Implementation notes: amd64, jasper, crypto_kem/kyber90s1024

Computer: jasper
Microarchitecture: amd64; Tremont (906c0)
Architecture: amd64
CPU ID: GenuineIntel-000906c0-20-bfebfbff
SUPERCOP version: 20240625
Operation: crypto_kem
Primitive: kyber90s1024
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
153029156180 0 074304 820 1720refclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
159098163744 0 082952 820 1720refclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
160189672636 0 093048 820 1720refclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
166976041789 0 058286 812 1720refclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
173143359461 0 079088 788 1752refgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
191353642320 0 060400 788 1752refgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
191987343130 0 061776 788 1752refgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
219138842842 0 060086 812 1720refclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625
265844640535 0 056664 780 1720refgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024070520240625

Compiler output


cbd.c: cbd.c:20:26: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'cbd2' that is compiled without support for 'avx'
cbd.c:   const __m256i mask55 = _mm256_set1_epi32(0x55555555);
cbd.c:                          ^
cbd.c: cbd.c:20:26: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
cbd.c: cbd.c:21:26: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'cbd2' that is compiled without support for 'avx'
cbd.c:   const __m256i mask33 = _mm256_set1_epi32(0x33333333);
cbd.c:                          ^
cbd.c: cbd.c:21:26: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
cbd.c: cbd.c:22:26: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'cbd2' that is compiled without support for 'avx'
cbd.c:   const __m256i mask03 = _mm256_set1_epi32(0x03030303);
cbd.c:                          ^
cbd.c: cbd.c:22:26: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
cbd.c: cbd.c:23:26: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'cbd2' that is compiled without support for 'avx'
cbd.c:   const __m256i mask0F = _mm256_set1_epi32(0x0F0F0F0F);
cbd.c:                          ^
cbd.c: cbd.c:23:26: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
cbd.c: cbd.c:26:10: error: always_inline function '_mm256_load_si256' requires target feature 'avx', but would be inlined into function 'cbd2' that is compiled without support for 'avx'
cbd.c:     f0 = _mm256_load_si256(&buf[i]);
cbd.c:          ^
cbd.c: cbd.c:26:10: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
cbd.c: cbd.c:28:10: error: always_inline function '_mm256_srli_epi16' requires target feature 'avx2', but would be inlined into function 'cbd2' that is compiled without support for 'avx2'
cbd.c:     f1 = _mm256_srli_epi16(f0, 1);
cbd.c:          ^
cbd.c: cbd.c:28:10: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
cbd.c: cbd.c:29:10: error: always_inline function '_mm256_and_si256' requires target feature 'avx2', but would be inlined into function 'cbd2' that is compiled without support for 'avx2'
cbd.c: ...

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
avx2clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
avx2clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
avx2clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
avx2clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output


aes256ctr.c: aes256ctr.c:90:3: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
aes256ctr.c:   BLOCK1(0x01);
aes256ctr.c:   ^
aes256ctr.c: aes256ctr.c:71:11: note: expanded from macro 'BLOCK1'
aes256ctr.c:   temp1 = _mm_aeskeygenassist_si128(temp2, IMM);                        \
aes256ctr.c:           ^
aes256ctr.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/__wmmintrin_aes.h:136:12: note: expanded from macro '_mm_aeskeygenassist_si128'
aes256ctr.c:   (__m128i)__builtin_ia32_aeskeygenassist128((__v2di)(__m128i)(C), (int)(R))
aes256ctr.c:            ^
aes256ctr.c: aes256ctr.c:91:3: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
aes256ctr.c:   BLOCK2(0x01);
aes256ctr.c:   ^
aes256ctr.c: aes256ctr.c:81:11: note: expanded from macro 'BLOCK2'
aes256ctr.c:   temp1 = _mm_aeskeygenassist_si128(temp0, IMM);                        \
aes256ctr.c:           ^
aes256ctr.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/__wmmintrin_aes.h:136:12: note: expanded from macro '_mm_aeskeygenassist_si128'
aes256ctr.c:   (__m128i)__builtin_ia32_aeskeygenassist128((__v2di)(__m128i)(C), (int)(R))
aes256ctr.c:            ^
aes256ctr.c: aes256ctr.c:93:3: error: '__builtin_ia32_aeskeygenassist128' needs target feature aes
aes256ctr.c:   BLOCK1(0x02);
aes256ctr.c:   ^
aes256ctr.c: aes256ctr.c:71:11: note: expanded from macro 'BLOCK1'
aes256ctr.c:   temp1 = _mm_aeskeygenassist_si128(temp2, IMM);                        \
aes256ctr.c:           ^
aes256ctr.c: /usr/lib/llvm-11/lib/clang/11.0.1/include/__wmmintrin_aes.h:136:12: note: expanded from macro '_mm_aeskeygenassist_si128'
aes256ctr.c: ...

Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
avx2clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)

Compiler output


cbd.c: cbd.c: In function 'cbd2':
cbd.c: cbd.c:20:17: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
cbd.c:    20 |   const __m256i mask55 = _mm256_set1_epi32(0x55555555);
cbd.c:       |                 ^~~~~~
cbd.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
cbd.c:                  from cbd.c:2:
cbd.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:1319:1: error: inlining failed in call to 'always_inline' '_mm256_set1_epi32': target specific option mismatch
cbd.c:  1319 | _mm256_set1_epi32 (int __A)
cbd.c:       | ^~~~~~~~~~~~~~~~~
cbd.c: cbd.c:23:26: note: called from here
cbd.c:    23 |   const __m256i mask0F = _mm256_set1_epi32(0x0F0F0F0F);
cbd.c:       |                          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cbd.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
cbd.c:                  from cbd.c:2:
cbd.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:1319:1: error: inlining failed in call to 'always_inline' '_mm256_set1_epi32': target specific option mismatch
cbd.c:  1319 | _mm256_set1_epi32 (int __A)
cbd.c:       | ^~~~~~~~~~~~~~~~~
cbd.c: cbd.c:22:26: note: called from here
cbd.c:    22 |   const __m256i mask03 = _mm256_set1_epi32(0x03030303);
cbd.c:       |                          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cbd.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
cbd.c:                  from cbd.c:2:
cbd.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:1319:1: error: inlining failed in call to 'always_inline' '_mm256_set1_epi32': target specific option mismatch
cbd.c:  1319 | _mm256_set1_epi32 (int __A)
cbd.c:       | ^~~~~~~~~~~~~~~~~
cbd.c: ...

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
avx2gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
avx2gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
avx2gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
avx2gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)

Passed TIMECOP


TIMECOP iterations: 1

Number of similar (implementation,compiler) pairs: 9, namely:
ImplementationCompiler
refclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
refclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
refclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
refclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
refclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)
refgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
refgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
refgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)
refgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)