phy: for 4.4 -rc
Subsystems (5)
Hi Greg,
Looks like this have not been merged yet. Can you merge it in this -rc cycle. Let me know if I have to change something.
Bulk of the changes are with respect to adding of_node_put in the PHY drivers, a fix in PHY core and a patch to fix randconfig error.
Thanks
Kishon
The following changes since commit 8005c49d9aea74d382f474ce11afbbc7d7130bec:
Linux 4.4-rc1 (2015-11-15 17:00:27 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/kishon/linux-phy.git tags/phy-for-4.4-rc
for you to fetch changes up to 708744628ba96ed4dfcac74a985eb66ad551f164:
phy: core: Get a refcount to phy in devm_of_phy_get_by_index() (2015-12-07 18:44:02 +0530)
----------------------------------------------------------------
phy: for 4.4 -rc
*) Add missing of_node_put in a bunch of PHY drivers *) Add get_device in devm_of_phy_get_by_index() *) Fix randconfig build error in sun9i usb driver
----------------------------------------------------------------
Arnd Bergmann (1):
phy: sun9i-usb: add USB dependency
Chunfeng Yun (1):
phy: core: Get a refcount to phy in devm_of_phy_get_by_index()
Julia Lawall (7):
phy: brcmstb-sata: add missing of_node_put
phy: mt65xx-usb3: add missing of_node_put
phy: berlin-sata: add missing of_node_put
phy: rockchip-usb: add missing of_node_put
phy: miphy28lp: add missing of_node_put
phy: miphy365x: add missing of_node_put
phy: cygnus: pcie: add missing of_node_put
drivers/phy/Kconfig | 1 + drivers/phy/phy-bcm-cygnus-pcie.c | 16 ++++++++++++---- drivers/phy/phy-berlin-sata.c | 20 ++++++++++++++------ drivers/phy/phy-brcmstb-sata.c | 17 ++++++++++++----- drivers/phy/phy-core.c | 21 +++++++++++++++------ drivers/phy/phy-miphy28lp.c | 16 +++++++++++----- drivers/phy/phy-miphy365x.c | 16 +++++++++++----- drivers/phy/phy-mt65xx-usb3.c | 20 +++++++++++++------- drivers/phy/phy-rockchip-usb.c | 17 ++++++++++++----- 9 files changed, 101 insertions(+), 43 deletions(-)
Signature · 1 line
1.7.9.5
PATCH 1/9 phy: sun9i-usb: add USB dependency
From: Arnd Bergmann <[email protected]>
The sun9i usb phy driver calls of_usb_get_phy_mode(), which is not available if USB is disabled:
drivers/built-in.o: In function `sun9i_usb_phy_probe':
:(.text+0x7fb0): undefined reference to `of_usb_get_phy_mode'
This adds a dependency to avoid the randconfig build errors.
--- drivers/phy/Kconfig | 1 + 1 file changed, 1 insertion(+)
drivers/phy/Kconfig +1 -0
@@ -233,6 +233,7 @@ config PHY_SUN9I_USB tristate "Allwinner sun9i SoC USB PHY driver" depends on ARCH_SUNXI && HAS_IOMEM && OF depends on RESET_CONTROLLER + depends on USB_COMMON select GENERIC_PHY help Enable this to support the transceiver that is part of Allwinner
Signature · 1 line
1.7.9.5
PATCH 2/9 phy: brcmstb-sata: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_available_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_available_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-brcmstb-sata.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
drivers/phy/phy-brcmstb-sata.c +12 -5
@@ -140,7 +140,7 @@ static int brcm_sata_phy_probe(struct platform_device *pdev)struct brcm_sata_phy *priv;struct resource *res;struct phy_provider *provider; -int count =0 ; +int ret, count =0 ;if (of_get_child_count (dn) ==0 )return -ENODEV ;
@@ -163,16 +163,19 @@ static int brcm_sata_phy_probe(struct platform_device *pdev)if (of_property_read_u32 (child,"reg" , &id)) {dev_err (dev,"missing reg property in node %s\n" , child->name); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; }if (id >=MAX_PORTS ) {dev_err (dev,"invalid reg: %u\n" , id); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; }if (priv->phys[id].phy) {dev_err (dev,"already registered port %u\n" , id); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; } port = &priv->phys[id];
@@ -182,7 +185,8 @@ static int brcm_sata_phy_probe(struct platform_device *pdev) port->ssc_en =of_property_read_bool (child,"brcm,enable-ssc" );if (IS_ERR (port->phy)) {dev_err (dev,"failed to create PHY\n" ); -return PTR_ERR (port->phy); + ret =PTR_ERR (port->phy); + goto put_child; }phy_set_drvdata (port->phy, port);
@@ -198,6 +202,9 @@ static int brcm_sata_phy_probe(struct platform_device *pdev)dev_info (dev,"registered %d port(s)\n" , count);return 0 ; +put_child: +of_node_put (child); +return ret; }static struct platform_driver brcm_sata_phy_driver = {
Signature · 1 line
1.7.9.5
PATCH 3/9 phy: mt65xx-usb3: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-mt65xx-usb3.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-)
drivers/phy/phy-mt65xx-usb3.c +13 -7
@@ -415,7 +415,7 @@ static int mt65xx_u3phy_probe(struct platform_device *pdev)struct resource *sif_res;struct mt65xx_u3phy *u3phy;struct resource res; -int port; +int port, retval; u3phy =devm_kzalloc (dev,sizeof (*u3phy),GFP_KERNEL );if (!u3phy)
@@ -447,31 +447,34 @@ static int mt65xx_u3phy_probe(struct platform_device *pdev)for_each_child_of_node (np, child_np) {struct mt65xx_phy_instance *instance;struct phy *phy; -int retval; instance =devm_kzalloc (dev,sizeof (*instance),GFP_KERNEL ); -if (!instance) -return -ENOMEM ; +if (!instance) { + retval = -ENOMEM ; + goto put_child; + } u3phy->phys[port] = instance; phy =devm_phy_create (dev, child_np, &mt65xx_u3phy_ops);if (IS_ERR (phy)) {dev_err (dev,"failed to create phy\n" ); -return PTR_ERR (phy); + retval =PTR_ERR (phy); + goto put_child; } retval =of_address_to_resource (child_np,0 , &res);if (retval) {dev_err (dev,"failed to get address resource(id-%d)\n" , port); -return retval; + goto put_child; } instance->port_base =devm_ioremap_resource (&phy->dev, &res);if (IS_ERR (instance->port_base)) {dev_err (dev,"failed to remap phy regs\n" ); -return PTR_ERR (instance->port_base); + retval =PTR_ERR (instance->port_base); + goto put_child; } instance->phy = phy;
@@ -483,6 +486,9 @@ static int mt65xx_u3phy_probe(struct platform_device *pdev) provider =devm_of_phy_provider_register (dev, mt65xx_phy_xlate);return PTR_ERR_OR_ZERO (provider); +put_child: +of_node_put (child_np); +return retval; }static const struct of_device_id mt65xx_u3phy_id_table[] = {
Signature · 1 line
1.7.9.5
PATCH 4/9 phy: berlin-sata: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_available_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_available_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-berlin-sata.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-)
drivers/phy/phy-berlin-sata.c +14 -6
@@ -195,7 +195,7 @@ static int phy_berlin_sata_probe(struct platform_device *pdev)struct phy_provider *phy_provider;struct phy_berlin_priv *priv;struct resource *res; -int i =0 ; +int ret, i =0 ;u32 phy_id; priv =devm_kzalloc (dev,sizeof (*priv),GFP_KERNEL );
@@ -237,22 +237,27 @@ static int phy_berlin_sata_probe(struct platform_device *pdev)if (of_property_read_u32 (child,"reg" , &phy_id)) {dev_err (dev,"missing reg property in node %s\n" , child->name); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; }if (phy_id >=ARRAY_SIZE (phy_berlin_power_down_bits)) {dev_err (dev,"invalid reg in node %s\n" , child->name); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; } phy_desc =devm_kzalloc (dev,sizeof (*phy_desc),GFP_KERNEL ); -if (!phy_desc) -return -ENOMEM ; +if (!phy_desc) { + ret = -ENOMEM ; + goto put_child; + } phy =devm_phy_create (dev,NULL , &phy_berlin_sata_ops);if (IS_ERR (phy)) {dev_err (dev,"failed to create PHY %d\n" , phy_id); -return PTR_ERR (phy); + ret =PTR_ERR (phy); + goto put_child; } phy_desc->phy = phy;
@@ -269,6 +274,9 @@ static int phy_berlin_sata_probe(struct platform_device *pdev) phy_provider =devm_of_phy_provider_register (dev, phy_berlin_sata_phy_xlate);return PTR_ERR_OR_ZERO (phy_provider); +put_child: +of_node_put (child); +return ret; }static const struct of_device_id phy_berlin_sata_of_match[] = {
Signature · 1 line
1.7.9.5
PATCH 5/9 phy: rockchip-usb: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_available_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_available_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-rockchip-usb.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
drivers/phy/phy-rockchip-usb.c +12 -5
@@ -108,13 +108,16 @@ static int rockchip_usb_phy_probe(struct platform_device *pdev)for_each_available_child_of_node (dev->of_node, child) { rk_phy =devm_kzalloc (dev,sizeof (*rk_phy),GFP_KERNEL ); -if (!rk_phy) -return -ENOMEM ; +if (!rk_phy) { + err = -ENOMEM ; + goto put_child; + }if (of_property_read_u32 (child,"reg" , ®_offset)) {dev_err (dev,"missing reg property in node %s\n" , child->name); -return -EINVAL ; + err = -EINVAL ; + goto put_child; } rk_phy->reg_offset = reg_offset;
@@ -127,18 +130,22 @@ static int rockchip_usb_phy_probe(struct platform_device *pdev) rk_phy->phy =devm_phy_create (dev, child, &ops);if (IS_ERR (rk_phy->phy)) {dev_err (dev,"failed to create PHY\n" ); -return PTR_ERR (rk_phy->phy); + err =PTR_ERR (rk_phy->phy); + goto put_child; }phy_set_drvdata (rk_phy->phy, rk_phy);/* only power up usb phy when it use, so disable it when init*/ err =rockchip_usb_phy_power (rk_phy,1 );if (err) -return err; + goto put_child; } phy_provider =devm_of_phy_provider_register (dev, of_phy_simple_xlate);return PTR_ERR_OR_ZERO (phy_provider); +put_child: +of_node_put (child); +return err; }static const struct of_device_id rockchip_usb_phy_dt_ids[] = {
Signature · 1 line
1.7.9.5
PATCH 6/9 phy: miphy28lp: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-miphy28lp.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-)
drivers/phy/phy-miphy28lp.c +11 -5
@@ -1226,15 +1226,18 @@ static int miphy28lp_probe(struct platform_device *pdev) miphy_phy =devm_kzalloc (&pdev->dev,sizeof (*miphy_phy),GFP_KERNEL ); -if (!miphy_phy) -return -ENOMEM ; +if (!miphy_phy) { + ret = -ENOMEM ; + goto put_child; + } miphy_dev->phys[port] = miphy_phy; phy =devm_phy_create (&pdev->dev, child, &miphy28lp_ops);if (IS_ERR (phy)) {dev_err (&pdev->dev,"failed to create PHY\n" ); -return PTR_ERR (phy); + ret =PTR_ERR (phy); + goto put_child; } miphy_dev->phys[port]->phy = phy;
@@ -1242,11 +1245,11 @@ static int miphy28lp_probe(struct platform_device *pdev) ret =miphy28lp_of_probe (child, miphy_phy);if (ret) -return ret; + goto put_child; ret =miphy28lp_probe_resets (child, miphy_dev->phys[port]);if (ret) -return ret; + goto put_child;phy_set_drvdata (phy, miphy_dev->phys[port]); port++;
@@ -1255,6 +1258,9 @@ static int miphy28lp_probe(struct platform_device *pdev) provider =devm_of_phy_provider_register (&pdev->dev, miphy28lp_xlate);return PTR_ERR_OR_ZERO (provider); +put_child: +of_node_put (child); +return ret; }static const struct of_device_id miphy28lp_of_match[] = {
Signature · 1 line
1.7.9.5
PATCH 7/9 phy: miphy365x: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-miphy365x.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-)
drivers/phy/phy-miphy365x.c +11 -5
@@ -566,22 +566,25 @@ static int miphy365x_probe(struct platform_device *pdev) miphy_phy =devm_kzalloc (&pdev->dev,sizeof (*miphy_phy),GFP_KERNEL ); -if (!miphy_phy) -return -ENOMEM ; +if (!miphy_phy) { + ret = -ENOMEM ; + goto put_child; + } miphy_dev->phys[port] = miphy_phy; phy =devm_phy_create (&pdev->dev, child, &miphy365x_ops);if (IS_ERR (phy)) {dev_err (&pdev->dev,"failed to create PHY\n" ); -return PTR_ERR (phy); + ret =PTR_ERR (phy); + goto put_child; } miphy_dev->phys[port]->phy = phy; ret =miphy365x_of_probe (child, miphy_phy);if (ret) -return ret; + goto put_child;phy_set_drvdata (phy, miphy_dev->phys[port]);
@@ -591,12 +594,15 @@ static int miphy365x_probe(struct platform_device *pdev) &miphy_phy->ctrlreg);if (ret) {dev_err (&pdev->dev,"No sysconfig offset found\n" ); -return ret; + goto put_child; } } provider =devm_of_phy_provider_register (&pdev->dev, miphy365x_xlate);return PTR_ERR_OR_ZERO (provider); +put_child: +of_node_put (child); +return ret; }static const struct of_device_id miphy365x_of_match[] = {
Signature · 1 line
1.7.9.5
PATCH 8/9 phy: cygnus: pcie: add missing of_node_put
In reply to Kishon Vijay Abraham IFrom: Julia Lawall <[email protected]>
for_each_available_child_of_node performs an of_node_get on each iteration, so a return from the middle of the loop requires an of_node_put.
A simplified version of the semantic patch that finds this problem is as follows (http://coccinelle.lip6.fr):
// <smpl>
@@
expression root,e;
local idexpression child;
@@
for_each_available_child_of_node(root, child) {
... when != of_node_put(child)
when != e = child
(
return child;
|
* return ...;
)
...
}
// </smpl>
--- drivers/phy/phy-bcm-cygnus-pcie.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)
drivers/phy/phy-bcm-cygnus-pcie.c +12 -4
@@ -128,6 +128,7 @@ static int cygnus_pcie_phy_probe(struct platform_device *pdev)struct phy_provider *provider;struct resource *res;unsigned cnt =0 ; +int ret;if (of_get_child_count (node) ==0 ) {dev_err (dev,"PHY no child node\n" );
@@ -154,24 +155,28 @@ static int cygnus_pcie_phy_probe(struct platform_device *pdev)if (of_property_read_u32 (child,"reg" , &id)) {dev_err (dev,"missing reg property for %s\n" , child->name); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; }if (id >=MAX_NUM_PHYS ) {dev_err (dev,"invalid PHY id: %u\n" , id); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; }if (core->phys[id].phy) {dev_err (dev,"duplicated PHY id: %u\n" , id); -return -EINVAL ; + ret = -EINVAL ; + goto put_child; } p = &core->phys[id]; p->phy =devm_phy_create (dev, child, &cygnus_pcie_phy_ops);if (IS_ERR (p->phy)) {dev_err (dev,"failed to create PHY\n" ); -return PTR_ERR (p->phy); + ret =PTR_ERR (p->phy); + goto put_child; } p->core = core;
@@ -191,6 +196,9 @@ static int cygnus_pcie_phy_probe(struct platform_device *pdev)dev_dbg (dev,"registered %u PCIe PHY(s)\n" , cnt);return 0 ; +put_child: +of_node_put (child); +return ret; }static const struct of_device_id cygnus_pcie_phy_match_table[] = {
Signature · 1 line
1.7.9.5
PATCH 9/9 phy: core: Get a refcount to phy in devm_of_phy_get_by_index()
In reply to Kishon Vijay Abraham IFrom: Chunfeng Yun <[email protected]>
On driver detach, devm_phy_release() will put a refcount to the phy, so gets a refconut to it before return.
--- drivers/phy/phy-core.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-)
drivers/phy/phy-core.c +15 -6
@@ -636,8 +636,9 @@ EXPORT_SYMBOL_GPL(devm_of_phy_get);* @np: node containing the phy * @index: index of the phy * -* Gets the phy using _of_phy_get(), and associates a device with it using -* devres. On driver detach, release function is invoked on the devres data, +* Gets the phy using _of_phy_get(), then gets a refcount to it, +* and associates a device with it using devres. On driver detach, +* release function is invoked on the devres data, * then, devres data is freed. * */
@@ -651,13 +652,21 @@ struct phy *devm_of_phy_get_by_index(struct device *dev, struct device_node *np,return ERR_PTR (-ENOMEM ); phy =_of_phy_get (np, index); -if (!IS_ERR (phy)) { - *ptr = phy; -devres_add (dev, ptr); - }else { +if (IS_ERR (phy)) {devres_free (ptr); +return phy; } +if (!try_module_get (phy->ops->owner)) { +devres_free (ptr); +return ERR_PTR (-EPROBE_DEFER ); + } + +get_device (&phy->dev); + + *ptr = phy; +devres_add (dev, ptr); +return phy; }EXPORT_SYMBOL_GPL (devm_of_phy_get_by_index);
Signature · 1 line
1.7.9.5
Quoted 11 lines · Kishon Vijay Abraham I
Hi Greg,
Looks like this have not been merged yet. Can you merge it in this -rc cycle. Let me know if I have to change something.
Bulk of the changes are with respect to adding
of_node_putin the PHY drivers, a fix in PHY core and a patch to fix randconfig error.Thanks
Kishon
Quoted 7 lines
The following changes since commit
8005c49d9aea74d382f474ce11afbbc7d7130bec:Linux 4.4-rc1 (2015-11-15 17:00:27 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/kishon/linux-phy.gittags/phy-for-4.4-rc
Sorry for the delay, now pulled and pushed out.
thanks,
greg k-h
Quoted 14 lines · Greg KH
Quoted 13 lines · Kishon Vijay Abraham I
Hi Greg,
Looks like this have not been merged yet. Can you merge it in this -rc cycle. Let me know if I have to change something.
Bulk of the changes are with respect to adding
of_node_putin the PHY drivers, a fix in PHY core and a patch to fix randconfig error.Thanks
KishonThe following changes since commit
8005c49d9aea74d382f474ce11afbbc7d7130bec:
Quoted 7 lines
Quoted 5 lines
Linux 4.4-rc1 (2015-11-15 17:00:27 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/kishon/linux-phy.gittags/phy-for-4.4-rcSorry for the delay, now pulled and pushed out.
Thank you for taking this Greg.
-Kishon