[cfp-interest 3533] JTC1/SC22/WG14/CFP 2025/06/11 Meeting minutes
Jerome Coonen
jcoonen at gmail.com
Thu Jun 12 11:17:53 PDT 2025
2025/06/11: 8:00 AM PDT/3:00 PM UTC
[Please submit proposed changes to these minutes to Jerome or to the group.
Revision changes appear at the bottom.]
*Attendees*
Jim Thomas, Rajan Bhakta, Jerome Coonen, Damian McGuckin, Joshua
Cranmer, Tue Ly, Fred Tydeman
*Updated agenda plus new items*
https://wiki.edg.com/pub/CFP/WebHome/CFP%20meeting%20agenda-20250611-update.pdf
–
These minutes should be read alongside the agenda, with its many reference
links.
*Previous meeting notes*
https://mailman.oakapple.net/pipermail/cfp-interest/2025-May/003479.html
*Study group logistics*
Next meeting: 9 July 2025, 8:00 AM PDT/3:00 PM UTC
ISO Zoom teleconference
Please notify the group if this time slot does not work.
*C documents*
The latest C2Y draft is N3550 May 2025
https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3550.pdf
C23 has been published ISO/IEC 9899, available for purchase.
https://www.iso.org/standard/82075.html
*IEEE 754 liaison*
Damian: Came to agreement on the meaning of "precision". Damian will
send a note elaborating. Jerome comments on the number o significant bits
in a binary type (traditionally, it's "precision") vs. the approximate
precision in decimal digits vs. the number of decimal digits required to
distinguish binary values in the given precision vs., yes, the number
decimals distinguished by the binary numbers. Damian will speak precisely
on this. Also, Damian and Jim Demmel and others will investigate complex
arithmetic.
*C++ liaison*
Reproducible floating point -- no update from Guy Davidson but David
Hough has sent a note to the group.
Joshua: Will send a note to work on tightening up the semantics.
The next C++ group meeting is next week.
*WG14 update*
The mailing deadline is 25 July 2025 for the next meeting over 24-29
August.
*TS-4 and TS-5 revisions*
Jim: MInor issue, to be discussed below.
*Carry-over action items from last meeting *
Damian: Update the proposal re. the wording of cimag() and creal().
Done.
Ly: Send email to CFP about the next____() issues pertaining to the
several functions inspired by nextafter().
Done.
Rajan: Monitor the new WG14 issue tracker re. TS 18661-4, the domain
and range error issue.
Rajan: Nothing has happened here, so I will raise the matter at
next WG14 meeting.
*Action items from last meeting *
Jim: Draft a note about the [[reproducible]] issue.
Done.
Jerome & Jim: Get a document # for the powr proposal, 3550, check the
minor updates, and submit to WG14.
Done.
Damian & Jim: Get a document # and submit the cproj proposal with minor
updates to WG14.
Done.
Jim: Review the complex.h proposal from Damian to confirm that the
changes listed in CFP 3448 are correct, and there if are no other
significant changes then submit it to WG14. Otherwise, CFP will review it
again next month.
Done.
Jerome: Search C2Y for instances of "return value" vs. "return type"
and summarize the usage across the whole document.
Done.
Rajan: Send note to Javier supporting his changes re. atan2 and atan2pi
range errors, and encouraging him to proceed with his proposal.
Done, and more.
Jim: Write up a suggestion re. semantic rules for constant evaluation.
Done, email 3465
<https://mailman.oakapple.net/pipermail/cfp-interest/2025-May/003480.html>
(linked here in case anyone missed it).
Ly: Send email on expanded complex functions.
Carry over.
Jim & Group: Write a note about the priorities list while group ponders
list.
Done.
*Discussion of issues:*
Reproducible floating point
Jim: With David unavailable, we'll carry over.
[[reproducible]] example in 6.7.13.8 ignores exceptions
Jim: No response from JeanHeyd. Should we just remove the example?
Rajan: The only way to get action is to write a paper. Perhaps
start with JH.
Jim: Will follow up.
powr(x,y) with x < 0
Jerome: The proposal reflects the decision last meeting to add the
language "all y (including NaN)" to bring the 3 relevant usages in C2Y into
harmony. Whether this is actually the best phrasing might come up in the
future.
Range error wording for atan2 and atan2pi
Damian: This proposal removes an ambiguity in the phrasing of the
elaborate conditional, using language that is consistent with other usage
in C2Y.
Jerome: As in the powr(x,y) case just above, this may not be the
best language, but it is clear and consistent.
Semantic rules for constant expression evaluation
Jim: The fix deals with one, maybe two, cases where constant
evaluation cannot match non-constant evaluation, which is otherwise
required by C2Y. Takes an action to write up a proposal.
Return value vs return type
Jerome: After surveying some 137 instances of "return type" and 49
instances of "return value", I found 3 more instances of "return value"
that should be fixed as Damian did as part of his completed N3537 proposal
for complex.h. Takes an action to write a proposal.
next____() issues
Ly: Re. exceptions in the functions, tried to follow Annex F
closely, but found too many cases. Implemented a compiler switch to give
results without exception.
Editorial suggestions incl wording for cimag and creal
Jim: Proposes rewording of the real floating type that gives the
real and imaginary components.
Damian: Takes action to rework and submit.
Imaginary I macro and i suffix issues, N3390 update
Jim: There has been no update to N3390. Suggests that Rajan bring
this up at WG14 meeting, allowing CFP to move forward with a response to
the change.
Rajan: Confirms that this issue has come up with WG14, but no
action yet.
Editorial consistency in Annex F
Damian: Summarizes the long document covering many different usages
and inconsistences. Mentions Vincent's comments re. quiet NaNs. Offers many
suggestions for more concise language. Evokes a discussion about footnotes,
when a specific issue arises multiple times but is covered by a footnote in
just its first instnace.
Jim: Propose splitting the large effort into blocks of manageable
size for the group, not to mention presentation to WG14. (1) Syntax for
special cases; (2) Use of qualifiers on variables, as in "integer n"; (3)
How to deal with signaling NaNs; (4) F.10.1.<several> separate topics; (5)
>From F.10.7 on, look at those topics farther on.
Jim: Signaling NaNs were nearly removed from 60559 in 2008. Let's
find language to handle them concisely, in one place, in C2Y. Proposes
action with Damian to take Annex F a step further.
Should pragmas be tied to headers?
Jim: Shows how the language of usage of pragmas is repeated
verbatim for each of the STDC pragmas. Suggests that a better organization
might be a single treatment of the STDC pragma mechanism, with each
individual pragma discussed apart from the header file that currently
provides general context but no definitions essential to the pragma. Takes
an action with Rajan to investigate.
frexp() in Annex F
Damian: Let's skip this for now. It will arise in the Annex F work.
More complex functions, e.g. ccbrt, csinpi
Jim: We will await email from Ly.
CFP list of post C23 issues
Jim: Presents a taxonomy of issues allowing CFP to organize tasks.
Given the limited time of monthly meetings and the size of the CFP group,
we need to choose our tasks well.
Jerome: Suggests adding 60559 to the list of originators of tasks.
*Other issues:*
Carried over
*Adjournment*
10:02 AM PDT
-----------------------------------------------
*Action items to be carried over:*
Ly: Send email on expanded complex functions.
*New action items:*
Jim: Add WG14 issues to the agenda.
Jim: Add discussion of Ly's note about fixed point arithmetic to the
agenda.
Damian: Send a note about the improved use of "precision" in 754.
Rajan: Raise TS issue at August meeting.
Jim: Follow up with JeanHeyd on the [[reproducible]] example.
Jim: Write up proposal re. constant evaluation semantics
Jerome: Write a proposal re. return type vs. value with changes along
the lines of N3537 for complex.h
Damian: Rework the creal and cimag proposal with updated wording from
the discussion.
Rajan: Bring the N3390 update issue up at WG14 meeting.
Damian & Jim: Break the Annex F work into individual items for the
committee.
Jim & Rajan: Investigate the change to decouple pragmas from headers.
*Discussion issues to be carried over:*
Reproducible floating point
Missing quotes in C standard
Magnitude or Absolute Value or |x| or ....
Rounding Direction Mode or Rounding Direction or Rounding Mode
hypot(x, NAN) in Annex F
canonicalise() in AnnexF - F.10.9.7
F10.1#14
F.10.9.1 - nextafter(x, y)
Preferred Style in Annex F - fromfp/fromfpx or nearbyint/rint
hypot(x, y) - C2Y F.10.5.4
Preferred quantum exponent
*Signoff*
Respectfully submitted.
-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/20250612/05ebd521/attachment-0001.htm>
More information about the cfp-interest
mailing list