OK, this is the first time you have said you are actually building a
combined RIP of machinekit-hal and machinekit-cnc, not machinekit.
The build_with_cnc script DOES copy the scripts.
First it removes the section of the paths in linuxcnc script, that
lead to the machinekit-cnc clone and edits it to point to
machinekit-hal
https://github.com/machinekit/machinekit-hal/blob/master/scripts/build_with_cnc#L56
Then it copies all scripts except 2 specifically excluded, to
machinekit-hal/scripts.
https://github.com/machinekit/machinekit-hal/blob/master/scripts/build_with_cnc#L57
If it did not, the combined build could never have worked.
I suspect that either you were not calling .
./scripts/rip-environment from the terminal session of the build and
then running from that session,
or that somehow your PATH var retained /bin ahead of the added paths
from the RIP and executed /bin first.
Run 'echo $PATH' before calling 'machinekit and see what is in it.
On 10/07/19 05:08, Ryan Helminen wrote:
I think I finally figured out where the source of my
initial problems were. ArcEye has been working on splitting up
the machinekit repo into machinekit-cnc and machinekit-hal.
The linuxcnc bash script that I was supposed to be calling
into does exist under the scripts folder in machinekit-cnc,
but that specific location doesn't get copied over to
machinekit-hal when running the build_with_cnc script under
the machinekit-hal/scripts folder. As a result, I was instead
calling the linuxcnc script under the /usr/bin folder which
had old settings in it.
I haven't tested my theory, but from looking through the
build_with_cnc script I am not seeing the machinekit-cnc
scripts folder get copied over to the machinekit-hal scripts
folder. This will cause problems because the correct linuxcnc
bash script is only in the machinekit-cnc repo.
On Friday, May 31, 2019 at 11:18:49 PM UTC-5, Ryan Helminen
wrote:
Hello Gentlemen,
Thank you for the guidance. In the end it looks like
the problem was that the linuxcnc bash script was not
setup to respect any RIP settings and had the paths
hardcoded to the root /usr directory.
I had to update the first part of the script to check
for the EMC2_HOME environment variable, and if it existed
set the paths accordingly. My updated linuxcnc executable
is at the following gist:
Best Regards,
Ryan
On Wednesday, May 29, 2019 at 6:16:08 AM UTC-5, Schooner
wrote:
Can you attach your
copy of the linuxcnc script.
That is where a lot of paths are set, the rpath looks
like the path for an install, not a RIP.
On 29/05/19 05:20, Ryan Helminen wrote:
Hello,
I am wondering how to get the mkwrapper
approach to using QtQuickVCP to respect RIP
environment variables so that the correct
environment is loaded. From what I can tell the
run.py script that I am using detects the RIP
environment variables, but for some reason that
isn't getting communicated during the call to the
linuxcnc executable.
I have a custom build that includes a bipodkins
RT module and when I try to start things up using
the attached python script I get the following
error message:
bipodsim.hal:5: insmod failed, returned -1:
do_load_cmd: dlopen: bipodkins.so: cannot open
shared object file: No such file or directory
rpath=/usr/lib/linuxcnc/rt-preempt
The rpath that is referenced is not the
path to the RIP files. I have tested that RIP is
correctly setup by manually running halrun and
doing a loadrt bipodkins which works as expected.
Does anyone have any tips on how I can force
mkwrapper to correctly use the RIP environment
that I have configured?
Thanks much,
Ryan
--
website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit
---
You received this message because you are subscribed
to the Google Groups "Machinekit" group.
To unsubscribe from this group and stop receiving
emails from it, send an email to [email protected].
Visit this group at https://groups.google.com/group/machinekit.
To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/6844b1f8-9718-4d8d-b022-1b7ef1d5dc49%40googlegroups.com.
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 [email protected].
Visit this group at https://groups.google.com/group/machinekit.
To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/6ab3bd6e-3a37-45c9-b8d0-e4717fcb3bae%40googlegroups.com.
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 [email protected].
Visit this group at https://groups.google.com/group/machinekit.
To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/5D258FB8.2070408%40gmail.com.
For more options, visit https://groups.google.com/d/optout.
|
