[U-Boot-Users] Intel LXT972A on FADS 852T
This is my second series of post on this subject. Current situation is that when I ping from uboot, now I do not get TX timeout, but get host is dead message.
On tracing, I can tell that uboot successfully return from fec_send (it does get ok that packet was transmitted.) by reading TX descriptor.
In fec_recv, it reads receive descriptor, and see no packet there.
I have done of the obvious and non-obvious checks. e.g. ethernet cable and stuff outside the box is fully operational. Infect I see arp packet going out on ethereal. I know that reply is sent by the pinged box.
I have read data sheet section of fec, and as far as I can tell, all the initialization is correct. I can talk to physical chip LXT972A, e.g. I can turn LEDs off and can set loopback mode, so that I am really talking to the chip.
My hardware guy has checked all the clocks and he assures me that they are all in order. He also made external loopback, i.e. ethernet cable with input going to output, this also works.
So what would be the cause of LXT972A not receiving anything. On LXT972A, I have checked configuration and status registers. On status register it says that nothing has been received.
In message 20040726063133.98061.qmail@web50505.mail.yahoo.com you wrote:
This is my second series of post on this subject. Current situation is that when I ping from uboot, now I do not get TX timeout, but get host is dead message.
Please use shorter lines (max. 70 characters or so).
What do you mean you get a "host is dead" message? U-Boot does not print such a message.
I have done of the obvious and non-obvious checks. e.g. ethernet cable and stuff outside the box is fully operational. Infect I see arp packet going out on ethereal. I know that reply is sent by the pinged box.
Do you also see the ICMP packet sent by U-Boot on the wire?
My hardware guy has checked all the clocks and he assures me that they are all in order. He also made external loopback, i.e. ethernet cable with input going to output, this also works.
How did you test this that you know it works?
--0-783479016-1090823493=:97116 Content-Type: text/html; charset=us-ascii
<DIV>
And please stop sending HTML!!!
Best regards,
Wolfgang Denk
--- Wolfgang Denk wd@denx.de wrote:
In message 20040726063133.98061.qmail@web50505.mail.yahoo.com you wrote:
This is my second series of post on this subject. Current
situation is that when I ping from uboot, now I do not get TX timeout, but get host is dead message.
Please use shorter lines (max. 70 characters or so).
What do you mean you get a "host is dead" message? U-Boot does not print such a message.
Actual message is host <ip address> is not alive
I have done of the obvious and non-obvious checks. e.g. ethernet
cable and stuff outside the box is fully operational. Infect I see arp packet going out on ethereal. I know that reply is sent by the pinged box.
Do you also see the ICMP packet sent by U-Boot on the wire?
No
My hardware guy has checked all the clocks and he assures me that
they are all in order. He also made external loopback, i.e. ethernet cable with input going to output, this also works.
How did you test this that you know it works?
Well, initially, when I was getting TX timeout, I stop getting timeout message with lopback cable.
--0-783479016-1090823493=:97116 Content-Type: text/html; charset=us-ascii
<DIV>
And please stop sending HTML!!!
Best regards,
Wolfgang Denk
-- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd@denx.de The manager will be continually amazed that policies he took for com- mon knowledge are totally unknown by some member of his team. Since his fundamental job is to keep everybody going in the same direction, his chief daily task will be communication, not decision-making. - Fred Brooks, "The Mythical Man Month"
This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_idG21&alloc_id%10040&op=click _______________________________________________ U-Boot-Users mailing list U-Boot-Users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/u-boot-users
In message 20040726165905.67524.qmail@web50501.mail.yahoo.com you wrote:
situation is that when I ping from uboot, now I do not get TX timeout, but get host is dead message.
...
Actual message is host <ip address> is not alive
Then please write EXACTLY this the next time you report a problem.
I have done of the obvious and non-obvious checks. e.g. ethernet
cable and stuff outside the box is fully operational. Infect I see arp packet going out on ethereal. I know that reply is sent by the pinged box.
How do you know that? Did you see the packet on the wire?
Did you see how it entered the Rx code in U-boot?
Do you also see the ICMP packet sent by U-Boot on the wire?
No
Did you trace that the Tx code was actually executed?
My hardware guy has checked all the clocks and he assures me that
they are all in order. He also made external loopback, i.e. ethernet cable with input going to output, this also works.
How did you test this that you know it works?
Well, initially, when I was getting TX timeout, I stop getting timeout message with lopback cable.
Arghh... And what exactly do you know from this test?
Best regards,
Wolfgang Denk
--- Wolfgang Denk wd@denx.de wrote:
I have done of the obvious and non-obvious checks. e.g. ethernet cable and stuff outside the box is fully operational. Infect I see arp packet going out on ethereal. I know that reply is sent by the pinged box.
How do you know that? Did you see the packet on the wire?
I have seen out going packet from PPC boc (can tell by mac address). Also seen reply from the host which is pinged
Did you see how it entered the Rx code in U-boot?
Yes, it enters RX code, and receive control (r_cntrl) register says there is nothing and it leaves big for (;;) loop. fec_recv is called about 20 - 25 times.
Do you also see the ICMP packet sent by U-Boot on the wire?
No
Did you trace that the Tx code was actually executed?
Yes, I did. After it receive ack that packet was actually sent, it returns and goes back to calls fec_recv
Wolfgang Denk
Abu Sana
In message 20040726215657.38202.qmail@web50509.mail.yahoo.com you wrote:
Yes, it enters RX code, and receive control (r_cntrl) register says there is nothing and it leaves big for (;;) loop. fec_recv is called
Stop here. You write: "receive control (r_cntrl) register says there is nothing".
Find out why this is the case, and fix it.
Best regards,
Wolfgang Denk
participants (2)
-
Knowledge Seeker -
Wolfgang Denk