merge main into amd-staging

Author: ronlieb

clang/include/clang/Basic/BuiltinsX86.td

@@ -276,21 +276,25 @@ let Features = "sse2", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] i
   def psllw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
   def pslld128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
   def psllq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">;
+  def pmaddwd128 : X86Builtin<"_Vector<4, int>(_Vector<8, short>, _Vector<8, short>)">;
+  def pslldqi128_byteshift : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
+  def psrldqi128_byteshift : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
+}
+
+let Features = "sse2",
+    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
+  def pmuludq128 : X86Builtin<"_Vector<2, long long int>(_Vector<4, int>, _Vector<4, int>)">;
+
   def psllwi128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, int)">;
   def pslldi128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, int)">;
   def psllqi128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, int)">;
+
   def psrlwi128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, int)">;
   def psrldi128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, int)">;
   def psrlqi128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, int)">;
+
   def psrawi128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, int)">;
   def psradi128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, int)">;
-  def pmaddwd128 : X86Builtin<"_Vector<4, int>(_Vector<8, short>, _Vector<8, short>)">;
-  def pslldqi128_byteshift : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
-  def psrldqi128_byteshift : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
-}
-
-let Features = "sse2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
-  def pmuludq128 : X86Builtin<"_Vector<2, long long int>(_Vector<4, int>, _Vector<4, int>)">;
 }
 
 let Features = "sse3", Attributes = [NoThrow] in {
@@ -595,23 +599,15 @@ let Features = "avx2", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] i
   def psignb256 : X86Builtin<"_Vector<32, char>(_Vector<32, char>, _Vector<32, char>)">;
   def psignw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<16, short>)">;
   def psignd256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<8, int>)">;
-  def psllwi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">;
   def psllw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
   def pslldqi256_byteshift : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
-  def pslldi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">;
   def pslld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
-  def psllqi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">;
   def psllq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
-  def psrawi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">;
   def psraw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
-  def psradi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">;
   def psrad256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
   def psrldqi256_byteshift : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
-  def psrlwi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">;
   def psrlw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
-  def psrldi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">;
   def psrld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
-  def psrlqi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">;
   def psrlq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
   def pblendd128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>, _Constant int)">;
   def pblendd256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<8, int>, _Constant int)">;
@@ -628,6 +624,19 @@ let Features = "avx2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWi
   def pmuldq256 : X86Builtin<"_Vector<4, long long int>(_Vector<8, int>, _Vector<8, int>)">;
   def pmuludq256 : X86Builtin<"_Vector<4, long long int>(_Vector<8, int>, _Vector<8, int>)">;
 
+  def psllwi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">;
+  def pslldi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">;
+  def psllqi256
+      : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">;
+
+  def psrlwi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">;
+  def psrldi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">;
+  def psrlqi256
+      : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">;
+
+  def psrawi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">;
+  def psradi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">;
+
   def pmulhuw256 : X86Builtin<"_Vector<16, unsigned short>(_Vector<16, unsigned short>, _Vector<16, unsigned short>)">;
   def pmulhw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<16, short>)">;
 
@@ -2038,43 +2047,22 @@ let Features = "avx512dq,evex512", Attributes = [NoThrow, Const, RequiredVectorW
 
 let Features = "avx512f,evex512", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
   def prold512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, _Constant int)">;
-  def prolq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Constant int)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
-  def prold128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Constant int)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
-  def prold256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Constant int)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
-  def prolq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
-  def prolq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
-}
-
-let Features = "avx512f,evex512", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
   def prord512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, _Constant int)">;
+  def prolq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Constant int)">;
   def prorq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Constant int)">;
 }
 
 let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
+  def prold128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Constant int)">;
   def prord128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Constant int)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
-  def prord256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Constant int)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
+  def prolq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
   def prorq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Constant int)">;
 }
 
 let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
+  def prold256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Constant int)">;
+  def prord256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Constant int)">;
+  def prolq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
   def prorq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
 }
 
@@ -2083,7 +2071,6 @@ let Features = "avx512bw,evex512", Attributes = [NoThrow, Const, RequiredVectorW
   def pshuflw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Constant int)">;
   def psllv32hi : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<32, short>)">;
   def psllw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
-  def psllwi512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, int)">;
 }
 
 let Features = "avx512bw,avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
@@ -2094,7 +2081,9 @@ let Features = "avx512bw,avx512vl", Attributes = [NoThrow, Const, RequiredVector
   def psllv8hi : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
 }
 
-let Features = "avx512f,evex512", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
+let Features = "avx512f,evex512",
+    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in {
+  def psllwi512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, int)">;
   def pslldi512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, int)">;
   def psllqi512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, int)">;
 }
@@ -2111,7 +2100,9 @@ let Features = "avx512bw,avx512vl", Attributes = [NoThrow, Const, RequiredVector
   def psrlv8hi : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
 }
 
-let Features = "avx512f,evex512", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
+let Features = "avx512f,evex512",
+    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in {
+  def psrlwi512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, int)">;
   def psrldi512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, int)">;
   def psrlqi512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, int)">;
 }
@@ -2137,10 +2128,10 @@ let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256
 }
 
 let Features = "avx512bw,evex512", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
-  def psraw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
-  def psrawi512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, int)">;
-  def psrlw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
-  def psrlwi512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, int)">;
+  def psraw512
+      : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
+  def psrlw512
+      : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
   def pslldqi512_byteshift : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Constant int)">;
   def psrldqi512_byteshift : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Constant int)">;
 }
@@ -2456,7 +2447,9 @@ let Features = "avx512f", Attributes = [NoThrow, Const, RequiredVectorWidth<128>
   def scalefss_round_mask : X86Builtin<"_Vector<4, float>(_Vector<4, float>, _Vector<4, float>, _Vector<4, float>, unsigned char, _Constant int)">;
 }
 
-let Features = "avx512f,evex512", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
+let Features = "avx512f,evex512",
+    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in {
+  def psrawi512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, int)">;
   def psradi512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, int)">;
   def psraqi512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, int)">;
 }
@@ -2469,11 +2462,13 @@ let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256
   def psraq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
 }
 
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
+let Features = "avx512vl",
+    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
   def psraqi128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, int)">;
 }
 
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
+let Features = "avx512vl",
+    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<256>] in {
   def psraqi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">;
 }
 

clang/include/clang/Basic/LangOptions.def

@@ -256,6 +256,7 @@ LANGOPT(HLSL, 1, 0, NotCompatible, "HLSL")
 ENUM_LANGOPT(HLSLVersion, HLSLLangStd, 16, HLSL_Unset, NotCompatible, "HLSL Version")
 LANGOPT(HLSLStrictAvailability, 1, 0, NotCompatible,
         "Strict availability diagnostic mode for HLSL built-in functions.")
+LANGOPT(HLSLSpvUseUnknownImageFormat, 1, 0, NotCompatible, "For storage images and texel buffers, sets the default format to 'Unknown' when not specified via the `vk::image_format` attribute. If this option is not used, the format is inferred from the resource's data type.")
 
 LANGOPT(CUDAIsDevice      , 1, 0, NotCompatible, "compiling for CUDA device")
 LANGOPT(CUDAAllowVariadicFunctions, 1, 0, NotCompatible, "allowing variadic functions in CUDA device code")

clang/include/clang/Driver/Options.td

@@ -9802,6 +9802,16 @@ def fvk_use_scalar_layout
     : DXCFlag<"fvk-use-scalar-layout">,
       HelpText<"Use scalar memory layout for Vulkan resources.">;
 
+def fhlsl_spv_use_unknown_image_format
+    : Flag<["-"], "fspv-use-unknown-image-format">,
+      Group<dxc_Group>,
+      Visibility<[CC1Option, DXCOption]>,
+      HelpText<"For storage images and texel buffers, sets the default format "
+               "to 'Unknown' when not specified via the `vk::image_format` "
+               "attribute. If this option is not used, the format is inferred "
+               "from the resource's data type.">,
+      MarshallingInfoFlag<LangOpts<"HLSLSpvUseUnknownImageFormat">>;
+
 def no_wasm_opt : Flag<["--"], "no-wasm-opt">,
   Group<m_Group>,
   HelpText<"Disable the wasm-opt optimizer">,

clang/lib/AST/ASTImporter.cpp

@@ -1740,21 +1740,10 @@ ExpectedType ASTNodeImporter::VisitDeducedTemplateSpecializationType(
 }
 
 ExpectedType ASTNodeImporter::VisitTagType(const TagType *T) {
-  TagDecl *DeclForType = T->getOriginalDecl();
-  Expected<TagDecl *> ToDeclOrErr = import(DeclForType);
+  Expected<TagDecl *> ToDeclOrErr = import(T->getOriginalDecl());
   if (!ToDeclOrErr)
     return ToDeclOrErr.takeError();
 
-  if (DeclForType->isUsed()) {
-    // If there is a definition of the 'OriginalDecl', it should be imported to
-    // have all information for the type in the "To" AST. (In some cases no
-    // other reference may exist to the definition decl and it would not be
-    // imported otherwise.)
-    Expected<TagDecl *> ToDefDeclOrErr = import(DeclForType->getDefinition());
-    if (!ToDefDeclOrErr)
-      return ToDefDeclOrErr.takeError();
-  }
-
   if (T->isCanonicalUnqualified())
     return Importer.getToContext().getCanonicalTagType(*ToDeclOrErr);
 

clang/lib/AST/ByteCode/Interp.h

@@ -3490,7 +3490,15 @@ inline bool AllocN(InterpState &S, CodePtr OpPC, PrimType T, const Expr *Source,
     S.Stk.push<Pointer>(0, nullptr);
     return true;
   }
-  assert(NumElements.isPositive());
+  if (NumElements.isNegative()) {
+    if (!IsNoThrow) {
+      S.FFDiag(S.Current->getSource(OpPC), diag::note_constexpr_new_negative)
+          << NumElements.toDiagnosticString(S.getASTContext());
+      return false;
+    }
+    S.Stk.push<Pointer>(0, nullptr);
+    return true;
+  }
 
   if (!CheckArraySize(S, OpPC, static_cast<uint64_t>(NumElements)))
     return false;