ARM: add __aeabi_d2h for truncation on AEABI systems

[oota-llvm.git] / test / CodeGen / ARM / vst2.ll

diff --git a/test/CodeGen/ARM/vst2.ll b/test/CodeGen/ARM/vst2.ll
index 497214316a9efb1875f8efda81965914f45e0a5e..2180259d57f79f452dfbb9f526a573c665695cbc 100644 (file)
--- a/test/CodeGen/ARM/vst2.ll
+++ b/test/CodeGen/ARM/vst2.ll
@@ -1,9 +1,9 @@
-; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
+; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
 
 define void @vst2i8(i8* %A, <8 x i8>* %B) nounwind {
-;CHECK: vst2i8:
+;CHECK-LABEL: vst2i8:
 ;Check the alignment value.  Max for this instruction is 128 bits:
-;CHECK: vst2.8 {d16, d17}, [r0, :64]
+;CHECK: vst2.8 {d16, d17}, [r0:64]
        %tmp1 = load <8 x i8>* %B
        call void @llvm.arm.neon.vst2.v8i8(i8* %A, <8 x i8> %tmp1, <8 x i8> %tmp1, i32 8)
        ret void
@@ -11,7 +11,7 @@ define void @vst2i8(i8* %A, <8 x i8>* %B) nounwind {
 
 ;Check for a post-increment updating store with register increment.
 define void @vst2i8_update(i8** %ptr, <8 x i8>* %B, i32 %inc) nounwind {
-;CHECK: vst2i8_update:
+;CHECK-LABEL: vst2i8_update:
 ;CHECK: vst2.8 {d16, d17}, [r1], r2
        %A = load i8** %ptr
        %tmp1 = load <8 x i8>* %B
@@ -22,9 +22,9 @@ define void @vst2i8_update(i8** %ptr, <8 x i8>* %B, i32 %inc) nounwind {
 }
 
 define void @vst2i16(i16* %A, <4 x i16>* %B) nounwind {
-;CHECK: vst2i16:
+;CHECK-LABEL: vst2i16:
 ;Check the alignment value.  Max for this instruction is 128 bits:
-;CHECK: vst2.16 {d16, d17}, [r0, :128]
+;CHECK: vst2.16 {d16, d17}, [r0:128]
        %tmp0 = bitcast i16* %A to i8*
        %tmp1 = load <4 x i16>* %B
        call void @llvm.arm.neon.vst2.v4i16(i8* %tmp0, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 32)
@@ -32,7 +32,7 @@ define void @vst2i16(i16* %A, <4 x i16>* %B) nounwind {
 }
 
 define void @vst2i32(i32* %A, <2 x i32>* %B) nounwind {
-;CHECK: vst2i32:
+;CHECK-LABEL: vst2i32:
 ;CHECK: vst2.32
        %tmp0 = bitcast i32* %A to i8*
        %tmp1 = load <2 x i32>* %B
@@ -41,7 +41,7 @@ define void @vst2i32(i32* %A, <2 x i32>* %B) nounwind {
 }
 
 define void @vst2f(float* %A, <2 x float>* %B) nounwind {
-;CHECK: vst2f:
+;CHECK-LABEL: vst2f:
 ;CHECK: vst2.32
        %tmp0 = bitcast float* %A to i8*
        %tmp1 = load <2 x float>* %B
@@ -50,9 +50,9 @@ define void @vst2f(float* %A, <2 x float>* %B) nounwind {
 }
 
 define void @vst2i64(i64* %A, <1 x i64>* %B) nounwind {
-;CHECK: vst2i64:
+;CHECK-LABEL: vst2i64:
 ;Check the alignment value.  Max for this instruction is 128 bits:
-;CHECK: vst1.64 {d16, d17}, [r0, :128]
+;CHECK: vst1.64 {d16, d17}, [r0:128]
        %tmp0 = bitcast i64* %A to i8*
        %tmp1 = load <1 x i64>* %B
        call void @llvm.arm.neon.vst2.v1i64(i8* %tmp0, <1 x i64> %tmp1, <1 x i64> %tmp1, i32 32)
@@ -61,8 +61,8 @@ define void @vst2i64(i64* %A, <1 x i64>* %B) nounwind {
 
 ;Check for a post-increment updating store.
 define void @vst2i64_update(i64** %ptr, <1 x i64>* %B) nounwind {
-;CHECK: vst2i64_update:
-;CHECK: vst1.64 {d16, d17}, [r1, :64]!
+;CHECK-LABEL: vst2i64_update:
+;CHECK: vst1.64 {d16, d17}, [r1:64]!
        %A = load i64** %ptr
        %tmp0 = bitcast i64* %A to i8*
        %tmp1 = load <1 x i64>* %B
@@ -73,18 +73,18 @@ define void @vst2i64_update(i64** %ptr, <1 x i64>* %B) nounwind {
 }
 
 define void @vst2Qi8(i8* %A, <16 x i8>* %B) nounwind {
-;CHECK: vst2Qi8:
+;CHECK-LABEL: vst2Qi8:
 ;Check the alignment value.  Max for this instruction is 256 bits:
-;CHECK: vst2.8 {d16, d17, d18, d19}, [r0, :64]
+;CHECK: vst2.8 {d16, d17, d18, d19}, [r0:64]
        %tmp1 = load <16 x i8>* %B
        call void @llvm.arm.neon.vst2.v16i8(i8* %A, <16 x i8> %tmp1, <16 x i8> %tmp1, i32 8)
        ret void
 }
 
 define void @vst2Qi16(i16* %A, <8 x i16>* %B) nounwind {
-;CHECK: vst2Qi16:
+;CHECK-LABEL: vst2Qi16:
 ;Check the alignment value.  Max for this instruction is 256 bits:
-;CHECK: vst2.16 {d16, d17, d18, d19}, [r0, :128]
+;CHECK: vst2.16 {d16, d17, d18, d19}, [r0:128]
        %tmp0 = bitcast i16* %A to i8*
        %tmp1 = load <8 x i16>* %B
        call void @llvm.arm.neon.vst2.v8i16(i8* %tmp0, <8 x i16> %tmp1, <8 x i16> %tmp1, i32 16)
@@ -92,9 +92,9 @@ define void @vst2Qi16(i16* %A, <8 x i16>* %B) nounwind {
 }
 
 define void @vst2Qi32(i32* %A, <4 x i32>* %B) nounwind {
-;CHECK: vst2Qi32:
+;CHECK-LABEL: vst2Qi32:
 ;Check the alignment value.  Max for this instruction is 256 bits:
-;CHECK: vst2.32 {d16, d17, d18, d19}, [r0, :256]
+;CHECK: vst2.32 {d16, d17, d18, d19}, [r0:256]
        %tmp0 = bitcast i32* %A to i8*
        %tmp1 = load <4 x i32>* %B
        call void @llvm.arm.neon.vst2.v4i32(i8* %tmp0, <4 x i32> %tmp1, <4 x i32> %tmp1, i32 64)
@@ -102,7 +102,7 @@ define void @vst2Qi32(i32* %A, <4 x i32>* %B) nounwind {
 }
 
 define void @vst2Qf(float* %A, <4 x float>* %B) nounwind {
-;CHECK: vst2Qf:
+;CHECK-LABEL: vst2Qf:
 ;CHECK: vst2.32
        %tmp0 = bitcast float* %A to i8*
        %tmp1 = load <4 x float>* %B
@@ -111,7 +111,7 @@ define void @vst2Qf(float* %A, <4 x float>* %B) nounwind {
 }
 
 define i8* @vst2update(i8* %out, <4 x i16>* %B) nounwind {
-;CHECK: vst2update
+;CHECK-LABEL: vst2update:
 ;CHECK: vst2.16 {d16, d17}, [r0]!
        %tmp1 = load <4 x i16>* %B
        tail call void @llvm.arm.neon.vst2.v4i16(i8* %out, <4 x i16> %tmp1, <4 x i16> %tmp1, i32 2)
@@ -119,6 +119,15 @@ define i8* @vst2update(i8* %out, <4 x i16>* %B) nounwind {
        ret i8* %t5
 }
 
+define i8* @vst2update2(i8 * %out, <4 x float> * %this) nounwind optsize ssp align 2 {
+;CHECK-LABEL: vst2update2:
+;CHECK: vst2.32 {d16, d17, d18, d19}, [r0]!
+  %tmp1 = load <4 x float>* %this
+  call void @llvm.arm.neon.vst2.v4f32(i8* %out, <4 x float> %tmp1, <4 x float> %tmp1, i32 4) nounwind
+  %tmp2 = getelementptr inbounds i8* %out, i32  32
+  ret i8* %tmp2
+}
+
 declare void @llvm.arm.neon.vst2.v8i8(i8*, <8 x i8>, <8 x i8>, i32) nounwind
 declare void @llvm.arm.neon.vst2.v4i16(i8*, <4 x i16>, <4 x i16>, i32) nounwind
 declare void @llvm.arm.neon.vst2.v2i32(i8*, <2 x i32>, <2 x i32>, i32) nounwind