#emc-devel | Logs for 2011-04-11

Back
[00:04:08] -!- theorbtwo has quit [Ping timeout: 260 seconds]
[00:04:14] theorb is now known as theorbtwo
[00:09:56] -!- EDocTooR has quit [Quit: Leaving]
[00:10:59] -!- Gensor has quit []
[00:12:12] -!- Gensor [[email protected]] has joined #emc-devel
[00:14:28] -!- EDocTooR [[email protected]] has joined #emc-devel
[00:16:42] -!- mozmck has quit [Quit: Leaving.]
[00:18:48] -!- mozmck [[email protected]] has joined #emc-devel
[01:09:50] -!- crazy_imp has quit [Ping timeout: 258 seconds]
[01:19:18] -!- EDocTooR_ [[email protected]] has joined #emc-devel
[01:22:19] -!- EDocTooR has quit [Ping timeout: 246 seconds]
[01:23:07] -!- uwe_ has quit [Ping timeout: 276 seconds]
[01:29:26] -!- uwe_ has quit [Ping timeout: 276 seconds]
[01:41:25] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[02:08:44] -!- EDocTooR__ [[email protected]] has joined #emc-devel
[02:11:41] -!- EDocTooR_ has quit [Ping timeout: 240 seconds]
[02:19:23] -!- JCSpresser has quit [Remote host closed the connection]
[02:50:38] EDocTooR__ is now known as EDocTooR
[02:55:19] -!- mhaberler [[email protected]] has joined #emc-devel
[02:57:58] -!- EDocTooR has quit [Quit: Leaving]
[02:58:21] -!- EDocTooR [[email protected]] has joined #emc-devel
[03:13:15] -!- mhaberler has quit [Quit: mhaberler]
[03:18:58] -!- EDocToor_ [[email protected]] has joined #emc-devel
[03:20:26] -!- EDocToor_ has quit [Client Quit]
[03:22:26] -!- EDocTooR has quit [Quit: Leaving]
[03:28:08] -!- EDocTooR [[email protected]] has joined #emc-devel
[03:37:52] -!- EDocTooR has quit [Quit: Leaving]
[03:38:23] -!- cblack001 [[email protected]] has joined #emc-devel
[03:38:38] -!- EDocTooR [[email protected]] has joined #emc-devel
[04:07:20] -!- stormlight has quit [Quit: stormlight]
[04:08:53] -!- mhaberler [[email protected]] has joined #emc-devel
[04:15:47] -!- Athlocatle has quit [Quit: �����]
[04:27:29] -!- EDocTooR has quit [Quit: Leaving]
[04:28:02] -!- EDocTooR [[email protected]] has joined #emc-devel
[04:28:34] -!- EDocTooR has quit [Read error: Connection reset by peer]
[04:29:06] -!- EDocTooR [[email protected]] has joined #emc-devel
[05:07:05] -!- ve7it has quit [Remote host closed the connection]
[05:26:02] -!- Valen has quit [Quit: Leaving.]
[05:39:53] -!- mhaberler has quit [Quit: mhaberler]
[06:00:24] -!- mozmck has quit [Quit: Leaving.]
[07:05:57] -!- isssy has quit [Client Quit]
[07:09:52] -!- capricorn_one has quit [Remote host closed the connection]
[07:13:25] -!- stormlight has quit [Quit: stormlight]
[07:17:05] -!- mhaberler [[email protected]] has joined #emc-devel
[07:28:36] -!- mhaberler has quit [Quit: mhaberler]
[08:09:18] -!- awallin_ [awallin_!~quassel@2001:708:110:1020:224:7eff:feda:7c7d] has joined #emc-devel
[08:18:39] -!- cradek has quit [Read error: Connection reset by peer]
[08:18:44] -!- cradek [[email protected]] has joined #emc-devel
[08:46:39] -!- robh__ [[email protected]] has joined #emc-devel
[08:51:45] -!- KimK has quit [Ping timeout: 252 seconds]
[08:54:40] -!- maximilian_h [[email protected]] has joined #emc-devel
[09:36:30] -!- Valen has quit [Client Quit]
[10:16:18] -!- Paragon39 has quit [Quit: ChatZilla 0.9.86.1 [Firefox 3.5.18/20110319140258]]
[10:24:45] -!- mhaberler [[email protected]] has joined #emc-devel
[10:25:10] -!- psha [[email protected]] has joined #emc-devel
[10:39:49] <jthornton> psha: the links are now working for me too
[10:56:31] -!- sumpfralle has quit [Read error: Operation timed out]
[11:02:31] <mhaberler> pdf too?
[11:04:07] <jthornton> that's what I was checking, I didn't know the html links were broken
[11:08:02] -!- mhaberler has quit [Quit: mhaberler]
[11:14:58] <psha> jthornton: nice
[11:15:01] <jthornton> they all seem to work now
[11:15:08] <psha> i've to fix images/other blocks
[11:15:15] <psha> i beth image links in pdf are broken
[11:15:23] <psha> bet
[11:16:23] <jthornton> yes, caption links are borked
[11:25:46] <jthornton> I still wonder what the red box is around all the links in the pdf's
[11:25:58] <jthornton> is it supposed to be highlighted or something?
[11:40:37] <psha> i guess yes
[11:40:43] <psha> it's supposed to be highlighted
[11:41:30] <jthornton> with the stock Ubuntu 10.04 viewer it shows up as a red box that surrounds the text
[11:41:52] <psha> same here
[11:41:53] <psha> (evince)
[11:42:03] <jthornton> yes evince
[11:42:19] <psha> in older docs links were just numbers or hyperlinks (clickable)?
[11:42:42] <psha> red border in xpdf too
[11:43:16] <jthornton> in the older docs it just showed the section number in blue and you could click on that
[11:56:28] <jthornton> what tool generates the pdf's?
[11:57:08] <psha> dblatex
[12:14:39] -!- atom1 has quit [Disconnected by services]
[12:14:45] Tom_sbc is now known as atom1
[12:16:01] -!- mhaberler [[email protected]] has joined #emc-devel
[12:17:31] <psha> jthornton: it's controlled in css
[12:17:34] <psha> oops
[12:17:36] <psha> in .styl
[12:17:37] <psha> sty
[12:20:16] -!- pjm has quit [Ping timeout: 276 seconds]
[12:24:34] <psha> hm, hal/rtcomps.txt is not included in Master_Hal?
[12:27:26] <psha> mhaberler: psha/doc-link-colors
[12:27:30] <psha> oops
[12:27:34] <psha> jthornton: that was for you
[12:29:14] <mhaberler> cool!
[12:31:47] <psha> mhaberler: maybe you'll commit your doc changes?
[12:31:53] <psha> then i'll rebase on top of you
[12:32:19] <mhaberler> tools or content?
[12:32:33] <psha> content
[12:32:40] <psha> tools rarely cause conflicts
[12:32:45] <mhaberler> tools still requires mscgen, content too
[12:35:36] <mhaberler> I could push the pngs and just refer to those for now
[12:38:51] <psha> do you think that embeding mcs in docs is better then out-of-doc .mcs files?
[12:40:44] <jthornton> psha: hal/rtcomps is part of the integrators manual
[12:41:03] <mhaberler> dont care about that - i need to write a script which just touches an existing png when mscgen not installed and creates a warning
[12:44:01] -!- skunkworks [skunkworks!447329d2@gateway/web/freenode/ip.68.115.41.210] has joined #emc-devel
[12:53:07] -!- mhaberler has quit [Quit: mhaberler]
[12:56:18] -!- mhaberler [[email protected]] has joined #emc-devel
[12:56:36] -!- mhaberler_ [[email protected]] has joined #emc-devel
[12:57:59] <mhaberler> psha: I though about it - no, too early too push. I need another week or so. Th content is inadaequate at this point.
[13:00:47] <psha> hm, then we have to somehow merge section blockid fix and your work
[13:10:44] -!- Valen has quit [Quit: Leaving.]
[13:14:37] <mhaberler> look, 'is it done'?
[13:14:54] <psha> most of section links are fixed
[13:15:08] <mhaberler> does the format complete?
[13:15:21] <psha> all sections have they blockid fixed (checked by script)
[13:15:34] <mhaberler> and how does one fix the rest - manually?
[13:15:46] <psha> rest - what? pictures etc?
[13:16:02] <mhaberler> the non-fixed blockid links
[13:18:25] <psha> blockid links are only used for sections/figures/tables
[13:18:31] <psha> all other links are just anchros
[13:18:46] <mhaberler> fine. So how the rest will be fixed
[13:19:18] <mhaberler> a later one-off edit run?
[13:19:34] <mhaberler> me or jt wading through files?
[13:31:21] <psha> i'll fix figs with script
[13:31:26] <psha> maybe tables too
[13:31:30] <mhaberler> ok
[13:41:30] -!- mk0 has quit [Ping timeout: 246 seconds]
[13:44:40] -!- mozmck [mozmck!~moses@client-173.225.233.207.dfwtx.partnershipbroadband.com] has joined #emc-devel
[13:45:37] <mhaberler> well I'll give it a second stab and then try to fix this msgen build thinge
[13:46:57] <mhaberler> psha: on a different from, it seems I need your mdi_execute_hook or similar :-/
[13:48:04] -!- stormlight has quit [Quit: stormlight]
[13:49:11] -!- stillme has quit [Ping timeout: 264 seconds]
[13:59:19] -!- nullie has quit [Quit: Ex-Chat]
[14:43:10] -!- awallin_ has quit [Remote host closed the connection]
[14:47:12] <psha> heh
[14:47:42] <mhaberler> however, it has to triggered from *witihn* interp()
[14:48:09] <mhaberler> I wish task and interp were separate threads
[14:48:55] <psha> heh, then you'll get lot of bugs :)
[14:49:27] <mhaberler> not so sure. The single process approach creates many, too. Just look at your mdi contortions.
[14:49:45] <mhaberler> a bit of wait() and signal() would go a long way
[14:51:02] <mhaberler> have a look at http://git.mah.priv.at/gitweb/emc2-dev.git/blob/c18e5e89480712e6f61dd8f321495e7d96117795:/src/emc/rs274ngc/interp_execute.cc , execute_handler()
[14:51:38] <mhaberler> after the INTERP_EXECUTE_FINISH I need to queue an execute_msg and wait from within interp
[14:52:43] <mhaberler> one option is to replicate your mdi call logic there and queue setwait/clearwait and execute from within interp
[14:54:00] <mhaberler> there are things to be done after the excute_handler() so I cant just release control there and have task do its thing
[14:54:59] <mhaberler> basically the whle thing is fine except the XECUTE_FINISH where it should , e.g., handle pin I/O or probing
[14:57:34] <mhaberler> psha: how would you grind that cat?
[15:02:39] <psha> [side note] with threads you'll easily get either deadlocks or reentrant problems...
[15:04:30] <mhaberler> well, I wasnt embarking on that journey anyway..
[15:05:07] <psha> execute finish need waiting on synch or not?
[15:05:15] <psha> or it's rescheduled?
[15:05:40] <mhaberler> I cant 'reschedule' because I'm not done in interp at that point
[15:05:43] <psha> heh, don't be upset about this - debugging complex multithreaded programs is great pain...
[15:06:01] <psha> but you may throw exec_finish?
[15:06:10] <psha> so you'll be rescheduled from above?
[15:06:25] <mhaberler> throw?
[15:06:29] <psha> return
[15:06:42] <psha> hm, or not...
[15:07:10] <mhaberler> ok, but then I would need to call into interp again when the wait is done
[15:07:12] <mhaberler> right?
[15:07:21] <mhaberler> I'll try replicating the logic from gcodemodule.cc parse_file to start with, see if that changes anything
[15:07:43] <psha> yes
[15:08:28] <psha> have you already implemented reentrant interp?
[15:08:40] <mhaberler> duh..
[15:08:56] <psha> let me refresh my memories ;)
[15:09:05] <psha> we are talking about using O-toolchangers?
[15:09:11] <mhaberler> yessir
[15:09:31] <psha> and about running something long and scary when M-code is executed?
[15:09:57] <mhaberler> yes
[15:10:07] <psha> you was cleaning Interp to allow serveral instances
[15:10:15] <psha> or it's not related to O-toolchanger?
[15:10:24] <mhaberler> and then we have all sorts of phun in the handler, pin I/O, probing, theworks
[15:10:33] <mhaberler> all queuebusters
[15:10:34] <psha> hm, wait
[15:10:56] <psha> as i recall there were problems with several "O-calls" on one line
[15:11:30] <psha> so when you encounter M-code you emulate O-call and place another frame on call stack
[15:11:33] <psha> correct?
[15:11:37] <psha> or not?
[15:11:41] <mhaberler> yes
[15:12:00] <mhaberler> actually a new _setup struct
[15:12:09] <mhaberler> and yes. a call frame
[15:12:11] <psha> so everything you need is correct 'return'
[15:12:17] <mhaberler> yes
[15:12:23] <mhaberler> if queue busted, that is
[15:12:33] <psha> interp is in nice shape - driven by rescheds from emctask
[15:12:39] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[15:12:46] <psha> until it need to remove that last frame
[15:12:48] <psha> right?
[15:13:07] <mhaberler> it all executes fine - my code just doesnt handle the interfacing to the outside properly - if I have an M66 in there, this one is excuted AFTER all said and done because it got queued
[15:28:16] <mhaberler> It could very well be I need to alternate read() and execute() because read() does exctly what I need it to do - read the external pins etc
[15:29:53] <mhaberler> maybe just the calling sequence is off - open the file, setup parameters and call read()
[15:43:41] <JT-Shop> http://www.linuxcnc.org/component/option,com_kunena/Itemid,20/func,view/catid,21/id,8700/lang,english/#8759
[15:44:26] <mhaberler> looks useful!
[15:45:16] <JT-Shop> yes it does
[15:46:07] <JT-Shop> maybe if no file is in the ini then open last file???
[15:51:36] <JT-Shop> hmm I think it does that...
[15:51:36] op is now known as micges_cnc
[15:52:02] -!- psha has quit [Read error: No route to host]
[15:52:24] -!- psha [[email protected]] has joined #emc-devel
[15:58:15] -!- micges_cnc has quit [Quit: Leaving]
[16:01:20] -!- ve7it [[email protected]] has joined #emc-devel
[16:05:02] -!- tom3p [[email protected]] has joined #emc-devel
[16:10:29] -!- stillme has quit [Ping timeout: 252 seconds]
[16:14:54] -!- psha has quit [Quit: leaving]
[16:33:59] <tom3p> cradek: hello, can i get an example of how to use HUD in vismach?
[16:34:37] -!- maximilian_h [[email protected]] has parted #emc-devel
[17:22:11] -!- stillme has quit [Ping timeout: 264 seconds]
[17:25:36] <tom3p> does loadusr motmod... num_dio=16 get 16 ins AND 16 outs or 16pins with the name determining direction
[17:25:43] <tom3p> AXIS configuration shows 16 of each but im not sure they co-exist
[17:29:09] -!- tom3p has quit [Read error: Connection reset by peer]
[17:30:04] -!- tom3p [[email protected]] has joined #emc-devel
[18:12:31] <mhaberler> separate ins in outs
[18:14:27] -!- stillme1 has quit [Quit: Leaving.]
[18:15:43] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100423140709]]
[18:16:39] <tom3p> mhaberler thx
[18:17:24] <mhaberler> re procedure-based toolchange - there's a serious bug I'm working on right now - basically M66 doesnt work
[18:17:43] <mhaberler> but it should in a few days the latest
[18:18:37] * mhaberler symptom: works ok if prcoedure called from MDI or auto; broken when called through M6_COMMAND/T_COMMAND
[18:19:07] <tom3p> oops i was just coding M66 in to check the shuttle rack was fully extended ( my rack is on an extending pneumatic slide, in emc vismach i use a step gen )
[18:19:12] <tom3p> will report any findings
[18:19:28] <mhaberler> for now, test as normal procedures
[18:19:39] <mhaberler> call from MDI or auto to debug stuff
[18:19:53] <mhaberler> when I'm done it will work within *_COMMAND as well
[18:20:14] <mhaberler> I know what the issue is, I'm meandering how to fix it ;-)
[18:33:28] -!- nullie has quit [Quit: Ex-Chat]
[18:41:04] -!- newbynobi has quit [Quit: Verlassend]
[18:45:53] -!- micges has quit [Ping timeout: 240 seconds]
[18:52:28] micges_ is now known as micges
[19:06:42] -!- Vladimirek [[email protected]] has joined #emc-devel
[19:22:44] -!- mhaberler_ has quit [Ping timeout: 248 seconds]
[19:27:39] -!- mhaberler has quit [Quit: mhaberler]
[19:32:05] -!- uwe_ has quit [Quit: leaving]
[19:37:20] <Vladimirek> Hi ! Can be EMC2 used on ARM procesors? In readme I read: "
[19:37:22] <Vladimirek> eliminating support for platforms
[19:37:23] <Vladimirek> other than Linux on x86, with either RTAI or RTLinux for
[19:37:25] <Vladimirek> realtime.
[19:37:26] <Vladimirek> "
[19:46:37] <cpresser> Vladimirek: there is a RTAI for various arm-platforms
[19:47:19] <cpresser> however, you might want to consider how to do IO operations
[19:52:49] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust1037.basl.cable.virginmedia.com] has joined #emc-devel
[19:58:45] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[20:07:18] <Vladimirek> cpr
[20:07:52] <Vladimirek> cpresser: is there any public ARM project ?
[20:08:14] <Vladimirek> I mean EMC on arm.
[20:09:25] -!- mhaberler [[email protected]] has joined #emc-devel
[20:09:32] <cpresser> sorry, i dont know about that. but since RTAI is avaiable, at least you should be able to compile emc2 on arm.
[20:10:52] <Vladimirek> ok, thanks.
[20:11:56] <cpresser> but be advised that there are many variations of arm-platforms out there
[20:19:13] <tom3p> mhaberler, very not done yet but beginning to be fun http://videobin.org/+471/4jo.html
[20:20:12] <mhaberler> very cool! how did you do this?
[20:20:14] <tom3p> uses #<named params> for positions, tooltable to describe shape, M64 65 66 to make sure element move fully before proceeding
[20:20:26] <tom3p> uh, with your code ;)
[20:20:29] <mhaberler> yay!
[20:20:46] <tom3p> NDY not done yet, will post
[20:21:35] <tom3p> i had limited success making the alpha channel of old tool disappear then new tool appear in chuk
[20:22:27] <tom3p> but will not bother with that (vismach) till the logical code done
[20:27:35] <andypugh> Vladimirek: I don't think EMC2 on ARM works, Jon Elson has been trying for a while with a Beagleboard.
[20:28:09] <andypugh> It might work on something like a RiscPC, if any are still working.
[20:30:07] <andypugh> (Forget that last bit, I am thinking of something else, and getting Alpha and ARM mixed up too). The EMC2 not yet working with Beagleboard part is true though.
[20:30:07] <Vladimirek> I woud like run it on this:http://armadeus.com/
[20:31:35] <Vladimirek> What is the main issue?
[20:31:36] <Jymmm> Vladimirek: How do you plan on interfacing motor drives with that?
[20:31:53] <andypugh> If you are doing it for fun, then carry on. If you want a small, low power, EMC2 system then Mini-ITX is pretty small.
[20:31:54] <Vladimirek> Via FPGA.
[20:32:12] <Jymmm> Vladimirek: I mena I/O, there is no USB support in EMC
[20:32:14] <Jymmm> mean
[20:34:18] <Vladimirek> No this wil not for fun. Hardware is selected (has another key role in project - CAM is only bonus).
[20:37:23] <andypugh> I think it is one of those "well, it ought to work" things.
[20:37:48] <andypugh> I imagine that there might have to be some work done on the makefiles.
[20:37:54] <Vladimirek> Can EMC run without X server ? Is there API to make custom vizualization (simple text mode)?
[20:38:18] <Jymmm> Keystick (text mode)
[20:38:59] -!- tom3p [[email protected]] has parted #emc-devel
[20:40:03] <Vladimirek> Jymmm: yes thanks this is fine.
[20:41:46] <Jymmm> Vladimirek: No, you are not understanding what I'm asking. How do you plan on connecting motor drivers to that board you linked to?
[20:43:54] <Vladimirek> Jymmm: Sorry, I have PCB convert FPGA IO to 422 levels connected to servo drivers.
[20:44:57] <Jymmm> RS-422 ???
[20:45:53] <Vladimirek> Jymmm: Yes, but it is not about serial, it is only about voltage levels - for Omron Smart Step.
[20:46:08] <Vladimirek> Jymmm: pulse/dir of course
[20:51:03] <Vladimirek> RTAI is used 'only' for moves right? It is posible out moves on some 'upper' level from EMC ? For example axis_X: direction, frequency for time T.
[20:53:36] <andypugh> You still need RTAI to provide the next update.
[20:53:55] <andypugh> However, you might want to consider moving the step generation to the FPGA.
[20:54:34] <andypugh> I think it ought to be possible to use the Mesa firmware in your own FPGA. (though it would be polite to ask first)
[20:54:54] -!- mhaberler has quit [Quit: mhaberler]
[21:02:40] -!- e3m has quit [Quit: Visitor from www.linuxcnc.org]
[21:04:44] -!- FinboySlick has quit [Quit: Leaving.]
[21:05:59] <Vladimirek> andypugh: Yes, exactly this i have in mind. I woul like to move step generators to FPGA. Is there some documented HAL/API (on EMC) for this?
[21:07:41] <andypugh> It is entirely possible that the Mesa firmwares (bitfiles) would install and run on any Spartan FPGA of the same model. I don't know.
[21:08:14] <andypugh> http://www.linuxcnc.org/docview/html/man/man9/hostmot2.9.html
[21:08:56] <andypugh> You might need a PCI bridge (or virtual parallel port) to use the standard EMC2 drivers.
[21:09:19] <Vladimirek> andypugh: Not this way. do some in FPGA is no problem for me, but what I need is some point in EMC where I can send data to FPGA.
[21:09:20] -!- mk0 has quit [Quit: Ухожу я от вас (xchat 2.4.5 или старше)]
[21:10:00] <andypugh> That's what the EMC2 Hostmot2 driver does.
[21:10:17] <Vladimirek> So as I understand some Hal module must be writen for my own FPGA.
[21:10:30] <Vladimirek> andypugh: ah so..
[21:10:44] <andypugh> If you want to do it differently, then you would write your own HAL driver to accept position updates, and convert to step rates.
[21:11:19] <andypugh> Look at the Motenc, Hostmot and Pluto drivers for examples.
[21:12:57] <Vladimirek> andypugh: Thanks a lot, this looks like good starting point.
[21:16:24] <andypugh> Somewhere in the sources you might even find the vhdl (?) for the FPGA
[21:16:55] <sarariman_seb> here's the vhdl for hostmot2: http://git.linuxcnc.org/gitweb?p=hostmot2-firmware.git;a=summary
[21:31:52] <Vladimirek> sarariman_seb: Thanks for your direction.
[21:40:34] -!- micges has quit [Quit: Ex-Chat]
[21:50:10] -!- Vladimirek has quit [Read error: Connection reset by peer]
[21:50:37] -!- Vladimirek [[email protected]] has joined #emc-devel
[21:52:55] -!- Fox_Muldr has quit [Ping timeout: 276 seconds]
[21:54:43] -!- skunkworks has quit [Ping timeout: 252 seconds]
[22:08:50] -!- Vladimirek has quit [Remote host closed the connection]
[22:24:29] -!- KimK [[email protected]] has joined #emc-devel
[22:37:11] -!- krushia has quit [Read error: Operation timed out]
[22:54:13] -!- acemi has quit [Quit: WeeChat 0.3.2]
[23:15:01] -!- jsr__ has quit [Quit: Leaving]
[23:56:49] -!- andypugh has quit [Quit: andypugh]