Implementation notes: amd64, hunsnivy, crypto_core/inv3sntrup857

Computer: hunsnivy
Microarchitecture: amd64; Ivy Bridge+AES (306a9)
Architecture: amd64
CPU ID: GenuineIntel-000306a9-bfebfbff
SUPERCOP version: 20240625
Operation: crypto_core
Primitive: inv3sntrup857

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
4741372	4198 0 0	17170 788 984	`ref`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
17100357	2535 0 0	16208 812 920	`ref`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
17150879	5797 0 0	21144 812 920	`ref`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
20848508	4958 0 0	18728 812 920	`ref`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
36295189	672 0 0	11686 804 920	`ref`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
39663506	855 0 0	12358 804 920	`ref`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
52182866	868 0 0	12898 788 984	`ref`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
57614329	817 0 0	12463 772 984	`ref`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625
59567029	675 0 0	11261 764 952	`ref`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240626	20240625

Compiler output

r3_recip.c: r3_recip.c:39:17: error: always_inline function '_mm256_unpacklo_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c0 = _mm256_unpacklo_epi32(b0,b1); /* 0 1 2 3 32 33 34 35 4 5 6 7 36 37 38 39 ... 55 */
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:40:17: error: always_inline function '_mm256_unpackhi_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c1 = _mm256_unpackhi_epi32(b0,b1); /* 8 9 10 11 40 41 42 43 ... 63 */
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:41:17: error: always_inline function '_mm256_unpacklo_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c2 = _mm256_unpacklo_epi32(b2,b3);
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:42:17: error: always_inline function '_mm256_unpackhi_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c3 = _mm256_unpackhi_epi32(b2,b3);
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:43:17: error: always_inline function '_mm256_unpacklo_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c4 = _mm256_unpacklo_epi32(b4,b5);
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:44:17: error: always_inline function '_mm256_unpackhi_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c5 = _mm256_unpackhi_epi32(b4,b5);
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:45:17: error: always_inline function '_mm256_unpacklo_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c6 = _mm256_unpacklo_epi32(b6,b7);
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:46:17: error: always_inline function '_mm256_unpackhi_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c:     vec256 c7 = _mm256_unpackhi_epi32(b6,b7);
r3_recip.c:                 ^
r3_recip.c: r3_recip.c:48:22: error: always_inline function '_mm256_slli_epi32' requires target feature 'avx2', but would be inlined into function 'vec256_frombits' that is compiled without support for 'avx2'
r3_recip.c: ...

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

Implementation	Compiler
`avx`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`avx`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`avx`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`avx`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`

Compiler output

r3_recip.c: r3_recip.c:418:11: error: always_inline function '_mm256_set_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c:   F0[0] = _mm256_set_epi32(0,0,0,0,0,0,0,1);
r3_recip.c:           ^
r3_recip.c: r3_recip.c:418:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
r3_recip.c: r3_recip.c:419:11: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c:   F0[1] = _mm256_set1_epi32(0);
r3_recip.c:           ^
r3_recip.c: r3_recip.c:419:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
r3_recip.c: r3_recip.c:420:11: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c:   F0[2] = _mm256_set1_epi32(0);
r3_recip.c:           ^
r3_recip.c: r3_recip.c:420:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
r3_recip.c: r3_recip.c:421:11: error: always_inline function '_mm256_set_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c:   F0[3] = _mm256_set_epi32(0,0,0,0,0,4194304,0,4194304);
r3_recip.c:           ^
r3_recip.c: r3_recip.c:421:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
r3_recip.c: r3_recip.c:422:11: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c:   F1[0] = _mm256_set1_epi32(0);
r3_recip.c:           ^
r3_recip.c: r3_recip.c:422:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
r3_recip.c: r3_recip.c:423:11: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c:   F1[1] = _mm256_set1_epi32(0);
r3_recip.c:           ^
r3_recip.c: r3_recip.c:423:11: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
r3_recip.c: r3_recip.c:424:11: error: always_inline function '_mm256_set1_epi32' requires target feature 'avx', but would be inlined into function 'crypto_core_inv3sntrup857_avx_constbranchindex' that is compiled without support for 'avx'
r3_recip.c: ...

Number of similar (implementation,compiler) pairs: 1, namely:

Implementation	Compiler
`avx`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`

Compiler output

r3_recip.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
r3_recip.c:                  from r3_recip.c:3:
r3_recip.c: r3_recip.c: In function 'vec256_frombits':
r3_recip.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avx2intrin.h:597:1: error: inlining failed in call to 'always_inline' '_mm256_shuffle_epi32': target specific option mismatch
r3_recip.c:   597 | _mm256_shuffle_epi32 (__m256i __A, const int __mask)
r3_recip.c:       | ^~~~~~~~~~~~~~~~~~~~
r3_recip.c: r3_recip.c:69:9: note: called from here
r3_recip.c:    69 |     h = _mm256_shuffle_epi32(h,0xd8);
r3_recip.c:       |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
r3_recip.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
r3_recip.c:                  from r3_recip.c:3:
r3_recip.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
r3_recip.c:  1068 | _mm256_permute4x64_epi64 (__m256i __X, const int __M)
r3_recip.c:       | ^~~~~~~~~~~~~~~~~~~~~~~~
r3_recip.c: r3_recip.c:68:9: note: called from here
r3_recip.c:    68 |     h = _mm256_permute4x64_epi64(h,0xd8);
r3_recip.c:       |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
r3_recip.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
r3_recip.c:                  from r3_recip.c:3:
r3_recip.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avx2intrin.h:588:1: error: inlining failed in call to 'always_inline' '_mm256_shuffle_epi8': target specific option mismatch
r3_recip.c:   588 | _mm256_shuffle_epi8 (__m256i __X, __m256i __Y)
r3_recip.c:       | ^~~~~~~~~~~~~~~~~~~
r3_recip.c: r3_recip.c:67:9: note: called from here
r3_recip.c:    67 |     h = _mm256_shuffle_epi8(h,TRANSPOSE);
r3_recip.c:       |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
r3_recip.c: ...

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

Implementation	Compiler
`avx`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`avx`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`avx`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`avx`	`gcc -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:

Implementation	Compiler
`ref`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`ref`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`ref`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`