Reprioritize tests for tail duplication to be aggressive about indirect

branches even when optimizing for code size.  Unless we find evidence to the
contrary in the future, the special treatment for indirect branches does not
have a significant effect on code size, and performance still matters with -Os.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90147 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/CodeGen/TailDuplication.cpp b/lib/CodeGen/TailDuplication.cpp
index 68fa2099c814f02350d7561ebe7cc48765cdcf80..9c0b596c33f5a6dddc7637507dad6e13861e3fdf 100644 (file)
--- a/lib/CodeGen/TailDuplication.cpp
+++ b/lib/CodeGen/TailDuplication.cpp
@@ -116,14 +116,14 @@ bool TailDuplicatePass::TailDuplicate(MachineBasicBlock *TailBB,
   // duplicate only one, because one branch instruction can be eliminated to
   // compensate for the duplication.
   unsigned MaxDuplicateCount;
-  if (MF.getFunction()->hasFnAttr(Attribute::OptimizeForSize))
-    MaxDuplicateCount = 1;
-  else if (!TailBB->empty() && TailBB->back().getDesc().isIndirectBranch())
+  if (!TailBB->empty() && TailBB->back().getDesc().isIndirectBranch())
     // If the target has hardware branch prediction that can handle indirect
     // branches, duplicating them can often make them predictable when there
     // are common paths through the code.  The limit needs to be high enough
     // to allow undoing the effects of tail merging.
     MaxDuplicateCount = 20;
+  else if (MF.getFunction()->hasFnAttr(Attribute::OptimizeForSize))
+    MaxDuplicateCount = 1;
   else
     MaxDuplicateCount = TailDuplicateSize;