merge main into amd-staging

Author: ronlieb

clang-tools-extra/clang-tidy/readability/UppercaseLiteralSuffixCheck.cpp

@@ -29,7 +29,7 @@ struct IntegerLiteralCheck {
   // integer (wb), and can be a complex number ('i', 'j'). In MS compatibility
   // mode, suffixes like i32 are supported.
   static constexpr llvm::StringLiteral Suffixes =
-      llvm::StringLiteral("uUlLzZwWbBiIjJ");
+      llvm::StringLiteral("uUlLzZwWiIjJ");
 };
 
 struct FloatingLiteralCheck {

clang-tools-extra/test/clang-tidy/checkers/cert/uppercase-literal-suffix-integer.cpp

@@ -128,3 +128,12 @@ void integer_suffix() {
   static_assert(is_same<decltype(v24), const unsigned long long>::value, "");
   static_assert(v24 == 1, "");
 }
+
+void no_warning_on_hex_literals() {
+  int a = 0xa;
+  int b = 0xb;
+  int c = 0xc;
+  int d = 0xd;
+  int e = 0xe;
+  int f = 0xf;
+}

clang/docs/OpenMPSupport.rst

@@ -443,7 +443,8 @@ implementation.
 +-------------------------------------------------------------+---------------------------+---------------------------+--------------------------------------------------------------------------+
 | Traits for default device envirable                         | :none:`unclaimed`         | :none:`unclaimed`         |                                                                          |
 +-------------------------------------------------------------+---------------------------+---------------------------+--------------------------------------------------------------------------+
-| Optionally omit array length expression                     | :good:`done`              | :none:`unclaimed`         | https://github.com/llvm/llvm-project/pull/148048                         |
+| Optionally omit array length expression                     | :good:`done`              | :none:`unclaimed`         | (Parse) https://github.com/llvm/llvm-project/pull/148048,                |
+|                                                             |                           |                           | (Sema) https://github.com/llvm/llvm-project/pull/152786                  |
 +-------------------------------------------------------------+---------------------------+---------------------------+--------------------------------------------------------------------------+
 | Canonical loop sequences                                    | :none:`unclaimed`         | :part:`In Progress`       |                                                                          |
 +-------------------------------------------------------------+---------------------------+---------------------------+--------------------------------------------------------------------------+

clang/docs/ReleaseNotes.rst

@@ -734,6 +734,7 @@ OpenMP Support
   modifier in the ``adjust_args`` clause.
 - Allow array length to be omitted in array section subscript expression.
 - Fixed non-contiguous strided update in the ``omp target update`` directive with the ``from`` clause.
+- Properly handle array section/assumed-size array privatization in C/C++.
 
 Improvements
 ^^^^^^^^^^^^

clang/include/clang/Basic/Attr.td

@@ -979,7 +979,7 @@ def AnalyzerNoReturn : InheritableAttr {
 }
 
 def InferredNoReturn : InheritableAttr {
-  let Spellings = []; 
+  let Spellings = [];
   let SemaHandler = 0;
   let Subjects = SubjectList<[Function], ErrorDiag>;
   let Documentation = [InternalOnly];

clang/include/clang/Basic/BuiltinsAMDGPU.def

@@ -675,6 +675,10 @@ TARGET_BUILTIN(__builtin_amdgcn_flat_load_monitor_b128, "V4iV4i*0Ii", "nc", "gfx
 TARGET_BUILTIN(__builtin_amdgcn_cluster_load_b32, "ii*1Iii", "nc", "gfx1250-insts,wavefrontsize32")
 TARGET_BUILTIN(__builtin_amdgcn_cluster_load_b64, "V2iV2i*1Iii", "nc", "gfx1250-insts,wavefrontsize32")
 TARGET_BUILTIN(__builtin_amdgcn_cluster_load_b128, "V4iV4i*1Iii", "nc", "gfx1250-insts,wavefrontsize32")
+TARGET_BUILTIN(__builtin_amdgcn_cluster_load_async_to_lds_b8, "vc*1c*3IiIii", "nc", "gfx1250-insts,wavefrontsize32")
+TARGET_BUILTIN(__builtin_amdgcn_cluster_load_async_to_lds_b32, "vi*1i*3IiIii", "nc", "gfx1250-insts,wavefrontsize32")
+TARGET_BUILTIN(__builtin_amdgcn_cluster_load_async_to_lds_b64, "vV2i*1V2i*3IiIii", "nc", "gfx1250-insts,wavefrontsize32")
+TARGET_BUILTIN(__builtin_amdgcn_cluster_load_async_to_lds_b128, "vV4i*1V4i*3IiIii", "nc", "gfx1250-insts,wavefrontsize32")
 TARGET_BUILTIN(__builtin_amdgcn_global_load_async_to_lds_b8, "vc*1c*3IiIi", "nc", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_global_load_async_to_lds_b32, "vi*1i*3IiIi", "nc", "gfx1250-insts")
 TARGET_BUILTIN(__builtin_amdgcn_global_load_async_to_lds_b64, "vV2i*1V2i*3IiIi", "nc", "gfx1250-insts")

clang/include/clang/Basic/DiagnosticSemaKinds.td

@@ -10694,7 +10694,7 @@ def warn_dangling_reference_captured_by_unknown : Warning<
 
 // Diagnostics based on the Lifetime safety analysis.
 def warn_lifetime_safety_loan_expires_permissive : Warning<
-   "object whose reference is captured does not live long enough">, 
+   "object whose reference is captured does not live long enough">,
    InGroup<LifetimeSafetyPermissive>, DefaultIgnore;
 def warn_lifetime_safety_loan_expires_strict : Warning<
    "object whose reference is captured may not live long enough">,

clang/include/clang/CIR/Dialect/Builder/CIRBaseBuilder.h

@@ -245,6 +245,11 @@ class CIRBaseBuilderTy : public mlir::OpBuilder {
     return createGetGlobal(global.getLoc(), global);
   }
 
+  /// Create a copy with inferred length.
+  cir::CopyOp createCopy(mlir::Value dst, mlir::Value src) {
+    return cir::CopyOp::create(*this, dst.getLoc(), dst, src);
+  }
+
   cir::StoreOp createStore(mlir::Location loc, mlir::Value val, mlir::Value dst,
                            bool isVolatile = false,
                            mlir::IntegerAttr align = {},

clang/include/clang/CIR/Dialect/IR/CIROps.td

@@ -2451,6 +2451,52 @@ def CIR_CallOp : CIR_CallOpBase<"call", [NoRegionArguments]> {
   ];
 }
 
+//===----------------------------------------------------------------------===//
+// CopyOp
+//===----------------------------------------------------------------------===//
+
+def CIR_CopyOp : CIR_Op<"copy",[
+  SameTypeOperands,
+  DeclareOpInterfaceMethods<PromotableMemOpInterface>
+]> {
+  let summary = "Copies contents from a CIR pointer to another";
+  let description = [{
+    Given two CIR pointers, `src` and `dst`, `cir.copy` will copy the memory
+    pointed by `src` to the memory pointed by `dst`.
+
+    The number of bytes copied is inferred from the pointee type. The pointee
+    type of `src` and `dst` must match and both must implement the
+    `DataLayoutTypeInterface`.
+
+    Examples:
+
+    ```mlir
+      // Copying contents from one record to another:
+      cir.copy %0 to %1 : !cir.ptr<!record_ty>
+    ```
+  }];
+
+  let arguments = (ins
+      Arg<CIR_PointerType, "", [MemWrite]>:$dst,
+      Arg<CIR_PointerType, "", [MemRead]>:$src
+  );
+
+  let assemblyFormat = [{$src `to` $dst
+                        attr-dict `:` qualified(type($dst))
+  }];
+  let hasVerifier = 1;
+
+  let extraClassDeclaration = [{
+    /// Returns the pointer type being copied.
+    cir::PointerType getType() { return getSrc().getType(); }
+
+    /// Returns the number of bytes to be copied.
+    unsigned getLength(const mlir::DataLayout &dt) {
+      return dt.getTypeSize(getType().getPointee());
+    }    
+  }];
+}
+
 //===----------------------------------------------------------------------===//
 // ReturnAddrOp and FrameAddrOp
 //===----------------------------------------------------------------------===//
@@ -3752,12 +3798,23 @@ class CIR_UnaryFPToFPBuiltinOp<string mnemonic, string llvmOpName>
   let llvmOp = llvmOpName;
 }
 
+def CIR_ACosOp : CIR_UnaryFPToFPBuiltinOp<"acos", "ACosOp"> {
+  let summary = "Computes the arcus cosine of the specified value";
+  let description = [{
+    `cir.acos`computes the arcus cosine of a given value and
+    returns a result of the same type.
+
+    Floating-point exceptions are ignored, and it does not set `errno`.
+  }];
+}
+
 def CIR_FAbsOp : CIR_UnaryFPToFPBuiltinOp<"fabs", "FAbsOp"> {
   let summary = "Computes the floating-point absolute value";
   let description = [{
     `cir.fabs` computes the absolute value of a floating-point operand
-    and returns a result of the same type, ignoring floating-point
-    exceptions. It does not set `errno`.
+    and returns a result of the same type.
+
+    Floating-point exceptions are ignored, and it does not set `errno`.
   }];
 }
 

clang/include/clang/CIR/MissingFeatures.h

@@ -168,6 +168,7 @@ struct MissingFeatures {
   // Misc
   static bool abiArgInfo() { return false; }
   static bool addHeapAllocSiteMetadata() { return false; }
+  static bool aggEmitFinalDestCopyRValue() { return false; }
   static bool aggValueSlot() { return false; }
   static bool aggValueSlotAlias() { return false; }
   static bool aggValueSlotDestructedFlag() { return false; }