#MT
1 messages · Page 1 of 1 (latest)
And now I'm working on getting the GICv3 working, since it's the interrupt controller that the virt-5.0 board uses.
creative name
Mutant NT
Didn't have any better name
Didn't want to waste my time coming up with a name either
Lol
mutant new technology
Are you assuming because I haven't pushed any commits yet on the GitHub?
Interesting project! 🙂
Thanks 😅
Lately I've involved a lot in school and I can't wait to resume development.
Here's my plan, I'll try to make my new project POSIX-like, and once it's finished to the point where it can run POSIX userland, I'll attempt to refactor the kernel so that it works similarly, if not the same as NT, with a POSIX subsystem.
stop bullying for fuck sake
also this project is cancelled
mean
i know
cuz @sick kestrel deemed riscv too difficult
it's not a good NT replacement
it's a great NT replacement
It's features include:
- UART
It's more secure than Microsoft NT

its doesn't have exploits
and never have because its abandoned 🥺
what
sry i just badly know english
ever
🥸
for the record PE has very little to do with the concept of an NT clone
you can easily have an NT clone using any arbitrary executable format including ELF
Well
yes
It is completely possible for an NT clone to be ELF
just not traditional
uh, not really
there is no such thing as "traditional" when it comes to NT
there is only one reference implementation which is proprietary
you can copy the design and make a clone like a few have done, but that's not tradition
Oh
it's not like saying that ELF is traditionally used amongst unix-likes
Well, it is commonly used
however I believe it would still be totally possible for a UNIX-like kernel to run PE executables
It would be cool 🤔
Oh
ReactOS crying in the corner
i wouldn't call ReactOS a "reference implementation" of NT
many actual NT experts have called it shite
why and who?
some actual nt researchers did contribute
some security researchers use it as a reference
well, according to reactos chat
they say it's not how NT is meant to work and their priority isn't to make an NT clone but to get win32 compat working
i don't remember exactly who but i am pretty certain hyenasky was in that bunch
but if he was, he's certainly not the only person that said that
now, it could be that i have been misled and it's actually a perfectly reasonable NT clone
i don't have enough NT expertise to tell
As far as I understand its main goal is compatibility with W32
So the kernel is a little neglected
It's kind of like SerenityOS except it's trying to recreate something
so if u will run the nativeshell for example, i think this will mean what u created the nt clone
(sry for my bad english)
Oh
newer project
wrote a basic EFI header, might make it C instead of C++ later
I just chose to use C++ because of it's namespace features
Not found
if that's all you use c++ for I'd rather use C
Lemme try to bring it back up
I can build the BOOTX64.EFI file on Windows but I can't make a disc image for it
fuck it
im not going to add any support for building on Windows
Alright, here's how the custom EFI library is going on, now I'll start actually implementing the bootloader stuff, as well as migrating from C++ to C because I dislike C++ (only reason why I chose to use C++ initially is for it's namespace features)
Looks like I forgot to use \r
I believe I have the EFI header advanced enough to start parsing PE executables and loading them into memory]
So I'll get started with that now
Was able to build a very basic kernel (ntoskrnl.exe), and I finished a function that loads the file into memory.
All I have to do now is parse the PE executable.
No, not the official NT kernel
This is a custom kernel I wrote which just hangs the system
but will it launch the official kernel too?