Implementation notes: armeabi, bblack, crypto_core/salsa208

Computer: bblack
Microarchitecture: armeabi; Cortex-A8 (413fc082)
Architecture: armeabi
CPU ID: unknown CPU ID
SUPERCOP version: 20240107
Operation: crypto_core
Primitive: salsa208

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
1792	380 0 0	12431 408 712	`armneon2`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1792	380 0 0	21143 408 728	`armneon2`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1792	380 0 0	8421 436 712	`armneon2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
1792	380 0 0	7897 428 712	`armneon2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
1832	380 0 0	12431 408 712	`armneon2`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1832	380 0 0	9502 440 712	`armneon2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
1832	380 0 0	8249 436 712	`armneon2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
1836	876 0 0	21631 408 728	`armneon`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1872	380 0 0	12391 408 720	`armneon2`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1872	380 0 0	10765 404 712	`armneon2`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1876	1052 0 0	9069 436 712	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
1916	1052 0 0	10150 440 712	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
1960	1228 0 0	21979 408 728	`ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1996	1228 0 0	13227 408 720	`ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
1996	1228 0 0	11601 404 712	`ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
2000	1228 0 0	13267 408 712	`ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
2000	1228 0 0	13267 408 712	`ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231218	20231212
2204	802 0 0	8297 428 712	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212
2332	820 0 0	8685 436 712	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231218	20231212

Compiler output

Implementation: armneon
Security model: constbranchindex
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

core.c: In file included from core.c:9:
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:28:2: error: "NEON support not enabled"
core.c: #error "NEON support not enabled"
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:48:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(8))) int8_t int8x8_t;
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:49:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(16))) int8_t int8x16_t;
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:50:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(4))) int16_t int16x4_t;
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:51:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(8))) int16_t int16x8_t;
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:52:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(2))) int32_t int32x2_t;
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:53:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(4))) int32_t int32x4_t;
core.c: ^
core.c: /usr/lib/llvm-7/lib/clang/7.0.1/include/arm_neon.h:54:24: error: 'neon_vector_type' attribute is not supported for this target
core.c: typedef __attribute__((neon_vector_type(1))) int64_t int64x1_t;
core.c: ^
core.c: ...

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

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

Compiler output

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

core.c: In file included from core.c:9:
core.c: core.c: In function 'crypto_core_salsa208_armneon_constbranchindex':
core.c: /usr/lib/gcc/arm-linux-gnueabihf/8/include/arm_neon.h:8866:1: error: inlining failed in call to always_inline 'vextq_u32': target specific option mismatch
core.c: vextq_u32 (uint32x4_t __a, uint32x4_t __b, const int __c)
core.c: ^~~~~~~~~
core.c: core.c:51:23: note: called from here
core.c: uint32x4_t start3 = vextq_u32(k2k3n3k7,k2k3n3k7,1);
core.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
core.c: In file included from core.c:9:
core.c: /usr/lib/gcc/arm-linux-gnueabihf/8/include/arm_neon.h:8866:1: error: inlining failed in call to always_inline 'vextq_u32': target specific option mismatch
core.c: vextq_u32 (uint32x4_t __a, uint32x4_t __b, const int __c)
core.c: ^~~~~~~~~
core.c: core.c:50:23: note: called from here
core.c: uint32x4_t start2 = vextq_u32(n1n2k6k1,n1n2k6k1,1);
core.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
core.c: In file included from core.c:9:
core.c: /usr/lib/gcc/arm-linux-gnueabihf/8/include/arm_neon.h:8866:1: error: inlining failed in call to always_inline 'vextq_u32': target specific option mismatch
core.c: vextq_u32 (uint32x4_t __a, uint32x4_t __b, const int __c)
core.c: ^~~~~~~~~
core.c: core.c:49:23: note: called from here
core.c: uint32x4_t start1 = vextq_u32(k4k5k0n0,k4k5k0n0,1);
core.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
core.c: In file included from core.c:9:
core.c: /usr/lib/gcc/arm-linux-gnueabihf/8/include/arm_neon.h:7192:1: error: inlining failed in call to always_inline 'vcombine_u32': target specific option mismatch
core.c: vcombine_u32 (uint32x2_t __a, uint32x2_t __b)
core.c: ...

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

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