RE: [sv-bc] MERGE REVIEW draft 2: Chapter 7

> 7.1.1 "By default, structures are unpacked, meaning that there is no
> defined representation on how each variable in the collection is
> stored."  Now a structure is said to be a collection of variables
> instead of a collection of types.

[SB] I would say that the representation of each variable is defined,
but their packing is not, as written in 7.1.1.1: "An unpacked structure
has an implementation-dependent packing, normally matching the C
compiler."

7.1.1.1 brings me to say that I dislike going down to 4th level
sub-clauses if it can be avoided. Instead of the following structure:

7.1 Structures and unions
  7.1.1 Structures
  7.1.2 Unions

how about simply

7.1 Structures
7.2 Unions ?


> 7.1.2 "A union is a data type that represents a single piece of
> storage
> which can store one of any variable type. Only one variable type can
> be
> in a union at one time."  What is a variable type?

[SB] The "one of" phrase is ambiguous.

Does it allow this example from the LRM?

typedef union packed { // default unsigned
                s_atmcell acell;
                bit [423:0] bit_slice;
                bit [52:0][7:0] byte_slice;
} u_atmcell;

I might think not.

Mark suggested, 
"A union is a data type that represents a single piece of storage which
can store several different variable types."

I think that is ambiguous too. A structure might answer to that
definition also.

BTW, is there any problem with 

union {
  integer a;
  integer b;
} cucu ;    ?


Shalom

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