[AMDGPU] Optimize AtomicOptimizer

[Flakebi]

Summary

Mark the ctpop as convergent so it does not get moved into the
single-lane basic block. This saves us currently one instruction.
Another way to save this instruction is reusing the saved exec register
from the inserted control flow (output of s_saveexec). This is
currently hard to do though it might work when GlobalISel gets used.

[foad]

LGTM.

[Flakebi]

Unfortunately this does not work anymore with the updated ballot intrinsic. I’ll leave this for later, see also D65088.

[arsenm]

llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp

530

This won't be preserved in any meaningful way to th backend, this should be removed

llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp

if (ValDivergent) {
     switch (Op) {
     default:
       llvm_unreachable("Unhandled atomic op");
 
     case AtomicRMWInst::Add:
     case AtomicRMWInst::Sub: {
       // The new value we will be contributing to the atomic operation is the
       // old value times the number of active lanes.
-      Value *const Ctpop = B.CreateIntCast(
-          B.CreateUnaryIntrinsic(Intrinsic::ctpop, Ballot), Ty, false);
+
+      // Mark the ctpop as convergent so it does not get moved into the
+      // single-lane basic block. This saves us currently one instruction.
+      // Another way to save this instruction is reusing the saved exec register
+      // from the inserted control flow (output of s_saveexec). This is
+      // currently hard to do though it might work when GlobalISel gets used.
+      CallInst *const CtpopCall = B.CreateUnaryIntrinsic(Intrinsic::ctpop, Ballot);
+      CtpopCall->addAttribute(AttributeList::FunctionIndex, Attribute::Convergent);
+      Value *const Ctpop = B.CreateIntCast(CtpopCall, Ty, false);
       NewV = B.CreateMul(V, Ctpop);
       break;
     }
 
     case AtomicRMWInst::And:
     case AtomicRMWInst::Or:
     case AtomicRMWInst::Max:
     case AtomicRMWInst::Min:

llvm/test/CodeGen/AMDGPU/atomic_optimizations_buffer.ll

 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX7LESS %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=tonga -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -amdgpu-dpp-combine=false -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,GFX89 %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -amdgpu-dpp-combine=false -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,GFX89 %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=-wavefrontsize32,+wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -amdgpu-dpp-combine=false -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64 %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -amdgpu-dpp-combine=false -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN32,GFX8MORE,GFX8MORE32 %s
 
 declare i32 @llvm.amdgcn.workitem.id.x()
 declare i32 @llvm.amdgcn.raw.buffer.atomic.add(i32, <4 x i32>, i32, i32, i32 immarg)
 declare i32 @llvm.amdgcn.struct.buffer.atomic.add(i32, <4 x i32>, i32, i32, i32, i32 immarg)
 declare i32 @llvm.amdgcn.raw.buffer.atomic.sub(i32, <4 x i32>, i32, i32, i32 immarg)
 
 ; Show what the atomic optimization pass will do for raw buffers.
 
 ; GCN-LABEL: add_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: buffer_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_constant(i32 addrspace(1)* %out, <4 x i32> %inout) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.add(i32 5, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: buffer_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_uniform(i32 addrspace(1)* %out, <4 x i32> %inout, i32 %additive) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.add(i32 %additive, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
entry:
   %lane = call i32 @llvm.amdgcn.workitem.id.x()
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.add(i32 1, <4 x i32> %inout, i32 %lane, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: buffer_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_constant(i32 addrspace(1)* %out, <4 x i32> %inout) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.sub(i32 5, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: buffer_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_uniform(i32 addrspace(1)* %out, <4 x i32> %inout, i32 %subitive) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.sub(i32 %subitive, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void

llvm/test/CodeGen/AMDGPU/atomic_optimizations_global_pointer.ll

 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX7LESS %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=tonga -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,DPPCOMB %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,DPPCOMB %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=-wavefrontsize32,+wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64 %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN32,GFX8MORE,GFX8MORE32 %s
 
 declare i32 @llvm.amdgcn.workitem.id.x()
 
 ; Show what the atomic optimization pass will do for global pointers.
 
 ; GCN-LABEL: add_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: {{flat|buffer|global}}_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_constant(i32 addrspace(1)* %out, i32 addrspace(1)* %inout) {
 entry:
   %old = atomicrmw add i32 addrspace(1)* %inout, i32 5 acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: {{flat|buffer|global}}_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_uniform(i32 addrspace(1)* %out, i32 addrspace(1)* %inout, i32 %additive) {
 entry:
   %old = atomicrmw add i32 addrspace(1)* %inout, i32 %additive acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
entry:
   %lane = call i32 @llvm.amdgcn.workitem.id.x()
   %old = atomicrmw add i32 addrspace(1)* %inout, i32 %lane acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i64_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_hi_u32_u24{{(_e[0-9]+)?}} v[[value_hi:[0-9]+]], s[[popcount]], 5
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value_lo:[0-9]+]], s[[popcount]], 5
 ; GCN: {{flat|buffer|global}}_atomic_add_x2 v{{\[}}[[value_lo]]:[[value_hi]]{{\]}}
 define amdgpu_kernel void @add_i64_constant(i64 addrspace(1)* %out, i64 addrspace(1)* %inout) {
 entry:
   %old = atomicrmw add i64 addrspace(1)* %inout, i64 5 acq_rel
   store i64 %old, i64 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i64_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: {{flat|buffer|global}}_atomic_add_x2 v{{\[}}{{[0-9]+}}:{{[0-9]+}}{{\]}}
 define amdgpu_kernel void @add_i64_uniform(i64 addrspace(1)* %out, i64 addrspace(1)* %inout, i64 %additive) {
 entry:
   %old = atomicrmw add i64 addrspace(1)* %inout, i64 %additive acq_rel
   store i64 %old, i64 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i64_varying:
 ; GCN-NOT: v_mbcnt_lo_u32_b32
 ; GCN-NOT: v_mbcnt_hi_u32_b32
 ; GCN-NOT: s_bcnt1_i32_b64
 ; GCN: {{flat|buffer|global}}_atomic_add_x2 v{{\[}}{{[0-9]+}}:{{[0-9]+}}{{\]}}
 define amdgpu_kernel void @add_i64_varying(i64 addrspace(1)* %out, i64 addrspace(1)* %inout) {
 entry:
   %lane = call i32 @llvm.amdgcn.workitem.id.x()
   %zext = zext i32 %lane to i64
   %old = atomicrmw add i64 addrspace(1)* %inout, i64 %zext acq_rel
   store i64 %old, i64 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: {{flat|buffer|global}}_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_constant(i32 addrspace(1)* %out, i32 addrspace(1)* %inout) {
 entry:
   %old = atomicrmw sub i32 addrspace(1)* %inout, i32 5 acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: {{flat|buffer|global}}_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_uniform(i32 addrspace(1)* %out, i32 addrspace(1)* %inout, i32 %subitive) {
 entry:
   %old = atomicrmw sub i32 addrspace(1)* %inout, i32 %subitive acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
entry:
   %lane = call i32 @llvm.amdgcn.workitem.id.x()
   %old = atomicrmw sub i32 addrspace(1)* %inout, i32 %lane acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i64_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_hi_u32_u24{{(_e[0-9]+)?}} v[[value_hi:[0-9]+]], s[[popcount]], 5
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value_lo:[0-9]+]], s[[popcount]], 5
 ; GCN: {{flat|buffer|global}}_atomic_sub_x2 v{{\[}}[[value_lo]]:[[value_hi]]{{\]}}
 define amdgpu_kernel void @sub_i64_constant(i64 addrspace(1)* %out, i64 addrspace(1)* %inout) {
 entry:
   %old = atomicrmw sub i64 addrspace(1)* %inout, i64 5 acq_rel
   store i64 %old, i64 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i64_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: {{flat|buffer|global}}_atomic_sub_x2 v{{\[}}{{[0-9]+}}:{{[0-9]+}}{{\]}}
 define amdgpu_kernel void @sub_i64_uniform(i64 addrspace(1)* %out, i64 addrspace(1)* %inout, i64 %subitive) {
 entry:
   %old = atomicrmw sub i64 addrspace(1)* %inout, i64 %subitive acq_rel
   store i64 %old, i64 addrspace(1)* %out
   ret void
 }
 

llvm/test/CodeGen/AMDGPU/atomic_optimizations_local_pointer.ll

 
 ; Show what the atomic optimization pass will do for local pointers.
 
 define amdgpu_kernel void @add_i32_constant(i32 addrspace(1)* %out) {
 ;
 ;
 ; GFX7LESS-LABEL: add_i32_constant:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[2:3], exec
 ; GFX7LESS-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x9
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1
-; GFX7LESS-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX7LESS-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB0_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX7LESS-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
-; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v2, s2, 5
+; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
+; GFX7LESS-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
-; GFX7LESS-NEXT:    ds_add_rtn_u32 v1, v1, v2
+; GFX7LESS-NEXT:    ds_add_rtn_u32 v1, v2, v1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_wbinvl1
 ; GFX7LESS-NEXT:  BB0_2:
-; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX7LESS-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX7LESS-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX7LESS-NEXT:    v_mad_u32_u24 v0, v0, 5, s2
 ; GFX7LESS-NEXT:    s_mov_b32 s2, -1
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: add_i32_constant:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[2:3], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1
-; GFX8-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB0_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX8-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
 ; GFX8-NEXT:    s_mov_b32 m0, -1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    ds_add_rtn_u32 v1, v2, v1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    buffer_wbinvl1_vol
 ; GFX8-NEXT:  BB0_2:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX8-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX8-NEXT:    v_mad_u32_u24 v0, v0, 5, s2
 ; GFX8-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX8-NEXT:    s_mov_b32 s2, -1
 ; GFX8-NEXT:    s_nop 1
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: add_i32_constant:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[2:3], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1
-; GFX9-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB0_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX9-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    ds_add_rtn_u32 v1, v2, v1
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    buffer_wbinvl1_vol
 ; GFX9-NEXT:  BB0_2:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX9-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX9-NEXT:    v_mad_u32_u24 v0, v0, 5, s2
 ; GFX9-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX9-NEXT:    s_mov_b32 s2, -1
 ; GFX9-NEXT:    s_nop 1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: add_i32_constant:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[2:3], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB0_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
+; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX1064-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
-; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_add_rtn_u32 v1, v2, v1
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv
 ; GFX1064-NEXT:    buffer_gl1_inv
 ; GFX1064-NEXT:  BB0_2:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX1064-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX1064-NEXT:    s_mov_b32 s3, 0x31016000
 ; GFX1064-NEXT:    v_mad_u32_u24 v0, v0, 5, s2
 ; GFX1064-NEXT:    s_mov_b32 s2, -1
 ; GFX1064-NEXT:    s_nop 1
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: add_i32_constant:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s2, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s3, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
-; GFX1032-NEXT:    s_and_saveexec_b32 s3, vcc_lo
+; GFX1032-NEXT:    s_and_saveexec_b32 s2, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB0_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s2, s2
+; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s3, 5
 ; GFX1032-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
-; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_add_rtn_u32 v1, v2, v1
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
 ; GFX1032-NEXT:    buffer_gl1_inv
 ; GFX1032-NEXT:  BB0_2:
 ; GFX1032-NEXT:    v_nop
-; GFX1032-NEXT:    s_or_b32 exec_lo, exec_lo, s3
+; GFX1032-NEXT:    s_or_b32 exec_lo, exec_lo, s2
 ; GFX1032-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX1032-NEXT:    s_mov_b32 s3, 0x31016000
 ; GFX1032-NEXT:    v_mad_u32_u24 v0, v0, 5, s2
 ; GFX1032-NEXT:    s_mov_b32 s2, -1
 ; GFX1032-NEXT:    s_nop 1
 ; GFX1032-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX1032-NEXT:    s_endpgm
 entry:
   %old = atomicrmw add i32 addrspace(3)* @local_var32, i32 5 acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 define amdgpu_kernel void @add_i32_uniform(i32 addrspace(1)* %out, i32 %additive) {
 ;
 ;
 ; GFX7LESS-LABEL: add_i32_uniform:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[6:7], exec
 ; GFX7LESS-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x9
 ; GFX7LESS-NEXT:    s_load_dword s2, s[0:1], 0xb
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s3, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1
 ; GFX7LESS-NEXT:    s_and_saveexec_b64 s[0:1], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB1_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s3, s[6:7]
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    s_mul_i32 s3, s2, s3
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v2, s3
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    ds_add_rtn_u32 v1, v1, v2
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)

 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: add_i32_uniform:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX8-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[2:3], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s1, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1
-; GFX8-NEXT:    s_and_saveexec_b64 s[6:7], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB1_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s1, s[2:3]
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    s_mul_i32 s1, s0, s1
 ; GFX8-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX8-NEXT:    v_mov_b32_e32 v2, s1
 ; GFX8-NEXT:    s_mov_b32 m0, -1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    ds_add_rtn_u32 v1, v1, v2
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    buffer_wbinvl1_vol
 ; GFX8-NEXT:  BB1_2:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[6:7]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    v_mul_lo_u32 v0, s0, v0
 ; GFX8-NEXT:    v_readfirstlane_b32 s0, v1
 ; GFX8-NEXT:    s_mov_b32 s7, 0xf000
 ; GFX8-NEXT:    s_mov_b32 s6, -1
 ; GFX8-NEXT:    v_add_u32_e32 v0, vcc, s0, v0
 ; GFX8-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: add_i32_uniform:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX9-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[2:3], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s1, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1
-; GFX9-NEXT:    s_and_saveexec_b64 s[6:7], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB1_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s1, s[2:3]
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_mul_i32 s1, s0, s1
 ; GFX9-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s1
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    ds_add_rtn_u32 v1, v1, v2
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    buffer_wbinvl1_vol
 ; GFX9-NEXT:  BB1_2:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[6:7]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    v_mul_lo_u32 v0, s0, v0
 ; GFX9-NEXT:    v_readfirstlane_b32 s0, v1
 ; GFX9-NEXT:    s_mov_b32 s7, 0xf000
 ; GFX9-NEXT:    s_mov_b32 s6, -1
 ; GFX9-NEXT:    v_add_u32_e32 v0, s0, v0
 ; GFX9-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: add_i32_uniform:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX1064-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[2:3], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s1, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[6:7], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB1_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s1, s[2:3]
-; GFX1064-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    s_mul_i32 s1, s0, s1
+; GFX1064-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1064-NEXT:    v_mov_b32_e32 v2, s1
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_add_rtn_u32 v1, v1, v2
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv
 ; GFX1064-NEXT:    buffer_gl1_inv
 ; GFX1064-NEXT:  BB1_2:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[6:7]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    v_mul_lo_u32 v0, s0, v0
 ; GFX1064-NEXT:    v_readfirstlane_b32 s0, v1
 ; GFX1064-NEXT:    s_mov_b32 s7, 0x31016000
 ; GFX1064-NEXT:    s_mov_b32 s6, -1
 ; GFX1064-NEXT:    v_add_nc_u32_e32 v0, s0, v0
 ; GFX1064-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: add_i32_uniform:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX1032-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s2, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s2, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s1, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB1_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s2, s2
-; GFX1032-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1032-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1032-NEXT:    s_mul_i32 s2, s0, s2
+; GFX1032-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1032-NEXT:    v_mov_b32_e32 v2, s2
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_add_rtn_u32 v1, v1, v2
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
 ; GFX1032-NEXT:    buffer_gl1_inv
 ; GFX1032-NEXT:  BB1_2:
entry:
   ret void
 }
 
 define amdgpu_kernel void @add_i64_constant(i64 addrspace(1)* %out) {
 ;
 ;
 ; GFX7LESS-LABEL: add_i64_constant:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[2:3], exec
 ; GFX7LESS-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x9
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1_vgpr2
-; GFX7LESS-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX7LESS-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB5_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
+; GFX7LESS-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
-; GFX7LESS-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_wbinvl1
 ; GFX7LESS-NEXT:  BB5_2:
-; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX7LESS-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX7LESS-NEXT:    v_readfirstlane_b32 s4, v2
 ; GFX7LESS-NEXT:    v_mul_hi_u32_u24_e32 v1, 5, v0
 ; GFX7LESS-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX7LESS-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v2, s4
 ; GFX7LESS-NEXT:    v_add_i32_e32 v0, vcc, s2, v0
 ; GFX7LESS-NEXT:    v_addc_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX7LESS-NEXT:    s_mov_b32 s2, -1
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: add_i64_constant:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[2:3], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1_vgpr2
-; GFX8-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB5_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX8-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX8-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX8-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX8-NEXT:    s_mov_b32 m0, -1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    buffer_wbinvl1_vol
 ; GFX8-NEXT:  BB5_2:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX8-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX8-NEXT:    v_readfirstlane_b32 s3, v2
 ; GFX8-NEXT:    v_mov_b32_e32 v1, s2
 ; GFX8-NEXT:    v_mov_b32_e32 v2, s3
 ; GFX8-NEXT:    v_mad_u64_u32 v[0:1], s[2:3], v0, 5, v[1:2]
 ; GFX8-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX8-NEXT:    s_mov_b32 s2, -1
 ; GFX8-NEXT:    s_nop 2
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: add_i64_constant:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[2:3], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1_vgpr2
-; GFX9-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB5_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX9-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX9-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX9-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    buffer_wbinvl1_vol
 ; GFX9-NEXT:  BB5_2:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX9-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX9-NEXT:    v_readfirstlane_b32 s3, v2
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s2
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s3
 ; GFX9-NEXT:    v_mad_u64_u32 v[0:1], s[2:3], v0, 5, v[1:2]
 ; GFX9-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX9-NEXT:    s_mov_b32 s2, -1
 ; GFX9-NEXT:    s_nop 2
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: add_i64_constant:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[2:3], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB5_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
+; GFX1064-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX1064-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
-; GFX1064-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv
 ; GFX1064-NEXT:    buffer_gl1_inv
 ; GFX1064-NEXT:  BB5_2:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX1064-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX1064-NEXT:    v_readfirstlane_b32 s3, v2
 ; GFX1064-NEXT:    v_mad_u64_u32 v[0:1], s[2:3], v0, 5, s[2:3]
 ; GFX1064-NEXT:    s_mov_b32 s3, 0x31016000
 ; GFX1064-NEXT:    s_mov_b32 s2, -1
 ; GFX1064-NEXT:    s_nop 2
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: add_i64_constant:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s3, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s3, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s2, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB5_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s3, s3
-; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    v_mul_hi_u32_u24_e64 v2, s3, 5
 ; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s3, 5
+; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
 ; GFX1032-NEXT:    buffer_gl1_inv
 ; GFX1032-NEXT:  BB5_2:
 ; GFX1032-NEXT:    v_nop
entry:
   ret void
 }
 
 define amdgpu_kernel void @add_i64_uniform(i64 addrspace(1)* %out, i64 %additive) {
 ;
 ;
 ; GFX7LESS-LABEL: add_i64_uniform:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[6:7], exec
 ; GFX7LESS-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX7LESS-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB6_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
-; GFX7LESS-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX7LESS-NEXT:    v_mul_hi_u32 v1, s2, v1
 ; GFX7LESS-NEXT:    s_mul_i32 s6, s2, s6
+; GFX7LESS-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX7LESS-NEXT:    v_add_i32_e32 v2, vcc, s7, v1
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_wbinvl1
 ; GFX7LESS-NEXT:  BB6_2:

 ; GFX7LESS-NEXT:    buffer_store_dwordx2 v[0:1], off, s[4:7], 0
 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: add_i64_uniform:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[6:7], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX8-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB6_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
 ; GFX8-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    v_mul_hi_u32 v1, s2, v1
 ; GFX8-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX8-NEXT:    s_mul_i32 s6, s2, s6
 ; GFX8-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX8-NEXT:    v_add_u32_e32 v2, vcc, s7, v1
 ; GFX8-NEXT:    v_mov_b32_e32 v1, s6

 ; GFX8-NEXT:    buffer_store_dwordx2 v[0:1], off, s[4:7], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: add_i64_uniform:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[6:7], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX9-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB6_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX9-NEXT:    s_mul_hi_u32 s8, s2, s6
 ; GFX9-NEXT:    s_add_i32 s8, s8, s7
 ; GFX9-NEXT:    s_mul_i32 s6, s2, s6
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s8
 ; GFX9-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo

 ; GFX9-NEXT:    v_addc_co_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX9-NEXT:    buffer_store_dwordx2 v[0:1], off, s[4:7], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: add_i64_uniform:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[6:7], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX1064-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB6_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
-; GFX1064-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX1064-NEXT:    s_mul_hi_u32 s8, s2, s6
 ; GFX1064-NEXT:    s_mul_i32 s6, s2, s6
+; GFX1064-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1064-NEXT:    s_add_i32 s8, s8, s7
 ; GFX1064-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX1064-NEXT:    v_mov_b32_e32 v2, s8
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv

 ; GFX1064-NEXT:    v_add_co_ci_u32_e32 v1, vcc, s5, v1, vcc
 ; GFX1064-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: add_i64_uniform:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s5, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s5, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s4, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB6_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s5, s5
-; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1032-NEXT:    s_mul_i32 s6, s3, s5
 ; GFX1032-NEXT:    s_mul_hi_u32 s7, s2, s5
 ; GFX1032-NEXT:    s_mul_i32 s5, s2, s5
+; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    s_add_i32 s7, s7, s6
 ; GFX1032-NEXT:    v_mov_b32_e32 v1, s5
 ; GFX1032-NEXT:    v_mov_b32_e32 v2, s7
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_add_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
entry:
   ret void
 }
 
 define amdgpu_kernel void @sub_i32_constant(i32 addrspace(1)* %out) {
 ;
 ;
 ; GFX7LESS-LABEL: sub_i32_constant:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[2:3], exec
 ; GFX7LESS-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x9
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1
-; GFX7LESS-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX7LESS-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB8_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX7LESS-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
-; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v2, s2, 5
+; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
+; GFX7LESS-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
-; GFX7LESS-NEXT:    ds_sub_rtn_u32 v1, v1, v2
+; GFX7LESS-NEXT:    ds_sub_rtn_u32 v1, v2, v1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_wbinvl1
 ; GFX7LESS-NEXT:  BB8_2:
-; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX7LESS-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX7LESS-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX7LESS-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX7LESS-NEXT:    v_sub_i32_e32 v0, vcc, s2, v0
 ; GFX7LESS-NEXT:    s_mov_b32 s2, -1
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: sub_i32_constant:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[2:3], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1
-; GFX8-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB8_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX8-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
 ; GFX8-NEXT:    s_mov_b32 m0, -1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    ds_sub_rtn_u32 v1, v2, v1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    buffer_wbinvl1_vol
 ; GFX8-NEXT:  BB8_2:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX8-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX8-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX8-NEXT:    v_sub_u32_e32 v0, vcc, s2, v0
 ; GFX8-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX8-NEXT:    s_mov_b32 s2, -1
 ; GFX8-NEXT:    s_nop 0
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: sub_i32_constant:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[2:3], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1
-; GFX9-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB8_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX9-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    ds_sub_rtn_u32 v1, v2, v1
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    buffer_wbinvl1_vol
 ; GFX9-NEXT:  BB8_2:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX9-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX9-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX9-NEXT:    v_sub_u32_e32 v0, s2, v0
 ; GFX9-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX9-NEXT:    s_mov_b32 s2, -1
 ; GFX9-NEXT:    s_nop 0
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: sub_i32_constant:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[2:3], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB8_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
+; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX1064-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
-; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_sub_rtn_u32 v1, v2, v1
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv
 ; GFX1064-NEXT:    buffer_gl1_inv
 ; GFX1064-NEXT:  BB8_2:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX1064-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX1064-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX1064-NEXT:    s_mov_b32 s3, 0x31016000
 ; GFX1064-NEXT:    v_sub_nc_u32_e32 v0, s2, v0
 ; GFX1064-NEXT:    s_mov_b32 s2, -1
 ; GFX1064-NEXT:    s_nop 0
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: sub_i32_constant:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s2, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s3, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
-; GFX1032-NEXT:    s_and_saveexec_b32 s3, vcc_lo
+; GFX1032-NEXT:    s_and_saveexec_b32 s2, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB8_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s2, s2
+; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s3, 5
 ; GFX1032-NEXT:    v_mov_b32_e32 v2, local_var32@abs32@lo
-; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_sub_rtn_u32 v1, v2, v1
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
 ; GFX1032-NEXT:    buffer_gl1_inv
 ; GFX1032-NEXT:  BB8_2:
 ; GFX1032-NEXT:    v_nop
-; GFX1032-NEXT:    s_or_b32 exec_lo, exec_lo, s3
+; GFX1032-NEXT:    s_or_b32 exec_lo, exec_lo, s2
 ; GFX1032-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX1032-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX1032-NEXT:    s_mov_b32 s3, 0x31016000
 ; GFX1032-NEXT:    v_sub_nc_u32_e32 v0, s2, v0
 ; GFX1032-NEXT:    s_mov_b32 s2, -1
 ; GFX1032-NEXT:    s_nop 0
 ; GFX1032-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX1032-NEXT:    s_endpgm
 entry:
   %old = atomicrmw sub i32 addrspace(3)* @local_var32, i32 5 acq_rel
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 define amdgpu_kernel void @sub_i32_uniform(i32 addrspace(1)* %out, i32 %subitive) {
 ;
 ;
 ; GFX7LESS-LABEL: sub_i32_uniform:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[6:7], exec
 ; GFX7LESS-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x9
 ; GFX7LESS-NEXT:    s_load_dword s2, s[0:1], 0xb
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s3, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1
 ; GFX7LESS-NEXT:    s_and_saveexec_b64 s[0:1], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB9_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s3, s[6:7]
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    s_mul_i32 s3, s2, s3
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v2, s3
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    ds_sub_rtn_u32 v1, v1, v2
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)

 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: sub_i32_uniform:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX8-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[2:3], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s1, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1
-; GFX8-NEXT:    s_and_saveexec_b64 s[6:7], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB9_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s1, s[2:3]
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    s_mul_i32 s1, s0, s1
 ; GFX8-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX8-NEXT:    v_mov_b32_e32 v2, s1
 ; GFX8-NEXT:    s_mov_b32 m0, -1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    ds_sub_rtn_u32 v1, v1, v2
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    buffer_wbinvl1_vol
 ; GFX8-NEXT:  BB9_2:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[6:7]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    v_mul_lo_u32 v0, s0, v0
 ; GFX8-NEXT:    v_readfirstlane_b32 s0, v1
 ; GFX8-NEXT:    s_mov_b32 s7, 0xf000
 ; GFX8-NEXT:    s_mov_b32 s6, -1
 ; GFX8-NEXT:    v_sub_u32_e32 v0, vcc, s0, v0
 ; GFX8-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: sub_i32_uniform:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX9-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[2:3], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s1, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1
-; GFX9-NEXT:    s_and_saveexec_b64 s[6:7], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB9_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s1, s[2:3]
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_mul_i32 s1, s0, s1
 ; GFX9-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s1
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    ds_sub_rtn_u32 v1, v1, v2
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    buffer_wbinvl1_vol
 ; GFX9-NEXT:  BB9_2:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[6:7]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    v_mul_lo_u32 v0, s0, v0
 ; GFX9-NEXT:    v_readfirstlane_b32 s0, v1
 ; GFX9-NEXT:    s_mov_b32 s7, 0xf000
 ; GFX9-NEXT:    s_mov_b32 s6, -1
 ; GFX9-NEXT:    v_sub_u32_e32 v0, s0, v0
 ; GFX9-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: sub_i32_uniform:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX1064-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[2:3], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s1, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[6:7], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB9_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s1, s[2:3]
-; GFX1064-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    s_mul_i32 s1, s0, s1
+; GFX1064-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1064-NEXT:    v_mov_b32_e32 v2, s1
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_sub_rtn_u32 v1, v1, v2
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv
 ; GFX1064-NEXT:    buffer_gl1_inv
 ; GFX1064-NEXT:  BB9_2:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[6:7]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    v_mul_lo_u32 v0, s0, v0
 ; GFX1064-NEXT:    v_readfirstlane_b32 s0, v1
 ; GFX1064-NEXT:    s_mov_b32 s7, 0x31016000
 ; GFX1064-NEXT:    s_mov_b32 s6, -1
 ; GFX1064-NEXT:    v_sub_nc_u32_e32 v0, s0, v0
 ; GFX1064-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: sub_i32_uniform:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
 ; GFX1032-NEXT:    s_load_dword s0, s[0:1], 0x2c
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s2, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s2, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s1, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB9_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s2, s2
-; GFX1032-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1032-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1032-NEXT:    s_mul_i32 s2, s0, s2
+; GFX1032-NEXT:    v_mov_b32_e32 v1, local_var32@abs32@lo
 ; GFX1032-NEXT:    v_mov_b32_e32 v2, s2
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_sub_rtn_u32 v1, v1, v2
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
 ; GFX1032-NEXT:    buffer_gl1_inv
 ; GFX1032-NEXT:  BB9_2:
entry:
   ret void
 }
 
 define amdgpu_kernel void @sub_i64_constant(i64 addrspace(1)* %out) {
 ;
 ;
 ; GFX7LESS-LABEL: sub_i64_constant:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[2:3], exec
 ; GFX7LESS-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x9
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1_vgpr2
-; GFX7LESS-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX7LESS-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB11_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
+; GFX7LESS-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
-; GFX7LESS-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX7LESS-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_wbinvl1
 ; GFX7LESS-NEXT:  BB11_2:
-; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX7LESS-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX7LESS-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX7LESS-NEXT:    v_readfirstlane_b32 s4, v2
 ; GFX7LESS-NEXT:    v_mul_hi_u32_u24_e32 v1, 5, v0
 ; GFX7LESS-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX7LESS-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v2, s4
 ; GFX7LESS-NEXT:    v_sub_i32_e32 v0, vcc, s2, v0
 ; GFX7LESS-NEXT:    v_subb_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX7LESS-NEXT:    s_mov_b32 s2, -1
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: sub_i64_constant:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[2:3], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1_vgpr2
-; GFX8-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB11_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX8-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX8-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX8-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX8-NEXT:    s_mov_b32 m0, -1
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX8-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX8-NEXT:    buffer_wbinvl1_vol
 ; GFX8-NEXT:  BB11_2:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX8-NEXT:    v_readfirstlane_b32 s3, v2
 ; GFX8-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX8-NEXT:    v_mul_hi_u32_u24_e32 v1, 5, v0
 ; GFX8-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX8-NEXT:    v_mov_b32_e32 v2, s3
 ; GFX8-NEXT:    v_sub_u32_e32 v0, vcc, s2, v0
 ; GFX8-NEXT:    v_subb_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX8-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX8-NEXT:    s_mov_b32 s2, -1
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: sub_i64_constant:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[2:3], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1_vgpr2
-; GFX9-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB11_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
-; GFX9-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
+; GFX9-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX9-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    buffer_wbinvl1_vol
 ; GFX9-NEXT:  BB11_2:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX9-NEXT:    v_readfirstlane_b32 s3, v2
 ; GFX9-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX9-NEXT:    v_mul_hi_u32_u24_e32 v1, 5, v0
 ; GFX9-NEXT:    v_mul_u32_u24_e32 v0, 5, v0
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s3
 ; GFX9-NEXT:    v_sub_co_u32_e32 v0, vcc, s2, v0
 ; GFX9-NEXT:    v_subb_co_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX9-NEXT:    s_mov_b32 s3, 0xf000
 ; GFX9-NEXT:    s_mov_b32 s2, -1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: sub_i64_constant:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[2:3], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s4, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[4:5], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[2:3], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB11_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s2, s[2:3]
+; GFX1064-NEXT:    v_mul_hi_u32_u24_e64 v2, s4, 5
+; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s4, 5
 ; GFX1064-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
-; GFX1064-NEXT:    v_mul_hi_u32_u24_e64 v2, s2, 5
-; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s2, 5
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv
 ; GFX1064-NEXT:    buffer_gl1_inv
 ; GFX1064-NEXT:  BB11_2:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[4:5]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[2:3]
 ; GFX1064-NEXT:    v_readfirstlane_b32 s2, v1
 ; GFX1064-NEXT:    v_mul_u32_u24_e32 v1, 5, v0
 ; GFX1064-NEXT:    v_readfirstlane_b32 s3, v2
 ; GFX1064-NEXT:    v_mul_hi_u32_u24_e32 v2, 5, v0
 ; GFX1064-NEXT:    v_sub_co_u32_e64 v0, vcc, s2, v1
 ; GFX1064-NEXT:    s_mov_b32 s2, -1
 ; GFX1064-NEXT:    v_sub_co_ci_u32_e32 v1, vcc, s3, v2, vcc
 ; GFX1064-NEXT:    s_mov_b32 s3, 0x31016000
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: sub_i64_constant:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s3, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s3, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s2, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB11_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s3, s3
-; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    v_mul_hi_u32_u24_e64 v2, s3, 5
 ; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s3, 5
+; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv
 ; GFX1032-NEXT:    buffer_gl1_inv
 ; GFX1032-NEXT:  BB11_2:
 ; GFX1032-NEXT:    v_nop
entry:
   ret void
 }
 
 define amdgpu_kernel void @sub_i64_uniform(i64 addrspace(1)* %out, i64 %subitive) {
 ;
 ;
 ; GFX7LESS-LABEL: sub_i64_uniform:
 ; GFX7LESS:       ; %bb.0: ; %entry
-; GFX7LESS-NEXT:    s_mov_b64 s[6:7], exec
 ; GFX7LESS-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
 ; GFX7LESS-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7LESS-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX7LESS-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7LESS-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX7LESS-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX7LESS-NEXT:    s_cbranch_execz BB12_2
 ; GFX7LESS-NEXT:  ; %bb.1:
-; GFX7LESS-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
-; GFX7LESS-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX7LESS-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX7LESS-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX7LESS-NEXT:    v_mul_hi_u32 v1, s2, v1
 ; GFX7LESS-NEXT:    s_mul_i32 s6, s2, s6
+; GFX7LESS-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX7LESS-NEXT:    v_add_i32_e32 v2, vcc, s7, v1
 ; GFX7LESS-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX7LESS-NEXT:    s_mov_b32 m0, -1
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX7LESS-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX7LESS-NEXT:    buffer_wbinvl1
 ; GFX7LESS-NEXT:  BB12_2:

 ; GFX7LESS-NEXT:    buffer_store_dwordx2 v[0:1], off, s[4:7], 0
 ; GFX7LESS-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: sub_i64_uniform:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[6:7], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX8-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB12_2
 ; GFX8-NEXT:  ; %bb.1:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
 ; GFX8-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX8-NEXT:    v_mul_hi_u32 v1, s2, v1
 ; GFX8-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX8-NEXT:    s_mul_i32 s6, s2, s6
 ; GFX8-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX8-NEXT:    v_add_u32_e32 v2, vcc, s7, v1
 ; GFX8-NEXT:    v_mov_b32_e32 v1, s6

 ; GFX8-NEXT:    buffer_store_dwordx2 v[0:1], off, s[4:7], 0
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: sub_i64_uniform:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[6:7], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX9-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB12_2
 ; GFX9-NEXT:  ; %bb.1:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX9-NEXT:    s_mul_hi_u32 s8, s2, s6
 ; GFX9-NEXT:    s_add_i32 s8, s8, s7
 ; GFX9-NEXT:    s_mul_i32 s6, s2, s6
 ; GFX9-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s8
 ; GFX9-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo

 ; GFX9-NEXT:    v_subb_co_u32_e32 v1, vcc, v2, v1, vcc
 ; GFX9-NEXT:    buffer_store_dwordx2 v[0:1], off, s[4:7], 0
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: sub_i64_uniform:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[6:7], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s6, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX1064-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB12_2
 ; GFX1064-NEXT:  ; %bb.1:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s6, s[6:7]
-; GFX1064-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1064-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1064-NEXT:    s_mul_i32 s7, s3, s6
 ; GFX1064-NEXT:    s_mul_hi_u32 s8, s2, s6
 ; GFX1064-NEXT:    s_mul_i32 s6, s2, s6
+; GFX1064-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1064-NEXT:    s_add_i32 s8, s8, s7
 ; GFX1064-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX1064-NEXT:    v_mov_b32_e32 v2, s8
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1064-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1064-NEXT:    buffer_gl0_inv

 ; GFX1064-NEXT:    v_sub_co_ci_u32_e32 v1, vcc, s5, v1, vcc
 ; GFX1064-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: sub_i64_uniform:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s5, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s5, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1_vgpr2
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s4, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB12_2
 ; GFX1032-NEXT:  ; %bb.1:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s5, s5
-; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX1032-NEXT:    s_mul_i32 s6, s3, s5
 ; GFX1032-NEXT:    s_mul_hi_u32 s7, s2, s5
 ; GFX1032-NEXT:    s_mul_i32 s5, s2, s5
+; GFX1032-NEXT:    v_mov_b32_e32 v3, local_var64@abs32@lo
 ; GFX1032-NEXT:    s_add_i32 s7, s7, s6
 ; GFX1032-NEXT:    v_mov_b32_e32 v1, s5
 ; GFX1032-NEXT:    v_mov_b32_e32 v2, s7
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    s_waitcnt_vscnt null, 0x0
 ; GFX1032-NEXT:    ds_sub_rtn_u64 v[1:2], v3, v[1:2]
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
 ; GFX1032-NEXT:    buffer_gl0_inv

llvm/test/CodeGen/AMDGPU/atomic_optimizations_pixelshader.ll

 define amdgpu_ps void @add_i32_constant(<4 x i32> inreg %out, <4 x i32> inreg %inout) {
 ; GFX7-LABEL: add_i32_constant:
 ; GFX7:       ; %bb.0: ; %entry
 ; GFX7-NEXT:    s_mov_b64 s[10:11], exec
 ; GFX7-NEXT:    ; implicit-def: $vgpr0
 ; GFX7-NEXT:    s_and_saveexec_b64 s[8:9], s[10:11]
 ; GFX7-NEXT:    s_cbranch_execz BB0_4
 ; GFX7-NEXT:  ; %bb.1:
-; GFX7-NEXT:    s_mov_b64 s[10:11], exec
 ; GFX7-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
 ; GFX7-NEXT:    v_mbcnt_hi_u32_b32_e32 v0, exec_hi, v0
+; GFX7-NEXT:    s_bcnt1_i32_b64 s12, exec
 ; GFX7-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX7-NEXT:    ; implicit-def: $vgpr1
-; GFX7-NEXT:    s_and_saveexec_b64 s[12:13], vcc
+; GFX7-NEXT:    s_and_saveexec_b64 s[10:11], vcc
 ; GFX7-NEXT:    s_cbranch_execz BB0_3
 ; GFX7-NEXT:  ; %bb.2:
-; GFX7-NEXT:    s_bcnt1_i32_b64 s10, s[10:11]
-; GFX7-NEXT:    v_mul_u32_u24_e64 v1, s10, 5
+; GFX7-NEXT:    v_mul_u32_u24_e64 v1, s12, 5
 ; GFX7-NEXT:    buffer_atomic_add v1, off, s[4:7], 0 glc
 ; GFX7-NEXT:  BB0_3:
-; GFX7-NEXT:    s_or_b64 exec, exec, s[12:13]
+; GFX7-NEXT:    s_or_b64 exec, exec, s[10:11]
 ; GFX7-NEXT:    s_waitcnt vmcnt(0)
 ; GFX7-NEXT:    v_readfirstlane_b32 s4, v1
 ; GFX7-NEXT:    v_mad_u32_u24 v0, v0, 5, s4
 ; GFX7-NEXT:  BB0_4: ; %Flow
 ; GFX7-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GFX7-NEXT:    s_wqm_b64 s[4:5], -1
 ; GFX7-NEXT:    s_andn2_b64 vcc, exec, s[4:5]
 ; GFX7-NEXT:    s_cbranch_vccnz BB0_6
 ; GFX7-NEXT:  ; %bb.5: ; %if
 ; GFX7-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX7-NEXT:  BB0_6: ; %UnifiedReturnBlock
 ; GFX7-NEXT:    s_endpgm
 ;
 ; GFX8-LABEL: add_i32_constant:
 ; GFX8:       ; %bb.0: ; %entry
 ; GFX8-NEXT:    s_mov_b64 s[10:11], exec
 ; GFX8-NEXT:    ; implicit-def: $vgpr0
 ; GFX8-NEXT:    s_and_saveexec_b64 s[8:9], s[10:11]
 ; GFX8-NEXT:    s_cbranch_execz BB0_4
 ; GFX8-NEXT:  ; %bb.1:
 ; GFX8-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX8-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX8-NEXT:    s_mov_b64 s[10:11], exec
+; GFX8-NEXT:    s_bcnt1_i32_b64 s12, exec
 ; GFX8-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX8-NEXT:    ; implicit-def: $vgpr1
-; GFX8-NEXT:    s_and_saveexec_b64 s[12:13], vcc
+; GFX8-NEXT:    s_and_saveexec_b64 s[10:11], vcc
 ; GFX8-NEXT:    s_cbranch_execz BB0_3
 ; GFX8-NEXT:  ; %bb.2:
-; GFX8-NEXT:    s_bcnt1_i32_b64 s10, s[10:11]
-; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s10, 5
+; GFX8-NEXT:    v_mul_u32_u24_e64 v1, s12, 5
 ; GFX8-NEXT:    buffer_atomic_add v1, off, s[4:7], 0 glc
 ; GFX8-NEXT:  BB0_3:
-; GFX8-NEXT:    s_or_b64 exec, exec, s[12:13]
+; GFX8-NEXT:    s_or_b64 exec, exec, s[10:11]
 ; GFX8-NEXT:    s_waitcnt vmcnt(0)
 ; GFX8-NEXT:    v_readfirstlane_b32 s4, v1
 ; GFX8-NEXT:    v_mad_u32_u24 v0, v0, 5, s4
 ; GFX8-NEXT:  BB0_4: ; %Flow
 ; GFX8-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GFX8-NEXT:    s_wqm_b64 s[4:5], -1
 ; GFX8-NEXT:    s_andn2_b64 vcc, exec, s[4:5]
 ; GFX8-NEXT:    s_cbranch_vccnz BB0_6
 ; GFX8-NEXT:  ; %bb.5: ; %if
 ; GFX8-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX8-NEXT:  BB0_6: ; %UnifiedReturnBlock
 ; GFX8-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: add_i32_constant:
 ; GFX9:       ; %bb.0: ; %entry
 ; GFX9-NEXT:    s_mov_b64 s[10:11], exec
 ; GFX9-NEXT:    ; implicit-def: $vgpr0
 ; GFX9-NEXT:    s_and_saveexec_b64 s[8:9], s[10:11]
 ; GFX9-NEXT:    s_cbranch_execz BB0_4
 ; GFX9-NEXT:  ; %bb.1:
 ; GFX9-NEXT:    v_mbcnt_lo_u32_b32 v0, exec_lo, 0
 ; GFX9-NEXT:    v_mbcnt_hi_u32_b32 v0, exec_hi, v0
-; GFX9-NEXT:    s_mov_b64 s[10:11], exec
+; GFX9-NEXT:    s_bcnt1_i32_b64 s12, exec
 ; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
 ; GFX9-NEXT:    ; implicit-def: $vgpr1
-; GFX9-NEXT:    s_and_saveexec_b64 s[12:13], vcc
+; GFX9-NEXT:    s_and_saveexec_b64 s[10:11], vcc
 ; GFX9-NEXT:    s_cbranch_execz BB0_3
 ; GFX9-NEXT:  ; %bb.2:
-; GFX9-NEXT:    s_bcnt1_i32_b64 s10, s[10:11]
-; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s10, 5
+; GFX9-NEXT:    v_mul_u32_u24_e64 v1, s12, 5
 ; GFX9-NEXT:    buffer_atomic_add v1, off, s[4:7], 0 glc
 ; GFX9-NEXT:  BB0_3:
-; GFX9-NEXT:    s_or_b64 exec, exec, s[12:13]
+; GFX9-NEXT:    s_or_b64 exec, exec, s[10:11]
 ; GFX9-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-NEXT:    v_readfirstlane_b32 s4, v1
 ; GFX9-NEXT:    v_mad_u32_u24 v0, v0, 5, s4
 ; GFX9-NEXT:  BB0_4: ; %Flow
 ; GFX9-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GFX9-NEXT:    s_wqm_b64 s[4:5], -1
 ; GFX9-NEXT:    s_andn2_b64 vcc, exec, s[4:5]
 ; GFX9-NEXT:    s_cbranch_vccnz BB0_6
 ; GFX9-NEXT:  ; %bb.5: ; %if
 ; GFX9-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX9-NEXT:  BB0_6: ; %UnifiedReturnBlock
 ; GFX9-NEXT:    s_endpgm
 ;
 ; GFX1064-LABEL: add_i32_constant:
 ; GFX1064:       ; %bb.0: ; %entry
 ; GFX1064-NEXT:    s_mov_b64 s[10:11], exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr0
 ; GFX1064-NEXT:    s_and_saveexec_b64 s[8:9], s[10:11]
 ; GFX1064-NEXT:    s_cbranch_execz BB0_4
 ; GFX1064-NEXT:  ; %bb.1:
 ; GFX1064-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1064-NEXT:    s_mov_b64 s[10:11], exec
+; GFX1064-NEXT:    s_bcnt1_i32_b64 s12, exec
 ; GFX1064-NEXT:    ; implicit-def: $vgpr1
 ; GFX1064-NEXT:    v_mbcnt_hi_u32_b32_e64 v0, exec_hi, v0
 ; GFX1064-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
-; GFX1064-NEXT:    s_and_saveexec_b64 s[12:13], vcc
+; GFX1064-NEXT:    s_and_saveexec_b64 s[30:31], vcc
 ; GFX1064-NEXT:    s_cbranch_execz BB0_3
 ; GFX1064-NEXT:  ; %bb.2:
-; GFX1064-NEXT:    s_bcnt1_i32_b64 s10, s[10:11]
-; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s10, 5
+; GFX1064-NEXT:    v_mul_u32_u24_e64 v1, s12, 5
 ; GFX1064-NEXT:    buffer_atomic_add v1, off, s[4:7], 0 glc
 ; GFX1064-NEXT:  BB0_3:
 ; GFX1064-NEXT:    v_nop
-; GFX1064-NEXT:    s_or_b64 exec, exec, s[12:13]
+; GFX1064-NEXT:    s_or_b64 exec, exec, s[30:31]
 ; GFX1064-NEXT:    s_waitcnt vmcnt(0)
 ; GFX1064-NEXT:    v_readfirstlane_b32 s4, v1
 ; GFX1064-NEXT:    v_mad_u32_u24 v0, v0, 5, s4
 ; GFX1064-NEXT:  BB0_4: ; %Flow
 ; GFX1064-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GFX1064-NEXT:    s_wqm_b64 s[4:5], -1
 ; GFX1064-NEXT:    s_andn2_b64 vcc, exec, s[4:5]
 ; GFX1064-NEXT:    s_cbranch_vccnz BB0_6
 ; GFX1064-NEXT:  ; %bb.5: ; %if
 ; GFX1064-NEXT:    buffer_store_dword v0, off, s[0:3], 0
 ; GFX1064-NEXT:  BB0_6: ; %UnifiedReturnBlock
 ; GFX1064-NEXT:    s_endpgm
 ;
 ; GFX1032-LABEL: add_i32_constant:
 ; GFX1032:       ; %bb.0: ; %entry
 ; GFX1032-NEXT:    s_mov_b32 s9, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vgpr0
 ; GFX1032-NEXT:    ; implicit-def: $vcc_hi
 ; GFX1032-NEXT:    s_and_saveexec_b32 s8, s9
 ; GFX1032-NEXT:    s_cbranch_execz BB0_4
 ; GFX1032-NEXT:  ; %bb.1:
 ; GFX1032-NEXT:    v_mbcnt_lo_u32_b32_e64 v0, exec_lo, 0
-; GFX1032-NEXT:    s_mov_b32 s10, exec_lo
+; GFX1032-NEXT:    s_bcnt1_i32_b32 s10, exec_lo
 ; GFX1032-NEXT:    ; implicit-def: $vgpr1
 ; GFX1032-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 0, v0
 ; GFX1032-NEXT:    s_and_saveexec_b32 s9, vcc_lo
 ; GFX1032-NEXT:    s_cbranch_execz BB0_3
 ; GFX1032-NEXT:  ; %bb.2:
-; GFX1032-NEXT:    s_bcnt1_i32_b32 s10, s10
 ; GFX1032-NEXT:    v_mul_u32_u24_e64 v1, s10, 5
 ; GFX1032-NEXT:    buffer_atomic_add v1, off, s[4:7], 0 glc
 ; GFX1032-NEXT:  BB0_3:
 ; GFX1032-NEXT:    v_nop
 ; GFX1032-NEXT:    s_or_b32 exec_lo, exec_lo, s9
 ; GFX1032-NEXT:    s_waitcnt vmcnt(0)
 ; GFX1032-NEXT:    v_readfirstlane_b32 s4, v1
 ; GFX1032-NEXT:    v_mad_u32_u24 v0, v0, 5, s4

llvm/test/CodeGen/AMDGPU/atomic_optimizations_raw_buffer.ll

 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX7LESS %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=tonga -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,DPPCOMB %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,DPPCOMB %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=-wavefrontsize32,+wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64 %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN32,GFX8MORE,GFX8MORE32 %s
 
 declare i32 @llvm.amdgcn.workitem.id.x()
 declare i32 @llvm.amdgcn.raw.buffer.atomic.add(i32, <4 x i32>, i32, i32, i32)
 declare i32 @llvm.amdgcn.raw.buffer.atomic.sub(i32, <4 x i32>, i32, i32, i32)
 
 ; Show what the atomic optimization pass will do for raw buffers.
 
 ; GCN-LABEL: add_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: buffer_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_constant(i32 addrspace(1)* %out, <4 x i32> %inout) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.add(i32 5, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: buffer_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_uniform(i32 addrspace(1)* %out, <4 x i32> %inout, i32 %additive) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.add(i32 %additive, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
entry:
   %lane = call i32 @llvm.amdgcn.workitem.id.x()
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.add(i32 1, <4 x i32> %inout, i32 %lane, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: buffer_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_constant(i32 addrspace(1)* %out, <4 x i32> %inout) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.sub(i32 5, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: buffer_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_uniform(i32 addrspace(1)* %out, <4 x i32> %inout, i32 %subitive) {
 entry:
   %old = call i32 @llvm.amdgcn.raw.buffer.atomic.sub(i32 %subitive, <4 x i32> %inout, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void

llvm/test/CodeGen/AMDGPU/atomic_optimizations_struct_buffer.ll

 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX7LESS %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=tonga -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,DPPCOMB %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64,DPPCOMB %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=-wavefrontsize32,+wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN64,GFX8MORE,GFX8MORE64 %s
 ; RUN: llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 -mattr=-flat-for-global -amdgpu-atomic-optimizations=true -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GCN32,GFX8MORE,GFX8MORE32 %s
 
 declare i32 @llvm.amdgcn.workitem.id.x()
 declare i32 @llvm.amdgcn.struct.buffer.atomic.add(i32, <4 x i32>, i32, i32, i32, i32)
 declare i32 @llvm.amdgcn.struct.buffer.atomic.sub(i32, <4 x i32>, i32, i32, i32, i32)
 
 ; Show what the atomic optimization pass will do for struct buffers.
 
 ; GCN-LABEL: add_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: buffer_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_constant(i32 addrspace(1)* %out, <4 x i32> %inout) {
 entry:
   %old = call i32 @llvm.amdgcn.struct.buffer.atomic.add(i32 5, <4 x i32> %inout, i32 0, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: add_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: buffer_atomic_add v[[value]]
 define amdgpu_kernel void @add_i32_uniform(i32 addrspace(1)* %out, <4 x i32> %inout, i32 %additive) {
 entry:
   %old = call i32 @llvm.amdgcn.struct.buffer.atomic.add(i32 %additive, <4 x i32> %inout, i32 0, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
entry:
   %lane = call i32 @llvm.amdgcn.workitem.id.x()
   %old = call i32 @llvm.amdgcn.struct.buffer.atomic.add(i32 1, <4 x i32> %inout, i32 0, i32 %lane, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_constant:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: v_mul_u32_u24{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[popcount]], 5
 ; GCN: buffer_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_constant(i32 addrspace(1)* %out, <4 x i32> %inout) {
 entry:
   %old = call i32 @llvm.amdgcn.struct.buffer.atomic.sub(i32 5, <4 x i32> %inout, i32 0, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void
 }
 
 ; GCN-LABEL: sub_i32_uniform:
 ; GCN: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt:[0-9]+]], exec_lo, 0
-; GCN64: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
+; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], exec_lo
+; GCN64-DAG: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], exec
+; GCN64-DAG: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt]], exec_hi, v[[mbcnt]]
 ; GCN: v_cmp_eq_u32{{(_e[0-9]+)?}} vcc{{(_lo)?}}, 0, v[[mbcnt]]
 ; GCN: s_and_saveexec_b{{32|64}} s[[exec:\[?[0-9:]+\]?]], vcc
-; GCN32: s_bcnt1_i32_b32 s[[popcount:[0-9]+]], s
-; GCN64: s_bcnt1_i32_b64 s[[popcount:[0-9]+]], s
 ; GCN: s_mul_i32 s[[scalar_value:[0-9]+]], s{{[0-9]+}}, s[[popcount]]
 ; GCN: v_mov_b32{{(_e[0-9]+)?}} v[[value:[0-9]+]], s[[scalar_value]]
 ; GCN: buffer_atomic_sub v[[value]]
 define amdgpu_kernel void @sub_i32_uniform(i32 addrspace(1)* %out, <4 x i32> %inout, i32 %subitive) {
 entry:
   %old = call i32 @llvm.amdgcn.struct.buffer.atomic.sub(i32 %subitive, <4 x i32> %inout, i32 0, i32 0, i32 0, i32 0)
   store i32 %old, i32 addrspace(1)* %out
   ret void