CodeGen: Cast alloca to expected address space

Alloca always returns a pointer in alloca address space, which may
be different from the type defined by the language. For example,
in C++ the auto variables are in the default address space. Therefore
cast alloca to the expected address space when necessary.

Differential Revision: https://reviews.llvm.org/D32248

llvm-svn: 303370

1 files changed, 44 insertions, 20 deletions

diff --git a/clang/lib/Basic/Targets.cpp b/clang/lib/Basic/Targets.cpp
index 92c561aa941..1b9fbed1773 100644
--- a/clang/lib/Basic/Targets.cpp
+++ b/clang/lib/Basic/Targets.cpp
@@ -2034,25 +2034,45 @@ ArrayRef<const char *> NVPTXTargetInfo::getGCCRegNames() const {
   return llvm::makeArrayRef(GCCRegNames);
 }
 
-static const LangAS::Map AMDGPUPrivateIsZeroMap = {
-    4,  // Default
-    1,  // opencl_global
-    3,  // opencl_local
-    2,  // opencl_constant
-    4,  // opencl_generic
-    1,  // cuda_device
-    2,  // cuda_constant
-    3   // cuda_shared
+static const LangAS::Map AMDGPUNonOpenCLPrivateIsZeroMap = {
+    4, // Default
+    1, // opencl_global
+    3, // opencl_local
+    2, // opencl_constant
+    4, // opencl_generic
+    1, // cuda_device
+    2, // cuda_constant
+    3  // cuda_shared
+};
+static const LangAS::Map AMDGPUNonOpenCLGenericIsZeroMap = {
+    0, // Default
+    1, // opencl_global
+    3, // opencl_local
+    2, // opencl_constant
+    0, // opencl_generic
+    1, // cuda_device
+    2, // cuda_constant
+    3  // cuda_shared
 };
-static const LangAS::Map AMDGPUGenericIsZeroMap = {
-    0,  // Default
-    1,  // opencl_global
-    3,  // opencl_local
-    2,  // opencl_constant
-    0,  // opencl_generic
-    1,  // cuda_device
-    2,  // cuda_constant
-    3   // cuda_shared
+static const LangAS::Map AMDGPUOpenCLPrivateIsZeroMap = {
+    0, // Default
+    1, // opencl_global
+    3, // opencl_local
+    2, // opencl_constant
+    4, // opencl_generic
+    1, // cuda_device
+    2, // cuda_constant
+    3  // cuda_shared
+};
+static const LangAS::Map AMDGPUOpenCLGenericIsZeroMap = {
+    5, // Default
+    1, // opencl_global
+    3, // opencl_local
+    2, // opencl_constant
+    0, // opencl_generic
+    1, // cuda_device
+    2, // cuda_constant
+    3  // cuda_shared
 };
 
 // If you edit the description strings, make sure you update
@@ -2149,8 +2169,12 @@ public:
                     : DataLayoutStringR600);
     assert(DataLayout->getAllocaAddrSpace() == AS.Private);
 
-    AddrSpaceMap = IsGenericZero ? &AMDGPUGenericIsZeroMap :
-        &AMDGPUPrivateIsZeroMap;
+    AddrSpaceMap =
+        llvm::StringSwitch<const LangAS::Map *>(Triple.getEnvironmentName())
+            .Case("opencl", &AMDGPUOpenCLPrivateIsZeroMap)
+            .Case("amdgiz", &AMDGPUNonOpenCLGenericIsZeroMap)
+            .Case("amdgizcl", &AMDGPUOpenCLGenericIsZeroMap)
+            .Default(&AMDGPUNonOpenCLPrivateIsZeroMap);
     UseAddrSpaceMapMangling = true;
   }