[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Supervisor mode & multitasking
On Thu, 19 Feb 1998, Katherine L Ellis wrote:
> compatibility reason. I 100% agree with konrad.
> This is an option very useful. DO you use old software that will leak
> through the desktop, or crash the system because of that? I do, and
> having singletasking mode will allow me to stay in mint for hours
> without rebooting.
I must disagree with this. Once a system call like this is allowed and
used, we can not call the system MiNT anymore. In the state that it is,
it's not utilizing the bonuses of MiNT. If some piece of software does not
work in a multitasking environment I'd say "Bad, luck, recode it". It
shouldn't be the kernel's problem if a software is badly coded. Once you
enter single-tasking mode you might aswell be in TOS. So OK, maybe you can
then return to MiNT quicker than by rebooting, but I'd call this a kernel
kludge.
> >>I also thought about expanding Pexec() for that - i.e. launch program in
> >>single mode. Comments?
> >
> >It sounds like MiNT goes to MagiC. What are the advantages of starting a
> >program in single mode? Everyone can reboot to SingleTOS if he/she needs a
> >single mode...
> Mint goes to magic, I think in life you gotta be intelligent, MiNT to
> magic is a long way that I don't even want to know about, I don't want
> mint as instable as magic, losing the FAT of my C partition, don't
> want to loose mintos, don't want to loose minix, don't want to loose
> minix partition, etc etc.
But the point is still valid. MagiC in the above argument represents a
more imperfect multitasking environment. A single-task syscall is
a more imperfect kernel solution and frankly reminds me of Windows95
(which is full of methods like this for "compatibility" reasons).
- ---------- = = ---------//--+
| / Kristoffer Lawson | www.fishpool.com
+-> | setok@fishpool.com | - - --+
|-- Fishpool Creations Ltd - / |
+-------- = - - - = --------- /~setok/