This is an archive of the discontinued LLVM Phabricator instance.

Differential D110469

[GlobalISel] Widen G_EXTRACT_VECTOR_ELT using anyext instead of sext.
ClosedPublic

Authored by aemerson on Sep 24 2021, 10:56 PM.

Details

Reviewers
arsenm
foad
Commits
rGdafcbfdaa0cd: [GlobalISel] Widen G_EXTRACT_VECTOR_ELT using anyext instead of sext.
Summary

G_SEXT seems to be unnecessary here, anyext will do.

Diff Detail

Event Timeline

aemerson created this revision.Sep 24 2021, 10:56 PM
Herald added subscribers: kerbowa, hiraditya, rovka and 2 others. · View Herald TranscriptSep 24 2021, 10:56 PM
aemerson requested review of this revision.Sep 24 2021, 10:56 PM
Herald added a subscriber: wdng. · View Herald TranscriptSep 24 2021, 10:56 PM
Harbormaster completed remote builds in B125685: Diff 375015.Sep 24 2021, 11:17 PM
aemerson added a comment.Oct 3 2021, 11:07 PM

ping

foad accepted this revision.Oct 4 2021, 1:10 AM

LGTM.

This revision is now accepted and ready to land.Oct 4 2021, 1:10 AM
This revision was landed with ongoing or failed builds.Oct 4 2021, 12:19 PM
Closed by commit rGdafcbfdaa0cd: [GlobalISel] Widen G_EXTRACT_VECTOR_ELT using anyext instead of sext. (authored by aemerson). · Explain Why
This revision was automatically updated to reflect the committed changes.
aemerson added a commit: rGdafcbfdaa0cd: [GlobalISel] Widen G_EXTRACT_VECTOR_ELT using anyext instead of sext..

Revision Contents

PathSize
llvm/
lib/
CodeGen/
GlobalISel/
2 lines
test/
CodeGen/
AArch64/
GlobalISel/
24 lines
AMDGPU/
GlobalISel/
76 lines
20 lines
460 lines

Diff 376991

llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp

  • This file is larger than 256 KB, so syntax highlighting is disabled by default.
Show First 20 Lines • Show All 2,429 Lines • ▼ Show 20 LinesLegalizerHelper::widenScalar(MachineInstr &MI, unsigned TypeIdx, LLT WideTy) {
case TargetOpcode::G_EXTRACT_VECTOR_ELT: { case TargetOpcode::G_EXTRACT_VECTOR_ELT: {
if (TypeIdx == 0) { if (TypeIdx == 0) {
Register VecReg = MI.getOperand(1).getReg(); Register VecReg = MI.getOperand(1).getReg();
LLT VecTy = MRI.getType(VecReg); LLT VecTy = MRI.getType(VecReg);
Observer.changingInstr(MI); Observer.changingInstr(MI);
widenScalarSrc( widenScalarSrc(
MI, LLT::vector(VecTy.getElementCount(), WideTy.getSizeInBits()), 1, MI, LLT::vector(VecTy.getElementCount(), WideTy.getSizeInBits()), 1,
TargetOpcode::G_SEXT); TargetOpcode::G_ANYEXT);
widenScalarDst(MI, WideTy, 0); widenScalarDst(MI, WideTy, 0);
Observer.changedInstr(MI); Observer.changedInstr(MI);
return Legalized; return Legalized;
} }
if (TypeIdx != 2) if (TypeIdx != 2)
return UnableToLegalize; return UnableToLegalize;
▲ Show 20 Lines • Show All 5,541 Lines • Show Last 20 Lines

llvm/test/CodeGen/AArch64/GlobalISel/legalize-extract-vector-elt.mir

Show All 22 Lines
body: | body: |
bb.0: bb.0:
liveins: $q0, $q1, $x0 liveins: $q0, $q1, $x0
; CHECK-LABEL: name: test_eve_v2s1 ; CHECK-LABEL: name: test_eve_v2s1
; CHECK: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $q0 ; CHECK: [[COPY:%[0-9]+]]:_(<2 x s64>) = COPY $q0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $q1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s64>) = COPY $q1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0 ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<2 x s64>) = G_ICMP intpred(eq), [[COPY]](<2 x s64>), [[COPY1]] ; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<2 x s64>) = G_ICMP intpred(eq), [[COPY]](<2 x s64>), [[COPY1]]
; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 63 ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s64) = G_EXTRACT_VECTOR_ELT [[ICMP]](<2 x s64>), [[COPY2]](s64)
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
; CHECK-NEXT: [[SHL:%[0-9]+]]:_(<2 x s64>) = G_SHL [[ICMP]], [[BUILD_VECTOR]](<2 x s64>)
; CHECK-NEXT: [[ASHR:%[0-9]+]]:_(<2 x s64>) = G_ASHR [[SHL]], [[BUILD_VECTOR]](<2 x s64>)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s64) = G_EXTRACT_VECTOR_ELT [[ASHR]](<2 x s64>), [[COPY2]](s64)
; CHECK-NEXT: $x0 = COPY [[EVEC]](s64) ; CHECK-NEXT: $x0 = COPY [[EVEC]](s64)
; CHECK-NEXT: RET_ReallyLR ; CHECK-NEXT: RET_ReallyLR
%0:_(<2 x s64>) = COPY $q0 %0:_(<2 x s64>) = COPY $q0
%1:_(<2 x s64>) = COPY $q1 %1:_(<2 x s64>) = COPY $q1
%2:_(s64) = COPY $x0 %2:_(s64) = COPY $x0
%3:_(<2 x s1>) = G_ICMP intpred(eq), %0(<2 x s64>), %1 %3:_(<2 x s1>) = G_ICMP intpred(eq), %0(<2 x s64>), %1
%4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<2 x s1>), %2:_(s64) %4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<2 x s1>), %2:_(s64)
%5:_(s64) = G_ANYEXT %4(s1) %5:_(s64) = G_ANYEXT %4(s1)
$x0 = COPY %5(s64) $x0 = COPY %5(s64)
RET_ReallyLR RET_ReallyLR
... ...
--- ---
name: test_eve_v4s1 name: test_eve_v4s1
body: | body: |
bb.0: bb.0:
liveins: $q0, $q1, $x0 liveins: $q0, $q1, $x0
; CHECK-LABEL: name: test_eve_v4s1 ; CHECK-LABEL: name: test_eve_v4s1
; CHECK: [[COPY:%[0-9]+]]:_(<4 x s32>) = COPY $q0 ; CHECK: [[COPY:%[0-9]+]]:_(<4 x s32>) = COPY $q0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s32>) = COPY $q1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s32>) = COPY $q1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0 ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<4 x s32>) = G_ICMP intpred(eq), [[COPY]](<4 x s32>), [[COPY1]] ; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<4 x s32>) = G_ICMP intpred(eq), [[COPY]](<4 x s32>), [[COPY1]]
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 31 ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[ICMP]](<4 x s32>), [[COPY2]](s64)
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
; CHECK-NEXT: [[SHL:%[0-9]+]]:_(<4 x s32>) = G_SHL [[ICMP]], [[BUILD_VECTOR]](<4 x s32>)
; CHECK-NEXT: [[ASHR:%[0-9]+]]:_(<4 x s32>) = G_ASHR [[SHL]], [[BUILD_VECTOR]](<4 x s32>)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[ASHR]](<4 x s32>), [[COPY2]](s64)
; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s64) = G_ANYEXT [[EVEC]](s32) ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s64) = G_ANYEXT [[EVEC]](s32)
; CHECK-NEXT: $x0 = COPY [[ANYEXT]](s64) ; CHECK-NEXT: $x0 = COPY [[ANYEXT]](s64)
; CHECK-NEXT: RET_ReallyLR ; CHECK-NEXT: RET_ReallyLR
%0:_(<4 x s32>) = COPY $q0 %0:_(<4 x s32>) = COPY $q0
%1:_(<4 x s32>) = COPY $q1 %1:_(<4 x s32>) = COPY $q1
%2:_(s64) = COPY $x0 %2:_(s64) = COPY $x0
%3:_(<4 x s1>) = G_ICMP intpred(eq), %0(<4 x s32>), %1 %3:_(<4 x s1>) = G_ICMP intpred(eq), %0(<4 x s32>), %1
%4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<4 x s1>), %2:_(s64) %4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<4 x s1>), %2:_(s64)
%5:_(s64) = G_ANYEXT %4(s1) %5:_(s64) = G_ANYEXT %4(s1)
$x0 = COPY %5(s64) $x0 = COPY %5(s64)
RET_ReallyLR RET_ReallyLR
... ...
--- ---
name: test_eve_v8s1 name: test_eve_v8s1
body: | body: |
bb.0: bb.0:
liveins: $q0, $q1, $x0 liveins: $q0, $q1, $x0
; CHECK-LABEL: name: test_eve_v8s1 ; CHECK-LABEL: name: test_eve_v8s1
; CHECK: [[COPY:%[0-9]+]]:_(<8 x s16>) = COPY $q0 ; CHECK: [[COPY:%[0-9]+]]:_(<8 x s16>) = COPY $q0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<8 x s16>) = COPY $q1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<8 x s16>) = COPY $q1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0 ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<8 x s16>) = G_ICMP intpred(eq), [[COPY]](<8 x s16>), [[COPY1]] ; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<8 x s16>) = G_ICMP intpred(eq), [[COPY]](<8 x s16>), [[COPY1]]
; CHECK-NEXT: [[C:%[0-9]+]]:_(s16) = G_CONSTANT i16 15 ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s16) = G_EXTRACT_VECTOR_ELT [[ICMP]](<8 x s16>), [[COPY2]](s64)
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[C]](s16), [[C]](s16), [[C]](s16), [[C]](s16), [[C]](s16), [[C]](s16), [[C]](s16), [[C]](s16)
; CHECK-NEXT: [[SHL:%[0-9]+]]:_(<8 x s16>) = G_SHL [[ICMP]], [[BUILD_VECTOR]](<8 x s16>)
; CHECK-NEXT: [[ASHR:%[0-9]+]]:_(<8 x s16>) = G_ASHR [[SHL]], [[BUILD_VECTOR]](<8 x s16>)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s16) = G_EXTRACT_VECTOR_ELT [[ASHR]](<8 x s16>), [[COPY2]](s64)
; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s64) = G_ANYEXT [[EVEC]](s16) ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s64) = G_ANYEXT [[EVEC]](s16)
; CHECK-NEXT: $x0 = COPY [[ANYEXT]](s64) ; CHECK-NEXT: $x0 = COPY [[ANYEXT]](s64)
; CHECK-NEXT: RET_ReallyLR ; CHECK-NEXT: RET_ReallyLR
%0:_(<8 x s16>) = COPY $q0 %0:_(<8 x s16>) = COPY $q0
%1:_(<8 x s16>) = COPY $q1 %1:_(<8 x s16>) = COPY $q1
%2:_(s64) = COPY $x0 %2:_(s64) = COPY $x0
%3:_(<8 x s1>) = G_ICMP intpred(eq), %0(<8 x s16>), %1 %3:_(<8 x s1>) = G_ICMP intpred(eq), %0(<8 x s16>), %1
%4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<8 x s1>), %2:_(s64) %4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<8 x s1>), %2:_(s64)
%5:_(s64) = G_ANYEXT %4(s1) %5:_(s64) = G_ANYEXT %4(s1)
$x0 = COPY %5(s64) $x0 = COPY %5(s64)
RET_ReallyLR RET_ReallyLR
... ...
--- ---
name: test_eve_v16s1 name: test_eve_v16s1
body: | body: |
bb.0: bb.0:
liveins: $q0, $q1, $x0 liveins: $q0, $q1, $x0
; CHECK-LABEL: name: test_eve_v16s1 ; CHECK-LABEL: name: test_eve_v16s1
; CHECK: [[COPY:%[0-9]+]]:_(<16 x s8>) = COPY $q0 ; CHECK: [[COPY:%[0-9]+]]:_(<16 x s8>) = COPY $q0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<16 x s8>) = COPY $q1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<16 x s8>) = COPY $q1
; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0 ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s64) = COPY $x0
; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<16 x s8>) = G_ICMP intpred(eq), [[COPY]](<16 x s8>), [[COPY1]] ; CHECK-NEXT: [[ICMP:%[0-9]+]]:_(<16 x s8>) = G_ICMP intpred(eq), [[COPY]](<16 x s8>), [[COPY1]]
; CHECK-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 7 ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s8) = G_EXTRACT_VECTOR_ELT [[ICMP]](<16 x s8>), [[COPY2]](s64)
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8)
; CHECK-NEXT: [[SHL:%[0-9]+]]:_(<16 x s8>) = G_SHL [[ICMP]], [[BUILD_VECTOR]](<16 x s8>)
; CHECK-NEXT: [[ASHR:%[0-9]+]]:_(<16 x s8>) = G_ASHR [[SHL]], [[BUILD_VECTOR]](<16 x s8>)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s8) = G_EXTRACT_VECTOR_ELT [[ASHR]](<16 x s8>), [[COPY2]](s64)
; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s64) = G_ANYEXT [[EVEC]](s8) ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(s64) = G_ANYEXT [[EVEC]](s8)
; CHECK-NEXT: $x0 = COPY [[ANYEXT]](s64) ; CHECK-NEXT: $x0 = COPY [[ANYEXT]](s64)
; CHECK-NEXT: RET_ReallyLR ; CHECK-NEXT: RET_ReallyLR
%0:_(<16 x s8>) = COPY $q0 %0:_(<16 x s8>) = COPY $q0
%1:_(<16 x s8>) = COPY $q1 %1:_(<16 x s8>) = COPY $q1
%2:_(s64) = COPY $x0 %2:_(s64) = COPY $x0
%3:_(<16 x s1>) = G_ICMP intpred(eq), %0(<16 x s8>), %1 %3:_(<16 x s1>) = G_ICMP intpred(eq), %0(<16 x s8>), %1
%4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<16 x s1>), %2:_(s64) %4:_(s1) = G_EXTRACT_VECTOR_ELT %3:_(<16 x s1>), %2:_(s64)
▲ Show 20 Lines • Show All 150 Lines • Show Last 20 Lines

llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-extract-vector-elt.mir

Show First 20 Lines • Show All 199 Lines • ▼ Show 20 Lines
--- ---
name: extract_vector_elt_0_v2i8_i32 name: extract_vector_elt_0_v2i8_i32
body: | body: |
bb.0: bb.0:
; CHECK-LABEL: name: extract_vector_elt_0_v2i8_i32 ; CHECK-LABEL: name: extract_vector_elt_0_v2i8_i32
; CHECK: [[DEF:%[0-9]+]]:_(<2 x s32>) = G_IMPLICIT_DEF ; CHECK: [[DEF:%[0-9]+]]:_(<2 x s32>) = G_IMPLICIT_DEF
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[DEF]](<2 x s32>) ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[DEF]](<2 x s32>), 0
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 8
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 8
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 0
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(<2 x s8>) = G_IMPLICIT_DEF %0:_(<2 x s8>) = G_IMPLICIT_DEF
%1:_(s32) = G_CONSTANT i32 0 %1:_(s32) = G_CONSTANT i32 0
%2:_(s8) = G_EXTRACT_VECTOR_ELT %0, %1 %2:_(s8) = G_EXTRACT_VECTOR_ELT %0, %1
%3:_(s32) = G_ANYEXT %2 %3:_(s32) = G_ANYEXT %2
$vgpr0 = COPY %3 $vgpr0 = COPY %3
... ...
Show All 19 Lines
--- ---
name: extract_vector_elt_0_v2i1_i32 name: extract_vector_elt_0_v2i1_i32
body: | body: |
bb.0: bb.0:
; CHECK-LABEL: name: extract_vector_elt_0_v2i1_i32 ; CHECK-LABEL: name: extract_vector_elt_0_v2i1_i32
; CHECK: [[DEF:%[0-9]+]]:_(<2 x s32>) = G_IMPLICIT_DEF ; CHECK: [[DEF:%[0-9]+]]:_(<2 x s32>) = G_IMPLICIT_DEF
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[DEF]](<2 x s32>) ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[DEF]](<2 x s32>), 0
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 1
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 1
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 0
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(<2 x s1>) = G_IMPLICIT_DEF %0:_(<2 x s1>) = G_IMPLICIT_DEF
%1:_(s32) = G_CONSTANT i32 0 %1:_(s32) = G_CONSTANT i32 0
%2:_(s1) = G_EXTRACT_VECTOR_ELT %0, %1 %2:_(s1) = G_EXTRACT_VECTOR_ELT %0, %1
%3:_(s32) = G_ANYEXT %2 %3:_(s32) = G_ANYEXT %2
$vgpr0 = COPY %3 $vgpr0 = COPY %3
... ...
--- ---
name: extract_vector_elt_0_v2i1_i1 name: extract_vector_elt_0_v2i1_i1
body: | body: |
bb.0: bb.0:
; CHECK-LABEL: name: extract_vector_elt_0_v2i1_i1 ; CHECK-LABEL: name: extract_vector_elt_0_v2i1_i1
; CHECK: [[DEF:%[0-9]+]]:_(<2 x s32>) = G_IMPLICIT_DEF ; CHECK: [[DEF:%[0-9]+]]:_(<2 x s32>) = G_IMPLICIT_DEF
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[DEF]](<2 x s32>) ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[DEF]](<2 x s32>), 0
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 1
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 1
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 0
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(<2 x s1>) = G_IMPLICIT_DEF %0:_(<2 x s1>) = G_IMPLICIT_DEF
%1:_(s1) = G_CONSTANT i1 false %1:_(s1) = G_CONSTANT i1 false
%2:_(s1) = G_EXTRACT_VECTOR_ELT %0, %1 %2:_(s1) = G_EXTRACT_VECTOR_ELT %0, %1
%3:_(s32) = G_ANYEXT %2 %3:_(s32) = G_ANYEXT %2
$vgpr0 = COPY %3 $vgpr0 = COPY %3
... ...
--- ---
name: extract_vector_elt_v2s8_varidx_i32 name: extract_vector_elt_v2s8_varidx_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; CHECK-LABEL: name: extract_vector_elt_v2s8_varidx_i32 ; CHECK-LABEL: name: extract_vector_elt_v2s8_varidx_i32
; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0 ; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8 ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8
; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32) ; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY]], 8 ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[LSHR]](s32)
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 8
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<2 x s32>), [[COPY1]](s32) ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<2 x s32>), [[COPY1]](s32)
; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32)
%0:_(s32) = COPY $vgpr0 %0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1 %1:_(s32) = COPY $vgpr1
%2:_(s16) = G_TRUNC %0 %2:_(s16) = G_TRUNC %0
%3:_(<2 x s8>) = G_BITCAST %2 %3:_(<2 x s8>) = G_BITCAST %2
%4:_(s8) = G_EXTRACT_VECTOR_ELT %3, %1 %4:_(s8) = G_EXTRACT_VECTOR_ELT %3, %1
%5:_(s32) = G_ANYEXT %4 %5:_(s32) = G_ANYEXT %4
$vgpr0 = COPY %5 $vgpr0 = COPY %5
... ...
--- ---
name: extract_vector_elt_v2s8_constidx_0_i32 name: extract_vector_elt_v2s8_constidx_0_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0 liveins: $vgpr0
; CHECK-LABEL: name: extract_vector_elt_v2s8_constidx_0_i32 ; CHECK-LABEL: name: extract_vector_elt_v2s8_constidx_0_i32
; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0 ; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8 ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8
; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32) ; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY]], 8 ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[LSHR]](s32)
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 8
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 0 ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 0
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(s32) = COPY $vgpr0 %0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1 %1:_(s32) = COPY $vgpr1
%2:_(s16) = G_TRUNC %0 %2:_(s16) = G_TRUNC %0
%3:_(<2 x s8>) = G_BITCAST %2 %3:_(<2 x s8>) = G_BITCAST %2
%4:_(s32) = G_CONSTANT i32 0 %4:_(s32) = G_CONSTANT i32 0
%5:_(s8) = G_EXTRACT_VECTOR_ELT %3, %4 %5:_(s8) = G_EXTRACT_VECTOR_ELT %3, %4
%6:_(s32) = G_ANYEXT %5 %6:_(s32) = G_ANYEXT %5
$vgpr0 = COPY %6 $vgpr0 = COPY %6
... ...
--- ---
name: extract_vector_elt_v2s8_constidx_1_i32 name: extract_vector_elt_v2s8_constidx_1_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0 liveins: $vgpr0
; CHECK-LABEL: name: extract_vector_elt_v2s8_constidx_1_i32 ; CHECK-LABEL: name: extract_vector_elt_v2s8_constidx_1_i32
; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0 ; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8 ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8
; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32) ; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY]], 8 ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[LSHR]](s32)
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 8
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 32 ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<2 x s32>), 32
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(s32) = COPY $vgpr0 %0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1 %1:_(s32) = COPY $vgpr1
%2:_(s16) = G_TRUNC %0 %2:_(s16) = G_TRUNC %0
%3:_(<2 x s8>) = G_BITCAST %2 %3:_(<2 x s8>) = G_BITCAST %2
%4:_(s32) = G_CONSTANT i32 1 %4:_(s32) = G_CONSTANT i32 1
%5:_(s8) = G_EXTRACT_VECTOR_ELT %3, %4 %5:_(s8) = G_EXTRACT_VECTOR_ELT %3, %4
Show All 12 Lines bb.0:
; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0 ; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 4 ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 4
; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32) ; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32)
; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 8 ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 8
; CHECK-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C1]](s32) ; CHECK-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C1]](s32)
; CHECK-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 12 ; CHECK-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 12
; CHECK-NEXT: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C2]](s32) ; CHECK-NEXT: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C2]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY]], 4 ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[LSHR]](s32), [[LSHR1]](s32), [[LSHR2]](s32)
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 4
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR1]], 4
; CHECK-NEXT: [[SEXT_INREG3:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR2]], 4
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32), [[SEXT_INREG3]](s32)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<4 x s32>), [[COPY1]](s32) ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<4 x s32>), [[COPY1]](s32)
; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32)
%0:_(s32) = COPY $vgpr0 %0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1 %1:_(s32) = COPY $vgpr1
%2:_(s16) = G_TRUNC %0 %2:_(s16) = G_TRUNC %0
%3:_(<4 x s4>) = G_BITCAST %2 %3:_(<4 x s4>) = G_BITCAST %2
%4:_(s4) = G_EXTRACT_VECTOR_ELT %3, %1 %4:_(s4) = G_EXTRACT_VECTOR_ELT %3, %1
%5:_(s32) = G_ANYEXT %4 %5:_(s32) = G_ANYEXT %4
$vgpr0 = COPY %5 $vgpr0 = COPY %5
... ...
--- ---
name: extract_vector_elt_v3s8_varidx_i32 name: extract_vector_elt_v3s8_varidx_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1_vgpr2, $vgpr3 liveins: $vgpr0_vgpr1_vgpr2, $vgpr3
; CHECK-LABEL: name: extract_vector_elt_v3s8_varidx_i32 ; CHECK-LABEL: name: extract_vector_elt_v3s8_varidx_i32
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 ; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr3 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr3
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<3 x s32>) ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[COPY]](<3 x s32>), [[COPY1]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 8
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 8
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV2]], 8
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<3 x s32>), [[COPY1]](s32)
; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 %0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(s32) = COPY $vgpr3 %1:_(s32) = COPY $vgpr3
%2:_(<3 x s8>) = G_TRUNC %0 %2:_(<3 x s8>) = G_TRUNC %0
%3:_(s8) = G_EXTRACT_VECTOR_ELT %2, %1 %3:_(s8) = G_EXTRACT_VECTOR_ELT %2, %1
%4:_(s32) = G_ANYEXT %3 %4:_(s32) = G_ANYEXT %3
$vgpr0 = COPY %4 $vgpr0 = COPY %4
... ...
▲ Show 20 Lines • Show All 602 Lines • ▼ Show 20 Lines
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1_vgpr2, $vgpr3 liveins: $vgpr0_vgpr1_vgpr2, $vgpr3
; CHECK-LABEL: name: extract_vector_elt_v3s16_varidx_i32 ; CHECK-LABEL: name: extract_vector_elt_v3s16_varidx_i32
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 ; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr3 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr3
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<3 x s32>) ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[COPY]](<3 x s32>), [[COPY1]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 16
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 16
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV2]], 16
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<3 x s32>), [[COPY1]](s32)
; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 %0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(s32) = COPY $vgpr3 %1:_(s32) = COPY $vgpr3
%2:_(<3 x s16>) = G_TRUNC %0 %2:_(<3 x s16>) = G_TRUNC %0
%3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1 %3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1
%4:_(s32) = G_ANYEXT %3 %4:_(s32) = G_ANYEXT %3
$vgpr0 = COPY %4 $vgpr0 = COPY %4
... ...
--- ---
name: extract_vector_elt_v3s16_idx0_i32 name: extract_vector_elt_v3s16_idx0_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1_vgpr2 liveins: $vgpr0_vgpr1_vgpr2
; CHECK-LABEL: name: extract_vector_elt_v3s16_idx0_i32 ; CHECK-LABEL: name: extract_vector_elt_v3s16_idx0_i32
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 ; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<3 x s32>) ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 0
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 16
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 16
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV2]], 16
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 0
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 %0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(s32) = G_CONSTANT i32 0 %1:_(s32) = G_CONSTANT i32 0
%2:_(<3 x s16>) = G_TRUNC %0 %2:_(<3 x s16>) = G_TRUNC %0
%3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1 %3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1
%4:_(s32) = G_ANYEXT %3 %4:_(s32) = G_ANYEXT %3
$vgpr0 = COPY %4 $vgpr0 = COPY %4
... ...
--- ---
name: extract_vector_elt_v3s16_idx1_i32 name: extract_vector_elt_v3s16_idx1_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1_vgpr2 liveins: $vgpr0_vgpr1_vgpr2
; CHECK-LABEL: name: extract_vector_elt_v3s16_idx1_i32 ; CHECK-LABEL: name: extract_vector_elt_v3s16_idx1_i32
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 ; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<3 x s32>) ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 32
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 16
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 16
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV2]], 16
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 32
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 %0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(s32) = G_CONSTANT i32 1 %1:_(s32) = G_CONSTANT i32 1
%2:_(<3 x s16>) = G_TRUNC %0 %2:_(<3 x s16>) = G_TRUNC %0
%3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1 %3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1
%4:_(s32) = G_ANYEXT %3 %4:_(s32) = G_ANYEXT %3
$vgpr0 = COPY %4 $vgpr0 = COPY %4
... ...
--- ---
name: extract_vector_elt_v3s16_idx2_i32 name: extract_vector_elt_v3s16_idx2_i32
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1_vgpr2 liveins: $vgpr0_vgpr1_vgpr2
; CHECK-LABEL: name: extract_vector_elt_v3s16_idx2_i32 ; CHECK-LABEL: name: extract_vector_elt_v3s16_idx2_i32
; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 ; CHECK: [[COPY:%[0-9]+]]:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[COPY]](<3 x s32>) ; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[COPY]](<3 x s32>), 64
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV]], 16
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV1]], 16
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[UV2]], 16
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64
; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EXTRACT]](s32)
%0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2 %0:_(<3 x s32>) = COPY $vgpr0_vgpr1_vgpr2
%1:_(s32) = G_CONSTANT i32 2 %1:_(s32) = G_CONSTANT i32 2
%2:_(<3 x s16>) = G_TRUNC %0 %2:_(<3 x s16>) = G_TRUNC %0
%3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1 %3:_(s16) = G_EXTRACT_VECTOR_ELT %2, %1
%4:_(s32) = G_ANYEXT %3 %4:_(s32) = G_ANYEXT %3
$vgpr0 = COPY %4 $vgpr0 = COPY %4
... ...
▲ Show 20 Lines • Show All 931 Lines • ▼ Show 20 Lines bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; CHECK-LABEL: name: extract_vector_elt_v3s8_varidx_s32 ; CHECK-LABEL: name: extract_vector_elt_v3s8_varidx_s32
; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0 ; CHECK: [[COPY:%[0-9]+]]:_(s32) = COPY $vgpr0
; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1 ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $vgpr1
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8 ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 8
; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32) ; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C]](s32)
; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; CHECK-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C1]](s32) ; CHECK-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[COPY]], [[C1]](s32)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[COPY]], 8 ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[LSHR]](s32), [[LSHR1]](s32)
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 8
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR1]], 8
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<3 x s32>), [[COPY1]](s32) ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<3 x s32>), [[COPY1]](s32)
; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32) ; CHECK-NEXT: $vgpr0 = COPY [[EVEC]](s32)
%0:_(s32) = COPY $vgpr0 %0:_(s32) = COPY $vgpr0
%1:_(s32) = COPY $vgpr1 %1:_(s32) = COPY $vgpr1
%2:_(s24) = G_TRUNC %0 %2:_(s24) = G_TRUNC %0
%3:_(<3 x s8>) = G_BITCAST %2 %3:_(<3 x s8>) = G_BITCAST %2
%4:_(s8) = G_EXTRACT_VECTOR_ELT %3, %1 %4:_(s8) = G_EXTRACT_VECTOR_ELT %3, %1
%5:_(s32) = G_ANYEXT %4 %5:_(s32) = G_ANYEXT %4
$vgpr0 = COPY %5 $vgpr0 = COPY %5
... ...

llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-shuffle-vector.mir

Show First 20 Lines • Show All 296 Lines • ▼ Show 20 Lines bb.0:
; CHECK-NEXT: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0 ; CHECK-NEXT: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0
; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
; CHECK-NEXT: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0 ; CHECK-NEXT: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; CHECK-NEXT: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>) ; CHECK-NEXT: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>)
; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>) ; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; CHECK-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; CHECK-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>) ; CHECK-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST]], 16 ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST]](s32), [[LSHR]](s32), [[BITCAST1]](s32)
; CHECK-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 16
; CHECK-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST1]], 16
; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; CHECK-NEXT: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64 ; CHECK-NEXT: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64
; CHECK-NEXT: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0 ; CHECK-NEXT: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; CHECK-NEXT: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>) ; CHECK-NEXT: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>)
; CHECK-NEXT: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>) ; CHECK-NEXT: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; CHECK-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32) ; CHECK-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; CHECK-NEXT: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>) ; CHECK-NEXT: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; CHECK-NEXT: [[SEXT_INREG3:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST2]], 16 ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST2]](s32), [[LSHR1]](s32), [[BITCAST3]](s32)
; CHECK-NEXT: [[SEXT_INREG4:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR1]], 16
; CHECK-NEXT: [[SEXT_INREG5:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST3]], 16
; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG3]](s32), [[SEXT_INREG4]](s32), [[SEXT_INREG5]](s32)
; CHECK-NEXT: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32 ; CHECK-NEXT: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32
; CHECK-NEXT: [[INSERT2:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0 ; CHECK-NEXT: [[INSERT2:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; CHECK-NEXT: [[UV4:%[0-9]+]]:_(<2 x s16>), [[UV5:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT2]](<4 x s16>) ; CHECK-NEXT: [[UV4:%[0-9]+]]:_(<2 x s16>), [[UV5:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT2]](<4 x s16>)
; CHECK-NEXT: [[BITCAST4:%[0-9]+]]:_(s32) = G_BITCAST [[UV4]](<2 x s16>) ; CHECK-NEXT: [[BITCAST4:%[0-9]+]]:_(s32) = G_BITCAST [[UV4]](<2 x s16>)
; CHECK-NEXT: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST4]], [[C]](s32) ; CHECK-NEXT: [[LSHR2:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST4]], [[C]](s32)
; CHECK-NEXT: [[BITCAST5:%[0-9]+]]:_(s32) = G_BITCAST [[UV5]](<2 x s16>) ; CHECK-NEXT: [[BITCAST5:%[0-9]+]]:_(s32) = G_BITCAST [[UV5]](<2 x s16>)
; CHECK-NEXT: [[SEXT_INREG6:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST4]], 16 ; CHECK-NEXT: [[BUILD_VECTOR2:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST4]](s32), [[LSHR2]](s32), [[BITCAST5]](s32)
; CHECK-NEXT: [[SEXT_INREG7:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR2]], 16
; CHECK-NEXT: [[SEXT_INREG8:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST5]], 16
; CHECK-NEXT: [[BUILD_VECTOR2:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG6]](s32), [[SEXT_INREG7]](s32), [[SEXT_INREG8]](s32)
; CHECK-NEXT: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR2]](<3 x s32>), 0 ; CHECK-NEXT: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR2]](<3 x s32>), 0
; CHECK-NEXT: [[INSERT3:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0 ; CHECK-NEXT: [[INSERT3:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; CHECK-NEXT: [[UV6:%[0-9]+]]:_(<2 x s16>), [[UV7:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT3]](<4 x s16>) ; CHECK-NEXT: [[UV6:%[0-9]+]]:_(<2 x s16>), [[UV7:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT3]](<4 x s16>)
; CHECK-NEXT: [[BITCAST6:%[0-9]+]]:_(s32) = G_BITCAST [[UV6]](<2 x s16>) ; CHECK-NEXT: [[BITCAST6:%[0-9]+]]:_(s32) = G_BITCAST [[UV6]](<2 x s16>)
; CHECK-NEXT: [[LSHR3:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST6]], [[C]](s32) ; CHECK-NEXT: [[LSHR3:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST6]], [[C]](s32)
; CHECK-NEXT: [[BITCAST7:%[0-9]+]]:_(s32) = G_BITCAST [[UV7]](<2 x s16>) ; CHECK-NEXT: [[BITCAST7:%[0-9]+]]:_(s32) = G_BITCAST [[UV7]](<2 x s16>)
; CHECK-NEXT: [[SEXT_INREG9:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST6]], 16 ; CHECK-NEXT: [[BUILD_VECTOR3:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST6]](s32), [[LSHR3]](s32), [[BITCAST7]](s32)
; CHECK-NEXT: [[SEXT_INREG10:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR3]], 16
; CHECK-NEXT: [[SEXT_INREG11:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST7]], 16
; CHECK-NEXT: [[BUILD_VECTOR3:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG9]](s32), [[SEXT_INREG10]](s32), [[SEXT_INREG11]](s32)
; CHECK-NEXT: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR3]](<3 x s32>), 0 ; CHECK-NEXT: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR3]](<3 x s32>), 0
; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; CHECK-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[EXTRACT2]], [[C1]] ; CHECK-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[EXTRACT2]], [[C1]]
; CHECK-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[EXTRACT3]], [[C1]] ; CHECK-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[EXTRACT3]], [[C1]]
; CHECK-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32) ; CHECK-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; CHECK-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; CHECK-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; CHECK-NEXT: [[BITCAST8:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; CHECK-NEXT: [[BITCAST8:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; CHECK-NEXT: [[AND2:%[0-9]+]]:_(s32) = G_AND [[EXTRACT4]], [[C1]] ; CHECK-NEXT: [[AND2:%[0-9]+]]:_(s32) = G_AND [[EXTRACT4]], [[C1]]
Show All 14 Lines

llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-shuffle-vector.s16.mir

# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=fiji -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX8 %s # RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=fiji -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX8 %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX9 %s # RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX9 %s
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1010 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX9 %s # RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1010 -run-pass=legalizer -o - %s | FileCheck -check-prefix=GFX9 %s
--- ---
name: shufflevector_v2s16_v2s16_undef_undef name: shufflevector_v2s16_v2s16_undef_undef
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, undef)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_undef ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_undef
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[C]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[C]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C]], [[SHL]]
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, undef)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, undef) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, undef)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_0_undef name: shufflevector_v2s16_v2s16_0_undef
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, undef)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_undef ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_undef
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32) ; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY2]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY2]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, undef)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, undef) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, undef)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_undef_0 name: shufflevector_v2s16_v2s16_undef_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, 0)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_0 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_0
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32) ; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY2]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY2]], [[SHL]]
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, 0)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 0) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 0)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_0_1 name: shufflevector_v2s16_v2s16_0_1
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_1
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, 1)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_1 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_1
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_1
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, 1)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 1) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 1)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_1_0 name: shufflevector_v2s16_v2s16_1_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_1_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(1, 0)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_1_0 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_1_0
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_1_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(1, 0)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_0_0 name: shufflevector_v2s16_v2s16_0_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, 0)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_0 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_0
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, 0)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 0) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 0)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_1_1 name: shufflevector_v2s16_v2s16_1_1
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_1_1
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(1, 1)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_1_1 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_1_1
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_1_1
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(1, 1)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 1) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 1)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_2_2 name: shufflevector_v2s16_v2s16_2_2
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(2, 2)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_2 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_2
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(2, 2)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 2) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 2)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_2_undef name: shufflevector_v2s16_v2s16_2_undef
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(2, undef)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_undef ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_undef
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32) ; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY2]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[COPY2]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(2, undef)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, undef) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, undef)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_undef_2 name: shufflevector_v2s16_v2s16_undef_2
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, 2)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_2 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_2
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32) ; GFX8-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY2]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[COPY2]], [[SHL]]
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, 2)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 2) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 2)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_2_3 name: shufflevector_v2s16_v2s16_2_3
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(2, 3)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_3 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_3
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(2, 3)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 3) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 3)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_3_2 name: shufflevector_v2s16_v2s16_3_2
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(3, 2)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_2 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_2
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(3, 2)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 2) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 2)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_undef_3 name: shufflevector_v2s16_v2s16_undef_3
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, 3)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_3 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_undef_3
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND]], [[C]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND]], [[C]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C1]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[C1]], [[SHL]]
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_undef_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(undef, 3)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 3) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(undef, 3)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_3_undef name: shufflevector_v2s16_v2s16_3_undef
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(3, undef)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_undef ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_undef
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[C2]], [[C]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[C2]], [[C]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST1]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_undef
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(3, undef)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, undef) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, undef)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_3_3 name: shufflevector_v2s16_v2s16_3_3
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(3, 3)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_3 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_3
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(3, 3)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 3) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 3)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_0_2 name: shufflevector_v2s16_v2s16_0_2
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_2 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_2
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 2) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 2)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_2_0 name: shufflevector_v2s16_v2s16_2_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, 0)
; GFX9: $vgpr0 = COPY [[SHUF]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_0 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_0
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C2]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[SHUF:%[0-9]+]]:_(<2 x s16>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s16>), [[COPY1]], shufflemask(0, 0)
; GFX9-NEXT: $vgpr0 = COPY [[SHUF]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 0) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 0)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_3_0 name: shufflevector_v2s16_v2s16_3_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_0 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_3_0
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_3_0
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 0) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(3, 0)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_0_3 name: shufflevector_v2s16_v2s16_0_3
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_3 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_0_3
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_0_3
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 3) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(0, 3)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_1_2 name: shufflevector_v2s16_v2s16_1_2
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_1_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_1_2 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_1_2
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_1_2
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C1]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 2) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 2)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v2s16_2_1 name: shufflevector_v2s16_v2s16_2_1
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0, $vgpr1 liveins: $vgpr0, $vgpr1
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_1
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_1 ; GFX8-LABEL: name: shufflevector_v2s16_v2s16_2_1
; GFX8: liveins: $vgpr0, $vgpr1 ; GFX8: liveins: $vgpr0, $vgpr1
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v2s16_2_1
; GFX9: liveins: $vgpr0, $vgpr1
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr0
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<2 x s16>) = COPY $vgpr1
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[COPY1]](<2 x s16>)
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[COPY]](<2 x s16>)
; GFX9-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST1]], [[C1]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<2 x s16>) = COPY $vgpr0 %0:_(<2 x s16>) = COPY $vgpr0
%1:_(<2 x s16>) = COPY $vgpr1 %1:_(<2 x s16>) = COPY $vgpr1
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 1) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(2, 1)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v4s16_v3s16_2_0 name: shufflevector_v4s16_v3s16_2_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3 liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9-LABEL: name: shufflevector_v4s16_v3s16_2_0
; GFX9: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9: [[EXTRACT:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY]](<4 x s16>), 0
; GFX9: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0
; GFX9: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
; GFX9: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; GFX9: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>)
; GFX9: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; GFX9: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST]], 16
; GFX9: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 16
; GFX9: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST1]], 16
; GFX9: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; GFX9: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64
; GFX9: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; GFX9: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>)
; GFX9: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; GFX9: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; GFX9: [[SEXT_INREG3:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST2]], 16
; GFX9: [[SEXT_INREG4:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR1]], 16
; GFX9: [[SEXT_INREG5:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST3]], 16
; GFX9: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG3]](s32), [[SEXT_INREG4]](s32), [[SEXT_INREG5]](s32)
; GFX9: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32
; GFX9: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 0
; GFX9: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 0
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[EXTRACT2]](s32), [[EXTRACT3]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC1:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[EXTRACT4]](s32), [[EXTRACT5]](s32)
; GFX9: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BUILD_VECTOR_TRUNC]](<2 x s16>), [[BUILD_VECTOR_TRUNC1]](<2 x s16>)
; GFX9: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
; GFX8-LABEL: name: shufflevector_v4s16_v3s16_2_0 ; GFX8-LABEL: name: shufflevector_v4s16_v3s16_2_0
; GFX8: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3 ; GFX8: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX8-NEXT: [[EXTRACT:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY]](<4 x s16>), 0 ; GFX8-NEXT: [[EXTRACT:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY]](<4 x s16>), 0
; GFX8-NEXT: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0 ; GFX8-NEXT: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0
; GFX8-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF ; GFX8-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
; GFX8-NEXT: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0 ; GFX8-NEXT: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; GFX8-NEXT: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>) ; GFX8-NEXT: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>)
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; GFX8-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST]], 16 ; GFX8-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST]](s32), [[LSHR]](s32), [[BITCAST1]](s32)
; GFX8-NEXT: [[SEXT_INREG1:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR]], 16
; GFX8-NEXT: [[SEXT_INREG2:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST1]], 16
; GFX8-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG]](s32), [[SEXT_INREG1]](s32), [[SEXT_INREG2]](s32)
; GFX8-NEXT: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64 ; GFX8-NEXT: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64
; GFX8-NEXT: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0 ; GFX8-NEXT: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; GFX8-NEXT: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>) ; GFX8-NEXT: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>)
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; GFX8-NEXT: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>) ; GFX8-NEXT: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; GFX8-NEXT: [[SEXT_INREG3:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST2]], 16 ; GFX8-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST2]](s32), [[LSHR1]](s32), [[BITCAST3]](s32)
; GFX8-NEXT: [[SEXT_INREG4:%[0-9]+]]:_(s32) = G_SEXT_INREG [[LSHR1]], 16
; GFX8-NEXT: [[SEXT_INREG5:%[0-9]+]]:_(s32) = G_SEXT_INREG [[BITCAST3]], 16
; GFX8-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[SEXT_INREG3]](s32), [[SEXT_INREG4]](s32), [[SEXT_INREG5]](s32)
; GFX8-NEXT: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32 ; GFX8-NEXT: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32
; GFX8-NEXT: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 0 ; GFX8-NEXT: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 0
; GFX8-NEXT: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 0 ; GFX8-NEXT: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[EXTRACT2]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[EXTRACT2]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[EXTRACT3]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[EXTRACT3]], [[C1]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST4:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST4:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: [[AND2:%[0-9]+]]:_(s32) = G_AND [[EXTRACT4]], [[C1]] ; GFX8-NEXT: [[AND2:%[0-9]+]]:_(s32) = G_AND [[EXTRACT4]], [[C1]]
; GFX8-NEXT: [[AND3:%[0-9]+]]:_(s32) = G_AND [[EXTRACT5]], [[C1]] ; GFX8-NEXT: [[AND3:%[0-9]+]]:_(s32) = G_AND [[EXTRACT5]], [[C1]]
; GFX8-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[AND3]], [[C]](s32) ; GFX8-NEXT: [[SHL1:%[0-9]+]]:_(s32) = G_SHL [[AND3]], [[C]](s32)
; GFX8-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[AND2]], [[SHL1]] ; GFX8-NEXT: [[OR1:%[0-9]+]]:_(s32) = G_OR [[AND2]], [[SHL1]]
; GFX8-NEXT: [[BITCAST5:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR1]](s32) ; GFX8-NEXT: [[BITCAST5:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR1]](s32)
; GFX8-NEXT: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BITCAST4]](<2 x s16>), [[BITCAST5]](<2 x s16>) ; GFX8-NEXT: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BITCAST4]](<2 x s16>), [[BITCAST5]](<2 x s16>)
; GFX8-NEXT: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>) ; GFX8-NEXT: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
; GFX9-LABEL: name: shufflevector_v4s16_v3s16_2_0
; GFX9: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9-NEXT: [[EXTRACT:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY]](<4 x s16>), 0
; GFX9-NEXT: [[EXTRACT1:%[0-9]+]]:_(<3 x s16>) = G_EXTRACT [[COPY1]](<4 x s16>), 0
; GFX9-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
; GFX9-NEXT: [[INSERT:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT1]](<3 x s16>), 0
; GFX9-NEXT: [[UV:%[0-9]+]]:_(<2 x s16>), [[UV1:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT]](<4 x s16>)
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(s32) = G_BITCAST [[UV]](<2 x s16>)
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST]], [[C]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(s32) = G_BITCAST [[UV1]](<2 x s16>)
; GFX9-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST]](s32), [[LSHR]](s32), [[BITCAST1]](s32)
; GFX9-NEXT: [[EXTRACT2:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 64
; GFX9-NEXT: [[INSERT1:%[0-9]+]]:_(<4 x s16>) = G_INSERT [[DEF]], [[EXTRACT]](<3 x s16>), 0
; GFX9-NEXT: [[UV2:%[0-9]+]]:_(<2 x s16>), [[UV3:%[0-9]+]]:_(<2 x s16>) = G_UNMERGE_VALUES [[INSERT1]](<4 x s16>)
; GFX9-NEXT: [[BITCAST2:%[0-9]+]]:_(s32) = G_BITCAST [[UV2]](<2 x s16>)
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[BITCAST2]], [[C]](s32)
; GFX9-NEXT: [[BITCAST3:%[0-9]+]]:_(s32) = G_BITCAST [[UV3]](<2 x s16>)
; GFX9-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<3 x s32>) = G_BUILD_VECTOR [[BITCAST2]](s32), [[LSHR1]](s32), [[BITCAST3]](s32)
; GFX9-NEXT: [[EXTRACT3:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 32
; GFX9-NEXT: [[EXTRACT4:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR]](<3 x s32>), 0
; GFX9-NEXT: [[EXTRACT5:%[0-9]+]]:_(s32) = G_EXTRACT [[BUILD_VECTOR1]](<3 x s32>), 0
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[EXTRACT2]](s32), [[EXTRACT3]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC1:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[EXTRACT4]](s32), [[EXTRACT5]](s32)
; GFX9-NEXT: [[CONCAT_VECTORS:%[0-9]+]]:_(<4 x s16>) = G_CONCAT_VECTORS [[BUILD_VECTOR_TRUNC]](<2 x s16>), [[BUILD_VECTOR_TRUNC1]](<2 x s16>)
; GFX9-NEXT: $vgpr0_vgpr1 = COPY [[CONCAT_VECTORS]](<4 x s16>)
%0:_(<4 x s16>) = COPY $vgpr0_vgpr1 %0:_(<4 x s16>) = COPY $vgpr0_vgpr1
%1:_(<4 x s16>) = COPY $vgpr2_vgpr3 %1:_(<4 x s16>) = COPY $vgpr2_vgpr3
%2:_(<3 x s16>) = G_EXTRACT %0, 0 %2:_(<3 x s16>) = G_EXTRACT %0, 0
%3:_(<3 x s16>) = G_EXTRACT %1, 0 %3:_(<3 x s16>) = G_EXTRACT %1, 0
%4:_(<4 x s16>) = G_SHUFFLE_VECTOR %2, %3, shufflemask(5, 1, 3, 0) %4:_(<4 x s16>) = G_SHUFFLE_VECTOR %2, %3, shufflemask(5, 1, 3, 0)
$vgpr0_vgpr1 = COPY %4 $vgpr0_vgpr1 = COPY %4
... ...
--- ---
name: shufflevector_v2s16_v4s16_1_0 name: shufflevector_v2s16_v4s16_1_0
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3 liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9-LABEL: name: shufflevector_v2s16_v4s16_1_0
; GFX9: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0
; GFX9: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C1]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 0
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v4s16_1_0 ; GFX8-LABEL: name: shufflevector_v2s16_v4s16_1_0
; GFX8: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3 ; GFX8: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX8-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0 ; GFX8-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C1]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C1]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX8-NEXT: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 0 ; GFX8-NEXT: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 0
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32)
; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C2:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C2]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C2]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C1]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v4s16_1_0
; GFX9: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
; GFX9-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0
; GFX9-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C1]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9-NEXT: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 0
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<4 x s16>) = COPY $vgpr0_vgpr1 %0:_(<4 x s16>) = COPY $vgpr0_vgpr1
%1:_(<4 x s16>) = COPY $vgpr2_vgpr3 %1:_(<4 x s16>) = COPY $vgpr2_vgpr3
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 0)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...
--- ---
name: shufflevector_v2s16_v4s16_1_3 name: shufflevector_v2s16_v4s16_1_3
tracksRegLiveness: true tracksRegLiveness: true
body: | body: |
bb.0: bb.0:
liveins: $vgpr0_vgpr1, $vgpr2_vgpr3 liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9-LABEL: name: shufflevector_v2s16_v4s16_1_3
; GFX9: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0
; GFX9: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C]](s32)
; GFX9: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 32
; GFX9: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32)
; GFX9: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
; GFX8-LABEL: name: shufflevector_v2s16_v4s16_1_3 ; GFX8-LABEL: name: shufflevector_v2s16_v4s16_1_3
; GFX8: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3 ; GFX8: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX8-NEXT: {{ $}} ; GFX8-NEXT: {{ $}}
; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1 ; GFX8-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3 ; GFX8-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>) ; GFX8-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX8-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0 ; GFX8-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0
; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16 ; GFX8-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C]](s32) ; GFX8-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C]](s32)
; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>) ; GFX8-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX8-NEXT: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 32 ; GFX8-NEXT: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 32
; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32) ; GFX8-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32)
; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535 ; GFX8-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 65535
; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]] ; GFX8-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[LSHR]], [[C1]]
; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]] ; GFX8-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[LSHR1]], [[C1]]
; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32) ; GFX8-NEXT: [[SHL:%[0-9]+]]:_(s32) = G_SHL [[AND1]], [[C]](s32)
; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]] ; GFX8-NEXT: [[OR:%[0-9]+]]:_(s32) = G_OR [[AND]], [[SHL]]
; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32) ; GFX8-NEXT: [[BITCAST2:%[0-9]+]]:_(<2 x s16>) = G_BITCAST [[OR]](s32)
; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>) ; GFX8-NEXT: $vgpr0 = COPY [[BITCAST2]](<2 x s16>)
; GFX9-LABEL: name: shufflevector_v2s16_v4s16_1_3
; GFX9: liveins: $vgpr0_vgpr1, $vgpr2_vgpr3
; GFX9-NEXT: {{ $}}
; GFX9-NEXT: [[COPY:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr0_vgpr1
; GFX9-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s16>) = COPY $vgpr2_vgpr3
; GFX9-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9-NEXT: [[EXTRACT:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST]](<2 x s32>), 0
; GFX9-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 16
; GFX9-NEXT: [[LSHR:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT]], [[C]](s32)
; GFX9-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s32>) = G_BITCAST [[COPY]](<4 x s16>)
; GFX9-NEXT: [[EXTRACT1:%[0-9]+]]:_(s32) = G_EXTRACT [[BITCAST1]](<2 x s32>), 32
; GFX9-NEXT: [[LSHR1:%[0-9]+]]:_(s32) = G_LSHR [[EXTRACT1]], [[C]](s32)
; GFX9-NEXT: [[BUILD_VECTOR_TRUNC:%[0-9]+]]:_(<2 x s16>) = G_BUILD_VECTOR_TRUNC [[LSHR]](s32), [[LSHR1]](s32)
; GFX9-NEXT: $vgpr0 = COPY [[BUILD_VECTOR_TRUNC]](<2 x s16>)
%0:_(<4 x s16>) = COPY $vgpr0_vgpr1 %0:_(<4 x s16>) = COPY $vgpr0_vgpr1
%1:_(<4 x s16>) = COPY $vgpr2_vgpr3 %1:_(<4 x s16>) = COPY $vgpr2_vgpr3
%2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 3) %2:_(<2 x s16>) = G_SHUFFLE_VECTOR %0, %1, shufflemask(1, 3)
$vgpr0 = COPY %2 $vgpr0 = COPY %2
... ...