Hi,
On 22-03-16 08:51, Michael Haas wrote:
The RTL8211C found on the A20-OlinuXino-Lime2 does not word in slave
mode. This patch disables master/slave mode autonegotiation and forces
master mode.
The RTL8211C identifies itself as RTL8211B via its UID. This patch uses
the revision number taken from the PHYID2 register to distinguish the
two. The NetBSD driver uses the same approach.
CC: fra...@gmail.com
CC: mer...@debian.org
CC: hdego...@redhat.com
CC: i...@hellion.org.uk
CC: joe.hershber...@ni.com
Signed-off-by: Michael Haas <h...@computerlinguist.org>
---
drivers/net/phy/realtek.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index 259a87f..97d5712 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -54,6 +54,19 @@ static int rtl8211x_config(struct phy_device *phydev)
phy_write(phydev, MDIO_DEVAD_NONE, MIIM_RTL8211x_PHY_INER,
MIIM_RTL8211x_PHY_INTR_DIS);
+ int reg = phy_read(phydev, MDIO_DEVAD_NONE, MII_PHYSID1);
+ int rev = reg & 0xf;
+ if (rev == 3 && phydev->phy_id == 0x1cc912) {
+ /* RTL8211C and RTL8211C are distinguished only by
+ their revision number */
RTL8211C and RTL8211C -> RTL8211B and RTL8211C ?
Also I would put this comment above the:
int reg = phy_read(phydev, MDIO_DEVAD_NONE, MII_PHYSID1);
Line, and put a comment like this above the if :
/* The RTL8211C is unreliable in slave mode, force master mode */
Code wise everything looks good :)
+ reg = phy_read(phydev, MDIO_DEVAD_NONE, MII_CTRL1000);
+ /* force manual master/slave configuration */
+ reg |= (1 << 12);
+ /* force master mode */
+ reg | = (1 << 11);
+ phy_write(phydev, MDIO_DEVAD_NONE, MII_CTRL1000, reg);
+ }
+
/* read interrupt status just to clear it */
phy_read(phydev, MDIO_DEVAD_NONE, MIIM_RTL8211x_PHY_INER);
Regards,
Hans
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot
