Implementation notes: amd64, waldorf, crypto_hash/echo256

Computer: waldorf
Architecture: amd64
CPU ID: GenuineIntel-000106e5-bfebfbff
SUPERCOP version: 20160715
Operation: crypto_hash
Primitive: echo256

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
77484	`core2/core2v64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
80236	`core2/core2v64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
83996	`athlon/athlon64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
84272	`core2/core2v64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
86160	`athlon/athlon64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
88676	`core2/core2v64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
90160	`athlon/athlon64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
92568	`athlon/athlon64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
99932	`generic/opt64`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
103956	`sphlib`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
104504	`sphlib`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
104516	`sphlib-small`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
105076	`sphlib`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
107680	`ccalik/vperm`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
107872	`generic/opt64`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
108956	`generic/opt64`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
109116	`sphlib`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
111844	`ccalik/vperm`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
112632	`sphlib-small`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
113952	`pentium/pentium4v64v1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
113956	`ccalik/bitsliced`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
114620	`pentium/pentium4v64v2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
114700	`ccalik/vperm`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
115220	`pentium/pentium4v64v2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
116608	`ccalik/bitsliced`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
117072	`ccalik/bitsliced`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
117868	`pentium/pentium4v64v1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
118024	`sphlib-small`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
119164	`pentium/pentium4v64v1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
119324	`pentium/pentium4v64v1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
119616	`ccalik/vperm`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
120172	`ccalik/vperm`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
120208	`sphlib`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
120408	`pentium/pentium4v64v2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
120888	`pentium/pentium4v64v2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
123028	`generic/opt64`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
123364	`generic/opt32`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
124956	`ccalik/bitsliced`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
125852	`sphlib-small`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
126268	`generic/opt32`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
126876	`generic/opt32`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
131196	`generic/opt32`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
131884	`sphlib-small`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
132900	`generic/opt32`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
134428	`ccalik/bitsliced`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
142656	`powerpc/pp32cv2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
165004	`powerpc/pp32cv2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
180544	`generic/opt64`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
182296	`powerpc/pp32cv1`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
185816	`powerpc/pp32cv2`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
194188	`powerpc/pp32cv2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
197908	`powerpc/pp32cv2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
204640	`powerpc/pp32cv1`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
243308	`powerpc/pp32cv1`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
244216	`powerpc/pp32cv1`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
255244	`powerpc/pp32cv1`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715

Test failure

Implementation: crypto_hash/echo256/aes/aes64
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

error 111

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	aes/aes64
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	aes/aes64
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	aes/aes64
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	aes/aes64

Compiler output

Implementation: crypto_hash/echo256/pentium/pentium4v64v1
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

echo32.s: echo32.c:545:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:545:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:545:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:545:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:546:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:546:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:546:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:546:2: error: invalid instruction mnemonic 'addd'
echo32.s: addd [SHA3_CNT], 1
echo32.s: ^
echo32.s: echo32.c:547:2: error: invalid instruction mnemonic 'addd'
echo32.s: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	pentium/pentium4v64v1

Compiler output

Implementation: crypto_hash/echo256/generic/opt64
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

echo64.c: echo64.c:501:29: warning: passing 'const unsigned char *' to parameter of type 'BitSequence *' (aka 'unsigned char *') discards qualifiers [-Wincompatible-pointer-types-discards-qualifiers]
echo64.c: if (Hash(CRYPTO_BYTES * 8,in,inlen * 8,out) == SUCCESS) return 0;
echo64.c: ^~
echo64.c: echo64.c:488:46: note: passing argument to parameter 'data' here
echo64.c: HashReturn Hash(int hashbitlen, BitSequence *data, DataLength databitlen, BitSequence *hashval)
echo64.c: ^
echo64.c: 1 warning generated.

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	generic/opt64

Compiler output

Implementation: crypto_hash/echo256/core2/core2v64
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, ah
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, al
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, bl
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, bh
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, al
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, ah
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, bh
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, bl
echo64.s: ^
echo64.s: echo64.c:473:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	core2/core2v64

Compiler output

Implementation: crypto_hash/echo256/athlon/athlon64
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, ah
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, al
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, bl
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, bh
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, al
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, ah
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl ebp, bh
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: movzbl esi, bl
echo64.s: ^
echo64.s: echo64.c:519:2: error: invalid instruction mnemonic 'movzbl'
echo64.s: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	athlon/athlon64

Compiler output

Implementation: crypto_hash/echo256/pentium/pentium4v64v2
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

echo64.s: echo64.c:571:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:571:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:571:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:571:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:572:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:572:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:572:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:572:2: error: invalid instruction mnemonic 'addd'
echo64.s: addd [SHA3_CNT], 1
echo64.s: ^
echo64.s: echo64.c:573:2: error: invalid instruction mnemonic 'addd'
echo64.s: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	pentium/pentium4v64v2

Compiler output

Implementation: crypto_hash/echo256/aes/aes64
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

echo64.s: echo64.c:636:14: error: invalid instruction mnemonic 'noprefix'
echo64.s: .att_syntax noprefix
echo64.s: ^

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	aes/aes64

Compiler output

Implementation: crypto_hash/echo256/ccalik/aesni
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

hash.c: In file included from hash.c:25:
hash.c: /usr/include/clang/3.5.0/include/wmmintrin.h:30:3: error: "AES/PCLMUL instructions not enabled"
hash.c: # error "AES/PCLMUL instructions not enabled"
hash.c: ^
hash.c: hash.c:295:4: warning: implicit declaration of function '_mm_aesenc_si128' is invalid in C99 [-Wimplicit-function-declaration]
hash.c: ECHO_ROUND_UNROLL2;
hash.c: ^
hash.c: hash.c:185:4: note: expanded from macro 'ECHO_ROUND_UNROLL2'
hash.c: ECHO_SUBBYTES(_state, 0, 0);\
hash.c: ^
hash.c: hash.c:114:19: note: expanded from macro 'ECHO_SUBBYTES'
hash.c: state[i][j] = _mm_aesenc_si128(state[i][j], k1);\
hash.c: ^
hash.c: hash.c:295:4: error: assigning to '__m128i' (vector of 2 'long long' values) from incompatible type 'int'
hash.c: ECHO_ROUND_UNROLL2;
hash.c: ^~~~~~~~~~~~~~~~~~
hash.c: hash.c:185:4: note: expanded from macro 'ECHO_ROUND_UNROLL2'
hash.c: ECHO_SUBBYTES(_state, 0, 0);\
hash.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~
hash.c: hash.c:114:17: note: expanded from macro 'ECHO_SUBBYTES'
hash.c: state[i][j] = _mm_aesenc_si128(state[i][j], k1);\
hash.c: ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
hash.c: hash.c:295:4: error: assigning to '__m128i' (vector of 2 'long long' values) from incompatible type 'int'
hash.c: ECHO_ROUND_UNROLL2;
hash.c: ^~~~~~~~~~~~~~~~~~
hash.c: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ccalik/aesni

Compiler output

Implementation: crypto_hash/echo256/generic/opt64
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

echo64.c: echo64.c: In function 'crypto_hash_echo256_generic_opt64':
echo64.c: echo64.c:501:29: warning: passing argument 2 of 'Hash' discards 'const' qualifier from pointer target type
echo64.c: if (Hash(CRYPTO_BYTES * 8,in,inlen * 8,out) == SUCCESS) return 0;
echo64.c: ^
echo64.c: echo64.c:488:12: note: expected 'BitSequence *' but argument is of type 'const unsigned char *'
echo64.c: HashReturn Hash(int hashbitlen, BitSequence *data, DataLength databitlen, BitSequence *hashval)
echo64.c: ^

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	generic/opt64
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	generic/opt64
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	generic/opt64
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	generic/opt64

Compiler output

Implementation: crypto_hash/echo256/ccalik/aesni
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

hash.c: In file included from hash.c:25:0:
hash.c: hash.c: In function 'Compress':
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/wmmintrin.h:61:1: error: inlining failed in call to always_inline '_mm_aesenc_si128': target specific option mismatch
hash.c: _mm_aesenc_si128 (__m128i __X, __m128i __Y)
hash.c: ^
hash.c: hash.c:114:19: error: called from here
hash.c: state[i][j] = _mm_aesenc_si128(state[i][j], k1);\
hash.c: ^
hash.c: hash.c:185:4: note: in expansion of macro 'ECHO_SUBBYTES'
hash.c: ECHO_SUBBYTES(_state, 0, 0);\
hash.c: ^
hash.c: hash.c:295:4: note: in expansion of macro 'ECHO_ROUND_UNROLL2'
hash.c: ECHO_ROUND_UNROLL2;
hash.c: ^
hash.c: In file included from hash.c:25:0:
hash.c: /usr/lib/gcc/x86_64-linux-gnu/4.9/include/wmmintrin.h:61:1: error: inlining failed in call to always_inline '_mm_aesenc_si128': target specific option mismatch
hash.c: _mm_aesenc_si128 (__m128i __X, __m128i __Y)
hash.c: ^
hash.c: hash.c:115:19: error: called from here
hash.c: state[i][j] = _mm_aesenc_si128(state[i][j], M128(zero));\
hash.c: ^
hash.c: hash.c:185:4: note: in expansion of macro 'ECHO_SUBBYTES'
hash.c: ECHO_SUBBYTES(_state, 0, 0);\
hash.c: ^
hash.c: hash.c:295:4: note: in expansion of macro 'ECHO_ROUND_UNROLL2'
hash.c: ...

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

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