This commit changes how inline information is stored in SILDebugScope
from a tree to a linear chain of inlined call sites (similar to what
LLVM is using). This makes creating inlined SILDebugScopes slightly
more expensive, but makes lowering SILDebugScopes into LLVM metadata
much faster because entire inlined-at chains can now be cached. This
means that SIL is no longer preserve the inlining history (i.e., ((a
was inlined into b) was inlined into c) is represented the same as (a
was inlined into (b was inlined into c)), but this information was not
used by anyone.
On my late 2012 i7 iMac, this saves about 4 seconds when compiling the
RelWithDebInfo x86_64 swift standard library — or 40% of IRGen time.
rdar://problem/28311051
In particular, support the following optimizations:
- owned-to-guaranteed
- dead argument elimination
Argument explosion is disabled for generics at the moment as it usually leads to a slower code.
With -sil-serialize-all, we might deserialize and specialize
a private function from multiple TUs which are then linked
together.
Since private symbols have unique mangling, this is fine.
It's hard to make a test for this since it only happens with
-sil-serialize-all, which I'd like to kill soon anyway.
Also, add a third [serializable] state for functions whose bodies we
*can* serialize, but only do so if they're referenced from another
serialized function.
This will be used for bodies synthesized for imported definitions,
such as init(rawValue:), etc, and various thunks, but for now this
change is NFC.
This is useful for optimizations (like AllocBoxToStack) which create (de-)alloc_stack instructions.
They can just insert the new instructions anywhere without worrying about nesting and correct the nesting afterwards.
Use -sil-partial-specialization-with-generic-substitutions to enable the partial specialization even in cases of substitutions containing generic replacement types.
This improves the existing logic which is used to stop specialization for types that are too big to handle. It catches some pathological cases which hang the compiler.
Fixes rdar://30938882
Re-applying this commit, which was speculatively reverted. It turned out that that performance tests issues were unrelated.
Previously it was part of swiftBasic.
The demangler library does not depend on llvm (except some header-only utilities like StringRef). Putting it into its own library makes sure that no llvm stuff will be linked into clients which use the demangler library.
This change also contains other refactoring, like moving demangler code into different files. This makes it easier to remove the old demangler from the runtime library when we switch to the new symbol mangling.
Also in this commit: remove some unused API functions from the demangler Context.
fixes rdar://problem/30503344
This improves the existing logic which is used to stop specialization for types that are too big to handle. It catches some pathological cases which hang the compiler.
Fixes rdar://30938882
Change emitApplyOfLibraryIntrinsic() to take a SubstitutionMap,
and use the correct abstractions to build the map.
This gets rid of the last remaining uses of gatherAllSubstitutions()
in SIL.
