Subject: SV_CC meeting minutes for 15 Jan, 2003
From: Amouroux, John (john_amouroux@mentorg.com)
Date: Wed Jan 15 2003 - 10:20:01 PST
SV-CC Meeting Minutes for January 15, 2003:
Attendees::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
John Amouroux (Mentor)
John Stickley (Mentor)
Doug Warmke (Mentor)
Michael Rohleder (Motorola)
Swapnajit Mittra (SGI)
Joao Geada (Synopsys)
Ghassan Khoory (Synopsys)
Andrzej Litwiniuk (Synopsys)
Agenda:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
- General procedural work - Swapnajit [5 minutes]
- Andrzej's modified proposal - Andrzej [35 minutes]
http://www.eda.org/sv-cc/hm/0643.html
- Issue 1.4 - Michael [20 minutes]
General::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
The 1/14 meeting minutes were proposed and seconded for acceptance by John
A. and Michael.
Swapnajit informs the group that his Friday will be a chair's meeting. He
will report that we are basically on schedule.
Andrzej's modified proposal::::::::::::::::::::::::::::::::::::::::
Andrzej has cleaned up the doc as per various member's suggestions.
He goes through these with the group...
Michael is not sure about some aspects of naming [functions?]; he wants to
ensure that names don't collide.
ACTION: Michael to check this out (maybe with Francoise's help).
[JA - I think this the action item Swapnajit refers to later in the
meeting.]
Doug is not clear on normalizing indexing with packed arrays because (n-1:0)
indexing doesn't make sense in C.
Andrzej says that this is from the SVerilog point of view.
Andrzej has added the const qualifier to all input arguments. Not really
needed for int's, but using it still makes things clearer and more
consistent.
Doug thinks that for clarity we should explain that the "denormalizing" of
indexing is possible.
ACTION: Doug to give Andrzej a few extra sentences to help with this.
Andrzej brings up the naming convention of the header files svc_bin.h and
svc_src.h. While there was some discussion of changes here, the concensus
is to leave the names as is unless someone suggests better names off-line.
Doug suggests use of the word "chunks" instead of words for the sizing
functions. Andrzej suggests "canonical size" and the group agrees.
Now Andrzej discusses restrictions...
Sized arrays have direct access issues when mapping to/from C. Unsized
array do not have this issue because they have abstract access methods.
svGetArrayPtr issue... Cannot always get a contiguous C representation of
arrays due to layout restrictions. This function will return a NULL in
these cases and the GetArrElem functions must be used.
ACTION: (according to Swapnajit) Michael to check on bit part select
definition. [JA - This is the action from above.(?)]
ACTION: For ALL, Swapnajit reminds us that the poll on the assertions
document is due today (1/15) by midnight.
Issue 1.4 (Michael's)::::::::::::::::::::::::::::::::::::::::::::::::
No peer relationship to other API's.
Michael says this is now much clearer w.r.t. assertions and directc, and
thinks this issue has been mostly solved.
Reminder, the "pure" keyword means that there are no side-effects, even
though pli/vpi read functions can be called. Doug suggests another
qualifier that says NO pli/vpi functions will be called. Andrzej notes that
we should have the keyword on if pli/vpi functions will be called and have
the default be for no pli/vpi calls allowed.
ACTION: Michael to come up with a proposal by the end of this week.
Meeting end::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
Roll call taken.
Face-to-face meeting will be on 1/23, hosted by Synopsys.
This archive was generated by hypermail 2b28 : Wed Jan 15 2003 - 10:20:53 PST