[cfp-interest 3824] JTC1/SC22/WG14/CFP 2026/03/18 Meeting notes
Jerome Coonen
jcoonen at gmail.com
Thu Mar 19 12:45:10 PDT 2026
2026/02/18: 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, Fred Tydeman, David Hough, Ariel Burton
*Updated agenda plus new items*
https://cfp-wiki.esi.com.au/pub/CFP/WebHome/CFP%20meeting%20agenda-20260318-update4.pdf?_t=1773865597
–
These minutes should be read alongside the agenda, in this case a rare 4th
update, with its many reference links.
*Previous meeting notes*
https://mailman.oakapple.net/pipermail/cfp-interest/2026-February/003807.html
2026/02/18 Meeting notes
*Study group logistics*
Next meeting: 15 April 2026, 8:00 AM PDT/3:00 PM UTC
ISO Zoom teleconference
Please notify the group if this time slot does not work.
David: Computer centers in Brazil rent time to bad players who attack
system around the world, often for "protection" money. Penniless though it
may be, our reflector has suffered collateral damage in the chaos.
Damian: Be aware that when you upload an updated file (with the same
name) to the wiki, the upload may be slow after the Nth copy, for N seven
to nine. This is due in part to the automated backup mechanism.
*C documents*
The latest C2Y draft is N3783 Jan. 2026
https://www.open-std.org/jtc1/sc22/wg14/www/docs/n3783.pdf
C23 has been published ISO/IEC 9899, available for purchase.
https://www.iso.org/standard/82075.html
*IEEE 754 liaison*
Damian: Trying nail Round-To-Odd and language. There's a new
probabilistic rounding proposal. Propagation of NaNs is a continuing issue.
At David Keaton's suggestion, Damian is proposing to be liaison between 754
and WG14/CFP.
*C++ liaison*
Joshua: Next meeting is the week of 23 March. Four papers are of
interest:
1) http://wg21.link/p3938 -- clarifies the wording, like "infinity"
and "NaN", needed to talk about the values floating point types might take.
2) http://wg21.link/p3899 -- clarify the behavior on floating point
overflow.
3) http://wg21.link/p3935 -- update <cmath.h> to align with C23
<math.h>.
4) http://wg21.link/p3864 -- Use the prefix "cr_" to add
correctly-rounded math functions to the library.
Joshua & Group: These functions use "to nearest with ties to even"
rounding regardless of any prevailing rounding direction modes.
*WG14 update*
Rajan: Reviews detailed email summary (linked in the agenda). Most
items pass with just 5 minutes of discussion apiece: preferred quantum
exponent (N3703); clean up frexp, et al. (N3704); range bounds for math
functions (with editorial changes) (N3731); floating expressions in
translation time (N3732).
Rajan: In our proposals until now, we use a red strikethrough typeface
for deletion and green for insertion. WG14 requests that we also underline
insertions. The issue is red-green color blindness.
Group: Perhaps use blue for insertion. Will try blue underline in new
proposals this month.
Rajan: The contentious issue, which took most of the CFP time in the
meeting, was the proposal to refine the language of error reporting (3731).
The proposal italicizes "occurs" to make precise the notion that what we
generally view as a mathematical exceptional case "occurs as an error" in
the C environment in just specific circumstances. WG14 said that defining
that usage of "occur(s)" might affect other usage of the word in the
standard. The motion passed, with the removal of the italics for "occurs".
Jerome: Takes action to revisit this proposal and this usage.
*TS-4 and TS-5 revisions*
Jim: Joseph Meyers has emailed about the error in TS-4 that CFP
discovered. It acknowledges the issue and the proposed fix, though there is
no planned action yet.
*News*
None
*Carryover action items from last meeting *
Group: Think about where CFP is headed.
Carry over
*Action items from last meeting *
Damian: Investigate the current state of default rounding(s) in 60559
and whatever is being discussed in the committee about changes, especially
in decimal.
Closed -- there is the option to have the default rounding break
ties away from 0.
Damian: Arrange to take the email archive from David. Discuss the
status going forward.
Carry over
Damian: Document the wiki backup protocol.
Done
Jim & Jerome: Draft proposal re. double_t semantics and send a note to
Joseph about the proposed solution.
Done
Jerome & Damian: Write a proposal to repair fromfp and enhance Table
F.2.
Carry over
Jim & Damian: Look at the language around rounding direction in
Damian's Annex F paper and make a proposal to CFP about how to go forward.
Done
*Discussion of issues*
Conflicting requirements for double_t
Rajan: Editorially, prefer to have underscore along with color for
additions.
Group: Good to submit. Action.
Annex F special cases
Damian: Reviews many special cases, including some worked worked
out with Fred. Will use square brackets for rationale items interleaved
with proposal items.
Rajan & Group: In such a long proposal, burying the rationale at
the end does not help the reader.
Jim: In frexp, "value" is the parameter, not "x".
Jim: modf is like frexp, not bound to a 60559 function, so could
signal “inexact” or “underflow” unless specified to not signal any
exceptions. The parameter "value" needs to be rationalized, too.
Damian: Amid the pages of editorial changes, there are just two
technical changes: scalbn with a signaling NaN and the addition of a
compoundn special case Fred noticed was missing.
Jim & Rajan: The changes are so small that we can just have one
paper with both (2) technical changes and (many) editorial changes.
Jim: Suggests making the uniform syntax more explicit at the start
of the proposal, giving the reader a better sense of the broad cleanup
undertaken in the proposal.
Rajan: WG14 doesn't like long papers, but the items are small here
so it may get through.
Jim: Encourages group to spend the time to review the next draft in
detail. The challenge here is to get it all right -- and to convince WG14
that we have it all right.
Rounding direction language
Jim: Reviews his paper with Damian presenting terms to be used
going forward.
Damian: Already in use in Annex F documents.
Jim: There is no proposal to update the whole standard with this
language.
Group: Locution "standard/binary and decimal types" is a bit
misleading.
Rajan & Group: Good clarification
*Other issues*
None
*Adjournment*
10:03 AM PDT
-----------------------------------------------
*Action items to be carried over*
Group: Think about where CFP is headed.
Jerome & Damian: Write a proposal to repair fromfp and enhance Table
F.2.
*New action items*
Jim: Add agenda item to fix range bounds problem for complex fcns (as
done for real functions in N3731).
'
Jerome: Research usage of "occur(s)" and how it applies to proposal
n3737.
Jim: Submit doble_t proposal.
Damian: In the Annex F special cases paper, add the function header to
the code examples (which will clarify the usage of the parameters).
Jim & Damian: Review the wording of F.10.4.7 and F.10.4.9 and propose
any changes.
Damian: Update the Annex F Special Cases paper as technical plus
editorial.
Damian & Jerome: Send a note to CFP about the usage of "current
rounding direction".
Damian: Draft a proposal for a new table (F.3) in Annex F mapping the
rounding direction attributes from 60559 to the language of C2Y.
*Carryover discussion items*
Fixes to symbols: minus signs and hyphens, etc.
Complex and signaling NaNs
INFINITY and _Float16
Function result vs return value
CFP future
Editorial issues F.2.2
Editorial issues F.3, etc.
Editorial issues F.10.1
*Signoff*
Respectfully submitted.
-Jerome Coonen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.oakapple.net/pipermail/cfp-interest/attachments/20260319/1dcecbfa/attachment-0001.htm>
More information about the cfp-interest
mailing list