[clang] 30faf19 - [SME2] Add LUTI2 and LUTI4 double Builtins and Intrinsics (#73305)
via cfe-commits
cfe-commits at lists.llvm.org
Wed Dec 6 06:35:16 PST 2023
Author: Matthew Devereau
Date: 2023-12-06T14:35:11Z
New Revision: 30faf19a88c56d064bc372fa325345d35d82902c
URL: https://github.com/llvm/llvm-project/commit/30faf19a88c56d064bc372fa325345d35d82902c
DIFF: https://github.com/llvm/llvm-project/commit/30faf19a88c56d064bc372fa325345d35d82902c.diff
LOG: [SME2] Add LUTI2 and LUTI4 double Builtins and Intrinsics (#73305)
See https://github.com/ARM-software/acle/pull/217
Patch by: Hassnaa Hamdi <hassnaa.hamdi at arm.com>
Added:
clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c
clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti4_lane_zt_x2.c
llvm/test/CodeGen/AArch64/sme2-intrinsics-luti2-lane-x2.ll
llvm/test/CodeGen/AArch64/sme2-intrinsics-luti4-lane-x2.ll
Modified:
clang/include/clang/Basic/arm_sme.td
clang/test/Sema/aarch64-sme2-intrinsics/acle_sme2_imm.cpp
llvm/include/llvm/IR/IntrinsicsAArch64.td
llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
Removed:
################################################################################
diff --git a/clang/include/clang/Basic/arm_sme.td b/clang/include/clang/Basic/arm_sme.td
index 08d484acb3c10..1520e8ad8cebd 100644
--- a/clang/include/clang/Basic/arm_sme.td
+++ b/clang/include/clang/Basic/arm_sme.td
@@ -330,9 +330,18 @@ let TargetGuard = "sme2" in {
def SVZERO_ZT : Inst<"svzero_zt", "vi", "", MergeNone, "aarch64_sme_zero_zt", [IsOverloadNone, IsStreamingCompatible, IsSharedZA], [ImmCheck<0, ImmCheck0_0>]>;
}
+//
// lookup table expand four contiguous registers
//
let TargetGuard = "sme2" in {
def SVLUTI2_LANE_ZT_X4 : Inst<"svluti2_lane_zt_{d}_x4", "4.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt_x4", [IsStreaming, IsSharedZA, IsPreservesZA], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_3>]>;
def SVLUTI4_LANE_ZT_X4 : Inst<"svluti4_lane_zt_{d}_x4", "4.di[i", "sUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt_x4", [IsStreaming, IsSharedZA, IsPreservesZA], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_1>]>;
}
+
+//
+// lookup table expand two contiguous registers
+//
+let TargetGuard = "sme2" in {
+ def SVLUTI2_LANE_ZT_X2 : Inst<"svluti2_lane_zt_{d}_x2", "2.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti2_lane_zt_x2", [IsStreaming, IsSharedZA, IsPreservesZA], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_7>]>;
+ def SVLUTI4_LANE_ZT_X2 : Inst<"svluti4_lane_zt_{d}_x2", "2.di[i", "cUcsUsiUibhf", MergeNone, "aarch64_sme_luti4_lane_zt_x2", [IsStreaming, IsSharedZA, IsPreservesZA], [ImmCheck<0, ImmCheck0_0>, ImmCheck<2, ImmCheck0_3>]>;
+}
diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c
new file mode 100644
index 0000000000000..04f37af46767a
--- /dev/null
+++ b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti2_lane_zt_x2.c
@@ -0,0 +1,209 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+
+// REQUIRES: aarch64-registered-target
+
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
+
+#include <arm_sme_draft_spec_subject_to_change.h>
+
+// CHECK-LABEL: @test_svluti2_lane_zt_u8(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z23test_svluti2_lane_zt_u8u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CPP-CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+svuint8x2_t test_svluti2_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_u8_x2(0, zn, 7);
+}
+
+
+// CHECK-LABEL: @test_svluti2_lane_zt_s8(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z23test_svluti2_lane_zt_s8u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CPP-CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+svint8x2_t test_svluti2_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_s8_x2(0, zn, 7);
+}
+
+// CHECK-LABEL: @test_svluti2_lane_zt_u16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti2_lane_zt_u16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+svuint16x2_t test_svluti2_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_u16_x2(0, zn, 7);
+}
+
+
+// CHECK-LABEL: @test_svluti2_lane_zt_s16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti2_lane_zt_s16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+svint16x2_t test_svluti2_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_s16_x2(0, zn, 7);
+}
+
+// CHECK-LABEL: @test_svluti2_lane_zt_f16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8f16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> poison, <vscale x 8 x half> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> [[TMP2]], <vscale x 8 x half> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x half> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti2_lane_zt_f16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8f16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> poison, <vscale x 8 x half> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> [[TMP2]], <vscale x 8 x half> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x half> [[TMP4]]
+//
+svfloat16x2_t test_svluti2_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_f16_x2(0, zn, 7);
+}
+
+// CHECK-LABEL: @test_svluti2_lane_zt_bf16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8bf16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x bfloat> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z25test_svluti2_lane_zt_bf16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8bf16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x bfloat> [[TMP4]]
+//
+svbfloat16x2_t test_svluti2_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_bf16_x2(0, zn, 7);
+}
+
+// CHECK-LABEL: @test_svluti2_lane_zt_u32(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti2_lane_zt_u32u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+svuint32x2_t test_svluti2_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_u32_x2(0, zn, 7);
+}
+
+// CHECK-LABEL: @test_svluti2_lane_zt_s32(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti2_lane_zt_s32u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+svint32x2_t test_svluti2_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_s32_x2(0, zn, 7);
+}
+
+// CHECK-LABEL: @test_svluti2_lane_zt_f32(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4f32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> poison, <vscale x 4 x float> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> [[TMP2]], <vscale x 4 x float> [[TMP3]], i64 4)
+// CHECK-NEXT: ret <vscale x 8 x float> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti2_lane_zt_f32u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4f32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 7)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> poison, <vscale x 4 x float> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> [[TMP2]], <vscale x 4 x float> [[TMP3]], i64 4)
+// CPP-CHECK-NEXT: ret <vscale x 8 x float> [[TMP4]]
+//
+svfloat32x2_t test_svluti2_lane_zt_f32(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti2_lane_zt_f32_x2(0, zn, 7);
+}
diff --git a/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti4_lane_zt_x2.c b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti4_lane_zt_x2.c
new file mode 100644
index 0000000000000..4c181dd9123c5
--- /dev/null
+++ b/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_luti4_lane_zt_x2.c
@@ -0,0 +1,209 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+
+// REQUIRES: aarch64-registered-target
+
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
+// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
+
+#include <arm_sme_draft_spec_subject_to_change.h>
+
+// CHECK-LABEL: @test_svluti4_lane_zt_u8(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z23test_svluti4_lane_zt_u8u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CPP-CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+svuint8x2_t test_svluti4_lane_zt_u8(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_u8_x2(0, zn, 3);
+}
+
+
+// CHECK-LABEL: @test_svluti4_lane_zt_s8(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z23test_svluti4_lane_zt_s8u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> [[TMP2]], <vscale x 16 x i8> [[TMP3]], i64 16)
+// CPP-CHECK-NEXT: ret <vscale x 32 x i8> [[TMP4]]
+//
+svint8x2_t test_svluti4_lane_zt_s8(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_s8_x2(0, zn, 3);
+}
+
+// CHECK-LABEL: @test_svluti4_lane_zt_u16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti4_lane_zt_u16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+svuint16x2_t test_svluti4_lane_zt_u16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_u16_x2(0, zn, 3);
+}
+
+
+// CHECK-LABEL: @test_svluti4_lane_zt_s16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti4_lane_zt_s16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> [[TMP2]], <vscale x 8 x i16> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x i16> [[TMP4]]
+//
+svint16x2_t test_svluti4_lane_zt_s16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_s16_x2(0, zn, 3);
+}
+
+// CHECK-LABEL: @test_svluti4_lane_zt_f16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8f16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> poison, <vscale x 8 x half> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> [[TMP2]], <vscale x 8 x half> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x half> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti4_lane_zt_f16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8f16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> poison, <vscale x 8 x half> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x half> @llvm.vector.insert.nxv16f16.nxv8f16(<vscale x 16 x half> [[TMP2]], <vscale x 8 x half> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x half> [[TMP4]]
+//
+svfloat16x2_t test_svluti4_lane_zt_f16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_f16_x2(0, zn, 3);
+}
+
+// CHECK-LABEL: @test_svluti4_lane_zt_bf16(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8bf16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], i64 8)
+// CHECK-NEXT: ret <vscale x 16 x bfloat> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z25test_svluti4_lane_zt_bf16u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8bf16(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], i64 8)
+// CPP-CHECK-NEXT: ret <vscale x 16 x bfloat> [[TMP4]]
+//
+svbfloat16x2_t test_svluti4_lane_zt_bf16(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_bf16_x2(0, zn, 3);
+}
+
+// CHECK-LABEL: @test_svluti4_lane_zt_u32(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti4_lane_zt_u32u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+svuint32x2_t test_svluti4_lane_zt_u32(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_u32_x2(0, zn, 3);
+}
+
+// CHECK-LABEL: @test_svluti4_lane_zt_s32(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti4_lane_zt_s32u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> [[TMP2]], <vscale x 4 x i32> [[TMP3]], i64 4)
+// CPP-CHECK-NEXT: ret <vscale x 8 x i32> [[TMP4]]
+//
+svint32x2_t test_svluti4_lane_zt_s32(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_s32_x2(0, zn, 3);
+}
+
+// CHECK-LABEL: @test_svluti4_lane_zt_f32(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4f32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 0
+// CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> poison, <vscale x 4 x float> [[TMP1]], i64 0)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 1
+// CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> [[TMP2]], <vscale x 4 x float> [[TMP3]], i64 4)
+// CHECK-NEXT: ret <vscale x 8 x float> [[TMP4]]
+//
+// CPP-CHECK-LABEL: @_Z24test_svluti4_lane_zt_f32u11__SVUint8_t(
+// CPP-CHECK-NEXT: entry:
+// CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4f32(i32 0, <vscale x 16 x i8> [[ZN:%.*]], i32 3)
+// CPP-CHECK-NEXT: [[TMP1:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 0
+// CPP-CHECK-NEXT: [[TMP2:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> poison, <vscale x 4 x float> [[TMP1]], i64 0)
+// CPP-CHECK-NEXT: [[TMP3:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]], 1
+// CPP-CHECK-NEXT: [[TMP4:%.*]] = tail call <vscale x 8 x float> @llvm.vector.insert.nxv8f32.nxv4f32(<vscale x 8 x float> [[TMP2]], <vscale x 4 x float> [[TMP3]], i64 4)
+// CPP-CHECK-NEXT: ret <vscale x 8 x float> [[TMP4]]
+//
+svfloat32x2_t test_svluti4_lane_zt_f32(svuint8_t zn) __arm_streaming __arm_shared_za __arm_preserves_za {
+ return svluti4_lane_zt_f32_x2(0, zn, 3);
+}
diff --git a/clang/test/Sema/aarch64-sme2-intrinsics/acle_sme2_imm.cpp b/clang/test/Sema/aarch64-sme2-intrinsics/acle_sme2_imm.cpp
index d344d0e459a1c..2f42435804aa3 100644
--- a/clang/test/Sema/aarch64-sme2-intrinsics/acle_sme2_imm.cpp
+++ b/clang/test/Sema/aarch64-sme2-intrinsics/acle_sme2_imm.cpp
@@ -77,3 +77,57 @@ void test_svluti4_lane_zt_x4(svuint8_t zn) __arm_streaming __arm_shared_za __arm
// Test index value range
svluti4_lane_zt_f32_x4(0, zn, 2); // expected-error {{argument value 2 is outside the valid range [0, 1]}}
}
+
+void test_svluti2_lane_zt_x2(svuint8_t zn_u8) __arm_streaming __arm_shared_za __arm_preserves_za {
+ // Test Reg Offset
+ svluti2_lane_zt_u8_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti2_lane_zt_u8_x2(0, zn_u8, 8); // expected-error {{argument value 8 is outside the valid range [0, 7]}}
+ // Test Reg Offset
+ svluti2_lane_zt_u16_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti2_lane_zt_u16_x2(0, zn_u8, 8); // expected-error {{argument value 8 is outside the valid range [0, 7]}}
+ // Test Reg Offset
+ svluti2_lane_zt_u32_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti2_lane_zt_u32_x2(0, zn_u8, 8); // expected-error {{argument value 8 is outside the valid range [0, 7]}}
+ // Test Reg Offset
+ svluti2_lane_zt_f16_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti2_lane_zt_f16_x2(0, zn_u8, 8); // expected-error {{argument value 8 is outside the valid range [0, 7]}}
+ // Test Reg Offset
+ svluti2_lane_zt_bf16_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti2_lane_zt_bf16_x2(0, zn_u8, 8); // expected-error {{argument value 8 is outside the valid range [0, 7]}}
+ // Test Reg Offset
+ svluti2_lane_zt_f32_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti2_lane_zt_f32_x2(0, zn_u8, 8); // expected-error {{argument value 8 is outside the valid range [0, 7]}}
+}
+
+void test_svluti4_lane_zt_x2(svuint8_t zn_u8) __arm_streaming __arm_shared_za __arm_preserves_za {
+ // Test Reg Offset
+ svluti4_lane_zt_u8_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti4_lane_zt_u8_x2(0, zn_u8, 4); // expected-error {{argument value 4 is outside the valid range [0, 3]}}
+ // Test Reg Offset
+ svluti4_lane_zt_u16_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti4_lane_zt_u16_x2(0, zn_u8, 4); // expected-error {{argument value 4 is outside the valid range [0, 3]}}
+ // Test Reg Offset
+ svluti4_lane_zt_u32_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti4_lane_zt_u32_x2(0, zn_u8, 4); // expected-error {{argument value 4 is outside the valid range [0, 3]}}
+ // Test Reg Offset
+ svluti4_lane_zt_f16_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti4_lane_zt_f16_x2(0, zn_u8, 4); // expected-error {{argument value 4 is outside the valid range [0, 3]}}
+ // Test Reg Offset
+ svluti4_lane_zt_bf16_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti4_lane_zt_bf16_x2(0, zn_u8, 4); // expected-error {{argument value 4 is outside the valid range [0, 3]}}
+ // Test Reg Offset
+ svluti4_lane_zt_f32_x2(1, zn_u8, 2); // expected-error {{argument value 1 is outside the valid range [0, 0]}}
+ // Test index value range
+ svluti4_lane_zt_f32_x2(0, zn_u8, 4); // expected-error {{argument value 4 is outside the valid range [0, 3]}}
+}
diff --git a/llvm/include/llvm/IR/IntrinsicsAArch64.td b/llvm/include/llvm/IR/IntrinsicsAArch64.td
index 430f1bc4920b3..0e6bf76ef0847 100644
--- a/llvm/include/llvm/IR/IntrinsicsAArch64.td
+++ b/llvm/include/llvm/IR/IntrinsicsAArch64.td
@@ -3560,6 +3560,17 @@ let TargetPrefix = "aarch64" in {
: DefaultAttrsIntrinsic<[llvm_anyvector_ty, LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>],
[llvm_i32_ty, llvm_nxv16i8_ty, llvm_i32_ty],
[ImmArg<ArgIndex<0>>, ImmArg<ArgIndex<2>>, IntrReadMem]>;
+
+ //
+ // Lookup table expand two registers
+ //
+ def int_aarch64_sme_luti2_lane_zt_x2
+ : DefaultAttrsIntrinsic<[llvm_anyvector_ty, LLVMMatchType<0>], [llvm_i32_ty, llvm_nxv16i8_ty, llvm_i32_ty],
+ [ImmArg<ArgIndex<0>>, ImmArg<ArgIndex<2>>, IntrReadMem]>;
+ def int_aarch64_sme_luti4_lane_zt_x2
+ : DefaultAttrsIntrinsic<[llvm_anyvector_ty, LLVMMatchType<0>], [llvm_i32_ty, llvm_nxv16i8_ty, llvm_i32_ty],
+ [ImmArg<ArgIndex<0>>, ImmArg<ArgIndex<2>>, IntrReadMem]>;
+
}
// SVE2.1 - ZIPQ1, ZIPQ2, UZPQ1, UZPQ2
diff --git a/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp b/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
index 8c108fd6ec4c4..d05d22a7164e7 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
@@ -5120,6 +5120,24 @@ void AArch64DAGToDAGISel::Select(SDNode *Node) {
SelectMultiVectorLuti(Node, 4, Opc, 1);
return;
}
+ case Intrinsic::aarch64_sme_luti2_lane_zt_x2: {
+ if (auto Opc = SelectOpcodeFromVT<SelectTypeKind::AnyType>(
+ Node->getValueType(0),
+ {AArch64::LUTI2_2ZTZI_B, AArch64::LUTI2_2ZTZI_H,
+ AArch64::LUTI2_2ZTZI_S}))
+ // Second Immediate must be <= 7:
+ SelectMultiVectorLuti(Node, 2, Opc, 7);
+ return;
+ }
+ case Intrinsic::aarch64_sme_luti4_lane_zt_x2: {
+ if (auto Opc = SelectOpcodeFromVT<SelectTypeKind::AnyType>(
+ Node->getValueType(0),
+ {AArch64::LUTI4_2ZTZI_B, AArch64::LUTI4_2ZTZI_H,
+ AArch64::LUTI4_2ZTZI_S}))
+ // Second Immediate must be <= 3:
+ SelectMultiVectorLuti(Node, 2, Opc, 3);
+ return;
+ }
}
} break;
case ISD::INTRINSIC_WO_CHAIN: {
diff --git a/llvm/test/CodeGen/AArch64/sme2-intrinsics-luti2-lane-x2.ll b/llvm/test/CodeGen/AArch64/sme2-intrinsics-luti2-lane-x2.ll
new file mode 100644
index 0000000000000..8f9993cf00c59
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/sme2-intrinsics-luti2-lane-x2.ll
@@ -0,0 +1,65 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2 < %s | FileCheck %s
+
+; lookup table expand one register
+
+define {<vscale x 16 x i8>, <vscale x 16 x i8>} @luti2_i8(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti2_i8:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti2 { z0.b, z1.b }, zt0, z0[7]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> %x, i32 7)
+ ret {<vscale x 16 x i8>, <vscale x 16 x i8>} %res
+}
+
+define {<vscale x 8 x i16>, <vscale x 8 x i16>} @luti2_i16(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti2_i16:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti2 { z0.h, z1.h }, zt0, z0[7]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> %x, i32 7)
+ ret {<vscale x 8 x i16>, <vscale x 8 x i16>} %res
+}
+
+define {<vscale x 4 x i32>, <vscale x 4 x i32>} @luti2_i32(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti2_i32:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti2 { z0.s, z1.s }, zt0, z0[7]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> %x, i32 7)
+ ret {<vscale x 4 x i32>, <vscale x 4 x i32>} %res
+}
+
+define {<vscale x 8 x half>, <vscale x 8 x half>} @luti2_f16(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti2_f16:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti2 { z0.h, z1.h }, zt0, z0[7]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8f16(i32 0, <vscale x 16 x i8> %x, i32 7)
+ ret {<vscale x 8 x half>, <vscale x 8 x half>} %res
+}
+
+define {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} @luti2_bf16(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti2_bf16:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti2 { z0.h, z1.h }, zt0, z0[7]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8bf16(i32 0, <vscale x 16 x i8> %x, i32 7)
+ ret {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} %res
+}
+
+define {<vscale x 4 x float>, <vscale x 4 x float>} @luti2_f32(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti2_f32:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti2 { z0.s, z1.s }, zt0, z0[7]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4f32(i32 0, <vscale x 16 x i8> %x, i32 7)
+ ret {<vscale x 4 x float>, <vscale x 4 x float>} %res
+}
+
+declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv16i8(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8i16(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4i32(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8f16(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv8bf16(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.aarch64.sme.luti2.lane.zt.x2.nxv4f32(i32, <vscale x 16 x i8>, i32)
diff --git a/llvm/test/CodeGen/AArch64/sme2-intrinsics-luti4-lane-x2.ll b/llvm/test/CodeGen/AArch64/sme2-intrinsics-luti4-lane-x2.ll
new file mode 100644
index 0000000000000..1be2253dd9239
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/sme2-intrinsics-luti4-lane-x2.ll
@@ -0,0 +1,65 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2 < %s | FileCheck %s
+
+; lookup table expand one register
+
+define {<vscale x 16 x i8>, <vscale x 16 x i8>} @luti4_i8(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti4_i8:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti4 { z0.b, z1.b }, zt0, z0[3]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv16i8(i32 0, <vscale x 16 x i8> %x, i32 3)
+ ret {<vscale x 16 x i8>, <vscale x 16 x i8>} %res
+}
+
+define {<vscale x 8 x i16>, <vscale x 8 x i16>} @luti4_i16(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti4_i16:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti4 { z0.h, z1.h }, zt0, z0[3]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8i16(i32 0, <vscale x 16 x i8> %x, i32 3)
+ ret {<vscale x 8 x i16>, <vscale x 8 x i16>} %res
+}
+
+define {<vscale x 4 x i32>, <vscale x 4 x i32>} @luti4_i32(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti4_i32:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti4 { z0.s, z1.s }, zt0, z0[3]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4i32(i32 0, <vscale x 16 x i8> %x, i32 3)
+ ret {<vscale x 4 x i32>, <vscale x 4 x i32>} %res
+}
+
+define {<vscale x 8 x half>, <vscale x 8 x half>} @luti4_f16(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti4_f16:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti4 { z0.h, z1.h }, zt0, z0[3]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8f16(i32 0, <vscale x 16 x i8> %x, i32 3)
+ ret {<vscale x 8 x half>, <vscale x 8 x half>} %res
+}
+
+define {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} @luti4_bf16(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti4_bf16:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti4 { z0.h, z1.h }, zt0, z0[3]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8bf16(i32 0, <vscale x 16 x i8> %x, i32 3)
+ ret {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} %res
+}
+
+define {<vscale x 4 x float>, <vscale x 4 x float>} @luti4_f32(<vscale x 16 x i8> %x) {
+; CHECK-LABEL: luti4_f32:
+; CHECK: // %bb.0:
+; CHECK-NEXT: luti4 { z0.s, z1.s }, zt0, z0[3]
+; CHECK-NEXT: ret
+ %res = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4f32(i32 0, <vscale x 16 x i8> %x, i32 3)
+ ret {<vscale x 4 x float>, <vscale x 4 x float>} %res
+}
+
+declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv16i8(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8i16(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4i32(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8f16(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 8 x bfloat>, <vscale x 8 x bfloat>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv8bf16(i32, <vscale x 16 x i8>, i32)
+declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.aarch64.sme.luti4.lane.zt.x2.nxv4f32(i32, <vscale x 16 x i8>, i32)
More information about the cfe-commits
mailing list