Implementation notes: amd64, hertz, crypto_sign/dilithium2
Computer: hertz
Microarchitecture: amd64; Zen 4 (a60f12)
Architecture: amd64
CPU ID: AuthenticAMD-00a60f12-178bfbff
SUPERCOP version: 20240716
Operation: crypto_sign
Primitive: dilithium2
Time | Object size | Test size | Implementation | Compiler | Benchmark date | SUPERCOP version |
269241 | 128387 64 0 | 160769 916 1832 | avx2 | clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
269378 | 124314 64 0 | 156785 916 1832 | avx2 | clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
283291 | 76457 64 0 | 102963 908 1832 | avx2 | clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
293617 | 104446 64 0 | 133140 892 1896 | avx2 | gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
295481 | 65218 64 0 | 91948 892 1832 | avx2 | gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
299509 | 58999 64 0 | 84172 884 1800 | avx2 | gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
1379853 | 18703 0 0 | 45667 836 1832 | ref | clang_-march=native_-Os_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
1396793 | 80047 0 0 | 112377 844 1832 | ref | clang_-march=native_-O3_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
1399532 | 76747 0 0 | 108953 844 1832 | ref | clang_-march=native_-O2_-fwrapv_-Qunused-arguments_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
1401821 | 61379 0 0 | 90060 820 1896 | ref | gcc_-march=native_-mtune=native_-O3_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
1441491 | 22228 0 0 | 48852 820 1832 | ref | gcc_-march=native_-mtune=native_-O2_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
1597882 | 18747 0 0 | 43860 812 1800 | ref | gcc_-march=native_-mtune=native_-Os_-fwrapv_-fPIC_-fPIE_-gdwarf-4_-Wall | 20240716 | 20240716 |
Compiler output
rejsample.c: rejsample.c:316:62: warning: argument 'buf' of type 'const uint8_t[168]' (aka 'const unsigned char[168]') with mismatched bound [-Warray-parameter]
rejsample.c: 316 | unsigned int rej_eta_avx(int32_t * restrict r, const uint8_t buf[REJ_UNIFORM_ETA_BUFLEN]) {
rejsample.c: | ^
rejsample.c: ./rejsample.h:25:52: note: previously declared as 'const uint8_t[840]' (aka 'const unsigned char[840]') here
rejsample.c: 25 | unsigned int rej_eta_avx(int32_t *r, const uint8_t buf[REJ_UNIFORM_BUFLEN]);
rejsample.c: | ^
rejsample.c: 1 warning generated.
Number of similar (implementation,compiler) pairs: 3, namely:
Implementation | Compiler |
avx2 | clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
avx2 | clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
avx2 | clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
Compiler output
poly.c: poly.c: In function 'crypto_sign_dilithium2_avx2_constbranchindex_poly_uniform_eta_4x':
poly.c: <command-line>: warning: 'crypto_sign_dilithium2_avx2_constbranchindex_rej_eta_avx' reading 840 bytes from a region of size 768 [-Wstringop-overread]
poly.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
poly.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
poly.c: | ^~~~~~~~~~~~~~~~
poly.c: rejsample.h:24:21: note: in expansion of macro 'DILITHIUM_NAMESPACE'
poly.c: 24 | #define rej_eta_avx DILITHIUM_NAMESPACE(rej_eta_avx)
poly.c: | ^~~~~~~~~~~~~~~~~~~
poly.c: poly.c:594:10: note: in expansion of macro 'rej_eta_avx'
poly.c: 594 | ctr0 = rej_eta_avx(a0->coeffs, buf[0].coeffs);
poly.c: | ^~~~~~~~~~~
poly.c: <command-line>: note: referencing argument 2 of type 'const uint8_t[840]' {aka 'const unsigned char[840]'}
poly.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
poly.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
poly.c: | ^~~~~~~~~~~~~~~~
poly.c: rejsample.h:24:21: note: in expansion of macro 'DILITHIUM_NAMESPACE'
poly.c: 24 | #define rej_eta_avx DILITHIUM_NAMESPACE(rej_eta_avx)
poly.c: | ^~~~~~~~~~~~~~~~~~~
poly.c: poly.c:594:10: note: in expansion of macro 'rej_eta_avx'
poly.c: 594 | ctr0 = rej_eta_avx(a0->coeffs, buf[0].coeffs);
poly.c: | ^~~~~~~~~~~
poly.c: <command-line>: note: in a call to function 'crypto_sign_dilithium2_avx2_constbranchindex_rej_eta_avx'
poly.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
poly.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
poly.c: | ^~~~~~~~~~~~~~~~
poly.c: ...
polyvec.c: polyvec.c: In function 'crypto_sign_dilithium2_avx2_constbranchindex_polyvecl_uniform_gamma1':
polyvec.c: <command-line>: warning: 'crypto_sign_dilithium2_avx2_constbranchindex_poly_uniform_gamma1' reading 48 bytes from a region of size 32 [-Wstringop-overread]
polyvec.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
polyvec.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
polyvec.c: | ^~~~~~~~~~~~~~~~
polyvec.c: poly.h:55:29: note: in expansion of macro 'DILITHIUM_NAMESPACE'
polyvec.c: 55 | #define poly_uniform_gamma1 DILITHIUM_NAMESPACE(poly_uniform_gamma1)
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: polyvec.c:279:5: note: in expansion of macro 'poly_uniform_gamma1'
polyvec.c: 279 | poly_uniform_gamma1(&v->vec[i], seed, L*nonce + i);
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: <command-line>: note: referencing argument 2 of type 'const uint8_t[48]' {aka 'const unsigned char[48]'}
polyvec.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
polyvec.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
polyvec.c: | ^~~~~~~~~~~~~~~~
polyvec.c: poly.h:55:29: note: in expansion of macro 'DILITHIUM_NAMESPACE'
polyvec.c: 55 | #define poly_uniform_gamma1 DILITHIUM_NAMESPACE(poly_uniform_gamma1)
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: polyvec.c:279:5: note: in expansion of macro 'poly_uniform_gamma1'
polyvec.c: 279 | poly_uniform_gamma1(&v->vec[i], seed, L*nonce + i);
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: <command-line>: note: in a call to function 'crypto_sign_dilithium2_avx2_constbranchindex_poly_uniform_gamma1'
polyvec.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
polyvec.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
polyvec.c: | ^~~~~~~~~~~~~~~~
polyvec.c: ...
rejsample.c: rejsample.c:316:62: warning: argument 2 of type 'const uint8_t[168]' {aka 'const unsigned char[168]'} with mismatched bound [-Warray-parameter=]
rejsample.c: 316 | unsigned int rej_eta_avx(int32_t * restrict r, const uint8_t buf[REJ_UNIFORM_ETA_BUFLEN]) {
rejsample.c: | ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
rejsample.c: In file included from rejsample.c:4:
rejsample.c: rejsample.h:25:52: note: previously declared as 'const uint8_t[840]' {aka 'const unsigned char[840]'}
rejsample.c: 25 | unsigned int rej_eta_avx(int32_t *r, const uint8_t buf[REJ_UNIFORM_BUFLEN]);
rejsample.c: | ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
Number of similar (implementation,compiler) pairs: 3, namely:
Implementation | Compiler |
avx2 | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
avx2 | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
avx2 | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
Compiler output
polyvec.c: polyvec.c: In function 'crypto_sign_dilithium2_ref_constbranchindex_polyvecl_uniform_gamma1':
polyvec.c: <command-line>: warning: 'crypto_sign_dilithium2_ref_constbranchindex_poly_uniform_gamma1' reading 48 bytes from a region of size 32 [-Wstringop-overread]
polyvec.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
polyvec.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
polyvec.c: | ^~~~~~~~~~~~~~~~
polyvec.c: poly.h:51:29: note: in expansion of macro 'DILITHIUM_NAMESPACE'
polyvec.c: 51 | #define poly_uniform_gamma1 DILITHIUM_NAMESPACE(poly_uniform_gamma1)
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: polyvec.c:47:5: note: in expansion of macro 'poly_uniform_gamma1'
polyvec.c: 47 | poly_uniform_gamma1(&v->vec[i], seed, L*nonce + i);
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: <command-line>: note: referencing argument 2 of type 'const uint8_t[48]' {aka 'const unsigned char[48]'}
polyvec.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
polyvec.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
polyvec.c: | ^~~~~~~~~~~~~~~~
polyvec.c: poly.h:51:29: note: in expansion of macro 'DILITHIUM_NAMESPACE'
polyvec.c: 51 | #define poly_uniform_gamma1 DILITHIUM_NAMESPACE(poly_uniform_gamma1)
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: polyvec.c:47:5: note: in expansion of macro 'poly_uniform_gamma1'
polyvec.c: 47 | poly_uniform_gamma1(&v->vec[i], seed, L*nonce + i);
polyvec.c: | ^~~~~~~~~~~~~~~~~~~
polyvec.c: <command-line>: note: in a call to function 'crypto_sign_dilithium2_ref_constbranchindex_poly_uniform_gamma1'
polyvec.c: config.h:3:32: note: in expansion of macro 'CRYPTO_NAMESPACE'
polyvec.c: 3 | #define DILITHIUM_NAMESPACE(s) CRYPTO_NAMESPACE(s)
polyvec.c: | ^~~~~~~~~~~~~~~~
polyvec.c: ...
Number of similar (implementation,compiler) pairs: 3, namely:
Implementation | Compiler |
ref | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
ref | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
ref | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
TIMECOP error (can be valgrind bug)
error 111
Process terminating with default action of signal 4 (SIGILL)
Illegal opcode at address 0x10B066
at 0x...: core (try-anything.c:61)
by 0x...: salsa20 (try-anything.c:101)
by 0x...: testvector (try-anything.c:124)
by 0x...: myrandom (try-anything.c:132)
by 0x...: test (try.c:124)
by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 2, namely:
Implementation | Compiler |
avx2 | clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
ref | clang -march=native -O2 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
TIMECOP error (can be valgrind bug)
error 111
Process terminating with default action of signal 4 (SIGILL)
Illegal opcode at address 0x10BB84
at 0x...: salsa20 (try-anything.c:90)
by 0x...: canary (try-anything.c:148)
by 0x...: output_prepare (try-anything.c:178)
by 0x...: test (try.c:126)
by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 2, namely:
Implementation | Compiler |
avx2 | clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
ref | clang -march=native -O3 -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
TIMECOP error (can be valgrind bug)
error 111
Process terminating with default action of signal 4 (SIGILL)
Illegal opcode at address 0x10A22E
at 0x...: core (try-anything.c:64)
by 0x...: salsa20 (try-anything.c:101)
by 0x...: testvector (try-anything.c:124)
by 0x...: myrandom (try-anything.c:132)
by 0x...: test (try.c:124)
by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 2, namely:
Implementation | Compiler |
avx2 | clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
ref | clang -march=native -Os -fwrapv -Qunused-arguments -fPIC -fPIE -gdwarf-4 -Wall (Ubuntu_Clang_18.1.3_(1ubuntu1)) |
TIMECOP error (can be valgrind bug)
error 111
Process terminating with default action of signal 4 (SIGILL)
Illegal opcode at address 0x10A267
at 0x...: core (try-anything.c:68)
by 0x...: salsa20.part.0 (try-anything.c:101)
by 0x...: salsa20 (try-anything.c:85)
by 0x...: testvector (try-anything.c:124)
by 0x...: myrandom (try-anything.c:132)
by 0x...: test (try.c:124)
by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 2, namely:
Implementation | Compiler |
avx2 | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
ref | gcc -march=native -mtune=native -O2 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
TIMECOP error (can be valgrind bug)
error 111
Process terminating with default action of signal 4 (SIGILL)
Illegal opcode at address 0x10A2BC
at 0x...: core (try-anything.c:64)
by 0x...: salsa20 (try-anything.c:101)
by 0x...: salsa20 (try-anything.c:81)
by 0x...: testvector (try-anything.c:124)
by 0x...: myrandom (try-anything.c:132)
by 0x...: test (try.c:124)
by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 2, namely:
Implementation | Compiler |
avx2 | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
ref | gcc -march=native -mtune=native -O3 -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
TIMECOP error (can be valgrind bug)
error 111
Process terminating with default action of signal 4 (SIGILL)
Illegal opcode at address 0x10A064
at 0x...: core (try-anything.c:64)
by 0x...: salsa20.part.0 (try-anything.c:101)
by 0x...: salsa20 (try-anything.c:85)
by 0x...: testvector (try-anything.c:124)
by 0x...: myrandom (try-anything.c:132)
by 0x...: test (try.c:124)
by 0x...: main (try-anything.c:345)
Number of similar (implementation,compiler) pairs: 2, namely:
Implementation | Compiler |
avx2 | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |
ref | gcc -march=native -mtune=native -Os -fwrapv -fPIC -fPIE -gdwarf-4 -Wall (13.2.0) |