[SYCL] Fix kernel name mangling for CUDA/HIP offload with shared libraries

SYCL kernel names are generated during template instantiation using
MangleContext. When using SYCL offload to CUDA/HIP, kernel names must
be consistent across compilation units, including shared libraries
compiled separately.

Previous approach only handled cross-ABI scenarios (Microsoft host +
Itanium device). This left same-ABI scenarios (Itanium host + Itanium
device) broken for shared libraries, as each compilation unit would
generate different lambda numbering, resulting in runtime errors:

  No kernel named _ZTSZZ21performIncrementationENK... was found

Solution:
For SYCL offload compilation, always use createDeviceMangleContext()
instead of createMangleContext(). Device mangling context uses
DeviceLambdaManglingNumbers which provides globally consistent lambda
numbering across compilation units.

This ensures identical kernel names for:
- Cross-ABI: Microsoft host + Itanium device (Windows + CUDA/HIP)
- Same-ABI: Itanium host + Itanium device (Linux + CUDA/HIP)
- Shared libraries compiled separately (.so/.dll)

Fixes CMPLRLLVM-69642
Fixes CMPLRLLVM-69415

Author: kekaczma

a.out

@@ -7734,18 +7734,19 @@ static void processFunctionInstantiation(Sema &S,
 }
 
 static std::unique_ptr<MangleContext> createSYCLMangleContext(ASTContext &Ctx) {
-  // For SYCL compilation, ensure kernel name mangling is consistent between
-  // host and device compilations when targeting CUDA/HIP backends.
-  // When the host uses Microsoft ABI and device uses Itanium ABI, we need
-  // to use device-aware mangling context on the host side.
+  // For SYCL offload compilation (CUDA/HIP), always use device mangling context
+  // to ensure consistent kernel name mangling across compilation units.
+  // Device lambda numbering is tracked globally via DeviceLambdaManglingNumbers,
+  // which provides stable numbering even for shared libraries compiled separately.
+  // This solves both cross-ABI scenarios (Microsoft host + Itanium device) and
+  // same-ABI scenarios (Itanium host + Itanium device) where separate compilation
+  // would otherwise produce different lambda numbers.
   const TargetInfo *AuxTarget = Ctx.getAuxTargetInfo();
-  if (AuxTarget && Ctx.getTargetInfo().getCXXABI().isMicrosoft() &&
-      AuxTarget->getCXXABI().isItaniumFamily()) {
-    // Host compilation with Microsoft ABI targeting Itanium device
+  if (AuxTarget) {
     return std::unique_ptr<MangleContext>(
         Ctx.createDeviceMangleContext(*AuxTarget));
   }
-  // All other cases: use standard mangling for the primary target
+  // No offload target: use standard mangling
   return std::unique_ptr<MangleContext>(Ctx.createMangleContext());
 }
 

clang/lib/Sema/SemaTemplateInstantiateDecl.cpp

@@ -1,6 +1,3 @@
-// UNSUPPORTED: cuda || hip
-// UNSUPPORTED-TRACKER: CMPLRLLVM-69415
-
 // DEFINE: %{fPIC_flag} =  %if windows %{%} %else %{-fPIC%}
 // DEFINE: %{shared_lib_ext} = %if windows %{dll%} %else %{so%}