[InstrProf] Fix bug when merging empty profile with multiple threads

[ellis]

Summary

When merging profiles with multiple threads, the mergeWriterContexts() function is used to merge profile data between writers. This must be in sync with loadInput() which merges profiles to a single writer. This diff merges the profile kind correctly in mergeWriterContexts() to fix a subtle bug.

[ellis]

Expand comment.

[phosek]

LGTM

[ellis]

Rebase.

llvm/test/tools/llvm-profdata/merge-incompatible.test

 RUN: not llvm-profdata merge %p/Inputs/fe-basic.proftext %p/Inputs/ir-basic.proftext -o /dev/null 2>&1 | FileCheck %s
 CHECK: ir-basic.proftext: Merge IR generated profile with Clang generated profile.
+
+// We get a slightly different error message when using multiple threads because
+// we do not know which files have incompatible kinds.
+RUN: not llvm-profdata merge %p/Inputs/fe-basic.proftext %p/Inputs/ir-basic.proftext --num-threads=2 -o /dev/null 2>&1 | FileCheck %s --check-prefix=THREADS
+THREADS: unsupported instrumentation profile format version

llvm/test/tools/llvm-profdata/merge_empty_profile.test

 # Tests for merge of empty profile files.
 
 RUN: touch %t_empty.proftext
-RUN: llvm-profdata merge -text -o %t_clang.proftext %t_empty.proftext %p/Inputs/clang_profile.proftext
-RUN: FileCheck --input-file=%t_clang.proftext %s -check-prefix=CLANG_PROF_TEXT
+RUN: llvm-profdata merge -text -o - %t_empty.proftext %p/Inputs/clang_profile.proftext | FileCheck %s -check-prefix=CLANG_PROF_TEXT
+RUN: llvm-profdata merge -text -o - %t_empty.proftext %p/Inputs/clang_profile.proftext --num-threads=2 | FileCheck %s -check-prefix=CLANG_PROF_TEXT
 CLANG_PROF_TEXT: main
 CLANG_PROF_TEXT: 0
 CLANG_PROF_TEXT: 1
 CLANG_PROF_TEXT: 1
 
-RUN: llvm-profdata merge -text -o %t_ir.proftext %t_empty.proftext %p/Inputs/IR_profile.proftext
-RUN: FileCheck --input-file=%t_ir.proftext %s -check-prefix=IR_PROF_TEXT
+RUN: llvm-profdata merge -text -o - %t_empty.proftext %p/Inputs/IR_profile.proftext | FileCheck %s -check-prefix=IR_PROF_TEXT
+RUN: llvm-profdata merge -text -o - %t_empty.proftext %p/Inputs/IR_profile.proftext --num-threads=2 | FileCheck %s -check-prefix=IR_PROF_TEXT
 IR_PROF_TEXT: :ir
 IR_PROF_TEXT: main
 IR_PROF_TEXT: 0
 IR_PROF_TEXT: 1
 IR_PROF_TEXT: 1

llvm/tools/llvm-profdata/llvm-profdata.cpp

 }
 
 /// Merge the \p Src writer context into \p Dst.
 static void mergeWriterContexts(WriterContext *Dst, WriterContext *Src) {
   for (auto &ErrorPair : Src->Errors)
     Dst->Errors.push_back(std::move(ErrorPair));
   Src->Errors.clear();
 
+  if (Error E = Dst->Writer.mergeProfileKind(Src->Writer.getProfileKind()))
+    exitWithError(std::move(E));
+
   Dst->Writer.mergeRecordsFromWriter(std::move(Src->Writer), [&](Error E) {
     instrprof_error IPE = InstrProfError::take(std::move(E));
     std::unique_lock<std::mutex> ErrGuard{Dst->ErrLock};
     bool firstTime = Dst->WriterErrorCodes.insert(IPE).second;
     if (firstTime)
       warn(toString(make_error<InstrProfError>(IPE)));
   });
 }
static void mergeInstrProfile(const WeightedFileVector &Inputs,
 
   std::mutex ErrorLock;
   SmallSet<instrprof_error, 4> WriterErrorCodes;
 
   // If NumThreads is not specified, auto-detect a good default.
   if (NumThreads == 0)
     NumThreads = std::min(hardware_concurrency().compute_thread_count(),
                           unsigned((Inputs.size() + 1) / 2));
-  // FIXME: There's a bug here, where setting NumThreads = Inputs.size() fails
-  // the merge_empty_profile.test because the InstrProfWriter.ProfileKind isn't
-  // merged, thus the emitted file ends up with a PF_Unknown kind.
 
   // Initialize the writer contexts.
   SmallVector<std::unique_ptr<WriterContext>, 4> Contexts;
   for (unsigned I = 0; I < NumThreads; ++I)
     Contexts.emplace_back(std::make_unique<WriterContext>(
         OutputSparse, ErrorLock, WriterErrorCodes));
 
   if (NumThreads == 1) {