[Bug tree-optimization/102462] vectorization breaks diagnostic for array out of bound detect.

[crazylht at gmail dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102462

--- Comment #3 from Hongtao.liu <crazylht at gmail dot com> ---
case3:
struct A1
{
  char n;
  char a[1];                    // { dg-message "destination object" "note" }
};
void sink (void*);
struct A1 a1i_1 = { 0, { 1 } };

void ga1i_1 (void)
{
  a1i_1.a[0] = 0;
  a1i_1.a[1] = 1;               // { dg-warning "\\\[-Wstringop-overflow" }
  a1i_1.a[2] = 2;               // { dg-warning "\\\[-Wstringop-overflow" }

  struct A1 a = { 0, { 1 } };
  a.a[0] = 1;
  a.a[1] = 2;                   // { dg-warning "\\\[-Wstringop-overflow" }
  a.a[2] = 3;                   // { dg-warning "\\\[-Wstringop-overflow" }
  sink (&a);
}

I think i was wrong, there's no case3, case3 is just case2, part of access
inbound, part is not, now the warning message is from struct A1 a = { 0, { 1 }
}; which is recorded as the vectorized stmt lineno

  [case3.c:15:13] MEM <vector(4) char> [(char *)&a] = { 0, 1, 2, 3 };