#linuxcnc-devel Logs

Jun 15 2021

#linuxcnc-devel Calendar

01:32 AM -!- #linuxcnc-devel mode set to +v by ChanServ
01:48 AM mrec: does anyone know what halui is-auto is for?
02:49 AM mrec: ok small issues and they're fixed
02:49 AM mrec: https://forum.linuxcnc.org/24-hal-components/42740-xhc-whb04b-4-6-issues#212091
02:49 AM mrec: https://github.com/LinuxCNC/linuxcnc/issues/1187
03:50 AM Tom_L: halui.mode.auto bit in
03:50 AM Tom_L: pin for requesting auto mode
03:50 AM Tom_L: halui.mode.is-auto bit out
03:50 AM Tom_L: pin for auto mode is on
04:26 AM mrec: I fixed it already.
04:26 AM mrec: the modifications are in the forum for discussion
04:27 AM mrec: the pendant works smooth now but I'd like to fix my backlash now and that's not working it seems
05:53 AM linuxcnc-build: build #6165 of 4008.deb-precise-amd64 is complete: Failure [4failed apt-get-update shell_2] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4008.deb-precise-amd64/builds/6165 blamelist: alkabal <33371787+alkabal@users.noreply.github.com>, andypugh <andy@bodgesoc.org>
05:53 AM linuxcnc-build: build #6165 of 4007.deb-precise-i386 is complete: Failure [4failed apt-get-update shell_2] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4007.deb-precise-i386/builds/6165 blamelist: alkabal <33371787+alkabal@users.noreply.github.com>, andypugh <andy@bodgesoc.org>
05:53 AM linuxcnc-build: build #4996 of 4009.deb-precise-rtai-i386 is complete: Failure [4failed apt-get-update shell_2] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4009.deb-precise-rtai-i386/builds/4996 blamelist: alkabal <33371787+alkabal@users.noreply.github.com>, andypugh <andy@bodgesoc.org>
11:38 AM mrec: andypugh: G71 does not allow to decrease X on a lathe?
11:38 AM mrec: I just swapped my object around and G71 won't do a thing... G70 would work
11:39 AM mrec: if X is getting bigger (the radius is getting bigger G71 also works..)
11:42 AM mrec: endless loop...
11:44 AM mrec: axis 1.9gb
11:44 AM mrec: memory
11:44 AM mrec: holy G71 ...
11:44 AM mrec: I think I'll trash the idea of using it and build my own path
11:58 AM andypugh: There is an issue raised about the endless loop. I haven’t got round to looking at it. (Not my code, you know)
11:59 AM andypugh: mrec: Don’t suppose you feel like looking at the code to see where it goes wrong? Seeing as how you have it all set up and showing the problem?
12:49 PM mrec: I sorted it out it was my mistake; well first you need to know the bugs in order to fix them
12:50 PM mrec: I did not retract X at the end of it - however the endless loop which catches up the memory should be avoided with some sanity check
12:52 PM mrec: XHC-WHB04B that one works smooth now I'm using it for a few hours already
12:52 PM mrec: https://github.com/LinuxCNC/linuxcnc/issues/1187
01:35 PM norias: hi
02:51 PM rene-dev: andypugh if you have a running rtai system, can you test this branch? https://github.com/LinuxCNC/linuxcnc/issues/1188
03:14 PM andypugh: rene-dev: I do have an RTAI system, but it has a Yapps problem. And a tkinter problem
03:25 PM andypugh: https://paste.ubuntu.com/p/VZjhSXxtXp/
03:25 PM andypugh: I only get the tkinter problem with python3, but the yapps problem is there with any python and both your branch and master,
03:25 PM rene-dev: is python3-tk installed?
03:25 PM rene-dev: and install python-yapps python3-yapps and yapps2
03:30 PM andypugh: I didn’t have python3-tk, so I think that needs to be added to the debian control file
03:30 PM rene-dev: maybe make clean should also invoke pyclean
03:31 PM rene-dev: I should have known this, I had this issue many times when I did the python stuff
03:39 PM andypugh: The new branch still fails with Pluto Servo: https://paste.ubuntu.com/p/pxrMyqRvDj/
03:53 PM rene-dev: argh
03:56 PM rene-dev: I was just guessing
03:56 PM rene-dev: I dont really understand what the code does
03:56 PM andypugh: It seems to generate firmware.
03:57 PM andypugh: We could just expunge Pluto. I doubt it has many current users.
04:03 PM rene-dev: while I agree, its probably just a simple 1 line fix
04:03 PM rene-dev: I do have 2 plutos :D
04:11 PM andypugh: I am pretty sure that it’s the $< > $@ in the Submakefile
04:14 PM andypugh: Maybe not, it only looks like gibberish, but is apparently makefile gibberish.
04:17 PM rene-dev: how can it be a makefile problem?
04:21 PM andypugh: rbf2h.py is called by the Submakedfile and it looks like one of the passed parameters isn’t right.
04:22 PM andypugh: (This is a guess)
04:24 PM andypugh: in rbf2h.py there is: h = open(args[0], 'rb').read();
04:25 PM andypugh: And that is trying to parse arguments from: https://github.com/LinuxCNC/linuxcnc/blob/master/src/hal/drivers/Submakefile#L1
04:28 PM jepler: andypugh: I can look at python3 problems in rbf2h if that's helpful
04:29 PM andypugh: That would be helpful, yes.
04:29 PM jepler: this is for python3-only world right? no 2/3 compat needed?
04:30 PM andypugh: Is it possible that ord(c) was never working but that py2 was more forgiving?
04:30 PM jepler: or we can delete it ....?
04:31 PM jepler: the elements of a b''-string in python2 are themselves length-1 b''-string, while the elements of a b''-string in python3 are simply a number.
04:31 PM andypugh: Pluto in general or rbf2h?
04:31 PM jepler: delete pluto
04:32 PM jepler: pluto doesn't work without rbf2h since that's integral to how the firwmare gets to the board on each startup
04:32 PM jepler: andypugh: what branch should I PR to?
04:33 PM andypugh: Ask rene-dev :-)
04:33 PM jepler: rene-dev: ^^^
04:33 PM rene-dev: If it fixes the problem master
04:33 PM jepler: master or master-gtk3 I assume?
04:33 PM andypugh: It’s either master or rtai-python3
04:33 PM jepler: OK
04:33 PM jepler: it'll break python2
04:33 PM rene-dev: Then master-gtk3
04:34 PM rene-dev: now you have 3 branches to pick ;D
04:38 PM jepler: I see now that andypugh had almost the whole solution already :-/
04:38 PM jepler: thanks for letting me feel special thouhg
04:39 PM Tom_L: special? sure... divine... maybe not
04:40 PM Tom_L: i took the bot off freenode
04:41 PM Tom_L: seems it's collapsed anyway
04:41 PM andypugh: Well, you could explain what “$< > $@“ does in this case. I found out that $< will become the file name (hal/drivers/pluto_servo_rbf.h) and that $@ becomes the name of the file being built, but the > in the middle has me stumped. Is that passed as a parameter or does it pipe?
04:51 PM jepler: that's the shell redirection character ">"
04:51 PM jepler: so it gives the input filename as a positional parameter to the script (argv) and takes the output and puts it in the target file
04:52 PM andypugh: OK, and that runs before passing the result as a paramter to the rbf2h.py file?
04:53 PM andypugh: That’s a bit circuitous...
04:53 PM jepler: It's a fair cop
04:57 PM andypugh: Well, it worked for many years, but then confused me. I am easily confused.
05:08 PM jepler: echo hi > filename.txt
05:09 PM jepler: once you have disentangled it as the shell redirection character rather than a special make character I hope it's not too big a secret .. but the script & makefile can always be improved if you have a yen to
05:23 PM andypugh: I am rather short of yens where Pluto is concerned.
05:24 PM andypugh: I have lots of other things queued up.
05:32 PM rene-dev: thanks jepler
05:47 PM andypugh: Aye, I dont think I would have got to that solution.