From 6ee99d6a8279d0d1d3e7ed7a6527d051f7cd742d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=BD=D0=B0=D0=B1?= Date: Tue, 17 May 2022 20:26:25 +0200 Subject: [PATCH] Note what happened at the meeting; add draft of p1 --- README.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/README.md b/README.md index 9064549..b0d0a4b 100644 --- a/README.md +++ b/README.md @@ -4,3 +4,11 @@ PhD [put me up](//twitter.com/__phantomderp/status/1468040630932750343) to writi ## [PDFs](//git.sr.ht/~nabijaczleweli/wg14/tree/output/item/README.md) ### [*n2878*: `nsec_t && timespec::tv_nsec`](//srhtcdn.githack.com/~nabijaczleweli/wg14/blob/output/n2878.pdf) +Discussed on [17.05.2022](//www.open-std.org/jtc1/sc22/wg14/www/docs/n2985.htm); verdict: delayed; quoth PhD: + - implementation-defined **SIGNED** integer type (must be signed) + - talk to Nick Cannon (Austin Group / POSIX) about timespec change since POSIX owns it + - should not have a name (nsec_t is used in user code, we could break people) + - Denied for C23, but can be allowed for C2Y/C3a (whenever the hell that is) + +### [*n????* (p1): `timespec::tv_nsec`](//srhtcdn.githack.com/~nabijaczleweli/wg14/blob/output/N2878P1.pdf) +`nsec_t` -> `/* see below */` + "shall be of an implementation-defined signed integer type capable of…" -- 2.38.5