Ethernet problems on RCM3700
Started by ●February 25, 2005
Hi everyone!
I have a big problem with ethernet on my RCM3700 core module: even on
its dev board, I can compile everything, and all works fine. But when
I make an init, a sock_init for example, i.e. when I try to access the
ethenet port, the software is regularry downloaded in the rcm flash,
but when the sw goes in execution I have no power on the eth. status
led, and everything gets stucked, appearing the timeout message "while
debugging: timeout while waiting response from target".
It seems impossible to me that I've burned the ethernet and not the
rcm core!
All the other functions of the core module are working, is just the
ethernet that after 2 months of going well, it just stopped to work.
I have reinstalled everything (DYN C) on another PC, a new one, and
even using the original demo programs, the ethernet is dead
Thanks to everyone who can help me, I don't know what to do...
best regards,
Cirutech
Reply by ●February 25, 20052005-02-25
--- In rabbit-semi@rabb..., "cirutech" <cirutech@l...> wrote:
>
> Hi everyone!
> I have a big problem with ethernet on my RCM3700 core module: even
on
> its dev board, I can compile everything, and all works fine. But
when
> I make an init, a sock_init for example, i.e. when I try to access
the
> ethenet port, the software is regularry downloaded in the rcm flash,
> but when the sw goes in execution I have no power on the eth. status
> led, and everything gets stucked, appearing the timeout
message "while
> debugging: timeout while waiting response from target".
>
> It seems impossible to me that I've burned the ethernet and not the
> rcm core!
It's possible. Recently, I momentarily applied power backwards to a
RabbitNet board hooked to an SBC with an RCM3200. A spark and a puff
of smoke later, I could still program and run the SBC, but the RN1100
and the Ethernet on the RCM were out of commission.
Reply by ●February 25, 20052005-02-25
Hi,
I connected the dev board via its connector, and it is supposed to be
impossible to apply power backwards. And no puffs or bad smells, as I
hope!
Anyway thanks for answering, best regards,
C.
>
> It's possible. Recently, I momentarily applied power backwards to a
> RabbitNet board hooked to an SBC with an RCM3200. A spark and a puff
> of smoke later, I could still program and run the SBC, but the RN1100
> and the Ethernet on the RCM were out of commission.
Reply by ●February 25, 20052005-02-25
--- In rabbit-semi@rabb..., "cirutech" <cirutech@l...> wrote:
>
> Hi,
> I connected the dev board via its connector, and it is supposed to
be
> impossible to apply power backwards. And no puffs or bad smells, as
I
> hope!
>
> Anyway thanks for answering, best regards,
>
> C.
>
Try to check out board ID stored in flash memory ID block. If it got
corrupted then your compiler might be confused, thinking its another
board (i.e. one from TCP/IP dev.kit in most cases). Also, this may
be indicated by compiler warning when compiling code.
There are differences in how the Ethernet chip is connected to the
Rabbit on various board types. DC might have choosen wrong functions
to talk to it, so it seems like the CPU has no answer from LAN chip.
Well, I hope this one or any other less permanent "out of commission
state" would be your case.
J.
Reply by ●February 25, 20052005-02-25
On Feb 25, 2005, at 10:00 AM, cirutech wrote:
> I connected the dev board via its connector, and it is supposed to be
> impossible to apply power backwards. And no puffs or bad smells, as I
> hope!
I've fried a unit with static electricity before, so that's a
possibility to consider. If I leave my desk and come back, I always
ground myself before touching my development board or the cables
attached to it.
--
Tom Collins - tom@tom@...
Certified Rabbit Semiconductor Consultant based in Napa, California
Info on the Sniffter, a rechargeable, hand-held Network Tester
powered by the RCM2210: http://sniffter.com/
> I connected the dev board via its connector, and it is supposed to be
> impossible to apply power backwards. And no puffs or bad smells, as I
> hope!
I've fried a unit with static electricity before, so that's a
possibility to consider. If I leave my desk and come back, I always
ground myself before touching my development board or the cables
attached to it.
--
Tom Collins - tom@tom@...
Certified Rabbit Semiconductor Consultant based in Napa, California
Info on the Sniffter, a rechargeable, hand-held Network Tester
powered by the RCM2210: http://sniffter.com/
