[PowerPC] Simplify and improve loading into TOC register

During an indirect function call sequence on the 64-bit SVR4 ABI,
generate code must load and then restore the TOC register.

This does not use a regular LOAD instruction since the TOC
register r2 is marked as reserved.  Instead, the are two
special instruction patterns:

 let RST = 2, DS = 2 in
 def LDinto_toc: DSForm_1a<58, 0, (outs), (ins g8rc:$reg),
                     "ld 2, 8($reg)", IIC_LdStLD,
                     [(PPCload_toc i64:$reg)]>, isPPC64;
 
 let RST = 2, DS = 10, RA = 1 in
 def LDtoc_restore : DSForm_1a<58, 0, (outs), (ins),
                     "ld 2, 40(1)", IIC_LdStLD,
                     [(PPCtoc_restore)]>, isPPC64;

Note that these not only restrict the destination of the
load to r2, but they also restrict the *source* of the
load to particular address combinations.  The latter is
a problem when we want to support the ELFv2 ABI, since
there the TOC save slot is no longer at 40(1).

This patch replaces those two instructions with a single
instruction pattern that only hard-codes r2 as destination,
but supports generic addresses as source.  This will allow
supporting the ELFv2 ABI, and also helps generate more
efficient code for calls to absolute addresses (allowing
simplification of the ppc64-calls.ll test case).



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@211193 91177308-0d34-0410-b5e6-96231b3b80d8

6 files changed, 25 insertions, 40 deletions

diff --git a/lib/Target/PowerPC/PPCFrameLowering.h b/lib/Target/PowerPC/PPCFrameLowering.h
index 1280dec864..ca1ca56d08 100644
--- a/lib/Target/PowerPC/PPCFrameLowering.h
+++ b/lib/Target/PowerPC/PPCFrameLowering.h
@@ -74,6 +74,12 @@ public:
     return isPPC64 ? 16 : 4;
   }
 
+  /// getTOCSaveOffset - Return the previous frame offset to save the
+  /// TOC register -- 64-bit SVR4 ABI only.
+  static unsigned getTOCSaveOffset(void) {
+    return 40;
+  }
+
   /// getFramePointerSaveOffset - Return the previous frame offset to save the
   /// frame pointer.
   static unsigned getFramePointerSaveOffset(bool isPPC64, bool isDarwinABI) {
diff --git a/lib/Target/PowerPC/PPCISelLowering.cpp b/lib/Target/PowerPC/PPCISelLowering.cpp
index e815d615ac..201d124199 100644
--- a/lib/Target/PowerPC/PPCISelLowering.cpp
+++ b/lib/Target/PowerPC/PPCISelLowering.cpp
@@ -773,7 +773,6 @@ const char *PPCTargetLowering::getTargetNodeName(unsigned Opcode) const {
   case PPCISD::Hi:              return "PPCISD::Hi";
   case PPCISD::Lo:              return "PPCISD::Lo";
   case PPCISD::TOC_ENTRY:       return "PPCISD::TOC_ENTRY";
-  case PPCISD::TOC_RESTORE:     return "PPCISD::TOC_RESTORE";
   case PPCISD::LOAD:            return "PPCISD::LOAD";
   case PPCISD::LOAD_TOC:        return "PPCISD::LOAD_TOC";
   case PPCISD::DYNALLOC:        return "PPCISD::DYNALLOC";
@@ -3544,8 +3543,10 @@ unsigned PrepareCall(SelectionDAG &DAG, SDValue &Callee, SDValue &InFlag,
       // additional register being allocated and an unnecessary move instruction
       // being generated.
       VTs = DAG.getVTList(MVT::Other, MVT::Glue);
+      SDValue TOCOff = DAG.getIntPtrConstant(8);
+      SDValue AddTOC = DAG.getNode(ISD::ADD, dl, MVT::i64, Callee, TOCOff);
       SDValue LoadTOCPtr = DAG.getNode(PPCISD::LOAD_TOC, dl, VTs, Chain,
-                                       Callee, InFlag);
+                                       AddTOC, InFlag);
       Chain = LoadTOCPtr.getValue(0);
       InFlag = LoadTOCPtr.getValue(1);
 
@@ -3729,7 +3730,12 @@ PPCTargetLowering::FinishCall(CallingConv::ID CallConv, SDLoc dl,
 
   if (needsTOCRestore) {
     SDVTList VTs = DAG.getVTList(MVT::Other, MVT::Glue);
-    Chain = DAG.getNode(PPCISD::TOC_RESTORE, dl, VTs, Chain, InFlag);
+    EVT PtrVT = DAG.getTargetLoweringInfo().getPointerTy();
+    SDValue StackPtr = DAG.getRegister(PPC::X1, PtrVT);
+    unsigned TOCSaveOffset = PPCFrameLowering::getTOCSaveOffset();
+    SDValue TOCOff = DAG.getIntPtrConstant(TOCSaveOffset);
+    SDValue AddTOC = DAG.getNode(ISD::ADD, dl, MVT::i64, StackPtr, TOCOff);
+    Chain = DAG.getNode(PPCISD::LOAD_TOC, dl, VTs, Chain, AddTOC, InFlag);
     InFlag = Chain.getValue(1);
   }
 
@@ -4388,7 +4394,8 @@ PPCTargetLowering::LowerCall_64SVR4(SDValue Chain, SDValue Callee,
     // Load r2 into a virtual register and store it to the TOC save area.
     SDValue Val = DAG.getCopyFromReg(Chain, dl, PPC::X2, MVT::i64);
     // TOC save area offset.
-    SDValue PtrOff = DAG.getIntPtrConstant(40);
+    unsigned TOCSaveOffset = PPCFrameLowering::getTOCSaveOffset();
+    SDValue PtrOff = DAG.getIntPtrConstant(TOCSaveOffset);
     SDValue AddPtr = DAG.getNode(ISD::ADD, dl, PtrVT, StackPtr, PtrOff);
     Chain = DAG.getStore(Val.getValue(1), dl, Val, AddPtr, MachinePointerInfo(),
                          false, false, 0);
diff --git a/lib/Target/PowerPC/PPCISelLowering.h b/lib/Target/PowerPC/PPCISelLowering.h
index 9b85c99726..2b69208fea 100644
--- a/lib/Target/PowerPC/PPCISelLowering.h
+++ b/lib/Target/PowerPC/PPCISelLowering.h
@@ -70,19 +70,14 @@ namespace llvm {
 
       TOC_ENTRY,
 
-      /// The following three target-specific nodes are used for calls through
+      /// The following two target-specific nodes are used for calls through
       /// function pointers in the 64-bit SVR4 ABI.
 
-      /// Restore the TOC from the TOC save area of the current stack frame.
-      /// This is basically a hard coded load instruction which additionally
-      /// takes/produces a flag.
-      TOC_RESTORE,
-
       /// Like a regular LOAD but additionally taking/producing a flag.
       LOAD,
 
-      /// LOAD into r2 (also taking/producing a flag). Like TOC_RESTORE, this is
-      /// a hard coded load instruction.
+      /// Like LOAD (taking/producing a flag), but using r2 as hard-coded
+      /// destination.
       LOAD_TOC,
 
       /// OPRC, CHAIN = DYNALLOC(CHAIN, NEGSIZE, FRAME_INDEX)
diff --git a/lib/Target/PowerPC/PPCInstr64Bit.td b/lib/Target/PowerPC/PPCInstr64Bit.td
index b71c09ea8e..9318f70f43 100644
--- a/lib/Target/PowerPC/PPCInstr64Bit.td
+++ b/lib/Target/PowerPC/PPCInstr64Bit.td
@@ -802,17 +802,11 @@ def LDtocCPT: Pseudo<(outs g8rc:$rD), (ins tocentry:$disp, g8rc:$reg),
                   [(set i64:$rD,
                      (PPCtoc_entry tconstpool:$disp, i64:$reg))]>, isPPC64;
 
-let hasSideEffects = 1, isCodeGenOnly = 1 in {
-let RST = 2, DS = 2 in
-def LDinto_toc: DSForm_1a<58, 0, (outs), (ins g8rc:$reg),
-                    "ld 2, 8($reg)", IIC_LdStLD,
-                    [(PPCload_toc i64:$reg)]>, isPPC64;
-                    
-let RST = 2, DS = 10, RA = 1 in
-def LDtoc_restore : DSForm_1a<58, 0, (outs), (ins),
-                    "ld 2, 40(1)", IIC_LdStLD,
-                    [(PPCtoc_restore)]>, isPPC64;
-}
+let hasSideEffects = 1, isCodeGenOnly = 1, RST = 2 in
+def LDinto_toc: DSForm_1<58, 0, (outs), (ins memrix:$src),
+                    "ld 2, $src", IIC_LdStLD,
+                    [(PPCload_toc ixaddr:$src)]>, isPPC64;
+
 def LDX  : XForm_1<31,  21, (outs g8rc:$rD), (ins memrr:$src),
                    "ldx $rD, $src", IIC_LdStLD,
                    [(set i64:$rD, (load xaddr:$src))]>, isPPC64;
diff --git a/lib/Target/PowerPC/PPCInstrFormats.td b/lib/Target/PowerPC/PPCInstrFormats.td
index 7fed2c65da..1e4396cd10 100644
--- a/lib/Target/PowerPC/PPCInstrFormats.td
+++ b/lib/Target/PowerPC/PPCInstrFormats.td
@@ -360,20 +360,6 @@ class DSForm_1<bits<6> opcode, bits<2> xo, dag OOL, dag IOL, string asmstr,
   let Inst{30-31} = xo;
 }
 
-class DSForm_1a<bits<6> opcode, bits<2> xo, dag OOL, dag IOL, string asmstr,
-                InstrItinClass itin, list<dag> pattern>
-         : I<opcode, OOL, IOL, asmstr, itin> {
-   bits<5>  RST;
-   bits<14> DS;
-   bits<5>  RA;
- 
-   let Pattern = pattern;
-   
-   let Inst{6-10}  = RST;
-   let Inst{11-15} = RA;
-   let Inst{16-29} = DS;
-   let Inst{30-31} = xo;
-}
 
 // 1.7.6 X-Form
 class XForm_base_r3xo<bits<6> opcode, bits<10> xo, dag OOL, dag IOL, string asmstr, 
diff --git a/lib/Target/PowerPC/PPCInstrInfo.td b/lib/Target/PowerPC/PPCInstrInfo.td
index e421f8e69e..c2e3382b3e 100644
--- a/lib/Target/PowerPC/PPCInstrInfo.td
+++ b/lib/Target/PowerPC/PPCInstrInfo.td
@@ -141,9 +141,6 @@ def PPCload   : SDNode<"PPCISD::LOAD", SDTypeProfile<1, 1, []>,
 def PPCload_toc : SDNode<"PPCISD::LOAD_TOC", SDTypeProfile<0, 1, []>,
                           [SDNPHasChain, SDNPSideEffect,
                            SDNPInGlue, SDNPOutGlue]>;
-def PPCtoc_restore : SDNode<"PPCISD::TOC_RESTORE", SDTypeProfile<0, 0, []>,
-                            [SDNPHasChain, SDNPSideEffect,
-                             SDNPInGlue, SDNPOutGlue]>;
 def PPCmtctr      : SDNode<"PPCISD::MTCTR", SDT_PPCCall,
                            [SDNPHasChain, SDNPOptInGlue, SDNPOutGlue]>;
 def PPCbctrl : SDNode<"PPCISD::BCTRL", SDTNone,