#emc-devel | Logs for 2008-05-19

[14:40:02] <micges> I've noticed that cvs server is not responding
[14:42:24] <SWPadnos> it seems to be working right now
[14:43:05] <SWPadnos> oh man - I didn't notice tha tseb had committed all the PDF files for every Mesa card
[14:44:24] <jepler> micges: the cvs hosting site had some connectivity problems earlier. it looks like they may be resolved, or at least lessened.
[14:44:29] <jepler> micges: are you still having trouble now?
[14:44:37] <jepler> SWPadnos: ouch -- those should probably not be there
[14:45:41] <SWPadnos> each is <200k, so it's not too bad
[14:46:03] <SWPadnos> but we can't change them, and they're likely copyrighted by Mesa
[14:46:35] <micges> jepler: its ok now, thanks
[14:46:54] <jepler> micges: good. thanks for mentioning the problem though
[14:47:01] <micges> np
[14:47:09] <rayh> Darn same topic on both lists.
[14:48:15] <rayh> SWPadnos, The earlier m5i20 stuff was also mesa copyright but I think PeterW allowed us to gpl it.
[14:48:55] <SWPadnos> I'm just talking about the manuals - there are 6 PDFs in the hal/drivers/mesa-hostmots/docs dir
[14:49:02] <SWPadnos> err - hostmot2
[14:49:11] <rayh> Oh okay.
[14:49:40] <rayh> That's right I did see those go by and wondered why they were at that location.
[14:50:59] <SWPadnos> normally we'd have the lyx (or whatever) source, but we can't modify these
[14:52:02] <rayh> Yes we should move them into our usual document process.
[14:52:21] <SWPadnos> well, we can't - we don't hav ethe source :)
[14:52:26] <SWPadnos> and we don't want it I think
[14:52:40] <SWPadnos> unless we want to include the manuals for all supported hardware
[14:55:28] <rayh> looking
[14:57:42] <jepler> I would sure not say "no" if somebody wanted to ask permission permission, and then bundle as much third-party documentation as possible. but I don't think there's much value in putting it alongside the source code. Probably it should be a separate location and when/if it is packaged it should be a separate package.
[14:58:02] <SWPadnos> like emc2-vendor-docs :)
[14:58:40] <cradek> a carefully maintained web page of links would be better
[15:02:10] <rayh> Yep links or a separate package. Links has the advantage of the producer of the stuff keeping it up to date.
[15:02:22] <cradek> that avoids all problems of permission and I think is the simplest way for people to get and read the docs
[15:03:29] <rayh> And since these are unmodified mesa docs I see no need to keep them with our code even though it uses the devices described by them.
[15:04:02] <rayh> If they were unique to us, like the drivers are, then I'd say let's put them in our documents.
[15:04:55] <rayh> I don't see any copyright notice so we could certainly ask to borrow stuff it it helps our documentation of our drivers.
[15:11:25] <skunkworks_> http://www.altera.com/products/devkits/altera/kit-cyc2-2C20N.html
[15:14:13] <skunkworks_> a guy here is getting one as an early christmas present.
[15:15:50] <skunkworks_> (from his brother that does some 'high level shit' (actual quote from him) at mayo clinic)
[16:05:03] <jepler> skunkworks_: fun. looks like that chip can handle designs that are about 10x more complex than the fpga chip on the m5i20
[16:07:01] <jepler> skunkworks_: I dunno if you saw it in the logs, but this weekend I hooked pluto-step up to zenbot; it worked flawlessly for me while I ran gcodes in the air for about 20 minutes.
[16:08:58] <cradek> cool, I didn't see that either
[16:09:09] <alex_joni> I saw it, but it's still cool :)
[16:26:34] <cradek> jepler: in emc2.2.5 sim/axis, no limit switches are hooked up, but I still get an active override limits checkbox in AXIS
[16:27:42] <jepler> cradek: I think that behavior improvement may only be on trunk
[16:27:57] <cradek> oh ok
[16:28:23] <cradek> I'm anxious for it; that confuses a lot of folks
[16:32:31] <jepler> enough that it's a bug fix?
[16:35:01] <cradek> jepler: I'd ask the release manager for his opinion on that
[16:37:04] <cradek> I wonder what the chances of backporting it correctly would be. It seems like it touched a lot of things.
[16:40:23] <jepler> I only found two commits for it
[16:40:23] <jepler> http://cvs.linuxcnc.org/cvs/emc2/share/axis/tcl/axis.tcl.diff?r1=1.58;r2=1.59 http://cvs.linuxcnc.org/cvs/emc2/src/emc/usr_intf/axis/scripts/axis.py.diff?r1=1.149;r2=1.150
[16:41:04] <jepler> cradek: the release manager says that if somebody wants to backport those changes to 2.2 and test it, and it's not more than those two files, then it's OK.
[16:41:27] <cradek> ok, thanks for asking him for me
[16:42:45] <micges> cradek: who is the "release manager" ?
[16:42:57] <cradek> micges: jepler
[16:43:08] <micges> ok I get it :P
[16:43:09] <cradek> for emc2.1 I am; for emc2.2 jepler is
[16:43:35] <cradek> nobody has been foolhardy enough to volunteer for 2.3 yet
[16:43:48] <jepler> oh it'll be alex
[16:43:52] <cradek> yay
[16:43:53] <jepler> we make him do all the jobs that are like work
[16:43:59] <cradek> sounds like alex just volunteered
[16:44:25] <micges> did Alex know ?:P
[16:44:34] <cradek> oh never mind that
[16:45:13] <cradek> that is a minor detail
[16:46:24] <micges> cradek: what bugfixes/features are candidates for backports ? are there some rules ?
[16:46:45] <cradek> bugfixes are good candidates if they are unlikely to break things
[16:47:08] <cradek> sometimes new drivers are backported because they cannot generally break anything else
[16:47:19] <cradek> new features are almost never backported
[16:47:49] <SWPadnos> features which require configuration changes are never* backported
[16:47:53] <SWPadnos> * for some values of never
[16:48:26] <cradek> yes we never break existing configurations in minor (bugfix) releases
[16:48:47] <micges> I see
[17:25:09] <alex_joni> hmm.. you might want to reconsider that
[17:25:22] <alex_joni> otherwise it'll be a long time till 2.3.0
[17:29:24] <alex_joni> SWPadnos: did you see the optimus tactus ?
[17:29:48] <SWPadnos> no - multi-touch?
[17:30:53] <SWPadnos> heh - cool
[17:30:59] <SWPadnos> I Bet it's hard to type on though :)
[17:31:35] <SWPadnos> bbl
[17:32:18] <alex_joni> yeah, I thought so too.. but it looks nice :)
[17:32:42] <alex_joni> I bet it'll be cheaper than maximus :D