Implementation notes: amd64, waldorf, crypto_hash/keccak

Computer: waldorf
Architecture: amd64
CPU ID: GenuineIntel-000106e5-bfebfbff
SUPERCOP version: 20160715
Operation: crypto_hash
Primitive: keccak

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
45780	`opt64lcu24shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
46176	`opt64lcu6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
47500	`x86_64_asm`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
47848	`opt64lcu24`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
47928	`opt64lcu6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
48920	`x86_64_shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
49016	`opt64u6`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
49424	`opt64lcu6`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
49636	`opt64lcu24`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
50116	`x86_64_asm`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
50120	`opt64lcu24`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
50496	`x86_64_shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
50536	`opt64lcu24shld`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
50624	`opt64lcu24`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
50720	`x86_64_shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
50956	`simple`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
51508	`opt64lcu24shld`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
51708	`opt64lcu24`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
51876	`x86_64_asm`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
52028	`simple`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
52700	`opt64lcu6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
52708	`inplace`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
53020	`x86_64_asm`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
53032	`opt64u6`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
53212	`x86_64_shld`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
53716	`opt64u6`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
53728	`opt64u6`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
53804	`inplace`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
53892	`simple`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
54320	`opt64lcu6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
54488	`sseu2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
54640	`inplace`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
54872	`sseu2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
55412	`inplace`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
55456	`inplace`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
55632	`sseu2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
55768	`opt64lcu24shld`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
55972	`sseu2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
57972	`simple`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
58152	`opt64lcu24shld`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
58328	`opt64u6`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
58580	`simple`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
61884	`mmxu1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
63128	`sseu2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
64792	`mmxu1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
65980	`mmxu1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
68736	`mmxu1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
83032	`mmxu1`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
100908	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
104900	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
107760	`opt32bi-rvku2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
110512	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
111828	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
112372	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
113644	`inplace32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
116600	`opt32biT-s2lcu4`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
118736	`opt32biT-s2lcu4`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
119920	`simple32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
119976	`compact`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
120428	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
121540	`opt32bi-s2lcu4`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
121920	`inplace32bi`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
122256	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
123976	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
127344	`simple32bi`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
129188	`opt32bi-rvku2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
130192	`opt32bi-s2lcu4`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
130252	`simple32bi`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
130568	`opt32bi-rvku2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
131880	`inplace32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
132852	`simple32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
132884	`simple32bi`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
134808	`inplace32bi`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
135832	`compact`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
140508	`inplace32bi`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
213576	`compact`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
222240	`compact`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
228204	`compact`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
397852	`compact8`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
476768	`compact8`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
491180	`compact8`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
504612	`compact8`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
937500	`compact8`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715

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: warning: implicit declaration of function '_mm_roti_epi64' is invalid in C99 [-Wimplicit-function-declaration]
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: KeccakF-1600-opt64.c:90:50: note: expanded from macro 'XOR128'
KeccakF-1600-opt64.c: #define XOR128(a, b) _mm_xor_si128(a, b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: error: passing 'int' to parameter of incompatible type '__m128i' (vector of 2 'long long' values)
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: ...

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:51:16: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ apInput, %rsi
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: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: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:59:16: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ rpState, %rdi
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: .equ rpStack, %rsp
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:62:12: error: unknown token in expression
KeccakF-1600-x86-64-gas.s: .equ rDa, %rbx
KeccakF-1600-x86-64-gas.s: ^
KeccakF-1600-x86-64-gas.s: KeccakF-1600-x86-64-gas.s:63:12: 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:51:16: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ apInput, %rsi
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: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: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:59:16: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ rpState, %rdi
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: .equ rpStack, %rsp
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:62:12: error: unknown token in expression
KeccakF-1600-x86-64-shld-gas.s: .equ rDa, %rbx
KeccakF-1600-x86-64-shld-gas.s: ^
KeccakF-1600-x86-64-shld-gas.s: KeccakF-1600-x86-64-shld-gas.s:63:12: 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/4.9/include/x86intrin.h:52:0,
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/4.9/include/xopintrin.h:266:1: error: inlining failed in call to always_inline '_mm_roti_epi64': target specific option mismatch
KeccakF-1600-opt64.c: _mm_roti_epi64(__m128i __A, const int __B)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:90:33: error: called from here
KeccakF-1600-opt64.c: #define XOR128(a, b) _mm_xor_si128(a, b)
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:40:11: note: in expansion of macro 'XOR128'
KeccakF-1600-opt64.c: Dei = XOR128(Cae, ROL6464same(Cio, 1)); \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-xop.macros:53:5: note: in expansion of macro 'computeD'
KeccakF-1600-opt64.c: computeD \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-unrolling.macros:17:5: note: in expansion of macro 'thetaRhoPiChiIotaPrepareTheta'
KeccakF-1600-opt64.c: thetaRhoPiChiIotaPrepareTheta( 0, A, E) \
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: KeccakF-1600-opt64.c:185:5: note: in expansion of macro 'rounds'
KeccakF-1600-opt64.c: rounds
KeccakF-1600-opt64.c: ^
KeccakF-1600-opt64.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/4.9/include/x86intrin.h:52:0,
KeccakF-1600-opt64.c: from KeccakF-1600-opt64.c:74:
KeccakF-1600-opt64.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/xopintrin.h:266:1: error: inlining failed in call to always_inline '_mm_roti_epi64': target specific option mismatch
KeccakF-1600-opt64.c: _mm_roti_epi64(__m128i __A, const int __B)
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