Prefer cheap registers for busy live ranges.

On the x86-64 and thumb2 targets, some registers are more expensive to encode
than others in the same register class.

Add a CostPerUse field to the TableGen register description, and make it
available from TRI->getCostPerUse. This represents the cost of a REX prefix or a
32-bit instruction encoding required by choosing a high register.

Teach the greedy register allocator to prefer cheap registers for busy live
ranges (as indicated by spill weight).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129864 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 3 insertions, 0 deletions

diff --git a/lib/Target/ARM/ARMRegisterInfo.td b/lib/Target/ARM/ARMRegisterInfo.td
index 22d15b572d..7d57b496fb 100644
--- a/lib/Target/ARM/ARMRegisterInfo.td
+++ b/lib/Target/ARM/ARMRegisterInfo.td
@@ -70,6 +70,8 @@ def R4  : ARMReg< 4, "r4">,  DwarfRegNum<[4]>;
 def R5  : ARMReg< 5, "r5">,  DwarfRegNum<[5]>;
 def R6  : ARMReg< 6, "r6">,  DwarfRegNum<[6]>;
 def R7  : ARMReg< 7, "r7">,  DwarfRegNum<[7]>;
+// These require 32-bit instructions.
+let CostPerUse = 1 in {
 def R8  : ARMReg< 8, "r8">,  DwarfRegNum<[8]>;
 def R9  : ARMReg< 9, "r9">,  DwarfRegNum<[9]>;
 def R10 : ARMReg<10, "r10">, DwarfRegNum<[10]>;
@@ -78,6 +80,7 @@ def R12 : ARMReg<12, "r12">, DwarfRegNum<[12]>;
 def SP  : ARMReg<13, "sp">,  DwarfRegNum<[13]>;
 def LR  : ARMReg<14, "lr">,  DwarfRegNum<[14]>;
 def PC  : ARMReg<15, "pc">,  DwarfRegNum<[15]>;
+}
 
 // Float registers
 def S0  : ARMFReg< 0, "s0">;  def S1  : ARMFReg< 1, "s1">;