Implementation notes: amd64, scw1b63b1, crypto_core/multsntrup761

Computer: scw1b63b1
Architecture: amd64
CPU ID: GenuineIntel-000506f1-0f8bfbff
SUPERCOP version: 20191017
Operation: crypto_core
Primitive: multsntrup761
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
5499561587 0 013313 784 760refclang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2019121420191017
5510221587 0 012489 784 760refclang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2019121420191017
5554741587 0 012489 784 760refclang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2019121420191017
1629528590 0 010579 776 760refclang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2019121420191017
16459862315 0 014113 784 760refclang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2019121420191017
1794492597 0 011451 800 832refgcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2019121420191017
18638022563 0 015828 808 832refgcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2019121420191017
1878328624 0 011723 800 832refgcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2019121420191017
2333074485 0 010379 792 800refgcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2019121420191017

Compiler output

Implementation: crypto_core/multsntrup761/round1
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
mult.c: mult.c:146:22: error: invalid output size for constraint '=&x'
mult.c: MULSTEP_fromzero(0,h0,h1,h2,h3,h4)
mult.c: ^
mult.c: mult.c:148:26: error: invalid output size for constraint '+x'
mult.c: MULSTEP_noload(j + 1,h1,h2,h3,h4,h0)
mult.c: ^
mult.c: mult.c:149:26: error: invalid output size for constraint '+x'
mult.c: MULSTEP_noload(j + 2,h2,h3,h4,h0,h1)
mult.c: ^
mult.c: mult.c:150:26: error: invalid output size for constraint '+x'
mult.c: ...
mult.c: #define MULSTEP MULSTEP_asm
mult.c: ^
mult.c: mult.c:185:26: error: invalid output size for constraint '+x'
mult.c: MULSTEP_noload(j + 3,h3,h4,h0,h1,h2)
mult.c: ^
mult.c: mult.c:186:26: error: invalid output size for constraint '+x'
mult.c: MULSTEP_noload(j + 4,h4,h0,h1,h2,h3)
mult.c: ^
mult.c: fatal error: too many errors emitted, stopping now [-ferror-limit=]
mult.c: 20 errors generated.

Number of similar (compiler,implementation) pairs: 5, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE round1
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE round1
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE round1
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE round1
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE round1

Compiler output

Implementation: crypto_core/multsntrup761/avx
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
mult768.c: mult768.c:266:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'xsave', but would be inlined into function 'crypto_core_multsntrup761_avx' that is compiled without support for 'xsave'
mult768.c: x = const_x16(0);
mult768.c: ^
mult768.c: mult768.c:10:19: note: expanded from macro 'const_x16'
mult768.c: #define const_x16 _mm256_set1_epi16
mult768.c: ^
mult768.c: mult768.c:267:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'xsave', but would be inlined into function 'crypto_core_multsntrup761_avx' that is compiled without support for 'xsave'
mult768.c: for (i = p&~15;i < 768;i += 16) store_x16(&f[i],x);
mult768.c: ^
mult768.c: mult768.c:9:24: note: expanded from macro 'store_x16'
mult768.c: ...
mult768.c: mult768.c:11:17: note: expanded from macro 'add_x16'
mult768.c: #define add_x16 _mm256_add_epi16
mult768.c: ^
mult768.c: mult768.c:292:5: error: always_inline function '_mm256_storeu_si256' requires target feature 'xsave', but would be inlined into function 'crypto_core_multsntrup761_avx' that is compiled without support for 'xsave'
mult768.c: store_x16(&h[i],x);
mult768.c: ^
mult768.c: mult768.c:9:24: note: expanded from macro 'store_x16'
mult768.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c: ^
mult768.c: 11 errors generated.

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE avx
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE avx
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE avx
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE avx

Compiler output

Implementation: crypto_core/multsntrup761/avx
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
mult768.c: mult768.c:266:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'crypto_core_multsntrup761_avx' that is compiled without support for 'sse4.2'
mult768.c: x = const_x16(0);
mult768.c: ^
mult768.c: mult768.c:10:19: note: expanded from macro 'const_x16'
mult768.c: #define const_x16 _mm256_set1_epi16
mult768.c: ^
mult768.c: mult768.c:267:35: error: always_inline function '_mm256_storeu_si256' requires target feature 'sse4.2', but would be inlined into function 'crypto_core_multsntrup761_avx' that is compiled without support for 'sse4.2'
mult768.c: for (i = p&~15;i < 768;i += 16) store_x16(&f[i],x);
mult768.c: ^
mult768.c: mult768.c:9:24: note: expanded from macro 'store_x16'
mult768.c: ...
mult768.c: mult768.c:11:17: note: expanded from macro 'add_x16'
mult768.c: #define add_x16 _mm256_add_epi16
mult768.c: ^
mult768.c: mult768.c:292:5: error: always_inline function '_mm256_storeu_si256' requires target feature 'sse4.2', but would be inlined into function 'crypto_core_multsntrup761_avx' that is compiled without support for 'sse4.2'
mult768.c: store_x16(&h[i],x);
mult768.c: ^
mult768.c: mult768.c:9:24: note: expanded from macro 'store_x16'
mult768.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult768.c: ^
mult768.c: 11 errors generated.

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE avx

Compiler output

Implementation: crypto_core/multsntrup761/round1
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE
mult.c: mult.c: In function ‘add’:
mult.c: mult.c:81:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
mult.c: {
mult.c: ^
mult.c: mult.c: In function ‘squeezeadd16’:
mult.c: mult.c:109:23: note: The ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
mult.c: static inline __m256i squeezeadd16(__m256i x,__m256i y)
mult.c: ^~~~~~~~~~~~
mult.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:79:0,
mult.c: from mult.c:2:
mult.c: ...
mult.c: return _mm256_fmadd_ps(y,v1_float,x);
mult.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mult.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:41:0,
mult.c: from mult.c:2:
mult.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avxintrin.h:1285:1: error: inlining failed in call to always_inline ‘_mm256_set1_ps’: target specific option mismatch
mult.c: _mm256_set1_ps (float __A)
mult.c: ^~~~~~~~~~~~~~
mult.c: mult.c:87:10: note: called from here
mult.c: return _mm256_fmadd_ps(y,v1_float,x);
mult.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE round1
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE round1
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE round1
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE round1

Compiler output

Implementation: crypto_core/multsntrup761/avx
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE
mult768.c: mult768.c: In function ‘squeeze_4591_x16’:
mult768.c: mult768.c:19:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
mult768.c: {
mult768.c: ^
mult768.c: mult768.c: In function ‘squeeze_7681_x16’:
mult768.c: mult768.c:23:17: note: The ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
mult768.c: static int16x16 squeeze_7681_x16(int16x16 x)
mult768.c: ^~~~~~~~~~~~~~~~
mult768.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
mult768.c: from mult768.c:2:
mult768.c: ...
mult768.c: from mult768.c:2:
mult768.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avxintrin.h:1300:1: error: inlining failed in call to always_inline ‘_mm256_set1_epi16’: target specific option mismatch
mult768.c: _mm256_set1_epi16 (short __A)
mult768.c: ^~~~~~~~~~~~~~~~~
mult768.c: mult768.c:12:17: note: called from here
mult768.c: #define sub_x16 _mm256_sub_epi16
mult768.c: ^
mult768.c: mult768.c:20:10: note: in expansion of macro ‘sub_x16’
mult768.c: return sub_x16(x,mullo_x16(mulhrs_x16(x,const_x16(7)),const_x16(4591)));
mult768.c: ^~~~~~~

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE avx
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE avx
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE avx
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE avx