diff options
| author | Jason Merrill <jason@redhat.com> | 2021-05-19 21:13:43 -0400 |
|---|---|---|
| committer | Jason Merrill <jason@redhat.com> | 2021-05-20 16:59:17 -0400 |
| commit | 885035eacb36b5bf1aa3b0d05f675ab89665d7be (patch) | |
| tree | ff4cd9403ff68ee97e53d5ac5d48d23502412479 /gcc/collect2.c | |
| parent | 84fd1b5dff70cd74aee7e8b18f66959d8b8e1ce7 (diff) | |
| download | gcc-885035eacb36b5bf1aa3b0d05f675ab89665d7be.zip gcc-885035eacb36b5bf1aa3b0d05f675ab89665d7be.tar.gz gcc-885035eacb36b5bf1aa3b0d05f675ab89665d7be.tar.bz2 | |
c++: designators in single-element init lists
While looking at PR100489, it occurred to me that places that currently
use an initializer-list with a single element to initialize an object of the
same type shouldn't do that if the element has a designator.
gcc/cp/ChangeLog:
* call.c (reference_binding): Check for designator.
(implicit_conversion_1, build_special_member_call): Likewise.
* decl.c (reshape_init_r): Likewise.
* pt.c (do_class_deduction): Likewise.
* typeck2.c (digest_init_r): Likewise.
gcc/testsuite/ChangeLog:
* g++.dg/cpp2a/desig19.C: New test.
Diffstat (limited to 'gcc/collect2.c')
0 files changed, 0 insertions, 0 deletions