Implementation notes: amd64, comet, crypto_auth/siphash24

Computer: comet
Microarchitecture: amd64; Comet Lake (806ec)
Architecture: amd64
CPU ID: GenuineIntel-000806ec-bfebfbff
SUPERCOP version: 20240107
Operation: crypto_auth
Primitive: siphash24

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
4357	753 0 0	13879 844 960	`T:sandy2`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4365	525 0 0	12849 852 928	`T:sandy`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4385	769 0 0	15401 852 928	`T:sandy2`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4389	820 0 0	17113 852 960	`T:little2`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4390	820 0 0	16817 852 928	`T:little2`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4391	801 0 0	13927 844 960	`T:little2`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4404	809 0 0	13113 852 928	`T:little2`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4408	768 0 0	16769 852 928	`T:sandy2`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4408	758 0 0	13065 852 928	`T:sandy2`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4412	768 0 0	17065 852 960	`T:sandy2`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4432	897 0 0	15492 780 992	`T:little2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4435	703 0 0	15337 852 928	`T:little2`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4441	797 0 0	17113 852 960	`T:little`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4444	733 0 0	15385 852 928	`T:sandy`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4449	600 0 0	13324 780 992	`T:little`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4449	749 0 0	16769 852 928	`T:sandy`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4457	777 0 0	13027 772 992	`T:sandy2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4459	749 0 0	17065 852 960	`T:sandy`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4462	797 0 0	16817 852 928	`T:little`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4462	948 0 0	13652 780 992	`T:ref_le`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4468	685 0 0	15337 852 928	`T:little`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4470	810 0 0	13059 772 992	`T:little2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4485	840 0 0	13540 780 992	`T:little2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4487	779 0 0	15388 780 992	`T:little`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4493	948 0 0	15540 780 992	`T:ref_le`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4494	800 0 0	13492 780 992	`T:sandy2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4500	865 0 0	15460 780 992	`T:sandy2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4502	763 0 0	15372 780 992	`T:sandy`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4507	573 0 0	11695 756 960	`T:sandy`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4516	537 0 0	11647 756 960	`T:little`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4539	662 0 0	12939 772 992	`T:little`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4541	550 0 0	13703 844 960	`T:little`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4548	646 0 0	12923 772 992	`T:sandy`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4609	874 0 0	11967 756 960	`T:ref_le`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4650	753 0 0	11839 756 960	`T:sandy2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
4801	573 0 0	12897 852 928	`T:little`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4988	901 0 0	14031 844 960	`T:ref_le`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4994	911 0 0	16921 852 928	`T:ref_le`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
4999	937 0 0	13203 772 992	`T:ref_le`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
5011	911 0 0	17217 852 960	`T:ref_le`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
5011	911 0 0	13225 852 928	`T:ref_le`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
5115	807 0 0	15449 852 928	`T:ref_le`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
5161	702 0 0	11799 756 960	`T:little2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
5174	600 0 0	13324 780 992	`T:sandy`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
5178	510 0 0	13671 844 960	`T:sandy`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
7012	1261 0 0	13489 852 928	`T:sse2-1`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
7050	1117 0 0	17449 852 960	`T:sse2-1`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
7071	1117 0 0	17153 852 928	`T:sse2-1`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
7126	766 0 0	13492 780 992	`T:mmx`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7145	702 0 0	13879 844 960	`T:sse2-1`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
7165	1197 0 0	15865 852 928	`T:sse2-1`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
7183	735 0 0	13019 772 992	`T:mmx`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7190	1081 0 0	15692 780 992	`T:mmx`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7250	682 0 0	11815 756 960	`T:mmx`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7286	687 0 0	11823 756 960	`T:sse2-1`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7356	806 0 0	13532 780 992	`T:sse2-1`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7389	741 0 0	13027 772 992	`T:sse2-1`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
7407	1097 0 0	15708 780 992	`T:sse2-1`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
8846	2020 0 0	18249 852 960	`T:sse41`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
8984	1101 0 0	17137 852 928	`T:sse41`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
9009	1049 0 0	15660 780 992	`T:sse41`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
9052	1292 0 0	13537 852 928	`T:sse41`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
9128	690 0 0	13863 844 960	`T:sse41`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
9181	658 0 0	11791 756 960	`T:sse41`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
9190	774 0 0	13500 780 992	`T:sse41`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
9510	712 0 0	13011 772 992	`T:sse41`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20240105	20231222
10344	997 0 0	15657 852 928	`T:mmx`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
10354	1029 0 0	17057 852 928	`T:mmx`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
10383	1029 0 0	17353 852 960	`T:mmx`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
10420	642 0 0	13799 844 960	`T:mmx`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222
10433	677 0 0	13009 852 928	`T:mmx`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20240105	20231222

Compiler output

Implementation: T:sse2-1
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

siphash.c: siphash.c:24:9: warning: '_mm_roti_epi64' macro redefined [-Wmacro-redefined]
siphash.c: #define _mm_roti_epi64(x, c) ((16 == (c)) ? _mm_shufflelo_epi16((x), _MM_SHUFFLE(2,1,0,3)) : _mm_xor_si128(_mm_slli_epi64((x), (c)), _mm_srli_epi64((x), 64-(c))))
siphash.c: ^
siphash.c: /usr/lib/llvm-14/lib/clang/14.0.6/include/xopintrin.h:236:9: note: previous definition is here
siphash.c: #define _mm_roti_epi64(A, N) \
siphash.c: ^
siphash.c: 1 warning generated.

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse2-1
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse2-1
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse2-1
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse2-1
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse2-1

Compiler output

Implementation: T:sse41
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse41
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse41
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse41
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse41

Compiler output

Implementation: T:sse41
Security model: timingleaks
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

siphash.c: siphash.c:27:9: warning: '_mm_roti_epi64' macro redefined [-Wmacro-redefined]
siphash.c: #define _mm_roti_epi64(x, c) (((c) == 16) ? rotate16((x)) : _mm_xor_si128(_mm_slli_epi64((x), (c)), _mm_srli_epi64((x), 64-(c))))
siphash.c: ^
siphash.c: /usr/lib/llvm-14/lib/clang/14.0.6/include/xopintrin.h:236:9: note: previous definition is here
siphash.c: #define _mm_roti_epi64(A, N) \
siphash.c: ^
siphash.c: siphash.c:70:39: error: '__builtin_ia32_pblendw128' needs target feature sse4.1
siphash.c: for(k = 0; k < SIPHASH_ROUNDS; ++k) COMPRESS(v02,v13);
siphash.c: ^
siphash.c: siphash.c:60:3: note: expanded from macro 'COMPRESS'
siphash.c: HALF_ROUND(v02,v13,13,16); \
siphash.c: ^
siphash.c: siphash.c:53:60: note: expanded from macro 'HALF_ROUND'
siphash.c: b1 = _mm_roti_epi64(b, s); b2 = _mm_roti_epi64(b, t); b = _mm_blend_epi16(b1, b2, 0xF0); \
siphash.c: ^
siphash.c: /usr/lib/llvm-14/lib/clang/14.0.6/include/smmintrin.h:520:14: note: expanded from macro '_mm_blend_epi16'
siphash.c: ((__m128i) __builtin_ia32_pblendw128 ((__v8hi)(__m128i)(V1), \
siphash.c: ^
siphash.c: siphash.c:70:39: error: '__builtin_ia32_pblendw128' needs target feature sse4.1
siphash.c: siphash.c:62:3: note: expanded from macro 'COMPRESS'
siphash.c: HALF_ROUND(v02,v13,17,21); \
siphash.c: ^
siphash.c: siphash.c:53:60: note: expanded from macro 'HALF_ROUND'
siphash.c: b1 = _mm_roti_epi64(b, s); b2 = _mm_roti_epi64(b, t); b = _mm_blend_epi16(b1, b2, 0xF0); \
siphash.c: ^
siphash.c: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse41