Implementation notes: amd64, hydra8, crypto_core/multsntrup1277

Computer: hydra8
Microarchitecture: amd64; Ivy Bridge+AES (306a9)
Architecture: amd64
CPU ID: GenuineIntel-000306a9-bfebfbff
SUPERCOP version: 20240808
Operation: crypto_core
Primitive: multsntrup1277

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
569064	3652 0 0	17549 804 992	`ref`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
815054	2114 0 0	15784 860 928	`compact`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
830675	4040 0 0	19648 860 928	`compact`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
944720	1582 0 0	15485 804 992	`compact`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
1075765	1726 0 0	15384 860 928	`ref`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
1084429	2318 0 0	17920 860 928	`ref`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
1466116	1687 0 0	15952 860 928	`ref`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
1608632	1504 0 0	15776 860 928	`compact`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
3372615	467 0 0	12384 860 928	`compact`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
3679375	674 0 0	12592 860 928	`ref`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
3699046	611 0 0	12202 852 928	`ref`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
3969660	722 0 0	13389 804 992	`ref`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
4000275	670 0 0	13028 796 992	`ref`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
4685403	564 0 0	11784 780 960	`ref`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
4718868	1315 0 0	12890 852 928	`compact`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
4922293	572 0 0	13237 804 992	`compact`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
4986172	521 0 0	11728 780 960	`compact`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808
5136855	564 0 0	12908 796 992	`compact`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240809	20240808

Compiler output

mult1280.c: mult1280.c:450:21: error: always_inline function '_mm256_add_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_multsntrup1277_avx_constbranchindex' that is compiled without support for 'avx2'
mult1280.c:     x = add_x16(fgi,add_x16(fgip,fgip1));
mult1280.c:                     ^
mult1280.c: mult1280.c:12:17: note: expanded from macro 'add_x16'
mult1280.c: #define add_x16 _mm256_add_epi16
mult1280.c:                 ^
mult1280.c: mult1280.c:450:9: error: always_inline function '_mm256_add_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_multsntrup1277_avx_constbranchindex' that is compiled without support for 'avx2'
mult1280.c:     x = add_x16(fgi,add_x16(fgip,fgip1));
mult1280.c:         ^
mult1280.c: mult1280.c:12:17: note: expanded from macro 'add_x16'
mult1280.c: #define add_x16 _mm256_add_epi16
mult1280.c:                 ^
mult1280.c: 2 errors generated.

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

Implementation	Compiler
`avx`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`avx`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`avx`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`avx`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`

Compiler output

mult1280.c: mult1280.c:426:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup1277_avx_constbranchindex' that is compiled without support for 'avx'
mult1280.c:   x = const_x16(0);
mult1280.c:       ^
mult1280.c: mult1280.c:11:19: note: expanded from macro 'const_x16'
mult1280.c: #define const_x16 _mm256_set1_epi16
mult1280.c:                   ^
mult1280.c: mult1280.c:426:7: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult1280.c: mult1280.c:11:19: note: expanded from macro 'const_x16'
mult1280.c: #define const_x16 _mm256_set1_epi16
mult1280.c:                   ^
mult1280.c: mult1280.c:427:36: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup1277_avx_constbranchindex' that is compiled without support for 'avx'
mult1280.c:   for (i = p&~15;i < 1280;i += 16) store_x16(&f[i],x);
mult1280.c:                                    ^
mult1280.c: mult1280.c:10:24: note: expanded from macro 'store_x16'
mult1280.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult1280.c:                        ^
mult1280.c: mult1280.c:427:36: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult1280.c: mult1280.c:10:24: note: expanded from macro 'store_x16'
mult1280.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult1280.c:                        ^
mult1280.c: mult1280.c:428:36: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup1277_avx_constbranchindex' that is compiled without support for 'avx'
mult1280.c:   for (i = p&~15;i < 1280;i += 16) store_x16(&g[i],x);
mult1280.c:                                    ^
mult1280.c: mult1280.c:10:24: note: expanded from macro 'store_x16'
mult1280.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult1280.c: ...

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

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

Compiler output

mult1280.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
mult1280.c:                  from mult1280.c:2:
mult1280.c: mult1280.c: In function 'squeeze_7879_x16':
mult1280.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:808:1: error: inlining failed in call to 'always_inline' '_mm256_sub_epi16': target specific option mismatch
mult1280.c:   808 | _mm256_sub_epi16 (__m256i __A, __m256i __B)
mult1280.c:       | ^~~~~~~~~~~~~~~~
mult1280.c: mult1280.c:13:17: note: called from here
mult1280.c:    13 | #define sub_x16 _mm256_sub_epi16
mult1280.c:       |                 ^
mult1280.c: mult1280.c:21:10: note: in expansion of macro 'sub_x16'
mult1280.c:    21 |   return sub_x16(x,mullo_x16(mulhrs_x16(x,const_x16(4)),const_x16(7879)));
mult1280.c:       |          ^~~~~~~
mult1280.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
mult1280.c:                  from mult1280.c:2:
mult1280.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:553:1: error: inlining failed in call to 'always_inline' '_mm256_mullo_epi16': target specific option mismatch
mult1280.c:   553 | _mm256_mullo_epi16 (__m256i __A, __m256i __B)
mult1280.c:       | ^~~~~~~~~~~~~~~~~~
mult1280.c: mult1280.c:13:17: note: called from here
mult1280.c:    13 | #define sub_x16 _mm256_sub_epi16
mult1280.c:       |                 ^
mult1280.c: mult1280.c:21:10: note: in expansion of macro 'sub_x16'
mult1280.c:    21 |   return sub_x16(x,mullo_x16(mulhrs_x16(x,const_x16(4)),const_x16(7879)));
mult1280.c:       |          ^~~~~~~
mult1280.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
mult1280.c:                  from mult1280.c:2:
mult1280.c: ...

Number of similar (implementation,compiler) pairs: 8, namely:

Implementation	Compiler
`avx`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx800`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx800`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx800`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`avx800`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`

Compiler output

mult1280.c: mult1280.c:450:21: error: always_inline function '_mm256_add_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_multsntrup1277_avx800_constbranchindex' that is compiled without support for 'avx2'
mult1280.c:     x = add_x16(fgi,add_x16(fgip,fgip1));
mult1280.c:                     ^
mult1280.c: mult1280.c:12:17: note: expanded from macro 'add_x16'
mult1280.c: #define add_x16 _mm256_add_epi16
mult1280.c:                 ^
mult1280.c: mult1280.c:450:9: error: always_inline function '_mm256_add_epi16' requires target feature 'avx2', but would be inlined into function 'crypto_core_multsntrup1277_avx800_constbranchindex' that is compiled without support for 'avx2'
mult1280.c:     x = add_x16(fgi,add_x16(fgip,fgip1));
mult1280.c:         ^
mult1280.c: mult1280.c:12:17: note: expanded from macro 'add_x16'
mult1280.c: #define add_x16 _mm256_add_epi16
mult1280.c:                 ^
mult1280.c: 2 errors generated.

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

Implementation	Compiler
`avx800`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`avx800`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`avx800`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`avx800`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`

Compiler output

mult1280.c: mult1280.c:426:7: error: always_inline function '_mm256_set1_epi16' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup1277_avx800_constbranchindex' that is compiled without support for 'avx'
mult1280.c:   x = const_x16(0);
mult1280.c:       ^
mult1280.c: mult1280.c:11:19: note: expanded from macro 'const_x16'
mult1280.c: #define const_x16 _mm256_set1_epi16
mult1280.c:                   ^
mult1280.c: mult1280.c:426:7: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult1280.c: mult1280.c:11:19: note: expanded from macro 'const_x16'
mult1280.c: #define const_x16 _mm256_set1_epi16
mult1280.c:                   ^
mult1280.c: mult1280.c:427:36: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup1277_avx800_constbranchindex' that is compiled without support for 'avx'
mult1280.c:   for (i = p&~15;i < 1280;i += 16) store_x16(&f[i],x);
mult1280.c:                                    ^
mult1280.c: mult1280.c:10:24: note: expanded from macro 'store_x16'
mult1280.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult1280.c:                        ^
mult1280.c: mult1280.c:427:36: error: AVX vector argument of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
mult1280.c: mult1280.c:10:24: note: expanded from macro 'store_x16'
mult1280.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult1280.c:                        ^
mult1280.c: mult1280.c:428:36: error: always_inline function '_mm256_storeu_si256' requires target feature 'avx', but would be inlined into function 'crypto_core_multsntrup1277_avx800_constbranchindex' that is compiled without support for 'avx'
mult1280.c:   for (i = p&~15;i < 1280;i += 16) store_x16(&g[i],x);
mult1280.c:                                    ^
mult1280.c: mult1280.c:10:24: note: expanded from macro 'store_x16'
mult1280.c: #define store_x16(p,v) _mm256_storeu_si256((int16x16 *) (p),(v))
mult1280.c: ...

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

Implementation	Compiler
`avx800`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`

Passed TIMECOP

TIMECOP iterations: 1

Number of similar (implementation,compiler) pairs: 18, namely:

Implementation	Compiler
`compact`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`compact`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`compact`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`compact`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`compact`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`compact`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`compact`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`compact`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`compact`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`ref`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`ref`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`ref`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`ref`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`ref`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)`
`ref`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`ref`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`ref`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`
`ref`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)`