Implementation notes: amd64, par, crypto_stream/salsa2012

Computer: par
Architecture: amd64
CPU ID: GenuineIntel-000406c3-bfebfbff
SUPERCOP version: 20161026
Operation: crypto_stream
Primitive: salsa2012

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
6640	`dolbeau/amd64-xmm6int`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
6920	`dolbeau/amd64-xmm6int`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
6920	`dolbeau/amd64-xmm6int`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
6960	`e/amd64-xmm2`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
6960	`e/amd64-xmm2`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
6960	`e/amd64-xmm2`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
6960	`e/amd64-xmm2`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
6980	`e/amd64-xmm2`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
6980	`e/amd64-xmm2`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
7060	`dolbeau/amd64-xmm6int`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
7360	`e/amd64-xmm`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
7360	`e/amd64-xmm`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
7360	`e/amd64-xmm`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
7360	`e/amd64-xmm`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
7360	`e/amd64-xmm`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
7380	`e/amd64-xmm`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
7460	`dolbeau/amd64-xmm6int`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
7780	`amd64_xmm6`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
7780	`amd64_xmm6`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
7780	`amd64_xmm6`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
7780	`amd64_xmm6`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
7780	`amd64_xmm6`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
7800	`amd64_xmm6`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
7800	`dolbeau/amd64-xmm6int`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
8020	`e/amd64-xmm5`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
8020	`e/amd64-xmm5`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
8020	`e/amd64-xmm5`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
8040	`e/amd64-xmm5`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
8040	`e/amd64-xmm5`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
8040	`e/amd64-xmm5`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
8060	`e/amd64-xmm6`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
8060	`e/amd64-xmm6`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
8060	`e/amd64-xmm6`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
8060	`e/amd64-xmm6`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
8080	`e/amd64-xmm6`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
8080	`e/amd64-xmm6`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13620	`e/amd64-3`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
13660	`e/amd64-3`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
13660	`e/amd64-3`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
13660	`e/amd64-3`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
13660	`e/amd64-3`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
13720	`e/amd64-3`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14080	`e/ref`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14500	`e/regs`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14740	`e/amd64-2`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
14760	`e/amd64-2`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14760	`e/merged`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
14760	`e/amd64-2`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
14760	`e/amd64-2`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
14760	`e/amd64-2`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
14800	`e/amd64-2`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14820	`e/merged`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
14820	`e/ref`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
15000	`e/regs`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
15020	`e/merged`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
15080	`e/merged`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
15800	`e/merged`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
15980	`e/merged`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
16580	`ref`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
16660	`ref`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
17460	`e/amd64-1`	`gcc -funroll-loops -march=native -mcpu=native -O3`	20161214	20161026
17500	`e/amd64-1`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
17520	`e/amd64-1`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
17520	`e/amd64-1`	`gcc -march=native -mcpu=native -O3`	20161214	20161026
17520	`e/amd64-1`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
17620	`e/amd64-1`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
18660	`e/ref`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
18820	`e/regs`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
19860	`e/regs`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
20980	`ref`	`gcc -funroll-loops -march=native -mcpu=native -O2`	20161214	20161026
21980	`ref`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
23440	`ref`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
23440	`ref`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
23780	`e/regs`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
24180	`e/ref`	`gcc -march=native -mcpu=native -O2`	20161214	20161026
27960	`e/ref`	`gcc -funroll-loops -march=native -mcpu=native -Os`	20161214	20161026
29000	`e/regs`	`gcc -march=native -mcpu=native -Os`	20161214	20161026
33380	`e/ref`	`gcc -march=native -mcpu=native -Os`	20161214	20161026

Compiler output

Implementation: crypto_stream/salsa2012/dolbeau/amd64-xmm6int
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

api.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
salsa20-xmm6int.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

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

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	dolbeau/amd64-xmm6int
gcc -funroll-loops -march=native -mcpu=native -O3	dolbeau/amd64-xmm6int
gcc -funroll-loops -march=native -mcpu=native -Os	dolbeau/amd64-xmm6int
gcc -march=native -mcpu=native -O2	dolbeau/amd64-xmm6int
gcc -march=native -mcpu=native -O3	dolbeau/amd64-xmm6int
gcc -march=native -mcpu=native -Os	dolbeau/amd64-xmm6int

Compiler output

Implementation: crypto_stream/salsa2012/e/merged
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

api.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
salsa20.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 18, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	e/merged e/ref e/regs
gcc -funroll-loops -march=native -mcpu=native -O3	e/merged e/ref e/regs
gcc -funroll-loops -march=native -mcpu=native -Os	e/merged e/ref e/regs
gcc -march=native -mcpu=native -O2	e/merged e/ref e/regs
gcc -march=native -mcpu=native -O3	e/merged e/ref e/regs
gcc -march=native -mcpu=native -Os	e/merged e/ref e/regs

Compiler output

Implementation: crypto_stream/salsa2012/e/amd64-1
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

api.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
salsa20.s: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

Number of similar (compiler,implementation) pairs: 42, namely:

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	e/amd64-1 e/amd64-2 e/amd64-3 e/amd64-xmm e/amd64-xmm2 e/amd64-xmm5 e/amd64-xmm6
gcc -funroll-loops -march=native -mcpu=native -O3	e/amd64-1 e/amd64-2 e/amd64-3 e/amd64-xmm e/amd64-xmm2 e/amd64-xmm5 e/amd64-xmm6
gcc -funroll-loops -march=native -mcpu=native -Os	e/amd64-1 e/amd64-2 e/amd64-3 e/amd64-xmm e/amd64-xmm2 e/amd64-xmm5 e/amd64-xmm6
gcc -march=native -mcpu=native -O2	e/amd64-1 e/amd64-2 e/amd64-3 e/amd64-xmm e/amd64-xmm2 e/amd64-xmm5 e/amd64-xmm6
gcc -march=native -mcpu=native -O3	e/amd64-1 e/amd64-2 e/amd64-3 e/amd64-xmm e/amd64-xmm2 e/amd64-xmm5 e/amd64-xmm6
gcc -march=native -mcpu=native -Os	e/amd64-1 e/amd64-2 e/amd64-3 e/amd64-xmm e/amd64-xmm2 e/amd64-xmm5 e/amd64-xmm6

Compiler output

Implementation: crypto_stream/salsa2012/ref
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

stream.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
xor.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

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

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	ref
gcc -funroll-loops -march=native -mcpu=native -O3	ref
gcc -funroll-loops -march=native -mcpu=native -Os	ref
gcc -march=native -mcpu=native -O2	ref
gcc -march=native -mcpu=native -O3	ref
gcc -march=native -mcpu=native -Os	ref

Compiler output

Implementation: crypto_stream/salsa2012/amd64_xmm6
Compiler: gcc -funroll-loops -march=native -mcpu=native -O2

stream.s: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
try.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead
measure.c: gcc: warning: '-mcpu=' is deprecated; use '-mtune=' or '-march=' instead

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

Compiler	Implementations
gcc -funroll-loops -march=native -mcpu=native -O2	amd64_xmm6
gcc -funroll-loops -march=native -mcpu=native -O3	amd64_xmm6
gcc -funroll-loops -march=native -mcpu=native -Os	amd64_xmm6
gcc -march=native -mcpu=native -O2	amd64_xmm6
gcc -march=native -mcpu=native -O3	amd64_xmm6
gcc -march=native -mcpu=native -Os	amd64_xmm6