How to improve the performance of this code? - c

Question

I'm writing a program which operates with arrays of Big Integer Numbers and does basic operations.
I'm worried about the performance of my code. It executes in 47 seconds when compiled with:
gcc -Ofast -funroll-all-loops -ftree-vectorize -fopt-info-vec -g -lm $1 -o ${2}.Ofast
I thought the key to solving my problem is changing the data type of my arrays to unsigned long long, but when I do the result is different.
Any suggestion is welcome, even if it changes my whole program or data types as long as it doesn't change the outcome of my program.
Also, I have been able to visualize that my performance problem is mainly in the LongNumAddition and LongNumAddDigit functions, how can I improve my code? Thanks for your attention.
#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include <string.h>
#include <limits.h>
// Variable used to generate pseudo-random numbers
unsigned int seed;
unsigned int temp;
unsigned int var1 = 214013;
unsigned int var2 = 2531011;
#define val13 13
#define ten 10
// Function to generate pseudo-random numbers
inline int myRandom() {
temp = var1*seed;
seed = temp + var2;
return (seed>>val13);
}
void LongNumInit( uint8_t *L, size_t N )
{
for(size_t i = 0; i < N; ++i){
L[i] = myRandom() % 10;
}
}
void LongNumPrint( uint8_t *L, size_t N, uint8_t *Name )
{
printf("%s:", Name);
for ( size_t i=N; i>0;--i )
{
printf("%d", L[i-1]);
}
printf("\n");
}
void LongNumSet( uint8_t *L, size_t N){
memset(L,0,N);
}
void LongNumCopy( const uint8_t *Vin, uint8_t *Vout, size_t N )
{
memcpy(Vout,Vin,N);
}
uint8_t LongNumAddition( uint8_t *Vin1, uint8_t *Vin2,uint8_t *Vout, size_t N)
{
uint8_t CARRY = 0;
for ( size_t i=0; i< N; ++i )
{
Vout[i] = Vin1[i] + Vin2[i] + CARRY;
CARRY = (Vout[i] > 9);
if(CARRY){
Vout[i] -= ten;
}
}
return CARRY;
}
uint8_t LongNumAddDigit( uint8_t *V, uint8_t digit, size_t N )
{
size_t i=0;
V[0] += digit;
if ( V[0] < ten){
return 0;
}
V[0] -=ten;
// add carry, maybe iteratively for all digits
while ((++i < N) && (V[i] >= 9))
{
V[i] = 0;
}
if((i != N) && (V[i] < 9)){
V[i]++;
return 0;
}
return 1;
}
uint8_t LongNumHorizAdd( uint8_t *Vin, uint8_t *Vout, size_t N )
{
uint8_t CARRY = 0;
LongNumSet ( Vout, N);
for ( size_t i=0; i< N; ++i )
{
LongNumAddDigit ( Vout, Vin[i], N );
}
return 0; // CARRY can never be set
}
uint8_t LongNumConstMult( uint8_t *V, size_t N, uint8_t digit )
{
uint8_t CARRY = 0;
for ( size_t i=0; i< N; ++i )
{
V[i] = V[i] * digit + CARRY;
CARRY = ((u_int32_t)V[i] * (u_int32_t)0xCCCD) >> 19;
V[i] -= (CARRY << 3) + (CARRY << 1);
}
return CARRY; // may be from 0 to 9
}
void LongNumMultiply( uint8_t *Vin1, uint8_t *Vin2, uint8_t *VoutH, uint8_t *VoutL, size_t N)
{
// Create Temporal Long Integer with double size
uint8_t *TEMP = (uint8_t*) malloc(2*N*sizeof(uint8_t));
uint8_t *RES = (uint8_t*) malloc( 2*N*sizeof(uint8_t));
LongNumSet ( RES, 2*N); // Set RES to 0
for ( size_t i=0; i<N; ++i )
{
LongNumSet ( TEMP, 2*N); // Set TEMP to 0
LongNumCopy ( Vin1, TEMP+i, N ); // Copy Vin1 -> TEMP, with offset i
LongNumConstMult( TEMP, 2*N, Vin2[i] ); // TEMP * Vin2[i] -> TEMP
LongNumAddition ( TEMP, RES, RES, 2*N ); // TEMP + RES -> RES
}
// Result goes to VoutH-VoutL
LongNumCopy ( RES, VoutL, N ); // Copy RES -> VoutL
LongNumCopy ( RES+N, VoutH, N ); // Copy RES+N -> VoutH
}
int main (int argc, char **argv)
{
int i, sum1, sum2, sum3, N=10000, Rep=50;
seed = 12345;
// obtain parameters at run time
if (argc>1) { N = atoi(argv[1]); }
if (argc>2) { Rep = atoi(argv[2]); }
printf("Challenge #3: Vector size is %d. Repeat %d times\n", N, Rep);
// Create Long Nums
unsigned char *V1= (unsigned char*) malloc( N*sizeof(unsigned char) );
unsigned char *V2= (unsigned char*) malloc( N*sizeof(unsigned char) );
unsigned char *V3= (unsigned char*) malloc( N*sizeof(unsigned char) );
unsigned char *V4= (unsigned char*) malloc( N*sizeof(unsigned char) );
LongNumInit ( V1, N ); LongNumInit ( V2, N ); LongNumInit ( V3, N );
// Repeat
for (i=0; i<Rep; i++)
{
LongNumAddition ( V1, V2, V4, N );
LongNumMultiply ( V3, V4, V2, V1, N );
LongNumHorizAdd ( V1, V2, N );
LongNumAddDigit ( V3, V2[0], N );
}
// Print last 32 digits of Long Numbers
LongNumPrint( V1, 32, "V1" );
LongNumPrint( V2, 32, "V2" );
LongNumPrint( V3, 32, "V3" );
LongNumPrint( V4, 32, "V4" );
free(V1); free(V2); free(V3); free(V4);
return 0;
}

Answer 1

Use a profiler - strongly recommend kcachegrind
https://kcachegrind.github.io/html/Usage.html
first, install valgrind and kcachegrind, then compile your binary using '-g' using gcc, and then run your binary using
valgrind --tool=callgrind ./yourbinary parameters ...
then, run kcachgrind in the current directory, switch view to display line-by-line timing information, like this
https://kcachegrind.github.io/html/Shot4Large.html
this tells you which lines of your code took most of the runtime.