6 files changed, 110 insertions, 12 deletions
diff --git a/clang/test/CXX/lex/lex.literal/lex.ccon/p1.cpp b/clang/test/CXX/lex/lex.literal/lex.ccon/p1.cpp
index 7b65f7ee832..6df035d63f1 100644
--- a/clang/test/CXX/lex/lex.literal/lex.ccon/p1.cpp
+++ b/clang/test/CXX/lex/lex.literal/lex.ccon/p1.cpp
@@ -1,4 +1,5 @@
-// RUN: %clang_cc1 -fsyntax-only -verify %s
+// RUN: %clang_cc1 -std=c++0x -fsyntax-only -verify %s
+// Runs in c++0x mode so that char16_t and char32_t are available.
 
 // Check types of char literals
 extern char a;
@@ -7,3 +8,7 @@ extern int b;
 extern __typeof('asdf') b;
 extern wchar_t c;
 extern __typeof(L'a') c;
+extern char16_t d;
+extern __typeof(u'a') d;
+extern char32_t e;
+extern __typeof(U'a') e;
diff --git a/clang/test/CodeGen/char-literal.c b/clang/test/CodeGen/char-literal.c
index 322041c0049..014f6eb4fb0 100644
--- a/clang/test/CodeGen/char-literal.c
+++ b/clang/test/CodeGen/char-literal.c
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -x c++ -triple i386-unknown-unknown -emit-llvm %s -o - | FileCheck %s
-// Runs in c++ mode so that wchar_t is available.
+// RUN: %clang_cc1 -x c++ -std=c++0x -triple i386-unknown-unknown -emit-llvm %s -o - | FileCheck %s
+// Runs in c++0x mode so that wchar_t, char16_t, and char32_t are available.
 
 int main() {
   // CHECK: store i8 97
@@ -16,6 +16,20 @@ int main() {
   // CHECK: store i32 98
   wchar_t wb = L'ab';
 
+  // CHECK: store i16 97
+  char16_t ua = u'a';
+
+  // Should pick second character.
+  // CHECK: store i16 98
+  char16_t ub = u'ab';
+
+  // CHECK: store i32 97
+  char32_t Ua = U'a';
+
+  // Should pick second character.
+  // CHECK: store i32 98
+  char32_t Ub = U'ab';
+
   // Should pick last character and store its lowest byte.
   // This does not match gcc, which takes the last character, converts it to
   // utf8, and then picks the second-lowest byte of that (they probably store
@@ -26,10 +40,36 @@ int main() {
   // CHECK: store i32 61451
   wchar_t wc = L'\uF00B';
 
+  // -4085 == 0xf00b
+  // CHECK: store i16 -4085
+  char16_t uc = u'\uF00B';
+
+  // CHECK: store i32 61451
+  char32_t Uc = U'\uF00B';
+
   // CHECK: store i32 1110027
   wchar_t wd = L'\U0010F00B';
 
+  // Should take lower word of the 4byte UNC sequence. This does not match
+  // gcc. I don't understand what gcc does (it looks like it converts to utf16,
+  // then takes the second (!) utf16 word, swaps the lower two nibbles, and
+  // stores that?).
+  // CHECK: store i16 -4085
+  char16_t ud = u'\U0010F00B';  // has utf16 encoding dbc8 dcb0
+
+  // CHECK: store i32 1110027
+  char32_t Ud = U'\U0010F00B';
+
   // Should pick second character.
   // CHECK: store i32 1110027
   wchar_t we = L'\u1234\U0010F00B';
+
+  // Should pick second character.
+  // CHECK: store i16 -4085
+  char16_t ue = u'\u1234\U0010F00B';
+
+  // Should pick second character.
+  // CHECK: store i32 1110027
+  char32_t Ue = U'\u1234\U0010F00B';
+
 }
diff --git a/clang/test/CodeGen/string-literal.c b/clang/test/CodeGen/string-literal.c
index cc6c0943d95..6d14330a0b9 100644
--- a/clang/test/CodeGen/string-literal.c
+++ b/clang/test/CodeGen/string-literal.c
@@ -1,4 +1,5 @@
-// RUN: %clang_cc1 -triple i386-unknown-unknown -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -x c++ -std=c++0x -triple i386-unknown-unknown -emit-llvm %s -o - | FileCheck %s
+// Runs in c++0x mode so that wchar_t, char16_t, and char32_t are available.
 
 int main() {
   // CHECK: internal unnamed_addr constant [10 x i8] c"abc\00\00\00\00\00\00\00", align 1
@@ -9,8 +10,24 @@ int main() {
   char b[10] = "\u1120\u0220\U00102030";
 
   // CHECK: private unnamed_addr constant [12 x i8] c"A\00\00\00B\00\00\00\00\00\00\00", align 1
-  void *foo = L"AB";
+  const wchar_t *foo = L"AB";
 
   // CHECK: private unnamed_addr constant [12 x i8] c"4\12\00\00\0B\F0\10\00\00\00\00\00", align 1
-  void *bar = L"\u1234\U0010F00B";
+  const wchar_t *bar = L"\u1234\U0010F00B";
+
+  // CHECK: private unnamed_addr constant [12 x i8] c"C\00\00\00D\00\00\00\00\00\00\00", align 1
+  const char32_t *c = U"CD";
+
+  // CHECK: private unnamed_addr constant [12 x i8] c"5\12\00\00\0C\F0\10\00\00\00\00\00", align 1
+  const char32_t *d = U"\u1235\U0010F00C";
+
+  // CHECK: private unnamed_addr constant [6 x i8] c"E\00F\00\00\00", align 1
+  const char16_t *e = u"EF";
+
+  // This should convert to utf16.
+  // CHECK: private unnamed_addr constant [10 x i8] c" \11 \02\C8\DB0\DC\00\00", align 1
+  const char16_t *f = u"\u1120\u0220\U00102030";
+
+  // CHECK: private unnamed_addr constant [4 x i8] c"def\00", align 1
+  const char *g = u8"def";
 }
diff --git a/clang/test/Lexer/wchar.c b/clang/test/Lexer/wchar.c
index ac82c1f73b4..648a38ef3f9 100644
--- a/clang/test/Lexer/wchar.c
+++ b/clang/test/Lexer/wchar.c
@@ -5,8 +5,8 @@ void f() {
 
   (void)L'\U00010000';  // expected-warning {{character unicode escape sequence too long for its type}}
 
-  (void)L'ab';  // expected-warning {{extraneous characters in wide character constant ignored}}
+  (void)L'ab';  // expected-warning {{extraneous characters in character constant ignored}}
 
-  (void)L'a\u1000';  // expected-warning {{extraneous characters in wide character constant ignored}}
+  (void)L'a\u1000';  // expected-warning {{extraneous characters in character constant ignored}}
 }
 
diff --git a/clang/test/Parser/char-literal-printing.c b/clang/test/Parser/char-literal-printing.c
index 5843e5f4015..a0cafd64410 100644
--- a/clang/test/Parser/char-literal-printing.c
+++ b/clang/test/Parser/char-literal-printing.c
@@ -1,6 +1,5 @@
-// RUN: %clang_cc1 -ast-print %s
-
-#include <stddef.h>
+// RUN: %clang_cc1 -x c++ -std=c++0x -ast-print %s
+// Runs in c++0x mode so that wchar_t, char16_t, and char32_t are available.
 
 char    test1(void) { return '\\'; }
 wchar_t test2(void) { return L'\\'; }
@@ -29,3 +28,35 @@ char    test23(void) { return '\x3'; }
 wchar_t test24(void) { return L'\x3'; }
 
 wchar_t test25(void) { return L'\x333'; }
+
+char16_t test26(void) { return u'\\'; }
+char16_t test27(void) { return u'\''; }
+char16_t test28(void) { return u'\a'; }
+char16_t test29(void) { return u'\b'; }
+char16_t test30(void) { return u'\e'; }
+char16_t test31(void) { return u'\f'; }
+char16_t test32(void) { return u'\n'; }
+char16_t test33(void) { return u'\r'; }
+char16_t test34(void) { return u'\t'; }
+char16_t test35(void) { return u'\v'; }
+
+char16_t test36(void) { return u'c'; }
+char16_t test37(void) { return u'\x3'; }
+
+char16_t test38(void) { return u'\x333'; }
+
+char32_t test39(void) { return U'\\'; }
+char32_t test40(void) { return U'\''; }
+char32_t test41(void) { return U'\a'; }
+char32_t test42(void) { return U'\b'; }
+char32_t test43(void) { return U'\e'; }
+char32_t test44(void) { return U'\f'; }
+char32_t test45(void) { return U'\n'; }
+char32_t test46(void) { return U'\r'; }
+char32_t test47(void) { return U'\t'; }
+char32_t test48(void) { return U'\v'; }
+
+char32_t test49(void) { return U'c'; }
+char32_t test50(void) { return U'\x3'; }
+
+char32_t test51(void) { return U'\x333'; }
diff --git a/clang/test/SemaCXX/type-convert-construct.cpp b/clang/test/SemaCXX/type-convert-construct.cpp
index 479af21476b..a367633e485 100644
--- a/clang/test/SemaCXX/type-convert-construct.cpp
+++ b/clang/test/SemaCXX/type-convert-construct.cpp
@@ -1,4 +1,5 @@
-// RUN: %clang_cc1 -fsyntax-only -verify %s 
+// RUN: %clang_cc1 -std=gnu++0x -fsyntax-only -verify %s 
+// Runs in c++0x mode so that char16_t and char32_t are available.
 
 void f() {
   float v1 = float(1);
@@ -14,4 +15,8 @@ void f() {
   str = "a string"; // expected-warning{{conversion from string literal to 'char *' is deprecated}}
   wchar_t *wstr;
   wstr = L"a wide string"; // expected-warning{{conversion from string literal to 'wchar_t *' is deprecated}}
+  char16_t *ustr;
+  ustr = u"a UTF-16 string"; // expected-error {{assigning to 'char16_t *' from incompatible type 'const char16_t [16]'}}
+  char32_t *Ustr;
+  Ustr = U"a UTF-32 string"; // expected-error {{assigning to 'char32_t *' from incompatible type 'const char32_t [16]'}}
 }