diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.softwqm.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.softwqm.ll
index 787a15b6dc7275..efdae50e1f127e 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.softwqm.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.softwqm.ll
@@ -1,13 +1,19 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=CHECK %s
 
 ; Check that WQM is not triggered by the softwqm intrinsic alone.
 ;
-;CHECK-LABEL: {{^}}test1:
-;CHECK-NOT: s_wqm_b64 exec, exec
-;CHECK: buffer_load_dword
-;CHECK: buffer_load_dword
-;CHECK: v_add_f32_e32
 define amdgpu_ps float @test1(i32 inreg %idx0, i32 inreg %idx1) {
+; CHECK-LABEL: test1:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    v_mov_b32_e32 v1, s1
+; CHECK-NEXT:    buffer_load_dword v0, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    buffer_load_dword v1, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v0, v0, v1
+; CHECK-NEXT:    ; kill: def $vgpr0 killed $vgpr0 killed $exec killed $exec
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %src0 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
   %src1 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx1, i32 0, i32 0, i32 0)
@@ -18,12 +24,17 @@ main_body:
 
 ; Check that the softwqm intrinsic works correctly for integers.
 ;
-;CHECK-LABEL: {{^}}test2:
-;CHECK-NOT: s_wqm_b64 exec, exec
-;CHECK: buffer_load_dword
-;CHECK: buffer_load_dword
-;CHECK: v_add_f32_e32
 define amdgpu_ps float @test2(i32 inreg %idx0, i32 inreg %idx1) {
+; CHECK-LABEL: test2:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    v_mov_b32_e32 v1, s1
+; CHECK-NEXT:    buffer_load_dword v0, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    buffer_load_dword v1, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v0, v0, v1
+; CHECK-NEXT:    ; kill: def $vgpr0 killed $vgpr0 killed $exec killed $exec
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %src0 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
   %src1 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx1, i32 0, i32 0, i32 0)
@@ -36,14 +47,20 @@ main_body:
 
 ; Make sure the transition from WQM to Exact to softwqm does not trigger WQM.
 ;
-;CHECK-LABEL: {{^}}test_softwqm1:
-;CHECK-NOT: s_wqm_b64 exec, exec
-;CHECK: buffer_load_dword
-;CHECK: buffer_load_dword
-;CHECK: buffer_store_dword
-;CHECK-NOT; s_wqm_b64 exec, exec
-;CHECK: v_add_f32_e32
 define amdgpu_ps float @test_softwqm1(i32 inreg %idx0, i32 inreg %idx1) {
+; CHECK-LABEL: test_softwqm1:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    v_mov_b32_e32 v2, s1
+; CHECK-NEXT:    buffer_load_dword v1, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    buffer_load_dword v2, v2, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v1, v1, v2
+; CHECK-NEXT:    buffer_store_dword v1, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    v_add_f32_e32 v0, v1, v1
+; CHECK-NEXT:    ; kill: def $vgpr0 killed $vgpr0 killed $exec killed $exec
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %src0 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
   %src1 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx1, i32 0, i32 0, i32 0)
@@ -56,16 +73,26 @@ main_body:
 
 ; Make sure the transition from WQM to Exact to softwqm does trigger WQM.
 ;
-;CHECK-LABEL: {{^}}test_softwqm2:
-;CHECK: s_mov_b64 [[ORIG:s\[[0-9]+:[0-9]+\]]], exec
-;CHECK: s_wqm_b64 exec, exec
-;CHECK: buffer_load_dword
-;CHECK: buffer_load_dword
-;CHECK: v_add_f32_e32
-;CHECK: v_add_f32_e32
-;CHECK: s_and_b64 exec, exec, [[ORIG]]
-;CHECK: buffer_store_dword
 define amdgpu_ps float @test_softwqm2(i32 inreg %idx0, i32 inreg %idx1) {
+; CHECK-LABEL: test_softwqm2:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    s_mov_b64 s[2:3], exec
+; CHECK-NEXT:    s_wqm_b64 exec, exec
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    v_mov_b32_e32 v2, s1
+; CHECK-NEXT:    buffer_load_dword v1, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    buffer_load_dword v2, v2, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v1, v1, v2
+; CHECK-NEXT:    v_mov_b32_e32 v2, v1
+; CHECK-NEXT:    v_add_f32_e32 v1, v1, v1
+; CHECK-NEXT:    s_and_b64 exec, exec, s[2:3]
+; CHECK-NEXT:    buffer_store_dword v2, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    s_wqm_b64 exec, exec
+; CHECK-NEXT:    v_mov_b32_e32 v0, v1
+; CHECK-NEXT:    s_and_b64 exec, exec, s[2:3]
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %src0 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
   %src1 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx1, i32 0, i32 0, i32 0)
@@ -80,17 +107,25 @@ main_body:
 ; NOTE: llvm.amdgcn.wwm is deprecated, use llvm.amdgcn.strict.wwm instead.
 ; Make sure the transition from Exact to STRICT_WWM then softwqm does not trigger WQM.
 ;
-;CHECK-LABEL: {{^}}test_wwm1:
-;CHECK: s_or_saveexec_b64 [[ORIG0:s\[[0-9]+:[0-9]+\]]], -1
-;CHECK: buffer_load_dword
-;CHECK: s_mov_b64 exec, [[ORIG0]]
-;CHECK: buffer_store_dword
-;CHECK: s_or_saveexec_b64 [[ORIG1:s\[[0-9]+:[0-9]+\]]], -1
-;CHECK: buffer_load_dword
-;CHECK: v_add_f32_e32
-;CHECK: s_mov_b64 exec, [[ORIG1]]
-;CHECK-NOT: s_wqm_b64
 define amdgpu_ps float @test_wwm1(i32 inreg %idx0, i32 inreg %idx1) {
+; CHECK-LABEL: test_wwm1:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    s_or_saveexec_b64 s[2:3], -1
+; CHECK-NEXT:    v_mov_b32_e32 v1, s0
+; CHECK-NEXT:    buffer_load_dword v2, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_mov_b64 exec, s[2:3]
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    buffer_store_dword v2, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_or_saveexec_b64 s[2:3], -1
+; CHECK-NEXT:    v_mov_b32_e32 v1, s1
+; CHECK-NEXT:    buffer_load_dword v1, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v1, v2, v1
+; CHECK-NEXT:    s_mov_b64 exec, s[2:3]
+; CHECK-NEXT:    v_mov_b32_e32 v0, v1
+; CHECK-NEXT:    v_add_f32_e32 v0, v0, v0
+; CHECK-NEXT:    ; kill: def $vgpr0 killed $vgpr0 killed $exec killed $exec
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %src0 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
   call void @llvm.amdgcn.struct.buffer.store.f32(float %src0, <4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
@@ -104,17 +139,25 @@ main_body:
 
 ; Make sure the transition from Exact to STRICT_WWM then softwqm does not trigger WQM.
 ;
-;CHECK-LABEL: {{^}}test_strict_wwm1:
-;CHECK: s_or_saveexec_b64 [[ORIG0:s\[[0-9]+:[0-9]+\]]], -1
-;CHECK: buffer_load_dword
-;CHECK: s_mov_b64 exec, [[ORIG0]]
-;CHECK: buffer_store_dword
-;CHECK: s_or_saveexec_b64 [[ORIG1:s\[[0-9]+:[0-9]+\]]], -1
-;CHECK: buffer_load_dword
-;CHECK: v_add_f32_e32
-;CHECK: s_mov_b64 exec, [[ORIG1]]
-;CHECK-NOT: s_wqm_b64
 define amdgpu_ps float @test_strict_wwm1(i32 inreg %idx0, i32 inreg %idx1) {
+; CHECK-LABEL: test_strict_wwm1:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    s_or_saveexec_b64 s[2:3], -1
+; CHECK-NEXT:    v_mov_b32_e32 v1, s0
+; CHECK-NEXT:    buffer_load_dword v2, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_mov_b64 exec, s[2:3]
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    buffer_store_dword v2, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_or_saveexec_b64 s[2:3], -1
+; CHECK-NEXT:    v_mov_b32_e32 v1, s1
+; CHECK-NEXT:    buffer_load_dword v1, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v1, v2, v1
+; CHECK-NEXT:    s_mov_b64 exec, s[2:3]
+; CHECK-NEXT:    v_mov_b32_e32 v0, v1
+; CHECK-NEXT:    v_add_f32_e32 v0, v0, v0
+; CHECK-NEXT:    ; kill: def $vgpr0 killed $vgpr0 killed $exec killed $exec
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %src0 = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
   call void @llvm.amdgcn.struct.buffer.store.f32(float %src0, <4 x i32> undef, i32 %idx0, i32 0, i32 0, i32 0)
@@ -129,15 +172,32 @@ main_body:
 
 ; Check that softwqm on one case of branch does not trigger WQM for shader.
 ;
-;CHECK-LABEL: {{^}}test_control_flow_0:
-;CHECK-NEXT: ; %main_body
-;CHECK-NOT: s_wqm_b64 exec, exec
-;CHECK: %ELSE
-;CHECK: store
-;CHECK: %IF
-;CHECK: buffer_load
-;CHECK: buffer_load
 define amdgpu_ps float @test_control_flow_0(<8 x i32> inreg %rsrc, <4 x i32> inreg %sampler, i32 inreg %idx0, i32 inreg %idx1, i32 %c, i32 %z, float %data) {
+; CHECK-LABEL: test_control_flow_0:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    v_cmp_ne_u32_e32 vcc, 0, v1
+; CHECK-NEXT:    s_and_saveexec_b64 s[0:1], vcc
+; CHECK-NEXT:    s_xor_b64 s[0:1], exec, s[0:1]
+; CHECK-NEXT:    s_cbranch_execz .LBB6_2
+; CHECK-NEXT:  ; %bb.1: ; %ELSE
+; CHECK-NEXT:    buffer_store_dword v2, v0, s[0:3], 0 idxen
+; CHECK-NEXT:  .LBB6_2: ; %Flow
+; CHECK-NEXT:    s_or_saveexec_b64 s[0:1], s[0:1]
+; CHECK-NEXT:    s_xor_b64 exec, exec, s[0:1]
+; CHECK-NEXT:    s_cbranch_execz .LBB6_4
+; CHECK-NEXT:  ; %bb.3: ; %IF
+; CHECK-NEXT:    v_mov_b32_e32 v0, s12
+; CHECK-NEXT:    v_mov_b32_e32 v1, s13
+; CHECK-NEXT:    buffer_load_dword v0, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    buffer_load_dword v1, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v2, v0, v1
+; CHECK-NEXT:    ; kill: def $vgpr2 killed $vgpr2 killed $exec killed $exec
+; CHECK-NEXT:  .LBB6_4: ; %END
+; CHECK-NEXT:    s_or_b64 exec, exec, s[0:1]
+; CHECK-NEXT:    v_mov_b32_e32 v0, v2
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %cmp = icmp eq i32 %z, 0
   br i1 %cmp, label %IF, label %ELSE
@@ -160,20 +220,41 @@ END:
 
 ; Check that softwqm on one case of branch is treated as WQM in WQM shader.
 ;
-;CHECK-LABEL: {{^}}test_control_flow_1:
-;CHECK-NEXT: ; %main_body
-;CHECK-NEXT: s_mov_b64 [[ORIG:s\[[0-9]+:[0-9]+\]]], exec
-;CHECK-NEXT: s_wqm_b64 exec, exec
-;CHECK: %ELSE
-;CHECK: s_and_saveexec_b64 [[SAVED:s\[[0-9]+:[0-9]+\]]], [[ORIG]]
-;CHECK: store
-;CHECK: s_mov_b64 exec, [[SAVED]]
-;CHECK: %IF
-;CHECK-NOT: s_and_saveexec_b64
-;CHECK-NOT: s_and_b64 exec
-;CHECK: buffer_load
-;CHECK: buffer_load
 define amdgpu_ps float @test_control_flow_1(<8 x i32> inreg %rsrc, <4 x i32> inreg %sampler, i32 inreg %idx0, i32 inreg %idx1, i32 %c, i32 %z, float %data) {
+; CHECK-LABEL: test_control_flow_1:
+; CHECK:       ; %bb.0: ; %main_body
+; CHECK-NEXT:    s_mov_b64 s[14:15], exec
+; CHECK-NEXT:    s_wqm_b64 exec, exec
+; CHECK-NEXT:    v_cmp_ne_u32_e32 vcc, 0, v1
+; CHECK-NEXT:    s_and_saveexec_b64 s[16:17], vcc
+; CHECK-NEXT:    s_xor_b64 s[16:17], exec, s[16:17]
+; CHECK-NEXT:    s_cbranch_execz .LBB7_2
+; CHECK-NEXT:  ; %bb.1: ; %ELSE
+; CHECK-NEXT:    image_sample v1, v0, s[0:7], s[8:11] dmask:0x1
+; CHECK-NEXT:    s_and_saveexec_b64 s[18:19], s[14:15]
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    image_sample v1, v1, s[0:7], s[8:11] dmask:0x1
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    buffer_store_dword v1, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    s_mov_b64 exec, s[18:19]
+; CHECK-NEXT:  .LBB7_2: ; %Flow
+; CHECK-NEXT:    s_or_saveexec_b64 s[0:1], s[16:17]
+; CHECK-NEXT:    s_xor_b64 exec, exec, s[0:1]
+; CHECK-NEXT:    s_cbranch_execz .LBB7_4
+; CHECK-NEXT:  ; %bb.3: ; %IF
+; CHECK-NEXT:    v_mov_b32_e32 v0, s12
+; CHECK-NEXT:    v_mov_b32_e32 v1, s13
+; CHECK-NEXT:    buffer_load_dword v0, v0, s[0:3], 0 idxen
+; CHECK-NEXT:    buffer_load_dword v1, v1, s[0:3], 0 idxen
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_add_f32_e32 v2, v0, v1
+; CHECK-NEXT:    ; kill: def $vgpr2 killed $vgpr2 killed $exec killed $exec
+; CHECK-NEXT:  .LBB7_4: ; %END
+; CHECK-NEXT:    s_or_b64 exec, exec, s[0:1]
+; CHECK-NEXT:    s_and_b64 exec, exec, s[14:15]
+; CHECK-NEXT:    v_mov_b32_e32 v0, v2
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    ; return to shader part epilog
 main_body:
   %c.bc = bitcast i32 %c to float
   %tex = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %c.bc, <8 x i32> %rsrc, <4 x i32> %sampler, i1 0, i32 0, i32 0) #0
@@ -205,7 +286,6 @@ declare void @llvm.amdgcn.struct.buffer.store.v4f32(<4 x float>, <4 x i32>, i32,
 declare float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32>, i32, i32, i32, i32 immarg) #3
 declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #3
 declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #3
-declare void @llvm.amdgcn.kill(i1) #1
 declare float @llvm.amdgcn.wqm.f32(float) #3
 declare float @llvm.amdgcn.softwqm.f32(float) #3
 declare i32 @llvm.amdgcn.softwqm.i32(i32) #3