[Cfp-interest 2467] Re: Likely upcoming work for CFP from WG14'smeeting today

[Mike Cowlishaw]

If there is confusion, then indeed clarification is justified.  An explicit
statement of the quantum exponent makes sense, even if technically it's not
necessary.
 
Mike


  _____  

From: Cfp-interest [mailto:cfp-interest-bounces at oakapple.net] On Behalf Of
Rajan Bhakta
Sent: 21 July 2022 18:06
To: cfp-interest at oakapple.net
Subject: [Cfp-interest 2466] Likely upcoming work for CFP from WG14'smeeting
today



Hello,

 

In today's WG14 meeting the ++(E) quantum exponent issue came up again (that
we discussion the March meeting this year).

Given later changes to the C standard wording (for _BitInt I believe) and
what was already there for postfix, we will need to revisit this.

Joseph has a suggestion in a chat he sent me, but that of course is not
necessarily the only way to resolve this issue.

 

Here are the comments I got:

 

For ++E DFP quantum exponent issue:

>From Joseph:

The issue is that the wording now consistently no longer says 1 in program
font, it says 1 in appropriate type.

 

Based on the comment I made:

For the +1 quantum exponent issue, CFP discussed this before (in the March
2022 meeting) and this was our discussion (and conclusion):

  C/C++ defines ++ as adding the value 1 to the operand. For DFP, what is
the quantum exponent of that 1?

    Zero is the quantum exponent. This follows from ++E is equivalent to
(E)+=1 where the 1 is in program font meaning an int constant. The C
conversion from int to decimal gives a preferred quantum exponent of zero.
So the rules cover this case already.

    No other result seems to make sense anyways.

    Perhaps say it is adding the integer one instead of using the font in
the C text for that ++E to clarify?

 

  Joseph: Postfix ++ and -- said "appropriate type" all along. Prefix
changed for _BitInt, because 1 of type int is inappropriate for _BitInt. 1
of appropriate type i fine for both prefix and postfix *if* an additional
explicit statement of quantum exponent is added for the DFP case. (And it
should be fine for that to be quantum exponent 0, as if integer 1 were
used.)

 

Regards,

Rajan Bhakta
z/OS XL C/C++ Compiler Technical Architect
ISO C Standards Representative (Canada, USA), PL22.11 Chair
C/C++ Compiler Development

rbhakta at us.ibm.com

 

IBM

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.oakapple.net/pipermail/cfp-interest/attachments/20220721/1f568af8/attachment-0001.htm>