Implementation notes: amd64, h9trinity, crypto_core/invsntrup761

Computer: h9trinity
Architecture: amd64
CPU ID: AuthenticAMD-00610f01-178bfbff
SUPERCOP version: 20200618
Operation: crypto_core
Primitive: invsntrup761

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
8744598	4275 0 0	17384 784 832	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200410	20200409
41789869	3024 0 0	15057 776 776	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200410	20200409
41849324	3024 0 0	15057 776 776	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200410	20200409
43213980	6208 0 0	19809 776 776	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200410	20200409
44756986	4031 0 0	16545 776 760	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200410	20200409
60145324	1130 0 0	11845 768 832	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200410	20200409
60534109	868 0 0	10773 760 800	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200410	20200409
62819542	900 0 0	10399 768 760	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200410	20200409
65531004	1046 0 0	11789 768 832	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200410	20200409

Compiler output

Implementation: avx
Security model: unknown
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

recip.c: recip.c:85:23: error: always_inline function '_mm256_mullo_epi16' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: __m256i f0vecqinv = _mm256_mullo_epi16(f0vec,qinvvec);
recip.c: ^
recip.c: recip.c:86:23: error: always_inline function '_mm256_mullo_epi16' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: __m256i g0vecqinv = _mm256_mullo_epi16(g0vec,qinvvec);
recip.c: ^
recip.c: recip.c:92:21: error: always_inline function '_mm256_blendv_epi8' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: __m256i finew = _mm256_blendv_epi8(fi,gi,maskvec);
recip.c: ^
recip.c: recip.c:93:21: error: always_inline function '_mm256_blendv_epi8' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: __m256i ginew = _mm256_blendv_epi8(gi,fi,maskvec);
recip.c: ^
recip.c: recip.c:94:13: error: always_inline function '_mm256_sub_epi16' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: ginew = _mm256_sub_epi16(montproduct(ginew,f0vec,f0vecqinv),montproduct(finew,g0vec,g0vecqinv));
recip.c: ^
recip.c: recip.c:75:7: error: always_inline function '_mm256_mullo_epi16' requires target feature 'avx2', but would be inlined into function 'montproduct' that is compiled without support for 'avx2'
recip.c: d = _mm256_mullo_epi16(x,yqinv);
recip.c: ^
recip.c: recip.c:76:8: error: always_inline function '_mm256_mulhi_epi16' requires target feature 'avx2', but would be inlined into function 'montproduct' that is compiled without support for 'avx2'
recip.c: hi = _mm256_mulhi_epi16(x,y);
recip.c: ^
recip.c: recip.c:77:7: error: always_inline function '_mm256_mulhi_epi16' requires target feature 'avx2', but would be inlined into function 'montproduct' that is compiled without support for 'avx2'
recip.c: e = _mm256_mulhi_epi16(d,qvec);
recip.c: ^
recip.c: recip.c:78:10: error: always_inline function '_mm256_sub_epi16' requires target feature 'avx2', but would be inlined into function 'montproduct' that is compiled without support for 'avx2'
recip.c: ...

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

Compiler	Implementations
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: avx
Security model: unknown
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

recip.c: recip.c:83:19: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'sse4.2'
recip.c: __m256i f0vec = _mm256_set1_epi16(f0);
recip.c: ^
recip.c: recip.c:84:19: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'sse4.2'
recip.c: __m256i g0vec = _mm256_set1_epi16(g0);
recip.c: ^
recip.c: recip.c:85:23: error: always_inline function '_mm256_mullo_epi16' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: __m256i f0vecqinv = _mm256_mullo_epi16(f0vec,qinvvec);
recip.c: ^
recip.c: recip.c:85:48: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'sse4.2'
recip.c: __m256i f0vecqinv = _mm256_mullo_epi16(f0vec,qinvvec);
recip.c: ^
recip.c: recip.c:69:17: note: expanded from macro 'qinvvec'
recip.c: #define qinvvec _mm256_set1_epi16(qinv)
recip.c: ^
recip.c: recip.c:86:23: error: always_inline function '_mm256_mullo_epi16' requires target feature 'avx2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'avx2'
recip.c: __m256i g0vecqinv = _mm256_mullo_epi16(g0vec,qinvvec);
recip.c: ^
recip.c: recip.c:86:48: error: always_inline function '_mm256_set1_epi16' requires target feature 'sse4.2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'sse4.2'
recip.c: __m256i g0vecqinv = _mm256_mullo_epi16(g0vec,qinvvec);
recip.c: ^
recip.c: recip.c:69:17: note: expanded from macro 'qinvvec'
recip.c: #define qinvvec _mm256_set1_epi16(qinv)
recip.c: ^
recip.c: recip.c:87:21: error: always_inline function '_mm256_set1_epi32' requires target feature 'sse4.2', but would be inlined into function 'vectormodq_swapeliminate' that is compiled without support for 'sse4.2'
recip.c: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx

Compiler output

Implementation: avx
Security model: unknown
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

recip.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
recip.c: from recip.c:1:
recip.c: recip.c: In function 'montproduct':
recip.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:808:1: error: inlining failed in call to always_inline '_mm256_sub_epi16': target specific option mismatch
recip.c: _mm256_sub_epi16 (__m256i __A, __m256i __B)
recip.c: ^~~~~~~~~~~~~~~~
recip.c: recip.c:78:10: note: called from here
recip.c: return _mm256_sub_epi16(hi,e);
recip.c: ^~~~~~~~~~~~~~~~~~~~~~
recip.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
recip.c: from recip.c:1:
recip.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:546:1: error: inlining failed in call to always_inline '_mm256_mulhi_epi16': target specific option mismatch
recip.c: _mm256_mulhi_epi16 (__m256i __A, __m256i __B)
recip.c: ^~~~~~~~~~~~~~~~~~
recip.c: recip.c:77:5: note: called from here
recip.c: e = _mm256_mulhi_epi16(d,qvec);
recip.c: ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~
recip.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
recip.c: from recip.c:1:
recip.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:546:1: error: inlining failed in call to always_inline '_mm256_mulhi_epi16': target specific option mismatch
recip.c: _mm256_mulhi_epi16 (__m256i __A, __m256i __B)
recip.c: ^~~~~~~~~~~~~~~~~~
recip.c: recip.c:76:6: note: called from here
recip.c: hi = _mm256_mulhi_epi16(x,y);
recip.c: ~~~^~~~~~~~~~~~~~~~~~~~~~~~~
recip.c: ...

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

Compiler	Implementations
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

Compiler output

Implementation: jumpdivsteps
Security model: unknown
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

avx-768.c: avx-768.c:1406:12: error: always_inline function '_mm256_mulhi_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_invsntrup761_jumpdivsteps_crt768' that is compiled without support for 'avx2'
avx-768.c: g[0] = _mm256_mulhi_epi16(f[0],qrecip_769_x16);
avx-768.c: ^
avx-768.c: avx-768.c:1407:12: error: always_inline function '_mm256_mulhi_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_invsntrup761_jumpdivsteps_crt768' that is compiled without support for 'avx2'
avx-768.c: g[1] = _mm256_mulhi_epi16(f[1],qrecip_7681_x16);
avx-768.c: ^
avx-768.c: avx-768.c:1408:12: error: always_inline function '_mm256_mulhi_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_invsntrup761_jumpdivsteps_crt768' that is compiled without support for 'avx2'
avx-768.c: g[2] = _mm256_mulhi_epi16(f[2],qrecip_10753_x16);
avx-768.c: ^
avx-768.c: avx-768.c:1412:12: error: always_inline function '_mm256_srai_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_invsntrup761_jumpdivsteps_crt768' that is compiled without support for 'avx2'
avx-768.c: g[0] = _mm256_srai_epi16(g[0],8);
avx-768.c: ^
avx-768.c: 4 errors generated.

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	jumpdivsteps
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	jumpdivsteps
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	jumpdivsteps
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	jumpdivsteps

Compiler output

Implementation: jumpdivsteps
Security model: unknown
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

avx-768.c: avx-768.c:544:36: error: invalid output size for constraint '+x'
avx-768.c: __asm__("vpsubw %1,%0,%0" : "+x"(a),"+x"(b));
avx-768.c: ^
avx-768.c: avx-768.c:550:36: error: invalid output size for constraint '+x'
avx-768.c: __asm__("vpaddw %1,%0,%0" : "+x"(a),"+x"(b));
avx-768.c: ^
avx-768.c: 2 errors generated.

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	jumpdivsteps

Compiler output

Implementation: jumpdivsteps
Security model: unknown
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

avx-768.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
avx-768.c: from avx-768.c:4:
avx-768.c: avx-768.c: In function 'reduce_x16':
avx-768.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:553:1: error: inlining failed in call to always_inline '_mm256_mullo_epi16': target specific option mismatch
avx-768.c: _mm256_mullo_epi16 (__m256i __A, __m256i __B)
avx-768.c: ^~~~~~~~~~~~~~~~~~
avx-768.c: avx-768.c:558:5: note: called from here
avx-768.c: y = _mm256_mullo_epi16(y,q_x16);
avx-768.c: ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
avx-768.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
avx-768.c: from avx-768.c:4:
avx-768.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:531:1: error: inlining failed in call to always_inline '_mm256_mulhrs_epi16': target specific option mismatch
avx-768.c: _mm256_mulhrs_epi16 (__m256i __X, __m256i __Y)
avx-768.c: ^~~~~~~~~~~~~~~~~~~
avx-768.c: avx-768.c:557:5: note: called from here
avx-768.c: y = _mm256_mulhrs_epi16(y,qshift_x16);
avx-768.c: ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
avx-768.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/7/include/immintrin.h:43:0,
avx-768.c: from avx-768.c:4:
avx-768.c: /usr/lib/gcc/x86_64-linux-gnu/7/include/avx2intrin.h:546:1: error: inlining failed in call to always_inline '_mm256_mulhi_epi16': target specific option mismatch
avx-768.c: _mm256_mulhi_epi16 (__m256i __A, __m256i __B)
avx-768.c: ^~~~~~~~~~~~~~~~~~
avx-768.c: avx-768.c:556:11: note: called from here
avx-768.c: __m256i y = _mm256_mulhi_epi16(x,qrecip_x16);
avx-768.c: ^
avx-768.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	jumpdivsteps
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	jumpdivsteps
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	jumpdivsteps
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	jumpdivsteps