backslashxx

ready to squash I guess. I'll squash after another approving review Not so attention grabbing gcc-14  clang-19-git 

 Heres how it looks if a builder doesnt backport path_umount. might want to make it noticeable or something

> +1 but why don't use latest code ? the backport becomes bloody as path_mounted also is needed the easiest backports are on either 5.9 or 5.10.9 but yeah let's...

> Lol, it seems that non-gki support will be completely removed, I hope won't be applied https://github.com/tiann/KernelSU/tree/drop_nongki lmao, time to start backporting fsopen

@hmtheboy154 lsm hook hack is kill damn, ksu has to hook prctl and setuid on kprobe for LKM

its still doable, the issue is that they have to move the hooks to kprobe (prctl, setuid) not great when theres actual apps that probe prctl nowadays, its just gonna...

nomount and notmpfs can be straight up ported. https://github.com/tiann/KernelSU/commit/697ec7e7c470271c9c106523657856b0f157bd8a nosparse has missing parts. We'll have to skip all image creation and add new logic for handling modules_update. Install seems fine...

> Maybe creating a refresh function after action button was pressed is a good idea. it gets refreshed on module updates/customize.sh triggers but not on action ofcourse this is also...

@rsuntk https://github.com/user-attachments/assets/19cf0079-dd1c-4ad3-be2f-0d8be9df4c69 seems same bug on Apatch maybe thats where I should report this as the change came from them according to #2156

you'll probably come across the same issue again once AKM24 / SAE-EXT-KEY gets enabled on openwrt better to just make a wpa2-only ssid for those devices. actually I am interested...