Add target hook for setting symbol prefix and section of unicode

string literals.

llvm-svn: 68363

clang/include/clang/Basic/TargetInfo.h

@@ -257,6 +257,18 @@ public:
     return "";
   }
 
+  /// getUnicodeStringSymbolPrefix - Get the default symbol prefix to
+  /// use for string literals.
+  virtual const char *getUnicodeStringSymbolPrefix() const { 
+    return ".str";
+  }
+
+  /// getUnicodeStringSymbolPrefix - Get the default symbol prefix to
+  /// use for string literals.
+  virtual const char *getUnicodeStringSection() const { 
+    return 0;
+  }
+
   /// getCFStringSection - Return the section to use for the CFString
   /// literals, or 0 if no special section is used.
   virtual const char *getCFStringSection() const { 

clang/lib/Basic/Targets.cpp

@@ -658,6 +658,14 @@ public:
     return IsConstant ? "\01LC" : "\01lC";
   }
 
+  virtual const char *getUnicodeStringSymbolPrefix() const { 
+    return "__utf16_string_";
+  }
+
+  virtual const char *getUnicodeStringSection() const { 
+    return "__TEXT,__ustring";
+  }
+
   virtual const char *getCFStringSymbolPrefix() const { 
     return "\01LC";
   }
@@ -810,6 +818,14 @@ public:
     return IsConstant ? "\01LC" : "\01lC";
   }
 
+  virtual const char *getUnicodeStringSymbolPrefix() const { 
+    return "__utf16_string_";
+  }
+
+  virtual const char *getUnicodeStringSection() const { 
+    return "__TEXT,__ustring";
+  }
+
   virtual const char *getCFStringSymbolPrefix() const { 
     return "\01L_unnamed_cfstring_";
   }

clang/lib/CodeGen/CodeGenModule.cpp

@@ -1082,13 +1082,31 @@ GetAddrOfConstantCFString(const StringLiteral *Literal) {
   CurField = NextField;
   NextField = *Field++;
   llvm::Constant *C = llvm::ConstantArray::get(str);
+
+  const char *Sect, *Prefix;
+  bool isConstant;
+  if (isUTF16) {
+    Prefix = getContext().Target.getUnicodeStringSymbolPrefix();
+    Sect = getContext().Target.getUnicodeStringSection();
+    // FIXME: Why does GCC not set constant here?
+    isConstant = false;
+  } else {
+    Prefix = getContext().Target.getStringSymbolPrefix(true);
+    Sect = getContext().Target.getCFStringDataSection();
+    // FIXME: -fwritable-strings should probably affect this, but we
+    // are following gcc here.
+    isConstant = true;
+  }
   llvm::GlobalVariable *GV = 
-    new llvm::GlobalVariable(C->getType(), true, 
+    new llvm::GlobalVariable(C->getType(), isConstant, 
                              llvm::GlobalValue::InternalLinkage,
-                             C, getContext().Target.getStringSymbolPrefix(true),
-                             &getModule());
-  if (const char *Sect = getContext().Target.getCFStringDataSection())
+                             C, Prefix, &getModule());
+  if (Sect)
     GV->setSection(Sect);
+  if (isUTF16) {
+    unsigned Align = getContext().getTypeAlign(getContext().ShortTy)/8;
+    GV->setAlignment(Align); 
+  }
   appendFieldAndPadding(*this, Fields, CurField, NextField,
                         llvm::ConstantExpr::getGetElementPtr(GV, Zeros, 2),
                         CFRD, STy);

clang/test/CodeGen/darwin-string-literals.c

@@ -2,8 +2,9 @@
 
 // RUN: grep -F '@"\01LC" = internal constant [8 x i8] c"string0\00"' %t &&
 // RUN: grep -F '@"\01LC1" = internal constant [8 x i8] c"string1\00", section "__TEXT,__cstring,cstring_literals"' %t &&
-
+// RUN: grep -F '@__utf16_string_ = internal global [35 x i8] c"h\00e\00l\00l\00o\00 \00\92! \00\03& \00\90! \00w\00o\00r\00l\00d\00\00", section "__TEXT,__ustring", align 2' %t &&
 // RUN: true
 
 const char *g0 = "string0";
 const void *g1 = __builtin___CFStringMakeConstantString("string1");
+const void *g2 = __builtin___CFStringMakeConstantString("hello \u2192 \u2603 \u2190 world");