#emc-devel | Logs for 2010-03-02

[09:40:21] <micges_work1> micges_work1 is now known as micges
[11:30:42] <micges_work> seb_kuzminsky: around?
[12:36:26] <micges_work1> micges_work1 is now known as micges
[14:46:56] <jepler> hm, O<xxx> call doesn't work when xxx is in xxx.ngc and you use ~ in PROGRAM_PREFIX
[14:47:02] <jepler> open("/usr/local/jepler/src/emc2/configs/sim/~/xxx.ngc", O_RDONLY) = -1 ENOENT (No such file or directory)
[14:47:05] <jepler> open("/xxx.ngc", O_RDONLY) = -1 ENOENT (No such file or directory)
[14:47:12] <jepler> here's what axis tries to open when generating the preview
[14:49:36] <SWPadnos> I guess you need to expand the PREFIX before using it
[14:49:52] <jepler> something like that
[14:50:36] <seb_kuzminsky> micges_work: i'm running around trying to get a bunch of kids off to school, i'll be around more in about an hour, hour and a half
[14:51:04] <seb_kuzminsky> are you thinking about a mandriva buildslave? if so start here: http://emc2-buildbot.colorado.edu/buildslave-admin-guide.html
[14:51:52] <micges_work> seb_kuzminsky: important: after 1h testing hm2 encoder velocity works ok
[14:52:10] <micges_work> seems that your and cradek last patch fixed it properly
[14:52:22] <micges_work> more test will be done tomorrow
[14:52:42] <seb_kuzminsky> well i'm glad to hear its working, but it's really not a fix, it just tries to detect the problem and warn about it instead of setting vel to inf or nan
[14:52:50] <seb_kuzminsky> did you get any warnings in dmesg?
[14:53:09] <micges_work> no warnings
[14:53:35] <seb_kuzminsky> hmm, i think that means the problem just never happened
[14:54:01] <micges_work> I'm talking about patch from november
[14:54:36] <micges_work> latest master, I have no messages at all
[14:54:58] <micges_work> seb_kuzminsky: bbliab
[15:07:30] <jepler> I hate the sf bugtracker
[15:07:45] <jepler> for one thing, how much would it have cost them to make the "comment" box taller?
[15:08:26] <jepler> for another, comments should either have rudimentary formatting or be in whitespace-preserving fixed-width font
[15:08:49] <jepler> oh well, the price is right at sf
[15:09:20] <SWPadnos> but at what cost?
[15:09:54] <jepler> the cost of my own annoyance is pretty low for anyone not in the same room with me
[15:10:49] <SWPadnos> heh
[15:11:19] <jepler> and lately I'm working on Windows during the day so I'm near my annoyance ceiling without sf
[15:12:00] <SWPadnos> oh man. that's too bad
[15:27:36] <micges_work1> why last G0 line desn't show on preview? http://www.pastebin.ca/1819108 (master)
[15:48:16] <jepler> micges_work1: Line N3650 shows here. I'm on a slightly old master, adfaf4e screenshot: http://emergent.unpy.net/files/sandbox/lastrapid.png
[15:50:51] <micges_work1> ok I got it.. silly .axisrc
[15:51:02] <micges_work1> sorry for bothering ;P
[16:05:46] <jepler> glad you figured it out
[19:12:15] <alex_joni> jepler: chrome has one really neat figure which makes me wonder where it's been all this time
[19:12:50] <alex_joni> when you are writing in a scrollbox (I believe it's called) you can drag the bottom-right corner and just scale it as you please
[19:12:59] <alex_joni> make it huge, edit what you like, then drag it back
[19:26:55] <jepler> sounds handy
[19:27:01] <jepler> so would having a vi-compatible editor embedded
[19:27:50] <SWPadnos> edit in vi, copy/paste
[19:45:39] <alex_joni> SWPadnos: too much trouble
[20:52:50] <skunkworks_> http://www.linuxcnc.org/content/view/5/5/lang,english/
[20:53:08] <skunkworks_> not that I should say anything about spelling.. ;)
[20:53:09] <skunkworks_> 2022HTML Documentation for the soon to be releaded 2.4 version
[20:55:48] <jepler> yeah, the reason it's late is that one of the leads got broken off
[20:55:56] <jepler> we had to send it back for rework; they'll solder new leads on
[20:56:01] <jepler> once it's releaded, it'll be ready for use
[20:57:53] <skunkworks_> heh
[21:10:45] <alex_joni> oh, and we need to put the magic smoke back inside
[21:10:49] <alex_joni> after the releading
[21:14:20] <skunkworks_> Thanks alex
[21:26:04] <alex_joni> skunkworks_: what for?
[21:26:36] <alex_joni> hmm.. seems the latest chrome beta does automatic website translation
[21:28:59] <skunkworks_> I just figured you would be the one to change it / and that you probably already did.
[21:29:48] <alex_joni> nope, wasn't me
[21:29:56] <alex_joni> if it's changed ;)
[21:30:06] <alex_joni> "Intel demoed its experimental 48 core chip at Cebit this morning..."
[21:32:09] <alex_joni> skunkworks_: fixed now
[21:34:19] <skunkworks_> See above thanks
[21:34:22] <skunkworks_> ;)
[22:13:40] <micges> sim on 2.4 branch: make clean cause that I must run ./configure each time after that
[22:15:49] <micges> there wasn't such behaviour earlier
[22:19:42] <jepler> micges: have a look at 1a48491
[22:21:07] <micges> ah ok
[22:21:20] <micges> didn't noticed that
[22:22:00] <jepler> the rationale seems like a good one to me, but if it's inconvenient then we can look at other solutions like a 'make distclean' target or putting the removal rule in debian/rules instead of src/Makefile
[22:22:29] <jepler> bbl
[22:30:36] <CIA-2> EMC: 03seb 07v2.4_branch * rf76f7ae53256 10/src/ (6 files in 2 dirs): Support kernels 2.6.29 and newer (ie Lucid Lynx)
[22:43:57] <micges> for 2.4: patch adding synced analog output (m67) to motion queue: http://www.pastebin.ca/1820103
[22:49:30] <cradek> micges: 2.4 feature freeze was Feb 1. Is there a very strong reason to put this in 2.4?
[22:50:57] <micges> this is fix I was trying to add while messing with m67 but I was unable to do this then
[22:53:26] <micges> patch touches only m67,nothing else
[23:11:24] <jepler> #
[23:11:25] <jepler> + if (syncaio.anychanged != 0) {
[23:11:30] <jepler> #
[23:11:30] <jepler> + } else {
[23:11:31] <jepler> #
[23:11:32] <jepler> + tc.syncaio.anychanged = 0;
[23:11:43] <jepler> oh, hm, maybe this makes sense
[23:12:47] <jepler> apparently it's an error in this implementation to set more than one aio before a given move. however, the error is not diagnosed.
[23:14:06] <jepler> the arguments to tpSetAout frankly don't make much sense (the names of the arguments don't tell me anything, the expressions they come from in the call don't mean anything to me, and one argument isn't even used.. why?
[23:15:00] <jepler> here, you break error handling -- granted tpClearxIOs don't error now, but that's no excuse:
[23:15:02] <jepler> - return tpClearDIOs();
[23:15:04] <jepler> + tpClearDIOs();
[23:15:07] <jepler> + tpClearAIOs();
[23:15:10] <jepler> + return 0;
[23:16:44] <micges> implementation was done as a copy of behaviour of synvced dio and its working
[23:17:02] <micges> multily m67 before move also works
[23:17:41] <micges> error handling break: yeas I was follow 'return 0'
[23:18:30] <micges> jepler: same questions can be asked about synced dio implementation
[23:18:40] <jepler> the difference being that synced dio already got in and this isn't in yet
[23:18:49] <jepler> but ok, let's fix these things in synced dio in master as well
[23:18:50] <jepler> bbl