Implementation notes: aarch64, a53, crypto_aead/aezv4

Computer: a53
Architecture: aarch64
CPU ID: unknown CPU ID
SUPERCOP version: 20160731
Operation: crypto_aead
Primitive: aezv4

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
1211355	`ref`	`gcc -mcpu=cortex-a53 -O3 -fomit-frame-pointer -fwrapv`	20160801	20160731
1780000	`ref`	`clang -mcpu=cortex-a53 -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160801	20160731
2404000	`ref`	`gcc -mcpu=cortex-a53 -O2 -fomit-frame-pointer -fwrapv`	20160801	20160731
2446400	`ref`	`gcc -mcpu=cortex-a53 -Os -fomit-frame-pointer -fwrapv`	20160801	20160731
2545600	`ref`	`gcc -mcpu=cortex-a53 -O -fomit-frame-pointer -fwrapv`	20160801	20160731

Compiler output

Implementation: crypto_aead/aezv4/aesni
Compiler: clang -mcpu=cortex-a53 -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

encrypt.c: encrypt.c:234:31: error: unknown type name 'size_t'
encrypt.c: static int blake2b(void *out, size_t outlen,
encrypt.c: ^
encrypt.c: encrypt.c:235:37: error: unknown type name 'size_t'
encrypt.c: const void *key, size_t keylen,
encrypt.c: ^
encrypt.c: encrypt.c:236:36: error: unknown type name 'size_t'
encrypt.c: const void *in, size_t inlen);
encrypt.c: ^
encrypt.c: encrypt.c:712:5: error: unknown type name 'size_t'
encrypt.c: size_t c;
encrypt.c: ^
encrypt.c: encrypt.c:713:5: error: unknown type name 'size_t'
encrypt.c: size_t outlen;
encrypt.c: ^
encrypt.c: encrypt.c:795:44: error: unknown type name 'size_t'
encrypt.c: const void *in, size_t inlen)
encrypt.c: ^
encrypt.c: encrypt.c:797:5: error: use of undeclared identifier 'size_t'; did you mean 'sizeof'?
encrypt.c: size_t i;
encrypt.c: ^~~~~~
encrypt.c: sizeof
encrypt.c: encrypt.c:797:12: error: use of undeclared identifier 'i'
encrypt.c: size_t i;
encrypt.c: ^
encrypt.c: ...

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

Compiler	Implementations
clang -mcpu=cortex-a53 -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	aesni

Compiler output

Implementation: crypto_aead/aezv4/aesni
Compiler: gcc -mcpu=cortex-a53 -O2 -fomit-frame-pointer -fwrapv

encrypt.c: encrypt.c: In function 'srl4':
encrypt.c: encrypt.c:130:5: note: use -flax-vector-conversions to permit conversions between vectors with differing element types or numbers of subparts
encrypt.c: uint8x16_t tmp = vandq_u8(vshrq_n_s8(vextq_u8(x, x, 1),4),mask);
encrypt.c: ^
encrypt.c: encrypt.c:130:31: error: incompatible type for argument 1 of 'vshrq_n_s8'
encrypt.c: uint8x16_t tmp = vandq_u8(vshrq_n_s8(vextq_u8(x, x, 1),4),mask);
encrypt.c: ^
encrypt.c: In file included from encrypt.c:113:0:
encrypt.c: /usr/lib/gcc/aarch64-linux-gnu/4.9/include/arm_neon.h:21834:1: note: expected 'int8x16_t' but argument is of type 'uint8x16_t'
encrypt.c: vshrq_n_s8 (int8x16_t __a, const int __b)
encrypt.c: ^
encrypt.c: encrypt.c: In function 'sll4':
encrypt.c: encrypt.c:136:22: error: incompatible type for argument 1 of 'vshlq_n_s8'
encrypt.c: uint8x16_t tmp = vshlq_n_s8(vandq_u8(vextq_u8(x, x, 15),mask),4);
encrypt.c: ^
encrypt.c: In file included from encrypt.c:113:0:
encrypt.c: /usr/lib/gcc/aarch64-linux-gnu/4.9/include/arm_neon.h:21544:1: note: expected 'int8x16_t' but argument is of type 'uint8x16_t'
encrypt.c: vshlq_n_s8 (int8x16_t __a, const int __b)
encrypt.c: ^
encrypt.c: encrypt.c: In function 'double_block':
encrypt.c: encrypt.c:144:17: error: incompatible types when initializing type 'uint8x16_t' using type 'int8x16_t'
encrypt.c: block tmp = vshrq_n_s8((int8x16_t)b,7);
encrypt.c: ^
encrypt.c: encrypt.c: At top level:
encrypt.c: encrypt.c:234:31: error: unknown type name 'size_t'
encrypt.c: ...

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

Compiler	Implementations
gcc -mcpu=cortex-a53 -O2 -fomit-frame-pointer -fwrapv	aesni
gcc -mcpu=cortex-a53 -O3 -fomit-frame-pointer -fwrapv	aesni
gcc -mcpu=cortex-a53 -O -fomit-frame-pointer -fwrapv	aesni
gcc -mcpu=cortex-a53 -Os -fomit-frame-pointer -fwrapv	aesni