I have a question

A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?

RTOS's are designed in such a way that the timing of events is highly deterministic

The response time of an RTOS like Vx-works / QNX is close to zero as compared to an General Purpose OS like Windows / Linux (Redhat/Fedora)

: I have a question
:
: A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?
:
:
:

Can you please explain what you mean in more detail? I don't understand what you mean by preemption "at the kernel level". An OS such as Linux and Windows do have preemption so I'm not sure waht you are referring to. I assume you mean the classic definition of preemption referring to the scheduler suspending a process involuntarily and scheduling another process to run. This does exist in RTOSs as well as general purpose OSs so please clarify if I hvae misunderstood.

Till now linux didn't have preemption points...unlike fedora (radhat 10.0).Guess they have included them.


: : I have a question
: :
: : A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?
: :
: :
: :
:
: Can you please explain what you mean in more detail? I don't understand what you mean by preemption "at the kernel level". An OS such as Linux and Windows do have preemption so I'm not sure waht you are referring to. I assume you mean the classic definition of preemption referring to the scheduler suspending a process involuntarily and scheduling another process to run. This does exist in RTOSs as well as general purpose OSs so please clarify if I hvae misunderstood.
:
:

1> This is not as simple as it may sound. Modern general purpose
>operating system kernals are very large, with several million lines of code.
>It can be difficult to trace through them to find all the possible sources of
>delay in response. An RTOS tends to be much smaller than a general purpose OS
>making guarantying the response time more practical. <

this is one of the difference

: I have a question
:
: A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?
:
:
:

All of the systems that I program are considered real time embedded applications. For the DOS based systems I use Datalight's ROM-DOS, but I do not believe it would fall into the preemptive catagory. For those issues that need to be addressed immediately, interupts work just fine. I this what you are referring to?

: : I have a question
: :
: : A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?
: :
: :
: :
:
: All of the systems that I program are considered real time embedded applications. For the DOS based systems I use Datalight's ROM-DOS, but I do not believe it would fall into the preemptive catagory. For those issues that need to be addressed immediately, interupts work just fine. I this what you are referring to?
:


DOS is no RTOS sence no timing is guaranteed. An application using DOS is therefore not a real-time application, even if you wish it was.

Here is a good description of a RTOS:
http://en.wikipedia.org/wiki/Real-time_operating_system

: : : I have a question
: : :
: : : A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?
: : :
: : :
: : :
: :
: : All of the systems that I program are considered real time embedded applications. For the DOS based systems I use Datalight's ROM-DOS, but I do not believe it would fall into the preemptive catagory. For those issues that need to be addressed immediately, interupts work just fine. I this what you are referring to?
: :
:
:
: DOS is no RTOS sence no timing is guaranteed. An application using DOS is therefore not a real-time application, even if you wish it was.
:
: Here is a good description of a RTOS:
: http://en.wikipedia.org/wiki/Real-time_operating_system
:
:

Thanks for correcting me and for the link. From that link I copied the following:

An operation within a larger dynamic system is called a real-time operation if the combined reaction- and operation-time of a task is shorter than the maximum delay that is allowed, in view of circumstances outside the operation.

Which is substantially similar to what I had read before. So if we say that even if DOS is not a RTOS, could we not conclude that it is capable of real time operations? Example:One of the machines I programmed monitors motor load in a seperate thread from the main thread. When the motor load exceeds 115% rated load the computer shuts down operations quickly enough to avoid damage. Isn't this a real time operation?

: : : : I have a question
: : : :
: : : : A normal OS doesnt have preemption at the kernel level whereas the RTOS has a preemptive kernel. For this, the scheduler has to be re-written in such a way that it takes care of pre-emption. Is there any other feature(s) specific to a RTOS?
: : : :
: : : :
: : : :
: : :
: : : All of the systems that I program are considered real time embedded applications. For the DOS based systems I use Datalight's ROM-DOS, but I do not believe it would fall into the preemptive catagory. For those issues that need to be addressed immediately, interupts work just fine. I this what you are referring to?
: : :
: :
: :
: : DOS is no RTOS sence no timing is guaranteed. An application using DOS is therefore not a real-time application, even if you wish it was.
: :
: : Here is a good description of a RTOS:
: : http://en.wikipedia.org/wiki/Real-time_operating_system
: :
: :
:
: Thanks for correcting me and for the link. From that link I copied the following:
:
: An operation within a larger dynamic system is called a real-time operation if the combined reaction- and operation-time of a task is shorter than the maximum delay that is allowed, in view of circumstances outside the operation.
:
: Which is substantially similar to what I had read before. So if we say that even if DOS is not a RTOS, could we not conclude that it is capable of real time operations? Example:One of the machines I programmed monitors motor load in a seperate thread from the main thread. When the motor load exceeds 115% rated load the computer shuts down operations quickly enough to avoid damage. Isn't this a real time operation?
:
:


Real time only means that the system must be guaranteed to respond in a certain time. So if the surrounding system is sluggish and we are talking about seconds or slower in response time, then I don't think DOS will fail to respond. But in the embedded industry such applications are rare though, a milisecond is an eternity on most industrial/stearing/telecom applications.

And I question if DOS is an operative system or just a programmer/user interface. When I want to add an OS to an application, the only demand I have is that it can handle multitasking and scheduling, and do so rather fast. And that is also the most common definition of the mimimum requirement on an OS.

: : Which is substantially similar to what I had read before. So if we say that even if DOS is not a RTOS, could we not conclude that it is capable of real time operations? Example:One of the machines I programmed monitors motor load in a seperate thread from the main thread. When the motor load exceeds 115% rated load the computer shuts down operations quickly enough to avoid damage. Isn't this a real time operation?

dssuresh6,

You must do this in firmware, or can you implement a solution to shut down the motor in a hardware solution?

-tankarray

: : : Which is substantially similar to what I had read before. So if we say that even if DOS is not a RTOS, could we not conclude that it is capable of real time operations? Example:One of the machines I programmed monitors motor load in a seperate thread from the main thread. When the motor load exceeds 115% rated load the computer shuts down operations quickly enough to avoid damage. Isn't this a real time operation?
:
: dssuresh6,
:
: You must do this in firmware, or can you implement a solution to shut down the motor in a hardware solution?
:
: -tankarray
:

In this case there are several layers of safety:

1) inverter speed control has embedded firmware which shuts down drive on overload,

2) Motion control software monitors motor load via analog input and can shut down more quickly as desired,

3) Motors are electrically protected through fuses,

4) Motor shafts have sheer pins or other clutched type design to yield at certain loads.

#3 and #4 are the worst cases and are only present should controls experience complete failure.

: : : : Which is substantially similar to what I had read before. So if we say that even if DOS is not a RTOS, could we not conclude that it is capable of real time operations? Example:One of the machines I programmed monitors motor load in a seperate thread from the main thread. When the motor load exceeds 115% rated load the computer shuts down operations quickly enough to avoid damage. Isn't this a real time operation?
: :
: : dssuresh6,
I would say that generally a real time operation is a system that reacts to 'interrupt' conditions, which is pretty much what you describe, so yes, I would say it is a real time operation.

A non-real time operation would be a system where there are no 'hard-deadlines' as far as reaction time to an event goes. I suppose it's kind of a gray area and what needs to be defined is 'How fast is fast?'

Hope that was a 'hard' enough answer, dssuresh6.

tankarray