Implementation notes: aarch64, pi3aplus, crypto_hashblocks/sha512

Computer: pi3aplus
Microarchitecture: aarch64; Cortex-A53 (410fd034)
Architecture: aarch64
CPU ID: 410fd034
SUPERCOP version: 20240107
Operation: crypto_hashblocks
Primitive: sha512

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
17018	10856 0 0	21152 840 744	`inplace`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
17065	10832 0 0	21136 840 744	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
17089	2608 0 0	12920 840 744	`wflip`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
17133	2600 0 0	11607 832 728	`wflip`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
18278	2348 0 0	13285 768 736	`wflip`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
18339	4124 0 0	14448 840 744	`compact`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
18358	10980 0 0	19999 832 728	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
18371	11000 0 0	20023 832 728	`inplace`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
19288	2632 0 0	10735 816 720	`wflip`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
19744	11876 0 0	22821 768 736	`inplace`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
19902	10888 0 0	19023 816 720	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
19950	10884 0 0	19015 816 720	`inplace`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
19953	11904 0 0	22829 768 736	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
20048	3560 0 0	12607 832 728	`compact`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
21752	3532 0 0	11671 816 720	`compact`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
21975	2636 0 0	11591 832 728	`wflip`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
24145	3964 0 0	14901 768 736	`compact`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
26617	11184 0 0	20167 832 728	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
26708	11184 0 0	20175 832 728	`inplace`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
27791	1736 0 0	12677 768 736	`compact4`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
28882	1968 0 0	12280 840 744	`compact4`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
30278	3660 0 0	12647 832 728	`compact`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
34140	1352 0 0	10375 832 728	`compact4`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
37766	1208 0 0	9343 816 720	`compact4`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
41377	1496 0 0	10519 832 728	`compact3`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
43914	2084 0 0	12392 840 744	`compact3`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
44112	1716 0 0	12661 768 736	`compact3`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
46214	2228 0 0	12552 840 744	`compact2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
46579	1388 0 0	10367 832 728	`compact4`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
47971	1664 0 0	10711 832 728	`compact2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
52389	2216 0 0	13157 768 736	`compact2`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
66768	1568 0 0	9719 816 720	`compact2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
73915	1348 0 0	9487 816 720	`compact3`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
75947	1700 0 0	10695 832 728	`compact2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
79609	1568 0 0	10551 832 728	`compact3`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212

Compiler output

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

inner.c: In file included from inner.c:1:
inner.c: In file included from /usr/lib/llvm-7/lib/clang/7.0.1/include/immintrin.h:28:
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:64:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_vec_init_v2si(__i, 0);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:143:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_packsswb((__v4hi)__m1, (__v4hi)__m2);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:173:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_packssdw((__v2si)__m1, (__v2si)__m2);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:203:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_packuswb((__v4hi)__m1, (__v4hi)__m2);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:230:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_punpckhbw((__v8qi)__m1, (__v8qi)__m2);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:253:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_punpckhwd((__v4hi)__m1, (__v4hi)__m2);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:274:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_punpckhdq((__v2si)__m1, (__v2si)__m2);
inner.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
inner.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/mmintrin.h:301:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
inner.c: return (__m64)__builtin_ia32_punpcklbw((__v8qi)__m1, (__v8qi)__m2);
inner.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: constbranchindex
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

inner.c: inner.c:1:10: fatal error: immintrin.h: No such file or directory
inner.c: #include <immintrin.h>
inner.c: ^~~~~~~~~~~~~
inner.c: compilation terminated.

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: avx2
Security model: constbranchindex
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

inner.S: inner.S:422:5: error: unknown token in expression
inner.S: mov %rsp,%r11
inner.S: ^
inner.S: inner.S:422:5: error: invalid operand
inner.S: mov %rsp,%r11
inner.S: ^
inner.S: inner.S:423:5: error: invalid token in expression
inner.S: and $511,%r11
inner.S: ^
inner.S: inner.S:423:5: error: invalid operand
inner.S: and $511,%r11
inner.S: ^
inner.S: inner.S:424:5: error: invalid token in expression
inner.S: add $416,%r11
inner.S: ^
inner.S: inner.S:424:5: error: invalid operand
inner.S: add $416,%r11
inner.S: ^
inner.S: inner.S:425:5: error: unknown token in expression
inner.S: sub %r11,%rsp
inner.S: ^
inner.S: inner.S:425:5: error: invalid operand
inner.S: sub %r11,%rsp
inner.S: ^
inner.S: inner.S:430:5: error: unknown token in expression
inner.S: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	avx2

Compiler output

Implementation: avx2
Security model: constbranchindex
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

inner.S: inner.S: Assembler messages:
inner.S: inner.S:422: Error: operand 1 must be an integer register -- `mov %rsp,%r11'
inner.S: inner.S:423: Error: operand 1 must be an integer or stack pointer register -- `and $511,%r11'
inner.S: inner.S:424: Error: operand 1 must be an integer or stack pointer register -- `add $416,%r11'
inner.S: inner.S:425: Error: operand 1 must be an integer or stack pointer register -- `sub %r11,%rsp'
inner.S: inner.S:430: Error: operand 1 must be an integer register -- `mov %rcx,%r8'
inner.S: inner.S:435: Error: unknown mnemonic `vmovupd' -- `vmovupd 640(%rcx),%ymm0'
inner.S: inner.S:440: Error: unknown mnemonic `vmovupd' -- `vmovupd 0(%rdi),%ymm1'
inner.S: inner.S:445: Error: operand 1 must be an integer register -- `mov %rdi,%rcx'
inner.S: inner.S:450: Error: unknown mnemonic `vmovupd' -- `vmovupd 32(%rdi),%ymm2'
inner.S: inner.S:455: Error: unknown mnemonic `vpshufb' -- `vpshufb %ymm0,%ymm1,%ymm1'
inner.S: inner.S:460: Error: unknown mnemonic `vpshufb' -- `vpshufb %ymm0,%ymm2,%ymm2'
inner.S: inner.S:465: Error: unknown mnemonic `vmovapd' -- `vmovapd %ymm1,0(%rsp)'
inner.S: inner.S:470: Error: unknown mnemonic `movq' -- `movq %r11,320(%rsp)'
inner.S: inner.S:475: Error: unknown mnemonic `vmovapd' -- `vmovapd %ymm2,32(%rsp)'
inner.S: inner.S:480: Error: unknown mnemonic `movq' -- `movq %r13,328(%rsp)'
inner.S: inner.S:485: Error: unknown mnemonic `movq' -- `movq %r12,336(%rsp)'
inner.S: inner.S:490: Error: unknown mnemonic `movq' -- `movq %r14,344(%rsp)'
inner.S: inner.S:495: Error: unknown mnemonic `movq' -- `movq %rbx,352(%rsp)'
inner.S: inner.S:500: Error: unknown mnemonic `movq' -- `movq %r15,360(%rsp)'
inner.S: inner.S:505: Error: unknown mnemonic `movq' -- `movq %rbp,368(%rsp)'
inner.S: inner.S:510: Error: unknown mnemonic `movq' -- `movq %rdx,376(%rsp)'
inner.S: inner.S:515: Error: operand 1 must be an integer register -- `mov %rsi,%rdi'
inner.S: inner.S:520: Error: unknown mnemonic `movq' -- `movq %rcx,384(%rsp)'
inner.S: inner.S:525: Error: unknown mnemonic `movq' -- `movq 0(%rsp),%rsi'
inner.S: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	avx2