| Message ID | 20200924083746.GA9410@laureti-dev |
|---|---|
| State | Accepted |
| Delegated to: | David Miller |
| Headers | show |
| Series | net: dsa: microchip: really look for phy-mode in port nodes | expand |
From: Helmut Grohne <helmut.grohne@intenta.de> Date: Thu, 24 Sep 2020 10:37:47 +0200 > The previous implementation failed to account for the "ports" node. The > actual port nodes are not child nodes of the switch node, but a "ports" > node sits in between. > > Fixes: edecfa98f602 ("net: dsa: microchip: look for phy-mode in port nodes") > Signed-off-by: Helmut Grohne <helmut.grohne@intenta.de> Applied and queued up for -stable. > I am very sorry that I need to send a fixup. It turned out that my > testing methodology was flawed. When I reintegrated Linus' master > branch, I noticed that it didn't work. You should be testing against the 'net' GIT tree, not Linus's master branch.
diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c index 8e755b50c9c1..c796d42730ba 100644 --- a/drivers/net/dsa/microchip/ksz_common.c +++ b/drivers/net/dsa/microchip/ksz_common.c @@ -387,8 +387,8 @@ EXPORT_SYMBOL(ksz_switch_alloc); int ksz_switch_register(struct ksz_device *dev, const struct ksz_dev_ops *ops) { + struct device_node *port, *ports; phy_interface_t interface; - struct device_node *port; unsigned int port_num; int ret; @@ -429,13 +429,17 @@ int ksz_switch_register(struct ksz_device *dev, ret = of_get_phy_mode(dev->dev->of_node, &interface); if (ret == 0) dev->compat_interface = interface; - for_each_available_child_of_node(dev->dev->of_node, port) { - if (of_property_read_u32(port, "reg", &port_num)) - continue; - if (port_num >= dev->port_cnt) - return -EINVAL; - of_get_phy_mode(port, &dev->ports[port_num].interface); - } + ports = of_get_child_by_name(dev->dev->of_node, "ports"); + if (ports) + for_each_available_child_of_node(ports, port) { + if (of_property_read_u32(port, "reg", + &port_num)) + continue; + if (port_num >= dev->port_cnt) + return -EINVAL; + of_get_phy_mode(port, + &dev->ports[port_num].interface); + } dev->synclko_125 = of_property_read_bool(dev->dev->of_node, "microchip,synclko-125"); }
The previous implementation failed to account for the "ports" node. The actual port nodes are not child nodes of the switch node, but a "ports" node sits in between. Fixes: edecfa98f602 ("net: dsa: microchip: look for phy-mode in port nodes") Signed-off-by: Helmut Grohne <helmut.grohne@intenta.de> --- drivers/net/dsa/microchip/ksz_common.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) I am very sorry that I need to send a fixup. It turned out that my testing methodology was flawed. When I reintegrated Linus' master branch, I noticed that it didn't work. It turned out that our hardware now correctly implements hardware reset. As a consequence, the correct setting of the phy-mode now became essential to operating the device. I'm also looking forward to see "net: dsa: microchip: Improve phy mode message" (from net-next) being merged. That would have helped me spot this earlier. Helmut