[PATCH] PR c++/90532 Ensure __is_constructible(T[]) is false

[Jonathan Wakely <jwakely at redhat dot com>]

An array of an unknown bound is an incomplete type, so no object of such
a type can be constructed. This means __is_constructible should always
be false for an array of unknown bound.

This patch also changes the std::is_default_constructible trait to use
std::is_constructible, which now gives the right answer for arrays of
unknown bound.

gcc/cp:

	PR c++/90532 Ensure __is_constructible(T[]) is false
	* method.c (is_xible_helper): Return error_mark_node for construction
	of an array of unknown bound.

gcc/testsuite:

	PR c++/90532 Ensure __is_constructible(T[]) is false
	* g++.dg/ext/90532.C: New test.

libstdc++-v3:

	PR c++/90532 Ensure __is_constructible(T[]) is false
	* include/std/type_traits (__do_is_default_constructible_impl)
	(__is_default_constructible_atom, __is_default_constructible_safe):
	Remove.
	(is_default_constructible): Use is_constructible.
	* testsuite/20_util/is_constructible/value.cc: Check int[] case.
	* testsuite/20_util/is_default_constructible/value.cc: Likewise.
	* testsuite/20_util/is_trivially_constructible/value.cc: Likewise.
	* testsuite/20_util/is_trivially_default_constructible/value.cc:
	Likewise.

Tested powerpc64le-linux. Is the G++ change OK for trunk?

This fixes a regression on gcc-9-branch and gcc-8-branch, because
changing std::is_constructible to use __is_constructible caused it to
start giving the wrong answer for arrays of unknown bound. OK for the
branches too, after appropriate testing?

Attachment: patch.txt
Description: Text document