Implementation notes: amd64, wolfdale, crypto_stream/aes256ctr

Computer: wolfdale
Microarchitecture: amd64; Core 2 45nm (1067a)
Architecture: amd64
CPU ID: GenuineIntel-0001067a-bfebfbff
SUPERCOP version: 20240808
Operation: crypto_stream
Primitive: aes256ctr
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
2966419003 2800 01658414 145348 10424T:cryptoppg++_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
2967620936 2808 01660865 145332 10424T:cryptoppg++_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
297849424 1480 01358070 145340 10328T:cryptoppclang++_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
2982218675 2800 01657308 145348 10424T:cryptoppg++_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
2983012339 2064 01363461 145348 10328T:cryptoppclang++_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
299028296 3272 01650334 146412 10392T:cryptoppg++_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3000712251 2064 01362593 145348 10328T:cryptoppclang++_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3042917833 6592 01370706 148748 10328T:cryptoppclang++_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
31367437 0 013913 860 952T:opensslclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
31398428 0 016123 868 952T:opensslclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
31718654 0 015744 852 984T:opensslgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
31740358 0 012731 828 984T:opensslgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
31899437 0 014104 852 984T:opensslgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
32025428 0 016427 868 952T:opensslclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
32152423 0 013377 860 952T:opensslclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
32468472 0 014520 852 984T:opensslgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
32649428 0 015611 868 952T:opensslclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625

Compiler output


aesenc-int.c: aesenc-int.c:81:20: warning: unused function 'aesni_encrypt1' [-Wunused-function]
aesenc-int.c: static inline void aesni_encrypt1(unsigned char *out, unsigned char *n, __m128i rkeys[16]) {
aesenc-int.c:                    ^
aesenc-int.c: aesenc-int.c:97:20: warning: unused function 'incle' [-Wunused-function]
aesenc-int.c: static inline void incle(unsigned char n[16]) {
aesenc-int.c:                    ^
aesenc-int.c: aesenc-int.c:195:1: warning: unused function 'aesni_encrypt4' [-Wunused-function]
aesenc-int.c: FUNC(4, MAKE4)
aesenc-int.c: ^
aesenc-int.c: aesenc-int.c:172:22: note: expanded from macro 'FUNC'
aesenc-int.c:   static inline void aesni_encrypt##N(unsigned char *out, unsigned char *n, __m128i rkeys[16]) { \
aesenc-int.c:                      ^
aesenc-int.c: <scratch space>:224:1: note: expanded from here
aesenc-int.c: aesni_encrypt4
aesenc-int.c: ^
aesenc-int.c: aesenc-int.c:196:1: warning: unused function 'aesni_encrypt6' [-Wunused-function]
aesenc-int.c: FUNC(6, MAKE6)
aesenc-int.c: ^
aesenc-int.c: aesenc-int.c:172:22: note: expanded from macro 'FUNC'
aesenc-int.c:   static inline void aesni_encrypt##N(unsigned char *out, unsigned char *n, __m128i rkeys[16]) { \
aesenc-int.c:                      ^
aesenc-int.c: <scratch space>:273:1: note: expanded from here
aesenc-int.c: aesni_encrypt6
aesenc-int.c: ^
aesenc-int.c: aesenc-int.c:197:1: warning: unused function 'aesni_encrypt7' [-Wunused-function]
aesenc-int.c: ...

Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
dolbeau/aesenc-intclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/aesenc-intclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/aesenc-intclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/aesenc-intclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/aesenc-intclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output


aesenc-int.c: aesenc-int.c: In function 'aesni_encrypt1':
aesenc-int.c: aesenc-int.c:85: warning: ignoring '#pragma unroll ' [-Wunknown-pragmas]
aesenc-int.c:    85 | #pragma unroll(13)
aesenc-int.c:       |
aesenc-int.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:39,
aesenc-int.c:                  from aesenc-int.c:12:
aesenc-int.c: aesenc-int.c: In function 'aesni_key256_expand':
aesenc-int.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/wmmintrin.h:87:1: error: inlining failed in call to 'always_inline' '_mm_aeskeygenassist_si128': target specific option mismatch
aesenc-int.c:    87 | _mm_aeskeygenassist_si128 (__m128i __X, const int __C)
aesenc-int.c:       | ^~~~~~~~~~~~~~~~~~~~~~~~~
aesenc-int.c: aesenc-int.c:39:11: note: called from here
aesenc-int.c:    39 |   temp1 = _mm_aeskeygenassist_si128(temp2, IMM); \
aesenc-int.c:       |           ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
aesenc-int.c: aesenc-int.c:76:3: note: in expansion of macro 'BLOCK1'
aesenc-int.c:    76 |   BLOCK1(0x40);
aesenc-int.c:       |   ^~~~~~
aesenc-int.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:39,
aesenc-int.c:                  from aesenc-int.c:12:
aesenc-int.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/wmmintrin.h:87:1: error: inlining failed in call to 'always_inline' '_mm_aeskeygenassist_si128': target specific option mismatch
aesenc-int.c:    87 | _mm_aeskeygenassist_si128 (__m128i __X, const int __C)
aesenc-int.c:       | ^~~~~~~~~~~~~~~~~~~~~~~~~
aesenc-int.c: aesenc-int.c:49:11: note: called from here
aesenc-int.c:    49 |   temp1 = _mm_aeskeygenassist_si128(temp0, IMM); \
aesenc-int.c:       |           ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
aesenc-int.c: aesenc-int.c:74:3: note: in expansion of macro 'BLOCK2'
aesenc-int.c: ...

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
dolbeau/aesenc-intgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
dolbeau/aesenc-intgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
dolbeau/aesenc-intgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
dolbeau/aesenc-intgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)

Compiler output


vaesenc-int.c: vaesenc-int.c:15:2: error: "This code requires AVX enabled in the compiler."
vaesenc-int.c: #error "This code requires AVX enabled in the compiler."
vaesenc-int.c:  ^
vaesenc-int.c: vaesenc-int.c:19:2: error: "This code requires VAES enabled in the compiler."
vaesenc-int.c: #error "This code requires VAES enabled in the compiler."
vaesenc-int.c:  ^
vaesenc-int.c: 2 errors generated.

Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
dolbeau/vaesenc-intclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/vaesenc-intclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/vaesenc-intclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/vaesenc-intclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
dolbeau/vaesenc-intclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output


vaesenc-int.c: vaesenc-int.c:15:2: error: #error "This code requires AVX enabled in the compiler."
vaesenc-int.c:    15 | #error "This code requires AVX enabled in the compiler."
vaesenc-int.c:       |  ^~~~~
vaesenc-int.c: vaesenc-int.c:19:2: error: #error "This code requires VAES enabled in the compiler."
vaesenc-int.c:    19 | #error "This code requires VAES enabled in the compiler."
vaesenc-int.c:       |  ^~~~~
vaesenc-int.c: vaesenc-int.c: In function 'aesni_encrypt1':
vaesenc-int.c: vaesenc-int.c:123: warning: ignoring '#pragma unroll ' [-Wunknown-pragmas]
vaesenc-int.c:   123 | #pragma unroll(13)
vaesenc-int.c:       |
vaesenc-int.c: vaesenc-int.c: In function 'aesni_key256_expand':
vaesenc-int.c: vaesenc-int.c:71:16: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
vaesenc-int.c:    71 |   rkeys[idx++] = _mmAES_broadcast_i32x4(key0);
vaesenc-int.c:       |                ^

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
dolbeau/vaesenc-intgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
dolbeau/vaesenc-intgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
dolbeau/vaesenc-intgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
dolbeau/vaesenc-intgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)