https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83920
--- Comment #7 from Tom de Vries <vries at gcc dot gnu.org> ---
(In reply to Tom de Vries from comment #6)
> (In reply to Tom de Vries from comment #3)
> > Likewise, reversing the if-then-else order in gemm.f90 makes the testcase
> > fail on trunk without this patch.
>
> Minimal version:
An even more minimal c version:
...
/* { dg-do run } */
extern void abort (void);
#define n 10
static void __attribute__((noinline)) __attribute__((noclone))
gemm (int beta, int *c)
{
#pragma acc parallel copy(c[0:(n * n) - 1]) num_gangs(2)
#pragma acc loop gang
for (int j = 0; j < n; ++j)
if (beta != 1)
{
#pragma acc loop vector
for (int i = 0; i < n; ++i)
c[i + (j * n)] = 0;
}
}
int
main (void)
{
int c[n * n];
c[0] = 1;
gemm (0.0, c);
if (c[0] != 0)
abort ();
}
...
Passes at O0, fails at O2.
