Fix the InstCombine to reserve the VP metadata and sets correct call count.

Summary: Currently the VP metadata was dropped when InstCombine converts a call to direct call. This patch converts the VP metadata to branch_weights so that its hotness is recorded.

Reviewers: eraman, davidxl

Reviewed By: davidxl

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D31344

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@299228 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/IR/Metadata.cpp b/lib/IR/Metadata.cpp
index ec96b0fcc9367c3ff7a0a027ff0330738f7da3f9..7228de3d23702d140e2254f86a24f25937fd1caf 100644 (file)
--- a/lib/IR/Metadata.cpp
+++ b/lib/IR/Metadata.cpp
@@ -1337,17 +1337,26 @@ bool Instruction::extractProfTotalWeight(uint64_t &TotalVal) const {
     return false;
 
   auto *ProfDataName = dyn_cast<MDString>(ProfileData->getOperand(0));
-  if (!ProfDataName || !ProfDataName->getString().equals("branch_weights"))
+  if (!ProfDataName)
     return false;
 
-  TotalVal = 0;
-  for (unsigned i = 1; i < ProfileData->getNumOperands(); i++) {
-    auto *V = mdconst::dyn_extract<ConstantInt>(ProfileData->getOperand(i));
-    if (!V)
-      return false;
-    TotalVal += V->getValue().getZExtValue();
+  if (ProfDataName->getString().equals("branch_weights")) {
+    TotalVal = 0;
+    for (unsigned i = 1; i < ProfileData->getNumOperands(); i++) {
+      auto *V = mdconst::dyn_extract<ConstantInt>(ProfileData->getOperand(i));
+      if (!V)
+        return false;
+      TotalVal += V->getValue().getZExtValue();
+    }
+    return true;
+  } else if (ProfDataName->getString().equals("VP") &&
+             ProfileData->getNumOperands() > 3) {
+    TotalVal = mdconst::dyn_extract<ConstantInt>(ProfileData->getOperand(2))
+                   ->getValue()
+                   .getZExtValue();
+    return true;
   }
-  return true;
+  return false;
 }
 
 void Instruction::clearMetadataHashEntries() {

diff --git a/lib/Transforms/InstCombine/InstCombineCalls.cpp b/lib/Transforms/InstCombine/InstCombineCalls.cpp
index f3fc1e2d6316db6bdbba3ef40915bb3a3f832831..71ca14dee58473a33a1c509d803617dc74724874 100644 (file)
--- a/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -4143,6 +4143,12 @@ bool InstCombiner::transformConstExprCastCall(CallSite CS) {
     CallInst *CI = cast<CallInst>(Caller);
     NC = Builder->CreateCall(Callee, Args, OpBundles);
     NC->takeName(CI);
+    // Preserve the weight metadata for the new call instruction. The metadata
+    // is used by SamplePGO to check callsite's hotness.
+    uint64_t W;
+    if (CI->extractProfTotalWeight(W))
+      NC->setProfWeight(W);
+
     cast<CallInst>(NC)->setTailCallKind(CI->getTailCallKind());
     cast<CallInst>(NC)->setCallingConv(CI->getCallingConv());
     cast<CallInst>(NC)->setAttributes(NewCallerPAL);

diff --git a/test/Transforms/InstCombine/cast-call-combine-prof.ll b/test/Transforms/InstCombine/cast-call-combine-prof.ll
new file mode 100644 (file)
index 0000000..e48c14c
--- /dev/null
+++ b/test/Transforms/InstCombine/cast-call-combine-prof.ll
@@ -0,0 +1,38 @@
+; RUN: opt -instcombine -inline -S -inline-threshold=0 -hot-callsite-threshold=100 < %s | FileCheck %s
+; Checks if VP profile is used for hotness checks in inlining after instcombine
+; converted the call to a direct call.
+
+declare void @bar(i16 *)
+
+define void @foo(i16* %a) {
+  call void @bar(i16* %a)
+  call void @bar(i16* %a)
+  ret void
+}
+
+; CHECK-LABEL: @test()
+; CHECK-NEXT: call void @bar
+; CHECK-NEXT: call void @bar
+define void @test() {
+  call void bitcast (void (i16*)* @foo to void (i8*)*) (i8* null), !prof !0
+  ret void
+}
+
+!0 = !{!"VP", i32 0, i64 2000, i64 -3913987384944532146, i64 2000}
+
+!llvm.module.flags = !{!1}
+
+!1 = !{i32 1, !"ProfileSummary", !2}
+!2 = !{!3, !4, !5, !6, !7, !8, !9, !10}
+!3 = !{!"ProfileFormat", !"InstrProf"}
+!4 = !{!"TotalCount", i64 10000}
+!5 = !{!"MaxCount", i64 1000}
+!6 = !{!"MaxInternalCount", i64 1}
+!7 = !{!"MaxFunctionCount", i64 1000}
+!8 = !{!"NumCounts", i64 3}
+!9 = !{!"NumFunctions", i64 3}
+!10 = !{!"DetailedSummary", !11}
+!11 = !{!12, !13, !14}
+!12 = !{i32 10000, i64 1000, i32 1}
+!13 = !{i32 999000, i64 1000, i32 1}
+!14 = !{i32 999999, i64 1, i32 2}