[pull] main from llvm:main

.github/workflows/issue-release-workflow.yml

@@ -46,7 +46,7 @@ jobs:
           # If there are multiple /cherry-pick commands they will all be included in the output.
           # This is fine since the github-automation.py script can handle multiple cherry-pick commands.
           validated_comment=$(\
-            grep -o -e '/cherry-pick \(https://github.com/llvm/llvm-project/commit/\)\?[a-f0-9 ]\+' <<< $COMMENT_BODY \
+            grep -o -e '/cherry-pick \(\(https://github.com/llvm/llvm-project/commit/\)\?[a-f0-9]\+ *\)\+' <<< $COMMENT_BODY \
           )
           echo "Validated Comment is:"
           echo ${validated_comment}

.github/workflows/libcxx-pr-benchmark.yml

@@ -48,43 +48,38 @@ jobs:
     permissions:
       pull-requests: write
     steps:
-      - uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1 # v6.3.0
-        with:
-          python-version: '3.14'
-
-      - name: Setup virtual environment
-        run: |
-          python3 -m venv .venv
-          source .venv/bin/activate
-          python -m pip install pygithub==2.8.1
-
       - name: Extract information from the PR
         id: vars
-        env:
-          COMMENT_BODY: ${{ github.event.comment.body }}
-        run: |
-          source .venv/bin/activate
-          cat <<EOF | python >> ${GITHUB_OUTPUT}
-          import github
-          repo = github.Github(auth=github.Auth.Token("${{ github.token }}")).get_repo("${{ github.repository }}")
-          pr = repo.get_pull(${{ github.event.issue.number }})
-          print(f"pr_base={pr.base.sha}")
-          print(f"pr_head={pr.head.sha}")
-          EOF
-          BENCHMARKS=$(echo "$COMMENT_BODY" | sed -nE 's/\/libcxx-bot benchmark (.+)/\1/p')
-          echo "benchmarks=${BENCHMARKS}" >> ${GITHUB_OUTPUT}
+        uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
+        with:
+          script: |
+            const pr = await github.rest.pulls.get({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              pull_number: context.issue.number,
+            });
+            core.setOutput('pr_base', pr.data.base.sha);
+            core.setOutput('pr_head', pr.data.head.sha);
+            const match = context.payload.comment.body.match(/\/libcxx-bot benchmark (.+)/);
+            core.setOutput('benchmarks', match ? match[1] : '');
 
       - name: Update comment with link to the run
-        run: |
-          source .venv/bin/activate
-          cat <<EOF | python
-          import github
-          repo = github.Github(auth=github.Auth.Token("${{ github.token }}")).get_repo("${{ github.repository }}")
-          pr = repo.get_pull(${{ github.event.issue.number }})
-          comment = pr.get_issue_comment(${{ github.event.comment.id }})
-          add_text = "> _Running benchmarks in ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}_"
-          comment.edit('\n\n'.join([comment.body, add_text]))
-          EOF
+        uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
+        with:
+          script: |
+            const { data: comment } = await github.rest.issues.getComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              comment_id: context.payload.comment.id,
+            });
+            const run_url = `${context.serverUrl}/${context.repo.owner}/${context.repo.repo}/actions/runs/${context.runId}`;
+            const note = `> _Running benchmarks in ${run_url}_`;
+            await github.rest.issues.updateComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              comment_id: context.payload.comment.id,
+              body: [comment.body, note].join('\n\n'),
+            });
     outputs:
       pr_base: ${{ steps.vars.outputs.pr_base }}
       pr_head: ${{ steps.vars.outputs.pr_head }}
@@ -139,7 +134,6 @@ jobs:
           python3 -m venv .venv
           source .venv/bin/activate
           python -m pip install -r libcxx/utils/requirements.txt
-          python -m pip install pygithub==2.8.1
 
       - name: Build the baseline and the candidate
         env:
@@ -178,30 +172,54 @@ jobs:
           ./libcxx/utils/compare-benchmarks baseline.lnt candidate.lnt | tee results.txt
 
       - name: Update comment with results
-        run: |
-          source .venv/bin/activate
-          cat <<EOF | python
-          import github
-          repo = github.Github(auth=github.Auth.Token("${{ github.token }}")).get_repo("${{ github.repository }}")
-          pr = repo.get_pull(${{ github.event.issue.number }})
-          comment = pr.get_issue_comment(${{ github.event.comment.id }})
-          with open('results.txt', 'r') as f:
-            benchmark_results = f.read()
-
-          new_comment_text = f"""
-          {comment.body}
-
-          <details>
-          <summary>
-          Benchmark results for ${{ matrix.platform }}:
-          </summary>
-
-          \`\`\`
-          {benchmark_results}
-          \`\`\`
-
-          </details>
-          """
-
-          comment.edit(new_comment_text)
-          EOF
+        uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
+        env:
+          PLATFORM: ${{ matrix.platform }}
+        with:
+          script: |
+            const fs = require('fs');
+            const results = fs.readFileSync('results.txt', 'utf8');
+            const { data: comment } = await github.rest.issues.getComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              comment_id: context.payload.comment.id,
+            });
+            const details = [
+              '<details>',
+              '<summary>',
+              `Benchmark results for ${process.env.PLATFORM}:`,
+              '</summary>',
+              '',
+              '```',
+              results,
+              '```',
+              '',
+              '</details>',
+            ].join('\n');
+            await github.rest.issues.updateComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              comment_id: context.payload.comment.id,
+              body: [comment.body, details].join('\n\n'),
+            });
+
+      - name: Report failure in the comment
+        if: failure()
+        uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
+        env:
+          PLATFORM: ${{ matrix.platform }}
+        with:
+          script: |
+            const { data: comment } = await github.rest.issues.getComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              comment_id: context.payload.comment.id,
+            });
+            const run_url = `${context.serverUrl}/${context.repo.owner}/${context.repo.repo}/actions/runs/${context.runId}`;
+            const note = `> _:x: Benchmarks for ${process.env.PLATFORM} failed. See ${run_url} for details._`;
+            await github.rest.issues.updateComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              comment_id: context.payload.comment.id,
+              body: [comment.body, note].join('\n\n'),
+            });

bolt/lib/Core/BinaryFunction.cpp

@@ -2874,6 +2874,7 @@ struct CFISnapshot {
     case MCCFIInstruction::OpLabel:
     case MCCFIInstruction::OpValOffset:
     case MCCFIInstruction::OpNegateRAState:
+    case MCCFIInstruction::OpLLVMSetRAState:
       reportFatalUsageError("unsupported CFI opcode");
     case MCCFIInstruction::OpLLVMRegisterPair:
     case MCCFIInstruction::OpLLVMVectorRegisters:
@@ -3019,6 +3020,7 @@ struct CFISnapshotDiff : public CFISnapshot {
     case MCCFIInstruction::OpLabel:
     case MCCFIInstruction::OpValOffset:
     case MCCFIInstruction::OpNegateRAState:
+    case MCCFIInstruction::OpLLVMSetRAState:
       reportFatalUsageError("unsupported CFI opcode");
     case MCCFIInstruction::OpLLVMRegisterPair:
     case MCCFIInstruction::OpLLVMVectorRegisters:
@@ -3174,6 +3176,7 @@ BinaryFunction::unwindCFIState(int32_t FromState, int32_t ToState,
     case MCCFIInstruction::OpLabel:
     case MCCFIInstruction::OpValOffset:
     case MCCFIInstruction::OpNegateRAState:
+    case MCCFIInstruction::OpLLVMSetRAState:
       reportFatalUsageError("unsupported CFI opcode");
     case MCCFIInstruction::OpLLVMRegisterPair:
     case MCCFIInstruction::OpLLVMVectorRegisters:

clang/cmake/caches/Fuchsia-stage2.cmake

@@ -205,7 +205,7 @@ if(FUCHSIA_SDK)
   set(FUCHSIA_x86_64-unknown-fuchsia_NAME x64)
   set(FUCHSIA_riscv64-unknown-fuchsia_NAME riscv64)
   foreach(target i386-unknown-fuchsia;x86_64-unknown-fuchsia;aarch64-unknown-fuchsia;arm-unknown-fuchsia;riscv64-unknown-fuchsia)
-    set(FUCHSIA_${target}_COMPILER_FLAGS "--target=${target} -I${FUCHSIA_SDK}/pkg/sync/include -I${FUCHSIA_SDK}/pkg/fdio/include")
+    set(FUCHSIA_${target}_COMPILER_FLAGS "--target=${target} -I${FUCHSIA_SDK}/pkg/sync/include -I${FUCHSIA_SDK}/pkg/fdio/include -fstack-size-section -fexperimental-call-graph-section")
     set(FUCHSIA_${target}_LINKER_FLAGS "-L${FUCHSIA_SDK}/arch/${FUCHSIA_${target}_NAME}/lib")
     set(FUCHSIA_${target}_SYSROOT "${FUCHSIA_SDK}/arch/${FUCHSIA_${target}_NAME}/sysroot")
   endforeach()
@@ -326,7 +326,7 @@ foreach(target armv6m-none-eabi;armv7m-none-eabi;armv7em-none-eabi;armv8m.main-n
     if(${target} STREQUAL "armv8.1m.main-none-eabi")
       set(BUILTINS_${target}_CMAKE_${lang}_local_flags "${BUILTINS_${target}_CMAKE_${lang}_local_flags} -mfloat-abi=hard -march=armv8.1-m.main+mve.fp+fp.dp -mcpu=cortex-m55")
     endif()
-    set(BUILTINS_${target}_CMAKE_${lang}_FLAGS "${BUILTINS_${target}_CMAKE_${lang}_local_flags}" CACHE STRING "")
+    set(BUILTINS_${target}_CMAKE_${lang}_FLAGS "${BUILTINS_${target}_CMAKE_${lang}_local_flags} -fstack-size-section -fexperimental-call-graph-section" CACHE STRING "")
   endforeach()
   foreach(type SHARED;MODULE;EXE)
     set(BUILTINS_${target}_CMAKE_${type}_LINKER_FLAGS "-fuse-ld=lld" CACHE STRING "")
@@ -352,7 +352,7 @@ foreach(target armv6m-none-eabi;armv7m-none-eabi;armv7em-none-eabi;armv8m.main-n
     if(${target} STREQUAL "armv8.1m.main-none-eabi")
       set(RUNTIMES_${target}_CMAKE_${lang}_local_flags "${RUNTIMES_${target}_CMAKE_${lang}_local_flags} -mfloat-abi=hard -march=armv8.1-m.main+mve.fp+fp.dp -mcpu=cortex-m55")
     endif()
-    set(RUNTIMES_${target}_CMAKE_${lang}_FLAGS "${RUNTIMES_${target}_CMAKE_${lang}_local_flags}" CACHE STRING "")
+    set(RUNTIMES_${target}_CMAKE_${lang}_FLAGS "${RUNTIMES_${target}_CMAKE_${lang}_local_flags} -fstack-size-section -fexperimental-call-graph-section" CACHE STRING "")
   endforeach()
   foreach(type SHARED;MODULE;EXE)
     set(RUNTIMES_${target}_CMAKE_${type}_LINKER_FLAGS "-fuse-ld=lld" CACHE STRING "")
@@ -394,7 +394,7 @@ foreach(target riscv32-unknown-elf)
   set(BUILTINS_${target}_CMAKE_SYSROOT "" CACHE STRING "")
   set(BUILTINS_${target}_CMAKE_BUILD_TYPE MinSizeRel CACHE STRING "")
   foreach(lang C;CXX;ASM)
-    set(BUILTINS_${target}_CMAKE_${lang}_FLAGS "--target=${target} -march=rv32imafc -mabi=ilp32f" CACHE STRING "")
+    set(BUILTINS_${target}_CMAKE_${lang}_FLAGS "--target=${target} -march=rv32imafc -mabi=ilp32f -fstack-size-section -fexperimental-call-graph-section" CACHE STRING "")
   endforeach()
   foreach(type SHARED;MODULE;EXE)
     set(BUILTINS_${target}_CMAKE_${type}_LINKER_FLAGS "-fuse-ld=lld" CACHE STRING "")
@@ -410,7 +410,7 @@ foreach(target riscv32-unknown-elf)
   foreach(lang C;CXX;ASM)
     # TODO: The preprocessor defines workaround various issues in libc and libc++ integration.
     # These should be addressed and removed over time.
-    set(RUNTIMES_${target}_CMAKE_${lang}_FLAGS "--target=${target} -march=rv32imafc -mabi=ilp32f -Wno-atomic-alignment" CACHE STRING "")
+    set(RUNTIMES_${target}_CMAKE_${lang}_FLAGS "--target=${target} -march=rv32imafc -mabi=ilp32f -Wno-atomic-alignment -fstack-size-section -fexperimental-call-graph-section" CACHE STRING "")
   endforeach()
   foreach(type SHARED;MODULE;EXE)
     set(RUNTIMES_${target}_CMAKE_${type}_LINKER_FLAGS "-fuse-ld=lld" CACHE STRING "")

clang/include/clang/Basic/DiagnosticSemaKinds.td

@@ -13393,8 +13393,7 @@ def err_builtin_is_within_lifetime_invalid_arg : Error<
 def err_builtin_invalid_arg_type: Error<
   "%ordinal0 argument must be a "
   // First component: scalar or container types
-  "%select{|scalar|vector|matrix|vector of|scalar or vector of|"
-  "scalar, vector, or matrix of}1"
+  "%select{|scalar|vector|matrix|vector of|scalar or vector of}1"
   // A comma after generic vector/matrix types if there are non-empty second
   // and third components, to initiate a list.
   "%plural{[2,3]:%plural{0:|:%plural{0:|:,}2}3|:}1"

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

@@ -2258,6 +2258,9 @@ def CIR_WhileOp : CIR_WhileOpBase<"while"> {
     mlir::Region *maybeGetCleanup() {
       return getCleanup().empty() ? nullptr : &getCleanup();
     }
+    cir::CleanupKindAttr maybeGetCleanupKind() {
+      return getCleanupKindAttr();
+    }
     llvm::SmallVector<mlir::Region *> getRegionsInExecutionOrder() {
       llvm::SmallVector<mlir::Region *> regions{&getCond(), &getBody()};
       if (mlir::Region *cleanup = maybeGetCleanup())
@@ -2386,6 +2389,9 @@ def CIR_ForOp : CIR_LoopOpBase<"for"> {
     mlir::Region *maybeGetCleanup() {
       return getCleanup().empty() ? nullptr : &getCleanup();
     }
+    cir::CleanupKindAttr maybeGetCleanupKind() {
+      return getCleanupKindAttr();
+    }
     llvm::SmallVector<mlir::Region *> getRegionsInExecutionOrder() {
       llvm::SmallVector<mlir::Region *> regions{&getCond(), &getBody(),
                                                 &getStep()};

clang/include/clang/CIR/Interfaces/CIRLoopOpInterface.h

@@ -13,12 +13,13 @@
 #ifndef CLANG_CIR_INTERFACES_CIRLOOPOPINTERFACE_H
 #define CLANG_CIR_INTERFACES_CIRLOOPOPINTERFACE_H
 
-#include "llvm/ADT/APInt.h"
 #include "mlir/IR/BuiltinTypes.h"
 #include "mlir/IR/OpDefinition.h"
 #include "mlir/IR/Operation.h"
 #include "mlir/Interfaces/ControlFlowInterfaces.h"
 #include "mlir/Interfaces/LoopLikeInterface.h"
+#include "clang/CIR/Dialect/IR/CIRAttrs.h"
+#include "llvm/ADT/APInt.h"
 
 using llvm::APInt;
 namespace cir {

clang/include/clang/CIR/Interfaces/CIRLoopOpInterface.td

@@ -58,6 +58,17 @@ def LoopOpInterface : OpInterface<"LoopOpInterface", [
       /*methodBody=*/"",
       /*defaultImplementation=*/"return nullptr;"
     >,
+    InterfaceMethod<[{
+        Returns the cleanup-kind attribute for the loop's per-iteration
+        cleanup region, or a null attribute if the loop has no cleanup kind
+        (e.g. do-while, or a for/while without a cleanup region).
+      }],
+      /*retTy=*/"::cir::CleanupKindAttr",
+      /*methodName=*/"maybeGetCleanupKind",
+      /*args=*/(ins),
+      /*methodBody=*/"",
+      /*defaultImplementation=*/"return {};"
+    >,
     InterfaceMethod<[{
         Returns the first region to be executed in the loop.
       }],

clang/include/clang/Options/Options.td

@@ -9923,6 +9923,9 @@ def dxc_gis : DXCFlag<"Gis">,
 def : Option<["/", "-"], "Qembed_debug", KIND_FLAG>, Group<dxc_Group>,
   Flags<[Ignored]>, Visibility<[DXCOption]>,
   HelpText<"Embed PDB in shader container (ignored)">;
+def dxc_Qpdb_in_private : DXCFlag<"Qpdb_in_private">,
+                          Flags<[HelpHidden]>,
+                          HelpText<"Store PDB in private user data">;
 def spirv : DXCFlag<"spirv">,
   HelpText<"Generate SPIR-V code">;
 def metal : DXCFlag<"metal">, HelpText<"Generate Metal library">;

clang/lib/CIR/CodeGen/CIRGenClass.cpp

@@ -901,6 +901,20 @@ void CIRGenFunction::emitImplicitAssignmentOperatorBody(FunctionArgList &args) {
   assert(!cir::MissingFeatures::incrementProfileCounter());
   assert(!cir::MissingFeatures::runCleanupsScope());
 
+  // A defaulted union copy/move assignment has an empty synthesized body:
+  // Sema skips union fields (the FIXME in SemaDeclCXX::buildSingleCopyAssign),
+  // so there is no AST expression for the implied whole-object memcpy.
+  // Emitting that body would silently drop the copy, so report NYI instead.
+  // Struct/array memcpy-equivalent assignments carry the implicit memberwise
+  // copies in the AST (per-field assignment expressions, or a builtin memcpy
+  // call for array members) and lower correctly through the loop below.
+  if (assignOp->isMemcpyEquivalentSpecialMember(getContext()) &&
+      assignOp->getParent()->isUnion()) {
+    cgm.errorNYI(assignOp->getSourceRange(),
+                 "defaulted union copy/move assignment operator");
+    return;
+  }
+
   // Classic codegen uses a special class to attempt to replace member
   // initializers with memcpy. We could possibly defer that to the
   // lowering or optimization phases to keep the memory accesses more

clang/lib/CIR/CodeGen/CIRGenExprConstant.cpp

@@ -1309,9 +1309,10 @@ mlir::Attribute ConstantEmitter::emitForMemory(CIRGenModule &cgm,
     cgm.errorNYI("emitForMemory: zero-extend HLSL bool vectors");
   }
 
-  if (destType->isBitIntType()) {
-    cgm.errorNYI("emitForMemory: _BitInt type");
-  }
+  // CIR represents source types as literally as possible.  Some types, such as
+  // bool and _BitInt(N), are kept at their literal width here and expanded to
+  // their wider "in memory" types during lowering to the LLVM dialect, so the
+  // constant is already in the right form and needs no adjustment.
 
   return c;
 }

clang/lib/CIR/CodeGen/CIRGenOpenACCRecipe.cpp

@@ -155,10 +155,11 @@ mlir::Value OpenACCRecipeBuilderBase::makeBoundsAlloca(
   // Collect the 'do we have any allocas needed after this type' list.
   llvm::SmallVector<bool> allocasLeftArr;
   llvm::ArrayRef<QualType> resultTypes = boundTypes.drop_front();
-  std::transform_inclusive_scan(
-      resultTypes.begin(), resultTypes.end(),
-      std::back_inserter(allocasLeftArr), std::plus<bool>{},
-      [](QualType ty) { return !ty->isConstantArrayType(); }, false);
+  bool accumulator = false;
+  for (QualType ty : resultTypes) {
+    accumulator = accumulator || !ty->isConstantArrayType();
+    allocasLeftArr.push_back(accumulator);
+  }
 
   // Keep track of the number of 'elements' that we're allocating. Individual
   // allocas should multiply this by the size of its current allocation.

clang/lib/CIR/Dialect/Transforms/CallConvLoweringPass.cpp

@@ -115,13 +115,13 @@ static bool isSupportedType(mlir::Type ty) {
     // Unions and packed / padded records each need classification this bridge
     // does not implement (a union widen fixup and pad-aware eightbyte
     // classification), so reject them here and report NYI rather than
-    // misclassify.  Empty-for-ABI records classify as Ignore, which is also
-    // deferred: a C empty struct is a zero-field record, and CIRGen lays out
-    // an empty C++ class as a single padded byte (caught by the padded check).
+    // misclassify.  A zero-field record (a C empty struct) classifies as
+    // Ignore and is dropped from the lowered signature.  CIRGen lays out an
+    // empty C++ class as a single padded byte, which the padded check rejects.
     // A real one-byte struct such as `{char[1]}` has a field and is not
     // padded, so it is classified normally.
     if (recTy.isUnion() || !recTy.isComplete() || recTy.getPacked() ||
-        recTy.getPadded() || recTy.getMembers().empty())
+        recTy.getPadded())
       return false;
     return llvm::all_of(recTy.getMembers(),
                         [](mlir::Type m) { return isSupportedType(m); });
@@ -257,8 +257,8 @@ static const llvm::abi::Type *mapCIRType(mlir::Type type,
 /// Indirect: an aggregate that does not fit in registers is passed via a
 /// pointer (sret for returns, byval for arguments).
 ///
-/// Ignore: a void return has no register or stack slot.  (Empty-for-ABI
-/// records are rejected by isSupportedType, so they never reach here.)
+/// Ignore: a void return has no register or stack slot, and a zero-field
+/// (empty) record is dropped from the signature.
 static std::optional<ArgClassification>
 convertABIArgInfo(const llvm::abi::ArgInfo &info, MLIRContext *ctx,
                   mlir::Type origTy) {