Implementation notes: aarch64, pi4b, crypto_hashblocks/sha256

Computer: pi4b
Microarchitecture: aarch64; Cortex-A72 (410fd083)
Architecture: aarch64
CPU ID: 410fd083
SUPERCOP version: 20240107
Operation: crypto_hashblocks
Primitive: sha256

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
18258	8212 0 0	17636 816 728	`inplace`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
18400	8224 0 0	17644 816 728	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
18408	8456 0 0	18836 816 736	`inplace`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
18476	8472 0 0	18836 816 736	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
21688	8248 0 0	16700 800 720	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
21730	8392 0 0	17739 808 728	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
21853	8244 0 0	16708 800 720	`inplace`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
22205	8380 0 0	17731 808 728	`inplace`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231224	20231222
22375	9032 0 0	20490 840 728	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231224	20231222
22600	9044 0 0	20514 840 728	`inplace`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231224	20231222

Test failure

Implementation: dolbeau/armv8crypto
Security model: constbranchindex
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

error 111

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	dolbeau/armv8crypto

Compiler output

Implementation: dolbeau/amd64-sha
Security model: constbranchindex
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

blocks.c: In file included from blocks.c:37:
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/immintrin.h:14:2: error: "This header is only meant to be used on x86 and x64 architecture"
blocks.c: #error "This header is only meant to be used on x86 and x64 architecture"
blocks.c: ^
blocks.c: In file included from blocks.c:37:
blocks.c: In file included from /usr/lib/llvm-14/lib/clang/14.0.0/include/immintrin.h:17:
blocks.c: In file included from /usr/lib/llvm-14/lib/clang/14.0.0/include/x86gprintrin.h:15:
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/hresetintrin.h:42:27: error: invalid input constraint 'a' in asm
blocks.c: __asm__ ("hreset $0" :: "a"(__eax));
blocks.c: ^
blocks.c: In file included from blocks.c:37:
blocks.c: In file included from /usr/lib/llvm-14/lib/clang/14.0.0/include/immintrin.h:21:
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/mmintrin.h:14:2: error: "This header is only meant to be used on x86 and x64 architecture"
blocks.c: #error "This header is only meant to be used on x86 and x64 architecture"
blocks.c: ^
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/mmintrin.h:54:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
blocks.c: return (__m64)__builtin_ia32_vec_init_v2si(__i, 0);
blocks.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/mmintrin.h:133:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
blocks.c: return (__m64)__builtin_ia32_packsswb((__v4hi)__m1, (__v4hi)__m2);
blocks.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/mmintrin.h:163:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
blocks.c: return (__m64)__builtin_ia32_packssdw((__v2si)__m1, (__v2si)__m2);
blocks.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
blocks.c: /usr/lib/llvm-14/lib/clang/14.0.0/include/mmintrin.h:193:12: error: invalid conversion between vector type '__m64' (vector of 1 'long long' value) and integer type 'int' of different size
blocks.c: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	dolbeau/amd64-sha

Compiler output

Implementation: dolbeau/amd64-sha
Security model: constbranchindex
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

blocks.c: blocks.c:37:10: fatal error: immintrin.h: No such file or directory
blocks.c: 37 | #include <immintrin.h>
blocks.c: | ^~~~~~~~~~~~~
blocks.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	dolbeau/amd64-sha
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/amd64-sha
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/amd64-sha
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/amd64-sha

Compiler output

Implementation: dolbeau/armv8crypto
Security model: constbranchindex
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

blocks.c: In file included from blocks.c:37:
blocks.c: blocks.c: In function 'crypto_hashblocks_sha256_dolbeau_armv8crypto_constbranchindex':
blocks.c: /usr/lib/gcc/aarch64-linux-gnu/11/include/arm_neon.h:25281:1: error: inlining failed in call to 'always_inline' 'vsha256h2q_u32': target specific option mismatch
blocks.c: 25281 | vsha256h2q_u32 (uint32x4_t __hash_efgh, uint32x4_t __hash_abcd, uint32x4_t __wk)
blocks.c: | ^~~~~~~~~~~~~~
blocks.c: blocks.c:118:22: note: called from here
blocks.c: 118 | s1 = vsha256h2q_u32(x1, x0, h1)
blocks.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~
blocks.c: blocks.c:146:17: note: in expansion of macro 'DO16ROUNDS'
blocks.c: 146 | DO16ROUNDS(j0, j1, j2, j3, cc, cd, ce, cf);
blocks.c: | ^~~~~~~~~~
blocks.c: In file included from blocks.c:37:
blocks.c: /usr/lib/gcc/aarch64-linux-gnu/11/include/arm_neon.h:25273:1: error: inlining failed in call to 'always_inline' 'vsha256hq_u32': target specific option mismatch
blocks.c: 25273 | vsha256hq_u32 (uint32x4_t __hash_abcd, uint32x4_t __hash_efgh, uint32x4_t __wk)
blocks.c: | ^~~~~~~~~~~~~
blocks.c: blocks.c:117:22: note: called from here
blocks.c: 117 | s0 = vsha256hq_u32(x0, x1, h1); \
blocks.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~
blocks.c: blocks.c:146:17: note: in expansion of macro 'DO16ROUNDS'
blocks.c: 146 | DO16ROUNDS(j0, j1, j2, j3, cc, cd, ce, cf);
blocks.c: | ^~~~~~~~~~
blocks.c: In file included from blocks.c:37:
blocks.c: /usr/lib/gcc/aarch64-linux-gnu/11/include/arm_neon.h:25281:1: error: inlining failed in call to 'always_inline' 'vsha256h2q_u32': target specific option mismatch
blocks.c: 25281 | vsha256h2q_u32 (uint32x4_t __hash_efgh, uint32x4_t __hash_abcd, uint32x4_t __wk)
blocks.c: | ^~~~~~~~~~~~~~
blocks.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/armv8crypto
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/armv8crypto
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/armv8crypto
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	dolbeau/armv8crypto