Index: llvm/lib/Transforms/Scalar/LICM.cpp
===================================================================
--- llvm/lib/Transforms/Scalar/LICM.cpp
+++ llvm/lib/Transforms/Scalar/LICM.cpp
@@ -75,6 +75,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/raw_ostream.h"
+#include "llvm/Target/TargetOptions.h"
 #include "llvm/Transforms/Scalar.h"
 #include "llvm/Transforms/Utils/AssumeBundleBuilder.h"
 #include "llvm/Transforms/Utils/BasicBlockUtils.h"
@@ -2111,7 +2112,7 @@
   // stores along paths which originally didn't have them without violating the
   // memory model.
   if (!SafeToInsertStore) {
-    if (IsKnownThreadLocalObject)
+    if (IsKnownThreadLocalObject || ThreadModel::Single)
       SafeToInsertStore = true;
     else {
       Value *Object = getUnderlyingObject(SomePtr);
Index: llvm/test/Transforms/LICM/hoist-load-without-store.ll
===================================================================
--- llvm/test/Transforms/LICM/hoist-load-without-store.ll
+++ llvm/test/Transforms/LICM/hoist-load-without-store.ll
@@ -26,13 +26,15 @@
 ; CHECK-NEXT:    [[TOBOOL_NOT:%.*]] = icmp eq i32 [[TMP0]], 0
 ; CHECK-NEXT:    br i1 [[TOBOOL_NOT]], label [[IF_END]], label [[FOR_BODY_CLEANUP1_CRIT_EDGE:%.*]]
 ; CHECK:       if.end:
-; CHECK-NEXT:    store i32 1, i32* [[PTR]], align 4
 ; CHECK-NEXT:    [[INC]] = add nuw nsw i32 [[I_08]], 1
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[INC]], [[N]]
 ; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP1_CRIT_EDGE:%.*]]
 ; CHECK:       for.body.cleanup1_crit_edge:
+; CHECK-NEXT:    [[DOTLCSSA:%.*]] = phi i32 [ [[TMP0]], [[FOR_BODY]] ]
+; CHECK-NEXT:    store i32 [[DOTLCSSA]], i32* [[PTR]], align 4
 ; CHECK-NEXT:    br label [[CLEANUP1]]
 ; CHECK:       for.cond.cleanup1_crit_edge:
+; CHECK-NEXT:    store i32 1, i32* [[PTR]], align 4
 ; CHECK-NEXT:    br label [[CLEANUP1]]
 ; CHECK:       cleanup1:
 ; CHECK-NEXT:    ret void
Index: llvm/test/Transforms/LICM/promote-capture.ll
===================================================================
--- llvm/test/Transforms/LICM/promote-capture.ll
+++ llvm/test/Transforms/LICM/promote-capture.ll
@@ -120,7 +120,6 @@
 ; CHECK-NEXT:    br i1 [[COND]], label [[IF:%.*]], label [[LATCH]]
 ; CHECK:       if:
 ; CHECK-NEXT:    [[C_INC:%.*]] = add i32 [[C_INC2]], 1
-; CHECK-NEXT:    store i32 [[C_INC]], i32* [[COUNT]], align 4
 ; CHECK-NEXT:    br label [[LATCH]]
 ; CHECK:       latch:
 ; CHECK-NEXT:    [[C_INC1]] = phi i32 [ [[C_INC]], [[IF]] ], [ [[C_INC2]], [[LOOP]] ]
@@ -128,6 +127,8 @@
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[I_NEXT]], [[LEN:%.*]]
 ; CHECK-NEXT:    br i1 [[CMP]], label [[EXIT:%.*]], label [[LOOP]]
 ; CHECK:       exit:
+; CHECK-NEXT:    [[C_INC1_LCSSA:%.*]] = phi i32 [ [[C_INC1]], [[LATCH]] ]
+; CHECK-NEXT:    store i32 [[C_INC1_LCSSA]], i32* [[COUNT]], align 4
 ; CHECK-NEXT:    ret void
 ;
 entry:
Index: llvm/test/Transforms/LICM/reg-promote.ll
===================================================================
--- /dev/null
+++ llvm/test/Transforms/LICM/reg-promote.ll
@@ -0,0 +1,135 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -licm -S %s | FileCheck %s
+
+target triple = "aarch64-unknown-linux"
+
+@u = dso_local global i32 0, align 4
+@v = dso_local global i32 0, align 4
+@restrict = dso_local global i32 0, align 4
+@i = dso_local global i32 0, align 4
+
+; Function Attrs: mustprogress nounwind uwtable
+define dso_local void @_Z1fPiS_i(ptr noundef %0, ptr noundef %1, i32 noundef %2) #0 {
+; CHECK-LABEL: @_Z1fPiS_i(
+; CHECK-NEXT:    [[TMP4:%.*]] = alloca ptr, align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = alloca ptr, align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = alloca i32, align 4
+; CHECK-NEXT:    store ptr [[TMP0:%.*]], ptr [[TMP4]], align 8
+; CHECK-NEXT:    store ptr [[TMP1:%.*]], ptr [[TMP5]], align 8
+; CHECK-NEXT:    store i32 [[TMP2:%.*]], ptr [[TMP6]], align 4
+; CHECK-NEXT:    [[TMP7:%.*]] = load i32, ptr @restrict, align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = zext i32 [[TMP7]] to i64
+; CHECK-NEXT:    [[TMP9:%.*]] = load i32, ptr @restrict, align 4
+; CHECK-NEXT:    [[TMP10:%.*]] = zext i32 [[TMP9]] to i64
+; CHECK-NEXT:    store i32 0, ptr @i, align 4
+; CHECK-NEXT:    [[TMP11:%.*]] = load i32, ptr [[TMP6]], align 4
+; CHECK-NEXT:    [[TMP12:%.*]] = load ptr, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP13:%.*]] = load ptr, ptr [[TMP5]], align 8
+; CHECK-NEXT:    br label [[TMP14:%.*]]
+; CHECK:       14:
+; CHECK-NEXT:    [[TMP15:%.*]] = load i32, ptr @i, align 4
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp slt i32 [[TMP15]], [[TMP11]]
+; CHECK-NEXT:    br i1 [[TMP16]], label [[TMP17:%.*]], label [[DOTLOOPEXIT:%.*]]
+; CHECK:       17:
+; CHECK-NEXT:    [[TMP18:%.*]] = load i32, ptr @i, align 4
+; CHECK-NEXT:    [[TMP19:%.*]] = sext i32 [[TMP18]] to i64
+; CHECK-NEXT:    [[TMP20:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i64 [[TMP19]]
+; CHECK-NEXT:    [[TMP21:%.*]] = load i32, ptr [[TMP20]], align 4
+; CHECK-NEXT:    [[TMP22:%.*]] = icmp ne i32 [[TMP21]], 0
+; CHECK-NEXT:    br i1 [[TMP22]], label [[TMP23:%.*]], label [[TMP26:%.*]]
+; CHECK:       23:
+; CHECK-NEXT:    [[TMP24:%.*]] = load i32, ptr @u, align 4
+; CHECK-NEXT:    [[TMP25:%.*]] = add nsw i32 [[TMP24]], 1
+; CHECK-NEXT:    store i32 [[TMP25]], ptr @u, align 4
+; CHECK-NEXT:    br label [[TMP41:%.*]]
+; CHECK:       26:
+; CHECK-NEXT:    [[TMP27:%.*]] = load i32, ptr @u, align 4
+; CHECK-NEXT:    [[TMP28:%.*]] = add nsw i32 [[TMP27]], 1
+; CHECK-NEXT:    store i32 [[TMP28]], ptr @u, align 4
+; CHECK-NEXT:    [[TMP29:%.*]] = load i32, ptr @i, align 4
+; CHECK-NEXT:    [[TMP30:%.*]] = sext i32 [[TMP29]] to i64
+; CHECK-NEXT:    [[TMP31:%.*]] = getelementptr inbounds i32, ptr [[TMP13]], i64 [[TMP30]]
+; CHECK-NEXT:    [[TMP32:%.*]] = load i32, ptr [[TMP31]], align 4
+; CHECK-NEXT:    [[TMP33:%.*]] = icmp ne i32 [[TMP32]], 0
+; CHECK-NEXT:    br i1 [[TMP33]], label [[TMP34:%.*]], label [[TMP37:%.*]]
+; CHECK:       34:
+; CHECK-NEXT:    [[TMP35:%.*]] = load i32, ptr @v, align 4
+; CHECK-NEXT:    [[TMP36:%.*]] = add nsw i32 [[TMP35]], 1
+; CHECK-NEXT:    store i32 [[TMP36]], ptr @v, align 4
+; CHECK-NEXT:    br label [[TMP37]]
+; CHECK:       37:
+; CHECK-NEXT:    br label [[TMP38:%.*]]
+; CHECK:       38:
+; CHECK-NEXT:    [[TMP39:%.*]] = load i32, ptr @i, align 4
+; CHECK-NEXT:    [[TMP40:%.*]] = add nsw i32 [[TMP39]], 1
+; CHECK-NEXT:    store i32 [[TMP40]], ptr @i, align 4
+; CHECK-NEXT:    br label [[TMP14]]
+; CHECK:       .loopexit:
+; CHECK-NEXT:    br label [[TMP41]]
+; CHECK:       41:
+; CHECK-NEXT:    ret void
+;
+  %4 = alloca ptr, align 8
+  %5 = alloca ptr, align 8
+  %6 = alloca i32, align 4
+  store ptr %0, ptr %4, align 8
+  store ptr %1, ptr %5, align 8
+  store i32 %2, ptr %6, align 4
+  %7 = load i32, ptr @restrict, align 4
+  %8 = zext i32 %7 to i64
+  %9 = load i32, ptr @restrict, align 4
+  %10 = zext i32 %9 to i64
+  store i32 0, ptr @i, align 4
+  br label %11
+
+11:                                               ; preds = %38, %3
+  %12 = load i32, ptr @i, align 4
+  %13 = load i32, ptr %6, align 4
+  %14 = icmp slt i32 %12, %13
+  br i1 %14, label %15, label %41
+
+15:                                               ; preds = %11
+  %16 = load ptr, ptr %4, align 8
+  %17 = load i32, ptr @i, align 4
+  %18 = sext i32 %17 to i64
+  %19 = getelementptr inbounds i32, ptr %16, i64 %18
+  %20 = load i32, ptr %19, align 4
+  %21 = icmp ne i32 %20, 0
+  br i1 %21, label %22, label %25
+
+22:                                               ; preds = %15
+  %23 = load i32, ptr @u, align 4
+  %24 = add nsw i32 %23, 1
+  store i32 %24, ptr @u, align 4
+  br label %41
+
+25:                                               ; preds = %15
+  %26 = load i32, ptr @u, align 4
+  %27 = add nsw i32 %26, 1
+  store i32 %27, ptr @u, align 4
+  %28 = load ptr, ptr %5, align 8
+  %29 = load i32, ptr @i, align 4
+  %30 = sext i32 %29 to i64
+  %31 = getelementptr inbounds i32, ptr %28, i64 %30
+  %32 = load i32, ptr %31, align 4
+  %33 = icmp ne i32 %32, 0
+  br i1 %33, label %34, label %37
+
+34:                                               ; preds = %25
+  %35 = load i32, ptr @v, align 4
+  %36 = add nsw i32 %35, 1
+  store i32 %36, ptr @v, align 4
+  br label %37
+
+37:                                               ; preds = %34, %25
+  br label %38
+
+38:                                               ; preds = %37
+  %39 = load i32, ptr @i, align 4
+  %40 = add nsw i32 %39, 1
+  store i32 %40, ptr @i, align 4
+  br label %11
+
+41:                                               ; preds = %22, %11
+  ret void
+}
Index: llvm/test/Transforms/LICM/scalar-promote-memmodel.ll
===================================================================
--- llvm/test/Transforms/LICM/scalar-promote-memmodel.ll
+++ llvm/test/Transforms/LICM/scalar-promote-memmodel.ll
@@ -22,13 +22,14 @@
 ; CHECK-NEXT:    br i1 [[TOBOOL]], label [[FOR_INC]], label [[IF_THEN:%.*]]
 ; CHECK:       if.then:
 ; CHECK-NEXT:    [[INC:%.*]] = add nsw i32 [[INC2]], 1
-; CHECK-NEXT:    store i32 [[INC]], i32* @g, align 4
 ; CHECK-NEXT:    br label [[FOR_INC]]
 ; CHECK:       for.inc:
 ; CHECK-NEXT:    [[INC1]] = phi i32 [ [[INC]], [[IF_THEN]] ], [ [[INC2]], [[FOR_BODY]] ]
 ; CHECK-NEXT:    [[INC5]] = add nsw i32 [[I_0]], 1
 ; CHECK-NEXT:    br label [[FOR_COND]]
 ; CHECK:       for.end:
+; CHECK-NEXT:    [[INC2_LCSSA:%.*]] = phi i32 [ [[INC2]], [[FOR_COND]] ]
+; CHECK-NEXT:    store i32 [[INC2_LCSSA]], i32* @g, align 4
 ; CHECK-NEXT:    ret void
 ;
 entry:
Index: llvm/test/Transforms/LICM/scalar-promote-opaque-ptrs.ll
===================================================================
--- llvm/test/Transforms/LICM/scalar-promote-opaque-ptrs.ll
+++ llvm/test/Transforms/LICM/scalar-promote-opaque-ptrs.ll
@@ -323,7 +323,6 @@
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[X2]], 0
 ; CHECK-NEXT:    br i1 [[CMP]], label [[IF:%.*]], label [[ELSE]]
 ; CHECK:       if:
-; CHECK-NEXT:    store i32 [[X2]], ptr [[LOCAL]], align 4
 ; CHECK-NEXT:    br label [[ELSE]]
 ; CHECK:       else:
 ; CHECK-NEXT:    [[X21]] = phi i32 [ [[X2]], [[IF]] ], [ [[X22]], [[LOOP]] ]
@@ -331,6 +330,8 @@
 ; CHECK-NEXT:    [[COND:%.*]] = icmp eq i32 [[NEXT]], 0
 ; CHECK-NEXT:    br i1 [[COND]], label [[EXIT:%.*]], label [[LOOP]]
 ; CHECK:       exit:
+; CHECK-NEXT:    [[X21_LCSSA:%.*]] = phi i32 [ [[X21]], [[ELSE]] ]
+; CHECK-NEXT:    store i32 [[X21_LCSSA]], ptr [[LOCAL]], align 4
 ; CHECK-NEXT:    [[RET:%.*]] = load i32, ptr [[LOCAL]], align 4
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
Index: llvm/test/Transforms/LICM/scalar-promote.ll
===================================================================
--- llvm/test/Transforms/LICM/scalar-promote.ll
+++ llvm/test/Transforms/LICM/scalar-promote.ll
@@ -325,7 +325,6 @@
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[X2]], 0
 ; CHECK-NEXT:    br i1 [[CMP]], label [[IF:%.*]], label [[ELSE]]
 ; CHECK:       if:
-; CHECK-NEXT:    store i32 [[X2]], i32* [[LOCAL]], align 4
 ; CHECK-NEXT:    br label [[ELSE]]
 ; CHECK:       else:
 ; CHECK-NEXT:    [[X21]] = phi i32 [ [[X2]], [[IF]] ], [ [[X22]], [[LOOP]] ]
@@ -333,6 +332,8 @@
 ; CHECK-NEXT:    [[COND:%.*]] = icmp eq i32 [[NEXT]], 0
 ; CHECK-NEXT:    br i1 [[COND]], label [[EXIT:%.*]], label [[LOOP]]
 ; CHECK:       exit:
+; CHECK-NEXT:    [[X21_LCSSA:%.*]] = phi i32 [ [[X21]], [[ELSE]] ]
+; CHECK-NEXT:    store i32 [[X21_LCSSA]], i32* [[LOCAL]], align 4
 ; CHECK-NEXT:    [[RET:%.*]] = load i32, i32* [[LOCAL]], align 4
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
@@ -611,11 +612,12 @@
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i8 [[I]], 4
 ; CHECK-NEXT:    br i1 [[CMP]], label [[LOOP_LATCH]], label [[EXIT:%.*]]
 ; CHECK:       loop.latch:
-; CHECK-NEXT:    store i8 [[INC1]], i8* [[PTR]], align 1
 ; CHECK-NEXT:    [[ADD]] = add i8 [[I]], [[INC1]]
 ; CHECK-NEXT:    br label [[LOOP_HEADER]]
 ; CHECK:       exit:
+; CHECK-NEXT:    [[INC1_LCSSA:%.*]] = phi i8 [ [[INC1]], [[LOOP_HEADER]] ]
 ; CHECK-NEXT:    [[I_LCSSA:%.*]] = phi i8 [ [[I]], [[LOOP_HEADER]] ]
+; CHECK-NEXT:    store i8 [[INC1_LCSSA]], i8* [[PTR]], align 1
 ; CHECK-NEXT:    ret i8 [[I_LCSSA]]
 ;
 entry: