See #59144 for more on why this is a bad idea.
Patch out the synthesized file unit accessor to only clear the source cache, then patch up all the places that were assuming they could iterate over the module's file list and see synthesized files.
rdar://94164512
PublicCMOSymbols stores symbols which are made public by cross-module-optimizations.
Those symbols are primarily stored in SILModule and eventually used by TBD generation and validation.
Add SymbolObjectCodeRequest, which emits object
code for a specific set of symbols by querying
TBDGen for the source of the symbols, and asking
SILGen (and eventually IRGen) to emit them.
Rename the member on ASTLoweringDescriptor and
IRGenDescriptor to make it more explicit it returns
the files that need emitting, rather than just the
files that happen to be present. This distinction
will become important once we start emitting code
only for a specific set of symbols.
With an inverted pipeline, IRGen needs to be able
to compute the linker directives itself, so sink
it down such that it can be computed by the
`IRGenDescriptor`.
Split off the notion of "recording" dependencies from the notion of
"collecting" dependencies. This corrects an oversight in the previous
design where dependency replay and recording were actually not "free" in
WMO where we actually never track dependencies. This architecture also
lays the groundwork for the removal of the referenced name trackers.
The algorithm builds upon the infrastructure for dependency sources and
sinks laid down during the cut over to request-based dependency tracking
in #30723.
The idea of the naive algorithm is this:
For a chain of requests A -> B* -> C -> D* -> ... -> L where L is a lookup
request and all starred requests are cached, once L writes into the
dependency collector, the active stack is walked and at each cache-point
the results of dependency collection are associated with the request
itself (in this example, B* and D* have all the names L found associated
with them). Subsequent evaluations of these cached requests (B* and D*
et al) will then *replay* the previous lookup results from L into the
active referenced name tracker. One complication is, suppose the
evaluation of a cached request involves multiple downstream name
lookups. More concretely, suppose we have the following request trace:
A* -> B -> L
|
-> C -> L
|
-> D -> L
|
-> ...
Then A* must see the union of the results of each L. If this reminds
anyone of a union-find, that is no accident! A persistent union-find
a la Conchon and Filliatre is probably in order to help bring down peak
heap usage...
Define a new type DependencyCollector that abstracts over the
incremental dependency gathering logic. This will insulate the
request-based name tracking code from future work on private,
intransitive dependencies.
swift::GeneratedModule encapsulates an llvm::Module, llvm::LLVMContext
pair that must live and die together. It has convenient accessors for
projecting the module and context components. The meat of this type is
the two conversion functions, which transfer ownership of either the
module component to the caller or the module and context to ORCJIT.
This is because ORC enforces an ownership contract that is distinct from
LLVM's rather wild ownership story for modules and their associated
contexts. See http://llvm.org/docs/ORCv2.html#how-to-use-threadsafemodule-and-threadsafecontext
