x25519 0.0.0 → 0.1.0

data/ext/x25519/ref10/fe_mul121666.c

@@ -0,0 +1,69 @@
+#include "fe.h"
+
+/*
+h = f * 121666
+Can overlap h with f.
+
+Preconditions:
+   |f| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
+
+Postconditions:
+   |h| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
+*/
+
+void fe_mul121666(fe h,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];
+  int64_t h0 = f0 * (int64_t) 121666;
+  int64_t h1 = f1 * (int64_t) 121666;
+  int64_t h2 = f2 * (int64_t) 121666;
+  int64_t h3 = f3 * (int64_t) 121666;
+  int64_t h4 = f4 * (int64_t) 121666;
+  int64_t h5 = f5 * (int64_t) 121666;
+  int64_t h6 = f6 * (int64_t) 121666;
+  int64_t h7 = f7 * (int64_t) 121666;
+  int64_t h8 = f8 * (int64_t) 121666;
+  int64_t h9 = f9 * (int64_t) 121666;
+  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;
+
+  carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
+  carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
+  carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
+  carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
+  carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
+
+  carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
+  carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
+  carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
+  carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
+  carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 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/x25519/ref10/fe_sq.c

@@ -0,0 +1,148 @@
+#include "fe.h"
+
+/*
+h = f * f
+Can overlap h with f.
+
+Preconditions:
+   |f| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
+
+Postconditions:
+   |h| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
+*/
+
+/*
+See fe_mul.c for discussion of implementation strategy.
+*/
+
+void fe_sq(fe h,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; /* 1.31*2^30 */
+  int32_t f6_19 = 19 * f6; /* 1.31*2^30 */
+  int32_t f7_38 = 38 * f7; /* 1.31*2^30 */
+  int32_t f8_19 = 19 * f8; /* 1.31*2^30 */
+  int32_t f9_38 = 38 * f9; /* 1.31*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;
+
+  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/x25519/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,fe f,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/x25519/ref10/fe_tobytes.c

@@ -0,0 +1,119 @@
+#include "fe.h"
+
+/*
+Preconditions:
+  |h| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,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|<2^230 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,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/x25519/ref10/montgomery.h

@@ -0,0 +1,140 @@
+
+/* qhasm: fe X2 */
+
+/* qhasm: fe Z2 */
+
+/* qhasm: fe X3 */
+
+/* qhasm: fe Z3 */
+
+/* qhasm: fe X4 */
+
+/* qhasm: fe Z4 */
+
+/* qhasm: fe X5 */
+
+/* qhasm: fe Z5 */
+
+/* qhasm: fe A */
+
+/* qhasm: fe B */
+
+/* qhasm: fe C */
+
+/* qhasm: fe D */
+
+/* qhasm: fe E */
+
+/* qhasm: fe AA */
+
+/* qhasm: fe BB */
+
+/* qhasm: fe DA */
+
+/* qhasm: fe CB */
+
+/* qhasm: fe t0 */
+
+/* qhasm: fe t1 */
+
+/* qhasm: fe t2 */
+
+/* qhasm: fe t3 */
+
+/* qhasm: fe t4 */
+
+/* qhasm: enter ladder */
+
+/* qhasm: D = X3-Z3 */
+/* asm 1: fe_sub(>D=fe#5,<X3=fe#3,<Z3=fe#4); */
+/* asm 2: fe_sub(>D=tmp0,<X3=x3,<Z3=z3); */
+fe_sub(tmp0,x3,z3);
+
+/* qhasm: B = X2-Z2 */
+/* asm 1: fe_sub(>B=fe#6,<X2=fe#1,<Z2=fe#2); */
+/* asm 2: fe_sub(>B=tmp1,<X2=x2,<Z2=z2); */
+fe_sub(tmp1,x2,z2);
+
+/* qhasm: A = X2+Z2 */
+/* asm 1: fe_add(>A=fe#1,<X2=fe#1,<Z2=fe#2); */
+/* asm 2: fe_add(>A=x2,<X2=x2,<Z2=z2); */
+fe_add(x2,x2,z2);
+
+/* qhasm: C = X3+Z3 */
+/* asm 1: fe_add(>C=fe#2,<X3=fe#3,<Z3=fe#4); */
+/* asm 2: fe_add(>C=z2,<X3=x3,<Z3=z3); */
+fe_add(z2,x3,z3);
+
+/* qhasm: DA = D*A */
+/* asm 1: fe_mul(>DA=fe#4,<D=fe#5,<A=fe#1); */
+/* asm 2: fe_mul(>DA=z3,<D=tmp0,<A=x2); */
+fe_mul(z3,tmp0,x2);
+
+/* qhasm: CB = C*B */
+/* asm 1: fe_mul(>CB=fe#2,<C=fe#2,<B=fe#6); */
+/* asm 2: fe_mul(>CB=z2,<C=z2,<B=tmp1); */
+fe_mul(z2,z2,tmp1);
+
+/* qhasm: BB = B^2 */
+/* asm 1: fe_sq(>BB=fe#5,<B=fe#6); */
+/* asm 2: fe_sq(>BB=tmp0,<B=tmp1); */
+fe_sq(tmp0,tmp1);
+
+/* qhasm: AA = A^2 */
+/* asm 1: fe_sq(>AA=fe#6,<A=fe#1); */
+/* asm 2: fe_sq(>AA=tmp1,<A=x2); */
+fe_sq(tmp1,x2);
+
+/* qhasm: t0 = DA+CB */
+/* asm 1: fe_add(>t0=fe#3,<DA=fe#4,<CB=fe#2); */
+/* asm 2: fe_add(>t0=x3,<DA=z3,<CB=z2); */
+fe_add(x3,z3,z2);
+
+/* qhasm: assign x3 to t0 */
+
+/* qhasm: t1 = DA-CB */
+/* asm 1: fe_sub(>t1=fe#2,<DA=fe#4,<CB=fe#2); */
+/* asm 2: fe_sub(>t1=z2,<DA=z3,<CB=z2); */
+fe_sub(z2,z3,z2);
+
+/* qhasm: X4 = AA*BB */
+/* asm 1: fe_mul(>X4=fe#1,<AA=fe#6,<BB=fe#5); */
+/* asm 2: fe_mul(>X4=x2,<AA=tmp1,<BB=tmp0); */
+fe_mul(x2,tmp1,tmp0);
+
+/* qhasm: E = AA-BB */
+/* asm 1: fe_sub(>E=fe#6,<AA=fe#6,<BB=fe#5); */
+/* asm 2: fe_sub(>E=tmp1,<AA=tmp1,<BB=tmp0); */
+fe_sub(tmp1,tmp1,tmp0);
+
+/* qhasm: t2 = t1^2 */
+/* asm 1: fe_sq(>t2=fe#2,<t1=fe#2); */
+/* asm 2: fe_sq(>t2=z2,<t1=z2); */
+fe_sq(z2,z2);
+
+/* qhasm: t3 = a24*E */
+/* asm 1: fe_mul121666(>t3=fe#4,<E=fe#6); */
+/* asm 2: fe_mul121666(>t3=z3,<E=tmp1); */
+fe_mul121666(z3,tmp1);
+
+/* qhasm: X5 = t0^2 */
+/* asm 1: fe_sq(>X5=fe#3,<t0=fe#3); */
+/* asm 2: fe_sq(>X5=x3,<t0=x3); */
+fe_sq(x3,x3);
+
+/* qhasm: t4 = BB+t3 */
+/* asm 1: fe_add(>t4=fe#5,<BB=fe#5,<t3=fe#4); */
+/* asm 2: fe_add(>t4=tmp0,<BB=tmp0,<t3=z3); */
+fe_add(tmp0,tmp0,z3);
+
+/* qhasm: Z5 = X1*t2 */
+/* asm 1: fe_mul(>Z5=fe#4,x1,<t2=fe#2); */
+/* asm 2: fe_mul(>Z5=z3,x1,<t2=z2); */
+fe_mul(z3,x1,z2);
+
+/* qhasm: Z4 = E*t4 */
+/* asm 1: fe_mul(>Z4=fe#2,<E=fe#6,<t4=fe#5); */
+/* asm 2: fe_mul(>Z4=z2,<E=tmp1,<t4=tmp0); */
+fe_mul(z2,tmp1,tmp0);
+
+/* qhasm: return */