Subject: Re: [sv-ec] Email Vote 3 on proposed changes
From: Francoise Martinolle (fm@cadence.com)
Date: Fri Feb 07 2003 - 12:28:01 PST
Dave,
I think it is very difficult to review chapters 13 and 14 change by change
because
we modified it so much. I think it would be better if we were given the new
chapters
13 and 14 all together and review them again as a whole to get a better
global picture.
I am very confused because I don't have the whole new write up to build a basic
view of the clocking domain and clocking variables operations.
Anyway here is my vote on these items. I would like to read again the
chapters after all the
modifications have been made to put clarity in my mind about the new
terminology
syntax and semantics.
At 10:20 AM 2/5/2003 -0800, David W. Smith wrote:
>The following are the list of changes that need approval. Please consider
>each and vote on them.
>
>__yes _X_no 1. Approve CH-17
I don't like the fact that persistent events are declared a different way
(event bit) that regular
events but the trigger operation is the same for persistent and regular events.
There may be other ways of accomplish this without creating a new type of
events.
>X__yes __no 2. Approve CH-18
In example 2:
Complete the example. Where are these clocking domains declared in?
>_X_yes __no 3. Approve CH-20
>__yes _X_no 4. Approve CH-47
If you look at the bnf for constant_expression A.8.3, a constant expression
can be
much more that what you allow for skew value. Specifically
constant_expression contains
string. Constant expression also includes constant_primary and a
constant_primary can be
a concatenation, a function call, a genvar, a specparam, a parameter etc...
I think that the sentence needs to be rewritten to say something like:
it must be a constant_expression of type unsigned int or time.
What happen if skews are not specified? What are their default values?
>__yes _X_no 5. Approve CH-53
Minor rewording: replaces statement with declaration. I believe that a
default clocking
is a declaration and not a statement.
>__yes _X_no 6. Approve CH-60
Replaces "If the input skew is zero then the value sampled corresponds to
the signal value at the start of the verification phase." with:
"If the input skew is zero then the value sampled corresponds to the signal
value at the clock domain event"
>_X_yes __no 7. Approve CH-83
>__yes _X_no 8. Approve CH-84
Leave as it was previously said
>_X_yes __no 9. Approve CH-85
>_X_yes __no 10. Approve CH-86
>_X_yes __no 11. Approve CH-93
>__yes _X_no 12. Approve CH-94
Should refer to regular event control but not include it here
>_X_yes __no 13. Approve CH-95
>__yes _x_no 14. Approve CH-96
add bnf for event_count.
What about using non blocking drives? Do they disappear?
>_X_yes __no 15. Approve CH_97
>__yes _X_no 16. Approve CH-98
If the the conflicting drives are only for net and not for variables, I
would insert the
additional proposed paragraph before instead of after.
>__yes _X_no 17. Approve CH-99
What is the verification phase?
In the example:
isn't bus.data = 0 supposed to be bus.data <= 0? I thought we were only
allowing
0 delay non blocking drives.
>__yes _X_no 18. Approve CH-100
I suggest that we use:
wait <block_name>
so that if we just name the fork parallel block we can just use
that name to wait for all the spawned processes to complete.
Also use:
disable <block_name> to disable the fork
I don't understand the difference between $terminate and disable.
I don't see the need for $suspend_thread if this is equivalent to a #0,
just use #0
>__yes __no 19. Approve CH_101
Does it mean that we are merging 13 and 14?
>
>Voting on these items will close on Wednesday 12 February 2003. Same
>voting rules as last time.
>
>I expect that some of these will not pass since there are some significant
>changes in response to the action items. I am doing this to prune done the
>list of items to the ones that we need to discuss to help focus our time
>in meetings.
>
>Regards
>David
This archive was generated by hypermail 2b28 : Fri Feb 07 2003 - 12:29:16 PST