electrodeyt

This appears to have sourced a lot of code from Minoca without Attribution.

From OSDev discord: To give an example, there's a striking similarity between these two files: <https://github.com/xt-sys/exectos/blob/master/xtoskrnl/ke/sy...> <https://github.com/xt-sys/exectos/blob/master/xtoskrnl/ke/ru...> and these: <https://github.com/minoca/os/blob/master/kernel/ke/sysres.c> <https://github.com/minoca/os/blob/master/kernel/ke/runlevel....>

Edit: also found this: https://lists.reactos.org/hyperkitty/list/[email protected]...

Took them long enough

I wonder if they (23andMe) can sleep at night after telling innocent people that something was their fault

This feels like an ad; the blog post and the README feel like they were written to sound complicated and smart rather than tell the whole truth. Like, "On boot, the kernel has full access to the memory and is allowed to write where it should not (its own code, for example)"? Apart from the fact that ring 0 will always have full control, the MMU also exists. Also, the "48k" (the kernel has 34324 lines of non-comment code, the rest are in the other repos I assume) LOC are obvious when you consider code like src/syscall/mod.rs:717 is present.

To be fair, this is impressive, but its a basic monolithic kernel written for a school project, with the "twist" that it is in Rust and uses Linux syscalls.

For anyone who is interesting in more Rust UNIX-like kernels, Aero (https://github.com/Andy-Python-Programmer/aero) is farther ahead supports running quite a lot of recompiled Linux userspace, including dwm and WebKit. Managarm, a microkernel I personally contribute to, also supports a lot of the same things, but is in C++ and is also mostly asynchronous.