Implementation notes: amd64, jasper3, crypto_aead/norx6461v3

Computer: jasper3
Microarchitecture: amd64; Tremont (906c0)
Architecture: amd64
CPU ID: GenuineIntel-000906c0-20-bfebfbff
SUPERCOP version: 20240625
Operation: crypto_aead
Primitive: norx6461v3

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
40972	18724 8 0	38991 780 1080	`T:xmm`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
42727	17415 8 0	39400 788 1080	`T:xmm`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
42940	14406 8 0	33219 764 1048	`T:xmm`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
43265	18212 8 0	39016 788 1080	`T:xmm`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
43832	19110 8 0	42656 820 1016	`T:xmm`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
43881	19078 8 0	40320 820 1016	`T:xmm`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
44250	18135 8 0	37494 812 1016	`T:xmm`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
48338	20703 8 0	42904 820 1016	`T:xmm`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
48710	19475 8 0	39278 812 1016	`T:xmm`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
60019	5908 16 0	27988 796 1080	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
60243	4732 16 0	25620 796 1080	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
61277	3373 16 0	22316 772 1048	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
63582	4797 16 0	27216 828 1016	`T:ref`	`clang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
63903	3985 16 0	23518 820 1016	`T:ref`	`clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
64643	4745 16 0	28536 828 1016	`T:ref`	`clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
65236	4921 16 0	26408 828 1016	`T:ref`	`clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
67430	4166 16 0	24528 788 1080	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625
68068	4544 16 0	24662 820 1016	`T:ref`	`clang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall`	20240627	20240625

Compiler output

norx.c: norx.c:391:23: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'norx_aead_encrypt' that is compiled without support for 'avx'
norx.c:     const __m256i K = LOADU(key);
norx.c:                       ^
norx.c: norx.c:53:19: note: expanded from macro 'LOADU'
norx.c: #define LOADU(in) _mm256_loadu_si256((__m256i*)(in))
norx.c:                   ^
norx.c: norx.c:391:23: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
norx.c: norx.c:53:19: note: expanded from macro 'LOADU'
norx.c: #define LOADU(in) _mm256_loadu_si256((__m256i*)(in))
norx.c:                   ^
norx.c: norx.c:395:5: error: always_inline function '_mm256_loadu_si256' requires target feature 'avx', but would be inlined into function 'norx_aead_encrypt' that is compiled without support for 'avx'
norx.c:     INITIALISE(A, B, C, D, nonce, K);
norx.c:     ^
norx.c: norx.c:293:9: note: expanded from macro 'INITIALISE'
norx.c:     A = LOADU(NONCE);                                                      \
norx.c:         ^
norx.c: norx.c:53:19: note: expanded from macro 'LOADU'
norx.c: #define LOADU(in) _mm256_loadu_si256((__m256i*)(in))
norx.c:                   ^
norx.c: norx.c:395:5: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
norx.c: norx.c:293:9: note: expanded from macro 'INITIALISE'
norx.c:     A = LOADU(NONCE);                                                      \
norx.c:         ^
norx.c: norx.c:53:19: note: expanded from macro 'LOADU'
norx.c: #define LOADU(in) _mm256_loadu_si256((__m256i*)(in))
norx.c: ...

Number of similar (implementation,compiler) pairs: 5, namely:

Implementation	Compiler
`T:ymm`	`clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`T:ymm`	`clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`T:ymm`	`clang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`T:ymm`	`clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`
`T:ymm`	`clang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Debian_Clang_11.0.1)`

Compiler output

norx.c: norx.c: In function 'norx_aead_encrypt':
norx.c: norx.c:391:19: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
norx.c:   391 |     const __m256i K = LOADU(key);
norx.c:       |                   ^
norx.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:53,
norx.c:                  from norx.c:25:
norx.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avx2intrin.h:913:1: error: inlining failed in call to 'always_inline' '_mm256_xor_si256': target specific option mismatch
norx.c:   913 | _mm256_xor_si256 (__m256i __A, __m256i __B)
norx.c:       | ^~~~~~~~~~~~~~~~
norx.c: norx.c:72:19: note: called from here
norx.c:    72 | #define XOR(A, B) _mm256_xor_si256((A), (B))
norx.c:       |                   ^~~~~~~~~~~~~~~~~~~~~~~~~~
norx.c: norx.c:297:9: note: in expansion of macro 'XOR'
norx.c:   297 |     D = XOR(D, _mm256_set_epi64x(NORX_T, NORX_P, NORX_L, NORX_W));         \
norx.c:       |         ^~~
norx.c: norx.c:395:5: note: in expansion of macro 'INITIALISE'
norx.c:   395 |     INITIALISE(A, B, C, D, nonce, K);
norx.c:       |     ^~~~~~~~~~
norx.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:51,
norx.c:                  from norx.c:25:
norx.c: /usr/lib/gcc/x86_64-linux-gnu/10/include/avxintrin.h:1296:1: error: inlining failed in call to 'always_inline' '_mm256_set_epi64x': target specific option mismatch
norx.c:  1296 | _mm256_set_epi64x (long long __A, long long __B, long long __C,
norx.c:       | ^~~~~~~~~~~~~~~~~
norx.c: norx.c:72:19: note: called from here
norx.c:    72 | #define XOR(A, B) _mm256_xor_si256((A), (B))
norx.c: ...

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

Implementation	Compiler
`T:ymm`	`gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`T:ymm`	`gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`T:ymm`	`gcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`
`T:ymm`	`gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (10.2.1_20210110)`