60 changes: 30 additions & 30 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-shl.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -211,28 +211,28 @@ body: |
; GFX6-LABEL: name: shl_s64_sv
; GFX6: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr0_sgpr1
; GFX6: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX6: [[V_LSHL_B64_:%[0-9]+]]:vreg_64 = V_LSHL_B64 [[COPY]], [[COPY1]], implicit $exec
; GFX6: S_ENDPGM 0, implicit [[V_LSHL_B64_]]
; GFX6: [[V_LSHL_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHL_B64_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX6: S_ENDPGM 0, implicit [[V_LSHL_B64_e64_]]
; GFX7-LABEL: name: shl_s64_sv
; GFX7: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr0_sgpr1
; GFX7: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX7: [[V_LSHL_B64_:%[0-9]+]]:vreg_64 = V_LSHL_B64 [[COPY]], [[COPY1]], implicit $exec
; GFX7: S_ENDPGM 0, implicit [[V_LSHL_B64_]]
; GFX7: [[V_LSHL_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHL_B64_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX7: S_ENDPGM 0, implicit [[V_LSHL_B64_e64_]]
; GFX8-LABEL: name: shl_s64_sv
; GFX8: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr0_sgpr1
; GFX8: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX8: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX8: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
; GFX9-LABEL: name: shl_s64_sv
; GFX9: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr0_sgpr1
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX9: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
; GFX10-LABEL: name: shl_s64_sv
; GFX10: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr0_sgpr1
; GFX10: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX10: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
%0:sgpr(s64) = COPY $sgpr0_sgpr1
%1:vgpr(s32) = COPY $vgpr0
%2:vgpr(s64) = G_SHL %0, %1
Expand All @@ -250,28 +250,28 @@ body: |
; GFX6-LABEL: name: shl_s64_vs
; GFX6: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX6: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX6: [[V_LSHL_B64_:%[0-9]+]]:vreg_64 = V_LSHL_B64 [[COPY]], [[COPY1]], implicit $exec
; GFX6: S_ENDPGM 0, implicit [[V_LSHL_B64_]]
; GFX6: [[V_LSHL_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHL_B64_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX6: S_ENDPGM 0, implicit [[V_LSHL_B64_e64_]]
; GFX7-LABEL: name: shl_s64_vs
; GFX7: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX7: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX7: [[V_LSHL_B64_:%[0-9]+]]:vreg_64 = V_LSHL_B64 [[COPY]], [[COPY1]], implicit $exec
; GFX7: S_ENDPGM 0, implicit [[V_LSHL_B64_]]
; GFX7: [[V_LSHL_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHL_B64_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX7: S_ENDPGM 0, implicit [[V_LSHL_B64_e64_]]
; GFX8-LABEL: name: shl_s64_vs
; GFX8: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX8: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX8: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX8: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
; GFX9-LABEL: name: shl_s64_vs
; GFX9: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX9: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX9: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
; GFX10-LABEL: name: shl_s64_vs
; GFX10: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX10: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX10: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX10: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
%0:vgpr(s64) = COPY $vgpr0_vgpr1
%1:sgpr(s32) = COPY $sgpr0
%2:vgpr(s64) = G_SHL %0, %1
Expand All @@ -289,28 +289,28 @@ body: |
; GFX6-LABEL: name: shl_s64_vv
; GFX6: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX6: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX6: [[V_LSHL_B64_:%[0-9]+]]:vreg_64 = V_LSHL_B64 [[COPY]], [[COPY1]], implicit $exec
; GFX6: S_ENDPGM 0, implicit [[V_LSHL_B64_]]
; GFX6: [[V_LSHL_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHL_B64_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX6: S_ENDPGM 0, implicit [[V_LSHL_B64_e64_]]
; GFX7-LABEL: name: shl_s64_vv
; GFX7: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX7: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX7: [[V_LSHL_B64_:%[0-9]+]]:vreg_64 = V_LSHL_B64 [[COPY]], [[COPY1]], implicit $exec
; GFX7: S_ENDPGM 0, implicit [[V_LSHL_B64_]]
; GFX7: [[V_LSHL_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHL_B64_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX7: S_ENDPGM 0, implicit [[V_LSHL_B64_e64_]]
; GFX8-LABEL: name: shl_s64_vv
; GFX8: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX8: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX8: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX8: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
; GFX9-LABEL: name: shl_s64_vv
; GFX9: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX9: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX9: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
; GFX10-LABEL: name: shl_s64_vv
; GFX10: [[COPY:%[0-9]+]]:vreg_64 = COPY $vgpr0_vgpr1
; GFX10: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX10: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_LSHLREV_B64_]]
; GFX10: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_LSHLREV_B64_e64_]]
%0:vgpr(s64) = COPY $vgpr0_vgpr1
%1:vgpr(s32) = COPY $vgpr2
%2:vgpr(s64) = G_SHL %0, %1
Expand Down
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-shl.s16.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -176,8 +176,8 @@ body: |
; GFX10: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX10: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX10: [[V_LSHLREV_B16_e64_:%[0-9]+]]:vgpr_32 = V_LSHLREV_B16_e64 [[COPY1]], [[COPY]], implicit $exec
; GFX10: [[V_BFE_U32_:%[0-9]+]]:vgpr_32 = V_BFE_U32 [[V_LSHLREV_B16_e64_]], 0, 16, implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_BFE_U32_]]
; GFX10: [[V_BFE_U32_e64_:%[0-9]+]]:vgpr_32 = V_BFE_U32_e64 [[V_LSHLREV_B16_e64_]], 0, 16, implicit $exec
; GFX10: S_ENDPGM 0, implicit [[V_BFE_U32_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s16) = G_TRUNC %0
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-shuffle-vector.v2s16.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -251,8 +251,8 @@ body: |
; GFX9-LABEL: name: v_shufflevector_v2s16_v2s16_1_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[V_ALIGNBIT_B32_:%[0-9]+]]:vgpr_32 = V_ALIGNBIT_B32 [[COPY]], [[COPY]], 16, implicit $exec
; GFX9: $vgpr0 = COPY [[V_ALIGNBIT_B32_]]
; GFX9: [[V_ALIGNBIT_B32_e64_:%[0-9]+]]:vgpr_32 = V_ALIGNBIT_B32_e64 [[COPY]], [[COPY]], 16, implicit $exec
; GFX9: $vgpr0 = COPY [[V_ALIGNBIT_B32_e64_]]
%0:vgpr(<2 x s16>) = COPY $vgpr0
%1:vgpr(<2 x s16>) = COPY $vgpr1
%2:vgpr(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0)
Expand Down Expand Up @@ -338,8 +338,8 @@ body: |
; GFX9-LABEL: name: v_shufflevector_v2s16_v2s16_3_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[V_ALIGNBIT_B32_:%[0-9]+]]:vgpr_32 = V_ALIGNBIT_B32 [[COPY]], [[COPY]], 16, implicit $exec
; GFX9: $vgpr0 = COPY [[V_ALIGNBIT_B32_]]
; GFX9: [[V_ALIGNBIT_B32_e64_:%[0-9]+]]:vgpr_32 = V_ALIGNBIT_B32_e64 [[COPY]], [[COPY]], 16, implicit $exec
; GFX9: $vgpr0 = COPY [[V_ALIGNBIT_B32_e64_]]
%0:vgpr(<2 x s16>) = COPY $vgpr0
%1:vgpr(<2 x s16>) = COPY $vgpr1
%2:vgpr(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 2)
Expand Down
24 changes: 12 additions & 12 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-smulh.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -44,13 +44,13 @@ body: |
; SI-LABEL: name: smulh_s32_sv
; SI: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
; SI: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; SI: [[V_MUL_HI_I32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_I32_]]
; SI: [[V_MUL_HI_I32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32_e64 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_I32_e64_]]
; GFX9-LABEL: name: smulh_s32_sv
; GFX9: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[V_MUL_HI_I32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_I32_]]
; GFX9: [[V_MUL_HI_I32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_I32_e64_]]
%0:sgpr(s32) = COPY $sgpr0
%1:vgpr(s32) = COPY $vgpr0
%2:vgpr(s32) = G_SMULH %0, %1
Expand All @@ -69,13 +69,13 @@ body: |
; SI-LABEL: name: smulh_s32_vs
; SI: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; SI: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; SI: [[V_MUL_HI_I32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_I32_]]
; SI: [[V_MUL_HI_I32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32_e64 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_I32_e64_]]
; GFX9-LABEL: name: smulh_s32_vs
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX9: [[V_MUL_HI_I32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_I32_]]
; GFX9: [[V_MUL_HI_I32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_I32_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:sgpr(s32) = COPY $sgpr0
%2:vgpr(s32) = G_SMULH %0, %1
Expand All @@ -94,13 +94,13 @@ body: |
; SI-LABEL: name: smulh_s32_vv
; SI: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; SI: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; SI: [[V_MUL_HI_I32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_I32_]]
; SI: [[V_MUL_HI_I32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32_e64 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_I32_e64_]]
; GFX9-LABEL: name: smulh_s32_vv
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[V_MUL_HI_I32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_I32_]]
; GFX9: [[V_MUL_HI_I32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_I32_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_I32_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s32) = G_SMULH %0, %1
Expand Down
24 changes: 12 additions & 12 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-umulh.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -44,13 +44,13 @@ body: |
; SI-LABEL: name: umulh_s32_sv
; SI: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
; SI: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; SI: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_U32_]]
; SI: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_U32_e64_]]
; GFX9-LABEL: name: umulh_s32_sv
; GFX9: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_U32_]]
; GFX9: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_U32_e64_]]
%0:sgpr(s32) = COPY $sgpr0
%1:vgpr(s32) = COPY $vgpr0
%2:vgpr(s32) = G_UMULH %0, %1
Expand All @@ -69,13 +69,13 @@ body: |
; SI-LABEL: name: umulh_s32_vs
; SI: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; SI: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; SI: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_U32_]]
; SI: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_U32_e64_]]
; GFX9-LABEL: name: umulh_s32_vs
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr0
; GFX9: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_U32_]]
; GFX9: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_U32_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:sgpr(s32) = COPY $sgpr0
%2:vgpr(s32) = G_UMULH %0, %1
Expand All @@ -94,13 +94,13 @@ body: |
; SI-LABEL: name: umulh_s32_vv
; SI: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; SI: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; SI: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_U32_]]
; SI: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY]], [[COPY1]], implicit $exec
; SI: S_ENDPGM 0, implicit [[V_MUL_HI_U32_e64_]]
; GFX9-LABEL: name: umulh_s32_vv
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_U32_]]
; GFX9: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MUL_HI_U32_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s32) = G_UMULH %0, %1
Expand Down
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-zext.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -128,8 +128,8 @@ body: |
; GCN-LABEL: name: zext_vgpr_s16_to_vgpr_s32
; GCN: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GCN: [[V_BFE_U32_:%[0-9]+]]:vgpr_32 = V_BFE_U32 [[COPY]], 0, 16, implicit $exec
; GCN: $vgpr0 = COPY [[V_BFE_U32_]]
; GCN: [[V_BFE_U32_e64_:%[0-9]+]]:vgpr_32 = V_BFE_U32_e64 [[COPY]], 0, 16, implicit $exec
; GCN: $vgpr0 = COPY [[V_BFE_U32_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s16) = G_TRUNC %0
%2:vgpr(s32) = G_ZEXT %1
Expand Down
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.raw.buffer.load.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -407,8 +407,8 @@ define amdgpu_ps float @raw_buffer_load_i8__sgpr_rsrc__vgpr_voffset__sgpr_soffse
; CHECK: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr6
; CHECK: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2, [[COPY3]], %subreg.sub3
; CHECK: [[BUFFER_LOAD_UBYTE_OFFEN:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_UBYTE_OFFEN [[COPY4]], [[REG_SEQUENCE]], [[COPY5]], 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load 1 from custom "TargetCustom7", addrspace 4)
; CHECK: [[V_BFE_I32_:%[0-9]+]]:vgpr_32 = V_BFE_I32 [[BUFFER_LOAD_UBYTE_OFFEN]], 0, 8, implicit $exec
; CHECK: $vgpr0 = COPY [[V_BFE_I32_]]
; CHECK: [[V_BFE_I32_e64_:%[0-9]+]]:vgpr_32 = V_BFE_I32_e64 [[BUFFER_LOAD_UBYTE_OFFEN]], 0, 8, implicit $exec
; CHECK: $vgpr0 = COPY [[V_BFE_I32_e64_]]
; CHECK: SI_RETURN_TO_EPILOG implicit $vgpr0
%val = call i8 @llvm.amdgcn.raw.buffer.load.i8(<4 x i32> %rsrc, i32 %voffset, i32 %soffset, i32 0)
%zext = sext i8 %val to i32
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.struct.buffer.load.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -250,8 +250,8 @@ define amdgpu_ps float @struct_buffer_load_i8_sext__sgpr_rsrc__vgpr_vindex__vgpr
; CHECK: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2, [[COPY3]], %subreg.sub3
; CHECK: [[REG_SEQUENCE1:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY4]], %subreg.sub0, [[COPY5]], %subreg.sub1
; CHECK: [[BUFFER_LOAD_UBYTE_BOTHEN:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_UBYTE_BOTHEN [[REG_SEQUENCE1]], [[REG_SEQUENCE]], [[COPY6]], 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load 1 from custom "TargetCustom7", addrspace 4)
; CHECK: [[V_BFE_I32_:%[0-9]+]]:vgpr_32 = V_BFE_I32 [[BUFFER_LOAD_UBYTE_BOTHEN]], 0, 8, implicit $exec
; CHECK: $vgpr0 = COPY [[V_BFE_I32_]]
; CHECK: [[V_BFE_I32_e64_:%[0-9]+]]:vgpr_32 = V_BFE_I32_e64 [[BUFFER_LOAD_UBYTE_BOTHEN]], 0, 8, implicit $exec
; CHECK: $vgpr0 = COPY [[V_BFE_I32_e64_]]
; CHECK: SI_RETURN_TO_EPILOG implicit $vgpr0
%val = call i8 @llvm.amdgcn.struct.buffer.load.i8(<4 x i32> %rsrc, i32 %vindex, i32 %voffset, i32 %soffset, i32 0)
%ext = sext i8 %val to i32
Expand Down Expand Up @@ -295,8 +295,8 @@ define amdgpu_ps float @struct_buffer_load_i16_sext__sgpr_rsrc__vgpr_vindex__vgp
; CHECK: [[REG_SEQUENCE:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1, [[COPY2]], %subreg.sub2, [[COPY3]], %subreg.sub3
; CHECK: [[REG_SEQUENCE1:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY4]], %subreg.sub0, [[COPY5]], %subreg.sub1
; CHECK: [[BUFFER_LOAD_USHORT_BOTHEN:%[0-9]+]]:vgpr_32 = BUFFER_LOAD_USHORT_BOTHEN [[REG_SEQUENCE1]], [[REG_SEQUENCE]], [[COPY6]], 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load 2 from custom "TargetCustom7", align 1, addrspace 4)
; CHECK: [[V_BFE_I32_:%[0-9]+]]:vgpr_32 = V_BFE_I32 [[BUFFER_LOAD_USHORT_BOTHEN]], 0, 16, implicit $exec
; CHECK: $vgpr0 = COPY [[V_BFE_I32_]]
; CHECK: [[V_BFE_I32_e64_:%[0-9]+]]:vgpr_32 = V_BFE_I32_e64 [[BUFFER_LOAD_USHORT_BOTHEN]], 0, 16, implicit $exec
; CHECK: $vgpr0 = COPY [[V_BFE_I32_e64_]]
; CHECK: SI_RETURN_TO_EPILOG implicit $vgpr0
%val = call i16 @llvm.amdgcn.struct.buffer.load.i16(<4 x i32> %rsrc, i32 %vindex, i32 %voffset, i32 %soffset, i32 0)
%ext = sext i16 %val to i32
Expand Down
430 changes: 215 additions & 215 deletions llvm/test/CodeGen/AMDGPU/accvgpr-copy.mir
View file

Large diffs are not rendered by default.

12 changes: 6 additions & 6 deletions llvm/test/CodeGen/AMDGPU/clamp-omod-special-case.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,7 @@ body: |
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%14 = S_MOV_B32 2
%26 = V_LSHL_B64 killed %25, 2, implicit $exec
%26 = V_LSHL_B64_e64 killed %25, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%18 = COPY %26
%17 = BUFFER_LOAD_DWORD_ADDR64 %26, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -114,7 +114,7 @@ body: |
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%14 = S_MOV_B32 2
%26 = V_LSHL_B64 killed %25, 2, implicit $exec
%26 = V_LSHL_B64_e64 killed %25, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%18 = COPY %26
%17 = BUFFER_LOAD_DWORD_ADDR64 %26, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -177,7 +177,7 @@ body: |
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%14 = S_MOV_B32 2
%26 = V_LSHL_B64 killed %25, 2, implicit $exec
%26 = V_LSHL_B64_e64 killed %25, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%18 = COPY %26
%17 = BUFFER_LOAD_DWORD_ADDR64 %26, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -242,7 +242,7 @@ body: |
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%14 = S_MOV_B32 2
%26 = V_LSHL_B64 killed %25, 2, implicit $exec
%26 = V_LSHL_B64_e64 killed %25, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%18 = COPY %26
%17 = BUFFER_LOAD_DWORD_ADDR64 %26, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -319,7 +319,7 @@ body: |
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%14 = S_MOV_B32 2
%26 = V_LSHL_B64 killed %25, 2, implicit $exec
%26 = V_LSHL_B64_e64 killed %25, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%18 = COPY %26
%17 = BUFFER_LOAD_DWORD_ADDR64 %26, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -384,7 +384,7 @@ body: |
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%14 = S_MOV_B32 2
%26 = V_LSHL_B64 killed %25, 2, implicit $exec
%26 = V_LSHL_B64_e64 killed %25, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%18 = COPY %26
%17 = BUFFER_LOAD_DWORD_ADDR64 %26, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/coalescer-subranges-another-copymi-not-live.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ body: |
bb.8:
successors: %bb.9(0x40000000), %bb.11(0x40000000)
%18:vgpr_32 = V_MUL_LO_I32 %15.sub1, target-flags(amdgpu-gotprel32-lo) 7, implicit $exec
%18:vgpr_32 = V_MUL_LO_I32_e64 %15.sub1, target-flags(amdgpu-gotprel32-lo) 7, implicit $exec
S_CBRANCH_SCC1 %bb.11, implicit undef $scc
S_BRANCH %bb.9
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/coalescer-subranges-another-prune-error.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -260,7 +260,7 @@ body: |
%110:vgpr_32 = IMAGE_SAMPLE_V1_V2 killed %107, killed %109, undef %111:sgpr_128, 8, 0, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load 16 from constant-pool, addrspace 4)
%112:vgpr_32 = nofpexcept V_MUL_F32_e32 0, killed %110, implicit $mode, implicit $exec
%113:vgpr_32 = nofpexcept V_MUL_F32_e32 0, killed %112, implicit $mode, implicit $exec
%114:vgpr_32 = nofpexcept V_MAD_F32 0, killed %113, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%114:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %113, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%115:vgpr_32 = nofpexcept V_MAX_F32_e32 0, killed %114, implicit $mode, implicit $exec
%116:vgpr_32 = nofpexcept V_CVT_PKRTZ_F16_F32_e64 0, killed %115, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
EXP 0, undef %117:vgpr_32, killed %116, undef %118:vgpr_32, undef %119:vgpr_32, -1, -1, 15, implicit $exec
Expand Down
10 changes: 5 additions & 5 deletions llvm/test/CodeGen/AMDGPU/coalescer-subregjoin-fullcopy.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -144,23 +144,23 @@ body: |
bb.16:
successors: %bb.17(0x80000000)
%39:vgpr_32 = nofpexcept V_FMA_F32 0, undef %40:vgpr_32, 0, killed %37.sub0, 0, undef %41:vgpr_32, 0, 0, implicit $mode, implicit $exec
%42:vgpr_32 = nofpexcept V_FMA_F32 0, undef %43:vgpr_32, 0, undef %44:vgpr_32, 0, killed %39, 0, 0, implicit $mode, implicit $exec
%45:vgpr_32 = nofpexcept V_FMA_F32 0, undef %46:vgpr_32, 0, undef %47:vgpr_32, 0, killed %42, 0, 0, implicit $mode, implicit $exec
%39:vgpr_32 = nofpexcept V_FMA_F32_e64 0, undef %40:vgpr_32, 0, killed %37.sub0, 0, undef %41:vgpr_32, 0, 0, implicit $mode, implicit $exec
%42:vgpr_32 = nofpexcept V_FMA_F32_e64 0, undef %43:vgpr_32, 0, undef %44:vgpr_32, 0, killed %39, 0, 0, implicit $mode, implicit $exec
%45:vgpr_32 = nofpexcept V_FMA_F32_e64 0, undef %46:vgpr_32, 0, undef %47:vgpr_32, 0, killed %42, 0, 0, implicit $mode, implicit $exec
dead %48:vgpr_32 = nofpexcept V_MUL_F32_e32 undef %49:vgpr_32, killed %45, implicit $mode, implicit $exec
%50:vgpr_32 = nofpexcept V_MUL_F32_e32 undef %51:vgpr_32, undef %52:vgpr_32, implicit $mode, implicit $exec
undef %53.sub1:vreg_128 = COPY %50
%38:vreg_128 = COPY killed %53
bb.17:
%54:vreg_128 = COPY killed %38
%55:vgpr_32 = nofpexcept V_FMA_F32 0, killed %54.sub1, 0, target-flags(amdgpu-gotprel32-lo) 1056964608, 0, 1056964608, 0, 0, implicit $mode, implicit $exec
%55:vgpr_32 = nofpexcept V_FMA_F32_e64 0, killed %54.sub1, 0, target-flags(amdgpu-gotprel32-lo) 1056964608, 0, 1056964608, 0, 0, implicit $mode, implicit $exec
EXP 1, undef %56:vgpr_32, killed %55, undef %57:vgpr_32, undef %58:vgpr_32, -1, 0, 15, implicit $exec
S_ENDPGM 0
bb.18:
successors: %bb.7(0x80000000)
dead %59:vgpr_32 = nofpexcept V_FMA_F32 0, killed %9.sub2, 0, undef %60:vgpr_32, 0, undef %61:vgpr_32, 0, 0, implicit $mode, implicit $exec
dead %59:vgpr_32 = nofpexcept V_FMA_F32_e64 0, killed %9.sub2, 0, undef %60:vgpr_32, 0, undef %61:vgpr_32, 0, 0, implicit $mode, implicit $exec
dead %62:vgpr_32 = BUFFER_LOAD_DWORD_OFFEN undef %63:vgpr_32, undef %64:sgpr_128, undef %65:sreg_32, 0, 0, 0, 0, 0, 0, implicit $exec
undef %66.sub1:vreg_128 = COPY %13.sub1
%66.sub2:vreg_128 = COPY %13.sub2
Expand Down
16 changes: 8 additions & 8 deletions llvm/test/CodeGen/AMDGPU/coalescer-with-subregs-bad-identical.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -48,10 +48,10 @@ body: |
%4.sub6:sgpr_256 = COPY %1
%4.sub7:sgpr_256 = COPY killed %1
%5:vgpr_32 = IMAGE_LOAD_V1_V4 killed %3, killed %4, 1, -1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load 16 from constant-pool, addrspace 4)
%6:vgpr_32 = nofpexcept V_MAD_F32 0, killed %5, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%6:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %5, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%7:vgpr_32 = nofpexcept V_RCP_F32_e32 killed %6, implicit $mode, implicit $exec
%8:vgpr_32 = nofpexcept V_MUL_F32_e32 0, killed %7, implicit $mode, implicit $exec
%9:vgpr_32 = nofpexcept V_MAD_F32 0, killed %8, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%9:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %8, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
dead %10:vgpr_32 = nofpexcept V_MAC_F32_e32 undef %11:vgpr_32, undef %12:vgpr_32, undef %10, implicit $mode, implicit $exec
undef %13.sub0:vreg_128 = COPY %9
%14:vgpr_32 = V_MOV_B32_e32 -1, implicit $exec
Expand All @@ -65,7 +65,7 @@ body: |
bb.4:
successors: %bb.5(0x40000000), %bb.7(0x40000000)
%17:vgpr_32 = nofpexcept V_MAD_F32 0, killed %9, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%17:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %9, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%18:vgpr_32 = nofpexcept V_MIN_F32_e32 1065353216, killed %17, implicit $mode, implicit $exec
%19:sreg_64_xexec = nofpexcept V_CMP_NEQ_F32_e64 0, 1065353216, 0, killed %18, 0, implicit $mode, implicit $exec
%20:vgpr_32 = V_MOV_B32_e32 2143289344, implicit $exec
Expand Down Expand Up @@ -140,13 +140,13 @@ body: |
bb.14:
successors: %bb.15(0x40000000), %bb.16(0x40000000)
%38:vgpr_32 = nofpexcept V_MAD_F32 0, killed %36.sub0, 0, target-flags(amdgpu-gotprel) 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%39:vgpr_32 = nofpexcept V_MAD_F32 0, killed %38, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%40:vgpr_32 = nofpexcept V_MAD_F32 0, killed %39, 0, -1090519040, 0, 1056964608, 0, 0, implicit $mode, implicit $exec
%41:vgpr_32 = nofpexcept V_MAD_F32 0, killed %40, 0, 0, 0, -1090519040, 0, 0, implicit $mode, implicit $exec
%38:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %36.sub0, 0, target-flags(amdgpu-gotprel) 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%39:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %38, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%40:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %39, 0, -1090519040, 0, 1056964608, 0, 0, implicit $mode, implicit $exec
%41:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %40, 0, 0, 0, -1090519040, 0, 0, implicit $mode, implicit $exec
%42:vgpr_32 = nofpexcept V_CVT_I32_F32_e32 killed %41, implicit $mode, implicit $exec
%43:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM undef %44:sgpr_128, 12, 0, 0 :: (dereferenceable invariant load 4)
%45:vgpr_32 = V_MUL_LO_I32 killed %42, killed %43, implicit $exec
%45:vgpr_32 = V_MUL_LO_I32_e64 killed %42, killed %43, implicit $exec
%46:vgpr_32 = V_LSHLREV_B32_e32 2, killed %45, implicit $exec
%47:vgpr_32 = BUFFER_LOAD_FORMAT_X_IDXEN killed %46, undef %48:sgpr_128, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load 4 from constant-pool, align 1, addrspace 4)
%49:sreg_64 = V_CMP_NE_U32_e64 0, killed %47, implicit $exec
Expand Down
10 changes: 5 additions & 5 deletions llvm/test/CodeGen/AMDGPU/constant-fold-imm-immreg.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -74,7 +74,7 @@ body: |
%4:sreg_64_xexec = S_LOAD_DWORDX2_IMM %0, 36, 0, 0
%13:vgpr_32 = V_ASHRREV_I32_e64 31, %3, implicit $exec
%14:vreg_64 = REG_SEQUENCE %3, %subreg.hi16, %13, %subreg.lo16
%15:vreg_64 = V_LSHLREV_B64 2, killed %14, implicit $exec
%15:vreg_64 = V_LSHLREV_B64_e64 2, killed %14, implicit $exec
%5:sreg_32_xm0 = COPY %4.sub1
%20:vgpr_32 = V_ADD_CO_U32_e32 %4.sub0, %15.sub0, implicit-def $vcc, implicit $exec
%18:vgpr_32 = COPY killed %5
Expand Down Expand Up @@ -204,7 +204,7 @@ body: |
%3:sreg_64_xexec = S_LOAD_DWORDX2_IMM %0, 36, 0, 0
%15:vgpr_32 = V_ASHRREV_I32_e64 31, %2, implicit $exec
%16:vreg_64 = REG_SEQUENCE %2, %subreg.hi16, %15, %subreg.lo16
%17:vreg_64 = V_LSHLREV_B64 2, killed %16, implicit $exec
%17:vreg_64 = V_LSHLREV_B64_e64 2, killed %16, implicit $exec
%9:sreg_32_xm0 = COPY %3.sub1
%21:vgpr_32 = V_ADD_CO_U32_e32 %3.sub0, %17.sub0, implicit-def $vcc, implicit $exec
%19:vgpr_32 = COPY killed %9
Expand Down Expand Up @@ -328,7 +328,7 @@ body: |
%3:sreg_64_xexec = S_LOAD_DWORDX2_IMM %0, 36, 0, 0
%15:vgpr_32 = V_ASHRREV_I32_e64 31, %2, implicit $exec
%16:vreg_64 = REG_SEQUENCE %2, %subreg.hi16, %15, %subreg.lo16
%17:vreg_64 = V_LSHLREV_B64 2, killed %16, implicit $exec
%17:vreg_64 = V_LSHLREV_B64_e64 2, killed %16, implicit $exec
%9:sreg_32_xm0 = COPY %3.sub1
%21:vgpr_32 = V_ADD_CO_U32_e32 %3.sub0, %17.sub0, implicit-def $vcc, implicit $exec
%19:vgpr_32 = COPY killed %9
Expand Down Expand Up @@ -520,7 +520,7 @@ body: |
%4:sreg_64_xexec = S_LOAD_DWORDX2_IMM %0, 36, 0, 0
%13:vgpr_32 = V_ASHRREV_I32_e64 31, %3, implicit $exec
%14:vreg_64 = REG_SEQUENCE %3, %subreg.sub0, %13, %subreg.sub1
%15:vreg_64 = V_LSHLREV_B64 2, killed %14, implicit $exec
%15:vreg_64 = V_LSHLREV_B64_e64 2, killed %14, implicit $exec
%5:sreg_32_xm0 = COPY %4.sub1
%20:vgpr_32 = V_ADD_CO_U32_e32 %4.sub0, %15.sub0, implicit-def $vcc, implicit $exec
%18:vgpr_32 = COPY killed %5
Expand Down Expand Up @@ -727,7 +727,7 @@ body: |
%3:sreg_64_xexec = S_LOAD_DWORDX2_IMM %0, 36, 0, 0 :: (non-temporal dereferenceable invariant load 8 from `i64 addrspace(4)* undef`, addrspace 4)
%15:vgpr_32 = V_ASHRREV_I32_e64 31, %2, implicit $exec
%16:vreg_64 = REG_SEQUENCE %2, %subreg.sub0, %15, %subreg.sub1
%17:vreg_64 = V_LSHLREV_B64 2, killed %16, implicit $exec
%17:vreg_64 = V_LSHLREV_B64_e64 2, killed %16, implicit $exec
%9:sreg_32_xm0 = COPY %3.sub1
%21:vgpr_32 = V_ADD_CO_U32_e32 %3.sub0, %17.sub0, implicit-def $vcc, implicit $exec
%19:vgpr_32 = COPY killed %9
Expand Down
16 changes: 8 additions & 8 deletions llvm/test/CodeGen/AMDGPU/couldnt-join-subrange-3.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ body: |
; GCN: %3:vgpr_32 = nofpexcept V_TRUNC_F32_e32 undef %4:vgpr_32, implicit $mode, implicit $exec
; GCN: %5:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 %3, implicit $mode, implicit $exec
; GCN: [[V_LSHRREV_B32_e32_:%[0-9]+]]:vgpr_32 = V_LSHRREV_B32_e32 4, %5, implicit $exec
; GCN: undef %11.sub0:vreg_128 = V_MUL_LO_I32 [[V_LSHRREV_B32_e32_]], 3, implicit $exec
; GCN: undef %11.sub0:vreg_128 = V_MUL_LO_I32_e64 [[V_LSHRREV_B32_e32_]], 3, implicit $exec
; GCN: %11.sub3:vreg_128 = COPY %11.sub0
; GCN: [[S_MOV_B32_:%[0-9]+]]:sreg_32_xm0 = S_MOV_B32 0
; GCN: bb.1:
Expand Down Expand Up @@ -49,9 +49,9 @@ body: |
; GCN: bb.5:
; GCN: %21:vgpr_32 = nofpexcept V_MUL_F32_e32 target-flags(amdgpu-gotprel) 0, %11.sub0, implicit $mode, implicit $exec
; GCN: %22:vgpr_32 = nofpexcept V_MIN_F32_e32 1106771968, %21, implicit $mode, implicit $exec
; GCN: %23:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32 0, %22, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
; GCN: %24:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32 0, %23, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
; GCN: %25:vgpr_32 = nofpexcept V_MAD_F32 0, %24, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
; GCN: %23:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32_e64 0, %22, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
; GCN: %24:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32_e64 0, %23, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
; GCN: %25:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %24, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
; GCN: %26:vgpr_32 = nofpexcept V_CVT_PKRTZ_F16_F32_e64 0, %25, 0, undef %27:vgpr_32, 0, 0, implicit $mode, implicit $exec
; GCN: EXP_DONE 0, %26, undef %28:vgpr_32, undef %29:vgpr_32, undef %30:vgpr_32, -1, -1, 15, implicit $exec
; GCN: S_ENDPGM 0
Expand All @@ -61,7 +61,7 @@ body: |
%10:vgpr_32 = nofpexcept V_TRUNC_F32_e32 undef %11:vgpr_32, implicit $mode, implicit $exec
%12:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 killed %10, implicit $mode, implicit $exec
%50:vgpr_32 = V_LSHRREV_B32_e32 4, killed %12, implicit $exec
%51:vgpr_32 = V_MUL_LO_I32 killed %50, 3, implicit $exec
%51:vgpr_32 = V_MUL_LO_I32_e64 killed %50, 3, implicit $exec
undef %52.sub0:vreg_128 = COPY %51
%52.sub3:vreg_128 = COPY %51
%9:sreg_32_xm0 = S_MOV_B32 0
Expand Down Expand Up @@ -104,9 +104,9 @@ body: |
bb.5:
%39:vgpr_32 = nofpexcept V_MUL_F32_e32 target-flags(amdgpu-gotprel) 0, killed %55.sub0, implicit $mode, implicit $exec
%41:vgpr_32 = nofpexcept V_MIN_F32_e32 1106771968, killed %39, implicit $mode, implicit $exec
%42:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32 0, killed %41, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%43:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32 0, killed %42, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%44:vgpr_32 = nofpexcept V_MAD_F32 0, killed %43, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%42:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32_e64 0, killed %41, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%43:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32_e64 0, killed %42, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%44:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %43, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%45:vgpr_32 = nofpexcept V_CVT_PKRTZ_F16_F32_e64 0, killed %44, 0, undef %46:vgpr_32, 0, 0, implicit $mode, implicit $exec
EXP_DONE 0, killed %45, undef %47:vgpr_32, undef %48:vgpr_32, undef %49:vgpr_32, -1, -1, 15, implicit $exec
S_ENDPGM 0
Expand Down
12 changes: 6 additions & 6 deletions llvm/test/CodeGen/AMDGPU/debug-value-scheduler-crash.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -68,9 +68,9 @@ body: |
; CHECK: dead $sgpr30_sgpr31 = SI_CALL [[DEF14]], @foo, csr_amdgpu_highregs, implicit undef $sgpr0_sgpr1_sgpr2_sgpr3, implicit killed $sgpr4, implicit $vgpr0, implicit $vgpr1, implicit $vgpr2, implicit-def $vgpr0
; CHECK: %25:vgpr_32 = nofpexcept V_ADD_F32_e32 %9, [[DEF8]], implicit $mode, implicit $exec
; CHECK: %25:vgpr_32 = nofpexcept V_MAC_F32_e32 [[DEF12]], [[DEF9]], %25, implicit $mode, implicit $exec
; CHECK: dead %26:vgpr_32 = nofpexcept V_MAD_F32 0, %25, 0, [[DEF4]], 0, [[DEF1]], 0, 0, implicit $mode, implicit $exec
; CHECK: dead %27:vgpr_32 = nofpexcept V_MAD_F32 0, %25, 0, [[DEF5]], 0, [[DEF2]], 0, 0, implicit $mode, implicit $exec
; CHECK: dead %28:vgpr_32 = nofpexcept V_MAD_F32 0, %25, 0, [[DEF6]], 0, [[DEF3]], 0, 0, implicit $mode, implicit $exec
; CHECK: dead %26:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %25, 0, [[DEF4]], 0, [[DEF1]], 0, 0, implicit $mode, implicit $exec
; CHECK: dead %27:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %25, 0, [[DEF5]], 0, [[DEF2]], 0, 0, implicit $mode, implicit $exec
; CHECK: dead %28:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %25, 0, [[DEF6]], 0, [[DEF3]], 0, 0, implicit $mode, implicit $exec
; CHECK: GLOBAL_STORE_DWORD [[DEF]], [[DEF10]], 0, 0, 0, 0, implicit $exec
; CHECK: S_ENDPGM 0
bb.0:
Expand Down Expand Up @@ -126,9 +126,9 @@ body: |
dead $sgpr30_sgpr31 = SI_CALL %24, @foo, csr_amdgpu_highregs, implicit undef $sgpr0_sgpr1_sgpr2_sgpr3, implicit killed $sgpr4, implicit killed $vgpr0, implicit killed $vgpr1, implicit killed $vgpr2, implicit-def $vgpr0
%25:vgpr_32 = nofpexcept V_ADD_F32_e32 %9, %8, implicit $mode, implicit $exec
%25:vgpr_32 = nofpexcept V_MAC_F32_e32 %15, %10, %25, implicit $mode, implicit $exec
%26:vgpr_32 = nofpexcept V_MAD_F32 0, %25, 0, %4, 0, %1, 0, 0, implicit $mode, implicit $exec
%27:vgpr_32 = nofpexcept V_MAD_F32 0, %25, 0, %5, 0, %2, 0, 0, implicit $mode, implicit $exec
%28:vgpr_32 = nofpexcept V_MAD_F32 0, %25, 0, %6, 0, %3, 0, 0, implicit $mode, implicit $exec
%26:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %25, 0, %4, 0, %1, 0, 0, implicit $mode, implicit $exec
%27:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %25, 0, %5, 0, %2, 0, 0, implicit $mode, implicit $exec
%28:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %25, 0, %6, 0, %3, 0, 0, implicit $mode, implicit $exec
GLOBAL_STORE_DWORD %0, %11, 0, 0, 0, 0, implicit $exec
S_ENDPGM 0
Expand Down
36 changes: 18 additions & 18 deletions llvm/test/CodeGen/AMDGPU/fdiv-nofpexcept.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -11,23 +11,23 @@ define float @fdiv_f32(float %a, float %b) #0 {
; GCN: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr30_sgpr31
; GCN: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GCN: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GCN: %6:vgpr_32, %7:sreg_64 = nofpexcept V_DIV_SCALE_F32 0, [[COPY2]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %8:vgpr_32, %9:sreg_64 = nofpexcept V_DIV_SCALE_F32 0, [[COPY1]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %6:vgpr_32, %7:sreg_64 = nofpexcept V_DIV_SCALE_F32_e64 0, [[COPY2]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %8:vgpr_32, %9:sreg_64 = nofpexcept V_DIV_SCALE_F32_e64 0, [[COPY1]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %10:vgpr_32 = nofpexcept V_RCP_F32_e64 0, %8, 0, 0, implicit $mode, implicit $exec
; GCN: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 3
; GCN: [[S_MOV_B32_1:%[0-9]+]]:sgpr_32 = S_MOV_B32 1065353216
; GCN: [[S_MOV_B32_2:%[0-9]+]]:sreg_32 = S_MOV_B32 0
; GCN: S_SETREG_B32_mode killed [[S_MOV_B32_]], 2305, implicit-def $mode, implicit $mode
; GCN: %14:vgpr_32 = nofpexcept V_FMA_F32 1, %8, 0, %10, 0, killed [[S_MOV_B32_1]], 0, 0, implicit $mode, implicit $exec
; GCN: %15:vgpr_32 = nofpexcept V_FMA_F32 0, killed %14, 0, %10, 0, %10, 0, 0, implicit $mode, implicit $exec
; GCN: %14:vgpr_32 = nofpexcept V_FMA_F32_e64 1, %8, 0, %10, 0, killed [[S_MOV_B32_1]], 0, 0, implicit $mode, implicit $exec
; GCN: %15:vgpr_32 = nofpexcept V_FMA_F32_e64 0, killed %14, 0, %10, 0, %10, 0, 0, implicit $mode, implicit $exec
; GCN: %16:vgpr_32 = nofpexcept V_MUL_F32_e64 0, %6, 0, %15, 0, 0, implicit $mode, implicit $exec
; GCN: %17:vgpr_32 = nofpexcept V_FMA_F32 1, %8, 0, %16, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: %18:vgpr_32 = nofpexcept V_FMA_F32 0, killed %17, 0, %15, 0, %16, 0, 0, implicit $mode, implicit $exec
; GCN: %19:vgpr_32 = nofpexcept V_FMA_F32 1, %8, 0, %18, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: %17:vgpr_32 = nofpexcept V_FMA_F32_e64 1, %8, 0, %16, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: %18:vgpr_32 = nofpexcept V_FMA_F32_e64 0, killed %17, 0, %15, 0, %16, 0, 0, implicit $mode, implicit $exec
; GCN: %19:vgpr_32 = nofpexcept V_FMA_F32_e64 1, %8, 0, %18, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: S_SETREG_B32_mode killed [[S_MOV_B32_2]], 2305, implicit-def dead $mode, implicit $mode
; GCN: $vcc = COPY %7
; GCN: %20:vgpr_32 = nofpexcept V_DIV_FMAS_F32 0, killed %19, 0, %15, 0, %18, 0, 0, implicit $mode, implicit $vcc, implicit $exec
; GCN: %21:vgpr_32 = nofpexcept V_DIV_FIXUP_F32 0, killed %20, 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %20:vgpr_32 = nofpexcept V_DIV_FMAS_F32_e64 0, killed %19, 0, %15, 0, %18, 0, 0, implicit $mode, implicit $vcc, implicit $exec
; GCN: %21:vgpr_32 = nofpexcept V_DIV_FIXUP_F32_e64 0, killed %20, 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY]]
; GCN: $vgpr0 = COPY %21
; GCN: [[COPY4:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY3]]
Expand All @@ -44,23 +44,23 @@ define float @fdiv_nnan_f32(float %a, float %b) #0 {
; GCN: [[COPY:%[0-9]+]]:sreg_64 = COPY $sgpr30_sgpr31
; GCN: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GCN: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GCN: %6:vgpr_32, %7:sreg_64 = nnan nofpexcept V_DIV_SCALE_F32 0, [[COPY2]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %8:vgpr_32, %9:sreg_64 = nnan nofpexcept V_DIV_SCALE_F32 0, [[COPY1]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %6:vgpr_32, %7:sreg_64 = nnan nofpexcept V_DIV_SCALE_F32_e64 0, [[COPY2]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %8:vgpr_32, %9:sreg_64 = nnan nofpexcept V_DIV_SCALE_F32_e64 0, [[COPY1]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %10:vgpr_32 = nnan nofpexcept V_RCP_F32_e64 0, %8, 0, 0, implicit $mode, implicit $exec
; GCN: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 3
; GCN: [[S_MOV_B32_1:%[0-9]+]]:sgpr_32 = S_MOV_B32 1065353216
; GCN: [[S_MOV_B32_2:%[0-9]+]]:sreg_32 = S_MOV_B32 0
; GCN: S_SETREG_B32_mode killed [[S_MOV_B32_]], 2305, implicit-def $mode, implicit $mode
; GCN: %14:vgpr_32 = nnan nofpexcept V_FMA_F32 1, %8, 0, %10, 0, killed [[S_MOV_B32_1]], 0, 0, implicit $mode, implicit $exec
; GCN: %15:vgpr_32 = nnan nofpexcept V_FMA_F32 0, killed %14, 0, %10, 0, %10, 0, 0, implicit $mode, implicit $exec
; GCN: %14:vgpr_32 = nnan nofpexcept V_FMA_F32_e64 1, %8, 0, %10, 0, killed [[S_MOV_B32_1]], 0, 0, implicit $mode, implicit $exec
; GCN: %15:vgpr_32 = nnan nofpexcept V_FMA_F32_e64 0, killed %14, 0, %10, 0, %10, 0, 0, implicit $mode, implicit $exec
; GCN: %16:vgpr_32 = nnan nofpexcept V_MUL_F32_e64 0, %6, 0, %15, 0, 0, implicit $mode, implicit $exec
; GCN: %17:vgpr_32 = nnan nofpexcept V_FMA_F32 1, %8, 0, %16, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: %18:vgpr_32 = nnan nofpexcept V_FMA_F32 0, killed %17, 0, %15, 0, %16, 0, 0, implicit $mode, implicit $exec
; GCN: %19:vgpr_32 = nnan nofpexcept V_FMA_F32 1, %8, 0, %18, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: %17:vgpr_32 = nnan nofpexcept V_FMA_F32_e64 1, %8, 0, %16, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: %18:vgpr_32 = nnan nofpexcept V_FMA_F32_e64 0, killed %17, 0, %15, 0, %16, 0, 0, implicit $mode, implicit $exec
; GCN: %19:vgpr_32 = nnan nofpexcept V_FMA_F32_e64 1, %8, 0, %18, 0, %6, 0, 0, implicit $mode, implicit $exec
; GCN: S_SETREG_B32_mode killed [[S_MOV_B32_2]], 2305, implicit-def dead $mode, implicit $mode
; GCN: $vcc = COPY %7
; GCN: %20:vgpr_32 = nnan nofpexcept V_DIV_FMAS_F32 0, killed %19, 0, %15, 0, %18, 0, 0, implicit $mode, implicit $vcc, implicit $exec
; GCN: %21:vgpr_32 = nnan nofpexcept V_DIV_FIXUP_F32 0, killed %20, 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: %20:vgpr_32 = nnan nofpexcept V_DIV_FMAS_F32_e64 0, killed %19, 0, %15, 0, %18, 0, 0, implicit $mode, implicit $vcc, implicit $exec
; GCN: %21:vgpr_32 = nnan nofpexcept V_DIV_FIXUP_F32_e64 0, killed %20, 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $mode, implicit $exec
; GCN: [[COPY3:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY]]
; GCN: $vgpr0 = COPY %21
; GCN: [[COPY4:%[0-9]+]]:ccr_sgpr_64 = COPY [[COPY3]]
Expand Down
16 changes: 8 additions & 8 deletions llvm/test/CodeGen/AMDGPU/fold-immediate-output-mods.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ body: |
%13 = REG_SEQUENCE killed %12, 1, killed %11, 2
%14 = REG_SEQUENCE killed %5, 17, %13, 18
%15 = S_MOV_B32 2
%29 = V_LSHL_B64 killed %28, killed %15, implicit $exec
%29 = V_LSHL_B64_e64 killed %28, killed %15, implicit $exec
%17 = REG_SEQUENCE killed %6, 17, %13, 18
%18 = REG_SEQUENCE killed %4, 17, %13, 18
%20 = COPY %29
Expand Down Expand Up @@ -127,7 +127,7 @@ body: |
%13 = REG_SEQUENCE killed %12, 1, killed %11, 2
%14 = REG_SEQUENCE killed %5, 17, %13, 18
%15 = S_MOV_B32 2
%29 = V_LSHL_B64 killed %28, killed %15, implicit $exec
%29 = V_LSHL_B64_e64 killed %28, killed %15, implicit $exec
%17 = REG_SEQUENCE killed %6, 17, %13, 18
%18 = REG_SEQUENCE killed %4, 17, %13, 18
%20 = COPY %29
Expand All @@ -144,7 +144,7 @@ body: |
---
# GCN: name: no_fold_imm_madak_mad_clamp_f32
# GCN: %23:vgpr_32 = V_MOV_B32_e32 1090519040, implicit $exec
# GCN: %24:vgpr_32 = nofpexcept V_MAD_F32 0, killed %19, 0, killed %21, 0, %23, 1, 0, implicit $mode, implicit $exec
# GCN: %24:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %19, 0, killed %21, 0, %23, 1, 0, implicit $mode, implicit $exec

name: no_fold_imm_madak_mad_clamp_f32
tracksRegLiveness: true
Expand Down Expand Up @@ -198,15 +198,15 @@ body: |
%13 = REG_SEQUENCE killed %12, 1, killed %11, 2
%14 = REG_SEQUENCE killed %5, 17, %13, 18
%15 = S_MOV_B32 2
%29 = V_LSHL_B64 killed %28, killed %15, implicit $exec
%29 = V_LSHL_B64_e64 killed %28, killed %15, implicit $exec
%17 = REG_SEQUENCE killed %6, 17, %13, 18
%18 = REG_SEQUENCE killed %4, 17, %13, 18
%20 = COPY %29
%19 = BUFFER_LOAD_DWORD_ADDR64 %20, killed %14, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%22 = COPY %29
%21 = BUFFER_LOAD_DWORD_ADDR64 %22, killed %17, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%23 = V_MOV_B32_e32 1090519040, implicit $exec
%24 = nofpexcept V_MAD_F32 0, killed %19, 0, killed %21, 0, %23, 1, 0, implicit $mode, implicit $exec
%24 = nofpexcept V_MAD_F32_e64 0, killed %19, 0, killed %21, 0, %23, 1, 0, implicit $mode, implicit $exec
%26 = COPY %29
BUFFER_STORE_DWORD_ADDR64 killed %24, %26, killed %18, 0, 0, 0, 0, 0, 0, 0, implicit $exec
S_ENDPGM 0
Expand All @@ -215,7 +215,7 @@ body: |
---
# GCN: name: no_fold_imm_madak_mad_omod_f32
# GCN: %23:vgpr_32 = V_MOV_B32_e32 1090519040, implicit $exec
# GCN: %24:vgpr_32 = nofpexcept V_MAD_F32 0, killed %19, 0, killed %21, 0, %23, 0, 1, implicit $mode, implicit $exec
# GCN: %24:vgpr_32 = nofpexcept V_MAD_F32_e64 0, killed %19, 0, killed %21, 0, %23, 0, 1, implicit $mode, implicit $exec

name: no_fold_imm_madak_mad_omod_f32
tracksRegLiveness: true
Expand Down Expand Up @@ -269,15 +269,15 @@ body: |
%13 = REG_SEQUENCE killed %12, 1, killed %11, 2
%14 = REG_SEQUENCE killed %5, 17, %13, 18
%15 = S_MOV_B32 2
%29 = V_LSHL_B64 killed %28, killed %15, implicit $exec
%29 = V_LSHL_B64_e64 killed %28, killed %15, implicit $exec
%17 = REG_SEQUENCE killed %6, 17, %13, 18
%18 = REG_SEQUENCE killed %4, 17, %13, 18
%20 = COPY %29
%19 = BUFFER_LOAD_DWORD_ADDR64 %20, killed %14, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%22 = COPY %29
%21 = BUFFER_LOAD_DWORD_ADDR64 %22, killed %17, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%23 = V_MOV_B32_e32 1090519040, implicit $exec
%24 = nofpexcept V_MAD_F32 0, killed %19, 0, killed %21, 0, %23, 0, 1, implicit $mode, implicit $exec
%24 = nofpexcept V_MAD_F32_e64 0, killed %19, 0, killed %21, 0, %23, 0, 1, implicit $mode, implicit $exec
%26 = COPY %29
BUFFER_STORE_DWORD_ADDR64 killed %24, %26, killed %18, 0, 0, 0, 0, 0, 0, 0, implicit $exec
S_ENDPGM 0
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/fold-vgpr-copy.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,7 @@ body: |

# GCN-LABEL: name: fma_sgpr_use
# GCN: %0:sreg_64_xexec = IMPLICIT_DEF
# GCN-NEXT: %4:vgpr_32 = nnan ninf nsz arcp contract afn reassoc V_FMA_F32 2, %0.sub0, 0, 1073741824, 0, %0.sub1, 0, 0, implicit $mode, implicit $exec
# GCN-NEXT: %4:vgpr_32 = nnan ninf nsz arcp contract afn reassoc V_FMA_F32_e64 2, %0.sub0, 0, 1073741824, 0, %0.sub1, 0, 0, implicit $mode, implicit $exec
---
name: fma_sgpr_use
body: |
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/fold_16bit_imm.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ body: |
; GCN-LABEL: name: fold_aimm_16_sub_to_phys
; GCN: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 0
; GCN: $agpr0 = V_ACCVGPR_WRITE_B32 0, implicit $exec
; GCN: $agpr0 = V_ACCVGPR_WRITE_B32_e64 0, implicit $exec
; GCN: SI_RETURN_TO_EPILOG $agpr0_lo16
%0:sreg_32 = S_MOV_B32 0
$agpr0_lo16 = COPY killed %0.lo16
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/hazard-pass-ordering.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ name: mai_hazard_pass_ordering_optimize_vcc_branch
body: |
bb.0:
$vgpr0 = V_MOV_B32_e32 1, implicit $exec
$vgpr2 = V_ACCVGPR_READ_B32 killed $agpr0, implicit $exec
$vgpr2 = V_ACCVGPR_READ_B32_e64 killed $agpr0, implicit $exec
$sgpr8_sgpr9 = S_MOV_B64 -1
$vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
$vcc = S_ANDN2_B64 $exec, killed renamable $sgpr8_sgpr9, implicit-def dead $scc
Expand Down
10 changes: 5 additions & 5 deletions llvm/test/CodeGen/AMDGPU/inserted-wait-states.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -41,22 +41,22 @@ name: div_fmas
body: |
bb.0:
$vcc = S_MOV_B64 0
$vgpr0 = V_DIV_FMAS_F32 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
$vgpr0 = V_DIV_FMAS_F32_e64 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
S_BRANCH %bb.1
bb.1:
implicit $vcc = V_CMP_EQ_I32_e32 $vgpr1, $vgpr2, implicit $exec
$vgpr0 = V_DIV_FMAS_F32 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
$vgpr0 = V_DIV_FMAS_F32_e64 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
S_BRANCH %bb.2
bb.2:
$vcc = V_CMP_EQ_I32_e64 $vgpr1, $vgpr2, implicit $exec
$vgpr0 = V_DIV_FMAS_F32 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
$vgpr0 = V_DIV_FMAS_F32_e64 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
S_BRANCH %bb.3
bb.3:
$vgpr4, $vcc = V_DIV_SCALE_F32 0, $vgpr1, 0, $vgpr1, 0, $vgpr3, 0, 0, implicit $mode, implicit $exec
$vgpr0 = V_DIV_FMAS_F32 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
$vgpr4, $vcc = V_DIV_SCALE_F32_e64 0, $vgpr1, 0, $vgpr1, 0, $vgpr3, 0, 0, implicit $mode, implicit $exec
$vgpr0 = V_DIV_FMAS_F32_e64 0, $vgpr1, 0, $vgpr2, 0, $vgpr3, 0, 0, implicit $mode, implicit $vcc, implicit $exec
S_ENDPGM 0
...
Expand Down
4 changes: 4 additions & 0 deletions llvm/test/CodeGen/AMDGPU/llvm.amdgcn.fma.legacy.ll
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@ define float @v_fma(float %a, float %b, float %c) {
; GCN-NEXT: v_fmac_legacy_f32_e64 v2, v0, v1
; GCN-NEXT: v_mov_b32_e32 v0, v2
; GCN-NEXT: s_setpc_b64 s[30:31]
;
%fma = call float @llvm.amdgcn.fma.legacy(float %a, float %b, float %c)
ret float %fma
}
Expand All @@ -21,6 +22,7 @@ define float @v_fabs_fma(float %a, float %b, float %c) {
; GCN-NEXT: s_waitcnt_vscnt null, 0x0
; GCN-NEXT: v_fma_legacy_f32 v0, |v0|, v1, v2
; GCN-NEXT: s_setpc_b64 s[30:31]
;
%fabs.a = call float @llvm.fabs.f32(float %a)
%fma = call float @llvm.amdgcn.fma.legacy(float %fabs.a, float %b, float %c)
ret float %fma
Expand All @@ -33,6 +35,7 @@ define float @v_fneg_fabs_fma(float %a, float %b, float %c) {
; GCN-NEXT: s_waitcnt_vscnt null, 0x0
; GCN-NEXT: v_fma_legacy_f32 v0, v0, -|v1|, v2
; GCN-NEXT: s_setpc_b64 s[30:31]
;
%fabs.b = call float @llvm.fabs.f32(float %b)
%neg.fabs.b = fneg float %fabs.b
%fma = call float @llvm.amdgcn.fma.legacy(float %a, float %neg.fabs.b, float %c)
Expand All @@ -46,6 +49,7 @@ define float @v_fneg_fma(float %a, float %b, float %c) {
; GCN-NEXT: s_waitcnt_vscnt null, 0x0
; GCN-NEXT: v_fma_legacy_f32 v0, v0, v1, -v2
; GCN-NEXT: s_setpc_b64 s[30:31]
;
%neg.c = fneg float %c
%fma = call float @llvm.amdgcn.fma.legacy(float %a, float %b, float %neg.c)
ret float %fma
Expand Down
166 changes: 83 additions & 83 deletions llvm/test/CodeGen/AMDGPU/mai-hazards.mir
View file

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/memory-legalizer-atomic-insert-end.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ body: |
$sgpr4_sgpr5 = S_LOAD_DWORDX2_IMM $sgpr0_sgpr1, 11, 0, 0 :: (non-temporal dereferenceable invariant load 8 from `i64 addrspace(4)* undef`)
$vgpr1 = V_ASHRREV_I32_e32 31, $vgpr0, implicit $exec
$vgpr1_vgpr2 = V_LSHL_B64 $vgpr0_vgpr1, 3, implicit $exec
$vgpr1_vgpr2 = V_LSHL_B64_e64 $vgpr0_vgpr1, 3, implicit $exec
$sgpr7 = S_MOV_B32 61440
$sgpr6 = S_MOV_B32 0
S_WAITCNT 127
Expand Down
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/merge-load-store-vreg.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -115,7 +115,7 @@ body: |
%2:vgpr_32 = V_CNDMASK_B32_e64 0, 0, 0, 1, %1, implicit $exec
V_CMP_NE_U32_e32 1, %2, implicit-def $vcc, implicit $exec
DS_WRITE_B32 %0.sub0, %0.sub0, 1024, 0, implicit $m0, implicit $exec :: (store 4 into %ir.tmp)
%3:vreg_64 = V_LSHLREV_B64 0, 0, implicit $exec
%3:vreg_64 = V_LSHLREV_B64_e64 0, 0, implicit $exec
DS_WRITE_B32 %0.sub0, %3.sub0, 1056, 0, implicit $m0, implicit $exec :: (store 4 into %ir.tmp1)
%4:vgpr_32 = DS_READ_B32 %3.sub0, 1088, 0, implicit $m0, implicit $exec :: (load 4 from %ir.tmp2)
%5:vgpr_32 = DS_READ_B32 %3.sub0, 1120, 0, implicit $m0, implicit $exec :: (load 4 from %ir.tmp3)
Expand Down Expand Up @@ -146,7 +146,7 @@ body: |
%2:vgpr_32 = V_CNDMASK_B32_e64 0, 0, 0, 1, %1, implicit $exec
V_CMP_NE_U32_e32 1, %2, implicit-def $vcc, implicit $exec
DS_WRITE_B32 %0.sub0, %0.sub0, 0, 0, implicit $m0, implicit $exec :: (store 4 into %ir.tmp)
%3:vreg_64 = V_LSHLREV_B64 0, 0, implicit $exec
%3:vreg_64 = V_LSHLREV_B64_e64 0, 0, implicit $exec
DS_WRITE_B32 %0.sub0, %3.sub0, 32, 0, implicit $m0, implicit $exec :: (store 4 into %ir.tmp1)
%4:vgpr_32 = DS_READ_B32 %3.sub0, 0, 0, implicit $m0, implicit $exec :: (load 4 from %ir.tmp2)
%5:vgpr_32 = DS_READ_B32 %3.sub0, 32, 0, implicit $m0, implicit $exec :: (load 4 from %ir.tmp3)
Expand Down
176 changes: 88 additions & 88 deletions llvm/test/CodeGen/AMDGPU/pei-build-spill-partial-agpr.mir
View file

Large diffs are not rendered by default.

1,848 changes: 924 additions & 924 deletions llvm/test/CodeGen/AMDGPU/pei-build-spill.mir
View file

Large diffs are not rendered by default.

4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/power-sched-no-instr-sunit.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,8 +14,8 @@ body: |
$sgpr10_sgpr11 = S_MOV_B64 $sgpr2_sgpr3, implicit-def $sgpr8_sgpr9_sgpr10_sgpr11, implicit $sgpr0_sgpr1_sgpr2_sgpr3
$sgpr8_sgpr9 = S_MOV_B64 $sgpr0_sgpr1, implicit killed $sgpr0_sgpr1_sgpr2_sgpr3
S_BARRIER
$agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X1F32 undef $vgpr0, undef $vgpr0, 0, 0, 0, 2, implicit $mode, implicit $exec
$vgpr0 = V_ACCVGPR_READ_B32 $agpr31, implicit $exec
$agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X1F32_e64 undef $vgpr0, undef $vgpr0, 0, 0, 0, 2, implicit $mode, implicit $exec
$vgpr0 = V_ACCVGPR_READ_B32_e64 $agpr31, implicit $exec
BUFFER_STORE_DWORD_OFFEN killed $vgpr0, undef $vgpr0, $sgpr8_sgpr9_sgpr10_sgpr11, $sgpr6, 0, 0, 0, 0, 0, 0, implicit $exec
...
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/promote-constOffset-to-imm-gfx10.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_32_xm0_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_32_xm0_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_32_xm0_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
%25:sgpr_32 = S_MOV_B32 4096
Expand Down Expand Up @@ -80,7 +80,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_32_xm0_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_32_xm0_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_32_xm0_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
%25:sgpr_32 = S_MOV_B32 8000
Expand Down Expand Up @@ -137,7 +137,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_32_xm0_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_32_xm0_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_32_xm0_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
%25:sgpr_32 = S_MOV_B32 6144
Expand Down Expand Up @@ -182,7 +182,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_32_xm0_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_32_xm0_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_32_xm0_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/promote-constOffset-to-imm.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_64_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_64_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_64_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
%25:sgpr_32 = S_MOV_B32 4096
Expand Down Expand Up @@ -80,7 +80,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_64_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_64_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_64_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
%25:sgpr_32 = S_MOV_B32 8000
Expand Down Expand Up @@ -133,7 +133,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_64_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_64_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_64_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
%25:sgpr_32 = S_MOV_B32 6144
Expand Down Expand Up @@ -178,7 +178,7 @@ body: |
%16:vgpr_32 = COPY %12
%17:vgpr_32, dead %18:sreg_64_xexec = V_ADDC_U32_e64 %16, %13, killed %15, 0, implicit $exec
%19:vreg_64 = REG_SEQUENCE %14, %subreg.sub0, %17, %subreg.sub1
%20:vreg_64 = V_LSHLREV_B64 3, %9, implicit $exec
%20:vreg_64 = V_LSHLREV_B64_e64 3, %9, implicit $exec
%21:vgpr_32, %22:sreg_64_xexec = V_ADD_CO_U32_e64 %14, %20.sub0, 0, implicit $exec
%23:vgpr_32, dead %24:sreg_64_xexec = V_ADDC_U32_e64 %17, %20.sub1, killed %22, 0, implicit $exec
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/regbank-reassign.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -530,8 +530,8 @@ body: |
# GCN: $vgpr36_vgpr37_vgpr38_vgpr39 = IMPLICIT_DEF
# GCN: $vgpr40_vgpr41_vgpr42_vgpr43 = IMPLICIT_DEF
# GCN: $vgpr44_vgpr45_vgpr46_vgpr47 = IMPLICIT_DEF
# GCN: $vgpr0_vgpr1 = V_ADD_F64 0, $vgpr11_vgpr12, 0, killed $vgpr16_vgpr17, 0, 0, implicit $mode, implicit $exec
# GCN: $vgpr0_vgpr1 = V_ADD_F64 0, $vgpr9_vgpr10, 0, killed $vgpr14_vgpr15, 0, 0, implicit $mode, implicit $exec
# GCN: $vgpr0_vgpr1 = V_ADD_F64_e64 0, $vgpr11_vgpr12, 0, killed $vgpr16_vgpr17, 0, 0, implicit $mode, implicit $exec
# GCN: $vgpr0_vgpr1 = V_ADD_F64_e64 0, $vgpr9_vgpr10, 0, killed $vgpr14_vgpr15, 0, 0, implicit $mode, implicit $exec
---
name: vgpr_sub_dependence
tracksRegLiveness: true
Expand Down Expand Up @@ -568,7 +568,7 @@ body: |
%13 = IMPLICIT_DEF
%14 = IMPLICIT_DEF
%15 = IMPLICIT_DEF
%3 = V_ADD_F64 0, %0.sub2_sub3:vreg_128, 0, %1:vreg_64, 0, 0, implicit $mode, implicit $exec
%4 = V_ADD_F64 0, %0.sub0_sub1:vreg_128, 0, %2:vreg_64, 0, 0, implicit $mode, implicit $exec
%3 = V_ADD_F64_e64 0, %0.sub2_sub3:vreg_128, 0, %1:vreg_64, 0, 0, implicit $mode, implicit $exec
%4 = V_ADD_F64_e64 0, %0.sub0_sub1:vreg_128, 0, %2:vreg_64, 0, 0, implicit $mode, implicit $exec
S_ENDPGM 0
...
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/regcoal-subrange-join-seg.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -184,7 +184,7 @@ body: |
bb.28:
%9 = S_FF1_I32_B32 undef %10
%13 = V_MAD_U32_U24 killed %9, 48, 32, 0, implicit $exec
%13 = V_MAD_U32_U24_e64 killed %9, 48, 32, 0, implicit $exec
%45 = BUFFER_LOAD_DWORD_OFFEN killed %13, undef %15, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable invariant load 4)
%46 = V_AND_B32_e32 1, killed %45, implicit $exec
%21 = S_BUFFER_LOAD_DWORD_SGPR undef %22, undef %23, 0, 0 :: (dereferenceable invariant load 4)
Expand All @@ -211,7 +211,7 @@ body: |
S_BRANCH %bb.31
bb.30:
%33 = nofpexcept V_MAD_F32 1, killed %53.sub0, 0, undef %34, 0, 0, 0, 0, implicit $mode, implicit $exec
%33 = nofpexcept V_MAD_F32_e64 1, killed %53.sub0, 0, undef %34, 0, 0, 0, 0, implicit $mode, implicit $exec
%35 = nofpexcept V_MAC_F32_e32 killed %33, undef %36, undef %35, implicit $mode, implicit $exec
%38 = nofpexcept V_MAX_F32_e32 0, killed %35, implicit $mode, implicit $exec
%39 = nofpexcept V_LOG_F32_e32 killed %38, implicit $mode, implicit $exec
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/regcoalesce-dbg.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -68,7 +68,7 @@ body: |
%12.sub1 = COPY killed %10
undef %13.sub0_sub1 = COPY killed %4
%13.sub2_sub3 = COPY killed %12
%20 = V_LSHL_B64 killed %19, 2, implicit $exec
%20 = V_LSHL_B64_e64 killed %19, 2, implicit $exec
%16 = COPY killed %5
BUFFER_STORE_DWORD_ADDR64 killed %16, killed %20, killed %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (store 4 into %ir.out)
S_ENDPGM 0
Expand Down
6 changes: 3 additions & 3 deletions llvm/test/CodeGen/AMDGPU/regcoalescing-remove-partial-redundancy-assert.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -48,8 +48,8 @@ body: |
bb.6:
%36:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM undef %37:sgpr_128, 2708, 0, 0 :: (dereferenceable invariant load 4)
%39:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32 0, killed %110.sub1, 0, target-flags(amdgpu-gotprel32-lo) 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%40:vgpr_32 = nofpexcept V_MAD_F32 0, %111.sub1, 0, target-flags(amdgpu-gotprel32-lo) 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%39:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32_e64 0, killed %110.sub1, 0, target-flags(amdgpu-gotprel32-lo) 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%40:vgpr_32 = nofpexcept V_MAD_F32_e64 0, %111.sub1, 0, target-flags(amdgpu-gotprel32-lo) 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%41:vgpr_32 = nofpexcept V_MUL_F32_e64 0, 0, 0, killed %40, 1, 0, implicit $mode, implicit $exec
%43:vgpr_32 = nofpexcept V_MUL_F32_e32 0, %39, implicit $mode, implicit $exec
%44:vgpr_32 = COPY killed %43
Expand Down Expand Up @@ -159,7 +159,7 @@ body: |
bb.28:
dead %77:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
%78:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32 0, killed %113.sub1, 0, target-flags(amdgpu-gotprel32-lo) 0, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
%78:vgpr_32 = nnan arcp contract reassoc nofpexcept V_MAD_F32_e64 0, killed %113.sub1, 0, target-flags(amdgpu-gotprel32-lo) 0, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
dead %80:sreg_32_xm0 = S_MOV_B32 0
dead %82:vgpr_32 = nofpexcept V_MUL_F32_e32 killed %78, %78, implicit $mode, implicit $exec
dead %126:vgpr_32 = V_MOV_B32_e32 2143289344, implicit $exec
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/rename-independent-subregs.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -80,7 +80,7 @@ body: |
undef %0.sub0:vreg_64 = IMPLICIT_DEF
bb.1:
undef %0.sub1:vreg_64 = V_ALIGNBIT_B32 %0.sub0:vreg_64, %0.sub0:vreg_64, 16, implicit $exec
undef %0.sub1:vreg_64 = V_ALIGNBIT_B32_e64 %0.sub0:vreg_64, %0.sub0:vreg_64, 16, implicit $exec
INLINEASM &"", 32, 327690, def undef %0.sub0:vreg_64, 327690, def %0.sub1:vreg_64, 2147483657, undef %0.sub0:vreg_64(tied-def 3), 2147549193, %0.sub1:vreg_64(tied-def 5)
S_BRANCH %bb.1
Expand Down
14 changes: 7 additions & 7 deletions llvm/test/CodeGen/AMDGPU/s_add_co_pseudo_lowering.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,18 +17,18 @@ body: |
; GCN: [[COPY4:%[0-9]+]]:sreg_32 = COPY $sgpr1
; GCN: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2
; GCN: [[COPY6:%[0-9]+]]:sgpr_32 = COPY [[COPY3]]
; GCN: [[V_MUL_LO_U32_:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32 [[COPY]], [[COPY4]], implicit $exec
; GCN: [[V_ADD_CO_U32_e64_:%[0-9]+]]:vgpr_32, [[V_ADD_CO_U32_e64_1:%[0-9]+]]:sreg_64_xexec = V_ADD_CO_U32_e64 killed [[V_MUL_LO_U32_]], [[COPY6]], 0, implicit $exec
; GCN: [[V_MUL_LO_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32_e64 [[COPY]], [[COPY4]], implicit $exec
; GCN: [[V_ADD_CO_U32_e64_:%[0-9]+]]:vgpr_32, [[V_ADD_CO_U32_e64_1:%[0-9]+]]:sreg_64_xexec = V_ADD_CO_U32_e64 killed [[V_MUL_LO_U32_e64_]], [[COPY6]], 0, implicit $exec
; GCN: [[S_MUL_HI_U32_:%[0-9]+]]:sreg_32 = S_MUL_HI_U32 [[COPY4]], [[COPY5]]
; GCN: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 -614296167
; GCN: [[V_MUL_LO_U32_1:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32 [[COPY]], [[COPY3]], implicit $exec
; GCN: [[V_MUL_LO_U32_e64_1:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32_e64 [[COPY]], [[COPY3]], implicit $exec
; GCN: [[COPY7:%[0-9]+]]:vgpr_32 = COPY killed [[S_MOV_B32_]]
; GCN: [[V_ADDC_U32_e64_:%[0-9]+]]:vgpr_32, [[V_ADDC_U32_e64_1:%[0-9]+]]:sreg_64_xexec = V_ADDC_U32_e64 killed [[V_MUL_LO_U32_1]], [[COPY7]], [[V_ADD_CO_U32_e64_1]], 0, implicit $exec
; GCN: [[V_MUL_HI_U32_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32 [[COPY4]], [[V_ADDC_U32_e64_]], implicit $exec
; GCN: [[V_ADDC_U32_e64_:%[0-9]+]]:vgpr_32, [[V_ADDC_U32_e64_1:%[0-9]+]]:sreg_64_xexec = V_ADDC_U32_e64 killed [[V_MUL_LO_U32_e64_1]], [[COPY7]], [[V_ADD_CO_U32_e64_1]], 0, implicit $exec
; GCN: [[V_MUL_HI_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_HI_U32_e64 [[COPY4]], [[V_ADDC_U32_e64_]], implicit $exec
; GCN: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 -181084736
; GCN: [[V_MUL_LO_U32_2:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32 [[V_MUL_HI_U32_]], [[S_MOV_B32_1]], implicit $exec
; GCN: [[V_MUL_LO_U32_e64_2:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32_e64 [[V_MUL_HI_U32_e64_]], [[S_MOV_B32_1]], implicit $exec
; GCN: [[COPY8:%[0-9]+]]:vgpr_32 = COPY killed [[S_MOV_B32_1]]
; GCN: [[V_ADDC_U32_e64_2:%[0-9]+]]:vgpr_32, [[V_ADDC_U32_e64_3:%[0-9]+]]:sreg_64_xexec = V_ADDC_U32_e64 [[COPY8]], killed [[V_MUL_LO_U32_2]], [[V_ADDC_U32_e64_1]], 0, implicit $exec
; GCN: [[V_ADDC_U32_e64_2:%[0-9]+]]:vgpr_32, [[V_ADDC_U32_e64_3:%[0-9]+]]:sreg_64_xexec = V_ADDC_U32_e64 [[COPY8]], killed [[V_MUL_LO_U32_e64_2]], [[V_ADDC_U32_e64_1]], 0, implicit $exec
%0:vgpr_32 = COPY $vgpr0
%6:sreg_32 = COPY %0
%1:vgpr_32 = COPY $vgpr1
Expand Down
6 changes: 3 additions & 3 deletions llvm/test/CodeGen/AMDGPU/sched-assert-onlydbg-value-empty-region.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -49,9 +49,9 @@ body: |
; CHECK: [[DEF1]].sub0:vreg_64 = GLOBAL_LOAD_DWORD [[DEF3]], 0, 0, 0, 0, implicit $exec
; CHECK: dead %20:vgpr_32 = GLOBAL_LOAD_DWORD %11, 0, 0, 0, 0, implicit $exec
; CHECK: dead %21:vgpr_32 = GLOBAL_LOAD_DWORD [[DEF6]], 0, 0, 0, 0, implicit $exec
; CHECK: [[V_LSHLREV_B64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64 2, [[DEF1]], implicit $exec
; CHECK: [[V_LSHLREV_B64_e64_:%[0-9]+]]:vreg_64 = V_LSHLREV_B64_e64 2, [[DEF1]], implicit $exec
; CHECK: dead %22:vgpr_32 = GLOBAL_LOAD_DWORD [[DEF7]], 0, 0, 0, 0, implicit $exec
; CHECK: S_NOP 0, implicit [[DEF5]], implicit [[V_LSHLREV_B64_]].sub0, implicit [[DEF4]], implicit [[V_MOV_B32_e32_]]
; CHECK: S_NOP 0, implicit [[DEF5]], implicit [[V_LSHLREV_B64_e64_]].sub0, implicit [[DEF4]], implicit [[V_MOV_B32_e32_]]
; CHECK: GLOBAL_STORE_DWORD [[DEF7]], [[V_MOV_B32_e32_1]], 0, 0, 0, 0, implicit $exec
; CHECK: bb.1:
; CHECK: successors: %bb.2(0x80000000)
Expand Down Expand Up @@ -96,7 +96,7 @@ body: |
%20:vgpr_32 = GLOBAL_LOAD_DWORD %11, 0, 0, 0, 0, implicit $exec
%21:vgpr_32 = GLOBAL_LOAD_DWORD %14, 0, 0, 0, 0, implicit $exec
%22:vgpr_32 = GLOBAL_LOAD_DWORD %15, 0, 0, 0, 0, implicit $exec
%23:vreg_64 = V_LSHLREV_B64 2, %8, implicit $exec
%23:vreg_64 = V_LSHLREV_B64_e64 2, %8, implicit $exec
S_NOP 0, implicit %13, implicit %23.sub0, implicit %12, implicit %17
GLOBAL_STORE_DWORD %15, %18, 0, 0, 0, 0, implicit $exec
Expand Down
34 changes: 17 additions & 17 deletions llvm/test/CodeGen/AMDGPU/sched-crash-dbg-value.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -208,17 +208,17 @@ body: |
%10:sreg_64_xexec = S_LOAD_DWORDX2_IMM %3, 4, 0, 0
%11:sreg_32_xm0 = S_LSHR_B32 %10.sub0, 16, implicit-def dead $scc
%12:sreg_32_xm0 = S_MUL_I32 %11, %10.sub1
%13:vgpr_32 = V_MUL_LO_I32 0, %0, implicit $exec
%14:vgpr_32 = V_MUL_LO_I32 %1, %10.sub1, implicit $exec
%13:vgpr_32 = V_MUL_LO_I32_e64 0, %0, implicit $exec
%14:vgpr_32 = V_MUL_LO_I32_e64 %1, %10.sub1, implicit $exec
%15:vgpr_32 = V_ADD_CO_U32_e32 0, %13, implicit-def dead $vcc, implicit $exec
%16:vgpr_32 = V_ADD_CO_U32_e32 0, %15, implicit-def dead $vcc, implicit $exec
%17:vgpr_32 = IMPLICIT_DEF
%18:sreg_64 = S_MOV_B64 0
%19:sreg_32_xm0_xexec = IMPLICIT_DEF
%20:vgpr_32 = V_ADD_CO_U32_e32 %19, %0, implicit-def dead $vcc, implicit $exec
%21:vreg_64, dead %22:sreg_64 = V_MAD_I64_I32 %20, 12, %7, 0, implicit $exec
%21:vreg_64, dead %22:sreg_64 = V_MAD_I64_I32_e64 %20, 12, %7, 0, implicit $exec
%23:vgpr_32 = GLOBAL_LOAD_DWORD %21, 4, 0, 0, 0, implicit $exec
%24:vreg_64, dead %25:sreg_64 = V_MAD_I64_I32 %20, 48, %8, 0, implicit $exec
%24:vreg_64, dead %25:sreg_64 = V_MAD_I64_I32_e64 %20, 48, %8, 0, implicit $exec
%26:vreg_128 = IMPLICIT_DEF
undef %27.sub0:sreg_64_xexec = S_LOAD_DWORD_IMM %6, 0, 0, 0
%27.sub1:sreg_64_xexec = S_MOV_B32 0
Expand All @@ -231,25 +231,25 @@ body: |
%32:sreg_32_xm0 = S_ADD_U32 0, %31.sub0, implicit-def $scc
%33:sgpr_32 = S_ADDC_U32 %5.sub1, %31.sub1, implicit-def dead $scc, implicit killed $scc
%34:vgpr_32 = IMPLICIT_DEF
%35:vreg_64, dead %36:sreg_64 = V_MAD_I64_I32 %23, %34, 0, 0, implicit $exec
%35:vreg_64, dead %36:sreg_64 = V_MAD_I64_I32_e64 %23, %34, 0, 0, implicit $exec
%37:vreg_64 = GLOBAL_LOAD_DWORDX2 %35, 32, 0, 0, 0, implicit $exec
undef %38.sub1:vreg_64 = V_ASHRREV_I32_e32 31, %37.sub0, implicit $exec
%38.sub0:vreg_64 = COPY %37.sub0
%39:vreg_64 = V_LSHLREV_B64 3, %38, implicit $exec
%39:vreg_64 = V_LSHLREV_B64_e64 3, %38, implicit $exec
undef %40.sub0:vreg_64, %41:sreg_64_xexec = V_ADD_CO_U32_e64 0, %39.sub0, 0, implicit $exec
%42:vgpr_32 = COPY %33
%40.sub1:vreg_64, dead %43:sreg_64_xexec = V_ADDC_U32_e64 %42, %39.sub1, %41, 0, implicit $exec
%44:vreg_64 = GLOBAL_LOAD_DWORDX2 %40, 0, 0, 0, 0, implicit $exec :: (load 8 from %ir.tmp34)
undef %45.sub1:vreg_64 = IMPLICIT_DEF
%45.sub0:vreg_64 = COPY %37.sub1
%46:vreg_64 = V_LSHLREV_B64 3, %45, implicit $exec
%46:vreg_64 = V_LSHLREV_B64_e64 3, %45, implicit $exec
undef %47.sub0:vreg_64, %48:sreg_64_xexec = V_ADD_CO_U32_e64 %32, %46.sub0, 0, implicit $exec
%49:vgpr_32 = COPY %33
%47.sub1:vreg_64, dead %50:sreg_64_xexec = V_ADDC_U32_e64 %49, %46.sub1, %48, 0, implicit $exec
%51:vreg_64 = IMPLICIT_DEF
undef %52.sub0:vreg_64 = GLOBAL_LOAD_DWORD %35, 40, 0, 0, 0, implicit $exec :: (load 4 from %ir.18 + 8)
%52.sub1:vreg_64 = IMPLICIT_DEF
%53:vreg_64 = V_LSHLREV_B64 3, %52, implicit $exec
%53:vreg_64 = V_LSHLREV_B64_e64 3, %52, implicit $exec
undef %54.sub0:vreg_64, %55:sreg_64_xexec = V_ADD_CO_U32_e64 0, %53.sub0, 0, implicit $exec
%56:vgpr_32 = COPY %33
%54.sub1:vreg_64, dead %57:sreg_64_xexec = V_ADDC_U32_e64 0, %53.sub1, %55, 0, implicit $exec
Expand Down Expand Up @@ -288,9 +288,9 @@ body: |
%87:vgpr_32 = IMPLICIT_DEF
%88:vgpr_32 = IMPLICIT_DEF
%90:vgpr_32 = IMPLICIT_DEF
%91:vgpr_32, dead %92:sreg_64 = nofpexcept V_DIV_SCALE_F32 0, %90, 0, %90, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
%95:vgpr_32 = nofpexcept V_FMA_F32 0, 0, 0, 0, 0, undef %93:vgpr_32, 0, 0, implicit $mode, implicit $exec
%96:vgpr_32, %97:sreg_64 = nofpexcept V_DIV_SCALE_F32 0, 1065353216, 0, %90, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
%91:vgpr_32, dead %92:sreg_64 = nofpexcept V_DIV_SCALE_F32_e64 0, %90, 0, %90, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
%95:vgpr_32 = nofpexcept V_FMA_F32_e64 0, 0, 0, 0, 0, undef %93:vgpr_32, 0, 0, implicit $mode, implicit $exec
%96:vgpr_32, %97:sreg_64 = nofpexcept V_DIV_SCALE_F32_e64 0, 1065353216, 0, %90, 0, 1065353216, 0, 0, implicit $mode, implicit $exec
%98:vgpr_32 = IMPLICIT_DEF
%99:vgpr_32 = IMPLICIT_DEF
%100:vgpr_32 = IMPLICIT_DEF
Expand All @@ -299,18 +299,18 @@ body: |
%103:vgpr_32 = IMPLICIT_DEF
%104:vgpr_32 = IMPLICIT_DEF
%105:vgpr_32 = IMPLICIT_DEF
%106:vgpr_32, dead %107:sreg_64 = nofpexcept V_DIV_SCALE_F32 0, %90, 0, %90, 0, %105, 0, 0, implicit $mode, implicit $exec
%106:vgpr_32, dead %107:sreg_64 = nofpexcept V_DIV_SCALE_F32_e64 0, %90, 0, %90, 0, %105, 0, 0, implicit $mode, implicit $exec
%108:vgpr_32 = nofpexcept V_RCP_F32_e32 0, implicit $mode, implicit $exec
%109:vgpr_32 = IMPLICIT_DEF
%110:vgpr_32 = nofpexcept V_FMA_F32 0, 0, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%111:vgpr_32, %112:sreg_64 = nofpexcept V_DIV_SCALE_F32 0, 0, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%110:vgpr_32 = nofpexcept V_FMA_F32_e64 0, 0, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%111:vgpr_32, %112:sreg_64 = nofpexcept V_DIV_SCALE_F32_e64 0, 0, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
%113:vgpr_32 = nofpexcept V_MUL_F32_e32 0, %110, implicit $mode, implicit $exec
%114:vgpr_32 = IMPLICIT_DEF
%115:vgpr_32 = IMPLICIT_DEF
%116:vgpr_32 = IMPLICIT_DEF
$vcc = IMPLICIT_DEF
%117:vgpr_32 = nofpexcept V_DIV_FMAS_F32 0, %116, 0, %110, 0, %115, 0, 0, implicit killed $vcc, implicit $mode, implicit $exec
%118:vgpr_32 = nofpexcept V_DIV_FIXUP_F32 0, %117, 0, %90, 0, %105, 0, 0, implicit $mode, implicit $exec
%117:vgpr_32 = nofpexcept V_DIV_FMAS_F32_e64 0, %116, 0, %110, 0, %115, 0, 0, implicit killed $vcc, implicit $mode, implicit $exec
%118:vgpr_32 = nofpexcept V_DIV_FIXUP_F32_e64 0, %117, 0, %90, 0, %105, 0, 0, implicit $mode, implicit $exec
%119:vgpr_32 = IMPLICIT_DEF
%120:vgpr_32 = IMPLICIT_DEF
%121:vgpr_32 = IMPLICIT_DEF
Expand All @@ -328,7 +328,7 @@ body: |
$vgpr3 = COPY %126
dead $sgpr30_sgpr31 = SI_CALL %127, @func, csr_amdgpu_highregs, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr4, implicit $vgpr0, implicit $vgpr1_vgpr2, implicit killed $vgpr3
ADJCALLSTACKDOWN 0, 0, implicit-def $scc, implicit-def $sgpr32, implicit $sgpr32
%128:vreg_64, dead %129:sreg_64 = V_MAD_I64_I32 %20, %34, 0, 0, implicit $exec
%128:vreg_64, dead %129:sreg_64 = V_MAD_I64_I32_e64 %20, %34, 0, 0, implicit $exec
S_ENDPGM 0
...
12 changes: 6 additions & 6 deletions llvm/test/CodeGen/AMDGPU/sched-handleMoveUp-subreg-def-across-subreg-def.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -56,13 +56,13 @@ body: |
; CHECK: [[V_ADD_U32_e32_:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 1, [[DEF2]], implicit $exec
; CHECK: [[COPY1:%[0-9]+]]:vgpr_32 = COPY [[S_LOAD_DWORDX2_IMM]].sub1
; CHECK: [[DEF]].sub1:vreg_64 = COPY [[V_MOV_B32_e32_]]
; CHECK: [[V_MUL_LO_U32_:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32 [[V_ADD_U32_e32_]], [[S_MOV_B32_]], implicit $exec
; CHECK: [[V_MUL_LO_U32_e64_:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32_e64 [[V_ADD_U32_e32_]], [[S_MOV_B32_]], implicit $exec
; CHECK: [[V_CMP_GT_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_GT_U32_e64 64, [[V_ADD_U32_e32_]], implicit $exec
; CHECK: [[V_CNDMASK_B32_e64_:%[0-9]+]]:vgpr_32 = V_CNDMASK_B32_e64 0, 0, 0, [[V_ADD_U32_e32_]], [[V_CMP_GT_U32_e64_]], implicit $exec
; CHECK: [[V_SUB_U32_e32_:%[0-9]+]]:vgpr_32 = V_SUB_U32_e32 [[V_MUL_LO_U32_]], [[DEF1]], implicit $exec
; CHECK: [[V_MUL_LO_U32_1:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32 [[V_CNDMASK_B32_e64_]], [[S_MOV_B32_]], implicit $exec
; CHECK: [[V_SUB_U32_e32_:%[0-9]+]]:vgpr_32 = V_SUB_U32_e32 [[V_MUL_LO_U32_e64_]], [[DEF1]], implicit $exec
; CHECK: [[V_MUL_LO_U32_e64_1:%[0-9]+]]:vgpr_32 = V_MUL_LO_U32_e64 [[V_CNDMASK_B32_e64_]], [[S_MOV_B32_]], implicit $exec
; CHECK: [[V_ADD_U32_e32_1:%[0-9]+]]:vgpr_32 = V_ADD_U32_e32 [[V_SUB_U32_e32_]], [[DEF]].sub0, implicit $exec
; CHECK: [[V_SUB_U32_e32_1:%[0-9]+]]:vgpr_32 = V_SUB_U32_e32 [[V_MUL_LO_U32_1]], [[V_MUL_LO_U32_]], implicit $exec
; CHECK: [[V_SUB_U32_e32_1:%[0-9]+]]:vgpr_32 = V_SUB_U32_e32 [[V_MUL_LO_U32_e64_1]], [[V_MUL_LO_U32_e64_]], implicit $exec
; CHECK: [[DEF]].sub0:vreg_64 = V_ADD_U32_e32 [[V_SUB_U32_e32_1]], [[V_ADD_U32_e32_1]], implicit $exec
; CHECK: undef %38.sub0:vreg_64, %39:sreg_64_xexec = V_ADD_CO_U32_e64 [[S_LOAD_DWORDX2_IMM]].sub0, [[DEF]].sub0, 0, implicit $exec
; CHECK: undef %40.sub1:vreg_64, dead %41:sreg_64_xexec = V_ADDC_U32_e64 [[COPY1]], [[DEF]].sub1, %39, 0, implicit $exec
Expand Down Expand Up @@ -109,10 +109,10 @@ body: |
DS_WRITE_B32_gfx9 undef %29:vgpr_32, %22, 0, 0, implicit $exec :: (store 4, addrspace 3)
DS_WRITE_B64_gfx9 undef %30:vgpr_32, %5, 0, 0, implicit $exec :: (store 8, addrspace 3)
undef %31.sub1:vreg_64 = FLAT_LOAD_DWORD undef %32:vreg_64, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4)
%33:vgpr_32 = V_MUL_LO_U32 %25, %4, implicit $exec
%33:vgpr_32 = V_MUL_LO_U32_e64 %25, %4, implicit $exec
%10:vgpr_32 = V_CNDMASK_B32_e64 0, 0, 0, %25, %26, implicit $exec
%34:vgpr_32 = V_SUB_U32_e32 %33, %9, implicit $exec
%9:vgpr_32 = V_MUL_LO_U32 %10, %4, implicit $exec
%9:vgpr_32 = V_MUL_LO_U32_e64 %10, %4, implicit $exec
%35:vgpr_32 = V_ADD_U32_e32 %34, %8.sub0, implicit $exec
%36:vgpr_32 = V_SUB_U32_e32 %9, %33, implicit $exec
%37:vgpr_32 = COPY %3.sub1
Expand Down
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/sched-prefer-non-mfma.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@
name: unrelated_mfma
body: |
bb.0.entry:
renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X1F32 $vgpr67, $vgpr66, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
renamable $agpr32_agpr33_agpr34_agpr35_agpr36_agpr37_agpr38_agpr39_agpr40_agpr41_agpr42_agpr43_agpr44_agpr45_agpr46_agpr47_agpr48_agpr49_agpr50_agpr51_agpr52_agpr53_agpr54_agpr55_agpr56_agpr57_agpr58_agpr59_agpr60_agpr61_agpr62_agpr63 = V_MFMA_F32_32X32X1F32 $vgpr69, $vgpr68, killed $agpr32_agpr33_agpr34_agpr35_agpr36_agpr37_agpr38_agpr39_agpr40_agpr41_agpr42_agpr43_agpr44_agpr45_agpr46_agpr47_agpr48_agpr49_agpr50_agpr51_agpr52_agpr53_agpr54_agpr55_agpr56_agpr57_agpr58_agpr59_agpr60_agpr61_agpr62_agpr63, 0, 0, 0, implicit $mode, implicit $exec
renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X1F32_e64 $vgpr67, $vgpr66, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
renamable $agpr32_agpr33_agpr34_agpr35_agpr36_agpr37_agpr38_agpr39_agpr40_agpr41_agpr42_agpr43_agpr44_agpr45_agpr46_agpr47_agpr48_agpr49_agpr50_agpr51_agpr52_agpr53_agpr54_agpr55_agpr56_agpr57_agpr58_agpr59_agpr60_agpr61_agpr62_agpr63 = V_MFMA_F32_32X32X1F32_e64 $vgpr69, $vgpr68, killed $agpr32_agpr33_agpr34_agpr35_agpr36_agpr37_agpr38_agpr39_agpr40_agpr41_agpr42_agpr43_agpr44_agpr45_agpr46_agpr47_agpr48_agpr49_agpr50_agpr51_agpr52_agpr53_agpr54_agpr55_agpr56_agpr57_agpr58_agpr59_agpr60_agpr61_agpr62_agpr63, 0, 0, 0, implicit $mode, implicit $exec
renamable $sgpr2 = S_ADD_U32 renamable $sgpr2, 4, implicit-def $scc
renamable $sgpr3 = S_ADDC_U32 renamable $sgpr3, 0, implicit-def dead $scc, implicit killed $scc
S_CMP_LG_U32 renamable $sgpr2, 64, implicit-def $scc
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/schedule-barrier.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,8 @@
# multiplies, despite the presence of a barrier in the function.
# CHECK: BUFFER_LOAD_DWORD_OFFSET
# CHECK: BUFFER_LOAD_DWORD_OFFSET
# CHECK: V_MUL_LO_U32
# CHECK: V_MUL_LO_U32
# CHECK: V_MUL_LO_U32_e64
# CHECK: V_MUL_LO_U32_e64
name: test
tracksRegLiveness: true
body: |
Expand All @@ -31,14 +31,14 @@ body: |
%33.sub2:sgpr_128 = V_READFIRSTLANE_B32 %45.sub2, implicit $exec
%33.sub3:sgpr_128 = V_READFIRSTLANE_B32 %46.sub3, implicit $exec
%15:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET %33, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%39:vgpr_32 = V_MUL_LO_U32 %15, %15, implicit $exec
%39:vgpr_32 = V_MUL_LO_U32_e64 %15, %15, implicit $exec
undef %27.sub0:sgpr_128 = V_READFIRSTLANE_B32 %26.sub0, implicit $exec
%27.sub1:sgpr_128 = V_READFIRSTLANE_B32 %41.sub1, implicit $exec
%27.sub2:sgpr_128 = V_READFIRSTLANE_B32 %42.sub2, implicit $exec
%27.sub3:sgpr_128 = V_READFIRSTLANE_B32 %43.sub3, implicit $exec
%19:vgpr_32 = BUFFER_LOAD_DWORD_OFFSET %27, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%40:vgpr_32 = V_MUL_LO_U32 %19, %19, implicit $exec
%40:vgpr_32 = V_MUL_LO_U32_e64 %19, %19, implicit $exec
%23:vgpr_32 = V_ADD_U32_e32 %39, %40, implicit $exec
GLOBAL_STORE_DWORD %38, %23, 0, 0, 0, 0, implicit $exec
Expand Down
24 changes: 12 additions & 12 deletions llvm/test/CodeGen/AMDGPU/sdwa-peephole-instr-gfx10.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -236,51 +236,51 @@ body: |
%12 = V_AND_B32_e32 %6, %11, implicit $exec
%13 = V_LSHLREV_B32_e64 16, %12, implicit $exec
%14 = V_LSHRREV_B32_e64 16, %13, implicit $exec
%15 = V_BFE_U32 %13, 8, 8, implicit $exec
%15 = V_BFE_U32_e64 %13, 8, 8, implicit $exec
%16 = V_ADD_F32_e32 %14, %15, implicit $mode, implicit $exec
%17 = V_LSHLREV_B32_e64 16, %16, implicit $exec
%18 = V_LSHRREV_B32_e64 16, %17, implicit $exec
%19 = V_BFE_U32 %17, 8, 8, implicit $exec
%19 = V_BFE_U32_e64 %17, 8, 8, implicit $exec
%20 = V_SUB_F16_e32 %18, %19, implicit $mode, implicit $exec
%21 = V_LSHLREV_B32_e64 16, %20, implicit $exec
%22 = V_BFE_U32 %20, 8, 8, implicit $exec
%22 = V_BFE_U32_e64 %20, 8, 8, implicit $exec
%23 = V_FMAC_F32_e32 %21, %22, %22, implicit $mode, implicit $exec
%24 = V_LSHLREV_B32_e64 16, %23, implicit $exec
%25 = V_LSHRREV_B32_e64 16, %24, implicit $exec
%26 = V_BFE_U32 %24, 8, 8, implicit $exec
%26 = V_BFE_U32_e64 %24, 8, 8, implicit $exec
%27 = V_FMAC_F16_e32 %25, %26, %26, implicit $mode, implicit $exec
%28 = V_LSHLREV_B32_e64 16, %27, implicit $exec
%29 = V_LSHRREV_B32_e64 16, %28, implicit $exec
%30 = V_AND_B32_e64 23, %29, implicit $exec
%31 = V_LSHLREV_B32_e64 16, %30, implicit $exec
%32 = V_LSHRREV_B32_e64 16, %31, implicit $exec
%33 = V_BFE_U32 %31, 8, 8, implicit $exec
%33 = V_BFE_U32_e64 %31, 8, 8, implicit $exec
%34 = V_ADD_F32_e64 0, %32, 0, %33, 0, 0, implicit $mode, implicit $exec
%35 = V_LSHLREV_B32_e64 16, %34, implicit $exec
%37 = V_BFE_U32 %35, 8, 8, implicit $exec
%37 = V_BFE_U32_e64 %35, 8, 8, implicit $exec
%38 = V_SUB_F16_e64 0, 23, 0, %37, 0, 0, implicit $mode, implicit $exec
%39 = V_LSHLREV_B32_e64 16, %38, implicit $exec
%40 = V_BFE_U32 %39, 8, 8, implicit $exec
%40 = V_BFE_U32_e64 %39, 8, 8, implicit $exec
%41 = V_FMAC_F32_e64 0, 23, 0, %40, 0, %40, 0, 0, implicit $mode, implicit $exec
%42 = V_LSHLREV_B32_e64 16, %41, implicit $exec
%43 = V_LSHRREV_B32_e64 16, %42, implicit $exec
%44 = V_BFE_U32 %42, 8, 8, implicit $exec
%44 = V_BFE_U32_e64 %42, 8, 8, implicit $exec
%45 = V_FMAC_F16_e64 0, %43, 0, %44, 0, %44, 0, 0, implicit $mode, implicit $exec
%46 = V_LSHLREV_B32_e64 16, %45, implicit $exec
%47 = V_LSHRREV_B32_e64 16, %46, implicit $exec
%48 = V_BFE_U32 %46, 8, 8, implicit $exec
%48 = V_BFE_U32_e64 %46, 8, 8, implicit $exec
%49 = V_ADD_F32_e64 0, %47, 1, %48, 0, 0, implicit $mode, implicit $exec
%50 = V_LSHLREV_B32_e64 16, %49, implicit $exec
%51 = V_BFE_U32 %50, 8, 8, implicit $exec
%51 = V_BFE_U32_e64 %50, 8, 8, implicit $exec
%52 = V_SUB_F16_e64 1, 23, 1, %51, 0, 0, implicit $mode, implicit $exec
%53 = V_LSHLREV_B32_e64 16, %52, implicit $exec
%54 = V_BFE_U32 %53, 8, 8, implicit $exec
%54 = V_BFE_U32_e64 %53, 8, 8, implicit $exec
%55 = V_FMAC_F32_e64 1, 23, 1, %54, 1, %54, 1, 0, implicit $mode, implicit $exec
%56 = V_LSHLREV_B32_e64 16, %55, implicit $exec
%57 = V_LSHRREV_B32_e64 16, %56, implicit $exec
%58 = V_BFE_U32 %56, 8, 8, implicit $exec
%58 = V_BFE_U32_e64 %56, 8, 8, implicit $exec
%59 = V_FMAC_F16_e64 1, %57, 1, %58, 1, %58, 0, 2, implicit $mode, implicit $exec
%60 = V_LSHLREV_B32_e64 16, %59, implicit $exec
Expand Down
26 changes: 13 additions & 13 deletions llvm/test/CodeGen/AMDGPU/sdwa-peephole-instr.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -265,51 +265,51 @@ body: |
%12 = V_AND_B32_e32 %6, %11, implicit $exec
%13 = V_LSHLREV_B32_e64 16, %12, implicit $exec
%14 = V_LSHRREV_B32_e64 16, %13, implicit $exec
%15 = V_BFE_U32 %13, 8, 8, implicit $exec
%15 = V_BFE_U32_e64 %13, 8, 8, implicit $exec
%16 = V_ADD_F32_e32 %14, %15, implicit $mode, implicit $exec
%17 = V_LSHLREV_B32_e64 16, %16, implicit $exec
%18 = V_LSHRREV_B32_e64 16, %17, implicit $exec
%19 = V_BFE_U32 %17, 8, 8, implicit $exec
%19 = V_BFE_U32_e64 %17, 8, 8, implicit $exec
%20 = V_SUB_F16_e32 %18, %19, implicit $mode, implicit $exec
%21 = V_LSHLREV_B32_e64 16, %20, implicit $exec
%22 = V_BFE_U32 %20, 8, 8, implicit $exec
%22 = V_BFE_U32_e64 %20, 8, 8, implicit $exec
%23 = V_MAC_F32_e32 %21, %22, %22, implicit $mode, implicit $exec
%24 = V_LSHLREV_B32_e64 16, %23, implicit $exec
%25 = V_LSHRREV_B32_e64 16, %24, implicit $exec
%26 = V_BFE_U32 %24, 8, 8, implicit $exec
%26 = V_BFE_U32_e64 %24, 8, 8, implicit $exec
%27 = V_MAC_F16_e32 %25, %26, %26, implicit $mode, implicit $exec
%28 = V_LSHLREV_B32_e64 16, %27, implicit $exec
%29 = V_LSHRREV_B32_e64 16, %28, implicit $exec
%30 = V_AND_B32_e64 23, %29, implicit $exec
%31 = V_LSHLREV_B32_e64 16, %30, implicit $exec
%32 = V_LSHRREV_B32_e64 16, %31, implicit $exec
%33 = V_BFE_U32 %31, 8, 8, implicit $exec
%33 = V_BFE_U32_e64 %31, 8, 8, implicit $exec
%34 = V_ADD_F32_e64 0, %32, 0, %33, 0, 0, implicit $mode, implicit $exec
%35 = V_LSHLREV_B32_e64 16, %34, implicit $exec
%37 = V_BFE_U32 %35, 8, 8, implicit $exec
%37 = V_BFE_U32_e64 %35, 8, 8, implicit $exec
%38 = V_SUB_F16_e64 0, 23, 0, %37, 0, 0, implicit $mode, implicit $exec
%39 = V_LSHLREV_B32_e64 16, %38, implicit $exec
%40 = V_BFE_U32 %39, 8, 8, implicit $exec
%40 = V_BFE_U32_e64 %39, 8, 8, implicit $exec
%41 = V_MAC_F32_e64 0, 23, 0, %40, 0, %40, 0, 0, implicit $mode, implicit $exec
%42 = V_LSHLREV_B32_e64 16, %41, implicit $exec
%43 = V_LSHRREV_B32_e64 16, %42, implicit $exec
%44 = V_BFE_U32 %42, 8, 8, implicit $exec
%44 = V_BFE_U32_e64 %42, 8, 8, implicit $exec
%45 = V_MAC_F16_e64 0, %43, 0, %44, 0, %44, 0, 0, implicit $mode, implicit $exec
%46 = V_LSHLREV_B32_e64 16, %45, implicit $exec
%47 = V_LSHRREV_B32_e64 16, %46, implicit $exec
%48 = V_BFE_U32 %46, 8, 8, implicit $exec
%48 = V_BFE_U32_e64 %46, 8, 8, implicit $exec
%49 = V_ADD_F32_e64 0, %47, 1, %48, 0, 0, implicit $mode, implicit $exec
%50 = V_LSHLREV_B32_e64 16, %49, implicit $exec
%51 = V_BFE_U32 %50, 8, 8, implicit $exec
%51 = V_BFE_U32_e64 %50, 8, 8, implicit $exec
%52 = V_SUB_F16_e64 1, 23, 1, %51, 0, 0, implicit $mode, implicit $exec
%53 = V_LSHLREV_B32_e64 16, %52, implicit $exec
%54 = V_BFE_U32 %53, 8, 8, implicit $exec
%54 = V_BFE_U32_e64 %53, 8, 8, implicit $exec
%55 = V_MAC_F32_e64 1, 23, 1, %54, 1, %54, 1, 0, implicit $mode, implicit $exec
%56 = V_LSHLREV_B32_e64 16, %55, implicit $exec
%57 = V_LSHRREV_B32_e64 16, %56, implicit $exec
%58 = V_BFE_U32 %56, 8, 8, implicit $exec
%58 = V_BFE_U32_e64 %56, 8, 8, implicit $exec
%59 = V_MAC_F16_e64 1, %57, 1, %58, 1, %58, 0, 2, implicit $mode, implicit $exec
%60 = V_LSHLREV_B32_e64 16, %59, implicit $exec
Expand Down Expand Up @@ -463,7 +463,7 @@ body: |
%3:vgpr_32 = V_AND_B32_e32 %1, %2, implicit $exec
%4:vgpr_32 = V_LSHLREV_B32_e64 16, %3, implicit $exec
%5:vgpr_32 = V_LSHRREV_B32_e64 16, %4, implicit $exec
%6:vgpr_32 = V_BFE_U32 %4, 8, 8, implicit $exec
%6:vgpr_32 = V_BFE_U32_e64 %4, 8, 8, implicit $exec
%7:vgpr_32 = nnan nofpexcept V_ADD_F32_e32 %5, %6, implicit $mode, implicit $exec
S_ENDPGM 0, implicit %7
Expand Down
2 changes: 1 addition & 1 deletion llvm/test/CodeGen/AMDGPU/sdwa-preserve.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,7 @@ body: |
%5 = V_AND_B32_e32 65535, %3, implicit $exec
%6 = V_LSHRREV_B32_e64 16, %4, implicit $exec
%7 = V_BFE_U32 %3, 8, 8, implicit $exec
%7 = V_BFE_U32_e64 %3, 8, 8, implicit $exec
%8 = V_LSHRREV_B32_e32 24, %4, implicit $exec
%9 = V_ADD_F16_e64 0, %5, 0, %6, 0, 0, implicit $mode, implicit $exec
Expand Down
8 changes: 4 additions & 4 deletions llvm/test/CodeGen/AMDGPU/sdwa-scalar-ops.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -222,7 +222,7 @@ body: |
%16 = REG_SEQUENCE %14, %subreg.sub0, %15, %subreg.sub1
%18 = COPY %16
%17 = FLAT_LOAD_DWORD %18, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.uglygep45)
%60 = V_BFE_U32 %17, 8, 8, implicit $exec
%60 = V_BFE_U32_e64 %17, 8, 8, implicit $exec
%61 = V_LSHLREV_B32_e32 2, killed %60, implicit $exec
%70 = V_ADD_CO_U32_e32 %7.sub0, %61, implicit-def $vcc, implicit $exec
%66 = COPY %13
Expand All @@ -235,7 +235,7 @@ body: |
%72 = COPY killed %38
%41 = REG_SEQUENCE killed %71, %subreg.sub0, killed %72, %subreg.sub1
%40 = FLAT_LOAD_DWORD killed %41, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.scevgep)
%73 = V_BFE_U32 %40, 8, 8, implicit $exec
%73 = V_BFE_U32_e64 %40, 8, 8, implicit $exec
%74 = V_LSHLREV_B32_e32 2, killed %73, implicit $exec
%83 = V_ADD_CO_U32_e32 %7.sub0, %74, implicit-def $vcc, implicit $exec
%78 = V_ADDC_U32_e32 0, %66, implicit-def $vcc, implicit $vcc, implicit $exec
Expand Down Expand Up @@ -385,7 +385,7 @@ body: |
%16 = REG_SEQUENCE %14, %subreg.sub0, %15, %subreg.sub1
%18 = COPY %16
%17 = FLAT_LOAD_DWORD %18, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.uglygep45)
%60 = V_BFE_U32 %17, 8, 8, implicit $exec
%60 = V_BFE_U32_e64 %17, 8, 8, implicit $exec
%61 = V_LSHLREV_B32_e32 %84, killed %60, implicit $exec
%70 = V_ADD_CO_U32_e32 %7.sub0, %61, implicit-def $vcc, implicit $exec
%66 = COPY %13
Expand All @@ -398,7 +398,7 @@ body: |
%72 = COPY killed %38
%41 = REG_SEQUENCE killed %71, %subreg.sub0, killed %72, %subreg.sub1
%40 = FLAT_LOAD_DWORD killed %41, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.scevgep)
%73 = V_BFE_U32 %40, 8, 8, implicit $exec
%73 = V_BFE_U32_e64 %40, 8, 8, implicit $exec
%74 = V_LSHLREV_B32_e32 %84, killed %73, implicit $exec
%83 = V_ADD_CO_U32_e32 %7.sub0, %74, implicit-def $vcc, implicit $exec
%78 = V_ADDC_U32_e32 0, %66, implicit-def $vcc, implicit $vcc, implicit $exec
Expand Down
12 changes: 6 additions & 6 deletions llvm/test/CodeGen/AMDGPU/shrink-vop3-carry-out.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -79,7 +79,7 @@ body: |
%11 = S_MOV_B32 0
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%28 = V_LSHL_B64 killed %27, 2, implicit $exec
%28 = V_LSHL_B64_e64 killed %27, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%17 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%19 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 4, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -163,7 +163,7 @@ body: |
%11 = S_MOV_B32 0
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%28 = V_LSHL_B64 killed %27, 2, implicit $exec
%28 = V_LSHL_B64_e64 killed %27, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%17 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%19 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 4, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -247,7 +247,7 @@ body: |
%11 = S_MOV_B32 0
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%28 = V_LSHL_B64 killed %27, 2, implicit $exec
%28 = V_LSHL_B64_e64 killed %27, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%17 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%19 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 4, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -330,7 +330,7 @@ body: |
%11 = S_MOV_B32 0
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%28 = V_LSHL_B64 killed %27, 2, implicit $exec
%28 = V_LSHL_B64_e64 killed %27, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%17 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%19 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 4, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -415,7 +415,7 @@ body: |
%11 = S_MOV_B32 0
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%28 = V_LSHL_B64 killed %27, 2, implicit $exec
%28 = V_LSHL_B64_e64 killed %27, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%17 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%19 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 4, 0, 0, 0, 0, 0, implicit $exec
Expand Down Expand Up @@ -500,7 +500,7 @@ body: |
%11 = S_MOV_B32 0
%12 = REG_SEQUENCE killed %11, 1, killed %10, 2
%13 = REG_SEQUENCE killed %5, 17, %12, 18
%28 = V_LSHL_B64 killed %27, 2, implicit $exec
%28 = V_LSHL_B64_e64 killed %27, 2, implicit $exec
%16 = REG_SEQUENCE killed %4, 17, %12, 18
%17 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 0, 0, 0, 0, 0, 0, implicit $exec
%19 = BUFFER_LOAD_DWORD_ADDR64 %28, %13, 0, 4, 0, 0, 0, 0, 0, implicit $exec
Expand Down
4 changes: 2 additions & 2 deletions llvm/test/CodeGen/AMDGPU/smem-no-clause-coalesced.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -38,8 +38,8 @@ body: |
%3.sub2:sgpr_128 = S_MOV_B32 -1
%7.sub0:sreg_64_xexec = S_LOAD_DWORD_IMM %7, 48, 0, 0 :: (load 4 from `i8 addrspace(4)* undef`, addrspace 4)
%8:sreg_64_xexec = S_BUFFER_LOAD_DWORDX2_IMM %3, 640, 0, 0 :: (dereferenceable invariant load 8)
undef %9.sub0:vreg_128 = V_LSHL_ADD_U32 %6, 4, %4, implicit $exec
%9.sub1:vreg_128 = V_LSHL_ADD_U32 %5, 4, %0, implicit $exec
undef %9.sub0:vreg_128 = V_LSHL_ADD_U32_e64 %6, 4, %4, implicit $exec
%9.sub1:vreg_128 = V_LSHL_ADD_U32_e64 %5, 4, %0, implicit $exec
S_ENDPGM 0
...
Expand Down
12 changes: 6 additions & 6 deletions llvm/test/CodeGen/AMDGPU/spill-agpr-partially-undef.mir
View file
Original file line number Diff line number Diff line change
Expand Up @@ -16,9 +16,9 @@ body: |
; CHECK-LABEL: name: spill_a64_kill
; CHECK: liveins: $agpr0_agpr1
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32 killed $agpr0, implicit $exec, implicit-def $agpr0_agpr1
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32_e64 killed $agpr0, implicit $exec, implicit-def $agpr0_agpr1
; CHECK: BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr32, 0, 0, 0, 0, 0, 0, implicit $exec, implicit $agpr0_agpr1 :: (store 4 into %stack.0, addrspace 5)
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32 killed $agpr1, implicit $exec
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32_e64 killed $agpr1, implicit $exec
; CHECK: BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr32, 4, 0, 0, 0, 0, 0, implicit $exec, implicit killed $agpr0_agpr1 :: (store 4 into %stack.0 + 4, addrspace 5)
SI_SPILL_A64_SAVE killed $agpr0_agpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store 8 into %stack.0, addrspace 5)
...
Expand All @@ -40,9 +40,9 @@ body: |
; CHECK-LABEL: name: spill_a64_undef_sub1_killed
; CHECK: liveins: $agpr0
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32 killed $agpr0, implicit $exec, implicit-def $agpr0_agpr1
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32_e64 killed $agpr0, implicit $exec, implicit-def $agpr0_agpr1
; CHECK: BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr32, 0, 0, 0, 0, 0, 0, implicit $exec, implicit $agpr0_agpr1 :: (store 4 into %stack.0, addrspace 5)
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32 killed $agpr1, implicit $exec
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32_e64 killed $agpr1, implicit $exec
; CHECK: BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr32, 4, 0, 0, 0, 0, 0, implicit $exec, implicit killed $agpr0_agpr1 :: (store 4 into %stack.0 + 4, addrspace 5)
SI_SPILL_A64_SAVE killed $agpr0_agpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store 8 into %stack.0, addrspace 5)
...
Expand All @@ -62,9 +62,9 @@ body: |
; CHECK-LABEL: name: spill_a64_undef_sub0_killed
; CHECK: liveins: $agpr1
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32 killed $agpr0, implicit $exec, implicit-def $agpr0_agpr1
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32_e64 killed $agpr0, implicit $exec, implicit-def $agpr0_agpr1
; CHECK: BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr32, 0, 0, 0, 0, 0, 0, implicit $exec, implicit $agpr0_agpr1 :: (store 4 into %stack.0, addrspace 5)
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32 killed $agpr1, implicit $exec
; CHECK: $vgpr0 = V_ACCVGPR_READ_B32_e64 killed $agpr1, implicit $exec
; CHECK: BUFFER_STORE_DWORD_OFFSET killed $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr32, 4, 0, 0, 0, 0, 0, implicit $exec, implicit killed $agpr0_agpr1 :: (store 4 into %stack.0 + 4, addrspace 5)
SI_SPILL_A64_SAVE killed $agpr0_agpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store 8 into %stack.0, addrspace 5)
...
Loading