I’m getting similar findings:

  #include <time.h>
  #include <stdio.h>
  #include <stdint.h>

  int main() {
        struct timespec foo;
        int z;
        uint8_t buffer[512];

        for(z=0;z<128;z++) {
                clock_gettime(CLOCK_REALTIME,&foo);
                buffer[z * 4] = (foo.tv_nsec >> 24) & 0xff;
                buffer[z * 4 + 1] = (foo.tv_nsec >> 16) & 0xff;
                buffer[z * 4 + 2] = (foo.tv_nsec >> 8) & 0xff;
                buffer[z * 4 + 3] = (foo.tv_nsec) & 0xff;
        }
        for(z=0;z<512;z++) {
                printf("%02x ",buffer[z]);
                if(z % 16 == 15) {puts("");}
        }
        return 0;
  }

(code is public domain)

Here, we see, running it on Windows, at least 1 but of entropy per clock_gettime() call. For people who argue kernel entropy is somehow more secure, perhaps they should become familiar with how kernels before Linux 5.6 or so on some devices had issues where (u)random wouldn’t provide enough entropy to be really secure (people would use haveged to make sure they had enough entropy).