diff --git a/openmp/libomptarget/DeviceRTL/include/Debug.h b/openmp/libomptarget/DeviceRTL/include/Debug.h
--- a/openmp/libomptarget/DeviceRTL/include/Debug.h
+++ b/openmp/libomptarget/DeviceRTL/include/Debug.h
@@ -12,16 +12,30 @@
 #ifndef OMPTARGET_DEVICERTL_DEBUG_H
 #define OMPTARGET_DEVICERTL_DEBUG_H
 
+#include "Configuration.h"
+#include "Utils.h"
+
 /// Assertion
 ///
 /// {
 extern "C" {
-void __assert_assume(bool cond, const char *exp, const char *file, int line);
+void __assert_assume(bool cond, bool enabled, bool print, const char *exp,
+                     const char *file, int line);
 void __assert_fail(const char *assertion, const char *file, unsigned line,
                    const char *function);
 }
 
-#define ASSERT(e) __assert_assume(e, #e, __FILE__, __LINE__)
+/// Helper expansion for ASSERT. \p Cond is the expression in the assertion,
+/// \p Enabled determines if assertions are enabled at all, \p ID is a unique
+/// identifier for the assertion so we can ensure it is only reported once.
+#define ASSERT_IMPL(Cond, Enabled)                                             \
+  __assert_assume(Cond, Enabled, Enabled &&utils::SingletonFlag::testAndSet(), \
+                  #Cond, __FILE__, __LINE__)
+
+/// Assert \p Cond holds. If assertions are enabled it will check it, otherwise
+/// simply assume it holds.
+#define ASSERT(Cond)                                                           \
+  ASSERT_IMPL(Cond, config::isDebugMode(config::DebugKind::Assertion))
 
 ///}
 
diff --git a/openmp/libomptarget/DeviceRTL/include/Synchronization.h b/openmp/libomptarget/DeviceRTL/include/Synchronization.h
--- a/openmp/libomptarget/DeviceRTL/include/Synchronization.h
+++ b/openmp/libomptarget/DeviceRTL/include/Synchronization.h
@@ -74,6 +74,10 @@
 /// Atomically add \p V to \p *Addr with \p Ordering semantics.
 uint64_t add(uint64_t *Addr, uint64_t V, int Ordering);
 
+/// Atomically write \p V to \p *Addr with \p Ordering semantics and return the
+/// old value of \p *Addr.
+uint32_t exchange(uint32_t *Addr, uint32_t V, int Ordering);
+
 } // namespace atomic
 
 } // namespace _OMP
diff --git a/openmp/libomptarget/DeviceRTL/include/Utils.h b/openmp/libomptarget/DeviceRTL/include/Utils.h
--- a/openmp/libomptarget/DeviceRTL/include/Utils.h
+++ b/openmp/libomptarget/DeviceRTL/include/Utils.h
@@ -12,6 +12,7 @@
 #ifndef OMPTARGET_DEVICERTL_UTILS_H
 #define OMPTARGET_DEVICERTL_UTILS_H
 
+#include "Synchronization.h"
 #include "Types.h"
 
 namespace _OMP {
@@ -72,6 +73,24 @@
   return V - V % Align;
 }
 
+/// Helper class to perform an action only once.
+///
+/// Using this is probably costly even if it is not executed. It should be
+/// guarded such that release mode execution will not be impacted.
+template <uint32_t ID> struct SingletonFlagImpl {
+
+  /// Each SingletonFlag instantiation with the same ID has an internal flag.
+  /// This function will return true if the flag was not set before, otherwise
+  /// it will return false. In either case the flag is set afterwards.
+  static bool testAndSet() {
+    static uint32_t DoOnceFlag = 0;
+    return 1 != atomic::exchange(&DoOnceFlag, 1, __ATOMIC_ACQ_REL);
+  }
+};
+
+/// Helper to hide the __COUNTER__ use away.
+#define SingletonFlag SingletonFlagImpl<__COUNTER__>
+
 #define OMP_LIKELY(EXPR) __builtin_expect((bool)(EXPR), true)
 #define OMP_UNLIKELY(EXPR) __builtin_expect((bool)(EXPR), false)
 
diff --git a/openmp/libomptarget/DeviceRTL/src/Debug.cpp b/openmp/libomptarget/DeviceRTL/src/Debug.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Debug.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Debug.cpp
@@ -20,9 +20,12 @@
 #pragma omp declare target
 
 extern "C" {
-void __assert_assume(bool cond, const char *exp, const char *file, int line) {
-  if (!cond && config::isDebugMode(config::DebugKind::Assertion)) {
-    PRINTF("ASSERTION failed: %s at %s, line %d\n", exp, file, line);
+void __assert_assume(bool cond, bool enabled, bool print, const char *exp,
+                     const char *file, int line) {
+  // TODO: Consider printing the thread coordinates as well.
+  if (enabled && !cond) {
+    if (print)
+      PRINTF("ASSERTION failed: %s at %s, line %d\n", exp, file, line);
     __builtin_trap();
   }
 
diff --git a/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp b/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
@@ -259,6 +259,10 @@
   return impl::atomicAdd(Addr, V, Ordering);
 }
 
+uint32_t atomic::exchange(uint32_t *Addr, uint32_t V, int Ordering) {
+  return impl::atomicExchange(Addr, V, Ordering);
+}
+
 extern "C" {
 void __kmpc_ordered(IdentTy *Loc, int32_t TId) {}