# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 6 # RUN: llc -mtriple=amdgcn -mcpu=gfx1250 -run-pass=machine-scheduler -amdgpu-sched-strategy=coexec -verify-misched %s -o - | FileCheck -check-prefix=DEFAULT %s --- | define void @tensor_load_tensor_load() #0 { ret void } define void @tensor_wait_tensor_load() #0 { ret void } define void @tensor_wait_tensor_wait() #0 { ret void } define void @tensor_load_tensor_wait() #0 { ret void } define void @async_load_async_load() #0 { ret void } define void @async_wait_async_load() #0 { ret void } define void @async_wait_async_wait() #0 { ret void } define void @async_load_async_wait() #0 { ret void } attributes #0 = { "amdgpu-waves-per-eu"="1,1" } ... --- name: tensor_load_tensor_load tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: tensor_load_tensor_load ; DEFAULT: [[DEF:%[0-9]+]]:sgpr_128 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF1:%[0-9]+]]:sgpr_256 = IMPLICIT_DEF ; DEFAULT-NEXT: TENSOR_LOAD_TO_LDS_d2 [[DEF]], [[DEF1]], 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt ; DEFAULT-NEXT: TENSOR_LOAD_TO_LDS_d2 [[DEF]], [[DEF1]], 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: S_ENDPGM 0 %0:sgpr_128 = IMPLICIT_DEF %1:sgpr_256 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF TENSOR_LOAD_TO_LDS_d2 %0:sgpr_128, %1:sgpr_256, 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt TENSOR_LOAD_TO_LDS_d2 %0:sgpr_128, %1:sgpr_256, 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec S_ENDPGM 0 ... --- name: tensor_wait_tensor_load tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: tensor_wait_tensor_load ; DEFAULT: S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt ; DEFAULT-NEXT: [[DEF:%[0-9]+]]:sgpr_128 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF1:%[0-9]+]]:sgpr_256 = IMPLICIT_DEF ; DEFAULT-NEXT: TENSOR_LOAD_TO_LDS_d2 [[DEF]], [[DEF1]], 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: S_ENDPGM 0 %0:sgpr_128 = IMPLICIT_DEF %1:sgpr_256 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt TENSOR_LOAD_TO_LDS_d2 %0:sgpr_128, %1:sgpr_256, 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec S_ENDPGM 0 ... --- name: tensor_wait_tensor_wait tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: tensor_wait_tensor_wait ; DEFAULT: S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt ; DEFAULT-NEXT: S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt ; DEFAULT-NEXT: dead [[DEF:%[0-9]+]]:sgpr_128 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[DEF1:%[0-9]+]]:sgpr_256 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: S_ENDPGM 0 %0:sgpr_128 = IMPLICIT_DEF %1:sgpr_256 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec S_ENDPGM 0 ... --- name: tensor_load_tensor_wait tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: tensor_load_tensor_wait ; DEFAULT: [[DEF:%[0-9]+]]:sgpr_128 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF1:%[0-9]+]]:sgpr_256 = IMPLICIT_DEF ; DEFAULT-NEXT: TENSOR_LOAD_TO_LDS_d2 [[DEF]], [[DEF1]], 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt ; DEFAULT-NEXT: S_ENDPGM 0 %0:sgpr_128 = IMPLICIT_DEF %1:sgpr_256 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF TENSOR_LOAD_TO_LDS_d2 %0:sgpr_128, %1:sgpr_256, 0, 0, implicit-def dead $tensorcnt, implicit $exec, implicit $tensorcnt S_WAIT_TENSORCNT 3, implicit-def dead $tensorcnt, implicit $tensorcnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec S_ENDPGM 0 ... --- name: async_load_async_load tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: async_load_async_load ; DEFAULT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF1:%[0-9]+]]:sreg_64 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF4:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR [[DEF]], [[DEF1]], [[DEF4]], 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR [[DEF]], [[DEF1]], [[DEF4]], 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: SCHED_GROUP_BARRIER 32, 1, 0 ; DEFAULT-NEXT: SCHED_GROUP_BARRIER 2, 6, 0 ; DEFAULT-NEXT: S_ENDPGM 0 %0:vgpr_32 = IMPLICIT_DEF %1:sreg_64 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF %4:vgpr_32 = IMPLICIT_DEF GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR %0:vgpr_32, %1:sreg_64, %4:vgpr_32, 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR %0:vgpr_32, %1:sreg_64, %4:vgpr_32, 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec SCHED_GROUP_BARRIER 32, 1, 0 SCHED_GROUP_BARRIER 2, 6, 0 S_ENDPGM 0 ... --- name: async_wait_async_load tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: async_wait_async_load ; DEFAULT: S_WAIT_ASYNCCNT 2, implicit-def dead $asynccnt, implicit $asynccnt ; DEFAULT-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF1:%[0-9]+]]:sreg_64 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: [[DEF4:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR [[DEF]], [[DEF1]], [[DEF4]], 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt ; DEFAULT-NEXT: S_ENDPGM 0 %0:vgpr_32 = IMPLICIT_DEF %1:sreg_64 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF %4:vgpr_32 = IMPLICIT_DEF S_WAIT_ASYNCCNT 2, implicit-def dead $asynccnt, implicit $asynccnt GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR %0:vgpr_32, %1:sreg_64, %4:vgpr_32, 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec S_ENDPGM 0 ... --- name: async_wait_async_wait tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: async_wait_async_wait ; DEFAULT: S_WAIT_ASYNCCNT 2, implicit-def dead $asynccnt, implicit $asynccnt ; DEFAULT-NEXT: S_WAIT_ASYNCCNT 1, implicit-def dead $asynccnt, implicit $asynccnt ; DEFAULT-NEXT: dead [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[DEF1:%[0-9]+]]:sreg_64 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[DEF4:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: S_ENDPGM 0 %0:vgpr_32 = IMPLICIT_DEF %1:sreg_64 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF %4:vgpr_32 = IMPLICIT_DEF S_WAIT_ASYNCCNT 2, implicit-def dead $asynccnt, implicit $asynccnt S_WAIT_ASYNCCNT 1, implicit-def dead $asynccnt, implicit $asynccnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec S_ENDPGM 0 ... --- name: async_load_async_wait tracksRegLiveness: true body: | bb.0: ; DEFAULT-LABEL: name: async_load_async_wait ; DEFAULT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF1:%[0-9]+]]:sreg_64 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF2:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF3:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: [[DEF4:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF ; DEFAULT-NEXT: GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR [[DEF]], [[DEF1]], [[DEF4]], 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_2:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_3:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_4:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: dead [[V_ADD_U32_e32_5:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[DEF2]], [[DEF3]], implicit $exec ; DEFAULT-NEXT: SCHED_GROUP_BARRIER 32, 1, 0 ; DEFAULT-NEXT: SCHED_GROUP_BARRIER 2, 6, 0 ; DEFAULT-NEXT: S_WAIT_ASYNCCNT 2, implicit-def dead $asynccnt, implicit $asynccnt ; DEFAULT-NEXT: S_ENDPGM 0 %0:vgpr_32 = IMPLICIT_DEF %1:sreg_64 = IMPLICIT_DEF %2:vgpr_32 = IMPLICIT_DEF %3:vgpr_32 = IMPLICIT_DEF %4:vgpr_32 = IMPLICIT_DEF GLOBAL_LOAD_ASYNC_TO_LDS_B64_SADDR %0:vgpr_32, %1:sreg_64, %4:vgpr_32, 0, 0, implicit-def dead $asynccnt, implicit $exec, implicit $asynccnt S_WAIT_ASYNCCNT 2, implicit-def dead $asynccnt, implicit $asynccnt %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %4:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %5:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %6:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %7:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec %8:vgpr_32 = V_ADD_U32_e32 %2, %3, implicit $exec SCHED_GROUP_BARRIER 32, 1, 0 SCHED_GROUP_BARRIER 2, 6, 0 S_ENDPGM 0 ...