Slight performance difference between RDSEED and RDRAND

Question

Recent Intel chips (Ivy Bridge and up) have instructions for generating (pseudo) random bits. RDSEEDoutputs the “true” random bits generated by the entropy collected from the sensor on the chip. RDRANDoutputs bits generated by a pseudo-random number generator, seeded by a true random number generator. According to Intel documentation , it RDSEEDruns slower, since entropy collection is expensive. Thus, it is RDRANDoffered as a cheaper alternative, and its output is quite safe for most cryptographic applications. (This is similar /dev/randomcompared /dev/urandomto Unix systems.)

I was interested to know the difference in performance between the two instructions, so I wrote code to compare them. To my surprise, I believe that there is practically no difference in performance . Can someone give an explanation? The following are code and system data.

Benchmark

/* Compare the performance of RDSEED and RDRAND.
 *
 * Compute the CPU time used to fill a buffer with (pseudo) random bits 
 * using each instruction.
 *
 * Compile with: gcc -mdrnd -mdseed
 */
#include <time.h>
#include <stdio.h>
#include <stdlib.h>
#include <x86intrin.h>

#define BUFSIZE (1<<24)

int main() {

  unsigned int ok, i;
  unsigned long long *rand = malloc(BUFSIZE*sizeof(unsigned long long)), 
                     *seed = malloc(BUFSIZE*sizeof(unsigned long long)); 

  clock_t start, end, bm;

  // RDRAND (the benchmark)
  start = clock();
  for (i = 0; i < BUFSIZE; i++) {
    ok  = _rdrand64_step(&rand[i]);
  }
  bm = clock() - start;
  printf("RDRAND: %li\n", bm);

  // RDSEED
  start = clock();
  for (i = 0; i < BUFSIZE; i++) {
    ok = _rdseed64_step(&seed[i]);
  }
  end = clock();
  printf("RDSEED: %li, %.2lf\n", end - start, (double)(end-start)/bm);

  free(rand);
  free(seed);
  return 0;
}

System Information

• Intel Core i7-6700 CPU @ 3.40 GHz
• Ubuntu 16.04
• gcc 5.4.0

Answer 1

You do not check the return value, so you do not know how many actual random numbers you created. With repetition, as Florian suggested , the version is RDSEEDmore than 3 times slower:

RDRAND: 1989817
RDSEED: 6636792, 3.34 

, , , RDSEED , . RDRAND, , , , .

:

  // RDRAND (the benchmark)
  start = clock();
  for (i = 0; i < BUFSIZE; i++) {
    while (!_rdrand64_step(&rand[i]))
        ;
  }
  bm = clock() - start;
  printf("RDRAND: %li\n", bm);

  // RDSEED
  start = clock();
  for (i = 0; i < BUFSIZE; i++) {
    while (!_rdseed64_step(&seed[i]))
        ;
  }
  end = clock();

Answer 2

, m7-6Y75, RDSEED ( assert (ok); s, - ). , RDRAND. ( RDRAND, , , , RDRAND .)