diff options
| author | Dimitry Andric <dim@FreeBSD.org> | 2017-06-01 20:58:49 +0000 |
|---|---|---|
| committer | Dimitry Andric <dim@FreeBSD.org> | 2017-06-01 20:58:49 +0000 |
| commit | 416ada0f75bab22b084a1776deb229cd4a669c4d (patch) | |
| tree | 6eb65f3790434471361628af6199b07a4de92de7 /test | |
| parent | 550ae89a710bf458d47e5b1d183f5e7039c2b384 (diff) | |
Notes
Diffstat (limited to 'test')
37 files changed, 1488 insertions, 679 deletions
diff --git a/test/CodeGen/arm_neon_intrinsics.c b/test/CodeGen/arm_neon_intrinsics.c index ae7c78e08f86..62888dd73339 100644 --- a/test/CodeGen/arm_neon_intrinsics.c +++ b/test/CodeGen/arm_neon_intrinsics.c @@ -3,8 +3,6 @@ // RUN: -disable-O0-optnone -emit-llvm -o - %s \ // RUN: | opt -S -mem2reg | FileCheck %s -// REQUIRES: long-tests - #include <arm_neon.h> // CHECK-LABEL: @test_vaba_s8( @@ -3481,11 +3479,11 @@ float32_t test_vgetq_lane_f32(float32x4_t a) { } // CHECK-LABEL: @test_vgetq_lane_f16( -// CHECK: [[__REINT_244:%.*]] = alloca <8 x half>, align 8 +// CHECK: [[__REINT_244:%.*]] = alloca <8 x half>, align 16 // CHECK: [[__REINT1_244:%.*]] = alloca i16, align 2 -// CHECK: store <8 x half> %a, <8 x half>* [[__REINT_244]], align 8 +// CHECK: store <8 x half> %a, <8 x half>* [[__REINT_244]], align 16 // CHECK: [[TMP0:%.*]] = bitcast <8 x half>* [[__REINT_244]] to <8 x i16>* -// CHECK: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* [[TMP0]], align 8 +// CHECK: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* [[TMP0]], align 16 // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> [[TMP1]] to <16 x i8> // CHECK: [[TMP3:%.*]] = bitcast <16 x i8> [[TMP2]] to <8 x i16> // CHECK: [[VGET_LANE:%.*]] = extractelement <8 x i16> [[TMP3]], i32 3 @@ -4543,7 +4541,7 @@ poly16x4_t test_vld1_lane_p16(poly16_t const * a, poly16x4_t b) { } // CHECK-LABEL: @test_vld2q_u8( -// CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET]] to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <16 x i8>, <16 x i8> uint8x16x2_t test_vld2q_u8(uint8_t const * a) { @@ -4551,7 +4549,7 @@ uint8x16x2_t test_vld2q_u8(uint8_t const * a) { } // CHECK-LABEL: @test_vld2q_u16( -// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <8 x i16>, <8 x i16> @@ -4560,7 +4558,7 @@ uint16x8x2_t test_vld2q_u16(uint16_t const * a) { } // CHECK-LABEL: @test_vld2q_u32( -// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <4 x i32>, <4 x i32> @@ -4569,7 +4567,7 @@ uint32x4x2_t test_vld2q_u32(uint32_t const * a) { } // CHECK-LABEL: @test_vld2q_s8( -// CHECK: [[__RET:%.*]] = alloca %struct.int8x16x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__RET]] to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <16 x i8>, <16 x i8> int8x16x2_t test_vld2q_s8(int8_t const * a) { @@ -4577,7 +4575,7 @@ int8x16x2_t test_vld2q_s8(int8_t const * a) { } // CHECK-LABEL: @test_vld2q_s16( -// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <8 x i16>, <8 x i16> @@ -4586,7 +4584,7 @@ int16x8x2_t test_vld2q_s16(int16_t const * a) { } // CHECK-LABEL: @test_vld2q_s32( -// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <4 x i32>, <4 x i32> @@ -4595,7 +4593,7 @@ int32x4x2_t test_vld2q_s32(int32_t const * a) { } // CHECK-LABEL: @test_vld2q_f16( -// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <8 x i16>, <8 x i16> @@ -4604,7 +4602,7 @@ float16x8x2_t test_vld2q_f16(float16_t const * a) { } // CHECK-LABEL: @test_vld2q_f32( -// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <4 x float>, <4 x float> @@ -4613,7 +4611,7 @@ float32x4x2_t test_vld2q_f32(float32_t const * a) { } // CHECK-LABEL: @test_vld2q_p8( -// CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET]] to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <16 x i8>, <16 x i8> poly8x16x2_t test_vld2q_p8(poly8_t const * a) { @@ -4621,7 +4619,7 @@ poly8x16x2_t test_vld2q_p8(poly8_t const * a) { } // CHECK-LABEL: @test_vld2q_p16( -// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD2Q_V:%.*]] = call { <8 x i16>, <8 x i16> @@ -4840,24 +4838,24 @@ poly16x4x2_t test_vld2_dup_p16(poly16_t const * a) { } // CHECK-LABEL: @test_vld2q_lane_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -4867,24 +4865,24 @@ uint16x8x2_t test_vld2q_lane_u16(uint16_t const * a, uint16x8x2_t b) { } // CHECK-LABEL: @test_vld2q_lane_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x i32>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32> @@ -4894,24 +4892,24 @@ uint32x4x2_t test_vld2q_lane_u32(uint32_t const * a, uint32x4x2_t b) { } // CHECK-LABEL: @test_vld2q_lane_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.int16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -4921,24 +4919,24 @@ int16x8x2_t test_vld2q_lane_s16(int16_t const * a, int16x8x2_t b) { } // CHECK-LABEL: @test_vld2q_lane_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x i32>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.int32x4x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32> @@ -4948,24 +4946,24 @@ int32x4x2_t test_vld2q_lane_s32(int32_t const * a, int32x4x2_t b) { } // CHECK-LABEL: @test_vld2q_lane_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x half>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.float16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x half> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x half> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -4975,24 +4973,24 @@ float16x8x2_t test_vld2q_lane_f16(float16_t const * a, float16x8x2_t b) { } // CHECK-LABEL: @test_vld2q_lane_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x float>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.float32x4x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x float> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x float> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x float> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x float> @@ -5002,24 +5000,24 @@ float32x4x2_t test_vld2q_lane_f32(float32_t const * a, float32x4x2_t b) { } // CHECK-LABEL: @test_vld2q_lane_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -5284,7 +5282,7 @@ poly16x4x2_t test_vld2_lane_p16(poly16_t const * a, poly16x4x2_t b) { } // CHECK-LABEL: @test_vld3q_u8( -// CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x3_t* [[__RET]] to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> uint8x16x3_t test_vld3q_u8(uint8_t const * a) { @@ -5292,7 +5290,7 @@ uint8x16x3_t test_vld3q_u8(uint8_t const * a) { } // CHECK-LABEL: @test_vld3q_u16( -// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> @@ -5301,7 +5299,7 @@ uint16x8x3_t test_vld3q_u16(uint16_t const * a) { } // CHECK-LABEL: @test_vld3q_u32( -// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> @@ -5310,7 +5308,7 @@ uint32x4x3_t test_vld3q_u32(uint32_t const * a) { } // CHECK-LABEL: @test_vld3q_s8( -// CHECK: [[__RET:%.*]] = alloca %struct.int8x16x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int8x16x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x3_t* [[__RET]] to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> int8x16x3_t test_vld3q_s8(int8_t const * a) { @@ -5318,7 +5316,7 @@ int8x16x3_t test_vld3q_s8(int8_t const * a) { } // CHECK-LABEL: @test_vld3q_s16( -// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> @@ -5327,7 +5325,7 @@ int16x8x3_t test_vld3q_s16(int16_t const * a) { } // CHECK-LABEL: @test_vld3q_s32( -// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> @@ -5336,7 +5334,7 @@ int32x4x3_t test_vld3q_s32(int32_t const * a) { } // CHECK-LABEL: @test_vld3q_f16( -// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> @@ -5345,7 +5343,7 @@ float16x8x3_t test_vld3q_f16(float16_t const * a) { } // CHECK-LABEL: @test_vld3q_f32( -// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <4 x float>, <4 x float>, <4 x float> @@ -5354,7 +5352,7 @@ float32x4x3_t test_vld3q_f32(float32_t const * a) { } // CHECK-LABEL: @test_vld3q_p8( -// CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x3_t* [[__RET]] to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> poly8x16x3_t test_vld3q_p8(poly8_t const * a) { @@ -5362,7 +5360,7 @@ poly8x16x3_t test_vld3q_p8(poly8_t const * a) { } // CHECK-LABEL: @test_vld3q_p16( -// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x3_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x3_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD3Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> @@ -5581,28 +5579,28 @@ poly16x4x3_t test_vld3_dup_p16(poly16_t const * a) { } // CHECK-LABEL: @test_vld3q_lane_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.uint16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -5613,28 +5611,28 @@ uint16x8x3_t test_vld3q_lane_u16(uint16_t const * a, uint16x8x3_t b) { } // CHECK-LABEL: @test_vld3q_lane_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x i32>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.uint32x4x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <4 x i32> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32> @@ -5645,28 +5643,28 @@ uint32x4x3_t test_vld3q_lane_u32(uint32_t const * a, uint32x4x3_t b) { } // CHECK-LABEL: @test_vld3q_lane_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.int16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -5677,28 +5675,28 @@ int16x8x3_t test_vld3q_lane_s16(int16_t const * a, int16x8x3_t b) { } // CHECK-LABEL: @test_vld3q_lane_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x i32>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.int32x4x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <4 x i32> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32> @@ -5709,28 +5707,28 @@ int32x4x3_t test_vld3q_lane_s32(int32_t const * a, int32x4x3_t b) { } // CHECK-LABEL: @test_vld3q_lane_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x half>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.float16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x half> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x half> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x half> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -5741,28 +5739,28 @@ float16x8x3_t test_vld3q_lane_f16(float16_t const * a, float16x8x3_t b) { } // CHECK-LABEL: @test_vld3q_lane_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x float>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.float32x4x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x float> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x float> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <4 x float> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x float> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x float> @@ -5773,28 +5771,28 @@ float32x4x3_t test_vld3q_lane_f32(float32_t const * a, float32x4x3_t b) { } // CHECK-LABEL: @test_vld3q_lane_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.poly16x8x3_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -6104,7 +6102,7 @@ poly16x4x3_t test_vld3_lane_p16(poly16_t const * a, poly16x4x3_t b) { } // CHECK-LABEL: @test_vld4q_u8( -// CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x4_t* [[__RET]] to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> uint8x16x4_t test_vld4q_u8(uint8_t const * a) { @@ -6112,7 +6110,7 @@ uint8x16x4_t test_vld4q_u8(uint8_t const * a) { } // CHECK-LABEL: @test_vld4q_u16( -// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> @@ -6121,7 +6119,7 @@ uint16x8x4_t test_vld4q_u16(uint16_t const * a) { } // CHECK-LABEL: @test_vld4q_u32( -// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> @@ -6130,7 +6128,7 @@ uint32x4x4_t test_vld4q_u32(uint32_t const * a) { } // CHECK-LABEL: @test_vld4q_s8( -// CHECK: [[__RET:%.*]] = alloca %struct.int8x16x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int8x16x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x4_t* [[__RET]] to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> int8x16x4_t test_vld4q_s8(int8_t const * a) { @@ -6138,7 +6136,7 @@ int8x16x4_t test_vld4q_s8(int8_t const * a) { } // CHECK-LABEL: @test_vld4q_s16( -// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> @@ -6147,7 +6145,7 @@ int16x8x4_t test_vld4q_s16(int16_t const * a) { } // CHECK-LABEL: @test_vld4q_s32( -// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> @@ -6156,7 +6154,7 @@ int32x4x4_t test_vld4q_s32(int32_t const * a) { } // CHECK-LABEL: @test_vld4q_f16( -// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> @@ -6165,7 +6163,7 @@ float16x8x4_t test_vld4q_f16(float16_t const * a) { } // CHECK-LABEL: @test_vld4q_f32( -// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <4 x float>, <4 x float>, <4 x float>, <4 x float> @@ -6174,7 +6172,7 @@ float32x4x4_t test_vld4q_f32(float32_t const * a) { } // CHECK-LABEL: @test_vld4q_p8( -// CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x4_t* [[__RET]] to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> poly8x16x4_t test_vld4q_p8(poly8_t const * a) { @@ -6182,7 +6180,7 @@ poly8x16x4_t test_vld4q_p8(poly8_t const * a) { } // CHECK-LABEL: @test_vld4q_p16( -// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x4_t, align 8 +// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x4_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8* // CHECK: [[VLD4Q_V:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> @@ -6401,32 +6399,32 @@ poly16x4x4_t test_vld4_dup_p16(poly16_t const * a) { } // CHECK-LABEL: @test_vld4q_lane_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.uint16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <8 x i16> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -6438,32 +6436,32 @@ uint16x8x4_t test_vld4q_lane_u16(uint16_t const * a, uint16x8x4_t b) { } // CHECK-LABEL: @test_vld4q_lane_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x i32>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.uint32x4x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <4 x i32> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <4 x i32> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32> @@ -6475,32 +6473,32 @@ uint32x4x4_t test_vld4q_lane_u32(uint32_t const * a, uint32x4x4_t b) { } // CHECK-LABEL: @test_vld4q_lane_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.int16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.int16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <8 x i16> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -6512,32 +6510,32 @@ int16x8x4_t test_vld4q_lane_s16(int16_t const * a, int16x8x4_t b) { } // CHECK-LABEL: @test_vld4q_lane_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.int32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x i32>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.int32x4x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <4 x i32> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <4 x i32> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32> @@ -6549,32 +6547,32 @@ int32x4x4_t test_vld4q_lane_s32(int32_t const * a, int32x4x4_t b) { } // CHECK-LABEL: @test_vld4q_lane_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.float16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x half>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.float16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x half> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x half> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x half> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <8 x half> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -6586,32 +6584,32 @@ float16x8x4_t test_vld4q_lane_f16(float16_t const * a, float16x8x4_t b) { } // CHECK-LABEL: @test_vld4q_lane_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.float32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x float>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.float32x4x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <4 x float> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <4 x float> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <4 x float> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <4 x float> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x float> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x float> @@ -6623,32 +6621,32 @@ float32x4x4_t test_vld4q_lane_f32(float32_t const * a, float32x4x4_t b) { } // CHECK-LABEL: @test_vld4q_lane_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 8 -// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 16 +// CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast %struct.poly16x8x4_t* [[__RET]] to i8* // CHECK: [[TMP4:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP11:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP11:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP12:%.*]] = bitcast <8 x i16> [[TMP11]] to <16 x i8> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16> // CHECK: [[TMP14:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16> @@ -14549,21 +14547,21 @@ float32x4_t test_vsetq_lane_f32(float32_t a, float32x4_t b) { // CHECK-LABEL: @test_vsetq_lane_f16( // CHECK: [[__REINT_248:%.*]] = alloca half, align 2 -// CHECK: [[__REINT1_248:%.*]] = alloca <8 x half>, align 8 -// CHECK: [[__REINT2_248:%.*]] = alloca <8 x i16>, align 8 +// CHECK: [[__REINT1_248:%.*]] = alloca <8 x half>, align 16 +// CHECK: [[__REINT2_248:%.*]] = alloca <8 x i16>, align 16 // CHECK: [[TMP0:%.*]] = load half, half* %a, align 2 // CHECK: store half [[TMP0]], half* [[__REINT_248]], align 2 -// CHECK: store <8 x half> %b, <8 x half>* [[__REINT1_248]], align 8 +// CHECK: store <8 x half> %b, <8 x half>* [[__REINT1_248]], align 16 // CHECK: [[TMP1:%.*]] = bitcast half* [[__REINT_248]] to i16* // CHECK: [[TMP2:%.*]] = load i16, i16* [[TMP1]], align 2 // CHECK: [[TMP3:%.*]] = bitcast <8 x half>* [[__REINT1_248]] to <8 x i16>* -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[TMP3]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[TMP3]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[TMP6:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[VSET_LANE:%.*]] = insertelement <8 x i16> [[TMP6]], i16 [[TMP2]], i32 3 -// CHECK: store <8 x i16> [[VSET_LANE]], <8 x i16>* [[__REINT2_248]], align 8 +// CHECK: store <8 x i16> [[VSET_LANE]], <8 x i16>* [[__REINT2_248]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16>* [[__REINT2_248]] to <8 x half>* -// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[TMP7]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[TMP7]], align 16 // CHECK: ret <8 x half> [[TMP8]] float16x8_t test_vsetq_lane_f16(float16_t *a, float16x8_t b) { return vsetq_lane_f16(*a, b, 3); @@ -16194,20 +16192,20 @@ void test_vst1_lane_p16(poly16_t * a, poly16x4_t b) { } // CHECK-LABEL: @test_vst2q_u8( -// CHECK: [[B:%.*]] = alloca %struct.uint8x16x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint8x16x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint8x16x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint8x16x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <16 x i8>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint8x16x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint8x16x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: call void @llvm.arm.neon.vst2.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i32 1) // CHECK: ret void void test_vst2q_u8(uint8_t * a, uint8x16x2_t b) { @@ -16215,22 +16213,22 @@ void test_vst2q_u8(uint8_t * a, uint8x16x2_t b) { } // CHECK-LABEL: @test_vst2q_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16241,22 +16239,22 @@ void test_vst2q_u16(uint16_t * a, uint16x8x2_t b) { } // CHECK-LABEL: @test_vst2q_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x i32>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -16267,20 +16265,20 @@ void test_vst2q_u32(uint32_t * a, uint32x4x2_t b) { } // CHECK-LABEL: @test_vst2q_s8( -// CHECK: [[B:%.*]] = alloca %struct.int8x16x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int8x16x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int8x16x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int8x16x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <16 x i8>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int8x16x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int8x16x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: call void @llvm.arm.neon.vst2.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i32 1) // CHECK: ret void void test_vst2q_s8(int8_t * a, int8x16x2_t b) { @@ -16288,22 +16286,22 @@ void test_vst2q_s8(int8_t * a, int8x16x2_t b) { } // CHECK-LABEL: @test_vst2q_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16314,22 +16312,22 @@ void test_vst2q_s16(int16_t * a, int16x8x2_t b) { } // CHECK-LABEL: @test_vst2q_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x i32>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -16340,22 +16338,22 @@ void test_vst2q_s32(int32_t * a, int32x4x2_t b) { } // CHECK-LABEL: @test_vst2q_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x half>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16366,22 +16364,22 @@ void test_vst2q_f16(float16_t * a, float16x8x2_t b) { } // CHECK-LABEL: @test_vst2q_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x float>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float> @@ -16392,20 +16390,20 @@ void test_vst2q_f32(float32_t * a, float32x4x2_t b) { } // CHECK-LABEL: @test_vst2q_p8( -// CHECK: [[B:%.*]] = alloca %struct.poly8x16x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly8x16x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly8x16x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly8x16x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <16 x i8>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly8x16x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly8x16x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: call void @llvm.arm.neon.vst2.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i32 1) // CHECK: ret void void test_vst2q_p8(poly8_t * a, poly8x16x2_t b) { @@ -16413,22 +16411,22 @@ void test_vst2q_p8(poly8_t * a, poly8x16x2_t b) { } // CHECK-LABEL: @test_vst2q_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16736,22 +16734,22 @@ void test_vst2_p16(poly16_t * a, poly16x4x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16762,22 +16760,22 @@ void test_vst2q_lane_u16(uint16_t * a, uint16x8x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x i32>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -16788,22 +16786,22 @@ void test_vst2q_lane_u32(uint32_t * a, uint32x4x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16814,22 +16812,22 @@ void test_vst2q_lane_s16(int16_t * a, int16x8x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x i32>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -16840,22 +16838,22 @@ void test_vst2q_lane_s32(int32_t * a, int32x4x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x half>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -16866,22 +16864,22 @@ void test_vst2q_lane_f16(float16_t * a, float16x8x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <4 x float>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float> @@ -16892,22 +16890,22 @@ void test_vst2q_lane_f32(float32_t * a, float32x4x2_t b) { } // CHECK-LABEL: @test_vst2q_lane_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x2_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x2_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [2 x <8 x i16>]* [[COERCE_DIVE]] to [4 x i64]* -// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 8 +// CHECK: store [4 x i64] [[B]].coerce, [4 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x2_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x2_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 32, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17163,23 +17161,23 @@ void test_vst2_lane_p16(poly16_t * a, poly16x4x2_t b) { } // CHECK-LABEL: @test_vst3q_u8( -// CHECK: [[B:%.*]] = alloca %struct.uint8x16x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint8x16x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint8x16x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint8x16x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <16 x i8>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint8x16x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint8x16x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16 // CHECK: call void @llvm.arm.neon.vst3.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i32 1) // CHECK: ret void void test_vst3q_u8(uint8_t * a, uint8x16x3_t b) { @@ -17187,26 +17185,26 @@ void test_vst3q_u8(uint8_t * a, uint8x16x3_t b) { } // CHECK-LABEL: @test_vst3q_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17218,26 +17216,26 @@ void test_vst3q_u16(uint16_t * a, uint16x8x3_t b) { } // CHECK-LABEL: @test_vst3q_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x i32>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -17249,23 +17247,23 @@ void test_vst3q_u32(uint32_t * a, uint32x4x3_t b) { } // CHECK-LABEL: @test_vst3q_s8( -// CHECK: [[B:%.*]] = alloca %struct.int8x16x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int8x16x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int8x16x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int8x16x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <16 x i8>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int8x16x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int8x16x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16 // CHECK: call void @llvm.arm.neon.vst3.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i32 1) // CHECK: ret void void test_vst3q_s8(int8_t * a, int8x16x3_t b) { @@ -17273,26 +17271,26 @@ void test_vst3q_s8(int8_t * a, int8x16x3_t b) { } // CHECK-LABEL: @test_vst3q_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17304,26 +17302,26 @@ void test_vst3q_s16(int16_t * a, int16x8x3_t b) { } // CHECK-LABEL: @test_vst3q_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x i32>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -17335,26 +17333,26 @@ void test_vst3q_s32(int32_t * a, int32x4x3_t b) { } // CHECK-LABEL: @test_vst3q_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x half>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x half> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17366,26 +17364,26 @@ void test_vst3q_f16(float16_t * a, float16x8x3_t b) { } // CHECK-LABEL: @test_vst3q_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x float>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x float> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float> @@ -17397,23 +17395,23 @@ void test_vst3q_f32(float32_t * a, float32x4x3_t b) { } // CHECK-LABEL: @test_vst3q_p8( -// CHECK: [[B:%.*]] = alloca %struct.poly8x16x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly8x16x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly8x16x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly8x16x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <16 x i8>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly8x16x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly8x16x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16 // CHECK: call void @llvm.arm.neon.vst3.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i32 1) // CHECK: ret void void test_vst3q_p8(poly8_t * a, poly8x16x3_t b) { @@ -17421,26 +17419,26 @@ void test_vst3q_p8(poly8_t * a, poly8x16x3_t b) { } // CHECK-LABEL: @test_vst3q_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17803,26 +17801,26 @@ void test_vst3_p16(poly16_t * a, poly16x4x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17834,26 +17832,26 @@ void test_vst3q_lane_u16(uint16_t * a, uint16x8x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x i32>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -17865,26 +17863,26 @@ void test_vst3q_lane_u32(uint32_t * a, uint32x4x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17896,26 +17894,26 @@ void test_vst3q_lane_s16(int16_t * a, int16x8x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x i32>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -17927,26 +17925,26 @@ void test_vst3q_lane_s32(int32_t * a, int32x4x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x half>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x half> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -17958,26 +17956,26 @@ void test_vst3q_lane_f16(float16_t * a, float16x8x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <4 x float>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x float> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float> @@ -17989,26 +17987,26 @@ void test_vst3q_lane_f32(float32_t * a, float32x4x3_t b) { } // CHECK-LABEL: @test_vst3q_lane_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x3_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x3_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [3 x <8 x i16>]* [[COERCE_DIVE]] to [6 x i64]* -// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 8 +// CHECK: store [6 x i64] [[B]].coerce, [6 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x3_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x3_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 48, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -18309,26 +18307,26 @@ void test_vst3_lane_p16(poly16_t * a, poly16x4x3_t b) { } // CHECK-LABEL: @test_vst4q_u8( -// CHECK: [[B:%.*]] = alloca %struct.uint8x16x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint8x16x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint8x16x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint8x16x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <16 x i8>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint8x16x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint8x16x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16 // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16 // CHECK: call void @llvm.arm.neon.vst4.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i32 1) // CHECK: ret void void test_vst4q_u8(uint8_t * a, uint8x16x4_t b) { @@ -18336,30 +18334,30 @@ void test_vst4q_u8(uint8_t * a, uint8x16x4_t b) { } // CHECK-LABEL: @test_vst4q_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -18372,30 +18370,30 @@ void test_vst4q_u16(uint16_t * a, uint16x8x4_t b) { } // CHECK-LABEL: @test_vst4q_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x i32>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <4 x i32> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -18408,26 +18406,26 @@ void test_vst4q_u32(uint32_t * a, uint32x4x4_t b) { } // CHECK-LABEL: @test_vst4q_s8( -// CHECK: [[B:%.*]] = alloca %struct.int8x16x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int8x16x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int8x16x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int8x16x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <16 x i8>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int8x16x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int8x16x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16 // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16 // CHECK: call void @llvm.arm.neon.vst4.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i32 1) // CHECK: ret void void test_vst4q_s8(int8_t * a, int8x16x4_t b) { @@ -18435,30 +18433,30 @@ void test_vst4q_s8(int8_t * a, int8x16x4_t b) { } // CHECK-LABEL: @test_vst4q_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -18471,30 +18469,30 @@ void test_vst4q_s16(int16_t * a, int16x8x4_t b) { } // CHECK-LABEL: @test_vst4q_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x i32>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <4 x i32> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -18507,30 +18505,30 @@ void test_vst4q_s32(int32_t * a, int32x4x4_t b) { } // CHECK-LABEL: @test_vst4q_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x half>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x half> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x half> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -18543,30 +18541,30 @@ void test_vst4q_f16(float16_t * a, float16x8x4_t b) { } // CHECK-LABEL: @test_vst4q_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x float>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x float> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <4 x float> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float> @@ -18579,26 +18577,26 @@ void test_vst4q_f32(float32_t * a, float32x4x4_t b) { } // CHECK-LABEL: @test_vst4q_p8( -// CHECK: [[B:%.*]] = alloca %struct.poly8x16x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly8x16x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly8x16x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly8x16x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <16 x i8>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly8x16x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly8x16x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16 // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16 // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16 // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16 // CHECK: call void @llvm.arm.neon.vst4.p0i8.v16i8(i8* %a, <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i32 1) // CHECK: ret void void test_vst4q_p8(poly8_t * a, poly8x16x4_t b) { @@ -18606,30 +18604,30 @@ void test_vst4q_p8(poly8_t * a, poly8x16x4_t b) { } // CHECK-LABEL: @test_vst4q_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -19047,30 +19045,30 @@ void test_vst4_p16(poly16_t * a, poly16x4x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_u16( -// CHECK: [[B:%.*]] = alloca %struct.uint16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -19083,30 +19081,30 @@ void test_vst4q_lane_u16(uint16_t * a, uint16x8x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_u32( -// CHECK: [[B:%.*]] = alloca %struct.uint32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.uint32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x i32>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.uint32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.uint32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <4 x i32> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -19119,30 +19117,30 @@ void test_vst4q_lane_u32(uint32_t * a, uint32x4x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_s16( -// CHECK: [[B:%.*]] = alloca %struct.int16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -19155,30 +19153,30 @@ void test_vst4q_lane_s16(int16_t * a, int16x8x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_s32( -// CHECK: [[B:%.*]] = alloca %struct.int32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.int32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x i32>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.int32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.int32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i32* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <4 x i32> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32> @@ -19191,30 +19189,30 @@ void test_vst4q_lane_s32(int32_t * a, int32x4x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_f16( -// CHECK: [[B:%.*]] = alloca %struct.float16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x half>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast half* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x half> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x half> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -19227,30 +19225,30 @@ void test_vst4q_lane_f16(float16_t * a, float16x8x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_f32( -// CHECK: [[B:%.*]] = alloca %struct.float32x4x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.float32x4x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <4 x float>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.float32x4x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.float32x4x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast float* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <4 x float> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <4 x float> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float> @@ -19263,30 +19261,30 @@ void test_vst4q_lane_f32(float32_t * a, float32x4x4_t b) { } // CHECK-LABEL: @test_vst4q_lane_p16( -// CHECK: [[B:%.*]] = alloca %struct.poly16x8x4_t, align 8 -// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 8 +// CHECK: [[B:%.*]] = alloca %struct.poly16x8x4_t, align 16 +// CHECK: [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 16 // CHECK: [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[B]], i32 0, i32 0 // CHECK: [[TMP0:%.*]] = bitcast [4 x <8 x i16>]* [[COERCE_DIVE]] to [8 x i64]* -// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 8 +// CHECK: store [8 x i64] [[B]].coerce, [8 x i64]* [[TMP0]], align 16 // CHECK: [[TMP1:%.*]] = bitcast %struct.poly16x8x4_t* [[__S1]] to i8* // CHECK: [[TMP2:%.*]] = bitcast %struct.poly16x8x4_t* [[B]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP1]], i8* [[TMP2]], i32 64, i32 16, i1 false) // CHECK: [[TMP3:%.*]] = bitcast i16* %a to i8* // CHECK: [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i32 0, i32 0 -// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 8 +// CHECK: [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16 // CHECK: [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8> // CHECK: [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i32 0, i32 1 -// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 8 +// CHECK: [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16 // CHECK: [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8> // CHECK: [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i32 0, i32 2 -// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 8 +// CHECK: [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16 // CHECK: [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8> // CHECK: [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0 // CHECK: [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i32 0, i32 3 -// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 8 +// CHECK: [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16 // CHECK: [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8> // CHECK: [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16> // CHECK: [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16> @@ -20631,7 +20629,7 @@ poly16x4x2_t test_vtrn_p16(poly16x4_t a, poly16x4_t b) { } // CHECK-LABEL: @test_vtrnq_s8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VTRN_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 16, i32 2, i32 18, i32 4, i32 20, i32 6, i32 22, i32 8, i32 24, i32 10, i32 26, i32 12, i32 28, i32 14, i32 30> @@ -20641,14 +20639,14 @@ poly16x4x2_t test_vtrn_p16(poly16x4_t a, poly16x4_t b) { // CHECK: store <16 x i8> [[VTRN1_I]], <16 x i8>* [[TMP2]], !noalias !30 // CHECK: [[TMP3:%.*]] = bitcast %struct.int8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.int8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void int8x16x2_t test_vtrnq_s8(int8x16_t a, int8x16_t b) { return vtrnq_s8(a, b); } // CHECK-LABEL: @test_vtrnq_s16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -20660,14 +20658,14 @@ int8x16x2_t test_vtrnq_s8(int8x16_t a, int8x16_t b) { // CHECK: store <8 x i16> [[VTRN1_I]], <8 x i16>* [[TMP4]], !noalias !33 // CHECK: [[TMP5:%.*]] = bitcast %struct.int16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.int16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void int16x8x2_t test_vtrnq_s16(int16x8_t a, int16x8_t b) { return vtrnq_s16(a, b); } // CHECK-LABEL: @test_vtrnq_s32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8> @@ -20679,14 +20677,14 @@ int16x8x2_t test_vtrnq_s16(int16x8_t a, int16x8_t b) { // CHECK: store <4 x i32> [[VTRN1_I]], <4 x i32>* [[TMP4]], !noalias !36 // CHECK: [[TMP5:%.*]] = bitcast %struct.int32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.int32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void int32x4x2_t test_vtrnq_s32(int32x4_t a, int32x4_t b) { return vtrnq_s32(a, b); } // CHECK-LABEL: @test_vtrnq_u8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VTRN_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 16, i32 2, i32 18, i32 4, i32 20, i32 6, i32 22, i32 8, i32 24, i32 10, i32 26, i32 12, i32 28, i32 14, i32 30> @@ -20696,14 +20694,14 @@ int32x4x2_t test_vtrnq_s32(int32x4_t a, int32x4_t b) { // CHECK: store <16 x i8> [[VTRN1_I]], <16 x i8>* [[TMP2]], !noalias !39 // CHECK: [[TMP3:%.*]] = bitcast %struct.uint8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void uint8x16x2_t test_vtrnq_u8(uint8x16_t a, uint8x16_t b) { return vtrnq_u8(a, b); } // CHECK-LABEL: @test_vtrnq_u16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -20715,14 +20713,14 @@ uint8x16x2_t test_vtrnq_u8(uint8x16_t a, uint8x16_t b) { // CHECK: store <8 x i16> [[VTRN1_I]], <8 x i16>* [[TMP4]], !noalias !42 // CHECK: [[TMP5:%.*]] = bitcast %struct.uint16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void uint16x8x2_t test_vtrnq_u16(uint16x8_t a, uint16x8_t b) { return vtrnq_u16(a, b); } // CHECK-LABEL: @test_vtrnq_u32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8> @@ -20734,14 +20732,14 @@ uint16x8x2_t test_vtrnq_u16(uint16x8_t a, uint16x8_t b) { // CHECK: store <4 x i32> [[VTRN1_I]], <4 x i32>* [[TMP4]], !noalias !45 // CHECK: [[TMP5:%.*]] = bitcast %struct.uint32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void uint32x4x2_t test_vtrnq_u32(uint32x4_t a, uint32x4_t b) { return vtrnq_u32(a, b); } // CHECK-LABEL: @test_vtrnq_f32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x float> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x float> %b to <16 x i8> @@ -20753,14 +20751,14 @@ uint32x4x2_t test_vtrnq_u32(uint32x4_t a, uint32x4_t b) { // CHECK: store <4 x float> [[VTRN1_I]], <4 x float>* [[TMP4]], !noalias !48 // CHECK: [[TMP5:%.*]] = bitcast %struct.float32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.float32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void float32x4x2_t test_vtrnq_f32(float32x4_t a, float32x4_t b) { return vtrnq_f32(a, b); } // CHECK-LABEL: @test_vtrnq_p8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.poly8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.poly8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VTRN_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 16, i32 2, i32 18, i32 4, i32 20, i32 6, i32 22, i32 8, i32 24, i32 10, i32 26, i32 12, i32 28, i32 14, i32 30> @@ -20770,14 +20768,14 @@ float32x4x2_t test_vtrnq_f32(float32x4_t a, float32x4_t b) { // CHECK: store <16 x i8> [[VTRN1_I]], <16 x i8>* [[TMP2]], !noalias !51 // CHECK: [[TMP3:%.*]] = bitcast %struct.poly8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void poly8x16x2_t test_vtrnq_p8(poly8x16_t a, poly8x16_t b) { return vtrnq_p8(a, b); } // CHECK-LABEL: @test_vtrnq_p16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -20789,7 +20787,7 @@ poly8x16x2_t test_vtrnq_p8(poly8x16_t a, poly8x16_t b) { // CHECK: store <8 x i16> [[VTRN1_I]], <8 x i16>* [[TMP4]], !noalias !54 // CHECK: [[TMP5:%.*]] = bitcast %struct.poly16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void poly16x8x2_t test_vtrnq_p16(poly16x8_t a, poly16x8_t b) { return vtrnq_p16(a, b); @@ -21125,7 +21123,7 @@ poly16x4x2_t test_vuzp_p16(poly16x4_t a, poly16x4_t b) { } // CHECK-LABEL: @test_vuzpq_s8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VUZP_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14, i32 16, i32 18, i32 20, i32 22, i32 24, i32 26, i32 28, i32 30> @@ -21135,14 +21133,14 @@ poly16x4x2_t test_vuzp_p16(poly16x4_t a, poly16x4_t b) { // CHECK: store <16 x i8> [[VUZP1_I]], <16 x i8>* [[TMP2]], !noalias !84 // CHECK: [[TMP3:%.*]] = bitcast %struct.int8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.int8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void int8x16x2_t test_vuzpq_s8(int8x16_t a, int8x16_t b) { return vuzpq_s8(a, b); } // CHECK-LABEL: @test_vuzpq_s16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -21154,14 +21152,14 @@ int8x16x2_t test_vuzpq_s8(int8x16_t a, int8x16_t b) { // CHECK: store <8 x i16> [[VUZP1_I]], <8 x i16>* [[TMP4]], !noalias !87 // CHECK: [[TMP5:%.*]] = bitcast %struct.int16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.int16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void int16x8x2_t test_vuzpq_s16(int16x8_t a, int16x8_t b) { return vuzpq_s16(a, b); } // CHECK-LABEL: @test_vuzpq_s32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8> @@ -21173,14 +21171,14 @@ int16x8x2_t test_vuzpq_s16(int16x8_t a, int16x8_t b) { // CHECK: store <4 x i32> [[VUZP1_I]], <4 x i32>* [[TMP4]], !noalias !90 // CHECK: [[TMP5:%.*]] = bitcast %struct.int32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.int32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void int32x4x2_t test_vuzpq_s32(int32x4_t a, int32x4_t b) { return vuzpq_s32(a, b); } // CHECK-LABEL: @test_vuzpq_u8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VUZP_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14, i32 16, i32 18, i32 20, i32 22, i32 24, i32 26, i32 28, i32 30> @@ -21190,14 +21188,14 @@ int32x4x2_t test_vuzpq_s32(int32x4_t a, int32x4_t b) { // CHECK: store <16 x i8> [[VUZP1_I]], <16 x i8>* [[TMP2]], !noalias !93 // CHECK: [[TMP3:%.*]] = bitcast %struct.uint8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void uint8x16x2_t test_vuzpq_u8(uint8x16_t a, uint8x16_t b) { return vuzpq_u8(a, b); } // CHECK-LABEL: @test_vuzpq_u16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -21209,14 +21207,14 @@ uint8x16x2_t test_vuzpq_u8(uint8x16_t a, uint8x16_t b) { // CHECK: store <8 x i16> [[VUZP1_I]], <8 x i16>* [[TMP4]], !noalias !96 // CHECK: [[TMP5:%.*]] = bitcast %struct.uint16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void uint16x8x2_t test_vuzpq_u16(uint16x8_t a, uint16x8_t b) { return vuzpq_u16(a, b); } // CHECK-LABEL: @test_vuzpq_u32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8> @@ -21228,14 +21226,14 @@ uint16x8x2_t test_vuzpq_u16(uint16x8_t a, uint16x8_t b) { // CHECK: store <4 x i32> [[VUZP1_I]], <4 x i32>* [[TMP4]], !noalias !99 // CHECK: [[TMP5:%.*]] = bitcast %struct.uint32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void uint32x4x2_t test_vuzpq_u32(uint32x4_t a, uint32x4_t b) { return vuzpq_u32(a, b); } // CHECK-LABEL: @test_vuzpq_f32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x float> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x float> %b to <16 x i8> @@ -21247,14 +21245,14 @@ uint32x4x2_t test_vuzpq_u32(uint32x4_t a, uint32x4_t b) { // CHECK: store <4 x float> [[VUZP1_I]], <4 x float>* [[TMP4]], !noalias !102 // CHECK: [[TMP5:%.*]] = bitcast %struct.float32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.float32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void float32x4x2_t test_vuzpq_f32(float32x4_t a, float32x4_t b) { return vuzpq_f32(a, b); } // CHECK-LABEL: @test_vuzpq_p8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.poly8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.poly8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VUZP_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14, i32 16, i32 18, i32 20, i32 22, i32 24, i32 26, i32 28, i32 30> @@ -21264,14 +21262,14 @@ float32x4x2_t test_vuzpq_f32(float32x4_t a, float32x4_t b) { // CHECK: store <16 x i8> [[VUZP1_I]], <16 x i8>* [[TMP2]], !noalias !105 // CHECK: [[TMP3:%.*]] = bitcast %struct.poly8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void poly8x16x2_t test_vuzpq_p8(poly8x16_t a, poly8x16_t b) { return vuzpq_p8(a, b); } // CHECK-LABEL: @test_vuzpq_p16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -21283,7 +21281,7 @@ poly8x16x2_t test_vuzpq_p8(poly8x16_t a, poly8x16_t b) { // CHECK: store <8 x i16> [[VUZP1_I]], <8 x i16>* [[TMP4]], !noalias !108 // CHECK: [[TMP5:%.*]] = bitcast %struct.poly16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void poly16x8x2_t test_vuzpq_p16(poly16x8_t a, poly16x8_t b) { return vuzpq_p16(a, b); @@ -21455,7 +21453,7 @@ poly16x4x2_t test_vzip_p16(poly16x4_t a, poly16x4_t b) { } // CHECK-LABEL: @test_vzipq_s8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VZIP_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23> @@ -21465,14 +21463,14 @@ poly16x4x2_t test_vzip_p16(poly16x4_t a, poly16x4_t b) { // CHECK: store <16 x i8> [[VZIP1_I]], <16 x i8>* [[TMP2]], !noalias !138 // CHECK: [[TMP3:%.*]] = bitcast %struct.int8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.int8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void int8x16x2_t test_vzipq_s8(int8x16_t a, int8x16_t b) { return vzipq_s8(a, b); } // CHECK-LABEL: @test_vzipq_s16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -21484,14 +21482,14 @@ int8x16x2_t test_vzipq_s8(int8x16_t a, int8x16_t b) { // CHECK: store <8 x i16> [[VZIP1_I]], <8 x i16>* [[TMP4]], !noalias !141 // CHECK: [[TMP5:%.*]] = bitcast %struct.int16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.int16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void int16x8x2_t test_vzipq_s16(int16x8_t a, int16x8_t b) { return vzipq_s16(a, b); } // CHECK-LABEL: @test_vzipq_s32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.int32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.int32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8> @@ -21503,14 +21501,14 @@ int16x8x2_t test_vzipq_s16(int16x8_t a, int16x8_t b) { // CHECK: store <4 x i32> [[VZIP1_I]], <4 x i32>* [[TMP4]], !noalias !144 // CHECK: [[TMP5:%.*]] = bitcast %struct.int32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.int32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void int32x4x2_t test_vzipq_s32(int32x4_t a, int32x4_t b) { return vzipq_s32(a, b); } // CHECK-LABEL: @test_vzipq_u8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VZIP_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23> @@ -21520,14 +21518,14 @@ int32x4x2_t test_vzipq_s32(int32x4_t a, int32x4_t b) { // CHECK: store <16 x i8> [[VZIP1_I]], <16 x i8>* [[TMP2]], !noalias !147 // CHECK: [[TMP3:%.*]] = bitcast %struct.uint8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void uint8x16x2_t test_vzipq_u8(uint8x16_t a, uint8x16_t b) { return vzipq_u8(a, b); } // CHECK-LABEL: @test_vzipq_u16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -21539,14 +21537,14 @@ uint8x16x2_t test_vzipq_u8(uint8x16_t a, uint8x16_t b) { // CHECK: store <8 x i16> [[VZIP1_I]], <8 x i16>* [[TMP4]], !noalias !150 // CHECK: [[TMP5:%.*]] = bitcast %struct.uint16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void uint16x8x2_t test_vzipq_u16(uint16x8_t a, uint16x8_t b) { return vzipq_u16(a, b); } // CHECK-LABEL: @test_vzipq_u32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.uint32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.uint32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8> @@ -21558,14 +21556,14 @@ uint16x8x2_t test_vzipq_u16(uint16x8_t a, uint16x8_t b) { // CHECK: store <4 x i32> [[VZIP1_I]], <4 x i32>* [[TMP4]], !noalias !153 // CHECK: [[TMP5:%.*]] = bitcast %struct.uint32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void uint32x4x2_t test_vzipq_u32(uint32x4_t a, uint32x4_t b) { return vzipq_u32(a, b); } // CHECK-LABEL: @test_vzipq_f32( -// CHECK: [[__RET_I:%.*]] = alloca %struct.float32x4x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.float32x4x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <4 x float> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <4 x float> %b to <16 x i8> @@ -21577,14 +21575,14 @@ uint32x4x2_t test_vzipq_u32(uint32x4_t a, uint32x4_t b) { // CHECK: store <4 x float> [[VZIP1_I]], <4 x float>* [[TMP4]], !noalias !156 // CHECK: [[TMP5:%.*]] = bitcast %struct.float32x4x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.float32x4x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void float32x4x2_t test_vzipq_f32(float32x4_t a, float32x4_t b) { return vzipq_f32(a, b); } // CHECK-LABEL: @test_vzipq_p8( -// CHECK: [[__RET_I:%.*]] = alloca %struct.poly8x16x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.poly8x16x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <16 x i8>* // CHECK: [[VZIP_I:%.*]] = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23> @@ -21594,14 +21592,14 @@ float32x4x2_t test_vzipq_f32(float32x4_t a, float32x4_t b) { // CHECK: store <16 x i8> [[VZIP1_I]], <16 x i8>* [[TMP2]], !noalias !159 // CHECK: [[TMP3:%.*]] = bitcast %struct.poly8x16x2_t* %agg.result to i8* // CHECK: [[TMP4:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP3]], i8* [[TMP4]], i32 32, i32 16, i1 false) // CHECK: ret void poly8x16x2_t test_vzipq_p8(poly8x16_t a, poly8x16_t b) { return vzipq_p8(a, b); } // CHECK-LABEL: @test_vzipq_p16( -// CHECK: [[__RET_I:%.*]] = alloca %struct.poly16x8x2_t, align 8 +// CHECK: [[__RET_I:%.*]] = alloca %struct.poly16x8x2_t, align 16 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET_I]] to i8* // CHECK: [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8> // CHECK: [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8> @@ -21613,7 +21611,7 @@ poly8x16x2_t test_vzipq_p8(poly8x16_t a, poly8x16_t b) { // CHECK: store <8 x i16> [[VZIP1_I]], <8 x i16>* [[TMP4]], !noalias !162 // CHECK: [[TMP5:%.*]] = bitcast %struct.poly16x8x2_t* %agg.result to i8* // CHECK: [[TMP6:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET_I]] to i8* -// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 8, i1 false) +// CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* [[TMP5]], i8* [[TMP6]], i32 32, i32 16, i1 false) // CHECK: ret void poly16x8x2_t test_vzipq_p16(poly16x8_t a, poly16x8_t b) { return vzipq_p16(a, b); diff --git a/test/CodeGen/ubsan-pointer-overflow.m b/test/CodeGen/ubsan-pointer-overflow.m new file mode 100644 index 000000000000..c83d527e624e --- /dev/null +++ b/test/CodeGen/ubsan-pointer-overflow.m @@ -0,0 +1,171 @@ +// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -w -emit-llvm -o - %s -fsanitize=pointer-overflow | FileCheck %s + +// CHECK-LABEL: define void @unary_arith +void unary_arith(char *p) { + // CHECK: [[BASE:%.*]] = ptrtoint i8* {{.*}} to i64, !nosanitize + // CHECK-NEXT: [[COMPGEP:%.*]] = add i64 [[BASE]], 1, !nosanitize + // CHECK-NEXT: [[POSVALID:%.*]] = icmp uge i64 [[COMPGEP]], [[BASE]], !nosanitize + // CHECK-NEXT: [[NEGVALID:%.*]] = icmp ult i64 [[COMPGEP]], [[BASE]], !nosanitize + // CHECK-NEXT: [[DIFFVALID:%.*]] = select i1 true, i1 [[POSVALID]], i1 [[NEGVALID]], !nosanitize + // CHECK-NEXT: [[VALID:%.*]] = and i1 true, [[DIFFVALID]], !nosanitize + // CHECK-NEXT: br i1 [[VALID]]{{.*}}, !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}}, i64 [[BASE]], i64 [[COMPGEP]]){{.*}}, !nosanitize + ++p; + + // CHECK: ptrtoint i8* {{.*}} to i64, !nosanitize + // CHECK-NEXT: add i64 {{.*}}, -1, !nosanitize + // CHECK: select i1 false{{.*}}, !nosanitize + // CHECK-NEXT: and i1 true{{.*}}, !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + --p; + + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + p++; + + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + p--; +} + +// CHECK-LABEL: define void @binary_arith +void binary_arith(char *p, int i) { + // CHECK: [[SMUL:%.*]] = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 1, i64 %{{.*}}), !nosanitize + // CHECK-NEXT: [[SMULOFLOW:%.*]] = extractvalue { i64, i1 } [[SMUL]], 1, !nosanitize + // CHECK-NEXT: [[OFFSETOFLOW:%.*]] = or i1 false, [[SMULOFLOW]], !nosanitize + // CHECK-NEXT: [[SMULVAL:%.*]] = extractvalue { i64, i1 } [[SMUL]], 0, !nosanitize + // CHECK-NEXT: [[BASE:%.*]] = ptrtoint i8* {{.*}} to i64, !nosanitize + // CHECK-NEXT: [[COMPGEP:%.*]] = add i64 [[BASE]], [[SMULVAL]], !nosanitize + // CHECK-NEXT: [[POSVALID:%.*]] = icmp uge i64 [[COMPGEP]], [[BASE]], !nosanitize + // CHECK-NEXT: [[NEGVALID:%.*]] = icmp ult i64 [[COMPGEP]], [[BASE]], !nosanitize + // CHECK-NEXT: [[POSOFFSET:%.*]] = icmp sge i64 [[SMULVAL]], 0, !nosanitize + // CHECK-NEXT: [[OFFSETVALID:%.*]] = xor i1 [[OFFSETOFLOW]], true, !nosanitize + // CHECK-NEXT: [[DIFFVALID:%.*]] = select i1 [[POSOFFSET]], i1 [[POSVALID]], i1 [[NEGVALID]], !nosanitize + // CHECK-NEXT: [[VALID:%.*]] = and i1 [[OFFSETVALID]], [[DIFFVALID]], !nosanitize + // CHECK-NEXT: br i1 [[VALID]]{{.*}}, !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}}, i64 [[BASE]], i64 [[COMPGEP]]){{.*}}, !nosanitize + p + i; + + // CHECK: [[OFFSET:%.*]] = sub i64 0, {{.*}} + // CHECK-NEXT: getelementptr inbounds {{.*}} [[OFFSET]] + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + p - i; +} + +// CHECK-LABEL: define void @fixed_len_array +void fixed_len_array(int k) { + // CHECK: getelementptr inbounds [10 x [10 x i32]], [10 x [10 x i32]]* [[ARR:%.*]], i64 0, i64 [[IDXPROM:%.*]] + // CHECK-NEXT: [[SMUL:%.*]] = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 40, i64 [[IDXPROM]]), !nosanitize + // CHECK-NEXT: [[SMULOFLOW:%.*]] = extractvalue { i64, i1 } [[SMUL]], 1, !nosanitize + // CHECK-NEXT: [[OFFSETOFLOW:%.*]] = or i1 false, [[SMULOFLOW]], !nosanitize + // CHECK-NEXT: [[SMULVAL:%.*]] = extractvalue { i64, i1 } [[SMUL]], 0, !nosanitize + // CHECK-NEXT: [[BASE:%.*]] = ptrtoint [10 x [10 x i32]]* [[ARR]] to i64, !nosanitize + // CHECK-NEXT: [[COMPGEP:%.*]] = add i64 [[BASE]], [[SMULVAL]], !nosanitize + // CHECK-NEXT: [[POSVALID:%.*]] = icmp uge i64 [[COMPGEP]], [[BASE]], !nosanitize + // CHECK-NEXT: [[NEGVALID:%.*]] = icmp ult i64 [[COMPGEP]], [[BASE]], !nosanitize + // CHECK-NEXT: [[POSOFFSET:%.*]] = icmp sge i64 [[SMULVAL]], 0, !nosanitize + // CHECK-NEXT: [[OFFSETVALID:%.*]] = xor i1 [[OFFSETOFLOW]], true, !nosanitize + // CHECK-NEXT: [[DIFFVALID:%.*]] = select i1 [[POSOFFSET]], i1 [[POSVALID]], i1 [[NEGVALID]], !nosanitize + // CHECK-NEXT: [[VALID:%.*]] = and i1 [[OFFSETVALID]], [[DIFFVALID]], !nosanitize + // CHECK-NEXT: br i1 [[VALID]]{{.*}}, !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}}, i64 [[BASE]], i64 [[COMPGEP]]){{.*}}, !nosanitize + + // CHECK: getelementptr inbounds [10 x i32], [10 x i32]* {{.*}}, i64 0, i64 [[IDXPROM1:%.*]] + // CHECK-NEXT: @llvm.smul.with.overflow.i64(i64 4, i64 [[IDXPROM1]]), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + + int arr[10][10]; + arr[k][k]; +} + +// CHECK-LABEL: define void @variable_len_array +void variable_len_array(int n, int k) { + // CHECK: getelementptr inbounds i32, i32* {{.*}}, i64 [[IDXPROM:%.*]] + // CHECK-NEXT: @llvm.smul.with.overflow.i64(i64 4, i64 [[IDXPROM]]), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + + // CHECK: getelementptr inbounds i32, i32* {{.*}}, i64 [[IDXPROM1:%.*]] + // CHECK-NEXT: @llvm.smul.with.overflow.i64(i64 4, i64 [[IDXPROM1]]), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + + int arr[n][n]; + arr[k][k]; +} + +// CHECK-LABEL: define void @pointer_array +void pointer_array(int **arr, int k) { + // CHECK: @llvm.smul.with.overflow.i64(i64 8, i64 {{.*}}), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + + // CHECK: @llvm.smul.with.overflow.i64(i64 4, i64 {{.*}}), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + + arr[k][k]; +} + +struct S1 { + int pad1; + union { + char leaf; + struct S1 *link; + } u; + struct S1 *arr; +}; + +// TODO: Currently, structure GEPs are not checked, so there are several +// potentially unsafe GEPs here which we don't instrument. +// +// CHECK-LABEL: define void @struct_index +void struct_index(struct S1 *p) { + // CHECK: getelementptr inbounds %struct.S1, %struct.S1* [[P:%.*]], i64 10 + // CHECK-NEXT: [[BASE:%.*]] = ptrtoint %struct.S1* [[P]] to i64, !nosanitize + // CHECK-NEXT: [[COMPGEP:%.*]] = add i64 [[BASE]], 240, !nosanitize + // CHECK: @__ubsan_handle_pointer_overflow{{.*}} i64 [[BASE]], i64 [[COMPGEP]]) {{.*}}, !nosanitize + + // CHECK-NOT: @__ubsan_handle_pointer_overflow + + p->arr[10].u.link->u.leaf; +} + +typedef void (*funcptr_t)(void); + +// CHECK-LABEL: define void @function_pointer_arith +void function_pointer_arith(funcptr_t *p, int k) { + // CHECK: add i64 {{.*}}, 8, !nosanitize + // CHECK: @__ubsan_handle_pointer_overflow{{.*}} + ++p; + + // CHECK: @llvm.smul.with.overflow.i64(i64 8, i64 {{.*}}), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + p + k; +} + +// CHECK-LABEL: define void @variable_len_array_arith +void variable_len_array_arith(int n, int k) { + int vla[n]; + int (*p)[n] = &vla; + + // CHECK: getelementptr inbounds i32, i32* {{.*}}, i64 [[INC:%.*]] + // CHECK: @llvm.smul.with.overflow.i64(i64 4, i64 [[INC]]), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + ++p; + + // CHECK: getelementptr inbounds i32, i32* {{.*}}, i64 [[IDXPROM:%.*]] + // CHECK: @llvm.smul.with.overflow.i64(i64 4, i64 [[IDXPROM]]), !nosanitize + // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}} + p + k; +} + +// CHECK-LABEL: define void @objc_id +void objc_id(id *p) { + // CHECK: add i64 {{.*}}, 8, !nosanitize + // CHECK: @__ubsan_handle_pointer_overflow{{.*}} + p++; +} + +// CHECK-LABEL: define void @dont_emit_checks_for_no_op_GEPs +// CHECK-NOT: __ubsan_handle_pointer_overflow +void dont_emit_checks_for_no_op_GEPs(char *p) { + &p[0]; + + int arr[10][10]; + &arr[0][0]; +} diff --git a/test/CodeGenCXX/stmtexpr.cpp b/test/CodeGenCXX/stmtexpr.cpp index 5885a1663e63..5bd9908d6c25 100644 --- a/test/CodeGenCXX/stmtexpr.cpp +++ b/test/CodeGenCXX/stmtexpr.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -Wno-unused-value -triple %itanium_abi_triple -emit-llvm -o - %s | FileCheck %s +// RUN: %clang_cc1 -Wno-unused-value -triple i686-linux-gnu -emit-llvm -o - %s | FileCheck %s // rdar: //8540501 extern "C" int printf(...); extern "C" void abort(); @@ -139,6 +139,34 @@ extern "C" int cleanup_exit_lvalue(bool cond) { // CHECK: %[[v:[^ ]*]] = load i32*, i32** %[[tmp]] // CHECK-NEXT: store i32* %[[v]], i32** %r +// Bind the reference to a byval argument. It is not an instruction or Constant, +// so it's a bit of a corner case. +struct ByVal { int x[3]; }; +extern "C" int cleanup_exit_lvalue_byval(bool cond, ByVal arg) { + ByVal &r = (A(1), ({ if (cond) return 0; (void)ByVal(); }), arg); + return r.x[0]; +} +// CHECK-LABEL: define{{.*}} i32 @cleanup_exit_lvalue_byval({{.*}}, %struct.ByVal* byval align 4 %arg) +// CHECK: call {{.*}} @_ZN1AC1Ei +// CHECK: call {{.*}} @_ZN1AD1Ev +// CHECK: switch +// CHECK: store %struct.ByVal* %arg, %struct.ByVal** %r + +// Bind the reference to a local variable. We don't need to spill it. Binding a +// reference to it doesn't generate any instructions. +extern "C" int cleanup_exit_lvalue_local(bool cond) { + int local = 42; + int &r = (A(1), ({ if (cond) return 0; (void)0; }), local); + return r; +} +// CHECK-LABEL: define{{.*}} i32 @cleanup_exit_lvalue_local({{.*}}) +// CHECK: %local = alloca i32 +// CHECK: store i32 42, i32* %local +// CHECK: call {{.*}} @_ZN1AC1Ei +// CHECK-NOT: store i32* %local +// CHECK: call {{.*}} @_ZN1AD1Ev +// CHECK: switch +// CHECK: store i32* %local, i32** %r, align 4 // We handle ExprWithCleanups for complex evaluation type separately, and it had // the same bug. diff --git a/test/CodeGenCXX/strict-vtable-pointers.cpp b/test/CodeGenCXX/strict-vtable-pointers.cpp index 928817bfb1f5..c3798920abdd 100644 --- a/test/CodeGenCXX/strict-vtable-pointers.cpp +++ b/test/CodeGenCXX/strict-vtable-pointers.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 %s -I%S -triple=x86_64-apple-darwin10 -fstrict-vtable-pointers -disable-llvm-passes -O2 -emit-llvm -o %t.ll +// RUN: %clang_cc1 %s -I%S -triple=x86_64-apple-darwin10 -fstrict-vtable-pointers -std=c++11 -disable-llvm-passes -O2 -emit-llvm -o %t.ll // RUN: FileCheck --check-prefix=CHECK-CTORS %s < %t.ll // RUN: FileCheck --check-prefix=CHECK-NEW %s < %t.ll // RUN: FileCheck --check-prefix=CHECK-DTORS %s < %t.ll @@ -180,6 +180,119 @@ struct DynamicFromStatic; // CHECK-CTORS-NOT: @llvm.invariant.group.barrier( // CHECK-CTORS-LABEL: {{^}}} +struct A { + virtual void foo(); +}; +struct B : A { + virtual void foo(); +}; + +union U { + A a; + B b; +}; + +void changeToB(U *u); +void changeToA(U *u); + +void g2(A *a) { + a->foo(); +} +// We have to guard access to union fields with invariant.group, because +// it is very easy to skip the barrier with unions. In this example the inlined +// g2 will produce loads with the same !invariant.group metadata, and +// u->a and u->b would use the same pointer. +// CHECK-NEW-LABEL: define void @_Z14UnionsBarriersP1U +void UnionsBarriers(U *u) { + // CHECK-NEW: call void @_Z9changeToBP1U( + changeToB(u); + // CHECK-NEW: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: call void @_Z2g2P1A(%struct.A* + g2(&u->b); + // CHECK-NEW: call void @_Z9changeToAP1U(%union.U* + changeToA(u); + // CHECK-NEW: call i8* @llvm.invariant.group.barrier(i8* + // call void @_Z2g2P1A(%struct.A* %a) + g2(&u->a); + // CHECK-NEW-NOT: call i8* @llvm.invariant.group.barrier(i8* +} + +struct HoldingVirtuals { + A a; +}; + +struct Empty {}; +struct AnotherEmpty { + Empty e; +}; +union NoVptrs { + int a; + AnotherEmpty empty; +}; +void take(AnotherEmpty &); + +// CHECK-NEW-LABEL: noBarriers +void noBarriers(NoVptrs &noVptrs) { + // CHECK-NEW-NOT: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: 42 + noVptrs.a += 42; + // CHECK-NEW-NOT: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: call void @_Z4takeR12AnotherEmpty( + take(noVptrs.empty); +} + +union U2 { + HoldingVirtuals h; + int z; +}; +void take(HoldingVirtuals &); + +// CHECK-NEW-LABEL: define void @_Z15UnionsBarriers2R2U2 +void UnionsBarriers2(U2 &u) { + // CHECK-NEW-NOT: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: 42 + u.z += 42; + // CHECK-NEW: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: call void @_Z4takeR15HoldingVirtuals( + take(u.h); +} + +struct VirtualInBase : HoldingVirtuals, Empty { +}; + +struct VirtualInVBase : virtual Empty, virtual HoldingVirtuals { +}; + +// It has vtable by virtual inheritance. +struct VirtualInheritance : virtual Empty { +}; + +union U3 { + VirtualInBase v1; + VirtualInBase v2; + VirtualInheritance v3; + int z; +}; + +void take(VirtualInBase &); +void take(VirtualInVBase &); +void take(VirtualInheritance &); + +void UnionsBarrier3(U3 &u) { + // CHECK-NEW-NOT: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: 42 + u.z += 42; + // CHECK-NEW: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: call void @_Z4takeR13VirtualInBase( + take(u.v1); + // CHECK-NEW: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: call void @_Z4takeR13VirtualInBase( + take(u.v2); + + // CHECK-NEW: call i8* @llvm.invariant.group.barrier(i8* + // CHECK-NEW: call void @_Z4takeR18VirtualInheritance( + take(u.v3); +} /** DTORS **/ // CHECK-DTORS-LABEL: define linkonce_odr void @_ZN10StaticBaseD2Ev( diff --git a/test/CodeGenCXX/vtable-available-externally.cpp b/test/CodeGenCXX/vtable-available-externally.cpp index ae81959a75e9..db99f73d9e72 100644 --- a/test/CodeGenCXX/vtable-available-externally.cpp +++ b/test/CodeGenCXX/vtable-available-externally.cpp @@ -12,6 +12,7 @@ // RUN: FileCheck --check-prefix=CHECK-TEST14 %s < %t.opt // RUN: FileCheck --check-prefix=CHECK-TEST15 %s < %t.opt // RUN: FileCheck --check-prefix=CHECK-TEST16 %s < %t.opt +// RUN: FileCheck --check-prefix=CHECK-TEST17 %s < %t.opt #include <typeinfo> @@ -274,8 +275,8 @@ struct C { virtual D& operator=(const D&); }; -// Cannot emit B's vtable available_externally, because we cannot create -// a reference to the inline virtual B::operator= function. +// Cannot emit D's vtable available_externally, because we cannot create +// a reference to the inline virtual D::operator= function. // CHECK-TEST11: @_ZTVN6Test111DE = external unnamed_addr constant struct D : C { virtual void key(); @@ -391,3 +392,30 @@ void test() { } } +namespace Test17 { +// This test checks if we emit vtables opportunistically. +// CHECK-TEST17-DAG: @_ZTVN6Test171AE = available_externally +// CHECK-TEST17-DAG: @_ZTVN6Test171BE = external + +struct A { + virtual void key(); + virtual void bar() {} +}; + +// We won't gonna use deleting destructor for this type, which will disallow +// emitting vtable as available_externally +struct B { + virtual void key(); + virtual ~B() {} +}; + +void testcaseA() { + A a; + a.bar(); // this forces to emit definition of bar +} + +void testcaseB() { + B b; // This only forces emitting of complete object destructor +} + +} // namespace Test17 diff --git a/test/CodeGenCXX/vtable-linkage.cpp b/test/CodeGenCXX/vtable-linkage.cpp index c4ebf75b999a..0c55eb26f7f1 100644 --- a/test/CodeGenCXX/vtable-linkage.cpp +++ b/test/CodeGenCXX/vtable-linkage.cpp @@ -145,12 +145,14 @@ void use_F() { // F<int> is an explicit template instantiation declaration without a // key function, so its vtable should have external linkage. // CHECK-DAG: @_ZTV1FIiE = external unnamed_addr constant -// CHECK-OPT-DAG: @_ZTV1FIiE = external unnamed_addr constant +// CHECK-OPT-DAG: @_ZTV1FIiE = available_externally unnamed_addr constant // E<int> is an explicit template instantiation declaration. It has a // key function is not instantiated, so we know that vtable definition // will be generated in TU where key function will be defined -// so we can mark it as available_externally (only with optimizations) +// so we can mark it as external (without optimizations) and +// available_externally (with optimizations) because all of the inline +// virtual functions have been emitted. // CHECK-DAG: @_ZTV1EIiE = external unnamed_addr constant // CHECK-OPT-DAG: @_ZTV1EIiE = available_externally unnamed_addr constant diff --git a/test/CodeGenCoroutines/coro-await-domination.cpp b/test/CodeGenCoroutines/coro-await-domination.cpp new file mode 100644 index 000000000000..5df22374a6e7 --- /dev/null +++ b/test/CodeGenCoroutines/coro-await-domination.cpp @@ -0,0 +1,38 @@ +// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -emit-llvm %s -o - | FileCheck %s +#include "Inputs/coroutine.h" + +using namespace std::experimental; + +struct coro { + struct promise_type { + coro get_return_object(); + suspend_never initial_suspend(); + suspend_never final_suspend(); + void return_void(); + static void unhandled_exception(); + }; +}; + +struct A { + ~A(); + bool await_ready(); + int await_resume() { return 8; } + template <typename F> void await_suspend(F); +}; + +extern "C" void consume(int); + +// Verifies that domination is properly built during cleanup. +// Without CGCleanup.cpp fix verifier was reporting: +// Instruction does not dominate all uses! +// %tmp.exprcleanup = alloca i32*, align 8 +// store i32* %x, i32** %tmp.exprcleanup, align 8 + + +// CHECK-LABEL: f( +extern "C" coro f(int) { + int x = 42; + x = co_await A{}; + consume(x); +} + diff --git a/test/CodeGenObjC/parameterized_classes.m b/test/CodeGenObjC/parameterized_classes.m index b75cf2e3ad2d..8fe5c52b8d39 100644 --- a/test/CodeGenObjC/parameterized_classes.m +++ b/test/CodeGenObjC/parameterized_classes.m @@ -68,3 +68,31 @@ void blockTest(NSMutableArray<void (^)(void)> *array, NSString *name) { // CHECK: call i8* @objc_retainBlock // CHECK: ret void } + +// CHECK-LABEL: define internal void @"\01-[Derived setDest:] +// CHECK: %[[SELFADDR:.*]] = alloca %[[SELFTY:.*]]* +// CHECK: %[[AADDR:.*]] = alloca %[[IVARTY:.*]]* +// CHECK: %[[V2:.*]] = load %[[IVARTY]]*, %[[IVARTY]]** %[[AADDR]] +// CHECK: %[[V3:.*]] = load %[[SELFTY]]*, %[[SELFTY]]** %[[SELFADDR]] +// CHECK: %[[IVAR:.*]] = load i64, i64* @"OBJC_IVAR_$_Base._destination" +// CHECK: %[[V4:.*]] = bitcast %[[SELFTY]]* %[[V3]] to i8* +// CHECK: %[[ADDPTR:.*]] = getelementptr inbounds i8, i8* %[[V4]], i64 %[[IVAR]] +// CHECK: %[[V5:.*]] = bitcast i8* %[[ADDPTR]] to %[[IVARTY]]** +// CHECK: %[[V6:.*]] = bitcast %[[IVARTY]]** %[[V5]] to i8** +// CHECK: %[[V7:.*]] = bitcast %[[IVARTY]]* %[[V2]] to i8* +// CHECK: call void @objc_storeStrong(i8** %[[V6]], i8* %[[V7]]) + +@interface Base<DestType> : NSObject { + DestType _destination; +} +@end + +@interface Derived : Base<NSObject *> +- (void)setDest:(NSObject *)a; +@end + +@implementation Derived +- (void)setDest:(NSObject *)a { + _destination = a; +} +@end diff --git a/test/CodeGenOpenCL/bool_cast.cl b/test/CodeGenOpenCL/bool_cast.cl index ab40eccf571f..72926eb141ab 100644 --- a/test/CodeGenOpenCL/bool_cast.cl +++ b/test/CodeGenOpenCL/bool_cast.cl @@ -4,7 +4,7 @@ typedef unsigned char uchar4 __attribute((ext_vector_type(4))); typedef unsigned int int4 __attribute((ext_vector_type(4))); typedef float float4 __attribute((ext_vector_type(4))); -// CHECK-LABEL: define void @ker() +// CHECK-LABEL: define spir_kernel void @ker() void kernel ker() { bool t = true; int4 vec4 = (int4)t; diff --git a/test/CodeGenOpenCL/kernel-attributes.cl b/test/CodeGenOpenCL/kernel-attributes.cl index e61a75f7b537..eff2e57d8c9b 100644 --- a/test/CodeGenOpenCL/kernel-attributes.cl +++ b/test/CodeGenOpenCL/kernel-attributes.cl @@ -3,13 +3,13 @@ typedef unsigned int uint4 __attribute__((ext_vector_type(4))); kernel __attribute__((vec_type_hint(int))) __attribute__((reqd_work_group_size(1,2,4))) void kernel1(int a) {} -// CHECK: define void @kernel1(i32 {{[^%]*}}%a) {{[^{]+}} !vec_type_hint ![[MD1:[0-9]+]] !reqd_work_group_size ![[MD2:[0-9]+]] +// CHECK: define spir_kernel void @kernel1(i32 {{[^%]*}}%a) {{[^{]+}} !vec_type_hint ![[MD1:[0-9]+]] !reqd_work_group_size ![[MD2:[0-9]+]] kernel __attribute__((vec_type_hint(uint4))) __attribute__((work_group_size_hint(8,16,32))) void kernel2(int a) {} -// CHECK: define void @kernel2(i32 {{[^%]*}}%a) {{[^{]+}} !vec_type_hint ![[MD3:[0-9]+]] !work_group_size_hint ![[MD4:[0-9]+]] +// CHECK: define spir_kernel void @kernel2(i32 {{[^%]*}}%a) {{[^{]+}} !vec_type_hint ![[MD3:[0-9]+]] !work_group_size_hint ![[MD4:[0-9]+]] kernel __attribute__((intel_reqd_sub_group_size(8))) void kernel3(int a) {} -// CHECK: define void @kernel3(i32 {{[^%]*}}%a) {{[^{]+}} !intel_reqd_sub_group_size ![[MD5:[0-9]+]] +// CHECK: define spir_kernel void @kernel3(i32 {{[^%]*}}%a) {{[^{]+}} !intel_reqd_sub_group_size ![[MD5:[0-9]+]] // CHECK: [[MD1]] = !{i32 undef, i32 1} // CHECK: [[MD2]] = !{i32 1, i32 2, i32 4} diff --git a/test/CodeGenOpenCL/kernel-metadata.cl b/test/CodeGenOpenCL/kernel-metadata.cl index 4165f1fa0ce5..95be43015aa9 100644 --- a/test/CodeGenOpenCL/kernel-metadata.cl +++ b/test/CodeGenOpenCL/kernel-metadata.cl @@ -6,5 +6,5 @@ void normal_function() { __kernel void kernel_function() { } -// CHECK: define void @kernel_function() {{[^{]+}} !kernel_arg_addr_space ![[MD:[0-9]+]] !kernel_arg_access_qual ![[MD]] !kernel_arg_type ![[MD]] !kernel_arg_base_type ![[MD]] !kernel_arg_type_qual ![[MD]] { +// CHECK: define spir_kernel void @kernel_function() {{[^{]+}} !kernel_arg_addr_space ![[MD:[0-9]+]] !kernel_arg_access_qual ![[MD]] !kernel_arg_type ![[MD]] !kernel_arg_base_type ![[MD]] !kernel_arg_type_qual ![[MD]] { // CHECK: ![[MD]] = !{} diff --git a/test/CodeGenOpenCL/kernels-have-spir-cc-by-default.cl b/test/CodeGenOpenCL/kernels-have-spir-cc-by-default.cl new file mode 100644 index 000000000000..5bb52e9beb51 --- /dev/null +++ b/test/CodeGenOpenCL/kernels-have-spir-cc-by-default.cl @@ -0,0 +1,65 @@ +// RUN: %clang_cc1 %s -cl-std=CL1.2 -emit-llvm -triple x86_64-unknown-unknown -o - | FileCheck %s +// RUN: %clang_cc1 %s -cl-std=CL1.2 -emit-llvm -triple amdgcn-unknown-unknown -o - | FileCheck -check-prefixes=AMDGCN %s +// Test that the kernels always use the SPIR calling convention +// to have unambiguous mapping of arguments to feasibly implement +// clSetKernelArg(). + +typedef struct int_single { + int a; +} int_single; + +typedef struct int_pair { + long a; + long b; +} int_pair; + +typedef struct test_struct { + int elementA; + int elementB; + long elementC; + char elementD; + long elementE; + float elementF; + short elementG; + double elementH; +} test_struct; + +kernel void test_single(int_single input, global int* output) { +// CHECK: spir_kernel +// AMDGCN: define amdgpu_kernel void @test_single +// CHECK: struct.int_single* byval nocapture +// CHECK: i32* nocapture %output + output[0] = input.a; +} + +kernel void test_pair(int_pair input, global int* output) { +// CHECK: spir_kernel +// AMDGCN: define amdgpu_kernel void @test_pair +// CHECK: struct.int_pair* byval nocapture +// CHECK: i32* nocapture %output + output[0] = (int)input.a; + output[1] = (int)input.b; +} + +kernel void test_kernel(test_struct input, global int* output) { +// CHECK: spir_kernel +// AMDGCN: define amdgpu_kernel void @test_kernel +// CHECK: struct.test_struct* byval nocapture +// CHECK: i32* nocapture %output + output[0] = input.elementA; + output[1] = input.elementB; + output[2] = (int)input.elementC; + output[3] = (int)input.elementD; + output[4] = (int)input.elementE; + output[5] = (int)input.elementF; + output[6] = (int)input.elementG; + output[7] = (int)input.elementH; +}; + +void test_function(int_pair input, global int* output) { +// CHECK-NOT: spir_kernel +// AMDGCN-NOT: define amdgpu_kernel void @test_function +// CHECK: i64 %input.coerce0, i64 %input.coerce1, i32* nocapture %output + output[0] = (int)input.a; + output[1] = (int)input.b; +} diff --git a/test/CodeGenOpenCL/pipe_types.cl b/test/CodeGenOpenCL/pipe_types.cl index f5b42e2a490f..7c11f74ad7b9 100644 --- a/test/CodeGenOpenCL/pipe_types.cl +++ b/test/CodeGenOpenCL/pipe_types.cl @@ -28,7 +28,7 @@ void test5(read_only pipe int4 p) { typedef read_only pipe int MyPipe; kernel void test6(MyPipe p) { -// CHECK: define void @test6(%opencl.pipe_t* %p) +// CHECK: define spir_kernel void @test6(%opencl.pipe_t* %p) } struct Person { diff --git a/test/CodeGenOpenCL/ptx-calls.cl b/test/CodeGenOpenCL/ptx-calls.cl index bde00bc3d73a..2a3400371edf 100644 --- a/test/CodeGenOpenCL/ptx-calls.cl +++ b/test/CodeGenOpenCL/ptx-calls.cl @@ -7,7 +7,7 @@ void device_function() { __kernel void kernel_function() { device_function(); } -// CHECK-LABEL: define void @kernel_function() +// CHECK-LABEL: define spir_kernel void @kernel_function() // CHECK: call void @device_function() // CHECK: !{{[0-9]+}} = !{void ()* @kernel_function, !"kernel", i32 1} diff --git a/test/CodeGenOpenCL/ptx-kernels.cl b/test/CodeGenOpenCL/ptx-kernels.cl index fc6de4f3d517..b9e1c224c7ce 100644 --- a/test/CodeGenOpenCL/ptx-kernels.cl +++ b/test/CodeGenOpenCL/ptx-kernels.cl @@ -6,6 +6,6 @@ void device_function() { __kernel void kernel_function() { } -// CHECK-LABEL: define void @kernel_function() +// CHECK-LABEL: define spir_kernel void @kernel_function() // CHECK: !{{[0-9]+}} = !{void ()* @kernel_function, !"kernel", i32 1} diff --git a/test/Driver/arm-cortex-cpus.c b/test/Driver/arm-cortex-cpus.c index c52e643f0337..80912678e029 100644 --- a/test/Driver/arm-cortex-cpus.c +++ b/test/Driver/arm-cortex-cpus.c @@ -120,11 +120,11 @@ // RUN: %clang -target armv7a-linux-gnueabi -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V7A %s // RUN: %clang -target arm-linux-gnueabi -march=armv7-a -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V7A %s -// CHECK-V7A: "-cc1"{{.*}} "-triple" "armv7-{{.*}} "-target-cpu" "cortex-a8" +// CHECK-V7A: "-cc1"{{.*}} "-triple" "armv7-{{.*}} "-target-cpu" "generic" // RUN: %clang -target armv7a-linux-gnueabi -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V7A-THUMB %s // RUN: %clang -target arm-linux-gnueabi -march=armv7-a -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V7A-THUMB %s -// CHECK-V7A-THUMB: "-cc1"{{.*}} "-triple" "thumbv7-{{.*}} "-target-cpu" "cortex-a8" +// CHECK-V7A-THUMB: "-cc1"{{.*}} "-triple" "thumbv7-{{.*}} "-target-cpu" "generic" // RUN: %clang -target armv7r-linux-gnueabi -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V7R %s // RUN: %clang -target arm-linux-gnueabi -march=armv7-r -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V7R %s @@ -144,7 +144,7 @@ // RUN: %clang -target armv8a -mlittle-endian -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A %s // RUN: %clang -target arm -march=armv8a -mlittle-endian -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A %s // RUN: %clang -target arm -mlittle-endian -march=armv8-a -mlittle-endian -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A %s -// CHECK-V8A: "-cc1"{{.*}} "-triple" "armv8-{{.*}}" "-target-cpu" "cortex-a53" +// CHECK-V8A: "-cc1"{{.*}} "-triple" "armv8-{{.*}}" "-target-cpu" "generic" // RUN: %clang -target armv8r-linux-gnueabi -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8R %s // RUN: %clang -target arm -march=armv8r -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8R %s @@ -189,7 +189,7 @@ // RUN: %clang -target armv8a -mbig-endian -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A %s // RUN: %clang -target arm -march=armebv8a -mbig-endian -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A %s // RUN: %clang -target arm -march=armebv8-a -mbig-endian -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A %s -// CHECK-BE-V8A: "-cc1"{{.*}} "-triple" "armebv8-{{.*}}" "-target-cpu" "cortex-a53" +// CHECK-BE-V8A: "-cc1"{{.*}} "-triple" "armebv8-{{.*}}" "-target-cpu" "generic" // RUN: %clang -target armv8 -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A-THUMB %s // RUN: %clang -target arm -march=armv8 -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A-THUMB %s @@ -199,7 +199,7 @@ // RUN: %clang -target arm -march=armv8 -mlittle-endian -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A-THUMB %s // RUN: %clang -target armv8a -mlittle-endian -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A-THUMB %s // RUN: %clang -target arm -march=armv8a -mlittle-endian -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V8A-THUMB %s -// CHECK-V8A-THUMB: "-cc1"{{.*}} "-triple" "thumbv8-{{.*}}" "-target-cpu" "cortex-a53" +// CHECK-V8A-THUMB: "-cc1"{{.*}} "-triple" "thumbv8-{{.*}}" "-target-cpu" "generic" // RUN: %clang -target armebv8 -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A-THUMB %s // RUN: %clang -target armeb -march=armebv8 -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A-THUMB %s @@ -209,7 +209,7 @@ // RUN: %clang -target arm -march=armebv8 -mbig-endian -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A-THUMB %s // RUN: %clang -target armv8a -mbig-endian -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A-THUMB %s // RUN: %clang -target arm -march=armebv8a -mbig-endian -mthumb -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-BE-V8A-THUMB %s -// CHECK-BE-V8A-THUMB: "-cc1"{{.*}} "-triple" "thumbebv8-{{.*}}" "-target-cpu" "cortex-a53" +// CHECK-BE-V8A-THUMB: "-cc1"{{.*}} "-triple" "thumbebv8-{{.*}}" "-target-cpu" "generic" // RUN: %clang -target arm -march=armv8.1a -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V81A %s // RUN: %clang -target armv8.1a -### -c %s 2>&1 | FileCheck -check-prefix=CHECK-V81A %s diff --git a/test/Driver/cl-cc-flags.c b/test/Driver/cl-cc-flags.c new file mode 100644 index 000000000000..76f116e199e2 --- /dev/null +++ b/test/Driver/cl-cc-flags.c @@ -0,0 +1,33 @@ +// Note: %s must be preceded by --, otherwise it may be interpreted as a +// command-line option, e.g. on Mac where %s is commonly under /Users. + +// RUN: %clang_cl --target=i686-windows-msvc /Gd -### -- %s 2>&1 | FileCheck --check-prefix=CDECL %s +// CDECL: -fdefault-calling-conv=cdecl + +// RUN: %clang_cl --target=i686-windows-msvc /Gr -### -- %s 2>&1 | FileCheck --check-prefix=FASTCALL %s +// FASTCALL: -fdefault-calling-conv=fastcall + +// RUN: %clang_cl --target=i686-windows-msvc /Gz -### -- %s 2>&1 | FileCheck --check-prefix=STDCALL %s +// STDCALL: -fdefault-calling-conv=stdcall + +// RUN: %clang_cl --target=i686-windows-msvc /Gv -### -- %s 2>&1 | FileCheck --check-prefix=VECTORCALL %s +// VECTORCALL: -fdefault-calling-conv=vectorcall + +// Last one should win: + +// RUN: %clang_cl --target=i686-windows-msvc /Gd /Gv -### -- %s 2>&1 | FileCheck --check-prefix=LASTWINS_VECTOR %s +// LASTWINS_VECTOR: -fdefault-calling-conv=vectorcall + +// RUN: %clang_cl --target=i686-windows-msvc /Gv /Gd -### -- %s 2>&1 | FileCheck --check-prefix=LASTWINS_CDECL %s +// LASTWINS_CDECL: -fdefault-calling-conv=cdecl + +// No fastcall or stdcall on x86_64: + +// RUN: %clang_cl -Wno-msvc-not-found --target=x86_64-windows-msvc /Gr -### -- %s 2>&1 | FileCheck --check-prefix=UNSUPPORTED %s +// RUN: %clang_cl -Wno-msvc-not-found --target=x86_64-windows-msvc /Gz -### -- %s 2>&1 | FileCheck --check-prefix=UNSUPPORTED %s +// RUN: %clang_cl -Wno-msvc-not-found --target=thumbv7-windows-msvc /Gv -### -- %s 2>&1 | FileCheck --check-prefix=UNSUPPORTED %s + +// UNSUPPORTED-NOT: error: +// UNSUPPORTED-NOT: warning: +// UNSUPPORTED-NOT: -fdefault-calling-conv= + diff --git a/test/Driver/cl-diagnostics.c b/test/Driver/cl-diagnostics.c new file mode 100644 index 000000000000..975a8cf8bb54 --- /dev/null +++ b/test/Driver/cl-diagnostics.c @@ -0,0 +1,28 @@ +// Note: %s must be preceded by --, otherwise it may be interpreted as a +// command-line option, e.g. on Mac where %s is commonly under /Users. + +// RUN: %clang_cl /diagnostics:classic -### -- %s 2>&1 | FileCheck %s --check-prefix=CLASSIC +// CLASSIC: -fno-caret-diagnostics +// CLASSIC: -fno-show-column + +// RUN: %clang_cl /diagnostics:column -### -- %s 2>&1 | FileCheck %s --check-prefix=COLUMN +// COLUMN: -fno-caret-diagnostics +// COLUMN-NOT: -fno-show-column + +// RUN: %clang_cl /diagnostics:caret -### -- %s 2>&1 | FileCheck %s --check-prefix=CARET +// CARET-NOT: -fno-caret-diagnostics +// CARET-NOT: -fno-show-column + +// RUN: not %clang_cl -fms-compatibility-version=19 /diagnostics:classic /Zs -c -- %s 2>&1 | FileCheck %s --check-prefix=OUTPUT_CLASSIC + +// OUTPUT_CLASSIC: cl-diagnostics.c({{[0-9]+}}): error: "asdf" +// OUTPUT_CLASSIC-NOT: #error + +// RUN: not %clang_cl -fms-compatibility-version=19 /diagnostics:caret /Zs -c -- %s 2>&1 | FileCheck %s --check-prefix=OUTPUT_CARET + +// OUTPUT_CARET: cl-diagnostics.c({{[0-9]+,[0-9]+}}): error: "asdf" +// OUTPUT_CARET-NEXT: #error "asdf" +// OUTPUT_CARET-NEXT: ^ + + +#error "asdf" diff --git a/test/Driver/cl-include.c b/test/Driver/cl-include.c new file mode 100644 index 000000000000..d3dc006e575f --- /dev/null +++ b/test/Driver/cl-include.c @@ -0,0 +1,14 @@ +// Note: %s must be preceded by --, otherwise it may be interpreted as a +// command-line option, e.g. on Mac where %s is commonly under /Users. + +// RUN: %clang_cl -### -- %s 2>&1 | FileCheck %s --check-prefix=BUILTIN +// BUILTIN: "-internal-isystem" "{{.*lib.*clang.*include}}" + +// RUN: %clang_cl -nobuiltininc -### -- %s 2>&1 | FileCheck %s --check-prefix=NOBUILTIN +// NOBUILTIN-NOT: "-internal-isystem" "{{.*lib.*clang.*include}}" + +// RUN: env INCLUDE=/my/system/inc %clang_cl -### -- %s 2>&1 | FileCheck %s --check-prefix=STDINC +// STDINC: "-internal-isystem" "/my/system/inc" + +// RUN: env INCLUDE=/my/system/inc %clang_cl -nostdinc -### -- %s 2>&1 | FileCheck %s --check-prefix=NOSTDINC +// NOSTDINC-NOT: "-internal-isystem" "/my/system/inc" diff --git a/test/Driver/cl-zc.cpp b/test/Driver/cl-zc.cpp index 26496293201e..4414eb6ebba2 100644 --- a/test/Driver/cl-zc.cpp +++ b/test/Driver/cl-zc.cpp @@ -53,6 +53,11 @@ // RUN: %clang_cl /c -### /Zc:inline- -- %s 2>&1 | FileCheck -check-prefix=INLINE-OFF %s // INLINE-OFF: argument unused during compilation +// RUN: %clang_cl /c -### /Zc:ternary -- %s 2>&1 | FileCheck -check-prefix=TERNARY-ON %s +// TERNARY-ON-NOT: argument unused during compilation +// RUN: %clang_cl /c -### /Zc:ternary- -- %s 2>&1 | FileCheck -check-prefix=TERNARY-OFF %s +// TERNARY-OFF: argument unused during compilation + // These never warn, but don't have an effect yet. diff --git a/test/Driver/fsanitize.c b/test/Driver/fsanitize.c index 41f573aa7316..786f517f5371 100644 --- a/test/Driver/fsanitize.c +++ b/test/Driver/fsanitize.c @@ -3,18 +3,18 @@ // RUN: %clang -target x86_64-linux-gnu -fsanitize=undefined -fsanitize-undefined-trap-on-error %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-TRAP // RUN: %clang -target x86_64-linux-gnu -fsanitize=undefined-trap -fsanitize-undefined-trap-on-error %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-TRAP // RUN: %clang -target x86_64-linux-gnu -fsanitize-undefined-trap-on-error -fsanitize=undefined-trap %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-TRAP -// CHECK-UNDEFINED-TRAP: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute|function),?){18}"}} -// CHECK-UNDEFINED-TRAP: "-fsanitize-trap=alignment,array-bounds,bool,enum,float-cast-overflow,float-divide-by-zero,function,integer-divide-by-zero,nonnull-attribute,null,object-size,return,returns-nonnull-attribute,shift-base,shift-exponent,signed-integer-overflow,unreachable,vla-bound" -// CHECK-UNDEFINED-TRAP2: "-fsanitize-trap=alignment,array-bounds,bool,enum,float-cast-overflow,float-divide-by-zero,function,integer-divide-by-zero,nonnull-attribute,null,object-size,return,returns-nonnull-attribute,shift-base,shift-exponent,unreachable,vla-bound" +// CHECK-UNDEFINED-TRAP: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|pointer-overflow|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute|function),?){19}"}} +// CHECK-UNDEFINED-TRAP: "-fsanitize-trap=alignment,array-bounds,bool,enum,float-cast-overflow,float-divide-by-zero,function,integer-divide-by-zero,nonnull-attribute,null,object-size,pointer-overflow,return,returns-nonnull-attribute,shift-base,shift-exponent,signed-integer-overflow,unreachable,vla-bound" +// CHECK-UNDEFINED-TRAP2: "-fsanitize-trap=alignment,array-bounds,bool,enum,float-cast-overflow,float-divide-by-zero,function,integer-divide-by-zero,nonnull-attribute,null,object-size,pointer-overflow,return,returns-nonnull-attribute,shift-base,shift-exponent,unreachable,vla-bound" // RUN: %clang -target x86_64-linux-gnu -fsanitize=undefined %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED -// CHECK-UNDEFINED: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|function|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|vptr|object-size|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){19}"}} +// CHECK-UNDEFINED: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|function|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|vptr|object-size|pointer-overflow|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){20}"}} // RUN: %clang -target x86_64-apple-darwin10 -fsanitize=undefined %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-DARWIN -// CHECK-UNDEFINED-DARWIN: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){17}"}} +// CHECK-UNDEFINED-DARWIN: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|pointer-overflow|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){18}"}} // RUN: %clang -target i386-unknown-openbsd -fsanitize=undefined %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-OPENBSD -// CHECK-UNDEFINED-OPENBSD: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){17}"}} +// CHECK-UNDEFINED-OPENBSD: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|pointer-overflow|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){18}"}} // RUN: %clang -target i386-pc-win32 -fsanitize=undefined %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-WIN --check-prefix=CHECK-UNDEFINED-WIN32 // RUN: %clang -target i386-pc-win32 -fsanitize=undefined -x c++ %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-UNDEFINED-WIN --check-prefix=CHECK-UNDEFINED-WIN32 --check-prefix=CHECK-UNDEFINED-WIN-CXX @@ -23,7 +23,7 @@ // CHECK-UNDEFINED-WIN32: "--dependent-lib={{[^"]*}}ubsan_standalone-i386.lib" // CHECK-UNDEFINED-WIN64: "--dependent-lib={{[^"]*}}ubsan_standalone-x86_64.lib" // CHECK-UNDEFINED-WIN-CXX: "--dependent-lib={{[^"]*}}ubsan_standalone_cxx{{[^"]*}}.lib" -// CHECK-UNDEFINED-WIN-SAME: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){17}"}} +// CHECK-UNDEFINED-WIN-SAME: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|pointer-overflow|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){18}"}} // RUN: %clang -target i386-pc-win32 -fsanitize-coverage=bb %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-COVERAGE-WIN32 // CHECK-COVERAGE-WIN32: "--dependent-lib={{[^"]*}}ubsan_standalone-i386.lib" @@ -43,7 +43,7 @@ // CHECK-FNO-SANITIZE-ALL: "-fsanitize=thread" // RUN: %clang -target x86_64-linux-gnu -fsanitize=thread,undefined -fno-sanitize=thread -fno-sanitize=float-cast-overflow,vptr,bool,enum %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-PARTIAL-UNDEFINED -// CHECK-PARTIAL-UNDEFINED: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|function|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|array-bounds|returns-nonnull-attribute|nonnull-attribute),?){15}"}} +// CHECK-PARTIAL-UNDEFINED: "-fsanitize={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|function|shift-base|shift-exponent|unreachable|return|vla-bound|alignment|null|object-size|pointer-overflow|array-bounds|returns-nonnull-attribute|nonnull-attribute),?){16}"}} // RUN: %clang -target x86_64-linux-gnu -fsanitize=shift -fno-sanitize=shift-base %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-FSANITIZE-SHIFT-PARTIAL // CHECK-FSANITIZE-SHIFT-PARTIAL: "-fsanitize=shift-exponent" @@ -217,7 +217,7 @@ // RUN: %clang -target x86_64-linux-gnu %s -fsanitize=undefined -fno-sanitize-recover=undefined -### 2>&1 | FileCheck %s --check-prefix=CHECK-NO-RECOVER-UBSAN // RUN: %clang -target x86_64-linux-gnu %s -fsanitize=undefined -fno-sanitize-recover=all -fsanitize-recover=thread -### 2>&1 | FileCheck %s --check-prefix=CHECK-NO-RECOVER-UBSAN // RUN: %clang -target x86_64-linux-gnu %s -fsanitize=undefined -fsanitize-recover=all -fno-sanitize-recover=undefined -### 2>&1 | FileCheck %s --check-prefix=CHECK-NO-RECOVER-UBSAN -// CHECK-RECOVER-UBSAN: "-fsanitize-recover={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|function|shift-base|shift-exponent|vla-bound|alignment|null|vptr|object-size|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){17}"}} +// CHECK-RECOVER-UBSAN: "-fsanitize-recover={{((signed-integer-overflow|integer-divide-by-zero|float-divide-by-zero|function|shift-base|shift-exponent|vla-bound|alignment|null|vptr|object-size|pointer-overflow|float-cast-overflow|array-bounds|enum|bool|returns-nonnull-attribute|nonnull-attribute),?){18}"}} // CHECK-NO-RECOVER-UBSAN-NOT: sanitize-recover // RUN: %clang -target x86_64-linux-gnu %s -fsanitize=undefined -fno-sanitize-recover=all -fsanitize-recover=object-size,shift-base -### 2>&1 | FileCheck %s --check-prefix=CHECK-PARTIAL-RECOVER diff --git a/test/Driver/gold-lto.c b/test/Driver/gold-lto.c index 050b1ef18be1..b8eca5112b5a 100644 --- a/test/Driver/gold-lto.c +++ b/test/Driver/gold-lto.c @@ -19,7 +19,7 @@ // RUN: -march=armv7a -Wl,-plugin-opt=foo -O0 \ // RUN: | FileCheck %s --check-prefix=CHECK-ARM-V7A // CHECK-ARM-V7A: "-plugin" "{{.*}}/LLVMgold.so" -// CHECK-ARM-V7A: "-plugin-opt=mcpu=cortex-a8" +// CHECK-ARM-V7A: "-plugin-opt=mcpu=generic" // CHECK-ARM-V7A: "-plugin-opt=O0" // CHECK-ARM-V7A: "-plugin-opt=foo" // diff --git a/test/Driver/nacl-direct.c b/test/Driver/nacl-direct.c index f71e14f996ee..3c8e33b180b0 100644 --- a/test/Driver/nacl-direct.c +++ b/test/Driver/nacl-direct.c @@ -46,7 +46,7 @@ // RUN: | FileCheck --check-prefix=CHECK-ARM %s // CHECK-ARM: {{.*}}clang{{.*}}" "-cc1" // CHECK-ARM: "-fuse-init-array" -// CHECK-ARM: "-target-cpu" "cortex-a8" +// CHECK-ARM: "-target-cpu" "generic" // CHECK-ARM: "-target-abi" "aapcs-linux" // CHECK-ARM: "-mfloat-abi" "hard" // CHECK-ARM: "-resource-dir" "foo" diff --git a/test/Driver/openmp-offload.c b/test/Driver/openmp-offload.c index 097dc7d17faa..39eb41e6ac08 100644 --- a/test/Driver/openmp-offload.c +++ b/test/Driver/openmp-offload.c @@ -210,14 +210,16 @@ // CHK-LKS-ST: INPUT([[T2BIN:.+\.out-openmp-x86_64-pc-linux-gnu]]) // CHK-LKS: SECTIONS // CHK-LKS: { -// CHK-LKS: .omp_offloading : +// CHK-LKS: .omp_offloading.powerpc64le-ibm-linux-gnu : // CHK-LKS: ALIGN(0x10) // CHK-LKS: { -// CHK-LKS: . = ALIGN(0x10); // CHK-LKS: PROVIDE_HIDDEN(.omp_offloading.img_start.powerpc64le-ibm-linux-gnu = .); // CHK-LKS: [[T1BIN]] // CHK-LKS: PROVIDE_HIDDEN(.omp_offloading.img_end.powerpc64le-ibm-linux-gnu = .); -// CHK-LKS: . = ALIGN(0x10); +// CHK-LKS: } +// CHK-LKS: .omp_offloading.x86_64-pc-linux-gnu : +// CHK-LKS: ALIGN(0x10) +// CHK-LKS: { // CHK-LKS: PROVIDE_HIDDEN(.omp_offloading.img_start.x86_64-pc-linux-gnu = .); // CHK-LKS: [[T2BIN]] // CHK-LKS: PROVIDE_HIDDEN(.omp_offloading.img_end.x86_64-pc-linux-gnu = .); diff --git a/test/Misc/diag-mapping2.c b/test/Misc/diag-mapping2.c index 672d0548ad82..c3afea94b370 100644 --- a/test/Misc/diag-mapping2.c +++ b/test/Misc/diag-mapping2.c @@ -4,6 +4,7 @@ // This should not emit anything. // RUN: %clang_cc1 %s -w 2>&1 | not grep diagnostic // RUN: %clang_cc1 %s -Wno-#warnings 2>&1 | not grep diagnostic +// RUN: %clang_cc1 %s -Wno-cpp 2>&1 | not grep diagnostic // -Werror can map all warnings to error. // RUN: not %clang_cc1 %s -Werror 2>&1 | grep "error: foo" diff --git a/test/Modules/odr_hash.cpp b/test/Modules/odr_hash.cpp index 947583bcfd21..a6a0b74743aa 100644 --- a/test/Modules/odr_hash.cpp +++ b/test/Modules/odr_hash.cpp @@ -866,6 +866,40 @@ S9 s9; #endif } +namespace TemplateSpecializationType { +#if defined(FIRST) +template <class T1> struct U1 {}; +struct S1 { + U1<int> u; +}; +#elif defined(SECOND) +template <class T1, class T2> struct U1 {}; +struct S1 { + U1<int, int> u; +}; +#else +S1 s1; +// expected-error@first.h:* {{'TemplateSpecializationType::S1::u' from module 'FirstModule' is not present in definition of 'TemplateSpecializationType::S1' in module 'SecondModule'}} +// expected-note@second.h:* {{declaration of 'u' does not match}} +#endif + +#if defined(FIRST) +template <class T1> struct U2 {}; +struct S2 { + U2<int> u; +}; +#elif defined(SECOND) +template <class T1> struct V1 {}; +struct S2 { + V1<int> u; +}; +#else +S2 s2; +// expected-error@first.h:* {{'TemplateSpecializationType::S2::u' from module 'FirstModule' is not present in definition of 'TemplateSpecializationType::S2' in module 'SecondModule'}} +// expected-note@second.h:* {{declaration of 'u' does not match}} +#endif +} + // Interesting cases that should not cause errors. struct S should not error // while struct T should error at the access specifier mismatch at the end. namespace AllDecls { diff --git a/test/Modules/preprocess-module.cpp b/test/Modules/preprocess-module.cpp index eaab3136938f..9d1a2bb6801f 100644 --- a/test/Modules/preprocess-module.cpp +++ b/test/Modules/preprocess-module.cpp @@ -14,8 +14,6 @@ // RUN: FileCheck %s --input-file %t/rewrite.ii --check-prefix=CHECK --check-prefix=REWRITE // Check that we can build a module from the preprocessed output. -// FIXME: For now, we need the headers to exist. -// RUN: touch %t/file.h %t/file2.h // RUN: %clang_cc1 -fmodules -fmodule-name=file -fmodule-file=%t/fwd.pcm -x c++-module-map-cpp-output %t/no-rewrite.ii -emit-module -o %t/no-rewrite.pcm // RUN: %clang_cc1 -fmodules -fmodule-name=file -fmodule-file=%t/fwd.pcm -x c++-module-map-cpp-output %t/rewrite.ii -emit-module -o %t/rewrite.pcm @@ -27,6 +25,8 @@ // Check the module we built works. // RUN: %clang_cc1 -fmodules -fmodule-file=%t/no-rewrite.pcm %s -I%t -verify -fno-modules-error-recovery // RUN: %clang_cc1 -fmodules -fmodule-file=%t/rewrite.pcm %s -I%t -verify -fno-modules-error-recovery -DREWRITE +// RUN: %clang_cc1 -fmodules -fmodule-file=%t/no-rewrite.pcm %s -I%t -verify -fno-modules-error-recovery -DINCLUDE -I%S/Inputs/preprocess +// RUN: %clang_cc1 -fmodules -fmodule-file=%t/rewrite.pcm %s -I%t -verify -fno-modules-error-recovery -DREWRITE -DINCLUDE -I%S/Inputs/preprocess // == module map @@ -102,7 +102,11 @@ __FILE *a; // expected-error {{declaration of '__FILE' must be imported}} // expected-note@no-rewrite.ii:1 {{here}} #endif +#ifdef INCLUDE +#include "file.h" +#else #pragma clang module import file +#endif FILE *b; -int x = file2; +int x = file2; // ok, found in file2.h, even under -DINCLUDE diff --git a/test/Modules/preprocess-nested.cpp b/test/Modules/preprocess-nested.cpp index 8fccf137e94f..f26b65507ece 100644 --- a/test/Modules/preprocess-nested.cpp +++ b/test/Modules/preprocess-nested.cpp @@ -8,8 +8,6 @@ // RUN: FileCheck %s --input-file %t/rewrite.ii --check-prefix=CHECK --check-prefix=REWRITE // Check that we can build a module from the preprocessed output. -// FIXME: For now, the files need to exist. -// RUN: touch %t/a.h %t/b.h %t/c.h // RUN: %clang_cc1 -fmodules -fmodules-local-submodule-visibility -fmodule-name=nested -x c++-module-map-cpp-output %t/no-rewrite.ii -emit-module -o %t/no-rewrite.pcm // RUN: %clang_cc1 -fmodules -fmodules-local-submodule-visibility -fmodule-name=nested -x c++-module-map-cpp-output %t/rewrite.ii -emit-module -o %t/rewrite.pcm diff --git a/test/Modules/preprocess-unavailable.cpp b/test/Modules/preprocess-unavailable.cpp index e568cd7b5251..cfd18253daf7 100644 --- a/test/Modules/preprocess-unavailable.cpp +++ b/test/Modules/preprocess-unavailable.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -x c++-module-map %s -fmodule-name=a -verify +// RUN: %clang_cc1 -x c++-module-map %s -fmodule-name=a -verify -std=c++98 module a { module b { requires cplusplus11 diff --git a/test/SemaCXX/attr-require-constant-initialization.cpp b/test/SemaCXX/attr-require-constant-initialization.cpp index 3ed51071cd13..0df9f2e88029 100644 --- a/test/SemaCXX/attr-require-constant-initialization.cpp +++ b/test/SemaCXX/attr-require-constant-initialization.cpp @@ -7,9 +7,9 @@ #define ATTR __attribute__((require_constant_initialization)) // expected-note 0+ {{expanded from macro}} -int ReturnInt(); +int ReturnInt(); // expected-note 0+ {{declared here}} -struct PODType { +struct PODType { // expected-note 0+ {{declared here}} int value; int value2; }; @@ -20,20 +20,20 @@ struct PODType { struct LitType { constexpr LitType() : value(0) {} constexpr LitType(int x) : value(x) {} - LitType(void *) : value(-1) {} + LitType(void *) : value(-1) {} // expected-note 0+ {{declared here}} int value; }; #endif -struct NonLit { +struct NonLit { // expected-note 0+ {{declared here}} #if __cplusplus >= 201402L constexpr NonLit() : value(0) {} constexpr NonLit(int x) : value(x) {} #else - NonLit() : value(0) {} + NonLit() : value(0) {} // expected-note 0+ {{declared here}} NonLit(int x) : value(x) {} #endif - NonLit(void *) : value(-1) {} + NonLit(void *) : value(-1) {} // expected-note 0+ {{declared here}} ~NonLit() {} int value; }; @@ -43,7 +43,7 @@ struct StoresNonLit { constexpr StoresNonLit() : obj() {} constexpr StoresNonLit(int x) : obj(x) {} #else - StoresNonLit() : obj() {} + StoresNonLit() : obj() {} // expected-note 0+ {{declared here}} StoresNonLit(int x) : obj(x) {} #endif StoresNonLit(void *p) : obj(p) {} @@ -82,6 +82,12 @@ void test_basic_start_static_2_1() { const int non_global = 42; ATTR static const int &local_init = non_global; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +#if __cplusplus >= 201103L + // expected-note@-3 {{reference to 'non_global' is not a constant expression}} + // expected-note@-5 {{declared here}} +#else + // expected-note@-6 {{subexpression not valid in a constant expression}} +#endif ATTR static const int &global_init = glvalue_int; ATTR static const int &temp_init = 42; } @@ -89,8 +95,18 @@ void test_basic_start_static_2_1() { ATTR const int &temp_ref = 42; ATTR const int &temp_ref2 = ReturnInt(); // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +#if __cplusplus >= 201103L +// expected-note@-3 {{non-constexpr function 'ReturnInt' cannot be used in a constant expression}} +#else +// expected-note@-5 {{subexpression not valid in a constant expression}} +#endif ATTR const NonLit &nl_temp_ref = 42; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +#if __cplusplus >= 201103L +// expected-note@-3 {{non-literal type 'const NonLit' cannot be used in a constant expression}} +#else +// expected-note@-5 {{subexpression not valid in a constant expression}} +#endif #if __cplusplus >= 201103L ATTR const LitType &lit_temp_ref = 42; @@ -99,6 +115,11 @@ ATTR const int &subobj_ref = LitType{}.value; ATTR const int &nl_subobj_ref = NonLit().value; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +#if __cplusplus >= 201103L +// expected-note-re@-3 {{non-literal type '{{.*}}' cannot be used in a constant expression}} +#else +// expected-note@-5 {{subexpression not valid in a constant expression}} +#endif struct TT1 { ATTR static const int &no_init; @@ -116,6 +137,8 @@ const int &TT1::subobj_init = PODType().value; #if __cplusplus >= 201103L thread_local const int &TT1::tl_glvalue_init = glvalue_int; thread_local const int &TT1::tl_temp_init = 42; // expected-error {{variable does not have a constant initializer}} +// expected-note@-1 {{reference to temporary is not a constant expression}} +// expected-note@-2 {{temporary created here}} #endif // [basic.start.static]p2.2 @@ -129,17 +152,25 @@ void test_basic_start_static_2_2() { #else ATTR static PODType pod; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'PODType' cannot be used in a constant expression}} #endif ATTR static PODType pot2 = {ReturnInt()}; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +#if __cplusplus >= 201103L +// expected-note@-3 {{non-constexpr function 'ReturnInt' cannot be used in a constant expression}} +#else +// expected-note@-5 {{subexpression not valid in a constant expression}} +#endif #if __cplusplus >= 201103L constexpr LitType l; ATTR static LitType static_lit = l; ATTR static LitType static_lit2 = (void *)0; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} + // expected-note@-2 {{non-constexpr constructor 'LitType' cannot be used in a constant expression}} ATTR static LitType static_lit3 = ReturnInt(); // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} + // expected-note@-2 {{non-constexpr function 'ReturnInt' cannot be used in a constant expression}} ATTR thread_local LitType tls = 42; #endif } @@ -157,15 +188,23 @@ struct TT2 { ATTR static const NonLit non_lit_copy_init; // expected-note {{required by 'require_constant_initialization' attribute here}} #endif }; -PODType TT2::pod_noinit; +PODType TT2::pod_noinit; // expected-note 0+ {{declared here}} #if __cplusplus >= 201103L // expected-error@-2 {{variable does not have a constant initializer}} +// expected-note@-3 {{non-constexpr constructor 'PODType' cannot be used in a constant expression}} #endif PODType TT2::pod_copy_init(TT2::pod_noinit); // expected-error {{variable does not have a constant initializer}} +#if __cplusplus >= 201103L +// expected-note@-2 {{read of non-constexpr variable 'pod_noinit' is not allowed in a constant expression}} +// expected-note@-3 {{in call to 'PODType(pod_noinit)'}} +#else +// expected-note@-5 {{subexpression not valid in a constant expression}} +#endif #if __cplusplus >= 201402L const NonLit TT2::non_lit(42); const NonLit TT2::non_lit_list_init = {42}; const NonLit TT2::non_lit_copy_init = 42; // expected-error {{variable does not have a constant initializer}} +// expected-note@-1 {{subexpression not valid in a constant expression}} #endif #if __cplusplus >= 201103L @@ -183,19 +222,25 @@ ATTR StoresNonLit snl; #else ATTR NonLit nl_ctor; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'NonLit' cannot be used in a constant expression}} ATTR NonLit nl_ctor2{}; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'NonLit' cannot be used in a constant expression}} ATTR NonLit nl_ctor3 = {}; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'NonLit' cannot be used in a constant expression}} ATTR thread_local NonLit nl_ctor_tl = {}; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'NonLit' cannot be used in a constant expression}} ATTR StoresNonLit snl; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'StoresNonLit' cannot be used in a constant expression}} #endif // Non-literal types cannot appear in the initializer of a non-literal type. ATTR int nl_in_init = NonLit{42}.value; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-literal type 'NonLit' cannot be used in a constant expression}} ATTR int lit_in_init = LitType{42}.value; #endif @@ -218,6 +263,11 @@ ATTR PODType pod_missing_init = {42 /* should have second arg */}; ATTR PODType pod_full_init = {1, 2}; ATTR PODType pod_non_constexpr_init = {1, ReturnInt()}; // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +#if __cplusplus >= 201103L +// expected-note@-3 {{non-constexpr function 'ReturnInt' cannot be used in a constant expression}} +#else +// expected-note@-5 {{subexpression not valid in a constant expression}} +#endif #if __cplusplus >= 201103L ATTR int val_init{}; @@ -233,15 +283,17 @@ typedef const char *StrType; // initializer struct NotC { constexpr NotC(void *) {} - NotC(int) {} + NotC(int) {} // expected-note 0+ {{declared here}} }; template <class T> struct TestCtor { constexpr TestCtor(int x) : value(x) {} + // expected-note@-1 {{non-constexpr constructor 'NotC' cannot be used in a constant expression}} T value; }; ATTR TestCtor<NotC> t(42); // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{in call to 'TestCtor(42)'}} #endif // Test various array types @@ -261,9 +313,11 @@ struct TestCtor { ATTR LitType non_const_lit(nullptr); // expected-error {{variable does not have a constant initializer}} // expected-note@-1 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-2 {{non-constexpr constructor 'LitType' cannot be used in a constant expression}} ATTR NonLit non_const(nullptr); // expected-error {{variable does not have a constant initializer}} // expected-warning@-1 {{declaration requires a global destructor}} // expected-note@-2 {{required by 'require_constant_initialization' attribute here}} +// expected-note@-3 {{non-constexpr constructor 'NonLit' cannot be used in a constant expression}} LitType const_init_lit(nullptr); // expected-warning {{declaration requires a global constructor}} NonLit const_init{42}; // expected-warning {{declaration requires a global destructor}} constexpr TestCtor<NotC> inval_constexpr(42); // expected-error {{must be initialized by a constant expression}} diff --git a/test/SemaCXX/coreturn.cpp b/test/SemaCXX/coreturn.cpp index 0ec94d1b5997..7265d7c19c2e 100644 --- a/test/SemaCXX/coreturn.cpp +++ b/test/SemaCXX/coreturn.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -triple x86_64-apple-darwin9 %s -std=c++14 -fcoroutines-ts -fsyntax-only -Wignored-qualifiers -Wno-error=return-type -verify -fblocks -Wno-unreachable-code -Wno-unused-value +// RUN: %clang_cc1 -triple x86_64-apple-darwin9 %s -std=c++14 -fcoroutines-ts -fsyntax-only -Wignored-qualifiers -Wno-error=return-type -verify -fblocks -Wall -Wextra -Wno-error=unreachable-code #include "Inputs/std-coroutine.h" using std::experimental::suspend_always; diff --git a/test/SemaCXX/coroutine-uninitialized-warning-crash.cpp b/test/SemaCXX/coroutine-uninitialized-warning-crash.cpp new file mode 100644 index 000000000000..5bdb232d5307 --- /dev/null +++ b/test/SemaCXX/coroutine-uninitialized-warning-crash.cpp @@ -0,0 +1,44 @@ +// RUN: %clang_cc1 -triple x86_64-apple-darwin9 %s -std=c++14 -fcoroutines-ts -fsyntax-only -Wall -Wextra -Wuninitialized -fblocks +#include "Inputs/std-coroutine.h" + +using namespace std::experimental; + + +struct A { + bool await_ready() { return true; } + int await_resume() { return 42; } + template <typename F> + void await_suspend(F) {} +}; + + +struct coro_t { + struct promise_type { + coro_t get_return_object() { return {}; } + suspend_never initial_suspend() { return {}; } + suspend_never final_suspend() { return {}; } + A yield_value(int) { return {}; } + void return_void() {} + static void unhandled_exception() {} + }; +}; + +coro_t f(int n) { + if (n == 0) + co_return; + co_yield 42; + int x = co_await A{}; +} + +template <class Await> +coro_t g(int n) { + if (n == 0) + co_return; + co_yield 42; + int x = co_await Await{}; +} + +int main() { + f(0); + g<A>(0); +} diff --git a/test/SemaCXX/coroutines.cpp b/test/SemaCXX/coroutines.cpp index 47ad86e5b02f..a867cf68a6e1 100644 --- a/test/SemaCXX/coroutines.cpp +++ b/test/SemaCXX/coroutines.cpp @@ -840,12 +840,12 @@ coro<bad_promise_no_return_func> no_return_value_or_return_void() { struct bad_await_suspend_return { bool await_ready(); - // expected-error@+1 {{the return type of 'await_suspend' is required to be 'void' or 'bool' (have 'char')}} + // expected-error@+1 {{return type of 'await_suspend' is required to be 'void' or 'bool' (have 'char')}} char await_suspend(std::experimental::coroutine_handle<>); void await_resume(); }; struct bad_await_ready_return { - // expected-note@+1 {{the return type of 'await_ready' is required to be contextually convertible to 'bool'}} + // expected-note@+1 {{return type of 'await_ready' is required to be contextually convertible to 'bool'}} void await_ready(); bool await_suspend(std::experimental::coroutine_handle<>); void await_resume(); @@ -858,6 +858,14 @@ struct await_ready_explicit_bool { void await_suspend(std::experimental::coroutine_handle<>); void await_resume(); }; +template <class SuspendTy> +struct await_suspend_type_test { + bool await_ready(); + // expected-error@+2 {{return type of 'await_suspend' is required to be 'void' or 'bool' (have 'bool &')}} + // expected-error@+1 {{return type of 'await_suspend' is required to be 'void' or 'bool' (have 'bool &&')}} + SuspendTy await_suspend(std::experimental::coroutine_handle<>); + void await_resume(); +}; void test_bad_suspend() { { // FIXME: The actual error emitted here is terrible, and no number of notes can save it. @@ -873,4 +881,14 @@ void test_bad_suspend() { await_ready_explicit_bool c; co_await c; // OK } + { + await_suspend_type_test<bool &&> a; + await_suspend_type_test<bool &> b; + await_suspend_type_test<const void> c; + await_suspend_type_test<const volatile bool> d; + co_await a; // expected-note {{call to 'await_suspend' implicitly required by coroutine function here}} + co_await b; // expected-note {{call to 'await_suspend' implicitly required by coroutine function here}} + co_await c; // OK + co_await d; // OK + } } diff --git a/test/SemaCXX/type-traits.cpp b/test/SemaCXX/type-traits.cpp index 919122576222..8a994f01ba1a 100644 --- a/test/SemaCXX/type-traits.cpp +++ b/test/SemaCXX/type-traits.cpp @@ -252,6 +252,11 @@ void is_pod() { int arr[F(__is_pod(void))]; } { int arr[F(__is_pod(cvoid))]; } // { int arr[F(__is_pod(NonPODUnion))]; } + + { int arr[T(__is_pod(ACompleteType))]; } + { int arr[F(__is_pod(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_pod(AnIncompleteType[]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_pod(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} } typedef Empty EmptyAr[10]; @@ -287,6 +292,11 @@ void is_empty() { int arr[F(__is_empty(IntArNB))]; } { int arr[F(__is_empty(HasAnonymousUnion))]; } // { int arr[F(__is_empty(DerivesVirt))]; } + + { int arr[T(__is_empty(ACompleteType))]; } + { int arr[F(__is_empty(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_empty(AnIncompleteType[]))]; } + { int arr[F(__is_empty(AnIncompleteType[1]))]; } } typedef Derives ClassType; @@ -511,6 +521,8 @@ void is_aggregate() constexpr bool TrueAfterCpp14 = __cplusplus > 201402L; __is_aggregate(AnIncompleteType); // expected-error {{incomplete type}} + __is_aggregate(AnIncompleteType[]); // expected-error {{incomplete type}} + __is_aggregate(AnIncompleteType[1]); // expected-error {{incomplete type}} __is_aggregate(AnIncompleteTypeAr); // expected-error {{incomplete type}} __is_aggregate(AnIncompleteTypeArNB); // expected-error {{incomplete type}} __is_aggregate(AnIncompleteTypeArMB); // expected-error {{incomplete type}} @@ -1220,6 +1232,13 @@ void is_trivial2() int t32[F(__is_trivial(SuperNonTrivialStruct))]; int t33[F(__is_trivial(NonTCStruct))]; int t34[F(__is_trivial(ExtDefaulted))]; + + int t40[T(__is_trivial(ACompleteType))]; + int t41[F(__is_trivial(AnIncompleteType))]; // expected-error {{incomplete type}} + int t42[F(__is_trivial(AnIncompleteType[]))]; // expected-error {{incomplete type}} + int t43[F(__is_trivial(AnIncompleteType[1]))]; // expected-error {{incomplete type}} + int t44[F(__is_trivial(void))]; + int t45[F(__is_trivial(const volatile void))]; } void is_trivially_copyable2() @@ -1257,6 +1276,13 @@ void is_trivially_copyable2() int t34[T(__is_trivially_copyable(const int))]; int t35[T(__is_trivially_copyable(volatile int))]; + + int t40[T(__is_trivially_copyable(ACompleteType))]; + int t41[F(__is_trivially_copyable(AnIncompleteType))]; // expected-error {{incomplete type}} + int t42[F(__is_trivially_copyable(AnIncompleteType[]))]; // expected-error {{incomplete type}} + int t43[F(__is_trivially_copyable(AnIncompleteType[1]))]; // expected-error {{incomplete type}} + int t44[F(__is_trivially_copyable(void))]; + int t45[F(__is_trivially_copyable(const volatile void))]; } struct CStruct { @@ -1320,6 +1346,13 @@ void is_standard_layout() int t15[F(__is_standard_layout(CppStructNonStandardByBaseAr))]; int t16[F(__is_standard_layout(CppStructNonStandardBySameBase))]; int t17[F(__is_standard_layout(CppStructNonStandardBy2ndVirtBase))]; + + int t40[T(__is_standard_layout(ACompleteType))]; + int t41[F(__is_standard_layout(AnIncompleteType))]; // expected-error {{incomplete type}} + int t42[F(__is_standard_layout(AnIncompleteType[]))]; // expected-error {{incomplete type}} + int t43[F(__is_standard_layout(AnIncompleteType[1]))]; // expected-error {{incomplete type}} + int t44[F(__is_standard_layout(void))]; + int t45[F(__is_standard_layout(const volatile void))]; } void is_signed() @@ -2133,6 +2166,13 @@ void trivial_checks() TrivialMoveButNotCopy)))]; } { int arr[T((__is_assignable(TrivialMoveButNotCopy &, TrivialMoveButNotCopy &&)))]; } + + { int arr[T(__is_assignable(ACompleteType, ACompleteType))]; } + { int arr[F(__is_assignable(AnIncompleteType, AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_assignable(AnIncompleteType[], AnIncompleteType[]))]; } + { int arr[F(__is_assignable(AnIncompleteType[1], AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_assignable(void, void))]; } + { int arr[F(__is_assignable(const volatile void, const volatile void))]; } } void constructible_checks() { @@ -2164,6 +2204,19 @@ void constructible_checks() { // PR25513 { int arr[F(__is_constructible(int(int)))]; } + + { int arr[T(__is_constructible(ACompleteType))]; } + { int arr[T(__is_nothrow_constructible(ACompleteType))]; } + { int arr[F(__is_constructible(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_nothrow_constructible(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_constructible(AnIncompleteType[]))]; } + { int arr[F(__is_nothrow_constructible(AnIncompleteType[]))]; } + { int arr[F(__is_constructible(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_nothrow_constructible(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_constructible(void))]; } + { int arr[F(__is_nothrow_constructible(void))]; } + { int arr[F(__is_constructible(const volatile void))]; } + { int arr[F(__is_nothrow_constructible(const volatile void))]; } } // Instantiation of __is_trivially_constructible @@ -2192,6 +2245,13 @@ void is_trivially_constructible_test() { { int arr[F((is_trivially_constructible<NonTrivialDefault>::value))]; } { int arr[F((is_trivially_constructible<ThreeArgCtor, int*, char*, int&>::value))]; } { int arr[F((is_trivially_constructible<Abstract>::value))]; } // PR19178 + + { int arr[T(__is_trivially_constructible(ACompleteType))]; } + { int arr[F(__is_trivially_constructible(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_trivially_constructible(AnIncompleteType[]))]; } + { int arr[F(__is_trivially_constructible(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_trivially_constructible(void))]; } + { int arr[F(__is_trivially_constructible(const volatile void))]; } } void array_rank() { @@ -2218,6 +2278,13 @@ void is_destructible_test() { { int arr[F(__is_destructible(AllDeleted))]; } { int arr[T(__is_destructible(ThrowingDtor))]; } { int arr[T(__is_destructible(NoThrowDtor))]; } + + { int arr[T(__is_destructible(ACompleteType))]; } + { int arr[F(__is_destructible(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_destructible(AnIncompleteType[]))]; } + { int arr[F(__is_destructible(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_destructible(void))]; } + { int arr[F(__is_destructible(const volatile void))]; } } void is_nothrow_destructible_test() { @@ -2234,4 +2301,33 @@ void is_nothrow_destructible_test() { { int arr[F(__is_nothrow_destructible(ThrowingDtor))]; } { int arr[T(__is_nothrow_destructible(NoExceptDtor))]; } { int arr[T(__is_nothrow_destructible(NoThrowDtor))]; } + + { int arr[T(__is_nothrow_destructible(ACompleteType))]; } + { int arr[F(__is_nothrow_destructible(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_nothrow_destructible(AnIncompleteType[]))]; } + { int arr[F(__is_nothrow_destructible(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_nothrow_destructible(void))]; } + { int arr[F(__is_nothrow_destructible(const volatile void))]; } +} + +void is_trivially_destructible_test() { + { int arr[T(__is_trivially_destructible(int))]; } + { int arr[T(__is_trivially_destructible(int[2]))]; } + { int arr[F(__is_trivially_destructible(int[]))]; } + { int arr[F(__is_trivially_destructible(void))]; } + { int arr[T(__is_trivially_destructible(int &))]; } + { int arr[F(__is_trivially_destructible(HasDest))]; } + { int arr[F(__is_trivially_destructible(AllPrivate))]; } + { int arr[F(__is_trivially_destructible(SuperNonTrivialStruct))]; } + { int arr[T(__is_trivially_destructible(AllDefaulted))]; } + { int arr[F(__is_trivially_destructible(AllDeleted))]; } + { int arr[F(__is_trivially_destructible(ThrowingDtor))]; } + { int arr[F(__is_trivially_destructible(NoThrowDtor))]; } + + { int arr[T(__is_trivially_destructible(ACompleteType))]; } + { int arr[F(__is_trivially_destructible(AnIncompleteType))]; } // expected-error {{incomplete type}} + { int arr[F(__is_trivially_destructible(AnIncompleteType[]))]; } + { int arr[F(__is_trivially_destructible(AnIncompleteType[1]))]; } // expected-error {{incomplete type}} + { int arr[F(__is_trivially_destructible(void))]; } + { int arr[F(__is_trivially_destructible(const volatile void))]; } } diff --git a/test/SemaObjC/attr-deprecated.m b/test/SemaObjC/attr-deprecated.m index b0613851ddaa..28031997af7a 100644 --- a/test/SemaObjC/attr-deprecated.m +++ b/test/SemaObjC/attr-deprecated.m @@ -83,7 +83,7 @@ int t5() { } -__attribute ((deprecated)) // expected-note 2 {{'DEPRECATED' has been explicitly marked deprecated here}} +__attribute ((deprecated)) // expected-note {{'DEPRECATED' has been explicitly marked deprecated here}} @interface DEPRECATED { @public int ivar; DEPRECATED *ivar2; // no warning. @@ -98,9 +98,17 @@ __attribute ((deprecated)) // expected-note 2 {{'DEPRECATED' has been explicitly @end @interface DEPRECATED (Category2) // no warning. +- (id)meth; @end -@implementation DEPRECATED (Category2) // expected-warning {{'DEPRECATED' is deprecated}} +__attribute__((deprecated)) +void depr_function(); + +@implementation DEPRECATED (Category2) // no warning +- (id)meth { + depr_function(); // no warning. + return 0; +} @end @interface NS : DEPRECATED // expected-warning {{'DEPRECATED' is deprecated}} diff --git a/test/SemaObjC/class-unavail-warning.m b/test/SemaObjC/class-unavail-warning.m index 268d51910b0c..f7d8f569ca67 100644 --- a/test/SemaObjC/class-unavail-warning.m +++ b/test/SemaObjC/class-unavail-warning.m @@ -2,7 +2,7 @@ // rdar://9092208 __attribute__((unavailable("not available"))) -@interface MyClass { // expected-note 8 {{'MyClass' has been explicitly marked unavailable here}} +@interface MyClass { // expected-note 7 {{'MyClass' has been explicitly marked unavailable here}} @public void *_test; MyClass *ivar; // no error. @@ -28,7 +28,7 @@ __attribute__((unavailable("not available"))) @interface MyClass (Cat2) // no error. @end -@implementation MyClass (Cat2) // expected-error {{unavailable}} +@implementation MyClass (Cat2) // no error. @end int main() { diff --git a/test/SemaObjC/warn-deprecated-implementations.m b/test/SemaObjC/warn-deprecated-implementations.m index 0c341165b0f0..df2557b9cd52 100644 --- a/test/SemaObjC/warn-deprecated-implementations.m +++ b/test/SemaObjC/warn-deprecated-implementations.m @@ -28,15 +28,14 @@ - (void) G {} // No warning, implementing its own deprecated method @end -__attribute__((deprecated)) // expected-note 2 {{'CL' has been explicitly marked deprecated here}} +__attribute__((deprecated)) // expected-note {{'CL' has been explicitly marked deprecated here}} @interface CL // expected-note 2 {{class declared here}} @end @implementation CL // expected-warning {{Implementing deprecated class}} @end -@implementation CL ( SomeCategory ) // expected-warning {{'CL' is deprecated}} \ - // expected-warning {{Implementing deprecated category}} +@implementation CL (SomeCategory) // expected-warning {{Implementing deprecated category}} @end @interface CL_SUB : CL // expected-warning {{'CL' is deprecated}} |