Implementation notes: amd64, icelake, crypto_hash/blake3

Computer: icelake
Architecture: amd64
CPU ID: GenuineIntel-000706e5-bfebfbff
SUPERCOP version: 20200826
Operation: crypto_hash
Primitive: blake3

Time	Object size	Test size	Implementation	Compiler	Benchmark date	SUPERCOP version
4493	17272 272 0	28672 4096 0	`T:avx512`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
4508	18139 272 0	32768 4096 0	`T:avx512`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
4538	16766 272 0	28672 4096 0	`T:avx512`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
4555	18790 272 0	32768 4096 0	`T:avx512`	`clang_-march=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
4558	18726 272 0	32768 4096 0	`T:avx512`	`clang_-march=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
4559	17019 272 0	28672 4096 0	`T:avx512`	`clang_-march=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
5570	22529 480 0	36864 4096 0	`T:avx2`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
5572	23284 480 0	36864 4096 0	`T:avx2`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
5577	12933 176 0	24576 4096 0	`T:sse41`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
5592	13504 176 0	28672 4096 0	`T:sse41`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
5622	22007 480 0	32768 4096 0	`T:avx2`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
5626	12427 176 0	24576 4096 0	`T:sse41`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
5647	23967 480 0	36864 4096 0	`T:avx2`	`clang_-march=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
5648	24031 480 0	36864 4096 0	`T:avx2`	`clang_-march=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
5649	14387 176 0	28672 4096 0	`T:sse41`	`clang_-march=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
5649	14451 176 0	28672 4096 0	`T:sse41`	`clang_-march=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
5652	22260 480 0	32768 4096 0	`T:avx2`	`clang_-march=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
5652	12680 176 0	24576 4096 0	`T:sse41`	`clang_-march=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
7326	10286 0 0	24576 4096 0	`T:portable`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
7326	10801 0 0	24576 4096 0	`T:portable`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
7463	11406 0 0	24576 4096 0	`T:portable`	`clang_-march=icelake-client_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
7511	10290 0 0	20480 4096 0	`T:portable`	`clang_-march=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826
7617	9971 0 0	20480 4096 0	`T:portable`	`opt-local-bin-gcc_-march=icelake-client_-mtune=icelake-client_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE`	20200826	20200826
7679	11294 0 0	24576 4096 0	`T:portable`	`clang_-march=icelake-client_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE`	20200826	20200826

Compiler output

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

blake3.c: In file included from blake3.c:12:
blake3.c: ./blake3_static_dispatch.h:17:9: warning: 'MAX_SIMD_DEGREE' macro redefined [-Wmacro-redefined]
blake3.c: #define MAX_SIMD_DEGREE 4
blake3.c: ^
blake3.c: ./blake3_impl.h:49:9: note: previous definition is here
blake3.c: #define MAX_SIMD_DEGREE 16
blake3.c: ^
blake3.c: In file included from blake3.c:12:
blake3.c: ./blake3_static_dispatch.h:18:9: warning: 'MAX_SIMD_DEGREE_OR_2' macro redefined [-Wmacro-redefined]
blake3.c: #define MAX_SIMD_DEGREE_OR_2 4
blake3.c: ^
blake3.c: ./blake3_impl.h:58:9: note: previous definition is here
blake3.c: #define MAX_SIMD_DEGREE_OR_2 (MAX_SIMD_DEGREE > 2 ? MAX_SIMD_DEGREE : 2)
blake3.c: ^
blake3.c: 2 warnings generated.
blake3_neon.c: In file included from blake3_neon.c:3:
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:28:2: error: "NEON support not enabled"
blake3_neon.c: #error "NEON support not enabled"
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:48:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(8))) int8_t int8x8_t;
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:49:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(16))) int8_t int8x16_t;
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:50:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(4))) int16_t int16x4_t;
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:51:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(8))) int16_t int16x8_t;
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:52:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(2))) int32_t int32x2_t;
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:53:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(4))) int32_t int32x4_t;
blake3_neon.c: ^
blake3_neon.c: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/11.0.3/include/arm_neon.h:54:24: error: 'neon_vector_type' attribute is not supported for this target
blake3_neon.c: typedef __attribute__((neon_vector_type(1))) int64_t int64x1_t;
blake3_neon.c: ^
blake3_neon.c: ...

Number of similar (compiler,implementation) pairs: 3, namely:

Compiler	Implementations
clang -march=icelake-client -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:neon
clang -march=icelake-client -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:neon
clang -march=icelake-client -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE	T:neon

Compiler output

Implementation: T:neon
Security model: timingleaks
Compiler: opt-local-bin-gcc -march=icelake-client -mtune=icelake-client -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE

blake3.c: In file included from blake3.c:12:
blake3.c: blake3_static_dispatch.h:17: warning: "MAX_SIMD_DEGREE" redefined
blake3.c: 17 | #define MAX_SIMD_DEGREE 4
blake3.c: |
blake3.c: In file included from blake3.c:6:
blake3.c: blake3_impl.h:49: note: this is the location of the previous definition
blake3.c: 49 | #define MAX_SIMD_DEGREE 16
blake3.c: |
blake3.c: In file included from blake3.c:12:
blake3.c: blake3_static_dispatch.h:18: warning: "MAX_SIMD_DEGREE_OR_2" redefined
blake3.c: 18 | #define MAX_SIMD_DEGREE_OR_2 4
blake3.c: |
blake3.c: In file included from blake3.c:6:
blake3.c: blake3_impl.h:58: note: this is the location of the previous definition
blake3.c: 58 | #define MAX_SIMD_DEGREE_OR_2 (MAX_SIMD_DEGREE > 2 ? MAX_SIMD_DEGREE : 2)
blake3.c: |
blake3_neon.c: blake3_neon.c:3:10: fatal error: arm_neon.h: No such file or directory
blake3_neon.c: 3 | #include <arm_neon.h>
blake3_neon.c: | ^~~~~~~~~~~~
blake3_neon.c: compilation terminated.

Number of similar (compiler,implementation) pairs: 3, namely:

Compiler	Implementations
opt-local-bin-gcc -march=icelake-client -mtune=icelake-client -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:neon
opt-local-bin-gcc -march=icelake-client -mtune=icelake-client -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:neon
opt-local-bin-gcc -march=icelake-client -mtune=icelake-client -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE	T:neon