karatsuba

static void karatsuba_simple(const uint16_t *a_1, const uint16_t *b_1, uint16_t *result_final) {
    //Serial.println("Enter karatsuba");
     uint16_t d01[KARATSUBA_N / 2 - 1];
     uint16_t d0123[KARATSUBA_N / 2 - 1];
     uint16_t d23[KARATSUBA_N / 2 - 1];
     uint16_t result_d01[KARATSUBA_N - 1];

    // Inizializzazione delle variabili
    memset(result_d01, 0, sizeof(result_d01));
    memset(d01, 0, sizeof(d01));
    memset(d0123, 0, sizeof(d0123));
    memset(d23, 0, sizeof(d23));
    memset(result_final, 0, (2 * KARATSUBA_N - 1) * sizeof(uint16_t));



    for (int i = 0; i < KARATSUBA_N / 4; i++) {

        uint16_t acc1 = a_1[i];                      // a0
        uint16_t acc2 = a_1[i + KARATSUBA_N / 4];    // a1
        uint16_t acc3 = a_1[i + 2 * KARATSUBA_N / 4];// a2
        uint16_t acc4 = a_1[i + 3 * KARATSUBA_N / 4];// a3
        for (int j = 0; j < KARATSUBA_N / 4; j++) {

             //Nutro il wtd
             if (j == 16 ) {
                ESP.wdtFeed();
             }

            uint16_t acc5 = b_1[j];                  // b0
            uint16_t acc6 = b_1[j + KARATSUBA_N / 4];// b1

            // Utilizzare registri per accumulare i risultati
            uint16_t mul1 = OVERFLOWING_MUL(acc1, acc5);
            uint16_t mul2 = OVERFLOWING_MUL(acc2, acc6);

            result_final[i + j] += mul1;
            result_final[i + j + 2 * KARATSUBA_N / 4] += mul2;
            uint16_t acc7 = acc5 + acc6; // b01
            uint16_t acc8 = acc1 + acc2; // a01
            d01[i + j] += acc7 * acc8;

            acc7 = b_1[j + 2 * KARATSUBA_N / 4]; // b2
            acc8 = b_1[j + 3 * KARATSUBA_N / 4]; // b3
            uint16_t mul3 = OVERFLOWING_MUL(acc7, acc3);
            uint16_t mul4 = OVERFLOWING_MUL(acc8, acc4);

            result_final[i + j + 4 * KARATSUBA_N / 4] += mul3;
            result_final[i + j + 6 * KARATSUBA_N / 4] += mul4;

            uint16_t acc9 = acc3 + acc4;
            uint16_t acc10 = acc7 + acc8;
            d23[i + j] += OVERFLOWING_MUL(acc9, acc10);

            acc5 += acc7; // b02
            acc7 = acc1 + acc3; // a02
            result_d01[i + j] += OVERFLOWING_MUL(acc5, acc7);

            acc6 += acc8; // b13
            acc8 = acc2 + acc4;
            result_d01[i + j + 2 * KARATSUBA_N / 4] += OVERFLOWING_MUL(acc6, acc8);
            acc5 += acc6;
            acc7 += acc8;
            d0123[i + j] += OVERFLOWING_MUL(acc5, acc7);

        }
    }



    // 2nd last stage
    for (int i = 0; i < KARATSUBA_N / 2 - 1; i++) {
        d0123[i] -= result_d01[i] + result_d01[i + 2 * KARATSUBA_N / 4];
        d01[i] -= result_final[i] + result_final[i + 2 * KARATSUBA_N / 4];
        d23[i] -= result_final[i + 4 * KARATSUBA_N / 4] + result_final[i + 6 * KARATSUBA_N / 4];
    }




    for (int i = 0; i < KARATSUBA_N / 2 - 1; i++) {
        result_d01[i + KARATSUBA_N / 4] += d0123[i];
        result_final[i + KARATSUBA_N / 4] += d01[i];
        result_final[i + 5 * KARATSUBA_N / 4] += d23[i];
    }

    //2


    // Last stage
    for (int i = 0; i < KARATSUBA_N - 1; i++) {
        result_d01[i] -= result_final[i] + result_final[i + KARATSUBA_N];
    }

    //2

    for (int i = 0; i < KARATSUBA_N - 1; i++) {
        result_final[i + KARATSUBA_N / 2] += result_d01[i];
    }
        //Serial.println("Exit karatsuba");

}