sleeping_kangaroo12 0.0.1

data/ext/xkcp/lib/low/Xoodoo-times8/AVX512/Xoodoo-times8-SnP.h

@@ -0,0 +1,48 @@
+/*
+The eXtended Keccak Code Package (XKCP)
+https://github.com/XKCP/XKCP
+
+The Xoodoo permutation, designed by Joan Daemen, Seth Hoffert, Gilles Van Assche and Ronny Van Keer.
+
+Implementation by Ronny Van Keer, hereby denoted as "the implementer".
+
+For more information, feedback or questions, please refer to the Keccak Team website:
+https://keccak.team/
+
+To the extent possible under law, the implementer has waived all copyright
+and related or neighboring rights to the source code in this file.
+http://creativecommons.org/publicdomain/zero/1.0/
+*/
+
+#ifndef _Xoodoo_times8_SnP_h_
+#define _Xoodoo_times8_SnP_h_
+
+/** For the documentation, see PlSnP-documentation.h.
+ */
+
+#define Xoodootimes8_implementation        "512-bit SIMD (AVX-512) implementation"
+#define Xoodootimes8_statesSizeInBytes     (8*3*4*4)
+#define Xoodootimes8_statesAlignment       64
+
+#define Xoodootimes8_StaticInitialize()
+void Xoodootimes8_InitializeAll(void *states);
+#define Xoodootimes8_AddByte(states, instanceIndex, byte, offset) \
+    ((unsigned char*)(states))[(instanceIndex)*4 + ((offset)/4)*8*4 + (offset)%4] ^= (byte)
+void Xoodootimes8_AddBytes(void *states, unsigned int instanceIndex, const unsigned char *data, unsigned int offset, unsigned int length);
+void Xoodootimes8_AddLanesAll(void *states, const unsigned char *data, unsigned int laneCount, unsigned int laneOffset);
+void Xoodootimes8_OverwriteBytes(void *states, unsigned int instanceIndex, const unsigned char *data, unsigned int offset, unsigned int length);
+void Xoodootimes8_OverwriteLanesAll(void *states, const unsigned char *data, unsigned int laneCount, unsigned int laneOffset);
+void Xoodootimes8_OverwriteWithZeroes(void *states, unsigned int instanceIndex, unsigned int byteCount);
+void Xoodootimes8_PermuteAll_6rounds(void *states);
+void Xoodootimes8_PermuteAll_12rounds(void *states);
+void Xoodootimes8_ExtractBytes(const void *states, unsigned int instanceIndex, unsigned char *data, unsigned int offset, unsigned int length);
+void Xoodootimes8_ExtractLanesAll(const void *states, unsigned char *data, unsigned int laneCount, unsigned int laneOffset);
+void Xoodootimes8_ExtractAndAddBytes(const void *states, unsigned int instanceIndex,  const unsigned char *input, unsigned char *output, unsigned int offset, unsigned int length);
+void Xoodootimes8_ExtractAndAddLanesAll(const void *states, const unsigned char *input, unsigned char *output, unsigned int laneCount, unsigned int laneOffset);
+
+#define Xoodootimes8_FastXoofff_supported
+void Xooffftimes8_AddIs(unsigned char *output, const unsigned char *input, size_t bitLen);
+size_t Xooffftimes8_CompressFastLoop(unsigned char *k, unsigned char *x, const unsigned char *input, size_t length);
+size_t Xooffftimes8_ExpandFastLoop(unsigned char *yAccu, const unsigned char *kRoll, unsigned char *output, size_t length);
+
+#endif

data/ext/xkcp/lib/low/Xoodoo-times8/fallback-on1/Xoodoo-times8-SnP.h

@@ -0,0 +1,45 @@
+/*
+The eXtended Keccak Code Package (XKCP)
+https://github.com/XKCP/XKCP
+
+The Xoodoo permutation, designed by Joan Daemen, Seth Hoffert, Gilles Van Assche and Ronny Van Keer.
+
+Implementation by Ronny Van Keer, hereby denoted as "the implementer".
+
+For more information, feedback or questions, please refer to the Keccak Team website:
+https://keccak.team/
+
+To the extent possible under law, the implementer has waived all copyright
+and related or neighboring rights to the source code in this file.
+http://creativecommons.org/publicdomain/zero/1.0/
+*/
+
+#ifndef _Xoodoo_times8_SnP_h_
+#define _Xoodoo_times8_SnP_h_
+
+#include "Xoodoo-SnP.h"
+
+/** For the documentation, see PlSnP-documentation.h.
+ */
+
+#define Xoodootimes8_implementation        "fallback on serial implementation (" Xoodoo_implementation ")"
+#define Xoodootimes8_statesSizeInBytes     (((Xoodoo_stateSizeInBytes+(Xoodoo_stateAlignment-1))/Xoodoo_stateAlignment)*Xoodoo_stateAlignment*8)
+#define Xoodootimes8_statesAlignment       Xoodoo_stateAlignment
+#define Xoodootimes8_isFallback
+
+void Xoodootimes8_StaticInitialize(void);
+void Xoodootimes8_InitializeAll(void *states);
+void Xoodootimes8_AddByte(void *states, unsigned int instanceIndex, unsigned char data, unsigned int offset);
+void Xoodootimes8_AddBytes(void *states, unsigned int instanceIndex, const unsigned char *data, unsigned int offset, unsigned int length);
+void Xoodootimes8_AddLanesAll(void *states, const unsigned char *data, unsigned int laneCount, unsigned int laneOffset);
+void Xoodootimes8_OverwriteBytes(void *states, unsigned int instanceIndex, const unsigned char *data, unsigned int offset, unsigned int length);
+void Xoodootimes8_OverwriteLanesAll(void *states, const unsigned char *data, unsigned int laneCount, unsigned int laneOffset);
+void Xoodootimes8_OverwriteWithZeroes(void *states, unsigned int instanceIndex, unsigned int byteCount);
+void Xoodootimes8_PermuteAll_6rounds(void *states);
+void Xoodootimes8_PermuteAll_12rounds(void *states);
+void Xoodootimes8_ExtractBytes(const void *states, unsigned int instanceIndex, unsigned char *data, unsigned int offset, unsigned int length);
+void Xoodootimes8_ExtractLanesAll(const void *states, unsigned char *data, unsigned int laneCount, unsigned int laneOffset);
+void Xoodootimes8_ExtractAndAddBytes(const void *states, unsigned int instanceIndex,  const unsigned char *input, unsigned char *output, unsigned int offset, unsigned int length);
+void Xoodootimes8_ExtractAndAddLanesAll(const void *states, const unsigned char *input, unsigned char *output, unsigned int laneCount, unsigned int laneOffset);
+
+#endif

data/ext/xkcp/lib/low/Xoodoo-times8/fallback-on1/Xoodoo-times8-on1.c

@@ -0,0 +1,37 @@
+/*
+The eXtended Keccak Code Package (XKCP)
+https://github.com/XKCP/XKCP
+
+The Xoodoo permutation, designed by Joan Daemen, Seth Hoffert, Gilles Van Assche and Ronny Van Keer.
+
+Implementation by Ronny Van Keer, hereby denoted as "the implementer".
+
+For more information, feedback or questions, please refer to the Keccak Team website:
+https://keccak.team/
+
+To the extent possible under law, the implementer has waived all copyright
+and related or neighboring rights to the source code in this file.
+http://creativecommons.org/publicdomain/zero/1.0/
+
+---
+
+This file implements Xoodoo×8 in a PlSnP-compatible way.
+Please refer to PlSnP-documentation.h for more details.
+
+This implementation comes with Xoodoo-times8-SnP.h in the same folder.
+Please refer to LowLevel.build for the exact list of other files it must be combined with.
+*/
+
+#include "Xoodoo-SnP.h"
+
+#define prefix                          Xoodootimes8
+#define PlSnP_baseParallelism           1
+#define PlSnP_targetParallelism         8
+#define SnP_laneLengthInBytes           4
+#define SnP                             Xoodoo
+#define SnP_Permute                     Xoodoo_Permute_6rounds
+#define SnP_Permute_12rounds            Xoodoo_Permute_12rounds
+#define PlSnP_PermuteAll                Xoodootimes8_PermuteAll_6rounds
+#define PlSnP_PermuteAll_12rounds       Xoodootimes8_PermuteAll_12rounds
+
+#include "PlSnP-Fallback.inc"

data/ext/xkcp/lib/low/common/PlSnP-Fallback.inc

@@ -0,0 +1,290 @@
+/*
+The eXtended Keccak Code Package (XKCP)
+https://github.com/XKCP/XKCP
+
+Implementation by Gilles Van Assche, hereby denoted as "the implementer".
+
+For more information, feedback or questions, please refer to the Keccak Team website:
+https://keccak.team/
+
+To the extent possible under law, the implementer has waived all copyright
+and related or neighboring rights to the source code in this file.
+http://creativecommons.org/publicdomain/zero/1.0/
+
+---
+
+This file contains macros that help make a PlSnP-compatible implementation by
+serially falling back on a SnP-compatible implementation or on a PlSnP-compatible
+implementation of lower parallism degree.
+
+Please refer to PlSnP-documentation.h for more details.
+*/
+
+/* expect PlSnP_baseParallelism, PlSnP_targetParallelism */
+/* expect SnP_stateSizeInBytes, SnP_stateAlignment */
+/* expect prefix */
+/* expect SnP_* */
+
+#define JOIN0(a, b)                     a ## b
+#define JOIN(a, b)                      JOIN0(a, b)
+
+#define PlSnP_StaticInitialize          JOIN(prefix, _StaticInitialize)
+#define PlSnP_InitializeAll             JOIN(prefix, _InitializeAll)
+#define PlSnP_AddByte                   JOIN(prefix, _AddByte)
+#define PlSnP_AddBytes                  JOIN(prefix, _AddBytes)
+#define PlSnP_AddLanesAll               JOIN(prefix, _AddLanesAll)
+#define PlSnP_OverwriteBytes            JOIN(prefix, _OverwriteBytes)
+#define PlSnP_OverwriteLanesAll         JOIN(prefix, _OverwriteLanesAll)
+#define PlSnP_OverwriteWithZeroes       JOIN(prefix, _OverwriteWithZeroes)
+#define PlSnP_ExtractBytes              JOIN(prefix, _ExtractBytes)
+#define PlSnP_ExtractLanesAll           JOIN(prefix, _ExtractLanesAll)
+#define PlSnP_ExtractAndAddBytes        JOIN(prefix, _ExtractAndAddBytes)
+#define PlSnP_ExtractAndAddLanesAll     JOIN(prefix, _ExtractAndAddLanesAll)
+
+#if (PlSnP_baseParallelism == 1)
+    #define SnP_stateSizeInBytes            JOIN(SnP, _stateSizeInBytes)
+    #define SnP_stateAlignment              JOIN(SnP, _stateAlignment)
+#else
+    #define SnP_stateSizeInBytes            JOIN(SnP, _statesSizeInBytes)
+    #define SnP_stateAlignment              JOIN(SnP, _statesAlignment)
+#endif
+#define PlSnP_factor ((PlSnP_targetParallelism)/(PlSnP_baseParallelism))
+#define SnP_stateOffset (((SnP_stateSizeInBytes+(SnP_stateAlignment-1))/SnP_stateAlignment)*SnP_stateAlignment)
+#define stateWithIndex(i) ((unsigned char *)states+((i)*SnP_stateOffset))
+
+#define SnP_StaticInitialize            JOIN(SnP, _StaticInitialize)
+#define SnP_Initialize                  JOIN(SnP, _Initialize)
+#define SnP_InitializeAll               JOIN(SnP, _InitializeAll)
+#define SnP_AddByte                     JOIN(SnP, _AddByte)
+#define SnP_AddBytes                    JOIN(SnP, _AddBytes)
+#define SnP_AddLanesAll                 JOIN(SnP, _AddLanesAll)
+#define SnP_OverwriteBytes              JOIN(SnP, _OverwriteBytes)
+#define SnP_OverwriteLanesAll           JOIN(SnP, _OverwriteLanesAll)
+#define SnP_OverwriteWithZeroes         JOIN(SnP, _OverwriteWithZeroes)
+#define SnP_ExtractBytes                JOIN(SnP, _ExtractBytes)
+#define SnP_ExtractLanesAll             JOIN(SnP, _ExtractLanesAll)
+#define SnP_ExtractAndAddBytes          JOIN(SnP, _ExtractAndAddBytes)
+#define SnP_ExtractAndAddLanesAll       JOIN(SnP, _ExtractAndAddLanesAll)
+
+void PlSnP_StaticInitialize( void )
+{
+    SnP_StaticInitialize();
+}
+
+void PlSnP_InitializeAll(void *states)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++)
+    #if (PlSnP_baseParallelism == 1)
+        SnP_Initialize(stateWithIndex(i));
+    #else
+        SnP_InitializeAll(stateWithIndex(i));
+    #endif
+}
+
+void PlSnP_AddByte(void *states, unsigned int instanceIndex, unsigned char byte, unsigned int offset)
+{
+    #if (PlSnP_baseParallelism == 1)
+        SnP_AddByte(stateWithIndex(instanceIndex), byte, offset);
+    #else
+        SnP_AddByte(stateWithIndex(instanceIndex/PlSnP_baseParallelism), instanceIndex%PlSnP_baseParallelism, byte, offset);
+    #endif
+}
+
+void PlSnP_AddBytes(void *states, unsigned int instanceIndex, const unsigned char *data, unsigned int offset, unsigned int length)
+{
+    #if (PlSnP_baseParallelism == 1)
+        SnP_AddBytes(stateWithIndex(instanceIndex), data, offset, length);
+    #else
+        SnP_AddBytes(stateWithIndex(instanceIndex/PlSnP_baseParallelism), instanceIndex%PlSnP_baseParallelism, data, offset, length);
+    #endif
+}
+
+void PlSnP_AddLanesAll(void *states, const unsigned char *data, unsigned int laneCount, unsigned int laneOffset)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_AddBytes(stateWithIndex(i), data, 0, laneCount*SnP_laneLengthInBytes);
+        #else
+            SnP_AddLanesAll(stateWithIndex(i), data, laneCount, laneOffset);
+        #endif
+        data += PlSnP_baseParallelism*laneOffset*SnP_laneLengthInBytes;
+    }
+}
+
+void PlSnP_OverwriteBytes(void *states, unsigned int instanceIndex, const unsigned char *data, unsigned int offset, unsigned int length)
+{
+    #if (PlSnP_baseParallelism == 1)
+        SnP_OverwriteBytes(stateWithIndex(instanceIndex), data, offset, length);
+    #else
+        SnP_OverwriteBytes(stateWithIndex(instanceIndex/PlSnP_baseParallelism), instanceIndex%PlSnP_baseParallelism, data, offset, length);
+    #endif
+}
+
+void PlSnP_OverwriteLanesAll(void *states, const unsigned char *data, unsigned int laneCount, unsigned int laneOffset)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_OverwriteBytes(stateWithIndex(i), data, 0, laneCount*SnP_laneLengthInBytes);
+        #else
+            SnP_OverwriteLanesAll(stateWithIndex(i), data, laneCount, laneOffset);
+        #endif
+        data += PlSnP_baseParallelism*laneOffset*SnP_laneLengthInBytes;
+    }
+}
+
+void PlSnP_OverwriteWithZeroes(void *states, unsigned int instanceIndex, unsigned int byteCount)
+{
+    #if (PlSnP_baseParallelism == 1)
+        SnP_OverwriteWithZeroes(stateWithIndex(instanceIndex), byteCount);
+    #else
+        SnP_OverwriteWithZeroes(stateWithIndex(instanceIndex/PlSnP_baseParallelism), instanceIndex%PlSnP_baseParallelism, byteCount);
+    #endif
+}
+
+void PlSnP_PermuteAll(void *states)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_Permute(stateWithIndex(i));
+        #else
+            SnP_PermuteAll(stateWithIndex(i));
+        #endif
+    }
+}
+
+#if (defined(SnP_Permute_12rounds) || defined(SnP_PermuteAll_12rounds))
+void PlSnP_PermuteAll_12rounds(void *states)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_Permute_12rounds(stateWithIndex(i));
+        #else
+            SnP_PermuteAll_12rounds(stateWithIndex(i));
+        #endif
+    }
+}
+#endif
+
+#if (defined(SnP_Permute_Nrounds) || defined(SnP_PermuteAll_6rounds))
+void PlSnP_PermuteAll_6rounds(void *states)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_Permute_Nrounds(stateWithIndex(i), 6);
+        #else
+            SnP_PermuteAll_6rounds(stateWithIndex(i));
+        #endif
+    }
+}
+#endif
+
+#if (defined(SnP_Permute_Nrounds) || defined(SnP_PermuteAll_4rounds))
+void PlSnP_PermuteAll_4rounds(void *states)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_Permute_Nrounds(stateWithIndex(i), 4);
+        #else
+            SnP_PermuteAll_4rounds(stateWithIndex(i));
+        #endif
+    }
+}
+#endif
+
+void PlSnP_ExtractBytes(void *states, unsigned int instanceIndex, unsigned char *data, unsigned int offset, unsigned int length)
+{
+    #if (PlSnP_baseParallelism == 1)
+        SnP_ExtractBytes(stateWithIndex(instanceIndex), data, offset, length);
+    #else
+        SnP_ExtractBytes(stateWithIndex(instanceIndex/PlSnP_baseParallelism), instanceIndex%PlSnP_baseParallelism, data, offset, length);
+    #endif
+}
+
+void PlSnP_ExtractLanesAll(const void *states, unsigned char *data, unsigned int laneCount, unsigned int laneOffset)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_ExtractBytes(stateWithIndex(i), data, 0, laneCount*SnP_laneLengthInBytes);
+        #else
+            SnP_ExtractLanesAll(stateWithIndex(i), data, laneCount, laneOffset);
+        #endif
+        data += laneOffset*SnP_laneLengthInBytes*PlSnP_baseParallelism;
+    }
+}
+
+void PlSnP_ExtractAndAddBytes(void *states, unsigned int instanceIndex, const unsigned char *input, unsigned char *output, unsigned int offset, unsigned int length)
+{
+    #if (PlSnP_baseParallelism == 1)
+        SnP_ExtractAndAddBytes(stateWithIndex(instanceIndex), input, output, offset, length);
+    #else
+        SnP_ExtractAndAddBytes(stateWithIndex(instanceIndex/PlSnP_baseParallelism), instanceIndex%PlSnP_baseParallelism, input, output, offset, length);
+    #endif
+}
+
+void PlSnP_ExtractAndAddLanesAll(const void *states, const unsigned char *input, unsigned char *output, unsigned int laneCount, unsigned int laneOffset)
+{
+    unsigned int i;
+
+    for(i=0; i<PlSnP_factor; i++) {
+        #if (PlSnP_baseParallelism == 1)
+            SnP_ExtractAndAddBytes(stateWithIndex(i), input, output, 0, laneCount*SnP_laneLengthInBytes);
+        #else
+            SnP_ExtractAndAddLanesAll(stateWithIndex(i), input, output, laneCount, laneOffset);
+        #endif
+        input += laneOffset*SnP_laneLengthInBytes*PlSnP_baseParallelism;
+        output += laneOffset*SnP_laneLengthInBytes*PlSnP_baseParallelism;
+    }
+}
+
+#undef PlSnP_factor
+#undef SnP_stateOffset
+#undef stateWithIndex
+#undef JOIN0
+#undef JOIN
+#undef PlSnP_StaticInitialize
+#undef PlSnP_InitializeAll
+#undef PlSnP_AddByte
+#undef PlSnP_AddBytes
+#undef PlSnP_AddLanesAll
+#undef PlSnP_OverwriteBytes
+#undef PlSnP_OverwriteLanesAll
+#undef PlSnP_OverwriteWithZeroes
+#undef PlSnP_PermuteAll
+#undef PlSnP_ExtractBytes
+#undef PlSnP_ExtractLanesAll
+#undef PlSnP_ExtractAndAddBytes
+#undef PlSnP_ExtractAndAddLanesAll
+#undef SnP_stateAlignment
+#undef SnP_stateSizeInBytes
+#undef PlSnP_factor
+#undef SnP_stateOffset
+#undef stateWithIndex
+#undef SnP_StaticInitialize
+#undef SnP_Initialize
+#undef SnP_InitializeAll
+#undef SnP_AddByte
+#undef SnP_AddBytes
+#undef SnP_AddLanesAll
+#undef SnP_OverwriteBytes
+#undef SnP_OverwriteWithZeroes
+#undef SnP_OverwriteLanesAll
+#undef SnP_ExtractBytes
+#undef SnP_ExtractLanesAll
+#undef SnP_ExtractAndAddBytes
+#undef SnP_ExtractAndAddLanesAll

data/ext/xkcp/lib/low/common/SnP-Relaned.h

@@ -0,0 +1,141 @@
+/*
+The eXtended Keccak Code Package (XKCP)
+https://github.com/XKCP/XKCP
+
+Implementation by Gilles Van Assche and Ronny Van Keer, hereby denoted as "the implementer".
+
+For more information, feedback or questions, please refer to the Keccak Team website:
+https://keccak.team/
+
+To the extent possible under law, the implementer has waived all copyright
+and related or neighboring rights to the source code in this file.
+http://creativecommons.org/publicdomain/zero/1.0/
+
+---
+
+This file contains macros that help implement a permutation in a SnP-compatible way.
+It converts an implementation that implement state input/output functions
+in a lane-oriented fashion (i.e., using SnP_AddLanes() and SnP_AddBytesInLane,
+and similarly for Overwite, Extract and ExtractAndAdd) to the byte-oriented SnP.
+Please refer to SnP-documentation.h for more details.
+*/
+
+#ifndef _SnP_Relaned_h_
+#define _SnP_Relaned_h_
+
+#define SnP_AddBytes(state, data, offset, length, SnP_AddLanes, SnP_AddBytesInLane, SnP_laneLengthInBytes) \
+    { \
+        if ((offset) == 0) { \
+            SnP_AddLanes(state, data, (length)/SnP_laneLengthInBytes); \
+            SnP_AddBytesInLane(state, \
+                (length)/SnP_laneLengthInBytes, \
+                (data)+((length)/SnP_laneLengthInBytes)*SnP_laneLengthInBytes, \
+                0, \
+                (length)%SnP_laneLengthInBytes); \
+        } \
+        else { \
+            unsigned int _sizeLeft = (length); \
+            unsigned int _lanePosition = (offset)/SnP_laneLengthInBytes; \
+            unsigned int _offsetInLane = (offset)%SnP_laneLengthInBytes; \
+            const unsigned char *_curData = (data); \
+            while(_sizeLeft > 0) { \
+                unsigned int _bytesInLane = SnP_laneLengthInBytes - _offsetInLane; \
+                if (_bytesInLane > _sizeLeft) \
+                    _bytesInLane = _sizeLeft; \
+                SnP_AddBytesInLane(state, _lanePosition, _curData, _offsetInLane, _bytesInLane); \
+                _sizeLeft -= _bytesInLane; \
+                _lanePosition++; \
+                _offsetInLane = 0; \
+                _curData += _bytesInLane; \
+            } \
+        } \
+    }
+
+#define SnP_OverwriteBytes(state, data, offset, length, SnP_OverwriteLanes, SnP_OverwriteBytesInLane, SnP_laneLengthInBytes) \
+    { \
+        if ((offset) == 0) { \
+            SnP_OverwriteLanes(state, data, (length)/SnP_laneLengthInBytes); \
+            SnP_OverwriteBytesInLane(state, \
+                (length)/SnP_laneLengthInBytes, \
+                (data)+((length)/SnP_laneLengthInBytes)*SnP_laneLengthInBytes, \
+                0, \
+                (length)%SnP_laneLengthInBytes); \
+        } \
+        else { \
+            unsigned int _sizeLeft = (length); \
+            unsigned int _lanePosition = (offset)/SnP_laneLengthInBytes; \
+            unsigned int _offsetInLane = (offset)%SnP_laneLengthInBytes; \
+            const unsigned char *_curData = (data); \
+            while(_sizeLeft > 0) { \
+                unsigned int _bytesInLane = SnP_laneLengthInBytes - _offsetInLane; \
+                if (_bytesInLane > _sizeLeft) \
+                    _bytesInLane = _sizeLeft; \
+                SnP_OverwriteBytesInLane(state, _lanePosition, _curData, _offsetInLane, _bytesInLane); \
+                _sizeLeft -= _bytesInLane; \
+                _lanePosition++; \
+                _offsetInLane = 0; \
+                _curData += _bytesInLane; \
+            } \
+        } \
+    }
+
+#define SnP_ExtractBytes(state, data, offset, length, SnP_ExtractLanes, SnP_ExtractBytesInLane, SnP_laneLengthInBytes) \
+    { \
+        if ((offset) == 0) { \
+            SnP_ExtractLanes(state, data, (length)/SnP_laneLengthInBytes); \
+            SnP_ExtractBytesInLane(state, \
+                (length)/SnP_laneLengthInBytes, \
+                (data)+((length)/SnP_laneLengthInBytes)*SnP_laneLengthInBytes, \
+                0, \
+                (length)%SnP_laneLengthInBytes); \
+        } \
+        else { \
+            unsigned int _sizeLeft = (length); \
+            unsigned int _lanePosition = (offset)/SnP_laneLengthInBytes; \
+            unsigned int _offsetInLane = (offset)%SnP_laneLengthInBytes; \
+            unsigned char *_curData = (data); \
+            while(_sizeLeft > 0) { \
+                unsigned int _bytesInLane = SnP_laneLengthInBytes - _offsetInLane; \
+                if (_bytesInLane > _sizeLeft) \
+                    _bytesInLane = _sizeLeft; \
+                SnP_ExtractBytesInLane(state, _lanePosition, _curData, _offsetInLane, _bytesInLane); \
+                _sizeLeft -= _bytesInLane; \
+                _lanePosition++; \
+                _offsetInLane = 0; \
+                _curData += _bytesInLane; \
+            } \
+        } \
+    }
+
+#define SnP_ExtractAndAddBytes(state, input, output, offset, length, SnP_ExtractAndAddLanes, SnP_ExtractAndAddBytesInLane, SnP_laneLengthInBytes) \
+    { \
+        if ((offset) == 0) { \
+            SnP_ExtractAndAddLanes(state, input, output, (length)/SnP_laneLengthInBytes); \
+            SnP_ExtractAndAddBytesInLane(state, \
+                (length)/SnP_laneLengthInBytes, \
+                (input)+((length)/SnP_laneLengthInBytes)*SnP_laneLengthInBytes, \
+                (output)+((length)/SnP_laneLengthInBytes)*SnP_laneLengthInBytes, \
+                0, \
+                (length)%SnP_laneLengthInBytes); \
+        } \
+        else { \
+            unsigned int _sizeLeft = (length); \
+            unsigned int _lanePosition = (offset)/SnP_laneLengthInBytes; \
+            unsigned int _offsetInLane = (offset)%SnP_laneLengthInBytes; \
+            const unsigned char *_curInput = (input); \
+            unsigned char *_curOutput = (output); \
+            while(_sizeLeft > 0) { \
+                unsigned int _bytesInLane = SnP_laneLengthInBytes - _offsetInLane; \
+                if (_bytesInLane > _sizeLeft) \
+                    _bytesInLane = _sizeLeft; \
+                SnP_ExtractAndAddBytesInLane(state, _lanePosition, _curInput, _curOutput, _offsetInLane, _bytesInLane); \
+                _sizeLeft -= _bytesInLane; \
+                _lanePosition++; \
+                _offsetInLane = 0; \
+                _curInput += _bytesInLane; \
+                _curOutput += _bytesInLane; \
+            } \
+        } \
+    }
+
+#endif

data/ext/xkcp/support/Build/ExpandProducts.xsl

@@ -0,0 +1,79 @@
+<?xml version='1.0' encoding="UTF-8"?>
+<!--
+The eXtended Keccak Code Package (XKCP)
+https://github.com/XKCP/XKCP
+
+Implementation by Gilles Van Assche, hereby denoted as "the implementer".
+
+For more information, feedback or questions, please refer to the Keccak Team website:
+https://keccak.team/
+
+To the extent possible under law, the implementer has waived all copyright
+and related or neighboring rights to the source code in this file.
+http://creativecommons.org/publicdomain/zero/1.0/
+-->
+<xsl:stylesheet
+    xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+    version='1.0'>
+
+<xsl:template name="expand">
+    <xsl:param name="first"/>
+    <xsl:param name="second"/>
+    <xsl:param name="first-explored" select="''"/>
+    <xsl:param name="second-explored" select="''"/>
+    <xsl:param name="delimiter"/>
+    <xsl:variable name="first-cur" select="substring-before(concat($first, ' '), ' ')"/>
+    <xsl:variable name="first-rest" select="substring-after($first, ' ')"/>
+    <xsl:variable name="second-cur" select="substring-before(concat($second, ' '), ' ')"/>
+    <xsl:variable name="second-rest" select="substring-after($second, ' ')"/>
+
+    <xsl:if test="($first-cur!='') and ($second-cur!='')">
+        <target>
+            <xsl:attribute name="name"><xsl:value-of select="concat($first-cur, $delimiter, $second-cur)"/></xsl:attribute>
+            <xsl:attribute name="inherits"><xsl:value-of select="concat($first-cur, ' ', $second-cur)"/></xsl:attribute>
+        </target>
+
+        <xsl:choose>
+            <xsl:when test="normalize-space($second-rest)!=''">
+                <xsl:call-template name="expand">
+                    <xsl:with-param name="first" select="$first"/>
+                    <xsl:with-param name="first-explored" select="$first-explored"/>
+                    <xsl:with-param name="second" select="$second-rest"/>
+                    <xsl:with-param name="second-explored" select="normalize-space(concat($second-explored, ' ', $second-cur))"/>
+                    <xsl:with-param name="delimiter" select="$delimiter"/>
+                </xsl:call-template>
+            </xsl:when>
+            <xsl:when test="normalize-space($first-rest)!=''">
+                <xsl:call-template name="expand">
+                    <xsl:with-param name="first" select="$first-rest"/>
+                    <xsl:with-param name="first-explored" select="normalize-space(concat($first-explored, ' ', $first-cur))"/>
+                    <xsl:with-param name="second" select="normalize-space(concat($second-explored, ' ', $second-cur))"/>
+                    <xsl:with-param name="second-explored" select="''"/>
+                    <xsl:with-param name="delimiter" select="$delimiter"/>
+                </xsl:call-template>
+            </xsl:when>
+        </xsl:choose>
+    </xsl:if>
+</xsl:template>
+
+<xsl:template match="product">
+    <xsl:call-template name="expand">
+        <xsl:with-param name="first" select="normalize-space(factor[1]/@set)"/>
+        <xsl:with-param name="second" select="normalize-space(factor[2]/@set)"/>
+        <xsl:with-param name="delimiter">
+            <xsl:choose>
+                <xsl:when test="@delimiter!=''"><xsl:value-of select="@delimiter"/></xsl:when>
+                <xsl:otherwise>+</xsl:otherwise>
+            </xsl:choose>
+        </xsl:with-param>
+    </xsl:call-template>
+</xsl:template>
+
+<xsl:template match="*">
+    <xsl:copy>
+        <xsl:copy-of select="@*"/>
+        <xsl:apply-templates/>
+    </xsl:copy>
+</xsl:template>
+
+</xsl:stylesheet>