See
https://youtrack.jetbrains.com/issue/KT-19251https://github.com/puniverse/quasar/issues/280https://bugs.openjdk.java.net/browse/JDK-8046233
Inline function calls (as well as try/catch expressions) in constructor
arguments produce bytecode that spills stack, and stores uninitialized
objects (created by 'NEW C', but not initialized by 'C.<init>') to
local variables. Such bytecode is valid according to the JVM spec, but
confuses Quasar (and other bytecode postprocessing tools),
and fails to verify under some (buggy) versions of JDK 8.
In order to avoid that, we apply 'processUnitializedStores' already
implemented for coroutines. It moves 'NEW' instructions after the
constructor arguments evaluation, producing code like
<initialize class C using Class.forName>
<evaluate constructor arguments>
<store constructor arguments to variables>
NEW C
DUP
<load constructor arguments from variables>
INVOKESPECIAL C.<init>(...)
NB some other expressions, such as break/continue in the constructor
arguments, also can produce "weird" bytecode: object is created by a
'NEW C' instruction, but later (conditionally) POPped from stack and
left uninitialized. This, as we know, also can screw bytecode
postprocessing. However, it looks like we can get away with it ATM.
Otherwise it looks like we'd have to analyze constructor arguments, see
if the evaluation can "jump out", and perform argument linearization in
codegen.
Similar to enum entry initialization, when we have a companion object
in an interface, its constructor (or clinit) initializes its state
before the instance field in corresponding interface is initialized.
So, interface companion object must be accessed via a captured object
reference (#0, or #0.this$0 for inner anonymous objects).
Synthesized 'copy' introduces default values for parameters, which is
prohibited for regular overrides.
Report warning in language version 1.2-, error in 1.3+.
The main changes are in jvm_package_table.proto and ModuleMapping.kt.
With JvmPackageName, package parts can now have a JVM package name that
differs from their Kotlin name. So, in addition to the old package parts
which were stored as short names + short name of multifile facade (we
can't change this because of compatibility with old compilers), we now
store separately those package parts, which have a different JVM package
name. The format is optimized to avoid storing any package name more
than once as a string.
Another notable change is in KotlinCliJavaFileManagerImpl, where we now
load .kotlin_module files when determining whether or not a package
exists. Before this change, no PsiPackage (and thus, no JavaPackage and
eventually, no LazyJavaPackageFragment) was created unless there was at
least one file in the corresponding directory. Now we also create
packages if they are "mapped" to other JVM packages, i.e. if all package
parts in them have been annotated with JvmPackageName.
Most of the other changes are refactorings to allow internal names of
package parts/multifile classes where previously there were only short
names.
This annotation is currently internal because we only commit to its
support for our own libraries. It will be used to change JVM package
names of declarations in JDK-specific stdlib additions (now called
kotlin-stdlib-jre7/8), both to preserve source compatibility of the old
Kotlin code and to solve the split package problem (KT-19258)
Before this change, we were computing the visibility of an inherited
private property setter, and ISE at AsmUtil.getVisibilityAccessFlag
happened ("invisible_fake is not a valid visibility in backend")
Enum entries are "special" kind of singletons that should be
referenced as a captured 'this' instance inside during entry
initialization, because corresponding static fields in enum class
are not initialized yet.
#KT-7257 Fixed
Use it for char boxing/unboxing and unit materialization.
Possible to use for other purposes, for example, to add type checks
to dynamics.
See KT-18793, KT-17915, KT-19081, KT-18216, KT-12970, KT-17014,
KT-13932, KT-13930
In an inner class of the enum entry class, enum entry reference should
be generated as an outer 'this', not as a enum entry access, because
enum entry itself may be not initialized yet.
In Kotlin 1.1 and before, there were no nullability assertions on
extension receivers, because receiver is resolved with NO_EXPECTED_TYPE.
So, if an expression of platform type is passed as an extension receiver
to a non-private function, it would fail with IllegalArgumentException.
However, if the function is private, then we generated no parameter
assertions under assumption that such function can be called from Kotlin
only, and all arguments are checked on the call site. Thus 'null' could
propagate indefinitely.
In Kotlin 1.2, we do the following:
- Generate nullability assertions for expression receivers.
NB nullability assertions are stored for ReceiverValue instances, not
for expressions: given expression can act as receiver in different
calls, each with an expected receiver type of its own.
- Generate nullability assertions for extension receivers of private
operator functions.
NB it still can throw NPE for some particular "optimized" cases, but at
least those nulls would not propagate indefinitely.
This behavior is disabled by an "advanced" command-line option
'-Xno-receiver-assertions'.
In the case the single parameter of override has `Integer` type instead
of `int` type (while in common case it would be just `int`)
See the comment inside forceSingleValueParameterBoxing for clarification
#KT-19892 Fixed
Lateinit local vars are guaranteed to be non-null after store.
So we mark such stores as storing non-null value
(could be useful for some other constructs, too),
and optimize null checks accordingly.
It's necessary for generic inline suspend as a codegen
for it uses binding slice SUSPEND_FUNCTION_TO_JVM_VIEW
to generate fake continuation parameter, so all the
descriptors that are used for body generation must be
obtained from the SUSPEND_FUNCTION_TO_JVM_VIEW
#KT-19528 Fixed
Dmitry Petrov