[Cfp-interest 1264] Re: Fw: WANT macros

[Rajan Bhakta]

Hi Jim,

Agreed.

Regards,

Rajan Bhakta
z/OS XL C/C++ Compiler Technical Architect
ISO C Standards Representative for Canada, PL22.11 Chair (USA)
C Compiler Development
Contact: rbhakta at us.ibm.com, Rajan Bhakta/Houston/IBM



From:   Jim Thomas <jaswthomas at sbcglobal.net>
To:     Rajan Bhakta <rbhakta at us.ibm.com>
Cc:     Cfp <cfp-interest at oakapple.net>
Date:   02/19/2019 10:47 AM
Subject:        Re: [Cfp-interest 1242]  Fw: WANT macros





On Jan 29, 2019, at 6:39 AM, Rajan Bhakta <rbhakta at us.ibm.com> wrote:

Hello,

Some feedback re our position to remove the WANT macros.
I intend to push back on the header version macros saying it needs a 
proposal.
Anything else the group wants me to bring forward?

I think we should give our opinion on renaming math functions, if it comes 
up.

I do intend on writing our status report after our next meeting which I 
hope we can review via this list before I submit it.
The current text is:
  - Working closely with 754 (strong liaison, one member is chair and one 
is the editor)
  - Working on creating updated TS's for the ones voted in (parts 2 and 3) 
for inclusion into the working C2X draft

Only parts 1 and 2 are voted in so far. Right? Part 1 is integrated (by 
Jens) into a C2X working draft. The integration draft for part 2 will be 
submitted to Jens. Part 3 as annex will be submitted to WG14. A new 
proposal for part 4a will be submitted WG14. This should all be done 
before the next WG14 mailing, assuming CFP approves.

- Jim Thomas

  - Will continue binding IEEE-754 2019 to the TS's that remain after C2X

Regards,

Rajan Bhakta
z/OS XL C/C++ Compiler Technical Architect
ISO C Standards Representative for Canada, PL22.11 Chair (USA)
C Compiler Development
Contact: rbhakta at us.ibm.com, Rajan Bhakta/Houston/IBM
----- Forwarded by Rajan Bhakta/Houston/IBM on 01/29/2019 08:32 AM -----

From:        Jens Gustedt <jens.gustedt at inria.fr>
To:        "Rajan Bhakta" <rbhakta at us.ibm.com>
Cc:        David Keaton <dmk at dmk.com>, Larry Jones <
lawrence.jones at siemens.com>
Date:        01/29/2019 02:28 AM
Subject:        Re: WANT macros



Hello,

On Mon, 28 Jan 2019 17:49:32 -0600, Rajan Bhakta wrote:

> Also, just so it's not a surprise, the CFP group will be stating our 
> preference of removing the WANT macros as we believe they are needed
> for TS's but not for actual standards. I intend to write up our
> position in the CFP group status report in time for the London
> mailing.

That is good to hear, so we mainly agree, here.

The only thing that I'd like to note to that position is that it would
make more sense to make the distinction between mandatory and optional
parts of the standard.

So my personal position would be

- no WANT macros in the numbered clauses

- WANT macros in Annex F and perhaps Annex G (don't remember if we
  were adding interfaces, here)

This is to avoid exposure of symbols of optional parts of the standard
to code that doesn't expect them. This would keep us in line with the
current policy in Annex K.

What I also would like to add is version macros for the headers. This
is because implementations and users will otherwise have a difficult
time getting things straight during a transition to C2x. Nowadays,
compiler and library often come from different parties and a global
version macro for the standard version is not appropriate to match
that situation. POSIX does this and has good experience with that, I
think.

So I would like to have __STDC_MATH_VERSION__ and similar, for all
library clauses where we add interfaces.

> From:   David Keaton <dmk at dmk.com>
> ....

>       One thing you could do is to merge the TSes the way you hope it 
> works, and let the committee see how that looks.  However, that might
> be a lot of work, so you don't have to do that.

The first is now merged as had been proposed (modulo bugs,
probably). Removing these parts is not a big deal, editorially, but I
wouldn't want to change that before London, unless you are telling me
that removing WANT is what everybody voted for implicitly.

What we could do is to add a warning in the preamble to early adopters
that this WANT thing is something that is likely to change in the
process.




Thanks
Jens


-- 
:: INRIA Nancy Grand Est ::: Camus ::::::: ICube/ICPS :::
:: ::::::::::::::: office Strasbourg : +33 368854536   ::
:: :::::::::::::::::::::: gsm France : +33 651400183   ::
:: ::::::::::::::: gsm international : +49 15737185122 ::
:: http://icube-icps.unistra.fr/index.php/Jens_Gustedt::


<attvi5yg.dat>_______________________________________________
Cfp-interest mailing list
Cfp-interest at oakapple.net
http://mailman.oakapple.net/mailman/listinfo/cfp-interest




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.oakapple.net/pipermail/cfp-interest/attachments/20190219/58859dd3/attachment-0001.html