Use the generic type lowering algorithm described in
"docs/CallingConvention.rst#physical-lowering" to map from IRGen's explosion
type to the type expected by the ABI.
Change IRGen to use the swift calling convention (swiftcc) for native swift
functions.
Use the 'swiftself' attribute on self parameters and for closures contexts.
Use the 'swifterror' parameter for swift error parameters.
Change functions in the runtime that are called as native swift functions to use
the swift calling convention.
rdar://19978563
This seems to more than fix a performance regression that we
detected on a metadata-allocation microbenchmark.
A few months ago, I improved the metadata cache representation
and changed the metadata allocation scheme to primarily use malloc.
Previously, we'd been using malloc in the concurrent tree data
structure but a per-cache slab allocator for the metadata itself.
At the time, I was concerned about the overhead of per-cache
allocators, since many metadata patterns see only a small number
of instantiations. That's still an important factor, so in the
new scheme we're using a global allocator; but instead of using
malloc for individual allocations, we're using a slab allocator,
which should have better peak, single-thread performance, at the
cost of not easily supporting deallocation. Deallocation is
only used for metadata when there's contention on the cache, and
specifically only when there's contention for the same key, so
leaking a little isn't the worst thing in the world.
The initial slab is a 64K globally-allocated buffer.
Successive slabs are 16K and allocated with malloc.
rdar://28189496
There are a few different use cases here:
1. In Raw SIL, no return folding may not have been run yet implying that a call
to a no-return function /can/ have arbitrary control flow after it (consider
mandatory inlined functions). We need to recognize that the region of code that
is strictly post dominated by the no-return function is "transitively
unreachable" and thus leaking is ok from that point. *Footnote 1*.
2. In Canonical and Raw SIL, we must recognize that unreachables and no-return
functions constitute places where we are allowed to leak.
rdar://29791263
----
*Footnote 1*: The reason why this is done is since we want to emit unreachable
code diagnostics when we run no-return folding. By leaving in the relevant code,
we have preserved all of the SILLocations on that code allowing us to create
really nice diagnostics.
The mangler never produces a mangling here, the demangler doesn't
demangle anything here, the remangler punted or asserted, and type
reconstruction did something very wrong. Delete this code.
Previously looking up an extension would result in all extensions for
types with the same name (nested or not) being deserialized; this
could even bring in base types that had not been deserialized yet. Add
in a string to distinguish an extension's base type; in the top-level
case this is just a module name, but for nested types it's a full
mangled name.
This is a little heavier than I'd like it to be, since it means we
mangle names and then throw them away, and since it means there's a
whole bunch of extra string data in the module just for uniquely
identifying a declaration. But it's correct, and does less work than
before, and fixes a circularity issue with a nested type A.B.A that
apparently used to work.
https://bugs.swift.org/browse/SR-3915
This reverts commit 1b3d29a163, reversing
changes made to b32424953e.
We're seeing a handful of issues from turning on inlining of generics,
so I'm reverting to unblock the bots.
Make the addSubstitution() and addConformance() methods private,
and declare GenericEnvironment and GenericSignature as friends of
SubstitutionMap.
At some point in the future, we can switch to a more efficient
representation of SubstitutionMap, where instead of storing
multiple hashtables, we store arrays; the keys are pre-determined.
This operator++ is a post-increment operator but it was implemented with pre-increment behavior.
This bug was exposed by the changes in llvm r294339 and caused the master-next branch to stop
building.
This essentially undoes the implementation in 51da51dfc0, which
implicitly did a substitution of the Self type in a protocol's
requirement signature by threading around the replacement PA. This is
brittle because every part of the code needs to take and pass around the
argument. By preemptively substituting, the whole requirement is in the
right form from the time it enters `addRequirement`.
The infrastructure here also allows simplifying some code.
Extensive cross-language tooling support needs to bridge decl names between two different languages more freely. This SourceKit request is designed to translate Objc names to Swift names and vice versa. Working similarly to cursor-info requisition, the name translation request requires a Swift reference to a Swift/Clang decl, and the preferred name to translate from, and language kind that the given name belongs to. If the translation succeeds, SourceKit service responds with the corresponding name than belongs to the other kind of language.
Newly introduced keys:
“key.namekind": “source.lang.name.kind.objc” | "source.lang.name.kind.swift"
“key.basename”: “name"
“key.argnames”: [“name"]
“key.selectorpieces”: [“name[:]"]
This commit only implements translation from Objc to Swift.
Instead of a global demangleSymbolAsNode, which returns a reference-counted NodePointer, there is now a Context class which owns the nodes.
So now demangleSymbolAsNode is a member of Context and the returned NodePointer is alive as long as the Context is alive.
This is still a NFC: the new ABI still maps to the old functions.
The purpose of this change is to let lldb adapt to the new API and then we can switch to the new implementation.
It is now possible to check for any apply if it can be devirtualized without actually performing the deirtualization. This could be used e.g. by inlining heuristics.
Remove the pre-expansion of all of the archetypes in a generic
environment; they can be constructed lazily from interface types.
Note that this only concerns the construction of the archetypes
themselves. The archetype builder is still pre-expanding all
*potential* archetypes.
"Fixes" rdar://problem/30351514, in the sense that the eager code and
the assertion that was getting tripped up are being eliminated
completely.
Rather than serializing the complete structure of all archetypes
(which is completely redundant), serialize a reference to their owning
generic environment as well as their interface type. The archetype
itself will be reconsituted by mapping the interface type into that
generic environment.
