Test results for amd64, h4atom, crypto_core/multsntrup653

[Page version: 20251115 13:22:58]

Measurements for amd64, h4atom, crypto_core Test results for amd64, h4atom, crypto_core Test results for crypto_core/multsntrup653
Computer: h4atom
Microarchitecture: amd64; Bonnell (106ca)
Architecture: amd64
CPU ID: GenuineIntel-000106ca-bfe9fbff
SUPERCOP version: 20250922
Operation: crypto_core
Primitive: multsntrup653
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
8850314206 0 018142 780 1000refgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
19205712942 0 020999 820 936refclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
20369332839 0 016055 820 936refclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
20450622839 0 017199 820 936refclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
20735542639 0 016614 780 1000compactgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
25740881020 0 015359 820 936compactclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
25763971020 0 014215 820 936compactclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
28436581372 0 019431 820 936compactclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
3915368458 0 012821 772 1000compactgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
41672301460 0 014174 780 1000refgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
4336610459 0 011641 812 936compactclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
4338135463 0 012247 820 936compactclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
4384572613 0 011793 812 936refclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
4769139486 0 013230 780 1000compactgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
4777098410 0 010993 756 968compactgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
4839298465 0 011065 756 968refgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
5031685671 0 012447 820 936refclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall2025082920250415
6327506583 0 012965 772 1000refgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall2025082920250415

Compiler output

mult768.c: mult768.c:267:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_avx_constbranchindex' that is compiled without support for 'avx'
mult768.c:   267 |   x = const_x16(0);
mult768.c:       |       ^
mult768.c: mult768.c:12:19: note: expanded from macro 'const_x16'
mult768.c:    12 | #define const_x16 _mm256_set1_epi16
mult768.c:       |                   ^
mult768.c: mult768.c:267:7: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult768.c: mult768.c:12:19: note: expanded from macro 'const_x16'
mult768.c:    12 | #define const_x16 _mm256_set1_epi16
mult768.c:       |                   ^
mult768.c: mult768.c:268:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_avx_constbranchindex' that is compiled without support for 'avx'
mult768.c:   268 |   for (i = p&~15;i < 768;i += 16) store_x16(&f[i],x);
mult768.c:       |                                   ^
mult768.c: mult768.c:11:24: note: expanded from macro 'store_x16'
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^
mult768.c: mult768.c:268:35: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult768.c: mult768.c:11:24: note: expanded from macro 'store_x16'
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^
mult768.c: mult768.c:269:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_avx_constbranchindex' that is compiled without support for 'avx'
mult768.c:   269 |   for (i = p&~15;i < 768;i += 16) store_x16(&g[i],x);
mult768.c:       |                                   ^
mult768.c: mult768.c:11:24: note: expanded from macro 'store_x16'
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c: ...
Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
avxclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avxclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avxclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avxclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avxclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))

Compiler output

mult768.c: mult768.c: In function 'squeeze_4621_x16':
mult768.c: mult768.c:21:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
mult768.c:    21 | {
mult768.c:       | ^
mult768.c: mult768.c: In function 'squeeze_7681_x16':
mult768.c: mult768.c:25:24: note: the ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
mult768.c:    25 | static inline int16x16 squeeze_7681_x16(int16x16 x)
mult768.c:       |                        ^~~~~~~~~~~~~~~~
mult768.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/14/include/immintrin.h:43,
mult768.c:                  from mult768.c:3:
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/14/include/avxintrin.h: In function 'good':
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/14/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
mult768.c:   933 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
mult768.c:       | ^~~~~~~~~~~~~~~~~~~
mult768.c: mult768.c:11:24: note: called from here
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mult768.c: mult768.c:77:5: note: in expansion of macro 'store_x16'
mult768.c:    77 |     store_x16(&fpad[2][j],(f0&mask2)|(f1&mask0));
mult768.c:       |     ^~~~~~~~~
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/14/include/avxintrin.h:1282:1: error: inlining failed in call to 'always_inline' '_mm256_set_epi16': target specific option mismatch
mult768.c:  1282 | _mm256_set_epi16 (short __q15, short __q14, short __q13, short __q12,
mult768.c:       | ^~~~~~~~~~~~~~~~
mult768.c: mult768.c:62:15: note: called from here
mult768.c:    62 | #define mask0 _mm256_set_epi16(-1,0,0,-1,0,0,-1,0,0,-1,0,0,-1,0,0,-1)
mult768.c: ...
Number of similar (implementation,compiler) pairs: 8, namely:
ImplementationCompiler
avxgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avxgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avxgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avxgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avx800gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avx800gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avx800gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
avx800gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)

Compiler output

mult768.c: mult768.c:267:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_avx800_constbranchindex' that is compiled without support for 'avx'
mult768.c:   267 |   x = const_x16(0);
mult768.c:       |       ^
mult768.c: mult768.c:12:19: note: expanded from macro 'const_x16'
mult768.c:    12 | #define const_x16 _mm256_set1_epi16
mult768.c:       |                   ^
mult768.c: mult768.c:267:7: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult768.c: mult768.c:12:19: note: expanded from macro 'const_x16'
mult768.c:    12 | #define const_x16 _mm256_set1_epi16
mult768.c:       |                   ^
mult768.c: mult768.c:268:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_avx800_constbranchindex' that is compiled without support for 'avx'
mult768.c:   268 |   for (i = p&~15;i < 768;i += 16) store_x16(&f[i],x);
mult768.c:       |                                   ^
mult768.c: mult768.c:11:24: note: expanded from macro 'store_x16'
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^
mult768.c: mult768.c:268:35: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult768.c: mult768.c:11:24: note: expanded from macro 'store_x16'
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^
mult768.c: mult768.c:269:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_avx800_constbranchindex' that is compiled without support for 'avx'
mult768.c:   269 |   for (i = p&~15;i < 768;i += 16) store_x16(&g[i],x);
mult768.c:       |                                   ^
mult768.c: mult768.c:11:24: note: expanded from macro 'store_x16'
mult768.c:    11 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c: ...
Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
avx800clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avx800clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avx800clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avx800clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
avx800clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))

Compiler output

mult768.c: mult768.c:268:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_round2_constbranchindex' that is compiled without support for 'avx'
mult768.c:   268 |   x = const_x16(0);
mult768.c:       |       ^
mult768.c: mult768.c:11:19: note: expanded from macro 'const_x16'
mult768.c:    11 | #define const_x16 _mm256_set1_epi16
mult768.c:       |                   ^
mult768.c: mult768.c:268:7: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult768.c: mult768.c:11:19: note: expanded from macro 'const_x16'
mult768.c:    11 | #define const_x16 _mm256_set1_epi16
mult768.c:       |                   ^
mult768.c: mult768.c:269:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_round2_constbranchindex' that is compiled without support for 'avx'
mult768.c:   269 |   for (i = p&~15;i < 768;i += 16) store_x16(&f[i],x);
mult768.c:       |                                   ^
mult768.c: mult768.c:10:24: note: expanded from macro 'store_x16'
mult768.c:    10 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^
mult768.c: mult768.c:269:35: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult768.c: mult768.c:10:24: note: expanded from macro 'store_x16'
mult768.c:    10 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^
mult768.c: mult768.c:270:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup653_round2_constbranchindex' that is compiled without support for 'avx'
mult768.c:   270 |   for (i = p&~15;i < 768;i += 16) store_x16(&g[i],x);
mult768.c:       |                                   ^
mult768.c: mult768.c:10:24: note: expanded from macro 'store_x16'
mult768.c:    10 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c: ...
Number of similar (implementation,compiler) pairs: 5, namely:
ImplementationCompiler
round2clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
round2clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
round2clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
round2clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
round2clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))

Compiler output

mult768.c: mult768.c: In function 'squeeze_4621_x16':
mult768.c: mult768.c:20:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
mult768.c:    20 | {
mult768.c:       | ^
mult768.c: mult768.c: In function 'squeeze_7681_x16':
mult768.c: mult768.c:24:24: note: the ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
mult768.c:    24 | static inline int16x16 squeeze_7681_x16(int16x16 x)
mult768.c:       |                        ^~~~~~~~~~~~~~~~
mult768.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/14/include/immintrin.h:43,
mult768.c:                  from mult768.c:3:
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/14/include/avxintrin.h: In function 'good':
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/14/include/avxintrin.h:933:1: error: inlining failed in call to 'always_inline' '_mm256_storeu_si256': target specific option mismatch
mult768.c:   933 | _mm256_storeu_si256 (__m256i_u *__P, __m256i __A)
mult768.c:       | ^~~~~~~~~~~~~~~~~~~
mult768.c: mult768.c:10:24: note: called from here
mult768.c:    10 | #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c:       |                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mult768.c: mult768.c:76:5: note: in expansion of macro 'store_x16'
mult768.c:    76 |     store_x16(&fpad[2][j],(f0&mask2)|(f1&mask0));
mult768.c:       |     ^~~~~~~~~
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/14/include/avxintrin.h:1282:1: error: inlining failed in call to 'always_inline' '_mm256_set_epi16': target specific option mismatch
mult768.c:  1282 | _mm256_set_epi16 (short __q15, short __q14, short __q13, short __q12,
mult768.c:       | ^~~~~~~~~~~~~~~~
mult768.c: mult768.c:61:15: note: called from here
mult768.c:    61 | #define mask0 _mm256_set_epi16(-1,0,0,-1,0,0,-1,0,0,-1,0,0,-1,0,0,-1)
mult768.c: ...
Number of similar (implementation,compiler) pairs: 4, namely:
ImplementationCompiler
round2gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
round2gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
round2gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
round2gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)

Passed TIMECOP

TIMECOP iterations: 1
Number of similar (implementation,compiler) pairs: 18, namely:
ImplementationCompiler
compactclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
compactclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
compactclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
compactclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
compactclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
compactgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
compactgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
compactgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
compactgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
refclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
refclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
refclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
refclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
refclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_19.1.7_(3+b1))
refgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
refgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
refgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)
refgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (14.2.0)