[U-Boot-Users] PQ2FADS-VR 2nd Ethernet

[Jihua Cheng]

Yuli,

Thanks so much for your help.
Let me explan the experiments which I did.

1) 172.18.100.54 is a redhat linux. Netmask is 255:255:255:0. I verified by use of ifconfig command.

1) Here is the environment variable for PQ2FADS-VR.
=> printenv
bootdelay=5
baudrate=38400
ethaddr=00:30:d4:00:01:02
g=bootm 100000
ipaddr=172.18.100.99
serverip=172.18.100.54
stdin=serial
stdout=serial
stderr=serial

2) I use arp command and found "172.18.100.99 ether 00:30:d4:00:01:02 C eth0" is in the ARP display. 
When I use ifconfig, I found RX packets and Tx packets increase the same number.

It seems Linux host received the ping from Pq2FADS-VR board and replied.

4) Why I suspect some hardware setting is probabily quite right is 
	I use same U-boot environment parameter and linux machine(172.18.100.54) for ethernet port1 with FCC2, the ping works fine.
	I only the u-boot to use FCC3 for ethernet port2, ping fails.
5) I check the RxBD status. I found the Empty bit is 1, means no data received in Rx.

Could I suspect something is wrong for FCC3 Rx portion? 
Could you please tell me what I must to check, or what I missed?

I greatly appreciate your help.

Reagrds,

Charles
-----Original Message-----
From: Yuli Barcohen [mailto:yuli at arabellasw.com]
Sent: Monday, July 19, 2004 10:55 AM
To: Jihua Cheng
Cc: u-boot-users at lists.sourceforge.net
Subject: RE: [U-Boot-Users] PQ2FADS-VR 2nd Ethernet


>>>>> Jihua Cheng writes:

    Jihua> Yuli, Thanks very much for your instruction.  It seems I move
    Jihua> further, however I still do not make Ethernet port2 works
    Jihua> fine. Following is the my modification and the symptom.

    Jihua> 1) I changed the BCSRs bit FETHIEN2 to 0, reset the Ethernet
    Jihua>    Port2 (Phy addr 3) to enable auto-negatiation. Now, the
    Jihua>    MII reg0 = 0x3100, reg1=0x782D.  IP Addr 172.18.100.54
    Jihua>    does exist. when I enable FCC2 and ping 172.18.100.54,
    Jihua>    ping success.  When I "ping 172.18.100.54", the message
    Jihua>    "fex: tx error" is gone. It seems TX is fine. Is it
    Jihua>    correct?

Probably it is. Try a sniffer (ethereal, tcpdump) on the host to see if
anything is transmitted by the board.

    Jihua> 2) However, "ping 172.18.100.54" fail. the message "ping
    Jihua>    failed; host 172.18.100.54" is not alive.  FCC3 Rx Clock
    Jihua>    is set to CLK15. PortB4~17, PortC16~17 are enabled.

Why do you think that the problem is still at the hardware level? What
are the board's IP address, MAC address, subnet mask, etc. (just type
printenv). You write that 172.18.100.54 exists but you don't write
anything about the host to which this address belongs. Is it a Linux
machine? Does it respond to pings at all? What's about subnets? There
are too many possible reasons for ping failure.

[...code deleted...]

-- 
========================================================================
 Yuli Barcohen       | Phone +972-9-765-1788 |  Software Project Leader
 yuli at arabellasw.com | Fax   +972-9-765-7494 | Arabella Software, Israel
========================================================================