[clang][CodeGen] Allow memcpy replace with trivial auto var init

clang/lib/CodeGen/CGDecl.cpp

@@ -1242,27 +1242,38 @@ static void emitStoresForConstant(CodeGenModule &CGM, const VarDecl &D,
     return;
   }
 
-  // If the initializer is small, use a handful of stores.
+  // If the initializer is small or trivialAutoVarInit is set, use a handful of
+  // stores.
+  bool IsTrivialAutoVarInitPattern =
+      CGM.getContext().getLangOpts().getTrivialAutoVarInit() ==
+      LangOptions::TrivialAutoVarInitKind::Pattern;
   if (shouldSplitConstantStore(CGM, ConstantSize)) {
     if (auto *STy = dyn_cast<llvm::StructType>(Ty)) {
-      const llvm::StructLayout *Layout =
-          CGM.getDataLayout().getStructLayout(STy);
-      for (unsigned i = 0; i != constant->getNumOperands(); i++) {
-        CharUnits CurOff = CharUnits::fromQuantity(Layout->getElementOffset(i));
-        Address EltPtr = Builder.CreateConstInBoundsByteGEP(
-            Loc.withElementType(CGM.Int8Ty), CurOff);
-        emitStoresForConstant(CGM, D, EltPtr, isVolatile, Builder,
-                              constant->getAggregateElement(i), IsAutoInit);
+      if (STy == Loc.getElementType() ||
+          (STy != Loc.getElementType() && IsTrivialAutoVarInitPattern)) {
+        const llvm::StructLayout *Layout =
+            CGM.getDataLayout().getStructLayout(STy);
+        for (unsigned i = 0; i != constant->getNumOperands(); i++) {
+          CharUnits CurOff =
+              CharUnits::fromQuantity(Layout->getElementOffset(i));
+          Address EltPtr = Builder.CreateConstInBoundsByteGEP(
+              Loc.withElementType(CGM.Int8Ty), CurOff);
+          emitStoresForConstant(CGM, D, EltPtr, isVolatile, Builder,
+                                constant->getAggregateElement(i), IsAutoInit);
+        }
+        return;
       }
-      return;
     } else if (auto *ATy = dyn_cast<llvm::ArrayType>(Ty)) {
-      for (unsigned i = 0; i != ATy->getNumElements(); i++) {
-        Address EltPtr = Builder.CreateConstGEP(
-            Loc.withElementType(ATy->getElementType()), i);
-        emitStoresForConstant(CGM, D, EltPtr, isVolatile, Builder,
-                              constant->getAggregateElement(i), IsAutoInit);
+      if (ATy == Loc.getElementType() ||
+          (ATy != Loc.getElementType() && IsTrivialAutoVarInitPattern)) {
+        for (unsigned i = 0; i != ATy->getNumElements(); i++) {
+          Address EltPtr = Builder.CreateConstGEP(
+              Loc.withElementType(ATy->getElementType()), i);
+          emitStoresForConstant(CGM, D, EltPtr, isVolatile, Builder,
+                                constant->getAggregateElement(i), IsAutoInit);
+        }
+        return;
       }
-      return;
     }
   }
 

clang/test/CodeGen/aapcs-align.cpp

@@ -134,8 +134,8 @@ void g6() {
   f6m(1, 2, 3, 4, 5, s);
 }
 // CHECK: define{{.*}} void @g6
-// CHECK: call void @f6(i32 noundef 1, [4 x i32] [i32 6, i32 7, i32 0, i32 0])
-// CHECK: call void @f6m(i32 noundef 1, i32 noundef 2, i32 noundef 3, i32 noundef 4, i32 noundef 5, [4 x i32] [i32 6, i32 7, i32 0, i32 0])
+// CHECK: call void @f6(i32 noundef 1, [4 x i32] [i32 6, i32 7, i32 0, i32 undef])
+// CHECK: call void @f6m(i32 noundef 1, i32 noundef 2, i32 noundef 3, i32 noundef 4, i32 noundef 5, [4 x i32] [i32 6, i32 7, i32 0, i32 undef])
 // CHECK: declare void @f6(i32 noundef, [4 x i32])
 // CHECK: declare void @f6m(i32 noundef, i32 noundef, i32 noundef, i32 noundef, i32 noundef, [4 x i32])
 }

clang/test/CodeGen/aapcs64-align.cpp

@@ -75,8 +75,8 @@ void g4() {
   f4m(1, 2, 3, 4, 5, s);
 }
 // CHECK: define{{.*}} void @g4()
-// CHECK: call void @f4(i32 noundef 1, [2 x i64] %{{.*}})
-// CHECK: void @f4m(i32 noundef 1, i32 noundef 2, i32 noundef 3, i32 noundef 4, i32 noundef 5, [2 x i64] %{{.*}})
+// CHECK: call void @f4(i32 noundef 1, [2 x i64] [i64 30064771078, i64 0])
+// CHECK: void @f4m(i32 noundef 1, i32 noundef 2, i32 noundef 3, i32 noundef 4, i32 noundef 5, [2 x i64] [i64 30064771078, i64 0])
 // CHECK: declare void @f4(i32 noundef, [2 x i64])
 // CHECK: declare void @f4m(i32 noundef, i32 noundef, i32 noundef, i32 noundef, i32 noundef, [2 x i64])
 
@@ -95,8 +95,8 @@ void f5m(int, int, int, int, int, P16);
     f5m(1, 2, 3, 4, 5, s);
 }
 // CHECK: define{{.*}} void @g5()
-// CHECK: call void @f5(i32 noundef 1, [2 x i64] %{{.*}})
-// CHECK: void @f5m(i32 noundef 1, i32 noundef 2, i32 noundef 3, i32 noundef 4, i32 noundef 5, [2 x i64] %{{.*}})
+// CHECK: call void @f5(i32 noundef 1, [2 x i64] [i64 30064771078, i64 0])
+// CHECK: void @f5m(i32 noundef 1, i32 noundef 2, i32 noundef 3, i32 noundef 4, i32 noundef 5, [2 x i64] [i64 30064771078, i64 0])
 // CHECK: declare void @f5(i32 noundef, [2 x i64])
 // CHECK: declare void @f5m(i32 noundef, i32 noundef, i32 noundef, i32 noundef, i32 noundef, [2 x i64])
 

clang/test/CodeGen/attr-counted-by.c

@@ -1314,17 +1314,10 @@ int test14(int idx) {
 // NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i32 @test15(
 // NO-SANITIZE-WITH-ATTR-SAME: i32 noundef [[IDX:%.*]]) local_unnamed_addr #[[ATTR4]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
-// NO-SANITIZE-WITH-ATTR-NEXT:    [[FOO:%.*]] = alloca [[STRUCT_ANON_8:%.*]], align 4
-// NO-SANITIZE-WITH-ATTR-NEXT:    call void @llvm.lifetime.start.p0(i64 8, ptr nonnull [[FOO]]) #[[ATTR12]]
-// NO-SANITIZE-WITH-ATTR-NEXT:    store i32 1, ptr [[FOO]], align 4
-// NO-SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 4
-// NO-SANITIZE-WITH-ATTR-NEXT:    store i32 2, ptr [[TMP0]], align 4
-// NO-SANITIZE-WITH-ATTR-NEXT:    [[BLAH:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 8
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[IDXPROM:%.*]] = sext i32 [[IDX]] to i64
-// NO-SANITIZE-WITH-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [0 x i32], ptr [[BLAH]], i64 0, i64 [[IDXPROM]]
-// NO-SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA2]]
-// NO-SANITIZE-WITH-ATTR-NEXT:    call void @llvm.lifetime.end.p0(i64 8, ptr nonnull [[FOO]]) #[[ATTR12]]
-// NO-SANITIZE-WITH-ATTR-NEXT:    ret i32 [[TMP1]]
+// NO-SANITIZE-WITH-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [0 x i32], ptr getelementptr inbounds ([[STRUCT_ANON_8:%.*]], ptr @__const.test15.foo, i64 1, i32 0), i64 0, i64 [[IDXPROM]]
+// NO-SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA2]]
+// NO-SANITIZE-WITH-ATTR-NEXT:    ret i32 [[TMP0]]
 //
 // SANITIZE-WITHOUT-ATTR-LABEL: define dso_local i32 @test15(
 // SANITIZE-WITHOUT-ATTR-SAME: i32 noundef [[IDX:%.*]]) local_unnamed_addr #[[ATTR0]] {
@@ -1342,17 +1335,10 @@ int test14(int idx) {
 // NO-SANITIZE-WITHOUT-ATTR-LABEL: define dso_local i32 @test15(
 // NO-SANITIZE-WITHOUT-ATTR-SAME: i32 noundef [[IDX:%.*]]) local_unnamed_addr #[[ATTR1]] {
 // NO-SANITIZE-WITHOUT-ATTR-NEXT:  entry:
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[FOO:%.*]] = alloca [[STRUCT_ANON_8:%.*]], align 4
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    call void @llvm.lifetime.start.p0(i64 8, ptr nonnull [[FOO]]) #[[ATTR9]]
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    store i32 1, ptr [[FOO]], align 4
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 4
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    store i32 2, ptr [[TMP0]], align 4
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[BLAH:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 8
 // NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[IDXPROM:%.*]] = sext i32 [[IDX]] to i64
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [0 x i32], ptr [[BLAH]], i64 0, i64 [[IDXPROM]]
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA2]]
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    call void @llvm.lifetime.end.p0(i64 8, ptr nonnull [[FOO]]) #[[ATTR9]]
-// NO-SANITIZE-WITHOUT-ATTR-NEXT:    ret i32 [[TMP1]]
+// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [0 x i32], ptr getelementptr inbounds ([[STRUCT_ANON_8:%.*]], ptr @__const.test15.foo, i64 1, i32 0), i64 0, i64 [[IDXPROM]]
+// NO-SANITIZE-WITHOUT-ATTR-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA2]]
+// NO-SANITIZE-WITHOUT-ATTR-NEXT:    ret i32 [[TMP0]]
 //
 int test15(int idx) {
   struct {

clang/test/CodeGenCXX/auto-var-init.cpp

@@ -1,8 +1,8 @@
 // RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks %s -emit-llvm -o - | FileCheck %s -check-prefixes=CHECK,CHECK-O0
 // RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks -ftrivial-auto-var-init=pattern %s -emit-llvm -o - | FileCheck %s -check-prefixes=CHECK-O0,PATTERN,PATTERN-O0
-// RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks -ftrivial-auto-var-init=pattern %s -O1 -emit-llvm -o - | FileCheck %s -check-prefixes=CHECK-O1,PATTERN,PATTERN-O1
+// RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks -ftrivial-auto-var-init=pattern %s -O1 -emit-llvm -o - | FileCheck %s -check-prefixes=PATTERN,PATTERN-O1
 // RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks -ftrivial-auto-var-init=zero %s -emit-llvm -o - | FileCheck %s -check-prefixes=CHECK-O0,ZERO,ZERO-O0
-// RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks -ftrivial-auto-var-init=zero %s -O1 -emit-llvm -o - | FileCheck %s -check-prefixes=CHECK-O1,ZERO,ZERO-O1
+// RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown -fblocks -ftrivial-auto-var-init=zero %s -O1 -emit-llvm -o - | FileCheck %s -check-prefixes=ZERO,ZERO-O1
 // RUN: %clang_cc1 -std=c++14 -triple i386-unknown-unknown -fblocks -ftrivial-auto-var-init=pattern %s -emit-llvm -o - | FileCheck %s -check-prefixes=CHECK-O0,PATTERN,PATTERN-O0
 
 #pragma clang diagnostic ignored "-Winaccessible-base"
@@ -1303,9 +1303,10 @@ TEST_CUSTOM(semivolatile, semivolatile, { 0x44444444, 0x44444444 });
 // CHECK-O0:  call void @llvm.memcpy
 // CHECK-NOT:   !annotation
 // CHECK-O0:  call void @{{.*}}used{{.*}}%custom)
-// CHECK-O1:  store i32 1145324612, ptr %custom, align 4
-// CHECK-O1-NEXT:  %[[I:[^ ]*]] = getelementptr inbounds i8, ptr %custom, i64 4
-// CHECK-O1-NEXT:  store i32 1145324612, ptr %[[I]], align 4
+// PATTERN-O1:       store i32 1145324612, ptr %custom, align 4
+// PATTERN-O1-NEXT:  %[[I:[^ ]*]] = getelementptr inbounds i8, ptr %custom, i64 4
+// PATTERN-O1-NEXT:  store i32 1145324612, ptr %[[I]], align 4
+// ZERO-O1:          store i64 4919131752989213764, ptr %custom, align 8
 // CHECK-NOT:   !annotation
 
 TEST_UNINIT(semivolatileinit, semivolatileinit);
@@ -1418,7 +1419,8 @@ TEST_CUSTOM(matching, matching, { .f = 0xf00f });
 // CHECK-O0:  call void @llvm.memcpy
 // CHECK-NOT:   !annotation
 // CHECK-O0:  call void @{{.*}}used{{.*}}%custom)
-// CHECK-O1:  store float 6.145500e+04, ptr {{.*}}, align 4
+// PATTERN-O1:  store float 6.145500e+04, ptr {{.*}}, align 4
+// ZERO-O1:     store i32 1198526208, ptr %custom, align 4
 // CHECK-NOT:   !annotation
 
 TEST_UNINIT(matchingreverse, matchingreverse);
@@ -1445,7 +1447,8 @@ TEST_CUSTOM(matchingreverse, matchingreverse, { .i = 0xf00f });
 // CHECK-O0:    call void @llvm.memcpy
 // CHECK-NOT:   !annotation
 // CHECK-O0:    call void @{{.*}}used{{.*}}%custom)
-// CHECK-O1:    store i32 61455, ptr %custom, align 4
+// PATTERN-O1:  store i32 61455, ptr %custom, align 4
+// ZERO-O1:     store i32 61455, ptr %custom, align 4
 // CHECK-NOT:   !annotation
 
 TEST_UNINIT(unmatched, unmatched);
@@ -1471,7 +1474,8 @@ TEST_CUSTOM(unmatched, unmatched, { .i = 0x3badbeef });
 // CHECK-O0:    call void @llvm.memcpy
 // CHECK-NOT:   !annotation
 // CHECK-O0:    call void @{{.*}}used{{.*}}%custom)
-// CHECK-O1:    store i32 1001242351, ptr {{.*}}, align 4
+// PATTERN-O1:  store i32 1001242351, ptr {{.*}}, align 4
+// ZERO-O1:     store i32 1001242351, ptr {{.*}}, align 4
 // CHECK-NOT:   !annotation
 
 TEST_UNINIT(unmatchedreverse, unmatchedreverse);
@@ -1504,9 +1508,7 @@ TEST_CUSTOM(unmatchedreverse, unmatchedreverse, { .c = 42  });
 // PATTERN-O1-NEXT:  store i8 -86, ptr %[[I]], align {{.*}}
 // PATTERN-O1-NEXT:  %[[I:[^ ]*]] = getelementptr inbounds i8, ptr %custom, i64 3
 // PATTERN-O1-NEXT:  store i8 -86, ptr %[[I]], align {{.*}}
-// ZERO-O1:     store i8 42, ptr {{.*}}, align 4
-// ZERO-O1-NEXT:  %[[I:[^ ]*]] = getelementptr inbounds i8, ptr %custom, i64 1
-// ZERO-O1-NEXT:  call void @llvm.memset.{{.*}}({{.*}}, i8 0, i64 3, {{.*}})
+// ZERO-O1:     store i32 42, ptr {{.*}}, align 4
 
 TEST_UNINIT(unmatchedfp, unmatchedfp);
 // CHECK-LABEL: @test_unmatchedfp_uninit()
@@ -1531,7 +1533,8 @@ TEST_CUSTOM(unmatchedfp, unmatchedfp, { .d = 3.1415926535897932384626433 });
 // CHECK-O0:    call void @llvm.memcpy
 // CHECK-NOT:   !annotation
 // CHECK-O0:    call void @{{.*}}used{{.*}}%custom)
-// CHECK-O1:    store double 0x400921FB54442D18, ptr %custom, align 8
+// PATTERN-O1:  store double 0x400921FB54442D18, ptr %custom, align 8
+// ZERO-O1:     store i64 4614256656552045848, ptr %custom, align 8
 // CHECK-NOT:   !annotation
 
 TEST_UNINIT(emptyenum, emptyenum);

clang/test/CodeGenOpenCL/amdgpu-printf.cl

@@ -30,14 +30,7 @@ __kernel void test_printf_int(int i) {
 // CHECK-NEXT:    [[S:%.*]] = alloca [4 x i8], align 1, addrspace(5)
 // CHECK-NEXT:    store i32 [[I:%.*]], ptr addrspace(5) [[I_ADDR]], align 4, !tbaa [[TBAA8]]
 // CHECK-NEXT:    call void @llvm.lifetime.start.p5(i64 4, ptr addrspace(5) [[S]]) #[[ATTR5:[0-9]+]]
-// CHECK-NEXT:    [[LOC0:%.*]] = getelementptr i8, ptr addrspace(5) [[S]], i64 0
-// CHECK-NEXT:    store i8 102, ptr addrspace(5) [[LOC0]], align 1
-// CHECK-NEXT:    [[LOC1:%.*]] = getelementptr i8, ptr addrspace(5) [[S]], i64 1
-// CHECK-NEXT:    store i8 111, ptr addrspace(5) [[LOC1]], align 1
-// CHECK-NEXT:    [[LOC2:%.*]] = getelementptr i8, ptr addrspace(5) [[S]], i64 2
-// CHECK-NEXT:    store i8 111, ptr addrspace(5) [[LOC2]], align 1
-// CHECK-NEXT:    [[LOC3:%.*]] = getelementptr i8, ptr addrspace(5) [[S]], i64 3
-// CHECK-NEXT:    store i8 0, ptr addrspace(5) [[LOC3]], align 1
+// CHECK-NEXT:    call void @llvm.memcpy.p5.p4.i64(ptr addrspace(5) align 1 [[S]], ptr addrspace(4) align 1 @__const.test_printf_str_int.s, i64 4, i1 false)
 // CHECK-NEXT:    [[ARRAYDECAY:%.*]] = getelementptr inbounds [4 x i8], ptr addrspace(5) [[S]], i64 0, i64 0
 // CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr addrspace(5) [[I_ADDR]], align 4, !tbaa [[TBAA8]]
 // CHECK-NEXT:    [[CALL:%.*]] = call i32 (ptr addrspace(4), ...) @printf(ptr addrspace(4) noundef @.str.2, ptr addrspace(5) noundef [[ARRAYDECAY]], i32 noundef [[TMP2]]) #[[ATTR4]]

clang/test/OpenMP/bug54082.c

@@ -69,9 +69,7 @@ void foo() {
 // CHECK-NEXT:    [[X_TRAITS:%.*]] = alloca [1 x %struct.omp_alloctrait_t], align 16
 // CHECK-NEXT:    [[X_ALLOC:%.*]] = alloca i64, align 8
 // CHECK-NEXT:    call void @llvm.lifetime.start.p0(i64 16, ptr nonnull [[X_TRAITS]]) #[[ATTR5:[0-9]+]]
-// CHECK-NEXT:    store i32 2, ptr [[X_TRAITS]], align 16
-// CHECK-NEXT:    [[LOC0:%.*]] = getelementptr inbounds i8, ptr [[X_TRAITS]], i64 8
-// CHECK-NEXT:    store i64 64, ptr [[LOC0]], align 8
+// CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr noundef nonnull align 16 dereferenceable(16) [[X_TRAITS]], ptr noundef nonnull align 16 dereferenceable(16) @__const.foo.x_traits, i64 16, i1 false)
 // CHECK-NEXT:    call void @llvm.lifetime.start.p0(i64 8, ptr nonnull [[X_ALLOC]]) #[[ATTR5]]
 // CHECK-NEXT:    [[CALL:%.*]] = call i64 @omp_init_allocator(i64 noundef 0, i32 noundef 1, ptr noundef nonnull [[X_TRAITS]]) #[[ATTR5]]
 // CHECK-NEXT:    store i64 [[CALL]], ptr [[X_ALLOC]], align 8, !tbaa [[TBAA3:![0-9]+]]