diff --git a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
index 7cb93f3c5b050..fc7f17003f99c 100644
--- a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
+++ b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
@@ -7733,9 +7733,26 @@ static void processFunctionInstantiation(Sema &S,
   FD->setInstantiationIsPending(false);
 }
 
+static std::unique_ptr<MangleContext>
+createCrossABIMangleContext(ASTContext &Ctx) {
+  // For offload compilation with cross-ABI scenarios (e.g., Microsoft host
+  // + Itanium device), use device mangling context to ensure consistent
+  // symbol name mangling between host and device compilations.
+  const TargetInfo *AuxTarget = Ctx.getAuxTargetInfo();
+  if (AuxTarget && Ctx.getTargetInfo().getCXXABI().isMicrosoft() &&
+      AuxTarget->getCXXABI().isItaniumFamily()) {
+    return std::unique_ptr<MangleContext>(
+        Ctx.createDeviceMangleContext(*AuxTarget));
+  }
+  // Same ABI or no offload target: use standard mangling
+  return std::unique_ptr<MangleContext>(Ctx.createMangleContext());
+}
+
 void Sema::PerformPendingInstantiations(bool LocalOnly, bool AtEndOfTU) {
+  // Use cross-ABI aware mangling context to handle offload scenarios where
+  // host and device may use different ABIs (e.g., Windows+CUDA, Windows+HIP).
   std::unique_ptr<MangleContext> MangleCtx(
-      getASTContext().createMangleContext());
+      createCrossABIMangleContext(getASTContext()));
   std::deque<PendingImplicitInstantiation> DelayedImplicitInstantiations;
   while (!PendingLocalImplicitInstantiations.empty() ||
          (!LocalOnly && !PendingInstantiations.empty())) {
diff --git a/sycl/test-e2e/Printf/mixed-address-space.cpp b/sycl/test-e2e/Printf/mixed-address-space.cpp
index 1e256f5619288..126891644535c 100644
--- a/sycl/test-e2e/Printf/mixed-address-space.cpp
+++ b/sycl/test-e2e/Printf/mixed-address-space.cpp
@@ -2,8 +2,6 @@
 // for constant and generic address space can be used in the same module.
 //
 // UNSUPPORTED: target-amd
-// XFAIL: cuda && windows
-// XFAIL-TRACKER: https://github.com/intel/llvm/issues/14733
 // FIXME: Drop the test once generic AS support is considered stable and the
 //        dedicated constant AS overload of printf is removed from the library.
 //
diff --git a/sycl/test-e2e/Printf/percent-symbol.cpp b/sycl/test-e2e/Printf/percent-symbol.cpp
index 364016220a62f..60f483f880e04 100644
--- a/sycl/test-e2e/Printf/percent-symbol.cpp
+++ b/sycl/test-e2e/Printf/percent-symbol.cpp
@@ -5,8 +5,6 @@
 // [1]: https://en.cppreference.com/w/cpp/io/c/fprintf
 //
 // UNSUPPORTED: target-amd
-// XFAIL: cuda && windows
-// XFAIL-TRACKER: https://github.com/intel/llvm/issues/14733
 // RUN: %{build} -o %t.out
 // RUN: %{run} %t.out | FileCheck %s
 // FIXME: Remove dedicated constant address space testing once generic AS
diff --git a/test_cross_abi_mangling.cpp b/test_cross_abi_mangling.cpp
new file mode 100644
index 0000000000000..bb4ada1ea0c08
--- /dev/null
+++ b/test_cross_abi_mangling.cpp
@@ -0,0 +1,19 @@
+// Test case for cross-ABI kernel name mangling (Windows+CUDA scenario)
+// This should compile and link correctly on Windows targeting CUDA
+
+#include <iostream>
+
+template <typename T> class KernelName;
+
+// Simulate SYCL kernel lambda - this will be instantiated
+template <typename T> void kernel_func() {
+  // Empty kernel body
+}
+
+int main() {
+  // Instantiate the kernel template
+  kernel_func<KernelName<int>>();
+
+  std::cout << "Cross-ABI mangling test passed!" << std::endl;
+  return 0;
+}