The standard library uses `_precondition` calls which have a message argument.
Allow disabling the generated cond_fail by these message arguments.
For example:
_precondition(source >= (0 as T), "Negative value is not representable")
Results in a `cond_fail "Negative value is not representable"`.
This commit allows for specifying a file that contains these messages on
each line.
/path/to/disable_cond_fails:
```
Negative value is not representable
Array index is out of range
```
The optimizer will remove these cond_fails if the swift frontend is invoked with
`-Xllvm -cond-fail-config-file=/path/to/disable_cond_fails`.
* Reimplement most of the logic in Swift as an Instruction simplification and remove the old code from SILCombine
* support more cases of existential archetype replacements:
For example:
```
%0 = alloc_stack $any P
%1 = init_existential_addr %0, $T
use %1
```
is transformed to
```
%0 = alloc_stack $T
use %0
```
Also, if the alloc_stack is already an opened existential and the concrete type is known,
replace it as well:
```
%0 = metatype $@thick T.Type
%1 = init_existential_metatype %0, $@thick any P.Type
%2 = open_existential_metatype %1 : $@thick any P.Type to $@thick (@opened("X", P) Self).Type
...
%3 = alloc_stack $@opened("X", any P) Self
use %3
```
is transformed to
```
...
%3 = alloc_stack $T
use %3
```
At one point, OpenedArchetypeType did not exist as a separate subclass
of ArchetypeType, so this method did something. Now, it's just
equivalent to calling is<> or isa<>.
I also removed a couple of asserts that were obvious no-ops as a result.
Canonicalize a `fix_lifetime` from an address to a `load` + `fix_lifetime`:
```
%1 = alloc_stack $T
...
fix_lifetime %1
```
->
```
%1 = alloc_stack $T
...
%2 = load %1
fix_lifetime %2
```
This transformation is done for `alloc_stack` and `store_borrow` (which always has an `alloc_stack` operand).
The benefit of this transformation is that it enables other optimizations, like mem2reg.
This peephole optimization was already done in SILCombine, but it didn't handle store_borrow.
A good opportunity to make an instruction simplification out of it.
This is part of fixing regressions when enabling OSSA modules:
rdar://140229560
* Remove dead `load_borrow` instructions (replaces the old peephole optimization in SILCombine)
* If the `load_borrow` is followed by a `copy_value`, combine both into a `load [copy]`
The generality of the `AvailabilityContext` name made it seem like it
encapsulates more than it does. Really it just augments `VersionRange` with
additional set algebra operations that are useful for availability
computations. The `AvailabilityContext` name should be reserved for something
pulls together more than just a single version.
Although I don't plan to bring over new assertions wholesale
into the current qualification branch, it's entirely possible
that various minor changes in main will use the new assertions;
having this basic support in the release branch will simplify that.
(This is why I'm adding the includes as a separate pass from
rewriting the individual assertions)
This fixes a crash at runtime when destroying a Swift array of values of a C++ foreign reference type.
Swift optimizes the amount of metadata emitted for `_ContiguousArrayStorage<Element>` by reusing `_ContiguousArrayStorage<AnyObject>` whenever possible (see `getContiguousArrayStorageType`). However, C++ foreign reference types are not `AnyObject`s, since they have custom retain/release operations.
This change disables the `_ContiguousArrayStorage` metadata optimization for C++ reference types, which makes sure that `swift_arrayDestroy` will call the correct release operation for elements of `[MyCxxRefType]`.
rdar://127154770
If stores of empty values to the address don't dominate the
inject_enum_addr, the inject_enum_addr can't be eliminated--it's the
only indication of the case that's in the address.
Compute, update and handle borrowed-from instruction in various utilities and passes.
Also, used borrowed-from to simplify `gatherBorrowIntroducers` and `gatherEnclosingValues`.
Replace those utilities by `Value.getBorrowIntroducers` and `Value.getEnclosingValues`, which return a lazily computed Sequence of borrowed/enclosing values.
Don't attempt a SILCombine transform on `select_enum` that inserts a copy when an
enum is noncopyable. Adjust the cleanup handling for a consuming optional chain to
ensure a `destroy_value` still gets emitted on the `none` path; this shouldn't
actually matter since `none` is a trivial case, but the memory verifier isn't
that fancy during OSSA, and we can optimize it later. Fixes rdar://124426918.
It's better to ask SILType if it is MoveOnly than go to the AST type and
ask if it is noncopyable, because some types in SIL do not have a
well-defined notion of conformance in the AST.
Such destroys are part of the lifetime of lexical values inserted to
prevent shortening lifetimes over deinit baarriers by
CanonicalizeOSSALifetime. After OSSALifetimeCompletion is enabled, more
instructions will need to be preserved.
I think from SIL's perspective, it should only worry about whether the
type is move-only. That includes MoveOnlyWrapped SILTypes and regular
types that cannot be copied.
Most of the code querying `SILType::isPureMoveOnly` is in SILGen, where
it's very likely that the original AST type is sitting around already.
In such cases, I think it's fine to ask the AST type if it is
noncopyable. The clarity of only asking the ASTType if it's noncopyable
is beneficial, I think.
There are 2 reasons to disable this:
1. OSSA rauw creates copies for replacement, copies created in this case
cannot be canonicalized away due to the presence of pointer escape.
(See test test_mark_dependence_ossa_silcombine1 after -sil-combine -copy-propagation)
2. SILCombine of mark_dependence (enum can cause a potential miscompile.
ClosureLifetimeFixup specifically inserts enums to lifetime extend the closure
lifetime until the end of the function. Replacing mark_dependence's base operand
with a copy of the enum's operand can end up shortening the lifetime of the base.
(See test test_mark_dependence_ossa_silcombine2 after -sil-combine -dce -semantic-arc-opts)
The `bare` attribute indicates that the object header is not used throughout the lifetime of the object.
This means, no reference counting operations are performed on the object and its metadata is not used.
The header of bare objects doesn't need to be initialized.
This is phase-1 of switching from llvm::Optional to std::optional in the
next rebranch. llvm::Optional was removed from upstream LLVM, so we need
to migrate off rather soon. On Darwin, std::optional, and llvm::Optional
have the same layout, so we don't need to be as concerned about ABI
beyond the name mangling. `llvm::Optional` is only returned from one
function in
```
getStandardTypeSubst(StringRef TypeName,
bool allowConcurrencyManglings);
```
It's the return value, so it should not impact the mangling of the
function, and the layout is the same as `std::optional`, so it should be
mostly okay. This function doesn't appear to have users, and the ABI was
already broken 2 years ago for concurrency and no one seemed to notice
so this should be "okay".
I'm doing the migration incrementally so that folks working on main can
cherry-pick back to the release/5.9 branch. Once 5.9 is done and locked
away, then we can go through and finish the replacement. Since `None`
and `Optional` show up in contexts where they are not `llvm::None` and
`llvm::Optional`, I'm preparing the work now by going through and
removing the namespace unwrapping and making the `llvm` namespace
explicit. This should make it fairly mechanical to go through and
replace llvm::Optional with std::optional, and llvm::None with
std::nullopt. It's also a change that can be brought onto the
release/5.9 with minimal impact. This should be an NFC change.
The release needs to be preserved in case a user-defined deinit is
present in the released type. Checking for move-only is slightly
conservative.
Fixes rdar://109846094 ([move-only] SILCombine eliminates struct deinitialization)
