Merge upstream LLVM into amd-gfx12

Author: SC llvm team

bolt/include/bolt/Profile/DataAggregator.h

@@ -502,6 +502,9 @@ class DataAggregator : public DataReader {
   /// entries).
   void imputeFallThroughs();
 
+  /// Register profiled functions for lite mode.
+  void registerProfiledFunctions();
+
   /// Debugging dump methods
   void dump() const;
   void dump(const PerfBranchSample &Sample) const;

bolt/lib/Profile/DataAggregator.cpp

@@ -581,6 +581,26 @@ void DataAggregator::imputeFallThroughs() {
     outs() << "BOLT-INFO: imputed " << InferredTraces << " traces\n";
 }
 
+void DataAggregator::registerProfiledFunctions() {
+  DenseSet<uint64_t> Addrs;
+  for (const auto &Trace : llvm::make_first_range(Traces)) {
+    if (Trace.Branch != Trace::FT_ONLY &&
+        Trace.Branch != Trace::FT_EXTERNAL_ORIGIN)
+      Addrs.insert(Trace.Branch);
+    Addrs.insert(Trace.From);
+  }
+
+  for (const auto [PC, _] : BasicSamples)
+    Addrs.insert(PC);
+
+  for (const PerfMemSample &MemSample : MemSamples)
+    Addrs.insert(MemSample.PC);
+
+  for (const uint64_t Addr : Addrs)
+    if (BinaryFunction *Func = getBinaryFunctionContainingAddress(Addr))
+      Func->setHasProfileAvailable();
+}
+
 Error DataAggregator::preprocessProfile(BinaryContext &BC) {
   this->BC = &BC;
 
@@ -603,6 +623,7 @@ Error DataAggregator::preprocessProfile(BinaryContext &BC) {
       exit(0);
   }
 
+  registerProfiledFunctions();
   return Error::success();
 }
 
@@ -1347,10 +1368,6 @@ std::error_code DataAggregator::parseAggregatedLBREntry() {
   }
 
   const uint64_t FromOffset = Addr[0]->Offset;
-  BinaryFunction *FromFunc = getBinaryFunctionContainingAddress(FromOffset);
-  if (FromFunc)
-    FromFunc->setHasProfileAvailable();
-
   int64_t Count = Counters[0];
   int64_t Mispreds = Counters[1];
 
@@ -1361,11 +1378,6 @@ std::error_code DataAggregator::parseAggregatedLBREntry() {
     return std::error_code();
   }
 
-  const uint64_t ToOffset = Addr[1]->Offset;
-  BinaryFunction *ToFunc = getBinaryFunctionContainingAddress(ToOffset);
-  if (ToFunc)
-    ToFunc->setHasProfileAvailable();
-
   /// For fall-through types, adjust locations to match Trace container.
   if (Type == FT || Type == FT_EXTERNAL_ORIGIN || Type == FT_EXTERNAL_RETURN) {
     Addr[2] = Location(Addr[1]->Offset); // Trace To
@@ -1613,9 +1625,6 @@ std::error_code DataAggregator::parseBranchEvents() {
   Traces.reserve(TraceMap.size());
   for (const auto &[Trace, Info] : TraceMap) {
     Traces.emplace_back(Trace, Info);
-    for (const uint64_t Addr : {Trace.Branch, Trace.From})
-      if (BinaryFunction *BF = getBinaryFunctionContainingAddress(Addr))
-        BF->setHasProfileAvailable();
   }
   clear(TraceMap);
 
@@ -1676,9 +1685,6 @@ std::error_code DataAggregator::parseBasicEvents() {
       continue;
     ++NumTotalSamples;
 
-    if (BinaryFunction *BF = getBinaryFunctionContainingAddress(Sample->PC))
-      BF->setHasProfileAvailable();
-
     ++BasicSamples[Sample->PC];
     EventNames.insert(Sample->EventName);
   }
@@ -1716,9 +1722,6 @@ std::error_code DataAggregator::parseMemEvents() {
     if (std::error_code EC = Sample.getError())
       return EC;
 
-    if (BinaryFunction *BF = getBinaryFunctionContainingAddress(Sample->PC))
-      BF->setHasProfileAvailable();
-
     MemSamples.emplace_back(std::move(Sample.get()));
   }
 

clang/lib/Sema/SemaOverload.cpp

@@ -8042,8 +8042,8 @@ static void AddTemplateOverloadCandidateImmediately(
 
     Candidate.IgnoreObjectArgument =
         isa<CXXMethodDecl>(Candidate.Function) &&
-        cast<CXXMethodDecl>(Candidate.Function)
-            ->isImplicitObjectMemberFunction() &&
+        !cast<CXXMethodDecl>(Candidate.Function)
+             ->isExplicitObjectMemberFunction() &&
         !isa<CXXConstructorDecl>(Candidate.Function);
 
     Candidate.ExplicitCallArguments = Args.size();

clang/test/SemaCXX/cxx2b-deducing-this.cpp

@@ -1357,3 +1357,35 @@ void Bar(this int) { // expected-note {{candidate function}}
 }
 
 }
+
+namespace GH147046_regression {
+
+template <typename z> struct ai {
+    ai(z::ah);
+};
+
+template <typename z> struct ak {
+    template <typename am> void an(am, z);
+    template <typename am> static void an(am, ai<z>);
+};
+template <typename> struct ao {};
+
+template <typename ap>
+auto ar(ao<ap> at) -> decltype(ak<ap>::an(at, 0));
+// expected-note@-1 {{candidate template ignored: substitution failure [with ap = GH147046_regression::ay]: no matching function for call to 'an'}}
+
+class aw;
+struct ax {
+    typedef int ah;
+};
+struct ay {
+    typedef aw ah;
+};
+
+ao<ay> az ;
+ai<ax> bd(0);
+void f() {
+    ar(az); // expected-error {{no matching function for call to 'ar'}}
+}
+
+}

lldb/source/Plugins/Process/Linux/NativeRegisterContextLinux_arm.h

@@ -109,7 +109,6 @@ class NativeRegisterContextLinux_arm : public NativeRegisterContextLinux {
                            // occurred.
     lldb::addr_t real_addr; // Address value that should cause target to stop.
     uint32_t control;       // Breakpoint/watchpoint control value.
-    uint32_t refcount;      // Serves as enable/disable and reference counter.
   };
 
   struct DREG m_hbr_regs[16]; // Arm native linux hardware breakpoints

llvm/include/llvm/Support/DebugLog.h

@@ -40,8 +40,8 @@ namespace llvm {
   DEBUGLOG_WITH_STREAM_TYPE_AND_FILE(STREAM, TYPE, __SHORT_FILE__)
 #else
 #define DEBUGLOG_WITH_STREAM_AND_TYPE(STREAM, TYPE)                            \
-  DEBUGLOG_WITH_STREAM_TYPE_AND_FILE(                                          \
-      STREAM, TYPE, ::llvm::impl::LogWithNewline::getShortFileName(__FILE__))
+  DEBUGLOG_WITH_STREAM_TYPE_AND_FILE(STREAM, TYPE,                             \
+                                     ::llvm::impl::getShortFileName(__FILE__))
 #endif
 
 namespace impl {

llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp

@@ -13,7 +13,6 @@
 #include "AMDGPU.h"
 #include "GCNSubtarget.h"
 #include "Utils/AMDGPUBaseInfo.h"
-#include "llvm/CodeGen/TargetPassConfig.h"
 #include "llvm/IR/IntrinsicsAMDGPU.h"
 #include "llvm/IR/IntrinsicsR600.h"
 #include "llvm/Target/TargetMachine.h"

llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp

@@ -708,10 +708,14 @@ static Instruction *shrinkSplatShuffle(TruncInst &Trunc,
   auto *Shuf = dyn_cast<ShuffleVectorInst>(Trunc.getOperand(0));
   if (Shuf && Shuf->hasOneUse() && match(Shuf->getOperand(1), m_Undef()) &&
       all_equal(Shuf->getShuffleMask()) &&
-      Shuf->getType() == Shuf->getOperand(0)->getType()) {
+      ElementCount::isKnownGE(Shuf->getType()->getElementCount(),
+                              cast<VectorType>(Shuf->getOperand(0)->getType())
+                                  ->getElementCount())) {
     // trunc (shuf X, Undef, SplatMask) --> shuf (trunc X), Poison, SplatMask
     // trunc (shuf X, Poison, SplatMask) --> shuf (trunc X), Poison, SplatMask
-    Value *NarrowOp = Builder.CreateTrunc(Shuf->getOperand(0), Trunc.getType());
+    Type *NewTruncTy = Shuf->getOperand(0)->getType()->getWithNewType(
+        Trunc.getType()->getScalarType());
+    Value *NarrowOp = Builder.CreateTrunc(Shuf->getOperand(0), NewTruncTy);
     return new ShuffleVectorInst(NarrowOp, Shuf->getShuffleMask());
   }
 

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

@@ -1363,11 +1363,15 @@ class LoopVectorizationCostModel {
                       TTI.hasActiveVectorLength() && !EnableVPlanNativePath;
     if (EVLIsLegal)
       return;
-    // If for some reason EVL mode is unsupported, fallback to
-    // DataWithoutLaneMask to try to vectorize the loop with folded tail
-    // in a generic way.
-    ChosenTailFoldingStyle = {TailFoldingStyle::DataWithoutLaneMask,
-                              TailFoldingStyle::DataWithoutLaneMask};
+    // If for some reason EVL mode is unsupported, fallback to a scalar epilogue
+    // if it's allowed, or DataWithoutLaneMask otherwise.
+    if (ScalarEpilogueStatus == CM_ScalarEpilogueAllowed ||
+        ScalarEpilogueStatus == CM_ScalarEpilogueNotNeededUsePredicate)
+      ChosenTailFoldingStyle = {TailFoldingStyle::None, TailFoldingStyle::None};
+    else
+      ChosenTailFoldingStyle = {TailFoldingStyle::DataWithoutLaneMask,
+                                TailFoldingStyle::DataWithoutLaneMask};
+
     LLVM_DEBUG(
         dbgs() << "LV: Preference for VP intrinsics indicated. Will "
                   "not try to generate VP Intrinsics "
@@ -4500,19 +4504,17 @@ VectorizationFactor LoopVectorizationPlanner::selectEpilogueVectorizationFactor(
   Type *TCType = Legal->getWidestInductionType();
   const SCEV *RemainingIterations = nullptr;
   unsigned MaxTripCount = 0;
-  if (MainLoopVF.isFixed()) {
-    // TODO: extend to support scalable VFs.
-    const SCEV *TC = vputils::getSCEVExprForVPValue(
-        getPlanFor(MainLoopVF).getTripCount(), SE);
-    assert(!isa<SCEVCouldNotCompute>(TC) &&
-           "Trip count SCEV must be computable");
-    RemainingIterations = SE.getURemExpr(
-        TC, SE.getConstant(TCType, MainLoopVF.getFixedValue() * IC));
-
-    // No iterations left to process in the epilogue.
-    if (RemainingIterations->isZero())
-      return Result;
+  const SCEV *TC =
+      vputils::getSCEVExprForVPValue(getPlanFor(MainLoopVF).getTripCount(), SE);
+  assert(!isa<SCEVCouldNotCompute>(TC) && "Trip count SCEV must be computable");
+  RemainingIterations =
+      SE.getURemExpr(TC, SE.getElementCount(TCType, MainLoopVF * IC));
+
+  // No iterations left to process in the epilogue.
+  if (RemainingIterations->isZero())
+    return Result;
 
+  if (MainLoopVF.isFixed()) {
     MaxTripCount = MainLoopVF.getFixedValue() * IC - 1;
     if (SE.isKnownPredicate(CmpInst::ICMP_ULT, RemainingIterations,
                             SE.getConstant(TCType, MaxTripCount))) {