Change memcpy/memset/memmove to have dest and source alignments.

This is a follow on from a similar LLVM commit: r253511.

Note, this was reviewed (and more details are in) http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html

These intrinsics currently have an explicit alignment argument which is
required to be a constant integer.  It represents the alignment of the
source and dest, and so must be the minimum of those.

This change allows source and dest to each have their own alignments
by using the alignment attribute on their arguments.  The alignment
argument itself is removed.

The only code change to clang is hidden in CGBuilder.h which now passes
both dest and source alignment to IRBuilder, instead of taking the minimum of
dest and source alignments.

Reviewed by Hal Finkel.

llvm-svn: 253512

15 files changed, 61 insertions, 61 deletions

diff --git a/clang/test/OpenMP/atomic_write_codegen.c b/clang/test/OpenMP/atomic_write_codegen.c
index 66172af07e8..e100ccbe3d1 100644
--- a/clang/test/OpenMP/atomic_write_codegen.c
+++ b/clang/test/OpenMP/atomic_write_codegen.c
@@ -134,7 +134,7 @@ int main() {
   dx = dv;
 // CHECK: [[LD:%.+]] = load x86_fp80, x86_fp80*
 // CHECK: [[BITCAST:%.+]] = bitcast x86_fp80* [[LDTEMP:%.*]] to i8*
-// CHECK: call void @llvm.memset.p0i8.i64(i8* [[BITCAST]], i8 0, i64 16, i32 16, i1 false)
+// CHECK: call void @llvm.memset.p0i8.i64(i8* align 16 [[BITCAST]], i8 0, i64 16, i1 false)
 // CHECK: store x86_fp80 [[LD]], x86_fp80* [[LDTEMP]]
 // CHECK: [[BITCAST:%.+]] = bitcast x86_fp80* [[LDTEMP:%.*]] to i128*
 // CHECK: [[LD:%.+]] = load i128, i128* [[BITCAST]]
@@ -223,7 +223,7 @@ int main() {
 // CHECK: load i64, i64*
 // CHECK: [[VAL:%.+]] = uitofp i64 %{{.+}} to x86_fp80
 // CHECK: [[BITCAST:%.+]] = bitcast x86_fp80* [[TEMP:%.+]] to i8*
-// CHECK: call void @llvm.memset.p0i8.i64(i8* [[BITCAST]], i8 0, i64 16, i32 16, i1 false)
+// CHECK: call void @llvm.memset.p0i8.i64(i8* align 16 [[BITCAST]], i8 0, i64 16, i1 false)
 // CHECK: store x86_fp80 [[VAL]], x86_fp80* [[TEMP]]
 // CHECK: [[BITCAST:%.+]] = bitcast x86_fp80* [[TEMP]] to i128*
 // CHECK: [[VAL:%.+]] = load i128, i128* [[BITCAST]]
diff --git a/clang/test/OpenMP/for_firstprivate_codegen.cpp b/clang/test/OpenMP/for_firstprivate_codegen.cpp
index 01a93559451..f4039e27a8c 100644
--- a/clang/test/OpenMP/for_firstprivate_codegen.cpp
+++ b/clang/test/OpenMP/for_firstprivate_codegen.cpp
@@ -222,7 +222,7 @@ int main() {
 
 // firstprivate vec(vec)
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
 
 // firstprivate s_arr(s_arr)
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
@@ -291,7 +291,7 @@ int main() {
 // firstprivate vec(vec)
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 [[VEC_SRC]],
 
 // firstprivate s_arr(s_arr)
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
diff --git a/clang/test/OpenMP/for_lastprivate_codegen.cpp b/clang/test/OpenMP/for_lastprivate_codegen.cpp
index f268df61e04..e9907a604e5 100644
--- a/clang/test/OpenMP/for_lastprivate_codegen.cpp
+++ b/clang/test/OpenMP/for_lastprivate_codegen.cpp
@@ -276,7 +276,7 @@ int main() {
 // original vec[]=private_vec[];
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 [[VEC_SRC]],
 
 // original s_arr[]=private_s_arr[];
 // CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
@@ -472,7 +472,7 @@ int main() {
 // original vec[]=private_vec[];
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 128 [[VEC_DEST]], i8* align 128 [[VEC_SRC]],
 
 // original s_arr[]=private_s_arr[];
 // CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
diff --git a/clang/test/OpenMP/for_reduction_codegen.cpp b/clang/test/OpenMP/for_reduction_codegen.cpp
index 59d785480ea..7d895bd6b99 100644
--- a/clang/test/OpenMP/for_reduction_codegen.cpp
+++ b/clang/test/OpenMP/for_reduction_codegen.cpp
@@ -277,7 +277,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @{{.+}}([[S_FLOAT_TY]]* [[VAR_REF]], [[S_FLOAT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1 = var1.operator &&(var1_reduction);
 // CHECK: [[TO_FLOAT:%.+]] = call float @{{.+}}([[S_FLOAT_TY]]* [[VAR1_REF]])
@@ -293,7 +293,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_FLOAT_TY]]* [[COND_LVALUE:%.+]], float [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1 = min(t_var1, t_var1_reduction);
 // CHECK: [[T_VAR1_VAL:%.+]] = load float, float* [[T_VAR1_REF]],
@@ -331,7 +331,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @{{.+}}([[S_FLOAT_TY]]* [[VAR_REF]], [[S_FLOAT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // var1 = var1.operator &&(var1_reduction);
@@ -349,7 +349,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_FLOAT_TY]]* [[COND_LVALUE:%.+]], float [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // t_var1 = min(t_var1, t_var1_reduction);
@@ -436,7 +436,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @{{.+}}([[S_FLOAT_TY]]* [[VAR_LHS]], [[S_FLOAT_TY]]* dereferenceable(4) [[VAR_RHS]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1_lhs = var1_lhs.operator &&(var1_rhs);
 // CHECK: [[TO_FLOAT:%.+]] = call float @{{.+}}([[S_FLOAT_TY]]* [[VAR1_LHS]])
@@ -452,7 +452,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_FLOAT_TY]]* [[COND_LVALUE:%.+]], float [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR1_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1_lhs = min(t_var1_lhs, t_var1_rhs);
 // CHECK: [[T_VAR1_LHS_VAL:%.+]] = load float, float* [[T_VAR1_LHS]],
@@ -558,7 +558,7 @@ int main() {
 // CHECK: phi [[S_FLOAT_TY]]*
 // CHECK: [[AND:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @_ZN1SIfEanERKS0_([[S_FLOAT_TY]]* %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: [[BITCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[AND]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* [[BITCAST]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %{{.+}}, i8* align 4 [[BITCAST]], i64 4, i1 false)
 // CHECK: [[DONE:%.+]] = icmp eq [[S_FLOAT_TY]]* %{{.+}}, [[END]]
 // CHECK: br i1 [[DONE]],
 
@@ -588,7 +588,7 @@ int main() {
 // CHECK: call void @__kmpc_critical(
 // CHECK: [[AND:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @_ZN1SIfEanERKS0_([[S_FLOAT_TY]]* %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: [[BITCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[AND]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* [[BITCAST]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %{{.+}}, i8* align 4 [[BITCAST]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 // CHECK: [[DONE:%.+]] = icmp eq [[S_FLOAT_TY]]* %{{.+}}, [[END]]
 // CHECK: br i1 [[DONE]],
@@ -664,7 +664,7 @@ int main() {
 // CHECK: phi [[S_FLOAT_TY]]*
 // CHECK: [[AND:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @_ZN1SIfEanERKS0_([[S_FLOAT_TY]]* %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: [[BITCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[AND]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* [[BITCAST]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %{{.+}}, i8* align 4 [[BITCAST]], i64 4, i1 false)
 // CHECK: [[DONE:%.+]] = icmp eq [[S_FLOAT_TY]]* %{{.+}}, [[END]]
 // CHECK: br i1 [[DONE]],
 
@@ -753,7 +753,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_REF]], [[S_INT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1 = var1.operator &&(var1_reduction);
 // CHECK: [[TO_INT:%.+]] = call i{{[0-9]+}} @{{.+}}([[S_INT_TY]]* [[VAR1_REF]])
@@ -769,7 +769,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1 = min(t_var1, t_var1_reduction);
 // CHECK: [[T_VAR1_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR1_REF]],
@@ -795,7 +795,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_REF]], [[S_INT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // var1 = var1.operator &&(var1_reduction);
@@ -813,7 +813,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // t_var1 = min(t_var1, t_var1_reduction);
@@ -881,7 +881,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_LHS]], [[S_INT_TY]]* dereferenceable(4) [[VAR_RHS]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1_lhs = var1_lhs.operator &&(var1_rhs);
 // CHECK: [[TO_INT:%.+]] = call i{{[0-9]+}} @{{.+}}([[S_INT_TY]]* [[VAR1_LHS]])
@@ -897,7 +897,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1_lhs = min(t_var1_lhs, t_var1_rhs);
 // CHECK: [[T_VAR1_LHS_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR1_LHS]],
diff --git a/clang/test/OpenMP/parallel_codegen.cpp b/clang/test/OpenMP/parallel_codegen.cpp
index 0a09bada432..d8a2cc9e4cb 100644
--- a/clang/test/OpenMP/parallel_codegen.cpp
+++ b/clang/test/OpenMP/parallel_codegen.cpp
@@ -41,7 +41,7 @@ int main (int argc, char **argv) {
 // CHECK-DEBUG:       [[LOC_2_ADDR:%.+]] = alloca %ident_t
 // CHECK-DEBUG:       [[KMPC_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[LOC_2_ADDR]] to i8*
 // CHECK-DEBUG-NEXT:  [[KMPC_DEFAULT_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[DEF_LOC_2]] to i8*
-// CHECK-DEBUG-NEXT:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[KMPC_LOC_VOIDPTR]], i8* [[KMPC_DEFAULT_LOC_VOIDPTR]], i64 24, i32 8, i1 false)
+// CHECK-DEBUG-NEXT:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[KMPC_LOC_VOIDPTR]], i8* align 8 [[KMPC_DEFAULT_LOC_VOIDPTR]], i64 24, i1 false)
 // CHECK-DEBUG:       store i32 %argc, i32* [[ARGC_ADDR:%.+]],
 // CHECK-DEBUG:       [[KMPC_LOC_PSOURCE_REF:%.+]] = getelementptr inbounds %ident_t, %ident_t* [[LOC_2_ADDR]], i32 0, i32 4
 // CHECK-DEBUG-NEXT:  store i8* getelementptr inbounds ([{{.+}} x i8], [{{.+}} x i8]* [[LOC1]], i32 0, i32 0), i8** [[KMPC_LOC_PSOURCE_REF]]
@@ -88,7 +88,7 @@ int main (int argc, char **argv) {
 // CHECK-DEBUG-DAG:   [[LOC_2_ADDR:%.+]] = alloca %ident_t
 // CHECK-DEBUG:       [[KMPC_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[LOC_2_ADDR]] to i8*
 // CHECK-DEBUG-NEXT:  [[KMPC_DEFAULT_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[DEF_LOC_2]] to i8*
-// CHECK-DEBUG-NEXT:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[KMPC_LOC_VOIDPTR]], i8* [[KMPC_DEFAULT_LOC_VOIDPTR]], i64 24, i32 8, i1 false)
+// CHECK-DEBUG-NEXT:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[KMPC_LOC_VOIDPTR]], i8* align 8 [[KMPC_DEFAULT_LOC_VOIDPTR]], i64 24, i1 false)
 // CHECK-DEBUG-NEXT:  store i8** %argc, i8*** [[ARGC_ADDR:%.+]],
 // CHECK-DEBUG:  [[KMPC_LOC_PSOURCE_REF:%.+]] = getelementptr inbounds %ident_t, %ident_t* [[LOC_2_ADDR]], i32 0, i32 4
 // CHECK-DEBUG-NEXT:  store i8* getelementptr inbounds ([{{.+}} x i8], [{{.+}} x i8]* [[LOC2]], i32 0, i32 0), i8** [[KMPC_LOC_PSOURCE_REF]]
diff --git a/clang/test/OpenMP/parallel_copyin_codegen.cpp b/clang/test/OpenMP/parallel_copyin_codegen.cpp
index 9a7449ded0f..3a99a679546 100644
--- a/clang/test/OpenMP/parallel_copyin_codegen.cpp
+++ b/clang/test/OpenMP/parallel_copyin_codegen.cpp
@@ -277,10 +277,10 @@ int main() {
 
 // threadprivate_vec = vec;
 // CHECK: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[VEC]]
-// CHECK: call void @llvm.memcpy{{.*}}(i8* %{{.+}}, i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
+// CHECK: call void @llvm.memcpy{{.*}}(i8* align 4 %{{.+}}, i8* align 4 bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
 
 // TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x i32]* [[MASTER_REF2]] to i8*
-// TLS-CHECK: call void @llvm.memcpy{{.*}}(i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*), i8* [[MASTER_CAST]]
+// TLS-CHECK: call void @llvm.memcpy{{.*}}(i8* align 4 bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*), i8* align 4 [[MASTER_CAST]]
 
 // threadprivate_s_arr = s_arr;
 // CHECK: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[S_ARR]]
@@ -398,10 +398,10 @@ int main() {
 
 // threadprivate_vec = vec;
 // CHECK: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[TMAIN_VEC]]
-// CHECK: call {{.*}}void @llvm.memcpy{{.*}}(i8* %{{.+}}, i8* bitcast ([2 x i{{[0-9]+}}]* [[TMAIN_VEC]] to i8*),
+// CHECK: call {{.*}}void @llvm.memcpy{{.*}}(i8* align 128 %{{.+}}, i8* align 128 bitcast ([2 x i{{[0-9]+}}]* [[TMAIN_VEC]] to i8*),
 
 // TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x i32]* [[MASTER_REF1]] to i8*
-// TLS-CHECK: call void @llvm.memcpy{{.*}}(i8* bitcast ([2 x i{{[0-9]+}}]* [[TMAIN_VEC]] to i8*), i8* [[MASTER_CAST]]
+// TLS-CHECK: call void @llvm.memcpy{{.*}}(i8* align 128 bitcast ([2 x i{{[0-9]+}}]* [[TMAIN_VEC]] to i8*), i8* align 128 [[MASTER_CAST]]
 
 // threadprivate_s_arr = s_arr;
 // CHECK: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[TMAIN_S_ARR]]
@@ -491,13 +491,13 @@ void array_func() {
   static St s[2];
 // ARRAY: @__kmpc_fork_call(
 // ARRAY: call i8* @__kmpc_threadprivate_cached(
-// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* bitcast ([2 x i32]* @{{.+}} to i8*), i64 8, i32 4, i1 false)
+// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 %{{.+}}, i8* align 4 bitcast ([2 x i32]* @{{.+}} to i8*), i64 8, i1 false)
 // ARRAY: call dereferenceable(8) %struct.St* @{{.+}}(%struct.St* %{{.+}}, %struct.St* dereferenceable(8) %{{.+}})
 
 // TLS-ARRAY: @__kmpc_fork_call(
 // TLS-ARRAY: [[REFT:%.+]] = load [2 x i32]*, [2 x i32]** [[ADDR:%.+]],
 // TLS-ARRAY: [[REF:%.+]] = bitcast [2 x i32]* [[REFT]] to i8*
-// TLS-ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* bitcast ([2 x i32]* @{{.+}} to i8*), i8* [[REF]], i64 8, i32 4, i1 false)
+// TLS-ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 bitcast ([2 x i32]* @{{.+}} to i8*), i8* align 4 [[REF]], i64 8, i1 false)
 // TLS-ARRAY: call dereferenceable(8) %struct.St* @{{.+}}(%struct.St* %{{.+}}, %struct.St* dereferenceable(8) %{{.+}})
 
 #pragma omp threadprivate(a, s)
diff --git a/clang/test/OpenMP/parallel_firstprivate_codegen.cpp b/clang/test/OpenMP/parallel_firstprivate_codegen.cpp
index e44c64b7c95..c329bc12229 100644
--- a/clang/test/OpenMP/parallel_firstprivate_codegen.cpp
+++ b/clang/test/OpenMP/parallel_firstprivate_codegen.cpp
@@ -185,7 +185,7 @@ int main() {
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 [[VEC_SRC]],
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_BEGIN:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[S_ARR_REF]] to [[S_FLOAT_TY]]*
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
@@ -233,7 +233,7 @@ int main() {
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]], align 128
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]], i{{[0-9]+}} {{[0-9]+}}, i{{[0-9]+}} 128,
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 128 [[VEC_DEST]], i8* align 128 [[VEC_SRC]], i{{[0-9]+}} {{[0-9]+}},
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_BEGIN:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[S_ARR_REF]] to [[S_INT_TY]]*
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
@@ -283,7 +283,7 @@ void array_func(float a[3], St s[2], int n, long double vla1[n]) {
 // ARRAY-DAG: store double* %{{.+}}, double** [[PRIV_VLA2]],
 // ARRAY: call i8* @llvm.stacksave()
 // ARRAY: [[SIZE:%.+]] = mul nuw i64 %{{.+}}, 8
-// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* %{{.+}}, i64 [[SIZE]], i32 128, i1 false)
+// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 %{{.+}}, i8* align 128 %{{.+}}, i64 [[SIZE]], i1 false)
 #pragma omp parallel firstprivate(a, s, vla1, vla2)
   s[0].St_func(s, n, vla1);
   ;
@@ -299,7 +299,7 @@ void array_func(float a[3], St s[2], int n, long double vla1[n]) {
 // ARRAY-DAG: store double* %{{.+}}, double** [[PRIV_VLA2]],
 // ARRAY: call i8* @llvm.stacksave()
 // ARRAY: [[SIZE:%.+]] = mul nuw i64 %{{.+}}, 8
-// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* %{{.+}}, i64 [[SIZE]], i32 128, i1 false)
+// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 %{{.+}}, i8* align 128 %{{.+}}, i64 [[SIZE]], i1 false)
 #endif
 
 
diff --git a/clang/test/OpenMP/parallel_reduction_codegen.cpp b/clang/test/OpenMP/parallel_reduction_codegen.cpp
index 0fea14fcd79..70418caa478 100644
--- a/clang/test/OpenMP/parallel_reduction_codegen.cpp
+++ b/clang/test/OpenMP/parallel_reduction_codegen.cpp
@@ -245,7 +245,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @{{.+}}([[S_FLOAT_TY]]* [[VAR_REF]], [[S_FLOAT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1 = var1.operator &&(var1_reduction);
 // CHECK: [[TO_FLOAT:%.+]] = call float @{{.+}}([[S_FLOAT_TY]]* [[VAR1_REF]])
@@ -261,7 +261,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_FLOAT_TY]]* [[COND_LVALUE:%.+]], float [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1 = min(t_var1, t_var1_reduction);
 // CHECK: [[T_VAR1_VAL:%.+]] = load float, float* [[T_VAR1_REF]],
@@ -299,7 +299,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @{{.+}}([[S_FLOAT_TY]]* [[VAR_REF]], [[S_FLOAT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // var1 = var1.operator &&(var1_reduction);
@@ -317,7 +317,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_FLOAT_TY]]* [[COND_LVALUE:%.+]], float [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // t_var1 = min(t_var1, t_var1_reduction);
@@ -400,7 +400,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_FLOAT_TY]]* @{{.+}}([[S_FLOAT_TY]]* [[VAR_LHS]], [[S_FLOAT_TY]]* dereferenceable(4) [[VAR_RHS]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1_lhs = var1_lhs.operator &&(var1_rhs);
 // CHECK: [[TO_FLOAT:%.+]] = call float @{{.+}}([[S_FLOAT_TY]]* [[VAR1_LHS]])
@@ -416,7 +416,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_FLOAT_TY]]* [[COND_LVALUE:%.+]], float [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR1_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_FLOAT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1_lhs = min(t_var1_lhs, t_var1_rhs);
 // CHECK: [[T_VAR1_LHS_VAL:%.+]] = load float, float* [[T_VAR1_LHS]],
@@ -503,7 +503,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_REF]], [[S_INT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1 = var1.operator &&(var1_reduction);
 // CHECK: [[TO_INT:%.+]] = call i{{[0-9]+}} @{{.+}}([[S_INT_TY]]* [[VAR1_REF]])
@@ -518,7 +518,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1 = min(t_var1, t_var1_reduction);
 // CHECK: [[T_VAR1_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR1_REF]],
@@ -544,7 +544,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_REF]], [[S_INT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // var1 = var1.operator &&(var1_reduction);
@@ -562,7 +562,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // t_var1 = min(t_var1, t_var1_reduction);
@@ -631,7 +631,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_LHS]], [[S_INT_TY]]* dereferenceable(4) [[VAR_RHS]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1_lhs = var1_lhs.operator &&(var1_rhs);
 // CHECK: [[TO_INT:%.+]] = call i{{[0-9]+}} @{{.+}}([[S_INT_TY]]* [[VAR1_LHS]])
@@ -647,7 +647,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 128 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1_lhs = min(t_var1_lhs, t_var1_rhs);
 // CHECK: [[T_VAR1_LHS_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR1_LHS]],
diff --git a/clang/test/OpenMP/sections_firstprivate_codegen.cpp b/clang/test/OpenMP/sections_firstprivate_codegen.cpp
index f673597f660..effe59b78f4 100644
--- a/clang/test/OpenMP/sections_firstprivate_codegen.cpp
+++ b/clang/test/OpenMP/sections_firstprivate_codegen.cpp
@@ -215,7 +215,7 @@ int main() {
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
 // firstprivate vec(vec)
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
 
 // firstprivate s_arr(s_arr)
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
@@ -281,7 +281,7 @@ int main() {
 // firstprivate vec(vec)
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 [[VEC_SRC]],
 
 // firstprivate s_arr(s_arr)
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
diff --git a/clang/test/OpenMP/sections_lastprivate_codegen.cpp b/clang/test/OpenMP/sections_lastprivate_codegen.cpp
index b1d242aad93..dbb49b0f1a1 100644
--- a/clang/test/OpenMP/sections_lastprivate_codegen.cpp
+++ b/clang/test/OpenMP/sections_lastprivate_codegen.cpp
@@ -338,7 +338,7 @@ int main() {
 // original vec[]=private_vec[];
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 [[VEC_SRC]],
 
 // original s_arr[]=private_s_arr[];
 // CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
diff --git a/clang/test/OpenMP/sections_reduction_codegen.cpp b/clang/test/OpenMP/sections_reduction_codegen.cpp
index 2904d97c82c..8402e82e300 100644
--- a/clang/test/OpenMP/sections_reduction_codegen.cpp
+++ b/clang/test/OpenMP/sections_reduction_codegen.cpp
@@ -300,7 +300,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_REF]], [[S_INT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1 = var1.operator &&(var1_reduction);
 // CHECK: [[TO_INT:%.+]] = call i{{[0-9]+}} @{{.+}}([[S_INT_TY]]* [[VAR1_REF]])
@@ -316,7 +316,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1 = min(t_var1, t_var1_reduction);
 // CHECK: [[T_VAR1_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR1_REF]],
@@ -342,7 +342,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_REF]], [[S_INT_TY]]* dereferenceable(4) [[VAR_PRIV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // var1 = var1.operator &&(var1_reduction);
@@ -360,7 +360,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_REF]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 // CHECK: call void @__kmpc_end_critical(
 
 // t_var1 = min(t_var1, t_var1_reduction);
@@ -428,7 +428,7 @@ int main() {
 // CHECK: [[UP:%.+]] = call dereferenceable(4) [[S_INT_TY]]* @{{.+}}([[S_INT_TY]]* [[VAR_LHS]], [[S_INT_TY]]* dereferenceable(4) [[VAR_RHS]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[UP]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // var1_lhs = var1_lhs.operator &&(var1_rhs);
 // CHECK: [[TO_INT:%.+]] = call i{{[0-9]+}} @{{.+}}([[S_INT_TY]]* [[VAR1_LHS]])
@@ -444,7 +444,7 @@ int main() {
 // CHECK:  call void @{{.+}}([[S_INT_TY]]* [[COND_LVALUE:%.+]], i32 [[CONV]])
 // CHECK: [[BC1:%.+]] = bitcast [[S_INT_TY]]* [[VAR1_LHS]] to i8*
 // CHECK: [[BC2:%.+]] = bitcast [[S_INT_TY]]* [[COND_LVALUE]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[BC1]], i8* [[BC2]], i64 4, i32 4, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[BC1]], i8* align 4 [[BC2]], i64 4, i1 false)
 
 // t_var1_lhs = min(t_var1_lhs, t_var1_rhs);
 // CHECK: [[T_VAR1_LHS_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR1_LHS]],
diff --git a/clang/test/OpenMP/single_codegen.cpp b/clang/test/OpenMP/single_codegen.cpp
index 61a93a5a1b7..decaebd0978 100644
--- a/clang/test/OpenMP/single_codegen.cpp
+++ b/clang/test/OpenMP/single_codegen.cpp
@@ -137,7 +137,7 @@ int main() {
 // CHECK: [[DST_A2_ADDR:%.+]] = load i8*, i8** [[DST_A2_ADDR_REF]],
 // CHECK: [[SRC_A2_ADDR_REF:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[SRC_ADDR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3
 // CHECK: [[SRC_A2_ADDR:%.+]] = load i8*, i8** [[SRC_A2_ADDR_REF]],
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[DST_A2_ADDR]], i8* [[SRC_A2_ADDR]], i{{[0-9]+}} 2, i{{[0-9]+}} 1, i1 false)
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 1 [[DST_A2_ADDR]], i8* align 1 [[SRC_A2_ADDR]], i{{[0-9]+}} 2, i1 false)
 // CHECK: [[DST_TC2_ADDR_REF:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[DST_ADDR]], i{{[0-9]+}} 0, i{{[0-9]+}} 4
 // CHECK: [[DST_TC2_ADDR_VOID_PTR:%.+]] = load i8*, i8** [[DST_TC2_ADDR_REF]],
 // CHECK: [[DST_TC2_ADDR:%.+]] = bitcast i8* [[DST_TC2_ADDR_VOID_PTR]] to [[TEST_CLASS_TY]]*
diff --git a/clang/test/OpenMP/single_firstprivate_codegen.cpp b/clang/test/OpenMP/single_firstprivate_codegen.cpp
index 316595b74bc..dc77239fc93 100644
--- a/clang/test/OpenMP/single_firstprivate_codegen.cpp
+++ b/clang/test/OpenMP/single_firstprivate_codegen.cpp
@@ -187,7 +187,7 @@ int main() {
 
 // firstprivate vec(vec)
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
 
 // firstprivate s_arr(s_arr)
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
@@ -251,7 +251,7 @@ int main() {
 // firstprivate vec(vec)
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
-// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
+// CHECK: call void @llvm.memcpy.{{.+}}(i8* align 4 [[VEC_DEST]], i8* align 4 [[VEC_SRC]],
 
 // firstprivate s_arr(s_arr)
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
diff --git a/clang/test/OpenMP/task_codegen.cpp b/clang/test/OpenMP/task_codegen.cpp
index ed59b3d3814..077ee25ba6e 100644
--- a/clang/test/OpenMP/task_codegen.cpp
+++ b/clang/test/OpenMP/task_codegen.cpp
@@ -34,7 +34,7 @@ int main() {
 // CHECK: [[SHAREDS_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR:%.+]], i32 0, i32 0
 // CHECK: [[SHAREDS_REF:%.+]] = load i8*, i8** [[SHAREDS_REF_PTR]]
 // CHECK: [[BITCAST:%.+]] = bitcast [[STRUCT_SHAREDS]]* [[CAPTURES]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[SHAREDS_REF]], i8* [[BITCAST]], i64 16, i32 8, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[SHAREDS_REF]], i8* align 8 [[BITCAST]], i64 16, i1 false)
 // CHECK: [[DESTRUCTORS_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR]], i32 0, i32 3
 // CHECK: store i32 (i32, i8*)* null, i32 (i32, i8*)** [[DESTRUCTORS_REF_PTR]]
 // CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
@@ -50,7 +50,7 @@ int main() {
 // CHECK: [[SHAREDS_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR:%.+]], i32 0, i32 0
 // CHECK: [[SHAREDS_REF:%.+]] = load i8*, i8** [[SHAREDS_REF_PTR]]
 // CHECK: [[BITCAST:%.+]] = bitcast [[STRUCT_SHAREDS1]]* [[CAPTURES]] to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[SHAREDS_REF]], i8* [[BITCAST]], i64 8, i32 8, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[SHAREDS_REF]], i8* align 8 [[BITCAST]], i64 8, i1 false)
 // CHECK: [[DESTRUCTORS_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR]], i32 0, i32 3
 // CHECK: store i32 (i32, i8*)* null, i32 (i32, i8*)** [[DESTRUCTORS_REF_PTR]]
 // CHECK: [[DEP:%.*]] = getelementptr inbounds [4 x [[KMP_DEPEND_INFO]]], [4 x [[KMP_DEPEND_INFO]]]* [[DEPENDENCIES:%.*]], i64 0, i64 0
diff --git a/clang/test/OpenMP/task_firstprivate_codegen.cpp b/clang/test/OpenMP/task_firstprivate_codegen.cpp
index e2244140d1e..29b0946d49c 100644
--- a/clang/test/OpenMP/task_firstprivate_codegen.cpp
+++ b/clang/test/OpenMP/task_firstprivate_codegen.cpp
@@ -201,7 +201,7 @@ int main() {
 // CHECK: [[SHAREDS_REF_ADDR:%.+]] = getelementptr inbounds [[KMP_TASK_T_TY]], [[KMP_TASK_T_TY]]* [[TASK]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[SHAREDS_REF:%.+]] = load i8*, i8** [[SHAREDS_REF_ADDR]],
 // CHECK: [[CAPTURES_ADDR:%.+]] = bitcast [[CAP_MAIN_TY]]* %{{.+}} to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[SHAREDS_REF]], i8* [[CAPTURES_ADDR]], i64 40, i32 8, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[SHAREDS_REF]], i8* align 8 [[CAPTURES_ADDR]], i64 40, i1 false)
 
 // Initialize kmp_task_t->privates with default values (no init for simple types, default constructors for classes).
 // Also copy address of private copy to the corresponding shareds reference.
@@ -355,7 +355,7 @@ int main() {
 // CHECK: [[SHAREDS_REF_ADDR:%.+]] = getelementptr inbounds [[KMP_TASK_T_TY]], [[KMP_TASK_T_TY]]* [[TASK]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[SHAREDS_REF:%.+]] = load i8*, i8** [[SHAREDS_REF_ADDR]],
 // CHECK: [[CAPTURES_ADDR:%.+]] = bitcast [[CAP_TMAIN_TY]]* %{{.+}} to i8*
-// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[SHAREDS_REF]], i8* [[CAPTURES_ADDR]], i64 32, i32 8, i1 false)
+// CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[SHAREDS_REF]], i8* align 8 [[CAPTURES_ADDR]], i64 32, i1 false)
 
 // Initialize kmp_task_t->privates with default values (no init for simple types, default constructors for classes).
 // CHECK: [[PRIVATES:%.+]] = getelementptr inbounds [[KMP_TASK_TMAIN_TY]], [[KMP_TASK_TMAIN_TY]]* [[RES_KMP_TASK]], i{{[0-9]+}} 0, i{{[0-9]+}} 2