ADD: added other eigen lib

libs/eigen/bench/btl/libs/STL/STL_interface.hh

@@ -84,9 +84,12 @@ public :
     for (int j=0;j<N;j++){
       for (int i=0;i<N;i++){
         somme=0.0;
+        if(i>=j)
+        {
         for (int k=0;k<N;k++)
           somme += A[i][k]*A[j][k];
         X[j][i]=somme;
+	}
       }
     }
   }

libs/eigen/bench/dense_solvers.cpp

@@ -38,8 +38,6 @@ void bench(int id, int rows, int size = Size)
     A = A*A.adjoint();
   BenchTimer t_llt, t_ldlt, t_lu, t_fplu, t_qr, t_cpqr, t_cod, t_fpqr, t_jsvd, t_bdcsvd;
 
-  int svd_opt = ComputeThinU|ComputeThinV;
-  
   int tries = 5;
   int rep = 1000/size;
   if(rep==0) rep = 1;
@@ -53,8 +51,8 @@ void bench(int id, int rows, int size = Size)
   ColPivHouseholderQR<Mat> cpqr(A.rows(),A.cols());
   CompleteOrthogonalDecomposition<Mat> cod(A.rows(),A.cols());
   FullPivHouseholderQR<Mat> fpqr(A.rows(),A.cols());
-  JacobiSVD<MatDyn> jsvd(A.rows(),A.cols());
-  BDCSVD<MatDyn> bdcsvd(A.rows(),A.cols());
+  JacobiSVD<MatDyn, ComputeThinU|ComputeThinV> jsvd(A.rows(),A.cols());
+  BDCSVD<MatDyn, ComputeThinU|ComputeThinV> bdcsvd(A.rows(),A.cols());
   
   BENCH(t_llt, tries, rep, compute_norm_equation(llt,A));
   BENCH(t_ldlt, tries, rep, compute_norm_equation(ldlt,A));
@@ -67,9 +65,9 @@ void bench(int id, int rows, int size = Size)
   if(size*rows<=10000000)
     BENCH(t_fpqr, tries, rep, compute(fpqr,A));
   if(size<500) // JacobiSVD is really too slow for too large matrices
-    BENCH(t_jsvd, tries, rep, jsvd.compute(A,svd_opt));
+    BENCH(t_jsvd, tries, rep, jsvd.compute(A));
 //   if(size*rows<=20000000)
-    BENCH(t_bdcsvd, tries, rep, bdcsvd.compute(A,svd_opt));
+    BENCH(t_bdcsvd, tries, rep, bdcsvd.compute(A));
   
   results["LLT"][id] = t_llt.best();
   results["LDLT"][id] = t_ldlt.best();

libs/eigen/bench/spbench/CMakeLists.txt

@@ -1,7 +1,7 @@
 
 
-set(BLAS_FOUND TRUE)
-set(LAPACK_FOUND TRUE)
+set(BLAS_FOUND EIGEN_BUILD_BLAS)
+set(LAPACK_FOUND EIGEN_BUILD_LAPACK)
 set(BLAS_LIBRARIES eigen_blas_static)
 set(LAPACK_LIBRARIES eigen_lapack_static)
 

libs/eigen/bench/tensors/tensor_benchmarks.h

@@ -219,14 +219,8 @@ template <typename Device, typename T> class BenchmarkSuite {
     size_b[1] = m_;
     TensorMap<Tensor<T, 2>, Eigen::Aligned> B(b_, size_b);
 
-#if defined(EIGEN_HAS_INDEX_LIST)
     Eigen::IndexPairList<Eigen::type2indexpair<0, 0>,
                          Eigen::type2indexpair<2, 1> > paddings;
-#else
-    Eigen::array<Eigen::IndexPair<TensorIndex>, 2> paddings;
-    paddings[0] = Eigen::IndexPair<TensorIndex>(0, 0);
-    paddings[1] = Eigen::IndexPair<TensorIndex>(2, 1);
-#endif
 #ifdef EIGEN_USE_SYCL // warmup for sycl
     for (int iter = 0; iter < 10; ++iter) {
       B.device(device_) = A.pad(paddings);
@@ -251,15 +245,7 @@ template <typename Device, typename T> class BenchmarkSuite {
     size_b[1] = k_/2;
     TensorMap<Tensor<T, 2>, Eigen::Aligned> B(b_, size_b);
 
-#ifndef EIGEN_HAS_INDEX_LIST
-    Eigen::array<TensorIndex, 2> strides;
-    strides[0] = 1;
-    strides[1] = 2;
-#else
-    // Take advantage of cxx11 to give the compiler information it can use to
-    // optimize the code.
     Eigen::IndexList<Eigen::type2index<1>, Eigen::type2index<2> > strides;
-#endif
 
 #ifdef EIGEN_USE_SYCL // warmup for sycl
     for (int iter = 0; iter < 10; ++iter) {
@@ -284,17 +270,8 @@ template <typename Device, typename T> class BenchmarkSuite {
     size_c[0] = m_;
     size_c[1] = n_;
     TensorMap<Tensor<T, 2>, Eigen::Aligned> C(c_, size_c);
-
-#ifndef EIGEN_HAS_INDEX_LIST
-    Eigen::array<int, 2> broadcast;
-    broadcast[0] = 1;
-    broadcast[1] = n_;
-#else
-    // Take advantage of cxx11 to give the compiler information it can use to
-    // optimize the code.
     Eigen::IndexList<Eigen::type2index<1>, int> broadcast;
     broadcast.set(1, n_);
-#endif
 
 #ifdef EIGEN_USE_SYCL // warmup for sycl
     for (int iter = 0; iter < 10; ++iter) {
@@ -385,15 +362,7 @@ for (int iter = 0; iter < 10; ++iter) {
     Eigen::array<TensorIndex, 1> output_size;
     output_size[0] = n_;
     TensorMap<Tensor<T, 1, 0, TensorIndex>, Eigen::Aligned> C(c_, output_size);
-
-#ifndef EIGEN_HAS_INDEX_LIST
-    Eigen::array<TensorIndex, 1> sum_along_dim;
-    sum_along_dim[0] = 0;
-#else
-    // Take advantage of cxx11 to give the compiler information it can use to
-    // optimize the code.
     Eigen::IndexList<Eigen::type2index<0>> sum_along_dim;
-#endif
 #ifdef EIGEN_USE_SYCL // warmup for sycl
   for (int iter = 0; iter < 10; ++iter) {
     C.device(device_) = B.sum(sum_along_dim);
@@ -564,9 +533,9 @@ for (int iter = 0; iter < 10; ++iter) {
 
     // Initialize the content of the memory pools to prevent asan from
     // complaining.
-    device_.memset(a_, 12, m_ * k_ * sizeof(T));
-    device_.memset(b_, 23, k_ * n_ * sizeof(T));
-    device_.memset(c_, 31, m_ * n_ * sizeof(T));
+    device_.fill(a_, a_ + m_ * k_, T(12));
+    device_.fill(b_, b_ + k_ * n_, T(23));
+    device_.fill(c_, c_ + m_ * n_, T(31));
 
   }
 

libs/eigen/bench/tensors/tensor_contract_sycl_bench.cc

@@ -56,9 +56,9 @@ void contraction(const Device& device_, TensorIndex num_iters, TensorIndex m_, T
 
   // Initialize the content of the memory pools to prevent asan from
   // complaining.
-  device_.memset(a_, 12, m_ * k_ * sizeof(T));
-  device_.memset(b_, 23, k_ * n_ * sizeof(T));
-  device_.memset(c_, 31, m_ * n_ * sizeof(T));
+  device_.fill(a_, m_ * k_, T(12));
+  device_.fill(b_, k_ * n_, T(23));
+  device_.fill(c_, m_ * n_, T(31));
 
   Eigen::array<TensorIndex, 2> sizeA;
   sizeA[0] = m_;