IEC%20TC93%20Liaison%20Report%20DASC%20SC%20Alex%20Zamfirescu%20IEC%20USNC%20Technical%20Aadviser%20TC93%20Convener%20IEC%20TC93%20WG2%20September,%2019%202005

1
IEC TC93 Liaison Report DASC SCAlex Zamfirescu
IEC USNC Technical Aadviser TC93Convener IEC
TC93 WG2September, 19 2005
2
News from IEC TC93

• TC93 meetings took place during the week of
  September 4 to 9, 2005
• Hosted by Japan NC in Nara
• TC93 WG1 and WG2 discussions included
• Interoperability of international standards
• Maintenance of IEC standards
• Future candidates for IS
• Analog VHDL (IEEE 1076.1)
• SystemC (IEEE P1666)
• SystemVerilog (IEEE P1800)
• IEEE 1647 IEEE 1850 IEEE 1500 - 2005
  (Sponsor CSS) Extend 1603 for Radiation
  Hardening (RHR)
• Register Description Language Numerics Ontology
  as PAS

3
Good progress on IEEE DASC Dual Logo Submissions

• All DL Submissions were voted and approved by the
  P member countries
• IEEE Std 1497-2001 IEEE Standard for Standard
  Delay Format (SDF) for the Electronic Design
  Process IEC 61523-3
• IEEE Std 1076-2002 IEEE Standard VHDL Language
  Reference Manual .. ..IEC 61691-5
• IEEE Std 1364-2001 IEEE Standard Verilog
  Hardware Description Language ..
  ..IEC 61691-4
• IEEE Std 1076.4-2000 IEEE Standard for VITAL
  ASIC Modeling Specification.. ...IEC
  61691-6
• IEEE Std 1603-2003 IEEE Standard for Advanced
  Library Format (ALF) Describing Integrated
  Circuit (IC) Technology, Cells, and Blocks
  ................................................
  .............. IEC 62265 - Ed. 1.0
• IEEE 1076.6 ......IEC 62050
• IEEE 1364.1 ..IEC 62142

4
IEC TC93 Std Maintenance Extract
IEC Publication Number Kind Feeder Mnemonics IEC Maint.
IEC 61691-1-1 (2004-10) DL IEEE 1076 VHDL 2007
IEC 61691-2 (2001-06) IEEE 1149 MVL9 2004
IEC 61691-3-2 (2001-06) IEEE 1076.2 VHDLMATH 2004
IEC 61691-3-3 (2001-06) IEEE 1076.3 VHDLNUM 2000
IEC 61691-4 (2004-10) DL IEEE 1364 VERILOG 2006
5
Discussion about VHDL IEEE/IEC DL Maintenance

• The first four standards (on previous slide) are
  in the process of being merged into one LRM with
  additional VHPI documentation
• The group (WG2) agreed that the only alternative
  left for IEC participation is to wait until IEEE
  draft comes for ballot, form Accellera now
  jointly working on the documentation with the
  IEEE VASG.
• Note that IEC was not invited to form a joint
  working group as required by the DL procedures
  agreed upon in the IEEE-IEC MOU on DL maintenance
  (summary of the MOU was circulated twice in DASC
  SC minutes, and attached again to this report
  next 4 slides).
• No draft was circulated so far to IEC (possibly
  in contradiction to DASC SC assigned actions).

6
An IEC-TC/SC needs to revise an IEC-IEEE standard
IEEE needs to revise an IEC-IEEE standard
MCR
PAR
INFO
IEC and IEEE agree on a common
need
IEC disagrees see note 1
IEEE disagrees see note 2
YES
Dual logo maintenance team (DLMT) set up with
members from IEC and IEEE and a convenor jointly
accepted ( see note 3 )
Maintenance Procedure
Dual logo maintenance process see pages 2/3
7
Maintenance Procedure
8
Positive results
Positive results
NO
NO
INFO
See note 6
YES
See note 6
IEEE final ballot. Once approved IEC submits the
FDIS for final vote (2 months) See note 7
Maintenance Procedure
Revised IEC-IEEE standard
The DLMT prepares a common compilation of
comments on the CDV together with proposals to
solve them
IEC-TC/SC takes final decisions on the comments
by email or at a meeting
IEEE takes final decisions on the comments
received
INFO
The DLMT prepares the FDIS based on commonly
agreed modifications.
9
Maintenance Procedure Notes

• Note 1 IEC may decide to continue its own
  route. If any material is used from the IEC/IEEE
  standard, a separate license agreement will be
  required.
• Note 2 IEEE may decide to continue its own
  route leading to publish a revised single IEEE
  logo standard replacing the previous one. If so
  IEC will withdraw the existing IEC-IEEE standard.
  Then IEEE may submit to IEC this new standard as
  a candidate for a possible new IEC-IEEE standard
  according to the agreement.
• Note 3 The organisation which initiates the
  process, proposes a convenor first. If the
  proposal is not accepted by the other
  organisation, another convenor has to be chosen
  amongst the members of the DLMT by mutual
  agreement.
• Note 4 At this stage, IEC or IEEE are requested
  to task the DLMT to provide a new draft for
  comments
• Note 5 The 2 ballot processes must be carried
  out in such a way that the deadlines are the same
• Note 6 At this stage, IEC or IEEE may decide
  either to continue their respective route or to
  task the DLMT to provide a new draft for vote by
  both organizations. If IEC decides to continue
  its own route, then note 1 applies. If IEEE
  decides to continue its own route then note 2
  applies.
• Note 7 The comments made at this stage must be
  limited to corrections of misspelling or obvious
  errors which may have been generated during the
  preparation of the FDIS.

10
Discussion about Verilog IEEE/IEC DL Maintenance

• The Verilog situation is unfortunately similar.
• The IEC was not contacted to form a joint working
  group. However from the reports of WG1 we (the
  IEC TC93) know that some unprecedented
  deprecation took out nearly half of the LRM and
  the semantic of important functions used to
  interoperate tools and integrate applications.
• Therefore, there is a proposal to re-affirm the
  IEC 61691-4 Verilog (in 2006) without
  modifications. That will make sure that a
  standard including the IEEE version will still be
  a live standard.
• Decision should be reached by end of September
  (to keep Verilog as is in the IEC, to seek
  standardizing at the international level the new
  IEEE version or to do nothing).

11
Future Work

• Towards clarifying DL Std maintenance issues
• Develop roadmap for next generation standards
• Language independent numerics, TC93 WG2 will
  participate in DASC/DATC 2005 Workshop
• Analyze EDA relevance of efforts like SPIRIT and
  SAE Architecture Std.
• Standards to support Statistical EDA, 3D, Farming
• Plans to simplify standardization by separating
  PASs from STDs
• Involve all available experts from around the
  globe