Implementation notes: armeabi, a15, crypto_sign/sphincs256

Computer: a15
Architecture: armeabi
CPU ID: unknown CPU ID
SUPERCOP version: 20161026
Operation: crypto_sign
Primitive: sphincs256

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
353562173	`ref`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161217	20161026
361585986	`ref`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161217	20161026
522031683	`ref`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161217	20161026
626605444	`ref`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161217	20161026
629095590	`ref`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161217	20161026
638186719	`ref`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161217	20161026

Compiler output

Implementation: crypto_sign/sphincs256/avx2
Compiler: gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2

genfullchain8x.s: genfullchain8x.s: Assembler messages:
genfullchain8x.s: genfullchain8x.s:104: Error: ARM register expected -- `mov %rsp,%r11'
genfullchain8x.s: genfullchain8x.s:105: Error: ARM register expected -- `and $31,%r11'
genfullchain8x.s: genfullchain8x.s:106: Error: ARM register expected -- `add $256,%r11'
genfullchain8x.s: genfullchain8x.s:107: Error: immediate expression requires a # prefix -- `sub %r11,%rsp'
genfullchain8x.s: genfullchain8x.s:112: Error: bad instruction `vmovupd _rotate8,%ymm0'
genfullchain8x.s: genfullchain8x.s:117: Error: bad instruction `vmovupd _rotate16,%ymm1'
genfullchain8x.s: genfullchain8x.s:122: Error: bad instruction `lea hashc8x(%rip),%rdx'
genfullchain8x.s: genfullchain8x.s:127: Error: ARM register expected -- `mov %rsi,%rsi'
genfullchain8x.s: genfullchain8x.s:132: Error: bad instruction `vmovupd 0(%rdi),%ymm2'
genfullchain8x.s: ...
genfullchain8x.s: genfullchain8x.s:5139: Error: bad instruction `vmovupd %ymm8,0(%rdi)'
genfullchain8x.s: genfullchain8x.s:5144: Error: bad instruction `vmovupd %ymm6,32(%rdi)'
genfullchain8x.s: genfullchain8x.s:5149: Error: bad instruction `vmovupd %ymm5,64(%rdi)'
genfullchain8x.s: genfullchain8x.s:5154: Error: bad instruction `vmovupd %ymm4,96(%rdi)'
genfullchain8x.s: genfullchain8x.s:5159: Error: bad instruction `vmovupd %ymm3,128(%rdi)'
genfullchain8x.s: genfullchain8x.s:5164: Error: bad instruction `vmovupd %ymm1,160(%rdi)'
genfullchain8x.s: genfullchain8x.s:5169: Error: bad instruction `vmovupd %ymm2,192(%rdi)'
genfullchain8x.s: genfullchain8x.s:5174: Error: bad instruction `vmovupd %ymm0,224(%rdi)'
genfullchain8x.s: genfullchain8x.s:5177: Error: immediate expression requires a # prefix -- `add %r11,%rsp'
genfullchain8x.s: genfullchain8x.s:5178: Error: bad instruction `ret'

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

Compiler	Implementations
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2	avx2
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3	avx2
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os	avx2
gcc -mcpu=native -mfpu=neon-vfpv4 -O2	avx2
gcc -mcpu=native -mfpu=neon-vfpv4 -Os	avx2

Compiler output

Implementation: crypto_sign/sphincs256/avx2
Compiler: gcc -mcpu=native -mfpu=neon-vfpv4 -O3

sign.c: sign.c: In function 'crypto_sign_sphincs256_avx2':
sign.c: sign.c:382:1: internal compiler error: Segmentation fault
sign.c: }
sign.c: ^
sign.c: Please submit a full bug report,
sign.c: with preprocessed source if appropriate.
sign.c: See gt; for instructions.

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

Compiler	Implementations
gcc -mcpu=native -mfpu=neon-vfpv4 -O3	avx2