[𝘀𝗽𝗿] initial versionusers/aaupov/spr/boltnfc-register-profiled-functions-once-1

Created using spr 1.3.4

4 files changed, 42 insertions, 38 deletions

diff --git a/bolt/include/bolt/Profile/DataAggregator.h b/bolt/include/bolt/Profile/DataAggregator.h
index db0f690..cb1b87f 100644
--- a/bolt/include/bolt/Profile/DataAggregator.h
+++ b/bolt/include/bolt/Profile/DataAggregator.h
@@ -502,6 +502,9 @@ private:
   /// entries).
   void imputeFallThroughs();
 
+  /// Register profiled functions for lite mode.
+  void registerProfiledFunctions();
+
   /// Debugging dump methods
   void dump() const;
   void dump(const PerfBranchSample &Sample) const;
diff --git a/bolt/lib/Profile/DataAggregator.cpp b/bolt/lib/Profile/DataAggregator.cpp
index 905728de..c13fa6d 100644
--- a/bolt/lib/Profile/DataAggregator.cpp
+++ b/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();
 }
 
@@ -906,11 +927,10 @@ DataAggregator::getFallthroughsInTrace(BinaryFunction &BF, const Trace &Trace,
   if (BF.isPseudo())
     return Branches;
 
-  if (!BF.isSimple())
+  // Can only record traces in CFG state
+  if (!BF.hasCFG())
     return std::nullopt;
 
-  assert(BF.hasCFG() && "can only record traces in CFG state");
-
   const BinaryBasicBlock *FromBB = BF.getBasicBlockContainingOffset(From);
   const BinaryBasicBlock *ToBB = BF.getBasicBlockContainingOffset(To);
 
@@ -1348,10 +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];
 
@@ -1362,11 +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
@@ -1614,9 +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);
 
@@ -1677,9 +1685,6 @@ std::error_code DataAggregator::parseBasicEvents() {
       continue;
     ++NumTotalSamples;
 
-    if (BinaryFunction *BF = getBinaryFunctionContainingAddress(Sample->PC))
-      BF->setHasProfileAvailable();
-
     ++BasicSamples[Sample->PC];
     EventNames.insert(Sample->EventName);
   }
@@ -1717,9 +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()));
   }
 
diff --git a/bolt/lib/Rewrite/RewriteInstance.cpp b/bolt/lib/Rewrite/RewriteInstance.cpp
index f1b94e4..9f243a1 100644
--- a/bolt/lib/Rewrite/RewriteInstance.cpp
+++ b/bolt/lib/Rewrite/RewriteInstance.cpp
@@ -714,21 +714,6 @@ Error RewriteInstance::run() {
 
   preprocessProfileData();
 
-  // Skip disassembling if we have a translation table and we are running an
-  // aggregation job.
-  if (opts::AggregateOnly && BAT->enabledFor(InputFile)) {
-    // YAML profile in BAT mode requires CFG for .bolt.org.text functions
-    if (!opts::SaveProfile.empty() ||
-        opts::ProfileFormat == opts::ProfileFormatKind::PF_YAML) {
-      selectFunctionsToProcess();
-      disassembleFunctions();
-      processMetadataPreCFG();
-      buildFunctionsCFG();
-    }
-    processProfileData();
-    return Error::success();
-  }
-
   selectFunctionsToProcess();
 
   readDebugInfo();
diff --git a/bolt/test/X86/unclaimed-jt-entries.s b/bolt/test/X86/unclaimed-jt-entries.s
index 1102e4a..31b72c4 100644
--- a/bolt/test/X86/unclaimed-jt-entries.s
+++ b/bolt/test/X86/unclaimed-jt-entries.s
@@ -18,6 +18,18 @@
 
 # RUN: llvm-mc -filetype=obj -triple x86_64-unknown-unknown %s -o %t.o
 # RUN: %clang %cflags -no-pie %t.o -o %t.exe -Wl,-q
+
+## Check that non-simple function profile is emitted in perf2bolt mode
+# RUN: link_fdata %s %t.exe %t.pa PREAGG
+# RUN: llvm-strip -N L5 -N L5_ret %t.exe
+# RUN: perf2bolt %t.exe -p %t.pa --pa -o %t.fdata -strict=0 -print-profile \
+# RUN:   -print-only=main | FileCheck %s --check-prefix=CHECK-P2B
+# CHECK-P2B: PERF2BOLT: traces mismatching disassembled function contents: 0
+# CHECK-P2B: Binary Function "main"
+# CHECK-P2B: IsSimple : 0
+# RUN: FileCheck %s --input-file %t.fdata --check-prefix=CHECK-FDATA
+# CHECK-FDATA: 1 main 0 1 main 7 0 1
+
 # RUN: llvm-bolt %t.exe -v=1 -o %t.out 2>&1 | FileCheck %s
 
 # CHECK: BOLT-WARNING: unclaimed data to code reference (possibly an unrecognized jump table entry) to .Ltmp[[#]] in main
@@ -33,8 +45,10 @@
   .size main, .Lend-main
 main:
   jmp *L4-24(,%rdi,8)
-.L5:
+# PREAGG: T #main# #L5# #L5_ret# 1
+L5:
   movl $4, %eax
+L5_ret:
   ret
 .L9:
   movl $2, %eax
@@ -58,7 +72,7 @@ L4:
   .quad .L3
   .quad .L6
   .quad .L3
-  .quad .L5
+  .quad L5
   .quad .L3
   .quad .L3
   .quad .L3