diff --git a/src/XKCP/lib/LowLevel.build b/src/XKCP/lib/LowLevel.build
index f504190eedab500cd0a9291a3ef6a5956ef82534..d6d14985844c52d6b5a58834181623af25db2828 100755
--- a/src/XKCP/lib/LowLevel.build
+++ b/src/XKCP/lib/LowLevel.build
@@ -62,11 +62,9 @@ The fragments below allow to select the desired implementation of the permutatio
* optimized1600lcu6: same as optimized1600u6 but using the lane complementing technique, which is useful for platforms that do not have a "and not" instruction
* optimized1600lcufull: same as optimized1600lcu6 but with all rounds unrolled
* optimized1600lcufullshld: same as optimized1600lcufull but with the rotation implementation with the 'shld' instruction, as it is faster on some platforms (e.g., SandyBridge)
-
* optimized1600AsmX86-64: an assembly-optimized implementation for x86_64
* optimized1600AsmX86-64shld: same as optimized1600AsmX86-64, but with the 'shld' instruction
* optimized1600AsmX86-64Apple: same as optimized1600AsmX86-64, but with a syntax that works better on some Apple platforms
-
* optimized1600AVX2: an optimized implementation taking advantage of the AVX2 instruction set
* optimized1600AVX512a: an optimized implementation taking advantage of the AVX512 instruction set (in assembler)
* optimized1600AVX512c: an optimized implementation taking advantage of the AVX512 instruction set (in C)
diff --git a/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.c b/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.c
index 3336f68a36f9e6b8c66b909cc8f3187161becd7f..6dc94232cc3102ffaa11d2b01aaa824d9ebc1b9f 100755
--- a/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.c
+++ b/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.c
@@ -16,25 +16,6 @@ int SHAKE128(unsigned char *output, size_t outputByteLen, const unsigned char *i
return KeccakWidth1600_Sponge(1344, 256, input, inputByteLen, 0x1F, output, outputByteLen);
}
-void SHAKE128_InitAbsorb(Keccak_HashInstance *ks, const unsigned char *input, size_t inputByteLen)
-{
- Keccak_HashInitialize_SHAKE128( ks );
- Keccak_HashUpdate( &ks, input, inputByteLen * 8 );
- Keccak_HashFinal( &ks, input );
-}
-
-void SHAKE256_InitAbsorb(Keccak_HashInstance *ks, const unsigned char *input, size_t inputByteLen)
-{
- Keccak_HashInitialize_SHAKE256( ks );
- Keccak_HashUpdate( &ks, input, inputByteLen * 8 );
- Keccak_HashFinal( &ks, input );
-}
-
-void KECCAK_HashSqueeze(Keccak_HashInstance *ks, const unsigned char *out, size_t outByteLen)
-{
- Keccak_HashSqueeze( &ks, (unsigned char *)out, outByteLen * 8 );
-}
-
int SHAKE256(unsigned char *output, size_t outputByteLen, const unsigned char *input, size_t inputByteLen)
{
return KeccakWidth1600_Sponge(1088, 512, input, inputByteLen, 0x1F, output, outputByteLen);
diff --git a/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.h b/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.h
index a2fa94413903d51d0ccc5a70872ab1344e0d548b..b3c68ae842fc4f9667a4769a7b9f5e3d6f2deec8 100755
--- a/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.h
+++ b/src/XKCP/lib/high/Keccak/FIPS202/SimpleFIPS202.h
@@ -12,7 +12,6 @@ http://creativecommons.org/publicdomain/zero/1.0/
#ifndef _SimpleFIPS202_h_
#define _SimpleFIPS202_h_
-#include "KeccakHash.h"
#include "KeccakSpongeWidth1600.h"
#include <string.h>
@@ -23,6 +22,7 @@ http://creativecommons.org/publicdomain/zero/1.0/
* @param inputByteLen The length of the input message in bytes.
* @return 0 if successful, 1 otherwise.
*/
+int SHAKE128(unsigned char *output, size_t outputByteLen, const unsigned char *input, size_t inputByteLen);
/** Implementation of the SHAKE256 extendable output function (XOF) [FIPS 202].
* @param output Pointer to the output buffer.
@@ -31,6 +31,7 @@ http://creativecommons.org/publicdomain/zero/1.0/
* @param inputByteLen The length of the input message in bytes.
* @return 0 if successful, 1 otherwise.
*/
+int SHAKE256(unsigned char *output, size_t outputByteLen, const unsigned char *input, size_t inputByteLen);
/** Implementation of SHA3-224 [FIPS 202].
* @param output Pointer to the output buffer (28 bytes).
@@ -38,6 +39,7 @@ http://creativecommons.org/publicdomain/zero/1.0/
* @param inputByteLen The length of the input message in bytes.
* @return 0 if successful, 1 otherwise.
*/
+int SHA3_224(unsigned char *output, const unsigned char *input, size_t inputByteLen);
/** Implementation of SHA3-256 [FIPS 202].
* @param output Pointer to the output buffer (32 bytes).
@@ -45,6 +47,7 @@ http://creativecommons.org/publicdomain/zero/1.0/
* @param inputByteLen The length of the input message in bytes.
* @return 0 if successful, 1 otherwise.
*/
+int SHA3_256(unsigned char *output, const unsigned char *input, size_t inputByteLen);
/** Implementation of SHA3-384 [FIPS 202].
* @param output Pointer to the output buffer (48 bytes).
@@ -52,6 +55,7 @@ http://creativecommons.org/publicdomain/zero/1.0/
* @param inputByteLen The length of the input message in bytes.
* @return 0 if successful, 1 otherwise.
*/
+int SHA3_384(unsigned char *output, const unsigned char *input, size_t inputByteLen);
/** Implementation of SHA3-512 [FIPS 202].
* @param output Pointer to the output buffer (64 bytes).
@@ -59,17 +63,6 @@ http://creativecommons.org/publicdomain/zero/1.0/
* @param inputByteLen The length of the input message in bytes.
* @return 0 if successful, 1 otherwise.
*/
-
-int SHAKE128(unsigned char *output, size_t outputByteLen, const unsigned char *input, size_t inputByteLen);
-int SHAKE256(unsigned char *output, size_t outputByteLen, const unsigned char *input, size_t inputByteLen);
-int SHA3_224(unsigned char *output, const unsigned char *input, size_t inputByteLen);
-int SHA3_256(unsigned char *output, const unsigned char *input, size_t inputByteLen);
-int SHA3_384(unsigned char *output, const unsigned char *input, size_t inputByteLen);
int SHA3_512(unsigned char *output, const unsigned char *input, size_t inputByteLen);
-void SHAKE128_InitAbsorb( Keccak_HashInstance *ks, const unsigned char *input, size_t inputByteLen );
-void SHAKE256_InitAbsorb( Keccak_HashInstance *ks, const unsigned char *input, size_t inputByteLen );
-void KECCAK_HashSqueeze(Keccak_HashInstance *ks, const unsigned char *out, size_t outByteLen);
-
-
#endif
diff --git a/src/XKCP/lib/high/Keccak/KeccakSponge-common.h b/src/XKCP/lib/high/Keccak/KeccakSponge-common.h
index e2b73e896ab18284e31ce2b29694a0e150d143d9..2939f42c76150f8f2050a89a46fdae9713b0ece0 100755
--- a/src/XKCP/lib/high/Keccak/KeccakSponge-common.h
+++ b/src/XKCP/lib/high/Keccak/KeccakSponge-common.h
@@ -18,7 +18,7 @@ http://creativecommons.org/publicdomain/zero/1.0/
#include "align.h"
#define KCP_DeclareSpongeStructure(prefix, size, alignment) \
- ALIGN(64) typedef struct prefix##_SpongeInstanceStruct { \
+ ALIGN(alignment) typedef struct prefix##_SpongeInstanceStruct { \
unsigned char state[size]; \
unsigned int rate; \
unsigned int byteIOIndex; \
diff --git a/src/XKCP/lib/low/KeccakP-1600-times2/SIMD128/KeccakP-1600-times2-SIMD128.c b/src/XKCP/lib/low/KeccakP-1600-times2/SIMD128/KeccakP-1600-times2-SIMD128.c
index 47d4ca5de60b39c77cdbdbea5fc8533de38e7eec..99e59c69604f3a7659e1a57ed3db12fb3c99cd5a 100755
--- a/src/XKCP/lib/low/KeccakP-1600-times2/SIMD128/KeccakP-1600-times2-SIMD128.c
+++ b/src/XKCP/lib/low/KeccakP-1600-times2/SIMD128/KeccakP-1600-times2-SIMD128.c
@@ -21,16 +21,6 @@ Please refer to LowLevel.build for the exact list of other files it must be comb
#include <stdlib.h>
#include <string.h>
#include <x86intrin.h>
-// #include <x86intrin.h>
-
-//#include <smmintrin.h>
-//#include <wmmintrin.h>
-//#include <immintrin.h>
-//#include <emmintrin.h>
-
-// #include <mmintrin.h>
-// #include <emmintrin.h>
-
#include "align.h"
#include "KeccakP-1600-times2-SnP.h"
#include "SIMD128-config.h"
diff --git a/src/XKCP/lib/low/KeccakP-1600-times4/SIMD256/KeccakP-1600-times4-SIMD256.c b/src/XKCP/lib/low/KeccakP-1600-times4/SIMD256/KeccakP-1600-times4-SIMD256.c
index 75c692ea26a8b4bd459480a77723e663f5157ed8..cea7584b2815439a8901d3490822c8bbe1874dbb 100755
--- a/src/XKCP/lib/low/KeccakP-1600-times4/SIMD256/KeccakP-1600-times4-SIMD256.c
+++ b/src/XKCP/lib/low/KeccakP-1600-times4/SIMD256/KeccakP-1600-times4-SIMD256.c
@@ -45,7 +45,7 @@ typedef __m256i V256;
#if defined(KeccakP1600times4_useAVX2)
#define ANDnu256(a, b) _mm256_andnot_si256(a, b)
#define CONST256(a) _mm256_load_si256((const V256 *)&(a))
- #define CONST256_64(a) (V256)_mm256_broadcast_sd((const double*)(&a))
+ #define CONST256_64(a) _mm256_set1_epi64x(a)
#define LOAD256(a) _mm256_load_si256((const V256 *)&(a))
#define LOAD256u(a) _mm256_loadu_si256((const V256 *)&(a))
#define LOAD4_64(a, b, c, d) _mm256_set_epi64x((UINT64)(a), (UINT64)(b), (UINT64)(c), (UINT64)(d))
@@ -56,13 +56,13 @@ static const UINT64 rho8[4] = {0x0605040302010007, 0x0E0D0C0B0A09080F, 0x1615141
static const UINT64 rho56[4] = {0x0007060504030201, 0x080F0E0D0C0B0A09, 0x1017161514131211, 0x181F1E1D1C1B1A19};
#define STORE256(a, b) _mm256_store_si256((V256 *)&(a), b)
#define STORE256u(a, b) _mm256_storeu_si256((V256 *)&(a), b)
- #define STORE2_128(ah, al, v) _mm256_storeu2_m128d((V128*)&(ah), (V128*)&(al), v)
+ #define STORE2_128(ah, al, v) _mm256_storeu2_m128i(&(ah), &(al), v)
#define XOR256(a, b) _mm256_xor_si256(a, b)
#define XOReq256(a, b) a = _mm256_xor_si256(a, b)
#define UNPACKL( a, b ) _mm256_unpacklo_epi64((a), (b))
#define UNPACKH( a, b ) _mm256_unpackhi_epi64((a), (b))
- #define PERM128( a, b, c ) (V256)_mm256_permute2f128_ps((__m256)(a), (__m256)(b), c)
- #define SHUFFLE64( a, b, c ) (V256)_mm256_shuffle_pd((__m256d)(a), (__m256d)(b), c)
+ #define PERM128( a, b, c ) _mm256_permute2f128_si256((a), (b), c)
+ #define SHUFFLE64( a, b, c ) _mm256_castpd_si256(_mm256_shuffle_pd(_mm256_castsi256_pd(a), _mm256_castsi256_pd(b), c))
#define UNINTLEAVE() lanesL01 = UNPACKL( lanes0, lanes1 ), \
lanesH01 = UNPACKH( lanes0, lanes1 ), \
diff --git a/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forHaswell/KeccakP-1600-opt64.s b/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forHaswell/KeccakP-1600-opt64.s
index 2712a48ae6b777e16ac992c375cd507712b3df60..31826dbb8231060d33d3316a026af1fe2ae381e0 100755
--- a/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forHaswell/KeccakP-1600-opt64.s
+++ b/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forHaswell/KeccakP-1600-opt64.s
@@ -13762,4 +13762,4 @@ KeccakF1600_FastLoop_Absorb:
.size KeccakF1600_FastLoop_Absorb, .-KeccakF1600_FastLoop_Absorb
.p2align 4,,15
.ident "GCC: (SUSE Linux) 4.7.4"
-# .section .note.GNU-stack,"",@progbits
+ .section .note.GNU-stack,"",@progbits
diff --git a/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forSandyBridge/KeccakP-1600-opt64.s b/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forSandyBridge/KeccakP-1600-opt64.s
index be377b4c23a78b534c432309a7656f3bdd5754f5..d36c3fd0b3d0c58f3e94815b17f357a1846705d4 100755
--- a/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forSandyBridge/KeccakP-1600-opt64.s
+++ b/src/XKCP/lib/low/KeccakP-1600/Optimized64/CompiledByGCC474forSandyBridge/KeccakP-1600-opt64.s
@@ -1,9 +1,8 @@
.file "KeccakP-1600-opt64.c"
.text
- .section .text$KeccakP1600_StaticInitialize,"x"
.p2align 4,,15
.globl KeccakP1600_StaticInitialize
- #.type KeccakP1600_StaticInitialize, @function
+ .type KeccakP1600_StaticInitialize, @function
KeccakP1600_StaticInitialize:
.LFB22:
.cfi_startproc
@@ -11,11 +10,10 @@ KeccakP1600_StaticInitialize:
ret
.cfi_endproc
.LFE22:
- #.size KeccakP1600_StaticInitialize, .-KeccakP1600_StaticInitialize
- .section .text$KeccakP1600_Initialize,"x"
+ .size KeccakP1600_StaticInitialize, .-KeccakP1600_StaticInitialize
.p2align 4,,15
.globl KeccakP1600_Initialize
- #.type KeccakP1600_Initialize, @function
+ .type KeccakP1600_Initialize, @function
KeccakP1600_Initialize:
.LFB23:
.cfi_startproc
@@ -79,10 +77,10 @@ KeccakP1600_Initialize:
jmp .L5
.cfi_endproc
.LFE23:
- #.size KeccakP1600_Initialize, .-KeccakP1600_Initialize
+ .size KeccakP1600_Initialize, .-KeccakP1600_Initialize
.p2align 4,,15
.globl KeccakP1600_AddBytesInLane
- #.type KeccakP1600_AddBytesInLane, @function
+ .type KeccakP1600_AddBytesInLane, @function
KeccakP1600_AddBytesInLane:
.LFB24:
.cfi_startproc
@@ -131,10 +129,10 @@ KeccakP1600_AddBytesInLane:
jmp .L34
.cfi_endproc
.LFE24:
- #.size KeccakP1600_AddBytesInLane, .-KeccakP1600_AddBytesInLane
+ .size KeccakP1600_AddBytesInLane, .-KeccakP1600_AddBytesInLane
.p2align 4,,15
.globl KeccakP1600_AddLanes
- #.type KeccakP1600_AddLanes, @function
+ .type KeccakP1600_AddLanes, @function
KeccakP1600_AddLanes:
.LFB25:
.cfi_startproc
@@ -244,10 +242,10 @@ KeccakP1600_AddLanes:
jmp .L45
.cfi_endproc
.LFE25:
- #.size KeccakP1600_AddLanes, .-KeccakP1600_AddLanes
+ .size KeccakP1600_AddLanes, .-KeccakP1600_AddLanes
.p2align 4,,15
.globl KeccakP1600_AddByte
- #.type KeccakP1600_AddByte, @function
+ .type KeccakP1600_AddByte, @function
KeccakP1600_AddByte:
.LFB26:
.cfi_startproc
@@ -261,10 +259,10 @@ KeccakP1600_AddByte:
ret
.cfi_endproc
.LFE26:
- #.size KeccakP1600_AddByte, .-KeccakP1600_AddByte
+ .size KeccakP1600_AddByte, .-KeccakP1600_AddByte
.p2align 4,,15
.globl KeccakP1600_AddBytes
- #.type KeccakP1600_AddBytes, @function
+ .type KeccakP1600_AddBytes, @function
KeccakP1600_AddBytes:
.LFB27:
.cfi_startproc
@@ -377,10 +375,10 @@ KeccakP1600_AddBytes:
jmp .L66
.cfi_endproc
.LFE27:
- #.size KeccakP1600_AddBytes, .-KeccakP1600_AddBytes
+ .size KeccakP1600_AddBytes, .-KeccakP1600_AddBytes
.p2align 4,,15
.globl KeccakP1600_OverwriteBytesInLane
- #.type KeccakP1600_OverwriteBytesInLane, @function
+ .type KeccakP1600_OverwriteBytesInLane, @function
KeccakP1600_OverwriteBytesInLane:
.LFB28:
.cfi_startproc
@@ -430,10 +428,10 @@ KeccakP1600_OverwriteBytesInLane:
jmp memcpy
.cfi_endproc
.LFE28:
- #.size KeccakP1600_OverwriteBytesInLane, .-KeccakP1600_OverwriteBytesInLane
+ .size KeccakP1600_OverwriteBytesInLane, .-KeccakP1600_OverwriteBytesInLane
.p2align 4,,15
.globl KeccakP1600_OverwriteLanes
- #.type KeccakP1600_OverwriteLanes, @function
+ .type KeccakP1600_OverwriteLanes, @function
KeccakP1600_OverwriteLanes:
.LFB29:
.cfi_startproc
@@ -478,10 +476,10 @@ KeccakP1600_OverwriteLanes:
ret
.cfi_endproc
.LFE29:
- #.size KeccakP1600_OverwriteLanes, .-KeccakP1600_OverwriteLanes
+ .size KeccakP1600_OverwriteLanes, .-KeccakP1600_OverwriteLanes
.p2align 4,,15
.globl KeccakP1600_OverwriteBytes
- #.type KeccakP1600_OverwriteBytes, @function
+ .type KeccakP1600_OverwriteBytes, @function
KeccakP1600_OverwriteBytes:
.LFB30:
.cfi_startproc
@@ -695,10 +693,10 @@ KeccakP1600_OverwriteBytes:
jmp .L110
.cfi_endproc
.LFE30:
- #.size KeccakP1600_OverwriteBytes, .-KeccakP1600_OverwriteBytes
+ .size KeccakP1600_OverwriteBytes, .-KeccakP1600_OverwriteBytes
.p2align 4,,15
.globl KeccakP1600_OverwriteWithZeroes
- #.type KeccakP1600_OverwriteWithZeroes, @function
+ .type KeccakP1600_OverwriteWithZeroes, @function
KeccakP1600_OverwriteWithZeroes:
.LFB31:
.cfi_startproc
@@ -774,10 +772,10 @@ KeccakP1600_OverwriteWithZeroes:
ret
.cfi_endproc
.LFE31:
- #.size KeccakP1600_OverwriteWithZeroes, .-KeccakP1600_OverwriteWithZeroes
+ .size KeccakP1600_OverwriteWithZeroes, .-KeccakP1600_OverwriteWithZeroes
.p2align 4,,15
.globl KeccakP1600_Permute_24rounds
- #.type KeccakP1600_Permute_24rounds, @function
+ .type KeccakP1600_Permute_24rounds, @function
KeccakP1600_Permute_24rounds:
.LFB32:
.cfi_startproc
@@ -8217,10 +8215,10 @@ KeccakP1600_Permute_24rounds:
ret
.cfi_endproc
.LFE32:
- #.size KeccakP1600_Permute_24rounds, .-KeccakP1600_Permute_24rounds
+ .size KeccakP1600_Permute_24rounds, .-KeccakP1600_Permute_24rounds
.p2align 4,,15
.globl KeccakP1600_Permute_12rounds
- #.type KeccakP1600_Permute_12rounds, @function
+ .type KeccakP1600_Permute_12rounds, @function
KeccakP1600_Permute_12rounds:
.LFB33:
.cfi_startproc
@@ -11945,10 +11943,10 @@ KeccakP1600_Permute_12rounds:
ret
.cfi_endproc
.LFE33:
- #.size KeccakP1600_Permute_12rounds, .-KeccakP1600_Permute_12rounds
+ .size KeccakP1600_Permute_12rounds, .-KeccakP1600_Permute_12rounds
.p2align 4,,15
.globl KeccakP1600_ExtractBytesInLane
- #.type KeccakP1600_ExtractBytesInLane, @function
+ .type KeccakP1600_ExtractBytesInLane, @function
KeccakP1600_ExtractBytesInLane:
.LFB34:
.cfi_startproc
@@ -11991,10 +11989,10 @@ KeccakP1600_ExtractBytesInLane:
jmp .L162
.cfi_endproc
.LFE34:
- #.size KeccakP1600_ExtractBytesInLane, .-KeccakP1600_ExtractBytesInLane
+ .size KeccakP1600_ExtractBytesInLane, .-KeccakP1600_ExtractBytesInLane
.p2align 4,,15
.globl KeccakP1600_ExtractLanes
- #.type KeccakP1600_ExtractLanes, @function
+ .type KeccakP1600_ExtractLanes, @function
KeccakP1600_ExtractLanes:
.LFB35:
.cfi_startproc
@@ -12038,10 +12036,10 @@ KeccakP1600_ExtractLanes:
ret
.cfi_endproc
.LFE35:
- #.size KeccakP1600_ExtractLanes, .-KeccakP1600_ExtractLanes
+ .size KeccakP1600_ExtractLanes, .-KeccakP1600_ExtractLanes
.p2align 4,,15
.globl KeccakP1600_ExtractBytes
- #.type KeccakP1600_ExtractBytes, @function
+ .type KeccakP1600_ExtractBytes, @function
KeccakP1600_ExtractBytes:
.LFB36:
.cfi_startproc
@@ -12204,10 +12202,10 @@ KeccakP1600_ExtractBytes:
jmp .L180
.cfi_endproc
.LFE36:
- #.size KeccakP1600_ExtractBytes, .-KeccakP1600_ExtractBytes
+ .size KeccakP1600_ExtractBytes, .-KeccakP1600_ExtractBytes
.p2align 4,,15
.globl KeccakP1600_ExtractAndAddBytesInLane
- #.type KeccakP1600_ExtractAndAddBytesInLane, @function
+ .type KeccakP1600_ExtractAndAddBytesInLane, @function
KeccakP1600_ExtractAndAddBytesInLane:
.LFB37:
.cfi_startproc
@@ -12263,10 +12261,10 @@ KeccakP1600_ExtractAndAddBytesInLane:
jmp .L193
.cfi_endproc
.LFE37:
- #.size KeccakP1600_ExtractAndAddBytesInLane, .-KeccakP1600_ExtractAndAddBytesInLane
+ .size KeccakP1600_ExtractAndAddBytesInLane, .-KeccakP1600_ExtractAndAddBytesInLane
.p2align 4,,15
.globl KeccakP1600_ExtractAndAddLanes
- #.type KeccakP1600_ExtractAndAddLanes, @function
+ .type KeccakP1600_ExtractAndAddLanes, @function
KeccakP1600_ExtractAndAddLanes:
.LFB38:
.cfi_startproc
@@ -12352,10 +12350,10 @@ KeccakP1600_ExtractAndAddLanes:
jmp .L213
.cfi_endproc
.LFE38:
- #.size KeccakP1600_ExtractAndAddLanes, .-KeccakP1600_ExtractAndAddLanes
+ .size KeccakP1600_ExtractAndAddLanes, .-KeccakP1600_ExtractAndAddLanes
.p2align 4,,15
.globl KeccakP1600_ExtractAndAddBytes
- #.type KeccakP1600_ExtractAndAddBytes, @function
+ .type KeccakP1600_ExtractAndAddBytes, @function
KeccakP1600_ExtractAndAddBytes:
.LFB39:
.cfi_startproc
@@ -12562,10 +12560,10 @@ KeccakP1600_ExtractAndAddBytes:
jmp .L243
.cfi_endproc
.LFE39:
- #.size KeccakP1600_ExtractAndAddBytes, .-KeccakP1600_ExtractAndAddBytes
+ .size KeccakP1600_ExtractAndAddBytes, .-KeccakP1600_ExtractAndAddBytes
.p2align 4,,15
.globl KeccakF1600_FastLoop_Absorb
- #.type KeccakF1600_FastLoop_Absorb, @function
+ .type KeccakF1600_FastLoop_Absorb, @function
KeccakF1600_FastLoop_Absorb:
.LFB40:
.cfi_startproc
@@ -20324,6 +20322,6 @@ KeccakF1600_FastLoop_Absorb:
jmp .L255
.cfi_endproc
.LFE40:
- #.size KeccakF1600_FastLoop_Absorb, .-KeccakF1600_FastLoop_Absorb
-# .ident "GCC: (SUSE Linux) 4.7.4"
-# .section .note.GNU-stack,"",@progbits
+ .size KeccakF1600_FastLoop_Absorb, .-KeccakF1600_FastLoop_Absorb
+ .ident "GCC: (SUSE Linux) 4.7.4"
+ .section .note.GNU-stack,"",@progbits
diff --git a/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-gas.s b/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-gas.s
index 2eea40444c356b7ec383ac27904ad9e789c1e77a..65aff45cb064fb1773bc06dea717e50cdd70c88d 100755
--- a/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-gas.s
+++ b/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-gas.s
@@ -22,7 +22,7 @@
.text
# conditional assembly settings
-.equ UseSIMD, 1
+.equ UseSIMD, 0
.equ InlinePerm, 1
# offsets in state
@@ -474,7 +474,7 @@
.if InlinePerm == 1
mKeccakPermutation24
.else
- callq KeccakP1600_Permute_24rounds
+ callq KeccakP1600_Permute_24rounds@PLT
.endif
.endm
@@ -749,7 +749,7 @@ KeccakLaneComplementTable:
.type KeccakP1600_OverwriteBytes, %function
KeccakP1600_OverwriteBytes:
addq arg3, arg1
- leaq KeccakLaneComplementTable, arg5
+ leaq KeccakLaneComplementTable(%rip), arg5
addq arg3, arg5
subq $8, arg4
jc KeccakP1600_OverwriteBytes_Bytes
@@ -786,7 +786,7 @@ KeccakP1600_OverwriteBytes_Exit:
.global KeccakP1600_OverwriteWithZeroes
.type KeccakP1600_OverwriteWithZeroes, %function
KeccakP1600_OverwriteWithZeroes:
- leaq KeccakLaneComplementTable, arg5
+ leaq KeccakLaneComplementTable(%rip), arg5
subq $8, arg2
jc KeccakP1600_OverwriteWithZeroes_Bytes
KeccakP1600_OverwriteWithZeroes_LanesLoop:
@@ -821,7 +821,7 @@ KeccakP1600_OverwriteWithZeroes_Exit:
.type KeccakP1600_ExtractBytes, %function
KeccakP1600_ExtractBytes:
addq arg3, arg1
- leaq KeccakLaneComplementTable, arg5
+ leaq KeccakLaneComplementTable(%rip), arg5
addq arg3, arg5
subq $8, arg4
jc KeccakP1600_ExtractBytes_Bytes
@@ -859,7 +859,7 @@ KeccakP1600_ExtractBytes_Exit:
.type KeccakP1600_ExtractAndAddBytes, %function
KeccakP1600_ExtractAndAddBytes:
addq arg4, arg1
- leaq KeccakLaneComplementTable, arg6
+ leaq KeccakLaneComplementTable(%rip), arg6
addq arg4, arg6
subq $8, arg5
jc KeccakP1600_ExtractAndAddBytes_Bytes
@@ -980,7 +980,8 @@ KeccakP1600_Permute_Nrounds:
movq rT2a, _su(rpStack)
KeccakP1600_Permute_Nrounds_Dispatch:
shlq $3, rT1
- jmp *KeccakP1600_Permute_NroundsTable-8(rT1)
+ leaq KeccakP1600_Permute_NroundsTable-8(%rip), rT2a
+ jmp *(rT1, rT2a)
KeccakP1600_Permute_Nrounds24:
mKeccakRound rpState, rpStack, 0x0000000000000001, 0
@@ -1174,13 +1175,13 @@ KeccakF1600_FastLoop_Absorb_VariableLaneCountLoop:
shlq $3, arg4
movq arg3, arg2 # data pointer
xorq arg3, arg3 # offset = 0
- callq KeccakP1600_AddBytes # (void *state, const unsigned char *data, unsigned int offset, unsigned int length)
+ callq KeccakP1600_AddBytes@PLT # (void *state, const unsigned char *data, unsigned int offset, unsigned int length)
movq arg2, arg3 # updated data pointer
movq 24(%rsp), rT1a # xor trailingBits
xorq rT1a, (arg1)
popq arg1
pushq arg3
- callq KeccakP1600_Permute_24rounds
+ callq KeccakP1600_Permute_24rounds@PLT
popq arg3
popq arg2
popq arg4
diff --git a/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-shld-gas.s b/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-shld-gas.s
index 9e3e573274c139f29f5301393330e64a3f9271e8..cf4a1b1e877e4864747521a93d60803fbe235d3e 100755
--- a/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-shld-gas.s
+++ b/src/XKCP/lib/low/KeccakP-1600/OptimizedAsmX86-64/KeccakP-1600-x86-64-shld-gas.s
@@ -474,7 +474,7 @@
.if InlinePerm == 1
mKeccakPermutation24
.else
- callq KeccakP1600_Permute_24rounds
+ callq KeccakP1600_Permute_24rounds@PLT
.endif
.endm
@@ -749,7 +749,7 @@ KeccakLaneComplementTable:
.type KeccakP1600_OverwriteBytes, %function
KeccakP1600_OverwriteBytes:
addq arg3, arg1
- leaq KeccakLaneComplementTable, arg5
+ leaq KeccakLaneComplementTable(%rip), arg5
addq arg3, arg5
subq $8, arg4
jc KeccakP1600_OverwriteBytes_Bytes
@@ -786,7 +786,7 @@ KeccakP1600_OverwriteBytes_Exit:
.global KeccakP1600_OverwriteWithZeroes
.type KeccakP1600_OverwriteWithZeroes, %function
KeccakP1600_OverwriteWithZeroes:
- leaq KeccakLaneComplementTable, arg5
+ leaq KeccakLaneComplementTable(%rip), arg5
subq $8, arg2
jc KeccakP1600_OverwriteWithZeroes_Bytes
KeccakP1600_OverwriteWithZeroes_LanesLoop:
@@ -821,7 +821,7 @@ KeccakP1600_OverwriteWithZeroes_Exit:
.type KeccakP1600_ExtractBytes, %function
KeccakP1600_ExtractBytes:
addq arg3, arg1
- leaq KeccakLaneComplementTable, arg5
+ leaq KeccakLaneComplementTable(%rip), arg5
addq arg3, arg5
subq $8, arg4
jc KeccakP1600_ExtractBytes_Bytes
@@ -859,7 +859,7 @@ KeccakP1600_ExtractBytes_Exit:
.type KeccakP1600_ExtractAndAddBytes, %function
KeccakP1600_ExtractAndAddBytes:
addq arg4, arg1
- leaq KeccakLaneComplementTable, arg6
+ leaq KeccakLaneComplementTable(%rip), arg6
addq arg4, arg6
subq $8, arg5
jc KeccakP1600_ExtractAndAddBytes_Bytes
@@ -980,7 +980,8 @@ KeccakP1600_Permute_Nrounds:
movq rT2a, _su(rpStack)
KeccakP1600_Permute_Nrounds_Dispatch:
shlq $3, rT1
- jmp *KeccakP1600_Permute_NroundsTable-8(rT1)
+ leaq KeccakP1600_Permute_NroundsTable-8(%rip), rT2a
+ jmp *(rT1, rT2a)
KeccakP1600_Permute_Nrounds24:
mKeccakRound rpState, rpStack, 0x0000000000000001, 0
@@ -1174,13 +1175,13 @@ KeccakF1600_FastLoop_Absorb_VariableLaneCountLoop:
shlq $3, arg4
movq arg3, arg2 # data pointer
xorq arg3, arg3 # offset = 0
- callq KeccakP1600_AddBytes # (void *state, const unsigned char *data, unsigned int offset, unsigned int length)
+ callq KeccakP1600_AddBytes@PLT # (void *state, const unsigned char *data, unsigned int offset, unsigned int length)
movq arg2, arg3 # updated data pointer
movq 24(%rsp), rT1a # xor trailingBits
xorq rT1a, (arg1)
popq arg1
pushq arg3
- callq KeccakP1600_Permute_24rounds
+ callq KeccakP1600_Permute_24rounds@PLT
popq arg3
popq arg2
popq arg4
diff --git a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-armcc.s b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-armcc.s
index 403058ef294de23e976125af6ba24781b07d46b1..b0e6dc83048ba6abed8344483d02a7ea3a584d55 100755
--- a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-armcc.s
+++ b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-armcc.s
@@ -224,7 +224,7 @@ Xoodoo_ExtractAndAddBytes_Exit
Xoodoo_Permute_6rounds PROC
vpush {q4-q5}
vldmia r0, {q0-q2}
- ldr r1, = _rc6
+ adr r1, _rc6
mRound
mRound
mRound
@@ -260,7 +260,7 @@ _rc6
Xoodoo_Permute_12rounds PROC
vpush {q4-q5}
vldmia r0, {q0-q2}
- ldr r1, = _rc12
+ adr r1, _rc12
mRound
mRound
mRound
@@ -371,7 +371,7 @@ Xoofff_CompressFastLoop PROC
vld1.32 {q11}, [r1]
Xoofff_CompressFastLoop_Loop
vld1.32 {q0,q1}, [r2]! ; get input
- ldr r1, = _rc6
+ adr r1, _rc6b
vld1.32 {q2}, [r2]!
veor.32 q0, q0, q6
veor.32 q1, q1, q7
@@ -403,7 +403,15 @@ Xoofff_CompressFastLoop_Loop
sub r0, r2, r6 ; return number of bytes processed
vpop {q4-q7}
pop {r4,r5,r6,pc}
+ LTORG
align 8
+_rc6b
+ dcq 0x00000060
+ dcq 0x0000002C
+ dcq 0x00000380
+ dcq 0x000000F0
+ dcq 0x000001A0
+ dcq 0x00000012
ENDP
; ----------------------------------------------------------------------------
@@ -425,7 +433,7 @@ Xoofff_ExpandFastLoop_Loop
vmov q0, q9
vmov q1, q10
vmov q2, q11
- ldr r1, = _rc6
+ adr r1, _rc6b
mRound ; permutation
mRound
mRound
diff --git a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-gcc.s b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-gcc.s
index bd6e75667e60f22dab156e082418644f5b265002..9b523ca0030607ac4e6c4a91498e22fc462cfc76 100755
--- a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-gcc.s
+++ b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodoo-uf-armv7a-neon-le-gcc.s
@@ -230,7 +230,7 @@ Xoodoo_ExtractAndAddBytes_Exit:
Xoodoo_Permute_6rounds:
vpush {q4-q5}
vldmia r0, {q0-q2}
- ldr r1, = _rc6
+ adr r1, _rc6
mRound
mRound
mRound
@@ -267,7 +267,7 @@ _rc6:
Xoodoo_Permute_12rounds:
vpush {q4-q5}
vldmia r0, {q0-q2}
- ldr r1, = _rc12
+ adr r1, _rc12
mRound
mRound
mRound
@@ -380,7 +380,7 @@ Xoofff_CompressFastLoop:
vld1.32 {q11}, [r1]
Xoofff_CompressFastLoop_Loop:
vld1.32 {q0,q1}, [r2]! @ get input
- ldr r1, = _rc6
+ adr r1, _rc6b
vld1.32 {q2}, [r2]!
veor.32 q0, q0, q6
veor.32 q1, q1, q7
@@ -412,7 +412,15 @@ Xoofff_CompressFastLoop_Loop:
sub r0, r2, r6 @ return number of bytes processed
vpop {q4-q7}
pop {r4,r5,r6,pc}
+ .ltorg
.align 8
+_rc6b:
+ .quad 0x00000060
+ .quad 0x0000002C
+ .quad 0x00000380
+ .quad 0x000000F0
+ .quad 0x000001A0
+ .quad 0x00000012
@ ----------------------------------------------------------------------------
@@ -435,7 +443,7 @@ Xoofff_ExpandFastLoop_Loop:
vmov q0, q9
vmov q1, q10
vmov q2, q11
- ldr r1, = _rc6
+ adr r1, _rc6b
mRound @ permutation
mRound
mRound
diff --git a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-armcc.s b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-armcc.s
index df5c5d31aaaa765ac8e477d9bf5ffa7c7dcc5b24..b4937f0af08e9c6264a3560d564de008dd1a8ac7 100755
--- a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-armcc.s
+++ b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-armcc.s
@@ -64,7 +64,7 @@ Xoodyak_Rhash equ 16
;
align 8
Xoodoo_Permute_12roundsAsm PROC
- ldr r1, =_rc12
+ adr r1, _rc12
mRound
mRound
mRound
diff --git a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-gcc.s b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-gcc.s
index 2b9b018f057b36ed1a85b07dd71012bad6e6182c..d701e08b1c9b029524a9faa71a303c0b65c95090 100755
--- a/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-gcc.s
+++ b/src/XKCP/lib/low/Xoodoo/OptimizedAsmARMv7A/Xoodyak-uf-armv7a-neon-le-gcc.s
@@ -64,7 +64,7 @@
.align 8
.type Xoodoo_Permute_12roundsAsm, %function;
Xoodoo_Permute_12roundsAsm:
- ldr r1, =_rc12
+ adr r1, _rc12
mRound
mRound
mRound
diff --git a/src/XKCP/tests/UnitTests/timing.h b/src/XKCP/tests/UnitTests/timing.h
index 2fdd0c141389272e7a51de54120517f2463efcb2..eed7291989f370aaea701be158ddd54f4d9a883b 100755
--- a/src/XKCP/tests/UnitTests/timing.h
+++ b/src/XKCP/tests/UnitTests/timing.h
@@ -80,7 +80,7 @@ static uint_32t HiResTime(void) /* return the current value of time st
uint32_t cycle_count;
asm volatile("MRS %0, pmevcntr0_el0" : "=r" (cycle_count));
return cycle_count;
- #elif defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A_)
+ #elif defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__)
/* Implement Aarch32 bits on PMU : Tested on Cortex a-53 */
uint32_t cycle_count;
asm volatile("MRC p15, 0, %0, c9, c13, 0 \t\n" : "=r"(cycle_count));
@@ -115,10 +115,10 @@ static uint_32t calibrate()
uint32_t r = 0;
asm volatile("mrs %0, pmcntenset_el0" : "=r" (r));
asm volatile("msr pmcntenset_el0, %0" : : "r" (r|1));
- #elif defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A_)
+ #elif defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__)
/* Implement Aarch32 bits on PMU : Tested on Cortex a-53 */
/* Enable counters in Control Register and reset cycle count and event count */
- printf("PMU32 Enable.. \n");
+ printf("PMU32 Enabled... \n");
asm volatile("MCR p15, 0, %0, c9, c12, 0" : : "r"(0x00000007));
/* Event counter selection register, which counter to access */
asm volatile("MCR p15, 0, %0, c9, c12, 5" : : "r"(0x0));