Implementation notes: amd64, genji262, crypto_stream/nlsv2

Computer: genji262
Architecture: amd64
CPU ID: AuthenticAMD-00800f12-178bfbff
SUPERCOP version: 20191017
Operation: crypto_stream
Primitive: nlsv2

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
4972	37152 4096 4	54543 4952 832	`e/v2/sync/2`	`icc_-march=broadwell_-mtune=broadwell_-O2_-fomit-frame-pointer`	20191024	20191017
5016	45400 4096 4	61167 4952 832	`e/v2/sync/2`	`icc_-march=corei7_-mtune=corei7_-O3_-fomit-frame-pointer`	20191024	20191017
5016	46136 4096 4	63127 4952 832	`e/v2/sync/2`	`icc_-march=skylake_-mtune=skylake_-O3_-fomit-frame-pointer`	20191024	20191017
5038	46136 4096 4	63127 4952 832	`e/v2/sync/2`	`icc_-march=broadwell_-mtune=broadwell_-O3_-fomit-frame-pointer`	20191024	20191017
5038	37152 4096 4	54543 4952 832	`e/v2/sync/2`	`icc_-march=core-avx2_-mtune=core-avx2_-O2_-fomit-frame-pointer`	20191024	20191017
5038	37152 4096 4	54543 4952 832	`e/v2/sync/2`	`icc_-march=haswell_-mtune=haswell_-O2_-fomit-frame-pointer`	20191024	20191017
5038	46136 4096 4	63127 4952 832	`e/v2/sync/2`	`icc_-march=haswell_-mtune=haswell_-O3_-fomit-frame-pointer`	20191024	20191017
5060	46136 4096 4	63127 4952 832	`e/v2/sync/2`	`icc_-march=core-avx2_-mtune=core-avx2_-O3_-fomit-frame-pointer`	20191024	20191017
5060	37152 4096 4	54543 4952 832	`e/v2/sync/2`	`icc_-march=skylake_-mtune=skylake_-O2_-fomit-frame-pointer`	20191024	20191017
5126	37096 4096 4	52807 4952 832	`e/v2/sync/2`	`icc_-march=corei7_-mtune=corei7_-O2_-fomit-frame-pointer`	20191024	20191017
5346	22992 4096 4	36527 4912 864	`e/v2/sync/2`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
5456	21000 4096 4	32999 4912 864	`e/v2/sync/2`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
5566	22751 4096 4	34535 4912 864	`e/v2/sync/2`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
5874	18154 4096 4	28778 4888 832	`e/v2/sync/2`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
5896	48512 4096 4	65551 4952 832	`e/v2/sync/2`	`icc_-march=corei7-avx_-mtune=corei7-avx_-O3_-fomit-frame-pointer`	20191024	20191017
5918	48512 4096 4	65551 4952 832	`e/v2/sync/2`	`icc_-march=core-avx-i_-mtune=core-avx-i_-O3_-fomit-frame-pointer`	20191024	20191017
5918	39480 4096 4	56487 4952 832	`e/v2/sync/2`	`icc_-march=corei7-avx_-mtune=corei7-avx_-O2_-fomit-frame-pointer`	20191024	20191017
5918	39480 4096 4	56487 4952 832	`e/v2/sync/2`	`icc_-march=ivybridge_-mtune=ivybridge_-O2_-fomit-frame-pointer`	20191024	20191017
5918	48512 4096 4	65551 4952 832	`e/v2/sync/2`	`icc_-march=ivybridge_-mtune=ivybridge_-O3_-fomit-frame-pointer`	20191024	20191017
5940	39480 4096 4	56487 4952 832	`e/v2/sync/2`	`icc_-march=sandybridge_-mtune=sandybridge_-O2_-fomit-frame-pointer`	20191024	20191017
5940	48512 4096 4	65551 4952 832	`e/v2/sync/2`	`icc_-march=sandybridge_-mtune=sandybridge_-O3_-fomit-frame-pointer`	20191024	20191017
5962	39480 4096 4	56487 4952 832	`e/v2/sync/2`	`icc_-march=core-avx-i_-mtune=core-avx-i_-O2_-fomit-frame-pointer`	20191024	20191017
10670	21848 4096 4	37559 4952 832	`e/v2/sync/1`	`icc_-march=corei7_-mtune=corei7_-O2_-fomit-frame-pointer`	20191024	20191017
10890	27440 4096 4	43207 4952 832	`e/v2/sync/1`	`icc_-march=corei7_-mtune=corei7_-O3_-fomit-frame-pointer`	20191024	20191017
11110	22584 4096 4	39975 4952 832	`e/v2/sync/1`	`icc_-march=core-avx2_-mtune=core-avx2_-O2_-fomit-frame-pointer`	20191024	20191017
11154	22584 4096 4	39975 4952 832	`e/v2/sync/1`	`icc_-march=skylake_-mtune=skylake_-O2_-fomit-frame-pointer`	20191024	20191017
11264	22584 4096 4	39975 4952 832	`e/v2/sync/1`	`icc_-march=haswell_-mtune=haswell_-O2_-fomit-frame-pointer`	20191024	20191017
11286	22584 4096 4	39975 4952 832	`e/v2/sync/1`	`icc_-march=broadwell_-mtune=broadwell_-O2_-fomit-frame-pointer`	20191024	20191017
11462	28008 4096 4	44999 4952 832	`e/v2/sync/1`	`icc_-march=broadwell_-mtune=broadwell_-O3_-fomit-frame-pointer`	20191024	20191017
11528	28008 4096 4	44999 4952 832	`e/v2/sync/1`	`icc_-march=skylake_-mtune=skylake_-O3_-fomit-frame-pointer`	20191024	20191017
11550	24560 4096 4	41567 4952 832	`e/v2/sync/1`	`icc_-march=ivybridge_-mtune=ivybridge_-O2_-fomit-frame-pointer`	20191024	20191017
11550	24560 4096 4	41567 4952 832	`e/v2/sync/1`	`icc_-march=sandybridge_-mtune=sandybridge_-O2_-fomit-frame-pointer`	20191024	20191017
11572	28008 4096 4	44999 4952 832	`e/v2/sync/1`	`icc_-march=haswell_-mtune=haswell_-O3_-fomit-frame-pointer`	20191024	20191017
11616	28008 4096 4	44999 4952 832	`e/v2/sync/1`	`icc_-march=core-avx2_-mtune=core-avx2_-O3_-fomit-frame-pointer`	20191024	20191017
11616	24560 4096 4	41567 4952 832	`e/v2/sync/1`	`icc_-march=corei7-avx_-mtune=corei7-avx_-O2_-fomit-frame-pointer`	20191024	20191017
11682	24560 4096 4	41567 4952 832	`e/v2/sync/1`	`icc_-march=core-avx-i_-mtune=core-avx-i_-O2_-fomit-frame-pointer`	20191024	20191017
11770	30712 4096 4	47751 4952 832	`e/v2/sync/1`	`icc_-march=ivybridge_-mtune=ivybridge_-O3_-fomit-frame-pointer`	20191024	20191017
11814	30712 4096 4	47751 4952 832	`e/v2/sync/1`	`icc_-march=core-avx-i_-mtune=core-avx-i_-O3_-fomit-frame-pointer`	20191024	20191017
11814	30712 4096 4	47751 4952 832	`e/v2/sync/1`	`icc_-march=corei7-avx_-mtune=corei7-avx_-O3_-fomit-frame-pointer`	20191024	20191017
11814	30712 4096 4	47751 4952 832	`e/v2/sync/1`	`icc_-march=sandybridge_-mtune=sandybridge_-O3_-fomit-frame-pointer`	20191024	20191017
13684	5248 4096 4	17247 4912 864	`e/v2/sync/1`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
14586	4948 4096 4	16727 4912 864	`e/v2/sync/1`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
14784	3849 4096 4	14482 4888 832	`e/v2/sync/1`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017
17930	8400 4096 4	21927 4912 864	`e/v2/sync/1`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191024	20191017

Test failure

Implementation: e/v2/sync/1
Security model: unknown
Compiler: icc -march=cannonlake -mtune=cannonlake -O2 -fomit-frame-pointer

error 111

Number of similar (compiler,implementation) pairs: 12, namely:

Compiler	Implementations
icc -march=cannonlake -mtune=cannonlake -O2 -fomit-frame-pointer	e/v2/sync/1
icc -march=cannonlake -mtune=cannonlake -O3 -fomit-frame-pointer	e/v2/sync/1
icc -march=icelake-client -mtune=icelake-client -O2 -fomit-frame-pointer	e/v2/sync/1
icc -march=icelake-client -mtune=icelake-client -O3 -fomit-frame-pointer	e/v2/sync/1
icc -march=skylake-avx512 -mtune=skylake-avx512 -O2 -fomit-frame-pointer	e/v2/sync/1
icc -march=skylake-avx512 -mtune=skylake-avx512 -O3 -fomit-frame-pointer	e/v2/sync/1
icc -march=cannonlake -mtune=cannonlake -O2 -fomit-frame-pointer	e/v2/sync/2
icc -march=cannonlake -mtune=cannonlake -O3 -fomit-frame-pointer	e/v2/sync/2
icc -march=icelake-client -mtune=icelake-client -O2 -fomit-frame-pointer	e/v2/sync/2
icc -march=icelake-client -mtune=icelake-client -O3 -fomit-frame-pointer	e/v2/sync/2
icc -march=skylake-avx512 -mtune=skylake-avx512 -O2 -fomit-frame-pointer	e/v2/sync/2
icc -march=skylake-avx512 -mtune=skylake-avx512 -O3 -fomit-frame-pointer	e/v2/sync/2

Compiler output

Implementation: e/v2/sync/1
Security model: unknown
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

api.c: In file included from api.c:2:
api.c: In file included from /scratch_gpfs_na2700_60/bdolbeaur/SC/supercop-20191017/supercop-data/genji262/amd64/include/estream-convert-api.h:7:
api.c: In file included from ./e/ecrypt-sync.h:15:
api.c: In file included from /scratch_gpfs_na2700_60/bdolbeaur/SC/supercop-20191017/supercop-data/genji262/amd64/include/ecrypt-portable.h:23:
api.c: In file included from /scratch_gpfs_na2700_60/bdolbeaur/SC/supercop-20191017/supercop-data/genji262/amd64/include/ecrypt-config.h:302:
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:52:1: error: conflicting types for '_mm_prefetch'
api.c: _mm_prefetch (const void *__P, enum _mm_hint __I)
api.c: ^
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:52:1: note: '_mm_prefetch' is a builtin with type 'void (const char *, int)'
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:52:1: error: definition of builtin function '_mm_prefetch'
api.c: _mm_prefetch (const void *__P, enum _mm_hint __I)
api.c: ^
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:54:3: error: argument to '__builtin_prefetch' must be a constant integer
api.c: __builtin_prefetch (__P, (__I & 0x4) >> 2, __I & 0x3);
api.c: ^ ~~~~~~~~~~~~~~~~
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:130:10: error: invalid conversion between vector type '__m128' (vector of 4 'float' values) and integer type 'int' of different size
api.c: return (__m128) __builtin_ia32_addss ((__v4sf)__A, (__v4sf)__B);
api.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:136:10: error: invalid conversion between vector type '__m128' (vector of 4 'float' values) and integer type 'int' of different size
api.c: return (__m128) __builtin_ia32_subss ((__v4sf)__A, (__v4sf)__B);
api.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:142:10: error: invalid conversion between vector type '__m128' (vector of 4 'float' values) and integer type 'int' of different size
api.c: return (__m128) __builtin_ia32_mulss ((__v4sf)__A, (__v4sf)__B);
api.c: ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
api.c: /software/compilers/gcc/gcc-9.2.0-full+isl+binutils/lib/gcc/x86_64-pc-linux-gnu/9.2.0/include/xmmintrin.h:148:10: error: invalid conversion between vector type '__m128' (vector of 4 'float' values) and integer type 'int' of different size
api.c: ...

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/1
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/1
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/1
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/1
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/1
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/2
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/2
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/2
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/2
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	e/v2/sync/2