Re: [Machinekit] Re: RPI gipo help

['schoone...@btinternet.com' via Machinekit]

Without precise details of what you are doing, it is impossible to judge.

It is well known that big gcode files will cause Axis to eat all memory 
on small boards when doing a preview, but that is NOT a leak.


Whether that happens more quickly on Stretch, which may well be a 
'heavier' system, I don't know.


You can mitigate it to an extent by disabling previews, has been 
discussed many times on this list.
The only full solution is to use a more capable computer / board or a 
different GUI.


On 12/08/18 14:34, Timothy March wrote:

On RPI using RT kernel with Stretch and it dosen’t matter which RT kernel you 
use. I use Htop to monitor system memory climbs until it is out. This happens 
on RPI and Tinkerboard but not when using Jessie it is a very slow leak

Tim



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: RPI gipo help

['schoone...@btinternet.com' via Machinekit]

Can you please quantify the assertion that Stretch has a memory leak 
with Axis.


I have been using Stretch since it was alpha status.  Usually any memory 
leak is with a program and will manifest irrespective of the distro.



On 8/11/2018 8:43 PM, Timothy March wrote:

I have a question for you are you using Jessie or Stretch? Stretch has a memory 
leak with Axis GUI it will eventually use up all memory. I have post in this 
thread on how to setup Jessie light check it out



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] machinekit-dev broken?

['schoone...@btinternet.com' via Machinekit]

Just checked, the comp binary and includes etc are in the 
machinekit- not the main one.

It was a while back, forgot which.

The Makefile.inc etc are per flavour, to stop the problems we had with 
machinekit-dev always having a posix flavour

because of the build setup for that package.


On 11/08/18 12:22, Charles Steinkuehler wrote:

I've got the latest machinekit package installed, but I don't see a
comp command.  There's an instcomp binary, but it doesn't like the two
*.comp files I'm trying to compile, reporting:

machinekit@beaglebone:~/embedded/beaglebone$
EXTRA_LDFLAGS="-lprussdrv" instcomp --compile raster.comp
Error: Unrecognized file type for mode compile: 'raster.comp'
machinekit@beaglebone:~/embedded/beaglebone$ instcomp --compile
--userspace BBB_listener.comp
Error: Unrecognized file type for mode compile: 'BBB_listener.comp'

On 8/11/2018 1:10 AM, 'schoone...@btinternet.com' via Machinekit wrote:

HI Charles


Yes Comp, plus headers, Makefile.inc and everything else in the -dev
package was moved into the main machinekit package some time ago.

There are only four machinekit-dev packages left in the repo and they
match the also deprecated wheezy and raspian packages in the repo,

which are retained for completeness of support rather than usefullness.


Regards


On 8/10/2018 10:36 PM, Charles Steinkuehler wrote:

I'm trying to migrate to a fresh install on a BBB and am having
problems building some custom hal components.  The first problem is
the comp command is missing, so I figure I'll just install
machinekit-dev package.  But apt-get complains about broken packages
and how I need a different version of the machinekit package than I
have installed.

I checked on machienkit.io and machinekit-dev version is:

machinekit-dev_0.1.1498834902

...while my installed machinekit package (and the most recent
available on machinekit.io) is:

machinekit/machinekit_0.1.1532446277

So...are the machinekit-dev packages broken, or did the comp utility
move somewhere else and machinekit-dev is now deprecated?





--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] machinekit-dev broken?

['schoone...@btinternet.com' via Machinekit]

HI Charles


Yes Comp, plus headers, Makefile.inc and everything else in the -dev 
package was moved into the main machinekit package some time ago.


There are only four machinekit-dev packages left in the repo and they 
match the also deprecated wheezy and raspian packages in the repo,


which are retained for completeness of support rather than usefullness.


Regards


On 8/10/2018 10:36 PM, Charles Steinkuehler wrote:

I'm trying to migrate to a fresh install on a BBB and am having
problems building some custom hal components.  The first problem is
the comp command is missing, so I figure I'll just install
machinekit-dev package.  But apt-get complains about broken packages
and how I need a different version of the machinekit package than I
have installed.

I checked on machienkit.io and machinekit-dev version is:

machinekit-dev_0.1.1498834902

...while my installed machinekit package (and the most recent
available on machinekit.io) is:

machinekit/machinekit_0.1.1532446277

So...are the machinekit-dev packages broken, or did the comp utility
move somewhere else and machinekit-dev is now deprecated?



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] launching machinekit after build

['schoone...@btinternet.com' via Machinekit]

  
  
If you are building from sources please read the documentation

http://www.machinekit.io/docs/developing/machinekit-developing/

Sounds like you probably have not set the pathspec, which is covered
in this document.

On 08/08/18 11:43, mngr wrote:


  
Hi,


I have build machinekit from source, now I am trying to
  launch it.


marco@tmarco:~/machinekit$ ./bin/machinekit 
  ./bin/machinekit: line 1: linuxcnc: command not found

I tried to make a simbolic link to tcl/linuxcnc.so in
  /usr/lib/ but nothing changed
I am sure I am simply noob, and don't know the standard
  procedure... can you please help me?


now I don't have a kernel-rt, but I will only do test, does
  this change anything?



mngr

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] mklauncher "Can not find -sec DISPLAY -var INTRO_GRAPHIC"

['schoone...@btinternet.com' via Machinekit]

On 08/08/18 03:43, Cameron McQuinn wrote:

..was able to successfully start the config several months ago, 
but I had to set the project aside for a while. When I returned, I 
upgraded the kernel from 4.9 to 4.14, and upgraded machinekit and 
QtQuickVcp to the latest versions, but didnt' change anything else.


The error in your title is consequential not causal.

The error is here.
I will leave those who know more about it to pinpoint what caused it.  I 
would tend to suspect the kernel change and wonder if the firmware is 
present or correct.

Aug  8 02:01:24 socfpga rtapi:0: 4:rtapi_app:1890:user hm2_soc_ol: 
soc_program_fpga
Aug  8 02:01:24 socfpga rtapi:0: 4:rtapi_app:1890:user hm2_soc_ol: FPGA overlay 
status: applied
Aug  8 02:01:26 socfpga rtapi:0: 1:rtapi_app:1890:user hm2_soc_ol: failed to 
map hm2-socfpga0 to /dev/uioX
Aug  8 02:01:26 socfpga rtapi:0: 1:rtapi_app:1890:user hm2_soc_ol: 
soc_mmap_fail hm2-socfpga0
Aug  8 02:01:26 socfpga rtapi:0: 1:rtapi_app:1890:user hm2/foo: failed to 
program fpga, aborting hm2_register
Aug  8 02:01:26 socfpga rtapi:0: 1:rtapi_app:1890:user foo: hm2_soc_ol_board 
fails HM2 registration
Aug  8 02:01:26 socfpga rtapi:0: 1:rtapi_app:1890:user hm2_soc_ol: error 
registering UIO driver: -22
Aug  8 02:01:26 socfpga rtapi:0: hal_call_usrfunct(newinst,-- 
config="firmware=zynq/dtbo/replicookie_7z020_ol.dtbo num_pwmgens=5 
num_stepgens=5" debug=1 already_programmed=1) failed: -1 - Operation not permitted
Aug  8 02:01:26 socfpga rtapi:0: 1:rtapi_app:1890:user hal_call_usrfunct(newinst,-- 
config="firmware=zynq/dtbo/replicookie_7z020_ol.dtbo num_pwmgens=5 
num_stepgens=5" debug=1 already_programmed=1) failed: -1 - Operation not permitted
Aug  8 02:01:26 socfpga msgd:0: hal_lib:1904:user halg_exit:293 HAL: removing 
component 99 'halcmd1904'


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

[Machinekit] Machinekit Issue tracker

['schoone...@btinternet.com' via Machinekit]

Dear All,

Some while back I undertook work to split out machinekit into its HAL 
stack and its CNC stack.


This work has reached maturity, to an extent that if the machinekit-hal 
package is installed and then the machinekit-cnc package is installed,

a fully functioning machinekit is available, in tests thus far anyway :)

In preparation for a two package distribution, the existing open Issues 
in machinekit have been copied to either machinekit-hal, or
machinekit-cnc (in a few cases both, with a master issue in 
machinekit-cnc, where they apply to both).


There were 6 exceptions, which were too large to copy over without 
risking data loss and remain in machinekit.
The remaining Open Issues are now Closed, with a reference to the new 
location to be used for future comment.


So, if you have opened, contributed to or just read an Issue in 
machinekit, it can now be found in either

machinekit-hal: https://github.com/machinekit/machinekit-hal/issues
or
machinekit-cnc: https://github.com/machinekit/machinekit-cnc/issues

If you wish to open an issue, please do so in the appropriate repos 
Issues section


If unsure which repo it should come under, generally a look through the 
source files should tell you, there are no duplicates between the repos 
regards actual code.


If still unsure just ask.

If you wish to test the machinekit-hal and / or machinekit-cnc packages, 
please do so, they are available in the main machinekit repo.

Any feedback gratefully received.


regards

--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: DE10-DE0-Nano - Invitation to collaborate

['schoone...@btinternet.com' via Machinekit]

  
  

On 27/05/18 11:26, Michael Brown wrote:


  

  I dont know if you are aware that capitalization matters
the correct filenames are:
  
  
  /lib/firmware/socfpga/DE10_Nano_FB_Cramps.3x24_cap.rbf
  
  /lib/firmware/socfpga/DE10_Nano_SoC_FB_DB25.7I76_7I76_7I76_7I76.rbf
  
   
  But these will first be in the socfpga-rbf.deb package a
few hours after Charles (or someone else) gets around to
merging.
  :-) 
  

  

You requested review by @cdsteinkuehler, but I have no idea as to
his availability at present.

I am happy to merge if you require the files available and we can
revert if there transpires to be a problem.




  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Any interest in Thermocouple support?

['schoone...@btinternet.com' via Machinekit]

  
  
Hi

Yes, we are trying to get the HAL side of Machinekit as generic as
possible, there is a lot more could be done aside from CNC machine
control.

If you fork machinekit, clone to your computer and then create a
suitably named branch and put your driver in
https://github.com/machinekit/machinekit/tree/master/lib/python/drivers,
that should be about right.

Would like to see a document or example config which shows it's use
too.
If a short doc, it can go in the same directory, if a full config,
needs a suitable entry in
https://github.com/machinekit/machinekit/tree/master/configs
exact placement depending upon whether it requires specific machine
/ hardware etc.
Looks like there is unlikely to be a sim, as it should require
hardware present?

The python string on line one, should specify python2.
This is so that distros that ship with python3 as default can build
without problems.

As for licence, GPL2 is fine
A general boiler plate would be


  

  # This
  is a HAL component
  to 




  


  

  # Copyright date
name 



  


  

  #



  


  

  # This program is
  free software; you can redistribute it and/or modify



  


  

  # it under the
  terms of the GNU General Public License as published by



  


  

  # the Free Software
  Foundation; either version 2 of the License, or



  


  

  # (at your option)
  any later version.



  


  

  #



  


  

  # This program is
  distributed in the hope that it will be useful,



  


  

  # but WITHOUT ANY
  WARRANTY; without even the implied warranty of



  


  

  # MERCHANTABILITY
  or FITNESS FOR A PARTICULAR PURPOSE. See the



  


  

  # GNU General
  Public License for more details.



  


  

  #



  


  

  # You should have
  received a copy of the GNU General Public License



  


  

  # along with this
  program; if not, write to the Free Software



  


  

  # Foundation, Inc.,
  59 Temple Place, Suite 330, Boston, MA 02111-1307 USA



  



Then push your new branch back to your fork, and create a pull
request from your fork of machinekit in github, to machinekit /
machinekit
and we will check it, test build and authorise.

If that is beyond your current experience curve, then reply direct
when you have your fork with the relevant files in and I will talk
you through it

regards


On 24/05/18 01:50, Karl Nilsson wrote:


  Hello All,
  A while back, I created a HAL component for interfacing
thermocouple amplifiers over the SPI bus (specifically the
MAX31855 and MAX6675). I was curious whether there was any
interest in including it in machinekit proper, and if so, what
do I need to do in order to get it included. Specifically, is
there any preferred copyright license for user-contributed
components?

Here's the code, feedback is appreciated!

https://github.com/karl-nilsson/MK_config/blob/master/spi_tc.py

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 

Re: [Machinekit] Re: Help requested with Mesa custom 7i80HD-25 bit file compatibility

['schoone...@btinternet.com' via Machinekit]

  
  

On 18/05/18 22:14, fairbanks-john
  wrote:


  
To day I decided that the cause of my problem is most
  likely the version of  gmoccapy available with machinekit. I
  had modified my configuration files to make gmoccapy happy but
  it appears that there are still incompatibilities and do not
  wish to spend more time on.
  


The last update was here
https://github.com/machinekit/machinekit/commit/df4b3b68aa1c9a64e62096350afdf9eb6d09e05c#diff-489f43a022a37a46dc11a435ee80955b
where the author updated to the same state as the then current
Linuxcnc version.

To get this 'up to date' would require someone who was interested in
using it and spending the time removing joints-axes stuff before
merging into machinekit.

I personally don't use it, as it adds nothing in terms of
functionality and visually reminds me of awful early KDE programs,
with their big clunky picture icon push buttons, 
where a simple word to denote its use would suffice.  But that is
just my taste.


  


I used my Mesa 7I80 flash development boards along with HAL
  show, and tested each I/O pin individually. I was able to
  verify that all 72 i/o pins were being read correctly. So my
  linuxcnc Mesa flashed configuration is indeed fully compatible
  with Machinekit. I also wired in several TTL shaft angle
  encoders into the X and Y flashed encoder inputs and verified
  that position data was being read correctly..
  


Good, that all seems to be OK.  There should have been no reason why
a valid bitfile would have rendered the board unusable with
machinekit.

  


The old gmoccapy version is causing to many issues for me
  so I will need to switch to an up to date remote gui for
  testing and further development work. I will start digging for
  information on the best way to do that. The remote Gui will be
  run on a Window 7 box.

 


  


I am confused by your use of 'remote'.  Gmoccapy is not a remote
GUI, but you were using it.  
Are you just referring to accessing the controller machine remotely
from another computer?

Or are you intending to use an actual remote interface via a loader
and machinetalk, like Alex's Machineface or Cetus?


  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Problem with new install errors, warnings and notes

['schoone...@btinternet.com' via Machinekit]

  
  

On 17/05/18 09:45, fairbanks-john
  wrote:


  
Schooner I think that I pushed a wrong buttons while
  posting and messed things up a bit. I got a new computer and
  build a new configuration since I think the old machine may
  have gotten corrupted while attempting to solve issues. I will
  rebuild it later. Machine kit loadsrand runs now but I'm
  confused on how machinekit handles Mesa bit files. I have read
  that they need to be placed in a folder somewhere as a txt
  file which confuses me more since a bit file that is loaded
  into a Mesa FPGA board is not a text file.
  


Without seeing what you read, hard to comment.  All the docs come
from Linuxcnc and have not been changed.
Mesa cards are re-flashed with mesaflash if needs be, to change the
config. 
hm2_pci just needs to be passed the number of stepgens etc in the
config string.

You might be getting mixed up with the DE0-NANO-Soc FPGA boards,
which take a config string which specifies a .dtbo file on a given
path.


  


Here are the steps I used to install this new installation
  using Debian 9.4.0 stretch.


# Desktop computer with run in place
  $ installed Debian 9.4.0 amd64 stretch cd1 free only version
  $ su usermod -a -G sudo "my user name"
  $ sudo apt-key adv --keyserver hkp://pool.sks-keyservers.net
  --recv 43DDF224
  $ sudo sh -c "echo 'deb http://deb.machinekit.io/debian
  stretch main' > /etc/apt/sources.list.d/machinekit.list
  $ sudo apt-get update
  $ sudo apt-get install linux-image-rt-amd64
  $ reboot
  $ cd /home
  $ git clone https://github.com/machinekit/machinekit.git
  machinekit-dev
    
  ===
  # in machinekit-dev folder
  $ debian/configure -r
  $ sudo mk-build-deps -ir
  # say yes to install depends
  $ sudo apt-get install dirmngr
  $ sudo apt-get install git dpkg-dev 
  $ sudo apt-get install devscripts
  $ sudo apt-get install python-avahi 
  $ sudo apt-get install python-dbus python-dbus-dev
  python-dbus-dbg
  # some in the above line may not be required
  $ sudo apt-get install python-glade2 python-gtkglext1
  python-gtksourceview2
  $ sudo apt-get install python-vte
  $ sudo apt-get install --no-install-recommends devscripts
  equivs
  $ sudo apt-get install memlockd  # may not be required
  # the above line may not be required
  $ cd src
  $ ./autogen.sh
  $ ./configure --with-rt-preempt --enable-build-documentation
  $ make clean  (only if needed to completely make fresh
  version)
  $ make
  $ sudo make setuid
  $ ../scripts/check-system-configuration.sh
  $ sudo cp rtapi/rsyslogd-linuxcnc.conf
  /etc/rsyslog.d/linuxcnc.conf
  $ sudo service rsyslog restart
  $ sudo cp rtapi/shmdrv/limits.d-machinekit.conf
  /etc/security/limits.d/machinekit.conf
  $ sudo touch /var/log/linuxcnc.log
  $ sudo service rsyslog restart
  $ cd ..
  $ . ./scripts/rip-environment
  $ export DEBUG=5
  $ machinekit /home/cnc/machinekit-dev/configs/cnc1/cnc1.ini
  
  # gmoccapy gui comes up ok but its  version is too old to be
  usefull
  

  


Looks like you are mostly there.
Don't know where the path for the protobuf files came from in the
last print, maybe you had added it to the PATH at some point?

Gmoccapy has not been updated for a while and frankly very few
people use it.
It starts hitting problems when you get to Buster and when packages
for it get backported to Stretch.
I had to disable the attempts by it to play audio files in Buster,
because the library it relies upon has been deprecated.

Anything based upon python2 is probably on its last legs, despite
the python.orgs spectacular own goal in coming up with 
a non backwardly compatible python3 and expecting everyone to just
adopt it.

One of the major widget libs used in pyvcp is due for deprecation
shortly too, which is going to force a major re-think on GUIs
both here and in Linuxcnc.



  


  
  -- 
  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 

Re: [Machinekit] Can't install machine kit on Beagle Bone stock image

['schoone...@btinternet.com' via Machinekit]

  
  

On 16/05/18 16:39, Eric wrote:


  I'm guessing that updating the kernel on the
Machinekit image isn't as trivial as installing an app? (can you
tell I'm new to Linux! :)


I don't think you have ever said what image you have, so no idea
what kernel is in it

If it is the 3.8 Xenomai one, then no you can't update it.
Later kernels use Xenomai 3 and Machinekit does not support it

You will need to use a rt-preempt > 4.1.x based image to get
module support for that Texas Instruments chip.
The 4.4.113-ti-r149 kernel seems to be the favorite, but there are
other later kernels too.

There are several images marked 'all Beagle Bone variants' and 'BBB
and BBW', but you need advice as to which one to use.
Make sure you search the forum posts, several people have gone down
this route before.

Failing that, RCN is the expert, I don't have any BBx boards or
capes and s**t  :-P 


  


Where would I find Robert Nelson?

  


http://www.machinekit.io/docs/getting-started/machinekit-images/


  

  On Tuesday, May 15, 2018 at 10:43:13 AM UTC-4, Schooner wrote:
  
 
  On 14/05/18 19:10, Eric wrote:
  
  
 BeagleBone wireless, including the wifi
  chip: http://octavosystems.com/2016/09/27/the-beaglebone-black-wireless-and-beagleboard-compatibility/
  
  
  Getting wifi up and running on an image I didn't
create (and admittedly don't fully understand yet)
seemed like a pretty hard task. 
  

  
  
  Looks like you need the firmware-ti-connectivity
  package, which is backported all the way back to Jessie
  now
  https://packages.debian.org/jessie-backports/firmware-ti-connectivity
  
  Needs a 4.1 > kernel by the looks of it though, to get
  the kernel module, so if the image you tried uses the 3.8
  xenomai kernel or similar, that is part of the problem
  
  All the information is out there, but I would suggest you
  contact Robert Nelson who make the images, he will know
  which one you need.
  
  
  

  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Problem with new install errors, warnings and notes

['schoone...@btinternet.com' via Machinekit]

On 15/05/18 08:29, fairbanks-john wrote:
I just made a fresh install from the git hub clone today and wondering 
if I did something wrong since there are a lot of warnings and notes 
in the installation log which I have attached..
 The .hal file I have used before for testing fails to complete 
loading at the below error message:




This bit of the print clearly shows the problem is with your config, 
which we have no details of


/home/cnc/machinekit-dev/configs/cnc1/cnc1.ini:81: executing 'import sys
sys.path.insert(0,"python")'
PythonPlugin: Python  '2.7.12 (default, Dec  4 2017, 14:50:18)
[GCC 5.4.0 20160609]'
emcTaskInit: using builtin interpreter
/home/cnc/machinekit-dev/configs/cnc1/cnc1.ini:81: executing 'import sys
sys.path.insert(0,"python")'
PythonPlugin: Python  '2.7.12 (default, Dec  4 2017, 14:50:18)
[GCC 5.4.0 20160609]'
is_callable(__init__) = FALSE
is_callable(oword.on_abort) = FALSE
is_callable(oword.on_abort) = FALSE
/home/cnc/machinekit-dev/configs/cnc1/cnc1.ini:81: executing 'import sys
sys.path.insert(0,"python")'
PythonPlugin: Python  '2.7.12 (default, Dec  4 2017, 14:50:18)
[GCC 5.4.0 20160609]'
Traceback (most recent call last):
  File "/home/cnc/machinekit-dev/bin/gmoccapy", line 30, in 
import hal_glib   # needed to make our own hal pins
  File "/home/cnc/machinekit-dev/lib/python/hal_glib.py", line 20, in 
from machinetalk.protobuf.message_pb2 import Container
  File "build/bdist.linux-x86_64/egg/machinetalk/protobuf/message_pb2.py", line 16, 
in 
  File "build/bdist.linux-x86_64/egg/machinetalk/protobuf/nanopb_pb2.py", line 26, in 

TypeError: __init__() got an unexpected keyword argument 'syntax'
Shutting down and cleaning up Machinekit...

Run an axis_mm sim, and I expect it will run perfectly, the problem is 
with your config / gmoccapy / python


syntax='proto2' has been in there for 2 years
https://github.com/machinekit/machinekit/blob/master/src/machinetalk/proto/src/machinetalk/protobuf/nanopb.proto#L8

but I don't use python, protobuf or gmoccapy, so can't help there, other 
than wonder why your path to the machinetalk/protobuf files is 
completely different

They should be in

/home/cnc/machinekit-dev/lib/python/machinetalk/protobuf

Are you using files from another build?


Warning:  Logging appears to be not working

  Found rsyslogd configuration and /var/log/linuxcnc.log
  exists, but a test did not appear in the log.
  Please investigate.

Makefile:842: recipe for target 'setuid' failed


This occurs as it says.  Sometimes the rsyslogd daemon has died, the 
linuxcnc.log has the wrong permissions and cannot be written to, or similar.
You also do need to check that the setuid part worked, but there is no 
indication it did not when realtime was started



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Can't install machine kit on Beagle Bone stock image

['schoone...@btinternet.com' via Machinekit]

  
  

On 14/05/18 19:10, Eric wrote:


   BeagleBone wireless, including the wifi
chip: http://octavosystems.com/2016/09/27/the-beaglebone-black-wireless-and-beagleboard-compatibility/


Getting wifi up and running on an image I didn't create
  (and admittedly don't fully understand yet) seemed like a
  pretty hard task. 

  


Looks like you need the firmware-ti-connectivity package,
which is backported all the way back to Jessie now
https://packages.debian.org/jessie-backports/firmware-ti-connectivity

Needs a 4.1 > kernel by the looks of it though, to get the kernel
module, so if the image you tried uses the 3.8 xenomai kernel or
similar, that is part of the problem

All the information is out there, but I would suggest you contact
Robert Nelson who make the images, he will know which one you need.



  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Can't install machine kit on Beagle Bone stock image

['schoone...@btinternet.com' via Machinekit]

  
  

On 14/05/18 12:39, Eric wrote:


  Ahh, ok so that got me farther, but now I'm
getting a "no keyserver available" error... 
  


Unless you detail what you did and exactly what the error was,
no-one will be able to help

If you search for gpg 'no keyserver available' errors you will see
there can be a variety of reasons, from 
the dirmngr program having problems with the DNS server you use,  to
trying to do this behind a proxy.

I think you are going about this all wrong.

Instead of getting your wireless NIC recognised and module and
firmware installed for it in the Machinekit image,
(which incidently we don't produce and queries about it would best
addressed to RCN's site where you probably downloaded it from)
you are now trying to install Machinekit into an image which does
not have a RT kernel, so won't run Machinekit as is.

Wifi modules are not built in to kernels normally, they are loaded
as required.
Some of it is non-free and not in distributions for legal reasons,
but it is available all the same.

The first action, identify your wifi chipset
Internet search will 
find
https://www.cyberciti.biz/faq/linux-find-wireless-driver-chipset/ 
for example


  
On Monday, May 14, 2018 at 1:33:23 AM UTC-4, Bas de Bruijn
wrote:

  
  
  > On 14 May 2018, at 01:29, Eric 
  wrote:
  
  > 
  > Ok, so if you've seen my other posts you're probably
  figuring out that I'm new to pretty new to Linux. I've fooled
  around with Pis so I have some base knowledge, but that's
  about it.
  
  > 
  > I'm trying to use a stock BeagleBone wireless image that
  I'm running. The image I have on it is this: https://beagleboard.org/latest-images
  (Stretch LXQT)
  
  > 
  > I've been trying to use these commands to install
  MachineKit:
  
  > 
  > http://www.machinekit.io/docs/getting-started/APT-packages-stretch/
  
  > 
  > I keep getting "failed to start the dirmngr
  '/tmp/apt-key-gpdhome.ck58yIIcSL/S.dirmngr' failed: No
  such file or directory"
  
  > 
  > What's going on here?
  
  
  Use:
  
  Sudo apt update
  
  Sudo apt install dirmngr
  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Docker upgrade failing

['schoone...@btinternet.com' via Machinekit]

  
  
> Apr 27 11:13:30 D525MW dockerd[4167]: Error starting daemon:
Error initializing network controller: Error creating default…e
network
The print is truncated, there should be a more detailed error
explanation following a colon after the second ' Error'

eg.
Error starting daemon: Error
initializing network controller: could not delete the default
bridge network: network bridge has active endpoints

Have a look here, there are several people with similar issues and
some resolutions
https://github.com/moby/moby/issues/18283

Otherwise try searching with the full error string


On 27/04/18 22:13, mugginsac wrote:


  Schooner,

I uninstalled 18 and tried installing 17.0.3 and it failed too.
Must be something wierd with my setup.

Here is the screen from that effort. This was booted in to a
non-realtime kernel (4.9.0.6 AMD64).

acondit@D525MW:~$ sudo apt-get install
docker-ce=17.03.2~ce-0~debian-stretch
Reading package lists... Done
Building dependency tree   
Reading state information... Done
The following package was automatically installed and is no
longer required:
  pigz
Use 'sudo apt autoremove' to remove it.
The following NEW packages will be installed:
  docker-ce
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 21.3 MB of archives.
After this operation, 105 MB of additional disk space will be
used.
Get:1 https://download.docker.com/linux/debian stretch/stable
amd64 docker-ce amd64 17.03.2~ce-0~debian-stretch [21.3 MB]
Fetched 21.3 MB in 5s (4,250 kB/s)    
Selecting previously unselected package docker-ce.
(Reading database ... 325458 files and directories currently
installed.)
Preparing to unpack
.../docker-ce_17.03.2~ce-0~debian-stretch_amd64.deb ...
Unpacking docker-ce (17.03.2~ce-0~debian-stretch) ...
Setting up docker-ce (17.03.2~ce-0~debian-stretch) ...
Created symlink
/etc/systemd/system/multi-user.target.wants/docker.service →
/lib/systemd/system/docker.service.
Created symlink
/etc/systemd/system/sockets.target.wants/docker.socket →
/lib/systemd/system/docker.socket.
insserv: warning: script 'realtime' missing LSB tags and
overrides
Job for docker.service failed because the control process exited
with error code.
See "systemctl status docker.service" and "journalctl -xe" for
details.
invoke-rc.d: initscript docker, action "start" failed.
● docker.service - Docker Application Container Engine
   Loaded: loaded (/lib/systemd/system/docker.service; enabled;
vendor preset: enabled)
  Drop-In: /etc/systemd/system/docker.service.d
   └─override.conf
   Active: failed (Result: exit-code) since Fri 2018-04-27
11:13:30 PDT; 37ms ago
 Docs: https://docs.docker.com
  Process: 4167 ExecStart=/usr/bin/dockerd -H tcp://0.0.0.0:2375
(code=exited, status=1/FAILURE)
 Main PID: 4167 (code=exited, status=1/FAILURE)
  CPU: 515ms

Apr 27 11:13:30 D525MW dockerd[4167]:
time="2018-04-27T11:13:30.838786754-07:00" level=debug
msg="ReleaseAddress(LocalDefa….17.0.1)"
Apr 27 11:13:30 D525MW dockerd[4167]:
time="2018-04-27T11:13:30.838859527-07:00" level=debug
msg="ReleasePool(LocalDefault….0.0/16)"
Apr 27 11:13:30 D525MW dockerd[4167]:
time="2018-04-27T11:13:30.838944383-07:00" level=debug
msg="start clean shutdown of …meout..."
Apr 27 11:13:30 D525MW dockerd[4167]:
time="2018-04-27T11:13:30.878010057-07:00" level=debug
msg="Cleaning up old mountid : start."
Apr 27 11:13:30 D525MW dockerd[4167]:
time="2018-04-27T11:13:30.879990648-07:00" level=debug
msg="Cleaning up old mountid : done."
Apr 27 11:13:30 D525MW dockerd[4167]: Error starting daemon:
Error initializing network controller: Error creating default…e
network
Apr 27 11:13:30 D525MW systemd[1]: docker.service: Main process
exited, code=exited, status=1/FAILURE
Apr 27 11:13:30 D525MW systemd[1]: Failed to start Docker
Application Container Engine.
Apr 27 11:13:30 D525MW systemd[1]: docker.service: Unit entered
failed state.
Apr 27 11:13:30 D525MW systemd[1]: docker.service: Failed with
result 'exit-code'.
Hint: Some lines were ellipsized, use -l to show in full.
dpkg: error processing package docker-ce (--configure):
 subprocess installed post-installation script returned error
exit status 1
Processing triggers for 

Re: [Machinekit] Docker upgrade failing

['schoone...@btinternet.com' via Machinekit]

I am running v18.03 on an amd64 Buster desktop and v17.03 on my amd64 
Stretch server without problems.


As Charles says, it will not run on other than a POSIX kernel, had 
problems in the past when I booted into the RT

kernel without thinking about it.

There may be something else in the wind though, this thread is recent 
and unresolved

https://github.com/moby/moby/issues/36804

Worth checking the changelogs to see if they have updated without 
changing major version number recently?


The obvious first advice is remove it and install your previous version 
again via dpkg

Don't think there is anything magic about v18

On 26/04/18 23:02, Charles Steinkuehler wrote:

There's not really enough to tell exactly what went wrong, but IIRC I
ran into similar problems trying to get Docker running on one of my
machines I was also using for CNC tests.

It turns out my problem was with the kernel version or configuration.
The real-time kernel I was using wouldn't work with Docker and the
daemon would just refuse to start.  Switching back to a stock Debian
kernel got everything working as expected.

On 4/26/2018 4:50 PM, mugginsac wrote:

I have spent a couple days trying to get docker to run. I keep getting a
failure on the installation. Can anybody help?

acondit@D525MW:~$ sudo apt-get dist-upgrade
Reading package lists... Done
Building dependency tree
Reading state information... Done
Calculating upgrade... Done
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
1 not fully installed or removed.
After this operation, 0 B of additional disk space will be used.
Do you want to continue? [Y/n]
Setting up docker-ce (18.03.0~ce-0~debian) ...
insserv: warning: script 'realtime' missing LSB tags and overrides
Job for docker.service failed because the control process exited with error
code.
See "systemctl status docker.service" and "journalctl -xe" for details.
invoke-rc.d: initscript docker, action "start" failed.
● docker.service - Docker Application Container Engine
Loaded: loaded (/lib/systemd/system/docker.service; enabled; vendor
preset: enabled)
   Drop-In: /etc/systemd/system/docker.service.d
└─override.conf
Active: activating (auto-restart) (Result: exit-code) since Thu
2018-04-26 14:40:38 PDT; 66ms ago
  Docs: https://docs.docker.com
   Process: 9301 ExecStart=/usr/bin/dockerd -H tcp://0.0.0.0:2375
(code=exited, status=1/FAILURE)
  Main PID: 9301 (code=exited, status=1/FAILURE)
   CPU: 340ms

Apr 26 14:40:38 D525MW systemd[1]: Failed to start Docker Application
Container Engine.
Apr 26 14:40:38 D525MW systemd[1]: docker.service: Unit entered failed
state.
Apr 26 14:40:38 D525MW systemd[1]: docker.service: Failed with result
'exit-code'.
dpkg: error processing package docker-ce (--configure):
  subprocess installed post-installation script returned error exit status 1
Errors were encountered while processing:
  docker-ce
E: Sub-process /usr/bin/dpkg returned an error code (1)
acondit@D525MW:~$ sudo systemctl status docker.service
● docker.service - Docker Application Container Engine
Loaded: loaded (/lib/systemd/system/docker.service; enabled; vendor
preset: enabled)
   Drop-In: /etc/systemd/system/docker.service.d
└─override.conf
Active: failed (Result: exit-code) since Thu 2018-04-26 14:40:40 PDT;
42s ago
  Docs: https://docs.docker.com
   Process: 9394 ExecStart=/usr/bin/dockerd -H tcp://0.0.0.0:2375
(code=exited, status=1/FAILURE)
  Main PID: 9394 (code=exited, status=1/FAILURE)
   CPU: 338ms






--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] switching between kinematics modes in G code

['schoone...@btinternet.com' via Machinekit]

  
  
Have a look at rs274ngc_interp.hh:L233

The functions starting comp_get_current(...) appear
to get and set the values in the initial settings struct.
Initialised in rs274ngc_pre:L113 and struct detail in
interp_setup.cc / interp_internal.hh

I haven't looked much at it, but assuming this struct is updated
throughout, to hold the current values for all the settings, it may
be of use.

If you have a sim which can switch kinematic modes using your
components, I may be able to assist more.

I don't have any machines that use anything other than trivkins.

On 24/04/18 14:12,
  j...@allwinedesigns.com wrote:


  I believe I found the culprit. While everything
seems to be in order in the motion controller, I believe the
GCode interpreter still believes it's at the previous position
after a kinematics mode change. Manually executing a g0 to the
currently displayed coordinates after a change and then
executing another command seems to work perfectly. Now to find
where to change that...
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] MESA 7i80 hm2_eth unable to read static internal configuration

['schoone...@btinternet.com' via Machinekit]

  
  

On 21/04/18 09:29, fairbanks-john
  wrote:


  I'm finally continuing my work on building a new
gui and switching this in progress work from linuxcnc to
machinekit. I'm using my ACRA milling machine as a test bed
which I do need to be able to use normally while this work
progresses. It runs on linuxcnc 2.8pre with gmoccapy. 

The interface board between the machine is a Mesa 7i80hd board
with a custom pre-reloaded and stored bit file file. Linuxcnc
has no problem reading the internal programming of the Mesa
board and runs mill just fine. Machine kit on the other hand
appears to be a little confused as to handle the programming of
the board.  This to me is strange since the driver that was
developed for this board is known to very reliable. So I suspect
that machine kit may read and write to this card in a slightly
different manner.

There does appear to be some differences in the hm_eth driver 
files used by machinekit and linuxcnc which could be causing a
incompatibility issue. As I recall coping the hm_eth driver .c
and .h files and recompiling causes a make to fail. 
  


There are major differences between the hm2_eth files in Linuxcnc
and those in Machinekit

At the most trivial basic level, it uses rtapi_xxx types which are
just redefinitions of standard types, but sufficient to prevent
compilation.
Getting past those, there are major incompatibilities.

Just a glance at hm2_eth.h, shows that there is a big hm2_eth_t
struct that did not exist before.
The hm2_eth.c file is 70% larger, partly as a consequence.

Errors indicate that hm2_lowlevel_io_t has changed markedly
too and within this and elsewhere, there are other functions used
which only exist in Linuxcnc,
such as need_soft_reboot(), rtapi_spawn_as_root().

If your 7i80 with its custom config works with Linuxcnc, I would use
use that.

Other people may be using 7i80s and be able to see why the existing
driver does not like your bitfile, I don't own one, so cannot test
anything.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit as a general purpose motor controller?

['schoone...@btinternet.com' via Machinekit]

Please see also https://github.com/machinekit/machinekit-hal

This is a standalone Machinekit HAL stack without interpreter, motion, 
NML etc etc.


At present the python vcp stuff is in the corresponding CNC stack at
https://github.com/machinekit/machinekit-cnc

If you need virtual panels etc for your project they could be moved 
across to machinekit-hal.
Their current location is simply because the primary mainstream use is 
within the Axis GUI,

but they can be wherever is most useful.

On 20/04/18 14:29, Charles Steinkuehler wrote:

On 4/19/2018 10:02 PM, dschr...@eng.ucsd.edu wrote:

Hi! I was curious about people using machinekit as a general purpose motor
controller? I have been developing my own 8 axis motor driver on the
DEO-NANO-SOC, but this seems to have many advanced features.

You could easily use the Machinekit HAL framework and the hostmot2 VHDL
to control whatever mix of hardware you need.  Assuming you don't want
to use the existing CNC stack, you should be able to fairly easily
connect your high-level planning software up to HAL using the C or
Python bindings.  Note that the ring-buffers available in Machinekit
make interfacing to non-real-time user-mode code a *LOT* easier than it
used to be in LinuxCNC.

Depending on exactly what you need to do, you may not even need anything
outside of HAL and a few limit2/limit3 blocks.  I'm controlling a couple
custom (not CNC, not 3D printer) machines using just Machinekit HAL and
a couple small HAL components I wrote for specific tasks.



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] switching between kinematics modes in G code

['schoone...@btinternet.com' via Machinekit]

  
  

On 12/04/18 16:00,
  j...@allwinedesigns.com wrote:


  I definitely have to stick with Machinekit as I'm
running the machine with a Beaglebone. What is the nature of the
changes made to LinuxCNC that allowed this to work? Do you think
I could get this going on Machinekit?
  


It is completely non-trivial and I am far from sure if it exists as
a series of commits which are easily accessible.
Linuxcnc did not have a Github repo until recently and I don't know
how they converted across.

These are just the commits that actually specify 'JA' in the title
(238 of them)
https://github.com/LinuxCNC/linuxcnc/search?q=JA=Commits

If the commit was relevant but did not have that tag in the commit
text, the search would not find it.
It may also have picked up 'jam' and the likes, Github
searches are not very good.

You are welcome to try cherry-picking the commits and resolving all
the major conflicts occurring because machinekit has completely
diverged from linuxcnc in many respects.  I would not recommend it.

If I were you I would get an old computer with a parport, install
Linuxcnc from a rtai kernel distro.
Then you can find out if it actually works as you want.

If it does, why do you need to use a BBB?

I don't have an axe to grind either way, whatever works best IS
best.


  
On Thursday, April 12, 2018 at 8:28:29 AM UTC-6, Schooner wrote:

   
On 12/04/18 14:59, jo...@allwinedesigns.com
  wrote:


  Thanks for the advice!


I started implementing this the other day and am
  running into follow errors when switching between
  modes using a hal pin. 
  


Yes I did predict that it would likely cause problems

  "The actual instant of changeover could reek havoc if it
  occurred midway through setting the joints say, so that
  would need to be catered for with a function
 that acted upon the changeover pin changing state,
  clearing old data and establishing new defaults or
  whatever."
  
You might be able to get around it by forcing
the position_fb to whatever the actual position changes to,
perhaps by unlinking position_fb at the right point
then re-linking,
with it set to the same as the actual position after the
move.


  
I found this thread on the LinuxCNC forums which
  discusses the same desired behavior and the same
  errors I'm encountering. It seems like they were able
  to resolve the issues, so I'll give what they tried a
  shot: https://forum.linuxcnc.org/10-advanced-configuration/31813-tcp-5-axis-kinematics?limitstart=0

  


We don't have the joints / axes work backported into
Machinekit, it caused a lot of problems in Linuxcnc
initially, so interesting that it seems to help this
succeed.

Hopefully it will work for you too.


  

  On Thursday, April 12, 2018 at 2:48:29 AM UTC-6,
  Schooner wrote:
  
 
  On 12/04/18 09:29, 'schoo...@btinternet.com'
via Machinekit wrote:
  
   

This code snippet from my lathe pulley speed
setting script shows the process to convert and
use case / esac to switch the input to the
correct call


  
  
  Just re-read and that gives the impression you
  need to use a case / esac switch
  
  All you need of course is a single call in place
  of the switch, as P contains the value to pass
  
  eg.    halcmd setp combokins.state $int
  
   



On 11/04/18 21:30, jo...@allwinedesigns.com
  wrote:


  Is there a way to switch
between kinematics modules using G code in
   

Re: [Machinekit] switching between kinematics modes in G code

['schoone...@btinternet.com' via Machinekit]

  
  

On 12/04/18 09:29,
  'schoone...@btinternet.com' via Machinekit wrote:


  
  
  
  This code snippet from my lathe pulley speed setting script shows
  the process to convert and use case / esac to switch the input to
  the correct call
  
  


Just re-read and that gives the impression you need to use a case /
esac switch

All you need of course is a single call in place of the switch, as P
contains the value to pass

eg.    halcmd setp combokins.state $int


  
  
  
  
  On 11/04/18 21:30, j...@allwinedesigns.com
wrote:
  
  
Is there a way to switch between kinematics
  modules using G code in Machinekit? The desired effect is to
  be able to switch between trivkins and a 5 axis inverse
  kinematics module that allows for TCPC (Haas uses G234 to
  enable TCPC, see https://www.youtube.com/watch?v=HxPjH4v5iEg).
  I've implemented the 5 axis kinematics module for my machine.
  I was thinking I could incorporate the trivial kinematics into
  the same module and add a hal pin that would switch between
  the two. Then I could write a python script that flips the hal
  pin, and bind that to a custom G code. Would that work? Is
  there another way to do it?
-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] switching between kinematics modes in G code

['schoone...@btinternet.com' via Machinekit]

  
  
Interesting concept.

In theory there should be nothing to prevent the combination of two
kinematics types
and a conditional test based upon whether a pin is set, to decide
which is used.

Implementation might be a little more complicated.  
The data structures would need combining so that any data storage
required was available irrespective of which mode it started in.
The actual instant of changeover could reek havoc if it occurred
midway through setting the joints say, so that would need to be
catered for with a function
that acted upon the changeover pin changing state, clearing old data
and establishing new defaults or whatever.

Regards activating, that is the simple bit.
Create a bit pin, called say combokins.state
Then just create a user M code which is a bash script which calls
`setp combokins.state 1` or 0 as required
The 1 or 0 can be passed in the first arg (P) to the M code, just
remember it will be passed as fp and needs conversion to an integer.

This code snippet from my lathe pulley speed setting script shows
the process to convert and use case / esac to switch the input to
the correct call

Like to hear how it goes.

~

#!/bin/bash

## sets scale gain to make output speed accurate
## for each of 4 pulley settings

if [ ! $# -ge 1 ]; then
  echo "Usage: M101 n - where n is speed range 1 - 4"
  exit 1
fi

echo "param 2 is $2"
## emc takes the M101 P3 for example and passes 
## $1 as 3.00 and $2 as -1.00 (because Q was not set)
## to this script!!!
## need to convert to int first

float=$1
int=${float/\.*}
# DEBUG    echo "$float converted to $int"


case $int in

    1 ) halcmd setp scale.0.gain 0.595;;
#   accurate at 1200 in M4 for turning

    2 ) halcmd setp scale.0.gain 1.16;;
#   accurate at 600 in M4 for turning

    3 ) halcmd setp scale.0.gain 2.175;;
#   accurate at 300 in M3 for threading

    4 ) halcmd setp scale.0.gain 4.3;;
#   accurate at 150 in M3 for threading
esac

exit 0




On 11/04/18 21:30,
  j...@allwinedesigns.com wrote:


  Is there a way to switch between kinematics modules
using G code in Machinekit? The desired effect is to be able to
switch between trivkins and a 5 axis inverse kinematics module
that allows for TCPC (Haas uses G234 to enable TCPC,
see https://www.youtube.com/watch?v=HxPjH4v5iEg). I've
implemented the 5 axis kinematics module for my machine. I was
thinking I could incorporate the trivial kinematics into the
same module and add a hal pin that would switch between the two.
Then I could write a python script that flips the hal pin, and
bind that to a custom G code. Would that work? Is there another
way to do it?
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: DE10-DE0-Nano - Invitation to collaborate

['schoone...@btinternet.com' via Machinekit]

  
  
Well done Michael,

Just tested it and works fine on a DE0-NANO too.

If you have no objections I will add them to the images at
http://deb.machinekit.io/uploads/de0-nano/

Notes for anyone looking to test.

Instructions are in the file readme-console-image.md
DO use bmaptool to write to SD card.  dd does not seem to cope too
well, but bmap works fine.

I used the console image and once the ethaddr etc was set etc. just
needed to log in;
$ sudo apt update
$ sudo apt install machinekit-rt-preempt machinekit socfpga-rbf
and I was up and running the socfpga-stepper sim, easy as that!



On 07/04/18 00:56, Michael Brown wrote:


  Just a FYI that I have released 2 new stretch based
images that are much more stable than the previous Jessie ones.


U-boot 2018.01,  4.9.76-ltsi-rt kernel


Also one is a desktop image running LXQT with kwin and even
  though there is some latency in loading, I find the windows to
  be much more responsive
with instantainios drag around etc.
And axis seems to run quite well with preview (see video
  demo).


   
https://github.com/the-snowwhite/soc-image-buildscripts/releases/tag/Release_2.0



  On Tuesday, 29 August 2017 15:05:32 UTC+2, Michael Brown
  wrote:
  

  

  

  Michael Brown has invited you to contribute
  to the following shared folder:
  


  DE10-DE0-Nano
  
  DE10-SoC
  Machinekit demo image with framebuffer
  This image also works with the Atlas
  (DE0-Nano-Soc) board(tested)
  Open

  
  

  

  

  






  
  






  
  






  

  




  

  






  
  






  
  






  

  

  

  
  

  
Google
  Drive: Have all your files within reach from
  any device. 
  Google Inc. 1600 Amphitheatre Parkway,
  Mountain View, CA 94043, USA

  

  

  

  

  
  -- 
  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 

Re: [Machinekit] LinuxCNC and BeagleBone Black - a road well travelled?

['schoone...@btinternet.com' via Machinekit]

  
  

On 05/04/18 10:07, 'fishy' via
  Machinekit wrote:


  

On Wednesday, 4 April 2018 16:41:46 UTC+1, Schooner wrote:

  
  
  It is a community project, but thus far everyone (with minimal
  
  exceptions) who has stood where you are and bemoaned the lack
  of 
  detailed personalised instructions for their
  
  hardware combination of choice, has subsequently failed to
  contribute 
  anything to make it easier for the next person to tread that
  path.
  
  
  If you were to break that trend, I would be delighted.
  
  



I am not sure if that is what was described by "tragedy of
  the commons" ?
  


I think it is, certainly it is a common trend.


  


I guess most open source projects get less than 1% of users
  actually contributing something back to the project?


I wonder if some autodoc system could be made to force
  people to contribute in order to get support from a real
  person? :lol


Unfortunately 99% of people are here to get free cnc
  software installed for zero cost, they are not interested in
  playing with software development as a hobby.
  


Yes and as someone who would not touch Microsoft or Apple with a
bargepole, it does gall somewhat.


  


As a start does machinekit have a easy access wiki system?
  (http://www.machinekit.io/community/contributing/)
  


Wiki's unfortunately are for people who commit a mind dump to a page
and then move on, leaving it orphaned and unmaintained.

Linuxcnc's documentation is full of wikis which refer to Ubuntu
Hardy and other long since deceased distros and you pretty
much need to know enough not to need the wiki in the first place, to
sort the wheat from the chaff.

  


Of course this forum may not be the best place to find
  linuxcnc help, as it is actually  here for discussion of
  machinekit, so maybe the original poster should start by
  updating the wiki
  (http://wiki.linuxcnc.org/cgi-bin/wiki.pl?BeagleBone) with all
  the information they have used and the steps they took to
  install machinekit on the beaglebone.  Then maybe some of the
  experts here could take time away from their most important
  work to review and analyse the problem?
  


If you read the wiki, it was originally largely written by Michael
Haberler, before he left Linuxcnc to found Machinekit.
It is well out of date.

It was the refusal of Linuxcnc to incorporate changes that would
allow ARM boards to run Xenomai and rt-preempt kernels so as to make
the use of an ARM board
possible, amongst many other changes, that led to the split and fork
to Machinekit.
They have since seamlessly incorporated some of the changes, as if
they had come up with them ;-) 

If someone has something to contribute and refuses to use the tools
provided because some perceived difficulty or other bias, I don't
care.

If they create a wiki and provide a link to it, we can cut and paste
the text into a proper asciidoc and incorporate it into the website.
When they move to other things, the wiki will wither and die from
lack of relevance and maintenance, but we will have a document that
could be updated as required.

There's the challenge, Word document, wiki, compilation cut and
paste from a number of forum posts (all of which can be searched
from the website anyway),
or any other format.

If it is of use we will use it.



  


  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit 

Re: [Machinekit] LinuxCNC and BeagleBone Black - a road well travelled?

['schoone...@btinternet.com' via Machinekit]

On 28 Mar 2018, at 22:02, Rick Mann  wrote:

I'm just starting on building a CNC router using MachineKit and a BBB. You'll 
still need a board of some kind, if for no other reason than to isolate the BBB 
from the rest of your system. I bought an OPTOCAPE ($165) to experiment with, 
but I will be designing and building my own board at the same time (mostly 
because I want all the right connectors on it).


The big problem to get across is that Machinekit is nothing to do with BBB.

The fact that you can get a BBB to work with Machinekit and 'X' 
hardware, does not mean it is the desirable or optimum choice.


For the money you spent on a cape and BBB, you could buy an ex-corporate 
dual core pentium computer and a Mesa card and have a setup which is 50 
times more capable

and far easier to configure and work with.


- Stepper Configuration Wizard: Better introductory information than anything else so 
far, but says it's for "standard parallel port." Does that mean I can't use it 
on BBB?

Yes, the BBB does not have a parallel port. Machinekit is not a beaglebone 
project. So there will always be options not applicable to your hardware



The documentation is legacy Linuxcnc, which was rtai kernel and parport 
or rt-preempt kernel and hardware FPGA cards only.
Where the docs diverge to include machinekit only areas, depends wholly 
upon people writing them and submitting them for general usage.


It is a community project, but thus far everyone (with minimal 
exceptions) who has stood where you are and bemoaned the lack of 
detailed personalised instructions for their
hardware combination of choice, has subsequently failed to contribute 
anything to make it easier for the next person to tread that path.


If you were to break that trend, I would be delighted.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Happy Birthday Machinekit!

['schoone...@btinternet.com' via Machinekit]

Yes, Happy Birthday to us

What is your inbox like if you have 3 years of these saved to reply to 
John !!!



On 04/04/18 00:37, John Morris wrote:

Happy Birthday #4 Machinekit!

While I've probably *contributed to* Machinekit *less* in the last 
year than ever before, it turns out I've *used* Machinekit *more*, 
whether for the EMC app on my 3D printer, or whether for straight HAL 
and QtQuickVCP for Golibox, autoclave and robot controls. Thanks to 
all the devs and maintainers for the great work.


John


On 04/03/2017 12:19 AM, John Morris wrote:

It's been three years.  Happy Birthday, Machinekit!

 John


On 04/03/2016 04:48 PM, John Morris wrote:

Happy two-year birthday, Machinekit! Another great year and more great
leaps forward.  Keep it up!

John


On 04/03/2015 03:59 AM, John Morris wrote:

A year ago today on April 3, Michael posted the [Machinekit
announcement] on the Emc-developers list.  IMO, that marks the 
official

birthday of Machinekit, so,

Happy Birthday!

I'm incredibly pleased (and surprised!) by the huge leaps the project
has made in such a short time, and by the explosive growth of the
project's community.  In the [ZeroMQ Guide Chap. 6], Pieter Hintjens
writes that the two go hand-in-hand, and says "It's hard to
overemphasize the power and persistence of a working open source
community."  That's exactly what I see here, an amazing bunch of folks
continually attracting new amazing folks, and the result is 
brilliance.

  Congratulations to everybody, both those participating in the
Machinekit project today, and also those who contributed to the 15+ 
year

legacy of the code base.  I can't wait to see what the next year will
bring.

[Machinekit announcement]:
http://sourceforge.net/p/emc/mailman/message/32182244/

[ZeroMQ Guide Chap. 6]: http://zguide.zeromq.org/php:chapter6

 John





--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: DE0-Nano-SoC install guide

['schoone...@btinternet.com' via Machinekit]

On 04/04/18 13:05, Richard Thornton wrote:

Thanks, much appreciated!

So I just do a "sudo apt-get update" and then a sudo apt-get upgrade"
to upgrade socfpga-rbf?


Just 'apt install socfpga-rbf' will update the package you are 
interested in.


Doing an dist-upgrade on a SD card image should not be done routinely, 
if it works, leave it alone!




The thing is I might never use the DE0-Nano_DB25 (I might not have the
ability to solder it, I might never get round to it) and I want to get
it working well before then, I'm only talking about connecting two
pins between the G540 and the DE0, I have all I need as the G540 can
accept 3.3v.


You will need to look at the schematics Charles sent you for the 40 pin 
header that the DB25 board connects to
20 pins will be the P3 connector and 20 pins the P2 (17 pins plus a 
couple of ground/5v+)


Then preferably take them from the header to a buffered BOB and connect 
to that.


You don't want to be connecting directly to a £90 board if you can help 
it, too easy to fry something.


On 4 April 2018 at 18:51, 'schoone...@btinternet.com' via Machinekit
<machinekit@googlegroups.com> wrote:

On 04/04/18 08:48, Richard Thornton wrote:

Hi,

Apologies, I'm a real newbie and would really appreciate some help,
where I'm up to...

Got the latest image from here:


http://deb.machinekit.io/uploads/de0-nano/debian-9.2-console-armhf-2017-10-29/debian-9.2-console-armhf-2017-10-29-5.0gb.img.xz

I used an 8GB sandisk ultra

In linux:

sudo bmaptool copy ./debian-9.2-console-armhf-2017-10-29-5.0gb.img.xz
/dev/mmcblk0

Connected up using SerialTools on the Mac, select usbserial-X
and baud rate 115200 and hit connect.

Press any key while booting:

=> env default -a
## Resetting to default environment
=> setenv ethaddr ba:d0:4a:9c:4e:ce
=> saveenv
Saving Environment to MMC...
Writing to MMC(0)... done
=> reset

l:p is machinekit:machinekit

uname -a shows:

Linux mksocfpga 4.1.22-ltsi-rt23-ge3c1da3 #1 SMP PREEMPT RT Tue Aug 23
09:56:10 UTC 2016 armv7l GNU/Linux

cat /etc/issue
"Debian GNU/Linux 8 \n \l

Machinekit SoC/FPGA Debian Image, Altera CycloneV 2016-11-21"

ifconfig should show the MAC you set and your IP address should be
stable with DHCP

I can ssh in from my Mac Pro (install XQuartz) using "ssh -X
machinekit@x.x.x.x"

I change directory into machinekit and run machinekit

I'm presented with the Machinekit Configuration Selector, woohoo!

Is 4.1.22-ltsi-rt23-ge3c1da3 correct for debian-9.2, isn't it supposed
to be 4.9, also /etc/issue shows 8 and the dates are all in 2016,
seems weird?

It is an upgraded Jessie image.

There is no Debian kernel for the DE0 so the kernel is one that works


What to do now with machinekit, no idea where to start, Charles made
me a G540 config:


https://github.com/machinekit/mksocfpga/commit/7e81953c7a7971daaa08c533ef1cc828dbd9fcdd

How do I get that on to this DE0?


Upgrade the socfpga-rbf package

Then check /lib/firmware/socfpga/dtbo for the config you need

Specify this one in your .ini file


So I have my G540 and a stepper, all I want to do is cable up the X
axis to verify everything works (I'm using jumper wires) so can
somebody please give me a newbie guide to what I do next, my thoughts:

DE0 to G540 DB25, on the G540 it seem I only need to connect pins 2
(X-AXIS STEP) & 3 (X-AXIS DIRECTION), which pins will I connect that
to on the DE0 once I import the Charles G540 config, direct DE0 pin to
G540 pin connections as I don't have an assembled DE0-Nano_DB25 (I got
three from osh park but I'm waiting for the parts to come from
digi-key and I will have to do a bunch of practice before I even
attempt to solder 0603)?

DB9 for X AXIS I need to connect the four wires from the stepper to
pins 6, 7, 8 & 9, 6 & 7 are the A coil, 8 & 9 the B coil, how to work
out the order as the stepper wires (black/green are one coil and
red/blue are another according to a piece of paper that came with the
stepper)?


Plenty of sources on the net for this


Once I have the G540 config and have wired the above simple test what
do I do from the "Machinekit Configuration Selector" page to just test
one stepper, is that possible, if so, how?


I would suggest that you modify the configs/hm2-soc-stepper/5i25-socfpga.ini
and hal sim for now
Don't worry about attaching steppers until you have a DB25, because the
pinout on the headers is different and you will get in a right mess quite
easily


Sorry to be a pain but this would really be a lot of help.

Thanks.

Richard


On 2 March 2018 at 18:37, Richard Thornton <richie.thorn...@gmail.com>
wrote:

Update

Image installed and successfully booted

I lost some time figuring out that I had written to an SD card partition
instead of the raw device...

On 28 February 2018 at 23:11, Richard Thornton
<richie.thorn...@gmail.com> wrote:

Hi,

Thanks for all the help.

So I have the DE0-Nano-SoC, G540 and steppers are

Re: [Machinekit] Re: Rotary Axis on BBB and Xylotex BBB_DB25/26

['schoone...@btinternet.com' via Machinekit]

  
  
It is floating point, otherwise it could only ever be accurate to
the nearest whole unit

On 30/03/18 08:59, John Dammeyer wrote:


  

  
I'll answer my own question. I was searching through
  the HAL docs rather than the ini docs.
  

 
http://www.machinekit.io/docs/config/ini_config/#sub:AXIS-section
In that document I found ANGULAR_UNITS = degrees and has
  nothing to do with NC code.  
"'ANGULAR_UNITS =
  ' - Specifies the 'machine units' for
  rotational axes. Possible choices are 'deg', 'degree' (360
  per circle), 'rad', 'radian' (2pi per circle), 'grad', or
  'gon' (400 per circle). This does not affect the angular
  units of NC code. In RS274NGC, A-, B- and C- words are
  always expressed in degrees."
It doesn't state if the SCALE
is a floating point number like 75,000 pulses per 360
degrees = 208.3 per 1 degree or if it has to be 208.


  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: rtapi_loadrt 'hal_pru_generic'... failed: Operation after apt update upgrade (4.14.20-ti-rt-r36)

['schoone...@btinternet.com' via Machinekit]

  
  
The linuxcnc log only shows after the error, need to see what it was
that produced 'operation not permitted' from insmod

We don't know what you upgraded from exactly, but for quite some
time now the /usr/lib/linuxcnc/prubin/*.bin have been symlinked
through to 
/usr/lib/linuxcnc//*.bin
Check both locations exist and no extra locations lingering from old
installs are there, but it does not sound like it is missing.
insmod errors can be pretty vague, but normally it is clear if the
module to be loaded does not exist.

It would not matter if the pru_generic.bin came from a xenomai
package, the binary format is the same across flavours.

Before I changed the packaging to build the pru binaries for all
armhf packages, people used to just manually copy them 
across from a xenomai package, if they wanted to use rt-preempt.


On 29/03/18 09:00, Karl Jacobs wrote:


  Are you sure that you prucode=prubin points to
rt-preempt/pru_generic.bin and not xenomai/pru_generic.bin ? The
Stretch Machinekit image runs on rt-preempt.

Am Mittwoch, 28. März 2018 22:24:08 UTC+2 schrieb Seb hl:

  Hi,
I've recently downloaded the Stretch Machinekit image
  for the Beaglebone (weekly build
  05.03.2018, 4.14.20-ti-rt-r36). Now I did an apt
  udate + upgrade and my hal-standalone config using the
  python hal modul no longer starts.



  $ export Debug=5
  $ python hal_test.py




  
  
Traceback (most recent call last):
  File "printjob_hal.py", line 305, in
  
    init_hardware()
  File "printjob_hal.py", line 211, in
  init_hardware
    rt.loadrt('hal_pru_generic', pru=1,
  disabled=0, num_stepgens=3, num_pwmgens=6,
  prucode=prubin, halname='hpg1', pru_period=1000)
  File "hal/cython/machinekit/rtapi.pyx",
  line 219, in machinekit.rtapi.RTAPIcommand.loadrt
  (hal/cython/machinekit/rtapi.c:5233)
RuntimeError: rtapi_loadrt '('hal_pru_generic',
  'disabled=0', 'pru_period=1000', 'num_stepgens=3',
  'pru=1', 'halname=hpg1',
  'prucode=/usr/lib/linuxcnc/prubin/pru_generic.bin',
  'num_pwmgens=6')' failed: Operation not permitted


  

  
  

linuxcnc.log shows



  
  
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5746:user halg_exit:293 HAL: removing
  component 76 'halcmd5746'
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5746:user ulapi_hal_lib_cleanup:235 HAL:
  lib_module_id=74
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5746:user halg_exit:293 HAL: removing
  component 74 'hal_lib5746'
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5746:user halg_exit:315 HAL:
  hal_errorcount()=0
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5746:user halg_exit:316 HAL: _halerrno=0
Mar 28 16:27:26 beaglebone rtapi:0:
  4:rtapi_app:5740:user pid=5740 flavor=rt-preempt
  gcc=6.3.0 20170516 git=v0.1~-~e37c70e
Mar 28 16:27:26 beaglebone msgd:0:
  ulapi:5750:user _ulapi_init(): ulapi rt-preempt
  v0.1~-~e37c70e loaded
Mar 28 16:27:26 beaglebone msgd:0:
  ulapi:5750:user halg_xinitfv:271 HAL: singleton
  component 'hal_lib5750' id=78 initialized
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5750:user --halcmd stop
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5750:user hal_stop_threads:357 HAL:
  threads stopped
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5750:user halg_exit:293 HAL: removing
  component 80 'halcmd5750'
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5750:user ulapi_hal_lib_cleanup:235 HAL:
  lib_module_id=78
Mar 28 16:27:26 beaglebone msgd:0:
  hal_lib:5750:user 

Re: [Machinekit] Pico-Systems ppmc setup

['schoone...@btinternet.com' via Machinekit]

  
  

On 12/03/18 14:48, Maxime Lemonnier
  wrote:


  Hi MachineKit,


It's been almost 8 years since I used my pico systems
  univpwm system (5 axis brushless servos). At the time I was in
  the process of rewriting Axis, that I found too monolithic to
  maintain or extend. I had a PyQt4 gui working, jogging,
  mdi-ing with a rather raw openGL display. I recently
  resurrected all this code and made it work with the latest
  linuxcnc version. Then I learned about the machinekit project,
  which I find fascinating. Especially the multicore effort and
  the more recent backend(hal)/frontend(not-hal) separation.
  Which leads to my questions:


What is the state of the ppmc driver in all this? anyone
  used it?
  


I am pretty sure no-one uses it, because it is not built.
https://github.com/machinekit/machinekit/blob/master/src/Makefile#L1440

The actual code is just the same as linuxcnc, it has not been
updated to run as instantiable or use any of the new HAL stuff.


  
Is there an effort/interest in modernizing the front-end
  API?
  


Alex has done quite a lot with the python API.

Or do you mean the front end GUI, as you mentioned rewriting Axis?

Alex has written Machineface, a remote capable Axis clone plus
QtQuick and Qml based VCP stuff.

I have a C++ Qt5 Axis replacement (which is not released because I
don't want to have to support it),
plus other C++ VCP stuff.


  
Any interest in ROS, expecially ROS2?
  


There is definitely some interest in ROS, I expect those concerned
will chime in.


  


At the present I use my machine as a robot to precisely
  move a sensor. I have years of experience in modern
  Python/C++/Qt/QML and lots of experience in real-time,
  math-heavy programming. I *may* find some time to contribute a
  bit to your project.
  


You will be most welcome.

Look forward to seeing what direction interests you.


  




Thanks!
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: how to add toolchanger funtion to BBB with Xylotex cape

['schoone...@btinternet.com' via Machinekit]

On 03/03/18 16:17, Aurelien wrote:

Many thanks for reply

yes this is your code but modified by Mike Ross for 6 tool turret

For the Axis i have set as you says LINEAR and with trial and error 
have found the good SCALE for turn 360° exactly



i don't understand "You simply home axis A" because with this this 
component this is impossible without following error from classic setting,


You just set the homing values to 0, then A will show as homed with no 
movement.

(read www.machinekit.io/docs/config/ini_homing/)

If you have connected up properly, when homed the relevant pin will be 
true and allow the M6TN command without movement to initialise.


also impossible to move the motor by GUI. (hijack stepgen) (i think 
you know that lol)


It is not impossible, just load a config without the toolchanger 
component and A connected as normal to work out moves required



There is so much commented out in your component, including the tool 
number setting on homing provision, I am not going to get into
trouble shooting something that seriously diverges from the component I 
wrote.


I would get the original component working by reducing the tool numbers 
to 6 and changing the angles / steps between stations accordingly.


Even if you managed to get the ATC seeking the home switch, you have to 
update machinekit with the current tool number.


The method my component uses is simple, because it sets the number via 
M6, whilst not attempting to process the command.


It could be adapted to rotate the ATC until the home switch is triggered 
when A is homed, then trigger a halui MDI command to set the tool number.


What you need to be wary of is un-commanded movement and whether it will 
always be safe or if it is possible to foul a workpiece
or injure the operator when the ATC suddenly starts rotating during a 
homing procedure.



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: how to add toolchanger funtion to BBB with Xylotex cape

['schoone...@btinternet.com' via Machinekit]

  
  
It would be very helpful if you actually said what you had found and
provided a link to it.

From the comments in the code, I can tell that it is one I wrote
some time ago and posted here
http://www.machinekit.io/docs/developing/toolchangers/
amongst other places, including on the Linuxcnc contributed
components wiki.

There are several problems with this type of ATC, that need to be
overcome. 

First is that it is extremely unlikely that your tool tips will be a
perfect '360 / number_of_tools' apart.
I found I had to move different amounts depending upon whether it
was a conventional cutting tool
or a VDI 20 mounted tool or drill.

Next there is how rotary axes are dealt with.
Unless care is taken, and incremental moves only are programmed, the
interpreter will try to take the shortest path to the required
position, which may entail running in reverse.
eg. if you are at 30 deg and program in 330 deg, it will attempt to
move 60 deg in reverse. You need to program for instance G90 A300 
to avoid ramming back against the pawl and stalling.

The only sure way to determine the movement amounts is to experiment
using MDI commands until you have consistent results.
Every ATC will be different, the one this was written for had 8
tools, not 6. (4 cutting 4 boring)

On my current lathe I abandoned degrees altogether and calculated
the number of steps required to move the distance, programmed A as a
linear axis 
and just moved by outputting incremental position_cmd.  Far simpler
to account for any differences in move between tool numbers.

To answer your second email

There is no provision for a ATC home switch, because the one it was
originally written for did not have one.

The code tells you how to home

                    // axis does not remember the current tool
number, so prompt for it when A axis homed.
    if(ishomed && !currenttoolnumber
&& !bWarn)
    {
    bWarn = true;  // just warn once, its not an
error as such but INFO won't display unless debugging is set 3+
    rtapi_print_msg(RTAPI_MSG_ERR, "No tool
selected. Use M6Tx to set current tool");
    break;..
    }

You simply home axis A, and if no tool number is set, you are
prompted to do a M6TN to set the tool number.

If you want an automatic rotation until the home switch is found
etc, you need to use elements of the carousel toolchanger in the
same page.

I think that one uses an amended iocontrol component however, which
I have no intention of publishing or supporting, so I would just use
the method outlined above
unless your programming is up to doing similar.

regards

On 02/03/18 22:39, Aurelien wrote:


  Hi


Now i have found good config for moving the turret but i
  like to report 2 issue.


First :
When you move from tool 1 to 6 with actual setting the past
  to pawl move too less compared to real amount moved 5x63° =
  .315     fudge factor is 14 so 315-14 = 301° !!!


If i'm not wrong :



  pin io float odd_move = 64      "distance from odd tool
station to even one";
  pin io float even_move = 64     "distance from even tool
station to odd one";
  pin io float divisor = 2        "used in calculating
reverse move to lock";
  pin io float fudge_factor = 24      "additional move to
ensure locking";  
  // fudge factor need to be selected higher than
"additional angle" x( nTool-1)
  // so 4°x(6-1) = 20  i have add +4 for make really sure.



In this way 5x64° = 320° - 24 = 296° so i'm sure the final
  position is 300° on the pawl with 4° more for locking


Please are you agree with this ?




Second :
Is not a real issue but need to be potnted for noob like
  me 


 if(!fjog  && (delay_index < (delay * 30)) )  
if your stepper is too slow the past to pawl not append so
  you need to increase this delay.




Br






  
  -- 
  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 

Re: [Machinekit] DE0-Nano-SoC install guide

['schoone...@btinternet.com' via Machinekit]

  
  

On 28/02/18 15:31,
  'schoone...@btinternet.com' via Machinekit wrote:


  
  
  On 28/02/18 12:11, Richard Thornton
wrote:
  
  
Hi,
  
  
  Thanks for all the help.


So I have the DE0-Nano-SoC, G540 and
  steppers are there instructions on how to install
  machinekit on the SoC and mesa on the FPGA?
  

  
  
  Use the existing image and change the firmware config in the .ini
  file to the firmware Charles has provided for the 540
  
  I put a snippet of my ini file here which shows that line.


Here even, that link somehow didn't make it

http://blog.machinekit.io/2016/11/you-will-recall-that-while-back-charles.html

 
  Easiest to adapt the 5i25-socfpga.ini etc to start with, get a sim
  working to show the module and the firmware is loading is loaded
  OK and then progress to a real config
  
  If you are not familiar with Mesa 5i25 pin outs, it will take a
  while, and as Charles alluded to, the pin designations from GPIO
  out to DB25 pins take a bit of getting your head around
  
  I made the attached image when I used  one without a 7i76, to
  connect to a stepper powered small mill using normal stepper
  drivers
  Just the P3 output pinout, but when you match it to the diagrams
  Charles linked to you will begin to see the light.
  (It is already in the forum posts if you search DE0-NANO, as with
  lots of other info, some out of date however - beware)
  
  

  


I should be able to wire it all up OK, I'm
  basically going to start by getting one stepper moving as
  instructed, I'm calling this phase 1.


So it will be DE0-Nano-SoC connected to G540
  connected to a stepper.


I will just install Cetus on Android.


Am I taking the path of least resistance?
  

  
  
  If you are a masochist.  :-P 
  
  Just connect it via a cat5 cable to your router and `ssh -X` into
  it to start with, you can struggle with remote apps as a second
  issue once you have it running.
  
  

  


Any documents, tips or tricks that will help
  me?
  

  
  
  There are several blogs on the DE0 starting here
  http://blog.machinekit.io/2016/04/altera-soc-running-machinekit-with.html
  
  

  


My computer skills are great so as long as I
  have instructions I will be fine with that.


I know very little about CAD/CAM or
  emc2/linuxcnc/machinekit.


Thanks again.


Richard






  

-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] DE0-Nano-SoC install guide

['schoone...@btinternet.com' via Machinekit]

  
  

On 28/02/18 12:11, Richard Thornton
  wrote:


  Hi,


Thanks for all the help.
  
  
  So I have the DE0-Nano-SoC, G540 and steppers
are there instructions on how to install machinekit on the
SoC and mesa on the FPGA?

  


Use the existing image and change the firmware config in the .ini
file to the firmware Charles has provided for the 540

I put a snippet of my ini file here which shows that line.

Easiest to adapt the 5i25-socfpga.ini etc to start with, get a sim
working to show the module and the firmware is loading is loaded OK
and then progress to a real config

If you are not familiar with Mesa 5i25 pin outs, it will take a
while, and as Charles alluded to, the pin designations from GPIO
out to DB25 pins take a bit of getting your head around

I made the attached image when I used  one without a 7i76, to
connect to a stepper powered small mill using normal stepper drivers
Just the P3 output pinout, but when you match it to the diagrams
Charles linked to you will begin to see the light.
(It is already in the forum posts if you search DE0-NANO, as with
lots of other info, some out of date however - beware)


  

  
  
  I should be able to wire it all up OK, I'm
basically going to start by getting one stepper moving as
instructed, I'm calling this phase 1.
  
  
  So it will be DE0-Nano-SoC connected to G540
connected to a stepper.
  
  
  I will just install Cetus on Android.
  
  
  Am I taking the path of least resistance?

  


If you are a masochist.  :-P 

Just connect it via a cat5 cable to your router and `ssh -X` into it
to start with, you can struggle with remote apps as a second issue
once you have it running.


  

  
  
  Any documents, tips or tricks that will help
me?

  


There are several blogs on the DE0 starting here
http://blog.machinekit.io/2016/04/altera-soc-running-machinekit-with.html


  

  
  
  My computer skills are great so as long as I
have instructions I will be fine with that.
  
  
  I know very little about CAD/CAM or
emc2/linuxcnc/machinekit.
  
  
  Thanks again.
  
  
  Richard
  
  
  
  
  
  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit XHC-WHB04B-6

['schoone...@btinternet.com' via Machinekit]

  
  

On 23/02/18 07:26,
  'schoone...@btinternet.com' via Machinekit wrote:


  
  
  
  
  On 23/02/2018 01:12, Aurelien wrote:
  
  
this message is sended by the driver i have found
  the output text error in source code
  
  
  
 xhc.hal =
  (xhc_hal_t *)hal_malloc(sizeof(xhc_hal_t));
 if (xhc.hal ==
  NULL) {
 fprintf(stderr,
  "%s: ERROR: unable to allocate HAL shared memory\n",
  modname);
 exit(1);
  

  
  The most common mistake that gives this error, is calling
  hal_malloc() when realtime is not started.
  Does not look to be the case.
  
  I have no familiarity with this code and only have a standard xhc
  pendant, but will look later.


I don't know what you are doing, but when I build it and run, it
works (as far as it can with no pendant attached)

You will get the error you had when realtime is not started.

The memory allocation will work in sim mode because that uses
calloc() not hal_malloc()


user@INTEL-i7:/usr/src/sandbox/machinekit# DEBUG=5 realtime restart
user@INTEL-i7:/usr/src/sandbox/machinekit# halcmd loadusr
xhc-whb04b-6 -Hup
:0: Program 'xhc-whb04b-6' started
user@INTEL-i7:/usr/src/sandbox/machinekit# hal   initialize HAL
component in HAL mode xhc-whb04b-6 ... ok
hal   initialize shared HAL memory for component id  80 ... ok
hal   bit   out whb.button.reset
hal   bit   out whb.button.macro-11
hal   bit   out whb.button.stop
hal   bit   out whb.button.macro-12
hal   bit   out whb.button.start-pause
hal   bit   out whb.button.macro-13
hal   bit   out whb.button.feed-plus
hal   bit   out whb.button.macro-1
hal   bit   out whb.button.feed-minus
hal   bit   out whb.button.macro-2
hal   bit   out whb.button.spindle-plus
hal   bit   out whb.button.macro-3
hal   bit   out whb.button.spindle-minus
hal   bit   out whb.button.macro-4
hal   bit   out whb.button.m-home
hal   bit   out whb.button.macro-5
hal   bit   out whb.button.safe-z
hal   bit   out whb.button.macro-6
hal   bit   out whb.button.w-home
hal   bit   out whb.button.macro-7
hal   bit   out whb.button.s-on-off
hal   bit   out whb.button.macro-8
hal   bit   out whb.button.fn
hal   bit   out whb.button.probe-z
hal   bit   out whb.button.macro-9
hal   bit   out whb.button.macro-10
hal   bit   out whb.button.macro-14
hal   bit   out whb.button.mode-continuous
hal   bit   out whb.button.macro-15
hal   bit   out whb.button.mode-step
hal   bit   out whb.button.macro-16
hal   s32   out whb.axis.0.jog-counts
hal   bit   out whb.axis.0.jog-enable
hal   float out whb.axis.0.jog-scale
hal   bit   out whb.axis.0.jog-vel-mode
hal   s32   out whb.axis.1.jog-counts
hal   bit   out whb.axis.1.jog-enable
hal   float out whb.axis.1.jog-scale
hal   bit   out whb.axis.1.jog-vel-mode
hal   s32   out whb.axis.2.jog-counts
hal   bit   out whb.axis.2.jog-enable
hal   float out whb.axis.2.jog-scale
hal   bit   out whb.axis.2.jog-vel-mode
hal   s32   out whb.axis.3.jog-counts
hal   bit   out whb.axis.3.jog-enable
hal   float out whb.axis.3.jog-scale
hal   bit   out whb.axis.3.jog-vel-mode
hal   s32   out whb.axis.4.jog-counts
hal   bit   out whb.axis.4.jog-enable
hal   float out whb.axis.4.jog-scale
hal   bit   out whb.axis.4.jog-vel-mode
hal   s32   out whb.axis.5.jog-counts
hal   bit   out whb.axis.5.jog-enable
hal   float out whb.axis.5.jog-scale
hal   bit   out whb.axis.5.jog-vel-mode
hal   bit   out whb.pendant.is-sleeping
hal   bit   out whb.pendant.is-connected
hal   float in  whb.halui.axis.0.pos-feedback
hal   float in  whb.halui.axis.1.pos-feedback
hal   float in  whb.halui.axis.2.pos-feedback
hal   float in  whb.halui.axis.3.pos-feedback
hal   float in  whb.halui.axis.4.pos-feedback
hal   float in  whb.halui.axis.5.pos-feedback
hal   float in  whb.halui.axis.0.pos-relative
hal   float in  whb.halui.axis.1.pos-relative
hal   float in  whb.halui.axis.2.pos-relative
hal   float in  whb.halui.axis.3.pos-relative
hal   float in  whb.halui.axis.4.pos-relative
hal   float in  whb.halui.axis.5.pos-relative
hal   float in  whb.stepgen.00.maxvel
hal   float in  whb.stepgen.00.position-scale
hal   float in  whb.stepgen.01.maxvel
hal   float in  whb.stepgen.01.position-scale
hal   float in  whb.stepgen.02.maxvel
hal   float in  whb.stepgen.02.position-scale
hal   float in  whb.stepgen.03.maxvel
hal   float in  whb.stepgen.03.position-scale
hal   float in  whb.stepgen.04.maxvel
hal   float in  whb.stepgen.04.position-scale
h

Re: [Machinekit] Machinekit XHC-WHB04B-6

['schoone...@btinternet.com' via Machinekit]

On 23/02/2018 01:12, Aurelien wrote:
this message is sended by the driver i have found the output text 
error in source code


xhc.hal = (xhc_hal_t *)hal_malloc(sizeof(xhc_hal_t));
if (xhc.hal == NULL) {
fprintf(stderr, "%s: ERROR: unable to allocate HAL shared memory\n", 
modname);

exit(1);
The most common mistake that gives this error, is calling hal_malloc() 
when realtime is not started.

Does not look to be the case.

I have no familiarity with this code and only have a standard xhc 
pendant, but will look later.





--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BiggleBone Red from element 14 and Machinekit with furaday cape

['schoone...@btinternet.com' via Machinekit]

  
  
Hopefully with this info, someone who knows about the BBB will be
able to see what is likely happening

A SWAG from me would be that the correct device tree file is not
being used, but what do I know 
;-) 

On 18/02/18 12:25, Marius Alksnys
  wrote:


    
  machinekit@beaglebone:~/CRAMPS$ DEBUG=5 realtime status
  
  /bin/bash: warning: setlocale: LC_ALL: cannot change locale
  (en_GB.utf8)
  
  msgd:0 stopped
  
  rtapi:0 stopped
  
  machinekit@beaglebone:~/CRAMPS$ DEBUG=5 realtime restart

  /bin/bash: warning: setlocale: LC_ALL: cannot change locale
  (en_GB.utf8)
  
  machinekit@beaglebone:~/CRAMPS$ halcmd loadrt hal_bb_gpio
output_pins=816 input_pins=807
  
  E: 18-02-18 12:20:36 dangling 'ROUTER' socket created at
  rtapi/rt-preempt/rtapi_app.cc:1301
  
  :0: insmod failed, returned -1:
  
  rtapi_rpc(): reply timeout
  
  See /var/log/linuxcnc.log for more information.
  
  machinekit@beaglebone:~/CRAMPS$ cat /var/log/linuxcnc.log
   
  Feb 18 12:20:06 beaglebone msgd:0: msgd:0: got signal 15 -
  sending SIGTERM to rtapi (pid 17329)
  
  Feb 18 12:20:06 beaglebone msgd:0: msgd:0: Terminated -
  shutting down
  
  Feb 18 12:20:06 beaglebone msgd:0: zeroconf: unregistering
  'Log service on beaglebone.local pid 17324'
  
  Feb 18 12:20:06 beaglebone msgd:0: log buffer hwm: 0% (8 msgs,
  665 bytes out of 524288)
  
  Feb 18 12:20:06 beaglebone msgd:0: sent SIGTERM to rtapi (pid
  17329)
  
  Feb 18 12:20:06 beaglebone msgd:0: normal shutdown - global
  segment detached
  
  Feb 18 12:20:24 beaglebone msgd:0: startup pid=17433
  flavor=rt-preempt rtlevel=5 usrlevel=5 halsize=524288
  shm=Posix cc=gcc 6.3.0 201
  70516  version=v0.1~-~57844e1
  
  Feb 18 12:20:24 beaglebone msgd:0: ØMQ=4.2.1 czmq=4.0.2
  protobuf=3.0.0 atomics=gcc intrinsics    libwebsockets=2.0.3
  
  Feb 18 12:20:24 beaglebone msgd:0: configured: sha=57844e1
  
  Feb 18 12:20:24 beaglebone msgd:0: built:  Dec  2 2017
  11:50:00 sha=57844e1
  
  Feb 18 12:20:24 beaglebone msgd:0: register_stuff: actual
  hostname as announced by avahi='beaglebone.local'
  
  Feb 18 12:20:24 beaglebone msgd:0: zeroconf: registering: 'Log
  service on beaglebone.local pid 17433'
  
  Feb 18 12:20:24 beaglebone rtapi:0: 2:rtapi_app:17438:user
  rtapi:0: cannot create core dumps - /proc/sys/fs/suid_dumpable
  contains 0
  
  Feb 18 12:20:24 beaglebone rtapi:0: 2:rtapi_app:17438:user you
  might have to run 'echo 1 > /proc/sys/fs/suid_dumpable' as
  root to ena
  ble rtapi_app core dumps
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  rtapi.so default iparms: ''
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  RTAPI:0  rt-preempt v0.1~-~57844e1 init
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  rtapi: loaded from rtapi.so
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  hal_lib.so default iparms: ''
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  rtapi_app_main:195 HAL: initializing RT hal_lib support
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  halg_xinitfv:90 HAL: initializing component 'hal_lib' type=4
  arg1=0 arg2=0
  /0x0
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  hal_heap_addmem:58 HAL: extending arena by 262144 bytes
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  halg_export_xfunctfv:85 HAL: exporting function 'newinst' type
  2 fp=0 owne
  r=66
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  halg_export_xfunctfv:85 HAL: exporting function 'delinst' type
  2 fp=0 owne
  r=66
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  halg_xinitfv:271 HAL: singleton component 'hal_lib' id=66
  initialized
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  rtapi_app_main:199 HAL: RT hal_lib support initialized rc=66
  
  Feb 18 12:20:24 beaglebone rtapi:0: 4:rtapi_app:17438:user
  hal_lib: loaded from 

Re: [Machinekit] Re: BiggleBone Red from element 14 and Machinekit with furaday cape

['schoone...@btinternet.com' via Machinekit]

  
  

On 18/02/18 12:01, Marius Alksnys
  wrote:


  Line 28:
loadrt hal_bb_gpio
  output_pins=816,822,823,824,825,826,914,923,925
  input_pins=807,808,809,810,817,911,913

  


I know nothing about BBB specifically

To debug it, use 

$ DEBUG=5 realtime restart
$ halcmd loadrt hal_bb_gpio
output_pins=816,822,823,824,825,826,914,923,925
input_pins=807,808,809,810,817,911,913

 to launch and look at /var/log/linuxcnc.log at the point of failure
It will contain the rtapi error message prints

If you cannot do this because the whole thing relies upon previously
loaded hardware drivers etc, then run
$ DEBUG=5 machinekit .ini
and look at the log after that fails

Also check dmesg in case of a GP fault or similar, albeit I see no
hint of that.


  

  2018 m. vasaris 18 d., sekmadienis 13:56:18 UTC+2, Schooner
  rašė:
  

On 18/02/18 11:37, Marius Alksnys wrote:

>

> CRAMPS.hal:28: insmod failed, returned -1:


Whatever you are trying to load at line 28 of your hal file
is what is 
producing the error


>

>

> That dangling 'DEALER' socket error sometimes shows up
even in BB Black.


Don't worry about that.

Since we switched to the new zmq API, that comes up when
machinekit does 
not shut down cleanly.


Something that used to be deleted automatically is not now.

  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BiggleBone Red from element 14 and Machinekit with furaday cape

['schoone...@btinternet.com' via Machinekit]

On 18/02/18 11:37, Marius Alksnys wrote:


CRAMPS.hal:28: insmod failed, returned -1:


Whatever you are trying to load at line 28 of your hal file is what is 
producing the error





That dangling 'DEALER' socket error sometimes shows up even in BB Black.


Don't worry about that.
Since we switched to the new zmq API, that comes up when machinekit does 
not shut down cleanly.


Something that used to be deleted automatically is not now.

--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] 5-axis simultaneous, G-codes?

['schoone...@btinternet.com' via Machinekit]

Not a subject I know much about, this seems to give the best overview of 
the kinematics.


http://linuxcnc.org/docs/ja/html/motion/5-axis-kinematics.html

I think the XYZAC example at the very end is nearest what you are 
proposing, with A and C rotating in relation to Z


The G codes remain the same, but the plane in which the axis operates 
may limit them.


Regards post processors, usually one for Mach is a good starting point, 
being generally compatible.


Quite a few CAM packages have linuxcnc posts, either in the package or 
as something available either from their forum or the linuxcnc forum. So 
check either before you choose a CAM or before you write your own.




On 17/02/2018 23:51, Damien D wrote:

Hi,

I'm planning on building a 5axis plasma cutter similar to this one 
 but 
with one more axis to tilt the torch head.
I'm thinking of starting from the existing maxkins kinematics (XYZBC) 
and then modify it to my needs.


Sorry if some questions seem 'stupid' but I'm quite new in CAM.. I'm 
googling around and try to understand how the 5axis interface works 
between CAM softwares and CNC controller (Machinekit).


What I'm struggling to understand is what will be the G-codes 
supported by Machinekit if I use the "maxkins" configuration?


Does Machinekit support RTCP (Rotating Tool Center Point) like in this 
video . If yes, what are 
the G-codes commands Machinekit can understand to handle RTCP?


I would need to know that to chose/configure the post-processor.
Or Maybe I should rather ask what post-processor (configuration) 
should I use to produce G-code that can be understood by Machinekit?


Thanks for your help,

/Damien
--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Machinekit and closed loop

['schoone...@btinternet.com' via Machinekit]

Ha, very good!


On 17/02/18 11:41, fritz wrote:
I first read that it had a name on the webcomic XKCD, but it appears 
that this phenomena now has its own website:


http://xyproblem.info/

On 02/16/2018 11:21 AM, 'schoone...@btinternet.com' via Machinekit wrote:

Now we know what you are doing, it is easier to assist.

You really need to ask a question about your problem, not how to 
effect the solution you have thought of.




--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Machinekit and closed loop

['schoone...@btinternet.com' via Machinekit]

  
  
Now we know what you are doing, it is easier to assist.

You really need to ask a question about your problem, not how to
effect the solution you have thought of.

The usual way to test backlash with a mill / lathe is to use a dial
gauge and move the bed a distance within the gauges range.
No need for a closed loop stepper.
May take several bites, but backlash should be taken up by the
initial movement in either direction and be pretty constant.

If you are getting a loss of position at a particular point in the
bed travel aside from the initial backlash, that could be caused by
binding, flexing or similar.
Insufficiently robust, loose or out of square / out of parallel,
rails, screws or many other mechanical components could be to blame.

A very common one is where the bed runs on 2 rails that are not
perfectly parallel and there is a pinch point at one or even both
ends.
Another is with Chinese castings and gib strips that have not been
scraped and mated properly
( I bought one once where the gib strips had obviously been
'finished' with an angle grinder!)

The ini file has facility to read a compensation file for backlash
on a per axis basis.
http://www.machinekit.io/docs/config/ini_config/#axis_-num-section-a-id-sub-axis-section-a

That should take care of backlash, which can differ in opposing
directions, especially on the Z axis where head mass comes into play
and may pre-load some moves and play catch-up on others.

Have a good search on their forum first, you won't be the first or
last person with a particular problem, it is defining what the
problem is that is key.

Good luck.


On 16/02/18 15:23, pasteur wrote:


  
Ok thank you guys, it was very useful all the data that you
  provided to me. 
Maybe I'm gonna to search or post in LinuxCNC forums... 
I read very interested the PID algorithm , but I'm using
  steppers motors and it isnt too eficient I guess, for my
  problem. My problem is simple I need to create a close loop
  because the backlash on my machine it isn't equal in all
  points and I need to measure and correct the errors...


I'm gonna to search in LinuxCNC forums and  post a question
  if it is necessary 


Thank you a lot.
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit and closed loop

['schoone...@btinternet.com' via Machinekit]

I suggest you read the posts on the linuxcnc forum.

More words have been written on this than most other subjects.

For the most part, the people who ask, want to exceed the limits of 
their stepper motors but escape the consequences.


https://www.forum.linuxcnc.org/forum/27-driver-boards/29357-real-closed-loop-steppers

Above is a starter



On 15/02/2018 19:45, pasteur wrote:
Hello everybody. I wanna make a simple question. How I can use a ruler 
or anything (anything that have sense) to create a closed loop for 
Machinekit?
I mean , I read many post  that explain that  it is possible but I 
didn't find a post that explain how do it (sorry about that :( ). I 
mean , I don´t know even where I must begin...
After read some Machinekit´s documentation  I guess that I need to 
program a HAL component , but also I dont know if that component 
already exists and I gonna to re invent the wheel.


Can someone give me a clue about this?

Thanks!
--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Build MK for non-debian distribution

['schoone...@btinternet.com' via Machinekit]

  
  

On 15/02/18 07:37, Filip Kotouček
  wrote:


  Hello everyone,


I would like to ask if there is possibility to build
  Machinekit to other Linux distribution than Debian. To be
  specific we have platform based on (am335x) Beaglebone. I'm
  currently using official TI kernel 4.9.59 Arago. There is no
  packaging system and because of Buildroot I don't have native
  toolchain in the target.


The ideal scenario will be to build the Machine kit by
  cross-compilation and just paste some files into filesystem.
  


This may or may not be of use, I don't / won't use any boards that
are not capable of building their own software.

If you use Debian multistrap on your desktop, not only will you have
a chroot system you can compile in, it downloads a base Debian root
filesystem tarball for the version/arch

You could also use this for the basis of the filesystem on your
board, with just enough libs etc to be able to run the binaries, but
with the build tools and any superfluous stuff stripped out.
Then you do have a packaging system, so building Machinekit
becomes optional

I have chroot builders for amd64/i386/armhf in
Wheezy/Jessie/Stretch/Buster set up on my desktop which work very
well.
The multistrap filesystems are also at the core of the sysroot
docker builders used to produce the Machinekit packages.

regards

  


Does anybody have some advice how to accomplish this? Or
  should I switch to Debian? My main reason to use Arago and
  buildroot is that I need to keep the system as small and fast
  as possible. Stock Debian for BBB boots about 40 seconds and
  it is not acceptable for our application. 


Many thanks  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Will DE0-Nano image work on a DE10-Nano board?

['schoone...@btinternet.com' via Machinekit]

First thing to check is probably that your dip switches are correctly set.

Seem to recall my DE0 board arrived with them in FPGA programming mode 
positions



On 08/02/2018 05:09, mugginsac wrote:

Thanks Schooner and Michael,

I followed the instructions through the setenv default -a; setenv 
ethaddr ba:d0:4a:9c:4e:ce; saveenv; reset.


Now I get a little farther. Do I need to remove the 'quiet' in 
extlinux.conf following the 'rootwait' so that I can see what is going 
on in the starting kernel?
=~=~=~=~=~=~=~=~=~=~=~= PuTTY log 2018.02.07 20:16:46 
=~=~=~=~=~=~=~=~=~=~=~=


U-Boot SPL 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14)
drivers/ddr/altera/sequencer.c: Preparing to start memory calibration
drivers/ddr/altera/sequencer.c: CALIBRATION PASSED
drivers/ddr/altera/sequencer.c: Calibration complete
Trying to boot from MMC1

U-Boot 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14 -0800)

CPU:   Altera SoCFPGA Platform
FPGA:  Altera Cyclone V, SE/A6 or SX/C6 or ST/D6, version 0x0
BOOT:  SD/MMC Internal Transceiver (3.0V)
   Watchdog enabled
I2C:   ready
DRAM:  1 GiB
MMC:   dwmmc0@ff704000: 0
In:    serial
Out:   serial
Err:   serial
Model: Terasic DE10-Nano
Net:   eth0: ethernet@ff702000
Hit any key to stop autoboot:  0
switch to partitions #0, OK
mmc0 is current device
Scanning mmc 0:2...
Found /boot/extlinux/extlinux.conf
Retrieving file: /boot/extlinux/extlinux.conf
179 bytes read in 21 ms (7.8 KiB/s)
1:    Linux 4.9.68-socfpga-r3
Retrieving file: /boot/vmlinuz-4.9.68-socfpga-r3
5455176 bytes read in 380 ms (13.7 MiB/s)
append: root=/dev/mmcblk0p2 ro rootfstype=ext4 rootwait quiet
Retrieving file: 
/boot/dtbs/4.9.68-socfpga-r3/socfpga_cyclone5_de10_nano.dtb

14997 bytes read in 45 ms (325.2 KiB/s)
## Flattened Device Tree blob at 0200
   Booting using the fdt blob at 0x200
   reserving fdt memory region: addr=0 size=1000
   Loading Device Tree to 03ff9000, end 03fffa94 ... OK

Starting kernel ...

**

U-Boot SPL 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14)
drivers/ddr/altera/sequencer.c: Preparing to start memory calibration
drivers/ddr/altera/sequencer.c: CALIBRATION PASSED
drivers/ddr/altera/sequencer.c: Calibration complete
Trying to boot from MMC1


U-Boot 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14 -0800)

CPU:   Altera SoCFPGA Platform
FPGA:  Altera Cyclone V, SE/A6 or SX/C6 or ST/D6, version 0x0
BOOT:  SD/MMC Internal Transceiver (3.0V)
   Watchdog enabled
I2C:   ready
DRAM:  1 GiB
MMC:   dwmmc0@ff704000: 0
In:    serial
Out:   serial
Err:   serial
Model: Terasic DE10-Nano
Net:   eth0: ethernet@ff702000
Hit any key to stop autoboot:  2 1 0
=>


--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Will DE0-Nano image work on a DE10-Nano board?

['schoone...@btinternet.com' via Machinekit]

  
  

On 07/02/18 07:03,
  'schoone...@btinternet.com' via Machinekit wrote:


  
  Read the gist that Michael did for the DE0

Specifically here
https://gist.github.com/mhaberler/89a813dc70688e35d8848e8e467a1337

where you will see exactly the same error message you report,
because the MAC address is not set.

  There is a link to it in the blogs.
  You have to set the NIC MAC address etc and then restart
  
  
  On 06/02/2018 23:54, mugginsac wrote:
  
  
I finally got the USB console working under
  stretch. 
  I didn't have the dtb file that it was looking
  (socfpga_cyclone5_de10_nano.dtb) for in the
  boot/dtbs/4.9.68-socfpga-r3/ directory.
  After I fixed that it is either loading the kernel or trying
  to load the kernel. It is getting a crc error and then
  starting over to try and load the kernel.
  
  Any help interpreting what it is telling me would be greatly
  appreciated. The log is below:
  
  =~=~=~=~=~=~=~=~=~=~=~= PuTTY log 2018.02.06 15:47:15
  =~=~=~=~=~=~=~=~=~=~=~=
  
  U-Boot SPL 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14)
  drivers/ddr/altera/sequencer.c: Preparing to start memory
  calibration
  drivers/ddr/altera/sequencer.c: CALIBRATION PASSED
  drivers/ddr/altera/sequencer.c: Calibration complete
  Trying to boot from MMC1
  
  
  U-Boot 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14 -0800)
  
  CPU:   Altera SoCFPGA Platform
  FPGA:  Altera Cyclone V, SE/A6 or SX/C6 or ST/D6, version 0x0
  BOOT:  SD/MMC Internal Transceiver (3.0V)
     Watchdog enabled
  I2C:   ready
  DRAM:  1 GiB
  MMC:   dwmmc0@ff704000: 0
  *** Warning - bad CRC, using default environment
  
  In:    serial
  Out:   serial
  Err:   serial
  Model: Terasic DE10-Nano
  Net:   
  Error: ethernet@ff702000 address not set.
  No ethernet found.
  Hit any key to stop autoboot:  2 1 0 
  switch to partitions #0, OK
  mmc0 is current device
  Scanning mmc 0:2...
  Found /boot/extlinux/extlinux.conf
  Retrieving file: /boot/extlinux/extlinux.conf
  179 bytes read in 21 ms (7.8 KiB/s)
  1:    Linux 4.9.68-socfpga-r3
  Retrieving file: /boot/vmlinuz-4.9.68-socfpga-r3
  5455176 bytes read in 380 ms (13.7 MiB/s)
  append: root=/dev/mmcblk0p2 ro rootfstype=ext4 rootwait quiet
  Retrieving file:
  /boot/dtbs/4.9.68-socfpga-r3/socfpga_cyclone5_de10_nano.dtb
  14997 bytes read in 45 ms (325.2 KiB/s)
  ## Flattened Device Tree blob at 0200
     Booting using the fdt blob at 0x200
     reserving fdt memory region: addr=0 size=1000
     Loading Device Tree to 03ff9000, end 03fffa94 ... OK
  
  Starting kernel ...
  
  
  U-Boot SPL 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14)
  drivers/ddr/altera/sequencer.c: Preparing to start memory
  calibration
  drivers/ddr/altera/sequencer.c: CALIBRATION PASSED
  drivers/ddr/altera/sequencer.c: Calibration complete
  Trying to boot from MMC1
  
  
  U-Boot 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14 -0800)
  
  CPU:   Altera SoCFPGA Platform
  FPGA:  Altera Cyclone V, SE/A6 or SX/C6 or ST/D6, version 0x0
  BOOT:  SD/MMC Internal Transceiver (3.0V)
     Watchdog enabled
  I2C:   ready
  DRAM:  1 GiB
  MMC:   dwmmc0@ff704000: 0
  *** Warning - bad CRC, using default environment
  
  In:    serial
  Out:   serial
  Err:   serial
  Model: Terasic DE10-Nano
  Net:   
  Error: ethernet@ff702000 address not set.
  No ethernet found.
  Hit any key to stop autoboot:  2 0 
  

-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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.
 

Re: [Machinekit] Re: Will DE0-Nano image work on a DE10-Nano board?

['schoone...@btinternet.com' via Machinekit]

Read the gist that Michael did for the DE0

There is a link to it in the blogs.

You have to set the NIC MAC address etc and then restart


On 06/02/2018 23:54, mugginsac wrote:

I finally got the USB console working under stretch.
I didn't have the dtb file that it was looking 
(socfpga_cyclone5_de10_nano.dtb) for in the 
boot/dtbs/4.9.68-socfpga-r3/ directory.
After I fixed that it is either loading the kernel or trying to load 
the kernel. It is getting a crc error and then starting over to try 
and load the kernel.


Any help interpreting what it is telling me would be greatly 
appreciated. The log is below:


=~=~=~=~=~=~=~=~=~=~=~= PuTTY log 2018.02.06 15:47:15 
=~=~=~=~=~=~=~=~=~=~=~=


U-Boot SPL 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14)
drivers/ddr/altera/sequencer.c: Preparing to start memory calibration
drivers/ddr/altera/sequencer.c: CALIBRATION PASSED
drivers/ddr/altera/sequencer.c: Calibration complete
Trying to boot from MMC1


U-Boot 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14 -0800)

CPU:   Altera SoCFPGA Platform
FPGA:  Altera Cyclone V, SE/A6 or SX/C6 or ST/D6, version 0x0
BOOT:  SD/MMC Internal Transceiver (3.0V)
   Watchdog enabled
I2C:   ready
DRAM:  1 GiB
MMC:   dwmmc0@ff704000: 0
*** Warning - bad CRC, using default environment

In:    serial
Out:   serial
Err:   serial
Model: Terasic DE10-Nano
Net:
Error: ethernet@ff702000 address not set.
No ethernet found.
Hit any key to stop autoboot:  2 1 0
switch to partitions #0, OK
mmc0 is current device
Scanning mmc 0:2...
Found /boot/extlinux/extlinux.conf
Retrieving file: /boot/extlinux/extlinux.conf
179 bytes read in 21 ms (7.8 KiB/s)
1:    Linux 4.9.68-socfpga-r3
Retrieving file: /boot/vmlinuz-4.9.68-socfpga-r3
5455176 bytes read in 380 ms (13.7 MiB/s)
append: root=/dev/mmcblk0p2 ro rootfstype=ext4 rootwait quiet
Retrieving file: 
/boot/dtbs/4.9.68-socfpga-r3/socfpga_cyclone5_de10_nano.dtb

14997 bytes read in 45 ms (325.2 KiB/s)
## Flattened Device Tree blob at 0200
   Booting using the fdt blob at 0x200
   reserving fdt memory region: addr=0 size=1000
   Loading Device Tree to 03ff9000, end 03fffa94 ... OK

Starting kernel ...


U-Boot SPL 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14)
drivers/ddr/altera/sequencer.c: Preparing to start memory calibration
drivers/ddr/altera/sequencer.c: CALIBRATION PASSED
drivers/ddr/altera/sequencer.c: Calibration complete
Trying to boot from MMC1


U-Boot 2018.01-rc1-dirty (Jan 25 2018 - 15:56:14 -0800)

CPU:   Altera SoCFPGA Platform
FPGA:  Altera Cyclone V, SE/A6 or SX/C6 or ST/D6, version 0x0
BOOT:  SD/MMC Internal Transceiver (3.0V)
   Watchdog enabled
I2C:   ready
DRAM:  1 GiB
MMC:   dwmmc0@ff704000: 0
*** Warning - bad CRC, using default environment

In:    serial
Out:   serial
Err:   serial
Model: Terasic DE10-Nano
Net:
Error: ethernet@ff702000 address not set.
No ethernet found.
Hit any key to stop autoboot:  2 0

--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit XHC-WHB04B-6

[schoone...@btinternet.com]

You need to contact the author.

The support in Machinekit is for the standard pendant.


On 03/02/2018 23:01, Aurelien wrote:

Hi All

Another newbee question :

I have found some link about the possible integration MPG XHC-WHB04B-6 
but for now i think ther is a missing file, if someone can help me to 
understand how make it work.


https://github.com/rubienr/machinekit-docs/blob/master/docs/man/man1/xhc-whb04b-6.asciidoc
https://github.com/rubienr/bbb-linuxcnc-config/tree/master/machinekit/configs/ARM.BeagleBone.Panther

Linuxcnc driver for this unit : 
https://github.com/smalla/linuxcnc/tree/master/configs/sim/axis/whb04b


i think the last file is only a link but i dont found in any folder.

can you help with this ?

Best regards
--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] problems installing xenomai kernel

[schoone...@btinternet.com]

  
  
The docs are a bit ambivalent, because the same link is used for all
kernels and it says what is available in Stretch (ie rt-preempt) 
but does not say Xenomai is not available in Stretch

I'll look at clarifying them in a bit, sorry for the confusion, over
familiarity on my part did not register any problem with them, 
because I would never have considered installing Xenomai

On 20/01/18 18:11,
  schoone...@btinternet.com wrote:


  
  Hi
  
  There are no Xenomai packages for Stretch.
  
  Xenomai was initially a way of getting better latency on ARM
  platforms, notably the BBB
  Now the rt-prempt kernels (4.4 at least) provide sufficiently good
  latency for a servo thread, so the later images use rt-preempt.
  Also later xenomai kernels moved to xenomai3, which is not
  supported in Machinekit.
  
  Xenomai has never been of any use for x86 computers.
  It does not have good enough latency for software stepping and is
  completely unnecessary for hardware stepping, where rt-preempt is
  much simpler and perfectly good.
  
  If you want to do software stepping you need to use linuxcnc which
  still has a rtai kernel
  
  Otherwise use rt-preempt with Mesa boards
  
  regards
  
  On 20/01/18 17:20, Raul Nieves Pardo
wrote:
  
  
hi!
  
  i installed debian stretch on a pentium E5300 (features PAE)
  i run these commands
  http://www.machinekit.io/docs/getting-started/APT-packages-stretch/
  all fine
  
  but trying to install the xenomai kernel (http://www.machinekit.io/docs/getting-started/install-rt-kernel-amd64/)
  "sudo apt-get install linux-image-xenomai.x86-686-pae"
  returns a not found package error
  
  "sudo apt-cache search linux-image-xenomai.x86-686-pae"
  and
  "sudo apt-cache search machinekit-xenomai"
  return  nothing
  
  any ideas?
  
  thanks!!!
  Raul

-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] problems installing xenomai kernel

[schoone...@btinternet.com]

  
  
Hi

There are no Xenomai packages for Stretch.

Xenomai was initially a way of getting better latency on ARM
platforms, notably the BBB
Now the rt-prempt kernels (4.4 at least) provide sufficiently good
latency for a servo thread, so the later images use rt-preempt.
Also later xenomai kernels moved to xenomai3, which is not supported
in Machinekit.

Xenomai has never been of any use for x86 computers.
It does not have good enough latency for software stepping and is
completely unnecessary for hardware stepping, where rt-preempt is
much simpler and perfectly good.

If you want to do software stepping you need to use linuxcnc which
still has a rtai kernel

Otherwise use rt-preempt with Mesa boards

regards

On 20/01/18 17:20, Raul Nieves Pardo
  wrote:


  hi!

i installed debian stretch on a pentium E5300 (features PAE)
i run these commands
http://www.machinekit.io/docs/getting-started/APT-packages-stretch/
all fine

but trying to install the xenomai kernel
(http://www.machinekit.io/docs/getting-started/install-rt-kernel-amd64/)
"sudo apt-get install linux-image-xenomai.x86-686-pae"
returns a not found package error

"sudo apt-cache search linux-image-xenomai.x86-686-pae"
and
"sudo apt-cache search machinekit-xenomai"
return  nothing

any ideas?

thanks!!!
Raul
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: comp missing

[schoone...@btinternet.com]

  
  
The last packages for that OS (assuming you are using Xenomai) are
here
http://deb.machinekit.io/debian/pool/main/m/machinekit/machinekit-xenomai_0.1.1507308757-1mk.travis.master.gitdbf37ca2~1wheezy_armhf.deb
http://deb.machinekit.io/debian/pool/main/m/machinekit/machinekit_0.1.1507308757-1mk.travis.master.gitdbf37ca2~1wheezy_armhf.deb

The machinekit-xenomai package contains comp

There are no machinekit-dev packages in the repo any more, I forgot
that I removed them once several of the new packages had been
generated for each distro.
Not doing so would have caused all sorts of problems, by making it
possible to install the old dev package which was defective, over
the top of the fully functioning
binaries and files from the new 'all in one' packages.

It might be an idea to remove machinekit, machinekit-xenomai and
machinekit-dev (if in the image) and then install again, making sure
there are no packages in /var/cache/apt/archives.

Then instead of trying to upgrade something, you are just installing
which should just pull the latest package for that OS/kernel
ie
apt purge machinekit machinekit-xenomai {machinekit-dev if
installed}
apt clean
apt update
apt install machinekit machinekit-xenomai

On 20/01/18 15:28, gary wrote:


  
sorry its the pre build package for the BBB


Like i said i can't do any updates seems after do then
  machinekit will not run even with the latest per build
  version.
I did try to do a install for machinekit-dev but get the
  message can't fine so that didn't go either.
I am trying to get a vel extrusion version working with my
  cramps board and machineface.


will try a newer install again and not do an upgrade see
  how that go's


thanks gary 
  
  On Friday, January 19, 2018 at 8:38:59 PM UTC-5, gary wrote:

  
comp missing, i installed a machinekit iso package and
  i am getting the error comp missing
its a older package from debian 7.9,  is it missing in
  the older packages?.


I don't seem to be able to do an update, upgrade every
  time i do then i can't boot machinekit it errors out and
  aborts 
tried a number of times reinstalled the package and
  then did update upgrade and never could run machinekit
  after that


any one run into this?


thanks gary
  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] comp missing

[schoone...@btinternet.com]

On 20/01/2018 01:38, gary wrote:
comp missing, i installed a machinekit iso package and i am getting 
the error comp missing
its a older package from debian 7.9,  is it missing in the older 
packages?.


There is no machinekit .iso package, that would be a CD or DVD image, 
can you please be precise about what you installed and on what.




I don't seem to be able to do an update, upgrade every time i do then 
i can't boot machinekit it errors out and aborts
tried a number of times reinstalled the package and then did update 
upgrade and never could run machinekit after that


any one run into this?


comp was originally in a separate package called machinekit-dev, more 
recently it has been incorporated into the main machinekit package.


If there is some reason to use Wheezy packages (which were discontinued 
and are no longer supported)

you need to install the corresponding machinekit-dev package

As for the error messages, they are telling you what is wrong, unless 
you share them along with details of what you are doing and on what 
platform.




thanks gary
--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] debian stretch can not launch machinekit, rtapi: no service UUID

[schoone...@btinternet.com]

  
  
The error comes from here
https://github.com/machinekit/machinekit/blob/master/src/rtapi/rtapi_app.cc#L1717
(note it tries to parse the ini file, what is in yours?)

Without the full errors and view of what you are trying to run can't
help further.


On 17/01/18 09:59, Letian Fang wrote:


      Hi guys:
    I have installed my new debian stretch on my laptop pc.
  I have compiled machinekit using posix flavor as followed the
  instruction by www.machinekit.io/docs/:
    debian/configure -p  // just want
use posix thread to run a simulation
    sudo mk-build-deps -ir

    cd src

    ./autogen.sh

    ./configure --with-posix

    ... ...
    after everything is done. I launched machinekit using
  shell command:machinekit
    And it turn's out :
    Starting Machinekit:
    rtapi: no service UUID 
    rtapi_app startup failed 
    halcmd: no service UUID ...
    I've used sudo make setuid, and still It does not work!


    So please, May I ask why? And How to solve this
  problem?
   
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: newbie help request: 'comp' compiling modules out of tree (kinematics and trajectory)

[schoone...@btinternet.com]

  
  

On 12/01/18 18:38, 'fishy' via
  Machinekit wrote:


  okay I have made a huge terrible hack that seems to
work for kinematics comp of genserkins out of tree but someone
needs to add command line args to comp to scan filelist to add
extra %s-objs and search in machinekit tree for paths etc..


https://github.com/machinekit/machinekit/blob/master/src/hal/utils/comp.g#L677


add in the following code

  
  
      print >>f, "obj-m +=
%s " % objname
     
nname=os.path.basename(os.path.splitext(filename)[0] )
      print >>f, "%s-objs :=
%s " % (nname, objname)
      print >>f, "%s-objs +=
machinekit/src/libnml/posemath/gomath.o "% nname
      print >>f, "%s-objs +=
machinekit/src/libnml/posemath/sincos.o "% nname
      print >>f, " "
  
  
      print >>f, "include
%s" % find_modinc()
  
  
      print >>f,
"$(RTLIBDIR)/%s$(MODULE_EXT): $(addprefix
$(OBJDIR)/,$(%s-objs))" % (nname,nname)
  
  

  
  

I may get time to do this over weekend unless someone else
  has better idea or more free time?,
  comments/advice/ideas..

  

  

Unfortunately that is not a tenable method for general use.
Comp is a pre-processor to allow the writing of fairly simple
components in psuedo-code, without any in depth knowledge of C.
It can be used to compile other code, but that is not its purpose

Anyway comp already has the provision to specify extra compile args
and extra link args, they just aren't documented.
https://github.com/machinekit/machinekit/blob/master/src/hal/utils/comp.g#L658

I submitted a patch for this before I joined Machinekit, but even
Linuxcnc only documents the 'extra_link_args' option.
http://linuxcnc.org/docs/html/hal/comp.html#_options
Need to update our docs I suppose :)

What you are doing is creating a mini Makefile within the comp
build, when what you should really be doing, is as Charles first
suggested, 
is just simply change the existing Makefile and Submakefile to build
your module in the source tree.

At every instance where you find genserkins, copy it and substitute
your own kins name - nothing magic about it

https://github.com/machinekit/machinekit/blob/master/src/emc/kinematics/Submakefile#L3
https://github.com/machinekit/machinekit/blob/master/src/emc/kinematics/Submakefile#L22
https://github.com/machinekit/machinekit/blob/master/src/Makefile#L504
https://github.com/machinekit/machinekit/blob/master/src/Makefile#L1581
https://github.com/machinekit/machinekit/blob/master/src/Makefile#L1897

If the real problem is that your platform is incapable of building
Machinekit from sources due to processor and memory shortcomings,
that is a different matter.

If you don't want to use a more capable platform, you can either use
your hacked comp, script the steps to build as you have done just
using gcc 
or longer term perhaps become familiar with chroot / Docker / VM
builds and build whatever component on a proper computer then copy
to your platform.

If you can convince us there is any general merit to your '..kins'
component, it could even be added to the source code and get built
into packages for all to use.

regards

regards

  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] newbie help request: 'comp' compiling modules out of tree (kinematics and trajectory)

[schoone...@btinternet.com]

On 12/01/18 16:05, John Morris wrote:




Is the old `Makefile.inc` still installed with packages? 


Yes it is.  The machinekit-dev package was scrubbed and now the main 
machinekit- package contains it at /usr/share/linuxcnc


That generated file is supposed to include enough stuff that it can be 
included in an out-of-tree C-language component's Makefile, and (I 
think) you can add your component's rules like they are in the main 
source tree's `src/Makefile`. Cribbing from there, this is unlikely to 
work as is, but I'd hope the basic idea would work:


# Include the build parameters
include /usr/share/linuxcnc/Makefile.inc

# Set up the module rules
obj-$(CONFIG_MYKINS) += tp.o tpmain.o
mykins-objs := tp.o tpmain.o $(MATHSTUB)
mykins$(MODULE_EXT): $(mykins-objs)

# Add the module build to a higher-level target
all: obj-$(CONFIG_MYKINS)

John



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] What's the current status of Machinekit on DE0 / DE10 -nano-SOC?

[schoone...@btinternet.com]

  
  
My opinion, use the DE0-NANO-Soc

It is known to work well and there are other people using it who can
help you if you get stuck.

The DE10 does not have a confirmed SD image ready to burn (not on
our site anyway) and the 'development'
was getting it to do the same as the DE0 already did, AFAICT.

At some stage people will have to use the DE10, because the DE0 has
been discontinued, but if you can get one use it

regards


On 10/01/18 13:13, Johan Jakobsson
  wrote:


  Hi,


I'm plan on using Machinekit on a desktop CNC-router and
  thought I'd run it on either a terrasic DE0 or DE10 which I'd
  connect to a MESA 7i76 using Charles DB25 interface board 
Reading through conversations I haven't got a good idea of
  how well Machinekit runs on those and which one is more
  stable.
Which one would you recommend? Seems the DE0-nano-SCO works
  well and there is a guide on how to set it up on the website.
  On the other hand, DE10 is the newer board so I guess most of
  the development is focused that one.. or? 


Regards
Johan Jakobsson


  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] share memory created usrmotintf.cc and motion.c

[schoone...@btinternet.com]

  
  
I am far from an expert on the emc stuff, but basically the shared
memory is a method of passing data from userspace to realtime.

The memory is allocated to the size of a specific structure, (say
emcmot_struct_t) then pointers are allocated to elements within the
structure.

These pointers are then used to pass and access the values.

As for the specific connection, motion.c creates the structure for
realtime functions.
usmotintf.cc, does not create a different structure, it you look it
gets the emcmot SHMEM_KEY (line 64) of the realtime structure and
uses that to create its local version (line 690)

This is cutting edge MIT 1980s coding, only the sheer complexity of
replacing it has prevented simpler and more efficient code.

That and the fact that it works, so don't fix it :)


Search the pointers created, for instance emcmotCommand, and you
will find it accessed by taskintf.cc and command.c
On 09/01/18 03:46,
  steve.bette...@gmail.com wrote:


  Hi everyone,


  I found that two blocks of share memory of
  emcmot_struct_t has been created, but how does they get
  associated?


usrmotintf.cc

  
int
  usrmotInit(const char *modname)
{
 int
  retval;


 module_id = rtapi_init(modname);
 if
  (module_id < 0) {
fprintf(stderr,
 "usrmotintf:
ERROR: rtapi init failed\n");
return
  -1;
 }
 /* get
shared memory block from RTAPI */
 shmem_id = rtapi_shmem_new(SHMEM_KEY, module_id, sizeof(emcmot_struct_t));
 if
  (shmem_id < 0) {
fprintf(stderr,
 "usrmotintf:
ERROR: could not open shared memory\n");
rtapi_exit(module_id);
return
  -1;
 }
 /* get
address of shared memory area */
 retval = rtapi_shmem_getptr(shmem_id,
  (void **)
  );
 if
  (retval < 0) {
rtapi_print_msg(RTAPI_MSG_ERR,
 "usrmotintf:
ERROR: could not access shared memory\n");
rtapi_exit(module_id);
return
  -1;
 }

  
motion.c

  
static int init_comm_buffers(void)
{
 int
  joint_num, axis_num, n;
 emcmot_joint_t
  *joint;
 int retval;

 rtapi_print_msg(RTAPI_MSG_INFO,
  "MOTION:
init_comm_buffers() starting...\n");

 emcmotStruct = 0;
 emcmotDebug = 0;
 emcmotStatus = 0;
 emcmotCommand = 0;
 emcmotConfig = 0;

 /* allocate and
initialize the shared memory structure */
 emc_shmem_id = rtapi_shmem_new(key,
  mot_comp_id, sizeof(emcmot_struct_t));
 if
  (emc_shmem_id < 0)
  {
rtapi_print_msg(RTAPI_MSG_ERR,
 "MOTION:
rtapi_shmem_new failed, returned %d\n", emc_shmem_id);
return -1;
 }
 retval = rtapi_shmem_getptr(emc_shmem_id,
  (void **)
  );
 if (retval
  < 0) {
rtapi_print_msg(RTAPI_MSG_ERR,
 "MOTION:
rtapi_shmem_getptr failed, returned %d\n", retval);
return -1;
 }
  

  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 

Re: [Machinekit] Re: BBB and new cramps board delta homing problems

[schoone...@btinternet.com]

  
  
I refer you to the docs
http://www.machinekit.io/docs/config/ini_homing/#home

Your config at present effectively just has 3 home switches, all the
other inputs for limits are commented out, presumably for testing?

I would suggest that you:

Number each axis HOME_SEQUENCE differently, 0, 1, 2 for now and only
home a specified axis, not 'home all'

Reduce the SEARCH_VEL and LATCH_VEL considerably ( I have 3.5 search
and 1 latch on my mill)

Comment out the HOME and HOME_OFFSET values (not sure about those at
all)

Open a HalMeter dialog for each of the home switch signals

Then home just one axis at a time, watching the relevant HalMeter
for the switch being tripped and latched.
(you have inverted your signals at hal:L265, is that actually
correct?  Need to be sure)

Once you get it homing on the switch, you can add stuff to make it
move elsewhere.
Currently given the speed it is moving at and the huge offset, which
may or may not be correctly signed, it will be
difficult to see what is going on.

regards

On 07/01/18 14:18, gary wrote:


  
https://pastebin.com/wrHza5td my
  cramps.ini file
https://pastebin.com/W7u7LX6c my
  cramps.hal file


sorry just thought there was a simple fix.. but here is my
  2 files


Thanks gary
  
  On Saturday, January 6, 2018 at 8:42:16 PM UTC-5, gary wrote:

  
I just connected a BBB and Cramps board to my delta
  printer, but i have been having homing problem
I have optic switches installed if i go into the hal
  configuration when machinekit it booted, and then to
  the show  window with xmax ymax and zmax displayed  then
  when axis's are off the switches there red, and when i
  trigger the switches they turn yellow
So i assume the switches are working and being read by
  the software.


But when i do home all the motors they move up to the
  switches in the correct direction but never stop just
  keeps running into the optic end stops and i have to power
  down..


any suggestions?


Thanks gary
  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] When machinekit icon is clicked, linuxcnc error message results

[schoone...@btinternet.com]

The only way anyone can assist is if you attach the errors.

How did you save them?  You can attach text or images to the forum posts 
by replying to the email.


Have you tried running a sample config, say axis_mm from the sims?
If that works it is your config, or lack of it that is the problem.

On 19/12/17 00:22, John Hammond wrote:

Linuxcnc error terminal window appears, but nothing runs.
I can start linuxcnc by opening terminal window and entering “linuxcnc &” . 
This stops when it does not find my configuration file.
I have copied the linuxcnc error message along with two other files that are 
created. I am willing to attach those files if shown how.
Thank you,
John Hammond
hammond.jo...@gmail.com



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit+iMX6+ubuntu12.04

[schoone...@btinternet.com]

  
  
Unless you are specific in saying exactly what you did, what
tutorial, on what 'official' site you followed and exactly what
compilation errors you got,
no-one is going to be able to even start to help you.

On 12/12/17 10:59, Savoy yang wrote:


  hey,man
  I recently wanted to transplant Machinekit to i.MX6
running on Ubuntu 12.04 already  with Xenomai(run
successful). I checked a lot of information and I know that
Machinekit can run on BBB and Raspberry Pi, both of which
are armhf ,also iMX6 , I follow the
tutorial given by the official website to install the
dependencies appeared a lot of problems, and finally all the
dependencies are installed, but the compilation MachineKit
error.
 I do not know if this road works, please give me
some advice, big brother, thank you very much!!!
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BeagleBone Black compiling hidcomp

[schoone...@btinternet.com]

  
  
The headers have always been in the -dev package and more recently
in the main machinekit package itself.

My crystal ball obviously needs a polish.

On 04/12/17 07:42, Rob M wrote:


  

Would have been quicker if someone had of chimed in and said
"the headers are in the debian update" hahahahahaha wouldn't
have wasted all that time with docker
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit installation on i386, need help...

[schoone...@btinternet.com]

  
  
I have been able to log in to the server and everything looks up and
running.

I have run a Jessie partition, which uses the machinekit repo and it
`apt updates` fine, with no errors and reaches the machinekit repo.
The ping result may be a red herring, the owner might have closed
down all ports except those specifically in use (80, 22, 25 etc)

Why are you installing Wheezy?

There are no more builds for it from machinekit and Debian is
dropping it completely next year.

I would install Jessie, which will give you a few more years of
support.

regards

On 04/12/17 10:43,
  schoone...@btinternet.com wrote:


  
  There appears to be some problem with the server
  
  Whilst http appears OK, a direct ping of the url or the
  deb.machinekit.io comes back as `Unreachable`
  
  You can download the packages directly via http and install with
  dpkg for now, or wait for it to be fixed.
  Will need to contact the server owner, who may not be about at
  present.
  
  
  On 04/12/17 08:29, kevin.zhang0...@gmail.com
wrote:
  
  
Hi guys
  I am new to Machinekit, and trying to install it on my
i386 PC platform. However, when I following the installation
guide:
  
  
  sudo apt-key adv --keyserver
  hkp://keyserver.ubuntu.com:80 --recv 43DDF224 sudo sh -c \ "echo 'deb http://deb.machinekit.io/debian
  wheezy main' > \ 
  /etc/apt/sources.list.d/machinekit.list" sudo apt-get
  update
  
  
  
  It doesn't work and the terminal shows that:
  
  
  
Err http://deb.machinekit.io
  wheezy Release.gpg                                       
                                                           
                                             
  Could not connect to deb.machinekit.io:80
  (86.59.12.252). - connect (111: Connection refused)
Reading package lists... Done                        
W: Failed to fetch http://deb.machinekit.io/debian/dists/wheezy/Release.gpg 
  Could not connect to deb.machinekit.io:80 (86.59.12.252).
  - connect (111: Connection refused)


W: Some index files failed to download. They have been
  ignored, or old ones used instead.
  
  
  
  Seems like that Machinekit group has changed their server
address, or maybe I did something wrong. Could any one
please help me?
  
  
  Thanks for your time!!
  
  
  Kevin

-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Machinekit installation on i386, need help...

[schoone...@btinternet.com]

  
  
There appears to be some problem with the server

Whilst http appears OK, a direct ping of the url or the
deb.machinekit.io comes back as `Unreachable`

You can download the packages directly via http and install with
dpkg for now, or wait for it to be fixed.
Will need to contact the server owner, who may not be about at
present.


On 04/12/17 08:29,
  kevin.zhang0...@gmail.com wrote:


  Hi guys
I am new to Machinekit, and trying to install it on my i386
  PC platform. However, when I following the installation guide:


sudo apt-key adv
--keyserver hkp://keyserver.ubuntu.com:80 --recv 43DDF224
sudo sh -c \ "echo 'deb http://deb.machinekit.io/debian wheezy
main' > \
  
/etc/apt/sources.list.d/machinekit.list"
sudo apt-get update



It doesn't work and the terminal shows that:



  Err http://deb.machinekit.io wheezy Release.gpg         
                                                           
                                                           
           
    Could not connect to deb.machinekit.io:80
(86.59.12.252). - connect (111: Connection refused)
  Reading package lists... Done                        
  W: Failed to fetch
http://deb.machinekit.io/debian/dists/wheezy/Release.gpg 
Could not connect to deb.machinekit.io:80 (86.59.12.252). -
connect (111: Connection refused)
  
  
  W: Some index files failed to download. They have been
ignored, or old ones used instead.



Seems like that Machinekit group has changed their server
  address, or maybe I did something wrong. Could any one please
  help me?


Thanks for your time!!


Kevin
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BeagleBone Black compiling hidcomp

[schoone...@btinternet.com]

Sounds good.

Everything is possible if you have the patience ;-)


On 04/12/2017 01:56, Rob M wrote:



Ok,

A little update:

Updated my system, 
bone-debian-7.11-machinekit-armhf-2017-01-22-4gb.img.xz, using apt-get 
update, (had to use apt-offline as no net in the shed).


Ended up with:
*machinekit_0.1.1507308757-1mk.travis.master.gitdbf37ca2~1wheezy_armhf.deb
*which had the needed include files.

To get everything built (after installing the Qt libs & dev stuff and 
libsub-dev) I built and installed included libusb and went on my way ( 
a few changes to the make file and one include file) and hidconfig & 
hidcomp successfully natively compiled on the BBB. Wasn't as slow as 
my 33 year old Radio Shack PC4 pocket computer (I think I got it when 
I was 12), which is an rebranded 80's Casio.


I did a quick test of hidconfig, that operated as per normal once 
perms were set for my usbdevice.


loaded up hidcomp via halrun and did a few tests and all appears to be 
doing what it should, all the pins showed up via show pin.


I have only tested buttons and LEDs as that is all I have on my 
device.no LCD or other items.
As I'm a bit mentally stuffed, late night & early morning, I'll test 
it on a live system on the morrow.


So far it has just been a test of Franks s\w to see if it will 
compile, then load under halrun and toggle a few pins.

**
--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BeagleBone Black compiling hidcomp

[schoone...@btinternet.com]

  
  

On 02/12/17 21:13, Rob M wrote:


  

Ok, I think I may have over thought what I wanted to do.

The goal:
Add some extra I\O pins for a simple control panel.to get
away from the keyboard.

Now that I've thought about it, would using 2 MCP2017 (I2C i\o
expanders) and using the supplied user component get me to where
I want to be without going the whole USB route on the BBB ?
  


Outside my experience I'm afraid, hopefully someone else will chime
in.

I would still like to see the repo with hidcomp in it, to assess
whether it is worth including it anyway though.


  
Sorry to have been wasting your time guys. (I'm kinda feeling
pretty silly now)
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] M66 question

[schoone...@btinternet.com]

  
  

On 02/12/17 17:20, Sag ich Dir nich
  wrote:


  okay thank you, i will try that.


one more question, does the next M66 reset #5399?

  


You will need to experiment, set a short timeout M66 and don't send
a signal, to set it to -1, then set a longer one and send a signal
before
it times out.
#5399 should contain 0 or 1 typically but never -1

I don't know if #5399 is set to anything before M66 exits, I would
suspect it would hold the last entered value until M66 exits for
whatever reason
and sets the return value accordingly.
You will need to verify, but suspect #5399 is meaningless until M66
does exit.


  

  Am Samstag, 2. Dezember 2017 16:54:35 UTC+1 schrieb Schooner:
  
 It appears that the Q
  input is of type double, which means it will only
  wait for 28,561.6 years before moving again
  
  So set it to a high number and effectively it will never
  move again without an input.
  
  Or you can enter a shorter period and test the value of
  #5399 to see if it is -1 indicating a timeout.
  In that event you can abort.
  
  
  On 02/12/17 15:15, Sag ich Dir nich wrote:
  
  
The manual says "M66
wait on an input stops further execution of the
program, until the selected event (or the programmed
timeout) occurs."
  
  
  so that means if the machine waits for an input
but then the timeout occurs, the machine continues?
Thats not logical to me. I would like the machine to
stop if the timeout occurs. 

  
  In case you were wondering why i bother, its for
an automatic toolchanger to let the machine know if
the pull stud is in the correct position to prevent
expensiveness, if it is not in the correct position,
the machine waits until it is... or gets an timeout
and stops further execution. 
  
  
  thanks

-- 
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 machinekit+...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  

  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] M66 question

[schoone...@btinternet.com]

  
  
It appears that the Q input is of type double, which means it will only
wait for 28,561.6 years before moving again

So set it to a high number and effectively it will never move again
without an input.

Or you can enter a shorter period and test the value of #5399 to see
if it is -1 indicating a timeout.
In that event you can abort.


On 02/12/17 15:15, Sag ich Dir nich
  wrote:


  The manual says "M66 wait on
  an input stops further execution of the program, until the
  selected event (or the programmed timeout) occurs."


so that means if the machine waits for an input but then
  the timeout occurs, the machine continues? Thats not logical
  to me. I would like the machine to stop if the timeout
  occurs. 
  

In case you were wondering why i bother, its for an
  automatic toolchanger to let the machine know if the pull stud
  is in the correct position to prevent expensiveness, if it is
  not in the correct position, the machine waits until it is...
  or gets an timeout and stops further execution. 


thanks
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BeagleBone Black compiling hidcomp

[schoone...@btinternet.com]

Probably, over Wheezy anyway.

The problem with czmq, was that they changed the API so that the 
Machinekit code only worked with earlier versions.


We changed the code and Debian backported the czmq packages to Jessie 
from Stretch,


which means that Jessie and Stretch are supported by Debian packages.

Wheezy is stuck with earlier packages and the old code.

This, plus the imminent 'end of life' is why Wheezy packages are no 
longer being produced



On 01/12/2017 22:10, 'fishy' via Machinekit wrote:

Is jessie the best debian to use now?

I got the feeling that the czmq or zmq versions in 'configure; 
autobuild were upgraded to require the stretch versions and it broke 
back compatability with wheezy and jessie?






On Thursday, 30 November 2017 16:14:12 UTC, Schooner wrote:


On 30/11/17 14:23, Rob M wrote:



Schooner,

Cheers for your reply. Going by what you said maybe running on a
Jessie image maybe my best bet ?

Depends completely upon what you are trying to do.

The image is actually probably Wheezy.


So I managed to get a docker image to build machinekit packages
for jessie, so I guess that is a plus. This is the resource I
used https://github.com/Dovetail-Automata/mk-cross-builder
 .


Yeah, that is what we use in the CI Builder


I guess now the best thing to do would be start running jessie,
install/upgrade machine with the packages built in the docker
container. With a large enough sd card I could install a build
enviroment and use that to build hidcomp on ?


I would be minded to fix hidcomp on the image you have first, then
look at upgrading the image to something else.
There are other problems with using Jessie and later kernels that
are best left until after this.

Trying to build anything natively on the BBB is a major pain, it
has the processing power of an 80's Casio watch. (minor exaggeration)
That is without the added problem of the Makefile.modinc from the
earlier -dev packages previously mentioned.

I would cross-build the altered Machinekit repo for Wheezy armhf 
and copy the relevant modules across to test them.

When that is proved, if it fits nicely with the rest of the repo,
it could be included and then you can just install the new
Machinekit packages onto whatever
platform you are using, but would need to see the code first.



I'm not looking to use the LCD portion so that's not a real issue
I'm guessing ?

I must admit I'm not a "debian guy" Slackware is more my cup of
tea. Would I be correct in guessing before I build the
binary-debs I should build a source package that I can decompress
later and use to build hidcomp against ?

I'm really appreciating your time.

Rob
-- 
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 machinekit+...@googlegroups.com .
Visit this group at https://groups.google.com/group/machinekit
.
For more options, visit https://groups.google.com/d/optout
.


--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: BeagleBone Black compiling hidcomp

[schoone...@btinternet.com]

  
  

On 30/11/17 14:23, Rob M wrote:


  

Schooner,

Cheers for your reply. Going by what you said maybe running on a
Jessie image maybe my best bet ?
  

Depends completely upon what you are trying to do.

The image is actually probably Wheezy.

  
So I managed to get a docker image to build machinekit packages
for jessie, so I guess that is a plus. This is the resource I
used https://github.com/Dovetail-Automata/mk-cross-builder .
  


Yeah, that is what we use in the CI Builder

  
I guess now the best thing to do would be start running jessie,
install/upgrade machine with the packages built in the docker
container. With a large enough sd card I could install a build
enviroment and use that to build hidcomp on ?
  


I would be minded to fix hidcomp on the image you have first, then
look at upgrading the image to something else.
There are other problems with using Jessie and later kernels that
are best left until after this.

Trying to build anything natively on the BBB is a major pain, it has
the processing power of an 80's Casio watch. (minor exaggeration)
That is without the added problem of the Makefile.modinc from the
earlier -dev packages previously mentioned.

I would cross-build the altered Machinekit repo for Wheezy armhf 
and copy the relevant modules across to test them.

When that is proved, if it fits nicely with the rest of the repo, it
could be included and then you can just install the new Machinekit
packages onto whatever
platform you are using, but would need to see the code first.


  
I'm not looking to use the LCD portion so that's not a real
issue I'm guessing ?

I must admit I'm not a "debian guy" Slackware is more my cup of
tea. Would I be correct in guessing before I build the
binary-debs I should build a source package that I can
decompress later and use to build hidcomp against ?

I'm really appreciating your time.

Rob
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] BeagleBone Black compiling hidcomp

[schoone...@btinternet.com]

  
  

On 30/11/17 02:05, Rob M wrote:


  Hi,

Was wondering what would the procedure be to compile hidcomp http://hidcomp.sourceforge.net/
with this image that I am currently running, bone-debian-7.11-machinekit-armhf-2017-01-22-4gb.img.xz.



There will probably be a problem compiling in-situ with that vintage
image.

An unforeseen consequence of the old build system, now fixed, was
that in using posix to build the machinekit-dev package, it set the
flavour to posix in Makefile.modinc
irrespective of the running kernel flavour.

  
  I have the sources for both hidcomp ( pulled from
  https://github.com/ftkalcevic/hidcomp) and machinekit (pulled
  from github a couple of months ago) and with a few changes,
  just copying over the machinekit hal.h to the hidcomp source
  tree and adding a couple of defines I got it to compile and
  run on a x86 laptop with a RIP posix build. I use the laptop
  to test configs and such, easier than the BBB in the garage.



If you push your working machinekit tree up to Github and send me a
link, I will build it for armhf (Jessie?) and send you the binaries
to add to your image and test.

There may be merit in adding it to the machinekit repo and
supporting it in packages hereafter, but will need to see the code.
Rather looks from the wiki as if changes which might not be wanted
are required. [ http://hidcomp.sourceforge.net/ ]

The other potential issue is that your image is very old and pre a
lot of hal changes
If you have run it on x86 with a few month old machinekit pull, it
probably is not an issue, but need to check


  
  I've lashed up a test device on a sparkfun ProMicro using the
  LUFA libs and I can talk to the deivce using both halinput
  & hidcomp. The reason I would like to use hidcomp is to be
  able to have more than 11 leds (linux kernel limit with the
  input subsystem for keyboard LEDs). 
  
  The reason for this avenue is to be able to have a hardware
  control panel for some basic switches & indicators. As it
  is halinput is ok with the ProMirco, but eventually I would
  like to use a teensy for a bit more I\O.
  
  If a docker container is the go for cross compiling, is anyone
  able to supply some step by step instructions (really simple
  ones).I had a go last night had some issues.
  
  Or should I just burn the image I have to a SD card, download
  all the required tools & libs to build, and go from there.
  
  Ordoes anyone have some pointers to using python and
  pysub, if this is a viable option ? 
  
  Thanks 
  Rob 
  

  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: rtapi:0 stopped not root

[schoone...@btinternet.com]

  
  

On 18/11/17 22:12, Jeff Washburn wrote:


  

   
Add a 5i25 - 7i76 combo and a rt-preempt kernel on Stretch
with x86, will run perfectly, I have it in my workshop.


  


  I have a Geckodrive G540 that I would like to use.
  
  To be clear, Schooner, you are recommending switching to x86
  by:
  
  
removing amd64 rt-preempt 

Installing the 686 RT-Preempt kernel
installing 5i25 to which I will connect a Geckodrive
  G540 (I got it).
there doesn't seem to be an i386 specific version of
  Machinekit.
  
  Is there a need to switch to an x86 version of Stretch?
  

  

Sorry, I am obviously confusing you, I just used x86 as a generic
term to mean an Intel / AMD processor rather than ARM for instance.

What I am suggesting is that if you want software stepgen, you use
the Linuxcnc distro ( which is 686, but will install on a x86_64 )
Then your gecko drive will work fine with a parport and software
stepgen.

I tested xenomai kernels on Intel computers and never found them any
good, certainly nothing like as good as rtai kernels for low latency
and would not recommend them for software stepping, when a far
better alternative exists.

If you install a 5i25, then you can use a x86_64 Jessie or Stretch
system, with a rt-preempt kernel and it will work fine with
Machinekit.
I am sure there are configs out there for matching 5i25 / Gecko 540.

regards


  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: rtapi:0 stopped not root

[schoone...@btinternet.com]

  
  

On 18/11/17 19:14, Jeff Washburn wrote:


  I installed according to the machinekit.io Getting
Started page for Debian Stretch on amd64. That means apt
configure, intall kernel and then runtime packages. I did not
build machinekit from sources. Beyond that I have not installed
anything associated with Machinekit. After installing runtime
packages, I ran Latency Test from the drop down menu in Gnome
and then Xfce. I did run latency-test  25us and 1ms. That gave
jitter number in both servo and base, but according to
linuxcnc.log,rtapi:0 stopped so the numbers are simulation
based.

As for being directed to LinuxCNC earlier,  I don't know when
that happened, but I am not as young as I once was.

As for LinuxCNC, their documentation states that Preempt RT is
necessary for 64 bit (amd64). That is probably not the same
kernel as RT-Preempt.
  


All computers for the last 15 years approx have been amd64.  You can
still run i386 on them however.

  
I am unclear as to why software stepping on my core i5 with
rt-preempt kernel won't work. Perhaps the Debian Stretch with
"amd64" platform should notify potential users that amd64 does
not include Intel 64 bit chips.
  


It is nothing to do with the processor, what matters is the kernel

I have a single core Pentium P4 which will run all day with a
homebrew 3.5.7 rtai kernel at 5k max jitter

machinekit is not an alternative to Linuxcnc that supports later
distros, 
It has ended up specialising in supporting boards and processors
using hardware stepping on ubiquitous kernels (rt-preempt and
xenomai)

Add a 5i25 - 7i76 combo and a rt-preempt kernel on Stretch with x86,
will run perfectly, I have it in my workshop.

If you want software stepping, rtai is the only way to go and
machinekit no longer supports it.


  
On Saturday, November 18, 2017 at 11:08:08 AM UTC-7, Schooner
wrote:

  Can you please elaborate

Have you installed packages, if so which ones?

Rather sounds as if you have built machinekit from sources,
if you have somewhere to try running 'sudo make setuid' ?
In which case you presumably have not completed the building
properly.

The whole latency-test issue rather suggests you are
operating under the misapprehension that you can do software
stepping on an x86 with rt-preempt kernel.
You can't.

The reason there is no base thread reading by default in
latency-test, is that the various hardware step generation
methods used ( BBB PRU, DE0-NANO-Soc, Mesa boards)
do not require a base thread.

The rtai kernel based Linuxcnc distro you were pointed to
earlier is where you need to go if you want to run a machine
off the parport with software stepgen.

On Saturday, 18 November 2017 17:35:44 UTC, Jeff Washburn
wrote:

  I have loaded Machinekit on an Intel Core
i5 3400 series and on Debian Stretch. Flavor is
rt-preempt. When I run Latency Test, I get the subject
message in the linuxcnc.log file as it is starting up.
When I run latency-test in root, then I get a mesage
saying the rtapi_app is not root. Thus, lack of root
privilege stops rtapi_app.

How do I give rtapi_app root privilege? Somebody might
suggest 'sudo make setuid' running from /src. I've tried
it and get an error: 'no rule to make target 'setuid''.
I am guessing that make operates with setuid, which I
don't have.
  

  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 

Re: [Machinekit] Re: Machinekit package repo

[schoone...@btinternet.com]

On 27/10/17 18:14, Robert Nelson wrote:

On Fri, Oct 27, 2017 at 12:11 PM, schoone...@btinternet.com
<schoone...@btinternet.com> wrote:

Hi Robert

Whilst you are looking at this can I pick your brains for the best of your
rt-preempt kernels to upgrade the DE0-NANO-Soc
mksocfpga image with.

I have already upgraded the existing SD image to Stretch and have it
working, but with the old 4.1.22-ltsi-rt23 kernel

I have crossbuild tools installed and can quite happily build one from your
sources if required

I'd use their 4.9.x branch:

https://github.com/altera-opensource/linux-socfpga/branches/all

as that has rt right now:

https://www.kernel.org/pub/linux/kernel/projects/rt/

But as soon as they open up socfpga-4.13 i'd move to that..

Regards,


That turned out to be a PITA

The socfpga patches do not play nicely with the rt patches.
Lots of failures and some code seems to bear little relation to the patch.

Fixed up 4.9 manually only to find it really did not like some of the 
patches.
Never had so much trouble with a rt-preempt kernel, they normally patch 
and build without problems.


Messed around some more with 4.6 etc but similar problems.

I have stuck with the 4.1.22-ltsi-rt23 kernel, which works perfectly 
well. :-P


The main intention was to get it upgraded to Stretch and able to use the new
czmq4 enabled code, which it now does.

I will post the image in http://deb.machinekit.io/uploads/de0-nano/
once cleaned up and minimised as far as possible.





--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

[Machinekit] Re: Machinekit package repo

[schoone...@btinternet.com]

Hi Robert

Whilst you are looking at this can I pick your brains for the best of 
your rt-preempt kernels to upgrade the DE0-NANO-Soc

mksocfpga image with.

I have already upgraded the existing SD image to Stretch and have it 
working, but with the old 4.1.22-ltsi-rt23 kernel


I have crossbuild tools installed and can quite happily build one from 
your sources if required


regards


On 27/10/17 14:39, Robert Nelson wrote:

On Fri, Oct 27, 2017 at 7:49 AM, Lewis Cobb  wrote:

Hi Robert -

This is great - thanks very much.

I'm pretty new to all this but looking at the Uboot link you provided it
looks like I just have to make sure the Uboot is a recent one, and then that
the overlays are enabled ?

The easiest thing to do is just run this script and report it's output..

sudo /opt/scripts/tools/version.sh

Then we can walk you thru what you need to do..


Now - here's a real noob question for you - are Uboot overlays and Device
Tree Overlays the same thing ?  I've just started to read about the DT stuff
and my head is spinning, so if you have a link to some educational material
on the uboot overlay stuff it would be much appreciated.

They are exactly the same.  We've just EOL'ed kernel overlays and have
u-boot doing all the overlay loading work. (its more reliable/etc..)

Regards,



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Machinekit package repo

[schoone...@btinternet.com]

  
  
What RCN does is very much down to him :)

The way is open for him to produce a rt-preempt kernel 4.x.x image
using the packages, if he is so minded


On 26/10/17 18:03, Lewis Cobb wrote:


  Hi Schooner
What does this mean for luddites like myself that barely know
what they are doing - i.e. - will this mean there's a new SD
card image that may appear soon for machinekit/BBB ?  I've been
using Robert Nelson's image linked to on the machinekit.io docs
page dated 2017-02-12.
Thanks (and thanks for your patience with an enthusiastic
machinekit noob ...)
Lewis

On Thursday, October 26, 2017 at 4:39:45 AM UTC-3, Schooner
wrote:

   Hi All,

The changes to switch over to John Morris's brilliant
cross-builder docker container, are complete for machinekit
(https://hub.docker.com/r/dovetailautomata/mk-cross-builder/)

In conjunction with the recent changes to machinekit to use
the new czmq API (John Morris, Alex Rossler and myself)
this means the following for users.


  There are now Debian Stretch packages of
machinekit for amd64, i386 and armhf

These do not require any other special packages, all
dependencies are available from the standard Debian repos.
(save a rt-preempt kernel for armhf - see below)

There are no Xenomai packages for Stretch.
The 4.x.x-rt-preempt kernels are good enough to provide a
stable servo thread on BBB or DE0-NANO-Soc
Stretch would require Xenomai3, there are none available and
machinekit would not run with them without a lot of work.
The PRU on a BBB or the FPGA on the DE0-NANO provides
hardware stepgen etc, so  rt-preempt is perfectly adequate.


  There are a full set of Debian Jessie packages
for machinekit for amd64, i386 and armhf

These do require a special build version of libczmq4,
  which is in the repo
They no longer require special rt-preempt kernels from
  the repo or elsewhere for i386 or amd64, because Debian
  have backported their Stretch kernels
  Simply add the jessie-backports stanza to your apt sources
  http://www.machinekit.io/docs/getting-started/install-rt-kernel-amd64/

Xenomai packages are still provided for Jessie for
  backward compatibility.

Raspbian packages however are no longer specifically
  provided.
  This is because whilst Raspbian continues as a distro, it
  uses Jessie armhf binaries.
  It was hugely wasteful of time and disk space to build and
  hold 2 sets of packages that were byte identical
  except for the distro name.

Just change your apt sources stanza from raspbian to
  jessie if you want to continue to update packages.
  It does work, a user Petros tried it for us on this thread
  https://groups.google.com/forum/#!topic/machinekit/vSPXdPspyEs

All the existing packages will be retained, so if you
  want, you can just do nothing.
  http://www.machinekit.io/docs/getting-started/APT-packages-raspbian/


  Debian Wheezy is no longer supported or being
built for, as per the previous announcement

All support from Debian for Wheezy will cease in May 2018
  anyway and it currently has absolutely minimal LTS, with
  no updates or backports.
  Machinekit would not now work unless a special backport of
  the czmq library were made and given its life expectancy,
  that would be unproductive.

All the existing packages will be retained, there will
  just not be any updates.

Important re armhf:

 There were difficulties in upgrading on armhf in the
  initial packages - this was caused by apt running the
  postrm script from the old package after
  the new package was installed, deleting some elements.

This is now fixed but if you have an affected package
  already installed on armhf, the upgrade may fail.
  You simply need to remove it first and then re-install
  packages.

sudo apt-get purge machinekit*

  sudo apt-get update

sudo apt-get 

[Machinekit] Machinekit package repo

[schoone...@btinternet.com]

  
  
Hi All,

The changes to switch over to John Morris's brilliant cross-builder
docker container, are complete for machinekit
(https://hub.docker.com/r/dovetailautomata/mk-cross-builder/)

In conjunction with the recent changes to machinekit to use the new
czmq API (John Morris, Alex Rossler and myself)
this means the following for users.


  There are now Debian Stretch packages of machinekit
for amd64, i386 and armhf

These do not require any other special packages, all dependencies
are available from the standard Debian repos.
(save a rt-preempt kernel for armhf - see below)

There are no Xenomai packages for Stretch.
The 4.x.x-rt-preempt kernels are good enough to provide a stable
servo thread on BBB or DE0-NANO-Soc
Stretch would require Xenomai3, there are none available and
machinekit would not run with them without a lot of work.
The PRU on a BBB or the FPGA on the DE0-NANO provides hardware
stepgen etc, so  rt-preempt is perfectly adequate.


  There are a full set of Debian Jessie packages for
machinekit for amd64, i386 and armhf

These do require a special build version of libczmq4, which is in
  the repo
They no longer require special rt-preempt kernels from the repo
  or elsewhere for i386 or amd64, because Debian have backported
  their Stretch kernels
  Simply add the jessie-backports stanza to your apt sources
  http://www.machinekit.io/docs/getting-started/install-rt-kernel-amd64/

Xenomai packages are still provided for Jessie for backward
  compatibility.

Raspbian packages however are no longer specifically provided.
  This is because whilst Raspbian continues as a distro, it uses
  Jessie armhf binaries.
  It was hugely wasteful of time and disk space to build and hold 2
  sets of packages that were byte identical
  except for the distro name.

Just change your apt sources stanza from raspbian to jessie if
  you want to continue to update packages.
  It does work, a user Petros tried it for us on this thread
  https://groups.google.com/forum/#!topic/machinekit/vSPXdPspyEs

All the existing packages will be retained, so if you want, you
  can just do nothing.
  http://www.machinekit.io/docs/getting-started/APT-packages-raspbian/


  Debian Wheezy is no longer supported or being built
for, as per the previous announcement

All support from Debian for Wheezy will cease in May 2018 anyway
  and it currently has absolutely minimal LTS, with no updates or
  backports.
  Machinekit would not now work unless a special backport of the
  czmq library were made and given its life expectancy, that would
  be unproductive.

All the existing packages will be retained, there will just not
  be any updates.

Important re armhf:

 There were difficulties in upgrading on armhf in the initial
  packages - this was caused by apt running the postrm script from
  the old package after
  the new package was installed, deleting some elements.

This is now fixed but if you have an affected package already
  installed on armhf, the upgrade may fail.
  You simply need to remove it first and then re-install packages.

sudo apt-get purge machinekit*

  sudo apt-get update

sudo apt-get install machinekit machinekit-

Subsequent upgrades will work as intended. 
regards




  

  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Installation for Rpi3 using machinekit.io site instruction broken

[schoone...@btinternet.com]

  
  
The repo is stabilised now, although there is still some work to do

The current serial version for machinekit packages is
0.1.1508673250.gitf811885-1dovetailautomatamkcrossbuilderamd648.HEAD~1jessie
I just did a test install into Jessie amd64 partition and all is
good to go, package versions match

(Jessie-sda2-chroot)root@INTEL-i7:/# apt-get install machinekit
machinekit-posix
Reading package lists... Done
Building dependency tree   
Reading state information... Done
The following extra packages will be installed:
  libczmq4 libzmq3 python-zmq
The following NEW packages will be installed:
  libczmq4 libzmq3 machinekit machinekit-posix python-zmq
0 upgraded, 5 newly installed, 0 to remove and 45 not upgraded.
Need to get 5,603 kB/6,391 kB of archives.
After this operation, 39.8 MB of additional disk space will be used.
Do you want to continue? [Y/n] 
Get:1 http://deb.machinekit.io/debian/ jessie/main machinekit-posix
amd64
0.1.1508673250.gitf811885-1dovetailautomatamkcrossbuilderamd648.HEAD~1jessie
[1,042 kB]
Get:2 http://deb.machinekit.io/debian/ jessie/main machinekit amd64
0.1.1508673250.gitf811885-1dovetailautomatamkcrossbuilderamd648.HEAD~1jessie
[4,560 kB]

Make sure you have purged old packages

apt-get purge machinekit*
apt-get purge libczmq*
apt-get purge libzmq*

apt-get update
apt-get install machinekit machinekit-



On 22/10/17 13:42,
  schoone...@btinternet.com wrote:


  
  The repo is in a state of flux.
  
  We are switching over to a new package builder.
  I am aware that the machinekit packages serials are not matching
  the flavour serials
  
  When a build currently under way completes, hopefully that will be
  resolved
  
  Please bear with us
  
  On 22/10/17 13:03, Petros Makris
wrote:
  
  
The exact same way, the instruction for
  installing in jessie 64bit are broken as well.
  
  
  
  On Monday, October 16, 2017 at 11:41:06 PM UTC+3, Petros
  Makris wrote:
  
Hello,
  
  If booting to the latest 8.x jessie raspbian and following
  the instructions in machinekit.io you get
  
  (after installing the latest rt-preempt kernel)
  
  The following packages have unmet dependencies:
  
  machinekit-rt-preempt:
Depends: machinekit (=0.1.1507881226-1mk.travis.master.git15d5a3f6)

but is not going to be installed
  
  Any ideas? 
  
  

  

-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Installation for Rpi3 using machinekit.io site instruction broken

[schoone...@btinternet.com]

  
  
The repo is in a state of flux.

We are switching over to a new package builder.
I am aware that the machinekit packages serials are not matching the
flavour serials

When a build currently under way completes, hopefully that will be
resolved

Please bear with us

On 22/10/17 13:03, Petros Makris wrote:


  The exact same way, the instruction for installing
in jessie 64bit are broken as well.



On Monday, October 16, 2017 at 11:41:06 PM UTC+3, Petros Makris
wrote:

  Hello,

If booting to the latest 8.x jessie raspbian and following
the instructions in machinekit.io you get

(after installing the latest rt-preempt kernel)

The following packages have unmet dependencies:

machinekit-rt-preempt:
  Depends: machinekit (=0.1.1507881226-1mk.travis.master.git15d5a3f6)
  but is not going to be installed

Any ideas? 


  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Installation for Rpi3 using machinekit.io site instruction broken

[schoone...@btinternet.com]

Thanks for reporting back.

It is good to get hands on confirmation of something that is 
theoretically obvious, but I dont have the hardware to test.


The documented 'raspbian' tagged route should work fine too now.


On 17/10/2017 20:54, Petros Makris wrote:

[SOLVED]

Hi. Fisrt of, thanks.

I did the following

1) Boot into clean 2017-07-05-raspbian-jessie
2) Added the machinekit repo as in machinekit.io
3) installed rt-preempt kernel via apt-get and reboot
4) changed distro tag in /etc/apt/sources.list.d/machinekit to 'jessie'
5) apt install machinekit machinekit-rt-preempt


And I managed to get machinekit installed and starting up.

(Although without much stress on the system I got 262ms jitter with 
median ~100ms!)



-Petros

On Monday, October 16, 2017 at 11:41:06 PM UTC+3, Petros Makris wrote:

Hello,

If booting to the latest 8.x jessie raspbian and following the
instructions in machinekit.io  you get

(after installing the latest rt-preempt kernel)

The following packages have unmet dependencies:

machinekit-rt-preempt: Depends: machinekit
(=0.1.1507881226-1mk.travis.master.git15d5a3f6) but is not going
to be installed

Any ideas?


--
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 machinekit+unsubscr...@googlegroups.com 
.

Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.


--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Installation for Rpi3 using machinekit.io site instruction broken

[schoone...@btinternet.com]

  
  
I have now included the packages for czmq and zmq from Jessie into
the Raspbian distro
They are still tagged ~jessie, but that does not matter, because
that is what the dependencies specify anyway.
It is certainly a much simpler solution to re-building a complete
set of binaries just to have a ~raspbian tag in them
http://deb.machinekit.io/debian/dists/raspbian/main/binary-armhf/Packages

If you have the time and inclination, try my suggestion of switching
to the Jessie distro first, it should just work.
The only difference between a package tagged ~jessie-armhf and
~raspbian-armhf is the name in the control file etc.
You can download one from each distro folder and run debdiff on them
to show this.
In future distributions (Stretch onwards) there will not be specific
raspbian packages.

Otherwise, just purge all your old packages, 
apt-get update
apt-get install machinekit machinekit-rt-preempt
and let the dependencies list pull in libczmq, libzmq and python-zmq


On 17/10/17 08:59, Petros Makris wrote:


  
Thanks!, I will
  give it a try this evening and let you now.

  On Tue, Oct 17, 2017 at 10:49 AM, schoone...@btinternet.com
<schoone...@btinternet.com>
wrote:

   
All the packages of version 0.1.1507881226-1mk for raspbian are
there

Unfortunately raspbian should no longer exist as a
distro, because the packages are binary identical to
jessie-armhf

Installing machinekit first, will find
machinekit-rt-preempt, but it is probably not going to
find libczmq4 and libzmq3, because they have jessie
tags.

You should find that if you purge your packages
(machinekit, machinekit-rt-preempt libczmq libzmq
python-zmq), 
change the distro tag in your /etc/sources.list.d/machinekit.list
to jessie
do
apt-get update
apt-get install machinekit machinekit-rt-preempt

It will install and work

Otherwise, later I should be able to produce some libs
that are specifically tagged ~raspbian.

On
  16/10/17 22:10, Petros Makris wrote:


  All I can see is that machinekit
depends on libczmq4 which is available in stretch.

On Monday, October 16, 2017 at 11:41:06 PM UTC+3,
Petros Makris wrote:

  Hello,

If booting to the latest 8.x jessie raspbian and
following the instructions in machinekit.io you get

(after installing the latest rt-preempt kernel)

The following packages have unmet dependencies:

machinekit-rt-preempt:

  Depends: machinekit (=0.1.1507881226-1mk.travis.master.git15d5a3f6)

  but is not going to be installed

Any ideas? 

 
  
   
 
   -- 
  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 machinekit+unsubscribe@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.

 
  
  
  -- 
  website: http://www.machinekit.io
  blog: http://blog.machinekit.io
  github: https://github.com/machinekit
  --- 
  You received this message because you are subscribed
  to a topic in the Google Groups "Machinekit" group.
  To unsubscribe from this topic, visit https://groups.google.com/d/topic/machinek

Re: [Machinekit] Re: Installation for Rpi3 using machinekit.io site instruction broken

[schoone...@btinternet.com]

  
  

All the packages of version 0.1.1507881226-1mk
for raspbian are there

Unfortunately raspbian should no longer exist as a distro, because
the packages are binary identical to jessie-armhf

Installing machinekit first, will find machinekit-rt-preempt, but it
is probably not going to find libczmq4 and libzmq3, because they
have jessie tags.

You should find that if you purge your packages (machinekit,
machinekit-rt-preempt libczmq libzmq python-zmq), 
change the distro tag in your /etc/sources.list.d/machinekit.list to
jessie
do
apt-get update
apt-get install machinekit machinekit-rt-preempt

It will install and work

Otherwise, later I should be able to produce some libs that are
specifically tagged ~raspbian.

On 16/10/17 22:10, Petros Makris wrote:


  All I can see is that machinekit depends on
libczmq4 which is available in stretch.

On Monday, October 16, 2017 at 11:41:06 PM UTC+3, Petros Makris
wrote:

  Hello,

If booting to the latest 8.x jessie raspbian and following
the instructions in machinekit.io you get

(after installing the latest rt-preempt kernel)

The following packages have unmet dependencies:

machinekit-rt-preempt:
  Depends: machinekit (=0.1.1507881226-1mk.travis.master.git15d5a3f6)
  but is not going to be installed

Any ideas? 


  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

[Machinekit] Announcement: Machinekit repo down from 1130hrs BST Weds 11th Oct 2017

[schoone...@btinternet.com]

Just a notification NOT to update packages or do a fresh pull of 
machinekit packages or any czmq / zeromq libs until further advised


If you do, you are liable to get machinekit and lib versions which do 
not match and will not run.


This is to allow building of new packages and insertion of related libs 
to the repo.


Thanks for your co-operation

--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Error with CRAMPS.hal

[schoone...@btinternet.com]

  
  

On 09/10/17 14:34, 'Dimitar Dobrev' via
  Machinekit wrote:


  My packages are from:

http://deb.mgware.co.uk/
  


that is my unofficial testing repo

  
They are for POSIX and rt-preempt only.
About the index, I actually don't have iio:device0
  under /sys/bus/iio/devices/
  so that's the problem.
  I could try changing the kernel on my existing installation, I
  hope it's possible. Another option would be to build
  Machinekit from source for Stretch. Do you think it's
  possible?



Alex has just replied, so follow his advice, he knows far more about
the BBB

  
On Monday, October 9, 2017 at 4:25:15 PM UTC+3, Schooner wrote:

   
On 09/10/17 12:57, 'Dimitar Dobrev' via Machinekit
  wrote:


  I created /usr/lib/linuxcnc/xenomai and
copied pru_generic.bin there. After trying again I got:

Traceback (most recent call last):
  File "/usr/bin/hal_temp_bbb", line 198, in

    checkAdcInput(pin)
  File "/usr/bin/hal_temp_bbb", line 156, in
checkAdcInput
    pin.filename = tempName[0]
IndexError: list index out of range
  


This is not an error linked to pru_generic.bin AFAIK, but
you need someone who knows about BBB.

If it is saying that an index of 0 is out of range, 
you need to look at what it is trying to do and work out how
it can be empty.
https://github.com/machinekit/machinekit/blob/master/src/hal/user_comps/hal_temp_bbb.py#L156


  
In case the pru_generic.bin for rt-preempt cannot work
in xenomai, 


The pru_generic.bin is identical for all kernel flavours so
you can just copy it across

The rt-preempt component modules may work with xenomai,
there has been discussion as to whether they are actually
different, 
but the bigger problem is that you will lack
ulapi_xenomai.so and rtapi_app_xenomai.

There are no xenomai packages for Stretch either, as Stretch
is on Xenomai3, which does not work with Machinekit
I assume you have manually installed the Jessie ones?

I suggest you read this thread
https://groups.google.com/forum/#!topic/machinekit/RrNLUo4ASP4
and contact Darren or one of the others to find where you
can download an image with the rt-preempt kernel, 
or even just the kernel, modules etc and install it into
your existing image


  do I correctly understand that switching
from xenomai to rt-preempt on an existing installation
is impossible and I need to install a new image instead?

On Saturday, October 7, 2017 at 5:06:38 PM UTC+3,
Schooner wrote:

   I don't use a
BBB, it looks like Debian only provide i386 and
amd64 rt-preempt kernels for Stretch.

I don't know which rt kernel others have used,
probably one from here
https://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Mainline_.284.14.x_lts.29

The one I used on my server is 4.4.68-ti-rt-r112,
but there are probably newer ones.
Looks like linux-image-4.9.40-bone-rt-r7 is quite
new, the BBG people have to use rt-preempt I believe
to get everything working.

Someone should be able to give you a link
In fact if you have a root around here https://rcn-ee.com/rootfs/bb.org/testing/
you may find an image which suits

rt-preempt kernels are quite easy to build, I have
even built one on a 4 core aarm64, but that has much
more processing power
than a BBB, might take quite a while on a BBB

Is there any reason why you have to use a BBB at
all?
Life is much simpler with a proper computer ;\

On 07/10/17 14:39, 'Dimitar Dobrev' via
  Machinekit wrote:


  

Re: [Machinekit] Error with CRAMPS.hal

[schoone...@btinternet.com]

  
  

On 09/10/17 12:57, 'Dimitar Dobrev' via
  Machinekit wrote:


  I created /usr/lib/linuxcnc/xenomai and copied
pru_generic.bin there. After trying again I got:

Traceback (most recent call last):
  File "/usr/bin/hal_temp_bbb", line 198, in 
    checkAdcInput(pin)
  File "/usr/bin/hal_temp_bbb", line 156, in checkAdcInput
    pin.filename = tempName[0]
IndexError: list index out of range
  


This is not an error linked to pru_generic.bin AFAIK, but you need
someone who knows about BBB.

If it is saying that an index of 0 is out of range, 
you need to look at what it is trying to do and work out how it can
be empty.
https://github.com/machinekit/machinekit/blob/master/src/hal/user_comps/hal_temp_bbb.py#L156


  
In case the pru_generic.bin for rt-preempt cannot work in
xenomai, 


The pru_generic.bin is identical for all kernel flavours so you can
just copy it across

The rt-preempt component modules may work with xenomai, there has
been discussion as to whether they are actually different, 
but the bigger problem is that you will lack ulapi_xenomai.so and
rtapi_app_xenomai.

There are no xenomai packages for Stretch either, as Stretch is on
Xenomai3, which does not work with Machinekit
I assume you have manually installed the Jessie ones?

I suggest you read this thread
https://groups.google.com/forum/#!topic/machinekit/RrNLUo4ASP4
and contact Darren or one of the others to find where you can
download an image with the rt-preempt kernel, 
or even just the kernel, modules etc and install it into your
existing image


  do I correctly understand that switching from
xenomai to rt-preempt on an existing installation is impossible
and I need to install a new image instead?

On Saturday, October 7, 2017 at 5:06:38 PM UTC+3, Schooner
wrote:

   I don't use a BBB, it
looks like Debian only provide i386 and amd64 rt-preempt
kernels for Stretch.

I don't know which rt kernel others have used, probably one
from here
https://elinux.org/Beagleboard:BeagleBoneBlack_Debian#Mainline_.284.14.x_lts.29

The one I used on my server is 4.4.68-ti-rt-r112, but there
are probably newer ones.
Looks like linux-image-4.9.40-bone-rt-r7 is quite new, the
BBG people have to use rt-preempt I believe to get
everything working.

Someone should be able to give you a link
In fact if you have a root around here https://rcn-ee.com/rootfs/bb.org/testing/
you may find an image which suits

rt-preempt kernels are quite easy to build, I have even
built one on a 4 core aarm64, but that has much more
processing power
than a BBB, might take quite a while on a BBB

Is there any reason why you have to use a BBB at all?
Life is much simpler with a proper computer ;\

On 07/10/17 14:39, 'Dimitar Dobrev' via Machinekit
  wrote:


  Schooner, thank you very much for all of
your help. I am definitely going to try it. Would you
have some directions about the switch to rt-preempt? Do
I need to compile it from source code or is it just a
matter of installing packages?

On Saturday, October 7, 2017 at 4:25:21 PM UTC+3,
Schooner wrote:

   In that case,
get hold of the rt-preempt kernel from Stretch and
use that.

Then you will be able to install
machinekit-rt-preempt for stretch-armhf from my repo
and it should work, with the pru_* binaries built
and
symlinked to the /usr/lib/linuxcnc/rt-preempt
directory where they can be found by hal_pru_generic

Xenomai is becoming of lessening importance, as the
rt-preempt kernel latency improves.
Because the BBB uses the PRU for step generation
etc, all that is required is a stable servo thread
which rt-preempt should be able to provide.

Bas is already using rt-preempt on his BBB, as has
Charles.
The reason for the recent commit to make
pru_generic.* available to all armhf kernel
flavours, was to allow this, 

Re: [Machinekit] Homann Designs MODIO on the BBB.

[schoone...@btinternet.com]

  
  
If you do try to build, Michaels repo which is listed in the wiki is
defunct by the look of it

You can get modbus.c and modbus.h from here 
https://github.com/stephane/libmodbus/tree/master/src

This details the agonies Kirk went through to get it working and may
be of use
https://sourceforge.net/p/emc/mailman/message/29416863/

I would just copy Makefile.modinc to your working dir and include it
specifically in the Makefile and just tell the Makefile where the
$BINDIR is

Don't know if those scripts he lists exist any more, albeit there is
something similar in comp
https://github.com/machinekit/machinekit/blob/master/src/hal/utils/comp.g#L640

Good luck.

On 09/10/17 11:00,
  schoone...@btinternet.com wrote:


  
  Does not look as though it has been touched for quite a while, you
  would need to try building the elements and see where you get.
  
  Alternately, proper MPGs are now so cheap (£13) , is it worth the
  bother?
  http://www.ebay.co.uk/itm/CNC-Manual-Rotary-Hand-Wheel-100MPG-ISM8060-5V-DC-60mm-2-36-High-Quality-/332338204332?hash=item4d60e68eac:g:1qQAAOSwHLJZjAq1
  
  On 09/10/17 01:39, John Dammeyer
wrote:
  
  

  Hi,
  I'm running the BBB with a Xylotex DB-25 Cape driving
Servos.  I've also got a Homann Designs MODIO that could
work as an MPG.  The web site: 
  http://wiki.linuxcnc.org/cgi-bin/wiki.pl?ModIO 
  has drivers for LinuxCNC from 2012 which is a while ago
now.
  A search of MachineKit group doesn't have any hits for
MODIO so I'm wondering if it's possible to add this device
to a Beagle running MachineKit. 
  Thanks
  John

-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Cross-building Debian Multi-Arch packages

[schoone...@btinternet.com]

Hi John,

The best possibility I can come up with is dh-virtualenv
http://dh-virtualenv.readthedocs.io/en/1.0/usage.html

I am far from a python guru and it looks fiendishly complicated, but 
appears to allow a virtual python environment during a package build.


Not sure however whether it requires python3.x or if it can be used for 
the exact purpose you require.


regards


On 04/10/17 19:26, John Morris wrote:
It occurs to me there's another possibility for cross-building 
Multi-Arch packages that might be somewhat simpler than the `gcc 
--sysroot` system I've been working on with Mick, but maybe a python 
guru would know better about the feasibility.


Under normal Multi-Arch cross-building circumstances, we would use 
e.g. `mk-build-deps --host-arch armhf -i`, and all armhf-arch 
dependency packages would be installed alongside the amd64 versions.  
Then we'd build the package as usual, while specifying the host arch, 
e.g. `dpkg-buildpackage -a armhf`.


However, because MK uses many python dependencies, and building 
Multi-Arch python packages is non-trivial [1], the `mk-build-deps` 
command uninstalls the amd64 build-arch `python2.7` and other packages 
in order to install the armhf host-arch deps (python-simplejson, 
libboost-python-dev, cython, python-tk, python-protobuf).  Of course 
this would break the build, since the amd64 build-arch python would be 
missing.


I wonder if a build-arch python could somehow be installed outside the 
packaging system, for example in /usr/local, and just for the package 
build, we point to that python.  Then the packaging system could be 
fooled into cross-building armhf packages, and the only antics we'd 
have to perform would be installing the alternative python version.


Does anyone think they know how to do that, install python outside of 
the packaging system?  Maybe that means build from source, or maybe it 
means install the amd64-arch package first and move it out of the way, 
or maybe use some other creative means.  This would be a really 
valuable thing to have going forward, since it would clear the path 
for cross-building packages for many Debian releases to come without 
the difficult antics of the currently proposed system.


[1]: https://wiki.debian.org/Python/MultiArch

John



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Error with CRAMPS.hal

[schoone...@btinternet.com]

  
  
Thanks for the extra info, that makes things clearer
You have installed an old Xenomai2 kernel onto Stretch
My testing repo for Stretch does not contain any packages for
  Xenomai and so you appear to have installed machinekit-rt-preempt
  instead
So when machinekit looks for the pru_generic.bin file, based upon
  the kernel in use, it does not find it and you get this error
Oct  6 23:33:14 beaglebone rtapi:0: 1:rtapi_app:10176:user
  hal_pru_generic: cant find
  /usr/lib/linuxcnc/xenomai/pru_generic.bin in
  /home/machinekit/CRAMPS or /lib/firmware/pru/
  Oct  6 23:33:14 beaglebone rtapi:0: 1:rtapi_app:10176:user hpg:
  ERROR: failed to initialize PRU

So use the official, supported image and all will work and leave
  you free to do whatever you intended with the board,
  rather than mess around trying to get the wrong packages working
  on the wrong kernel.
  Use the Jessie SD card image from Robert Nelsons site ( as
  indicated in the 'Getting Started' section of the docs)

There is no reason to use Stretch and currently many reasons not to,
which is why we are working using an unofficial testing repo at
present.
Anything there may or may not work and is subject to change or
removal without notice.

regards

On 07/10/2017 02:03, 'Dimitar Dobrev'
  via Machinekit wrote:


  
  Kernel:

Linux beaglebone 3.8.13-xenomai-r83 #1 Wed Jan 4 23:10:15 UTC
2017 armv7l GNU/Linux

Packages from:

http://deb.mgware.co.uk/

Log:

Attached.

I already have pru_generic.bin:

/usr/lib/linuxcnc/rt-preempt/pru_generic.bin
/usr/lib/linuxcnc/prubin/pru_generic.bin

Please let me know if you need anything else.


On Thursday, October 5, 2017 at 1:46:48 PM UTC+3, Schooner
wrote:

   
On 05/10/17 10:55, 'Dimitar Dobrev' via Machinekit
  wrote:


  
Hello all,



My name is Dimitar Dobrev. I work as an application
  programmer and I have tried using CRAMPS. I have
  encountered an error I would like to ask for your help
  about.

I have a BeagleBone Black with Debian Stretch. 
  


kernel?  kernel version? git version of Machinekit? etc.
etc.


  
I need to start Machinekit on to connect to a real
  machine. I started run.py and got one error which I
  solved but I am now stuck at:



CRAMPS.hal:28: insmod failed, returned -1;

rtapi_app_main(hal_pru_generic): -1 Operation
  not permitted



  


What is the rest of the error output (/var/log/linuxcnc.log
or dmesg if it caused a segfault)
insmod errors are rather vague and can be from several
causes.

Try running with 'DEBUG=5 preceding the machinekit command
and change the inifile 'debug = ' stanza to a higher number

if you don't get a cause printed.
Make sure you have rsyslogd running to log output

If you are running an older pull of machinekit, on a
rt_preempt kernel, it won't have my recent
commits which make pru_generic.bin etc available
https://github.com/machinekit/machinekit/commit/a6ac2576fc9a7f9d8a9a3f0e9da0eb244b8336d3

That is one possibility, but impossible to say with this
little info.


  
My Machinekit is installed from packages rather
  than built from source. 
  


Where did you get the packages?

There is no official support for Stretch yet.

The only packages for Stretch are in my unofficial repo.
They HAVE been tested on a BBB and work


  
I have been able to find https://groups.google.com/forum/#!msg/machinekit/RrNLUo4ASP4/840ireJEBgAJ
  and https://groups.google.com/forum/#!topic/machinekit/qncQ_Dh7V9s
  but I haven't been able to find a solution there. Any
  direction you can give me will be both useful and
  appreciated.

Thank you for 

Re: [Machinekit] Error with CRAMPS.hal

[schoone...@btinternet.com]

  
  

On 05/10/17 10:55, 'Dimitar Dobrev' via
  Machinekit wrote:


  
Hello all,



My name is Dimitar Dobrev. I work as an application
  programmer and I have tried using CRAMPS. I have encountered
  an error I would like to ask for your help about.

I have a BeagleBone Black with Debian Stretch. 
  


kernel?  kernel version? git version of Machinekit? etc. etc.


  
I need to start Machinekit on to connect to a real machine.
  I started run.py and got one error which I solved but I am now
  stuck at:



CRAMPS.hal:28: insmod failed, returned -1;

rtapi_app_main(hal_pru_generic): -1 Operation not permitted



  


What is the rest of the error output (/var/log/linuxcnc.log or dmesg
if it caused a segfault)
insmod errors are rather vague and can be from several causes.

Try running with 'DEBUG=5 preceding the machinekit command and
change the inifile 'debug = ' stanza to a higher number 
if you don't get a cause printed.
Make sure you have rsyslogd running to log output

If you are running an older pull of machinekit, on a rt_preempt
kernel, it won't have my recent
commits which make pru_generic.bin etc available
https://github.com/machinekit/machinekit/commit/a6ac2576fc9a7f9d8a9a3f0e9da0eb244b8336d3

That is one possibility, but impossible to say with this little
info.


  
My Machinekit is installed from packages rather than built
  from source. 
  


Where did you get the packages?

There is no official support for Stretch yet.

The only packages for Stretch are in my unofficial repo.
They HAVE been tested on a BBB and work


  
I have been able to find https://groups.google.com/forum/#!msg/machinekit/RrNLUo4ASP4/840ireJEBgAJ
  and https://groups.google.com/forum/#!topic/machinekit/qncQ_Dh7V9s
  but I haven't been able to find a solution there. Any
  direction you can give me will be both useful and appreciated.

Thank you for your time.



Dimitar Dobrev,

https://gitlab.com/ddobrev/QtSharp
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Example Gantry Configs

[schoone...@btinternet.com]

On 19/09/17 12:28, Charles Steinkuehler wrote:

Also, you can find a bit of documentation regarding instantiated
components here:

http://www.machinekit.io/docs/hal/instcomp_writing_a_component/

...but it looks like the newinst command didn't get added to the
online HAL docs.

Every component that is instantiated, has a description in its man page 
of how to load it.


Covers loadrt method and the newinst method
eg
http://www.machinekit.io/docs/man/man9/debounce/

However it is not covered here
http://www.machinekit.io/docs/hal/basic_hal/
and probably should be

--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Running 3D Printer using Machinekit + CRAMPS + BBB

[schoone...@btinternet.com]

  
  

On 04/09/17 16:46,
  schoone...@btinternet.com wrote:


  
  
  On 04/09/17 16:38, 'machinen' via
Machinekit wrote:
  
  
Oh, thanks.
  
  I think that the 14-something serial is from the February
  image, trying on newer I have gotten something like:
  
  
  Depends:
  machinekit (= 0.1.1504202989-1mk.travis.luminizepatch1.git59e91621~1jessie) but 0.1.1504208693-1mk.travis.pr1265.git00ce6961~1jessie is to be
  installed



  
  
  Yes that is the error you should have had.
  Neither of those packages should be in the repo.
  
  There is a new build in progress to bring all the serials up to a
  new number, which should solve the issues.
  Takes a couple of hours to fully complete however.


The new packages are built.
The serial string you should now get is 0.1.1504538312-1mk.travis.master.gitde24b276

Note machinekit-xenomai for Jessie armhf now depends upon the
correct serial version of machinekit
Package: machinekit-xenomai
Source: machinekit
Version: 0.1.1504538312-1mk.travis.master.gitde24b276~1jessie
Architecture: armhf
Maintainer: John Morris <j...@dovetail-automata.com>
Installed-Size: 6529
Depends: machinekit (= 0.1.1504538312-1mk.travis.master.gitde24b276~1jessie), libavahi-client3 (>= 0.6.16), libavahi-common3 (>= 0.6.16), libc6 (>= 2.15), libczmq2 (>= 2.2.0), libgcc1 (>= 1:4.4.0), libprotobuf9, libstdc++6 (>= 4.3.0), libudev1 (>= 183), libuuid1 (>= 2.16), libxenomai1, libzmq3 (>= 3.2.3+dfsg), yapps2-runtime, xenomai-runtime
Enhances: machinekit
Breaks: machinekit-dev
Provides: machinekit-rt-threads
Priority: extra
Section: misc
Filename: pool/main/m/machinekit/machinekit-xenomai_0.1.1504538312-1mk.travis.master.gitde24b276~1jessie_armhf.deb
Size: 1008094
SHA256: 164585cc5441756c2b2be72c26477079cce42526365d6219152264e3f0fc6f14
SHA1: 159446e0c8f1f2b2950d3a5de1b167df39c10daa
MD5sum: e0814dc15139ef0372e44d4baa24ee14
Description: PC based motion controller for real-time Linux
 Machinekit is the next-generation Enhanced Machine Controller which
 provides motion control for CNC machine tools and robotic
 applications (milling, cutting, routing, etc.).
 .
 This package provides components and drivers that run on a Xenomai
 realtime system, with userspace threads.

Hopefully your problems will be one fewer now  ;-) 

  
  

  
  
  On Monday, September 4, 2017 at 5:13:34 PM UTC+2, Schooner
  wrote:
  
 OK I have found the
  problem.
  You were unlucky, because it does not affect every flavour
  and OS
  
  A problem whereby a PR submission got  pushed into the
  repo before merge, has screwed some of the packages files.
  The package serials for machinekit-xenomai and machinekit
  don't match so dependencies can never be met
  
  I cannot however find any reference to 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie

  and have no idea where that
  came from.
  
  I am going to have to rebuild the repo manually, removing
  the rogue packages back to the last proper merge.
  
  Don't attempt any further update etc. until you hear back.
  
  
  
  On 04/09/17 15:48, 'machinen' via Machinekit wrote:
  
  
I actually looked at the repository and
  source list as first thing, but I didn't see anything
  odd. I am for this endeavour using the "latest"
  official machinekit image 
  bone-debian-8.7-machinekit-armhf-2017-02-12-4gb.img.xz from elinux.org/Beagleboard:BeagleBoneBlack_Debian which is from february so I am not surprised that it is old.

However in sources.list there is:

deb http://httpredir.debian.org/debian/ jessie main contrib non-free
#deb-src http://httpredir.debian.org/debian/ jessie main contrib non-free

deb http://httpredir.debian.org/debian/ jessie-updates main contrib non-free
#deb-src http://httpredir.debian.org/debian/ jessie-updates main contrib non-free

deb http://security.debian.org/ jessie/updates main contrib non-free
#deb-src http://security.debian.org/ jessie/updates main contrib non-free

#deb http://httpredir.debian.org/debian jessie-backports main contrib non-free
##deb-src http://httpredir.debian.org/debian jessie-backports main contrib non-free

deb [arch=armhf] http://deb.machinekit.io/debian jessie main
#deb-src [arch=armhf] http://deb.machinekit.io/debian jessie main

#Kernel source (repos.rcn-ee.com) : https://github.com/RobertCNe

Re: [Machinekit] Re: Running 3D Printer using Machinekit + CRAMPS + BBB

[schoone...@btinternet.com]

  
  

On 04/09/17 16:38, 'machinen' via
  Machinekit wrote:


  Oh, thanks.

I think that the 14-something serial is from the February image,
trying on newer I have gotten something like:


Depends:
machinekit (= 0.1.1504202989-1mk.travis.luminizepatch1.git59e91621~1jessie) but 0.1.1504208693-1mk.travis.pr1265.git00ce6961~1jessie is to be
installed
  
  
  


Yes that is the error you should have had.
Neither of those packages should be in the repo.

There is a new build in progress to bring all the serials up to a
new number, which should solve the issues.
Takes a couple of hours to fully complete however.


  


On Monday, September 4, 2017 at 5:13:34 PM UTC+2, Schooner
wrote:

   OK I have found the
problem.
You were unlucky, because it does not affect every flavour
and OS

A problem whereby a PR submission got  pushed into the repo
before merge, has screwed some of the packages files.
The package serials for machinekit-xenomai and machinekit
don't match so dependencies can never be met

I cannot however find any reference to 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie
and have no idea where that
came from.

I am going to have to rebuild the repo manually, removing
the rogue packages back to the last proper merge.

Don't attempt any further update etc. until you hear back.



On 04/09/17 15:48, 'machinen' via Machinekit wrote:


  I actually looked at the repository and
source list as first thing, but I didn't see anything
odd. I am for this endeavour using the "latest" official
machinekit image 
bone-debian-8.7-machinekit-armhf-2017-02-12-4gb.img.xz from elinux.org/Beagleboard:BeagleBoneBlack_Debian which is from february so I am not surprised that it is old.

However in sources.list there is:

deb http://httpredir.debian.org/debian/ jessie main contrib non-free
#deb-src http://httpredir.debian.org/debian/ jessie main contrib non-free

deb http://httpredir.debian.org/debian/ jessie-updates main contrib non-free
#deb-src http://httpredir.debian.org/debian/ jessie-updates main contrib non-free

deb http://security.debian.org/ jessie/updates main contrib non-free
#deb-src http://security.debian.org/ jessie/updates main contrib non-free

#deb http://httpredir.debian.org/debian jessie-backports main contrib non-free
##deb-src http://httpredir.debian.org/debian jessie-backports main contrib non-free

deb [arch=armhf] http://deb.machinekit.io/debian jessie main
#deb-src [arch=armhf] http://deb.machinekit.io/debian jessie main

#Kernel source (repos.rcn-ee.com) : https://github.com/RobertCNelson/linux-stable-rcn-ee
#
#git clone https://github.com/RobertCNelson/linux-stable-rcn-ee
#cd ./linux-stable-rcn-ee
#git checkout `uname -r` -b tmp
#
deb [arch=armhf] http://repos.rcn-ee.com/debian/ jessie main
#deb-src [arch=armhf] http://repos.rcn-ee.com/debian/ jessie main

And the sources.list.d folder is empty. And I didn't change anything.




On Monday, September 4, 2017 at 3:14:57 PM UTC+2,
Schooner wrote:

   Look at http://deb.machinekit.io/debian/pool/main/m/machinekit/

There is no package 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie
  that

  serial is ancient.

You need to check your /etc/apt/sources.list and the
files in /etc/apt/sources.list.d and see where this
is coming from

I don't know how you can be getting this error, if
your one and only machinekit reference is deb.machinekit.io, as it should
be.
The package list there only refers to 0.1.1504202989

Have to suspect that your image is old and also
points to deb.dovetail-automata.com or similar,
but that serial is not there either AFAICT

As far as FDM stuff goes I have no idea I'm afraid,
don't use it, hopefully someone else can talk you
through it once you get machinekit back running.



On 04/09/17 13:38, 'machinen' via Machinekit
  wrote:
 

Re: [Machinekit] Re: Running 3D Printer using Machinekit + CRAMPS + BBB

[schoone...@btinternet.com]

  
  
OK I have found the problem.
You were unlucky, because it does not affect every flavour and OS

A problem whereby a PR submission got  pushed into the repo before
merge, has screwed some of the packages files.
The package serials for machinekit-xenomai and machinekit don't
match so dependencies can never be met

I cannot however find any reference to 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie and have
no idea where that came from.

I am going to have to rebuild the repo manually, removing the rogue
packages back to the last proper merge.

Don't attempt any further update etc. until you hear back.



On 04/09/17 15:48, 'machinen' via
  Machinekit wrote:


  I actually looked at the repository and source list
as first thing, but I didn't see anything odd. I am for this
endeavour using the "latest" official machinekit image 
bone-debian-8.7-machinekit-armhf-2017-02-12-4gb.img.xz from elinux.org/Beagleboard:BeagleBoneBlack_Debian which is from february so I am not surprised that it is old.

However in sources.list there is:

deb http://httpredir.debian.org/debian/ jessie main contrib non-free
#deb-src http://httpredir.debian.org/debian/ jessie main contrib non-free

deb http://httpredir.debian.org/debian/ jessie-updates main contrib non-free
#deb-src http://httpredir.debian.org/debian/ jessie-updates main contrib non-free

deb http://security.debian.org/ jessie/updates main contrib non-free
#deb-src http://security.debian.org/ jessie/updates main contrib non-free

#deb http://httpredir.debian.org/debian jessie-backports main contrib non-free
##deb-src http://httpredir.debian.org/debian jessie-backports main contrib non-free

deb [arch=armhf] http://deb.machinekit.io/debian jessie main
#deb-src [arch=armhf] http://deb.machinekit.io/debian jessie main

#Kernel source (repos.rcn-ee.com) : https://github.com/RobertCNelson/linux-stable-rcn-ee
#
#git clone https://github.com/RobertCNelson/linux-stable-rcn-ee
#cd ./linux-stable-rcn-ee
#git checkout `uname -r` -b tmp
#
deb [arch=armhf] http://repos.rcn-ee.com/debian/ jessie main
#deb-src [arch=armhf] http://repos.rcn-ee.com/debian/ jessie main

And the sources.list.d folder is empty. And I didn't change anything.




On Monday, September 4, 2017 at 3:14:57 PM UTC+2, Schooner
wrote:

   Look at http://deb.machinekit.io/debian/pool/main/m/machinekit/

There is no package 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie
  that
  serial is ancient.

You need to check your /etc/apt/sources.list and the files
in /etc/apt/sources.list.d and see where this is coming from

I don't know how you can be getting this error, if your one
and only machinekit reference is deb.machinekit.io, as it should be.
The package list there only refers to 0.1.1504202989

Have to suspect that your image is old and also points to deb.dovetail-automata.com or similar, but that
serial is not there either AFAICT

As far as FDM stuff goes I have no idea I'm afraid, don't
use it, hopefully someone else can talk you through it once
you get machinekit back running.



On 04/09/17 13:38, 'machinen' via Machinekit wrote:


  Somehow I didn't have the fdm folder on my
system - probably from mucking around. However I did a
new flash of clean bone-debian-8.7-machinekit-armhf-2017-02-12-4gb.img.xz
image and sure there it is. Unfortunately when the first
thing when I log into this new install I do is run
machinekit and select the Fabrikator Mini CRAMPS
configuration, then try the ./run.py script I still get
the same error.

And the combination of sudo apt-get remove machinekit
machinekit-xenomai/update/install
machinekit-xenomai only removes machinekit packages from
system and when installing it still falls on the same
error. I will try purge next with new clean install.

On Monday, September 4, 2017 at 10:02:12 AM UTC+2,
Schooner wrote:

   
On 03/09/17 23:37, 'machinen' via Machinekit
  wrote:


  The dist-aupgrade removes
machinekit packages from system. There is this
problem:

The following packages
have 

Re: [Machinekit] Re: Running 3D Printer using Machinekit + CRAMPS + BBB

[schoone...@btinternet.com]

  
  
Look at http://deb.machinekit.io/debian/pool/main/m/machinekit/

There is no package 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie
  that serial is
  ancient.

You need to check your /etc/apt/sources.list and the files in
/etc/apt/sources.list.d and see where this is coming from

I don't know how you can be getting this error, if your one and only
machinekit reference is deb.machinekit.io, as it should be.
The package list there only refers to 0.1.1504202989

Have to suspect that your image is old and also points to
deb.dovetail-automata.com or similar, but that serial is not there
either AFAICT

As far as FDM stuff goes I have no idea I'm afraid, don't use it,
hopefully someone else can talk you through it once you get
machinekit back running.



On 04/09/17 13:38, 'machinen' via
  Machinekit wrote:


  Somehow I didn't have the fdm folder on my system -
probably from mucking around. However I did a new flash of clean
bone-debian-8.7-machinekit-armhf-2017-02-12-4gb.img.xz image and
sure there it is. Unfortunately when the first thing when I log
into this new install I do is run machinekit and select the
Fabrikator Mini CRAMPS configuration, then try the ./run.py
script I still get the same error.

And the combination of sudo apt-get remove machinekit
machinekit-xenomai/update/install machinekit-xenomai only
removes machinekit packages from system and when installing it
still falls on the same error. I will try purge next with new
clean install.

On Monday, September 4, 2017 at 10:02:12 AM UTC+2, Schooner
wrote:

   
On 03/09/17 23:37, 'machinen' via Machinekit wrote:


  The dist-aupgrade removes machinekit
packages from system. There is this problem:

The following packages have
unmet dependencies:
 machinekit-xenomai : Depends: machinekit (= 0.1.1504202989-1mk.travis.luminizepatch1.git59e91621~1jessie) but 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie is to be installed
E: Unable to correct problems, you have held broken
packages.
  
  

Problem is that as I don't speak Python, I don't know
where the file (assembly?) fdm.config should be or
better yet from which file it is attempting to
load/import.
  


I don't speak parseltongue either, but this is nothing to do
with python

This is the package list for Jessie armhf
http://deb.machinekit.io/debian/dists/wheezy/main/binary-armhf/Packages

As you will see the machinekit-xenomai and machinekit
versions are both at 0.1.1504202989, so if your apt
is trying to fetch 0.1.1486851828
  it is working from an old package list,
which means it has not been updated properly.

apt-get remove machinekit machinekit-xenomai
apt-get update
apt-get install machinekit-xenomai

should hopefully restore things.  ( If apt uninstalls other
packages when you do the uninstall, make sure you install
those again too)

As for where the fdm stuff is installed, /usr/lib/python2.7/dist-packages/fdm
It has not changed for yonks, so updating packages achieves
nothing.



 -- 
  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 machinekit+...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit 

Re: [Machinekit] Re: Running 3D Printer using Machinekit + CRAMPS + BBB

[schoone...@btinternet.com]

  
  

On 04/09/17 09:02,
  schoone...@btinternet.com wrote:


  
  
  On 03/09/17 23:37, 'machinen' via
Machinekit wrote:
  
  
The dist-aupgrade removes machinekit packages
  from system. There is this problem:
  
  The
  following packages have unmet dependencies:
   machinekit-xenomai : Depends:
  machinekit (= 0.1.1504202989-1mk.travis.luminizepatch1.git59e91621~1jessie) but 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie is to be
  installed
  E: Unable to
  correct problems, you have held
  broken packages.


  
  Problem is that as I don't speak Python, I don't know where
  the file (assembly?) fdm.config should be or better yet from
  which file it is attempting to load/import.

  
  
  I don't speak parseltongue either, but this is nothing to do with
  python
  
  This is the package list for Jessie armhf
  http://deb.machinekit.io/debian/dists/wheezy/main/binary-armhf/Packages
  
  As you will see the machinekit-xenomai and machinekit versions are
  both at 0.1.1504202989, so if your apt is trying to fetch
  0.1.1486851828 it is working from an old
package list, which means it has not
  been updated properly.
  
  apt-get remove machinekit machinekit-xenomai
  apt-get update
  apt-get install machinekit-xenomai
  
  should hopefully restore things.  ( If apt uninstalls other
  packages when you do the uninstall, make sure you install those
  again too)
  
  As for where the fdm stuff is installed, /usr/lib/python2.7/dist-packages/fdm
  It has not changed for yonks, so updating packages achieves
  nothing.


FYI, as I noted to someone else, but off forum, in case this is how
it has come about.

"The situation described could normally only be encountered where
the user has not updated apt-get, but tries to install a later
version of a package directly, via dpkg.
This results in the dependencies being unable to be met, because the
packages known about do not include it."

 
  
  
   -- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.
  
  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Running 3D Printer using Machinekit + CRAMPS + BBB

[schoone...@btinternet.com]

  
  

On 03/09/17 23:37, 'machinen' via
  Machinekit wrote:


  The dist-aupgrade removes machinekit packages from
system. There is this problem:

The following packages
have unmet dependencies:
 machinekit-xenomai : Depends:
machinekit (= 0.1.1504202989-1mk.travis.luminizepatch1.git59e91621~1jessie) but 0.1.1486851828-1mk.travis.master.gitcbd2dfd3~1jessie is to be
installed
E: Unable to
correct problems, you have held broken
packages.
  
  

Problem is that as I don't speak Python, I don't know where the
file (assembly?) fdm.config should be or better yet from which
file it is attempting to load/import.
  


I don't speak parseltongue either, but this is nothing to do with
python

This is the package list for Jessie armhf
http://deb.machinekit.io/debian/dists/wheezy/main/binary-armhf/Packages

As you will see the machinekit-xenomai and machinekit versions are
both at 0.1.1504202989, so if your apt is trying to fetch 0.1.1486851828 it is working from an old package
  list, which means it has not been updated
properly.

apt-get remove machinekit machinekit-xenomai
apt-get update
apt-get install machinekit-xenomai

should hopefully restore things.  ( If apt uninstalls other packages
when you do the uninstall, make sure you install those again too)

As for where the fdm stuff is installed, /usr/lib/python2.7/dist-packages/fdm
It has not changed for yonks, so updating packages achieves nothing.




  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: DE10-DE0-Nano - Invitation to collaborate

[schoone...@btinternet.com]

On 01/09/17 15:45, Charles Steinkuehler wrote:
...but since it's already working, I think programming the FPGA via 
overlays should remain supported for folks who aren't trying to use 
things like HDMI. 


My thoughts for what they are worth.

I would have no intention of ever using HDMI from a board with 1GB SDRAM 
and no GPU.

http://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=English=205=1046=2

I am not sure we should even encourage it by making it available!
Just asks for loads of "Axis locks up when I load my half billion line 
printer file" threads.


It certainly needs to be able to be disabled or not used, with 
preferably the inconvenience of reboot etc. being for those seeking to 
use HDMI :-P



--
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] Re: Will DE0-Nano image work on a DE10-Nano board?

[schoone...@btinternet.com]

  
  

On 18/08/17 05:06, mugginsac wrote:


  By the way it with with the USB cable connected to
the D525MW and running lsusb it shows up on bus 2 as device 32
and is correctly identified as Altera (09fb:6810).
But I don't know how to get a /dev/ttyUSB? or some other serial
device handle.
  


cat /sys/class/tty/ttyUSB*
will usually bring up the ports that exist (ie are in use with a
device plugged into them)

The DE0-Nano usually just came up as /dev/ttyUSB0 or ttyUSB1 AFAIR

Certainly no exotic ports like some of the Arduinos

regards

  
Alan

  
  -- 
  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 machinekit+unsubscr...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  




-- 
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 machinekit+unsubscr...@googlegroups.com.
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Re: [Machinekit] how to compile source in machinekit

[schoone...@btinternet.com]

  
  

On 15/08/17 03:29, thangle wrote:


  

  
The module appears to have built from your printout, but you
also look to be building a RIP and then executing a binary
installation
Look at the rpath
  

 
i dont know what exactly i should do?



   
You need to build the RIP, set the current terminal session
environment using . ./scripts/rip-environment  from the root
of the RIP
and then launch machinekit from that terminal
session.
  



when i launch machinekit from terminal, the CRAMPS worked.
  but when i launch machinekit from the one (in the screenshot)
  it's bugged. why are they difference?
  


Because you have changed one but not the other.
You have 2 different machinekit installations, one in the binary
file system and another, which you built, in your home
directory.

If you want to test your new *kins, run the one which contains it,
in your home dir, from a terminal.

I won't even try to explain how to add your new module to the binary
file system, there is no need to do it.

The whole point about a RIP build is that it is self contained and
you can experiment without messing up the main installation.


  



   
Either that or copy the new module to your binary install
and run ldconfig manually and check it is in the cache
  



i dont know where is the new module, binary place? how to
  check it in cache by ldconfig?


thank you for your help very much.   

   

  

       rpath=/usr/lib/linuxcnc/xenomai
       See /var/log/linuxcnc.log for more
information.
       10047
       PID TTY      STAT   TIME COMMAND  
       Stopping realtime threads
       Unloading hal components
  
  
  i'm using BBB and machinekit. sorry for my trash
English.
  
  
  Vào 15:41:12 UTC+7 Thứ Hai, ngày 14 tháng 8 năm
2017, Schooner đã viết:

   Going to
need a lot more info than that and to see the
source.

You basically need to know how to write
Makefiles, to alter the existing ones to build
your stuff too, assuming it is compatible.

On 14/08/17 03:39, thangle wrote:


  Hi,


Does anyone know how to compile a new
  source? example, i coppy a new  C project
  from an old project in machinekit src and
  rename, what should i do to make it work.
  please help me. tks 
  
  -- 
  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 machinekit+...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  

  

  
  -- 
  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 machinekit+...@googlegroups.com.
  Visit this group at https://groups.google.com/group/machinekit.
  For more options, visit https://groups.google.com/d/optout.


  
 

Re: [Machinekit] how to compile source in machinekit

[schoone...@btinternet.com]

  
  

On 14/08/17 17:02, thangle wrote:


  sorry, here is my makefile. i just add 
         
obj-m += trivkinstest.o
  trivkinstest-objs := emc/kinematics/trivkinstest.o
  $(RTLIBDIR)/trivkinstest$(MODULE_EXT): $(addprefix
$(OBJDIR)/,$(trivkinstest-objs))
  
  
  
  can you give more details how should i do to build a RIP
  

  


You are already doing it
- 3rd: use terminal go to /home/machinekit/machinekit/src and
  "make" then type command "sudo make setuid"
  
You need to run what you built, not what is installed,
as per my previous reply


  

  or run idconfig mannually. Sorry, i 'm a new in linux and
BBB, trying to learn them

  


man ldconfig  (starts with an L, not an i )

  

  
  
  
  
  
  Vào 22:39:17 UTC+7 Thứ Hai, ngày 14 tháng 8 năm 2017, Schooner
  đã viết:
  
 
  On 14/08/17 16:22, thangle wrote:
  
  
Here is what i did:
  
    - 1st: i make a coppy trivkins.c then
  rename it become trivkinstest.c  and changing line
  88 to: static const char *name = "trivkinstest";


    - 2nd: i follow this 
         https://groups.google.com/forum/#!msg/machinekit/yrBnTxQL7_E/IrPHxKn3BQAJ;context-place=forum/machinekit
  

  
  
  I wrote that, so I am familiar with it
  
  

  
         + i pass Submakefile
         + just change something in Makefile 

  

  
  
  'just change something in makefile' is not information
  
  

  
    - 3rd: use terminal go to /home/machinekit/machinekit/src
  and "make" then type command "sudo make setuid"

      - 4th: i try to test if trivkinstest work. i
  open CRAMPS.hal, change 
           loadrt trivkins -->loadrt
  trivkinstest 
           loadrt [EMCMOT]EMCMOT
  servo_period_nsec=[EMCMOT]SERVO_PERIOD
  num_joints=[TRAJ]AXES tp=tp
  kins=trivkins--->trivkinstest


    - finally: i try run CRAMPS but it's
  bugged 
     Debug file information:
     stat: No such file or directory
     CRAMPS.hal:18: insmod failed, returned -1:
     do_load_cmd: dlopen: trivkinstest.so:
  cannot open shared object file: No such file or
  directory
  

  
  
  The module appears to have built from your printout, but
  you also look to be building a RIP and then executing a
  binary installation
  Look at the rpath
  
  You need to build the RIP, set the current terminal
  session environment using . ./scripts/rip-environment 
  from the root of the RIP
  and then launch machinekit from that terminal
  session
  
  Either that or copy the new module to your binary install
  and run ldconfig manually and check it is in the cache
  
  

  
     rpath=/usr/lib/linuxcnc/xenomai
     See /var/log/linuxcnc.log for more
  information.
     10047
     PID TTY      STAT   TIME COMMAND  
     Stopping realtime threads
     Unloading hal components


i'm using BBB and machinekit. sorry for my
  trash English.


Vào 15:41:12 UTC+7 Thứ Hai, ngày 14 tháng 8 năm
  2017, Schooner đã viết:
  
 Going to
  need a lot more info than that and to see the
  source.
  
  You basically need to know how to write
  Makefiles, to alter the existing ones to build