| Author |
Message |
ricky
Guest
|
 Posted:
Mon Oct 11, 2004 10:30 pm Post subject:
X's during simulation |
|
|
Hi,
I have a problem, I got X's in the simulation results. Could anybody explain
me the ways to trace the origin of X's. I am using NCsim.
Your help is very much appreciated.
Thanks in advance.
|
|
| Back to top |
|
 |
Kai Harrekilde-Petersen
Guest
|
| Posted:
Tue Oct 12, 2004 1:55 am Post subject:
Re: X's during simulation |
|
|
"ricky" <ricky_100@yahoo.com> writes:
| Quote: | Hi,
I have a problem, I got X's in the simulation results. Could anybody explain
me the ways to trace the origin of X's. I am using NCsim.
Your help is very much appreciated.
|
Check the output of your rams; it is usually 'X' after reset. Search
the netlist from the point where you have the 'X' and work backwards
looking at all inputs.
Kai
--
Kai Harrekilde-Petersen <khp(at)harrekilde(dot)dk> |
|
| Back to top |
|
|
Narendran Kumaraguru Nath
Guest
|
| Posted:
Tue Oct 12, 2004 10:13 am Post subject:
Re: X's during simulation |
|
|
ricky wrote:
| Quote: | Hi,
I have a problem, I got X's in the simulation results. Could anybody explain
me the ways to trace the origin of X's. I am using NCsim.
Your help is very much appreciated.
Thanks in advance.
|
Hi ricky,
The general procedure is to find wether things after reset happen as
it is expected to. for example let's assume you are simulating a processor.
1. first you have to check wether you've the processor reset properly
2. then wether the processor starts fetching from the start address
3. if yes wether is it getting proper instruction
4. if yes wether it is behaving as it is supposed to ...
generally see wether some important signals like reset, test/functional
mode select input .... pins are in proper values...
If the system goes to 'x' after some time of the startup, then look at
the first error message in your transcript and start working from there...
another cause may be your PLL. It should be fed proper clock to get back
the multiplied clock. Also its inputs should be in proper values...
Check your crystal oscillator too...
HTH
Narendran Kumaragruru Nathan
TooMuch Semiconductor Solutions
www.toomuchsemi.com
A Bangalore based startup specialising on services in EDA &
Verification.
|
|
| Back to top |
|
|
Guest
|
| Posted:
Wed Oct 13, 2004 1:17 am Post subject:
Re: X's during simulation |
|
|
Hi Ricky,
There is no need to get annoyed with X's showing up in the simulation,
since it is the "model" that is responsible, rather than the simulator
(except the few cases mentioned below in the paragraph).
An "X" is just a way, by means of which a model communicates its behavior
in the same way it does with 0's and 1's. In general X's are used to
describe unknown state of the circuit(which could either be a "0" or "1"
in a digital logic) at that particular point in time. The "X" originates
most of the time from the model to reflect the intended behavior of the
circuit. An "X" could also be generated by the simulator as the result of
a system task that captures a timing violation. It is also possible that
you are having multiple drivers on the same net without a resolution function
defined, which could make the simulator generate an "X" instead.
Tracing the origin of "X" is similar to tracing an intended output
as a result of the relevant inputs, except that in this case, the intended
output is an "X". As long as you know basic circuit theory and "some"
understanding of how the HDL's work, it should not be a difficult task.
It might also benefit, if you could understand the model a bit to verify
how the intended behavior is implemented(and also to validate the correctness
of the model).
Regards
Raman
"ricky" <ricky_100@yahoo.com> wrote in message news:<ckejoq$84r$1@reader10.wxs.nl>...
| Quote: | Hi,
I have a problem, I got X's in the simulation results. Could anybody explain
me the ways to trace the origin of X's. I am using NCsim.
Your help is very much appreciated.
Thanks in advance. |
|
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in