Include static prof data when collecting loop BBs

Summary:
If the programmer adds static profile data to a branch---i.e. uses
"__builtin_expect()" or similar---then we should honor it. Otherwise,
"__builtin_expect()" is ignored in crucial situations. So we trust that
the programmer knows what they're doing until proven wrong.

Subscribers: hiraditya, JDevlieghere, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D74809

1 files changed, 8 insertions, 3 deletions

diff --git a/llvm/lib/CodeGen/MachineBlockPlacement.cpp b/llvm/lib/CodeGen/MachineBlockPlacement.cpp
index 3149151..5ce4390 100644
--- a/llvm/lib/CodeGen/MachineBlockPlacement.cpp
+++ b/llvm/lib/CodeGen/MachineBlockPlacement.cpp
@@ -2506,9 +2506,14 @@ 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. This needs precise profile data and we only do this when
-  // profile data is available.
-  if (F->getFunction().hasProfileData() || ForceLoopColdBlock) {
+  // 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()) {
     BlockFrequency LoopFreq(0);
     for (auto LoopPred : L.getHeader()->predecessors())
       if (!L.contains(LoopPred))