diff options
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/optimize-if-exec-masking.mir | 314 |
1 files changed, 42 insertions, 272 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/optimize-if-exec-masking.mir b/llvm/test/CodeGen/AMDGPU/optimize-if-exec-masking.mir index 576c1ce9874..a9a0bba2fb4 100644 --- a/llvm/test/CodeGen/AMDGPU/optimize-if-exec-masking.mir +++ b/llvm/test/CodeGen/AMDGPU/optimize-if-exec-masking.mir @@ -1,29 +1,21 @@ # RUN: llc -march=amdgcn -verify-machineinstrs -run-pass si-optimize-exec-masking -o - %s | FileCheck %s --- | - define amdgpu_kernel void @optimize_if_and_saveexec_xor(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_and_saveexec_xor(i32 %z, i32 %v) { main_body: - %id = call i32 @llvm.amdgcn.workitem.id.x() - %cc = icmp eq i32 %id, 0 - %0 = call { i1, i64 } @llvm.amdgcn.if(i1 %cc) - %1 = extractvalue { i1, i64 } %0, 0 - %2 = extractvalue { i1, i64 } %0, 1 - br i1 %1, label %if, label %end + br i1 undef, label %if, label %end if: ; preds = %main_body %v.if = load volatile i32, i32 addrspace(1)* undef br label %end end: ; preds = %if, %main_body - %r = phi i32 [ 4, %main_body ], [ %v.if, %if ] - call void @llvm.amdgcn.end.cf(i64 %2) - store i32 %r, i32 addrspace(1)* undef ret void } - define amdgpu_kernel void @optimize_if_and_saveexec(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_and_saveexec(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -32,9 +24,9 @@ ret void } - define amdgpu_kernel void @optimize_if_or_saveexec(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_or_saveexec(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -43,31 +35,20 @@ ret void } - - define amdgpu_kernel void @optimize_if_and_saveexec_xor_valu_middle(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_and_saveexec_xor_valu_middle(i32 %z, i32 %v) { main_body: - %id = call i32 @llvm.amdgcn.workitem.id.x() - %cc = icmp eq i32 %id, 0 - %0 = call { i1, i64 } @llvm.amdgcn.if(i1 %cc) - %1 = extractvalue { i1, i64 } %0, 0 - %2 = extractvalue { i1, i64 } %0, 1 - store i32 %id, i32 addrspace(1)* undef - br i1 %1, label %if, label %end + br i1 undef, label %if, label %end if: ; preds = %main_body - %v.if = load volatile i32, i32 addrspace(1)* undef br label %end end: ; preds = %if, %main_body - %r = phi i32 [ 4, %main_body ], [ %v.if, %if ] - call void @llvm.amdgcn.end.cf(i64 %2) - store i32 %r, i32 addrspace(1)* undef ret void } - define amdgpu_kernel void @optimize_if_and_saveexec_xor_wrong_reg(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_and_saveexec_xor_wrong_reg(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -76,9 +57,9 @@ ret void } - define amdgpu_kernel void @optimize_if_and_saveexec_xor_modify_copy_to_exec(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_and_saveexec_xor_modify_copy_to_exec(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -87,9 +68,9 @@ ret void } - define amdgpu_kernel void @optimize_if_and_saveexec_xor_live_out_setexec(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_and_saveexec_xor_live_out_setexec(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -98,9 +79,9 @@ ret void } - define amdgpu_kernel void @optimize_if_unknown_saveexec(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_unknown_saveexec(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -109,9 +90,9 @@ ret void } - define amdgpu_kernel void @optimize_if_andn2_saveexec(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_andn2_saveexec(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -120,9 +101,9 @@ ret void } - define amdgpu_kernel void @optimize_if_andn2_saveexec_no_commute(i32 %z, i32 %v) #0 { + define amdgpu_kernel void @optimize_if_andn2_saveexec_no_commute(i32 %z, i32 %v) { main_body: - br i1 undef, label %if, label %end + br i1 undef, label %if, label %end if: br label %end @@ -131,17 +112,6 @@ ret void } - ; Function Attrs: nounwind readnone - declare i32 @llvm.amdgcn.workitem.id.x() #1 - - declare { i1, i64 } @llvm.amdgcn.if(i1) - - declare void @llvm.amdgcn.end.cf(i64) - - - attributes #0 = { nounwind } - attributes #1 = { nounwind readnone } - ... --- # CHECK-LABEL: name: optimize_if_and_saveexec_xor{{$}} @@ -150,28 +120,8 @@ # CHECK-NEXT: SI_MASK_BRANCH name: optimize_if_and_saveexec_xor -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -190,7 +140,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -198,7 +148,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -208,28 +158,8 @@ body: | # CHECK-NEXT: SI_MASK_BRANCH name: optimize_if_and_saveexec -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -247,7 +177,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -255,7 +185,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -265,28 +195,8 @@ body: | # CHECK-NEXT: SI_MASK_BRANCH name: optimize_if_or_saveexec -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -304,7 +214,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -312,40 +222,20 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... --- # CHECK-LABEL: name: optimize_if_and_saveexec_xor_valu_middle # CHECK: $sgpr2_sgpr3 = S_AND_B64 $sgpr0_sgpr1, killed $vcc, implicit-def $scc -# CHECK-NEXT: BUFFER_STORE_DWORD_OFFSET $vgpr0, undef $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`, addrspace 1) +# CHECK-NEXT: BUFFER_STORE_DWORD_OFFSET $vgpr0, undef $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec # CHECK-NEXT: $sgpr0_sgpr1 = S_XOR_B64 $sgpr2_sgpr3, killed $sgpr0_sgpr1, implicit-def $scc # CHECK-NEXT: $exec = COPY killed $sgpr2_sgpr3 # CHECK-NEXT: SI_MASK_BRANCH name: optimize_if_and_saveexec_xor_valu_middle -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -354,7 +244,7 @@ body: | $vcc = V_CMP_EQ_I32_e64 0, killed $vgpr0, implicit $exec $vgpr0 = V_MOV_B32_e32 4, implicit $exec $sgpr2_sgpr3 = S_AND_B64 $sgpr0_sgpr1, killed $vcc, implicit-def $scc - BUFFER_STORE_DWORD_OFFSET $vgpr0, undef $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET $vgpr0, undef $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec $sgpr0_sgpr1 = S_XOR_B64 $sgpr2_sgpr3, killed $sgpr0_sgpr1, implicit-def $scc $exec = S_MOV_B64_term killed $sgpr2_sgpr3 SI_MASK_BRANCH %bb.2, implicit $exec @@ -365,7 +255,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -373,7 +263,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -384,28 +274,8 @@ body: | # CHECK-NEXT: $exec = COPY $sgpr0_sgpr1 # CHECK-NEXT: SI_MASK_BRANCH %bb.2, implicit $exec name: optimize_if_and_saveexec_xor_wrong_reg -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -423,7 +293,7 @@ body: | bb.1.if: liveins: $sgpr0_sgpr1 , $sgpr4_sgpr5_sgpr6_sgpr7 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1, $sgpr4_sgpr5_sgpr6_sgpr7 @@ -431,7 +301,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -444,28 +314,8 @@ body: | # CHECK-NEXT: SI_MASK_BRANCH %bb.2, implicit $exec name: optimize_if_and_saveexec_xor_modify_copy_to_exec -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -485,7 +335,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -495,7 +345,7 @@ body: | $sgpr1 = S_MOV_B32 1 $sgpr2 = S_MOV_B32 -1 $sgpr3 = S_MOV_B32 61440 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -506,28 +356,8 @@ body: | # CHECK-NEXT: $exec = COPY $sgpr2_sgpr3 # CHECK-NEXT: SI_MASK_BRANCH name: optimize_if_and_saveexec_xor_live_out_setexec -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -546,7 +376,7 @@ body: | S_SLEEP 0, implicit $sgpr2_sgpr3 $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -554,7 +384,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -566,28 +396,8 @@ body: | # CHECK-NEXT: SI_MASK_BRANCH %bb.2, implicit $exec name: optimize_if_unknown_saveexec -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -605,7 +415,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -613,7 +423,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -623,28 +433,8 @@ body: | # CHECK-NEXT: SI_MASK_BRANCH name: optimize_if_andn2_saveexec -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -662,7 +452,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -670,7 +460,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... @@ -680,28 +470,8 @@ body: | # CHECK-NEXT: $exec = COPY killed $sgpr2_sgpr3 # CHECK-NEXT: SI_MASK_BRANCH %bb.2, implicit $exec name: optimize_if_andn2_saveexec_no_commute -alignment: 0 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true liveins: - { reg: '$vgpr0' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false body: | bb.0.main_body: liveins: $vgpr0 @@ -719,7 +489,7 @@ body: | $sgpr7 = S_MOV_B32 61440 $sgpr6 = S_MOV_B32 -1 - $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec :: (volatile load 4 from `i32 addrspace(1)* undef`) + $vgpr0 = BUFFER_LOAD_DWORD_OFFSET $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, 0, implicit $exec bb.2.end: liveins: $vgpr0, $sgpr0_sgpr1 @@ -727,7 +497,7 @@ body: | $exec = S_OR_B64 $exec, killed $sgpr0_sgpr1, implicit-def $scc $sgpr3 = S_MOV_B32 61440 $sgpr2 = S_MOV_B32 -1 - BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into `i32 addrspace(1)* undef`) + BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 0, 0, 0, 0, implicit $exec S_ENDPGM ... |
