This ensures correct generation of generic signatures in the resulting
byte code, but it _is_ a work in progress: the actual type *arguments*
passed for these parameters during compilation are dummy `Any?` types.
Sites that need more work are indicated with TODO's.
- copy type parameters of interfaces to methods moved to DefaultImpls
- implement type parameter renaming scheme from JVM, with proper
renaming and substitution.
- adjust call sites in bridges in classes->DefaultImpls
- adjust call sites in bridges from DefaultImpls->Interface
- adjust call sites in bridges from DefaultImpls->DefaultImpls
- adjust super calls ->DefaultImpls
- adjust calls in code of Interfaces->DefaultImpls
Otherwise, it's complicated to work with nested lambdas:
outer/inner ones may be replaced independently and still refer
to the old one instances.
So in the changed test, if we don't apply the commit, will remain
implicit return and receiver types for the nested lambda
Anyway, once we decided to leave immutable semantics,
replacements are not necessary anymore
This way we're making sure that the copied field has some associated
property, where we can get the value of isConst flag from. That flag is
later used in StaticInitializersLowering to determine whether we need to
erase initializer of a field. The tests are unmuted because now the
initializer is correctly _not_ erased. (They were passing before
switching master to 1.4 because without
NoConstantValueAttributeForNonConstVals, we treated all static fields
with primitive/string values as const, and never erased initializers
because of that.)
Generate the expression with the original type and then insert
an implicit conversion. That matches the behavior of psi2ir
better and therefore avoids breaking backend assumptions.
In particular, IrGetValue expects the type of the underlying
symbol and the type of the IrGetValue to be the same.
When a property backed by a final field is the delegate for interface
implementation by delegation, the backend should reuse the backing
field rather than introduce a new, extraneous `$$delegate_n` field.
This change adds D8 in addition to Dx which is being deprecated.
At some point after the official Dx deprecation, we should remove
the Dx checker.
Moving to D8 has the additional benefit that D8 can dex code
using java 8 features without using a separate desugaring tool.
This fixes an issue with lateinit properties where the metadata from
the original field was not copied to the nullable field in
LateinitLowering. Also consolidated related tests.
Before this commit, FIR2IR expected that FIR will have no receiver
in this case. However, now FIR *has* receiver, so we have to convert
it to object correctly
The issue was, that built IR function does not have a PSI element,
which is required to report error on suspend functions inside monitors.
In this case, use PSI element of the class, containing the function,
which is consistent with old BE.
It is easier to introduce a new lowering so the codegen will emit code for the old
tail-call optimizer to understand. Also, this is more flexible and would allow to
optimize cases, which are now feasible with the old optimizer.
Note, that because of bytecode inlining, we cannot replace the old one, but we cannot
emit code, that is simpler for it to optimize.
Since the markers replace ALOAD 0 as continuations, passed to suspend calls, in
JVM_IR we do not need this, since in JVM_IR all inline lambdas are static
functions.
Kristoffer Andersen