blob: c66d334d16b0be6cbf74d6644a4a1bd4f6437e55 [file] [log] [blame]
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# WARNING: update_mir_test_checks.py does not include the constant pools output,
# so this test requires manual fixing up after running the script.
# RUN: llc -mtriple=aarch64-- -O0 -run-pass=instruction-select -verify-machineinstrs %s -global-isel-abort=1 -o - | FileCheck %s
--- |
target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
target triple = "aarch64"
define <2 x float> @shuffle_v2f32(<2 x float> %a, <2 x float> %b) {
%shuf = shufflevector <2 x float> %a, <2 x float> %b, <2 x i32> <i32 1, i32 0>
ret <2 x float> %shuf
}
define <4 x i32> @shuffle_v4i32(<4 x i32> %a, <4 x i32> %b) {
%shuf = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 0, i32 1, i32 3, i32 0>
ret <4 x i32> %shuf
}
define <4 x i32> @shuffle_tbl_v4i32(<4 x i32> %a, <4 x i32> %b) {
%shuf = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 5, i32 7, i32 1, i32 0>
ret <4 x i32> %shuf
}
define <2 x i64> @shuffle_v2i64(<2 x i64> %a, <2 x i64> %b) {
%shuf = shufflevector <2 x i64> %a, <2 x i64> %b, <2 x i32> zeroinitializer
ret <2 x i64> %shuf
}
...
---
name: shuffle_v2f32
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1 (%ir-block.0):
liveins: $d0, $d1
; CHECK-LABEL: name: shuffle_v2f32
; CHECK: constants:
; CHECK: - id: 0
; CHECK: value: '<8 x i8> <i8 4, i8 5, i8 6, i8 7, i8 0, i8 1, i8 2, i8 3>'
; CHECK: alignment: 8
; CHECK: liveins: $d0, $d1
; CHECK: [[COPY:%[0-9]+]]:fpr64 = COPY $d0
; CHECK: [[COPY1:%[0-9]+]]:fpr64 = COPY $d1
; CHECK: [[ADRP:%[0-9]+]]:gpr64common = ADRP target-flags(aarch64-page) %const.0
; CHECK: [[LDRDui:%[0-9]+]]:fpr64 = LDRDui [[ADRP]], target-flags(aarch64-pageoff, aarch64-nc) %const.0
; CHECK: [[DEF:%[0-9]+]]:fpr128 = IMPLICIT_DEF
; CHECK: [[INSERT_SUBREG:%[0-9]+]]:fpr128 = INSERT_SUBREG [[DEF]], [[COPY]], %subreg.dsub
; CHECK: [[DEF1:%[0-9]+]]:fpr128 = IMPLICIT_DEF
; CHECK: [[INSERT_SUBREG1:%[0-9]+]]:fpr128 = INSERT_SUBREG [[DEF1]], [[COPY1]], %subreg.dsub
; CHECK: [[INSvi64lane:%[0-9]+]]:fpr128 = INSvi64lane [[INSERT_SUBREG]], 1, [[INSERT_SUBREG1]], 0
; CHECK: [[DEF2:%[0-9]+]]:fpr128 = IMPLICIT_DEF
; CHECK: [[INSERT_SUBREG2:%[0-9]+]]:fpr128 = INSERT_SUBREG [[DEF2]], [[LDRDui]], %subreg.dsub
; CHECK: [[TBLv16i8One:%[0-9]+]]:fpr128 = TBLv16i8One [[INSvi64lane]], [[INSERT_SUBREG2]]
; CHECK: [[COPY2:%[0-9]+]]:fpr64 = COPY [[TBLv16i8One]].dsub
; CHECK: $d0 = COPY [[COPY2]]
; CHECK: RET_ReallyLR implicit $d0
%0:fpr(<2 x s32>) = COPY $d0
%1:fpr(<2 x s32>) = COPY $d1
%2:fpr(<2 x s32>) = G_SHUFFLE_VECTOR %0(<2 x s32>), %1, shufflemask(1, 0)
$d0 = COPY %2(<2 x s32>)
RET_ReallyLR implicit $d0
...
---
name: shuffle_v4i32
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1 (%ir-block.0):
liveins: $q0, $q1
; CHECK-LABEL: name: shuffle_v4i32
; CHECK: constants:
; CHECK: value: '<16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 12, i8 13, i8 14, i8 15, i8 0, i8 1, i8 2, i8 3>'
; CHECK: alignment: 16
; CHECK: isTargetSpecific: false
; CHECK: liveins: $q0, $q1
; CHECK: [[COPY:%[0-9]+]]:fpr128 = COPY $q0
; CHECK: [[COPY1:%[0-9]+]]:fpr128 = COPY $q1
; CHECK: [[ADRP:%[0-9]+]]:gpr64common = ADRP target-flags(aarch64-page) %const.0
; CHECK: [[LDRQui:%[0-9]+]]:fpr128 = LDRQui [[ADRP]], target-flags(aarch64-pageoff, aarch64-nc) %const.0
; CHECK: [[REG_SEQUENCE:%[0-9]+]]:qq = REG_SEQUENCE [[COPY]], %subreg.qsub0, [[COPY1]], %subreg.qsub1
; CHECK: [[TBLv16i8Two:%[0-9]+]]:fpr128 = TBLv16i8Two [[REG_SEQUENCE]], [[LDRQui]]
; CHECK: $q0 = COPY [[TBLv16i8Two]]
; CHECK: RET_ReallyLR implicit $q0
%0:fpr(<4 x s32>) = COPY $q0
%1:fpr(<4 x s32>) = COPY $q1
%2:fpr(<4 x s32>) = G_SHUFFLE_VECTOR %0(<4 x s32>), %1, shufflemask(0, 1, 3, 0)
$q0 = COPY %2(<4 x s32>)
RET_ReallyLR implicit $q0
...
---
name: shuffle_tbl_v4i32
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1 (%ir-block.0):
liveins: $q0, $q1
; CHECK-LABEL: name: shuffle_tbl_v4i32
; CHECK: constants:
; CHECK: value: '<16 x i8> <i8 20, i8 21, i8 22, i8 23, i8 28, i8 29, i8 30, i8 31, i8 4, i8 5, i8 6, i8 7, i8 0, i8 1, i8 2, i8 3>'
; CHECK: alignment: 16
; CHECK: isTargetSpecific: false
; CHECK: liveins: $q0, $q1
; CHECK: [[COPY:%[0-9]+]]:fpr128 = COPY $q0
; CHECK: [[COPY1:%[0-9]+]]:fpr128 = COPY $q1
; CHECK: [[ADRP:%[0-9]+]]:gpr64common = ADRP target-flags(aarch64-page) %const.0
; CHECK: [[LDRQui:%[0-9]+]]:fpr128 = LDRQui [[ADRP]], target-flags(aarch64-pageoff, aarch64-nc) %const.0
; CHECK: [[REG_SEQUENCE:%[0-9]+]]:qq = REG_SEQUENCE [[COPY]], %subreg.qsub0, [[COPY1]], %subreg.qsub1
; CHECK: [[TBLv16i8Two:%[0-9]+]]:fpr128 = TBLv16i8Two [[REG_SEQUENCE]], [[LDRQui]]
; CHECK: $q0 = COPY [[TBLv16i8Two]]
; CHECK: RET_ReallyLR implicit $q0
%0:fpr(<4 x s32>) = COPY $q0
%1:fpr(<4 x s32>) = COPY $q1
%2:fpr(<4 x s32>) = G_SHUFFLE_VECTOR %0(<4 x s32>), %1, shufflemask(5, 7, 1, 0)
$q0 = COPY %2(<4 x s32>)
RET_ReallyLR implicit $q0
...
---
name: shuffle_v2i64
alignment: 4
legalized: true
regBankSelected: true
tracksRegLiveness: true
body: |
bb.1 (%ir-block.0):
liveins: $q0, $q1
; CHECK-LABEL: name: shuffle_v2i64
; CHECK: constants:
; CHECK: value: '<16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7>'
; CHECK: alignment: 16
; CHECK: isTargetSpecific: false
; CHECK: liveins: $q0, $q1
; CHECK: [[COPY:%[0-9]+]]:fpr128 = COPY $q0
; CHECK: [[COPY1:%[0-9]+]]:fpr128 = COPY $q1
; CHECK: [[ADRP:%[0-9]+]]:gpr64common = ADRP target-flags(aarch64-page) %const.0
; CHECK: [[LDRQui:%[0-9]+]]:fpr128 = LDRQui [[ADRP]], target-flags(aarch64-pageoff, aarch64-nc) %const.0
; CHECK: [[REG_SEQUENCE:%[0-9]+]]:qq = REG_SEQUENCE [[COPY]], %subreg.qsub0, [[COPY1]], %subreg.qsub1
; CHECK: [[TBLv16i8Two:%[0-9]+]]:fpr128 = TBLv16i8Two [[REG_SEQUENCE]], [[LDRQui]]
; CHECK: $q0 = COPY [[TBLv16i8Two]]
; CHECK: RET_ReallyLR implicit $q0
%0:fpr(<2 x s64>) = COPY $q0
%1:fpr(<2 x s64>) = COPY $q1
%2:fpr(<2 x s64>) = G_SHUFFLE_VECTOR %0(<2 x s64>), %1, shufflemask(0, 0)
$q0 = COPY %2(<2 x s64>)
RET_ReallyLR implicit $q0
...