| Message ID | afba85f5-987b-4449-83cc-350438af7fe7@gmail.com (mailing list archive) |
|---|---|
| State | New |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [v2,net] r8169: remove redundant hwmon support | expand |
On 1/9/2025 2:43 PM, Heiner Kallweit wrote: > The temperature sensor is actually part of the integrated PHY and available > also on the standalone versions of the PHY. Therefore hwmon support will > be added to the Realtek PHY driver and can be removed here. > > Fixes: 1ffcc8d41306 ("r8169: add support for the temperature sensor being available from RTL8125B") > Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> > --- > v2: > - rebase patch on net In the future, please avoid re-sending within 24 hours: https://docs.kernel.org/process/maintainer-netdev.html#resending-after-review Thanks, Jake
On 1/9/2025 2:43 PM, Heiner Kallweit wrote: > The temperature sensor is actually part of the integrated PHY and available > also on the standalone versions of the PHY. Therefore hwmon support will > be added to the Realtek PHY driver and can be removed here. > > Fixes: 1ffcc8d41306 ("r8169: add support for the temperature sensor being available from RTL8125B") > Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> > --- Can you explain what user-facing issues this fixes? Do we get conflicts when multiple hwmon sensors are registered? I'm not sure this counts as a 'net' fix, unless we can identify the user-facing behavior that is problematic?
On 10.01.2025 02:14, Jacob Keller wrote: > > > On 1/9/2025 2:43 PM, Heiner Kallweit wrote: >> The temperature sensor is actually part of the integrated PHY and available >> also on the standalone versions of the PHY. Therefore hwmon support will >> be added to the Realtek PHY driver and can be removed here. >> >> Fixes: 1ffcc8d41306 ("r8169: add support for the temperature sensor being available from RTL8125B") >> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> >> --- > > Can you explain what user-facing issues this fixes? Do we get conflicts > when multiple hwmon sensors are registered? I'm not sure this counts as > a 'net' fix, unless we can identify the user-facing behavior that is > problematic? The same sensor would be exposed twice, by the MAC and the PHY driver. Support in the MAC driver was added for 6.13, so it's still in the rc version only and not yet part of a released kernel version.
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 739707a7b..8a3959bb2 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -16,7 +16,6 @@ #include <linux/clk.h> #include <linux/delay.h> #include <linux/ethtool.h> -#include <linux/hwmon.h> #include <linux/phy.h> #include <linux/if_vlan.h> #include <linux/in.h> @@ -5347,43 +5346,6 @@ static bool rtl_aspm_is_safe(struct rtl8169_private *tp) return false; } -static umode_t r8169_hwmon_is_visible(const void *drvdata, - enum hwmon_sensor_types type, - u32 attr, int channel) -{ - return 0444; -} - -static int r8169_hwmon_read(struct device *dev, enum hwmon_sensor_types type, - u32 attr, int channel, long *val) -{ - struct rtl8169_private *tp = dev_get_drvdata(dev); - int val_raw; - - val_raw = phy_read_paged(tp->phydev, 0xbd8, 0x12) & 0x3ff; - if (val_raw >= 512) - val_raw -= 1024; - - *val = 1000 * val_raw / 2; - - return 0; -} - -static const struct hwmon_ops r8169_hwmon_ops = { - .is_visible = r8169_hwmon_is_visible, - .read = r8169_hwmon_read, -}; - -static const struct hwmon_channel_info * const r8169_hwmon_info[] = { - HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT), - NULL -}; - -static const struct hwmon_chip_info r8169_hwmon_chip_info = { - .ops = &r8169_hwmon_ops, - .info = r8169_hwmon_info, -}; - static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct rtl8169_private *tp; @@ -5563,12 +5525,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) return rc; - /* The temperature sensor is available from RTl8125B */ - if (IS_REACHABLE(CONFIG_HWMON) && tp->mac_version >= RTL_GIGA_MAC_VER_63) - /* ignore errors */ - devm_hwmon_device_register_with_info(&pdev->dev, "nic_temp", tp, - &r8169_hwmon_chip_info, - NULL); rc = register_netdev(dev); if (rc) return rc;
The temperature sensor is actually part of the integrated PHY and available also on the standalone versions of the PHY. Therefore hwmon support will be added to the Realtek PHY driver and can be removed here. Fixes: 1ffcc8d41306 ("r8169: add support for the temperature sensor being available from RTL8125B") Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> --- v2: - rebase patch on net --- drivers/net/ethernet/realtek/r8169_main.c | 44 ----------------------- 1 file changed, 44 deletions(-)