Implementation notes: amd64, hydra8, crypto_core/weightsntrup1013

Computer: hydra8
Microarchitecture: amd64; Ivy Bridge+AES (306a9)
Architecture: amd64
CPU ID: GenuineIntel-000306a9-bfebfbff
SUPERCOP version: 20240716
Operation: crypto_core
Primitive: weightsntrup1013
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
1711537 0 016768 860 928refclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
268307 0 014200 860 928refclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
277631 0 013912 860 928refclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
278323 0 014325 804 992refgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
2091188 0 012933 804 992refgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
2095103 0 011672 860 928refclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
214390 0 011298 852 928refclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
2145190 0 012684 796 992refgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716
2446168 0 011472 780 960refgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071720240716

Compiler output


weight.c: weight.c:26:11: error: always_inline function '_mm256_add_epi8' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:     sum = _mm256_add_epi8(sum,bits);
weight.c:           ^
weight.c: weight.c:31:11: error: always_inline function '_mm256_srli_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:   sumhi = _mm256_srli_epi16(sum,8);
weight.c:           ^
weight.c: weight.c:33:9: error: always_inline function '_mm256_add_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:   sum = _mm256_add_epi16(sum,sumhi);
weight.c:         ^
weight.c: weight.c:36:9: error: always_inline function '_mm256_hadd_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:   sum = _mm256_hadd_epi16(sum,sum);
weight.c:         ^
weight.c: weight.c:38:9: error: always_inline function '_mm256_hadd_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:   sum = _mm256_hadd_epi16(sum,sum);
weight.c:         ^
weight.c: weight.c:40:9: error: always_inline function '_mm256_hadd_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:   sum = _mm256_hadd_epi16(sum,sum);
weight.c:         ^
weight.c: 6 errors generated.

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
avxclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)

Compiler output


weight.c: weight.c:20:9: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx'
weight.c:   sum = _mm256_loadu_si256((__m256i *) (in+p-32));
weight.c:         ^
weight.c: weight.c:20:9: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
weight.c: weight.c:21:10: error: always_inline function '_mm256_set_epi8' requires target feature 'avx', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx'
weight.c:   sum &= endingmask;
weight.c:          ^
weight.c: ./params.h:2:20: note: expanded from macro 'endingmask'
weight.c: #define endingmask _mm256_set_epi8(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0)
weight.c:                    ^
weight.c: weight.c:21:10: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
weight.c: ./params.h:2:20: note: expanded from macro 'endingmask'
weight.c: #define endingmask _mm256_set_epi8(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0)
weight.c:                    ^
weight.c: weight.c:24:20: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx'
weight.c:     __m256i bits = _mm256_loadu_si256((__m256i *) in);
weight.c:                    ^
weight.c: weight.c:24:20: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
weight.c: weight.c:25:13: error: always_inline function '_mm256_set1_epi8' requires target feature 'avx', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx'
weight.c:     bits &= _mm256_set1_epi8(1);
weight.c:             ^
weight.c: weight.c:25:13: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
weight.c: weight.c:26:11: error: always_inline function '_mm256_add_epi8' requires target feature 'avx2', but would be inlined into function 'crypto_core_weightsntrup1013_avx_constbranchindex' that is compiled without support for 'avx2'
weight.c:     sum = _mm256_add_epi8(sum,bits);
weight.c:           ^
weight.c: ...

Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
avxclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)

Compiler output


weight.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
weight.c:                  from weight.c:1:
weight.c: weight.c: In function 'crypto_core_weightsntrup1013_avx_constbranchindex':
weight.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:105:1: error: inlining failed in call to 'always_inline' '_mm256_add_epi8': target specific option mismatch
weight.c:   105 | _mm256_add_epi8 (__m256i __A, __m256i __B)
weight.c:       | ^~~~~~~~~~~~~~~
weight.c: weight.c:26:11: note: called from here
weight.c:    26 |     sum = _mm256_add_epi8(sum,bits);
weight.c:       |           ^~~~~~~~~~~~~~~~~~~~~~~~~
weight.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
weight.c:                  from weight.c:1:
weight.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:105:1: error: inlining failed in call to 'always_inline' '_mm256_add_epi8': target specific option mismatch
weight.c:   105 | _mm256_add_epi8 (__m256i __A, __m256i __B)
weight.c:       | ^~~~~~~~~~~~~~~
weight.c: weight.c:26:11: note: called from here
weight.c:    26 |     sum = _mm256_add_epi8(sum,bits);
weight.c:       |           ^~~~~~~~~~~~~~~~~~~~~~~~~
weight.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
weight.c:                  from weight.c:1:
weight.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:287:1: error: inlining failed in call to 'always_inline' '_mm256_hadd_epi16': target specific option mismatch
weight.c:   287 | _mm256_hadd_epi16 (__m256i __X, __m256i __Y)
weight.c:       | ^~~~~~~~~~~~~~~~~
weight.c: weight.c:40:9: note: called from here
weight.c:    40 |   sum = _mm256_hadd_epi16(sum,sum);
weight.c:       |         ^~~~~~~~~~~~~~~~~~~~~~~~~~
weight.c: ...

Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
avxgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
avxgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
avxgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
avxgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)

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 (Ubuntu_Clang_14.0.0)
refclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
refgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
refgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
refgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)