forked from OSchip/llvm-project
228 lines
7.7 KiB
LLVM
228 lines
7.7 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=thumbv8.1m.main-none-eabi -mattr=+mve --verify-machineinstrs %s -o - | FileCheck %s
|
|
|
|
define void @vctp8(i32 %arg, <16 x i8> *%in, <16 x i8>* %out) {
|
|
; CHECK-LABEL: vctp8:
|
|
; CHECK: @ %bb.0:
|
|
; CHECK-NEXT: vldrw.u32 q1, [r1]
|
|
; CHECK-NEXT: vctp.8 r0
|
|
; CHECK-NEXT: vmov.i32 q0, #0x0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q0, q1
|
|
; CHECK-NEXT: vstrw.32 q0, [r2]
|
|
; CHECK-NEXT: bx lr
|
|
%pred = call <16 x i1> @llvm.arm.mve.vctp8(i32 %arg)
|
|
%ld = load <16 x i8>, <16 x i8>* %in
|
|
%res = select <16 x i1> %pred, <16 x i8> %ld, <16 x i8> zeroinitializer
|
|
store <16 x i8> %res, <16 x i8>* %out
|
|
ret void
|
|
}
|
|
|
|
define void @vctp16(i32 %arg, <8 x i16> *%in, <8 x i16>* %out) {
|
|
; CHECK-LABEL: vctp16:
|
|
; CHECK: @ %bb.0:
|
|
; CHECK-NEXT: vldrw.u32 q1, [r1]
|
|
; CHECK-NEXT: vctp.16 r0
|
|
; CHECK-NEXT: vmov.i32 q0, #0x0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q0, q1
|
|
; CHECK-NEXT: vstrw.32 q0, [r2]
|
|
; CHECK-NEXT: bx lr
|
|
%pred = call <8 x i1> @llvm.arm.mve.vctp16(i32 %arg)
|
|
%ld = load <8 x i16>, <8 x i16>* %in
|
|
%res = select <8 x i1> %pred, <8 x i16> %ld, <8 x i16> zeroinitializer
|
|
store <8 x i16> %res, <8 x i16>* %out
|
|
ret void
|
|
}
|
|
|
|
define void @vctp32(i32 %arg, <4 x i32> *%in, <4 x i32>* %out) {
|
|
; CHECK-LABEL: vctp32:
|
|
; CHECK: @ %bb.0:
|
|
; CHECK-NEXT: vldrw.u32 q1, [r1]
|
|
; CHECK-NEXT: vctp.32 r0
|
|
; CHECK-NEXT: vmov.i32 q0, #0x0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q0, q1
|
|
; CHECK-NEXT: vstrw.32 q0, [r2]
|
|
; CHECK-NEXT: bx lr
|
|
%pred = call <4 x i1> @llvm.arm.mve.vctp32(i32 %arg)
|
|
%ld = load <4 x i32>, <4 x i32>* %in
|
|
%res = select <4 x i1> %pred, <4 x i32> %ld, <4 x i32> zeroinitializer
|
|
store <4 x i32> %res, <4 x i32>* %out
|
|
ret void
|
|
}
|
|
|
|
define void @vctp64(i32 %arg, <2 x i64> *%in, <2 x i64>* %out) {
|
|
; CHECK-LABEL: vctp64:
|
|
; CHECK: @ %bb.0:
|
|
; CHECK-NEXT: vldrw.u32 q1, [r1]
|
|
; CHECK-NEXT: vctp.64 r0
|
|
; CHECK-NEXT: vmov.i32 q0, #0x0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q0, q1
|
|
; CHECK-NEXT: vstrw.32 q0, [r2]
|
|
; CHECK-NEXT: bx lr
|
|
%pred = call <2 x i1> @llvm.arm.mve.vctp64(i32 %arg)
|
|
%ld = load <2 x i64>, <2 x i64>* %in
|
|
%res = select <2 x i1> %pred, <2 x i64> %ld, <2 x i64> zeroinitializer
|
|
store <2 x i64> %res, <2 x i64>* %out
|
|
ret void
|
|
}
|
|
|
|
define arm_aapcs_vfpcc <4 x i32> @vcmp_ult_v4i32(i32 %n, <4 x i32> %a, <4 x i32> %b) {
|
|
; CHECK-LABEL: vcmp_ult_v4i32:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: vctp.32 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <4 x i32> undef, i32 %n, i32 0
|
|
%ns = shufflevector <4 x i32> %i, <4 x i32> undef, <4 x i32> zeroinitializer
|
|
%c = icmp ult <4 x i32> <i32 0, i32 1, i32 2, i32 3>, %ns
|
|
%s = select <4 x i1> %c, <4 x i32> %a, <4 x i32> %b
|
|
ret <4 x i32> %s
|
|
}
|
|
|
|
define arm_aapcs_vfpcc <4 x i32> @vcmp_uge_v4i32(i32 %n, <4 x i32> %a, <4 x i32> %b) {
|
|
; CHECK-LABEL: vcmp_uge_v4i32:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: vctp.32 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <4 x i32> undef, i32 %n, i32 0
|
|
%ns = shufflevector <4 x i32> %i, <4 x i32> undef, <4 x i32> zeroinitializer
|
|
%c = icmp uge <4 x i32> %ns, <i32 0, i32 1, i32 2, i32 3>
|
|
%s = select <4 x i1> %c, <4 x i32> %a, <4 x i32> %b
|
|
ret <4 x i32> %s
|
|
}
|
|
|
|
define arm_aapcs_vfpcc <4 x i32> @vcmp_ult_v4i32_undef(i32 %n, <4 x i32> %a, <4 x i32> %b) {
|
|
; CHECK-LABEL: vcmp_ult_v4i32_undef:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: vctp.32 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <4 x i32> undef, i32 %n, i32 0
|
|
%ns = shufflevector <4 x i32> %i, <4 x i32> undef, <4 x i32> zeroinitializer
|
|
%c = icmp ult <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>, %ns
|
|
%s = select <4 x i1> %c, <4 x i32> %a, <4 x i32> %b
|
|
ret <4 x i32> %s
|
|
}
|
|
|
|
|
|
define arm_aapcs_vfpcc <8 x i16> @vcmp_ult_v8i16(i16 %n, <8 x i16> %a, <8 x i16> %b) {
|
|
; CHECK-LABEL: vcmp_ult_v8i16:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: uxth r0, r0
|
|
; CHECK-NEXT: vctp.16 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <8 x i16> undef, i16 %n, i32 0
|
|
%ns = shufflevector <8 x i16> %i, <8 x i16> undef, <8 x i32> zeroinitializer
|
|
%c = icmp ult <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, %ns
|
|
%s = select <8 x i1> %c, <8 x i16> %a, <8 x i16> %b
|
|
ret <8 x i16> %s
|
|
}
|
|
|
|
define arm_aapcs_vfpcc <8 x i16> @vcmp_uge_v8i16(i16 %n, <8 x i16> %a, <8 x i16> %b) {
|
|
; CHECK-LABEL: vcmp_uge_v8i16:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: uxth r0, r0
|
|
; CHECK-NEXT: vctp.16 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <8 x i16> undef, i16 %n, i32 0
|
|
%ns = shufflevector <8 x i16> %i, <8 x i16> undef, <8 x i32> zeroinitializer
|
|
%c = icmp uge <8 x i16> %ns, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>
|
|
%s = select <8 x i1> %c, <8 x i16> %a, <8 x i16> %b
|
|
ret <8 x i16> %s
|
|
}
|
|
|
|
|
|
define arm_aapcs_vfpcc <16 x i8> @vcmp_ult_v16i8(i8 %n, <16 x i8> %a, <16 x i8> %b) {
|
|
; CHECK-LABEL: vcmp_ult_v16i8:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: uxtb r0, r0
|
|
; CHECK-NEXT: vctp.8 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <16 x i8> undef, i8 %n, i32 0
|
|
%ns = shufflevector <16 x i8> %i, <16 x i8> undef, <16 x i32> zeroinitializer
|
|
%c = icmp ult <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, %ns
|
|
%s = select <16 x i1> %c, <16 x i8> %a, <16 x i8> %b
|
|
ret <16 x i8> %s
|
|
}
|
|
|
|
define arm_aapcs_vfpcc <16 x i8> @vcmp_uge_v16i8(i8 %n, <16 x i8> %a, <16 x i8> %b) {
|
|
; CHECK-LABEL: vcmp_uge_v16i8:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: uxtb r0, r0
|
|
; CHECK-NEXT: vctp.8 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <16 x i8> undef, i8 %n, i32 0
|
|
%ns = shufflevector <16 x i8> %i, <16 x i8> undef, <16 x i32> zeroinitializer
|
|
%c = icmp uge <16 x i8> %ns, <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>
|
|
%s = select <16 x i1> %c, <16 x i8> %a, <16 x i8> %b
|
|
ret <16 x i8> %s
|
|
}
|
|
|
|
|
|
define arm_aapcs_vfpcc <2 x i64> @vcmp_ult_v2i64(i64 %n, <2 x i64> %a, <2 x i64> %b) {
|
|
; CHECK-LABEL: vcmp_ult_v2i64:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: vctp.64 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <2 x i64> undef, i64 %n, i32 0
|
|
%ns = shufflevector <2 x i64> %i, <2 x i64> undef, <2 x i32> zeroinitializer
|
|
%c = icmp ult <2 x i64> <i64 0, i64 1>, %ns
|
|
%s = select <2 x i1> %c, <2 x i64> %a, <2 x i64> %b
|
|
ret <2 x i64> %s
|
|
}
|
|
|
|
define arm_aapcs_vfpcc <2 x i64> @vcmp_uge_v2i64(i64 %n, <2 x i64> %a, <2 x i64> %b) {
|
|
; CHECK-LABEL: vcmp_uge_v2i64:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: vctp.64 r0
|
|
; CHECK-NEXT: vpst
|
|
; CHECK-NEXT: vmovt q1, q0
|
|
; CHECK-NEXT: vmov q0, q1
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%i = insertelement <2 x i64> undef, i64 %n, i32 0
|
|
%ns = shufflevector <2 x i64> %i, <2 x i64> undef, <2 x i32> zeroinitializer
|
|
%c = icmp uge <2 x i64> %ns, <i64 0, i64 1>
|
|
%s = select <2 x i1> %c, <2 x i64> %a, <2 x i64> %b
|
|
ret <2 x i64> %s
|
|
}
|
|
|
|
|
|
declare <16 x i1> @llvm.arm.mve.vctp8(i32)
|
|
declare <8 x i1> @llvm.arm.mve.vctp16(i32)
|
|
declare <4 x i1> @llvm.arm.mve.vctp32(i32)
|
|
declare <2 x i1> @llvm.arm.mve.vctp64(i32)
|