Once we move to a copy-on-write implementation of existential value buffers we
can no longer consume or destroy values of an opened existential unless the
buffer is uniquely owned.
Therefore we need to track the allowed operation on opened values.
Add qualifiers "mutable_access" and "immutable_access" to open_existential_addr
instructions to indicate the allowed access to the opened value.
Once we move to a copy-on-write implementation, an "open_existential_addr
mutable_access" instruction will ensure unique ownership of the value buffer.
As per John, WritebackScope was always an unfortunate name. Generally these
scopes are meant for formal evaluations of inout parameters. The cases that I am
interested in generalizing them to be used for are borrows of the base of a
class that will then be used as an lvalue.
This also eliminates the out of line vector of lvalue writebacks.
rdar://29791263
SubstitutionList is going to be a more compact representation of
a SubstitutionMap, suitable for inline allocation inside another
object.
For now, it's just a typedef for ArrayRef<Substitution>.
Most of this involved sprinkling ValueOwnershipKind::Owned in many places. In
some of these places, I am sure I was too cavalier and I expect some of them to
be trivial. The verifier will help me to track those down.
On the other hand, I do expect there to be some places where we are willing to
accept guaranteed+trivial or owned+trivial. In those cases, I am going to
provide an aggregate ValueOwnershipKind that will then tell SILArgument that it
should disambiguate using the type. This will eliminate the ackwardness from
such code.
I am going to use a verifier to fix such cases.
This commit also begins the serialization of ValueOwnershipKind of arguments,
but does not implement parsing of value ownership kinds. That and undef are the
last places that we still use ValueOwnershipKind::Any.
rdar://29791263
We preserve the current behavior of assuming Any ownership always and use
default arguments to hide this change most of the time. There are asserts now in
the SILBasicBlock::{create,replace,insert}{PHI,Function}Argument to ensure that
the people can only create SILFunctionArguments in entry blocks and
SILPHIArguments in non-entry blocks. This will ensure that the code in tree
maintains the API distinction even if we are not using the full distinction in
between the two.
Once the verifier is finished being upstreamed, I am going to audit the
createPHIArgument cases for the proper ownership. This is b/c I will be able to
use the verifier to properly debug the code. At that point, I will also start
serializing/printing/parsing the ownershipkind of SILPHIArguments, but lets take
things one step at a time and move incrementally.
In the process, I also discovered a CSE bug. I am not sure how it ever worked.
Basically we replace an argument with a new argument type but return the uses of
the old argument to refer to the old argument instead of a new argument.
rdar://29671437
Changes:
* Terminate all namespaces with the correct closing comment.
* Make sure argument names in comments match the corresponding parameter name.
* Remove redundant get() calls on smart pointers.
* Prefer using "override" or "final" instead of "virtual". Remove "virtual" where appropriate.
The RValue(ArrayRef<ManagedValue>, CanType) constructor was intended as a semi-private interface for building an RValue from a pre-exploded array of elements, but was (understandably) widely being misused as a general ManagedValue-to-RValue constructor, causing crashes when working with tuples in various contexts where RValue's methods expected them to be exploded. Make the constructor private and update most improper uses of it to use the exploding RValue constructor, or to use a new `RValue::withPreExplodedElements` static method that more explicitly communicates the intent of the constructor. Fixes rdar://problem/29500731.
Before this commit all code relating to handling arguments in SILBasicBlock had
somewhere in the name BB. This is redundant given that the class's name is
already SILBasicBlock. This commit drops those names.
Some examples:
getBBArg() => getArgument()
BBArgList => ArgumentList
bbarg_begin() => args_begin()
Keep in mind that these are approximations that will not impact correctness
since in all cases I ensured that the SIL will be the same after the
OwnershipModelEliminator has run. The cases that I was unsure of I commented
with SEMANTIC ARC TODO. Once we have the verifier any confusion that may have
occurred here will be dealt with.
rdar://28685236
This ensures that ownership is properly propagated forward through the use-def
graph.
This was the work that was stymied by issues relating to SILBuilder performing
local ARC dataflow. I ripped out that local dataflow in 6f4e2ab and added a
cheap ARC guaranteed dataflow pass that performs the same optimization.
Also in the process of doing this work, I found that there were many SILGen
tests that were either pattern matching in the wrong functions or had wrong
CHECK lines (for instance CHECK_NEXT). I fixed all of these issues and also
expanded many of the tests so that they verify ownership. The only work I left
for a future PR is that there are certain places in tests where we are using the
projection from an original value, instead of a copy. I marked those with a
message SEMANTIC ARC TODO so that they are easy to find.
rdar://28685236
Today, loads and stores are treated as having @unowned(unsafe) ownership
semantics. This leaves the user to specify ownership changes on the loaded or
stored value independently of the load/store by inserting ARC operations. With
the change to Semantic SIL, this will no longer be true. Instead loads, stores
have ownership semantics that one must reason about such as copy, take, and
trivial.
This change moves us closer to that world by eliminating the default
OwnershipQualification argument from create{Load,Store}. This means that the
compiler developer cannot ignore reasoning about the ownership semantics of the
memory operation that they are creating.
Operationally, this is a NFC change since I have just gone through the compiler
and updated all places where we create loads, stores to pass in the former
default argument ({Load,Store}OwnershipQualifier::Unqualified), to
SILBuilder::create{Load,Store}(...). For now, one can just do that in situations
where one needs to create loads/stores, but over time, I am going to tighten the
semantics up via the verifier.
rdar://28685236
This is still a pretty bad code-generation pattern, but the layering
stuff makes it challenging to do the right thing.
Also, bridge non-optional NSErrors to Error using init_existential_ref
instead of going through the runtime function.
rdar://27810321
Naively wrapping a T in an Any would require metadata for T, but we can go to AnyObject first and put the AnyObject in the Any without T's metadata. This fixes a regression in ObjC generic extension methods when they try to pass their generic parameters to id parameters as Any. rdar://problem/27526877
When emitting an existential erasure to Error from an archetype, use
the _getEmbeddedNSError() witness. If it produces an NSError, erase
that; otherwise, go through the normal erasure path.
Of course, make NSError and CFError implement _getEmbeddedNSError() so
this kicks in for the obvious cases as well as the more obscure ones.
Fixes the rest of SR-1562 / rdar://problem/26370984.
Imported Cocoa error types are represented by structs wrapping an
NSError. The conversion from these structs to Error would end up
boxing the structs in _SwiftNativeNSError, losing identity and leading
to a wrapping loop.
Instead, extract the embedded NSError if there is one. In the Swift
runtime, do this as part of the dynamic cast to NSError, using a (new,
defaulted) requirement in the Error type so we can avoid an extra
runtime lookup of the protocol. In SILGEn, do this by looking for the
_BridgedStoredNSError protocol conformance when erasing to an Error
type. Fixes SR-1562 / rdar://problem/26370984.
If a `T?` is converted to `Any?` to pass to a `_Nullable id` interface in ObjC, avoid the intermediate conversion and bridge straight to AnyObject? too.
Being generic, the '_unwrapped' intrinsics force trafficking through memory, and while they're transparent so always get inlined, we don't do memory promotion in -Onone. Emitting the branch inline lets loadable optionals stay values leading to better -Onone codegen. (It also lets us throw away a surprising amount of support code for these optional intrinsics.)
Similarly to how we've always handled parameter types, we
now recursively expand tuples in result types and separately
determine a result convention for each result.
The most important code-generation change here is that
indirect results are now returned separately from each
other and from any direct results. It is generally far
better, when receiving an indirect result, to receive it
as an independent result; the caller is much more likely
to be able to directly receive the result in the address
they want to initialize, rather than having to receive it
in temporary memory and then copy parts of it into the
target.
The most important conceptual change here that clients and
producers of SIL must be aware of is the new distinction
between a SILFunctionType's *parameters* and its *argument
list*. The former is just the formal parameters, derived
purely from the parameter types of the original function;
indirect results are no longer in this list. The latter
includes the indirect result arguments; as always, all
the indirect results strictly precede the parameters.
Apply instructions and entry block arguments follow the
argument list, not the parameter list.
A relatively minor change is that there can now be multiple
direct results, each with its own result convention.
This is a minor change because I've chosen to leave
return instructions as taking a single operand and
apply instructions as producing a single result; when
the type describes multiple results, they are implicitly
bound up in a tuple. It might make sense to split these
up and allow e.g. return instructions to take a list
of operands; however, it's not clear what to do on the
caller side, and this would be a major change that can
be separated out from this already over-large patch.
Unsurprisingly, the most invasive changes here are in
SILGen; this requires substantial reworking of both call
emission and reabstraction. It also proved important
to switch several SILGen operations over to work with
RValue instead of ManagedValue, since otherwise they
would be forced to spuriously "implode" buffers.
And use the new project_existential_box to get to the address value.
SILGen now generates a project_existential_box for each alloc_existential_box.
And IRGen re-uses the address value from the alloc_existential_box if the operand of project_existential_box is an alloc_existential_box.
This lets the generated code be the same as before.
This eliminates some minor overheads, but mostly it eliminates
a lot of conceptual complexity due to the overhead basically
appearing outside of its context.
The main idea here is that we really, really want to be
able to recover the protocol requirement of a conformance
reference even if it's abstract due to the conforming type
being abstract (e.g. an archetype). I've made the conversion
from ProtocolConformance* explicit to discourage casual
contamination of the Ref with a null value.
As part of this change, always make conformance arrays in
Substitutions fully parallel to the requirements, as opposed
to occasionally being empty when the conformances are abstract.
As another part of this, I've tried to proactively fix
prospective bugs with partially-concrete conformances, which I
believe can happen with concretely-bound archetypes.
In addition to just giving us stronger invariants, this is
progress towards the removal of the archetype from Substitution.
