Implementation notes: amd64, h3neo, crypto_kem/mceliece348864f

Computer: h3neo
Microarchitecture: amd64; K10 45nm (100f63)
Architecture: amd64
CPU ID: AuthenticAMD-00100f63-078bfbff
SUPERCOP version: 20240625
Operation: crypto_kem
Primitive: mceliece348864f
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
95574499930 0 0142639 900 1728T:vecclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
95729999930 0 0149247 900 1728vecclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
989622105978 0 0149127 900 1728T:vecclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
989867105978 0 0155735 900 1728vecclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
99647468575 0 0110431 900 1728T:vecclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
99689268575 0 0116975 900 1728vecclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
103644199594 0 0149344 876 1760vecgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
103656399594 0 0142736 876 1760T:vecgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
111136741274 0 082319 908 1728T:vecclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
112287541274 0 088863 908 1728vecclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
129015940816 0 089400 876 1760vecgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
130573340816 0 082856 876 1760T:vecgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
135189138018 0 084825 900 1728vecclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
135922738018 0 078217 900 1728T:vecclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
161046334817 0 074456 868 1728T:vecgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
162166134817 0 081064 868 1728vecgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
182233039845 0 080672 876 1760T:vecgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
182592539845 0 087280 876 1760vecgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
10517758342158 0 081866 884 1760refgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
10973923142158 0 075258 884 1760T:refgcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16218179444628 0 077055 900 1728T:refclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16218546744628 0 083663 900 1728refclang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16246461025956 0 057447 900 1728T:refclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16246887325956 0 064055 900 1728refclang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16450800247951 0 080823 900 1728T:refclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16452231947951 0 087431 900 1728refclang_-mcpu=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16837235916274 0 048106 884 1760T:refgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
16842270516274 0 054650 884 1760refgcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17047153112170 0 041586 876 1728T:refgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17075305112834 0 042819 900 1728T:refclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17075316912834 0 049363 900 1728refclang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17158125912170 0 048130 876 1728refgcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17327489815085 0 052208 876 1760refgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17330148615085 0 045600 876 1760T:refgcc_-march=native_-mtune=native_-O_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17526232914658 0 051895 900 1728refclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625
17740856114658 0 045287 900 1728T:refclang_-march=native_-O_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall2024071620240625

Compiler output


decrypt.c: decrypt.c:220:11: warning: AVX vector return of type 'vec256' (aka '__m256i') without 'avx' enabled changes the ABI [-Wpsabi]
decrypt.c:         allone = vec256_set1_16b(0xFFFF);
decrypt.c:                  ^
decrypt.c: decrypt.c:224:17: warning: AVX vector return of type 'vec256' (aka '__m256i') without 'avx' enabled changes the ABI [-Wpsabi]
decrypt.c:                 error256[i] = vec256_or_reduce(eval[i]);
decrypt.c:                               ^
decrypt.c: decrypt.c:225:17: warning: AVX vector argument of type 'vec256' (aka '__m256i') without 'avx' enabled changes the ABI [-Wpsabi]
decrypt.c:                 error256[i] = vec256_xor(error256[i], allone);
decrypt.c:                               ^
decrypt.c: decrypt.c:153:12: warning: AVX vector return of type 'vec256' (aka '__m256i') without 'avx' enabled changes the ABI [-Wpsabi]
decrypt.c:                 out[i] = vec256_set4x(v[0], v[1], v[2], v[3]);
decrypt.c:                          ^
decrypt.c: In file included from decrypt.c:15:
decrypt.c: In file included from ./fft_tr.h:11:
decrypt.c: ./vec256.h:35:10: error: always_inline function '_mm256_set_epi64x' requires target feature 'avx', but would be inlined into function 'vec256_set4x' that is compiled without support for 'avx'
decrypt.c:         return  _mm256_set_epi64x(a3, a2, a1, a0);
decrypt.c:                 ^
decrypt.c: ./vec256.h:35:10: error: AVX vector return of type '__m256i' (vector of 4 'long long' values) without 'avx' enabled changes the ABI
decrypt.c: decrypt.c:59:15: warning: AVX vector argument of type 'vec256' (aka '__m256i') without 'avx' enabled changes the ABI [-Wpsabi]
decrypt.c:                 out[i][j] = vec256_and(inv[i][j], recv[i]);
decrypt.c:                             ^
decrypt.c: In file included from decrypt.c:15:
decrypt.c: In file included from ./fft_tr.h:11:
decrypt.c: ./vec256.h:74:9: error: always_inline function '_mm256_and_si256' requires target feature 'avx2', but would be inlined into function 'vec256_and' that is compiled without support for 'avx2'
decrypt.c:         return _mm256_and_si256(a, b);
decrypt.c: ...

Number of similar (implementation,compiler) pairs: 10, namely:
ImplementationCompiler
avxclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:avxclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:avxclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:avxclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:avxclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
avxclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:avxclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)

Compiler output


benes.c: In file included from transpose.h:11,
benes.c:                  from benes.c:14:
benes.c: vec256.h: In function 'vec256_load':
benes.c: vec256.h:14:1: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
benes.c:    14 | {
benes.c:       | ^
bm.c: In file included from vec128.h:9,
bm.c:                  from bm.h:12,
bm.c:                  from bm.c:24:
bm.c: bm.c: In function 'get_coefs':
bm.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/smmintrin.h:455:1: error: inlining failed in call to 'always_inline' '_mm_extract_epi64': target specific option mismatch
bm.c:   455 | _mm_extract_epi64 (__m128i __X, const int __N)
bm.c:       | ^~~~~~~~~~~~~~~~~
bm.c: In file included from bm.h:12,
bm.c:                  from bm.c:24:
bm.c: vec128.h:23:42: note: called from here
bm.c:    23 | #define vec128_extract(a, i) ((uint64_t) _mm_extract_epi64((vec128) (a), (i)))
bm.c:       |                                          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
bm.c: bm.c:122:44: note: in expansion of macro 'vec128_extract'
bm.c:   122 |                 out[ (4*1 + k)*16 + i ] = (vec128_extract(buf[i], 1) >> (k*16)) & GFMASK;
bm.c:       |                                            ^~~~~~~~~~~~~~
bm.c: In file included from vec128.h:9,
bm.c:                  from bm.h:12,
bm.c:                  from bm.c:24:
bm.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/smmintrin.h:455:1: error: inlining failed in call to 'always_inline' '_mm_extract_epi64': target specific option mismatch
bm.c:   455 | _mm_extract_epi64 (__m128i __X, const int __N)
bm.c:       | ^~~~~~~~~~~~~~~~~
bm.c: In file included from bm.h:12,
bm.c:                  from bm.c:24:
bm.c: vec128.h:23:42: note: called from here
bm.c:    23 | #define vec128_extract(a, i) ((uint64_t) _mm_extract_epi64((vec128) (a), (i)))
bm.c: ...

Number of similar (implementation,compiler) pairs: 8, namely:
ImplementationCompiler
avxgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:avxgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
avxgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:avxgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
avxgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:avxgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
avxgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:avxgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)

Compiler output


decrypt.c: In file included from decrypt.c:9:
decrypt.c: In file included from ./vec128_gf.h:10:
decrypt.c: ./vec128.h:33:9: error: always_inline function '_mm_testz_si128' requires target feature 'sse4.1', but would be inlined into function 'vec128_testz' that is compiled without support for 'sse4.1'
decrypt.c:         return _mm_testz_si128(a, a);
decrypt.c:                ^
decrypt.c: 1 error generated.

Number of similar (implementation,compiler) pairs: 10, namely:
ImplementationCompiler
sseclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:sseclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
sseclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:sseclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
sseclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:sseclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
sseclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:sseclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
sseclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
T:sseclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)

Compiler output


bm.c: bm.c:120:18: warning: argument 1 of type 'uint64_t[12]' {aka 'long unsigned int[12]'} with mismatched bound [-Warray-parameter=]
bm.c:   120 | void bm(uint64_t out[ GFBITS ], vec128 in[ GFBITS ])
bm.c:       |         ~~~~~~~~~^~~~~~~~~~~~~
bm.c: In file included from bm.c:15:
bm.c: bm.h:15:19: note: previously declared as 'uint64_t *' {aka 'long unsigned int *'}
bm.c:    15 | void bm(uint64_t *out, vec128 *in);
bm.c:       |         ~~~~~~~~~~^~~
bm.c: bm.c:120:40: warning: argument 2 of type 'vec128[12]' {aka '__m128i[12]'} with mismatched bound [-Warray-parameter=]
bm.c:   120 | void bm(uint64_t out[ GFBITS ], vec128 in[ GFBITS ])
bm.c:       |                                 ~~~~~~~^~~~~~~~~~~~
bm.c: In file included from bm.c:15:
bm.c: bm.h:15:32: note: previously declared as 'vec128 *' {aka '__m128i *'}
bm.c:    15 | void bm(uint64_t *out, vec128 *in);
bm.c:       |                        ~~~~~~~~^~
bm.c: In file included from vec128.h:10,
bm.c:                  from bm.h:13,
bm.c:                  from bm.c:15:
bm.c: bm.c: In function 'get_coefs':
bm.c: /usr/lib/gcc/x86_64-linux-gnu/11/include/smmintrin.h:455:1: error: inlining failed in call to 'always_inline' '_mm_extract_epi64': target specific option mismatch
bm.c:   455 | _mm_extract_epi64 (__m128i __X, const int __N)
bm.c:       | ^~~~~~~~~~~~~~~~~
bm.c: In file included from bm.h:13,
bm.c:                  from bm.c:15:
bm.c: vec128.h:29:42: note: called from here
bm.c:    29 | #define vec128_extract(a, i) ((uint64_t) _mm_extract_epi64((vec128) (a), (i)))
bm.c: ...

Number of similar (implementation,compiler) pairs: 8, namely:
ImplementationCompiler
ssegcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:ssegcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
ssegcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:ssegcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
ssegcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:ssegcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
ssegcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
T:ssegcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)

TIMECOP error (can be valgrind bug)


error 111

Process terminating with default action of signal 4 (SIGILL)
 Illegal opcode at address 0x10E48B
   at 0x...: syndrome (encrypt.c:116)
   by 0x...: crypto_kem_mceliece348864f_ref_constbranchindex_pke_encrypt (encrypt.c:134)
   by 0x...: crypto_kem_mceliece348864f_ref_constbranchindex_operation_enc (kem_enc.c:29)
   by 0x...: test (try.c:141)
   by 0x...: main (try-anything.c:345)

Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
refclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)

TIMECOP error (can be valgrind bug)


error 111

Process terminating with default action of signal 4 (SIGILL)
 Illegal opcode at address 0x11002B
   at 0x...: syndrome (encrypt.c:116)
   by 0x...: crypto_kem_mceliece348864f_ref_constbranchindex_pke_encrypt (encrypt.c:134)
   by 0x...: crypto_kem_mceliece348864f_ref_constbranchindex_operation_enc (kem_enc.c:29)
   by 0x...: test (try.c:141)
   by 0x...: main (try-anything.c:345)

Number of similar (implementation,compiler) pairs: 1, namely:
ImplementationCompiler
refclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)

Passed TIMECOP


TIMECOP iterations: 1

Number of similar (implementation,compiler) pairs: 16, namely:
ImplementationCompiler
refclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
refgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
refgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
refgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
refgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
vecclang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
vecclang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
vecclang -march=native -O -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
vecclang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
vecclang -mcpu=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_14.0.0)
vecgcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
vecgcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
vecgcc -march=native -mtune=native -O -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)
vecgcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (11.4.0)