Messages in this thread |  | | From | Rosen Penev <> | | Date | Mon, 27 Apr 2026 12:18:52 -0700 | | Subject | Re: [PATCH net-next] net: enc28j60: use of_get_ethdev_address |
| |
On Mon, Apr 27, 2026 at 7:19 AM Andrew Lunn <andrew@lunn.ch> wrote: > > On Sun, Apr 26, 2026 at 10:25:33PM -0700, Rosen Penev wrote: > > Since this is an OF only driver, of_ instead of device_ allows nvmem to > > be used to specify the MAC address. > > Now that is usual, somebody converting device_ to of_. It is normally > the other way around, for not particularly good reasons. Nice. > > > - if (device_get_ethdev_address(&spi->dev, dev)) > > + ret = of_get_ethdev_address(spi->dev.of_node, dev); > > + if (ret == -EPROBE_DEFER) > > + return ret; > > /** > * device_get_mac_address - Get the MAC for a given device > * @dev: Pointer to the device > * @addr: Address of buffer to store the MAC in > */ > int device_get_mac_address(struct device *dev, char *addr) > { > if (!fwnode_get_mac_address(dev_fwnode(dev), addr)) > return 0; > > return nvmem_get_mac_address(dev, addr); > } > EXPORT_SYMBOL(device_get_mac_address); > > This appears to support nvmem. Would device_get_mac_address() actually > work? I'm not against swapping to of_get_ethdev_address, but the > commit message should be accurate. I seem to have missed https://lore.kernel.org/all/20250912140332.35395-3-wahrenst@gmx.net/ > > Andrew
|  |