Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Iteration limit error on ModelSim

903 views
Skip to first unread message

Yongduck Seo

unread,
Nov 2, 2001, 9:35:02 AM11/2/01
to
Hi,

I got the message "Iteration limit reached. Possible zero
delay oscillation. See the manual." after simulation of
results of gate level synthesis by synopsis without any
sdf file. But that message did not come out when
I tried RTL level funtional simulation, of course, using
the same ModelSim. And I cannot find where that happen
in my code befer or after the systhesis.

What's the problem?
Does somebody have any idea?

Andy Peters

unread,
Nov 2, 2001, 11:32:50 AM11/2/01
to
You've got a race somewhere.

Why aren't you using the post-simulation SDF file? You may be getting
the races because you don't have the delays.

--a

Andrew Brown

unread,
Nov 2, 2001, 2:50:56 PM11/2/01
to
Or you could just have a very long delta - increase the limit above 5000 (i
think thats the default) and see if that helps.
It's unlikely you have a race which would generally mean a combinational
loop of some kind - this should produce a warning when synthesising (is the
netlist vhdl?).

A.

"Andy Peters" <an...@exponentmedia.deletethis.com> wrote in message
news:3BE2CAAC...@exponentmedia.deletethis.com...

Srinivasan Venkataramanan

unread,
Nov 6, 2001, 3:22:58 AM11/6/01
to
Hi,
You may want to try with +delay_mode_unit option, that may solve
the problem temporarily before you get your SDF.

Good Luck,
Srinivasan

--
Srinivasan Venkataramanan
ASIC Design Engineer
Software & Silicon Systems India Pvt. Ltd. (An Intel company)
Bangalore, India, Visit: http://www.simputer.org)


"Yongduck Seo" <yd...@pixelplus.co.kr> wrote in message
news:5829290d.01110...@posting.google.com...

Mike Treseler

unread,
Nov 6, 2001, 12:19:52 PM11/6/01
to
Yongduck Seo wrote:
>
> Hi,
>
> I got the message "Iteration limit reached. Possible zero
> delay oscillation"

Sounds like you've enabled an
asychronous oscillator.
Why not bring up the gui,
view source, trace the code
and see what's happening?

Might have to add a register or two.


--Mike Treseler

Narendran Kumaraguru Nathan

unread,
Nov 20, 2001, 4:53:34 AM11/20/01
to
Man,
I seriously doubt a non-exit loop. Please check you code if any two
process trigger each other and keep on running
without incrementing of the simulation time. Please check if the
simulation time advances.
~Naren.
0 new messages