OSDev.org

tjmonk15 wrote:I have been browsing these forums (and the Wiki) for literally years (6+, I'm only 24 ) and have completed the basic osdev checklist (environment set up, booting, running my code, etc. I followed JamesM's tutorials but converted it to c++ w/namespaces.)

Ahh, one of those "follow this scheme and you have written an OS"-guys. FYI, I don't think writing a successful OS can be done by following some OS-for-dummies tutorials. It's true creative work that needs some skill beyond adapting existing code.

tjmonk15 wrote:And I have to say, the recent "appearance" of rdos, and his "My OS is perfect" ideology, is hilarious. Every time I see a comment from him, i read it just to see how off topic, or close minded it is.

Recent? I started with my OS when you still used diapers, and I had no OS-for-dummies tutorials to follow. I didn't even have a 32-bit linker, much less a cross-compiler or C compiler that produced 32-bit code. Linux was not invented yet.

tjmonk15 wrote:rdos: You do not write a general purpose hobby OS. You write a commercial OS targeted to an extremely small (laughably small) subset of computers.

Not quite. RDOS is not a commercial OS, but it is used by commercial applications.

Offtopic Response About Offtopic Response

rdos wrote:Recent? I started with my OS when you still used diapers, and I had no OS-for-dummies tutorials to follow. I didn't even have a 32-bit linker, much less a cross-compiler or C compiler that produced 32-bit code. Linux was not invented yet.

As cool and rad as that is, you do deserve some consideration as being the few who were in your shoes to do that, however.

Antti wrote:rdos posts make this forum more interesting.

I think so as well, But

tjmonk15 wrote: Every time I see a comment from him, i read it just to see how off topic, or close minded it is.

Is exactly what i do (I know i am not the only one), and its Not a good thing.
The flame was probably needed. <- Come on, post it. Post your whining about that line and the rule's and how i need time in jail for posting it. (lol)

Just wanted to clear one thing up.

rdos wrote:

tjmonk15 wrote:I have been browsing these forums (and the Wiki) for literally years (6+, I'm only 24 ) and have completed the basic osdev checklist (environment set up, booting, running my code, etc. I followed JamesM's tutorials but converted it to c++ w/namespaces.)

Ahh, one of those "follow this scheme and you have written an OS"-guys. FYI, I don't think writing a successful OS can be done by following some OS-for-dummies tutorials. It's true creative work that needs some skill beyond adapting existing code.

Not really. That was to show how little I know, and that I am far from an "expert" on the subject, but [insert rest of post here]

- Monk

tjmonk15 wrote:Every time I see a comment from him, i read it just to see how off topic, or close minded it is.

Every time I see a comment from rdos, I'll be mostly reading it to see if I have a legitimate reason to hit "report this post".

Which just happened, once more.

Combuster wrote:

tjmonk15 wrote:Every time I see a comment from him, i read it just to see how off topic, or close minded it is.

Every time I see a comment from rdos, I'll be mostly reading it to see if I have a legitimate reason to hit "report this post".

Which just happened, once more.

Yup - I even then watch the thread for a bit just to see how bad it gets.

So you wrote a system in only assembly language ?
I think at this time, any OS developper had to work with only an assembler and the Intel's manuals.
I was probably not even born at this time. I think you all should first write a whole system with only Intel's manuals, an assembler, and a console text editor (I don't think there were mouses nor GUI's at the time) without any base before criticizing.

AbstractYouShudNow wrote:I think you all should first write a whole system with only Intel's manuals, an assembler, and a console text editor (I don't think there were mouses nor GUI's at the time) without any base before criticizing.

You won't use spoon for steak, it's not a proof of intelligence.

I generally don't like to post in idiotic renamed threads that should not have been created (rather the posts should have been deleted), but I make an exception to answer the below

AbstractYouShudNow wrote:So you wrote a system in only assembly language ?
I think at this time, any OS developper had to work with only an assembler and the Intel's manuals.

Originally, I used MASM, and Microsofts linker that naturally outputed DOS executables. Since the linker assumed the executable was for real-mode, any segment relocations would be faulty, so the implied rule was that no "SEG" operators was allowed. The linker couldn't combine several modules into a kernel either as it would be confined to 64k, which was too little even for an assembly-only project. That was initially why the syscalls and between kernel-modules was written like they were in order to be able to do run-time relocation by patching the code. Since quite some time now, using DOS executables for device-drivers is no longer used (except for the initial "kernel", which must use this format in order to be compatible with loaders. Today, device-drivers are native and can be both 16-bit and 32-bit segmented (but not flat). The new C device-drivers are 32-bit, and so is the PE loader (which is in assembly).

This background also lead me to write an IA32 emulator. As this was before Bochs and Virtual PC, that was the only way to debug the startup-code. The same code was also used to emulate the first page in DOS, as this was required when running multiple DOS processes. I still have a device-driver that can emulate instructions.

AbstractYouShudNow wrote:I think you all should first write a whole system with only Intel's manuals, an assembler, and a console text editor (I don't think there were mouses nor GUI's at the time) without any base before criticizing.

Come on, with an assembler? Those intel manuals also contains the opcodes, so all anybody should need is a hex editor and the manuals!

Go on and I'll follow !