Hi!
It has been reported that RGMII is broken in fixed-link, and that is not surprising considering the fact that no PHY is attached to the MAC in that case, but a switch.
Okay, but there's something wrong in the code.
This brings us to the topic of the patch: the enetc driver should have not enabled the optional in-band status signaling for RGMII unconditionally, but should have forced the speed and duplex to what was resolved by phylink.
+++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -494,13 +494,20 @@ static void enetc_configure_port_mac(str static void enetc_mac_config(struct enetc_hw *hw, phy_interface_t phy_mode) {
...
- u32 val;
- if (phy_interface_mode_is_rgmii(phy_mode)) {
val = enetc_port_rd(hw, ENETC_PM0_IF_MODE);val &= ~ENETC_PM0_IFM_EN_AUTO;val &= ENETC_PM0_IFM_IFMODE_MASK;val |= ENETC_PM0_IFM_IFMODE_GMII | ENETC_PM0_IFM_RG;enetc_port_wr(hw, ENETC_PM0_IF_MODE, val);- }
val clears ENETC_PM0_IFM_EN_AUTO bit, then the bit is cleared again, preserving just IFMODE bits, then ors the IFMODE bits with new values.
I believe this is needed:
Signed-off-by: Pavel Machek (CIP) pavel@denx.de Pavel
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c index 83187cd59fdd..446ad4c43fab 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -499,7 +499,7 @@ static void enetc_mac_config(struct enetc_hw *hw, phy_interface_t phy_mode) if (phy_interface_mode_is_rgmii(phy_mode)) { val = enetc_port_rd(hw, ENETC_PM0_IF_MODE); val &= ~ENETC_PM0_IFM_EN_AUTO; - val &= ENETC_PM0_IFM_IFMODE_MASK; + val &= ~ENETC_PM0_IFM_IFMODE_MASK; val |= ENETC_PM0_IFM_IFMODE_GMII | ENETC_PM0_IFM_RG; enetc_port_wr(hw, ENETC_PM0_IF_MODE, val); }