net: fix phydev reference leaks

This series fixes a number of phydev reference leaks (and one of_node leak) due to failure to put the reference taken by of_phy_find_device().

Note that I did not try to fix drivers/net/phy/xilinx_gmii2rgmii.c which still leaks a reference.

Against net but should apply just as fine to net-next.

Thanks,
Johan

v2:
- use put_device() instead of phy_dev_free() to put the references taken in net/dsa (patch 1/4).
- add four new patches fixing similar leaks

Johan Hovold (5):
net: dsa: fix fixed-link-phy device leaks
net: bcmgenet: fix phydev reference leak
net: fsl/fman: fix phydev reference leak
net: fsl/fman: fix fixed-link-phydev reference leak
net: qcom/emac: fix of_node and phydev leaks

 drivers/net/ethernet/broadcom/genet/bcmmii.c     | 4 +++-
 drivers/net/ethernet/freescale/fman/fman_memac.c | 3 +++
 drivers/net/ethernet/freescale/fman/mac.c        | 2 ++
 drivers/net/ethernet/qualcomm/emac/emac-phy.c    | 1 +
 drivers/net/ethernet/qualcomm/emac/emac.c        | 4 ++++
 net/dsa/dsa.c                                    | 5 ++++-
 6 files changed, 17 insertions(+), 2 deletions(-)

2.7.3

Make sure to drop the reference taken by of_phy_find_device() when registering and deregistering the fixed-link PHY-device.

speeds/duplex")

---
 net/dsa/dsa.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

index a6902c1e2f28..cb0091b99592 100644
@@ -233,6 +233,8 @@ int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev,
 		genphy_read_status(phydev);
 		if (ds->ops->adjust_link)
 			ds->ops->adjust_link(ds, port, phydev);
+
+		put_device(&phydev->mdio.dev);
 	}
 
 	return 0;

@@ -509,8 +511,9 @@ void dsa_cpu_dsa_destroy(struct device_node *port_dn)
 	if (of_phy_is_fixed_link(port_dn)) {
 		phydev = of_phy_find_device(port_dn);
 		if (phydev) {
-			phy_device_free(phydev);
 			fixed_phy_unregister(phydev);
+			put_device(&phydev->mdio.dev);
+			phy_device_free(phydev);
 		}
 	}
 }

2.7.3

Make sure to drop the reference taken by of_phy_find_device() when initialising MOCA PHYs.

all MoCA PHYs")

---
 drivers/net/ethernet/broadcom/genet/bcmmii.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

index 457c3bc8cfff..2e745bd51df4 100644
@@ -542,8 +542,10 @@ static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv)
 	/* Make sure we initialize MoCA PHYs with a link down */
 	if (phy_mode == PHY_INTERFACE_MODE_MOCA) {
 		phydev = of_phy_find_device(dn);
-		if (phydev)
+		if (phydev) {
 			phydev->link = 0;
+			put_device(&phydev->mdio.dev);
+		}
 	}
 
 	return 0;

2.7.3

Make sure to drop the reference taken by of_phy_find_device() during initialisation when later freeing the struct fman_mac.

---
 drivers/net/ethernet/freescale/fman/fman_memac.c | 3 +++
 1 file changed, 3 insertions(+)

index 53ef51e3bd9e..71a5ded9d1de 100644
@@ -1107,6 +1107,9 @@ int memac_free(struct fman_mac *memac)
 {
 	free_init_resources(memac);
 
+	if (memac->pcsphy)
+		put_device(&memac->pcsphy->mdio.dev);
+
 	kfree(memac->memac_drv_param);
 	kfree(memac);
 

2.7.3

Make sure to drop the reference taken by of_phy_find_device() when looking up a fixed-link phydev during probe.

---
 drivers/net/ethernet/freescale/fman/mac.c | 2 ++
 1 file changed, 2 insertions(+)

index 8fe6b3e253fa..736db9d9b0ad 100644
@@ -892,6 +892,8 @@ static int mac_probe(struct platform_device *_of_dev)
 		priv->fixed_link->duplex = phy->duplex;
 		priv->fixed_link->pause = phy->pause;
 		priv->fixed_link->asym_pause = phy->asym_pause;
+
+		put_device(&phy->mdio.dev);
 	}
 
 	err = mac_dev->init(mac_dev);

2.7.3

Thank you,
Madalin

Make sure to drop the reference taken by of_phy_find_device() during probe on probe errors and on driver unbind.

Also drop the of_node reference taken by of_parse_phandle() in the same path.

---
 drivers/net/ethernet/qualcomm/emac/emac-phy.c | 1 +
 drivers/net/ethernet/qualcomm/emac/emac.c     | 4 ++++
 2 files changed, 5 insertions(+)

index da4e90db4d98..99a14df28b96 100644
@@ -212,6 +212,7 @@ int emac_phy_config(struct platform_device *pdev, struct emac_adapter *adpt)
 
 		phy_np = of_parse_phandle(np, "phy-handle", 0);
 		adpt->phydev = of_phy_find_device(phy_np);
+		of_node_put(phy_np);
 	}
 
 	if (!adpt->phydev) {

index 4fede4b86538..57b35aeac51a 100644
@@ -711,6 +711,8 @@ static int emac_probe(struct platform_device *pdev)
 err_undo_napi:
 	netif_napi_del(&adpt->rx_q.napi);
 err_undo_mdiobus:
+	if (!has_acpi_companion(&pdev->dev))
+		put_device(&adpt->phydev->mdio.dev);
 	mdiobus_unregister(adpt->mii_bus);
 err_undo_clocks:
 	emac_clks_teardown(adpt);

@@ -730,6 +732,8 @@ static int emac_remove(struct platform_device *pdev)
 
 	emac_clks_teardown(adpt);
 
+	if (!has_acpi_companion(&pdev->dev))
+		put_device(&adpt->phydev->mdio.dev);
 	mdiobus_unregister(adpt->mii_bus);
 	free_netdev(netdev);
 

2.7.3

When I first wrote this code, I had a lot of trouble getting the PHY to re-probe when the driver was unloaded and reloaded, and it was because of some odd problems with get/put of of_nodes.

Since I know that you didn't test this driver on real hardware, I have to test these patches myself before I can ACK them.

Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the
Code Aurora Forum, hosted by The Linux Foundation.

Yeah, that's better.

Thanks,
Johan

From: Johan Hovold <[email protected]>
Date: Thu, 24 Nov 2016 19:21:26 +0100

...

Series applied, thanks.

I was really hoping you'd give me the chance to test the patches before applying them.

Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the
Code Aurora Forum, hosted by The Linux Foundation.

From: Timur Tabi <[email protected]>
Date: Sun, 27 Nov 2016 20:11:17 -0600

Sorry, if anything is broken I will happily revert if it isn't fixed promptly.