To: J3 Members J3/17-117r2
From: Van Snyder & Bill Long
Subject: Comments concerning atomic subroutines
References: 17-007
Date: 2017 February 14
1. Introduction
===============
Several of the atomic subroutines in 16.9.20 - 16.9.30 include a
requirement of the form "The values of VALUE and ATOM + VALUE shall be
representable in ATOMIC_INT_KIND."
These are nonsense because ATOMIC_INT_KIND is a named constant, if the
phrase actually refers to the ATOMIC_INT_KIND defined in the intrinsic
module ISO_FORTRAN_ENV. This interpretation of ATOMIC_INT_KIND is
already stated earlier in the same paragraph.
2. Edits
========
[360:19] In 16.9.20 ATOMIC_ADD, the description of VALUE, replace "in
ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[360:34] In 16.9.21 ATOMIC_AND, the description of VALUE, replace "in
ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[362:5] In 16.9.24 ATOMIC_FETCH_ADD, the description of VALUE, replace
"in ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[362:23] In 16.9.25 ATOMIC_FETCH_AND, the description of VALUE,
replace "in ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[362:41] In 16.9.26 ATOMIC_FETCH_OR, the description of VALUE, replace
"in ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[363:15] In 16.9.27 ATOMIC_FETCH_XOR, the description of VALUE,
replace "in ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[363:33] In 16.9.28 ATOMIC_OR, the description of VALUE, replace "in
ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".
[364:21] In 16.9.30 ATOMIC_XOR, the description of VALUE, replace "in
ATOMIC_INT_KIND" with "in kind ATOMIC_INT_KIND".