This eliminates all inline creation of SILBasicBlock via placement new.
There are a few reasons to do this:
1. A SILBasicBlock is always created with a parent function. This commit
formalizes this into the SILBasicBlock API by only allowing for SILFunctions to
create SILBasicBlocks. This is implemented via the type system by making all
SILBasicBlock constructors private. Since SILFunction is a friend of
SILBasicBlock, SILFunction can still create a SILBasicBlock without issue.
2. Since all SILBasicBlocks will be created in only a few functions, it becomes
very easy to determine using instruments the amount of memory being allocated
for SILBasicBlocks by simply inverting the call tree in Allocations.
With LTO+PGO, normal inlining can occur if profitable so there shouldn't be
overhead that we care about in shipping compilers.
This lets us get to the goal of +0 guaranteed closure contexts. NFC yet, just add the under-the-hood ability for partial_apply instructions producing callee-guaranteed closures to be parsed, printed, and serialized.
Till now there was no way in SIL to explicitly express a dependency of an instruction on any opened archetypes used by it. This was a cause of many errors and correctness issues. In many cases the code was moved around without taking into account these dependencies, which resulted in breaking the invariant that any uses of an opened archetype should be dominated by the definition of this archetype.
This patch does the following:
- Map opened archetypes to the instructions defining them, i.e. to open_existential instructions.
- Introduce a helper class SILOpenedArchetypesTracker for creating and maintaining such mappings.
- Introduce a helper class SILOpenedArchetypesState for providing a read-only API for looking up available opened archetypes.
- Each SIL instruction which uses an opened archetype as a type gets an additional opened archetype operand representing a dependency of the instruction on this archetype. These opened archetypes operands are an in-memory representation. They are not serialized. Instead, they are re-constructed when reading binary or textual SIL files.
- SILVerifier was extended to conduct more thorough checks related to the usage of opened archetypes.
Till now there was no way in SIL to explicitly express a dependency of an instruction on any opened archetypes used by it. This was a cause of many errors and correctness issues. In many cases the code was moved around without taking into account these dependencies, which resulted in breaking the invariant that any uses of an opened archetype should be dominated by the definition of this archetype.
This patch does the following:
- Map opened archetypes to the instructions defining them, i.e. to open_existential instructions.
- Introduce a helper class SILOpenedArchetypesTracker for creating and maintaining such mappings.
- Introduce a helper class SILOpenedArchetypesState for providing a read-only API for looking up available opened archetypes.
- Each SIL instruction which uses an opened archetype as a type gets an additional opened archetype operand representing a dependency of the instruction on this archetype. These opened archetypes operands are an in-memory representation. They are not serialized. Instead, they are re-constructed when reading binary or textual SIL files.
- SILVerifier was extended to conduct more thorough checks related to the usage of opened archetypes.
Till now there was no way in SIL to explicitly express a dependency of an instruction on any opened archetypes used by it. This was a cause of many errors and correctness issues. In many cases the code was moved around without taking into account these dependencies, which resulted in breaking the invariant that any uses of an opened archetype should be dominated by the definition of this archetype.
This patch does the following:
- Map opened archetypes to the instructions defining them, i.e. to open_existential instructions.
- Introduce a helper class SILOpenedArchetypesTracker for creating and maintaining such mappings.
- Introduce a helper class SILOpenedArchetypesState for providing a read-only API for looking up available opened archetypes.
- Each SIL instruction which uses an opened archetype as a type gets an additional opened archetype operand representing a dependency of the instruction on this archetype. These opened archetypes operands are an in-memory representation. They are not serialized. Instead, they are re-constructed when reading binary or textual SIL files.
- SILVerifier was extended to conduct more thorough checks related to the usage of opened archetypes.
not have access to their type arguments at runtime. Use this to
fix the emission of native thunks for imported ObjC-generic
initializers, since they may need to perform bridging.
For now, pseudo-genericity is all-or-nothing, but we may want to
make it apply only to certain type arguments.
Also, clean up some code that was using dead mangling nodes.
We do clever tricks to avoid having to serialize the full type of a
partial_apply instruction, but those conflicted with our clever tricks
for handling unusual return conventions. Make sure we do all type
calculations for partial_apply in one place. (Credit to Slava for
both introducing the problem and immediately seeing how to fix it.)
This should allow re-application of e6a519f and 86dcce1.
The overhead of uniquing the locations in a Densemap isn't worth any of
the potential memory savings: While this adds an extra pointer and
unsigned to each SILInstruction, any extra memory is completely lost in
the noise (measured on a release -emit-ir build of the x86_64 stdlib).
This is not too surpising as the ratio between SILInstructions and unique
SILLocations is not very high and the DenseMap also needs space.
<rdar://problem/22706994>
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.
The drivers for this change are providing a simpler API to SIL pass
authors, having a more efficient of the in-memory representation,
and ruling out an entire class of common bugs that usually result
in hard-to-debug backend crashes.
Summary
-------
SILInstruction
Old New
+---------------+ +------------------+ +-----------------+
|SILInstruction | |SILInstruction | |SILDebugLocation |
+---------------+ +------------------+ +-----------------+
| ... | | ... | | ... |
|SILLocation | |SILDebugLocation *| -> |SILLocation |
|SILDebugScope *| +------------------+ |SILDebugScope * |
+---------------+ +-----------------+
We’re introducing a new class SILDebugLocation which represents the
combination of a SILLocation and a SILDebugScope.
Instead of storing an inline SILLocation and a SILDebugScope pointer,
SILInstruction now only has one SILDebugLocation pointer. The APIs of
SILBuilder and SILDebugLocation guarantees that every SILInstruction
has a nonempty SILDebugScope.
Developer-visible changes include:
SILBuilder
----------
In the old design SILBuilder populated the InsertedInstrs list to
allow setting the debug scopes of all built instructions in bulk
at the very end (as the responsibility of the user). In the new design,
SILBuilder now carries a "current debug scope" state and immediately
sets the debug scope when an instruction is inserted.
This fixes a use-after-free issue with with SIL passes that delete
instructions before destroying the SILBuilder that created them.
Because of this, SILBuilderWithScopes no longer needs to be a template,
which simplifies its call sites.
SILInstruction
--------------
It is neither possible or necessary to manually call setDebugScope()
on a SILInstruction any more. The function still exists as a private
method, but is only used when splicing instructions from one function
to another.
Efficiency
----------
In addition to dropping 20 bytes from each SILInstruction,
SILDebugLocations are now allocated in the SILModule's bump pointer
allocator and are uniqued by SILBuilder. Unfortunately repeat compiles
of the standard library already vary by about 5% so I couldn’t yet
produce reliable numbers for how much this saves overall.
rdar://problem/22017421
This improves support for promoting to and generating
unchecked_ref_cast so we no longer need unchecked_ref_bit_cast, which
will just go away in the next commit.
Swift SVN r32597
Unless we can prove that the input and output are layout identical we must use either trivial_bit_cast or bitwise_cast.
This changes the lowering to be more conservative for:
- Builtin.reinterpretCast
- address_to_pointer -> pointer_to_address
This also makes LoadStoreOpts more conservative in:
- forwardAddrToUncheckedCastToLd
In order to commit this without massive performance regressions, I added SILType::canBitCastAsSingleRef. When this is true for both sides
of a bitwise case, we can convert it to a RefBitCast, which has RC identity.
Swift SVN r30172
emit{StrongRelease,ReleaseValue} => emit{StrongRelease,ReleaseValue}AndFold.
Then introduce a new method emit{StrongRelease,ReleaseValue} that returns a
PointerUnion containing the increment to be deleted if it exists. This obviates
the need for the callback.
Swift SVN r27804
These aren't really orthogonal concerns--you'll never have a @thick @cc(objc_method), or an @objc_block @cc(witness_method)--and we have gross decision trees all over the codebase that try to hopscotch between the subset of combinations that make sense. Stop the madness by eliminating AbstractCC and folding its states into SILFunctionTypeRepresentation. This cleans up a ton of code across the compiler.
I couldn't quite eliminate AbstractCC's information from AST function types, since SIL type lowering transiently created AnyFunctionTypes with AbstractCCs set, even though these never occur at the source level. To accommodate type lowering, allow AnyFunctionType::ExtInfo to carry a SILFunctionTypeRepresentation, and arrange for the overlapping representations to share raw values.
In order to avoid disturbing test output, AST and SILFunctionTypes are still printed and parsed using the existing @thin/@thick/@objc_block and @cc() attributes, which is kind of gross, but lets me stage in the real source-breaking change separately.
Swift SVN r27095
"Autoclosure" is uninteresting to SIL. "noescape" isn't currently used by SIL and we shouldn't have it until it has a meaningful effect on SIL. "throws" should be adequately represented by a SIL function type having an error result.
Swift SVN r27023
This change permits SILGen to make smarter decisions about
block placement by keeping related blocks together instead
of always inserting to the end to the function. The
flipside is that SILGen needs to be somewhat careful to
create blocks in the right order. Counter-intuitively,
that order is the reverse of the order in which the blocks
should be laid out, since blocks created later will be
inserted before blocks created earlier. Note, however,
that this produces the right results for recursive
emission.
To that end, adjust a couple of places in SILGen to
create blocks in properly nested order.
All of the block-order differences in the tests seem
to be desirable; several of them even had confused
comments wondering how on earth a block got injected
where it did.
Also, fix the implementation of SILBuilder::moveBlockTo,
and fix a latent bug in epilogue emission where epilogBB
was erased from its parent (deleting it) and then
queried multiple times (!).
Swift SVN r26428
This is part of rdar://16323038. Because this hasn't been fully design reviewed and
implemented, I'm naming it as __nocapture for now. It is blocking finishing off the
"improved let model" work.
Swift SVN r24079
without a valid SILDebugScope. An assertion in IRGenSIL prevents future
optimizations from regressing in this regard.
Introducing SILBuilderWithScope and SILBuilderwithPostprocess to ease the
transition.
This patch is large, but mostly mechanical.
<rdar://problem/18494573> Swift: Debugger is not stopping at the set breakpoint
Swift SVN r22978
These bits are orthogonal to each other, so combine them into one, and diagnose attempts to produce a type that's both. Spot-fix a bunch of places this revealed by inspection that we would have crashed in SILGen or IRGen if blocks were be handled.
Swift SVN r16088
retain_value instructions, just like we do for strong_retain/release.
This wraps up rdar://15889208, which is admittedly more of a moral
victory than a practical one.
Swift SVN r15804
The 'metatype' instruction can produce either a thick or Objective-C
metatype value for a class. Sometimes SILGen picks the "wrong" one,
and ends up converting the value directly, which is sub-optimal. Clean
this up in SILBuilder.
I didn't add a corresponding SILCombine optimization, nor did I add
this for class_metatype, because it seems unlikely that those will
occur in practice.
Swift SVN r14183
There are some straggling references to the context generic param list, but nothing uses the non-interface param or result types anymore!
Swift SVN r13725
Lower types for SILDeclRefs from the interface types of their referents, dragging the old type along for the ride so we can still offer the context to clients that haven't been weaned off of it. Make SILFunctionType's interface types and generic signature independent arguments of its Derive the context types of SILFunctionType from the interface types, instead of the other way around. Do a bunch of annoying inseparable work in the AST and IRGen to accommodate the switchover.
Swift SVN r12536
In nongeneric contexts, or contexts where we only care about the indirectness of parameters or have already substituted the generic parameters for a function, the interface types are interchangeable, so just switch over.
Swift SVN r12044
