Implementation notes: amd64, hydra4, crypto_aead/deoxysi128v141

Computer: hydra4
Microarchitecture: amd64; K10 32nm (300f10)
Architecture: amd64
CPU ID: AuthenticAMD-00300f10-178bfbff
SUPERCOP version: 20240107
Operation: crypto_aead
Primitive: deoxysi128v141

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
176283	34141 0 547	47542 804 1592	`T:table`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
187511	35661 0 592	50480 780 1688	`T:table`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
187629	35004 0 547	51368 812 1592	`T:table`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
188197	35004 0 547	51368 812 1592	`T:table`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
189297	37301 0 592	53288 780 1688	`T:table`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
192262	34971 0 547	50504 812 1592	`T:table`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
194255	32949 0 592	45923 756 1656	`T:table`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
196769	34425 0 592	48656 780 1688	`T:table`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
214622	33622 0 547	47646 804 1592	`T:table`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
1186492	26664 0 0	43064 812 1016	`T:ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
1190587	26664 0 0	43064 812 1016	`T:ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
1336859	29589 0 0	45616 780 1080	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
1372252	24983 0 0	40536 812 1016	`T:ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
2972628	25615 0 0	40472 780 1080	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
3583137	23174 0 0	37222 804 1016	`T:ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
3654115	22472 0 0	35435 756 1048	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
3677913	23994 0 0	37414 804 1016	`T:ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
3738908	23860 0 0	38088 780 1080	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212

Compiler output

Implementation: T:aesni
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

deoxys.c: deoxys.c:98:13: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'ssse3'
deoxys.c: tmp = permute( tmp, H_PERMUTATION );
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:7:22: note: expanded from macro 'permute'
deoxys.c: #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: ^
deoxys.c: deoxys.c:105:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'ssse3'
deoxys.c: TWEAKEY_SCHEDULE2( tsubkeys,subkeys, key);
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:39:3: note: expanded from macro 'TWEAKEY_SCHEDULE2'
deoxys.c: ONE_KEY_ROUND( subkeys[ 0], subkeys[ 1] ); ts[ 1] = xor( subkeys[ 1], RCONS[ 1] ); \
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:35:13: note: expanded from macro 'ONE_KEY_ROUND'
deoxys.c: new_key = permute( new_key, H_PERMUTATION);
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:7:22: note: expanded from macro 'permute'
deoxys.c: #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: ^
deoxys.c: deoxys.c:105:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'ssse3'
deoxys.c: ./tweakable-cipher.macros:40:3: note: expanded from macro 'TWEAKEY_SCHEDULE2'
deoxys.c: ONE_KEY_ROUND( subkeys[ 1], subkeys[ 2] ); ts[ 2] = xor( subkeys[ 2], RCONS[ 2] ); \
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:35:13: note: expanded from macro 'ONE_KEY_ROUND'
deoxys.c: new_key = permute( new_key, H_PERMUTATION);
deoxys.c: ^
deoxys.c: ...

Number of similar (compiler,implementation) pairs: 5, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesni
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesni
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesni
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesni
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesni

Compiler output

Implementation: T:aesni
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

deoxys.c: In file included from deoxys.c:29:
deoxys.c: deoxys.c: In function 'deoxys_aead_encrypt':
deoxys.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
deoxys.c: | ^~~~~~~~~~~~~~~~
deoxys.c: In file included from deoxys.c:30:
deoxys.c: tweakable-cipher.macros:7:22: note: called from here
deoxys.c: 7 | #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: | ^~~~~~~~~~~~~~~~~~~~~
deoxys.c: deoxys.c:98:13: note: in expansion of macro 'permute'
deoxys.c: 98 | tmp = permute( tmp, H_PERMUTATION );
deoxys.c: | ^~~~~~~
deoxys.c: In file included from deoxys.c:29:
deoxys.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
deoxys.c: | ^~~~~~~~~~~~~~~~
deoxys.c: In file included from deoxys.c:30:
deoxys.c: tweakable-cipher.macros:7:22: note: called from here
deoxys.c: 7 | #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: | ^~~~~~~~~~~~~~~~~~~~~
deoxys.c: deoxys.c:98:13: note: in expansion of macro 'permute'
deoxys.c: 98 | tmp = permute( tmp, H_PERMUTATION );
deoxys.c: | ^~~~~~~
deoxys.c: In file included from deoxys.c:29:
deoxys.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesni
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesni
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesni
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesni

Compiler output

Implementation: T:aesnis
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

deoxys.c: deoxys.c:81:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'ssse3'
deoxys.c: TWEAKEY_SCHEDULE2(subkeys, key, tmp, tmp1);
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:33:3: note: expanded from macro 'TWEAKEY_SCHEDULE2'
deoxys.c: ONE_KEY_ROUND( tmp1, tmp2 ); subkeys[ 1] = xor( tmp2, RCONST( 1) ); \
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:27:13: note: expanded from macro 'ONE_KEY_ROUND'
deoxys.c: new_key = permute( new_key, H_PERMUTATION);
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:7:22: note: expanded from macro 'permute'
deoxys.c: #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: ^
deoxys.c: deoxys.c:81:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'ssse3'
deoxys.c: ./tweakable-cipher.macros:34:3: note: expanded from macro 'TWEAKEY_SCHEDULE2'
deoxys.c: ONE_KEY_ROUND( tmp2, tmp1 ); subkeys[ 2] = xor( tmp1, RCONST( 2) ); \
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:27:13: note: expanded from macro 'ONE_KEY_ROUND'
deoxys.c: new_key = permute( new_key, H_PERMUTATION);
deoxys.c: ^
deoxys.c: ./tweakable-cipher.macros:7:22: note: expanded from macro 'permute'
deoxys.c: #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: ^
deoxys.c: deoxys.c:81:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt' that is compiled without support for 'ssse3'
deoxys.c: ./tweakable-cipher.macros:35:3: note: expanded from macro 'TWEAKEY_SCHEDULE2'
deoxys.c: ONE_KEY_ROUND( tmp1, tmp2 ); subkeys[ 3] = xor( tmp2, RCONST( 3) ); \
deoxys.c: ...

Number of similar (compiler,implementation) pairs: 5, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesnis
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesnis
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesnis
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesnis
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:aesnis

Compiler output

Implementation: T:aesnis
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

deoxys.c: In file included from deoxys.c:29:
deoxys.c: deoxys.c: In function 'deoxys_aead_encrypt':
deoxys.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
deoxys.c: | ^~~~~~~~~~~~~~~~
deoxys.c: In file included from deoxys.c:30:
deoxys.c: tweakable-cipher.macros:7:22: note: called from here
deoxys.c: 7 | #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: | ^~~~~~~~~~~~~~~~~~~~~
deoxys.c: tweakable-cipher.macros:27:13: note: in expansion of macro 'permute'
deoxys.c: 27 | new_key = permute( new_key, H_PERMUTATION);
deoxys.c: | ^~~~~~~
deoxys.c: tweakable-cipher.macros:46:3: note: in expansion of macro 'ONE_KEY_ROUND'
deoxys.c: 46 | ONE_KEY_ROUND( tmp2, tmp1 ); subkeys[14] = xor( tmp1, RCONST(14) );
deoxys.c: | ^~~~~~~~~~~~~
deoxys.c: deoxys.c:81:5: note: in expansion of macro 'TWEAKEY_SCHEDULE2'
deoxys.c: 81 | TWEAKEY_SCHEDULE2(subkeys, key, tmp, tmp1);
deoxys.c: | ^~~~~~~~~~~~~~~~~
deoxys.c: In file included from deoxys.c:29:
deoxys.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
deoxys.c: | ^~~~~~~~~~~~~~~~
deoxys.c: In file included from deoxys.c:30:
deoxys.c: tweakable-cipher.macros:7:22: note: called from here
deoxys.c: 7 | #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesnis
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesnis
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesnis
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:aesnis

Compiler output

Implementation: T:bitslice
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

deoxys_8.c: deoxys_8.c:178:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt_8' that is compiled without support for 'ssse3'
deoxys_8.c: KEY_SCHEDULE(key, subkey);
deoxys_8.c: ^
deoxys_8.c: ./deoxys.macros:76:5: note: expanded from macro 'KEY_SCHEDULE'
deoxys_8.c: packing(subkey[0]);\
deoxys_8.c: ^
deoxys_8.c: ./deoxys.macros:401:14: note: expanded from macro 'packing'
deoxys_8.c: (x)[0] = shuffle_pack((x)[0]);\
deoxys_8.c: ^
deoxys_8.c: ./deoxys.macros:32:25: note: expanded from macro 'shuffle_pack'
deoxys_8.c: #define shuffle_pack(a) permute(a, SET8(15,11,7,3,14,10,6,2,13,9,5,1,12,8,4,0) )
deoxys_8.c: ^
deoxys_8.c: ./deoxys.macros:31:25: note: expanded from macro 'permute'
deoxys_8.c: #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys_8.c: ^
deoxys_8.c: deoxys_8.c:178:5: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'deoxys_aead_encrypt_8' that is compiled without support for 'ssse3'
deoxys_8.c: ./deoxys.macros:76:5: note: expanded from macro 'KEY_SCHEDULE'
deoxys_8.c: packing(subkey[0]);\
deoxys_8.c: ^
deoxys_8.c: ./deoxys.macros:402:14: note: expanded from macro 'packing'
deoxys_8.c: (x)[1] = shuffle_pack((x)[1]);\
deoxys_8.c: ^
deoxys_8.c: ./deoxys.macros:32:25: note: expanded from macro 'shuffle_pack'
deoxys_8.c: #define shuffle_pack(a) permute(a, SET8(15,11,7,3,14,10,6,2,13,9,5,1,12,8,4,0) )
deoxys_8.c: ^
deoxys_8.c: ...

Number of similar (compiler,implementation) pairs: 5, namely:

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:bitslice
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:bitslice
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:bitslice
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:bitslice
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:bitslice

Compiler output

Implementation: T:bitslice
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

deoxys_8.c: In file included from deoxys_8.c:29:
deoxys_8.c: deoxys_8.c: In function 'deoxys_aead_encrypt_8':
deoxys_8.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys_8.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
deoxys_8.c: | ^~~~~~~~~~~~~~~~
deoxys_8.c: In file included from deoxys_8.c:30:
deoxys_8.c: deoxys.macros:31:25: note: called from here
deoxys_8.c: 31 | #define permute(a,b) _mm_shuffle_epi8(a,b)
deoxys_8.c: | ^~~~~~~~~~~~~~~~~~~~~
deoxys_8.c: deoxys.macros:32:25: note: in expansion of macro 'permute'
deoxys_8.c: 32 | #define shuffle_pack(a) permute(a, SET8(15,11,7,3,14,10,6,2,13,9,5,1,12,8,4,0) )
deoxys_8.c: | ^~~~~~~
deoxys_8.c: deoxys.macros:408:14: note: in expansion of macro 'shuffle_pack'
deoxys_8.c: 408 | (x)[7] = shuffle_pack((x)[7]);\
deoxys_8.c: | ^~~~~~~~~~~~
deoxys_8.c: deoxys.macros:76:5: note: in expansion of macro 'packing'
deoxys_8.c: 76 | packing(subkey[0]);\
deoxys_8.c: | ^~~~~~~
deoxys_8.c: deoxys_8.c:178:5: note: in expansion of macro 'KEY_SCHEDULE'
deoxys_8.c: 178 | KEY_SCHEDULE(key, subkey);
deoxys_8.c: | ^~~~~~~~~~~~
deoxys_8.c: In file included from deoxys_8.c:29:
deoxys_8.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/tmmintrin.h:136:1: error: inlining failed in call to 'always_inline' '_mm_shuffle_epi8': target specific option mismatch
deoxys_8.c: 136 | _mm_shuffle_epi8 (__m128i __X, __m128i __Y)
deoxys_8.c: | ^~~~~~~~~~~~~~~~
deoxys_8.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:bitslice
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:bitslice
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:bitslice
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:bitslice