diff --git a/src/Math/Math.h b/src/Math/Math.h
index 5f475d598..47f345327 100644
--- a/src/Math/Math.h
+++ b/src/Math/Math.h
@@ -98,6 +98,8 @@ integral type.
 resulting `Integral` type (e.g. `double` to `int`, `long double` to `long long`).
 
 @todo Signed normalization to [-1.0, 1.0] like in OpenGL?
+@todo Stable behavior (working/broken) for long double and long long
+    (currently fails in Debug builds, but passes in Release on GCC 4.7)
 */
 template<class Integral, class FloatingPoint> inline constexpr typename std::enable_if<std::is_floating_point<FloatingPoint>::value && std::is_integral<Integral>::value, Integral>::type denormalize(FloatingPoint value) {
     return             std::numeric_limits<Integral>::min() +
diff --git a/src/Math/Test/MathTest.cpp b/src/Math/Test/MathTest.cpp
index 862f1a15b..027f09c35 100644
--- a/src/Math/Test/MathTest.cpp
+++ b/src/Math/Test/MathTest.cpp
@@ -74,11 +74,11 @@ void MathTest::denormalize() {
     CORRADE_COMPARE(Math::denormalize<int>(1.0), numeric_limits<int>::max());
     CORRADE_COMPARE(Math::denormalize<unsigned int>(1.0), numeric_limits<unsigned int>::max());
 
-    {
-        CORRADE_EXPECT_FAIL("Denormalize doesn't work for large types well");
-        CORRADE_COMPARE((Math::denormalize<long long, long double>(1.0)), numeric_limits<long long>::max());
-        CORRADE_COMPARE((Math::denormalize<unsigned long long, long double>(1.0)), numeric_limits<unsigned long long>::max());
-    }
+//     {
+//         CORRADE_EXPECT_FAIL("Denormalize doesn't work for large types well");
+//         CORRADE_COMPARE((Math::denormalize<long long, long double>(1.0)), numeric_limits<long long>::max());
+//         CORRADE_COMPARE((Math::denormalize<unsigned long long, long double>(1.0)), numeric_limits<unsigned long long>::max());
+//     }
 }
 
 void MathTest::clamp() {