• Welcome to OS2World OLD-STATIC-BACKUP Forum.
 

News:

This is an old OS2World backup forum for reference only. IT IS READ ONLY!!!

If you need help with OS/2 - eComStation visit http://www.os2world.com/forum

Main Menu

OS2 & eCS kernel

Started by miturbide, 2007.05.16, 22:38:54

Previous topic - Next topic

miturbide

Supporting thread for the OS2 & eCS kernel bounty. The original and old bounty thread can be found here.


Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

saborion2

Are there any updated information available on the Bounty that was posted for a "New" OS/2 & eCS Kernel?

Thanks.

SAB.

RobertM

SAB,

This bounty post OS2 & eCS kernel describes the general goal of the bounty.

This thread is to discuss specifically the less general goals that should be part of the successful completion of the bounty (ie: not drivers, not PM, not WPS, etc - except as related to how they interact with the kernel).

Thus, based off the general goals in the bounty, here is where the discussion should take place as to what needs to be satisfied to complete/satisfy the bounty.

So, for instance (to start off this thread):


  • Do we want a platform independent setup, similar in that aspect to the Mach kernel? (or is our goal currently Intel only?)
  • How do we wish to handle 32bit compatibility, and should we be handling that in the kernel? Should we have a virtualized 32 bit API? Or should we use something that remaps the APIs in a similar fashion that Odin does?

Robert


|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


Blonde Guy

I am contributing to this bounty in partnership with Craig Colby.

I hope kernel authors will claim this bounty when a suitable kernel is ready.

I envision a kernel able to access modern computer features that can still offer a 32-bit OS/2 environment, but with specific extensions for 64-bit and beyond.
Expert Consulting for OS/2 and eComStation

demetrioussharpe

Does this kernel have to be written from scratch or can this be a kernel that's been modified to replace the OS/2 kernel?
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

miturbide

Hi demetrioussharpe

What do you mean by "modified" ?

According to the bounty rules it has to be something that can be legally distributed and that can have an Open Source approved license. The bounty sponsors can also give some conditions on this forum.

Martin
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

demetrioussharpe

Well, there're so many opensource kernels out there. In theory, any kernel would work, as long as you have a booting system, the kernel, & doscall(1).dll; all done in LX format. I have an associate who was trying to modify FreeBSD into an OS/2 replacement. Later, he changed the initial base system to DragonflyBSD. I think that he eventually gave up. It's very hard to find some of the necessary info to complete such a task. But, with a lot of effort, this could work. I'd try to tackle this myself, but I'm busy working on GL/2.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

cytan

There's something called osfree at
      http://www.osfree.org

who are trying to create a new os/2 clone using the L4 microkernel if you're interested. There's lots of activity by OS/2 standards  :) . The last SVN repository change was a few days ago on the 19th of Nov. They are at version 0.05 and trying to get the file system API sorted out.



demetrioussharpe

If I recall correctly, their target is OS/2 PPC - the product that never made it to the masses. I'm more concerned about Merlin & Aurora. I think one of the BSD kernels would be a perfect fit.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

miturbide

what do you mean by "their target is OS/2 PPC". They are actually working on making a OS/2 clone that runs on x86. Check the project. It is worthy to keep an eye on it. OSfreeldr had made nice progress booting from USB and CD-Rom.
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

demetrioussharpe

Yeh, I keep a constant watch on that project. They're using the software architecture of OS/2 PowerPC. It's a good design, but I'm not ready to see the Merlin/Aurora architecture go away yet. It's far too easy to get microkernel OS's wrong & too easy to get monolithic kernels right. IBM threw money & developers at a fundamentally sound architecture, but in the end, it just didn't work out. I'm not downing osfree's work, I think that they should keep going. But, even if they succeed, there's still the allure of being able to replace a kernel & a hand full of dll files, verses having to replace your whole OS/2 setup. So, regardless of other projects, is it acceptable for someone to plug in a replacement kernel & a few replacement dll's to close this bounty or at least to consider it a success?
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

cytan

Hi demetrioussharpe,
   First I am not a programmer but I've read enough to be dangerous  :) . It sounds impossible what you are trying to do: replacing the OS/2 kernel with something else without a complete rewrite of the other subsystems. For example, Reactos. And Reactos has been in alpha for so many years and is not even complete yet!

Perhaps I am misinterpreting what you are saying. A longer explanation might be useful for lay people like us.

   

demetrioussharpe

Quote from: cytan on 2010.11.24, 04:06:20
Hi demetrioussharpe,
   First I am not a programmer but I've read enough to be dangerous  :) . It sounds impossible what you are trying to do: replacing the OS/2 kernel with something else without a complete rewrite of the other subsystems. For example, Reactos. And Reactos has been in alpha for so many years and is not even complete yet!

Perhaps I am misinterpreting what you are saying. A longer explanation might be useful for lay people like us.

   


Ok, here's the basics of it:

All programs, whether they're apps or system services, communicate with the kernel through the support DLLs. For OS/2, all system calls will eventually find their way to DLLs like doscall.dll or doscall1.dll. These kinds of DLL are the entrypoint.  They serve to isolate the app from the kernel. As long as these DLLs behave as normal, the kernel can be changed & none of the apps will notice a difference. So, you change the kernel & you create the support DLLs for the apps to communicate with the kernel through.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

lpino

Quote from: demetrioussharpe on 2010.11.24, 06:17:30
Quote from: cytan on 2010.11.24, 04:06:20
Hi demetrioussharpe,
   First I am not a programmer but I've read enough to be dangerous  :) . It sounds impossible what you are trying to do: replacing the OS/2 kernel with something else without a complete rewrite of the other subsystems. For example, Reactos. And Reactos has been in alpha for so many years and is not even complete yet!

Perhaps I am misinterpreting what you are saying. A longer explanation might be useful for lay people like us.

   


Ok, here's the basics of it:

All programs, whether they're apps or system services, communicate with the kernel through the support DLLs. For OS/2, all system calls will eventually find their way to DLLs like doscall.dll or doscall1.dll. These kinds of DLL are the entrypoint.  They serve to isolate the app from the kernel. As long as these DLLs behave as normal, the kernel can be changed & none of the apps will notice a difference. So, you change the kernel & you create the support DLLs for the apps to communicate with the kernel through.

Demetrious, I'm glad you are around and helping to develop new things for this great OS. You seem to know your stuff and I'm in no position to question your intentions. Beware of comments and opinions other people post here.
I still remember a good programmer that created a great multimedia extension and after many critics and misinformed remarks he decided to leave, never to return to this day. His classes are still one of the best examples of why OS/2 is still a superior system.
Anyway, I hope you continue your work on OpenGL and reach your goal. We need all the help we can have I personally I don't think we are in a position to say this is important and this is not. If we want something specific to be done then we should put money into it, otherwise I see no point in the argument

Thanks and good coding

cytan

Quote from: lpino on 2010.11.24, 14:02:15
Quote from: demetrioussharpe on 2010.11.24, 06:17:30
Quote from: cytan on 2010.11.24, 04:06:20
Hi demetrioussharpe,
   First I am not a programmer but I've read enough to be dangerous  :) . It sounds impossible what you are trying to do: replacing the OS/2 kernel with something else without a complete rewrite of the other subsystems. For example, Reactos. And Reactos has been in alpha for so many years and is not even complete yet!

Perhaps I am misinterpreting what you are saying. A longer explanation might be useful for lay people like us.

   


Ok, here's the basics of it:

All programs, whether they're apps or system services, communicate with the kernel through the support DLLs. For OS/2, all system calls will eventually find their way to DLLs like doscall.dll or doscall1.dll. These kinds of DLL are the entrypoint.  They serve to isolate the app from the kernel. As long as these DLLs behave as normal, the kernel can be changed & none of the apps will notice a difference. So, you change the kernel & you create the support DLLs for the apps to communicate with the kernel through.

Demetrious, I'm glad you are around and helping to develop new things for this great OS. You seem to know your stuff and I'm in no position to question your intentions. Beware of comments and opinions other people post here.
I still remember a good programmer that created a great multimedia extension and after many critics and misinformed remarks he decided to leave, never to return to this day. His classes are still one of the best examples of why OS/2 is still a superior system.
Anyway, I hope you continue your work on OpenGL and reach your goal. We need all the help we can have I personally I don't think we are in a position to say this is important and this is not. If we want something specific to be done then we should put money into it, otherwise I see no point in the argument

Thanks and good coding


Hi guys,
   I hope I don't come across as a doubting thomas or overly critical about what demetrious is proposing. My intent was just to get a clarification of what he had in mind, nothing more.