After an accidental change in 846537b367, we started generating metadata
into anonymous classes for callable references. This metadata contained
the Kotlin representation of the referenced function. In KT-53794, this
led to a problem where Java's protected visibility could not be
represented in Kotlin terms, which crashed the backend.
But also, this metadata is useless because there's no real use case for
interpreting it (since you already have a function reference object at
runtime with all the needed information), so it would take some extra
space in the bytecode.
#KT-53794 Fixed
Ensure that when .entries is accessed from an inline function body
or lambda argument, EntriesMapping are properly generated and used
without excessive mappings and duplicated fields
#KT-53236
Leveraging the same mechanism with $EntriesMapping as Java enums.
Old (compiled with LV/AV < 1.8) enums are detected by looking for
static special <get-entries> method that cannot be introduced on
Kotlin enums otherwise
#KT-53236
instead of bytecode text ones. Check the content of continuation
object instead of bytecode, since this is less prone to changes during
changes in coroutines codegen.
#KT-48678
For FE 1.0, we just change super qualifier symbol to
base class, if it's possible.
For FIR, we introduce using of super qualifier symbols for
field accesses and repeat FE 1.0 behavior here.
#KT-49507 Fixed
E.g. in `x + f()` where `f` is an inline lambda, the instructions for
`+` should have the line number of that expression (while previously
they instead had the line number of the last line of the lambda).
^KT-51738 Fixed
If we do not do this, the state-machine builder will not know the type
of the ACONST_NULL, defaulting to Object, leading to VerifyError.
Alternatively, we could use LVT to deduce the type, but getting types
from LVT is something I got rid of long time ago, and I have no desire
to return it back.
Generating CHECKCAST hints the state-machine builder the type of the
variable avoiding the issue of VerifyError. However, this CHECKCAST
replaces StrictBasicValue.NULL_VALUE with BasicValue in
OptimizationBasicInterpreter. To preserve optimization on not-spilling
known nulls, introduce BasicValues, which represent typed nulls and
create BasicInterpreter, which is aware of them. This way we have the
best of two worlds - we do not spill known nulls, and we know the type
of ACONST_NULL.
#KT-51718 Fixed
* Change 1.6 to 1.7 constants
* Fix SAFE_CALL_WILL_CHANGE_NULLABILITY for testData
* Change EXPOSED_PROPERTY_TYPE_IN_CONSTRUCTOR_WARNING to EXPOSED_PROPERTY_TYPE_IN_CONSTRUCTOR_ERROR
* Change NON_EXHAUSTIVE_WHEN_STATEMENT to NO_ELSE_IN_WHEN
* Fix testData for SafeCallsAreAlwaysNullable
* Change T -> T & Any in test dumps
* Change INVALID_CHARACTERS_NATIVE_WARNING -> INVALID_CHARACTERS_NATIVE_ERROR
* TYPECHECKER_HAS_RUN_INTO_RECURSIVE_PROBLEM_WARNING -> TYPECHECKER_HAS_RUN_INTO_RECURSIVE_PROBLEM_ERROR
Do not generate linenumber for the start of the finally block, because
that is usually where the only word 'finally' is located. Instead,
generate linenumber for the first expression inside the finally block.
Not generating this linenumber fixes an issue in code coverage tools
which would consider such finally uncovered. Although this might be
technically considered as designed, it makes more sense to NOT detect it
as uncovered because semantics of the finally block shouldn't really
differ whether it's executed normally or because an exception happened.
It's also beneficial for the tool support to behave like javac, which
doesn't generate the linenumber either.
#KT-50973 Fixed
This is a partial revert of e857966edb.
Specifically, behavior is restored in the old backend, which will allow
to support language version 1.3, where this language feature was not
enabled yet. There are no changes in the JVM IR backend, because to
enable JVM IR with LV 1.3, you need to pass the compiler argument
`-Xuse-ir` which is not stable and we don't guarantee anything about it.
#KT-50251
We are going to deprecate `WITH_RUNTIME` directive. The main reason
behind this change is that `WITH_STDLIB` directive better describes
its meaning, specifically it will add kotlin stdlib to test's classpath.
In general, we would rather prefer a range-based for loop to look like
a counter loop in Java ('for (i = start; i < end; ++i) { <BODY> }').
This corresponds to
i = start;
do {
if (i >= end) break;
<BODY>
} while ( { ++i; true } )
However, HotSpot doesn't recognize Kotlin unsigned integer comparison
in 'if (i >= end) break;' as a counter loop condition. Thus, the loop
doesn't get optimized as a counter loop, resulting in a performance
regression.
If we use exclusive range-based for loop instead, then we actually use
unsigned integer equality instead of unsigned integer comparison, which
is Ok for HotSpot.
KT-49444
Alexander Udalov