Commit Graph

13 Commits

Author SHA1 Message Date
pyos d66be3f82d FIR CFG: do not assume all arguments to inline funs are called in place
Only values for non-noinline, non-crossinline, functional type
parameters qualify.
2022-12-08 10:19:32 +00:00
pyos c4c05f5248 FIR CFG: remove ordering from control flow through in-place lambdas
Old graph:

  arg -> lambda enter -> ... -> lambda exit -> lambda enter -> ... ->
   -> lambda exit -> call

New graph:

  arg -+-> lambda enter -> ... -> lambda exit -+-> call
       \-> lambda enter -> ... -> lambda exit -/
2022-12-08 10:19:31 +00:00
pyos a9be27e330 FIR CFG: add union nodes
Quick quiz:

 Q: In a CFG, what does `a -> b -> c -> d` mean?
 A: `a`, then `b`, then `c`, then `d`.

 Q: In a CFG, what does `a -> b -> d; a -> c -> d` mean?
 A: `a`, then `b` or `c`, then `d`.

 Q: So how do you encode "a, then (b, then c) or (c, then b), then d`?
 A: You can't.

Problem is, you need to, because that's what `a; run2({ b }, { c }); d`
does when `run2` has a contract that it calls both its lambda arguments
in-place: `shuffle(listOf(block1, block2)).forEach { it() }` is a
perfectly valid implementation for it, as little sense as that makes.

So that's what union nodes solve. When a node implements
`UnionNodeMarker`, its inputs are interpreted as "all visited in some
order" instead of the normal "one of the inputs is visited".

Currently this is used for data flow. It *should* also be used for
control flow, but it isn't. But it should be. But that's not so easy.

BTW, `try` exit is NOT a union node; although lambdas in one branch can
be completed according to types' of lambdas in another, data does not
flow between the branches anyway (since we don't know how much of the
`try` executed before jumping into `catch`, and `catch`es are mutually
exclusive) so a `try` expression is more like `when` than a function
call with called-in-place-exactly-once arguments. The fact that
`exitTryExpression` used `processUnionOfArguments` in a weird way
should've hinted at that, but now we know for certain.
2022-12-08 10:19:29 +00:00
pyos 5a2ec4a0d5 FIR CFG: merge data flow if called-in-place lambda may not be called 2022-07-11 18:11:30 +03:00
Andrey Zinovyev a6984c5198 [FIR] Add NO_RETURN_IN_FUNCTION_WITH_BLOCK_BODY diagnostic 2021-07-19 13:40:28 +03:00
Dmitriy Novozhilov 4225813d79 [FIR] Update CFG dumps according to changed order of visiting class children 2021-06-29 21:03:29 +03:00
Ilya Kirillov 0cf00d0f72 FIR: fix FirDefaultPropertyAccessor phase to BODY_RESOLVE 2021-04-15 15:23:56 +03:00
Dmitriy Novozhilov 5ebd24eac5 [FIR] Save inline status of lambda after resolution 2021-04-06 12:30:34 +03:00
Denis Zharkov 65119adb6a FIR: Adjust test data. FakeOverride -> SubssitutionOverride 2020-11-06 11:32:39 +03:00
Dmitriy Novozhilov f794ced888 [FIR] Fix incorrect cluster creating in CFG dumps 2020-10-12 11:55:05 +03:00
Oleg Ivanov cc9c5b9e3c [FIR] Add CFG nodes, add multiple subGraphs for CFGOwner 2020-08-11 16:17:01 +03:00
Dmitriy Novozhilov 58af8d68a1 [FIR] Check for isFun flag in SAM resolution 2020-07-08 12:13:32 +03:00
Dmitriy Novozhilov 29849b1330 [FIR] Resolve rhs of += in dependent context 2020-07-08 12:13:31 +03:00