Implementation notes: amd64, genji346, crypto_stream/tpy

Computer: genji346
Architecture: amd64
CPU ID: AuthenticAMD-00830f10-178bfbff
SUPERCOP version: 20191017
Operation: crypto_stream
Primitive: tpy

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
9020	6544 224 34084	22045 1088 35200	`e/tpy`	`icc_-march=corei7_-mtune=corei7_-O2_-fomit-frame-pointer`	20191025	20191017
9060	7136 224 34084	23917 1088 35200	`e/tpy`	`icc_-march=haswell_-mtune=haswell_-O3_-fomit-frame-pointer`	20191025	20191017
9080	7064 224 34084	24245 1088 35200	`e/tpy`	`icc_-march=core-avx2_-mtune=core-avx2_-O2_-fomit-frame-pointer`	20191025	20191017
9140	7136 224 34084	23917 1088 35200	`e/tpy`	`icc_-march=broadwell_-mtune=broadwell_-O3_-fomit-frame-pointer`	20191025	20191017
9140	6568 224 34084	22125 1088 35200	`e/tpy`	`icc_-march=corei7_-mtune=corei7_-O3_-fomit-frame-pointer`	20191025	20191017
9140	7136 224 34084	23917 1088 35200	`e/tpy`	`icc_-march=skylake_-mtune=skylake_-O3_-fomit-frame-pointer`	20191025	20191017
9160	7136 224 34084	23917 1088 35200	`e/tpy`	`icc_-march=core-avx2_-mtune=core-avx2_-O3_-fomit-frame-pointer`	20191025	20191017
9180	7064 224 34084	24245 1088 35200	`e/tpy`	`icc_-march=haswell_-mtune=haswell_-O2_-fomit-frame-pointer`	20191025	20191017
9200	7064 224 34084	24245 1088 35200	`e/tpy`	`icc_-march=broadwell_-mtune=broadwell_-O2_-fomit-frame-pointer`	20191025	20191017
9200	7064 224 34084	24245 1088 35200	`e/tpy`	`icc_-march=skylake_-mtune=skylake_-O2_-fomit-frame-pointer`	20191025	20191017
9860	4132 218 34084	15884 1034 35232	`e/tpy`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191025	20191017
9880	4560 218 34084	18585 1042 35232	`e/tpy`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191025	20191017
9920	6904 224 34084	23749 1088 35200	`e/tpy`	`icc_-march=core-avx-i_-mtune=core-avx-i_-O3_-fomit-frame-pointer`	20191025	20191017
9920	6912 224 34084	23725 1088 35200	`e/tpy`	`icc_-march=ivybridge_-mtune=ivybridge_-O2_-fomit-frame-pointer`	20191025	20191017
9920	6904 224 34084	23749 1088 35200	`e/tpy`	`icc_-march=ivybridge_-mtune=ivybridge_-O3_-fomit-frame-pointer`	20191025	20191017
9940	6912 224 34084	23725 1088 35200	`e/tpy`	`icc_-march=core-avx-i_-mtune=core-avx-i_-O2_-fomit-frame-pointer`	20191025	20191017
9960	6912 224 34084	23725 1088 35200	`e/tpy`	`icc_-march=corei7-avx_-mtune=corei7-avx_-O2_-fomit-frame-pointer`	20191025	20191017
9960	6904 224 34084	23749 1088 35200	`e/tpy`	`icc_-march=corei7-avx_-mtune=corei7-avx_-O3_-fomit-frame-pointer`	20191025	20191017
9980	6912 224 34084	23725 1088 35200	`e/tpy`	`icc_-march=sandybridge_-mtune=sandybridge_-O2_-fomit-frame-pointer`	20191025	20191017
10000	3953 218 34084	15452 1034 35232	`e/tpy`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191025	20191017
10020	6904 224 34084	23749 1088 35200	`e/tpy`	`icc_-march=sandybridge_-mtune=sandybridge_-O3_-fomit-frame-pointer`	20191025	20191017
10260	3479 218 34084	13892 1018 35200	`e/tpy`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20191025	20191017

Test failure

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

error 111

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

Compiler	Implementations
icc -march=cannonlake -mtune=cannonlake -O2 -fomit-frame-pointer	e/tpy
icc -march=cannonlake -mtune=cannonlake -O3 -fomit-frame-pointer	e/tpy
icc -march=icelake-client -mtune=icelake-client -O2 -fomit-frame-pointer	e/tpy
icc -march=icelake-client -mtune=icelake-client -O3 -fomit-frame-pointer	e/tpy
icc -march=skylake-avx512 -mtune=skylake-avx512 -O2 -fomit-frame-pointer	e/tpy
icc -march=skylake-avx512 -mtune=skylake-avx512 -O3 -fomit-frame-pointer	e/tpy

Compiler output

Implementation: e/tpy
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/genji346/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/genji346/amd64/include/ecrypt-portable.h:23:
api.c: In file included from /scratch_gpfs_na2700_60/bdolbeaur/SC/supercop-20191017/supercop-data/genji346/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: 5, namely:

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