From: Steven Sharp (sharp@cadence.com)
Date: Thu Mar 03 2005 - 10:50:00 PST
The following reply was made to PR errata/655; it has been noted by GNATS.
From: Steven Sharp <sharp@cadence.com>
To: etf-bugs@boyd.com, Shalom.Bresticker@freescale.com
Cc:
Subject: Re: errata/655: 2005D6, 7.1.6: ambiguities in instance array port connection rules
Date: Thu, 3 Mar 2005 14:08:09 -0500 (EST)
> However, the wording could potentially be interpreted as saying that it
> connects 1 of the 4 bits in that case to each instance (4 bits divided by
> 4 instances) and that the intent of the last rule is to generate an error if
> the vector width is not divisible by 4, not allowing to connect the same
> number of bits to each instance.
So the third rule should be reworded to make it clear that the part select
attached to each instance will be the same width as the instance port. So
something like
"If bit lengths are different, each instance shall get a part-select of the
port expression, of a bit length equal to the instance port bit length. The
least significant bit of the port expression shall be connected to the
instance corresponding to the right-hand index of the array range."
This rewording also addresses another ambiguity in the rule. It says that
the connection of part-selects starts with the right-hand index (presumably
of the array range), but does not say which end of the port expression is
connected to that start point.
Steven Sharp
sharp@cadence.com
This archive was generated by hypermail 2.1.4
: Thu Mar 03 2005 - 10:50:04 PST
and
sponsored by Boyd Technology, Inc.