ed25519 1.0.0-jruby

data/ext/ed25519_ref10/fe_sq2.c

@@ -0,0 +1,159 @@
+#include "fe.h"
+
+/*
+h = 2 * f * f
+Can overlap h with f.
+
+Preconditions:
+   |f| bounded by 1.65*2^26,1.65*2^25,1.65*2^26,1.65*2^25,etc.
+
+Postconditions:
+   |h| bounded by 1.01*2^25,1.01*2^24,1.01*2^25,1.01*2^24,etc.
+*/
+
+/*
+See fe_mul.c for discussion of implementation strategy.
+*/
+
+void fe_sq2(fe h,const fe f)
+{
+  int32_t f0 = f[0];
+  int32_t f1 = f[1];
+  int32_t f2 = f[2];
+  int32_t f3 = f[3];
+  int32_t f4 = f[4];
+  int32_t f5 = f[5];
+  int32_t f6 = f[6];
+  int32_t f7 = f[7];
+  int32_t f8 = f[8];
+  int32_t f9 = f[9];
+  int32_t f0_2 = 2 * f0;
+  int32_t f1_2 = 2 * f1;
+  int32_t f2_2 = 2 * f2;
+  int32_t f3_2 = 2 * f3;
+  int32_t f4_2 = 2 * f4;
+  int32_t f5_2 = 2 * f5;
+  int32_t f6_2 = 2 * f6;
+  int32_t f7_2 = 2 * f7;
+  int32_t f5_38 = 38 * f5; /* 1.959375*2^30 */
+  int32_t f6_19 = 19 * f6; /* 1.959375*2^30 */
+  int32_t f7_38 = 38 * f7; /* 1.959375*2^30 */
+  int32_t f8_19 = 19 * f8; /* 1.959375*2^30 */
+  int32_t f9_38 = 38 * f9; /* 1.959375*2^30 */
+  int64_t f0f0    = f0   * (int64_t) f0;
+  int64_t f0f1_2  = f0_2 * (int64_t) f1;
+  int64_t f0f2_2  = f0_2 * (int64_t) f2;
+  int64_t f0f3_2  = f0_2 * (int64_t) f3;
+  int64_t f0f4_2  = f0_2 * (int64_t) f4;
+  int64_t f0f5_2  = f0_2 * (int64_t) f5;
+  int64_t f0f6_2  = f0_2 * (int64_t) f6;
+  int64_t f0f7_2  = f0_2 * (int64_t) f7;
+  int64_t f0f8_2  = f0_2 * (int64_t) f8;
+  int64_t f0f9_2  = f0_2 * (int64_t) f9;
+  int64_t f1f1_2  = f1_2 * (int64_t) f1;
+  int64_t f1f2_2  = f1_2 * (int64_t) f2;
+  int64_t f1f3_4  = f1_2 * (int64_t) f3_2;
+  int64_t f1f4_2  = f1_2 * (int64_t) f4;
+  int64_t f1f5_4  = f1_2 * (int64_t) f5_2;
+  int64_t f1f6_2  = f1_2 * (int64_t) f6;
+  int64_t f1f7_4  = f1_2 * (int64_t) f7_2;
+  int64_t f1f8_2  = f1_2 * (int64_t) f8;
+  int64_t f1f9_76 = f1_2 * (int64_t) f9_38;
+  int64_t f2f2    = f2   * (int64_t) f2;
+  int64_t f2f3_2  = f2_2 * (int64_t) f3;
+  int64_t f2f4_2  = f2_2 * (int64_t) f4;
+  int64_t f2f5_2  = f2_2 * (int64_t) f5;
+  int64_t f2f6_2  = f2_2 * (int64_t) f6;
+  int64_t f2f7_2  = f2_2 * (int64_t) f7;
+  int64_t f2f8_38 = f2_2 * (int64_t) f8_19;
+  int64_t f2f9_38 = f2   * (int64_t) f9_38;
+  int64_t f3f3_2  = f3_2 * (int64_t) f3;
+  int64_t f3f4_2  = f3_2 * (int64_t) f4;
+  int64_t f3f5_4  = f3_2 * (int64_t) f5_2;
+  int64_t f3f6_2  = f3_2 * (int64_t) f6;
+  int64_t f3f7_76 = f3_2 * (int64_t) f7_38;
+  int64_t f3f8_38 = f3_2 * (int64_t) f8_19;
+  int64_t f3f9_76 = f3_2 * (int64_t) f9_38;
+  int64_t f4f4    = f4   * (int64_t) f4;
+  int64_t f4f5_2  = f4_2 * (int64_t) f5;
+  int64_t f4f6_38 = f4_2 * (int64_t) f6_19;
+  int64_t f4f7_38 = f4   * (int64_t) f7_38;
+  int64_t f4f8_38 = f4_2 * (int64_t) f8_19;
+  int64_t f4f9_38 = f4   * (int64_t) f9_38;
+  int64_t f5f5_38 = f5   * (int64_t) f5_38;
+  int64_t f5f6_38 = f5_2 * (int64_t) f6_19;
+  int64_t f5f7_76 = f5_2 * (int64_t) f7_38;
+  int64_t f5f8_38 = f5_2 * (int64_t) f8_19;
+  int64_t f5f9_76 = f5_2 * (int64_t) f9_38;
+  int64_t f6f6_19 = f6   * (int64_t) f6_19;
+  int64_t f6f7_38 = f6   * (int64_t) f7_38;
+  int64_t f6f8_38 = f6_2 * (int64_t) f8_19;
+  int64_t f6f9_38 = f6   * (int64_t) f9_38;
+  int64_t f7f7_38 = f7   * (int64_t) f7_38;
+  int64_t f7f8_38 = f7_2 * (int64_t) f8_19;
+  int64_t f7f9_76 = f7_2 * (int64_t) f9_38;
+  int64_t f8f8_19 = f8   * (int64_t) f8_19;
+  int64_t f8f9_38 = f8   * (int64_t) f9_38;
+  int64_t f9f9_38 = f9   * (int64_t) f9_38;
+  int64_t h0 = f0f0  +f1f9_76+f2f8_38+f3f7_76+f4f6_38+f5f5_38;
+  int64_t h1 = f0f1_2+f2f9_38+f3f8_38+f4f7_38+f5f6_38;
+  int64_t h2 = f0f2_2+f1f1_2 +f3f9_76+f4f8_38+f5f7_76+f6f6_19;
+  int64_t h3 = f0f3_2+f1f2_2 +f4f9_38+f5f8_38+f6f7_38;
+  int64_t h4 = f0f4_2+f1f3_4 +f2f2   +f5f9_76+f6f8_38+f7f7_38;
+  int64_t h5 = f0f5_2+f1f4_2 +f2f3_2 +f6f9_38+f7f8_38;
+  int64_t h6 = f0f6_2+f1f5_4 +f2f4_2 +f3f3_2 +f7f9_76+f8f8_19;
+  int64_t h7 = f0f7_2+f1f6_2 +f2f5_2 +f3f4_2 +f8f9_38;
+  int64_t h8 = f0f8_2+f1f7_4 +f2f6_2 +f3f5_4 +f4f4   +f9f9_38;
+  int64_t h9 = f0f9_2+f1f8_2 +f2f7_2 +f3f6_2 +f4f5_2;
+  int64_t carry0;
+  int64_t carry1;
+  int64_t carry2;
+  int64_t carry3;
+  int64_t carry4;
+  int64_t carry5;
+  int64_t carry6;
+  int64_t carry7;
+  int64_t carry8;
+  int64_t carry9;
+
+  h0 += h0;
+  h1 += h1;
+  h2 += h2;
+  h3 += h3;
+  h4 += h4;
+  h5 += h5;
+  h6 += h6;
+  h7 += h7;
+  h8 += h8;
+  h9 += h9;
+
+  carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
+  carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
+
+  carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
+  carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
+
+  carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
+  carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
+
+  carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
+  carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
+
+  carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
+  carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
+
+  carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
+
+  carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
+
+  h[0] = (int32_t)h0;
+  h[1] = (int32_t)h1;
+  h[2] = (int32_t)h2;
+  h[3] = (int32_t)h3;
+  h[4] = (int32_t)h4;
+  h[5] = (int32_t)h5;
+  h[6] = (int32_t)h6;
+  h[7] = (int32_t)h7;
+  h[8] = (int32_t)h8;
+  h[9] = (int32_t)h9;
+}

data/ext/ed25519_ref10/fe_sub.c

@@ -0,0 +1,57 @@
+#include "fe.h"
+
+/*
+h = f - g
+Can overlap h with f or g.
+
+Preconditions:
+   |f| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
+   |g| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
+
+Postconditions:
+   |h| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
+*/
+
+void fe_sub(fe h,const fe f,const fe g)
+{
+  int32_t f0 = f[0];
+  int32_t f1 = f[1];
+  int32_t f2 = f[2];
+  int32_t f3 = f[3];
+  int32_t f4 = f[4];
+  int32_t f5 = f[5];
+  int32_t f6 = f[6];
+  int32_t f7 = f[7];
+  int32_t f8 = f[8];
+  int32_t f9 = f[9];
+  int32_t g0 = g[0];
+  int32_t g1 = g[1];
+  int32_t g2 = g[2];
+  int32_t g3 = g[3];
+  int32_t g4 = g[4];
+  int32_t g5 = g[5];
+  int32_t g6 = g[6];
+  int32_t g7 = g[7];
+  int32_t g8 = g[8];
+  int32_t g9 = g[9];
+  int32_t h0 = f0 - g0;
+  int32_t h1 = f1 - g1;
+  int32_t h2 = f2 - g2;
+  int32_t h3 = f3 - g3;
+  int32_t h4 = f4 - g4;
+  int32_t h5 = f5 - g5;
+  int32_t h6 = f6 - g6;
+  int32_t h7 = f7 - g7;
+  int32_t h8 = f8 - g8;
+  int32_t h9 = f9 - g9;
+  h[0] = h0;
+  h[1] = h1;
+  h[2] = h2;
+  h[3] = h3;
+  h[4] = h4;
+  h[5] = h5;
+  h[6] = h6;
+  h[7] = h7;
+  h[8] = h8;
+  h[9] = h9;
+}

data/ext/ed25519_ref10/fe_tobytes.c

@@ -0,0 +1,119 @@
+#include "fe.h"
+
+/*
+Preconditions:
+  |h| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
+
+Write p=2^255-19; q=floor(h/p).
+Basic claim: q = floor(2^(-255)(h + 19 2^(-25)h9 + 2^(-1))).
+
+Proof:
+  Have |h|<=p so |q|<=1 so |19^2 2^(-255) q|<1/4.
+  Also have |h-2^230 h9|<2^231 so |19 2^(-255)(h-2^230 h9)|<1/4.
+
+  Write y=2^(-1)-19^2 2^(-255)q-19 2^(-255)(h-2^230 h9).
+  Then 0<y<1.
+
+  Write r=h-pq.
+  Have 0<=r<=p-1=2^255-20.
+  Thus 0<=r+19(2^-255)r<r+19(2^-255)2^255<=2^255-1.
+
+  Write x=r+19(2^-255)r+y.
+  Then 0<x<2^255 so floor(2^(-255)x) = 0 so floor(q+2^(-255)x) = q.
+
+  Have q+2^(-255)x = 2^(-255)(h + 19 2^(-25) h9 + 2^(-1))
+  so floor(2^(-255)(h + 19 2^(-25) h9 + 2^(-1))) = q.
+*/
+
+void fe_tobytes(unsigned char *s,const fe h)
+{
+  int32_t h0 = h[0];
+  int32_t h1 = h[1];
+  int32_t h2 = h[2];
+  int32_t h3 = h[3];
+  int32_t h4 = h[4];
+  int32_t h5 = h[5];
+  int32_t h6 = h[6];
+  int32_t h7 = h[7];
+  int32_t h8 = h[8];
+  int32_t h9 = h[9];
+  int32_t q;
+  int32_t carry0;
+  int32_t carry1;
+  int32_t carry2;
+  int32_t carry3;
+  int32_t carry4;
+  int32_t carry5;
+  int32_t carry6;
+  int32_t carry7;
+  int32_t carry8;
+  int32_t carry9;
+
+  q = (19 * h9 + (((int32_t) 1) << 24)) >> 25;
+  q = (h0 + q) >> 26;
+  q = (h1 + q) >> 25;
+  q = (h2 + q) >> 26;
+  q = (h3 + q) >> 25;
+  q = (h4 + q) >> 26;
+  q = (h5 + q) >> 25;
+  q = (h6 + q) >> 26;
+  q = (h7 + q) >> 25;
+  q = (h8 + q) >> 26;
+  q = (h9 + q) >> 25;
+
+  /* Goal: Output h-(2^255-19)q, which is between 0 and 2^255-20. */
+  h0 += 19 * q;
+  /* Goal: Output h-2^255 q, which is between 0 and 2^255-20. */
+
+  carry0 = h0 >> 26; h1 += carry0; h0 -= carry0 << 26;
+  carry1 = h1 >> 25; h2 += carry1; h1 -= carry1 << 25;
+  carry2 = h2 >> 26; h3 += carry2; h2 -= carry2 << 26;
+  carry3 = h3 >> 25; h4 += carry3; h3 -= carry3 << 25;
+  carry4 = h4 >> 26; h5 += carry4; h4 -= carry4 << 26;
+  carry5 = h5 >> 25; h6 += carry5; h5 -= carry5 << 25;
+  carry6 = h6 >> 26; h7 += carry6; h6 -= carry6 << 26;
+  carry7 = h7 >> 25; h8 += carry7; h7 -= carry7 << 25;
+  carry8 = h8 >> 26; h9 += carry8; h8 -= carry8 << 26;
+  carry9 = h9 >> 25;               h9 -= carry9 << 25;
+                  /* h10 = carry9 */
+
+  /*
+  Goal: Output h0+...+2^255 h10-2^255 q, which is between 0 and 2^255-20.
+  Have h0+...+2^230 h9 between 0 and 2^255-1;
+  evidently 2^255 h10-2^255 q = 0.
+  Goal: Output h0+...+2^230 h9.
+  */
+
+  s[0] = h0 >> 0;
+  s[1] = h0 >> 8;
+  s[2] = h0 >> 16;
+  s[3] = (h0 >> 24) | (h1 << 2);
+  s[4] = h1 >> 6;
+  s[5] = h1 >> 14;
+  s[6] = (h1 >> 22) | (h2 << 3);
+  s[7] = h2 >> 5;
+  s[8] = h2 >> 13;
+  s[9] = (h2 >> 21) | (h3 << 5);
+  s[10] = h3 >> 3;
+  s[11] = h3 >> 11;
+  s[12] = (h3 >> 19) | (h4 << 6);
+  s[13] = h4 >> 2;
+  s[14] = h4 >> 10;
+  s[15] = h4 >> 18;
+  s[16] = h5 >> 0;
+  s[17] = h5 >> 8;
+  s[18] = h5 >> 16;
+  s[19] = (h5 >> 24) | (h6 << 1);
+  s[20] = h6 >> 7;
+  s[21] = h6 >> 15;
+  s[22] = (h6 >> 23) | (h7 << 3);
+  s[23] = h7 >> 5;
+  s[24] = h7 >> 13;
+  s[25] = (h7 >> 21) | (h8 << 4);
+  s[26] = h8 >> 4;
+  s[27] = h8 >> 12;
+  s[28] = (h8 >> 20) | (h9 << 6);
+  s[29] = h9 >> 2;
+  s[30] = h9 >> 10;
+  s[31] = h9 >> 18;
+}

data/ext/ed25519_ref10/ge.h

@@ -0,0 +1,95 @@
+#ifndef GE_H
+#define GE_H
+
+/*
+ge means group element.
+
+Here the group is the set of pairs (x,y) of field elements (see fe.h)
+satisfying -x^2 + y^2 = 1 + d x^2y^2
+where d = -121665/121666.
+
+Representations:
+  ge_p2 (projective): (X:Y:Z) satisfying x=X/Z, y=Y/Z
+  ge_p3 (extended): (X:Y:Z:T) satisfying x=X/Z, y=Y/Z, XY=ZT
+  ge_p1p1 (completed): ((X:Z),(Y:T)) satisfying x=X/Z, y=Y/T
+  ge_precomp (Duif): (y+x,y-x,2dxy)
+*/
+
+#include "fe.h"
+
+typedef struct {
+  fe X;
+  fe Y;
+  fe Z;
+} ge_p2;
+
+typedef struct {
+  fe X;
+  fe Y;
+  fe Z;
+  fe T;
+} ge_p3;
+
+typedef struct {
+  fe X;
+  fe Y;
+  fe Z;
+  fe T;
+} ge_p1p1;
+
+typedef struct {
+  fe yplusx;
+  fe yminusx;
+  fe xy2d;
+} ge_precomp;
+
+typedef struct {
+  fe YplusX;
+  fe YminusX;
+  fe Z;
+  fe T2d;
+} ge_cached;
+
+#define ge_frombytes_negate_vartime crypto_sign_ed25519_ref10_ge_frombytes_negate_vartime
+#define ge_tobytes crypto_sign_ed25519_ref10_ge_tobytes
+#define ge_p3_tobytes crypto_sign_ed25519_ref10_ge_p3_tobytes
+
+#define ge_p2_0 crypto_sign_ed25519_ref10_ge_p2_0
+#define ge_p3_0 crypto_sign_ed25519_ref10_ge_p3_0
+#define ge_precomp_0 crypto_sign_ed25519_ref10_ge_precomp_0
+#define ge_p3_to_p2 crypto_sign_ed25519_ref10_ge_p3_to_p2
+#define ge_p3_to_cached crypto_sign_ed25519_ref10_ge_p3_to_cached
+#define ge_p1p1_to_p2 crypto_sign_ed25519_ref10_ge_p1p1_to_p2
+#define ge_p1p1_to_p3 crypto_sign_ed25519_ref10_ge_p1p1_to_p3
+#define ge_p2_dbl crypto_sign_ed25519_ref10_ge_p2_dbl
+#define ge_p3_dbl crypto_sign_ed25519_ref10_ge_p3_dbl
+
+#define ge_madd crypto_sign_ed25519_ref10_ge_madd
+#define ge_msub crypto_sign_ed25519_ref10_ge_msub
+#define ge_add crypto_sign_ed25519_ref10_ge_add
+#define ge_sub crypto_sign_ed25519_ref10_ge_sub
+#define ge_scalarmult_base crypto_sign_ed25519_ref10_ge_scalarmult_base
+#define ge_double_scalarmult_vartime crypto_sign_ed25519_ref10_ge_double_scalarmult_vartime
+
+extern void ge_tobytes(unsigned char *,const ge_p2 *);
+extern void ge_p3_tobytes(unsigned char *,const ge_p3 *);
+extern int ge_frombytes_negate_vartime(ge_p3 *,const unsigned char *);
+
+extern void ge_p2_0(ge_p2 *);
+extern void ge_p3_0(ge_p3 *);
+extern void ge_precomp_0(ge_precomp *);
+extern void ge_p3_to_p2(ge_p2 *,const ge_p3 *);
+extern void ge_p3_to_cached(ge_cached *,const ge_p3 *);
+extern void ge_p1p1_to_p2(ge_p2 *,const ge_p1p1 *);
+extern void ge_p1p1_to_p3(ge_p3 *,const ge_p1p1 *);
+extern void ge_p2_dbl(ge_p1p1 *,const ge_p2 *);
+extern void ge_p3_dbl(ge_p1p1 *,const ge_p3 *);
+
+extern void ge_madd(ge_p1p1 *,const ge_p3 *,const ge_precomp *);
+extern void ge_msub(ge_p1p1 *,const ge_p3 *,const ge_precomp *);
+extern void ge_add(ge_p1p1 *,const ge_p3 *,const ge_cached *);
+extern void ge_sub(ge_p1p1 *,const ge_p3 *,const ge_cached *);
+extern void ge_scalarmult_base(ge_p3 *,const unsigned char *);
+extern void ge_double_scalarmult_vartime(ge_p2 *,const unsigned char *,const ge_p3 *,const unsigned char *);
+
+#endif

data/ext/ed25519_ref10/ge_add.c

@@ -0,0 +1,11 @@
+#include "ge.h"
+
+/*
+r = p + q
+*/
+
+void ge_add(ge_p1p1 *r,const ge_p3 *p,const ge_cached *q)
+{
+  fe t0;
+#include "ge_add.h"
+}

data/ext/ed25519_ref10/ge_add.h

@@ -0,0 +1,97 @@
+
+/* qhasm: enter ge_add */
+
+/* qhasm: fe X1 */
+
+/* qhasm: fe Y1 */
+
+/* qhasm: fe Z1 */
+
+/* qhasm: fe Z2 */
+
+/* qhasm: fe T1 */
+
+/* qhasm: fe ZZ */
+
+/* qhasm: fe YpX2 */
+
+/* qhasm: fe YmX2 */
+
+/* qhasm: fe T2d2 */
+
+/* qhasm: fe X3 */
+
+/* qhasm: fe Y3 */
+
+/* qhasm: fe Z3 */
+
+/* qhasm: fe T3 */
+
+/* qhasm: fe YpX1 */
+
+/* qhasm: fe YmX1 */
+
+/* qhasm: fe A */
+
+/* qhasm: fe B */
+
+/* qhasm: fe C */
+
+/* qhasm: fe D */
+
+/* qhasm: YpX1 = Y1+X1 */
+/* asm 1: fe_add(>YpX1=fe#1,<Y1=fe#12,<X1=fe#11); */
+/* asm 2: fe_add(>YpX1=r->X,<Y1=p->Y,<X1=p->X); */
+fe_add(r->X,p->Y,p->X);
+
+/* qhasm: YmX1 = Y1-X1 */
+/* asm 1: fe_sub(>YmX1=fe#2,<Y1=fe#12,<X1=fe#11); */
+/* asm 2: fe_sub(>YmX1=r->Y,<Y1=p->Y,<X1=p->X); */
+fe_sub(r->Y,p->Y,p->X);
+
+/* qhasm: A = YpX1*YpX2 */
+/* asm 1: fe_mul(>A=fe#3,<YpX1=fe#1,<YpX2=fe#15); */
+/* asm 2: fe_mul(>A=r->Z,<YpX1=r->X,<YpX2=q->YplusX); */
+fe_mul(r->Z,r->X,q->YplusX);
+
+/* qhasm: B = YmX1*YmX2 */
+/* asm 1: fe_mul(>B=fe#2,<YmX1=fe#2,<YmX2=fe#16); */
+/* asm 2: fe_mul(>B=r->Y,<YmX1=r->Y,<YmX2=q->YminusX); */
+fe_mul(r->Y,r->Y,q->YminusX);
+
+/* qhasm: C = T2d2*T1 */
+/* asm 1: fe_mul(>C=fe#4,<T2d2=fe#18,<T1=fe#14); */
+/* asm 2: fe_mul(>C=r->T,<T2d2=q->T2d,<T1=p->T); */
+fe_mul(r->T,q->T2d,p->T);
+
+/* qhasm: ZZ = Z1*Z2 */
+/* asm 1: fe_mul(>ZZ=fe#1,<Z1=fe#13,<Z2=fe#17); */
+/* asm 2: fe_mul(>ZZ=r->X,<Z1=p->Z,<Z2=q->Z); */
+fe_mul(r->X,p->Z,q->Z);
+
+/* qhasm: D = 2*ZZ */
+/* asm 1: fe_add(>D=fe#5,<ZZ=fe#1,<ZZ=fe#1); */
+/* asm 2: fe_add(>D=t0,<ZZ=r->X,<ZZ=r->X); */
+fe_add(t0,r->X,r->X);
+
+/* qhasm: X3 = A-B */
+/* asm 1: fe_sub(>X3=fe#1,<A=fe#3,<B=fe#2); */
+/* asm 2: fe_sub(>X3=r->X,<A=r->Z,<B=r->Y); */
+fe_sub(r->X,r->Z,r->Y);
+
+/* qhasm: Y3 = A+B */
+/* asm 1: fe_add(>Y3=fe#2,<A=fe#3,<B=fe#2); */
+/* asm 2: fe_add(>Y3=r->Y,<A=r->Z,<B=r->Y); */
+fe_add(r->Y,r->Z,r->Y);
+
+/* qhasm: Z3 = D+C */
+/* asm 1: fe_add(>Z3=fe#3,<D=fe#5,<C=fe#4); */
+/* asm 2: fe_add(>Z3=r->Z,<D=t0,<C=r->T); */
+fe_add(r->Z,t0,r->T);
+
+/* qhasm: T3 = D-C */
+/* asm 1: fe_sub(>T3=fe#4,<D=fe#5,<C=fe#4); */
+/* asm 2: fe_sub(>T3=r->T,<D=t0,<C=r->T); */
+fe_sub(r->T,t0,r->T);
+
+/* qhasm: return */