Has anyone choosen a partition type id for SFS yet?

[Kevin]

Yeah, the problem is if you write a bootloader... you know what your goal is and you write it to do it's job as efficient as possible. But this is one creation that tries to be an OS and a bootloader but the latter only on a 386+. To complicate things even futher, theres no reason the 32bit OS couldn't be loaded by grub as a multiboot kernel directly. Am I driving you insane yet?

Actually no. I mean, the concept of such an bootloader/OS hybrid doesn't make a lot of sense to me, but thats okay. The goal doesn't have to make sense, it's all just for fun anyway. What matters is that the way it's done makes sense.

What does a microkernel really mean through the context of PM which has protection mechanisms? It's just a design philosophy

I always considered privilege separation the most important aspect of microkernels.

And having different kernels for ever little difference in architecture or hardware is just silly.

But that can be solved with a (modularised) monolithic kernel as well.

I still don't really know where to draw the line for RM OSes. Would DOS be a hybrid because some drivers are baked into the kernel and others are loaded as TSRs? I image after being loaded they both work exactly the same way, though (mostly just hooking themselves up as interrupt handlers).

[Nutterts]

The goal doesn't have to make sense, it's all just for fun anyway. What matters is that the way it's done makes sense.

Theres still alot of design decisions to make. Feel free to be as critical as you want, whenever you want. I'm not fragile and I appreciate input like that.

I always considered privilege separation the most important aspect of microkernels.

Well it is the most important aspect! It's where microkernels have the opportunity to shine. But it's not like microkernels are useless without it.

And having different kernels for ever little difference in architecture or hardware is just silly.

But that can be solved with a (modularised) monolithic kernel as well.

Yes that could work.. but I basicly want to throw out of memory every bit of code not needed after booting has finished. Conceptionally it's easier for me to wrap my head around it when looking at it as a micro-kernel.

I still don't really know where to draw the line for RM OSes. Would DOS be a hybrid because some drivers are baked into the kernel and others are loaded as TSRs?

Somewhere between monolithic and hybrid I guess. But honestly? I just feel tsr, network redirector and drivers were a horrible kludge to keep users happy.

A good example of how not to write a kernel. So inflexible that it took some ugly hacks to keep adding new features without losing backwards compatibility. The death of dos was sealed with how they designed the network redirector interface. Never a good design to let some interface depend on internal data structures that change between versions.

[Combuster]

Somewhat back to topic - did someone ever generate a EFI GUID for an SFS partition? I suggest the first one to be posted one will be remembered forever

[Nutterts]

Yeah sorry, it's gone way off topic for this part of the forum.

Somewhat back to topic - did someone ever generate a EFI GUID for an SFS partition? I suggest the first one to be posted one will be remembered forever

Haha, well lets make it official.
My sugestion would be.. *drum roll*..

MBR partition ID: 0x1A
GPT GUID: {4ebf0e06-11bf-450c-1a06-534653534653}

Motivation: The MBR partition ID is free acording to the wikipedia page. The guid was generated randomly but I made a few changes. 1a06 is for the partition id and the year it was first designed. (afaik) The last 6 bytes read "SFSSFS".

Altho I'm a great fan of the hitchhikers guide to the galaxy, the id 0x42 is technically allready taken by both linux and windows 2000. I'm sorry Combuster...altho.. it's your call. I won't complain, just doesn't seem right.

[Kazinsal]

If you ask fdisk, it lists SFS (a different one, but hey why not?) under partition IDs. However if you actually use that, an installed Windows on the same machine refuses to boot

Bit late on the draw on this one, but Windows uses 0x42 as a whole-disk container partition for dynamic disks (up to 2000 volumes, software RAID, non-contiguous volumes, disk spanning) and will refuse to boot if it can't find a dynamic volume database on that partition.