Back
[00:00:35] -!- KimK_3 has quit [Ping timeout: 246 seconds]
[00:01:02] -!- KimK_3 [KimK_3!~kkirwan@ip68-102-65-151.ks.ok.cox.net] has joined #linuxcnc-devel
[00:03:11] -!- Nick001-Shop has quit [Remote host closed the connection]
[00:19:46] -!- jfire has quit [Quit: Leaving.]
[00:22:36] -!- asdfasd has quit [Ping timeout: 256 seconds]
[00:23:58] -!- rob_h has quit [Ping timeout: 252 seconds]
[00:32:20] -!- mourner has quit [Quit: mourner]
[00:37:10] -!- BJfreeman has quit [Ping timeout: 252 seconds]
[00:38:28] -!- jpk has quit [Ping timeout: 256 seconds]
[00:43:55] -!- rosenjon has quit [Quit: rosenjon]
[00:44:49] -!- erictheise has quit [Quit: erictheise]
[01:01:38] -!- gabewillen has quit [Ping timeout: 252 seconds]
[01:11:05] <jepler> this looks neat: change a debian live iso (cd) image so that it persists data on the same memory stick where it resides:
http://syn.theti.ca/2013/06/22/tutorial-wheezy-live-iso-hybrid-with-persistence-on-usb/
[01:16:46] <skunkworks> if we get linuxcnc into debian - do we lose the livecd style testing?
[01:26:52] -!- skorasaurus has quit [Quit: Elvis has left the building.]
[01:29:10] <jepler> we would probably still have to build our own bootable image, like we do with ubuntu
[01:30:35] <jepler> debian iso-hybrid is a lot like ubuntu live cd, at least on paper: you can boot it to desktop and you can install onto the hard drive from it
[01:30:43] <jepler> I have never used a debian iso-hybrid image myself
[01:42:19] <pcw_home> skunkworks: looks like my rtk-8139 based rtnet setup has the same problem your Intel one does:
[01:42:21] <pcw_home> the start net and stop net script work (and rtping works for me) but linuxcnc doesn't see the driver as loaded
[01:42:22] <pcw_home> maybe a race condition of some kind
[01:46:19] <skunkworks> pcw_home: 12.04?
[01:46:24] <skunkworks> 64bit
[01:46:26] <skunkworks> ?
[01:48:09] <pcw_home> 12.04 32 bits
[01:48:28] <pcw_home> dont think its a 32/64 bit issue
[01:50:19] <skunkworks> ah - well - I have a 10.04 32 bit installed with xenomi. tomorrow I will git linuxcnc and build it with the patch. He seemed to think that it was a 64bit issue but maybe it is a 12.04 issue
[01:51:32] <skunkworks> just to make sure it works here with micges same testbed
[01:51:42] <skunkworks> (10.04 32bit)
[01:52:23] <skunkworks> pcw_home: so it fails loading the hm2_eth hal componant?
[01:53:27] <pcw_home> Yes. You might verify that rtping works just to test the driver
[01:53:29] <pcw_home> you also need to rmmod the non rt driver if it gets loaded
[01:53:56] <skunkworks> pcw_home: yes - have been rmmoding the e100 driver on my end
[01:54:01] <pcw_home> (my system crashed without doing that)
[01:54:24] <skunkworks> mine just didn't insmod the rt_eepro100.ko
[01:54:30] <skunkworks> iirc
[01:54:57] <pcw_home> well didn't really crash but maybe wedged one of the PCI interrupts
[01:55:02] <skunkworks> heh
[01:55:05] -!- tmcw has quit [Remote host closed the connection]
[01:55:27] <skunkworks> I feel like we are | | close
[01:56:09] -!- jrabbit has quit [Changing host]
[01:57:01] <seb_kuzminsky> zultron's xenomai kernel builder is chugging away...
[01:57:25] <skunkworks> I hope to edit the wiki - there are a lot of extra things that need to be done. Like the adding of the kernels headers and for some reason you need to symlink the headers to a build directory under the xenomi directory for rtnet to build
[01:57:36] <skunkworks> seb_kuzminsky: cool!
[01:57:47] <skunkworks> he is one smart cookie
[01:58:34] <skunkworks> (and installing ncurses
[01:58:36] <skunkworks> _
[01:58:37] <skunkworks> )
[01:58:43] -!- sumpfralle has quit [Ping timeout: 264 seconds]
[02:01:39] <skunkworks> could rtnet be added to linuxcnc?
[02:01:46] * skunkworks doesn't know how that all works
[02:06:04] <jepler> skunkworks: someone should build the rtnet thing into a debian package that we would put on the linuxcnc.org debian package server
[02:06:53] <jepler> from what I know that makes more sense than putting it "in linuxcnc", which I take to mean "in the same git repository as the main linuxcnc software" or "in the same debian package as the main linuxcnc software"
[02:06:57] <skunkworks> ah - so it isn't a deb. duh
[02:07:31] <skunkworks> like classicladder.... ;)
[02:08:40] <skunkworks> (only different... rtnet as a deb would probably work just fine for what we needed - classic ladder needed to be modified to hook into linuxcnc)
[02:09:46] <jepler> right, classicladder hooks into linuxcnc; linuxcnc hooks into (wants to hook into) rtnet
[02:09:52] <jepler> so the relationship is not the same
[02:17:31] <skunkworks> hmm git clone git://rtnet.git.sourceforge.net/gitroot/rtnet/rtnet rtnet-0.9.13
[02:17:56] <skunkworks> that gets the master version of rtnet doesn't it? I wonder what version it actually is
[02:20:44] <skunkworks> eh - it looks like it is 0.9.13 the best I can tell
[02:33:09] -!- chesepuff has quit [Ping timeout: 248 seconds]
[02:46:53] -!- Laremere has quit [Ping timeout: 240 seconds]
[02:51:27] -!- FinboySlick has quit [Quit: Leaving.]
[03:01:42] -!- PetefromTn has quit [Remote host closed the connection]
[03:04:59] -!- Loetmichel has quit [Ping timeout: 260 seconds]
[03:15:49] -!- jfire has quit [Quit: Leaving.]
[03:34:54] -!- gonzo_ has quit [Read error: Connection reset by peer]
[03:45:03] -!- tjb1 has quit [Quit: tjb1]
[03:51:38] -!- tjb1 has quit [Quit: tjb1]
[04:10:59] <zultron> Hey seb_kuzminsky, if your single CPU core is fast as Michael's machine, you might be almost done with the first set of packages :P, maybe Precise i386? Any problems yet?
[04:13:21] <zultron> skunkworks & jepler, RTnet might best be added to the Debian pkg build scripts I'm trying to dump in Seb's lap. ;) Those scripts' function is exactly to build kernels and matching (non-LinuxCNC) kernel modules.
[04:14:46] <zultron> seb_kuzminsky, I'm not worried too much about packaging for Fedora. If there's a lot of interest, I'll have folks volunteering to co-maintain in no time.
[04:15:04] <zultron> (and if no interest, support is a moot question ;)
[04:17:55] -!- ravenlock has quit [Ping timeout: 264 seconds]
[04:35:25] -!- rosenjon has quit [Quit: rosenjon]
[04:48:34] -!- rosenjon has quit [Quit: rosenjon]
[04:54:13] -!- gabewillen has quit [Quit: gabewillen]
[04:58:28] -!- workmank has quit [Ping timeout: 246 seconds]
[05:03:03] -!- Fox_Muldr has quit [Ping timeout: 260 seconds]
[05:34:45] -!- rosenjon has quit [Quit: rosenjon]
[05:47:06] -!- Tecan has quit [Quit: Live Long And Phosphor!]
[05:52:46] <seb_kuzminsky> zultron: worked like a charm
[05:52:55] <seb_kuzminsky> precise i386
[06:15:37] -!- dhoovie has quit [Read error: Connection reset by peer]
[06:19:45] -!- gambakufu has quit [Ping timeout: 248 seconds]
[06:39:25] -!- mourner has quit [Quit: mourner]
[06:51:31] -!- KimK_2 [KimK_2!~kkirwan@ip68-102-65-151.ks.ok.cox.net] has joined #linuxcnc-devel
[06:55:13] -!- rob_h [rob_h!~rob_h@2.124.43.146] has joined #linuxcnc-devel
[06:59:02] -!- mhaberler [mhaberler!~mhaberler@69.199.158.1] has joined #linuxcnc-devel
[07:33:16] -!- KimK_2 has quit [Remote host closed the connection]
[07:57:14] -!- maximilian_h [maximilian_h!~bonsai@pD9585DE4.dip0.t-ipconnect.de] has joined #linuxcnc-devel
[08:02:39] -!- psha[work] [psha[work]!~psha@195.135.238.205] has joined #linuxcnc-devel
[08:06:47] _BJFreeman is now known as BJfreeman
[08:12:04] -!- odogono has quit [Read error: Connection reset by peer]
[08:24:09] BJfreeman is now known as Guest23096
[08:24:16] _BJFreeman is now known as BJfreeman
[08:24:56] -!- mhaberler has quit [Quit: mhaberler]
[08:24:59] -!- Guest23096 has quit [Ping timeout: 252 seconds]
[08:44:02] -!- JT-Shop has quit [Read error: Connection reset by peer]
[08:44:02] -!- jthornton has quit [Read error: Connection reset by peer]
[08:44:25] -!- JT-Shop [JT-Shop!~John@99.198.67.11] has joined #linuxcnc-devel
[08:44:27] -!- jthornton [jthornton!~john@99.198.67.11] has joined #linuxcnc-devel
[08:48:52] -!- b_b has quit [Changing host]
[08:50:13] -!- stsydow has quit [Remote host closed the connection]
[08:54:42] -!- jthornton has quit [Ping timeout: 246 seconds]
[08:54:42] -!- JT-Shop has quit [Ping timeout: 246 seconds]
[08:55:38] -!- jthornton [jthornton!~john@99.198.67.11] has joined #linuxcnc-devel
[08:55:39] -!- JT-Shop [JT-Shop!~John@99.198.67.11] has joined #linuxcnc-devel
[08:59:05] -!- BJfreeman has quit [Ping timeout: 252 seconds]
[09:03:21] _BJFreeman is now known as BJfreeman
[09:13:37] -!- jthornton has quit [Ping timeout: 248 seconds]
[09:13:53] -!- JT-Shop has quit [Ping timeout: 252 seconds]
[09:29:46] -!- JT-Shop [JT-Shop!~John@99.198.67.11] has joined #linuxcnc-devel
[09:29:48] -!- jthornton [jthornton!~john@99.198.67.11] has joined #linuxcnc-devel
[09:38:05] -!- BJfreeman has quit [Quit: had a good time]
[09:43:14] -!- cpresser has quit [Quit: Lost terminal]
[09:45:36] -!- maximilian_h has quit [Quit: Leaving.]
[10:05:57] -!- sumpfralle has quit [Ping timeout: 240 seconds]
[10:06:26] -!- tjb1 has quit [Client Quit]
[10:14:31] -!- carper64_lb [carper64_lb!~quassel@2.217.116.99] has joined #linuxcnc-devel
[10:18:44] -!- stsydow has quit [Read error: Operation timed out]
[10:27:11] -!- JT-Shop has quit [Read error: Connection reset by peer]
[10:27:19] -!- jthornton has quit [Read error: Connection reset by peer]
[10:27:25] -!- jthornton_ [jthornton_!~john@99.198.67.11] has joined #linuxcnc-devel
[10:37:52] jthornton_ is now known as jthornton
[10:41:59] -!- mung [mung!~mung@81.168.69.141] has joined #linuxcnc-devel
[10:43:22] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[10:44:31] -!- skunkworks has quit [Ping timeout: 256 seconds]
[10:49:23] -!- stsydow has quit [Remote host closed the connection]
[10:50:54] -!- vladimirek has quit [Remote host closed the connection]
[10:55:38] -!- sumpfralle has quit [Ping timeout: 252 seconds]
[11:09:30] -!- adb has quit [Read error: Operation timed out]
[11:10:20] -!- mung has quit [Remote host closed the connection]
[11:11:36] -!- stsydow has quit [Remote host closed the connection]
[11:20:43] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[11:20:57] -!- gabewillen has quit [Quit: gabewillen]
[11:33:48] -!- jfire has quit [Client Quit]
[11:36:50] <skunkworks> pcw_home, zultron, micges,
http://pastebin.ca/2410789
[11:37:16] <skunkworks> So - it works on 10.04 32bit (started with the linuxcnc livecd)
[11:58:07] -!- fomox has quit [Ping timeout: 264 seconds]
[12:07:56] saarlaender is now known as cpresser
[12:08:32] <jepler> probably the userspace version is missing a link-time library that provides rt_net_* in userspace
[12:08:43] <jepler> i.e., -lsomething needs to go in some Makefile or Submakefile
[12:08:51] <jepler> but that's just a guess from the peanut gallery
[12:11:17] -!- dhoovie has quit [Read error: Connection reset by peer]
[12:12:20] -!- mourner has quit [Quit: mourner]
[12:15:41] -!- micges [micges!~micges@esx224.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[12:15:52] <micges> skunkworks: hi
[12:16:16] <micges> skunkworks: your problem is not related to x64 platform
[12:16:52] <micges> skunkworks: Peter has same problems with 32 bit
[12:20:54] <micges> ok I've just readback
[12:21:02] <micges> cool!!
[12:31:07] -!- ravenlock has quit [Ping timeout: 264 seconds]
[12:34:41] -!- stsydow has quit [Read error: Connection reset by peer]
[12:43:45] -!- stsydow has quit [Remote host closed the connection]
[12:45:56] -!- b_b has quit [Changing host]
[12:46:19] <skunkworks> micges, does what jepler said make sense?
[12:47:19] -!- fomox has quit [Ping timeout: 264 seconds]
[12:48:31] <micges> yes, but I don't know how to check it
[12:50:43] <micges> important is that is system problem not lcnc or driver problem
[12:50:57] <micges> we'll figure it out
[12:52:02] -!- Laremere has quit [Ping timeout: 240 seconds]
[12:54:05] <skunkworks> Neat!
[13:00:43] -!- gambakufu has quit []
[13:03:14] <jepler> if your component is building inside the linuxcnc source tree, and the symbol rt_net_* comes from -lfoo, then I think you want to write: ../rtlib/yourcomponent.so: LDFLAGS += -lfoo
[13:03:25] <jepler> but .. untested and I don't have an appropriate system to test it on
[13:05:31] <micges> jepler: thanks, I'll check it
[13:07:46] <jepler> remove yourcomponent.so before doing make again; make won't figure out that it needs to be remade
[13:08:23] -!- shurshur has quit [Read error: Operation timed out]
[13:08:45] <jepler> anyway, I can sort-of test. I added this to Makefile:
[13:08:45] <jepler> +../rtlib/genserkins$(MODULE_EXT): LDFLAGS += -lfoo
[13:08:51] <jepler> and then when building I get an error:
[13:08:58] <jepler> /usr/bin/ld.bfd.real: cannot find -lfoo
[13:09:01] <jepler> make: *** [../rtlib/genserkins.so] Error 1
[13:09:20] <jepler> so it looks like this will add the lib to the linker line
[13:16:18] <skunkworks> any ideas?
[13:16:19] <skunkworks> http://pastebin.ca/2410829
[13:16:28] -!- `Nerobro has quit [Ping timeout: 256 seconds]
[13:16:37] -!- rosenjon has quit [Quit: rosenjon]
[13:18:59] -!- rosenjon has quit [Client Quit]
[13:20:50] -!- fomox has quit [Ping timeout: 252 seconds]
[13:22:44] <micges> skunkworks: do you have encoder connected?
[13:22:50] <skunkworks> no
[13:26:48] <skunkworks> This is the hal file I am testing
[13:26:48] <skunkworks> http://pastebin.ca/2410841
[13:27:46] -!- rosenjon has quit [Client Quit]
[13:28:07] <micges> move pet_watchdog to first place
[13:30:36] <skunkworks> I think the watchdog hit was from me doing things manually. I have not had it bite since I have been loading the test hal file
[13:32:50] <skunkworks> the only errors I am getting are the encoder errors. Is that noise or an issue?
[13:36:02] <micges> no idea
[13:36:20] <micges> try to enable few encoders and see if yuo still have errors
[13:37:09] <skunkworks> I don't think encoders are enable/disableable...
[13:39:32] -!- fomox has quit [Ping timeout: 252 seconds]
[13:40:20] <micges> change num_encoders to some number
[13:46:07] <skunkworks> ok
[13:49:59] -!- mourner has quit [Client Quit]
[13:52:06] <skunkworks> 1 or 6 encoders does the same thing
[13:52:37] <skunkworks> is this because it is getting weird data back from the card? or not parcing the data correctly?
[13:52:47] * skunkworks is talking out of his a$$ again
[13:53:42] <micges> probably weird data
[13:54:53] _BJFreeman is now known as BJfreeman
[13:57:08] -!- fomox has quit [Ping timeout: 252 seconds]
[13:59:29] -!- jerryitt has quit [Quit: Leaving.]
[14:04:45] <pcw_home> changing num_encoder should change the tram read sizes though
[14:04:48] -!- mk0 has quit [Quit: Leaving]
[14:05:19] -!- mhaberler [mhaberler!~mhaberler@69.199.158.1] has joined #linuxcnc-devel
[14:05:35] <pcw_home> (looks like its reading all of them even if told there are none to read)
[14:07:17] <seb_kuzminsky> zultron: the lucid i386 xenomai build worked too (though i didn't stay up to watch it)
[14:07:30] <seb_kuzminsky> that's a really cool thing you made
[14:07:48] <cradek> ooh I can just put a new kernel on my lucid i386 machine and try it?
[14:08:50] -!- mhaberler has quit [Client Quit]
[14:14:37] -!- JT-Shop [JT-Shop!~John@99.198.67.11] has joined #linuxcnc-devel
[14:21:14] <skunkworks> cradek, I did..
[14:21:20] -!- fomox has quit [Ping timeout: 252 seconds]
[14:28:04] <seb_kuzminsky> i'm rsyncing the debs to a public place, hold on
[14:28:17] * seb_kuzminsky needs better wifi at home
[14:28:31] <skunkworks> I can twittle a bit!
[14:29:23] <seb_kuzminsky> i dont know what that means, but i'm glad for you
[14:29:25] <cradek> seb_kuzminsky: day off?
[14:29:28] <cradek> haha
[14:29:59] <seb_kuzminsky> all my days off were in wichita ;-)
[14:30:20] -!- mhaberler [mhaberler!~mhaberler@69.199.158.1] has joined #linuxcnc-devel
[14:30:21] <seb_kuzminsky> speaking of wifi, i bought one of these and it works great with the stock precise kernel:
http://www.amazon.com/Edimax-EW-7811Un-Wireless-Adapter-Wizard/dp/B003MTTJOY
[14:30:29] <seb_kuzminsky> it's amazingly tiny
[14:30:46] <cradek> my vacation renews monday, yay
[14:31:03] <seb_kuzminsky> nice :-)
[14:31:32] <cradek> isn't your machine new enough to have internal wifi? even mine are...
[14:31:37] <skunkworks> pcw_home, It doesn't look like I can use the first plug as gpio? There is no .is_output or .out
[14:31:46] <skunkworks> even with everything set to 0
[14:31:57] <seb_kuzminsky> i'm going into vacation debt and taking the kids to europe later this summer for this:
https://ohm2013.org
[14:32:44] <cradek> ooh, take me too!
[14:32:49] <skunkworks> Neat
[14:32:58] <seb_kuzminsky> my laptop has fine wifi built in, this is for the beagle bone black i bought, and conveniently it also works on an older desktop i'm using to build zultron's kernels
[14:33:12] <skunkworks> seb_kuzminsky, I was setting output .024 true/false
[14:33:15] <pcw_home> Looks like the drivers enable 0 through n of encoders, pwmgens, etc is not working
[14:33:23] <skunkworks> pcw_home, ah
[14:34:15] <skunkworks> is num_encoders=0 spelled wrong? :)
[14:34:17] <pcw_home> a clue is that the TRAM read writes still indicated encoder reads and PWM writes even though you had them set to 0
[14:34:19] <seb_kuzminsky> http://highlab.com/~seb/linuxcnc/zultron-debs/
[14:34:34] <seb_kuzminsky> lucid debs are up, precise is still transferring
[14:34:42] <skunkworks> there are now xenomi debs?
[14:34:46] <seb_kuzminsky> i have not tried to install or boot these kernels! use at your own risk!
[14:34:55] <skunkworks> cool
[14:39:22] <seb_kuzminsky> i kept hearing people at the hackfest pronounce xenomai as "zen-omie', i always thought it was 'zen-oh-my', and googling it just now i found many other alternatives :-/
[14:40:14] -!- pierp has quit [Quit: He just can't jump...]
[14:40:37] <seb_kuzminsky> my fave so far is 'ecks-no-my'
[14:41:32] <cradek> I pronounce it xanadu
[14:42:21] <seb_kuzminsky> huh, xenomai started as an independent project, then merged with RTAI in 2003, then forked off from RTAIN in 2005
[14:42:35] <seb_kuzminsky> s/N//
[14:42:38] <cradek> (actually I always assumed zen-oh/uh-my)
[14:43:40] <seb_kuzminsky> my bus ride to work is almost over and now i have to go be a sarariman :-(
[14:45:08] <jepler> seb_kuzminsky: do good work
[14:46:04] -!- mhaberler has quit [Quit: mhaberler]
[14:51:02] -!- fomox has quit [Ping timeout: 252 seconds]
[14:54:38] -!- Heinz_60 has quit [Quit: ChatZilla 0.9.90 [Firefox 20.0/20130329043827]]
[14:56:14] -!- psha[work] has quit [Quit: Lost terminal]
[14:57:41] -!- Heinz_60 has quit [Client Quit]
[14:59:46] -!- odogono has quit [Read error: Connection reset by peer]
[15:14:41] -!- rosenjon has quit [Quit: rosenjon]
[15:17:05] -!- tmcw has quit [Remote host closed the connection]
[15:22:16] -!- krusty_ar has quit [Quit: No Ping reply in 180 seconds.]
[15:27:54] -!- Kup has quit [Read error: Connection reset by peer]
[15:31:52] -!- shurshur has quit [Remote host closed the connection]
[15:32:40] -!- mhaberler [mhaberler!~mhaberler@69.199.158.1] has joined #linuxcnc-devel
[15:37:29] <skunkworks> if the watchdog bites - shoudn't the outputs float?
[15:38:49] <pcw_home> Yes
[15:39:16] -!- ink has quit [Disconnected by services]
[15:42:12] -!- rosenjon has quit [Quit: rosenjon]
[15:42:14] <skunkworks> hmm - I have a led on gpio 24 and set to open collector. When I pull the network cable - the watchdog bits and the led stays on
[15:43:14] -!- skorasaurus has quit [Quit: Elvis has left the building.]
[15:45:08] <pcw_home> maybe the watchdog is not really enabled
[15:46:18] <pcw_home> the encoder error stuff kind of indicates a data alignment problem or something similar
[15:47:30] -!- `Nerobro has quit [Ping timeout: 252 seconds]
[15:51:52] <skunkworks> ah
[15:52:11] <pcw_home> I dont think there any way the watchdog can fail to operate if its enabled but there may still be bugs in the driver/packet interface. I can check the hardware later today but I expect this is a driver issue
[15:53:26] -!- Thetawaves has quit [Remote host closed the connection]
[15:54:59] <skunkworks> http://pastebin.ca/2410889
[15:55:58] -!- rosenjon has quit [Quit: rosenjon]
[15:57:15] <pcw_home> 1155090 hm2_7i80.0.read.tmax
[15:57:16] <pcw_home> is no good...
[15:57:21] <skunkworks> saw that...
[15:58:56] <micges> skunkworks: try to reset tmax and see if it will rise again
[15:59:07] <micges> simply write 0
[15:59:43] <skunkworks> I restarted halcmd
[16:00:02] <jepler> this is also troubling, if it represents a 'typical' value: 354986 hm2_7i80.0.read.time
[16:00:26] <skunkworks> it is at 851000 at the moment
[16:00:42] <micges> no good
[16:00:49] -!- Heinz_60 has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Wibbly Wobbly IRC]
[16:02:20] <skunkworks> jepler, after restarting hal - it is right back to 354000ish
[16:03:10] <skunkworks> I should run the latency test with the rtnet ko's loaded....
[16:03:35] <skunkworks> without those running - I was getting <10us
[16:04:26] <pcw_home> doubt if they will make any difference (unless you rtping the 7I80 while running the latency test)
[16:04:28] <skunkworks> with all the rtnet modules running latency <10us so far
[16:04:59] <jepler> it's the hm2_7i80.0.read which is taking a long time
[16:05:22] <jepler> latency is one thing, the time that it takes for the realtime code to execute is another
[16:06:46] <jepler> but there's "something" not working right and maybe that same something is what is making hm2_7i80.0.read take a long time to execute
[16:06:51] <jepler> it's too soon to know with any certainty
[16:06:55] <pcw_home> thats probably expected at this point, I'm not sure all the HM2 random access botches have been patched
[16:07:14] <micges> not all
[16:07:23] <micges> 40%-60% of them
[16:07:25] <pcw_home> so there may be more than 3 packets per thread invocation
[16:08:09] <micges> I havent such high numbers on rt8139
[16:09:08] <micges> I had ~400k
[16:12:01] <skunkworks> what kind of numbers would we be looking for?
[16:12:05] <pcw_home> Well if i can get mine going I can verify.
[16:12:07] <pcw_home> Problems I see:
[16:12:08] <pcw_home> 1. TRAM transfer sizes not using enabled hardware for transfer count (is this wrong for PCI.EPP also?)
[16:12:10] <pcw_home> 2. probably more than 3 packets so more hm2 undoing TRAM botches
[16:12:11] <pcw_home> 3. move to 2 packet mode to subtract wire /turnaround time from reads
[16:17:33] <skunkworks> pcw_home, heh - when you ping the card - the 4 led's count up binary
[16:18:23] <pcw_home> The default (EEPROM) LED pointer is to RXPacketCount
[16:18:38] <skunkworks> rt_ping is pretty cool - it gives the time as about 60us
[16:19:08] <skunkworks> ah - so ping is slow enought to see the packet count increas
[16:19:31] <jepler> by default ping sends one per second (I don't know about rt_ping)
[16:19:45] <skunkworks> I was wondering why sometimes some leds are on and some off
[16:21:10] <pcw_home> you can set the EEPROM so the LEDS are "owned" by hm2
[16:22:04] <pcw_home> also the INIT LED indicates parse errors
[16:22:05] <skunkworks> oh - so that is the hm2_7i80.0.led.CR01 -04
[16:22:18] -!- dway has quit [Quit: NOOOOOOooooooooo……]
[16:22:19] <skunkworks> why I could not seem to set them?
[16:23:03] <pcw_home> Yes the EEPROM default is *RXPacketCount
[16:23:09] -!- Nick001-Shop has quit [Client Quit]
[16:23:38] <pcw_home> not hm2
[16:23:43] <skunkworks> ah
[16:25:25] -!- chesepuff has quit [Quit: Lost terminal]
[16:26:20] <skunkworks> running a few minutes and the init led is still off
[16:26:38] <skunkworks> (running through hal)
[16:28:26] -!- `Nerobro_ has quit [Remote host closed the connection]
[16:29:06] -!- `Nerobro__ has quit [Read error: Connection reset by peer]
[16:29:31] <micges> there shouldn't be parse error
[16:31:32] -!- mhaberler has quit [Quit: mhaberler]
[16:33:43] <pcw_home> No, you should never see a parse error
[16:39:08] * KimK adds "ZEE-no-my" to the earlier list
[16:39:18] <skunkworks> with just the watchdog thread running thread tmax is 26599
[16:39:44] <skunkworks> *function
[16:43:14] -!- theorbtwo has quit [Remote host closed the connection]
[16:48:33] -!- b_b has quit [Changing host]
[16:54:56] -!- andypugh [andypugh!~andy2@cpc16-basl9-2-0-cust685.20-1.cable.virginmedia.com] has joined #linuxcnc-devel
[17:01:19] -!- rosenjon has quit [Quit: rosenjon]
[17:01:35] <seb_kuzminsky> did micges (or someone) add actual tram support to hostmot2? if not (if it's using the code in 2.5 or master) the driver collects tram allocation requests from the subdrivers and then blithely issues lots of individual reads/writes for the different regions
[17:03:03] <pcw_home> this is whats we are finding :-(
[17:03:19] -!- mourner has quit [Quit: mourner]
[17:04:01] <seb_kuzminsky> i think all the info of how the tram should be organized is collected by the hostmot2 driver, just not sent to the anyio boards
[17:04:34] <seb_kuzminsky> and the infrastructure for distributing the different parts of the tram region to the subdrivers is all there
[17:05:30] <seb_kuzminsky> so all that's needed is to have the hostmot2 driver communicate the tram setup to the anyio board, and read/write from/to tram in one chunk in the main hm2_read()/hm2_write() functions
[17:07:06] <andypugh> As far as I know the drivers all act like there is TRAM, and then there isn't.
[17:07:35] <seb_kuzminsky> all the subdrivers (encoder, pwmgen, etc) allocate tram regions from the hostmot2 mother-driver
[17:08:06] <seb_kuzminsky> this allocation takes a range of register addresses in the fpga and returns a pointer to the memory that will be transfered back and forth
[17:08:35] <seb_kuzminsky> the mother-driver collects all these tram requests from the sub-drivers, but then stops at the last second and doesnt actually set up tram in the fpga
[17:09:00] <andypugh> Yeah. That's not ideal, is it?
[17:09:07] <seb_kuzminsky> when hm2_read() runs, it walks this list of tram regions and reads each one in turn, and puts the data into the memory that it told the subdriver about
[17:09:11] <seb_kuzminsky> no, it's not ideal :-/
[17:09:41] <andypugh> Actually, micges _might_ have changed that, but as far as I have seen his work isn't merged yet?
[17:09:42] <seb_kuzminsky> but it works really well on pci, and pretty well on epp
[17:09:52] -!- md-2 has quit [Remote host closed the connection]
[17:09:52] <seb_kuzminsky> i think micges work is not merged yet
[17:10:22] <skunkworks> http://wiki.linuxcnc.org/uploads/hm2_7i80_support_on_rtos_merged_to_master.patch
[17:10:44] <seb_kuzminsky> i think it depends on rtnet, which i think depends on xenomai? i'm not sure.. but i bet it doesnt work with the rtai we ship for 2.5 on lucid & hardy
[17:11:34] <andypugh> That sounds like an adequate explanation.
[17:11:44] <andypugh> Anyway, I have places to go and people to see.
[17:12:01] -!- andypugh has quit [Quit: andypugh]
[17:13:52] <micges> in hm2_eth I'm collecting all write data into one eth packet
[17:14:09] <micges> but there is no TRAM use atm, just tram list like it was earler
[17:15:20] <micges> I must go
[17:15:22] <micges> see you
[17:15:23] <skunkworks> - Linux kernel 2.6.x
[17:15:24] <skunkworks> 22 - Xenomai 2.0 or better
[17:15:24] <skunkworks> 23 RTAI 3.3-cv or better
[17:15:35] -!- micges has quit [Quit: Leaving]
[17:31:53] -!- PetefromTn has quit [Remote host closed the connection]
[17:32:27] -!- IchGuckLive [IchGuckLive!~chatzilla@95-89-97-51-dynip.superkabel.de] has joined #linuxcnc-devel
[17:32:50] <IchGuckLive> does vismach only run unter git invirement
[17:48:05] -!- ReadError_ has quit [Excess Flood]
[17:49:24] <seb_kuzminsky> IchGuckLive: i've never tried it, but i think it should work when running installed-from-deb as well
[17:49:38] <seb_kuzminsky> if it doesn't, that's probably a bug
[17:51:00] <IchGuckLive> B)
[17:51:15] <IchGuckLive> i try it soon
[17:51:52] <mozmck> seb_kuzminsky: you should have a box today.
[17:51:59] <IchGuckLive> seb_kuzminsky: inspiration at 1:25 into video
http://www.youtube.com/watch?feature=player_embedded&v=R1DCXe9t3UE
[17:53:50] <skunkworks> does seb have a new arm? ;)
[17:54:00] <mozmck> fixin' to!
[17:59:01] -!- mackerski has quit [Quit: mackerski]
[18:01:10] -!- `Nerobro has quit [Remote host closed the connection]
[18:03:15] <IchGuckLive> seb_kuzminsky: vismach is on messurments Inch as internal as i belive
[18:04:49] -!- kanzure has quit [Ping timeout: 248 seconds]
[18:14:18] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 20.0/20130329043827]]
[18:21:56] -!- erasmo [erasmo!~erasmo@77-255-238-64.adsl.inetia.pl] has joined #linuxcnc-devel
[18:23:04] <seb_kuzminsky> mozmck: awesome!
[18:25:38] <seb_kuzminsky> mozmck: i just tried to send you a private message but you (or your irc client) said you're away
[18:25:42] <CaptHindsight> pcw_home:
http://cubieboard.org/ cubieboard2
http://cubieboard.org/2013/06/19/cubieboard2-is-here/
[18:28:49] -!- Heinz_60 has quit [Ping timeout: 248 seconds]
[18:30:04] <cradek> pcw_home: the 7i30 is rated 48V. does that mean I can use a 48v regulated switcher to run it, or should I use more like 36-40?
[18:30:34] -!- AndChat|681225 has quit [Client Quit]
[18:35:43] <cradek> I think my stepper power supply is 42, I oughta use that (these are really expensive on mouser)
[18:36:09] -!- BJfreeman has quit [Ping timeout: 252 seconds]
[18:36:21] _BJFreeman is now known as BJfreeman
[18:44:49] <jepler> cradek: I think the power supply at workshop was a 24V one tuned to the high end ~27V
[18:45:29] <cradek> I'd hope to get a lot more torque than that setup gave me - it's enough to move, but not much more
[18:45:58] <jepler> yeah if you are thinking of retrofitting max you need more torque than I need on zenbot for pcb milling too
[18:46:05] <cradek> the current limit was maybe on the low setting on the 7i30?
[18:46:12] <jepler> that's entirely possible
[18:46:40] <cradek> so I could have 42/27th the voltage and 3x the current
[18:46:49] <cradek> maybe
[18:55:05] -!- tmcw has quit [Remote host closed the connection]
[18:59:09] -!- adb [adb!~IonMoldom@178.211.237.94] has joined #linuxcnc-devel
[19:01:03] -!- psha [psha!~psha@213.208.162.67] has joined #linuxcnc-devel
[19:02:19] <cradek> EMC2 sensationally good!
[19:02:27] <cradek> http://www.youtube.com/watch?v=5MOSnFSx8JQ
[19:05:19] <jepler> oooh lazors
[19:17:13] -!- BJfreeman has quit [Ping timeout: 252 seconds]
[19:28:10] -!- micges [micges!~micges@ewy185.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[19:28:31] <seb_kuzminsky> cradek: that's very cool
[19:29:02] _BJFreeman is now known as BJfreeman
[19:32:00] -!- micges has quit [Client Quit]
[19:32:23] <cradek> I oughta use my surplus rotary thingies to make a delta arrangement
[19:40:40] -!- zzolo has quit [Ping timeout: 276 seconds]
[19:43:02] <skunkworks> rotory thingies?
[19:44:03] <skunkworks> A hexapot would be cooler.
[19:44:06] <skunkworks> heh
[19:44:09] <skunkworks> hexapod
[19:44:32] <cradek> I'm not sure exactly how many of them I have that are complete enough to use
[19:44:42] -!- tmcw has quit [Remote host closed the connection]
[19:44:46] <cradek> way more than 3, maybe or maybe not 6
[19:45:27] <skunkworks> what are they?
[19:47:27] <psha> skunkworks: yea, 3 actuators to control arm and 3 - for leg :]
[19:48:40] <cradek> I can't find a picture. imagine an 8ish inch rotary table, take off the table part, add toothed pulleys and a servo and encoder
[19:49:00] <skunkworks> cradek, did you see the delta was running the torcher program?
[19:49:09] <jepler> torture
[19:49:10] <jepler> ?
[19:49:16] <cradek> yes
[19:49:19] <jepler> argh, why can't I stop myself from correcting spelling mistakes
[19:49:23] <jepler> it's a terrible habit
[19:49:38] <skunkworks> heh - wow - I did mange that one...
[19:49:54] <cradek> haha
[19:50:52] <skunkworks> my wife does feel sorry for you guys...
[19:54:15] <seb_kuzminsky> and we feel sorry for your wife
[19:54:24] <seb_kuzminsky> just kidding! ;-)
[19:54:38] <CaptHindsight> http://www.youtube.com/watch?v=AYs6jASd_Ww Rostock delta robot 3D printer vertical accuracy, currently within 0.01 inch / 0.25 mm.
[19:55:57] <jepler> yeah I've been talking to cradek about delta 3d printers all day
[19:57:06] <CaptHindsight> http://www.adept.com/products/robots/parallel/quattro-s650h/technical-data not too far from the Adept specs
[19:57:56] <skunkworks> seb_kuzminsky, heh
[19:58:09] <seb_kuzminsky> :-)
[20:00:08] -!- zzolo has quit [Ping timeout: 252 seconds]
[20:02:08] <cradek> "A programmer's wife tells him: `Run to the store and pick up a loaf of bread. If they have eggs, get a dozen.' The programmer comes home with 12 loaves of bread."
[20:02:34] <jepler> now that's a joke
[20:03:54] <cradek> "It's hard to explain puns to kleptomaniacs because they always take things literally."
[20:04:02] <seb_kuzminsky> not 13 loaves?
[20:04:05] <cradek> oh these are great
[20:04:09] <seb_kuzminsky> there's a bug in your joke
[20:04:46] <jepler> hm I can certainly see the argument for 13
[20:04:51] <cradek> seb_kuzminsky: I agree that her spec was incomplete, but may or may not agree there is a bug in the joke
[20:05:07] <seb_kuzminsky> coding to a spec is like walking on water:
[20:05:12] <seb_kuzminsky> it's easier if it's frozen
[20:09:35] -!- gimpswork has quit [Ping timeout: 240 seconds]
[20:16:08] -!- jthornton has quit [Read error: Connection reset by peer]
[20:16:34] -!- jthornton [jthornton!~john@99.198.67.11] has joined #linuxcnc-devel
[20:27:42] -!- skunkworks has quit [Read error: Connection reset by peer]
[20:41:48] -!- TekniQue has quit [Ping timeout: 245 seconds]
[20:42:13] <cradek> in the break room, approximately 3 man-hours were just spent arguing about the bread/eggs joke
[20:42:49] -!- TekniQue has quit [Changing host]
[20:45:59] <seb_kuzminsky> heh
[20:53:06] <cradek> An infinite number of mathematicians walk into a bar. The first one orders one beer, the second orders half a beer, the next orders one quarter of a beer, and so on. The bartender pours two pints, sets them on the counter, and says "you mathematicians just don't know your limits."
[20:55:01] <jepler> > 5 out of 10 math teachers think they should be 1 out of 2 math teachers.
[20:55:57] -!- jfire has quit [Quit: Leaving.]
[20:57:16] <jepler> > You won't get my Haskell joke until you need to.
[20:59:10] -!- DaViruz has quit [Read error: Connection reset by peer]
[21:00:00] -!- skunkworks [skunkworks!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[21:00:59] -!- FinboySlick has quit [Quit: Leaving.]
[21:03:26] -!- DJ9DJ has quit [Quit: bye]
[21:07:26] -!- psha has quit [Quit: Lost terminal]
[21:07:37] -!- zzolo has quit [Ping timeout: 246 seconds]
[21:15:49] -!- tmcw has quit [Remote host closed the connection]
[21:17:53] -!- gimpswork has quit [Ping timeout: 248 seconds]
[21:20:11] -!- motioncontrol has quit [Remote host closed the connection]
[21:23:07] -!- mozmck has quit [Quit: Leaving.]
[21:26:26] -!- zzolo has quit [Quit: zzolo]
[21:27:02] <seb_kuzminsky> the best thing about udp jokes is, it's ok if you don't get it
[21:28:28] -!- mourner has quit [Quit: mourner]
[21:34:26] -!- [1]DaveB [[1]DaveB!~MVcalypso@c-24-4-122-138.hsd1.ca.comcast.net] has joined #linuxcnc-devel
[21:35:21] -!- [1]DaveB has quit [Client Quit]
[21:43:22] -!- chillly has quit [Quit: Leaving]
[21:45:03] -!- rosenjon has quit [Quit: rosenjon]
[21:46:57] -!- tjb1 has quit [Client Quit]
[21:47:19] -!- Tecan has quit [Changing host]
[21:49:58] -!- mourner has quit [Client Quit]
[21:55:23] -!- erasmo has quit [Remote host closed the connection]
[21:57:31] -!- stsydow has quit [Quit: Leaving]
[22:02:11] -!- `Nerobro has quit [Remote host closed the connection]
[22:02:38] -!- mourner has quit [Client Quit]
[22:03:42] -!- IwfY has quit [Quit: Leaving]
[22:07:27] -!- `Nerobro_ has quit [Ping timeout: 256 seconds]
[22:09:37] -!- louis____ has quit [Ping timeout: 248 seconds]
[22:13:47] -!- lomach has quit [Quit: Page closed]
[22:26:23] -!- Meduza has quit [Quit: http://quassel-irc.org - Chatta smidigt. Överallt.]
[22:28:40] -!- DaveCVI [DaveCVI!~DaveCVI@c-24-4-122-138.hsd1.ca.comcast.net] has joined #linuxcnc-devel
[22:32:01] -!- theorbtwo has quit [Ping timeout: 248 seconds]
[22:32:16] -!- tmcw has quit [Ping timeout: 276 seconds]
[22:32:39] -!- syyl has quit [Quit: Leaving]
[22:42:41] -!- |1li has quit [Quit: Leaving]
[22:44:49] -!- DaveCVI has quit [Read error: Connection reset by peer]
[22:45:00] -!- DaveCVI [DaveCVI!~DaveCVI@c-24-4-122-138.hsd1.ca.comcast.net] has joined #linuxcnc-devel
[22:56:36] -!- odogono has quit [Quit: odogono]
[22:59:26] -!- tommy80 has quit [Quit: Page closed]
[23:50:08] -!- mpictor has quit [Ping timeout: 245 seconds]
[23:52:02] -!- asdfasd has quit [Ping timeout: 240 seconds]
[23:56:13] -!- DaveCVI has quit [Ping timeout: 252 seconds]
[23:56:32] -!- DaveCVI [DaveCVI!~DaveCVI@c-24-4-122-138.hsd1.ca.comcast.net] has joined #linuxcnc-devel