diff options
| author | Benjamin Kramer <benny.kra@googlemail.com> | 2012-05-05 12:49:14 +0000 |
|---|---|---|
| committer | Benjamin Kramer <benny.kra@googlemail.com> | 2012-05-05 12:49:14 +0000 |
| commit | aaf723dd2bccc052d2dd28e3cc4db76f2a3e2fb0 (patch) | |
| tree | 009d318741a34f16b1136257e05c64d7141facef /lib/Target/X86/X86ISelLowering.cpp | |
| parent | af878315192a9fa5b534364e327c24aeb8d73b5a (diff) | |
Add a new target hook "predictableSelectIsExpensive".
This will be used to determine whether it's profitable to turn a select into a
branch when the branch is likely to be predicted.
Currently enabled for everything but Atom on X86 and Cortex-A9 devices on ARM.
I'm not entirely happy with the name of this flag, suggestions welcome ;)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@156233 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/X86/X86ISelLowering.cpp')
| -rw-r--r-- | lib/Target/X86/X86ISelLowering.cpp | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp index 7872a492f8..24d95a9977 100644 --- a/lib/Target/X86/X86ISelLowering.cpp +++ b/lib/Target/X86/X86ISelLowering.cpp @@ -1243,6 +1243,9 @@ X86TargetLowering::X86TargetLowering(X86TargetMachine &TM) setPrefLoopAlignment(4); // 2^4 bytes. benefitFromCodePlacementOpt = true; + // Predictable cmov don't hurt on atom because it's in-order. + predictableSelectIsExpensive = !Subtarget->isAtom(); + setPrefFunctionAlignment(4); // 2^4 bytes. } |