[ARM] Implement isTruncateFree

[samparker]

Summary

Implement the isTruncateFree hooks, lifted from AArch64, that are used by TargetTransformInfo. This allows simplifycfg to reduce the test case into a single basic block.

[SjoerdMeijer]

Looks reasonable to me.

lib/Target/ARM/ARMISelLowering.cpp

12182 ↗

(On Diff #114007)

Nit: perhaps a comment why on ARM this is free as well? There's a comment about this in AArch64.

test/Transforms/SimplifyCFG/ARM/select-trunc-i64.ll

13 ↗

(On Diff #114007)

Nit: indent off by 1

llvm/trunk/lib/Target/ARM/ARMISelLowering.h

bool allowsMisalignedMemoryAccesses(EVT VT, unsigned AddrSpace,
                                         bool *Fast) const override;
 
     EVT getOptimalMemOpType(uint64_t Size,
                             unsigned DstAlign, unsigned SrcAlign,
                             bool IsMemset, bool ZeroMemset,
                             bool MemcpyStrSrc,
                             MachineFunction &MF) const override;
 
-    using TargetLowering::isZExtFree;
+    bool isTruncateFree(Type *SrcTy, Type *DstTy) const override;
+    bool isTruncateFree(EVT SrcVT, EVT DstVT) const override;
     bool isZExtFree(SDValue Val, EVT VT2) const override;
 
     bool isVectorLoadExtDesirable(SDValue ExtVal) const override;
 
     bool allowTruncateForTailCall(Type *Ty1, Type *Ty2) const override;
 
 
     /// isLegalAddressingMode - Return true if the addressing mode represented

llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp

if (Size >= 16 &&
       return MVT::f64;
     }
   }
 
   // Let the target-independent logic figure it out.
   return MVT::Other;
 }
 
+// 64-bit integers are split into their high and low parts and held in two
+// different registers, so the trunc is free since the low register can just
+// be used.
+bool ARMTargetLowering::isTruncateFree(Type *SrcTy, Type *DstTy) const {
+  if (!SrcTy->isIntegerTy() || !DstTy->isIntegerTy())
+    return false;
+  unsigned NumBits1 = SrcTy->getPrimitiveSizeInBits();
+  unsigned NumBits2 = DstTy->getPrimitiveSizeInBits();
+  return NumBits1 > NumBits2;
+}
+
+bool ARMTargetLowering::isTruncateFree(EVT SrcVT, EVT DstVT) const {
+  if (SrcVT.isVector() || DstVT.isVector() || !SrcVT.isInteger() ||
+      !DstVT.isInteger())
+    return false;
+  unsigned NumBits1 = SrcVT.getSizeInBits();
+  unsigned NumBits2 = DstVT.getSizeInBits();
+  return NumBits1 > NumBits2;
+}
+
 bool ARMTargetLowering::isZExtFree(SDValue Val, EVT VT2) const {
   if (Val.getOpcode() != ISD::LOAD)
     return false;
 
   EVT VT1 = Val.getValueType();
   if (!VT1.isSimple() || !VT1.isInteger() ||
       !VT2.isSimple() || !VT2.isInteger())
     return false;

llvm/trunk/test/Transforms/SimplifyCFG/ARM/select-trunc-i64.ll

+;RUN: opt -S -simplifycfg -mtriple=arm < %s | FileCheck %s
+target datalayout = "e-m:e-p:32:32-i64:64-v128:64:128-a:0:32-n32-S64"
+
+; CHECK-LABEL: select_trunc_i64
+; CHECK-NOT: br
+; CHECK: select
+; CHECK: select
+define arm_aapcscc i32 @select_trunc_i64(i32 %a, i32 %b) {
+entry:
+  %conv = sext i32 %a to i64
+  %conv1 = sext i32 %b to i64
+  %add = add nsw i64 %conv1, %conv
+  %cmp = icmp sgt i64 %add, 2147483647
+  br i1 %cmp, label %cond.end7, label %cond.false
+
+cond.false:                                       ; preds = %entry
+  %0 = icmp sgt i64 %add, -2147483648
+  %cond = select i1 %0, i64 %add, i64 -2147483648
+  %extract.t = trunc i64 %cond to i32
+  br label %cond.end7
+
+cond.end7:                                        ; preds = %cond.false, %entry
+  %cond8.off0 = phi i32 [ 2147483647, %entry ], [ %extract.t, %cond.false ]
+  ret i32 %cond8.off0
+}