Implementation notes: amd64, rumba5, crypto_kem/ntskem13136

Computer: rumba5
Microarchitecture: amd64; Zen (800f11)
Architecture: amd64
CPU ID: AuthenticAMD-00800f11-178bfbff
SUPERCOP version: 20240625
Operation: crypto_kem
Primitive: ntskem13136

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
1872823	148917 84 16	169565 980 1728	`T:avx2`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
1969300	100671 84 16	121477 980 1728	`T:avx2`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2010810	94297 84 16	115213 980 1728	`T:sse2`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2017501	107501 84 16	125629 980 1728	`T:avx2`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2091590	140100 84 16	160925 980 1728	`T:sse2`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2119283	75738 84 16	94423 972 1824	`T:sse2`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2156533	100688 84 16	118821 980 1728	`T:sse2`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2273306	87887 84 16	105879 956 1792	`T:avx2`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2275721	81901 84 16	100455 972 1824	`T:avx2`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2277394	112435 84 16	132295 956 1792	`T:avx2`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2325538	90447 84 16	109191 956 1792	`T:avx2`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2482750	123609 84 16	145941 980 1728	`T:opt`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
2877084	83238 84 16	100326 948 1760	`T:avx2`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3074599	75403 84 16	95863 956 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3129555	165809 84 16	187413 980 1728	`T:opt`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3133889	99967 84 16	121535 956 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3189562	92546 84 16	114965 980 1728	`T:opt`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3252376	72052 84 16	92007 972 1824	`T:opt`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3266730	72923 84 16	92647 956 1792	`T:opt`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3279335	96659 84 16	116317 980 1728	`T:opt`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
3800309	69946 84 16	88534 948 1760	`T:opt`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
23661353	38672 76 16	60581 964 1728	`T:ref`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
24719649	40692 76 16	63829 964 1728	`T:ref`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
24889639	38064 76 16	60789 964 1728	`T:ref`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
26675545	22846 76 16	43031 956 1824	`T:ref`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
27278757	34265 76 16	55839 924 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
27323246	29589 76 16	49205 964 1728	`T:ref`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
28090086	26942 76 16	47351 924 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
28188523	21157 76 16	39718 916 1760	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
29876481	23869 76 16	43559 924 1792	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625

Compiler output

bitslice_bma_128.c: bitslice_bma_128.c:315:21: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'bitslice_bma' that is compiled without support for 'ssse3'
bitslice_bma_128.c:         out[1][j] = _mm_shuffle_epi8(out[0][j], mask);
bitslice_bma_128.c:                     ^
bitslice_bma_128.c: bitslice_bma_128.c:316:21: error: '__builtin_ia32_palignr128' needs target feature ssse3
bitslice_bma_128.c:         out[0][j] = _mm_alignr_epi8(out[0][j], psi[0][j], 15);
bitslice_bma_128.c:                     ^
bitslice_bma_128.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/tmmintrin.h:152:13: note: expanded from macro '_mm_alignr_epi8'
bitslice_bma_128.c:   ((__m128i)__builtin_ia32_palignr128((__v16qi)(__m128i)(a), \
bitslice_bma_128.c:             ^
bitslice_bma_128.c: 2 errors generated.

Number of similar (implementation,compiler) pairs: 2, namely:

Implementation	Compiler
`T:avx2`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`T:sse2`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`

Compiler output

m4r.c: m4r.c: In function 'zero_vector':
m4r.c: m4r.c:85:20: error: incompatible types when assigning to type 'vector' {aka '__m128i'} from type '__m256i'
m4r.c:    85 |         *vec_ptr = _mm256_setzero_si256(); vec_ptr++;
m4r.c:       |                    ^~~~~~~~~~~~~~~~~~~~
m4r.c: m4r.c:86:20: error: incompatible types when assigning to type 'vector' {aka '__m128i'} from type '__m256i'
m4r.c:    86 |         *vec_ptr = _mm256_setzero_si256(); vec_ptr++;
m4r.c:       |                    ^~~~~~~~~~~~~~~~~~~~
m4r.c: m4r.c: In function '_m4ri_make_table_rev':
m4r.c: m4r.c:147:12: error: incompatible types when assigning to type 'vector' {aka '__m128i'} from type '__m256i'
m4r.c:   147 |     mask = _mm256_set_epi64x(v[3], v[2], v[1], v[0]);
m4r.c:       |            ^~~~~~~~~~~~~~~~~
m4r.c: m4r.c:196:46: error: incompatible type for argument 1 of '_mm256_and_si256'
m4r.c:   196 |     S_ptr[nblocks-1] = _mm256_and_si256(S_ptr[nblocks-1], mask);
m4r.c:       |                                         ~~~~~^~~~~~~~~~~
m4r.c:       |                                              |
m4r.c:       |                                              vector {aka __m128i}
m4r.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
m4r.c:                  from bits.h:28,
m4r.c:                  from m4r.c:26:
m4r.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:179:27: note: expected '__m256i' but argument is of type 'vector' {aka '__m128i'}
m4r.c:   179 | _mm256_and_si256 (__m256i __A, __m256i __B)
m4r.c:       |                   ~~~~~~~~^~~
m4r.c: m4r.c:196:59: error: incompatible type for argument 2 of '_mm256_and_si256'
m4r.c:   196 |     S_ptr[nblocks-1] = _mm256_and_si256(S_ptr[nblocks-1], mask);
m4r.c:       |                                                           ^~~~
m4r.c: ...

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

Implementation	Compiler
`T:sse2`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`T:sse2`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`T:sse2`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`T:sse2`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`