AMDGPU: Split flat offsets that don't fit in DAG

We handle it this way for some other address spaces.

Since r349196, SILoadStoreOptimizer has been trying to do this. This
is after SIFoldOperands runs, which can change the addressing
patterns. It's simpler to just split this earlier.

llvm-svn: 375366

12 files changed, 443 insertions, 376 deletions

diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
index 62904dc2380..f768a10a720 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
@@ -262,6 +262,8 @@ private:
 
   SDValue getHi16Elt(SDValue In) const;
 
+  SDValue getMaterializedScalarImm32(int64_t Val, const SDLoc &DL) const;
+
   void SelectADD_SUB_I64(SDNode *N);
   void SelectAddcSubb(SDNode *N);
   void SelectUADDO_USUBO(SDNode *N);
@@ -963,6 +965,14 @@ bool AMDGPUDAGToDAGISel::SelectADDRIndirect(SDValue Addr, SDValue &Base,
   return true;
 }
 
+SDValue AMDGPUDAGToDAGISel::getMaterializedScalarImm32(int64_t Val,
+                                                       const SDLoc &DL) const {
+  SDNode *Mov = CurDAG->getMachineNode(
+    AMDGPU::S_MOV_B32, DL, MVT::i32,
+    CurDAG->getTargetConstant(Val, DL, MVT::i32));
+  return SDValue(Mov, 0);
+}
+
 // FIXME: Should only handle addcarry/subcarry
 void AMDGPUDAGToDAGISel::SelectADD_SUB_I64(SDNode *N) {
   SDLoc DL(N);
@@ -1630,13 +1640,80 @@ bool AMDGPUDAGToDAGISel::SelectFlatOffset(SDNode *N,
       CurDAG->isBaseWithConstantOffset(Addr)) {
     SDValue N0 = Addr.getOperand(0);
     SDValue N1 = Addr.getOperand(1);
-    int64_t COffsetVal = cast<ConstantSDNode>(N1)->getSExtValue();
+    uint64_t COffsetVal = cast<ConstantSDNode>(N1)->getSExtValue();
 
     const SIInstrInfo *TII = Subtarget->getInstrInfo();
-    if (TII->isLegalFLATOffset(COffsetVal, findMemSDNode(N)->getAddressSpace(),
-                               IsSigned)) {
+    unsigned AS = findMemSDNode(N)->getAddressSpace();
+    if (TII->isLegalFLATOffset(COffsetVal, AS, IsSigned)) {
       Addr = N0;
       OffsetVal = COffsetVal;
+    } else {
+      // If the offset doesn't fit, put the low bits into the offset field and
+      // add the rest.
+
+      SDLoc DL(N);
+      uint64_t ImmField;
+      const unsigned NumBits = TII->getNumFlatOffsetBits(AS, IsSigned);
+      if (IsSigned) {
+        ImmField = SignExtend64(COffsetVal, NumBits);
+
+        // Don't use a negative offset field if the base offset is positive.
+        // Since the scheduler currently relies on the offset field, doing so
+        // could result in strange scheduling decisions.
+
+        // TODO: Should we not do this in the opposite direction as well?
+        if (static_cast<int64_t>(COffsetVal) > 0) {
+          if (static_cast<int64_t>(ImmField) < 0) {
+            const uint64_t OffsetMask = maskTrailingOnes<uint64_t>(NumBits - 1);
+            ImmField = COffsetVal & OffsetMask;
+          }
+        }
+      } else {
+        // TODO: Should we do this for a negative offset?
+        const uint64_t OffsetMask = maskTrailingOnes<uint64_t>(NumBits);
+        ImmField = COffsetVal & OffsetMask;
+      }
+
+      uint64_t RemainderOffset = COffsetVal - ImmField;
+
+      assert(TII->isLegalFLATOffset(ImmField, AS, IsSigned));
+      assert(RemainderOffset + ImmField == COffsetVal);
+
+      OffsetVal = ImmField;
+
+      // TODO: Should this try to use a scalar add pseudo if the base address is
+      // uniform and saddr is usable?
+      SDValue Sub0 = CurDAG->getTargetConstant(AMDGPU::sub0, DL, MVT::i32);
+      SDValue Sub1 = CurDAG->getTargetConstant(AMDGPU::sub1, DL, MVT::i32);
+
+      SDNode *N0Lo = CurDAG->getMachineNode(TargetOpcode::EXTRACT_SUBREG,
+                                            DL, MVT::i32, N0, Sub0);
+      SDNode *N0Hi = CurDAG->getMachineNode(TargetOpcode::EXTRACT_SUBREG,
+                                            DL, MVT::i32, N0, Sub1);
+
+      SDValue AddOffsetLo
+        = getMaterializedScalarImm32(Lo_32(RemainderOffset), DL);
+      SDValue AddOffsetHi
+        = getMaterializedScalarImm32(Hi_32(RemainderOffset), DL);
+
+      SDVTList VTs = CurDAG->getVTList(MVT::i32, MVT::i1);
+      SDValue Clamp = CurDAG->getTargetConstant(0, DL, MVT::i1);
+
+      SDNode *Add = CurDAG->getMachineNode(
+        AMDGPU::V_ADD_I32_e64, DL, VTs,
+        {AddOffsetLo, SDValue(N0Lo, 0), Clamp});
+
+      SDNode *Addc = CurDAG->getMachineNode(
+        AMDGPU::V_ADDC_U32_e64, DL, VTs,
+        {AddOffsetHi, SDValue(N0Hi, 0), SDValue(Add, 1), Clamp});
+
+      SDValue RegSequenceArgs[] = {
+        CurDAG->getTargetConstant(AMDGPU::VReg_64RegClassID, DL, MVT::i32),
+        SDValue(Add, 0), Sub0, SDValue(Addc, 0), Sub1
+      };
+
+      Addr = SDValue(CurDAG->getMachineNode(AMDGPU::REG_SEQUENCE, DL,
+                                            MVT::i64, RegSequenceArgs), 0);
     }
   }
 
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
index 6bc2a0ca08e..88dc938e2b8 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
@@ -6275,6 +6275,20 @@ bool SIInstrInfo::isBufferSMRD(const MachineInstr &MI) const {
   return RI.getRegClass(RCID)->hasSubClassEq(&AMDGPU::SGPR_128RegClass);
 }
 
+unsigned SIInstrInfo::getNumFlatOffsetBits(unsigned AddrSpace,
+                                           bool Signed) const {
+  if (!ST.hasFlatInstOffsets())
+    return 0;
+
+  if (ST.hasFlatSegmentOffsetBug() && AddrSpace == AMDGPUAS::FLAT_ADDRESS)
+    return 0;
+
+  if (ST.getGeneration() >= AMDGPUSubtarget::GFX10)
+    return Signed ? 12 : 11;
+
+  return Signed ? 13 : 12;
+}
+
 bool SIInstrInfo::isLegalFLATOffset(int64_t Offset, unsigned AddrSpace,
                                     bool Signed) const {
   // TODO: Should 0 be special cased?
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.h b/llvm/lib/Target/AMDGPU/SIInstrInfo.h
index 72073e9ce39..be463442c88 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.h
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.h
@@ -1004,6 +1004,8 @@ public:
     return isUInt<12>(Imm);
   }
 
+  unsigned getNumFlatOffsetBits(unsigned AddrSpace, bool Signed) const;
+
   /// Returns if \p Offset is legal for the subtarget as the offset to a FLAT
   /// encoded instruction. If \p Signed, this is for an instruction that
   /// interprets the offset as signed.
diff --git a/llvm/test/CodeGen/AMDGPU/cgp-addressing-modes.ll b/llvm/test/CodeGen/AMDGPU/cgp-addressing-modes.ll
index 47ca6054f52..1cdf1d391f2 100644
--- a/llvm/test/CodeGen/AMDGPU/cgp-addressing-modes.ll
+++ b/llvm/test/CodeGen/AMDGPU/cgp-addressing-modes.ll
@@ -44,7 +44,10 @@ done:
 ; GCN-LABEL: {{^}}test_sink_global_small_max_i32_ds_offset:
 ; GCN: s_and_saveexec_b64
 ; SICIVI: buffer_load_sbyte {{v[0-9]+}}, off, {{s\[[0-9]+:[0-9]+\]}}, s{{[0-9]+$}}
-; GFX9: global_load_sbyte {{v[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, off{{$}}
+
+; GFX9: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0xf000,
+; GFX9: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, 0,
+; GFX9: global_load_sbyte {{v[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, off offset:4095{{$}}
 ; GCN: {{^}}BB1_2:
 ; GCN: s_or_b64 exec
 define amdgpu_kernel void @test_sink_global_small_max_i32_ds_offset(i32 addrspace(1)* %out, i8 addrspace(1)* %in) {
diff --git a/llvm/test/CodeGen/AMDGPU/flat-address-space.ll b/llvm/test/CodeGen/AMDGPU/flat-address-space.ll
index 5c45528f9df..e48e8c96cb5 100644
--- a/llvm/test/CodeGen/AMDGPU/flat-address-space.ll
+++ b/llvm/test/CodeGen/AMDGPU/flat-address-space.ll
@@ -1,14 +1,7 @@
-; RUN: llc -O0 -mtriple=amdgcn-mesa-mesa3d -mcpu=bonaire < %s | FileCheck -check-prefixes=CHECK,CIVI %s
-; RUN: llc -O0 -mtriple=amdgcn-mesa-mesa3d -mcpu=tonga -mattr=-flat-for-global < %s | FileCheck -check-prefixes=CHECK,CIVI %s
-; RUN: llc -O0 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global < %s | FileCheck -check-prefixes=CHECK,HSA %s
-; RUN: llc -O0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=CHECK,HSA,GFX9 %s
-
-; Disable optimizations in case there are optimizations added that
-; specialize away generic pointer accesses.
-
-
-; These testcases might become useless when there are optimizations to
-; remove generic pointers.
+; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=bonaire < %s | FileCheck -check-prefixes=CHECK,CIVI %s
+; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=tonga -mattr=-flat-for-global < %s | FileCheck -check-prefixes=CHECK,CIVI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global < %s | FileCheck -check-prefixes=CHECK,CIVI,HSA %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global < %s | FileCheck -check-prefixes=CHECK,HSA,GFX9 %s
 
 ; CHECK-LABEL: {{^}}store_flat_i32:
 ; CHECK-DAG: s_load_dwordx2 s{{\[}}[[LO_SREG:[0-9]+]]:[[HI_SREG:[0-9]+]]],
@@ -191,7 +184,11 @@ define amdgpu_kernel void @store_flat_i8_max_offset_p1(i8* %fptr, i8 %x) #0 {
 }
 
 ; CHECK-LABEL: {{^}}store_flat_i8_neg_offset:
-; CHECK: flat_store_byte v{{\[[0-9]+:[0-9]+\]}}, v{{[0-9]+}}{{$}}
+; CIVI: flat_store_byte v{{\[[0-9]+:[0-9]+\]}}, v{{[0-9]+}}{{$}}
+
+; GFX9: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0xfffff000, v
+; GFX9: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, -1,
+; GFX9: flat_store_byte v{{\[[0-9]+:[0-9]+\]}}, v{{[0-9]+}} offset:4094{{$}}
 define amdgpu_kernel void @store_flat_i8_neg_offset(i8* %fptr, i8 %x) #0 {
   %fptr.offset = getelementptr inbounds i8, i8* %fptr, i64 -2
   store volatile i8 %x, i8* %fptr.offset
@@ -216,7 +213,11 @@ define amdgpu_kernel void @load_flat_i8_max_offset_p1(i8* %fptr) #0 {
 }
 
 ; CHECK-LABEL: {{^}}load_flat_i8_neg_offset:
-; CHECK: flat_load_ubyte v{{[0-9]+}}, v{{\[[0-9]+:[0-9]+\]}}{{$}}
+; CIVI: flat_load_ubyte v{{[0-9]+}}, v{{\[[0-9]+:[0-9]+\]}}{{$}}
+
+; GFX9: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0xfffff000, v
+; GFX9: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, -1,
+; GFX9: flat_load_ubyte v{{[0-9]+}}, v{{\[[0-9]+:[0-9]+\]}} offset:4094{{$}}
 define amdgpu_kernel void @load_flat_i8_neg_offset(i8* %fptr) #0 {
   %fptr.offset = getelementptr inbounds i8, i8* %fptr, i64 -2
   %val = load volatile i8, i8* %fptr.offset
diff --git a/llvm/test/CodeGen/AMDGPU/global-saddr.ll b/llvm/test/CodeGen/AMDGPU/global-saddr.ll
index b21fd985226..4df1ad683df 100644
--- a/llvm/test/CodeGen/AMDGPU/global-saddr.ll
+++ b/llvm/test/CodeGen/AMDGPU/global-saddr.ll
@@ -46,7 +46,7 @@ entry:
 
 ; Test various offset boundaries.
 ; GFX9: global_load_dwordx2 v{{\[[0-9]+:[0-9]+\]}}, v{{\[[0-9]+:[0-9]+\]}}, s[{{[0-9]+}}:{{[0-9]+}}] offset:4088{{$}}
-; GFX9: global_load_dwordx2 v{{\[[0-9]+:[0-9]+\]}}, v{{\[[0-9]+:[0-9]+\]}}, off{{$}}
+; GFX9: global_load_dwordx2 v{{\[[0-9]+:[0-9]+\]}}, v{{\[[0-9]+:[0-9]+\]}}, off offset:4088{{$}}
 ; GFX9: global_load_dwordx4 v{{\[[0-9]+:[0-9]+\]}}, v{{\[[0-9]+:[0-9]+\]}}, s[{{[0-9]+}}:{{[0-9]+}}] offset:2040{{$}}
   %gep11 = getelementptr inbounds i64, i64 addrspace(1)* %gep, i64 511
   %load11 = load i64, i64 addrspace(1)* %gep11
diff --git a/llvm/test/CodeGen/AMDGPU/global_atomics.ll b/llvm/test/CodeGen/AMDGPU/global_atomics.ll
index 72f748578b3..618c7008307 100644
--- a/llvm/test/CodeGen/AMDGPU/global_atomics.ll
+++ b/llvm/test/CodeGen/AMDGPU/global_atomics.ll
@@ -25,7 +25,9 @@ entry:
 ; SIVI: s_mov_b32 [[SREG:s[0-9]+]], 0x8ca0
 ; SIVI: buffer_atomic_add v{{[0-9]+}}, off, s[{{[0-9]+}}:{{[0-9]+}}], [[SREG]]{{$}}
 
-; GFX9: global_atomic_add v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}}, off{{$}}
+; GFX9: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0x8000,
+; GFX9-NEXT: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, 0, v{{[0-9]+}}, vcc
+; GFX9: global_atomic_add v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}}, off offset:3232{{$}}
 define amdgpu_kernel void @atomic_add_i32_soffset(i32 addrspace(1)* %out, i32 %in) {
 entry:
   %gep = getelementptr i32, i32 addrspace(1)* %out, i64 9000
@@ -40,7 +42,10 @@ entry:
 
 ; VI: flat_atomic_add
 
-; GFX9: global_atomic_add v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}}, off{{$}}
+; GFX9: v_mov_b32_e32 [[HIGH_K:v[0-9]+]], 0xabcd
+; GFX9: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0xd000,
+; GFX9-NEXT: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, [[HIGH_K]], v{{[0-9]+}}, vcc
+; GFX9: global_atomic_add v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}}, off offset:3756{{$}}
 define amdgpu_kernel void @atomic_add_i32_huge_offset(i32 addrspace(1)* %out, i32 %in) {
 entry:
   %gep = getelementptr i32, i32 addrspace(1)* %out, i64 47224239175595
diff --git a/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll b/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll
index 0ef58fc1fb8..b996646a098 100644
--- a/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll
+++ b/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll
@@ -991,7 +991,9 @@ entry:
 ; CIVI: s_mov_b32 [[SREG:s[0-9]+]], 0x11940
 ; CIVI: buffer_atomic_cmpswap_x2 v[{{[0-9]+}}:{{[0-9]+}}], off, s[{{[0-9]+}}:{{[0-9]+}}], [[SREG]]{{$}}
 
-; GFX9: global_atomic_cmpswap_x2 v[{{[0-9]+:[0-9]+}}], v{{\[[0-9]+:[0-9]+\]}}, off{{$}}
+; GFX9: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0x11000,
+; GFX9: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, 0, v{{[0-9]+}}, vcc
+; GFX9: global_atomic_cmpswap_x2 v[{{[0-9]+:[0-9]+}}], v{{\[[0-9]+:[0-9]+\]}}, off offset:2368{{$}}
 define amdgpu_kernel void @atomic_cmpxchg_i64_soffset(i64 addrspace(1)* %out, i64 %in, i64 %old) {
 entry:
   %gep = getelementptr i64, i64 addrspace(1)* %out, i64 9000
diff --git a/llvm/test/CodeGen/AMDGPU/offset-split-flat.ll b/llvm/test/CodeGen/AMDGPU/offset-split-flat.ll
index 5763c53ff39..f7538c081e6 100644
--- a/llvm/test/CodeGen/AMDGPU/offset-split-flat.ll
+++ b/llvm/test/CodeGen/AMDGPU/offset-split-flat.ll
@@ -78,9 +78,9 @@ define i8 @flat_inst_valu_offset_13bit_max(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_13bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -103,9 +103,9 @@ define i8 @flat_inst_valu_offset_neg_11bit_max(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_neg_11bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfffff800, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfffff000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -201,9 +201,9 @@ define i8 @flat_inst_valu_offset_2x_12bit_max(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_2x_12bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -226,9 +226,9 @@ define i8 @flat_inst_valu_offset_2x_13bit_max(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_2x_13bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x3fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x3000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -327,9 +327,9 @@ define i8 @flat_inst_valu_offset_64bit_11bit_split0(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_11bit_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x7ff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -353,9 +353,9 @@ define i8 @flat_inst_valu_offset_64bit_11bit_split1(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_11bit_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x800, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -379,9 +379,9 @@ define i8 @flat_inst_valu_offset_64bit_12bit_split0(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_12bit_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -431,9 +431,9 @@ define i8 @flat_inst_valu_offset_64bit_13bit_split0(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_13bit_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -483,10 +483,10 @@ define i8 @flat_inst_valu_offset_64bit_11bit_neg_high_split0(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_11bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x7ff, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -510,10 +510,10 @@ define i8 @flat_inst_valu_offset_64bit_11bit_neg_high_split1(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_11bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x800, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -537,10 +537,10 @@ define i8 @flat_inst_valu_offset_64bit_12bit_neg_high_split0(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_12bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfff, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -566,7 +566,7 @@ define i8 @flat_inst_valu_offset_64bit_12bit_neg_high_split1(i8* %p) {
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
@@ -591,10 +591,10 @@ define i8 @flat_inst_valu_offset_64bit_13bit_neg_high_split0(i8* %p) {
 ; GFX9-LABEL: flat_inst_valu_offset_64bit_13bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -620,7 +620,7 @@ define i8 @flat_inst_valu_offset_64bit_13bit_neg_high_split1(i8* %p) {
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
@@ -738,11 +738,11 @@ define amdgpu_kernel void @flat_inst_salu_offset_13bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -771,11 +771,11 @@ define amdgpu_kernel void @flat_inst_salu_offset_neg_11bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfffff800
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfffff000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -804,10 +804,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_neg_12bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfffff000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfffff000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -837,10 +837,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_neg_13bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xffffe000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xffffe000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -901,11 +901,11 @@ define amdgpu_kernel void @flat_inst_salu_offset_2x_12bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -934,11 +934,11 @@ define amdgpu_kernel void @flat_inst_salu_offset_2x_13bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x3fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x3000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -967,10 +967,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_2x_neg_11bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfffff000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfffff000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -1000,10 +1000,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_2x_neg_12bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xffffe000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xffffe000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -1033,10 +1033,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_2x_neg_13bit_max(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xffffc000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xffffc000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -1067,11 +1067,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_11bit_split0(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x7ff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1101,11 +1100,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_11bit_split1(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x800
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1135,11 +1133,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_12bit_split0(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1169,10 +1166,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_12bit_split1(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -1203,11 +1200,11 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_13bit_split0(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1237,10 +1234,10 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_13bit_split1(i8* %p) {
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x2000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -1270,12 +1267,12 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_11bit_neg_high_split0(i8*
 ; GFX9-LABEL: flat_inst_salu_offset_64bit_11bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x7ff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1304,12 +1301,12 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_11bit_neg_high_split1(i8*
 ; GFX9-LABEL: flat_inst_salu_offset_64bit_11bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x800
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1338,12 +1335,12 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_12bit_neg_high_split0(i8*
 ; GFX9-LABEL: flat_inst_salu_offset_64bit_12bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1372,11 +1369,12 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_12bit_neg_high_split1(i8*
 ; GFX9-LABEL: flat_inst_salu_offset_64bit_12bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
@@ -1406,12 +1404,13 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_13bit_neg_high_split0(i8*
 ; GFX9-LABEL: flat_inst_salu_offset_64bit_13bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    flat_load_ubyte v0, v[0:1] offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
 ; GFX9-NEXT:    s_endpgm
@@ -1440,11 +1439,12 @@ define amdgpu_kernel void @flat_inst_salu_offset_64bit_13bit_neg_high_split1(i8*
 ; GFX9-LABEL: flat_inst_salu_offset_64bit_13bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x2000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
 ; GFX9-NEXT:    flat_load_ubyte v0, v[0:1]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    flat_store_byte v[0:1], v0
diff --git a/llvm/test/CodeGen/AMDGPU/offset-split-global.ll b/llvm/test/CodeGen/AMDGPU/offset-split-global.ll
index ad32b739bdb..add4e687926 100644
--- a/llvm/test/CodeGen/AMDGPU/offset-split-global.ll
+++ b/llvm/test/CodeGen/AMDGPU/offset-split-global.ll
@@ -59,10 +59,10 @@ define i8 @global_inst_valu_offset_12bit_max(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0xfff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 4095
@@ -74,9 +74,9 @@ define i8 @global_inst_valu_offset_13bit_max(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_13bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -84,10 +84,10 @@ define i8 @global_inst_valu_offset_13bit_max(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1fff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 8191
@@ -176,10 +176,10 @@ define i8 @global_inst_valu_offset_2x_11bit_max(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0xfff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 4095
@@ -191,9 +191,9 @@ define i8 @global_inst_valu_offset_2x_12bit_max(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_2x_12bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -201,10 +201,10 @@ define i8 @global_inst_valu_offset_2x_12bit_max(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1fff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 8191
@@ -216,9 +216,9 @@ define i8 @global_inst_valu_offset_2x_13bit_max(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_2x_13bit_max:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x3fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x3000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -226,10 +226,10 @@ define i8 @global_inst_valu_offset_2x_13bit_max(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x3fff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x3800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 16383
@@ -315,9 +315,9 @@ define i8 @global_inst_valu_offset_64bit_11bit_split0(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_64bit_11bit_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x7ff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -325,10 +325,10 @@ define i8 @global_inst_valu_offset_64bit_11bit_split0(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x7ff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 2, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 8589936639
@@ -341,9 +341,9 @@ define i8 @global_inst_valu_offset_64bit_11bit_split1(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_64bit_11bit_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x800, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -367,9 +367,9 @@ define i8 @global_inst_valu_offset_64bit_12bit_split0(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_64bit_12bit_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -377,10 +377,10 @@ define i8 @global_inst_valu_offset_64bit_12bit_split0(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0xfff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 2, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 8589938687
@@ -419,9 +419,9 @@ define i8 @global_inst_valu_offset_64bit_13bit_split0(i8 addrspace(1)* %p) {
 ; GFX9-LABEL: global_inst_valu_offset_64bit_13bit_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -429,10 +429,10 @@ define i8 @global_inst_valu_offset_64bit_13bit_split0(i8 addrspace(1)* %p) {
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1fff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1800, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 2, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 8589942783
@@ -471,10 +471,10 @@ define i8 @global_inst_valu_offset_64bit_11bit_neg_high_split0(i8 addrspace(1)*
 ; GFX9-LABEL: global_inst_valu_offset_64bit_11bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x7ff, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -482,10 +482,10 @@ define i8 @global_inst_valu_offset_64bit_11bit_neg_high_split0(i8 addrspace(1)*
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x7ff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 -9223372036854773761
@@ -498,10 +498,10 @@ define i8 @global_inst_valu_offset_64bit_11bit_neg_high_split1(i8 addrspace(1)*
 ; GFX9-LABEL: global_inst_valu_offset_64bit_11bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x800, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -509,10 +509,10 @@ define i8 @global_inst_valu_offset_64bit_11bit_neg_high_split1(i8 addrspace(1)*
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x800, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1000, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:-2048
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 -9223372036854773760
@@ -525,10 +525,10 @@ define i8 @global_inst_valu_offset_64bit_12bit_neg_high_split0(i8 addrspace(1)*
 ; GFX9-LABEL: global_inst_valu_offset_64bit_12bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xfff, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -536,10 +536,10 @@ define i8 @global_inst_valu_offset_64bit_12bit_neg_high_split0(i8 addrspace(1)*
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0xfff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1000, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:-1
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 -9223372036854771713
@@ -552,10 +552,10 @@ define i8 @global_inst_valu_offset_64bit_12bit_neg_high_split1(i8 addrspace(1)*
 ; GFX9-LABEL: global_inst_valu_offset_64bit_12bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:-4096
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -579,10 +579,10 @@ define i8 @global_inst_valu_offset_64bit_13bit_neg_high_split0(i8 addrspace(1)*
 ; GFX9-LABEL: global_inst_valu_offset_64bit_13bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1fff, v0
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:-1
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
@@ -590,10 +590,10 @@ define i8 @global_inst_valu_offset_64bit_13bit_neg_high_split0(i8 addrspace(1)*
 ; GFX10:       ; %bb.0:
 ; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX10-NEXT:    s_waitcnt_vscnt null, 0x0
-; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1fff, v0
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x2000, v0
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:-1
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    s_setpc_b64 s[30:31]
   %gep = getelementptr i8, i8 addrspace(1)* %p, i64 -9223372036854767617
@@ -608,7 +608,7 @@ define i8 @global_inst_valu_offset_64bit_13bit_neg_high_split1(i8 addrspace(1)*
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
 ; GFX9-NEXT:    v_bfrev_b32_e32 v2, 1
-; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
@@ -703,11 +703,9 @@ define amdgpu_kernel void @global_inst_salu_offset_12bit_max(i8 addrspace(1)* %p
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 0, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -722,11 +720,11 @@ define amdgpu_kernel void @global_inst_salu_offset_13bit_max(i8 addrspace(1)* %p
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -736,11 +734,9 @@ define amdgpu_kernel void @global_inst_salu_offset_13bit_max(i8 addrspace(1)* %p
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x1800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 0, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -796,10 +792,8 @@ define amdgpu_kernel void @global_inst_salu_offset_neg_12bit_max(i8 addrspace(1)
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xfffff000
-; GFX10-NEXT:    s_addc_u32 s1, s1, -1
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0xfffff000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, -1, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -815,10 +809,10 @@ define amdgpu_kernel void @global_inst_salu_offset_neg_13bit_max(i8 addrspace(1)
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xffffe000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xffffe000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
@@ -829,10 +823,8 @@ define amdgpu_kernel void @global_inst_salu_offset_neg_13bit_max(i8 addrspace(1)
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xffffe000
-; GFX10-NEXT:    s_addc_u32 s1, s1, -1
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0xffffe000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, -1, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -860,11 +852,9 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_11bit_max(i8 addrspace(1)*
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 0, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -879,11 +869,11 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_12bit_max(i8 addrspace(1)*
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -893,11 +883,9 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_12bit_max(i8 addrspace(1)*
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x1800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 0, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -912,11 +900,11 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_13bit_max(i8 addrspace(1)*
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x3fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x3000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 0, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -926,11 +914,9 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_13bit_max(i8 addrspace(1)*
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x3fff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x3800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 0, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -957,10 +943,8 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_neg_11bit_max(i8 addrspace
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xfffff000
-; GFX10-NEXT:    s_addc_u32 s1, s1, -1
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0xfffff000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, -1, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -976,10 +960,10 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_neg_12bit_max(i8 addrspace
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xffffe000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xffffe000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
@@ -990,10 +974,8 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_neg_12bit_max(i8 addrspace
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xffffe000
-; GFX10-NEXT:    s_addc_u32 s1, s1, -1
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0xffffe000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, -1, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -1009,10 +991,10 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_neg_13bit_max(i8 addrspace
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xffffc000
-; GFX9-NEXT:    s_addc_u32 s1, s1, -1
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0xffffc000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, -1, v1, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
@@ -1023,10 +1005,8 @@ define amdgpu_kernel void @global_inst_salu_offset_2x_neg_13bit_max(i8 addrspace
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xffffc000
-; GFX10-NEXT:    s_addc_u32 s1, s1, -1
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0xffffc000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, -1, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -1043,11 +1023,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_split0(i8 addrspa
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x7ff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1057,11 +1036,9 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_split0(i8 addrspa
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x7ff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 2
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 2, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1077,11 +1054,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_split1(i8 addrspa
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x800
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1091,10 +1067,8 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_split1(i8 addrspa
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x800
-; GFX10-NEXT:    s_addc_u32 s1, s1, 2
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 2, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -1111,11 +1085,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_split0(i8 addrspa
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1125,11 +1098,9 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_split0(i8 addrspa
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 2
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 2, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1145,10 +1116,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_split1(i8 addrspa
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
@@ -1159,10 +1130,8 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_split1(i8 addrspa
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x1000
-; GFX10-NEXT:    s_addc_u32 s1, s1, 2
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x1000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 2, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -1179,11 +1148,11 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_split0(i8 addrspa
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x1000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1193,11 +1162,9 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_split0(i8 addrspa
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 2
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x1800, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 2, s1, s0
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1213,10 +1180,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_split1(i8 addrspa
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x2000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 2
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, 2, v1, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
@@ -1227,10 +1194,8 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_split1(i8 addrspa
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x2000
-; GFX10-NEXT:    s_addc_u32 s1, s1, 2
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
-; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, s0, 0x2000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e64 v1, s0, 2, s1, s0
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -1246,12 +1211,12 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_neg_high_split0(i
 ; GFX9-LABEL: global_inst_salu_offset_64bit_11bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x7ff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1261,11 +1226,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_neg_high_split0(i
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x7ff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:2047
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1280,12 +1244,12 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_neg_high_split1(i
 ; GFX9-LABEL: global_inst_salu_offset_64bit_11bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x800
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:2048
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1295,11 +1259,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_11bit_neg_high_split1(i
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x800
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:-2048
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1314,12 +1277,12 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_neg_high_split0(i
 ; GFX9-LABEL: global_inst_salu_offset_64bit_12bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e64 v0, vcc, 0, s0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:4095
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1329,11 +1292,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_neg_high_split0(i
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0xfff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:-1
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1348,12 +1310,13 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_neg_high_split1(i
 ; GFX9-LABEL: global_inst_salu_offset_64bit_12bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:-4096
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1363,10 +1326,9 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_12bit_neg_high_split1(i
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x1000
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x1000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
@@ -1382,12 +1344,13 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_neg_high_split0(i
 ; GFX9-LABEL: global_inst_salu_offset_64bit_13bit_neg_high_split0:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
-; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
+; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off offset:-1
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX9-NEXT:    s_endpgm
@@ -1397,11 +1360,10 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_neg_high_split0(i
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x1fff
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX10-NEXT:    v_mov_b32_e32 v1, s1
-; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x2000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
+; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off offset:-1
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
 ; GFX10-NEXT:    s_endpgm
@@ -1416,11 +1378,12 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_neg_high_split1(i
 ; GFX9-LABEL: global_inst_salu_offset_64bit_13bit_neg_high_split1:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
+; GFX9-NEXT:    v_bfrev_b32_e32 v1, 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_add_u32 s0, s0, 0x2000
-; GFX9-NEXT:    s_addc_u32 s1, s1, 0x80000000
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s0
-; GFX9-NEXT:    v_mov_b32_e32 v1, s1
+; GFX9-NEXT:    v_mov_b32_e32 v2, s1
+; GFX9-NEXT:    v_add_co_u32_e32 v0, vcc, 0x2000, v0
+; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v1, v2, vcc
 ; GFX9-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    global_store_byte v[0:1], v0, off
@@ -1431,10 +1394,9 @@ define amdgpu_kernel void @global_inst_salu_offset_64bit_13bit_neg_high_split1(i
 ; GFX10-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX10-NEXT:    ; implicit-def: $vcc_hi
 ; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX10-NEXT:    s_add_u32 s0, s0, 0x2000
-; GFX10-NEXT:    s_addc_u32 s1, s1, 0x80000000
-; GFX10-NEXT:    v_mov_b32_e32 v0, s0
 ; GFX10-NEXT:    v_mov_b32_e32 v1, s1
+; GFX10-NEXT:    v_add_co_u32_e64 v0, vcc_lo, 0x2000, s0
+; GFX10-NEXT:    v_add_co_ci_u32_e32 v1, vcc_lo, 0x80000000, v1, vcc_lo
 ; GFX10-NEXT:    global_load_ubyte v0, v[0:1], off
 ; GFX10-NEXT:    s_waitcnt vmcnt(0)
 ; GFX10-NEXT:    global_store_byte v[0:1], v0, off
diff --git a/llvm/test/CodeGen/AMDGPU/promote-constOffset-to-imm.ll b/llvm/test/CodeGen/AMDGPU/promote-constOffset-to-imm.ll
index a9fc318ce0e..d9a3d29785b 100644
--- a/llvm/test/CodeGen/AMDGPU/promote-constOffset-to-imm.ll
+++ b/llvm/test/CodeGen/AMDGPU/promote-constOffset-to-imm.ll
@@ -14,14 +14,15 @@ define amdgpu_kernel void @clmem_read_simplified(i8 addrspace(1)*  %buffer) {
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ;
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+
 entry:
   %call = tail call i64 @_Z13get_global_idj(i32 0)
   %conv = and i64 %call, 255
@@ -75,15 +76,15 @@ define hidden amdgpu_kernel void @clmem_read(i8 addrspace(1)*  %buffer) {
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ;
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
 entry:
@@ -184,7 +185,7 @@ while.end:                                        ; preds = %while.cond.loopexit
 }
 
 ; using 32bit address.
-define amdgpu_kernel void @Address32(i8 addrspace(1)*  %buffer) {
+define amdgpu_kernel void @Address32(i8 addrspace(1)* %buffer) {
 ; GCN-LABEL: Address32:
 ; GFX8:    flat_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}]
@@ -197,16 +198,16 @@ define amdgpu_kernel void @Address32(i8 addrspace(1)*  %buffer) {
 ; GFX8:    flat_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}]
 ;
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:1024
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:1024
 ; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:2048
 ; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:3072
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:-4096
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:-3072
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:-1024
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off{{$}}
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:1024
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:3072
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off{{$}}
 entry:
    %call = tail call i64 @_Z13get_global_idj(i32 0)
    %conv = and i64 %call, 255
@@ -265,10 +266,10 @@ define amdgpu_kernel void @Offset64(i8 addrspace(1)*  %buffer) {
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ;
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 entry:
   %call = tail call i64 @_Z13get_global_idj(i32 0)
   %conv = and i64 %call, 255
@@ -306,10 +307,10 @@ define amdgpu_kernel void @p32Offset64(i8 addrspace(1)*  %buffer) {
 ; GFX8:    flat_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}]
 ;
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:-1024
-; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off offset:3072
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off{{$}}
+; GFX9:    global_load_dword {{v[0-9]+}}, v[{{[0-9]+:[0-9]+}}], off{{$}}
 entry:
   %call = tail call i64 @_Z13get_global_idj(i32 0)
   %conv = and i64 %call, 255
@@ -347,13 +348,13 @@ define amdgpu_kernel void @DiffBase(i8 addrspace(1)* %buffer1,
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
-;
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
+
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
                                     i8 addrspace(1)* %buffer2) {
 entry:
   %call = tail call i64 @_Z13get_global_idj(i32 0)
@@ -403,13 +404,13 @@ define amdgpu_kernel void @ReverseOrder(i8 addrspace(1)* %buffer) {
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ; GFX8:    flat_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}]
 ;
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:-4096
-; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
+; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off{{$}}
 ; GFX9:    global_load_dwordx2 v[{{[0-9]+:[0-9]+}}], v[{{[0-9]+:[0-9]+}}], off offset:2048
 entry:
   %call = tail call i64 @_Z13get_global_idj(i32 0)
diff --git a/llvm/test/CodeGen/AMDGPU/store-hi16.ll b/llvm/test/CodeGen/AMDGPU/store-hi16.ll
index 3791337446c..e8d6b24efd3 100644
--- a/llvm/test/CodeGen/AMDGPU/store-hi16.ll
+++ b/llvm/test/CodeGen/AMDGPU/store-hi16.ll
@@ -1,5 +1,5 @@
-; RUN: llc -march=amdgcn -mcpu=gfx900 -amdgpu-sroa=0 -mattr=-promote-alloca -verify-machineinstrs < %s | FileCheck -allow-deprecated-dag-overlap -check-prefixes=GCN,GFX900 %s
-; RUN: llc -march=amdgcn -mcpu=gfx906 -amdgpu-sroa=0 -mattr=-promote-alloca,+sram-ecc -verify-machineinstrs < %s | FileCheck -allow-deprecated-dag-overlap -check-prefixes=GCN,GFX906,NO-D16-HI %s
+; RUN: llc -march=amdgcn -mcpu=gfx900 -amdgpu-sroa=0 -mattr=-promote-alloca -verify-machineinstrs < %s | FileCheck -allow-deprecated-dag-overlap -check-prefixes=GCN,GFX900,GFX9 %s
+; RUN: llc -march=amdgcn -mcpu=gfx906 -amdgpu-sroa=0 -mattr=-promote-alloca,+sram-ecc -verify-machineinstrs < %s | FileCheck -allow-deprecated-dag-overlap -check-prefixes=GCN,GFX906,GFX9,NO-D16-HI %s
 ; RUN: llc -march=amdgcn -mcpu=fiji -amdgpu-sroa=0 -mattr=-promote-alloca -verify-machineinstrs < %s | FileCheck -allow-deprecated-dag-overlap -check-prefixes=GCN,GFX803,NO-D16-HI %s
 
 ; GCN-LABEL: {{^}}store_global_hi_v2i16:
@@ -311,16 +311,16 @@ entry:
 
 ; GCN-LABEL: {{^}}store_flat_hi_v2i16_neg_offset:
 ; GCN: s_waitcnt
-; GCN: v_add{{(_co)?}}_{{i|u}}32_e32
-
+; GFX803: v_add{{(_co)?}}_{{i|u}}32_e32
 ; GFX803: v_addc_u32_e32
-; GFX900: v_addc_co_u32_e32
 
-; GFX906-NEXT: v_lshrrev_b32_e32
-; GFX906-NEXT: v_addc_co_u32_e32
-; GFX906: flat_store_short v[0:1], v2
+; GFX9-DAG: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0xfffff000, v
+; GFX9-DAG: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, -1, v
 
-; GFX900-NEXT: flat_store_short_d16_hi v[0:1], v2{{$}}
+; GFX906-DAG: v_lshrrev_b32_e32
+; GFX906: flat_store_short v[0:1], v2 offset:2050{{$}}
+
+; GFX900-NEXT: flat_store_short_d16_hi v[0:1], v2 offset:2050{{$}}
 ; GFX803: flat_store_short v[0:1], v2{{$}}
 ; GCN-NEXT: s_waitcnt
 ; GCN-NEXT: s_setpc_b64
@@ -359,17 +359,17 @@ entry:
 
 ; GCN-LABEL: {{^}}store_flat_hi_v2i16_i8_neg_offset:
 ; GCN: s_waitcnt
-; GCN-DAG: v_add{{(_co)?}}_{{i|u}}32_e32
 
+; GFX803-DAG: v_add_u32_e32
 ; GFX803-DAG: v_addc_u32_e32
-; GFX900-DAG: v_addc_co_u32_e32
-; GFX906-DAG: v_add_co_u32_e32
 
-; GFX900-NEXT: flat_store_byte_d16_hi v[0:1], v2{{$}}
+; GFX9-DAG: v_add_co_u32_e32 v{{[0-9]+}}, vcc, 0xfffff000, v
+; GFX9-DAG: v_addc_co_u32_e32 v{{[0-9]+}}, vcc, -1, v{{[0-9]+}}, vcc
 
-; GFX906-NEXT: v_lshrrev_b32_e32 v2, 16, v2
-; GFX906-NEXT: v_addc_co_u32_e32
-; GFX906-NEXT: flat_store_byte v[0:1], v2{{$}}
+; GFX900-NEXT: flat_store_byte_d16_hi v[0:1], v2 offset:1{{$}}
+
+; GFX906-DAG: v_lshrrev_b32_e32 v2, 16, v2
+; GFX906: flat_store_byte v[0:1], v2 offset:1{{$}}
 
 ; GFX803-DAG: v_lshrrev_b32_e32 v2, 16, v2
 ; GFX803: flat_store_byte v[0:1], v2{{$}}