OSDev.org

Hi,

version 1.02 of dex-os is available at http://sourceforge.net/projects/dex-os/.

I have been working on the project for quite some time but it is only now that I have released it to the public. Currently it features a 32-bit multitasking kernel with some extensibility features like a dynamic scheduler and a replaceable kernel malloc function. It has a different device driver model than Unix though.

Currently there is support for IDE/ATA/ATAPI CDROM drives and has built-in drivers for FAT12/16/32 and ISO9660/Joliet(buggy) filesystems. There is read/write support for all FAT systems but write support for hard disks is disabled for obvious reasons (but you can enable it if you're up to it). There is also an experimental devfs filesystem for emulating Unix style devices. You can mount devices using the mount command (ex. mount fat hd1p1 /my_hd_dir or
mount cdfs cdp1 /mycd) for a list of devices detected in your system, use the lsdev command.

You should try both the 1.02 image and the release 1 images since they have different applications installed. The 1.02 image contains a C-compiler and a really bad text editor, you can already write simple C programs under it.

For other commands to try, type help at the command prompt.

Thank you very much.

i'm testing on a toshiba laptop (HDD+CDRom, 48MB, Pentium 1 processor)
The boot process went quite fine, and the shell looks to work fine too.
I started DexEdit, but it apparently crashed with

"DexEdit - untitled.txt col:531952512 row:-1
075848788dex32_kernel: user mode page fault.
recovery mode active. system may become unstable.
shutting down application..
/ramdisk/ %"

i also tried to mount hdp0p0 /windoze (reported as "Windows95 FAT32 on partition 0 at hdp0; hdp0 is reported as ATA p master HDD (READ-ONLY), but it says "mount not successful"
i tried to mkdir /windoze first, but that doesn't look to work ...

oh. ok, my fault. forgot the "fat" argument ...

i tried ed.exe /windoze/netlog.txt, but it was obviously a poor idea (still in the same session).

process ID : 1279461901
=== memory access info ===
invalid access : 0x11738a
page directory entry= 0xffffffff
page table entry= 0x7580a15f
addr of faulting instr=0x1034cd
kernel PD: 0x301f000 process PD: 0x646e6957 (looks like a string, imho)
=== registers >===
a=0xf0458c58 b=affff9f0 c=282 d=13d470 bp=9000fb67 di=0 si=0
cs:38 ds=93 es=93 ss=18 fs=93 gs=93

=== api ===
process top of heap: 2073776f
kernel top of heap c039d000
last system call (1) 204e4f49 (2-last) 0x394e4957
context device #7 (def_sched) function #8 (scheduler())
fault occured during system call
press anykey ...

okay, where's that "any" key again ....
btw, the ED.EXE took a *long* time to load (about 30 seconds)

oops. i shouldn't have hit the spacebar instead of "any": the computer rebooted :-/

I have DEX in my claws and fed my Laptop with it.

It boots up, displays looots of things in a fine manner (cool, all this cpu detection stuff ), loads the ramdisk, for which it takes some time, as Pype has stated. Then it drops me to a shell:

Welcome to DEX-OS 1.0
Type help on the command prompt to
display help

press any key to continue or q to quit.

/ %'

Keys are taken, but the mapping is completely screwed up. It dislays f. ex. e with accent circumflex when I am pressing 'i'. It displays both scancodes: Down and Up. Nothing happens if I press enter.

If it is of any help: the Laptop is a toshiba satellite 1700-300 with intel celeron 650 mhz, 128 mb ram, graphics adapter: ati rage mobility 128 (no VGA Bios!!) and the other usual laptop mumbo jumbo.

hth.

've found out that it is indeed the keymap. could it be by any chance that you are using an azerty or dvorak keymap?

I have not only misplaced keys, but f. ex. enter on the key 'a'.

It's a german keyboard I am using.

beyond infinity wrote: 've found out that it is indeed the keymap. could it be by any chance that you are using an azerty or dvorak keymap?

I have not only misplaced keys, but f. ex. enter on the key 'a'.

The enter key is on the same place on /ALL/ keyboards. At least, all PS/2 keyboards. No matter whether they're azerty, qwertz, the standard qwerty or dvorak for that matter. The only thing I can think of is the keymap setting that he tries to modify (to AT or something like that, set 2 or 3), and which your keyboard cannot change (so it sticks to 1). Also, the other way around might happen, but I doubt that more.

Candy *tztztz*

I *know* this. I am not a moron *grmpf*.

[me] is just trying to express something which is a bit hard to express in a foreign language.

The more, you might have noticed, that it is dependant of the keymap you provide, at which key the scancode for ENTER is recognized as *enter* - i.e. sthg like \n\r or so ... You can easily misplace it in the keymap.

beyond infinity wrote: Candy *tztztz*

I *know* this. I am not a moron *grmpf*.

Sorry, didn't intend to be offensive.

[me] is just trying to express something which is a bit hard to express in a foreign language.

I'm kind of asleep right now, I'm somehow incapable of seeing what foreign language you mean? Or do you mean English?

The more, you might have noticed, that it is dependant of the keymap you provide, at which key the scancode for ENTER is recognized as *enter* - i.e. sthg like \n\r or so ... You can easily misplace it in the keymap.

That is true. However, seeing other people working just fine with that map (Pype), and seeing you working NOT fine with that map, I'm searching it in the hardware corner. I'm assuming the author didn't include a heuristic to determine what keymap to use.

My opinion:
Well, when running in bochs, it just loads and loads, and then I get "mounting boot device" and then it just does nothing
I waited for 10 minutes, and nothing else appeared

Hi everyone,

Dennis, what is your hardware spec? I think it could possibly have something to do with your floppy drive hardware... well I'm not sure...

As for the keyboard mapping problem, dex-os is set to use set 1...seems there are some keyboards that don't support it, I'll send out a fix real soon (support for set 3) in version 1.03.

Thanks everyone.

jedld wrote: Hi everyone,

Dennis, what is your hardware spec? I think it could possibly have something to do with your floppy drive hardware... well I'm not sure...

As for the keyboard mapping problem, dex-os is set to use set 1...seems there are some keyboards that don't support it, I'll send out a fix real soon (support for set 3) in version 1.03.

Thanks everyone.

It's in Bochs, as already mentioned

Dennis, I was able to replicate your problem in bochs.
Are you using bochs-2.1.pre1 or earlier? If so, then you may have to try it in bochs-2.1.1 (I am using it and it works). I don't know if it is a bug in bochs or my OS, in any case I will investigate and fix it up as soon as possible. Thank you very much.

Bochs x86 Pentium+ Emulator
Updated: : Sun Feb 8 17:22:07 CET 2004
Version: 2.1.1

So, I use version 2.1.1
Anyway, I would like to use your OS, to see what it does