Please describe your issue in as much detail as possible:
When trying to end a game through the "stop game" button in the library, on RT3, games that use multiple processes (and I think a bootstrapper that starts something else and then dies is I think a prerequisite), then the game doesn't seem to be killed anymore. Instead, another process of the game gets killed, and if it's an unimportant process, the game just recovers (like a Chromium Render process). I'm not sure what exactly happens.
This has only been happening today, and yesterday it was okay. I assume a beta update triggered it.
Wine seems also generally not be hoisted as a child process of steam anymore, the pressure vessel, I think what it's called, is suddenly under Systemd. Steam is always under systemd --user for me, because I use systemd-run, which has never really impacted it before.
Steps for reproducing this issue:
Start a game that has a launcher that dies quickly and isn't the main process anymore (my example: HoyoPlay's HYP launcher. Start launcher.ese, what is left behind is HYP.exe)
Hit the button on steam that quits the game (the one under here)
It doesn't react the first time, often, but another try will usually "succeed"
Normally, all game processes will now be quit, but on RT3, the launcher only fidgets. It seems like one of its render processes is being killed but restarted. Steam won't show the game as launched anymore.
Your system information
Please describe your issue in as much detail as possible:
When trying to end a game through the "stop game" button in the library, on RT3, games that use multiple processes (and I think a bootstrapper that starts something else and then dies is I think a prerequisite), then the game doesn't seem to be killed anymore. Instead, another process of the game gets killed, and if it's an unimportant process, the game just recovers (like a Chromium Render process). I'm not sure what exactly happens.
This has only been happening today, and yesterday it was okay. I assume a beta update triggered it.
Wine seems also generally not be hoisted as a child process of steam anymore, the pressure vessel, I think what it's called, is suddenly under Systemd. Steam is always under systemd --user for me, because I use systemd-run, which has never really impacted it before.
Steps for reproducing this issue:
It doesn't react the first time, often, but another try will usually "succeed"