Back
[00:03:15] -!- mhaberler has quit [Quit: mhaberler]
[00:03:15] -!- logger[mah] has quit [Remote host closed the connection]
[00:03:22] -!- logger[mah] [logger[mah]!~loggermah@mail.mah.priv.at] has joined #linuxcnc-devel
[00:05:28] -!- Nick001-Shop has quit [Remote host closed the connection]
[00:09:09] -!- ve7it has quit [Remote host closed the connection]
[00:14:10] -!- rob_h [rob_h!~rob_h@5e083ea0.bb.sky.com] has joined #linuxcnc-devel
[00:19:32] -!- adb [adb!~IonMoldom@178-211-235-11.dhcp.voenergies.net] has joined #linuxcnc-devel
[00:41:00] -!- gmagno has quit [Quit: Enough small talk...]
[00:41:07] -!- broofa has quit [Quit: Computer has gone to sleep.]
[00:45:09] -!- broofa has quit [Client Quit]
[01:05:26] -!- theos has quit [Disconnected by services]
[01:13:12] -!- broofa has quit [Ping timeout: 272 seconds]
[01:16:55] -!- andypugh has quit [Quit: andypugh]
[01:22:42] -!- rob_h has quit [Ping timeout: 256 seconds]
[01:25:02] -!- RussianKid1 has quit [Remote host closed the connection]
[01:29:28] -!- archivist has quit [Read error: Operation timed out]
[01:32:13] -!- the_wench has quit [Ping timeout: 245 seconds]
[01:39:13] -!- sumpfralle has quit [Ping timeout: 252 seconds]
[01:40:48] -!- odogono has quit [Read error: Connection reset by peer]
[02:00:39] -!- adb has quit [Ping timeout: 265 seconds]
[02:12:04] -!- broofa has quit [Quit: Computer has gone to sleep.]
[02:14:30] -!- broofa has quit [Client Quit]
[02:48:20] -!- gmouer has quit []
[03:00:10] -!- Keknom has quit [Quit: Leaving.]
[03:38:50] -!- zzolo has quit [Quit: zzolo]
[03:48:16] -!- Valen has quit [Quit: Leaving.]
[04:25:01] -!- bradsimantel has quit [Quit: bradsimantel]
[04:28:49] -!- kmiyashiro has quit [Quit: kmiyashiro]
[04:53:09] -!- FinboySlick has quit [Quit: Leaving.]
[04:53:16] -!- kmiyashiro has quit [Quit: kmiyashiro]
[05:00:47] -!- tjb1 has quit [Quit: tjb1]
[05:02:19] -!- Fox_Muldr has quit [Ping timeout: 260 seconds]
[05:12:34] -!- bradsimantel has quit [Quit: bradsimantel]
[05:18:23] -!- wsjr has quit [Read error: Connection reset by peer]
[05:29:15] -!- wsjr has quit [Read error: Connection reset by peer]
[05:54:40] -!- mhaberler [mhaberler!~mhaberler@extern-183.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[06:11:32] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[06:13:15] -!- psha[work] [psha[work]!~psha@195.135.238.205] has joined #linuxcnc-devel
[06:34:58] -!- skorket has quit [Quit: Leaving]
[06:46:00] -!- x10 has quit [Client Quit]
[06:48:45] -!- Cylly has quit []
[07:10:49] -!- kmiyashiro has quit [Quit: kmiyashiro]
[07:30:51] -!- grummund has quit [Ping timeout: 245 seconds]
[07:34:09] -!- racycle has quit [Quit: racycle]
[07:56:19] -!- kmiyashiro has quit [Quit: kmiyashiro]
[08:04:33] -!- mhaberler has quit [Quit: mhaberler]
[08:05:21] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[08:07:14] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[08:18:32] -!- rob_h [rob_h!~rob_h@5e083ea0.bb.sky.com] has joined #linuxcnc-devel
[08:30:40] -!- JT-Shop has quit [Ping timeout: 246 seconds]
[08:30:41] -!- jthornton has quit [Ping timeout: 245 seconds]
[08:32:38] -!- gmagno has quit [Quit: Enough small talk...]
[08:50:51] -!- maximilian_h [maximilian_h!~bonsai@130.255.104.21] has joined #linuxcnc-devel
[08:56:09] -!- sumpfralle has quit [Ping timeout: 276 seconds]
[09:03:35] -!- jdh has quit [Quit: changing universes]
[09:06:35] -!- Pilla_CNCpc has quit [Quit: ChatZilla 0.9.89 [Firefox 11.0/20120310193829]]
[09:08:39] -!- V0idExp has quit [Ping timeout: 246 seconds]
[09:12:29] -!- plushy has quit [Ping timeout: 256 seconds]
[09:20:03] -!- V0idExp has quit [Ping timeout: 260 seconds]
[09:25:26] -!- maximilian_h has quit [Quit: Leaving.]
[09:26:59] -!- maximilian_h [maximilian_h!~bonsai@130.255.104.21] has joined #linuxcnc-devel
[10:14:25] -!- factor has quit [Read error: Operation timed out]
[10:21:56] -!- dimas_ has quit [Ping timeout: 245 seconds]
[10:22:08] -!- rob__H [rob__H!~rob_h@5e083ea0.bb.sky.com] has joined #linuxcnc-devel
[10:22:33] -!- toastydeath has quit [Read error: Connection reset by peer]
[10:24:46] -!- t12 has quit [Ping timeout: 252 seconds]
[10:24:58] -!- GargantuaSauce has quit [Quit: No Ping reply in 180 seconds.]
[10:25:43] -!- rob_h has quit [Ping timeout: 245 seconds]
[10:26:45] -!- Tom_itx has quit [Ping timeout: 272 seconds]
[10:29:08] -!- JT-Shop [JT-Shop!~John@184.63.140.99] has joined #linuxcnc-devel
[10:29:16] -!- jthornton [jthornton!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[10:37:01] -!- Thetawaves has quit [Remote host closed the connection]
[10:41:13] -!- ChanServ has quit [*.net *.split]
[10:41:13] -!- cradek has quit [*.net *.split]
[10:41:14] -!- r00t4rd3d has quit [*.net *.split]
[10:41:14] -!- phantoxeD has quit [*.net *.split]
[10:41:14] -!- trygve has quit [*.net *.split]
[10:41:14] -!- olsson has quit [*.net *.split]
[10:41:14] -!- cherry_lin has quit [*.net *.split]
[10:41:15] -!- jamesp has quit [*.net *.split]
[10:41:15] -!- archivist has quit [*.net *.split]
[10:41:15] -!- MattyMatt has quit [*.net *.split]
[10:41:15] -!- pjm__ has quit [*.net *.split]
[10:41:15] -!- fragalot has quit [*.net *.split]
[10:41:16] -!- wsjr has quit [*.net *.split]
[10:41:16] -!- Tecan has quit [*.net *.split]
[10:41:16] -!- KGB-linuxcnc has quit [*.net *.split]
[10:41:16] -!- Ekkeri has quit [*.net *.split]
[10:41:16] -!- dimas has quit [*.net *.split]
[10:41:16] -!- LeelooMinai has quit [*.net *.split]
[10:41:17] -!- GargantuaSauce_ has quit [*.net *.split]
[10:41:17] -!- djheinzNO has quit [*.net *.split]
[10:41:17] -!- bengan has quit [*.net *.split]
[10:41:17] -!- crib has quit [*.net *.split]
[10:41:17] -!- jepler has quit [*.net *.split]
[10:41:18] -!- steves_logging has quit [*.net *.split]
[10:41:18] -!- RagingComputer has quit [*.net *.split]
[10:41:18] -!- pilla_ has quit [*.net *.split]
[10:41:18] -!- maximilian_h has quit [*.net *.split]
[10:41:19] -!- the_wench has quit [*.net *.split]
[10:41:19] -!- SWPadnos has quit [*.net *.split]
[10:41:19] -!- PCW has quit [*.net *.split]
[10:41:19] -!- archivist_herron has quit [*.net *.split]
[10:41:19] -!- _bwg_ has quit [*.net *.split]
[10:41:20] -!- Connor has quit [*.net *.split]
[10:41:20] -!- rob__H has quit [*.net *.split]
[10:41:20] -!- Loetmichel has quit [*.net *.split]
[10:41:20] -!- mozmck has quit [*.net *.split]
[10:41:20] -!- djdelorie has quit [*.net *.split]
[10:41:21] -!- NeZinovaS has quit [*.net *.split]
[10:41:21] -!- log_ybon has quit [*.net *.split]
[10:41:21] -!- ReadError has quit [*.net *.split]
[10:41:21] -!- hardik has quit [*.net *.split]
[10:41:21] -!- dimas_ has quit [*.net *.split]
[10:41:22] -!- Valen has quit [*.net *.split]
[10:41:22] -!- theos has quit [*.net *.split]
[10:41:22] -!- zlog has quit [*.net *.split]
[10:41:23] -!- TekniQue has quit [*.net *.split]
[10:44:57] -!- rob__H [rob__H!~rob_h@5e083ea0.bb.sky.com] has joined #linuxcnc-devel
[10:44:57] -!- SWPadnos [SWPadnos!~Me@emc/developer/SWPadnos] has joined #linuxcnc-devel
[10:44:57] -!- zlog [zlog!~zlog@ip68-102-196-2.ks.ok.cox.net] has joined #linuxcnc-devel
[10:44:57] -!- mozmck [mozmck!~moses@client-204.235.45.161.wcfltx.partnershipbroadband.com] has joined #linuxcnc-devel
[10:44:57] -!- cradek [cradek!~chris@emc/board-of-directors/cradek] has joined #linuxcnc-devel
[10:44:57] -!- KGB-linuxcnc [KGB-linuxcnc!~kgb-linux@git.linuxcnc.org] has joined #linuxcnc-devel
[10:44:57] -!- Ekkeri [Ekkeri!ekke@hilla.kapsi.fi] has joined #linuxcnc-devel
[10:44:57] -!- jepler [jepler!~jepler@emc/developer/pdpc.professional.jepler] has joined #linuxcnc-devel
[10:44:57] -!- steves_logging [steves_logging!~Steve@wsip-70-168-134-18.dc.dc.cox.net] has joined #linuxcnc-devel
[10:44:57] -!- mode/#linuxcnc-devel [+vv SWPadnos] by pratchett.freenode.net
[10:50:36] -!- bmwyss has quit [Read error: Connection reset by peer]
[10:57:38] -!- pilla_ has quit [Ping timeout: 245 seconds]
[11:03:16] -!- sumpfralle1 has quit [Ping timeout: 248 seconds]
[11:48:42] -!- dhoovie has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
[12:00:29] plushy1 is now known as plushy
[12:04:04] -!- bmwyss has quit [Ping timeout: 246 seconds]
[12:04:05] bmwyss_ is now known as bmwyss
[12:34:43] -!- archivist_herron has quit [Ping timeout: 245 seconds]
[12:35:51] -!- factor has quit [Quit: Leaving]
[12:36:33] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[12:46:20] <skunkworks> JT-Shop, technically - I bet mach does run some profiles faster... If it is using full accelleration. From what I have read though mach doesn't alway obey machine constraints..
[12:49:39] -!- gmagno has quit [Read error: Operation timed out]
[12:52:47] <skunkworks> chris tried to recently allow linuxcnc to use full accelleration but decided there was too many issues in its current archetecture.
[12:59:07] <jthornton> what is full acceleration?
[12:59:38] <skunkworks> during CV - linuxcnc only used 1/2accelleration.
[13:00:02] <skunkworks> uses
[13:00:13] <jthornton> Constant Velocity?
[13:00:17] <skunkworks> yes
[13:00:21] <skunkworks> g64
[13:00:32] <jthornton> so that is lathe only?
[13:00:39] <skunkworks> everything...
[13:00:40] <jthornton> ok
[13:01:50] <jthornton> g64 = go as fast as you can no matter where you go
[13:02:47] <jthornton> I found with G64 Pn on my plasma it runs much faster and smoother than with the default G64
[13:03:54] <skunkworks> right - because it take short segments and combines them into longer segments. so it can keep the velocity up. It is still only using 1/2 acc for those moves.
[13:04:16] <jthornton> I didn't know that
[13:04:26] <jthornton> or I forgot lol
[13:04:35] <skunkworks> heh ;)
[13:05:10] <jthornton> is that what the guy on the forum was talking about with the math forumulas?
[13:06:10] <jthornton> http://linuxcnc.org/index.php/english/component/kunena/?func=view&catid=20&id=24948&limit=6&start=12#25133
[13:06:13] <skunkworks> On cnczone?
[13:06:24] <skunkworks> the kflop guy?
[13:06:34] <jthornton> no the link above
[13:06:47] <jthornton> diegoroman17
[13:07:03] <skunkworks> oh - that could be...
[13:07:39] <jthornton> he is saying there might be a bug in the manual with the formula
[13:08:28] <skunkworks> huh - someone in the know should probably look at that.
[13:08:52] <jthornton> or someone with the math skills to understand it
[13:09:05] <skunkworks> Maybe if we keep chris in banjo parts - he would look into the acc patch more ;)
[13:09:10] <jthornton> I've asked a couple of times for someone to look at it
[13:09:24] <jthornton> banjo parts?
[13:09:43] <skunkworks> chris plays banjo
[13:09:50] <jthornton> ah ok
[13:10:21] <jthornton> I know which end is which but my skills with a banjo stop there
[13:10:57] <skunkworks> same here
[13:11:04] <skunkworks> (for any musical instrument)
[13:12:31] <jthornton> yea me too, tried once to play the guitar but that never panned out although I used to play the piano when I was little
[13:12:50] <jthornton> not very well so my sister says
[13:12:59] <skunkworks> heh
[13:13:14] <skunkworks> I spent most of my time taking things apart.
[13:13:18] -!- arvind_khadri has quit [Ping timeout: 272 seconds]
[13:41:16] -!- mk0 has quit [Read error: Connection reset by peer]
[13:42:14] -!- karavanjoW has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/]
[14:07:13] <skunkworks> but you do get FO, MV, feed hold and other stuff that works instantaniously...
[14:09:02] -!- mutilator has quit [Read error: Connection reset by peer]
[14:16:08] -!- Valen has quit [Quit: Leaving.]
[14:23:36] <skunkworks> and linuxcnc is awesome
[14:23:43] -!- odogono has quit [Ping timeout: 252 seconds]
[14:38:09] <skunkworks> logger[ps
[14:38:11] <skunkworks> logger[psha],
[14:42:03] -!- odogono has quit [Quit: odogono]
[14:46:10] -!- bradsimantel has quit [Quit: bradsimantel]
[14:55:57] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[14:59:00] -!- broofa has quit [Quit: Computer has gone to sleep.]
[14:59:22] -!- bradsimantel has quit [Client Quit]
[15:43:03] -!- psha[work] has quit [Quit: Lost terminal]
[15:49:43] -!- motioncontrol has quit [Client Quit]
[15:58:29] -!- joe9 has quit [Quit: leaving]
[16:10:23] -!- dimas_ has quit [Ping timeout: 244 seconds]
[16:11:14] -!- JT-Shop-2 [JT-Shop-2!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[16:11:15] -!- JT-Shop has quit [Read error: Connection reset by peer]
[16:11:15] -!- jthornton has quit [Read error: Connection reset by peer]
[16:11:16] -!- jthornton_ [jthornton_!~john@184.63.140.99] has joined #linuxcnc-devel
[16:11:47] -!- sumpfralle has quit [Ping timeout: 265 seconds]
[16:12:02] -!- bradsimantel has quit [Quit: bradsimantel]
[16:13:45] -!- bradsimantel has quit [Client Quit]
[16:14:03] -!- V0idExp has quit [Remote host closed the connection]
[16:15:10] JT-Shop-2 is now known as JT-Shop
[16:29:49] -!- siji has quit [Client Quit]
[16:30:38] -!- automata_ has quit [Ping timeout: 265 seconds]
[16:34:10] -!- wsjr has quit [Read error: Connection reset by peer]
[16:39:24] <mhaberler> automata and me have been working on xenomai x86 kernels - looks promising: we have a linxux-3.2.21-xenomai-2.6.1 kernel and it doesnt go beyond 18us latency on a dualcore atom with isolcpus=1
[16:39:50] <mhaberler> box loaded to the gills: 2 glxgears, over vnc, load avg 5, and firefox on top
[16:45:32] -!- karavanjoW has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/]
[16:46:35] <awallin> sounds great, what latency does the same box give with rtai?
[16:46:58] <awallin> 7us?
[16:57:33] -!- jamesp has quit [Ping timeout: 245 seconds]
[17:01:23] <mhaberler> need to look, hold on
[17:07:01] <mhaberler> well, under rtai the box cant even take that load, vnc doesnt respond after second glxgear.. need to do on real console
[17:08:27] jthornton_ is now known as jthornton
[17:10:37] -!- mutilator has quit [Read error: Connection reset by peer]
[17:12:01] -!- gmouer has quit [Client Quit]
[17:12:54] <mhaberler> goes up to about 15uS
[17:13:13] <mhaberler> the worst case with xenomai latency was 21 uS eventually
[17:13:43] -!- phantoxe has quit []
[17:16:50] -!- mattswe has quit [Remote host closed the connection]
[17:16:52] -!- phillipadsmith has quit [Remote host closed the connection]
[17:19:54] <mhaberler> and while I pushed it, the RTAI kernel oopsed with an illegal instruction in the X server
[17:21:30] <mhaberler> fairly similar results on a Intel(R) Pentium(R) 4 CPU 2.53GHz MSI board, around 18uS
[17:25:21] <mhaberler> you guys really should see this - where the RTAI UI becomes unusable xenomai Desktop still perfectly responsive
[17:25:37] <mhaberler> that kernel btw runs fine with 12.04, even under virtualbox
[17:27:29] <mhaberler> some load scenarios which run fine with xenomai I just can reproduce with RTAI, eg VNC immediately hangs up after connecting
[17:32:07] <skunkworks> sounds neet - how much work to get it working with linuxcnc? I see the rtai wrapper isn't an option anymore..
[17:38:21] <mhaberler> automata is working on integrating it, and it has been done before, so it is definitely possible
[17:38:43] <mhaberler> if anybody wants to give that kernel a spin: the debs are here:
http://static.mah.priv.at/public/debian/xenomai-3.2.21-x86-xenomai-2.6.1/
[17:46:59] <skunkworks> so - install 12.04 -> install above debs -> run latency test (how?)
[17:47:53] -!- bmwyss has quit [Quit: bmwyss]
[18:08:57] -!- adb [adb!~IonMoldom@178-211-235-11.dhcp.voenergies.net] has joined #linuxcnc-devel
[18:09:46] -!- vladimirek has quit [Remote host closed the connection]
[18:09:53] -!- Gast950 has quit [Quit: Bye Bye]
[18:10:32] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[18:18:28] -!- morfic has quit [Ping timeout: 248 seconds]
[18:20:01] <skunkworks> looks like /usr/xenomai/bin/latency
[18:20:08] -!- LeelooMinai has quit [Ping timeout: 255 seconds]
[18:21:14] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 14.0.1/20120713225625]]
[18:51:41] -!- mattswe has quit [Remote host closed the connection]
[18:58:08] <mhaberler> yes, sorry
[18:58:14] <mhaberler> that should go on the path
[18:58:38] -!- phillipadsmith has quit [Read error: Connection reset by peer]
[18:58:38] -!- mattswe has quit [Remote host closed the connection]
[18:58:38] <mhaberler> what's your results?
[19:01:42] <skunkworks> not that far.. ;)
[19:02:11] <skunkworks> the 12.04 cd I had was old and didn't boot on some new hardware I had. (amd - asus motherboard)
[19:02:23] <skunkworks> installing 12.04 right now
[19:03:18] <skunkworks> it will be interesting if all work what the latency is - with rtai - it is around 30k
[19:03:23] <skunkworks> 30us
[19:04:24] -!- mattswe has quit [Remote host closed the connection]
[19:04:43] -!- siji has quit [Quit: Leaving]
[19:06:00] <skunkworks> mhaberler, is there an easy way to install all the debs?
[19:06:28] <mhaberler> dpkg -i *.deb
[19:06:33] <skunkworks> cool
[19:06:59] <mhaberler> sorry, I dont have a debian ppa or somesuch
[19:07:19] <skunkworks> another stupid question. The kernel has to be correct - right?
[19:07:36] <skunkworks> 3.2.21
[19:07:47] <skunkworks> is that the latest on 10.04?
[19:07:53] <skunkworks> *12.04
[19:08:17] <mhaberler> no, it is unrelated to ubuntu kernels, thats a debian kernel - will run fine though
[19:08:24] <skunkworks> ok
[19:08:50] <mhaberler> need to choose a kernel version which fits the rt patches, so ubuntu kernel as base is not an option
[19:09:24] <mhaberler> I'm trying 2.6.38.8-xenomai-2.6.1 now as well and seems to work, too
[19:10:10] <skunkworks> wait - so - I cannot try this on 12.04 ubuntu?
[19:10:29] <mhaberler> yes you can
[19:10:32] <skunkworks> heh - ok
[19:10:46] <mhaberler> that's the whole point to start with..
[19:10:51] <mhaberler> and then some
[19:11:34] <skunkworks> I thought I understood what was going on ;)
[19:13:38] <mozmck> mhaberler: That is good to hear. I've been wanting to try making LinuxCNC run on xenomai for a while, but just have had no time.
[19:14:10] <mhaberler> this a running kernel. This is NOT 'linuxCNC on xenomai' yet.
[19:14:18] <mozmck> Oh, I see.
[19:14:34] <mozmck> How much work does it look to get linuxCNC to work?
[19:14:42] <mozmck> I might can carve out a little time to help.
[19:17:09] <mhaberler> it will be work on rtapi; automata committed to doing it and there are examples to start from, any hand would be welcome!
[19:17:40] -!- KimK has quit [Ping timeout: 248 seconds]
[19:21:10] <mhaberler> there's a some work needed for the kernels (both xenomai and rt-preempt) and arranging with seb for buildbot setup
[19:27:57] <mozmck> What will the buildbot do with the kernel? Once it's built it shouldn't need to be built again?
[19:28:21] -!- zzolo has quit [Quit: zzolo]
[19:28:50] <mhaberler> let me turn the question around: find the parts and instructions to reproduce the existing RTAI kernel.
[19:28:51] <mozmck> I did see there is a xenomai_rtapi.c now, is that where the work is happening?
[19:29:29] <mozmck> I built the 10.04 rtai kernel and the live-cd.
[19:29:38] <mhaberler> if you build on your box and push a .deb , nobody knows how to rebuild it . It's also about documenting the work
[19:29:48] <mozmck> true.
[19:30:20] <mhaberler> where exactly do you see xenomai_rtapi.c
[19:30:21] <mozmck> I believe my notes on building the kernel are on the wiki, but they could be better.
[19:31:20] <mhaberler> it is not in the linuxcnc repo, afaict
[19:31:47] <mozmck> xenomai_rtapi.c is in the rtapi dir in the master branch.
[19:32:32] <mozmck> The file date is 2010, but I don't think it was there before.
[19:32:50] <mhaberler> please post a liml?
[19:33:09] <mozmck> badly worded: I don't think it was there a couple months ago.
[19:33:12] <mhaberler> no in my master:
http://git.mah.priv.at/gitweb/emc2-dev.git/tree/ce55978c1af6941c101cd2a7191c3fe810888fd1:/src/rtapi
[19:33:13] <mozmck> liml?
[19:33:18] <mhaberler> link
[19:34:22] <mhaberler> btw, just got linux-2.6.38-8-xenomai-2.6.1 installed to, runs fine as well - important to have a second option with the kernel (yes, that is an innovation ;)
[19:34:55] <mhaberler> also reliably under 25usec on an atom/isolcpus under a load which makes rtai unusable
[19:35:22] <mozmck> interesting.
[19:35:56] <mozmck> I was looking at xenomai a while back because it supports some ARM chips that RTAI does not. The one that's on the beagleboard for instance.
[19:36:19] <mhaberler> ah! yes! we want to get rid of the x86 lock-in!
[19:36:30] <mhaberler> that was three days 'operation iron butt' to figure how to build these things, I dont do that every day
[19:38:29] <mozmck> I went through that building the kernel for 10.04 :)
[19:39:02] <mhaberler> and then ran out of steam documenting it, hm ;-?
[19:39:19] <mozmck> I bought an AMD 6-core because each compile and package build took almost 2 hours. The 6-core took the time down to 17 minutes.
[19:40:03] <mhaberler> here are the working 2.6.38.8 kernels :
http://static.mah.priv.at/public/debian/xenomai-2.6.38.8-2.6.1/
[19:44:03] <mhaberler> on the P4 under 15uS
[19:44:44] <mhaberler> well in summary I fail to understand why this community has been putting up with RTAI and all its drawbacks for so long.
[19:45:40] <mozmck> I think because xenomai did not have good enough latencies before?
[19:46:22] <skunkworks> Hmm - latency does nothing..
[19:46:39] <skunkworks> I think I installed all the debs.. \
[19:46:51] <mozmck> And they have not been this slow supporting new kernels before that I've seen.
[19:47:22] <mhaberler> skunkworks: you need to create an initrd image:
[19:47:30] <skunkworks> heh - ok.
[19:47:35] <skunkworks> how do I do that?
[19:48:05] <mhaberler> update-initramfs -c -k 3.2.21-xenomai-2.6.1
[19:48:09] <mhaberler> then run
[19:48:13] <mhaberler> update-grub
[19:48:38] <mhaberler> for the 2.6 kernel its similar:
[19:48:57] <mhaberler> update-initramfs -c -k 2.6.38.8-xenomai-2.6.1
[19:49:28] <mhaberler> it should come with the deb, for some reason I dont understand it doesnt
[19:52:33] <mhaberler> bbl
[19:55:09] -!- broofa has quit [Quit: Computer has gone to sleep.]
[19:55:48] <mozmck> mhaberler: here's my kernel notes for 10.04:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Ubuntu10.04Notes
[19:56:55] -!- mhaberler has quit [Ping timeout: 246 seconds]
[19:57:37] -!- KimK [KimK!~Kim__@wsip-184-176-200-171.ks.ks.cox.net] has joined #linuxcnc-devel
[19:58:50] <skunkworks> heh
[19:59:10] <skunkworks> I did the above - now - in the olden days I needed to pick the kernel I wanted to run.
[20:03:27] -!- mhaberler [mhaberler!~mhaberler@extern-183.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[20:03:57] -!- empire [empire!~chatzilla@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[20:04:18] <empire> this looks good
[20:04:22] <empire> Linux empire-System-Product-Name 3.2.21-xenomai-2.6.1 #1 SMP PREEMPT Fri Oct 12
[20:04:32] empire is now known as skunkworks_
[20:04:35] skunkworks_ is now known as skunkworks__
[20:04:51] <skunkworks__> but the latency test doesn't seem to be there.
[20:05:31] -!- tjb1 has quit [Quit: tjb1]
[20:05:47] <mhaberler> look in /usr/xenomai/bin
[20:05:50] <mhaberler> need to be root
[20:05:57] <mhaberler> ./latency-test
[20:07:07] <skunkworks__> there is not xenomai in usr
[20:07:25] <mhaberler> then you need to download this:
[20:07:53] <mhaberler> http://static.mah.priv.at/public/debian/xenomai-3.2.21-x86-xenomai-2.6.1/xenomai-runtime_2.6.1_i386.deb
[20:08:05] <mhaberler> and 'dpkg -i thisfile'
[20:09:44] <skunkworks__> done - but still not a xenomai in /usr
[20:10:15] <mozmck> mhaberler: here's my kernel notes for 10.04:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Ubuntu10.04Notes
[20:10:15] <mhaberler> let me check, it might have been the wrong one
[20:10:49] <skunkworks__> Preparing to replace xenomai-runtime 2.6.1 (using xenomai-runtime_2.6.1_i386.deb) ...
[20:10:56] <skunkworks__> (is what installed)
[20:12:59] -!- wboykinm has quit [Remote host closed the connection]
[20:13:32] <mozmck> ok, I just figured out that the xenomai_rtapi.c is one that I made a while back :) I don't think I did much with it.
[20:14:21] <mhaberler> skunkworks: it might be under /usr/lib/xenomai
[20:14:36] <skunkworks> will look
[20:14:49] <mhaberler> mozmck: did you do any work on it which can be salvaged?
[20:15:15] <mozmck> I don't know. I may have only copied over the rtai files.
[20:17:17] <mhaberler> that wiki page: is that the basis of the current kernel, or your attempt to recreate it?
[20:20:14] <skunkworks__> got it.
[20:20:24] <skunkworks__> bad latency on this thing.. 283us
[20:20:40] <skunkworks__> let me try a video card
[20:20:43] <skunkworks__> bbl
[20:21:17] -!- skunkworks__ has quit [Remote host closed the connection]
[20:22:53] <mozmck> That is the steps I used to create the kernel for 10.04
[20:25:15] <mhaberler> you mean the one which now everybody uses as 2.6.32-122-rtai ?
[20:25:21] <mozmck> yes
[20:25:32] <skunkworks> hmm - I don't have any video cards...
[20:27:00] <mozmck> The nice thing about using the ubuntu kernel is that the build system gives you ready to install .deb packages without too much trouble.
[20:27:06] <mhaberler> are you definitely running this kernel - what does uname -a say?
[20:27:18] <skunkworks> Me?
[20:27:24] <mhaberler> yes
[20:27:31] <mhaberler> just to make sure
[20:27:41] -!- zzolo has quit [Quit: zzolo]
[20:27:49] <skunkworks> empire-System-Product-Name 3.2.21-xenomai-2.6.1 #1 SMP PREEMPT Fri Oct 12
[20:27:56] <mhaberler> sounds right
[20:28:05] <mhaberler> what cpu is that?
[20:28:19] <skunkworks> like I said - this machine was 30-40us on rtai
[20:29:18] <skunkworks> amd.. a6
[20:29:23] <skunkworks> tripple core
[20:30:12] <mhaberler> hn, that build was for a generic intel p4 classic
[20:30:57] <mhaberler> mozmck: is the result of your work anywhere in a public git repo?
[20:31:17] <mozmck> result of my work?
[20:31:41] -!- djcoin has quit [Quit: WeeChat 0.3.7]
[20:32:06] <mhaberler> well I understand you produced the kernel which everybody uses, and you obviously used git, so I wonder if there's a repo where your steps are recorded
[20:32:08] <mozmck> The kernel packages are the result of my work. If you mean the kernel after it was patched and configured, no.
[20:33:26] <mozmck> I'm sure I can reproduce it pretty easily.
[20:34:19] <mhaberler> so assume somebody finds a bug, how does he aquire history?
[20:35:23] <mhaberler> get the source package and poke around in the blob?
[20:35:51] <mhaberler> I just want to find out what the status actually is.
[20:36:28] <mozmck> Hmm. I believe the kernel repository is still here: git://kernel.ubuntu.com/ubuntu/ubuntu-lucid.git ubuntu-lucid and I'm pretty sure all the steps I used to make the current kernel are in the wiki page I mentioned earlier.
[20:37:31] <mhaberler> well sure, I was asking about the delta between the ubuntu repo and what everybody is running,
[20:37:40] <mozmck> I applied the rtai patch, made one fix due to the ubuntu changes, changed the ubuntu generic kernel config as stated in the wiki page, and compiling.
[20:37:50] <mozmck> compiled.
[20:37:56] <mhaberler> assuming you had a git repo you worked in, do you still have it?
[20:39:13] <mozmck> I believe the ubuntu version was 2.6.32-22 and that I changed mine to 122 so it would not replace the generic version and would be the first kernel in the list so it would load automatically.
[20:39:45] <mozmck> I don't think I have that git repo anymore. I should probably have saved it or done a diff.
[20:40:00] <mhaberler> aaaaaaaaargh
[20:45:07] -!- skunkworks__ [skunkworks__!~chatzilla@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[20:45:34] <skunkworks__> put the hd in a intel machine
[20:46:00] <skunkworks__> Pentium(R) Dual-Core CPU E5800 @ 3.20GHz × 2
[20:46:07] <skunkworks__> 20us
[20:48:50] <skunkworks__> http://imagebin.org/index.php?mode=image&id=231774
[20:48:56] <skunkworks__> seems to be leveling off there.
[20:49:23] <mhaberler> mucho better
[20:49:33] <skunkworks__> yes
[20:49:45] <skunkworks__> I don't know what this computer would be with rtai
[20:49:56] <skunkworks__> it has onboard intel graphics
[20:50:05] <mhaberler> ah, that would have been my next question;)
[20:50:20] <skunkworks__> I could test - give me a second
[20:50:26] <mhaberler> sure
[20:51:11] <skunkworks__> 20.561 - now re-booting
[20:51:19] -!- skunkworks__ has quit [Remote host closed the connection]
[20:54:47] -!- motioncontrol has quit [Quit: Sto andando via]
[20:55:24] -!- FinboySlick has quit [Quit: Leaving.]
[20:57:41] -!- Brandonian has quit [Quit: Brandonian]
[21:02:40] <skunkworks> http://imagebin.org/index.php?mode=image&id=231777
[21:02:48] -!- rob__H has quit [Ping timeout: 240 seconds]
[21:02:58] <skunkworks> that was only running a few minutes
[21:04:07] <mhaberler> you mean the same box?
[21:05:32] <skunkworks> same box booted with the linuxcnc livecd
[21:05:45] -!- rob_h [rob_h!~rob_h@5e083ea0.bb.sky.com] has joined #linuxcnc-devel
[21:05:51] <skunkworks> so - pretty close to the same
[21:07:37] <mhaberler> that was my impression too, I saw no differences which matter much in practice
[21:08:35] <mhaberler> it is interesting to see how it performs under load - run stuff like many glxgears as long as its still usable; then try the same after booting rtai
[21:09:30] <skunkworks> with the amd system - the linuxcnc latency test ran for a day or to and stayed below 40us. (283us with xenomi)
[21:09:44] -!- DJ9DJ has quit [Quit: bye]
[21:10:16] <skunkworks> I will have to play
[21:10:23] <mhaberler> that probably needs some experimentation with options, because that's abnormal
[21:10:38] <mhaberler> dont need an rt kernel for that ;)
[21:10:48] <skunkworks> heh
[21:13:09] <skunkworks> pretty damn cool booting the HD on a different computer/processor and it worked no issues. try that with xp
[21:16:10] <mhaberler> well, that's the intel monoculture. no such thing in arm/embedded land
[21:18:38] <mozmck> I thought it was a MS windows thing.
[21:19:43] <mhaberler> that on top;)
[21:19:53] -!- skunkworks has quit [Read error: Connection reset by peer]
[21:20:14] -!- rob_h has quit [Quit: Leaving]
[21:32:15] -!- chillly has quit [Quit: Leaving]
[21:33:46] -!- rob_h [rob_h!~rob_h@5e083ea0.bb.sky.com] has joined #linuxcnc-devel
[21:47:53] toudi_ is now known as micges
[22:00:16] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[22:05:38] -!- zzolo has quit [Quit: zzolo]
[22:12:51] -!- racycle has quit [Quit: racycle]
[22:16:54] -!- micges has quit [Ping timeout: 240 seconds]
[22:18:06] -!- vladimirek has quit [Remote host closed the connection]
[22:26:18] -!- skunkworks__ [skunkworks__!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[22:34:44] -!- cradek has quit [Read error: Operation timed out]
[22:35:17] -!- KGB-linuxcnc has quit [Ping timeout: 255 seconds]
[22:38:23] -!- KGB-linuxcnc [KGB-linuxcnc!~kgb-linux@git.linuxcnc.org] has joined #linuxcnc-devel
[22:40:57] -!- cradek [cradek!~chris@outpost.timeguy.com] has joined #linuxcnc-devel
[22:41:11] -!- cradek has quit [Changing host]
[22:41:11] -!- cradek [cradek!~chris@emc/board-of-directors/cradek] has joined #linuxcnc-devel
[22:41:11] -!- mode/#linuxcnc-devel [+v cradek] by ChanServ
[22:46:33] -!- wsjr has quit [Read error: Connection reset by peer]
[22:48:54] -!- i_tarzan has quit [Ping timeout: 240 seconds]
[23:07:37] -!- mutilator has quit [Read error: Connection reset by peer]
[23:20:27] -!- Loetmichel has quit [Ping timeout: 240 seconds]
[23:46:26] -!- toudi_ has quit [Read error: Connection reset by peer]
[23:46:27] -!- tjb1 has quit [Read error: Connection reset by peer]
[23:52:29] -!- ve7it [ve7it!~LawrenceG@S0106002401cb41eb.pk.shawcable.net] has joined #linuxcnc-devel
[23:56:41] -!- gmagno has quit [Ping timeout: 245 seconds]