In void*-to-ptr casts, the type of the pointed-to object in the source
operand needs to be compared to the target pointee type.
If a block was created for a `new`/`new[]`/`std::allocator` expression,
then a pointer needs to be stripped from the type of the expression
(which points to the single-object allocation or first element of the
allocation) to get the former.
`Descriptor::getType` did not do this and `Descriptor::getDataType`
returns an array type for array allocations. Therefore this introduces a
new function `Descriptor::getDataElemType` with the same behavior as
`Descriptor::getDataType`, except that it always produces the element
type in the array case and avoids the need for an `ASTContext`
reference. Make `Pointer::getType` use this function instead.
Fixes#174131
As mentioned a few times in the past, the previous handling using a
`optional<pair<bool, std::shared_ptr<>>>` was confusing and nobody ever
remembered what the optional being unset meant or what the bool stood
for.
Add an `InitMapPtr` struct that wraps a `uintptr_t` that either holds a
pointer to a valid `InitMap` instance _or_ one of two special values.
The struct has meaningful accessors for the various special cases that
were confusing before.
Save them as a pointer intead of using a shared_ptr. This we we can use
the pointer integer value to differentiate the "no initmap yet" and "all
values initialzed" cases.
This regresses one test case in const-eval.c, but as it turns out, that
only worked coincidentally before.
We only call this when we just pushed a new pointer to the stack, so try
to save the folling PopPtr op by removing the pointer inside
emitDestruction directly, e.g. by letting the Call op just remove it.
This way, we can check a single uint8_t for != 0 to know whether this
block is accessible or not. If not, we still need to figure out why not
and diagnose appropriately of course.
This is a major change on how we represent nested name qualifications in
the AST.
* The nested name specifier itself and how it's stored is changed. The
prefixes for types are handled within the type hierarchy, which makes
canonicalization for them super cheap, no memory allocation required.
Also translating a type into nested name specifier form becomes a no-op.
An identifier is stored as a DependentNameType. The nested name
specifier gains a lightweight handle class, to be used instead of
passing around pointers, which is similar to what is implemented for
TemplateName. There is still one free bit available, and this handle can
be used within a PointerUnion and PointerIntPair, which should keep
bit-packing aficionados happy.
* The ElaboratedType node is removed, all type nodes in which it could
previously apply to can now store the elaborated keyword and name
qualifier, tail allocating when present.
* TagTypes can now point to the exact declaration found when producing
these, as opposed to the previous situation of there only existing one
TagType per entity. This increases the amount of type sugar retained,
and can have several applications, for example in tracking module
ownership, and other tools which care about source file origins, such as
IWYU. These TagTypes are lazily allocated, in order to limit the
increase in AST size.
This patch offers a great performance benefit.
It greatly improves compilation time for
[stdexec](https://github.com/NVIDIA/stdexec). For one datapoint, for
`test_on2.cpp` in that project, which is the slowest compiling test,
this patch improves `-c` compilation time by about 7.2%, with the
`-fsyntax-only` improvement being at ~12%.
This has great results on compile-time-tracker as well:
This patch also further enables other optimziations in the future, and
will reduce the performance impact of template specialization resugaring
when that lands.
It has some other miscelaneous drive-by fixes.
About the review: Yes the patch is huge, sorry about that. Part of the
reason is that I started by the nested name specifier part, before the
ElaboratedType part, but that had a huge performance downside, as
ElaboratedType is a big performance hog. I didn't have the steam to go
back and change the patch after the fact.
There is also a lot of internal API changes, and it made sense to remove
ElaboratedType in one go, versus removing it from one type at a time, as
that would present much more churn to the users. Also, the nested name
specifier having a different API avoids missing changes related to how
prefixes work now, which could make existing code compile but not work.
How to review: The important changes are all in
`clang/include/clang/AST` and `clang/lib/AST`, with also important
changes in `clang/lib/Sema/TreeTransform.h`.
The rest and bulk of the changes are mostly consequences of the changes
in API.
PS: TagType::getDecl is renamed to `getOriginalDecl` in this patch, just
for easier to rebasing. I plan to rename it back after this lands.
Fixes#136624
Fixes https://github.com/llvm/llvm-project/issues/43179
Fixes https://github.com/llvm/llvm-project/issues/68670
Fixes https://github.com/llvm/llvm-project/issues/92757
For mutable and const fields, we have two bits in InlineDescriptor,
which both get inherited down the hierarchy. When a field is both const
and mutable, we CAN read from it if it is a mutable-in-const field, but
we _can't_ read from it if it is a const-in-mutable field. We need
another bit to distinguish the two cases.
We sometimes used to have a long list of
```
GetLocalPtr
PopPtr
[...]
```
ops at the end of scopes, because we first got a pointer to a local
variable and only then did we figure out that we didn't actually want to
call the destructor for it. Add a new function that allows us to just
ask the `Descriptor` whether we need to call its destructor.
…types usi… (#144676)"
This reverts commit 68471d29eed2c49f9b439e505b3f24d387d54f97.
IntegralAP contains a union:
union {
uint64_t *Memory = nullptr;
uint64_t Val;
};
On 64bit systems, both Memory and Val have the same size. However, on 32
bit system, Val is 64bit and Memory only 32bit. Which means the default
initializer for Memory will only zero half of Val. We fixed this by
zero-initializing Val explicitly in the IntegralAP(unsigned BitWidth)
constructor.
See also the discussion in
https://github.com/llvm/llvm-project/pull/144246
Both `APInt` and `APFloat` will heap-allocate memory themselves using
the system allocator when the size of their data exceeds 64 bits.
This is why clang has `APNumericStorage`, which allocates its memory
using an allocator (via `ASTContext`) instead. Calling `getValue()` on
an ast node like that will then create a new `APInt`/`APFloat` , which
will copy the data (in the `APFloat` case, we even copy it twice).
That's sad but whatever.
In the bytecode interpreter, we have a similar problem. Large integers
and floating-point values are placement-new allocated into the
`InterpStack` (or into the bytecode, which is a `vector<std::byte>`).
When we then later interrupt interpretation, we don't run the destructor
for all items on the stack, which means we leak the memory the
`APInt`/`APFloat` (which backs the `IntegralAP`/`Floating` the
interpreter uses).
Fix this by using an approach similar to the one used in the AST. Add an
allocator to `InterpState`, which is used for temporaries and local
values. Those values will be freed at the end of interpretation. For
global variables, we need to promote the values to global lifetime,
which we do via `InitGlobal` and `FinishInitGlobal` ops.
Interestingly, this results in a slight _improvement_ in compile times:
https://llvm-compile-time-tracker.com/compare.php?from=6bfcdda9b1ddf0900f82f7e30cb5e3253a791d50&to=88d1d899127b408f0fb0f385c2c58e6283195049&stat=instructions:u
(but don't ask me why).
Fixes https://github.com/llvm/llvm-project/issues/139012
For
```c++
struct S {
constexpr S(int=0) : i(1) {}
int i;
};
constexpr volatile S vs;
```
reading from `vs.i` is not allowed, even though `i` is not volatile
qualified. Propagate the IsVolatile bit down the hierarchy, so we know
reading from `vs.i` is a volatile read.
This returns the type of data in the Block, which might be different
than the type of the expression or declaration we created the block for.
This lets us remove some special cases from CheckNewDeleteForms() and
CheckNewTypeMismatch().
When creating descriptor for array element types, we only save the
original source, e.g. int[2][2][2]. So later calls to getType() of the
element descriptors will also return int[2][2][2], instead of e.g.
int[2][2] for the second dimension.
Fix this by explicitly tracking the array types.
The last attached test case used to have an lvalue offset of 32 instead
of 24.
We should do this for more desriptor types though and not just composite
array, but I'm leaving that to a later patch.
Note that PointerUnion::dyn_cast has been soft deprecated in
PointerUnion.h:
// FIXME: Replace the uses of is(), get() and dyn_cast() with
// isa<T>, cast<T> and the llvm::dyn_cast<T>
Literal migration would result in dyn_cast_if_present (see the
definition of PointerUnion::dyn_cast), but this patch uses dyn_cast
because we expect Source to be nonnull.
