samvv/bolt
1
0
Fork 0
Code Issues 24 Pull requests Projects Releases Packages Wiki Activity Actions
bolt/deps/llvm-18.1.8/llvm/test/CodeGen/AMDGPU/GlobalISel/irtranslator-call-implicit-args.ll
Sam Vervaeck 174b6f4d4e
Embed LLVM 18.1.8
2025-02-14 19:21:04 +01:00

1231 lines
82 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
; RUN: llc -global-isel -stop-after=irtranslator -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -verify-machineinstrs -o - %s | FileCheck -enable-var-scope -check-prefix=GFX900 %s
; RUN: llc -global-isel -stop-after=irtranslator -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx908 -verify-machineinstrs -o - %s | FileCheck -enable-var-scope -check-prefix=GFX908 %s
; Workitem IDs are passed to the kernel differently for gfx908
declare hidden void @external_void_func_void() #0
declare hidden void @external_void_func_i32(i32) #0
declare hidden void @external_void_func_v32i32(<32 x i32>) #0
define amdgpu_kernel void @test_call_external_void_func_i32([17 x i8]) #0 {
; GFX900-LABEL: name: test_call_external_void_func_i32
; GFX900: bb.1 (%ir-block.1):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr1, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX900-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX900-NEXT: [[COPY2:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY7:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: [[INT:%[0-9]+]]:_(p4) = G_INTRINSIC intrinsic(@llvm.amdgcn.kernarg.segment.ptr)
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(p4) = COPY [[COPY8]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 20
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY10]], [[C1]](s64)
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s64) = COPY [[COPY6]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY2]](s32)
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY16]], [[C2]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY15]], [[SHL]]
; GFX900-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX900-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[COPY17]], [[C3]](s32)
; GFX900-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[OR]], [[SHL1]]
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY18:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY18]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY9]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY11]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY14]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR1]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_call_external_void_func_i32
; GFX908: bb.1 (%ir-block.1):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr1, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX908-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX908-NEXT: [[COPY2:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY7:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: [[INT:%[0-9]+]]:_(p4) = G_INTRINSIC intrinsic(@llvm.amdgcn.kernarg.segment.ptr)
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(p4) = COPY [[COPY8]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 20
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY10]], [[C1]](s64)
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s64) = COPY [[COPY6]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY2]](s32)
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY16]], [[C2]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY15]], [[SHL]]
; GFX908-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX908-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[COPY17]], [[C3]](s32)
; GFX908-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[OR]], [[SHL1]]
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY18:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY18]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY9]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY11]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY14]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR1]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
define void @test_func_call_external_void_func_i32() #0 {
; GFX900-LABEL: name: test_func_call_external_void_func_i32
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $sgpr15, $vgpr31, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9, $sgpr10_sgpr11
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr31
; GFX900-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr15
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr10_sgpr11
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY7:%[0-9]+]]:sgpr_64(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[COPY8:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 99
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY8]]
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX900-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY18:%[0-9]+]]:_(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY18]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY9]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[COPY10]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[COPY11]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY12]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY14]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY15]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[COPY16]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[COPY17]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: SI_RETURN
;
; GFX908-LABEL: name: test_func_call_external_void_func_i32
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $sgpr15, $vgpr31, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9, $sgpr10_sgpr11
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr31
; GFX908-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr15
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr10_sgpr11
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY7:%[0-9]+]]:sgpr_64(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[COPY8:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 99
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY8]]
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX908-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY18:%[0-9]+]]:_(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY18]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY9]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[COPY10]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[COPY11]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY12]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY14]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY15]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[COPY16]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[COPY17]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: SI_RETURN
call void @external_void_func_i32(i32 99)
ret void
}
; Explicit argument is split between registers ad the stack due to v31
; being used for workitem IDs.
define amdgpu_kernel void @test_call_external_void_func_v32i32([17 x i8]) #0 {
; GFX900-LABEL: name: test_call_external_void_func_v32i32
; GFX900: bb.1 (%ir-block.1):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr1, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX900-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX900-NEXT: [[COPY2:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY7:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
; GFX900-NEXT: [[INT:%[0-9]+]]:_(p4) = G_INTRINSIC intrinsic(@llvm.amdgcn.kernarg.segment.ptr)
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_v32i32
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(p4) = COPY [[COPY8]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 20
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY10]], [[C1]](s64)
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s64) = COPY [[COPY6]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY2]](s32)
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY16]], [[C2]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY15]], [[SHL]]
; GFX900-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX900-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[COPY17]], [[C3]](s32)
; GFX900-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[OR]], [[SHL1]]
; GFX900-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32), [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32), [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32), [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32), [[UV12:%[0-9]+]]:_(s32), [[UV13:%[0-9]+]]:_(s32), [[UV14:%[0-9]+]]:_(s32), [[UV15:%[0-9]+]]:_(s32), [[UV16:%[0-9]+]]:_(s32), [[UV17:%[0-9]+]]:_(s32), [[UV18:%[0-9]+]]:_(s32), [[UV19:%[0-9]+]]:_(s32), [[UV20:%[0-9]+]]:_(s32), [[UV21:%[0-9]+]]:_(s32), [[UV22:%[0-9]+]]:_(s32), [[UV23:%[0-9]+]]:_(s32), [[UV24:%[0-9]+]]:_(s32), [[UV25:%[0-9]+]]:_(s32), [[UV26:%[0-9]+]]:_(s32), [[UV27:%[0-9]+]]:_(s32), [[UV28:%[0-9]+]]:_(s32), [[UV29:%[0-9]+]]:_(s32), [[UV30:%[0-9]+]]:_(s32), [[UV31:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[BUILD_VECTOR]](<32 x s32>)
; GFX900-NEXT: [[AMDGPU_WAVE_ADDRESS:%[0-9]+]]:_(p5) = G_AMDGPU_WAVE_ADDRESS $sp_reg
; GFX900-NEXT: [[C4:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p5) = G_PTR_ADD [[AMDGPU_WAVE_ADDRESS]], [[C4]](s32)
; GFX900-NEXT: G_STORE [[UV31]](s32), [[PTR_ADD1]](p5) :: (store (s32) into stack, align 16, addrspace 5)
; GFX900-NEXT: $vgpr0 = COPY [[UV]](s32)
; GFX900-NEXT: $vgpr1 = COPY [[UV1]](s32)
; GFX900-NEXT: $vgpr2 = COPY [[UV2]](s32)
; GFX900-NEXT: $vgpr3 = COPY [[UV3]](s32)
; GFX900-NEXT: $vgpr4 = COPY [[UV4]](s32)
; GFX900-NEXT: $vgpr5 = COPY [[UV5]](s32)
; GFX900-NEXT: $vgpr6 = COPY [[UV6]](s32)
; GFX900-NEXT: $vgpr7 = COPY [[UV7]](s32)
; GFX900-NEXT: $vgpr8 = COPY [[UV8]](s32)
; GFX900-NEXT: $vgpr9 = COPY [[UV9]](s32)
; GFX900-NEXT: $vgpr10 = COPY [[UV10]](s32)
; GFX900-NEXT: $vgpr11 = COPY [[UV11]](s32)
; GFX900-NEXT: $vgpr12 = COPY [[UV12]](s32)
; GFX900-NEXT: $vgpr13 = COPY [[UV13]](s32)
; GFX900-NEXT: $vgpr14 = COPY [[UV14]](s32)
; GFX900-NEXT: $vgpr15 = COPY [[UV15]](s32)
; GFX900-NEXT: $vgpr16 = COPY [[UV16]](s32)
; GFX900-NEXT: $vgpr17 = COPY [[UV17]](s32)
; GFX900-NEXT: $vgpr18 = COPY [[UV18]](s32)
; GFX900-NEXT: $vgpr19 = COPY [[UV19]](s32)
; GFX900-NEXT: $vgpr20 = COPY [[UV20]](s32)
; GFX900-NEXT: $vgpr21 = COPY [[UV21]](s32)
; GFX900-NEXT: $vgpr22 = COPY [[UV22]](s32)
; GFX900-NEXT: $vgpr23 = COPY [[UV23]](s32)
; GFX900-NEXT: $vgpr24 = COPY [[UV24]](s32)
; GFX900-NEXT: $vgpr25 = COPY [[UV25]](s32)
; GFX900-NEXT: $vgpr26 = COPY [[UV26]](s32)
; GFX900-NEXT: $vgpr27 = COPY [[UV27]](s32)
; GFX900-NEXT: $vgpr28 = COPY [[UV28]](s32)
; GFX900-NEXT: $vgpr29 = COPY [[UV29]](s32)
; GFX900-NEXT: $vgpr30 = COPY [[UV30]](s32)
; GFX900-NEXT: [[COPY18:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY18]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY9]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY11]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY14]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR1]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_v32i32, csr_amdgpu, implicit $vgpr0, implicit $vgpr1, implicit $vgpr2, implicit $vgpr3, implicit $vgpr4, implicit $vgpr5, implicit $vgpr6, implicit $vgpr7, implicit $vgpr8, implicit $vgpr9, implicit $vgpr10, implicit $vgpr11, implicit $vgpr12, implicit $vgpr13, implicit $vgpr14, implicit $vgpr15, implicit $vgpr16, implicit $vgpr17, implicit $vgpr18, implicit $vgpr19, implicit $vgpr20, implicit $vgpr21, implicit $vgpr22, implicit $vgpr23, implicit $vgpr24, implicit $vgpr25, implicit $vgpr26, implicit $vgpr27, implicit $vgpr28, implicit $vgpr29, implicit $vgpr30, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 4, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_call_external_void_func_v32i32
; GFX908: bb.1 (%ir-block.1):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr1, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX908-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX908-NEXT: [[COPY2:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY7:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
; GFX908-NEXT: [[INT:%[0-9]+]]:_(p4) = G_INTRINSIC intrinsic(@llvm.amdgcn.kernarg.segment.ptr)
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_v32i32
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(p4) = COPY [[COPY8]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 20
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY10]], [[C1]](s64)
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s64) = COPY [[COPY6]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY2]](s32)
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY16]], [[C2]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY15]], [[SHL]]
; GFX908-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX908-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[COPY17]], [[C3]](s32)
; GFX908-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[OR]], [[SHL1]]
; GFX908-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32), [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32), [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32), [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32), [[UV12:%[0-9]+]]:_(s32), [[UV13:%[0-9]+]]:_(s32), [[UV14:%[0-9]+]]:_(s32), [[UV15:%[0-9]+]]:_(s32), [[UV16:%[0-9]+]]:_(s32), [[UV17:%[0-9]+]]:_(s32), [[UV18:%[0-9]+]]:_(s32), [[UV19:%[0-9]+]]:_(s32), [[UV20:%[0-9]+]]:_(s32), [[UV21:%[0-9]+]]:_(s32), [[UV22:%[0-9]+]]:_(s32), [[UV23:%[0-9]+]]:_(s32), [[UV24:%[0-9]+]]:_(s32), [[UV25:%[0-9]+]]:_(s32), [[UV26:%[0-9]+]]:_(s32), [[UV27:%[0-9]+]]:_(s32), [[UV28:%[0-9]+]]:_(s32), [[UV29:%[0-9]+]]:_(s32), [[UV30:%[0-9]+]]:_(s32), [[UV31:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[BUILD_VECTOR]](<32 x s32>)
; GFX908-NEXT: [[AMDGPU_WAVE_ADDRESS:%[0-9]+]]:_(p5) = G_AMDGPU_WAVE_ADDRESS $sp_reg
; GFX908-NEXT: [[C4:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p5) = G_PTR_ADD [[AMDGPU_WAVE_ADDRESS]], [[C4]](s32)
; GFX908-NEXT: G_STORE [[UV31]](s32), [[PTR_ADD1]](p5) :: (store (s32) into stack, align 16, addrspace 5)
; GFX908-NEXT: $vgpr0 = COPY [[UV]](s32)
; GFX908-NEXT: $vgpr1 = COPY [[UV1]](s32)
; GFX908-NEXT: $vgpr2 = COPY [[UV2]](s32)
; GFX908-NEXT: $vgpr3 = COPY [[UV3]](s32)
; GFX908-NEXT: $vgpr4 = COPY [[UV4]](s32)
; GFX908-NEXT: $vgpr5 = COPY [[UV5]](s32)
; GFX908-NEXT: $vgpr6 = COPY [[UV6]](s32)
; GFX908-NEXT: $vgpr7 = COPY [[UV7]](s32)
; GFX908-NEXT: $vgpr8 = COPY [[UV8]](s32)
; GFX908-NEXT: $vgpr9 = COPY [[UV9]](s32)
; GFX908-NEXT: $vgpr10 = COPY [[UV10]](s32)
; GFX908-NEXT: $vgpr11 = COPY [[UV11]](s32)
; GFX908-NEXT: $vgpr12 = COPY [[UV12]](s32)
; GFX908-NEXT: $vgpr13 = COPY [[UV13]](s32)
; GFX908-NEXT: $vgpr14 = COPY [[UV14]](s32)
; GFX908-NEXT: $vgpr15 = COPY [[UV15]](s32)
; GFX908-NEXT: $vgpr16 = COPY [[UV16]](s32)
; GFX908-NEXT: $vgpr17 = COPY [[UV17]](s32)
; GFX908-NEXT: $vgpr18 = COPY [[UV18]](s32)
; GFX908-NEXT: $vgpr19 = COPY [[UV19]](s32)
; GFX908-NEXT: $vgpr20 = COPY [[UV20]](s32)
; GFX908-NEXT: $vgpr21 = COPY [[UV21]](s32)
; GFX908-NEXT: $vgpr22 = COPY [[UV22]](s32)
; GFX908-NEXT: $vgpr23 = COPY [[UV23]](s32)
; GFX908-NEXT: $vgpr24 = COPY [[UV24]](s32)
; GFX908-NEXT: $vgpr25 = COPY [[UV25]](s32)
; GFX908-NEXT: $vgpr26 = COPY [[UV26]](s32)
; GFX908-NEXT: $vgpr27 = COPY [[UV27]](s32)
; GFX908-NEXT: $vgpr28 = COPY [[UV28]](s32)
; GFX908-NEXT: $vgpr29 = COPY [[UV29]](s32)
; GFX908-NEXT: $vgpr30 = COPY [[UV30]](s32)
; GFX908-NEXT: [[COPY18:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY18]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY9]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY11]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY14]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR1]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_v32i32, csr_amdgpu, implicit $vgpr0, implicit $vgpr1, implicit $vgpr2, implicit $vgpr3, implicit $vgpr4, implicit $vgpr5, implicit $vgpr6, implicit $vgpr7, implicit $vgpr8, implicit $vgpr9, implicit $vgpr10, implicit $vgpr11, implicit $vgpr12, implicit $vgpr13, implicit $vgpr14, implicit $vgpr15, implicit $vgpr16, implicit $vgpr17, implicit $vgpr18, implicit $vgpr19, implicit $vgpr20, implicit $vgpr21, implicit $vgpr22, implicit $vgpr23, implicit $vgpr24, implicit $vgpr25, implicit $vgpr26, implicit $vgpr27, implicit $vgpr28, implicit $vgpr29, implicit $vgpr30, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 4, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_v32i32(<32 x i32> zeroinitializer)
ret void
}
define void @test_func_call_external_void_func_v32i32([17 x i8]) #0 {
; GFX900-LABEL: name: test_func_call_external_void_func_v32i32
; GFX900: bb.1 (%ir-block.1):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $sgpr15, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8, $vgpr9, $vgpr10, $vgpr11, $vgpr12, $vgpr13, $vgpr14, $vgpr15, $vgpr16, $vgpr31, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9, $sgpr10_sgpr11
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr31
; GFX900-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr15
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr10_sgpr11
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY7:%[0-9]+]]:sgpr_64(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[COPY8:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX900-NEXT: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[COPY9]](s32)
; GFX900-NEXT: [[TRUNC1:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC]](s16)
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX900-NEXT: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[COPY10]](s32)
; GFX900-NEXT: [[TRUNC3:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC2]](s16)
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY $vgpr2
; GFX900-NEXT: [[TRUNC4:%[0-9]+]]:_(s16) = G_TRUNC [[COPY11]](s32)
; GFX900-NEXT: [[TRUNC5:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC4]](s16)
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY $vgpr3
; GFX900-NEXT: [[TRUNC6:%[0-9]+]]:_(s16) = G_TRUNC [[COPY12]](s32)
; GFX900-NEXT: [[TRUNC7:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC6]](s16)
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY $vgpr4
; GFX900-NEXT: [[TRUNC8:%[0-9]+]]:_(s16) = G_TRUNC [[COPY13]](s32)
; GFX900-NEXT: [[TRUNC9:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC8]](s16)
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY $vgpr5
; GFX900-NEXT: [[TRUNC10:%[0-9]+]]:_(s16) = G_TRUNC [[COPY14]](s32)
; GFX900-NEXT: [[TRUNC11:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC10]](s16)
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY $vgpr6
; GFX900-NEXT: [[TRUNC12:%[0-9]+]]:_(s16) = G_TRUNC [[COPY15]](s32)
; GFX900-NEXT: [[TRUNC13:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC12]](s16)
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY $vgpr7
; GFX900-NEXT: [[TRUNC14:%[0-9]+]]:_(s16) = G_TRUNC [[COPY16]](s32)
; GFX900-NEXT: [[TRUNC15:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC14]](s16)
; GFX900-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY $vgpr8
; GFX900-NEXT: [[TRUNC16:%[0-9]+]]:_(s16) = G_TRUNC [[COPY17]](s32)
; GFX900-NEXT: [[TRUNC17:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC16]](s16)
; GFX900-NEXT: [[COPY18:%[0-9]+]]:_(s32) = COPY $vgpr9
; GFX900-NEXT: [[TRUNC18:%[0-9]+]]:_(s16) = G_TRUNC [[COPY18]](s32)
; GFX900-NEXT: [[TRUNC19:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC18]](s16)
; GFX900-NEXT: [[COPY19:%[0-9]+]]:_(s32) = COPY $vgpr10
; GFX900-NEXT: [[TRUNC20:%[0-9]+]]:_(s16) = G_TRUNC [[COPY19]](s32)
; GFX900-NEXT: [[TRUNC21:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC20]](s16)
; GFX900-NEXT: [[COPY20:%[0-9]+]]:_(s32) = COPY $vgpr11
; GFX900-NEXT: [[TRUNC22:%[0-9]+]]:_(s16) = G_TRUNC [[COPY20]](s32)
; GFX900-NEXT: [[TRUNC23:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC22]](s16)
; GFX900-NEXT: [[COPY21:%[0-9]+]]:_(s32) = COPY $vgpr12
; GFX900-NEXT: [[TRUNC24:%[0-9]+]]:_(s16) = G_TRUNC [[COPY21]](s32)
; GFX900-NEXT: [[TRUNC25:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC24]](s16)
; GFX900-NEXT: [[COPY22:%[0-9]+]]:_(s32) = COPY $vgpr13
; GFX900-NEXT: [[TRUNC26:%[0-9]+]]:_(s16) = G_TRUNC [[COPY22]](s32)
; GFX900-NEXT: [[TRUNC27:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC26]](s16)
; GFX900-NEXT: [[COPY23:%[0-9]+]]:_(s32) = COPY $vgpr14
; GFX900-NEXT: [[TRUNC28:%[0-9]+]]:_(s16) = G_TRUNC [[COPY23]](s32)
; GFX900-NEXT: [[TRUNC29:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC28]](s16)
; GFX900-NEXT: [[COPY24:%[0-9]+]]:_(s32) = COPY $vgpr15
; GFX900-NEXT: [[TRUNC30:%[0-9]+]]:_(s16) = G_TRUNC [[COPY24]](s32)
; GFX900-NEXT: [[TRUNC31:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC30]](s16)
; GFX900-NEXT: [[COPY25:%[0-9]+]]:_(s32) = COPY $vgpr16
; GFX900-NEXT: [[TRUNC32:%[0-9]+]]:_(s16) = G_TRUNC [[COPY25]](s32)
; GFX900-NEXT: [[TRUNC33:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC32]](s16)
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_v32i32
; GFX900-NEXT: [[COPY26:%[0-9]+]]:_(p4) = COPY [[COPY8]]
; GFX900-NEXT: [[COPY27:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX900-NEXT: [[COPY28:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX900-NEXT: [[COPY29:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY30:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY31:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY32:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[COPY33:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX900-NEXT: [[COPY34:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32), [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32), [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32), [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32), [[UV12:%[0-9]+]]:_(s32), [[UV13:%[0-9]+]]:_(s32), [[UV14:%[0-9]+]]:_(s32), [[UV15:%[0-9]+]]:_(s32), [[UV16:%[0-9]+]]:_(s32), [[UV17:%[0-9]+]]:_(s32), [[UV18:%[0-9]+]]:_(s32), [[UV19:%[0-9]+]]:_(s32), [[UV20:%[0-9]+]]:_(s32), [[UV21:%[0-9]+]]:_(s32), [[UV22:%[0-9]+]]:_(s32), [[UV23:%[0-9]+]]:_(s32), [[UV24:%[0-9]+]]:_(s32), [[UV25:%[0-9]+]]:_(s32), [[UV26:%[0-9]+]]:_(s32), [[UV27:%[0-9]+]]:_(s32), [[UV28:%[0-9]+]]:_(s32), [[UV29:%[0-9]+]]:_(s32), [[UV30:%[0-9]+]]:_(s32), [[UV31:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[BUILD_VECTOR]](<32 x s32>)
; GFX900-NEXT: [[AMDGPU_WAVE_ADDRESS:%[0-9]+]]:_(p5) = G_AMDGPU_WAVE_ADDRESS $sgpr32
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p5) = G_PTR_ADD [[AMDGPU_WAVE_ADDRESS]], [[C1]](s32)
; GFX900-NEXT: G_STORE [[UV31]](s32), [[PTR_ADD]](p5) :: (store (s32) into stack, align 16, addrspace 5)
; GFX900-NEXT: $vgpr0 = COPY [[UV]](s32)
; GFX900-NEXT: $vgpr1 = COPY [[UV1]](s32)
; GFX900-NEXT: $vgpr2 = COPY [[UV2]](s32)
; GFX900-NEXT: $vgpr3 = COPY [[UV3]](s32)
; GFX900-NEXT: $vgpr4 = COPY [[UV4]](s32)
; GFX900-NEXT: $vgpr5 = COPY [[UV5]](s32)
; GFX900-NEXT: $vgpr6 = COPY [[UV6]](s32)
; GFX900-NEXT: $vgpr7 = COPY [[UV7]](s32)
; GFX900-NEXT: $vgpr8 = COPY [[UV8]](s32)
; GFX900-NEXT: $vgpr9 = COPY [[UV9]](s32)
; GFX900-NEXT: $vgpr10 = COPY [[UV10]](s32)
; GFX900-NEXT: $vgpr11 = COPY [[UV11]](s32)
; GFX900-NEXT: $vgpr12 = COPY [[UV12]](s32)
; GFX900-NEXT: $vgpr13 = COPY [[UV13]](s32)
; GFX900-NEXT: $vgpr14 = COPY [[UV14]](s32)
; GFX900-NEXT: $vgpr15 = COPY [[UV15]](s32)
; GFX900-NEXT: $vgpr16 = COPY [[UV16]](s32)
; GFX900-NEXT: $vgpr17 = COPY [[UV17]](s32)
; GFX900-NEXT: $vgpr18 = COPY [[UV18]](s32)
; GFX900-NEXT: $vgpr19 = COPY [[UV19]](s32)
; GFX900-NEXT: $vgpr20 = COPY [[UV20]](s32)
; GFX900-NEXT: $vgpr21 = COPY [[UV21]](s32)
; GFX900-NEXT: $vgpr22 = COPY [[UV22]](s32)
; GFX900-NEXT: $vgpr23 = COPY [[UV23]](s32)
; GFX900-NEXT: $vgpr24 = COPY [[UV24]](s32)
; GFX900-NEXT: $vgpr25 = COPY [[UV25]](s32)
; GFX900-NEXT: $vgpr26 = COPY [[UV26]](s32)
; GFX900-NEXT: $vgpr27 = COPY [[UV27]](s32)
; GFX900-NEXT: $vgpr28 = COPY [[UV28]](s32)
; GFX900-NEXT: $vgpr29 = COPY [[UV29]](s32)
; GFX900-NEXT: $vgpr30 = COPY [[UV30]](s32)
; GFX900-NEXT: [[COPY35:%[0-9]+]]:_(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY35]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY26]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[COPY27]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[COPY28]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY29]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY30]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY31]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY32]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[COPY33]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[COPY34]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_v32i32, csr_amdgpu, implicit $vgpr0, implicit $vgpr1, implicit $vgpr2, implicit $vgpr3, implicit $vgpr4, implicit $vgpr5, implicit $vgpr6, implicit $vgpr7, implicit $vgpr8, implicit $vgpr9, implicit $vgpr10, implicit $vgpr11, implicit $vgpr12, implicit $vgpr13, implicit $vgpr14, implicit $vgpr15, implicit $vgpr16, implicit $vgpr17, implicit $vgpr18, implicit $vgpr19, implicit $vgpr20, implicit $vgpr21, implicit $vgpr22, implicit $vgpr23, implicit $vgpr24, implicit $vgpr25, implicit $vgpr26, implicit $vgpr27, implicit $vgpr28, implicit $vgpr29, implicit $vgpr30, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 4, implicit-def $scc
; GFX900-NEXT: SI_RETURN
;
; GFX908-LABEL: name: test_func_call_external_void_func_v32i32
; GFX908: bb.1 (%ir-block.1):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $sgpr15, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8, $vgpr9, $vgpr10, $vgpr11, $vgpr12, $vgpr13, $vgpr14, $vgpr15, $vgpr16, $vgpr31, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9, $sgpr10_sgpr11
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr31
; GFX908-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr15
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr10_sgpr11
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY7:%[0-9]+]]:sgpr_64(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[COPY8:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(s32) = COPY $vgpr0
; GFX908-NEXT: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[COPY9]](s32)
; GFX908-NEXT: [[TRUNC1:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC]](s16)
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY $vgpr1
; GFX908-NEXT: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[COPY10]](s32)
; GFX908-NEXT: [[TRUNC3:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC2]](s16)
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY $vgpr2
; GFX908-NEXT: [[TRUNC4:%[0-9]+]]:_(s16) = G_TRUNC [[COPY11]](s32)
; GFX908-NEXT: [[TRUNC5:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC4]](s16)
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY $vgpr3
; GFX908-NEXT: [[TRUNC6:%[0-9]+]]:_(s16) = G_TRUNC [[COPY12]](s32)
; GFX908-NEXT: [[TRUNC7:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC6]](s16)
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY $vgpr4
; GFX908-NEXT: [[TRUNC8:%[0-9]+]]:_(s16) = G_TRUNC [[COPY13]](s32)
; GFX908-NEXT: [[TRUNC9:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC8]](s16)
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY $vgpr5
; GFX908-NEXT: [[TRUNC10:%[0-9]+]]:_(s16) = G_TRUNC [[COPY14]](s32)
; GFX908-NEXT: [[TRUNC11:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC10]](s16)
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY $vgpr6
; GFX908-NEXT: [[TRUNC12:%[0-9]+]]:_(s16) = G_TRUNC [[COPY15]](s32)
; GFX908-NEXT: [[TRUNC13:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC12]](s16)
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(s32) = COPY $vgpr7
; GFX908-NEXT: [[TRUNC14:%[0-9]+]]:_(s16) = G_TRUNC [[COPY16]](s32)
; GFX908-NEXT: [[TRUNC15:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC14]](s16)
; GFX908-NEXT: [[COPY17:%[0-9]+]]:_(s32) = COPY $vgpr8
; GFX908-NEXT: [[TRUNC16:%[0-9]+]]:_(s16) = G_TRUNC [[COPY17]](s32)
; GFX908-NEXT: [[TRUNC17:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC16]](s16)
; GFX908-NEXT: [[COPY18:%[0-9]+]]:_(s32) = COPY $vgpr9
; GFX908-NEXT: [[TRUNC18:%[0-9]+]]:_(s16) = G_TRUNC [[COPY18]](s32)
; GFX908-NEXT: [[TRUNC19:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC18]](s16)
; GFX908-NEXT: [[COPY19:%[0-9]+]]:_(s32) = COPY $vgpr10
; GFX908-NEXT: [[TRUNC20:%[0-9]+]]:_(s16) = G_TRUNC [[COPY19]](s32)
; GFX908-NEXT: [[TRUNC21:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC20]](s16)
; GFX908-NEXT: [[COPY20:%[0-9]+]]:_(s32) = COPY $vgpr11
; GFX908-NEXT: [[TRUNC22:%[0-9]+]]:_(s16) = G_TRUNC [[COPY20]](s32)
; GFX908-NEXT: [[TRUNC23:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC22]](s16)
; GFX908-NEXT: [[COPY21:%[0-9]+]]:_(s32) = COPY $vgpr12
; GFX908-NEXT: [[TRUNC24:%[0-9]+]]:_(s16) = G_TRUNC [[COPY21]](s32)
; GFX908-NEXT: [[TRUNC25:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC24]](s16)
; GFX908-NEXT: [[COPY22:%[0-9]+]]:_(s32) = COPY $vgpr13
; GFX908-NEXT: [[TRUNC26:%[0-9]+]]:_(s16) = G_TRUNC [[COPY22]](s32)
; GFX908-NEXT: [[TRUNC27:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC26]](s16)
; GFX908-NEXT: [[COPY23:%[0-9]+]]:_(s32) = COPY $vgpr14
; GFX908-NEXT: [[TRUNC28:%[0-9]+]]:_(s16) = G_TRUNC [[COPY23]](s32)
; GFX908-NEXT: [[TRUNC29:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC28]](s16)
; GFX908-NEXT: [[COPY24:%[0-9]+]]:_(s32) = COPY $vgpr15
; GFX908-NEXT: [[TRUNC30:%[0-9]+]]:_(s16) = G_TRUNC [[COPY24]](s32)
; GFX908-NEXT: [[TRUNC31:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC30]](s16)
; GFX908-NEXT: [[COPY25:%[0-9]+]]:_(s32) = COPY $vgpr16
; GFX908-NEXT: [[TRUNC32:%[0-9]+]]:_(s16) = G_TRUNC [[COPY25]](s32)
; GFX908-NEXT: [[TRUNC33:%[0-9]+]]:_(s8) = G_TRUNC [[TRUNC32]](s16)
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_v32i32
; GFX908-NEXT: [[COPY26:%[0-9]+]]:_(p4) = COPY [[COPY8]]
; GFX908-NEXT: [[COPY27:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX908-NEXT: [[COPY28:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX908-NEXT: [[COPY29:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY30:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY31:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY32:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[COPY33:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX908-NEXT: [[COPY34:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32), [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32), [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32), [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32), [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32), [[UV12:%[0-9]+]]:_(s32), [[UV13:%[0-9]+]]:_(s32), [[UV14:%[0-9]+]]:_(s32), [[UV15:%[0-9]+]]:_(s32), [[UV16:%[0-9]+]]:_(s32), [[UV17:%[0-9]+]]:_(s32), [[UV18:%[0-9]+]]:_(s32), [[UV19:%[0-9]+]]:_(s32), [[UV20:%[0-9]+]]:_(s32), [[UV21:%[0-9]+]]:_(s32), [[UV22:%[0-9]+]]:_(s32), [[UV23:%[0-9]+]]:_(s32), [[UV24:%[0-9]+]]:_(s32), [[UV25:%[0-9]+]]:_(s32), [[UV26:%[0-9]+]]:_(s32), [[UV27:%[0-9]+]]:_(s32), [[UV28:%[0-9]+]]:_(s32), [[UV29:%[0-9]+]]:_(s32), [[UV30:%[0-9]+]]:_(s32), [[UV31:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[BUILD_VECTOR]](<32 x s32>)
; GFX908-NEXT: [[AMDGPU_WAVE_ADDRESS:%[0-9]+]]:_(p5) = G_AMDGPU_WAVE_ADDRESS $sgpr32
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p5) = G_PTR_ADD [[AMDGPU_WAVE_ADDRESS]], [[C1]](s32)
; GFX908-NEXT: G_STORE [[UV31]](s32), [[PTR_ADD]](p5) :: (store (s32) into stack, align 16, addrspace 5)
; GFX908-NEXT: $vgpr0 = COPY [[UV]](s32)
; GFX908-NEXT: $vgpr1 = COPY [[UV1]](s32)
; GFX908-NEXT: $vgpr2 = COPY [[UV2]](s32)
; GFX908-NEXT: $vgpr3 = COPY [[UV3]](s32)
; GFX908-NEXT: $vgpr4 = COPY [[UV4]](s32)
; GFX908-NEXT: $vgpr5 = COPY [[UV5]](s32)
; GFX908-NEXT: $vgpr6 = COPY [[UV6]](s32)
; GFX908-NEXT: $vgpr7 = COPY [[UV7]](s32)
; GFX908-NEXT: $vgpr8 = COPY [[UV8]](s32)
; GFX908-NEXT: $vgpr9 = COPY [[UV9]](s32)
; GFX908-NEXT: $vgpr10 = COPY [[UV10]](s32)
; GFX908-NEXT: $vgpr11 = COPY [[UV11]](s32)
; GFX908-NEXT: $vgpr12 = COPY [[UV12]](s32)
; GFX908-NEXT: $vgpr13 = COPY [[UV13]](s32)
; GFX908-NEXT: $vgpr14 = COPY [[UV14]](s32)
; GFX908-NEXT: $vgpr15 = COPY [[UV15]](s32)
; GFX908-NEXT: $vgpr16 = COPY [[UV16]](s32)
; GFX908-NEXT: $vgpr17 = COPY [[UV17]](s32)
; GFX908-NEXT: $vgpr18 = COPY [[UV18]](s32)
; GFX908-NEXT: $vgpr19 = COPY [[UV19]](s32)
; GFX908-NEXT: $vgpr20 = COPY [[UV20]](s32)
; GFX908-NEXT: $vgpr21 = COPY [[UV21]](s32)
; GFX908-NEXT: $vgpr22 = COPY [[UV22]](s32)
; GFX908-NEXT: $vgpr23 = COPY [[UV23]](s32)
; GFX908-NEXT: $vgpr24 = COPY [[UV24]](s32)
; GFX908-NEXT: $vgpr25 = COPY [[UV25]](s32)
; GFX908-NEXT: $vgpr26 = COPY [[UV26]](s32)
; GFX908-NEXT: $vgpr27 = COPY [[UV27]](s32)
; GFX908-NEXT: $vgpr28 = COPY [[UV28]](s32)
; GFX908-NEXT: $vgpr29 = COPY [[UV29]](s32)
; GFX908-NEXT: $vgpr30 = COPY [[UV30]](s32)
; GFX908-NEXT: [[COPY35:%[0-9]+]]:_(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY35]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY26]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[COPY27]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[COPY28]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY29]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY30]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY31]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY32]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[COPY33]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[COPY34]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_v32i32, csr_amdgpu, implicit $vgpr0, implicit $vgpr1, implicit $vgpr2, implicit $vgpr3, implicit $vgpr4, implicit $vgpr5, implicit $vgpr6, implicit $vgpr7, implicit $vgpr8, implicit $vgpr9, implicit $vgpr10, implicit $vgpr11, implicit $vgpr12, implicit $vgpr13, implicit $vgpr14, implicit $vgpr15, implicit $vgpr16, implicit $vgpr17, implicit $vgpr18, implicit $vgpr19, implicit $vgpr20, implicit $vgpr21, implicit $vgpr22, implicit $vgpr23, implicit $vgpr24, implicit $vgpr25, implicit $vgpr26, implicit $vgpr27, implicit $vgpr28, implicit $vgpr29, implicit $vgpr30, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 4, implicit-def $scc
; GFX908-NEXT: SI_RETURN
call void @external_void_func_v32i32(<32 x i32> zeroinitializer)
ret void
}
define amdgpu_kernel void @test_only_workitem_id_x() #0 !reqd_work_group_size !0 {
; GFX900-LABEL: name: test_only_workitem_id_x
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX900-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY6:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY [[COPY5]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY8]], [[C1]](s64)
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(s64) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY14]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY7]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY9]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY10]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY11]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_only_workitem_id_x
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX908-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY6:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY [[COPY5]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY8]], [[C1]](s64)
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(s64) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY14]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY7]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY9]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY10]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY11]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
define amdgpu_kernel void @test_only_workitem_id_y() #0 !reqd_work_group_size !1 {
; GFX900-LABEL: name: test_only_workitem_id_y
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr1, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX900-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY6:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY [[COPY5]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY8]], [[C1]](s64)
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(s64) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY13]], [[C3]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C2]], [[SHL]]
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY14]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY7]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY9]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY10]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY11]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_only_workitem_id_y
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr1, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX908-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY6:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY [[COPY5]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY8]], [[C1]](s64)
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(s64) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY13]], [[C3]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C2]], [[SHL]]
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY14]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY7]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY9]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY10]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY11]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
define amdgpu_kernel void @test_only_workitem_id_z() #0 !reqd_work_group_size !2 {
; GFX900-LABEL: name: test_only_workitem_id_z
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX900-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY6:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY [[COPY5]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY8]], [[C1]](s64)
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(s64) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY13]], [[C3]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C2]], [[SHL]]
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY14]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY7]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY9]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY10]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY11]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_only_workitem_id_z
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX908-NEXT: [[COPY1:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY6:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY [[COPY5]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY8]], [[C1]](s64)
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(s64) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY1]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY13]], [[C3]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C2]], [[SHL]]
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY14]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY7]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY9]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY10]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY11]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
define amdgpu_kernel void @test_only_workitem_id_xy() #0 !reqd_work_group_size !3 {
; GFX900-LABEL: name: test_only_workitem_id_xy
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr1, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX900-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY9]], [[C1]](s64)
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY15]], [[C2]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY14]], [[SHL]]
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY16]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY8]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY10]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY11]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_only_workitem_id_xy
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr1, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX908-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY9]], [[C1]](s64)
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY15]], [[C2]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY14]], [[SHL]]
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY16]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY8]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY10]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY11]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
define amdgpu_kernel void @test_only_workitem_id_yz() #0 !reqd_work_group_size !4 {
; GFX900-LABEL: name: test_only_workitem_id_yz
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr1, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX900-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY9]], [[C1]](s64)
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX900-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY14]], [[C3]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C2]], [[SHL]]
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C4:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX900-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[COPY15]], [[C4]](s32)
; GFX900-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[OR]], [[SHL1]]
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY16]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY8]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY10]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY11]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR1]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_only_workitem_id_yz
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr1, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX908-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr1
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY9]], [[C1]](s64)
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX908-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 10
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY14]], [[C3]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C2]], [[SHL]]
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C4:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX908-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[COPY15]], [[C4]](s32)
; GFX908-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[OR]], [[SHL1]]
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY16]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY8]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY10]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY11]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR1]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
define amdgpu_kernel void @test_only_workitem_id_xz() #0 !reqd_work_group_size !5 {
; GFX900-LABEL: name: test_only_workitem_id_xz
; GFX900: bb.1 (%ir-block.0):
; GFX900-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX900-NEXT: {{ $}}
; GFX900-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX900-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX900-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX900-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX900-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX900-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX900-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX900-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX900-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX900-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX900-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX900-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX900-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX900-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX900-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY9]], [[C1]](s64)
; GFX900-NEXT: [[COPY10:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX900-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX900-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX900-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX900-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX900-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX900-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX900-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX900-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY15]], [[C2]](s32)
; GFX900-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY14]], [[SHL]]
; GFX900-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX900-NEXT: [[COPY16:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX900-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY16]](<4 x s32>)
; GFX900-NEXT: $sgpr4_sgpr5 = COPY [[COPY8]](p4)
; GFX900-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX900-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX900-NEXT: $sgpr10_sgpr11 = COPY [[COPY10]](s64)
; GFX900-NEXT: $sgpr12 = COPY [[COPY11]](s32)
; GFX900-NEXT: $sgpr13 = COPY [[COPY12]](s32)
; GFX900-NEXT: $sgpr14 = COPY [[COPY13]](s32)
; GFX900-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX900-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX900-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX900-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX900-NEXT: S_ENDPGM 0
;
; GFX908-LABEL: name: test_only_workitem_id_xz
; GFX908: bb.1 (%ir-block.0):
; GFX908-NEXT: liveins: $sgpr12, $sgpr13, $sgpr14, $vgpr0, $vgpr2, $sgpr4_sgpr5, $sgpr6_sgpr7, $sgpr8_sgpr9
; GFX908-NEXT: {{ $}}
; GFX908-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr2
; GFX908-NEXT: [[COPY1:%[0-9]+]]:vgpr_32(s32) = COPY $vgpr0
; GFX908-NEXT: [[COPY2:%[0-9]+]]:sgpr_32 = COPY $sgpr14
; GFX908-NEXT: [[COPY3:%[0-9]+]]:sgpr_32 = COPY $sgpr13
; GFX908-NEXT: [[COPY4:%[0-9]+]]:sgpr_32 = COPY $sgpr12
; GFX908-NEXT: [[COPY5:%[0-9]+]]:sgpr_64 = COPY $sgpr8_sgpr9
; GFX908-NEXT: [[COPY6:%[0-9]+]]:sgpr_64 = COPY $sgpr4_sgpr5
; GFX908-NEXT: [[COPY7:%[0-9]+]]:_(p4) = COPY $sgpr6_sgpr7
; GFX908-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
; GFX908-NEXT: ADJCALLSTACKUP 0, 0, implicit-def $scc
; GFX908-NEXT: [[GV:%[0-9]+]]:_(p0) = G_GLOBAL_VALUE @external_void_func_i32
; GFX908-NEXT: [[COPY8:%[0-9]+]]:_(p4) = COPY [[COPY6]]
; GFX908-NEXT: [[DEF:%[0-9]+]]:_(p4) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY9:%[0-9]+]]:_(p4) = COPY [[COPY7]](p4)
; GFX908-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
; GFX908-NEXT: [[PTR_ADD:%[0-9]+]]:_(p4) = G_PTR_ADD [[COPY9]], [[C1]](s64)
; GFX908-NEXT: [[COPY10:%[0-9]+]]:_(s64) = COPY [[COPY5]]
; GFX908-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[COPY4]]
; GFX908-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[COPY3]]
; GFX908-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[COPY2]]
; GFX908-NEXT: [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; GFX908-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[COPY1]](s32)
; GFX908-NEXT: [[COPY15:%[0-9]+]]:_(s32) = COPY [[COPY]](s32)
; GFX908-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 20
; GFX908-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY15]], [[C2]](s32)
; GFX908-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY14]], [[SHL]]
; GFX908-NEXT: $vgpr0 = COPY [[C]](s32)
; GFX908-NEXT: [[COPY16:%[0-9]+]]:_(<4 x s32>) = COPY $private_rsrc_reg
; GFX908-NEXT: $sgpr0_sgpr1_sgpr2_sgpr3 = COPY [[COPY16]](<4 x s32>)
; GFX908-NEXT: $sgpr4_sgpr5 = COPY [[COPY8]](p4)
; GFX908-NEXT: $sgpr6_sgpr7 = COPY [[DEF]](p4)
; GFX908-NEXT: $sgpr8_sgpr9 = COPY [[PTR_ADD]](p4)
; GFX908-NEXT: $sgpr10_sgpr11 = COPY [[COPY10]](s64)
; GFX908-NEXT: $sgpr12 = COPY [[COPY11]](s32)
; GFX908-NEXT: $sgpr13 = COPY [[COPY12]](s32)
; GFX908-NEXT: $sgpr14 = COPY [[COPY13]](s32)
; GFX908-NEXT: $sgpr15 = COPY [[DEF1]](s32)
; GFX908-NEXT: $vgpr31 = COPY [[OR]](s32)
; GFX908-NEXT: $sgpr30_sgpr31 = noconvergent G_SI_CALL [[GV]](p0), @external_void_func_i32, csr_amdgpu, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4_sgpr5, implicit $sgpr6_sgpr7, implicit $sgpr8_sgpr9, implicit $sgpr10_sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $vgpr31
; GFX908-NEXT: ADJCALLSTACKDOWN 0, 0, implicit-def $scc
; GFX908-NEXT: S_ENDPGM 0
call void @external_void_func_i32(i32 42)
ret void
}
declare i32 @llvm.amdgcn.workitem.id.x() #1
declare i32 @llvm.amdgcn.workitem.id.y() #1
declare i32 @llvm.amdgcn.workitem.id.z() #1
attributes #0 = { nounwind }
attributes #1 = { nounwind readnone speculatable willreturn }
!llvm.module.flags = !{!6}
!0 = !{i32 64, i32 1, i32 1}
!1 = !{i32 1, i32 64, i32 1}
!2 = !{i32 1, i32 1, i32 64}
!3 = !{i32 32, i32 2, i32 1}
!4 = !{i32 1, i32 32, i32 2}
!5 = !{i32 32, i32 1, i32 2}
!6 = !{i32 1, !"amdgpu_code_object_version", i32 500}
Reference in a new issue View git blame Copy permalink