[sv-ac] RE: 1730 question

The syntactic ambiguity (between a sequence instance and a nested
sequence expr) is resolvable once we know what the 'clk1' symbol means,
although that requires deferring the check until later in the parser. 

It would be better if the ambiguity could be removed at the syntax level
(rather than requiring some semantic analysis), e.g. by requiring parens
around clk: '@(clk1)'.  Unfortunately, this requirement would be unique
to passing it as an argument. 

 

Lisa

 

________________________________

From: Eduard Cerny [mailto:Eduard.Cerny@synopsys.com] 
Sent: Tuesday, May 22, 2007 3:03 PM
To: Lisa Piper; Eduard Cerny
Cc: sv-ac@eda.org
Subject: RE: 1730 question

 

Hi Lisa,

 

I suppose you mean that @clk1 is used as event control in procedural
clock, for example? I suppose that since all identifiers must be unique,
it is possible to determine that clk1 is not a sequence, but a variable
(or the other way round). If it is a variable, then it is an error, if
clk1 is a sequence then it is OK if it has the right formal arg.

Would that work?

 

Best...

ed

	 

	
________________________________


	From: Lisa Piper [mailto:piper@cadence.com] 
	Sent: Tuesday, May 22, 2007 2:55 PM
	To: Eduard Cerny
	Cc: sv-ac@eda.org
	Subject: 1730 question

	Hi Ed,

	What happens if I have:

	@clk1(@clk2  c + d)

	How do I know if  clk1 is a sequence with an argument that is a
sequence expr or a clock?  I am wondering if we are going to have a
problem now that we allow sequence and property expressions as arguments
due to the @ sign that is used to pass clocks.

	Lisa


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.