Implementation notes: amd64, saber214, crypto_core/weightsntrup1277

Computer: saber214
Microarchitecture: amd64; Bulldozer (600f20)
Architecture: amd64
CPU ID: AuthenticAMD-00600f20-1789c3f5
SUPERCOP version: 20240625
Operation: crypto_core
Primitive: weightsntrup1277
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
338656 0 015348 816 776refclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
340656 0 012956 816 760refclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
346411 0 013844 816 760refclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
358419 0 013622 776 832refgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
260490 0 010742 808 760refclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
2609189 0 011773 768 832refgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
2612187 0 011934 776 832refgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
3850103 0 011132 816 760refclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625
3852168 0 010601 752 800refgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024062720240625

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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0)
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(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0)
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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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_weightsntrup1277_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)

TIMECOP error (can be valgrind bug)

Process terminating with default action of signal 4 (SIGILL)
 Illegal opcode at address 0x109FBD
   at 0x...: salsa20.part.0 (try-anything.c:102)
   by 0x...: salsa20 (try-anything.c:85)
   by 0x...: canary (try-anything.c:148)
   by 0x...: output_prepare (try-anything.c:178)
   by 0x...: test (try.c:99)
   by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
refgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)

TIMECOP error (can be valgrind bug)

Process terminating with default action of signal 4 (SIGILL)
 Illegal opcode at address 0x109C0A
   at 0x...: core (try-anything.c:53)
   by 0x...: salsa20.part.0 (try-anything.c:101)
   by 0x...: salsa20 (try-anything.c:85)
   by 0x...: canary (try-anything.c:148)
   by 0x...: output_prepare (try-anything.c:178)
   by 0x...: test (try.c:99)
   by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
refgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)

TIMECOP error (can be valgrind bug)

Process terminating with default action of signal 4 (SIGILL)
 Illegal opcode at address 0x109804
   at 0x...: memcpy (string_fortified.h:29)
   by 0x...: test (try.c:149)
   by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
refgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)

Passed TIMECOP

TIMECOP iterations: 1
Number of similar (implementation,compiler) pairs: 6, 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 -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)