Technical debt

dpctl/apis/include/dpctl4pybind11.hpp

@@ -27,9 +27,9 @@
 
 #include "dpctl_capi.h"
 #include <complex>
-#include <exception>
 #include <memory>
 #include <pybind11/pybind11.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>

dpctl/tensor/libtensor/include/kernels/accumulators.hpp

@@ -26,6 +26,7 @@
 #include <array>
 #include <cstdint>
 #include <limits>
+#include <new>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>

dpctl/tensor/libtensor/include/kernels/copy_as_contiguous.hpp

@@ -173,6 +173,7 @@ as_c_contiguous_array_generic_impl(sycl::queue &exec_q,
 
     sycl::event copy_ev = exec_q.submit([&](sycl::handler &cgh) {
         cgh.depends_on(depends);
+        cgh.use_kernel_bundle(kb);
 
         const sycl::range<1> gRange{n_groups * lws};
         const sycl::range<1> lRange{lws};

dpctl/tensor/libtensor/include/kernels/elementwise_functions/common.hpp

@@ -25,6 +25,7 @@
 #pragma once
 #include <cstddef>
 #include <cstdint>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 

dpctl/tensor/libtensor/include/kernels/elementwise_functions/common_inplace.hpp

@@ -26,6 +26,7 @@
 #pragma once
 #include <cstddef>
 #include <cstdint>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 
 #include "kernels/alignment.hpp"

dpctl/tensor/libtensor/include/kernels/linalg_functions/dot_product.hpp

@@ -27,6 +27,7 @@
 #include <complex>
 #include <cstddef>
 #include <cstdint>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <type_traits>
 #include <utility>

dpctl/tensor/libtensor/include/kernels/linalg_functions/gemm.hpp

@@ -27,6 +27,7 @@
 #include <complex>
 #include <cstddef>
 #include <cstdint>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <type_traits>
 #include <utility>
@@ -1349,6 +1350,8 @@ sycl::event _gemm_batch_nm_impl(sycl::queue &exec_q,
     sycl::event gemm_ev = exec_q.submit([&](sycl::handler &cgh) {
         cgh.depends_on(depends);
 
+        cgh.use_kernel_bundle(kb);
+
         using LocAccT1 = sycl::local_accessor<resTy, 1>;
         LocAccT1 local_A_block(wg_delta_n * wi_delta_n * wi_delta_k, cgh);
 

dpctl/tensor/libtensor/include/kernels/reductions.hpp

@@ -26,6 +26,7 @@
 #include <complex>
 #include <cstddef>
 #include <cstdint>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <type_traits>
 #include <utility>

dpctl/tensor/libtensor/include/utils/output_validation.hpp

@@ -25,6 +25,8 @@
 //===----------------------------------------------------------------------===//
 
 #pragma once
+#include <stdexcept>
+
 #include "dpctl4pybind11.hpp"
 #include <pybind11/pybind11.h>
 

dpctl/tensor/libtensor/include/utils/type_dispatch.hpp

@@ -23,6 +23,7 @@
 //===----------------------------------------------------------------------===//
 
 #pragma once
+#include <stdexcept>
 
 #include "dpctl4pybind11.hpp"
 #include "type_dispatch_building.hpp"

dpctl/tensor/libtensor/include/utils/type_utils.hpp

@@ -24,7 +24,7 @@
 
 #pragma once
 #include <complex>
-#include <exception>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 

dpctl/tensor/libtensor/source/accumulators.cpp

@@ -22,15 +22,17 @@
 /// This file defines functions of dpctl.tensor._tensor_impl extensions
 //===----------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <cstdint>
 #include <limits>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "kernels/accumulators.hpp"
 #include "simplify_iteration_space.hpp"
 #include "utils/memory_overlap.hpp"

dpctl/tensor/libtensor/source/accumulators/accumulate_over_axis.hpp

@@ -24,16 +24,18 @@
 
 #pragma once
 
-#include "dpctl4pybind11.hpp"
 #include <cstdint>
 #include <limits>
-#include <pybind11/numpy.h>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/numpy.h>
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "kernels/accumulators.hpp"
 #include "simplify_iteration_space.hpp"
 #include "utils/memory_overlap.hpp"

dpctl/tensor/libtensor/source/boolean_advanced_indexing.cpp

@@ -23,15 +23,17 @@
 /// dpctl.tensor.extract, dpctl.tensor.nonzero
 //===----------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <cstdint>
 #include <limits>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "boolean_advanced_indexing.hpp"
 #include "kernels/boolean_advanced_indexing.hpp"
 #include "simplify_iteration_space.hpp"

dpctl/tensor/libtensor/source/clip.cpp

@@ -23,14 +23,16 @@
 /// dpctl.tensor.clip
 //===----------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <complex>
 #include <cstdint>
+#include <stdexcept>
+#include <sycl/sycl.hpp>
+#include <utility>
+
+#include "dpctl4pybind11.hpp"
 #include <pybind11/complex.h>
 #include <pybind11/pybind11.h>
 #include <pybind11/stl.h>
-#include <sycl/sycl.hpp>
-#include <utility>
 
 #include "clip.hpp"
 #include "kernels/clip.hpp"

dpctl/tensor/libtensor/source/copy_and_cast_usm_to_usm.cpp

@@ -25,16 +25,17 @@
 #include <algorithm>
 #include <complex>
 #include <cstdint>
-#include <pybind11/complex.h>
-#include <pybind11/numpy.h>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <thread>
 #include <type_traits>
 #include <utility>
 
 #include "dpctl4pybind11.hpp"
+#include <pybind11/complex.h>
+#include <pybind11/numpy.h>
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
 
 #include "kernels/copy_and_cast.hpp"
 #include "utils/memory_overlap.hpp"

dpctl/tensor/libtensor/source/copy_for_reshape.cpp

@@ -22,18 +22,20 @@
 /// This file defines functions of dpctl.tensor._tensor_impl extensions
 //===----------------------------------------------------------------------===//
 
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
-#include "copy_for_reshape.hpp"
 #include "dpctl4pybind11.hpp"
+#include <pybind11/pybind11.h>
+
+#include "copy_for_reshape.hpp"
 #include "kernels/copy_and_cast.hpp"
 #include "utils/offset_utils.hpp"
 #include "utils/output_validation.hpp"
 #include "utils/sycl_alloc_utils.hpp"
 #include "utils/type_dispatch.hpp"
-#include <pybind11/pybind11.h>
 
 namespace dpctl
 {

dpctl/tensor/libtensor/source/copy_for_roll.cpp

@@ -22,18 +22,20 @@
 /// This file defines functions of dpctl.tensor._tensor_impl extensions
 //===----------------------------------------------------------------------===//
 
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
-#include "copy_for_roll.hpp"
 #include "dpctl4pybind11.hpp"
+#include <pybind11/pybind11.h>
+
+#include "copy_for_roll.hpp"
 #include "kernels/copy_and_cast.hpp"
 #include "utils/offset_utils.hpp"
 #include "utils/output_validation.hpp"
 #include "utils/sycl_alloc_utils.hpp"
 #include "utils/type_dispatch.hpp"
-#include <pybind11/pybind11.h>
 
 #include "simplify_iteration_space.hpp"
 

dpctl/tensor/libtensor/source/elementwise_functions/elementwise_functions.hpp

@@ -24,13 +24,15 @@
 //===----------------------------------------------------------------------===//
 #pragma once
 
+#include <stdexcept>
+#include <sycl/sycl.hpp>
+#include <utility>
+#include <vector>
+
 #include "dpctl4pybind11.hpp"
 #include <pybind11/numpy.h>
 #include <pybind11/pybind11.h>
 #include <pybind11/stl.h>
-#include <sycl/sycl.hpp>
-#include <utility>
-#include <vector>
 
 #include "elementwise_functions_type_utils.hpp"
 #include "kernels/alignment.hpp"

dpctl/tensor/libtensor/source/elementwise_functions/true_divide.cpp

@@ -23,16 +23,18 @@
 /// specifically functions for elementwise operations.
 //===----------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <complex>
 #include <cstdint>
-#include <pybind11/numpy.h>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/numpy.h>
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "elementwise_functions.hpp"
 #include "simplify_iteration_space.hpp"
 #include "true_divide.hpp"

dpctl/tensor/libtensor/source/full_ctor.cpp

@@ -22,14 +22,16 @@
 /// This file defines functions of dpctl.tensor._tensor_impl extensions
 //===--------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <complex>
-#include <pybind11/complex.h>
-#include <pybind11/pybind11.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/complex.h>
+#include <pybind11/pybind11.h>
+
 #include "kernels/constructors.hpp"
 #include "utils/output_validation.hpp"
 #include "utils/type_dispatch.hpp"

dpctl/tensor/libtensor/source/integer_advanced_indexing.cpp

@@ -26,13 +26,15 @@
 #include <algorithm>
 #include <complex>
 #include <cstdint>
-#include <pybind11/complex.h>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 
 #include "dpctl4pybind11.hpp"
+#include <pybind11/complex.h>
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "kernels/integer_advanced_indexing.hpp"
 #include "utils/memory_overlap.hpp"
 #include "utils/offset_utils.hpp"

dpctl/tensor/libtensor/source/linalg_functions/dot.cpp

@@ -22,15 +22,17 @@
 /// This file defines functions of dpctl.tensor._tensor_impl extensions
 //===--------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <cstdint>
-#include <pybind11/numpy.h>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/numpy.h>
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "dot.hpp"
 #include "dot_atomic_support.hpp"
 #include "dot_dispatch.hpp"

dpctl/tensor/libtensor/source/reductions/reduction_over_axis.hpp

@@ -27,6 +27,7 @@
 
 #include <algorithm>
 #include <cstdint>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <type_traits>
 #include <utility>

dpctl/tensor/libtensor/source/repeat.cpp

@@ -22,15 +22,17 @@
 /// This file defines functions of dpctl.tensor._tensor_impl extensions
 //===--------------------------------------------------------------------===//
 
-#include "dpctl4pybind11.hpp"
 #include <cstdint>
 #include <limits>
-#include <pybind11/pybind11.h>
-#include <pybind11/stl.h>
+#include <stdexcept>
 #include <sycl/sycl.hpp>
 #include <utility>
 #include <vector>
 
+#include "dpctl4pybind11.hpp"
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+
 #include "kernels/repeat.hpp"
 #include "simplify_iteration_space.hpp"
 #include "utils/memory_overlap.hpp"