Implementation notes: amd64, gcc14, crypto_sign/luov8117404

Computer: gcc14
Architecture: amd64
CPU ID: GenuineIntel-00010676-bfebfbff
SUPERCOP version: 20220506
Operation: crypto_sign
Primitive: luov8117404
TimeObject sizeTest sizeImplementationCompilerBenchmark dateSUPERCOP version
155213758558973 0 0118992 876 1768T:portablegcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
159522066554095 0 0114479 860 1736T:portableclang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
160437392553951 0 0114159 860 1736T:portableclang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
161197188553807 0 0113575 860 1736T:portableclang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
161877422553807 0 0113575 860 1736T:portableclang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
166258507549048 0 0109507 844 1736T:portableclang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
166646423552662 0 0113104 876 1768T:portablegcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
183996076551832 0 0112688 876 1768T:portablegcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
200205356548166 0 0108936 868 1736T:portablegcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
926093861412627 36 0235271 860 1736T:refclang_-mcpu=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
929766275412755 36 0235559 860 1736T:refclang_-march=native_-O3_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
938813487410719 36 0234232 876 1768T:refgcc_-march=native_-mtune=native_-O_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
948130878412547 36 0234591 860 1736T:refclang_-march=native_-O_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
957556619412547 36 0234591 860 1736T:refclang_-march=native_-O2_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
961550590412017 36 0234992 876 1768T:refgcc_-march=native_-mtune=native_-O2_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
1024174585408692 36 0231587 844 1736T:refclang_-march=native_-Os_-fomit-frame-pointer_-fwrapv_-Qunused-arguments_-fPIC_-fPIE2021011620210114
1029695395408571 36 0231880 868 1736T:refgcc_-march=native_-mtune=native_-Os_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114
1215657232417037 36 0238744 876 1768T:refgcc_-march=native_-mtune=native_-O3_-fomit-frame-pointer_-fwrapv_-fPIC_-fPIE2021011620210114

Compiler output

Implementation: T:avx2
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
LUOV.c: LUOV.c:110:17: error: '__builtin_ia32_permti256' needs target feature avx2
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^
LUOV.c: /usr/lib/llvm-3.8/bin/../lib/clang/3.8.1/include/avx2intrin.h:882:12: note: expanded from macro '_mm256_permute2x128_si256'
LUOV.c: (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (M)); })
LUOV.c: ^
LUOV.c: LUOV.c:110:43: error: always_inline function '_mm256_loadu_si256' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^
LUOV.c: LUOV.c:110:77: error: always_inline function '_mm256_setzero_si256' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^
LUOV.c: LUOV.c:115:20: error: always_inline function '_mm256_set1_epi8' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: __m256i tttt = _mm256_set1_epi8(t[k/8]);
LUOV.c: ^
LUOV.c: LUOV.c:117:20: error: always_inline function '_mm256_cmpeq_epi8' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: __m256i t1t2 = _mm256_cmpeq_epi8(tttt & masks[0],_mm256_setzero_si256());
LUOV.c: ^
LUOV.c: LUOV.c:117:54: error: always_inline function '_mm256_setzero_si256' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: __m256i t1t2 = _mm256_cmpeq_epi8(tttt & masks[0],_mm256_setzero_si256());
LUOV.c: ^
LUOV.c: LUOV.c:118:39: error: always_inline function '_mm256_andnot_si256' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: *((__m256i *)&TempMat[i][k+0]) ^= _mm256_andnot_si256(t1t2,rr);
LUOV.c: ^
LUOV.c: LUOV.c:120:20: error: always_inline function '_mm256_cmpeq_epi8' requires target feature 'xsave', but would be inlined into function 'calculateQ2' that is compiled without support for 'xsave'
LUOV.c: ...

Number of similar (compiler,implementation) pairs: 4, namely:
CompilerImplementations
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
LUOV.c: LUOV.c:110:17: error: '__builtin_ia32_permti256' needs target feature avx2
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^
LUOV.c: /usr/lib/llvm-3.8/bin/../lib/clang/3.8.1/include/avx2intrin.h:882:12: note: expanded from macro '_mm256_permute2x128_si256'
LUOV.c: (__m256i)__builtin_ia32_permti256((__m256i)(V1), (__m256i)(V2), (M)); })
LUOV.c: ^
LUOV.c: LUOV.c:110:43: error: always_inline function '_mm256_loadu_si256' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^
LUOV.c: LUOV.c:110:77: error: always_inline function '_mm256_setzero_si256' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^
LUOV.c: LUOV.c:115:20: error: always_inline function '_mm256_set1_epi8' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: __m256i tttt = _mm256_set1_epi8(t[k/8]);
LUOV.c: ^
LUOV.c: LUOV.c:117:20: error: always_inline function '_mm256_cmpeq_epi8' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: __m256i t1t2 = _mm256_cmpeq_epi8(tttt & masks[0],_mm256_setzero_si256());
LUOV.c: ^
LUOV.c: LUOV.c:117:54: error: always_inline function '_mm256_setzero_si256' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: __m256i t1t2 = _mm256_cmpeq_epi8(tttt & masks[0],_mm256_setzero_si256());
LUOV.c: ^
LUOV.c: LUOV.c:118:39: error: always_inline function '_mm256_andnot_si256' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: *((__m256i *)&TempMat[i][k+0]) ^= _mm256_andnot_si256(t1t2,rr);
LUOV.c: ^
LUOV.c: LUOV.c:120:20: error: always_inline function '_mm256_cmpeq_epi8' requires target feature 'sse4.2', but would be inlined into function 'calculateQ2' that is compiled without support for 'sse4.2'
LUOV.c: ...

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
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
LUOV.c: LUOV.c: In function 'calculateQ2':
LUOV.c: LUOV.c:110:12: warning: AVX vector return without AVX enabled changes the ABI [-Wpsabi]
LUOV.c: __m256i rr = _mm256_permute2x128_si256(_mm256_loadu_si256((__m256i *)&r),_mm256_setzero_si256(),0);
LUOV.c: ^~
LUOV.c: In file included from LinearAlgebra.h:9:0,
LUOV.c: from LUOV.h:13,
LUOV.c: from LUOV.c:1:
LUOV.c: AVX_Operations.h: In function 'scalarMul_ct':
LUOV.c: AVX_Operations.h:529:6: note: The ABI for passing parameters with 32-byte alignment has changed in GCC 4.6
LUOV.c: void scalarMul_ct(__m256i *Out, __m256i A, FELT b){
LUOV.c: ^~~~~~~~~~~~
LUOV.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
LUOV.c: from LUOV.h:7,
LUOV.c: from LUOV.c:1:
LUOV.c: AVX_Operations.h: In function 'addScalarProductAVX':
LUOV.c: /usr/lib/gcc/x86_64-linux-gnu/6/include/avx2intrin.h:186:1: error: inlining failed in call to always_inline '_mm256_andnot_si256': target specific option mismatch
LUOV.c: _mm256_andnot_si256 (__m256i __A, __m256i __B)
LUOV.c: ^~~~~~~~~~~~~~~~~~~
LUOV.c: In file included from LinearAlgebra.h:9:0,
LUOV.c: from LUOV.h:13,
LUOV.c: from LUOV.c:1:
LUOV.c: AVX_Operations.h:87:7: note: called from here
LUOV.c: avx4 = _mm256_andnot_si256(avx4,aa);
LUOV.c: ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
LUOV.c: In file included from /usr/lib/gcc/x86_64-linux-gnu/6/include/immintrin.h:43:0,
LUOV.c: ...

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

Namespace violations

Implementation: T:portable
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
Bitcontainer.o deserialize_bitcontainer T
Bitcontainer.o flipBit T
Bitcontainer.o getBit T
Bitcontainer.o randomBitcontainer T
Bitcontainer.o serialize_bitcontainer T
Bitcontainer.o squeezeBitcontainerArray T
Bitcontainer.o xor T
F16Field.o f16EucildeanDivision T
F16Field.o f16ExtendedEuclideanAlgorithm T
F16Field.o f16antilog T
F16Field.o f16deserialize_FELT T
F16Field.o f16inverse T
F16Field.o f16log T
F16Field.o f16multiply T
F16Field.o f16polyAdd T
F16Field.o f16polyCopy T
F16Field.o f16polyMult T
F16Field.o f16polyOne T
F16Field.o f16polyZero T
F16Field.o f16printFELT T
F16Field.o f16scalarMultiply T
F16Field.o f16serialize_FELT T
F16Field.o isArrayOfZeros T
F32Field.o f32add T
F32Field.o f32addInPlace T
F32Field.o f32deserialize_FELT T
F32Field.o f32inverse T
F32Field.o f32isEqual T
F32Field.o f32multiply T
F32Field.o f32multiplyOld T
F32Field.o f32printFELT T
F32Field.o f32scalarMultiply T
F32Field.o f32serialize_FELT T
F32Field.o newF32FELT T
F48Field.o f48add T
F48Field.o f48addInPlace T
F48Field.o f48deserialize_FELT T
F48Field.o f48inverse T
F48Field.o f48isEqual T
F48Field.o f48multiply T
F48Field.o f48printFELT T
F48Field.o f48scalarMultiply T
F48Field.o f48serialize_FELT T
F48Field.o newF48FELT T
F64Field.o f64add T
F64Field.o f64addInPlace T
F64Field.o f64deserialize_FELT T
F64Field.o f64inverse T
F64Field.o f64isEqual T
F64Field.o f64multiply T
F64Field.o f64printFELT T
F64Field.o f64scalarMultiply T
F64Field.o f64serialize_FELT T
F64Field.o newF64FELT T
F80Field.o f80Scalarmultiply T
F80Field.o f80add T
F80Field.o f80addInPlace T
F80Field.o f80deserialize_FELT T
F80Field.o f80inverse T
F80Field.o f80isEqual T
F80Field.o f80multiply T
F80Field.o f80printFELT T
F80Field.o f80serialize_FELT T
F80Field.o newF80FELT T
F8Field.o f8antilog T
F8Field.o f8deserialize_FELT T
F8Field.o f8inverse T
F8Field.o f8log T
F8Field.o f8multiply T
F8Field.o f8printFELT T
F8Field.o f8serialize_FELT T
LUOV.o BuildAugmentedMatrix T
LUOV.o _addScalarProduct1 T
LUOV.o _addScalarProduct3 T
LUOV.o addScalarProduct T
LUOV.o addScalarProduct3 T
LUOV.o calculateQ2 T
LUOV.o computeTarget T
LUOV.o deserialize_PublicKey T
LUOV.o deserialize_SecretKey T
LUOV.o deserialize_signature T
LUOV.o destroy_PublicKey T
LUOV.o destroy_SecretKey T
LUOV.o destroy_signature T
LUOV.o evaluatePublicMap T
LUOV.o expandTable T
LUOV.o extractMessage T
LUOV.o generateKeyPair T
LUOV.o repeatTable T
LUOV.o serialize_PublicKey T
LUOV.o serialize_SecretKey T
LUOV.o serialize_signature T
LUOV.o signDocument T
LUOV.o solvePrivateUOVSystem T
LUOV.o verify T
LinearAlgebra.o destroy_matrix T
LinearAlgebra.o getUniqueSolution T
LinearAlgebra.o newMatrix T
LinearAlgebra.o printMatrix T
LinearAlgebra.o rowEchelonAugmented T
LinearAlgebra.o rowOp T
LinearAlgebra.o scaleRow T
LinearAlgebra.o swapRows T
LinearAlgebra.o zeroMatrix T
buffer.o deserialize_uint64_t T
buffer.o newReader T
buffer.o newWriter T
buffer.o readBit T
buffer.o serialize_uint64_t T
buffer.o transcribe T
buffer.o writeBit T
intermediateValues.o printAugmentedMatrix T
intermediateValues.o printEvaluation T
intermediateValues.o printPrivateSolution T
intermediateValues.o printVinegarValues T
intermediateValues.o reportSolutionFound T
keccakrng.o initializeAndAbsorb T
keccakrng.o squeezeVector T
keccakrng.o squeezeuint64_t T

Number of similar (compiler,implementation) pairs: 9, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:portable
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:portable
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:portable
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:portable
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:portable
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:portable
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:portable
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE T:portable
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE T:portable

Namespace violations

Implementation: T:ref
Security model: timingleaks
Compiler: clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE
Bitcontainer.o deserialize_bitcontainer T
Bitcontainer.o flipBit T
Bitcontainer.o getBit T
Bitcontainer.o randomBitcontainer T
Bitcontainer.o serialize_bitcontainer T
Bitcontainer.o squeezeBitcontainerArray T
Bitcontainer.o xor T
F16Field.o f16EucildeanDivision T
F16Field.o f16ExtendedEuclideanAlgorithm T
F16Field.o f16antilog T
F16Field.o f16deserialize_FELT T
F16Field.o f16inverse T
F16Field.o f16log T
F16Field.o f16multiply T
F16Field.o f16polyAdd T
F16Field.o f16polyCopy T
F16Field.o f16polyMult T
F16Field.o f16polyOne T
F16Field.o f16polyZero T
F16Field.o f16printFELT T
F16Field.o f16scalarMultiply T
F16Field.o f16serialize_FELT T
F16Field.o isArrayOfZeros T
F32Field.o f32add T
F32Field.o f32deserialize_FELT T
F32Field.o f32inverse T
F32Field.o f32isEqual T
F32Field.o f32multiply T
F32Field.o f32multiplyOld T
F32Field.o f32printFELT T
F32Field.o f32scalarMultiply T
F32Field.o f32serialize_FELT T
F32Field.o newF32FELT T
F48Field.o f48add T
F48Field.o f48deserialize_FELT T
F48Field.o f48inverse T
F48Field.o f48isEqual T
F48Field.o f48multiply T
F48Field.o f48printFELT T
F48Field.o f48scalarMultiply T
F48Field.o f48serialize_FELT T
F48Field.o newF48FELT T
F64Field.o f64add T
F64Field.o f64deserialize_FELT T
F64Field.o f64inverse T
F64Field.o f64isEqual T
F64Field.o f64multiply T
F64Field.o f64printFELT T
F64Field.o f64scalarMultiply T
F64Field.o f64serialize_FELT T
F64Field.o newF64FELT T
F80Field.o f80Scalarmultiply T
F80Field.o f80add T
F80Field.o f80deserialize_FELT T
F80Field.o f80inverse T
F80Field.o f80isEqual T
F80Field.o f80multiply T
F80Field.o f80printFELT T
F80Field.o f80serialize_FELT T
F80Field.o newF80FELT T
F8Field.o f8antilog T
F8Field.o f8deserialize_FELT T
F8Field.o f8inverse T
F8Field.o f8log T
F8Field.o f8multiply T
F8Field.o f8printFELT T
F8Field.o f8serialize_FELT T
LUOV.o BuildAugmentedMatrix T
LUOV.o calculateQ2 T
LUOV.o computeTarget T
LUOV.o deserialize_PublicKey T
LUOV.o deserialize_SecretKey T
LUOV.o deserialize_signature T
LUOV.o destroy_PublicKey T
LUOV.o destroy_SecretKey T
LUOV.o destroy_signature T
LUOV.o evaluatePublicMap T
LUOV.o extractMessage T
LUOV.o generateKeyPair T
LUOV.o serialize_PublicKey T
LUOV.o serialize_SecretKey T
LUOV.o serialize_signature T
LUOV.o signDocument T
LUOV.o solvePrivateUOVSystem T
LUOV.o verify T
LinearAlgebra.o destroy_matrix T
LinearAlgebra.o getUniqueSolution T
LinearAlgebra.o newMatrix T
LinearAlgebra.o printMatrix T
LinearAlgebra.o rowEchelonAugmented T
LinearAlgebra.o rowOp T
LinearAlgebra.o scaleRow T
LinearAlgebra.o swapRows T
LinearAlgebra.o zeroMatrix T
buffer.o deserialize_uint64_t T
buffer.o newReader T
buffer.o newWriter T
buffer.o readBit T
buffer.o serialize_uint64_t T
buffer.o transcribe T
buffer.o writeBit T
intermediateValues.o printAugmentedMatrix T
intermediateValues.o printEvaluation T
intermediateValues.o printPrivateSolution T
intermediateValues.o printVinegarValues T
intermediateValues.o reportSolutionFound T
keccakrng.o initializeAndAbsorb T
keccakrng.o squeezeVector T
keccakrng.o squeezeuint64_t T

Number of similar (compiler,implementation) pairs: 9, namely:
CompilerImplementations
clang -march=native -O2 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -march=native -O -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -march=native -Os -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
clang -mcpu=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments -fPIC -fPIE T:ref
gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref
gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref
gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref
gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv -fPIC -fPIE T:ref