[AArch64] prevent (shl (srl x, c1), c2) -> (and (shift x, c3)) when load

Currently, process of replacing bitwise operations consisting of
`(shl (srl x, c1), c2)` with `And` is performed by `DAGCombiner`.

However, in certain case like `(shl (srl, x, c1) 2)` is do not
need to transform to `AND` if it was used to `Load` Target.

Consider following case:
```
        lsr x8, x8, llvm#56
        and x8, x8, #0xfc
        ldr w0, [x2, x8]
        ret
```

In this case, we can remove the `AND` by changing the target of `LDR`
to `[X2, X8, LSL llvm#2]` and right-shifting amount change to 56 to 58.

after changed:
```
        lsr x8, x8, llvm#58
        ldr w0, [x2, x8, lsl llvm#2]
        ret
```

This patch checks to see if the `(shl (srl x, c1) 2)` operation on
`load` target can be prevent transform to `And`.

llvm/lib/Target/AArch64/AArch64ISelLowering.cpp

@@ -16918,6 +16918,24 @@ bool AArch64TargetLowering::shouldFoldConstantShiftPairToMask(
     return (!C1 || !C2 || C1->getZExtValue() >= C2->getZExtValue());
   }
 
+  // We do not need to fold when this shifting used in specific load case:
+  // (ldr x, (add x, (shl (srl x, c1) 2)))
+  if (N->getOpcode() == ISD::SHL) {
+    auto C2 = dyn_cast_or_null<ConstantSDNode>(N->getOperand(1));
+    if (C2 && C2->getZExtValue() == 2) {
+      auto ShouldADD =
+          dyn_cast_or_null<SDNode>(N->use_begin().getUse().getUser());
+      if (ShouldADD && ShouldADD->getOpcode() == ISD::ADD) {
+        auto ShouldLOAD = dyn_cast_or_null<LoadSDNode>(
+            ShouldADD->use_begin().getUse().getUser());
+        if (ShouldLOAD) {
+          if (isIndexedLoadLegal(ISD::PRE_INC, ShouldLOAD->getMemoryVT()))
+            return false;
+        }
+      }
+    }
+  }
+
   return true;
 }
 

llvm/test/CodeGen/AArch64/arm64-fold-lshr.ll

@@ -19,9 +19,8 @@ define i32 @load_shr63(i64 %a, i64 %b, ptr %table) {
 ; CHECK-LABEL: load_shr63:
 ; CHECK:       // %bb.0: // %entry
 ; CHECK-NEXT:    mul x8, x1, x0
-; CHECK-NEXT:    lsr x8, x8, #61
-; CHECK-NEXT:    and x8, x8, #0x4
-; CHECK-NEXT:    ldr w0, [x2, x8]
+; CHECK-NEXT:    lsr x8, x8, #63
+; CHECK-NEXT:    ldr w0, [x2, x8, lsl #2]
 ; CHECK-NEXT:    ret
 entry:
   %mul = mul i64 %b, %a
@@ -35,8 +34,8 @@ define i32 @load_shr2(i64 %a, i64 %b, ptr %table) {
 ; CHECK-LABEL: load_shr2:
 ; CHECK:       // %bb.0: // %entry
 ; CHECK-NEXT:    mul x8, x1, x0
-; CHECK-NEXT:    and x8, x8, #0xfffffffffffffffc
-; CHECK-NEXT:    ldr w0, [x2, x8]
+; CHECK-NEXT:    lsr x8, x8, #2
+; CHECK-NEXT:    ldr w0, [x2, x8, lsl #2]
 ; CHECK-NEXT:    ret
 entry:
   %mul = mul i64 %b, %a
@@ -50,9 +49,8 @@ define i32 @load_shr1(i64 %a, i64 %b, ptr %table) {
 ; CHECK-LABEL: load_shr1:
 ; CHECK:       // %bb.0: // %entry
 ; CHECK-NEXT:    mul x8, x1, x0
-; CHECK-NEXT:    lsl x8, x8, #1
-; CHECK-NEXT:    and x8, x8, #0xfffffffffffffffc
-; CHECK-NEXT:    ldr w0, [x2, x8]
+; CHECK-NEXT:    lsr x8, x8, #1
+; CHECK-NEXT:    ldr w0, [x2, x8, lsl #2]
 ; CHECK-NEXT:    ret
 entry:
   %mul = mul i64 %b, %a