Actually, for the debian rt-preempt there is an apt repo to add .... all of rt-preempt is not mainlined yet, but it's the goal. But I have been running an rt-preempt kernel on my laptop for a while, and quite arguably it doesn't affect performance much, until I do the tuning of the system, in which case it doesn't sleep and so on.
Also there seems to be some confusion on rt-preempt and RT-Linux .... they are different beasts. I don't think it's the high end board thing, it's more how the system handles context switches, and the device drivers, the main problem is locking in the rest of the kernel. RT-Preempt runs the interrupts as threads, in order to do priority inversion and such, so things got a lot better after the BKL went away. There are problems with proprietary grafix drivers also, ( NVidia ) .... but on the whole it's a lot more of a regular linux kernel, most kernels are running preempt at the moment, it's just the rt part that is specific. Like any other system it's sensitive to tuning, but I think there is a project for profiling problems, check out the rt-wiki https://rt.wiki.kernel.org/index.php/Main_Page , there are a lot of tools and hints, so please have a look before bashing. There are also a bit of examples, it's not 100% up to date but nice reading. The mailing list is really good, if you have a solid problem, ask and you shall recieve .... I suggest we check with the mutex bug on that list. / regards, Lars Segerlund. 2012/8/28 Michael Haberler <[email protected]>: > > Am 28.08.2012 um 09:00 schrieb Jan de Kruyf: > >> I fully share the urgency Michael feels about the issue, but I think that >> the RT issue is not easily tackled from LinuxCNC2. >> >> So in LinuxCNC2 I would rather vote for getting Xenomai to work (again) as >> a real time alternative. If you look in the building scripts there always >> was that option, it has been neglected a bit lately. >> On top of it, there is the issue of the software stepgeneraration for a >> majority of systems, for which RTPREEMPT is not the perfect vehicle yet, >> unless you invest in a high-end board. > > Jan, > > I did not say "let us break software step generation". > > I said: the LinuxCNC project needs an *alternative* realtime kernel which > runs on every washer and dryer, and it better be mainstream, that is: > installable with apt-get and no extra apt repo to add. That in my view is the > top priority - "simplify, simplify, simplify - and get rid of this goddamn > guru factor". > > RTAI will continue to run until Paolo's fate happens to turn to the worse, > plus some delta-T. So I do not understand your issue. And parports are > legacy, too. > > Wrt to performance discussions: I love them, especially if they are based on > facts. So let's get this port done, and do some measuring, or out again we > are in "conjecture and musing land". I really suggest we put this agonizing > habit to rest. > > Personally I do not think that high-rate software stepgen needs to be > supported on every conceivable platform combination. That said: what Lars > says hints me that reasonable software stepgen is entirely feasible with > RT_PREEMPT. > > I also dont say "forget Xenomai" - if someone does the patches, fine. I'm > just not sure the upside for supporting it long term is high enough. > > Wrt to cost: I dont get kickbacks from Mesanet or Pico systems, but I think > the cost of their hardware to do fast stepgen is negligible in the context of > overall project cost. > > We need to stop making "sound technical decisions" which might be anywhere > from not-so-great to reckless from a project lifecycle and adoption point of > view. > > - Michael > > >> >> My experience with other systems shows that RTPREEMPT is quite stable on >> the 3 series kernel, provided some rules are followed in your design. >> >> So personally I am not prepared to fantasize about backhacking RT into >> LCNC2, a few people have been at it now and nothing was finished into a >> workable product or perhaps the interest of the community is not there, I >> do not know. >> >> cheers, >> >> j. >> >> >> >> On Mon, Aug 27, 2012 at 10:04 PM, Michael Haberler <[email protected]>wrote: >> >>> I am late to get the gist on options for RT with Linux, but what I found >>> leaves me a bit speechless. >>> >>> My understanding is the following: >>> >>> LinuxCNC supports two RT linux kernels, RTLinux and RTAI. >>> >>> RTLinux is dead in the water. Website doesnt even respond. Product >>> abandoned by last owner, Wind River Systems. >>> >>> RTAI is a mostly one-man-show project locked into a single platform. >>> >>> The hope for RTAI on non-PC platform is just that - somebody seems to have >>> noticed a dead end here. >>> >>> RT_PREEMPT is where the Linux mainstream is heading, as is Xenomai stated >>> direction. >>> >>> There is no coordinated attempt on getting LinuxCNC to run out of the box >>> on RT_PREEMPT or Xenomai, or any other alternative, despite some work as at >>> least a starting point being available. >>> >>> --- >>> >>> Please tell me where I'm wrong: is this really it - Paolo walks into the >>> wrong car, and the LinuxCNC project has no viable and believably maintained >>> kernel alternative at hand? >>> >>> If I'm right: this must be changed ASAP, and this is a LinuxCNC2 issue. >>> >>> At the risk of hurting some feelings: if LinuxCNC were my company, and my >>> chief tech told me "this is our strategy", I would have this fellow empty >>> his desk the very same day. >>> >>> - Michael >>> >>> >>> ps: I am leaving software stepgen type performance issues out of the >>> picture for now, and on purpose - I dont think this is a terribly >>> important property going forward given the hardware on the market at a >>> reasonable price point and with good integration into LinuxCNC. >>> >>> ------------------------------------------------------------------------------ >>> Live Security Virtual Conference >>> Exclusive live event will cover all the ways today's security and >>> threat landscape has changed and how IT managers can respond. Discussions >>> will include endpoint security, mobile security and the latest in malware >>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ >>> _______________________________________________ >>> Emc-developers mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/emc-developers >>> >> ------------------------------------------------------------------------------ >> Live Security Virtual Conference >> Exclusive live event will cover all the ways today's security and >> threat landscape has changed and how IT managers can respond. Discussions >> will include endpoint security, mobile security and the latest in malware >> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ >> _______________________________________________ >> Emc-developers mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/emc-developers > > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Emc-developers mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-developers ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
