diff --git a/crypto/fipsmodule/bcm.c b/crypto/fipsmodule/bcm.c
index b506b43..fb16215 100644
--- a/crypto/fipsmodule/bcm.c
+++ b/crypto/fipsmodule/bcm.c
@@ -67,10 +67,10 @@
 #include "ec/ec_montgomery.c"
 #include "ec/oct.c"
 #include "ec/p224-64.c"
-#include "ec/p256-64.c"
+#include "../../third_party/fiat/p256.c"
 #include "ec/p256-x86_64.c"
 #include "ec/simple.c"
-#include "ec/util-64.c"
+#include "ec/util.c"
 #include "ec/wnaf.c"
 #include "hmac/hmac.c"
 #include "md4/md4.c"
diff --git a/crypto/fipsmodule/ec/ec.c b/crypto/fipsmodule/ec/ec.c
index 41c2540..ed54554 100644
--- a/crypto/fipsmodule/ec/ec.c
+++ b/crypto/fipsmodule/ec/ec.c
@@ -215,13 +215,6 @@
     0xB7, 0x1E, 0x91, 0x38, 0x64, 0x09,
 };
 
-// MSan appears to have a bug that causes code to be miscompiled in opt mode.
-// While that is being looked at, don't run the uint128_t code under MSan.
-#if defined(OPENSSL_64_BIT) && !defined(OPENSSL_WINDOWS) && \
-    !defined(MEMORY_SANITIZER)
-#define BORINGSSL_USE_INT128_CODE
-#endif
-
 DEFINE_METHOD_FUNCTION(struct built_in_curves, OPENSSL_built_in_curves) {
   // 1.3.132.0.35
   static const uint8_t kOIDP521[] = {0x2b, 0x81, 0x04, 0x00, 0x23};
@@ -253,16 +246,19 @@
   out->curves[2].param_len = 32;
   out->curves[2].params = kP256Params;
   out->curves[2].method =
-#if defined(BORINGSSL_USE_INT128_CODE)
+// MSan appears to have a bug that causes code to be miscompiled in opt mode.
+// While that is being looked at, don't run the uint128_t code under MSan.
 #if !defined(OPENSSL_NO_ASM) && defined(OPENSSL_X86_64) && \
-    !defined(OPENSSL_SMALL)
+    !defined(OPENSSL_SMALL) && !defined(MEMORY_SANITIZER)
       EC_GFp_nistz256_method();
 #else
+#if defined(OPENSSL_32_BIT) || \
+    (defined(OPENSSL_64_BIT) && !defined(MEMORY_SANITIZER))
       EC_GFp_nistp256_method();
-#endif
 #else
       EC_GFp_mont_method();
 #endif
+#endif
 
   // 1.3.132.0.33
   static const uint8_t kOIDP224[] = {0x2b, 0x81, 0x04, 0x00, 0x21};
@@ -273,7 +269,8 @@
   out->curves[3].param_len = 28;
   out->curves[3].params = kP224Params;
   out->curves[3].method =
-#if defined(BORINGSSL_USE_INT128_CODE) && !defined(OPENSSL_SMALL)
+#if defined(OPENSSL_64_BIT) && !defined(OPENSSL_WINDOWS) && \
+    !defined(MEMORY_SANITIZER) && !defined(OPENSSL_SMALL)
       EC_GFp_nistp224_method();
 #else
       EC_GFp_mont_method();
@@ -883,6 +880,24 @@
   return ret;
 }
 
+int ec_point_mul_scalar_public(const EC_GROUP *group, EC_POINT *r,
+                               const EC_SCALAR *g_scalar, const EC_POINT *p,
+                               const EC_SCALAR *p_scalar, BN_CTX *ctx) {
+  if ((g_scalar == NULL && p_scalar == NULL) ||
+      (p == NULL) != (p_scalar == NULL))  {
+    OPENSSL_PUT_ERROR(EC, ERR_R_PASSED_NULL_PARAMETER);
+    return 0;
+  }
+
+  if (EC_GROUP_cmp(group, r->group, NULL) != 0 ||
+      (p != NULL && EC_GROUP_cmp(group, p->group, NULL) != 0)) {
+    OPENSSL_PUT_ERROR(EC, EC_R_INCOMPATIBLE_OBJECTS);
+    return 0;
+  }
+
+  return group->meth->mul_public(group, r, g_scalar, p, p_scalar, ctx);
+}
+
 int ec_point_mul_scalar(const EC_GROUP *group, EC_POINT *r,
                         const EC_SCALAR *g_scalar, const EC_POINT *p,
                         const EC_SCALAR *p_scalar, BN_CTX *ctx) {
diff --git a/crypto/fipsmodule/ec/ec_montgomery.c b/crypto/fipsmodule/ec/ec_montgomery.c
index 6670b84..898cf07 100644
--- a/crypto/fipsmodule/ec/ec_montgomery.c
+++ b/crypto/fipsmodule/ec/ec_montgomery.c
@@ -270,6 +270,7 @@
   out->group_set_curve = ec_GFp_mont_group_set_curve;
   out->point_get_affine_coordinates = ec_GFp_mont_point_get_affine_coordinates;
   out->mul = ec_wNAF_mul /* XXX: Not constant time. */;
+  out->mul_public = ec_wNAF_mul;
   out->field_mul = ec_GFp_mont_field_mul;
   out->field_sqr = ec_GFp_mont_field_sqr;
   out->field_encode = ec_GFp_mont_field_encode;
diff --git a/crypto/fipsmodule/ec/internal.h b/crypto/fipsmodule/ec/internal.h
index 1b860c6..145c5c4 100644
--- a/crypto/fipsmodule/ec/internal.h
+++ b/crypto/fipsmodule/ec/internal.h
@@ -115,6 +115,12 @@
   // non-null.
   int (*mul)(const EC_GROUP *group, EC_POINT *r, const EC_SCALAR *g_scalar,
              const EC_POINT *p, const EC_SCALAR *p_scalar, BN_CTX *ctx);
+  // mul_public performs the same computation as mul. It further assumes that
+  // the inputs are public so there is no concern about leaking their values
+  // through timing.
+  int (*mul_public)(const EC_GROUP *group, EC_POINT *r,
+                    const EC_SCALAR *g_scalar, const EC_POINT *p,
+                    const EC_SCALAR *p_scalar, BN_CTX *ctx);
 
   // 'field_mul' and 'field_sqr' can be used by 'add' and 'dbl' so that the
   // same implementations of point operations can be used with different
@@ -195,6 +201,13 @@
                         const EC_SCALAR *g_scalar, const EC_POINT *p,
                         const EC_SCALAR *p_scalar, BN_CTX *ctx);
 
+// ec_point_mul_scalar_public performs the same computation as
+// ec_point_mul_scalar.  It further assumes that the inputs are public so
+// there is no concern about leaking their values through timing.
+int ec_point_mul_scalar_public(const EC_GROUP *group, EC_POINT *r,
+                               const EC_SCALAR *g_scalar, const EC_POINT *p,
+                               const EC_SCALAR *p_scalar, BN_CTX *ctx);
+
 int ec_wNAF_mul(const EC_GROUP *group, EC_POINT *r, const EC_SCALAR *g_scalar,
                 const EC_POINT *p, const EC_SCALAR *p_scalar, BN_CTX *ctx);
 
diff --git a/crypto/fipsmodule/ec/p224-64.c b/crypto/fipsmodule/ec/p224-64.c
index ba25d22..d0285d6 100644
--- a/crypto/fipsmodule/ec/p224-64.c
+++ b/crypto/fipsmodule/ec/p224-64.c
@@ -1122,6 +1122,7 @@
   out->point_get_affine_coordinates =
       ec_GFp_nistp224_point_get_affine_coordinates;
   out->mul = ec_GFp_nistp224_points_mul;
+  out->mul_public = ec_GFp_nistp224_points_mul;
   out->field_mul = ec_GFp_simple_field_mul;
   out->field_sqr = ec_GFp_simple_field_sqr;
   out->field_encode = NULL;
diff --git a/crypto/fipsmodule/ec/p256-64.c b/crypto/fipsmodule/ec/p256-64.c
deleted file mode 100644
index d4a8ff6..0000000
--- a/crypto/fipsmodule/ec/p256-64.c
+++ /dev/null
@@ -1,1674 +0,0 @@
-/* Copyright (c) 2015, Google Inc.
- *
- * Permission to use, copy, modify, and/or distribute this software for any
- * purpose with or without fee is hereby granted, provided that the above
- * copyright notice and this permission notice appear in all copies.
- *
- * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
- * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
- * SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
- * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
- * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */
-
-// A 64-bit implementation of the NIST P-256 elliptic curve point
-// multiplication
-//
-// OpenSSL integration was taken from Emilia Kasper's work in ecp_nistp224.c.
-// Otherwise based on Emilia's P224 work, which was inspired by my curve25519
-// work which got its smarts from Daniel J. Bernstein's work on the same.
-
-#include <openssl/base.h>
-
-#if defined(OPENSSL_64_BIT) && !defined(OPENSSL_WINDOWS)
-
-#include <openssl/bn.h>
-#include <openssl/ec.h>
-#include <openssl/err.h>
-#include <openssl/mem.h>
-
-#include <string.h>
-
-#include "../delocate.h"
-#include "../../internal.h"
-#include "internal.h"
-
-
-// The underlying field. P256 operates over GF(2^256-2^224+2^192+2^96-1). We
-// can serialise an element of this field into 32 bytes. We call this an
-// felem_bytearray.
-typedef uint8_t felem_bytearray[32];
-
-// The representation of field elements.
-// ------------------------------------
-//
-// We represent field elements with either four 128-bit values, eight 128-bit
-// values, or four 64-bit values. The field element represented is:
-//   v[0]*2^0 + v[1]*2^64 + v[2]*2^128 + v[3]*2^192  (mod p)
-// or:
-//   v[0]*2^0 + v[1]*2^64 + v[2]*2^128 + ... + v[8]*2^512  (mod p)
-//
-// 128-bit values are called 'limbs'. Since the limbs are spaced only 64 bits
-// apart, but are 128-bits wide, the most significant bits of each limb overlap
-// with the least significant bits of the next.
-//
-// A field element with four limbs is an 'felem'. One with eight limbs is a
-// 'longfelem'
-//
-// A field element with four, 64-bit values is called a 'smallfelem'. Small
-// values are used as intermediate values before multiplication.
-
-#define NLIMBS 4
-
-typedef uint128_t limb;
-typedef limb felem[NLIMBS];
-typedef limb longfelem[NLIMBS * 2];
-typedef uint64_t smallfelem[NLIMBS];
-
-// This is the value of the prime as four 64-bit words, little-endian.
-static const uint64_t kPrime[4] = {0xfffffffffffffffful, 0xffffffff, 0,
-                              0xffffffff00000001ul};
-static const uint64_t bottom63bits = 0x7ffffffffffffffful;
-
-static uint64_t load_u64(const uint8_t in[8]) {
-  uint64_t ret;
-  OPENSSL_memcpy(&ret, in, sizeof(ret));
-  return ret;
-}
-
-static void store_u64(uint8_t out[8], uint64_t in) {
-  OPENSSL_memcpy(out, &in, sizeof(in));
-}
-
-// bin32_to_felem takes a little-endian byte array and converts it into felem
-// form. This assumes that the CPU is little-endian.
-static void bin32_to_felem(felem out, const uint8_t in[32]) {
-  out[0] = load_u64(&in[0]);
-  out[1] = load_u64(&in[8]);
-  out[2] = load_u64(&in[16]);
-  out[3] = load_u64(&in[24]);
-}
-
-// smallfelem_to_bin32 takes a smallfelem and serialises into a little endian,
-// 32 byte array. This assumes that the CPU is little-endian.
-static void smallfelem_to_bin32(uint8_t out[32], const smallfelem in) {
-  store_u64(&out[0], in[0]);
-  store_u64(&out[8], in[1]);
-  store_u64(&out[16], in[2]);
-  store_u64(&out[24], in[3]);
-}
-
-// To preserve endianness when using BN_bn2bin and BN_bin2bn.
-static void flip_endian(uint8_t *out, const uint8_t *in, size_t len) {
-  for (size_t i = 0; i < len; ++i) {
-    out[i] = in[len - 1 - i];
-  }
-}
-
-// BN_to_felem converts an OpenSSL BIGNUM into an felem.
-static int BN_to_felem(felem out, const BIGNUM *bn) {
-  if (BN_is_negative(bn)) {
-    OPENSSL_PUT_ERROR(EC, EC_R_BIGNUM_OUT_OF_RANGE);
-    return 0;
-  }
-
-  felem_bytearray b_out;
-  // BN_bn2bin eats leading zeroes
-  OPENSSL_memset(b_out, 0, sizeof(b_out));
-  size_t num_bytes = BN_num_bytes(bn);
-  if (num_bytes > sizeof(b_out)) {
-    OPENSSL_PUT_ERROR(EC, EC_R_BIGNUM_OUT_OF_RANGE);
-    return 0;
-  }
-
-  felem_bytearray b_in;
-  num_bytes = BN_bn2bin(bn, b_in);
-  flip_endian(b_out, b_in, num_bytes);
-  bin32_to_felem(out, b_out);
-  return 1;
-}
-
-// felem_to_BN converts an felem into an OpenSSL BIGNUM.
-static BIGNUM *smallfelem_to_BN(BIGNUM *out, const smallfelem in) {
-  felem_bytearray b_in, b_out;
-  smallfelem_to_bin32(b_in, in);
-  flip_endian(b_out, b_in, sizeof(b_out));
-  return BN_bin2bn(b_out, sizeof(b_out), out);
-}
-
-// Field operations.
-
-static void felem_assign(felem out, const felem in) {
-  out[0] = in[0];
-  out[1] = in[1];
-  out[2] = in[2];
-  out[3] = in[3];
-}
-
-// felem_sum sets out = out + in.
-static void felem_sum(felem out, const felem in) {
-  out[0] += in[0];
-  out[1] += in[1];
-  out[2] += in[2];
-  out[3] += in[3];
-}
-
-// felem_small_sum sets out = out + in.
-static void felem_small_sum(felem out, const smallfelem in) {
-  out[0] += in[0];
-  out[1] += in[1];
-  out[2] += in[2];
-  out[3] += in[3];
-}
-
-// felem_scalar sets out = out * scalar
-static void felem_scalar(felem out, const uint64_t scalar) {
-  out[0] *= scalar;
-  out[1] *= scalar;
-  out[2] *= scalar;
-  out[3] *= scalar;
-}
-
-// longfelem_scalar sets out = out * scalar
-static void longfelem_scalar(longfelem out, const uint64_t scalar) {
-  out[0] *= scalar;
-  out[1] *= scalar;
-  out[2] *= scalar;
-  out[3] *= scalar;
-  out[4] *= scalar;
-  out[5] *= scalar;
-  out[6] *= scalar;
-  out[7] *= scalar;
-}
-
-#define two105m41m9 ((((limb)1) << 105) - (((limb)1) << 41) - (((limb)1) << 9))
-#define two105 (((limb)1) << 105)
-#define two105m41p9 ((((limb)1) << 105) - (((limb)1) << 41) + (((limb)1) << 9))
-
-// zero105 is 0 mod p
-static const felem zero105 = {two105m41m9, two105, two105m41p9, two105m41p9};
-
-// smallfelem_neg sets |out| to |-small|
-// On exit:
-//   out[i] < out[i] + 2^105
-static void smallfelem_neg(felem out, const smallfelem small) {
-  // In order to prevent underflow, we subtract from 0 mod p.
-  out[0] = zero105[0] - small[0];
-  out[1] = zero105[1] - small[1];
-  out[2] = zero105[2] - small[2];
-  out[3] = zero105[3] - small[3];
-}
-
-// felem_diff subtracts |in| from |out|
-// On entry:
-//   in[i] < 2^104
-// On exit:
-//   out[i] < out[i] + 2^105.
-static void felem_diff(felem out, const felem in) {
-  // In order to prevent underflow, we add 0 mod p before subtracting.
-  out[0] += zero105[0];
-  out[1] += zero105[1];
-  out[2] += zero105[2];
-  out[3] += zero105[3];
-
-  out[0] -= in[0];
-  out[1] -= in[1];
-  out[2] -= in[2];
-  out[3] -= in[3];
-}
-
-#define two107m43m11 \
-  ((((limb)1) << 107) - (((limb)1) << 43) - (((limb)1) << 11))
-#define two107 (((limb)1) << 107)
-#define two107m43p11 \
-  ((((limb)1) << 107) - (((limb)1) << 43) + (((limb)1) << 11))
-
-// zero107 is 0 mod p
-static const felem zero107 = {two107m43m11, two107, two107m43p11, two107m43p11};
-
-// An alternative felem_diff for larger inputs |in|
-// felem_diff_zero107 subtracts |in| from |out|
-// On entry:
-//   in[i] < 2^106
-// On exit:
-//   out[i] < out[i] + 2^107.
-static void felem_diff_zero107(felem out, const felem in) {
-  // In order to prevent underflow, we add 0 mod p before subtracting.
-  out[0] += zero107[0];
-  out[1] += zero107[1];
-  out[2] += zero107[2];
-  out[3] += zero107[3];
-
-  out[0] -= in[0];
-  out[1] -= in[1];
-  out[2] -= in[2];
-  out[3] -= in[3];
-}
-
-// longfelem_diff subtracts |in| from |out|
-// On entry:
-//   in[i] < 7*2^67
-// On exit:
-//   out[i] < out[i] + 2^70 + 2^40.
-static void longfelem_diff(longfelem out, const longfelem in) {
-  static const limb two70m8p6 =
-      (((limb)1) << 70) - (((limb)1) << 8) + (((limb)1) << 6);
-  static const limb two70p40 = (((limb)1) << 70) + (((limb)1) << 40);
-  static const limb two70 = (((limb)1) << 70);
-  static const limb two70m40m38p6 = (((limb)1) << 70) - (((limb)1) << 40) -
-                                    (((limb)1) << 38) + (((limb)1) << 6);
-  static const limb two70m6 = (((limb)1) << 70) - (((limb)1) << 6);
-
-  // add 0 mod p to avoid underflow
-  out[0] += two70m8p6;
-  out[1] += two70p40;
-  out[2] += two70;
-  out[3] += two70m40m38p6;
-  out[4] += two70m6;
-  out[5] += two70m6;
-  out[6] += two70m6;
-  out[7] += two70m6;
-
-  // in[i] < 7*2^67 < 2^70 - 2^40 - 2^38 + 2^6
-  out[0] -= in[0];
-  out[1] -= in[1];
-  out[2] -= in[2];
-  out[3] -= in[3];
-  out[4] -= in[4];
-  out[5] -= in[5];
-  out[6] -= in[6];
-  out[7] -= in[7];
-}
-
-#define two64m0 ((((limb)1) << 64) - 1)
-#define two110p32m0 ((((limb)1) << 110) + (((limb)1) << 32) - 1)
-#define two64m46 ((((limb)1) << 64) - (((limb)1) << 46))
-#define two64m32 ((((limb)1) << 64) - (((limb)1) << 32))
-
-// zero110 is 0 mod p.
-static const felem zero110 = {two64m0, two110p32m0, two64m46, two64m32};
-
-// felem_shrink converts an felem into a smallfelem. The result isn't quite
-// minimal as the value may be greater than p.
-//
-// On entry:
-//   in[i] < 2^109
-// On exit:
-//   out[i] < 2^64.
-static void felem_shrink(smallfelem out, const felem in) {
-  felem tmp;
-  uint64_t a, b, mask;
-  int64_t high, low;
-  static const uint64_t kPrime3Test =
-      0x7fffffff00000001ul;  // 2^63 - 2^32 + 1
-
-  // Carry 2->3
-  tmp[3] = zero110[3] + in[3] + ((uint64_t)(in[2] >> 64));
-  // tmp[3] < 2^110
-
-  tmp[2] = zero110[2] + (uint64_t)in[2];
-  tmp[0] = zero110[0] + in[0];
-  tmp[1] = zero110[1] + in[1];
-  // tmp[0] < 2**110, tmp[1] < 2^111, tmp[2] < 2**65
-
-  // We perform two partial reductions where we eliminate the high-word of
-  // tmp[3]. We don't update the other words till the end.
-  a = tmp[3] >> 64;  // a < 2^46
-  tmp[3] = (uint64_t)tmp[3];
-  tmp[3] -= a;
-  tmp[3] += ((limb)a) << 32;
-  // tmp[3] < 2^79
-
-  b = a;
-  a = tmp[3] >> 64;  // a < 2^15
-  b += a;            // b < 2^46 + 2^15 < 2^47
-  tmp[3] = (uint64_t)tmp[3];
-  tmp[3] -= a;
-  tmp[3] += ((limb)a) << 32;
-  // tmp[3] < 2^64 + 2^47
-
-  // This adjusts the other two words to complete the two partial
-  // reductions.
-  tmp[0] += b;
-  tmp[1] -= (((limb)b) << 32);
-
-  // In order to make space in tmp[3] for the carry from 2 -> 3, we
-  // conditionally subtract kPrime if tmp[3] is large enough.
-  high = tmp[3] >> 64;
-  // As tmp[3] < 2^65, high is either 1 or 0
-  high = ~(high - 1);
-  // high is:
-  //   all ones   if the high word of tmp[3] is 1
-  //   all zeros  if the high word of tmp[3] if 0
-  low = tmp[3];
-  mask = low >> 63;
-  // mask is:
-  //   all ones   if the MSB of low is 1
-  //   all zeros  if the MSB of low if 0
-  low &= bottom63bits;
-  low -= kPrime3Test;
-  // if low was greater than kPrime3Test then the MSB is zero
-  low = ~low;
-  low >>= 63;
-  // low is:
-  //   all ones   if low was > kPrime3Test
-  //   all zeros  if low was <= kPrime3Test
-  mask = (mask & low) | high;
-  tmp[0] -= mask & kPrime[0];
-  tmp[1] -= mask & kPrime[1];
-  // kPrime[2] is zero, so omitted
-  tmp[3] -= mask & kPrime[3];
-  // tmp[3] < 2**64 - 2**32 + 1
-
-  tmp[1] += ((uint64_t)(tmp[0] >> 64));
-  tmp[0] = (uint64_t)tmp[0];
-  tmp[2] += ((uint64_t)(tmp[1] >> 64));
-  tmp[1] = (uint64_t)tmp[1];
-  tmp[3] += ((uint64_t)(tmp[2] >> 64));
-  tmp[2] = (uint64_t)tmp[2];
-  // tmp[i] < 2^64
-
-  out[0] = tmp[0];
-  out[1] = tmp[1];
-  out[2] = tmp[2];
-  out[3] = tmp[3];
-}
-
-// smallfelem_expand converts a smallfelem to an felem
-static void smallfelem_expand(felem out, const smallfelem in) {
-  out[0] = in[0];
-  out[1] = in[1];
-  out[2] = in[2];
-  out[3] = in[3];
-}
-
-// smallfelem_square sets |out| = |small|^2
-// On entry:
-//   small[i] < 2^64
-// On exit:
-//   out[i] < 7 * 2^64 < 2^67
-static void smallfelem_square(longfelem out, const smallfelem small) {
-  limb a;
-  uint64_t high, low;
-
-  a = ((uint128_t)small[0]) * small[0];
-  low = a;
-  high = a >> 64;
-  out[0] = low;
-  out[1] = high;
-
-  a = ((uint128_t)small[0]) * small[1];
-  low = a;
-  high = a >> 64;
-  out[1] += low;
-  out[1] += low;
-  out[2] = high;
-
-  a = ((uint128_t)small[0]) * small[2];
-  low = a;
-  high = a >> 64;
-  out[2] += low;
-  out[2] *= 2;
-  out[3] = high;
-
-  a = ((uint128_t)small[0]) * small[3];
-  low = a;
-  high = a >> 64;
-  out[3] += low;
-  out[4] = high;
-
-  a = ((uint128_t)small[1]) * small[2];
-  low = a;
-  high = a >> 64;
-  out[3] += low;
-  out[3] *= 2;
-  out[4] += high;
-
-  a = ((uint128_t)small[1]) * small[1];
-  low = a;
-  high = a >> 64;
-  out[2] += low;
-  out[3] += high;
-
-  a = ((uint128_t)small[1]) * small[3];
-  low = a;
-  high = a >> 64;
-  out[4] += low;
-  out[4] *= 2;
-  out[5] = high;
-
-  a = ((uint128_t)small[2]) * small[3];
-  low = a;
-  high = a >> 64;
-  out[5] += low;
-  out[5] *= 2;
-  out[6] = high;
-  out[6] += high;
-
-  a = ((uint128_t)small[2]) * small[2];
-  low = a;
-  high = a >> 64;
-  out[4] += low;
-  out[5] += high;
-
-  a = ((uint128_t)small[3]) * small[3];
-  low = a;
-  high = a >> 64;
-  out[6] += low;
-  out[7] = high;
-}
-
-//felem_square sets |out| = |in|^2
-// On entry:
-//   in[i] < 2^109
-// On exit:
-//   out[i] < 7 * 2^64 < 2^67.
-static void felem_square(longfelem out, const felem in) {
-  uint64_t small[4];
-  felem_shrink(small, in);
-  smallfelem_square(out, small);
-}
-
-// smallfelem_mul sets |out| = |small1| * |small2|
-// On entry:
-//   small1[i] < 2^64
-//   small2[i] < 2^64
-// On exit:
-//   out[i] < 7 * 2^64 < 2^67.
-static void smallfelem_mul(longfelem out, const smallfelem small1,
-                           const smallfelem small2) {
-  limb a;
-  uint64_t high, low;
-
-  a = ((uint128_t)small1[0]) * small2[0];
-  low = a;
-  high = a >> 64;
-  out[0] = low;
-  out[1] = high;
-
-  a = ((uint128_t)small1[0]) * small2[1];
-  low = a;
-  high = a >> 64;
-  out[1] += low;
-  out[2] = high;
-
-  a = ((uint128_t)small1[1]) * small2[0];
-  low = a;
-  high = a >> 64;
-  out[1] += low;
-  out[2] += high;
-
-  a = ((uint128_t)small1[0]) * small2[2];
-  low = a;
-  high = a >> 64;
-  out[2] += low;
-  out[3] = high;
-
-  a = ((uint128_t)small1[1]) * small2[1];
-  low = a;
-  high = a >> 64;
-  out[2] += low;
-  out[3] += high;
-
-  a = ((uint128_t)small1[2]) * small2[0];
-  low = a;
-  high = a >> 64;
-  out[2] += low;
-  out[3] += high;
-
-  a = ((uint128_t)small1[0]) * small2[3];
-  low = a;
-  high = a >> 64;
-  out[3] += low;
-  out[4] = high;
-
-  a = ((uint128_t)small1[1]) * small2[2];
-  low = a;
-  high = a >> 64;
-  out[3] += low;
-  out[4] += high;
-
-  a = ((uint128_t)small1[2]) * small2[1];
-  low = a;
-  high = a >> 64;
-  out[3] += low;
-  out[4] += high;
-
-  a = ((uint128_t)small1[3]) * small2[0];
-  low = a;
-  high = a >> 64;
-  out[3] += low;
-  out[4] += high;
-
-  a = ((uint128_t)small1[1]) * small2[3];
-  low = a;
-  high = a >> 64;
-  out[4] += low;
-  out[5] = high;
-
-  a = ((uint128_t)small1[2]) * small2[2];
-  low = a;
-  high = a >> 64;
-  out[4] += low;
-  out[5] += high;
-
-  a = ((uint128_t)small1[3]) * small2[1];
-  low = a;
-  high = a >> 64;
-  out[4] += low;
-  out[5] += high;
-
-  a = ((uint128_t)small1[2]) * small2[3];
-  low = a;
-  high = a >> 64;
-  out[5] += low;
-  out[6] = high;
-
-  a = ((uint128_t)small1[3]) * small2[2];
-  low = a;
-  high = a >> 64;
-  out[5] += low;
-  out[6] += high;
-
-  a = ((uint128_t)small1[3]) * small2[3];
-  low = a;
-  high = a >> 64;
-  out[6] += low;
-  out[7] = high;
-}
-
-// felem_mul sets |out| = |in1| * |in2|
-// On entry:
-//   in1[i] < 2^109
-//   in2[i] < 2^109
-// On exit:
-//   out[i] < 7 * 2^64 < 2^67
-static void felem_mul(longfelem out, const felem in1, const felem in2) {
-  smallfelem small1, small2;
-  felem_shrink(small1, in1);
-  felem_shrink(small2, in2);
-  smallfelem_mul(out, small1, small2);
-}
-
-// felem_small_mul sets |out| = |small1| * |in2|
-// On entry:
-//   small1[i] < 2^64
-//   in2[i] < 2^109
-// On exit:
-//   out[i] < 7 * 2^64 < 2^67
-static void felem_small_mul(longfelem out, const smallfelem small1,
-                            const felem in2) {
-  smallfelem small2;
-  felem_shrink(small2, in2);
-  smallfelem_mul(out, small1, small2);
-}
-
-#define two100m36m4 ((((limb)1) << 100) - (((limb)1) << 36) - (((limb)1) << 4))
-#define two100 (((limb)1) << 100)
-#define two100m36p4 ((((limb)1) << 100) - (((limb)1) << 36) + (((limb)1) << 4))
-
-// zero100 is 0 mod p
-static const felem zero100 = {two100m36m4, two100, two100m36p4, two100m36p4};
-
-// Internal function for the different flavours of felem_reduce.
-// felem_reduce_ reduces the higher coefficients in[4]-in[7].
-// On entry:
-//   out[0] >= in[6] + 2^32*in[6] + in[7] + 2^32*in[7]
-//   out[1] >= in[7] + 2^32*in[4]
-//   out[2] >= in[5] + 2^32*in[5]
-//   out[3] >= in[4] + 2^32*in[5] + 2^32*in[6]
-// On exit:
-//   out[0] <= out[0] + in[4] + 2^32*in[5]
-//   out[1] <= out[1] + in[5] + 2^33*in[6]
-//   out[2] <= out[2] + in[7] + 2*in[6] + 2^33*in[7]
-//   out[3] <= out[3] + 2^32*in[4] + 3*in[7]
-static void felem_reduce_(felem out, const longfelem in) {
-  int128_t c;
-  // combine common terms from below
-  c = in[4] + (in[5] << 32);
-  out[0] += c;
-  out[3] -= c;
-
-  c = in[5] - in[7];
-  out[1] += c;
-  out[2] -= c;
-
-  // the remaining terms
-  // 256: [(0,1),(96,-1),(192,-1),(224,1)]
-  out[1] -= (in[4] << 32);
-  out[3] += (in[4] << 32);
-
-  // 320: [(32,1),(64,1),(128,-1),(160,-1),(224,-1)]
-  out[2] -= (in[5] << 32);
-
-  // 384: [(0,-1),(32,-1),(96,2),(128,2),(224,-1)]
-  out[0] -= in[6];
-  out[0] -= (in[6] << 32);
-  out[1] += (in[6] << 33);
-  out[2] += (in[6] * 2);
-  out[3] -= (in[6] << 32);
-
-  // 448: [(0,-1),(32,-1),(64,-1),(128,1),(160,2),(192,3)]
-  out[0] -= in[7];
-  out[0] -= (in[7] << 32);
-  out[2] += (in[7] << 33);
-  out[3] += (in[7] * 3);
-}
-
-// felem_reduce converts a longfelem into an felem.
-// To be called directly after felem_square or felem_mul.
-// On entry:
-//   in[0] < 2^64, in[1] < 3*2^64, in[2] < 5*2^64, in[3] < 7*2^64
-//   in[4] < 7*2^64, in[5] < 5*2^64, in[6] < 3*2^64, in[7] < 2*64
-// On exit:
-//   out[i] < 2^101
-static void felem_reduce(felem out, const longfelem in) {
-  out[0] = zero100[0] + in[0];
-  out[1] = zero100[1] + in[1];
-  out[2] = zero100[2] + in[2];
-  out[3] = zero100[3] + in[3];
-
-  felem_reduce_(out, in);
-
-  // out[0] > 2^100 - 2^36 - 2^4 - 3*2^64 - 3*2^96 - 2^64 - 2^96 > 0
-  // out[1] > 2^100 - 2^64 - 7*2^96 > 0
-  // out[2] > 2^100 - 2^36 + 2^4 - 5*2^64 - 5*2^96 > 0
-  // out[3] > 2^100 - 2^36 + 2^4 - 7*2^64 - 5*2^96 - 3*2^96 > 0
-  //
-  // out[0] < 2^100 + 2^64 + 7*2^64 + 5*2^96 < 2^101
-  // out[1] < 2^100 + 3*2^64 + 5*2^64 + 3*2^97 < 2^101
-  // out[2] < 2^100 + 5*2^64 + 2^64 + 3*2^65 + 2^97 < 2^101
-  // out[3] < 2^100 + 7*2^64 + 7*2^96 + 3*2^64 < 2^101
-}
-
-// felem_reduce_zero105 converts a larger longfelem into an felem.
-// On entry:
-//   in[0] < 2^71
-// On exit:
-//   out[i] < 2^106
-static void felem_reduce_zero105(felem out, const longfelem in) {
-    out[0] = zero105[0] + in[0];
-    out[1] = zero105[1] + in[1];
-    out[2] = zero105[2] + in[2];
-    out[3] = zero105[3] + in[3];
-
-    felem_reduce_(out, in);
-
-    // out[0] > 2^105 - 2^41 - 2^9 - 2^71 - 2^103 - 2^71 - 2^103 > 0
-    // out[1] > 2^105 - 2^71 - 2^103 > 0
-    // out[2] > 2^105 - 2^41 + 2^9 - 2^71 - 2^103 > 0
-    // out[3] > 2^105 - 2^41 + 2^9 - 2^71 - 2^103 - 2^103 > 0
-    //
-    // out[0] < 2^105 + 2^71 + 2^71 + 2^103 < 2^106
-    // out[1] < 2^105 + 2^71 + 2^71 + 2^103 < 2^106
-    // out[2] < 2^105 + 2^71 + 2^71 + 2^71 + 2^103 < 2^106
-    // out[3] < 2^105 + 2^71 + 2^103 + 2^71 < 2^106
-}
-
-// subtract_u64 sets *result = *result - v and *carry to one if the
-// subtraction underflowed.
-static void subtract_u64(uint64_t *result, uint64_t *carry, uint64_t v) {
-  uint128_t r = *result;
-  r -= v;
-  *carry = (r >> 64) & 1;
-  *result = (uint64_t)r;
-}
-
-// felem_contract converts |in| to its unique, minimal representation. On
-// entry: in[i] < 2^109.
-static void felem_contract(smallfelem out, const felem in) {
-  uint64_t all_equal_so_far = 0, result = 0;
-
-  felem_shrink(out, in);
-  // small is minimal except that the value might be > p
-
-  all_equal_so_far--;
-  // We are doing a constant time test if out >= kPrime. We need to compare
-  // each uint64_t, from most-significant to least significant. For each one, if
-  // all words so far have been equal (m is all ones) then a non-equal
-  // result is the answer. Otherwise we continue.
-  for (size_t i = 3; i < 4; i--) {
-    uint64_t equal;
-    uint128_t a = ((uint128_t)kPrime[i]) - out[i];
-    // if out[i] > kPrime[i] then a will underflow and the high 64-bits
-    // will all be set.
-    result |= all_equal_so_far & ((uint64_t)(a >> 64));
-
-    // if kPrime[i] == out[i] then |equal| will be all zeros and the
-    // decrement will make it all ones.
-    equal = kPrime[i] ^ out[i];
-    equal--;
-    equal &= equal << 32;
-    equal &= equal << 16;
-    equal &= equal << 8;
-    equal &= equal << 4;
-    equal &= equal << 2;
-    equal &= equal << 1;
-    equal = ((int64_t)equal) >> 63;
-
-    all_equal_so_far &= equal;
-  }
-
-  // if all_equal_so_far is still all ones then the two values are equal
-  // and so out >= kPrime is true.
-  result |= all_equal_so_far;
-
-  // if out >= kPrime then we subtract kPrime.
-  uint64_t carry;
-  subtract_u64(&out[0], &carry, result & kPrime[0]);
-  subtract_u64(&out[1], &carry, carry);
-  subtract_u64(&out[2], &carry, carry);
-  subtract_u64(&out[3], &carry, carry);
-
-  subtract_u64(&out[1], &carry, result & kPrime[1]);
-  subtract_u64(&out[2], &carry, carry);
-  subtract_u64(&out[3], &carry, carry);
-
-  subtract_u64(&out[2], &carry, result & kPrime[2]);
-  subtract_u64(&out[3], &carry, carry);
-
-  subtract_u64(&out[3], &carry, result & kPrime[3]);
-}
-
-// felem_is_zero returns a limb with all bits set if |in| == 0 (mod p) and 0
-// otherwise.
-// On entry:
-//   small[i] < 2^64
-static limb smallfelem_is_zero(const smallfelem small) {
-  limb result;
-  uint64_t is_p;
-
-  uint64_t is_zero = small[0] | small[1] | small[2] | small[3];
-  is_zero--;
-  is_zero &= is_zero << 32;
-  is_zero &= is_zero << 16;
-  is_zero &= is_zero << 8;
-  is_zero &= is_zero << 4;
-  is_zero &= is_zero << 2;
-  is_zero &= is_zero << 1;
-  is_zero = ((int64_t)is_zero) >> 63;
-
-  is_p = (small[0] ^ kPrime[0]) | (small[1] ^ kPrime[1]) |
-         (small[2] ^ kPrime[2]) | (small[3] ^ kPrime[3]);
-  is_p--;
-  is_p &= is_p << 32;
-  is_p &= is_p << 16;
-  is_p &= is_p << 8;
-  is_p &= is_p << 4;
-  is_p &= is_p << 2;
-  is_p &= is_p << 1;
-  is_p = ((int64_t)is_p) >> 63;
-
-  is_zero |= is_p;
-
-  result = is_zero;
-  result |= ((limb)is_zero) << 64;
-  return result;
-}
-
-// felem_inv calculates |out| = |in|^{-1}
-//
-// Based on Fermat's Little Theorem:
-//   a^p = a (mod p)
-//   a^{p-1} = 1 (mod p)
-//   a^{p-2} = a^{-1} (mod p)
-static void felem_inv(felem out, const felem in) {
-  felem ftmp, ftmp2;
-  // each e_I will hold |in|^{2^I - 1}
-  felem e2, e4, e8, e16, e32, e64;
-  longfelem tmp;
-
-  felem_square(tmp, in);
-  felem_reduce(ftmp, tmp);  // 2^1
-  felem_mul(tmp, in, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^2 - 2^0
-  felem_assign(e2, ftmp);
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^3 - 2^1
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^4 - 2^2
-  felem_mul(tmp, ftmp, e2);
-  felem_reduce(ftmp, tmp);  // 2^4 - 2^0
-  felem_assign(e4, ftmp);
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^5 - 2^1
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^6 - 2^2
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^7 - 2^3
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);  // 2^8 - 2^4
-  felem_mul(tmp, ftmp, e4);
-  felem_reduce(ftmp, tmp);  // 2^8 - 2^0
-  felem_assign(e8, ftmp);
-  for (size_t i = 0; i < 8; i++) {
-    felem_square(tmp, ftmp);
-    felem_reduce(ftmp, tmp);
-  }  // 2^16 - 2^8
-  felem_mul(tmp, ftmp, e8);
-  felem_reduce(ftmp, tmp);  // 2^16 - 2^0
-  felem_assign(e16, ftmp);
-  for (size_t i = 0; i < 16; i++) {
-    felem_square(tmp, ftmp);
-    felem_reduce(ftmp, tmp);
-  }  // 2^32 - 2^16
-  felem_mul(tmp, ftmp, e16);
-  felem_reduce(ftmp, tmp);  // 2^32 - 2^0
-  felem_assign(e32, ftmp);
-  for (size_t i = 0; i < 32; i++) {
-    felem_square(tmp, ftmp);
-    felem_reduce(ftmp, tmp);
-  }  // 2^64 - 2^32
-  felem_assign(e64, ftmp);
-  felem_mul(tmp, ftmp, in);
-  felem_reduce(ftmp, tmp);  // 2^64 - 2^32 + 2^0
-  for (size_t i = 0; i < 192; i++) {
-    felem_square(tmp, ftmp);
-    felem_reduce(ftmp, tmp);
-  }  // 2^256 - 2^224 + 2^192
-
-  felem_mul(tmp, e64, e32);
-  felem_reduce(ftmp2, tmp);  // 2^64 - 2^0
-  for (size_t i = 0; i < 16; i++) {
-    felem_square(tmp, ftmp2);
-    felem_reduce(ftmp2, tmp);
-  }  // 2^80 - 2^16
-  felem_mul(tmp, ftmp2, e16);
-  felem_reduce(ftmp2, tmp);  // 2^80 - 2^0
-  for (size_t i = 0; i < 8; i++) {
-    felem_square(tmp, ftmp2);
-    felem_reduce(ftmp2, tmp);
-  }  // 2^88 - 2^8
-  felem_mul(tmp, ftmp2, e8);
-  felem_reduce(ftmp2, tmp);  // 2^88 - 2^0
-  for (size_t i = 0; i < 4; i++) {
-    felem_square(tmp, ftmp2);
-    felem_reduce(ftmp2, tmp);
-  }  // 2^92 - 2^4
-  felem_mul(tmp, ftmp2, e4);
-  felem_reduce(ftmp2, tmp);  // 2^92 - 2^0
-  felem_square(tmp, ftmp2);
-  felem_reduce(ftmp2, tmp);  // 2^93 - 2^1
-  felem_square(tmp, ftmp2);
-  felem_reduce(ftmp2, tmp);  // 2^94 - 2^2
-  felem_mul(tmp, ftmp2, e2);
-  felem_reduce(ftmp2, tmp);  // 2^94 - 2^0
-  felem_square(tmp, ftmp2);
-  felem_reduce(ftmp2, tmp);  // 2^95 - 2^1
-  felem_square(tmp, ftmp2);
-  felem_reduce(ftmp2, tmp);  // 2^96 - 2^2
-  felem_mul(tmp, ftmp2, in);
-  felem_reduce(ftmp2, tmp);  // 2^96 - 3
-
-  felem_mul(tmp, ftmp2, ftmp);
-  felem_reduce(out, tmp);  // 2^256 - 2^224 + 2^192 + 2^96 - 3
-}
-
-// Group operations
-// ----------------
-//
-// Building on top of the field operations we have the operations on the
-// elliptic curve group itself. Points on the curve are represented in Jacobian
-// coordinates.
-
-// point_double calculates 2*(x_in, y_in, z_in)
-//
-// The method is taken from:
-//   http://hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-3.html#doubling-dbl-2001-b
-//
-// Outputs can equal corresponding inputs, i.e., x_out == x_in is allowed.
-// while x_out == y_in is not (maybe this works, but it's not tested).
-static void point_double(felem x_out, felem y_out, felem z_out,
-                         const felem x_in, const felem y_in, const felem z_in) {
-  longfelem tmp, tmp2;
-  felem delta, gamma, beta, alpha, ftmp, ftmp2;
-  smallfelem small1, small2;
-
-  felem_assign(ftmp, x_in);
-  // ftmp[i] < 2^106
-  felem_assign(ftmp2, x_in);
-  // ftmp2[i] < 2^106
-
-  // delta = z^2
-  felem_square(tmp, z_in);
-  felem_reduce(delta, tmp);
-  // delta[i] < 2^101
-
-  // gamma = y^2
-  felem_square(tmp, y_in);
-  felem_reduce(gamma, tmp);
-  // gamma[i] < 2^101
-  felem_shrink(small1, gamma);
-
-  // beta = x*gamma
-  felem_small_mul(tmp, small1, x_in);
-  felem_reduce(beta, tmp);
-  // beta[i] < 2^101
-
-  // alpha = 3*(x-delta)*(x+delta)
-  felem_diff(ftmp, delta);
-  // ftmp[i] < 2^105 + 2^106 < 2^107
-  felem_sum(ftmp2, delta);
-  // ftmp2[i] < 2^105 + 2^106 < 2^107
-  felem_scalar(ftmp2, 3);
-  // ftmp2[i] < 3 * 2^107 < 2^109
-  felem_mul(tmp, ftmp, ftmp2);
-  felem_reduce(alpha, tmp);
-  // alpha[i] < 2^101
-  felem_shrink(small2, alpha);
-
-  // x' = alpha^2 - 8*beta
-  smallfelem_square(tmp, small2);
-  felem_reduce(x_out, tmp);
-  felem_assign(ftmp, beta);
-  felem_scalar(ftmp, 8);
-  // ftmp[i] < 8 * 2^101 = 2^104
-  felem_diff(x_out, ftmp);
-  // x_out[i] < 2^105 + 2^101 < 2^106
-
-  // z' = (y + z)^2 - gamma - delta
-  felem_sum(delta, gamma);
-  // delta[i] < 2^101 + 2^101 = 2^102
-  felem_assign(ftmp, y_in);
-  felem_sum(ftmp, z_in);
-  // ftmp[i] < 2^106 + 2^106 = 2^107
-  felem_square(tmp, ftmp);
-  felem_reduce(z_out, tmp);
-  felem_diff(z_out, delta);
-  // z_out[i] < 2^105 + 2^101 < 2^106
-
-  // y' = alpha*(4*beta - x') - 8*gamma^2
-  felem_scalar(beta, 4);
-  // beta[i] < 4 * 2^101 = 2^103
-  felem_diff_zero107(beta, x_out);
-  // beta[i] < 2^107 + 2^103 < 2^108
-  felem_small_mul(tmp, small2, beta);
-  // tmp[i] < 7 * 2^64 < 2^67
-  smallfelem_square(tmp2, small1);
-  // tmp2[i] < 7 * 2^64
-  longfelem_scalar(tmp2, 8);
-  // tmp2[i] < 8 * 7 * 2^64 = 7 * 2^67
-  longfelem_diff(tmp, tmp2);
-  // tmp[i] < 2^67 + 2^70 + 2^40 < 2^71
-  felem_reduce_zero105(y_out, tmp);
-  // y_out[i] < 2^106
-}
-
-// point_double_small is the same as point_double, except that it operates on
-// smallfelems.
-static void point_double_small(smallfelem x_out, smallfelem y_out,
-                               smallfelem z_out, const smallfelem x_in,
-                               const smallfelem y_in, const smallfelem z_in) {
-  felem felem_x_out, felem_y_out, felem_z_out;
-  felem felem_x_in, felem_y_in, felem_z_in;
-
-  smallfelem_expand(felem_x_in, x_in);
-  smallfelem_expand(felem_y_in, y_in);
-  smallfelem_expand(felem_z_in, z_in);
-  point_double(felem_x_out, felem_y_out, felem_z_out, felem_x_in, felem_y_in,
-               felem_z_in);
-  felem_shrink(x_out, felem_x_out);
-  felem_shrink(y_out, felem_y_out);
-  felem_shrink(z_out, felem_z_out);
-}
-
-// p256_copy_conditional copies in to out iff mask is all ones.
-static void p256_copy_conditional(felem out, const felem in, limb mask) {
-  for (size_t i = 0; i < NLIMBS; ++i) {
-    const limb tmp = mask & (in[i] ^ out[i]);
-    out[i] ^= tmp;
-  }
-}
-
-// copy_small_conditional copies in to out iff mask is all ones.
-static void copy_small_conditional(felem out, const smallfelem in, limb mask) {
-  const uint64_t mask64 = mask;
-  for (size_t i = 0; i < NLIMBS; ++i) {
-    out[i] = ((limb)(in[i] & mask64)) | (out[i] & ~mask);
-  }
-}
-
-// point_add calcuates (x1, y1, z1) + (x2, y2, z2)
-//
-// The method is taken from:
-//   http://hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-3.html#addition-add-2007-bl,
-// adapted for mixed addition (z2 = 1, or z2 = 0 for the point at infinity).
-//
-// This function includes a branch for checking whether the two input points
-// are equal, (while not equal to the point at infinity). This case never
-// happens during single point multiplication, so there is no timing leak for
-// ECDH or ECDSA signing.
-static void point_add(felem x3, felem y3, felem z3, const felem x1,
-                      const felem y1, const felem z1, const int mixed,
-                      const smallfelem x2, const smallfelem y2,
-                      const smallfelem z2) {
-  felem ftmp, ftmp2, ftmp3, ftmp4, ftmp5, ftmp6, x_out, y_out, z_out;
-  longfelem tmp, tmp2;
-  smallfelem small1, small2, small3, small4, small5;
-  limb x_equal, y_equal, z1_is_zero, z2_is_zero;
-
-  felem_shrink(small3, z1);
-
-  z1_is_zero = smallfelem_is_zero(small3);
-  z2_is_zero = smallfelem_is_zero(z2);
-
-  // ftmp = z1z1 = z1**2
-  smallfelem_square(tmp, small3);
-  felem_reduce(ftmp, tmp);
-  // ftmp[i] < 2^101
-  felem_shrink(small1, ftmp);
-
-  if (!mixed) {
-    // ftmp2 = z2z2 = z2**2
-    smallfelem_square(tmp, z2);
-    felem_reduce(ftmp2, tmp);
-    // ftmp2[i] < 2^101
-    felem_shrink(small2, ftmp2);
-
-    felem_shrink(small5, x1);
-
-    // u1 = ftmp3 = x1*z2z2
-    smallfelem_mul(tmp, small5, small2);
-    felem_reduce(ftmp3, tmp);
-    // ftmp3[i] < 2^101
-
-    // ftmp5 = z1 + z2
-    felem_assign(ftmp5, z1);
-    felem_small_sum(ftmp5, z2);
-    // ftmp5[i] < 2^107
-
-    // ftmp5 = (z1 + z2)**2 - (z1z1 + z2z2) = 2z1z2
-    felem_square(tmp, ftmp5);
-    felem_reduce(ftmp5, tmp);
-    // ftmp2 = z2z2 + z1z1
-    felem_sum(ftmp2, ftmp);
-    // ftmp2[i] < 2^101 + 2^101 = 2^102
-    felem_diff(ftmp5, ftmp2);
-    // ftmp5[i] < 2^105 + 2^101 < 2^106
-
-    // ftmp2 = z2 * z2z2
-    smallfelem_mul(tmp, small2, z2);
-    felem_reduce(ftmp2, tmp);
-
-    // s1 = ftmp2 = y1 * z2**3
-    felem_mul(tmp, y1, ftmp2);
-    felem_reduce(ftmp6, tmp);
-    // ftmp6[i] < 2^101
-  } else {
-    // We'll assume z2 = 1 (special case z2 = 0 is handled later).
-
-    // u1 = ftmp3 = x1*z2z2
-    felem_assign(ftmp3, x1);
-    // ftmp3[i] < 2^106
-
-    // ftmp5 = 2z1z2
-    felem_assign(ftmp5, z1);
-    felem_scalar(ftmp5, 2);
-    // ftmp5[i] < 2*2^106 = 2^107
-
-    // s1 = ftmp2 = y1 * z2**3
-    felem_assign(ftmp6, y1);
-    // ftmp6[i] < 2^106
-  }
-
-  // u2 = x2*z1z1
-  smallfelem_mul(tmp, x2, small1);
-  felem_reduce(ftmp4, tmp);
-
-  // h = ftmp4 = u2 - u1
-  felem_diff_zero107(ftmp4, ftmp3);
-  // ftmp4[i] < 2^107 + 2^101 < 2^108
-  felem_shrink(small4, ftmp4);
-
-  x_equal = smallfelem_is_zero(small4);
-
-  // z_out = ftmp5 * h
-  felem_small_mul(tmp, small4, ftmp5);
-  felem_reduce(z_out, tmp);
-  // z_out[i] < 2^101
-
-  // ftmp = z1 * z1z1
-  smallfelem_mul(tmp, small1, small3);
-  felem_reduce(ftmp, tmp);
-
-  // s2 = tmp = y2 * z1**3
-  felem_small_mul(tmp, y2, ftmp);
-  felem_reduce(ftmp5, tmp);
-
-  // r = ftmp5 = (s2 - s1)*2
-  felem_diff_zero107(ftmp5, ftmp6);
-  // ftmp5[i] < 2^107 + 2^107 = 2^108
-  felem_scalar(ftmp5, 2);
-  // ftmp5[i] < 2^109
-  felem_shrink(small1, ftmp5);
-  y_equal = smallfelem_is_zero(small1);
-
-  if (x_equal && y_equal && !z1_is_zero && !z2_is_zero) {
-    point_double(x3, y3, z3, x1, y1, z1);
-    return;
-  }
-
-  // I = ftmp = (2h)**2
-  felem_assign(ftmp, ftmp4);
-  felem_scalar(ftmp, 2);
-  // ftmp[i] < 2*2^108 = 2^109
-  felem_square(tmp, ftmp);
-  felem_reduce(ftmp, tmp);
-
-  // J = ftmp2 = h * I
-  felem_mul(tmp, ftmp4, ftmp);
-  felem_reduce(ftmp2, tmp);
-
-  // V = ftmp4 = U1 * I
-  felem_mul(tmp, ftmp3, ftmp);
-  felem_reduce(ftmp4, tmp);
-
-  // x_out = r**2 - J - 2V
-  smallfelem_square(tmp, small1);
-  felem_reduce(x_out, tmp);
-  felem_assign(ftmp3, ftmp4);
-  felem_scalar(ftmp4, 2);
-  felem_sum(ftmp4, ftmp2);
-  // ftmp4[i] < 2*2^101 + 2^101 < 2^103
-  felem_diff(x_out, ftmp4);
-  // x_out[i] < 2^105 + 2^101
-
-  // y_out = r(V-x_out) - 2 * s1 * J
-  felem_diff_zero107(ftmp3, x_out);
-  // ftmp3[i] < 2^107 + 2^101 < 2^108
-  felem_small_mul(tmp, small1, ftmp3);
-  felem_mul(tmp2, ftmp6, ftmp2);
-  longfelem_scalar(tmp2, 2);
-  // tmp2[i] < 2*2^67 = 2^68
-  longfelem_diff(tmp, tmp2);
-  // tmp[i] < 2^67 + 2^70 + 2^40 < 2^71
-  felem_reduce_zero105(y_out, tmp);
-  // y_out[i] < 2^106
-
-  copy_small_conditional(x_out, x2, z1_is_zero);
-  p256_copy_conditional(x_out, x1, z2_is_zero);
-  copy_small_conditional(y_out, y2, z1_is_zero);
-  p256_copy_conditional(y_out, y1, z2_is_zero);
-  copy_small_conditional(z_out, z2, z1_is_zero);
-  p256_copy_conditional(z_out, z1, z2_is_zero);
-  felem_assign(x3, x_out);
-  felem_assign(y3, y_out);
-  felem_assign(z3, z_out);
-}
-
-// point_add_small is the same as point_add, except that it operates on
-// smallfelems.
-static void point_add_small(smallfelem x3, smallfelem y3, smallfelem z3,
-                            smallfelem x1, smallfelem y1, smallfelem z1,
-                            smallfelem x2, smallfelem y2, smallfelem z2) {
-  felem felem_x3, felem_y3, felem_z3;
-  felem felem_x1, felem_y1, felem_z1;
-  smallfelem_expand(felem_x1, x1);
-  smallfelem_expand(felem_y1, y1);
-  smallfelem_expand(felem_z1, z1);
-  point_add(felem_x3, felem_y3, felem_z3, felem_x1, felem_y1, felem_z1, 0, x2,
-            y2, z2);
-  felem_shrink(x3, felem_x3);
-  felem_shrink(y3, felem_y3);
-  felem_shrink(z3, felem_z3);
-}
-
-// Base point pre computation
-// --------------------------
-//
-// Two different sorts of precomputed tables are used in the following code.
-// Each contain various points on the curve, where each point is three field
-// elements (x, y, z).
-//
-// For the base point table, z is usually 1 (0 for the point at infinity).
-// This table has 2 * 16 elements, starting with the following:
-// index | bits    | point
-// ------+---------+------------------------------
-//     0 | 0 0 0 0 | 0G
-//     1 | 0 0 0 1 | 1G
-//     2 | 0 0 1 0 | 2^64G
-//     3 | 0 0 1 1 | (2^64 + 1)G
-//     4 | 0 1 0 0 | 2^128G
-//     5 | 0 1 0 1 | (2^128 + 1)G
-//     6 | 0 1 1 0 | (2^128 + 2^64)G
-//     7 | 0 1 1 1 | (2^128 + 2^64 + 1)G
-//     8 | 1 0 0 0 | 2^192G
-//     9 | 1 0 0 1 | (2^192 + 1)G
-//    10 | 1 0 1 0 | (2^192 + 2^64)G
-//    11 | 1 0 1 1 | (2^192 + 2^64 + 1)G
-//    12 | 1 1 0 0 | (2^192 + 2^128)G
-//    13 | 1 1 0 1 | (2^192 + 2^128 + 1)G
-//    14 | 1 1 1 0 | (2^192 + 2^128 + 2^64)G
-//    15 | 1 1 1 1 | (2^192 + 2^128 + 2^64 + 1)G
-// followed by a copy of this with each element multiplied by 2^32.
-//
-// The reason for this is so that we can clock bits into four different
-// locations when doing simple scalar multiplies against the base point,
-// and then another four locations using the second 16 elements.
-//
-// Tables for other points have table[i] = iG for i in 0 .. 16.
-
-// g_pre_comp is the table of precomputed base points
-static const smallfelem g_pre_comp[2][16][3] = {
-    {{{0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}},
-     {{0xf4a13945d898c296, 0x77037d812deb33a0, 0xf8bce6e563a440f2,
-       0x6b17d1f2e12c4247},
-      {0xcbb6406837bf51f5, 0x2bce33576b315ece, 0x8ee7eb4a7c0f9e16,
-       0x4fe342e2fe1a7f9b},
-      {1, 0, 0, 0}},
-     {{0x90e75cb48e14db63, 0x29493baaad651f7e, 0x8492592e326e25de,
-       0x0fa822bc2811aaa5},
-      {0xe41124545f462ee7, 0x34b1a65050fe82f5, 0x6f4ad4bcb3df188b,
-       0xbff44ae8f5dba80d},
-      {1, 0, 0, 0}},
-     {{0x93391ce2097992af, 0xe96c98fd0d35f1fa, 0xb257c0de95e02789,
-       0x300a4bbc89d6726f},
-      {0xaa54a291c08127a0, 0x5bb1eeada9d806a5, 0x7f1ddb25ff1e3c6f,
-       0x72aac7e0d09b4644},
-      {1, 0, 0, 0}},
-     {{0x57c84fc9d789bd85, 0xfc35ff7dc297eac3, 0xfb982fd588c6766e,
-       0x447d739beedb5e67},
-      {0x0c7e33c972e25b32, 0x3d349b95a7fae500, 0xe12e9d953a4aaff7,
-       0x2d4825ab834131ee},
-      {1, 0, 0, 0}},
-     {{0x13949c932a1d367f, 0xef7fbd2b1a0a11b7, 0xddc6068bb91dfc60,
-       0xef9519328a9c72ff},
-      {0x196035a77376d8a8, 0x23183b0895ca1740, 0xc1ee9807022c219c,
-       0x611e9fc37dbb2c9b},
-      {1, 0, 0, 0}},
-     {{0xcae2b1920b57f4bc, 0x2936df5ec6c9bc36, 0x7dea6482e11238bf,
-       0x550663797b51f5d8},
-      {0x44ffe216348a964c, 0x9fb3d576dbdefbe1, 0x0afa40018d9d50e5,
-       0x157164848aecb851},
-      {1, 0, 0, 0}},
-     {{0xe48ecafffc5cde01, 0x7ccd84e70d715f26, 0xa2e8f483f43e4391,
-       0xeb5d7745b21141ea},
-      {0xcac917e2731a3479, 0x85f22cfe2844b645, 0x0990e6a158006cee,
-       0xeafd72ebdbecc17b},
-      {1, 0, 0, 0}},
-     {{0x6cf20ffb313728be, 0x96439591a3c6b94a, 0x2736ff8344315fc5,
-       0xa6d39677a7849276},
-      {0xf2bab833c357f5f4, 0x824a920c2284059b, 0x66b8babd2d27ecdf,
-       0x674f84749b0b8816},
-      {1, 0, 0, 0}},
-     {{0x2df48c04677c8a3e, 0x74e02f080203a56b, 0x31855f7db8c7fedb,
-       0x4e769e7672c9ddad},
-      {0xa4c36165b824bbb0, 0xfb9ae16f3b9122a5, 0x1ec0057206947281,
-       0x42b99082de830663},
-      {1, 0, 0, 0}},
-     {{0x6ef95150dda868b9, 0xd1f89e799c0ce131, 0x7fdc1ca008a1c478,
-       0x78878ef61c6ce04d},
-      {0x9c62b9121fe0d976, 0x6ace570ebde08d4f, 0xde53142c12309def,
-       0xb6cb3f5d7b72c321},
-      {1, 0, 0, 0}},
-     {{0x7f991ed2c31a3573, 0x5b82dd5bd54fb496, 0x595c5220812ffcae,
-       0x0c88bc4d716b1287},
-      {0x3a57bf635f48aca8, 0x7c8181f4df2564f3, 0x18d1b5b39c04e6aa,
-       0xdd5ddea3f3901dc6},
-      {1, 0, 0, 0}},
-     {{0xe96a79fb3e72ad0c, 0x43a0a28c42ba792f, 0xefe0a423083e49f3,
-       0x68f344af6b317466},
-      {0xcdfe17db3fb24d4a, 0x668bfc2271f5c626, 0x604ed93c24d67ff3,
-       0x31b9c405f8540a20},
-      {1, 0, 0, 0}},
-     {{0xd36b4789a2582e7f, 0x0d1a10144ec39c28, 0x663c62c3edbad7a0,
-       0x4052bf4b6f461db9},
-      {0x235a27c3188d25eb, 0xe724f33999bfcc5b, 0x862be6bd71d70cc8,
-       0xfecf4d5190b0fc61},
-      {1, 0, 0, 0}},
-     {{0x74346c10a1d4cfac, 0xafdf5cc08526a7a4, 0x123202a8f62bff7a,
-       0x1eddbae2c802e41a},
-      {0x8fa0af2dd603f844, 0x36e06b7e4c701917, 0x0c45f45273db33a0,
-       0x43104d86560ebcfc},
-      {1, 0, 0, 0}},
-     {{0x9615b5110d1d78e5, 0x66b0de3225c4744b, 0x0a4a46fb6aaf363a,
-       0xb48e26b484f7a21c},
-      {0x06ebb0f621a01b2d, 0xc004e4048b7b0f98, 0x64131bcdfed6f668,
-       0xfac015404d4d3dab},
-      {1, 0, 0, 0}}},
-    {{{0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}},
-     {{0x3a5a9e22185a5943, 0x1ab919365c65dfb6, 0x21656b32262c71da,
-       0x7fe36b40af22af89},
-      {0xd50d152c699ca101, 0x74b3d5867b8af212, 0x9f09f40407dca6f1,
-       0xe697d45825b63624},
-      {1, 0, 0, 0}},
-     {{0xa84aa9397512218e, 0xe9a521b074ca0141, 0x57880b3a18a2e902,
-       0x4a5b506612a677a6},
-      {0x0beada7a4c4f3840, 0x626db15419e26d9d, 0xc42604fbe1627d40,
-       0xeb13461ceac089f1},
-      {1, 0, 0, 0}},
-     {{0xf9faed0927a43281, 0x5e52c4144103ecbc, 0xc342967aa815c857,
-       0x0781b8291c6a220a},
-      {0x5a8343ceeac55f80, 0x88f80eeee54a05e3, 0x97b2a14f12916434,
-       0x690cde8df0151593},
-      {1, 0, 0, 0}},
-     {{0xaee9c75df7f82f2a, 0x9e4c35874afdf43a, 0xf5622df437371326,
-       0x8a535f566ec73617},
-      {0xc5f9a0ac223094b7, 0xcde533864c8c7669, 0x37e02819085a92bf,
-       0x0455c08468b08bd7},
-      {1, 0, 0, 0}},
-     {{0x0c0a6e2c9477b5d9, 0xf9a4bf62876dc444, 0x5050a949b6cdc279,
-       0x06bada7ab77f8276},
-      {0xc8b4aed1ea48dac9, 0xdebd8a4b7ea1070f, 0x427d49101366eb70,
-       0x5b476dfd0e6cb18a},
-      {1, 0, 0, 0}},
-     {{0x7c5c3e44278c340a, 0x4d54606812d66f3b, 0x29a751b1ae23c5d8,
-       0x3e29864e8a2ec908},
-      {0x142d2a6626dbb850, 0xad1744c4765bd780, 0x1f150e68e322d1ed,
-       0x239b90ea3dc31e7e},
-      {1, 0, 0, 0}},
-     {{0x78c416527a53322a, 0x305dde6709776f8e, 0xdbcab759f8862ed4,
-       0x820f4dd949f72ff7},
-      {0x6cc544a62b5debd4, 0x75be5d937b4e8cc4, 0x1b481b1b215c14d3,
-       0x140406ec783a05ec},
-      {1, 0, 0, 0}},
-     {{0x6a703f10e895df07, 0xfd75f3fa01876bd8, 0xeb5b06e70ce08ffe,
-       0x68f6b8542783dfee},
-      {0x90c76f8a78712655, 0xcf5293d2f310bf7f, 0xfbc8044dfda45028,
-       0xcbe1feba92e40ce6},
-      {1, 0, 0, 0}},
-     {{0xe998ceea4396e4c1, 0xfc82ef0b6acea274, 0x230f729f2250e927,
-       0xd0b2f94d2f420109},
-      {0x4305adddb38d4966, 0x10b838f8624c3b45, 0x7db2636658954e7a,
-       0x971459828b0719e5},
-      {1, 0, 0, 0}},
-     {{0x4bd6b72623369fc9, 0x57f2929e53d0b876, 0xc2d5cba4f2340687,
-       0x961610004a866aba},
-      {0x49997bcd2e407a5e, 0x69ab197d92ddcb24, 0x2cf1f2438fe5131c,
-       0x7acb9fadcee75e44},
-      {1, 0, 0, 0}},
-     {{0x254e839423d2d4c0, 0xf57f0c917aea685b, 0xa60d880f6f75aaea,
-       0x24eb9acca333bf5b},
-      {0xe3de4ccb1cda5dea, 0xfeef9341c51a6b4f, 0x743125f88bac4c4d,
-       0x69f891c5acd079cc},
-      {1, 0, 0, 0}},
-     {{0xeee44b35702476b5, 0x7ed031a0e45c2258, 0xb422d1e7bd6f8514,
-       0xe51f547c5972a107},
-      {0xa25bcd6fc9cf343d, 0x8ca922ee097c184e, 0xa62f98b3a9fe9a06,
-       0x1c309a2b25bb1387},
-      {1, 0, 0, 0}},
-     {{0x9295dbeb1967c459, 0xb00148833472c98e, 0xc504977708011828,
-       0x20b87b8aa2c4e503},
-      {0x3063175de057c277, 0x1bd539338fe582dd, 0x0d11adef5f69a044,
-       0xf5c6fa49919776be},
-      {1, 0, 0, 0}},
-     {{0x8c944e760fd59e11, 0x3876cba1102fad5f, 0xa454c3fad83faa56,
-       0x1ed7d1b9332010b9},
-      {0xa1011a270024b889, 0x05e4d0dcac0cd344, 0x52b520f0eb6a2a24,
-       0x3a2b03f03217257a},
-      {1, 0, 0, 0}},
-     {{0xf20fc2afdf1d043d, 0xf330240db58d5a62, 0xfc7d229ca0058c3b,
-       0x15fee545c78dd9f6},
-      {0x501e82885bc98cda, 0x41ef80e5d046ac04, 0x557d9f49461210fb,
-       0x4ab5b6b2b8753f81},
-      {1, 0, 0, 0}}}};
-
-// select_point selects the |idx|th point from a precomputation table and
-// copies it to out.
-static void select_point(const uint64_t idx, size_t size,
-                         const smallfelem pre_comp[/*size*/][3],
-                         smallfelem out[3]) {
-  uint64_t *outlimbs = &out[0][0];
-  OPENSSL_memset(outlimbs, 0, 3 * sizeof(smallfelem));
-
-  for (size_t i = 0; i < size; i++) {
-    const uint64_t *inlimbs = (const uint64_t *)&pre_comp[i][0][0];
-    uint64_t mask = i ^ idx;
-    mask |= mask >> 4;
-    mask |= mask >> 2;
-    mask |= mask >> 1;
-    mask &= 1;
-    mask--;
-    for (size_t j = 0; j < NLIMBS * 3; j++) {
-      outlimbs[j] |= inlimbs[j] & mask;
-    }
-  }
-}
-
-// get_bit returns the |i|th bit in |in|
-static char get_bit(const felem_bytearray in, int i) {
-  if (i < 0 || i >= 256) {
-    return 0;
-  }
-  return (in[i >> 3] >> (i & 7)) & 1;
-}
-
-// Interleaved point multiplication using precomputed point multiples: The
-// small point multiples 0*P, 1*P, ..., 17*P are in p_pre_comp, the scalar
-// in p_scalar, if non-NULL. If g_scalar is non-NULL, we also add this multiple
-// of the generator, using certain (large) precomputed multiples in g_pre_comp.
-// Output point (X, Y, Z) is stored in x_out, y_out, z_out.
-static void batch_mul(felem x_out, felem y_out, felem z_out,
-                      const uint8_t *p_scalar, const uint8_t *g_scalar,
-                      const smallfelem p_pre_comp[17][3]) {
-  felem nq[3], ftmp;
-  smallfelem tmp[3];
-  uint64_t bits;
-  uint8_t sign, digit;
-
-  // set nq to the point at infinity
-  OPENSSL_memset(nq, 0, 3 * sizeof(felem));
-
-  // Loop over both scalars msb-to-lsb, interleaving additions of multiples
-  // of the generator (two in each of the last 32 rounds) and additions of p
-  // (every 5th round).
-
-  int skip = 1;  // save two point operations in the first round
-  size_t i = p_scalar != NULL ? 255 : 31;
-  for (;;) {
-    // double
-    if (!skip) {
-      point_double(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2]);
-    }
-
-    // add multiples of the generator
-    if (g_scalar != NULL && i <= 31) {
-      // first, look 32 bits upwards
-      bits = get_bit(g_scalar, i + 224) << 3;
-      bits |= get_bit(g_scalar, i + 160) << 2;
-      bits |= get_bit(g_scalar, i + 96) << 1;
-      bits |= get_bit(g_scalar, i + 32);
-      // select the point to add, in constant time
-      select_point(bits, 16, g_pre_comp[1], tmp);
-
-      if (!skip) {
-        point_add(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2], 1 /* mixed */,
-                  tmp[0], tmp[1], tmp[2]);
-      } else {
-        smallfelem_expand(nq[0], tmp[0]);
-        smallfelem_expand(nq[1], tmp[1]);
-        smallfelem_expand(nq[2], tmp[2]);
-        skip = 0;
-      }
-
-      // second, look at the current position
-      bits = get_bit(g_scalar, i + 192) << 3;
-      bits |= get_bit(g_scalar, i + 128) << 2;
-      bits |= get_bit(g_scalar, i + 64) << 1;
-      bits |= get_bit(g_scalar, i);
-      // select the point to add, in constant time
-      select_point(bits, 16, g_pre_comp[0], tmp);
-      point_add(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2], 1 /* mixed */, tmp[0],
-                tmp[1], tmp[2]);
-    }
-
-    // do other additions every 5 doublings
-    if (p_scalar != NULL && i % 5 == 0) {
-      bits = get_bit(p_scalar, i + 4) << 5;
-      bits |= get_bit(p_scalar, i + 3) << 4;
-      bits |= get_bit(p_scalar, i + 2) << 3;
-      bits |= get_bit(p_scalar, i + 1) << 2;
-      bits |= get_bit(p_scalar, i) << 1;
-      bits |= get_bit(p_scalar, i - 1);
-      ec_GFp_nistp_recode_scalar_bits(&sign, &digit, bits);
-
-      // select the point to add or subtract, in constant time.
-      select_point(digit, 17, p_pre_comp, tmp);
-      smallfelem_neg(ftmp, tmp[1]);  // (X, -Y, Z) is the negative
-                                     // point
-      copy_small_conditional(ftmp, tmp[1], (((limb)sign) - 1));
-      felem_contract(tmp[1], ftmp);
-
-      if (!skip) {
-        point_add(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2], 0 /* mixed */,
-                  tmp[0], tmp[1], tmp[2]);
-      } else {
-        smallfelem_expand(nq[0], tmp[0]);
-        smallfelem_expand(nq[1], tmp[1]);
-        smallfelem_expand(nq[2], tmp[2]);
-        skip = 0;
-      }
-    }
-
-    if (i == 0) {
-      break;
-    }
-    --i;
-  }
-  felem_assign(x_out, nq[0]);
-  felem_assign(y_out, nq[1]);
-  felem_assign(z_out, nq[2]);
-}
-
-// OPENSSL EC_METHOD FUNCTIONS
-
-// Takes the Jacobian coordinates (X, Y, Z) of a point and returns (X', Y') =
-// (X/Z^2, Y/Z^3).
-static int ec_GFp_nistp256_point_get_affine_coordinates(const EC_GROUP *group,
-                                                        const EC_POINT *point,
-                                                        BIGNUM *x, BIGNUM *y,
-                                                        BN_CTX *ctx) {
-  felem z1, z2, x_in, y_in;
-  smallfelem x_out, y_out;
-  longfelem tmp;
-
-  if (EC_POINT_is_at_infinity(group, point)) {
-    OPENSSL_PUT_ERROR(EC, EC_R_POINT_AT_INFINITY);
-    return 0;
-  }
-  if (!BN_to_felem(x_in, &point->X) ||
-      !BN_to_felem(y_in, &point->Y) ||
-      !BN_to_felem(z1, &point->Z)) {
-    return 0;
-  }
-  felem_inv(z2, z1);
-  felem_square(tmp, z2);
-  felem_reduce(z1, tmp);
-
-  if (x != NULL) {
-    felem_mul(tmp, x_in, z1);
-    felem_reduce(x_in, tmp);
-    felem_contract(x_out, x_in);
-    if (!smallfelem_to_BN(x, x_out)) {
-      OPENSSL_PUT_ERROR(EC, ERR_R_BN_LIB);
-      return 0;
-    }
-  }
-
-  if (y != NULL) {
-    felem_mul(tmp, z1, z2);
-    felem_reduce(z1, tmp);
-    felem_mul(tmp, y_in, z1);
-    felem_reduce(y_in, tmp);
-    felem_contract(y_out, y_in);
-    if (!smallfelem_to_BN(y, y_out)) {
-      OPENSSL_PUT_ERROR(EC, ERR_R_BN_LIB);
-      return 0;
-    }
-  }
-
-  return 1;
-}
-
-static int ec_GFp_nistp256_points_mul(const EC_GROUP *group, EC_POINT *r,
-                                      const EC_SCALAR *g_scalar,
-                                      const EC_POINT *p,
-                                      const EC_SCALAR *p_scalar, BN_CTX *ctx) {
-  int ret = 0;
-  BN_CTX *new_ctx = NULL;
-  BIGNUM *x, *y, *z, *tmp_scalar;
-  smallfelem p_pre_comp[17][3];
-  smallfelem x_in, y_in, z_in;
-  felem x_out, y_out, z_out;
-
-  if (ctx == NULL) {
-    ctx = new_ctx = BN_CTX_new();
-    if (ctx == NULL) {
-      return 0;
-    }
-  }
-
-  BN_CTX_start(ctx);
-  if ((x = BN_CTX_get(ctx)) == NULL ||
-      (y = BN_CTX_get(ctx)) == NULL ||
-      (z = BN_CTX_get(ctx)) == NULL ||
-      (tmp_scalar = BN_CTX_get(ctx)) == NULL) {
-    goto err;
-  }
-
-  if (p != NULL && p_scalar != NULL) {
-    // We treat NULL scalars as 0, and NULL points as points at infinity, i.e.,
-    // they contribute nothing to the linear combination.
-    OPENSSL_memset(&p_pre_comp, 0, sizeof(p_pre_comp));
-    // Precompute multiples.
-    if (!BN_to_felem(x_out, &p->X) ||
-        !BN_to_felem(y_out, &p->Y) ||
-        !BN_to_felem(z_out, &p->Z)) {
-      goto err;
-    }
-    felem_shrink(p_pre_comp[1][0], x_out);
-    felem_shrink(p_pre_comp[1][1], y_out);
-    felem_shrink(p_pre_comp[1][2], z_out);
-    for (size_t j = 2; j <= 16; ++j) {
-      if (j & 1) {
-        point_add_small(p_pre_comp[j][0], p_pre_comp[j][1],
-                        p_pre_comp[j][2], p_pre_comp[1][0],
-                        p_pre_comp[1][1], p_pre_comp[1][2],
-                        p_pre_comp[j - 1][0], p_pre_comp[j - 1][1],
-                        p_pre_comp[j - 1][2]);
-      } else {
-        point_double_small(p_pre_comp[j][0], p_pre_comp[j][1],
-                           p_pre_comp[j][2], p_pre_comp[j / 2][0],
-                           p_pre_comp[j / 2][1], p_pre_comp[j / 2][2]);
-      }
-    }
-  }
-
-  batch_mul(x_out, y_out, z_out,
-            (p != NULL && p_scalar != NULL) ? p_scalar->bytes : NULL,
-            g_scalar != NULL ? g_scalar->bytes : NULL,
-            (const smallfelem(*)[3]) & p_pre_comp);
-
-  // reduce the output to its unique minimal representation
-  felem_contract(x_in, x_out);
-  felem_contract(y_in, y_out);
-  felem_contract(z_in, z_out);
-  if (!smallfelem_to_BN(x, x_in) ||
-      !smallfelem_to_BN(y, y_in) ||
-      !smallfelem_to_BN(z, z_in)) {
-    OPENSSL_PUT_ERROR(EC, ERR_R_BN_LIB);
-    goto err;
-  }
-  ret = ec_point_set_Jprojective_coordinates_GFp(group, r, x, y, z, ctx);
-
-err:
-  BN_CTX_end(ctx);
-  BN_CTX_free(new_ctx);
-  return ret;
-}
-
-DEFINE_METHOD_FUNCTION(EC_METHOD, EC_GFp_nistp256_method) {
-  out->group_init = ec_GFp_simple_group_init;
-  out->group_finish = ec_GFp_simple_group_finish;
-  out->group_set_curve = ec_GFp_simple_group_set_curve;
-  out->point_get_affine_coordinates =
-      ec_GFp_nistp256_point_get_affine_coordinates;
-  out->mul = ec_GFp_nistp256_points_mul;
-  out->field_mul = ec_GFp_simple_field_mul;
-  out->field_sqr = ec_GFp_simple_field_sqr;
-  out->field_encode = NULL;
-  out->field_decode = NULL;
-};
-
-#endif  // 64_BIT && !WINDOWS
diff --git a/crypto/fipsmodule/ec/p256-x86_64.c b/crypto/fipsmodule/ec/p256-x86_64.c
index a9b603a..0e79b6d 100644
--- a/crypto/fipsmodule/ec/p256-x86_64.c
+++ b/crypto/fipsmodule/ec/p256-x86_64.c
@@ -446,6 +446,7 @@
   out->group_set_curve = ec_GFp_mont_group_set_curve;
   out->point_get_affine_coordinates = ecp_nistz256_get_affine;
   out->mul = ecp_nistz256_points_mul;
+  out->mul_public = ecp_nistz256_points_mul;
   out->field_mul = ec_GFp_mont_field_mul;
   out->field_sqr = ec_GFp_mont_field_sqr;
   out->field_encode = ec_GFp_mont_field_encode;
diff --git a/crypto/fipsmodule/ec/util-64.c b/crypto/fipsmodule/ec/util.c
similarity index 97%
rename from crypto/fipsmodule/ec/util-64.c
rename to crypto/fipsmodule/ec/util.c
index 0cb117b..7303a15 100644
--- a/crypto/fipsmodule/ec/util-64.c
+++ b/crypto/fipsmodule/ec/util.c
@@ -14,9 +14,6 @@
 
 #include <openssl/base.h>
 
-
-#if defined(OPENSSL_64_BIT) && !defined(OPENSSL_WINDOWS)
-
 #include <openssl/ec.h>
 
 #include "internal.h"
@@ -105,5 +102,3 @@
   *sign = s & 1;
   *digit = d;
 }
-
-#endif  // 64_BIT && !WINDOWS
diff --git a/crypto/fipsmodule/ecdsa/ecdsa.c b/crypto/fipsmodule/ecdsa/ecdsa.c
index 6571c94..9e038de 100644
--- a/crypto/fipsmodule/ecdsa/ecdsa.c
+++ b/crypto/fipsmodule/ecdsa/ecdsa.c
@@ -275,7 +275,7 @@
     OPENSSL_PUT_ERROR(ECDSA, ERR_R_MALLOC_FAILURE);
     goto err;
   }
-  if (!ec_point_mul_scalar(group, point, &u1, pub_key, &u2, ctx)) {
+  if (!ec_point_mul_scalar_public(group, point, &u1, pub_key, &u2, ctx)) {
     OPENSSL_PUT_ERROR(ECDSA, ERR_R_EC_LIB);
     goto err;
   }
diff --git a/third_party/fiat/p256.c b/third_party/fiat/p256.c
new file mode 100644
index 0000000..19a8284
--- /dev/null
+++ b/third_party/fiat/p256.c
@@ -0,0 +1,1725 @@
+// The MIT License (MIT)
+//
+// Copyright (c) 2015-2016 the fiat-crypto authors (see the AUTHORS file).
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in all
+// copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+// SOFTWARE.
+
+// The field arithmetic code is generated by Fiat
+// (https://github.com/mit-plv/fiat-crypto), which is MIT licensed.
+//
+// An implementation of the NIST P-256 elliptic curve point multiplication.
+// 256-bit Montgomery form, generated using fiat-crypto, for 64 and 32-bit.
+// Field operations with inputs in [0,p) return outputs in [0,p).
+
+#include <openssl/base.h>
+
+// MSVC does not implement uint128_t, and crashes with intrinsics
+#if defined(OPENSSL_64_BIT) && !defined(OPENSSL_WINDOWS)
+#define BORINGSSL_NISTP256_64BIT 1
+#endif
+
+#include <openssl/bn.h>
+#include <openssl/ec.h>
+#include <openssl/err.h>
+#include <openssl/mem.h>
+
+#include <string.h>
+
+#include "../../crypto/fipsmodule/delocate.h"
+#include "../../crypto/internal.h"
+#include "../../crypto/fipsmodule/ec/internal.h"
+
+
+// "intrinsics"
+
+#if defined(BORINGSSL_NISTP256_64BIT)
+
+static uint64_t mulx_u64(uint64_t a, uint64_t b, uint64_t *high) {
+  uint128_t x = (uint128_t)a * b;
+  *high = (uint64_t) (x >> 64);
+  return (uint64_t) x;
+}
+
+static uint64_t addcarryx_u64(uint8_t c, uint64_t a, uint64_t b, uint64_t *low) {
+  uint128_t x = (uint128_t)a + b + c;
+  *low = (uint64_t) x;
+  return (uint64_t) (x>>64);
+}
+
+static uint64_t subborrow_u64(uint8_t c, uint64_t a, uint64_t b, uint64_t *low) {
+  uint128_t t = ((uint128_t) b + c);
+  uint128_t x = a-t;
+  *low = (uint64_t) x;
+  return (uint8_t) (x>>127);
+}
+
+static uint64_t cmovznz_u64(uint64_t t, uint64_t z, uint64_t nz) {
+  t = -!!t; // all set if nonzero, 0 if 0
+  return (t&nz) | ((~t)&z);
+}
+
+#else
+
+static uint32_t mulx_u32(uint32_t a, uint32_t b, uint32_t *high) {
+  uint64_t x = (uint64_t)a * b;
+  *high = (uint32_t) (x >> 32);
+  return (uint32_t) x;
+}
+
+static uint32_t addcarryx_u32(uint8_t c, uint32_t a, uint32_t b, uint32_t *low) {
+  uint64_t x = (uint64_t)a + b + c;
+  *low = (uint32_t) x;
+  return (uint32_t) (x>>32);
+}
+
+static uint32_t subborrow_u32(uint8_t c, uint32_t a, uint32_t b, uint32_t *low) {
+  uint64_t t = ((uint64_t) b + c);
+  uint64_t x = a-t;
+  *low = (uint32_t) x;
+  return (uint8_t) (x>>63);
+}
+
+static uint32_t cmovznz_u32(uint32_t t, uint32_t z, uint32_t nz) {
+  t = -!!t; // all set if nonzero, 0 if 0
+  return (t&nz) | ((~t)&z);
+}
+
+#endif
+
+// fiat-crypto generated code
+
+#if defined(BORINGSSL_NISTP256_64BIT)
+
+static void fe_add(uint64_t out[4], const uint64_t in1[4], const uint64_t in2[4]) {
+  { const uint64_t x8 = in1[3];
+  { const uint64_t x9 = in1[2];
+  { const uint64_t x7 = in1[1];
+  { const uint64_t x5 = in1[0];
+  { const uint64_t x14 = in2[3];
+  { const uint64_t x15 = in2[2];
+  { const uint64_t x13 = in2[1];
+  { const uint64_t x11 = in2[0];
+  { uint64_t x17; uint8_t x18 = addcarryx_u64(0x0, x5, x11, &x17);
+  { uint64_t x20; uint8_t x21 = addcarryx_u64(x18, x7, x13, &x20);
+  { uint64_t x23; uint8_t x24 = addcarryx_u64(x21, x9, x15, &x23);
+  { uint64_t x26; uint8_t x27 = addcarryx_u64(x24, x8, x14, &x26);
+  { uint64_t x29; uint8_t x30 = subborrow_u64(0x0, x17, 0xffffffffffffffffL, &x29);
+  { uint64_t x32; uint8_t x33 = subborrow_u64(x30, x20, 0xffffffff, &x32);
+  { uint64_t x35; uint8_t x36 = subborrow_u64(x33, x23, 0x0, &x35);
+  { uint64_t x38; uint8_t x39 = subborrow_u64(x36, x26, 0xffffffff00000001L, &x38);
+  { uint64_t _1; uint8_t x42 = subborrow_u64(x39, x27, 0x0, &_1);
+  { uint64_t x43 = cmovznz_u64(x42, x38, x26);
+  { uint64_t x44 = cmovznz_u64(x42, x35, x23);
+  { uint64_t x45 = cmovznz_u64(x42, x32, x20);
+  { uint64_t x46 = cmovznz_u64(x42, x29, x17);
+  out[0] = x46;
+  out[1] = x45;
+  out[2] = x44;
+  out[3] = x43;
+  }}}}}}}}}}}}}}}}}}}}}
+}
+
+// fe_op sets out = -in
+static void fe_opp(uint64_t out[4], const uint64_t in1[4]) {
+  const uint64_t x5 = in1[3];
+  const uint64_t x6 = in1[2];
+  const uint64_t x4 = in1[1];
+  const uint64_t x2 = in1[0];
+  uint64_t x8; uint8_t x9 = subborrow_u64(0x0, 0x0, x2, &x8);
+  uint64_t x11; uint8_t x12 = subborrow_u64(x9, 0x0, x4, &x11);
+  uint64_t x14; uint8_t x15 = subborrow_u64(x12, 0x0, x6, &x14);
+  uint64_t x17; uint8_t x18 = subborrow_u64(x15, 0x0, x5, &x17);
+  uint64_t x19 = (uint64_t)cmovznz_u64(x18, 0x0, 0xffffffffffffffffL);
+  uint64_t x20 = (x19 & 0xffffffffffffffffL);
+  uint64_t x22; uint8_t x23 = addcarryx_u64(0x0, x8, x20, &x22);
+  uint64_t x24 = (x19 & 0xffffffff);
+  uint64_t x26; uint8_t x27 = addcarryx_u64(x23, x11, x24, &x26);
+  uint64_t x29; uint8_t x30 = addcarryx_u64(x27, x14, 0x0, &x29);
+  uint64_t x31 = (x19 & 0xffffffff00000001L);
+  uint64_t x33; addcarryx_u64(x30, x17, x31, &x33);
+  out[0] = x22;
+  out[1] = x26;
+  out[2] = x29;
+  out[3] = x33;
+}
+
+static void fe_mul(uint64_t out[4], const uint64_t in1[4], const uint64_t in2[4]) {
+  const uint64_t x8 = in1[3];
+  const uint64_t x9 = in1[2];
+  const uint64_t x7 = in1[1];
+  const uint64_t x5 = in1[0];
+  const uint64_t x14 = in2[3];
+  const uint64_t x15 = in2[2];
+  const uint64_t x13 = in2[1];
+  const uint64_t x11 = in2[0];
+  uint64_t x18;  uint64_t x17 = mulx_u64(x5, x11, &x18);
+  uint64_t x21;  uint64_t x20 = mulx_u64(x5, x13, &x21);
+  uint64_t x24;  uint64_t x23 = mulx_u64(x5, x15, &x24);
+  uint64_t x27;  uint64_t x26 = mulx_u64(x5, x14, &x27);
+  uint64_t x29; uint8_t x30 = addcarryx_u64(0x0, x18, x20, &x29);
+  uint64_t x32; uint8_t x33 = addcarryx_u64(x30, x21, x23, &x32);
+  uint64_t x35; uint8_t x36 = addcarryx_u64(x33, x24, x26, &x35);
+  uint64_t x38; addcarryx_u64(0x0, x36, x27, &x38);
+  uint64_t x42;  uint64_t x41 = mulx_u64(x17, 0xffffffffffffffffL, &x42);
+  uint64_t x45;  uint64_t x44 = mulx_u64(x17, 0xffffffff, &x45);
+  uint64_t x48;  uint64_t x47 = mulx_u64(x17, 0xffffffff00000001L, &x48);
+  uint64_t x50; uint8_t x51 = addcarryx_u64(0x0, x42, x44, &x50);
+  uint64_t x53; uint8_t x54 = addcarryx_u64(x51, x45, 0x0, &x53);
+  uint64_t x56; uint8_t x57 = addcarryx_u64(x54, 0x0, x47, &x56);
+  uint64_t x59; addcarryx_u64(0x0, x57, x48, &x59);
+  uint64_t _2; uint8_t x63 = addcarryx_u64(0x0, x17, x41, &_2);
+  uint64_t x65; uint8_t x66 = addcarryx_u64(x63, x29, x50, &x65);
+  uint64_t x68; uint8_t x69 = addcarryx_u64(x66, x32, x53, &x68);
+  uint64_t x71; uint8_t x72 = addcarryx_u64(x69, x35, x56, &x71);
+  uint64_t x74; uint8_t x75 = addcarryx_u64(x72, x38, x59, &x74);
+  uint64_t x78;  uint64_t x77 = mulx_u64(x7, x11, &x78);
+  uint64_t x81;  uint64_t x80 = mulx_u64(x7, x13, &x81);
+  uint64_t x84;  uint64_t x83 = mulx_u64(x7, x15, &x84);
+  uint64_t x87;  uint64_t x86 = mulx_u64(x7, x14, &x87);
+  uint64_t x89; uint8_t x90 = addcarryx_u64(0x0, x78, x80, &x89);
+  uint64_t x92; uint8_t x93 = addcarryx_u64(x90, x81, x83, &x92);
+  uint64_t x95; uint8_t x96 = addcarryx_u64(x93, x84, x86, &x95);
+  uint64_t x98; addcarryx_u64(0x0, x96, x87, &x98);
+  uint64_t x101; uint8_t x102 = addcarryx_u64(0x0, x65, x77, &x101);
+  uint64_t x104; uint8_t x105 = addcarryx_u64(x102, x68, x89, &x104);
+  uint64_t x107; uint8_t x108 = addcarryx_u64(x105, x71, x92, &x107);
+  uint64_t x110; uint8_t x111 = addcarryx_u64(x108, x74, x95, &x110);
+  uint64_t x113; uint8_t x114 = addcarryx_u64(x111, x75, x98, &x113);
+  uint64_t x117;  uint64_t x116 = mulx_u64(x101, 0xffffffffffffffffL, &x117);
+  uint64_t x120;  uint64_t x119 = mulx_u64(x101, 0xffffffff, &x120);
+  uint64_t x123;  uint64_t x122 = mulx_u64(x101, 0xffffffff00000001L, &x123);
+  uint64_t x125; uint8_t x126 = addcarryx_u64(0x0, x117, x119, &x125);
+  uint64_t x128; uint8_t x129 = addcarryx_u64(x126, x120, 0x0, &x128);
+  uint64_t x131; uint8_t x132 = addcarryx_u64(x129, 0x0, x122, &x131);
+  uint64_t x134; addcarryx_u64(0x0, x132, x123, &x134);
+  uint64_t _3; uint8_t x138 = addcarryx_u64(0x0, x101, x116, &_3);
+  uint64_t x140; uint8_t x141 = addcarryx_u64(x138, x104, x125, &x140);
+  uint64_t x143; uint8_t x144 = addcarryx_u64(x141, x107, x128, &x143);
+  uint64_t x146; uint8_t x147 = addcarryx_u64(x144, x110, x131, &x146);
+  uint64_t x149; uint8_t x150 = addcarryx_u64(x147, x113, x134, &x149);
+  uint8_t x151 = (x150 + x114);
+  uint64_t x154;  uint64_t x153 = mulx_u64(x9, x11, &x154);
+  uint64_t x157;  uint64_t x156 = mulx_u64(x9, x13, &x157);
+  uint64_t x160;  uint64_t x159 = mulx_u64(x9, x15, &x160);
+  uint64_t x163;  uint64_t x162 = mulx_u64(x9, x14, &x163);
+  uint64_t x165; uint8_t x166 = addcarryx_u64(0x0, x154, x156, &x165);
+  uint64_t x168; uint8_t x169 = addcarryx_u64(x166, x157, x159, &x168);
+  uint64_t x171; uint8_t x172 = addcarryx_u64(x169, x160, x162, &x171);
+  uint64_t x174; addcarryx_u64(0x0, x172, x163, &x174);
+  uint64_t x177; uint8_t x178 = addcarryx_u64(0x0, x140, x153, &x177);
+  uint64_t x180; uint8_t x181 = addcarryx_u64(x178, x143, x165, &x180);
+  uint64_t x183; uint8_t x184 = addcarryx_u64(x181, x146, x168, &x183);
+  uint64_t x186; uint8_t x187 = addcarryx_u64(x184, x149, x171, &x186);
+  uint64_t x189; uint8_t x190 = addcarryx_u64(x187, x151, x174, &x189);
+  uint64_t x193;  uint64_t x192 = mulx_u64(x177, 0xffffffffffffffffL, &x193);
+  uint64_t x196;  uint64_t x195 = mulx_u64(x177, 0xffffffff, &x196);
+  uint64_t x199;  uint64_t x198 = mulx_u64(x177, 0xffffffff00000001L, &x199);
+  uint64_t x201; uint8_t x202 = addcarryx_u64(0x0, x193, x195, &x201);
+  uint64_t x204; uint8_t x205 = addcarryx_u64(x202, x196, 0x0, &x204);
+  uint64_t x207; uint8_t x208 = addcarryx_u64(x205, 0x0, x198, &x207);
+  uint64_t x210; addcarryx_u64(0x0, x208, x199, &x210);
+  uint64_t _4; uint8_t x214 = addcarryx_u64(0x0, x177, x192, &_4);
+  uint64_t x216; uint8_t x217 = addcarryx_u64(x214, x180, x201, &x216);
+  uint64_t x219; uint8_t x220 = addcarryx_u64(x217, x183, x204, &x219);
+  uint64_t x222; uint8_t x223 = addcarryx_u64(x220, x186, x207, &x222);
+  uint64_t x225; uint8_t x226 = addcarryx_u64(x223, x189, x210, &x225);
+  uint8_t x227 = (x226 + x190);
+  uint64_t x230;  uint64_t x229 = mulx_u64(x8, x11, &x230);
+  uint64_t x233;  uint64_t x232 = mulx_u64(x8, x13, &x233);
+  uint64_t x236;  uint64_t x235 = mulx_u64(x8, x15, &x236);
+  uint64_t x239;  uint64_t x238 = mulx_u64(x8, x14, &x239);
+  uint64_t x241; uint8_t x242 = addcarryx_u64(0x0, x230, x232, &x241);
+  uint64_t x244; uint8_t x245 = addcarryx_u64(x242, x233, x235, &x244);
+  uint64_t x247; uint8_t x248 = addcarryx_u64(x245, x236, x238, &x247);
+  uint64_t x250; addcarryx_u64(0x0, x248, x239, &x250);
+  uint64_t x253; uint8_t x254 = addcarryx_u64(0x0, x216, x229, &x253);
+  uint64_t x256; uint8_t x257 = addcarryx_u64(x254, x219, x241, &x256);
+  uint64_t x259; uint8_t x260 = addcarryx_u64(x257, x222, x244, &x259);
+  uint64_t x262; uint8_t x263 = addcarryx_u64(x260, x225, x247, &x262);
+  uint64_t x265; uint8_t x266 = addcarryx_u64(x263, x227, x250, &x265);
+  uint64_t x269;  uint64_t x268 = mulx_u64(x253, 0xffffffffffffffffL, &x269);
+  uint64_t x272;  uint64_t x271 = mulx_u64(x253, 0xffffffff, &x272);
+  uint64_t x275;  uint64_t x274 = mulx_u64(x253, 0xffffffff00000001L, &x275);
+  uint64_t x277; uint8_t x278 = addcarryx_u64(0x0, x269, x271, &x277);
+  uint64_t x280; uint8_t x281 = addcarryx_u64(x278, x272, 0x0, &x280);
+  uint64_t x283; uint8_t x284 = addcarryx_u64(x281, 0x0, x274, &x283);
+  uint64_t x286; addcarryx_u64(0x0, x284, x275, &x286);
+  uint64_t _5; uint8_t x290 = addcarryx_u64(0x0, x253, x268, &_5);
+  uint64_t x292; uint8_t x293 = addcarryx_u64(x290, x256, x277, &x292);
+  uint64_t x295; uint8_t x296 = addcarryx_u64(x293, x259, x280, &x295);
+  uint64_t x298; uint8_t x299 = addcarryx_u64(x296, x262, x283, &x298);
+  uint64_t x301; uint8_t x302 = addcarryx_u64(x299, x265, x286, &x301);
+  uint8_t x303 = (x302 + x266);
+  uint64_t x305; uint8_t x306 = subborrow_u64(0x0, x292, 0xffffffffffffffffL, &x305);
+  uint64_t x308; uint8_t x309 = subborrow_u64(x306, x295, 0xffffffff, &x308);
+  uint64_t x311; uint8_t x312 = subborrow_u64(x309, x298, 0x0, &x311);
+  uint64_t x314; uint8_t x315 = subborrow_u64(x312, x301, 0xffffffff00000001L, &x314);
+  uint64_t _6; uint8_t x318 = subborrow_u64(x315, x303, 0x0, &_6);
+  uint64_t x319 = cmovznz_u64(x318, x314, x301);
+  uint64_t x320 = cmovznz_u64(x318, x311, x298);
+  uint64_t x321 = cmovznz_u64(x318, x308, x295);
+  uint64_t x322 = cmovznz_u64(x318, x305, x292);
+  out[0] = x322;
+  out[1] = x321;
+  out[2] = x320;
+  out[3] = x319;
+}
+
+static void fe_sub(uint64_t out[4], const uint64_t in1[4], const uint64_t in2[4]) {
+  const uint64_t x8 = in1[3];
+  const uint64_t x9 = in1[2];
+  const uint64_t x7 = in1[1];
+  const uint64_t x5 = in1[0];
+  const uint64_t x14 = in2[3];
+  const uint64_t x15 = in2[2];
+  const uint64_t x13 = in2[1];
+  const uint64_t x11 = in2[0];
+  uint64_t x17; uint8_t x18 = subborrow_u64(0x0, x5, x11, &x17);
+  uint64_t x20; uint8_t x21 = subborrow_u64(x18, x7, x13, &x20);
+  uint64_t x23; uint8_t x24 = subborrow_u64(x21, x9, x15, &x23);
+  uint64_t x26; uint8_t x27 = subborrow_u64(x24, x8, x14, &x26);
+  uint64_t x28 = (uint64_t)cmovznz_u64(x27, 0x0, 0xffffffffffffffffL);
+  uint64_t x29 = (x28 & 0xffffffffffffffffL);
+  uint64_t x31; uint8_t x32 = addcarryx_u64(0x0, x17, x29, &x31);
+  uint64_t x33 = (x28 & 0xffffffff);
+  uint64_t x35; uint8_t x36 = addcarryx_u64(x32, x20, x33, &x35);
+  uint64_t x38; uint8_t x39 = addcarryx_u64(x36, x23, 0x0, &x38);
+  uint64_t x40 = (x28 & 0xffffffff00000001L);
+  uint64_t x42; addcarryx_u64(x39, x26, x40, &x42);
+  out[0] = x31;
+  out[1] = x35;
+  out[2] = x38;
+  out[3] = x42;
+}
+
+#else // 64BIT, else 32BIT
+
+static void fe_add(uint32_t out[8], const uint32_t in1[8], const uint32_t in2[8]) {
+  const uint32_t x16 = in1[7];
+  const uint32_t x17 = in1[6];
+  const uint32_t x15 = in1[5];
+  const uint32_t x13 = in1[4];
+  const uint32_t x11 = in1[3];
+  const uint32_t x9 = in1[2];
+  const uint32_t x7 = in1[1];
+  const uint32_t x5 = in1[0];
+  const uint32_t x30 = in2[7];
+  const uint32_t x31 = in2[6];
+  const uint32_t x29 = in2[5];
+  const uint32_t x27 = in2[4];
+  const uint32_t x25 = in2[3];
+  const uint32_t x23 = in2[2];
+  const uint32_t x21 = in2[1];
+  const uint32_t x19 = in2[0];
+  uint32_t x33; uint8_t x34 = addcarryx_u32(0x0, x5, x19, &x33);
+  uint32_t x36; uint8_t x37 = addcarryx_u32(x34, x7, x21, &x36);
+  uint32_t x39; uint8_t x40 = addcarryx_u32(x37, x9, x23, &x39);
+  uint32_t x42; uint8_t x43 = addcarryx_u32(x40, x11, x25, &x42);
+  uint32_t x45; uint8_t x46 = addcarryx_u32(x43, x13, x27, &x45);
+  uint32_t x48; uint8_t x49 = addcarryx_u32(x46, x15, x29, &x48);
+  uint32_t x51; uint8_t x52 = addcarryx_u32(x49, x17, x31, &x51);
+  uint32_t x54; uint8_t x55 = addcarryx_u32(x52, x16, x30, &x54);
+  uint32_t x57; uint8_t x58 = subborrow_u32(0x0, x33, 0xffffffff, &x57);
+  uint32_t x60; uint8_t x61 = subborrow_u32(x58, x36, 0xffffffff, &x60);
+  uint32_t x63; uint8_t x64 = subborrow_u32(x61, x39, 0xffffffff, &x63);
+  uint32_t x66; uint8_t x67 = subborrow_u32(x64, x42, 0x0, &x66);
+  uint32_t x69; uint8_t x70 = subborrow_u32(x67, x45, 0x0, &x69);
+  uint32_t x72; uint8_t x73 = subborrow_u32(x70, x48, 0x0, &x72);
+  uint32_t x75; uint8_t x76 = subborrow_u32(x73, x51, 0x1, &x75);
+  uint32_t x78; uint8_t x79 = subborrow_u32(x76, x54, 0xffffffff, &x78);
+  uint32_t _; uint8_t x82 = subborrow_u32(x79, x55, 0x0, &_);
+  uint32_t x83 = cmovznz_u32(x82, x78, x54);
+  uint32_t x84 = cmovznz_u32(x82, x75, x51);
+  uint32_t x85 = cmovznz_u32(x82, x72, x48);
+  uint32_t x86 = cmovznz_u32(x82, x69, x45);
+  uint32_t x87 = cmovznz_u32(x82, x66, x42);
+  uint32_t x88 = cmovznz_u32(x82, x63, x39);
+  uint32_t x89 = cmovznz_u32(x82, x60, x36);
+  uint32_t x90 = cmovznz_u32(x82, x57, x33);
+  out[0] = x90;
+  out[1] = x89;
+  out[2] = x88;
+  out[3] = x87;
+  out[4] = x86;
+  out[5] = x85;
+  out[6] = x84;
+  out[7] = x83;
+}
+
+static void fe_mul(uint32_t out[8], const uint32_t in1[8], const uint32_t in2[8]) {
+  const uint32_t x16 = in1[7];
+  const uint32_t x17 = in1[6];
+  const uint32_t x15 = in1[5];
+  const uint32_t x13 = in1[4];
+  const uint32_t x11 = in1[3];
+  const uint32_t x9 = in1[2];
+  const uint32_t x7 = in1[1];
+  const uint32_t x5 = in1[0];
+  const uint32_t x30 = in2[7];
+  const uint32_t x31 = in2[6];
+  const uint32_t x29 = in2[5];
+  const uint32_t x27 = in2[4];
+  const uint32_t x25 = in2[3];
+  const uint32_t x23 = in2[2];
+  const uint32_t x21 = in2[1];
+  const uint32_t x19 = in2[0];
+  uint32_t x34;  uint32_t x33 = mulx_u32(x5, x19, &x34);
+  uint32_t x37;  uint32_t x36 = mulx_u32(x5, x21, &x37);
+  uint32_t x40;  uint32_t x39 = mulx_u32(x5, x23, &x40);
+  uint32_t x43;  uint32_t x42 = mulx_u32(x5, x25, &x43);
+  uint32_t x46;  uint32_t x45 = mulx_u32(x5, x27, &x46);
+  uint32_t x49;  uint32_t x48 = mulx_u32(x5, x29, &x49);
+  uint32_t x52;  uint32_t x51 = mulx_u32(x5, x31, &x52);
+  uint32_t x55;  uint32_t x54 = mulx_u32(x5, x30, &x55);
+  uint32_t x57; uint8_t x58 = addcarryx_u32(0x0, x34, x36, &x57);
+  uint32_t x60; uint8_t x61 = addcarryx_u32(x58, x37, x39, &x60);
+  uint32_t x63; uint8_t x64 = addcarryx_u32(x61, x40, x42, &x63);
+  uint32_t x66; uint8_t x67 = addcarryx_u32(x64, x43, x45, &x66);
+  uint32_t x69; uint8_t x70 = addcarryx_u32(x67, x46, x48, &x69);
+  uint32_t x72; uint8_t x73 = addcarryx_u32(x70, x49, x51, &x72);
+  uint32_t x75; uint8_t x76 = addcarryx_u32(x73, x52, x54, &x75);
+  uint32_t x78; addcarryx_u32(0x0, x76, x55, &x78);
+  uint32_t x82;  uint32_t x81 = mulx_u32(x33, 0xffffffff, &x82);
+  uint32_t x85;  uint32_t x84 = mulx_u32(x33, 0xffffffff, &x85);
+  uint32_t x88;  uint32_t x87 = mulx_u32(x33, 0xffffffff, &x88);
+  uint32_t x91;  uint32_t x90 = mulx_u32(x33, 0xffffffff, &x91);
+  uint32_t x93; uint8_t x94 = addcarryx_u32(0x0, x82, x84, &x93);
+  uint32_t x96; uint8_t x97 = addcarryx_u32(x94, x85, x87, &x96);
+  uint32_t x99; uint8_t x100 = addcarryx_u32(x97, x88, 0x0, &x99);
+  uint8_t x101 = (0x0 + 0x0);
+  uint32_t _1; uint8_t x104 = addcarryx_u32(0x0, x33, x81, &_1);
+  uint32_t x106; uint8_t x107 = addcarryx_u32(x104, x57, x93, &x106);
+  uint32_t x109; uint8_t x110 = addcarryx_u32(x107, x60, x96, &x109);
+  uint32_t x112; uint8_t x113 = addcarryx_u32(x110, x63, x99, &x112);
+  uint32_t x115; uint8_t x116 = addcarryx_u32(x113, x66, x100, &x115);
+  uint32_t x118; uint8_t x119 = addcarryx_u32(x116, x69, x101, &x118);
+  uint32_t x121; uint8_t x122 = addcarryx_u32(x119, x72, x33, &x121);
+  uint32_t x124; uint8_t x125 = addcarryx_u32(x122, x75, x90, &x124);
+  uint32_t x127; uint8_t x128 = addcarryx_u32(x125, x78, x91, &x127);
+  uint8_t x129 = (x128 + 0x0);
+  uint32_t x132;  uint32_t x131 = mulx_u32(x7, x19, &x132);
+  uint32_t x135;  uint32_t x134 = mulx_u32(x7, x21, &x135);
+  uint32_t x138;  uint32_t x137 = mulx_u32(x7, x23, &x138);
+  uint32_t x141;  uint32_t x140 = mulx_u32(x7, x25, &x141);
+  uint32_t x144;  uint32_t x143 = mulx_u32(x7, x27, &x144);
+  uint32_t x147;  uint32_t x146 = mulx_u32(x7, x29, &x147);
+  uint32_t x150;  uint32_t x149 = mulx_u32(x7, x31, &x150);
+  uint32_t x153;  uint32_t x152 = mulx_u32(x7, x30, &x153);
+  uint32_t x155; uint8_t x156 = addcarryx_u32(0x0, x132, x134, &x155);
+  uint32_t x158; uint8_t x159 = addcarryx_u32(x156, x135, x137, &x158);
+  uint32_t x161; uint8_t x162 = addcarryx_u32(x159, x138, x140, &x161);
+  uint32_t x164; uint8_t x165 = addcarryx_u32(x162, x141, x143, &x164);
+  uint32_t x167; uint8_t x168 = addcarryx_u32(x165, x144, x146, &x167);
+  uint32_t x170; uint8_t x171 = addcarryx_u32(x168, x147, x149, &x170);
+  uint32_t x173; uint8_t x174 = addcarryx_u32(x171, x150, x152, &x173);
+  uint32_t x176; addcarryx_u32(0x0, x174, x153, &x176);
+  uint32_t x179; uint8_t x180 = addcarryx_u32(0x0, x106, x131, &x179);
+  uint32_t x182; uint8_t x183 = addcarryx_u32(x180, x109, x155, &x182);
+  uint32_t x185; uint8_t x186 = addcarryx_u32(x183, x112, x158, &x185);
+  uint32_t x188; uint8_t x189 = addcarryx_u32(x186, x115, x161, &x188);
+  uint32_t x191; uint8_t x192 = addcarryx_u32(x189, x118, x164, &x191);
+  uint32_t x194; uint8_t x195 = addcarryx_u32(x192, x121, x167, &x194);
+  uint32_t x197; uint8_t x198 = addcarryx_u32(x195, x124, x170, &x197);
+  uint32_t x200; uint8_t x201 = addcarryx_u32(x198, x127, x173, &x200);
+  uint32_t x203; uint8_t x204 = addcarryx_u32(x201, x129, x176, &x203);
+  uint32_t x207;  uint32_t x206 = mulx_u32(x179, 0xffffffff, &x207);
+  uint32_t x210;  uint32_t x209 = mulx_u32(x179, 0xffffffff, &x210);
+  uint32_t x213;  uint32_t x212 = mulx_u32(x179, 0xffffffff, &x213);
+  uint32_t x216;  uint32_t x215 = mulx_u32(x179, 0xffffffff, &x216);
+  uint32_t x218; uint8_t x219 = addcarryx_u32(0x0, x207, x209, &x218);
+  uint32_t x221; uint8_t x222 = addcarryx_u32(x219, x210, x212, &x221);
+  uint32_t x224; uint8_t x225 = addcarryx_u32(x222, x213, 0x0, &x224);
+  uint8_t x226 = (0x0 + 0x0);
+  uint32_t _2; uint8_t x229 = addcarryx_u32(0x0, x179, x206, &_2);
+  uint32_t x231; uint8_t x232 = addcarryx_u32(x229, x182, x218, &x231);
+  uint32_t x234; uint8_t x235 = addcarryx_u32(x232, x185, x221, &x234);
+  uint32_t x237; uint8_t x238 = addcarryx_u32(x235, x188, x224, &x237);
+  uint32_t x240; uint8_t x241 = addcarryx_u32(x238, x191, x225, &x240);
+  uint32_t x243; uint8_t x244 = addcarryx_u32(x241, x194, x226, &x243);
+  uint32_t x246; uint8_t x247 = addcarryx_u32(x244, x197, x179, &x246);
+  uint32_t x249; uint8_t x250 = addcarryx_u32(x247, x200, x215, &x249);
+  uint32_t x252; uint8_t x253 = addcarryx_u32(x250, x203, x216, &x252);
+  uint8_t x254 = (x253 + x204);
+  uint32_t x257;  uint32_t x256 = mulx_u32(x9, x19, &x257);
+  uint32_t x260;  uint32_t x259 = mulx_u32(x9, x21, &x260);
+  uint32_t x263;  uint32_t x262 = mulx_u32(x9, x23, &x263);
+  uint32_t x266;  uint32_t x265 = mulx_u32(x9, x25, &x266);
+  uint32_t x269;  uint32_t x268 = mulx_u32(x9, x27, &x269);
+  uint32_t x272;  uint32_t x271 = mulx_u32(x9, x29, &x272);
+  uint32_t x275;  uint32_t x274 = mulx_u32(x9, x31, &x275);
+  uint32_t x278;  uint32_t x277 = mulx_u32(x9, x30, &x278);
+  uint32_t x280; uint8_t x281 = addcarryx_u32(0x0, x257, x259, &x280);
+  uint32_t x283; uint8_t x284 = addcarryx_u32(x281, x260, x262, &x283);
+  uint32_t x286; uint8_t x287 = addcarryx_u32(x284, x263, x265, &x286);
+  uint32_t x289; uint8_t x290 = addcarryx_u32(x287, x266, x268, &x289);
+  uint32_t x292; uint8_t x293 = addcarryx_u32(x290, x269, x271, &x292);
+  uint32_t x295; uint8_t x296 = addcarryx_u32(x293, x272, x274, &x295);
+  uint32_t x298; uint8_t x299 = addcarryx_u32(x296, x275, x277, &x298);
+  uint32_t x301; addcarryx_u32(0x0, x299, x278, &x301);
+  uint32_t x304; uint8_t x305 = addcarryx_u32(0x0, x231, x256, &x304);
+  uint32_t x307; uint8_t x308 = addcarryx_u32(x305, x234, x280, &x307);
+  uint32_t x310; uint8_t x311 = addcarryx_u32(x308, x237, x283, &x310);
+  uint32_t x313; uint8_t x314 = addcarryx_u32(x311, x240, x286, &x313);
+  uint32_t x316; uint8_t x317 = addcarryx_u32(x314, x243, x289, &x316);
+  uint32_t x319; uint8_t x320 = addcarryx_u32(x317, x246, x292, &x319);
+  uint32_t x322; uint8_t x323 = addcarryx_u32(x320, x249, x295, &x322);
+  uint32_t x325; uint8_t x326 = addcarryx_u32(x323, x252, x298, &x325);
+  uint32_t x328; uint8_t x329 = addcarryx_u32(x326, x254, x301, &x328);
+  uint32_t x332;  uint32_t x331 = mulx_u32(x304, 0xffffffff, &x332);
+  uint32_t x335;  uint32_t x334 = mulx_u32(x304, 0xffffffff, &x335);
+  uint32_t x338;  uint32_t x337 = mulx_u32(x304, 0xffffffff, &x338);
+  uint32_t x341;  uint32_t x340 = mulx_u32(x304, 0xffffffff, &x341);
+  uint32_t x343; uint8_t x344 = addcarryx_u32(0x0, x332, x334, &x343);
+  uint32_t x346; uint8_t x347 = addcarryx_u32(x344, x335, x337, &x346);
+  uint32_t x349; uint8_t x350 = addcarryx_u32(x347, x338, 0x0, &x349);
+  uint8_t x351 = (0x0 + 0x0);
+  uint32_t _3; uint8_t x354 = addcarryx_u32(0x0, x304, x331, &_3);
+  uint32_t x356; uint8_t x357 = addcarryx_u32(x354, x307, x343, &x356);
+  uint32_t x359; uint8_t x360 = addcarryx_u32(x357, x310, x346, &x359);
+  uint32_t x362; uint8_t x363 = addcarryx_u32(x360, x313, x349, &x362);
+  uint32_t x365; uint8_t x366 = addcarryx_u32(x363, x316, x350, &x365);
+  uint32_t x368; uint8_t x369 = addcarryx_u32(x366, x319, x351, &x368);
+  uint32_t x371; uint8_t x372 = addcarryx_u32(x369, x322, x304, &x371);
+  uint32_t x374; uint8_t x375 = addcarryx_u32(x372, x325, x340, &x374);
+  uint32_t x377; uint8_t x378 = addcarryx_u32(x375, x328, x341, &x377);
+  uint8_t x379 = (x378 + x329);
+  uint32_t x382;  uint32_t x381 = mulx_u32(x11, x19, &x382);
+  uint32_t x385;  uint32_t x384 = mulx_u32(x11, x21, &x385);
+  uint32_t x388;  uint32_t x387 = mulx_u32(x11, x23, &x388);
+  uint32_t x391;  uint32_t x390 = mulx_u32(x11, x25, &x391);
+  uint32_t x394;  uint32_t x393 = mulx_u32(x11, x27, &x394);
+  uint32_t x397;  uint32_t x396 = mulx_u32(x11, x29, &x397);
+  uint32_t x400;  uint32_t x399 = mulx_u32(x11, x31, &x400);
+  uint32_t x403;  uint32_t x402 = mulx_u32(x11, x30, &x403);
+  uint32_t x405; uint8_t x406 = addcarryx_u32(0x0, x382, x384, &x405);
+  uint32_t x408; uint8_t x409 = addcarryx_u32(x406, x385, x387, &x408);
+  uint32_t x411; uint8_t x412 = addcarryx_u32(x409, x388, x390, &x411);
+  uint32_t x414; uint8_t x415 = addcarryx_u32(x412, x391, x393, &x414);
+  uint32_t x417; uint8_t x418 = addcarryx_u32(x415, x394, x396, &x417);
+  uint32_t x420; uint8_t x421 = addcarryx_u32(x418, x397, x399, &x420);
+  uint32_t x423; uint8_t x424 = addcarryx_u32(x421, x400, x402, &x423);
+  uint32_t x426; addcarryx_u32(0x0, x424, x403, &x426);
+  uint32_t x429; uint8_t x430 = addcarryx_u32(0x0, x356, x381, &x429);
+  uint32_t x432; uint8_t x433 = addcarryx_u32(x430, x359, x405, &x432);
+  uint32_t x435; uint8_t x436 = addcarryx_u32(x433, x362, x408, &x435);
+  uint32_t x438; uint8_t x439 = addcarryx_u32(x436, x365, x411, &x438);
+  uint32_t x441; uint8_t x442 = addcarryx_u32(x439, x368, x414, &x441);
+  uint32_t x444; uint8_t x445 = addcarryx_u32(x442, x371, x417, &x444);
+  uint32_t x447; uint8_t x448 = addcarryx_u32(x445, x374, x420, &x447);
+  uint32_t x450; uint8_t x451 = addcarryx_u32(x448, x377, x423, &x450);
+  uint32_t x453; uint8_t x454 = addcarryx_u32(x451, x379, x426, &x453);
+  uint32_t x457;  uint32_t x456 = mulx_u32(x429, 0xffffffff, &x457);
+  uint32_t x460;  uint32_t x459 = mulx_u32(x429, 0xffffffff, &x460);
+  uint32_t x463;  uint32_t x462 = mulx_u32(x429, 0xffffffff, &x463);
+  uint32_t x466;  uint32_t x465 = mulx_u32(x429, 0xffffffff, &x466);
+  uint32_t x468; uint8_t x469 = addcarryx_u32(0x0, x457, x459, &x468);
+  uint32_t x471; uint8_t x472 = addcarryx_u32(x469, x460, x462, &x471);
+  uint32_t x474; uint8_t x475 = addcarryx_u32(x472, x463, 0x0, &x474);
+  uint8_t x476 = (0x0 + 0x0);
+  uint32_t _4; uint8_t x479 = addcarryx_u32(0x0, x429, x456, &_4);
+  uint32_t x481; uint8_t x482 = addcarryx_u32(x479, x432, x468, &x481);
+  uint32_t x484; uint8_t x485 = addcarryx_u32(x482, x435, x471, &x484);
+  uint32_t x487; uint8_t x488 = addcarryx_u32(x485, x438, x474, &x487);
+  uint32_t x490; uint8_t x491 = addcarryx_u32(x488, x441, x475, &x490);
+  uint32_t x493; uint8_t x494 = addcarryx_u32(x491, x444, x476, &x493);
+  uint32_t x496; uint8_t x497 = addcarryx_u32(x494, x447, x429, &x496);
+  uint32_t x499; uint8_t x500 = addcarryx_u32(x497, x450, x465, &x499);
+  uint32_t x502; uint8_t x503 = addcarryx_u32(x500, x453, x466, &x502);
+  uint8_t x504 = (x503 + x454);
+  uint32_t x507;  uint32_t x506 = mulx_u32(x13, x19, &x507);
+  uint32_t x510;  uint32_t x509 = mulx_u32(x13, x21, &x510);
+  uint32_t x513;  uint32_t x512 = mulx_u32(x13, x23, &x513);
+  uint32_t x516;  uint32_t x515 = mulx_u32(x13, x25, &x516);
+  uint32_t x519;  uint32_t x518 = mulx_u32(x13, x27, &x519);
+  uint32_t x522;  uint32_t x521 = mulx_u32(x13, x29, &x522);
+  uint32_t x525;  uint32_t x524 = mulx_u32(x13, x31, &x525);
+  uint32_t x528;  uint32_t x527 = mulx_u32(x13, x30, &x528);
+  uint32_t x530; uint8_t x531 = addcarryx_u32(0x0, x507, x509, &x530);
+  uint32_t x533; uint8_t x534 = addcarryx_u32(x531, x510, x512, &x533);
+  uint32_t x536; uint8_t x537 = addcarryx_u32(x534, x513, x515, &x536);
+  uint32_t x539; uint8_t x540 = addcarryx_u32(x537, x516, x518, &x539);
+  uint32_t x542; uint8_t x543 = addcarryx_u32(x540, x519, x521, &x542);
+  uint32_t x545; uint8_t x546 = addcarryx_u32(x543, x522, x524, &x545);
+  uint32_t x548; uint8_t x549 = addcarryx_u32(x546, x525, x527, &x548);
+  uint32_t x551; addcarryx_u32(0x0, x549, x528, &x551);
+  uint32_t x554; uint8_t x555 = addcarryx_u32(0x0, x481, x506, &x554);
+  uint32_t x557; uint8_t x558 = addcarryx_u32(x555, x484, x530, &x557);
+  uint32_t x560; uint8_t x561 = addcarryx_u32(x558, x487, x533, &x560);
+  uint32_t x563; uint8_t x564 = addcarryx_u32(x561, x490, x536, &x563);
+  uint32_t x566; uint8_t x567 = addcarryx_u32(x564, x493, x539, &x566);
+  uint32_t x569; uint8_t x570 = addcarryx_u32(x567, x496, x542, &x569);
+  uint32_t x572; uint8_t x573 = addcarryx_u32(x570, x499, x545, &x572);
+  uint32_t x575; uint8_t x576 = addcarryx_u32(x573, x502, x548, &x575);
+  uint32_t x578; uint8_t x579 = addcarryx_u32(x576, x504, x551, &x578);
+  uint32_t x582;  uint32_t x581 = mulx_u32(x554, 0xffffffff, &x582);
+  uint32_t x585;  uint32_t x584 = mulx_u32(x554, 0xffffffff, &x585);
+  uint32_t x588;  uint32_t x587 = mulx_u32(x554, 0xffffffff, &x588);
+  uint32_t x591;  uint32_t x590 = mulx_u32(x554, 0xffffffff, &x591);
+  uint32_t x593; uint8_t x594 = addcarryx_u32(0x0, x582, x584, &x593);
+  uint32_t x596; uint8_t x597 = addcarryx_u32(x594, x585, x587, &x596);
+  uint32_t x599; uint8_t x600 = addcarryx_u32(x597, x588, 0x0, &x599);
+  uint8_t x601 = (0x0 + 0x0);
+  uint32_t _5; uint8_t x604 = addcarryx_u32(0x0, x554, x581, &_5);
+  uint32_t x606; uint8_t x607 = addcarryx_u32(x604, x557, x593, &x606);
+  uint32_t x609; uint8_t x610 = addcarryx_u32(x607, x560, x596, &x609);
+  uint32_t x612; uint8_t x613 = addcarryx_u32(x610, x563, x599, &x612);
+  uint32_t x615; uint8_t x616 = addcarryx_u32(x613, x566, x600, &x615);
+  uint32_t x618; uint8_t x619 = addcarryx_u32(x616, x569, x601, &x618);
+  uint32_t x621; uint8_t x622 = addcarryx_u32(x619, x572, x554, &x621);
+  uint32_t x624; uint8_t x625 = addcarryx_u32(x622, x575, x590, &x624);
+  uint32_t x627; uint8_t x628 = addcarryx_u32(x625, x578, x591, &x627);
+  uint8_t x629 = (x628 + x579);
+  uint32_t x632;  uint32_t x631 = mulx_u32(x15, x19, &x632);
+  uint32_t x635;  uint32_t x634 = mulx_u32(x15, x21, &x635);
+  uint32_t x638;  uint32_t x637 = mulx_u32(x15, x23, &x638);
+  uint32_t x641;  uint32_t x640 = mulx_u32(x15, x25, &x641);
+  uint32_t x644;  uint32_t x643 = mulx_u32(x15, x27, &x644);
+  uint32_t x647;  uint32_t x646 = mulx_u32(x15, x29, &x647);
+  uint32_t x650;  uint32_t x649 = mulx_u32(x15, x31, &x650);
+  uint32_t x653;  uint32_t x652 = mulx_u32(x15, x30, &x653);
+  uint32_t x655; uint8_t x656 = addcarryx_u32(0x0, x632, x634, &x655);
+  uint32_t x658; uint8_t x659 = addcarryx_u32(x656, x635, x637, &x658);
+  uint32_t x661; uint8_t x662 = addcarryx_u32(x659, x638, x640, &x661);
+  uint32_t x664; uint8_t x665 = addcarryx_u32(x662, x641, x643, &x664);
+  uint32_t x667; uint8_t x668 = addcarryx_u32(x665, x644, x646, &x667);
+  uint32_t x670; uint8_t x671 = addcarryx_u32(x668, x647, x649, &x670);
+  uint32_t x673; uint8_t x674 = addcarryx_u32(x671, x650, x652, &x673);
+  uint32_t x676; addcarryx_u32(0x0, x674, x653, &x676);
+  uint32_t x679; uint8_t x680 = addcarryx_u32(0x0, x606, x631, &x679);
+  uint32_t x682; uint8_t x683 = addcarryx_u32(x680, x609, x655, &x682);
+  uint32_t x685; uint8_t x686 = addcarryx_u32(x683, x612, x658, &x685);
+  uint32_t x688; uint8_t x689 = addcarryx_u32(x686, x615, x661, &x688);
+  uint32_t x691; uint8_t x692 = addcarryx_u32(x689, x618, x664, &x691);
+  uint32_t x694; uint8_t x695 = addcarryx_u32(x692, x621, x667, &x694);
+  uint32_t x697; uint8_t x698 = addcarryx_u32(x695, x624, x670, &x697);
+  uint32_t x700; uint8_t x701 = addcarryx_u32(x698, x627, x673, &x700);
+  uint32_t x703; uint8_t x704 = addcarryx_u32(x701, x629, x676, &x703);
+  uint32_t x707;  uint32_t x706 = mulx_u32(x679, 0xffffffff, &x707);
+  uint32_t x710;  uint32_t x709 = mulx_u32(x679, 0xffffffff, &x710);
+  uint32_t x713;  uint32_t x712 = mulx_u32(x679, 0xffffffff, &x713);
+  uint32_t x716;  uint32_t x715 = mulx_u32(x679, 0xffffffff, &x716);
+  uint32_t x718; uint8_t x719 = addcarryx_u32(0x0, x707, x709, &x718);
+  uint32_t x721; uint8_t x722 = addcarryx_u32(x719, x710, x712, &x721);
+  uint32_t x724; uint8_t x725 = addcarryx_u32(x722, x713, 0x0, &x724);
+  uint8_t x726 = (0x0 + 0x0);
+  uint32_t _6; uint8_t x729 = addcarryx_u32(0x0, x679, x706, &_6);
+  uint32_t x731; uint8_t x732 = addcarryx_u32(x729, x682, x718, &x731);
+  uint32_t x734; uint8_t x735 = addcarryx_u32(x732, x685, x721, &x734);
+  uint32_t x737; uint8_t x738 = addcarryx_u32(x735, x688, x724, &x737);
+  uint32_t x740; uint8_t x741 = addcarryx_u32(x738, x691, x725, &x740);
+  uint32_t x743; uint8_t x744 = addcarryx_u32(x741, x694, x726, &x743);
+  uint32_t x746; uint8_t x747 = addcarryx_u32(x744, x697, x679, &x746);
+  uint32_t x749; uint8_t x750 = addcarryx_u32(x747, x700, x715, &x749);
+  uint32_t x752; uint8_t x753 = addcarryx_u32(x750, x703, x716, &x752);
+  uint8_t x754 = (x753 + x704);
+  uint32_t x757;  uint32_t x756 = mulx_u32(x17, x19, &x757);
+  uint32_t x760;  uint32_t x759 = mulx_u32(x17, x21, &x760);
+  uint32_t x763;  uint32_t x762 = mulx_u32(x17, x23, &x763);
+  uint32_t x766;  uint32_t x765 = mulx_u32(x17, x25, &x766);
+  uint32_t x769;  uint32_t x768 = mulx_u32(x17, x27, &x769);
+  uint32_t x772;  uint32_t x771 = mulx_u32(x17, x29, &x772);
+  uint32_t x775;  uint32_t x774 = mulx_u32(x17, x31, &x775);
+  uint32_t x778;  uint32_t x777 = mulx_u32(x17, x30, &x778);
+  uint32_t x780; uint8_t x781 = addcarryx_u32(0x0, x757, x759, &x780);
+  uint32_t x783; uint8_t x784 = addcarryx_u32(x781, x760, x762, &x783);
+  uint32_t x786; uint8_t x787 = addcarryx_u32(x784, x763, x765, &x786);
+  uint32_t x789; uint8_t x790 = addcarryx_u32(x787, x766, x768, &x789);
+  uint32_t x792; uint8_t x793 = addcarryx_u32(x790, x769, x771, &x792);
+  uint32_t x795; uint8_t x796 = addcarryx_u32(x793, x772, x774, &x795);
+  uint32_t x798; uint8_t x799 = addcarryx_u32(x796, x775, x777, &x798);
+  uint32_t x801; addcarryx_u32(0x0, x799, x778, &x801);
+  uint32_t x804; uint8_t x805 = addcarryx_u32(0x0, x731, x756, &x804);
+  uint32_t x807; uint8_t x808 = addcarryx_u32(x805, x734, x780, &x807);
+  uint32_t x810; uint8_t x811 = addcarryx_u32(x808, x737, x783, &x810);
+  uint32_t x813; uint8_t x814 = addcarryx_u32(x811, x740, x786, &x813);
+  uint32_t x816; uint8_t x817 = addcarryx_u32(x814, x743, x789, &x816);
+  uint32_t x819; uint8_t x820 = addcarryx_u32(x817, x746, x792, &x819);
+  uint32_t x822; uint8_t x823 = addcarryx_u32(x820, x749, x795, &x822);
+  uint32_t x825; uint8_t x826 = addcarryx_u32(x823, x752, x798, &x825);
+  uint32_t x828; uint8_t x829 = addcarryx_u32(x826, x754, x801, &x828);
+  uint32_t x832;  uint32_t x831 = mulx_u32(x804, 0xffffffff, &x832);
+  uint32_t x835;  uint32_t x834 = mulx_u32(x804, 0xffffffff, &x835);
+  uint32_t x838;  uint32_t x837 = mulx_u32(x804, 0xffffffff, &x838);
+  uint32_t x841;  uint32_t x840 = mulx_u32(x804, 0xffffffff, &x841);
+  uint32_t x843; uint8_t x844 = addcarryx_u32(0x0, x832, x834, &x843);
+  uint32_t x846; uint8_t x847 = addcarryx_u32(x844, x835, x837, &x846);
+  uint32_t x849; uint8_t x850 = addcarryx_u32(x847, x838, 0x0, &x849);
+  uint8_t x851 = (0x0 + 0x0);
+  uint32_t _7; uint8_t x854 = addcarryx_u32(0x0, x804, x831, &_7);
+  uint32_t x856; uint8_t x857 = addcarryx_u32(x854, x807, x843, &x856);
+  uint32_t x859; uint8_t x860 = addcarryx_u32(x857, x810, x846, &x859);
+  uint32_t x862; uint8_t x863 = addcarryx_u32(x860, x813, x849, &x862);
+  uint32_t x865; uint8_t x866 = addcarryx_u32(x863, x816, x850, &x865);
+  uint32_t x868; uint8_t x869 = addcarryx_u32(x866, x819, x851, &x868);
+  uint32_t x871; uint8_t x872 = addcarryx_u32(x869, x822, x804, &x871);
+  uint32_t x874; uint8_t x875 = addcarryx_u32(x872, x825, x840, &x874);
+  uint32_t x877; uint8_t x878 = addcarryx_u32(x875, x828, x841, &x877);
+  uint8_t x879 = (x878 + x829);
+  uint32_t x882;  uint32_t x881 = mulx_u32(x16, x19, &x882);
+  uint32_t x885;  uint32_t x884 = mulx_u32(x16, x21, &x885);
+  uint32_t x888;  uint32_t x887 = mulx_u32(x16, x23, &x888);
+  uint32_t x891;  uint32_t x890 = mulx_u32(x16, x25, &x891);
+  uint32_t x894;  uint32_t x893 = mulx_u32(x16, x27, &x894);
+  uint32_t x897;  uint32_t x896 = mulx_u32(x16, x29, &x897);
+  uint32_t x900;  uint32_t x899 = mulx_u32(x16, x31, &x900);
+  uint32_t x903;  uint32_t x902 = mulx_u32(x16, x30, &x903);
+  uint32_t x905; uint8_t x906 = addcarryx_u32(0x0, x882, x884, &x905);
+  uint32_t x908; uint8_t x909 = addcarryx_u32(x906, x885, x887, &x908);
+  uint32_t x911; uint8_t x912 = addcarryx_u32(x909, x888, x890, &x911);
+  uint32_t x914; uint8_t x915 = addcarryx_u32(x912, x891, x893, &x914);
+  uint32_t x917; uint8_t x918 = addcarryx_u32(x915, x894, x896, &x917);
+  uint32_t x920; uint8_t x921 = addcarryx_u32(x918, x897, x899, &x920);
+  uint32_t x923; uint8_t x924 = addcarryx_u32(x921, x900, x902, &x923);
+  uint32_t x926; addcarryx_u32(0x0, x924, x903, &x926);
+  uint32_t x929; uint8_t x930 = addcarryx_u32(0x0, x856, x881, &x929);
+  uint32_t x932; uint8_t x933 = addcarryx_u32(x930, x859, x905, &x932);
+  uint32_t x935; uint8_t x936 = addcarryx_u32(x933, x862, x908, &x935);
+  uint32_t x938; uint8_t x939 = addcarryx_u32(x936, x865, x911, &x938);
+  uint32_t x941; uint8_t x942 = addcarryx_u32(x939, x868, x914, &x941);
+  uint32_t x944; uint8_t x945 = addcarryx_u32(x942, x871, x917, &x944);
+  uint32_t x947; uint8_t x948 = addcarryx_u32(x945, x874, x920, &x947);
+  uint32_t x950; uint8_t x951 = addcarryx_u32(x948, x877, x923, &x950);
+  uint32_t x953; uint8_t x954 = addcarryx_u32(x951, x879, x926, &x953);
+  uint32_t x957;  uint32_t x956 = mulx_u32(x929, 0xffffffff, &x957);
+  uint32_t x960;  uint32_t x959 = mulx_u32(x929, 0xffffffff, &x960);
+  uint32_t x963;  uint32_t x962 = mulx_u32(x929, 0xffffffff, &x963);
+  uint32_t x966;  uint32_t x965 = mulx_u32(x929, 0xffffffff, &x966);
+  uint32_t x968; uint8_t x969 = addcarryx_u32(0x0, x957, x959, &x968);
+  uint32_t x971; uint8_t x972 = addcarryx_u32(x969, x960, x962, &x971);
+  uint32_t x974; uint8_t x975 = addcarryx_u32(x972, x963, 0x0, &x974);
+  uint8_t x976 = (0x0 + 0x0);
+  uint32_t _8; uint8_t x979 = addcarryx_u32(0x0, x929, x956, &_8);
+  uint32_t x981; uint8_t x982 = addcarryx_u32(x979, x932, x968, &x981);
+  uint32_t x984; uint8_t x985 = addcarryx_u32(x982, x935, x971, &x984);
+  uint32_t x987; uint8_t x988 = addcarryx_u32(x985, x938, x974, &x987);
+  uint32_t x990; uint8_t x991 = addcarryx_u32(x988, x941, x975, &x990);
+  uint32_t x993; uint8_t x994 = addcarryx_u32(x991, x944, x976, &x993);
+  uint32_t x996; uint8_t x997 = addcarryx_u32(x994, x947, x929, &x996);
+  uint32_t x999; uint8_t x1000 = addcarryx_u32(x997, x950, x965, &x999);
+  uint32_t x1002; uint8_t x1003 = addcarryx_u32(x1000, x953, x966, &x1002);
+  uint8_t x1004 = (x1003 + x954);
+  uint32_t x1006; uint8_t x1007 = subborrow_u32(0x0, x981, 0xffffffff, &x1006);
+  uint32_t x1009; uint8_t x1010 = subborrow_u32(x1007, x984, 0xffffffff, &x1009);
+  uint32_t x1012; uint8_t x1013 = subborrow_u32(x1010, x987, 0xffffffff, &x1012);
+  uint32_t x1015; uint8_t x1016 = subborrow_u32(x1013, x990, 0x0, &x1015);
+  uint32_t x1018; uint8_t x1019 = subborrow_u32(x1016, x993, 0x0, &x1018);
+  uint32_t x1021; uint8_t x1022 = subborrow_u32(x1019, x996, 0x0, &x1021);
+  uint32_t x1024; uint8_t x1025 = subborrow_u32(x1022, x999, 0x1, &x1024);
+  uint32_t x1027; uint8_t x1028 = subborrow_u32(x1025, x1002, 0xffffffff, &x1027);
+  uint32_t _9; uint8_t x1031 = subborrow_u32(x1028, x1004, 0x0, &_9);
+  uint32_t x1032 = cmovznz_u32(x1031, x1027, x1002);
+  uint32_t x1033 = cmovznz_u32(x1031, x1024, x999);
+  uint32_t x1034 = cmovznz_u32(x1031, x1021, x996);
+  uint32_t x1035 = cmovznz_u32(x1031, x1018, x993);
+  uint32_t x1036 = cmovznz_u32(x1031, x1015, x990);
+  uint32_t x1037 = cmovznz_u32(x1031, x1012, x987);
+  uint32_t x1038 = cmovznz_u32(x1031, x1009, x984);
+  uint32_t x1039 = cmovznz_u32(x1031, x1006, x981);
+  out[0] = x1039;
+  out[1] = x1038;
+  out[2] = x1037;
+  out[3] = x1036;
+  out[4] = x1035;
+  out[5] = x1034;
+  out[6] = x1033;
+  out[7] = x1032;
+}
+
+// NOTE: the following functions are generated from fiat-crypto, from the same
+// template as their 64-bit counterparts above, but the correctness proof of
+// the template was not composed with the correctness proof of the
+// specialization pipeline. This is because Coq unexplainedly loops on trying
+// to synthesize opp and sub using the normal pipeline.
+
+static void fe_sub(uint32_t out[8], const uint32_t in1[8], const uint32_t in2[8]) {
+  const uint32_t x14 = in1[7];
+  const uint32_t x15 = in1[6];
+  const uint32_t x13 = in1[5];
+  const uint32_t x11 = in1[4];
+  const uint32_t x9 = in1[3];
+  const uint32_t x7 = in1[2];
+  const uint32_t x5 = in1[1];
+  const uint32_t x3 = in1[0];
+  const uint32_t x28 = in2[7];
+  const uint32_t x29 = in2[6];
+  const uint32_t x27 = in2[5];
+  const uint32_t x25 = in2[4];
+  const uint32_t x23 = in2[3];
+  const uint32_t x21 = in2[2];
+  const uint32_t x19 = in2[1];
+  const uint32_t x17 = in2[0];
+  uint32_t x31; uint8_t x32 = subborrow_u32(0x0, x3, x17, &x31);
+  uint32_t x34; uint8_t x35 = subborrow_u32(x32, x5, x19, &x34);
+  uint32_t x37; uint8_t x38 = subborrow_u32(x35, x7, x21, &x37);
+  uint32_t x40; uint8_t x41 = subborrow_u32(x38, x9, x23, &x40);
+  uint32_t x43; uint8_t x44 = subborrow_u32(x41, x11, x25, &x43);
+  uint32_t x46; uint8_t x47 = subborrow_u32(x44, x13, x27, &x46);
+  uint32_t x49; uint8_t x50 = subborrow_u32(x47, x15, x29, &x49);
+  uint32_t x52; uint8_t x53 = subborrow_u32(x50, x14, x28, &x52);
+  uint32_t x54 = cmovznz_u32(x53, 0x0, 0xffffffff);
+  uint32_t x56; uint8_t x57 = addcarryx_u32(0x0, x31, (x54 & 0xffffffff), &x56);
+  uint32_t x59; uint8_t x60 = addcarryx_u32(x57, x34, (x54 & 0xffffffff), &x59);
+  uint32_t x62; uint8_t x63 = addcarryx_u32(x60, x37, (x54 & 0xffffffff), &x62);
+  uint32_t x65; uint8_t x66 = addcarryx_u32(x63, x40, 0x0, &x65);
+  uint32_t x68; uint8_t x69 = addcarryx_u32(x66, x43, 0x0, &x68);
+  uint32_t x71; uint8_t x72 = addcarryx_u32(x69, x46, 0x0, &x71);
+  uint32_t x74; uint8_t x75 = addcarryx_u32(x72, x49, ((uint8_t)x54 & 0x1), &x74);
+  uint32_t x77; addcarryx_u32(x75, x52, (x54 & 0xffffffff), &x77);
+  out[0] = x56;
+  out[1] = x59;
+  out[2] = x62;
+  out[3] = x65;
+  out[4] = x68;
+  out[5] = x71;
+  out[6] = x74;
+  out[7] = x77;
+}
+
+// fe_op sets out = -in
+static void fe_opp(uint32_t out[8], const uint32_t in1[8]) {
+  const uint32_t x12 = in1[7];
+  const uint32_t x13 = in1[6];
+  const uint32_t x11 = in1[5];
+  const uint32_t x9 = in1[4];
+  const uint32_t x7 = in1[3];
+  const uint32_t x5 = in1[2];
+  const uint32_t x3 = in1[1];
+  const uint32_t x1 = in1[0];
+  uint32_t x15; uint8_t x16 = subborrow_u32(0x0, 0x0, x1, &x15);
+  uint32_t x18; uint8_t x19 = subborrow_u32(x16, 0x0, x3, &x18);
+  uint32_t x21; uint8_t x22 = subborrow_u32(x19, 0x0, x5, &x21);
+  uint32_t x24; uint8_t x25 = subborrow_u32(x22, 0x0, x7, &x24);
+  uint32_t x27; uint8_t x28 = subborrow_u32(x25, 0x0, x9, &x27);
+  uint32_t x30; uint8_t x31 = subborrow_u32(x28, 0x0, x11, &x30);
+  uint32_t x33; uint8_t x34 = subborrow_u32(x31, 0x0, x13, &x33);
+  uint32_t x36; uint8_t x37 = subborrow_u32(x34, 0x0, x12, &x36);
+  uint32_t x38 = cmovznz_u32(x37, 0x0, 0xffffffff);
+  uint32_t x40; uint8_t x41 = addcarryx_u32(0x0, x15, (x38 & 0xffffffff), &x40);
+  uint32_t x43; uint8_t x44 = addcarryx_u32(x41, x18, (x38 & 0xffffffff), &x43);
+  uint32_t x46; uint8_t x47 = addcarryx_u32(x44, x21, (x38 & 0xffffffff), &x46);
+  uint32_t x49; uint8_t x50 = addcarryx_u32(x47, x24, 0x0, &x49);
+  uint32_t x52; uint8_t x53 = addcarryx_u32(x50, x27, 0x0, &x52);
+  uint32_t x55; uint8_t x56 = addcarryx_u32(x53, x30, 0x0, &x55);
+  uint32_t x58; uint8_t x59 = addcarryx_u32(x56, x33, ((uint8_t)x38 & 0x1), &x58);
+  uint32_t x61; addcarryx_u32(x59, x36, (x38 & 0xffffffff), &x61);
+  out[0] = x40;
+  out[1] = x43;
+  out[2] = x46;
+  out[3] = x49;
+  out[4] = x52;
+  out[5] = x55;
+  out[6] = x58;
+  out[7] = x61;
+}
+
+#endif
+
+// utility functions, handwritten
+
+#define NBYTES 32
+
+#if defined(BORINGSSL_NISTP256_64BIT)
+
+#define NLIMBS 4
+typedef uint64_t limb_t;
+#define cmovznz_limb cmovznz_u64
+typedef uint64_t fe[NLIMBS];
+#else // 64BIT; else 32BIT
+
+#define NLIMBS 8
+typedef uint32_t limb_t;
+#define cmovznz_limb cmovznz_u32
+typedef uint32_t fe[NLIMBS];
+
+#endif // 64BIT
+
+static limb_t fe_nz(const limb_t in1[NLIMBS]) {
+  limb_t ret = 0;
+  for (int i = 0; i < NLIMBS; i++) {
+    ret |= in1[i];
+  }
+  return ret;
+}
+
+static void fe_copy(limb_t out[NLIMBS], const limb_t in1[NLIMBS]) {
+  for (int i = 0; i < NLIMBS; i++) {
+    out[i] = in1[i];
+  }
+}
+
+static void fe_cmovznz(limb_t out[NLIMBS], limb_t t, const limb_t z[NLIMBS],
+                       const limb_t nz[NLIMBS]) {
+  for (int i = 0; i < NLIMBS; i++) {
+    out[i] = cmovznz_limb(t, z[i], nz[i]);
+  }
+}
+
+static void fe_sqr(fe out, const fe in) {
+  fe_mul(out, in, in);
+}
+
+static void fe_tobytes(uint8_t out[NBYTES], const fe in) {
+  for (int i = 0; i<NBYTES; i++) {
+    out[i] = (uint8_t)(in[i/sizeof(in[0])] >> (8*(i%sizeof(in[0]))));
+  }
+}
+
+static void fe_frombytes(fe out, const uint8_t in[NBYTES]) {
+  for (int i = 0; i<NLIMBS; i++) {
+    out[i] = 0;
+  }
+  for (int i = 0; i<NBYTES; i++) {
+    out[i/sizeof(out[0])] |= ((limb_t)in[i]) << (8*(i%sizeof(out[0])));
+  }
+}
+
+static void fe_from_montgomery(fe x) {
+  static const limb_t kOne[NLIMBS] = {1, 0};
+  fe_mul(x, x, kOne);
+}
+
+// BN_* compatability wrappers
+
+static int BN_to_fe(fe out, const BIGNUM *bn) {
+  uint8_t tmp[NBYTES];
+  if (!BN_bn2le_padded(tmp, NBYTES, bn)) {
+    return 0;
+  }
+  fe_frombytes(out, tmp);
+  return 1;
+}
+
+static BIGNUM *fe_to_BN(BIGNUM *out, const fe in) {
+  uint8_t tmp[NBYTES];
+  fe_tobytes(tmp, in);
+  return BN_le2bn(tmp, NBYTES, out);
+}
+
+// fe_inv calculates |out| = |in|^{-1}
+//
+// Based on Fermat's Little Theorem:
+//   a^p = a (mod p)
+//   a^{p-1} = 1 (mod p)
+//   a^{p-2} = a^{-1} (mod p)
+static void fe_inv(fe out, const fe in) {
+  fe ftmp, ftmp2;
+  // each e_I will hold |in|^{2^I - 1}
+  fe e2, e4, e8, e16, e32, e64;
+
+  fe_sqr(ftmp, in);  // 2^1
+  fe_mul(ftmp, in, ftmp);  // 2^2 - 2^0
+  fe_copy(e2, ftmp);
+  fe_sqr(ftmp, ftmp);  // 2^3 - 2^1
+  fe_sqr(ftmp, ftmp);  // 2^4 - 2^2
+  fe_mul(ftmp, ftmp, e2);  // 2^4 - 2^0
+  fe_copy(e4, ftmp);
+  fe_sqr(ftmp, ftmp);  // 2^5 - 2^1
+  fe_sqr(ftmp, ftmp);  // 2^6 - 2^2
+  fe_sqr(ftmp, ftmp);  // 2^7 - 2^3
+  fe_sqr(ftmp, ftmp);  // 2^8 - 2^4
+  fe_mul(ftmp, ftmp, e4);  // 2^8 - 2^0
+  fe_copy(e8, ftmp);
+  for (size_t i = 0; i < 8; i++) {
+    fe_sqr(ftmp, ftmp);
+  }  // 2^16 - 2^8
+  fe_mul(ftmp, ftmp, e8);  // 2^16 - 2^0
+  fe_copy(e16, ftmp);
+  for (size_t i = 0; i < 16; i++) {
+    fe_sqr(ftmp, ftmp);
+  }  // 2^32 - 2^16
+  fe_mul(ftmp, ftmp, e16);  // 2^32 - 2^0
+  fe_copy(e32, ftmp);
+  for (size_t i = 0; i < 32; i++) {
+    fe_sqr(ftmp, ftmp);
+  }  // 2^64 - 2^32
+  fe_copy(e64, ftmp);
+  fe_mul(ftmp, ftmp, in);  // 2^64 - 2^32 + 2^0
+  for (size_t i = 0; i < 192; i++) {
+    fe_sqr(ftmp, ftmp);
+  }  // 2^256 - 2^224 + 2^192
+
+  fe_mul(ftmp2, e64, e32);  // 2^64 - 2^0
+  for (size_t i = 0; i < 16; i++) {
+    fe_sqr(ftmp2, ftmp2);
+  }  // 2^80 - 2^16
+  fe_mul(ftmp2, ftmp2, e16);  // 2^80 - 2^0
+  for (size_t i = 0; i < 8; i++) {
+    fe_sqr(ftmp2, ftmp2);
+  }  // 2^88 - 2^8
+  fe_mul(ftmp2, ftmp2, e8);  // 2^88 - 2^0
+  for (size_t i = 0; i < 4; i++) {
+    fe_sqr(ftmp2, ftmp2);
+  }  // 2^92 - 2^4
+  fe_mul(ftmp2, ftmp2, e4);  // 2^92 - 2^0
+  fe_sqr(ftmp2, ftmp2);  // 2^93 - 2^1
+  fe_sqr(ftmp2, ftmp2);  // 2^94 - 2^2
+  fe_mul(ftmp2, ftmp2, e2);  // 2^94 - 2^0
+  fe_sqr(ftmp2, ftmp2);  // 2^95 - 2^1
+  fe_sqr(ftmp2, ftmp2);  // 2^96 - 2^2
+  fe_mul(ftmp2, ftmp2, in);  // 2^96 - 3
+
+  fe_mul(out, ftmp2, ftmp);  // 2^256 - 2^224 + 2^192 + 2^96 - 3
+}
+
+// Group operations
+// ----------------
+//
+// Building on top of the field operations we have the operations on the
+// elliptic curve group itself. Points on the curve are represented in Jacobian
+// coordinates.
+
+// point_double calculates 2*(x_in, y_in, z_in)
+//
+// The method is taken from:
+//   http://hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-3.html#doubling-dbl-2001-b
+//
+// Outputs can equal corresponding inputs, i.e., x_out == x_in is allowed.
+// while x_out == y_in is not (maybe this works, but it's not tested).
+static void point_double(fe x_out, fe y_out, fe z_out,
+                         const fe x_in, const fe y_in, const fe z_in) {
+  fe delta, gamma, beta, ftmp, ftmp2, tmptmp, alpha, fourbeta;
+  // delta = z^2
+  fe_sqr(delta, z_in);
+  // gamma = y^2
+  fe_sqr(gamma, y_in);
+  // beta = x*gamma
+  fe_mul(beta, x_in, gamma);
+
+  // alpha = 3*(x-delta)*(x+delta)
+  fe_sub(ftmp, x_in, delta);
+  fe_add(ftmp2, x_in, delta);
+
+  fe_add(tmptmp, ftmp2, ftmp2);
+  fe_add(ftmp2, ftmp2, tmptmp);
+  fe_mul(alpha, ftmp, ftmp2);
+
+  // x' = alpha^2 - 8*beta
+  fe_sqr(x_out, alpha);
+  fe_add(fourbeta, beta, beta);
+  fe_add(fourbeta, fourbeta, fourbeta);
+  fe_add(tmptmp, fourbeta, fourbeta);
+  fe_sub(x_out, x_out, tmptmp);
+
+  // z' = (y + z)^2 - gamma - delta
+  fe_add(delta, gamma, delta);
+  fe_add(ftmp, y_in, z_in);
+  fe_sqr(z_out, ftmp);
+  fe_sub(z_out, z_out, delta);
+
+  // y' = alpha*(4*beta - x') - 8*gamma^2
+  fe_sub(y_out, fourbeta, x_out);
+  fe_add(gamma, gamma, gamma);
+  fe_sqr(gamma, gamma);
+  fe_mul(y_out, alpha, y_out);
+  fe_add(gamma, gamma, gamma);
+  fe_sub(y_out, y_out, gamma);
+}
+
+// point_add calcuates (x1, y1, z1) + (x2, y2, z2)
+//
+// The method is taken from:
+//   http://hyperelliptic.org/EFD/g1p/auto-shortw-jacobian-3.html#addition-add-2007-bl,
+// adapted for mixed addition (z2 = 1, or z2 = 0 for the point at infinity).
+//
+// This function includes a branch for checking whether the two input points
+// are equal, (while not equal to the point at infinity). This case never
+// happens during single point multiplication, so there is no timing leak for
+// ECDH or ECDSA signing.
+static void point_add(fe x3, fe y3, fe z3, const fe x1,
+                      const fe y1, const fe z1, const int mixed,
+                      const fe x2, const fe y2, const fe z2) {
+  fe x_out, y_out, z_out;
+  limb_t z1nz = fe_nz(z1);
+  limb_t z2nz = fe_nz(z2);
+
+  // z1z1 = z1z1 = z1**2
+  fe z1z1; fe_sqr(z1z1, z1);
+
+  fe u1, s1, two_z1z2;
+  if (!mixed) {
+    // ftmp2 = z2z2 = z2**2
+    fe z2z2; fe_sqr(z2z2, z2);
+
+    // u1 = ftmp3 = x1*z2z2
+    fe_mul(u1, x1, z2z2);
+
+    // two_z1z2 = (z1 + z2)**2 - (z1z1 + z2z2) = 2z1z2
+    fe_add(two_z1z2, z1, z2);
+    fe_sqr(two_z1z2, two_z1z2);
+    fe_sub(two_z1z2, two_z1z2, z1z1);
+    fe_sub(two_z1z2, two_z1z2, z2z2);
+
+    // s1 = ftmp2 = y1 * z2**3
+    fe_mul(s1, z2, z2z2);
+    fe_mul(s1, s1, y1);
+  } else {
+    // We'll assume z2 = 1 (special case z2 = 0 is handled later).
+
+    // u1 = ftmp3 = x1*z2z2
+    fe_copy(u1, x1);
+    // two_z1z2 = 2z1z2
+    fe_add(two_z1z2, z1, z1);
+    // s1 = ftmp2 = y1 * z2**3
+    fe_copy(s1, y1);
+  }
+
+  // u2 = x2*z1z1
+  fe u2; fe_mul(u2, x2, z1z1);
+
+  // h = ftmp4 = u2 - u1
+  fe h; fe_sub(h, u2, u1);
+
+  limb_t xneq = fe_nz(h);
+
+  // z_out = two_z1z2 * h
+  fe_mul(z_out, h, two_z1z2);
+
+  // z1z1z1 = z1 * z1z1
+  fe z1z1z1; fe_mul(z1z1z1, z1, z1z1);
+
+  // s2 = tmp = y2 * z1**3
+  fe s2; fe_mul(s2, y2, z1z1z1);
+
+  // r = (s2 - s1)*2
+  fe r;
+  fe_sub(r, s2, s1);
+  fe_add(r, r, r);
+
+  limb_t yneq = fe_nz(r);
+
+  if (!xneq && !yneq && z1nz && z2nz) {
+    point_double(x_out, y_out, z_out, x1, y1, z1);
+    return;
+  }
+
+  // I = (2h)**2
+  fe i;
+  fe_add(i, h, h);
+  fe_sqr(i, i);
+
+  // J = ftmp2 = h * I
+  fe j; fe_mul(j, h, i);
+
+  // V = ftmp4 = U1 * I
+  fe v; fe_mul(v, u1, i);
+
+  // x_out = r**2 - J - 2V
+  fe_sqr(x_out, r);
+  fe_sub(x_out, x_out, j);
+  fe_sub(x_out, x_out, v);
+  fe_sub(x_out, x_out, v);
+
+  // y_out = r(V-x_out) - 2 * s1 * J
+  fe_sub(y_out, v, x_out);
+  fe_mul(y_out, y_out, r);
+  fe s1j;
+  fe_mul(s1j, s1, j);
+  fe_sub(y_out, y_out, s1j);
+  fe_sub(y_out, y_out, s1j);
+
+  fe_cmovznz(x_out, z1nz, x2, x_out);
+  fe_cmovznz(x3, z2nz, x1, x_out);
+  fe_cmovznz(y_out, z1nz, y2, y_out);
+  fe_cmovznz(y3, z2nz, y1, y_out);
+  fe_cmovznz(z_out, z1nz, z2, z_out);
+  fe_cmovznz(z3, z2nz, z1, z_out);
+}
+
+// Base point pre computation
+// --------------------------
+//
+// Two different sorts of precomputed tables are used in the following code.
+// Each contain various points on the curve, where each point is three field
+// elements (x, y, z).
+//
+// For the base point table, z is usually 1 (0 for the point at infinity).
+// This table has 2 * 16 elements, starting with the following:
+// index | bits    | point
+// ------+---------+------------------------------
+//     0 | 0 0 0 0 | 0G
+//     1 | 0 0 0 1 | 1G
+//     2 | 0 0 1 0 | 2^64G
+//     3 | 0 0 1 1 | (2^64 + 1)G
+//     4 | 0 1 0 0 | 2^128G
+//     5 | 0 1 0 1 | (2^128 + 1)G
+//     6 | 0 1 1 0 | (2^128 + 2^64)G
+//     7 | 0 1 1 1 | (2^128 + 2^64 + 1)G
+//     8 | 1 0 0 0 | 2^192G
+//     9 | 1 0 0 1 | (2^192 + 1)G
+//    10 | 1 0 1 0 | (2^192 + 2^64)G
+//    11 | 1 0 1 1 | (2^192 + 2^64 + 1)G
+//    12 | 1 1 0 0 | (2^192 + 2^128)G
+//    13 | 1 1 0 1 | (2^192 + 2^128 + 1)G
+//    14 | 1 1 1 0 | (2^192 + 2^128 + 2^64)G
+//    15 | 1 1 1 1 | (2^192 + 2^128 + 2^64 + 1)G
+// followed by a copy of this with each element multiplied by 2^32.
+//
+// The reason for this is so that we can clock bits into four different
+// locations when doing simple scalar multiplies against the base point,
+// and then another four locations using the second 16 elements.
+//
+// Tables for other points have table[i] = iG for i in 0 .. 16.
+
+// g_pre_comp is the table of precomputed base points
+#if defined(BORINGSSL_NISTP256_64BIT)
+static const fe g_pre_comp[2][16][3] = {
+    {{{0x0, 0x0, 0x0, 0x0}, {0x0, 0x0, 0x0, 0x0}, {0x0, 0x0, 0x0, 0x0}},
+     {{0x79e730d418a9143c, 0x75ba95fc5fedb601, 0x79fb732b77622510,
+       0x18905f76a53755c6},
+      {0xddf25357ce95560a, 0x8b4ab8e4ba19e45c, 0xd2e88688dd21f325,
+       0x8571ff1825885d85},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x4f922fc516a0d2bb, 0xd5cc16c1a623499, 0x9241cf3a57c62c8b,
+       0x2f5e6961fd1b667f},
+      {0x5c15c70bf5a01797, 0x3d20b44d60956192, 0x4911b37071fdb52,
+       0xf648f9168d6f0f7b},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x9e566847e137bbbc, 0xe434469e8a6a0bec, 0xb1c4276179d73463,
+       0x5abe0285133d0015},
+      {0x92aa837cc04c7dab, 0x573d9f4c43260c07, 0xc93156278e6cc37,
+       0x94bb725b6b6f7383},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x62a8c244bfe20925, 0x91c19ac38fdce867, 0x5a96a5d5dd387063,
+       0x61d587d421d324f6},
+      {0xe87673a2a37173ea, 0x2384800853778b65, 0x10f8441e05bab43e,
+       0xfa11fe124621efbe},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x1c891f2b2cb19ffd, 0x1ba8d5bb1923c23, 0xb6d03d678ac5ca8e,
+       0x586eb04c1f13bedc},
+      {0xc35c6e527e8ed09, 0x1e81a33c1819ede2, 0x278fd6c056c652fa,
+       0x19d5ac0870864f11},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x62577734d2b533d5, 0x673b8af6a1bdddc0, 0x577e7c9aa79ec293,
+       0xbb6de651c3b266b1},
+      {0xe7e9303ab65259b3, 0xd6a0afd3d03a7480, 0xc5ac83d19b3cfc27,
+       0x60b4619a5d18b99b},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xbd6a38e11ae5aa1c, 0xb8b7652b49e73658, 0xb130014ee5f87ed,
+       0x9d0f27b2aeebffcd},
+      {0xca9246317a730a55, 0x9c955b2fddbbc83a, 0x7c1dfe0ac019a71,
+       0x244a566d356ec48d},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x56f8410ef4f8b16a, 0x97241afec47b266a, 0xa406b8e6d9c87c1,
+       0x803f3e02cd42ab1b},
+      {0x7f0309a804dbec69, 0xa83b85f73bbad05f, 0xc6097273ad8e197f,
+       0xc097440e5067adc1},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x846a56f2c379ab34, 0xa8ee068b841df8d1, 0x20314459176c68ef,
+       0xf1af32d5915f1f30},
+      {0x99c375315d75bd50, 0x837cffbaf72f67bc, 0x613a41848d7723f,
+       0x23d0f130e2d41c8b},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xed93e225d5be5a2b, 0x6fe799835934f3c6, 0x4314092622626ffc,
+       0x50bbb4d97990216a},
+      {0x378191c6e57ec63e, 0x65422c40181dcdb2, 0x41a8099b0236e0f6,
+       0x2b10011801fe49c3},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xfc68b5c59b391593, 0xc385f5a2598270fc, 0x7144f3aad19adcbb,
+       0xdd55899983fbae0c},
+      {0x93b88b8e74b82ff4, 0xd2e03c4071e734c9, 0x9a7a9eaf43c0322a,
+       0xe6e4c551149d6041},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x5fe14bfe80ec21fe, 0xf6ce116ac255be82, 0x98bc5a072f4a5d67,
+       0xfad27148db7e63af},
+      {0x90c0b6ac29ab05b3, 0x37a9a83c4e251ae6, 0xa7dc875c2aade7d,
+       0x77387de39f0e1a84},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x1e9ecc49a56c0dd7, 0xa5cffcd846086c74, 0x8f7a1408f505aece,
+       0xb37b85c0bef0c47e},
+      {0x3596b6e4cc0e6a8f, 0xfd6d4bbf6b388f23, 0xaba453fac39cef4e,
+       0x9c135ac8f9f628d5},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xa1c729495c8f8be, 0x2961c4803bf362bf, 0x9e418403df63d4ac,
+       0xc109f9cb91ece900},
+      {0xc2d095d058945705, 0xb9083d96ddeb85c0, 0x84692b8d7a40449b,
+       0x9bc3344f2eee1ee1},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xd5ae35642913074, 0x55491b2748a542b1, 0x469ca665b310732a,
+       0x29591d525f1a4cc1},
+      {0xe76f5b6bb84f983f, 0xbe7eef419f5f84e1, 0x1200d49680baa189,
+       0x6376551f18ef332c},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}}},
+    {{{0x0, 0x0, 0x0, 0x0}, {0x0, 0x0, 0x0, 0x0}, {0x0, 0x0, 0x0, 0x0}},
+     {{0x202886024147519a, 0xd0981eac26b372f0, 0xa9d4a7caa785ebc8,
+       0xd953c50ddbdf58e9},
+      {0x9d6361ccfd590f8f, 0x72e9626b44e6c917, 0x7fd9611022eb64cf,
+       0x863ebb7e9eb288f3},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x4fe7ee31b0e63d34, 0xf4600572a9e54fab, 0xc0493334d5e7b5a4,
+       0x8589fb9206d54831},
+      {0xaa70f5cc6583553a, 0x879094ae25649e5, 0xcc90450710044652,
+       0xebb0696d02541c4f},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xabbaa0c03b89da99, 0xa6f2d79eb8284022, 0x27847862b81c05e8,
+       0x337a4b5905e54d63},
+      {0x3c67500d21f7794a, 0x207005b77d6d7f61, 0xa5a378104cfd6e8,
+       0xd65e0d5f4c2fbd6},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xd433e50f6d3549cf, 0x6f33696ffacd665e, 0x695bfdacce11fcb4,
+       0x810ee252af7c9860},
+      {0x65450fe17159bb2c, 0xf7dfbebe758b357b, 0x2b057e74d69fea72,
+       0xd485717a92731745},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xce1f69bbe83f7669, 0x9f8ae8272877d6b, 0x9548ae543244278d,
+       0x207755dee3c2c19c},
+      {0x87bd61d96fef1945, 0x18813cefb12d28c3, 0x9fbcd1d672df64aa,
+       0x48dc5ee57154b00d},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xef0f469ef49a3154, 0x3e85a5956e2b2e9a, 0x45aaec1eaa924a9c,
+       0xaa12dfc8a09e4719},
+      {0x26f272274df69f1d, 0xe0e4c82ca2ff5e73, 0xb9d8ce73b7a9dd44,
+       0x6c036e73e48ca901},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xe1e421e1a47153f0, 0xb86c3b79920418c9, 0x93bdce87705d7672,
+       0xf25ae793cab79a77},
+      {0x1f3194a36d869d0c, 0x9d55c8824986c264, 0x49fb5ea3096e945e,
+       0x39b8e65313db0a3e},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xe3417bc035d0b34a, 0x440b386b8327c0a7, 0x8fb7262dac0362d1,
+       0x2c41114ce0cdf943},
+      {0x2ba5cef1ad95a0b1, 0xc09b37a867d54362, 0x26d6cdd201e486c9,
+       0x20477abf42ff9297},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xf121b41bc0a67d2, 0x62d4760a444d248a, 0xe044f1d659b4737,
+       0x8fde365250bb4a8},
+      {0xaceec3da848bf287, 0xc2a62182d3369d6e, 0x3582dfdc92449482,
+       0x2f7e2fd2565d6cd7},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xa0122b5178a876b, 0x51ff96ff085104b4, 0x50b31ab14f29f76,
+       0x84abb28b5f87d4e6},
+      {0xd5ed439f8270790a, 0x2d6cb59d85e3f46b, 0x75f55c1b6c1e2212,
+       0xe5436f6717655640},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xc2965ecc9aeb596d, 0x1ea03e7023c92b4, 0x4704b4b62e013961,
+       0xca8fd3f905ea367},
+      {0x92523a42551b2b61, 0x1eb7a89c390fcd06, 0xe7f1d2be0392a63e,
+       0x96dca2644ddb0c33},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x231c210e15339848, 0xe87a28e870778c8d, 0x9d1de6616956e170,
+       0x4ac3c9382bb09c0b},
+      {0x19be05516998987d, 0x8b2376c4ae09f4d6, 0x1de0b7651a3f933d,
+       0x380d94c7e39705f4},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x3685954b8c31c31d, 0x68533d005bf21a0c, 0xbd7626e75c79ec9,
+       0xca17754742c69d54},
+      {0xcc6edafff6d2dbb2, 0xfd0d8cbd174a9d18, 0x875e8793aa4578e8,
+       0xa976a7139cab2ce6},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0xce37ab11b43ea1db, 0xa7ff1a95259d292, 0x851b02218f84f186,
+       0xa7222beadefaad13},
+      {0xa2ac78ec2b0a9144, 0x5a024051f2fa59c5, 0x91d1eca56147ce38,
+       0xbe94d523bc2ac690},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}},
+     {{0x2d8daefd79ec1a0f, 0x3bbcd6fdceb39c97, 0xf5575ffc58f61a95,
+       0xdbd986c4adf7b420},
+      {0x81aa881415f39eb7, 0x6ee2fcf5b98d976c, 0x5465475dcf2f717d,
+       0x8e24d3c46860bbd0},
+      {0x1, 0xffffffff00000000, 0xffffffffffffffff, 0xfffffffe}}}};
+#else
+static const fe g_pre_comp[2][16][3] = {
+    {{{0x0,0x0, 0x0,0x0, 0x0,0x0, 0x0,0x0},
+      {0x0,0x0, 0x0,0x0, 0x0,0x0, 0x0,0x0},
+      {0x0,0x0, 0x0,0x0, 0x0,0x0, 0x0,0x0}},
+     {{0x18a9143c,0x79e730d4, 0x5fedb601,0x75ba95fc, 0x77622510,0x79fb732b,
+       0xa53755c6,0x18905f76},
+      {0xce95560a,0xddf25357, 0xba19e45c,0x8b4ab8e4, 0xdd21f325,0xd2e88688,
+       0x25885d85,0x8571ff18},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x16a0d2bb,0x4f922fc5, 0x1a623499,0xd5cc16c, 0x57c62c8b,0x9241cf3a,
+       0xfd1b667f,0x2f5e6961},
+      {0xf5a01797,0x5c15c70b, 0x60956192,0x3d20b44d, 0x71fdb52,0x4911b37,
+       0x8d6f0f7b,0xf648f916},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xe137bbbc,0x9e566847, 0x8a6a0bec,0xe434469e, 0x79d73463,0xb1c42761,
+       0x133d0015,0x5abe0285},
+      {0xc04c7dab,0x92aa837c, 0x43260c07,0x573d9f4c, 0x78e6cc37,0xc931562,
+       0x6b6f7383,0x94bb725b},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xbfe20925,0x62a8c244, 0x8fdce867,0x91c19ac3, 0xdd387063,0x5a96a5d5,
+       0x21d324f6,0x61d587d4},
+      {0xa37173ea,0xe87673a2, 0x53778b65,0x23848008, 0x5bab43e,0x10f8441e,
+       0x4621efbe,0xfa11fe12},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x2cb19ffd,0x1c891f2b, 0xb1923c23,0x1ba8d5b, 0x8ac5ca8e,0xb6d03d67,
+       0x1f13bedc,0x586eb04c},
+      {0x27e8ed09,0xc35c6e5, 0x1819ede2,0x1e81a33c, 0x56c652fa,0x278fd6c0,
+       0x70864f11,0x19d5ac08},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xd2b533d5,0x62577734, 0xa1bdddc0,0x673b8af6, 0xa79ec293,0x577e7c9a,
+       0xc3b266b1,0xbb6de651},
+      {0xb65259b3,0xe7e9303a, 0xd03a7480,0xd6a0afd3, 0x9b3cfc27,0xc5ac83d1,
+       0x5d18b99b,0x60b4619a},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x1ae5aa1c,0xbd6a38e1, 0x49e73658,0xb8b7652b, 0xee5f87ed,0xb130014,
+       0xaeebffcd,0x9d0f27b2},
+      {0x7a730a55,0xca924631, 0xddbbc83a,0x9c955b2f, 0xac019a71,0x7c1dfe0,
+       0x356ec48d,0x244a566d},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xf4f8b16a,0x56f8410e, 0xc47b266a,0x97241afe, 0x6d9c87c1,0xa406b8e,
+       0xcd42ab1b,0x803f3e02},
+      {0x4dbec69,0x7f0309a8, 0x3bbad05f,0xa83b85f7, 0xad8e197f,0xc6097273,
+       0x5067adc1,0xc097440e},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xc379ab34,0x846a56f2, 0x841df8d1,0xa8ee068b, 0x176c68ef,0x20314459,
+       0x915f1f30,0xf1af32d5},
+      {0x5d75bd50,0x99c37531, 0xf72f67bc,0x837cffba, 0x48d7723f,0x613a418,
+       0xe2d41c8b,0x23d0f130},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xd5be5a2b,0xed93e225, 0x5934f3c6,0x6fe79983, 0x22626ffc,0x43140926,
+       0x7990216a,0x50bbb4d9},
+      {0xe57ec63e,0x378191c6, 0x181dcdb2,0x65422c40, 0x236e0f6,0x41a8099b,
+       0x1fe49c3,0x2b100118},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x9b391593,0xfc68b5c5, 0x598270fc,0xc385f5a2, 0xd19adcbb,0x7144f3aa,
+       0x83fbae0c,0xdd558999},
+      {0x74b82ff4,0x93b88b8e, 0x71e734c9,0xd2e03c40, 0x43c0322a,0x9a7a9eaf,
+       0x149d6041,0xe6e4c551},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x80ec21fe,0x5fe14bfe, 0xc255be82,0xf6ce116a, 0x2f4a5d67,0x98bc5a07,
+       0xdb7e63af,0xfad27148},
+      {0x29ab05b3,0x90c0b6ac, 0x4e251ae6,0x37a9a83c, 0xc2aade7d,0xa7dc875,
+       0x9f0e1a84,0x77387de3},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xa56c0dd7,0x1e9ecc49, 0x46086c74,0xa5cffcd8, 0xf505aece,0x8f7a1408,
+       0xbef0c47e,0xb37b85c0},
+      {0xcc0e6a8f,0x3596b6e4, 0x6b388f23,0xfd6d4bbf, 0xc39cef4e,0xaba453fa,
+       0xf9f628d5,0x9c135ac8},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x95c8f8be,0xa1c7294, 0x3bf362bf,0x2961c480, 0xdf63d4ac,0x9e418403,
+       0x91ece900,0xc109f9cb},
+      {0x58945705,0xc2d095d0, 0xddeb85c0,0xb9083d96, 0x7a40449b,0x84692b8d,
+       0x2eee1ee1,0x9bc3344f},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x42913074,0xd5ae356, 0x48a542b1,0x55491b27, 0xb310732a,0x469ca665,
+       0x5f1a4cc1,0x29591d52},
+      {0xb84f983f,0xe76f5b6b, 0x9f5f84e1,0xbe7eef41, 0x80baa189,0x1200d496,
+       0x18ef332c,0x6376551f},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}}},
+    {{{0x0,0x0, 0x0,0x0, 0x0,0x0, 0x0,0x0},
+      {0x0,0x0, 0x0,0x0, 0x0,0x0, 0x0,0x0},
+      {0x0,0x0, 0x0,0x0, 0x0,0x0, 0x0,0x0}},
+     {{0x4147519a,0x20288602, 0x26b372f0,0xd0981eac, 0xa785ebc8,0xa9d4a7ca,
+       0xdbdf58e9,0xd953c50d},
+      {0xfd590f8f,0x9d6361cc, 0x44e6c917,0x72e9626b, 0x22eb64cf,0x7fd96110,
+       0x9eb288f3,0x863ebb7e},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xb0e63d34,0x4fe7ee31, 0xa9e54fab,0xf4600572, 0xd5e7b5a4,0xc0493334,
+       0x6d54831,0x8589fb92},
+      {0x6583553a,0xaa70f5cc, 0xe25649e5,0x879094a, 0x10044652,0xcc904507,
+       0x2541c4f,0xebb0696d},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x3b89da99,0xabbaa0c0, 0xb8284022,0xa6f2d79e, 0xb81c05e8,0x27847862,
+       0x5e54d63,0x337a4b59},
+      {0x21f7794a,0x3c67500d, 0x7d6d7f61,0x207005b7, 0x4cfd6e8,0xa5a3781,
+       0xf4c2fbd6,0xd65e0d5},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x6d3549cf,0xd433e50f, 0xfacd665e,0x6f33696f, 0xce11fcb4,0x695bfdac,
+       0xaf7c9860,0x810ee252},
+      {0x7159bb2c,0x65450fe1, 0x758b357b,0xf7dfbebe, 0xd69fea72,0x2b057e74,
+       0x92731745,0xd485717a},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xe83f7669,0xce1f69bb, 0x72877d6b,0x9f8ae82, 0x3244278d,0x9548ae54,
+       0xe3c2c19c,0x207755de},
+      {0x6fef1945,0x87bd61d9, 0xb12d28c3,0x18813cef, 0x72df64aa,0x9fbcd1d6,
+       0x7154b00d,0x48dc5ee5},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xf49a3154,0xef0f469e, 0x6e2b2e9a,0x3e85a595, 0xaa924a9c,0x45aaec1e,
+       0xa09e4719,0xaa12dfc8},
+      {0x4df69f1d,0x26f27227, 0xa2ff5e73,0xe0e4c82c, 0xb7a9dd44,0xb9d8ce73,
+       0xe48ca901,0x6c036e73},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xa47153f0,0xe1e421e1, 0x920418c9,0xb86c3b79, 0x705d7672,0x93bdce87,
+       0xcab79a77,0xf25ae793},
+      {0x6d869d0c,0x1f3194a3, 0x4986c264,0x9d55c882, 0x96e945e,0x49fb5ea3,
+       0x13db0a3e,0x39b8e653},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x35d0b34a,0xe3417bc0, 0x8327c0a7,0x440b386b, 0xac0362d1,0x8fb7262d,
+       0xe0cdf943,0x2c41114c},
+      {0xad95a0b1,0x2ba5cef1, 0x67d54362,0xc09b37a8, 0x1e486c9,0x26d6cdd2,
+       0x42ff9297,0x20477abf},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xbc0a67d2,0xf121b41, 0x444d248a,0x62d4760a, 0x659b4737,0xe044f1d,
+       0x250bb4a8,0x8fde365},
+      {0x848bf287,0xaceec3da, 0xd3369d6e,0xc2a62182, 0x92449482,0x3582dfdc,
+       0x565d6cd7,0x2f7e2fd2},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x178a876b,0xa0122b5, 0x85104b4,0x51ff96ff, 0x14f29f76,0x50b31ab,
+       0x5f87d4e6,0x84abb28b},
+      {0x8270790a,0xd5ed439f, 0x85e3f46b,0x2d6cb59d, 0x6c1e2212,0x75f55c1b,
+       0x17655640,0xe5436f67},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x9aeb596d,0xc2965ecc, 0x23c92b4,0x1ea03e7, 0x2e013961,0x4704b4b6,
+       0x905ea367,0xca8fd3f},
+      {0x551b2b61,0x92523a42, 0x390fcd06,0x1eb7a89c, 0x392a63e,0xe7f1d2be,
+       0x4ddb0c33,0x96dca264},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x15339848,0x231c210e, 0x70778c8d,0xe87a28e8, 0x6956e170,0x9d1de661,
+       0x2bb09c0b,0x4ac3c938},
+      {0x6998987d,0x19be0551, 0xae09f4d6,0x8b2376c4, 0x1a3f933d,0x1de0b765,
+       0xe39705f4,0x380d94c7},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x8c31c31d,0x3685954b, 0x5bf21a0c,0x68533d00, 0x75c79ec9,0xbd7626e,
+       0x42c69d54,0xca177547},
+      {0xf6d2dbb2,0xcc6edaff, 0x174a9d18,0xfd0d8cbd, 0xaa4578e8,0x875e8793,
+       0x9cab2ce6,0xa976a713},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0xb43ea1db,0xce37ab11, 0x5259d292,0xa7ff1a9, 0x8f84f186,0x851b0221,
+       0xdefaad13,0xa7222bea},
+      {0x2b0a9144,0xa2ac78ec, 0xf2fa59c5,0x5a024051, 0x6147ce38,0x91d1eca5,
+       0xbc2ac690,0xbe94d523},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}},
+     {{0x79ec1a0f,0x2d8daefd, 0xceb39c97,0x3bbcd6fd, 0x58f61a95,0xf5575ffc,
+       0xadf7b420,0xdbd986c4},
+      {0x15f39eb7,0x81aa8814, 0xb98d976c,0x6ee2fcf5, 0xcf2f717d,0x5465475d,
+       0x6860bbd0,0x8e24d3c4},
+      {0x1,0x0, 0x0,0xffffffff, 0xffffffff,0xffffffff, 0xfffffffe,0x0}}}};
+#endif
+
+// select_point selects the |idx|th point from a precomputation table and
+// copies it to out.
+static void select_point(const limb_t idx, size_t size,
+                         const fe pre_comp[/*size*/][3],
+                         fe out[3]) {
+  OPENSSL_memset(out, 0, sizeof(fe) * 3);
+  for (size_t i = 0; i < size; i++) {
+    limb_t mismatch = i ^ idx;
+    fe_cmovznz(out[0], mismatch, pre_comp[i][0], out[0]);
+    fe_cmovznz(out[1], mismatch, pre_comp[i][1], out[1]);
+    fe_cmovznz(out[2], mismatch, pre_comp[i][2], out[2]);
+  }
+}
+
+// get_bit returns the |i|th bit in |in|
+static char get_bit(const uint8_t *in, int i) {
+  if (i < 0 || i >= 256) {
+    return 0;
+  }
+  return (in[i >> 3] >> (i & 7)) & 1;
+}
+
+// Interleaved point multiplication using precomputed point multiples: The
+// small point multiples 0*P, 1*P, ..., 17*P are in p_pre_comp, the scalar
+// in p_scalar, if non-NULL. If g_scalar is non-NULL, we also add this multiple
+// of the generator, using certain (large) precomputed multiples in g_pre_comp.
+// Output point (X, Y, Z) is stored in x_out, y_out, z_out.
+static void batch_mul(fe x_out, fe y_out, fe z_out,
+                      const uint8_t *p_scalar, const uint8_t *g_scalar,
+                      const fe p_pre_comp[17][3]) {
+  // set nq to the point at infinity
+  fe nq[3] = {{0},{0},{0}}, ftmp, tmp[3];
+  uint64_t bits;
+  uint8_t sign, digit;
+
+  // Loop over both scalars msb-to-lsb, interleaving additions of multiples
+  // of the generator (two in each of the last 32 rounds) and additions of p
+  // (every 5th round).
+
+  int skip = 1;  // save two point operations in the first round
+  size_t i = p_scalar != NULL ? 255 : 31;
+  for (;;) {
+    // double
+    if (!skip) {
+      point_double(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2]);
+    }
+
+    // add multiples of the generator
+    if (g_scalar != NULL && i <= 31) {
+      // first, look 32 bits upwards
+      bits = get_bit(g_scalar, i + 224) << 3;
+      bits |= get_bit(g_scalar, i + 160) << 2;
+      bits |= get_bit(g_scalar, i + 96) << 1;
+      bits |= get_bit(g_scalar, i + 32);
+      // select the point to add, in constant time
+      select_point(bits, 16, g_pre_comp[1], tmp);
+
+      if (!skip) {
+        point_add(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2], 1 /* mixed */,
+                  tmp[0], tmp[1], tmp[2]);
+      } else {
+        fe_copy(nq[0], tmp[0]);
+        fe_copy(nq[1], tmp[1]);
+        fe_copy(nq[2], tmp[2]);
+        skip = 0;
+      }
+
+      // second, look at the current position
+      bits = get_bit(g_scalar, i + 192) << 3;
+      bits |= get_bit(g_scalar, i + 128) << 2;
+      bits |= get_bit(g_scalar, i + 64) << 1;
+      bits |= get_bit(g_scalar, i);
+      // select the point to add, in constant time
+      select_point(bits, 16, g_pre_comp[0], tmp);
+      point_add(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2], 1 /* mixed */, tmp[0],
+                tmp[1], tmp[2]);
+    }
+
+    // do other additions every 5 doublings
+    if (p_scalar != NULL && i % 5 == 0) {
+      bits = get_bit(p_scalar, i + 4) << 5;
+      bits |= get_bit(p_scalar, i + 3) << 4;
+      bits |= get_bit(p_scalar, i + 2) << 3;
+      bits |= get_bit(p_scalar, i + 1) << 2;
+      bits |= get_bit(p_scalar, i) << 1;
+      bits |= get_bit(p_scalar, i - 1);
+      ec_GFp_nistp_recode_scalar_bits(&sign, &digit, bits);
+
+      // select the point to add or subtract, in constant time.
+      select_point(digit, 17, p_pre_comp, tmp);
+      fe_opp(ftmp, tmp[1]);  // (X, -Y, Z) is the negative point.
+      fe_cmovznz(tmp[1], sign, tmp[1], ftmp);
+
+      if (!skip) {
+        point_add(nq[0], nq[1], nq[2], nq[0], nq[1], nq[2], 0 /* mixed */,
+                  tmp[0], tmp[1], tmp[2]);
+      } else {
+        fe_copy(nq[0], tmp[0]);
+        fe_copy(nq[1], tmp[1]);
+        fe_copy(nq[2], tmp[2]);
+        skip = 0;
+      }
+    }
+
+    if (i == 0) {
+      break;
+    }
+    --i;
+  }
+  fe_copy(x_out, nq[0]);
+  fe_copy(y_out, nq[1]);
+  fe_copy(z_out, nq[2]);
+}
+
+// OPENSSL EC_METHOD FUNCTIONS
+
+// Takes the Jacobian coordinates (X, Y, Z) of a point and returns (X', Y') =
+// (X/Z^2, Y/Z^3).
+static int ec_GFp_nistp256_point_get_affine_coordinates(const EC_GROUP *group,
+                                                        const EC_POINT *point,
+                                                        BIGNUM *x_out,
+                                                        BIGNUM *y_out,
+                                                        BN_CTX *ctx) {
+  fe x, y, z1, z2;
+
+  if (EC_POINT_is_at_infinity(group, point)) {
+    OPENSSL_PUT_ERROR(EC, EC_R_POINT_AT_INFINITY);
+    return 0;
+  }
+  if (!BN_to_fe(x, &point->X) ||
+      !BN_to_fe(y, &point->Y) ||
+      !BN_to_fe(z1, &point->Z)) {
+    return 0;
+  }
+
+  fe_inv(z2, z1);
+  fe_sqr(z1, z2);
+
+  if (x_out != NULL) {
+    fe_mul(x, x, z1);
+    fe_from_montgomery(x);
+    if (!fe_to_BN(x_out, x)) {
+      OPENSSL_PUT_ERROR(EC, ERR_R_BN_LIB);
+      return 0;
+    }
+  }
+
+  if (y_out != NULL) {
+    fe_mul(z1, z1, z2);
+    fe_mul(y, y, z1);
+    fe_from_montgomery(y);
+    if (!fe_to_BN(y_out, y)) {
+      OPENSSL_PUT_ERROR(EC, ERR_R_BN_LIB);
+      return 0;
+    }
+  }
+
+  return 1;
+}
+
+static int ec_GFp_nistp256_points_mul(const EC_GROUP *group, EC_POINT *r,
+                                      const EC_SCALAR *g_scalar,
+                                      const EC_POINT *p,
+                                      const EC_SCALAR *p_scalar,
+                                      BN_CTX *unused_ctx) {
+  fe p_pre_comp[17][3];
+  fe x_out, y_out, z_out;
+
+  if (p != NULL && p_scalar != NULL) {
+    // We treat NULL scalars as 0, and NULL points as points at infinity, i.e.,
+    // they contribute nothing to the linear combination.
+    OPENSSL_memset(&p_pre_comp, 0, sizeof(p_pre_comp));
+    // Precompute multiples.
+    if (!BN_to_fe(p_pre_comp[1][0], &p->X) ||
+        !BN_to_fe(p_pre_comp[1][1], &p->Y) ||
+        !BN_to_fe(p_pre_comp[1][2], &p->Z)) {
+      return 0;
+    }
+    for (size_t j = 2; j <= 16; ++j) {
+      if (j & 1) {
+        point_add(p_pre_comp[j][0], p_pre_comp[j][1],
+                  p_pre_comp[j][2], p_pre_comp[1][0],
+                  p_pre_comp[1][1], p_pre_comp[1][2],
+                  0,
+                  p_pre_comp[j - 1][0], p_pre_comp[j - 1][1],
+                  p_pre_comp[j - 1][2]);
+      } else {
+        point_double(p_pre_comp[j][0], p_pre_comp[j][1],
+                     p_pre_comp[j][2], p_pre_comp[j / 2][0],
+                     p_pre_comp[j / 2][1], p_pre_comp[j / 2][2]);
+      }
+    }
+  }
+
+  batch_mul(x_out, y_out, z_out,
+            (p != NULL && p_scalar != NULL) ? p_scalar->bytes : NULL,
+            g_scalar != NULL ? g_scalar->bytes : NULL,
+            (const fe (*) [3])p_pre_comp);
+
+  if (!fe_to_BN(&r->X, x_out) ||
+      !fe_to_BN(&r->Y, y_out) ||
+      !fe_to_BN(&r->Z, z_out)) {
+    OPENSSL_PUT_ERROR(EC, ERR_R_BN_LIB);
+    return 0;
+  }
+  return 1;
+}
+
+DEFINE_METHOD_FUNCTION(EC_METHOD, EC_GFp_nistp256_method) {
+  out->group_init = ec_GFp_mont_group_init;
+  out->group_finish = ec_GFp_mont_group_finish;
+  out->group_set_curve = ec_GFp_mont_group_set_curve;
+  out->point_get_affine_coordinates =
+    ec_GFp_nistp256_point_get_affine_coordinates;
+  out->mul = ec_GFp_nistp256_points_mul;
+// The variable-time wNAF point multiplication uses fewer field operations than
+// the constant-time implementation here, but the 64-bit field arithmetic in
+// this file is much faster than the generic BIGNUM-based field arithmetic used
+// by wNAF. For 32-bit, the wNAF code is overall ~60% faster on non-precomputed
+// points, so we use it for public inputs.
+#if defined(BORINGSSL_NISTP256_64BIT)
+  out->mul_public = ec_GFp_nistp256_points_mul;
+#else
+  out->mul_public = ec_wNAF_mul;
+#endif
+  out->field_mul = ec_GFp_mont_field_mul;
+  out->field_sqr = ec_GFp_mont_field_sqr;
+  out->field_encode = ec_GFp_mont_field_encode;
+  out->field_decode = ec_GFp_mont_field_decode;
+};
+
+#undef BORINGSSL_NISTP256_64BIT