[profcheck] Add unknown branch weights to expanded cmpxchg loop.

Author: Jin Huang

llvm/lib/CodeGen/AtomicExpandPass.cpp

@@ -38,6 +38,7 @@
 #include "llvm/IR/MDBuilder.h"
 #include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Module.h"
+#include "llvm/IR/ProfDataUtils.h"
 #include "llvm/IR/Type.h"
 #include "llvm/IR/User.h"
 #include "llvm/IR/Value.h"
@@ -1259,8 +1260,7 @@ Value *AtomicExpandImpl::insertRMWLLSCLoop(
   BasicBlock *BB = Builder.GetInsertBlock();
   Function *F = BB->getParent();
 
-  assert(AddrAlign >=
-             F->getDataLayout().getTypeStoreSize(ResultTy) &&
+  assert(AddrAlign >= F->getDataLayout().getTypeStoreSize(ResultTy) &&
          "Expected at least natural alignment at this point.");
 
   // Given: atomicrmw some_op iN* %addr, iN %incr ordering
@@ -1680,7 +1680,12 @@ Value *AtomicExpandImpl::insertRMWCmpXchgLoop(
 
   Loaded->addIncoming(NewLoaded, LoopBB);
 
-  Builder.CreateCondBr(Success, ExitBB, LoopBB);
+  Instruction *CondBr = Builder.CreateCondBr(Success, ExitBB, LoopBB);
+
+  // Atomic RMW expands to a cmpxchg loop, Since precise branch weights
+  // cannot be easily determined here, we mark the branch as "unknown" (50/50)
+  // to prevent misleading optimizations.
+  setExplicitlyUnknownBranchWeightsIfProfiled(*CondBr, *F, DEBUG_TYPE);
 
   Builder.SetInsertPoint(ExitBB, ExitBB->begin());
   return NewLoaded;

llvm/test/Transforms/AtomicExpand/AArch64/atomicrmw-fp.ll

@@ -1,8 +1,9 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt -S -mtriple=aarch64-linux-gnu -passes=atomic-expand %s | FileCheck %s
 
-define float @test_atomicrmw_fadd_f32(ptr %ptr, float %value) {
+define float @test_atomicrmw_fadd_f32(ptr %ptr, float %value) !prof !0 {
 ; CHECK-LABEL: @test_atomicrmw_fadd_f32(
+; CHECK: !prof [[PROF_0:![0-9]+]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = load float, ptr [[PTR:%.*]], align 4
 ; CHECK-NEXT:    br label [[ATOMICRMW_START:%.*]]
 ; CHECK:       atomicrmw.start:
@@ -14,7 +15,7 @@ define float @test_atomicrmw_fadd_f32(ptr %ptr, float %value) {
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP4]], 1
 ; CHECK-NEXT:    [[NEWLOADED:%.*]] = extractvalue { i32, i1 } [[TMP4]], 0
 ; CHECK-NEXT:    [[TMP5]] = bitcast i32 [[NEWLOADED]] to float
-; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
+; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]], !prof [[PROF_1:[0-9]+]]
 ; CHECK:       atomicrmw.end:
 ; CHECK-NEXT:    ret float [[TMP5]]
 ;
@@ -336,3 +337,7 @@ define <2 x half> @atomicrmw_fminimum_2_x_half(ptr %ptr, <2 x half> %val) {
   %res = atomicrmw fminimum ptr %ptr, <2 x half> %val seq_cst
   ret <2 x half> %res
 }
+
+!0 = !{!"function_entry_count", i64 1000}
+; CHECK: [[PROF_0]] = !{!"function_entry_count", i64 1000}
+; CHECK: [[PROF_1]] = !{!"unknown", !"atomic-expand"}