enzoic 1.0.3

data/ext/phc-winner-argon2/src/blake2/blamka-round-opt.h

@@ -0,0 +1,328 @@
+/*
+ * Argon2 reference source code package - reference C implementations
+ *
+ * Copyright 2015
+ * Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
+ *
+ * You may use this work under the terms of a Creative Commons CC0 1.0 
+ * License/Waiver or the Apache Public License 2.0, at your option. The terms of
+ * these licenses can be found at:
+ *
+ * - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
+ * - Apache 2.0        : http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * You should have received a copy of both of these licenses along with this
+ * software. If not, they may be obtained at the above URLs.
+ */
+
+#ifndef BLAKE_ROUND_MKA_OPT_H
+#define BLAKE_ROUND_MKA_OPT_H
+
+#include "blake2-impl.h"
+
+#include <emmintrin.h>
+#if defined(__SSSE3__)
+#include <tmmintrin.h> /* for _mm_shuffle_epi8 and _mm_alignr_epi8 */
+#endif
+
+#if defined(__XOP__) && (defined(__GNUC__) || defined(__clang__))
+#include <x86intrin.h>
+#endif
+
+#if !defined(__AVX2__)
+#if !defined(__XOP__)
+#if defined(__SSSE3__)
+#define r16                                                                    \
+    (_mm_setr_epi8(2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9))
+#define r24                                                                    \
+    (_mm_setr_epi8(3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10))
+#define _mm_roti_epi64(x, c)                                                   \
+    (-(c) == 32)                                                               \
+        ? _mm_shuffle_epi32((x), _MM_SHUFFLE(2, 3, 0, 1))                      \
+        : (-(c) == 24)                                                         \
+              ? _mm_shuffle_epi8((x), r24)                                     \
+              : (-(c) == 16)                                                   \
+                    ? _mm_shuffle_epi8((x), r16)                               \
+                    : (-(c) == 63)                                             \
+                          ? _mm_xor_si128(_mm_srli_epi64((x), -(c)),           \
+                                          _mm_add_epi64((x), (x)))             \
+                          : _mm_xor_si128(_mm_srli_epi64((x), -(c)),           \
+                                          _mm_slli_epi64((x), 64 - (-(c))))
+#else /* defined(__SSE2__) */
+#define _mm_roti_epi64(r, c)                                                   \
+    _mm_xor_si128(_mm_srli_epi64((r), -(c)), _mm_slli_epi64((r), 64 - (-(c))))
+#endif
+#else
+#endif
+
+static BLAKE2_INLINE __m128i fBlaMka(__m128i x, __m128i y) {
+    const __m128i z = _mm_mul_epu32(x, y);
+    return _mm_add_epi64(_mm_add_epi64(x, y), _mm_add_epi64(z, z));
+}
+
+#define G1(A0, B0, C0, D0, A1, B1, C1, D1)                                     \
+    do {                                                                       \
+        A0 = fBlaMka(A0, B0);                                                  \
+        A1 = fBlaMka(A1, B1);                                                  \
+                                                                               \
+        D0 = _mm_xor_si128(D0, A0);                                            \
+        D1 = _mm_xor_si128(D1, A1);                                            \
+                                                                               \
+        D0 = _mm_roti_epi64(D0, -32);                                          \
+        D1 = _mm_roti_epi64(D1, -32);                                          \
+                                                                               \
+        C0 = fBlaMka(C0, D0);                                                  \
+        C1 = fBlaMka(C1, D1);                                                  \
+                                                                               \
+        B0 = _mm_xor_si128(B0, C0);                                            \
+        B1 = _mm_xor_si128(B1, C1);                                            \
+                                                                               \
+        B0 = _mm_roti_epi64(B0, -24);                                          \
+        B1 = _mm_roti_epi64(B1, -24);                                          \
+    } while ((void)0, 0)
+
+#define G2(A0, B0, C0, D0, A1, B1, C1, D1)                                     \
+    do {                                                                       \
+        A0 = fBlaMka(A0, B0);                                                  \
+        A1 = fBlaMka(A1, B1);                                                  \
+                                                                               \
+        D0 = _mm_xor_si128(D0, A0);                                            \
+        D1 = _mm_xor_si128(D1, A1);                                            \
+                                                                               \
+        D0 = _mm_roti_epi64(D0, -16);                                          \
+        D1 = _mm_roti_epi64(D1, -16);                                          \
+                                                                               \
+        C0 = fBlaMka(C0, D0);                                                  \
+        C1 = fBlaMka(C1, D1);                                                  \
+                                                                               \
+        B0 = _mm_xor_si128(B0, C0);                                            \
+        B1 = _mm_xor_si128(B1, C1);                                            \
+                                                                               \
+        B0 = _mm_roti_epi64(B0, -63);                                          \
+        B1 = _mm_roti_epi64(B1, -63);                                          \
+    } while ((void)0, 0)
+
+#if defined(__SSSE3__)
+#define DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1)                            \
+    do {                                                                       \
+        __m128i t0 = _mm_alignr_epi8(B1, B0, 8);                               \
+        __m128i t1 = _mm_alignr_epi8(B0, B1, 8);                               \
+        B0 = t0;                                                               \
+        B1 = t1;                                                               \
+                                                                               \
+        t0 = C0;                                                               \
+        C0 = C1;                                                               \
+        C1 = t0;                                                               \
+                                                                               \
+        t0 = _mm_alignr_epi8(D1, D0, 8);                                       \
+        t1 = _mm_alignr_epi8(D0, D1, 8);                                       \
+        D0 = t1;                                                               \
+        D1 = t0;                                                               \
+    } while ((void)0, 0)
+
+#define UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1)                          \
+    do {                                                                       \
+        __m128i t0 = _mm_alignr_epi8(B0, B1, 8);                               \
+        __m128i t1 = _mm_alignr_epi8(B1, B0, 8);                               \
+        B0 = t0;                                                               \
+        B1 = t1;                                                               \
+                                                                               \
+        t0 = C0;                                                               \
+        C0 = C1;                                                               \
+        C1 = t0;                                                               \
+                                                                               \
+        t0 = _mm_alignr_epi8(D0, D1, 8);                                       \
+        t1 = _mm_alignr_epi8(D1, D0, 8);                                       \
+        D0 = t1;                                                               \
+        D1 = t0;                                                               \
+    } while ((void)0, 0)
+#else /* SSE2 */
+#define DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1)                            \
+    do {                                                                       \
+        __m128i t0 = D0;                                                       \
+        __m128i t1 = B0;                                                       \
+        D0 = C0;                                                               \
+        C0 = C1;                                                               \
+        C1 = D0;                                                               \
+        D0 = _mm_unpackhi_epi64(D1, _mm_unpacklo_epi64(t0, t0));               \
+        D1 = _mm_unpackhi_epi64(t0, _mm_unpacklo_epi64(D1, D1));               \
+        B0 = _mm_unpackhi_epi64(B0, _mm_unpacklo_epi64(B1, B1));               \
+        B1 = _mm_unpackhi_epi64(B1, _mm_unpacklo_epi64(t1, t1));               \
+    } while ((void)0, 0)
+
+#define UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1)                          \
+    do {                                                                       \
+        __m128i t0, t1;                                                        \
+        t0 = C0;                                                               \
+        C0 = C1;                                                               \
+        C1 = t0;                                                               \
+        t0 = B0;                                                               \
+        t1 = D0;                                                               \
+        B0 = _mm_unpackhi_epi64(B1, _mm_unpacklo_epi64(B0, B0));               \
+        B1 = _mm_unpackhi_epi64(t0, _mm_unpacklo_epi64(B1, B1));               \
+        D0 = _mm_unpackhi_epi64(D0, _mm_unpacklo_epi64(D1, D1));               \
+        D1 = _mm_unpackhi_epi64(D1, _mm_unpacklo_epi64(t1, t1));               \
+    } while ((void)0, 0)
+#endif
+
+#define BLAKE2_ROUND(A0, A1, B0, B1, C0, C1, D0, D1)                           \
+    do {                                                                       \
+        G1(A0, B0, C0, D0, A1, B1, C1, D1);                                    \
+        G2(A0, B0, C0, D0, A1, B1, C1, D1);                                    \
+                                                                               \
+        DIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1);                           \
+                                                                               \
+        G1(A0, B0, C0, D0, A1, B1, C1, D1);                                    \
+        G2(A0, B0, C0, D0, A1, B1, C1, D1);                                    \
+                                                                               \
+        UNDIAGONALIZE(A0, B0, C0, D0, A1, B1, C1, D1);                         \
+    } while ((void)0, 0)
+#else /* __AVX2__ */
+
+#include <immintrin.h>
+
+#define rotr32(x)   _mm256_shuffle_epi32(x, _MM_SHUFFLE(2, 3, 0, 1))
+#define rotr24(x)   _mm256_shuffle_epi8(x, _mm256_setr_epi8(3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10, 3, 4, 5, 6, 7, 0, 1, 2, 11, 12, 13, 14, 15, 8, 9, 10))
+#define rotr16(x)   _mm256_shuffle_epi8(x, _mm256_setr_epi8(2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9, 2, 3, 4, 5, 6, 7, 0, 1, 10, 11, 12, 13, 14, 15, 8, 9))
+#define rotr63(x)   _mm256_xor_si256(_mm256_srli_epi64((x), 63), _mm256_add_epi64((x), (x)))
+
+#define G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+    do { \
+        __m256i ml = _mm256_mul_epu32(A0, B0); \
+        ml = _mm256_add_epi64(ml, ml); \
+        A0 = _mm256_add_epi64(A0, _mm256_add_epi64(B0, ml)); \
+        D0 = _mm256_xor_si256(D0, A0); \
+        D0 = rotr32(D0); \
+        \
+        ml = _mm256_mul_epu32(C0, D0); \
+        ml = _mm256_add_epi64(ml, ml); \
+        C0 = _mm256_add_epi64(C0, _mm256_add_epi64(D0, ml)); \
+        \
+        B0 = _mm256_xor_si256(B0, C0); \
+        B0 = rotr24(B0); \
+        \
+        ml = _mm256_mul_epu32(A1, B1); \
+        ml = _mm256_add_epi64(ml, ml); \
+        A1 = _mm256_add_epi64(A1, _mm256_add_epi64(B1, ml)); \
+        D1 = _mm256_xor_si256(D1, A1); \
+        D1 = rotr32(D1); \
+        \
+        ml = _mm256_mul_epu32(C1, D1); \
+        ml = _mm256_add_epi64(ml, ml); \
+        C1 = _mm256_add_epi64(C1, _mm256_add_epi64(D1, ml)); \
+        \
+        B1 = _mm256_xor_si256(B1, C1); \
+        B1 = rotr24(B1); \
+    } while((void)0, 0);
+
+#define G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+    do { \
+        __m256i ml = _mm256_mul_epu32(A0, B0); \
+        ml = _mm256_add_epi64(ml, ml); \
+        A0 = _mm256_add_epi64(A0, _mm256_add_epi64(B0, ml)); \
+        D0 = _mm256_xor_si256(D0, A0); \
+        D0 = rotr16(D0); \
+        \
+        ml = _mm256_mul_epu32(C0, D0); \
+        ml = _mm256_add_epi64(ml, ml); \
+        C0 = _mm256_add_epi64(C0, _mm256_add_epi64(D0, ml)); \
+        B0 = _mm256_xor_si256(B0, C0); \
+        B0 = rotr63(B0); \
+        \
+        ml = _mm256_mul_epu32(A1, B1); \
+        ml = _mm256_add_epi64(ml, ml); \
+        A1 = _mm256_add_epi64(A1, _mm256_add_epi64(B1, ml)); \
+        D1 = _mm256_xor_si256(D1, A1); \
+        D1 = rotr16(D1); \
+        \
+        ml = _mm256_mul_epu32(C1, D1); \
+        ml = _mm256_add_epi64(ml, ml); \
+        C1 = _mm256_add_epi64(C1, _mm256_add_epi64(D1, ml)); \
+        B1 = _mm256_xor_si256(B1, C1); \
+        B1 = rotr63(B1); \
+    } while((void)0, 0);
+
+#define DIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
+    do { \
+        B0 = _mm256_permute4x64_epi64(B0, _MM_SHUFFLE(0, 3, 2, 1)); \
+        C0 = _mm256_permute4x64_epi64(C0, _MM_SHUFFLE(1, 0, 3, 2)); \
+        D0 = _mm256_permute4x64_epi64(D0, _MM_SHUFFLE(2, 1, 0, 3)); \
+        \
+        B1 = _mm256_permute4x64_epi64(B1, _MM_SHUFFLE(0, 3, 2, 1)); \
+        C1 = _mm256_permute4x64_epi64(C1, _MM_SHUFFLE(1, 0, 3, 2)); \
+        D1 = _mm256_permute4x64_epi64(D1, _MM_SHUFFLE(2, 1, 0, 3)); \
+    } while((void)0, 0);
+
+#define DIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
+    do { \
+        __m256i tmp1 = _mm256_blend_epi32(B0, B1, 0xCC); \
+        __m256i tmp2 = _mm256_blend_epi32(B0, B1, 0x33); \
+        B1 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
+        B0 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
+        \
+        tmp1 = C0; \
+        C0 = C1; \
+        C1 = tmp1; \
+        \
+        tmp1 = _mm256_blend_epi32(D0, D1, 0xCC); \
+        tmp2 = _mm256_blend_epi32(D0, D1, 0x33); \
+        D0 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
+        D1 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
+    } while(0);
+
+#define UNDIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
+    do { \
+        B0 = _mm256_permute4x64_epi64(B0, _MM_SHUFFLE(2, 1, 0, 3)); \
+        C0 = _mm256_permute4x64_epi64(C0, _MM_SHUFFLE(1, 0, 3, 2)); \
+        D0 = _mm256_permute4x64_epi64(D0, _MM_SHUFFLE(0, 3, 2, 1)); \
+        \
+        B1 = _mm256_permute4x64_epi64(B1, _MM_SHUFFLE(2, 1, 0, 3)); \
+        C1 = _mm256_permute4x64_epi64(C1, _MM_SHUFFLE(1, 0, 3, 2)); \
+        D1 = _mm256_permute4x64_epi64(D1, _MM_SHUFFLE(0, 3, 2, 1)); \
+    } while((void)0, 0);
+
+#define UNDIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
+    do { \
+        __m256i tmp1 = _mm256_blend_epi32(B0, B1, 0xCC); \
+        __m256i tmp2 = _mm256_blend_epi32(B0, B1, 0x33); \
+        B0 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
+        B1 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
+        \
+        tmp1 = C0; \
+        C0 = C1; \
+        C1 = tmp1; \
+        \
+        tmp1 = _mm256_blend_epi32(D0, D1, 0x33); \
+        tmp2 = _mm256_blend_epi32(D0, D1, 0xCC); \
+        D0 = _mm256_permute4x64_epi64(tmp1, _MM_SHUFFLE(2,3,0,1)); \
+        D1 = _mm256_permute4x64_epi64(tmp2, _MM_SHUFFLE(2,3,0,1)); \
+    } while((void)0, 0);
+
+#define BLAKE2_ROUND_1(A0, A1, B0, B1, C0, C1, D0, D1) \
+    do{ \
+        G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        \
+        DIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
+        \
+        G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        \
+        UNDIAGONALIZE_1(A0, B0, C0, D0, A1, B1, C1, D1) \
+    } while((void)0, 0);
+
+#define BLAKE2_ROUND_2(A0, A1, B0, B1, C0, C1, D0, D1) \
+    do{ \
+        G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        \
+        DIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        \
+        G1_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        G2_AVX2(A0, A1, B0, B1, C0, C1, D0, D1) \
+        \
+        UNDIAGONALIZE_2(A0, A1, B0, B1, C0, C1, D0, D1) \
+    } while((void)0, 0);
+
+#endif /* __AVX2__ */
+#endif /* BLAKE_ROUND_MKA_OPT_H */

data/ext/phc-winner-argon2/src/blake2/blamka-round-ref.h

@@ -0,0 +1,56 @@
+/*
+ * Argon2 reference source code package - reference C implementations
+ *
+ * Copyright 2015
+ * Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
+ *
+ * You may use this work under the terms of a Creative Commons CC0 1.0 
+ * License/Waiver or the Apache Public License 2.0, at your option. The terms of
+ * these licenses can be found at:
+ *
+ * - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
+ * - Apache 2.0        : http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * You should have received a copy of both of these licenses along with this
+ * software. If not, they may be obtained at the above URLs.
+ */
+
+#ifndef BLAKE_ROUND_MKA_H
+#define BLAKE_ROUND_MKA_H
+
+#include "blake2.h"
+#include "blake2-impl.h"
+
+/*designed by the Lyra PHC team */
+static BLAKE2_INLINE uint64_t fBlaMka(uint64_t x, uint64_t y) {
+    const uint64_t m = UINT64_C(0xFFFFFFFF);
+    const uint64_t xy = (x & m) * (y & m);
+    return x + y + 2 * xy;
+}
+
+#define G(a, b, c, d)                                                          \
+    do {                                                                       \
+        a = fBlaMka(a, b);                                                     \
+        d = rotr64(d ^ a, 32);                                                 \
+        c = fBlaMka(c, d);                                                     \
+        b = rotr64(b ^ c, 24);                                                 \
+        a = fBlaMka(a, b);                                                     \
+        d = rotr64(d ^ a, 16);                                                 \
+        c = fBlaMka(c, d);                                                     \
+        b = rotr64(b ^ c, 63);                                                 \
+    } while ((void)0, 0)
+
+#define BLAKE2_ROUND_NOMSG(v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11,   \
+                           v12, v13, v14, v15)                                 \
+    do {                                                                       \
+        G(v0, v4, v8, v12);                                                    \
+        G(v1, v5, v9, v13);                                                    \
+        G(v2, v6, v10, v14);                                                   \
+        G(v3, v7, v11, v15);                                                   \
+        G(v0, v5, v10, v15);                                                   \
+        G(v1, v6, v11, v12);                                                   \
+        G(v2, v7, v8, v13);                                                    \
+        G(v3, v4, v9, v14);                                                    \
+    } while ((void)0, 0)
+
+#endif

data/ext/phc-winner-argon2/src/core.c

@@ -0,0 +1,635 @@
+/*
+ * Argon2 reference source code package - reference C implementations
+ *
+ * Copyright 2015
+ * Daniel Dinu, Dmitry Khovratovich, Jean-Philippe Aumasson, and Samuel Neves
+ *
+ * You may use this work under the terms of a Creative Commons CC0 1.0 
+ * License/Waiver or the Apache Public License 2.0, at your option. The terms of
+ * these licenses can be found at:
+ *
+ * - CC0 1.0 Universal : http://creativecommons.org/publicdomain/zero/1.0
+ * - Apache 2.0        : http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * You should have received a copy of both of these licenses along with this
+ * software. If not, they may be obtained at the above URLs.
+ */
+
+/*For memory wiping*/
+#ifdef _MSC_VER
+#include <windows.h>
+#include <winbase.h> /* For SecureZeroMemory */
+#endif
+#if defined __STDC_LIB_EXT1__
+#define __STDC_WANT_LIB_EXT1__ 1
+#endif
+#define VC_GE_2005(version) (version >= 1400)
+
+#include <inttypes.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "core.h"
+#include "thread.h"
+#include "blake2/blake2.h"
+#include "blake2/blake2-impl.h"
+
+#ifdef GENKAT
+#include "genkat.h"
+#endif
+
+#if defined(__clang__)
+#if __has_attribute(optnone)
+#define NOT_OPTIMIZED __attribute__((optnone))
+#endif
+#elif defined(__GNUC__)
+#define GCC_VERSION                                                            \
+    (__GNUC__ * 10000 + __GNUC_MINOR__ * 100 + __GNUC_PATCHLEVEL__)
+#if GCC_VERSION >= 40400
+#define NOT_OPTIMIZED __attribute__((optimize("O0")))
+#endif
+#endif
+#ifndef NOT_OPTIMIZED
+#define NOT_OPTIMIZED
+#endif
+
+/***************Instance and Position constructors**********/
+void init_block_value(block *b, uint8_t in) { memset(b->v, in, sizeof(b->v)); }
+
+void copy_block(block *dst, const block *src) {
+    memcpy(dst->v, src->v, sizeof(uint64_t) * ARGON2_QWORDS_IN_BLOCK);
+}
+
+void xor_block(block *dst, const block *src) {
+    int i;
+    for (i = 0; i < ARGON2_QWORDS_IN_BLOCK; ++i) {
+        dst->v[i] ^= src->v[i];
+    }
+}
+
+static void load_block(block *dst, const void *input) {
+    unsigned i;
+    for (i = 0; i < ARGON2_QWORDS_IN_BLOCK; ++i) {
+        dst->v[i] = load64((const uint8_t *)input + i * sizeof(dst->v[i]));
+    }
+}
+
+static void store_block(void *output, const block *src) {
+    unsigned i;
+    for (i = 0; i < ARGON2_QWORDS_IN_BLOCK; ++i) {
+        store64((uint8_t *)output + i * sizeof(src->v[i]), src->v[i]);
+    }
+}
+
+/***************Memory functions*****************/
+
+int allocate_memory(const argon2_context *context, uint8_t **memory,
+                    size_t num, size_t size) {
+    size_t memory_size = num*size;
+    if (memory == NULL) {
+        return ARGON2_MEMORY_ALLOCATION_ERROR;
+    }
+
+    /* 1. Check for multiplication overflow */
+    if (size != 0 && memory_size / size != num) {
+        return ARGON2_MEMORY_ALLOCATION_ERROR;
+    }
+
+    /* 2. Try to allocate with appropriate allocator */
+    if (context->allocate_cbk) {
+        (context->allocate_cbk)(memory, memory_size);
+    } else {
+        *memory = malloc(memory_size);
+    }
+
+    if (*memory == NULL) {
+        return ARGON2_MEMORY_ALLOCATION_ERROR;
+    }
+
+    return ARGON2_OK;
+}
+
+void free_memory(const argon2_context *context, uint8_t *memory,
+                 size_t num, size_t size) {
+    size_t memory_size = num*size;
+    clear_internal_memory(memory, memory_size);
+    if (context->free_cbk) {
+        (context->free_cbk)(memory, memory_size);
+    } else {
+        free(memory);
+    }
+}
+
+void NOT_OPTIMIZED secure_wipe_memory(void *v, size_t n) {
+#if defined(_MSC_VER) && VC_GE_2005(_MSC_VER)
+    SecureZeroMemory(v, n);
+#elif defined memset_s
+    memset_s(v, n, 0, n);
+#elif defined(__OpenBSD__)
+    explicit_bzero(v, n);
+#else
+    static void *(*const volatile memset_sec)(void *, int, size_t) = &memset;
+    memset_sec(v, 0, n);
+#endif
+}
+
+/* Memory clear flag defaults to true. */
+int FLAG_clear_internal_memory = 1;
+void clear_internal_memory(void *v, size_t n) {
+  if (FLAG_clear_internal_memory && v) {
+    secure_wipe_memory(v, n);
+  }
+}
+
+void finalize(const argon2_context *context, argon2_instance_t *instance) {
+    if (context != NULL && instance != NULL) {
+        block blockhash;
+        uint32_t l;
+
+        copy_block(&blockhash, instance->memory + instance->lane_length - 1);
+
+        /* XOR the last blocks */
+        for (l = 1; l < instance->lanes; ++l) {
+            uint32_t last_block_in_lane =
+                l * instance->lane_length + (instance->lane_length - 1);
+            xor_block(&blockhash, instance->memory + last_block_in_lane);
+        }
+
+        /* Hash the result */
+        {
+            uint8_t blockhash_bytes[ARGON2_BLOCK_SIZE];
+            store_block(blockhash_bytes, &blockhash);
+            blake2b_long(context->out, context->outlen, blockhash_bytes,
+                         ARGON2_BLOCK_SIZE);
+            /* clear blockhash and blockhash_bytes */
+            clear_internal_memory(blockhash.v, ARGON2_BLOCK_SIZE);
+            clear_internal_memory(blockhash_bytes, ARGON2_BLOCK_SIZE);
+        }
+
+#ifdef GENKAT
+        print_tag(context->out, context->outlen);
+#endif
+
+        free_memory(context, (uint8_t *)instance->memory,
+                    instance->memory_blocks, sizeof(block));
+    }
+}
+
+uint32_t index_alpha(const argon2_instance_t *instance,
+                     const argon2_position_t *position, uint32_t pseudo_rand,
+                     int same_lane) {
+    /*
+     * Pass 0:
+     *      This lane : all already finished segments plus already constructed
+     * blocks in this segment
+     *      Other lanes : all already finished segments
+     * Pass 1+:
+     *      This lane : (SYNC_POINTS - 1) last segments plus already constructed
+     * blocks in this segment
+     *      Other lanes : (SYNC_POINTS - 1) last segments
+     */
+    uint32_t reference_area_size;
+    uint64_t relative_position;
+    uint32_t start_position, absolute_position;
+
+    if (0 == position->pass) {
+        /* First pass */
+        if (0 == position->slice) {
+            /* First slice */
+            reference_area_size =
+                position->index - 1; /* all but the previous */
+        } else {
+            if (same_lane) {
+                /* The same lane => add current segment */
+                reference_area_size =
+                    position->slice * instance->segment_length +
+                    position->index - 1;
+            } else {
+                reference_area_size =
+                    position->slice * instance->segment_length +
+                    ((position->index == 0) ? (-1) : 0);
+            }
+        }
+    } else {
+        /* Second pass */
+        if (same_lane) {
+            reference_area_size = instance->lane_length -
+                                  instance->segment_length + position->index -
+                                  1;
+        } else {
+            reference_area_size = instance->lane_length -
+                                  instance->segment_length +
+                                  ((position->index == 0) ? (-1) : 0);
+        }
+    }
+
+    /* 1.2.4. Mapping pseudo_rand to 0..<reference_area_size-1> and produce
+     * relative position */
+    relative_position = pseudo_rand;
+    relative_position = relative_position * relative_position >> 32;
+    relative_position = reference_area_size - 1 -
+                        (reference_area_size * relative_position >> 32);
+
+    /* 1.2.5 Computing starting position */
+    start_position = 0;
+
+    if (0 != position->pass) {
+        start_position = (position->slice == ARGON2_SYNC_POINTS - 1)
+                             ? 0
+                             : (position->slice + 1) * instance->segment_length;
+    }
+
+    /* 1.2.6. Computing absolute position */
+    absolute_position = (start_position + relative_position) %
+                        instance->lane_length; /* absolute position */
+    return absolute_position;
+}
+
+/* Single-threaded version for p=1 case */
+static int fill_memory_blocks_st(argon2_instance_t *instance) {
+    uint32_t r, s, l;
+
+    for (r = 0; r < instance->passes; ++r) {
+        for (s = 0; s < ARGON2_SYNC_POINTS; ++s) {
+            for (l = 0; l < instance->lanes; ++l) {
+                argon2_position_t position = {r, l, (uint8_t)s, 0};
+                fill_segment(instance, position);
+            }
+        }
+#ifdef GENKAT
+        internal_kat(instance, r); /* Print all memory blocks */
+#endif
+    }
+    return ARGON2_OK;
+}
+
+#if !defined(ARGON2_NO_THREADS)
+
+#ifdef _WIN32
+static unsigned __stdcall fill_segment_thr(void *thread_data)
+#else
+static void *fill_segment_thr(void *thread_data)
+#endif
+{
+    argon2_thread_data *my_data = thread_data;
+    fill_segment(my_data->instance_ptr, my_data->pos);
+    argon2_thread_exit();
+    return 0;
+}
+
+/* Multi-threaded version for p > 1 case */
+static int fill_memory_blocks_mt(argon2_instance_t *instance) {
+    uint32_t r, s;
+    argon2_thread_handle_t *thread = NULL;
+    argon2_thread_data *thr_data = NULL;
+    int rc = ARGON2_OK;
+
+    /* 1. Allocating space for threads */
+    thread = calloc(instance->lanes, sizeof(argon2_thread_handle_t));
+    if (thread == NULL) {
+        rc = ARGON2_MEMORY_ALLOCATION_ERROR;
+        goto fail;
+    }
+
+    thr_data = calloc(instance->lanes, sizeof(argon2_thread_data));
+    if (thr_data == NULL) {
+        rc = ARGON2_MEMORY_ALLOCATION_ERROR;
+        goto fail;
+    }
+
+    for (r = 0; r < instance->passes; ++r) {
+        for (s = 0; s < ARGON2_SYNC_POINTS; ++s) {
+            uint32_t l;
+
+            /* 2. Calling threads */
+            for (l = 0; l < instance->lanes; ++l) {
+                argon2_position_t position;
+
+                /* 2.1 Join a thread if limit is exceeded */
+                if (l >= instance->threads) {
+                    if (argon2_thread_join(thread[l - instance->threads])) {
+                        rc = ARGON2_THREAD_FAIL;
+                        goto fail;
+                    }
+                }
+
+                /* 2.2 Create thread */
+                position.pass = r;
+                position.lane = l;
+                position.slice = (uint8_t)s;
+                position.index = 0;
+                thr_data[l].instance_ptr =
+                    instance; /* preparing the thread input */
+                memcpy(&(thr_data[l].pos), &position,
+                       sizeof(argon2_position_t));
+                if (argon2_thread_create(&thread[l], &fill_segment_thr,
+                                         (void *)&thr_data[l])) {
+                    rc = ARGON2_THREAD_FAIL;
+                    goto fail;
+                }
+
+                /* fill_segment(instance, position); */
+                /*Non-thread equivalent of the lines above */
+            }
+
+            /* 3. Joining remaining threads */
+            for (l = instance->lanes - instance->threads; l < instance->lanes;
+                 ++l) {
+                if (argon2_thread_join(thread[l])) {
+                    rc = ARGON2_THREAD_FAIL;
+                    goto fail;
+                }
+            }
+        }
+
+#ifdef GENKAT
+        internal_kat(instance, r); /* Print all memory blocks */
+#endif
+    }
+
+fail:
+    if (thread != NULL) {
+        free(thread);
+    }
+    if (thr_data != NULL) {
+        free(thr_data);
+    }
+    return rc;
+}
+
+#endif /* ARGON2_NO_THREADS */
+
+int fill_memory_blocks(argon2_instance_t *instance) {
+	if (instance == NULL || instance->lanes == 0) {
+	    return ARGON2_INCORRECT_PARAMETER;
+    }
+#if defined(ARGON2_NO_THREADS)
+    return fill_memory_blocks_st(instance);
+#else
+    return instance->threads == 1 ?
+			fill_memory_blocks_st(instance) : fill_memory_blocks_mt(instance);
+#endif
+}
+
+int validate_inputs(const argon2_context *context) {
+    if (NULL == context) {
+        return ARGON2_INCORRECT_PARAMETER;
+    }
+
+    if (NULL == context->out) {
+        return ARGON2_OUTPUT_PTR_NULL;
+    }
+
+    /* Validate output length */
+    if (ARGON2_MIN_OUTLEN > context->outlen) {
+        return ARGON2_OUTPUT_TOO_SHORT;
+    }
+
+    if (ARGON2_MAX_OUTLEN < context->outlen) {
+        return ARGON2_OUTPUT_TOO_LONG;
+    }
+
+    /* Validate password (required param) */
+    if (NULL == context->pwd) {
+        if (0 != context->pwdlen) {
+            return ARGON2_PWD_PTR_MISMATCH;
+        }
+    }
+
+    if (ARGON2_MIN_PWD_LENGTH > context->pwdlen) {
+      return ARGON2_PWD_TOO_SHORT;
+    }
+
+    if (ARGON2_MAX_PWD_LENGTH < context->pwdlen) {
+        return ARGON2_PWD_TOO_LONG;
+    }
+
+    /* Validate salt (required param) */
+    if (NULL == context->salt) {
+        if (0 != context->saltlen) {
+            return ARGON2_SALT_PTR_MISMATCH;
+        }
+    }
+
+    if (ARGON2_MIN_SALT_LENGTH > context->saltlen) {
+        return ARGON2_SALT_TOO_SHORT;
+    }
+
+    if (ARGON2_MAX_SALT_LENGTH < context->saltlen) {
+        return ARGON2_SALT_TOO_LONG;
+    }
+
+    /* Validate secret (optional param) */
+    if (NULL == context->secret) {
+        if (0 != context->secretlen) {
+            return ARGON2_SECRET_PTR_MISMATCH;
+        }
+    } else {
+        if (ARGON2_MIN_SECRET > context->secretlen) {
+            return ARGON2_SECRET_TOO_SHORT;
+        }
+        if (ARGON2_MAX_SECRET < context->secretlen) {
+            return ARGON2_SECRET_TOO_LONG;
+        }
+    }
+
+    /* Validate associated data (optional param) */
+    if (NULL == context->ad) {
+        if (0 != context->adlen) {
+            return ARGON2_AD_PTR_MISMATCH;
+        }
+    } else {
+        if (ARGON2_MIN_AD_LENGTH > context->adlen) {
+            return ARGON2_AD_TOO_SHORT;
+        }
+        if (ARGON2_MAX_AD_LENGTH < context->adlen) {
+            return ARGON2_AD_TOO_LONG;
+        }
+    }
+
+    /* Validate memory cost */
+    if (ARGON2_MIN_MEMORY > context->m_cost) {
+        return ARGON2_MEMORY_TOO_LITTLE;
+    }
+
+    if (ARGON2_MAX_MEMORY < context->m_cost) {
+        return ARGON2_MEMORY_TOO_MUCH;
+    }
+
+    if (context->m_cost < 8 * context->lanes) {
+        return ARGON2_MEMORY_TOO_LITTLE;
+    }
+
+    /* Validate time cost */
+    if (ARGON2_MIN_TIME > context->t_cost) {
+        return ARGON2_TIME_TOO_SMALL;
+    }
+
+    if (ARGON2_MAX_TIME < context->t_cost) {
+        return ARGON2_TIME_TOO_LARGE;
+    }
+
+    /* Validate lanes */
+    if (ARGON2_MIN_LANES > context->lanes) {
+        return ARGON2_LANES_TOO_FEW;
+    }
+
+    if (ARGON2_MAX_LANES < context->lanes) {
+        return ARGON2_LANES_TOO_MANY;
+    }
+
+    /* Validate threads */
+    if (ARGON2_MIN_THREADS > context->threads) {
+        return ARGON2_THREADS_TOO_FEW;
+    }
+
+    if (ARGON2_MAX_THREADS < context->threads) {
+        return ARGON2_THREADS_TOO_MANY;
+    }
+
+    if (NULL != context->allocate_cbk && NULL == context->free_cbk) {
+        return ARGON2_FREE_MEMORY_CBK_NULL;
+    }
+
+    if (NULL == context->allocate_cbk && NULL != context->free_cbk) {
+        return ARGON2_ALLOCATE_MEMORY_CBK_NULL;
+    }
+
+    return ARGON2_OK;
+}
+
+void fill_first_blocks(uint8_t *blockhash, const argon2_instance_t *instance) {
+    uint32_t l;
+    /* Make the first and second block in each lane as G(H0||0||i) or
+       G(H0||1||i) */
+    uint8_t blockhash_bytes[ARGON2_BLOCK_SIZE];
+    for (l = 0; l < instance->lanes; ++l) {
+
+        store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH, 0);
+        store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH + 4, l);
+        blake2b_long(blockhash_bytes, ARGON2_BLOCK_SIZE, blockhash,
+                     ARGON2_PREHASH_SEED_LENGTH);
+        load_block(&instance->memory[l * instance->lane_length + 0],
+                   blockhash_bytes);
+
+        store32(blockhash + ARGON2_PREHASH_DIGEST_LENGTH, 1);
+        blake2b_long(blockhash_bytes, ARGON2_BLOCK_SIZE, blockhash,
+                     ARGON2_PREHASH_SEED_LENGTH);
+        load_block(&instance->memory[l * instance->lane_length + 1],
+                   blockhash_bytes);
+    }
+    clear_internal_memory(blockhash_bytes, ARGON2_BLOCK_SIZE);
+}
+
+void initial_hash(uint8_t *blockhash, argon2_context *context,
+                  argon2_type type) {
+    blake2b_state BlakeHash;
+    uint8_t value[sizeof(uint32_t)];
+
+    if (NULL == context || NULL == blockhash) {
+        return;
+    }
+
+    blake2b_init(&BlakeHash, ARGON2_PREHASH_DIGEST_LENGTH);
+
+    store32(&value, context->lanes);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    store32(&value, context->outlen);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    store32(&value, context->m_cost);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    store32(&value, context->t_cost);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    store32(&value, context->version);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    store32(&value, (uint32_t)type);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    store32(&value, context->pwdlen);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    if (context->pwd != NULL) {
+        blake2b_update(&BlakeHash, (const uint8_t *)context->pwd,
+                       context->pwdlen);
+
+        if (context->flags & ARGON2_FLAG_CLEAR_PASSWORD) {
+            secure_wipe_memory(context->pwd, context->pwdlen);
+            context->pwdlen = 0;
+        }
+    }
+
+    store32(&value, context->saltlen);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    if (context->salt != NULL) {
+        blake2b_update(&BlakeHash, (const uint8_t *)context->salt,
+                       context->saltlen);
+    }
+
+    store32(&value, context->secretlen);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    if (context->secret != NULL) {
+        blake2b_update(&BlakeHash, (const uint8_t *)context->secret,
+                       context->secretlen);
+
+        if (context->flags & ARGON2_FLAG_CLEAR_SECRET) {
+            secure_wipe_memory(context->secret, context->secretlen);
+            context->secretlen = 0;
+        }
+    }
+
+    store32(&value, context->adlen);
+    blake2b_update(&BlakeHash, (const uint8_t *)&value, sizeof(value));
+
+    if (context->ad != NULL) {
+        blake2b_update(&BlakeHash, (const uint8_t *)context->ad,
+                       context->adlen);
+    }
+
+    blake2b_final(&BlakeHash, blockhash, ARGON2_PREHASH_DIGEST_LENGTH);
+}
+
+int initialize(argon2_instance_t *instance, argon2_context *context) {
+    uint8_t blockhash[ARGON2_PREHASH_SEED_LENGTH];
+    int result = ARGON2_OK;
+
+    if (instance == NULL || context == NULL)
+        return ARGON2_INCORRECT_PARAMETER;
+    instance->context_ptr = context;
+
+    /* 1. Memory allocation */
+    result = allocate_memory(context, (uint8_t **)&(instance->memory),
+                             instance->memory_blocks, sizeof(block));
+    if (result != ARGON2_OK) {
+        return result;
+    }
+
+    /* 2. Initial hashing */
+    /* H_0 + 8 extra bytes to produce the first blocks */
+    /* uint8_t blockhash[ARGON2_PREHASH_SEED_LENGTH]; */
+    /* Hashing all inputs */
+    initial_hash(blockhash, context, instance->type);
+    /* Zeroing 8 extra bytes */
+    clear_internal_memory(blockhash + ARGON2_PREHASH_DIGEST_LENGTH,
+                          ARGON2_PREHASH_SEED_LENGTH -
+                              ARGON2_PREHASH_DIGEST_LENGTH);
+
+#ifdef GENKAT
+    initial_kat(blockhash, context, instance->type);
+#endif
+
+    /* 3. Creating first blocks, we always have at least two blocks in a slice
+     */
+    fill_first_blocks(blockhash, instance);
+    /* Clearing the hash */
+    clear_internal_memory(blockhash, ARGON2_PREHASH_SEED_LENGTH);
+
+    return ARGON2_OK;
+}