Even though acquire/release pattern guarantees memory visibility across
threads, it doesn't prevents concurrent access to critical section (i.e.
to force-resolve of the corresponding body).
This can lead to multiple resolution passes over one and the same PSI in
IDE, which, in turn, leads to 'rewrite at slice'-exceptions. See
KT-30030 for case description and details.
^KT-30030 Fixed
Move logic regarding expect classes to the only relevant call site at
PackageCodegenImpl, and return the list of members to remove code
duplication at remaining call sites
The root problem is the fact that ConstantExpressionEvaluator returns
null for values such as infinity and NaN loaded from cls psi (see
IDEA-207252). This commit simply reverts a part of 8ab9226805 where we
started to compute default values more often than needed. In
LazyJavaClassMemberScope, we only need to check whether or not there
_is_ a default value, not compute its value.
#KT-29792 Fixed
Previously only `VariableDescriptor` was key of `ControlFlowInfo`.
Now it's generic parameter `K: Any`, so there is a possibility of using
different object's inside CFA
There is two methods added into `AbstractLoadJavaTest`:
- `configureEnvironment(KotlinCoreEnvironment environment)` allow to
configure environment that will be used two compilation of source test
data and loading generated `.class` files (e.g. you can register extensions)
- `getExtraClasspath()` allow to add custom libraries to classpath of
compiler
Similar methods exists in `AbstractDiagnosticsTest`
Since these tests compile Kotlin code against compiled Java code, it's
only reasonable to use the same .class file reading implementation as is
used in production for the same purpose
All bytecode text tests are run with stdlib in the classpath and only
for JVM backend, therefore directives WITH_RUNTIME, TARGET_BACKEND,
IGNORE_BACKEND are not needed
Test data of crossinline_1_2.txt has changed because now we do not
automatically enable 1.2 (with _all_ of its features) when we're
executing a common coroutines test with old coroutines; and so the
NewCapturedReceiverFieldNamingConvention language feature (which is
introduced in 1.3) is now enabled in this test
Dmitry Savvinov