Implementation notes: armeabi, c1, crypto_hash/blake512

Computer: c1
Architecture: armeabi
CPU ID: unknown CPU ID
SUPERCOP version: 20161026
Operation: crypto_hash
Primitive: blake512

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
67105	`vect128-neon`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
67106	`vect128-neon`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
67155	`vect128-neon`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
67344	`vect128-neon`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
68861	`vect128-neon`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
72650	`vect128-neon`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
183209	`bswap`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
195300	`ref`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
195775	`ref`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
196205	`sphlib-small`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
200981	`sphlib-small`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
203040	`sphlib`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
208989	`bswap`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
214232	`sphlib`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
224237	`regs`	`gcc -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
238938	`regs`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os`	20161210	20161026
247791	`sphlib-small`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
255245	`sphlib`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
255836	`sphlib`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
258083	`sphlib-small`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
262941	`sphlib-small`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
267463	`ref`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
273875	`ref`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
291911	`sphlib`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
295011	`ref`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
308725	`sphlib-small`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
310217	`sphlib`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
310997	`ref`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
335253	`bswap`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
340893	`regs`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
349644	`regs`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
373406	`bswap`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
373627	`bswap`	`gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
386561	`bswap`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026
397319	`regs`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O2`	20161210	20161026
413696	`regs`	`gcc -mcpu=native -mfpu=neon-vfpv4 -O3`	20161210	20161026

Compiler output

Implementation: crypto_hash/blake512/vect128
Compiler: gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2

vector.c: /tmp/ccSyiqvd.s: Assembler messages:
vector.c: /tmp/ccSyiqvd.s:282: Error: invalid instruction shape -- `vmov d28,s6'
vector.c: /tmp/ccSyiqvd.s:283: Error: invalid instruction shape -- `vmov d29,s22'
vector.c: /tmp/ccSyiqvd.s:313: Error: invalid instruction shape -- `vmov d10,s26'
vector.c: /tmp/ccSyiqvd.s:338: Error: invalid instruction shape -- `vmov d18,s24'
vector.c: /tmp/ccSyiqvd.s:355: Error: invalid instruction shape -- `vmov d12,s26'
vector.c: /tmp/ccSyiqvd.s:356: Error: invalid instruction shape -- `vmov d13,s24'
vector.c: /tmp/ccSyiqvd.s:366: Error: invalid instruction shape -- `vmov d11,s6'
vector.c: /tmp/ccSyiqvd.s:415: Error: invalid instruction shape -- `vmov d6,s14'
vector.c: /tmp/ccSyiqvd.s:528: Error: invalid instruction shape -- `vmov d20,s2'
vector.c: /tmp/ccSyiqvd.s:529: Error: invalid instruction shape -- `vmov d21,s18'
vector.c: /tmp/ccSyiqvd.s:544: Error: invalid instruction shape -- `vmov d30,s30'
vector.c: /tmp/ccSyiqvd.s:616: Error: invalid instruction shape -- `vmov d27,s22'
vector.c: /tmp/ccSyiqvd.s:684: Error: invalid instruction shape -- `vmov d27,s10'
vector.c: /tmp/ccSyiqvd.s:693: Error: invalid instruction shape -- `vmov d15,s26'
vector.c: /tmp/ccSyiqvd.s:743: Error: invalid instruction shape -- `vmov d30,s22'
vector.c: /tmp/ccSyiqvd.s:744: Error: invalid instruction shape -- `vmov d31,s2'
vector.c: /tmp/ccSyiqvd.s:756: Error: invalid instruction shape -- `vmov d4,s20'
vector.c: /tmp/ccSyiqvd.s:781: Error: invalid instruction shape -- `vmov d4,s10'
vector.c: /tmp/ccSyiqvd.s:793: Error: invalid instruction shape -- `vmov d26,s22'
vector.c: /tmp/ccSyiqvd.s:840: Error: invalid instruction shape -- `vmov d10,s20'
vector.c: /tmp/ccSyiqvd.s:862: Error: invalid instruction shape -- `vmov d29,s14'
vector.c: /tmp/ccSyiqvd.s:918: Error: invalid instruction shape -- `vmov d30,s28'
vector.c: /tmp/ccSyiqvd.s:919: Error: invalid instruction shape -- `vmov d31,s18'
vector.c: /tmp/ccSyiqvd.s:932: Error: invalid instruction shape -- `vmov d2,s30'
vector.c: ...

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

Compiler	Implementations
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2	vect128

Compiler output

Implementation: crypto_hash/blake512/vect128-inplace
Compiler: gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2

vector.c: In file included from vector.c:588:0:
vector.c: vector.c: In function 'round512':
vector.c: perm512.h:6:10: warning: implicit declaration of function 'v64_interleavel' [-Wimplicit-function-declaration]
vector.c: m0 = v64_interleavel(mm0, mm1); \
vector.c: ^
vector.c: vector.h:5:19: note: in expansion of macro 'PERM_512_0'
vector.c: #define CAT(x, y) x##y
vector.c: ^
vector.c: vector.h:6:19: note: in expansion of macro 'CAT'
vector.c: #define XCAT(x,y) CAT(x,y)
vector.c: ^~~
vector.c: perm512.h:1:17: note: in expansion of macro 'XCAT'
vector.c: #define PERM(i) XCAT(PERM_512_,i)
vector.c: ^~~~
vector.c: vector.c:670:5: note: in expansion of macro 'PERM'
vector.c: PERM(i); \
vector.c: ^~~~
vector.c: vector.c:753:3: note: in expansion of macro 'ROUND'
vector.c: ROUND( 0); ROUND( 1); ROUND( 2); ROUND( 3);
vector.c: ^~~~~
vector.c: perm512.h:6:8: error: incompatible types when assigning to type 'v64' from type 'int'
vector.c: m0 = v64_interleavel(mm0, mm1); \
vector.c: ^
vector.c: vector.h:5:19: note: in expansion of macro 'PERM_512_0'
vector.c: #define CAT(x, y) x##y
vector.c: ...

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

Compiler	Implementations
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O2	vect128-inplace
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3	vect128-inplace
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os	vect128-inplace
gcc -mcpu=native -mfpu=neon-vfpv4 -O2	vect128-inplace
gcc -mcpu=native -mfpu=neon-vfpv4 -O3	vect128-inplace
gcc -mcpu=native -mfpu=neon-vfpv4 -Os	vect128-inplace

Compiler output

Implementation: crypto_hash/blake512/vect128
Compiler: gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3

vector.c: /tmp/ccvRKCax.s: Assembler messages:
vector.c: /tmp/ccvRKCax.s:282: Error: invalid instruction shape -- `vmov d28,s6'
vector.c: /tmp/ccvRKCax.s:283: Error: invalid instruction shape -- `vmov d29,s22'
vector.c: /tmp/ccvRKCax.s:313: Error: invalid instruction shape -- `vmov d10,s26'
vector.c: /tmp/ccvRKCax.s:338: Error: invalid instruction shape -- `vmov d18,s24'
vector.c: /tmp/ccvRKCax.s:355: Error: invalid instruction shape -- `vmov d12,s26'
vector.c: /tmp/ccvRKCax.s:356: Error: invalid instruction shape -- `vmov d13,s24'
vector.c: /tmp/ccvRKCax.s:366: Error: invalid instruction shape -- `vmov d11,s6'
vector.c: /tmp/ccvRKCax.s:415: Error: invalid instruction shape -- `vmov d6,s14'
vector.c: /tmp/ccvRKCax.s:528: Error: invalid instruction shape -- `vmov d20,s2'
vector.c: /tmp/ccvRKCax.s:529: Error: invalid instruction shape -- `vmov d21,s18'
vector.c: /tmp/ccvRKCax.s:544: Error: invalid instruction shape -- `vmov d30,s30'
vector.c: /tmp/ccvRKCax.s:616: Error: invalid instruction shape -- `vmov d27,s22'
vector.c: /tmp/ccvRKCax.s:684: Error: invalid instruction shape -- `vmov d27,s10'
vector.c: /tmp/ccvRKCax.s:693: Error: invalid instruction shape -- `vmov d15,s26'
vector.c: /tmp/ccvRKCax.s:743: Error: invalid instruction shape -- `vmov d30,s22'
vector.c: /tmp/ccvRKCax.s:744: Error: invalid instruction shape -- `vmov d31,s2'
vector.c: /tmp/ccvRKCax.s:756: Error: invalid instruction shape -- `vmov d4,s20'
vector.c: /tmp/ccvRKCax.s:781: Error: invalid instruction shape -- `vmov d4,s10'
vector.c: /tmp/ccvRKCax.s:793: Error: invalid instruction shape -- `vmov d26,s22'
vector.c: /tmp/ccvRKCax.s:840: Error: invalid instruction shape -- `vmov d10,s20'
vector.c: /tmp/ccvRKCax.s:862: Error: invalid instruction shape -- `vmov d29,s14'
vector.c: /tmp/ccvRKCax.s:918: Error: invalid instruction shape -- `vmov d30,s28'
vector.c: /tmp/ccvRKCax.s:919: Error: invalid instruction shape -- `vmov d31,s18'
vector.c: /tmp/ccvRKCax.s:932: Error: invalid instruction shape -- `vmov d2,s30'
vector.c: ...

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

Compiler	Implementations
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -O3	vect128

Compiler output

Implementation: crypto_hash/blake512/vect128
Compiler: gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os

vector.c: /tmp/cckkkiLJ.s: Assembler messages:
vector.c: /tmp/cckkkiLJ.s:221: Error: invalid instruction shape -- `vmov d22,s2'
vector.c: /tmp/cckkkiLJ.s:232: Error: invalid instruction shape -- `vmov d20,s4'
vector.c: /tmp/cckkkiLJ.s:233: Error: invalid instruction shape -- `vmov d21,s6'
vector.c: /tmp/cckkkiLJ.s:245: Error: invalid instruction shape -- `vmov d12,s8'
vector.c: /tmp/cckkkiLJ.s:246: Error: invalid instruction shape -- `vmov d13,s10'
vector.c: /tmp/cckkkiLJ.s:259: Error: invalid instruction shape -- `vmov d10,s12'
vector.c: /tmp/cckkkiLJ.s:260: Error: invalid instruction shape -- `vmov d11,s14'
vector.c: /tmp/cckkkiLJ.s:279: Error: invalid instruction shape -- `vmov d8,s16'
vector.c: /tmp/cckkkiLJ.s:280: Error: invalid instruction shape -- `vmov d9,s18'
vector.c: /tmp/cckkkiLJ.s:563: Error: invalid instruction shape -- `vmov d12,s22'
vector.c: /tmp/cckkkiLJ.s:574: Error: invalid instruction shape -- `vmov d6,s26'
vector.c: /tmp/cckkkiLJ.s:575: Error: invalid instruction shape -- `vmov d7,s30'
vector.c: /tmp/cckkkiLJ.s:588: Error: invalid instruction shape -- `vmov d26,s18'
vector.c: /tmp/cckkkiLJ.s:593: Error: invalid instruction shape -- `vmov d29,s20'
vector.c: /tmp/cckkkiLJ.s:596: Error: invalid instruction shape -- `vmov d22,s16'
vector.c: /tmp/cckkkiLJ.s:597: Error: invalid instruction shape -- `vmov d23,s28'
vector.c: /tmp/cckkkiLJ.s:600: Error: invalid instruction shape -- `vmov d14,s22'
vector.c: /tmp/cckkkiLJ.s:601: Error: invalid instruction shape -- `vmov d15,s24'
vector.c: /tmp/cckkkiLJ.s:615: Error: invalid instruction shape -- `vmov d8,s26'
vector.c: /tmp/cckkkiLJ.s:616: Error: invalid instruction shape -- `vmov d9,s18'
vector.c: /tmp/cckkkiLJ.s:852: Error: invalid instruction shape -- `vmov d4,s2'
vector.c: /tmp/cckkkiLJ.s:863: Error: invalid instruction shape -- `vmov d6,s6'
vector.c: /tmp/cckkkiLJ.s:864: Error: invalid instruction shape -- `vmov d7,s14'
vector.c: /tmp/cckkkiLJ.s:880: Error: invalid instruction shape -- `vmov d27,s4'
vector.c: ...

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

Compiler	Implementations
gcc -funroll-loops -mcpu=native -mfpu=neon-vfpv4 -Os	vect128

Compiler output

Implementation: crypto_hash/blake512/vect128
Compiler: gcc -mcpu=native -mfpu=neon-vfpv4 -O2

vector.c: /tmp/ccXtI8LC.s: Assembler messages:
vector.c: /tmp/ccXtI8LC.s:215: Error: invalid instruction shape -- `vmov d20,s26'
vector.c: /tmp/ccXtI8LC.s:271: Error: invalid instruction shape -- `vmov d29,s14'
vector.c: /tmp/ccXtI8LC.s:288: Error: invalid instruction shape -- `vmov d31,s20'
vector.c: /tmp/ccXtI8LC.s:295: Error: invalid instruction shape -- `vmov d6,s12'
vector.c: /tmp/ccXtI8LC.s:296: Error: invalid instruction shape -- `vmov d7,s10'
vector.c: /tmp/ccXtI8LC.s:307: Error: invalid instruction shape -- `vmov d10,s20'
vector.c: /tmp/ccXtI8LC.s:308: Error: invalid instruction shape -- `vmov d11,s8'
vector.c: /tmp/ccXtI8LC.s:322: Error: invalid instruction shape -- `vmov d7,s16'
vector.c: /tmp/ccXtI8LC.s:334: Error: invalid instruction shape -- `vmov d8,s18'
vector.c: /tmp/ccXtI8LC.s:335: Error: invalid instruction shape -- `vmov d9,s24'
vector.c: /tmp/ccXtI8LC.s:344: Error: invalid instruction shape -- `vmov d6,s10'
vector.c: /tmp/ccXtI8LC.s:354: Error: invalid instruction shape -- `vmov d4,s10'
vector.c: /tmp/ccXtI8LC.s:355: Error: invalid instruction shape -- `vmov d5,s28'
vector.c: /tmp/ccXtI8LC.s:638: Error: invalid instruction shape -- `vmov d24,s24'
vector.c: /tmp/ccXtI8LC.s:639: Error: invalid instruction shape -- `vmov d25,s2'
vector.c: /tmp/ccXtI8LC.s:698: Error: invalid instruction shape -- `vmov d30,s26'
vector.c: /tmp/ccXtI8LC.s:719: Error: invalid instruction shape -- `vmov d28,s24'
vector.c: /tmp/ccXtI8LC.s:734: Error: invalid instruction shape -- `vmov d31,s18'
vector.c: /tmp/ccXtI8LC.s:747: Error: invalid instruction shape -- `vmov d28,s26'
vector.c: /tmp/ccXtI8LC.s:748: Error: invalid instruction shape -- `vmov d29,s18'
vector.c: /tmp/ccXtI8LC.s:807: Error: invalid instruction shape -- `vmov d4,s2'
vector.c: /tmp/ccXtI8LC.s:808: Error: invalid instruction shape -- `vmov d5,s6'
vector.c: /tmp/ccXtI8LC.s:819: Error: invalid instruction shape -- `vmov d28,s6'
vector.c: /tmp/ccXtI8LC.s:873: Error: invalid instruction shape -- `vmov d31,s26'
vector.c: ...

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

Compiler	Implementations
gcc -mcpu=native -mfpu=neon-vfpv4 -O2	vect128

Compiler output

Implementation: crypto_hash/blake512/vect128
Compiler: gcc -mcpu=native -mfpu=neon-vfpv4 -O3

vector.c: /tmp/ccKkglJZ.s: Assembler messages:
vector.c: /tmp/ccKkglJZ.s:284: Error: invalid instruction shape -- `vmov d16,s20'
vector.c: /tmp/ccKkglJZ.s:285: Error: invalid instruction shape -- `vmov d17,s22'
vector.c: /tmp/ccKkglJZ.s:312: Error: invalid instruction shape -- `vmov d22,s24'
vector.c: /tmp/ccKkglJZ.s:371: Error: invalid instruction shape -- `vmov d12,s24'
vector.c: /tmp/ccKkglJZ.s:382: Error: invalid instruction shape -- `vmov d17,s6'
vector.c: /tmp/ccKkglJZ.s:411: Error: invalid instruction shape -- `vmov d26,s14'
vector.c: /tmp/ccKkglJZ.s:521: Error: invalid instruction shape -- `vmov d28,s2'
vector.c: /tmp/ccKkglJZ.s:522: Error: invalid instruction shape -- `vmov d29,s16'
vector.c: /tmp/ccKkglJZ.s:610: Error: invalid instruction shape -- `vmov d26,s24'
vector.c: /tmp/ccKkglJZ.s:754: Error: invalid instruction shape -- `vmov d30,s22'
vector.c: /tmp/ccKkglJZ.s:755: Error: invalid instruction shape -- `vmov d31,s2'
vector.c: /tmp/ccKkglJZ.s:758: Error: invalid instruction shape -- `vmov d4,s20'
vector.c: /tmp/ccKkglJZ.s:785: Error: invalid instruction shape -- `vmov d4,s10'
vector.c: /tmp/ccKkglJZ.s:789: Error: invalid instruction shape -- `vmov d26,s22'
vector.c: /tmp/ccKkglJZ.s:810: Error: invalid instruction shape -- `vmov d26,s20'
vector.c: /tmp/ccKkglJZ.s:928: Error: invalid instruction shape -- `vmov d30,s16'
vector.c: /tmp/ccKkglJZ.s:929: Error: invalid instruction shape -- `vmov d31,s18'
vector.c: /tmp/ccKkglJZ.s:941: Error: invalid instruction shape -- `vmov d2,s30'
vector.c: /tmp/ccKkglJZ.s:942: Error: invalid instruction shape -- `vmov d3,s14'
vector.c: /tmp/ccKkglJZ.s:950: Error: invalid instruction shape -- `vmov d30,s16'
vector.c: /tmp/ccKkglJZ.s:959: Error: invalid instruction shape -- `vmov d26,s12'
vector.c: /tmp/ccKkglJZ.s:972: Error: invalid instruction shape -- `vmov d30,s16'
vector.c: /tmp/ccKkglJZ.s:973: Error: invalid instruction shape -- `vmov d31,s18'
vector.c: /tmp/ccKkglJZ.s:976: Error: invalid instruction shape -- `vmov d26,s30'
vector.c: ...

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

Compiler	Implementations
gcc -mcpu=native -mfpu=neon-vfpv4 -O3	vect128

Compiler output

Implementation: crypto_hash/blake512/vect128
Compiler: gcc -mcpu=native -mfpu=neon-vfpv4 -Os

vector.c: /tmp/cc9MA1p5.s: Assembler messages:
vector.c: /tmp/cc9MA1p5.s:592: Error: invalid instruction shape -- `vmov d14,s16'
vector.c: /tmp/cc9MA1p5.s:593: Error: invalid instruction shape -- `vmov d15,s18'
vector.c: /tmp/cc9MA1p5.s:628: Error: invalid instruction shape -- `vmov d8,s16'
vector.c: /tmp/cc9MA1p5.s:629: Error: invalid instruction shape -- `vmov d9,s18'
vector.c: /tmp/cc9MA1p5.s:2720: Error: invalid instruction shape -- `vmov d14,s28'
vector.c: /tmp/cc9MA1p5.s:2721: Error: invalid instruction shape -- `vmov d15,s30'
vector.c: /tmp/cc9MA1p5.s:3003: Error: invalid instruction shape -- `vmov d14,s28'
vector.c: /tmp/cc9MA1p5.s:3004: Error: invalid instruction shape -- `vmov d15,s30'
vector.c: /tmp/cc9MA1p5.s:3431: Error: invalid instruction shape -- `vmov d0,s6'
vector.c: /tmp/cc9MA1p5.s:3432: Error: invalid instruction shape -- `vmov d1,s4'
vector.c: /tmp/cc9MA1p5.s:3474: Error: invalid instruction shape -- `vmov d2,s6'
vector.c: /tmp/cc9MA1p5.s:3475: Error: invalid instruction shape -- `vmov d3,s4'

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

Compiler	Implementations
gcc -mcpu=native -mfpu=neon-vfpv4 -Os	vect128