Always use relative jump table encodings on PowerPC64.

For the default, small and medium code model, use the existing
difference from the jump table towards the label. For all other code
models, setup the picbase and use the difference between the picbase and
the block address.

Overall, this results in smaller data tables at the expensive of one or
two more arithmetic operation at the jump site. Given that we only create
jump tables with a lot more than two entries, it is a net win in size.
For larger code models the assumption remains that individual functions
are no larger than 2GB.

Differential Revision: https://reviews.llvm.org/D26336


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287059 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Target/PowerPC/PPCISelLowering.cpp

@@ -27,6 +27,7 @@
 #include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineFunction.h"
 #include "llvm/CodeGen/MachineInstrBuilder.h"
+#include "llvm/CodeGen/MachineJumpTableInfo.h"
 #include "llvm/CodeGen/MachineLoopInfo.h"
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/CodeGen/SelectionDAG.h"
@@ -2179,6 +2180,55 @@ SDValue PPCTargetLowering::LowerConstantPool(SDValue Op,
   return LowerLabelRef(CPIHi, CPILo, IsPIC, DAG);
 }
 
+// For 64-bit PowerPC, prefer the more compact relative encodings.
+// This trades 32 bits per jump table entry for one or two instructions
+// on the jump site.
+unsigned PPCTargetLowering::getJumpTableEncoding() const {
+  if (isJumpTableRelative())
+    return MachineJumpTableInfo::EK_LabelDifference32;
+
+  return TargetLowering::getJumpTableEncoding();
+}
+
+bool PPCTargetLowering::isJumpTableRelative() const {
+  if (Subtarget.isPPC64())
+    return true;
+  return TargetLowering::isJumpTableRelative();
+}
+
+SDValue PPCTargetLowering::getPICJumpTableRelocBase(SDValue Table,
+                                                    SelectionDAG &DAG) const {
+  if (!Subtarget.isPPC64())
+    return TargetLowering::getPICJumpTableRelocBase(Table, DAG);
+
+  switch (getTargetMachine().getCodeModel()) {
+  case CodeModel::Default:
+  case CodeModel::Small:
+  case CodeModel::Medium:
+    return TargetLowering::getPICJumpTableRelocBase(Table, DAG);
+  default:
+    return DAG.getNode(PPCISD::GlobalBaseReg, SDLoc(),
+                       getPointerTy(DAG.getDataLayout()));
+  }
+}
+
+const MCExpr *
+PPCTargetLowering::getPICJumpTableRelocBaseExpr(const MachineFunction *MF,
+                                                unsigned JTI,
+                                                MCContext &Ctx) const {
+  if (!Subtarget.isPPC64())
+    return TargetLowering::getPICJumpTableRelocBaseExpr(MF, JTI, Ctx);
+
+  switch (getTargetMachine().getCodeModel()) {
+  case CodeModel::Default:
+  case CodeModel::Small:
+  case CodeModel::Medium:
+    return TargetLowering::getPICJumpTableRelocBaseExpr(MF, JTI, Ctx);
+  default:
+    return MCSymbolRefExpr::create(MF->getPICBaseSymbol(), Ctx);
+  }
+}
+
 SDValue PPCTargetLowering::LowerJumpTable(SDValue Op, SelectionDAG &DAG) const {
   EVT PtrVT = Op.getValueType();
   JumpTableSDNode *JT = cast<JumpTableSDNode>(Op);

lib/Target/PowerPC/PPCISelLowering.h

@@ -770,6 +770,15 @@ namespace llvm {
     void insertSSPDeclarations(Module &M) const override;
 
     bool isFPImmLegal(const APFloat &Imm, EVT VT) const override;
+
+    unsigned getJumpTableEncoding() const override;
+    bool isJumpTableRelative() const override;
+    SDValue getPICJumpTableRelocBase(SDValue Table,
+                                     SelectionDAG &DAG) const override;
+    const MCExpr *getPICJumpTableRelocBaseExpr(const MachineFunction *MF,
+                                               unsigned JTI,
+                                               MCContext &Ctx) const override;
+
   private:
     struct ReuseLoadInfo {
       SDValue Ptr;

test/CodeGen/PowerPC/mcm-5.ll

@@ -1,5 +1,5 @@
 ; RUN: llc -verify-machineinstrs -mcpu=pwr7 -code-model=medium <%s | FileCheck %s
-; RUN: llc -verify-machineinstrs -mcpu=pwr7 -code-model=large <%s | FileCheck %s
+; RUN: llc -verify-machineinstrs -mcpu=pwr7 -code-model=large <%s | FileCheck -check-prefix=LARGE  %s
 
 ; Test correct code generation for medium and large code model
 ; for loading the address of a jump table from the TOC.
@@ -50,11 +50,30 @@ sw.epilog:                                        ; preds = %sw.bb3, %sw.default
   %5 = load i32, i32* %i.addr, align 4
   ret i32 %5
 }
-
 ; CHECK-LABEL: test_jump_table:
-; CHECK: addis [[REG1:[0-9]+]], 2, .LC[[TOCNUM:[0-9]+]]@toc@ha
-; CHECK: ld [[REG2:[0-9]+]], .LC[[TOCNUM]]@toc@l([[REG1]])
-; CHECK: ldx {{[0-9]+}}, {{[0-9]+}}, [[REG2]]
-; CHECK: .section .toc
-; CHECK: .LC[[TOCNUM]]:
-; CHECK: .tc {{[a-z0-9A-Z_.]+}}[TC],{{[a-z0-9A-Z_.]+}}
+; CHECK-NOT:       bl .L0$pb
+
+; CHECK:       addis [[REG1:[0-9]+]], 2, .LC[[TOCNUM:[0-9]+]]@toc@ha
+; CHECK:       ld [[REG2:[0-9]+]], .LC[[TOCNUM]]@toc@l([[REG1]])
+; CHECK:       lwax [[REG3:[0-9]+]], {{[0-9]+}}, [[REG2]]
+; CHECK-NEXT:  add  [[REG4:[0-9]+]], [[REG3]], [[REG2]]
+; CHECK-NEXT:  mtctr [[REG4]]
+; CHECK-NEXT:  bctr
+
+; CHECK-LABEL: .LJTI0_0:
+; CHECK-NEXT: .long	.LBB0_{{[0-9]+}}-.LJTI0_0
+
+; LARGE-LABEL: test_jump_table:
+; LARGE:       bl .L0$pb
+; LARGE-NEXT:  .L0$pb:
+; LARGE:       mflr [[REGBASE:[0-9]+]]
+
+; LARGE:       addis [[REG1:[0-9]+]], 2, .LC[[TOCNUM:[0-9]+]]@toc@ha
+; LARGE:       ld [[REG2:[0-9]+]], .LC[[TOCNUM]]@toc@l([[REG1]])
+; LARGE:       lwax [[REG3:[0-9]+]], {{[0-9]+}}, [[REG2]]
+; LARGE-NEXT:  add  [[REG4:[0-9]+]], [[REG3]], [[REGBASE]]
+; LARGE-NEXT:  mtctr [[REG4]]
+; LARGE-NEXT:  bctr
+
+; LARGE-LABEL: .LJTI0_0:
+; LARGE-NEXT: .long	.LBB0_{{[0-9]+}}-.L0$pb