Comment by saynsedit

Nah, correct solution is simply to use memcpy(), works on all compilers, all platforms, all versions, with SSE and with any flags specified:

  #include <stdlib.h>
  #include <stdint.h>

  uint64_t sum (char *p, size_t nwords)
  {
      uint64_t res = 0;
      size_t i;
      for (i = 0; i < nwords; i += 8) {
        uint64_t tmp;
        memcpy(&tmp, &p[i], sizeof(tmp));
        res += tmp;
      }
      return res;
  }