[PGO] Value profile for size of memory intrinsic calls

This patch annotates the valuesites profile to memory intrinsics.

Differential Revision: http://reviews.llvm.org/D31002

llvm-svn: 298110

llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp

@@ -121,6 +121,13 @@ static cl::opt<unsigned> MaxNumAnnotations(
     cl::desc("Max number of annotations for a single indirect "
              "call callsite"));
 
+// Command line option to set the maximum number of value annotations
+// to write to the metadata for a single memop intrinsic.
+static cl::opt<unsigned> MaxNumMemOPAnnotations(
+    "memop-max-annotations", cl::init(4), cl::Hidden, cl::ZeroOrMore,
+    cl::desc("Max number of preicise value annotations for a single memop"
+             "intrinsic"));
+
 // Command line option to control appending FunctionHash to the name of a COMDAT
 // function. This is to avoid the hash mismatch caused by the preinliner.
 static cl::opt<bool> DoComdatRenaming(
@@ -250,6 +257,7 @@ struct MemIntrinsicVisitor : public InstVisitor<MemIntrinsicVisitor> {
   GlobalVariable *FuncNameVar = nullptr;
   uint64_t FuncHash = 0;
   PGOUseFunc *UseFunc = nullptr;
+  std::vector<Instruction *> Candidates;
 
   MemIntrinsicVisitor(Function &Func) : F(Func) {}
 
@@ -258,6 +266,7 @@ struct MemIntrinsicVisitor : public InstVisitor<MemIntrinsicVisitor> {
     Mode = VM_counting;
     visit(Func);
   }
+
   void instrumentMemIntrinsics(Function &Func, unsigned TotalNC,
                                GlobalVariable *FNV, uint64_t FHash) {
     Mode = VM_instrument;
@@ -267,6 +276,13 @@ struct MemIntrinsicVisitor : public InstVisitor<MemIntrinsicVisitor> {
     visit(Func);
   }
 
+  std::vector<Instruction *> findMemIntrinsics(Function &Func) {
+    Candidates.clear();
+    Mode = VM_annotate;
+    visit(Func);
+    return Candidates;
+  }
+
   // Visit the IR stream and annotate all mem intrinsic call instructions.
   void instrumentOneMemIntrinsic(MemIntrinsic &MI);
   // Visit \p MI instruction and perform tasks according to visit mode.
@@ -432,6 +448,7 @@ template <class Edge, class BBInfo> class FuncPGOInstrumentation {
     NumOfPGOSelectInsts += SIVisitor.getNumOfSelectInsts();
     NumOfPGOMemIntrinsics += MIVisitor.getNumOfMemIntrinsics();
     ValueSites[IPVK_IndirectCallTarget] = findIndirectCallSites(Func);
+    ValueSites[IPVK_MemOPSize] = MIVisitor.findMemIntrinsics(Func);
 
     FuncName = getPGOFuncName(F);
     computeCFGHash();
@@ -1153,7 +1170,8 @@ void MemIntrinsicVisitor::visitMemIntrinsic(MemIntrinsic &MI) {
     instrumentOneMemIntrinsic(MI);
     return;
   case VM_annotate:
-    break;
+    Candidates.push_back(&MI);
+    return;
   }
   llvm_unreachable("Unknown visiting mode");
 }
@@ -1167,9 +1185,7 @@ void PGOUseFunc::annotateValueSites() {
   createPGOFuncNameMetadata(F, FuncInfo.FuncName);
 
   for (uint32_t Kind = IPVK_First; Kind <= IPVK_Last; ++Kind)
-    // TBD: Only handle IPVK_IndirectCallTarget for now.
-    if (Kind == IPVK_IndirectCallTarget)
-      annotateValueSites(Kind);
+    annotateValueSites(Kind);
 }
 
 // Annotate the instructions for a specific value kind.
@@ -1193,7 +1209,8 @@ void PGOUseFunc::annotateValueSites(uint32_t Kind) {
                  << NumValueSites << "\n");
     annotateValueSite(*M, *I, ProfileRecord,
                       static_cast<InstrProfValueKind>(Kind), ValueSiteIndex,
-                      MaxNumAnnotations);
+                      Kind == IPVK_MemOPSize ? MaxNumMemOPAnnotations
+                                             : MaxNumAnnotations);
     ValueSiteIndex++;
   }
 }

llvm/test/Transforms/PGOProfile/Inputs/memop_size_annotation.proftext

@@ -0,0 +1,27 @@
+# IR level Instrumentation Flag
+:ir
+foo
+# Func Hash:
+53929068288
+# Num Counters:
+3
+# Counter Values:
+556
+20
+1
+# Num Value Kinds:
+1
+# ValueKind = IPVK_MemOPSize:
+1
+# NumValueSites:
+1
+9
+7:33
+2:88
+9:72
+4:66
+1:99
+5:55
+6:44
+3:77
+8:22

llvm/test/Transforms/PGOProfile/memop_size_annotation.ll

@@ -0,0 +1,59 @@
+; RUN: llvm-profdata merge %S/Inputs/memop_size_annotation.proftext -o %t.profdata
+; RUN: opt < %s -pgo-instr-use -memop-max-annotations=9 -pgo-test-profile-file=%t.profdata -S | FileCheck %s --check-prefixes=MEMOP_ANNOTATION,MEMOP_ANNOTATION9
+; RUN: opt < %s -passes=pgo-instr-use -memop-max-annotations=9 -pgo-test-profile-file=%t.profdata -S | FileCheck %s --check-prefixes=MEMOP_ANNOTATION,MEMOP_ANNOTATION9
+; RUN: opt < %s -pgo-instr-use -pgo-test-profile-file=%t.profdata -S | FileCheck %s --check-prefixes=MEMOP_ANNOTATION,MEMOP_ANNOTATION4
+; RUN: opt < %s -passes=pgo-instr-use -pgo-test-profile-file=%t.profdata -S | FileCheck %s --check-prefixes=MEMOP_ANNOTATION,MEMOP_ANNOTATION4
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+define void @foo(i8* %dst, i8* %src, i32* %a, i32 %n) {
+entry:
+  br label %for.cond
+
+for.cond:
+  %i.0 = phi i32 [ 0, %entry ], [ %inc5, %for.inc4 ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end6
+
+for.body:
+  br label %for.cond1
+
+for.cond1:
+  %j.0 = phi i32 [ 0, %for.body ], [ %inc, %for.inc ]
+  %idx.ext = sext i32 %i.0 to i64
+  %add.ptr = getelementptr inbounds i32, i32* %a, i64 %idx.ext
+  %0 = load i32, i32* %add.ptr, align 4
+  %cmp2 = icmp slt i32 %j.0, %0
+  br i1 %cmp2, label %for.body3, label %for.end
+
+for.body3:
+  %add = add nsw i32 %i.0, 1
+  %conv = sext i32 %add to i64
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %conv, i32 1, i1 false)
+; MEMOP_ANNOTATION: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %conv, i32 1, i1 false)
+; MEMOP_ANNOTATION-SAME: !prof ![[MEMOP_VALUESITE:[0-9]+]]
+; MEMOP_ANNOTATION9: ![[MEMOP_VALUESITE]] = !{!"VP", i32 1, i64 556, i64 1, i64 99, i64 2, i64 88, i64 3, i64 77, i64 9, i64 72, i64 4, i64 66, i64 5, i64 55, i64 6, i64 44, i64 7, i64 33, i64 8, i64 22}
+; MEMOP_ANNOTATION4: ![[MEMOP_VALUESITE]] = !{!"VP", i32 1, i64 556, i64 1, i64 99, i64 2, i64 88, i64 3, i64 77, i64 9, i64 72}
+  br label %for.inc
+
+for.inc:
+  %inc = add nsw i32 %j.0, 1
+  br label %for.cond1
+
+for.end:
+  br label %for.inc4
+
+for.inc4:
+  %inc5 = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end6:
+  ret void
+}
+
+declare void @llvm.lifetime.start(i64, i8* nocapture)
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i32, i1)
+
+declare void @llvm.lifetime.end(i64, i8* nocapture)