https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93566
--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
void
foo (int *x)
{
int c[10] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
int bar (int i) { return c[i]; }
#pragma omp parallel for reduction(+:x[:10])
for (int i = 0; i < 10; i++)
x[i] = bar (i);
}
ICEs too.
