A commit a year ago in MK legacy addresses something with libmodbus. https://github.com/machinekit/machinekit/tree/master/src/hal/user_comps/mb2hal
I don't know that I was asking specifically about LinuxCNC's version unless there is a more recent change that matters. The version that was in MK is probably fine if it's useable. On Saturday, December 28, 2019 at 12:10:25 PM UTC-5, [email protected] wrote: > > Well, looking at the mb2hal in LinuxCNC repository, I - at first glance - > don't see any reason why it should not work in the Machinekit (maybe with > some simple and generally automatic code changes). > > So I would really like to know the reason behind removing it. I think > (trying to remember) that there was some issue with the libmodbus library > and Debian package which mb2hal is wrapper around. > > Cern. > > > Dec 28, 2019, 16:40 by [email protected] <javascript:>: > > > Just navigating the split repo directories, "user" comps still exist in > the typical directory of MK-hal, tho mb2hal is missing. Nothing like that > seems to exist in MK-CNC so I assumed the MO was to keep the user comps in > MK-HAL as is. mb2hal also works over serial which aside from Mesa > smartserial I don't think has ever been considered "RT" in linuxCNC likely > because of baud rates. While I'm sure RT modbus is probably quite useful I > don't think the main use for it is usually RT critical. Generally it would > take the place of other serial protocols with the convenience of TCP as > well for interfacing with something else that is running faster at a > hardware level, like talking to a VFD or commanding a servo drive. I think > getting it in there in it's current form if possible is probably more > important than addressing RT changes, which could always be added later I > suppose. > > > > On Saturday, December 28, 2019 at 9:50:53 AM UTC-5, [email protected] > wrote: > > > >> I remember that there was some problem with Modbus module which I > thought was solved by ArcEye. (Maybe he will chime in.) But given that it > was the so called "userspace" component, it would be in machinekit-cnc, not > machinekit-hal. But it doesn't look like it is there. > >> > >> The problem with mb2hal is that it presents itself as a non-real-time > component (Modbus is generally considered as a non-RT), but as far as I > know the Modbus RTU can be used in RT, the TCP because of inner working of > TCP communication cannot. > >> > >> Given the "RT" and "userspace" components definition in Machinekit is a > prehistoric vintage from LinuxCNC which outlived it's usefuness, maybe it > is time to have something capable both of RT and non-RT functioning. > >> > >> I am planning something for which I will need the modbus communication, > problem is, I plan in years. > >> > >> Cern. > >> > >> > >> Dec 28, 2019, 02:36 by >> [email protected] <>>> : > >> > >> > Maybe a slight misunderstanding as I just realized mb2hal does exist > in the (legacy?) MK repo but is missing from the mk-hal split repo as is a > couple other user comps. > >> > > >> > I'm not requesting a special lcnc flavored mb2hal, As modbus is a > protocol I would expect a working modbus component to work generically. > What I'm asking is is there a particular reason it's not in the split repo? > If it fails to build for whatever reason I'm sure I'm not the person anyone > wants modifying the code and PR'ing it. I could however help test it. > >> > > >> > On Friday, December 27, 2019 at 6:50:45 PM UTC-5, Michael Brown > wrote: > >> > > >> >> As things are currently I'm certain that if you convert any "simple" > existing mk userspace component and replace the functional code with > >> >> that lcnc modbus component you are requesting, test it, get it to > function as expected and lastly create a PR.. > >> >> Someone eventually will forward the pr request into the repo ... :-) > >> >> > >> >> Best wishes > >> >> Michael Brown > >> >> > >> >> On Friday, December 27, 2019 at 4:46:09 PM UTC+1, justin White > wrote: > >> >> > >> >>> Bump...not alotta interest in modbus in my eh? > >> >>> > >> > > >> > > >> > > >> > -- > >> > website: > >> http://www.machinekit.io>> > blog: > >> > http://blog.machinekit.io>> > github: > >> https://github.com/machinekit>> > > >> > --- > >> > You received this message because you are subscribed to the Google > Groups "Machinekit" group. > >> > To unsubscribe from this group and stop receiving emails from it, > send an email to > >> machi...@>> googlegroups.com <>>> > . > >> > To view this discussion on the web visit > >> > https://groups.google.com/d/>> msgid/machinekit/acc411f1->> > 36e1-49c5-9c5b-22bc7b47e2b6%>> 40googlegroups.com>> <>> > https://groups.google.com/d/>> msgid/machinekit/acc411f1->> > 36e1-49c5-9c5b-22bc7b47e2b6%>> 40googlegroups.com?utm_medium=>> > email&utm_source=footer>> >> . > >> > > >> > >> > > > > > > > > -- > > website: > http://www.machinekit.io> blog: > http://blog.machinekit.io> > github: > https://github.com/machinekit > > --- > > You received this message because you are subscribed to the Google > Groups "Machinekit" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to > [email protected] <javascript:>> . > > To view this discussion on the web visit > > https://groups.google.com/d/msgid/machinekit/c7557316-a094-4fda-ad54-5639be36167c%40googlegroups.com > > < > https://groups.google.com/d/msgid/machinekit/c7557316-a094-4fda-ad54-5639be36167c%40googlegroups.com?utm_medium=email&utm_source=footer>> > > . > > > > -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/9fe74eaa-3e40-4070-a7df-c69a41152010%40googlegroups.com.
