search

sustrik / libmill

Go-style concurrency in C
MIT License
2.74k stars 204 forks source link

Context switching points #164

Closed paulofaria closed 7 years ago

paulofaria commented 7 years ago

Hello, Martin.

As you might remember I maintain https://github.com/Zewo/Venice. A Swift library that wraps libmill.

Unfortunately we're having a bit of a leakage problem on macOS. Swift uses Foundation under the covers on Darwin which in turn uses ARC. The problem is that because of libmill context switches the calls to push/pop on the autorelease pool get unbalanced.

To deal with this we basically need a hook for right before any resume of a co-routine and a hook for right after it’s suspended and a guarantee that those hooks are always 1:1.

We tried inspecting the code ourselves, but I'm afraid we couldn't find all the places. Could you help us pinpoint the actual places the context switches occur, or the best place to put those hooks, given we need 1:1?

raedwulf commented 7 years ago

Just to note, I checked CLibVenice and it uses the setjmp context switching method still. @paulofaria - libmill HEAD uses an optimised context switcher on x64.

paulofaria commented 7 years ago

@raedwulf yeah! we we're talking about updating it just now. haha. awesome work by the way. 😊

raedwulf commented 7 years ago

Thanks :) not sure if this helps but setjmp occurs when the coroutine is suspended and longjmp performs the actual switching to another coroutine. When a setjmp occurs again it is this latter coroutine being suspended now. Maybe hooking these functions will catch all cases easily?

raedwulf commented 7 years ago

I'm not familiar with Swift but I think it might be difficult to wrap my optimised implementation and I think the interop would kill any of the benefits gained. But no worries, the implementation with setjmp and longjmp is not going anywhere.

paulofaria commented 7 years ago

I think @antonmes had a try with your context switching. Anton, do you remember if there were performance drops?

paulofaria commented 7 years ago

Or actually, not much gain.

robertjpayne commented 7 years ago

@raedwulf appreciate the tips, using setjmp/longjump nearly works, but theres still somewhere else where execution drops off that we're missing.

antonmes commented 7 years ago

@paulofaria yes, i had the same numbers with this benchmark, but i guess the bottleneck in this case is memory allocation for each coroutine.

paulofaria commented 7 years ago

This is what we have so far:

https://github.com/Zewo/CLibvenice/pull/8/files

sustrik commented 7 years ago

Sorry for the delay.

Some background info: mill_suspend() is called to suspend current coroutine (IIRC it's also called -- without the corresponding resume -- when coroutine finishes. mill_resume() is called to re-schedule a suspended coroutine. It doesn't mean it will run immediately though.

That being said, I would say the best way to hook into libmill's context switching are millsetjmp and milllongjmp macros. They are the only way to switch contexts in libmill, so putting the hooks there should catch all context switches.

paulofaria commented 7 years ago

Thanks! And sorry for the late reply as well!

raedwulf commented 7 years ago

@sustrik and I have been doing some work on libdill's context switching/stack allocation code. I'm not sure if you've tried using it in Venice yet, but it might be worth a try to see if the problems still exist. However some functionality is split between libdill and dsock, so you might not be able to do a full port as yet if it does work.