Implementation notes: amd64, waldorf, crypto_aead/pi64cipher256v1

Computer: waldorf
Architecture: amd64
CPU ID: GenuineIntel-000106e5-bfebfbff
SUPERCOP version: 20160715
Operation: crypto_aead
Primitive: pi64cipher256v1

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
156492	`optimized_SSE`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160718	20160715
157508	`optimized_SSE`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
161336	`optimized_SSE`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
161612	`optimized_SSE`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
161848	`optimized_SSE`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
194128	`optimized_nonSSE`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
236248	`optimized_nonSSE`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
238188	`optimized_nonSSE`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715
249996	`optimized_nonSSE`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
335900	`ref`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160718	20160715
353880	`ref`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160718	20160715
371088	`ref`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160718	20160715
372496	`ref`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160718	20160715

Test failure

Implementation: crypto_aead/pi64cipher256v1/optimized_AVX
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

error 111

Number of similar (compiler,implementation) pairs: 10, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	optimized_AVX optimized_AVX2
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	optimized_AVX optimized_AVX2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	optimized_AVX optimized_AVX2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	optimized_AVX optimized_AVX2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	optimized_AVX optimized_AVX2

Test failure

Implementation: crypto_aead/pi64cipher256v1/optimized_nonSSE
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

error 111
crypto_aead_decrypt returns nonzero

Number of similar (compiler,implementation) pairs: 2, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	optimized_nonSSE ref

Compiler output

Implementation: crypto_aead/pi64cipher256v1/ref
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

encrypt.c: encrypt.c:231:42: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: InternalState8[i1] = InternalState8[i1++] ^ ad[b+i];
encrypt.c: ~~ ^
encrypt.c: encrypt.c:333:68: 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:478:42: 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: 1, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	ref

Compiler output

Implementation: crypto_aead/pi64cipher256v1/optimized_nonSSE
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

encrypt.c: encrypt.c:346:42: warning: unsequenced modification and access to 'i1' [-Wunsequenced]
encrypt.c: InternalState8[i1] = InternalState8[i1++] ^ ad[b+i];
encrypt.c: ~~ ^
encrypt.c: encrypt.c:448:68: 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:593:42: 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: 1, namely:

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	optimized_nonSSE