Ryan C. Gordon
Ryan C. Gordon
> This is pretty clear that we shouldn't do this by default and that SDL_SHUTDOWN_DBUS_ON_QUIT should only be set by people trying to do memory leak detection. Ugh, okay. :/
For the XrmInititalize leak, I'm thinking we should try other approaches first (GDK's Xsettings key, if not the GDK_SCALE environment variable), and try Xft.dpi as a last resort, so we...
It looks like Sam made good progress here, but how do we deal with the leaks out of our control? I'd like to find a solution, both so they don't...
We definitely don't want to _hide_ app leaks, or leaks in SDL itself. I'll try to reproduce it today.
Should be lowercase m, imho.
I'll have to check when I'm not on a phone, but I _think_ this was intended to deal with platforms that didn't offer recursive mutexes--which Mac OS X didn't at...
I don’t know if I’ll add this yet, but I’ve moved this to SDL_net and reopened it for now.
Passing on this.
> is there a way to get the USB VID/PID of the audio device, or can this be done by name? It...might be difficult to get this information, in most...
SDL_AUDIODRIVER=pulseaudio fixed it here, too. The culprit appears to be PipeWire, but it's not clear to me if this is a bug in sdl2-compat, SDL3, the Steam Runtime, or PipeWire...