Implementation notes: amd64, h4atom, crypto_hash/keccak

Computer: h4atom
Architecture: amd64
CPU ID: GenuineIntel-000106ca-bfe9fbff
SUPERCOP version: 20160806
Operation: crypto_hash
Primitive: keccak

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
37728	`opt64lcu24`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
38200	`opt64lcu6`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
40584	`opt64lcu24`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
41784	`opt64u6`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
42136	`opt64lcu6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
43784	`opt64lcu6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
44080	`opt64lcu24`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
44112	`opt64lcu24`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
44752	`opt64lcu6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
45560	`simple`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
45784	`opt64u6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
45936	`inplace`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
46736	`simple`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
47544	`simple`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
48408	`inplace`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
48504	`x86_64_asm`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
48504	`x86_64_asm`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
48560	`x86_64_asm`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
48800	`x86_64_asm`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
49288	`opt64u6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
49416	`inplace`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
50176	`opt64u6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
51296	`inplace`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
51440	`simple`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
56736	`opt64lcu24`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
58376	`opt64lcu6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
60808	`sseu2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
64768	`sseu2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
64776	`sseu2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
67624	`sseu2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
68360	`simple`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
69032	`opt64u6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
70664	`inplace`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
79928	`sseu2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
85568	`mmxu1`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
86968	`mmxu1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
88184	`mmxu1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
88208	`mmxu1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
94808	`opt32bi-s2lcu4`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
98880	`mmxu1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
101952	`simple32bi`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
102552	`opt32bi-rvku2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
104544	`inplace32bi`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
105952	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
106512	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
107936	`opt32bi-rvku2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
108840	`simple32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
110584	`opt32biT-s2lcu4`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
114888	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
115888	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
116472	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
119472	`inplace32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
120792	`simple32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
121176	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
123288	`compact`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
124960	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
125224	`simple32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
126184	`inplace32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
127048	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
131832	`opt64lcu24shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
135152	`opt64lcu24shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
135192	`opt64lcu24shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
135208	`inplace32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
137384	`opt64lcu24shld`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
139056	`x86_64_shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
139072	`x86_64_shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
139160	`x86_64_shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
139344	`x86_64_shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
143760	`opt64lcu24shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
145816	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
156176	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
169136	`simple32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
169568	`inplace32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
171088	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
171128	`compact`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
235944	`compact`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
261552	`compact`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
291216	`compact`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806
400104	`compact8`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160812	20160806
470632	`compact8`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160812	20160806
493280	`compact8`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160812	20160806
552824	`compact8`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160812	20160806
611720	`compact8`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160812	20160806

Compiler output

Implementation: crypto_hash/keccak/xopu24
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: error: '__builtin_ia32_vprotqi' needs target feature xop
KeccakF-1600-opt64.c: rounds
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ./KeccakF-1600-unrolling.macros:17:5: note: expanded from macro 'rounds'
KeccakF-1600-opt64.c: thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:53:5: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
KeccakF-1600-opt64.c: computeD \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:40:23: note: expanded from macro 'computeD'
KeccakF-1600-opt64.c: Dei = XOR128(Cae, ROL6464same(Cio, 1)); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:102:33: note: expanded from macro 'ROL6464same'
KeccakF-1600-opt64.c: #define ROL6464same(a, o) _mm_roti_epi64(a, o)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: /usr/lib/llvm-3.8/bin/../lib/clang/3.8.0/include/xopintrin.h:250:12: note: expanded from macro '_mm_roti_epi64'
KeccakF-1600-opt64.c: (__m128i)__builtin_ia32_vprotqi((__v2di)(__m128i)(A), (N)); })
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: error: '__builtin_ia32_vprotqi' needs target feature xop
KeccakF-1600-opt64.c: ./KeccakF-1600-unrolling.macros:17:5: note: expanded from macro 'rounds'
KeccakF-1600-opt64.c: thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ./KeccakF-1600-xop.macros:53:5: note: expanded from macro 'thetaRhoPiChiIotaPrepareTheta'
KeccakF-1600-opt64.c: computeD \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 1, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	xopu24

Compiler output

Implementation: crypto_hash/keccak/x86_64_asm
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:50:16: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ apState, %rdi
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:50:16: error: missing expression
KeccakF-1600-x86-64-gas.s: .equ apState, %rdi
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:52:18: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ aNbrWords, %rdx
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:52:18: error: missing expression
KeccakF-1600-x86-64-gas.s: .equ aNbrWords, %rdx
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:55:16: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ xpState, %r9
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:55:16: error: missing expression
KeccakF-1600-x86-64-gas.s: .equ xpState, %r9
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:58:12: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ rT1, %rax
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:58:12: error: missing expression
KeccakF-1600-x86-64-gas.s: .equ rT1, %rax
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:60:16: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: ...

Number of similar (compiler,implementation) pairs: 1, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	x86_64_asm

Compiler output

Implementation: crypto_hash/keccak/x86_64_shld
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:50:16: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ apState, %rdi
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:50:16: error: missing expression
KeccakF-1600-x86-64-shld-gas.s: .equ apState, %rdi
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:52:18: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ aNbrWords, %rdx
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:52:18: error: missing expression
KeccakF-1600-x86-64-shld-gas.s: .equ aNbrWords, %rdx
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:55:16: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ xpState, %r9
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:55:16: error: missing expression
KeccakF-1600-x86-64-shld-gas.s: .equ xpState, %r9
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:58:12: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ rT1, %rax
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:58:12: error: missing expression
KeccakF-1600-x86-64-shld-gas.s: .equ rT1, %rax
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:60:16: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: ...

Number of similar (compiler,implementation) pairs: 1, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	x86_64_shld

Compiler output

Implementation: crypto_hash/keccak/xopu24
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/5/include/x86intrin.h:41:0,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:74:
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c: In function 'KeccakPermutationOnWords':
KeccakF-1600-opt64.c: /usr/lib/gcc/x86_64-linux-gnu/5/include/smmintrin.h:216:1: error: inlining failed in call to always_inline '_mm_blend_pd': target specific option mismatch
KeccakF-1600-opt64.c: _mm_blend_pd (__m128d __X, __m128d __Y, const int __M)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:97:43: error: called from here
KeccakF-1600-opt64.c: #define GET64LOHI(a, b) ((__m128i)_mm_blend_pd((__m128d)a, (__m128d)b, 2))
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:94:51: note: in definition of macro 'SWAP64'
KeccakF-1600-opt64.c: #define SWAP64(a) _mm_shuffle_epi32(a, 0x4E)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:98:40: note: in expansion of macro 'GET64LOHI'
KeccakF-1600-opt64.c: #define GET64HILO(a, b) SWAP64(GET64LOHI(b, a))
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:106:13: note: in expansion of macro 'GET64HILO'
KeccakF-1600-opt64.c: Bsosu = GET64HILO(Bsiso, Bsusa); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:123:36: note: in expansion of macro 'thetaRhoPiChiIotaPrepareTheta'
KeccakF-1600-opt64.c: #define thetaRhoPiChiIota(i, A, E) thetaRhoPiChiIotaPrepareTheta(i, A, E)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-unrolling.macros:40:5: note: in expansion of macro 'thetaRhoPiChiIota'
KeccakF-1600-opt64.c: thetaRhoPiChiIota(23, E, A) \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: note: in expansion of macro 'rounds'
KeccakF-1600-opt64.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	xopu24
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	xopu24
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	xopu24
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	xopu24