch);
> >  		else			/* B-Peripheral */
> > @@ -257,6 +248,32 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> >  			rcar_gen3_init_for_peri(ch);
> >  	}
> >
> > +	return 0;
> > +}
> > +
> > +static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> > +			  const char *buf, size_t count)
> > +{
> > +	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > +	enum phy_mode mode;
> > +	ssize_t ret = -EIO;
> > +
> > +	if (!ch->phy->init_count)
> > +		return -EIO;
> > +
> > +	if (!strncmp(buf, "host", strlen("host")))
> > +		mode = PHY_MODE_USB_HOST;
> > +	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> > +		mode = PHY_MODE_USB_DEVICE;
> > +	else
> > +		return -EINVAL;
> > +
> > +	if (ch->has_otg)
> > +		ret = rcar_gen3_otg_change_role(ch, mode);
> > +
> > +	if (ret < 0)
> > +		return ret;
> > +
> >  	return count;
> >  }
> >
> > @@ -264,12 +281,17 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
> >  			 char *buf)
> >  {
> >  	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > +	bool is_host;
> >
> > -	if (!ch->has_otg || !ch->phy->init_count)
> > +	if (!ch->phy->init_count)
> > +		return -EIO;
> > +
> > +	if (ch->has_otg)
> > +		is_host = rcar_gen3_is_host(ch);
> > +	else
> >  		return -EIO;
> >
> > -	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
> > -							    "peripheral");
> > +	return sprintf(buf, "%s\n", is_host ? "host" : "peripheral");
> >  }
> 
> I'm not sure why the above hunk is necessary.
> They function seems logically the same before and after.

Indeed. So, I will keep rcar_gen3_is_host() as bool and
remove the role_show() modification.

Best regards,
Yoshihiro Shimoda

> >  static DEVICE_ATTR_RW(role);
> >
> > --
> > 1.9.1
> >

RE: [PATCH v4 2/4] phy: rcar-gen3-usb2: cleanup the role_{store,show}()

[Yoshihiro Shimoda]

Hi Simon-san,

Thank you for the comments!

> From: Simon Horman, Sent: Tuesday, October 10, 2017 4:46 PM
> 
> On Tue, Oct 10, 2017 at 02:58:28PM +0900, Yoshihiro Shimoda wrote:
> > This patch cleanups the role_{store,show}() and replaces the local
> > "bool" for host/device mode selection with the "enum phy_mode" in
> > the role_store().
> >
> > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh-zM6kxYcvzFBBDgjK7y7TUQ@public.gmane.org>
> > ---
> >  drivers/phy/renesas/phy-rcar-gen3-usb2.c | 58 ++++++++++++++++++++++----------
> >  1 file changed, 40 insertions(+), 18 deletions(-)
> >
> > diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > index e00e99a..7619468 100644
> > --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > @@ -219,14 +219,10 @@ static bool rcar_gen3_is_host(struct rcar_gen3_chan *ch)
> >  	return !(readl(ch->base + USB2_COMMCTRL) & USB2_COMMCTRL_OTG_PERI);
> >  }
> >
> > -static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> > -			  const char *buf, size_t count)
> > +static ssize_t rcar_gen3_otg_change_role(struct rcar_gen3_chan *ch,
> > +					 enum phy_mode mode)
> >  {
> > -	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > -	bool is_b_device, is_host, new_mode_is_host;
> > -
> > -	if (!ch->has_otg || !ch->phy->init_count)
> > -		return -EIO;
> > +	bool is_b_device, is_host;
> >
> >  	/*
> >  	 * is_b_device: true is B-Device. false is A-Device.
> > @@ -234,18 +230,13 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> >  	 */
> >  	is_b_device = rcar_gen3_check_id(ch);
> >  	is_host = rcar_gen3_is_host(ch);
> > -	if (!strncmp(buf, "host", strlen("host")))
> > -		new_mode_is_host = true;
> > -	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> > -		new_mode_is_host = false;
> > -	else
> > -		return -EINVAL;
> >
> >  	/* If current and new mode is the same, this returns the error */
> > -	if (is_host == new_mode_is_host)
> > +	if ((is_host && mode == PHY_MODE_USB_HOST) ||
> > +	    (!is_host && mode == PHY_MODE_USB_DEVICE))
> 
> Maybe it is not worth the effort, but it seems that if
> rcar_gen3_is_host returned enum phy_mode then the above
> could be changed into a single comparison.

It's a nice idea! Since role_show() below should not be changed,
I will add rcar_gen3_is_phy_mode_host() and changed the above code
into a single comparison.

> >  		return -EINVAL;
> >
> > -	if (new_mode_is_host) {		/* And is_host must be false */
> > +	if (mode == PHY_MODE_USB_HOST) { /* And is_host must be false */
> >  		if (!is_b_device)	/* A-Peripheral */
> >  			rcar_gen3_init_from_a_peri_to_a_host(ch);
> >  		else			/* B-Peripheral */
> > @@ -257,6 +248,32 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> >  			rcar_gen3_init_for_peri(ch);
> >  	}
> >
> > +	return 0;
> > +}
> > +
> > +static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> > +			  const char *buf, size_t count)
> > +{
> > +	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > +	enum phy_mode mode;
> > +	ssize_t ret = -EIO;
> > +
> > +	if (!ch->phy->init_count)
> > +		return -EIO;
> > +
> > +	if (!strncmp(buf, "host", strlen("host")))
> > +		mode = PHY_MODE_USB_HOST;
> > +	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> > +		mode = PHY_MODE_USB_DEVICE;
> > +	else
> > +		return -EINVAL;
> > +
> > +	if (ch->has_otg)
> > +		ret = rcar_gen3_otg_change_role(ch, mode);
> > +
> > +	if (ret < 0)
> > +		return ret;
> > +
> >  	return count;
> >  }
> >
> > @@ -264,12 +281,17 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
> >  			 char *buf)
> >  {
> >  	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > +	bool is_host;
> >
> > -	if (!ch->has_otg || !ch->phy->init_count)
> > +	if (!ch->phy->init_count)
> > +		return -EIO;
> > +
> > +	if (ch->has_otg)
> > +		is_host = rcar_gen3_is_host(ch);
> > +	else
> >  		return -EIO;
> >
> > -	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
> > -							    "peripheral");
> > +	return sprintf(buf, "%s\n", is_host ? "host" : "peripheral");
> >  }
> 
> I'm not sure why the above hunk is necessary.
> They function seems logically the same before and after.

Indeed. So, I will keep rcar_gen3_is_host() as bool and
remove the role_show() modification.

Best regards,
Yoshihiro Shimoda

> >  static DEVICE_ATTR_RW(role);
> >
> > --
> > 1.9.1
> >
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: [PATCH v4 2/4] phy: rcar-gen3-usb2: cleanup the role_{store,show}()

[Yoshihiro Shimoda]

Hi Simon-san,

Thank you for the comments!

> From: Simon Horman, Sent: Tuesday, October 10, 2017 4:46 PM
> 
> On Tue, Oct 10, 2017 at 02:58:28PM +0900, Yoshihiro Shimoda wrote:
> > This patch cleanups the role_{store,show}() and replaces the local
> > "bool" for host/device mode selection with the "enum phy_mode" in
> > the role_store().
> >
> > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> > ---
> >  drivers/phy/renesas/phy-rcar-gen3-usb2.c | 58 ++++++++++++++++++++++----------
> >  1 file changed, 40 insertions(+), 18 deletions(-)
> >
> > diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > index e00e99a..7619468 100644
> > --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c
> > @@ -219,14 +219,10 @@ static bool rcar_gen3_is_host(struct rcar_gen3_chan *ch)
> >  	return !(readl(ch->base + USB2_COMMCTRL) & USB2_COMMCTRL_OTG_PERI);
> >  }
> >
> > -static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> > -			  const char *buf, size_t count)
> > +static ssize_t rcar_gen3_otg_change_role(struct rcar_gen3_chan *ch,
> > +					 enum phy_mode mode)
> >  {
> > -	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > -	bool is_b_device, is_host, new_mode_is_host;
> > -
> > -	if (!ch->has_otg || !ch->phy->init_count)
> > -		return -EIO;
> > +	bool is_b_device, is_host;
> >
> >  	/*
> >  	 * is_b_device: true is B-Device. false is A-Device.
> > @@ -234,18 +230,13 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> >  	 */
> >  	is_b_device = rcar_gen3_check_id(ch);
> >  	is_host = rcar_gen3_is_host(ch);
> > -	if (!strncmp(buf, "host", strlen("host")))
> > -		new_mode_is_host = true;
> > -	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> > -		new_mode_is_host = false;
> > -	else
> > -		return -EINVAL;
> >
> >  	/* If current and new mode is the same, this returns the error */
> > -	if (is_host == new_mode_is_host)
> > +	if ((is_host && mode == PHY_MODE_USB_HOST) ||
> > +	    (!is_host && mode == PHY_MODE_USB_DEVICE))
> 
> Maybe it is not worth the effort, but it seems that if
> rcar_gen3_is_host returned enum phy_mode then the above
> could be changed into a single comparison.

It's a nice idea! Since role_show() below should not be changed,
I will add rcar_gen3_is_phy_mode_host() and changed the above code
into a single comparison.

> >  		return -EINVAL;
> >
> > -	if (new_mode_is_host) {		/* And is_host must be false */
> > +	if (mode == PHY_MODE_USB_HOST) { /* And is_host must be false */
> >  		if (!is_b_device)	/* A-Peripheral */
> >  			rcar_gen3_init_from_a_peri_to_a_host(ch);
> >  		else			/* B-Peripheral */
> > @@ -257,6 +248,32 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> >  			rcar_gen3_init_for_peri(ch);
> >  	}
> >
> > +	return 0;
> > +}
> > +
> > +static ssize_t role_store(struct device *dev, struct device_attribute *attr,
> > +			  const char *buf, size_t count)
> > +{
> > +	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > +	enum phy_mode mode;
> > +	ssize_t ret = -EIO;
> > +
> > +	if (!ch->phy->init_count)
> > +		return -EIO;
> > +
> > +	if (!strncmp(buf, "host", strlen("host")))
> > +		mode = PHY_MODE_USB_HOST;
> > +	else if (!strncmp(buf, "peripheral", strlen("peripheral")))
> > +		mode = PHY_MODE_USB_DEVICE;
> > +	else
> > +		return -EINVAL;
> > +
> > +	if (ch->has_otg)
> > +		ret = rcar_gen3_otg_change_role(ch, mode);
> > +
> > +	if (ret < 0)
> > +		return ret;
> > +
> >  	return count;
> >  }
> >
> > @@ -264,12 +281,17 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
> >  			 char *buf)
> >  {
> >  	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
> > +	bool is_host;
> >
> > -	if (!ch->has_otg || !ch->phy->init_count)
> > +	if (!ch->phy->init_count)
> > +		return -EIO;
> > +
> > +	if (ch->has_otg)
> > +		is_host = rcar_gen3_is_host(ch);
> > +	else
> >  		return -EIO;
> >
> > -	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
> > -							    "peripheral");
> > +	return sprintf(buf, "%s\n", is_host ? "host" : "peripheral");
> >  }
> 
> I'm not sure why the above hunk is necessary.
> They function seems logically the same before and after.

Indeed. So, I will keep rcar_gen3_is_host() as bool and
remove the role_show() modification.

Best regards,
Yoshihiro Shimoda

> >  static DEVICE_ATTR_RW(role);
> >
> > --
> > 1.9.1
> >