https://github.com/arsenm created
https://github.com/llvm/llvm-project/pull/140607
None
>From 4021541f1a05550bf27348b67f551d4ba73ef9e1 Mon Sep 17 00:00:00 2001
From: Matt Arsenault <matthew.arsena...@amd.com>
Date: Wed, 14 May 2025 08:50:59 +0200
Subject: [PATCH] AMDGPU: Add baseline tests for #139317
---
.../CodeGen/AMDGPU/fold-imm-copy-agpr.mir | 320 ++++++++++++++
.../AMDGPU/fold-short-64-bit-literals.mir | 392 +++++++++++++++++-
...issue139317-bad-opsel-reg-sequence-fold.ll | 66 +++
.../si-fold-operands-subreg-imm.gfx942.mir | 202 +++++++++
.../AMDGPU/si-fold-operands-subreg-imm.mir | 26 ++
5 files changed, 1001 insertions(+), 5 deletions(-)
create mode 100644 llvm/test/CodeGen/AMDGPU/fold-imm-copy-agpr.mir
create mode 100644
llvm/test/CodeGen/AMDGPU/issue139317-bad-opsel-reg-sequence-fold.ll
create mode 100644
llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.gfx942.mir
diff --git a/llvm/test/CodeGen/AMDGPU/fold-imm-copy-agpr.mir
b/llvm/test/CodeGen/AMDGPU/fold-imm-copy-agpr.mir
new file mode 100644
index 0000000000000..3021761f099fa
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/fold-imm-copy-agpr.mir
@@ -0,0 +1,320 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
UTC_ARGS: --version 5
+# RUN: llc -mtriple=amdgcn -mcpu=gfx90a -run-pass=si-fold-operands %s -o - |
FileCheck -check-prefix=GCN %s
+
+---
+name: v_mov_b64_pseudo_imm_0_copy_to_areg_64
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_0_copy_to_areg_64
+ ; GCN: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit
$exec
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:areg_64_align2 = COPY [[V_MOV_B]]
+ ; GCN-NEXT: $agpr0_agpr1 = COPY [[COPY]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
+ %1:areg_64_align2 = COPY %0
+ $agpr0_agpr1 = COPY %1
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_neg1_copy_to_areg_64
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_neg1_copy_to_areg_64
+ ; GCN: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO -1, implicit
$exec
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:areg_64_align2 = COPY [[V_MOV_B]]
+ ; GCN-NEXT: $agpr0_agpr1 = COPY [[COPY]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO -1, implicit $exec
+ %1:areg_64_align2 = COPY %0
+ $agpr0_agpr1 = COPY %1
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_literal_copy_to_areg_64
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_literal_copy_to_areg_64
+ ; GCN: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO 999, implicit
$exec
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:areg_64_align2 = COPY [[V_MOV_B]]
+ ; GCN-NEXT: $agpr0_agpr1 = COPY [[COPY]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 999, implicit $exec
+ %1:areg_64_align2 = COPY %0
+ $agpr0_agpr1 = COPY %1
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_0_copy_sub0_to_agpr_32
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_0_copy_sub0_to_agpr_32
+ ; GCN: [[V_ACCVGPR_WRITE_B32_e64_:%[0-9]+]]:agpr_32 =
V_ACCVGPR_WRITE_B32_e64 0, implicit $exec
+ ; GCN-NEXT: $agpr0 = COPY [[V_ACCVGPR_WRITE_B32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
+ %1:agpr_32 = COPY %0.sub0
+ $agpr0 = COPY %1
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_0_copy_sub1_to_agpr_32
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_0_copy_sub1_to_agpr_32
+ ; GCN: [[V_ACCVGPR_WRITE_B32_e64_:%[0-9]+]]:agpr_32 =
V_ACCVGPR_WRITE_B32_e64 0, implicit $exec
+ ; GCN-NEXT: $agpr0 = COPY [[V_ACCVGPR_WRITE_B32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
+ %1:agpr_32 = COPY %0.sub1
+ $agpr0 = COPY %1
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_lit_copy_sub0_to_agpr_32
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_lit_copy_sub0_to_agpr_32
+ ; GCN: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
4290672329592, implicit $exec
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:agpr_32 = COPY [[V_MOV_B]].sub0
+ ; GCN-NEXT: $agpr0 = COPY [[COPY]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 4290672329592, implicit $exec
+ %1:agpr_32 = COPY %0.sub0
+ $agpr0 = COPY %1
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_lit_copy_sub1_to_agpr_32
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_lit_copy_sub1_to_agpr_32
+ ; GCN: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
4290672329592, implicit $exec
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:agpr_32 = COPY [[V_MOV_B]].sub1
+ ; GCN-NEXT: $agpr0 = COPY [[COPY]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 4290672329592, implicit $exec
+ %1:agpr_32 = COPY %0.sub1
+ $agpr0 = COPY %1
+ S_ENDPGM 0
+
+...
+
+
+---
+name: v_mov_b64_pseudo_imm_0_copy_to_av_64
+tracksRegLiveness: true
+body: |
+ bb.0:
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_0_copy_to_av_64
+ ; GCN: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit
$exec
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:av_64_align2 = COPY [[V_MOV_B]]
+ ; GCN-NEXT: S_ENDPGM 0, implicit [[COPY]]
+ %0:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
+ %1:av_64_align2 = COPY %0
+ S_ENDPGM 0, implicit %1
+
+...
+
+
+# XXX need f64 use
+
+
+
+
+---
+name: v_mov_b64_pseudo_imm_0_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_0_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, 0, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp64_1_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp64_1_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, 4607182418800017408, 0, 0, implicit $mode,
implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_lit_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_lit_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
4290672329592, implicit $exec
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, [[V_MOV_B]], 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4290672329592, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp32_1_splat_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
4575657222473777152, implicit $exec
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, [[V_MOV_B]], 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub0_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub0_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vgpr_32 = COPY %1.sub0
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub1_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub1_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vgpr_32 = COPY %1.sub1
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub0_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub0_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 0, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
+ %2:vgpr_32 = COPY %1.sub0
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub1_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub1_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_MOV_B32_e32_:%[0-9]+]]:vgpr_32 = V_MOV_B32_e32 1072693248,
implicit $exec
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, [[V_MOV_B32_e32_]], 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
+ %2:vgpr_32 = COPY %1.sub1
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
diff --git a/llvm/test/CodeGen/AMDGPU/fold-short-64-bit-literals.mir
b/llvm/test/CodeGen/AMDGPU/fold-short-64-bit-literals.mir
index 1f15066264c74..cfb42de4456f3 100644
--- a/llvm/test/CodeGen/AMDGPU/fold-short-64-bit-literals.mir
+++ b/llvm/test/CodeGen/AMDGPU/fold-short-64-bit-literals.mir
@@ -1,5 +1,6 @@
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
UTC_ARGS: --version 3
-# RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs
-run-pass=si-fold-operands -o - %s | FileCheck --check-prefix=GCN %s
+# RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs
-run-pass=si-fold-operands -o - %s | FileCheck --check-prefixes=GCN,GFX10 %s
+# RUN: llc -mtriple=amdgcn -mcpu=gfx900 -verify-machineinstrs
-run-pass=si-fold-operands -o - %s | FileCheck --check-prefixes=GCN,GFX9 %s
---
name: no_fold_fp_64bit_literal_sgpr
@@ -41,10 +42,16 @@ tracksRegLiveness: true
body: |
bb.0:
- ; GCN-LABEL: name: fold_fp_32bit_literal_sgpr
- ; GCN: [[DEF:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
- ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64 = V_ADD_F64_e64 0,
4636737291354636288, 0, [[DEF]], 0, 0, implicit $mode, implicit $exec
- ; GCN-NEXT: SI_RETURN_TO_EPILOG [[V_ADD_F64_e64_]]
+ ; GFX10-LABEL: name: fold_fp_32bit_literal_sgpr
+ ; GFX10: [[DEF:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
+ ; GFX10-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64 = V_ADD_F64_e64 0,
4636737291354636288, 0, [[DEF]], 0, 0, implicit $mode, implicit $exec
+ ; GFX10-NEXT: SI_RETURN_TO_EPILOG [[V_ADD_F64_e64_]]
+ ;
+ ; GFX9-LABEL: name: fold_fp_32bit_literal_sgpr
+ ; GFX9: [[DEF:%[0-9]+]]:vreg_64 = IMPLICIT_DEF
+ ; GFX9-NEXT: [[S_MOV_B:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO
4636737291354636288
+ ; GFX9-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64 = V_ADD_F64_e64 0,
[[S_MOV_B]], 0, [[DEF]], 0, 0, implicit $mode, implicit $exec
+ ; GFX9-NEXT: SI_RETURN_TO_EPILOG [[V_ADD_F64_e64_]]
%0:vreg_64 = IMPLICIT_DEF
%1:sreg_64 = S_MOV_B64_IMM_PSEUDO 4636737291354636288
%2:vreg_64 = V_ADD_F64_e64 0, %1, 0, %0, 0, 0, implicit $mode, implicit
$exec
@@ -136,3 +143,378 @@ body: |
%2:vreg_64 = V_PK_ADD_F32 0, %0, 0, %1, 0, 0, 0, 0, 0, implicit $mode,
implicit $exec
SI_RETURN_TO_EPILOG %2
...
+
+---
+name: v_mov_b64_pseudo_imm_0_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_0_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, 0, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp64_1_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp64_1_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, 4607182418800017408, 0, 0, implicit $mode,
implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_lit_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_lit_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
4290672329592, implicit $exec
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, [[V_MOV_B]], 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4290672329592, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_full_copy_to_f64_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp32_1_splat_full_copy_to_f64_use
+ ; GCN: liveins: $vgpr0_vgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vreg_64_align2 = COPY $vgpr0_vgpr1
+ ; GCN-NEXT: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
4575657222473777152, implicit $exec
+ ; GCN-NEXT: [[V_ADD_F64_e64_:%[0-9]+]]:vreg_64_align2 = nofpexcept
V_ADD_F64_e64 0, [[COPY]], 1, [[V_MOV_B]], 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0_vgpr1 = COPY [[V_ADD_F64_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vreg_64_align2 = COPY $vgpr0_vgpr1
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vreg_64_align2 = COPY %1
+ %3:vreg_64_align2 = nofpexcept V_ADD_F64_e64 0, %0, 1, %2, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0_vgpr1 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub0_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub0_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vgpr_32 = COPY %1.sub0
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub1_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_splat_copy_extract_sub1_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vgpr_32 = COPY %1.sub1
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub0_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub0_to_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 0, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
+ %2:vgpr_32 = COPY %1.sub0
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub1_to_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GFX10-LABEL: name:
v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub1_to_f32_use
+ ; GFX10: liveins: $vgpr0
+ ; GFX10-NEXT: {{ $}}
+ ; GFX10-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GFX10-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept
V_ADD_F32_e64 0, [[COPY]], 1, 1072693248, 0, 0, implicit $mode, implicit $exec
+ ; GFX10-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GFX10-NEXT: S_ENDPGM 0
+ ;
+ ; GFX9-LABEL: name:
v_mov_b64_pseudo_imm_fp64_1_copy_extract_sub1_to_f32_use
+ ; GFX9: liveins: $vgpr0
+ ; GFX9-NEXT: {{ $}}
+ ; GFX9-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GFX9-NEXT: [[V_MOV_B32_e32_:%[0-9]+]]:vgpr_32 = V_MOV_B32_e32
1072693248, implicit $exec
+ ; GFX9-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, [[V_MOV_B32_e32_]], 0, 0, implicit $mode, implicit $exec
+ ; GFX9-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GFX9-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
+ %2:vgpr_32 = COPY %1.sub1
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_fp32_2_copy_extract_sub0_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_fp32_2_copy_extract_sub0_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1073741824, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222482165760, implicit $exec
+ %2:vgpr_32 = COPY %1.sub0
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_fp32_2_copy_extract_sub1_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_fp32_2_copy_extract_sub1_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222482165760, implicit $exec
+ %2:vgpr_32 = COPY %1.sub1
+ %3:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %2, 0, 0, implicit $mode,
implicit $exec
+ $vgpr0 = COPY %3
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_direct_subreg_sub0_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_splat_direct_subreg_sub0_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %1.sub0, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0 = COPY %2
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_splat_direct_subreg_sub1_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_splat_direct_subreg_sub1_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222473777152, implicit $exec
+ %2:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %1.sub1, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0 = COPY %2
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_lit_direct_subreg_sub0_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GFX10-LABEL: name: v_mov_b64_pseudo_lit_direct_subreg_sub0_f32_use
+ ; GFX10: liveins: $vgpr0
+ ; GFX10-NEXT: {{ $}}
+ ; GFX10-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GFX10-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept
V_ADD_F32_e64 0, [[COPY]], 1, 8888, 0, 0, implicit $mode, implicit $exec
+ ; GFX10-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GFX10-NEXT: S_ENDPGM 0
+ ;
+ ; GFX9-LABEL: name: v_mov_b64_pseudo_lit_direct_subreg_sub0_f32_use
+ ; GFX9: liveins: $vgpr0
+ ; GFX9-NEXT: {{ $}}
+ ; GFX9-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GFX9-NEXT: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
42945378001592, implicit $exec
+ ; GFX9-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, [[V_MOV_B]].sub0, 0, 0, implicit $mode, implicit $exec
+ ; GFX9-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GFX9-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 42945378001592, implicit $exec
+ %2:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %1.sub0, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0 = COPY %2
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_lit_direct_subreg_sub1_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GFX10-LABEL: name: v_mov_b64_pseudo_lit_direct_subreg_sub1_f32_use
+ ; GFX10: liveins: $vgpr0
+ ; GFX10-NEXT: {{ $}}
+ ; GFX10-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GFX10-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept
V_ADD_F32_e64 0, [[COPY]], 1, 9999, 0, 0, implicit $mode, implicit $exec
+ ; GFX10-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GFX10-NEXT: S_ENDPGM 0
+ ;
+ ; GFX9-LABEL: name: v_mov_b64_pseudo_lit_direct_subreg_sub1_f32_use
+ ; GFX9: liveins: $vgpr0
+ ; GFX9-NEXT: {{ $}}
+ ; GFX9-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GFX9-NEXT: [[V_MOV_B:%[0-9]+]]:vreg_64_align2 = V_MOV_B64_PSEUDO
42945378001592, implicit $exec
+ ; GFX9-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, [[V_MOV_B]].sub1, 0, 0, implicit $mode, implicit $exec
+ ; GFX9-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GFX9-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 42945378001592, implicit $exec
+ %2:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %1.sub1, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0 = COPY %2
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_fp32_2_direct_subreg_sub0_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_fp32_2_direct_subreg_sub0_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1073741824, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222482165760, implicit $exec
+ %2:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %1.sub0, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0 = COPY %2
+ S_ENDPGM 0
+
+...
+
+---
+name: v_mov_b64_pseudo_imm_fp32_1_fp32_2_direct_subreg_sub1_f32_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+ ; GCN-LABEL: name:
v_mov_b64_pseudo_imm_fp32_1_fp32_2_direct_subreg_sub1_f32_use
+ ; GCN: liveins: $vgpr0
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; GCN-NEXT: [[V_ADD_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_ADD_F32_e64
0, [[COPY]], 1, 1065353216, 0, 0, implicit $mode, implicit $exec
+ ; GCN-NEXT: $vgpr0 = COPY [[V_ADD_F32_e64_]]
+ ; GCN-NEXT: S_ENDPGM 0
+ %0:vgpr_32 = COPY $vgpr0
+ %1:vreg_64_align2 = V_MOV_B64_PSEUDO 4575657222482165760, implicit $exec
+ %2:vgpr_32 = nofpexcept V_ADD_F32_e64 0, %0, 1, %1.sub1, 0, 0, implicit
$mode, implicit $exec
+ $vgpr0 = COPY %2
+ S_ENDPGM 0
+
+...
diff --git
a/llvm/test/CodeGen/AMDGPU/issue139317-bad-opsel-reg-sequence-fold.ll
b/llvm/test/CodeGen/AMDGPU/issue139317-bad-opsel-reg-sequence-fold.ll
new file mode 100644
index 0000000000000..7d1ea68e63241
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/issue139317-bad-opsel-reg-sequence-fold.ll
@@ -0,0 +1,66 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
UTC_ARGS: --version 5
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 < %s | FileCheck
-check-prefix=GFX942 %s
+
+; Check for correct folding of the constants produced by the
+; stepvector into the fadd. The value should not get lost when folding
+; through subregister extracts of reg_sequence.
+define amdgpu_kernel void
@stepper_test_kernel_DType_I6A6AcB6A6AsA6A6A_68a5362b97a102776ef47f0e8e894a38(ptr
addrspace(1) readonly captures(none) %.global, ptr addrspace(1) writeonly
captures(none) %.global1, i32 %arg2) {
+; GFX942-LABEL:
stepper_test_kernel_DType_I6A6AcB6A6AsA6A6A_68a5362b97a102776ef47f0e8e894a38:
+; GFX942: ; %bb.0: ; %bb
+; GFX942-NEXT: s_load_dword s6, s[4:5], 0x10
+; GFX942-NEXT: s_mov_b32 s7, 0
+; GFX942-NEXT: s_waitcnt lgkmcnt(0)
+; GFX942-NEXT: s_cmp_eq_u32 s6, 0
+; GFX942-NEXT: s_cbranch_scc1 .LBB0_3
+; GFX942-NEXT: ; %bb.1: ; %.lr.ph.preheader
+; GFX942-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x0
+; GFX942-NEXT: s_mov_b32 s8, 0x47004600
+; GFX942-NEXT: s_mov_b32 s9, 0x45004400
+; GFX942-NEXT: s_mov_b64 s[4:5], 0
+; GFX942-NEXT: v_mov_b32_e32 v2, 0
+; GFX942-NEXT: v_mov_b64_e32 v[0:1], s[6:7]
+; GFX942-NEXT: .LBB0_2: ; %.lr.ph
+; GFX942-NEXT: ; =>This Inner Loop Header: Depth=1
+; GFX942-NEXT: s_waitcnt lgkmcnt(0)
+; GFX942-NEXT: global_load_dwordx4 v[4:7], v2, s[2:3]
+; GFX942-NEXT: s_add_u32 s4, s4, 8
+; GFX942-NEXT: s_addc_u32 s5, s5, 0
+; GFX942-NEXT: v_cmp_lt_u64_e32 vcc, s[4:5], v[0:1]
+; GFX942-NEXT: s_waitcnt vmcnt(0)
+; GFX942-NEXT: v_pk_add_f16 v7, v7, s8
+; GFX942-NEXT: v_pk_add_f16 v6, v6, s9
+; GFX942-NEXT: v_pk_add_f16 v5, v5, 0
+; GFX942-NEXT: v_pk_add_f16 v4, v4, 1.0 op_sel:[0,1] op_sel_hi:[1,0]
+; GFX942-NEXT: global_store_dwordx4 v2, v[4:7], s[0:1]
+; GFX942-NEXT: s_add_u32 s0, s0, 16
+; GFX942-NEXT: s_addc_u32 s1, s1, 0
+; GFX942-NEXT: s_add_u32 s2, s2, 16
+; GFX942-NEXT: s_addc_u32 s3, s3, 0
+; GFX942-NEXT: s_cbranch_vccnz .LBB0_2
+; GFX942-NEXT: .LBB0_3: ; %._crit_edge
+; GFX942-NEXT: s_endpgm
+bb:
+ %i = tail call <8 x i32> @llvm.stepvector.v8i32()
+ %i3 = sitofp <8 x i32> %i to <8 x half>
+ %i4 = zext i32 %arg2 to i64
+ %.not = icmp eq i32 %arg2, 0
+ br i1 %.not, label %._crit_edge, label %.lr.ph
+
+.lr.ph: ; preds = %.lr.ph, %bb
+ %i5 = phi i64 [ %i6, %.lr.ph ], [ 0, %bb ]
+ %i6 = add nuw nsw i64 %i5, 8
+ %i7 = getelementptr inbounds nuw half, ptr addrspace(1) %.global1, i64 %i5
+ %i8 = load <8 x half>, ptr addrspace(1) %i7, align 2
+ %i9 = fadd <8 x half> %i8, %i3
+ %i10 = getelementptr inbounds nuw half, ptr addrspace(1) %.global, i64 %i5
+ store <8 x half> %i9, ptr addrspace(1) %i10, align 2
+ %i11 = icmp samesign ult i64 %i6, %i4
+ br i1 %i11, label %.lr.ph, label %._crit_edge
+
+._crit_edge: ; preds = %.lr.ph, %bb
+ ret void
+}
+
+declare <8 x i32> @llvm.stepvector.v8i32() #0
+
+attributes #0 = { nocallback nofree nosync nounwind willreturn memory(none) }
diff --git a/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.gfx942.mir
b/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.gfx942.mir
new file mode 100644
index 0000000000000..c1ce216955dca
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.gfx942.mir
@@ -0,0 +1,202 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
UTC_ARGS: --version 5
+# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 -run-pass=si-fold-operands
-o - %s | FileCheck %s
+
+# Test behavior of folding into op_sel operands through reg_sequence
+# with subregister uses
+
+---
+name: issue139317
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 1107312640
+ ; CHECK-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 1006632960
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_1]], %subreg.sub0, killed [[S_MOV_B32_]], %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 8, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_32 = S_MOV_B32 1107312640
+ %2:sreg_32 = S_MOV_B32 1006632960
+ %3:sgpr_64 = REG_SEQUENCE killed %2:sreg_32, %subreg.sub0, killed
%1:sreg_32, %subreg.sub1
+ %4:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %3.sub1:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %4
+...
+
+---
+name: issue139317_sub0
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_sub0
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 1107312640
+ ; CHECK-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 1006632960
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_1]], %subreg.sub0, killed [[S_MOV_B32_]], %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 4, 15360, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_32 = S_MOV_B32 1107312640
+ %2:sreg_32 = S_MOV_B32 1006632960
+ %3:sgpr_64 = REG_SEQUENCE killed %2:sreg_32, %subreg.sub0, killed
%1:sreg_32, %subreg.sub1
+ %4:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %3.sub0:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %4
+...
+
+---
+name: issue139317_nested_reg_sequence
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_nested_reg_sequence
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 1107312640
+ ; CHECK-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 1006632960
+ ; CHECK-NEXT: [[S_MOV_B32_2:%[0-9]+]]:sreg_32 = S_MOV_B32 1191200256
+ ; CHECK-NEXT: [[S_MOV_B32_3:%[0-9]+]]:sreg_32 = S_MOV_B32 1157645312
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_1]], %subreg.sub0, killed [[S_MOV_B32_]], %subreg.sub1
+ ; CHECK-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_2]], %subreg.sub0, killed [[S_MOV_B32_3]], %subreg.sub1
+ ; CHECK-NEXT: [[REG_SEQUENCE2:%[0-9]+]]:sgpr_128 = REG_SEQUENCE
[[REG_SEQUENCE]], %subreg.sub0_sub1, [[REG_SEQUENCE1]], %subreg.sub2_sub3
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, killed [[COPY]], 8, [[REG_SEQUENCE2]].sub1, 0, 0, 0, 0, 0, implicit $mode,
implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_32 = S_MOV_B32 1107312640
+ %2:sreg_32 = S_MOV_B32 1006632960
+ %3:sreg_32 = S_MOV_B32 1191200256
+ %4:sreg_32 = S_MOV_B32 1157645312
+ %5:sgpr_64 = REG_SEQUENCE killed %2, %subreg.sub0, killed %1, %subreg.sub1
+ %6:sgpr_64 = REG_SEQUENCE killed %3, %subreg.sub0, killed %4, %subreg.sub1
+ %7:sgpr_128 = REG_SEQUENCE %5, %subreg.sub0_sub1, %6, %subreg.sub2_sub3
+ %8:vgpr_32 = nofpexcept V_PK_ADD_F16 8, killed %0, 8, %7.sub1, 0, 0, 0, 0,
0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %8
+
+...
+
+---
+name: issue139317_foldable_neg_imm
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_foldable_neg_imm
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 -16
+ ; CHECK-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 1006632960
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_1]], %subreg.sub0, killed [[S_MOV_B32_]], %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 8, 4294967295, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_32 = S_MOV_B32 -16
+ %2:sreg_32 = S_MOV_B32 1006632960
+ %3:sgpr_64 = REG_SEQUENCE killed %2:sreg_32, %subreg.sub0, killed
%1:sreg_32, %subreg.sub1
+ %4:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %3.sub1:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %4
+...
+
+---
+name: issue139317_foldable_fp_imm_0
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_foldable_fp_imm_0
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 15360
+ ; CHECK-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 1006632960
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_1]], %subreg.sub0, killed [[S_MOV_B32_]], %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 8, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_32 = S_MOV_B32 15360
+ %2:sreg_32 = S_MOV_B32 1006632960
+ %3:sgpr_64 = REG_SEQUENCE killed %2:sreg_32, %subreg.sub0, killed
%1:sreg_32, %subreg.sub1
+ %4:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %3.sub1:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %4
+...
+
+---
+name: issue139317_foldable_fp_imm_1
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_foldable_fp_imm_1
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 1006632960
+ ; CHECK-NEXT: [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 15360
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B32_1]], %subreg.sub0, killed [[S_MOV_B32_]], %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 8, 0, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_32 = S_MOV_B32 1006632960
+ %2:sreg_32 = S_MOV_B32 15360
+ %3:sgpr_64 = REG_SEQUENCE killed %2:sreg_32, %subreg.sub0, killed
%1:sreg_32, %subreg.sub1
+ %4:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %3.sub1:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %4
+...
+
+---
+name: issue139317_reg_sequence_subreg_use
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_reg_sequence_subreg_use
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO
4755871576254054400
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B]].sub0, %subreg.sub0, [[S_MOV_B]].sub1, %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 8, [[REG_SEQUENCE]].sub1, 0, 0, 0, 0, 0, implicit $mode, implicit
$exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_64 = S_MOV_B64_IMM_PSEUDO 4755871576254054400
+ %2:sgpr_64 = REG_SEQUENCE killed %1.sub0, %subreg.sub0, %1.sub1,
%subreg.sub1
+ %3:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %2.sub1:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %3
+...
+
+---
+name: issue139317_reg_sequence_subreg_use_foldable_imm
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $vgpr0
+
+ ; CHECK-LABEL: name: issue139317_reg_sequence_subreg_use_foldable_imm
+ ; CHECK: liveins: $vgpr0
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+ ; CHECK-NEXT: [[S_MOV_B:%[0-9]+]]:sreg_64 = S_MOV_B64_IMM_PSEUDO
65971704299520
+ ; CHECK-NEXT: [[REG_SEQUENCE:%[0-9]+]]:sgpr_64 = REG_SEQUENCE killed
[[S_MOV_B]].sub0, %subreg.sub0, [[S_MOV_B]].sub1, %subreg.sub1
+ ; CHECK-NEXT: [[V_PK_ADD_F16_:%[0-9]+]]:vgpr_32 = nofpexcept V_PK_ADD_F16
8, [[COPY]], 8, 15360, 0, 0, 0, 0, 0, implicit $mode, implicit $exec
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[V_PK_ADD_F16_]]
+ %0:vgpr_32 = COPY $vgpr0
+ %1:sreg_64 = S_MOV_B64_IMM_PSEUDO 65971704299520
+ %2:sgpr_64 = REG_SEQUENCE killed %1.sub0, %subreg.sub0, %1.sub1,
%subreg.sub1
+ %3:vgpr_32 = nofpexcept V_PK_ADD_F16 8, %0:vgpr_32, 8, %2.sub1:sgpr_64, 0,
0, 0, 0, 0, implicit $mode, implicit $exec
+ S_ENDPGM 0, implicit %3
+...
diff --git a/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.mir
b/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.mir
index 8d6c3efb5ded5..38b4533a14895 100644
--- a/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.mir
+++ b/llvm/test/CodeGen/AMDGPU/si-fold-operands-subreg-imm.mir
@@ -149,3 +149,29 @@ body: |
S_ENDPGM 0, implicit $scc
...
+
+---
+name: s_mov_b64_copy_subreg_into_reg_sequence_user
+tracksRegLiveness: true
+body: |
+ bb.0:
+ liveins: $sgpr8, $sgpr9
+
+ ; CHECK-LABEL: name: s_mov_b64_copy_subreg_into_reg_sequence_user
+ ; CHECK: liveins: $sgpr8, $sgpr9
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr8
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:sreg_32 = COPY $sgpr9
+ ; CHECK-NEXT: [[S_ADD_U32_:%[0-9]+]]:sreg_32 = S_ADD_U32 [[COPY]], 16,
implicit-def $scc
+ ; CHECK-NEXT: [[S_ADDC_U32_:%[0-9]+]]:sreg_32 = S_ADDC_U32 [[COPY1]], 0,
implicit-def $scc, implicit $scc
+ ; CHECK-NEXT: S_ENDPGM 0, implicit [[S_ADD_U32_]], implicit [[S_ADDC_U32_]]
+ %0:sreg_32 = COPY $sgpr8
+ %1:sreg_32 = COPY $sgpr9
+ %2:sreg_64 = S_MOV_B64 16
+ %3:sreg_32 = COPY %2.sub0
+ %4:sreg_32 = COPY %2.sub1
+ %5:sreg_32 = S_ADD_U32 %0, %3, implicit-def $scc
+ %6:sreg_32 = S_ADDC_U32 %1, %4, implicit-def $scc, implicit $scc
+ S_ENDPGM 0, implicit %5, implicit %6
+
+...
_______________________________________________
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
