Back
[00:13:44] -!- rob_h [rob_h!~rob_h@027c16a0.bb.sky.com] has joined #linuxcnc-devel
[00:37:38] -!- odogono has quit [Read error: Connection reset by peer]
[00:39:07] -!- rob_h has quit [Ping timeout: 264 seconds]
[00:54:20] -!- linlin has quit [Read error: Connection reset by peer]
[01:06:41] -!- Brandonian has quit [Quit: Brandonian]
[01:08:47] -!- odogono has quit [Quit: odogono]
[01:10:30] -!- linlin has quit [Remote host closed the connection]
[01:24:10] -!- adb has quit [Ping timeout: 260 seconds]
[01:26:45] -!- zzolo has quit [Quit: zzolo]
[01:27:15] -!- jfire1 has quit [Quit: Leaving.]
[01:39:20] -!- Dupe has quit [Read error: Connection reset by peer]
[01:54:57] -!- andypugh has quit [Quit: andypugh]
[01:57:18] -!- pfred1 has quit [Quit: bye]
[02:18:46] -!- tjb1 has quit [Ping timeout: 245 seconds]
[02:27:31] -!- tjb1 has quit [Ping timeout: 245 seconds]
[02:31:20] -!- jthornton has quit [Ping timeout: 252 seconds]
[02:31:21] -!- JT-Shop has quit [Read error: Connection reset by peer]
[02:31:29] -!- jthornton_ [jthornton_!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[02:31:32] -!- JT-Shop-2 [JT-Shop-2!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[02:32:46] -!- JT-Shop-3 [JT-Shop-3!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[02:32:46] -!- jthornton__ [jthornton__!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[02:33:35] -!- jthornton_ has quit [Read error: Connection reset by peer]
[02:33:36] -!- JT-Shop-3 has quit [Read error: Connection reset by peer]
[02:33:36] -!- jthornton__ has quit [Read error: Connection reset by peer]
[02:33:37] -!- JT-Shop-2 has quit [Read error: Connection reset by peer]
[02:34:27] -!- morfic- has quit [Remote host closed the connection]
[02:54:05] -!- skorasaurus has quit [Quit: left the building.]
[03:00:03] -!- RangerRick has quit [Remote host closed the connection]
[03:01:53] -!- Keknom has quit [Quit: Leaving.]
[03:07:40] Guest711 is now known as mephux
[03:07:55] -!- mephux has quit [Changing host]
[03:13:20] -!- gimps has quit [Read error: Connection reset by peer]
[03:14:38] -!- L33TG33KG34R has quit [Read error: Connection reset by peer]
[03:20:08] -!- grummund has quit [Ping timeout: 255 seconds]
[03:30:53] -!- jfire has quit [Quit: Leaving.]
[03:36:23] -!- Patang has quit [Ping timeout: 245 seconds]
[04:27:56] -!- sumpfralle has quit [Ping timeout: 252 seconds]
[04:29:39] -!- hdokes has quit [Read error: No route to host]
[04:31:44] -!- FinboySlick has quit [Quit: Leaving.]
[04:32:29] -!- yhojeyisaac has quit [Ping timeout: 245 seconds]
[04:44:42] -!- yhoje has quit [Read error: Connection reset by peer]
[04:48:28] -!- Tecan has quit [Read error: Connection reset by peer]
[05:05:16] -!- JT-Shop-3 [JT-Shop-3!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[05:05:17] -!- jthornton__ [jthornton__!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[05:13:03] -!- RagingComputer has quit [Ping timeout: 245 seconds]
[05:16:15] -!- yhoje has quit [Ping timeout: 245 seconds]
[05:20:30] -!- jfire has quit [Quit: Leaving.]
[05:31:23] -!- AR_ has quit [Ping timeout: 245 seconds]
[05:42:51] -!- kwallace1 [kwallace1!~kwallace@63.83.75.11] has parted #linuxcnc-devel
[05:50:12] -!- tripod| has quit [Ping timeout: 248 seconds]
[05:54:55] -!- pfred1 has quit [Quit: nite]
[06:02:36] -!- Fox_Muldr has quit [Ping timeout: 264 seconds]
[06:05:09] -!- ve7it has quit [Remote host closed the connection]
[06:21:27] -!- psha[work] [psha[work]!~psha@195.135.238.205] has joined #linuxcnc-devel
[06:26:59] -!- tripod| has quit [Ping timeout: 260 seconds]
[06:27:23] -!- dhoovie has quit [Read error: Connection reset by peer]
[06:37:44] -!- morfic has quit [Remote host closed the connection]
[07:05:41] -!- tjb1 has quit [Read error: Connection reset by peer]
[07:26:29] -!- racycle has quit [Quit: racycle]
[07:28:20] -!- firephoto has quit [Quit: ZNC - http://znc.in]
[07:31:32] -!- tripod| has quit [Ping timeout: 248 seconds]
[07:46:23] -!- mhaberler has quit [Quit: mhaberler]
[07:51:17] -!- emel has quit [Excess Flood]
[07:59:49] -!- emel has quit [Excess Flood]
[08:04:11] -!- Dupe has quit [Quit: Leaving]
[08:37:06] -!- tjb1 has quit [Ping timeout: 245 seconds]
[08:44:27] -!- tomate_ [tomate_!~quassel@2a01:e34:ee61:a5f0:14c8:6f5d:f51b:a07c] has joined #linuxcnc-devel
[08:46:56] -!- tjb1 has quit [Quit: tjb1]
[08:58:41] Cylly is now known as Loetmichel
[09:07:25] -!- Tecan has quit [Quit: Ex-Chat]
[09:31:37] -!- b_b has quit [Changing host]
[09:38:40] -!- rob_h [rob_h!~rob_h@027c16a0.bb.sky.com] has joined #linuxcnc-devel
[10:19:38] -!- maximilian_h [maximilian_h!~bonsai@93.132.177.85] has joined #linuxcnc-devel
[10:37:38] -!- skunkworks has quit [Remote host closed the connection]
[10:47:50] -!- Simooon has quit [Remote host closed the connection]
[10:49:08] -!- dhoovie has quit [Ping timeout: 246 seconds]
[11:00:21] jthornton__ is now known as jthornton
[11:09:03] -!- cherry_l1n has quit [Read error: Connection reset by peer]
[11:37:31] -!- seb_kuzm1nsky [seb_kuzm1nsky!~seb@174-16-204-191.hlrn.qwest.net] has joined #linuxcnc-devel
[11:38:15] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[11:40:30] -!- joe9_ [joe9_!~joe9@c-24-98-97-215.hsd1.ga.comcast.net] has joined #linuxcnc-devel
[11:40:44] -!- korsi_ [korsi_!~korsi@adsl-85-217-43-107.kotinet.com] has joined #linuxcnc-devel
[11:43:46] -!- theorbtwo has quit [*.net *.split]
[11:43:46] -!- toxx has quit [*.net *.split]
[11:43:47] -!- korsi has quit [*.net *.split]
[11:43:47] -!- tronwizard has quit [*.net *.split]
[11:43:47] -!- joe9 has quit [*.net *.split]
[11:43:47] -!- IceKiller has quit [*.net *.split]
[11:43:47] -!- KimK has quit [*.net *.split]
[11:43:48] -!- Gene34 has quit [*.net *.split]
[11:43:48] -!- Thetawaves has quit [*.net *.split]
[11:43:48] -!- seb_kuzminsky has quit [*.net *.split]
[11:43:48] -!- kanzure has quit [*.net *.split]
[11:43:48] -!- ssi has quit [*.net *.split]
[11:43:56] theorb is now known as theorbtwo
[11:44:11] -!- jthornton has quit [Quit: Ex-Chat]
[11:45:21] -!- jthornton [jthornton!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[11:47:42] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[11:49:33] -!- dhoovie|2 has quit [Read error: Connection reset by peer]
[11:49:54] -!- KimK [KimK!~Kim__@wsip-184-176-200-171.ks.ks.cox.net] has joined #linuxcnc-devel
[12:29:01] -!- adb [adb!~IonMoldom@178.211.237.94] has joined #linuxcnc-devel
[12:44:20] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[13:15:45] hashfail is now known as gimps
[13:16:13] -!- JT-Shop-3 has quit [Read error: Connection reset by peer]
[13:18:55] -!- JT-Shop [JT-Shop!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[13:28:40] -!- Valen has quit [Quit: Leaving.]
[13:37:18] -!- tmcw has quit [Remote host closed the connection]
[13:38:09] -!- jpk has quit [Ping timeout: 248 seconds]
[13:51:53] -!- maximilian_h has quit [Quit: Leaving.]
[14:16:14] -!- adb has quit [Ping timeout: 255 seconds]
[14:16:55] Icekille- is now known as IceKiller
[14:17:00] -!- IceKiller has quit [Changing host]
[14:19:05] -!- zzolo has quit [Quit: zzolo]
[14:24:03] -!- adb [adb!~IonMoldom@178.211.237.94] has joined #linuxcnc-devel
[14:27:10] <mozmck> Morris mentions getting the LCNC code into the big distributions: what is the reasoning behind this?
[14:28:10] <seb_kuzm1nsky> good question!
[14:28:52] <mozmck> sounds like it will be a *lot* of trouble, but I can't currently see the benefit?
[14:31:23] <jepler> it would take a lot of work off our shoulders if linuxcnc installation was: download $major_distribution's installer; install it; select linuxcnc from the package management system
[14:31:46] <jepler> you know what a pain it is to prepare a kernel and an OS installer image
[14:31:53] <mozmck> What about a PPA?
[14:32:22] <jepler> ppa is ubuntu lockin crap
[14:32:31] <mozmck> :) The kernel is a pain, but the OS was not too bad.
[14:33:04] <mozmck> PPA is somewhat limiting.
[14:34:32] <jepler> but anyway before it's worth having a package in a distribution we would have to run in a way that is fully useful on some kernel that distribution ships .. which is why before this new rtos integration stuff it was basically a non-starter
[14:35:14] <jepler> and honestly I haven't been following the rtos stuff closely enough to know what the situation is (except to find that the debian wheezy -rt kernel doesn't even start on my hardware..)
[14:35:28] <mozmck> so do distributions ship xenomai patched kernels? or just preempt-rt...
[14:35:38] <mhaberler> with a unified binary build, in the minimum you get a 'sim' running; pull a stock rt-preempt kernel - you get rt (modulo rt-preempt performance) - that's a lot already
[14:35:45] -!- psha[work] has quit [Quit: Lost terminal]
[14:36:17] <mhaberler> you can expect an rt-preempt kernel being available stock; for something better you'd need a xenomai or rtai kernel from say here or the xenomai site
[14:37:16] <mozmck> but you can't pull a mesa 5i25 or similar from the distribution, so the preempt-rt help you much :)
[14:37:17] <jepler> "Xenomai is officially included in Debian" --
http://www.xenomai.org/index.php/Building_Debian_packages
[14:37:38] <mhaberler> that is just the userland runtime methinks
[14:37:38] <jepler> though that seems to be in the form of linux-patch-xenomai, so you're stuck building your own and we all know how much (little) fun that is
[14:37:50] <mhaberler> ah, the ptach
[14:38:05] <mozmck> debian also had rtai packages - don't know if they still do though.
[14:38:32] <mhaberler> yes, thats's why John wrapped the 3.5.7 kernel into a binary package
[14:38:50] <jepler> and it doesn't look that fresh, either. debian wheezy is based on the 3.2 kernel but the description of linux-patch-xenomai only lists versions up to 2.6.32.15, so ... stale
[14:38:56] <mozmck> which is what we have already been doing.
[14:40:22] <jepler> but if it was possible to run linuxcnc for sim and for servo-cycle-only configs using only packages from debian main, and do software stepgen by adding only a kernel image from linuxcnc.org -- that would really be nice
[14:40:25] <mhaberler> that's something we need to think about - if the debian xenomai userland support package is too old; we have our own right now
[14:40:40] <mhaberler> that's what I had in mind, yes
[14:40:55] <jepler> (assuming you think debian is a good target, which is the direction my biases point me this year)
[14:41:29] <mhaberler> mine too, plus mint for the UI
[14:41:39] <mozmck> haha. I like debian, ran it for years. Never got into fedora much.
[14:41:57] <mozmck> I run ubuntu with cinnamon for the UI right now.
[14:42:19] <mhaberler> I feel the Ubuntu folks are veering off into mobile/tablet space; not sure how that will pan out for desktops
[14:42:33] <mozmck> gnome3 too
[14:44:17] <seb_kuzm1nsky> i favor debian-based distros
[14:44:30] <seb_kuzm1nsky> i don't mind ubuntu as much as some folks seem to
[14:44:34] <seb_kuzm1nsky> at least not yet...
[14:45:34] seb_kuzm1nsky is now known as seb_kuzminsky
[14:47:28] <seb_kuzminsky> so, speaking of the new rtos branches....
[14:47:59] <seb_kuzminsky> the main obstacle to merging that into master, as i see it, is to unify the configure command
[14:48:23] <seb_kuzminsky> i would really prefer to have a single set of instructions for building any version of our software
[14:48:35] <seb_kuzminsky> "any": v2.5_branch or newer
[14:48:50] <seb_kuzminsky> currently we do "./autogen.sh; ./configure $CONFIGURE_ARGS; make -j"
[14:49:18] <seb_kuzminsky> CONFIGURE_ARGS is either "--enable-simulator" or "" (for rtai)
[14:49:21] <mhaberler> I see why - you will still have to give a kernel version for it
[14:49:35] <mhaberler> otherwise configure cant tell which flavor to build
[14:50:12] <seb_kuzminsky> we could teach it to try to autodetect the common situations, while allowing the user to override in the uncommon/complicated situations
[14:50:25] <mhaberler> so you do './configure' - how is that going to guess this is for rtai, xenomai (which thread style) or rt-preempt
[14:50:37] <seb_kuzminsky> we sort of do that currently: if you don't give a --with-realtime=$RT_DIR, it autodetects that
[14:50:39] <mhaberler> autodetect: from where
[14:50:48] <mhaberler> the running kernel?
[14:51:01] <seb_kuzminsky> the running kernel would be one reasonable option
[14:51:14] <seb_kuzminsky> another is to look in some well-known places for realtime kernels
[14:51:15] <mhaberler> well that is in place anyway
[14:51:23] <seb_kuzminsky> we currently use /usr/realtime-* for rtai
[14:52:12] <mhaberler> that isnt the best bet IMO; the way I did this is to point to a kernel config in /boot - that gives reliable detection
[14:52:20] <seb_kuzminsky> mhaberler: please remind me what the configure command line looks like in the current rtos branch
[14:52:31] <mhaberler> ./configure
[14:52:40] <seb_kuzminsky> you specify "thread style" and something else?
[14:52:46] <mhaberler> it will build for the running kernel
[14:52:57] <mhaberler> for xenomai this defaults to xenomai-user
[14:53:18] <mhaberler> or you override it, for which theres no reason because it's on a deprecation path
[14:53:32] <mhaberler> please pull it and try it out
[14:54:14] <mhaberler> this is the one you want:
http://git.mah.priv.at/gitweb?p=emc2-dev.git;a=shortlog;h=refs/heads/rtos-integration-preview3
[14:54:19] <seb_kuzminsky> for the xenomai kernel you support user threads and kernel threads, but you've deprecated kernel threads, is that right?
[14:54:37] <mhaberler> yes
[14:54:50] <mhaberler> they work, but there is no point in using them
[14:55:06] <mhaberler> that's dropping off the xenomai roadmap
[14:55:14] <seb_kuzminsky> ok
[14:55:37] <seb_kuzminsky> what thread styles do you support for preempt-rt? which do you recommend?
[14:55:43] <mhaberler> just ./configure ; make sould be fine for rtai, xenomai, rt-preempt, vanilla
[14:55:48] <mhaberler> there's only one
[14:55:51] <mhaberler> user
[14:55:52] <seb_kuzminsky> vanilla == sim?
[14:56:01] <seb_kuzminsky> ok, that sounds nice and simple
[14:56:03] <mhaberler> yes
[14:56:39] <mhaberler> to force a sim build: use ./configure --with-threads=posix
[14:56:59] <mhaberler> assume you have rt-preempt, rtai, xenomai, vanilla kernels installed
[14:57:21] <mhaberler> you can build for either one regardless what you are running by point to the kconfig like so:
[14:57:44] <mhaberler> ../configure --with-kernel=/boot/config-my-random-rt-kernels-config
[14:57:58] <seb_kuzminsky> cool
[14:58:16] <mhaberler> only on raspberry and beaglebone you need ./configure --with-platform=raspberry|beaglebone
[14:58:28] <mhaberler> (platform=pc by default)
[14:59:09] <mhaberler> the idea of pointing to /usr/src/-kernel-headers* is not a good one because you dont need them for xeno-user, and rt-preempt
[14:59:16] <seb_kuzminsky> what does --with-platform do?
[14:59:37] <mhaberler> set config.h #define, and make.inc for conditional driver build
[14:59:50] <seb_kuzminsky> ah, conditional drivers, that makes sense
[15:00:00] <mhaberler> no point for hal_parport on say raspberry, doesnt have that
[15:00:06] <mhaberler> mesa - same thing
[15:00:36] <mhaberler> the implicit assumption of kernel headers being present had to go away
[15:01:13] <mhaberler> however, in the rtai and xeno-kernel they will be used of course because module build would fail otherwise
[15:01:31] <seb_kuzminsky> yeah
[15:01:56] <mhaberler> that's why I detoured to configure (which also works from the running kernel from /proc/config.gz)
[15:02:45] <seb_kuzminsky> is the rtos-integration-preview3-merged-into-master branch up to date?
[15:03:07] <mhaberler> but fact is: without pointing to a kconfig or autodetecting the running on configure wont know what you want
[15:03:24] <mhaberler> I need to roll it forward, need to see when that last happened
[15:03:28] <mhaberler> I'll look into it
[15:03:31] <seb_kuzminsky> that's where i'd like to start, then backport the minimal changes to make the build homogenous to 2.5
[15:03:50] <mhaberler> ok, I'll roll it forward then
[15:04:09] -!- kwallace [kwallace!~kwallace@smb-4.sonnet.com] has joined #linuxcnc-devel
[15:04:19] <seb_kuzminsky> hm, i wonder what the xenomai and preempt-rt buildslaves should do when we ask them to build 2.5
[15:05:07] <mhaberler> it would build just fine off the rtos-integration-preview3 branch
[15:05:45] <mhaberler> with stock v2.5_branch - well if you install rtai headers it might work
[15:05:57] <mhaberler> sim will work anyway
[15:07:00] <mhaberler> there is no link between 'style x' and 'running kernel y' - if you have kheaders and in the case of xenomai the userland support package, build should be fine; runtests - not fine
[15:07:21] <seb_kuzminsky> each buildslave builds for a particular platform (meaning distro/cpu/realtime-or-not), and this target platform is recorded in the deb
[15:07:45] <mhaberler> do you have separate vms for each flavor?
[15:07:59] <seb_kuzminsky> the xenomai & preempt buildslaves won't be able to make debs in 2.5, so we need to not do the build in that case
[15:08:07] <seb_kuzminsky> yes, one VM for each supported platform
[15:09:01] <mhaberler> they would be if rtos-ntegration-preview3 would be merged into v2.5_branch, but thats an open question
[15:09:28] <mhaberler> in that case you'd have a lot more uniformity.
[15:09:35] <seb_kuzminsky> maybe after we have ./configure autodetect what it wants to build based on /boot/config-$(uname), we could could compare that to a per-buildslave expected value
[15:09:41] <mhaberler> (and a better configure..)
[15:10:03] <seb_kuzminsky> imo this change does not belong in 2.5
[15:10:14] <seb_kuzminsky> even though it would be more convenient for build purposes
[15:12:00] <mhaberler> well the third option is to build packages for both rtos-ntegration-preview3 and rtos-ntegration-preview3-merged-into master; otherwise you force everybody to switch to master - which I feel confident with, but I'm not the user majority
[15:12:06] <seb_kuzminsky> the history of the rtos branch is confusing and hard to review :-(
[15:12:29] <mhaberler> it helps if you try it out, lots of things will clear up
[15:13:03] <mhaberler> if you want something really confusing to review, look at rtapi in v2.5_branch versus the current one
[15:13:48] <seb_kuzminsky> it really helps reviewers if you rebase your feature branch on top of the parent branch instead of merging the parent branch into it
[15:14:03] <seb_kuzminsky> then all the new commits are all together at the top of a known foundation
[15:14:28] <seb_kuzminsky> brb
[15:16:48] -!- roh has quit [Ping timeout: 245 seconds]
[15:18:18] -!- mhaberler has quit [Quit: mhaberler]
[15:29:49] yellowfrog is now known as bluefr0g
[15:32:31] -!- mk0 has quit [Quit: Leaving]
[15:45:54] -!- stsydow has quit [Ping timeout: 252 seconds]
[15:46:33] -!- V0idExp has quit [Quit: Leaving.]
[15:53:50] -!- jst has quit [Read error: Connection reset by peer]
[15:54:16] -!- fatpandas has quit [Ping timeout: 264 seconds]
[15:54:48] -!- mrsun has quit [Ping timeout: 264 seconds]
[16:00:31] -!- ve7it [ve7it!~LawrenceG@S01060014d19d0b68.pk.shawcable.net] has joined #linuxcnc-devel
[16:04:21] -!- roh has quit [Read error: Operation timed out]
[16:07:24] firephoto_ is now known as firephoto
[16:21:54] -!- jthornton has quit [Read error: Connection reset by peer]
[16:21:55] -!- JT-Shop has quit [Read error: Connection reset by peer]
[16:22:22] -!- JT-Shop [JT-Shop!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[16:22:23] -!- jthornton [jthornton!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[16:48:15] <seb_kuzminsky> matt shaver for the win
[16:50:26] -!- jfire has quit [Quit: Leaving.]
[16:50:35] <cradek> he's awesome
[16:53:33] <skunkworks> wow
[17:00:00] -!- psha [psha!~psha@213.208.162.67] has joined #linuxcnc-devel
[17:01:14] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[17:06:36] -!- r00t4rd3d has quit [Quit: Leaving]
[17:14:14] -!- Blorb has quit [Remote host closed the connection]
[17:16:54] -!- davec has quit [Ping timeout: 264 seconds]
[17:23:34] -!- ler_hydra has quit [Remote host closed the connection]
[17:29:13] <skunkworks> isn't it easier to ask for forgiveness?
[17:31:59] -!- mackerski has quit [Quit: mackerski]
[17:47:17] -!- tripod| has quit [Ping timeout: 255 seconds]
[17:50:58] <seb_kuzminsky> not from lawyers
[17:51:00] * seb_kuzminsky spits
[17:51:38] -!- psha has quit [Quit: Lost terminal]
[17:51:58] <skunkworks> heh
[17:52:24] <cradek> http://www.zeromq.org/area:licensing
[17:52:56] <cradek> have people actually read this first google hit? what we want to do is clearly allowed
[17:53:11] <seb_kuzminsky> yeah that sounds pretty good for us
[17:53:20] <seb_kuzminsky> 0mq is the suggested replacement for nml, is that right?
[17:53:36] <cradek> yes
[17:54:37] <cradek> they do not specify a version of LGPL where they declare the LGPL is the 'base' license
[17:54:49] <cradek> the link goes to the LGPL3 text, but that URL probably has changed over time
[17:55:09] <cradek> if they are LGPL2 there's no problem and there never was, even without the special exception
[17:56:30] <cradek> ok, the COPYING.LESSER file in the current source is LGPL3
[17:59:03] -!- andypugh [andypugh!~andy2@cpc16-basl9-2-0-cust685.20-1.cable.virginmedia.com] has joined #linuxcnc-devel
[18:03:07] -!- tomate_ has quit [Remote host closed the connection]
[18:06:28] -!- V0idExp has quit [Quit: Leaving.]
[18:07:29] <cradek> I answered gene on -devel
[18:07:38] <cradek> I wish we could all move on
[18:11:35] -!- MarkusBec has quit [Ping timeout: 260 seconds]
[18:11:41] -!- dway has quit [Quit: NOOOOOOooooooooo……]
[18:19:40] <seb_kuzminsky> new rtai debs at
http://highlab.com/~seb/linuxcnc/rtai-for-3.5-prerelease fix the "panic on rmmod" problem i was having
[18:19:48] <cradek> !!
[18:19:52] <cradek> what fixed it?
[18:19:59] <seb_kuzminsky> but i don't understand why i'm running in to this problem on real hardware but not in a vm
[18:20:26] <cradek> you're just lucky. paolo's not seeing it either, right?
[18:20:38] <seb_kuzminsky> it's a null pointer dereference, my patch just papers over it by checking for null first
[18:20:51] <seb_kuzminsky> i dont understand why it's null, i think it shouldnt be
[18:21:08] <cradek> ah I see your email now
[18:21:52] <seb_kuzminsky> i think i'll leave it to paolo and the folks who actually know what's supposed to happen there to fix it for real
[18:22:32] <seb_kuzminsky> i still get this panic, which seems suspiciously similar (timer-related):
http://highlab.com/~seb/linuxcnc/rtai-for-3.5-prerelease/IMAG0577.jpg
[18:22:50] <seb_kuzminsky> this one's in our code, btw
[18:23:11] <cradek> count2nano is ours?
[18:23:26] <seb_kuzminsky> i wonder if my rtai ./configure line is wrong and it's lacking support for some timer hardware that my real hardware has, that my vm doesnt need
[18:23:59] <cradek> you were smart to give him your configure information
[18:24:06] -!- jfire has quit [Quit: Leaving.]
[18:24:37] <seb_kuzminsky> it changed a lot since the 2.6.32 days
[18:24:54] <cradek> count2nano isn't ours
[18:24:59] <seb_kuzminsky> i see that you're right
[18:25:15] <cradek> it takes and returns an integer. I doubt we can be calling it wrongly
[18:25:43] <seb_kuzminsky> it's provided by rtai
[18:26:40] <cradek> AX=BX=CX=DX=0 (but I guess that doesn't mean we called it with 0, considering 0x5b/0xb0)
[18:27:20] -!- AR_ has quit [Ping timeout: 255 seconds]
[18:29:29] <seb_kuzminsky> rtai's count2nano does a bunch of innocent-looking stuff, the only sketchy bit i see is a division by TIMER_FREQ, which is a constant from the config
[18:29:30] -!- uncrtnmind has quit [Quit: Leaving...]
[18:29:51] <seb_kuzminsky> i wonder which timer we're using, there are 3 options, rtc, lapic, and 8254
[18:30:29] <cradek> disassemble it
[18:31:06] <seb_kuzminsky> 0 12:30:50 seb@precise-x86 /home/seb/linux-rtai/rtai-3.9+2013.02.27> grep CONFIG_RTAI_RTC_FREQ config.log
[18:31:09] <cradek> I don't think /0 would panic
[18:31:09] <seb_kuzminsky> #define CONFIG_RTAI_RTC_FREQ 0
[18:31:27] <seb_kuzminsky> check the first line of the output
[18:31:35] <seb_kuzminsky> divide error: 0000
[18:31:37] <cradek> oh!
[18:31:50] <cradek> so .....
[18:32:14] <cradek> so it is a config/build problem?
[18:32:16] <seb_kuzminsky> i wonder if my kernel lacks rtc support? or there's a mismatch between the timers my kernel provides and the timers rtai knows to read?
[18:32:28] <seb_kuzminsky> seems so right now...
[18:32:34] <cradek> is your kernel smp?
[18:32:36] <seb_kuzminsky> but my confidence is low still
[18:32:38] <seb_kuzminsky> yeah
[18:32:48] <cradek> oh ok (I think he said he'd only tested on smp)
[18:32:48] <seb_kuzminsky> ok, i
[18:33:01] <seb_kuzminsky> i'll pick up here maybe tomorrow night
[18:33:08] <seb_kuzminsky> i need to work :-/
[18:33:31] <seb_kuzminsky> and hey, maybe if i wait paolo will come back with suggestions!
[18:33:41] <cradek> yeah, seems clues are everywhere
[18:34:24] -!- tmcw has quit [Remote host closed the connection]
[18:34:38] <seb_kuzminsky> hmm, rtai's ./configure has a --enable-rtc-freq...
[18:35:00] <seb_kuzminsky> ok fine, *one* more recompile/test, *then* i'll work...
[18:36:37] <seb_kuzminsky> hmm, it wants a number, and i dont know what to give it
[18:41:38] -!- jfire has quit [Quit: Leaving.]
[18:59:51] -!- syyl_ws has quit [Quit: Verlassend]
[19:14:46] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 18.0/20130108033621]]
[19:18:27] -!- cncbasher_ [cncbasher_!~quassel@cpc15-hart9-2-0-cust101.11-3.cable.virginmedia.com] has joined #linuxcnc-devel
[19:29:09] -!- stsydow has quit [Remote host closed the connection]
[19:41:52] -!- ler_hydra has quit [Remote host closed the connection]
[19:45:01] s1dev_ is now known as s1dev
[19:50:50] -!- syyl has quit [Ping timeout: 245 seconds]
[19:57:05] -!- tjb1 has quit [Quit: tjb1]
[19:58:34] -!- skunkworks has quit [Read error: Connection reset by peer]
[20:15:22] -!- asw has quit [Quit: Ухожу я от вас]
[20:16:35] -!- nunya has quit [Quit: Leaving]
[20:19:38] -!- Santolina has quit [Ping timeout: 245 seconds]
[21:00:05] -!- BHSPiMonkey has quit [Changing host]
[21:01:32] -!- FinboySlick has quit [Quit: Leaving.]
[21:07:59] -!- skunkworks [skunkworks!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[21:12:23] <alex_joni> seb_kuzminsky: 42
[21:19:31] -!- tshirtisinthemai has quit [Client Quit]
[21:22:56] -!- odogono has quit [Quit: odogono]
[21:26:53] -!- syyl_ has quit [Quit: Leaving]
[21:29:11] -!- _ink has quit [Ping timeout: 245 seconds]
[21:55:21] <seb_kuzminsky> actually i now think the rtc frequency is not the problem at all
[21:55:36] <seb_kuzminsky> first of all my machine seems to be using the apic for timing, not the rtc
[21:55:50] <seb_kuzminsky> and second of all i think it should be autodetecting the frequency, i shouldnt have to tell it
[21:56:06] <seb_kuzminsky> it autodetects it correctly on my VM, and gets it wrong on the real computer that runs my mill
[21:56:25] <seb_kuzminsky> i've mailed the rtai list about it, along with some debugging info & logs & stuff, we'll see what comes of it
[21:56:48] -!- tjb1 has quit [Quit: tjb1]
[21:57:49] -!- Brandonian has quit [Quit: Brandonian]
[22:01:08] -!- chillly has quit [Quit: Leaving]
[22:07:23] -!- Loetmichel has quit [Ping timeout: 255 seconds]
[22:08:48] Cylly is now known as Loetmichel
[22:14:01] <skunkworks> seb_kuzminsky: thanks for plugging away at this..
[22:19:28] -!- xyzee has quit [Remote host closed the connection]
[22:22:09] -!- DJ9DJ has quit [Quit: bye]
[22:27:41] <seb_kuzminsky> sure! it feels like we're close to fixing this particular bug
[22:27:53] -!- skorasaurus has quit [Ping timeout: 252 seconds]
[22:27:59] <seb_kuzminsky> there's no knowing how many are lurking elsewhere in the code...
[22:33:00] -!- pfred1 has quit [Quit: switching shells]
[22:37:11] -!- zzolo has quit [Quit: zzolo]
[22:40:47] -!- stsydow has quit [Quit: Leaving]
[22:41:40] -!- ravenlock has quit [Quit: Leaving]
[22:46:40] -!- tjb1 has quit [Client Quit]
[22:58:11] -!- mhaberler has quit [Read error: Operation timed out]
[22:59:18] -!- skorasaurus has quit [Remote host closed the connection]
[23:01:51] <andypugh> Oh my! This is a bug I can reproduce...
http://www.linuxcnc.org/index.php/english/forum/10-advanced-configuration/26272-external-jogging?limitstart=0
[23:05:12] <JT-Shop> really
[23:05:21] <andypugh> It seems that setting halui.joint.3.select and then halui.jog.selected.plus jogs joint 2
[23:05:28] <JT-Shop> I've been out of the shop...
[23:05:33] <JT-Shop> ouch
[23:06:18] <andypugh> Now to switch to 2.5 and see if it is the same there
[23:07:07] asdfasd is now known as Guest25877
[23:09:01] -!- Guest25877 has quit [Ping timeout: 240 seconds]
[23:10:49] shdhdfghd is now known as asdfasd
[23:16:14] -!- fomox has quit [Ping timeout: 255 seconds]
[23:25:13] -!- V0idExp has quit [Quit: Leaving.]
[23:26:09] -!- Tecan has quit [Quit: Ex-Chat]
[23:28:03] -!- tmcw has quit [Remote host closed the connection]
[23:34:24] -!- MyDrupalJourney has quit [Quit: Gotta run!]
[23:38:44] <kwallace> In case you missed out on the last one, here is another:
http://www.ebay.com/itm/111027258474
[23:53:37] -!- Nick001-Shop has quit [Remote host closed the connection]
[23:57:16] <seb_kuzminsky> sounds like paolo has some ideas about the rtai timer problem i ran in to
[23:58:36] <andypugh> And halui jogging _does_ work in 2.5
[23:58:47] <andypugh> Houston, we have a bug
[23:58:53] -!- tmcw has quit [Remote host closed the connection]
