[SYCL][E2E] Use fine-grained includes in Matrix/* tests (#13845)

Author: aelovikov-intel

sycl/test-e2e/Matrix/elemwise_irreg_size_ops_bf16.cpp

@@ -15,7 +15,8 @@
 // RUN: %{run} %t.out
 
 #include <iostream>
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 using namespace sycl;
 using namespace sycl::ext::oneapi::experimental::matrix;

sycl/test-e2e/Matrix/joint_matrix_gemm_cuda.hpp

@@ -12,7 +12,8 @@
 // supported for the Nvidia matrix extension, although some JIT optimizations
 // are performed at the level of the PTX assembly code.
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 using namespace sycl;
 using namespace sycl::ext::oneapi;

sycl/test-e2e/Matrix/joint_matrix_hip_apply.hpp

@@ -6,7 +6,8 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 #include <cstddef>
 #include <cstdint>

sycl/test-e2e/Matrix/joint_matrix_hip_copy.hpp

@@ -6,7 +6,8 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 #include <cstddef>
 #include <cstdint>

sycl/test-e2e/Matrix/joint_matrix_hip_fill.hpp

@@ -6,7 +6,8 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 #include <cstddef>
 #include <cstdint>

sycl/test-e2e/Matrix/joint_matrix_hip_mfma.hpp

@@ -6,7 +6,8 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 #include <cstddef>
 #include <cstdint>

sycl/test-e2e/Matrix/joint_matrix_query_default.cpp

@@ -13,7 +13,8 @@
 // RUN: %{run} %t.out
 
 #include <iostream>
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 using namespace sycl;
 using namespace sycl::ext::oneapi::experimental::matrix;

sycl/test-e2e/Matrix/runtime_query_hip_gfx90a.cpp

@@ -10,7 +10,8 @@
 // RUN: %{build} -Xsycl-target-backend=amdgcn-amd-amdhsa --offload-arch=gfx90a -o %t.out
 // RUN: %{run} %t.out
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 using namespace sycl::ext::oneapi::experimental::matrix;
 

sycl/test-e2e/Matrix/runtime_query_spr.cpp

@@ -10,7 +10,8 @@
 // RUN: %{build} -o %t.out
 // RUN: %{run} %t.out
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 using namespace sycl::ext::oneapi::experimental::matrix;
 

sycl/test-e2e/Matrix/runtime_query_tensorcores.cpp

@@ -10,7 +10,8 @@
 // RUN: %{build} -o %t.out
 // RUN: %{run} %t.out
 
-#include <sycl/sycl.hpp>
+#include <sycl/detail/core.hpp>
+#include <sycl/ext/oneapi/matrix/matrix.hpp>
 
 using namespace sycl::ext::oneapi::experimental::matrix;