Isn’t it a memory leak with OOM?

# announcements

[orangy]

Isn’t it a memory leak with OOM?

[groostav]

orangy: whats OOM? Is there something I can do to make kotlinc's life easier?

[orangy]

OutOfMemoryException 🙂

[groostav]

oh lol

[groostav]

but yeah thats what its looking like

[groostav]

the code generated appears to have some kind of nasty complexity between the number of recievers deep it is

[groostav]

if you look at the second chunk of the generated code its a wall of $LOCALVARIABLE asm stuff, far more than I would think necessary to run the relatively small block of source code its generating for

[groostav]

but that could be a misnomer, I've never written a line of jbc by hand and Im certainly no expert

[orangy]

you could minimize it down to some repro and submit an issue, we will look

[groostav]

you dont think the gist i uploaded has enough information?

[groostav]

it reads to me like somebody's written an O(n^2) memory complexity component in code with reciever lambda's

[orangy]

Gist doesn't have an "assignee field" ;)

[orangy]

kotl.in/issue is where team works, so we ask to put requests there.

[groostav]

alright well ill throw what I have up there

[groostav]

I really don't want to wrangle this down to an SSCCE because im pretty sure thats going to involve a lot of aggressive tuning of the JVM and its memory

[groostav]

run it with a max heap of 1MB

[groostav]

kind''ve thing

[orangy]

Post here a link when you're done

[groostav]

thanks @orangy https://youtrack.jetbrains.com/issue/KT-16493

[groostav]

Can I ask somebody at JB to take a cursory look at this for me? Its a real pain in the ass and my only current solution is to drop the relevant file from the build process.

[orangy]

You already asked by submitting an issue. It was assigned to a developer and he will take a look.