ARM/Dwarf: correctly align stack before callee-saved VPRs

We were making an attempt to do this by adding an extra callee-saved GPR (so
that there was an even number in the list), but when that failed we went ahead
and pushed anyway.

This had a couple of potential issues:
  + The .cfi directives we emit misplaced dN because they were based on
    PrologEpilogInserter's calculation.
  + Unaligned stores can be less efficient.
  + Unaligned stores can actually fault (likely only an issue in niche cases,
    but possible).

This adds a final explicit stack adjustment if all other options fail, so that
the actual locations of the registers match up with where they should be.

llvm-svn: 221320

2 files changed, 26 insertions, 5 deletions

diff --git a/llvm/lib/Target/ARM/ARMFrameLowering.cpp b/llvm/lib/Target/ARM/ARMFrameLowering.cpp
index 74f686507df..4589799933b 100644
--- a/llvm/lib/Target/ARM/ARMFrameLowering.cpp
+++ b/llvm/lib/Target/ARM/ARMFrameLowering.cpp
@@ -260,10 +260,11 @@ void ARMFrameLowering::emitPrologue(MachineFunction &MF) const {
 
   // Determine starting offsets of spill areas.
   bool HasFP = hasFP(MF);
-  unsigned DPRCSOffset  = NumBytes - (ArgRegsSaveSize + GPRCS1Size
-                                      + GPRCS2Size + DPRCSSize);
-  unsigned GPRCS2Offset = DPRCSOffset + DPRCSSize;
-  unsigned GPRCS1Offset = GPRCS2Offset + GPRCS2Size;
+  unsigned GPRCS1Offset = NumBytes - ArgRegsSaveSize - GPRCS1Size;
+  unsigned GPRCS2Offset = GPRCS1Offset - GPRCS2Size;
+  unsigned DPRAlign = DPRCSSize ? std::min(8U, Align) : 4U;
+  unsigned DPRGapSize = (GPRCS1Size + GPRCS2Size + ArgRegsSaveSize) % DPRAlign;
+  unsigned DPRCSOffset = GPRCS2Offset - DPRGapSize - DPRCSSize;
   int FramePtrOffsetInPush = 0;
   if (HasFP) {
     FramePtrOffsetInPush = MFI->getObjectOffset(FramePtrSpillFI)
@@ -279,6 +280,15 @@ void ARMFrameLowering::emitPrologue(MachineFunction &MF) const {
   if (GPRCS2Size > 0)
     GPRCS2Push = LastPush = MBBI++;
 
+  // Prolog/epilog inserter assumes we correctly align DPRs on the stack, so our
+  // .cfi_offset operations will reflect that.
+  if (DPRGapSize) {
+    assert(DPRGapSize == 4 && "unexpected alignment requirements for DPRs");
+    if (!tryFoldSPUpdateIntoPushPop(STI, MF, LastPush, DPRGapSize))
+      emitSPUpdate(isARM, MBB, MBBI, dl, TII, -DPRGapSize,
+                   MachineInstr::FrameSetup);
+  }
+
   // Move past area 3.
   if (DPRCSSize > 0) {
     DPRCSPush = MBBI;
@@ -508,6 +518,7 @@ void ARMFrameLowering::emitPrologue(MachineFunction &MF) const {
 
   AFI->setGPRCalleeSavedArea1Size(GPRCS1Size);
   AFI->setGPRCalleeSavedArea2Size(GPRCS2Size);
+  AFI->setDPRCalleeSavedGapSize(DPRGapSize);
   AFI->setDPRCalleeSavedAreaSize(DPRCSSize);
 
   // If we need dynamic stack realignment, do it here. Be paranoid and make
@@ -613,6 +624,7 @@ void ARMFrameLowering::emitEpilogue(MachineFunction &MF,
     NumBytes -= (ArgRegsSaveSize +
                  AFI->getGPRCalleeSavedArea1Size() +
                  AFI->getGPRCalleeSavedArea2Size() +
+                 AFI->getDPRCalleeSavedGapSize() +
                  AFI->getDPRCalleeSavedAreaSize());
 
     // Reset SP based on frame pointer only if the stack frame extends beyond
@@ -661,6 +673,12 @@ void ARMFrameLowering::emitEpilogue(MachineFunction &MF,
       while (MBBI->getOpcode() == ARM::VLDMDIA_UPD)
         MBBI++;
     }
+    if (AFI->getDPRCalleeSavedGapSize()) {
+      assert(AFI->getDPRCalleeSavedGapSize() == 4 &&
+             "unexpected DPR alignment gap");
+      emitSPUpdate(isARM, MBB, MBBI, dl, TII, AFI->getDPRCalleeSavedGapSize());
+    }
+
     if (AFI->getGPRCalleeSavedArea2Size()) MBBI++;
     if (AFI->getGPRCalleeSavedArea1Size()) MBBI++;
   }
diff --git a/llvm/lib/Target/ARM/ARMMachineFunctionInfo.h b/llvm/lib/Target/ARM/ARMMachineFunctionInfo.h
index 4f9ea7abb0d..4e67fa13638 100644
--- a/llvm/lib/Target/ARM/ARMMachineFunctionInfo.h
+++ b/llvm/lib/Target/ARM/ARMMachineFunctionInfo.h
@@ -86,6 +86,7 @@ class ARMFunctionInfo : public MachineFunctionInfo {
   /// areas.
   unsigned GPRCS1Size;
   unsigned GPRCS2Size;
+  unsigned DPRCSAlignGapSize;
   unsigned DPRCSSize;
 
   /// NumAlignedDPRCS2Regs - The number of callee-saved DPRs that are saved in
@@ -134,7 +135,7 @@ public:
     RestoreSPFromFP(false),
     LRSpilledForFarJump(false),
     FramePtrSpillOffset(0), GPRCS1Offset(0), GPRCS2Offset(0), DPRCSOffset(0),
-    GPRCS1Size(0), GPRCS2Size(0), DPRCSSize(0),
+    GPRCS1Size(0), GPRCS2Size(0), DPRCSAlignGapSize(0), DPRCSSize(0),
     NumAlignedDPRCS2Regs(0),
     JumpTableUId(0), PICLabelUId(0),
     VarArgsFrameIndex(0), HasITBlocks(false), GlobalBaseReg(0) {}
@@ -183,10 +184,12 @@ public:
 
   unsigned getGPRCalleeSavedArea1Size() const { return GPRCS1Size; }
   unsigned getGPRCalleeSavedArea2Size() const { return GPRCS2Size; }
+  unsigned getDPRCalleeSavedGapSize() const   { return DPRCSAlignGapSize; }
   unsigned getDPRCalleeSavedAreaSize()  const { return DPRCSSize; }
 
   void setGPRCalleeSavedArea1Size(unsigned s) { GPRCS1Size = s; }
   void setGPRCalleeSavedArea2Size(unsigned s) { GPRCS2Size = s; }
+  void setDPRCalleeSavedGapSize(unsigned s)   { DPRCSAlignGapSize = s; }
   void setDPRCalleeSavedAreaSize(unsigned s)  { DPRCSSize = s; }
 
   unsigned getArgumentStackSize() const { return ArgumentStackSize; }