Index: lib/CodeGen/AsmPrinter/AsmPrinter.cpp
===================================================================
--- lib/CodeGen/AsmPrinter/AsmPrinter.cpp
+++ lib/CodeGen/AsmPrinter/AsmPrinter.cpp
@@ -1336,6 +1336,7 @@
       OutStreamer->SwitchSection(TLOF.getDataSection());
       const DataLayout &DL = M.getDataLayout();
 
+      EmitAlignment(Log2_32(DL.getPointerSize()));
       for (const auto &Stub : Stubs) {
         OutStreamer->EmitLabel(Stub.first);
         OutStreamer->EmitSymbolValue(Stub.second.getPointer(),
Index: test/Object/X86/align-stub.ll
===================================================================
--- /dev/null
+++ test/Object/X86/align-stub.ll
@@ -0,0 +1,85 @@
+; RUN: llc %s -mtriple=x86_64-unknown-unknown -relocation-model=pic -o - \
+; RUN: | FileCheck %s
+
+; Check that the stub symbol in .data section is aligned at 8 bytes.
+; https://bugs.llvm.org/show_bug.cgi?id=36885
+; CHECK:      .data{{[[:space:]]}} .p2align 3
+; CHECK-NEXT: .L_ZTI3bar.DW.stub:
+
+
+%class.bar = type { i8 }
+
+$_ZTS3bar = comdat any
+
+$_ZTI3bar = comdat any
+
+@str = dso_local global [3 x i8] c"..\00", align 1
+@_ZTVN10__cxxabiv117__class_type_infoE = external global i8*
+@_ZTS3bar = linkonce_odr dso_local constant [5 x i8] c"3bar\00", comdat
+@_ZTI3bar = linkonce_odr dso_local constant { i8*, i8* } { i8* bitcast (i8** getelementptr inbounds (i8*, i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([5 x i8], [5 x i8]* @_ZTS3bar, i32 0, i32 0) }, comdat
+
+; Function Attrs: noinline nounwind optnone uwtable
+define dso_local i32 @_Z3foov() #0 {
+entry:
+  ret i32 0
+}
+
+; Function Attrs: noinline norecurse optnone uwtable
+define dso_local i32 @main() #1 personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) {
+entry:
+  %retval = alloca i32, align 4
+  %exn.slot = alloca i8*
+  %ehselector.slot = alloca i32
+  %e = alloca %class.bar*, align 8
+  store i32 0, i32* %retval, align 4
+  %call = invoke i32 @_Z3foov()
+          to label %invoke.cont unwind label %lpad
+
+invoke.cont:                                      ; preds = %entry
+  store i32 %call, i32* %retval, align 4
+  br label %try.cont
+
+lpad:                                             ; preds = %entry
+  %0 = landingpad { i8*, i32 }
+          catch i8* bitcast ({ i8*, i8* }* @_ZTI3bar to i8*)
+  %1 = extractvalue { i8*, i32 } %0, 0
+  store i8* %1, i8** %exn.slot, align 8
+  %2 = extractvalue { i8*, i32 } %0, 1
+  store i32 %2, i32* %ehselector.slot, align 4
+  br label %catch.dispatch
+
+catch.dispatch:                                   ; preds = %lpad
+  %sel = load i32, i32* %ehselector.slot, align 4
+  %3 = call i32 @llvm.eh.typeid.for(i8* bitcast ({ i8*, i8* }* @_ZTI3bar to i8*)) #3
+  %matches = icmp eq i32 %sel, %3
+  br i1 %matches, label %catch, label %eh.resume
+
+catch:                                            ; preds = %catch.dispatch
+  %exn = load i8*, i8** %exn.slot, align 8
+  %4 = call i8* @__cxa_begin_catch(i8* %exn) #3
+  %exn.byref = bitcast i8* %4 to %class.bar*
+  store %class.bar* %exn.byref, %class.bar** %e, align 8
+  store i32 1, i32* %retval, align 4
+  call void @__cxa_end_catch()
+  br label %try.cont
+
+try.cont:                                         ; preds = %invoke.cont, %catch
+  %5 = load i32, i32* %retval, align 4
+  ret i32 %5
+
+eh.resume:                                        ; preds = %catch.dispatch
+  %exn1 = load i8*, i8** %exn.slot, align 8
+  %sel2 = load i32, i32* %ehselector.slot, align 4
+  %lpad.val = insertvalue { i8*, i32 } undef, i8* %exn1, 0
+  %lpad.val3 = insertvalue { i8*, i32 } %lpad.val, i32 %sel2, 1
+  resume { i8*, i32 } %lpad.val3
+}
+
+declare i32 @__gxx_personality_v0(...)
+
+; Function Attrs: nounwind readnone
+declare i32 @llvm.eh.typeid.for(i8*) #2
+
+declare i8* @__cxa_begin_catch(i8*)
+
+declare void @__cxa_end_catch()