Index: clang/lib/CodeGen/CGCall.cpp
===================================================================
--- clang/lib/CodeGen/CGCall.cpp
+++ clang/lib/CodeGen/CGCall.cpp
@@ -2077,6 +2077,7 @@
     hasUsedSRet = true;
     if (RetAI.getInReg())
       SRETAttrs.addAttribute(llvm::Attribute::InReg);
+    SRETAttrs.addAlignmentAttr(RetAI.getIndirectAlign().getQuantity());
     ArgAttrs[IRFunctionArgs.getSRetArgNo()] =
         llvm::AttributeSet::get(getLLVMContext(), SRETAttrs);
   }
Index: clang/test/CodeGen/2006-05-19-SingleEltReturn.c
===================================================================
--- clang/test/CodeGen/2006-05-19-SingleEltReturn.c
+++ clang/test/CodeGen/2006-05-19-SingleEltReturn.c
@@ -24,7 +24,7 @@
 
 
 // X86_32: define void @foo(%struct.Y* %P)
-// X86_32:   call void @bar(%struct.Y* sret %{{[^),]*}})
+// X86_32:   call void @bar(%struct.Y* sret align 4 %{{[^),]*}})
 
-// X86_32: define void @bar(%struct.Y* noalias sret %{{[^,)]*}})
+// X86_32: define void @bar(%struct.Y* noalias sret align 4 %{{[^,)]*}})
 // X86_32:   ret void
Index: clang/test/CodeGen/aarch64-varargs.c
===================================================================
--- clang/test/CodeGen/aarch64-varargs.c
+++ clang/test/CodeGen/aarch64-varargs.c
@@ -639,7 +639,7 @@
   __int128 val;
 } overaligned_int128_struct;
 overaligned_int128_struct overaligned_int128_struct_test() {
-// CHECK-LABEL: define void @overaligned_int128_struct_test(%struct.overaligned_int128_struct* noalias sret %agg.result)
+// CHECK-LABEL: define void @overaligned_int128_struct_test(%struct.overaligned_int128_struct* noalias sret align 32 %agg.result)
   return va_arg(the_list, overaligned_int128_struct);
 // CHECK: [[GR_OFFS:%[a-z_0-9]+]] = load i32, i32* getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 3)
 // CHECK: [[EARLY_ONSTACK:%[a-z_0-9]+]] = icmp sge i32 [[GR_OFFS]], 0
@@ -853,7 +853,7 @@
   __int128 val __attribute__((aligned(32)));
 } overaligned_int128_struct_member;
 overaligned_int128_struct_member overaligned_int128_struct_member_test() {
-// CHECK-LABEL: define void @overaligned_int128_struct_member_test(%struct.overaligned_int128_struct_member* noalias sret %agg.result)
+// CHECK-LABEL: define void @overaligned_int128_struct_member_test(%struct.overaligned_int128_struct_member* noalias sret align 32 %agg.result)
   return va_arg(the_list, overaligned_int128_struct_member);
 // CHECK: [[GR_OFFS:%[a-z_0-9]+]] = load i32, i32* getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 3)
 // CHECK: [[EARLY_ONSTACK:%[a-z_0-9]+]] = icmp sge i32 [[GR_OFFS]], 0
Index: clang/test/CodeGen/aggregate-assign-call.c
===================================================================
--- clang/test/CodeGen/aggregate-assign-call.c
+++ clang/test/CodeGen/aggregate-assign-call.c
@@ -62,8 +62,8 @@
     // O1-NEWPM: %[[TMP3:.*]] = bitcast %struct.S* %[[TMP2_ALLOCA]] to i8*
     // O1-NEWPM: call void @llvm.lifetime.end.p0i8({{[^,]*}}, i8* nonnull %[[P]])
     //
-    // O1-LEGACY: call void @foo_int(%struct.S* sret %[[TMP1_ALLOCA]],
-    // O1-NEWPM: call void @foo_int(%struct.S* nonnull sret %[[TMP1_ALLOCA]],
+    // O1-LEGACY: call void @foo_int(%struct.S* sret align 4 %[[TMP1_ALLOCA]],
+    // O1-NEWPM: call void @foo_int(%struct.S* nonnull sret align 4 %[[TMP1_ALLOCA]],
     // O1: call void @llvm.memcpy
     // O1-LEGACY: %[[P:[^ ]+]] = bitcast %struct.S* %[[TMP1_ALLOCA]] to i8*
     // O1-LEGACY: call void @llvm.lifetime.end.p0i8({{[^,]*}}, i8* %[[P]])
@@ -71,8 +71,8 @@
     // O1-LEGACY: %[[P:[^ ]+]] = bitcast %struct.S* %[[TMP2_ALLOCA]] to i8*
     // O1-LEGACY: call void @llvm.lifetime.start.p0i8({{[^,]*}}, i8* %[[P]])
     // O1-NEWPM: call void @llvm.lifetime.start.p0i8({{[^,]*}}, i8* nonnull %[[TMP3]])
-    // O1-LEGACY: call void @foo_int(%struct.S* sret %[[TMP2_ALLOCA]],
-    // O1-NEWPM: call void @foo_int(%struct.S* nonnull sret %[[TMP2_ALLOCA]],
+    // O1-LEGACY: call void @foo_int(%struct.S* sret align 4 %[[TMP2_ALLOCA]],
+    // O1-NEWPM: call void @foo_int(%struct.S* nonnull sret align 4 %[[TMP2_ALLOCA]],
     // O1: call void @llvm.memcpy
     // O1-LEGACY: %[[P:[^ ]+]] = bitcast %struct.S* %[[TMP2_ALLOCA]] to i8*
     // O1-LEGACY: call void @llvm.lifetime.end.p0i8({{[^,]*}}, i8* %[[P]])
Index: clang/test/CodeGen/aligned-sret.c
===================================================================
--- /dev/null
+++ clang/test/CodeGen/aligned-sret.c
@@ -0,0 +1,10 @@
+// RUN: %clang_cc1 -triple x86_64-apple-macos %s -S -emit-llvm -o- | FileCheck %s
+
+typedef __attribute__((__ext_vector_type__(4),__aligned__(16))) double simd_double4;
+typedef struct { simd_double4 columns[4]; } simd_double4x4;
+typedef simd_double4x4 matrix_double4x4;
+
+// CHECK: define void @ident(%struct.simd_double4x4* noalias sret align 16 %agg.result
+matrix_double4x4 ident(matrix_double4x4 x) {
+  return x;
+}
Index: clang/test/CodeGen/arc/arguments.c
===================================================================
--- clang/test/CodeGen/arc/arguments.c
+++ clang/test/CodeGen/arc/arguments.c
@@ -22,7 +22,7 @@
 typedef struct {
   int cc;
 } s2;
-// CHECK: define void @f2(%struct.s2* noalias sret %agg.result)
+// CHECK: define void @f2(%struct.s2* noalias sret align 4 %agg.result)
 s2 f2() {
   s2 foo;
   return foo;
@@ -32,7 +32,7 @@
   int cc;
   int dd;
 } s3;
-// CHECK: define void @f3(%struct.s3* noalias sret %agg.result)
+// CHECK: define void @f3(%struct.s3* noalias sret align 4 %agg.result)
 s3 f3() {
   s3 foo;
   return foo;
@@ -128,8 +128,8 @@
 
 // 1 sret + 1 i32 + 2*(i32 coerce) + 4*(i32 coerce) + 1 byval
 s16 st4(int x, s8 a, s16 b, s16 c) { return b; }
-// CHECK: define void @st4(%struct.s16* noalias sret %agg.result, i32 inreg %x, i32 inreg %a.coerce0, i32 inreg %a.coerce1, i32 inreg %b.coerce0, i32 inreg %b.coerce1, i32 inreg %b.coerce2, i32 inreg %b.coerce3, { i32, i32, i32, i32 } %c.coerce)
+// CHECK: define void @st4(%struct.s16* noalias sret align 4 %agg.result, i32 inreg %x, i32 inreg %a.coerce0, i32 inreg %a.coerce1, i32 inreg %b.coerce0, i32 inreg %b.coerce1, i32 inreg %b.coerce2, i32 inreg %b.coerce3, { i32, i32, i32, i32 } %c.coerce)
 
 // 1 sret + 2*(i32 coerce) + 4*(i32 coerce) + 4*(i32 coerce)
 s16 st5(s8 a, s16 b, s16 c) { return b; }
-// CHECK: define void @st5(%struct.s16* noalias sret %agg.result, i32 inreg %a.coerce0, i32 inreg %a.coerce1, i32 inreg %b.coerce0, i32 inreg %b.coerce1, i32 inreg %b.coerce2, i32 inreg %b.coerce3, { i32, i32, i32, i32 } %c.coerce)
+// CHECK: define void @st5(%struct.s16* noalias sret align 4 %agg.result, i32 inreg %a.coerce0, i32 inreg %a.coerce1, i32 inreg %b.coerce0, i32 inreg %b.coerce1, i32 inreg %b.coerce2, i32 inreg %b.coerce3, { i32, i32, i32, i32 } %c.coerce)
Index: clang/test/CodeGen/arm-aapcs-vfp.c
===================================================================
--- clang/test/CodeGen/arm-aapcs-vfp.c
+++ clang/test/CodeGen/arm-aapcs-vfp.c
@@ -125,7 +125,7 @@
 // CHECK: define arm_aapcs_vfpcc void @test_vfp_stack_gpr_split_2(double %a, double %b, double %c, double %d, double %e, double %f, double %g, double %h, double %i, i32 %j, [2 x i64] %k.coerce)
 void test_vfp_stack_gpr_split_2(double a, double b, double c, double d, double e, double f, double g, double h, double i, int j, struct_long_long_int k) {}
 
-// CHECK: define arm_aapcs_vfpcc void @test_vfp_stack_gpr_split_3(%struct.struct_long_long_int* noalias sret %agg.result, double %a, double %b, double %c, double %d, double %e, double %f, double %g, double %h, double %i, [2 x i64] %k.coerce)
+// CHECK: define arm_aapcs_vfpcc void @test_vfp_stack_gpr_split_3(%struct.struct_long_long_int* noalias sret align 8 %agg.result, double %a, double %b, double %c, double %d, double %e, double %f, double %g, double %h, double %i, [2 x i64] %k.coerce)
 struct_long_long_int test_vfp_stack_gpr_split_3(double a, double b, double c, double d, double e, double f, double g, double h, double i, struct_long_long_int k) {}
 
 typedef struct { int a; int b:4; int c; } struct_int_bitfield_int;
Index: clang/test/CodeGen/arm-homogenous.c
===================================================================
--- clang/test/CodeGen/arm-homogenous.c
+++ clang/test/CodeGen/arm-homogenous.c
@@ -27,7 +27,7 @@
 void test_return_union_with_first_floats(void) {
   g_u_f = returns_union_with_first_floats();
 }
-// CHECK: declare arm_aapcs_vfpcc void @returns_union_with_first_floats(%union.union_with_first_floats* sret)
+// CHECK: declare arm_aapcs_vfpcc void @returns_union_with_first_floats(%union.union_with_first_floats* sret align 4)
 
 /* This is not a homogenous aggregate - fundamental types are different */
 typedef union {
@@ -47,7 +47,7 @@
 void test_return_union_with_non_first_floats(void) {
   g_u_nf_f = returns_union_with_non_first_floats();
 }
-// CHECK: declare arm_aapcs_vfpcc void @returns_union_with_non_first_floats(%union.union_with_non_first_floats* sret)
+// CHECK: declare arm_aapcs_vfpcc void @returns_union_with_non_first_floats(%union.union_with_non_first_floats* sret align 4)
 
 /* This is not a homogenous aggregate - fundamental types are different */
 typedef struct {
@@ -67,7 +67,7 @@
 void test_return_struct_with_union_with_first_floats(void) {
   g_s_f = returns_struct_with_union_with_first_floats();
 }
-// CHECK: declare arm_aapcs_vfpcc void @returns_struct_with_union_with_first_floats(%struct.struct_with_union_with_first_floats* sret)
+// CHECK: declare arm_aapcs_vfpcc void @returns_struct_with_union_with_first_floats(%struct.struct_with_union_with_first_floats* sret align 4)
 
 /* This is not a homogenous aggregate - fundamental types are different */
 typedef struct {
@@ -87,7 +87,7 @@
 void test_return_struct_with_union_with_non_first_floats(void) {
   g_s_nf_f = returns_struct_with_union_with_non_first_floats();
 }
-// CHECK: declare arm_aapcs_vfpcc void @returns_struct_with_union_with_non_first_floats(%struct.struct_with_union_with_non_first_floats* sret)
+// CHECK: declare arm_aapcs_vfpcc void @returns_struct_with_union_with_non_first_floats(%struct.struct_with_union_with_non_first_floats* sret align 4)
 
 /* Plain array is not a homogenous aggregate */
 extern void takes_array_of_floats(float a[4]);
Index: clang/test/CodeGen/arm-neon-vld.c
===================================================================
--- clang/test/CodeGen/arm-neon-vld.c
+++ clang/test/CodeGen/arm-neon-vld.c
@@ -9,7 +9,7 @@
 
 // CHECK-LABEL: @test_vld1_f16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float16x4x2_t, align 8
-// CHECK-A32: %struct.float16x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float16x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float16x4x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8*
@@ -29,7 +29,7 @@
 
 // CHECK-LABEL: @test_vld1_f16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float16x4x3_t, align 8
-// CHECK-A32: %struct.float16x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float16x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float16x4x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8*
@@ -49,7 +49,7 @@
 
 // CHECK-LABEL: @test_vld1_f16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float16x4x4_t, align 8
-// CHECK-A32: %struct.float16x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float16x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float16x4x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8*
@@ -69,7 +69,7 @@
 
 // CHECK-LABEL: @test_vld1_f32_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float32x2x2_t, align 8
-// CHECK-A32: %struct.float32x2x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float32x2x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float32x2x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x2x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8*
@@ -89,7 +89,7 @@
 
 // CHECK-LABEL: @test_vld1_f32_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float32x2x3_t, align 8
-// CHECK-A32: %struct.float32x2x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float32x2x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float32x2x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x2x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8*
@@ -108,7 +108,7 @@
 
 // CHECK-LABEL: @test_vld1_f32_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float32x2x4_t, align 8
-// CHECK-A32: %struct.float32x2x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float32x2x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float32x2x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x2x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8*
@@ -128,7 +128,7 @@
 
 // CHECK-LABEL: @test_vld1_p16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly16x4x2_t, align 8
-// CHECK-A32: %struct.poly16x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly16x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly16x4x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -148,7 +148,7 @@
 
 // CHECK-LABEL: @test_vld1_p16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly16x4x3_t, align 8
-// CHECK-A32: %struct.poly16x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly16x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly16x4x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -168,7 +168,7 @@
 
 // CHECK-LABEL: @test_vld1_p16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly16x4x4_t, align 8
-// CHECK-A32: %struct.poly16x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly16x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly16x4x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -188,7 +188,7 @@
 
 // CHECK-LABEL: @test_vld1_p8_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly8x8x2_t, align 8
-// CHECK-A32: %struct.poly8x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly8x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly8x8x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x8x2_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x2|arm.neon.vld1x2}}.v8i8.p0i8(i8* %a)
@@ -206,7 +206,7 @@
 
 // CHECK-LABEL: @test_vld1_p8_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly8x8x3_t, align 8
-// CHECK-A32: %struct.poly8x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly8x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly8x8x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x8x3_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x3|arm.neon.vld1x3}}.v8i8.p0i8(i8* %a)
@@ -224,7 +224,7 @@
 
 // CHECK-LABEL: @test_vld1_p8_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly8x8x4_t, align 8
-// CHECK-A32: %struct.poly8x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly8x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly8x8x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x8x4_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x4|arm.neon.vld1x4}}.v8i8.p0i8(i8* %a)
@@ -242,7 +242,7 @@
 
 // CHECK-LABEL: @test_vld1_s16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int16x4x2_t, align 8
-// CHECK-A32: %struct.int16x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int16x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int16x4x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -262,7 +262,7 @@
 
 // CHECK-LABEL: @test_vld1_s16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int16x4x3_t, align 8
-// CHECK-A32: %struct.int16x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int16x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int16x4x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -282,7 +282,7 @@
 
 // CHECK-LABEL: @test_vld1_s16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int16x4x4_t, align 8
-// CHECK-A32: %struct.int16x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int16x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int16x4x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -302,7 +302,7 @@
 
 // CHECK-LABEL: @test_vld1_s32_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int32x2x2_t, align 8
-// CHECK-A32: %struct.int32x2x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int32x2x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int32x2x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x2x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -322,7 +322,7 @@
 
 // CHECK-LABEL: @test_vld1_s32_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int32x2x3_t, align 8
-// CHECK-A32: %struct.int32x2x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int32x2x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int32x2x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x2x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -342,7 +342,7 @@
 
 // CHECK-LABEL: @test_vld1_s32_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int32x2x4_t, align 8
-// CHECK-A32: %struct.int32x2x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int32x2x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int32x2x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x2x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -362,7 +362,7 @@
 
 // CHECK-LABEL: @test_vld1_s64_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int64x1x2_t, align 8
-// CHECK-A32: %struct.int64x1x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int64x1x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int64x1x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int64x1x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -382,7 +382,7 @@
 
 // CHECK-LABEL: @test_vld1_s64_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int64x1x3_t, align 8
-// CHECK-A32: %struct.int64x1x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int64x1x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int64x1x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int64x1x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -402,7 +402,7 @@
 
 // CHECK-LABEL: @test_vld1_s64_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int64x1x4_t, align 8
-// CHECK-A32: %struct.int64x1x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int64x1x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int64x1x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int64x1x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -422,7 +422,7 @@
 
 // CHECK-LABEL: @test_vld1_s8_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int8x8x2_t, align 8
-// CHECK-A32: %struct.int8x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int8x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int8x8x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x8x2_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x2|arm.neon.vld1x2}}.v8i8.p0i8(i8* %a)
@@ -440,7 +440,7 @@
 
 // CHECK-LABEL: @test_vld1_s8_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int8x8x3_t, align 8
-// CHECK-A32: %struct.int8x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int8x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int8x8x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x8x3_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x3|arm.neon.vld1x3}}.v8i8.p0i8(i8* %a)
@@ -458,7 +458,7 @@
 
 // CHECK-LABEL: @test_vld1_s8_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int8x8x4_t, align 8
-// CHECK-A32: %struct.int8x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int8x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int8x8x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x8x4_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x4|arm.neon.vld1x4}}.v8i8.p0i8(i8* %a)
@@ -476,7 +476,7 @@
 
 // CHECK-LABEL: @test_vld1_u16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint16x4x2_t, align 8
-// CHECK-A32: %struct.uint16x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint16x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint16x4x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -496,7 +496,7 @@
 
 // CHECK-LABEL: @test_vld1_u16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint16x4x3_t, align 8
-// CHECK-A32: %struct.uint16x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint16x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint16x4x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -516,7 +516,7 @@
 
 // CHECK-LABEL: @test_vld1_u16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint16x4x4_t, align 8
-// CHECK-A32: %struct.uint16x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint16x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint16x4x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -536,7 +536,7 @@
 
 // CHECK-LABEL: @test_vld1_u32_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint32x2x2_t, align 8
-// CHECK-A32: %struct.uint32x2x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint32x2x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint32x2x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x2x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -556,7 +556,7 @@
 
 // CHECK-LABEL: @test_vld1_u32_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint32x2x3_t, align 8
-// CHECK-A32: %struct.uint32x2x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint32x2x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint32x2x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x2x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -576,7 +576,7 @@
 
 // CHECK-LABEL: @test_vld1_u32_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint32x2x4_t, align 8
-// CHECK-A32: %struct.uint32x2x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint32x2x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint32x2x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x2x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -596,7 +596,7 @@
 
 // CHECK-LABEL: @test_vld1_u64_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint64x1x2_t, align 8
-// CHECK-A32: %struct.uint64x1x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint64x1x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint64x1x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint64x1x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -616,7 +616,7 @@
 
 // CHECK-LABEL: @test_vld1_u64_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint64x1x3_t, align 8
-// CHECK-A32: %struct.uint64x1x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint64x1x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint64x1x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint64x1x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -636,7 +636,7 @@
 
 // CHECK-LABEL: @test_vld1_u64_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint64x1x4_t, align 8
-// CHECK-A32: %struct.uint64x1x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint64x1x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint64x1x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint64x1x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -656,7 +656,7 @@
 
 // CHECK-LABEL: @test_vld1_u8_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint8x8x2_t, align 8
-// CHECK-A32: %struct.uint8x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint8x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint8x8x2_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x8x2_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x2|arm.neon.vld1x2}}.v8i8.p0i8(i8* %a)
@@ -674,7 +674,7 @@
 
 // CHECK-LABEL: @test_vld1_u8_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint8x8x3_t, align 8
-// CHECK-A32: %struct.uint8x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint8x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint8x8x3_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x8x3_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x3|arm.neon.vld1x3}}.v8i8.p0i8(i8* %a)
@@ -692,7 +692,7 @@
 
 // CHECK-LABEL: @test_vld1_u8_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint8x8x4_t, align 8
-// CHECK-A32: %struct.uint8x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint8x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint8x8x4_t, align 8
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x8x4_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.{{aarch64.neon.ld1x4|arm.neon.vld1x4}}.v8i8.p0i8(i8* %a)
@@ -710,7 +710,7 @@
 
 // CHECK-LABEL: @test_vld1q_f16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float16x8x2_t, align 16
-// CHECK-A32: %struct.float16x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float16x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float16x8x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x8x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8*
@@ -730,7 +730,7 @@
 
 // CHECK-LABEL: @test_vld1q_f16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float16x8x3_t, align 16
-// CHECK-A32: %struct.float16x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float16x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float16x8x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x8x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8*
@@ -750,7 +750,7 @@
 
 // CHECK-LABEL: @test_vld1q_f16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float16x8x4_t, align 16
-// CHECK-A32: %struct.float16x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float16x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float16x8x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float16x8x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast half* %a to i8*
@@ -770,7 +770,7 @@
 
 // CHECK-LABEL: @test_vld1q_f32_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float32x4x2_t, align 16
-// CHECK-A32: %struct.float32x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float32x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float32x4x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8*
@@ -790,7 +790,7 @@
 
 // CHECK-LABEL: @test_vld1q_f32_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float32x4x3_t, align 16
-// CHECK-A32: %struct.float32x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float32x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float32x4x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8*
@@ -810,7 +810,7 @@
 
 // CHECK-LABEL: @test_vld1q_f32_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.float32x4x4_t, align 16
-// CHECK-A32: %struct.float32x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.float32x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.float32x4x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.float32x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast float* %a to i8*
@@ -830,7 +830,7 @@
 
 // CHECK-LABEL: @test_vld1q_p16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly16x8x2_t, align 16
-// CHECK-A32: %struct.poly16x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly16x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -850,7 +850,7 @@
 
 // CHECK-LABEL: @test_vld1q_p16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly16x8x3_t, align 16
-// CHECK-A32: %struct.poly16x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly16x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -870,7 +870,7 @@
 
 // CHECK-LABEL: @test_vld1q_p16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly16x8x4_t, align 16
-// CHECK-A32: %struct.poly16x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly16x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly16x8x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly16x8x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -890,7 +890,7 @@
 
 // CHECK-LABEL: @test_vld1q_p8_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly8x16x2_t, align 16
-// CHECK-A32: %struct.poly8x16x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly8x16x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x2|arm.neon.vld1x2}}.v16i8.p0i8(i8* %a)
@@ -908,7 +908,7 @@
 
 // CHECK-LABEL: @test_vld1q_p8_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly8x16x3_t, align 16
-// CHECK-A32: %struct.poly8x16x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly8x16x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x3_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x3|arm.neon.vld1x3}}.v16i8.p0i8(i8* %a)
@@ -926,7 +926,7 @@
 
 // CHECK-LABEL: @test_vld1q_p8_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.poly8x16x4_t, align 16
-// CHECK-A32: %struct.poly8x16x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.poly8x16x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.poly8x16x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.poly8x16x4_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x4|arm.neon.vld1x4}}.v16i8.p0i8(i8* %a)
@@ -944,7 +944,7 @@
 
 // CHECK-LABEL: @test_vld1q_s16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int16x8x2_t, align 16
-// CHECK-A32: %struct.int16x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int16x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int16x8x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -964,7 +964,7 @@
 
 // CHECK-LABEL: @test_vld1q_s16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int16x8x3_t, align 16
-// CHECK-A32: %struct.int16x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int16x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int16x8x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -984,7 +984,7 @@
 
 // CHECK-LABEL: @test_vld1q_s16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int16x8x4_t, align 16
-// CHECK-A32: %struct.int16x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int16x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int16x8x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int16x8x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -1004,7 +1004,7 @@
 
 // CHECK-LABEL: @test_vld1q_s32_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int32x4x2_t, align 16
-// CHECK-A32: %struct.int32x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int32x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int32x4x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -1024,7 +1024,7 @@
 
 // CHECK-LABEL: @test_vld1q_s32_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int32x4x3_t, align 16
-// CHECK-A32: %struct.int32x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int32x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int32x4x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -1044,7 +1044,7 @@
 
 // CHECK-LABEL: @test_vld1q_s32_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int32x4x4_t, align 16
-// CHECK-A32: %struct.int32x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int32x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int32x4x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int32x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -1064,7 +1064,7 @@
 
 // CHECK-LABEL: @test_vld1q_s64_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int64x2x2_t, align 16
-// CHECK-A32: %struct.int64x2x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int64x2x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int64x2x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int64x2x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -1084,7 +1084,7 @@
 
 // CHECK-LABEL: @test_vld1q_s64_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int64x2x3_t, align 16
-// CHECK-A32: %struct.int64x2x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int64x2x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int64x2x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int64x2x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -1104,7 +1104,7 @@
 
 // CHECK-LABEL: @test_vld1q_s64_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int64x2x4_t, align 16
-// CHECK-A32: %struct.int64x2x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int64x2x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int64x2x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int64x2x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -1124,7 +1124,7 @@
 
 // CHECK-LABEL: @test_vld1q_s8_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int8x16x2_t, align 16
-// CHECK-A32: %struct.int8x16x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int8x16x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int8x16x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x2|arm.neon.vld1x2}}.v16i8.p0i8(i8* %a)
@@ -1142,7 +1142,7 @@
 
 // CHECK-LABEL: @test_vld1q_s8_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int8x16x3_t, align 16
-// CHECK-A32: %struct.int8x16x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int8x16x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int8x16x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x3_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x3|arm.neon.vld1x3}}.v16i8.p0i8(i8* %a)
@@ -1160,7 +1160,7 @@
 
 // CHECK-LABEL: @test_vld1q_s8_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.int8x16x4_t, align 16
-// CHECK-A32: %struct.int8x16x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.int8x16x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.int8x16x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.int8x16x4_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x4|arm.neon.vld1x4}}.v16i8.p0i8(i8* %a)
@@ -1178,7 +1178,7 @@
 
 // CHECK-LABEL: @test_vld1q_u16_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint16x8x2_t, align 16
-// CHECK-A32: %struct.uint16x8x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint16x8x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -1198,7 +1198,7 @@
 
 // CHECK-LABEL: @test_vld1q_u16_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint16x8x3_t, align 16
-// CHECK-A32: %struct.uint16x8x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint16x8x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -1218,7 +1218,7 @@
 
 // CHECK-LABEL: @test_vld1q_u16_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint16x8x4_t, align 16
-// CHECK-A32: %struct.uint16x8x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint16x8x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint16x8x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint16x8x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i16* %a to i8*
@@ -1238,7 +1238,7 @@
 
 // CHECK-LABEL: @test_vld1q_u32_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint32x4x2_t, align 16
-// CHECK-A32: %struct.uint32x4x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint32x4x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -1258,7 +1258,7 @@
 
 // CHECK-LABEL: @test_vld1q_u32_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint32x4x3_t, align 16
-// CHECK-A32: %struct.uint32x4x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint32x4x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -1278,7 +1278,7 @@
 
 // CHECK-LABEL: @test_vld1q_u32_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint32x4x4_t, align 16
-// CHECK-A32: %struct.uint32x4x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint32x4x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint32x4x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint32x4x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i32* %a to i8*
@@ -1298,7 +1298,7 @@
 
 // CHECK-LABEL: @test_vld1q_u64_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint64x2x2_t, align 16
-// CHECK-A32: %struct.uint64x2x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint64x2x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint64x2x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint64x2x2_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -1318,7 +1318,7 @@
 
 // CHECK-LABEL: @test_vld1q_u64_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint64x2x3_t, align 16
-// CHECK-A32: %struct.uint64x2x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint64x2x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint64x2x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint64x2x3_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -1338,7 +1338,7 @@
 
 // CHECK-LABEL: @test_vld1q_u64_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint64x2x4_t, align 16
-// CHECK-A32: %struct.uint64x2x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint64x2x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint64x2x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint64x2x4_t* [[__RET]] to i8*
 // CHECK: [[TMP1:%.*]] = bitcast i64* %a to i8*
@@ -1358,7 +1358,7 @@
 
 // CHECK-LABEL: @test_vld1q_u8_x2(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint8x16x2_t, align 16
-// CHECK-A32: %struct.uint8x16x2_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint8x16x2_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x2_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x2|arm.neon.vld1x2}}.v16i8.p0i8(i8* %a)
@@ -1376,7 +1376,7 @@
 
 // CHECK-LABEL: @test_vld1q_u8_x3(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint8x16x3_t, align 16
-// CHECK-A32: %struct.uint8x16x3_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint8x16x3_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x3_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x3_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x3|arm.neon.vld1x3}}.v16i8.p0i8(i8* %a)
@@ -1394,7 +1394,7 @@
 
 // CHECK-LABEL: @test_vld1q_u8_x4(
 // CHECK-A64: [[RETVAL:%.*]] = alloca %struct.uint8x16x4_t, align 16
-// CHECK-A32: %struct.uint8x16x4_t* noalias sret [[RETVAL:%.*]],
+// CHECK-A32: %struct.uint8x16x4_t* noalias sret align 8 [[RETVAL:%.*]],
 // CHECK: [[__RET:%.*]] = alloca %struct.uint8x16x4_t, align {{16|8}}
 // CHECK: [[TMP0:%.*]] = bitcast %struct.uint8x16x4_t* [[__RET]] to i8*
 // CHECK: [[VLD1XN:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.{{aarch64.neon.ld1x4|arm.neon.vld1x4}}.v16i8.p0i8(i8* %a)
Index: clang/test/CodeGen/arm-varargs.c
===================================================================
--- clang/test/CodeGen/arm-varargs.c
+++ clang/test/CodeGen/arm-varargs.c
@@ -24,7 +24,7 @@
 };
 
 struct bigstruct simple_struct(void) {
-// CHECK-LABEL: define void @simple_struct(%struct.bigstruct* noalias sret %agg.result)
+// CHECK-LABEL: define void @simple_struct(%struct.bigstruct* noalias sret align 4 %agg.result)
   return va_arg(the_list, struct bigstruct);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[NEXT:%[a-z0-9._]+]] = getelementptr inbounds i8, i8* [[CUR]], i32 40
@@ -42,7 +42,7 @@
 };
 
 struct aligned_bigstruct simple_aligned_struct(void) {
-// CHECK-LABEL: define void @simple_aligned_struct(%struct.aligned_bigstruct* noalias sret %agg.result)
+// CHECK-LABEL: define void @simple_aligned_struct(%struct.aligned_bigstruct* noalias sret align 8 %agg.result)
   return va_arg(the_list, struct aligned_bigstruct);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[CUR_INT:%[a-z0-9._]+]] = ptrtoint i8* [[CUR]] to i32
@@ -78,7 +78,7 @@
 };
 
 struct hfa simple_hfa(void) {
-// CHECK-LABEL: define void @simple_hfa(%struct.hfa* noalias sret %agg.result)
+// CHECK-LABEL: define void @simple_hfa(%struct.hfa* noalias sret align 4 %agg.result)
   return va_arg(the_list, struct hfa);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[NEXT:%[a-z0-9._]+]] = getelementptr inbounds i8, i8* [[CUR]], i32 8
@@ -185,7 +185,7 @@
   int val;
 } overaligned_int_struct;
 overaligned_int_struct overaligned_int_struct_test() {
-// CHECK-LABEL: define void @overaligned_int_struct_test(%struct.overaligned_int_struct* noalias sret %agg.result)
+// CHECK-LABEL: define void @overaligned_int_struct_test(%struct.overaligned_int_struct* noalias sret align 16 %agg.result)
   return va_arg(the_list, overaligned_int_struct);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[NEXT:%[a-z0-9._]+]] = getelementptr inbounds i8, i8* [[CUR]], i32 16
@@ -201,7 +201,7 @@
   long long val;
 } underaligned_long_long_struct;
 underaligned_long_long_struct underaligned_long_long_struct_test() {
-// CHECK-LABEL: define void @underaligned_long_long_struct_test(%struct.underaligned_long_long_struct* noalias sret %agg.result)
+// CHECK-LABEL: define void @underaligned_long_long_struct_test(%struct.underaligned_long_long_struct* noalias sret align 2 %agg.result)
   return va_arg(the_list, underaligned_long_long_struct);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[NEXT:%[a-z0-9._]+]] = getelementptr inbounds i8, i8* [[CUR]], i32 8
@@ -217,7 +217,7 @@
   long long val;
 } overaligned_long_long_struct;
 overaligned_long_long_struct overaligned_long_long_struct_test() {
-// CHECK-LABEL: define void @overaligned_long_long_struct_test(%struct.overaligned_long_long_struct* noalias sret %agg.result)
+// CHECK-LABEL: define void @overaligned_long_long_struct_test(%struct.overaligned_long_long_struct* noalias sret align 16 %agg.result)
   return va_arg(the_list, overaligned_long_long_struct);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[CUR_INT:%[a-z0-9._]+]] = ptrtoint i8* [[CUR]] to i32
@@ -259,7 +259,7 @@
   int val __attribute__((aligned(16)));
 } overaligned_int_struct_member;
 overaligned_int_struct_member overaligned_int_struct_member_test() {
-// CHECK-LABEL: define void @overaligned_int_struct_member_test(%struct.overaligned_int_struct_member* noalias sret %agg.result)
+// CHECK-LABEL: define void @overaligned_int_struct_member_test(%struct.overaligned_int_struct_member* noalias sret align 16 %agg.result)
   return va_arg(the_list, overaligned_int_struct_member);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[CUR_INT:%[a-z0-9._]+]] = ptrtoint i8* [[CUR]] to i32
@@ -279,7 +279,7 @@
   long long val __attribute__((packed,aligned(2)));
 } underaligned_long_long_struct_member;
 underaligned_long_long_struct_member underaligned_long_long_struct_member_test() {
-// CHECK-LABEL: define void @underaligned_long_long_struct_member_test(%struct.underaligned_long_long_struct_member* noalias sret %agg.result)
+// CHECK-LABEL: define void @underaligned_long_long_struct_member_test(%struct.underaligned_long_long_struct_member* noalias sret align 2 %agg.result)
   return va_arg(the_list, underaligned_long_long_struct_member);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[NEXT:%[a-z0-9._]+]] = getelementptr inbounds i8, i8* [[CUR]], i32 8
@@ -295,7 +295,7 @@
   long long val __attribute__((aligned(16)));
 } overaligned_long_long_struct_member;
 overaligned_long_long_struct_member overaligned_long_long_struct_member_test() {
-// CHECK-LABEL: define void @overaligned_long_long_struct_member_test(%struct.overaligned_long_long_struct_member* noalias sret %agg.result)
+// CHECK-LABEL: define void @overaligned_long_long_struct_member_test(%struct.overaligned_long_long_struct_member* noalias sret align 16 %agg.result)
   return va_arg(the_list, overaligned_long_long_struct_member);
 // CHECK: [[CUR:%[a-z0-9._]+]] = load i8*, i8** getelementptr inbounds (%struct.__va_list, %struct.__va_list* @the_list, i32 0, i32 0), align 4
 // CHECK: [[CUR_INT:%[a-z0-9._]+]] = ptrtoint i8* [[CUR]] to i32
Index: clang/test/CodeGen/arm-vector-arguments.c
===================================================================
--- clang/test/CodeGen/arm-vector-arguments.c
+++ clang/test/CodeGen/arm-vector-arguments.c
@@ -9,7 +9,7 @@
 
 #include <arm_neon.h>
 
-// CHECK: define void @f0(%struct.int8x16x2_t* noalias sret %agg.result, <16 x i8> %{{.*}}, <16 x i8> %{{.*}})
+// CHECK: define void @f0(%struct.int8x16x2_t* noalias sret align 16 %agg.result, <16 x i8> %{{.*}}, <16 x i8> %{{.*}})
 int8x16x2_t f0(int8x16_t a0, int8x16_t a1) {
   return vzipq_s8(a0, a1);
 }
@@ -25,7 +25,7 @@
 T_float32x2 f1_0(T_float32x2 a0) { return a0; }
 // CHECK: define <4 x float> @f1_1(<4 x float> %{{.*}})
 T_float32x4 f1_1(T_float32x4 a0) { return a0; }
-// CHECK: define void @f1_2(<8 x float>* noalias sret %{{.*}}, <8 x float> %{{.*}})
+// CHECK: define void @f1_2(<8 x float>* noalias sret align 32 %{{.*}}, <8 x float> %{{.*}})
 T_float32x8 f1_2(T_float32x8 a0) { return a0; }
-// CHECK: define void @f1_3(<16 x float>* noalias sret %{{.*}}, <16 x float> %{{.*}})
+// CHECK: define void @f1_3(<16 x float>* noalias sret align 64 %{{.*}}, <16 x float> %{{.*}})
 T_float32x16 f1_3(T_float32x16 a0) { return a0; }
Index: clang/test/CodeGen/arm-vfp16-arguments.c
===================================================================
--- clang/test/CodeGen/arm-vfp16-arguments.c
+++ clang/test/CodeGen/arm-vfp16-arguments.c
@@ -71,6 +71,6 @@
 
 hfa_t ghfa;
 hfa_t test_ret_hfa(void) { return ghfa; }
-// CHECK-SOFT: define void @test_ret_hfa(%struct.hfa_t* noalias nocapture sret %agg.result)
+// CHECK-SOFT: define void @test_ret_hfa(%struct.hfa_t* noalias nocapture sret align 8 %agg.result)
 // CHECK-HARD: define arm_aapcs_vfpcc [2 x <2 x i32>] @test_ret_hfa()
 // CHECK-FULL: define arm_aapcs_vfpcc %struct.hfa_t @test_ret_hfa()
Index: clang/test/CodeGen/arm-vfp16-arguments2.cpp
===================================================================
--- clang/test/CodeGen/arm-vfp16-arguments2.cpp
+++ clang/test/CodeGen/arm-vfp16-arguments2.cpp
@@ -37,27 +37,27 @@
   B1 M[1];
 };
 
-// CHECK-SOFT: define void @_Z2f12S1(%struct.S1* noalias nocapture sret %agg.result, [2 x i64] %s1.coerce)
+// CHECK-SOFT: define void @_Z2f12S1(%struct.S1* noalias nocapture sret align 8 %agg.result, [2 x i64] %s1.coerce)
 // CHECK-HARD: define arm_aapcs_vfpcc [2 x <2 x i32>] @_Z2f12S1([2 x <2 x i32>] returned %s1.coerce)
 // CHECK-FULL: define arm_aapcs_vfpcc %struct.S1 @_Z2f12S1(%struct.S1 returned %s1.coerce)
 struct S1 f1(struct S1 s1) { return s1; }
 
-// CHECK-SOFT: define void @_Z2f22S2(%struct.S2* noalias nocapture sret %agg.result, [4 x i32] %s2.coerce)
+// CHECK-SOFT: define void @_Z2f22S2(%struct.S2* noalias nocapture sret align 8 %agg.result, [4 x i32] %s2.coerce)
 // CHECK-HARD: define arm_aapcs_vfpcc [2 x <2 x i32>] @_Z2f22S2([2 x <2 x i32>] returned %s2.coerce)
 // CHECK-FULL: define arm_aapcs_vfpcc %struct.S2 @_Z2f22S2(%struct.S2 returned %s2.coerce)
 struct S2 f2(struct S2 s2) { return s2; }
 
-// CHECK-SOFT: define void @_Z2f32S3(%struct.S3* noalias nocapture sret %agg.result, [2 x i64] %s3.coerce)
+// CHECK-SOFT: define void @_Z2f32S3(%struct.S3* noalias nocapture sret align 8 %agg.result, [2 x i64] %s3.coerce)
 // CHECK-HARD: define arm_aapcs_vfpcc [2 x <2 x i32>] @_Z2f32S3([2 x <2 x i32>] returned %s3.coerce)
 // CHECK-FULL: define arm_aapcs_vfpcc %struct.S3 @_Z2f32S3(%struct.S3 returned %s3.coerce)
 struct S3 f3(struct S3 s3) { return s3; }
 
-// CHECK-SOFT: define void @_Z2f42S4(%struct.S4* noalias nocapture sret %agg.result, [2 x i64] %s4.coerce)
+// CHECK-SOFT: define void @_Z2f42S4(%struct.S4* noalias nocapture sret align 8 %agg.result, [2 x i64] %s4.coerce)
 // CHECK-HARD: define arm_aapcs_vfpcc [2 x <2 x i32>] @_Z2f42S4([2 x <2 x i32>] returned %s4.coerce)
 // CHECK-FULL: define arm_aapcs_vfpcc %struct.S4 @_Z2f42S4(%struct.S4 returned %s4.coerce)
 struct S4 f4(struct S4 s4) { return s4; }
 
-// CHECK-SOFT: define void @_Z2f52S5(%struct.S5* noalias nocapture sret %agg.result, [2 x i64] %s5.coerce)
+// CHECK-SOFT: define void @_Z2f52S5(%struct.S5* noalias nocapture sret align 8 %agg.result, [2 x i64] %s5.coerce)
 // CHECK-HARD: define arm_aapcs_vfpcc %struct.S5 @_Z2f52S5(%struct.S5 returned %s5.coerce)
 // CHECK-FULL: define arm_aapcs_vfpcc %struct.S5 @_Z2f52S5(%struct.S5 returned %s5.coerce)
 struct S5 f5(struct S5 s5) { return s5; }
Index: clang/test/CodeGen/arm64-arguments.c
===================================================================
--- clang/test/CodeGen/arm64-arguments.c
+++ clang/test/CodeGen/arm64-arguments.c
@@ -181,9 +181,9 @@
 // CHECK: define <4 x float> @f1_1(<4 x float> %{{.*}})
 T_float32x4 f1_1(T_float32x4 a0) { return a0; }
 // Vector with length bigger than 16-byte is illegal and is passed indirectly.
-// CHECK: define void @f1_2(<8 x float>* noalias sret  %{{.*}}, <8 x float>* %0)
+// CHECK: define void @f1_2(<8 x float>* noalias sret align 16 %{{.*}}, <8 x float>* %0)
 T_float32x8 f1_2(T_float32x8 a0) { return a0; }
-// CHECK: define void @f1_3(<16 x float>* noalias sret %{{.*}}, <16 x float>* %0)
+// CHECK: define void @f1_3(<16 x float>* noalias sret align 16 %{{.*}}, <16 x float>* %0)
 T_float32x16 f1_3(T_float32x16 a0) { return a0; }
 
 // Testing alignment with aggregates: HFA, aggregates with size <= 16 bytes and
Index: clang/test/CodeGen/arm64-microsoft-arguments.cpp
===================================================================
--- clang/test/CodeGen/arm64-microsoft-arguments.cpp
+++ clang/test/CodeGen/arm64-microsoft-arguments.cpp
@@ -28,8 +28,8 @@
 }
 
 // Pass and return for type size > 16 bytes.
-// CHECK: define {{.*}} void @{{.*}}f3{{.*}}(%struct.S3* noalias sret %agg.result)
-// CHECK: call void {{.*}}func3{{.*}}(%struct.S3* sret %agg.result, %struct.S3* %agg.tmp)
+// CHECK: define {{.*}} void @{{.*}}f3{{.*}}(%struct.S3* noalias sret align 4 %agg.result)
+// CHECK: call void {{.*}}func3{{.*}}(%struct.S3* sret align 4 %agg.result, %struct.S3* %agg.tmp)
 struct S3 {
   int a[5];
 };
@@ -42,8 +42,8 @@
 
 // Pass and return aggregate (of size < 16 bytes) with non-trivial destructor.
 // Passed directly but returned indirectly.
-// CHECK: define {{.*}} void {{.*}}f4{{.*}}(%struct.S4* inreg noalias sret %agg.result)
-// CHECK: call void {{.*}}func4{{.*}}(%struct.S4* inreg sret %agg.result, [2 x i64] %5)
+// CHECK: define {{.*}} void {{.*}}f4{{.*}}(%struct.S4* inreg noalias sret align 4 %agg.result)
+// CHECK: call void {{.*}}func4{{.*}}(%struct.S4* inreg sret align 4 %agg.result, [2 x i64] %5)
 struct S4 {
   int a[3];
   ~S4();
@@ -56,8 +56,8 @@
 }
 
 // Pass and return from instance method called from instance method.
-// CHECK: define {{.*}} void @{{.*}}bar@Q1{{.*}}(%class.Q1* %this, %class.P1* inreg noalias sret %agg.result)
-// CHECK: call void {{.*}}foo@P1{{.*}}(%class.P1* %ref.tmp, %class.P1* inreg sret %agg.result, i8 %1)
+// CHECK: define {{.*}} void @{{.*}}bar@Q1{{.*}}(%class.Q1* %this, %class.P1* inreg noalias sret align 1 %agg.result)
+// CHECK: call void {{.*}}foo@P1{{.*}}(%class.P1* %ref.tmp, %class.P1* inreg sret align 1 %agg.result, i8 %1)
 
 class P1 {
 public:
@@ -76,7 +76,7 @@
 
 // Pass and return from instance method called from free function.
 // CHECK: define {{.*}} void {{.*}}bar{{.*}}()
-// CHECK: call void {{.*}}foo@P2{{.*}}(%class.P2* %ref.tmp, %class.P2* inreg sret %retval, i8 %0)
+// CHECK: call void {{.*}}foo@P2{{.*}}(%class.P2* %ref.tmp, %class.P2* inreg sret align 1 %retval, i8 %0)
 class P2 {
 public:
   P2 foo(P2 x);
@@ -89,8 +89,8 @@
 
 // Pass and return an object with a user-provided constructor (passed directly,
 // returned indirectly)
-// CHECK: define {{.*}} void @{{.*}}f5{{.*}}(%struct.S5* inreg noalias sret %agg.result)
-// CHECK: call void {{.*}}func5{{.*}}(%struct.S5* inreg sret %agg.result, i64 {{.*}})
+// CHECK: define {{.*}} void @{{.*}}f5{{.*}}(%struct.S5* inreg noalias sret align 4 %agg.result)
+// CHECK: call void {{.*}}func5{{.*}}(%struct.S5* inreg sret align 4 %agg.result, i64 {{.*}})
 struct S5 {
   S5();
   int x;
@@ -146,8 +146,8 @@
   int y;
 };
 
-// CHECK: define {{.*}} void {{.*}}?f8{{.*}}(%struct.S8* inreg noalias sret {{.*}})
-// CHECK: call void {{.*}}func8{{.*}}(%struct.S8* inreg sret {{.*}}, i64 {{.*}})
+// CHECK: define {{.*}} void {{.*}}?f8{{.*}}(%struct.S8* inreg noalias sret align 4 {{.*}})
+// CHECK: call void {{.*}}func8{{.*}}(%struct.S8* inreg sret align 4 {{.*}}, i64 {{.*}})
 S8 func8(S8 x);
 S8 f8() {
   S8 x;
@@ -157,8 +157,8 @@
 
 // Pass and return an object with a non-trivial copy-assignment operator and
 // a trivial copy constructor (passed directly, returned indirectly)
-// CHECK: define {{.*}} void @"?f9@@YA?AUS9@@XZ"(%struct.S9* inreg noalias sret {{.*}})
-// CHECK: call void {{.*}}func9{{.*}}(%struct.S9* inreg sret {{.*}}, i64 {{.*}})
+// CHECK: define {{.*}} void @"?f9@@YA?AUS9@@XZ"(%struct.S9* inreg noalias sret align 4 {{.*}})
+// CHECK: call void {{.*}}func9{{.*}}(%struct.S9* inreg sret align 4 {{.*}}, i64 {{.*}})
 struct S9 {
   S9& operator=(const S9&);
   int x;
@@ -174,8 +174,8 @@
 
 // Pass and return an object with a base class (passed directly, returned
 // indirectly).
-// CHECK: define dso_local void {{.*}}f10{{.*}}(%struct.S10* inreg noalias sret {{.*}})
-// CHECK: call void {{.*}}func10{{.*}}(%struct.S10* inreg sret {{.*}}, [2 x i64] {{.*}})
+// CHECK: define dso_local void {{.*}}f10{{.*}}(%struct.S10* inreg noalias sret align 4 {{.*}})
+// CHECK: call void {{.*}}func10{{.*}}(%struct.S10* inreg sret align 4 {{.*}}, [2 x i64] {{.*}})
 struct S10 : public S1 {
   int x;
 };
@@ -189,8 +189,8 @@
 
 // Pass and return a non aggregate object exceeding > 128 bits (passed
 // indirectly, returned indirectly)
-// CHECK: define dso_local void {{.*}}f11{{.*}}(%struct.S11* inreg noalias sret {{.*}})
-// CHECK: call void {{.*}}func11{{.*}}(%struct.S11* inreg sret {{.*}}, %struct.S11* {{.*}})
+// CHECK: define dso_local void {{.*}}f11{{.*}}(%struct.S11* inreg noalias sret align 8 {{.*}})
+// CHECK: call void {{.*}}func11{{.*}}(%struct.S11* inreg sret align 8 {{.*}}, %struct.S11* {{.*}})
 struct S11 {
   virtual void f();
   int a[5];
Index: clang/test/CodeGen/arm64_32.c
===================================================================
--- clang/test/CodeGen/arm64_32.c
+++ clang/test/CodeGen/arm64_32.c
@@ -27,4 +27,4 @@
 
 typedef float __attribute__((ext_vector_type(16))) v16f32;
 v16f32 func(v16f32 in) { return in; }
-// CHECK: define void @func(<16 x float>* noalias sret {{%.*}}, <16 x float> {{%.*}})
+// CHECK: define void @func(<16 x float>* noalias sret align 16 {{%.*}}, <16 x float> {{%.*}})
Index: clang/test/CodeGen/arm_neon_intrinsics.c
===================================================================
--- clang/test/CodeGen/arm_neon_intrinsics.c
+++ clang/test/CodeGen/arm_neon_intrinsics.c
@@ -20223,7 +20223,7 @@
   return vtbx4_p8(a, b, c);
 }
 
-// CHECK: @test_vtrn_s8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_s8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VTRN_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
@@ -20236,7 +20236,7 @@
   return vtrn_s8(a, b);
 }
 
-// CHECK: @test_vtrn_s16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_s16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20251,7 +20251,7 @@
   return vtrn_s16(a, b);
 }
 
-// CHECK: @test_vtrn_s32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_s32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x i32> %b to <8 x i8>
@@ -20266,7 +20266,7 @@
   return vtrn_s32(a, b);
 }
 
-// CHECK: @test_vtrn_u8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_u8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VTRN_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
@@ -20279,7 +20279,7 @@
   return vtrn_u8(a, b);
 }
 
-// CHECK: @test_vtrn_u16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_u16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20294,7 +20294,7 @@
   return vtrn_u16(a, b);
 }
 
-// CHECK: @test_vtrn_u32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_u32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x i32> %b to <8 x i8>
@@ -20309,7 +20309,7 @@
   return vtrn_u32(a, b);
 }
 
-// CHECK: @test_vtrn_f32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_f32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x float> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x float> %b to <8 x i8>
@@ -20324,7 +20324,7 @@
   return vtrn_f32(a, b);
 }
 
-// CHECK: @test_vtrn_p8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_p8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VTRN_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
@@ -20337,7 +20337,7 @@
   return vtrn_p8(a, b);
 }
 
-// CHECK: @test_vtrn_p16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrn_p16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20352,7 +20352,7 @@
   return vtrn_p16(a, b);
 }
 
-// CHECK: @test_vtrnq_s8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_s8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[AGG_RESULT]] 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>
@@ -20365,7 +20365,7 @@
   return vtrnq_s8(a, b);
 }
 
-// CHECK: @test_vtrnq_s16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_s16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -20380,7 +20380,7 @@
   return vtrnq_s16(a, b);
 }
 
-// CHECK: @test_vtrnq_s32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_s32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8>
@@ -20395,7 +20395,7 @@
   return vtrnq_s32(a, b);
 }
 
-// CHECK: @test_vtrnq_u8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_u8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[AGG_RESULT]] 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>
@@ -20408,7 +20408,7 @@
   return vtrnq_u8(a, b);
 }
 
-// CHECK: @test_vtrnq_u16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_u16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -20423,7 +20423,7 @@
   return vtrnq_u16(a, b);
 }
 
-// CHECK: @test_vtrnq_u32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_u32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8>
@@ -20438,7 +20438,7 @@
   return vtrnq_u32(a, b);
 }
 
-// CHECK: @test_vtrnq_f32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_f32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x float> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x float> %b to <16 x i8>
@@ -20453,7 +20453,7 @@
   return vtrnq_f32(a, b);
 }
 
-// CHECK: @test_vtrnq_p8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_p8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[AGG_RESULT]] 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>
@@ -20466,7 +20466,7 @@
   return vtrnq_p8(a, b);
 }
 
-// CHECK: @test_vtrnq_p16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vtrnq_p16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -20645,7 +20645,7 @@
   return vtstq_p16(a, b);
 }
 
-// CHECK: @test_vuzp_s8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_s8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VUZP_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
@@ -20658,7 +20658,7 @@
   return vuzp_s8(a, b);
 }
 
-// CHECK: @test_vuzp_s16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_s16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20673,7 +20673,7 @@
   return vuzp_s16(a, b);
 }
 
-// CHECK: @test_vuzp_s32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_s32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x i32> %b to <8 x i8>
@@ -20688,7 +20688,7 @@
   return vuzp_s32(a, b);
 }
 
-// CHECK: @test_vuzp_u8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_u8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VUZP_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
@@ -20701,7 +20701,7 @@
   return vuzp_u8(a, b);
 }
 
-// CHECK: @test_vuzp_u16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_u16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20716,7 +20716,7 @@
   return vuzp_u16(a, b);
 }
 
-// CHECK: @test_vuzp_u32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_u32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x i32> %b to <8 x i8>
@@ -20731,7 +20731,7 @@
   return vuzp_u32(a, b);
 }
 
-// CHECK: @test_vuzp_f32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_f32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x float> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x float> %b to <8 x i8>
@@ -20746,7 +20746,7 @@
   return vuzp_f32(a, b);
 }
 
-// CHECK: @test_vuzp_p8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_p8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VUZP_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
@@ -20759,7 +20759,7 @@
   return vuzp_p8(a, b);
 }
 
-// CHECK: @test_vuzp_p16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzp_p16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20774,7 +20774,7 @@
   return vuzp_p16(a, b);
 }
 
-// CHECK: @test_vuzpq_s8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_s8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[AGG_RESULT]] 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>
@@ -20787,7 +20787,7 @@
   return vuzpq_s8(a, b);
 }
 
-// CHECK: @test_vuzpq_s16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_s16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -20802,7 +20802,7 @@
   return vuzpq_s16(a, b);
 }
 
-// CHECK: @test_vuzpq_s32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_s32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8>
@@ -20817,7 +20817,7 @@
   return vuzpq_s32(a, b);
 }
 
-// CHECK: @test_vuzpq_u8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_u8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[AGG_RESULT]] 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>
@@ -20830,7 +20830,7 @@
   return vuzpq_u8(a, b);
 }
 
-// CHECK: @test_vuzpq_u16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_u16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -20845,7 +20845,7 @@
   return vuzpq_u16(a, b);
 }
 
-// CHECK: @test_vuzpq_u32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_u32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8>
@@ -20860,7 +20860,7 @@
   return vuzpq_u32(a, b);
 }
 
-// CHECK: @test_vuzpq_f32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_f32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x float> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x float> %b to <16 x i8>
@@ -20875,7 +20875,7 @@
   return vuzpq_f32(a, b);
 }
 
-// CHECK: @test_vuzpq_p8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_p8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[AGG_RESULT]] 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>
@@ -20888,7 +20888,7 @@
   return vuzpq_p8(a, b);
 }
 
-// CHECK: @test_vuzpq_p16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vuzpq_p16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -20903,7 +20903,7 @@
   return vuzpq_p16(a, b);
 }
 
-// CHECK: @test_vzip_s8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_s8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VZIP_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
@@ -20916,7 +20916,7 @@
   return vzip_s8(a, b);
 }
 
-// CHECK: @test_vzip_s16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_s16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20931,7 +20931,7 @@
   return vzip_s16(a, b);
 }
 
-// CHECK: @test_vzip_s32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_s32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x i32> %b to <8 x i8>
@@ -20946,7 +20946,7 @@
   return vzip_s32(a, b);
 }
 
-// CHECK: @test_vzip_u8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_u8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VZIP_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
@@ -20959,7 +20959,7 @@
   return vzip_u8(a, b);
 }
 
-// CHECK: @test_vzip_u16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_u16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -20974,7 +20974,7 @@
   return vzip_u16(a, b);
 }
 
-// CHECK: @test_vzip_u32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_u32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x i32> %b to <8 x i8>
@@ -20989,7 +20989,7 @@
   return vzip_u32(a, b);
 }
 
-// CHECK: @test_vzip_f32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_f32({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x float> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <2 x float> %b to <8 x i8>
@@ -21004,7 +21004,7 @@
   return vzip_f32(a, b);
 }
 
-// CHECK: @test_vzip_p8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_p8({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to <8 x i8>*
 // CHECK:   [[VZIP_I:%.*]] = shufflevector <8 x i8> %a, <8 x i8> %b, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
@@ -21017,7 +21017,7 @@
   return vzip_p8(a, b);
 }
 
-// CHECK: @test_vzip_p16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzip_p16({{.*}} sret align 8 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %a to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i16> %b to <8 x i8>
@@ -21032,7 +21032,7 @@
   return vzip_p16(a, b);
 }
 
-// CHECK: @test_vzipq_s8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_s8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[AGG_RESULT]] 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>
@@ -21045,7 +21045,7 @@
   return vzipq_s8(a, b);
 }
 
-// CHECK: @test_vzipq_s16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_s16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -21060,7 +21060,7 @@
   return vzipq_s16(a, b);
 }
 
-// CHECK: @test_vzipq_s32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_s32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8>
@@ -21075,7 +21075,7 @@
   return vzipq_s32(a, b);
 }
 
-// CHECK: @test_vzipq_u8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_u8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[AGG_RESULT]] 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>
@@ -21088,7 +21088,7 @@
   return vzipq_u8(a, b);
 }
 
-// CHECK: @test_vzipq_u16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_u16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
@@ -21103,7 +21103,7 @@
   return vzipq_u16(a, b);
 }
 
-// CHECK: @test_vzipq_u32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_u32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x i32> %b to <16 x i8>
@@ -21118,7 +21118,7 @@
   return vzipq_u32(a, b);
 }
 
-// CHECK: @test_vzipq_f32({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_f32({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x float> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <4 x float> %b to <16 x i8>
@@ -21133,7 +21133,7 @@
   return vzipq_f32(a, b);
 }
 
-// CHECK: @test_vzipq_p8({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_p8({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[AGG_RESULT]] 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>
@@ -21146,7 +21146,7 @@
   return vzipq_p8(a, b);
 }
 
-// CHECK: @test_vzipq_p16({{.*}} sret [[AGG_RESULT:%[0-9a-zA-Z.]+]],
+// CHECK: @test_vzipq_p16({{.*}} sret align 16 [[AGG_RESULT:%[0-9a-zA-Z.]+]],
 // CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[AGG_RESULT]] to i8*
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %a to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i16> %b to <16 x i8>
Index: clang/test/CodeGen/blocks.c
===================================================================
--- clang/test/CodeGen/blocks.c
+++ clang/test/CodeGen/blocks.c
@@ -18,7 +18,7 @@
   int a[64];
 };
 
-// CHECK: define internal void @__f2_block_invoke(%struct.s0* noalias sret {{%.*}}, i8* {{%.*}}, %struct.s0* byval(%struct.s0) align 4 {{.*}})
+// CHECK: define internal void @__f2_block_invoke(%struct.s0* noalias sret align 4 {{%.*}}, i8* {{%.*}}, %struct.s0* byval(%struct.s0) align 4 {{.*}})
 struct s0 f2(struct s0 a0) {
   return ^(struct s0 a1){ return a1; }(a0);
 }
Index: clang/test/CodeGen/c11atomics-ios.c
===================================================================
--- clang/test/CodeGen/c11atomics-ios.c
+++ clang/test/CodeGen/c11atomics-ios.c
@@ -203,7 +203,7 @@
 }
 
 PS test_promoted_load(_Atomic(PS) *addr) {
-  // CHECK-LABEL: @test_promoted_load(%struct.PS* noalias sret %agg.result, { %struct.PS, [2 x i8] }* %addr)
+  // CHECK-LABEL: @test_promoted_load(%struct.PS* noalias sret align 2 %agg.result, { %struct.PS, [2 x i8] }* %addr)
   // CHECK:   [[ADDR_ARG:%.*]] = alloca { %struct.PS, [2 x i8] }*, align 4
   // CHECK:   [[ATOMIC_RES:%.*]] = alloca { %struct.PS, [2 x i8] }, align 8
   // CHECK:   store { %struct.PS, [2 x i8] }* %addr, { %struct.PS, [2 x i8] }** [[ADDR_ARG]], align 4
@@ -245,7 +245,7 @@
 }
 
 PS test_promoted_exchange(_Atomic(PS) *addr, PS *val) {
-  // CHECK-LABEL: @test_promoted_exchange(%struct.PS* noalias sret %agg.result, { %struct.PS, [2 x i8] }* %addr, %struct.PS* %val)
+  // CHECK-LABEL: @test_promoted_exchange(%struct.PS* noalias sret align 2 %agg.result, { %struct.PS, [2 x i8] }* %addr, %struct.PS* %val)
   // CHECK:   [[ADDR_ARG:%.*]] = alloca { %struct.PS, [2 x i8] }*, align 4
   // CHECK:   [[VAL_ARG:%.*]] = alloca %struct.PS*, align 4
   // CHECK:   [[NONATOMIC_TMP:%.*]] = alloca %struct.PS, align 2
Index: clang/test/CodeGen/c11atomics.c
===================================================================
--- clang/test/CodeGen/c11atomics.c
+++ clang/test/CodeGen/c11atomics.c
@@ -368,7 +368,7 @@
 }
 
 PS test_promoted_load(_Atomic(PS) *addr) {
-  // CHECK-LABEL: @test_promoted_load(%struct.PS* noalias sret %agg.result, { %struct.PS, [2 x i8] }* %addr)
+  // CHECK-LABEL: @test_promoted_load(%struct.PS* noalias sret align 2 %agg.result, { %struct.PS, [2 x i8] }* %addr)
   // CHECK:   [[ADDR_ARG:%.*]] = alloca { %struct.PS, [2 x i8] }*, align 4
   // CHECK:   [[ATOMIC_RES:%.*]] = alloca { %struct.PS, [2 x i8] }, align 8
   // CHECK:   store { %struct.PS, [2 x i8] }* %addr, { %struct.PS, [2 x i8] }** [[ADDR_ARG]], align 4
@@ -411,7 +411,7 @@
 }
 
 PS test_promoted_exchange(_Atomic(PS) *addr, PS *val) {
-  // CHECK-LABEL: @test_promoted_exchange(%struct.PS* noalias sret %agg.result, { %struct.PS, [2 x i8] }* %addr, %struct.PS* %val)
+  // CHECK-LABEL: @test_promoted_exchange(%struct.PS* noalias sret align 2 %agg.result, { %struct.PS, [2 x i8] }* %addr, %struct.PS* %val)
   // CHECK:   [[ADDR_ARG:%.*]] = alloca { %struct.PS, [2 x i8] }*, align 4
   // CHECK:   [[VAL_ARG:%.*]] = alloca %struct.PS*, align 4
   // CHECK:   [[NONATOMIC_TMP:%.*]] = alloca %struct.PS, align 2
Index: clang/test/CodeGen/lanai-arguments.c
===================================================================
--- clang/test/CodeGen/lanai-arguments.c
+++ clang/test/CodeGen/lanai-arguments.c
@@ -16,7 +16,7 @@
 typedef struct {
   int cc;
 } s2;
-// CHECK: define void @f2(%struct.s2* noalias sret %agg.result)
+// CHECK: define void @f2(%struct.s2* noalias sret align 4 %agg.result)
 s2 f2() {
   s2 foo;
   return foo;
@@ -26,7 +26,7 @@
   int cc;
   int dd;
 } s3;
-// CHECK: define void @f3(%struct.s3* noalias sret %agg.result)
+// CHECK: define void @f3(%struct.s3* noalias sret align 4 %agg.result)
 s3 f3() {
   s3 foo;
   return foo;
Index: clang/test/CodeGen/le32-arguments.c
===================================================================
--- clang/test/CodeGen/le32-arguments.c
+++ clang/test/CodeGen/le32-arguments.c
@@ -17,7 +17,7 @@
   int cc;
 } s2;
 // Structs should be returned sret and not simplified by the frontend
-// CHECK-LABEL: define void @f2(%struct.s2* noalias sret %agg.result)
+// CHECK-LABEL: define void @f2(%struct.s2* noalias sret align 4 %agg.result)
 s2 f2() {
   s2 foo;
   return foo;
Index: clang/test/CodeGen/mcu-struct-return.c
===================================================================
--- clang/test/CodeGen/mcu-struct-return.c
+++ clang/test/CodeGen/mcu-struct-return.c
@@ -42,7 +42,7 @@
 struct S2 bar2() { return s2; }
 struct S1 bar3(union U1 u) { return s1; }
 // CHECK: define void @foo1()
-// CHECK: define void @foo2([[UNION2_TYPE]]* noalias sret %{{.+}})
+// CHECK: define void @foo2([[UNION2_TYPE]]* noalias sret align 4 %{{.+}})
 // CHECK: define i32 @foo3()
 // CHECK: define void @bar1()
 // CHECK: define i32 @bar2()
@@ -62,7 +62,7 @@
   // CHECK: [[Y1:%.+]] = alloca [[STRUCT1_TYPE]]
   // CHECK: [[Y2:%.+]] = alloca [[STRUCT2_TYPE]]
   // CHECK: call void @foo1()
-  // CHECK: call void @foo2([[UNION2_TYPE]]* sret [[X2]])
+  // CHECK: call void @foo2([[UNION2_TYPE]]* sret align 4 [[X2]])
   // CHECK: {{.+}} = call i32 @foo3()
   // CHECK: call void @bar1()
   // CHECK: {{.+}} = call i32 @bar2()
Index: clang/test/CodeGen/mingw-long-double.c
===================================================================
--- clang/test/CodeGen/mingw-long-double.c
+++ clang/test/CodeGen/mingw-long-double.c
@@ -32,15 +32,15 @@
   return x * x;
 }
 // GNU32: define dso_local x86_fp80 @TestLD(x86_fp80 %x)
-// GNU64: define dso_local void @TestLD(x86_fp80* noalias sret %agg.result, x86_fp80* %0)
+// GNU64: define dso_local void @TestLD(x86_fp80* noalias sret align 16 %agg.result, x86_fp80* %0)
 // MSC64: define dso_local double @TestLD(double %x)
 
 long double _Complex TestLDC(long double _Complex x) {
   return x * x;
 }
-// GNU32: define dso_local void @TestLDC({ x86_fp80, x86_fp80 }* noalias sret %agg.result, { x86_fp80, x86_fp80 }* byval({ x86_fp80, x86_fp80 }) align 4 %x)
-// GNU64: define dso_local void @TestLDC({ x86_fp80, x86_fp80 }* noalias sret %agg.result, { x86_fp80, x86_fp80 }* %x)
-// MSC64: define dso_local void @TestLDC({ double, double }* noalias sret %agg.result, { double, double }* %x)
+// GNU32: define dso_local void @TestLDC({ x86_fp80, x86_fp80 }* noalias sret align 4 %agg.result, { x86_fp80, x86_fp80 }* byval({ x86_fp80, x86_fp80 }) align 4 %x)
+// GNU64: define dso_local void @TestLDC({ x86_fp80, x86_fp80 }* noalias sret align 16 %agg.result, { x86_fp80, x86_fp80 }* %x)
+// MSC64: define dso_local void @TestLDC({ double, double }* noalias sret align 8 %agg.result, { double, double }* %x)
 
 // GNU32: declare dso_local void @__mulxc3
 // GNU64: declare dso_local void @__mulxc3
Index: clang/test/CodeGen/mips-zero-sized-struct.c
===================================================================
--- clang/test/CodeGen/mips-zero-sized-struct.c
+++ clang/test/CodeGen/mips-zero-sized-struct.c
@@ -19,7 +19,7 @@
 // RUN: %clang_cc1 -triple mipsisa64r6-unknown-linux-gnuabi64 -S -emit-llvm -o - %s | FileCheck -check-prefix=N64 %s
 // RUN: %clang_cc1 -triple mipsisa64r6el-unknown-linux-gnuabi64 -S -emit-llvm -o - %s | FileCheck -check-prefix=N64 %s
 
-// O32: define void @fn28(%struct.T2* noalias sret %agg.result, i8 signext %arg0)
+// O32: define void @fn28(%struct.T2* noalias sret align 1 %agg.result, i8 signext %arg0)
 // N32: define void @fn28(i8 signext %arg0)
 // N64: define void @fn28(i8 signext %arg0)
 
Index: clang/test/CodeGen/mips64-padding-arg.c
===================================================================
--- clang/test/CodeGen/mips64-padding-arg.c
+++ clang/test/CodeGen/mips64-padding-arg.c
@@ -33,9 +33,9 @@
 
 // Insert padding after hidden argument.
 //
-// N64-LABEL: define void @foo5(%struct.S0* noalias sret %agg.result, i64 %0, fp128 %a0)
-// N64: call void @foo6(%struct.S0* sret %agg.result, i32 signext 1, i32 signext 2, i64 undef, fp128 %a0)
-// N64: declare void @foo6(%struct.S0* sret, i32 signext, i32 signext, i64, fp128)
+// N64-LABEL: define void @foo5(%struct.S0* noalias sret align 16 %agg.result, i64 %0, fp128 %a0)
+// N64: call void @foo6(%struct.S0* sret align 16 %agg.result, i32 signext 1, i32 signext 2, i64 undef, fp128 %a0)
+// N64: declare void @foo6(%struct.S0* sret align 16, i32 signext, i32 signext, i64, fp128)
 
 extern S0 foo6(int, int, long double);
 
Index: clang/test/CodeGen/ms_abi.c
===================================================================
--- clang/test/CodeGen/ms_abi.c
+++ clang/test/CodeGen/ms_abi.c
@@ -155,7 +155,7 @@
 };
 
 __attribute__((ms_abi)) struct i128 f7(struct i128 a) {
-  // WIN64: define dso_local void @f7(%struct.i128* noalias sret %agg.result, %struct.i128* %a)
-  // FREEBSD: define win64cc void @f7(%struct.i128* noalias sret %agg.result, %struct.i128* %a)
+  // WIN64: define dso_local void @f7(%struct.i128* noalias sret align 8 %agg.result, %struct.i128* %a)
+  // FREEBSD: define win64cc void @f7(%struct.i128* noalias sret align 8 %agg.result, %struct.i128* %a)
   return a;
 }
Index: clang/test/CodeGen/ppc64-align-struct.c
===================================================================
--- clang/test/CodeGen/ppc64-align-struct.c
+++ clang/test/CodeGen/ppc64-align-struct.c
@@ -48,7 +48,7 @@
 {
 }
 
-// CHECK: define void @test1va(%struct.test1* noalias sret %[[AGG_RESULT:.*]], i32 signext %x, ...)
+// CHECK: define void @test1va(%struct.test1* noalias sret align 4 %[[AGG_RESULT:.*]], i32 signext %x, ...)
 // CHECK: %[[CUR:[^ ]+]] = load i8*, i8** %ap
 // CHECK: %[[NEXT:[^ ]+]] = getelementptr inbounds i8, i8* %[[CUR]], i64 8
 // CHECK: store i8* %[[NEXT]], i8** %ap
@@ -66,7 +66,7 @@
   return y;
 }
 
-// CHECK: define void @test2va(%struct.test2* noalias sret %[[AGG_RESULT:.*]], i32 signext %x, ...)
+// CHECK: define void @test2va(%struct.test2* noalias sret align 16 %[[AGG_RESULT:.*]], i32 signext %x, ...)
 // CHECK: %[[CUR:[^ ]+]] = load i8*, i8** %ap
 // CHECK: %[[TMP0:[^ ]+]] = ptrtoint i8* %[[CUR]] to i64
 // CHECK: %[[TMP1:[^ ]+]] = add i64 %[[TMP0]], 15
@@ -88,7 +88,7 @@
   return y;
 }
 
-// CHECK: define void @test3va(%struct.test3* noalias sret %[[AGG_RESULT:.*]], i32 signext %x, ...)
+// CHECK: define void @test3va(%struct.test3* noalias sret align 32 %[[AGG_RESULT:.*]], i32 signext %x, ...)
 // CHECK: %[[CUR:[^ ]+]] = load i8*, i8** %ap
 // CHECK: %[[TMP0:[^ ]+]] = ptrtoint i8* %[[CUR]] to i64
 // CHECK: %[[TMP1:[^ ]+]] = add i64 %[[TMP0]], 15
@@ -110,7 +110,7 @@
   return y;
 }
 
-// CHECK: define void @test4va(%struct.test4* noalias sret %[[AGG_RESULT:.*]], i32 signext %x, ...)
+// CHECK: define void @test4va(%struct.test4* noalias sret align 4 %[[AGG_RESULT:.*]], i32 signext %x, ...)
 // CHECK: %[[CUR:[^ ]+]] = load i8*, i8** %ap
 // CHECK: %[[NEXT:[^ ]+]] = getelementptr inbounds i8, i8* %[[CUR]], i64 16
 // CHECK: store i8* %[[NEXT]], i8** %ap
@@ -128,7 +128,7 @@
   return y;
 }
 
-// CHECK: define void @testva_longdouble(%struct.test_longdouble* noalias sret %[[AGG_RESULT:.*]], i32 signext %x, ...)
+// CHECK: define void @testva_longdouble(%struct.test_longdouble* noalias sret align 16 %[[AGG_RESULT:.*]], i32 signext %x, ...)
 // CHECK: %[[CUR:[^ ]+]] = load i8*, i8** %ap
 // CHECK: %[[NEXT:[^ ]+]] = getelementptr inbounds i8, i8* %[[CUR]], i64 16
 // CHECK: store i8* %[[NEXT]], i8** %ap
@@ -147,7 +147,7 @@
   return y;
 }
 
-// CHECK: define void @testva_vector(%struct.test_vector* noalias sret %[[AGG_RESULT:.*]], i32 signext %x, ...)
+// CHECK: define void @testva_vector(%struct.test_vector* noalias sret align 16 %[[AGG_RESULT:.*]], i32 signext %x, ...)
 // CHECK: %[[CUR:[^ ]+]] = load i8*, i8** %ap
 // CHECK: %[[TMP0:[^ ]+]] = ptrtoint i8* %[[CUR]] to i64
 // CHECK: %[[TMP1:[^ ]+]] = add i64 %[[TMP0]], 15
Index: clang/test/CodeGen/ppc64-elf-abi.c
===================================================================
--- clang/test/CodeGen/ppc64-elf-abi.c
+++ clang/test/CodeGen/ppc64-elf-abi.c
@@ -17,7 +17,7 @@
 // RUN: %clang_cc1 -triple powerpc64le-unknown-linux-gnu -emit-llvm -o - %s \
 // RUN:   -target-abi elfv2 | FileCheck %s --check-prefix=CHECK-ELFv2
 
-// CHECK-ELFv1: define void @func_fab(%struct.fab* noalias sret %agg.result, i64 %x.coerce)
+// CHECK-ELFv1: define void @func_fab(%struct.fab* noalias sret align 4 %agg.result, i64 %x.coerce)
 // CHECK-ELFv2: define [2 x float] @func_fab([2 x float] %x.coerce)
 struct fab { float a; float b; };
 struct fab func_fab(struct fab x) { return x; }
Index: clang/test/CodeGen/ppc64-qpx-vector.c
===================================================================
--- clang/test/CodeGen/ppc64-qpx-vector.c
+++ clang/test/CodeGen/ppc64-qpx-vector.c
@@ -24,6 +24,6 @@
 // QPX-LABEL: define <4 x double> @foo2(<4 x double> inreg %a.coerce, <4 x double> %b, [2 x i256] %c.coerce)
 // QPX: ret <4 x double>
 
-// NORMAL-LABEL: define void @foo2(<4 x double>* noalias sret %agg.result, [2 x i128] %a.coerce, <4 x double>* %0, [4 x i128] %c.coerce)
+// NORMAL-LABEL: define void @foo2(<4 x double>* noalias sret align 32 %agg.result, [2 x i128] %a.coerce, <4 x double>* %0, [4 x i128] %c.coerce)
 // NORMAL: ret void
 
Index: clang/test/CodeGen/ppc64-soft-float.c
===================================================================
--- clang/test/CodeGen/ppc64-soft-float.c
+++ clang/test/CodeGen/ppc64-soft-float.c
@@ -30,53 +30,53 @@
 struct f2a2b { float a[2]; float b[2]; };
 
 // CHECK-LE: define i32 @func_f1(float inreg %x.coerce)
-// CHECK-BE: define void @func_f1(%struct.f1* noalias sret %agg.result, float inreg %x.coerce)
+// CHECK-BE: define void @func_f1(%struct.f1* noalias sret align 4 %agg.result, float inreg %x.coerce)
 struct f1 func_f1(struct f1 x) { return x; }
 
 // CHECK-LE: define i64 @func_f2(i64 %x.coerce)
-// CHECK-BE: define void @func_f2(%struct.f2* noalias sret %agg.result, i64 %x.coerce)
+// CHECK-BE: define void @func_f2(%struct.f2* noalias sret align 4 %agg.result, i64 %x.coerce)
 struct f2 func_f2(struct f2 x) { return x; }
 
 // CHECK-LE: define { i64, i64 } @func_f3([2 x i64] %x.coerce)
-// CHECK-BE: define void @func_f3(%struct.f3* noalias sret %agg.result, [2 x i64] %x.coerce)
+// CHECK-BE: define void @func_f3(%struct.f3* noalias sret align 4 %agg.result, [2 x i64] %x.coerce)
 struct f3 func_f3(struct f3 x) { return x; }
 
 // CHECK-LE: define { i64, i64 } @func_f4([2 x i64] %x.coerce)
-// CHECK-BE: define void @func_f4(%struct.f4* noalias sret %agg.result, [2 x i64] %x.coerce)
+// CHECK-BE: define void @func_f4(%struct.f4* noalias sret align 4 %agg.result, [2 x i64] %x.coerce)
 struct f4 func_f4(struct f4 x) { return x; }
 
-// CHECK: define void @func_f5(%struct.f5* noalias sret %agg.result, [3 x i64] %x.coerce)
+// CHECK: define void @func_f5(%struct.f5* noalias sret align 4 %agg.result, [3 x i64] %x.coerce)
 struct f5 func_f5(struct f5 x) { return x; }
 
-// CHECK: define void @func_f6(%struct.f6* noalias sret %agg.result, [3 x i64] %x.coerce)
+// CHECK: define void @func_f6(%struct.f6* noalias sret align 4 %agg.result, [3 x i64] %x.coerce)
 struct f6 func_f6(struct f6 x) { return x; }
 
-// CHECK: define void @func_f7(%struct.f7* noalias sret %agg.result, [4 x i64] %x.coerce)
+// CHECK: define void @func_f7(%struct.f7* noalias sret align 4 %agg.result, [4 x i64] %x.coerce)
 struct f7 func_f7(struct f7 x) { return x; }
 
-// CHECK: define void @func_f8(%struct.f8* noalias sret %agg.result, [4 x i64] %x.coerce)
+// CHECK: define void @func_f8(%struct.f8* noalias sret align 4 %agg.result, [4 x i64] %x.coerce)
 struct f8 func_f8(struct f8 x) { return x; }
 
-// CHECK: define void @func_f9(%struct.f9* noalias sret %agg.result, [5 x i64] %x.coerce)
+// CHECK: define void @func_f9(%struct.f9* noalias sret align 4 %agg.result, [5 x i64] %x.coerce)
 struct f9 func_f9(struct f9 x) { return x; }
 
 // CHECK-LE: define i64 @func_fab(i64 %x.coerce)
-// CHECK-BE: define void @func_fab(%struct.fab* noalias sret %agg.result, i64 %x.coerce)
+// CHECK-BE: define void @func_fab(%struct.fab* noalias sret align 4 %agg.result, i64 %x.coerce)
 struct fab func_fab(struct fab x) { return x; }
 
 // CHECK-LE: define { i64, i64 } @func_fabc([2 x i64] %x.coerce)
-// CHECK-BE: define void @func_fabc(%struct.fabc* noalias sret %agg.result, [2 x i64] %x.coerce)
+// CHECK-BE: define void @func_fabc(%struct.fabc* noalias sret align 4 %agg.result, [2 x i64] %x.coerce)
 struct fabc func_fabc(struct fabc x) { return x; }
 
 // CHECK-LE: define { i64, i64 } @func_f2a2b([2 x i64] %x.coerce)
-// CHECK-BE: define void @func_f2a2b(%struct.f2a2b* noalias sret %agg.result, [2 x i64] %x.coerce)
+// CHECK-BE: define void @func_f2a2b(%struct.f2a2b* noalias sret align 4 %agg.result, [2 x i64] %x.coerce)
 struct f2a2b func_f2a2b(struct f2a2b x) { return x; }
 
 // CHECK-LABEL: @call_f1
 // CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f1, align 4
 // CHECK: %[[TMP:[^ ]+]] = load float, float* getelementptr inbounds (%struct.f1, %struct.f1* @global_f1, i32 0, i32 0, i32 0), align 4
 // CHECK-LE: call i32 @func_f1(float inreg %[[TMP]])
-// CHECK-BE: call void @func_f1(%struct.f1* sret %[[TMP0]], float inreg %[[TMP]])
+// CHECK-BE: call void @func_f1(%struct.f1* sret align 4 %[[TMP0]], float inreg %[[TMP]])
 struct f1 global_f1;
 void call_f1(void) { global_f1 = func_f1(global_f1); }
 
@@ -84,7 +84,7 @@
 // CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f2, align 4
 // CHECK: %[[TMP:[^ ]+]] = load i64, i64* bitcast (%struct.f2* @global_f2 to i64*), align 4
 // CHECK-LE: call i64 @func_f2(i64 %[[TMP]])
-// CHECK-BE: call void @func_f2(%struct.f2* sret %[[TMP0]], i64 %[[TMP]])
+// CHECK-BE: call void @func_f2(%struct.f2* sret align 4 %[[TMP0]], i64 %[[TMP]])
 struct f2 global_f2;
 void call_f2(void) { global_f2 = func_f2(global_f2); }
 
@@ -95,7 +95,7 @@
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %[[TMP2]], i8* align 4 bitcast (%struct.f3* @global_f3 to i8*), i64 12, i1 false)
 // CHECK: %[[TMP3:[^ ]+]] = load [2 x i64], [2 x i64]* %[[TMP1]]
 // CHECK-LE: call { i64, i64 } @func_f3([2 x i64] %[[TMP3]])
-// CHECK-BE: call void @func_f3(%struct.f3* sret %[[TMP0]], [2 x i64] %[[TMP3]])
+// CHECK-BE: call void @func_f3(%struct.f3* sret align 4 %[[TMP0]], [2 x i64] %[[TMP3]])
 struct f3 global_f3;
 void call_f3(void) { global_f3 = func_f3(global_f3); }
 
@@ -103,7 +103,7 @@
 // CHECK-BE: %[[TMP0:[^ ]+]] = alloca %struct.f4, align 4
 // CHECK: %[[TMP:[^ ]+]] = load [2 x i64], [2 x i64]* bitcast (%struct.f4* @global_f4 to [2 x i64]*), align 4
 // CHECK-LE: call { i64, i64 } @func_f4([2 x i64] %[[TMP]])
-// CHECK-BE: call void @func_f4(%struct.f4* sret %[[TMP0]], [2 x i64] %[[TMP]])
+// CHECK-BE: call void @func_f4(%struct.f4* sret align 4 %[[TMP0]], [2 x i64] %[[TMP]])
 struct f4 global_f4;
 void call_f4(void) { global_f4 = func_f4(global_f4); }
 
@@ -113,14 +113,14 @@
 // CHECK: %[[TMP2:[^ ]+]] = bitcast [3 x i64]* %[[TMP1]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %[[TMP2]], i8* align 4 bitcast (%struct.f5* @global_f5 to i8*), i64 20, i1 false)
 // CHECK: %[[TMP3:[^ ]+]] = load [3 x i64], [3 x i64]* %[[TMP1]]
-// CHECK: call void @func_f5(%struct.f5* sret %[[TMP0]], [3 x i64] %[[TMP3]])
+// CHECK: call void @func_f5(%struct.f5* sret align 4 %[[TMP0]], [3 x i64] %[[TMP3]])
 struct f5 global_f5;
 void call_f5(void) { global_f5 = func_f5(global_f5); }
 
 // CHECK-LABEL: @call_f6
 // CHECK: %[[TMP0:[^ ]+]] = alloca %struct.f6, align 4
 // CHECK: %[[TMP:[^ ]+]] = load [3 x i64], [3 x i64]* bitcast (%struct.f6* @global_f6 to [3 x i64]*), align 4
-// CHECK: call void @func_f6(%struct.f6* sret %[[TMP0]], [3 x i64] %[[TMP]])
+// CHECK: call void @func_f6(%struct.f6* sret align 4 %[[TMP0]], [3 x i64] %[[TMP]])
 struct f6 global_f6;
 void call_f6(void) { global_f6 = func_f6(global_f6); }
 
@@ -130,14 +130,14 @@
 // CHECK: %[[TMP2:[^ ]+]] = bitcast [4 x i64]* %[[TMP1]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %[[TMP2]], i8* align 4 bitcast (%struct.f7* @global_f7 to i8*), i64 28, i1 false)
 // CHECK: %[[TMP3:[^ ]+]] = load [4 x i64], [4 x i64]* %[[TMP1]], align 8
-// CHECK: call void @func_f7(%struct.f7* sret %[[TMP0]], [4 x i64] %[[TMP3]])
+// CHECK: call void @func_f7(%struct.f7* sret align 4 %[[TMP0]], [4 x i64] %[[TMP3]])
 struct f7 global_f7;
 void call_f7(void) { global_f7 = func_f7(global_f7); }
 
 // CHECK-LABEL: @call_f8
 // CHECK: %[[TMP0:[^ ]+]] = alloca %struct.f8, align 4
 // CHECK: %[[TMP:[^ ]+]] = load [4 x i64], [4 x i64]* bitcast (%struct.f8* @global_f8 to [4 x i64]*), align 4
-// CHECK: call void @func_f8(%struct.f8* sret %[[TMP0]], [4 x i64] %[[TMP]])
+// CHECK: call void @func_f8(%struct.f8* sret align 4 %[[TMP0]], [4 x i64] %[[TMP]])
 struct f8 global_f8;
 void call_f8(void) { global_f8 = func_f8(global_f8); }
 
@@ -146,7 +146,7 @@
 // CHECK: %[[TMP2:[^ ]+]] = bitcast [5 x i64]* %[[TMP1]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %[[TMP2]], i8* align 4 bitcast (%struct.f9* @global_f9 to i8*), i64 36, i1 false)
 // CHECK: %[[TMP3:[^ ]+]] = load [5 x i64], [5 x i64]* %[[TMP1]]
-// CHECK: call void @func_f9(%struct.f9* sret %{{[^ ]+}}, [5 x i64] %[[TMP3]])
+// CHECK: call void @func_f9(%struct.f9* sret align 4 %{{[^ ]+}}, [5 x i64] %[[TMP3]])
 struct f9 global_f9;
 void call_f9(void) { global_f9 = func_f9(global_f9); }
 
@@ -154,7 +154,7 @@
 // CHECK: %[[TMP0:[^ ]+]] = alloca %struct.fab, align 4
 // CHECK: %[[TMP:[^ ]+]] = load i64, i64* bitcast (%struct.fab* @global_fab to i64*), align 4
 // CHECK-LE: %call = call i64 @func_fab(i64 %[[TMP]])
-// CHECK-BE: call void @func_fab(%struct.fab* sret %[[TMP0]], i64 %[[TMP]])
+// CHECK-BE: call void @func_fab(%struct.fab* sret align 4 %[[TMP0]], i64 %[[TMP]])
 struct fab global_fab;
 void call_fab(void) { global_fab = func_fab(global_fab); }
 
@@ -165,7 +165,7 @@
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %[[TMP2]], i8* align 4 bitcast (%struct.fabc* @global_fabc to i8*), i64 12, i1 false)
 // CHECK: %[[TMP3:[^ ]+]] = load [2 x i64], [2 x i64]* %[[TMP0]], align 8
 // CHECK-LE: %call = call { i64, i64 } @func_fabc([2 x i64] %[[TMP3]])
-// CHECK-BE: call void @func_fabc(%struct.fabc* sret %[[TMPX]], [2 x i64] %[[TMP3]])
+// CHECK-BE: call void @func_fabc(%struct.fabc* sret align 4 %[[TMPX]], [2 x i64] %[[TMP3]])
 struct fabc global_fabc;
 void call_fabc(void) { global_fabc = func_fabc(global_fabc); }
 
Index: clang/test/CodeGen/ppc64-vector.c
===================================================================
--- clang/test/CodeGen/ppc64-vector.c
+++ clang/test/CodeGen/ppc64-vector.c
@@ -39,13 +39,13 @@
   return x;
 }
 
-// CHECK: define void @test_v16i16(<16 x i16>* noalias sret %agg.result, <16 x i16>* %0)
+// CHECK: define void @test_v16i16(<16 x i16>* noalias sret align 32 %agg.result, <16 x i16>* %0)
 v16i16 test_v16i16(v16i16 x)
 {
   return x;
 }
 
-// CHECK: define void @test_struct_v16i16(%struct.v16i16* noalias sret %agg.result, [2 x i128] %x.coerce)
+// CHECK: define void @test_struct_v16i16(%struct.v16i16* noalias sret align 32 %agg.result, [2 x i128] %x.coerce)
 struct v16i16 test_struct_v16i16(struct v16i16 x)
 {
   return x;
Index: clang/test/CodeGen/ppc64le-aggregates.c
===================================================================
--- clang/test/CodeGen/ppc64le-aggregates.c
+++ clang/test/CodeGen/ppc64le-aggregates.c
@@ -41,7 +41,7 @@
 // CHECK: define [8 x float] @func_f8([8 x float] %x.coerce)
 struct f8 func_f8(struct f8 x) { return x; }
 
-// CHECK: define void @func_f9(%struct.f9* noalias sret %agg.result, [5 x i64] %x.coerce)
+// CHECK: define void @func_f9(%struct.f9* noalias sret align 4 %agg.result, [5 x i64] %x.coerce)
 struct f9 func_f9(struct f9 x) { return x; }
 
 // CHECK: define [2 x float] @func_fab([2 x float] %x.coerce)
@@ -106,7 +106,7 @@
 // CHECK: %[[TMP2:[^ ]+]] = bitcast [5 x i64]* %[[TMP1]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 %[[TMP2]], i8* align 4 bitcast (%struct.f9* @global_f9 to i8*), i64 36, i1 false)
 // CHECK: %[[TMP3:[^ ]+]] = load [5 x i64], [5 x i64]* %[[TMP1]]
-// CHECK: call void @func_f9(%struct.f9* sret %{{[^ ]+}}, [5 x i64] %[[TMP3]])
+// CHECK: call void @func_f9(%struct.f9* sret align 4 %{{[^ ]+}}, [5 x i64] %[[TMP3]])
 struct f9 global_f9;
 void call_f9(void) { global_f9 = func_f9(global_f9); }
 
@@ -162,7 +162,7 @@
 // CHECK: define [8 x <4 x i32>] @func_v8([8 x <4 x i32>] %x.coerce)
 struct v8 func_v8(struct v8 x) { return x; }
 
-// CHECK: define void @func_v9(%struct.v9* noalias sret %agg.result, %struct.v9* byval(%struct.v9) align 16 %x)
+// CHECK: define void @func_v9(%struct.v9* noalias sret align 16 %agg.result, %struct.v9* byval(%struct.v9) align 16 %x)
 struct v9 func_v9(struct v9 x) { return x; }
 
 // CHECK: define [2 x <4 x i32>] @func_vab([2 x <4 x i32>] %x.coerce)
@@ -220,7 +220,7 @@
 void call_v8(void) { global_v8 = func_v8(global_v8); }
 
 // CHECK-LABEL: @call_v9
-// CHECK: call void @func_v9(%struct.v9* sret %{{[^ ]+}}, %struct.v9* byval(%struct.v9) align 16 @global_v9)
+// CHECK: call void @func_v9(%struct.v9* sret align 16 %{{[^ ]+}}, %struct.v9* byval(%struct.v9) align 16 @global_v9)
 struct v9 global_v9;
 void call_v9(void) { global_v9 = func_v9(global_v9); }
 
@@ -279,7 +279,7 @@
 // CHECK: define [8 x <4 x float>] @func_v3f8([8 x <4 x float>] %x.coerce)
 struct v3f8 func_v3f8(struct v3f8 x) { return x; }
 
-// CHECK: define void @func_v3f9(%struct.v3f9* noalias sret %agg.result, %struct.v3f9* byval(%struct.v3f9) align 16 %x)
+// CHECK: define void @func_v3f9(%struct.v3f9* noalias sret align 16 %agg.result, %struct.v3f9* byval(%struct.v3f9) align 16 %x)
 struct v3f9 func_v3f9(struct v3f9 x) { return x; }
 
 // CHECK: define [2 x <4 x float>] @func_v3fab([2 x <4 x float>] %x.coerce)
@@ -337,7 +337,7 @@
 void call_v3f8(void) { global_v3f8 = func_v3f8(global_v3f8); }
 
 // CHECK-LABEL: @call_v3f9
-// CHECK: call void @func_v3f9(%struct.v3f9* sret %{{[^ ]+}}, %struct.v3f9* byval(%struct.v3f9) align 16 @global_v3f9)
+// CHECK: call void @func_v3f9(%struct.v3f9* sret align 16 %{{[^ ]+}}, %struct.v3f9* byval(%struct.v3f9) align 16 @global_v3f9)
 struct v3f9 global_v3f9;
 void call_v3f9(void) { global_v3f9 = func_v3f9(global_v3f9); }
 
Index: clang/test/CodeGen/ppc64le-f128Aggregates.c
===================================================================
--- clang/test/CodeGen/ppc64le-f128Aggregates.c
+++ clang/test/CodeGen/ppc64le-f128Aggregates.c
@@ -42,7 +42,7 @@
 // CHECK: define [8 x fp128] @func_f8([8 x fp128] %x.coerce)
 struct fp8 func_f8(struct fp8 x) { return x; }
 
-// CHECK: define void @func_f9(%struct.fp9* noalias sret %agg.result, %struct.fp9* byval(%struct.fp9) align 16 %x)
+// CHECK: define void @func_f9(%struct.fp9* noalias sret align 16 %agg.result, %struct.fp9* byval(%struct.fp9) align 16 %x)
 struct fp9 func_f9(struct fp9 x) { return x; }
 
 // CHECK: define [2 x fp128] @func_fab([2 x fp128] %x.coerce)
@@ -104,7 +104,7 @@
 
 // CHECK-LABEL: @call_fp9
 // CHECK: %[[TMP1:[^ ]+]] = alloca %struct.fp9, align 16
-// CHECK: call void @func_f9(%struct.fp9* sret %[[TMP2:[^ ]+]], %struct.fp9* byval(%struct.fp9) align 16 @global_f9
+// CHECK: call void @func_f9(%struct.fp9* sret align 16 %[[TMP2:[^ ]+]], %struct.fp9* byval(%struct.fp9) align 16 @global_f9
 // CHECK: %[[TMP3:[^ ]+]] = bitcast %struct.fp9* %[[TMP2]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 bitcast (%struct.fp9* @global_f9 to i8*), i8* align 16 %[[TMP3]], i64 144, i1 false
 // CHECK: ret void
Index: clang/test/CodeGen/regparm-struct.c
===================================================================
--- clang/test/CodeGen/regparm-struct.c
+++ clang/test/CodeGen/regparm-struct.c
@@ -159,7 +159,7 @@
 }
 
 __attribute__((regparm(3))) struct s12 f17(int a, int b, int c);
-// CHECK: declare void @f17(%struct.s12* inreg sret, i32 inreg, i32 inreg, i32)
+// CHECK: declare void @f17(%struct.s12* inreg sret align 4, i32 inreg, i32 inreg, i32)
 void g17(void) {
   f17(41, 42, 43);
 }
Index: clang/test/CodeGen/renderscript.c
===================================================================
--- clang/test/CodeGen/renderscript.c
+++ clang/test/CodeGen/renderscript.c
@@ -83,15 +83,15 @@
 // and coerced to [a x iNN] for 64-bit RenderScript
 // =============================================================================
 
-// CHECK-RS32: void @retShortCharShort(%struct.sShortCharShort* noalias sret %agg.result)
+// CHECK-RS32: void @retShortCharShort(%struct.sShortCharShort* noalias sret align 2 %agg.result)
 // CHECK-RS64: [3 x i16] @retShortCharShort()
 sShortCharShort retShortCharShort() { sShortCharShort r; return r; }
 
-// CHECK-RS32: void @retIntShortChar(%struct.sIntShortChar* noalias sret %agg.result)
+// CHECK-RS32: void @retIntShortChar(%struct.sIntShortChar* noalias sret align 4 %agg.result)
 // CHECK-RS64: [2 x i32] @retIntShortChar()
 sIntShortChar retIntShortChar() { sIntShortChar r; return r; }
 
-// CHECK-RS32: void @retLongInt(%struct.sLongInt* noalias sret %agg.result)
+// CHECK-RS32: void @retLongInt(%struct.sLongInt* noalias sret align 8 %agg.result)
 // CHECK-RS64: [2 x i64] @retLongInt()
 sLongInt retLongInt() { sLongInt r; return r; }
 
@@ -116,12 +116,12 @@
 // 64-bit RenderScript
 // =============================================================================
 
-// CHECK-RS32: void @retInt5(%struct.sInt5* noalias sret %agg.result)
-// CHECK-RS64: void @retInt5(%struct.sInt5* noalias sret %agg.result)
+// CHECK-RS32: void @retInt5(%struct.sInt5* noalias sret align 4 %agg.result)
+// CHECK-RS64: void @retInt5(%struct.sInt5* noalias sret align 4 %agg.result)
 sInt5 retInt5() { sInt5 r; return r;}
 
-// CHECK-RS32: void @retLong2Char(%struct.sLong2Char* noalias sret %agg.result)
-// CHECK-RS64: void @retLong2Char(%struct.sLong2Char* noalias sret %agg.result)
+// CHECK-RS32: void @retLong2Char(%struct.sLong2Char* noalias sret align 8 %agg.result)
+// CHECK-RS64: void @retLong2Char(%struct.sLong2Char* noalias sret align 8 %agg.result)
 sLong2Char retLong2Char() { sLong2Char r; return r;}
 
 // =============================================================================
@@ -135,6 +135,6 @@
 // CHECK-RS64: void @argLong9(%struct.sLong9* %s)
 void argLong9(sLong9 s) {}
 
-// CHECK-RS32: void @retLong9(%struct.sLong9* noalias sret %agg.result)
-// CHECK-RS64: void @retLong9(%struct.sLong9* noalias sret %agg.result)
+// CHECK-RS32: void @retLong9(%struct.sLong9* noalias sret align 8 %agg.result)
+// CHECK-RS64: void @retLong9(%struct.sLong9* noalias sret align 8 %agg.result)
 sLong9 retLong9() { sLong9 r; return r; }
Index: clang/test/CodeGen/riscv32-ilp32-abi.c
===================================================================
--- clang/test/CodeGen/riscv32-ilp32-abi.c
+++ clang/test/CodeGen/riscv32-ilp32-abi.c
@@ -35,7 +35,7 @@
 // the presence of large return values that consume a register due to the need
 // to pass a pointer.
 
-// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret %agg.result, float %a, i64 %b, double %c, fp128 %d, i8 zeroext %e, i8 %f, i8 %g)
+// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret align 4 %agg.result, float %a, i64 %b, double %c, fp128 %d, i8 zeroext %e, i8 %f, i8 %g)
 struct large f_scalar_stack_2(float a, int64_t b, double c, long double d,
                               uint8_t e, int8_t f, uint8_t g) {
   return (struct large){a, e, f, g};
Index: clang/test/CodeGen/riscv32-ilp32-ilp32f-abi.c
===================================================================
--- clang/test/CodeGen/riscv32-ilp32-ilp32f-abi.c
+++ clang/test/CodeGen/riscv32-ilp32-ilp32f-abi.c
@@ -37,7 +37,7 @@
 // the presence of large return values that consume a register due to the need
 // to pass a pointer.
 
-// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret %agg.result, i32 %a, i64 %b, double %c, fp128 %d, i8 zeroext %e, i8 %f, i8 %g)
+// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret align 4 %agg.result, i32 %a, i64 %b, double %c, fp128 %d, i8 zeroext %e, i8 %f, i8 %g)
 struct large f_scalar_stack_2(int32_t a, int64_t b, double c, long double d,
                               uint8_t e, int8_t f, uint8_t g) {
   return (struct large){a, e, f, g};
Index: clang/test/CodeGen/riscv32-ilp32-ilp32f-ilp32d-abi.c
===================================================================
--- clang/test/CodeGen/riscv32-ilp32-ilp32f-ilp32d-abi.c
+++ clang/test/CodeGen/riscv32-ilp32-ilp32f-ilp32d-abi.c
@@ -177,7 +177,7 @@
 
 // The address where the struct should be written to will be the first
 // argument
-// CHECK-LABEL: define void @f_agg_large_ret(%struct.large* noalias sret %agg.result, i32 %i, i8 signext %j)
+// CHECK-LABEL: define void @f_agg_large_ret(%struct.large* noalias sret align 4 %agg.result, i32 %i, i8 signext %j)
 struct large f_agg_large_ret(int32_t i, int8_t j) {
   return (struct large){1, 2, 3, 4};
 }
@@ -189,7 +189,7 @@
   x[0] = x[7];
 }
 
-// CHECK-LABEL: define void @f_vec_large_v16i8_ret(<16 x i8>* noalias sret %agg.result)
+// CHECK-LABEL: define void @f_vec_large_v16i8_ret(<16 x i8>* noalias sret align 16 %agg.result)
 v16i8 f_vec_large_v16i8_ret() {
   return (v16i8){1, 2, 3, 4, 5, 6, 7, 8};
 }
@@ -207,7 +207,7 @@
 // the presence of large return values that consume a register due to the need
 // to pass a pointer.
 
-// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret %agg.result, i32 %a, i64 %b, i64 %c, fp128 %d, i8 zeroext %e, i8 %f, i8 %g)
+// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret align 4 %agg.result, i32 %a, i64 %b, i64 %c, fp128 %d, i8 zeroext %e, i8 %f, i8 %g)
 struct large f_scalar_stack_2(int32_t a, int64_t b, int64_t c, long double d,
                               uint8_t e, int8_t f, uint8_t g) {
   return (struct large){a, e, f, g};
Index: clang/test/CodeGen/riscv32-ilp32d-abi.c
===================================================================
--- clang/test/CodeGen/riscv32-ilp32d-abi.c
+++ clang/test/CodeGen/riscv32-ilp32d-abi.c
@@ -119,7 +119,7 @@
 // CHECK: define void @f_double_int64_s_arg(%struct.double_int64_s* %a)
 void f_double_int64_s_arg(struct double_int64_s a) {}
 
-// CHECK: define void @f_ret_double_int64_s(%struct.double_int64_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_double_int64_s(%struct.double_int64_s* noalias sret align 8 %agg.result)
 struct double_int64_s f_ret_double_int64_s() {
   return (struct double_int64_s){1.0, 2};
 }
@@ -243,7 +243,7 @@
 // CHECK: define void @f_int_double_int_s_arg(%struct.int_double_int_s* %a)
 void f_int_double_int_s_arg(struct int_double_int_s a) {}
 
-// CHECK: define void @f_ret_int_double_int_s(%struct.int_double_int_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_int_double_int_s(%struct.int_double_int_s* noalias sret align 8 %agg.result)
 struct int_double_int_s f_ret_int_double_int_s() {
   return (struct int_double_int_s){1, 2.0, 3};
 }
@@ -253,7 +253,7 @@
 // CHECK: define void @f_int64_double_s_arg(%struct.int64_double_s* %a)
 void f_int64_double_s_arg(struct int64_double_s a) {}
 
-// CHECK: define void @f_ret_int64_double_s(%struct.int64_double_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_int64_double_s(%struct.int64_double_s* noalias sret align 8 %agg.result)
 struct int64_double_s f_ret_int64_double_s() {
   return (struct int64_double_s){1, 2.0};
 }
@@ -263,7 +263,7 @@
 // CHECK-LABEL: define void @f_char_char_double_s_arg(%struct.char_char_double_s* %a)
 void f_char_char_double_s_arg(struct char_char_double_s a) {}
 
-// CHECK: define void @f_ret_char_char_double_s(%struct.char_char_double_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_char_char_double_s(%struct.char_char_double_s* noalias sret align 8 %agg.result)
 struct char_char_double_s f_ret_char_char_double_s() {
   return (struct char_char_double_s){1, 2, 3.0};
 }
Index: clang/test/CodeGen/riscv32-ilp32f-abi.c
===================================================================
--- clang/test/CodeGen/riscv32-ilp32f-abi.c
+++ clang/test/CodeGen/riscv32-ilp32f-abi.c
@@ -26,7 +26,7 @@
 // CHECK: define void @f_double_double_s_arg(%struct.double_double_s* %a)
 void f_double_double_s_arg(struct double_double_s a) {}
 
-// CHECK: define void @f_ret_double_double_s(%struct.double_double_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_double_double_s(%struct.double_double_s* noalias sret align 8 %agg.result)
 struct double_double_s f_ret_double_double_s() {
   return (struct double_double_s){1.0, 2.0};
 }
@@ -38,7 +38,7 @@
 // CHECK: define void @f_int_double_s_arg(%struct.int_double_s* %a)
 void f_int_double_s_arg(struct int_double_s a) {}
 
-// CHECK: define void @f_ret_int_double_s(%struct.int_double_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_int_double_s(%struct.int_double_s* noalias sret align 8 %agg.result)
 struct int_double_s f_ret_int_double_s() {
   return (struct int_double_s){1, 2.0};
 }
Index: clang/test/CodeGen/riscv32-ilp32f-ilp32d-abi.c
===================================================================
--- clang/test/CodeGen/riscv32-ilp32f-ilp32d-abi.c
+++ clang/test/CodeGen/riscv32-ilp32f-ilp32d-abi.c
@@ -112,7 +112,7 @@
 // CHECK: define void @f_float_int64_s_arg(%struct.float_int64_s* %a)
 void f_float_int64_s_arg(struct float_int64_s a) {}
 
-// CHECK: define void @f_ret_float_int64_s(%struct.float_int64_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_float_int64_s(%struct.float_int64_s* noalias sret align 8 %agg.result)
 struct float_int64_s f_ret_float_int64_s() {
   return (struct float_int64_s){1.0, 2};
 }
@@ -236,7 +236,7 @@
 // CHECK: define void @f_int_float_int_s_arg(%struct.int_float_int_s* %a)
 void f_int_float_int_s_arg(struct int_float_int_s a) {}
 
-// CHECK: define void @f_ret_int_float_int_s(%struct.int_float_int_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_int_float_int_s(%struct.int_float_int_s* noalias sret align 4 %agg.result)
 struct int_float_int_s f_ret_int_float_int_s() {
   return (struct int_float_int_s){1, 2.0, 3};
 }
@@ -246,7 +246,7 @@
 // CHECK: define void @f_int64_float_s_arg(%struct.int64_float_s* %a)
 void f_int64_float_s_arg(struct int64_float_s a) {}
 
-// CHECK: define void @f_ret_int64_float_s(%struct.int64_float_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_int64_float_s(%struct.int64_float_s* noalias sret align 8 %agg.result)
 struct int64_float_s f_ret_int64_float_s() {
   return (struct int64_float_s){1, 2.0};
 }
Index: clang/test/CodeGen/riscv64-lp64-abi.c
===================================================================
--- clang/test/CodeGen/riscv64-lp64-abi.c
+++ clang/test/CodeGen/riscv64-lp64-abi.c
@@ -25,7 +25,7 @@
 // the presence of large return values that consume a register due to the need
 // to pass a pointer.
 
-// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret %agg.result, double %a, i128 %b, fp128 %c, <32 x i8>* %0, i8 zeroext %e, i8 %f, i8 %g)
+// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret align 8 %agg.result, double %a, i128 %b, fp128 %c, <32 x i8>* %0, i8 zeroext %e, i8 %f, i8 %g)
 struct large f_scalar_stack_2(double a, __int128_t b, long double c, v32i8 d,
                               uint8_t e, int8_t f, uint8_t g) {
   return (struct large){a, e, f, g};
Index: clang/test/CodeGen/riscv64-lp64-lp64f-abi.c
===================================================================
--- clang/test/CodeGen/riscv64-lp64-lp64f-abi.c
+++ clang/test/CodeGen/riscv64-lp64-lp64f-abi.c
@@ -27,7 +27,7 @@
 // the presence of large return values that consume a register due to the need
 // to pass a pointer.
 
-// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret %agg.result, double %a, i128 %b, fp128 %c, <32 x i8>* %0, i8 zeroext %e, i8 %f, i8 %g)
+// CHECK-LABEL: define void @f_scalar_stack_2(%struct.large* noalias sret align 8 %agg.result, double %a, i128 %b, fp128 %c, <32 x i8>* %0, i8 zeroext %e, i8 %f, i8 %g)
 struct large f_scalar_stack_2(double a, __int128_t b, long double c, v32i8 d,
                               uint8_t e, int8_t f, uint8_t g) {
   return (struct large){a, e, f, g};
Index: clang/test/CodeGen/riscv64-lp64-lp64f-lp64d-abi.c
===================================================================
--- clang/test/CodeGen/riscv64-lp64-lp64f-lp64d-abi.c
+++ clang/test/CodeGen/riscv64-lp64-lp64f-lp64d-abi.c
@@ -166,7 +166,7 @@
 
 // The address where the struct should be written to will be the first
 // argument
-// CHECK-LABEL: define void @f_agg_large_ret(%struct.large* noalias sret %agg.result, i32 signext %i, i8 signext %j)
+// CHECK-LABEL: define void @f_agg_large_ret(%struct.large* noalias sret align 8 %agg.result, i32 signext %i, i8 signext %j)
 struct large f_agg_large_ret(int32_t i, int8_t j) {
   return (struct large){1, 2, 3, 4};
 }
@@ -178,7 +178,7 @@
   x[0] = x[7];
 }
 
-// CHECK-LABEL: define void @f_vec_large_v32i8_ret(<32 x i8>* noalias sret %agg.result)
+// CHECK-LABEL: define void @f_vec_large_v32i8_ret(<32 x i8>* noalias sret align 32 %agg.result)
 v32i8 f_vec_large_v32i8_ret() {
   return (v32i8){1, 2, 3, 4, 5, 6, 7, 8};
 }
@@ -202,7 +202,7 @@
 // the presence of large return values that consume a register due to the need
 // to pass a pointer.
 
-// CHECK-LABEL: define void @f_scalar_stack_3(%struct.large* noalias sret %agg.result, i32 signext %a, i128 %b, fp128 %c, <32 x i8>* %0, i8 zeroext %e, i8 %f, i8 %g)
+// CHECK-LABEL: define void @f_scalar_stack_3(%struct.large* noalias sret align 8 %agg.result, i32 signext %a, i128 %b, fp128 %c, <32 x i8>* %0, i8 zeroext %e, i8 %f, i8 %g)
 struct large f_scalar_stack_3(uint32_t a, __int128_t b, long double c, v32i8 d,
                               uint8_t e, int8_t f, uint8_t g) {
   return (struct large){a, e, f, g};
Index: clang/test/CodeGen/riscv64-lp64d-abi.c
===================================================================
--- clang/test/CodeGen/riscv64-lp64d-abi.c
+++ clang/test/CodeGen/riscv64-lp64d-abi.c
@@ -243,7 +243,7 @@
 // CHECK: define void @f_int_double_int_s_arg(%struct.int_double_int_s* %a)
 void f_int_double_int_s_arg(struct int_double_int_s a) {}
 
-// CHECK: define void @f_ret_int_double_int_s(%struct.int_double_int_s* noalias sret %agg.result)
+// CHECK: define void @f_ret_int_double_int_s(%struct.int_double_int_s* noalias sret align 8 %agg.result)
 struct int_double_int_s f_ret_int_double_int_s() {
   return (struct int_double_int_s){1, 2.0, 3};
 }
Index: clang/test/CodeGen/sparcv9-abi.c
===================================================================
--- clang/test/CodeGen/sparcv9-abi.c
+++ clang/test/CodeGen/sparcv9-abi.c
@@ -53,7 +53,7 @@
   int x;
 };
 
-// CHECK-LABEL: define void @f_large(%struct.large* noalias sret %agg.result, %struct.large* %x)
+// CHECK-LABEL: define void @f_large(%struct.large* noalias sret align 8 %agg.result, %struct.large* %x)
 struct large f_large(struct large x) {
   x.a += *x.b;
   x.b = 0;
Index: clang/test/CodeGen/struct-passing.c
===================================================================
--- clang/test/CodeGen/struct-passing.c
+++ clang/test/CodeGen/struct-passing.c
@@ -18,8 +18,8 @@
 
 // CHECK: declare i32 @f0() [[RN:#[0-9]+]]
 // CHECK: declare i32 @f1() [[RO:#[0-9]+]]
-// CHECK: declare void @f2({{.*}} sret)
-// CHECK: declare void @f3({{.*}} sret)
+// CHECK: declare void @f2({{.*}} sret align 4)
+// CHECK: declare void @f3({{.*}} sret align 4)
 // CHECK: declare void @f4({{.*}} byval({{.*}}) align 4)
 // CHECK: declare void @f5({{.*}} byval({{.*}}) align 4)
 
Index: clang/test/CodeGen/systemz-abi-vector.c
===================================================================
--- clang/test/CodeGen/systemz-abi-vector.c
+++ clang/test/CodeGen/systemz-abi-vector.c
@@ -50,91 +50,91 @@
 // CHECK-VECTOR: @align = global i32 8
 
 v1i8 pass_v1i8(v1i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1i8(<1 x i8>* noalias sret %{{.*}}, <1 x i8>* %0)
+// CHECK-LABEL: define void @pass_v1i8(<1 x i8>* noalias sret align 1 %{{.*}}, <1 x i8>* %0)
 // CHECK-VECTOR-LABEL: define <1 x i8> @pass_v1i8(<1 x i8> %{{.*}})
 
 v2i8 pass_v2i8(v2i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v2i8(<2 x i8>* noalias sret %{{.*}}, <2 x i8>* %0)
+// CHECK-LABEL: define void @pass_v2i8(<2 x i8>* noalias sret align 2 %{{.*}}, <2 x i8>* %0)
 // CHECK-VECTOR-LABEL: define <2 x i8> @pass_v2i8(<2 x i8> %{{.*}})
 
 v4i8 pass_v4i8(v4i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v4i8(<4 x i8>* noalias sret %{{.*}}, <4 x i8>* %0)
+// CHECK-LABEL: define void @pass_v4i8(<4 x i8>* noalias sret align 4 %{{.*}}, <4 x i8>* %0)
 // CHECK-VECTOR-LABEL: define <4 x i8> @pass_v4i8(<4 x i8> %{{.*}})
 
 v8i8 pass_v8i8(v8i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v8i8(<8 x i8>* noalias sret %{{.*}}, <8 x i8>* %0)
+// CHECK-LABEL: define void @pass_v8i8(<8 x i8>* noalias sret align 8 %{{.*}}, <8 x i8>* %0)
 // CHECK-VECTOR-LABEL: define <8 x i8> @pass_v8i8(<8 x i8> %{{.*}})
 
 v16i8 pass_v16i8(v16i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v16i8(<16 x i8>* noalias sret %{{.*}}, <16 x i8>* %0)
+// CHECK-LABEL: define void @pass_v16i8(<16 x i8>* noalias sret align 16 %{{.*}}, <16 x i8>* %0)
 // CHECK-VECTOR-LABEL: define <16 x i8> @pass_v16i8(<16 x i8> %{{.*}})
 
 v32i8 pass_v32i8(v32i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v32i8(<32 x i8>* noalias sret %{{.*}}, <32 x i8>* %0)
-// CHECK-VECTOR-LABEL: define void @pass_v32i8(<32 x i8>* noalias sret %{{.*}}, <32 x i8>* %0)
+// CHECK-LABEL: define void @pass_v32i8(<32 x i8>* noalias sret align 32 %{{.*}}, <32 x i8>* %0)
+// CHECK-VECTOR-LABEL: define void @pass_v32i8(<32 x i8>* noalias sret align 8 %{{.*}}, <32 x i8>* %0)
 
 v1i16 pass_v1i16(v1i16 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1i16(<1 x i16>* noalias sret %{{.*}}, <1 x i16>* %0)
+// CHECK-LABEL: define void @pass_v1i16(<1 x i16>* noalias sret align 2 %{{.*}}, <1 x i16>* %0)
 // CHECK-VECTOR-LABEL: define <1 x i16> @pass_v1i16(<1 x i16> %{{.*}})
 
 v2i16 pass_v2i16(v2i16 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v2i16(<2 x i16>* noalias sret %{{.*}}, <2 x i16>* %0)
+// CHECK-LABEL: define void @pass_v2i16(<2 x i16>* noalias sret align 4 %{{.*}}, <2 x i16>* %0)
 // CHECK-VECTOR-LABEL: define <2 x i16> @pass_v2i16(<2 x i16> %{{.*}})
 
 v4i16 pass_v4i16(v4i16 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v4i16(<4 x i16>* noalias sret %{{.*}}, <4 x i16>* %0)
+// CHECK-LABEL: define void @pass_v4i16(<4 x i16>* noalias sret align 8 %{{.*}}, <4 x i16>* %0)
 // CHECK-VECTOR-LABEL: define <4 x i16> @pass_v4i16(<4 x i16> %{{.*}})
 
 v8i16 pass_v8i16(v8i16 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v8i16(<8 x i16>* noalias sret %{{.*}}, <8 x i16>* %0)
+// CHECK-LABEL: define void @pass_v8i16(<8 x i16>* noalias sret align 16 %{{.*}}, <8 x i16>* %0)
 // CHECK-VECTOR-LABEL: define <8 x i16> @pass_v8i16(<8 x i16> %{{.*}})
 
 v1i32 pass_v1i32(v1i32 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1i32(<1 x i32>* noalias sret %{{.*}}, <1 x i32>* %0)
+// CHECK-LABEL: define void @pass_v1i32(<1 x i32>* noalias sret align 4 %{{.*}}, <1 x i32>* %0)
 // CHECK-VECTOR-LABEL: define <1 x i32> @pass_v1i32(<1 x i32> %{{.*}})
 
 v2i32 pass_v2i32(v2i32 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v2i32(<2 x i32>* noalias sret %{{.*}}, <2 x i32>* %0)
+// CHECK-LABEL: define void @pass_v2i32(<2 x i32>* noalias sret align 8 %{{.*}}, <2 x i32>* %0)
 // CHECK-VECTOR-LABEL: define <2 x i32> @pass_v2i32(<2 x i32> %{{.*}})
 
 v4i32 pass_v4i32(v4i32 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v4i32(<4 x i32>* noalias sret %{{.*}}, <4 x i32>* %0)
+// CHECK-LABEL: define void @pass_v4i32(<4 x i32>* noalias sret align 16 %{{.*}}, <4 x i32>* %0)
 // CHECK-VECTOR-LABEL: define <4 x i32> @pass_v4i32(<4 x i32> %{{.*}})
 
 v1i64 pass_v1i64(v1i64 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1i64(<1 x i64>* noalias sret %{{.*}}, <1 x i64>* %0)
+// CHECK-LABEL: define void @pass_v1i64(<1 x i64>* noalias sret align 8 %{{.*}}, <1 x i64>* %0)
 // CHECK-VECTOR-LABEL: define <1 x i64> @pass_v1i64(<1 x i64> %{{.*}})
 
 v2i64 pass_v2i64(v2i64 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v2i64(<2 x i64>* noalias sret %{{.*}}, <2 x i64>* %0)
+// CHECK-LABEL: define void @pass_v2i64(<2 x i64>* noalias sret align 16 %{{.*}}, <2 x i64>* %0)
 // CHECK-VECTOR-LABEL: define <2 x i64> @pass_v2i64(<2 x i64> %{{.*}})
 
 v1i128 pass_v1i128(v1i128 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1i128(<1 x i128>* noalias sret %{{.*}}, <1 x i128>* %0)
+// CHECK-LABEL: define void @pass_v1i128(<1 x i128>* noalias sret align 16 %{{.*}}, <1 x i128>* %0)
 // CHECK-VECTOR-LABEL: define <1 x i128> @pass_v1i128(<1 x i128> %{{.*}})
 
 v1f32 pass_v1f32(v1f32 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1f32(<1 x float>* noalias sret %{{.*}}, <1 x float>* %0)
+// CHECK-LABEL: define void @pass_v1f32(<1 x float>* noalias sret align 4 %{{.*}}, <1 x float>* %0)
 // CHECK-VECTOR-LABEL: define <1 x float> @pass_v1f32(<1 x float> %{{.*}})
 
 v2f32 pass_v2f32(v2f32 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v2f32(<2 x float>* noalias sret %{{.*}}, <2 x float>* %0)
+// CHECK-LABEL: define void @pass_v2f32(<2 x float>* noalias sret align 8 %{{.*}}, <2 x float>* %0)
 // CHECK-VECTOR-LABEL: define <2 x float> @pass_v2f32(<2 x float> %{{.*}})
 
 v4f32 pass_v4f32(v4f32 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v4f32(<4 x float>* noalias sret %{{.*}}, <4 x float>* %0)
+// CHECK-LABEL: define void @pass_v4f32(<4 x float>* noalias sret align 16 %{{.*}}, <4 x float>* %0)
 // CHECK-VECTOR-LABEL: define <4 x float> @pass_v4f32(<4 x float> %{{.*}})
 
 v1f64 pass_v1f64(v1f64 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1f64(<1 x double>* noalias sret %{{.*}}, <1 x double>* %0)
+// CHECK-LABEL: define void @pass_v1f64(<1 x double>* noalias sret align 8 %{{.*}}, <1 x double>* %0)
 // CHECK-VECTOR-LABEL: define <1 x double> @pass_v1f64(<1 x double> %{{.*}})
 
 v2f64 pass_v2f64(v2f64 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v2f64(<2 x double>* noalias sret %{{.*}}, <2 x double>* %0)
+// CHECK-LABEL: define void @pass_v2f64(<2 x double>* noalias sret align 16 %{{.*}}, <2 x double>* %0)
 // CHECK-VECTOR-LABEL: define <2 x double> @pass_v2f64(<2 x double> %{{.*}})
 
 v1f128 pass_v1f128(v1f128 arg) { return arg; }
-// CHECK-LABEL: define void @pass_v1f128(<1 x fp128>* noalias sret %{{.*}}, <1 x fp128>* %0)
+// CHECK-LABEL: define void @pass_v1f128(<1 x fp128>* noalias sret align 16 %{{.*}}, <1 x fp128>* %0)
 // CHECK-VECTOR-LABEL: define <1 x fp128> @pass_v1f128(<1 x fp128> %{{.*}})
 
 
@@ -142,62 +142,62 @@
 
 struct agg_v1i8 { v1i8 a; };
 struct agg_v1i8 pass_agg_v1i8(struct agg_v1i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_v1i8(%struct.agg_v1i8* noalias sret %{{.*}}, i8 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_v1i8(%struct.agg_v1i8* noalias sret %{{.*}}, <1 x i8> %{{.*}})
+// CHECK-LABEL: define void @pass_agg_v1i8(%struct.agg_v1i8* noalias sret align 1 %{{.*}}, i8 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_v1i8(%struct.agg_v1i8* noalias sret align 1 %{{.*}}, <1 x i8> %{{.*}})
 
 struct agg_v2i8 { v2i8 a; };
 struct agg_v2i8 pass_agg_v2i8(struct agg_v2i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_v2i8(%struct.agg_v2i8* noalias sret %{{.*}}, i16 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_v2i8(%struct.agg_v2i8* noalias sret %{{.*}}, <2 x i8> %{{.*}})
+// CHECK-LABEL: define void @pass_agg_v2i8(%struct.agg_v2i8* noalias sret align 2 %{{.*}}, i16 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_v2i8(%struct.agg_v2i8* noalias sret align 2 %{{.*}}, <2 x i8> %{{.*}})
 
 struct agg_v4i8 { v4i8 a; };
 struct agg_v4i8 pass_agg_v4i8(struct agg_v4i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_v4i8(%struct.agg_v4i8* noalias sret %{{.*}}, i32 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_v4i8(%struct.agg_v4i8* noalias sret %{{.*}}, <4 x i8> %{{.*}})
+// CHECK-LABEL: define void @pass_agg_v4i8(%struct.agg_v4i8* noalias sret align 4 %{{.*}}, i32 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_v4i8(%struct.agg_v4i8* noalias sret align 4 %{{.*}}, <4 x i8> %{{.*}})
 
 struct agg_v8i8 { v8i8 a; };
 struct agg_v8i8 pass_agg_v8i8(struct agg_v8i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_v8i8(%struct.agg_v8i8* noalias sret %{{.*}}, i64 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_v8i8(%struct.agg_v8i8* noalias sret %{{.*}}, <8 x i8> %{{.*}})
+// CHECK-LABEL: define void @pass_agg_v8i8(%struct.agg_v8i8* noalias sret align 8 %{{.*}}, i64 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_v8i8(%struct.agg_v8i8* noalias sret align 8 %{{.*}}, <8 x i8> %{{.*}})
 
 struct agg_v16i8 { v16i8 a; };
 struct agg_v16i8 pass_agg_v16i8(struct agg_v16i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_v16i8(%struct.agg_v16i8* noalias sret %{{.*}}, %struct.agg_v16i8* %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_v16i8(%struct.agg_v16i8* noalias sret %{{.*}}, <16 x i8> %{{.*}})
+// CHECK-LABEL: define void @pass_agg_v16i8(%struct.agg_v16i8* noalias sret align 16 %{{.*}}, %struct.agg_v16i8* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_v16i8(%struct.agg_v16i8* noalias sret align 8 %{{.*}}, <16 x i8> %{{.*}})
 
 struct agg_v32i8 { v32i8 a; };
 struct agg_v32i8 pass_agg_v32i8(struct agg_v32i8 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_v32i8(%struct.agg_v32i8* noalias sret %{{.*}}, %struct.agg_v32i8* %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_v32i8(%struct.agg_v32i8* noalias sret %{{.*}}, %struct.agg_v32i8* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_v32i8(%struct.agg_v32i8* noalias sret align 32 %{{.*}}, %struct.agg_v32i8* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_v32i8(%struct.agg_v32i8* noalias sret align 8 %{{.*}}, %struct.agg_v32i8* %{{.*}})
 
 
 // Verify that the following are *not* vector-like aggregate types
 
 struct agg_novector1 { v4i8 a; v4i8 b; };
 struct agg_novector1 pass_agg_novector1(struct agg_novector1 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_novector1(%struct.agg_novector1* noalias sret %{{.*}}, i64 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_novector1(%struct.agg_novector1* noalias sret %{{.*}}, i64 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_novector1(%struct.agg_novector1* noalias sret align 4 %{{.*}}, i64 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_novector1(%struct.agg_novector1* noalias sret align 4 %{{.*}}, i64 %{{.*}})
 
 struct agg_novector2 { v4i8 a; float b; };
 struct agg_novector2 pass_agg_novector2(struct agg_novector2 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_novector2(%struct.agg_novector2* noalias sret %{{.*}}, i64 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_novector2(%struct.agg_novector2* noalias sret %{{.*}}, i64 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_novector2(%struct.agg_novector2* noalias sret align 4 %{{.*}}, i64 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_novector2(%struct.agg_novector2* noalias sret align 4 %{{.*}}, i64 %{{.*}})
 
 struct agg_novector3 { v4i8 a; int : 0; };
 struct agg_novector3 pass_agg_novector3(struct agg_novector3 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_novector3(%struct.agg_novector3* noalias sret %{{.*}}, i32 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_novector3(%struct.agg_novector3* noalias sret %{{.*}}, i32 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_novector3(%struct.agg_novector3* noalias sret align 4 %{{.*}}, i32 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_novector3(%struct.agg_novector3* noalias sret align 4 %{{.*}}, i32 %{{.*}})
 
 struct agg_novector4 { v4i8 a __attribute__((aligned (8))); };
 struct agg_novector4 pass_agg_novector4(struct agg_novector4 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_novector4(%struct.agg_novector4* noalias sret %{{.*}}, i64 %{{.*}})
-// CHECK-VECTOR-LABEL: define void @pass_agg_novector4(%struct.agg_novector4* noalias sret %{{.*}}, i64 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_novector4(%struct.agg_novector4* noalias sret align 8 %{{.*}}, i64 %{{.*}})
+// CHECK-VECTOR-LABEL: define void @pass_agg_novector4(%struct.agg_novector4* noalias sret align 8 %{{.*}}, i64 %{{.*}})
 
 
 // Accessing variable argument lists
 
 v1i8 va_v1i8(__builtin_va_list l) { return __builtin_va_arg(l, v1i8); }
-// CHECK-LABEL: define void @va_v1i8(<1 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_v1i8(<1 x i8>* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -229,7 +229,7 @@
 // CHECK-VECTOR: ret <1 x i8> [[RET]]
 
 v2i8 va_v2i8(__builtin_va_list l) { return __builtin_va_arg(l, v2i8); }
-// CHECK-LABEL: define void @va_v2i8(<2 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_v2i8(<2 x i8>* noalias sret align 2 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -261,7 +261,7 @@
 // CHECK-VECTOR: ret <2 x i8> [[RET]]
 
 v4i8 va_v4i8(__builtin_va_list l) { return __builtin_va_arg(l, v4i8); }
-// CHECK-LABEL: define void @va_v4i8(<4 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_v4i8(<4 x i8>* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -293,7 +293,7 @@
 // CHECK-VECTOR: ret <4 x i8> [[RET]]
 
 v8i8 va_v8i8(__builtin_va_list l) { return __builtin_va_arg(l, v8i8); }
-// CHECK-LABEL: define void @va_v8i8(<8 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_v8i8(<8 x i8>* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -325,7 +325,7 @@
 // CHECK-VECTOR: ret <8 x i8> [[RET]]
 
 v16i8 va_v16i8(__builtin_va_list l) { return __builtin_va_arg(l, v16i8); }
-// CHECK-LABEL: define void @va_v16i8(<16 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_v16i8(<16 x i8>* noalias sret align 16 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -357,7 +357,7 @@
 // CHECK-VECTOR: ret <16 x i8> [[RET]]
 
 v32i8 va_v32i8(__builtin_va_list l) { return __builtin_va_arg(l, v32i8); }
-// CHECK-LABEL: define void @va_v32i8(<32 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_v32i8(<32 x i8>* noalias sret align 32 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -379,7 +379,7 @@
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi <32 x i8>** [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: [[INDIRECT_ARG:%[^ ]+]] = load <32 x i8>*, <32 x i8>** [[VA_ARG_ADDR]]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_v32i8(<32 x i8>* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_v32i8(<32 x i8>* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK-VECTOR: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK-VECTOR: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -403,7 +403,7 @@
 // CHECK-VECTOR: ret void
 
 struct agg_v1i8 va_agg_v1i8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_v1i8); }
-// CHECK-LABEL: define void @va_agg_v1i8(%struct.agg_v1i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_agg_v1i8(%struct.agg_v1i8* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -424,7 +424,7 @@
 // CHECK: store i8* [[OVERFLOW_ARG_AREA2]], i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi %struct.agg_v1i8* [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_agg_v1i8(%struct.agg_v1i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_agg_v1i8(%struct.agg_v1i8* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 2
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA:%[^ ]+]] = load i8*, i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK-VECTOR: [[MEM_ADDR:%[^ ]+]] = bitcast i8* [[OVERFLOW_ARG_AREA]] to %struct.agg_v1i8*
@@ -433,7 +433,7 @@
 // CHECK-VECTOR: ret void
 
 struct agg_v2i8 va_agg_v2i8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_v2i8); }
-// CHECK-LABEL: define void @va_agg_v2i8(%struct.agg_v2i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_agg_v2i8(%struct.agg_v2i8* noalias sret align 2 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -454,7 +454,7 @@
 // CHECK: store i8* [[OVERFLOW_ARG_AREA2]], i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi %struct.agg_v2i8* [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_agg_v2i8(%struct.agg_v2i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_agg_v2i8(%struct.agg_v2i8* noalias sret align 2 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 2
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA:%[^ ]+]] = load i8*, i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK-VECTOR: [[MEM_ADDR:%[^ ]+]] = bitcast i8* [[OVERFLOW_ARG_AREA]] to %struct.agg_v2i8*
@@ -463,7 +463,7 @@
 // CHECK-VECTOR: ret void
 
 struct agg_v4i8 va_agg_v4i8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_v4i8); }
-// CHECK-LABEL: define void @va_agg_v4i8(%struct.agg_v4i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_agg_v4i8(%struct.agg_v4i8* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -484,7 +484,7 @@
 // CHECK: store i8* [[OVERFLOW_ARG_AREA2]], i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi %struct.agg_v4i8* [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_agg_v4i8(%struct.agg_v4i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_agg_v4i8(%struct.agg_v4i8* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 2
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA:%[^ ]+]] = load i8*, i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK-VECTOR: [[MEM_ADDR:%[^ ]+]] = bitcast i8* [[OVERFLOW_ARG_AREA]] to %struct.agg_v4i8*
@@ -493,7 +493,7 @@
 // CHECK-VECTOR: ret void
 
 struct agg_v8i8 va_agg_v8i8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_v8i8); }
-// CHECK-LABEL: define void @va_agg_v8i8(%struct.agg_v8i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_agg_v8i8(%struct.agg_v8i8* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -514,7 +514,7 @@
 // CHECK: store i8* [[OVERFLOW_ARG_AREA2]], i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi %struct.agg_v8i8* [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_agg_v8i8(%struct.agg_v8i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_agg_v8i8(%struct.agg_v8i8* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 2
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA:%[^ ]+]] = load i8*, i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK-VECTOR: [[MEM_ADDR:%[^ ]+]] = bitcast i8* [[OVERFLOW_ARG_AREA]] to %struct.agg_v8i8*
@@ -523,7 +523,7 @@
 // CHECK-VECTOR: ret void
 
 struct agg_v16i8 va_agg_v16i8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_v16i8); }
-// CHECK-LABEL: define void @va_agg_v16i8(%struct.agg_v16i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_agg_v16i8(%struct.agg_v16i8* noalias sret align 16 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -545,7 +545,7 @@
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi %struct.agg_v16i8** [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: [[INDIRECT_ARG:%[^ ]+]] = load %struct.agg_v16i8*, %struct.agg_v16i8** [[VA_ARG_ADDR]]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_agg_v16i8(%struct.agg_v16i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_agg_v16i8(%struct.agg_v16i8* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 2
 // CHECK-VECTOR: [[OVERFLOW_ARG_AREA:%[^ ]+]] = load i8*, i8** [[OVERFLOW_ARG_AREA_PTR]]
 // CHECK-VECTOR: [[MEM_ADDR:%[^ ]+]] = bitcast i8* [[OVERFLOW_ARG_AREA]] to %struct.agg_v16i8*
@@ -554,7 +554,7 @@
 // CHECK-VECTOR: ret void
 
 struct agg_v32i8 va_agg_v32i8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_v32i8); }
-// CHECK-LABEL: define void @va_agg_v32i8(%struct.agg_v32i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_agg_v32i8(%struct.agg_v32i8* noalias sret align 32 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -576,7 +576,7 @@
 // CHECK: [[VA_ARG_ADDR:%[^ ]+]] = phi %struct.agg_v32i8** [ [[REG_ADDR]], %{{.*}} ], [ [[MEM_ADDR]], %{{.*}} ]
 // CHECK: [[INDIRECT_ARG:%[^ ]+]] = load %struct.agg_v32i8*, %struct.agg_v32i8** [[VA_ARG_ADDR]]
 // CHECK: ret void
-// CHECK-VECTOR-LABEL: define void @va_agg_v32i8(%struct.agg_v32i8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-VECTOR-LABEL: define void @va_agg_v32i8(%struct.agg_v32i8* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK-VECTOR: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK-VECTOR: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK-VECTOR: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
Index: clang/test/CodeGen/systemz-abi.c
===================================================================
--- clang/test/CodeGen/systemz-abi.c
+++ clang/test/CodeGen/systemz-abi.c
@@ -36,7 +36,7 @@
 // CHECK-LABEL: define i64 @pass_longlong(i64 %{{.*}})
 
 __int128 pass_int128(__int128 arg) { return arg; }
-// CHECK-LABEL: define void @pass_int128(i128* noalias sret %{{.*}}, i128* %0)
+// CHECK-LABEL: define void @pass_int128(i128* noalias sret align 16 %{{.*}}, i128* %0)
 
 float pass_float(float arg) { return arg; }
 // CHECK-LABEL: define float @pass_float(float %{{.*}})
@@ -45,114 +45,114 @@
 // CHECK-LABEL: define double @pass_double(double %{{.*}})
 
 long double pass_longdouble(long double arg) { return arg; }
-// CHECK-LABEL: define void @pass_longdouble(fp128* noalias sret %{{.*}}, fp128* %0)
+// CHECK-LABEL: define void @pass_longdouble(fp128* noalias sret align 8 %{{.*}}, fp128* %0)
 
 
 // Complex types
 
 _Complex char pass_complex_char(_Complex char arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_char({ i8, i8 }* noalias sret %{{.*}}, { i8, i8 }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_char({ i8, i8 }* noalias sret align 1 %{{.*}}, { i8, i8 }* %{{.*}}arg)
 
 _Complex short pass_complex_short(_Complex short arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_short({ i16, i16 }* noalias sret %{{.*}}, { i16, i16 }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_short({ i16, i16 }* noalias sret align 2 %{{.*}}, { i16, i16 }* %{{.*}}arg)
 
 _Complex int pass_complex_int(_Complex int arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_int({ i32, i32 }* noalias sret %{{.*}}, { i32, i32 }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_int({ i32, i32 }* noalias sret align 4 %{{.*}}, { i32, i32 }* %{{.*}}arg)
 
 _Complex long pass_complex_long(_Complex long arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_long({ i64, i64 }* noalias sret %{{.*}}, { i64, i64 }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_long({ i64, i64 }* noalias sret align 8 %{{.*}}, { i64, i64 }* %{{.*}}arg)
 
 _Complex long long pass_complex_longlong(_Complex long long arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_longlong({ i64, i64 }* noalias sret %{{.*}}, { i64, i64 }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_longlong({ i64, i64 }* noalias sret align 8 %{{.*}}, { i64, i64 }* %{{.*}}arg)
 
 _Complex float pass_complex_float(_Complex float arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_float({ float, float }* noalias sret %{{.*}}, { float, float }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_float({ float, float }* noalias sret align 4 %{{.*}}, { float, float }* %{{.*}}arg)
 
 _Complex double pass_complex_double(_Complex double arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_double({ double, double }* noalias sret %{{.*}}, { double, double }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_double({ double, double }* noalias sret align 8 %{{.*}}, { double, double }* %{{.*}}arg)
 
 _Complex long double pass_complex_longdouble(_Complex long double arg) { return arg; }
-// CHECK-LABEL: define void @pass_complex_longdouble({ fp128, fp128 }* noalias sret %{{.*}}, { fp128, fp128 }* %{{.*}}arg)
+// CHECK-LABEL: define void @pass_complex_longdouble({ fp128, fp128 }* noalias sret align 8 %{{.*}}, { fp128, fp128 }* %{{.*}}arg)
 
 
 // Aggregate types
 
 struct agg_1byte { char a[1]; };
 struct agg_1byte pass_agg_1byte(struct agg_1byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_1byte(%struct.agg_1byte* noalias sret %{{.*}}, i8 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_1byte(%struct.agg_1byte* noalias sret align 1 %{{.*}}, i8 %{{.*}})
 
 struct agg_2byte { char a[2]; };
 struct agg_2byte pass_agg_2byte(struct agg_2byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_2byte(%struct.agg_2byte* noalias sret %{{.*}}, i16 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_2byte(%struct.agg_2byte* noalias sret align 1 %{{.*}}, i16 %{{.*}})
 
 struct agg_3byte { char a[3]; };
 struct agg_3byte pass_agg_3byte(struct agg_3byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_3byte(%struct.agg_3byte* noalias sret %{{.*}}, %struct.agg_3byte* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_3byte(%struct.agg_3byte* noalias sret align 1 %{{.*}}, %struct.agg_3byte* %{{.*}})
 
 struct agg_4byte { char a[4]; };
 struct agg_4byte pass_agg_4byte(struct agg_4byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_4byte(%struct.agg_4byte* noalias sret %{{.*}}, i32 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_4byte(%struct.agg_4byte* noalias sret align 1 %{{.*}}, i32 %{{.*}})
 
 struct agg_5byte { char a[5]; };
 struct agg_5byte pass_agg_5byte(struct agg_5byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_5byte(%struct.agg_5byte* noalias sret %{{.*}}, %struct.agg_5byte* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_5byte(%struct.agg_5byte* noalias sret align 1 %{{.*}}, %struct.agg_5byte* %{{.*}})
 
 struct agg_6byte { char a[6]; };
 struct agg_6byte pass_agg_6byte(struct agg_6byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_6byte(%struct.agg_6byte* noalias sret %{{.*}}, %struct.agg_6byte* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_6byte(%struct.agg_6byte* noalias sret align 1 %{{.*}}, %struct.agg_6byte* %{{.*}})
 
 struct agg_7byte { char a[7]; };
 struct agg_7byte pass_agg_7byte(struct agg_7byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_7byte(%struct.agg_7byte* noalias sret %{{.*}}, %struct.agg_7byte* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_7byte(%struct.agg_7byte* noalias sret align 1 %{{.*}}, %struct.agg_7byte* %{{.*}})
 
 struct agg_8byte { char a[8]; };
 struct agg_8byte pass_agg_8byte(struct agg_8byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_8byte(%struct.agg_8byte* noalias sret %{{.*}}, i64 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_8byte(%struct.agg_8byte* noalias sret align 1 %{{.*}}, i64 %{{.*}})
 
 struct agg_16byte { char a[16]; };
 struct agg_16byte pass_agg_16byte(struct agg_16byte arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_16byte(%struct.agg_16byte* noalias sret %{{.*}}, %struct.agg_16byte* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_16byte(%struct.agg_16byte* noalias sret align 1 %{{.*}}, %struct.agg_16byte* %{{.*}})
 
 
 // Float-like aggregate types
 
 struct agg_float { float a; };
 struct agg_float pass_agg_float(struct agg_float arg) { return arg; }
-// HARD-FLOAT-LABEL: define void @pass_agg_float(%struct.agg_float* noalias sret %{{.*}}, float %{{.*}})
-// SOFT-FLOAT-LABEL: define void @pass_agg_float(%struct.agg_float* noalias sret %{{.*}}, i32 %{{.*}})
+// HARD-FLOAT-LABEL: define void @pass_agg_float(%struct.agg_float* noalias sret align 4 %{{.*}}, float %{{.*}})
+// SOFT-FLOAT-LABEL: define void @pass_agg_float(%struct.agg_float* noalias sret align 4 %{{.*}}, i32 %{{.*}})
 
 struct agg_double { double a; };
 struct agg_double pass_agg_double(struct agg_double arg) { return arg; }
-// HARD-FLOAT-LABEL: define void @pass_agg_double(%struct.agg_double* noalias sret %{{.*}}, double %{{.*}})
-// SOFT-FLOAT-LABEL: define void @pass_agg_double(%struct.agg_double* noalias sret %{{.*}}, i64 %{{.*}})
+// HARD-FLOAT-LABEL: define void @pass_agg_double(%struct.agg_double* noalias sret align 8 %{{.*}}, double %{{.*}})
+// SOFT-FLOAT-LABEL: define void @pass_agg_double(%struct.agg_double* noalias sret align 8 %{{.*}}, i64 %{{.*}})
 
 struct agg_longdouble { long double a; };
 struct agg_longdouble pass_agg_longdouble(struct agg_longdouble arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_longdouble(%struct.agg_longdouble* noalias sret %{{.*}}, %struct.agg_longdouble* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_longdouble(%struct.agg_longdouble* noalias sret align 8 %{{.*}}, %struct.agg_longdouble* %{{.*}})
 
 struct agg_float_a8 { float a __attribute__((aligned (8))); };
 struct agg_float_a8 pass_agg_float_a8(struct agg_float_a8 arg) { return arg; }
-// HARD-FLOAT-LABEL: define void @pass_agg_float_a8(%struct.agg_float_a8* noalias sret %{{.*}}, double %{{.*}})
-// SOFT-FLOAT-LABEL: define void @pass_agg_float_a8(%struct.agg_float_a8* noalias sret %{{.*}}, i64 %{{.*}})
+// HARD-FLOAT-LABEL: define void @pass_agg_float_a8(%struct.agg_float_a8* noalias sret align 8 %{{.*}}, double %{{.*}})
+// SOFT-FLOAT-LABEL: define void @pass_agg_float_a8(%struct.agg_float_a8* noalias sret align 8 %{{.*}}, i64 %{{.*}})
 
 struct agg_float_a16 { float a __attribute__((aligned (16))); };
 struct agg_float_a16 pass_agg_float_a16(struct agg_float_a16 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_float_a16(%struct.agg_float_a16* noalias sret %{{.*}}, %struct.agg_float_a16* %{{.*}})
+// CHECK-LABEL: define void @pass_agg_float_a16(%struct.agg_float_a16* noalias sret align 16 %{{.*}}, %struct.agg_float_a16* %{{.*}})
 
 
 // Verify that the following are *not* float-like aggregate types
 
 struct agg_nofloat1 { float a; float b; };
 struct agg_nofloat1 pass_agg_nofloat1(struct agg_nofloat1 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_nofloat1(%struct.agg_nofloat1* noalias sret %{{.*}}, i64 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_nofloat1(%struct.agg_nofloat1* noalias sret align 4 %{{.*}}, i64 %{{.*}})
 
 struct agg_nofloat2 { float a; int b; };
 struct agg_nofloat2 pass_agg_nofloat2(struct agg_nofloat2 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_nofloat2(%struct.agg_nofloat2* noalias sret %{{.*}}, i64 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_nofloat2(%struct.agg_nofloat2* noalias sret align 4 %{{.*}}, i64 %{{.*}})
 
 struct agg_nofloat3 { float a; int : 0; };
 struct agg_nofloat3 pass_agg_nofloat3(struct agg_nofloat3 arg) { return arg; }
-// CHECK-LABEL: define void @pass_agg_nofloat3(%struct.agg_nofloat3* noalias sret %{{.*}}, i32 %{{.*}})
+// CHECK-LABEL: define void @pass_agg_nofloat3(%struct.agg_nofloat3* noalias sret align 4 %{{.*}}, i32 %{{.*}})
 
 
 // Accessing variable argument lists
@@ -257,7 +257,7 @@
 // CHECK: ret double [[RET]]
 
 long double va_longdouble(__builtin_va_list l) { return __builtin_va_arg(l, long double); }
-// CHECK-LABEL: define void @va_longdouble(fp128* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}})
+// CHECK-LABEL: define void @va_longdouble(fp128* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}})
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -283,7 +283,7 @@
 // CHECK: ret void
 
 _Complex char va_complex_char(__builtin_va_list l) { return __builtin_va_arg(l, _Complex char); }
-// CHECK-LABEL: define void @va_complex_char({ i8, i8 }* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_complex_char({ i8, i8 }* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -307,7 +307,7 @@
 // CHECK: ret void
 
 struct agg_1byte va_agg_1byte(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_1byte); }
-// CHECK-LABEL: define void @va_agg_1byte(%struct.agg_1byte* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_1byte(%struct.agg_1byte* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -330,7 +330,7 @@
 // CHECK: ret void
 
 struct agg_2byte va_agg_2byte(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_2byte); }
-// CHECK-LABEL: define void @va_agg_2byte(%struct.agg_2byte* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_2byte(%struct.agg_2byte* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -353,7 +353,7 @@
 // CHECK: ret void
 
 struct agg_3byte va_agg_3byte(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_3byte); }
-// CHECK-LABEL: define void @va_agg_3byte(%struct.agg_3byte* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_3byte(%struct.agg_3byte* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -377,7 +377,7 @@
 // CHECK: ret void
 
 struct agg_4byte va_agg_4byte(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_4byte); }
-// CHECK-LABEL: define void @va_agg_4byte(%struct.agg_4byte* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_4byte(%struct.agg_4byte* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -400,7 +400,7 @@
 // CHECK: ret void
 
 struct agg_8byte va_agg_8byte(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_8byte); }
-// CHECK-LABEL: define void @va_agg_8byte(%struct.agg_8byte* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_8byte(%struct.agg_8byte* noalias sret align 1 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -423,7 +423,7 @@
 // CHECK: ret void
 
 struct agg_float va_agg_float(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_float); }
-// CHECK-LABEL: define void @va_agg_float(%struct.agg_float* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_float(%struct.agg_float* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // HARD-FLOAT: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 1
 // SOFT-FLOAT: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
@@ -449,7 +449,7 @@
 // CHECK: ret void
 
 struct agg_double va_agg_double(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_double); }
-// CHECK-LABEL: define void @va_agg_double(%struct.agg_double* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_double(%struct.agg_double* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // HARD-FLOAT: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 1
 // SOFT-FLOAT: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
@@ -475,7 +475,7 @@
 // CHECK: ret void
 
 struct agg_longdouble va_agg_longdouble(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_longdouble); }
-// CHECK-LABEL: define void @va_agg_longdouble(%struct.agg_longdouble* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_longdouble(%struct.agg_longdouble* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -499,7 +499,7 @@
 // CHECK: ret void
 
 struct agg_float_a8 va_agg_float_a8(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_float_a8); }
-// CHECK-LABEL: define void @va_agg_float_a8(%struct.agg_float_a8* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_float_a8(%struct.agg_float_a8* noalias sret align 8 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // HARD-FLOAT: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 1
 // SOFT-FLOAT: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
@@ -525,7 +525,7 @@
 // CHECK: ret void
 
 struct agg_float_a16 va_agg_float_a16(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_float_a16); }
-// CHECK-LABEL: define void @va_agg_float_a16(%struct.agg_float_a16* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_float_a16(%struct.agg_float_a16* noalias sret align 16 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -549,7 +549,7 @@
 // CHECK: ret void
 
 struct agg_nofloat1 va_agg_nofloat1(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_nofloat1); }
-// CHECK-LABEL: define void @va_agg_nofloat1(%struct.agg_nofloat1* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_nofloat1(%struct.agg_nofloat1* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -572,7 +572,7 @@
 // CHECK: ret void
 
 struct agg_nofloat2 va_agg_nofloat2(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_nofloat2); }
-// CHECK-LABEL: define void @va_agg_nofloat2(%struct.agg_nofloat2* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_nofloat2(%struct.agg_nofloat2* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
@@ -595,7 +595,7 @@
 // CHECK: ret void
 
 struct agg_nofloat3 va_agg_nofloat3(__builtin_va_list l) { return __builtin_va_arg(l, struct agg_nofloat3); }
-// CHECK-LABEL: define void @va_agg_nofloat3(%struct.agg_nofloat3* noalias sret %{{.*}}, %struct.__va_list_tag* %{{.*}}
+// CHECK-LABEL: define void @va_agg_nofloat3(%struct.agg_nofloat3* noalias sret align 4 %{{.*}}, %struct.__va_list_tag* %{{.*}}
 // CHECK: [[REG_COUNT_PTR:%[^ ]+]] = getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{.*}}, i32 0, i32 0
 // CHECK: [[REG_COUNT:%[^ ]+]] = load i64, i64* [[REG_COUNT_PTR]]
 // CHECK: [[FITS_IN_REGS:%[^ ]+]] = icmp ult i64 [[REG_COUNT]], 5
Index: clang/test/CodeGen/systemz-abi.cpp
===================================================================
--- clang/test/CodeGen/systemz-abi.cpp
+++ clang/test/CodeGen/systemz-abi.cpp
@@ -7,5 +7,5 @@
 
 struct agg_float_cpp { float a; int : 0; };
 struct agg_float_cpp pass_agg_float_cpp(struct agg_float_cpp arg) { return arg; }
-// CHECK-LABEL: define void @_Z18pass_agg_float_cpp13agg_float_cpp(%struct.agg_float_cpp* noalias sret %{{.*}}, float %{{.*}})
-// SOFT-FLOAT:  define void @_Z18pass_agg_float_cpp13agg_float_cpp(%struct.agg_float_cpp* noalias sret %{{.*}}, i32 %{{.*}})
+// CHECK-LABEL: define void @_Z18pass_agg_float_cpp13agg_float_cpp(%struct.agg_float_cpp* noalias sret align 4 %{{.*}}, float %{{.*}})
+// SOFT-FLOAT:  define void @_Z18pass_agg_float_cpp13agg_float_cpp(%struct.agg_float_cpp* noalias sret align 4 %{{.*}}, i32 %{{.*}})
Index: clang/test/CodeGen/systemz-inline-asm.c
===================================================================
--- clang/test/CodeGen/systemz-inline-asm.c
+++ clang/test/CodeGen/systemz-inline-asm.c
@@ -123,7 +123,7 @@
 long double test_f128(long double f, long double g) {
   asm("axbr %0, %2" : "=f" (f) : "0" (f), "f" (g));
   return f;
-// CHECK: define void @test_f128(fp128* noalias nocapture sret [[DEST:%.*]], fp128* nocapture readonly %0, fp128* nocapture readonly %1)
+// CHECK: define void @test_f128(fp128* noalias nocapture sret align 8 [[DEST:%.*]], fp128* nocapture readonly %0, fp128* nocapture readonly %1)
 // CHECK: %f = load fp128, fp128* %0
 // CHECK: %g = load fp128, fp128* %1
 // CHECK: [[RESULT:%.*]] = tail call fp128 asm "axbr $0, $2", "=f,0,f"(fp128 %f, fp128 %g)
Index: clang/test/CodeGen/vectorcall.c
===================================================================
--- clang/test/CodeGen/vectorcall.c
+++ clang/test/CodeGen/vectorcall.c
@@ -86,8 +86,8 @@
 // X64: define dso_local x86_vectorcallcc %struct.HVA4 @"\01hva6@@128"(%struct.HVA4 inreg %a.coerce, %struct.HVA4* %b)
 
 struct HVA5 __vectorcall hva7() {struct HVA5 a = {}; return a;}
-// X32: define dso_local x86_vectorcallcc void @"\01hva7@@0"(%struct.HVA5* inreg noalias sret %agg.result)
-// X64: define dso_local x86_vectorcallcc void @"\01hva7@@0"(%struct.HVA5* noalias sret %agg.result)
+// X32: define dso_local x86_vectorcallcc void @"\01hva7@@0"(%struct.HVA5* inreg noalias sret align 16 %agg.result)
+// X64: define dso_local x86_vectorcallcc void @"\01hva7@@0"(%struct.HVA5* noalias sret align 16 %agg.result)
 
 v4f32 __vectorcall hva8(v4f32 a, v4f32 b, v4f32 c, v4f32 d, int e, v4f32 f) {return f;}
 // X32: define dso_local x86_vectorcallcc <4 x float> @"\01hva8@@84"(<4 x float> inreg %a, <4 x float> inreg %b, <4 x float> inreg %c, <4 x float> inreg %d, i32 inreg %e, <4 x float> inreg %f)
Index: clang/test/CodeGen/wasm-arguments.c
===================================================================
--- clang/test/CodeGen/wasm-arguments.c
+++ clang/test/CodeGen/wasm-arguments.c
@@ -25,9 +25,9 @@
 void struct_arg(s1 i) {}
 
 // Structs should be returned sret and not simplified by the frontend.
-// WEBASSEMBLY32: define void @struct_ret(%struct.s1* noalias sret %agg.result)
+// WEBASSEMBLY32: define void @struct_ret(%struct.s1* noalias sret align 4 %agg.result)
 // WEBASSEMBLY32: ret void
-// WEBASSEMBLY64: define void @struct_ret(%struct.s1* noalias sret %agg.result)
+// WEBASSEMBLY64: define void @struct_ret(%struct.s1* noalias sret align 4 %agg.result)
 // WEBASSEMBLY64: ret void
 
 // Except with the experimental multivalue ABI, which returns structs by value
@@ -103,9 +103,9 @@
 void union_arg(union simple_union s) {}
 
 // Unions should be returned sret and not simplified by the frontend.
-// WEBASSEMBLY32: define void @union_ret(%union.simple_union* noalias sret %agg.result)
+// WEBASSEMBLY32: define void @union_ret(%union.simple_union* noalias sret align 4 %agg.result)
 // WEBASSEMBLY32: ret void
-// WEBASSEMBLY64: define void @union_ret(%union.simple_union* noalias sret %agg.result)
+// WEBASSEMBLY64: define void @union_ret(%union.simple_union* noalias sret align 4 %agg.result)
 // WEBASSEMBLY64: ret void
 
 // The experimental multivalue ABI returns them by value, though.
@@ -129,8 +129,8 @@
 void bitfield_arg(bitfield1 bf1) {}
 
 // And returned via sret pointers.
-// WEBASSEMBLY32: define void @bitfield_ret(%struct.bitfield1* noalias sret %agg.result)
-// WEBASSEMBLY64: define void @bitfield_ret(%struct.bitfield1* noalias sret %agg.result)
+// WEBASSEMBLY32: define void @bitfield_ret(%struct.bitfield1* noalias sret align 4 %agg.result)
+// WEBASSEMBLY64: define void @bitfield_ret(%struct.bitfield1* noalias sret align 4 %agg.result)
 
 // Except, of course, in the experimental multivalue ABI
 // EXPERIMENTAL-MV: define %struct.bitfield1 @bitfield_ret()
Index: clang/test/CodeGen/wasm-varargs.c
===================================================================
--- clang/test/CodeGen/wasm-varargs.c
+++ clang/test/CodeGen/wasm-varargs.c
@@ -80,7 +80,7 @@
   return v;
 }
 
-// CHECK:      define void @test_struct([[STRUCT_S:%[^,=]+]]*{{.*}} noalias sret [[AGG_RESULT:%.*]], i8*{{.*}} %fmt, ...) {{.*}} {
+// CHECK:      define void @test_struct([[STRUCT_S:%[^,=]+]]*{{.*}} noalias sret align 4 [[AGG_RESULT:%.*]], i8*{{.*}} %fmt, ...) {{.*}} {
 // CHECK:        [[FMT_ADDR:%[^,=]+]] = alloca i8*, align 4
 // CHECK-NEXT:   [[VA:%[^,=]+]] = alloca i8*, align 4
 // CHECK-NEXT:   store i8* %fmt, i8** [[FMT_ADDR]], align 4
@@ -112,7 +112,7 @@
   return v;
 }
 
-// CHECK:      define void @test_empty_struct([[STRUCT_S:%[^,=]+]]*{{.*}} noalias sret [[AGG_RESULT:%.*]], i8*{{.*}} %fmt, ...) {{.*}} {
+// CHECK:      define void @test_empty_struct([[STRUCT_S:%[^,=]+]]*{{.*}} noalias sret align 4 [[AGG_RESULT:%.*]], i8*{{.*}} %fmt, ...) {{.*}} {
 // CHECK:        [[FMT_ADDR:%[^,=]+]] = alloca i8*, align 4
 // CHECK-NEXT:   [[VA:%[^,=]+]] = alloca i8*, align 4
 // CHECK-NEXT:   [[U:%[^,=]+]] = alloca [[STRUCT_Z:%[^,=]+]], align 1
Index: clang/test/CodeGen/windows-struct-abi.c
===================================================================
--- clang/test/CodeGen/windows-struct-abi.c
+++ clang/test/CodeGen/windows-struct-abi.c
@@ -34,7 +34,7 @@
 
 struct f4 return_f4(void) { while (1); }
 
-// CHECK: define dso_local void @return_f4(%struct.f4* noalias sret %agg.result)
+// CHECK: define dso_local void @return_f4(%struct.f4* noalias sret align 4 %agg.result)
 
 void receive_f4(struct f4 a0) { }
 
Index: clang/test/CodeGen/x86_32-arguments-darwin.c
===================================================================
--- clang/test/CodeGen/x86_32-arguments-darwin.c
+++ clang/test/CodeGen/x86_32-arguments-darwin.c
@@ -71,7 +71,7 @@
 // Small vectors and 1 x {i64,double} are returned in registers
 
 // CHECK: i32 @f11()
-// CHECK: void @f12(<2 x i32>* noalias sret %agg.result)
+// CHECK: void @f12(<2 x i32>* noalias sret align 8 %agg.result)
 // CHECK: i64 @f13()
 // CHECK: i64 @f14()
 // CHECK: <2 x i64> @f15()
@@ -93,11 +93,11 @@
 // 128-bits).
 
 // CHECK: i32 @f17()
-// CHECK: void @f18(%{{.*}}* noalias sret %agg.result)
-// CHECK: void @f19(%{{.*}}* noalias sret %agg.result)
-// CHECK: void @f20(%{{.*}}* noalias sret %agg.result)
-// CHECK: void @f21(%{{.*}}* noalias sret %agg.result)
-// CHECK: void @f22(%{{.*}}* noalias sret %agg.result)
+// CHECK: void @f18(%{{.*}}* noalias sret align 8 %agg.result)
+// CHECK: void @f19(%{{.*}}* noalias sret align 8 %agg.result)
+// CHECK: void @f20(%{{.*}}* noalias sret align 8 %agg.result)
+// CHECK: void @f21(%{{.*}}* noalias sret align 16 %agg.result)
+// CHECK: void @f22(%{{.*}}* noalias sret align 16 %agg.result)
 struct { T11 a; } f17(void) { while (1) {} }
 struct { T12 a; } f18(void) { while (1) {} }
 struct { T13 a; } f19(void) { while (1) {} }
@@ -116,11 +116,11 @@
 
 // Small structures are handled recursively
 // CHECK: i32 @f26()
-// CHECK: void @f27(%struct.s27* noalias sret %agg.result)
+// CHECK: void @f27(%struct.s27* noalias sret align 1 %agg.result)
 struct s26 { struct { char a, b; } a; struct { char a, b; } b; } f26(void) { while (1) {} }
 struct s27 { struct { char a, b, c; } a; struct { char a; } b; } f27(void) { while (1) {} }
 
-// CHECK: void @f28(%struct.s28* noalias sret %agg.result)
+// CHECK: void @f28(%struct.s28* noalias sret align 4 %agg.result)
 struct s28 { int a; int b[]; } f28(void) { while (1) {} }
 
 // CHECK-LABEL: define i16 @f29()
@@ -150,7 +150,7 @@
 // CHECK-LABEL: define float @f37()
 struct s37 { float c[1][1]; } f37(void) { while (1) {} }
 
-// CHECK-LABEL: define void @f38(%struct.s38* noalias sret %agg.result)
+// CHECK-LABEL: define void @f38(%struct.s38* noalias sret align 2 %agg.result)
 struct s38 { char a[3]; short b; } f38(void) { while (1) {} }
 
 // CHECK-LABEL: define void @f39(%struct.s39* byval(%struct.s39) align 16 %x)
Index: clang/test/CodeGen/x86_32-arguments-iamcu.c
===================================================================
--- clang/test/CodeGen/x86_32-arguments-iamcu.c
+++ clang/test/CodeGen/x86_32-arguments-iamcu.c
@@ -58,7 +58,7 @@
 // CHECK-LABEL: define i64 @retPaddedStruct(i32 %r.coerce0, i32 %r.coerce1)
 st5_t retPaddedStruct(st5_t r) { return r; }
 
-// CHECK-LABEL: define void @retLargeStruct(%struct.st12_t* noalias sret %agg.result, i32 %i1, %struct.st12_t* byval(%struct.st12_t) align 4 %r)
+// CHECK-LABEL: define void @retLargeStruct(%struct.st12_t* noalias sret align 4 %agg.result, i32 %i1, %struct.st12_t* byval(%struct.st12_t) align 4 %r)
 st12_t retLargeStruct(int i1, st12_t r) { return r; }
 
 // CHECK-LABEL: define i32 @varArgs(i32 %i1, ...)
Index: clang/test/CodeGen/x86_64-arguments-nacl.c
===================================================================
--- clang/test/CodeGen/x86_64-arguments-nacl.c
+++ clang/test/CodeGen/x86_64-arguments-nacl.c
@@ -61,7 +61,7 @@
 
 // Check that sret parameter is accounted for when checking available integer
 // registers.
-// CHECK: define void @f13(%struct.s13_0* noalias sret %agg.result, i32 %a, i32 %b, i32 %c, i32 %d, {{.*}}* byval({{.*}}) align 8 %e, i32 %f)
+// CHECK: define void @f13(%struct.s13_0* noalias sret align 8 %agg.result, i32 %a, i32 %b, i32 %c, i32 %d, {{.*}}* byval({{.*}}) align 8 %e, i32 %f)
 
 struct s13_0 { long long f0[3]; };
 struct s13_1 { long long f0[2]; };
Index: clang/test/CodeGen/x86_64-arguments-win32.c
===================================================================
--- clang/test/CodeGen/x86_64-arguments-win32.c
+++ clang/test/CodeGen/x86_64-arguments-win32.c
@@ -27,5 +27,5 @@
 // CHECK-LABEL: define dso_local i64 @f7()
 _Complex float f7() { return 1.0; }
 
-// CHECK-LABEL: define dso_local void @f8({ double, double }* noalias sret %agg.result)
+// CHECK-LABEL: define dso_local void @f8({ double, double }* noalias sret align 8 %agg.result)
 _Complex double f8() { return 1.0; }
Index: clang/test/CodeGen/x86_64-arguments.c
===================================================================
--- clang/test/CodeGen/x86_64-arguments.c
+++ clang/test/CodeGen/x86_64-arguments.c
@@ -47,7 +47,7 @@
 
 // Test merging/passing of upper eightbyte with X87 class.
 //
-// CHECK-LABEL: define void @f8_1(%union.u8* noalias sret %agg.result)
+// CHECK-LABEL: define void @f8_1(%union.u8* noalias sret align 16 %agg.result)
 // CHECK-LABEL: define void @f8_2(%union.u8* byval(%union.u8) align 16 %a0)
 union u8 {
   long double a;
@@ -63,7 +63,7 @@
 struct s10 { int a; int b; int : 0; };
 void f10(struct s10 a0) {}
 
-// CHECK-LABEL: define void @f11(%union.anon* noalias sret %agg.result)
+// CHECK-LABEL: define void @f11(%union.anon* noalias sret align 16 %agg.result)
 union { long double a; float b; } f11() { while (1) {} }
 
 // CHECK-LABEL: define i32 @f12_0()
@@ -74,7 +74,7 @@
 
 // Check that sret parameter is accounted for when checking available integer
 // registers.
-// CHECK: define void @f13(%struct.s13_0* noalias sret %agg.result, i32 %a, i32 %b, i32 %c, i32 %d, {{.*}}* byval({{.*}}) align 8 %e, i32 %f)
+// CHECK: define void @f13(%struct.s13_0* noalias sret align 8 %agg.result, i32 %a, i32 %b, i32 %c, i32 %d, {{.*}}* byval({{.*}}) align 8 %e, i32 %f)
 
 struct s13_0 { long long f0[3]; };
 struct s13_1 { long long f0[2]; };
Index: clang/test/CodeGenCXX/arm-cc.cpp
===================================================================
--- clang/test/CodeGenCXX/arm-cc.cpp
+++ clang/test/CodeGenCXX/arm-cc.cpp
@@ -16,5 +16,5 @@
   zed(a);
 }
 
-// CHECK: declare void @_Z3fooPv(%class.SMLoc* sret, i8*)
+// CHECK: declare void @_Z3fooPv(%class.SMLoc* sret align 4, i8*)
 // CHECK: declare void @_Z3zed5SMLoc(%class.SMLoc*)
Index: clang/test/CodeGenCXX/builtin-source-location.cpp
===================================================================
--- clang/test/CodeGenCXX/builtin-source-location.cpp
+++ clang/test/CodeGenCXX/builtin-source-location.cpp
@@ -65,7 +65,7 @@
 //
 // CHECK-GLOBAL-TWO: define internal void @__cxx_global_var_init()
 // CHECK-GLOBAL-TWO-NOT: ret
-// CHECK-GLOBAL-TWO: call void @_ZN15source_location11bad_currentEjjPKcS1_(%struct.source_location* sret @runtime_init_global,
+// CHECK-GLOBAL-TWO: call void @_ZN15source_location11bad_currentEjjPKcS1_(%struct.source_location* sret align 8 @runtime_init_global,
 // CHECK-GLOBAL-TWO-SAME: i32 1100, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]],
 #line 1100 "test_runtime_init.cpp"
 SL runtime_init_global = SL::bad_current();
@@ -77,7 +77,7 @@
 // CHECK-LOCAL-ONE-DAG: @[[FILE:.*]] = {{.*}}c"test_current.cpp\00"
 // CHECK-LOCAL-ONE-DAG: @[[FUNC:.*]] = {{.*}}c"test_function\00"
 //
-// CHECK-LOCAL-ONE:  call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %local,
+// CHECK-LOCAL-ONE:  call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %local,
 // CHECK-LOCAL-ONE-SAME: i32 2100, i32 {{[0-9]+}},
 // CHECK-LOCAL-ONE-SAME: {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]],
 #line 2100 "test_current.cpp"
@@ -102,7 +102,7 @@
 // CHECK-CTOR-GLOBAL: define internal void @__cxx_global_var_init.{{[0-9]+}}()
 // CHECK-CTOR-GLOBAL-NOT: ret
 //
-// CHECK-CTOR-GLOBAL: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[TMP_ONE:[^,]*]],
+// CHECK-CTOR-GLOBAL: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[TMP_ONE:[^,]*]],
 // CHECK-CTOR-GLOBAL-SAME: i32 3400, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]],
 // CHECK-CTOR-GLOBAL-NEXT: call void @_ZN8TestInitC1E15source_location(%struct.TestInit* @GlobalInitVal, %struct.source_location* {{.*}}%[[TMP_ONE]])
 #line 3400 "GlobalInitVal.cpp"
@@ -117,7 +117,7 @@
 // CHECK-CTOR-LOCAL: define void @test_init_function()
 // CHECK-CTOR-LOCAL-NOT: ret
 //
-// CHECK-CTOR-LOCAL: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[TMP:[^,]*]],
+// CHECK-CTOR-LOCAL: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[TMP:[^,]*]],
 // CHECK-CTOR-LOCAL-SAME: i32 3500, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]],
 // CHECK-CTOR-LOCAL-NEXT: call void @_ZN8TestInitC1E15source_location(%struct.TestInit* %init_local, %struct.source_location* {{.*}}%[[TMP]])
 #line 3500 "LocalInitVal.cpp"
@@ -153,7 +153,7 @@
 // CHECK-CONSTEXPR-LOCAL-DAG: @[[FILE:.*]] = {{.*}}c"ConstexprLocal.cpp\00"
 //
 // CHECK-CONSTEXPR-LOCAL: define void @test_init_function_constexpr()
-// CHECK-CONSTEXPR-LOCAL: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[TMP:[^,]*]],
+// CHECK-CONSTEXPR-LOCAL: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[TMP:[^,]*]],
 // CHECK-CONSTEXPR-LOCAL-SAME: i32 4600, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]]
 // CHECK-CONSTEXPR-LOCAL: call void @_ZN17TestInitConstexprC1E15source_location(%struct.TestInitConstexpr* %local_val, {{.*}}%[[TMP]])
 #line 4600 "ConstexprLocal.cpp"
@@ -189,7 +189,7 @@
 //
 // CHECK-AGG-BRACE: define void @test_agg_init()
 // CHECK-AGG-BRACE: %[[I2:.*]] = getelementptr inbounds %struct.TestInitAgg, %struct.TestInitAgg* %local_brace_init, i32 0, i32 1
-// CHECK-AGG-BRACE-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[I2]],
+// CHECK-AGG-BRACE-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[I2]],
 // CHECK-AGG-BRACE-SAME: i32 5700, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]]
 #line 5600 "BraceInitStart.cpp"
   TestInitAgg local_brace_init{
@@ -203,7 +203,7 @@
 //
 // CHECK-AGG-EQUAL: define void @test_agg_init()
 // CHECK-AGG-EQUAL: %[[I2:.*]] = getelementptr inbounds %struct.TestInitAgg, %struct.TestInitAgg* %local_equal_init, i32 0, i32 1
-// CHECK-AGG-EQUAL-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[I2]],
+// CHECK-AGG-EQUAL-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[I2]],
 // CHECK-AGG-EQUAL-SAME: i32 5900, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]]
 #line 5800 "EqualInitStart.cpp"
   TestInitAgg local_equal_init =
@@ -220,11 +220,11 @@
 // CHECK-AGG-LIST: define void @test_agg_init()
 //
 // CHECK-AGG-LIST: %[[I1:.*]] =  getelementptr inbounds %struct.TestInitAgg, %struct.TestInitAgg* %local_list_init, i32 0, i32 0
-// CHECK-AGG-LIST-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[I1]],
+// CHECK-AGG-LIST-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[I1]],
 // CHECK-AGG-LIST-SAME: i32 6100, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE_ELEM]], {{[^@]*}}@[[FUNC]]
 //
 // CHECK-AGG-LIST: %[[I2:.*]] = getelementptr inbounds %struct.TestInitAgg, %struct.TestInitAgg* %local_list_init, i32 0, i32 1
-// CHECK-AGG-LIST-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[I2]],
+// CHECK-AGG-LIST-NEXT: call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[I2]],
 // CHECK-AGG-LIST-SAME: i32 6200, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE_DEFAULT]], {{[^@]*}}@[[FUNC]]
 #line 6000 "InitListStart.cpp"
   TestInitAgg local_list_init =
@@ -258,7 +258,7 @@
 // CHECK-TEMPL-NEXT: entry:
 // CHECK-TEMPL-NOT: ret
 //
-// CHECK-TEMPL:  call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret %[[TMP:[^,]*]],
+// CHECK-TEMPL:  call void @_ZN15source_location7currentEjjPKcS1_(%struct.source_location* sret align 8 %[[TMP:[^,]*]],
 // CHECK-TEMPL-SAME: i32 7300, i32 {{[0-9]+}}, {{[^@]*}}@[[FILE]], {{[^@]*}}@[[FUNC]]
 #line 7300 "local_templ.cpp"
   TestTemplate<T, V> local_templ;
Index: clang/test/CodeGenCXX/call-with-static-chain.cpp
===================================================================
--- clang/test/CodeGenCXX/call-with-static-chain.cpp
+++ clang/test/CodeGenCXX/call-with-static-chain.cpp
@@ -25,8 +25,8 @@
   // CHECK64: call i32 bitcast (i32 (i64, i64, i64, i64, i64, i64, %struct.A*)* @f1 to i32 (i8*, i64, i64, i64, i64, i64, i64, %struct.A*)*)(i8* nest bitcast (i32 (i64, i64, i64, i64, i64, i64, %struct.A*)* @f1 to i8*)
   __builtin_call_with_static_chain(f1(a, a, a, a), f1);
 
-  // CHECK32: call void bitcast (void (%struct.B*)* @f2 to void (%struct.B*, i8*)*)(%struct.B* sret %{{[0-9a-z]+}}, i8* nest bitcast (void (%struct.B*)* @f2 to i8*))
-  // CHECK64: call void bitcast (void (%struct.B*)* @f2 to void (%struct.B*, i8*)*)(%struct.B* sret %{{[0-9a-z]+}}, i8* nest bitcast (void (%struct.B*)* @f2 to i8*))
+  // CHECK32: call void bitcast (void (%struct.B*)* @f2 to void (%struct.B*, i8*)*)(%struct.B* sret align 4 %{{[0-9a-z]+}}, i8* nest bitcast (void (%struct.B*)* @f2 to i8*))
+  // CHECK64: call void bitcast (void (%struct.B*)* @f2 to void (%struct.B*, i8*)*)(%struct.B* sret align 8 %{{[0-9a-z]+}}, i8* nest bitcast (void (%struct.B*)* @f2 to i8*))
   __builtin_call_with_static_chain(f2(), f2);
 
   // CHECK32: call i64 bitcast (i64 ()* @f3 to i64 (i8*)*)(i8* nest bitcast (i64 ()* @f3 to i8*))
Index: clang/test/CodeGenCXX/conditional-gnu-ext.cpp
===================================================================
--- clang/test/CodeGenCXX/conditional-gnu-ext.cpp
+++ clang/test/CodeGenCXX/conditional-gnu-ext.cpp
@@ -94,7 +94,7 @@
   B test1() {
     // CHECK-LABEL:    define void @_ZN5test35test1Ev(
     // CHECK:      [[TEMP:%.*]] = alloca [[B]],
-    // CHECK:      call  void @_ZN5test312test1_helperEv([[B]]* sret [[TEMP]])
+    // CHECK:      call  void @_ZN5test312test1_helperEv([[B]]* sret align 1 [[TEMP]])
     // CHECK-NEXT: [[BOOL:%.*]] = call zeroext i1 @_ZN5test31BcvbEv([[B]]* [[TEMP]])
     // CHECK-NEXT: br i1 [[BOOL]]
     // CHECK:      call void @_ZN5test31BC1ERKS0_([[B]]* [[RESULT:%.*]], [[B]]* dereferenceable({{[0-9]+}}) [[TEMP]])
@@ -115,7 +115,7 @@
     // CHECK-NEXT: [[T0:%.*]] = load [[B]]*, [[B]]** [[X]]
     // CHECK-NEXT: [[BOOL:%.*]] = call zeroext i1 @_ZN5test31BcvbEv([[B]]* [[T0]])
     // CHECK-NEXT: br i1 [[BOOL]]
-    // CHECK:      call void @_ZN5test31BcvNS_1AEEv([[A:%.*]]* sret [[RESULT:%.*]], [[B]]* [[T0]])
+    // CHECK:      call void @_ZN5test31BcvNS_1AEEv([[A:%.*]]* sret align 1 [[RESULT:%.*]], [[B]]* [[T0]])
     // CHECK-NEXT: br label
     // CHECK:      call void @_ZN5test31AC1Ev([[A]]* [[RESULT]])
     // CHECK-NEXT: br label
@@ -126,10 +126,10 @@
   A test3() {
     // CHECK-LABEL:    define void @_ZN5test35test3Ev(
     // CHECK:      [[TEMP:%.*]] = alloca [[B]],
-    // CHECK:      call  void @_ZN5test312test3_helperEv([[B]]* sret [[TEMP]])
+    // CHECK:      call  void @_ZN5test312test3_helperEv([[B]]* sret align 1 [[TEMP]])
     // CHECK-NEXT: [[BOOL:%.*]] = call zeroext i1 @_ZN5test31BcvbEv([[B]]* [[TEMP]])
     // CHECK-NEXT: br i1 [[BOOL]]
-    // CHECK:      call void @_ZN5test31BcvNS_1AEEv([[A]]* sret [[RESULT:%.*]], [[B]]* [[TEMP]])
+    // CHECK:      call void @_ZN5test31BcvNS_1AEEv([[A]]* sret align 1 [[RESULT:%.*]], [[B]]* [[TEMP]])
     // CHECK-NEXT: br label
     // CHECK:      call void @_ZN5test31AC1Ev([[A]]* [[RESULT]])
     // CHECK-NEXT: br label
Index: clang/test/CodeGenCXX/cxx1z-copy-omission.cpp
===================================================================
--- clang/test/CodeGenCXX/cxx1z-copy-omission.cpp
+++ clang/test/CodeGenCXX/cxx1z-copy-omission.cpp
@@ -19,7 +19,7 @@
   // CHECK: %[[A:.*]] = alloca
   // CHECK-NOT: alloca
   // CHECK-NOT: call
-  // CHECK: call {{.*}} @_Z1fv({{.*}}* sret %[[A]])
+  // CHECK: call {{.*}} @_Z1fv({{.*}}* sret align 4 %[[A]])
   A a = A( A{ f() } );
   // CHECK-NOT: call
 
@@ -40,7 +40,7 @@
   // CHECK-NOT: alloca
   // CHECK-NOT: call
 
-  // CHECK: call {{.*}} @_Z1fv({{.*}}* sret %[[A]])
+  // CHECK: call {{.*}} @_Z1fv({{.*}}* sret align 4 %[[A]])
   // CHECK-NOT: call
   // CHECK: call {{.*}} @_Z1f1A({{.*}}* %[[A]])
   f(f());
Index: clang/test/CodeGenCXX/cxx1z-lambda-star-this.cpp
===================================================================
--- clang/test/CodeGenCXX/cxx1z-lambda-star-this.cpp
+++ clang/test/CodeGenCXX/cxx1z-lambda-star-this.cpp
@@ -10,7 +10,7 @@
 int X = A{}.foo()();
 } //end ns1
 
-//CHECK: @"?foo@A@@QAE?A?<auto>@@XZ"(%struct.A* %this, %class.anon* noalias sret %[[A_LAMBDA_RETVAL:.*]])
+//CHECK: @"?foo@A@@QAE?A?<auto>@@XZ"(%struct.A* %this, %class.anon* noalias sret align 8 %[[A_LAMBDA_RETVAL:.*]])
 // get the first object with the closure type, which is of type 'struct.A'
 //CHECK: %[[I0:.+]] = getelementptr inbounds %[[A_LAMBDA]], %[[A_LAMBDA]]* %[[A_LAMBDA_RETVAL]], i32 0, i32 0
 //CHECK: %[[I1:.+]] = bitcast %struct.A* %[[I0]] to i8*
@@ -26,6 +26,6 @@
 namespace ns2 {
 int X = B{}.bar()();
 }
-//CHECK: @"?bar@B@@QAE?A?<auto>@@XZ"(%struct.B* %this, %class.anon.0* noalias sret %agg.result)
+//CHECK: @"?bar@B@@QAE?A?<auto>@@XZ"(%struct.B* %this, %class.anon.0* noalias sret align 4 %agg.result)
 //CHECK: %[[I20:.+]] = getelementptr inbounds %class.anon.0, %class.anon.0* %agg.result, i32 0, i32 0
 //CHECK: store %struct.B* %this1, %struct.B** %[[I20]], align 4
Index: clang/test/CodeGenCXX/exceptions.cpp
===================================================================
--- clang/test/CodeGenCXX/exceptions.cpp
+++ clang/test/CodeGenCXX/exceptions.cpp
@@ -146,12 +146,12 @@
     // CHECK:      [[NEW:%.*]] = call noalias nonnull i8* @_Znwm(i64 8)
     // CHECK-NEXT: store i1 true, i1* [[ACTIVE]]
     // CHECK-NEXT: [[CAST:%.*]] = bitcast i8* [[NEW]] to [[A]]*
-    // CHECK-NEXT: invoke void @_ZN5test15makeBEv([[B:%.*]]* sret [[T0:%.*]])
+    // CHECK-NEXT: invoke void @_ZN5test15makeBEv([[B:%.*]]* sret align 4 [[T0:%.*]])
     // CHECK:      [[T1:%.*]] = invoke i32 @_ZN5test11BcviEv([[B]]* [[T0]])
     // CHECK:      invoke void @_ZN5test11AC1Ei([[A]]* [[CAST]], i32 [[T1]])
     // CHECK:      store i1 false, i1* [[ACTIVE]]
     // CHECK-NEXT: store [[A]]* [[CAST]], [[A]]** [[X]], align 8
-    // CHECK:      invoke void @_ZN5test15makeBEv([[B:%.*]]* sret [[T2:%.*]])
+    // CHECK:      invoke void @_ZN5test15makeBEv([[B:%.*]]* sret align 4 [[T2:%.*]])
     // CHECK:      [[RET:%.*]] = load [[A]]*, [[A]]** [[X]], align 8
 
     // CHECK98:      invoke void @_ZN5test11BD1Ev([[B]]* [[T2]])
@@ -239,7 +239,7 @@
     // CHECK-NEXT: store i8* [[FOO]], i8** [[SAVED1]]
     // CHECK-NEXT: store i1 true, i1* [[CLEANUPACTIVE]]
     // CHECK-NEXT: [[CAST:%.*]] = bitcast i8* [[NEW]] to [[A]]*
-    // CHECK-NEXT: invoke void @_ZN5test35makeAEv([[A]]* sret [[CAST]])
+    // CHECK-NEXT: invoke void @_ZN5test35makeAEv([[A]]* sret align 8 [[CAST]])
     // CHECK: br label
     //   -> cond.end
             new(foo(),10.0) A(makeA()) :
Index: clang/test/CodeGenCXX/homogeneous-aggregates.cpp
===================================================================
--- clang/test/CodeGenCXX/homogeneous-aggregates.cpp
+++ clang/test/CodeGenCXX/homogeneous-aggregates.cpp
@@ -38,10 +38,10 @@
 struct I3 : Base2 {};
 struct D5 : I1, I2, I3 {}; // homogeneous aggregate
 
-// PPC: define void @_Z7func_D12D1(%struct.D1* noalias sret %agg.result, [3 x i64] %x.coerce)
-// ARM32: define arm_aapcs_vfpcc void @_Z7func_D12D1(%struct.D1* noalias sret %agg.result, [3 x i64] %x.coerce)
-// ARM64: define void @_Z7func_D12D1(%struct.D1* noalias sret %agg.result, %struct.D1* %x)
-// X64: define dso_local x86_vectorcallcc void @"\01_Z7func_D12D1@@24"(%struct.D1* noalias sret %agg.result, %struct.D1* %x)
+// PPC: define void @_Z7func_D12D1(%struct.D1* noalias sret align 8 %agg.result, [3 x i64] %x.coerce)
+// ARM32: define arm_aapcs_vfpcc void @_Z7func_D12D1(%struct.D1* noalias sret align 8 %agg.result, [3 x i64] %x.coerce)
+// ARM64: define void @_Z7func_D12D1(%struct.D1* noalias sret align 8 %agg.result, %struct.D1* %x)
+// X64: define dso_local x86_vectorcallcc void @"\01_Z7func_D12D1@@24"(%struct.D1* noalias sret align 8 %agg.result, %struct.D1* %x)
 D1 CC func_D1(D1 x) { return x; }
 
 // PPC: define [3 x double] @_Z7func_D22D2([3 x double] %x.coerce)
@@ -50,9 +50,9 @@
 // X64: define dso_local x86_vectorcallcc %struct.D2 @"\01_Z7func_D22D2@@24"(%struct.D2 inreg %x.coerce)
 D2 CC func_D2(D2 x) { return x; }
 
-// PPC: define void @_Z7func_D32D3(%struct.D3* noalias sret %agg.result, [4 x i64] %x.coerce)
-// ARM32: define arm_aapcs_vfpcc void @_Z7func_D32D3(%struct.D3* noalias sret %agg.result, [4 x i64] %x.coerce)
-// ARM64: define void @_Z7func_D32D3(%struct.D3* noalias sret %agg.result, %struct.D3* %x)
+// PPC: define void @_Z7func_D32D3(%struct.D3* noalias sret align 8 %agg.result, [4 x i64] %x.coerce)
+// ARM32: define arm_aapcs_vfpcc void @_Z7func_D32D3(%struct.D3* noalias sret align 8 %agg.result, [4 x i64] %x.coerce)
+// ARM64: define void @_Z7func_D32D3(%struct.D3* noalias sret align 8 %agg.result, %struct.D3* %x)
 D3 CC func_D3(D3 x) { return x; }
 
 // PPC: define [4 x double] @_Z7func_D42D4([4 x double] %x.coerce)
Index: clang/test/CodeGenCXX/lambda-expressions.cpp
===================================================================
--- clang/test/CodeGenCXX/lambda-expressions.cpp
+++ clang/test/CodeGenCXX/lambda-expressions.cpp
@@ -194,8 +194,8 @@
 // CHECK-NEXT: call i32 @"_ZZ1fvENK3$_6clEii"
 // CHECK-NEXT: ret i32
 
-// CHECK-LABEL: define internal void @"_ZZ1hvEN4$_118__invokeEv"(%struct.A* noalias sret %agg.result) {{.*}} {
-// CHECK: call void @"_ZZ1hvENK4$_11clEv"(%struct.A* sret %agg.result,
+// CHECK-LABEL: define internal void @"_ZZ1hvEN4$_118__invokeEv"(%struct.A* noalias sret align 1 %agg.result) {{.*}} {
+// CHECK: call void @"_ZZ1hvENK4$_11clEv"(%struct.A* sret align 1 %agg.result,
 // CHECK-NEXT: ret void
 struct A { ~A(); };
 void h() {
Index: clang/test/CodeGenCXX/microsoft-abi-byval-sret.cpp
===================================================================
--- clang/test/CodeGenCXX/microsoft-abi-byval-sret.cpp
+++ clang/test/CodeGenCXX/microsoft-abi-byval-sret.cpp
@@ -49,7 +49,7 @@
 }
 
 // CHECK-LABEL: define dso_local x86_fastcallcc void @"?qux@B@@QAI?AUA@@U2@@Z"
-// CHECK:       (%struct.B* inreg %this, %struct.A* inreg noalias sret %agg.result, <{ %struct.A }>* inalloca %0)
+// CHECK:       (%struct.B* inreg %this, %struct.A* inreg noalias sret align 4 %agg.result, <{ %struct.A }>* inalloca %0)
 // CHECK:   ret void
 
 int main() {
@@ -67,4 +67,4 @@
 // CHECK: call x86_stdcallcc %struct.A* @"?baz@B@@QAG?AUA@@U2@@Z"
 // CHECK:       (<{ %struct.B*, %struct.A*, %struct.A }>* inalloca %{{[^,]*}})
 // CHECK: call x86_fastcallcc void @"?qux@B@@QAI?AUA@@U2@@Z"
-// CHECK:       (%struct.B* inreg %{{[^,]*}}, %struct.A* inreg sret %{{.*}}, <{ %struct.A }>* inalloca %{{[^,]*}})
+// CHECK:       (%struct.B* inreg %{{[^,]*}}, %struct.A* inreg sret align 4 %{{.*}}, <{ %struct.A }>* inalloca %{{[^,]*}})
Index: clang/test/CodeGenCXX/microsoft-abi-byval-thunks.cpp
===================================================================
--- clang/test/CodeGenCXX/microsoft-abi-byval-thunks.cpp
+++ clang/test/CodeGenCXX/microsoft-abi-byval-thunks.cpp
@@ -86,10 +86,10 @@
 // CHECK32-NEXT: ret %"struct.sret_thunk::Agg"* %[[rv]]
 
 // CHECK64-LABEL: define linkonce_odr dso_local void @"?foo@C@sret_thunk@@W7EAA?AUAgg@2@U32@@Z"
-// CHECK64:             (%"struct.sret_thunk::C"* %this, %"struct.sret_thunk::Agg"* noalias sret %agg.result, %"struct.sret_thunk::Agg"* %x)
+// CHECK64:             (%"struct.sret_thunk::C"* %this, %"struct.sret_thunk::Agg"* noalias sret align 4 %agg.result, %"struct.sret_thunk::Agg"* %x)
 // CHECK64:   getelementptr i8, i8* %{{.*}}, i32 -8
 // CHECK64:   call void @"?foo@C@sret_thunk@@UEAA?AUAgg@2@U32@@Z"
-// CHECK64:       (%"struct.sret_thunk::C"* %{{.*}}, %"struct.sret_thunk::Agg"* sret %agg.result, %"struct.sret_thunk::Agg"* %x)
+// CHECK64:       (%"struct.sret_thunk::C"* %{{.*}}, %"struct.sret_thunk::Agg"* sret align 4 %agg.result, %"struct.sret_thunk::Agg"* %x)
 // CHECK64-NOT: call
 // CHECK64:   ret void
 }
Index: clang/test/CodeGenCXX/microsoft-abi-cdecl-method-sret.cpp
===================================================================
--- clang/test/CodeGenCXX/microsoft-abi-cdecl-method-sret.cpp
+++ clang/test/CodeGenCXX/microsoft-abi-cdecl-method-sret.cpp
@@ -19,9 +19,9 @@
 S C::cdecl_sret() { return S(); }
 S C::byval_and_sret(S a) { return S(); }
 
-// CHECK: define dso_local void @"?variadic_sret@C@@QAA?AUS@@PBDZZ"(%struct.C* %this, %struct.S* noalias sret %agg.result, i8* %f, ...)
-// CHECK: define dso_local void @"?cdecl_sret@C@@QAA?AUS@@XZ"(%struct.C* %this, %struct.S* noalias sret %agg.result)
-// CHECK: define dso_local void @"?byval_and_sret@C@@QAA?AUS@@U2@@Z"(%struct.C* %this, %struct.S* noalias sret %agg.result, %struct.S* byval(%struct.S) align 4 %a)
+// CHECK: define dso_local void @"?variadic_sret@C@@QAA?AUS@@PBDZZ"(%struct.C* %this, %struct.S* noalias sret align 4 %agg.result, i8* %f, ...)
+// CHECK: define dso_local void @"?cdecl_sret@C@@QAA?AUS@@XZ"(%struct.C* %this, %struct.S* noalias sret align 4 %agg.result)
+// CHECK: define dso_local void @"?byval_and_sret@C@@QAA?AUS@@U2@@Z"(%struct.C* %this, %struct.S* noalias sret align 4 %agg.result, %struct.S* byval(%struct.S) align 4 %a)
 
 int main() {
   C c;
@@ -41,4 +41,4 @@
 S A::f(int x) {
   return S();
 }
-// CHECK-LABEL: define dso_local x86_fastcallcc void @"?f@A@@QAI?AUS@@H@Z"(%struct.A* inreg %this, %struct.S* inreg noalias sret %agg.result, i32 %x)
+// CHECK-LABEL: define dso_local x86_fastcallcc void @"?f@A@@QAI?AUS@@H@Z"(%struct.A* inreg %this, %struct.S* inreg noalias sret align 4 %agg.result, i32 %x)
Index: clang/test/CodeGenCXX/microsoft-abi-eh-cleanups.cpp
===================================================================
--- clang/test/CodeGenCXX/microsoft-abi-eh-cleanups.cpp
+++ clang/test/CodeGenCXX/microsoft-abi-eh-cleanups.cpp
@@ -18,9 +18,9 @@
 // WIN32-LABEL: define dso_local void @"?HasEHCleanup@@YAXXZ"() {{.*}} {
 // WIN32:   %[[base:.*]] = call i8* @llvm.stacksave()
 //    If this call throws, we have to restore the stack.
-// WIN32:   call void @"?getA@@YA?AUA@@XZ"(%struct.A* sret %{{.*}})
+// WIN32:   call void @"?getA@@YA?AUA@@XZ"(%struct.A* sret align 4 %{{.*}})
 //    If this call throws, we have to cleanup the first temporary.
-// WIN32:   invoke void @"?getA@@YA?AUA@@XZ"(%struct.A* sret %{{.*}})
+// WIN32:   invoke void @"?getA@@YA?AUA@@XZ"(%struct.A* sret align 4 %{{.*}})
 //    If this call throws, we have to cleanup the stacksave.
 // WIN32:   call i32 @"?TakesTwo@@YAHUA@@0@Z"
 // WIN32:   call void @llvm.stackrestore
Index: clang/test/CodeGenCXX/microsoft-abi-sret-and-byval.cpp
===================================================================
--- clang/test/CodeGenCXX/microsoft-abi-sret-and-byval.cpp
+++ clang/test/CodeGenCXX/microsoft-abi-sret-and-byval.cpp
@@ -84,45 +84,45 @@
 
 // Returning structs that fit into a register.
 Small small_return() { return Small(); }
-// LINUX-LABEL: define void @_Z12small_returnv(%struct.Small* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z12small_returnv(%struct.Small* noalias sret align 4 %agg.result)
 // WIN32: define dso_local i32 @"?small_return@@YA?AUSmall@@XZ"()
 // WIN64: define dso_local i32 @"?small_return@@YA?AUSmall@@XZ"()
 
 Medium medium_return() { return Medium(); }
-// LINUX-LABEL: define void @_Z13medium_returnv(%struct.Medium* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z13medium_returnv(%struct.Medium* noalias sret align 4 %agg.result)
 // WIN32: define dso_local i64 @"?medium_return@@YA?AUMedium@@XZ"()
 // WIN64: define dso_local i64 @"?medium_return@@YA?AUMedium@@XZ"()
 
 // Returning structs that fit into a register but are not POD.
 SmallCpp11NotCpp03Pod small_non_pod_return() { return SmallCpp11NotCpp03Pod(); }
-// LINUX-LABEL: define void @_Z20small_non_pod_returnv(%struct.SmallCpp11NotCpp03Pod* noalias sret %agg.result)
-// WIN32: define dso_local void @"?small_non_pod_return@@YA?AUSmallCpp11NotCpp03Pod@@XZ"(%struct.SmallCpp11NotCpp03Pod* noalias sret %agg.result)
-// WIN64: define dso_local void @"?small_non_pod_return@@YA?AUSmallCpp11NotCpp03Pod@@XZ"(%struct.SmallCpp11NotCpp03Pod* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z20small_non_pod_returnv(%struct.SmallCpp11NotCpp03Pod* noalias sret align 4 %agg.result)
+// WIN32: define dso_local void @"?small_non_pod_return@@YA?AUSmallCpp11NotCpp03Pod@@XZ"(%struct.SmallCpp11NotCpp03Pod* noalias sret align 4 %agg.result)
+// WIN64: define dso_local void @"?small_non_pod_return@@YA?AUSmallCpp11NotCpp03Pod@@XZ"(%struct.SmallCpp11NotCpp03Pod* noalias sret align 4 %agg.result)
 
 SmallWithCtor small_with_ctor_return() { return SmallWithCtor(); }
-// LINUX-LABEL: define void @_Z22small_with_ctor_returnv(%struct.SmallWithCtor* noalias sret %agg.result)
-// WIN32: define dso_local void @"?small_with_ctor_return@@YA?AUSmallWithCtor@@XZ"(%struct.SmallWithCtor* noalias sret %agg.result)
-// WIN64: define dso_local void @"?small_with_ctor_return@@YA?AUSmallWithCtor@@XZ"(%struct.SmallWithCtor* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z22small_with_ctor_returnv(%struct.SmallWithCtor* noalias sret align 4 %agg.result)
+// WIN32: define dso_local void @"?small_with_ctor_return@@YA?AUSmallWithCtor@@XZ"(%struct.SmallWithCtor* noalias sret align 4 %agg.result)
+// WIN64: define dso_local void @"?small_with_ctor_return@@YA?AUSmallWithCtor@@XZ"(%struct.SmallWithCtor* noalias sret align 4 %agg.result)
 // FIXME: The 'sret' mark here doesn't seem to be enough to convince LLVM to
 // preserve the hidden sret pointer in R0 across the function.
-// WOA: define dso_local arm_aapcs_vfpcc void @"?small_with_ctor_return@@YA?AUSmallWithCtor@@XZ"(%struct.SmallWithCtor* noalias sret %agg.result)
+// WOA: define dso_local arm_aapcs_vfpcc void @"?small_with_ctor_return@@YA?AUSmallWithCtor@@XZ"(%struct.SmallWithCtor* noalias sret align 4 %agg.result)
 
 SmallWithVftable small_with_vftable_return() { return SmallWithVftable(); }
-// LINUX-LABEL: define void @_Z25small_with_vftable_returnv(%struct.SmallWithVftable* noalias sret %agg.result)
-// WIN32: define dso_local void @"?small_with_vftable_return@@YA?AUSmallWithVftable@@XZ"(%struct.SmallWithVftable* noalias sret %agg.result)
-// WIN64: define dso_local void @"?small_with_vftable_return@@YA?AUSmallWithVftable@@XZ"(%struct.SmallWithVftable* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z25small_with_vftable_returnv(%struct.SmallWithVftable* noalias sret align 4 %agg.result)
+// WIN32: define dso_local void @"?small_with_vftable_return@@YA?AUSmallWithVftable@@XZ"(%struct.SmallWithVftable* noalias sret align 4 %agg.result)
+// WIN64: define dso_local void @"?small_with_vftable_return@@YA?AUSmallWithVftable@@XZ"(%struct.SmallWithVftable* noalias sret align 8 %agg.result)
 
 MediumWithCopyCtor medium_with_copy_ctor_return() { return MediumWithCopyCtor(); }
-// LINUX-LABEL: define void @_Z28medium_with_copy_ctor_returnv(%struct.MediumWithCopyCtor* noalias sret %agg.result)
-// WIN32: define dso_local void @"?medium_with_copy_ctor_return@@YA?AUMediumWithCopyCtor@@XZ"(%struct.MediumWithCopyCtor* noalias sret %agg.result)
-// WIN64: define dso_local void @"?medium_with_copy_ctor_return@@YA?AUMediumWithCopyCtor@@XZ"(%struct.MediumWithCopyCtor* noalias sret %agg.result)
-// WOA: define dso_local arm_aapcs_vfpcc void @"?medium_with_copy_ctor_return@@YA?AUMediumWithCopyCtor@@XZ"(%struct.MediumWithCopyCtor* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z28medium_with_copy_ctor_returnv(%struct.MediumWithCopyCtor* noalias sret align 4 %agg.result)
+// WIN32: define dso_local void @"?medium_with_copy_ctor_return@@YA?AUMediumWithCopyCtor@@XZ"(%struct.MediumWithCopyCtor* noalias sret align 4 %agg.result)
+// WIN64: define dso_local void @"?medium_with_copy_ctor_return@@YA?AUMediumWithCopyCtor@@XZ"(%struct.MediumWithCopyCtor* noalias sret align 4 %agg.result)
+// WOA: define dso_local arm_aapcs_vfpcc void @"?medium_with_copy_ctor_return@@YA?AUMediumWithCopyCtor@@XZ"(%struct.MediumWithCopyCtor* noalias sret align 4 %agg.result)
 
 // Returning a large struct that doesn't fit into a register.
 Big big_return() { return Big(); }
-// LINUX-LABEL: define void @_Z10big_returnv(%struct.Big* noalias sret %agg.result)
-// WIN32: define dso_local void @"?big_return@@YA?AUBig@@XZ"(%struct.Big* noalias sret %agg.result)
-// WIN64: define dso_local void @"?big_return@@YA?AUBig@@XZ"(%struct.Big* noalias sret %agg.result)
+// LINUX-LABEL: define void @_Z10big_returnv(%struct.Big* noalias sret align 4 %agg.result)
+// WIN32: define dso_local void @"?big_return@@YA?AUBig@@XZ"(%struct.Big* noalias sret align 4 %agg.result)
+// WIN64: define dso_local void @"?big_return@@YA?AUBig@@XZ"(%struct.Big* noalias sret align 4 %agg.result)
 
 
 void small_arg(Small s) {}
@@ -181,7 +181,7 @@
 
 // Test that the eligible non-aggregate is passed directly, but returned
 // indirectly on ARM64 Windows.
-// WOA64: define dso_local void @"?small_arg_with_private_member@@YA?AUSmallWithPrivate@@U1@@Z"(%struct.SmallWithPrivate* inreg noalias sret %agg.result, i64 %s.coerce) {{.*}} {
+// WOA64: define dso_local void @"?small_arg_with_private_member@@YA?AUSmallWithPrivate@@U1@@Z"(%struct.SmallWithPrivate* inreg noalias sret align 4 %agg.result, i64 %s.coerce) {{.*}} {
 SmallWithPrivate small_arg_with_private_member(SmallWithPrivate s) { return s; }
 
 void call_small_arg_with_dtor() {
@@ -281,24 +281,24 @@
 class Class {
  public:
   Small thiscall_method_small() { return Small(); }
-  // LINUX: define {{.*}} void @_ZN5Class21thiscall_method_smallEv(%struct.Small* noalias sret %agg.result, %class.Class* %this)
-  // WIN32: define {{.*}} x86_thiscallcc void @"?thiscall_method_small@Class@@QAE?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret %agg.result)
-  // WIN64: define linkonce_odr dso_local void @"?thiscall_method_small@Class@@QEAA?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret %agg.result)
+  // LINUX: define {{.*}} void @_ZN5Class21thiscall_method_smallEv(%struct.Small* noalias sret align 4 %agg.result, %class.Class* %this)
+  // WIN32: define {{.*}} x86_thiscallcc void @"?thiscall_method_small@Class@@QAE?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret align 4 %agg.result)
+  // WIN64: define linkonce_odr dso_local void @"?thiscall_method_small@Class@@QEAA?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret align 4 %agg.result)
 
   SmallWithCtor thiscall_method_small_with_ctor() { return SmallWithCtor(); }
-  // LINUX: define {{.*}} void @_ZN5Class31thiscall_method_small_with_ctorEv(%struct.SmallWithCtor* noalias sret %agg.result, %class.Class* %this)
-  // WIN32: define {{.*}} x86_thiscallcc void @"?thiscall_method_small_with_ctor@Class@@QAE?AUSmallWithCtor@@XZ"(%class.Class* %this, %struct.SmallWithCtor* noalias sret %agg.result)
-  // WIN64: define linkonce_odr dso_local void @"?thiscall_method_small_with_ctor@Class@@QEAA?AUSmallWithCtor@@XZ"(%class.Class* %this, %struct.SmallWithCtor* noalias sret %agg.result)
+  // LINUX: define {{.*}} void @_ZN5Class31thiscall_method_small_with_ctorEv(%struct.SmallWithCtor* noalias sret align 4 %agg.result, %class.Class* %this)
+  // WIN32: define {{.*}} x86_thiscallcc void @"?thiscall_method_small_with_ctor@Class@@QAE?AUSmallWithCtor@@XZ"(%class.Class* %this, %struct.SmallWithCtor* noalias sret align 4 %agg.result)
+  // WIN64: define linkonce_odr dso_local void @"?thiscall_method_small_with_ctor@Class@@QEAA?AUSmallWithCtor@@XZ"(%class.Class* %this, %struct.SmallWithCtor* noalias sret align 4 %agg.result)
 
   Small __cdecl cdecl_method_small() { return Small(); }
-  // LINUX: define {{.*}} void @_ZN5Class18cdecl_method_smallEv(%struct.Small* noalias sret %agg.result, %class.Class* %this)
-  // WIN32: define {{.*}} void @"?cdecl_method_small@Class@@QAA?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret %agg.result)
-  // WIN64: define linkonce_odr dso_local void @"?cdecl_method_small@Class@@QEAA?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret %agg.result)
+  // LINUX: define {{.*}} void @_ZN5Class18cdecl_method_smallEv(%struct.Small* noalias sret align 4 %agg.result, %class.Class* %this)
+  // WIN32: define {{.*}} void @"?cdecl_method_small@Class@@QAA?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret align 4 %agg.result)
+  // WIN64: define linkonce_odr dso_local void @"?cdecl_method_small@Class@@QEAA?AUSmall@@XZ"(%class.Class* %this, %struct.Small* noalias sret align 4 %agg.result)
 
   Big __cdecl cdecl_method_big() { return Big(); }
-  // LINUX: define {{.*}} void @_ZN5Class16cdecl_method_bigEv(%struct.Big* noalias sret %agg.result, %class.Class* %this)
-  // WIN32: define {{.*}} void @"?cdecl_method_big@Class@@QAA?AUBig@@XZ"(%class.Class* %this, %struct.Big* noalias sret %agg.result)
-  // WIN64: define linkonce_odr dso_local void @"?cdecl_method_big@Class@@QEAA?AUBig@@XZ"(%class.Class* %this, %struct.Big* noalias sret %agg.result)
+  // LINUX: define {{.*}} void @_ZN5Class16cdecl_method_bigEv(%struct.Big* noalias sret align 4 %agg.result, %class.Class* %this)
+  // WIN32: define {{.*}} void @"?cdecl_method_big@Class@@QAA?AUBig@@XZ"(%class.Class* %this, %struct.Big* noalias sret align 4 %agg.result)
+  // WIN64: define linkonce_odr dso_local void @"?cdecl_method_big@Class@@QEAA?AUBig@@XZ"(%class.Class* %this, %struct.Big* noalias sret align 4 %agg.result)
 
   void thiscall_method_arg(Empty s) {}
   // LINUX: define {{.*}} void @_ZN5Class19thiscall_method_argE5Empty(%class.Class* %this)
Index: clang/test/CodeGenCXX/microsoft-abi-vmemptr-conflicts.cpp
===================================================================
--- clang/test/CodeGenCXX/microsoft-abi-vmemptr-conflicts.cpp
+++ clang/test/CodeGenCXX/microsoft-abi-vmemptr-conflicts.cpp
@@ -65,7 +65,7 @@
 
 // CHECK-LABEL: define dso_local void @"?f@sret@@YAXPAUC@1@@Z"(%"struct.sret::C"* %c)
 // CHECK: call x86_thiscallcc i32 bitcast (void (%"struct.sret::C"*, ...)* @"??_9C@sret@@$BA@AE" to i32 (%"struct.sret::C"*)*)(%"struct.sret::C"* %{{.*}})
-// CHECK: call x86_thiscallcc void bitcast (void (%"struct.sret::C"*, ...)* @"??_9C@sret@@$BA@AE" to void (%"struct.sret::C"*, %"struct.sret::Big"*)*)(%"struct.sret::C"* %{{.*}}, %"struct.sret::Big"* sret %{{.*}})
+// CHECK: call x86_thiscallcc void bitcast (void (%"struct.sret::C"*, ...)* @"??_9C@sret@@$BA@AE" to void (%"struct.sret::C"*, %"struct.sret::Big"*)*)(%"struct.sret::C"* %{{.*}}, %"struct.sret::Big"* sret align 4 %{{.*}})
 
 // CHECK-LABEL: define linkonce_odr x86_thiscallcc void @"??_9C@sret@@$BA@AE"(%"struct.sret::C"* %this, ...) {{.*}} comdat
 // CHECK: musttail call x86_thiscallcc void (%"struct.sret::C"*, ...) %{{.*}}(%"struct.sret::C"* %{{.*}}, ...)
Index: clang/test/CodeGenCXX/regcall.cpp
===================================================================
--- clang/test/CodeGenCXX/regcall.cpp
+++ clang/test/CodeGenCXX/regcall.cpp
@@ -74,8 +74,8 @@
 // CHECK-WIN32-DAG: define dso_local x86_regcallcc zeroext i1 @"??8@Yw_NABVtest_class@@0@Z"
 
 test_class __regcall operator""_test_class (unsigned long long) { ++x; return test_class{};}
-// CHECK-LIN64-DAG: define x86_regcallcc void @_Zli11_test_classy(%class.test_class* noalias sret %agg.result, i64 %0)
-// CHECK-LIN32-DAG: define x86_regcallcc void @_Zli11_test_classy(%class.test_class* inreg noalias sret %agg.result, i64 %0)
+// CHECK-LIN64-DAG: define x86_regcallcc void @_Zli11_test_classy(%class.test_class* noalias sret align 4 %agg.result, i64 %0)
+// CHECK-LIN32-DAG: define x86_regcallcc void @_Zli11_test_classy(%class.test_class* inreg noalias sret align 4 %agg.result, i64 %0)
 // CHECK-WIN64-DAG: ??__K_test_class@@Yw?AVtest_class@@_K@Z"
 // CHECK-WIN32-DAG: ??__K_test_class@@Yw?AVtest_class@@_K@Z"
 
@@ -99,7 +99,7 @@
 long double _Complex __regcall foo(long double _Complex f) {
   return f;
 }
-// CHECK-LIN64-DAG: define x86_regcallcc void @_Z15__regcall3__fooCe({ x86_fp80, x86_fp80 }* noalias sret %agg.result, { x86_fp80, x86_fp80 }* byval({ x86_fp80, x86_fp80 }) align 16 %f)
-// CHECK-LIN32-DAG: define x86_regcallcc void @_Z15__regcall3__fooCe({ x86_fp80, x86_fp80 }* inreg noalias sret %agg.result, { x86_fp80, x86_fp80 }* byval({ x86_fp80, x86_fp80 }) align 4 %f)
+// CHECK-LIN64-DAG: define x86_regcallcc void @_Z15__regcall3__fooCe({ x86_fp80, x86_fp80 }* noalias sret align 16 %agg.result, { x86_fp80, x86_fp80 }* byval({ x86_fp80, x86_fp80 }) align 16 %f)
+// CHECK-LIN32-DAG: define x86_regcallcc void @_Z15__regcall3__fooCe({ x86_fp80, x86_fp80 }* inreg noalias sret align 4 %agg.result, { x86_fp80, x86_fp80 }* byval({ x86_fp80, x86_fp80 }) align 4 %f)
 // CHECK-WIN64-DAG: define dso_local x86_regcallcc { double, double } @"?foo@@YwU?$_Complex@O@__clang@@U12@@Z"(double %f.0, double %f.1)
 // CHECK-WIN32-DAG: define dso_local x86_regcallcc { double, double } @"?foo@@YwU?$_Complex@O@__clang@@U12@@Z"(double %f.0, double %f.1)
Index: clang/test/CodeGenCXX/stack-reuse-miscompile.cpp
===================================================================
--- clang/test/CodeGenCXX/stack-reuse-miscompile.cpp
+++ clang/test/CodeGenCXX/stack-reuse-miscompile.cpp
@@ -39,7 +39,7 @@
 // CHECK: call void @llvm.lifetime.start.p0i8(i64 16, i8* [[T3i8]])
 // CHECK: [[T5:%.*]] = call %class.T* @_ZN1TC1E1S(%class.T* [[T3]], [2 x i32] %{{.*}})
 //
-// CHECK: call void @_ZNK1T6concatERKS_(%class.T* sret [[T1]], %class.T* [[T2]], %class.T* dereferenceable(16) [[T3]])
+// CHECK: call void @_ZNK1T6concatERKS_(%class.T* sret align 4 [[T1]], %class.T* [[T2]], %class.T* dereferenceable(16) [[T3]])
 // CHECK: [[T6:%.*]] = call i8* @_ZNK1T3strEv(%class.T* [[T1]])
 //
 // CHECK: call void @llvm.lifetime.end.p0i8(
Index: clang/test/CodeGenCXX/stack-reuse.cpp
===================================================================
--- clang/test/CodeGenCXX/stack-reuse.cpp
+++ clang/test/CodeGenCXX/stack-reuse.cpp
@@ -135,7 +135,7 @@
 // CHECK: [[T2:%.*]] = alloca %struct.Combiner
 // CHECK: [[T1:%.*]] = alloca %struct.Combiner
 // CHECK: [[T3:%.*]] = call %struct.Combiner* @_ZN8CombinerC1E7S_large(%struct.Combiner* nonnull [[T1]], [9 x i32] %s.coerce)
-// CHECK: call void @_ZN8Combiner1fEv(%struct.Combiner* nonnull sret [[T2]], %struct.Combiner* nonnull [[T1]])
+// CHECK: call void @_ZN8Combiner1fEv(%struct.Combiner* nonnull sret align 4 [[T2]], %struct.Combiner* nonnull [[T1]])
 // CHECK: [[T4:%.*]] = getelementptr inbounds %struct.Combiner, %struct.Combiner* [[T2]], i32 0, i32 0, i32 0, i32 0
 // CHECK: [[T5:%.*]] = load i32, i32* [[T4]]
 // CHECK: ret i32 [[T5]]
Index: clang/test/CodeGenCXX/temporaries.cpp
===================================================================
--- clang/test/CodeGenCXX/temporaries.cpp
+++ clang/test/CodeGenCXX/temporaries.cpp
@@ -403,13 +403,13 @@
     // CHECK-NEXT: call void @_ZN7Elision1AC1Ev([[A]]* [[I]])
     A i = (foo(), A());
 
-    // CHECK-NEXT: call void @_ZN7Elision4fooAEv([[A]]* sret [[T0]])
+    // CHECK-NEXT: call void @_ZN7Elision4fooAEv([[A]]* sret align 8 [[T0]])
     // CHECK-NEXT: call void @_ZN7Elision1AC1Ev([[A]]* [[J]])
     // CHECK-NEXT: call void @_ZN7Elision1AD1Ev([[A]]* [[T0]])
     A j = (fooA(), A());
 
     // CHECK-NEXT: call void @_ZN7Elision1AC1Ev([[A]]* [[T1]])
-    // CHECK-NEXT: call void @_ZN7Elision4fooAEv([[A]]* sret [[K]])
+    // CHECK-NEXT: call void @_ZN7Elision4fooAEv([[A]]* sret align 8 [[K]])
     // CHECK-NEXT: call void @_ZN7Elision1AD1Ev([[A]]* [[T1]])
     A k = (A(), fooA());
 
@@ -436,7 +436,7 @@
     // CHECK-NEXT: call void @_ZN7Elision1AD1Ev([[A]]* [[I]])
   }
 
-  // CHECK: define void @_ZN7Elision5test2Ev([[A]]* noalias sret
+  // CHECK: define void @_ZN7Elision5test2Ev([[A]]* noalias sret align 8
   A test2() {
     // CHECK:      call void @_ZN7Elision3fooEv()
     // CHECK-NEXT: call void @_ZN7Elision1AC1Ev([[A]]* [[RET:%.*]])
@@ -444,7 +444,7 @@
     return (foo(), A());
   }
 
-  // CHECK: define void @_ZN7Elision5test3EiNS_1AE([[A]]* noalias sret
+  // CHECK: define void @_ZN7Elision5test3EiNS_1AE([[A]]* noalias sret align 8
   A test3(int v, A x) {
     if (v < 5)
     // CHECK:      call void @_ZN7Elision1AC1Ev([[A]]* [[RET:%.*]])
@@ -485,7 +485,7 @@
   }
 
   // rdar://problem/8433352
-  // CHECK: define void @_ZN7Elision5test5Ev([[A]]* noalias sret
+  // CHECK: define void @_ZN7Elision5test5Ev([[A]]* noalias sret align 8
   struct B { A a; B(); };
   A test5() {
     // CHECK:      [[AT0:%.*]] = alloca [[A]], align 8
@@ -523,7 +523,7 @@
   void test6(const C *x) {
     // CHECK:      [[T0:%.*]] = alloca [[A]], align 8
     // CHECK:      [[X:%.*]] = load [[C]]*, [[C]]** {{%.*}}, align 8
-    // CHECK-NEXT: call void @_ZNK7Elision1CcvNS_1AEEv([[A]]* sret [[T0]], [[C]]* [[X]])
+    // CHECK-NEXT: call void @_ZNK7Elision1CcvNS_1AEEv([[A]]* sret align 8 [[T0]], [[C]]* [[X]])
     // CHECK-NEXT: call void @_ZNK7Elision1A3fooEv([[A]]* [[T0]])
     // CHECK-NEXT: call void @_ZN7Elision1AD1Ev([[A]]* [[T0]])
     // CHECK-NEXT: ret void
Index: clang/test/CodeGenCXX/thiscall-struct-return.cpp
===================================================================
--- clang/test/CodeGenCXX/thiscall-struct-return.cpp
+++ clang/test/CodeGenCXX/thiscall-struct-return.cpp
@@ -34,8 +34,8 @@
 // CHECK: call void @_ZN1CC1Ev(%class.C* [[C:%.+]])
   C c;
 
-// CHECK: call x86_thiscallcc void @_ZNK1C5SmallEv(%struct.S* sret %{{.+}}, %class.C* [[C]])
+// CHECK: call x86_thiscallcc void @_ZNK1C5SmallEv(%struct.S* sret align 4 %{{.+}}, %class.C* [[C]])
   (void)c.Small();
-// CHECK: call x86_thiscallcc void @_ZNK1C6MediumEv(%struct.M* sret %{{.+}}, %class.C* [[C]])
+// CHECK: call x86_thiscallcc void @_ZNK1C6MediumEv(%struct.M* sret align 4 %{{.+}}, %class.C* [[C]])
   (void)c.Medium();
 }
Index: clang/test/CodeGenCXX/thunk-returning-memptr.cpp
===================================================================
--- clang/test/CodeGenCXX/thunk-returning-memptr.cpp
+++ clang/test/CodeGenCXX/thunk-returning-memptr.cpp
@@ -23,5 +23,5 @@
 // Because of the tail call, the return value cannot be copied into a local
 // alloca. (PR39901)
 
-// CHECK-LABEL: define linkonce_odr void @_ZThn4_N1C1fEv({ i32, i32 }* noalias sret %agg.result, %struct.C* %this)
-// CHECK: tail call void @_ZN1C1fEv({ i32, i32 }* sret %agg.result
+// CHECK-LABEL: define linkonce_odr void @_ZThn4_N1C1fEv({ i32, i32 }* noalias sret align 4 %agg.result, %struct.C* %this)
+// CHECK: tail call void @_ZN1C1fEv({ i32, i32 }* sret align 4 %agg.result
Index: clang/test/CodeGenCXX/thunks.cpp
===================================================================
--- clang/test/CodeGenCXX/thunks.cpp
+++ clang/test/CodeGenCXX/thunks.cpp
@@ -206,13 +206,13 @@
   // CHECK-LABEL: define void @_ZThn16_N5Test66Thunks1fEv
 	// CHECK-DBG-NOT: dbg.declare
   // CHECK-NOT: memcpy
-  // CHECK: {{call void @_ZN5Test66Thunks1fEv.*sret}}
+  // CHECK: {{call void @_ZN5Test66Thunks1fEv.*sret align 1}}
   // CHECK: ret void
   X Thunks::f() { return X(); }
 
-  // WIN64-LABEL: define linkonce_odr dso_local void @"?f@Thunks@Test6@@WBA@EAA?AUX@2@XZ"({{.*}} sret %{{.*}})
+  // WIN64-LABEL: define linkonce_odr dso_local void @"?f@Thunks@Test6@@WBA@EAA?AUX@2@XZ"({{.*}} sret align 1 %{{.*}})
   // WIN64-NOT: memcpy
-  // WIN64: tail call void @"?f@Thunks@Test6@@UEAA?AUX@2@XZ"({{.*}} sret %{{.*}})
+  // WIN64: tail call void @"?f@Thunks@Test6@@UEAA?AUX@2@XZ"({{.*}} sret align 1 %{{.*}})
 }
 
 namespace Test7 {
Index: clang/test/CodeGenCXX/trivial_abi.cpp
===================================================================
--- clang/test/CodeGenCXX/trivial_abi.cpp
+++ clang/test/CodeGenCXX/trivial_abi.cpp
@@ -126,7 +126,7 @@
 void testParamLarge(Large a) noexcept {
 }
 
-// CHECK: define void @_Z15testReturnLargev(%[[STRUCT_LARGE:.*]]* noalias sret %[[AGG_RESULT:.*]])
+// CHECK: define void @_Z15testReturnLargev(%[[STRUCT_LARGE:.*]]* noalias sret align 8 %[[AGG_RESULT:.*]])
 // CHECK: %[[CALL:.*]] = call %[[STRUCT_LARGE]]* @_ZN5LargeC1Ev(%[[STRUCT_LARGE]]* %[[AGG_RESULT]])
 // CHECK: ret void
 // CHECK: }
@@ -153,7 +153,7 @@
 
 // CHECK: define void @_Z14testCallLarge1v()
 // CHECK: %[[AGG_TMP:.*]] = alloca %[[STRUCT_LARGE:.*]], align 8
-// CHECK: call void @_Z15testReturnLargev(%[[STRUCT_LARGE]]* sret %[[AGG_TMP]])
+// CHECK: call void @_Z15testReturnLargev(%[[STRUCT_LARGE]]* sret align 8 %[[AGG_TMP]])
 // CHECK: call void @_Z14testParamLarge5Large(%[[STRUCT_LARGE]]* %[[AGG_TMP]])
 // CHECK: ret void
 // CHECK: }
@@ -164,7 +164,7 @@
 
 // CHECK: define void @_Z16testIgnoredLargev()
 // CHECK: %[[AGG_TMP_ENSURED:.*]] = alloca %[[STRUCT_LARGE:.*]], align 8
-// CHECK: call void @_Z15testReturnLargev(%[[STRUCT_LARGE]]* sret %[[AGG_TMP_ENSURED]])
+// CHECK: call void @_Z15testReturnLargev(%[[STRUCT_LARGE]]* sret align 8 %[[AGG_TMP_ENSURED]])
 // CHECK: %[[CALL:.*]] = call %[[STRUCT_LARGE]]* @_ZN5LargeD1Ev(%[[STRUCT_LARGE]]* %[[AGG_TMP_ENSURED]])
 // CHECK: ret void
 // CHECK: }
@@ -186,7 +186,7 @@
   return t;
 }
 
-// CHECK: define void @_Z23testReturnHasNonTrivialv(%[[STRUCT_NONTRIVIAL:.*]]* noalias sret %[[AGG_RESULT:.*]])
+// CHECK: define void @_Z23testReturnHasNonTrivialv(%[[STRUCT_NONTRIVIAL:.*]]* noalias sret align 4 %[[AGG_RESULT:.*]])
 // CHECK: %[[CALL:.*]] = call %[[STRUCT_NONTRIVIAL]]* @_ZN10NonTrivialC1Ev(%[[STRUCT_NONTRIVIAL]]* %[[AGG_RESULT]])
 // CHECK: ret void
 // CHECK: }
Index: clang/test/CodeGenCXX/unknown-anytype.cpp
===================================================================
--- clang/test/CodeGenCXX/unknown-anytype.cpp
+++ clang/test/CodeGenCXX/unknown-anytype.cpp
@@ -71,7 +71,7 @@
 };
 extern "C" __unknown_anytype test7_any(int);
 Test7 test7() {
-  // COMMON: call void @test7_any({{%.*}}* sret {{%.*}}, i32 5)
+  // COMMON: call void @test7_any({{%.*}}* sret align 1 {{%.*}}, i32 5)
   return (Test7) test7_any(5);
 }
 
Index: clang/test/CodeGenCXX/wasm-args-returns.cpp
===================================================================
--- clang/test/CodeGenCXX/wasm-args-returns.cpp
+++ clang/test/CodeGenCXX/wasm-args-returns.cpp
@@ -30,52 +30,52 @@
   double d, e;
 };
 test(two_fields);
-// CHECK: define void @_Z7forward10two_fields(%struct.two_fields* noalias nocapture sret %{{.*}}, %struct.two_fields* nocapture readonly byval(%struct.two_fields) align 8 %{{.*}})
+// CHECK: define void @_Z7forward10two_fields(%struct.two_fields* noalias nocapture sret align 8 %{{.*}}, %struct.two_fields* nocapture readonly byval(%struct.two_fields) align 8 %{{.*}})
 //
 // CHECK: define void @_Z15test_two_fieldsv()
 // CHECK: %[[tmp:.*]] = alloca %struct.two_fields, align 8
-// CHECK: call void @_Z14def_two_fieldsv(%struct.two_fields* nonnull sret %[[tmp]])
+// CHECK: call void @_Z14def_two_fieldsv(%struct.two_fields* nonnull sret align 8 %[[tmp]])
 // CHECK: call void @_Z3use10two_fields(%struct.two_fields* nonnull byval(%struct.two_fields) align 8 %[[tmp]])
 // CHECK: ret void
 //
 // CHECK: declare void @_Z3use10two_fields(%struct.two_fields* byval(%struct.two_fields) align 8)
-// CHECK: declare void @_Z14def_two_fieldsv(%struct.two_fields* sret)
+// CHECK: declare void @_Z14def_two_fieldsv(%struct.two_fields* sret align 8)
 
 struct copy_ctor {
   double d;
   copy_ctor(copy_ctor const &);
 };
 test(copy_ctor);
-// CHECK: define void @_Z7forward9copy_ctor(%struct.copy_ctor* noalias sret %{{.*}}, %struct.copy_ctor* nonnull %{{.*}})
+// CHECK: define void @_Z7forward9copy_ctor(%struct.copy_ctor* noalias sret align 8 %{{.*}}, %struct.copy_ctor* nonnull %{{.*}})
 //
 // CHECK: declare %struct.copy_ctor* @_ZN9copy_ctorC1ERKS_(%struct.copy_ctor* returned, %struct.copy_ctor* dereferenceable(8))
 //
 // CHECK: define void @_Z14test_copy_ctorv()
 // CHECK: %[[tmp:.*]] = alloca %struct.copy_ctor, align 8
-// CHECK: call void @_Z13def_copy_ctorv(%struct.copy_ctor* nonnull sret %[[tmp]])
+// CHECK: call void @_Z13def_copy_ctorv(%struct.copy_ctor* nonnull sret align 8 %[[tmp]])
 // CHECK: call void @_Z3use9copy_ctor(%struct.copy_ctor* nonnull %[[tmp]])
 // CHECK: ret void
 //
 // CHECK: declare void @_Z3use9copy_ctor(%struct.copy_ctor*)
-// CHECK: declare void @_Z13def_copy_ctorv(%struct.copy_ctor* sret)
+// CHECK: declare void @_Z13def_copy_ctorv(%struct.copy_ctor* sret align 8)
 
 struct __attribute__((aligned(16))) aligned_copy_ctor {
   double d, e;
   aligned_copy_ctor(aligned_copy_ctor const &);
 };
 test(aligned_copy_ctor);
-// CHECK: define void @_Z7forward17aligned_copy_ctor(%struct.aligned_copy_ctor* noalias sret %{{.*}}, %struct.aligned_copy_ctor* nonnull %{{.*}})
+// CHECK: define void @_Z7forward17aligned_copy_ctor(%struct.aligned_copy_ctor* noalias sret align 16 %{{.*}}, %struct.aligned_copy_ctor* nonnull %{{.*}})
 //
 // CHECK: declare %struct.aligned_copy_ctor* @_ZN17aligned_copy_ctorC1ERKS_(%struct.aligned_copy_ctor* returned, %struct.aligned_copy_ctor* dereferenceable(16))
 //
 // CHECK: define void @_Z22test_aligned_copy_ctorv()
 // CHECK: %[[tmp:.*]] = alloca %struct.aligned_copy_ctor, align 16
-// CHECK: call void @_Z21def_aligned_copy_ctorv(%struct.aligned_copy_ctor* nonnull sret %[[tmp]])
+// CHECK: call void @_Z21def_aligned_copy_ctorv(%struct.aligned_copy_ctor* nonnull sret align 16 %[[tmp]])
 // CHECK: call void @_Z3use17aligned_copy_ctor(%struct.aligned_copy_ctor* nonnull %[[tmp]])
 // CHECK: ret void
 //
 // CHECK: declare void @_Z3use17aligned_copy_ctor(%struct.aligned_copy_ctor*)
-// CHECK: declare void @_Z21def_aligned_copy_ctorv(%struct.aligned_copy_ctor* sret)
+// CHECK: declare void @_Z21def_aligned_copy_ctorv(%struct.aligned_copy_ctor* sret align 16)
 
 struct empty {};
 test(empty);
Index: clang/test/CodeGenCXX/x86_32-arguments.cpp
===================================================================
--- clang/test/CodeGenCXX/x86_32-arguments.cpp
+++ clang/test/CodeGenCXX/x86_32-arguments.cpp
@@ -6,7 +6,7 @@
   short s;
 };
 
-// CHECK-LABEL: define void @_Z1fv(%struct.S* noalias sret %
+// CHECK-LABEL: define void @_Z1fv(%struct.S* noalias sret align 2 %
 S f() { return S(); }
 // CHECK-LABEL: define void @_Z1f1S(%struct.S* %0)
 void f(S) { }
@@ -18,7 +18,7 @@
   double c;
 };
 
-// CHECK-LABEL: define void @_Z1gv(%class.C* noalias sret %
+// CHECK-LABEL: define void @_Z1gv(%class.C* noalias sret align 4 %
 C g() { return C(); }
 
 // CHECK-LABEL: define void @_Z1f1C(%class.C* %0) 
@@ -103,13 +103,13 @@
 struct s7 : s7_0, s7_1 { };
 s7 f7() { return s7(); }
 
-// CHECK-LABEL: define void @_Z2f8v(%struct.s8* noalias sret %agg.result)
+// CHECK-LABEL: define void @_Z2f8v(%struct.s8* noalias sret align 4 %agg.result)
 struct s8_0 { };
 struct s8_1 { double x; };
 struct s8 { s8_0 a; s8_1 b; };
 s8 f8() { return s8(); }
 
-// CHECK-LABEL: define void @_Z2f9v(%struct.s9* noalias sret %agg.result)
+// CHECK-LABEL: define void @_Z2f9v(%struct.s9* noalias sret align 4 %agg.result)
 struct s9_0 { unsigned : 0; };
 struct s9_1 { double x; };
 struct s9 { s9_0 a; s9_1 b; };
Index: clang/test/CodeGenCXX/x86_64-arguments.cpp
===================================================================
--- clang/test/CodeGenCXX/x86_64-arguments.cpp
+++ clang/test/CodeGenCXX/x86_64-arguments.cpp
@@ -176,7 +176,7 @@
   // CHECK: define void @_ZN5test93fooEPNS_1SEPNS_1TE([[S:%.*]]* %0, [[T:%.*]]* %1)
   void foo(S*, T*) {}
 
-  // CHECK: define void @_ZN5test91aEiiiiNS_1TEPv([[S]]* noalias sret {{%.*}}, i32 %0, i32 %1, i32 %2, i32 %3, [[T]]* byval([[T]]) align 8 %4, i8* %5)
+  // CHECK: define void @_ZN5test91aEiiiiNS_1TEPv([[S]]* noalias sret align 8 {{%.*}}, i32 %0, i32 %1, i32 %2, i32 %3, [[T]]* byval([[T]]) align 8 %4, i8* %5)
   S a(int, int, int, int, T, void*) {
     return S();
   }
@@ -186,7 +186,7 @@
     return sret;
   }
 
-  // CHECK: define void @_ZN5test91cEiiiNS_1TEPv([[S]]* noalias sret {{%.*}}, i32 %0, i32 %1, i32 %2, i8* {{%.*}}, i8* {{%.*}}, i8* %3)
+  // CHECK: define void @_ZN5test91cEiiiNS_1TEPv([[S]]* noalias sret align 8 {{%.*}}, i32 %0, i32 %1, i32 %2, i8* {{%.*}}, i8* {{%.*}}, i8* %3)
   S c(int, int, int, T, void*) {
     return S();
   }
Index: clang/test/CodeGenCoroutines/coro-await.cpp
===================================================================
--- clang/test/CodeGenCoroutines/coro-await.cpp
+++ clang/test/CodeGenCoroutines/coro-await.cpp
@@ -130,7 +130,7 @@
   // CHECK: %[[PROMISE:.+]] = alloca %"struct.std::experimental::coroutine_traits<void, int>::promise_type"
   // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
   co_yield 42;
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJviEE12promise_type11yield_valueEi(%struct.suspend_maybe* sret %[[AWAITER:.+]], %"struct.std::experimental::coroutine_traits<void, int>::promise_type"* %[[PROMISE]], i32 42)
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJviEE12promise_type11yield_valueEi(%struct.suspend_maybe* sret align 4 %[[AWAITER:.+]], %"struct.std::experimental::coroutine_traits<void, int>::promise_type"* %[[PROMISE]], i32 42)
 
   // See if we need to suspend:
   // --------------------------
@@ -197,20 +197,20 @@
 extern "C" void TestAggr() {
   UseAggr(co_await AggrAwaiter{});
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret %[[AwaitResume:.+]],
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret align 4 %[[AwaitResume:.+]],
   // CHECK: call void @UseAggr(%struct.Aggr* dereferenceable(12) %[[AwaitResume]])
   // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* %[[AwaitResume]])
   // CHECK: call void @Whatever()
 
   co_await AggrAwaiter{};
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret %[[AwaitResume2:.+]],
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret align 4 %[[AwaitResume2:.+]],
   // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* %[[AwaitResume2]])
   // CHECK: call void @Whatever()
 
   Aggr Val = co_await AggrAwaiter{};
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret %[[AwaitResume3:.+]],
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret align 4 %[[AwaitResume3:.+]],
   // CHECK: call void @Whatever()
   // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* %[[AwaitResume3]])
 }
@@ -253,7 +253,7 @@
 
   co_await MyAgg{};
   // CHECK: call void @_ZN5MyAggawEv(%struct.MyAgg* %
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret %
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret align 4 %
 }
 
 // CHECK-LABEL: EndlessLoop(
Index: clang/test/CodeGenCoroutines/coro-gro-nrvo.cpp
===================================================================
--- clang/test/CodeGenCoroutines/coro-gro-nrvo.cpp
+++ clang/test/CodeGenCoroutines/coro-gro-nrvo.cpp
@@ -34,14 +34,14 @@
 };
 
 // Verify that the NRVO is applied to the Gro object.
-// CHECK-LABEL: define void @_Z1fi(%struct.coro* noalias sret %agg.result, i32 %0)
+// CHECK-LABEL: define void @_Z1fi(%struct.coro* noalias sret align 8 %agg.result, i32 %0)
 coro f(int) {
 // CHECK: %call = call noalias nonnull i8* @_Znwm(
 // CHECK-NEXT: br label %[[CoroInit:.*]]
 
 // CHECK: {{.*}}[[CoroInit]]:
 // CHECK: store i1 false, i1* %gro.active
-// CHECK: call void @{{.*get_return_objectEv}}(%struct.coro* sret %agg.result
+// CHECK: call void @{{.*get_return_objectEv}}(%struct.coro* sret align 8 %agg.result
 // CHECK-NEXT: store i1 true, i1* %gro.active
   co_return;
 }
@@ -65,7 +65,7 @@
 };
 
 // Verify that the NRVO is applied to the Gro object.
-// CHECK-LABEL: define void @_Z1hi(%struct.coro_two* noalias sret %agg.result, i32 %0)
+// CHECK-LABEL: define void @_Z1hi(%struct.coro_two* noalias sret align 8 %agg.result, i32 %0)
  coro_two h(int) {
 
 // CHECK: %call = call noalias i8* @_ZnwmRKSt9nothrow_t
@@ -73,12 +73,12 @@
 // CHECK-NEXT: br i1 %[[CheckNull]], label %[[InitOnSuccess:.*]], label %[[InitOnFailure:.*]]
 
 // CHECK: {{.*}}[[InitOnFailure]]:
-// CHECK-NEXT: call void @{{.*get_return_object_on_allocation_failureEv}}(%struct.coro_two* sret %agg.result
+// CHECK-NEXT: call void @{{.*get_return_object_on_allocation_failureEv}}(%struct.coro_two* sret align 8 %agg.result
 // CHECK-NEXT: br label %[[RetLabel:.*]]
 
 // CHECK: {{.*}}[[InitOnSuccess]]:
 // CHECK: store i1 false, i1* %gro.active
-// CHECK: call void @{{.*get_return_objectEv}}(%struct.coro_two* sret %agg.result
+// CHECK: call void @{{.*get_return_objectEv}}(%struct.coro_two* sret align 8 %agg.result
 // CHECK-NEXT: store i1 true, i1* %gro.active
 
 // CHECK: [[RetLabel]]:
Index: clang/test/CodeGenObjC/arc.m
===================================================================
--- clang/test/CodeGenObjC/arc.m
+++ clang/test/CodeGenObjC/arc.m
@@ -1541,10 +1541,10 @@
   //
   // CHECK: %[[T:[^ ]+]] = bitcast %struct.AggDtor* %[[TMP1:[^ ]+]] to i8*
   // CHECK: call void @llvm.lifetime.start.p0i8({{[^,]+}}, i8* %[[T]])
-  // CHECK: call void @getAggDtor(%struct.AggDtor* sret %[[TMP1]])
+  // CHECK: call void @getAggDtor(%struct.AggDtor* sret align 8 %[[TMP1]])
   // CHECK: %[[T:[^ ]+]] = bitcast %struct.AggDtor* %[[TMP2:[^ ]+]] to i8*
   // CHECK: call void @llvm.lifetime.start.p0i8({{[^,]+}}, i8* %[[T]])
-  // CHECK: call void @getAggDtor(%struct.AggDtor* sret %[[TMP2]])
+  // CHECK: call void @getAggDtor(%struct.AggDtor* sret align 8 %[[TMP2]])
   // CHECK: %[[T:[^ ]+]] = bitcast %struct.AggDtor* %[[TMP2]] to i8**
   // CHECK: call void @__destructor_8_s40(i8** %[[T]])
   // CHECK: %[[T:[^ ]+]] = bitcast %struct.AggDtor* %[[TMP2:[^ ]+]] to i8*
Index: clang/test/CodeGenObjC/direct-method.m
===================================================================
--- clang/test/CodeGenObjC/direct-method.m
+++ clang/test/CodeGenObjC/direct-method.m
@@ -120,7 +120,7 @@
 
 // CHECK-LABEL: define hidden void @"\01-[Root getAggregate]"(
 - (struct my_aggregate_struct)getAggregate __attribute__((objc_direct)) {
-  // CHECK: %struct.my_aggregate_struct* noalias sret [[RETVAL:%[^,]*]],
+  // CHECK: %struct.my_aggregate_struct* noalias sret align 4 [[RETVAL:%[^,]*]],
 
   // loading parameters
   // CHECK-LABEL: entry:
Index: clang/test/CodeGenObjC/nontrivial-c-struct-exception.m
===================================================================
--- clang/test/CodeGenObjC/nontrivial-c-struct-exception.m
+++ clang/test/CodeGenObjC/nontrivial-c-struct-exception.m
@@ -41,8 +41,8 @@
 // CHECK: define void @testWeakException()
 // CHECK: %[[AGG_TMP:.*]] = alloca %[[STRUCT_WEAK]], align 8
 // CHECK: %[[AGG_TMP1:.*]] = alloca %[[STRUCT_WEAK]], align 8
-// CHECK: call void @genWeak(%[[STRUCT_WEAK]]* sret %[[AGG_TMP]])
-// CHECK: invoke void @genWeak(%[[STRUCT_WEAK]]* sret %[[AGG_TMP1]])
+// CHECK: call void @genWeak(%[[STRUCT_WEAK]]* sret align 8 %[[AGG_TMP]])
+// CHECK: invoke void @genWeak(%[[STRUCT_WEAK]]* sret align 8 %[[AGG_TMP1]])
 
 // CHECK: call void @calleeWeak(%[[STRUCT_WEAK]]* %[[AGG_TMP]], %[[STRUCT_WEAK]]* %[[AGG_TMP1]])
 // CHECK: ret void
Index: clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
===================================================================
--- clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
+++ clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
@@ -37,7 +37,7 @@
 
 void func1(TrivialBig *);
 
-// CHECK: define void @testTrivialBig(%[[STRUCT_TRIVIALBIG]]* noalias sret %[[AGG_RESULT:.*]])
+// CHECK: define void @testTrivialBig(%[[STRUCT_TRIVIALBIG]]* noalias sret align 4 %[[AGG_RESULT:.*]])
 // CHECK: call void @func1(%[[STRUCT_TRIVIALBIG]]* %[[AGG_RESULT]])
 // CHECK-NEXT: ret void
 
@@ -69,7 +69,7 @@
   return a;
 }
 
-// CHECK: define void @testWeak(%[[STRUCT_WEAK]]* noalias sret %[[AGG_RESULT:.*]])
+// CHECK: define void @testWeak(%[[STRUCT_WEAK]]* noalias sret align 8 %[[AGG_RESULT:.*]])
 // CHECK: %[[NRVO:.*]] = alloca i1, align 1
 // CHECK: %[[V0:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
 // CHECK: call void @__default_constructor_8_w0(i8** %[[V0]])
@@ -105,7 +105,7 @@
     return b;
 }
 
-// CHECK: define internal void @"\01-[C1 foo1]"(%[[STRUCT_WEAK]]* noalias sret %[[AGG_RESULT:.*]], %{{.*}}* %{{.*}}, i8* %{{.*}})
+// CHECK: define internal void @"\01-[C1 foo1]"(%[[STRUCT_WEAK]]* noalias sret align 8 %[[AGG_RESULT:.*]], %{{.*}}* %{{.*}}, i8* %{{.*}})
 // CHECK: %[[NRVO:.*]] = alloca i1, align 1
 // CHECK: %[[V0:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
 // CHECK: call void @__default_constructor_8_w0(i8** %[[V0]])
Index: clang/test/CodeGenObjC/stret-1.m
===================================================================
--- clang/test/CodeGenObjC/stret-1.m
+++ clang/test/CodeGenObjC/stret-1.m
@@ -14,19 +14,19 @@
 {
     struct stret s;
     s = [(id)(argc&~255) method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret [[T0:%[^,]+]]
+    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret align 4 [[T0:%[^,]+]]
     // CHECK: [[T0P:%.*]] = bitcast %struct.stret* [[T0]] to i8*
     // CHECK: call void @llvm.memset.p0i8.i64(i8* align 4 [[T0P]], i8 0, i64 400, i1 false)
 
     s = [Test method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret [[T1:%[^,]+]]
+    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret align 4 [[T1:%[^,]+]]
     // CHECK-NOT: call void @llvm.memset.p0i8.i64(
 
     [(id)(argc&~255) method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret [[T1:%[^,]+]]
+    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret align 4 [[T1:%[^,]+]]
     // CHECK-NOT: call void @llvm.memset.p0i8.i64(
 
     [Test method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret [[T1:%[^,]+]]
+    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret align 4 [[T1:%[^,]+]]
     // CHECK-NOT: call void @llvm.memset.p0i8.i64(
 }
Index: clang/test/CodeGenObjC/weak-in-c-struct.m
===================================================================
--- clang/test/CodeGenObjC/weak-in-c-struct.m
+++ clang/test/CodeGenObjC/weak-in-c-struct.m
@@ -179,7 +179,7 @@
   calleeWeak(*a);
 }
 
-// COMMON: define void @test_return_Weak(%[[STRUCT_WEAK]]* noalias sret %[[AGG_RESULT:.*]], %[[STRUCT_WEAK]]* %[[A:.*]])
+// COMMON: define void @test_return_Weak(%[[STRUCT_WEAK]]* noalias sret align {{.*}} %[[AGG_RESULT:.*]], %[[STRUCT_WEAK]]* %[[A:.*]])
 // COMMON: %[[A_ADDR:.*]] = alloca %[[STRUCT_WEAK]]*
 // COMMON: store %[[STRUCT_WEAK]]* %[[A]], %[[STRUCT_WEAK]]** %[[A_ADDR]]
 // COMMON: %[[V0:.*]] = load %[[STRUCT_WEAK]]*, %[[STRUCT_WEAK]]** %[[A_ADDR]]
Index: clang/test/CodeGenObjCXX/objc-struct-cxx-abi.mm
===================================================================
--- clang/test/CodeGenObjCXX/objc-struct-cxx-abi.mm
+++ clang/test/CodeGenObjCXX/objc-struct-cxx-abi.mm
@@ -90,7 +90,7 @@
   testParamStrongWeak(*a);
 }
 
-// CHECK: define void @_Z20testReturnStrongWeakP10StrongWeak(%[[STRUCT_STRONGWEAK:.*]]* noalias sret %[[AGG_RESULT:.*]], %[[STRUCT_STRONGWEAK]]* %[[A:.*]])
+// CHECK: define void @_Z20testReturnStrongWeakP10StrongWeak(%[[STRUCT_STRONGWEAK:.*]]* noalias sret align 8 %[[AGG_RESULT:.*]], %[[STRUCT_STRONGWEAK]]* %[[A:.*]])
 // CHECK: %[[A_ADDR:.*]] = alloca %[[STRUCT_STRONGWEAK]]*, align 8
 // CHECK: store %[[STRUCT_STRONGWEAK]]* %[[A]], %[[STRUCT_STRONGWEAK]]** %[[A_ADDR]], align 8
 // CHECK: %[[V0:.*]] = load %[[STRUCT_STRONGWEAK]]*, %[[STRUCT_STRONGWEAK]]** %[[A_ADDR]], align 8
Index: clang/test/CodeGenOpenCL/addr-space-struct-arg.cl
===================================================================
--- clang/test/CodeGenOpenCL/addr-space-struct-arg.cl
+++ clang/test/CodeGenOpenCL/addr-space-struct-arg.cl
@@ -43,7 +43,7 @@
 struct LargeStructOneMember g_s;
 #endif
 
-// X86-LABEL: define void @foo(%struct.Mat4X4* noalias sret %agg.result, %struct.Mat3X3* byval(%struct.Mat3X3) align 4 %in)
+// X86-LABEL: define void @foo(%struct.Mat4X4* noalias sret align 4 %agg.result, %struct.Mat3X3* byval(%struct.Mat3X3) align 4 %in)
 // AMDGCN-LABEL: define %struct.Mat4X4 @foo([9 x i32] %in.coerce)
 Mat4X4 __attribute__((noinline)) foo(Mat3X3 in) {
   Mat4X4 out;
@@ -63,8 +63,8 @@
   out[0] = foo(in[1]);
 }
 
-// X86-LABEL: define void @foo_large(%struct.Mat64X64* noalias sret %agg.result, %struct.Mat32X32* byval(%struct.Mat32X32) align 4 %in)
-// AMDGCN-LABEL: define void @foo_large(%struct.Mat64X64 addrspace(5)* noalias sret %agg.result, %struct.Mat32X32 addrspace(5)* byval(%struct.Mat32X32) align 4 %in)
+// X86-LABEL: define void @foo_large(%struct.Mat64X64* noalias sret align 4 %agg.result, %struct.Mat32X32* byval(%struct.Mat32X32) align 4 %in)
+// AMDGCN-LABEL: define void @foo_large(%struct.Mat64X64 addrspace(5)* noalias sret align 4 %agg.result, %struct.Mat32X32 addrspace(5)* byval(%struct.Mat32X32) align 4 %in)
 Mat64X64 __attribute__((noinline)) foo_large(Mat32X32 in) {
   Mat64X64 out;
   return out;
Index: clang/test/CodeGenOpenCL/amdgpu-abi-struct-coerce.cl
===================================================================
--- clang/test/CodeGenOpenCL/amdgpu-abi-struct-coerce.cl
+++ clang/test/CodeGenOpenCL/amdgpu-abi-struct-coerce.cl
@@ -404,14 +404,14 @@
   return s;
 }
 
-// CHECK: define void @func_ret_struct_arr32(%struct.struct_arr32 addrspace(5)* noalias nocapture sret %agg.result)
+// CHECK: define void @func_ret_struct_arr32(%struct.struct_arr32 addrspace(5)* noalias nocapture sret align 4 %agg.result)
 struct_arr32 func_ret_struct_arr32()
 {
   struct_arr32 s = { 0 };
   return s;
 }
 
-// CHECK: define void @func_ret_struct_arr33(%struct.struct_arr33 addrspace(5)* noalias nocapture sret %agg.result)
+// CHECK: define void @func_ret_struct_arr33(%struct.struct_arr33 addrspace(5)* noalias nocapture sret align 4 %agg.result)
 struct_arr33 func_ret_struct_arr33()
 {
   struct_arr33 s = { 0 };
@@ -440,7 +440,7 @@
   return s;
 }
 
-// CHECK: define void @func_flexible_array_ret(%struct.flexible_array addrspace(5)* noalias nocapture sret %agg.result)
+// CHECK: define void @func_flexible_array_ret(%struct.flexible_array addrspace(5)* noalias nocapture sret align 4 %agg.result)
 flexible_array func_flexible_array_ret()
 {
   flexible_array s = { 0 };
Index: clang/test/CodeGenOpenCLCXX/addrspace-of-this.cl
===================================================================
--- clang/test/CodeGenOpenCLCXX/addrspace-of-this.cl
+++ clang/test/CodeGenOpenCLCXX/addrspace-of-this.cl
@@ -114,7 +114,7 @@
 // Test the address space of 'this' when invoking the operator+
 // COMMON: [[C1GEN:%[.a-z0-9]+]] = addrspacecast %class.C* %c1 to %class.C addrspace(4)*
 // COMMON: [[C2GEN:%[.a-z0-9]+]] = addrspacecast %class.C* %c2 to %class.C addrspace(4)*
-// COMMON: call spir_func void @_ZNU3AS41CplERU3AS4KS_(%class.C* sret %c3, %class.C addrspace(4)* [[C1GEN]], %class.C addrspace(4)* dereferenceable(4) [[C2GEN]])
+// COMMON: call spir_func void @_ZNU3AS41CplERU3AS4KS_(%class.C* sret align 4 %c3, %class.C addrspace(4)* [[C1GEN]], %class.C addrspace(4)* dereferenceable(4) [[C2GEN]])
 
 // Test the address space of 'this' when invoking the move constructor
 // COMMON: [[C4GEN:%[.a-z0-9]+]] = addrspacecast %class.C* %c4 to %class.C addrspace(4)*
@@ -134,7 +134,7 @@
 
 // Tests address space of inline members
 //COMMON: @_ZNU3AS41C3getEv(%class.C addrspace(4)* %this)
-//COMMON: @_ZNU3AS41CplERU3AS4KS_(%class.C* noalias sret %agg.result, %class.C addrspace(4)* %this
+//COMMON: @_ZNU3AS41CplERU3AS4KS_(%class.C* noalias sret align 4 %agg.result, %class.C addrspace(4)* %this
 #define TEST(AS)             \
   __kernel void test##AS() { \
     AS C c;                  \
Index: clang/test/Modules/templates.mm
===================================================================
--- clang/test/Modules/templates.mm
+++ clang/test/Modules/templates.mm
@@ -125,7 +125,7 @@
 
 // Check that returnNonTrivial doesn't return Class0<S0> directly in registers.
 
-// CHECK: declare void @_Z16returnNonTrivialv(%struct.Class0* sret)
+// CHECK: declare void @_Z16returnNonTrivialv(%struct.Class0* sret align 8)
 
 @import template_nontrivial0;
 @import template_nontrivial1;