RE: [sv-bc] $bits question

I think we need to add a restriction that limits $bits to the bit-stream
types; i.e. only those types that allow us to access individual bits of
their data.

 

I seem to recall a big debate about whether to allow anonymous types in
places other than in the declaration of a variable, but I don't remember
the reason why we agreed to allow it and not require an intermediate
typedef. I think it is one of those things that would take more effort
to define than its worth, but I would support a proposal.

 

If we do agree the argument to $bits is legal, however odd, then the
enum label is created as a symbol in the current scope. So the reference
to x is legal, and the duplicate symbol is illegal.

 

Dave

 

 

________________________________

From: owner-sv-bc@eda.org [mailto:owner-sv-bc@eda.org] On Behalf Of
francoise martinolle
Sent: Friday, January 13, 2006 9:52 AM
To: sv-bc@eda.org
Subject: [sv-bc] $bits question

 

Is there any semantic restriction on the data type which can be passed
as an argument to $bits?

It looks to me like this argument can be syntactically a data_type, so
is this legal"

typedef bit [ $bits(enum { x }) : 0] mytype;

Where is the enumeration constant declared? In the scope containing the
declaration of "mytype"?

So is it legal to reference x outside the typedef?

mytype v;

initial

   v = mytype'(x);

 

and is the following illegal?

typedef bit [ $bits(enum { x }) : 0] mytype1;

typedef bit [ $bits(enum { x }) : 0] mytype2;  // duplicate symbol x