[Cfp-interest 3357] JTC1/SC22/WG14/CFP 2025/01/08 Meeting minutes -- Rev. 1

Thu Jan 9 23:08:58 PST 2025

2025/01/08: 8:00 AM PST
[Please submit proposed changes to these minutes to Jerome or to the group.
Changes in this revision are listed at the bottom.]

*Attendees*

    Jim Thomas, Rajan Bhakta, Jerome Coonen, Fred Tydeman, Joshua Cranmer,
Damian McGuckin, Hans Boehm, David Hough

*New agenda items*

    *https://wiki.edg.com/pub/CFP/WebHome/CFP%20meeting%20agenda-20250108-update.pdf
<https://wiki.edg.com/pub/CFP/WebHome/CFP%20meeting%20agenda-20250108-update.pdf>*

*Previous meeting notes*

    https://wiki.edg.com/pub/CFP/WebHome/n3445.pdf

*Next Meeting(s)*

    12 February 2025, 8:00 AM PST

    ISO Zoom teleconference

     Please notify the group if these time slots do not work.

*C documents*

    There is a new C2Y draft 3435 Jan 2025.
https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3435.pdf

    C23 has been published ISO/IEC 9899, available or purchase.

*IEEE 754 liaison*

   None.

*C++ liaison*

    Hans will present below.

*WG14*

    Next meeting is Feb 24-28.

    Robert Secord has asked if CFP would like to give a 754 revision update.

        Jim: Might be a good idea because the TS-4 and TS-5 revisions are
in ballot. This is a huge milestone.

        Rajan, et al.: The request really seems to be about the revision
work that has started in the last months. Given the target of 2029, it may
be too early than to give more than a cursory statement about the size and
activity level of the group, and some potential projects.

*C23 integration*

    C23 has been published. Has an October 2024 date, but referred to as
C23.

    https://www.iso.org/standard/82075.html

*Carry-over action items from last meeting *

    David: Draft reply to WG14 re midpoint and interpolation.

        Deferred to next meeting.

        __________

    Damian: Split doc into 3 documents: complex.h intro, cproj, and
editorial suggestions.
[Cfp-interest 3346] Re: The 'cproj' routine - 7.3.9.5 Damian McGuckin
        The cproj document is done and is a subject below.

*Action items from last meeting *

    Fred: Submit SIGFPE suggestions to WG14 for comment.
N3401 2024/11/24 https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3401.htm
Tydeman, SIGFPE and I/O, version 2
        Done.

    Jerome: Submit fadd, fsub, etc. domain errors proposal to WG14. (Will
require help from another member actually certified to submit to WG14.)
N3404 2024/12/01 https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3404.pdf
Thomas, Proposal for C2Y - Improved treatment of error conditions for
functions that round result to narrow type
        Done.

    Jerome: Submit exception "occurs" proposal to WG14 (again, with
support).
N3405 2024/12/01 https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3405.pdf
Thomas, Proposal for C2Y - Improved wording for treatment of error
conditions in <math.h>
        Done.

    Jim: Submit revised complex operators proposal to WG14.
N3406 2024/12/01 https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3406.pdf
Thomas, Proposal for C2Y - complex operators, updates N3384
        Done.

    All: Continue study of the imaginary I macro and i suffix issues,
noting that the I macro may be removed from C.
        Deferred, see below.

    Jim: Write a proposal for use of CMPLX to generate desired values with
signed 0 and inf when using the i suffix in a context with no imaginary
values, just real and complex.
[Cfp-interest 3337]
https://mailman.oakapple.net/pipermail/cfp-interest/2024-December/003352.html
Proposal for complex literals warning Jim Thomas

        Done, see below.

*TS-4 and TS-5 revisions*

    Part of WG14 discussion above.

*Discussion of issues*

    Fenv rounding modes
https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2024/p2746r6.pdf
        Hans: C++ group view is that it's better to pass a rounding mode to
a specific operation than to pass a mode to block (as suggested in 754).
Looked at a few examples of the struct, A major theme is that the C++
suggestion is not very C friendly.
        Jim: Noted a simpler way to evaluate the expression in the first
example.
        Hans: Not presented yet to 754 committee. Suggested it be kept in
the language domain.
        Jim: Have many comments, probably better to write them up.
        Rajan: Will come down to the compatibility study group.
        Hans: Reproducible arithmetic part of the conversation.
        Joshua: C++ doesn't want pragmas, but wants templates instead. Have
tried to propose, but have found resistance.
        Damian: If the templates do indeed have eleven arguments governing
their behavior, then we need to look at the overhead.
        Jim: The FENV_ROUND pragma is very new, so it's too early to ask
about implementation status.
        Hans: Rounding control will be via assembly snippets that change
the modes, all invisible to the compiler
        Joshua: C++ implementations allow something like FENV_ACCESS on a
per-function basis. (This is done via compiler extensions in all cases; the
standard itself says nothing here, and only a few implementations support
the FENV_ACCESS pragma itself.)

    Cproj cleanup
[Cfp-interest 3346]
https://mailman.oakapple.net/pipermail/cfp-interest/2025-January/003361.html
Re: The 'cproj' routine - 7.3.9.5 Damian McGuckin
        Damian: Have cleaned up the language and added some historical
rationale.
        Rajan: Suggest saving rationale for future reference.
        Jerome: One implementation years ago actually mapped to the Riemann
sphere in real 3-space. That implementation was changed to align with the
current specification, which is deemed more useful, so the discussion
probably need not be carried forward with the standard.

    Reply to WG14 re midpoint and interpolation
         David: Deferred but still in consideration.

    Use of CMPLX vs complex literals for special values
[Cfp-interest 3337]
https://mailman.oakapple.net/pipermail/cfp-interest/2024-December/003352.html
Proposal for complex literals warning Jim Thomas
        Rajan: Proposed removing reference to "execution time", but after
discussion withdrew.
        All: Discussion of other instances of execution time.

    Imaginary I
        Jim: Will keep on the agenda.

*Other issues:*

    Special cases for cpow
[Cfp-interest 3332]
https://mailman.oakapple.net/pipermail/cfp-interest/2024-November/003347.html
special cases of cpow Paul Zimmermann
[Cfp-interest 3336]
https://mailman.oakapple.net/pipermail/cfp-interest/2024-December/003351.html
Re: special cases of cpow Jim Thomas
        Jim: No reply yet to response to Paul Z. so deferred for now.

    glibc math functions
[Cfp-interest 3339]
https://mailman.oakapple.net/pipermail/cfp-interest/2024-December/003354.html
Fwd: More C23 math functions in glibc Jim Thomas
        Jim: Good news that GNU glibc is now supporting cospi, sinpi, and
tanpi.

*Adjournment*

    9:25 AM PST

-----------------------------------------------

*Action items to be carried over:*

    David: Reply to WG14 re midpoint and interpolation

*New action items:*

    Damian: Prepare cproj proposal for WG14.

    Jim: Submit n3352 related to complex literals to WG14.

        Rajan: OK to base on n3301 for now. Later, will move to the new
draft.

*Discussion issues to be carried over:*

    Imaginary I

Respectfully submitted.

Changes for Rev. 1:
    • Add Rajan to attendees. Use full names.
    • Drop "Do not distribute." from C2Y draft.
    • Edit Joshua's statement on FENV_ACCESS.
    • Refine Jerome's comment about an historical cproj mapping the the
true Riemann sphere.
    • Add Damian's comment about overhead in templates to handle rounding
control.
    • Fix several minor typos.

-Jerome Coonen
 650.996.4738
 jcoonen at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.oakapple.net/pipermail/cfp-interest/attachments/20250109/e358277c/attachment-0001.htm>