Dne pátek 23. listopadu 2018 8:12:02 UTC+1 Schooner napsal(a): > > Have you actually done what I recommended and purged all your packages, > updated and reinstalled? >
Yes, I did all of that. I will try to install new revision of Debian to be completely sure. > > The current packages work properly on amd64. > I am not disputing that. Given that it started with "Merge pull request #1409 <https://github.com/machinekit/machinekit/pull/1409> from ArcEye/remove-hard-kernel-threads" I would guess that it is specific to i386 platform. And I guess that nobody else is using the old i386 or if they are they are not upgrading it. But I will try it on other i386 PC to be sure. > > If you get those errors when building, your system isn't set up properly > anyway. > I was able to resolve and successfuly build. > > That should have happened on install. > > > On 11/22/2018 10:33 AM, [email protected] <javascript:> wrote: > > I have now booted my test Stretch partition and followed my own advice > > in full. > > > > There were a LOT of updates (227), I think some packages from sid have > > been backported recently, > > noticed a lot of updated packages recently on other machines. > > > > End result, all runs perfectly on amd64 Stretch with latest machinekit > > packages (rt-preempt-0.1.1542196491.git9c3423c-1~stretch) > > > > Cannot reproduce your problems. > > > > regards > > > > > > On 22/11/18 09:11, [email protected] <javascript:> wrote: > >> > >> On 22/11/18 00:59, [email protected] <javascript:> wrote: > >> > >> > Hello, > >> > isn't the newest Machinekit Debian package broken? I upgraded my > >> test system installation with apt-get upgrade (Debian 9 Stretch: > >> Linux machinekit 4.9.0-8-rt-686-pae #1 SMP > PREEMPT RT Debian > >> 4.9.130-2 (2018-10-27) i686 GNU/Linux) to the newest package > >> (0.1.1542196491.git9c3423c-1~stretch), however trying to start > >> machinekit with > >> > sim.axis/axis.ini I got the error: > >> > >> > >> I cannot immediately reproduce this > >> > >> A build of the current repo the package was made from produces a > >> normally running sim ( see below) > >> > >> I don't have a i386, but built on amd64 > >> > >> > >> > Trying to install older version, for example > >> machinekit_0.1.1537365199.gitf63b8a2-1~stretch_i386, solves the > >> problem and simulation normally turns on. > >> > >> That is quite a bit older, at which point does the package fail, does > >> this just affect the last package serial or some others before it? > >> > >> I will try a package install on a spare Stretch partition later. > >> > >> In the interim, can you try completely purging the installed > >> packages, including any dependencies that apt tells you can be > >> removed with > >> 'apt autoremove' because they are only used by machinekit. > >> Then delete the machinekit packages already downloaded and stored in > >> /var/cache/apt/archives > >> (if you don't have any need to retain any old packages and would like > >> the space back, just run 'apt clean' and delete the lot) > >> > >> Then do an 'apt update' and 'apt dist-update'. > >> > >> THEN install machinekit again. > >> > >> That should eliminate mixed version package conflicts etc. > >> > >> > >> If you still have problems, need that dmesg output to see what caused > >> the segfault. > >> core_sim.hal :9 is the main loading of threads, trivkins and tp > >> Nothing has changed regards these, so they are not directly the cause. > >> > > > -- 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]. Visit this group at https://groups.google.com/group/machinekit. For more options, visit https://groups.google.com/d/optout.
