Implementation notes: amd64, hunsnivy, crypto_aead/norx6461v1

Computer: hunsnivy
Microarchitecture: amd64; Ivy Bridge+AES (306a9)
Architecture: amd64
CPU ID: GenuineIntel-000306a9-bfebfbff
SUPERCOP version: 20240625
Operation: crypto_aead
Primitive: norx6461v1
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
3503310023 0 030784 780 1080T:xmmgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3507810023 0 031744 780 1080T:xmmgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3560410949 0 034648 812 1016T:xmmclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3562210923 0 032960 812 1016T:xmmclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
356449825 0 028595 756 1048T:xmmgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3576010453 0 030070 804 1016T:xmmclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3588210099 0 029302 804 1016T:xmmclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
3827412356 0 032711 772 1080T:xmmgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
4281215015 0 037128 812 1016T:xmmclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
428383377 8 022838 812 1016T:refclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
428836096 8 029944 820 1016T:refclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
429496144 8 028320 820 1016T:refclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
506423369 8 022268 764 1048T:refgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
510285688 8 027912 820 1016T:refclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
549416176 8 028028 788 1080T:refgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
554454880 8 025756 788 1080T:refgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
573793843 8 024288 780 1080T:refgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625
643604587 8 024558 812 1016T:refclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062820240625

Compiler output


norx.c: norx.c:355:5: error: '__builtin_ia32_insert128i256' needs target feature avx2
norx.c:     INITIALIZE(A, B, C, D, N, K);
norx.c:     ^
norx.c: norx.c:271:9: note: expanded from macro 'INITIALIZE'
norx.c:     A = _mm256_inserti128_si256(A, _mm_set_epi64x(U1, U0), 1);              \
norx.c:         ^
norx.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/avx2intrin.h:827:12: note: expanded from macro '_mm256_inserti128_si256'
norx.c:   (__m256i)__builtin_ia32_insert128i256((__v4di)(__m256i)(V1), \
norx.c:            ^
norx.c: norx.c:355:5: error: '__builtin_ia32_permdi256' needs target feature avx2
norx.c: norx.c:272:9: note: expanded from macro 'INITIALIZE'
norx.c:     A = _mm256_permute4x64_epi64(A, _MM_SHUFFLE(3, 1, 0, 2));               \
norx.c:         ^
norx.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/avx2intrin.h:818:12: note: expanded from macro '_mm256_permute4x64_epi64'
norx.c:   (__m256i)__builtin_ia32_permdi256((__v4di)(__m256i)(V), (int)(M))
norx.c:            ^
norx.c: norx.c:355:5: error: always_inline function '_mm256_xor_si256' requires target feature 'avx2', but would be inlined into function 'crypto_aead_norx6461v1_ymm_timingleaks_encrypt' that is compiled without support for 'avx2'
norx.c: norx.c:276:9: note: expanded from macro 'INITIALIZE'
norx.c:     D = XOR(D, _mm256_set_epi64x(0, NORX_PARAMETER, 0, 0));                 \
norx.c:         ^
norx.c: norx.c:66:19: note: expanded from macro 'XOR'
norx.c: #define XOR(A, B) _mm256_xor_si256((A), (B))
norx.c:                   ^
norx.c: norx.c:355:5: error: always_inline function '_mm256_xor_si256' requires target feature 'avx2', but would be inlined into function 'crypto_aead_norx6461v1_ymm_timingleaks_encrypt' that is compiled without support for 'avx2'
norx.c: norx.c:277:5: note: expanded from macro 'INITIALIZE'
norx.c: ...

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
T:ymmclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:ymmclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:ymmclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
T:ymmclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output


norx.c: norx.c:350:24: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_aead_norx6461v1_ymm_timingleaks_encrypt' that is compiled without support for 'avx'
norx.c:     const __m256i K  = LOADU(k + 0);
norx.c:                        ^
norx.c: norx.c:47:19: note: expanded from macro 'LOADU'
norx.c: #define LOADU(in) _mm256_loadu_si256((__m256i*)(in))
norx.c:                   ^
norx.c: norx.c:350:24: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
norx.c: norx.c:47:19: note: expanded from macro 'LOADU'
norx.c: #define LOADU(in) _mm256_loadu_si256((__m256i*)(in))
norx.c:                   ^
norx.c: norx.c:355:5: error: always_inline function '_mm256_castsi128_si256' requires target feature 'avx', but would be inlined into function 'crypto_aead_norx6461v1_ymm_timingleaks_encrypt' that is compiled without support for 'avx'
norx.c:     INITIALIZE(A, B, C, D, N, K);
norx.c:     ^
norx.c: norx.c:270:9: note: expanded from macro 'INITIALIZE'
norx.c:     A = _mm256_castsi128_si256(N);                                          \
norx.c:         ^
norx.c: norx.c:355:5: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
norx.c: norx.c:270:9: note: expanded from macro 'INITIALIZE'
norx.c:     A = _mm256_castsi128_si256(N);                                          \
norx.c:         ^
norx.c: norx.c:355:5: error: '__builtin_ia32_insert128i256' needs target feature avx2
norx.c: norx.c:271:9: note: expanded from macro 'INITIALIZE'
norx.c:     A = _mm256_inserti128_si256(A, _mm_set_epi64x(U1, U0), 1);              \
norx.c:         ^
norx.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/avx2intrin.h:827:12: note: expanded from macro '_mm256_inserti128_si256'
norx.c: ...

Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
T:ymmclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output


norx.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
norx.c:                  from /usr/lib/gcc/x86_64-linux-gnu/10/include/x86intrin.h:32,
norx.c:                  from norx.c:26:
norx.c: norx.c: In function 'crypto_aead_norx6461v1_ymm_timingleaks_encrypt':
norx.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avx2intrin.h:913:1: error: inlining failed in call to 'always_inline' '_mm256_xor_si256': target specific option mismatch
norx.c:   913 | _mm256_xor_si256 (__m256i __A, __m256i __B)
norx.c:       | ^~~~~~~~~~~~~~~~
norx.c: norx.c:66:19: note: called from here
norx.c:    66 | #define XOR(A, B) _mm256_xor_si256((A), (B))
norx.c:       |                   ^~~~~~~~~~~~~~~~~~~~~~~~~~
norx.c: norx.c:276:9: note: in expansion of macro 'XOR'
norx.c:   276 |     D = XOR(D, _mm256_set_epi64x(0, NORX_PARAMETER, 0, 0));                 \
norx.c:       |         ^~~
norx.c: norx.c:355:5: note: in expansion of macro 'INITIALIZE'
norx.c:   355 |     INITIALIZE(A, B, C, D, N, K);
norx.c:       |     ^~~~~~~~~~
norx.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
norx.c:                  from /usr/lib/gcc/x86_64-linux-gnu/10/include/x86intrin.h:32,
norx.c:                  from norx.c:26:
norx.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avx2intrin.h:1068:1: error: inlining failed in call to 'always_inline' '_mm256_permute4x64_epi64': target specific option mismatch
norx.c:  1068 | _mm256_permute4x64_epi64 (__m256i __X, const int __M)
norx.c:       | ^~~~~~~~~~~~~~~~~~~~~~~~
norx.c: norx.c:272:9: note: called from here
norx.c:   272 |     A = _mm256_permute4x64_epi64(A, _MM_SHUFFLE(3, 1, 0, 2));               \
norx.c:       |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
norx.c: ...

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