Revert "Include static prof data when collecting loop BBs"

This reverts commit 129c911.

Due to an internal benchmark regression.

llvm/include/llvm/CodeGen/MachineLoopInfo.h

@@ -67,10 +67,6 @@ class MachineLoop : public LoopBase<MachineBasicBlock, MachineLoop> {
   /// it returns an unknown location.
   DebugLoc getStartLoc() const;
 
-  /// Returns true if a machine loop has blocks that have static profiling
-  /// information---e.g. from '__builtin_expect()'.
-  bool hasStaticProfInfo() const;
-
   void dump() const;
 
 private:

llvm/lib/CodeGen/MachineBlockPlacement.cpp

@@ -2506,14 +2506,9 @@ MachineBlockPlacement::collectLoopBlockSet(const MachineLoop &L) {
   // its frequency and the frequency of the loop block. When it is too small,
   // don't add it to the loop chain. If there are outer loops, then this block
   // will be merged into the first outer loop chain for which this block is not
-  // cold anymore.
-  //
-  // If a block uses static profiling data (e.g. from '__builtin_expect()'),
-  // then the programmer is explicitly telling us which paths are hot and cold.
-  // There's no reason for the compiler to believe otherwise, unless
-  // '-fprofile-use' is specified.
-  if (F->getFunction().hasProfileData() || ForceLoopColdBlock ||
-      L.hasStaticProfInfo()) {
+  // cold anymore. This needs precise profile data and we only do this when
+  // profile data is available.
+  if (F->getFunction().hasProfileData() || ForceLoopColdBlock) {
     BlockFrequency LoopFreq(0);
     for (auto LoopPred : L.getHeader()->predecessors())
       if (!L.contains(LoopPred))

llvm/lib/CodeGen/MachineLoopInfo.cpp

@@ -111,13 +111,6 @@ DebugLoc MachineLoop::getStartLoc() const {
   return DebugLoc();
 }
 
-bool MachineLoop::hasStaticProfInfo() const {
-  return llvm::any_of(blocks(), [](const MachineBasicBlock *MBB){
-    const BasicBlock *BB = MBB->getBasicBlock();
-    return BB && BB->getTerminator()->hasMetadata(LLVMContext::MD_prof);
-  });
-}
-
 MachineBasicBlock *
 MachineLoopInfo::findLoopPreheader(MachineLoop *L,
                                    bool SpeculativePreheader) const {

llvm/test/CodeGen/Hexagon/prof-early-if.ll

@@ -1,8 +1,8 @@
 ; RUN: llc -O2 -march=hexagon < %s | FileCheck %s
 ; Rely on the comments generated by llc. Check that "if.then" was not predicated.
+; CHECK: b5
 ; CHECK: b2
 ; CHECK-NOT: if{{.*}}memd
-; CHECK: b5
 
 %s.0 = type { [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [24 x i32], [3 x i32], [24 x i32], [8 x %s.1], [5 x i32] }
 %s.1 = type { i32, i32 }

llvm/test/CodeGen/X86/block-placement.ll

@@ -1502,9 +1502,9 @@ define i32 @not_rotate_if_extra_branch(i32 %count) {
 ; CHECK: %.header
 ; CHECK: %.middle
 ; CHECK: %.backedge
+; CHECK: %.slow
 ; CHECK: %.bailout
 ; CHECK: %.stop
-; CHECK: %.slow
 .entry:
   %sum.0 = shl nsw i32 %count, 1
   br label %.header

llvm/test/CodeGen/X86/move_latch_to_loop_top.ll

@@ -173,8 +173,8 @@ exit:
 ;CHECK:       %header
 ;CHECK:       %true
 ;CHECK:       %latch
-;CHECK:       %exit
 ;CHECK:       %false
+;CHECK:       %exit
 define i32 @test4(i32 %t, i32* %p) {
 entry:
   br label %header

llvm/test/CodeGen/X86/ragreedy-bug.ll

@@ -10,29 +10,27 @@
 ; Mem-move
 ; CHECK-NEXT: movl
 ; CHECK-NEXT: andl
-; CHECK-NEXT: LBB0
-; CHECK-NEXT: in Loop
 ; CHECK-NEXT: testl
 ; CHECK-NEXT: jne
 ; CHECK: cond.true.i.i217
 ; CHECK-NEXT: in Loop
 ; Mem-move
 ; CHECK-NEXT: movl
 ; CHECK-NEXT: andl
-; CHECK-NEXT: LBB0
-; CHECK-NEXT: in Loop
 ; CHECK-NEXT: testl
 ; CHECK-NEXT: je
 ; CHECK: cond.false.i.i
 ; CHECK: maskrune
 ; CHECK-NEXT: movzbl
 ; CHECK-NEXT: movzbl
-; CHECK-NEXT: jmp
+; CHECK-NEXT: testl
+; CHECK-NEXT: je
 ; CHECK: cond.false.i.i219
 ; CHECK: maskrune
 ; CHECK-NEXT: movzbl
 ; CHECK-NEXT: movzbl
-; CHECK-NEXT: jmp
+; CHECK-NEXT: testl
+; CHECK-NEXT: jne
 
 %struct.List_o_links_struct = type { i32, i32, i32, %struct.List_o_links_struct* }
 %struct.Connector_struct = type { i16, i16, i8, i8, %struct.Connector_struct*, i8* }