Hello everybody out there using minix -

I'm doing a (free) operating system (just a hobby, won't be big and
professional like gnu) for 386(486) AT clones.  This has been brewing
since april, and is starting to get ready.  I'd like any feedback on
things people like/dislike in minix, as my OS resembles it somewhat
(same physical layout of the file-system (due to practical reasons)
among other things).

I've currently ported bash(1.08) and gcc(1.40), and things seem to work.
This implies that I'll get something practical within a few months, and
I'd like to know what features most people would want.  Any suggestions
are welcome, but I won't promise I'll implement them :-)

                Linus (torva...@kruuna.helsinki.fi)

PS.  Yes - it's free of any minix code, and it has a multi-threaded fs.
It is NOT protable (uses 386 task switching etc), and it probably never
will support anything other than AT-harddisks, as that's all I have :-(.

In article <1991Aug25.205708.9...@klaava.Helsinki.FI>, torvalds@klaava (Linus Benedict Torvalds) writes:

Tell us more!  Does it need a MMU?

How much of it is in C?  What difficulties will there be in porting?
Nobody will believe you about non-portability ;-), and I for one would
like to port it to my Amiga (Mach needs a MMU and Minix is not free).

As for the features; well, pseudo ttys, BSD sockets, user-mode
filesystems (so I can say cat /dev/tcp/kruuna.helsinki.fi/finger),
window size in the tty structure, system calls capable of supporting
POSIX.1.  Oh, and bsd-style long file names.

//Jyrki

Yes, it needs a MMU (sorry everybody), and it specifically needs a
386/486 MMU (see later).

Simply, I'd say that porting is impossible.  It's mostly in C, but most
people wouldn't call what I write C.  It uses every conceivable feature
of the 386 I could find, as it was also a project to teach me about the
386.  As already mentioned, it uses a MMU, for both paging (not to disk
yet) and segmentation. It's the segmentation that makes it REALLY 386
dependent (every task has a 64Mb segment for code & data - max 64 tasks
in 4Gb. Anybody who needs more than 64Mb/task - tough cookies).

It also uses every feature of gcc I could find, specifically the __asm__
directive, so that I wouldn't need so much assembly language objects.
Some of my "C"-files (specifically mm.c) are almost as much assembler as
C. It would be "interesting" even to port it to another compiler (though
why anybody would want to use anything other than gcc is a mystery).

Unlike minix, I also happen to LIKE interrupts, so interrupts are
handled without trying to hide the reason behind them (I especially like
my hard-disk-driver.  Anybody else make interrupts drive a state-
machine?).  All in all it's a porters nightmare.

Most of these seem possible (the tty structure already has stubs for
window size), except maybe for the user-mode filesystems. As to POSIX,
I'd be delighted to have it, but posix wants money for their papers, so
that's not currently an option. In any case these are things that won't
be supported for some time yet (first I'll make it a simple minix-
lookalike, keyword SIMPLE).

                Linus (torva...@kruuna.helsinki.fi)

PS. To make things really clear - yes I can run gcc on it, and bash, and
most of the gnu [bin/file]utilities, but it's not very debugged, and the
library is really minimal. It doesn't even support floppy-disks yet. It
won't be ready for distribution for a couple of months. Even then it
probably won't be able to do much more than minix, and much less in some
respects. It will be free though (probably under gnu-license or similar).

On a lower level:

I don't like the chmem mechanism of Minix. Processes should start with
a minimal size and grow as they need to until they run out of RAM or
disk space. Paging to disk would be nice, too.

If your OS is message based I would like to have arbitrarily large
messages. They could be implemented efficiently by mapping the pages
into the receivers address space (or just passing a pointer on 68k
systems without MMU). Oh, yes, and the addressing scheme for messages
should be different than in Minix. Messages should not be sent to
process slot numbers, but to ports. That way, user processes can use
messages, too, and it is easier to add your own servers.

PS: I am very interested in this OS. I have already thought of writing
my own OS, but decided I wouldn't have the time to write everything from
scratch. But I guess I could find the time to help raising a baby
OS :-)
--
|    _  | Peter J. Holzer                       | Think of it   |
| |_|_) | Technical University Vienna           | as evolution  |
| | |   | Dept. for Real-Time Systems           | in action!    |
| __/   | h...@vmars.tuwien.ac.at                 |     Tony Rand |

In article <1991Aug26.110602.19...@klaava.Helsinki.FI> torva...@klaava.Helsinki.FI (Linus Benedict Torvalds) writes:

Sure.  For one example, Alessandro Forin's Mach SCSI adapter drivers are
written this way.  A comment from his code:

/*
 * This layer works based on small simple 'scripts' that are installed
 * at the start of the command and drive the chip to completion.
 * The idea comes from the specs of the NCR 53C700 'script' processor.
 *
 * There are various reasons for this, mainly
 * - Performance: identify the common (successful) path, and follow it;
 *   at interrupt time no code is needed to find the current status
 * - Code size: it should be easy to compact common operations
 * - Adaptability: the code skeleton should adapt to different chips without
 *   terrible complications.
 * - Error handling: and it is easy to modify the actions performed
 *   by the scripts to cope with strange but well identified sequences
 *
 */

An interesting way to write a device driver.  I believe this is a very old
technique, too.

Good luck on your OS project, it sounds like a lot of fun.
Jaime
...........................................................................
: domain: j...@cs.umd.edu                                 James da Silva
: path:   uunet!mimsy!jds                   Systems Design & Analysis Group

One of the things that really bugs me about minix is the way device drivers
have to be compiled into the kernel. So, how about doing some sensible
installable device driver code (same goes for minix 2.0 whenever).

--
Adam David.
(ad...@rhi.hi.is)

In article <1991Aug26.110602.19...@klaava.Helsinki.FI> torva...@klaava.Helsinki.FI (Linus Benedict Torvalds) writes:

Is that max 64 64Mb tasks or max 64 tasks no matter what their size?
--
  Alan Barclay
  iT                                |        E-mail : a...@ukpoit.uucp
  Barker Lane                       |        BANG-STYLE : .....!ukc!ukpoit!alan
  CHESTERFIELD S40 1DY              |        VOICE : +44 246 214241

I'm afraid that is 64 tasks max (and one is used as swapper), no matter
how small they should be. Fragmentation is evil - this is how it was
handled. As the current opinion seems to be that 64 Mb is more than
enough, but 64 tasks might be a little crowded, I'll probably change the
limits be easily changed (to 32Mb/128 tasks for example) with just a
recompilation of the kernel. I don't want to be on the machine when
someone is spawning >64 processes, though :-)

                Linus