Implementation notes: amd64, h4atom, crypto_scalarmult/kummer

Computer: h4atom
Architecture: amd64
CPU ID: GenuineIntel-000106ca-bfe9fbff
SUPERCOP version: 20160806
Operation: crypto_scalarmult
Primitive: kummer

Time	Implementation	Compiler	Benchmark date	SUPERCOP version
6919208	`ref5u`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160814	20160806
6943720	`ref5`	`gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv`	20160814	20160806
6991968	`ref5`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160814	20160806
7037240	`ref5u`	`gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv`	20160814	20160806
7049192	`ref5u`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160814	20160806
7166816	`ref5`	`clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments`	20160814	20160806
7406440	`ref5`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160814	20160806
7475808	`ref5u`	`gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv`	20160814	20160806
7584184	`ref5`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160814	20160806
7624728	`ref5u`	`gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv`	20160814	20160806

Test failure

Implementation: crypto_scalarmult/kummer/avx
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

error 111

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	avx avx2
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	avx avx2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	avx avx2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	avx avx2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	avx avx2

Compiler output

Implementation: crypto_scalarmult/kummer/avx2int
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments

smult.c: smult.c:36:8: error: always_inline function '_mm256_mul_epi32' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t0 = _mm256_mul_epi32(a->gt;v[0],*b);
smult.c: ^
smult.c: smult.c:37:8: error: always_inline function '_mm256_mul_epi32' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t1 = _mm256_mul_epi32(a->gt;v[1],*b);
smult.c: ^
smult.c: smult.c:38:10: error: always_inline function '_mm256_add_epi64' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t1 = _mm256_add_epi64(t1,_mm256_srli_epi64(t0,26)); t0 &= mask26;
smult.c: ^
smult.c: smult.c:38:30: error: always_inline function '_mm256_srli_epi64' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t1 = _mm256_add_epi64(t1,_mm256_srli_epi64(t0,26)); t0 &= mask26;
smult.c: ^
smult.c: smult.c:39:8: error: always_inline function '_mm256_mul_epi32' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t2 = _mm256_mul_epi32(a->gt;v[2],*b);
smult.c: ^
smult.c: smult.c:40:8: error: always_inline function '_mm256_mul_epi32' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t3 = _mm256_mul_epi32(a->gt;v[3],*b);
smult.c: ^
smult.c: smult.c:41:8: error: always_inline function '_mm256_mul_epi32' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t4 = _mm256_mul_epi32(a->gt;v[4],*b);
smult.c: ^
smult.c: smult.c:42:10: error: always_inline function '_mm256_add_epi64' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: t3 = _mm256_add_epi64(t3,_mm256_srli_epi64(t2,26)); t2 &= mask26;
smult.c: ^
smult.c: smult.c:42:30: error: always_inline function '_mm256_srli_epi64' requires target feature 'xsave', but would be inlined into function 'gfe4x_mulconst' that is compiled without support for 'xsave'
smult.c: ...

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

Compiler	Implementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments	avx2int

Compiler output

Implementation: crypto_scalarmult/kummer/avx2int
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv

smult.c: smult.c: In function 'gfe4x_mulconst':
smult.c: smult.c:36:6: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
smult.c: t0 = _mm256_mul_epi32(a->gt;v[0],*b);
smult.c: ^
smult.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/5/include/immintrin.h:43:0,
smult.c: from gfe4x.h:5,
smult.c: from smult.c:3:
smult.c: /usr/lib/gcc/x86_64-linux-gnu/5/include/avx2intrin.h:126:1: error: inlining failed in call to always_inline '_mm256_add_epi64': target specific option mismatch
smult.c: _mm256_add_epi64 (__m256i __A, __m256i __B)
smult.c: ^
smult.c: smult.c:45:8: error: called from here
smult.c: t2 = _mm256_add_epi64(t2,_mm256_srli_epi64(t1,25)); t1 &= mask25;
smult.c: ^
smult.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/5/include/immintrin.h:43:0,
smult.c: from gfe4x.h:5,
smult.c: from smult.c:3:
smult.c: /usr/lib/gcc/x86_64-linux-gnu/5/include/avx2intrin.h:787:1: error: inlining failed in call to always_inline '_mm256_srli_epi64': target specific option mismatch
smult.c: _mm256_srli_epi64 (__m256i __A, int __B)
smult.c: ^
smult.c: smult.c:45:10: error: called from here
smult.c: t2 = _mm256_add_epi64(t2,_mm256_srli_epi64(t1,25)); t1 &= mask25;
smult.c: ^
smult.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/5/include/immintrin.h:43:0,
smult.c: from gfe4x.h:5,
smult.c: from smult.c:3:
smult.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv	avx2int
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv	avx2int
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv	avx2int
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv	avx2int