ELF is segment mapped, where the segment which contains a particular
section may be mapped to an address which does not correspond to the
address on disk. Since the reflection dumper does not use the loader to
load the image into memory, we must manually account for any section
offsets. Calculate this value when we query the mmap'ed image and wire
it through to the relative direct pointer accesses.
When switching to the linker table approach for the ELF metadata
introspection, this was uncovered as the segment containing the orphaned
sections was coalesced into a separate PT_LOAD header which had a non-0
offset for the mapping.
Currently when function types like `(_: Int...) -> Void` are mangled
their names are going to include enclosing sugar BoundGenericType(Array),
which is not necessary and doesn’t play well with `AnyFunctionType::Param`
which strips the sugar away.
Resolves: rdar://problem/34941557
The "superclass as associated type" modeling was put in to
maintain backward compatibility.
We just bumped the version number because of new mangling so
we may as well fix this sillyness too.
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 makes the demangler about 10 times faster.
It also changes the lifetimes of nodes. Previously nodes were reference-counted.
Now the returned demangle node-tree is owned by the Demangler class and it’s lifetime ends with the lifetime of the Demangler.
Therefore the old (and already deprecated) global functions demangleSymbolAsNode and demangleTypeAsNode are no longer available.
Another change is that the demangling for reflection now only supports the new mangling (which should be no problem because
we are generating only new mangled names for reflection).
These changes caused a number of issues:
1. No debug info is emitted when a release-debug info compiler is built.
2. OS X deployment target specification is broken.
3. Swift options were broken without any attempt any recreating that
functionality. The specific option in question is --force-optimized-typechecker.
Such refactorings should be done in a fashion that does not break existing
users and use cases.
This reverts commit e6ce2ff388.
This reverts commit e8645f3750.
This reverts commit 89b038ea7e.
This reverts commit 497cac64d9.
This reverts commit 953ad094da.
This reverts commit e096d1c033.
rdar://30549345
It also uses the new mangling for type names in meta-data (except for top-level non-generic classes).
lldb has now support for new mangled metadata type names.
This reinstates commit 21ba292943.
This patch splits add_swift_library into two functions one which handles
the simple case of adding a library that is part of the compiler being
built and the second handling the more complicated case of "target"
libraries, which may need to build for one or more targets.
The new add_swift_library is built using llvm_add_library, which re-uses
LLVM's CMake modules. In adapting to use LLVM's modules some of
add_swift_library's named parameters have been removed and
LINK_LIBRARIES has changed to LINK_LIBS, and LLVM_LINK_COMPONENTS
changed to LINK_COMPONENTS.
This patch also cleans up libswiftBasic's handling of UUID library and
headers, and how it interfaces with gyb sources.
add_swift_library also no longer has the FILE_DEPENDS parameter, which
doesn't matter because llvm_add_library's DEPENDS parameter has the same
behavior.
For this we are linking the new re-mangler instead of the old one into the swift runtime library.
Also we are linking the new de-mangling into the swift runtime library.
It also switches to the new mangling for class names of generic swift classes in the metadata.
Note that for non-generic class we still have to use the old mangling, because the ObjC runtime in the OS depends on it (it de-mangles the class names).
But names of generic classes are not handled by the ObjC runtime anyway, so there should be no problem to change the mangling for those.
The reason for this change is that it avoids linking the old re-mangler into the runtime library.
For generic multi-payload enums, we would proceed down the
dynamic layout path without checking for a builtin descriptor.
As a result Set and Dictionary were always reported as being
9 bytes in size and not 8. Oops...
Fixes <rdar://problem/30066015>.
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.
There were a few problems here with subclasses of Objective-C classes.
Use the InstanceStart field from rodata to correctly lay out instance
variables, and verify the results match with dynamic and static layout.
Better fix for <rdar://problem/27932061>.
The alignment was set to 0, which messed up the record layout
computations. Add an assert to catch this in the future.
Fixes <rdar://problem/29115967>.
Previously we would crash if we couldn't look up an associated
type witness for a concrete type.
Instead, propagate the failure up to type lowering, which returns
nullptr to the caller, just like when other metadata is missing.
This affects the computed stride for fixed-sized types in IRGen as well as the stored stride in value witness tables.
The reason is to let comparisons and difference operations work for pointers to zero-sized types.
(Currently this is achieved by using Builtin.strideof_nonzero in MemoryLayout.stride, but this requires a std::max(1, stride) operation after loading the stride)
To avoid crashing tools, we just propagate failure instead of
asserting. This makes debugging more difficult, though. When
this macro is enabled, messages are printed to standard error
to pinpoint the exact reason that type lowering bailed out.
The approach here is to split this into two cases:
- If all case payloads have a fixed size, spare bits may be
potentially used to differentiate between cases, and the
remote reflection library does not have enough information to
compute the layout itself.
However, the total size must be fixed, so IRGen just emits a
builtin type descriptor (which I need to rename to 'fixed type
descriptor' since these are also used for imported value types,
and now, certain enums).
- If at least one case has a size that depends on a generic
parameter or is a resilient type, IRGen does not know the size,
but this means fancy tricks with spare bits cannot be used either.
The remote reflection library uses the same approach as the
runtime, basically taking the maximum of the payload size and
alignment, and adding a tag byte.
As with single-payload enums, we produce a new kind of
RecordTypeInfo, this time with a field for every enum case.
All cases start at offset zero (but of course this might change,
if for example we put the enum tag before the address point).
Also, just as with single-payload enums, there is no remote
'project case index' operation on ReflectionContext yet.
So the the main benefit from this change is that we don't entirely
give up when doing layout of class instances containing enums;
however, tools still cannot look inside the enum values themselves,
except in the simplest cases involving optionals.
Notably, the remote reflection library finally understands all
of the standard library's collection types -- Array, Character,
Dictionary, Set, and String.
Attempt to lay out single-payload enums, using knowledge of extra
inhabitants where possible.
- The extra inhabitants of an aggregate are the extra inhabitants of
the first field. If the first field is empty, there are no extra
inhabitants, and subsequent fields do not affect anything.
- Function pointers and metatypes have different extra inhabitants
than Builtin.RawPointer, so have IRGen emit distinct builtin type
descriptors for those.
- Opaque existentials do not have extra inhabitants.
- Weak references do not have extra inhabitants.
Also, fix IRGen to emit more accurate enum reflection metadata in
these two cases:
- We now record whether enum cases are indirect or not. An indirect
case is the same as a payload case with Builtin.NativeObject.
- We now record whether a case is empty or not using the same logic
as the rest of IRGen. Previously, we would incorrectly emit a
payload type for a case with a payload that is an empty struct,
for example.
At this point we don't have a way to get the currently inhabited
enum case from a value. However, this is still an improvement because
we can still reflect other fields of aggregates containing enums,
instead of just giving up.
Finally make some methods on TypeCoverter private, and use 'friend'
to allow them to be accessed from other internal classes, making the
public API simpler.
As a first step to allowing the build script to build *only*
static library versions of the stdlib, change `add_swift_library`
such that callers must pass in `SHARED`, `STATIC`, or `OBJECT_LIBRARY`.
Ideally, only these flags would be used to determine whether to
build shared, static, or object libraries, but that is not currently
the case -- `add_swift_library` also checks whether the library
`IS_STDLIB` before performing certain additional actions. This will be
cleaned up in a future commit.
HOST_LIBRARY is supposed to mean "no matter what the defaults say, also build
this library for the host". FORCE_BUILD_FOR_HOST_SDK is a less confusing name.
@slava_pestov recently folded in @objc classes when building class field
descriptors - we just need to update the switch when considering records
for converting TypeRefs to TypeInfos.
rdar://problem/26594130
