#emc-devel | Logs for 2010-06-18

[01:28:42] <CIA-21> EMC: 03cmorley 07v2.4_branch * rfdb519f4d0ae 10/src/emc/usr_intf/pncconf/pncconf.py: fix instance names of abs,scale and mux8
[01:28:43] <CIA-21> EMC: 03cmorley 07v2.4_branch * r013db35c3280 10/src/emc/usr_intf/pncconf/pncconf.py: fix missing axis index-enable and add PID index-enable signals
[01:28:44] <CIA-21> EMC: 03cmorley 07v2.4_branch * refaf11229446 10/src/emc/usr_intf/pncconf/pncconf.py: fix loading of postgui_hal. fix signalnames add shutdown.hal
[01:28:45] <CIA-21> EMC: 03cmorley 07v2.4_branch * re3041ce361a8 10/src/emc/usr_intf/pncconf/pncconf.py: fix whitespace errors
[05:10:53] <ries_> ries_ is now known as ries
[05:21:18] <ries_> ries_ is now known as ries
[05:26:31] <ries_> ries_ is now known as ries
[06:21:47] <ries_> ries_ is now known as ries
[08:15:12] <ries_> ries_ is now known as ries
[08:19:51] <ries_> ries_ is now known as ries
[08:34:30] <ries_> ries_ is now known as ries
[10:44:54] <micges_work1> mozmck: it seems that my 10.04 rt installation was hanged by plugging and unplugging ethernet cable, since then it works fine (36h)
[10:47:11] <jthornton> making progress on 10.04rt?
[11:36:00] <micges_work1> jthornton: here works fine except one issue
[11:40:32] <jthornton> has he updated the kernel lately?
[11:40:52] <jthornton> mine shows 2.6.32-22-rtai
[11:41:25] <jthornton> aww I remember now I have 4 cores on this one
[13:12:13] <skunkworks> Well - I think all the ducks are in a row for the fest. :)
[13:18:04] <CIA-21> EMC: 03jthornton 07v2.4_branch * rf1f32273a81c 10/docs/src/config/ini_config.lyx: add a bit more info on axis numbers
[13:52:36] <cradek> jmkasunich: ja3 merges into master fairly trivially, and it runs - acceleration seems wrong (way too low) but I haven't looked into it
[13:54:21] <SWPadnos> oh, there was a question yesterday about combined XYZ + UVW moves
[13:54:36] <cradek> what about them?
[13:54:58] <SWPadnos> shouldn't those moves, or combined linear/rotary moves, still limit feed rate based on the accel/vel limits of all the axes in use?
[13:55:39] <SWPadnos> apparently they weren't doing that for someone with a hot wire cutter, using XYUV moves
[13:56:35] <cradek> do you know how to reproduce it?
[13:56:43] <SWPadnos> hmmm. not exactly
[13:57:02] <cradek> usually a complaint like this comes from a misunderstanding of what feed rate means for complicated moves like that
[13:57:02] <SWPadnos> I suspect that a sample config with low accel on UVW would show the problem
[13:57:06] <SWPadnos> sure
[13:57:20] <cradek> what version?
[13:57:22] <SWPadnos> the person said something like "the XY move completes, then UV catch up
[13:57:36] <SWPadnos> that I don't know either
[13:57:48] <cradek> I find that quite hard to believe :-/
[13:57:57] <SWPadnos> yeah, it sounded odd
[13:58:26] <cradek> software stepgen?
[13:59:12] <cradek> a silly accel constraint outside of emc combined with a silly high ferror limit could cause this
[13:59:24] <SWPadnos> ok, the user was tvc, yesterday around 16:30 EST
[13:59:41] <SWPadnos> err, 16:22
[14:00:17] <SWPadnos> no, 16:14-ish :)
[14:04:06] <cradek> SWPadnos: I don't read anything that says XY and UV lost synchronization and ended at different times
[14:04:43] <cradek> I'm having trouble understanding what he was actually seeing
[14:05:43] <SWPadnos> well, it is hard to understand
[14:06:00] <SWPadnos> [2010-06-17 16:14:59] <tvc> Seems like the feedrates are doing funny stuff on me, the X and Y are acting as masters and U and V are then forced to catch up, I would have expected the X Y to slow down but seems they are acting as masters. Does this sound familiar to anyone?
[14:06:21] <SWPadnos> I don't quite know what "catch up" means there, it's ambiguous
[14:07:04] <cradek> yes I don't know what "as masters" means
[14:07:05] <SWPadnos> or unclear at least
[14:07:22] <SWPadnos> I took that to mean that the feed rate is controlled by XYZ, which is as it should be
[14:08:51] <cradek> he did later quote the feed rate section of the manual which explicitly says in an XYUV move XY moves in specified F and UV follows it
[14:08:59] <cradek> yes, what you said
[14:09:08] <cradek> (I'm on hotel internet again)
[14:09:25] <SWPadnos> heh
[14:09:44] <cradek> ... so as to start and end in synchronization while obeying all axis constraints
[14:09:52] <SWPadnos> sadly, I usually get better internet access in hotels than I get at home - I've only got 2Mb/256k DSL here
[14:09:59] <SWPadnos> yep
[14:10:17] <skunkworks> cradek: are you alread there?
[14:10:21] <SWPadnos> where are you guys staying?
[14:10:27] <cradek> we're in cleveland
[14:10:28] <SWPadnos> (now, not in Ann Arbor)
[14:10:43] <skunkworks> ah - are you going to hgr?
[14:10:51] <cradek> yes, this afternoon
[14:11:03] <SWPadnos> oh, did you not stop at the radio store in Lima?
[14:11:07] <cradek> SWPadnos: a random super 8 kind of near jmk's house
[14:11:12] <SWPadnos> ah, ok
[14:11:22] <cradek> SWPadnos: no, maybe we'll go monday??
[14:11:35] <SWPadnos> there's a costco almost next door to him, so maybe we can get some good coffee :)
[14:11:46] <SWPadnos> oh, is it on the way-ish to Ann Arbor?
[14:11:49] <cradek> SWPadnos: fwiw, I put a low accel on V in axis_9axis.ini and everything looks right
[14:11:54] <SWPadnos> ok
[14:12:12] <cradek> SWPadnos: only for suitably low definitions of "ish"
[14:12:19] <SWPadnos> heh, yeah
[14:13:41] <skunkworks> ROAD TRIP!
[14:14:24] <SWPadnos> I guess I should pack or something
[14:14:37] <cradek> I started at 10pm the night before we left :-/
[14:14:43] <cradek> in other news, I'm an idiot
[14:14:45] <skunkworks> how much weight can jeplers car haul? ;)
[14:14:51] <SWPadnos> I would be starting at 10AM on the day I'm leaving ;)
[14:14:59] <cradek> skunkworks: at what speed?
[14:15:02] <skunkworks> heh
[14:15:02] <SWPadnos> since I head out tonight
[14:15:04] <SWPadnos> hah
[14:16:07] <cradek> I think he told me the capacity *including people* is 800lb or some crazy low number like that
[14:16:29] <cradek> at least I think that was him - maybe it was my car
[14:16:39] <SWPadnos> not surprising, my mini-van was either 1150 or 1500 lbs
[14:17:01] <SWPadnos> which is ludicrous for a "7-passenger van"
[14:17:18] <cradek> do you think they mean including people or not?
[14:17:24] <SWPadnos> yes, definitely
[14:17:30] <cradek> hmm
[14:17:31] <skunkworks> heh - actually I remember that from one the previous fest - - that's 2 americans or 6 of any other coutnry
[14:17:39] <cradek> 7 europeans maybe :-)
[14:17:44] <SWPadnos> heh
[14:18:10] <SWPadnos> since the way back seat can only hold 3 small children (or 2 cramped adults), it may be KO
[14:18:12] <SWPadnos> OK
[14:18:15] <cradek> I wonder if anyone's ever trieed to rigid tap with a lefthand tap
[14:18:26] <cradek> or tried
[14:18:27] <skunkworks> do you see a bug?
[14:18:45] <cradek> JT-Hardinge reported that lathe threading doesn't work in M4
[14:19:14] <JT-Hardinge> yes he did
[14:19:25] <cradek> good morning
[14:19:37] <JT-Hardinge> robh said he tried to left hand tap and it would not work either
[14:19:42] <JT-Hardinge> good morning
[14:19:54] <cradek> ok, I'm not surprised at that
[14:20:16] <mozmck_work> I tried to right hand tap and it didn't work either
[14:20:28] <cradek> what
[14:20:31] <JT-Hardinge> I make a lot of tie rods with left hand on one end out of hex material
[14:20:31] <skunkworks> well - someone should have said something.. ;)
[14:20:35] <mozmck_work> ...but that was by hand and I had a left hand tap :)
[14:20:42] <JT-Hardinge> lol
[14:20:48] <cradek> who even has a lefthand tap!?
[14:21:00] <SWPadnos> Careful Numbskull Control
[14:21:04] <JT-Hardinge> oh, me I have several
[14:21:07] <cradek> brb
[14:21:10] <mozmck_work> :) Had one laying about and didn't know it!
[14:21:29] <SWPadnos> that would sure make me question my coffee that day
[14:21:39] <SWPadnos> "how come it's not working?
[14:21:41] <SWPadnos> :
[14:21:43] <SWPadnos> L
[14:21:45] <SWPadnos> "
[14:21:50] <SWPadnos> like today apparently
[14:21:53] <mozmck_work> :)
[14:22:04] <SWPadnos> bbiab, time to pack or eat or make coffee or something
[14:25:43] <skunkworks> Eat or make coffee? that is hardcore
[14:26:09] <SWPadnos> eh, whichever works :)
[14:26:26] <SWPadnos> I don't have to pack coffee, because I can buy that locally
[14:50:29] <CIA-21> EMC: 03cradek 07v2.4_branch * r57c8a3892418 10/src/emc/kinematics/tp.c: Fix threading and tapping with spindle turning backward (M4)
[14:53:10] <JT-Hardinge> YEA!
[14:53:37] <cradek> how embarassing
[15:02:26] <JT-Hardinge> what is?
[15:02:40] <cradek> that I forgot about that for so long
[15:03:31] <JT-Hardinge> ah I wish mux16 and the direct feed overide was on 2.4 so I could use it on the hardine
[15:04:31] <JT-Hardinge> but for mechanical switches like mine it should be a grayscale converter instead of mux16
[15:04:52] <cradek> oh I can merge that into master if you want to test it (which would be great)
[15:05:07] <cradek> my fix, I mean
[15:05:12] <JT-Hardinge> ok, that would be cool
[15:06:33] <JT-Hardinge> * JT-Hardinge wobbles back out into the heat to put in the last three screws of the day fixing the garden shed...
[15:33:58] <CIA-21> EMC: 03cradek 07master * r672a3da080b6 10/src/emc/usr_intf/axis/scripts/image-to-gcode.py: Fix Python exception when using lace bounding
[15:34:01] <CIA-21> EMC: 03cradek 07master * ra9ed62ca7de5 10/docs/src/hal/rtcomps.lyx: add to description latency
[15:34:01] <CIA-21> EMC: 03cradek 07master * r176ab760c3f2 10/docs/src/code/emc2-motion-joint-controller-block-diag.png: try and get the image to display better in the pdf
[15:34:02] <CIA-21> EMC: 03cradek 07master * rc341121cd440 10/src/emc/usr_intf/emcrsh.cc: emcrsh: getopt_long() fix longopts termination
[15:34:03] <CIA-21> EMC: 03cradek 07master * rf6eb24834185 10/docs/src/code/Code_Notes.lyx: fix some spelling
[15:34:03] <CIA-21> EMC: 03cradek 07master * r0dea04057323 10/src/hal/components/stepgen.c: fix ctrl_type on certain 2.6.3x kernels; allow empty ctrl_types
[15:34:04] <CIA-21> EMC: 03cradek 07master * r2f96e7c39f23 10/tcl/bin/emccalib.tcl: emccalib: fix crash 'unmatched open quote in list'
[15:34:05] <CIA-21> EMC: 03cradek 07master * r61605f1d860c 10/nc_files/lathe_pawn.ngc: this code assumes radius mode
[15:34:06] <CIA-21> EMC: 03cradek 07master * racc8e11689a4 10/docs/src/gcode/main.lyx: file name requirements for called files, g2/3 offsets in g18
[15:34:07] <CIA-21> EMC: 03cradek 07master * ra8e9cb1989d8 10/docs/src/lathe/lathe-user.lyx: add info about arcs and update tool table info
[15:34:08] <CIA-21> EMC: 03cradek 07master * rfdb519f4d0ae 10/src/emc/usr_intf/pncconf/pncconf.py: fix instance names of abs,scale and mux8
[15:34:10] <CIA-21> EMC: 03cradek 07master * refaf11229446 10/src/emc/usr_intf/pncconf/pncconf.py: fix loading of postgui_hal. fix signalnames add shutdown.hal
[15:34:12] <CIA-21> EMC: 03cradek 07master * r013db35c3280 10/src/emc/usr_intf/pncconf/pncconf.py: fix missing axis index-enable and add PID index-enable signals
[15:34:13] <CIA-21> EMC: 03cradek 07master * r57c8a3892418 10/src/emc/kinematics/tp.c: Fix threading and tapping with spindle turning backward (M4)
[15:34:14] <CIA-21> EMC: 03cradek 07master * rf1f32273a81c 10/docs/src/config/ini_config.lyx: add a bit more info on axis numbers
[15:34:15] <CIA-21> EMC: 03cradek 07master * r3b250c99073d 10/ (13 files in 12 dirs): Merge remote branch 'origin/v2.4_branch'
[15:34:22] <CIA-21> EMC: 03cradek 07master * re3041ce361a8 10/src/emc/usr_intf/pncconf/pncconf.py: fix whitespace errors
[15:36:43] <micges> cradek: in ja3 you must define vel and acc to joints and axes and traj to achieve sane results, see sim/gantry_mm.ini in ja3
[15:40:00] <cradek> ok maybe they're defaulting to 1 or something (I ran sim/axis)
[15:40:36] <micges> probably yes
[15:53:13] <skunkworks> logger_dev: bookmark
[15:53:13] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emcdevel/2010-06-18.txt
[15:56:56] <JT-Hardinge> cradek: it just sits at the start point with M4 still
[15:57:43] <JT-Hardinge> I did a git pull but I'm in a branch off of master...
[15:58:23] <SWPadnos> cherry-pick the commit to your branch
[15:58:43] <jepler> JT-Hardinge: if you're in a branch besides v2.4_branch or master, then git won't have given you the fix automatically
[15:58:50] <JT-Hardinge> ok
[15:58:57] <JT-Hardinge> I was wondering about that
[15:59:23] <jepler> to get the change you could rebase onto master or you could cherry pick
[16:00:17] <JT-Hardinge> if I rebase will that overwrite changes that I have committed in my local branch?
[16:03:27] <jepler> rebase takes each change you've committed and re-does it but using a new starting point. The first pair of diagrams in 'git --help rebase' show what it will be doing in this case
[16:03:43] <JT-Hardinge> thanks jepler
[16:04:01] <jepler> right now you're at a place like C but you want the change introduced at F as well. Rebase takes ABC (your changes) and moves them so they're after G, so you get the change at F that you want to have
[16:04:51] <jepler> so make sure your work is committed, then 'git rebase origin/master'
[16:05:16] <JT-Hardinge> ok thanks... gotta write this good stuff down
[16:08:37] <JT-Hardinge> I see what you mean now when I look at it with git gui it put my changes at the top
[16:09:50] <skunkworks> JT-Hardinge: if you are running dev - could you look and see if your classic ladder has a print button in the gui?
[16:10:04] <JT-Hardinge> ok
[16:10:16] <JT-Hardinge> rebuilding at the moment
[16:11:05] <JT-Hardinge> yes it has a print button inbetween Preview and About
[16:11:32] <skunkworks> hmm - I wonder why mine doesnt.
[16:11:37] <skunkworks> It is in licid
[16:11:40] <skunkworks> lucid
[16:12:01] <JT-Hardinge> 8.04.4 here
[16:12:33] <jepler> it requires libgnomeprintuimumble-dev to get the print button
[16:12:44] <JT-Hardinge> cradek: M4 first test runs fine, I'll test more later I have to run
[16:12:55] <jepler> JT-Hardinge: thank you, see you later
[16:13:13] <JT-Hardinge> thanks for helping me wade through it :)
[16:13:13] <cradek> thanks for testing it
[16:13:19] <JT-Hardinge> bbl
[17:11:25] <skunkworks_> I am getting this now (trying to rebuild emc2) configure: error: Required GL library missing. Install it, or specify --disable-python to skip the parts of emc2 that depend on Python
[17:17:02] <micges> skunkworks_: it's rebuild of working run in place?
[17:17:24] <skunkworks> yes
[17:17:37] <micges> weird
[17:18:07] <micges> sudo apt-get build-dep emc2 should show missing packets
[17:21:16] <skunkworks_> E: Unable to find a source package for emc2
[17:23:44] <skunkworks> which I guess makes sense because there isn't a lucid package.
[17:23:46] <andypugh> It's broken, buy a new one...
[17:23:54] <skunkworks> heh
[17:25:01] <micges> skunkworks:you're on lucid rt trying to compile emc?
[17:25:09] <andypugh> sudio apt-get install emc2-dev ?
[17:25:24] <skunkworks_> http://pastebin.ca/1885888
[17:26:34] <skunkworks> nope - simulator
[17:27:49] <andypugh> You need someone with a clue, I will bow out and back to vanilla #emc
[17:27:58] <skunkworks> heh
[17:30:04] <SWPadnos> somewhere there should be a list of packages needed to build (or install) on lucid, considering that mozmck has made packages available
[17:30:37] <skunkworks> I was going by this http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Ubuntu10.04Notes
[17:30:50] <skunkworks> I got it to build once a week or so ago.
[17:30:53] <skunkworks> ;)
[17:31:23] <SWPadnos> is libglu1-mesa-dev still installed?
[17:31:25] <skunkworks> I was trying to rebuild it because I installed the libgnomeprintuimumble-dev
[17:32:24] <micges> skunkworks: please paste configure context of your error
[17:34:55] <micges> skunkworks: whole configure output I mean
[17:35:12] <skunkworks> ok
[17:36:23] <skunkworks_> http://pastebin.ca/1885892
[17:37:55] <micges> try libgl1-mesa-glx
[17:37:59] <micges> brb
[17:40:17] <skunkworks> says it is already at the newest version
[17:49:07] <SWPadnos> skunkworks, try installing xlibmesa-gl-dev and maybe xlibmesa-glu-dev (or xlibmesa3, or the other couple of xlibmesa* packages)
[17:49:46] <skunkworks> will do
[17:50:20] <SWPadnos> hmmm
[17:50:32] <SWPadnos> it looks like that only installs a doc file in /usr/share/doc
[17:52:54] <SWPadnos> have you tried make clean / make, or autogen?
[17:54:45] <skunkworks> I tried the make clean - then the ./autogen.sh
[17:54:54] <skunkworks> ./configure --enable-run-in-place --enable-simulator
[17:55:50] <skunkworks> I did all of those.. xlib - still same error
[17:59:02] <SWPadnos> strange. I don't get that error - just managed to build sim on my new laptop (for the first time :) )
[17:59:28] <SWPadnos> not bad - ~2 minutes, but that's without docs or hardware drivers
[18:01:15] <SWPadnos> oooh, even better with -j4, 1m 9s
[18:12:37] <micges> skunkworks: maybe try python-opengl
[18:12:49] <micges> it has some libs on dependencies
[18:24:06] <skunkworks> Hold on - I just installed the closed source ati driver and now no error. building
[18:24:15] <skunkworks> (the ubuntu certified)
[18:26:59] <skunkworks> hmm - now segfault ;)
[18:29:37] <skunkworks> now it is running
[18:30:32] <skunkworks> hey - and the print button is there!
[18:30:51] <skunkworks> jepler was right!
[18:42:58] <skunkworks> and now I uninstalled the closed source driver - and it still works
[18:44:09] <skunkworks> (re-boot fixed the segfault issue)
[19:31:50] <skunkworks> hmm ati has a new catlyist driver (10.6 - the one that quite working for me was 10.5) seems to install and boot :)
[19:32:17] <skunkworks> I had downloaded 10.5 just a few days ago.
[19:32:42] <skunkworks> emc still runs :)
[19:33:33] <micges> good
[20:23:55] <andypugh> I just noticed that the working bldc pattern for my motor is one that ought to be totally wrong, as it includes an invalid pattern, hence project Capacitor Death is on hold.
[20:48:00] <JT-Hardinge> cradek: I've tried threading every way I can think of inside, outside, left hand, right hand, M3, M4 all seem to work well
[20:49:55] <skunkworks> sweet!
[20:57:11] <JT-Hardinge> and so does G33.1 work for both left and right hand taps... as they say down in Arkansas Sweeeeeet!
[21:00:56] <KimK> JT-Hardinge: front turret/tool, back turret/tool? Lathe "right side up", lathe "upside down"?
[21:08:36] <JT-Hardinge> front turret I think it is upside down as the imaginary Y points toward the floor???
[21:08:53] <andypugh> "freqgen may be eliminated at any time, and almost certainly will be eliminated for the version 2.2 release of EMC."
[21:09:05] <JT-Hardinge> but I was doing some back threading and turning the other day
[21:09:55] <JT-Hardinge> KimK: it is a Hardinge CHNC I
[21:11:20] <SWPadnos> you can use stepgen in v mode instead of freqgen
[21:11:28] <KimK> JT-Hardinge: OK. I just wanted to encourage testing in every possible manner. And thanks for testing, btw.
[21:17:50] <JT-Hardinge> your welcome
[21:21:53] <KimK> Hi seb. So you're not coming to the workshop? Or have you decided yet?
[21:24:32] <seb_kuzminsky> I'm coming
[21:24:44] <seb_kuzminsky> i'll be there wednesday evening through saturday morning
[21:24:50] <seb_kuzminsky> i'm looking forward to it!
[21:26:38] <andypugh> Despite the threat from someone to gaffer-tape you to a chair until you do the SPI driver?
[21:27:14] <andypugh> (I think that was the driver, and the tape-type, and you. I could be wrong on all counts)
[21:28:48] <KimK> Great, seb, I'll look forward to seeing you there. (With or without the duct tape and chair.)
[21:29:01] <seb_kuzminsky> heh
[21:29:34] <seb_kuzminsky> i'd been planning to set up some lucid buildslaves, and maybe set up buildbots for heekscad and ocl, but maybe spi would be more useful
[21:30:52] <KimK> me: Hi seb, great to see you here at the workshop!
[21:30:54] <KimK> seb: Mmhph! Mmmph!
[22:17:14] <seb_kuzminsky> where are you all staying for the workshop? at the university?
[22:40:50] <SWPLinux_> SWPLinux_ is now known as SWPLinux
[22:59:55] <seb_kuzminsky> laters yo