Implementation notes: amd64, saber214, crypto_stream/simon128256ctr

Computer: saber214
Microarchitecture: amd64; Bulldozer (600f20)
Architecture: amd64
CPU ID: AuthenticAMD-00600f20-1789c3f5
SUPERCOP version: 20230530
Operation: crypto_stream
Primitive: simon128256ctr

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
11892	101782 0 0	114996 816 792	`T:sse4`	`clang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230607	20230530
12107	102502 0 0	118108 816 808	`T:sse4`	`clang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230607	20230530
12251	138109 0 0	153086 776 864	`T:sse4`	`gcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20230607	20230530
18550	138076 0 0	151229 768 864	`T:sse4`	`gcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20230607	20230530
18678	136495 0 0	149966 776 864	`T:sse4`	`gcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20230607	20230530
19092	101377 0 0	113126 808 792	`T:sse4`	`clang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230607	20230530
19215	106306 0 0	118380 816 792	`T:sse4`	`clang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20230607	20230530
20825	136188 0 0	148185 752 832	`T:sse4`	`gcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20230607	20230530

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

stream.c: stream.c:128:24: error: always_inline function '_mm256_add_epi64' requires target feature 'avx2', but would be inlined into function 'Encrypt' that is compiled without support for 'avx2'
stream.c: SET1(X[0],nonce[1]); SET4(Y[0],nonce[0]);
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:26:38: note: expanded from macro 'SET4'
stream.c: #define SET4(X,c) (X=SET(c,c,c,c), X=ADD(X,_q))
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:17:13: note: expanded from macro 'ADD'
stream.c: #define ADD _mm256_add_epi64
stream.c: ^
stream.c: stream.c:130:21: error: always_inline function '_mm256_slli_epi64' requires target feature 'avx2', but would be inlined into function 'Encrypt' that is compiled without support for 'avx2'
stream.c: if (numbytes==64) Enc(X,Y,rk,4);
stream.c: ^
stream.c: ./Simon128256AVX2.h:45:24: note: expanded from macro 'Enc'
stream.c: #define Enc(X,Y,rk,n) (R2x##n(X,Y,rk,0,1), R2x##n(X,Y,rk,2,3), R2x##n(X,Y,rk,4,5), R2x##n(X,Y,rk,6,7), R2x##n(X,Y,rk,8,9), \
stream.c: ^
stream.c: <scratch space>:227:1: note: expanded from here
stream.c: R2x4
stream.c: ^
stream.c: ./Simon128256AVX2.h:32:28: note: expanded from macro 'R2x4'
stream.c: #define R2x4(X,Y,rk,r,s) (R1x4(X,Y,rk,r), R1x4(Y,X,rk,s))
stream.c: ^
stream.c: note: (skipping 1 expansions in backtrace; use -fmacro-backtrace-limit=0 to see all)
stream.c: ./Simon128256AVX2.h:24:28: note: expanded from macro 'RD'
stream.c: #define RD(U,V,k) (V=XOR(V,ROL(U,2)), V=XOR(V,AND(ROL(U,1),ROL8(U))), V=XOR(V,k))
stream.c: ^
stream.c: ...

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

Compiler	Implementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:avx2
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:avx2
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:avx2
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:avx2

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

stream.c: stream.c:128:3: error: always_inline function '_mm256_set_epi64x' requires target feature 'avx', but would be inlined into function 'Encrypt' that is compiled without support for 'avx'
stream.c: SET1(X[0],nonce[1]); SET4(Y[0],nonce[0]);
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:25:22: note: expanded from macro 'SET1'
stream.c: #define SET1(X,c) (X=SET(c,c,c,c))
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:24:13: note: expanded from macro 'SET'
stream.c: #define SET _mm256_set_epi64x
stream.c: ^
stream.c: stream.c:128:3: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
stream.c: ./Intrinsics_AVX2_128block.h:25:22: note: expanded from macro 'SET1'
stream.c: #define SET1(X,c) (X=SET(c,c,c,c))
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:24:13: note: expanded from macro 'SET'
stream.c: #define SET _mm256_set_epi64x
stream.c: ^
stream.c: stream.c:128:24: error: always_inline function '_mm256_set_epi64x' requires target feature 'avx', but would be inlined into function 'Encrypt' that is compiled without support for 'avx'
stream.c: SET1(X[0],nonce[1]); SET4(Y[0],nonce[0]);
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:26:22: note: expanded from macro 'SET4'
stream.c: #define SET4(X,c) (X=SET(c,c,c,c), X=ADD(X,_q))
stream.c: ^
stream.c: ./Intrinsics_AVX2_128block.h:24:13: note: expanded from macro 'SET'
stream.c: #define SET _mm256_set_epi64x
stream.c: ^
stream.c: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:avx2

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

stream.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
stream.c: from Intrinsics_AVX2_128block.h:5,
stream.c: from Simon128256AVX2.h:17,
stream.c: from stream.c:20:
stream.c: stream.c: In function 'Transpose':
stream.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:913:1: error: inlining failed in call to 'always_inline' '_mm256_xor_si256': target specific option mismatch
stream.c: 913 | _mm256_xor_si256 (__m256i __A, __m256i __B)
stream.c: | ^~~~~~~~~~~~~~~~
stream.c: In file included from Simon128256AVX2.h:17,
stream.c: from stream.c:20:
stream.c: Intrinsics_AVX2_128block.h:15:13: note: called from here
stream.c: 15 | #define XOR _mm256_xor_si256
stream.c: stream.c:412:10: note: in expansion of macro 'XOR'
stream.c: 412 | T[6] = XOR(T[6], SL(W[3],1));
stream.c: | ^~~
stream.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:47,
stream.c: from Intrinsics_AVX2_128block.h:5,
stream.c: from Simon128256AVX2.h:17,
stream.c: from stream.c:20:
stream.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:696:1: error: inlining failed in call to 'always_inline' '_mm256_slli_epi64': target specific option mismatch
stream.c: 696 | _mm256_slli_epi64 (__m256i __A, int __B)
stream.c: | ^~~~~~~~~~~~~~~~~
stream.c: In file included from Simon128256AVX2.h:17,
stream.c: from stream.c:20:
stream.c: Intrinsics_AVX2_128block.h:15:13: note: called from here
stream.c: ...

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

Compiler	Implementations
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:avx2
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:avx2
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:avx2
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:avx2

Compiler output

Implementation: T:sse4
Security model: timingleaks
Compiler: clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE

stream.c: stream.c:307:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'ExpandKeyBS' that is compiled without support for 'ssse3'
stream.c: EKBS(rk);
stream.c: ^
stream.c: ./Simon128256SSE4.h:67:19: note: expanded from macro 'EKBS'
stream.c: #define EKBS(rk) (RKBS(rk,4,_D), RKBS(rk,5,_D), RKBS(rk,6,_C), RKBS(rk,7,_D), RKBS(rk,8,_C), RKBS(rk,9,_C), RKBS(rk,10,_C), RKBS(rk,11,_D), \
stream.c: ^
stream.c: ./Simon128256SSE4.h:57:52: note: expanded from macro 'RKBS'
stream.c: #define RKBS(rk,r,_V) (rk[r][7]= _D ^ rk[r-4][7] ^ ROR8(rk[r-1][2]) ^ rk[r-3][7] ^ ROR8(rk[r-1][3]) ^ ROR8(rk[r-3][0]), \
stream.c: ^
stream.c: ./Intrinsics_SSE4_128block.h:39:19: note: expanded from macro 'ROR8'
stream.c: #define ROR8(X) (SHFL(X,R8))
stream.c: ^
stream.c: ./Intrinsics_SSE4_128block.h:35:14: note: expanded from macro 'SHFL'
stream.c: #define SHFL _mm_shuffle_epi8
stream.c: ^
stream.c: stream.c:307:3: error: always_inline function '_mm_shuffle_epi8' requires target feature 'ssse3', but would be inlined into function 'ExpandKeyBS' that is compiled without support for 'ssse3'
stream.c: ./Simon128256SSE4.h:67:19: note: expanded from macro 'EKBS'
stream.c: #define EKBS(rk) (RKBS(rk,4,_D), RKBS(rk,5,_D), RKBS(rk,6,_C), RKBS(rk,7,_D), RKBS(rk,8,_C), RKBS(rk,9,_C), RKBS(rk,10,_C), RKBS(rk,11,_D), \
stream.c: ^
stream.c: ./Simon128256SSE4.h:57:85: note: expanded from macro 'RKBS'
stream.c: #define RKBS(rk,r,_V) (rk[r][7]= _D ^ rk[r-4][7] ^ ROR8(rk[r-1][2]) ^ rk[r-3][7] ^ ROR8(rk[r-1][3]) ^ ROR8(rk[r-3][0]), \
stream.c: ^
stream.c: ./Intrinsics_SSE4_128block.h:39:19: note: expanded from macro 'ROR8'
stream.c: #define ROR8(X) (SHFL(X,R8))
stream.c: ^
stream.c: ...

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

Compiler	Implementations
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:sse4