OSDev.org

Candy wrote: Don't attribute the method to me, I only heard it from somebody else, probably here even

srg wrote:BTW Dreamsmith, the machines you're talking about are embedded computers and that's the sort of thing that my Uni course is all about. The chip my Uni loves is the PIC16F877 wihich has 14336bytes of program memory, 368 of RAM and 256byte Data EEPROM (a bit small for your OS but the ATtiny15 has 512 words of flash for program and just it's 32 registers for variable storage), surely this is mor ethe sort of thing that what you listed would be using (I am still very much a noob at this though)?

srg wrote: Anyway, would C be enough or would asm be better?

invalidOpcodeHandler:
    push eax
    push ebx

%define .oldEIP      [esp+12]
%define .oldEFLAGS   [esp+8]
%define .oldEAX      [esp+4]
%define .oldEBX      [esp]

    mov eax,.oldEIP
    cmp word [eax],0x0F01       ;INVLPG
    je .gotINVLPG

    ;Test for any other emulated instructions here

.isInvalid:

    ;Start critical error handler here???

    pop ebx
    pop eax
    iretd


.gotINVLPG:
    mov bl,[eax+2]             ;bl = MOD R/M byte
    mov eax,cr3
    mov cr3,eax
    cmp bl,0xC0                ;Is the MOD R/M byte sane?
    jae .isInvalid             ; no, return to invalid opcode handler
    cmp bl,0xBC                ;Is there a 32 bit displacement and SIB?
    je .disp32SIB              ; yes
    cmp bl,0x80                ;Is there a 32 bit displacement?
    jae .disp32                ; yes
    cmp bl,0x7C                ;Is there an 8 bit displacement and SIB?
    je .disp8SIB               ; yes
    cmp bl,0x40                ;Is there an 8 bit displacement?
    jae .disp8SIB              ; yes
    cmp bl,0x3D                ;Is there a 32 bit displacement (with no register)?
    je .disp32                 ; yes
    cmp bl,0x3C                ;Is there a SIB without displacement?
    je .SIB                    ; yes

    pop ebx
    pop eax
    iretd

.disp32SIB:
    add dword .oldEIP,8
    pop ebx
    pop eax
    iretd

.disp32:
    add dword .oldEIP,7
    jmp .do

.disp8SIB:
    add dword .oldEIP,5
    pop ebx
    pop eax
    iretd

.SIB:
    add dword .oldEIP,3
    pop ebx
    pop eax
    iretd

I don't like OS's where the user is expected to compile the kernel before using it for best performance. It makes it hard for normal users (as opposed to programmers, system administrators) to install the OS. Also one of my goals is to have a single floppy disk which can be used to boot any supported computer with the absolute minimum of configuration required (for demonstration and installation purposes).

crc wrote: I get lucky here: my OS is written in assembly and Forth, so most of the drivers and applications are compiled on demand. It would be easy enough for me to add a couple of routines that scanned for known hardware and automatically compiled the drivers at boot time. Or drivers could be compiled/removed as needed.

Oh, and compiliation at boot or runtime won't waste much time in my OS. My compiler can compile a 200k source file in just over 4 seconds. All of my drivers don't even come close to 200k in size.

Now that is quite the sickest idea I've ever heard. It also applies to AtlantisOS as I think about it, but it's still sick. That's the more extreme form of plug&play. The most extreme would be that the OS looks for some driver sources online, downloads them, compiles them, links them in the kernel at runtime, and gives you a note: your hardware is ready to be used.

d*mn, good idea ... Although I think buffering sources on the box is more expensive than buffering compiled drivers, and buffering the compiled ones is too expensive (look at Microsoft).

On what sort of platform, and is it open source (IE: can I look at it without going to court with you)?

Legend wrote: I saw that the Project UDI seem to be dated some years ago...

And how about creating some new form of Project UDI, this time not driven by some companies, instead by some OS developers?