boxInline tests for suspend use StateMachineChecker and share the state
if compiled and run together. The single value accessed from all tests.
It is enough to reset state before the test is run.
`$$forInline` functions do not pass through the state machine generator,
and optimizing `Ref`s before that changes how assignments inside lambdas
passed to `suspendCoroutine`, etc. behave: without a `Ref`, the
assignment is not reflected in the continuation object, so the variable
has old value on resumption.
These functions will be optimized later, after they are inlined
somewhere and the state machine is generated.
^KT-52198 Fixed
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.
They may or may not be inlined later.
IDK how the test passes when both modules are compiled with the old
backend - perhaps this has something to do with the fact that when `f`
is compiled with the IR backend, the call to `x()` is followed by `pop`
and `getstatic kotlin/Unit.INSTANCE`? This is probably why the original
issue in kotlinx.coroutines reports that everything works fine with
kotlinx-coroutines-core:1.4.3.
^KT-46879 Fixed
^KT-48801 Fixed
When a suspend lambda does not capture crossinline lambda, it is
generated with as state-machine, since it does not inline anything.
However, when regenerating, the inliner used to remove all DebugMetadata
annotations to avoid duplication. This lead to missing annotation if
the lambda is regenerated, but state-machine is not regenerated.
This change fixes the blind spot by readding the annotation after
regeneration.
#KT-41789 Fixed
In general, `InliningContext.findAnonymousTransformationInfo` was not
reliable because it mapped each type to *some* info for that type,
preferring ones with `shouldRegenerate == true` if those exist. Thus, it
returned incorrect results if one type was regenerated multiple times,
e.g. in a nested inlining context or because of a `finally` (which
duplicates anonymous objects). The solution is to avoid a global map and
attach the current transformation info directly to the current inlining
context.
Otherwise, the assumption that coroutine codegen makes about every
inlined function already having the markers breaks and it is no longer
true that calls to inline lambdas do not require them.
This is one example of a function replaced by a lowering after
AddContinuationLowering mentioned in the last commit: the lowering that
makes default stubs static is further down. Although this is not a
lambda, calling getOrCreateSuspendFunctionViewIfNeeded on it in the
inliner is fatal all the same.
Private $$forInline versions of public suspend functions should not
be mangled. (Note that there are no $$forInline versions of private
suspend functions, as those are effectively inline-only.)
* unify various checks for whether a suspend function needs a
continuation;
* mark the continuation parameter with an origin (this also allows
correctly computing the offset of the local in codegen -- see the
new test);
* when wrapping `suspend fun main`, use a suspend reference instead of
a synthetic non-suspend lambda (required to correctly implement the
previous point, as previously the continuation parameter was passed
to main() itself correctly only because of very lenient checks in
AddContinuationLowering).
* TailCallOptimizationLowering should go into local classes in order to
transform their suspend methods;
* the check for invokes of noinline lambda arguments in codegen was
incorrect, as it also returned true for calls of lambdas stored in
local variables;
* IrInlineCodegen should mark non-inlinable arguments used as inline
suspend parameters;
* detection of suspend/inline call sites was incorrect (or maybe it's
the `compilationContextDescriptor` that was incorrect?..)
parameters of inline function. Otherwise, it leads to AnalyzerException,
when inlined lambda contains try-catch block. The reason is simple:
in try block, we leave some variables on stack, while on catch block the
stack is empty. Spilling the variables before try block does the trick.
#KT-34708 Fixed
Artem Kobzar