Push last bit of Point functionality into impls. (I hope.)

src/ed25519/constants.rs

@@ -10,7 +10,7 @@ pub struct Precomp {
 
 impl Precomp
 {
-    pub fn new() -> Precomp
+    fn new() -> Precomp
     {
         Precomp {
             yplusx: FieldElement::new(),
@@ -44,8 +44,46 @@ impl Precomp
         self.yminusx.cmov(&u.yminusx, b);
         self.xy2d.cmov(&u.xy2d, b);
     }
+
+    pub fn table_select(pos: i32, b: i8) -> Precomp
+    {
+        let mut minust = Precomp::new();
+        let mut res = Precomp::zero();
+        let bnegative = negative(b);
+        let babs = b - (((-(bnegative as i8)) & b) << 1);
+        
+        res.cmov(&K25519_PRECOMP[pos as usize][0], equal(babs, 1));
+        res.cmov(&K25519_PRECOMP[pos as usize][1], equal(babs, 2));
+        res.cmov(&K25519_PRECOMP[pos as usize][2], equal(babs, 3));
+        res.cmov(&K25519_PRECOMP[pos as usize][3], equal(babs, 4));
+        res.cmov(&K25519_PRECOMP[pos as usize][4], equal(babs, 5));
+        res.cmov(&K25519_PRECOMP[pos as usize][5], equal(babs, 6));
+        res.cmov(&K25519_PRECOMP[pos as usize][6], equal(babs, 7));
+        res.cmov(&K25519_PRECOMP[pos as usize][7], equal(babs, 8));
+        minust.yplusx.overwrite_with(&res.yminusx);
+        minust.yminusx.overwrite_with(&res.yplusx);
+        minust.xy2d = -&res.xy2d;
+        res.cmov(&minust, bnegative != 0);
+        res
+    }
 }
 
+fn equal(b: i8, c: i8) -> bool
+{
+    let ub = b;
+    let uc = c;
+    let x = ub ^ uc;  /* 0: yes; 1..255: no */
+    (x == 0)
+}
+
+fn negative(b: i8) -> u8
+{
+    let mut x = b as u32;
+    x >>= 31; /* 1: yes; 0: no */
+    x as u8
+}
+
+
 /* k25519Precomp[i][j] = (j+1)*256^i*B */
 pub const K25519_PRECOMP: [[Precomp; 8]; 32] = [
     [

src/ed25519/mod.rs

@@ -149,7 +149,7 @@ impl ED25519Public {
         a.invert();
         let h_digest = eddsa_digest(signature_r, &self.public, msg);
         let h = digest_scalar(&h_digest);
-        let r = ge_double_scalarmult_vartime(&h, &a, &signature_s);
+        let r = Point2::double_scalarmult_vartime(&h, &a, &signature_s);
         let r_check = r.encode();
         signature_r.to_vec() == r_check
     }

src/ed25519/point.rs

@@ -488,43 +488,6 @@ fn addsub() {
     });
 }
 
-fn equal(b: i8, c: i8) -> bool
-{
-    let ub = b;
-    let uc = c;
-    let x = ub ^ uc;  /* 0: yes; 1..255: no */
-    (x == 0)
-}
-
-fn negative(b: i8) -> u8
-{
-    let mut x = b as u32;
-    x >>= 31; /* 1: yes; 0: no */
-    x as u8
-}
-
-fn table_select(pos: i32, b: i8) -> Precomp
-{
-    let mut minust = Precomp::new();
-    let mut res = Precomp::zero();
-    let bnegative = negative(b);
-    let babs = b - (((-(bnegative as i8)) & b) << 1);
-    
-    res.cmov(&K25519_PRECOMP[pos as usize][0], equal(babs, 1));
-    res.cmov(&K25519_PRECOMP[pos as usize][1], equal(babs, 2));
-    res.cmov(&K25519_PRECOMP[pos as usize][2], equal(babs, 3));
-    res.cmov(&K25519_PRECOMP[pos as usize][3], equal(babs, 4));
-    res.cmov(&K25519_PRECOMP[pos as usize][4], equal(babs, 5));
-    res.cmov(&K25519_PRECOMP[pos as usize][5], equal(babs, 6));
-    res.cmov(&K25519_PRECOMP[pos as usize][6], equal(babs, 7));
-    res.cmov(&K25519_PRECOMP[pos as usize][7], equal(babs, 8));
-    minust.yplusx.overwrite_with(&res.yminusx);
-    minust.yminusx.overwrite_with(&res.yplusx);
-    minust.xy2d = -&res.xy2d;
-    res.cmov(&minust, bnegative != 0);
-    res
-}
-
 impl Point {
     /* h = a * B
      * where a = a[0]+256*a[1]+...+256^31 a[31]
@@ -557,7 +520,7 @@ impl Point {
   
       let mut h = Point::zero();
       for i in &[1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63] {
-          t = table_select(*i / 2, e[*i as usize]);
+          t = Precomp::table_select(*i / 2, e[*i as usize]);
           r = &h + &t;
           h = Point::from(&r);
       }
@@ -572,7 +535,7 @@ impl Point {
       h = Point::from(&r);
   
       for i in &[0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62] {
-          t = table_select(*i / 2, e[*i as usize]);
+          t = Precomp::table_select(*i / 2, e[*i as usize]);
           r = &h + &t;
           h = Point::from(&r);
       }
@@ -646,13 +609,15 @@ fn helper_slide() {
     });
 }
 
-/* r = a * A + b * B
+impl Point2
+{
+    /* r = a * A + b * B
      * where a = a[0]+256*a[1]+...+256^31 a[31].
      * and b = b[0]+256*b[1]+...+256^31 b[31].
      * B is the Ed25519 base point (x,4/5) with x positive. */
-#[allow(non_snake_case)]
-pub fn ge_double_scalarmult_vartime(a: &[u8], A: &Point, b: &[u8]) -> Point2
-{
+    #[allow(non_snake_case)]
+    pub fn double_scalarmult_vartime(a: &[u8], A: &Point, b: &[u8]) -> Point2
+    {
         let mut aslide: [i8; 256] = [0; 256];
         let mut bslide: [i8; 256] = [0; 256];
         #[allow(non_snake_case)]
@@ -729,6 +694,7 @@ pub fn ge_double_scalarmult_vartime(a: &[u8], A: &Point, b: &[u8]) -> Point2
         }
     
         r
+    }
 }
 
 #[cfg(test)]
@@ -744,7 +710,7 @@ fn double_scalarmult() {
         assert!(!nega && !negb && !negc && !negd);
         let b = Point::load_test_value(bbytes);
         let d = Point2::load_test_value(dbytes);
-        let mine = ge_double_scalarmult_vartime(&abytes, &b, &cbytes);
+        let mine = Point2::double_scalarmult_vartime(&abytes, &b, &cbytes);
         assert_eq!(mine, d);
     });
 }