Index: llvm/lib/Target/X86/X86InstrInfo.td
===================================================================
--- llvm/lib/Target/X86/X86InstrInfo.td
+++ llvm/lib/Target/X86/X86InstrInfo.td
@@ -1132,7 +1132,7 @@
   ISD::LoadExtType ExtType = LD->getExtensionType();
   if (ExtType == ISD::NON_EXTLOAD)
     return true;
-  if (ExtType == ISD::EXTLOAD)
+  if (ExtType == ISD::EXTLOAD && Subtarget->enableWideLoad())
     return LD->getAlignment() >= 4 && LD->isSimple();
   return false;
 }]>;
Index: llvm/lib/Target/X86/X86Subtarget.h
===================================================================
--- llvm/lib/Target/X86/X86Subtarget.h
+++ llvm/lib/Target/X86/X86Subtarget.h
@@ -907,6 +907,8 @@
   }
 
   bool enableAdvancedRASplitCost() const override { return true; }
+
+  bool enableWideLoad() const;
 };
 
 } // end namespace llvm
Index: llvm/lib/Target/X86/X86Subtarget.cpp
===================================================================
--- llvm/lib/Target/X86/X86Subtarget.cpp
+++ llvm/lib/Target/X86/X86Subtarget.cpp
@@ -45,6 +45,10 @@
 #define GET_SUBTARGETINFO_CTOR
 #include "X86GenSubtargetInfo.inc"
 
+static cl::opt<bool> EnableWideLoad("x86-enalbe-wide-load", cl::init(true),
+    cl::desc("If load an 8b or 16b value from 32b aligned address, and any "
+             "extended value is expected, change it to 32b load."), cl::Hidden);
+
 // Temporary option to control early if-conversion for x86 while adding machine
 // models.
 static cl::opt<bool>
@@ -384,3 +388,7 @@
 bool X86Subtarget::isPositionIndependent() const {
   return TM.isPositionIndependent();
 }
+
+bool X86Subtarget::enableWideLoad() const {
+  return EnableWideLoad;
+}
Index: llvm/test/CodeGen/X86/no-wide-load.ll
===================================================================
--- /dev/null
+++ llvm/test/CodeGen/X86/no-wide-load.ll
@@ -0,0 +1,22 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-- -x86-enalbe-wide-load=false | FileCheck %s
+
+%struct.S = type { i32, i16, i16 }
+
+define void @foo(%struct.S* %p, i16 signext %s) {
+; CHECK-LABEL: foo:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    movzwl 4(%rdi), %eax
+; CHECK-NEXT:    andl $-1121, %eax # imm = 0xFB9F
+; CHECK-NEXT:    orl $1024, %eax # imm = 0x400
+; CHECK-NEXT:    movw %ax, 4(%rdi)
+; CHECK-NEXT:    retq
+entry:
+  %f2 = getelementptr inbounds %struct.S, %struct.S* %p, i64 0, i32 1
+  %0 = load i16, i16* %f2, align 4
+  %1 = and i16 %0, -1121
+  %2 = or i16 %1, 1024
+  store i16 %2, i16* %f2, align 4
+  ret void
+}
+