diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/debug.h b/openmp/libomptarget/deviceRTLs/common/debug.h
rename from openmp/libomptarget/deviceRTLs/nvptx/src/debug.h
rename to openmp/libomptarget/deviceRTLs/common/debug.h
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/cancel.cu b/openmp/libomptarget/deviceRTLs/common/src/cancel.cu
rename from openmp/libomptarget/deviceRTLs/nvptx/src/cancel.cu
rename to openmp/libomptarget/deviceRTLs/common/src/cancel.cu
--- a/openmp/libomptarget/deviceRTLs/nvptx/src/cancel.cu
+++ b/openmp/libomptarget/deviceRTLs/common/src/cancel.cu
@@ -11,7 +11,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "interface.h"
-#include "debug.h"
+#include "common/debug.h"
 
 EXTERN int32_t __kmpc_cancellationpoint(kmp_Ident *loc, int32_t global_tid,
                                         int32_t cancelVal) {
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/critical.cu b/openmp/libomptarget/deviceRTLs/common/src/critical.cu
rename from openmp/libomptarget/deviceRTLs/nvptx/src/critical.cu
rename to openmp/libomptarget/deviceRTLs/common/src/critical.cu
--- a/openmp/libomptarget/deviceRTLs/nvptx/src/critical.cu
+++ b/openmp/libomptarget/deviceRTLs/common/src/critical.cu
@@ -11,7 +11,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "interface.h"
-#include "debug.h"
+#include "common/debug.h"
 
 EXTERN
 void __kmpc_critical(kmp_Ident *loc, int32_t global_tid,
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/loop.cu b/openmp/libomptarget/deviceRTLs/common/src/loop.cu
rename from openmp/libomptarget/deviceRTLs/nvptx/src/loop.cu
rename to openmp/libomptarget/deviceRTLs/common/src/loop.cu
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/task.cu b/openmp/libomptarget/deviceRTLs/common/src/task.cu
rename from openmp/libomptarget/deviceRTLs/nvptx/src/task.cu
rename to openmp/libomptarget/deviceRTLs/common/src/task.cu
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/state-queue.h b/openmp/libomptarget/deviceRTLs/common/state-queue.h
rename from openmp/libomptarget/deviceRTLs/nvptx/src/state-queue.h
rename to openmp/libomptarget/deviceRTLs/common/state-queue.h
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/state-queuei.h b/openmp/libomptarget/deviceRTLs/common/state-queuei.h
rename from openmp/libomptarget/deviceRTLs/nvptx/src/state-queuei.h
rename to openmp/libomptarget/deviceRTLs/common/state-queuei.h
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt b/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt
--- a/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt
+++ b/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt
@@ -38,6 +38,10 @@
 get_filename_component(devicertl_base_directory
   ${CMAKE_CURRENT_SOURCE_DIR}
   DIRECTORY)
+set(devicertl_common_directory
+  ${devicertl_base_directory}/common)
+set(devicertl_nvptx_directory
+  ${devicertl_base_directory}/nvptx)
 
 if(LIBOMPTARGET_DEP_CUDA_FOUND)
   libomptarget_say("Building CUDA offloading device RTL.")
@@ -47,17 +51,17 @@
   set(CUDA_PROPAGATE_HOST_FLAGS OFF)
 
   set(cuda_src_files
-      src/cancel.cu
-      src/critical.cu
+      ${devicertl_common_directory}/src/cancel.cu
+      ${devicertl_common_directory}/src/critical.cu
       src/data_sharing.cu
       src/libcall.cu
-      src/loop.cu
+      ${devicertl_common_directory}/src/loop.cu
       src/omptarget-nvptx.cu
       src/parallel.cu
       src/reduction.cu
       src/support.cu
       src/sync.cu
-      src/task.cu
+      ${devicertl_common_directory}/src/task.cu
   )
 
   set(omp_data_objects src/omp_data.cu)
@@ -88,7 +92,8 @@
   # yet supported by the CUDA toolchain on the device.
   set(BUILD_SHARED_LIBS OFF)
   set(CUDA_SEPARABLE_COMPILATION ON)
-  list(APPEND CUDA_NVCC_FLAGS -I${devicertl_base_directory})
+  list(APPEND CUDA_NVCC_FLAGS -I${devicertl_base_directory}
+                              -I${devicertl_nvptx_directory}/src)
   cuda_add_library(omptarget-nvptx STATIC ${cuda_src_files} ${omp_data_objects}
       OPTIONS ${CUDA_ARCH} ${CUDA_DEBUG})
 
@@ -123,7 +128,9 @@
 
     # Set flags for LLVM Bitcode compilation.
     set(bc_flags ${LIBOMPTARGET_NVPTX_SELECTED_CUDA_COMPILER_FLAGS}
-                 -I${devicertl_base_directory})
+                 -I${devicertl_base_directory}
+                 -I${devicertl_nvptx_directory}/src)
+
     if(${LIBOMPTARGET_NVPTX_DEBUG})
       set(bc_flags ${bc_flags} -DOMPTARGET_NVPTX_DEBUG=-1)
     else()
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/omptarget-nvptx.h b/openmp/libomptarget/deviceRTLs/nvptx/src/omptarget-nvptx.h
--- a/openmp/libomptarget/deviceRTLs/nvptx/src/omptarget-nvptx.h
+++ b/openmp/libomptarget/deviceRTLs/nvptx/src/omptarget-nvptx.h
@@ -21,9 +21,9 @@
 
 // local includes
 #include "target_impl.h"
-#include "debug.h"     // debug
+#include "common/debug.h"     // debug
 #include "interface.h" // interfaces with omp, compiler, and user
-#include "state-queue.h"
+#include "common/state-queue.h"
 #include "support.h"
 
 #define OMPTARGET_NVPTX_VERSION 1.1
diff --git a/openmp/libomptarget/deviceRTLs/nvptx/src/support.cu b/openmp/libomptarget/deviceRTLs/nvptx/src/support.cu
--- a/openmp/libomptarget/deviceRTLs/nvptx/src/support.cu
+++ b/openmp/libomptarget/deviceRTLs/nvptx/src/support.cu
@@ -11,7 +11,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "support.h"
-#include "debug.h"
+#include "common/debug.h"
 #include "omptarget-nvptx.h"
 
 ////////////////////////////////////////////////////////////////////////////////