AMDGPU: Make s34 the FP register

Make the FP register callee saved.

This is tricky because now the FP needs to be spilled in the prolog
relative to the incoming SP register, rather than the frame register
used throughout the rest of the function. I don't like how this
bypassess the standard mechanism for CSR spills just to get the
correct insert point. I may look for a better solution, since all CSR
VGPRs may also need to have all lanes activated. Another option might
be to make getFrameIndexReference change the base register if the
frame index is a CSR, and then try to figure out the right insertion
point in emitProlog.

If there is a free VGPR lane available for SGPR spilling, try to use
it for the FP. If that would require intrtoducing a new VGPR spill,
try to use a free call clobbered SGPR. Only fallback to introducing a
new VGPR spill as a last resort.

This also doesn't attempt to handle SGPR spilling with scalar stores.

llvm-svn: 365372

1 files changed, 8 insertions, 0 deletions

diff --git a/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.h b/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.h
index 73393634b31..2cbca8930a6 100644
--- a/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.h
+++ b/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.h
@@ -429,6 +429,12 @@ private:
   unsigned NumVGPRSpillLanes = 0;
   SmallVector<SGPRSpillVGPRCSR, 2> SpillVGPRs;
 
+public: // FIXME
+  /// If this is set, an SGPR used for save/restore of the register used for the
+  /// frame pointer.
+  unsigned SGPRForFPSaveRestoreCopy = 0;
+  Optional<int> FramePointerSaveIndex;
+
 public:
   SIMachineFunctionInfo(const MachineFunction &MF);
 
@@ -448,6 +454,8 @@ public:
     return Mode;
   }
 
+  bool haveFreeLanesForSGPRSpill(const MachineFunction &MF,
+                                 unsigned NumLane) const;
   bool allocateSGPRSpillToVGPR(MachineFunction &MF, int FI);
   void removeSGPRToVGPRFrameIndices(MachineFrameInfo &MFI);