/* { dg-do run } */
/* { dg-options "-O2 -ftree-loop-distribution -fdump-tree-ldist-details -mzarch -march=z13" } */

#include <cstring>
#include <cassert>
#include <cstdint>
#include <iostream>

template <typename T, T pattern>
__attribute__((noinline,noclone))
T* rawmemchr (T *s)
{
  while (*s != pattern)
    ++s;
  return s;
}

template <typename T, T pattern>
void doit()
{
  T *buf = new T[4096 * 2];
  assert (buf != NULL);
  memset (buf, 0xa, 4096 * 2);
  // ensure q is 4096-byte aligned
  T *q = buf + (4096 - ((uintptr_t)buf & 4095));
  T *p;

  // unaligned + block boundary + 1st load
  p = (T *) ((uintptr_t)q - 8);
  p[2] = pattern;
  assert ((rawmemchr<T, pattern> (&p[0]) == &p[2]));
  p[2] = (T) 0xaaaaaaaa;

  // unaligned + block boundary + 2nd load
  p = (T *) ((uintptr_t)q - 8);
  p[6] = pattern;
  assert ((rawmemchr<T, pattern> (&p[0]) == &p[6]));
  p[6] = (T) 0xaaaaaaaa;



  // unaligned + 1st load
  q[5] = pattern;
  assert ((rawmemchr<T, pattern>(&q[2]) == &q[5]));
  q[5] = (T) 0xaaaaaaaa;

  // unaligned + 2nd load
  q[14] = pattern;
  assert ((rawmemchr<T, pattern>(&q[2]) == &q[14]));
  q[14] = (T) 0xaaaaaaaa;

  // unaligned + 3rd load
  q[19] = pattern;
  assert ((rawmemchr<T, pattern>(&q[2]) == &q[19]));
  q[19] = (T) 0xaaaaaaaa;

  // unaligned + 4th load
  q[25] = pattern;
  assert ((rawmemchr<T, pattern>(&q[2]) == &q[25]));
  q[25] = (T) 0xaaaaaaaa;



  // aligned + 1st load
  q[5] = pattern;
  assert ((rawmemchr<T, pattern>(&q[0]) == &q[5]));
  q[5] = (T) 0xaaaaaaaa;

  // aligned + 2nd load
  q[14] = pattern;
  assert ((rawmemchr<T, pattern>(&q[0]) == &q[14]));
  q[14] = (T) 0xaaaaaaaa;

  // aligned + 3rd load
  q[19] = pattern;
  assert ((rawmemchr<T, pattern>(&q[0]) == &q[19]));
  q[19] = (T) 0xaaaaaaaa;

  // aligned + 4th load
  q[25] = pattern;
  assert ((rawmemchr<T, pattern>(&q[0]) == &q[25]));
  q[25] = (T) 0xaaaaaaaa;

  delete buf;
}

int main(void)
{
  doit<uint8_t, 0xde> ();
  doit<int8_t, (int8_t)0xde> ();
  doit<uint16_t, 0xdead> ();
  doit<int16_t, (int16_t)0xdead> ();
  doit<uint32_t, 0xdeadbeef> ();
  doit<int32_t, (int32_t)0xdeadbeef> ();
  return 0;
}

/* { dg-final { scan-tree-dump-times "generated rawmemchrqi" 2 "ldist" } } */
/* { dg-final { scan-tree-dump-times "generated rawmemchrhi" 2 "ldist" } } */
/* { dg-final { scan-tree-dump-times "generated rawmemchrsi" 2 "ldist" } } */