2013-12-14 12:49:06 +08:00
|
|
|
// RUN: %clang_cc1 -triple i686-pc-linux-gnu %s -o - -emit-llvm -verify | FileCheck %s
|
|
|
|
|
// expected-no-diagnostics
|
2009-11-16 13:16:40 +08:00
|
|
|
|
2011-01-27 17:37:56 +08:00
|
|
|
typedef __typeof(sizeof(int)) size_t;
|
2009-11-16 13:16:40 +08:00
|
|
|
|
2011-01-27 17:37:56 +08:00
|
|
|
namespace test1 {
|
|
|
|
|
struct A { void operator delete(void*,size_t); int x; };
|
|
|
|
|
|
2020-12-31 16:27:11 +08:00
|
|
|
// CHECK-LABEL: define{{.*}} void @_ZN5test11aEPNS_1AE(
|
2011-01-27 17:37:56 +08:00
|
|
|
void a(A *x) {
|
|
|
|
|
// CHECK: load
|
|
|
|
|
// CHECK-NEXT: icmp eq {{.*}}, null
|
|
|
|
|
// CHECK-NEXT: br i1
|
|
|
|
|
// CHECK: call void @_ZN5test11AdlEPvj(i8* %{{.*}}, i32 4)
|
|
|
|
|
delete x;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// Check that we make cookies for the two-arg delete even when using
|
|
|
|
|
// the global allocator and deallocator.
|
|
|
|
|
namespace test2 {
|
|
|
|
|
struct A {
|
|
|
|
|
int x;
|
|
|
|
|
void *operator new[](size_t);
|
|
|
|
|
void operator delete[](void *, size_t);
|
|
|
|
|
};
|
|
|
|
|
|
2020-12-31 16:27:11 +08:00
|
|
|
// CHECK: define{{.*}} [[A:%.*]]* @_ZN5test24testEv()
|
2011-01-27 17:37:56 +08:00
|
|
|
A *test() {
|
[clang] Annotating C++'s `operator new` with more attributes
Summary:
Right now we annotate C++'s `operator new` with `noalias` attribute,
which very much is healthy for optimizations.
However as per [[ http://eel.is/c++draft/basic.stc.dynamic.allocation | `[basic.stc.dynamic.allocation]` ]],
there are more promises on global `operator new`, namely:
* non-`std::nothrow_t` `operator new` *never* returns `nullptr`
* If `std::align_val_t align` parameter is taken, the pointer will also be `align`-aligned
* ~~global `operator new`-returned pointer is `__STDCPP_DEFAULT_NEW_ALIGNMENT__`-aligned ~~ It's more caveated than that.
Supplying this information may not cause immediate landslide effects
on any specific benchmarks, but it for sure will be healthy for optimizer
in the sense that the IR will better reflect the guarantees provided in the source code.
The caveat is `-fno-assume-sane-operator-new`, which currently prevents emitting `noalias`
attribute, and is automatically passed by Sanitizers ([[ https://bugs.llvm.org/show_bug.cgi?id=16386 | PR16386 ]]) - should it also cover these attributes?
The problem is that the flag is back-end-specific, as seen in `test/Modules/explicit-build-flags.cpp`.
But while it is okay to add `noalias` metadata in backend, we really should be adding at least
the alignment metadata to the AST, since that allows us to perform sema checks on it.
Reviewers: erichkeane, rjmccall, jdoerfert, eugenis, rsmith
Reviewed By: rsmith
Subscribers: xbolva00, jrtc27, atanasyan, nlopes, cfe-commits
Tags: #llvm, #clang
Differential Revision: https://reviews.llvm.org/D73380
2020-02-26 06:37:17 +08:00
|
|
|
// CHECK: [[NEW:%.*]] = call noalias nonnull i8* @_Znaj(i32 44)
|
2011-01-27 17:37:56 +08:00
|
|
|
// CHECK-NEXT: [[T0:%.*]] = bitcast i8* [[NEW]] to i32*
|
|
|
|
|
// CHECK-NEXT: store i32 10, i32* [[T0]]
|
Compute and preserve alignment more faithfully in IR-generation.
Introduce an Address type to bundle a pointer value with an
alignment. Introduce APIs on CGBuilderTy to work with Address
values. Change core APIs on CGF/CGM to traffic in Address where
appropriate. Require alignments to be non-zero. Update a ton
of code to compute and propagate alignment information.
As part of this, I've promoted CGBuiltin's EmitPointerWithAlignment
helper function to CGF and made use of it in a number of places in
the expression emitter.
The end result is that we should now be significantly more correct
when performing operations on objects that are locally known to
be under-aligned. Since alignment is not reliably tracked in the
type system, there are inherent limits to this, but at least we
are no longer confused by standard operations like derived-to-base
conversions and array-to-pointer decay. I've also fixed a large
number of bugs where we were applying the complete-object alignment
to a pointer instead of the non-virtual alignment, although most of
these were hidden by the very conservative approach we took with
member alignment.
Also, because IRGen now reliably asserts on zero alignments, we
should no longer be subject to an absurd but frustrating recurring
bug where an incomplete type would report a zero alignment and then
we'd naively do a alignmentAtOffset on it and emit code using an
alignment equal to the largest power-of-two factor of the offset.
We should also now be emitting much more aggressive alignment
attributes in the presence of over-alignment. In particular,
field access now uses alignmentAtOffset instead of min.
Several times in this patch, I had to change the existing
code-generation pattern in order to more effectively use
the Address APIs. For the most part, this seems to be a strict
improvement, like doing pointer arithmetic with GEPs instead of
ptrtoint. That said, I've tried very hard to not change semantics,
but it is likely that I've failed in a few places, for which I
apologize.
ABIArgInfo now always carries the assumed alignment of indirect and
indirect byval arguments. In order to cut down on what was already
a dauntingly large patch, I changed the code to never set align
attributes in the IR on non-byval indirect arguments. That is,
we still generate code which assumes that indirect arguments have
the given alignment, but we don't express this information to the
backend except where it's semantically required (i.e. on byvals).
This is likely a minor regression for those targets that did provide
this information, but it'll be trivial to add it back in a later
patch.
I partially punted on applying this work to CGBuiltin. Please
do not add more uses of the CreateDefaultAligned{Load,Store}
APIs; they will be going away eventually.
llvm-svn: 246985
2015-09-08 16:05:57 +08:00
|
|
|
// CHECK-NEXT: [[T1:%.*]] = getelementptr inbounds i8, i8* [[NEW]], i32 4
|
2011-01-27 17:37:56 +08:00
|
|
|
// CHECK-NEXT: [[T2:%.*]] = bitcast i8* [[T1]] to [[A]]*
|
|
|
|
|
// CHECK-NEXT: ret [[A]]* [[T2]]
|
|
|
|
|
return ::new A[10];
|
|
|
|
|
}
|
|
|
|
|
|
2020-12-31 16:27:11 +08:00
|
|
|
// CHECK-LABEL: define{{.*}} void @_ZN5test24testEPNS_1AE(
|
2011-01-27 17:37:56 +08:00
|
|
|
void test(A *p) {
|
|
|
|
|
// CHECK: [[P:%.*]] = alloca [[A]]*, align 4
|
|
|
|
|
// CHECK-NEXT: store [[A]]* {{%.*}}, [[A]]** [[P]], align 4
|
2015-02-28 05:19:58 +08:00
|
|
|
// CHECK-NEXT: [[T0:%.*]] = load [[A]]*, [[A]]** [[P]], align 4
|
2011-01-27 17:37:56 +08:00
|
|
|
// CHECK-NEXT: [[T1:%.*]] = icmp eq [[A]]* [[T0]], null
|
|
|
|
|
// CHECK-NEXT: br i1 [[T1]],
|
|
|
|
|
// CHECK: [[T2:%.*]] = bitcast [[A]]* [[T0]] to i8*
|
Compute and preserve alignment more faithfully in IR-generation.
Introduce an Address type to bundle a pointer value with an
alignment. Introduce APIs on CGBuilderTy to work with Address
values. Change core APIs on CGF/CGM to traffic in Address where
appropriate. Require alignments to be non-zero. Update a ton
of code to compute and propagate alignment information.
As part of this, I've promoted CGBuiltin's EmitPointerWithAlignment
helper function to CGF and made use of it in a number of places in
the expression emitter.
The end result is that we should now be significantly more correct
when performing operations on objects that are locally known to
be under-aligned. Since alignment is not reliably tracked in the
type system, there are inherent limits to this, but at least we
are no longer confused by standard operations like derived-to-base
conversions and array-to-pointer decay. I've also fixed a large
number of bugs where we were applying the complete-object alignment
to a pointer instead of the non-virtual alignment, although most of
these were hidden by the very conservative approach we took with
member alignment.
Also, because IRGen now reliably asserts on zero alignments, we
should no longer be subject to an absurd but frustrating recurring
bug where an incomplete type would report a zero alignment and then
we'd naively do a alignmentAtOffset on it and emit code using an
alignment equal to the largest power-of-two factor of the offset.
We should also now be emitting much more aggressive alignment
attributes in the presence of over-alignment. In particular,
field access now uses alignmentAtOffset instead of min.
Several times in this patch, I had to change the existing
code-generation pattern in order to more effectively use
the Address APIs. For the most part, this seems to be a strict
improvement, like doing pointer arithmetic with GEPs instead of
ptrtoint. That said, I've tried very hard to not change semantics,
but it is likely that I've failed in a few places, for which I
apologize.
ABIArgInfo now always carries the assumed alignment of indirect and
indirect byval arguments. In order to cut down on what was already
a dauntingly large patch, I changed the code to never set align
attributes in the IR on non-byval indirect arguments. That is,
we still generate code which assumes that indirect arguments have
the given alignment, but we don't express this information to the
backend except where it's semantically required (i.e. on byvals).
This is likely a minor regression for those targets that did provide
this information, but it'll be trivial to add it back in a later
patch.
I partially punted on applying this work to CGBuiltin. Please
do not add more uses of the CreateDefaultAligned{Load,Store}
APIs; they will be going away eventually.
llvm-svn: 246985
2015-09-08 16:05:57 +08:00
|
|
|
// CHECK-NEXT: [[T3:%.*]] = getelementptr inbounds i8, i8* [[T2]], i32 -4
|
2011-01-27 17:37:56 +08:00
|
|
|
// CHECK-NEXT: [[T4:%.*]] = bitcast i8* [[T3]] to i32*
|
2015-02-28 05:19:58 +08:00
|
|
|
// CHECK-NEXT: [[T5:%.*]] = load i32, i32* [[T4]]
|
2011-01-27 17:37:56 +08:00
|
|
|
// CHECK-NEXT: call void @_ZdaPv(i8* [[T3]])
|
|
|
|
|
// CHECK-NEXT: br label
|
|
|
|
|
::delete[] p;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// rdar://problem/8913519
|
|
|
|
|
namespace test3 {
|
|
|
|
|
struct A {
|
|
|
|
|
int x;
|
|
|
|
|
void operator delete[](void *, size_t);
|
[clang] Annotating C++'s `operator new` with more attributes
Summary:
Right now we annotate C++'s `operator new` with `noalias` attribute,
which very much is healthy for optimizations.
However as per [[ http://eel.is/c++draft/basic.stc.dynamic.allocation | `[basic.stc.dynamic.allocation]` ]],
there are more promises on global `operator new`, namely:
* non-`std::nothrow_t` `operator new` *never* returns `nullptr`
* If `std::align_val_t align` parameter is taken, the pointer will also be `align`-aligned
* ~~global `operator new`-returned pointer is `__STDCPP_DEFAULT_NEW_ALIGNMENT__`-aligned ~~ It's more caveated than that.
Supplying this information may not cause immediate landslide effects
on any specific benchmarks, but it for sure will be healthy for optimizer
in the sense that the IR will better reflect the guarantees provided in the source code.
The caveat is `-fno-assume-sane-operator-new`, which currently prevents emitting `noalias`
attribute, and is automatically passed by Sanitizers ([[ https://bugs.llvm.org/show_bug.cgi?id=16386 | PR16386 ]]) - should it also cover these attributes?
The problem is that the flag is back-end-specific, as seen in `test/Modules/explicit-build-flags.cpp`.
But while it is okay to add `noalias` metadata in backend, we really should be adding at least
the alignment metadata to the AST, since that allows us to perform sema checks on it.
Reviewers: erichkeane, rjmccall, jdoerfert, eugenis, rsmith
Reviewed By: rsmith
Subscribers: xbolva00, jrtc27, atanasyan, nlopes, cfe-commits
Tags: #llvm, #clang
Differential Revision: https://reviews.llvm.org/D73380
2020-02-26 06:37:17 +08:00
|
|
|
};
|
2011-01-27 17:37:56 +08:00
|
|
|
struct B : A {};
|
|
|
|
|
|
2020-12-31 16:27:11 +08:00
|
|
|
// CHECK-LABEL: define{{.*}} void @_ZN5test34testEv()
|
2011-01-27 17:37:56 +08:00
|
|
|
void test() {
|
[clang] Annotating C++'s `operator new` with more attributes
Summary:
Right now we annotate C++'s `operator new` with `noalias` attribute,
which very much is healthy for optimizations.
However as per [[ http://eel.is/c++draft/basic.stc.dynamic.allocation | `[basic.stc.dynamic.allocation]` ]],
there are more promises on global `operator new`, namely:
* non-`std::nothrow_t` `operator new` *never* returns `nullptr`
* If `std::align_val_t align` parameter is taken, the pointer will also be `align`-aligned
* ~~global `operator new`-returned pointer is `__STDCPP_DEFAULT_NEW_ALIGNMENT__`-aligned ~~ It's more caveated than that.
Supplying this information may not cause immediate landslide effects
on any specific benchmarks, but it for sure will be healthy for optimizer
in the sense that the IR will better reflect the guarantees provided in the source code.
The caveat is `-fno-assume-sane-operator-new`, which currently prevents emitting `noalias`
attribute, and is automatically passed by Sanitizers ([[ https://bugs.llvm.org/show_bug.cgi?id=16386 | PR16386 ]]) - should it also cover these attributes?
The problem is that the flag is back-end-specific, as seen in `test/Modules/explicit-build-flags.cpp`.
But while it is okay to add `noalias` metadata in backend, we really should be adding at least
the alignment metadata to the AST, since that allows us to perform sema checks on it.
Reviewers: erichkeane, rjmccall, jdoerfert, eugenis, rsmith
Reviewed By: rsmith
Subscribers: xbolva00, jrtc27, atanasyan, nlopes, cfe-commits
Tags: #llvm, #clang
Differential Revision: https://reviews.llvm.org/D73380
2020-02-26 06:37:17 +08:00
|
|
|
// CHECK: [[CALL:%.*]] = call noalias nonnull i8* @_Znaj(i32 24)
|
|
|
|
|
// CHECK-NEXT: bitcast i8* [[CALL]] to i32*
|
2011-01-27 17:37:56 +08:00
|
|
|
// CHECK-NEXT: store i32 5
|
|
|
|
|
(void) new B[5];
|
|
|
|
|
}
|
|
|
|
|
}
|