diff mbox

[RESEND] usb: ehci/ohci-exynos: Fix PHY getting sequence

Message ID 1407235148-31790-1-git-send-email-gautam.vivek@samsung.com (mailing list archive)
State New, archived
Headers show

Commit Message

Vivek Gautam Aug. 5, 2014, 10:39 a.m. UTC
Since we want to keep support for both older usb-phys as well as the
newer generic phys, lets first get the generic PHYs and fallback to
older USB-PHYs only when we fail to get the former.
This should fix the issue with ehci-exynos and ohci-exynos, wherein
in the absence of SAMSUNG_USB2PHY config symbol, we end up getting
the NOP_USB_XCEIV phy when the same is enabled. And thus the PHYs
are not configured properly.

Reported-by: Sachin Kamat <sachin.kamat@samsung.com>
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Jingoo Han <jg1.han@samsung.com>
---

Based on 'usb-next' branch.
Resending it after adding 'Reported-by' tag.

 drivers/usb/host/ehci-exynos.c |   40 +++++++++++++++++-----------------
 drivers/usb/host/ohci-exynos.c |   47 +++++++++++++++++++---------------------
 2 files changed, 42 insertions(+), 45 deletions(-)

Comments

Jingoo Han Aug. 5, 2014, 10:58 a.m. UTC | #1
On Tuesday, August 05, 2014 7:39 PM, Vivek Gautam wrote:
> 
> Since we want to keep support for both older usb-phys as well as the
> newer generic phys, lets first get the generic PHYs and fallback to
> older USB-PHYs only when we fail to get the former.
> This should fix the issue with ehci-exynos and ohci-exynos, wherein
> in the absence of SAMSUNG_USB2PHY config symbol, we end up getting
> the NOP_USB_XCEIV phy when the same is enabled. And thus the PHYs
> are not configured properly.
> 
> Reported-by: Sachin Kamat <sachin.kamat@samsung.com>
> Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
> Cc: Alan Stern <stern@rowland.harvard.edu>
> Cc: Jingoo Han <jg1.han@samsung.com>

Acked-by: Jingoo Han <jg1.han@samsung.com>

Right, we would get the generic PHYs first, then get the older
usb-phys. Then, the older one will be removed from the kernel.
Thank you.

Best regards,
Jingoo Han

> ---
> 
> Based on 'usb-next' branch.
> Resending it after adding 'Reported-by' tag.
> 
>  drivers/usb/host/ehci-exynos.c |   40 +++++++++++++++++-----------------
>  drivers/usb/host/ohci-exynos.c |   47 +++++++++++++++++++---------------------
>  2 files changed, 42 insertions(+), 45 deletions(-)
> 
> diff --git a/drivers/usb/host/ehci-exynos.c b/drivers/usb/host/ehci-exynos.c
> index cda0a2f..2eed9a4 100644
> --- a/drivers/usb/host/ehci-exynos.c
> +++ b/drivers/usb/host/ehci-exynos.c
> @@ -62,18 +62,6 @@ static int exynos_ehci_get_phy(struct device *dev,
>  	int phy_number;
>  	int ret = 0;
> 
> -	exynos_ehci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
> -	if (IS_ERR(exynos_ehci->phy)) {
> -		ret = PTR_ERR(exynos_ehci->phy);
> -		if (ret != -ENXIO && ret != -ENODEV) {
> -			dev_err(dev, "no usb2 phy configured\n");
> -			return ret;
> -		}
> -		dev_dbg(dev, "Failed to get usb2 phy\n");
> -	} else {
> -		exynos_ehci->otg = exynos_ehci->phy->otg;
> -	}
> -
>  	for_each_available_child_of_node(dev->of_node, child) {
>  		ret = of_property_read_u32(child, "reg", &phy_number);
>  		if (ret) {
> @@ -90,15 +78,27 @@ static int exynos_ehci_get_phy(struct device *dev,
> 
>  		phy = devm_of_phy_get(dev, child, NULL);
>  		of_node_put(child);
> -		if (IS_ERR(phy)) {
> -			ret = PTR_ERR(phy);
> -			if (ret != -ENOSYS && ret != -ENODEV) {
> -				dev_err(dev, "no usb2 phy configured\n");
> -				return ret;
> -			}
> -			dev_dbg(dev, "Failed to get usb2 phy\n");
> -		}
> +		if (IS_ERR(phy))
> +			/* Lets fallback to older USB-PHYs */
> +			goto usb_phy_old;
>  		exynos_ehci->phy_g[phy_number] = phy;
> +		/* Make the older PHYs unavailable */
> +		exynos_ehci->phy = ERR_PTR(-ENXIO);
> +	}
> +
> +	return 0;
> +
> +usb_phy_old:
> +	exynos_ehci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
> +	if (IS_ERR(exynos_ehci->phy)) {
> +		ret = PTR_ERR(exynos_ehci->phy);
> +		if (ret != -ENXIO && ret != -ENODEV) {
> +			dev_err(dev, "no usb2 phy configured\n");
> +			return ret;
> +		}
> +		dev_dbg(dev, "Failed to get usb2 phy\n");
> +	} else {
> +		exynos_ehci->otg = exynos_ehci->phy->otg;
>  	}
> 
>  	return ret;
> diff --git a/drivers/usb/host/ohci-exynos.c b/drivers/usb/host/ohci-exynos.c
> index a72ab8f..7c48e3f 100644
> --- a/drivers/usb/host/ohci-exynos.c
> +++ b/drivers/usb/host/ohci-exynos.c
> @@ -51,27 +51,12 @@ static int exynos_ohci_get_phy(struct device *dev,
>  	int phy_number;
>  	int ret = 0;
> 
> -	exynos_ohci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
> -	if (IS_ERR(exynos_ohci->phy)) {
> -		ret = PTR_ERR(exynos_ohci->phy);
> -		if (ret != -ENXIO && ret != -ENODEV) {
> -			dev_err(dev, "no usb2 phy configured\n");
> -			return ret;
> -		}
> -		dev_dbg(dev, "Failed to get usb2 phy\n");
> -	} else {
> -		exynos_ohci->otg = exynos_ohci->phy->otg;
> -	}
> -
>  	/*
>  	 * Getting generic phy:
>  	 * We are keeping both types of phys as a part of transiting OHCI
>  	 * to generic phy framework, so as to maintain backward compatibilty
> -	 * with old DTB.
> -	 * If there are existing devices using DTB files built from them,
> -	 * to remove the support for old bindings in this driver,
> -	 * we need to make sure that such devices have their DTBs
> -	 * updated to ones built from new DTS.
> +	 * with old DTB too.
> +	 * We fallback to older USB-PHYs when we fail to get generic PHYs.
>  	 */
>  	for_each_available_child_of_node(dev->of_node, child) {
>  		ret = of_property_read_u32(child, "reg", &phy_number);
> @@ -89,15 +74,27 @@ static int exynos_ohci_get_phy(struct device *dev,
> 
>  		phy = devm_of_phy_get(dev, child, NULL);
>  		of_node_put(child);
> -		if (IS_ERR(phy)) {
> -			ret = PTR_ERR(phy);
> -			if (ret != -ENOSYS && ret != -ENODEV) {
> -				dev_err(dev, "no usb2 phy configured\n");
> -				return ret;
> -			}
> -			dev_dbg(dev, "Failed to get usb2 phy\n");
> -		}
> +		if (IS_ERR(phy))
> +			/* Lets fallback to older USB-PHYs */
> +			goto usb_phy_old;
>  		exynos_ohci->phy_g[phy_number] = phy;
> +		/* Make the older PHYs unavailable */
> +		exynos_ohci->phy = ERR_PTR(-ENXIO);
> +	}
> +
> +	return 0;
> +
> +usb_phy_old:
> +	exynos_ohci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
> +	if (IS_ERR(exynos_ohci->phy)) {
> +		ret = PTR_ERR(exynos_ohci->phy);
> +		if (ret != -ENXIO && ret != -ENODEV) {
> +			dev_err(dev, "no usb2 phy configured\n");
> +			return ret;
> +		}
> +		dev_dbg(dev, "Failed to get usb2 phy\n");
> +	} else {
> +		exynos_ohci->otg = exynos_ohci->phy->otg;
>  	}
> 
>  	return ret;
> --
> 1.7.10.4

--
To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Sachin Kamat Aug. 5, 2014, 11:35 a.m. UTC | #2
Hi Vivek,

On Tue, Aug 5, 2014 at 4:09 PM, Vivek Gautam <gautam.vivek@samsung.com> wrote:
> Since we want to keep support for both older usb-phys as well as the
> newer generic phys, lets first get the generic PHYs and fallback to
> older USB-PHYs only when we fail to get the former.
> This should fix the issue with ehci-exynos and ohci-exynos, wherein
> in the absence of SAMSUNG_USB2PHY config symbol, we end up getting
> the NOP_USB_XCEIV phy when the same is enabled. And thus the PHYs
> are not configured properly.
>
> Reported-by: Sachin Kamat <sachin.kamat@samsung.com>
> Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
> Cc: Alan Stern <stern@rowland.harvard.edu>
> Cc: Jingoo Han <jg1.han@samsung.com>

Tested this patch on Exynos5250 based Arndale and Snow boards.
Fixes the said issue. Thanks.

Tested-by: Sachin Kamat <sachin.kamat@samsung.com>
Vivek Gautam Aug. 5, 2014, 11:44 a.m. UTC | #3
Hi Jingoo,


On Tuesday, August 05, 2014 4:28 PM, Jingoo Han wrote:
> On Tuesday, August 05, 2014 7:39 PM, Vivek Gautam wrote:
>>
>> Since we want to keep support for both older usb-phys as well as the
>> newer generic phys, lets first get the generic PHYs and fallback to
>> older USB-PHYs only when we fail to get the former.
>> This should fix the issue with ehci-exynos and ohci-exynos, wherein
>> in the absence of SAMSUNG_USB2PHY config symbol, we end up getting
>> the NOP_USB_XCEIV phy when the same is enabled. And thus the PHYs
>> are not configured properly.
>>
>> Reported-by: Sachin Kamat <sachin.kamat@samsung.com>
>> Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
>> Cc: Alan Stern <stern@rowland.harvard.edu>
>> Cc: Jingoo Han <jg1.han@samsung.com>
>
> Acked-by: Jingoo Han <jg1.han@samsung.com>
>
> Right, we would get the generic PHYs first, then get the older
> usb-phys. Then, the older one will be removed from the kernel.

Right, I have a set of patches to remove the older phy support for Exynos 
SoC series,
keeping the support for S3C64XX, which I shall send soon.

[snip]

Thanks
Vivek Gautam 

--
To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Alan Stern Aug. 5, 2014, 2:11 p.m. UTC | #4
On Tue, 5 Aug 2014, Vivek Gautam wrote:

> Since we want to keep support for both older usb-phys as well as the
> newer generic phys, lets first get the generic PHYs and fallback to
> older USB-PHYs only when we fail to get the former.
> This should fix the issue with ehci-exynos and ohci-exynos, wherein
> in the absence of SAMSUNG_USB2PHY config symbol, we end up getting
> the NOP_USB_XCEIV phy when the same is enabled. And thus the PHYs
> are not configured properly.
> 
> Reported-by: Sachin Kamat <sachin.kamat@samsung.com>
> Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
> Cc: Alan Stern <stern@rowland.harvard.edu>
> Cc: Jingoo Han <jg1.han@samsung.com>
> ---
> 
> Based on 'usb-next' branch.
> Resending it after adding 'Reported-by' tag.
> 
>  drivers/usb/host/ehci-exynos.c |   40 +++++++++++++++++-----------------
>  drivers/usb/host/ohci-exynos.c |   47 +++++++++++++++++++---------------------
>  2 files changed, 42 insertions(+), 45 deletions(-)
> 
> diff --git a/drivers/usb/host/ehci-exynos.c b/drivers/usb/host/ehci-exynos.c
> index cda0a2f..2eed9a4 100644
> --- a/drivers/usb/host/ehci-exynos.c
> +++ b/drivers/usb/host/ehci-exynos.c
> @@ -62,18 +62,6 @@ static int exynos_ehci_get_phy(struct device *dev,
>  	int phy_number;
>  	int ret = 0;
>  
> -	exynos_ehci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
> -	if (IS_ERR(exynos_ehci->phy)) {
> -		ret = PTR_ERR(exynos_ehci->phy);
> -		if (ret != -ENXIO && ret != -ENODEV) {
> -			dev_err(dev, "no usb2 phy configured\n");
> -			return ret;
> -		}
> -		dev_dbg(dev, "Failed to get usb2 phy\n");
> -	} else {
> -		exynos_ehci->otg = exynos_ehci->phy->otg;
> -	}
> -
>  	for_each_available_child_of_node(dev->of_node, child) {
>  		ret = of_property_read_u32(child, "reg", &phy_number);
>  		if (ret) {
> @@ -90,15 +78,27 @@ static int exynos_ehci_get_phy(struct device *dev,

What about the code that you skipped here?  Suppose the 
of_property_read_32() call returns an error.  Do you then want to abort 
the whole thing, or should you go on to try the old usb_phy?

Alan Stern

--
To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Vivek Gautam Aug. 14, 2014, 2:29 p.m. UTC | #5
Hi Alan,


On Tue, Aug 5, 2014 at 7:41 PM, Alan Stern <stern@rowland.harvard.edu> wrote:
> On Tue, 5 Aug 2014, Vivek Gautam wrote:
>
>> Since we want to keep support for both older usb-phys as well as the
>> newer generic phys, lets first get the generic PHYs and fallback to
>> older USB-PHYs only when we fail to get the former.
>> This should fix the issue with ehci-exynos and ohci-exynos, wherein
>> in the absence of SAMSUNG_USB2PHY config symbol, we end up getting
>> the NOP_USB_XCEIV phy when the same is enabled. And thus the PHYs
>> are not configured properly.
>>
>> Reported-by: Sachin Kamat <sachin.kamat@samsung.com>
>> Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
>> Cc: Alan Stern <stern@rowland.harvard.edu>
>> Cc: Jingoo Han <jg1.han@samsung.com>
>> ---
>>
>> Based on 'usb-next' branch.
>> Resending it after adding 'Reported-by' tag.
>>
>>  drivers/usb/host/ehci-exynos.c |   40 +++++++++++++++++-----------------
>>  drivers/usb/host/ohci-exynos.c |   47 +++++++++++++++++++---------------------
>>  2 files changed, 42 insertions(+), 45 deletions(-)
>>
>> diff --git a/drivers/usb/host/ehci-exynos.c b/drivers/usb/host/ehci-exynos.c
>> index cda0a2f..2eed9a4 100644
>> --- a/drivers/usb/host/ehci-exynos.c
>> +++ b/drivers/usb/host/ehci-exynos.c
>> @@ -62,18 +62,6 @@ static int exynos_ehci_get_phy(struct device *dev,
>>       int phy_number;
>>       int ret = 0;
>>
>> -     exynos_ehci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
>> -     if (IS_ERR(exynos_ehci->phy)) {
>> -             ret = PTR_ERR(exynos_ehci->phy);
>> -             if (ret != -ENXIO && ret != -ENODEV) {
>> -                     dev_err(dev, "no usb2 phy configured\n");
>> -                     return ret;
>> -             }
>> -             dev_dbg(dev, "Failed to get usb2 phy\n");
>> -     } else {
>> -             exynos_ehci->otg = exynos_ehci->phy->otg;
>> -     }
>> -
>>       for_each_available_child_of_node(dev->of_node, child) {
>>               ret = of_property_read_u32(child, "reg", &phy_number);
>>               if (ret) {
>> @@ -90,15 +78,27 @@ static int exynos_ehci_get_phy(struct device *dev,
>
> What about the code that you skipped here?  Suppose the
> of_property_read_32() call returns an error.  Do you then want to abort
> the whole thing, or should you go on to try the old usb_phy?

I have posted a patch-series [1] which cleans up the entire samsug
usb2 phy from older usb-phy
layer, and further removes the support for these older usb-phy from
ehci-exynos and ohci-exynos.
Please let me know your comments.


[1]:  [PATCH 0/7] usb-phy: samsung: Cleanup the unused drivers
diff mbox

Patch

diff --git a/drivers/usb/host/ehci-exynos.c b/drivers/usb/host/ehci-exynos.c
index cda0a2f..2eed9a4 100644
--- a/drivers/usb/host/ehci-exynos.c
+++ b/drivers/usb/host/ehci-exynos.c
@@ -62,18 +62,6 @@  static int exynos_ehci_get_phy(struct device *dev,
 	int phy_number;
 	int ret = 0;
 
-	exynos_ehci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
-	if (IS_ERR(exynos_ehci->phy)) {
-		ret = PTR_ERR(exynos_ehci->phy);
-		if (ret != -ENXIO && ret != -ENODEV) {
-			dev_err(dev, "no usb2 phy configured\n");
-			return ret;
-		}
-		dev_dbg(dev, "Failed to get usb2 phy\n");
-	} else {
-		exynos_ehci->otg = exynos_ehci->phy->otg;
-	}
-
 	for_each_available_child_of_node(dev->of_node, child) {
 		ret = of_property_read_u32(child, "reg", &phy_number);
 		if (ret) {
@@ -90,15 +78,27 @@  static int exynos_ehci_get_phy(struct device *dev,
 
 		phy = devm_of_phy_get(dev, child, NULL);
 		of_node_put(child);
-		if (IS_ERR(phy)) {
-			ret = PTR_ERR(phy);
-			if (ret != -ENOSYS && ret != -ENODEV) {
-				dev_err(dev, "no usb2 phy configured\n");
-				return ret;
-			}
-			dev_dbg(dev, "Failed to get usb2 phy\n");
-		}
+		if (IS_ERR(phy))
+			/* Lets fallback to older USB-PHYs */
+			goto usb_phy_old;
 		exynos_ehci->phy_g[phy_number] = phy;
+		/* Make the older PHYs unavailable */
+		exynos_ehci->phy = ERR_PTR(-ENXIO);
+	}
+
+	return 0;
+
+usb_phy_old:
+	exynos_ehci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
+	if (IS_ERR(exynos_ehci->phy)) {
+		ret = PTR_ERR(exynos_ehci->phy);
+		if (ret != -ENXIO && ret != -ENODEV) {
+			dev_err(dev, "no usb2 phy configured\n");
+			return ret;
+		}
+		dev_dbg(dev, "Failed to get usb2 phy\n");
+	} else {
+		exynos_ehci->otg = exynos_ehci->phy->otg;
 	}
 
 	return ret;
diff --git a/drivers/usb/host/ohci-exynos.c b/drivers/usb/host/ohci-exynos.c
index a72ab8f..7c48e3f 100644
--- a/drivers/usb/host/ohci-exynos.c
+++ b/drivers/usb/host/ohci-exynos.c
@@ -51,27 +51,12 @@  static int exynos_ohci_get_phy(struct device *dev,
 	int phy_number;
 	int ret = 0;
 
-	exynos_ohci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
-	if (IS_ERR(exynos_ohci->phy)) {
-		ret = PTR_ERR(exynos_ohci->phy);
-		if (ret != -ENXIO && ret != -ENODEV) {
-			dev_err(dev, "no usb2 phy configured\n");
-			return ret;
-		}
-		dev_dbg(dev, "Failed to get usb2 phy\n");
-	} else {
-		exynos_ohci->otg = exynos_ohci->phy->otg;
-	}
-
 	/*
 	 * Getting generic phy:
 	 * We are keeping both types of phys as a part of transiting OHCI
 	 * to generic phy framework, so as to maintain backward compatibilty
-	 * with old DTB.
-	 * If there are existing devices using DTB files built from them,
-	 * to remove the support for old bindings in this driver,
-	 * we need to make sure that such devices have their DTBs
-	 * updated to ones built from new DTS.
+	 * with old DTB too.
+	 * We fallback to older USB-PHYs when we fail to get generic PHYs.
 	 */
 	for_each_available_child_of_node(dev->of_node, child) {
 		ret = of_property_read_u32(child, "reg", &phy_number);
@@ -89,15 +74,27 @@  static int exynos_ohci_get_phy(struct device *dev,
 
 		phy = devm_of_phy_get(dev, child, NULL);
 		of_node_put(child);
-		if (IS_ERR(phy)) {
-			ret = PTR_ERR(phy);
-			if (ret != -ENOSYS && ret != -ENODEV) {
-				dev_err(dev, "no usb2 phy configured\n");
-				return ret;
-			}
-			dev_dbg(dev, "Failed to get usb2 phy\n");
-		}
+		if (IS_ERR(phy))
+			/* Lets fallback to older USB-PHYs */
+			goto usb_phy_old;
 		exynos_ohci->phy_g[phy_number] = phy;
+		/* Make the older PHYs unavailable */
+		exynos_ohci->phy = ERR_PTR(-ENXIO);
+	}
+
+	return 0;
+
+usb_phy_old:
+	exynos_ohci->phy = devm_usb_get_phy(dev, USB_PHY_TYPE_USB2);
+	if (IS_ERR(exynos_ohci->phy)) {
+		ret = PTR_ERR(exynos_ohci->phy);
+		if (ret != -ENXIO && ret != -ENODEV) {
+			dev_err(dev, "no usb2 phy configured\n");
+			return ret;
+		}
+		dev_dbg(dev, "Failed to get usb2 phy\n");
+	} else {
+		exynos_ohci->otg = exynos_ohci->phy->otg;
 	}
 
 	return ret;