FW: proposed UVM (1800.2) PAR

From: owner-stds-dasc@eda.org [mailto:owner-stds-dasc@eda.org] On Behalf Of Stan Krolikoski
Sent: Friday, September 05, 2014 7:30 PM
To: stds-dasc@eda.org
Subject: proposed UVM (1800.2) PAR

Hi All,

Attached is a proposed PAR for the "Universal Verification Methodology Language Reference Manual"  ("UVM").

As I suspect all of you know, UVM has over the past four or so years been developed in Accellera, and has been widely adopted in the world that does functional verification.  Accellera is now interested in potentially contributing UVM (the document, not the reference implementation) to the IEEE.  The attached is a proposed PAR that Tom Alsop and Hillel Miller developed with some help from me (with further additional help from Dennis and Yatin).  I am sending this potential PAR document to you for information purposes at this point.  We will discuss it at Thursday's DASC meeting, and will have a one week e-ballot ( that will end on the following Thursday) on whether to sponsor this PAR.

One point to note is the proposed numbering.  UVM, as it stands now, is very much defined in terms of SystemVerilog (IEEE 1800).  Other versions of UVM, e.g., a SystemC UVM, may follow in the coming years.  So, how do we organize and number this current effort?  One potential path would be to form a wholly new group- an umbrella  UVM WG- with its own unique number to handle all of the future UVMs.  I decided against this (after consulting with a number of knowledgeable people), since such "umbrella WGs" have not been our tradition in the DASC- we do not, for example, have  a single AMS WG or a single HDL WG.  Thus, I decided to make this new WG a "dot group" of the 1800 SystemVerilog WG.  A future SystemC UVM can, in the same way, be a dot group of the 1666 SystemC WG.

OK, you say, but why 1800.2? What happened to 1800.1?  To answer this, I again went back to DASC tradition.  We have 1076.1 (VHDL AMS), 1666.1 (SystemC AMS) and, presumably) in the future will have a SystemVerilog AMS WG that would be most sensibly numbered 1800.1.  In short, I think it most intuitive, given our 25+ year history", to reserve "wxyz.1" to be used for the AMS version of whatever IEEE wxyz is.  Thus, looking forward to a future 1800.1, I would like to request the number 1800.2 for the UVM WG.

Of course, there may be an IEEE "number czar" that will reject this approach in a pique of numerology, but we'll deal with that when if it happens.  I suspect that no one will object.

Any pre-Thursday's-meeting comments can be sent as a "reply all" to this email.


Stan

--
This message has been scanned for viruses and
dangerous content by MailScanner<http://www.mailscanner.info/>, and is
believed to be clean.

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