void
nmod_mat_mul_check(nmod_mat_t C, const nmod_mat_t A, const nmod_mat_t B)
{
long i, j, k;
mp_limb_t s0, s1, s2;
mp_limb_t t0, t1;
for (i = 0; i < A->r; i++)
{
for (j = 0; j < B->c; j++)
{
s0 = s1 = s2 = 0UL;
for (k = 0; k < A->c; k++)
{
umul_ppmm(t1, t0, A->rows[i][k], B->rows[k][j]);
add_sssaaaaaa(s2, s1, s0, s2, s1, s0, 0, t1, t0);
}
NMOD_RED(s2, s2, C->mod);
NMOD_RED3(s0, s2, s1, s0, C->mod);
C->rows[i][j] = s0;
}
}
}
void slow_way(uint64_t* r,uint64_t* s,uint64_t siz)
{
r[0]=r[1]=r[2]=0;
uint64_t i,j,z=0;
for(i=0; i<siz; i++)
{
j=2*i;
add_sssaaaaaa( r[2],r[1],r[0], r[2],r[1],r[0], 0, s[j], s[j+1] );
}
}
int
_nmod_vec_dot_bound_limbs(slong len, nmod_t mod)
{
mp_limb_t t2, t1, t0, u1, u0;
umul_ppmm(t1, t0, mod.n - 1, mod.n - 1);
umul_ppmm(t2, t1, t1, len);
umul_ppmm(u1, u0, t0, len);
add_sssaaaaaa(t2, t1, t0, t2, t1, UWORD(0), UWORD(0), u1, u0);
if (t2 != 0) return 3;
if (t1 != 0) return 2;
return (t0 != 0);
}
int main(void)
{
int i, j, result;
FLINT_TEST_INIT(state);
flint_printf("add_sssaaaaaa....");
fflush(stdout);
for (i = 0; i < 1000000; i++)
{
mp_limb_t s[3], t[3], a[3], b[3];
for (j = 0; j < 3; j++)
{
s[j] = n_randtest(state);
t[j] = n_randtest(state);
a[j] = n_randtest(state);
b[j] = n_randtest(state);
}
add_sssaaaaaa(s[2], s[1], s[0], a[2], a[1], a[0], b[2], b[1], b[0]);
mpn_add_n(t, a, b, 3);
result = ((s[2] == t[2]) && (s[1] == t[1]) && (s[0] == t[0]));
if (!result)
{
flint_printf("FAIL:\n");
flint_printf("a[2] = %wu, a[1] = %wu, a[0] = %wu\n", a[2], a[1], a[0]);
flint_printf("b[2] = %wu, b[1] = %wu, b[0] = %wu\n", b[2], b[1], b[0]);
flint_printf("s[2] = %wu, s[1] = %wu, s[0] = %wu\n", s[2], s[1], s[0]);
flint_printf("t[2] = %wu, t[1] = %wu, t[0] = %wu\n", t[2], t[1], t[0]);
abort();
}
}
FLINT_TEST_CLEANUP(state);
flint_printf("PASS\n");
return 0;
}
