Vendor import of clang release_50 branch r310316:vendor/clang/clang-release_50-r310316

https://llvm.org/svn/llvm-project/cfe/branches/release_50@310316

27 files changed, 413 insertions, 185 deletions

diff --git a/include/clang/AST/StmtCXX.h b/include/clang/AST/StmtCXX.h
index d6c9654fefa4..77f81838e5eb 100644
--- a/include/clang/AST/StmtCXX.h
+++ b/include/clang/AST/StmtCXX.h
@@ -317,6 +317,7 @@ class CoroutineBodyStmt final
   unsigned NumParams;
 
   friend class ASTStmtReader;
+  friend class ASTReader;
   friend TrailingObjects;
 
   Stmt **getStoredStmts() { return getTrailingObjects<Stmt *>(); }
@@ -347,6 +348,8 @@ private:
 
 public:
   static CoroutineBodyStmt *Create(const ASTContext &C, CtorArgs const &Args);
+  static CoroutineBodyStmt *Create(const ASTContext &C, EmptyShell,
+                                   unsigned NumParams);
 
   bool hasDependentPromiseType() const {
     return getPromiseDecl()->getType()->isDependentType();
@@ -444,6 +447,8 @@ public:
     SubStmts[SubStmt::PromiseCall] = PromiseCall;
   }
 
+  CoreturnStmt(EmptyShell) : CoreturnStmt({}, {}, {}) {}
+
   SourceLocation getKeywordLoc() const { return CoreturnLoc; }
 
   /// \brief Retrieve the operand of the 'co_return' statement. Will be nullptr
diff --git a/include/clang/Basic/BuiltinsX86.def b/include/clang/Basic/BuiltinsX86.def
index a516bf6bf06c..6d3a478ac360 100644
--- a/include/clang/Basic/BuiltinsX86.def
+++ b/include/clang/Basic/BuiltinsX86.def
@@ -338,8 +338,8 @@ TARGET_BUILTIN(__builtin_ia32_lfence, "v", "", "sse2")
 TARGET_HEADER_BUILTIN(_mm_lfence, "v", "h", "emmintrin.h", ALL_LANGUAGES, "sse2")
 TARGET_BUILTIN(__builtin_ia32_mfence, "v", "", "sse2")
 TARGET_HEADER_BUILTIN(_mm_mfence, "v", "h", "emmintrin.h", ALL_LANGUAGES, "sse2")
-TARGET_BUILTIN(__builtin_ia32_pause, "v", "", "sse2")
-TARGET_HEADER_BUILTIN(_mm_pause, "v", "h", "emmintrin.h", ALL_LANGUAGES, "sse2")
+TARGET_BUILTIN(__builtin_ia32_pause, "v", "", "")
+TARGET_HEADER_BUILTIN(_mm_pause, "v", "h", "emmintrin.h", ALL_LANGUAGES, "")
 TARGET_BUILTIN(__builtin_ia32_pmuludq128, "V2LLiV4iV4i", "", "sse2")
 TARGET_BUILTIN(__builtin_ia32_psraw128, "V8sV8sV8s", "", "sse2")
 TARGET_BUILTIN(__builtin_ia32_psrad128, "V4iV4iV4i", "", "sse2")
diff --git a/include/clang/Driver/Options.td b/include/clang/Driver/Options.td
index 753c178eec6a..3c0674f598d1 100644
--- a/include/clang/Driver/Options.td
+++ b/include/clang/Driver/Options.td
@@ -2019,10 +2019,6 @@ def mdspr2 : Flag<["-"], "mdspr2">, Group<m_Group>;
 def mno_dspr2 : Flag<["-"], "mno-dspr2">, Group<m_Group>;
 def msingle_float : Flag<["-"], "msingle-float">, Group<m_Group>;
 def mdouble_float : Flag<["-"], "mdouble-float">, Group<m_Group>;
-def mmadd4 : Flag<["-"], "mmadd4">, Group<m_Group>,
-  HelpText<"Enable the generation of 4-operand madd.s, madd.d and related instructions.">;
-def mno_madd4 : Flag<["-"], "mno-madd4">, Group<m_Group>,
-  HelpText<"Disable the generation of 4-operand madd.s, madd.d and related instructions.">;
 def mmsa : Flag<["-"], "mmsa">, Group<m_Group>,
   HelpText<"Enable MSA ASE (MIPS only)">;
 def mno_msa : Flag<["-"], "mno-msa">, Group<m_Group>,
diff --git a/include/clang/Serialization/ASTBitCodes.h b/include/clang/Serialization/ASTBitCodes.h
index 6b40781a1239..9227b33d2c53 100644
--- a/include/clang/Serialization/ASTBitCodes.h
+++ b/include/clang/Serialization/ASTBitCodes.h
@@ -1545,9 +1545,14 @@ namespace clang {
 
       // ARC
       EXPR_OBJC_BRIDGED_CAST,     // ObjCBridgedCastExpr
-      
+
       STMT_MS_DEPENDENT_EXISTS,   // MSDependentExistsStmt
-      EXPR_LAMBDA                 // LambdaExpr
+      EXPR_LAMBDA,                // LambdaExpr
+      STMT_COROUTINE_BODY,
+      STMT_CORETURN,
+      EXPR_COAWAIT,
+      EXPR_COYIELD,
+      EXPR_DEPENDENT_COAWAIT,
     };
 
     /// \brief The kinds of designators that can occur in a
diff --git a/lib/AST/ODRHash.cpp b/lib/AST/ODRHash.cpp
index b19135384cfd..121724a73152 100644
--- a/lib/AST/ODRHash.cpp
+++ b/lib/AST/ODRHash.cpp
@@ -378,8 +378,12 @@ void ODRHash::AddCXXRecordDecl(const CXXRecordDecl *Record) {
   assert(Record && Record->hasDefinition() &&
          "Expected non-null record to be a definition.");
 
-  if (isa<ClassTemplateSpecializationDecl>(Record)) {
-    return;
+  const DeclContext *DC = Record;
+  while (DC) {
+    if (isa<ClassTemplateSpecializationDecl>(DC)) {
+      return;
+    }
+    DC = DC->getParent();
   }
 
   AddDecl(Record);
diff --git a/lib/AST/StmtCXX.cpp b/lib/AST/StmtCXX.cpp
index 8466cd61f055..666f5dcc9d97 100644
--- a/lib/AST/StmtCXX.cpp
+++ b/lib/AST/StmtCXX.cpp
@@ -96,6 +96,20 @@ CoroutineBodyStmt *CoroutineBodyStmt::Create(
   return new (Mem) CoroutineBodyStmt(Args);
 }
 
+CoroutineBodyStmt *CoroutineBodyStmt::Create(const ASTContext &C, EmptyShell,
+                                             unsigned NumParams) {
+  std::size_t Size = totalSizeToAlloc<Stmt *>(
+      CoroutineBodyStmt::FirstParamMove + NumParams);
+
+  void *Mem = C.Allocate(Size, alignof(CoroutineBodyStmt));
+  auto *Result = new (Mem) CoroutineBodyStmt(CtorArgs());
+  Result->NumParams = NumParams;
+  auto *ParamBegin = Result->getStoredStmts() + SubStmt::FirstParamMove;
+  std::uninitialized_fill(ParamBegin, ParamBegin + NumParams,
+                          static_cast<Stmt *>(nullptr));
+  return Result;
+}
+
 CoroutineBodyStmt::CoroutineBodyStmt(CoroutineBodyStmt::CtorArgs const &Args)
     : Stmt(CoroutineBodyStmtClass), NumParams(Args.ParamMoves.size()) {
   Stmt **SubStmts = getStoredStmts();
diff --git a/lib/Basic/Targets.cpp b/lib/Basic/Targets.cpp
index 5d75aa5a7528..73be2e173fda 100644
--- a/lib/Basic/Targets.cpp
+++ b/lib/Basic/Targets.cpp
@@ -8050,7 +8050,6 @@ class MipsTargetInfo : public TargetInfo {
     NoDSP, DSP1, DSP2
   } DspRev;
   bool HasMSA;
-  bool DisableMadd4;
 
 protected:
   bool HasFP64;
@@ -8061,7 +8060,7 @@ public:
       : TargetInfo(Triple), IsMips16(false), IsMicromips(false),
         IsNan2008(false), IsSingleFloat(false), IsNoABICalls(false),
         CanUseBSDABICalls(false), FloatABI(HardFloat), DspRev(NoDSP),
-        HasMSA(false), DisableMadd4(false), HasFP64(false) {
+        HasMSA(false), HasFP64(false) {
     TheCXXABI.set(TargetCXXABI::GenericMIPS);
 
     setABI((getTriple().getArch() == llvm::Triple::mips ||
@@ -8307,9 +8306,6 @@ public:
     if (HasMSA)
       Builder.defineMacro("__mips_msa", Twine(1));
 
-    if (DisableMadd4)
-      Builder.defineMacro("__mips_no_madd4", Twine(1));
-
     Builder.defineMacro("_MIPS_SZPTR", Twine(getPointerWidth(0)));
     Builder.defineMacro("_MIPS_SZINT", Twine(getIntWidth()));
     Builder.defineMacro("_MIPS_SZLONG", Twine(getLongWidth()));
@@ -8472,8 +8468,6 @@ public:
         DspRev = std::max(DspRev, DSP2);
       else if (Feature == "+msa")
         HasMSA = true;
-      else if (Feature == "+nomadd4")
-        DisableMadd4 = true;
       else if (Feature == "+fp64")
         HasFP64 = true;
       else if (Feature == "-fp64")
diff --git a/lib/Driver/ToolChains/Arch/Mips.cpp b/lib/Driver/ToolChains/Arch/Mips.cpp
index b45dcd6db678..1da90d1dc7ba 100644
--- a/lib/Driver/ToolChains/Arch/Mips.cpp
+++ b/lib/Driver/ToolChains/Arch/Mips.cpp
@@ -297,8 +297,6 @@ void mips::getMIPSTargetFeatures(const Driver &D, const llvm::Triple &Triple,
 
   AddTargetFeature(Args, Features, options::OPT_mno_odd_spreg,
                    options::OPT_modd_spreg, "nooddspreg");
-  AddTargetFeature(Args, Features, options::OPT_mno_madd4, options::OPT_mmadd4,
-                   "nomadd4");
   AddTargetFeature(Args, Features, options::OPT_mlong_calls,
                    options::OPT_mno_long_calls, "long-calls");
   AddTargetFeature(Args, Features, options::OPT_mmt, options::OPT_mno_mt,"mt");
diff --git a/lib/Driver/ToolChains/OpenBSD.cpp b/lib/Driver/ToolChains/OpenBSD.cpp
index c5f266ec8fdc..1d54a1e9cbb0 100644
--- a/lib/Driver/ToolChains/OpenBSD.cpp
+++ b/lib/Driver/ToolChains/OpenBSD.cpp
@@ -133,6 +133,8 @@ void openbsd::Linker::ConstructJob(Compilation &C, const JobAction &JA,
     }
   }
 
+  if (Args.hasArg(options::OPT_pie))
+    CmdArgs.push_back("-pie");
   if (Args.hasArg(options::OPT_nopie))
     CmdArgs.push_back("-nopie");
 
diff --git a/lib/Headers/avx2intrin.h b/lib/Headers/avx2intrin.h
index 5d83a8db484b..576f761b2542 100644
--- a/lib/Headers/avx2intrin.h
+++ b/lib/Headers/avx2intrin.h
@@ -832,7 +832,8 @@ _mm256_xor_si256(__m256i __a, __m256i __b)
 static __inline__ __m256i __DEFAULT_FN_ATTRS
 _mm256_stream_load_si256(__m256i const *__V)
 {
-  return (__m256i)__builtin_nontemporal_load((const __v4di *)__V);
+  typedef __v4di __v4di_aligned __attribute__((aligned(32)));
+  return (__m256i)__builtin_nontemporal_load((const __v4di_aligned *)__V);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
diff --git a/lib/Headers/avx512fintrin.h b/lib/Headers/avx512fintrin.h
index b556d04efbb7..4ce694531100 100644
--- a/lib/Headers/avx512fintrin.h
+++ b/lib/Headers/avx512fintrin.h
@@ -4289,7 +4289,7 @@ static __inline__ __m512i __DEFAULT_FN_ATTRS
 _mm512_maskz_cvtps_epu32 ( __mmask16 __U, __m512 __A)
 {
   return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
-                  (__v16si) 
+                  (__v16si)
                   _mm512_setzero_si512 (),
                   (__mmask16) __U ,
                   _MM_FROUND_CUR_DIRECTION);
@@ -9035,25 +9035,29 @@ _mm512_kxor (__mmask16 __A, __mmask16 __B)
 static __inline__ void __DEFAULT_FN_ATTRS
 _mm512_stream_si512 (__m512i * __P, __m512i __A)
 {
-  __builtin_nontemporal_store((__v8di)__A, (__v8di*)__P);
+  typedef __v8di __v8di_aligned __attribute__((aligned(64)));
+  __builtin_nontemporal_store((__v8di_aligned)__A, (__v8di_aligned*)__P);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
 _mm512_stream_load_si512 (void *__P)
 {
-  return (__m512i) __builtin_nontemporal_load((const __v8di *)__P);
+  typedef __v8di __v8di_aligned __attribute__((aligned(64)));
+  return (__m512i) __builtin_nontemporal_load((const __v8di_aligned *)__P);
 }
 
 static __inline__ void __DEFAULT_FN_ATTRS
 _mm512_stream_pd (double *__P, __m512d __A)
 {
-  __builtin_nontemporal_store((__v8df)__A, (__v8df*)__P);
+  typedef __v8df __v8df_aligned __attribute__((aligned(64)));
+  __builtin_nontemporal_store((__v8df_aligned)__A, (__v8df_aligned*)__P);
 }
 
 static __inline__ void __DEFAULT_FN_ATTRS
 _mm512_stream_ps (float *__P, __m512 __A)
 {
-  __builtin_nontemporal_store((__v16sf)__A, (__v16sf*)__P);
+  typedef __v16sf __v16sf_aligned __attribute__((aligned(64)));
+  __builtin_nontemporal_store((__v16sf_aligned)__A, (__v16sf_aligned*)__P);
 }
 
 static __inline__ __m512d __DEFAULT_FN_ATTRS
@@ -9217,39 +9221,39 @@ _mm512_maskz_moveldup_ps (__mmask16 __U, __m512 __A)
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_move_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
 {
-  __m128 res = __A; 
+  __m128 res = __A;
   res[0] = (__U & 1) ? __B[0] : __W[0];
-  return res; 
+  return res;
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_move_ss (__mmask8 __U, __m128 __A, __m128 __B)
 {
-  __m128 res = __A; 
-  res[0] = (__U & 1) ? __B[0] : 0; 
-  return res; 
+  __m128 res = __A;
+  res[0] = (__U & 1) ? __B[0] : 0;
+  return res;
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_move_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
 {
-  __m128d res = __A; 
+  __m128d res = __A;
   res[0] = (__U & 1) ? __B[0] : __W[0];
-  return res; 
+  return res;
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_move_sd (__mmask8 __U, __m128d __A, __m128d __B)
 {
-  __m128d res = __A; 
-  res[0] = (__U & 1) ? __B[0] : 0; 
-  return res; 
+  __m128d res = __A;
+  res[0] = (__U & 1) ? __B[0] : 0;
+  return res;
 }
 
 static __inline__ void __DEFAULT_FN_ATTRS
 _mm_mask_store_ss (float * __W, __mmask8 __U, __m128 __A)
 {
-  __builtin_ia32_storess128_mask ((__v16sf *)__W, 
+  __builtin_ia32_storess128_mask ((__v16sf *)__W,
                 (__v16sf) _mm512_castps128_ps512(__A),
                 (__mmask16) __U & (__mmask16)1);
 }
@@ -9257,7 +9261,7 @@ _mm_mask_store_ss (float * __W, __mmask8 __U, __m128 __A)
 static __inline__ void __DEFAULT_FN_ATTRS
 _mm_mask_store_sd (double * __W, __mmask8 __U, __m128d __A)
 {
-  __builtin_ia32_storesd128_mask ((__v8df *)__W, 
+  __builtin_ia32_storesd128_mask ((__v8df *)__W,
                 (__v8df) _mm512_castpd128_pd512(__A),
                 (__mmask8) __U & 1);
 }
@@ -9606,7 +9610,7 @@ _mm_mask_cvtsd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128d __B)
 {
   return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)(__A),
                                              (__v2df)(__B),
-                                             (__v4sf)(__W), 
+                                             (__v4sf)(__W),
                                              (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
 }
 
@@ -9615,7 +9619,7 @@ _mm_maskz_cvtsd_ss (__mmask8 __U, __m128 __A, __m128d __B)
 {
   return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)(__A),
                                              (__v2df)(__B),
-                                             (__v4sf)_mm_setzero_ps(), 
+                                             (__v4sf)_mm_setzero_ps(),
                                              (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
 }
 
@@ -9680,7 +9684,7 @@ _mm_mask_cvtss_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128 __B)
   return __builtin_ia32_cvtss2sd_round_mask((__v2df)(__A),
                                               (__v4sf)(__B),
                                               (__v2df)(__W),
-                                              (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION); 
+                                              (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
@@ -9688,8 +9692,8 @@ _mm_maskz_cvtss_sd (__mmask8 __U, __m128d __A, __m128 __B)
 {
   return __builtin_ia32_cvtss2sd_round_mask((__v2df)(__A),
                                               (__v4sf)(__B),
-                                              (__v2df)_mm_setzero_pd(), 
-                                              (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION); 
+                                              (__v2df)_mm_setzero_pd(),
+                                              (__mmask8)(__U), _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
@@ -9935,7 +9939,7 @@ static __inline__ double __DEFAULT_FN_ATTRS _mm512_reduce_mul_pd(__m512d __W) {
 }
 
 // Vec512 - Vector with size 512.
-// Vec512Neutral - All vector elements set to the identity element. 
+// Vec512Neutral - All vector elements set to the identity element.
 // Identity element: {+,0},{*,1},{&,0xFFFFFFFFFFFFFFFF},{|,0}
 // Operator - Can be one of following: +,*,&,|
 // Mask - Intrinsic Mask
@@ -9965,19 +9969,19 @@ _mm512_mask_reduce_mul_epi64(__mmask8 __M, __m512i __W) {
 
 static __inline__ long long __DEFAULT_FN_ATTRS
 _mm512_mask_reduce_and_epi64(__mmask8 __M, __m512i __W) {
-  _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0xFFFFFFFFFFFFFFFF), 
+  _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0xFFFFFFFFFFFFFFFF),
                                     &, __M,  i, i, q);
 }
 
 static __inline__ long long __DEFAULT_FN_ATTRS
 _mm512_mask_reduce_or_epi64(__mmask8 __M, __m512i __W) {
-  _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0), |, __M, 
+  _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_epi64(0), |, __M,
                                     i, i, q);
 }
 
 static __inline__ double __DEFAULT_FN_ATTRS
 _mm512_mask_reduce_add_pd(__mmask8 __M, __m512d __W) {
-  _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_pd(0), +, __M, 
+  _mm512_mask_reduce_operator_64bit(__W, _mm512_set1_pd(0), +, __M,
                                     f, d, pd);
 }
 
@@ -10039,17 +10043,17 @@ _mm512_reduce_add_epi32(__m512i __W) {
   _mm512_reduce_operator_32bit(__W, +, i, i);
 }
 
-static __inline__ int __DEFAULT_FN_ATTRS 
+static __inline__ int __DEFAULT_FN_ATTRS
 _mm512_reduce_mul_epi32(__m512i __W) {
   _mm512_reduce_operator_32bit(__W, *, i, i);
 }
 
-static __inline__ int __DEFAULT_FN_ATTRS 
+static __inline__ int __DEFAULT_FN_ATTRS
 _mm512_reduce_and_epi32(__m512i __W) {
   _mm512_reduce_operator_32bit(__W, &, i, i);
 }
 
-static __inline__ int __DEFAULT_FN_ATTRS 
+static __inline__ int __DEFAULT_FN_ATTRS
 _mm512_reduce_or_epi32(__m512i __W) {
   _mm512_reduce_operator_32bit(__W, |, i, i);
 }
@@ -10065,7 +10069,7 @@ _mm512_reduce_mul_ps(__m512 __W) {
 }
 
 // Vec512 - Vector with size 512.
-// Vec512Neutral - All vector elements set to the identity element. 
+// Vec512Neutral - All vector elements set to the identity element.
 // Identity element: {+,0},{*,1},{&,0xFFFFFFFF},{|,0}
 // Operator - Can be one of following: +,*,&,|
 // Mask - Intrinsic Mask
@@ -10095,7 +10099,7 @@ _mm512_mask_reduce_mul_epi32( __mmask16 __M, __m512i __W) {
 
 static __inline__ int __DEFAULT_FN_ATTRS
 _mm512_mask_reduce_and_epi32( __mmask16 __M, __m512i __W) {
-  _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_epi32(0xFFFFFFFF), &, __M, 
+  _mm512_mask_reduce_operator_32bit(__W, _mm512_set1_epi32(0xFFFFFFFF), &, __M,
                                     i, i, d);
 }
 
@@ -10158,7 +10162,7 @@ _mm512_mask_reduce_mul_ps(__mmask16 __M, __m512 __W) {
     return Vec512[0];                                                          \
   })
 
-static __inline__ long long __DEFAULT_FN_ATTRS 
+static __inline__ long long __DEFAULT_FN_ATTRS
 _mm512_reduce_max_epi64(__m512i __V) {
   _mm512_reduce_maxMin_64bit(__V, max_epi64, i, i);
 }
@@ -10168,7 +10172,7 @@ _mm512_reduce_max_epu64(__m512i __V) {
   _mm512_reduce_maxMin_64bit(__V, max_epu64, i, i);
 }
 
-static __inline__ double __DEFAULT_FN_ATTRS 
+static __inline__ double __DEFAULT_FN_ATTRS
 _mm512_reduce_max_pd(__m512d __V) {
   _mm512_reduce_maxMin_64bit(__V, max_pd, d, f);
 }
@@ -10183,7 +10187,7 @@ _mm512_reduce_min_epu64(__m512i __V) {
   _mm512_reduce_maxMin_64bit(__V, min_epu64, i, i);
 }
 
-static __inline__ double __DEFAULT_FN_ATTRS 
+static __inline__ double __DEFAULT_FN_ATTRS
 _mm512_reduce_min_pd(__m512d __V) {
   _mm512_reduce_maxMin_64bit(__V, min_pd, d, f);
 }
diff --git a/lib/Headers/avxintrin.h b/lib/Headers/avxintrin.h
index 78bb70740bbf..dff5897b6bb6 100644
--- a/lib/Headers/avxintrin.h
+++ b/lib/Headers/avxintrin.h
@@ -3590,7 +3590,8 @@ _mm_maskstore_ps(float *__p, __m128i __m, __m128 __a)
 static __inline void __DEFAULT_FN_ATTRS
 _mm256_stream_si256(__m256i *__a, __m256i __b)
 {
-  __builtin_nontemporal_store((__v4di)__b, (__v4di*)__a);
+  typedef __v4di __v4di_aligned __attribute__((aligned(32)));
+  __builtin_nontemporal_store((__v4di_aligned)__b, (__v4di_aligned*)__a);
 }
 
 /// \brief Moves double-precision values from a 256-bit vector of [4 x double]
@@ -3609,7 +3610,8 @@ _mm256_stream_si256(__m256i *__a, __m256i __b)
 static __inline void __DEFAULT_FN_ATTRS
 _mm256_stream_pd(double *__a, __m256d __b)
 {
-  __builtin_nontemporal_store((__v4df)__b, (__v4df*)__a);
+  typedef __v4df __v4df_aligned __attribute__((aligned(32)));
+  __builtin_nontemporal_store((__v4df_aligned)__b, (__v4df_aligned*)__a);
 }
 
 /// \brief Moves single-precision floating point values from a 256-bit vector
@@ -3629,7 +3631,8 @@ _mm256_stream_pd(double *__a, __m256d __b)
 static __inline void __DEFAULT_FN_ATTRS
 _mm256_stream_ps(float *__p, __m256 __a)
 {
-  __builtin_nontemporal_store((__v8sf)__a, (__v8sf*)__p);
+  typedef __v8sf __v8sf_aligned __attribute__((aligned(32)));
+  __builtin_nontemporal_store((__v8sf_aligned)__a, (__v8sf_aligned*)__p);
 }
 
 /* Create vectors */
diff --git a/lib/Headers/float.h b/lib/Headers/float.h
index 0f453d87cbcb..502143d4e481 100644
--- a/lib/Headers/float.h
+++ b/lib/Headers/float.h
@@ -33,6 +33,15 @@
  */
 #if (defined(__APPLE__) || (defined(__MINGW32__) || defined(_MSC_VER))) && \
     __STDC_HOSTED__ && __has_include_next(<float.h>)
+
+/* Prior to Apple's 10.7 SDK, float.h SDK header used to apply an extra level
+ * of #include_next<float.h> to keep Metrowerks compilers happy. Avoid this
+ * extra indirection.
+ */
+#ifdef __APPLE__
+#define _FLOAT_H_
+#endif
+
 #  include_next <float.h>
 
 /* Undefine anything that we'll be redefining below. */
diff --git a/lib/Lex/ScratchBuffer.cpp b/lib/Lex/ScratchBuffer.cpp
index cd8a27e76c28..e0f3966fce48 100644
--- a/lib/Lex/ScratchBuffer.cpp
+++ b/lib/Lex/ScratchBuffer.cpp
@@ -35,6 +35,14 @@ SourceLocation ScratchBuffer::getToken(const char *Buf, unsigned Len,
                                        const char *&DestPtr) {
   if (BytesUsed+Len+2 > ScratchBufSize)
     AllocScratchBuffer(Len+2);
+  else {
+    // Clear out the source line cache if it's already been computed.
+    // FIXME: Allow this to be incrementally extended.
+    auto *ContentCache = const_cast<SrcMgr::ContentCache *>(
+        SourceMgr.getSLocEntry(SourceMgr.getFileID(BufferStartLoc))
+                 .getFile().getContentCache());
+    ContentCache->SourceLineCache = nullptr;
+  }
 
   // Prefix the token with a \n, so that it looks like it is the first thing on
   // its own virtual line in caret diagnostics.
diff --git a/lib/Sema/Sema.cpp b/lib/Sema/Sema.cpp
index 6f0db6ce1c6a..a18f71422fde 100644
--- a/lib/Sema/Sema.cpp
+++ b/lib/Sema/Sema.cpp
@@ -542,6 +542,9 @@ void Sema::getUndefinedButUsed(
     // __attribute__((weakref)) is basically a definition.
     if (ND->hasAttr<WeakRefAttr>()) continue;
 
+    if (isa<CXXDeductionGuideDecl>(ND))
+      continue;
+
     if (FunctionDecl *FD = dyn_cast<FunctionDecl>(ND)) {
       if (FD->isDefined())
         continue;
diff --git a/lib/Sema/SemaExpr.cpp b/lib/Sema/SemaExpr.cpp
index ead80b61586a..d3d7d8b67c70 100644
--- a/lib/Sema/SemaExpr.cpp
+++ b/lib/Sema/SemaExpr.cpp
@@ -8288,7 +8288,7 @@ QualType Sema::CheckVectorOperands(ExprResult &LHS, ExprResult &RHS,
     // type. Note that this is already done by non-compound assignments in
     // CheckAssignmentConstraints. If it's a scalar type, only bitcast for
     // <1 x T> -> T. The result is also a vector type.
-    } else if (OtherType->isExtVectorType() ||
+    } else if (OtherType->isExtVectorType() || OtherType->isVectorType() ||
                (OtherType->isScalarType() && VT->getNumElements() == 1)) {
       ExprResult *RHSExpr = &RHS;
       *RHSExpr = ImpCastExprToType(RHSExpr->get(), LHSType, CK_BitCast);
diff --git a/lib/Serialization/ASTReaderStmt.cpp b/lib/Serialization/ASTReaderStmt.cpp
index 21adcddd3a4a..3f5da029947c 100644
--- a/lib/Serialization/ASTReaderStmt.cpp
+++ b/lib/Serialization/ASTReaderStmt.cpp
@@ -367,28 +367,45 @@ void ASTStmtReader::VisitMSAsmStmt(MSAsmStmt *S) {
 }
 
 void ASTStmtReader::VisitCoroutineBodyStmt(CoroutineBodyStmt *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+  VisitStmt(S);
+  assert(Record.peekInt() == S->NumParams);
+  Record.skipInts(1);
+  auto *StoredStmts = S->getStoredStmts();
+  for (unsigned i = 0;
+       i < CoroutineBodyStmt::SubStmt::FirstParamMove + S->NumParams; ++i)
+    StoredStmts[i] = Record.readSubStmt();
 }
 
 void ASTStmtReader::VisitCoreturnStmt(CoreturnStmt *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+  VisitStmt(S);
+  S->CoreturnLoc = Record.readSourceLocation();
+  for (auto &SubStmt: S->SubStmts)
+    SubStmt = Record.readSubStmt();
+  S->IsImplicit = Record.readInt() != 0;
 }
 
-void ASTStmtReader::VisitCoawaitExpr(CoawaitExpr *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtReader::VisitCoawaitExpr(CoawaitExpr *E) {
+  VisitExpr(E);
+  E->KeywordLoc = ReadSourceLocation();
+  for (auto &SubExpr: E->SubExprs)
+    SubExpr = Record.readSubStmt();
+  E->OpaqueValue = cast_or_null<OpaqueValueExpr>(Record.readSubStmt());
+  E->setIsImplicit(Record.readInt() != 0);
 }
 
-void ASTStmtReader::VisitDependentCoawaitExpr(DependentCoawaitExpr *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtReader::VisitCoyieldExpr(CoyieldExpr *E) {
+  VisitExpr(E);
+  E->KeywordLoc = ReadSourceLocation();
+  for (auto &SubExpr: E->SubExprs)
+    SubExpr = Record.readSubStmt();
+  E->OpaqueValue = cast_or_null<OpaqueValueExpr>(Record.readSubStmt());
 }
 
-void ASTStmtReader::VisitCoyieldExpr(CoyieldExpr *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtReader::VisitDependentCoawaitExpr(DependentCoawaitExpr *E) {
+  VisitExpr(E);
+  E->KeywordLoc = ReadSourceLocation();
+  for (auto &SubExpr: E->SubExprs)
+    SubExpr = Record.readSubStmt();
 }
 
 void ASTStmtReader::VisitCapturedStmt(CapturedStmt *S) {
@@ -3947,6 +3964,29 @@ Stmt *ASTReader::ReadStmtFromStream(ModuleFile &F) {
       S = LambdaExpr::CreateDeserialized(Context, NumCaptures);
       break;
     }
+
+    case STMT_COROUTINE_BODY: {
+      unsigned NumParams = Record[ASTStmtReader::NumStmtFields];
+      S = CoroutineBodyStmt::Create(Context, Empty, NumParams);
+      break;
+    }
+
+    case STMT_CORETURN:
+      S = new (Context) CoreturnStmt(Empty);
+      break;
+
+    case EXPR_COAWAIT:
+      S = new (Context) CoawaitExpr(Empty);
+      break;
+
+    case EXPR_COYIELD:
+      S = new (Context) CoyieldExpr(Empty);
+      break;
+
+    case EXPR_DEPENDENT_COAWAIT:
+      S = new (Context) DependentCoawaitExpr(Empty);
+      break;
+
     }
 
     // We hit a STMT_STOP, so we're done with this expression.
diff --git a/lib/Serialization/ASTWriterStmt.cpp b/lib/Serialization/ASTWriterStmt.cpp
index ae2e0b88c311..6971339663f0 100644
--- a/lib/Serialization/ASTWriterStmt.cpp
+++ b/lib/Serialization/ASTWriterStmt.cpp
@@ -286,7 +286,7 @@ void ASTStmtWriter::VisitMSAsmStmt(MSAsmStmt *S) {
   }
 
   // Outputs
-  for (unsigned I = 0, N = S->getNumOutputs(); I != N; ++I) {      
+  for (unsigned I = 0, N = S->getNumOutputs(); I != N; ++I) {
     Record.AddStmt(S->getOutputExpr(I));
     Record.AddString(S->getOutputConstraint(I));
   }
@@ -300,29 +300,48 @@ void ASTStmtWriter::VisitMSAsmStmt(MSAsmStmt *S) {
   Code = serialization::STMT_MSASM;
 }
 
-void ASTStmtWriter::VisitCoroutineBodyStmt(CoroutineBodyStmt *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtWriter::VisitCoroutineBodyStmt(CoroutineBodyStmt *CoroStmt) {
+  VisitStmt(CoroStmt);
+  Record.push_back(CoroStmt->getParamMoves().size());
+  for (Stmt *S : CoroStmt->children())
+    Record.AddStmt(S);
+  Code = serialization::STMT_COROUTINE_BODY;
 }
 
 void ASTStmtWriter::VisitCoreturnStmt(CoreturnStmt *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+  VisitStmt(S);
+  Record.AddSourceLocation(S->getKeywordLoc());
+  Record.AddStmt(S->getOperand());
+  Record.AddStmt(S->getPromiseCall());
+  Record.push_back(S->isImplicit());
+  Code = serialization::STMT_CORETURN;
 }
 
-void ASTStmtWriter::VisitCoawaitExpr(CoawaitExpr *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtWriter::VisitCoroutineSuspendExpr(CoroutineSuspendExpr *E) {
+  VisitExpr(E);
+  Record.AddSourceLocation(E->getKeywordLoc());
+  for (Stmt *S : E->children())
+    Record.AddStmt(S);
+  Record.AddStmt(E->getOpaqueValue());
 }
 
-void ASTStmtWriter::VisitDependentCoawaitExpr(DependentCoawaitExpr *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtWriter::VisitCoawaitExpr(CoawaitExpr *E) {
+  VisitCoroutineSuspendExpr(E);
+  Record.push_back(E->isImplicit());
+  Code = serialization::EXPR_COAWAIT;
+}
+
+void ASTStmtWriter::VisitCoyieldExpr(CoyieldExpr *E) {
+  VisitCoroutineSuspendExpr(E);
+  Code = serialization::EXPR_COYIELD;
 }
 
-void ASTStmtWriter::VisitCoyieldExpr(CoyieldExpr *S) {
-  // FIXME: Implement coroutine serialization.
-  llvm_unreachable("unimplemented");
+void ASTStmtWriter::VisitDependentCoawaitExpr(DependentCoawaitExpr *E) {
+  VisitExpr(E);
+  Record.AddSourceLocation(E->getKeywordLoc());
+  for (Stmt *S : E->children())
+    Record.AddStmt(S);
+  Code = serialization::EXPR_DEPENDENT_COAWAIT;
 }
 
 void ASTStmtWriter::VisitCapturedStmt(CapturedStmt *S) {
diff --git a/test/CodeGen/mips-madd4.c b/test/CodeGen/mips-madd4.c
deleted file mode 100644
index bc7bb593f95d..000000000000
--- a/test/CodeGen/mips-madd4.c
+++ /dev/null
@@ -1,87 +0,0 @@
-// REQUIRES: mips-registered-target
-// RUN: %clang --target=mips64-unknown-linux -S -mmadd4    %s -o -| FileCheck %s -check-prefix=MADD4
-// RUN: %clang --target=mips64-unknown-linux -S -mno-madd4 %s -o -| FileCheck %s -check-prefix=NOMADD4
-// RUN: %clang --target=mips64-unknown-linux -S -mmadd4    -fno-honor-nans %s -o -| FileCheck %s -check-prefix=MADD4-NONAN
-// RUN: %clang --target=mips64-unknown-linux -S -mno-madd4 -fno-honor-nans %s -o -| FileCheck %s -check-prefix=NOMADD4-NONAN
- 
-float madd_s (float f, float g, float h)
-{
-  return (f * g) + h;
-}
-// MADD4:   madd.s
-// NOMADD4: mul.s
-// NOMADD4: add.s
-
-float msub_s (float f, float g, float h)
-{
-  return (f * g) - h;
-}
-// MADD4:   msub.s
-// NOMADD4: mul.s
-// NOMADD4: sub.s
-
-double madd_d (double f, double g, double h)
-{
-  return (f * g) + h;
-}
-// MADD4:   madd.d
-// NOMADD4: mul.d
-// NOMADD4: add.d
-
-double msub_d (double f, double g, double h)
-{
-  return (f * g) - h;
-}
-// MADD4:   msub.d
-// NOMADD4: mul.d
-// NOMADD4: sub.d
-
-
-float nmadd_s (float f, float g, float h)
-{
-  // FIXME: Zero has been explicitly placed to force generation of a positive
-  // zero in IR until pattern used to match this instruction is changed to
-  // comply with negative zero as well.
-  return 0-((f * g) + h);
-}
-// MADD4-NONAN:   nmadd.s
-// NOMADD4-NONAN: mul.s
-// NOMADD4-NONAN: add.s
-// NOMADD4-NONAN: sub.s
-
-float nmsub_s (float f, float g, float h)
-{
-  // FIXME: Zero has been explicitly placed to force generation of a positive
-  // zero in IR until pattern used to match this instruction is changed to
-  // comply with negative zero as well.
-  return 0-((f * g) - h);
-}
-// MADD4-NONAN:   nmsub.s
-// NOMADD4-NONAN: mul.s
-// NOMADD4-NONAN: sub.s
-// NOMADD4-NONAN: sub.s
-
-double nmadd_d (double f, double g, double h)
-{
-  // FIXME: Zero has been explicitly placed to force generation of a positive
-  // zero in IR until pattern used to match this instruction is changed to
-  // comply with negative zero as well.
-  return 0-((f * g) + h);
-}
-// MADD4-NONAN:   nmadd.d
-// NOMADD4-NONAN: mul.d
-// NOMADD4-NONAN: add.d
-// NOMADD4-NONAN: sub.d
-
-double nmsub_d (double f, double g, double h)
-{
-  // FIXME: Zero has been explicitly placed to force generation of a positive
-  // zero in IR until pattern used to match this instruction is changed to
-  // comply with negative zero as well.
-  return 0-((f * g) - h);
-}
-// MADD4-NONAN:   nmsub.d
-// NOMADD4-NONAN: mul.d
-// NOMADD4-NONAN: sub.d
-// NOMADD4-NONAN: sub.d
-
diff --git a/test/CodeGen/pause.c b/test/CodeGen/pause.c
new file mode 100644
index 000000000000..1a8e10aec043
--- /dev/null
+++ b/test/CodeGen/pause.c
@@ -0,0 +1,11 @@
+// RUN: %clang_cc1 -ffreestanding %s -triple=i386-pc-win32 -target-feature -sse2 -emit-llvm -o - -Wall -Werror | FileCheck %s
+// RUN: %clang_cc1 -ffreestanding %s -triple=i386-pc-win32 -target-feature +sse2 -emit-llvm -o - -Wall -Werror | FileCheck %s
+
+
+#include <x86intrin.h>
+
+void test_mm_pause() {
+  // CHECK-LABEL: test_mm_pause
+  // CHECK: call void @llvm.x86.sse2.pause()
+  return _mm_pause();
+}
diff --git a/test/CodeGen/x86-nontemporal.c b/test/CodeGen/x86-nontemporal.c
new file mode 100644
index 000000000000..5e9e42c9f204
--- /dev/null
+++ b/test/CodeGen/x86-nontemporal.c
@@ -0,0 +1,93 @@
+// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +sse4.1 -target-feature +avx -target-feature +avx2 -target-feature +avx512f -emit-llvm -o - -Wall -Werror | FileCheck %s --check-prefix=CHECK
+// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +sse4.1 -target-feature +avx -target-feature +avx2 -target-feature +avx512f -fno-signed-char -emit-llvm -o - -Wall -Werror | FileCheck %s --check-prefix=CHECK
+
+// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +sse4.1 -target-feature +avx -target-feature +avx2 -target-feature +avx512f -emit-llvm -o - -Wall -Werror -fmax-type-align=16 | FileCheck %s --check-prefix=CHECK
+// RUN: %clang_cc1 -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +sse4.1 -target-feature +avx -target-feature +avx2 -target-feature +avx512f -fno-signed-char -emit-llvm -o - -Wall -Werror -fmax-type-align=16 | FileCheck %s --check-prefix=CHECK
+
+#include <x86intrin.h>
+
+// (PR33830) Tests ensure the correct alignment of non-temporal load/stores on darwin targets where fmax-type-align is set to 16.
+
+//
+// 128-bit vectors
+//
+
+void test_mm_stream_pd(double* A, __m128d B) {
+  // CHECK-LABEL: test_mm_stream_pd
+  // CHECK: store <2 x double> %{{.*}}, <2 x double>* %{{.*}}, align 16, !nontemporal
+  _mm_stream_pd(A, B);
+}
+
+void test_mm_stream_ps(float* A, __m128 B) {
+  // CHECK16-LABEL: test_mm_stream_ps
+  // CHECK16: store <4 x float> %{{.*}}, <4 x float>* %{{.*}}, align 16, !nontemporal
+  _mm_stream_ps(A, B);
+}
+
+void test_mm_stream_si128(__m128i* A, __m128i B) {
+  // CHECK-LABEL: test_mm_stream_si128
+  // CHECK: store <2 x i64> %{{.*}}, <2 x i64>* %{{.*}}, align 16, !nontemporal
+  _mm_stream_si128(A, B);
+}
+
+__m128i test_mm_stream_load_si128(__m128i const *A) {
+  // CHECK-LABEL: test_mm_stream_load_si128
+  // CHECK: load <2 x i64>, <2 x i64>* %{{.*}}, align 16, !nontemporal
+  return _mm_stream_load_si128(A);
+}
+
+//
+// 256-bit vectors
+//
+
+void test_mm256_stream_pd(double* A, __m256d B) {
+  // CHECK-LABEL: test_mm256_stream_pd
+  // CHECK: store <4 x double> %{{.*}}, <4 x double>* %{{.*}}, align 32, !nontemporal
+  _mm256_stream_pd(A, B);
+}
+
+void test_mm256_stream_ps(float* A, __m256 B) {
+  // CHECK-LABEL: test_mm256_stream_ps
+  // CHECK: store <8 x float> %{{.*}}, <8 x float>* %{{.*}}, align 32, !nontemporal
+  _mm256_stream_ps(A, B);
+}
+
+void test_mm256_stream_si256(__m256i* A, __m256i B) {
+  // CHECK-LABEL: test_mm256_stream_si256
+  // CHECK: store <4 x i64> %{{.*}}, <4 x i64>* %{{.*}}, align 32, !nontemporal
+  _mm256_stream_si256(A, B);
+}
+
+__m256i test_mm256_stream_load_si256(__m256i const *A) {
+  // CHECK-LABEL: test_mm256_stream_load_si256
+  // CHECK: load <4 x i64>, <4 x i64>* %{{.*}}, align 32, !nontemporal
+  return _mm256_stream_load_si256(A);
+}
+
+//
+// 512-bit vectors
+//
+
+void test_mm512_stream_pd(double* A, __m512d B) {
+  // CHECK-LABEL: test_mm512_stream_pd
+  // CHECK: store <8 x double> %{{.*}}, <8 x double>* %{{.*}}, align 64, !nontemporal
+  _mm512_stream_pd(A, B);
+}
+
+void test_mm512_stream_ps(float* A, __m512 B) {
+  // CHECK-LABEL: test_mm512_stream_ps
+  // CHECK: store <16 x float> %{{.*}}, <16 x float>* %{{.*}}, align 64, !nontemporal
+  _mm512_stream_ps(A, B);
+}
+
+void test_mm512_stream_si512(__m512i* A, __m512i B) {
+  // CHECK-LABEL: test_mm512_stream_si512
+  // CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 64, !nontemporal
+  _mm512_stream_si512(A, B);
+}
+
+__m512i test_mm512_stream_load_si512(void *A) {
+  // CHECK-LABEL: test_mm512_stream_load_si512
+  // CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 64, !nontemporal
+  return _mm512_stream_load_si512(A);
+}
diff --git a/test/Driver/openbsd.c b/test/Driver/openbsd.c
index b4e2796c5d24..2bcd70377033 100644
--- a/test/Driver/openbsd.c
+++ b/test/Driver/openbsd.c
@@ -77,7 +77,9 @@
 // Check linking against correct startup code when (not) using PIE
 // RUN: %clang -no-canonical-prefixes -target i686-pc-openbsd %s -### 2>&1 \
 // RUN:   | FileCheck -check-prefix=CHECK-PIE %s
-// RUN: %clang -no-canonical-prefixes -target i686-pc-openbsd %s -fno-pie %s -### 2>&1 \
+// RUN: %clang -no-canonical-prefixes -target i686-pc-openbsd -pie %s -### 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHECK-PIE-FLAG %s
+// RUN: %clang -no-canonical-prefixes -target i686-pc-openbsd -fno-pie %s -### 2>&1 \
 // RUN:   | FileCheck -check-prefix=CHECK-PIE %s
 // RUN: %clang -no-canonical-prefixes -target i686-pc-openbsd -static %s -### 2>&1 \
 // RUN:   | FileCheck -check-prefix=CHECK-STATIC-PIE %s
@@ -93,6 +95,7 @@
 // RUN:   | FileCheck -check-prefix=CHECK-NOPIE %s
 // CHECK-PIE: "{{.*}}crt0.o"
 // CHECK-PIE-NOT: "-nopie"
+// CHECK-PIE-FLAG: "-pie"
 // CHECK-STATIC-PIE: "{{.*}}rcrt0.o"
 // CHECK-STATIC-PIE-NOT: "-nopie"
 // CHECK-NOPIE: "-nopie" "{{.*}}crt0.o"
diff --git a/test/Misc/caret-diags-multiline.cpp b/test/Misc/caret-diags-multiline.cpp
index 4826d9beaa3f..90b50ff398ac 100644
--- a/test/Misc/caret-diags-multiline.cpp
+++ b/test/Misc/caret-diags-multiline.cpp
@@ -232,3 +232,8 @@ void multiple_ranges(int a, int b) {
   b
   );
 }
+
+#define pr33902_a(b) #b
+#define pr33902_c(d) _Pragma(pr33902_a(d))
+#define pr33902_e(f) pr33902_c(GCC warning #f)
+pr33902_e() pr33902_e()
diff --git a/test/PCH/coroutines.cpp b/test/PCH/coroutines.cpp
new file mode 100644
index 000000000000..46a2872420bf
--- /dev/null
+++ b/test/PCH/coroutines.cpp
@@ -0,0 +1,77 @@
+// Test this without pch.
+// RUN: %clang_cc1 -include %s -verify -std=c++1z -fcoroutines-ts %s
+
+// Test with pch.
+// RUN: %clang_cc1 -std=c++1z -fcoroutines-ts  -emit-pch -o %t %s
+// RUN: %clang_cc1 -include-pch %t -verify -std=c++1z -fcoroutines-ts %s
+
+#ifndef HEADER
+#define HEADER
+
+namespace std::experimental {
+template <typename... T> struct coroutine_traits;
+
+template <class Promise = void> struct coroutine_handle {
+  coroutine_handle() = default;
+  static coroutine_handle from_address(void *) noexcept;
+};
+template <> struct coroutine_handle<void> {
+  static coroutine_handle from_address(void *) noexcept;
+  coroutine_handle() = default;
+  template <class PromiseType>
+  coroutine_handle(coroutine_handle<PromiseType>) noexcept;
+};
+}
+
+struct suspend_always {
+  bool await_ready() noexcept;
+  void await_suspend(std::experimental::coroutine_handle<>) noexcept;
+  void await_resume() noexcept;
+};
+
+template <typename... Args> struct std::experimental::coroutine_traits<void, Args...> {
+  struct promise_type {
+    void get_return_object() noexcept;
+    suspend_always initial_suspend() noexcept;
+    suspend_always final_suspend() noexcept;
+    void return_void() noexcept;
+    suspend_always yield_value(int) noexcept;
+    promise_type();
+    ~promise_type() noexcept;
+    void unhandled_exception() noexcept;
+  };
+};
+
+template <typename... Args> struct std::experimental::coroutine_traits<int, Args...> {
+  struct promise_type {
+    int get_return_object() noexcept;
+    suspend_always initial_suspend() noexcept;
+    suspend_always final_suspend() noexcept;
+    void return_value(int) noexcept;
+    promise_type();
+    ~promise_type() noexcept;
+    void unhandled_exception() noexcept;
+  };
+};
+
+template <typename T>
+void f(T x) {  // checks coawait_expr and coroutine_body_stmt
+  co_yield 42; // checks coyield_expr
+  co_await x;  // checks dependent_coawait
+  co_return;   // checks coreturn_stmt
+}
+
+template <typename T>
+int f2(T x) {  // checks coawait_expr and coroutine_body_stmt
+  co_return x;   // checks coreturn_stmt with expr
+}
+
+#else
+
+// expected-no-diagnostics
+void g() {
+  f(suspend_always{});
+  f2(42);
+}
+
+#endif
diff --git a/test/Preprocessor/init.c b/test/Preprocessor/init.c
index 5a77d06d2403..3a8e5dbd3fd8 100644
--- a/test/Preprocessor/init.c
+++ b/test/Preprocessor/init.c
@@ -4686,16 +4686,6 @@
 // RUN:   | FileCheck -match-full-lines -check-prefix MIPS-MSA %s
 // MIPS-MSA:#define __mips_msa 1
 //
-// RUN: %clang_cc1 -target-feature +nomadd4 \
-// RUN:   -E -dM -triple=mips-none-none < /dev/null \
-// RUN:   | FileCheck -match-full-lines -check-prefix MIPS-NOMADD4 %s
-// MIPS-NOMADD4:#define __mips_no_madd4 1
-//
-// RUN: %clang_cc1 \
-// RUN:   -E -dM -triple=mips-none-none < /dev/null \
-// RUN:   | FileCheck -match-full-lines -check-prefix MIPS-MADD4 %s
-// MIPS-MADD4-NOT:#define __mips_no_madd4 1
-//
 // RUN: %clang_cc1 -target-cpu mips32r3 -target-feature +nan2008 \
 // RUN:   -E -dM -triple=mips-none-none < /dev/null \
 // RUN:   | FileCheck -match-full-lines -check-prefix MIPS-NAN2008 %s
diff --git a/test/Sema/vector-cast.c b/test/Sema/vector-cast.c
index cf23eb2742c4..2bdc00707d4c 100644
--- a/test/Sema/vector-cast.c
+++ b/test/Sema/vector-cast.c
@@ -48,6 +48,9 @@ typedef float float2 __attribute__ ((vector_size (8)));
 typedef __attribute__((vector_size(8))) double float64x1_t;
 typedef __attribute__((vector_size(16))) double float64x2_t;
 float64x1_t vget_low_f64(float64x2_t __p0);
+typedef float float16 __attribute__((__vector_size__(16)));
+typedef signed int vSInt32 __attribute__((__vector_size__(16)));
+typedef unsigned int vUInt32 __attribute__((__vector_size__(16)));
 
 void f4() {
   float2 f2;
@@ -73,3 +76,8 @@ void f5() {
   v = ptr; // expected-error-re {{assigning to 'short_sizeof_pointer' (vector of {{[0-9]+}} 'short' values) from incompatible type 'void *'}}
   ptr = v; // expected-error {{assigning to 'void *' from incompatible type 'short_sizeof_pointer'}}
 }
+
+void f6(vSInt32 a0) {
+  vUInt32 counter = (float16){0.0f, 0.0f, 0.0f, 0.0f}; // expected-warning {{incompatible vector types initializing 'vUInt32' (vector of 4 'unsigned int' values) with an expression of type 'float16' (vector of 4 'float' values)}}
+  counter -= a0;
+}
diff --git a/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp b/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp
index 159b7072e61e..668c24280258 100644
--- a/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp
+++ b/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp
@@ -1,4 +1,9 @@
-// RUN: %clang_cc1 -std=c++1z -verify %s
+// RUN: %clang_cc1 -std=c++1z -verify %s -DERRORS
+// RUN: %clang_cc1 -std=c++1z -verify %s -UERRORS
+
+// This test is split into two because we only produce "undefined internal"
+// warnings if we didn't produce any errors.
+#if ERRORS
 
 namespace std {
   using size_t = decltype(sizeof(0));
@@ -280,3 +285,21 @@ namespace tuple_tests {
     scoped_lock l = {};
   }
 }
+
+#else
+
+// expected-no-diagnostics
+namespace undefined_warnings {
+  // Make sure we don't get an "undefined but used internal symbol" warning for the deduction guide here.
+  namespace {
+    template <typename T>
+    struct TemplDObj {
+      explicit TemplDObj(T func) noexcept {}
+    };
+    auto test1 = TemplDObj(0);
+
+    TemplDObj(float) -> TemplDObj<double>;
+    auto test2 = TemplDObj(.0f);
+  }
+}
+#endif