简体   繁体   中英

Fibonacci numbers with GMP lib

EDIT: solved. ans var needed to be set to 0 in every iteration.That was a mistake I overlooked.A mpz_set_ui(ans,0); at the beginning of each loop solves the problem.Thanks to Oli Charlesworth for his effort!

I'm coding a fibonacci function in C using the GMP library,I use the matrix multiplication algorithm,but for some unexpected reason,I get wrong results(I strongly believe that the algorithm is right).It must have something to do with Gmp structs and inits.Any ideas?

/*********************************************************
Find n th Fibonacci number.Matrix algorithm.
+--          --+     +--     --+
|f(n+1) f(n)   |     | 1     1 |  ^n
|              |  == |         |
|f(n)   f(n-1) |     | 1     0 |
+--          --+     +--     --+
*********************************************************/

int
fastfib(mpz_t result,int n)
{


    int cons[2][2]={{1,1},{1,0}};      //"constant" matrix
    mpz_t mat[2][2];                   //actual matrix holding fibonacci numbers
    mpz_t ans;                         //holds f(n+1)   
    mpz_t return_val;                  //used for calculations
    mpz_t temp;                        //used for calculations as well

    //initialize (automatically set to zero)
    mpz_init(ans);
    mpz_init(return_val);
    mpz_init(temp);
    mpz_init(mat[0][0]);
    mpz_init(mat[0][1]);
    mpz_init(mat[1][0]);
    mpz_init(mat[1][1]);

    //start with n=1
    mpz_set_ui(mat[0][0],1);
    mpz_set_ui(mat[1][0],1);
    mpz_set_ui(mat[0][1],1);
    mpz_set_ui(mat[1][1],0);

    //some trivial cases
    if(n==0){
        mpz_set_ui(result,0);
        return 0;
    }
    if(n==1){
        mpz_set_ui(result,1);
        return 0;
    }
    if(n==2){
        mpz_set_ui(result,1);
        return 0;
    }

    n--;


    while(n>1){

        //fib[n+1]
        //ans=mat[0][0]*cons[0][0]+mat[0][1]*cons[1][0];

        mpz_set_ui(ans,0);

        mpz_mul_ui(temp,mat[0][0],cons[0][0]);
        mpz_add(ans,ans,temp);
        mpz_mul_ui(temp,mat[0][1],cons[1][0]);
        mpz_add(ans,ans,temp);

        //update matrix


        mpz_set(mat[1][1],mat[1][0]);   //mat[1][1]=mat[1][0];
        mpz_set(mat[1][0],mat[0][0]);   //mat[0][1]=mat[1][0]=mat[0][0];
        mpz_set(mat[0][1],mat[0][0]);
        mpz_set(mat[0][0],ans);         //mat[0][0]=ans;

        n--;
    }


    //clear vars
    mpz_clear(ans);
    mpz_clear(return_val);
    mpz_clear(temp);

    mpz_set(result,mat[0][0]);
    return 0;

}

Some debug info

results for n=2       // expected
2 1                   //  2 1
1 1                   //  1 1

results for n=3
5 2                   //  3 2
2 1                   //  2 1

results for n=4
12 5                  //  5 3
5 2                   //  3 2

results for n=5
29 12                 //  8 5
12 5                  //  5 3

results for n=6
70 29                 //  13 8
29 12                 //  8  5

70

solved. ans var needed to be set to 0 in every iteration.That was a mistake I overlooked.A mpz_set_ui(ans,0); at the beginning of each loop solves the problem.Thanks to Oli Charlesworth for his effort!

The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM