[GlobalISel] Use replaceRegOrBuildCopy when legalizer-combining anyext(undef).

This avoids the unnecessary creation of some COPY nodes created from the CSE
builder.

Author: davemgreen

llvm/include/llvm/CodeGen/GlobalISel/LegalizationArtifactCombiner.h

@@ -112,7 +112,7 @@ class LegalizationArtifactCombiner {
         return true;
       }
     }
-    return tryFoldImplicitDef(MI, DeadInsts, UpdatedDefs);
+    return tryFoldImplicitDef(MI, DeadInsts, UpdatedDefs, Observer);
   }
 
   bool tryCombineZExt(MachineInstr &MI,
@@ -187,7 +187,7 @@ class LegalizationArtifactCombiner {
         return true;
       }
     }
-    return tryFoldImplicitDef(MI, DeadInsts, UpdatedDefs);
+    return tryFoldImplicitDef(MI, DeadInsts, UpdatedDefs, Observer);
   }
 
   bool tryCombineSExt(MachineInstr &MI,
@@ -252,7 +252,7 @@ class LegalizationArtifactCombiner {
       }
     }
 
-    return tryFoldImplicitDef(MI, DeadInsts, UpdatedDefs);
+    return tryFoldImplicitDef(MI, DeadInsts, UpdatedDefs, Observer);
   }
 
   bool tryCombineTrunc(MachineInstr &MI,
@@ -376,7 +376,8 @@ class LegalizationArtifactCombiner {
   /// Try to fold G_[ASZ]EXT (G_IMPLICIT_DEF).
   bool tryFoldImplicitDef(MachineInstr &MI,
                           SmallVectorImpl<MachineInstr *> &DeadInsts,
-                          SmallVectorImpl<Register> &UpdatedDefs) {
+                          SmallVectorImpl<Register> &UpdatedDefs,
+                          GISelObserverWrapper &Observer) {
     unsigned Opcode = MI.getOpcode();
     assert(Opcode == TargetOpcode::G_ANYEXT || Opcode == TargetOpcode::G_ZEXT ||
            Opcode == TargetOpcode::G_SEXT);
@@ -392,7 +393,9 @@ class LegalizationArtifactCombiner {
         if (!isInstLegal({TargetOpcode::G_IMPLICIT_DEF, {DstTy}}))
           return false;
         LLVM_DEBUG(dbgs() << ".. Combine G_ANYEXT(G_IMPLICIT_DEF): " << MI;);
-        Builder.buildInstr(TargetOpcode::G_IMPLICIT_DEF, {DstReg}, {});
+        auto Impl = Builder.buildInstr(TargetOpcode::G_IMPLICIT_DEF, {DstTy}, {});
+        replaceRegOrBuildCopy(DstReg, Impl.getReg(0), MRI, Builder, UpdatedDefs,
+                              Observer);
         UpdatedDefs.push_back(DstReg);
       } else {
         // G_[SZ]EXT (G_IMPLICIT_DEF) -> G_CONSTANT 0 because the top

llvm/test/CodeGen/AArch64/GlobalISel/combine-ext-debugloc.mir

@@ -2,7 +2,7 @@
 
 # Check that when we combine ZEXT/ANYEXT we assign the correct location.
 # CHECK: !8 = !DILocation(line: 23, column: 5, scope: !4)
-# CHECK:  G_AND %14, %15, debug-location !8
+# CHECK:  G_AND %16, %15, debug-location !8
 
 --- |
   target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"

llvm/test/CodeGen/AArch64/GlobalISel/legalize-build-vector.mir

@@ -141,8 +141,7 @@ body: |
   bb.0:
     ; CHECK-LABEL: name: widen_v2s8
     ; CHECK: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[DEF]](s32)
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32)
     ; CHECK-NEXT: $d0 = COPY [[BUILD_VECTOR]](<2 x s32>)
     ; CHECK-NEXT: RET_ReallyLR
     %0:_(s8) = G_IMPLICIT_DEF

llvm/test/CodeGen/AArch64/GlobalISel/legalize-concat-vectors.mir

@@ -75,10 +75,7 @@ body:             |
     ; CHECK-NEXT: %b:_(s32) = G_LOAD %a(p0) :: (load (s32))
     ; CHECK-NEXT: %c:_(<4 x s8>) = G_BITCAST %b(s32)
     ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s16) = COPY [[DEF]](s16)
-    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s16) = COPY [[DEF]](s16)
-    ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s16) = COPY [[DEF]](s16)
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[COPY]](s16), [[COPY1]](s16), [[COPY2]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
     ; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(<8 x s8>) = G_TRUNC [[BUILD_VECTOR]](<8 x s16>)
     ; CHECK-NEXT: [[UV:%[0-9]+]]:_(<4 x s8>), [[UV1:%[0-9]+]]:_(<4 x s8>) = G_UNMERGE_VALUES [[TRUNC]](<8 x s8>)
     ; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST %c(<4 x s8>)
@@ -113,8 +110,7 @@ body:             |
     ; CHECK-NEXT: %b:_(s32) = G_LOAD %a(p0) :: (load (s32))
     ; CHECK-NEXT: %c:_(<2 x s16>) = G_BITCAST %b(s32)
     ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
     ; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[BUILD_VECTOR]](<4 x s32>)
     ; CHECK-NEXT: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[TRUNC]](<4 x s16>)
     ; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST %c(<2 x s16>)