Implementation notes: amd64, hydra1, crypto_aead/morus640128v1

Computer: hydra1
Architecture: amd64
CPU ID: AuthenticAMD-00100fa0-178bfbff
SUPERCOP version: 20161220
Operation: crypto_aead
Primitive: morus640128v1

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
10696	`sse2`	`clang -mcpu=cortex-a9 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161020	20161009
10699	`sse2`	`clang -O3 -fomit-frame-pointer -Qunused-arguments`	20161020	20161009
10702	`sse2`	`clang -mcpu=native -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161020	20161009
10704	`sse2`	`clang -mcpu=cortex-a8 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161020	20161009
11672	`sse2`	`clang -O3 -fwrapv -march=native -fomit-frame-pointer -Qunused-arguments`	20161020	20161009
11685	`sse2`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20161020	20161009
13765	`sse2`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
13797	`sse2`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
13810	`sse2`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
13875	`sse2`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
14951	`sse2`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
14992	`sse2`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
15061	`sse2`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20161020	20161009
15064	`sse2`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20161020	20161009
15133	`sse2`	`gcc -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
15136	`sse2`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
15395	`sse2`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20161020	20161009
15507	`sse2`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20161020	20161009
15529	`sse2`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
15585	`sse2`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
15589	`sse2`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
15605	`sse2`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
15610	`sse2`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20161020	20161009
15645	`sse2`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20161020	20161009
15646	`sse2`	`gcc -m64 -O -fomit-frame-pointer`	20161020	20161009
15654	`sse2`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20161020	20161009
15662	`sse2`	`gcc -O -fomit-frame-pointer`	20161020	20161009
15682	`sse2`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20161020	20161009
15687	`sse2`	`gcc -funroll-loops -O -fomit-frame-pointer`	20161020	20161009
15736	`sse2`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20161020	20161009
15736	`sse2`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20161020	20161009
15755	`sse2`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20161020	20161009
15770	`sse2`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20161020	20161009
15784	`sse2`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20161020	20161009
15795	`sse2`	`gcc -march=nocona -O -fomit-frame-pointer`	20161020	20161009
15803	`sse2`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20161020	20161009
15850	`sse2`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20161020	20161009
15856	`sse2`	`gcc -march=k8 -O -fomit-frame-pointer`	20161020	20161009
16296	`sse2`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20161020	20161009
16342	`sse2`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20161020	20161009
16343	`sse2`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20161020	20161009
16343	`sse2`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20161020	20161009
16363	`sse2`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20161020	20161009
16368	`sse2`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
16370	`sse2`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
16476	`sse2`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
16499	`sse2`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
16618	`sse2`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
16624	`sse2`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
16795	`sse2`	`gcc -O3 -fomit-frame-pointer`	20161020	20161009
16795	`sse2`	`gcc -m64 -O3 -fomit-frame-pointer`	20161020	20161009
16797	`sse2`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20161020	20161009
16797	`sse2`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
16798	`sse2`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
17026	`sse2`	`gcc -m64 -march=corei7 -O2 -fomit-frame-pointer`	20161020	20161009
17341	`sse2`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20161020	20161009
17352	`sse2`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20161020	20161009
17365	`sse2`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20161020	20161009
17621	`sse2`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20161020	20161009
17646	`sse2`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20161020	20161009
17659	`sse2`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20161020	20161009
17661	`sse2`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
17680	`sse2`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
17715	`sse2`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
17738	`sse2`	`gcc -O2 -fomit-frame-pointer`	20161020	20161009
17738	`sse2`	`gcc -m64 -O2 -fomit-frame-pointer`	20161020	20161009
17850	`sse2`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20161020	20161009
17854	`sse2`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
17997	`sse2`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
17998	`sse2`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
18008	`sse2`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20161020	20161009
18011	`sse2`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20161020	20161009
18313	`sse2`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
18376	`sse2`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
26765	`ref`	`gcc -funroll-loops -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
26795	`ref`	`gcc -funroll-loops -m64 -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
26808	`ref`	`gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer`	20161020	20161009
26813	`ref`	`gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer`	20161020	20161009
26824	`ref`	`gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer`	20161020	20161009
26830	`ref`	`gcc -funroll-loops -m64 -Os -fomit-frame-pointer`	20161020	20161009
26837	`ref`	`gcc -funroll-loops -Os -fomit-frame-pointer`	20161020	20161009
26842	`ref`	`gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer`	20161020	20161009
26845	`ref`	`gcc -funroll-loops -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
26851	`ref`	`gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer`	20161020	20161009
26869	`ref`	`gcc -funroll-loops -m64 -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
26949	`ref`	`gcc -m64 -march=k8 -Os -fomit-frame-pointer`	20161020	20161009
26952	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20161020	20161009
26957	`ref`	`gcc -Os -fomit-frame-pointer`	20161020	20161009
26958	`ref`	`gcc -m64 -march=barcelona -Os -fomit-frame-pointer`	20161020	20161009
26963	`ref`	`gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer`	20161020	20161009
26974	`ref`	`gcc -m64 -Os -fomit-frame-pointer`	20161020	20161009
26976	`ref`	`gcc -fno-schedule-insns -Os -fomit-frame-pointer`	20161020	20161009
26976	`ref`	`gcc -march=barcelona -Os -fomit-frame-pointer`	20161020	20161009
26992	`ref`	`gcc -march=k8 -Os -fomit-frame-pointer`	20161020	20161009
27027	`ref`	`gcc -m64 -march=corei7 -Os -fomit-frame-pointer`	20161020	20161009
27066	`ref`	`gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer`	20161020	20161009
27077	`ref`	`gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer`	20161020	20161009
27099	`ref`	`gcc -m64 -march=core2 -Os -fomit-frame-pointer`	20161020	20161009
27112	`ref`	`gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer`	20161020	20161009
27141	`ref`	`gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer`	20161020	20161009
27192	`ref`	`gcc -march=nocona -Os -fomit-frame-pointer`	20161020	20161009
27210	`ref`	`gcc -m64 -march=nocona -Os -fomit-frame-pointer`	20161020	20161009
27211	`ref`	`gcc -m64 -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
27232	`ref`	`gcc -march=k8 -O2 -fomit-frame-pointer`	20161020	20161009
27251	`ref`	`gcc -m64 -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
27251	`ref`	`gcc -march=barcelona -O2 -fomit-frame-pointer`	20161020	20161009
27259	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20161020	20161009
27277	`ref`	`gcc -m64 -march=native -mtune=native -O2 -fomit-frame-pointer`	20161020	20161009
27981	`ref`	`gcc -funroll-loops -fno-schedule-insns -O2 -fomit-frame-pointer`	20161020	20161009
27990	`ref`	`gcc -funroll-loops -m64 -O2 -fomit-frame-pointer`	20161020	20161009
28008	`ref`	`gcc -funroll-loops -O2 -fomit-frame-pointer`	20161020	20161009
28325	`ref`	`gcc -m64 -O2 -fomit-frame-pointer`	20161020	20161009
28368	`ref`	`gcc -fno-schedule-insns -O2 -fomit-frame-pointer`	20161020	20161009
28400	`ref`	`gcc -O2 -fomit-frame-pointer`	20161020	20161009
28512	`ref`	`gcc -m64 -march=core2 -O2 -fomit-frame-pointer`	20161020	20161009
28528	`ref`	`gcc -m64 -march=core2 -msse4 -O2 -fomit-frame-pointer`	20161020	20161009
28534	`ref`	`gcc -m64 -march=core2 -msse4.1 -O2 -fomit-frame-pointer`	20161020	20161009
28667	`ref`	`gcc -m64 -march=corei7 -O2 -fomit-frame-pointer`	20161020	20161009
28862	`ref`	`gcc -funroll-loops -m64 -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
28905	`ref`	`gcc -funroll-loops -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
28946	`ref`	`gcc -funroll-loops -march=k8 -O -fomit-frame-pointer`	20161020	20161009
28959	`ref`	`gcc -funroll-loops -m64 -march=k8 -O -fomit-frame-pointer`	20161020	20161009
28981	`ref`	`gcc -funroll-loops -fno-schedule-insns -O -fomit-frame-pointer`	20161020	20161009
28987	`ref`	`gcc -funroll-loops -O -fomit-frame-pointer`	20161020	20161009
29021	`ref`	`gcc -funroll-loops -m64 -march=nocona -O -fomit-frame-pointer`	20161020	20161009
29035	`ref`	`gcc -funroll-loops -m64 -O -fomit-frame-pointer`	20161020	20161009
29082	`ref`	`gcc -funroll-loops -march=nocona -O -fomit-frame-pointer`	20161020	20161009
29203	`ref`	`gcc -funroll-loops -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
29205	`ref`	`gcc -funroll-loops -m64 -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
29212	`ref`	`gcc -march=k8 -O -fomit-frame-pointer`	20161020	20161009
29272	`ref`	`gcc -m64 -march=k8 -O -fomit-frame-pointer`	20161020	20161009
29285	`ref`	`gcc -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
29291	`ref`	`gcc -m64 -march=native -mtune=native -O -fomit-frame-pointer`	20161020	20161009
29323	`ref`	`gcc -m64 -march=barcelona -O -fomit-frame-pointer`	20161020	20161009
29330	`ref`	`gcc -march=nocona -O -fomit-frame-pointer`	20161020	20161009
29334	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20161020	20161009
29346	`ref`	`gcc -m64 -march=nocona -O -fomit-frame-pointer`	20161020	20161009
29429	`ref`	`gcc -m64 -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
29461	`ref`	`gcc -march=nocona -O2 -fomit-frame-pointer`	20161020	20161009
29520	`ref`	`gcc -m64 -march=core2 -msse4.1 -O -fomit-frame-pointer`	20161020	20161009
29523	`ref`	`gcc -fno-schedule-insns -O -fomit-frame-pointer`	20161020	20161009
29536	`ref`	`gcc -O -fomit-frame-pointer`	20161020	20161009
29539	`ref`	`gcc -m64 -march=core2 -msse4 -O -fomit-frame-pointer`	20161020	20161009
29549	`ref`	`gcc -m64 -march=core2 -O -fomit-frame-pointer`	20161020	20161009
29557	`ref`	`gcc -m64 -O -fomit-frame-pointer`	20161020	20161009
29576	`ref`	`gcc -m64 -march=corei7 -O -fomit-frame-pointer`	20161020	20161009
31953	`ref`	`gcc -funroll-loops -O3 -fomit-frame-pointer`	20161020	20161009
31963	`ref`	`gcc -funroll-loops -m64 -O3 -fomit-frame-pointer`	20161020	20161009
32017	`ref`	`gcc -funroll-loops -fno-schedule-insns -O3 -fomit-frame-pointer`	20161020	20161009
32523	`ref`	`gcc -fno-schedule-insns -O3 -fomit-frame-pointer`	20161020	20161009
32531	`ref`	`gcc -O3 -fomit-frame-pointer`	20161020	20161009
32538	`ref`	`gcc -m64 -O3 -fomit-frame-pointer`	20161020	20161009
32674	`ref`	`gcc -funroll-loops -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
32689	`ref`	`gcc -funroll-loops -m64 -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
32704	`ref`	`gcc -funroll-loops -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
32755	`ref`	`gcc -funroll-loops -m64 -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
32774	`ref`	`gcc -funroll-loops -m64 -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
32779	`ref`	`gcc -funroll-loops -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
33282	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20161020	20161009
33288	`ref`	`gcc -m64 -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
33288	`ref`	`gcc -march=barcelona -O3 -fomit-frame-pointer`	20161020	20161009
33298	`ref`	`gcc -m64 -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
33325	`ref`	`gcc -m64 -march=native -mtune=native -O3 -fomit-frame-pointer`	20161020	20161009
33330	`ref`	`gcc -march=k8 -O3 -fomit-frame-pointer`	20161020	20161009
33616	`ref`	`gcc -m64 -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
33617	`ref`	`gcc -march=nocona -O3 -fomit-frame-pointer`	20161020	20161009
198847	`sse2`	`gcc -funroll-loops`	20161020	20161009
198888	`sse2`	`cc`	20161020	20161009
198915	`sse2`	`gcc`	20161020	20161009

Test failure

Implementation: crypto_aead/morus640128v1/sse2
Compiler: gcc -Os -fomit-frame-pointer

error 111

Number of similar (compiler,implementation) pairs: 25, namely:

Compiler	Implementations
gcc -Os -fomit-frame-pointer	sse2
gcc -fno-schedule-insns -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -fno-schedule-insns -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -m64 -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -m64 -march=barcelona -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -m64 -march=k8 -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -m64 -march=nocona -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -march=barcelona -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -march=k8 -Os -fomit-frame-pointer	sse2
gcc -funroll-loops -march=nocona -Os -fomit-frame-pointer	sse2
gcc -m64 -Os -fomit-frame-pointer	sse2
gcc -m64 -march=core2 -O3 -fomit-frame-pointer	ref sse2
gcc -m64 -march=core2 -Os -fomit-frame-pointer	sse2
gcc -m64 -march=core2 -msse4.1 -Os -fomit-frame-pointer	sse2
gcc -m64 -march=core2 -msse4 -Os -fomit-frame-pointer	sse2
gcc -m64 -march=corei7 -Os -fomit-frame-pointer	sse2
gcc -m64 -march=k8 -Os -fomit-frame-pointer	sse2
gcc -m64 -march=native -mtune=native -Os -fomit-frame-pointer	sse2
gcc -m64 -march=nocona -Os -fomit-frame-pointer	sse2
gcc -march=barcelona -Os -fomit-frame-pointer	sse2
gcc -march=k8 -Os -fomit-frame-pointer	sse2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	sse2
gcc -march=nocona -Os -fomit-frame-pointer	sse2

Test failure

Implementation: crypto_aead/morus640128v1/sse2
Compiler: gcc -m64 -march=barcelona -Os -fomit-frame-pointer

error 111
error 111

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

Compiler	Implementations
gcc -m64 -march=barcelona -Os -fomit-frame-pointer	sse2

Compiler output

Implementation: crypto_aead/morus640128v1/ref
Compiler: cc

try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `morus_initialization':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_stateupdate'
try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `morus_tag_generation':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_stateupdate'
try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `morus_tag_verification':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_stateupdate'
try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `morus_enc_aut_partialblock':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_stateupdate'
try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `morus_dec_aut_partialblock':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_stateupdate'
try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `crypto_aead_morus640128v1_ref_encrypt':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_enc_aut_step'
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_enc_aut_step'
try.c: crypto_aead_morus640128v1.a(encrypt.o): In function `crypto_aead_morus640128v1_ref_decrypt':
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_enc_aut_step'
try.c: encrypt.c:(.text+0x...): undefined reference to `morus_dec_aut_step'
try.c: collect2: error: ld returned 1 exit status

Number of similar (compiler,implementation) pairs: 3, namely:

Compiler	Implementations
cc	ref
gcc	ref
gcc -funroll-loops	ref

Compiler output

Implementation: crypto_aead/morus640128v1/ref
Compiler: clang -O3 -fomit-frame-pointer -Qunused-arguments

Number of similar (compiler,implementation) pairs: 6, namely:

Compiler	Implementations
clang -O3 -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -march=native -fomit-frame-pointer -Qunused-arguments	ref
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ref
clang -mcpu=cortex-a8 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ref
clang -mcpu=cortex-a9 -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ref
clang -mcpu=native -mfpu=neon -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ref