diff --git a/lib/include/__clang_cuda_complex_builtins.h b/lib/include/__clang_cuda_complex_builtins.h
index 8c10ff6b46..d924487ab2 100644
--- a/lib/include/__clang_cuda_complex_builtins.h
+++ b/lib/include/__clang_cuda_complex_builtins.h
@@ -16,7 +16,7 @@
 // to work with CUDA and OpenMP target offloading [in C and C++ mode].)
 
 #pragma push_macro("__DEVICE__")
-#ifdef _OPENMP
+#ifdef __OPENMP_NVPTX__
 #pragma omp declare target
 #define __DEVICE__ __attribute__((noinline, nothrow, cold, weak))
 #else
@@ -26,7 +26,7 @@
 // To make the algorithms available for C and C++ in CUDA and OpenMP we select
 // different but equivalent function versions. TODO: For OpenMP we currently
 // select the native builtins as the overload support for templates is lacking.
-#if !defined(_OPENMP)
+#if !defined(__OPENMP_NVPTX__)
 #define _ISNANd std::isnan
 #define _ISNANf std::isnan
 #define _ISINFd std::isinf
@@ -250,7 +250,7 @@ __DEVICE__ float _Complex __divsc3(float __a, float __b, float __c, float __d) {
 #undef _LOGBd
 #undef _LOGBf
 
-#ifdef _OPENMP
+#ifdef __OPENMP_NVPTX__
 #pragma omp end declare target
 #endif
 
diff --git a/lib/include/openmp_wrappers/complex b/lib/include/openmp_wrappers/complex
index 1ed0b14879..d8dcd41670 100644
--- a/lib/include/openmp_wrappers/complex
+++ b/lib/include/openmp_wrappers/complex
@@ -18,7 +18,9 @@
 #include <cmath>
 
 #define __CUDA__
+#define __OPENMP_NVPTX__
 #include <__clang_cuda_complex_builtins.h>
+#undef __OPENMP_NVPTX__
 #endif
 
 // Grab the host header too.
diff --git a/lib/include/openmp_wrappers/complex.h b/lib/include/openmp_wrappers/complex.h
index 829c7a7857..00d278548f 100644
--- a/lib/include/openmp_wrappers/complex.h
+++ b/lib/include/openmp_wrappers/complex.h
@@ -18,7 +18,9 @@
 #include <math.h>
 
 #define __CUDA__
+#define __OPENMP_NVPTX__
 #include <__clang_cuda_complex_builtins.h>
+#undef __OPENMP_NVPTX__
 #endif
 
 // Grab the host header too.