Subject: RE: Call for Final Vote on Assertion Spec (version 1.8)
From: Dagan, Gail M (gail.m.dagan@intel.com)
Date: Fri Apr 05 2002 - 13:02:12 PST
I vote "yes", but have the following nits that should be cleaned
up in the document, in addition to the things David already mentioned.
I tried to go through it as if I'd never seen it before and wanted
to add assertions to my design today.
It's easy to miss the single line starting the new paragraph after
the syntax table on page 4. I originally wondered if I had a corrupted
copy because it looks like page 5 starts in the middle of the sentence.
Just add a blank line or force the new paragraph to start on a new
page after the table.
In the assert_msg syntax box on page 5 it's confusing whether ; is
required at the end of both fatal_func and msg_func. It's explicitely
quoted in on one case and not the other and is also included in the
higher level asert_msg definition.
On page 7 on the second line of the Sequential Assertions paragraph.
Can you change:
One way of doing this is to use nested immediate asertions:
to:
One way of doing this is to use nested immediate assertions where
each subsequent assertion is the pass statement of the previous
assertion
It will be clearer for someone reading this for the first time and will
also clarify the // no semicolon in the example.
Two line after the first example on page 7,
req must be false.
should read
req1 must be false.
On page 8 in the second line of text after the 3rd example (the
assert_strobe
example) change
it is common to use strobed assertions be used in combinational
always blocks.
to
it is common to use strobed assertions in combinational always
blocks.
On page 9 I have a question about the examples in the middle of the page
regarding valid step control expressions.
assert (a;b;c) @@(clk);
says it will be sampled on any edge of clk. The next example says:
assert (a;b;c) @@(ev1);
says it will be sampled when ev1 fires, which sounds like when it goes
high. Don't both of these mean that sampling will occur when the
signal changes, either edge?
On page 11 if the assertion control system tasks are used with arguments,
what is the syntax of the arguments other than the levels, wihch is a
number. Are the module names or individual assertion names strings or
not.
Thanks, Gail Dagan
Phone 503-712-1781
-----Original Message-----
From: David Lacey [mailto:dlacey@rsn.hp.com]
Sent: Thursday, April 04, 2002 4:31 PM
To: Accellera Assertion
Subject: Call for Final Vote on Assertion Spec (version 1.8)
The final version of the assertion spec was emailed out along with the
minutes from today's meeting (4/4/02). It is version 1.8. We need to
have a final vote on the document by the end of Friday.
Please respond to the reflector with your vote for or against the spec
as it is in version 1.8. The final tally will be made Friday at 6 PM
CST.
Thanks.
David
-- David J. Lacey Email: dlacey@rsn.hp.com Phone: 972-497-4114
This archive was generated by hypermail 2b28 : Fri Apr 05 2002 - 13:06:24 PST