LCS2016_IEEEStdUpdates < P1076

P1076 Web>VHDL2017>LCS2016_IEEEStdUpdates (2017-04-02, PatrickLehmann) (raw view)EditAttach
---+ Language Change Specification for Updated reference IEEE Standards
<noautolink>
<sticky>
---++ 

| <sticky><b>LCS Number:</b></sticky> | LCS-2016-IEEEStdUpdates |
| <sticky><b>Version:</b>   </sticky> | 2 |
| <sticky><b>Date:</b>      </sticky> | 30-Mar-2017  |
| <sticky><b>Status:</b>    </sticky> | Approved |
| <sticky><b>Author:</b>    </sticky> | [[Main.PatrickLehmann][Patrick Lehmann]] |
| <sticky><b>Summary:</b>   </sticky> | Updating the IEEE Std references. |

%TOC%

---++ Voting Results:

Yes:
   1 %USERSIG{SteveGrout - 2017-03-27}% - ver 1
   1 %USERSIG{MartinThompson - 2017-03-27}% - ver 1 (see comment about typo)
   1 %USERSIG{PatrickLehmann - 2017-03-30}% - ver 2
   1 %USERSIG{JimLewis - 2017-03-30}% - ver 2
   1 %USERSIG{RobGaddi - 2017-03-30}% - ver 2
   1 %USERSIG{LievenLemiengre - 2017-03-30}% - ver 2

No:

Abstain:

---++ Style Notes
Changes are shown in %RED%red font%ENDCOLOR%.%BR%
Deletions are %RED%<del>crossed out</del>%ENDCOLOR%.%BR%
Restructuring is shown in crossed out %GRAY%<del>gray font%ENDCOLOR% somewhere else in %GRAY%gray font%ENDCOLOR%.%BR%
Editing or reviewing notes in %GREEN%green font%ENDCOLOR%.
 
---++ Revisions
*Version 2:* 30-Mar-2017
   * Deleted "Annex E" entirely

*Version 1:* 27-Mar-2017
   * Fixed IEEE Std 754-1985 -> 754-2008
   * Fixed IEEE Std 854-1987 -> deleted, subsumed by 754-2008
   * Fixed IEEE Std 1076.6 -> deleted, withdrawn


---++ Details of Language Change
---+++ 2. Normative references
IEEE Std 754<sup>TM</sup>-%RED%2008 <del>1985 (Reaff 1990)</del>%ENDCOLOR%, IEEE Standard for %RED%<del>Binary</del>%ENDCOLOR% Floating-Point Arithmetic.<sup>3, 4</sup>

%RED%<del>IEEE Std 854<sup>TM</sup> -1987 (Reaff 1994), IEEE Standard for Radix-Independent Floating-Point Arithmetic</del>%ENDCOLOR%



---+++ 5.2.5.1 General (Floating-point types)
An implementation shall choose a representation for all floating-point types except for universal_real that
conforms %RED%<del>either</del>%ENDCOLOR% to IEEE Std 754-%RED%2008<del>1985 or to IEEE Std 854-1987</del>%ENDCOLOR%; %RED%<del>in either case,</del>%ENDCOLOR% a minimum representation
size of 64 bits is required for this chosen representation.


---+++ 6.5.2 Interface object declarations
NOTE 3 - Ports of mode linkage are used in the Boundary Scan Description Language (see IEEE Std 1149.1TM-%RED%2013 <del>2001</del>%ENDCOLOR% [B15]).


---+++ 16.8.1.1 Scope (Standard synthesis packages)
%RED%<del>Further related standard practices for synthesis of register-transfer level digital circuits are specified in IEEE Std 1076.6-2004 [B14].</del>%ENDCOLOR%


---+++ Changes from IEEE Std 1076-2002 (Annex E)

%GREEN%This section is deleted in its entirety.%ENDCOLOR%

---+++ G.5.2 Use model
The package includes a number of functions recommended by IEEE Std 754-%RED%2008 <del>1985 and IEEE Std 854-1987</del>%ENDCOLOR%. They are described in G.5.4.4.


---+++ G.5.4.2 Functions

| sqrt | Square root. Inputs: arg: float; round_style: round_type; guard: NATURAL; check_error: BOOLEAN; denormalize: BOOLEAN. Returns the square root of arg, as defined by IEEE Std 754-%RED%2008 <del>1985</del>%ENDCOLOR%. |


---+++ G.5.4.3 Conversion functions
| classfp | Find the classification of a floating-point number. Inputs: arg (float). Returns a value of the type valid_fpstate. Note that IEEE Std 754-%RED%2008 <del>1985 and IEEE Std 854-1987</del>%ENDCOLOR% \
            recommends%RED%s%ENDCOLOR% the name ?class? for this function. However, the floating-point package calls the function ?classfp? to avoid conflict with ?class? as a reserved word in a \
            future extension of VHDL. |

---+++ Glossary (Annex I)
<b>chosen implementation:</b> An implementation of floating-point types that conforms to %RED%<del>either</del>%ENDCOLOR% IEEE Std 754-%RED%2008 <del>1985 or to IEEE Std 854-1987</del>%ENDCOLOR% and
with a minimum representation size of 64 bits. (5.2.5.1)

%GREEN%[...]%ENDCOLOR%

<b>floating-point types:</b> A scalar type whose values approximate real numbers. The representation of a floating-point type conforms %RED%<del>either</del>%ENDCOLOR%
IEEE Std 754-%RED%2008 <del>1985 or to IEEE Std 854-1987</del>%ENDCOLOR% and has a minimum size of 64 bits. (5.2.1, 5.2.5)


---+++ Bibliography (Annex J)
%RED%<del>[B14] IEEE Std 1076.6TM-2004, IEEE Standard for VHDL Register-Transfer Level (RTL) Synthesis.</del>%ENDCOLOR%

[B15] IEEE Std 1149.1TM-%RED%2013 <del>2001</del>%ENDCOLOR%, IEEE Standard Test Access Port and Boundary Scan Architecture.


---++ Comments
---+++ Version 1




It seems that G5.4.3 has an extra red &#34;s&#34; in &#34;recommends&#34;

-- %BUBBLESIG{MartinThompson - 2017-03-28}%


In &#34;Changes from IEEE Std 1076-2008 2002 (Annex E)&#34; the first line:
&#34;This annex lists those clauses that have been changed from IEEE Std 1076-2002, during its revision.&#34;
should be:
&#34;This annex lists those clauses that have been changed from IEEE Std 1076-2008, during its revision.&#34;

-- %BUBBLESIG{Pablo2Blecua - 2017-03-28}%


Oh, and copying from PDF introduced some ? signs ...

-- %BUBBLESIG{PatrickLehmann - 2017-03-28}%
</sticky>
</noautolink>
Topic revision: r8 - 2017-04-02 - 16:28:19 - PatrickLehmann
P1076