[PM] Replace the Pass argument to SplitEdge with specific analyses used

[oota-llvm.git] / test / CodeGen / ARM / vld3.ll

diff --git a/test/CodeGen/ARM/vld3.ll b/test/CodeGen/ARM/vld3.ll
index ad63e1f716b20d51659bd04662b5155fdda7f2b9..171a03c24da12b15f22ce28c7017afe12b66be62 100644 (file)
--- a/test/CodeGen/ARM/vld3.ll
+++ b/test/CodeGen/ARM/vld3.ll
@@ -1,5 +1,5 @@
-; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
-; RUN: llc < %s -march=arm -mattr=+neon -regalloc=basic | FileCheck %s
+; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o -| FileCheck %s
+; RUN: llc -mtriple=arm-eabi -mattr=+neon -regalloc=basic %s -o - | FileCheck %s
 
 %struct.__neon_int8x8x3_t = type { <8 x i8>,  <8 x i8>,  <8 x i8> }
 %struct.__neon_int16x4x3_t = type { <4 x i16>, <4 x i16>, <4 x i16> }
@@ -13,7 +13,7 @@
 %struct.__neon_float32x4x3_t = type { <4 x float>, <4 x float>, <4 x float> }
 
 define <8 x i8> @vld3i8(i8* %A) nounwind {
-;CHECK: vld3i8:
+;CHECK-LABEL: vld3i8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld3.8 {d16, d17, d18}, [r0:64]
        %tmp1 = call %struct.__neon_int8x8x3_t @llvm.arm.neon.vld3.v8i8(i8* %A, i32 32)
@@ -24,7 +24,7 @@ define <8 x i8> @vld3i8(i8* %A) nounwind {
 }
 
 define <4 x i16> @vld3i16(i16* %A) nounwind {
-;CHECK: vld3i16:
+;CHECK-LABEL: vld3i16:
 ;CHECK: vld3.16
        %tmp0 = bitcast i16* %A to i8*
        %tmp1 = call %struct.__neon_int16x4x3_t @llvm.arm.neon.vld3.v4i16(i8* %tmp0, i32 1)
@@ -36,7 +36,7 @@ define <4 x i16> @vld3i16(i16* %A) nounwind {
 
 ;Check for a post-increment updating load with register increment.
 define <4 x i16> @vld3i16_update(i16** %ptr, i32 %inc) nounwind {
-;CHECK: vld3i16_update:
+;CHECK-LABEL: vld3i16_update:
 ;CHECK: vld3.16 {d16, d17, d18}, [{{r[0-9]+}}], {{r[0-9]+}}
        %A = load i16** %ptr
        %tmp0 = bitcast i16* %A to i8*
@@ -50,7 +50,7 @@ define <4 x i16> @vld3i16_update(i16** %ptr, i32 %inc) nounwind {
 }
 
 define <2 x i32> @vld3i32(i32* %A) nounwind {
-;CHECK: vld3i32:
+;CHECK-LABEL: vld3i32:
 ;CHECK: vld3.32
        %tmp0 = bitcast i32* %A to i8*
        %tmp1 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8* %tmp0, i32 1)
@@ -61,7 +61,7 @@ define <2 x i32> @vld3i32(i32* %A) nounwind {
 }
 
 define <2 x float> @vld3f(float* %A) nounwind {
-;CHECK: vld3f:
+;CHECK-LABEL: vld3f:
 ;CHECK: vld3.32
        %tmp0 = bitcast float* %A to i8*
        %tmp1 = call %struct.__neon_float32x2x3_t @llvm.arm.neon.vld3.v2f32(i8* %tmp0, i32 1)
@@ -72,7 +72,7 @@ define <2 x float> @vld3f(float* %A) nounwind {
 }
 
 define <1 x i64> @vld3i64(i64* %A) nounwind {
-;CHECK: vld3i64:
+;CHECK-LABEL: vld3i64:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld1.64 {d16, d17, d18}, [r0:64]
        %tmp0 = bitcast i64* %A to i8*
@@ -83,8 +83,21 @@ define <1 x i64> @vld3i64(i64* %A) nounwind {
        ret <1 x i64> %tmp4
 }
 
+define <1 x i64> @vld3i64_update(i64** %ptr, i64* %A) nounwind {
+;CHECK-LABEL: vld3i64_update:
+;CHECK: vld1.64        {d16, d17, d18}, [r1:64]!
+        %tmp0 = bitcast i64* %A to i8*
+        %tmp1 = call %struct.__neon_int64x1x3_t @llvm.arm.neon.vld3.v1i64(i8* %tmp0, i32 16)
+        %tmp5 = getelementptr i64* %A, i32 3
+        store i64* %tmp5, i64** %ptr
+        %tmp2 = extractvalue %struct.__neon_int64x1x3_t %tmp1, 0
+        %tmp3 = extractvalue %struct.__neon_int64x1x3_t %tmp1, 2
+        %tmp4 = add <1 x i64> %tmp2, %tmp3
+        ret <1 x i64> %tmp4
+}
+
 define <16 x i8> @vld3Qi8(i8* %A) nounwind {
-;CHECK: vld3Qi8:
+;CHECK-LABEL: vld3Qi8:
 ;Check the alignment value.  Max for this instruction is 64 bits:
 ;CHECK: vld3.8 {d16, d18, d20}, [r0:64]!
 ;CHECK: vld3.8 {d17, d19, d21}, [r0:64]
@@ -96,7 +109,7 @@ define <16 x i8> @vld3Qi8(i8* %A) nounwind {
 }
 
 define <8 x i16> @vld3Qi16(i16* %A) nounwind {
-;CHECK: vld3Qi16:
+;CHECK-LABEL: vld3Qi16:
 ;CHECK: vld3.16
 ;CHECK: vld3.16
        %tmp0 = bitcast i16* %A to i8*
@@ -108,7 +121,7 @@ define <8 x i16> @vld3Qi16(i16* %A) nounwind {
 }
 
 define <4 x i32> @vld3Qi32(i32* %A) nounwind {
-;CHECK: vld3Qi32:
+;CHECK-LABEL: vld3Qi32:
 ;CHECK: vld3.32
 ;CHECK: vld3.32
        %tmp0 = bitcast i32* %A to i8*
@@ -121,7 +134,7 @@ define <4 x i32> @vld3Qi32(i32* %A) nounwind {
 
 ;Check for a post-increment updating load. 
 define <4 x i32> @vld3Qi32_update(i32** %ptr) nounwind {
-;CHECK: vld3Qi32_update:
+;CHECK-LABEL: vld3Qi32_update:
 ;CHECK: vld3.32 {d16, d18, d20}, [r[[R:[0-9]+]]]!
 ;CHECK: vld3.32 {d17, d19, d21}, [r[[R]]]!
        %A = load i32** %ptr
@@ -136,7 +149,7 @@ define <4 x i32> @vld3Qi32_update(i32** %ptr) nounwind {
 }
 
 define <4 x float> @vld3Qf(float* %A) nounwind {
-;CHECK: vld3Qf:
+;CHECK-LABEL: vld3Qf:
 ;CHECK: vld3.32
 ;CHECK: vld3.32
        %tmp0 = bitcast float* %A to i8*