net: cpsw and mdio-gpio fixes for v3.15-final

These patches against v3.15-rc4 fix a few issues in the cpsw and mdio-gpio drivers.

Johan

Johan Hovold (5):
net: mdio-gpio: fix device-tree binding documentation
net: mdio-gpio: warn about missing bus alias id
Revert "net: eth: cpsw: Correctly attach to GPIO bitbang MDIO driver"
net: cpsw: fix null dereference at probe
net: cpsw: add missing of_node_put

 Documentation/devicetree/bindings/net/mdio-gpio.txt |  2 +-
 drivers/net/ethernet/ti/cpsw.c                      | 17 ++++++-----------
 drivers/net/phy/mdio-gpio.c                         |  4 ++++
 3 files changed, 11 insertions(+), 12 deletions(-)

1.8.3.2

Fix aliases syntax in device-tree binding example to avoid copy-paste errors (the alias would be dropped silently).

---
 Documentation/devicetree/bindings/net/mdio-gpio.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index c79bab025369..8dbcf8295c6c 100644
@@ -14,7 +14,7 @@ node.
 Example:
 
 aliases {
-	mdio-gpio0 = <&mdio0>;
+	mdio-gpio0 = &mdio0;
 };
 
 mdio0: mdio {

1.8.3.2

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

Use a sane default bus id (rather than -ENODEV) and print a warning when the bus alias id is missing.

---
 drivers/net/phy/mdio-gpio.c | 4 ++++
 1 file changed, 4 insertions(+)

index 9c4defdec67b..5f1a2250018f 100644
@@ -215,6 +215,10 @@ static int mdio_gpio_probe(struct platform_device *pdev)
 	if (pdev->dev.of_node) {
 		pdata = mdio_gpio_of_get_data(pdev);
 		bus_id = of_alias_get_id(pdev->dev.of_node, "mdio-gpio");
+		if (bus_id < 0) {
+			dev_warn(&pdev->dev, "failed to get alias id\n");
+			bus_id = 0;
+		}
 	} else {
 		pdata = dev_get_platdata(&pdev->dev);
 		bus_id = pdev->id;

1.8.3.2

Add missing of_node_put to avoid kref leak.

---
 drivers/net/ethernet/ti/cpsw.c | 1 +
 1 file changed, 1 insertion(+)

index db9360cd861c..c331b7ebc812 100644
@@ -1871,6 +1871,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 		mdio_node = of_find_node_by_phandle(be32_to_cpup(parp));
 		phyid = be32_to_cpup(parp+1);
 		mdio = of_find_device_by_node(mdio_node);
+		of_node_put(mdio_node);
 		if (!mdio) {
 			pr_err("Missing mdio platform device\n");
 			return -EINVAL;

1.8.3.2

These patches against v3.15-rc4 fix a few issues in the cpsw and mdio-gpio drivers.

Resend with proper stable CC (git send-email still fails to parse the #-part of the tag unless address is enclosed in <> it seems).

Sorry about the noise.

Johan

Johan Hovold (5):
net: mdio-gpio: fix device-tree binding documentation
net: mdio-gpio: warn about missing bus alias id
Revert "net: eth: cpsw: Correctly attach to GPIO bitbang MDIO driver"
net: cpsw: fix null dereference at probe
net: cpsw: add missing of_node_put

 Documentation/devicetree/bindings/net/mdio-gpio.txt |  2 +-
 drivers/net/ethernet/ti/cpsw.c                      | 17 ++++++-----------
 drivers/net/phy/mdio-gpio.c                         |  4 ++++
 3 files changed, 11 insertions(+), 12 deletions(-)

1.8.3.2

Fix aliases syntax in device-tree binding example to avoid copy-paste errors (the alias would be dropped silently).

---
 Documentation/devicetree/bindings/net/mdio-gpio.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index c79bab025369..8dbcf8295c6c 100644
@@ -14,7 +14,7 @@ node.
 Example:
 
 aliases {
-	mdio-gpio0 = <&mdio0>;
+	mdio-gpio0 = &mdio0;
 };
 
 mdio0: mdio {

1.8.3.2

Use a sane default bus id (rather than -ENODEV) and print a warning when the bus alias id is missing.

---
 drivers/net/phy/mdio-gpio.c | 4 ++++
 1 file changed, 4 insertions(+)

index 9c4defdec67b..5f1a2250018f 100644
@@ -215,6 +215,10 @@ static int mdio_gpio_probe(struct platform_device *pdev)
 	if (pdev->dev.of_node) {
 		pdata = mdio_gpio_of_get_data(pdev);
 		bus_id = of_alias_get_id(pdev->dev.of_node, "mdio-gpio");
+		if (bus_id < 0) {
+			dev_warn(&pdev->dev, "failed to get alias id\n");
+			bus_id = 0;
+		}
 	} else {
 		pdata = dev_get_platdata(&pdev->dev);
 		bus_id = pdev->id;

1.8.3.2

This reverts commit f8d56d8f892be43a2404356073e16401eb5a42e6.

Fix potential null-pointer dereference at probe if the mdio-gpio device has not been successfully probed yet.

The offending commit is plain wrong for a number of reasons. First of all it accesses internal driver data of an unrelated device. Neither does it check that the data is non-null (which it is in case the device has not been probed yet).

Furthermore, the decision on whether to treat any driver data according to the mdio-gpio driver's internals is made based on the node name. But the name is not compared against "mdio" which is the normal name for the node, but rather against "gpio" which the node does not have to be named (and shouldn't be according to the binding documentation). [ If this hack is to be kept out-of-tree it should at least be matching against the compatible property. ]

---
 drivers/net/ethernet/ti/cpsw.c | 14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

index 36aa109416c4..18d83d8d7f74 100644
@@ -1871,18 +1871,8 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 		mdio_node = of_find_node_by_phandle(be32_to_cpup(parp));
 		phyid = be32_to_cpup(parp+1);
 		mdio = of_find_device_by_node(mdio_node);
-
-		if (strncmp(mdio->name, "gpio", 4) == 0) {
-			/* GPIO bitbang MDIO driver attached */
-			struct mii_bus *bus = dev_get_drvdata(&mdio->dev);
-
-			snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
-				 PHY_ID_FMT, bus->id, phyid);
-		} else {
-			/* davinci MDIO driver attached */
-			snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
-				 PHY_ID_FMT, mdio->name, phyid);
-		}
+		snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
+			 PHY_ID_FMT, mdio->name, phyid);
 
 		mac_addr = of_get_mac_address(slave_node);
 		if (mac_addr)

1.8.3.2

Fix null-pointer dereference at probe when the mdio platform device is missing (e.g. when it has been disabled in DT).

---
 drivers/net/ethernet/ti/cpsw.c | 4 ++++
 1 file changed, 4 insertions(+)

index 18d83d8d7f74..db9360cd861c 100644
@@ -1871,6 +1871,10 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 		mdio_node = of_find_node_by_phandle(be32_to_cpup(parp));
 		phyid = be32_to_cpup(parp+1);
 		mdio = of_find_device_by_node(mdio_node);
+		if (!mdio) {
+			pr_err("Missing mdio platform device\n");
+			return -EINVAL;
+		}
 		snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
 			 PHY_ID_FMT, mdio->name, phyid);
 

1.8.3.2

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

Add missing of_node_put to avoid kref leak.

---
 drivers/net/ethernet/ti/cpsw.c | 1 +
 1 file changed, 1 insertion(+)

index db9360cd861c..c331b7ebc812 100644
@@ -1871,6 +1871,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
 		mdio_node = of_find_node_by_phandle(be32_to_cpup(parp));
 		phyid = be32_to_cpup(parp+1);
 		mdio = of_find_device_by_node(mdio_node);
+		of_node_put(mdio_node);
 		if (!mdio) {
 			pr_err("Missing mdio platform device\n");
 			return -EINVAL;

1.8.3.2

From: Johan Hovold <[email protected]>
Date: Thu, 8 May 2014 10:09:19 +0200

Series applied, thanks.