diff --git a/llvm/lib/Target/AMDGPU/AMDGPUReplaceLDSUseWithPointer.cpp b/llvm/lib/Target/AMDGPU/AMDGPUReplaceLDSUseWithPointer.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPUReplaceLDSUseWithPointer.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUReplaceLDSUseWithPointer.cpp
@@ -180,6 +180,37 @@
     return LDSPointer;
   }
 
+  // As an optimization, split the entry block of kernel K just after top most
+  // static alloca cluster.
+  //
+  // From the better code transformation/optimization perspective, it is
+  // expected that all static alloca appear as a single contiguous cluster at
+  // the start of the entry block. If this canonical form is *not* maintained,
+  // then few static alloca may become dynamic after the entry block split here.
+  Instruction *findEntryBlockSplitPoint(Function *K) {
+    auto &EBB = K->getEntryBlock();
+    auto Iter = EBB.getFirstInsertionPt();
+    auto *EI = &*Iter;
+    auto *TI = EBB.getTerminator();
+
+    // Locate the first static alloca.
+    while ((&*Iter != TI) && !isa<CallBase>(&*Iter) && !isa<AllocaInst>(&*Iter))
+      ++Iter;
+
+    // Top most static alloca cluster not found? Split the block at the first
+    // insertion point.
+    if (!isa<AllocaInst>(&*Iter))
+      return EI;
+
+    // We have found (at least one) static alloca at the top of the entry block.
+    // Skip them.
+    while (isa<AllocaInst>(&*Iter))
+      ++Iter;
+
+    // Split the block just after top most static alloca cluster.
+    return &*Iter;
+  }
+
   // Split entry basic block in such a way that only lane 0 of each wave does
   // the LDS pointer initialization, and return newly created basic block.
   BasicBlock *activateLaneZero(Function *K) {
@@ -189,9 +220,8 @@
     if (!BasicBlockEntry.second)
       return BasicBlockEntry.first->second;
 
-    // Split entry basic block of kernel K.
-    auto *EI = &(*(K->getEntryBlock().getFirstInsertionPt()));
-    IRBuilder<> Builder(EI);
+    // Split entry basic block of kernel K just after top static alloca cluster.
+    IRBuilder<> Builder(findEntryBlockSplitPoint(K));
 
     Value *Mbcnt =
         Builder.CreateIntrinsic(Intrinsic::amdgcn_mbcnt_lo, {},
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
@@ -295,7 +295,7 @@
 
 static cl::opt<bool> EnableLDSReplaceWithPointer(
     "amdgpu-enable-lds-replace-with-pointer",
-    cl::desc("Enable LDS replace with pointer pass"), cl::init(false),
+    cl::desc("Enable LDS replace with pointer pass"), cl::init(true),
     cl::Hidden);
 
 static cl::opt<bool, true> EnableLowerModuleLDS(
diff --git a/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll b/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
--- a/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
+++ b/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
@@ -42,6 +42,7 @@
 ; GCN-O0-NEXT:      Inliner for always_inline functions
 ; GCN-O0-NEXT:    A No-Op Barrier Pass
 ; GCN-O0-NEXT:    Lower OpenCL enqueued blocks
+; GCN-O0-NEXT:    Replace within non-kernel function use of LDS with pointer
 ; GCN-O0-NEXT:    Lower uses of LDS variables from non-kernel functions
 ; GCN-O0-NEXT:    FunctionPass Manager
 ; GCN-O0-NEXT:      Expand Atomic instructions
@@ -178,6 +179,7 @@
 ; GCN-O1-NEXT:      Inliner for always_inline functions
 ; GCN-O1-NEXT:    A No-Op Barrier Pass
 ; GCN-O1-NEXT:    Lower OpenCL enqueued blocks
+; GCN-O1-NEXT:    Replace within non-kernel function use of LDS with pointer
 ; GCN-O1-NEXT:    Lower uses of LDS variables from non-kernel functions
 ; GCN-O1-NEXT:    FunctionPass Manager
 ; GCN-O1-NEXT:      Infer address spaces
@@ -429,6 +431,7 @@
 ; GCN-O1-OPTS-NEXT:      Inliner for always_inline functions
 ; GCN-O1-OPTS-NEXT:    A No-Op Barrier Pass
 ; GCN-O1-OPTS-NEXT:    Lower OpenCL enqueued blocks
+; GCN-O1-OPTS-NEXT:    Replace within non-kernel function use of LDS with pointer
 ; GCN-O1-OPTS-NEXT:    Lower uses of LDS variables from non-kernel functions
 ; GCN-O1-OPTS-NEXT:    FunctionPass Manager
 ; GCN-O1-OPTS-NEXT:      Infer address spaces
@@ -713,6 +716,7 @@
 ; GCN-O2-NEXT:      Inliner for always_inline functions
 ; GCN-O2-NEXT:    A No-Op Barrier Pass
 ; GCN-O2-NEXT:    Lower OpenCL enqueued blocks
+; GCN-O2-NEXT:    Replace within non-kernel function use of LDS with pointer
 ; GCN-O2-NEXT:    Lower uses of LDS variables from non-kernel functions
 ; GCN-O2-NEXT:    FunctionPass Manager
 ; GCN-O2-NEXT:      Infer address spaces
@@ -999,6 +1003,7 @@
 ; GCN-O3-NEXT:      Inliner for always_inline functions
 ; GCN-O3-NEXT:    A No-Op Barrier Pass
 ; GCN-O3-NEXT:    Lower OpenCL enqueued blocks
+; GCN-O3-NEXT:    Replace within non-kernel function use of LDS with pointer
 ; GCN-O3-NEXT:    Lower uses of LDS variables from non-kernel functions
 ; GCN-O3-NEXT:    FunctionPass Manager
 ; GCN-O3-NEXT:      Infer address spaces
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-diamond-shape.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-diamond-shape.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-diamond-shape.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-diamond-shape.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-selected_functions.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-selected_functions.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-selected_functions.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-selected_functions.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-to-declare-only-func.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-to-declare-only-func.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-to-declare-only-func.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-call-to-declare-only-func.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-global-scope-use.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-global-scope-use.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-global-scope-use.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-global-scope-use.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-inline-asm-call.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-inline-asm-call.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-inline-asm-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-inline-asm-call.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-kernel-only-used-lds.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-kernel-only-used-lds.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-kernel-only-used-lds.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-kernel-only-used-lds.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION ;
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-not-reachable-lds.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-not-reachable-lds.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-not-reachable-lds.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-not-reachable-lds.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION ;
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-small-lds.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-small-lds.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-small-lds.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-ignore-small-lds.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION ;
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-diamond-shape.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-diamond-shape.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-diamond-shape.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-diamond-shape.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-selected_functions.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-selected_functions.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-selected_functions.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-selected_functions.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-signature-match.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-signature-match.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-signature-match.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-indirect-call-signature-match.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-split-entry-bb-after-top-static-alloca-cluster.ll
copy from llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll
copy to llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-split-entry-bb-after-top-static-alloca-cluster.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-split-entry-bb-after-top-static-alloca-cluster.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
@@ -7,6 +7,9 @@
 ; reachable from kernel. Hence nested constant expression should to be converted into a
 ; series of instructons and pointer replacement should take place.
 ;
+; Further the entry basic block of the kernel @k0 contains alloca instruction. Hence the
+; entry basic splitting for pointer initialization should happen after alloca.
+;
 
 ; Original LDS should exist.
 ; CHECK: @used_only_within_func = addrspace(3) global [4 x i32] undef, align 4
@@ -36,19 +39,31 @@
 ; Pointer initialization code shoud be added
 define amdgpu_kernel void @k0() {
 ; CHECK-LABEL: entry:
-; CHECK:   %0 = call i32 @llvm.amdgcn.mbcnt.lo(i32 -1, i32 0)
-; CHECK:   %1 = icmp eq i32 %0, 0
-; CHECK:   br i1 %1, label %2, label %3
+; CHECK:   %0 = alloca i64, align 8, addrspace(5)
+; CHECK:   %1 = alloca i64, align 8, addrspace(5)
+; CHECK:   %2 = alloca i64, align 8, addrspace(5)
+; CHECK:   %3 = call i32 @llvm.amdgcn.mbcnt.lo(i32 -1, i32 0)
+; CHECK:   %4 = icmp eq i32 %3, 0
+; CHECK:   br i1 %4, label %5, label %6
 ;
-; CHECK-LABEL: 2:
+; CHECK-LABEL: 5:
 ; CHECK:   store i16 ptrtoint ([4 x i32] addrspace(3)* @used_only_within_func to i16), i16 addrspace(3)* @used_only_within_func.ptr, align 2
-; CHECK:   br label %3
-;
-; CHECK-LABEL: 3:
+; CHECK:   br label %6
+
+; CHECK-LABEL: 6:
 ; CHECK:   call void @llvm.amdgcn.wave.barrier()
+; CHECK:   %7 = addrspacecast i64 addrspace(5)* %0 to i64*
+; CHECK:   %8 = addrspacecast i64 addrspace(5)* %1 to i64*
+; CHECK:   %9 = addrspacecast i64 addrspace(5)* %2 to i64*
 ; CHECK:   call void @f0(i32 0)
 ; CHECK:   ret void
 entry:
+  %0 = alloca i64, align 8, addrspace(5)
+  %1 = alloca i64, align 8, addrspace(5)
+  %2 = alloca i64, align 8, addrspace(5)
+  %3 = addrspacecast i64 addrspace(5)* %0 to i64*
+  %4 = addrspacecast i64 addrspace(5)* %1 to i64*
+  %5 = addrspacecast i64 addrspace(5)* %2 to i64*
   call void @f0(i32 0)
   ret void
 }
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-multiple-lds.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-multiple-lds.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-multiple-lds.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-multiple-lds.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-same-lds.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-same-lds.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-same-lds.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-same-lds.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr1.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr2.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr2.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr2.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-const-expr2.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ; There is one lds global defined here, and this lds is used within a single non-kernel
diff --git a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-phi-inst.ll b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-phi-inst.ll
--- a/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-phi-inst.ll
+++ b/llvm/test/CodeGen/AMDGPU/replace-lds-by-ptr-use-within-phi-inst.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -mtriple=amdgcn--  -amdgpu-replace-lds-use-with-pointer -amdgpu-enable-lds-replace-with-pointer=true < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn-- -amdgpu-replace-lds-use-with-pointer < %s | FileCheck %s
 
 ; DESCRIPTION:
 ;