Hi,
I volunteered to write a statement asking for more time for SV-CC
work. Charlie expects that we will look it over next meeting, finalize
it, and then he will bring our request to the working group. This is
my take on what we want and why:
SV-CC has reached an impasse in the area of how
the information model represents and accesses dynamic data and
achieving the
right balance between functionality and performance. Given the current
schedule
constraints we are faced with a specification in this area that either
will not
be adequate for users, or one with no consensus and that will
undoubtedly be a
ballot issue for the LRM. On the other hand, there is agreement that
this
is a very important area to get right and we have the willingness to
work on
the information model to better represent dynamic data and deal with
its
lifetime.
We are requesting until the end of March
to
complete this work.
Having long sought a compromise between
backward
compatibility with the pre-SV information model and extension to SV
dynamic
data types, we have come to the conclusion that we need to focus on a
more
faithful representation in the information model. We have to enable
applications like debugging and design browsing. The validity of PLI
handles and their interaction with the lifetime of dynamic data are
also a
concern. There is a balance we seek between good performance of the
simulator on
the model in the presence of PLI, acceptable performance of PLI
applications,
and safety considerations. We may even need PLI applications that
participate in the testbench (or other aspects of the model) where
dynamic data
is used in which PLI affects object lifetime. What we have clearly
recognized is the amount of attention yet required and importance of
doing it
right. We need the time.
Regards,
John Shields
Mentor Graphics
--
This message has been scanned for viruses and
dangerous content by
MailScanner, and is
believed to be clean.
Received on Fri Dec 21 10:18:44 2007