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

typedef uint8_t leg_t;
typedef uint16_t dleg_t;

#if 0
static void mp_sqr( leg_t *result, leg_t const * a, int legs )
{
  while( legs-- ) {
    leg_t *offs = result+2*legs+1;
    leg_t fac = a[legs];
    int leg = legs;
    dleg_t acc = (dleg_t)*offs + (dleg_t)fac * fac;

    *(offs--) = (leg_t)acc;
    acc >>= 8*sizeof(leg_t);

    while( leg-- )
    {
      dleg_t subresult = fac * a[leg];
      int carry = !!(subresult & (1<<(8*sizeof(dleg_t)-1)));

      subresult <<= 1;
      acc += subresult + *offs;

      *(offs--) = (leg_t)acc;
      acc >>= 8*sizeof(leg_t);
      acc += carry;
    }

    while( acc )
    {
      acc += *offs;
      *(offs--) = (leg_t)acc;
      acc >>= 8*sizeof(leg_t);
    }
  }
}
#else
static void mp_sqr( leg_t *result, leg_t const * a, int legs )
{
int i;
  while( legs-- ) {
    leg_t *offs = result+2*legs+1;
    leg_t fac = a[legs];
    int leg = legs,i;
    dleg_t acc = (dleg_t)*offs + (dleg_t)fac * (dleg_t)fac;

    *(offs--) = (leg_t)acc;
    acc >>= 8*sizeof(leg_t);

    while( leg-- )
    {
      dleg_t subresult = (dleg_t)fac * (dleg_t)a[leg];
      int carry = 1 & ( subresult >> (16*sizeof(leg_t)-1));

      subresult ^= carry << (16*sizeof(leg_t)-1);

      acc += 2 * subresult + (dleg_t)*offs;
      *(offs--) = (leg_t)acc;

      acc >>= 8*sizeof(leg_t);
      acc += (dleg_t)carry << 8*sizeof(leg_t);

printf( "%04X %04X %d\n", acc, subresult, carry );

    }
  for(i=0;i<6;++i)
    printf( "%02X", result[i] );

  putchar(10);
    while( acc )
    {
      acc += (dleg_t)*offs;
      *(offs--) = (leg_t)acc;
      acc >>= 8*sizeof(leg_t);
    }
  }
  for(i=0;i<6;++i)
    printf( "%02X", result[i] );
  putchar(10);
 }
#endif

int main() {
  leg_t a[3] = { 0xff, 0xff, 0xa1 };
  leg_t r[6] = {0,0,0,0,0,0};
  int i;

  mp_sqr( r, a, 3 );
  for(i=0;i<6;++i)
    printf( "%02X", r[i] );
  putchar(10);
}