https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127734
--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> ---
When testing on x86-64, a preexisting testcase fails with:
c-c++-common/gomp/declare-variant-5.c
:12:30: error: cannot convert 'float' to '__v4sf' {aka '__vector(4) float'}
:13:30: error: cannot convert 'float' to '__v8sf' {aka '__vector(8) float'}
:16:30: error: cannot convert 'int' to '__v4si' {aka '__vector(4) int'}
For C++ but not for C. There is clearly something amiss for C++.
However, this issue will be dealt with in PR 127850.
