Page 1 of 6

windows research kernel

Posted: Mon Mar 23, 2009 9:17 am
by RedDemon1970
Recently I found this "windows research kernel", I found also this http://www.microsoft.com/resources/shar ... ectoz.mspx

well is this suitable to build something with? Becouse when I search on the microsoft forums about it I don't see much interesting only a few unanswerded posts.

so can I build with some tools a basic os or is this not done?

Re: windows research kernel

Posted: Mon Mar 23, 2009 11:08 am
by Brynet-Inc
RedDemon1970 wrote:Recently I found this "windows research kernel", I found also this http://www.microsoft.com/resources/shar ... ectoz.mspx

well is this suitable to build something with? Becouse when I search on the microsoft forums about it I don't see much interesting only a few unanswerded posts.

so can I build with some tools a basic os or is this not done?
No you cannot base an operating system off of Microsoft's research kernel.. as far as I know, they release segments of the code to colleges and universities for academic study only.

It isn't open source, it's not freely redistributable.. and modifying it also out of the question.

If you like Windows for some unexplainable reason, join the ReactOS project..

Re: windows research kernel

Posted: Mon Mar 23, 2009 2:43 pm
by RedDemon1970
well i prefer to work with windows, becouse i feel confortabel with it. i don't have that with linux
i find it strange that microsoft shares a krenel and stuff for study if you can't use it for something...sounds more to me like "here is some code, look to it in your free time and shut up" or have i overlooked something?

Re: windows research kernel

Posted: Mon Mar 23, 2009 7:06 pm
by Troy Martin
RedDemon1970 wrote: ... or have i overlooked something?
Yes, you have. There is not enough code for an entire kernel, and what's missing could not be easily pieced together.

Re: windows research kernel

Posted: Tue Mar 24, 2009 1:51 am
by xyzzy
Troy Martin wrote:
RedDemon1970 wrote: ... or have i overlooked something?
Yes, you have. There is not enough code for an entire kernel, and what's missing could not be easily pieced together.
Yup. The WRK source code is not complete and comes with a bunch of precompiled objects for the parts they haven't included. AFAIK the HAL code and a lot of the architecture-dependent code is provided precompiled.

Re: windows research kernel

Posted: Tue Mar 24, 2009 4:21 am
by RedDemon1970
so the only reason that they made it public is to show how it's build up, but nothing more then that
btw maybe a not on topic question but when i have for example the barebones kernel in pascal of the wiki, can i build further on that?
I find the kernel in pascal easier to read then the one in C.

anywhay....i'm i wright if i say as conclusion that if you want to work with kernels that windows is only good as os to run programs?

Re: windows research kernel

Posted: Tue Mar 24, 2009 10:49 pm
by pcmattman
i'm i wright if i say as conclusion that if you want to work with kernels that windows is only good as os to run programs?
No.

EDIT: For a better answer, I need to understand exactly what you're asking, without having to guess (which I've done above).

Re: windows research kernel

Posted: Wed Mar 25, 2009 4:19 pm
by Love4Boobies
Actually, the HAL code is there. It is for the most part a working NT-OS/2 kernel (I think from the time they made the leap from OS/2). It's not the latest NT kernel but it gives you some insight on the architecture. And don't be surprised, Microsoft needs to hire developers that are familiar with the NT kernel so they need to train them somehow, right?

If you want to know what the latest NT kernel architecture looks like, get the Windows Internals 5th ed.

Re: windows research kernel

Posted: Thu Mar 26, 2009 1:26 am
by xyzzy
From what I can tell, the WRK is fairly recent - Windows Server 2003 SP1. See http://www.dcl.hpi.uni-potsdam.de/resea ... page_id=51

Re: windows research kernel

Posted: Thu Mar 26, 2009 5:54 am
by Love4Boobies
It is. Still, it's not the NT kernel :)

Re: windows research kernel

Posted: Thu Mar 26, 2009 8:57 am
by Colonel Kernel
Love4Boobies wrote:It is. Still, it's not the NT kernel :)
Why not? There hasn't been a version not based on the NT kernel since Windows ME (excluding WinCE).

Re: windows research kernel

Posted: Thu Mar 26, 2009 9:05 am
by Love4Boobies
Probably because they're not willing to risk letting the actual source code out of MS. Remember that part of the Windows 2000 source code got out there? Guess they weren't happy about it :) Anyway, I'm not sure if it's an early NT or late OS/2.

Re: windows research kernel

Posted: Thu Mar 26, 2009 3:36 pm
by dosfan
I have access to the WRK sources simply through being a member of the MSDNAA at work. I'm honestly surprised it hasn't leaked all over the place.

To be honest, I've never wanted to poison my mind by even looking at it!

PLEASE, NO PMs ASKING FOR IT. I WILL NOT DISTRIBUTE IT TO ANYONE. I DO NOT HAVE IT IN MY POSSESSION.

Re: windows research kernel

Posted: Thu Mar 26, 2009 10:43 pm
by Love4Boobies
It has leaked. "Poison" your mind? Man, you're lame :)

Re: windows research kernel

Posted: Fri Mar 27, 2009 12:28 am
by ThymeCypher
ProjectOZ is an operating systems project environment that uses the native kernel interfaces of Windows to provide simple, clean, user-mode abstractions of the CPU, MMU, trap mechanism, and physical memory that can be used to perform experiments in operating systems principles. ProjectOZ comes with a basic OS (BasicOZ) built on top of the SPACE abstractions.

Essentially, it sounds to me like it is nothing more than a simple NT 1.0 kernel with a full HAL. The purpose is isn't to rebuild windows or understand NT, rather to try new things (As my dumb @$$ has been doing since I found this site) with developing operating systems, and possibly stumble upon a black hole of speed and power never before tapped by an OS. That or find the hidden choc-o-chip cookie found in every processor through the HAL. Either way, it's probably the source for NT 1.0 with the HAL partially compiled.