Hi,
I'm using the RMII interface for the phy on a custom OMAP-L138 board. I'm using the files from the EVM but edited to our needs. In U-boot I get the following:
MMC: davinci: 0
MMC init failed
Using default environment
In: serial
Out: serial
Err: serial
ARM Clock : 300000000 Hz
DDR Clock : 75000000 Hz
SF: Unsupported manufacturer ff
Error - unable to probe SPI flash.
Invalid MAC address read.
Net: Ethernet PHY: GENERIC @ 0x00
DaVinci-EMAC
Then when linux boots, it says it can't connect to Phy 0..
PHY 0:00 not found
net eth0: could not connect to phy 0:00
ADDRCONF(NETDEV_UP): eth0: link is not ready
net eth0: DaVinci EMAC: ioctl not supported
eth0 no wireless extensions.
udhcpc (v1.13.2) started
Sending discover...
Sending discover...
Sending discover...
No lease, forking to background
done.
Setting up IP spoofing protection: rp_filter.
The Phy is a LAN8720 chip. I've enabled RMII in da850evm.h with:
#define CONFIG_DRIVER_TI_EMAC_USE_RMII
#undef CONFIG_MII
and edited da850evm.c with the following pinmix set-up:
static const struct pinmux_config emac_pins[] = {
{ pinmux(4), 8, 0 },
{ pinmux(4), 8, 1 },
{ pinmux(14), 8, 2 },
{ pinmux(14), 8, 3 },
{ pinmux(14), 8, 4 },
{ pinmux(14), 8, 5 },
{ pinmux(14), 8, 6 },
{ pinmux(14), 8, 7 },
{ pinmux(15), 8, 0 },
{ pinmux(15), 8, 1 }
};
I want the controller to source 50Mhz to the PHY..
I Noticed that there is a line in da850evm.h:
#define CONFIG_EMAC_MDIO_PHY_NUM 0
is that a correct value (0) ?
Does the above behaviour indicate an error with the PHY, or the configurations?
Any suggestions are appreciated. Btw, I've also edited the Kernel to support RMII.
-Koteich
