AMDGPU: Add sdst operand to VOP2b instructions

The VOP3 encoding of these allows any SGPR pair for the i1
output, but this was forced before to always use vcc.
This doesn't yet try to use this, but does add the operand
to the definitions so the main change is adding vcc to the
output of the VOP2 encoding.

llvm-svn: 246358

14 files changed, 61 insertions, 61 deletions

diff --git a/llvm/test/CodeGen/AMDGPU/add.ll b/llvm/test/CodeGen/AMDGPU/add.ll
index 655e75dbc1a..2ddfa9649ac 100644
--- a/llvm/test/CodeGen/AMDGPU/add.ll
+++ b/llvm/test/CodeGen/AMDGPU/add.ll
@@ -5,7 +5,7 @@
 ;FUNC-LABEL: {{^}}test1:
 ;EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-;SI: v_add_i32_e32 [[REG:v[0-9]+]], {{v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 [[REG:v[0-9]+]], vcc, {{v[0-9]+, v[0-9]+}}
 ;SI-NOT: [[REG]]
 ;SI: buffer_store_dword [[REG]],
 define void @test1(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
@@ -21,8 +21,8 @@ define void @test1(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
 ;EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ;EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-;SI: v_add_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-;SI: v_add_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
 
 define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
   %b_ptr = getelementptr <2 x i32>, <2 x i32> addrspace(1)* %in, i32 1
@@ -39,10 +39,10 @@ define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
 ;EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ;EG: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-;SI: v_add_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-;SI: v_add_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-;SI: v_add_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-;SI: v_add_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+;SI: v_add_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
 
 define void @test4(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
   %b_ptr = getelementptr <4 x i32>, <4 x i32> addrspace(1)* %in, i32 1
diff --git a/llvm/test/CodeGen/AMDGPU/array-ptr-calc-i32.ll b/llvm/test/CodeGen/AMDGPU/array-ptr-calc-i32.ll
index 8c2a0795860..67738f7daff 100644
--- a/llvm/test/CodeGen/AMDGPU/array-ptr-calc-i32.ll
+++ b/llvm/test/CodeGen/AMDGPU/array-ptr-calc-i32.ll
@@ -14,7 +14,7 @@ declare void @llvm.AMDGPU.barrier.local() nounwind noduplicate
 ; FIXME: We end up with zero argument for ADD, because
 ; SIRegisterInfo::eliminateFrameIndex() blindly replaces the frame index
 ; with the appropriate offset.  We should fold this into the store.
-; SI-ALLOCA: v_add_i32_e32 [[PTRREG:v[0-9]+]], 0, v{{[0-9]+}}
+; SI-ALLOCA: v_add_i32_e32 [[PTRREG:v[0-9]+]], vcc, 0, v{{[0-9]+}}
 ; SI-ALLOCA: buffer_store_dword {{v[0-9]+}}, [[PTRREG]], s[{{[0-9]+:[0-9]+}}]
 ;
 ; FIXME: The AMDGPUPromoteAlloca pass should be able to convert this
@@ -22,7 +22,7 @@ declare void @llvm.AMDGPU.barrier.local() nounwind noduplicate
 ; to interpret:
 ; getelementptr [4 x i32], [4 x i32]* %alloca, i32 1, i32 %b
 
-; SI-PROMOTE: v_add_i32_e32 [[PTRREG:v[0-9]+]], 16
+; SI-PROMOTE: v_add_i32_e32 [[PTRREG:v[0-9]+]], vcc, 16
 ; SI-PROMOTE: ds_write_b32 [[PTRREG]]
 define void @test_private_array_ptr_calc(i32 addrspace(1)* noalias %out, i32 addrspace(1)* noalias %inA, i32 addrspace(1)* noalias %inB) {
   %alloca = alloca [4 x i32], i32 4, align 16
diff --git a/llvm/test/CodeGen/AMDGPU/cvt_f32_ubyte.ll b/llvm/test/CodeGen/AMDGPU/cvt_f32_ubyte.ll
index 3399d9da29e..cd554ba256b 100644
--- a/llvm/test/CodeGen/AMDGPU/cvt_f32_ubyte.ll
+++ b/llvm/test/CodeGen/AMDGPU/cvt_f32_ubyte.ll
@@ -154,7 +154,7 @@ define void @load_v8i8_to_v8f32(<8 x float> addrspace(1)* noalias %out, <8 x i8>
 
 ; SI-LABEL: {{^}}i8_zext_inreg_i32_to_f32:
 ; SI: buffer_load_dword [[LOADREG:v[0-9]+]],
-; SI: v_add_i32_e32 [[ADD:v[0-9]+]], 2, [[LOADREG]]
+; SI: v_add_i32_e32 [[ADD:v[0-9]+]], vcc, 2, [[LOADREG]]
 ; SI-NEXT: v_cvt_f32_ubyte0_e32 [[CONV:v[0-9]+]], [[ADD]]
 ; SI: buffer_store_dword [[CONV]],
 define void @i8_zext_inreg_i32_to_f32(float addrspace(1)* noalias %out, i32 addrspace(1)* noalias %in) nounwind {
diff --git a/llvm/test/CodeGen/AMDGPU/ds-negative-offset-addressing-mode-loop.ll b/llvm/test/CodeGen/AMDGPU/ds-negative-offset-addressing-mode-loop.ll
index 5e4654abd91..ac17df11970 100644
--- a/llvm/test/CodeGen/AMDGPU/ds-negative-offset-addressing-mode-loop.ll
+++ b/llvm/test/CodeGen/AMDGPU/ds-negative-offset-addressing-mode-loop.ll
@@ -10,13 +10,13 @@ declare void @llvm.AMDGPU.barrier.local() #1
 ; CHECK: BB0_1:
 ; CHECK: v_add_i32_e32 [[VADDR:v[0-9]+]],
 ; SI-DAG: ds_read_b32 v{{[0-9]+}}, [[VADDR]]
-; SI-DAG: v_add_i32_e32 [[VADDR4:v[0-9]+]], 4, [[VADDR]]
+; SI-DAG: v_add_i32_e32 [[VADDR4:v[0-9]+]], vcc, 4, [[VADDR]]
 ; SI-DAG: ds_read_b32 v{{[0-9]+}}, [[VADDR4]]
-; SI-DAG: v_add_i32_e32 [[VADDR0x80:v[0-9]+]], 0x80, [[VADDR]]
+; SI-DAG: v_add_i32_e32 [[VADDR0x80:v[0-9]+]], vcc, 0x80, [[VADDR]]
 ; SI-DAG: ds_read_b32 v{{[0-9]+}}, [[VADDR0x80]]
-; SI-DAG: v_add_i32_e32 [[VADDR0x84:v[0-9]+]], 0x84, [[VADDR]]
+; SI-DAG: v_add_i32_e32 [[VADDR0x84:v[0-9]+]], vcc, 0x84, [[VADDR]]
 ; SI-DAG: ds_read_b32 v{{[0-9]+}}, [[VADDR0x84]]
-; SI-DAG: v_add_i32_e32 [[VADDR0x100:v[0-9]+]], 0x100, [[VADDR]]
+; SI-DAG: v_add_i32_e32 [[VADDR0x100:v[0-9]+]], vcc, 0x100, [[VADDR]]
 ; SI-DAG: ds_read_b32 v{{[0-9]+}}, [[VADDR0x100]]
 
 ; CI-DAG: ds_read2_b32 v{{\[[0-9]+:[0-9]+\]}}, [[VADDR]] offset1:1
diff --git a/llvm/test/CodeGen/AMDGPU/ds_read2st64.ll b/llvm/test/CodeGen/AMDGPU/ds_read2st64.ll
index e2e441214b4..00de55be852 100644
--- a/llvm/test/CodeGen/AMDGPU/ds_read2st64.ll
+++ b/llvm/test/CodeGen/AMDGPU/ds_read2st64.ll
@@ -65,7 +65,7 @@ define void @simple_read2st64_f32_max_offset(float addrspace(1)* %out, float add
 
 ; SI-LABEL: @simple_read2st64_f32_over_max_offset
 ; SI-NOT: ds_read2st64_b32
-; SI: v_add_i32_e32 [[BIGADD:v[0-9]+]], 0x10000, {{v[0-9]+}}
+; SI: v_add_i32_e32 [[BIGADD:v[0-9]+]], vcc, 0x10000, {{v[0-9]+}}
 ; SI: ds_read_b32 {{v[0-9]+}}, {{v[0-9]+}} offset:256
 ; SI: ds_read_b32 {{v[0-9]+}}, [[BIGADD]]
 ; SI: s_endpgm
@@ -197,7 +197,7 @@ define void @simple_read2st64_f64_max_offset(double addrspace(1)* %out, double a
 
 ; SI-LABEL: @simple_read2st64_f64_over_max_offset
 ; SI-NOT: ds_read2st64_b64
-; SI: v_add_i32_e32 [[BIGADD:v[0-9]+]], 0x10000, {{v[0-9]+}}
+; SI: v_add_i32_e32 [[BIGADD:v[0-9]+]], vcc, 0x10000, {{v[0-9]+}}
 ; SI: ds_read_b64 {{v\[[0-9]+:[0-9]+\]}}, {{v[0-9]+}} offset:512
 ; SI: ds_read_b64 {{v\[[0-9]+:[0-9]+\]}}, [[BIGADD]]
 ; SI: s_endpgm
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.bfe.i32.ll b/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.bfe.i32.ll
index 1168713ca66..d56b4845728 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.bfe.i32.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.AMDGPU.bfe.i32.ll
@@ -425,7 +425,7 @@ define void @bfe_sext_in_reg_i24(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
 ; SI: buffer_load_dword [[LOAD:v[0-9]+]]
 ; SI: v_bfe_i32 [[BFE:v[0-9]+]], [[LOAD]], 1, 16
 ; SI: v_lshrrev_b32_e32 [[TMP0:v[0-9]+]], 31, [[BFE]]
-; SI: v_add_i32_e32 [[TMP1:v[0-9]+]], [[TMP0]], [[BFE]]
+; SI: v_add_i32_e32 [[TMP1:v[0-9]+]], vcc, [[TMP0]], [[BFE]]
 ; SI: v_ashrrev_i32_e32 [[TMP2:v[0-9]+]], 1, [[TMP1]]
 ; SI: buffer_store_dword [[TMP2]]
 define void @simplify_demanded_bfe_sdiv(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounwind {
diff --git a/llvm/test/CodeGen/AMDGPU/local-memory-two-objects.ll b/llvm/test/CodeGen/AMDGPU/local-memory-two-objects.ll
index f501a7ac627..7f31ef45b62 100644
--- a/llvm/test/CodeGen/AMDGPU/local-memory-two-objects.ll
+++ b/llvm/test/CodeGen/AMDGPU/local-memory-two-objects.ll
@@ -30,7 +30,7 @@
 ; constant offsets.
 ; EG: LDS_READ_RET {{[*]*}} OQAP, {{PV|T}}[[ADDRR:[0-9]*\.[XYZW]]]
 ; EG-NOT: LDS_READ_RET {{[*]*}} OQAP, T[[ADDRR]]
-; SI: v_add_i32_e32 [[SIPTR:v[0-9]+]], 16, v{{[0-9]+}}
+; SI: v_add_i32_e32 [[SIPTR:v[0-9]+]], vcc, 16, v{{[0-9]+}}
 ; SI: ds_read_b32 {{v[0-9]+}}, [[SIPTR]]
 ; CI: ds_read_b32 {{v[0-9]+}}, [[ADDRR:v[0-9]+]] offset:16
 ; CI: ds_read_b32 {{v[0-9]+}}, [[ADDRR]]
diff --git a/llvm/test/CodeGen/AMDGPU/operand-folding.ll b/llvm/test/CodeGen/AMDGPU/operand-folding.ll
index 816755efb07..9e514ef9970 100644
--- a/llvm/test/CodeGen/AMDGPU/operand-folding.ll
+++ b/llvm/test/CodeGen/AMDGPU/operand-folding.ll
@@ -1,7 +1,7 @@
 ; RUN: llc < %s -march=amdgcn -mcpu=SI -verify-machineinstrs | FileCheck %s
 
 ; CHECK-LABEL: {{^}}fold_sgpr:
-; CHECK: v_add_i32_e32 v{{[0-9]+}}, s
+; CHECK: v_add_i32_e32 v{{[0-9]+}}, vcc, s
 define void @fold_sgpr(i32 addrspace(1)* %out, i32 %fold) {
 entry:
   %tmp0 = icmp ne i32 %fold, 0
diff --git a/llvm/test/CodeGen/AMDGPU/scratch-buffer.ll b/llvm/test/CodeGen/AMDGPU/scratch-buffer.ll
index 268869daaa3..d43de476605 100644
--- a/llvm/test/CodeGen/AMDGPU/scratch-buffer.ll
+++ b/llvm/test/CodeGen/AMDGPU/scratch-buffer.ll
@@ -51,7 +51,7 @@ done:
 
 ; GCN-LABEL: {{^}}legal_offset_fi_offset
 ; GCN: buffer_store_dword v{{[0-9]+}}, v{{[0-9]+}}, s[{{[0-9]+}}:{{[0-9]+}}], s{{[0-9]+}} offen
-; GCN: v_add_i32_e32 [[OFFSET:v[0-9]+]], 0x8000
+; GCN: v_add_i32_e32 [[OFFSET:v[0-9]+]], vcc, 0x8000
 ; GCN: buffer_store_dword v{{[0-9]+}}, [[OFFSET]], s[{{[0-9]+}}:{{[0-9]+}}], s{{[0-9]+}} offen{{$}}
 
 define void @legal_offset_fi_offset(i32 addrspace(1)* %out, i32 %cond, i32 addrspace(1)* %offsets, i32 %if_offset, i32 %else_offset) {
diff --git a/llvm/test/CodeGen/AMDGPU/shl_add_constant.ll b/llvm/test/CodeGen/AMDGPU/shl_add_constant.ll
index b1485bfaaeb..dfb2bf3383f 100644
--- a/llvm/test/CodeGen/AMDGPU/shl_add_constant.ll
+++ b/llvm/test/CodeGen/AMDGPU/shl_add_constant.ll
@@ -6,7 +6,7 @@ declare i32 @llvm.r600.read.tidig.x() #1
 
 ; FUNC-LABEL: {{^}}shl_2_add_9_i32:
 ; SI: v_lshlrev_b32_e32  [[REG:v[0-9]+]], 2, {{v[0-9]+}}
-; SI: v_add_i32_e32 [[RESULT:v[0-9]+]], 36, [[REG]]
+; SI: v_add_i32_e32 [[RESULT:v[0-9]+]], vcc, 36, [[REG]]
 ; SI: buffer_store_dword [[RESULT]]
 ; SI: s_endpgm
 define void @shl_2_add_9_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
@@ -20,7 +20,7 @@ define void @shl_2_add_9_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
 }
 
 ; FUNC-LABEL: {{^}}shl_2_add_9_i32_2_add_uses:
-; SI-DAG: v_add_i32_e32 [[ADDREG:v[0-9]+]], 9, {{v[0-9]+}}
+; SI-DAG: v_add_i32_e32 [[ADDREG:v[0-9]+]], vcc, 9, {{v[0-9]+}}
 ; SI-DAG: v_lshlrev_b32_e32 [[SHLREG:v[0-9]+]], 2, {{v[0-9]+}}
 ; SI-DAG: buffer_store_dword [[ADDREG]]
 ; SI-DAG: buffer_store_dword [[SHLREG]]
@@ -40,7 +40,7 @@ define void @shl_2_add_9_i32_2_add_uses(i32 addrspace(1)* %out0, i32 addrspace(1
 
 ; FUNC-LABEL: {{^}}shl_2_add_999_i32:
 ; SI: v_lshlrev_b32_e32  [[REG:v[0-9]+]], 2, {{v[0-9]+}}
-; SI: v_add_i32_e32 [[RESULT:v[0-9]+]], 0xf9c, [[REG]]
+; SI: v_add_i32_e32 [[RESULT:v[0-9]+]], vcc, 0xf9c, [[REG]]
 ; SI: buffer_store_dword [[RESULT]]
 ; SI: s_endpgm
 define void @shl_2_add_999_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
diff --git a/llvm/test/CodeGen/AMDGPU/shl_add_ptr.ll b/llvm/test/CodeGen/AMDGPU/shl_add_ptr.ll
index 6671e909cd1..ac94824bd61 100644
--- a/llvm/test/CodeGen/AMDGPU/shl_add_ptr.ll
+++ b/llvm/test/CodeGen/AMDGPU/shl_add_ptr.ll
@@ -35,7 +35,7 @@ define void @load_shl_base_lds_0(float addrspace(1)* %out, i32 addrspace(1)* %ad
 ; SI-LABEL: {{^}}load_shl_base_lds_1:
 ; SI: v_lshlrev_b32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
 ; SI: ds_read_b32 [[RESULT:v[0-9]+]], [[PTR]] offset:8
-; SI: v_add_i32_e32 [[ADDUSE:v[0-9]+]], 8, v{{[0-9]+}}
+; SI: v_add_i32_e32 [[ADDUSE:v[0-9]+]], vcc, 8, v{{[0-9]+}}
 ; SI-DAG: buffer_store_dword [[RESULT]]
 ; SI-DAG: buffer_store_dword [[ADDUSE]]
 ; SI: s_endpgm
diff --git a/llvm/test/CodeGen/AMDGPU/sub.ll b/llvm/test/CodeGen/AMDGPU/sub.ll
index b7fba0efa5b..9f9446a4e60 100644
--- a/llvm/test/CodeGen/AMDGPU/sub.ll
+++ b/llvm/test/CodeGen/AMDGPU/sub.ll
@@ -7,7 +7,7 @@ declare i32 @llvm.r600.read.tidig.x() readnone
 ; FUNC-LABEL: {{^}}test_sub_i32:
 ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-; SI: v_subrev_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
+; SI: v_subrev_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
 define void @test_sub_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
   %b_ptr = getelementptr i32, i32 addrspace(1)* %in, i32 1
   %a = load i32, i32 addrspace(1)* %in
@@ -22,8 +22,8 @@ define void @test_sub_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
 ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
+; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
 
 define void @test_sub_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
   %b_ptr = getelementptr <2 x i32>, <2 x i32> addrspace(1)* %in, i32 1
@@ -40,10 +40,10 @@ define void @test_sub_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)
 ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 ; EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
 
-; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
-; SI: v_sub_i32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
+; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
+; SI: v_sub_i32_e32 v{{[0-9]+, vcc, v[0-9]+, v[0-9]+}}
 
 define void @test_sub_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
   %b_ptr = getelementptr <4 x i32>, <4 x i32> addrspace(1)* %in, i32 1
diff --git a/llvm/test/CodeGen/AMDGPU/udivrem.ll b/llvm/test/CodeGen/AMDGPU/udivrem.ll
index b3837f28209..dc3e2e30297 100644
--- a/llvm/test/CodeGen/AMDGPU/udivrem.ll
+++ b/llvm/test/CodeGen/AMDGPU/udivrem.ll
@@ -30,19 +30,19 @@
 ; SI: v_rcp_iflag_f32_e32 [[RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[RCP_HI:v[0-9]+]], [[RCP]]
 ; SI-DAG: v_mul_lo_i32 [[RCP_LO:v[0-9]+]], [[RCP]]
-; SI-DAG: v_sub_i32_e32 [[NEG_RCP_LO:v[0-9]+]], 0, [[RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[NEG_RCP_LO:v[0-9]+]], vcc, 0, [[RCP_LO]]
 ; SI: v_cndmask_b32_e64
 ; SI: v_mul_hi_u32 [[E:v[0-9]+]], {{v[0-9]+}}, [[RCP]]
-; SI-DAG: v_add_i32_e32 [[RCP_A_E:v[0-9]+]], [[E]], [[RCP]]
-; SI-DAG: v_subrev_i32_e32 [[RCP_S_E:v[0-9]+]], [[E]], [[RCP]]
+; SI-DAG: v_add_i32_e32 [[RCP_A_E:v[0-9]+]], vcc, [[E]], [[RCP]]
+; SI-DAG: v_subrev_i32_e32 [[RCP_S_E:v[0-9]+]], vcc, [[E]], [[RCP]]
 ; SI: v_cndmask_b32_e64
 ; SI: v_mul_hi_u32 [[Quotient:v[0-9]+]]
 ; SI: v_mul_lo_i32 [[Num_S_Remainder:v[0-9]+]]
-; SI-DAG: v_sub_i32_e32 [[Remainder:v[0-9]+]], {{[vs][0-9]+}}, [[Num_S_Remainder]]
+; SI-DAG: v_sub_i32_e32 [[Remainder:v[0-9]+]], vcc, {{[vs][0-9]+}}, [[Num_S_Remainder]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
 ; SI: v_and_b32_e32 [[Tmp1:v[0-9]+]]
-; SI-DAG: v_add_i32_e32 [[Quotient_A_One:v[0-9]+]], 1, [[Quotient]]
+; SI-DAG: v_add_i32_e32 [[Quotient_A_One:v[0-9]+]], vcc, 1, [[Quotient]]
 ; SI-DAG: v_subrev_i32_e32 [[Quotient_S_One:v[0-9]+]],
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
@@ -110,15 +110,15 @@ define void @test_udivrem(i32 addrspace(1)* %out, i32 %x, i32 %y) {
 ; SI-DAG: v_rcp_iflag_f32_e32 [[FIRST_RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[FIRST_RCP_HI:v[0-9]+]], [[FIRST_RCP]]
 ; SI-DAG: v_mul_lo_i32 [[FIRST_RCP_LO:v[0-9]+]], [[FIRST_RCP]]
-; SI-DAG: v_sub_i32_e32 [[FIRST_NEG_RCP_LO:v[0-9]+]], 0, [[FIRST_RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[FIRST_NEG_RCP_LO:v[0-9]+]], vcc, 0, [[FIRST_RCP_LO]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[FIRST_E:v[0-9]+]], {{v[0-9]+}}, [[FIRST_RCP]]
-; SI-DAG: v_add_i32_e32 [[FIRST_RCP_A_E:v[0-9]+]], [[FIRST_E]], [[FIRST_RCP]]
-; SI-DAG: v_subrev_i32_e32 [[FIRST_RCP_S_E:v[0-9]+]], [[FIRST_E]], [[FIRST_RCP]]
+; SI-DAG: v_add_i32_e32 [[FIRST_RCP_A_E:v[0-9]+]], vcc, [[FIRST_E]], [[FIRST_RCP]]
+; SI-DAG: v_subrev_i32_e32 [[FIRST_RCP_S_E:v[0-9]+]], vcc, [[FIRST_E]], [[FIRST_RCP]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[FIRST_Quotient:v[0-9]+]]
 ; SI-DAG: v_mul_lo_i32 [[FIRST_Num_S_Remainder:v[0-9]+]]
-; SI-DAG: v_subrev_i32_e32 [[FIRST_Remainder:v[0-9]+]], [[FIRST_Num_S_Remainder]], v{{[0-9]+}}
+; SI-DAG: v_subrev_i32_e32 [[FIRST_Remainder:v[0-9]+]], vcc, [[FIRST_Num_S_Remainder]], v{{[0-9]+}}
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_and_b32_e32 [[FIRST_Tmp1:v[0-9]+]]
@@ -133,15 +133,15 @@ define void @test_udivrem(i32 addrspace(1)* %out, i32 %x, i32 %y) {
 ; SI-DAG: v_rcp_iflag_f32_e32 [[SECOND_RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[SECOND_RCP_HI:v[0-9]+]], [[SECOND_RCP]]
 ; SI-DAG: v_mul_lo_i32 [[SECOND_RCP_LO:v[0-9]+]], [[SECOND_RCP]]
-; SI-DAG: v_sub_i32_e32 [[SECOND_NEG_RCP_LO:v[0-9]+]], 0, [[SECOND_RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[SECOND_NEG_RCP_LO:v[0-9]+]], vcc, 0, [[SECOND_RCP_LO]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[SECOND_E:v[0-9]+]], {{v[0-9]+}}, [[SECOND_RCP]]
-; SI-DAG: v_add_i32_e32 [[SECOND_RCP_A_E:v[0-9]+]], [[SECOND_E]], [[SECOND_RCP]]
-; SI-DAG: v_subrev_i32_e32 [[SECOND_RCP_S_E:v[0-9]+]], [[SECOND_E]], [[SECOND_RCP]]
+; SI-DAG: v_add_i32_e32 [[SECOND_RCP_A_E:v[0-9]+]], vcc, [[SECOND_E]], [[SECOND_RCP]]
+; SI-DAG: v_subrev_i32_e32 [[SECOND_RCP_S_E:v[0-9]+]], vcc, [[SECOND_E]], [[SECOND_RCP]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[SECOND_Quotient:v[0-9]+]]
 ; SI-DAG: v_mul_lo_i32 [[SECOND_Num_S_Remainder:v[0-9]+]]
-; SI-DAG: v_subrev_i32_e32 [[SECOND_Remainder:v[0-9]+]], [[SECOND_Num_S_Remainder]], v{{[0-9]+}}
+; SI-DAG: v_subrev_i32_e32 [[SECOND_Remainder:v[0-9]+]], vcc, [[SECOND_Num_S_Remainder]], v{{[0-9]+}}
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_and_b32_e32 [[SECOND_Tmp1:v[0-9]+]]
@@ -260,15 +260,15 @@ define void @test_udivrem_v2(<2 x i32> addrspace(1)* %out, <2 x i32> %x, <2 x i3
 ; SI-DAG: v_rcp_iflag_f32_e32 [[FIRST_RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[FIRST_RCP_HI:v[0-9]+]], [[FIRST_RCP]]
 ; SI-DAG: v_mul_lo_i32 [[FIRST_RCP_LO:v[0-9]+]], [[FIRST_RCP]]
-; SI-DAG: v_sub_i32_e32 [[FIRST_NEG_RCP_LO:v[0-9]+]], 0, [[FIRST_RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[FIRST_NEG_RCP_LO:v[0-9]+]], vcc, 0, [[FIRST_RCP_LO]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[FIRST_E:v[0-9]+]], {{v[0-9]+}}, [[FIRST_RCP]]
-; SI-DAG: v_add_i32_e32 [[FIRST_RCP_A_E:v[0-9]+]], [[FIRST_E]], [[FIRST_RCP]]
-; SI-DAG: v_subrev_i32_e32 [[FIRST_RCP_S_E:v[0-9]+]], [[FIRST_E]], [[FIRST_RCP]]
+; SI-DAG: v_add_i32_e32 [[FIRST_RCP_A_E:v[0-9]+]], vcc, [[FIRST_E]], [[FIRST_RCP]]
+; SI-DAG: v_subrev_i32_e32 [[FIRST_RCP_S_E:v[0-9]+]], vcc, [[FIRST_E]], [[FIRST_RCP]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[FIRST_Quotient:v[0-9]+]]
 ; SI-DAG: v_mul_lo_i32 [[FIRST_Num_S_Remainder:v[0-9]+]]
-; SI-DAG: v_subrev_i32_e32 [[FIRST_Remainder:v[l0-9]+]], [[FIRST_Num_S_Remainder]], v{{[0-9]+}}
+; SI-DAG: v_subrev_i32_e32 [[FIRST_Remainder:v[l0-9]+]], vcc, [[FIRST_Num_S_Remainder]], v{{[0-9]+}}
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_and_b32_e32 [[FIRST_Tmp1:v[0-9]+]]
@@ -283,15 +283,15 @@ define void @test_udivrem_v2(<2 x i32> addrspace(1)* %out, <2 x i32> %x, <2 x i3
 ; SI-DAG: v_rcp_iflag_f32_e32 [[SECOND_RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[SECOND_RCP_HI:v[0-9]+]], [[SECOND_RCP]]
 ; SI-DAG: v_mul_lo_i32 [[SECOND_RCP_LO:v[0-9]+]], [[SECOND_RCP]]
-; SI-DAG: v_sub_i32_e32 [[SECOND_NEG_RCP_LO:v[0-9]+]], 0, [[SECOND_RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[SECOND_NEG_RCP_LO:v[0-9]+]], vcc, 0, [[SECOND_RCP_LO]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[SECOND_E:v[0-9]+]], {{v[0-9]+}}, [[SECOND_RCP]]
-; SI-DAG: v_add_i32_e32 [[SECOND_RCP_A_E:v[0-9]+]], [[SECOND_E]], [[SECOND_RCP]]
-; SI-DAG: v_subrev_i32_e32 [[SECOND_RCP_S_E:v[0-9]+]], [[SECOND_E]], [[SECOND_RCP]]
+; SI-DAG: v_add_i32_e32 [[SECOND_RCP_A_E:v[0-9]+]], vcc, [[SECOND_E]], [[SECOND_RCP]]
+; SI-DAG: v_subrev_i32_e32 [[SECOND_RCP_S_E:v[0-9]+]], vcc, [[SECOND_E]], [[SECOND_RCP]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[SECOND_Quotient:v[0-9]+]]
 ; SI-DAG: v_mul_lo_i32 [[SECOND_Num_S_Remainder:v[0-9]+]]
-; SI-DAG: v_subrev_i32_e32 [[SECOND_Remainder:v[0-9]+]], [[SECOND_Num_S_Remainder]], v{{[0-9]+}}
+; SI-DAG: v_subrev_i32_e32 [[SECOND_Remainder:v[0-9]+]], vcc, [[SECOND_Num_S_Remainder]], v{{[0-9]+}}
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_and_b32_e32 [[SECOND_Tmp1:v[0-9]+]]
@@ -306,15 +306,15 @@ define void @test_udivrem_v2(<2 x i32> addrspace(1)* %out, <2 x i32> %x, <2 x i3
 ; SI-DAG: v_rcp_iflag_f32_e32 [[THIRD_RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[THIRD_RCP_HI:v[0-9]+]], [[THIRD_RCP]]
 ; SI-DAG: v_mul_lo_i32 [[THIRD_RCP_LO:v[0-9]+]], [[THIRD_RCP]]
-; SI-DAG: v_sub_i32_e32 [[THIRD_NEG_RCP_LO:v[0-9]+]], 0, [[THIRD_RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[THIRD_NEG_RCP_LO:v[0-9]+]], vcc, 0, [[THIRD_RCP_LO]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[THIRD_E:v[0-9]+]], {{v[0-9]+}}, [[THIRD_RCP]]
-; SI-DAG: v_add_i32_e32 [[THIRD_RCP_A_E:v[0-9]+]], [[THIRD_E]], [[THIRD_RCP]]
-; SI-DAG: v_subrev_i32_e32 [[THIRD_RCP_S_E:v[0-9]+]], [[THIRD_E]], [[THIRD_RCP]]
+; SI-DAG: v_add_i32_e32 [[THIRD_RCP_A_E:v[0-9]+]], vcc, [[THIRD_E]], [[THIRD_RCP]]
+; SI-DAG: v_subrev_i32_e32 [[THIRD_RCP_S_E:v[0-9]+]], vcc, [[THIRD_E]], [[THIRD_RCP]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[THIRD_Quotient:v[0-9]+]]
 ; SI-DAG: v_mul_lo_i32 [[THIRD_Num_S_Remainder:v[0-9]+]]
-; SI-DAG: v_subrev_i32_e32 [[THIRD_Remainder:v[0-9]+]], [[THIRD_Num_S_Remainder]], {{v[0-9]+}}
+; SI-DAG: v_subrev_i32_e32 [[THIRD_Remainder:v[0-9]+]], vcc, [[THIRD_Num_S_Remainder]], {{v[0-9]+}}
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_and_b32_e32 [[THIRD_Tmp1:v[0-9]+]]
@@ -329,11 +329,11 @@ define void @test_udivrem_v2(<2 x i32> addrspace(1)* %out, <2 x i32> %x, <2 x i3
 ; SI-DAG: v_rcp_iflag_f32_e32 [[FOURTH_RCP:v[0-9]+]]
 ; SI-DAG: v_mul_hi_u32 [[FOURTH_RCP_HI:v[0-9]+]], [[FOURTH_RCP]]
 ; SI-DAG: v_mul_lo_i32 [[FOURTH_RCP_LO:v[0-9]+]], [[FOURTH_RCP]]
-; SI-DAG: v_sub_i32_e32 [[FOURTH_NEG_RCP_LO:v[0-9]+]], 0, [[FOURTH_RCP_LO]]
+; SI-DAG: v_sub_i32_e32 [[FOURTH_NEG_RCP_LO:v[0-9]+]], vcc, 0, [[FOURTH_RCP_LO]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI-DAG: v_mul_hi_u32 [[FOURTH_E:v[0-9]+]], {{v[0-9]+}}, [[FOURTH_RCP]]
-; SI-DAG: v_add_i32_e32 [[FOURTH_RCP_A_E:v[0-9]+]], [[FOURTH_E]], [[FOURTH_RCP]]
-; SI-DAG: v_subrev_i32_e32 [[FOURTH_RCP_S_E:v[0-9]+]], [[FOURTH_E]], [[FOURTH_RCP]]
+; SI-DAG: v_add_i32_e32 [[FOURTH_RCP_A_E:v[0-9]+]], vcc, [[FOURTH_E]], [[FOURTH_RCP]]
+; SI-DAG: v_subrev_i32_e32 [[FOURTH_RCP_S_E:v[0-9]+]], vcc, [[FOURTH_E]], [[FOURTH_RCP]]
 ; SI-DAG: v_cndmask_b32_e64
 ; SI: s_endpgm
 define void @test_udivrem_v4(<4 x i32> addrspace(1)* %out, <4 x i32> %x, <4 x i32> %y) {
diff --git a/llvm/test/CodeGen/AMDGPU/vop-shrink.ll b/llvm/test/CodeGen/AMDGPU/vop-shrink.ll
index 9b2f229c05a..2bfe1b2bd6e 100644
--- a/llvm/test/CodeGen/AMDGPU/vop-shrink.ll
+++ b/llvm/test/CodeGen/AMDGPU/vop-shrink.ll
@@ -3,8 +3,8 @@
 
 ; Test that we correctly commute a sub instruction
 ; FUNC-LABEL: {{^}}sub_rev:
-; SI-NOT: v_sub_i32_e32 v{{[0-9]+}}, s
-; SI: v_subrev_i32_e32 v{{[0-9]+}}, s
+; SI-NOT: v_sub_i32_e32 v{{[0-9]+}}, vcc, s
+; SI: v_subrev_i32_e32 v{{[0-9]+}}, vcc, s
 
 ; ModuleID = 'vop-shrink.ll'