Although I have implemented many projects in C, I am completely new to operating systems. I tried real time linux on Discovery board (STM32) and got the correct results for blinking LED but I didn't really understand the whole process since I just followed the steps and could not find whole description for each step on the internet.
I want to implement scheduling on real time linux. What is the best way to start? Any sites, books, tutorials available?
Complete RTLinux process description will be appreciated.
Thanks in adv.
The transition from "bare metal" to OS based programming is something that I experienced in reverse. I started out a complete software guy, totally into the OS side of things and over time I have moved to the opposite of that (even designing circuits in VHDL!). My advice would be to start simple. Linux is pretty complex, and everywhere you look there are many layers of things all working together to deliver the final product. If you are dead set on a real time linux extension, I'd be happy to suggest https://xenomai.org/ which is a real time extension for linux.
However, to more specifically address your question about implementing scheduling in Linux, you can, but it will be a large amount of work and can be very complicated. The OS uses a completely fair scheduling process ( http://en.wikipedia.org/wiki/Completely_Fair_Scheduler ) and whenever you spin up a thread, it simply gets added to the list to run. This can differ slightly if you implement your code in kernel space as a driver, rely on hardware interrupts, etc., but in general, this is how Linux works. Real time generally means that it has the ability to assign threads one of several different priorities and utilize thread preemption fully at any given time which are concepts that aren't really a part of vanilla Linux. It has some notion of this, but it has limitations that can cause problems when you are looking for real time behavior from Linux.
What may be helpful to you is an RTOS. If you are looking for a full on Real Time Operating System, check out FreeRTOS http://www.freertos.org/ . It has a large community and supports a lot of different devices out of the box with a large amount of example code. They even support your specific board with an example package, so you can give it a shot with nothing to lose! http://www.freertos.org/FreeRTOS-for-Cortex-M3-STM32-STM32F100-Discovery.html . It gives you access to many OS ish constructs like network APIs, memory management, and threading without the overhead and latency of a huge OS. With an RTOS, you create tasks and assign them priorities so you become the scheduler and are no longer at the mercy of the OS. You run the OS, not the OS runs you (if that makes sense). Plus, the constructs offered within an RTOS will feel much like bare metal code and thus will be much easier to follow, understand, and fully learn. It is a more simple world to learn the base building blocks of a full blown OS such as Linux or Windows. If this option sounds good, I would suggest looking through the supported devices on FreeRTOS website and picking one you would like to experiment with and then go for it. I would highly recommend this as a way to learn about scheduling and OS constructs in general as it is as simple as you can get and open source. Once you have the basics of an RTOS down, buying a book about Linux specifically wouldn't be a bad idea. Although there are many free resources on the web related to learning about Linux, they are commonly contradictory, and can be misleading. Pile on learning Linux specific knowledge along with OS in general, and it can feel overwhelming. Starting simpler will help keep you from getting burnt out and minimize the amount of time you spend feeling lost. Linux is definitely a learning process, but like with any learning process, start simple, keep your ultimate goal in mind, make a plan, and take small, manageable steps along that plan until you look up and find yourself exactly where you want to be. Then go tackle the next mountain!
The real-time Linux landscape is quite confusing. 99.99% of the information out there is just plain obsolete.
First, there are lots of "microkernels" that run Linux as one task. (Such as the defunct RTLinux). The problem is that you must write your real-time task to a different API, and can't depend on anything in Linux, because Linux will be frozen in the background while your task runs. So unless your task is dead-simple ("stop the motors when I press this button"), this approach will cause more pain than gain.
Next, there is the realtime Linux patch set. This hasn't been doing so well. because of the next item:
Lastly, the current Linux kernel has gotten rid of the problems that caused people to need realtime in the past. You can even turn off Linux on one of your processors to have full control of the CPU. See also this paper.
To answer your question: I see two different paths you could take:
1) Start with a normal 3.xx Linux kernel and explore the various APIs and realtime techniques (i.e. realtime priorities, memory pinning, etc.) This can get you "close enough" for 99% of what people want "realtime" for. If it's good enough for high frequency trading, it's probably good enough for you.
2) If you have a hard realtime requirement and you are worried that Linux won't cut it, then (as Nick mentioned above), just go buy a processor and write your realtime code with no OS. By splitting up your "realtime" and "non-realtime" code onto different CPUs, you will make the whole system simpler and much more robust.
If you want to learn real-time operating systems then I suggest that you get an FPGA, for example the Altera DE2, and experiment with your own operating system and ucos. You can read a good text about embedded RTOS here.
You could also get a Linux Raspberry and write your own operating system for that.
Related
Some background:
I am a Computer Engineering major attending school right now, and I just completed a project that created a microprocessor with a super simple instruction set that ran on an FPGA. I chose to implement a simple file storage scheme, a VGA text only display output, and a PS/2 keyboard input. I wrote two main programs, a firmware that was in ROM in the processor and a kernel that provided a bunch of library type functions, and was capable of loading and executing files from the filesystem. This project was challenging and overall a lot of fun.
My Question:
I want to do some super low level programming on a modern computer, but I can't seem to find any resources or documentation that help me get started. To be clear, I want to find the proper documentation that would help me to write a program in C, x86, or x86-64 assembly that I could compile, and format into some form of bootable data. I know this is a daunting task, and typically not something a hobbyist would take on, but I know that it's possible (Terry Davis's TempleOS).
Are there any websites or books that I can read that will contain the specifics needed to make something like this?
Look out, you might just catch the bug. OS Development, though having a very small demographic, is still quite a thriving hobby. Once you start, you may not ever give it up.
Since your subject line states 64-bit and you use the term modern hardware, be advised that modern hardware no longer has the older style BIOS, where the developer wrote the boot process which included the video out, file system in, and other standard routines. Modern hardware now uses an EFI firmware which does all of the booting for you, including the reading from the file system(s). For modern hardware, OS development really starts with the OS Loader, the part that loads the OS, and this is done in a high level language such as C/C++. Very little if any assembly at all, in fact that is its point/purpose.
Don't get too discouraged, currently a lot of computers still allow the old style boot. However, the old style boot starts in 16-bit mode, moves to 32-bit mode, then if desired, moves to long mode (64-bit). There also are emulators that you can use so you don't have to have a separate system, just to test your development. I prefer Bochs myself, but I am a little bias since I wrote some of the code for it, namely most of the (original) USB emulation.
If you wish to dip your toes into this hobby, there are numerous places to start. I personally wrote a few books on the subject. They show you how to start from when the time the POST gives up control to your boot code, up to the point of a minimal Round Robin style task/thread switching OS, with all the necessary hardware and software basics to do so. There is a forum to OS Development, along with its wiki.
Again, a project like this is not for the faint at heart, though it is an enjoyable hobby most have found to be a very good learning experience.
I have been programming in java for 3 years but I have no experience with other languages. I want to know what I need to study in order to be able to make an operating system. Am most likely going to make my operating system based on Linux kernel. What programming languages should I be familiar with and what aspects of the pc hardware should I study. if you know any online tutorials or good books please mention them.
The answer depends on how far you want to go and how much you want to write yourself vs using existing code.
The most straightforward way is to have a look at Arch Linux or Gentoo and build your own, custom Linux setup. Approximate time needed to create a minimal working system: ~2 hours
You could, otherwise, compile the Linux kernel, build some software packages and put it all together yourself and create your own Linux distro - i.e your own operating system in a sense. Linux From Scratch will be an invaluable resource if this is what you decide to do. Approximate time needed to create a minimal working system: ~2-5 days
Say that's too easy for you but you're not ready to delve into the nitty-gritty of kernel development. You can write your own applications that run on top of the Linux kernel. Typically, you'd need to know C/C++ but any language that supports running on Linux/compiling to a Linux executable will work. Heck, you could chuck a (or write your own) Java runtime on and write your whole 'operating system'/user-space in Java. Approximate time needed to create a minimal working system: ~6-12 months
What about if you want to get down to programming your own kernel from scratch? Heck, Linux is overrated and you want to write your own kernel and it's going to be the next best thing! You would want to learn a bit about the platform you're developing on. At the very least, you'll need to know the assembly instructions of some special operations for your platform that can't be done natively in C like switching CPU modes. You'll definitely need to read up on the OS dev wiki for this and you should have a fair decent computer science background.
At this point, you shouldn't need too much other than a good tutorial, C, a little assembly, reference manuals for the hardware you're hoping to support and 3+ years of computer science background to get you started. Your boot loader that's booting your kernel should handle most of the hardware initialisation. Bootloaders like GRUB (I'm assuming you're developing on a x86 system) does so much for you that it can probably just jump to your kernel main function straight away without you having to do too much. Again, if you wanted, you could port or write your own Java runtime in C and write the rest of your kernel in Java! Approximate time needed to create a minimal working system: ~3-5 years
But, let's just say you're screaming for more pain and you want to write an operating system really from scratch and you don't want no bootloader doing a lot of the legwork. What do you need? Firstly, you'll need a lot more reference manuals. And you'll need to read up on a lot more assembly. Especially for Intel processors, there's a lot of work involved bringing the system up from 16bit mode to 32/64bit protected mode with paging (which is what I assume you want). You'll also want to know about every tiny quirk and weirdness of your platform that will affect your OS (these are often not documented; hooray!). Plus, all of the above. In short, you will need to study everything. Approximate time needed to create a minimal working system: 5+ years
Of course, this post is just scraping the surface of what is needed to even bring up a basic operating system capable of say, opening a web browser. The approximates roughly assumes a minimal working system is something capable of running a graphical web browser and will vary depending on how much you want to write.
I don't mean to be condescending but this is the kind of reality you'll be facing if you decide to write your own operating system. Nevertheless, it is a valuable learning experience if you can break the initial barrier or even just trying to set up your own Linux system.
First, I would say that you install any Linux OS on your system and get accustomed with it.
Second, for OS development you have to know C language. As for the Assembly language, it depends from where you start the OS development. If you will be using available bootloaders, then I don't think that you will be requiring to learn assembly language.
This is a website on OS development: http://wiki.osdev.org/Main_Page
There you will find all the stuff you need to know for the OS development. And also how to develop OS step-by-step.
Now-a-days, a "Eudyptula Challenge" is going on. It is a series of programming exercise for the linux kernel. You can find more info here: http://eudyptula-challenge.org/
I'm taking an introductory course (3 months) about real time systems design, but any implementation.
I would like to build something that let me understand better what I'll learn in theory, but since I have never done any real time system I can't estimate how long will take any project. It would be a concept proof project, or something like that, given my available time and knowledge.
Please, could you give me some idea? Thank you in advance.
I programm in TSQL, Delphi and C#, but I'll not have any problem in learning another language.
Suggest you consider exploring the Real-Time Specification for Java (RTSJ). While it is not a traditional environment for constructing real-time software, it is an up-and-coming technology with a lot of interest. Even better, you can witness some of the ongoing debate about what matters and what doesn't in real-time systems.
Sun's JavaRTS is freely available for download, and has some interesting demonstrations available to show deterministic behavior, and show off their RT garbage collector.
In terms of a specific project, I suggest you start simple: 1) Build a work-generator that you can tune to consume a given amount of CPU time; 2) Put this into a framework that can produce a distribution of work-generator tasks (as threads, or as chunks of work executed in a thread) and a mechanism for logging the work produced; 3) Produce charts of the execution time, sojourn time, deadline, slack/overrun of these tasks versus their priority; 4) demonstrate that tasks running in the context of real-time threads (vice timesharing) behave differently.
Bonus points if you can measure the overhead in the scheduler by determining at what supplied load (total CPU time produced by your work generator tasks divided by wall-clock time) your tasks begin missing deadlines.
Try to think of real-time tasks that are time-critical, for instance video-playing, which fails if tasks are not finished (e.g. calculating the next frame) in time.
You can also think of some industrial solutions, but they are probably more difficult to study in your local environment.
You should definitely consider building your system using a hardware development board equipped with a small processor (ARM, PIC, AVR, any one will do). This really helped remove my fear of the low-level when I started developing. You'll have to use C or C++ though.
You will then have two alternatives : either go bare-metal, or use a real-time OS.
Going bare-metal, you can learn :
How to initalize your processor from scratch and most importantly how to use interrupts, which are the fastest way you have to respond to an externel event
How to implement lightweight threads with fast context switching, something every real-time OS implements
In order to ease this a bit, look for a dev kit which comes with lots of documentation and source code. I used Embedded Artists ARM boards and they give you a lot of material.
Going with the RT OS :
You'll fast-track your project, and will be able to learn how to fine-tune a RT OS
You may try your hand at an open-source OS, such as Linux or the BSDs, and learn a lot from the source code
Either choice is good, you will get a really cool hands-on project to show off and hopefully better understand your course material. Good luck!
As most realtime systems are still implemented in C or C++ it may be good to brush up your knowledge of these programming languages. Many realtime systems are also embedded systems, so you might want to play around with a cheap open source one like BeagleBoard (http://beagleboard.org/). This will also give you a chance to learn about cross compiling etc.
As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.
Closed 10 years ago.
I am toying with the idea of creating an completely new operating system and would like to hear what everyone on this forums take is on that? First is it too late are the big boys so entrenched in our lives that we will never be able to switch (wow - what a terrible thought...). But if this is not the case, what should a operating system do for you? What features are the most important? Should all the components be separate installations (in other words - should the base OS really have no user functionality and that gets added on by creating "plug-ins" kind of like a good flexible tool?)
Why do I want to do this... I am more curious about whether there is a demand and I am wondering, since the OSes we use most today (Linux, Windows, Mac OS X (Free BSD)) were actually written more than 20 years ago (and I am being generous - I mean dual and quad cores did not exist back then, buses were much slower, hardware was much more expensive, etc,...), I was just curious with the new technology if we would do anything differently?
I am anxious to read your comments.
To answer the first question: It's never too late. Especially when it comes to niche market segments and stuff like that.
Second though, before you start down the path of creating a new OS, you should understand the kind of undertaking it is: it'd be a massive project.
Is it just a normal programmer "scratch the itch" kind of project? If so, then by all means go ahead -- you might learn alot of things by doing it. But if you're doing it for the resulting product, then you shouldn't start down that path until you've looked at all the current OSes under development (there are alot more than you'd think at first) and figured out what you'd like to change in them.
Quite possibly the effort would be better spent improving/changing an existing open source system. Even for your own experimentation, it may be easier to get the results you want if you start out with something already in development.
First, a little story. In 1992, during the very first Win32 ( what would become the MS Professional Developers Conference ) conference, I had the opportunity to sit with over some lunch with one Mr. Dave Cutler ( Chief Architect of what most folks would now know as Windows NT,Windows 2000, XP, etc. ).
I was at the time working on the Multimedia group at IBM Boca Raton on what some of you might remember, OS/2. Having worked on OS/2 for several years, and recognizing "the writing on the wall" of where OSes were going, I asked him, "Dave, is Windows NT going to take us into the next century or are there other ideas on your mind ?". His answer to me was as follows:
"M...., Windows NT is the last operating system anyone will ever develop from scratch !". Then he looked over at me, took a sip of his beer, and said, "Then again, you could wake up next Saturday after a particularly good night out with your girl, and have a whole new approach for an operating system, that'll put this to shame."
Putting that conversation into context, and given the fact I'm back in college pursuing my Master's degree ( specializing in Operating Systems design ), I'd say there's TONS of room for new operating systems. The thing is to put things into perspective. What are your target goals for this operating system ? What problem space is it attempting to service ?
Putting this all into perspective will give you an indication of whether you're really setting your sights on an achievable goal.
That all being said, I second an earlier commenters note about looking into things like "Singularity" ( the focus of a talk I gave this past spring in one of my classes .... ), or if you really want to "sink your teeth into" an OS in its infancy....look at "ReactOS".
Then again, WebOSes, like gOS, and the like, are probably where we're headed over the next decade or so. Or then again, someone particularly bright could wake up after a particularly fruitful evening with their lady or guy friend, and have the "next big idea" in operating systems.
Why build the OS directly on a physical machine? You'll just be mucking around in assembly language ;). Sure, that's fun, but why not tackle an OS for a VM?
Say an OS that runs on the Java/.NET/Parrot (you name it) VM, that can easily be passed around over the net and can run a bunch of software.
What would it include?
Some way to store data (traditional FS won't cut it)
A model for processes / threads (or just hijack the stuff provided by the VM?)
Tools for interacting with these processes etc.
So, build a simple Platform that can be executed on a widely used virtual machine. Put in some cool functionality for a specific niche (cloud computing?). Go!
For more information on the micro- versus monolithic kernel, look up Linus' 'discussion' with Andrew Tanenbaum.
I would highly suggest looking at an early version on linux(0.01) to at least get your feet wet. You're going to mucking about with assembly and very obscure low-level stuff to even get started (especially getting into protected mode, multi-tasking, etc). And yes, it's probably true that the "big boys" already have the market cornered. I'm not telling you NOT to do it, but maybe doing some work on the linux kernel would be a better stepping stone.
Check out Cosmos and Singularity, these represent what I want from a futuristic operating system ;-)
Edit :
SharpOS is another managed OS effort. Suggested by yshuditelu
An OS should have no user functionality at all. User functionality should be added by separate projects, which does not at all mean that the projects should not work together!
If you are interested in user functionality maybe you should look into participating in existing Desktop Environment projects such as GNOME, KDE or something.
If you are interested in kernel-level functionality, either try hacking on a BSD derivate or on Linux, or try creating your own system -- but don't think too much about the user functionality then. Getting the core of an operating system right is hard and will take a long time -- wanting to reinvent everything does not make much sense and will get you nowhere.
You might want to join an existing OS implementation project first, or at least look at what other people have implemented.
For example AROS has been some 10 or more years in the making as a hobby OS, and is now quite usable in many ways.
Or how about something more niche? Check out Symbios, which is a fully multitasking desktop (in the style of Windows) operating system - for 4MHz Z80 CPUs (Amstrad CPC, MSX). Maybe you would want to write something like this, which is far less of a bite than a full next-generation operating system.
Bottom line...focus on your goals and even more importantly the goals of others...help to meet those needs. Never start with just technology.
I'd recommend against creating your own Operating System. (My own geeky interruption...Look into Cloud Computing and Amazon EC2)
I totally agree that it would first help by defining what your goals are. I am a big fan of User Experiences and thinking of not only your own goals but the goals of your audience/users/others. Once you have those goals, then move to the next step of how to meet it.
Now days what is an Operation System any way? kernal, Operating System, Virtual Server Instance, Linux, Windows Server, Windows Home, Ubuntu, AIX, zSeries OS/390, et al. I guess this is a good definition of OS... Wikipedia
I like Sun's slogan "the Network is the computer" also...but their company has really fallen in the past decade.
On that note of the Network is the computer... again, I highly recommend, checking out Amazon EC2 and more generally cloud computing.
I think that building a new OS from scratch to resemble the current OSes on the market is a waste of time. Instead, you should think about what Operating System will be like 10-20 years from now. My intuition is that they will be so different as to render them mostly unrecognizable by today's standards. Think of frameworks such as Facebook (gasp!) for models of how future OSes will operate.
I think you're right about our current operating systems being old. Someone said that all operating systems suck. And yes, don't we have problems with them? Call it BSOD, Sad Mac or a Kernel Panic. Our filesystems fail, there are security and reliability problems.
Microsoft pursued interesting approach with its Singularity kernel. It isolates processes in software, using a virtual machine similar to .NET, and formal verification methods. Basically all IPC seems to be formally specified and verified, even before a program is ran.
But there's another problem with it - Singularity is only a kernel. You can't run application not designed for it on it. This is a huge penalty, making eventual transition (Singularity is not public) quite hard. If you manage to produce something of similar technical advantages, but with a real transition plan (think about IPv4->IPv6 problems, or how Windows got so much market share on desktop), that could be huge!
But starting small is not a bad choice either. Linux started just like this, and there are many cases when it leads to better design. Small is beautiful. Easier to change. Easier to grow. Anyway, good luck!
checkout singularity project,
do something revolutionary
I've always wanted an operating system that was basically nothing but a fresh slate. It would have built in plugin support which allow you to build the user interface, applications, whatever you want.
This system would work much like a Lua sandbox to a game would work, minus the limitations. You could build a plugin or module system that would have access to a variety of subsystems that you would use. For example, if you were to write a web browser application, you would need to load the networking library and use that within your plugin script. Need 'security' ? Load the library.
The difference between this and Linux is that, Linux is an operating system but has a windows manager that runs over top of it. In this theoretical operating system, you would be able to implement the generic "look" and "feel" of a variety of windows within the plugin system, or could you create a custom interface.
The difference between this and Windows is that its fully customizable, and by fully I mean if you wanted to not implement any cryptography at all, you can do that, or if you wanted to customize an already existing window, you can do that. Nothing is closed to you.
In this theoretical operating system, there is an OS with a plugin system. The plugin system uses a simple and powerful language.
If you're asking what I'd like to see in an operating system, I can give you a list. I am just getting into programming so I'm not sure if any of this is possible, but I can give you my ideas.
I'd like to see a developed operating system (besides the main ones) in which it ISN'T a pain to get the wireless card to work. That is my #1 pet peeve with most of the ones I've tried out.
It would be cool to see an operating system designed by a programmer for other programmers. Have it so you can run programs for all different operating systems. I don't know if that's possible without having a copy of windows and OSX but it would be really damn cool if I could check the compatablity of programs I write with all operating systems.
You could also consider going with MINIX which is a good starting point.
To the originator of this forum, my hats off to you sir for daring to think in much bolder and idealistic terms regarding the IT industry. First and foremost, Your questions are precisely the kind you would think should engage a much broader audience given the flourishing Computer Sciences all over the globe & the openness taught to us by the Revolutionary Linux OS, which has only begun to win the hearts and minds of so many out there by way of strengthing its user-friendly interface. So kudos on pushing the envelope.
If I'm following correctly, you are supposing that given the fruits of our labor thus far, the development of further hardware & Software concoctions could or at least should be less conventional. The implication, of course, is that any new development would reach its goal faster than what is typical. The prospect, however, of an entirely new OS system #this time would be challenging - to say the least - only because there is so much friction out there already between Linux & Windows. It is really a battle between open source & the proprietary ideologies. Bart Roozendaal in a comment above proves my point nicely. Forget the idea of innovation and whatever possibilities may come from a much more contemporary based Operating System, for such things are secondary. What he is asking essentially is, are you going to be on the side of profit or no? He gives his position away easily here. As you know, Windows is notorious for its monopolistic approach regarding new markets, software, and other technology. It has maintained a deathgrip on its hegemony since its existence and sadly the windows os is racked with endless bugs & backdoors.
Again, I applaud you for your taking a road less travelled and hopefully forgeing ahead and not becoming discouraged. Personally, I'd like to see another OS out there...one much more contemporary.
I've been involved in embedded operating systems of one flavor or another, and have generally had to work with whatever the legacy system had. Now I have the chance to start from scratch on a new embedded project.
The primary constraints on the system are:
It needs a web-based interface.
Inputs are required to be processed in real-time (so a true RTOS is needed).
The memory available is 32MB of RAM and FLASH.
The operating systems that the team has used previously are VxWorks, ThreadX, uCos, pSOS, and Windows CE.
Does anyone have a comparison or trade study regarding operating system choice?
Are there any other operating systems that we should consider? (We've had eCos and RT-Linux suggested).
Edit - Thanks for all the responses to date. A pity I can't flag all as "accepted".
I think it would be wise to evaluate carefully what you mean by "RTOS". I have worked for years at a large company that builds high-performance embedded systems, and they refer to them as "real-time", although that's not what they really are. They are low-latency and have deterministic schedulers, and 9 times out of 10, that's what people are really after when they say RTOS.
True real-time requires hardware support and is likely not what you really mean. If all you want is low latency and deterministic scheduling (again, I think this is what people mean 90% of the time when they say "real-time"), then any Linux distribution would work just fine for you. You could probably even get by with Windows (I'm not sure how you control the Windows scheduler though...).
Again, just be careful what you mean by "Real-time".
It all depends on how much time was allocated for your team has to learn a "new" RTOS.
Are there any reasons you don't want to use something that people already have experience with?
I have plenty of experience with vxWorks and I like it, but disregard my opinion as I work for WindRiver.
uC/OS II has the advantage of being fully documented (as in the source code is actually explained) in Labrosse's Book. Don't know about Web Support though.
I know pSos is no longer available.
You can also take a look at this list of RTOSes
I worked with QNX many years ago, and have nothing but great things to say about it. Even back then, QNX 4 (which is positively chunky compared to the Neutrino microkernel) was perfectly suited for low memory situations (though 32MB is oodles compared to the 1-2MB that we had to play with), and while I didn't explicitly play with any web-based stuff, I know Apache was available.
I purchased some development hardware from netburner
It has been very easy to work with and very well documented. It is an RTOS running uCLinux. The company is great to work with.
It might be a wise decision to select an OS that your team is experienced with. However I would like to promote two good open source options:
eCos (has you mentioned)
RTEMS
Both have a lot of features and drivers for a wide variety of architectures. You haven't mentioned what architecture you will be using. They provide POSIX layers which is nice if you want to stay as portable as possible.
Also the license for both eCos and RTEMS is GPL but with an exception so that the executable that is produced by linking against the kernel is not covered by GPL.
The communities are very active and there are companies which provide commercial support and development.
We've been very happy with the Keil RTX system....light and fast and meets all of our tight real time constraints. It also has some nice debugging features built in to monitor stack overflow, etc.
I have been pretty happy with Windows CE, although it is 'heavier'.
Posting to agree with Ben Collins -- your really need to determine if you have a soft real-time requirement (primarily for human interaction) or hard real-time requirement (for interfacing with timing-sensitive devices).
Soft can also mean that you can tolerate some hiccups every once in a while.
What is the reliability requirements? My experience with more general-purpose operating systems like Linux in embedded is that they tend to experience random hiccups due to their smart average-case optimizations that try to avoid starvation and similar for individual tasks.
VxWorks is good:
good documentation;
friendly developing tool;
low latency;
deterministic scheduling.
However, I doubt that WindRiver would convert their major attention to Linux and WindRiver Linux would break into the market of WindRiver VxWorks.
Less market, less requirement of engineers.
Here is the latest study. The last one was done more than 8 years ago so this is most relevant. The tables can be used to add additional RTOS choices. You'll note that this comparison is focused on lighter machines but is equally applicable to heavier machines provided virtual memory is not required.
http://www.embedded.com/design/operating-systems/4425751/Comparing-microcontroller-real-time-operating-systems