On Thu, 10 Dec 2009 22:22:01 +0200 Bernt Kamark <[email protected]> wrote:
> 2009/12/10 Alf Kaanoken <[email protected]> > > > Hi, > > > hallo Hi, looks like you find more time , is it right ? > > > I wrote before about project state. > > I agree that the project has no designated purpose at the current moment. > > This is a mistake and there are a time to fix it. I think many people agree > > with me is that it makes no sense to > > develop a general purpose platform. > > The project team will focus on new goals (see below), however it will not > > interfere woth other developers to work in > > other directions and areas. So, new developers will have to choose between > > working in team and its own works in the project, > > this choice depends on the interest. In fact, it doesn't negate assistance > > for those who working outside of the team, all rules still > > the same. > > > im interested in as going to work within open source project team and it's > not for fun only. i will choose first. Good choice ;) > > > In fact, there is separation between the basic (base) and additional > > (contrib/extra) components. The set of basic component can be > > extended with additional ones in case this is not inconsistent with the > > project objectives, it also depends on the team resources > > (if something included it should be maintained by somebody). > > Well, the goal ... > > The main goal of the project is to create a platform for industrial > > computers use that are involved in the monitoring, > > controlling, managing and visualization of tthe industrial processes. Also, > > this goal includes creating a such kind > > of the platform that can be used on autonomous embedded systems (vessile > > drones for example). > > > it's a DEAL! it's not very interesting to deal with os only (gen. purpose is > deal for os only ) General purpose OSdev is a deal for those people who wants develop OS for OS. > > > I want to make a note: Jari OS should be given a possibility for other > > non-POSIX environments to run (i.e. the platform > > should provide a flexible architecture to implement it). > > > note: most of undustry ones are not free I know, I mean here the development team should take in mind that non-POSIX world exists ;) > > > That is, works will focus not on the implementation of a big set of > > drivers, protocols features etc..., but on the implementation > > a narrow set of tasks and providin necessary interfaces for the drivers, > > protocols, standards etc implementations. > > This doesn't mean that the system will not provide any functional, on the > > contrary, the system will has the minimum necessary functional > > and set of basic subsystems, but their amount will be strictly > > deterministic. > > In the future, it's a possible to extend range of supported drivers and > > protocols (due to the actual use of the platform), but this is > > not a priority. > > > it's a typical , don't care > > > In this case I want to modify plans for the first stable release and makes > > a comments on this: > > Networking. I will exclude networking subsystem from the first stable > > release(it moved to the next release) . Networking is required, but it > > hasn't a high priority. By the way, I will help with it for anybody who will > > start it, but my time is limited. > > > i think the system must contain a tcp/ip stack,also take a focus onto > DeviceNet, Profibus, CIA, well ancient modbus still in use. > existing tcp/ip will reduce some pain with industrial protos support,u may > take a google to learn more. > here > http://www.profibus.com/nc/downloads/downloads/profinet-technology-and-application-system-description/display/is > an overview of profinet. I will read this. > > > Ext2. This filesystem is not suitable for the industrial system use. We > > will switch to XFS (it has all functionality for us(actually this is the > > system that will supported, other filesystems will be in the extra > > components set)). Anyway, ext2 can be done in full functionality if somebody > > wants to take this task. > > > ?? "it has all functionality for us" - explain please. It's stable, has journaling, it well designed and has a killer feature: Guaranteed rate I/O The XFS guaranteed rate I/O system provides an API that allows applications to reserve bandwidth to the filesystem. XFS will dynamically calculate the performance available from the underlying storage devices, and will reserve bandwidth sufficient to meet the requested performance for a specified time. This feature is unique to the XFS file system. Guarantees can be hard or soft, representing a trade off between reliability and performance, though XFS will only allow hard guarantees if the underlying storage subsystem supports it. This facility is most used by real-time applications, such as video-streaming. I think that some industrial units use streaming (I know several examples). Well, it stands that XFS will "cover the ass" with features. > > > Security. We will leave DAC only for the first releases. > > > superish security for plc-like devices is a marketing bullshit. DAC is > needed to cover the ass ;) PLC devices hasn't any security, more PC-like PLCs going with QNX - and yep, QNX has DAC. But theoretically some applications might require more complex security. > > > RPC box. it excluded, IDL will exchange it. > > Sockets. it excluded, but like a networking this task moved to the next > > release. > > Currently we reviewing all subsystems, changing it to the IPCbox/IDL pair > > (instead of ugly parsing). > > Some later we will add driver model, and system bus. > > Our plans for the first release is a stable platform with strict > > functionality and set of the extensible API (IDL for RPC interfaces, > > driver model for drivers and so on ...). > > > > Any comments ? > > > how can i help now ? // "in the team" ;) I cannot allocate any time for XFS, I promise that libv2 API will not changed seriously. However, fs API will be extended (for guaranteed-IO/direct-IO and some other kinds of locating resources and for I/O). XFS has the OS-depended layer (checkout in linux or freebsd source tree). Well, could you take this task ? > > > > > Thanks. > > > > -- > > Alfeiks Kaänoken, > > Team Lead of the > > Jari OS R&D Team. > > http://jarios.org > > Free open-source > > microkernel-based > > multiservice RTOS > > _______________________________________________ > > Jarios-dev mailing list > > [email protected] > > http://lists.jarios.org/cgi-bin/mailman/listinfo/jarios-dev > > > -- Alfeiks Kaänoken, Team Lead of the Jari OS R&D Team. http://jarios.org Free open-source microkernel-based multiservice RTOS _______________________________________________ Jarios-dev mailing list [email protected] http://lists.jarios.org/cgi-bin/mailman/listinfo/jarios-dev
