Implementation notes: aarch64, gcc185, crypto_aead/aeadaes256ocbtaglen128v1

Computer: gcc185
Microarchitecture: aarch64; Skylark (503f0002)
Architecture: aarch64
CPU ID: 503f0002
SUPERCOP version: 20240107
Operation: crypto_aead
Primitive: aeadaes256ocbtaglen128v1

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
36150	7904 0 0	28054 840 880	`T:dolbeau/armv8crypto`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
37125	5192 0 0	24141 832 864	`T:dolbeau/armv8crypto`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
37350	5128 0 0	24093 832 864	`T:dolbeau/armv8crypto`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
38850	9460 21 0	29774 877 880	`T:opt`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
39750	4524 0 0	22613 816 856	`T:dolbeau/armv8crypto`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
42750	6488 21 0	25710 877 864	`T:opt`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
46575	6304 21 0	25494 877 864	`T:opt`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
48075	5516 21 0	23846 861 856	`T:opt`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
49350	5334 18 0	26783 898 856	`T:opt`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
49425	4894 18 0	24505 890 856	`T:opt`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
49425	5622 18 0	27695 898 872	`T:opt`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
49500	5622 18 0	27695 898 872	`T:opt`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
49500	5178 18 0	24879 898 856	`T:opt`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
213150	3440 0 0	24368 832 856	`T:ref`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
213450	3484 0 0	25024 832 872	`T:ref`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
213975	3484 0 0	25024 832 872	`T:ref`	`clang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
249000	6064 0 0	26572 872 880	`T:ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
251400	1792 0 0	20874 824 856	`T:ref`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
295275	2440 0 0	21616 832 856	`T:ref`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20231214	20231212
299550	2092 0 0	21411 864 864	`T:ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
301350	2872 0 0	22187 864 864	`T:ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212
309075	1796 0 0	20235 848 856	`T:ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20231214	20231212

Compiler output

Implementation: T:dolbeau/armv8crypto
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

encrypt.c: encrypt.c:81:16: error: always_inline function 'vaesimcq_u8' requires target feature 'aes', but would be inlined into function 'aes256ni_setkey_e2d' that is compiled without support for 'aes'
encrypt.c: 81 | dkeys[i] = vaesimcq_u8(ekeys[14-i]);
encrypt.c: | ^
encrypt.c: encrypt.c:118:10: error: always_inline function 'vaeseq_u8' requires target feature 'aes', but would be inlined into function 'aes256armv8_encrypt1_neon' that is compiled without support for 'aes'
encrypt.c: 118 | temp = vaeseq_u8(temp, vld1q_u8((rkeys+208)));
encrypt.c: | ^
encrypt.c: encrypt.c:281:1: error: always_inline function 'vaeseq_u8' requires target feature 'aes', but would be inlined into function 'aes256armv8_encrypt4_xor_neon' that is compiled without support for 'aes'
encrypt.c: 281 | FUNC(4, MAKE4, MAKE4COMMA)
encrypt.c: | ^
encrypt.c: encrypt.c:260:11: note: expanded from macro 'FUNC'
encrypt.c: 260 | MAKEN(AESENCLASTx); \
encrypt.c: | ^
encrypt.c: encrypt.c:281:1: error: always_inline function 'vaeseq_u8' requires target feature 'aes', but would be inlined into function 'aes256armv8_encrypt4_xor_neon' that is compiled without support for 'aes'
encrypt.c: encrypt.c:260:11: note: expanded from macro 'FUNC'
encrypt.c: 260 | MAKEN(AESENCLASTx); \
encrypt.c: | ^
encrypt.c: encrypt.c:281:1: error: always_inline function 'vaeseq_u8' requires target feature 'aes', but would be inlined into function 'aes256armv8_encrypt4_xor_neon' that is compiled without support for 'aes'
encrypt.c: encrypt.c:260:11: note: expanded from macro 'FUNC'
encrypt.c: 260 | MAKEN(AESENCLASTx); \
encrypt.c: | ^
encrypt.c: encrypt.c:281:1: error: always_inline function 'vaeseq_u8' requires target feature 'aes', but would be inlined into function 'aes256armv8_encrypt4_xor_neon' that is compiled without support for 'aes'
encrypt.c: encrypt.c:260:11: note: expanded from macro 'FUNC'
encrypt.c: 260 | MAKEN(AESENCLASTx); \
encrypt.c: | ^
encrypt.c: encrypt.c:284:1: error: always_inline function 'vaesdq_u8' requires target feature 'aes', but would be inlined into function 'aes256armv8_decrypt8_xor_neon' that is compiled without support for 'aes'
encrypt.c: ...

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:dolbeau/armv8crypto
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:dolbeau/armv8crypto
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:dolbeau/armv8crypto
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:dolbeau/armv8crypto
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:dolbeau/armv8crypto

Compiler output

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

ocb_vaes.c: In file included from ocb_vaes.c:71:
ocb_vaes.c: /usr/bin/../lib/clang/17/include/immintrin.h:14:2: error: "This header is only meant to be used on x86 and x64 architecture"
ocb_vaes.c: 14 | #error "This header is only meant to be used on x86 and x64 architecture"
ocb_vaes.c: | ^
ocb_vaes.c: In file included from ocb_vaes.c:71:
ocb_vaes.c: In file included from /usr/bin/../lib/clang/17/include/immintrin.h:17:
ocb_vaes.c: In file included from /usr/bin/../lib/clang/17/include/x86gprintrin.h:15:
ocb_vaes.c: /usr/bin/../lib/clang/17/include/hresetintrin.h:42:27: error: invalid input constraint 'a' in asm
ocb_vaes.c: 42 | __asm__ ("hreset $0" :: "a"(__eax));
ocb_vaes.c: | ^
ocb_vaes.c: In file included from ocb_vaes.c:71:
ocb_vaes.c: In file included from /usr/bin/../lib/clang/17/include/immintrin.h:21:
ocb_vaes.c: /usr/bin/../lib/clang/17/include/mmintrin.h:14:2: error: "This header is only meant to be used on x86 and x64 architecture"
ocb_vaes.c: 14 | #error "This header is only meant to be used on x86 and x64 architecture"
ocb_vaes.c: | ^
ocb_vaes.c: /usr/bin/../lib/clang/17/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
ocb_vaes.c: 54 | return (__m64)__builtin_ia32_vec_init_v2si(__i, 0);
ocb_vaes.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ocb_vaes.c: /usr/bin/../lib/clang/17/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
ocb_vaes.c: 133 | return (__m64)__builtin_ia32_packsswb((__v4hi)__m1, (__v4hi)__m2);
ocb_vaes.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ocb_vaes.c: /usr/bin/../lib/clang/17/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
ocb_vaes.c: 163 | return (__m64)__builtin_ia32_packssdw((__v2si)__m1, (__v2si)__m2);
ocb_vaes.c: | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ocb_vaes.c: /usr/bin/../lib/clang/17/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
ocb_vaes.c: ...

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

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

Compiler output

Implementation: T:vaes
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

ocb_vaes.c: ocb_vaes.c:71:10: fatal error: immintrin.h: No such file or directory
ocb_vaes.c: #include <immintrin.h>
ocb_vaes.c: ^~~~~~~~~~~~~
ocb_vaes.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	T:vaes
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:vaes
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:vaes
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:vaes