ISAC Minutes of meeting held 17 Feb 2005

I'm attaching the minutes. They are also available on the isac website.
Chuck



Minutes of ISAC meeting held via telecom on 
17 February 2005

Present: Peter Ashenden, Jim Lewis, Deepak Pant, Larry Soule, Chuck
Swart, Ajay Varikat

Absent: 

Next Meeting: Thursday 03 March 2005, 7 pm Pacific Time

TOPIC: IR2061 Default actions on severity flags is different between simulators

The consensus is to specify a recommendation that the simulator
continue execution on severity error or less.

ACTION: Chuck to update, all to vote.

TOPIC: Obtain list of VHDL vendors?

It was agreed that it would be useful for both ISAC and VASG to have a
general list of VHDL vendors, i.e., tools which either produce VHDL or
analyze/simulate VHDL. Accellera might be able to help produce such a
list.  Its probably more appropriate for VASG to prepare this list.

ACTION: Chuck to work with Steve to decide whether ISAC or VASG pursues this.

TOPIC: IR2013 Exact subtype "matching" for port associations

ISAC Approved.

ACTION: Chuck to forward to VASG.

TOPIC: IR2028 Clarify simulation cycle. Note: waiting for VHPI input

Still waiting for VHPI input.

TOPIC: Lack of line numbers in LRM makes analysis difficult.

It is cumbersome to describe problems in the LRM without the use of
line numbers. It might be possible to add paragraph numbers, or we
might be able to use a standard representation, e.g., pdf. The real
need is for visible notation finer than existing subclauses.

ACTION: Peter to work with IEEE on this.

TOPIC: IR2042 Architecture as a block causes problems

Peter is still working on this. There are several technical problems,
particularly with the visibility of the architecture name.

TOPIC: IR2048 Miscellaneous errors

    Problem 4: pure function which return access types.  Rather than
    forbid allocators in pure functions, better to forbid pure
    functions from returning an access value.

    Problem 8 signature of function whose base type is anonymous.
    Reword to make clear that the type mark used in the declaration is
    used.

    Problem 10 Constant parameter of access type in "=" Superseded by
    IR 2057

ACTION: Chuck to revise, then all to vote.


TOPIC: IR2051 Path_name and instance_name don't deal with operator symbols

Change "protected type simple name" to "shared_variable_simple_name"

ACTION: Chuck to make minor revisions, then all to vote.

TOPIC: IR2052 Path_name and instance_name don't deal with operator symbols

This omission can easily be repaired by replacing
<subprogram_>simple_name with designator.

ACTION: Chuck to update, then all to vote.

TOPIC IR2057 Access-typed parameters to predefined "=" and "/=" 

Several problems were uncovered with allowing general access types as
parameters to pure functions. The short term solution is to special
case the predefined "=" and "/=" operators. A note will be added
identifying issues for abstract data types in long term solutions.
The general idea is that purity is needed for resolution functions and
for staticness of expressions in elaboration.

ACTION: Chuck to update, then all to vote.

TOPIC: IR2058

Don't make the base type visible when the use clause references a
subtype.

Change "physical literals" to "physical units"

ACTION: Chuck to update, than all to vote.

TOPIC: IR2043 Numeric VALUE attribute parameter can't have sign

No action has been taken. Deepak will submit revision within a week or
so.

ACTION: Deepak to update.

TOPIC: IR2050 Definition of S'Last_Value was apparently broken in 1993

No work has been done.

ACTION: Peter to work on.

TOPIC: IR2054 Individ. assoc. rules for array formal are not valid

Larry is still working on.

ACTION: Larry to work on, Peter or Chuck available as resource.

TOPIC: IR2053 Minor Typos in VHDL 2002 part 2

Issue 6--Incorrect cross reference.  It isn't clear whether the term
[B11] refers to B for Bibliography or B for appendix B. Peter will
check the style guide.

ACTION: Peter to check style guide, Chuck to update, all to vote.
Note: Peter reports that the "B" refers to "Bibliography", not to the
specific annex.This has been incorporated into the IR.