[sv-bc] clarification request

We have a user who has found that different tools are disagreeing on
whether a mixture of procedural and continuous assignments are allowed
to different bits of a packed struct or array.

IEEE Std 1800-2005 6.7 has the statement

"It shall be an error to have a packed structure or array type written
with a mixture of procedural and continuous assignments."

This seems to forbid it, but could be interpreted as referring to an
assignment to the packed object name, rather than the members or
elements of it.  Also, it does not match the sentence just ahead of
this, which says "Multiple assignments made to independent elements of
a variable are examined individually."  Since different members of a
packed struct or different elements of a packed array are independent,
this says that they would be examined individually, so no restrictions
would apply between them.

There is also the description in the previous paragraph about the
restriction being based on the longest static prefix.  This restriction
does not distinguish between packed and unpacked types.

I do not see any technical reason to restrict this for packed structures
and arrays, any more than for unpacked ones.  The only difference I can
see is that a packed object can also be written as the whole vector.  That
means an additional way that the bits can be referenced, and this must be
taken into account when checking that the same bits are not written in
two separate ways.  Perhaps the quoted sentence was trying to say something
about that.  Does anyone know the original intent of this text and the
reasons behind it?

Steven Sharp
sharp@cadence.com


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