Subject: Re: OVL suggestions (win* endings)
From: Adam Krolnik (krolnik@lsil.com)
Date: Mon May 06 2002 - 13:43:36 PDT
Hi David;
A comment on the first topic only:
> OVL Window definitions:
>win_change assertions continue checking the test expression through >the last cycle of the check, while the assert_window stops checking
>immediately upon hitting the end_event condition.
Looking at my tool notes, it check the data expression during the
end event. We have a note that one could delay the end_event
1 cycle if one doesn't want to check the data during that cycle.
We also allow a static hold time specification after the end
event so that you could check the data value a few clocks after
the end event.
I would suggest that if you had delay functionality, one could
use that to prevent checking the last cycle. Otherwise, I would
suggest that a parameter be specified to check a specific number
of extra cycles after the end_event. A value of 0 would not
check the cycle end_event is valid; a value of 1 would check the
cycle end_event is valid; a value of 2 would check the current
and next cycle, etc. The default parameter value is 1.
>I think the best solution to this inconsistency is to
> add a parameter to these assertions that lets the
> user select if the assertion should check the test
> expression during the last cycle. Setting the parameter
> to a 0 would not check on the last cycle, while settting
> it a 1 would check the last cycle.
BTW, there is value in starting the checking a fixed number of
cycles after the start event. For example, you could say you want
to check the stability of an address from 1 cycle after the
beginning of a transaction until the end of the transaction.
You also want to check a value on the same cycle as the start event
occurrs. This is not in the current OVL win* assertions.
Adam Krolnik
Verification Mgr.
LSI Logic Corp.
Plano TX. 75074
This archive was generated by hypermail 2b28 : Mon May 06 2002 - 13:45:49 PDT