Back
[01:36:50] <KimK> I made a minor wiki edit but forgot to set a user name (KimK is fine) before saving, so it only shows an ip address as the editor. Sorry, it was me, no security breach or anything, no need to panic. Suggestion: on a save with default user name, say, "You didn't set a user name, would you like to?"
[04:41:38] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15andypugh commented on issue #59: Can you elaborate on the disablement of /proc ? Is that a kernel thing or a LinuxCNC thing?... 02
https://github.com/LinuxCNC/linuxcnc/issues/59#issuecomment-238497998
[05:20:56] <archivist> we have a basic bug in docs that means a google search for something finds old docs and nothing in current docs eg "linuxcnc ini variables" where are the new ones available in 2.7 :)
[05:25:22] <archivist> in my google results the following url is missing
http://linuxcnc.org/docs/html/config/ini-config.html
[07:29:54] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15jepler commented on issue #59: In the distant past, RTAPI + RTAI created entries in /proc. By kernel 3.10, the APIs for doing so changed incompatibly, and nobody updated the code to continue working -- I just disabled it instead. (ec38ac4566e07baa3e47e0a0463ac7384c92f3aa). But creating /proc entries is simply not possible for uspace-style realtime, so it's not a good solution for that either. 02
https://github.com/Li
[08:48:25] <cradek> archivist: I think we can't directly affect that, unfortunately. our current docs are always linked from the top of the front page of linuxcnc.org.
[08:49:09] <cradek> from there you can click on "ini configuration"
[09:15:06] <archivist> something fishy going on
[09:19:04] <archivist> it would also be helpful to be able to move from 2.6 to 2.7 by adjusting the url google gives eg
http://linuxcnc.org/docs/2.6/html/config/ini_config.html
[09:24:46] <archivist> even just removing the 2.6 fails
http://linuxcnc.org/docs/html/config/ini_config.html
[09:25:03] <archivist> silly change of _ to -
[09:25:36] <archivist> I meant
http://linuxcnc.org/docs/2.6/html/config/ini_config.html
[09:44:01] <cradek> I sympathize with wanting to find up-to-date documentation with google. it just never works very well for that.
[09:44:51] <cradek> I spent yesterday afternoon looking for cyanogenmod docs using google as my primary hammer and it sucked at it
[09:47:52] <archivist> just needs consistency
[15:25:23] <mozmck> Should a rotary axis slow down as you are jogging past the home position?
[15:37:17] <jepler> mozmck: sounds odd
[15:37:31] <cradek> I would sure not expect that
[15:45:32] <mozmck> Yeah, so here's what the guy says: "for some reason our b axis slows down when jogging past its home position, it will jog around both directions, but slows down while passing home, and I can see the jog federate dro change as it slows down"
[15:46:14] <mozmck> and, "also when I am jogging it around the “slow spot”, if I stop near the spot it will stop jogging and ALL other axis’ stop jogging around as well , until I push a home button for any axis on the GUI, then all returns to normal. This is happening reliably around the same spot."
[15:47:11] <cradek> and let me guess - when you try it with a sim config it doesn't happen?
[15:47:11] <mozmck> Which really sounds odd - I'm not sure I quite understand what he is even saying.
[15:50:11] <mozmck> Hmm, I have not done that yet
[15:52:46] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly 45b6c2d 06linuxcnc 10src/rtapi/uspace_rtapi_app.cc uspace: restore thread that reads out messagse from realtime * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=45b6c2d
[15:52:46] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly f93962b 06linuxcnc 10src/hal/user_comps/mb2hal/Submakefile 10src/hal/user_comps/mb2hal/mb2hal.h mb2hal: remove needless dependency on linuxcnc * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=f93962b
[15:52:46] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly f4322fd 06linuxcnc 10src/Makefile 10src/Makefile.inc.in 10src/configure.in WIP HACK: enable a 'hal only' build of linuxcnc * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=f4322fd
[16:08:22] <linuxcnc-build> build #4436 of 1300.rip-precise-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1300.rip-precise-i386/builds/4436 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:08:51] <linuxcnc-build> build #2597 of 1400.rip-wheezy-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1400.rip-wheezy-i386/builds/2597 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:09:28] <linuxcnc-build> build #1067 of 1530.rip-jessie-rtpreempt-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1530.rip-jessie-rtpreempt-amd64/builds/1067 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:10:03] <linuxcnc-build> build #4438 of 1306.rip-precise-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1306.rip-precise-amd64/builds/4438 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:10:23] <jepler> clearly I should have tested that a little more before pushing it .. looks like it wrecks a regular, full build
[16:10:34] <jepler> s/a little more/at all/
[16:11:05] <linuxcnc-build> build #2798 of 1404.rip-wheezy-rtpreempt-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1404.rip-wheezy-rtpreempt-amd64/builds/2798 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:11:18] <linuxcnc-build> build #2109 of 1402.rip-wheezy-rtpreempt-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1402.rip-wheezy-rtpreempt-i386/builds/2109 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:11:22] <linuxcnc-build> build #1065 of 1500.rip-jessie-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1500.rip-jessie-i386/builds/1065 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:11:27] <linuxcnc-build> build #1065 of 1520.rip-jessie-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1520.rip-jessie-amd64/builds/1065 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:11:31] <linuxcnc-build> build #3648 of 1301.rip-precise-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1301.rip-precise-rtai-i386/builds/3648 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:11:54] <linuxcnc-build> build #1065 of 1510.rip-jessie-rtpreempt-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1510.rip-jessie-rtpreempt-i386/builds/1065 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:12:15] <linuxcnc-build> build #2597 of 1403.rip-wheezy-amd64 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1403.rip-wheezy-amd64/builds/2597 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:13:42] <linuxcnc-build> build #2263 of 1401.rip-wheezy-rtai-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1401.rip-wheezy-rtai-i386/builds/2263 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:16:32] <linuxcnc-build> build #2631 of 1405.rip-wheezy-armhf is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1405.rip-wheezy-armhf/builds/2631 blamelist: Jeff Epler <jepler@unpythonic.net>
[16:16:33] <linuxcnc-build> build #4453 of 0000.checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4453 blamelist: Jeff Epler <jepler@unpythonic.net>
[17:08:27] <jepler> $ debian/configure uspace xenomai
[17:08:27] <jepler> E: You must put some 'source' URIs in your sources.list
[17:08:43] <jepler> sigh, ugh, some champion people set up sources.list without any deb-src listed
[17:08:55] <jepler> I recently made debian/configure depend on this not being broken
[17:10:26] <jepler> XENOMAI_ARCHITECTURE="$(apt-cache showsrc libxenomai-dev \
[17:10:28] <jepler> ...
[17:10:29] <jepler> )
[17:17:28] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly a252c97 06linuxcnc 10src/Makefile 10src/Makefile.inc.in 10src/configure.in WIP HACK: enable a 'hal only' build of linuxcnc * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=a252c97
[17:46:11] <linuxcnc-build> build #2110 of 1402.rip-wheezy-rtpreempt-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1402.rip-wheezy-rtpreempt-i386/builds/2110 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:52:20] <linuxcnc-build> build #4454 of 0000.checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4454 blamelist: dummy, Jeff Epler <jepler@unpythonic.net>
[17:55:49] <jepler> for some unfathomable reason, latency-histogram requires librs274.so to work
[17:56:34] <jepler> https://emergent.unpythonic.net/files/sandbox/bbone-xenomai-linuxcnc.png
[18:29:05] <jmk-mcfaul_> jepler, nice!
[18:46:16] <cradek> wow!
[19:08:25] <jepler> xenomai realtime worked the first time
[19:08:41] <jepler> no hardware drivers yet but a little bird told me that we might have someone who is motivated to develop at least a GPIO driver
[19:37:12] <cradek> woo
[19:50:32] <jepler> 44.2us latency after 1.5h
[19:50:43] <cradek> downright useful
[19:52:56] <skunkworks> Couldn't you use the machinekit bit file?
[19:53:23] <skunkworks> For the bb
[19:54:23] <skunkworks> jmk-mcfaul_: good to see you around
[20:00:34] <jepler> skunkworks: in fact starting with the hal_bb_gpio driver from machinekit looks very feasible
[20:06:14] <skunkworks> Cool
[20:39:09] <jepler> this doesn't need anything but a HAL driver, it's not using PRUs (or at least not customizing the software in the PRUs)
[20:48:38] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly 48578a4 06linuxcnc 10.gitignore 10docs/man/.gitignore git: only ignore .9 (component manual) pages in docs/man * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=48578a4
[20:48:38] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly 484982c 06linuxcnc 03docs/man/man3/rtapi_open_as_root.3rtapi 10src/hal/drivers/mesa-hostmot2/hm2_rpspi.c 10src/rtapi/rtapi.h 10src/rtapi/uspace_rtapi_app.cc rtapi: Add rtapi_open_as_root API * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=484982c
[20:48:38] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly 0e582f0 06linuxcnc 10src/Makefile 03src/hal/drivers/beaglebone_gpio.h 03src/hal/drivers/hal_bb_gpio.c hal_bb_gpio: new hardware driver * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=0e582f0
[20:48:42] <KGB-linuxcnc> 03Jeff Epler 05jepler/master/halonly d779099 06linuxcnc 03docs/src/man/man9/hal_bb_gpio.9 hal_bb_gpio: add manpage * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=d779099
[20:49:31] <cradek> well that was fast
[20:49:32] <jepler> well boy, merge that and we can claim to work on beaglebone as well as raspberry pi
[20:49:56] <jepler> the build time of linuxcnc "halonly" isn't even that bad, under 20 minutes
[20:49:59] <jepler> and rebuilds in under a minute
[20:50:07] <jepler> (change-one-file-and-builds)
[20:51:59] <jepler> I ran into some of the trouble that jmk-mcfaul_ alluded to, where specifying a "wrong" pin configuration just crashed rtapi_app
[20:52:02] <jepler> so boo for that
[20:54:22] <jepler> https://emergent.unpythonic.net/files/sandbox/hal_bb_gpio.9.txt
[21:10:30] <linuxcnc-build> build #2633 of 1405.rip-wheezy-armhf is complete: Failure [4failed compile] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1405.rip-wheezy-armhf/builds/2633 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:18:38] <linuxcnc-build> build #4455 of 0000.checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4455 blamelist: Jeff Epler <jepler@unpythonic.net>
[21:19:30] <jepler> one image shows that anybody who says to try putting your unix dev environment inside the new windows fake ubuntu is crazy:
https://openbenchmarking.org/embed.php?i=1608096-LO-BASHWINDO87&sha=8419652&p=2
[21:20:57] <cradek> for the low low price of a windows license you can slow down your work by almost a factor of 100?
[21:21:02] <cradek> how can anyone resist that
[21:21:46] <jepler> that's approximately right
[21:21:59] <jepler> and maybe you're stuck with ubuntu instead of your favorite linux, or maybe that's just the marketing talking
[21:22:45] <jepler> to be fair I don't know what that test is. this test of actually compiling php is only 1.2x as long using the windows fake ubuntu:
https://openbenchmarking.org/embed.php?i=1608096-LO-BASHWINDO87&sha=8b7a7ca&p=2
[23:07:38] <jepler> huh I wonder if the big gap in the servo thread timing is related to how long the fast thread takes to run?
https://emergent.unpythonic.net/files/sandbox/linuxcnc-bbb-xenomai-two-threads.png
