diff --git a/clang/lib/CodeGen/CGDeclCXX.cpp b/clang/lib/CodeGen/CGDeclCXX.cpp
--- a/clang/lib/CodeGen/CGDeclCXX.cpp
+++ b/clang/lib/CodeGen/CGDeclCXX.cpp
@@ -544,6 +544,8 @@
 
   llvm::GlobalVariable *COMDATKey =
       supportsCOMDAT() && D->isExternallyVisible() ? Addr : nullptr;
+  bool IsInstantiation =
+      isTemplateInstantiation(D->getTemplateSpecializationKind());
 
   if (D->getTLSKind()) {
     // FIXME: Should we support init_priority for thread_local?
@@ -557,7 +559,7 @@
     OrderGlobalInitsOrStermFinalizers Key(IPA->getPriority(),
                                           PrioritizedCXXGlobalInits.size());
     PrioritizedCXXGlobalInits.push_back(std::make_pair(Key, Fn));
-  } else if (isTemplateInstantiation(D->getTemplateSpecializationKind()) ||
+  } else if (IsInstantiation ||
              getContext().GetGVALinkageForVariable(D) == GVA_DiscardableODR ||
              D->hasAttr<SelectAnyAttr>()) {
     // C++ [basic.start.init]p2:
@@ -577,7 +579,7 @@
     // COMDAT group associated with the global, so the initializers get folded
     // too.
 
-    AddGlobalCtor(Fn, 65535, COMDATKey);
+    AddGlobalCtor(Fn, 65535, COMDATKey, IsInstantiation);
     if (COMDATKey && (getTriple().isOSBinFormatELF() ||
                       getTarget().getCXXABI().isMicrosoft())) {
       // When COMDAT is used on ELF or in the MS C++ ABI, the key must be in
diff --git a/clang/lib/CodeGen/CodeGenModule.h b/clang/lib/CodeGen/CodeGenModule.h
--- a/clang/lib/CodeGen/CodeGenModule.h
+++ b/clang/lib/CodeGen/CodeGenModule.h
@@ -1544,7 +1544,8 @@
 
   // FIXME: Hardcoding priority here is gross.
   void AddGlobalCtor(llvm::Function *Ctor, int Priority = 65535,
-                     llvm::Constant *AssociatedData = nullptr);
+                     llvm::Constant *AssociatedData = nullptr,
+                     bool InsertFront = false);
   void AddGlobalDtor(llvm::Function *Dtor, int Priority = 65535,
                      bool IsDtorAttrFunc = false);
 
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -1556,9 +1556,13 @@
 /// AddGlobalCtor - Add a function to the list that will be called before
 /// main() runs.
 void CodeGenModule::AddGlobalCtor(llvm::Function *Ctor, int Priority,
-                                  llvm::Constant *AssociatedData) {
+                                  llvm::Constant *AssociatedData,
+                                  bool InsertFront) {
   // FIXME: Type coercion of void()* types.
-  GlobalCtors.push_back(Structor(Priority, Ctor, AssociatedData));
+  if (InsertFront)
+    GlobalCtors.emplace(GlobalCtors.begin(), Priority, Ctor, AssociatedData);
+  else
+    GlobalCtors.emplace_back(Priority, Ctor, AssociatedData);
 }
 
 /// AddGlobalDtor - Add a function to the list that will be called
diff --git a/clang/test/CodeGenCXX/aix-static-init-temp-spec-and-inline-var.cpp b/clang/test/CodeGenCXX/aix-static-init-temp-spec-and-inline-var.cpp
--- a/clang/test/CodeGenCXX/aix-static-init-temp-spec-and-inline-var.cpp
+++ b/clang/test/CodeGenCXX/aix-static-init-temp-spec-and-inline-var.cpp
@@ -47,7 +47,7 @@
 // CHECK: @_ZGVN5test12t2E = linkonce_odr global i64 0, align 8
 // CHECK: @_ZGVN5test21AIvE8instanceE = weak_odr global i64 0, align 8
 // CHECK: @_ZGVN5test12t1IiEE = linkonce_odr global i64 0, align 8
-// CHECK: @llvm.global_ctors = appending global [4 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 65535, void ()* @__cxx_global_var_init.1, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @__cxx_global_var_init.2, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @__cxx_global_var_init.4, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @_GLOBAL__sub_I__, i8* null }]
+// CHECK: @llvm.global_ctors = appending global [4 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 65535, void ()* @__cxx_global_var_init.4, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @__cxx_global_var_init.2, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @__cxx_global_var_init.1, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @_GLOBAL__sub_I__, i8* null }]
 // CHECK: @llvm.global_dtors = appending global [4 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 65535, void ()* @__finalize__ZN5test12t2E, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @__finalize__ZN5test21AIvE8instanceE, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @__finalize__ZN5test12t1IiEE, i8* null }, { i32, void ()*, i8* } { i32 65535, void ()* @_GLOBAL__D_a, i8* null }]
 
 // CHECK: define internal void @__cxx_global_var_init() [[ATTR:#[0-9]+]] {
diff --git a/clang/test/CodeGenCXX/microsoft-abi-static-initializers.cpp b/clang/test/CodeGenCXX/microsoft-abi-static-initializers.cpp
--- a/clang/test/CodeGenCXX/microsoft-abi-static-initializers.cpp
+++ b/clang/test/CodeGenCXX/microsoft-abi-static-initializers.cpp
@@ -1,10 +1,10 @@
 // RUN: %clang_cc1 -no-opaque-pointers -fms-extensions -fno-threadsafe-statics -emit-llvm %s -o - -mconstructor-aliases -triple=i386-pc-win32 | FileCheck %s
 
 // CHECK: @llvm.global_ctors = appending global [5 x { i32, void ()*, i8* }] [
+// CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @"??__E?foo@?$B@H@@2VA@@A@@YAXXZ", i8* bitcast (%class.A* @"?foo@?$B@H@@2VA@@A" to i8*) },
+// CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @"??__E?s@?$ExportedTemplate@H@@2US@@A@@YAXXZ", i8* getelementptr inbounds (%struct.S, %struct.S* @"?s@?$ExportedTemplate@H@@2US@@A", i32 0, i32 0) },
 // CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @"??__Eselectany1@@YAXXZ", i8* getelementptr inbounds (%struct.S, %struct.S* @"?selectany1@@3US@@A", i32 0, i32 0) },
 // CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @"??__Eselectany2@@YAXXZ", i8* getelementptr inbounds (%struct.S, %struct.S* @"?selectany2@@3US@@A", i32 0, i32 0) },
-// CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @"??__E?s@?$ExportedTemplate@H@@2US@@A@@YAXXZ", i8* getelementptr inbounds (%struct.S, %struct.S* @"?s@?$ExportedTemplate@H@@2US@@A", i32 0, i32 0) },
-// CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @"??__E?foo@?$B@H@@2VA@@A@@YAXXZ", i8* bitcast (%class.A* @"?foo@?$B@H@@2VA@@A" to i8*) },
 // CHECK: { i32, void ()*, i8* } { i32 65535, void ()* @_GLOBAL__sub_I_microsoft_abi_static_initializers.cpp, i8* null }
 // CHECK: ]
 
diff --git a/clang/test/CodeGenCXX/static-member-variable-explicit-specialization.cpp b/clang/test/CodeGenCXX/static-member-variable-explicit-specialization.cpp
--- a/clang/test/CodeGenCXX/static-member-variable-explicit-specialization.cpp
+++ b/clang/test/CodeGenCXX/static-member-variable-explicit-specialization.cpp
@@ -29,32 +29,41 @@
 // ALL: @_ZN1AIbE1aE ={{.*}} global i32 10
 template<> int A<bool>::a = 10;
 
-// ALL: @llvm.global_ctors = appending global [8 x { i32, void ()*, i8* }]
+// ALL: @llvm.global_ctors = appending global [11 x { i32, void ()*, i8* }]
 
-// ELF: [{ i32, void ()*, i8* } { i32 65535, void ()* @[[unordered1:[^,]*]], i8* bitcast (i32* @_ZN1AIsE1aE to i8*) },
-// MACHO: [{ i32, void ()*, i8* } { i32 65535, void ()* @[[unordered1:[^,]*]], i8* null },
+// ELF:  [{ i32, void ()*, i8* } { i32 65535, void ()* @[[unordered10:[^,]*]], i8* bitcast (i32* @_ZN1RILi1EE1aE to i8*) },
+// MACHO: [{ i32, void ()*, i8* } { i32 65535, void ()* @[[unordered10:[^,]*]], i8* null },
 
-// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered2:[^,]*]], i8* bitcast (i16* @_Z1xIsE to i8*) },
-// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered2:[^,]*]], i8* null },
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered9:[^,]*]], i8* bitcast (i32* @_ZN1RILi2EE1aE to i8*) },
+// MACHO: { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered9:[^,]*]], i8* null },
 
-// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered3:[^,]*]], i8* bitcast (i32* @_ZN2ns1aIiE1iE to i8*) },
-// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered3:[^,]*]], i8* null },
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered8:[^,]*]], i8* bitcast (i32* @_ZN1RILi3EE1aE to i8*) },
+// MACHO: { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered8:[^,]*]], i8* null },
 
-// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered4:[^,]*]], i8* bitcast (i32* @_ZN2ns1b1iIiEE to i8*) },
-// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered4:[^,]*]], i8* null },
+// ALL:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered7:[^,]*]], i8* null },
+
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered6:[^,]*]], i8* @_Z1xIcE },
+// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered6:[^,]*]], i8* null },
 
 // ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered5:[^,]*]], i8* bitcast (i32* @_ZN1AIvE1aE to i8*) },
 // MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered5:[^,]*]], i8* null },
 
-// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered6:[^,]*]], i8* @_Z1xIcE },
-// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered6:[^,]*]], i8* null },
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered4:[^,]*]], i8* bitcast (i32* @_ZN2ns1b1iIiEE to i8*) },
+// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered4:[^,]*]], i8* null },
 
-// ALL:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered7:[^,]*]], i8* null },
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered3:[^,]*]], i8* bitcast (i32* @_ZN2ns1aIiE1iE to i8*) },
+// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered3:[^,]*]], i8* null },
+
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered2:[^,]*]], i8* bitcast (i16* @_Z1xIsE to i8*) },
+// MACHO:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered2:[^,]*]], i8* null },
+
+// ELF:  { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered1:[^,]*]], i8* bitcast (i32* @_ZN1AIsE1aE to i8*) },
+// MACHO: { i32, void ()*, i8* } { i32 65535, void ()* @[[unordered1:[^,]*]], i8* null },
 
 // ALL:  { i32, void ()*, i8* } { i32 65535, void ()* @_GLOBAL__sub_I_static_member_variable_explicit_specialization.cpp, i8* null }]
 
 /// llvm.used ensures SHT_INIT_ARRAY in a section group cannot be GCed.
-// ELF: @llvm.used = appending global [6 x i8*] [i8* bitcast (i32* @_ZN1AIsE1aE to i8*), i8* bitcast (i16* @_Z1xIsE to i8*), i8* bitcast (i32* @_ZN2ns1aIiE1iE to i8*), i8* bitcast (i32* @_ZN2ns1b1iIiEE to i8*), i8* bitcast (i32* @_ZN1AIvE1aE to i8*), i8* @_Z1xIcE]
+// ELF: @llvm.used = appending global [9 x i8*] [i8* bitcast (i32* @_ZN1AIsE1aE to i8*), i8* bitcast (i16* @_Z1xIsE to i8*), i8* bitcast (i32* @_ZN2ns1aIiE1iE to i8*), i8* bitcast (i32* @_ZN2ns1b1iIiEE to i8*), i8* bitcast (i32* @_ZN1AIvE1aE to i8*), i8* @_Z1xIcE, i8* bitcast (i32* @_ZN1RILi3EE1aE to i8*), i8* bitcast (i32* @_ZN1RILi2EE1aE to i8*), i8* bitcast (i32* @_ZN1RILi1EE1aE to i8*)]
 
 template int A<short>::a;  // Unordered
 int b = foo();
@@ -94,6 +103,11 @@
 }
 int *use_internal_a = &Internal<int>::a;
 
+template<int n> struct R { static int a; };
+template<> int R<0>::a = 0;
+template<int n> int R<n>::a = R<n - 1>::a + 1;
+int f = R<3>::a;
+
 #endif
 
 // ALL: define internal void @[[unordered1]](
@@ -131,6 +145,18 @@
 // ALL: store {{.*}} @_ZN12_GLOBAL__N_18InternalIiE1aE
 // ALL: ret
 
+// ALL: define internal void @[[unordered8]](
+// ALL: store {{.*}} @_ZGVN1RILi3EE1aE
+// ALL: ret
+
+// ALL: define internal void @[[unordered9]](
+// ALL: store {{.*}} @_ZGVN1RILi2EE1aE
+// ALL: ret
+
+// ALL: define internal void @[[unordered10]](
+// ALL: store {{.*}} @_ZGVN1RILi1EE1aE
+// ALL: ret
+
 // ALL: define internal void @_GLOBAL__sub_I_static_member_variable_explicit_specialization.cpp()
 //   We call unique stubs for every ordered dynamic initializer in the TU.
 // ALL: call
@@ -141,5 +167,6 @@
 // ALL: call
 // ALL: call
 // ALL: call
+// ALL: call
 // ALL-NOT: call
 // ALL: ret
diff --git a/clang/test/Modules/initializers.cpp b/clang/test/Modules/initializers.cpp
--- a/clang/test/Modules/initializers.cpp
+++ b/clang/test/Modules/initializers.cpp
@@ -154,12 +154,12 @@
 
 // It's OK if the order of the first 6 of these changes.
 // CHECK: @llvm.global_ctors = appending global
-// CHECK-SAME: @[[E_INIT:[^,]*]], {{[^@]*}} @[[E]]
-// CHECK-SAME: @[[F_INIT:[^,]*]], {{[^@]*}} @[[F]]
-// CHECK-SAME: @[[XA_INIT:[^,]*]], {{[^@]*}} @[[XA]]
-// CHECK-SAME: @[[XE_INIT:[^,]*]], {{[^@]*}} @[[XE]]
-// CHECK-SAME: @[[XF_INIT:[^,]*]], {{[^@]*}} @[[XF]]
 // CHECK-SAME: @[[XB_INIT:[^,]*]], {{[^@]*}} @[[XB]]
+// CHECK-SAME: @[[XF_INIT:[^,]*]], {{[^@]*}} @[[XF]]
+// CHECK-SAME: @[[XE_INIT:[^,]*]], {{[^@]*}} @[[XE]]
+// CHECK-SAME: @[[XA_INIT:[^,]*]], {{[^@]*}} @[[XA]]
+// CHECK-SAME: @[[F_INIT:[^,]*]], {{[^@]*}} @[[F]]
+// CHECK-SAME: @[[E_INIT:[^,]*]], {{[^@]*}} @[[E]]
 // CHECK-IMPORT-SAME: @[[TU_INIT:[^,]*]], i8* null }]
 
 // FIXME: Should this use __cxa_guard_acquire?