Implementation notes: amd64, skylake, crypto_aead/aes128poetv1aes128

Computer: skylake
Architecture: amd64
CPU ID: GenuineIntel-000506e3-bfebfbff
SUPERCOP version: 20161026
Operation: crypto_aead
Primitive: aes128poetv1aes128

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
16140	`ni`	`clang -O3 -fwrapv -march=native -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
16172	`ni`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161216	20161026
16194	`ni`	`clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
16238	`ni`	`clang -O3 -fwrapv -march=x86-64 -mcpu=core-avx2 -mavx2 -maes -mpclmul -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
30648	`ni`	`gcc -m64 -march=core-avx2 -O3 -fomit-frame-pointer`	20161216	20161026
30654	`ni`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20161216	20161026
30714	`ni`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20161216	20161026
30880	`ni`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20161216	20161026
30888	`ni`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20161216	20161026
31124	`ni`	`gcc -m64 -march=core-avx2 -O2 -fomit-frame-pointer`	20161216	20161026
31756	`ni`	`gcc -m64 -march=core-avx-i -O3 -fomit-frame-pointer`	20161216	20161026
31818	`ni`	`gcc -m64 -march=corei7-avx -O3 -fomit-frame-pointer`	20161216	20161026
31926	`ni`	`gcc -m64 -march=corei7-avx -O2 -fomit-frame-pointer`	20161216	20161026
32132	`ni`	`gcc -m64 -march=core-avx-i -O2 -fomit-frame-pointer`	20161216	20161026
33046	`ni`	`gcc -m64 -march=corei7-avx -O -fomit-frame-pointer`	20161216	20161026
33072	`ni`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20161216	20161026
33084	`ni`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20161216	20161026
33210	`ni`	`gcc -m64 -march=core-avx2 -O -fomit-frame-pointer`	20161216	20161026
33224	`ni`	`gcc -m64 -march=core-avx-i -O -fomit-frame-pointer`	20161216	20161026
43348	`ni`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20161216	20161026
43368	`ni`	`gcc -m64 -march=corei7-avx -Os -fomit-frame-pointer`	20161216	20161026
43494	`ni`	`gcc -m64 -march=core-avx-i -Os -fomit-frame-pointer`	20161216	20161026
43682	`ni`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20161216	20161026
44064	`ni`	`gcc -m64 -march=core-avx2 -Os -fomit-frame-pointer`	20161216	20161026
155278	`ref`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20161216	20161026
156128	`ref`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20161216	20161026
156404	`ref`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20161216	20161026
156524	`ref`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20161216	20161026
156858	`ref`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20161216	20161026
156906	`ref`	`gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer`	20161216	20161026
156920	`ref`	`gcc -m64 -march=corei7-avx -O3 -fomit-frame-pointer`	20161216	20161026
156982	`ref`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20161216	20161026
157236	`ref`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20161216	20161026
157474	`ref`	`gcc -m64 -march=corei7 -O3 -fomit-frame-pointer`	20161216	20161026
158280	`ref`	`gcc -m64 -march=core-avx2 -O3 -fomit-frame-pointer`	20161216	20161026
158458	`ref`	`gcc -m64 -march=core-avx-i -O3 -fomit-frame-pointer`	20161216	20161026
158578	`ref`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20161216	20161026
159700	`ref`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20161216	20161026
159816	`ref`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20161216	20161026
160366	`ref`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20161216	20161026
160450	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20161216	20161026
160460	`ref`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20161216	20161026
160508	`ref`	`gcc -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer`	20161216	20161026
161272	`ref`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20161216	20161026
161276	`ref`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20161216	20161026
161374	`ref`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20161216	20161026
161708	`ref`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20161216	20161026
161906	`ref`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20161216	20161026
161974	`ref`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20161216	20161026
162328	`ref`	`gcc -O3 -fomit-frame-pointer`	20161216	20161026
162576	`ref`	`gcc -m64 -O3 -fomit-frame-pointer`	20161216	20161026
162642	`ref`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20161216	20161026
162736	`ref`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20161216	20161026
162904	`ref`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20161216	20161026
163100	`ref`	`gcc -funroll-loops -O -fomit-frame-pointer`	20161216	20161026
163362	`ref`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20161216	20161026
163538	`ref`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20161216	20161026
163646	`ref`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20161216	20161026
163776	`ref`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20161216	20161026
163786	`ref`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20161216	20161026
165062	`ref`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20161216	20161026
165130	`ref`	`gcc -m64 -march=core2 -O3 -fomit-frame-pointer`	20161216	20161026
165210	`ref`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20161216	20161026
165788	`ref`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20161216	20161026
166286	`ref`	`gcc -O2 -fomit-frame-pointer`	20161216	20161026
166436	`ref`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20161216	20161026
166438	`ref`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20161216	20161026
166518	`ref`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20161216	20161026
167608	`ref`	`gcc -m64 -march=core-avx-i -O2 -fomit-frame-pointer`	20161216	20161026
167792	`ref`	`gcc -m64 -O2 -fomit-frame-pointer`	20161216	20161026
167948	`ref`	`clang -O3 -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
168008	`ref`	`clang -mcpu=native -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161216	20161026
168104	`ref`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20161216	20161026
168510	`ref`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20161216	20161026
168512	`ref`	`gcc -m64 -march=corei7-avx -O2 -fomit-frame-pointer`	20161216	20161026
168672	`ref`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20161216	20161026
169400	`ref`	`clang -mcpu=cortex-a9 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161216	20161026
169538	`ref`	`clang -O3 -fwrapv -march=x86-64 -mcpu=core-avx2 -mavx2 -maes -mpclmul -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
169552	`ref`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20161216	20161026
169574	`ref`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20161216	20161026
169708	`ref`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20161216	20161026
169864	`ref`	`clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
169874	`ref`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20161216	20161026
169906	`ref`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20161216	20161026
170108	`ref`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20161216	20161026
170184	`ref`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20161216	20161026
170268	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20161216	20161026
170528	`ref`	`clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
170556	`ref`	`gcc -m64 -march=corei7 -O2 -fomit-frame-pointer`	20161216	20161026
170592	`ref`	`clang -mcpu=cortex-a8 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161216	20161026
170806	`ref`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20161216	20161026
171054	`ref`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20161216	20161026
171086	`ref`	`clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
171202	`ref`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20161216	20161026
171246	`ref`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20161216	20161026
171438	`ref`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20161216	20161026
171566	`ref`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20161216	20161026
171750	`ref`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20161216	20161026
172062	`ref`	`gcc -m64 -march=core-avx2 -O2 -fomit-frame-pointer`	20161216	20161026
172156	`ref`	`gcc -m64 -march=corei7-avx -Os -fomit-frame-pointer`	20161216	20161026
172204	`ref`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20161216	20161026
172304	`ref`	`gcc -m64 -march=core-avx-i -Os -fomit-frame-pointer`	20161216	20161026
172332	`ref`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20161216	20161026
172684	`ref`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20161216	20161026
172738	`ref`	`gcc -O -fomit-frame-pointer`	20161216	20161026
172762	`ref`	`gcc -march=k8 -O -fomit-frame-pointer`	20161216	20161026
173054	`ref`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20161216	20161026
173094	`ref`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20161216	20161026
173166	`ref`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20161216	20161026
173300	`ref`	`gcc -march=nocona -Os -fomit-frame-pointer`	20161216	20161026
173426	`ref`	`gcc -m64 -O -fomit-frame-pointer`	20161216	20161026
173458	`ref`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20161216	20161026
173630	`ref`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20161216	20161026
173646	`ref`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20161216	20161026
173674	`ref`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20161216	20161026
173676	`ref`	`gcc -m64 -march=core-avx2 -Os -fomit-frame-pointer`	20161216	20161026
173746	`ref`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20161216	20161026
173748	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20161216	20161026
173806	`ref`	`gcc -march=k8 -Os -fomit-frame-pointer`	20161216	20161026
173816	`ref`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20161216	20161026
173862	`ref`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20161216	20161026
173926	`ref`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20161216	20161026
173928	`ref`	`gcc -Os -fomit-frame-pointer`	20161216	20161026
173938	`ref`	`gcc -m64 -Os -fomit-frame-pointer`	20161216	20161026
173982	`ref`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20161216	20161026
173986	`ref`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20161216	20161026
174042	`ref`	`gcc -m64 -march=core-avx-i -O -fomit-frame-pointer`	20161216	20161026
174046	`ref`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20161216	20161026
174314	`ref`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20161216	20161026
174388	`ref`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20161216	20161026
174646	`ref`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20161216	20161026
174762	`ref`	`gcc -m64 -march=corei7-avx -O -fomit-frame-pointer`	20161216	20161026
174856	`ref`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20161216	20161026
174940	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20161216	20161026
175074	`ref`	`gcc -m64 -march=core-avx2 -O -fomit-frame-pointer`	20161216	20161026
175478	`ref`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20161216	20161026
176042	`ref`	`gcc -march=barcelona -O -fomit-frame-pointer`	20161216	20161026
176044	`ref`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20161216	20161026
176460	`ref`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20161216	20161026
177400	`ref`	`clang -O3 -fwrapv -march=native -fomit-frame-pointer -Qunused-arguments`	20161216	20161026
178830	`ref`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161216	20161026
179062	`ref`	`gcc -march=nocona -O -fomit-frame-pointer`	20161216	20161026
418684	`ref`	`cc`	20161216	20161026
423542	`ref`	`gcc`	20161216	20161026
424428	`ref`	`gcc -funroll-loops`	20161216	20161026

Compiler output

Implementation: crypto_aead/aes128poetv1aes128/ni
Compiler: cc

poet.c: In file included from poet.c:31:0:
poet.c: poet.c: In function 'aes_encrypt':
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:69:1: error: inlining failed in call to always_inline '_mm_aesenclast_si128': target specific option mismatch
poet.c: _mm_aesenclast_si128 (__m128i __X, __m128i __Y)
poet.c: ^~~~~~~~~~~~~~~~~~~~
poet.c: poet.c:136:12: note: called from here
poet.c: return _mm_aesenclast_si128(x, k[10]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:61:1: error: inlining failed in call to always_inline '_mm_aesenc_si128': target specific option mismatch
poet.c: _mm_aesenc_si128 (__m128i __X, __m128i __Y)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:135:7: note: called from here
poet.c: x = _mm_aesenc_si128(x, k[9]);
poet.c: ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:61:1: error: inlining failed in call to always_inline '_mm_aesenc_si128': target specific option mismatch
poet.c: _mm_aesenc_si128 (__m128i __X, __m128i __Y)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:134:7: note: called from here
poet.c: x = _mm_aesenc_si128(x, k[8]);
poet.c: ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:61:1: error: inlining failed in call to always_inline '_mm_aesenc_si128': target specific option mismatch
poet.c: _mm_aesenc_si128 (__m128i __X, __m128i __Y)
poet.c: ...

Number of similar (compiler,implementation) pairs: 3, namely:

Compiler	Implementations
cc	ni
gcc	ni
gcc -funroll-loops	ni

Compiler output

Implementation: crypto_aead/aes128poetv1aes128/ni
Compiler: clang -O3 -fomit-frame-pointer -Qunused-arguments

poet.c: poet.c:701:16: error: always_inline function '_mm_testc_si128' requires target feature 'ssse3', but would be inlined into function 'decrypt_final' that is compiled without support for 'ssse3'
poet.c: return _mm_testc_si128(t, p) - 1;
poet.c: ^
poet.c: poet.c:713:16: error: always_inline function '_mm_testc_si128' requires target feature 'ssse3', but would be inlined into function 'decrypt_final' that is compiled without support for 'ssse3'
poet.c: return _mm_testc_si128(t, p) - 1;
poet.c: ^
poet.c: poet.c:728:21: error: always_inline function '_mm_testc_si128' requires target feature 'ssse3', but would be inlined into function 'decrypt_final' that is compiled without support for 'ssse3'
poet.c: int alpha = _mm_testc_si128(tau_alpha, tmp);
poet.c: ^
poet.c: poet.c:738:20: error: always_inline function '_mm_testc_si128' requires target feature 'ssse3', but would be inlined into function 'decrypt_final' that is compiled without support for 'ssse3'
poet.c: int beta = _mm_testc_si128(t_beta, tmp);
poet.c: ^
poet.c: 4 errors generated.

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

Compiler	Implementations
clang -O3 -fomit-frame-pointer -Qunused-arguments	ni
clang -mcpu=cortex-a8 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ni
clang -mcpu=cortex-a9 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ni
clang -mcpu=native -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ni

Compiler output

Implementation: crypto_aead/aes128poetv1aes128/ni
Compiler: clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments

poet.c: poet.c:81:13: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
poet.c: temp2 = _mm_aeskeygenassist_si128(temp1, 0x1);
poet.c: ^
poet.c: /usr/bin/../lib/clang/3.9.0/include/__wmmintrin_aes.h:147:12: note: expanded from macro '_mm_aeskeygenassist_si128'
poet.c: (__m128i)__builtin_ia32_aeskeygenassist128((__v2di)(__m128i)(C), (int)(R))
poet.c: ^
poet.c: poet.c:85:13: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
poet.c: temp2 = _mm_aeskeygenassist_si128(temp1, 0x2);
poet.c: ^
poet.c: /usr/bin/../lib/clang/3.9.0/include/__wmmintrin_aes.h:147:12: note: expanded from macro '_mm_aeskeygenassist_si128'
poet.c: (__m128i)__builtin_ia32_aeskeygenassist128((__v2di)(__m128i)(C), (int)(R))
poet.c: ^
poet.c: poet.c:89:13: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
poet.c: temp2 = _mm_aeskeygenassist_si128(temp1, 0x4);
poet.c: ^
poet.c: /usr/bin/../lib/clang/3.9.0/include/__wmmintrin_aes.h:147:12: note: expanded from macro '_mm_aeskeygenassist_si128'
poet.c: (__m128i)__builtin_ia32_aeskeygenassist128((__v2di)(__m128i)(C), (int)(R))
poet.c: ^
poet.c: poet.c:93:13: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
poet.c: temp2 = _mm_aeskeygenassist_si128(temp1, 0x8);
poet.c: ^
poet.c: /usr/bin/../lib/clang/3.9.0/include/__wmmintrin_aes.h:147:12: note: expanded from macro '_mm_aeskeygenassist_si128'
poet.c: (__m128i)__builtin_ia32_aeskeygenassist128((__v2di)(__m128i)(C), (int)(R))
poet.c: ^
poet.c: poet.c:97:13: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
poet.c: ...

Number of similar (compiler,implementation) pairs: 2, namely:

Compiler	Implementations
clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments	ni
clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments	ni

Compiler output

Implementation: crypto_aead/aes128poetv1aes128/ni
Compiler: gcc -O2 -fomit-frame-pointer

Number of similar (compiler,implementation) pairs: 84, namely:

Compiler	Implementations
gcc -O2 -fomit-frame-pointer	ni
gcc -O3 -fomit-frame-pointer	ni
gcc -O -fomit-frame-pointer	ni
gcc -Os -fomit-frame-pointer	ni
gcc -fno-schedule-insns -O2 -fomit-frame-pointer	ni
gcc -fno-schedule-insns -O3 -fomit-frame-pointer	ni
gcc -fno-schedule-insns -O -fomit-frame-pointer	ni
gcc -fno-schedule-insns -Os -fomit-frame-pointer	ni
gcc -funroll-loops -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -O -fomit-frame-pointer	ni
gcc -funroll-loops -Os -fomit-frame-pointer	ni
gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer	ni
gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -O -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -Os -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer	ni
gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer	ni
gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer	ni
gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer	ni
gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -march=k8 -O -fomit-frame-pointer	ni
gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer	ni
gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer	ni
gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer	ni
gcc -funroll-loops -march=nocona -O -fomit-frame-pointer	ni
gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer	ni
gcc -m64 -O2 -fomit-frame-pointer	ni
gcc -m64 -O3 -fomit-frame-pointer	ni
gcc -m64 -O -fomit-frame-pointer	ni
gcc -m64 -Os -fomit-frame-pointer	ni
gcc -m64 -march=core2 -O2 -fomit-frame-pointer	ni
gcc -m64 -march=core2 -O3 -fomit-frame-pointer	ni
gcc -m64 -march=core2 -O -fomit-frame-pointer	ni
gcc -m64 -march=core2 -Os -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4.1 -O3 -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4 -O3 -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer	ni
gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer	ni
gcc -m64 -march=corei7 -O2 -fomit-frame-pointer	ni
gcc -m64 -march=corei7 -O3 -fomit-frame-pointer	ni
gcc -m64 -march=corei7 -O -fomit-frame-pointer	ni
gcc -m64 -march=corei7 -Os -fomit-frame-pointer	ni
gcc -m64 -march=k8 -O2 -fomit-frame-pointer	ni
gcc -m64 -march=k8 -O3 -fomit-frame-pointer	ni
gcc -m64 -march=k8 -O -fomit-frame-pointer	ni
gcc -m64 -march=k8 -Os -fomit-frame-pointer	ni
gcc -m64 -march=nocona -O2 -fomit-frame-pointer	ni
gcc -m64 -march=nocona -O3 -fomit-frame-pointer	ni
gcc -m64 -march=nocona -O -fomit-frame-pointer	ni
gcc -m64 -march=nocona -Os -fomit-frame-pointer	ni
gcc -march=barcelona -O2 -fomit-frame-pointer	ni
gcc -march=barcelona -O3 -fomit-frame-pointer	ni
gcc -march=barcelona -O -fomit-frame-pointer	ni
gcc -march=barcelona -Os -fomit-frame-pointer	ni
gcc -march=k8 -O2 -fomit-frame-pointer	ni
gcc -march=k8 -O3 -fomit-frame-pointer	ni
gcc -march=k8 -O -fomit-frame-pointer	ni
gcc -march=k8 -Os -fomit-frame-pointer	ni
gcc -march=nocona -O2 -fomit-frame-pointer	ni
gcc -march=nocona -O3 -fomit-frame-pointer	ni
gcc -march=nocona -O -fomit-frame-pointer	ni
gcc -march=nocona -Os -fomit-frame-pointer	ni

Compiler output

Implementation: crypto_aead/aes128poetv1aes128/ni
Compiler: gcc -m64 -march=barcelona -O2 -fomit-frame-pointer

poet.c: In file included from poet.c:31:0:
poet.c: poet.c: In function 'aes_revert_key':
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:51:18: note: called from here
poet.c: dec_key[1] = _mm_aesimc_si128(enc_key[9]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:50:18: note: called from here
poet.c: dec_key[2] = _mm_aesimc_si128(enc_key[8]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:49:18: note: called from here
poet.c: dec_key[3] = _mm_aesimc_si128(enc_key[7]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ...
poet.c: In file included from poet.c:31:0:
poet.c: poet.c: In function 'aes_revert_key':
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:51:18: note: called from here
poet.c: dec_key[1] = _mm_aesimc_si128(enc_key[9]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:50:18: note: called from here
poet.c: dec_key[2] = _mm_aesimc_si128(enc_key[8]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ^~~~~~~~~~~~~~~~
poet.c: poet.c:49:18: note: called from here
poet.c: dec_key[3] = _mm_aesimc_si128(enc_key[7]);
poet.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
poet.c: In file included from poet.c:31:0:
poet.c: /usr/lib/gcc/x86_64-pc-linux-gnu/6.2.1/include/wmmintrin.h:77:1: error: inlining failed in call to always_inline '_mm_aesimc_si128': target specific option mismatch
poet.c: _mm_aesimc_si128 (__m128i __X)
poet.c: ...

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

Compiler	Implementations
gcc -m64 -march=barcelona -O2 -fomit-frame-pointer	ni
gcc -m64 -march=barcelona -O3 -fomit-frame-pointer	ni
gcc -m64 -march=barcelona -O -fomit-frame-pointer	ni
gcc -m64 -march=barcelona -Os -fomit-frame-pointer	ni