1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 03:02:36 +01:00
llvm-mirror/test/CodeGen/AMDGPU/GlobalISel/legalize-shuffle-vector.mir
Matt Arsenault e7dc03b43a GlobalISel: Implement bitcast action for G_EXTRACT_VECTOR_ELEMENT
For AMDGPU, vectors with elements < 32 bits should be indexed in
32-bit elements and the desired bits extracted from there. For
elements > 64-bits, these should be reduce to 64/32 elements to enable
the normal dynamic indexing paths.

In the dynamic index cases, this produces shorter code most of the
time. This does immediately regress the constant index cases, but this
should be fixed once we have the most basic of shift combines.

The element size > 64 case is pretty much ported from the exisiting
DAG implementation for extract element promote. The increasing element
size case is new.
2020-08-02 10:42:07 -04:00

373 lines
16 KiB
YAML

# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=fiji -O0 -run-pass=legalizer %s -o - | FileCheck %s
---
name: shufflevector_scalar_src
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; CHECK-LABEL: name: shufflevector_scalar_src
; CHECK: liveins: $vgpr0, $vgpr1
; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 1)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s32_0_1
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v2s32_0_1
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 0
; CHECK: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 32
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[EXTRACT]](s32), [[EXTRACT1]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 1)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s32_1_0
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v2s32_1_0
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 32
; CHECK: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 0
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[EXTRACT]](s32), [[EXTRACT1]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s32_0_0
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v2s32_0_0
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 0
; CHECK: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 0
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[EXTRACT]](s32), [[EXTRACT1]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 0)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s32_undef_undef
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v2s32_undef_undef
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; CHECK: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, undef)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s32_undef_0
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v2s32_undef_0
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; CHECK: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 0
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[EXTRACT]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 0)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s32_0_undef
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v2s32_0_undef
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $vgpr2_vgpr3
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<2 x s32>), 0
; CHECK: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[EXTRACT]](s32), [[DEF]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<2 x s32>) = COPY $vgpr0_vgpr1
%1:_(<2 x s32>) = COPY $vgpr2_vgpr3
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, undef)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v3s32_3_2_1
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1_vgpr2, $vgpr3_vgpr4_vgpr5
; CHECK-LABEL: name: shufflevector_v3s32_3_2_1
; CHECK: liveins: $vgpr0_vgpr1_vgpr2, $vgpr3_vgpr4_vgpr5
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK: [[COPY1:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr3_vgpr4_vgpr5
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY1]](<3 x s32>), 0
; CHECK: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 64
; CHECK: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 32
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[EXTRACT]](s32), [[EXTRACT1]](s32), [[EXTRACT2]](s32)
; CHECK: $vgpr0_vgpr1_vgpr2 = COPY [[BUILD_VECTOR]](<3 x s32>)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(<3 x s32>) = COPY $vgpr3_vgpr4_vgpr5
%2:_(<3 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 2, 1)
$vgpr0_vgpr1_vgpr2 = COPY %2
...
---
name: shufflevector_v3s32_3_2_1_smaller
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1_vgpr2, $vgpr3_vgpr4_vgpr5
; CHECK-LABEL: name: shufflevector_v3s32_3_2_1_smaller
; CHECK: liveins: $vgpr0_vgpr1_vgpr2, $vgpr3_vgpr4_vgpr5
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK: [[COPY1:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr3_vgpr4_vgpr5
; CHECK: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 64
; CHECK: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 32
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[EXTRACT]](s32), [[EXTRACT1]](s32)
; CHECK: $vgpr0_vgpr1 = COPY [[BUILD_VECTOR]](<2 x s32>)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(<3 x s32>) = COPY $vgpr3_vgpr4_vgpr5
%2:_(<2 x s32>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 1)
$vgpr0_vgpr1 = COPY %2
...
---
name: shufflevector_v2s16_0_1
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; CHECK-LABEL: name: shufflevector_v2s16_0_1
; CHECK: liveins: $vgpr0, $vgpr1
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; CHECK: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; CHECK: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; CHECK: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; CHECK: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; CHECK: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; CHECK: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; CHECK: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; CHECK: [[COPY2:%[0-9]+]]:_(s32) = COPY [[LSHR]](s32)
; CHECK: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY2]], [[C2]]
; CHECK: [[COPY3:%[0-9]+]]:_(s32) = COPY [[LSHR1]](s32)
; CHECK: [[AND1:%[0-9]+]]:_(s32) = G_AND [[COPY3]], [[C2]]
; CHECK: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; CHECK: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; CHECK: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; CHECK: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 1)
$vgpr0 = COPY %2
...
---
name: shufflevector_v2s16_1_0
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1
; CHECK-LABEL: name: shufflevector_v2s16_1_0
; CHECK: liveins: $vgpr0, $vgpr1
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; CHECK: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; CHECK: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; CHECK: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; CHECK: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; CHECK: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; CHECK: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; CHECK: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; CHECK: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; CHECK: [[COPY2:%[0-9]+]]:_(s32) = COPY [[LSHR]](s32)
; CHECK: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY2]], [[C2]]
; CHECK: [[COPY3:%[0-9]+]]:_(s32) = COPY [[LSHR1]](s32)
; CHECK: [[AND1:%[0-9]+]]:_(s32) = G_AND [[COPY3]], [[C2]]
; CHECK: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; CHECK: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; CHECK: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; CHECK: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0)
$vgpr0 = COPY %2
...
---
name: shufflevector_v3s16_2_0
tracksRegLiveness: true
body: |
bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK-LABEL: name: shufflevector_v3s16_2_0
; CHECK: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; CHECK: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; CHECK: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; CHECK: [[EXTRACT:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY]](<4 x s16>), 0
; CHECK: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0
; CHECK: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
; CHECK: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; CHECK: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>)
; CHECK: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; CHECK: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; CHECK: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; CHECK: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; CHECK: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; CHECK: [[COPY2:%[0-9]+]]:_(s32) = COPY [[BITCAST]](s32)
; CHECK: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY2]], 16
; CHECK: [[COPY3:%[0-9]+]]:_(s32) = COPY [[LSHR]](s32)
; CHECK: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY3]], 16
; CHECK: [[COPY4:%[0-9]+]]:_(s32) = COPY [[BITCAST1]](s32)
; CHECK: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY4]], 16
; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64
; CHECK: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; CHECK: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>)
; CHECK: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; CHECK: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; CHECK: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; CHECK: [[LSHR3:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST3]], [[C]](s32)
; CHECK: [[COPY5:%[0-9]+]]:_(s32) = COPY [[BITCAST2]](s32)
; CHECK: [[SEXT_INREG3:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY5]], 16
; CHECK: [[COPY6:%[0-9]+]]:_(s32) = COPY [[LSHR2]](s32)
; CHECK: [[SEXT_INREG4:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY6]], 16
; CHECK: [[COPY7:%[0-9]+]]:_(s32) = COPY [[BITCAST3]](s32)
; CHECK: [[SEXT_INREG5:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY7]], 16
; CHECK: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG3]](s32), [[SEXT_INREG4]](s32), [[SEXT_INREG5]](s32)
; CHECK: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32
; CHECK: [[INSERT2:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; CHECK: [[UV4:%[0-9]+]]:_(<2 x s16>), [[UV5:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT2]](<4 x s16>)
; CHECK: [[BITCAST4:%[0-9]+]]:_(s32) = G_BITCAST [[UV4]](<2 x s16>)
; CHECK: [[LSHR4:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST4]], [[C]](s32)
; CHECK: [[BITCAST5:%[0-9]+]]:_(s32) = G_BITCAST [[UV5]](<2 x s16>)
; CHECK: [[LSHR5:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST5]], [[C]](s32)
; CHECK: [[COPY8:%[0-9]+]]:_(s32) = COPY [[BITCAST4]](s32)
; CHECK: [[SEXT_INREG6:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY8]], 16
; CHECK: [[COPY9:%[0-9]+]]:_(s32) = COPY [[LSHR4]](s32)
; CHECK: [[SEXT_INREG7:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY9]], 16
; CHECK: [[COPY10:%[0-9]+]]:_(s32) = COPY [[BITCAST5]](s32)
; CHECK: [[SEXT_INREG8:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY10]], 16
; CHECK: [[BUILD_VECTOR2:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG6]](s32), [[SEXT_INREG7]](s32), [[SEXT_INREG8]](s32)
; CHECK: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR2]](<3 x s32>), 0
; CHECK: [[INSERT3:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; CHECK: [[UV6:%[0-9]+]]:_(<2 x s16>), [[UV7:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT3]](<4 x s16>)
; CHECK: [[BITCAST6:%[0-9]+]]:_(s32) = G_BITCAST [[UV6]](<2 x s16>)
; CHECK: [[LSHR6:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST6]], [[C]](s32)
; CHECK: [[BITCAST7:%[0-9]+]]:_(s32) = G_BITCAST [[UV7]](<2 x s16>)
; CHECK: [[LSHR7:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST7]], [[C]](s32)
; CHECK: [[COPY11:%[0-9]+]]:_(s32) = COPY [[BITCAST6]](s32)
; CHECK: [[SEXT_INREG9:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY11]], 16
; CHECK: [[COPY12:%[0-9]+]]:_(s32) = COPY [[LSHR6]](s32)
; CHECK: [[SEXT_INREG10:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY12]], 16
; CHECK: [[COPY13:%[0-9]+]]:_(s32) = COPY [[BITCAST7]](s32)
; CHECK: [[SEXT_INREG11:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY13]], 16
; CHECK: [[BUILD_VECTOR3:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG9]](s32), [[SEXT_INREG10]](s32), [[SEXT_INREG11]](s32)
; CHECK: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR3]](<3 x s32>), 0
; CHECK: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; CHECK: [[COPY14:%[0-9]+]]:_(s32) = COPY [[EXTRACT2]](s32)
; CHECK: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY14]], [[C1]]
; CHECK: [[COPY15:%[0-9]+]]:_(s32) = COPY [[EXTRACT3]](s32)
; CHECK: [[AND1:%[0-9]+]]:_(s32) = G_AND [[COPY15]], [[C1]]
; CHECK: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; CHECK: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; CHECK: [[BITCAST8:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; CHECK: [[COPY16:%[0-9]+]]:_(s32) = COPY [[EXTRACT4]](s32)
; CHECK: [[AND2:%[0-9]+]]:_(s32) = G_AND [[COPY16]], [[C1]]
; CHECK: [[COPY17:%[0-9]+]]:_(s32) = COPY [[EXTRACT5]](s32)
; CHECK: [[AND3:%[0-9]+]]:_(s32) = G_AND [[COPY17]], [[C1]]
; CHECK: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[AND3]], [[C]](s32)
; CHECK: [[OR1:%[0-9]+]]:_(s32) = G_OR [[AND2]], [[SHL1]]
; CHECK: [[BITCAST9:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR1]](s32)
; CHECK: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BITCAST8]](<2 x s16>), [[BITCAST9]](<2 x s16>)
; CHECK: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
%0:_(<4 x s16>) = COPY $vgpr0_vgpr1
%1:_(<4 x s16>) = COPY $vgpr2_vgpr3
%2:_(<3 x s16>) = G_EXTRACT %0, 0
%3:_(<3 x s16>) = G_EXTRACT %1, 0
%4:_(<4 x s16>) = G_SHUFFLE_VECTOR %2, %3, shufflemask(5, 1, 3, 0)
$vgpr0_vgpr1 = COPY %4
...