February 2006 - Bdi4emc-help - lists.ourproject.org

Re: [Bdi4emc-help] xfce and axis
by jepler＠unpythonic.net 17 Feb '06

17 Feb '06

On Thu, Feb 16, 2006 at 11:25:17PM -0500, Matt Timmermans wrote: > It looks like the problem is a call in line 640-something in axis, in the > update method. It tells something to call the same method again in 20ms > and then does the update. If it takes more than 20ms to do the update, bad > things ensue, including deadlocks on some semaphores. You might be able to > replicate the problem by setting the delay to 1ms or so. > > When I moved the enqueing call to the end of the update method, axis > started to work just fine. I don't know whether or not even that is > completely safe, however. Are "TkIntr" callbacks single-threaded? I'm glad you could find a change that made axis work for you. Knowing the change you made, I hope I'll be able to understand the bug better. Tk (and Tkinter, which just a Python wrapper for Tk) is based on an event loop, not threads. Normally, an event scheduled with "after" will be handled only after the code currently running returns to the event loop, so it should be safe to schedule the next "after" at any time in the current invocation. There are several Tk commands that can cause events to be handled immediately ("update", "vwait", and "tkwait" come to mind), but I didn't believe any of them were called in the affected code. The reason the next 'after' was scheduled at the very top is this: If an error happens somewhere inside the callback this time, I wanted to be confident that there was a next invocation already scheduled. However, in practice there are not runtime exceptions in this code, and ther are other ways to ensure this (like using try/finally). I intend to fix this bug in the upcoming AXIS 1.2 alpha 3, but I don't know if paul_c has any plans to package newer versions of AXIS for bdi. If you want to build your own version of axis on bdi, you first have to build a version of emc. To build the same version that is installed by default, follow these directions: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?BDI-4_Install (install extra development packages, then edit sources.list, then apt-get source emc, configure, and make), then follow the instructions for building axis here: http://axis.unpythonic.net/installing (those instructions refer to emc2 but should also work for the version of emc on bdi) If you have trouble building axis, please let me know, either directly or on this list. Thanks for your kind words about AXIS. Jeff

1 0

Lack of var values in debug output
by Gene Heskett 17 Feb '06

17 Feb '06

Hi All I've looked at that .h file with the debug defs in it, but I can't seem to find a bit that would, at each line of code, output the var number and its value for all vars used in the current line of gcode. Having that would make troubleshooting a lot easier. Something that looked like: line (# agrees with vim) #14=0.3119, #12=0.525, #13=0.075 Is there such a capability? -- Cheers, Gene People having trouble with vz bouncing email to me should add the word 'online' between the 'verizon', and the dot which bypasses vz's stupid bounce rules. I do use spamassassin too. :-) Yahoo.com and AOL/TW attorneys please note, additions to the above message by Gene Heskett are: Copyright 2006 by Maurice Eugene Heskett, all rights reserved.

1 0

Re: [Bdi4emc-help] xfce and axis
by jepler＠unpythonic.net 17 Feb '06

17 Feb '06

On Wed, Feb 15, 2006 at 11:12:25PM -0500, Matt Timmermans wrote: > Is there a known problem running axis under xfce? It starts up and eats > cpu, but there's no window. My screen resolution ix 800x600, if that could > possibly make a difference. I haven't used xfce, but AXIS should work just fine on 800x600 screens. If you start emc from a terminal, then hit control-c after AXIS has started, do you get a Python traceback? If so, what does it say? It might look something like this, and I want to see all the lines from 'Traceback' to 'KeyboardInterrupt': Traceback (most recent call last): File "/usr/bin/axis", line 2068, in ? o.mainloop() File "/usr/lib/python2.4/lib-tk/Tkinter.py", line 965, in mainloop self.tk.mainloop(n) File "/usr/lib/python2.4/lib-tk/Tkinter.py", line 1340, in __call__ def __call__(self, *args): KeyboardInterrupt (this is a traceback from a version of AXIS that was running normally at the time I hit control-c) On my own system (Ubuntu 5.10, emc2 CVS HEAD, AXIS 1.2a2, 1920x1200 screen), AXIS seems to work normally if I choose the 'xfce' session when I log in. However, this system is very differeent from a bdi system, so it may not be a very useful data point. Unfortunately, my only bdi system doesn't have an attached monitor, so it's difficult for me to test your configuration myself. Jeff

2 1

Quickstep driver
by Doug Goff 17 Feb '06

17 Feb '06

Hi Matt, Are you running the quickstep driver under the real-time kernel? As I understand it, the normal performance measurement programs are not aware of the real-time kernel usage, and may erroneously report it. Paul_C can you clarify? -- Regards, Doug Goff dgoff(a)comporium.net www.woodntractors.com Scale Tractors In Wood "The man who sets out to carry a cat by its tail learns something that will always be useful and which never will grow dim or doubtful." - Mark Twain

3 4

xfce and axis
by Matt Timmermans 16 Feb '06

16 Feb '06

Is there a known problem running axis under xfce? It starts up and eats cpu, but there's no window. My screen resolution ix 800x600, if that could possibly make a difference. -- Matt

1 0

BDI TESTING won't test
by Gene Heskett 15 Feb '06

15 Feb '06

Greetings John & company; I maybe didn't do it right, but testing won't run, returning this stanza on launching it, as root, built as gene after doing a chown gene:gene to the whole tree as root before I did the make clean && make as gene: ============ shop:/home/gene/emc2# scripts/emc Machine configuration directory is '/home/gene/emc2/configs/stepper/' Machine configuration file is 'stepper_inch.ini' Starting emc... insmod: can't read '/rtapi.ko': No such file or directory insmod: can't read '/hal_lib.ko': No such file or directory ERROR: Could not load 'rtapi' ERROR: Could not load 'hal_lib' Realtime system did not load Shutting down and cleaning up EMC... Could not find pid(s) for task classicladder Cleanup done ========== looks like that / in front of the name may be the perp? rtapi.ko exists with this timestamp in rtapi the dir: -rw-r--r-- 1 gene gene 31521 2006-02-14 22:33 rtlib/rtapi.ko And: shop:/home/gene/emc2# ls -l lib/* -rw-rw-r-- 1 gene gene 72296 2006-02-06 22:29 lib/hal_lib.o so it appears both modules are there, but not found because of the leading slash used makes it search / for those files. One thing I didn't do was to nuke the config.status etc files and rerun ./configure before doing the make, should I have? Its late or I'd test that yet tonight. -- Cheers, Gene People having trouble with vz bouncing email to me should add the word 'online' between the 'verizon', and the dot which bypasses vz's stupid bounce rules. I do use spamassassin too. :-) Yahoo.com and AOL/TW attorneys please note, additions to the above message by Gene Heskett are: Copyright 2006 by Maurice Eugene Heskett, all rights reserved.

4 8

bdi install
by Graham Foulkes 15 Feb '06

15 Feb '06

Hi There I am new to linux and the BDI community, brought up on assembler and DOS. I downloaded a copy of bdi 4.20 and installed it on a Dell optiplex with bootable cdrom bios support. It runs a couple of steppers great on the test bench but I want to install a copy on an older m/c which does not support cd boot. There must be a way to boot off floppy and then point to the cdrom drive to start to install. Sorry this is such a trivial question but the answer does not appear in faqs I have read and my knowledge of linux is very basic. Any advice? Regards. Graham Foulkes.

3 2

Installation Exception
by Peter Skerda 14 Feb '06

14 Feb '06

Greetings - This Old Gezzer has only recently tried to become familiar with Linux BDI & make Installation on a CNC System. Hope I got the picture right? Installation is on an OS free system no other systems on HD. Using Linux BDI & EMC 4.18 to install. Piece of cake - Right? Can get to the Graphical Interface (x) configuration, easy enough, then (hope you all are ready for this) get this window stating that an exception occurred. "an unhandled exception has occurred. This is most likely a bug." Also in another screen get the message - Init: cannot execute "/etc/X11/prefdm" - Init: Id "x" respawning too fast : disabled for 5 minutes. That's the end of the story & installation. Didn't mean to write a novel but any assistance or direction will be very much appreciated. Thank you all very much! Pepper

3 5

virtual machine
by Martin Monteverde 14 Feb '06

14 Feb '06

Hi, I'm new with EMC. My first question is if it's possible to run EMC inside a virtual machine. I know about the real time kernel and the real time nature of the application. I'm not thinking running EMC inside a VM to control a real machine, but for evaluation and (some) development purpose only. Thanks in advance and regards, Martin

2 1

parallel port init makes a step
by Matt Timmermans 13 Feb '06

13 Feb '06

When I shutdown EMC, and then start it up again, the process of releasing and re-acquiring the parallel port seems to generate a signal that get interpreted as a step. Is there a way to keep Linux from messing with the parallel port when EMC releases it? -- Matt

3 4