Implementation notes: aarch64, supercoplxc, crypto_core/aes128decrypt

Computer: supercoplxc
Architecture: aarch64
CPU ID: 410fd034
SUPERCOP version: 20190816
Operation: crypto_core
Primitive: aes128decrypt

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
320	1444 0 0	14782 800 792	`dolbeau/armv8crypto`	`clang_-mcpu=native_-mfpu=neon_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments`	20190828	20190816
880	5768 0 0	17841 888 784	`dolbeau/std-1rt-nodk`	`gcc_-O3_-fomit-frame-pointer`	20190828	20190816
880	5768 0 0	19225 888 784	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20190828	20190816
880	7844 0 0	21305 888 784	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20190828	20190816
880	11964 0 0	24041 888 784	`dolbeau/std-4rt-nodk`	`gcc_-O3_-fomit-frame-pointer`	20190828	20190816
880	11964 0 0	25417 888 784	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20190828	20190816
960	5200 0 0	15992 880 768	`dolbeau/std-1rt-nodk`	`gcc_-O2_-fomit-frame-pointer`	20190828	20190816
960	5780 0 0	18656 880 768	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20190828	20190816
960	7292 0 0	18096 880 768	`dolbeau/std-2rt-nodk`	`gcc_-O2_-fomit-frame-pointer`	20190828	20190816
960	7844 0 0	19921 888 784	`dolbeau/std-2rt-nodk`	`gcc_-O3_-fomit-frame-pointer`	20190828	20190816
960	7888 0 0	20752 880 768	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20190828	20190816
960	11404 0 0	22200 880 768	`dolbeau/std-4rt-nodk`	`gcc_-O2_-fomit-frame-pointer`	20190828	20190816
960	11984 0 0	24848 880 768	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20190828	20190816
1040	5072 0 0	18406 800 792	`dolbeau/std-1rt-nodk`	`clang_-mcpu=native_-mfpu=neon_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments`	20190828	20190816
1040	7136 0 0	20470 800 792	`dolbeau/std-2rt-nodk`	`clang_-mcpu=native_-mfpu=neon_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments`	20190828	20190816
1040	11276 0 0	24614 800 792	`dolbeau/std-4rt-nodk`	`clang_-mcpu=native_-mfpu=neon_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments`	20190828	20190816
1120	5080 0 0	18366 800 792	`dolbeau/std-1rt-nodk`	`clang_-O3_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	5080 0 0	18414 800 792	`dolbeau/std-1rt-nodk`	`clang_-O3_-fwrapv_-mavx2_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	5080 0 0	18414 800 792	`dolbeau/std-1rt-nodk`	`clang_-O3_-fwrapv_-mavx_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	5080 0 0	18414 800 792	`dolbeau/std-1rt-nodk`	`clang_-O3_-fwrapv_-mavx_-maes_-mpclmul_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	7144 0 0	20430 800 792	`dolbeau/std-2rt-nodk`	`clang_-O3_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	7144 0 0	20478 800 792	`dolbeau/std-2rt-nodk`	`clang_-O3_-fwrapv_-mavx2_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	7144 0 0	20478 800 792	`dolbeau/std-2rt-nodk`	`clang_-O3_-fwrapv_-mavx_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	7144 0 0	20478 800 792	`dolbeau/std-2rt-nodk`	`clang_-O3_-fwrapv_-mavx_-maes_-mpclmul_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	11280 0 0	24566 800 792	`dolbeau/std-4rt-nodk`	`clang_-O3_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	11280 0 0	24614 800 792	`dolbeau/std-4rt-nodk`	`clang_-O3_-fwrapv_-mavx2_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	11280 0 0	24614 800 792	`dolbeau/std-4rt-nodk`	`clang_-O3_-fwrapv_-mavx_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1120	11280 0 0	24614 800 792	`dolbeau/std-4rt-nodk`	`clang_-O3_-fwrapv_-mavx_-maes_-mpclmul_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1280	11376 0 0	21512 864 760	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20190828	20190816
1280	11936 0 0	24904 880 768	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1280	11920 0 0	25361 888 784	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1280	11376 0 0	21512 864 760	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1360	5156 0 0	15288 864 760	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20190828	20190816
1360	5724 0 0	18696 880 768	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1360	5712 0 0	19153 888 784	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1360	5156 0 0	15288 864 760	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1360	7232 0 0	17368 864 760	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20190828	20190816
1360	7788 0 0	21225 888 784	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1360	7232 0 0	17368 864 760	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1440	7812 0 0	20784 880 768	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1440	11376 0 0	21448 864 760	`dolbeau/std-4rt-nodk`	`gcc_-Os_-fomit-frame-pointer`	20190828	20190816
1440	11380 0 0	22144 880 768	`dolbeau/std-4rt-nodk`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1440	11376 0 0	21448 864 760	`dolbeau/std-4rt-nodk`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1520	11920 0 0	24001 888 784	`dolbeau/std-4rt-nodk`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1600	5156 0 0	15224 864 760	`dolbeau/std-1rt-nodk`	`gcc_-Os_-fomit-frame-pointer`	20190828	20190816
1600	5160 0 0	15920 880 768	`dolbeau/std-1rt-nodk`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1600	5712 0 0	17793 888 784	`dolbeau/std-1rt-nodk`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1600	5156 0 0	15224 864 760	`dolbeau/std-1rt-nodk`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1600	7232 0 0	17304 864 760	`dolbeau/std-2rt-nodk`	`gcc_-Os_-fomit-frame-pointer`	20190828	20190816
1600	7236 0 0	17992 880 768	`dolbeau/std-2rt-nodk`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1600	7788 0 0	19865 888 784	`dolbeau/std-2rt-nodk`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1600	7232 0 0	17304 864 760	`dolbeau/std-2rt-nodk`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1600	76 0 0	13568 816 792	`openssl`	`clang_-O3_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1600	76 0 0	13624 816 792	`openssl`	`clang_-O3_-fwrapv_-mavx2_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1600	76 0 0	13624 816 792	`openssl`	`clang_-O3_-fwrapv_-mavx_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1600	76 0 0	13624 816 792	`openssl`	`clang_-O3_-fwrapv_-mavx_-maes_-mpclmul_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
1600	76 0 0	13624 816 792	`openssl`	`clang_-mcpu=native_-mfpu=neon_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments`	20190828	20190816
1600	132 0 0	11146 896 768	`openssl`	`gcc_-O2_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	12427 904 784	`openssl`	`gcc_-O3_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	11250 896 768	`openssl`	`gcc_-O_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	10378 880 760	`openssl`	`gcc_-Os_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	11098 896 768	`openssl`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	12419 904 784	`openssl`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	11250 896 768	`openssl`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	10378 880 760	`openssl`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	13218 896 768	`openssl`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	13803 904 784	`openssl`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	14474 896 768	`openssl`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	10442 880 760	`openssl`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	13314 896 768	`openssl`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	13779 904 784	`openssl`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	14474 896 768	`openssl`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
1600	132 0 0	10442 880 760	`openssl`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
1680	132 0 0	13826 880 768	`openssl`	`cc`	20190828	20190816
1680	132 0 0	13826 880 768	`openssl`	`gcc`	20190828	20190816
1680	132 0 0	13826 880 768	`openssl`	`gcc_-funroll-loops`	20190828	20190816
1920	5256 0 0	16208 880 768	`dolbeau/std-1rt-nodk`	`gcc_-O_-fomit-frame-pointer`	20190828	20190816
1920	5256 0 0	16208 880 768	`dolbeau/std-1rt-nodk`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
1920	5844 0 0	20008 880 768	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20190828	20190816
1920	5844 0 0	20008 880 768	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
1920	7320 0 0	18272 880 768	`dolbeau/std-2rt-nodk`	`gcc_-O_-fomit-frame-pointer`	20190828	20190816
1920	7320 0 0	18272 880 768	`dolbeau/std-2rt-nodk`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
1920	7920 0 0	22096 880 768	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20190828	20190816
1920	7920 0 0	22096 880 768	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
2000	11492 0 0	22448 880 768	`dolbeau/std-4rt-nodk`	`gcc_-O_-fomit-frame-pointer`	20190828	20190816
2000	11492 0 0	22448 880 768	`dolbeau/std-4rt-nodk`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
2000	12104 0 0	26272 880 768	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20190828	20190816
2000	12104 0 0	26272 880 768	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
6480	3828 14336 0	17360 15200 768	`dolbeau/std-4rt-nodk`	`cc`	20190828	20190816
6480	3828 14336 0	17360 15200 768	`dolbeau/std-4rt-nodk`	`gcc`	20190828	20190816
6480	3828 14336 0	17360 15200 768	`dolbeau/std-4rt-nodk`	`gcc_-funroll-loops`	20190828	20190816
7760	4004 14336 0	17536 15200 768	`dolbeau/std-2rt-nodk`	`cc`	20190828	20190816
7760	4004 14336 0	17536 15200 768	`dolbeau/std-2rt-nodk`	`gcc`	20190828	20190816
7760	4004 14336 0	17536 15200 768	`dolbeau/std-2rt-nodk`	`gcc_-funroll-loops`	20190828	20190816
8400	4052 14336 0	17584 15200 768	`dolbeau/std-1rt-nodk`	`cc`	20190828	20190816
8400	4052 14336 0	17584 15200 768	`dolbeau/std-1rt-nodk`	`gcc`	20190828	20190816
8400	4052 14336 0	17584 15200 768	`dolbeau/std-1rt-nodk`	`gcc_-funroll-loops`	20190828	20190816
399600	5656 0 0	17785 888 784	`ref`	`gcc_-O3_-fomit-frame-pointer`	20190828	20190816
400720	5668 0 0	19177 888 784	`ref`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20190828	20190816
425280	5748 0 0	19233 888 784	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
457360	5712 0 0	17833 888 784	`ref`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20190828	20190816
476720	5364 0 0	18686 800 792	`ref`	`clang_-mcpu=native_-mfpu=neon_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments`	20190828	20190816
682880	5604 0 0	18870 800 792	`ref`	`clang_-O3_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
683040	5624 0 0	18942 800 792	`ref`	`clang_-O3_-fwrapv_-mavx_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
683040	5624 0 0	18942 800 792	`ref`	`clang_-O3_-fwrapv_-mavx_-maes_-mpclmul_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
687200	5624 0 0	18942 800 792	`ref`	`clang_-O3_-fwrapv_-mavx2_-fomit-frame-pointer_-Qunused-arguments`	20190828	20190816
700560	2976 0 0	15856 880 768	`ref`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20190828	20190816
701840	2980 0 0	15960 880 768	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
945440	3172 0 0	17312 880 768	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
945520	3172 0 0	17312 880 768	`ref`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20190828	20190816
2588800	2196 0 0	12312 864 760	`ref`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20190828	20190816
2589520	2196 0 0	12312 864 760	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
2611840	2096 0 0	12160 864 760	`ref`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20190828	20190816
2611920	2096 0 0	12160 864 760	`ref`	`gcc_-Os_-fomit-frame-pointer`	20190828	20190816
2741600	2228 0 0	13048 880 768	`ref`	`gcc_-O2_-fomit-frame-pointer`	20190828	20190816
2819760	2220 0 0	12992 880 768	`ref`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20190828	20190816
3278480	2332 0 0	13248 880 768	`ref`	`gcc_-O_-fomit-frame-pointer`	20190828	20190816
3278480	2332 0 0	13248 880 768	`ref`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20190828	20190816
14128000	4564 0 0	22192 864 768	`ref`	`gcc`	20190828	20190816
14132240	4564 0 0	22192 864 768	`ref`	`cc`	20190828	20190816
14137520	4564 0 0	22192 864 768	`ref`	`gcc_-funroll-loops`	20190828	20190816

Compiler output

Implementation: dolbeau/armv8crypto
Security model: unknown
Compiler: cc

core.c: In file included from core.c:9:
core.c: core.c: In function 'aes128_setkey_armv8_decrypt':
core.c: /usr/lib/gcc/aarch64-linux-gnu/8/include/arm_neon.h:12447:1: error: inlining failed in call to always_inline 'vaesimcq_u8': target specific option mismatch
core.c: vaesimcq_u8 (uint8x16_t data)
core.c: ^~~~~~~~~~~
core.c: core.c:21:16: note: called from here
core.c: rkeys[i] = vaesimcq_u8(tkeys[10-i]);
core.c: ^~~~~~~~~~~~~~~~~~~~~~~~

Number of similar (compiler,implementation) pairs: 19, namely:

Compiler	Implementations
cc	dolbeau/armv8crypto
gcc	dolbeau/armv8crypto
gcc -O2 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -O3 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -O -fomit-frame-pointer	dolbeau/armv8crypto
gcc -Os -fomit-frame-pointer	dolbeau/armv8crypto
gcc -fno-schedule-insns -O2 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -fno-schedule-insns -O3 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -fno-schedule-insns -O -fomit-frame-pointer	dolbeau/armv8crypto
gcc -fno-schedule-insns -Os -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops	dolbeau/armv8crypto
gcc -funroll-loops -O2 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -O3 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -O -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -Os -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer	dolbeau/armv8crypto
gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer	dolbeau/armv8crypto

Compiler output

Implementation: dolbeau/armv8crypto
Security model: unknown
Compiler: clang -O3 -fomit-frame-pointer -Qunused-arguments

core.c: core.c:21:16: warning: implicit declaration of function 'vaesimcq_u8' is invalid in C99 [-Wimplicit-function-declaration]
core.c: rkeys[i] = vaesimcq_u8(tkeys[10-i]);
core.c: ^
core.c: core.c:21:14: error: assigning to 'uint8x16_t' (vector of 16 'uint8_t' values) from incompatible type 'int'
core.c: rkeys[i] = vaesimcq_u8(tkeys[10-i]);
core.c: ^ ~~~~~~~~~~~~~~~~~~~~~~~~
core.c: core.c:58:10: warning: implicit declaration of function 'vaesdq_u8' is invalid in C99 [-Wimplicit-function-declaration]
core.c: temp = vaesdq_u8(temp, vld1q_u8((rkeys+144)));
core.c: ^
core.c: core.c:58:8: error: assigning to 'uint8x16_t' (vector of 16 'uint8_t' values) from incompatible type 'int'
core.c: temp = vaesdq_u8(temp, vld1q_u8((rkeys+144)));
core.c: ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
core.c: 2 warnings and 2 errors generated.

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

Compiler	Implementations
clang -O3 -fomit-frame-pointer -Qunused-arguments	dolbeau/armv8crypto
clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments	dolbeau/armv8crypto
clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments	dolbeau/armv8crypto
clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments	dolbeau/armv8crypto