Save _normalize_weak calls in group add methods

Author: peterdettman

src/group_impl.h

@@ -491,12 +491,12 @@ static void secp256k1_gej_add_ge_var(secp256k1_gej *r, const secp256k1_gej *a, c
     r->infinity = 0;
 
     secp256k1_fe_sqr(&z12, &a->z);
-    u1 = a->x; secp256k1_fe_normalize_weak(&u1);
+    u1 = a->x;
     secp256k1_fe_mul(&u2, &b->x, &z12);
-    s1 = a->y; secp256k1_fe_normalize_weak(&s1);
+    s1 = a->y;
     secp256k1_fe_mul(&s2, &b->y, &z12); secp256k1_fe_mul(&s2, &s2, &a->z);
-    secp256k1_fe_negate(&h, &u1, 1); secp256k1_fe_add(&h, &u2);
-    secp256k1_fe_negate(&i, &s1, 1); secp256k1_fe_add(&i, &s2);
+    secp256k1_fe_negate(&h, &u1, 6); secp256k1_fe_add(&h, &u2);
+    secp256k1_fe_negate(&i, &s1, 4); secp256k1_fe_add(&i, &s2);
     if (secp256k1_fe_normalizes_to_zero_var(&h)) {
         if (secp256k1_fe_normalizes_to_zero_var(&i)) {
             secp256k1_gej_double_var(r, a, rzr);
@@ -559,12 +559,12 @@ static void secp256k1_gej_add_zinv_var(secp256k1_gej *r, const secp256k1_gej *a,
     secp256k1_fe_mul(&az, &a->z, bzinv);
 
     secp256k1_fe_sqr(&z12, &az);
-    u1 = a->x; secp256k1_fe_normalize_weak(&u1);
+    u1 = a->x;
     secp256k1_fe_mul(&u2, &b->x, &z12);
-    s1 = a->y; secp256k1_fe_normalize_weak(&s1);
+    s1 = a->y;
     secp256k1_fe_mul(&s2, &b->y, &z12); secp256k1_fe_mul(&s2, &s2, &az);
-    secp256k1_fe_negate(&h, &u1, 1); secp256k1_fe_add(&h, &u2);
-    secp256k1_fe_negate(&i, &s1, 1); secp256k1_fe_add(&i, &s2);
+    secp256k1_fe_negate(&h, &u1, 6); secp256k1_fe_add(&h, &u2);
+    secp256k1_fe_negate(&i, &s1, 4); secp256k1_fe_add(&i, &s2);
     if (secp256k1_fe_normalizes_to_zero_var(&h)) {
         if (secp256k1_fe_normalizes_to_zero_var(&i)) {
             secp256k1_gej_double_var(r, a, NULL);
@@ -649,17 +649,17 @@ static void secp256k1_gej_add_ge(secp256k1_gej *r, const secp256k1_gej *a, const
      */
 
     secp256k1_fe_sqr(&zz, &a->z);                       /* z = Z1^2 */
-    u1 = a->x; secp256k1_fe_normalize_weak(&u1);        /* u1 = U1 = X1*Z2^2 (1) */
+    u1 = a->x;                                          /* u1 = U1 = X1*Z2^2 (6) */
     secp256k1_fe_mul(&u2, &b->x, &zz);                  /* u2 = U2 = X2*Z1^2 (1) */
-    s1 = a->y; secp256k1_fe_normalize_weak(&s1);        /* s1 = S1 = Y1*Z2^3 (1) */
+    s1 = a->y;                                          /* s1 = S1 = Y1*Z2^3 (4) */
     secp256k1_fe_mul(&s2, &b->y, &zz);                  /* s2 = Y2*Z1^2 (1) */
     secp256k1_fe_mul(&s2, &s2, &a->z);                  /* s2 = S2 = Y2*Z1^3 (1) */
-    t = u1; secp256k1_fe_add(&t, &u2);                  /* t = T = U1+U2 (2) */
-    m = s1; secp256k1_fe_add(&m, &s2);                  /* m = M = S1+S2 (2) */
+    t = u1; secp256k1_fe_add(&t, &u2);                  /* t = T = U1+U2 (7) */
+    m = s1; secp256k1_fe_add(&m, &s2);                  /* m = M = S1+S2 (5) */
     secp256k1_fe_sqr(&rr, &t);                          /* rr = T^2 (1) */
-    secp256k1_fe_negate(&m_alt, &u2, 1);                /* Malt = -X2*Z1^2 */
-    secp256k1_fe_mul(&tt, &u1, &m_alt);                 /* tt = -U1*U2 (2) */
-    secp256k1_fe_add(&rr, &tt);                         /* rr = R = T^2-U1*U2 (3) */
+    secp256k1_fe_negate(&m_alt, &u2, 1);                /* Malt = -X2*Z1^2 (2) */
+    secp256k1_fe_mul(&tt, &u1, &m_alt);                 /* tt = -U1*U2 (1) */
+    secp256k1_fe_add(&rr, &tt);                         /* rr = R = T^2-U1*U2 (2) */
     /** If lambda = R/M = 0/0 we have a problem (except in the "trivial"
      *  case that Z = z1z2 = 0, and this is special-cased later on). */
     degenerate = secp256k1_fe_normalizes_to_zero(&m) &
@@ -670,8 +670,8 @@ static void secp256k1_gej_add_ge(secp256k1_gej *r, const secp256k1_gej *a, const
      * non-indeterminate expression for lambda is (y1 - y2)/(x1 - x2),
      * so we set R/M equal to this. */
     rr_alt = s1;
-    secp256k1_fe_mul_int(&rr_alt, 2);       /* rr = Y1*Z2^3 - Y2*Z1^3 (2) */
-    secp256k1_fe_add(&m_alt, &u1);          /* Malt = X1*Z2^2 - X2*Z1^2 */
+    secp256k1_fe_mul_int(&rr_alt, 2);       /* rr = Y1*Z2^3 - Y2*Z1^3 (8) */
+    secp256k1_fe_add(&m_alt, &u1);          /* Malt = X1*Z2^2 - X2*Z1^2 (8) */
 
     secp256k1_fe_cmov(&rr_alt, &rr, !degenerate);
     secp256k1_fe_cmov(&m_alt, &m, !degenerate);
@@ -686,7 +686,7 @@ static void secp256k1_gej_add_ge(secp256k1_gej *r, const secp256k1_gej *a, const
      * zero (which is "computed" by cmov). So the cost is one squaring
      * versus two multiplications. */
     secp256k1_fe_sqr(&n, &n);
-    secp256k1_fe_cmov(&n, &m, degenerate);              /* n = M^3 * Malt (2) */
+    secp256k1_fe_cmov(&n, &m, degenerate);              /* n = M^3 * Malt (5) */
     secp256k1_fe_sqr(&t, &rr_alt);                      /* t = Ralt^2 (1) */
     secp256k1_fe_mul(&r->z, &a->z, &m_alt);             /* r->z = Malt*Z (1) */
     infinity = secp256k1_fe_normalizes_to_zero(&r->z) & ~a->infinity;
@@ -698,8 +698,8 @@ static void secp256k1_gej_add_ge(secp256k1_gej *r, const secp256k1_gej *a, const
     secp256k1_fe_mul_int(&t, 2);                        /* t = 2*x3 (2) */
     secp256k1_fe_add(&t, &q);                           /* t = 2*x3 - Q: (4) */
     secp256k1_fe_mul(&t, &t, &rr_alt);                  /* t = Ralt*(2*x3 - Q) (1) */
-    secp256k1_fe_add(&t, &n);                           /* t = Ralt*(2*x3 - Q) + M^3*Malt (3) */
-    secp256k1_fe_negate(&r->y, &t, 3);                  /* r->y = Ralt*(Q - 2x3) - M^3*Malt (4) */
+    secp256k1_fe_add(&t, &n);                           /* t = Ralt*(2*x3 - Q) + M^3*Malt (6) */
+    secp256k1_fe_negate(&r->y, &t, 6);                  /* r->y = Ralt*(Q - 2x3) - M^3*Malt (7) */
     secp256k1_fe_normalize_weak(&r->y);
     secp256k1_fe_mul_int(&r->x, 4);                     /* r->x = X3 = 4*(Ralt^2-Q) */
     secp256k1_fe_mul_int(&r->y, 4);                     /* r->y = Y3 = 4*Ralt*(Q - 2x3) - 4*M^3*Malt (4) */