* WIP: Objective-C support
* Further work on implementation
* ObjC dynamic cast
* Add swift stub class attribute
* Classes, protocols and ivars
* Fix compilation issues
* Fix objc ir codegen
* Add objc linker option
* Add swift stub classref get ir gen
* Minor cleanup
* Fix objc link flag being added on non-darwin platforms
* Refactor objc gen
* remove use of std::nullopt
* Emit protocol tables
* Remove unused variable
* Formatting
* Fix build in release mode. Thanks for nothing, c++.
* Fix consistency
* Fix dynamic casts
* Fix tocall parentfd ref and arm msgsend call
* Make instance variables work
* Implicitly add isa pointer to objc classes.
* Fix protocol referencing & allow pragma mangle
* Fix protocol linkage
* Fix direct call support
* always generate var type for methods
* Fix test 16096a
* Fix extern ivar symbol gen, retain method decls
* Remove arm32 and x86 support
* Check method and ivar info before pushing to member list
* Make ObjcMethod info untyped.
* Make ivar and method gen more robust
* Generate optional protocol symbols
* Use bitcasting instead of creating multiple type defs
* Fix invalid protocol list struct gen
* More codegen robustness
* emit protocol table as const
* Make protocol table anon struct
* Fix callable type, generate protocol_list_t properly.
* Cast vthis to argtype
* Handle protorefs and classrefs properly
* seperate label ref and deref
* Fix method lookup
* Enable objective-c tests
* Enable objc_call_static test
* Scan both classes and protocols for method ref
* Enable objective-c tests on arm as well.
* supress objc linker warning in tests
* Fix class and protocol gen structure
* Fix objc_protocol_sections test
* ObjcMethod only get callee for functions with bodies
* Fix protocol class method gen
* Make ObjcMethod anon again
* Fix missing emit calls
* Fix classref gen
* Implement some of the requested changes
* Enable compilable tests
* Fix property selector gen, ugly hack for final funcs.
* Fix segfault in referencing fd->type
* Refactor implementation
* Fix null references in class and method lookup
* include unordered_map
* Get functionality on-par with prev impl.
* Fix super context calls
* Move -L-w flag to d_do_test and use IN_LLVM in objc.d/h
* add LDC version tag to -L-w flag
* Update CHANGELOG.md
Init symbols, TypeInfos (classes only) & vtables - unless defined in a
root module. For explicitly exported aggregates, or, with
`-fvisibility=public`, all aggregates.
As *the* way to access the IrType associated with a Type via its `ctype`
field. Most importantly, it makes sure all access is redirected to the
*unqualified* type's `ctype`, which allows to get rid of the 'multiple
types' workaround for aggregates in DtoType(). Those were e.g. hit for
`shared struct SpinLock`, where the struct's type includes the `shared`
modifier...
The semantics were almost identical, except for DtoIsTemplateInstance()
checking the specified Dsymbol and its parents, while isInstantiated()
starts from its parent and ignores the symbol itself.
After a quick glance, we seem to have fed it with {Aggregate,Func,Var}
Declarations only, so should be fine with the default front-end variant.
I.e., *define* templated symbols in each referencing compilation unit
when using discardable linkonce_odr linkage, analogous to C++.
This makes each compilation unit self-sufficient wrt. templated symbols,
which also means increased opportunity for inlining and less need for
LTO. There should be no more undefined symbol issues caused by buggy
template culling.
The biggest advantage is that the optimizer can discard unused
linkonce_odr symbols early instead of optimizing and forwarding to the
assembler. So this is especially useful with -O to decrease compilation
times and can at least in some scenarios greatly outweigh the
(potentially very much) higher number of symbols defined by the glue
layer.
Libraries compiled with -linkonce-templates can generally not be linked
against dependent code compiled without -linkonce-templates; the other
way around works.
Compiling the rt.util.typeinfo unittests leads to base-typed forward
declarations of built-in TypeInfos colliding with the init symbols for
their TypeInfo_* classes in rt.util.typeinfo.
We've already had a workaround for this in one place (typinf.cpp);
extend it to the other side as well (iraggr.cpp).
Also make sure to emit the TypeInfo metadata even if the IR symbol
already exists as init symbol.
And finally, keep init symbols of built-in TypeInfo classes mutable just
like any other TypeInfo, so that e.g. synchronized() can be used on the
implicit monitor.
LLVM already provides suited RAII helper types to restore the IRBuilder
state. [They sadly aren't movable, so I've had to wrap them in a
unique_ptr.]
While at it, also minimally revise debuginfo generation for functions.
Replace the stack of IRScopes, each with its own IRBuilder, by directly
tampering with the state of a single IRBuilder.
This seems like the most feasible way to account for a breaking change
in LLVM 11, disallowing IRBuilder from being copied.
Whenever we need an IR function, we'd better make sure it exists. Handle
that in DtoCallee(), by invoking DtoDeclareFunction() by default,
instead of the previous DtoResolveFunction() + DtoCallee() combo.
DtoResolveFunction() usually declares the function, but somehow doesn't
for abstract and body-less functions.
I've also compared against the current DMD implementations, to make sure
we are in sync.
DMD doesn't populate `TypeInfo_Class.m_offTi` either. This seems like a
leftover of some unimplemented idea (the LDC commit introducing the
CMake option is from 2008); the normal `TypeInfo_Class.interfaces` array
contains the offsets nowadays.
Incl. getting rid of the dependence on an associated
TypeInfoStructDeclaration, which was only really needed for the mangled
name of the global.
Also slightly revise metadata generation.
