Sam Lantinga
Sam Lantinga
It looks like a stack issue again, which is strange since the code that was recursing before no longer exists. Any chance you can hook it up to MSVC torturethread...
Yes, probably, although the timer init should probably be made atomic. Let me see what I can do here.
I fixed the timer race condition in b378bc5c8. I'll do a more complete change for SDL3.
Okay, I have PRs for main and SDL2 that should take care of this. Can you take a look?
So I _think_ this is finally fixed now. :)
> > Just hit this in my fork in SDL2 branch: https://github.com/sezero/SDL/actions/runs/9910933216/job/27382477447 > > That looks the same as [my SDL2 crash](https://github.com/madebr/SDL/actions/runs/9899075063) I'm going to add sdlprocdump to SDL2 git...
Yeah, I'm not sure what's happening here. This appears to be happening in the setup before the first SDL call in main, and it's saying the stack isn't writable. Can...
Ohhhhhh, good idea. Maybe we just don't have enough stack space?