mask_loadu.c

#include <stdio.h>
#include <sys/mman.h>
#include <x86intrin.h>

int main() {
  void *p = (void *)0x100000000ULL;
  __m512i v, zero = {};
  p = mmap(p, 4096, PROT_READ | PROT_WRITE,
           MAP_FIXED | MAP_PRIVATE | MAP_POPULATE | MAP_ANONYMOUS, -1, 0);
  printf("p = %p\n", p);

  for (int i = 0; i < 512; i++)
    ((unsigned long long *)p)[i] = i;

  v = _mm512_mask_loadu_epi64(zero, 0x0f, p + 4096 - 32);
  printf("%llu %llu %llu %llu %llu %llu %llu %llu\n", v[0], v[1], v[2], v[3],
         v[4], v[5], v[6], v[7]);
  v = _mm512_mask_loadu_epi64(zero, 0xff, p + 4096 - 32);
  printf("%llu %llu %llu %llu %llu %llu %llu %llu\n", v[0], v[1], v[2], v[3],
         v[4], v[5], v[6], v[7]);
  return 0;
}