Add codegen support for NEON vld4lane intrinsics with 128-bit vectors.
Also fix some copy-and-paste errors in previous changes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@83590 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/CodeGen/ARM/vldlane.ll b/test/CodeGen/ARM/vldlane.ll
index e9eeba0..53881a3 100644
--- a/test/CodeGen/ARM/vldlane.ll
+++ b/test/CodeGen/ARM/vldlane.ll
@@ -209,6 +209,10 @@
%struct.__neon_int32x2x4_t = type { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }
%struct.__neon_float32x2x4_t = type { <2 x float>, <2 x float>, <2 x float>, <2 x float> }
+%struct.__neon_int16x8x4_t = type { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }
+%struct.__neon_int32x4x4_t = type { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }
+%struct.__neon_float32x4x4_t = type { <4 x float>, <4 x float>, <4 x float>, <4 x float> }
+
define <8 x i8> @vld4lanei8(i8* %A, <8 x i8>* %B) nounwind {
;CHECK: vld4lanei8:
;CHECK: vld4.8
@@ -269,7 +273,56 @@
ret <2 x float> %tmp9
}
+define <8 x i16> @vld4laneQi16(i16* %A, <8 x i16>* %B) nounwind {
+;CHECK: vld4laneQi16:
+;CHECK: vld4.16
+ %tmp1 = load <8 x i16>* %B
+ %tmp2 = call %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4lane.v8i16(i16* %A, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 1)
+ %tmp3 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 0
+ %tmp4 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 1
+ %tmp5 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 2
+ %tmp6 = extractvalue %struct.__neon_int16x8x4_t %tmp2, 3
+ %tmp7 = add <8 x i16> %tmp3, %tmp4
+ %tmp8 = add <8 x i16> %tmp5, %tmp6
+ %tmp9 = add <8 x i16> %tmp7, %tmp8
+ ret <8 x i16> %tmp9
+}
+
+define <4 x i32> @vld4laneQi32(i32* %A, <4 x i32>* %B) nounwind {
+;CHECK: vld4laneQi32:
+;CHECK: vld4.32
+ %tmp1 = load <4 x i32>* %B
+ %tmp2 = call %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4lane.v4i32(i32* %A, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 1)
+ %tmp3 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 0
+ %tmp4 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 1
+ %tmp5 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 2
+ %tmp6 = extractvalue %struct.__neon_int32x4x4_t %tmp2, 3
+ %tmp7 = add <4 x i32> %tmp3, %tmp4
+ %tmp8 = add <4 x i32> %tmp5, %tmp6
+ %tmp9 = add <4 x i32> %tmp7, %tmp8
+ ret <4 x i32> %tmp9
+}
+
+define <4 x float> @vld4laneQf(float* %A, <4 x float>* %B) nounwind {
+;CHECK: vld4laneQf:
+;CHECK: vld4.32
+ %tmp1 = load <4 x float>* %B
+ %tmp2 = call %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4lane.v4f32(float* %A, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, <4 x float> %tmp1, i32 1)
+ %tmp3 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 0
+ %tmp4 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 1
+ %tmp5 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 2
+ %tmp6 = extractvalue %struct.__neon_float32x4x4_t %tmp2, 3
+ %tmp7 = add <4 x float> %tmp3, %tmp4
+ %tmp8 = add <4 x float> %tmp5, %tmp6
+ %tmp9 = add <4 x float> %tmp7, %tmp8
+ ret <4 x float> %tmp9
+}
+
declare %struct.__neon_int8x8x4_t @llvm.arm.neon.vld4lane.v8i8(i8*, <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8>, i32) nounwind readonly
declare %struct.__neon_int16x4x4_t @llvm.arm.neon.vld4lane.v4i16(i8*, <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16>, i32) nounwind readonly
declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4lane.v2i32(i8*, <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, i32) nounwind readonly
declare %struct.__neon_float32x2x4_t @llvm.arm.neon.vld4lane.v2f32(i8*, <2 x float>, <2 x float>, <2 x float>, <2 x float>, i32) nounwind readonly
+
+declare %struct.__neon_int16x8x4_t @llvm.arm.neon.vld4lane.v8i16(i8*, <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16>, i32) nounwind readonly
+declare %struct.__neon_int32x4x4_t @llvm.arm.neon.vld4lane.v4i32(i8*, <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32>, i32) nounwind readonly
+declare %struct.__neon_float32x4x4_t @llvm.arm.neon.vld4lane.v4f32(i8*, <4 x float>, <4 x float>, <4 x float>, <4 x float>, i32) nounwind readonly