Merge branch 'llvm:main' into stacktrace23

Author: elsteveogrande

clang/docs/ReleaseNotes.rst

@@ -58,6 +58,8 @@ C++ Specific Potentially Breaking Changes
 - The type trait builtin ``__is_referenceable`` has been removed, since it has
   very few users and all the type traits that could benefit from it in the
   standard library already have their own bespoke builtins.
+- A workaround for libstdc++4.7 has been removed. Note that 4.8.3 remains the oldest
+  supported libstdc++ version.
 
 ABI Changes in This Version
 ---------------------------
@@ -118,6 +120,9 @@ C++23 Feature Support
 
 C++20 Feature Support
 ^^^^^^^^^^^^^^^^^^^^^
+- Fixed a crash with a defaulted spaceship (``<=>``) operator when the class
+  contains a member declaration of vector type. Vector types cannot yet be
+  compared directly, so this causes the operator to be deleted. (#GH137452)
 
 C++17 Feature Support
 ^^^^^^^^^^^^^^^^^^^^^
@@ -520,9 +525,10 @@ Improvements to Clang's diagnostics
 
 - Fixed a duplicate diagnostic when performing typo correction on function template
   calls with explicit template arguments. (#GH139226)
-
-- An error is now emitted when OpenMP ``collapse`` and ``ordered`` clauses have an
-  argument larger than what can fit within a 64-bit integer.
+  
+- Explanatory note is printed when ``assert`` fails during evaluation of a
+  constant expression. Prior to this, the error inaccurately implied that assert
+  could not be used at all in a constant expression (#GH130458)
 
 - A new off-by-default warning ``-Wms-bitfield-padding`` has been added to alert to cases where bit-field
   packing may differ under the MS struct ABI (#GH117428).
@@ -586,6 +592,8 @@ Bug Fixes in This Version
   ``#include`` directive. (#GH138094)
 - Fixed a crash during constant evaluation involving invalid lambda captures
   (#GH138832)
+- Fixed a crash when instantiating an invalid dependent friend template specialization.
+  (#GH139052)
 - Fixed a crash with an invalid member function parameter list with a default
   argument which contains a pragma. (#GH113722)
 - Fixed assertion failures when generating name lookup table in modules. (#GH61065, #GH134739)
@@ -935,6 +943,12 @@ OpenMP Support
 - Fixed a crashing bug with a malformed ``cancel`` directive. (#GH139360)
 - Fixed a crashing bug with ``omp distribute dist_schedule`` if the argument to
   ``dist_schedule`` was not strictly positive. (#GH139266)
+- Fixed two crashing bugs with a malformed ``metadirective`` directive. One was
+  a crash if the next token after ``metadirective`` was a paren, bracket, or
+  brace. The other was if the next token after the meta directive was not an
+  open parenthesis. (#GH139665)
+- An error is now emitted when OpenMP ``collapse`` and ``ordered`` clauses have
+  an argument larger than what can fit within a 64-bit integer.
 
 Improvements
 ^^^^^^^^^^^^

clang/docs/UsersManual.rst

@@ -2792,6 +2792,9 @@ usual build cycle when using sample profilers for optimization:
 
      $ llvm-profgen --binary=./code --output=code.prof --perfdata=perf.data
 
+   Please note, ``perf.data`` must be collected with ``-b`` flag to Linux ``perf``
+   for the above step to work.
+
    When using SEP the output is in the textual format corresponding to
    ``llvm-profgen --perfscript``. For example:
 

clang/include/clang/AST/Expr.h

@@ -4566,9 +4566,11 @@ class ShuffleVectorExpr : public Expr {
 
   void setExprs(const ASTContext &C, ArrayRef<Expr *> Exprs);
 
-  llvm::APSInt getShuffleMaskIdx(const ASTContext &Ctx, unsigned N) const {
+  llvm::APSInt getShuffleMaskIdx(unsigned N) const {
     assert((N < NumExprs - 2) && "Shuffle idx out of range!");
-    return getExpr(N+2)->EvaluateKnownConstInt(Ctx);
+    assert(isa<ConstantExpr>(getExpr(N + 2)) &&
+           "Index expression must be a ConstantExpr");
+    return cast<ConstantExpr>(getExpr(N + 2))->getAPValueResult().getInt();
   }
 
   // Iterators

clang/include/clang/Basic/BuiltinsNVPTX.td

@@ -620,6 +620,12 @@ def __nvvm_e2m3x2_to_f16x2_rn_relu : NVPTXBuiltinSMAndPTX<"_Vector<2, __fp16>(sh
 def __nvvm_e3m2x2_to_f16x2_rn : NVPTXBuiltinSMAndPTX<"_Vector<2, __fp16>(short)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
 def __nvvm_e3m2x2_to_f16x2_rn_relu : NVPTXBuiltinSMAndPTX<"_Vector<2, __fp16>(short)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
 
+def __nvvm_ff_to_e2m1x2_rn_satfinite : NVPTXBuiltinSMAndPTX<"short(float, float)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
+def __nvvm_ff_to_e2m1x2_rn_relu_satfinite : NVPTXBuiltinSMAndPTX<"short(float, float)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
+
+def __nvvm_e2m1x2_to_f16x2_rn : NVPTXBuiltinSMAndPTX<"_Vector<2, __fp16>(short)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
+def __nvvm_e2m1x2_to_f16x2_rn_relu : NVPTXBuiltinSMAndPTX<"_Vector<2, __fp16>(short)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
+
 def __nvvm_ff_to_ue8m0x2_rz : NVPTXBuiltinSMAndPTX<"short(float, float)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
 def __nvvm_ff_to_ue8m0x2_rz_satfinite : NVPTXBuiltinSMAndPTX<"short(float, float)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;
 def __nvvm_ff_to_ue8m0x2_rp : NVPTXBuiltinSMAndPTX<"short(float, float)", SM<"100a", [SM_101a, SM_120a]>, PTX86>;

clang/include/clang/Basic/BuiltinsWebAssembly.def

@@ -192,6 +192,7 @@ TARGET_BUILTIN(__builtin_wasm_replace_lane_f16x8, "V8hV8hIif", "nc", "fp16")
 // in which case the argument spec (second argument) is unused.
 
 TARGET_BUILTIN(__builtin_wasm_ref_null_extern, "i", "nct", "reference-types")
+TARGET_BUILTIN(__builtin_wasm_ref_is_null_extern, "ii", "nct", "reference-types")
 
 // A funcref represented as a function pointer with the funcref attribute
 // attached to the type, therefore SemaChecking will check for the right

clang/include/clang/Basic/DiagnosticSemaKinds.td

@@ -164,6 +164,8 @@ def err_ice_too_large : Error<
   "integer constant expression evaluates to value %0 that cannot be "
   "represented in a %1-bit %select{signed|unsigned}2 integer type">;
 def err_expr_not_string_literal : Error<"expression is not a string literal">;
+def note_constexpr_assert_failed : Note<
+  "assertion failed during evaluation of constant expression">;
 
 // Semantic analysis of constant literals.
 def ext_predef_outside_function : Warning<
@@ -10183,6 +10185,9 @@ def note_defaulted_comparison_no_viable_function_synthesized : Note<
 def note_defaulted_comparison_not_rewritten_callee : Note<
   "defaulted %0 is implicitly deleted because this non-rewritten comparison "
   "function would be the best match for the comparison">;
+def note_defaulted_comparison_vector_types : Note<
+  "defaulted %0 is implicitly deleted because defaulted comparison of vector "
+  "types is not supported">;
 def note_defaulted_comparison_not_rewritten_conversion : Note<
   "defaulted %0 is implicitly deleted because a builtin comparison function "
   "using this conversion would be the best match for the comparison">;
@@ -13012,6 +13017,8 @@ def err_wasm_reftype_multidimensional_array : Error<
   "multi-dimensional arrays of WebAssembly references are not allowed">;
 def err_wasm_builtin_arg_must_be_table_type : Error <
   "%ordinal0 argument must be a WebAssembly table">;
+def err_wasm_builtin_arg_must_be_externref_type : Error <
+  "%ordinal0 argument must be an externref">;
 def err_wasm_builtin_arg_must_match_table_element_type : Error <
   "%ordinal0 argument must match the element type of the WebAssembly table in the %ordinal1 argument">;
 def err_wasm_builtin_arg_must_be_integer_type : Error <