Implementation notes: aarch64, hikey960, crypto_aead/pi32cipher256v1

Computer: hikey960
Architecture: aarch64
CPU ID: 410fd034
SUPERCOP version: 20190816
Operation: crypto_aead
Primitive: pi32cipher256v1

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
71064	105076 0 0	119441 960 1024	`optimized_nonSSE`	`gcc_-O3_-fomit-frame-pointer`	20191129	20190816
79947	105844 0 0	120161 960 1024	`optimized_nonSSE`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20191129	20190816
106596	105204 0 0	119561 960 1024	`optimized_nonSSE`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv`	20191129	20190816
115479	45972 0 0	61891 944 928	`ref`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20191129	20190816
133200	106304 0 0	122281 960 1024	`optimized_nonSSE`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20191129	20190816
133260	28008 0 0	39879 928 928	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20191129	20190816
141525	105492 0 0	121136 952 992	`optimized_nonSSE`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20191129	20190816
159894	70128 0 0	82800 952 992	`optimized_nonSSE`	`gcc_-O2_-fomit-frame-pointer`	20191129	20190816
159894	56500 0 0	72067 944 928	`ref`	`gcc_-funroll-loops_-O2_-fomit-frame-pointer`	20191129	20190816
187088	104932 0 0	120608 952 992	`optimized_nonSSE`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20191129	20190816
191475	71064 0 0	83672 952 992	`optimized_nonSSE`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20191129	20190816
216450	107268 0 0	123209 960 1024	`optimized_nonSSE`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20191129	20190816
216450	102300 0 0	118296 952 992	`optimized_nonSSE`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20191129	20190816
216476	73304 0 0	85228 936 992	`optimized_nonSSE`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv`	20191129	20190816
224775	102300 0 0	118296 952 992	`optimized_nonSSE`	`gcc_-funroll-loops_-O_-fomit-frame-pointer`	20191129	20190816
224775	46712 0 0	60988 952 960	`ref`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv`	20191129	20190816
233100	73304 0 0	85252 936 992	`optimized_nonSSE`	`gcc_-funroll-loops_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20191129	20190816
248724	30916 0 0	43451 944 928	`ref`	`gcc_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20191129	20190816
258075	73304 0 0	85252 936 992	`optimized_nonSSE`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20191129	20190816
283050	58076 0 0	73675 944 928	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-O2_-fomit-frame-pointer`	20191129	20190816
308025	32496 0 0	45091 944 928	`ref`	`gcc_-O2_-fomit-frame-pointer`	20191129	20190816
308062	31012 0 0	43739 944 928	`ref`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20191129	20190816
324675	70904 0 0	83576 952 992	`optimized_nonSSE`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv`	20191129	20190816
327404	86492 0 0	99296 952 992	`optimized_nonSSE`	`gcc_-fno-schedule-insns_-O_-fomit-frame-pointer`	20191129	20190816
327404	49332 0 0	65196 952 960	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20191129	20190816
333000	46688 0 0	60972 952 960	`ref`	`gcc_-O3_-fomit-frame-pointer`	20191129	20190816
339097	86492 0 0	99296 952 992	`optimized_nonSSE`	`gcc_-O_-fomit-frame-pointer`	20191129	20190816
339203	28008 0 0	39855 928 928	`ref`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20191129	20190816
349650	47592 0 0	61828 952 960	`ref`	`gcc_-fno-schedule-insns_-O3_-fomit-frame-pointer`	20191129	20190816
373870	48356 0 0	64260 952 960	`ref`	`gcc_-funroll-loops_-O3_-fomit-frame-pointer`	20191129	20190816
381969	73304 0 0	85228 936 992	`optimized_nonSSE`	`gcc_-fno-schedule-insns_-Os_-fomit-frame-pointer`	20191129	20190816
432641	32492 0 0	45091 944 928	`ref`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv`	20191129	20190816
441225	45972 0 0	61891 944 928	`ref`	`gcc_-funroll-loops_-fno-schedule-insns_-O_-fomit-frame-pointer`	20191129	20190816
449550	28008 0 0	39855 928 928	`ref`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv`	20191129	20190816
466200	31012 0 0	43739 944 928	`ref`	`gcc_-O_-fomit-frame-pointer`	20191129	20190816
473073	28008 0 0	39855 928 928	`ref`	`gcc_-Os_-fomit-frame-pointer`	20191129	20190816
682732	31028 0 0	43755 944 928	`ref`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv`	20191129	20190816
736851	73304 0 0	85228 936 992	`optimized_nonSSE`	`gcc_-Os_-fomit-frame-pointer`	20191129	20190816
890775	28008 0 0	39879 928 928	`ref`	`gcc_-funroll-loops_-Os_-fomit-frame-pointer`	20191129	20190816
924186	86468 0 0	99272 952 992	`optimized_nonSSE`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv`	20191129	20190816
1548450	115948 0 0	141215 928 928	`ref`	`gcc_-funroll-loops`	20191129	20190816
1831500	302196 0 0	323444 936 992	`optimized_nonSSE`	`cc`	20191129	20190816
2174898	115948 0 0	141215 928 928	`ref`	`gcc`	20191129	20190816
4151595	302196 0 0	323444 936 992	`optimized_nonSSE`	`gcc`	20191129	20190816
5094900	115948 0 0	141215 928 928	`ref`	`cc`	20191129	20190816
5869886	302196 0 0	323444 936 992	`optimized_nonSSE`	`gcc_-funroll-loops`	20191129	20190816

Test failure

Implementation: optimized_nonSSE
Security model: unknown
Compiler: clang -O3 -fomit-frame-pointer -Qunused-arguments

error 111
crypto_aead_decrypt returns nonzero

Number of similar (compiler,implementation) pairs: 8, namely:

Compiler	Implementations
clang -O3 -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments	ref

Compiler output

Implementation: optimized_nonSSE
Security model: unknown
Compiler: clang -O3 -fomit-frame-pointer -Qunused-arguments

encrypt.c: encrypt.c:354:42: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: InternalState8[i1] = InternalState8[i1++] ^ ad[b+i];
encrypt.c: ~~ ^
encrypt.c: encrypt.c:468:68: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: c[CRYPTO_NSECBYTES+b+i] = InternalState8[i1] = InternalState8[i1++] ^ m[b+i];
encrypt.c: ~~ ^
encrypt.c: encrypt.c:622:42: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: InternalState8[i1] = InternalState8[i1++] ^ ad[b+i];
encrypt.c: ~~ ^
encrypt.c: 3 warnings generated.

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

Compiler	Implementations
clang -O3 -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE
clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments	optimized_nonSSE

Compiler output

Implementation: ref
Security model: unknown
Compiler: clang -O3 -fomit-frame-pointer -Qunused-arguments

encrypt.c: encrypt.c:240:42: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: InternalState8[i1] = InternalState8[i1++] ^ ad[b+i];
encrypt.c: ~~ ^
encrypt.c: encrypt.c:354:68: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: c[CRYPTO_NSECBYTES+b+i] = InternalState8[i1] = InternalState8[i1++] ^ m[b+i];
encrypt.c: ~~ ^
encrypt.c: encrypt.c:508:42: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: InternalState8[i1] = InternalState8[i1++] ^ ad[b+i];
encrypt.c: ~~ ^
encrypt.c: 3 warnings generated.

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

Compiler	Implementations
clang -O3 -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -mavx2 -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -mavx -fomit-frame-pointer -Qunused-arguments	ref
clang -O3 -fwrapv -mavx -maes -mpclmul -fomit-frame-pointer -Qunused-arguments	ref